Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "Here's the main drm pull request for 4.7, it's been a busy one, and
  I've been a bit more distracted in real life this merge window.  Lots
  more ARM drivers, not sure if it'll ever end.  I think I've at least
  one more coming the next merge window.

  But changes are all over the place, support for AMD Polaris GPUs is in
  here, some missing GM108 support for nouveau (found in some Lenovos),
  a bunch of MST and skylake fixes.

  I've also noticed a few fixes from Arnd in my inbox, that I'll try and
  get in asap, but I didn't think they should hold this up.

  New drivers:
   - Hisilicon kirin display driver
   - Mediatek MT8173 display driver
   - ARC PGU - bitstreamer on Synopsys ARC SDP boards
   - Allwinner A13 initial RGB output driver
   - Analogix driver for DisplayPort IP found in exynos and rockchip

  DRM Core:
   - UAPI headers fixes and C++ safety
   - DRM connector reference counting
   - DisplayID mode parsing for Dell 5K monitors
   - Removal of struct_mutex from drivers
   - Connector registration cleanups
   - MST robustness fixes
   - MAINTAINERS updates
   - Lockless GEM object freeing
   - Generic fbdev deferred IO support

  panel:
   - Support for a bunch of new panels

  i915:
   - VBT refactoring
   - PLL computation cleanups
   - DSI support for BXT
   - Color manager support
   - More atomic patches
   - GEM improvements
   - GuC fw loading fixes
   - DP detection fixes
   - SKL GPU hang fixes
   - Lots of BXT fixes

  radeon/amdgpu:
   - Initial Polaris support
   - GPUVM/Scheduler/Clock/Power improvements
   - ASYNC pageflip support
   - New mesa feature support

  nouveau:
   - GM108 support
   - Power sensor support improvements
   - GR init + ucode fixes.
   - Use GPU provided topology information

  vmwgfx:
   - Add host messaging support

  gma500:
   - Some cleanups and fixes

  atmel:
   - Bridge support
   - Async atomic commit support

  fsl-dcu:
   - Timing controller for LCD support
   - Pixel clock polarity support

  rcar-du:
   - Misc fixes

  exynos:
   - Pipeline clock support
   - Exynoss4533 SoC support
   - HW trigger mode support
   - export HDMI_PHY clock
   - DECON5433 fixes
   - Use generic prime functions
   - use DMA mapping APIs

  rockchip:
   - Lots of little fixes

  vc4:
   - Render node support
   - Gamma ramp support
   - DPI output support

  msm:
   - Mostly cleanups and fixes
   - Conversion to generic struct fence

  etnaviv:
   - Fix for prime buffer handling
   - Allow hangcheck to be coalesced with other wakeups

  tegra:
   - Gamme table size fix"

* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (1050 commits)
  drm/edid: add displayid detailed 1 timings to the modelist. (v1.1)
  drm/edid: move displayid validation to it's own function.
  drm/displayid: Iterate over all DisplayID blocks
  drm/edid: move displayid tiled block parsing into separate function.
  drm: Nuke ->vblank_disable_allowed
  drm/vmwgfx: Report vmwgfx version to vmware.log
  drm/vmwgfx: Add VMWare host messaging capability
  drm/vmwgfx: Kill some lockdep warnings
  drm/nouveau/gr/gf100-: fix race condition in fecs/gpccs ucode
  drm/nouveau/core: recognise GM108 chipsets
  drm/nouveau/gr/gm107-: fix touching non-existent ppcs in attrib cb setup
  drm/nouveau/gr/gk104-: share implementation of ppc exception init
  drm/nouveau/gr/gk104-: move rop_active_fbps init to nonctx
  drm/nouveau/bios/pll: check BIT table version before trying to parse it
  drm/nouveau/bios/pll: prevent oops when limits table can't be parsed
  drm/nouveau/volt/gk104: round up in gk104_volt_set
  drm/nouveau/fb/gm200: setup mmu debug buffer registers at init()
  drm/nouveau/fb/gk20a,gm20b: setup mmu debug buffer registers at init()
  drm/nouveau/fb/gf100-: allocate mmu debug buffers
  drm/nouveau/fb: allow chipset-specific actions for oneinit()
  ...

76 files changed, 12914 insertions, 9374 deletions

diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
index 20a5d0455e19..29a32b11953b 100644
--- a/drivers/gpu/drm/i915/Kconfig
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -56,3 +56,9 @@ config DRM_I915_USERPTR
 	  selected to enabled full userptr support.
 
 	  If in doubt, say "Y".
+
+menu "drm/i915 Debugging"
+depends on DRM_I915
+depends on EXPERT
+source drivers/gpu/drm/i915/Kconfig.debug
+endmenu
diff --git a/drivers/gpu/drm/i915/Kconfig.debug b/drivers/gpu/drm/i915/Kconfig.debug
new file mode 100644
index 000000000000..8f404103341d
--- /dev/null
+++ b/drivers/gpu/drm/i915/Kconfig.debug
@@ -0,0 +1,41 @@
+config DRM_I915_WERROR
+        bool "Force GCC to throw an error instead of a warning when compiling"
+        # As this may inadvertently break the build, only allow the user
+        # to shoot oneself in the foot iff they aim really hard
+        depends on EXPERT
+        # We use the dependency on !COMPILE_TEST to not be enabled in
+        # allmodconfig or allyesconfig configurations
+        depends on !COMPILE_TEST
+        default n
+        help
+          Add -Werror to the build flags for (and only for) i915.ko.
+          Do not enable this unless you are writing code for the i915.ko module.
+
+          Recommended for driver developers only.
+
+          If in doubt, say "N".
+
+config DRM_I915_DEBUG
+        bool "Enable additional driver debugging"
+        depends on DRM_I915
+        default n
+        help
+          Choose this option to turn on extra driver debugging that may affect
+          performance but will catch some internal issues.
+
+          Recommended for driver developers only.
+
+          If in doubt, say "N".
+
+config DRM_I915_DEBUG_GEM
+        bool "Insert extra checks into the GEM internals"
+        default n
+        depends on DRM_I915_WERROR
+        help
+          Enable extra sanity checks (including BUGs) along the GEM driver
+          paths that may slow the system down and if hit hang the machine.
+
+          Recommended for driver developers only.
+
+          If in doubt, say "N".
+
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 0851de07bd13..0b88ba0f3c1f 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -2,6 +2,8 @@
 # Makefile for the drm device driver.  This driver provides support for the
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
+subdir-ccflags-$(CONFIG_DRM_I915_WERROR) := -Werror
+
 # Please keep these build lists sorted!
 
 # core driver code
@@ -55,7 +57,9 @@ i915-y += intel_audio.o \
 	  intel_atomic.o \
 	  intel_atomic_plane.o \
 	  intel_bios.o \
+	  intel_color.o \
 	  intel_display.o \
+	  intel_dpll_mgr.o \
 	  intel_fbc.o \
 	  intel_fifo_underrun.o \
 	  intel_frontbuffer.o \
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 814d894ed925..a337f33bec5b 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -444,6 +444,7 @@ static const struct drm_i915_reg_descriptor gen7_render_regs[] = {
 	REG64(CL_PRIMITIVES_COUNT),
 	REG64(PS_INVOCATION_COUNT),
 	REG64(PS_DEPTH_COUNT),
+	REG64_IDX(RING_TIMESTAMP, RENDER_RING_BASE),
 	REG32(OACONTROL), /* Only allowed for LRI and SRM. See below. */
 	REG64(MI_PREDICATE_SRC0),
 	REG64(MI_PREDICATE_SRC1),
@@ -471,6 +472,25 @@ static const struct drm_i915_reg_descriptor gen7_render_regs[] = {
 	REG32(GEN7_L3SQCREG1),
 	REG32(GEN7_L3CNTLREG2),
 	REG32(GEN7_L3CNTLREG3),
+};
+
+static const struct drm_i915_reg_descriptor hsw_render_regs[] = {
+	REG64_IDX(HSW_CS_GPR, 0),
+	REG64_IDX(HSW_CS_GPR, 1),
+	REG64_IDX(HSW_CS_GPR, 2),
+	REG64_IDX(HSW_CS_GPR, 3),
+	REG64_IDX(HSW_CS_GPR, 4),
+	REG64_IDX(HSW_CS_GPR, 5),
+	REG64_IDX(HSW_CS_GPR, 6),
+	REG64_IDX(HSW_CS_GPR, 7),
+	REG64_IDX(HSW_CS_GPR, 8),
+	REG64_IDX(HSW_CS_GPR, 9),
+	REG64_IDX(HSW_CS_GPR, 10),
+	REG64_IDX(HSW_CS_GPR, 11),
+	REG64_IDX(HSW_CS_GPR, 12),
+	REG64_IDX(HSW_CS_GPR, 13),
+	REG64_IDX(HSW_CS_GPR, 14),
+	REG64_IDX(HSW_CS_GPR, 15),
 	REG32(HSW_SCRATCH1,
 	      .mask = ~HSW_SCRATCH1_L3_DATA_ATOMICS_DISABLE,
 	      .value = 0),
@@ -500,6 +520,33 @@ static const struct drm_i915_reg_descriptor hsw_master_regs[] = {
 #undef REG64
 #undef REG32
 
+struct drm_i915_reg_table {
+	const struct drm_i915_reg_descriptor *regs;
+	int num_regs;
+	bool master;
+};
+
+static const struct drm_i915_reg_table ivb_render_reg_tables[] = {
+	{ gen7_render_regs, ARRAY_SIZE(gen7_render_regs), false },
+	{ ivb_master_regs, ARRAY_SIZE(ivb_master_regs), true },
+};
+
+static const struct drm_i915_reg_table ivb_blt_reg_tables[] = {
+	{ gen7_blt_regs, ARRAY_SIZE(gen7_blt_regs), false },
+	{ ivb_master_regs, ARRAY_SIZE(ivb_master_regs), true },
+};
+
+static const struct drm_i915_reg_table hsw_render_reg_tables[] = {
+	{ gen7_render_regs, ARRAY_SIZE(gen7_render_regs), false },
+	{ hsw_render_regs, ARRAY_SIZE(hsw_render_regs), false },
+	{ hsw_master_regs, ARRAY_SIZE(hsw_master_regs), true },
+};
+
+static const struct drm_i915_reg_table hsw_blt_reg_tables[] = {
+	{ gen7_blt_regs, ARRAY_SIZE(gen7_blt_regs), false },
+	{ hsw_master_regs, ARRAY_SIZE(hsw_master_regs), true },
+};
+
 static u32 gen7_render_get_cmd_length_mask(u32 cmd_header)
 {
 	u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT;
@@ -555,7 +602,7 @@ static u32 gen7_blt_get_cmd_length_mask(u32 cmd_header)
 	return 0;
 }
 
-static bool validate_cmds_sorted(struct intel_engine_cs *ring,
+static bool validate_cmds_sorted(struct intel_engine_cs *engine,
 				 const struct drm_i915_cmd_table *cmd_tables,
 				 int cmd_table_count)
 {
@@ -577,7 +624,7 @@ static bool validate_cmds_sorted(struct intel_engine_cs *ring,
 
 			if (curr < previous) {
 				DRM_ERROR("CMD: table not sorted ring=%d table=%d entry=%d cmd=0x%08X prev=0x%08X\n",
-					  ring->id, i, j, curr, previous);
+					  engine->id, i, j, curr, previous);
 				ret = false;
 			}
 
@@ -611,11 +658,18 @@ static bool check_sorted(int ring_id,
 	return ret;
 }
 
-static bool validate_regs_sorted(struct intel_engine_cs *ring)
+static bool validate_regs_sorted(struct intel_engine_cs *engine)
 {
-	return check_sorted(ring->id, ring->reg_table, ring->reg_count) &&
-		check_sorted(ring->id, ring->master_reg_table,
-			     ring->master_reg_count);
+	int i;
+	const struct drm_i915_reg_table *table;
+
+	for (i = 0; i < engine->reg_table_count; i++) {
+		table = &engine->reg_tables[i];
+		if (!check_sorted(engine->id, table->regs, table->num_regs))
+			return false;
+	}
+
+	return true;
 }
 
 struct cmd_node {
@@ -639,13 +693,13 @@ struct cmd_node {
  */
 #define CMD_HASH_MASK STD_MI_OPCODE_MASK
 
-static int init_hash_table(struct intel_engine_cs *ring,
+static int init_hash_table(struct intel_engine_cs *engine,
 			   const struct drm_i915_cmd_table *cmd_tables,
 			   int cmd_table_count)
 {
 	int i, j;
 
-	hash_init(ring->cmd_hash);
+	hash_init(engine->cmd_hash);
 
 	for (i = 0; i < cmd_table_count; i++) {
 		const struct drm_i915_cmd_table *table = &cmd_tables[i];
@@ -660,7 +714,7 @@ static int init_hash_table(struct intel_engine_cs *ring,
 				return -ENOMEM;
 
 			desc_node->desc = desc;
-			hash_add(ring->cmd_hash, &desc_node->node,
+			hash_add(engine->cmd_hash, &desc_node->node,
 				 desc->cmd.value & CMD_HASH_MASK);
 		}
 	}
@@ -668,13 +722,13 @@ static int init_hash_table(struct intel_engine_cs *ring,
 	return 0;
 }
 
-static void fini_hash_table(struct intel_engine_cs *ring)
+static void fini_hash_table(struct intel_engine_cs *engine)
 {
 	struct hlist_node *tmp;
 	struct cmd_node *desc_node;
 	int i;
 
-	hash_for_each_safe(ring->cmd_hash, i, tmp, desc_node, node) {
+	hash_for_each_safe(engine->cmd_hash, i, tmp, desc_node, node) {
 		hash_del(&desc_node->node);
 		kfree(desc_node);
 	}
@@ -690,18 +744,18 @@ static void fini_hash_table(struct intel_engine_cs *ring)
  *
  * Return: non-zero if initialization fails
  */
-int i915_cmd_parser_init_ring(struct intel_engine_cs *ring)
+int i915_cmd_parser_init_ring(struct intel_engine_cs *engine)
 {
 	const struct drm_i915_cmd_table *cmd_tables;
 	int cmd_table_count;
 	int ret;
 
-	if (!IS_GEN7(ring->dev))
+	if (!IS_GEN7(engine->dev))
 		return 0;
 
-	switch (ring->id) {
+	switch (engine->id) {
 	case RCS:
-		if (IS_HASWELL(ring->dev)) {
+		if (IS_HASWELL(engine->dev)) {
 			cmd_tables = hsw_render_ring_cmds;
 			cmd_table_count =
 				ARRAY_SIZE(hsw_render_ring_cmds);
@@ -710,26 +764,23 @@ int i915_cmd_parser_init_ring(struct intel_engine_cs *ring)
 			cmd_table_count = ARRAY_SIZE(gen7_render_cmds);
 		}
 
-		ring->reg_table = gen7_render_regs;
-		ring->reg_count = ARRAY_SIZE(gen7_render_regs);
-
-		if (IS_HASWELL(ring->dev)) {
-			ring->master_reg_table = hsw_master_regs;
-			ring->master_reg_count = ARRAY_SIZE(hsw_master_regs);
+		if (IS_HASWELL(engine->dev)) {
+			engine->reg_tables = hsw_render_reg_tables;
+			engine->reg_table_count = ARRAY_SIZE(hsw_render_reg_tables);
 		} else {
-			ring->master_reg_table = ivb_master_regs;
-			ring->master_reg_count = ARRAY_SIZE(ivb_master_regs);
+			engine->reg_tables = ivb_render_reg_tables;
+			engine->reg_table_count = ARRAY_SIZE(ivb_render_reg_tables);
 		}
 
-		ring->get_cmd_length_mask = gen7_render_get_cmd_length_mask;
+		engine->get_cmd_length_mask = gen7_render_get_cmd_length_mask;
 		break;
 	case VCS:
 		cmd_tables = gen7_video_cmds;
 		cmd_table_count = ARRAY_SIZE(gen7_video_cmds);
-		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
+		engine->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
 	case BCS:
-		if (IS_HASWELL(ring->dev)) {
+		if (IS_HASWELL(engine->dev)) {
 			cmd_tables = hsw_blt_ring_cmds;
 			cmd_table_count = ARRAY_SIZE(hsw_blt_ring_cmds);
 		} else {
@@ -737,44 +788,41 @@ int i915_cmd_parser_init_ring(struct intel_engine_cs *ring)
 			cmd_table_count = ARRAY_SIZE(gen7_blt_cmds);
 		}
 
-		ring->reg_table = gen7_blt_regs;
-		ring->reg_count = ARRAY_SIZE(gen7_blt_regs);
-
-		if (IS_HASWELL(ring->dev)) {
-			ring->master_reg_table = hsw_master_regs;
-			ring->master_reg_count = ARRAY_SIZE(hsw_master_regs);
+		if (IS_HASWELL(engine->dev)) {
+			engine->reg_tables = hsw_blt_reg_tables;
+			engine->reg_table_count = ARRAY_SIZE(hsw_blt_reg_tables);
 		} else {
-			ring->master_reg_table = ivb_master_regs;
-			ring->master_reg_count = ARRAY_SIZE(ivb_master_regs);
+			engine->reg_tables = ivb_blt_reg_tables;
+			engine->reg_table_count = ARRAY_SIZE(ivb_blt_reg_tables);
 		}
 
-		ring->get_cmd_length_mask = gen7_blt_get_cmd_length_mask;
+		engine->get_cmd_length_mask = gen7_blt_get_cmd_length_mask;
 		break;
 	case VECS:
 		cmd_tables = hsw_vebox_cmds;
 		cmd_table_count = ARRAY_SIZE(hsw_vebox_cmds);
 		/* VECS can use the same length_mask function as VCS */
-		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
+		engine->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
 	default:
 		DRM_ERROR("CMD: cmd_parser_init with unknown ring: %d\n",
-			  ring->id);
+			  engine->id);
 		BUG();
 	}
 
-	BUG_ON(!validate_cmds_sorted(ring, cmd_tables, cmd_table_count));
-	BUG_ON(!validate_regs_sorted(ring));
+	BUG_ON(!validate_cmds_sorted(engine, cmd_tables, cmd_table_count));
+	BUG_ON(!validate_regs_sorted(engine));
 
-	WARN_ON(!hash_empty(ring->cmd_hash));
+	WARN_ON(!hash_empty(engine->cmd_hash));
 
-	ret = init_hash_table(ring, cmd_tables, cmd_table_count);
+	ret = init_hash_table(engine, cmd_tables, cmd_table_count);
 	if (ret) {
 		DRM_ERROR("CMD: cmd_parser_init failed!\n");
-		fini_hash_table(ring);
+		fini_hash_table(engine);
 		return ret;
 	}
 
-	ring->needs_cmd_parser = true;
+	engine->needs_cmd_parser = true;
 
 	return 0;
 }
@@ -786,21 +834,21 @@ int i915_cmd_parser_init_ring(struct intel_engine_cs *ring)
  * Releases any resources related to command parsing that may have been
  * initialized for the specified ring.
  */
-void i915_cmd_parser_fini_ring(struct intel_engine_cs *ring)
+void i915_cmd_parser_fini_ring(struct intel_engine_cs *engine)
 {
-	if (!ring->needs_cmd_parser)
+	if (!engine->needs_cmd_parser)
 		return;
 
-	fini_hash_table(ring);
+	fini_hash_table(engine);
 }
 
 static const struct drm_i915_cmd_descriptor*
-find_cmd_in_table(struct intel_engine_cs *ring,
+find_cmd_in_table(struct intel_engine_cs *engine,
 		  u32 cmd_header)
 {
 	struct cmd_node *desc_node;
 
-	hash_for_each_possible(ring->cmd_hash, desc_node, node,
+	hash_for_each_possible(engine->cmd_hash, desc_node, node,
 			       cmd_header & CMD_HASH_MASK) {
 		const struct drm_i915_cmd_descriptor *desc = desc_node->desc;
 		u32 masked_cmd = desc->cmd.mask & cmd_header;
@@ -822,18 +870,18 @@ find_cmd_in_table(struct intel_engine_cs *ring,
  * ring's default length encoding and returns default_desc.
  */
 static const struct drm_i915_cmd_descriptor*
-find_cmd(struct intel_engine_cs *ring,
+find_cmd(struct intel_engine_cs *engine,
 	 u32 cmd_header,
 	 struct drm_i915_cmd_descriptor *default_desc)
 {
 	const struct drm_i915_cmd_descriptor *desc;
 	u32 mask;
 
-	desc = find_cmd_in_table(ring, cmd_header);
+	desc = find_cmd_in_table(engine, cmd_header);
 	if (desc)
 		return desc;
 
-	mask = ring->get_cmd_length_mask(cmd_header);
+	mask = engine->get_cmd_length_mask(cmd_header);
 	if (!mask)
 		return NULL;
 
@@ -848,12 +896,31 @@ static const struct drm_i915_reg_descriptor *
 find_reg(const struct drm_i915_reg_descriptor *table,
 	 int count, u32 addr)
 {
-	if (table) {
-		int i;
+	int i;
+
+	for (i = 0; i < count; i++) {
+		if (i915_mmio_reg_offset(table[i].addr) == addr)
+			return &table[i];
+	}
 
-		for (i = 0; i < count; i++) {
-			if (i915_mmio_reg_offset(table[i].addr) == addr)
-				return &table[i];
+	return NULL;
+}
+
+static const struct drm_i915_reg_descriptor *
+find_reg_in_tables(const struct drm_i915_reg_table *tables,
+		   int count, bool is_master, u32 addr)
+{
+	int i;
+	const struct drm_i915_reg_table *table;
+	const struct drm_i915_reg_descriptor *reg;
+
+	for (i = 0; i < count; i++) {
+		table = &tables[i];
+		if (!table->master || is_master) {
+			reg = find_reg(table->regs, table->num_regs,
+				       addr);
+			if (reg != NULL)
+				return reg;
 		}
 	}
 
@@ -963,18 +1030,18 @@ unpin_src:
  *
  * Return: true if the ring requires software command parsing
  */
-bool i915_needs_cmd_parser(struct intel_engine_cs *ring)
+bool i915_needs_cmd_parser(struct intel_engine_cs *engine)
 {
-	if (!ring->needs_cmd_parser)
+	if (!engine->needs_cmd_parser)
 		return false;
 
-	if (!USES_PPGTT(ring->dev))
+	if (!USES_PPGTT(engine->dev))
 		return false;
 
 	return (i915.enable_cmd_parser == 1);
 }
 
-static bool check_cmd(const struct intel_engine_cs *ring,
+static bool check_cmd(const struct intel_engine_cs *engine,
 		      const struct drm_i915_cmd_descriptor *desc,
 		      const u32 *cmd, u32 length,
 		      const bool is_master,
@@ -1004,17 +1071,14 @@ static bool check_cmd(const struct intel_engine_cs *ring,
 		     offset += step) {
 			const u32 reg_addr = cmd[offset] & desc->reg.mask;
 			const struct drm_i915_reg_descriptor *reg =
-				find_reg(ring->reg_table, ring->reg_count,
-					 reg_addr);
-
-			if (!reg && is_master)
-				reg = find_reg(ring->master_reg_table,
-					       ring->master_reg_count,
-					       reg_addr);
+				find_reg_in_tables(engine->reg_tables,
+						   engine->reg_table_count,
+						   is_master,
+						   reg_addr);
 
 			if (!reg) {
 				DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
-						 reg_addr, *cmd, ring->id);
+						 reg_addr, *cmd, engine->id);
 				return false;
 			}
 
@@ -1087,7 +1151,7 @@ static bool check_cmd(const struct intel_engine_cs *ring,
 						 *cmd,
 						 desc->bits[i].mask,
 						 desc->bits[i].expected,
-						 dword, ring->id);
+						 dword, engine->id);
 				return false;
 			}
 		}
@@ -1113,7 +1177,7 @@ static bool check_cmd(const struct intel_engine_cs *ring,
  * Return: non-zero if the parser finds violations or otherwise fails; -EACCES
  * if the batch appears legal but should use hardware parsing
  */
-int i915_parse_cmds(struct intel_engine_cs *ring,
+int i915_parse_cmds(struct intel_engine_cs *engine,
 		    struct drm_i915_gem_object *batch_obj,
 		    struct drm_i915_gem_object *shadow_batch_obj,
 		    u32 batch_start_offset,
@@ -1147,7 +1211,7 @@ int i915_parse_cmds(struct intel_engine_cs *ring,
 		if (*cmd == MI_BATCH_BUFFER_END)
 			break;
 
-		desc = find_cmd(ring, *cmd, &default_desc);
+		desc = find_cmd(engine, *cmd, &default_desc);
 		if (!desc) {
 			DRM_DEBUG_DRIVER("CMD: Unrecognized command: 0x%08X\n",
 					 *cmd);
@@ -1179,7 +1243,7 @@ int i915_parse_cmds(struct intel_engine_cs *ring,
 			break;
 		}
 
-		if (!check_cmd(ring, desc, cmd, length, is_master,
+		if (!check_cmd(engine, desc, cmd, length, is_master,
 			       &oacontrol_set)) {
 			ret = -EINVAL;
 			break;
@@ -1223,6 +1287,7 @@ int i915_cmd_parser_get_version(void)
 	 * 3. Allow access to the GPGPU_THREADS_DISPATCHED register.
 	 * 4. L3 atomic chicken bits of HSW_SCRATCH1 and HSW_ROW_CHICKEN3.
 	 * 5. GPGPU dispatch compute indirect registers.
+	 * 6. TIMESTAMP register and Haswell CS GPR registers
 	 */
-	return 5;
+	return 6;
 }
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index e3f4c725a1c6..32690332d441 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -89,27 +89,34 @@ static int i915_capabilities(struct seq_file *m, void *data)
 	return 0;
 }
 
-static const char *get_pin_flag(struct drm_i915_gem_object *obj)
+static const char get_active_flag(struct drm_i915_gem_object *obj)
 {
-	if (obj->pin_display)
-		return "p";
-	else
-		return " ";
+	return obj->active ? '*' : ' ';
 }
 
-static const char *get_tiling_flag(struct drm_i915_gem_object *obj)
+static const char get_pin_flag(struct drm_i915_gem_object *obj)
+{
+	return obj->pin_display ? 'p' : ' ';
+}
+
+static const char get_tiling_flag(struct drm_i915_gem_object *obj)
 {
 	switch (obj->tiling_mode) {
 	default:
-	case I915_TILING_NONE: return " ";
-	case I915_TILING_X: return "X";
-	case I915_TILING_Y: return "Y";
+	case I915_TILING_NONE: return ' ';
+	case I915_TILING_X: return 'X';
+	case I915_TILING_Y: return 'Y';
 	}
 }
 
-static inline const char *get_global_flag(struct drm_i915_gem_object *obj)
+static inline const char get_global_flag(struct drm_i915_gem_object *obj)
 {
-	return i915_gem_obj_to_ggtt(obj) ? "g" : " ";
+	return i915_gem_obj_to_ggtt(obj) ? 'g' : ' ';
+}
+
+static inline const char get_pin_mapped_flag(struct drm_i915_gem_object *obj)
+{
+	return obj->mapping ? 'M' : ' ';
 }
 
 static u64 i915_gem_obj_total_ggtt_size(struct drm_i915_gem_object *obj)
@@ -129,23 +136,26 @@ static void
 describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct i915_vma *vma;
 	int pin_count = 0;
-	int i;
+	enum intel_engine_id id;
+
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
 
-	seq_printf(m, "%pK: %s%s%s%s %8zdKiB %02x %02x [ ",
+	seq_printf(m, "%pK: %c%c%c%c%c %8zdKiB %02x %02x [ ",
 		   &obj->base,
-		   obj->active ? "*" : " ",
+		   get_active_flag(obj),
 		   get_pin_flag(obj),
 		   get_tiling_flag(obj),
 		   get_global_flag(obj),
+		   get_pin_mapped_flag(obj),
 		   obj->base.size / 1024,
 		   obj->base.read_domains,
 		   obj->base.write_domain);
-	for_each_ring(ring, dev_priv, i)
+	for_each_engine_id(engine, dev_priv, id)
 		seq_printf(m, "%x ",
-				i915_gem_request_get_seqno(obj->last_read_req[i]));
+				i915_gem_request_get_seqno(obj->last_read_req[id]));
 	seq_printf(m, "] %x %x%s%s%s",
 		   i915_gem_request_get_seqno(obj->last_write_req),
 		   i915_gem_request_get_seqno(obj->last_fenced_req),
@@ -184,7 +194,7 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 	}
 	if (obj->last_write_req != NULL)
 		seq_printf(m, " (%s)",
-			   i915_gem_request_get_ring(obj->last_write_req)->name);
+			   i915_gem_request_get_engine(obj->last_write_req)->name);
 	if (obj->frontbuffer_bits)
 		seq_printf(m, " (frontbuffer: 0x%03x)", obj->frontbuffer_bits);
 }
@@ -202,8 +212,8 @@ static int i915_gem_object_list_info(struct seq_file *m, void *data)
 	uintptr_t list = (uintptr_t) node->info_ent->data;
 	struct list_head *head;
 	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_address_space *vm = &dev_priv->gtt.base;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_vma *vma;
 	u64 total_obj_size, total_gtt_size;
 	int count, ret;
@@ -216,11 +226,11 @@ static int i915_gem_object_list_info(struct seq_file *m, void *data)
 	switch (list) {
 	case ACTIVE_LIST:
 		seq_puts(m, "Active:\n");
-		head = &vm->active_list;
+		head = &ggtt->base.active_list;
 		break;
 	case INACTIVE_LIST:
 		seq_puts(m, "Inactive:\n");
-		head = &vm->inactive_list;
+		head = &ggtt->base.inactive_list;
 		break;
 	default:
 		mutex_unlock(&dev->struct_mutex);
@@ -397,15 +407,15 @@ static void print_batch_pool_stats(struct seq_file *m,
 {
 	struct drm_i915_gem_object *obj;
 	struct file_stats stats;
-	struct intel_engine_cs *ring;
-	int i, j;
+	struct intel_engine_cs *engine;
+	int j;
 
 	memset(&stats, 0, sizeof(stats));
 
-	for_each_ring(ring, dev_priv, i) {
-		for (j = 0; j < ARRAY_SIZE(ring->batch_pool.cache_list); j++) {
+	for_each_engine(engine, dev_priv) {
+		for (j = 0; j < ARRAY_SIZE(engine->batch_pool.cache_list); j++) {
 			list_for_each_entry(obj,
-					    &ring->batch_pool.cache_list[j],
+					    &engine->batch_pool.cache_list[j],
 					    batch_pool_link)
 				per_file_stats(0, obj, &stats);
 		}
@@ -429,11 +439,13 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 {
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	u32 count, mappable_count, purgeable_count;
 	u64 size, mappable_size, purgeable_size;
+	unsigned long pin_mapped_count = 0, pin_mapped_purgeable_count = 0;
+	u64 pin_mapped_size = 0, pin_mapped_purgeable_size = 0;
 	struct drm_i915_gem_object *obj;
-	struct i915_address_space *vm = &dev_priv->gtt.base;
 	struct drm_file *file;
 	struct i915_vma *vma;
 	int ret;
@@ -452,12 +464,12 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 		   count, mappable_count, size, mappable_size);
 
 	size = count = mappable_size = mappable_count = 0;
-	count_vmas(&vm->active_list, vm_link);
+	count_vmas(&ggtt->base.active_list, vm_link);
 	seq_printf(m, "  %u [%u] active objects, %llu [%llu] bytes\n",
 		   count, mappable_count, size, mappable_size);
 
 	size = count = mappable_size = mappable_count = 0;
-	count_vmas(&vm->inactive_list, vm_link);
+	count_vmas(&ggtt->base.inactive_list, vm_link);
 	seq_printf(m, "  %u [%u] inactive objects, %llu [%llu] bytes\n",
 		   count, mappable_count, size, mappable_size);
 
@@ -466,6 +478,14 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 		size += obj->base.size, ++count;
 		if (obj->madv == I915_MADV_DONTNEED)
 			purgeable_size += obj->base.size, ++purgeable_count;
+		if (obj->mapping) {
+			pin_mapped_count++;
+			pin_mapped_size += obj->base.size;
+			if (obj->pages_pin_count == 0) {
+				pin_mapped_purgeable_count++;
+				pin_mapped_purgeable_size += obj->base.size;
+			}
+		}
 	}
 	seq_printf(m, "%u unbound objects, %llu bytes\n", count, size);
 
@@ -483,6 +503,14 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 			purgeable_size += obj->base.size;
 			++purgeable_count;
 		}
+		if (obj->mapping) {
+			pin_mapped_count++;
+			pin_mapped_size += obj->base.size;
+			if (obj->pages_pin_count == 0) {
+				pin_mapped_purgeable_count++;
+				pin_mapped_purgeable_size += obj->base.size;
+			}
+		}
 	}
 	seq_printf(m, "%u purgeable objects, %llu bytes\n",
 		   purgeable_count, purgeable_size);
@@ -490,13 +518,20 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 		   mappable_count, mappable_size);
 	seq_printf(m, "%u fault mappable objects, %llu bytes\n",
 		   count, size);
+	seq_printf(m,
+		   "%lu [%lu] pin mapped objects, %llu [%llu] bytes [purgeable]\n",
+		   pin_mapped_count, pin_mapped_purgeable_count,
+		   pin_mapped_size, pin_mapped_purgeable_size);
 
 	seq_printf(m, "%llu [%llu] gtt total\n",
-		   dev_priv->gtt.base.total,
-		   (u64)dev_priv->gtt.mappable_end - dev_priv->gtt.base.start);
+		   ggtt->base.total, ggtt->mappable_end - ggtt->base.start);
 
 	seq_putc(m, '\n');
 	print_batch_pool_stats(m, dev_priv);
+
+	mutex_unlock(&dev->struct_mutex);
+
+	mutex_lock(&dev->filelist_mutex);
 	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
 		struct file_stats stats;
 		struct task_struct *task;
@@ -517,8 +552,7 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 		print_file_stats(m, task ? task->comm : "<unknown>", stats);
 		rcu_read_unlock();
 	}
-
-	mutex_unlock(&dev->struct_mutex);
+	mutex_unlock(&dev->filelist_mutex);
 
 	return 0;
 }
@@ -591,14 +625,13 @@ static int i915_gem_pageflip_info(struct seq_file *m, void *data)
 					   pipe, plane);
 			}
 			if (work->flip_queued_req) {
-				struct intel_engine_cs *ring =
-					i915_gem_request_get_ring(work->flip_queued_req);
+				struct intel_engine_cs *engine = i915_gem_request_get_engine(work->flip_queued_req);
 
 				seq_printf(m, "Flip queued on %s at seqno %x, next seqno %x [current breadcrumb %x], completed? %d\n",
-					   ring->name,
+					   engine->name,
 					   i915_gem_request_get_seqno(work->flip_queued_req),
 					   dev_priv->next_seqno,
-					   ring->get_seqno(ring, true),
+					   engine->get_seqno(engine),
 					   i915_gem_request_completed(work->flip_queued_req, true));
 			} else
 				seq_printf(m, "Flip not associated with any ring\n");
@@ -637,28 +670,28 @@ static int i915_gem_batch_pool_info(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	int total = 0;
-	int ret, i, j;
+	int ret, j;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
 		return ret;
 
-	for_each_ring(ring, dev_priv, i) {
-		for (j = 0; j < ARRAY_SIZE(ring->batch_pool.cache_list); j++) {
+	for_each_engine(engine, dev_priv) {
+		for (j = 0; j < ARRAY_SIZE(engine->batch_pool.cache_list); j++) {
 			int count;
 
 			count = 0;
 			list_for_each_entry(obj,
-					    &ring->batch_pool.cache_list[j],
+					    &engine->batch_pool.cache_list[j],
 					    batch_pool_link)
 				count++;
 			seq_printf(m, "%s cache[%d]: %d objects\n",
-				   ring->name, j, count);
+				   engine->name, j, count);
 
 			list_for_each_entry(obj,
-					    &ring->batch_pool.cache_list[j],
+					    &engine->batch_pool.cache_list[j],
 					    batch_pool_link) {
 				seq_puts(m, "   ");
 				describe_obj(m, obj);
@@ -681,26 +714,26 @@ static int i915_gem_request_info(struct seq_file *m, void *data)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct drm_i915_gem_request *req;
-	int ret, any, i;
+	int ret, any;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
 		return ret;
 
 	any = 0;
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		int count;
 
 		count = 0;
-		list_for_each_entry(req, &ring->request_list, list)
+		list_for_each_entry(req, &engine->request_list, list)
 			count++;
 		if (count == 0)
 			continue;
 
-		seq_printf(m, "%s requests: %d\n", ring->name, count);
-		list_for_each_entry(req, &ring->request_list, list) {
+		seq_printf(m, "%s requests: %d\n", engine->name, count);
+		list_for_each_entry(req, &engine->request_list, list) {
 			struct task_struct *task;
 
 			rcu_read_lock();
@@ -726,12 +759,12 @@ static int i915_gem_request_info(struct seq_file *m, void *data)
 }
 
 static void i915_ring_seqno_info(struct seq_file *m,
-				 struct intel_engine_cs *ring)
+				 struct intel_engine_cs *engine)
 {
-	if (ring->get_seqno) {
-		seq_printf(m, "Current sequence (%s): %x\n",
-			   ring->name, ring->get_seqno(ring, false));
-	}
+	seq_printf(m, "Current sequence (%s): %x\n",
+		   engine->name, engine->get_seqno(engine));
+	seq_printf(m, "Current user interrupts (%s): %x\n",
+		   engine->name, READ_ONCE(engine->user_interrupts));
 }
 
 static int i915_gem_seqno_info(struct seq_file *m, void *data)
@@ -739,16 +772,16 @@ static int i915_gem_seqno_info(struct seq_file *m, void *data)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int ret, i;
+	struct intel_engine_cs *engine;
+	int ret;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
 		return ret;
 	intel_runtime_pm_get(dev_priv);
 
-	for_each_ring(ring, dev_priv, i)
-		i915_ring_seqno_info(m, ring);
+	for_each_engine(engine, dev_priv)
+		i915_ring_seqno_info(m, engine);
 
 	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
@@ -762,7 +795,7 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	int ret, i, pipe;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
@@ -934,13 +967,13 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 		seq_printf(m, "Graphics Interrupt mask:		%08x\n",
 			   I915_READ(GTIMR));
 	}
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		if (INTEL_INFO(dev)->gen >= 6) {
 			seq_printf(m,
 				   "Graphics Interrupt mask (%s):	%08x\n",
-				   ring->name, I915_READ_IMR(ring));
+				   engine->name, I915_READ_IMR(engine));
 		}
-		i915_ring_seqno_info(m, ring);
+		i915_ring_seqno_info(m, engine);
 	}
 	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
@@ -981,12 +1014,12 @@ static int i915_hws_info(struct seq_file *m, void *data)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	const u32 *hws;
 	int i;
 
-	ring = &dev_priv->ring[(uintptr_t)node->info_ent->data];
-	hws = ring->status_page.page_addr;
+	engine = &dev_priv->engine[(uintptr_t)node->info_ent->data];
+	hws = engine->status_page.page_addr;
 	if (hws == NULL)
 		return 0;
 
@@ -1216,12 +1249,12 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		rpdeclimit = I915_READ(GEN6_RP_DOWN_THRESHOLD);
 
 		rpstat = I915_READ(GEN6_RPSTAT1);
-		rpupei = I915_READ(GEN6_RP_CUR_UP_EI);
-		rpcurup = I915_READ(GEN6_RP_CUR_UP);
-		rpprevup = I915_READ(GEN6_RP_PREV_UP);
-		rpdownei = I915_READ(GEN6_RP_CUR_DOWN_EI);
-		rpcurdown = I915_READ(GEN6_RP_CUR_DOWN);
-		rpprevdown = I915_READ(GEN6_RP_PREV_DOWN);
+		rpupei = I915_READ(GEN6_RP_CUR_UP_EI) & GEN6_CURICONT_MASK;
+		rpcurup = I915_READ(GEN6_RP_CUR_UP) & GEN6_CURBSYTAVG_MASK;
+		rpprevup = I915_READ(GEN6_RP_PREV_UP) & GEN6_CURBSYTAVG_MASK;
+		rpdownei = I915_READ(GEN6_RP_CUR_DOWN_EI) & GEN6_CURIAVG_MASK;
+		rpcurdown = I915_READ(GEN6_RP_CUR_DOWN) & GEN6_CURBSYTAVG_MASK;
+		rpprevdown = I915_READ(GEN6_RP_PREV_DOWN) & GEN6_CURBSYTAVG_MASK;
 		if (IS_GEN9(dev))
 			cagf = (rpstat & GEN9_CAGF_MASK) >> GEN9_CAGF_SHIFT;
 		else if (IS_HASWELL(dev) || IS_BROADWELL(dev))
@@ -1261,21 +1294,21 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		seq_printf(m, "RPDECLIMIT: 0x%08x\n", rpdeclimit);
 		seq_printf(m, "RPNSWREQ: %dMHz\n", reqf);
 		seq_printf(m, "CAGF: %dMHz\n", cagf);
-		seq_printf(m, "RP CUR UP EI: %dus\n", rpupei &
-			   GEN6_CURICONT_MASK);
-		seq_printf(m, "RP CUR UP: %dus\n", rpcurup &
-			   GEN6_CURBSYTAVG_MASK);
-		seq_printf(m, "RP PREV UP: %dus\n", rpprevup &
-			   GEN6_CURBSYTAVG_MASK);
+		seq_printf(m, "RP CUR UP EI: %d (%dus)\n",
+			   rpupei, GT_PM_INTERVAL_TO_US(dev_priv, rpupei));
+		seq_printf(m, "RP CUR UP: %d (%dus)\n",
+			   rpcurup, GT_PM_INTERVAL_TO_US(dev_priv, rpcurup));
+		seq_printf(m, "RP PREV UP: %d (%dus)\n",
+			   rpprevup, GT_PM_INTERVAL_TO_US(dev_priv, rpprevup));
 		seq_printf(m, "Up threshold: %d%%\n",
 			   dev_priv->rps.up_threshold);
 
-		seq_printf(m, "RP CUR DOWN EI: %dus\n", rpdownei &
-			   GEN6_CURIAVG_MASK);
-		seq_printf(m, "RP CUR DOWN: %dus\n", rpcurdown &
-			   GEN6_CURBSYTAVG_MASK);
-		seq_printf(m, "RP PREV DOWN: %dus\n", rpprevdown &
-			   GEN6_CURBSYTAVG_MASK);
+		seq_printf(m, "RP CUR DOWN EI: %d (%dus)\n",
+			   rpdownei, GT_PM_INTERVAL_TO_US(dev_priv, rpdownei));
+		seq_printf(m, "RP CUR DOWN: %d (%dus)\n",
+			   rpcurdown, GT_PM_INTERVAL_TO_US(dev_priv, rpcurdown));
+		seq_printf(m, "RP PREV DOWN: %d (%dus)\n",
+			   rpprevdown, GT_PM_INTERVAL_TO_US(dev_priv, rpprevdown));
 		seq_printf(m, "Down threshold: %d%%\n",
 			   dev_priv->rps.down_threshold);
 
@@ -1331,11 +1364,12 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	u64 acthd[I915_NUM_RINGS];
-	u32 seqno[I915_NUM_RINGS];
+	struct intel_engine_cs *engine;
+	u64 acthd[I915_NUM_ENGINES];
+	u32 seqno[I915_NUM_ENGINES];
 	u32 instdone[I915_NUM_INSTDONE_REG];
-	int i, j;
+	enum intel_engine_id id;
+	int j;
 
 	if (!i915.enable_hangcheck) {
 		seq_printf(m, "Hangcheck disabled\n");
@@ -1344,9 +1378,9 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 
 	intel_runtime_pm_get(dev_priv);
 
-	for_each_ring(ring, dev_priv, i) {
-		seqno[i] = ring->get_seqno(ring, false);
-		acthd[i] = intel_ring_get_active_head(ring);
+	for_each_engine_id(engine, dev_priv, id) {
+		acthd[id] = intel_ring_get_active_head(engine);
+		seqno[id] = engine->get_seqno(engine);
 	}
 
 	i915_get_extra_instdone(dev, instdone);
@@ -1360,19 +1394,22 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 	} else
 		seq_printf(m, "Hangcheck inactive\n");
 
-	for_each_ring(ring, dev_priv, i) {
-		seq_printf(m, "%s:\n", ring->name);
-		seq_printf(m, "\tseqno = %x [current %x]\n",
-			   ring->hangcheck.seqno, seqno[i]);
+	for_each_engine_id(engine, dev_priv, id) {
+		seq_printf(m, "%s:\n", engine->name);
+		seq_printf(m, "\tseqno = %x [current %x, last %x]\n",
+			   engine->hangcheck.seqno,
+			   seqno[id],
+			   engine->last_submitted_seqno);
+		seq_printf(m, "\tuser interrupts = %x [current %x]\n",
+			   engine->hangcheck.user_interrupts,
+			   READ_ONCE(engine->user_interrupts));
 		seq_printf(m, "\tACTHD = 0x%08llx [current 0x%08llx]\n",
-			   (long long)ring->hangcheck.acthd,
-			   (long long)acthd[i]);
-		seq_printf(m, "\tmax ACTHD = 0x%08llx\n",
-			   (long long)ring->hangcheck.max_acthd);
-		seq_printf(m, "\tscore = %d\n", ring->hangcheck.score);
-		seq_printf(m, "\taction = %d\n", ring->hangcheck.action);
-
-		if (ring->id == RCS) {
+			   (long long)engine->hangcheck.acthd,
+			   (long long)acthd[id]);
+		seq_printf(m, "\tscore = %d\n", engine->hangcheck.score);
+		seq_printf(m, "\taction = %d\n", engine->hangcheck.action);
+
+		if (engine->id == RCS) {
 			seq_puts(m, "\tinstdone read =");
 
 			for (j = 0; j < I915_NUM_INSTDONE_REG; j++)
@@ -1382,7 +1419,7 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 
 			for (j = 0; j < I915_NUM_INSTDONE_REG; j++)
 				seq_printf(m, " 0x%08x",
-					   ring->hangcheck.instdone[j]);
+					   engine->hangcheck.instdone[j]);
 
 			seq_puts(m, "\n");
 		}
@@ -1465,12 +1502,11 @@ static int i915_forcewake_domains(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_uncore_forcewake_domain *fw_domain;
-	int i;
 
 	spin_lock_irq(&dev_priv->uncore.lock);
-	for_each_fw_domain(fw_domain, dev_priv, i) {
+	for_each_fw_domain(fw_domain, dev_priv) {
 		seq_printf(m, "%s.wake_count = %u\n",
-			   intel_uncore_forcewake_domain_to_str(i),
+			   intel_uncore_forcewake_domain_to_str(fw_domain->id),
 			   fw_domain->wake_count);
 	}
 	spin_unlock_irq(&dev_priv->uncore.lock);
@@ -1897,6 +1933,11 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	struct intel_framebuffer *fbdev_fb = NULL;
 	struct drm_framebuffer *drm_fb;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
 
 #ifdef CONFIG_DRM_FBDEV_EMULATION
        if (to_i915(dev)->fbdev) {
@@ -1908,7 +1949,7 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
                          fbdev_fb->base.depth,
                          fbdev_fb->base.bits_per_pixel,
                          fbdev_fb->base.modifier[0],
-                         atomic_read(&fbdev_fb->base.refcount.refcount));
+                         drm_framebuffer_read_refcount(&fbdev_fb->base));
                describe_obj(m, fbdev_fb->obj);
                seq_putc(m, '\n');
        }
@@ -1926,11 +1967,12 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 			   fb->base.depth,
 			   fb->base.bits_per_pixel,
 			   fb->base.modifier[0],
-			   atomic_read(&fb->base.refcount.refcount));
+			   drm_framebuffer_read_refcount(&fb->base));
 		describe_obj(m, fb->obj);
 		seq_putc(m, '\n');
 	}
 	mutex_unlock(&dev->mode_config.fb_lock);
+	mutex_unlock(&dev->struct_mutex);
 
 	return 0;
 }
@@ -1948,9 +1990,10 @@ static int i915_context_status(struct seq_file *m, void *unused)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct intel_context *ctx;
-	int ret, i;
+	enum intel_engine_id id;
+	int ret;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
@@ -1968,13 +2011,13 @@ static int i915_context_status(struct seq_file *m, void *unused)
 
 		if (i915.enable_execlists) {
 			seq_putc(m, '\n');
-			for_each_ring(ring, dev_priv, i) {
+			for_each_engine_id(engine, dev_priv, id) {
 				struct drm_i915_gem_object *ctx_obj =
-					ctx->engine[i].state;
+					ctx->engine[id].state;
 				struct intel_ringbuffer *ringbuf =
-					ctx->engine[i].ringbuf;
+					ctx->engine[id].ringbuf;
 
-				seq_printf(m, "%s: ", ring->name);
+				seq_printf(m, "%s: ", engine->name);
 				if (ctx_obj)
 					describe_obj(m, ctx_obj);
 				if (ringbuf)
@@ -1995,22 +2038,22 @@ static int i915_context_status(struct seq_file *m, void *unused)
 
 static void i915_dump_lrc_obj(struct seq_file *m,
 			      struct intel_context *ctx,
-			      struct intel_engine_cs *ring)
+			      struct intel_engine_cs *engine)
 {
 	struct page *page;
 	uint32_t *reg_state;
 	int j;
-	struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
+	struct drm_i915_gem_object *ctx_obj = ctx->engine[engine->id].state;
 	unsigned long ggtt_offset = 0;
 
 	if (ctx_obj == NULL) {
 		seq_printf(m, "Context on %s with no gem object\n",
-			   ring->name);
+			   engine->name);
 		return;
 	}
 
-	seq_printf(m, "CONTEXT: %s %u\n", ring->name,
-		   intel_execlists_ctx_id(ctx, ring));
+	seq_printf(m, "CONTEXT: %s %u\n", engine->name,
+		   intel_execlists_ctx_id(ctx, engine));
 
 	if (!i915_gem_obj_ggtt_bound(ctx_obj))
 		seq_puts(m, "\tNot bound in GGTT\n");
@@ -2043,9 +2086,9 @@ static int i915_dump_lrc(struct seq_file *m, void *unused)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct intel_context *ctx;
-	int ret, i;
+	int ret;
 
 	if (!i915.enable_execlists) {
 		seq_printf(m, "Logical Ring Contexts are disabled\n");
@@ -2058,8 +2101,8 @@ static int i915_dump_lrc(struct seq_file *m, void *unused)
 
 	list_for_each_entry(ctx, &dev_priv->context_list, link)
 		if (ctx != dev_priv->kernel_context)
-			for_each_ring(ring, dev_priv, i)
-				i915_dump_lrc_obj(m, ctx, ring);
+			for_each_engine(engine, dev_priv)
+				i915_dump_lrc_obj(m, ctx, engine);
 
 	mutex_unlock(&dev->struct_mutex);
 
@@ -2071,15 +2114,14 @@ static int i915_execlists(struct seq_file *m, void *data)
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	u32 status_pointer;
 	u8 read_pointer;
 	u8 write_pointer;
 	u32 status;
 	u32 ctx_id;
 	struct list_head *cursor;
-	int ring_id, i;
-	int ret;
+	int i, ret;
 
 	if (!i915.enable_execlists) {
 		seq_puts(m, "Logical Ring Contexts are disabled\n");
@@ -2092,22 +2134,21 @@ static int i915_execlists(struct seq_file *m, void *data)
 
 	intel_runtime_pm_get(dev_priv);
 
-	for_each_ring(ring, dev_priv, ring_id) {
+	for_each_engine(engine, dev_priv) {
 		struct drm_i915_gem_request *head_req = NULL;
 		int count = 0;
-		unsigned long flags;
 
-		seq_printf(m, "%s\n", ring->name);
+		seq_printf(m, "%s\n", engine->name);
 
-		status = I915_READ(RING_EXECLIST_STATUS_LO(ring));
-		ctx_id = I915_READ(RING_EXECLIST_STATUS_HI(ring));
+		status = I915_READ(RING_EXECLIST_STATUS_LO(engine));
+		ctx_id = I915_READ(RING_EXECLIST_STATUS_HI(engine));
 		seq_printf(m, "\tExeclist status: 0x%08X, context: %u\n",
 			   status, ctx_id);
 
-		status_pointer = I915_READ(RING_CONTEXT_STATUS_PTR(ring));
+		status_pointer = I915_READ(RING_CONTEXT_STATUS_PTR(engine));
 		seq_printf(m, "\tStatus pointer: 0x%08X\n", status_pointer);
 
-		read_pointer = ring->next_context_status_buffer;
+		read_pointer = engine->next_context_status_buffer;
 		write_pointer = GEN8_CSB_WRITE_PTR(status_pointer);
 		if (read_pointer > write_pointer)
 			write_pointer += GEN8_CSB_ENTRIES;
@@ -2115,24 +2156,25 @@ static int i915_execlists(struct seq_file *m, void *data)
 			   read_pointer, write_pointer);
 
 		for (i = 0; i < GEN8_CSB_ENTRIES; i++) {
-			status = I915_READ(RING_CONTEXT_STATUS_BUF_LO(ring, i));
-			ctx_id = I915_READ(RING_CONTEXT_STATUS_BUF_HI(ring, i));
+			status = I915_READ(RING_CONTEXT_STATUS_BUF_LO(engine, i));
+			ctx_id = I915_READ(RING_CONTEXT_STATUS_BUF_HI(engine, i));
 
 			seq_printf(m, "\tStatus buffer %d: 0x%08X, context: %u\n",
 				   i, status, ctx_id);
 		}
 
-		spin_lock_irqsave(&ring->execlist_lock, flags);
-		list_for_each(cursor, &ring->execlist_queue)
+		spin_lock_bh(&engine->execlist_lock);
+		list_for_each(cursor, &engine->execlist_queue)
 			count++;
-		head_req = list_first_entry_or_null(&ring->execlist_queue,
-				struct drm_i915_gem_request, execlist_link);
-		spin_unlock_irqrestore(&ring->execlist_lock, flags);
+		head_req = list_first_entry_or_null(&engine->execlist_queue,
+						    struct drm_i915_gem_request,
+						    execlist_link);
+		spin_unlock_bh(&engine->execlist_lock);
 
 		seq_printf(m, "\t%d requests in queue\n", count);
 		if (head_req) {
 			seq_printf(m, "\tHead request id: %u\n",
-				   intel_execlists_ctx_id(head_req->ctx, ring));
+				   intel_execlists_ctx_id(head_req->ctx, engine));
 			seq_printf(m, "\tHead request tail: %u\n",
 				   head_req->tail);
 		}
@@ -2248,19 +2290,19 @@ static int per_file_ctx(int id, void *ptr, void *data)
 static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
-	int unused, i;
+	int i;
 
 	if (!ppgtt)
 		return;
 
-	for_each_ring(ring, dev_priv, unused) {
-		seq_printf(m, "%s\n", ring->name);
+	for_each_engine(engine, dev_priv) {
+		seq_printf(m, "%s\n", engine->name);
 		for (i = 0; i < 4; i++) {
-			u64 pdp = I915_READ(GEN8_RING_PDP_UDW(ring, i));
+			u64 pdp = I915_READ(GEN8_RING_PDP_UDW(engine, i));
 			pdp <<= 32;
-			pdp |= I915_READ(GEN8_RING_PDP_LDW(ring, i));
+			pdp |= I915_READ(GEN8_RING_PDP_LDW(engine, i));
 			seq_printf(m, "\tPDP%d 0x%016llx\n", i, pdp);
 		}
 	}
@@ -2269,19 +2311,22 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
 	if (INTEL_INFO(dev)->gen == 6)
 		seq_printf(m, "GFX_MODE: 0x%08x\n", I915_READ(GFX_MODE));
 
-	for_each_ring(ring, dev_priv, i) {
-		seq_printf(m, "%s\n", ring->name);
+	for_each_engine(engine, dev_priv) {
+		seq_printf(m, "%s\n", engine->name);
 		if (INTEL_INFO(dev)->gen == 7)
-			seq_printf(m, "GFX_MODE: 0x%08x\n", I915_READ(RING_MODE_GEN7(ring)));
-		seq_printf(m, "PP_DIR_BASE: 0x%08x\n", I915_READ(RING_PP_DIR_BASE(ring)));
-		seq_printf(m, "PP_DIR_BASE_READ: 0x%08x\n", I915_READ(RING_PP_DIR_BASE_READ(ring)));
-		seq_printf(m, "PP_DIR_DCLV: 0x%08x\n", I915_READ(RING_PP_DIR_DCLV(ring)));
+			seq_printf(m, "GFX_MODE: 0x%08x\n",
+				   I915_READ(RING_MODE_GEN7(engine)));
+		seq_printf(m, "PP_DIR_BASE: 0x%08x\n",
+			   I915_READ(RING_PP_DIR_BASE(engine)));
+		seq_printf(m, "PP_DIR_BASE_READ: 0x%08x\n",
+			   I915_READ(RING_PP_DIR_BASE_READ(engine)));
+		seq_printf(m, "PP_DIR_DCLV: 0x%08x\n",
+			   I915_READ(RING_PP_DIR_DCLV(engine)));
 	}
 	if (dev_priv->mm.aliasing_ppgtt) {
 		struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
@@ -2312,6 +2357,7 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 	else if (INTEL_INFO(dev)->gen >= 6)
 		gen6_ppgtt_info(m, dev);
 
+	mutex_lock(&dev->filelist_mutex);
 	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
 		struct drm_i915_file_private *file_priv = file->driver_priv;
 		struct task_struct *task;
@@ -2326,6 +2372,7 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 		idr_for_each(&file_priv->context_idr, per_file_ctx,
 			     (void *)(unsigned long)m);
 	}
+	mutex_unlock(&dev->filelist_mutex);
 
 out_put:
 	intel_runtime_pm_put(dev_priv);
@@ -2336,12 +2383,11 @@ out_put:
 
 static int count_irq_waiters(struct drm_i915_private *i915)
 {
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	int count = 0;
-	int i;
 
-	for_each_ring(ring, i915, i)
-		count += ring->irq_refcount;
+	for_each_engine(engine, i915)
+		count += engine->irq_refcount;
 
 	return count;
 }
@@ -2362,6 +2408,8 @@ static int i915_rps_boost_info(struct seq_file *m, void *data)
 		   intel_gpu_freq(dev_priv, dev_priv->rps.min_freq_softlimit),
 		   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq_softlimit),
 		   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq));
+
+	mutex_lock(&dev->filelist_mutex);
 	spin_lock(&dev_priv->rps.client_lock);
 	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
 		struct drm_i915_file_private *file_priv = file->driver_priv;
@@ -2384,6 +2432,7 @@ static int i915_rps_boost_info(struct seq_file *m, void *data)
 		   list_empty(&dev_priv->rps.mmioflips.link) ? "" : ", active");
 	seq_printf(m, "Kernel boosts: %d\n", dev_priv->rps.boosts);
 	spin_unlock(&dev_priv->rps.client_lock);
+	mutex_unlock(&dev->filelist_mutex);
 
 	return 0;
 }
@@ -2393,10 +2442,11 @@ static int i915_llc(struct seq_file *m, void *data)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	const bool edram = INTEL_GEN(dev_priv) > 8;
 
-	/* Size calculation for LLC is a bit of a pain. Ignore for now. */
 	seq_printf(m, "LLC: %s\n", yesno(HAS_LLC(dev)));
-	seq_printf(m, "eLLC: %zuMB\n", dev_priv->ellc_size);
+	seq_printf(m, "%s: %lluMB\n", edram ? "eDRAM" : "eLLC",
+		   intel_uncore_edram_size(dev_priv)/1024/1024);
 
 	return 0;
 }
@@ -2408,7 +2458,7 @@ static int i915_guc_load_status_info(struct seq_file *m, void *data)
 	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
 	u32 tmp, i;
 
-	if (!HAS_GUC_UCODE(dev_priv->dev))
+	if (!HAS_GUC_UCODE(dev_priv))
 		return 0;
 
 	seq_printf(m, "GuC firmware status:\n");
@@ -2449,9 +2499,8 @@ static void i915_guc_client_info(struct seq_file *m,
 				 struct drm_i915_private *dev_priv,
 				 struct i915_guc_client *client)
 {
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	uint64_t tot = 0;
-	uint32_t i;
 
 	seq_printf(m, "\tPriority %d, GuC ctx index: %u, PD offset 0x%x\n",
 		client->priority, client->ctx_index, client->proc_desc_offset);
@@ -2464,11 +2513,11 @@ static void i915_guc_client_info(struct seq_file *m,
 	seq_printf(m, "\tFailed doorbell: %u\n", client->b_fail);
 	seq_printf(m, "\tLast submission result: %d\n", client->retcode);
 
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		seq_printf(m, "\tSubmissions: %llu %s\n",
-				client->submissions[ring->guc_id],
-				ring->name);
-		tot += client->submissions[ring->guc_id];
+				client->submissions[engine->guc_id],
+				engine->name);
+		tot += client->submissions[engine->guc_id];
 	}
 	seq_printf(m, "\tTotal: %llu\n", tot);
 }
@@ -2480,11 +2529,10 @@ static int i915_guc_info(struct seq_file *m, void *data)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_guc guc;
 	struct i915_guc_client client = {};
-	struct intel_engine_cs *ring;
-	enum intel_ring_id i;
+	struct intel_engine_cs *engine;
 	u64 total = 0;
 
-	if (!HAS_GUC_SCHED(dev_priv->dev))
+	if (!HAS_GUC_SCHED(dev_priv))
 		return 0;
 
 	if (mutex_lock_interruptible(&dev->struct_mutex))
@@ -2504,11 +2552,11 @@ static int i915_guc_info(struct seq_file *m, void *data)
 	seq_printf(m, "GuC last action error code: %d\n", guc.action_err);
 
 	seq_printf(m, "\nGuC submissions:\n");
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		seq_printf(m, "\t%-24s: %10llu, last seqno 0x%08x\n",
-			ring->name, guc.submissions[ring->guc_id],
-			guc.last_seqno[ring->guc_id]);
-		total += guc.submissions[ring->guc_id];
+			engine->name, guc.submissions[engine->guc_id],
+			guc.last_seqno[engine->guc_id]);
+		total += guc.submissions[engine->guc_id];
 	}
 	seq_printf(m, "\t%s: %llu\n", "Total", total);
 
@@ -2688,10 +2736,8 @@ static int i915_runtime_pm_status(struct seq_file *m, void *unused)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (!HAS_RUNTIME_PM(dev)) {
-		seq_puts(m, "not supported\n");
-		return 0;
-	}
+	if (!HAS_RUNTIME_PM(dev_priv))
+		seq_puts(m, "Runtime power management not supported\n");
 
 	seq_printf(m, "GPU idle: %s\n", yesno(!dev_priv->mm.busy));
 	seq_printf(m, "IRQs disabled: %s\n",
@@ -2702,6 +2748,9 @@ static int i915_runtime_pm_status(struct seq_file *m, void *unused)
 #else
 	seq_printf(m, "Device Power Management (CONFIG_PM) disabled\n");
 #endif
+	seq_printf(m, "PCI device power state: %s [%d]\n",
+		   pci_power_name(dev_priv->dev->pdev->current_state),
+		   dev_priv->dev->pdev->current_state);
 
 	return 0;
 }
@@ -3114,9 +3163,10 @@ static int i915_semaphore_status(struct seq_file *m, void *unused)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	int num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
-	int i, j, ret;
+	enum intel_engine_id id;
+	int j, ret;
 
 	if (!i915_semaphore_is_enabled(dev)) {
 		seq_puts(m, "Semaphores are disabled\n");
@@ -3135,14 +3185,14 @@ static int i915_semaphore_status(struct seq_file *m, void *unused)
 		page = i915_gem_object_get_page(dev_priv->semaphore_obj, 0);
 
 		seqno = (uint64_t *)kmap_atomic(page);
-		for_each_ring(ring, dev_priv, i) {
+		for_each_engine_id(engine, dev_priv, id) {
 			uint64_t offset;
 
-			seq_printf(m, "%s\n", ring->name);
+			seq_printf(m, "%s\n", engine->name);
 
 			seq_puts(m, "  Last signal:");
 			for (j = 0; j < num_rings; j++) {
-				offset = i * I915_NUM_RINGS + j;
+				offset = id * I915_NUM_ENGINES + j;
 				seq_printf(m, "0x%08llx (0x%02llx) ",
 					   seqno[offset], offset * 8);
 			}
@@ -3150,7 +3200,7 @@ static int i915_semaphore_status(struct seq_file *m, void *unused)
 
 			seq_puts(m, "  Last wait:  ");
 			for (j = 0; j < num_rings; j++) {
-				offset = i + (j * I915_NUM_RINGS);
+				offset = id + (j * I915_NUM_ENGINES);
 				seq_printf(m, "0x%08llx (0x%02llx) ",
 					   seqno[offset], offset * 8);
 			}
@@ -3160,18 +3210,18 @@ static int i915_semaphore_status(struct seq_file *m, void *unused)
 		kunmap_atomic(seqno);
 	} else {
 		seq_puts(m, "  Last signal:");
-		for_each_ring(ring, dev_priv, i)
+		for_each_engine(engine, dev_priv)
 			for (j = 0; j < num_rings; j++)
 				seq_printf(m, "0x%08x\n",
-					   I915_READ(ring->semaphore.mbox.signal[j]));
+					   I915_READ(engine->semaphore.mbox.signal[j]));
 		seq_putc(m, '\n');
 	}
 
 	seq_puts(m, "\nSync seqno:\n");
-	for_each_ring(ring, dev_priv, i) {
-		for (j = 0; j < num_rings; j++) {
-			seq_printf(m, "  0x%08x ", ring->semaphore.sync_seqno[j]);
-		}
+	for_each_engine(engine, dev_priv) {
+		for (j = 0; j < num_rings; j++)
+			seq_printf(m, "  0x%08x ",
+				   engine->semaphore.sync_seqno[j]);
 		seq_putc(m, '\n');
 	}
 	seq_putc(m, '\n');
@@ -3193,8 +3243,8 @@ static int i915_shared_dplls_info(struct seq_file *m, void *unused)
 		struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i];
 
 		seq_printf(m, "DPLL%i: %s, id: %i\n", i, pll->name, pll->id);
-		seq_printf(m, " crtc_mask: 0x%08x, active: %d, on: %s\n",
-			   pll->config.crtc_mask, pll->active, yesno(pll->on));
+		seq_printf(m, " crtc_mask: 0x%08x, active: 0x%x, on: %s\n",
+			   pll->config.crtc_mask, pll->active_mask, yesno(pll->on));
 		seq_printf(m, " tracked hardware state:\n");
 		seq_printf(m, " dpll:    0x%08x\n", pll->config.hw_state.dpll);
 		seq_printf(m, " dpll_md: 0x%08x\n",
@@ -3212,11 +3262,12 @@ static int i915_wa_registers(struct seq_file *m, void *unused)
 {
 	int i;
 	int ret;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_workarounds *workarounds = &dev_priv->workarounds;
+	enum intel_engine_id id;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
@@ -3225,9 +3276,9 @@ static int i915_wa_registers(struct seq_file *m, void *unused)
 	intel_runtime_pm_get(dev_priv);
 
 	seq_printf(m, "Workarounds applied: %d\n", workarounds->count);
-	for_each_ring(ring, dev_priv, i)
+	for_each_engine_id(engine, dev_priv, id)
 		seq_printf(m, "HW whitelist count for %s: %d\n",
-			   ring->name, workarounds->hw_whitelist_count[i]);
+			   engine->name, workarounds->hw_whitelist_count[id]);
 	for (i = 0; i < workarounds->count; ++i) {
 		i915_reg_t addr;
 		u32 mask, value, read;
@@ -3417,7 +3468,8 @@ static int i915_dp_mst_info(struct seq_file *m, void *unused)
 		intel_dig_port = enc_to_dig_port(encoder);
 		if (!intel_dig_port->dp.can_mst)
 			continue;
-
+		seq_printf(m, "MST Source Port %c\n",
+			   port_name(intel_dig_port->port));
 		drm_dp_mst_dump_topology(m, &intel_dig_port->dp.mst_mgr);
 	}
 	drm_modeset_unlock_all(dev);
@@ -4693,7 +4745,7 @@ i915_wedged_get(void *data, u64 *val)
 	struct drm_device *dev = data;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	*val = atomic_read(&dev_priv->gpu_error.reset_counter);
+	*val = i915_terminally_wedged(&dev_priv->gpu_error);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 1c6d227aae7c..15615fb9bde6 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -50,6 +50,66 @@
 #include <linux/pm_runtime.h>
 #include <linux/oom.h>
 
+static unsigned int i915_load_fail_count;
+
+bool __i915_inject_load_failure(const char *func, int line)
+{
+	if (i915_load_fail_count >= i915.inject_load_failure)
+		return false;
+
+	if (++i915_load_fail_count == i915.inject_load_failure) {
+		DRM_INFO("Injecting failure at checkpoint %u [%s:%d]\n",
+			 i915.inject_load_failure, func, line);
+		return true;
+	}
+
+	return false;
+}
+
+#define FDO_BUG_URL "https://bugs.freedesktop.org/enter_bug.cgi?product=DRI"
+#define FDO_BUG_MSG "Please file a bug at " FDO_BUG_URL " against DRM/Intel " \
+		    "providing the dmesg log by booting with drm.debug=0xf"
+
+void
+__i915_printk(struct drm_i915_private *dev_priv, const char *level,
+	      const char *fmt, ...)
+{
+	static bool shown_bug_once;
+	struct device *dev = dev_priv->dev->dev;
+	bool is_error = level[1] <= KERN_ERR[1];
+	bool is_debug = level[1] == KERN_DEBUG[1];
+	struct va_format vaf;
+	va_list args;
+
+	if (is_debug && !(drm_debug & DRM_UT_DRIVER))
+		return;
+
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	dev_printk(level, dev, "[" DRM_NAME ":%ps] %pV",
+		   __builtin_return_address(0), &vaf);
+
+	if (is_error && !shown_bug_once) {
+		dev_notice(dev, "%s", FDO_BUG_MSG);
+		shown_bug_once = true;
+	}
+
+	va_end(args);
+}
+
+static bool i915_error_injected(struct drm_i915_private *dev_priv)
+{
+	return i915.inject_load_failure &&
+	       i915_load_fail_count == i915.inject_load_failure;
+}
+
+#define i915_load_error(dev_priv, fmt, ...)				     \
+	__i915_printk(dev_priv,						     \
+		      i915_error_injected(dev_priv) ? KERN_DEBUG : KERN_ERR, \
+		      fmt, ##__VA_ARGS__)
 
 static int i915_getparam(struct drm_device *dev, void *data,
 			 struct drm_file *file_priv)
@@ -87,16 +147,16 @@ static int i915_getparam(struct drm_device *dev, void *data,
 		value = 1;
 		break;
 	case I915_PARAM_HAS_BSD:
-		value = intel_ring_initialized(&dev_priv->ring[VCS]);
+		value = intel_engine_initialized(&dev_priv->engine[VCS]);
 		break;
 	case I915_PARAM_HAS_BLT:
-		value = intel_ring_initialized(&dev_priv->ring[BCS]);
+		value = intel_engine_initialized(&dev_priv->engine[BCS]);
 		break;
 	case I915_PARAM_HAS_VEBOX:
-		value = intel_ring_initialized(&dev_priv->ring[VECS]);
+		value = intel_engine_initialized(&dev_priv->engine[VECS]);
 		break;
 	case I915_PARAM_HAS_BSD2:
-		value = intel_ring_initialized(&dev_priv->ring[VCS2]);
+		value = intel_engine_initialized(&dev_priv->engine[VCS2]);
 		break;
 	case I915_PARAM_HAS_RELAXED_FENCING:
 		value = 1;
@@ -197,13 +257,6 @@ static int i915_get_bridge_dev(struct drm_device *dev)
 	return 0;
 }
 
-#define MCHBAR_I915 0x44
-#define MCHBAR_I965 0x48
-#define MCHBAR_SIZE (4*4096)
-
-#define DEVEN_REG 0x54
-#define   DEVEN_MCHBAR_EN (1 << 28)
-
 /* Allocate space for the MCH regs if needed, return nonzero on error */
 static int
 intel_alloc_mchbar_resource(struct drm_device *dev)
@@ -265,7 +318,7 @@ intel_setup_mchbar(struct drm_device *dev)
 	dev_priv->mchbar_need_disable = false;
 
 	if (IS_I915G(dev) || IS_I915GM(dev)) {
-		pci_read_config_dword(dev_priv->bridge_dev, DEVEN_REG, &temp);
+		pci_read_config_dword(dev_priv->bridge_dev, DEVEN, &temp);
 		enabled = !!(temp & DEVEN_MCHBAR_EN);
 	} else {
 		pci_read_config_dword(dev_priv->bridge_dev, mchbar_reg, &temp);
@@ -283,7 +336,7 @@ intel_setup_mchbar(struct drm_device *dev)
 
 	/* Space is allocated or reserved, so enable it. */
 	if (IS_I915G(dev) || IS_I915GM(dev)) {
-		pci_write_config_dword(dev_priv->bridge_dev, DEVEN_REG,
+		pci_write_config_dword(dev_priv->bridge_dev, DEVEN,
 				       temp | DEVEN_MCHBAR_EN);
 	} else {
 		pci_read_config_dword(dev_priv->bridge_dev, mchbar_reg, &temp);
@@ -296,17 +349,24 @@ intel_teardown_mchbar(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int mchbar_reg = INTEL_INFO(dev)->gen >= 4 ? MCHBAR_I965 : MCHBAR_I915;
-	u32 temp;
 
 	if (dev_priv->mchbar_need_disable) {
 		if (IS_I915G(dev) || IS_I915GM(dev)) {
-			pci_read_config_dword(dev_priv->bridge_dev, DEVEN_REG, &temp);
-			temp &= ~DEVEN_MCHBAR_EN;
-			pci_write_config_dword(dev_priv->bridge_dev, DEVEN_REG, temp);
+			u32 deven_val;
+
+			pci_read_config_dword(dev_priv->bridge_dev, DEVEN,
+					      &deven_val);
+			deven_val &= ~DEVEN_MCHBAR_EN;
+			pci_write_config_dword(dev_priv->bridge_dev, DEVEN,
+					       deven_val);
 		} else {
-			pci_read_config_dword(dev_priv->bridge_dev, mchbar_reg, &temp);
-			temp &= ~1;
-			pci_write_config_dword(dev_priv->bridge_dev, mchbar_reg, temp);
+			u32 mchbar_val;
+
+			pci_read_config_dword(dev_priv->bridge_dev, mchbar_reg,
+					      &mchbar_val);
+			mchbar_val &= ~1;
+			pci_write_config_dword(dev_priv->bridge_dev, mchbar_reg,
+					       mchbar_val);
 		}
 	}
 
@@ -370,6 +430,9 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
+	if (i915_inject_load_failure())
+		return -ENODEV;
+
 	ret = intel_bios_init(dev_priv);
 	if (ret)
 		DRM_INFO("failed to find VBIOS tables\n");
@@ -413,9 +476,6 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	intel_modeset_gem_init(dev);
 
-	/* Always safe in the mode setting case. */
-	/* FIXME: do pre/post-mode set stuff in core KMS code */
-	dev->vblank_disable_allowed = true;
 	if (INTEL_INFO(dev)->num_pipes == 0)
 		return 0;
 
@@ -444,7 +504,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 cleanup_gem:
 	mutex_lock(&dev->struct_mutex);
-	i915_gem_cleanup_ringbuffer(dev);
+	i915_gem_cleanup_engines(dev);
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
 cleanup_irq:
@@ -453,6 +513,7 @@ cleanup_irq:
 	intel_teardown_gmbus(dev);
 cleanup_csr:
 	intel_csr_ucode_fini(dev_priv);
+	intel_power_domains_fini(dev_priv);
 	vga_switcheroo_unregister_client(dev->pdev);
 cleanup_vga_client:
 	vga_client_register(dev->pdev, NULL, NULL, NULL);
@@ -465,6 +526,7 @@ static int i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
 {
 	struct apertures_struct *ap;
 	struct pci_dev *pdev = dev_priv->dev->pdev;
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	bool primary;
 	int ret;
 
@@ -472,8 +534,8 @@ static int i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
 	if (!ap)
 		return -ENOMEM;
 
-	ap->ranges[0].base = dev_priv->gtt.mappable_base;
-	ap->ranges[0].size = dev_priv->gtt.mappable_end;
+	ap->ranges[0].base = ggtt->mappable_base;
+	ap->ranges[0].size = ggtt->mappable_end;
 
 	primary =
 		pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW;
@@ -853,6 +915,10 @@ static void intel_device_info_runtime_init(struct drm_device *dev)
 	else if (INTEL_INFO(dev)->gen >= 9)
 		gen9_sseu_info_init(dev);
 
+	/* Snooping is broken on BXT A stepping. */
+	info->has_snoop = !info->has_llc;
+	info->has_snoop &= !IS_BXT_REVID(dev, 0, BXT_REVID_A1);
+
 	DRM_DEBUG_DRIVER("slice total: %u\n", info->slice_total);
 	DRM_DEBUG_DRIVER("subslice total: %u\n", info->subslice_total);
 	DRM_DEBUG_DRIVER("subslice per slice: %u\n", info->subslice_per_slice);
@@ -929,6 +995,84 @@ static void i915_workqueues_cleanup(struct drm_i915_private *dev_priv)
 	destroy_workqueue(dev_priv->wq);
 }
 
+/**
+ * i915_driver_init_early - setup state not requiring device access
+ * @dev_priv: device private
+ *
+ * Initialize everything that is a "SW-only" state, that is state not
+ * requiring accessing the device or exposing the driver via kernel internal
+ * or userspace interfaces. Example steps belonging here: lock initialization,
+ * system memory allocation, setting up device specific attributes and
+ * function hooks not requiring accessing the device.
+ */
+static int i915_driver_init_early(struct drm_i915_private *dev_priv,
+				  struct drm_device *dev,
+				  struct intel_device_info *info)
+{
+	struct intel_device_info *device_info;
+	int ret = 0;
+
+	if (i915_inject_load_failure())
+		return -ENODEV;
+
+	/* Setup the write-once "constant" device info */
+	device_info = (struct intel_device_info *)&dev_priv->info;
+	memcpy(device_info, info, sizeof(dev_priv->info));
+	device_info->device_id = dev->pdev->device;
+
+	spin_lock_init(&dev_priv->irq_lock);
+	spin_lock_init(&dev_priv->gpu_error.lock);
+	mutex_init(&dev_priv->backlight_lock);
+	spin_lock_init(&dev_priv->uncore.lock);
+	spin_lock_init(&dev_priv->mm.object_stat_lock);
+	spin_lock_init(&dev_priv->mmio_flip_lock);
+	mutex_init(&dev_priv->sb_lock);
+	mutex_init(&dev_priv->modeset_restore_lock);
+	mutex_init(&dev_priv->av_mutex);
+	mutex_init(&dev_priv->wm.wm_mutex);
+	mutex_init(&dev_priv->pps_mutex);
+
+	ret = i915_workqueues_init(dev_priv);
+	if (ret < 0)
+		return ret;
+
+	/* This must be called before any calls to HAS_PCH_* */
+	intel_detect_pch(dev);
+
+	intel_pm_setup(dev);
+	intel_init_dpio(dev_priv);
+	intel_power_domains_init(dev_priv);
+	intel_irq_init(dev_priv);
+	intel_init_display_hooks(dev_priv);
+	intel_init_clock_gating_hooks(dev_priv);
+	intel_init_audio_hooks(dev_priv);
+	i915_gem_load_init(dev);
+
+	intel_display_crc_init(dev);
+
+	i915_dump_device_info(dev_priv);
+
+	/* Not all pre-production machines fall into this category, only the
+	 * very first ones. Almost everything should work, except for maybe
+	 * suspend/resume. And we don't implement workarounds that affect only
+	 * pre-production machines. */
+	if (IS_HSW_EARLY_SDV(dev))
+		DRM_INFO("This is an early pre-production Haswell machine. "
+			 "It may not be fully functional.\n");
+
+	return 0;
+}
+
+/**
+ * i915_driver_cleanup_early - cleanup the setup done in i915_driver_init_early()
+ * @dev_priv: device private
+ */
+static void i915_driver_cleanup_early(struct drm_i915_private *dev_priv)
+{
+	i915_gem_load_cleanup(dev_priv->dev);
+	i915_workqueues_cleanup(dev_priv);
+}
+
 static int i915_mmio_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -970,97 +1114,87 @@ static void i915_mmio_cleanup(struct drm_device *dev)
 }
 
 /**
- * i915_driver_load - setup chip and create an initial config
- * @dev: DRM device
- * @flags: startup flags
+ * i915_driver_init_mmio - setup device MMIO
+ * @dev_priv: device private
  *
- * The driver load routine has to do several things:
- *   - drive output discovery via intel_modeset_init()
- *   - initialize the memory manager
- *   - allocate initial config memory
- *   - setup the DRM framebuffer with the allocated memory
+ * Setup minimal device state necessary for MMIO accesses later in the
+ * initialization sequence. The setup here should avoid any other device-wide
+ * side effects or exposing the driver via kernel internal or user space
+ * interfaces.
  */
-int i915_driver_load(struct drm_device *dev, unsigned long flags)
+static int i915_driver_init_mmio(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv;
-	struct intel_device_info *info, *device_info;
-	int ret = 0;
-	uint32_t aperture_size;
-
-	info = (struct intel_device_info *) flags;
-
-	dev_priv = kzalloc(sizeof(*dev_priv), GFP_KERNEL);
-	if (dev_priv == NULL)
-		return -ENOMEM;
-
-	dev->dev_private = dev_priv;
-	dev_priv->dev = dev;
+	struct drm_device *dev = dev_priv->dev;
+	int ret;
 
-	/* Setup the write-once "constant" device info */
-	device_info = (struct intel_device_info *)&dev_priv->info;
-	memcpy(device_info, info, sizeof(dev_priv->info));
-	device_info->device_id = dev->pdev->device;
+	if (i915_inject_load_failure())
+		return -ENODEV;
 
-	spin_lock_init(&dev_priv->irq_lock);
-	spin_lock_init(&dev_priv->gpu_error.lock);
-	mutex_init(&dev_priv->backlight_lock);
-	spin_lock_init(&dev_priv->uncore.lock);
-	spin_lock_init(&dev_priv->mm.object_stat_lock);
-	spin_lock_init(&dev_priv->mmio_flip_lock);
-	mutex_init(&dev_priv->sb_lock);
-	mutex_init(&dev_priv->modeset_restore_lock);
-	mutex_init(&dev_priv->av_mutex);
+	if (i915_get_bridge_dev(dev))
+		return -EIO;
 
-	ret = i915_workqueues_init(dev_priv);
+	ret = i915_mmio_setup(dev);
 	if (ret < 0)
-		goto out_free_priv;
+		goto put_bridge;
 
-	intel_pm_setup(dev);
+	intel_uncore_init(dev);
 
-	intel_runtime_pm_get(dev_priv);
+	return 0;
 
-	intel_display_crc_init(dev);
+put_bridge:
+	pci_dev_put(dev_priv->bridge_dev);
 
-	i915_dump_device_info(dev_priv);
+	return ret;
+}
 
-	/* Not all pre-production machines fall into this category, only the
-	 * very first ones. Almost everything should work, except for maybe
-	 * suspend/resume. And we don't implement workarounds that affect only
-	 * pre-production machines. */
-	if (IS_HSW_EARLY_SDV(dev))
-		DRM_INFO("This is an early pre-production Haswell machine. "
-			 "It may not be fully functional.\n");
+/**
+ * i915_driver_cleanup_mmio - cleanup the setup done in i915_driver_init_mmio()
+ * @dev_priv: device private
+ */
+static void i915_driver_cleanup_mmio(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
 
-	if (i915_get_bridge_dev(dev)) {
-		ret = -EIO;
-		goto out_runtime_pm_put;
-	}
+	intel_uncore_fini(dev);
+	i915_mmio_cleanup(dev);
+	pci_dev_put(dev_priv->bridge_dev);
+}
 
-	ret = i915_mmio_setup(dev);
-	if (ret < 0)
-		goto put_bridge;
+/**
+ * i915_driver_init_hw - setup state requiring device access
+ * @dev_priv: device private
+ *
+ * Setup state that requires accessing the device, but doesn't require
+ * exposing the driver via kernel internal or userspace interfaces.
+ */
+static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+	uint32_t aperture_size;
+	int ret;
 
-	/* This must be called before any calls to HAS_PCH_* */
-	intel_detect_pch(dev);
+	if (i915_inject_load_failure())
+		return -ENODEV;
 
-	intel_uncore_init(dev);
+	intel_device_info_runtime_init(dev);
 
-	ret = i915_gem_gtt_init(dev);
+	ret = i915_ggtt_init_hw(dev);
 	if (ret)
-		goto out_uncore_fini;
+		return ret;
 
 	/* WARNING: Apparently we must kick fbdev drivers before vgacon,
 	 * otherwise the vga fbdev driver falls over. */
 	ret = i915_kick_out_firmware_fb(dev_priv);
 	if (ret) {
 		DRM_ERROR("failed to remove conflicting framebuffer drivers\n");
-		goto out_gtt;
+		goto out_ggtt;
 	}
 
 	ret = i915_kick_out_vgacon(dev_priv);
 	if (ret) {
 		DRM_ERROR("failed to remove conflicting VGA console\n");
-		goto out_gtt;
+		goto out_ggtt;
 	}
 
 	pci_set_master(dev->pdev);
@@ -1080,26 +1214,27 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	if (IS_BROADWATER(dev) || IS_CRESTLINE(dev))
 		dma_set_coherent_mask(&dev->pdev->dev, DMA_BIT_MASK(32));
 
-	aperture_size = dev_priv->gtt.mappable_end;
+	aperture_size = ggtt->mappable_end;
 
-	dev_priv->gtt.mappable =
-		io_mapping_create_wc(dev_priv->gtt.mappable_base,
+	ggtt->mappable =
+		io_mapping_create_wc(ggtt->mappable_base,
 				     aperture_size);
-	if (dev_priv->gtt.mappable == NULL) {
+	if (!ggtt->mappable) {
 		ret = -EIO;
-		goto out_gtt;
+		goto out_ggtt;
 	}
 
-	dev_priv->gtt.mtrr = arch_phys_wc_add(dev_priv->gtt.mappable_base,
+	ggtt->mtrr = arch_phys_wc_add(ggtt->mappable_base,
 					      aperture_size);
 
-	intel_irq_init(dev_priv);
+	pm_qos_add_request(&dev_priv->pm_qos, PM_QOS_CPU_DMA_LATENCY,
+			   PM_QOS_DEFAULT_VALUE);
+
 	intel_uncore_sanitize(dev);
 
 	intel_opregion_setup(dev);
 
-	i915_gem_load_init(dev);
-	i915_gem_shrinker_init(dev_priv);
+	i915_gem_load_init_fences(dev_priv);
 
 	/* On the 945G/GM, the chipset reports the MSI capability on the
 	 * integrated graphics even though the support isn't actually there
@@ -1117,24 +1252,44 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 			DRM_DEBUG_DRIVER("can't enable MSI");
 	}
 
-	intel_device_info_runtime_init(dev);
+	return 0;
 
-	intel_init_dpio(dev_priv);
+out_ggtt:
+	i915_ggtt_cleanup_hw(dev);
 
-	if (INTEL_INFO(dev)->num_pipes) {
-		ret = drm_vblank_init(dev, INTEL_INFO(dev)->num_pipes);
-		if (ret)
-			goto out_gem_unload;
-	}
+	return ret;
+}
 
-	intel_power_domains_init(dev_priv);
+/**
+ * i915_driver_cleanup_hw - cleanup the setup done in i915_driver_init_hw()
+ * @dev_priv: device private
+ */
+static void i915_driver_cleanup_hw(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 
-	ret = i915_load_modeset_init(dev);
-	if (ret < 0) {
-		DRM_ERROR("failed to init modeset\n");
-		goto out_power_well;
-	}
+	if (dev->pdev->msi_enabled)
+		pci_disable_msi(dev->pdev);
+
+	pm_qos_remove_request(&dev_priv->pm_qos);
+	arch_phys_wc_del(ggtt->mtrr);
+	io_mapping_free(ggtt->mappable);
+	i915_ggtt_cleanup_hw(dev);
+}
 
+/**
+ * i915_driver_register - register the driver with the rest of the system
+ * @dev_priv: device private
+ *
+ * Perform any steps necessary to make the driver available via kernel
+ * internal or userspace interfaces.
+ */
+static void i915_driver_register(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	i915_gem_shrinker_init(dev_priv);
 	/*
 	 * Notify a valid surface after modesetting,
 	 * when running inside a VM.
@@ -1144,48 +1299,107 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 
 	i915_setup_sysfs(dev);
 
-	if (INTEL_INFO(dev)->num_pipes) {
+	if (INTEL_INFO(dev_priv)->num_pipes) {
 		/* Must be done after probing outputs */
 		intel_opregion_init(dev);
 		acpi_video_register();
 	}
 
-	if (IS_GEN5(dev))
+	if (IS_GEN5(dev_priv))
 		intel_gpu_ips_init(dev_priv);
 
-	intel_runtime_pm_enable(dev_priv);
-
 	i915_audio_component_init(dev_priv);
+}
+
+/**
+ * i915_driver_unregister - cleanup the registration done in i915_driver_regiser()
+ * @dev_priv: device private
+ */
+static void i915_driver_unregister(struct drm_i915_private *dev_priv)
+{
+	i915_audio_component_cleanup(dev_priv);
+	intel_gpu_ips_teardown();
+	acpi_video_unregister();
+	intel_opregion_fini(dev_priv->dev);
+	i915_teardown_sysfs(dev_priv->dev);
+	i915_gem_shrinker_cleanup(dev_priv);
+}
+
+/**
+ * i915_driver_load - setup chip and create an initial config
+ * @dev: DRM device
+ * @flags: startup flags
+ *
+ * The driver load routine has to do several things:
+ *   - drive output discovery via intel_modeset_init()
+ *   - initialize the memory manager
+ *   - allocate initial config memory
+ *   - setup the DRM framebuffer with the allocated memory
+ */
+int i915_driver_load(struct drm_device *dev, unsigned long flags)
+{
+	struct drm_i915_private *dev_priv;
+	int ret = 0;
+
+	dev_priv = kzalloc(sizeof(*dev_priv), GFP_KERNEL);
+	if (dev_priv == NULL)
+		return -ENOMEM;
+
+	dev->dev_private = dev_priv;
+	/* Must be set before calling __i915_printk */
+	dev_priv->dev = dev;
+
+	ret = i915_driver_init_early(dev_priv, dev,
+				     (struct intel_device_info *)flags);
+
+	if (ret < 0)
+		goto out_free_priv;
+
+	intel_runtime_pm_get(dev_priv);
+
+	ret = i915_driver_init_mmio(dev_priv);
+	if (ret < 0)
+		goto out_runtime_pm_put;
+
+	ret = i915_driver_init_hw(dev_priv);
+	if (ret < 0)
+		goto out_cleanup_mmio;
+
+	/*
+	 * TODO: move the vblank init and parts of modeset init steps into one
+	 * of the i915_driver_init_/i915_driver_register functions according
+	 * to the role/effect of the given init step.
+	 */
+	if (INTEL_INFO(dev)->num_pipes) {
+		ret = drm_vblank_init(dev, INTEL_INFO(dev)->num_pipes);
+		if (ret)
+			goto out_cleanup_hw;
+	}
+
+	ret = i915_load_modeset_init(dev);
+	if (ret < 0)
+		goto out_cleanup_vblank;
+
+	i915_driver_register(dev_priv);
+
+	intel_runtime_pm_enable(dev_priv);
 
 	intel_runtime_pm_put(dev_priv);
 
 	return 0;
 
-out_power_well:
-	intel_power_domains_fini(dev_priv);
+out_cleanup_vblank:
 	drm_vblank_cleanup(dev);
-out_gem_unload:
-	i915_gem_shrinker_cleanup(dev_priv);
-
-	if (dev->pdev->msi_enabled)
-		pci_disable_msi(dev->pdev);
-
-	intel_teardown_mchbar(dev);
-	pm_qos_remove_request(&dev_priv->pm_qos);
-	arch_phys_wc_del(dev_priv->gtt.mtrr);
-	io_mapping_free(dev_priv->gtt.mappable);
-out_gtt:
-	i915_global_gtt_cleanup(dev);
-out_uncore_fini:
-	intel_uncore_fini(dev);
-	i915_mmio_cleanup(dev);
-put_bridge:
-	pci_dev_put(dev_priv->bridge_dev);
-	i915_gem_load_cleanup(dev);
+out_cleanup_hw:
+	i915_driver_cleanup_hw(dev_priv);
+out_cleanup_mmio:
+	i915_driver_cleanup_mmio(dev_priv);
 out_runtime_pm_put:
 	intel_runtime_pm_put(dev_priv);
-	i915_workqueues_cleanup(dev_priv);
+	i915_driver_cleanup_early(dev_priv);
 out_free_priv:
+	i915_load_error(dev_priv, "Device initialization failed (%d)\n", ret);
+
 	kfree(dev_priv);
 
 	return ret;
@@ -1198,26 +1412,15 @@ int i915_driver_unload(struct drm_device *dev)
 
 	intel_fbdev_fini(dev);
 
-	i915_audio_component_cleanup(dev_priv);
-
 	ret = i915_gem_suspend(dev);
 	if (ret) {
 		DRM_ERROR("failed to idle hardware: %d\n", ret);
 		return ret;
 	}
 
-	intel_power_domains_fini(dev_priv);
-
-	intel_gpu_ips_teardown();
+	intel_display_power_get(dev_priv, POWER_DOMAIN_INIT);
 
-	i915_teardown_sysfs(dev);
-
-	i915_gem_shrinker_cleanup(dev_priv);
-
-	io_mapping_free(dev_priv->gtt.mappable);
-	arch_phys_wc_del(dev_priv->gtt.mtrr);
-
-	acpi_video_unregister();
+	i915_driver_unregister(dev_priv);
 
 	drm_vblank_cleanup(dev);
 
@@ -1246,31 +1449,24 @@ int i915_driver_unload(struct drm_device *dev)
 	cancel_delayed_work_sync(&dev_priv->gpu_error.hangcheck_work);
 	i915_destroy_error_state(dev);
 
-	if (dev->pdev->msi_enabled)
-		pci_disable_msi(dev->pdev);
-
-	intel_opregion_fini(dev);
-
 	/* Flush any outstanding unpin_work. */
 	flush_workqueue(dev_priv->wq);
 
 	intel_guc_ucode_fini(dev);
 	mutex_lock(&dev->struct_mutex);
-	i915_gem_cleanup_ringbuffer(dev);
+	i915_gem_cleanup_engines(dev);
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
 	intel_fbc_cleanup_cfb(dev_priv);
 
-	pm_qos_remove_request(&dev_priv->pm_qos);
+	intel_power_domains_fini(dev_priv);
 
-	i915_global_gtt_cleanup(dev);
+	i915_driver_cleanup_hw(dev_priv);
+	i915_driver_cleanup_mmio(dev_priv);
 
-	intel_uncore_fini(dev);
-	i915_mmio_cleanup(dev);
+	intel_display_power_put(dev_priv, POWER_DOMAIN_INIT);
 
-	i915_gem_load_cleanup(dev);
-	pci_dev_put(dev_priv->bridge_dev);
-	i915_workqueues_cleanup(dev_priv);
+	i915_driver_cleanup_early(dev_priv);
 	kfree(dev_priv);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 6d2fb3f4ac62..d37c0a671eed 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -66,6 +66,11 @@ static struct drm_driver driver;
 #define IVB_CURSOR_OFFSETS \
 	.cursor_offsets = { CURSOR_A_OFFSET, IVB_CURSOR_B_OFFSET, IVB_CURSOR_C_OFFSET }
 
+#define BDW_COLORS \
+	.color = { .degamma_lut_size = 512, .gamma_lut_size = 512 }
+#define CHV_COLORS \
+	.color = { .degamma_lut_size = 65, .gamma_lut_size = 257 }
+
 static const struct intel_device_info intel_i830_info = {
 	.gen = 2, .is_mobile = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
@@ -288,24 +293,28 @@ static const struct intel_device_info intel_haswell_m_info = {
 	.is_mobile = 1,
 };
 
+#define BDW_FEATURES \
+	HSW_FEATURES, \
+	BDW_COLORS
+
 static const struct intel_device_info intel_broadwell_d_info = {
-	HSW_FEATURES,
+	BDW_FEATURES,
 	.gen = 8,
 };
 
 static const struct intel_device_info intel_broadwell_m_info = {
-	HSW_FEATURES,
+	BDW_FEATURES,
 	.gen = 8, .is_mobile = 1,
 };
 
 static const struct intel_device_info intel_broadwell_gt3d_info = {
-	HSW_FEATURES,
+	BDW_FEATURES,
 	.gen = 8,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
 };
 
 static const struct intel_device_info intel_broadwell_gt3m_info = {
-	HSW_FEATURES,
+	BDW_FEATURES,
 	.gen = 8, .is_mobile = 1,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
 };
@@ -318,16 +327,17 @@ static const struct intel_device_info intel_cherryview_info = {
 	.display_mmio_offset = VLV_DISPLAY_BASE,
 	GEN_CHV_PIPEOFFSETS,
 	CURSOR_OFFSETS,
+	CHV_COLORS,
 };
 
 static const struct intel_device_info intel_skylake_info = {
-	HSW_FEATURES,
+	BDW_FEATURES,
 	.is_skylake = 1,
 	.gen = 9,
 };
 
 static const struct intel_device_info intel_skylake_gt3_info = {
-	HSW_FEATURES,
+	BDW_FEATURES,
 	.is_skylake = 1,
 	.gen = 9,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
@@ -345,18 +355,17 @@ static const struct intel_device_info intel_broxton_info = {
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
 	IVB_CURSOR_OFFSETS,
+	BDW_COLORS,
 };
 
 static const struct intel_device_info intel_kabylake_info = {
-	HSW_FEATURES,
-	.is_preliminary = 1,
+	BDW_FEATURES,
 	.is_kabylake = 1,
 	.gen = 9,
 };
 
 static const struct intel_device_info intel_kabylake_gt3_info = {
-	HSW_FEATURES,
-	.is_preliminary = 1,
+	BDW_FEATURES,
 	.is_kabylake = 1,
 	.gen = 9,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
@@ -504,6 +513,7 @@ void intel_detect_pch(struct drm_device *dev)
 				WARN_ON(!IS_SKYLAKE(dev) &&
 					!IS_KABYLAKE(dev));
 			} else if ((id == INTEL_PCH_P2X_DEVICE_ID_TYPE) ||
+				   (id == INTEL_PCH_P3X_DEVICE_ID_TYPE) ||
 				   ((id == INTEL_PCH_QEMU_DEVICE_ID_TYPE) &&
 				    pch->subsystem_vendor == 0x1af4 &&
 				    pch->subsystem_device == 0x1100)) {
@@ -557,10 +567,9 @@ static void intel_suspend_encoders(struct drm_i915_private *dev_priv)
 	drm_modeset_unlock_all(dev);
 }
 
-static int intel_suspend_complete(struct drm_i915_private *dev_priv);
 static int vlv_resume_prepare(struct drm_i915_private *dev_priv,
 			      bool rpm_resume);
-static int bxt_resume_prepare(struct drm_i915_private *dev_priv);
+static int vlv_suspend_complete(struct drm_i915_private *dev_priv);
 
 static bool suspend_to_idle(struct drm_i915_private *dev_priv)
 {
@@ -630,8 +639,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 
 	intel_display_set_init_power(dev_priv, false);
 
-	if (HAS_CSR(dev_priv))
-		flush_work(&dev_priv->csr.work);
+	intel_csr_ucode_suspend(dev_priv);
 
 out:
 	enable_rpm_wakeref_asserts(dev_priv);
@@ -647,7 +655,8 @@ static int i915_drm_suspend_late(struct drm_device *drm_dev, bool hibernation)
 
 	disable_rpm_wakeref_asserts(dev_priv);
 
-	fw_csr = suspend_to_idle(dev_priv) && dev_priv->csr.dmc_payload;
+	fw_csr = !IS_BROXTON(dev_priv) &&
+		suspend_to_idle(dev_priv) && dev_priv->csr.dmc_payload;
 	/*
 	 * In case of firmware assisted context save/restore don't manually
 	 * deinit the power domains. This also means the CSR/DMC firmware will
@@ -658,7 +667,13 @@ static int i915_drm_suspend_late(struct drm_device *drm_dev, bool hibernation)
 	if (!fw_csr)
 		intel_power_domains_suspend(dev_priv);
 
-	ret = intel_suspend_complete(dev_priv);
+	ret = 0;
+	if (IS_BROXTON(dev_priv))
+		bxt_enable_dc9(dev_priv);
+	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+		hsw_enable_pc8(dev_priv);
+	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+		ret = vlv_suspend_complete(dev_priv);
 
 	if (ret) {
 		DRM_ERROR("Suspend complete failed: %d\n", ret);
@@ -722,6 +737,8 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	disable_rpm_wakeref_asserts(dev_priv);
 
+	intel_csr_ucode_resume(dev_priv);
+
 	mutex_lock(&dev->struct_mutex);
 	i915_gem_restore_gtt_mappings(dev);
 	mutex_unlock(&dev->struct_mutex);
@@ -850,21 +867,25 @@ static int i915_drm_resume_early(struct drm_device *dev)
 
 	intel_uncore_early_sanitize(dev, true);
 
-	if (IS_BROXTON(dev))
-		ret = bxt_resume_prepare(dev_priv);
-	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+	if (IS_BROXTON(dev)) {
+		if (!dev_priv->suspended_to_idle)
+			gen9_sanitize_dc_state(dev_priv);
+		bxt_disable_dc9(dev_priv);
+	} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
 		hsw_disable_pc8(dev_priv);
+	}
 
 	intel_uncore_sanitize(dev);
 
-	if (!(dev_priv->suspended_to_idle && dev_priv->csr.dmc_payload))
+	if (IS_BROXTON(dev_priv) ||
+	    !(dev_priv->suspended_to_idle && dev_priv->csr.dmc_payload))
 		intel_power_domains_init_hw(dev_priv, true);
 
+	enable_rpm_wakeref_asserts(dev_priv);
+
 out:
 	dev_priv->suspended_to_idle = false;
 
-	enable_rpm_wakeref_asserts(dev_priv);
-
 	return ret;
 }
 
@@ -900,23 +921,32 @@ int i915_resume_switcheroo(struct drm_device *dev)
 int i915_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	bool simulated;
+	struct i915_gpu_error *error = &dev_priv->gpu_error;
+	unsigned reset_counter;
 	int ret;
 
 	intel_reset_gt_powersave(dev);
 
 	mutex_lock(&dev->struct_mutex);
 
-	i915_gem_reset(dev);
+	/* Clear any previous failed attempts at recovery. Time to try again. */
+	atomic_andnot(I915_WEDGED, &error->reset_counter);
+
+	/* Clear the reset-in-progress flag and increment the reset epoch. */
+	reset_counter = atomic_inc_return(&error->reset_counter);
+	if (WARN_ON(__i915_reset_in_progress(reset_counter))) {
+		ret = -EIO;
+		goto error;
+	}
 
-	simulated = dev_priv->gpu_error.stop_rings != 0;
+	i915_gem_reset(dev);
 
-	ret = intel_gpu_reset(dev);
+	ret = intel_gpu_reset(dev, ALL_ENGINES);
 
 	/* Also reset the gpu hangman. */
-	if (simulated) {
+	if (error->stop_rings != 0) {
 		DRM_INFO("Simulated gpu hang, resetting stop_rings\n");
-		dev_priv->gpu_error.stop_rings = 0;
+		error->stop_rings = 0;
 		if (ret == -ENODEV) {
 			DRM_INFO("Reset not implemented, but ignoring "
 				 "error for simulated gpu hangs\n");
@@ -928,9 +958,11 @@ int i915_reset(struct drm_device *dev)
 		pr_notice("drm/i915: Resetting chip after gpu hang\n");
 
 	if (ret) {
-		DRM_ERROR("Failed to reset chip: %i\n", ret);
-		mutex_unlock(&dev->struct_mutex);
-		return ret;
+		if (ret != -ENODEV)
+			DRM_ERROR("Failed to reset chip: %i\n", ret);
+		else
+			DRM_DEBUG_DRIVER("GPU reset disabled\n");
+		goto error;
 	}
 
 	intel_overlay_reset(dev_priv);
@@ -949,20 +981,14 @@ int i915_reset(struct drm_device *dev)
 	 * was running at the time of the reset (i.e. we weren't VT
 	 * switched away).
 	 */
-
-	/* Used to prevent gem_check_wedged returning -EAGAIN during gpu reset */
-	dev_priv->gpu_error.reload_in_reset = true;
-
 	ret = i915_gem_init_hw(dev);
-
-	dev_priv->gpu_error.reload_in_reset = false;
-
-	mutex_unlock(&dev->struct_mutex);
 	if (ret) {
 		DRM_ERROR("Failed hw init on reset %d\n", ret);
-		return ret;
+		goto error;
 	}
 
+	mutex_unlock(&dev->struct_mutex);
+
 	/*
 	 * rps/rc6 re-init is necessary to restore state lost after the
 	 * reset and the re-install of gt irqs. Skip for ironlake per
@@ -973,6 +999,11 @@ int i915_reset(struct drm_device *dev)
 		intel_enable_gt_powersave(dev);
 
 	return 0;
+
+error:
+	atomic_or(I915_WEDGED, &error->reset_counter);
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
 }
 
 static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
@@ -1079,44 +1110,6 @@ static int i915_pm_resume(struct device *dev)
 	return i915_drm_resume(drm_dev);
 }
 
-static int hsw_suspend_complete(struct drm_i915_private *dev_priv)
-{
-	hsw_enable_pc8(dev_priv);
-
-	return 0;
-}
-
-static int bxt_suspend_complete(struct drm_i915_private *dev_priv)
-{
-	struct drm_device *dev = dev_priv->dev;
-
-	/* TODO: when DC5 support is added disable DC5 here. */
-
-	broxton_ddi_phy_uninit(dev);
-	broxton_uninit_cdclk(dev);
-	bxt_enable_dc9(dev_priv);
-
-	return 0;
-}
-
-static int bxt_resume_prepare(struct drm_i915_private *dev_priv)
-{
-	struct drm_device *dev = dev_priv->dev;
-
-	/* TODO: when CSR FW support is added make sure the FW is loaded */
-
-	bxt_disable_dc9(dev_priv);
-
-	/*
-	 * TODO: when DC5 support is added enable DC5 here if the CSR FW
-	 * is available.
-	 */
-	broxton_init_cdclk(dev);
-	broxton_ddi_phy_init(dev);
-
-	return 0;
-}
-
 /*
  * Save all Gunit registers that may be lost after a D3 and a subsequent
  * S0i[R123] transition. The list of registers needing a save/restore is
@@ -1420,7 +1413,7 @@ static int vlv_suspend_complete(struct drm_i915_private *dev_priv)
 	if (err)
 		goto err2;
 
-	if (!IS_CHERRYVIEW(dev_priv->dev))
+	if (!IS_CHERRYVIEW(dev_priv))
 		vlv_save_gunit_s0ix_state(dev_priv);
 
 	err = vlv_force_gfx_clock(dev_priv, false);
@@ -1452,7 +1445,7 @@ static int vlv_resume_prepare(struct drm_i915_private *dev_priv,
 	 */
 	ret = vlv_force_gfx_clock(dev_priv, true);
 
-	if (!IS_CHERRYVIEW(dev_priv->dev))
+	if (!IS_CHERRYVIEW(dev_priv))
 		vlv_restore_gunit_s0ix_state(dev_priv);
 
 	err = vlv_allow_gt_wake(dev_priv, true);
@@ -1522,7 +1515,16 @@ static int intel_runtime_suspend(struct device *device)
 	intel_suspend_gt_powersave(dev);
 	intel_runtime_pm_disable_interrupts(dev_priv);
 
-	ret = intel_suspend_complete(dev_priv);
+	ret = 0;
+	if (IS_BROXTON(dev_priv)) {
+		bxt_display_core_uninit(dev_priv);
+		bxt_enable_dc9(dev_priv);
+	} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
+		hsw_enable_pc8(dev_priv);
+	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
+		ret = vlv_suspend_complete(dev_priv);
+	}
+
 	if (ret) {
 		DRM_ERROR("Runtime suspend failed, disabling it (%d)\n", ret);
 		intel_runtime_pm_enable_interrupts(dev_priv);
@@ -1596,12 +1598,17 @@ static int intel_runtime_resume(struct device *device)
 	if (IS_GEN6(dev_priv))
 		intel_init_pch_refclk(dev);
 
-	if (IS_BROXTON(dev))
-		ret = bxt_resume_prepare(dev_priv);
-	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+	if (IS_BROXTON(dev)) {
+		bxt_disable_dc9(dev_priv);
+		bxt_display_core_init(dev_priv, true);
+		if (dev_priv->csr.dmc_payload &&
+		    (dev_priv->csr.allowed_dc_mask & DC_STATE_EN_UPTO_DC5))
+			gen9_enable_dc5(dev_priv);
+	} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
 		hsw_disable_pc8(dev_priv);
-	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
 		ret = vlv_resume_prepare(dev_priv, true);
+	}
 
 	/*
 	 * No point of rolling back things in case of an error, as the best
@@ -1632,26 +1639,6 @@ static int intel_runtime_resume(struct device *device)
 	return ret;
 }
 
-/*
- * This function implements common functionality of runtime and system
- * suspend sequence.
- */
-static int intel_suspend_complete(struct drm_i915_private *dev_priv)
-{
-	int ret;
-
-	if (IS_BROXTON(dev_priv))
-		ret = bxt_suspend_complete(dev_priv);
-	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
-		ret = hsw_suspend_complete(dev_priv);
-	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		ret = vlv_suspend_complete(dev_priv);
-	else
-		ret = 0;
-
-	return ret;
-}
-
 static const struct dev_pm_ops i915_pm_ops = {
 	/*
 	 * S0ix (via system suspend) and S3 event handlers [PMSG_SUSPEND,
@@ -1772,10 +1759,8 @@ static int __init i915_init(void)
 	if (i915.modeset == 0)
 		driver.driver_features &= ~DRIVER_MODESET;
 
-#ifdef CONFIG_VGA_CONSOLE
 	if (vgacon_text_force() && i915.modeset == -1)
 		driver.driver_features &= ~DRIVER_MODESET;
-#endif
 
 	if (!(driver.driver_features & DRIVER_MODESET)) {
 		/* Silently fail loading to not upset userspace. */
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 5d7a7c4f5136..b87ca4fae20a 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -33,33 +33,40 @@
 #include <uapi/drm/i915_drm.h>
 #include <uapi/drm/drm_fourcc.h>
 
-#include <drm/drmP.h>
-#include "i915_params.h"
-#include "i915_reg.h"
-#include "intel_bios.h"
-#include "intel_ringbuffer.h"
-#include "intel_lrc.h"
-#include "i915_gem_gtt.h"
-#include "i915_gem_render_state.h"
 #include <linux/io-mapping.h>
 #include <linux/i2c.h>
 #include <linux/i2c-algo-bit.h>
-#include <drm/intel-gtt.h>
-#include <drm/drm_legacy.h> /* for struct drm_dma_handle */
-#include <drm/drm_gem.h>
 #include <linux/backlight.h>
 #include <linux/hashtable.h>
 #include <linux/intel-iommu.h>
 #include <linux/kref.h>
 #include <linux/pm_qos.h>
+#include <linux/shmem_fs.h>
+
+#include <drm/drmP.h>
+#include <drm/intel-gtt.h>
+#include <drm/drm_legacy.h> /* for struct drm_dma_handle */
+#include <drm/drm_gem.h>
+
+#include "i915_params.h"
+#include "i915_reg.h"
+
+#include "intel_bios.h"
+#include "intel_dpll_mgr.h"
 #include "intel_guc.h"
+#include "intel_lrc.h"
+#include "intel_ringbuffer.h"
+
+#include "i915_gem.h"
+#include "i915_gem_gtt.h"
+#include "i915_gem_render_state.h"
 
 /* General customization:
  */
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20160229"
+#define DRIVER_DATE		"20160425"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
@@ -97,6 +104,10 @@
 #define I915_STATE_WARN_ON(x)						\
 	I915_STATE_WARN((x), "%s", "WARN_ON(" __stringify(x) ")")
 
+bool __i915_inject_load_failure(const char *func, int line);
+#define i915_inject_load_failure() \
+	__i915_inject_load_failure(__func__, __LINE__)
+
 static inline const char *yesno(bool v)
 {
 	return v ? "yes" : "no";
@@ -122,9 +133,35 @@ enum transcoder {
 	TRANSCODER_B,
 	TRANSCODER_C,
 	TRANSCODER_EDP,
+	TRANSCODER_DSI_A,
+	TRANSCODER_DSI_C,
 	I915_MAX_TRANSCODERS
 };
-#define transcoder_name(t) ((t) + 'A')
+
+static inline const char *transcoder_name(enum transcoder transcoder)
+{
+	switch (transcoder) {
+	case TRANSCODER_A:
+		return "A";
+	case TRANSCODER_B:
+		return "B";
+	case TRANSCODER_C:
+		return "C";
+	case TRANSCODER_EDP:
+		return "EDP";
+	case TRANSCODER_DSI_A:
+		return "DSI A";
+	case TRANSCODER_DSI_C:
+		return "DSI C";
+	default:
+		return "<invalid>";
+	}
+}
+
+static inline bool transcoder_is_dsi(enum transcoder transcoder)
+{
+	return transcoder == TRANSCODER_DSI_A || transcoder == TRANSCODER_DSI_C;
+}
 
 /*
  * I915_MAX_PLANES in the enum below is the maximum (across all platforms)
@@ -176,6 +213,8 @@ enum intel_display_power_domain {
 	POWER_DOMAIN_TRANSCODER_B,
 	POWER_DOMAIN_TRANSCODER_C,
 	POWER_DOMAIN_TRANSCODER_EDP,
+	POWER_DOMAIN_TRANSCODER_DSI_A,
+	POWER_DOMAIN_TRANSCODER_DSI_C,
 	POWER_DOMAIN_PORT_DDI_A_LANES,
 	POWER_DOMAIN_PORT_DDI_B_LANES,
 	POWER_DOMAIN_PORT_DDI_C_LANES,
@@ -273,6 +312,10 @@ struct i915_hotplug {
 	     (__s) < INTEL_INFO(__dev_priv)->num_sprites[(__p)];	\
 	     (__s)++)
 
+#define for_each_port_masked(__port, __ports_mask) \
+	for ((__port) = PORT_A; (__port) < I915_MAX_PORTS; (__port)++)	\
+		for_each_if ((__ports_mask) & (1 << (__port)))
+
 #define for_each_crtc(dev, crtc) \
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
 
@@ -340,81 +383,6 @@ struct drm_i915_file_private {
 	unsigned int bsd_ring;
 };
 
-enum intel_dpll_id {
-	DPLL_ID_PRIVATE = -1, /* non-shared dpll in use */
-	/* real shared dpll ids must be >= 0 */
-	DPLL_ID_PCH_PLL_A = 0,
-	DPLL_ID_PCH_PLL_B = 1,
-	/* hsw/bdw */
-	DPLL_ID_WRPLL1 = 0,
-	DPLL_ID_WRPLL2 = 1,
-	DPLL_ID_SPLL = 2,
-
-	/* skl */
-	DPLL_ID_SKL_DPLL1 = 0,
-	DPLL_ID_SKL_DPLL2 = 1,
-	DPLL_ID_SKL_DPLL3 = 2,
-};
-#define I915_NUM_PLLS 3
-
-struct intel_dpll_hw_state {
-	/* i9xx, pch plls */
-	uint32_t dpll;
-	uint32_t dpll_md;
-	uint32_t fp0;
-	uint32_t fp1;
-
-	/* hsw, bdw */
-	uint32_t wrpll;
-	uint32_t spll;
-
-	/* skl */
-	/*
-	 * DPLL_CTRL1 has 6 bits for each each this DPLL. We store those in
-	 * lower part of ctrl1 and they get shifted into position when writing
-	 * the register.  This allows us to easily compare the state to share
-	 * the DPLL.
-	 */
-	uint32_t ctrl1;
-	/* HDMI only, 0 when used for DP */
-	uint32_t cfgcr1, cfgcr2;
-
-	/* bxt */
-	uint32_t ebb0, ebb4, pll0, pll1, pll2, pll3, pll6, pll8, pll9, pll10,
-		 pcsdw12;
-};
-
-struct intel_shared_dpll_config {
-	unsigned crtc_mask; /* mask of CRTCs sharing this PLL */
-	struct intel_dpll_hw_state hw_state;
-};
-
-struct intel_shared_dpll {
-	struct intel_shared_dpll_config config;
-
-	int active; /* count of number of active CRTCs (i.e. DPMS on) */
-	bool on; /* is the PLL actually active? Disabled during modeset */
-	const char *name;
-	/* should match the index in the dev_priv->shared_dplls array */
-	enum intel_dpll_id id;
-	/* The mode_set hook is optional and should be used together with the
-	 * intel_prepare_shared_dpll function. */
-	void (*mode_set)(struct drm_i915_private *dev_priv,
-			 struct intel_shared_dpll *pll);
-	void (*enable)(struct drm_i915_private *dev_priv,
-		       struct intel_shared_dpll *pll);
-	void (*disable)(struct drm_i915_private *dev_priv,
-			struct intel_shared_dpll *pll);
-	bool (*get_hw_state)(struct drm_i915_private *dev_priv,
-			     struct intel_shared_dpll *pll,
-			     struct intel_dpll_hw_state *hw_state);
-};
-
-#define SKL_DPLL0 0
-#define SKL_DPLL1 1
-#define SKL_DPLL2 2
-#define SKL_DPLL3 3
-
 /* Used by dp and fdi links */
 struct intel_link_m_n {
 	uint32_t	tu;
@@ -533,7 +501,8 @@ struct drm_i915_error_state {
 		u32 cpu_ring_head;
 		u32 cpu_ring_tail;
 
-		u32 semaphore_seqno[I915_NUM_RINGS - 1];
+		u32 last_seqno;
+		u32 semaphore_seqno[I915_NUM_ENGINES - 1];
 
 		/* Register state */
 		u32 start;
@@ -553,7 +522,7 @@ struct drm_i915_error_state {
 		u32 fault_reg;
 		u64 faddr;
 		u32 rc_psmi; /* sleep state */
-		u32 semaphore_mboxes[I915_NUM_RINGS - 1];
+		u32 semaphore_mboxes[I915_NUM_ENGINES - 1];
 
 		struct drm_i915_error_object {
 			int page_count;
@@ -561,6 +530,8 @@ struct drm_i915_error_state {
 			u32 *pages[0];
 		} *ringbuffer, *batchbuffer, *wa_batchbuffer, *ctx, *hws_page;
 
+		struct drm_i915_error_object *wa_ctx;
+
 		struct drm_i915_error_request {
 			long jiffies;
 			u32 seqno;
@@ -577,12 +548,12 @@ struct drm_i915_error_state {
 
 		pid_t pid;
 		char comm[TASK_COMM_LEN];
-	} ring[I915_NUM_RINGS];
+	} ring[I915_NUM_ENGINES];
 
 	struct drm_i915_error_buffer {
 		u32 size;
 		u32 name;
-		u32 rseqno[I915_NUM_RINGS], wseqno;
+		u32 rseqno[I915_NUM_ENGINES], wseqno;
 		u64 gtt_offset;
 		u32 read_domains;
 		u32 write_domain;
@@ -611,27 +582,12 @@ struct dpll;
 struct drm_i915_display_funcs {
 	int (*get_display_clock_speed)(struct drm_device *dev);
 	int (*get_fifo_size)(struct drm_device *dev, int plane);
-	/**
-	 * find_dpll() - Find the best values for the PLL
-	 * @limit: limits for the PLL
-	 * @crtc: current CRTC
-	 * @target: target frequency in kHz
-	 * @refclk: reference clock frequency in kHz
-	 * @match_clock: if provided, @best_clock P divider must
-	 *               match the P divider from @match_clock
-	 *               used for LVDS downclocking
-	 * @best_clock: best PLL values found
-	 *
-	 * Returns true on success, false on failure.
-	 */
-	bool (*find_dpll)(const struct intel_limit *limit,
-			  struct intel_crtc_state *crtc_state,
-			  int target, int refclk,
-			  struct dpll *match_clock,
-			  struct dpll *best_clock);
-	int (*compute_pipe_wm)(struct intel_crtc *crtc,
-			       struct drm_atomic_state *state);
-	void (*program_watermarks)(struct intel_crtc_state *cstate);
+	int (*compute_pipe_wm)(struct intel_crtc_state *cstate);
+	int (*compute_intermediate_wm)(struct drm_device *dev,
+				       struct intel_crtc *intel_crtc,
+				       struct intel_crtc_state *newstate);
+	void (*initial_watermarks)(struct intel_crtc_state *cstate);
+	void (*optimize_watermarks)(struct intel_crtc_state *cstate);
 	void (*update_wm)(struct drm_crtc *crtc);
 	int (*modeset_calc_cdclk)(struct drm_atomic_state *state);
 	void (*modeset_commit_cdclk)(struct drm_atomic_state *state);
@@ -662,6 +618,9 @@ struct drm_i915_display_funcs {
 	/* render clock increase/decrease */
 	/* display clock increase/decrease */
 	/* pll clock increase/decrease */
+
+	void (*load_csc_matrix)(struct drm_crtc_state *crtc_state);
+	void (*load_luts)(struct drm_crtc_state *crtc_state);
 };
 
 enum forcewake_domain_id {
@@ -681,6 +640,13 @@ enum forcewake_domains {
 			 FORCEWAKE_MEDIA)
 };
 
+#define FW_REG_READ  (1)
+#define FW_REG_WRITE (2)
+
+enum forcewake_domains
+intel_uncore_forcewake_for_reg(struct drm_i915_private *dev_priv,
+			       i915_reg_t reg, unsigned int op);
+
 struct intel_uncore_funcs {
 	void (*force_wake_get)(struct drm_i915_private *dev_priv,
 							enum forcewake_domains domains);
@@ -713,8 +679,9 @@ struct intel_uncore {
 	struct intel_uncore_forcewake_domain {
 		struct drm_i915_private *i915;
 		enum forcewake_domain_id id;
+		enum forcewake_domains mask;
 		unsigned wake_count;
-		struct timer_list timer;
+		struct hrtimer timer;
 		i915_reg_t reg_set;
 		u32 val_set;
 		u32 val_clear;
@@ -727,14 +694,14 @@ struct intel_uncore {
 };
 
 /* Iterate over initialised fw domains */
-#define for_each_fw_domain_mask(domain__, mask__, dev_priv__, i__) \
-	for ((i__) = 0, (domain__) = &(dev_priv__)->uncore.fw_domain[0]; \
-	     (i__) < FW_DOMAIN_ID_COUNT; \
-	     (i__)++, (domain__) = &(dev_priv__)->uncore.fw_domain[i__]) \
-		for_each_if (((mask__) & (dev_priv__)->uncore.fw_domains) & (1 << (i__)))
+#define for_each_fw_domain_masked(domain__, mask__, dev_priv__) \
+	for ((domain__) = &(dev_priv__)->uncore.fw_domain[0]; \
+	     (domain__) < &(dev_priv__)->uncore.fw_domain[FW_DOMAIN_ID_COUNT]; \
+	     (domain__)++) \
+		for_each_if ((mask__) & (domain__)->mask)
 
-#define for_each_fw_domain(domain__, dev_priv__, i__) \
-	for_each_fw_domain_mask(domain__, FORCEWAKE_ALL, dev_priv__, i__)
+#define for_each_fw_domain(domain__, dev_priv__) \
+	for_each_fw_domain_masked(domain__, FORCEWAKE_ALL, dev_priv__)
 
 #define CSR_VERSION(major, minor)	((major) << 16 | (minor))
 #define CSR_VERSION_MAJOR(version)	((version) >> 16)
@@ -750,6 +717,7 @@ struct intel_csr {
 	i915_reg_t mmioaddr[8];
 	uint32_t mmiodata[8];
 	uint32_t dc_state;
+	uint32_t allowed_dc_mask;
 };
 
 #define DEV_INFO_FOR_EACH_FLAG(func, sep) \
@@ -779,6 +747,7 @@ struct intel_csr {
 	func(overlay_needs_physical) sep \
 	func(supports_tv) sep \
 	func(has_llc) sep \
+	func(has_snoop) sep \
 	func(has_ddi) sep \
 	func(has_fpga_dbg)
 
@@ -810,6 +779,11 @@ struct intel_device_info {
 	u8 has_slice_pg:1;
 	u8 has_subslice_pg:1;
 	u8 has_eu_pg:1;
+
+	struct color_luts {
+		u16 degamma_lut_size;
+		u16 gamma_lut_size;
+	} color;
 };
 
 #undef DEFINE_FLAG
@@ -891,7 +865,7 @@ struct intel_context {
 		struct i915_vma *lrc_vma;
 		u64 lrc_desc;
 		uint32_t *lrc_reg_state;
-	} engine[I915_NUM_RINGS];
+	} engine[I915_NUM_ENGINES];
 
 	struct list_head link;
 };
@@ -1036,6 +1010,7 @@ struct intel_fbc_work;
 
 struct intel_gmbus {
 	struct i2c_adapter adapter;
+#define GMBUS_FORCE_BIT_RETRY (1U << 31)
 	u32 force_bit;
 	u32 reg0;
 	i915_reg_t gpio_reg;
@@ -1159,6 +1134,7 @@ struct intel_gen6_power_mgmt {
 	u8 efficient_freq;	/* AKA RPe. Pre-determined balanced frequency */
 	u8 rp1_freq;		/* "less than" RP0 power/freqency */
 	u8 rp0_freq;		/* Non-overclocked max frequency. */
+	u16 gpll_ref_freq;	/* vlv/chv GPLL reference frequency */
 
 	u8 up_threshold; /* Current %busy required to uplock */
 	u8 down_threshold; /* Current %busy required to downclock */
@@ -1298,6 +1274,7 @@ struct i915_gem_mm {
 	struct i915_hw_ppgtt *aliasing_ppgtt;
 
 	struct notifier_block oom_notifier;
+	struct notifier_block vmap_notifier;
 	struct shrinker shrinker;
 	bool shrinker_no_lock_stealing;
 
@@ -1423,9 +1400,6 @@ struct i915_gpu_error {
 
 	/* For missed irq/seqno simulation. */
 	unsigned int test_irq_rings;
-
-	/* Used to prevent gem_check_wedged returning -EAGAIN during gpu reset   */
-	bool reload_in_reset;
 };
 
 enum modeset_restore {
@@ -1482,21 +1456,23 @@ struct intel_vbt_data {
 	unsigned int lvds_use_ssc:1;
 	unsigned int display_clock_mode:1;
 	unsigned int fdi_rx_polarity_inverted:1;
-	unsigned int has_mipi:1;
+	unsigned int panel_type:4;
 	int lvds_ssc_freq;
 	unsigned int bios_lvds_val; /* initial [PCH_]LVDS reg val in VBIOS */
 
 	enum drrs_support_type drrs_type;
 
-	/* eDP */
-	int edp_rate;
-	int edp_lanes;
-	int edp_preemphasis;
-	int edp_vswing;
-	bool edp_initialized;
-	bool edp_support;
-	int edp_bpp;
-	struct edp_power_seq edp_pps;
+	struct {
+		int rate;
+		int lanes;
+		int preemphasis;
+		int vswing;
+		bool low_vswing;
+		bool initialized;
+		bool support;
+		int bpp;
+		struct edp_power_seq pps;
+	} edp;
 
 	struct {
 		bool full_link;
@@ -1516,7 +1492,6 @@ struct intel_vbt_data {
 
 	/* MIPI DSI */
 	struct {
-		u16 port;
 		u16 panel_id;
 		struct mipi_config *config;
 		struct mipi_pps_data *pps;
@@ -1532,6 +1507,7 @@ struct intel_vbt_data {
 	union child_device_config *child_dev;
 
 	struct ddi_vbt_port_info ddi_port_info[I915_MAX_PORTS];
+	struct sdvo_device_mapping sdvo_mappings[2];
 };
 
 enum intel_ddb_partitioning {
@@ -1706,7 +1682,7 @@ struct i915_wa_reg {
 struct i915_workarounds {
 	struct i915_wa_reg reg[I915_MAX_WA_REGS];
 	u32 count;
-	u32 hw_whitelist_count[I915_NUM_RINGS];
+	u32 hw_whitelist_count[I915_NUM_ENGINES];
 };
 
 struct i915_virtual_gpu {
@@ -1719,7 +1695,7 @@ struct i915_execbuffer_params {
 	uint32_t                        dispatch_flags;
 	uint32_t                        args_batch_start_offset;
 	uint64_t                        batch_obj_vm_offset;
-	struct intel_engine_cs          *ring;
+	struct intel_engine_cs *engine;
 	struct drm_i915_gem_object      *batch_obj;
 	struct intel_context            *ctx;
 	struct drm_i915_gem_request     *request;
@@ -1771,7 +1747,7 @@ struct drm_i915_private {
 	wait_queue_head_t gmbus_wait_queue;
 
 	struct pci_dev *bridge_dev;
-	struct intel_engine_cs ring[I915_NUM_RINGS];
+	struct intel_engine_cs engine[I915_NUM_ENGINES];
 	struct drm_i915_gem_object *semaphore_obj;
 	uint32_t last_seqno, next_seqno;
 
@@ -1829,6 +1805,7 @@ struct drm_i915_private {
 	unsigned int skl_boot_cdclk;
 	unsigned int cdclk_freq, max_cdclk_freq, atomic_cdclk_freq;
 	unsigned int max_dotclk_freq;
+	unsigned int rawclk_freq;
 	unsigned int hpll_freq;
 	unsigned int czclk_freq;
 
@@ -1855,7 +1832,7 @@ struct drm_i915_private {
 	struct drm_atomic_state *modeset_restore_state;
 
 	struct list_head vm_list; /* Global list of all address spaces */
-	struct i915_gtt gtt; /* VM representing the global address space */
+	struct i915_ggtt ggtt; /* VM representing the global address space */
 
 	struct i915_gem_mm mm;
 	DECLARE_HASHTABLE(mm_structs, 7);
@@ -1863,8 +1840,6 @@ struct drm_i915_private {
 
 	/* Kernel Modesetting */
 
-	struct sdvo_device_mapping sdvo_mappings[2];
-
 	struct drm_crtc *plane_to_crtc_mapping[I915_MAX_PIPES];
 	struct drm_crtc *pipe_to_crtc_mapping[I915_MAX_PIPES];
 	wait_queue_head_t pending_flip_queue;
@@ -1876,6 +1851,14 @@ struct drm_i915_private {
 	/* dpll and cdclk state is protected by connection_mutex */
 	int num_shared_dpll;
 	struct intel_shared_dpll shared_dplls[I915_NUM_PLLS];
+	const struct intel_dpll_mgr *dpll_mgr;
+
+	/*
+	 * dpll_lock serializes intel_{prepare,enable,disable}_shared_dpll.
+	 * Must be global rather than per dpll, because on some platforms
+	 * plls share registers.
+	 */
+	struct mutex dpll_lock;
 
 	unsigned int active_crtcs;
 	unsigned int min_pixclk[I915_MAX_PIPES];
@@ -1884,9 +1867,6 @@ struct drm_i915_private {
 
 	struct i915_workarounds workarounds;
 
-	/* Reclocking support */
-	bool render_reclock_avail;
-
 	struct i915_frontbuffer_tracking fb_tracking;
 
 	u16 orig_clock;
@@ -1896,7 +1876,7 @@ struct drm_i915_private {
 	struct intel_l3_parity l3_parity;
 
 	/* Cannot be determined by PCIID. You must always read a register. */
-	size_t ellc_size;
+	u32 edram_cap;
 
 	/* gen6+ rps state */
 	struct intel_gen6_power_mgmt rps;
@@ -1936,7 +1916,15 @@ struct drm_i915_private {
 
 	u32 fdi_rx_config;
 
+	/* Shadow for DISPLAY_PHY_CONTROL which can't be safely read */
 	u32 chv_phy_control;
+	/*
+	 * Shadows for CHV DPLL_MD regs to keep the state
+	 * checker somewhat working in the presence hardware
+	 * crappiness (can't read out DPLL_MD for pipes B & C).
+	 */
+	u32 chv_dpll_md[I915_MAX_PIPES];
+	u32 bxt_phy_grc;
 
 	u32 suspend_count;
 	bool suspended_to_idle;
@@ -1980,6 +1968,13 @@ struct drm_i915_private {
 		};
 
 		uint8_t max_level;
+
+		/*
+		 * Should be held around atomic WM register writing; also
+		 * protects * intel_crtc->wm.active and
+		 * cstate->wm.need_postvbl_update.
+		 */
+		struct mutex wm_mutex;
 	} wm;
 
 	struct i915_runtime_pm pm;
@@ -1989,15 +1984,13 @@ struct drm_i915_private {
 		int (*execbuf_submit)(struct i915_execbuffer_params *params,
 				      struct drm_i915_gem_execbuffer2 *args,
 				      struct list_head *vmas);
-		int (*init_rings)(struct drm_device *dev);
-		void (*cleanup_ring)(struct intel_engine_cs *ring);
-		void (*stop_ring)(struct intel_engine_cs *ring);
+		int (*init_engines)(struct drm_device *dev);
+		void (*cleanup_engine)(struct intel_engine_cs *engine);
+		void (*stop_engine)(struct intel_engine_cs *engine);
 	} gt;
 
 	struct intel_context *kernel_context;
 
-	bool edp_low_vswing;
-
 	/* perform PHY state sanity checks? */
 	bool chv_phy_assert[2];
 
@@ -2024,10 +2017,28 @@ static inline struct drm_i915_private *guc_to_i915(struct intel_guc *guc)
 	return container_of(guc, struct drm_i915_private, guc);
 }
 
-/* Iterate over initialised rings */
-#define for_each_ring(ring__, dev_priv__, i__) \
-	for ((i__) = 0; (i__) < I915_NUM_RINGS; (i__)++) \
-		for_each_if ((((ring__) = &(dev_priv__)->ring[(i__)]), intel_ring_initialized((ring__))))
+/* Simple iterator over all initialised engines */
+#define for_each_engine(engine__, dev_priv__) \
+	for ((engine__) = &(dev_priv__)->engine[0]; \
+	     (engine__) < &(dev_priv__)->engine[I915_NUM_ENGINES]; \
+	     (engine__)++) \
+		for_each_if (intel_engine_initialized(engine__))
+
+/* Iterator with engine_id */
+#define for_each_engine_id(engine__, dev_priv__, id__) \
+	for ((engine__) = &(dev_priv__)->engine[0], (id__) = 0; \
+	     (engine__) < &(dev_priv__)->engine[I915_NUM_ENGINES]; \
+	     (engine__)++) \
+		for_each_if (((id__) = (engine__)->id, \
+			      intel_engine_initialized(engine__)))
+
+/* Iterator over subset of engines selected by mask */
+#define for_each_engine_masked(engine__, dev_priv__, mask__) \
+	for ((engine__) = &(dev_priv__)->engine[0]; \
+	     (engine__) < &(dev_priv__)->engine[I915_NUM_ENGINES]; \
+	     (engine__)++) \
+		for_each_if (((mask__) & intel_engine_flag(engine__)) && \
+			     intel_engine_initialized(engine__))
 
 enum hdmi_force_audio {
 	HDMI_AUDIO_OFF_DVI = -2,	/* no aux data for HDMI-DVI converter */
@@ -2097,7 +2108,7 @@ struct drm_i915_gem_object {
 	struct drm_mm_node *stolen;
 	struct list_head global_list;
 
-	struct list_head ring_list[I915_NUM_RINGS];
+	struct list_head engine_list[I915_NUM_ENGINES];
 	/** Used in execbuf to temporarily hold a ref */
 	struct list_head obj_exec_link;
 
@@ -2108,7 +2119,7 @@ struct drm_i915_gem_object {
 	 * rendering and so a non-zero seqno), and is not set if it i s on
 	 * inactive (ready to be unbound) list.
 	 */
-	unsigned int active:I915_NUM_RINGS;
+	unsigned int active:I915_NUM_ENGINES;
 
 	/**
 	 * This is set if the object has been written to since last bound
@@ -2172,10 +2183,7 @@ struct drm_i915_gem_object {
 		struct scatterlist *sg;
 		int last;
 	} get_page;
-
-	/* prime dma-buf support */
-	void *dma_buf_vmapping;
-	int vmapping_count;
+	void *mapping;
 
 	/** Breadcrumb of last rendering to the buffer.
 	 * There can only be one writer, but we allow for multiple readers.
@@ -2187,7 +2195,7 @@ struct drm_i915_gem_object {
 	 * read request. This allows for the CPU to read from an active
 	 * buffer by only waiting for the write to complete.
 	 * */
-	struct drm_i915_gem_request *last_read_req[I915_NUM_RINGS];
+	struct drm_i915_gem_request *last_read_req[I915_NUM_ENGINES];
 	struct drm_i915_gem_request *last_write_req;
 	/** Breadcrumb of last fenced GPU access to the buffer. */
 	struct drm_i915_gem_request *last_fenced_req;
@@ -2242,7 +2250,8 @@ struct drm_i915_gem_request {
 
 	/** On Which ring this request was generated */
 	struct drm_i915_private *i915;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
+	unsigned reset_counter;
 
 	 /** GEM sequence number associated with the previous request,
 	  * when the HWS breadcrumb is equal to this the GPU is processing
@@ -2323,7 +2332,6 @@ struct drm_i915_gem_request {
 struct drm_i915_gem_request * __must_check
 i915_gem_request_alloc(struct intel_engine_cs *engine,
 		       struct intel_context *ctx);
-void i915_gem_request_cancel(struct drm_i915_gem_request *req);
 void i915_gem_request_free(struct kref *req_ref);
 int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
 				   struct drm_file *file);
@@ -2335,9 +2343,9 @@ i915_gem_request_get_seqno(struct drm_i915_gem_request *req)
 }
 
 static inline struct intel_engine_cs *
-i915_gem_request_get_ring(struct drm_i915_gem_request *req)
+i915_gem_request_get_engine(struct drm_i915_gem_request *req)
 {
-	return req ? req->ring : NULL;
+	return req ? req->engine : NULL;
 }
 
 static inline struct drm_i915_gem_request *
@@ -2351,7 +2359,7 @@ i915_gem_request_reference(struct drm_i915_gem_request *req)
 static inline void
 i915_gem_request_unreference(struct drm_i915_gem_request *req)
 {
-	WARN_ON(!mutex_is_locked(&req->ring->dev->struct_mutex));
+	WARN_ON(!mutex_is_locked(&req->engine->dev->struct_mutex));
 	kref_put(&req->ref, i915_gem_request_free);
 }
 
@@ -2363,7 +2371,7 @@ i915_gem_request_unreference__unlocked(struct drm_i915_gem_request *req)
 	if (!req)
 		return;
 
-	dev = req->ring->dev;
+	dev = req->engine->dev;
 	if (kref_put_mutex(&req->ref, i915_gem_request_free, &dev->struct_mutex))
 		mutex_unlock(&dev->struct_mutex);
 }
@@ -2493,6 +2501,7 @@ struct drm_i915_cmd_table {
 	__p; \
 })
 #define INTEL_INFO(p) 	(&__I915__(p)->info)
+#define INTEL_GEN(p)	(INTEL_INFO(p)->gen)
 #define INTEL_DEVID(p)	(INTEL_INFO(p)->device_id)
 #define INTEL_REVID(p)	(__I915__(p)->dev->pdev->revision)
 
@@ -2611,13 +2620,17 @@ struct drm_i915_cmd_table {
 #define BLT_RING		(1<<BCS)
 #define VEBOX_RING		(1<<VECS)
 #define BSD2_RING		(1<<VCS2)
+#define ALL_ENGINES		(~0)
+
 #define HAS_BSD(dev)		(INTEL_INFO(dev)->ring_mask & BSD_RING)
 #define HAS_BSD2(dev)		(INTEL_INFO(dev)->ring_mask & BSD2_RING)
 #define HAS_BLT(dev)		(INTEL_INFO(dev)->ring_mask & BLT_RING)
 #define HAS_VEBOX(dev)		(INTEL_INFO(dev)->ring_mask & VEBOX_RING)
 #define HAS_LLC(dev)		(INTEL_INFO(dev)->has_llc)
+#define HAS_SNOOP(dev)		(INTEL_INFO(dev)->has_snoop)
+#define HAS_EDRAM(dev)		(__I915__(dev)->edram_cap & EDRAM_ENABLED)
 #define HAS_WT(dev)		((IS_HASWELL(dev) || IS_BROADWELL(dev)) && \
-				 __I915__(dev)->ellc_size)
+				 HAS_EDRAM(dev))
 #define I915_NEED_GFX_HWS(dev)	(INTEL_INFO(dev)->need_gfx_hws)
 
 #define HAS_HW_CONTEXTS(dev)	(INTEL_INFO(dev)->gen >= 6)
@@ -2671,7 +2684,7 @@ struct drm_i915_cmd_table {
 #define HAS_RUNTIME_PM(dev)	(IS_GEN6(dev) || IS_HASWELL(dev) || \
 				 IS_BROADWELL(dev) || IS_VALLEYVIEW(dev) || \
 				 IS_CHERRYVIEW(dev) || IS_SKYLAKE(dev) || \
-				 IS_KABYLAKE(dev))
+				 IS_KABYLAKE(dev) || IS_BROXTON(dev))
 #define HAS_RC6(dev)		(INTEL_INFO(dev)->gen >= 6)
 #define HAS_RC6p(dev)		(INTEL_INFO(dev)->gen == 6 || IS_IVYBRIDGE(dev))
 
@@ -2696,6 +2709,7 @@ struct drm_i915_cmd_table {
 #define INTEL_PCH_SPT_DEVICE_ID_TYPE		0xA100
 #define INTEL_PCH_SPT_LP_DEVICE_ID_TYPE		0x9D00
 #define INTEL_PCH_P2X_DEVICE_ID_TYPE		0x7100
+#define INTEL_PCH_P3X_DEVICE_ID_TYPE		0x7000
 #define INTEL_PCH_QEMU_DEVICE_ID_TYPE		0x2900 /* qemu q35 has 2918 */
 
 #define INTEL_PCH_TYPE(dev) (__I915__(dev)->pch_type)
@@ -2727,6 +2741,13 @@ extern int i915_suspend_switcheroo(struct drm_device *dev, pm_message_t state);
 extern int i915_resume_switcheroo(struct drm_device *dev);
 
 /* i915_dma.c */
+void __printf(3, 4)
+__i915_printk(struct drm_i915_private *dev_priv, const char *level,
+	      const char *fmt, ...);
+
+#define i915_report_error(dev_priv, fmt, ...)				   \
+	__i915_printk(dev_priv, KERN_ERR, fmt, ##__VA_ARGS__)
+
 extern int i915_driver_load(struct drm_device *, unsigned long flags);
 extern int i915_driver_unload(struct drm_device *);
 extern int i915_driver_open(struct drm_device *dev, struct drm_file *file);
@@ -2739,9 +2760,11 @@ extern void i915_driver_postclose(struct drm_device *dev,
 extern long i915_compat_ioctl(struct file *filp, unsigned int cmd,
 			      unsigned long arg);
 #endif
-extern int intel_gpu_reset(struct drm_device *dev);
+extern int intel_gpu_reset(struct drm_device *dev, u32 engine_mask);
 extern bool intel_has_gpu_reset(struct drm_device *dev);
 extern int i915_reset(struct drm_device *dev);
+extern int intel_guc_reset(struct drm_i915_private *dev_priv);
+extern void intel_engine_init_hangcheck(struct intel_engine_cs *engine);
 extern unsigned long i915_chipset_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_mch_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_gfx_val(struct drm_i915_private *dev_priv);
@@ -2758,7 +2781,7 @@ bool intel_hpd_pin_to_port(enum hpd_pin pin, enum port *port);
 /* i915_irq.c */
 void i915_queue_hangcheck(struct drm_device *dev);
 __printf(3, 4)
-void i915_handle_error(struct drm_device *dev, bool wedged,
+void i915_handle_error(struct drm_device *dev, u32 engine_mask,
 		       const char *fmt, ...);
 
 extern void intel_irq_init(struct drm_i915_private *dev_priv);
@@ -2785,6 +2808,8 @@ void intel_uncore_forcewake_get__locked(struct drm_i915_private *dev_priv,
 					enum forcewake_domains domains);
 void intel_uncore_forcewake_put__locked(struct drm_i915_private *dev_priv,
 					enum forcewake_domains domains);
+u64 intel_uncore_edram_size(struct drm_i915_private *dev_priv);
+
 void assert_forcewakes_inactive(struct drm_i915_private *dev_priv);
 static inline bool intel_vgpu_active(struct drm_device *dev)
 {
@@ -2863,7 +2888,6 @@ int i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
 			     struct drm_file *file_priv);
 void i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 					struct drm_i915_gem_request *req);
-void i915_gem_execbuffer_retire_commands(struct i915_execbuffer_params *params);
 int i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 				   struct drm_i915_gem_execbuffer2 *args,
 				   struct list_head *vmas);
@@ -2894,6 +2918,7 @@ int i915_gem_wait_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 void i915_gem_load_init(struct drm_device *dev);
 void i915_gem_load_cleanup(struct drm_device *dev);
+void i915_gem_load_init_fences(struct drm_i915_private *dev_priv);
 void *i915_gem_object_alloc(struct drm_device *dev);
 void i915_gem_object_free(struct drm_i915_gem_object *obj);
 void i915_gem_object_init(struct drm_i915_gem_object *obj,
@@ -2978,12 +3003,46 @@ static inline void i915_gem_object_pin_pages(struct drm_i915_gem_object *obj)
 	BUG_ON(obj->pages == NULL);
 	obj->pages_pin_count++;
 }
+
 static inline void i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj)
 {
 	BUG_ON(obj->pages_pin_count == 0);
 	obj->pages_pin_count--;
 }
 
+/**
+ * i915_gem_object_pin_map - return a contiguous mapping of the entire object
+ * @obj - the object to map into kernel address space
+ *
+ * Calls i915_gem_object_pin_pages() to prevent reaping of the object's
+ * pages and then returns a contiguous mapping of the backing storage into
+ * the kernel address space.
+ *
+ * The caller must hold the struct_mutex, and is responsible for calling
+ * i915_gem_object_unpin_map() when the mapping is no longer required.
+ *
+ * Returns the pointer through which to access the mapped object, or an
+ * ERR_PTR() on error.
+ */
+void *__must_check i915_gem_object_pin_map(struct drm_i915_gem_object *obj);
+
+/**
+ * i915_gem_object_unpin_map - releases an earlier mapping
+ * @obj - the object to unmap
+ *
+ * After pinning the object and mapping its pages, once you are finished
+ * with your access, call i915_gem_object_unpin_map() to release the pin
+ * upon the mapping. Once the pin count reaches zero, that mapping may be
+ * removed.
+ *
+ * The caller must hold the struct_mutex.
+ */
+static inline void i915_gem_object_unpin_map(struct drm_i915_gem_object *obj)
+{
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
+	i915_gem_object_unpin_pages(obj);
+}
+
 int __must_check i915_mutex_lock_interruptible(struct drm_device *dev);
 int i915_gem_object_sync(struct drm_i915_gem_object *obj,
 			 struct intel_engine_cs *to,
@@ -3007,42 +3066,68 @@ i915_seqno_passed(uint32_t seq1, uint32_t seq2)
 static inline bool i915_gem_request_started(struct drm_i915_gem_request *req,
 					   bool lazy_coherency)
 {
-	u32 seqno = req->ring->get_seqno(req->ring, lazy_coherency);
-	return i915_seqno_passed(seqno, req->previous_seqno);
+	if (!lazy_coherency && req->engine->irq_seqno_barrier)
+		req->engine->irq_seqno_barrier(req->engine);
+	return i915_seqno_passed(req->engine->get_seqno(req->engine),
+				 req->previous_seqno);
 }
 
 static inline bool i915_gem_request_completed(struct drm_i915_gem_request *req,
 					      bool lazy_coherency)
 {
-	u32 seqno = req->ring->get_seqno(req->ring, lazy_coherency);
-	return i915_seqno_passed(seqno, req->seqno);
+	if (!lazy_coherency && req->engine->irq_seqno_barrier)
+		req->engine->irq_seqno_barrier(req->engine);
+	return i915_seqno_passed(req->engine->get_seqno(req->engine),
+				 req->seqno);
 }
 
 int __must_check i915_gem_get_seqno(struct drm_device *dev, u32 *seqno);
 int __must_check i915_gem_set_seqno(struct drm_device *dev, u32 seqno);
 
 struct drm_i915_gem_request *
-i915_gem_find_active_request(struct intel_engine_cs *ring);
+i915_gem_find_active_request(struct intel_engine_cs *engine);
 
 bool i915_gem_retire_requests(struct drm_device *dev);
-void i915_gem_retire_requests_ring(struct intel_engine_cs *ring);
-int __must_check i915_gem_check_wedge(struct i915_gpu_error *error,
-				      bool interruptible);
+void i915_gem_retire_requests_ring(struct intel_engine_cs *engine);
+
+static inline u32 i915_reset_counter(struct i915_gpu_error *error)
+{
+	return atomic_read(&error->reset_counter);
+}
+
+static inline bool __i915_reset_in_progress(u32 reset)
+{
+	return unlikely(reset & I915_RESET_IN_PROGRESS_FLAG);
+}
+
+static inline bool __i915_reset_in_progress_or_wedged(u32 reset)
+{
+	return unlikely(reset & (I915_RESET_IN_PROGRESS_FLAG | I915_WEDGED));
+}
+
+static inline bool __i915_terminally_wedged(u32 reset)
+{
+	return unlikely(reset & I915_WEDGED);
+}
 
 static inline bool i915_reset_in_progress(struct i915_gpu_error *error)
 {
-	return unlikely(atomic_read(&error->reset_counter)
-			& (I915_RESET_IN_PROGRESS_FLAG | I915_WEDGED));
+	return __i915_reset_in_progress(i915_reset_counter(error));
+}
+
+static inline bool i915_reset_in_progress_or_wedged(struct i915_gpu_error *error)
+{
+	return __i915_reset_in_progress_or_wedged(i915_reset_counter(error));
 }
 
 static inline bool i915_terminally_wedged(struct i915_gpu_error *error)
 {
-	return atomic_read(&error->reset_counter) & I915_WEDGED;
+	return __i915_terminally_wedged(i915_reset_counter(error));
 }
 
 static inline u32 i915_reset_count(struct i915_gpu_error *error)
 {
-	return ((atomic_read(&error->reset_counter) & ~I915_WEDGED) + 1) / 2;
+	return ((i915_reset_counter(error) & ~I915_WEDGED) + 1) / 2;
 }
 
 static inline bool i915_stop_ring_allow_ban(struct drm_i915_private *dev_priv)
@@ -3060,11 +3145,11 @@ static inline bool i915_stop_ring_allow_warn(struct drm_i915_private *dev_priv)
 void i915_gem_reset(struct drm_device *dev);
 bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, bool force);
 int __must_check i915_gem_init(struct drm_device *dev);
-int i915_gem_init_rings(struct drm_device *dev);
+int i915_gem_init_engines(struct drm_device *dev);
 int __must_check i915_gem_init_hw(struct drm_device *dev);
 int i915_gem_l3_remap(struct drm_i915_gem_request *req, int slice);
 void i915_gem_init_swizzling(struct drm_device *dev);
-void i915_gem_cleanup_ringbuffer(struct drm_device *dev);
+void i915_gem_cleanup_engines(struct drm_device *dev);
 int __must_check i915_gpu_idle(struct drm_device *dev);
 int __must_check i915_gem_suspend(struct drm_device *dev);
 void __i915_add_request(struct drm_i915_gem_request *req,
@@ -3075,7 +3160,6 @@ void __i915_add_request(struct drm_i915_gem_request *req,
 #define i915_add_request_no_flush(req) \
 	__i915_add_request(req, NULL, false)
 int __i915_wait_request(struct drm_i915_gem_request *req,
-			unsigned reset_counter,
 			bool interruptible,
 			s64 *timeout,
 			struct intel_rps_client *rps);
@@ -3155,13 +3239,9 @@ i915_gem_obj_to_ggtt(struct drm_i915_gem_object *obj)
 bool i915_gem_obj_is_pinned(struct drm_i915_gem_object *obj);
 
 /* Some GGTT VM helpers */
-#define i915_obj_to_ggtt(obj) \
-	(&((struct drm_i915_private *)(obj)->base.dev->dev_private)->gtt.base)
-
 static inline struct i915_hw_ppgtt *
 i915_vm_to_ppgtt(struct i915_address_space *vm)
 {
-	WARN_ON(i915_is_ggtt(vm));
 	return container_of(vm, struct i915_hw_ppgtt, base);
 }
 
@@ -3174,7 +3254,10 @@ static inline bool i915_gem_obj_ggtt_bound(struct drm_i915_gem_object *obj)
 static inline unsigned long
 i915_gem_obj_ggtt_size(struct drm_i915_gem_object *obj)
 {
-	return i915_gem_obj_size(obj, i915_obj_to_ggtt(obj));
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+
+	return i915_gem_obj_size(obj, &ggtt->base);
 }
 
 static inline int __must_check
@@ -3182,7 +3265,10 @@ i915_gem_obj_ggtt_pin(struct drm_i915_gem_object *obj,
 		      uint32_t alignment,
 		      unsigned flags)
 {
-	return i915_gem_object_pin(obj, i915_obj_to_ggtt(obj),
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+
+	return i915_gem_object_pin(obj, &ggtt->base,
 				   alignment, flags | PIN_GLOBAL);
 }
 
@@ -3297,6 +3383,7 @@ unsigned long i915_gem_shrink(struct drm_i915_private *dev_priv,
 #define I915_SHRINK_UNBOUND 0x2
 #define I915_SHRINK_BOUND 0x4
 #define I915_SHRINK_ACTIVE 0x8
+#define I915_SHRINK_VMAPS 0x10
 unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
 void i915_gem_shrinker_init(struct drm_i915_private *dev_priv);
 void i915_gem_shrinker_cleanup(struct drm_i915_private *dev_priv);
@@ -3343,7 +3430,7 @@ static inline void i915_error_state_buf_release(
 {
 	kfree(eb->buf);
 }
-void i915_capture_error_state(struct drm_device *dev, bool wedge,
+void i915_capture_error_state(struct drm_device *dev, u32 engine_mask,
 			      const char *error_msg);
 void i915_error_state_get(struct drm_device *dev,
 			  struct i915_error_state_file_priv *error_priv);
@@ -3355,10 +3442,10 @@ const char *i915_cache_level_str(struct drm_i915_private *i915, int type);
 
 /* i915_cmd_parser.c */
 int i915_cmd_parser_get_version(void);
-int i915_cmd_parser_init_ring(struct intel_engine_cs *ring);
-void i915_cmd_parser_fini_ring(struct intel_engine_cs *ring);
-bool i915_needs_cmd_parser(struct intel_engine_cs *ring);
-int i915_parse_cmds(struct intel_engine_cs *ring,
+int i915_cmd_parser_init_ring(struct intel_engine_cs *engine);
+void i915_cmd_parser_fini_ring(struct intel_engine_cs *engine);
+bool i915_needs_cmd_parser(struct intel_engine_cs *engine);
+int i915_parse_cmds(struct intel_engine_cs *engine,
 		    struct drm_i915_gem_object *batch_obj,
 		    struct drm_i915_gem_object *shadow_batch_obj,
 		    u32 batch_start_offset,
@@ -3392,6 +3479,12 @@ extern void intel_i2c_reset(struct drm_device *dev);
 /* intel_bios.c */
 int intel_bios_init(struct drm_i915_private *dev_priv);
 bool intel_bios_is_valid_vbt(const void *buf, size_t size);
+bool intel_bios_is_tv_present(struct drm_i915_private *dev_priv);
+bool intel_bios_is_lvds_present(struct drm_i915_private *dev_priv, u8 *i2c_pin);
+bool intel_bios_is_port_edp(struct drm_i915_private *dev_priv, enum port port);
+bool intel_bios_is_dsi_present(struct drm_i915_private *dev_priv, enum port *port);
+bool intel_bios_is_port_hpd_inverted(struct drm_i915_private *dev_priv,
+				     enum port port);
 
 /* intel_opregion.c */
 #ifdef CONFIG_ACPI
@@ -3403,6 +3496,7 @@ extern int intel_opregion_notify_encoder(struct intel_encoder *intel_encoder,
 					 bool enable);
 extern int intel_opregion_notify_adapter(struct drm_device *dev,
 					 pci_power_t state);
+extern int intel_opregion_get_panel_type(struct drm_device *dev);
 #else
 static inline int intel_opregion_setup(struct drm_device *dev) { return 0; }
 static inline void intel_opregion_init(struct drm_device *dev) { return; }
@@ -3418,6 +3512,10 @@ intel_opregion_notify_adapter(struct drm_device *dev, pci_power_t state)
 {
 	return 0;
 }
+static inline int intel_opregion_get_panel_type(struct drm_device *dev)
+{
+	return -ENODEV;
+}
 #endif
 
 /* intel_acpi.c */
@@ -3624,11 +3722,11 @@ wait_remaining_ms_from_jiffies(unsigned long timestamp_jiffies, int to_wait_ms)
 	}
 }
 
-static inline void i915_trace_irq_get(struct intel_engine_cs *ring,
+static inline void i915_trace_irq_get(struct intel_engine_cs *engine,
 				      struct drm_i915_gem_request *req)
 {
-	if (ring->trace_irq_req == NULL && ring->irq_get(ring))
-		i915_gem_request_assign(&ring->trace_irq_req, req);
+	if (engine->trace_irq_req == NULL && engine->irq_get(engine))
+		i915_gem_request_assign(&engine->trace_irq_req, req);
 }
 
 #endif
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 233adc31ef0c..94bbc4314ac5 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -32,14 +32,13 @@
 #include "i915_vgpu.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include "intel_mocs.h"
 #include <linux/shmem_fs.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 
-#define RQ_BUG_ON(expr)
-
 static void i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj);
 static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj);
 static void
@@ -85,9 +84,7 @@ i915_gem_wait_for_error(struct i915_gpu_error *error)
 {
 	int ret;
 
-#define EXIT_COND (!i915_reset_in_progress(error) || \
-		   i915_terminally_wedged(error))
-	if (EXIT_COND)
+	if (!i915_reset_in_progress(error))
 		return 0;
 
 	/*
@@ -96,17 +93,16 @@ i915_gem_wait_for_error(struct i915_gpu_error *error)
 	 * we should simply try to bail out and fail as gracefully as possible.
 	 */
 	ret = wait_event_interruptible_timeout(error->reset_queue,
-					       EXIT_COND,
+					       !i915_reset_in_progress(error),
 					       10*HZ);
 	if (ret == 0) {
 		DRM_ERROR("Timed out waiting for the gpu reset to complete\n");
 		return -EIO;
 	} else if (ret < 0) {
 		return ret;
+	} else {
+		return 0;
 	}
-#undef EXIT_COND
-
-	return 0;
 }
 
 int i915_mutex_lock_interruptible(struct drm_device *dev)
@@ -130,9 +126,9 @@ int
 i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 			    struct drm_file *file)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_get_aperture *args = data;
-	struct i915_gtt *ggtt = &dev_priv->gtt;
 	struct i915_vma *vma;
 	size_t pinned;
 
@@ -146,7 +142,7 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 			pinned += vma->node.size;
 	mutex_unlock(&dev->struct_mutex);
 
-	args->aper_size = dev_priv->gtt.base.total;
+	args->aper_size = ggtt->base.total;
 	args->aper_available_size = args->aper_size - pinned;
 
 	return 0;
@@ -211,11 +207,10 @@ i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj)
 	BUG_ON(obj->madv == __I915_MADV_PURGED);
 
 	ret = i915_gem_object_set_to_cpu_domain(obj, true);
-	if (ret) {
+	if (WARN_ON(ret)) {
 		/* In the event of a disaster, abandon all caches and
 		 * hope for the best.
 		 */
-		WARN_ON(ret != -EIO);
 		obj->base.read_domains = obj->base.write_domain = I915_GEM_DOMAIN_CPU;
 	}
 
@@ -700,7 +695,7 @@ i915_gem_pread_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -765,7 +760,8 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev,
 			 struct drm_i915_gem_pwrite *args,
 			 struct drm_file *file)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	ssize_t remain;
 	loff_t offset, page_base;
 	char __user *user_data;
@@ -807,7 +803,7 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev,
 		 * source page isn't available.  Return the error and we'll
 		 * retry in the slow path.
 		 */
-		if (fast_user_write(dev_priv->gtt.mappable, page_base,
+		if (fast_user_write(ggtt->mappable, page_base,
 				    page_offset, user_data, page_length)) {
 			ret = -EFAULT;
 			goto out_flush;
@@ -1053,7 +1049,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto put_rpm;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -1109,27 +1105,19 @@ put_rpm:
 	return ret;
 }
 
-int
-i915_gem_check_wedge(struct i915_gpu_error *error,
-		     bool interruptible)
+static int
+i915_gem_check_wedge(unsigned reset_counter, bool interruptible)
 {
-	if (i915_reset_in_progress(error)) {
+	if (__i915_terminally_wedged(reset_counter))
+		return -EIO;
+
+	if (__i915_reset_in_progress(reset_counter)) {
 		/* Non-interruptible callers can't handle -EAGAIN, hence return
 		 * -EIO unconditionally for these. */
 		if (!interruptible)
 			return -EIO;
 
-		/* Recovery complete, but the reset failed ... */
-		if (i915_terminally_wedged(error))
-			return -EIO;
-
-		/*
-		 * Check if GPU Reset is in progress - we need intel_ring_begin
-		 * to work properly to reinit the hw state while the gpu is
-		 * still marked as reset-in-progress. Handle this with a flag.
-		 */
-		if (!error->reload_in_reset)
-			return -EAGAIN;
+		return -EAGAIN;
 	}
 
 	return 0;
@@ -1141,9 +1129,9 @@ static void fake_irq(unsigned long data)
 }
 
 static bool missed_irq(struct drm_i915_private *dev_priv,
-		       struct intel_engine_cs *ring)
+		       struct intel_engine_cs *engine)
 {
-	return test_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings);
+	return test_bit(engine->id, &dev_priv->gpu_error.missed_irq_rings);
 }
 
 static unsigned long local_clock_us(unsigned *cpu)
@@ -1193,7 +1181,7 @@ static int __i915_spin_request(struct drm_i915_gem_request *req, int state)
 	 * takes to sleep on a request, on the order of a microsecond.
 	 */
 
-	if (req->ring->irq_refcount)
+	if (req->engine->irq_refcount)
 		return -EBUSY;
 
 	/* Only spin if we know the GPU is processing this request */
@@ -1223,7 +1211,6 @@ static int __i915_spin_request(struct drm_i915_gem_request *req, int state)
 /**
  * __i915_wait_request - wait until execution of request has finished
  * @req: duh!
- * @reset_counter: reset sequence associated with the given request
  * @interruptible: do an interruptible wait (normally yes)
  * @timeout: in - how long to wait (NULL forever); out - how much time remaining
  *
@@ -1238,16 +1225,15 @@ static int __i915_spin_request(struct drm_i915_gem_request *req, int state)
  * errno with remaining time filled in timeout argument.
  */
 int __i915_wait_request(struct drm_i915_gem_request *req,
-			unsigned reset_counter,
 			bool interruptible,
 			s64 *timeout,
 			struct intel_rps_client *rps)
 {
-	struct intel_engine_cs *ring = i915_gem_request_get_ring(req);
-	struct drm_device *dev = ring->dev;
+	struct intel_engine_cs *engine = i915_gem_request_get_engine(req);
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	const bool irq_test_in_progress =
-		ACCESS_ONCE(dev_priv->gpu_error.test_irq_rings) & intel_ring_flag(ring);
+		ACCESS_ONCE(dev_priv->gpu_error.test_irq_rings) & intel_engine_flag(engine);
 	int state = interruptible ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE;
 	DEFINE_WAIT(wait);
 	unsigned long timeout_expire;
@@ -1288,7 +1274,7 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 	if (ret == 0)
 		goto out;
 
-	if (!irq_test_in_progress && WARN_ON(!ring->irq_get(ring))) {
+	if (!irq_test_in_progress && WARN_ON(!engine->irq_get(engine))) {
 		ret = -ENODEV;
 		goto out;
 	}
@@ -1296,16 +1282,17 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 	for (;;) {
 		struct timer_list timer;
 
-		prepare_to_wait(&ring->irq_queue, &wait, state);
+		prepare_to_wait(&engine->irq_queue, &wait, state);
 
 		/* We need to check whether any gpu reset happened in between
-		 * the caller grabbing the seqno and now ... */
-		if (reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter)) {
-			/* ... but upgrade the -EAGAIN to an -EIO if the gpu
-			 * is truely gone. */
-			ret = i915_gem_check_wedge(&dev_priv->gpu_error, interruptible);
-			if (ret == 0)
-				ret = -EAGAIN;
+		 * the request being submitted and now. If a reset has occurred,
+		 * the request is effectively complete (we either are in the
+		 * process of or have discarded the rendering and completely
+		 * reset the GPU. The results of the request are lost and we
+		 * are free to continue on with the original operation.
+		 */
+		if (req->reset_counter != i915_reset_counter(&dev_priv->gpu_error)) {
+			ret = 0;
 			break;
 		}
 
@@ -1325,11 +1312,11 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 		}
 
 		timer.function = NULL;
-		if (timeout || missed_irq(dev_priv, ring)) {
+		if (timeout || missed_irq(dev_priv, engine)) {
 			unsigned long expire;
 
 			setup_timer_on_stack(&timer, fake_irq, (unsigned long)current);
-			expire = missed_irq(dev_priv, ring) ? jiffies + 1 : timeout_expire;
+			expire = missed_irq(dev_priv, engine) ? jiffies + 1 : timeout_expire;
 			mod_timer(&timer, expire);
 		}
 
@@ -1341,9 +1328,9 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 		}
 	}
 	if (!irq_test_in_progress)
-		ring->irq_put(ring);
+		engine->irq_put(engine);
 
-	finish_wait(&ring->irq_queue, &wait);
+	finish_wait(&engine->irq_queue, &wait);
 
 out:
 	trace_i915_gem_request_wait_end(req);
@@ -1370,7 +1357,6 @@ out:
 int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
 				   struct drm_file *file)
 {
-	struct drm_i915_private *dev_private;
 	struct drm_i915_file_private *file_priv;
 
 	WARN_ON(!req || !file || req->file_priv);
@@ -1381,7 +1367,6 @@ int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
 	if (req->file_priv)
 		return -EINVAL;
 
-	dev_private = req->ring->dev->dev_private;
 	file_priv = file->driver_priv;
 
 	spin_lock(&file_priv->mm.lock);
@@ -1434,7 +1419,7 @@ static void i915_gem_request_retire(struct drm_i915_gem_request *request)
 static void
 __i915_gem_request_retire__upto(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct drm_i915_gem_request *tmp;
 
 	lockdep_assert_held(&engine->dev->struct_mutex);
@@ -1459,26 +1444,15 @@ __i915_gem_request_retire__upto(struct drm_i915_gem_request *req)
 int
 i915_wait_request(struct drm_i915_gem_request *req)
 {
-	struct drm_device *dev;
-	struct drm_i915_private *dev_priv;
+	struct drm_i915_private *dev_priv = req->i915;
 	bool interruptible;
 	int ret;
 
-	BUG_ON(req == NULL);
-
-	dev = req->ring->dev;
-	dev_priv = dev->dev_private;
 	interruptible = dev_priv->mm.interruptible;
 
-	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-
-	ret = i915_gem_check_wedge(&dev_priv->gpu_error, interruptible);
-	if (ret)
-		return ret;
+	BUG_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
 
-	ret = __i915_wait_request(req,
-				  atomic_read(&dev_priv->gpu_error.reset_counter),
-				  interruptible, NULL, NULL);
+	ret = __i915_wait_request(req, interruptible, NULL, NULL);
 	if (ret)
 		return ret;
 
@@ -1505,14 +1479,14 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
 			if (ret)
 				return ret;
 
-			i = obj->last_write_req->ring->id;
+			i = obj->last_write_req->engine->id;
 			if (obj->last_read_req[i] == obj->last_write_req)
 				i915_gem_object_retire__read(obj, i);
 			else
 				i915_gem_object_retire__write(obj);
 		}
 	} else {
-		for (i = 0; i < I915_NUM_RINGS; i++) {
+		for (i = 0; i < I915_NUM_ENGINES; i++) {
 			if (obj->last_read_req[i] == NULL)
 				continue;
 
@@ -1522,7 +1496,7 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
 
 			i915_gem_object_retire__read(obj, i);
 		}
-		RQ_BUG_ON(obj->active);
+		GEM_BUG_ON(obj->active);
 	}
 
 	return 0;
@@ -1532,7 +1506,7 @@ static void
 i915_gem_object_retire_request(struct drm_i915_gem_object *obj,
 			       struct drm_i915_gem_request *req)
 {
-	int ring = req->ring->id;
+	int ring = req->engine->id;
 
 	if (obj->last_read_req[ring] == req)
 		i915_gem_object_retire__read(obj, ring);
@@ -1552,8 +1526,7 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_i915_gem_request *requests[I915_NUM_RINGS];
-	unsigned reset_counter;
+	struct drm_i915_gem_request *requests[I915_NUM_ENGINES];
 	int ret, i, n = 0;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
@@ -1562,12 +1535,6 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 	if (!obj->active)
 		return 0;
 
-	ret = i915_gem_check_wedge(&dev_priv->gpu_error, true);
-	if (ret)
-		return ret;
-
-	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
-
 	if (readonly) {
 		struct drm_i915_gem_request *req;
 
@@ -1577,7 +1544,7 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 
 		requests[n++] = i915_gem_request_reference(req);
 	} else {
-		for (i = 0; i < I915_NUM_RINGS; i++) {
+		for (i = 0; i < I915_NUM_ENGINES; i++) {
 			struct drm_i915_gem_request *req;
 
 			req = obj->last_read_req[i];
@@ -1589,9 +1556,9 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 	}
 
 	mutex_unlock(&dev->struct_mutex);
+	ret = 0;
 	for (i = 0; ret == 0 && i < n; i++)
-		ret = __i915_wait_request(requests[i], reset_counter, true,
-					  NULL, rps);
+		ret = __i915_wait_request(requests[i], true, NULL, rps);
 	mutex_lock(&dev->struct_mutex);
 
 	for (i = 0; i < n; i++) {
@@ -1640,7 +1607,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -1688,7 +1655,7 @@ i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -1735,7 +1702,7 @@ i915_gem_mmap_ioctl(struct drm_device *dev, void *data,
 	if (args->flags & I915_MMAP_WC && !boot_cpu_has(X86_FEATURE_PAT))
 		return -ENODEV;
 
-	obj = drm_gem_object_lookup(dev, file, args->handle);
+	obj = drm_gem_object_lookup(file, args->handle);
 	if (obj == NULL)
 		return -ENOENT;
 
@@ -1792,7 +1759,8 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct drm_i915_gem_object *obj = to_intel_bo(vma->vm_private_data);
 	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_ggtt_view view = i915_ggtt_view_normal;
 	pgoff_t page_offset;
 	unsigned long pfn;
@@ -1827,7 +1795,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	}
 
 	/* Use a partial view if the object is bigger than the aperture. */
-	if (obj->base.size >= dev_priv->gtt.mappable_end &&
+	if (obj->base.size >= ggtt->mappable_end &&
 	    obj->tiling_mode == I915_TILING_NONE) {
 		static const unsigned int chunk_size = 256; // 1 MiB
 
@@ -1855,7 +1823,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 		goto unpin;
 
 	/* Finally, remap it using the new GTT offset */
-	pfn = dev_priv->gtt.mappable_base +
+	pfn = ggtt->mappable_base +
 		i915_gem_obj_ggtt_offset_view(obj, &view);
 	pfn >>= PAGE_SHIFT;
 
@@ -1964,11 +1932,27 @@ out:
 void
 i915_gem_release_mmap(struct drm_i915_gem_object *obj)
 {
+	/* Serialisation between user GTT access and our code depends upon
+	 * revoking the CPU's PTE whilst the mutex is held. The next user
+	 * pagefault then has to wait until we release the mutex.
+	 */
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
+
 	if (!obj->fault_mappable)
 		return;
 
 	drm_vma_node_unmap(&obj->base.vma_node,
 			   obj->base.dev->anon_inode->i_mapping);
+
+	/* Ensure that the CPU's PTE are revoked and there are not outstanding
+	 * memory transactions from userspace before we return. The TLB
+	 * flushing implied above by changing the PTE above *should* be
+	 * sufficient, an extra barrier here just provides us with a bit
+	 * of paranoid documentation about our requirement to serialise
+	 * memory writes before touching registers / GSM.
+	 */
+	wmb();
+
 	obj->fault_mappable = false;
 }
 
@@ -2033,9 +2017,6 @@ static int i915_gem_object_create_mmap_offset(struct drm_i915_gem_object *obj)
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
 	int ret;
 
-	if (drm_vma_node_has_offset(&obj->base.vma_node))
-		return 0;
-
 	dev_priv->mm.shrinker_no_lock_stealing = true;
 
 	ret = drm_gem_create_mmap_offset(&obj->base);
@@ -2084,7 +2065,7 @@ i915_gem_mmap_gtt(struct drm_file *file,
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -2180,11 +2161,10 @@ i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj)
 	BUG_ON(obj->madv == __I915_MADV_PURGED);
 
 	ret = i915_gem_object_set_to_cpu_domain(obj, true);
-	if (ret) {
+	if (WARN_ON(ret)) {
 		/* In the event of a disaster, abandon all caches and
 		 * hope for the best.
 		 */
-		WARN_ON(ret != -EIO);
 		i915_gem_clflush_object(obj, true);
 		obj->base.read_domains = obj->base.write_domain = I915_GEM_DOMAIN_CPU;
 	}
@@ -2232,6 +2212,14 @@ i915_gem_object_put_pages(struct drm_i915_gem_object *obj)
 	 * lists early. */
 	list_del(&obj->global_list);
 
+	if (obj->mapping) {
+		if (is_vmalloc_addr(obj->mapping))
+			vunmap(obj->mapping);
+		else
+			kunmap(kmap_to_page(obj->mapping));
+		obj->mapping = NULL;
+	}
+
 	ops->put_pages(obj);
 	obj->pages = NULL;
 
@@ -2400,21 +2388,64 @@ i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 	return 0;
 }
 
+void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj)
+{
+	int ret;
+
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
+
+	ret = i915_gem_object_get_pages(obj);
+	if (ret)
+		return ERR_PTR(ret);
+
+	i915_gem_object_pin_pages(obj);
+
+	if (obj->mapping == NULL) {
+		struct page **pages;
+
+		pages = NULL;
+		if (obj->base.size == PAGE_SIZE)
+			obj->mapping = kmap(sg_page(obj->pages->sgl));
+		else
+			pages = drm_malloc_gfp(obj->base.size >> PAGE_SHIFT,
+					       sizeof(*pages),
+					       GFP_TEMPORARY);
+		if (pages != NULL) {
+			struct sg_page_iter sg_iter;
+			int n;
+
+			n = 0;
+			for_each_sg_page(obj->pages->sgl, &sg_iter,
+					 obj->pages->nents, 0)
+				pages[n++] = sg_page_iter_page(&sg_iter);
+
+			obj->mapping = vmap(pages, n, 0, PAGE_KERNEL);
+			drm_free_large(pages);
+		}
+		if (obj->mapping == NULL) {
+			i915_gem_object_unpin_pages(obj);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
+
+	return obj->mapping;
+}
+
 void i915_vma_move_to_active(struct i915_vma *vma,
 			     struct drm_i915_gem_request *req)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 
-	ring = i915_gem_request_get_ring(req);
+	engine = i915_gem_request_get_engine(req);
 
 	/* Add a reference if we're newly entering the active list. */
 	if (obj->active == 0)
 		drm_gem_object_reference(&obj->base);
-	obj->active |= intel_ring_flag(ring);
+	obj->active |= intel_engine_flag(engine);
 
-	list_move_tail(&obj->ring_list[ring->id], &ring->active_list);
-	i915_gem_request_assign(&obj->last_read_req[ring->id], req);
+	list_move_tail(&obj->engine_list[engine->id], &engine->active_list);
+	i915_gem_request_assign(&obj->last_read_req[engine->id], req);
 
 	list_move_tail(&vma->vm_link, &vma->vm->active_list);
 }
@@ -2422,8 +2453,8 @@ void i915_vma_move_to_active(struct i915_vma *vma,
 static void
 i915_gem_object_retire__write(struct drm_i915_gem_object *obj)
 {
-	RQ_BUG_ON(obj->last_write_req == NULL);
-	RQ_BUG_ON(!(obj->active & intel_ring_flag(obj->last_write_req->ring)));
+	GEM_BUG_ON(obj->last_write_req == NULL);
+	GEM_BUG_ON(!(obj->active & intel_engine_flag(obj->last_write_req->engine)));
 
 	i915_gem_request_assign(&obj->last_write_req, NULL);
 	intel_fb_obj_flush(obj, true, ORIGIN_CS);
@@ -2434,13 +2465,13 @@ i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring)
 {
 	struct i915_vma *vma;
 
-	RQ_BUG_ON(obj->last_read_req[ring] == NULL);
-	RQ_BUG_ON(!(obj->active & (1 << ring)));
+	GEM_BUG_ON(obj->last_read_req[ring] == NULL);
+	GEM_BUG_ON(!(obj->active & (1 << ring)));
 
-	list_del_init(&obj->ring_list[ring]);
+	list_del_init(&obj->engine_list[ring]);
 	i915_gem_request_assign(&obj->last_read_req[ring], NULL);
 
-	if (obj->last_write_req && obj->last_write_req->ring->id == ring)
+	if (obj->last_write_req && obj->last_write_req->engine->id == ring)
 		i915_gem_object_retire__write(obj);
 
 	obj->active &= ~(1 << ring);
@@ -2467,24 +2498,20 @@ static int
 i915_gem_init_seqno(struct drm_device *dev, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int ret, i, j;
+	struct intel_engine_cs *engine;
+	int ret;
 
 	/* Carefully retire all requests without writing to the rings */
-	for_each_ring(ring, dev_priv, i) {
-		ret = intel_ring_idle(ring);
+	for_each_engine(engine, dev_priv) {
+		ret = intel_engine_idle(engine);
 		if (ret)
 			return ret;
 	}
 	i915_gem_retire_requests(dev);
 
 	/* Finally reset hw state */
-	for_each_ring(ring, dev_priv, i) {
-		intel_ring_init_seqno(ring, seqno);
-
-		for (j = 0; j < ARRAY_SIZE(ring->semaphore.sync_seqno); j++)
-			ring->semaphore.sync_seqno[j] = 0;
-	}
+	for_each_engine(engine, dev_priv)
+		intel_ring_init_seqno(engine, seqno);
 
 	return 0;
 }
@@ -2542,7 +2569,7 @@ void __i915_add_request(struct drm_i915_gem_request *request,
 			struct drm_i915_gem_object *obj,
 			bool flush_caches)
 {
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct drm_i915_private *dev_priv;
 	struct intel_ringbuffer *ringbuf;
 	u32 request_start;
@@ -2551,8 +2578,8 @@ void __i915_add_request(struct drm_i915_gem_request *request,
 	if (WARN_ON(request == NULL))
 		return;
 
-	ring = request->ring;
-	dev_priv = ring->dev->dev_private;
+	engine = request->engine;
+	dev_priv = request->i915;
 	ringbuf = request->ringbuf;
 
 	/*
@@ -2579,6 +2606,28 @@ void __i915_add_request(struct drm_i915_gem_request *request,
 		WARN(ret, "*_ring_flush_all_caches failed: %d!\n", ret);
 	}
 
+	trace_i915_gem_request_add(request);
+
+	request->head = request_start;
+
+	/* Whilst this request exists, batch_obj will be on the
+	 * active_list, and so will hold the active reference. Only when this
+	 * request is retired will the the batch_obj be moved onto the
+	 * inactive_list and lose its active reference. Hence we do not need
+	 * to explicitly hold another reference here.
+	 */
+	request->batch_obj = obj;
+
+	/* Seal the request and mark it as pending execution. Note that
+	 * we may inspect this state, without holding any locks, during
+	 * hangcheck. Hence we apply the barrier to ensure that we do not
+	 * see a more recent value in the hws than we are tracking.
+	 */
+	request->emitted_jiffies = jiffies;
+	request->previous_seqno = engine->last_submitted_seqno;
+	smp_store_mb(engine->last_submitted_seqno, request->seqno);
+	list_add_tail(&request->list, &engine->request_list);
+
 	/* Record the position of the start of the request so that
 	 * should we detect the updated seqno part-way through the
 	 * GPU processing the request, we never over-estimate the
@@ -2587,33 +2636,16 @@ void __i915_add_request(struct drm_i915_gem_request *request,
 	request->postfix = intel_ring_get_tail(ringbuf);
 
 	if (i915.enable_execlists)
-		ret = ring->emit_request(request);
+		ret = engine->emit_request(request);
 	else {
-		ret = ring->add_request(request);
+		ret = engine->add_request(request);
 
 		request->tail = intel_ring_get_tail(ringbuf);
 	}
 	/* Not allowed to fail! */
 	WARN(ret, "emit|add_request failed: %d!\n", ret);
 
-	request->head = request_start;
-
-	/* Whilst this request exists, batch_obj will be on the
-	 * active_list, and so will hold the active reference. Only when this
-	 * request is retired will the the batch_obj be moved onto the
-	 * inactive_list and lose its active reference. Hence we do not need
-	 * to explicitly hold another reference here.
-	 */
-	request->batch_obj = obj;
-
-	request->emitted_jiffies = jiffies;
-	request->previous_seqno = ring->last_submitted_seqno;
-	ring->last_submitted_seqno = request->seqno;
-	list_add_tail(&request->list, &ring->request_list);
-
-	trace_i915_gem_request_add(request);
-
-	i915_queue_hangcheck(ring->dev);
+	i915_queue_hangcheck(engine->dev);
 
 	queue_delayed_work(dev_priv->wq,
 			   &dev_priv->mm.retire_work,
@@ -2680,7 +2712,7 @@ void i915_gem_request_free(struct kref *req_ref)
 
 	if (ctx) {
 		if (i915.enable_execlists && ctx != req->i915->kernel_context)
-			intel_lr_context_unpin(ctx, req->ring);
+			intel_lr_context_unpin(ctx, req->engine);
 
 		i915_gem_context_unreference(ctx);
 	}
@@ -2689,11 +2721,12 @@ void i915_gem_request_free(struct kref *req_ref)
 }
 
 static inline int
-__i915_gem_request_alloc(struct intel_engine_cs *ring,
+__i915_gem_request_alloc(struct intel_engine_cs *engine,
 			 struct intel_context *ctx,
 			 struct drm_i915_gem_request **req_out)
 {
-	struct drm_i915_private *dev_priv = to_i915(ring->dev);
+	struct drm_i915_private *dev_priv = to_i915(engine->dev);
+	unsigned reset_counter = i915_reset_counter(&dev_priv->gpu_error);
 	struct drm_i915_gem_request *req;
 	int ret;
 
@@ -2702,17 +2735,26 @@ __i915_gem_request_alloc(struct intel_engine_cs *ring,
 
 	*req_out = NULL;
 
+	/* ABI: Before userspace accesses the GPU (e.g. execbuffer), report
+	 * EIO if the GPU is already wedged, or EAGAIN to drop the struct_mutex
+	 * and restart.
+	 */
+	ret = i915_gem_check_wedge(reset_counter, dev_priv->mm.interruptible);
+	if (ret)
+		return ret;
+
 	req = kmem_cache_zalloc(dev_priv->requests, GFP_KERNEL);
 	if (req == NULL)
 		return -ENOMEM;
 
-	ret = i915_gem_get_seqno(ring->dev, &req->seqno);
+	ret = i915_gem_get_seqno(engine->dev, &req->seqno);
 	if (ret)
 		goto err;
 
 	kref_init(&req->ref);
 	req->i915 = dev_priv;
-	req->ring = ring;
+	req->engine = engine;
+	req->reset_counter = reset_counter;
 	req->ctx  = ctx;
 	i915_gem_context_reference(req->ctx);
 
@@ -2742,7 +2784,8 @@ __i915_gem_request_alloc(struct intel_engine_cs *ring,
 		 * fully prepared. Thus it can be cleaned up using the proper
 		 * free code.
 		 */
-		i915_gem_request_cancel(req);
+		intel_ring_reserved_space_cancel(req->ringbuf);
+		i915_gem_request_unreference(req);
 		return ret;
 	}
 
@@ -2779,19 +2822,12 @@ i915_gem_request_alloc(struct intel_engine_cs *engine,
 	return err ? ERR_PTR(err) : req;
 }
 
-void i915_gem_request_cancel(struct drm_i915_gem_request *req)
-{
-	intel_ring_reserved_space_cancel(req->ringbuf);
-
-	i915_gem_request_unreference(req);
-}
-
 struct drm_i915_gem_request *
-i915_gem_find_active_request(struct intel_engine_cs *ring)
+i915_gem_find_active_request(struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_request *request;
 
-	list_for_each_entry(request, &ring->request_list, list) {
+	list_for_each_entry(request, &engine->request_list, list) {
 		if (i915_gem_request_completed(request, false))
 			continue;
 
@@ -2801,38 +2837,38 @@ i915_gem_find_active_request(struct intel_engine_cs *ring)
 	return NULL;
 }
 
-static void i915_gem_reset_ring_status(struct drm_i915_private *dev_priv,
-				       struct intel_engine_cs *ring)
+static void i915_gem_reset_engine_status(struct drm_i915_private *dev_priv,
+				       struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_request *request;
 	bool ring_hung;
 
-	request = i915_gem_find_active_request(ring);
+	request = i915_gem_find_active_request(engine);
 
 	if (request == NULL)
 		return;
 
-	ring_hung = ring->hangcheck.score >= HANGCHECK_SCORE_RING_HUNG;
+	ring_hung = engine->hangcheck.score >= HANGCHECK_SCORE_RING_HUNG;
 
 	i915_set_reset_status(dev_priv, request->ctx, ring_hung);
 
-	list_for_each_entry_continue(request, &ring->request_list, list)
+	list_for_each_entry_continue(request, &engine->request_list, list)
 		i915_set_reset_status(dev_priv, request->ctx, false);
 }
 
-static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
-					struct intel_engine_cs *ring)
+static void i915_gem_reset_engine_cleanup(struct drm_i915_private *dev_priv,
+					struct intel_engine_cs *engine)
 {
 	struct intel_ringbuffer *buffer;
 
-	while (!list_empty(&ring->active_list)) {
+	while (!list_empty(&engine->active_list)) {
 		struct drm_i915_gem_object *obj;
 
-		obj = list_first_entry(&ring->active_list,
+		obj = list_first_entry(&engine->active_list,
 				       struct drm_i915_gem_object,
-				       ring_list[ring->id]);
+				       engine_list[engine->id]);
 
-		i915_gem_object_retire__read(obj, ring->id);
+		i915_gem_object_retire__read(obj, engine->id);
 	}
 
 	/*
@@ -2842,14 +2878,16 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 	 */
 
 	if (i915.enable_execlists) {
-		spin_lock_irq(&ring->execlist_lock);
+		/* Ensure irq handler finishes or is cancelled. */
+		tasklet_kill(&engine->irq_tasklet);
 
+		spin_lock_bh(&engine->execlist_lock);
 		/* list_splice_tail_init checks for empty lists */
-		list_splice_tail_init(&ring->execlist_queue,
-				      &ring->execlist_retired_req_list);
+		list_splice_tail_init(&engine->execlist_queue,
+				      &engine->execlist_retired_req_list);
+		spin_unlock_bh(&engine->execlist_lock);
 
-		spin_unlock_irq(&ring->execlist_lock);
-		intel_execlists_retire_requests(ring);
+		intel_execlists_retire_requests(engine);
 	}
 
 	/*
@@ -2859,10 +2897,10 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 	 * implicit references on things like e.g. ppgtt address spaces through
 	 * the request.
 	 */
-	while (!list_empty(&ring->request_list)) {
+	while (!list_empty(&engine->request_list)) {
 		struct drm_i915_gem_request *request;
 
-		request = list_first_entry(&ring->request_list,
+		request = list_first_entry(&engine->request_list,
 					   struct drm_i915_gem_request,
 					   list);
 
@@ -2876,28 +2914,29 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 	 * upon reset is less than when we start. Do one more pass over
 	 * all the ringbuffers to reset last_retired_head.
 	 */
-	list_for_each_entry(buffer, &ring->buffers, link) {
+	list_for_each_entry(buffer, &engine->buffers, link) {
 		buffer->last_retired_head = buffer->tail;
 		intel_ring_update_space(buffer);
 	}
+
+	intel_ring_init_seqno(engine, engine->last_submitted_seqno);
 }
 
 void i915_gem_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
 	/*
 	 * Before we free the objects from the requests, we need to inspect
 	 * them for finding the guilty party. As the requests only borrow
 	 * their reference to the objects, the inspection must be done first.
 	 */
-	for_each_ring(ring, dev_priv, i)
-		i915_gem_reset_ring_status(dev_priv, ring);
+	for_each_engine(engine, dev_priv)
+		i915_gem_reset_engine_status(dev_priv, engine);
 
-	for_each_ring(ring, dev_priv, i)
-		i915_gem_reset_ring_cleanup(dev_priv, ring);
+	for_each_engine(engine, dev_priv)
+		i915_gem_reset_engine_cleanup(dev_priv, engine);
 
 	i915_gem_context_reset(dev);
 
@@ -2910,19 +2949,19 @@ void i915_gem_reset(struct drm_device *dev)
  * This function clears the request list as sequence numbers are passed.
  */
 void
-i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
+i915_gem_retire_requests_ring(struct intel_engine_cs *engine)
 {
-	WARN_ON(i915_verify_lists(ring->dev));
+	WARN_ON(i915_verify_lists(engine->dev));
 
 	/* Retire requests first as we use it above for the early return.
 	 * If we retire requests last, we may use a later seqno and so clear
 	 * the requests lists without clearing the active list, leading to
 	 * confusion.
 	 */
-	while (!list_empty(&ring->request_list)) {
+	while (!list_empty(&engine->request_list)) {
 		struct drm_i915_gem_request *request;
 
-		request = list_first_entry(&ring->request_list,
+		request = list_first_entry(&engine->request_list,
 					   struct drm_i915_gem_request,
 					   list);
 
@@ -2936,45 +2975,44 @@ i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 	 * by the ringbuffer to the flushing/inactive lists as appropriate,
 	 * before we free the context associated with the requests.
 	 */
-	while (!list_empty(&ring->active_list)) {
+	while (!list_empty(&engine->active_list)) {
 		struct drm_i915_gem_object *obj;
 
-		obj = list_first_entry(&ring->active_list,
-				      struct drm_i915_gem_object,
-				      ring_list[ring->id]);
+		obj = list_first_entry(&engine->active_list,
+				       struct drm_i915_gem_object,
+				       engine_list[engine->id]);
 
-		if (!list_empty(&obj->last_read_req[ring->id]->list))
+		if (!list_empty(&obj->last_read_req[engine->id]->list))
 			break;
 
-		i915_gem_object_retire__read(obj, ring->id);
+		i915_gem_object_retire__read(obj, engine->id);
 	}
 
-	if (unlikely(ring->trace_irq_req &&
-		     i915_gem_request_completed(ring->trace_irq_req, true))) {
-		ring->irq_put(ring);
-		i915_gem_request_assign(&ring->trace_irq_req, NULL);
+	if (unlikely(engine->trace_irq_req &&
+		     i915_gem_request_completed(engine->trace_irq_req, true))) {
+		engine->irq_put(engine);
+		i915_gem_request_assign(&engine->trace_irq_req, NULL);
 	}
 
-	WARN_ON(i915_verify_lists(ring->dev));
+	WARN_ON(i915_verify_lists(engine->dev));
 }
 
 bool
 i915_gem_retire_requests(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	bool idle = true;
-	int i;
 
-	for_each_ring(ring, dev_priv, i) {
-		i915_gem_retire_requests_ring(ring);
-		idle &= list_empty(&ring->request_list);
+	for_each_engine(engine, dev_priv) {
+		i915_gem_retire_requests_ring(engine);
+		idle &= list_empty(&engine->request_list);
 		if (i915.enable_execlists) {
-			spin_lock_irq(&ring->execlist_lock);
-			idle &= list_empty(&ring->execlist_queue);
-			spin_unlock_irq(&ring->execlist_lock);
+			spin_lock_bh(&engine->execlist_lock);
+			idle &= list_empty(&engine->execlist_queue);
+			spin_unlock_bh(&engine->execlist_lock);
 
-			intel_execlists_retire_requests(ring);
+			intel_execlists_retire_requests(engine);
 		}
 	}
 
@@ -3011,25 +3049,21 @@ i915_gem_idle_work_handler(struct work_struct *work)
 	struct drm_i915_private *dev_priv =
 		container_of(work, typeof(*dev_priv), mm.idle_work.work);
 	struct drm_device *dev = dev_priv->dev;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
-	for_each_ring(ring, dev_priv, i)
-		if (!list_empty(&ring->request_list))
+	for_each_engine(engine, dev_priv)
+		if (!list_empty(&engine->request_list))
 			return;
 
 	/* we probably should sync with hangcheck here, using cancel_work_sync.
-	 * Also locking seems to be fubar here, ring->request_list is protected
+	 * Also locking seems to be fubar here, engine->request_list is protected
 	 * by dev->struct_mutex. */
 
 	intel_mark_idle(dev);
 
 	if (mutex_trylock(&dev->struct_mutex)) {
-		struct intel_engine_cs *ring;
-		int i;
-
-		for_each_ring(ring, dev_priv, i)
-			i915_gem_batch_pool_fini(&ring->batch_pool);
+		for_each_engine(engine, dev_priv)
+			i915_gem_batch_pool_fini(&engine->batch_pool);
 
 		mutex_unlock(&dev->struct_mutex);
 	}
@@ -3048,7 +3082,7 @@ i915_gem_object_flush_active(struct drm_i915_gem_object *obj)
 	if (!obj->active)
 		return 0;
 
-	for (i = 0; i < I915_NUM_RINGS; i++) {
+	for (i = 0; i < I915_NUM_ENGINES; i++) {
 		struct drm_i915_gem_request *req;
 
 		req = obj->last_read_req[i];
@@ -3093,11 +3127,9 @@ retire:
 int
 i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_wait *args = data;
 	struct drm_i915_gem_object *obj;
-	struct drm_i915_gem_request *req[I915_NUM_RINGS];
-	unsigned reset_counter;
+	struct drm_i915_gem_request *req[I915_NUM_ENGINES];
 	int i, n = 0;
 	int ret;
 
@@ -3108,7 +3140,7 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->bo_handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->bo_handle));
 	if (&obj->base == NULL) {
 		mutex_unlock(&dev->struct_mutex);
 		return -ENOENT;
@@ -3131,9 +3163,8 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	}
 
 	drm_gem_object_unreference(&obj->base);
-	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 
-	for (i = 0; i < I915_NUM_RINGS; i++) {
+	for (i = 0; i < I915_NUM_ENGINES; i++) {
 		if (obj->last_read_req[i] == NULL)
 			continue;
 
@@ -3144,7 +3175,7 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 
 	for (i = 0; i < n; i++) {
 		if (ret == 0)
-			ret = __i915_wait_request(req[i], reset_counter, true,
+			ret = __i915_wait_request(req[i], true,
 						  args->timeout_ns > 0 ? &args->timeout_ns : NULL,
 						  to_rps_client(file));
 		i915_gem_request_unreference__unlocked(req[i]);
@@ -3166,7 +3197,7 @@ __i915_gem_object_sync(struct drm_i915_gem_object *obj,
 	struct intel_engine_cs *from;
 	int ret;
 
-	from = i915_gem_request_get_ring(from_req);
+	from = i915_gem_request_get_engine(from_req);
 	if (to == from)
 		return 0;
 
@@ -3176,7 +3207,6 @@ __i915_gem_object_sync(struct drm_i915_gem_object *obj,
 	if (!i915_semaphore_is_enabled(obj->base.dev)) {
 		struct drm_i915_private *i915 = to_i915(obj->base.dev);
 		ret = __i915_wait_request(from_req,
-					  atomic_read(&i915->gpu_error.reset_counter),
 					  i915->mm.interruptible,
 					  NULL,
 					  &i915->rps.semaphores);
@@ -3260,7 +3290,7 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 		     struct drm_i915_gem_request **to_req)
 {
 	const bool readonly = obj->base.pending_write_domain == 0;
-	struct drm_i915_gem_request *req[I915_NUM_RINGS];
+	struct drm_i915_gem_request *req[I915_NUM_ENGINES];
 	int ret, i, n;
 
 	if (!obj->active)
@@ -3274,7 +3304,7 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 		if (obj->last_write_req)
 			req[n++] = obj->last_write_req;
 	} else {
-		for (i = 0; i < I915_NUM_RINGS; i++)
+		for (i = 0; i < I915_NUM_ENGINES; i++)
 			if (obj->last_read_req[i])
 				req[n++] = obj->last_read_req[i];
 	}
@@ -3297,9 +3327,6 @@ static void i915_gem_object_finish_gtt(struct drm_i915_gem_object *obj)
 	if ((obj->base.read_domains & I915_GEM_DOMAIN_GTT) == 0)
 		return;
 
-	/* Wait for any direct GTT access to complete */
-	mb();
-
 	old_read_domains = obj->base.read_domains;
 	old_write_domain = obj->base.write_domain;
 
@@ -3391,28 +3418,25 @@ int __i915_vma_unbind_no_wait(struct i915_vma *vma)
 int i915_gpu_idle(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int ret, i;
+	struct intel_engine_cs *engine;
+	int ret;
 
 	/* Flush everything onto the inactive list. */
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		if (!i915.enable_execlists) {
 			struct drm_i915_gem_request *req;
 
-			req = i915_gem_request_alloc(ring, NULL);
+			req = i915_gem_request_alloc(engine, NULL);
 			if (IS_ERR(req))
 				return PTR_ERR(req);
 
 			ret = i915_switch_context(req);
-			if (ret) {
-				i915_gem_request_cancel(req);
-				return ret;
-			}
-
 			i915_add_request_no_flush(req);
+			if (ret)
+				return ret;
 		}
 
-		ret = intel_ring_idle(ring);
+		ret = intel_engine_idle(engine);
 		if (ret)
 			return ret;
 	}
@@ -3466,7 +3490,8 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 			   uint64_t flags)
 {
 	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	u32 fence_alignment, unfenced_alignment;
 	u32 search_flag, alloc_flag;
 	u64 start, end;
@@ -3513,7 +3538,7 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 	start = flags & PIN_OFFSET_BIAS ? flags & PIN_OFFSET_MASK : 0;
 	end = vm->total;
 	if (flags & PIN_MAPPABLE)
-		end = min_t(u64, end, dev_priv->gtt.mappable_end);
+		end = min_t(u64, end, ggtt->mappable_end);
 	if (flags & PIN_ZONE_4G)
 		end = min_t(u64, end, (1ULL << 32) - PAGE_SIZE);
 
@@ -3720,6 +3745,9 @@ i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj)
 int
 i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 {
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	uint32_t old_write_domain, old_read_domains;
 	struct i915_vma *vma;
 	int ret;
@@ -3774,7 +3802,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 	vma = i915_gem_obj_to_ggtt(obj);
 	if (vma && drm_mm_node_allocated(&vma->node) && !obj->active)
 		list_move_tail(&vma->vm_link,
-			       &to_i915(obj->base.dev)->gtt.base.inactive_list);
+			       &ggtt->base.inactive_list);
 
 	return 0;
 }
@@ -3906,7 +3934,7 @@ int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data,
 	struct drm_i915_gem_caching *args = data;
 	struct drm_i915_gem_object *obj;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL)
 		return -ENOENT;
 
@@ -3949,7 +3977,7 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 		 * cacheline, whereas normally such cachelines would get
 		 * invalidated.
 		 */
-		if (IS_BXT_REVID(dev, 0, BXT_REVID_A1))
+		if (!HAS_LLC(dev) && !HAS_SNOOP(dev))
 			return -ENODEV;
 
 		level = I915_CACHE_LLC;
@@ -3967,7 +3995,7 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto rpm_put;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -4128,16 +4156,15 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 	unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES;
 	struct drm_i915_gem_request *request, *target = NULL;
-	unsigned reset_counter;
 	int ret;
 
 	ret = i915_gem_wait_for_error(&dev_priv->gpu_error);
 	if (ret)
 		return ret;
 
-	ret = i915_gem_check_wedge(&dev_priv->gpu_error, false);
-	if (ret)
-		return ret;
+	/* ABI: return -EIO if already wedged */
+	if (i915_terminally_wedged(&dev_priv->gpu_error))
+		return -EIO;
 
 	spin_lock(&file_priv->mm.lock);
 	list_for_each_entry(request, &file_priv->mm.request_list, client_list) {
@@ -4153,7 +4180,6 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 
 		target = request;
 	}
-	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 	if (target)
 		i915_gem_request_reference(target);
 	spin_unlock(&file_priv->mm.lock);
@@ -4161,7 +4187,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	if (target == NULL)
 		return 0;
 
-	ret = __i915_wait_request(target, reset_counter, true, NULL, NULL);
+	ret = __i915_wait_request(target, true, NULL, NULL);
 	if (ret == 0)
 		queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, 0);
 
@@ -4211,7 +4237,7 @@ void __i915_vma_set_map_and_fenceable(struct i915_vma *vma)
 		     (vma->node.start & (fence_alignment - 1)) == 0);
 
 	mappable = (vma->node.start + fence_size <=
-		    to_i915(obj->base.dev)->gtt.mappable_end);
+		    to_i915(obj->base.dev)->ggtt.mappable_end);
 
 	obj->map_and_fenceable = mappable && fenceable;
 }
@@ -4243,9 +4269,6 @@ i915_gem_object_do_pin(struct drm_i915_gem_object *obj,
 	vma = ggtt_view ? i915_gem_obj_to_ggtt_view(obj, ggtt_view) :
 			  i915_gem_obj_to_vma(obj, vm);
 
-	if (IS_ERR(vma))
-		return PTR_ERR(vma);
-
 	if (vma) {
 		if (WARN_ON(vma->pin_count == DRM_I915_GEM_OBJECT_MAX_PIN_COUNT))
 			return -EBUSY;
@@ -4308,10 +4331,13 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
 			 uint32_t alignment,
 			 uint64_t flags)
 {
-	if (WARN_ONCE(!view, "no view specified"))
-		return -EINVAL;
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+
+	BUG_ON(!view);
 
-	return i915_gem_object_do_pin(obj, i915_obj_to_ggtt(obj), view,
+	return i915_gem_object_do_pin(obj, &ggtt->base, view,
 				      alignment, flags | PIN_GLOBAL);
 }
 
@@ -4321,7 +4347,6 @@ i915_gem_object_ggtt_unpin_view(struct drm_i915_gem_object *obj,
 {
 	struct i915_vma *vma = i915_gem_obj_to_ggtt_view(obj, view);
 
-	BUG_ON(!vma);
 	WARN_ON(vma->pin_count == 0);
 	WARN_ON(!i915_gem_obj_ggtt_bound_view(obj, view));
 
@@ -4340,7 +4365,7 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -4359,15 +4384,15 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
 	if (obj->active) {
 		int i;
 
-		for (i = 0; i < I915_NUM_RINGS; i++) {
+		for (i = 0; i < I915_NUM_ENGINES; i++) {
 			struct drm_i915_gem_request *req;
 
 			req = obj->last_read_req[i];
 			if (req)
-				args->busy |= 1 << (16 + req->ring->exec_id);
+				args->busy |= 1 << (16 + req->engine->exec_id);
 		}
 		if (obj->last_write_req)
-			args->busy |= obj->last_write_req->ring->exec_id;
+			args->busy |= obj->last_write_req->engine->exec_id;
 	}
 
 unref:
@@ -4405,7 +4430,7 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file_priv, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file_priv, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
 		goto unlock;
@@ -4447,8 +4472,8 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj,
 	int i;
 
 	INIT_LIST_HEAD(&obj->global_list);
-	for (i = 0; i < I915_NUM_RINGS; i++)
-		INIT_LIST_HEAD(&obj->ring_list[i]);
+	for (i = 0; i < I915_NUM_ENGINES; i++)
+		INIT_LIST_HEAD(&obj->engine_list[i]);
 	INIT_LIST_HEAD(&obj->obj_exec_link);
 	INIT_LIST_HEAD(&obj->vma_list);
 	INIT_LIST_HEAD(&obj->batch_pool_link);
@@ -4623,14 +4648,15 @@ struct i915_vma *i915_gem_obj_to_vma(struct drm_i915_gem_object *obj,
 struct i915_vma *i915_gem_obj_to_ggtt_view(struct drm_i915_gem_object *obj,
 					   const struct i915_ggtt_view *view)
 {
-	struct i915_address_space *ggtt = i915_obj_to_ggtt(obj);
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_vma *vma;
 
-	if (WARN_ONCE(!view, "no view specified"))
-		return ERR_PTR(-EINVAL);
+	BUG_ON(!view);
 
 	list_for_each_entry(vma, &obj->vma_list, obj_link)
-		if (vma->vm == ggtt &&
+		if (vma->vm == &ggtt->base &&
 		    i915_ggtt_view_equal(&vma->ggtt_view, view))
 			return vma;
 	return NULL;
@@ -4653,14 +4679,13 @@ void i915_gem_vma_destroy(struct i915_vma *vma)
 }
 
 static void
-i915_gem_stop_ringbuffers(struct drm_device *dev)
+i915_gem_stop_engines(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
-	for_each_ring(ring, dev_priv, i)
-		dev_priv->gt.stop_ring(ring);
+	for_each_engine(engine, dev_priv)
+		dev_priv->gt.stop_engine(engine);
 }
 
 int
@@ -4676,7 +4701,7 @@ i915_gem_suspend(struct drm_device *dev)
 
 	i915_gem_retire_requests(dev);
 
-	i915_gem_stop_ringbuffers(dev);
+	i915_gem_stop_engines(dev);
 	mutex_unlock(&dev->struct_mutex);
 
 	cancel_delayed_work_sync(&dev_priv->gpu_error.hangcheck_work);
@@ -4697,8 +4722,8 @@ err:
 
 int i915_gem_l3_remap(struct drm_i915_gem_request *req, int slice)
 {
-	struct intel_engine_cs *ring = req->ring;
-	struct drm_device *dev = ring->dev;
+	struct intel_engine_cs *engine = req->engine;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 *remap_info = dev_priv->l3_parity.remap_info[slice];
 	int i, ret;
@@ -4716,12 +4741,12 @@ int i915_gem_l3_remap(struct drm_i915_gem_request *req, int slice)
 	 * at initialization time.
 	 */
 	for (i = 0; i < GEN7_L3LOG_SIZE / 4; i++) {
-		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit_reg(ring, GEN7_L3LOG(slice, i));
-		intel_ring_emit(ring, remap_info[i]);
+		intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit_reg(engine, GEN7_L3LOG(slice, i));
+		intel_ring_emit(engine, remap_info[i]);
 	}
 
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 
 	return ret;
 }
@@ -4778,7 +4803,7 @@ static void init_unused_rings(struct drm_device *dev)
 	}
 }
 
-int i915_gem_init_rings(struct drm_device *dev)
+int i915_gem_init_engines(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
@@ -4814,13 +4839,13 @@ int i915_gem_init_rings(struct drm_device *dev)
 	return 0;
 
 cleanup_vebox_ring:
-	intel_cleanup_ring_buffer(&dev_priv->ring[VECS]);
+	intel_cleanup_engine(&dev_priv->engine[VECS]);
 cleanup_blt_ring:
-	intel_cleanup_ring_buffer(&dev_priv->ring[BCS]);
+	intel_cleanup_engine(&dev_priv->engine[BCS]);
 cleanup_bsd_ring:
-	intel_cleanup_ring_buffer(&dev_priv->ring[VCS]);
+	intel_cleanup_engine(&dev_priv->engine[VCS]);
 cleanup_render_ring:
-	intel_cleanup_ring_buffer(&dev_priv->ring[RCS]);
+	intel_cleanup_engine(&dev_priv->engine[RCS]);
 
 	return ret;
 }
@@ -4829,8 +4854,8 @@ int
 i915_gem_init_hw(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int ret, i, j;
+	struct intel_engine_cs *engine;
+	int ret, j;
 
 	if (INTEL_INFO(dev)->gen < 6 && !intel_enable_gtt())
 		return -EIO;
@@ -4838,7 +4863,7 @@ i915_gem_init_hw(struct drm_device *dev)
 	/* Double layer security blanket, see i915_gem_init() */
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
-	if (dev_priv->ellc_size)
+	if (HAS_EDRAM(dev) && INTEL_GEN(dev_priv) < 9)
 		I915_WRITE(HSW_IDICR, I915_READ(HSW_IDICR) | IDIHASHMSK(0xf));
 
 	if (IS_HASWELL(dev))
@@ -4876,12 +4901,14 @@ i915_gem_init_hw(struct drm_device *dev)
 	}
 
 	/* Need to do basic initialisation of all rings first: */
-	for_each_ring(ring, dev_priv, i) {
-		ret = ring->init_hw(ring);
+	for_each_engine(engine, dev_priv) {
+		ret = engine->init_hw(engine);
 		if (ret)
 			goto out;
 	}
 
+	intel_mocs_init_l3cc_table(dev);
+
 	/* We can't enable contexts until all firmware is loaded */
 	if (HAS_GUC_UCODE(dev)) {
 		ret = intel_guc_ucode_load(dev);
@@ -4901,38 +4928,39 @@ i915_gem_init_hw(struct drm_device *dev)
 		goto out;
 
 	/* Now it is safe to go back round and do everything else: */
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		struct drm_i915_gem_request *req;
 
-		req = i915_gem_request_alloc(ring, NULL);
+		req = i915_gem_request_alloc(engine, NULL);
 		if (IS_ERR(req)) {
 			ret = PTR_ERR(req);
-			i915_gem_cleanup_ringbuffer(dev);
-			goto out;
+			break;
 		}
 
-		if (ring->id == RCS) {
-			for (j = 0; j < NUM_L3_SLICES(dev); j++)
-				i915_gem_l3_remap(req, j);
+		if (engine->id == RCS) {
+			for (j = 0; j < NUM_L3_SLICES(dev); j++) {
+				ret = i915_gem_l3_remap(req, j);
+				if (ret)
+					goto err_request;
+			}
 		}
 
 		ret = i915_ppgtt_init_ring(req);
-		if (ret && ret != -EIO) {
-			DRM_ERROR("PPGTT enable ring #%d failed %d\n", i, ret);
-			i915_gem_request_cancel(req);
-			i915_gem_cleanup_ringbuffer(dev);
-			goto out;
-		}
+		if (ret)
+			goto err_request;
 
 		ret = i915_gem_context_enable(req);
-		if (ret && ret != -EIO) {
-			DRM_ERROR("Context enable ring #%d failed %d\n", i, ret);
-			i915_gem_request_cancel(req);
-			i915_gem_cleanup_ringbuffer(dev);
-			goto out;
-		}
+		if (ret)
+			goto err_request;
 
+err_request:
 		i915_add_request_no_flush(req);
+		if (ret) {
+			DRM_ERROR("Failed to enable %s, error=%d\n",
+				  engine->name, ret);
+			i915_gem_cleanup_engines(dev);
+			break;
+		}
 	}
 
 out:
@@ -4952,14 +4980,14 @@ int i915_gem_init(struct drm_device *dev)
 
 	if (!i915.enable_execlists) {
 		dev_priv->gt.execbuf_submit = i915_gem_ringbuffer_submission;
-		dev_priv->gt.init_rings = i915_gem_init_rings;
-		dev_priv->gt.cleanup_ring = intel_cleanup_ring_buffer;
-		dev_priv->gt.stop_ring = intel_stop_ring_buffer;
+		dev_priv->gt.init_engines = i915_gem_init_engines;
+		dev_priv->gt.cleanup_engine = intel_cleanup_engine;
+		dev_priv->gt.stop_engine = intel_stop_engine;
 	} else {
 		dev_priv->gt.execbuf_submit = intel_execlists_submission;
-		dev_priv->gt.init_rings = intel_logical_rings_init;
-		dev_priv->gt.cleanup_ring = intel_logical_ring_cleanup;
-		dev_priv->gt.stop_ring = intel_logical_ring_stop;
+		dev_priv->gt.init_engines = intel_logical_rings_init;
+		dev_priv->gt.cleanup_engine = intel_logical_ring_cleanup;
+		dev_priv->gt.stop_engine = intel_logical_ring_stop;
 	}
 
 	/* This is just a security blanket to placate dragons.
@@ -4974,13 +5002,13 @@ int i915_gem_init(struct drm_device *dev)
 	if (ret)
 		goto out_unlock;
 
-	i915_gem_init_global_gtt(dev);
+	i915_gem_init_ggtt(dev);
 
 	ret = i915_gem_context_init(dev);
 	if (ret)
 		goto out_unlock;
 
-	ret = dev_priv->gt.init_rings(dev);
+	ret = dev_priv->gt.init_engines(dev);
 	if (ret)
 		goto out_unlock;
 
@@ -5003,29 +5031,52 @@ out_unlock:
 }
 
 void
-i915_gem_cleanup_ringbuffer(struct drm_device *dev)
+i915_gem_cleanup_engines(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
-	for_each_ring(ring, dev_priv, i)
-		dev_priv->gt.cleanup_ring(ring);
+	for_each_engine(engine, dev_priv)
+		dev_priv->gt.cleanup_engine(engine);
 
-    if (i915.enable_execlists)
-            /*
-             * Neither the BIOS, ourselves or any other kernel
-             * expects the system to be in execlists mode on startup,
-             * so we need to reset the GPU back to legacy mode.
-             */
-            intel_gpu_reset(dev);
+	if (i915.enable_execlists)
+		/*
+		 * Neither the BIOS, ourselves or any other kernel
+		 * expects the system to be in execlists mode on startup,
+		 * so we need to reset the GPU back to legacy mode.
+		 */
+		intel_gpu_reset(dev, ALL_ENGINES);
 }
 
 static void
-init_ring_lists(struct intel_engine_cs *ring)
+init_engine_lists(struct intel_engine_cs *engine)
 {
-	INIT_LIST_HEAD(&ring->active_list);
-	INIT_LIST_HEAD(&ring->request_list);
+	INIT_LIST_HEAD(&engine->active_list);
+	INIT_LIST_HEAD(&engine->request_list);
+}
+
+void
+i915_gem_load_init_fences(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	if (INTEL_INFO(dev_priv)->gen >= 7 && !IS_VALLEYVIEW(dev_priv) &&
+	    !IS_CHERRYVIEW(dev_priv))
+		dev_priv->num_fence_regs = 32;
+	else if (INTEL_INFO(dev_priv)->gen >= 4 || IS_I945G(dev_priv) ||
+		 IS_I945GM(dev_priv) || IS_G33(dev_priv))
+		dev_priv->num_fence_regs = 16;
+	else
+		dev_priv->num_fence_regs = 8;
+
+	if (intel_vgpu_active(dev))
+		dev_priv->num_fence_regs =
+				I915_READ(vgtif_reg(avail_rs.fence_num));
+
+	/* Initialize fence registers to zero */
+	i915_gem_restore_fences(dev);
+
+	i915_gem_detect_bit_6_swizzle(dev);
 }
 
 void
@@ -5055,8 +5106,8 @@ i915_gem_load_init(struct drm_device *dev)
 	INIT_LIST_HEAD(&dev_priv->mm.unbound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.bound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.fence_list);
-	for (i = 0; i < I915_NUM_RINGS; i++)
-		init_ring_lists(&dev_priv->ring[i]);
+	for (i = 0; i < I915_NUM_ENGINES; i++)
+		init_engine_lists(&dev_priv->engine[i]);
 	for (i = 0; i < I915_MAX_NUM_FENCES; i++)
 		INIT_LIST_HEAD(&dev_priv->fence_regs[i].lru_list);
 	INIT_DELAYED_WORK(&dev_priv->mm.retire_work,
@@ -5067,17 +5118,6 @@ i915_gem_load_init(struct drm_device *dev)
 
 	dev_priv->relative_constants_mode = I915_EXEC_CONSTANTS_REL_GENERAL;
 
-	if (INTEL_INFO(dev)->gen >= 7 && !IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev))
-		dev_priv->num_fence_regs = 32;
-	else if (INTEL_INFO(dev)->gen >= 4 || IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev))
-		dev_priv->num_fence_regs = 16;
-	else
-		dev_priv->num_fence_regs = 8;
-
-	if (intel_vgpu_active(dev))
-		dev_priv->num_fence_regs =
-				I915_READ(vgtif_reg(avail_rs.fence_num));
-
 	/*
 	 * Set initial sequence number for requests.
 	 * Using this number allows the wraparound to happen early,
@@ -5086,11 +5126,8 @@ i915_gem_load_init(struct drm_device *dev)
 	dev_priv->next_seqno = ((u32)~0 - 0x1100);
 	dev_priv->last_seqno = ((u32)~0 - 0x1101);
 
-	/* Initialize fence registers to zero */
 	INIT_LIST_HEAD(&dev_priv->mm.fence_list);
-	i915_gem_restore_fences(dev);
 
-	i915_gem_detect_bit_6_swizzle(dev);
 	init_waitqueue_head(&dev_priv->pending_flip_queue);
 
 	dev_priv->mm.interruptible = true;
@@ -5213,11 +5250,12 @@ u64 i915_gem_obj_offset(struct drm_i915_gem_object *o,
 u64 i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
 				  const struct i915_ggtt_view *view)
 {
-	struct i915_address_space *ggtt = i915_obj_to_ggtt(o);
+	struct drm_i915_private *dev_priv = to_i915(o->base.dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, &o->vma_list, obj_link)
-		if (vma->vm == ggtt &&
+		if (vma->vm == &ggtt->base &&
 		    i915_ggtt_view_equal(&vma->ggtt_view, view))
 			return vma->node.start;
 
@@ -5244,11 +5282,12 @@ bool i915_gem_obj_bound(struct drm_i915_gem_object *o,
 bool i915_gem_obj_ggtt_bound_view(struct drm_i915_gem_object *o,
 				  const struct i915_ggtt_view *view)
 {
-	struct i915_address_space *ggtt = i915_obj_to_ggtt(o);
+	struct drm_i915_private *dev_priv = to_i915(o->base.dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, &o->vma_list, obj_link)
-		if (vma->vm == ggtt &&
+		if (vma->vm == &ggtt->base &&
 		    i915_ggtt_view_equal(&vma->ggtt_view, view) &&
 		    drm_mm_node_allocated(&vma->node))
 			return true;
diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h
new file mode 100644
index 000000000000..8292e797d9b5
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem.h
@@ -0,0 +1,34 @@
+/*
+ * Copyright © 2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __I915_GEM_H__
+#define __I915_GEM_H__
+
+#ifdef CONFIG_DRM_I915_DEBUG_GEM
+#define GEM_BUG_ON(expr) BUG_ON(expr)
+#else
+#define GEM_BUG_ON(expr)
+#endif
+
+#endif /* __I915_GEM_H__ */
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 5dd84e148bba..e5acc3916f75 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -342,15 +342,15 @@ void i915_gem_context_reset(struct drm_device *dev)
 		struct intel_context *ctx;
 
 		list_for_each_entry(ctx, &dev_priv->context_list, link)
-			intel_lr_context_reset(dev, ctx);
+			intel_lr_context_reset(dev_priv, ctx);
 	}
 
-	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_engine_cs *ring = &dev_priv->ring[i];
+	for (i = 0; i < I915_NUM_ENGINES; i++) {
+		struct intel_engine_cs *engine = &dev_priv->engine[i];
 
-		if (ring->last_context) {
-			i915_gem_context_unpin(ring->last_context, ring);
-			ring->last_context = NULL;
+		if (engine->last_context) {
+			i915_gem_context_unpin(engine->last_context, engine);
+			engine->last_context = NULL;
 		}
 	}
 
@@ -413,7 +413,7 @@ void i915_gem_context_fini(struct drm_device *dev)
 		/* The only known way to stop the gpu from accessing the hw context is
 		 * to reset it. Do this as the very last operation to avoid confusing
 		 * other code, leading to spurious errors. */
-		intel_gpu_reset(dev);
+		intel_gpu_reset(dev, ALL_ENGINES);
 
 		/* When default context is created and switched to, base object refcount
 		 * will be 2 (+1 from object creation and +1 from do_switch()).
@@ -421,17 +421,17 @@ void i915_gem_context_fini(struct drm_device *dev)
 		 * to default context. So we need to unreference the base object once
 		 * to offset the do_switch part, so that i915_gem_context_unreference()
 		 * can then free the base object correctly. */
-		WARN_ON(!dev_priv->ring[RCS].last_context);
+		WARN_ON(!dev_priv->engine[RCS].last_context);
 
 		i915_gem_object_ggtt_unpin(dctx->legacy_hw_ctx.rcs_state);
 	}
 
-	for (i = I915_NUM_RINGS; --i >= 0;) {
-		struct intel_engine_cs *ring = &dev_priv->ring[i];
+	for (i = I915_NUM_ENGINES; --i >= 0;) {
+		struct intel_engine_cs *engine = &dev_priv->engine[i];
 
-		if (ring->last_context) {
-			i915_gem_context_unpin(ring->last_context, ring);
-			ring->last_context = NULL;
+		if (engine->last_context) {
+			i915_gem_context_unpin(engine->last_context, engine);
+			engine->last_context = NULL;
 		}
 	}
 
@@ -441,14 +441,14 @@ void i915_gem_context_fini(struct drm_device *dev)
 
 int i915_gem_context_enable(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	if (i915.enable_execlists) {
-		if (ring->init_context == NULL)
+		if (engine->init_context == NULL)
 			return 0;
 
-		ret = ring->init_context(req);
+		ret = engine->init_context(req);
 	} else
 		ret = i915_switch_context(req);
 
@@ -510,133 +510,147 @@ i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id)
 static inline int
 mi_set_context(struct drm_i915_gem_request *req, u32 hw_flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	u32 flags = hw_flags | MI_MM_SPACE_GTT;
 	const int num_rings =
 		/* Use an extended w/a on ivb+ if signalling from other rings */
-		i915_semaphore_is_enabled(ring->dev) ?
-		hweight32(INTEL_INFO(ring->dev)->ring_mask) - 1 :
+		i915_semaphore_is_enabled(engine->dev) ?
+		hweight32(INTEL_INFO(engine->dev)->ring_mask) - 1 :
 		0;
-	int len, i, ret;
+	int len, ret;
 
 	/* w/a: If Flush TLB Invalidation Mode is enabled, driver must do a TLB
 	 * invalidation prior to MI_SET_CONTEXT. On GEN6 we don't set the value
 	 * explicitly, so we rely on the value at ring init, stored in
 	 * itlb_before_ctx_switch.
 	 */
-	if (IS_GEN6(ring->dev)) {
-		ret = ring->flush(req, I915_GEM_GPU_DOMAINS, 0);
+	if (IS_GEN6(engine->dev)) {
+		ret = engine->flush(req, I915_GEM_GPU_DOMAINS, 0);
 		if (ret)
 			return ret;
 	}
 
 	/* These flags are for resource streamer on HSW+ */
-	if (IS_HASWELL(ring->dev) || INTEL_INFO(ring->dev)->gen >= 8)
+	if (IS_HASWELL(engine->dev) || INTEL_INFO(engine->dev)->gen >= 8)
 		flags |= (HSW_MI_RS_SAVE_STATE_EN | HSW_MI_RS_RESTORE_STATE_EN);
-	else if (INTEL_INFO(ring->dev)->gen < 8)
+	else if (INTEL_INFO(engine->dev)->gen < 8)
 		flags |= (MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN);
 
 
 	len = 4;
-	if (INTEL_INFO(ring->dev)->gen >= 7)
-		len += 2 + (num_rings ? 4*num_rings + 2 : 0);
+	if (INTEL_INFO(engine->dev)->gen >= 7)
+		len += 2 + (num_rings ? 4*num_rings + 6 : 0);
 
 	ret = intel_ring_begin(req, len);
 	if (ret)
 		return ret;
 
 	/* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw,bdw,chv */
-	if (INTEL_INFO(ring->dev)->gen >= 7) {
-		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE);
+	if (INTEL_INFO(engine->dev)->gen >= 7) {
+		intel_ring_emit(engine, MI_ARB_ON_OFF | MI_ARB_DISABLE);
 		if (num_rings) {
 			struct intel_engine_cs *signaller;
 
-			intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_rings));
-			for_each_ring(signaller, to_i915(ring->dev), i) {
-				if (signaller == ring)
+			intel_ring_emit(engine,
+					MI_LOAD_REGISTER_IMM(num_rings));
+			for_each_engine(signaller, to_i915(engine->dev)) {
+				if (signaller == engine)
 					continue;
 
-				intel_ring_emit_reg(ring, RING_PSMI_CTL(signaller->mmio_base));
-				intel_ring_emit(ring, _MASKED_BIT_ENABLE(GEN6_PSMI_SLEEP_MSG_DISABLE));
+				intel_ring_emit_reg(engine,
+						    RING_PSMI_CTL(signaller->mmio_base));
+				intel_ring_emit(engine,
+						_MASKED_BIT_ENABLE(GEN6_PSMI_SLEEP_MSG_DISABLE));
 			}
 		}
 	}
 
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_emit(ring, MI_SET_CONTEXT);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(req->ctx->legacy_hw_ctx.rcs_state) |
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_emit(engine, MI_SET_CONTEXT);
+	intel_ring_emit(engine,
+			i915_gem_obj_ggtt_offset(req->ctx->legacy_hw_ctx.rcs_state) |
 			flags);
 	/*
 	 * w/a: MI_SET_CONTEXT must always be followed by MI_NOOP
 	 * WaMiSetContext_Hang:snb,ivb,vlv
 	 */
-	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_emit(engine, MI_NOOP);
 
-	if (INTEL_INFO(ring->dev)->gen >= 7) {
+	if (INTEL_INFO(engine->dev)->gen >= 7) {
 		if (num_rings) {
 			struct intel_engine_cs *signaller;
+			i915_reg_t last_reg = {}; /* keep gcc quiet */
 
-			intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_rings));
-			for_each_ring(signaller, to_i915(ring->dev), i) {
-				if (signaller == ring)
+			intel_ring_emit(engine,
+					MI_LOAD_REGISTER_IMM(num_rings));
+			for_each_engine(signaller, to_i915(engine->dev)) {
+				if (signaller == engine)
 					continue;
 
-				intel_ring_emit_reg(ring, RING_PSMI_CTL(signaller->mmio_base));
-				intel_ring_emit(ring, _MASKED_BIT_DISABLE(GEN6_PSMI_SLEEP_MSG_DISABLE));
+				last_reg = RING_PSMI_CTL(signaller->mmio_base);
+				intel_ring_emit_reg(engine, last_reg);
+				intel_ring_emit(engine,
+						_MASKED_BIT_DISABLE(GEN6_PSMI_SLEEP_MSG_DISABLE));
 			}
+
+			/* Insert a delay before the next switch! */
+			intel_ring_emit(engine,
+					MI_STORE_REGISTER_MEM |
+					MI_SRM_LRM_GLOBAL_GTT);
+			intel_ring_emit_reg(engine, last_reg);
+			intel_ring_emit(engine, engine->scratch.gtt_offset);
+			intel_ring_emit(engine, MI_NOOP);
 		}
-		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE);
+		intel_ring_emit(engine, MI_ARB_ON_OFF | MI_ARB_ENABLE);
 	}
 
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 
 	return ret;
 }
 
-static inline bool should_skip_switch(struct intel_engine_cs *ring,
-				      struct intel_context *from,
-				      struct intel_context *to)
+static inline bool skip_rcs_switch(struct intel_engine_cs *engine,
+				   struct intel_context *to)
 {
 	if (to->remap_slice)
 		return false;
 
-	if (to->ppgtt && from == to &&
-	    !(intel_ring_flag(ring) & to->ppgtt->pd_dirty_rings))
-		return true;
+	if (!to->legacy_hw_ctx.initialized)
+		return false;
 
-	return false;
+	if (to->ppgtt &&
+	    !(intel_engine_flag(engine) & to->ppgtt->pd_dirty_rings))
+		return false;
+
+	return to == engine->last_context;
 }
 
 static bool
-needs_pd_load_pre(struct intel_engine_cs *ring, struct intel_context *to)
+needs_pd_load_pre(struct intel_engine_cs *engine, struct intel_context *to)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-
 	if (!to->ppgtt)
 		return false;
 
-	if (INTEL_INFO(ring->dev)->gen < 8)
+	if (engine->last_context == to &&
+	    !(intel_engine_flag(engine) & to->ppgtt->pd_dirty_rings))
+		return false;
+
+	if (engine->id != RCS)
 		return true;
 
-	if (ring != &dev_priv->ring[RCS])
+	if (INTEL_INFO(engine->dev)->gen < 8)
 		return true;
 
 	return false;
 }
 
 static bool
-needs_pd_load_post(struct intel_engine_cs *ring, struct intel_context *to,
-		u32 hw_flags)
+needs_pd_load_post(struct intel_context *to, u32 hw_flags)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-
 	if (!to->ppgtt)
 		return false;
 
-	if (!IS_GEN8(ring->dev))
-		return false;
-
-	if (ring != &dev_priv->ring[RCS])
+	if (!IS_GEN8(to->i915))
 		return false;
 
 	if (hw_flags & MI_RESTORE_INHIBIT)
@@ -645,58 +659,32 @@ needs_pd_load_post(struct intel_engine_cs *ring, struct intel_context *to,
 	return false;
 }
 
-static int do_switch(struct drm_i915_gem_request *req)
+static int do_rcs_switch(struct drm_i915_gem_request *req)
 {
 	struct intel_context *to = req->ctx;
-	struct intel_engine_cs *ring = req->ring;
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	struct intel_context *from = ring->last_context;
-	u32 hw_flags = 0;
-	bool uninitialized = false;
+	struct intel_engine_cs *engine = req->engine;
+	struct intel_context *from;
+	u32 hw_flags;
 	int ret, i;
 
-	if (from != NULL && ring == &dev_priv->ring[RCS]) {
-		BUG_ON(from->legacy_hw_ctx.rcs_state == NULL);
-		BUG_ON(!i915_gem_obj_is_pinned(from->legacy_hw_ctx.rcs_state));
-	}
-
-	if (should_skip_switch(ring, from, to))
+	if (skip_rcs_switch(engine, to))
 		return 0;
 
 	/* Trying to pin first makes error handling easier. */
-	if (ring == &dev_priv->ring[RCS]) {
-		ret = i915_gem_obj_ggtt_pin(to->legacy_hw_ctx.rcs_state,
-					    get_context_alignment(ring->dev), 0);
-		if (ret)
-			return ret;
-	}
+	ret = i915_gem_obj_ggtt_pin(to->legacy_hw_ctx.rcs_state,
+				    get_context_alignment(engine->dev),
+				    0);
+	if (ret)
+		return ret;
 
 	/*
 	 * Pin can switch back to the default context if we end up calling into
 	 * evict_everything - as a last ditch gtt defrag effort that also
 	 * switches to the default context. Hence we need to reload from here.
+	 *
+	 * XXX: Doing so is painfully broken!
 	 */
-	from = ring->last_context;
-
-	if (needs_pd_load_pre(ring, to)) {
-		/* Older GENs and non render rings still want the load first,
-		 * "PP_DCLV followed by PP_DIR_BASE register through Load
-		 * Register Immediate commands in Ring Buffer before submitting
-		 * a context."*/
-		trace_switch_mm(ring, to);
-		ret = to->ppgtt->switch_mm(to->ppgtt, req);
-		if (ret)
-			goto unpin_out;
-
-		/* Doing a PD load always reloads the page dirs */
-		to->ppgtt->pd_dirty_rings &= ~intel_ring_flag(ring);
-	}
-
-	if (ring != &dev_priv->ring[RCS]) {
-		if (from)
-			i915_gem_context_unreference(from);
-		goto done;
-	}
+	from = engine->last_context;
 
 	/*
 	 * Clear this page out of any CPU caches for coherent swap-in/out. Note
@@ -710,53 +698,37 @@ static int do_switch(struct drm_i915_gem_request *req)
 	if (ret)
 		goto unpin_out;
 
-	if (!to->legacy_hw_ctx.initialized || i915_gem_context_is_default(to)) {
-		hw_flags |= MI_RESTORE_INHIBIT;
+	if (needs_pd_load_pre(engine, to)) {
+		/* Older GENs and non render rings still want the load first,
+		 * "PP_DCLV followed by PP_DIR_BASE register through Load
+		 * Register Immediate commands in Ring Buffer before submitting
+		 * a context."*/
+		trace_switch_mm(engine, to);
+		ret = to->ppgtt->switch_mm(to->ppgtt, req);
+		if (ret)
+			goto unpin_out;
+	}
+
+	if (!to->legacy_hw_ctx.initialized || i915_gem_context_is_default(to))
 		/* NB: If we inhibit the restore, the context is not allowed to
 		 * die because future work may end up depending on valid address
 		 * space. This means we must enforce that a page table load
 		 * occur when this occurs. */
-	} else if (to->ppgtt &&
-		   (intel_ring_flag(ring) & to->ppgtt->pd_dirty_rings)) {
-		hw_flags |= MI_FORCE_RESTORE;
-		to->ppgtt->pd_dirty_rings &= ~intel_ring_flag(ring);
-	}
+		hw_flags = MI_RESTORE_INHIBIT;
+	else if (to->ppgtt &&
+		 intel_engine_flag(engine) & to->ppgtt->pd_dirty_rings)
+		hw_flags = MI_FORCE_RESTORE;
+	else
+		hw_flags = 0;
 
 	/* We should never emit switch_mm more than once */
-	WARN_ON(needs_pd_load_pre(ring, to) &&
-		needs_pd_load_post(ring, to, hw_flags));
-
-	ret = mi_set_context(req, hw_flags);
-	if (ret)
-		goto unpin_out;
+	WARN_ON(needs_pd_load_pre(engine, to) &&
+		needs_pd_load_post(to, hw_flags));
 
-	/* GEN8 does *not* require an explicit reload if the PDPs have been
-	 * setup, and we do not wish to move them.
-	 */
-	if (needs_pd_load_post(ring, to, hw_flags)) {
-		trace_switch_mm(ring, to);
-		ret = to->ppgtt->switch_mm(to->ppgtt, req);
-		/* The hardware context switch is emitted, but we haven't
-		 * actually changed the state - so it's probably safe to bail
-		 * here. Still, let the user know something dangerous has
-		 * happened.
-		 */
-		if (ret) {
-			DRM_ERROR("Failed to change address space on context switch\n");
-			goto unpin_out;
-		}
-	}
-
-	for (i = 0; i < MAX_L3_SLICES; i++) {
-		if (!(to->remap_slice & (1<<i)))
-			continue;
-
-		ret = i915_gem_l3_remap(req, i);
-		/* If it failed, try again next round */
+	if (to != from || (hw_flags & MI_FORCE_RESTORE)) {
+		ret = mi_set_context(req, hw_flags);
 		if (ret)
-			DRM_DEBUG_DRIVER("L3 remapping failed\n");
-		else
-			to->remap_slice &= ~(1<<i);
+			goto unpin_out;
 	}
 
 	/* The backing object for the context is done after switching to the
@@ -781,27 +753,51 @@ static int do_switch(struct drm_i915_gem_request *req)
 		i915_gem_object_ggtt_unpin(from->legacy_hw_ctx.rcs_state);
 		i915_gem_context_unreference(from);
 	}
+	i915_gem_context_reference(to);
+	engine->last_context = to;
+
+	/* GEN8 does *not* require an explicit reload if the PDPs have been
+	 * setup, and we do not wish to move them.
+	 */
+	if (needs_pd_load_post(to, hw_flags)) {
+		trace_switch_mm(engine, to);
+		ret = to->ppgtt->switch_mm(to->ppgtt, req);
+		/* The hardware context switch is emitted, but we haven't
+		 * actually changed the state - so it's probably safe to bail
+		 * here. Still, let the user know something dangerous has
+		 * happened.
+		 */
+		if (ret)
+			return ret;
+	}
 
-	uninitialized = !to->legacy_hw_ctx.initialized;
-	to->legacy_hw_ctx.initialized = true;
+	if (to->ppgtt)
+		to->ppgtt->pd_dirty_rings &= ~intel_engine_flag(engine);
 
-done:
-	i915_gem_context_reference(to);
-	ring->last_context = to;
+	for (i = 0; i < MAX_L3_SLICES; i++) {
+		if (!(to->remap_slice & (1<<i)))
+			continue;
+
+		ret = i915_gem_l3_remap(req, i);
+		if (ret)
+			return ret;
 
-	if (uninitialized) {
-		if (ring->init_context) {
-			ret = ring->init_context(req);
+		to->remap_slice &= ~(1<<i);
+	}
+
+	if (!to->legacy_hw_ctx.initialized) {
+		if (engine->init_context) {
+			ret = engine->init_context(req);
 			if (ret)
-				DRM_ERROR("ring init context: %d\n", ret);
+				return ret;
 		}
+		to->legacy_hw_ctx.initialized = true;
 	}
 
 	return 0;
 
 unpin_out:
-	if (ring->id == RCS)
-		i915_gem_object_ggtt_unpin(to->legacy_hw_ctx.rcs_state);
+	i915_gem_object_ggtt_unpin(to->legacy_hw_ctx.rcs_state);
 	return ret;
 }
 
@@ -820,23 +816,39 @@ unpin_out:
  */
 int i915_switch_context(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct intel_engine_cs *engine = req->engine;
+	struct drm_i915_private *dev_priv = req->i915;
 
 	WARN_ON(i915.enable_execlists);
 	WARN_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
 
-	if (req->ctx->legacy_hw_ctx.rcs_state == NULL) { /* We have the fake context */
-		if (req->ctx != ring->last_context) {
-			i915_gem_context_reference(req->ctx);
-			if (ring->last_context)
-				i915_gem_context_unreference(ring->last_context);
-			ring->last_context = req->ctx;
+	if (engine->id != RCS ||
+	    req->ctx->legacy_hw_ctx.rcs_state == NULL) {
+		struct intel_context *to = req->ctx;
+
+		if (needs_pd_load_pre(engine, to)) {
+			int ret;
+
+			trace_switch_mm(engine, to);
+			ret = to->ppgtt->switch_mm(to->ppgtt, req);
+			if (ret)
+				return ret;
+
+			/* Doing a PD load always reloads the page dirs */
+			to->ppgtt->pd_dirty_rings &= ~intel_engine_flag(engine);
 		}
+
+		if (to != engine->last_context) {
+			i915_gem_context_reference(to);
+			if (engine->last_context)
+				i915_gem_context_unreference(engine->last_context);
+			engine->last_context = to;
+		}
+
 		return 0;
 	}
 
-	return do_switch(req);
+	return do_rcs_switch(req);
 }
 
 static bool contexts_enabled(struct drm_device *dev)
@@ -937,7 +949,7 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data,
 		else if (to_i915(dev)->mm.aliasing_ppgtt)
 			args->value = to_i915(dev)->mm.aliasing_ppgtt->base.total;
 		else
-			args->value = to_i915(dev)->gtt.base.total;
+			args->value = to_i915(dev)->ggtt.base.total;
 		break;
 	default:
 		ret = -EINVAL;
diff --git a/drivers/gpu/drm/i915/i915_gem_debug.c b/drivers/gpu/drm/i915/i915_gem_debug.c
index 17299d04189f..a56516482394 100644
--- a/drivers/gpu/drm/i915/i915_gem_debug.c
+++ b/drivers/gpu/drm/i915/i915_gem_debug.c
@@ -36,29 +36,29 @@ i915_verify_lists(struct drm_device *dev)
 	static int warned;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_i915_gem_object *obj;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	int err = 0;
-	int i;
 
 	if (warned)
 		return 0;
 
-	for_each_ring(ring, dev_priv, i) {
-		list_for_each_entry(obj, &ring->active_list, ring_list[ring->id]) {
+	for_each_engine(engine, dev_priv) {
+		list_for_each_entry(obj, &engine->active_list,
+				    engine_list[engine->id]) {
 			if (obj->base.dev != dev ||
 			    !atomic_read(&obj->base.refcount.refcount)) {
 				DRM_ERROR("%s: freed active obj %p\n",
-					  ring->name, obj);
+					  engine->name, obj);
 				err++;
 				break;
 			} else if (!obj->active ||
-				   obj->last_read_req[ring->id] == NULL) {
+				   obj->last_read_req[engine->id] == NULL) {
 				DRM_ERROR("%s: invalid active obj %p\n",
-					  ring->name, obj);
+					  engine->name, obj);
 				err++;
 			} else if (obj->base.write_domain) {
 				DRM_ERROR("%s: invalid write obj %p (w %x)\n",
-					  ring->name,
+					  engine->name,
 					  obj, obj->base.write_domain);
 				err++;
 			}
diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
index 0506016e18e0..80bbe43a2e92 100644
--- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
@@ -95,14 +95,12 @@ static void i915_gem_unmap_dma_buf(struct dma_buf_attachment *attachment,
 {
 	struct drm_i915_gem_object *obj = dma_buf_to_obj(attachment->dmabuf);
 
-	mutex_lock(&obj->base.dev->struct_mutex);
-
 	dma_unmap_sg(attachment->dev, sg->sgl, sg->nents, dir);
 	sg_free_table(sg);
 	kfree(sg);
 
+	mutex_lock(&obj->base.dev->struct_mutex);
 	i915_gem_object_unpin_pages(obj);
-
 	mutex_unlock(&obj->base.dev->struct_mutex);
 }
 
@@ -110,51 +108,17 @@ static void *i915_gem_dmabuf_vmap(struct dma_buf *dma_buf)
 {
 	struct drm_i915_gem_object *obj = dma_buf_to_obj(dma_buf);
 	struct drm_device *dev = obj->base.dev;
-	struct sg_page_iter sg_iter;
-	struct page **pages;
-	int ret, i;
+	void *addr;
+	int ret;
 
 	ret = i915_mutex_lock_interruptible(dev);
 	if (ret)
 		return ERR_PTR(ret);
 
-	if (obj->dma_buf_vmapping) {
-		obj->vmapping_count++;
-		goto out_unlock;
-	}
-
-	ret = i915_gem_object_get_pages(obj);
-	if (ret)
-		goto err;
-
-	i915_gem_object_pin_pages(obj);
-
-	ret = -ENOMEM;
-
-	pages = drm_malloc_ab(obj->base.size >> PAGE_SHIFT, sizeof(*pages));
-	if (pages == NULL)
-		goto err_unpin;
-
-	i = 0;
-	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0)
-		pages[i++] = sg_page_iter_page(&sg_iter);
-
-	obj->dma_buf_vmapping = vmap(pages, i, 0, PAGE_KERNEL);
-	drm_free_large(pages);
-
-	if (!obj->dma_buf_vmapping)
-		goto err_unpin;
-
-	obj->vmapping_count = 1;
-out_unlock:
+	addr = i915_gem_object_pin_map(obj);
 	mutex_unlock(&dev->struct_mutex);
-	return obj->dma_buf_vmapping;
 
-err_unpin:
-	i915_gem_object_unpin_pages(obj);
-err:
-	mutex_unlock(&dev->struct_mutex);
-	return ERR_PTR(ret);
+	return addr;
 }
 
 static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr)
@@ -163,12 +127,7 @@ static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr)
 	struct drm_device *dev = obj->base.dev;
 
 	mutex_lock(&dev->struct_mutex);
-	if (--obj->vmapping_count == 0) {
-		vunmap(obj->dma_buf_vmapping);
-		obj->dma_buf_vmapping = NULL;
-
-		i915_gem_object_unpin_pages(obj);
-	}
+	i915_gem_object_unpin_map(obj);
 	mutex_unlock(&dev->struct_mutex);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index d4d7c88ab595..33df74d98269 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -313,7 +313,8 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 		   uint64_t target_offset)
 {
 	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	uint64_t delta = relocation_target(reloc, target_offset);
 	uint64_t offset;
 	void __iomem *reloc_page;
@@ -330,7 +331,7 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 	/* Map the page containing the relocation we're going to perform.  */
 	offset = i915_gem_obj_ggtt_offset(obj);
 	offset += reloc->offset;
-	reloc_page = io_mapping_map_atomic_wc(dev_priv->gtt.mappable,
+	reloc_page = io_mapping_map_atomic_wc(ggtt->mappable,
 					      offset & PAGE_MASK);
 	iowrite32(lower_32_bits(delta), reloc_page + offset_in_page(offset));
 
@@ -340,7 +341,7 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 		if (offset_in_page(offset) == 0) {
 			io_mapping_unmap_atomic(reloc_page);
 			reloc_page =
-				io_mapping_map_atomic_wc(dev_priv->gtt.mappable,
+				io_mapping_map_atomic_wc(ggtt->mappable,
 							 offset);
 		}
 
@@ -597,7 +598,7 @@ static bool only_mappable_for_reloc(unsigned int flags)
 
 static int
 i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
-				struct intel_engine_cs *ring,
+				struct intel_engine_cs *engine,
 				bool *need_reloc)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
@@ -711,7 +712,7 @@ eb_vma_misplaced(struct i915_vma *vma)
 }
 
 static int
-i915_gem_execbuffer_reserve(struct intel_engine_cs *ring,
+i915_gem_execbuffer_reserve(struct intel_engine_cs *engine,
 			    struct list_head *vmas,
 			    struct intel_context *ctx,
 			    bool *need_relocs)
@@ -721,10 +722,10 @@ i915_gem_execbuffer_reserve(struct intel_engine_cs *ring,
 	struct i915_address_space *vm;
 	struct list_head ordered_vmas;
 	struct list_head pinned_vmas;
-	bool has_fenced_gpu_access = INTEL_INFO(ring->dev)->gen < 4;
+	bool has_fenced_gpu_access = INTEL_INFO(engine->dev)->gen < 4;
 	int retry;
 
-	i915_gem_retire_requests_ring(ring);
+	i915_gem_retire_requests_ring(engine);
 
 	vm = list_first_entry(vmas, struct i915_vma, exec_list)->vm;
 
@@ -786,7 +787,9 @@ i915_gem_execbuffer_reserve(struct intel_engine_cs *ring,
 			if (eb_vma_misplaced(vma))
 				ret = i915_vma_unbind(vma);
 			else
-				ret = i915_gem_execbuffer_reserve_vma(vma, ring, need_relocs);
+				ret = i915_gem_execbuffer_reserve_vma(vma,
+								      engine,
+								      need_relocs);
 			if (ret)
 				goto err;
 		}
@@ -796,7 +799,8 @@ i915_gem_execbuffer_reserve(struct intel_engine_cs *ring,
 			if (drm_mm_node_allocated(&vma->node))
 				continue;
 
-			ret = i915_gem_execbuffer_reserve_vma(vma, ring, need_relocs);
+			ret = i915_gem_execbuffer_reserve_vma(vma, engine,
+							      need_relocs);
 			if (ret)
 				goto err;
 		}
@@ -819,7 +823,7 @@ static int
 i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 				  struct drm_i915_gem_execbuffer2 *args,
 				  struct drm_file *file,
-				  struct intel_engine_cs *ring,
+				  struct intel_engine_cs *engine,
 				  struct eb_vmas *eb,
 				  struct drm_i915_gem_exec_object2 *exec,
 				  struct intel_context *ctx)
@@ -908,7 +912,8 @@ i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 		goto err;
 
 	need_relocs = (args->flags & I915_EXEC_NO_RELOC) == 0;
-	ret = i915_gem_execbuffer_reserve(ring, &eb->vmas, ctx, &need_relocs);
+	ret = i915_gem_execbuffer_reserve(engine, &eb->vmas, ctx,
+					  &need_relocs);
 	if (ret)
 		goto err;
 
@@ -936,7 +941,7 @@ static int
 i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 				struct list_head *vmas)
 {
-	const unsigned other_rings = ~intel_ring_flag(req->ring);
+	const unsigned other_rings = ~intel_engine_flag(req->engine);
 	struct i915_vma *vma;
 	uint32_t flush_domains = 0;
 	bool flush_chipset = false;
@@ -946,7 +951,7 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 		struct drm_i915_gem_object *obj = vma->obj;
 
 		if (obj->active & other_rings) {
-			ret = i915_gem_object_sync(obj, req->ring, &req);
+			ret = i915_gem_object_sync(obj, req->engine, &req);
 			if (ret)
 				return ret;
 		}
@@ -958,7 +963,7 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 	}
 
 	if (flush_chipset)
-		i915_gem_chipset_flush(req->ring->dev);
+		i915_gem_chipset_flush(req->engine->dev);
 
 	if (flush_domains & I915_GEM_DOMAIN_GTT)
 		wmb();
@@ -1060,12 +1065,12 @@ validate_exec_list(struct drm_device *dev,
 
 static struct intel_context *
 i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
-			  struct intel_engine_cs *ring, const u32 ctx_id)
+			  struct intel_engine_cs *engine, const u32 ctx_id)
 {
 	struct intel_context *ctx = NULL;
 	struct i915_ctx_hang_stats *hs;
 
-	if (ring->id != RCS && ctx_id != DEFAULT_CONTEXT_HANDLE)
+	if (engine->id != RCS && ctx_id != DEFAULT_CONTEXT_HANDLE)
 		return ERR_PTR(-EINVAL);
 
 	ctx = i915_gem_context_get(file->driver_priv, ctx_id);
@@ -1078,8 +1083,8 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
 		return ERR_PTR(-EIO);
 	}
 
-	if (i915.enable_execlists && !ctx->engine[ring->id].state) {
-		int ret = intel_lr_context_deferred_alloc(ctx, ring);
+	if (i915.enable_execlists && !ctx->engine[engine->id].state) {
+		int ret = intel_lr_context_deferred_alloc(ctx, engine);
 		if (ret) {
 			DRM_DEBUG("Could not create LRC %u: %d\n", ctx_id, ret);
 			return ERR_PTR(ret);
@@ -1093,7 +1098,7 @@ void
 i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 				   struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = i915_gem_request_get_ring(req);
+	struct intel_engine_cs *engine = i915_gem_request_get_engine(req);
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, vmas, exec_list) {
@@ -1120,7 +1125,7 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 		if (entry->flags & EXEC_OBJECT_NEEDS_FENCE) {
 			i915_gem_request_assign(&obj->last_fenced_req, req);
 			if (entry->flags & __EXEC_OBJECT_HAS_FENCE) {
-				struct drm_i915_private *dev_priv = to_i915(ring->dev);
+				struct drm_i915_private *dev_priv = to_i915(engine->dev);
 				list_move_tail(&dev_priv->fence_regs[obj->fence_reg].lru_list,
 					       &dev_priv->mm.fence_list);
 			}
@@ -1130,11 +1135,11 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 	}
 }
 
-void
+static void
 i915_gem_execbuffer_retire_commands(struct i915_execbuffer_params *params)
 {
 	/* Unconditionally force add_request to emit a full flush. */
-	params->ring->gpu_caches_dirty = true;
+	params->engine->gpu_caches_dirty = true;
 
 	/* Add a breadcrumb for the completion of the batch buffer */
 	__i915_add_request(params->request, params->batch_obj, true);
@@ -1144,11 +1149,11 @@ static int
 i915_reset_gen7_sol_offsets(struct drm_device *dev,
 			    struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret, i;
 
-	if (!IS_GEN7(dev) || ring != &dev_priv->ring[RCS]) {
+	if (!IS_GEN7(dev) || engine != &dev_priv->engine[RCS]) {
 		DRM_DEBUG("sol reset is gen7/rcs only\n");
 		return -EINVAL;
 	}
@@ -1158,18 +1163,18 @@ i915_reset_gen7_sol_offsets(struct drm_device *dev,
 		return ret;
 
 	for (i = 0; i < 4; i++) {
-		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit_reg(ring, GEN7_SO_WRITE_OFFSET(i));
-		intel_ring_emit(ring, 0);
+		intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit_reg(engine, GEN7_SO_WRITE_OFFSET(i));
+		intel_ring_emit(engine, 0);
 	}
 
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 
 	return 0;
 }
 
 static struct drm_i915_gem_object*
-i915_gem_execbuffer_parse(struct intel_engine_cs *ring,
+i915_gem_execbuffer_parse(struct intel_engine_cs *engine,
 			  struct drm_i915_gem_exec_object2 *shadow_exec_entry,
 			  struct eb_vmas *eb,
 			  struct drm_i915_gem_object *batch_obj,
@@ -1181,12 +1186,12 @@ i915_gem_execbuffer_parse(struct intel_engine_cs *ring,
 	struct i915_vma *vma;
 	int ret;
 
-	shadow_batch_obj = i915_gem_batch_pool_get(&ring->batch_pool,
+	shadow_batch_obj = i915_gem_batch_pool_get(&engine->batch_pool,
 						   PAGE_ALIGN(batch_len));
 	if (IS_ERR(shadow_batch_obj))
 		return shadow_batch_obj;
 
-	ret = i915_parse_cmds(ring,
+	ret = i915_parse_cmds(engine,
 			      batch_obj,
 			      shadow_batch_obj,
 			      batch_start_offset,
@@ -1227,7 +1232,7 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 			       struct list_head *vmas)
 {
 	struct drm_device *dev = params->dev;
-	struct intel_engine_cs *ring = params->ring;
+	struct intel_engine_cs *engine = params->engine;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u64 exec_start, exec_len;
 	int instp_mode;
@@ -1242,8 +1247,8 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	if (ret)
 		return ret;
 
-	WARN(params->ctx->ppgtt && params->ctx->ppgtt->pd_dirty_rings & (1<<ring->id),
-	     "%s didn't clear reload\n", ring->name);
+	WARN(params->ctx->ppgtt && params->ctx->ppgtt->pd_dirty_rings & (1<<engine->id),
+	     "%s didn't clear reload\n", engine->name);
 
 	instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
 	instp_mask = I915_EXEC_CONSTANTS_MASK;
@@ -1251,7 +1256,7 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	case I915_EXEC_CONSTANTS_REL_GENERAL:
 	case I915_EXEC_CONSTANTS_ABSOLUTE:
 	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (instp_mode != 0 && ring != &dev_priv->ring[RCS]) {
+		if (instp_mode != 0 && engine != &dev_priv->engine[RCS]) {
 			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
 			return -EINVAL;
 		}
@@ -1278,17 +1283,17 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 		return -EINVAL;
 	}
 
-	if (ring == &dev_priv->ring[RCS] &&
+	if (engine == &dev_priv->engine[RCS] &&
 	    instp_mode != dev_priv->relative_constants_mode) {
 		ret = intel_ring_begin(params->request, 4);
 		if (ret)
 			return ret;
 
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit_reg(ring, INSTPM);
-		intel_ring_emit(ring, instp_mask << 16 | instp_mode);
-		intel_ring_advance(ring);
+		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit_reg(engine, INSTPM);
+		intel_ring_emit(engine, instp_mask << 16 | instp_mode);
+		intel_ring_advance(engine);
 
 		dev_priv->relative_constants_mode = instp_mode;
 	}
@@ -1306,7 +1311,7 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	if (exec_len == 0)
 		exec_len = params->batch_obj->base.size;
 
-	ret = ring->dispatch_execbuffer(params->request,
+	ret = engine->dispatch_execbuffer(params->request,
 					exec_start, exec_len,
 					params->dispatch_flags);
 	if (ret)
@@ -1315,7 +1320,6 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
 
 	i915_gem_execbuffer_move_to_active(vmas, params->request);
-	i915_gem_execbuffer_retire_commands(params);
 
 	return 0;
 }
@@ -1363,7 +1367,7 @@ eb_get_batch(struct eb_vmas *eb)
 
 #define I915_USER_RINGS (4)
 
-static const enum intel_ring_id user_ring_map[I915_USER_RINGS + 1] = {
+static const enum intel_engine_id user_ring_map[I915_USER_RINGS + 1] = {
 	[I915_EXEC_DEFAULT]	= RCS,
 	[I915_EXEC_RENDER]	= RCS,
 	[I915_EXEC_BLT]		= BCS,
@@ -1406,12 +1410,12 @@ eb_select_ring(struct drm_i915_private *dev_priv,
 			return -EINVAL;
 		}
 
-		*ring = &dev_priv->ring[_VCS(bsd_idx)];
+		*ring = &dev_priv->engine[_VCS(bsd_idx)];
 	} else {
-		*ring = &dev_priv->ring[user_ring_map[user_ring_id]];
+		*ring = &dev_priv->engine[user_ring_map[user_ring_id]];
 	}
 
-	if (!intel_ring_initialized(*ring)) {
+	if (!intel_engine_initialized(*ring)) {
 		DRM_DEBUG("execbuf with invalid ring: %u\n", user_ring_id);
 		return -EINVAL;
 	}
@@ -1425,12 +1429,13 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		       struct drm_i915_gem_execbuffer2 *args,
 		       struct drm_i915_gem_exec_object2 *exec)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_request *req = NULL;
 	struct eb_vmas *eb;
 	struct drm_i915_gem_object *batch_obj;
 	struct drm_i915_gem_exec_object2 shadow_exec_entry;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct intel_context *ctx;
 	struct i915_address_space *vm;
 	struct i915_execbuffer_params params_master; /* XXX: will be removed later */
@@ -1457,7 +1462,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (args->flags & I915_EXEC_IS_PINNED)
 		dispatch_flags |= I915_DISPATCH_PINNED;
 
-	ret = eb_select_ring(dev_priv, file, args, &ring);
+	ret = eb_select_ring(dev_priv, file, args, &engine);
 	if (ret)
 		return ret;
 
@@ -1471,9 +1476,9 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 			DRM_DEBUG("RS is only allowed for Haswell, Gen8 and above\n");
 			return -EINVAL;
 		}
-		if (ring->id != RCS) {
+		if (engine->id != RCS) {
 			DRM_DEBUG("RS is not available on %s\n",
-				 ring->name);
+				 engine->name);
 			return -EINVAL;
 		}
 
@@ -1486,7 +1491,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (ret)
 		goto pre_mutex_err;
 
-	ctx = i915_gem_validate_context(dev, file, ring, ctx_id);
+	ctx = i915_gem_validate_context(dev, file, engine, ctx_id);
 	if (IS_ERR(ctx)) {
 		mutex_unlock(&dev->struct_mutex);
 		ret = PTR_ERR(ctx);
@@ -1498,7 +1503,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (ctx->ppgtt)
 		vm = &ctx->ppgtt->base;
 	else
-		vm = &dev_priv->gtt.base;
+		vm = &ggtt->base;
 
 	memset(&params_master, 0x00, sizeof(params_master));
 
@@ -1520,7 +1525,8 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	/* Move the objects en-masse into the GTT, evicting if necessary. */
 	need_relocs = (args->flags & I915_EXEC_NO_RELOC) == 0;
-	ret = i915_gem_execbuffer_reserve(ring, &eb->vmas, ctx, &need_relocs);
+	ret = i915_gem_execbuffer_reserve(engine, &eb->vmas, ctx,
+					  &need_relocs);
 	if (ret)
 		goto err;
 
@@ -1529,7 +1535,8 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		ret = i915_gem_execbuffer_relocate(eb);
 	if (ret) {
 		if (ret == -EFAULT) {
-			ret = i915_gem_execbuffer_relocate_slow(dev, args, file, ring,
+			ret = i915_gem_execbuffer_relocate_slow(dev, args, file,
+								engine,
 								eb, exec, ctx);
 			BUG_ON(!mutex_is_locked(&dev->struct_mutex));
 		}
@@ -1545,16 +1552,16 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	}
 
 	params->args_batch_start_offset = args->batch_start_offset;
-	if (i915_needs_cmd_parser(ring) && args->batch_len) {
+	if (i915_needs_cmd_parser(engine) && args->batch_len) {
 		struct drm_i915_gem_object *parsed_batch_obj;
 
-		parsed_batch_obj = i915_gem_execbuffer_parse(ring,
-						      &shadow_exec_entry,
-						      eb,
-						      batch_obj,
-						      args->batch_start_offset,
-						      args->batch_len,
-						      file->is_master);
+		parsed_batch_obj = i915_gem_execbuffer_parse(engine,
+							     &shadow_exec_entry,
+							     eb,
+							     batch_obj,
+							     args->batch_start_offset,
+							     args->batch_len,
+							     file->is_master);
 		if (IS_ERR(parsed_batch_obj)) {
 			ret = PTR_ERR(parsed_batch_obj);
 			goto err;
@@ -1606,7 +1613,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		params->batch_obj_vm_offset = i915_gem_obj_offset(batch_obj, vm);
 
 	/* Allocate a request for this batch buffer nice and early. */
-	req = i915_gem_request_alloc(ring, ctx);
+	req = i915_gem_request_alloc(engine, ctx);
 	if (IS_ERR(req)) {
 		ret = PTR_ERR(req);
 		goto err_batch_unpin;
@@ -1614,7 +1621,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	ret = i915_gem_request_add_to_client(req, file);
 	if (ret)
-		goto err_batch_unpin;
+		goto err_request;
 
 	/*
 	 * Save assorted stuff away to pass through to *_submission().
@@ -1624,13 +1631,15 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	 */
 	params->dev                     = dev;
 	params->file                    = file;
-	params->ring                    = ring;
+	params->engine                    = engine;
 	params->dispatch_flags          = dispatch_flags;
 	params->batch_obj               = batch_obj;
 	params->ctx                     = ctx;
 	params->request                 = req;
 
 	ret = dev_priv->gt.execbuf_submit(params, args, &eb->vmas);
+err_request:
+	i915_gem_execbuffer_retire_commands(params);
 
 err_batch_unpin:
 	/*
@@ -1647,14 +1656,6 @@ err:
 	i915_gem_context_unreference(ctx);
 	eb_destroy(eb);
 
-	/*
-	 * If the request was created but not successfully submitted then it
-	 * must be freed again. If it was submitted then it is being tracked
-	 * on the active request list and no clean up is required here.
-	 */
-	if (ret && !IS_ERR_OR_NULL(req))
-		i915_gem_request_cancel(req);
-
 	mutex_unlock(&dev->struct_mutex);
 
 pre_mutex_err:
@@ -1773,11 +1774,9 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	exec2_list = kmalloc(sizeof(*exec2_list)*args->buffer_count,
-			     GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
-	if (exec2_list == NULL)
-		exec2_list = drm_malloc_ab(sizeof(*exec2_list),
-					   args->buffer_count);
+	exec2_list = drm_malloc_gfp(args->buffer_count,
+				    sizeof(*exec2_list),
+				    GFP_TEMPORARY);
 	if (exec2_list == NULL) {
 		DRM_DEBUG("Failed to allocate exec list for %d buffers\n",
 			  args->buffer_count);
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 49e4f26b79d8..0d666b3f7e9b 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -658,7 +658,7 @@ static int gen8_write_pdp(struct drm_i915_gem_request *req,
 			  unsigned entry,
 			  dma_addr_t addr)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	BUG_ON(entry >= 4);
@@ -667,13 +667,13 @@ static int gen8_write_pdp(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-	intel_ring_emit_reg(ring, GEN8_RING_PDP_UDW(ring, entry));
-	intel_ring_emit(ring, upper_32_bits(addr));
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-	intel_ring_emit_reg(ring, GEN8_RING_PDP_LDW(ring, entry));
-	intel_ring_emit(ring, lower_32_bits(addr));
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
+	intel_ring_emit_reg(engine, GEN8_RING_PDP_UDW(engine, entry));
+	intel_ring_emit(engine, upper_32_bits(addr));
+	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
+	intel_ring_emit_reg(engine, GEN8_RING_PDP_LDW(engine, entry));
+	intel_ring_emit(engine, lower_32_bits(addr));
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -706,8 +706,7 @@ static void gen8_ppgtt_clear_pte_range(struct i915_address_space *vm,
 				       uint64_t length,
 				       gen8_pte_t scratch_pte)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	gen8_pte_t *pt_vaddr;
 	unsigned pdpe = gen8_pdpe_index(start);
 	unsigned pde = gen8_pde_index(start);
@@ -746,7 +745,7 @@ static void gen8_ppgtt_clear_pte_range(struct i915_address_space *vm,
 			num_entries--;
 		}
 
-		kunmap_px(ppgtt, pt);
+		kunmap_px(ppgtt, pt_vaddr);
 
 		pte = 0;
 		if (++pde == I915_PDES) {
@@ -762,8 +761,7 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 				   uint64_t length,
 				   bool use_scratch)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	gen8_pte_t scratch_pte = gen8_pte_encode(px_dma(vm->scratch_page),
 						 I915_CACHE_LLC, use_scratch);
 
@@ -788,8 +786,7 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
 			      uint64_t start,
 			      enum i915_cache_level cache_level)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	gen8_pte_t *pt_vaddr;
 	unsigned pdpe = gen8_pdpe_index(start);
 	unsigned pde = gen8_pde_index(start);
@@ -829,8 +826,7 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 				      enum i915_cache_level cache_level,
 				      u32 unused)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	struct sg_page_iter sg_iter;
 
 	__sg_page_iter_start(&sg_iter, pages->sgl, sg_nents(pages->sgl), 0);
@@ -909,11 +905,10 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 static int gen8_ppgtt_notify_vgt(struct i915_hw_ppgtt *ppgtt, bool create)
 {
 	enum vgt_g2v_type msg;
-	struct drm_device *dev = ppgtt->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(ppgtt->base.dev);
 	int i;
 
-	if (USES_FULL_48BIT_PPGTT(dev)) {
+	if (USES_FULL_48BIT_PPGTT(dev_priv)) {
 		u64 daddr = px_dma(&ppgtt->pml4);
 
 		I915_WRITE(vgtif_reg(pdp[0].lo), lower_32_bits(daddr));
@@ -981,8 +976,7 @@ static void gen8_ppgtt_cleanup_4lvl(struct i915_hw_ppgtt *ppgtt)
 
 static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 
 	if (intel_vgpu_active(vm->dev))
 		gen8_ppgtt_notify_vgt(ppgtt, false);
@@ -1216,8 +1210,7 @@ static int gen8_alloc_va_range_3lvl(struct i915_address_space *vm,
 				    uint64_t start,
 				    uint64_t length)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	unsigned long *new_page_dirs, *new_page_tables;
 	struct drm_device *dev = vm->dev;
 	struct i915_page_directory *pd;
@@ -1329,8 +1322,7 @@ static int gen8_alloc_va_range_4lvl(struct i915_address_space *vm,
 				    uint64_t length)
 {
 	DECLARE_BITMAP(new_pdps, GEN8_PML4ES_PER_PML4);
-	struct i915_hw_ppgtt *ppgtt =
-			container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	struct i915_page_directory_pointer *pdp;
 	uint64_t pml4e;
 	int ret = 0;
@@ -1376,8 +1368,7 @@ err_out:
 static int gen8_alloc_va_range(struct i915_address_space *vm,
 			       uint64_t start, uint64_t length)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 
 	if (USES_FULL_48BIT_PPGTT(vm->dev))
 		return gen8_alloc_va_range_4lvl(vm, &ppgtt->pml4, start, length);
@@ -1629,6 +1620,7 @@ static void gen6_write_page_range(struct drm_i915_private *dev_priv,
 				  struct i915_page_directory *pd,
 				  uint32_t start, uint32_t length)
 {
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_page_table *pt;
 	uint32_t pde, temp;
 
@@ -1637,7 +1629,7 @@ static void gen6_write_page_range(struct drm_i915_private *dev_priv,
 
 	/* Make sure write is complete before other code can use this page
 	 * table. Also require for WC mapped PTEs */
-	readl(dev_priv->gtt.gsm);
+	readl(ggtt->gsm);
 }
 
 static uint32_t get_pd_offset(struct i915_hw_ppgtt *ppgtt)
@@ -1650,11 +1642,11 @@ static uint32_t get_pd_offset(struct i915_hw_ppgtt *ppgtt)
 static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 			 struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	/* NB: TLBs must be flushed and invalidated before a switch */
-	ret = ring->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
+	ret = engine->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
 	if (ret)
 		return ret;
 
@@ -1662,13 +1654,13 @@ static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(2));
-	intel_ring_emit_reg(ring, RING_PP_DIR_DCLV(ring));
-	intel_ring_emit(ring, PP_DIR_DCLV_2G);
-	intel_ring_emit_reg(ring, RING_PP_DIR_BASE(ring));
-	intel_ring_emit(ring, get_pd_offset(ppgtt));
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(2));
+	intel_ring_emit_reg(engine, RING_PP_DIR_DCLV(engine));
+	intel_ring_emit(engine, PP_DIR_DCLV_2G);
+	intel_ring_emit_reg(engine, RING_PP_DIR_BASE(engine));
+	intel_ring_emit(engine, get_pd_offset(ppgtt));
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -1676,22 +1668,22 @@ static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 static int vgpu_mm_switch(struct i915_hw_ppgtt *ppgtt,
 			  struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct drm_i915_private *dev_priv = to_i915(ppgtt->base.dev);
 
-	I915_WRITE(RING_PP_DIR_DCLV(ring), PP_DIR_DCLV_2G);
-	I915_WRITE(RING_PP_DIR_BASE(ring), get_pd_offset(ppgtt));
+	I915_WRITE(RING_PP_DIR_DCLV(engine), PP_DIR_DCLV_2G);
+	I915_WRITE(RING_PP_DIR_BASE(engine), get_pd_offset(ppgtt));
 	return 0;
 }
 
 static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
 			  struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	/* NB: TLBs must be flushed and invalidated before a switch */
-	ret = ring->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
+	ret = engine->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
 	if (ret)
 		return ret;
 
@@ -1699,17 +1691,17 @@ static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(2));
-	intel_ring_emit_reg(ring, RING_PP_DIR_DCLV(ring));
-	intel_ring_emit(ring, PP_DIR_DCLV_2G);
-	intel_ring_emit_reg(ring, RING_PP_DIR_BASE(ring));
-	intel_ring_emit(ring, get_pd_offset(ppgtt));
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(2));
+	intel_ring_emit_reg(engine, RING_PP_DIR_DCLV(engine));
+	intel_ring_emit(engine, PP_DIR_DCLV_2G);
+	intel_ring_emit_reg(engine, RING_PP_DIR_BASE(engine));
+	intel_ring_emit(engine, get_pd_offset(ppgtt));
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	/* XXX: RCS is the only one to auto invalidate the TLBs? */
-	if (ring->id != RCS) {
-		ret = ring->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
+	if (engine->id != RCS) {
+		ret = engine->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
 		if (ret)
 			return ret;
 	}
@@ -1720,15 +1712,15 @@ static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
 static int gen6_mm_switch(struct i915_hw_ppgtt *ppgtt,
 			  struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 
-	I915_WRITE(RING_PP_DIR_DCLV(ring), PP_DIR_DCLV_2G);
-	I915_WRITE(RING_PP_DIR_BASE(ring), get_pd_offset(ppgtt));
+	I915_WRITE(RING_PP_DIR_DCLV(engine), PP_DIR_DCLV_2G);
+	I915_WRITE(RING_PP_DIR_BASE(engine), get_pd_offset(ppgtt));
 
-	POSTING_READ(RING_PP_DIR_DCLV(ring));
+	POSTING_READ(RING_PP_DIR_DCLV(engine));
 
 	return 0;
 }
@@ -1736,12 +1728,11 @@ static int gen6_mm_switch(struct i915_hw_ppgtt *ppgtt,
 static void gen8_ppgtt_enable(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int j;
+	struct intel_engine_cs *engine;
 
-	for_each_ring(ring, dev_priv, j) {
+	for_each_engine(engine, dev_priv) {
 		u32 four_level = USES_FULL_48BIT_PPGTT(dev) ? GEN8_GFX_PPGTT_48B : 0;
-		I915_WRITE(RING_MODE_GEN7(ring),
+		I915_WRITE(RING_MODE_GEN7(engine),
 			   _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE | four_level));
 	}
 }
@@ -1749,9 +1740,8 @@ static void gen8_ppgtt_enable(struct drm_device *dev)
 static void gen7_ppgtt_enable(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	uint32_t ecochk, ecobits;
-	int i;
 
 	ecobits = I915_READ(GAC_ECO_BITS);
 	I915_WRITE(GAC_ECO_BITS, ecobits | ECOBITS_PPGTT_CACHE64B);
@@ -1765,9 +1755,9 @@ static void gen7_ppgtt_enable(struct drm_device *dev)
 	}
 	I915_WRITE(GAM_ECOCHK, ecochk);
 
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		/* GFX_MODE is per-ring on gen7+ */
-		I915_WRITE(RING_MODE_GEN7(ring),
+		I915_WRITE(RING_MODE_GEN7(engine),
 			   _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE));
 	}
 }
@@ -1796,8 +1786,7 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 				   uint64_t length,
 				   bool use_scratch)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	gen6_pte_t *pt_vaddr, scratch_pte;
 	unsigned first_entry = start >> PAGE_SHIFT;
 	unsigned num_entries = length >> PAGE_SHIFT;
@@ -1831,8 +1820,7 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 				      uint64_t start,
 				      enum i915_cache_level cache_level, u32 flags)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	gen6_pte_t *pt_vaddr;
 	unsigned first_entry = start >> PAGE_SHIFT;
 	unsigned act_pt = first_entry / GEN6_PTES;
@@ -1864,9 +1852,9 @@ static int gen6_alloc_va_range(struct i915_address_space *vm,
 {
 	DECLARE_BITMAP(new_page_tables, I915_PDES);
 	struct drm_device *dev = vm->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_ppgtt *ppgtt =
-				container_of(vm, struct i915_hw_ppgtt, base);
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	struct i915_page_table *pt;
 	uint32_t start, length, start_save, length_save;
 	uint32_t pde, temp;
@@ -1932,7 +1920,7 @@ static int gen6_alloc_va_range(struct i915_address_space *vm,
 
 	/* Make sure write is complete before other code can use this page
 	 * table. Also require for WC mapped PTEs */
-	readl(dev_priv->gtt.gsm);
+	readl(ggtt->gsm);
 
 	mark_tlbs_dirty(ppgtt);
 	return 0;
@@ -1978,8 +1966,7 @@ static void gen6_free_scratch(struct i915_address_space *vm)
 
 static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	struct i915_page_table *pt;
 	uint32_t pde;
 
@@ -1997,7 +1984,8 @@ static int gen6_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt)
 {
 	struct i915_address_space *vm = &ppgtt->base;
 	struct drm_device *dev = ppgtt->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	bool retried = false;
 	int ret;
 
@@ -2005,23 +1993,23 @@ static int gen6_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt)
 	 * allocator works in address space sizes, so it's multiplied by page
 	 * size. We allocate at the top of the GTT to avoid fragmentation.
 	 */
-	BUG_ON(!drm_mm_initialized(&dev_priv->gtt.base.mm));
+	BUG_ON(!drm_mm_initialized(&ggtt->base.mm));
 
 	ret = gen6_init_scratch(vm);
 	if (ret)
 		return ret;
 
 alloc:
-	ret = drm_mm_insert_node_in_range_generic(&dev_priv->gtt.base.mm,
+	ret = drm_mm_insert_node_in_range_generic(&ggtt->base.mm,
 						  &ppgtt->node, GEN6_PD_SIZE,
 						  GEN6_PD_ALIGN, 0,
-						  0, dev_priv->gtt.base.total,
+						  0, ggtt->base.total,
 						  DRM_MM_TOPDOWN);
 	if (ret == -ENOSPC && !retried) {
-		ret = i915_gem_evict_something(dev, &dev_priv->gtt.base,
+		ret = i915_gem_evict_something(dev, &ggtt->base,
 					       GEN6_PD_SIZE, GEN6_PD_ALIGN,
 					       I915_CACHE_NONE,
-					       0, dev_priv->gtt.base.total,
+					       0, ggtt->base.total,
 					       0);
 		if (ret)
 			goto err_out;
@@ -2034,7 +2022,7 @@ alloc:
 		goto err_out;
 
 
-	if (ppgtt->node.start < dev_priv->gtt.mappable_end)
+	if (ppgtt->node.start < ggtt->mappable_end)
 		DRM_DEBUG("Forced to use aperture for PDEs\n");
 
 	return 0;
@@ -2062,10 +2050,11 @@ static void gen6_scratch_va_range(struct i915_hw_ppgtt *ppgtt,
 static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_device *dev = ppgtt->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	int ret;
 
-	ppgtt->base.pte_encode = dev_priv->gtt.base.pte_encode;
+	ppgtt->base.pte_encode = ggtt->base.pte_encode;
 	if (IS_GEN6(dev)) {
 		ppgtt->switch_mm = gen6_mm_switch;
 	} else if (IS_HASWELL(dev)) {
@@ -2095,7 +2084,7 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 	ppgtt->pd.base.ggtt_offset =
 		ppgtt->node.start / PAGE_SIZE * sizeof(gen6_pte_t);
 
-	ppgtt->pd_addr = (gen6_pte_t __iomem *)dev_priv->gtt.gsm +
+	ppgtt->pd_addr = (gen6_pte_t __iomem *)ggtt->gsm +
 		ppgtt->pd.base.ggtt_offset / sizeof(gen6_pte_t);
 
 	gen6_scratch_va_range(ppgtt, 0, ppgtt->base.total);
@@ -2192,7 +2181,7 @@ int i915_ppgtt_init_hw(struct drm_device *dev)
 
 int i915_ppgtt_init_ring(struct drm_i915_gem_request *req)
 {
-	struct drm_i915_private *dev_priv = req->ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = req->i915;
 	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
 
 	if (i915.enable_execlists)
@@ -2263,9 +2252,10 @@ static bool needs_idle_maps(struct drm_device *dev)
 
 static bool do_idling(struct drm_i915_private *dev_priv)
 {
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	bool ret = dev_priv->mm.interruptible;
 
-	if (unlikely(dev_priv->gtt.do_idle_maps)) {
+	if (unlikely(ggtt->do_idle_maps)) {
 		dev_priv->mm.interruptible = false;
 		if (i915_gpu_idle(dev_priv->dev)) {
 			DRM_ERROR("Couldn't idle GPU\n");
@@ -2279,22 +2269,23 @@ static bool do_idling(struct drm_i915_private *dev_priv)
 
 static void undo_idling(struct drm_i915_private *dev_priv, bool interruptible)
 {
-	if (unlikely(dev_priv->gtt.do_idle_maps))
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+
+	if (unlikely(ggtt->do_idle_maps))
 		dev_priv->mm.interruptible = interruptible;
 }
 
 void i915_check_and_clear_faults(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
 	if (INTEL_INFO(dev)->gen < 6)
 		return;
 
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		u32 fault_reg;
-		fault_reg = I915_READ(RING_FAULT_REG(ring));
+		fault_reg = I915_READ(RING_FAULT_REG(engine));
 		if (fault_reg & RING_FAULT_VALID) {
 			DRM_DEBUG_DRIVER("Unexpected fault\n"
 					 "\tAddr: 0x%08lx\n"
@@ -2305,16 +2296,16 @@ void i915_check_and_clear_faults(struct drm_device *dev)
 					 fault_reg & RING_FAULT_GTTSEL_MASK ? "GGTT" : "PPGTT",
 					 RING_FAULT_SRCID(fault_reg),
 					 RING_FAULT_FAULT_TYPE(fault_reg));
-			I915_WRITE(RING_FAULT_REG(ring),
+			I915_WRITE(RING_FAULT_REG(engine),
 				   fault_reg & ~RING_FAULT_VALID);
 		}
 	}
-	POSTING_READ(RING_FAULT_REG(&dev_priv->ring[RCS]));
+	POSTING_READ(RING_FAULT_REG(&dev_priv->engine[RCS]));
 }
 
 static void i915_ggtt_flush(struct drm_i915_private *dev_priv)
 {
-	if (INTEL_INFO(dev_priv->dev)->gen < 6) {
+	if (INTEL_INFO(dev_priv)->gen < 6) {
 		intel_gtt_chipset_flush();
 	} else {
 		I915_WRITE(GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN);
@@ -2324,7 +2315,8 @@ static void i915_ggtt_flush(struct drm_i915_private *dev_priv)
 
 void i915_gem_suspend_gtt_mappings(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 
 	/* Don't bother messing with faults pre GEN6 as we have little
 	 * documentation supporting that it's a good idea.
@@ -2334,10 +2326,8 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev)
 
 	i915_check_and_clear_faults(dev);
 
-	dev_priv->gtt.base.clear_range(&dev_priv->gtt.base,
-				       dev_priv->gtt.base.start,
-				       dev_priv->gtt.base.total,
-				       true);
+	ggtt->base.clear_range(&ggtt->base, ggtt->base.start, ggtt->base.total,
+			     true);
 
 	i915_ggtt_flush(dev_priv);
 }
@@ -2367,10 +2357,11 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 				     uint64_t start,
 				     enum i915_cache_level level, u32 unused)
 {
-	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(vm->dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	unsigned first_entry = start >> PAGE_SHIFT;
 	gen8_pte_t __iomem *gtt_entries =
-		(gen8_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
+		(gen8_pte_t __iomem *)ggtt->gsm + first_entry;
 	int i = 0;
 	struct sg_page_iter sg_iter;
 	dma_addr_t addr = 0; /* shut up gcc */
@@ -2444,10 +2435,11 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 				     uint64_t start,
 				     enum i915_cache_level level, u32 flags)
 {
-	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(vm->dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	unsigned first_entry = start >> PAGE_SHIFT;
 	gen6_pte_t __iomem *gtt_entries =
-		(gen6_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
+		(gen6_pte_t __iomem *)ggtt->gsm + first_entry;
 	int i = 0;
 	struct sg_page_iter sg_iter;
 	dma_addr_t addr = 0;
@@ -2487,12 +2479,13 @@ static void gen8_ggtt_clear_range(struct i915_address_space *vm,
 				  uint64_t length,
 				  bool use_scratch)
 {
-	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(vm->dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	unsigned first_entry = start >> PAGE_SHIFT;
 	unsigned num_entries = length >> PAGE_SHIFT;
 	gen8_pte_t scratch_pte, __iomem *gtt_base =
-		(gen8_pte_t __iomem *) dev_priv->gtt.gsm + first_entry;
-	const int max_entries = gtt_total_entries(dev_priv->gtt) - first_entry;
+		(gen8_pte_t __iomem *)ggtt->gsm + first_entry;
+	const int max_entries = ggtt_total_entries(ggtt) - first_entry;
 	int i;
 	int rpm_atomic_seq;
 
@@ -2518,12 +2511,13 @@ static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 				  uint64_t length,
 				  bool use_scratch)
 {
-	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(vm->dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	unsigned first_entry = start >> PAGE_SHIFT;
 	unsigned num_entries = length >> PAGE_SHIFT;
 	gen6_pte_t scratch_pte, __iomem *gtt_base =
-		(gen6_pte_t __iomem *) dev_priv->gtt.gsm + first_entry;
-	const int max_entries = gtt_total_entries(dev_priv->gtt) - first_entry;
+		(gen6_pte_t __iomem *)ggtt->gsm + first_entry;
+	const int max_entries = ggtt_total_entries(ggtt) - first_entry;
 	int i;
 	int rpm_atomic_seq;
 
@@ -2613,32 +2607,31 @@ static int aliasing_gtt_bind_vma(struct i915_vma *vma,
 				 enum i915_cache_level cache_level,
 				 u32 flags)
 {
-	struct drm_device *dev = vma->vm->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_i915_gem_object *obj = vma->obj;
-	struct sg_table *pages = obj->pages;
-	u32 pte_flags = 0;
+	u32 pte_flags;
 	int ret;
 
 	ret = i915_get_ggtt_vma_pages(vma);
 	if (ret)
 		return ret;
-	pages = vma->ggtt_view.pages;
 
 	/* Currently applicable only to VLV */
-	if (obj->gt_ro)
+	pte_flags = 0;
+	if (vma->obj->gt_ro)
 		pte_flags |= PTE_READ_ONLY;
 
 
 	if (flags & GLOBAL_BIND) {
-		vma->vm->insert_entries(vma->vm, pages,
+		vma->vm->insert_entries(vma->vm,
+					vma->ggtt_view.pages,
 					vma->node.start,
 					cache_level, pte_flags);
 	}
 
 	if (flags & LOCAL_BIND) {
-		struct i915_hw_ppgtt *appgtt = dev_priv->mm.aliasing_ppgtt;
-		appgtt->base.insert_entries(&appgtt->base, pages,
+		struct i915_hw_ppgtt *appgtt =
+			to_i915(vma->vm->dev)->mm.aliasing_ppgtt;
+		appgtt->base.insert_entries(&appgtt->base,
+					    vma->ggtt_view.pages,
 					    vma->node.start,
 					    cache_level, pte_flags);
 	}
@@ -2717,8 +2710,8 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 	 * aperture.  One page should be enough to keep any prefetching inside
 	 * of the aperture.
 	 */
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_address_space *ggtt_vm = &dev_priv->gtt.base;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_mm_node *entry;
 	struct drm_i915_gem_object *obj;
 	unsigned long hole_start, hole_end;
@@ -2726,13 +2719,13 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 
 	BUG_ON(mappable_end > end);
 
-	ggtt_vm->start = start;
+	ggtt->base.start = start;
 
 	/* Subtract the guard page before address space initialization to
 	 * shrink the range used by drm_mm */
-	ggtt_vm->total = end - start - PAGE_SIZE;
-	i915_address_space_init(ggtt_vm, dev_priv);
-	ggtt_vm->total += PAGE_SIZE;
+	ggtt->base.total = end - start - PAGE_SIZE;
+	i915_address_space_init(&ggtt->base, dev_priv);
+	ggtt->base.total += PAGE_SIZE;
 
 	if (intel_vgpu_active(dev)) {
 		ret = intel_vgt_balloon(dev);
@@ -2741,36 +2734,36 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 	}
 
 	if (!HAS_LLC(dev))
-		ggtt_vm->mm.color_adjust = i915_gtt_color_adjust;
+		ggtt->base.mm.color_adjust = i915_gtt_color_adjust;
 
 	/* Mark any preallocated objects as occupied */
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		struct i915_vma *vma = i915_gem_obj_to_vma(obj, ggtt_vm);
+		struct i915_vma *vma = i915_gem_obj_to_vma(obj, &ggtt->base);
 
 		DRM_DEBUG_KMS("reserving preallocated space: %llx + %zx\n",
 			      i915_gem_obj_ggtt_offset(obj), obj->base.size);
 
 		WARN_ON(i915_gem_obj_ggtt_bound(obj));
-		ret = drm_mm_reserve_node(&ggtt_vm->mm, &vma->node);
+		ret = drm_mm_reserve_node(&ggtt->base.mm, &vma->node);
 		if (ret) {
 			DRM_DEBUG_KMS("Reservation failed: %i\n", ret);
 			return ret;
 		}
 		vma->bound |= GLOBAL_BIND;
 		__i915_vma_set_map_and_fenceable(vma);
-		list_add_tail(&vma->vm_link, &ggtt_vm->inactive_list);
+		list_add_tail(&vma->vm_link, &ggtt->base.inactive_list);
 	}
 
 	/* Clear any non-preallocated blocks */
-	drm_mm_for_each_hole(entry, &ggtt_vm->mm, hole_start, hole_end) {
+	drm_mm_for_each_hole(entry, &ggtt->base.mm, hole_start, hole_end) {
 		DRM_DEBUG_KMS("clearing unused GTT space: [%lx, %lx]\n",
 			      hole_start, hole_end);
-		ggtt_vm->clear_range(ggtt_vm, hole_start,
+		ggtt->base.clear_range(&ggtt->base, hole_start,
 				     hole_end - hole_start, true);
 	}
 
 	/* And finally clear the reserved guard page */
-	ggtt_vm->clear_range(ggtt_vm, end - PAGE_SIZE, PAGE_SIZE, true);
+	ggtt->base.clear_range(&ggtt->base, end - PAGE_SIZE, PAGE_SIZE, true);
 
 	if (USES_PPGTT(dev) && !USES_FULL_PPGTT(dev)) {
 		struct i915_hw_ppgtt *ppgtt;
@@ -2801,28 +2794,33 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 					true);
 
 		dev_priv->mm.aliasing_ppgtt = ppgtt;
-		WARN_ON(dev_priv->gtt.base.bind_vma != ggtt_bind_vma);
-		dev_priv->gtt.base.bind_vma = aliasing_gtt_bind_vma;
+		WARN_ON(ggtt->base.bind_vma != ggtt_bind_vma);
+		ggtt->base.bind_vma = aliasing_gtt_bind_vma;
 	}
 
 	return 0;
 }
 
-void i915_gem_init_global_gtt(struct drm_device *dev)
+/**
+ * i915_gem_init_ggtt - Initialize GEM for Global GTT
+ * @dev: DRM device
+ */
+void i915_gem_init_ggtt(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u64 gtt_size, mappable_size;
-
-	gtt_size = dev_priv->gtt.base.total;
-	mappable_size = dev_priv->gtt.mappable_end;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 
-	i915_gem_setup_global_gtt(dev, 0, mappable_size, gtt_size);
+	i915_gem_setup_global_gtt(dev, 0, ggtt->mappable_end, ggtt->base.total);
 }
 
-void i915_global_gtt_cleanup(struct drm_device *dev)
+/**
+ * i915_ggtt_cleanup_hw - Clean up GGTT hardware initialization
+ * @dev: DRM device
+ */
+void i915_ggtt_cleanup_hw(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_address_space *vm = &dev_priv->gtt.base;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 
 	if (dev_priv->mm.aliasing_ppgtt) {
 		struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
@@ -2832,15 +2830,15 @@ void i915_global_gtt_cleanup(struct drm_device *dev)
 
 	i915_gem_cleanup_stolen(dev);
 
-	if (drm_mm_initialized(&vm->mm)) {
+	if (drm_mm_initialized(&ggtt->base.mm)) {
 		if (intel_vgpu_active(dev))
 			intel_vgt_deballoon();
 
-		drm_mm_takedown(&vm->mm);
-		list_del(&vm->global_link);
+		drm_mm_takedown(&ggtt->base.mm);
+		list_del(&ggtt->base.global_link);
 	}
 
-	vm->cleanup(vm);
+	ggtt->base.cleanup(&ggtt->base);
 }
 
 static unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
@@ -2924,13 +2922,14 @@ static size_t gen9_get_stolen_size(u16 gen9_gmch_ctl)
 static int ggtt_probe_common(struct drm_device *dev,
 			     size_t gtt_size)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_page_scratch *scratch_page;
-	phys_addr_t gtt_phys_addr;
+	phys_addr_t ggtt_phys_addr;
 
 	/* For Modern GENs the PTEs and register space are split in the BAR */
-	gtt_phys_addr = pci_resource_start(dev->pdev, 0) +
-		(pci_resource_len(dev->pdev, 0) / 2);
+	ggtt_phys_addr = pci_resource_start(dev->pdev, 0) +
+			 (pci_resource_len(dev->pdev, 0) / 2);
 
 	/*
 	 * On BXT writes larger than 64 bit to the GTT pagetable range will be
@@ -2940,10 +2939,10 @@ static int ggtt_probe_common(struct drm_device *dev,
 	 * readback check when writing GTT PTE entries.
 	 */
 	if (IS_BROXTON(dev))
-		dev_priv->gtt.gsm = ioremap_nocache(gtt_phys_addr, gtt_size);
+		ggtt->gsm = ioremap_nocache(ggtt_phys_addr, gtt_size);
 	else
-		dev_priv->gtt.gsm = ioremap_wc(gtt_phys_addr, gtt_size);
-	if (!dev_priv->gtt.gsm) {
+		ggtt->gsm = ioremap_wc(ggtt_phys_addr, gtt_size);
+	if (!ggtt->gsm) {
 		DRM_ERROR("Failed to map the gtt page table\n");
 		return -ENOMEM;
 	}
@@ -2952,11 +2951,11 @@ static int ggtt_probe_common(struct drm_device *dev,
 	if (IS_ERR(scratch_page)) {
 		DRM_ERROR("Scratch setup failed\n");
 		/* iounmap will also get called at remove, but meh */
-		iounmap(dev_priv->gtt.gsm);
+		iounmap(ggtt->gsm);
 		return PTR_ERR(scratch_page);
 	}
 
-	dev_priv->gtt.base.scratch_page = scratch_page;
+	ggtt->base.scratch_page = scratch_page;
 
 	return 0;
 }
@@ -2977,7 +2976,7 @@ static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv)
 	      GEN8_PPAT(6, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(2)) |
 	      GEN8_PPAT(7, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(3));
 
-	if (!USES_PPGTT(dev_priv->dev))
+	if (!USES_PPGTT(dev_priv))
 		/* Spec: "For GGTT, there is NO pat_sel[2:0] from the entry,
 		 * so RTL will always use the value corresponding to
 		 * pat_sel = 000".
@@ -3034,20 +3033,16 @@ static void chv_setup_private_ppat(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN8_PRIVATE_PAT_HI, pat >> 32);
 }
 
-static int gen8_gmch_probe(struct drm_device *dev,
-			   u64 *gtt_total,
-			   size_t *stolen,
-			   phys_addr_t *mappable_base,
-			   u64 *mappable_end)
+static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u64 gtt_size;
+	struct drm_device *dev = ggtt->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	u16 snb_gmch_ctl;
 	int ret;
 
 	/* TODO: We're not aware of mappable constraints on gen8 yet */
-	*mappable_base = pci_resource_start(dev->pdev, 2);
-	*mappable_end = pci_resource_len(dev->pdev, 2);
+	ggtt->mappable_base = pci_resource_start(dev->pdev, 2);
+	ggtt->mappable_end = pci_resource_len(dev->pdev, 2);
 
 	if (!pci_set_dma_mask(dev->pdev, DMA_BIT_MASK(39)))
 		pci_set_consistent_dma_mask(dev->pdev, DMA_BIT_MASK(39));
@@ -3055,56 +3050,50 @@ static int gen8_gmch_probe(struct drm_device *dev,
 	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
 
 	if (INTEL_INFO(dev)->gen >= 9) {
-		*stolen = gen9_get_stolen_size(snb_gmch_ctl);
-		gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
+		ggtt->stolen_size = gen9_get_stolen_size(snb_gmch_ctl);
+		ggtt->size = gen8_get_total_gtt_size(snb_gmch_ctl);
 	} else if (IS_CHERRYVIEW(dev)) {
-		*stolen = chv_get_stolen_size(snb_gmch_ctl);
-		gtt_size = chv_get_total_gtt_size(snb_gmch_ctl);
+		ggtt->stolen_size = chv_get_stolen_size(snb_gmch_ctl);
+		ggtt->size = chv_get_total_gtt_size(snb_gmch_ctl);
 	} else {
-		*stolen = gen8_get_stolen_size(snb_gmch_ctl);
-		gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
+		ggtt->stolen_size = gen8_get_stolen_size(snb_gmch_ctl);
+		ggtt->size = gen8_get_total_gtt_size(snb_gmch_ctl);
 	}
 
-	*gtt_total = (gtt_size / sizeof(gen8_pte_t)) << PAGE_SHIFT;
+	ggtt->base.total = (ggtt->size / sizeof(gen8_pte_t)) << PAGE_SHIFT;
 
 	if (IS_CHERRYVIEW(dev) || IS_BROXTON(dev))
 		chv_setup_private_ppat(dev_priv);
 	else
 		bdw_setup_private_ppat(dev_priv);
 
-	ret = ggtt_probe_common(dev, gtt_size);
-
-	dev_priv->gtt.base.clear_range = gen8_ggtt_clear_range;
-	dev_priv->gtt.base.insert_entries = gen8_ggtt_insert_entries;
-	dev_priv->gtt.base.bind_vma = ggtt_bind_vma;
-	dev_priv->gtt.base.unbind_vma = ggtt_unbind_vma;
+	ret = ggtt_probe_common(dev, ggtt->size);
 
+	ggtt->base.clear_range = gen8_ggtt_clear_range;
 	if (IS_CHERRYVIEW(dev_priv))
-		dev_priv->gtt.base.insert_entries = gen8_ggtt_insert_entries__BKL;
+		ggtt->base.insert_entries = gen8_ggtt_insert_entries__BKL;
+	else
+		ggtt->base.insert_entries = gen8_ggtt_insert_entries;
+	ggtt->base.bind_vma = ggtt_bind_vma;
+	ggtt->base.unbind_vma = ggtt_unbind_vma;
 
 	return ret;
 }
 
-static int gen6_gmch_probe(struct drm_device *dev,
-			   u64 *gtt_total,
-			   size_t *stolen,
-			   phys_addr_t *mappable_base,
-			   u64 *mappable_end)
+static int gen6_gmch_probe(struct i915_ggtt *ggtt)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned int gtt_size;
+	struct drm_device *dev = ggtt->base.dev;
 	u16 snb_gmch_ctl;
 	int ret;
 
-	*mappable_base = pci_resource_start(dev->pdev, 2);
-	*mappable_end = pci_resource_len(dev->pdev, 2);
+	ggtt->mappable_base = pci_resource_start(dev->pdev, 2);
+	ggtt->mappable_end = pci_resource_len(dev->pdev, 2);
 
 	/* 64/512MB is the current min/max we actually know of, but this is just
 	 * a coarse sanity check.
 	 */
-	if ((*mappable_end < (64<<20) || (*mappable_end > (512<<20)))) {
-		DRM_ERROR("Unknown GMADR size (%llx)\n",
-			  dev_priv->gtt.mappable_end);
+	if ((ggtt->mappable_end < (64<<20) || (ggtt->mappable_end > (512<<20)))) {
+		DRM_ERROR("Unknown GMADR size (%llx)\n", ggtt->mappable_end);
 		return -ENXIO;
 	}
 
@@ -3112,37 +3101,32 @@ static int gen6_gmch_probe(struct drm_device *dev,
 		pci_set_consistent_dma_mask(dev->pdev, DMA_BIT_MASK(40));
 	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
 
-	*stolen = gen6_get_stolen_size(snb_gmch_ctl);
-
-	gtt_size = gen6_get_total_gtt_size(snb_gmch_ctl);
-	*gtt_total = (gtt_size / sizeof(gen6_pte_t)) << PAGE_SHIFT;
+	ggtt->stolen_size = gen6_get_stolen_size(snb_gmch_ctl);
+	ggtt->size = gen6_get_total_gtt_size(snb_gmch_ctl);
+	ggtt->base.total = (ggtt->size / sizeof(gen6_pte_t)) << PAGE_SHIFT;
 
-	ret = ggtt_probe_common(dev, gtt_size);
+	ret = ggtt_probe_common(dev, ggtt->size);
 
-	dev_priv->gtt.base.clear_range = gen6_ggtt_clear_range;
-	dev_priv->gtt.base.insert_entries = gen6_ggtt_insert_entries;
-	dev_priv->gtt.base.bind_vma = ggtt_bind_vma;
-	dev_priv->gtt.base.unbind_vma = ggtt_unbind_vma;
+	ggtt->base.clear_range = gen6_ggtt_clear_range;
+	ggtt->base.insert_entries = gen6_ggtt_insert_entries;
+	ggtt->base.bind_vma = ggtt_bind_vma;
+	ggtt->base.unbind_vma = ggtt_unbind_vma;
 
 	return ret;
 }
 
 static void gen6_gmch_remove(struct i915_address_space *vm)
 {
+	struct i915_ggtt *ggtt = container_of(vm, struct i915_ggtt, base);
 
-	struct i915_gtt *gtt = container_of(vm, struct i915_gtt, base);
-
-	iounmap(gtt->gsm);
+	iounmap(ggtt->gsm);
 	free_scratch_page(vm->dev, vm->scratch_page);
 }
 
-static int i915_gmch_probe(struct drm_device *dev,
-			   u64 *gtt_total,
-			   size_t *stolen,
-			   phys_addr_t *mappable_base,
-			   u64 *mappable_end)
+static int i915_gmch_probe(struct i915_ggtt *ggtt)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_device *dev = ggtt->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	int ret;
 
 	ret = intel_gmch_probe(dev_priv->bridge_dev, dev_priv->dev->pdev, NULL);
@@ -3151,15 +3135,16 @@ static int i915_gmch_probe(struct drm_device *dev,
 		return -EIO;
 	}
 
-	intel_gtt_get(gtt_total, stolen, mappable_base, mappable_end);
+	intel_gtt_get(&ggtt->base.total, &ggtt->stolen_size,
+		      &ggtt->mappable_base, &ggtt->mappable_end);
 
-	dev_priv->gtt.do_idle_maps = needs_idle_maps(dev_priv->dev);
-	dev_priv->gtt.base.insert_entries = i915_ggtt_insert_entries;
-	dev_priv->gtt.base.clear_range = i915_ggtt_clear_range;
-	dev_priv->gtt.base.bind_vma = ggtt_bind_vma;
-	dev_priv->gtt.base.unbind_vma = ggtt_unbind_vma;
+	ggtt->do_idle_maps = needs_idle_maps(dev_priv->dev);
+	ggtt->base.insert_entries = i915_ggtt_insert_entries;
+	ggtt->base.clear_range = i915_ggtt_clear_range;
+	ggtt->base.bind_vma = ggtt_bind_vma;
+	ggtt->base.unbind_vma = ggtt_unbind_vma;
 
-	if (unlikely(dev_priv->gtt.do_idle_maps))
+	if (unlikely(ggtt->do_idle_maps))
 		DRM_INFO("applying Ironlake quirks for intel_iommu\n");
 
 	return 0;
@@ -3170,41 +3155,53 @@ static void i915_gmch_remove(struct i915_address_space *vm)
 	intel_gmch_remove();
 }
 
-int i915_gem_gtt_init(struct drm_device *dev)
+/**
+ * i915_ggtt_init_hw - Initialize GGTT hardware
+ * @dev: DRM device
+ */
+int i915_ggtt_init_hw(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_gtt *gtt = &dev_priv->gtt;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	int ret;
 
 	if (INTEL_INFO(dev)->gen <= 5) {
-		gtt->gtt_probe = i915_gmch_probe;
-		gtt->base.cleanup = i915_gmch_remove;
+		ggtt->probe = i915_gmch_probe;
+		ggtt->base.cleanup = i915_gmch_remove;
 	} else if (INTEL_INFO(dev)->gen < 8) {
-		gtt->gtt_probe = gen6_gmch_probe;
-		gtt->base.cleanup = gen6_gmch_remove;
-		if (IS_HASWELL(dev) && dev_priv->ellc_size)
-			gtt->base.pte_encode = iris_pte_encode;
+		ggtt->probe = gen6_gmch_probe;
+		ggtt->base.cleanup = gen6_gmch_remove;
+
+		if (HAS_EDRAM(dev))
+			ggtt->base.pte_encode = iris_pte_encode;
 		else if (IS_HASWELL(dev))
-			gtt->base.pte_encode = hsw_pte_encode;
+			ggtt->base.pte_encode = hsw_pte_encode;
 		else if (IS_VALLEYVIEW(dev))
-			gtt->base.pte_encode = byt_pte_encode;
+			ggtt->base.pte_encode = byt_pte_encode;
 		else if (INTEL_INFO(dev)->gen >= 7)
-			gtt->base.pte_encode = ivb_pte_encode;
+			ggtt->base.pte_encode = ivb_pte_encode;
 		else
-			gtt->base.pte_encode = snb_pte_encode;
+			ggtt->base.pte_encode = snb_pte_encode;
 	} else {
-		dev_priv->gtt.gtt_probe = gen8_gmch_probe;
-		dev_priv->gtt.base.cleanup = gen6_gmch_remove;
+		ggtt->probe = gen8_gmch_probe;
+		ggtt->base.cleanup = gen6_gmch_remove;
 	}
 
-	gtt->base.dev = dev;
-	gtt->base.is_ggtt = true;
+	ggtt->base.dev = dev;
+	ggtt->base.is_ggtt = true;
 
-	ret = gtt->gtt_probe(dev, &gtt->base.total, &gtt->stolen_size,
-			     &gtt->mappable_base, &gtt->mappable_end);
+	ret = ggtt->probe(ggtt);
 	if (ret)
 		return ret;
 
+	if ((ggtt->base.total - 1) >> 32) {
+		DRM_ERROR("We never expected a Global GTT with more than 32bits"
+			  "of address space! Found %lldM!\n",
+			  ggtt->base.total >> 20);
+		ggtt->base.total = 1ULL << 32;
+		ggtt->mappable_end = min(ggtt->mappable_end, ggtt->base.total);
+	}
+
 	/*
 	 * Initialise stolen early so that we may reserve preallocated
 	 * objects for the BIOS to KMS transition.
@@ -3215,9 +3212,9 @@ int i915_gem_gtt_init(struct drm_device *dev)
 
 	/* GMADR is the PCI mmio aperture into the global GTT. */
 	DRM_INFO("Memory usable by graphics device = %lluM\n",
-		 gtt->base.total >> 20);
-	DRM_DEBUG_DRIVER("GMADR size = %lldM\n", gtt->mappable_end >> 20);
-	DRM_DEBUG_DRIVER("GTT stolen size = %zdM\n", gtt->stolen_size >> 20);
+		 ggtt->base.total >> 20);
+	DRM_DEBUG_DRIVER("GMADR size = %lldM\n", ggtt->mappable_end >> 20);
+	DRM_DEBUG_DRIVER("GTT stolen size = %zdM\n", ggtt->stolen_size >> 20);
 #ifdef CONFIG_INTEL_IOMMU
 	if (intel_iommu_gfx_mapped)
 		DRM_INFO("VT-d active for gfx access\n");
@@ -3234,33 +3231,30 @@ int i915_gem_gtt_init(struct drm_device *dev)
 	return 0;
 
 out_gtt_cleanup:
-	gtt->base.cleanup(&dev_priv->gtt.base);
+	ggtt->base.cleanup(&ggtt->base);
 
 	return ret;
 }
 
 void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_object *obj;
-	struct i915_address_space *vm;
 	struct i915_vma *vma;
 	bool flush;
 
 	i915_check_and_clear_faults(dev);
 
 	/* First fill our portion of the GTT with scratch pages */
-	dev_priv->gtt.base.clear_range(&dev_priv->gtt.base,
-				       dev_priv->gtt.base.start,
-				       dev_priv->gtt.base.total,
-				       true);
+	ggtt->base.clear_range(&ggtt->base, ggtt->base.start, ggtt->base.total,
+			       true);
 
 	/* Cache flush objects bound into GGTT and rebind them. */
-	vm = &dev_priv->gtt.base;
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
 		flush = false;
 		list_for_each_entry(vma, &obj->vma_list, obj_link) {
-			if (vma->vm != vm)
+			if (vma->vm != &ggtt->base)
 				continue;
 
 			WARN_ON(i915_vma_bind(vma, obj->cache_level,
@@ -3283,15 +3277,17 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 	}
 
 	if (USES_PPGTT(dev)) {
+		struct i915_address_space *vm;
+
 		list_for_each_entry(vm, &dev_priv->vm_list, global_link) {
 			/* TODO: Perhaps it shouldn't be gen6 specific */
 
-			struct i915_hw_ppgtt *ppgtt =
-					container_of(vm, struct i915_hw_ppgtt,
-						     base);
+			struct i915_hw_ppgtt *ppgtt;
 
-			if (i915_is_ggtt(vm))
+			if (vm->is_ggtt)
 				ppgtt = dev_priv->mm.aliasing_ppgtt;
+			else
+				ppgtt = i915_vm_to_ppgtt(vm);
 
 			gen6_write_page_range(dev_priv, &ppgtt->pd,
 					      0, ppgtt->base.total);
@@ -3350,19 +3346,13 @@ struct i915_vma *
 i915_gem_obj_lookup_or_create_ggtt_vma(struct drm_i915_gem_object *obj,
 				       const struct i915_ggtt_view *view)
 {
-	struct i915_address_space *ggtt = i915_obj_to_ggtt(obj);
-	struct i915_vma *vma;
-
-	if (WARN_ON(!view))
-		return ERR_PTR(-EINVAL);
-
-	vma = i915_gem_obj_to_ggtt_view(obj, view);
-
-	if (IS_ERR(vma))
-		return vma;
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+	struct i915_vma *vma = i915_gem_obj_to_ggtt_view(obj, view);
 
 	if (!vma)
-		vma = __i915_gem_vma_create(obj, ggtt, view);
+		vma = __i915_gem_vma_create(obj, &ggtt->base, view);
 
 	return vma;
 
@@ -3377,11 +3367,6 @@ rotate_pages(const dma_addr_t *in, unsigned int offset,
 	unsigned int column, row;
 	unsigned int src_idx;
 
-	if (!sg) {
-		st->nents = 0;
-		sg = st->sgl;
-	}
-
 	for (column = 0; column < width; column++) {
 		src_idx = stride * (height - 1) + column;
 		for (row = 0; row < height; row++) {
@@ -3405,7 +3390,7 @@ static struct sg_table *
 intel_rotate_fb_obj_pages(struct intel_rotation_info *rot_info,
 			  struct drm_i915_gem_object *obj)
 {
-	unsigned int size_pages = rot_info->size >> PAGE_SHIFT;
+	unsigned int size_pages = rot_info->plane[0].width * rot_info->plane[0].height;
 	unsigned int size_pages_uv;
 	struct sg_page_iter sg_iter;
 	unsigned long i;
@@ -3416,14 +3401,15 @@ intel_rotate_fb_obj_pages(struct intel_rotation_info *rot_info,
 	int ret = -ENOMEM;
 
 	/* Allocate a temporary list of source pages for random access. */
-	page_addr_list = drm_malloc_ab(obj->base.size / PAGE_SIZE,
-				       sizeof(dma_addr_t));
+	page_addr_list = drm_malloc_gfp(obj->base.size / PAGE_SIZE,
+					sizeof(dma_addr_t),
+					GFP_TEMPORARY);
 	if (!page_addr_list)
 		return ERR_PTR(ret);
 
 	/* Account for UV plane with NV12. */
 	if (rot_info->pixel_format == DRM_FORMAT_NV12)
-		size_pages_uv = rot_info->size_uv >> PAGE_SHIFT;
+		size_pages_uv = rot_info->plane[1].width * rot_info->plane[1].height;
 	else
 		size_pages_uv = 0;
 
@@ -3443,11 +3429,14 @@ intel_rotate_fb_obj_pages(struct intel_rotation_info *rot_info,
 		i++;
 	}
 
+	st->nents = 0;
+	sg = st->sgl;
+
 	/* Rotate the pages. */
 	sg = rotate_pages(page_addr_list, 0,
-		     rot_info->width_pages, rot_info->height_pages,
-		     rot_info->width_pages,
-		     st, NULL);
+			  rot_info->plane[0].width, rot_info->plane[0].height,
+			  rot_info->plane[0].width,
+			  st, sg);
 
 	/* Append the UV plane if NV12. */
 	if (rot_info->pixel_format == DRM_FORMAT_NV12) {
@@ -3459,18 +3448,15 @@ intel_rotate_fb_obj_pages(struct intel_rotation_info *rot_info,
 
 		rot_info->uv_start_page = uv_start_page;
 
-		rotate_pages(page_addr_list, uv_start_page,
-			     rot_info->width_pages_uv,
-			     rot_info->height_pages_uv,
-			     rot_info->width_pages_uv,
-			     st, sg);
+		sg = rotate_pages(page_addr_list, rot_info->uv_start_page,
+				  rot_info->plane[1].width, rot_info->plane[1].height,
+				  rot_info->plane[1].width,
+				  st, sg);
 	}
 
-	DRM_DEBUG_KMS(
-		      "Created rotated page mapping for object size %zu (pitch=%u, height=%u, pixel_format=0x%x, %ux%u tiles, %u pages (%u plane 0)).\n",
-		      obj->base.size, rot_info->pitch, rot_info->height,
-		      rot_info->pixel_format, rot_info->width_pages,
-		      rot_info->height_pages, size_pages + size_pages_uv,
+	DRM_DEBUG_KMS("Created rotated page mapping for object size %zu (%ux%u tiles, %u pages (%u plane 0)).\n",
+		      obj->base.size, rot_info->plane[0].width,
+		      rot_info->plane[0].height, size_pages + size_pages_uv,
 		      size_pages);
 
 	drm_free_large(page_addr_list);
@@ -3482,11 +3468,9 @@ err_sg_alloc:
 err_st_alloc:
 	drm_free_large(page_addr_list);
 
-	DRM_DEBUG_KMS(
-		      "Failed to create rotated mapping for object size %zu! (%d) (pitch=%u, height=%u, pixel_format=0x%x, %ux%u tiles, %u pages (%u plane 0))\n",
-		      obj->base.size, ret, rot_info->pitch, rot_info->height,
-		      rot_info->pixel_format, rot_info->width_pages,
-		      rot_info->height_pages, size_pages + size_pages_uv,
+	DRM_DEBUG_KMS("Failed to create rotated mapping for object size %zu! (%d) (%ux%u tiles, %u pages (%u plane 0))\n",
+		      obj->base.size, ret, rot_info->plane[0].width,
+		      rot_info->plane[0].height, size_pages + size_pages_uv,
 		      size_pages);
 	return ERR_PTR(ret);
 }
@@ -3634,7 +3618,7 @@ i915_ggtt_view_size(struct drm_i915_gem_object *obj,
 	if (view->type == I915_GGTT_VIEW_NORMAL) {
 		return obj->base.size;
 	} else if (view->type == I915_GGTT_VIEW_ROTATED) {
-		return view->params.rotated.size;
+		return intel_rotation_info_size(&view->params.rotated) << PAGE_SHIFT;
 	} else if (view->type == I915_GGTT_VIEW_PARTIAL) {
 		return view->params.partial.size << PAGE_SHIFT;
 	} else {
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 8774f1ba46e7..d7dd3d8a8758 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -42,7 +42,7 @@ typedef uint64_t gen8_pde_t;
 typedef uint64_t gen8_ppgtt_pdpe_t;
 typedef uint64_t gen8_ppgtt_pml4e_t;
 
-#define gtt_total_entries(gtt) ((gtt).base.total >> PAGE_SHIFT)
+#define ggtt_total_entries(ggtt) ((ggtt)->base.total >> PAGE_SHIFT)
 
 /* gen6-hsw has bit 11-4 for physical addr bit 39-32 */
 #define GEN6_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0xff0))
@@ -135,16 +135,13 @@ enum i915_ggtt_view_type {
 };
 
 struct intel_rotation_info {
-	unsigned int height;
-	unsigned int pitch;
 	unsigned int uv_offset;
 	uint32_t pixel_format;
-	uint64_t fb_modifier;
-	unsigned int width_pages, height_pages;
-	uint64_t size;
-	unsigned int width_pages_uv, height_pages_uv;
-	uint64_t size_uv;
 	unsigned int uv_start_page;
+	struct {
+		/* tiles */
+		unsigned int width, height;
+	} plane[2];
 };
 
 struct i915_ggtt_view {
@@ -342,13 +339,14 @@ struct i915_address_space {
  * and correct (in cases like swizzling). That region is referred to as GMADR in
  * the spec.
  */
-struct i915_gtt {
+struct i915_ggtt {
 	struct i915_address_space base;
 
 	size_t stolen_size;		/* Total size of stolen memory */
 	size_t stolen_usable_size;	/* Total size minus BIOS reserved */
 	size_t stolen_reserved_base;
 	size_t stolen_reserved_size;
+	size_t size;			/* Total size of Global GTT */
 	u64 mappable_end;		/* End offset that we can CPU map */
 	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
 	phys_addr_t mappable_base;	/* PA of our GMADR */
@@ -360,10 +358,7 @@ struct i915_gtt {
 
 	int mtrr;
 
-	/* global gtt ops */
-	int (*gtt_probe)(struct drm_device *dev, u64 *gtt_total,
-			  size_t *stolen, phys_addr_t *mappable_base,
-			  u64 *mappable_end);
+	int (*probe)(struct i915_ggtt *ggtt);
 };
 
 struct i915_hw_ppgtt {
@@ -518,10 +513,9 @@ i915_page_dir_dma_addr(const struct i915_hw_ppgtt *ppgtt, const unsigned n)
 		px_dma(ppgtt->base.scratch_pd);
 }
 
-int i915_gem_gtt_init(struct drm_device *dev);
-void i915_gem_init_global_gtt(struct drm_device *dev);
-void i915_global_gtt_cleanup(struct drm_device *dev);
-
+int i915_ggtt_init_hw(struct drm_device *dev);
+void i915_gem_init_ggtt(struct drm_device *dev);
+void i915_ggtt_cleanup_hw(struct drm_device *dev);
 
 int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt);
 int i915_ppgtt_init_hw(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.c b/drivers/gpu/drm/i915/i915_gem_render_state.c
index fc7e6d5c6251..71611bf21fca 100644
--- a/drivers/gpu/drm/i915/i915_gem_render_state.c
+++ b/drivers/gpu/drm/i915/i915_gem_render_state.c
@@ -169,15 +169,15 @@ void i915_gem_render_state_fini(struct render_state *so)
 	drm_gem_object_unreference(&so->obj->base);
 }
 
-int i915_gem_render_state_prepare(struct intel_engine_cs *ring,
+int i915_gem_render_state_prepare(struct intel_engine_cs *engine,
 				  struct render_state *so)
 {
 	int ret;
 
-	if (WARN_ON(ring->id != RCS))
+	if (WARN_ON(engine->id != RCS))
 		return -ENOENT;
 
-	ret = render_state_init(so, ring->dev);
+	ret = render_state_init(so, engine->dev);
 	if (ret)
 		return ret;
 
@@ -198,21 +198,21 @@ int i915_gem_render_state_init(struct drm_i915_gem_request *req)
 	struct render_state so;
 	int ret;
 
-	ret = i915_gem_render_state_prepare(req->ring, &so);
+	ret = i915_gem_render_state_prepare(req->engine, &so);
 	if (ret)
 		return ret;
 
 	if (so.rodata == NULL)
 		return 0;
 
-	ret = req->ring->dispatch_execbuffer(req, so.ggtt_offset,
+	ret = req->engine->dispatch_execbuffer(req, so.ggtt_offset,
 					     so.rodata->batch_items * 4,
 					     I915_DISPATCH_SECURE);
 	if (ret)
 		goto out;
 
 	if (so.aux_batch_size > 8) {
-		ret = req->ring->dispatch_execbuffer(req,
+		ret = req->engine->dispatch_execbuffer(req,
 						     (so.ggtt_offset +
 						      so.aux_batch_offset),
 						     so.aux_batch_size,
diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.h b/drivers/gpu/drm/i915/i915_gem_render_state.h
index e641bb093a90..6aaa3a10a630 100644
--- a/drivers/gpu/drm/i915/i915_gem_render_state.h
+++ b/drivers/gpu/drm/i915/i915_gem_render_state.h
@@ -43,7 +43,7 @@ struct render_state {
 
 int i915_gem_render_state_init(struct drm_i915_gem_request *req);
 void i915_gem_render_state_fini(struct render_state *so);
-int i915_gem_render_state_prepare(struct intel_engine_cs *ring,
+int i915_gem_render_state_prepare(struct intel_engine_cs *engine,
 				  struct render_state *so);
 
 #endif /* _I915_GEM_RENDER_STATE_H_ */
diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index d3c473ffb90a..425e721aac58 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -28,6 +28,7 @@
 #include <linux/swap.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
+#include <linux/vmalloc.h>
 #include <drm/drmP.h>
 #include <drm/i915_drm.h>
 
@@ -69,6 +70,10 @@ static bool swap_available(void)
 
 static bool can_release_pages(struct drm_i915_gem_object *obj)
 {
+	/* Only shmemfs objects are backed by swap */
+	if (!obj->base.filp)
+		return false;
+
 	/* Only report true if by unbinding the object and putting its pages
 	 * we can actually make forward progress towards freeing physical
 	 * pages.
@@ -166,6 +171,10 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 			    obj->madv != I915_MADV_DONTNEED)
 				continue;
 
+			if (flags & I915_SHRINK_VMAPS &&
+			    !is_vmalloc_addr(obj->mapping))
+				continue;
+
 			if ((flags & I915_SHRINK_ACTIVE) == 0 && obj->active)
 				continue;
 
@@ -246,7 +255,7 @@ i915_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 
 	count = 0;
 	list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list)
-		if (obj->pages_pin_count == 0)
+		if (can_release_pages(obj))
 			count += obj->base.size >> PAGE_SHIFT;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
@@ -288,67 +297,82 @@ i915_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc)
 	return freed;
 }
 
-static int
-i915_gem_shrinker_oom(struct notifier_block *nb, unsigned long event, void *ptr)
-{
-	struct drm_i915_private *dev_priv =
-		container_of(nb, struct drm_i915_private, mm.oom_notifier);
-	struct drm_device *dev = dev_priv->dev;
-	struct drm_i915_gem_object *obj;
-	unsigned long timeout = msecs_to_jiffies(5000) + 1;
-	unsigned long pinned, bound, unbound, freed_pages;
+struct shrinker_lock_uninterruptible {
 	bool was_interruptible;
 	bool unlock;
+};
+
+static bool
+i915_gem_shrinker_lock_uninterruptible(struct drm_i915_private *dev_priv,
+				       struct shrinker_lock_uninterruptible *slu,
+				       int timeout_ms)
+{
+	unsigned long timeout = msecs_to_jiffies(timeout_ms) + 1;
 
-	while (!i915_gem_shrinker_lock(dev, &unlock) && --timeout) {
+	while (!i915_gem_shrinker_lock(dev_priv->dev, &slu->unlock)) {
 		schedule_timeout_killable(1);
 		if (fatal_signal_pending(current))
-			return NOTIFY_DONE;
-	}
-	if (timeout == 0) {
-		pr_err("Unable to purge GPU memory due lock contention.\n");
-		return NOTIFY_DONE;
+			return false;
+		if (--timeout == 0) {
+			pr_err("Unable to lock GPU to purge memory.\n");
+			return false;
+		}
 	}
 
-	was_interruptible = dev_priv->mm.interruptible;
+	slu->was_interruptible = dev_priv->mm.interruptible;
 	dev_priv->mm.interruptible = false;
+	return true;
+}
 
-	freed_pages = i915_gem_shrink_all(dev_priv);
+static void
+i915_gem_shrinker_unlock_uninterruptible(struct drm_i915_private *dev_priv,
+					 struct shrinker_lock_uninterruptible *slu)
+{
+	dev_priv->mm.interruptible = slu->was_interruptible;
+	if (slu->unlock)
+		mutex_unlock(&dev_priv->dev->struct_mutex);
+}
+
+static int
+i915_gem_shrinker_oom(struct notifier_block *nb, unsigned long event, void *ptr)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(nb, struct drm_i915_private, mm.oom_notifier);
+	struct shrinker_lock_uninterruptible slu;
+	struct drm_i915_gem_object *obj;
+	unsigned long unevictable, bound, unbound, freed_pages;
 
-	dev_priv->mm.interruptible = was_interruptible;
+	if (!i915_gem_shrinker_lock_uninterruptible(dev_priv, &slu, 5000))
+		return NOTIFY_DONE;
+
+	freed_pages = i915_gem_shrink_all(dev_priv);
 
 	/* Because we may be allocating inside our own driver, we cannot
 	 * assert that there are no objects with pinned pages that are not
 	 * being pointed to by hardware.
 	 */
-	unbound = bound = pinned = 0;
+	unbound = bound = unevictable = 0;
 	list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list) {
-		if (!obj->base.filp) /* not backed by a freeable object */
-			continue;
-
-		if (obj->pages_pin_count)
-			pinned += obj->base.size;
+		if (!can_release_pages(obj))
+			unevictable += obj->base.size >> PAGE_SHIFT;
 		else
-			unbound += obj->base.size;
+			unbound += obj->base.size >> PAGE_SHIFT;
 	}
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		if (!obj->base.filp)
-			continue;
-
-		if (obj->pages_pin_count)
-			pinned += obj->base.size;
+		if (!can_release_pages(obj))
+			unevictable += obj->base.size >> PAGE_SHIFT;
 		else
-			bound += obj->base.size;
+			bound += obj->base.size >> PAGE_SHIFT;
 	}
 
-	if (unlock)
-		mutex_unlock(&dev->struct_mutex);
+	i915_gem_shrinker_unlock_uninterruptible(dev_priv, &slu);
 
 	if (freed_pages || unbound || bound)
-		pr_info("Purging GPU memory, %lu bytes freed, %lu bytes still pinned.\n",
-			freed_pages << PAGE_SHIFT, pinned);
+		pr_info("Purging GPU memory, %lu pages freed, "
+			"%lu pages still pinned.\n",
+			freed_pages, unevictable);
 	if (unbound || bound)
-		pr_err("%lu and %lu bytes still available in the "
+		pr_err("%lu and %lu pages still available in the "
 		       "bound and unbound GPU page lists.\n",
 		       bound, unbound);
 
@@ -356,6 +380,29 @@ i915_gem_shrinker_oom(struct notifier_block *nb, unsigned long event, void *ptr)
 	return NOTIFY_DONE;
 }
 
+static int
+i915_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(nb, struct drm_i915_private, mm.vmap_notifier);
+	struct shrinker_lock_uninterruptible slu;
+	unsigned long freed_pages;
+
+	if (!i915_gem_shrinker_lock_uninterruptible(dev_priv, &slu, 5000))
+		return NOTIFY_DONE;
+
+	freed_pages = i915_gem_shrink(dev_priv, -1UL,
+				      I915_SHRINK_BOUND |
+				      I915_SHRINK_UNBOUND |
+				      I915_SHRINK_ACTIVE |
+				      I915_SHRINK_VMAPS);
+
+	i915_gem_shrinker_unlock_uninterruptible(dev_priv, &slu);
+
+	*(unsigned long *)ptr += freed_pages;
+	return NOTIFY_DONE;
+}
+
 /**
  * i915_gem_shrinker_init - Initialize i915 shrinker
  * @dev_priv: i915 device
@@ -371,6 +418,9 @@ void i915_gem_shrinker_init(struct drm_i915_private *dev_priv)
 
 	dev_priv->mm.oom_notifier.notifier_call = i915_gem_shrinker_oom;
 	WARN_ON(register_oom_notifier(&dev_priv->mm.oom_notifier));
+
+	dev_priv->mm.vmap_notifier.notifier_call = i915_gem_shrinker_vmap;
+	WARN_ON(register_vmap_purge_notifier(&dev_priv->mm.vmap_notifier));
 }
 
 /**
@@ -381,6 +431,7 @@ void i915_gem_shrinker_init(struct drm_i915_private *dev_priv)
  */
 void i915_gem_shrinker_cleanup(struct drm_i915_private *dev_priv)
 {
+	WARN_ON(unregister_vmap_purge_notifier(&dev_priv->mm.vmap_notifier));
 	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
 	unregister_shrinker(&dev_priv->mm.shrinker);
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 2e6e9fb6f80d..b7ce963fb8f8 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -72,9 +72,11 @@ int i915_gem_stolen_insert_node(struct drm_i915_private *dev_priv,
 				struct drm_mm_node *node, u64 size,
 				unsigned alignment)
 {
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+
 	return i915_gem_stolen_insert_node_in_range(dev_priv, node, size,
-					alignment, 0,
-					dev_priv->gtt.stolen_usable_size);
+						    alignment, 0,
+						    ggtt->stolen_usable_size);
 }
 
 void i915_gem_stolen_remove_node(struct drm_i915_private *dev_priv,
@@ -87,14 +89,15 @@ void i915_gem_stolen_remove_node(struct drm_i915_private *dev_priv,
 
 static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct resource *r;
 	u32 base;
 
 	/* Almost universally we can find the Graphics Base of Stolen Memory
-	 * at offset 0x5c in the igfx configuration space. On a few (desktop)
-	 * machines this is also mirrored in the bridge device at different
-	 * locations, or in the MCHBAR.
+	 * at register BSM (0x5c) in the igfx configuration space. On a few
+	 * (desktop) machines this is also mirrored in the bridge device at
+	 * different locations, or in the MCHBAR.
 	 *
 	 * On 865 we just check the TOUD register.
 	 *
@@ -104,9 +107,11 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 	 */
 	base = 0;
 	if (INTEL_INFO(dev)->gen >= 3) {
-		/* Read Graphics Base of Stolen Memory directly */
-		pci_read_config_dword(dev->pdev, 0x5c, &base);
-		base &= ~((1<<20) - 1);
+		u32 bsm;
+
+		pci_read_config_dword(dev->pdev, BSM, &bsm);
+
+		base = bsm & BSM_MASK;
 	} else if (IS_I865G(dev)) {
 		u16 toud = 0;
 
@@ -134,7 +139,7 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 					 I85X_DRB3, &tmp);
 		tom = tmp * MB(32);
 
-		base = tom - tseg_size - dev_priv->gtt.stolen_size;
+		base = tom - tseg_size - ggtt->stolen_size;
 	} else if (IS_845G(dev)) {
 		u32 tseg_size = 0;
 		u32 tom;
@@ -158,7 +163,7 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 					 I830_DRB3, &tmp);
 		tom = tmp * MB(32);
 
-		base = tom - tseg_size - dev_priv->gtt.stolen_size;
+		base = tom - tseg_size - ggtt->stolen_size;
 	} else if (IS_I830(dev)) {
 		u32 tseg_size = 0;
 		u32 tom;
@@ -178,7 +183,7 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 					 I830_DRB3, &tmp);
 		tom = tmp * MB(32);
 
-		base = tom - tseg_size - dev_priv->gtt.stolen_size;
+		base = tom - tseg_size - ggtt->stolen_size;
 	}
 
 	if (base == 0)
@@ -189,41 +194,41 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 		struct {
 			u32 start, end;
 		} stolen[2] = {
-			{ .start = base, .end = base + dev_priv->gtt.stolen_size, },
-			{ .start = base, .end = base + dev_priv->gtt.stolen_size, },
+			{ .start = base, .end = base + ggtt->stolen_size, },
+			{ .start = base, .end = base + ggtt->stolen_size, },
 		};
-		u64 gtt_start, gtt_end;
+		u64 ggtt_start, ggtt_end;
 
-		gtt_start = I915_READ(PGTBL_CTL);
+		ggtt_start = I915_READ(PGTBL_CTL);
 		if (IS_GEN4(dev))
-			gtt_start = (gtt_start & PGTBL_ADDRESS_LO_MASK) |
-				(gtt_start & PGTBL_ADDRESS_HI_MASK) << 28;
+			ggtt_start = (ggtt_start & PGTBL_ADDRESS_LO_MASK) |
+				     (ggtt_start & PGTBL_ADDRESS_HI_MASK) << 28;
 		else
-			gtt_start &= PGTBL_ADDRESS_LO_MASK;
-		gtt_end = gtt_start + gtt_total_entries(dev_priv->gtt) * 4;
+			ggtt_start &= PGTBL_ADDRESS_LO_MASK;
+		ggtt_end = ggtt_start + ggtt_total_entries(ggtt) * 4;
 
-		if (gtt_start >= stolen[0].start && gtt_start < stolen[0].end)
-			stolen[0].end = gtt_start;
-		if (gtt_end > stolen[1].start && gtt_end <= stolen[1].end)
-			stolen[1].start = gtt_end;
+		if (ggtt_start >= stolen[0].start && ggtt_start < stolen[0].end)
+			stolen[0].end = ggtt_start;
+		if (ggtt_end > stolen[1].start && ggtt_end <= stolen[1].end)
+			stolen[1].start = ggtt_end;
 
 		/* pick the larger of the two chunks */
 		if (stolen[0].end - stolen[0].start >
 		    stolen[1].end - stolen[1].start) {
 			base = stolen[0].start;
-			dev_priv->gtt.stolen_size = stolen[0].end - stolen[0].start;
+			ggtt->stolen_size = stolen[0].end - stolen[0].start;
 		} else {
 			base = stolen[1].start;
-			dev_priv->gtt.stolen_size = stolen[1].end - stolen[1].start;
+			ggtt->stolen_size = stolen[1].end - stolen[1].start;
 		}
 
 		if (stolen[0].start != stolen[1].start ||
 		    stolen[0].end != stolen[1].end) {
 			DRM_DEBUG_KMS("GTT within stolen memory at 0x%llx-0x%llx\n",
-				      (unsigned long long) gtt_start,
-				      (unsigned long long) gtt_end - 1);
+				      (unsigned long long)ggtt_start,
+				      (unsigned long long)ggtt_end - 1);
 			DRM_DEBUG_KMS("Stolen memory adjusted to 0x%x-0x%x\n",
-				      base, base + (u32) dev_priv->gtt.stolen_size - 1);
+				      base, base + (u32)ggtt->stolen_size - 1);
 		}
 	}
 
@@ -233,7 +238,7 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 	 * kernel. So if the region is already marked as busy, something
 	 * is seriously wrong.
 	 */
-	r = devm_request_mem_region(dev->dev, base, dev_priv->gtt.stolen_size,
+	r = devm_request_mem_region(dev->dev, base, ggtt->stolen_size,
 				    "Graphics Stolen Memory");
 	if (r == NULL) {
 		/*
@@ -245,7 +250,7 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 		 * reservation starting from 1 instead of 0.
 		 */
 		r = devm_request_mem_region(dev->dev, base + 1,
-					    dev_priv->gtt.stolen_size - 1,
+					    ggtt->stolen_size - 1,
 					    "Graphics Stolen Memory");
 		/*
 		 * GEN3 firmware likes to smash pci bridges into the stolen
@@ -253,7 +258,7 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 		 */
 		if (r == NULL && !IS_GEN3(dev)) {
 			DRM_ERROR("conflict detected with stolen region: [0x%08x - 0x%08x]\n",
-				  base, base + (uint32_t)dev_priv->gtt.stolen_size);
+				  base, base + (uint32_t)ggtt->stolen_size);
 			base = 0;
 		}
 	}
@@ -274,11 +279,12 @@ void i915_gem_cleanup_stolen(struct drm_device *dev)
 static void g4x_get_stolen_reserved(struct drm_i915_private *dev_priv,
 				    unsigned long *base, unsigned long *size)
 {
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	uint32_t reg_val = I915_READ(IS_GM45(dev_priv) ?
 				     CTG_STOLEN_RESERVED :
 				     ELK_STOLEN_RESERVED);
 	unsigned long stolen_top = dev_priv->mm.stolen_base +
-		dev_priv->gtt.stolen_size;
+				   ggtt->stolen_size;
 
 	*base = (reg_val & G4X_STOLEN_RESERVED_ADDR2_MASK) << 16;
 
@@ -369,10 +375,11 @@ static void gen8_get_stolen_reserved(struct drm_i915_private *dev_priv,
 static void bdw_get_stolen_reserved(struct drm_i915_private *dev_priv,
 				    unsigned long *base, unsigned long *size)
 {
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	uint32_t reg_val = I915_READ(GEN6_STOLEN_RESERVED);
 	unsigned long stolen_top;
 
-	stolen_top = dev_priv->mm.stolen_base + dev_priv->gtt.stolen_size;
+	stolen_top = dev_priv->mm.stolen_base + ggtt->stolen_size;
 
 	*base = reg_val & GEN6_STOLEN_RESERVED_ADDR_MASK;
 
@@ -388,7 +395,8 @@ static void bdw_get_stolen_reserved(struct drm_i915_private *dev_priv,
 
 int i915_gem_init_stolen(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	unsigned long reserved_total, reserved_base = 0, reserved_size;
 	unsigned long stolen_top;
 
@@ -401,14 +409,14 @@ int i915_gem_init_stolen(struct drm_device *dev)
 	}
 #endif
 
-	if (dev_priv->gtt.stolen_size == 0)
+	if (ggtt->stolen_size == 0)
 		return 0;
 
 	dev_priv->mm.stolen_base = i915_stolen_to_physical(dev);
 	if (dev_priv->mm.stolen_base == 0)
 		return 0;
 
-	stolen_top = dev_priv->mm.stolen_base + dev_priv->gtt.stolen_size;
+	stolen_top = dev_priv->mm.stolen_base + ggtt->stolen_size;
 
 	switch (INTEL_INFO(dev_priv)->gen) {
 	case 2:
@@ -458,19 +466,18 @@ int i915_gem_init_stolen(struct drm_device *dev)
 		return 0;
 	}
 
-	dev_priv->gtt.stolen_reserved_base = reserved_base;
-	dev_priv->gtt.stolen_reserved_size = reserved_size;
+	ggtt->stolen_reserved_base = reserved_base;
+	ggtt->stolen_reserved_size = reserved_size;
 
 	/* It is possible for the reserved area to end before the end of stolen
 	 * memory, so just consider the start. */
 	reserved_total = stolen_top - reserved_base;
 
 	DRM_DEBUG_KMS("Memory reserved for graphics device: %zuK, usable: %luK\n",
-		      dev_priv->gtt.stolen_size >> 10,
-		      (dev_priv->gtt.stolen_size - reserved_total) >> 10);
+		      ggtt->stolen_size >> 10,
+		      (ggtt->stolen_size - reserved_total) >> 10);
 
-	dev_priv->gtt.stolen_usable_size = dev_priv->gtt.stolen_size -
-					   reserved_total;
+	ggtt->stolen_usable_size = ggtt->stolen_size - reserved_total;
 
 	/*
 	 * Basic memrange allocator for stolen space.
@@ -483,7 +490,7 @@ int i915_gem_init_stolen(struct drm_device *dev)
 	 * i915_gem_stolen_insert_node_in_range(). We may want to fix the fbcon
 	 * problem later.
 	 */
-	drm_mm_init(&dev_priv->mm.stolen, 0, dev_priv->gtt.stolen_usable_size);
+	drm_mm_init(&dev_priv->mm.stolen, 0, ggtt->stolen_usable_size);
 
 	return 0;
 }
@@ -492,12 +499,13 @@ static struct sg_table *
 i915_pages_create_for_stolen(struct drm_device *dev,
 			     u32 offset, u32 size)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct sg_table *st;
 	struct scatterlist *sg;
 
 	DRM_DEBUG_DRIVER("offset=0x%x, size=%d\n", offset, size);
-	BUG_ON(offset > dev_priv->gtt.stolen_size - size);
+	BUG_ON(offset > ggtt->stolen_size - size);
 
 	/* We hide that we have no struct page backing our stolen object
 	 * by wrapping the contiguous physical allocation with a fake
@@ -628,8 +636,8 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 					       u32 gtt_offset,
 					       u32 size)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_address_space *ggtt = &dev_priv->gtt.base;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_object *obj;
 	struct drm_mm_node *stolen;
 	struct i915_vma *vma;
@@ -675,7 +683,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 	if (gtt_offset == I915_GTT_OFFSET_NONE)
 		return obj;
 
-	vma = i915_gem_obj_lookup_or_create_vma(obj, ggtt);
+	vma = i915_gem_obj_lookup_or_create_vma(obj, &ggtt->base);
 	if (IS_ERR(vma)) {
 		ret = PTR_ERR(vma);
 		goto err;
@@ -688,8 +696,8 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 	 */
 	vma->node.start = gtt_offset;
 	vma->node.size = size;
-	if (drm_mm_initialized(&ggtt->mm)) {
-		ret = drm_mm_reserve_node(&ggtt->mm, &vma->node);
+	if (drm_mm_initialized(&ggtt->base.mm)) {
+		ret = drm_mm_reserve_node(&ggtt->base.mm, &vma->node);
 		if (ret) {
 			DRM_DEBUG_KMS("failed to allocate stolen GTT space\n");
 			goto err;
@@ -697,7 +705,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 
 		vma->bound |= GLOBAL_BIND;
 		__i915_vma_set_map_and_fenceable(vma);
-		list_add_tail(&vma->vm_link, &ggtt->inactive_list);
+		list_add_tail(&vma->vm_link, &ggtt->base.inactive_list);
 	}
 
 	list_add_tail(&obj->global_list, &dev_priv->mm.bound_list);
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 7410f6c962e7..b9bdb34032cd 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -166,7 +166,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 	struct drm_i915_gem_object *obj;
 	int ret = 0;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL)
 		return -ENOENT;
 
@@ -297,7 +297,7 @@ i915_gem_get_tiling(struct drm_device *dev, void *data,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
 	if (&obj->base == NULL)
 		return -ENOENT;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 4d30b60defda..32d9726e38b1 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -34,7 +34,7 @@
 
 struct i915_mm_struct {
 	struct mm_struct *mm;
-	struct drm_device *dev;
+	struct drm_i915_private *i915;
 	struct i915_mmu_notifier *mn;
 	struct hlist_node node;
 	struct kref kref;
@@ -49,6 +49,7 @@ struct i915_mmu_notifier {
 	struct hlist_node node;
 	struct mmu_notifier mn;
 	struct rb_root objects;
+	struct workqueue_struct *wq;
 };
 
 struct i915_mmu_object {
@@ -60,6 +61,37 @@ struct i915_mmu_object {
 	bool attached;
 };
 
+static void wait_rendering(struct drm_i915_gem_object *obj)
+{
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_gem_request *requests[I915_NUM_ENGINES];
+	int i, n;
+
+	if (!obj->active)
+		return;
+
+	n = 0;
+	for (i = 0; i < I915_NUM_ENGINES; i++) {
+		struct drm_i915_gem_request *req;
+
+		req = obj->last_read_req[i];
+		if (req == NULL)
+			continue;
+
+		requests[n++] = i915_gem_request_reference(req);
+	}
+
+	mutex_unlock(&dev->struct_mutex);
+
+	for (i = 0; i < n; i++)
+		__i915_wait_request(requests[i], false, NULL, NULL);
+
+	mutex_lock(&dev->struct_mutex);
+
+	for (i = 0; i < n; i++)
+		i915_gem_request_unreference(requests[i]);
+}
+
 static void cancel_userptr(struct work_struct *work)
 {
 	struct i915_mmu_object *mo = container_of(work, typeof(*mo), work);
@@ -75,13 +107,13 @@ static void cancel_userptr(struct work_struct *work)
 		struct i915_vma *vma, *tmp;
 		bool was_interruptible;
 
+		wait_rendering(obj);
+
 		was_interruptible = dev_priv->mm.interruptible;
 		dev_priv->mm.interruptible = false;
 
-		list_for_each_entry_safe(vma, tmp, &obj->vma_list, obj_link) {
-			int ret = i915_vma_unbind(vma);
-			WARN_ON(ret && ret != -EIO);
-		}
+		list_for_each_entry_safe(vma, tmp, &obj->vma_list, obj_link)
+			WARN_ON(i915_vma_unbind(vma));
 		WARN_ON(i915_gem_object_put_pages(obj));
 
 		dev_priv->mm.interruptible = was_interruptible;
@@ -140,7 +172,7 @@ static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 		 */
 		mo = container_of(it, struct i915_mmu_object, it);
 		if (kref_get_unless_zero(&mo->obj->base.refcount))
-			schedule_work(&mo->work);
+			queue_work(mn->wq, &mo->work);
 
 		list_add(&mo->link, &cancelled);
 		it = interval_tree_iter_next(it, start, end);
@@ -148,6 +180,8 @@ static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 	list_for_each_entry(mo, &cancelled, link)
 		del_object(mo);
 	spin_unlock(&mn->lock);
+
+	flush_workqueue(mn->wq);
 }
 
 static const struct mmu_notifier_ops i915_gem_userptr_notifier = {
@@ -167,10 +201,16 @@ i915_mmu_notifier_create(struct mm_struct *mm)
 	spin_lock_init(&mn->lock);
 	mn->mn.ops = &i915_gem_userptr_notifier;
 	mn->objects = RB_ROOT;
+	mn->wq = alloc_workqueue("i915-userptr-release", WQ_UNBOUND, 0);
+	if (mn->wq == NULL) {
+		kfree(mn);
+		return ERR_PTR(-ENOMEM);
+	}
 
 	 /* Protected by mmap_sem (write-lock) */
 	ret = __mmu_notifier_register(&mn->mn, mm);
 	if (ret) {
+		destroy_workqueue(mn->wq);
 		kfree(mn);
 		return ERR_PTR(ret);
 	}
@@ -205,13 +245,13 @@ i915_mmu_notifier_find(struct i915_mm_struct *mm)
 		return mn;
 
 	down_write(&mm->mm->mmap_sem);
-	mutex_lock(&to_i915(mm->dev)->mm_lock);
+	mutex_lock(&mm->i915->mm_lock);
 	if ((mn = mm->mn) == NULL) {
 		mn = i915_mmu_notifier_create(mm->mm);
 		if (!IS_ERR(mn))
 			mm->mn = mn;
 	}
-	mutex_unlock(&to_i915(mm->dev)->mm_lock);
+	mutex_unlock(&mm->i915->mm_lock);
 	up_write(&mm->mm->mmap_sem);
 
 	return mn;
@@ -256,6 +296,7 @@ i915_mmu_notifier_free(struct i915_mmu_notifier *mn,
 		return;
 
 	mmu_notifier_unregister(&mn->mn, mm);
+	destroy_workqueue(mn->wq);
 	kfree(mn);
 }
 
@@ -327,7 +368,7 @@ i915_gem_userptr_init__mm_struct(struct drm_i915_gem_object *obj)
 		}
 
 		kref_init(&mm->kref);
-		mm->dev = obj->base.dev;
+		mm->i915 = to_i915(obj->base.dev);
 
 		mm->mm = current->mm;
 		atomic_inc(&current->mm->mm_count);
@@ -362,7 +403,7 @@ __i915_mm_struct_free(struct kref *kref)
 
 	/* Protected by dev_priv->mm_lock */
 	hash_del(&mm->node);
-	mutex_unlock(&to_i915(mm->dev)->mm_lock);
+	mutex_unlock(&mm->i915->mm_lock);
 
 	INIT_WORK(&mm->work, __i915_mm_struct_free__worker);
 	schedule_work(&mm->work);
@@ -494,10 +535,7 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 	ret = -ENOMEM;
 	pinned = 0;
 
-	pvec = kmalloc(npages*sizeof(struct page *),
-		       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
-	if (pvec == NULL)
-		pvec = drm_malloc_ab(npages, sizeof(struct page *));
+	pvec = drm_malloc_gfp(npages, sizeof(struct page *), GFP_TEMPORARY);
 	if (pvec != NULL) {
 		struct mm_struct *mm = obj->userptr.mm->mm;
 
@@ -639,14 +677,11 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 	pvec = NULL;
 	pinned = 0;
 	if (obj->userptr.mm->mm == current->mm) {
-		pvec = kmalloc(num_pages*sizeof(struct page *),
-			       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
+		pvec = drm_malloc_gfp(num_pages, sizeof(struct page *),
+				      GFP_TEMPORARY);
 		if (pvec == NULL) {
-			pvec = drm_malloc_ab(num_pages, sizeof(struct page *));
-			if (pvec == NULL) {
-				__i915_gem_userptr_set_active(obj, false);
-				return -ENOMEM;
-			}
+			__i915_gem_userptr_set_active(obj, false);
+			return -ENOMEM;
 		}
 
 		pinned = __get_user_pages_fast(obj->userptr.ptr, num_pages,
@@ -763,6 +798,13 @@ i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file
 	int ret;
 	u32 handle;
 
+	if (!HAS_LLC(dev) && !HAS_SNOOP(dev)) {
+		/* We cannot support coherent userptr objects on hw without
+		 * LLC and broken snooping.
+		 */
+		return -ENODEV;
+	}
+
 	if (args->flags & ~(I915_USERPTR_READ_ONLY |
 			    I915_USERPTR_UNSYNCHRONIZED))
 		return -EINVAL;
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 831895b8cb75..89725c9efc25 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -198,7 +198,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 			   err->size,
 			   err->read_domains,
 			   err->write_domain);
-		for (i = 0; i < I915_NUM_RINGS; i++)
+		for (i = 0; i < I915_NUM_ENGINES; i++)
 			err_printf(m, "%02x ", err->rseqno[i]);
 
 		err_printf(m, "] %02x", err->wseqno);
@@ -230,8 +230,6 @@ static const char *hangcheck_action_to_str(enum intel_ring_hangcheck_action a)
 		return "wait";
 	case HANGCHECK_ACTIVE:
 		return "active";
-	case HANGCHECK_ACTIVE_LOOP:
-		return "active (loop)";
 	case HANGCHECK_KICK:
 		return "kick";
 	case HANGCHECK_HUNG:
@@ -298,6 +296,7 @@ static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
 		}
 	}
 	err_printf(m, "  seqno: 0x%08x\n", ring->seqno);
+	err_printf(m, "  last_seqno: 0x%08x\n", ring->last_seqno);
 	err_printf(m, "  waiting: %s\n", yesno(ring->waiting));
 	err_printf(m, "  ring->head: 0x%08x\n", ring->cpu_ring_head);
 	err_printf(m, "  ring->tail: 0x%08x\n", ring->cpu_ring_tail);
@@ -433,7 +432,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
 		obj = error->ring[i].batchbuffer;
 		if (obj) {
-			err_puts(m, dev_priv->ring[i].name);
+			err_puts(m, dev_priv->engine[i].name);
 			if (error->ring[i].pid != -1)
 				err_printf(m, " (submitted by %s [%d])",
 					   error->ring[i].comm,
@@ -447,14 +446,14 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 		obj = error->ring[i].wa_batchbuffer;
 		if (obj) {
 			err_printf(m, "%s (w/a) --- gtt_offset = 0x%08x\n",
-				   dev_priv->ring[i].name,
+				   dev_priv->engine[i].name,
 				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
 
 		if (error->ring[i].num_requests) {
 			err_printf(m, "%s --- %d requests\n",
-				   dev_priv->ring[i].name,
+				   dev_priv->engine[i].name,
 				   error->ring[i].num_requests);
 			for (j = 0; j < error->ring[i].num_requests; j++) {
 				err_printf(m, "  seqno 0x%08x, emitted %ld, tail 0x%08x\n",
@@ -466,7 +465,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 
 		if ((obj = error->ring[i].ringbuffer)) {
 			err_printf(m, "%s --- ringbuffer = 0x%08x\n",
-				   dev_priv->ring[i].name,
+				   dev_priv->engine[i].name,
 				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
@@ -480,7 +479,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 				hws_page = &obj->pages[LRC_PPHWSP_PN][0];
 			}
 			err_printf(m, "%s --- HW Status = 0x%08llx\n",
-				   dev_priv->ring[i].name, hws_offset);
+				   dev_priv->engine[i].name, hws_offset);
 			offset = 0;
 			for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
 				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
@@ -493,9 +492,31 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			}
 		}
 
+		obj = error->ring[i].wa_ctx;
+		if (obj) {
+			u64 wa_ctx_offset = obj->gtt_offset;
+			u32 *wa_ctx_page = &obj->pages[0][0];
+			struct intel_engine_cs *engine = &dev_priv->engine[RCS];
+			u32 wa_ctx_size = (engine->wa_ctx.indirect_ctx.size +
+					   engine->wa_ctx.per_ctx.size);
+
+			err_printf(m, "%s --- WA ctx batch buffer = 0x%08llx\n",
+				   dev_priv->engine[i].name, wa_ctx_offset);
+			offset = 0;
+			for (elt = 0; elt < wa_ctx_size; elt += 4) {
+				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
+					   offset,
+					   wa_ctx_page[elt + 0],
+					   wa_ctx_page[elt + 1],
+					   wa_ctx_page[elt + 2],
+					   wa_ctx_page[elt + 3]);
+				offset += 16;
+			}
+		}
+
 		if ((obj = error->ring[i].ctx)) {
 			err_printf(m, "%s --- HW Context = 0x%08x\n",
-				   dev_priv->ring[i].name,
+				   dev_priv->engine[i].name,
 				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
@@ -585,6 +606,7 @@ static void i915_error_state_free(struct kref *error_ref)
 		i915_error_object_free(error->ring[i].hws_page);
 		i915_error_object_free(error->ring[i].ctx);
 		kfree(error->ring[i].requests);
+		i915_error_object_free(error->ring[i].wa_ctx);
 	}
 
 	i915_error_object_free(error->semaphore_obj);
@@ -606,6 +628,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 			 struct drm_i915_gem_object *src,
 			 struct i915_address_space *vm)
 {
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_error_object *dst;
 	struct i915_vma *vma = NULL;
 	int num_pages;
@@ -632,7 +655,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 		vma = i915_gem_obj_to_ggtt(src);
 	use_ggtt = (src->cache_level == I915_CACHE_NONE &&
 		   vma && (vma->bound & GLOBAL_BIND) &&
-		   reloc_offset + num_pages * PAGE_SIZE <= dev_priv->gtt.mappable_end);
+		   reloc_offset + num_pages * PAGE_SIZE <= ggtt->mappable_end);
 
 	/* Cannot access stolen address directly, try to use the aperture */
 	if (src->stolen) {
@@ -642,12 +665,13 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 			goto unwind;
 
 		reloc_offset = i915_gem_obj_ggtt_offset(src);
-		if (reloc_offset + num_pages * PAGE_SIZE > dev_priv->gtt.mappable_end)
+		if (reloc_offset + num_pages * PAGE_SIZE > ggtt->mappable_end)
 			goto unwind;
 	}
 
 	/* Cannot access snooped pages through the aperture */
-	if (use_ggtt && src->cache_level != I915_CACHE_NONE && !HAS_LLC(dev_priv->dev))
+	if (use_ggtt && src->cache_level != I915_CACHE_NONE &&
+	    !HAS_LLC(dev_priv))
 		goto unwind;
 
 	dst->page_count = num_pages;
@@ -668,7 +692,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 			 * captures what the GPU read.
 			 */
 
-			s = io_mapping_map_atomic_wc(dev_priv->gtt.mappable,
+			s = io_mapping_map_atomic_wc(ggtt->mappable,
 						     reloc_offset);
 			memcpy_fromio(d, s, PAGE_SIZE);
 			io_mapping_unmap_atomic(s);
@@ -701,7 +725,7 @@ unwind:
 	return NULL;
 }
 #define i915_error_ggtt_object_create(dev_priv, src) \
-	i915_error_object_create((dev_priv), (src), &(dev_priv)->gtt.base)
+	i915_error_object_create((dev_priv), (src), &(dev_priv)->ggtt.base)
 
 static void capture_bo(struct drm_i915_error_buffer *err,
 		       struct i915_vma *vma)
@@ -711,7 +735,7 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 
 	err->size = obj->base.size;
 	err->name = obj->base.name;
-	for (i = 0; i < I915_NUM_RINGS; i++)
+	for (i = 0; i < I915_NUM_ENGINES; i++)
 		err->rseqno[i] = i915_gem_request_get_seqno(obj->last_read_req[i]);
 	err->wseqno = i915_gem_request_get_seqno(obj->last_write_req);
 	err->gtt_offset = vma->node.start;
@@ -726,7 +750,7 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 	err->purgeable = obj->madv != I915_MADV_WILLNEED;
 	err->userptr = obj->userptr.mm != NULL;
 	err->ring = obj->last_write_req ?
-			i915_gem_request_get_ring(obj->last_write_req)->id : -1;
+			i915_gem_request_get_engine(obj->last_write_req)->id : -1;
 	err->cache_level = obj->cache_level;
 }
 
@@ -788,7 +812,7 @@ static uint32_t i915_error_generate_code(struct drm_i915_private *dev_priv,
 	 * synchronization commands which almost always appear in the case
 	 * strictly a client bug. Use instdone to differentiate those some.
 	 */
-	for (i = 0; i < I915_NUM_RINGS; i++) {
+	for (i = 0; i < I915_NUM_ENGINES; i++) {
 		if (error->ring[i].hangcheck_action == HANGCHECK_HUNG) {
 			if (ring_id)
 				*ring_id = i;
@@ -821,11 +845,11 @@ static void i915_gem_record_fences(struct drm_device *dev,
 
 static void gen8_record_semaphore_state(struct drm_i915_private *dev_priv,
 					struct drm_i915_error_state *error,
-					struct intel_engine_cs *ring,
+					struct intel_engine_cs *engine,
 					struct drm_i915_error_ring *ering)
 {
 	struct intel_engine_cs *to;
-	int i;
+	enum intel_engine_id id;
 
 	if (!i915_semaphore_is_enabled(dev_priv->dev))
 		return;
@@ -835,68 +859,69 @@ static void gen8_record_semaphore_state(struct drm_i915_private *dev_priv,
 			i915_error_ggtt_object_create(dev_priv,
 						      dev_priv->semaphore_obj);
 
-	for_each_ring(to, dev_priv, i) {
+	for_each_engine_id(to, dev_priv, id) {
 		int idx;
 		u16 signal_offset;
 		u32 *tmp;
 
-		if (ring == to)
+		if (engine == to)
 			continue;
 
-		signal_offset = (GEN8_SIGNAL_OFFSET(ring, i) & (PAGE_SIZE - 1))
+		signal_offset = (GEN8_SIGNAL_OFFSET(engine, id) & (PAGE_SIZE - 1))
 				/ 4;
 		tmp = error->semaphore_obj->pages[0];
-		idx = intel_ring_sync_index(ring, to);
+		idx = intel_ring_sync_index(engine, to);
 
 		ering->semaphore_mboxes[idx] = tmp[signal_offset];
-		ering->semaphore_seqno[idx] = ring->semaphore.sync_seqno[idx];
+		ering->semaphore_seqno[idx] = engine->semaphore.sync_seqno[idx];
 	}
 }
 
 static void gen6_record_semaphore_state(struct drm_i915_private *dev_priv,
-					struct intel_engine_cs *ring,
+					struct intel_engine_cs *engine,
 					struct drm_i915_error_ring *ering)
 {
-	ering->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(ring->mmio_base));
-	ering->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(ring->mmio_base));
-	ering->semaphore_seqno[0] = ring->semaphore.sync_seqno[0];
-	ering->semaphore_seqno[1] = ring->semaphore.sync_seqno[1];
+	ering->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(engine->mmio_base));
+	ering->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(engine->mmio_base));
+	ering->semaphore_seqno[0] = engine->semaphore.sync_seqno[0];
+	ering->semaphore_seqno[1] = engine->semaphore.sync_seqno[1];
 
-	if (HAS_VEBOX(dev_priv->dev)) {
+	if (HAS_VEBOX(dev_priv)) {
 		ering->semaphore_mboxes[2] =
-			I915_READ(RING_SYNC_2(ring->mmio_base));
-		ering->semaphore_seqno[2] = ring->semaphore.sync_seqno[2];
+			I915_READ(RING_SYNC_2(engine->mmio_base));
+		ering->semaphore_seqno[2] = engine->semaphore.sync_seqno[2];
 	}
 }
 
 static void i915_record_ring_state(struct drm_device *dev,
 				   struct drm_i915_error_state *error,
-				   struct intel_engine_cs *ring,
+				   struct intel_engine_cs *engine,
 				   struct drm_i915_error_ring *ering)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (INTEL_INFO(dev)->gen >= 6) {
-		ering->rc_psmi = I915_READ(RING_PSMI_CTL(ring->mmio_base));
-		ering->fault_reg = I915_READ(RING_FAULT_REG(ring));
+		ering->rc_psmi = I915_READ(RING_PSMI_CTL(engine->mmio_base));
+		ering->fault_reg = I915_READ(RING_FAULT_REG(engine));
 		if (INTEL_INFO(dev)->gen >= 8)
-			gen8_record_semaphore_state(dev_priv, error, ring, ering);
+			gen8_record_semaphore_state(dev_priv, error, engine,
+						    ering);
 		else
-			gen6_record_semaphore_state(dev_priv, ring, ering);
+			gen6_record_semaphore_state(dev_priv, engine, ering);
 	}
 
 	if (INTEL_INFO(dev)->gen >= 4) {
-		ering->faddr = I915_READ(RING_DMA_FADD(ring->mmio_base));
-		ering->ipeir = I915_READ(RING_IPEIR(ring->mmio_base));
-		ering->ipehr = I915_READ(RING_IPEHR(ring->mmio_base));
-		ering->instdone = I915_READ(RING_INSTDONE(ring->mmio_base));
-		ering->instps = I915_READ(RING_INSTPS(ring->mmio_base));
-		ering->bbaddr = I915_READ(RING_BBADDR(ring->mmio_base));
+		ering->faddr = I915_READ(RING_DMA_FADD(engine->mmio_base));
+		ering->ipeir = I915_READ(RING_IPEIR(engine->mmio_base));
+		ering->ipehr = I915_READ(RING_IPEHR(engine->mmio_base));
+		ering->instdone = I915_READ(RING_INSTDONE(engine->mmio_base));
+		ering->instps = I915_READ(RING_INSTPS(engine->mmio_base));
+		ering->bbaddr = I915_READ(RING_BBADDR(engine->mmio_base));
 		if (INTEL_INFO(dev)->gen >= 8) {
-			ering->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(ring->mmio_base)) << 32;
-			ering->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(ring->mmio_base)) << 32;
+			ering->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(engine->mmio_base)) << 32;
+			ering->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(engine->mmio_base)) << 32;
 		}
-		ering->bbstate = I915_READ(RING_BBSTATE(ring->mmio_base));
+		ering->bbstate = I915_READ(RING_BBSTATE(engine->mmio_base));
 	} else {
 		ering->faddr = I915_READ(DMA_FADD_I8XX);
 		ering->ipeir = I915_READ(IPEIR);
@@ -904,20 +929,21 @@ static void i915_record_ring_state(struct drm_device *dev,
 		ering->instdone = I915_READ(GEN2_INSTDONE);
 	}
 
-	ering->waiting = waitqueue_active(&ring->irq_queue);
-	ering->instpm = I915_READ(RING_INSTPM(ring->mmio_base));
-	ering->seqno = ring->get_seqno(ring, false);
-	ering->acthd = intel_ring_get_active_head(ring);
-	ering->start = I915_READ_START(ring);
-	ering->head = I915_READ_HEAD(ring);
-	ering->tail = I915_READ_TAIL(ring);
-	ering->ctl = I915_READ_CTL(ring);
+	ering->waiting = waitqueue_active(&engine->irq_queue);
+	ering->instpm = I915_READ(RING_INSTPM(engine->mmio_base));
+	ering->acthd = intel_ring_get_active_head(engine);
+	ering->seqno = engine->get_seqno(engine);
+	ering->last_seqno = engine->last_submitted_seqno;
+	ering->start = I915_READ_START(engine);
+	ering->head = I915_READ_HEAD(engine);
+	ering->tail = I915_READ_TAIL(engine);
+	ering->ctl = I915_READ_CTL(engine);
 
 	if (I915_NEED_GFX_HWS(dev)) {
 		i915_reg_t mmio;
 
 		if (IS_GEN7(dev)) {
-			switch (ring->id) {
+			switch (engine->id) {
 			default:
 			case RCS:
 				mmio = RENDER_HWS_PGA_GEN7;
@@ -932,51 +958,51 @@ static void i915_record_ring_state(struct drm_device *dev,
 				mmio = VEBOX_HWS_PGA_GEN7;
 				break;
 			}
-		} else if (IS_GEN6(ring->dev)) {
-			mmio = RING_HWS_PGA_GEN6(ring->mmio_base);
+		} else if (IS_GEN6(engine->dev)) {
+			mmio = RING_HWS_PGA_GEN6(engine->mmio_base);
 		} else {
 			/* XXX: gen8 returns to sanity */
-			mmio = RING_HWS_PGA(ring->mmio_base);
+			mmio = RING_HWS_PGA(engine->mmio_base);
 		}
 
 		ering->hws = I915_READ(mmio);
 	}
 
-	ering->hangcheck_score = ring->hangcheck.score;
-	ering->hangcheck_action = ring->hangcheck.action;
+	ering->hangcheck_score = engine->hangcheck.score;
+	ering->hangcheck_action = engine->hangcheck.action;
 
 	if (USES_PPGTT(dev)) {
 		int i;
 
-		ering->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(ring));
+		ering->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(engine));
 
 		if (IS_GEN6(dev))
 			ering->vm_info.pp_dir_base =
-				I915_READ(RING_PP_DIR_BASE_READ(ring));
+				I915_READ(RING_PP_DIR_BASE_READ(engine));
 		else if (IS_GEN7(dev))
 			ering->vm_info.pp_dir_base =
-				I915_READ(RING_PP_DIR_BASE(ring));
+				I915_READ(RING_PP_DIR_BASE(engine));
 		else if (INTEL_INFO(dev)->gen >= 8)
 			for (i = 0; i < 4; i++) {
 				ering->vm_info.pdp[i] =
-					I915_READ(GEN8_RING_PDP_UDW(ring, i));
+					I915_READ(GEN8_RING_PDP_UDW(engine, i));
 				ering->vm_info.pdp[i] <<= 32;
 				ering->vm_info.pdp[i] |=
-					I915_READ(GEN8_RING_PDP_LDW(ring, i));
+					I915_READ(GEN8_RING_PDP_LDW(engine, i));
 			}
 	}
 }
 
 
-static void i915_gem_record_active_context(struct intel_engine_cs *ring,
+static void i915_gem_record_active_context(struct intel_engine_cs *engine,
 					   struct drm_i915_error_state *error,
 					   struct drm_i915_error_ring *ering)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	struct drm_i915_gem_object *obj;
 
 	/* Currently render ring is the only HW context user */
-	if (ring->id != RCS || !error->ccid)
+	if (engine->id != RCS || !error->ccid)
 		return;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
@@ -993,30 +1019,31 @@ static void i915_gem_record_active_context(struct intel_engine_cs *ring,
 static void i915_gem_record_rings(struct drm_device *dev,
 				  struct drm_i915_error_state *error)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_request *request;
 	int i, count;
 
-	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_engine_cs *ring = &dev_priv->ring[i];
+	for (i = 0; i < I915_NUM_ENGINES; i++) {
+		struct intel_engine_cs *engine = &dev_priv->engine[i];
 		struct intel_ringbuffer *rbuf;
 
 		error->ring[i].pid = -1;
 
-		if (ring->dev == NULL)
+		if (engine->dev == NULL)
 			continue;
 
 		error->ring[i].valid = true;
 
-		i915_record_ring_state(dev, error, ring, &error->ring[i]);
+		i915_record_ring_state(dev, error, engine, &error->ring[i]);
 
-		request = i915_gem_find_active_request(ring);
+		request = i915_gem_find_active_request(engine);
 		if (request) {
 			struct i915_address_space *vm;
 
 			vm = request->ctx && request->ctx->ppgtt ?
 				&request->ctx->ppgtt->base :
-				&dev_priv->gtt.base;
+				&ggtt->base;
 
 			/* We need to copy these to an anonymous buffer
 			 * as the simplest method to avoid being overwritten
@@ -1027,10 +1054,10 @@ static void i915_gem_record_rings(struct drm_device *dev,
 							 request->batch_obj,
 							 vm);
 
-			if (HAS_BROKEN_CS_TLB(dev_priv->dev))
+			if (HAS_BROKEN_CS_TLB(dev_priv))
 				error->ring[i].wa_batchbuffer =
 					i915_error_ggtt_object_create(dev_priv,
-							     ring->scratch.obj);
+							     engine->scratch.obj);
 
 			if (request->pid) {
 				struct task_struct *task;
@@ -1052,11 +1079,11 @@ static void i915_gem_record_rings(struct drm_device *dev,
 			 * executed).
 			 */
 			if (request)
-				rbuf = request->ctx->engine[ring->id].ringbuf;
+				rbuf = request->ctx->engine[engine->id].ringbuf;
 			else
-				rbuf = dev_priv->kernel_context->engine[ring->id].ringbuf;
+				rbuf = dev_priv->kernel_context->engine[engine->id].ringbuf;
 		} else
-			rbuf = ring->buffer;
+			rbuf = engine->buffer;
 
 		error->ring[i].cpu_ring_head = rbuf->head;
 		error->ring[i].cpu_ring_tail = rbuf->tail;
@@ -1065,12 +1092,19 @@ static void i915_gem_record_rings(struct drm_device *dev,
 			i915_error_ggtt_object_create(dev_priv, rbuf->obj);
 
 		error->ring[i].hws_page =
-			i915_error_ggtt_object_create(dev_priv, ring->status_page.obj);
+			i915_error_ggtt_object_create(dev_priv,
+						      engine->status_page.obj);
+
+		if (engine->wa_ctx.obj) {
+			error->ring[i].wa_ctx =
+				i915_error_ggtt_object_create(dev_priv,
+							      engine->wa_ctx.obj);
+		}
 
-		i915_gem_record_active_context(ring, error, &error->ring[i]);
+		i915_gem_record_active_context(engine, error, &error->ring[i]);
 
 		count = 0;
-		list_for_each_entry(request, &ring->request_list, list)
+		list_for_each_entry(request, &engine->request_list, list)
 			count++;
 
 		error->ring[i].num_requests = count;
@@ -1083,7 +1117,7 @@ static void i915_gem_record_rings(struct drm_device *dev,
 		}
 
 		count = 0;
-		list_for_each_entry(request, &ring->request_list, list) {
+		list_for_each_entry(request, &engine->request_list, list) {
 			struct drm_i915_error_request *erq;
 
 			if (count >= error->ring[i].num_requests) {
@@ -1272,7 +1306,7 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 
 static void i915_error_capture_msg(struct drm_device *dev,
 				   struct drm_i915_error_state *error,
-				   bool wedged,
+				   u32 engine_mask,
 				   const char *error_msg)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1295,7 +1329,7 @@ static void i915_error_capture_msg(struct drm_device *dev,
 	scnprintf(error->error_msg + len, sizeof(error->error_msg) - len,
 		  ", reason: %s, action: %s",
 		  error_msg,
-		  wedged ? "reset" : "continue");
+		  engine_mask ? "reset" : "continue");
 }
 
 static void i915_capture_gen_state(struct drm_i915_private *dev_priv,
@@ -1318,7 +1352,7 @@ static void i915_capture_gen_state(struct drm_i915_private *dev_priv,
  * out a structure which becomes available in debugfs for user level tools
  * to pick up.
  */
-void i915_capture_error_state(struct drm_device *dev, bool wedged,
+void i915_capture_error_state(struct drm_device *dev, u32 engine_mask,
 			      const char *error_msg)
 {
 	static bool warned;
@@ -1346,7 +1380,7 @@ void i915_capture_error_state(struct drm_device *dev, bool wedged,
 	error->overlay = intel_overlay_capture_error_state(dev);
 	error->display = intel_display_capture_error_state(dev);
 
-	i915_error_capture_msg(dev, error, wedged, error_msg);
+	i915_error_capture_msg(dev, error, engine_mask, error_msg);
 	DRM_INFO("%s\n", error->error_msg);
 
 	spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
diff --git a/drivers/gpu/drm/i915/i915_guc_reg.h b/drivers/gpu/drm/i915/i915_guc_reg.h
index e4ba5822289b..80786d9f9ad3 100644
--- a/drivers/gpu/drm/i915/i915_guc_reg.h
+++ b/drivers/gpu/drm/i915/i915_guc_reg.h
@@ -27,9 +27,12 @@
 /* Definitions of GuC H/W registers, bits, etc */
 
 #define GUC_STATUS			_MMIO(0xc000)
+#define   GS_RESET_SHIFT		0
+#define   GS_MIA_IN_RESET		  (0x01 << GS_RESET_SHIFT)
 #define   GS_BOOTROM_SHIFT		1
 #define   GS_BOOTROM_MASK		  (0x7F << GS_BOOTROM_SHIFT)
 #define   GS_BOOTROM_RSA_FAILED		  (0x50 << GS_BOOTROM_SHIFT)
+#define   GS_BOOTROM_JUMP_PASSED	  (0x76 << GS_BOOTROM_SHIFT)
 #define   GS_UKERNEL_SHIFT		8
 #define   GS_UKERNEL_MASK		  (0xFF << GS_UKERNEL_SHIFT)
 #define   GS_UKERNEL_LAPIC_DONE		  (0x30 << GS_UKERNEL_SHIFT)
@@ -37,7 +40,13 @@
 #define   GS_UKERNEL_READY		  (0xF0 << GS_UKERNEL_SHIFT)
 #define   GS_MIA_SHIFT			16
 #define   GS_MIA_MASK			  (0x07 << GS_MIA_SHIFT)
-#define   GS_MIA_CORE_STATE		  (1 << GS_MIA_SHIFT)
+#define   GS_MIA_CORE_STATE		  (0x01 << GS_MIA_SHIFT)
+#define   GS_MIA_HALT_REQUESTED		  (0x02 << GS_MIA_SHIFT)
+#define   GS_MIA_ISR_ENTRY		  (0x04 << GS_MIA_SHIFT)
+#define   GS_AUTH_STATUS_SHIFT		30
+#define   GS_AUTH_STATUS_MASK		  (0x03 << GS_AUTH_STATUS_SHIFT)
+#define   GS_AUTH_STATUS_BAD		  (0x01 << GS_AUTH_STATUS_SHIFT)
+#define   GS_AUTH_STATUS_GOOD		  (0x02 << GS_AUTH_STATUS_SHIFT)
 
 #define SOFT_SCRATCH(n)			_MMIO(0xc180 + (n) * 4)
 #define SOFT_SCRATCH_COUNT		16
diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index d7543efc8a5e..d40c13fb6643 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -179,15 +179,11 @@ static void guc_init_doorbell(struct intel_guc *guc,
 			      struct i915_guc_client *client)
 {
 	struct guc_doorbell_info *doorbell;
-	void *base;
 
-	base = kmap_atomic(i915_gem_object_get_page(client->client_obj, 0));
-	doorbell = base + client->doorbell_offset;
+	doorbell = client->client_base + client->doorbell_offset;
 
-	doorbell->db_status = 1;
+	doorbell->db_status = GUC_DOORBELL_ENABLED;
 	doorbell->cookie = 0;
-
-	kunmap_atomic(base);
 }
 
 static int guc_ring_doorbell(struct i915_guc_client *gc)
@@ -195,11 +191,9 @@ static int guc_ring_doorbell(struct i915_guc_client *gc)
 	struct guc_process_desc *desc;
 	union guc_doorbell_qw db_cmp, db_exc, db_ret;
 	union guc_doorbell_qw *db;
-	void *base;
 	int attempt = 2, ret = -EAGAIN;
 
-	base = kmap_atomic(i915_gem_object_get_page(gc->client_obj, 0));
-	desc = base + gc->proc_desc_offset;
+	desc = gc->client_base + gc->proc_desc_offset;
 
 	/* Update the tail so it is visible to GuC */
 	desc->tail = gc->wq_tail;
@@ -215,7 +209,7 @@ static int guc_ring_doorbell(struct i915_guc_client *gc)
 		db_exc.cookie = 1;
 
 	/* pointer of current doorbell cacheline */
-	db = base + gc->doorbell_offset;
+	db = gc->client_base + gc->doorbell_offset;
 
 	while (attempt--) {
 		/* lets ring the doorbell */
@@ -244,10 +238,6 @@ static int guc_ring_doorbell(struct i915_guc_client *gc)
 			db_exc.cookie = 1;
 	}
 
-	/* Finally, update the cached copy of the GuC's WQ head */
-	gc->wq_head = desc->head;
-
-	kunmap_atomic(base);
 	return ret;
 }
 
@@ -256,16 +246,12 @@ static void guc_disable_doorbell(struct intel_guc *guc,
 {
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
 	struct guc_doorbell_info *doorbell;
-	void *base;
 	i915_reg_t drbreg = GEN8_DRBREGL(client->doorbell_id);
 	int value;
 
-	base = kmap_atomic(i915_gem_object_get_page(client->client_obj, 0));
-	doorbell = base + client->doorbell_offset;
+	doorbell = client->client_base + client->doorbell_offset;
 
-	doorbell->db_status = 0;
-
-	kunmap_atomic(base);
+	doorbell->db_status = GUC_DOORBELL_DISABLED;
 
 	I915_WRITE(drbreg, I915_READ(drbreg) & ~GEN8_DRB_VALID);
 
@@ -341,10 +327,8 @@ static void guc_init_proc_desc(struct intel_guc *guc,
 			       struct i915_guc_client *client)
 {
 	struct guc_process_desc *desc;
-	void *base;
 
-	base = kmap_atomic(i915_gem_object_get_page(client->client_obj, 0));
-	desc = base + client->proc_desc_offset;
+	desc = client->client_base + client->proc_desc_offset;
 
 	memset(desc, 0, sizeof(*desc));
 
@@ -361,8 +345,6 @@ static void guc_init_proc_desc(struct intel_guc *guc,
 	desc->wq_size_bytes = client->wq_size;
 	desc->wq_status = WQ_STATUS_ACTIVE;
 	desc->priority = client->priority;
-
-	kunmap_atomic(base);
 }
 
 /*
@@ -376,12 +358,14 @@ static void guc_init_proc_desc(struct intel_guc *guc,
 static void guc_init_ctx_desc(struct intel_guc *guc,
 			      struct i915_guc_client *client)
 {
+	struct drm_i915_gem_object *client_obj = client->client_obj;
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct intel_context *ctx = client->owner;
 	struct guc_context_desc desc;
 	struct sg_table *sg;
-	int i;
+	enum intel_engine_id id;
+	u32 gfx_addr;
 
 	memset(&desc, 0, sizeof(desc));
 
@@ -390,8 +374,8 @@ static void guc_init_ctx_desc(struct intel_guc *guc,
 	desc.priority = client->priority;
 	desc.db_id = client->doorbell_id;
 
-	for_each_ring(ring, dev_priv, i) {
-		struct guc_execlist_context *lrc = &desc.lrc[ring->guc_id];
+	for_each_engine_id(engine, dev_priv, id) {
+		struct guc_execlist_context *lrc = &desc.lrc[engine->guc_id];
 		struct drm_i915_gem_object *obj;
 		uint64_t ctx_desc;
 
@@ -402,48 +386,44 @@ static void guc_init_ctx_desc(struct intel_guc *guc,
 		 * for now who owns a GuC client. But for future owner of GuC
 		 * client, need to make sure lrc is pinned prior to enter here.
 		 */
-		obj = ctx->engine[i].state;
+		obj = ctx->engine[id].state;
 		if (!obj)
 			break;	/* XXX: continue? */
 
-		ctx_desc = intel_lr_context_descriptor(ctx, ring);
+		ctx_desc = intel_lr_context_descriptor(ctx, engine);
 		lrc->context_desc = (u32)ctx_desc;
 
 		/* The state page is after PPHWSP */
-		lrc->ring_lcra = i915_gem_obj_ggtt_offset(obj) +
-				LRC_STATE_PN * PAGE_SIZE;
+		gfx_addr = i915_gem_obj_ggtt_offset(obj);
+		lrc->ring_lcra = gfx_addr + LRC_STATE_PN * PAGE_SIZE;
 		lrc->context_id = (client->ctx_index << GUC_ELC_CTXID_OFFSET) |
-				(ring->guc_id << GUC_ELC_ENGINE_OFFSET);
+				(engine->guc_id << GUC_ELC_ENGINE_OFFSET);
 
-		obj = ctx->engine[i].ringbuf->obj;
+		obj = ctx->engine[id].ringbuf->obj;
+		gfx_addr = i915_gem_obj_ggtt_offset(obj);
 
-		lrc->ring_begin = i915_gem_obj_ggtt_offset(obj);
-		lrc->ring_end = lrc->ring_begin + obj->base.size - 1;
-		lrc->ring_next_free_location = lrc->ring_begin;
+		lrc->ring_begin = gfx_addr;
+		lrc->ring_end = gfx_addr + obj->base.size - 1;
+		lrc->ring_next_free_location = gfx_addr;
 		lrc->ring_current_tail_pointer_value = 0;
 
-		desc.engines_used |= (1 << ring->guc_id);
+		desc.engines_used |= (1 << engine->guc_id);
 	}
 
 	WARN_ON(desc.engines_used == 0);
 
 	/*
-	 * The CPU address is only needed at certain points, so kmap_atomic on
-	 * demand instead of storing it in the ctx descriptor.
-	 * XXX: May make debug easier to have it mapped
+	 * The doorbell, process descriptor, and workqueue are all parts
+	 * of the client object, which the GuC will reference via the GGTT
 	 */
-	desc.db_trigger_cpu = 0;
-	desc.db_trigger_uk = client->doorbell_offset +
-		i915_gem_obj_ggtt_offset(client->client_obj);
-	desc.db_trigger_phy = client->doorbell_offset +
-		sg_dma_address(client->client_obj->pages->sgl);
-
-	desc.process_desc = client->proc_desc_offset +
-		i915_gem_obj_ggtt_offset(client->client_obj);
-
-	desc.wq_addr = client->wq_offset +
-		i915_gem_obj_ggtt_offset(client->client_obj);
-
+	gfx_addr = i915_gem_obj_ggtt_offset(client_obj);
+	desc.db_trigger_phy = sg_dma_address(client_obj->pages->sgl) +
+				client->doorbell_offset;
+	desc.db_trigger_cpu = (uintptr_t)client->client_base +
+				client->doorbell_offset;
+	desc.db_trigger_uk = gfx_addr + client->doorbell_offset;
+	desc.process_desc = gfx_addr + client->proc_desc_offset;
+	desc.wq_addr = gfx_addr + client->wq_offset;
 	desc.wq_size = client->wq_size;
 
 	/*
@@ -474,25 +454,16 @@ static void guc_fini_ctx_desc(struct intel_guc *guc,
 int i915_guc_wq_check_space(struct i915_guc_client *gc)
 {
 	struct guc_process_desc *desc;
-	void *base;
 	u32 size = sizeof(struct guc_wq_item);
 	int ret = -ETIMEDOUT, timeout_counter = 200;
 
 	if (!gc)
 		return 0;
 
-	/* Quickly return if wq space is available since last time we cache the
-	 * head position. */
-	if (CIRC_SPACE(gc->wq_tail, gc->wq_head, gc->wq_size) >= size)
-		return 0;
-
-	base = kmap_atomic(i915_gem_object_get_page(gc->client_obj, 0));
-	desc = base + gc->proc_desc_offset;
+	desc = gc->client_base + gc->proc_desc_offset;
 
 	while (timeout_counter-- > 0) {
-		gc->wq_head = desc->head;
-
-		if (CIRC_SPACE(gc->wq_tail, gc->wq_head, gc->wq_size) >= size) {
+		if (CIRC_SPACE(gc->wq_tail, desc->head, gc->wq_size) >= size) {
 			ret = 0;
 			break;
 		}
@@ -501,19 +472,19 @@ int i915_guc_wq_check_space(struct i915_guc_client *gc)
 			usleep_range(1000, 2000);
 	};
 
-	kunmap_atomic(base);
-
 	return ret;
 }
 
 static int guc_add_workqueue_item(struct i915_guc_client *gc,
 				  struct drm_i915_gem_request *rq)
 {
+	struct guc_process_desc *desc;
 	struct guc_wq_item *wqi;
 	void *base;
 	u32 tail, wq_len, wq_off, space;
 
-	space = CIRC_SPACE(gc->wq_tail, gc->wq_head, gc->wq_size);
+	desc = gc->client_base + gc->proc_desc_offset;
+	space = CIRC_SPACE(gc->wq_tail, desc->head, gc->wq_size);
 	if (WARN_ON(space < sizeof(struct guc_wq_item)))
 		return -ENOSPC; /* shouldn't happen */
 
@@ -542,11 +513,12 @@ static int guc_add_workqueue_item(struct i915_guc_client *gc,
 	wq_len = sizeof(struct guc_wq_item) / sizeof(u32) - 1;
 	wqi->header = WQ_TYPE_INORDER |
 			(wq_len << WQ_LEN_SHIFT) |
-			(rq->ring->guc_id << WQ_TARGET_SHIFT) |
+			(rq->engine->guc_id << WQ_TARGET_SHIFT) |
 			WQ_NO_WCFLUSH_WAIT;
 
 	/* The GuC wants only the low-order word of the context descriptor */
-	wqi->context_desc = (u32)intel_lr_context_descriptor(rq->ctx, rq->ring);
+	wqi->context_desc = (u32)intel_lr_context_descriptor(rq->ctx,
+							     rq->engine);
 
 	/* The GuC firmware wants the tail index in QWords, not bytes */
 	tail = rq->ringbuf->tail >> 3;
@@ -569,7 +541,7 @@ int i915_guc_submit(struct i915_guc_client *client,
 		    struct drm_i915_gem_request *rq)
 {
 	struct intel_guc *guc = client->guc;
-	unsigned int engine_id = rq->ring->guc_id;
+	unsigned int engine_id = rq->engine->guc_id;
 	int q_ret, b_ret;
 
 	q_ret = guc_add_workqueue_item(client, rq);
@@ -660,21 +632,28 @@ static void guc_client_free(struct drm_device *dev,
 	if (!client)
 		return;
 
-	if (client->doorbell_id != GUC_INVALID_DOORBELL_ID) {
-		/*
-		 * First disable the doorbell, then tell the GuC we've
-		 * finished with it, finally deallocate it in our bitmap
-		 */
-		guc_disable_doorbell(guc, client);
-		host2guc_release_doorbell(guc, client);
-		release_doorbell(guc, client->doorbell_id);
-	}
-
 	/*
 	 * XXX: wait for any outstanding submissions before freeing memory.
 	 * Be sure to drop any locks
 	 */
 
+	if (client->client_base) {
+		/*
+		 * If we got as far as setting up a doorbell, make sure
+		 * we shut it down before unmapping & deallocating the
+		 * memory. So first disable the doorbell, then tell the
+		 * GuC that we've finished with it, finally deallocate
+		 * it in our bitmap
+		 */
+		if (client->doorbell_id != GUC_INVALID_DOORBELL_ID) {
+			guc_disable_doorbell(guc, client);
+			host2guc_release_doorbell(guc, client);
+			release_doorbell(guc, client->doorbell_id);
+		}
+
+		kunmap(kmap_to_page(client->client_base));
+	}
+
 	gem_release_guc_obj(client->client_obj);
 
 	if (client->ctx_index != GUC_INVALID_CTX_ID) {
@@ -695,7 +674,7 @@ static void guc_client_free(struct drm_device *dev,
  * @ctx:	the context that owns the client (we use the default render
  * 		context)
  *
- * Return:	An i915_guc_client object if success.
+ * Return:	An i915_guc_client object if success, else NULL.
  */
 static struct i915_guc_client *guc_client_alloc(struct drm_device *dev,
 						uint32_t priority,
@@ -727,7 +706,9 @@ static struct i915_guc_client *guc_client_alloc(struct drm_device *dev,
 	if (!obj)
 		goto err;
 
+	/* We'll keep just the first (doorbell/proc) page permanently kmap'd. */
 	client->client_obj = obj;
+	client->client_base = kmap(i915_gem_object_get_page(obj, 0));
 	client->wq_offset = GUC_DB_SIZE;
 	client->wq_size = GUC_WQ_SIZE;
 
@@ -839,9 +820,9 @@ static void guc_create_ads(struct intel_guc *guc)
 	struct guc_ads *ads;
 	struct guc_policies *policies;
 	struct guc_mmio_reg_state *reg_state;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct page *page;
-	u32 size, i;
+	u32 size;
 
 	/* The ads obj includes the struct itself and buffers passed to GuC */
 	size = sizeof(struct guc_ads) + sizeof(struct guc_policies) +
@@ -867,11 +848,11 @@ static void guc_create_ads(struct intel_guc *guc)
 	 * so its address won't change after we've told the GuC where
 	 * to find it.
 	 */
-	ring = &dev_priv->ring[RCS];
-	ads->golden_context_lrca = ring->status_page.gfx_addr;
+	engine = &dev_priv->engine[RCS];
+	ads->golden_context_lrca = engine->status_page.gfx_addr;
 
-	for_each_ring(ring, dev_priv, i)
-		ads->eng_state_size[ring->guc_id] = intel_lr_context_size(ring);
+	for_each_engine(engine, dev_priv)
+		ads->eng_state_size[engine->guc_id] = intel_lr_context_size(engine);
 
 	/* GuC scheduling policies */
 	policies = (void *)ads + sizeof(struct guc_ads);
@@ -883,12 +864,12 @@ static void guc_create_ads(struct intel_guc *guc)
 	/* MMIO reg state */
 	reg_state = (void *)policies + sizeof(struct guc_policies);
 
-	for_each_ring(ring, dev_priv, i) {
-		reg_state->mmio_white_list[ring->guc_id].mmio_start =
-			ring->mmio_base + GUC_MMIO_WHITE_LIST_START;
+	for_each_engine(engine, dev_priv) {
+		reg_state->mmio_white_list[engine->guc_id].mmio_start =
+			engine->mmio_base + GUC_MMIO_WHITE_LIST_START;
 
 		/* Nothing to be saved or restored for now. */
-		reg_state->mmio_white_list[ring->guc_id].count = 0;
+		reg_state->mmio_white_list[engine->guc_id].count = 0;
 	}
 
 	ads->reg_state_addr = ads->scheduler_policies +
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 1c212205d0e7..2f6fd33c07ba 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -994,14 +994,15 @@ static void ironlake_rps_change_irq_handler(struct drm_device *dev)
 	return;
 }
 
-static void notify_ring(struct intel_engine_cs *ring)
+static void notify_ring(struct intel_engine_cs *engine)
 {
-	if (!intel_ring_initialized(ring))
+	if (!intel_engine_initialized(engine))
 		return;
 
-	trace_i915_gem_request_notify(ring);
+	trace_i915_gem_request_notify(engine);
+	engine->user_interrupts++;
 
-	wake_up_all(&ring->irq_queue);
+	wake_up_all(&engine->irq_queue);
 }
 
 static void vlv_c0_read(struct drm_i915_private *dev_priv,
@@ -1079,11 +1080,10 @@ static u32 vlv_wa_c0_ei(struct drm_i915_private *dev_priv, u32 pm_iir)
 
 static bool any_waiters(struct drm_i915_private *dev_priv)
 {
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
-	for_each_ring(ring, dev_priv, i)
-		if (ring->irq_refcount)
+	for_each_engine(engine, dev_priv)
+		if (engine->irq_refcount)
 			return true;
 
 	return false;
@@ -1219,7 +1219,7 @@ static void ivybridge_parity_work(struct work_struct *work)
 		i915_reg_t reg;
 
 		slice--;
-		if (WARN_ON_ONCE(slice >= NUM_L3_SLICES(dev_priv->dev)))
+		if (WARN_ON_ONCE(slice >= NUM_L3_SLICES(dev_priv)))
 			break;
 
 		dev_priv->l3_parity.which_slice &= ~(1<<slice);
@@ -1258,24 +1258,23 @@ static void ivybridge_parity_work(struct work_struct *work)
 out:
 	WARN_ON(dev_priv->l3_parity.which_slice);
 	spin_lock_irq(&dev_priv->irq_lock);
-	gen5_enable_gt_irq(dev_priv, GT_PARITY_ERROR(dev_priv->dev));
+	gen5_enable_gt_irq(dev_priv, GT_PARITY_ERROR(dev_priv));
 	spin_unlock_irq(&dev_priv->irq_lock);
 
 	mutex_unlock(&dev_priv->dev->struct_mutex);
 }
 
-static void ivybridge_parity_error_irq_handler(struct drm_device *dev, u32 iir)
+static void ivybridge_parity_error_irq_handler(struct drm_i915_private *dev_priv,
+					       u32 iir)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (!HAS_L3_DPF(dev))
+	if (!HAS_L3_DPF(dev_priv))
 		return;
 
 	spin_lock(&dev_priv->irq_lock);
-	gen5_disable_gt_irq(dev_priv, GT_PARITY_ERROR(dev));
+	gen5_disable_gt_irq(dev_priv, GT_PARITY_ERROR(dev_priv));
 	spin_unlock(&dev_priv->irq_lock);
 
-	iir &= GT_PARITY_ERROR(dev);
+	iir &= GT_PARITY_ERROR(dev_priv);
 	if (iir & GT_RENDER_L3_PARITY_ERROR_INTERRUPT_S1)
 		dev_priv->l3_parity.which_slice |= 1 << 1;
 
@@ -1285,102 +1284,85 @@ static void ivybridge_parity_error_irq_handler(struct drm_device *dev, u32 iir)
 	queue_work(dev_priv->wq, &dev_priv->l3_parity.error_work);
 }
 
-static void ilk_gt_irq_handler(struct drm_device *dev,
-			       struct drm_i915_private *dev_priv,
+static void ilk_gt_irq_handler(struct drm_i915_private *dev_priv,
 			       u32 gt_iir)
 {
 	if (gt_iir &
 	    (GT_RENDER_USER_INTERRUPT | GT_RENDER_PIPECTL_NOTIFY_INTERRUPT))
-		notify_ring(&dev_priv->ring[RCS]);
+		notify_ring(&dev_priv->engine[RCS]);
 	if (gt_iir & ILK_BSD_USER_INTERRUPT)
-		notify_ring(&dev_priv->ring[VCS]);
+		notify_ring(&dev_priv->engine[VCS]);
 }
 
-static void snb_gt_irq_handler(struct drm_device *dev,
-			       struct drm_i915_private *dev_priv,
+static void snb_gt_irq_handler(struct drm_i915_private *dev_priv,
 			       u32 gt_iir)
 {
 
 	if (gt_iir &
 	    (GT_RENDER_USER_INTERRUPT | GT_RENDER_PIPECTL_NOTIFY_INTERRUPT))
-		notify_ring(&dev_priv->ring[RCS]);
+		notify_ring(&dev_priv->engine[RCS]);
 	if (gt_iir & GT_BSD_USER_INTERRUPT)
-		notify_ring(&dev_priv->ring[VCS]);
+		notify_ring(&dev_priv->engine[VCS]);
 	if (gt_iir & GT_BLT_USER_INTERRUPT)
-		notify_ring(&dev_priv->ring[BCS]);
+		notify_ring(&dev_priv->engine[BCS]);
 
 	if (gt_iir & (GT_BLT_CS_ERROR_INTERRUPT |
 		      GT_BSD_CS_ERROR_INTERRUPT |
 		      GT_RENDER_CS_MASTER_ERROR_INTERRUPT))
 		DRM_DEBUG("Command parser error, gt_iir 0x%08x\n", gt_iir);
 
-	if (gt_iir & GT_PARITY_ERROR(dev))
-		ivybridge_parity_error_irq_handler(dev, gt_iir);
+	if (gt_iir & GT_PARITY_ERROR(dev_priv))
+		ivybridge_parity_error_irq_handler(dev_priv, gt_iir);
 }
 
 static __always_inline void
-gen8_cs_irq_handler(struct intel_engine_cs *ring, u32 iir, int test_shift)
+gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir, int test_shift)
 {
 	if (iir & (GT_RENDER_USER_INTERRUPT << test_shift))
-		notify_ring(ring);
+		notify_ring(engine);
 	if (iir & (GT_CONTEXT_SWITCH_INTERRUPT << test_shift))
-		intel_lrc_irq_handler(ring);
+		tasklet_schedule(&engine->irq_tasklet);
 }
 
-static irqreturn_t gen8_gt_irq_handler(struct drm_i915_private *dev_priv,
-				       u32 master_ctl)
+static irqreturn_t gen8_gt_irq_ack(struct drm_i915_private *dev_priv,
+				   u32 master_ctl,
+				   u32 gt_iir[4])
 {
 	irqreturn_t ret = IRQ_NONE;
 
 	if (master_ctl & (GEN8_GT_RCS_IRQ | GEN8_GT_BCS_IRQ)) {
-		u32 iir = I915_READ_FW(GEN8_GT_IIR(0));
-		if (iir) {
-			I915_WRITE_FW(GEN8_GT_IIR(0), iir);
+		gt_iir[0] = I915_READ_FW(GEN8_GT_IIR(0));
+		if (gt_iir[0]) {
+			I915_WRITE_FW(GEN8_GT_IIR(0), gt_iir[0]);
 			ret = IRQ_HANDLED;
-
-			gen8_cs_irq_handler(&dev_priv->ring[RCS],
-					iir, GEN8_RCS_IRQ_SHIFT);
-
-			gen8_cs_irq_handler(&dev_priv->ring[BCS],
-					iir, GEN8_BCS_IRQ_SHIFT);
 		} else
 			DRM_ERROR("The master control interrupt lied (GT0)!\n");
 	}
 
 	if (master_ctl & (GEN8_GT_VCS1_IRQ | GEN8_GT_VCS2_IRQ)) {
-		u32 iir = I915_READ_FW(GEN8_GT_IIR(1));
-		if (iir) {
-			I915_WRITE_FW(GEN8_GT_IIR(1), iir);
+		gt_iir[1] = I915_READ_FW(GEN8_GT_IIR(1));
+		if (gt_iir[1]) {
+			I915_WRITE_FW(GEN8_GT_IIR(1), gt_iir[1]);
 			ret = IRQ_HANDLED;
-
-			gen8_cs_irq_handler(&dev_priv->ring[VCS],
-					iir, GEN8_VCS1_IRQ_SHIFT);
-
-			gen8_cs_irq_handler(&dev_priv->ring[VCS2],
-					iir, GEN8_VCS2_IRQ_SHIFT);
 		} else
 			DRM_ERROR("The master control interrupt lied (GT1)!\n");
 	}
 
 	if (master_ctl & GEN8_GT_VECS_IRQ) {
-		u32 iir = I915_READ_FW(GEN8_GT_IIR(3));
-		if (iir) {
-			I915_WRITE_FW(GEN8_GT_IIR(3), iir);
+		gt_iir[3] = I915_READ_FW(GEN8_GT_IIR(3));
+		if (gt_iir[3]) {
+			I915_WRITE_FW(GEN8_GT_IIR(3), gt_iir[3]);
 			ret = IRQ_HANDLED;
-
-			gen8_cs_irq_handler(&dev_priv->ring[VECS],
-					iir, GEN8_VECS_IRQ_SHIFT);
 		} else
 			DRM_ERROR("The master control interrupt lied (GT3)!\n");
 	}
 
 	if (master_ctl & GEN8_GT_PM_IRQ) {
-		u32 iir = I915_READ_FW(GEN8_GT_IIR(2));
-		if (iir & dev_priv->pm_rps_events) {
+		gt_iir[2] = I915_READ_FW(GEN8_GT_IIR(2));
+		if (gt_iir[2] & dev_priv->pm_rps_events) {
 			I915_WRITE_FW(GEN8_GT_IIR(2),
-				      iir & dev_priv->pm_rps_events);
+				      gt_iir[2] & dev_priv->pm_rps_events);
 			ret = IRQ_HANDLED;
-			gen6_rps_irq_handler(dev_priv, iir);
 		} else
 			DRM_ERROR("The master control interrupt lied (PM)!\n");
 	}
@@ -1388,6 +1370,31 @@ static irqreturn_t gen8_gt_irq_handler(struct drm_i915_private *dev_priv,
 	return ret;
 }
 
+static void gen8_gt_irq_handler(struct drm_i915_private *dev_priv,
+				u32 gt_iir[4])
+{
+	if (gt_iir[0]) {
+		gen8_cs_irq_handler(&dev_priv->engine[RCS],
+				    gt_iir[0], GEN8_RCS_IRQ_SHIFT);
+		gen8_cs_irq_handler(&dev_priv->engine[BCS],
+				    gt_iir[0], GEN8_BCS_IRQ_SHIFT);
+	}
+
+	if (gt_iir[1]) {
+		gen8_cs_irq_handler(&dev_priv->engine[VCS],
+				    gt_iir[1], GEN8_VCS1_IRQ_SHIFT);
+		gen8_cs_irq_handler(&dev_priv->engine[VCS2],
+				    gt_iir[1], GEN8_VCS2_IRQ_SHIFT);
+	}
+
+	if (gt_iir[3])
+		gen8_cs_irq_handler(&dev_priv->engine[VECS],
+				    gt_iir[3], GEN8_VECS_IRQ_SHIFT);
+
+	if (gt_iir[2] & dev_priv->pm_rps_events)
+		gen6_rps_irq_handler(dev_priv, gt_iir[2]);
+}
+
 static bool bxt_port_hotplug_long_detect(enum port port, u32 val)
 {
 	switch (port) {
@@ -1627,9 +1634,9 @@ static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
 	if (INTEL_INFO(dev_priv)->gen >= 8)
 		return;
 
-	if (HAS_VEBOX(dev_priv->dev)) {
+	if (HAS_VEBOX(dev_priv)) {
 		if (pm_iir & PM_VEBOX_USER_INTERRUPT)
-			notify_ring(&dev_priv->ring[VECS]);
+			notify_ring(&dev_priv->engine[VECS]);
 
 		if (pm_iir & PM_VEBOX_CS_ERROR_INTERRUPT)
 			DRM_DEBUG("Command parser error, pm_iir 0x%08x\n", pm_iir);
@@ -1644,10 +1651,10 @@ static bool intel_pipe_handle_vblank(struct drm_device *dev, enum pipe pipe)
 	return true;
 }
 
-static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
+static void valleyview_pipestat_irq_ack(struct drm_device *dev, u32 iir,
+					u32 pipe_stats[I915_MAX_PIPES])
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pipe_stats[I915_MAX_PIPES] = { };
 	int pipe;
 
 	spin_lock(&dev_priv->irq_lock);
@@ -1701,6 +1708,13 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 			I915_WRITE(reg, pipe_stats[pipe]);
 	}
 	spin_unlock(&dev_priv->irq_lock);
+}
+
+static void valleyview_pipestat_irq_handler(struct drm_device *dev,
+					    u32 pipe_stats[I915_MAX_PIPES])
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	enum pipe pipe;
 
 	for_each_pipe(dev_priv, pipe) {
 		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS &&
@@ -1723,21 +1737,20 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 		gmbus_irq_handler(dev);
 }
 
-static void i9xx_hpd_irq_handler(struct drm_device *dev)
+static u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-	u32 pin_mask = 0, long_mask = 0;
 
-	if (!hotplug_status)
-		return;
+	if (hotplug_status)
+		I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
 
-	I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
-	/*
-	 * Make sure hotplug status is cleared before we clear IIR, or else we
-	 * may miss hotplug events.
-	 */
-	POSTING_READ(PORT_HOTPLUG_STAT);
+	return hotplug_status;
+}
+
+static void i9xx_hpd_irq_handler(struct drm_device *dev,
+				 u32 hotplug_status)
+{
+	u32 pin_mask = 0, long_mask = 0;
 
 	if (IS_G4X(dev) || IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
 		u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
@@ -1768,7 +1781,6 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 {
 	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 iir, gt_iir, pm_iir;
 	irqreturn_t ret = IRQ_NONE;
 
 	if (!intel_irqs_enabled(dev_priv))
@@ -1777,40 +1789,72 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 	/* IRQs are synced during runtime_suspend, we don't require a wakeref */
 	disable_rpm_wakeref_asserts(dev_priv);
 
-	while (true) {
-		/* Find, clear, then process each source of interrupt */
+	do {
+		u32 iir, gt_iir, pm_iir;
+		u32 pipe_stats[I915_MAX_PIPES] = {};
+		u32 hotplug_status = 0;
+		u32 ier = 0;
 
 		gt_iir = I915_READ(GTIIR);
-		if (gt_iir)
-			I915_WRITE(GTIIR, gt_iir);
-
 		pm_iir = I915_READ(GEN6_PMIIR);
-		if (pm_iir)
-			I915_WRITE(GEN6_PMIIR, pm_iir);
-
 		iir = I915_READ(VLV_IIR);
-		if (iir) {
-			/* Consume port before clearing IIR or we'll miss events */
-			if (iir & I915_DISPLAY_PORT_INTERRUPT)
-				i9xx_hpd_irq_handler(dev);
-			I915_WRITE(VLV_IIR, iir);
-		}
 
 		if (gt_iir == 0 && pm_iir == 0 && iir == 0)
-			goto out;
+			break;
 
 		ret = IRQ_HANDLED;
 
+		/*
+		 * Theory on interrupt generation, based on empirical evidence:
+		 *
+		 * x = ((VLV_IIR & VLV_IER) ||
+		 *      (((GT_IIR & GT_IER) || (GEN6_PMIIR & GEN6_PMIER)) &&
+		 *       (VLV_MASTER_IER & MASTER_INTERRUPT_ENABLE)));
+		 *
+		 * A CPU interrupt will only be raised when 'x' has a 0->1 edge.
+		 * Hence we clear MASTER_INTERRUPT_ENABLE and VLV_IER to
+		 * guarantee the CPU interrupt will be raised again even if we
+		 * don't end up clearing all the VLV_IIR, GT_IIR, GEN6_PMIIR
+		 * bits this time around.
+		 */
+		I915_WRITE(VLV_MASTER_IER, 0);
+		ier = I915_READ(VLV_IER);
+		I915_WRITE(VLV_IER, 0);
+
 		if (gt_iir)
-			snb_gt_irq_handler(dev, dev_priv, gt_iir);
+			I915_WRITE(GTIIR, gt_iir);
 		if (pm_iir)
-			gen6_rps_irq_handler(dev_priv, pm_iir);
+			I915_WRITE(GEN6_PMIIR, pm_iir);
+
+		if (iir & I915_DISPLAY_PORT_INTERRUPT)
+			hotplug_status = i9xx_hpd_irq_ack(dev_priv);
+
 		/* Call regardless, as some status bits might not be
 		 * signalled in iir */
-		valleyview_pipestat_irq_handler(dev, iir);
-	}
+		valleyview_pipestat_irq_ack(dev, iir, pipe_stats);
+
+		/*
+		 * VLV_IIR is single buffered, and reflects the level
+		 * from PIPESTAT/PORT_HOTPLUG_STAT, hence clear it last.
+		 */
+		if (iir)
+			I915_WRITE(VLV_IIR, iir);
+
+		I915_WRITE(VLV_IER, ier);
+		I915_WRITE(VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE);
+		POSTING_READ(VLV_MASTER_IER);
+
+		if (gt_iir)
+			snb_gt_irq_handler(dev_priv, gt_iir);
+		if (pm_iir)
+			gen6_rps_irq_handler(dev_priv, pm_iir);
+
+		if (hotplug_status)
+			i9xx_hpd_irq_handler(dev, hotplug_status);
+
+		valleyview_pipestat_irq_handler(dev, pipe_stats);
+	} while (0);
 
-out:
 	enable_rpm_wakeref_asserts(dev_priv);
 
 	return ret;
@@ -1820,7 +1864,6 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 {
 	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 master_ctl, iir;
 	irqreturn_t ret = IRQ_NONE;
 
 	if (!intel_irqs_enabled(dev_priv))
@@ -1830,6 +1873,12 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 	disable_rpm_wakeref_asserts(dev_priv);
 
 	do {
+		u32 master_ctl, iir;
+		u32 gt_iir[4] = {};
+		u32 pipe_stats[I915_MAX_PIPES] = {};
+		u32 hotplug_status = 0;
+		u32 ier = 0;
+
 		master_ctl = I915_READ(GEN8_MASTER_IRQ) & ~GEN8_MASTER_IRQ_CONTROL;
 		iir = I915_READ(VLV_IIR);
 
@@ -1838,25 +1887,49 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 
 		ret = IRQ_HANDLED;
 
+		/*
+		 * Theory on interrupt generation, based on empirical evidence:
+		 *
+		 * x = ((VLV_IIR & VLV_IER) ||
+		 *      ((GEN8_MASTER_IRQ & ~GEN8_MASTER_IRQ_CONTROL) &&
+		 *       (GEN8_MASTER_IRQ & GEN8_MASTER_IRQ_CONTROL)));
+		 *
+		 * A CPU interrupt will only be raised when 'x' has a 0->1 edge.
+		 * Hence we clear GEN8_MASTER_IRQ_CONTROL and VLV_IER to
+		 * guarantee the CPU interrupt will be raised again even if we
+		 * don't end up clearing all the VLV_IIR and GEN8_MASTER_IRQ_CONTROL
+		 * bits this time around.
+		 */
 		I915_WRITE(GEN8_MASTER_IRQ, 0);
+		ier = I915_READ(VLV_IER);
+		I915_WRITE(VLV_IER, 0);
 
-		/* Find, clear, then process each source of interrupt */
-
-		if (iir) {
-			/* Consume port before clearing IIR or we'll miss events */
-			if (iir & I915_DISPLAY_PORT_INTERRUPT)
-				i9xx_hpd_irq_handler(dev);
-			I915_WRITE(VLV_IIR, iir);
-		}
+		gen8_gt_irq_ack(dev_priv, master_ctl, gt_iir);
 
-		gen8_gt_irq_handler(dev_priv, master_ctl);
+		if (iir & I915_DISPLAY_PORT_INTERRUPT)
+			hotplug_status = i9xx_hpd_irq_ack(dev_priv);
 
 		/* Call regardless, as some status bits might not be
 		 * signalled in iir */
-		valleyview_pipestat_irq_handler(dev, iir);
+		valleyview_pipestat_irq_ack(dev, iir, pipe_stats);
+
+		/*
+		 * VLV_IIR is single buffered, and reflects the level
+		 * from PIPESTAT/PORT_HOTPLUG_STAT, hence clear it last.
+		 */
+		if (iir)
+			I915_WRITE(VLV_IIR, iir);
 
-		I915_WRITE(GEN8_MASTER_IRQ, DE_MASTER_IRQ_CONTROL);
+		I915_WRITE(VLV_IER, ier);
+		I915_WRITE(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
 		POSTING_READ(GEN8_MASTER_IRQ);
+
+		gen8_gt_irq_handler(dev_priv, gt_iir);
+
+		if (hotplug_status)
+			i9xx_hpd_irq_handler(dev, hotplug_status);
+
+		valleyview_pipestat_irq_handler(dev, pipe_stats);
 	} while (0);
 
 	enable_rpm_wakeref_asserts(dev_priv);
@@ -2217,9 +2290,9 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 		I915_WRITE(GTIIR, gt_iir);
 		ret = IRQ_HANDLED;
 		if (INTEL_INFO(dev)->gen >= 6)
-			snb_gt_irq_handler(dev, dev_priv, gt_iir);
+			snb_gt_irq_handler(dev_priv, gt_iir);
 		else
-			ilk_gt_irq_handler(dev, dev_priv, gt_iir);
+			ilk_gt_irq_handler(dev_priv, gt_iir);
 	}
 
 	de_iir = I915_READ(DEIIR);
@@ -2419,6 +2492,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 master_ctl;
+	u32 gt_iir[4] = {};
 	irqreturn_t ret;
 
 	if (!intel_irqs_enabled(dev_priv))
@@ -2435,7 +2509,8 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 	disable_rpm_wakeref_asserts(dev_priv);
 
 	/* Find, clear, then process each source of interrupt */
-	ret = gen8_gt_irq_handler(dev_priv, master_ctl);
+	ret = gen8_gt_irq_ack(dev_priv, master_ctl, gt_iir);
+	gen8_gt_irq_handler(dev_priv, gt_iir);
 	ret |= gen8_de_irq_handler(dev_priv, master_ctl);
 
 	I915_WRITE_FW(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
@@ -2449,8 +2524,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 static void i915_error_wake_up(struct drm_i915_private *dev_priv,
 			       bool reset_completed)
 {
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
 	/*
 	 * Notify all waiters for GPU completion events that reset state has
@@ -2460,8 +2534,8 @@ static void i915_error_wake_up(struct drm_i915_private *dev_priv,
 	 */
 
 	/* Wake up __wait_seqno, potentially holding dev->struct_mutex. */
-	for_each_ring(ring, dev_priv, i)
-		wake_up_all(&ring->irq_queue);
+	for_each_engine(engine, dev_priv)
+		wake_up_all(&engine->irq_queue);
 
 	/* Wake up intel_crtc_wait_for_pending_flips, holding crtc->mutex. */
 	wake_up_all(&dev_priv->pending_flip_queue);
@@ -2484,7 +2558,6 @@ static void i915_error_wake_up(struct drm_i915_private *dev_priv,
 static void i915_reset_and_wakeup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct i915_gpu_error *error = &dev_priv->gpu_error;
 	char *error_event[] = { I915_ERROR_UEVENT "=1", NULL };
 	char *reset_event[] = { I915_RESET_UEVENT "=1", NULL };
 	char *reset_done_event[] = { I915_ERROR_UEVENT "=0", NULL };
@@ -2502,7 +2575,7 @@ static void i915_reset_and_wakeup(struct drm_device *dev)
 	 * the reset in-progress bit is only ever set by code outside of this
 	 * work we don't need to worry about any other races.
 	 */
-	if (i915_reset_in_progress(error) && !i915_terminally_wedged(error)) {
+	if (i915_reset_in_progress(&dev_priv->gpu_error)) {
 		DRM_DEBUG_DRIVER("resetting chip\n");
 		kobject_uevent_env(&dev->primary->kdev->kobj, KOBJ_CHANGE,
 				   reset_event);
@@ -2530,25 +2603,9 @@ static void i915_reset_and_wakeup(struct drm_device *dev)
 
 		intel_runtime_pm_put(dev_priv);
 
-		if (ret == 0) {
-			/*
-			 * After all the gem state is reset, increment the reset
-			 * counter and wake up everyone waiting for the reset to
-			 * complete.
-			 *
-			 * Since unlock operations are a one-sided barrier only,
-			 * we need to insert a barrier here to order any seqno
-			 * updates before
-			 * the counter increment.
-			 */
-			smp_mb__before_atomic();
-			atomic_inc(&dev_priv->gpu_error.reset_counter);
-
+		if (ret == 0)
 			kobject_uevent_env(&dev->primary->kdev->kobj,
 					   KOBJ_CHANGE, reset_done_event);
-		} else {
-			atomic_or(I915_WEDGED, &error->reset_counter);
-		}
 
 		/*
 		 * Note: The wake_up also serves as a memory barrier so that
@@ -2653,14 +2710,14 @@ static void i915_report_and_clear_eir(struct drm_device *dev)
 /**
  * i915_handle_error - handle a gpu error
  * @dev: drm device
- *
+ * @engine_mask: mask representing engines that are hung
  * Do some basic checking of register state at error time and
  * dump it to the syslog.  Also call i915_capture_error_state() to make
  * sure we get a record and make it available in debugfs.  Fire a uevent
  * so userspace knows something bad happened (should trigger collection
  * of a ring dump etc.).
  */
-void i915_handle_error(struct drm_device *dev, bool wedged,
+void i915_handle_error(struct drm_device *dev, u32 engine_mask,
 		       const char *fmt, ...)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2671,10 +2728,10 @@ void i915_handle_error(struct drm_device *dev, bool wedged,
 	vscnprintf(error_msg, sizeof(error_msg), fmt, args);
 	va_end(args);
 
-	i915_capture_error_state(dev, wedged, error_msg);
+	i915_capture_error_state(dev, engine_mask, error_msg);
 	i915_report_and_clear_eir(dev);
 
-	if (wedged) {
+	if (engine_mask) {
 		atomic_or(I915_RESET_IN_PROGRESS_FLAG,
 				&dev_priv->gpu_error.reset_counter);
 
@@ -2805,10 +2862,10 @@ static void gen8_disable_vblank(struct drm_device *dev, unsigned int pipe)
 }
 
 static bool
-ring_idle(struct intel_engine_cs *ring, u32 seqno)
+ring_idle(struct intel_engine_cs *engine, u32 seqno)
 {
-	return (list_empty(&ring->request_list) ||
-		i915_seqno_passed(seqno, ring->last_submitted_seqno));
+	return i915_seqno_passed(seqno,
+				 READ_ONCE(engine->last_submitted_seqno));
 }
 
 static bool
@@ -2824,42 +2881,42 @@ ipehr_is_semaphore_wait(struct drm_device *dev, u32 ipehr)
 }
 
 static struct intel_engine_cs *
-semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr, u64 offset)
+semaphore_wait_to_signaller_ring(struct intel_engine_cs *engine, u32 ipehr,
+				 u64 offset)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	struct intel_engine_cs *signaller;
-	int i;
 
-	if (INTEL_INFO(dev_priv->dev)->gen >= 8) {
-		for_each_ring(signaller, dev_priv, i) {
-			if (ring == signaller)
+	if (INTEL_INFO(dev_priv)->gen >= 8) {
+		for_each_engine(signaller, dev_priv) {
+			if (engine == signaller)
 				continue;
 
-			if (offset == signaller->semaphore.signal_ggtt[ring->id])
+			if (offset == signaller->semaphore.signal_ggtt[engine->id])
 				return signaller;
 		}
 	} else {
 		u32 sync_bits = ipehr & MI_SEMAPHORE_SYNC_MASK;
 
-		for_each_ring(signaller, dev_priv, i) {
-			if(ring == signaller)
+		for_each_engine(signaller, dev_priv) {
+			if(engine == signaller)
 				continue;
 
-			if (sync_bits == signaller->semaphore.mbox.wait[ring->id])
+			if (sync_bits == signaller->semaphore.mbox.wait[engine->id])
 				return signaller;
 		}
 	}
 
 	DRM_ERROR("No signaller ring found for ring %i, ipehr 0x%08x, offset 0x%016llx\n",
-		  ring->id, ipehr, offset);
+		  engine->id, ipehr, offset);
 
 	return NULL;
 }
 
 static struct intel_engine_cs *
-semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
+semaphore_waits_for(struct intel_engine_cs *engine, u32 *seqno)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	u32 cmd, ipehr, head;
 	u64 offset = 0;
 	int i, backwards;
@@ -2881,11 +2938,11 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 	 * Therefore, this function does not support execlist mode in its
 	 * current form. Just return NULL and move on.
 	 */
-	if (ring->buffer == NULL)
+	if (engine->buffer == NULL)
 		return NULL;
 
-	ipehr = I915_READ(RING_IPEHR(ring->mmio_base));
-	if (!ipehr_is_semaphore_wait(ring->dev, ipehr))
+	ipehr = I915_READ(RING_IPEHR(engine->mmio_base));
+	if (!ipehr_is_semaphore_wait(engine->dev, ipehr))
 		return NULL;
 
 	/*
@@ -2896,8 +2953,8 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 	 * point at at batch, and semaphores are always emitted into the
 	 * ringbuffer itself.
 	 */
-	head = I915_READ_HEAD(ring) & HEAD_ADDR;
-	backwards = (INTEL_INFO(ring->dev)->gen >= 8) ? 5 : 4;
+	head = I915_READ_HEAD(engine) & HEAD_ADDR;
+	backwards = (INTEL_INFO(engine->dev)->gen >= 8) ? 5 : 4;
 
 	for (i = backwards; i; --i) {
 		/*
@@ -2905,10 +2962,10 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 		 * our ring is smaller than what the hardware (and hence
 		 * HEAD_ADDR) allows. Also handles wrap-around.
 		 */
-		head &= ring->buffer->size - 1;
+		head &= engine->buffer->size - 1;
 
 		/* This here seems to blow up */
-		cmd = ioread32(ring->buffer->virtual_start + head);
+		cmd = ioread32(engine->buffer->virtual_start + head);
 		if (cmd == ipehr)
 			break;
 
@@ -2918,32 +2975,32 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 	if (!i)
 		return NULL;
 
-	*seqno = ioread32(ring->buffer->virtual_start + head + 4) + 1;
-	if (INTEL_INFO(ring->dev)->gen >= 8) {
-		offset = ioread32(ring->buffer->virtual_start + head + 12);
+	*seqno = ioread32(engine->buffer->virtual_start + head + 4) + 1;
+	if (INTEL_INFO(engine->dev)->gen >= 8) {
+		offset = ioread32(engine->buffer->virtual_start + head + 12);
 		offset <<= 32;
-		offset = ioread32(ring->buffer->virtual_start + head + 8);
+		offset = ioread32(engine->buffer->virtual_start + head + 8);
 	}
-	return semaphore_wait_to_signaller_ring(ring, ipehr, offset);
+	return semaphore_wait_to_signaller_ring(engine, ipehr, offset);
 }
 
-static int semaphore_passed(struct intel_engine_cs *ring)
+static int semaphore_passed(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	struct intel_engine_cs *signaller;
 	u32 seqno;
 
-	ring->hangcheck.deadlock++;
+	engine->hangcheck.deadlock++;
 
-	signaller = semaphore_waits_for(ring, &seqno);
+	signaller = semaphore_waits_for(engine, &seqno);
 	if (signaller == NULL)
 		return -1;
 
 	/* Prevent pathological recursion due to driver bugs */
-	if (signaller->hangcheck.deadlock >= I915_NUM_RINGS)
+	if (signaller->hangcheck.deadlock >= I915_NUM_ENGINES)
 		return -1;
 
-	if (i915_seqno_passed(signaller->get_seqno(signaller, false), seqno))
+	if (i915_seqno_passed(signaller->get_seqno(signaller), seqno))
 		return 1;
 
 	/* cursory check for an unkickable deadlock */
@@ -2956,23 +3013,22 @@ static int semaphore_passed(struct intel_engine_cs *ring)
 
 static void semaphore_clear_deadlocks(struct drm_i915_private *dev_priv)
 {
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
-	for_each_ring(ring, dev_priv, i)
-		ring->hangcheck.deadlock = 0;
+	for_each_engine(engine, dev_priv)
+		engine->hangcheck.deadlock = 0;
 }
 
-static bool subunits_stuck(struct intel_engine_cs *ring)
+static bool subunits_stuck(struct intel_engine_cs *engine)
 {
 	u32 instdone[I915_NUM_INSTDONE_REG];
 	bool stuck;
 	int i;
 
-	if (ring->id != RCS)
+	if (engine->id != RCS)
 		return true;
 
-	i915_get_extra_instdone(ring->dev, instdone);
+	i915_get_extra_instdone(engine->dev, instdone);
 
 	/* There might be unstable subunit states even when
 	 * actual head is not moving. Filter out the unstable ones by
@@ -2981,49 +3037,44 @@ static bool subunits_stuck(struct intel_engine_cs *ring)
 	 */
 	stuck = true;
 	for (i = 0; i < I915_NUM_INSTDONE_REG; i++) {
-		const u32 tmp = instdone[i] | ring->hangcheck.instdone[i];
+		const u32 tmp = instdone[i] | engine->hangcheck.instdone[i];
 
-		if (tmp != ring->hangcheck.instdone[i])
+		if (tmp != engine->hangcheck.instdone[i])
 			stuck = false;
 
-		ring->hangcheck.instdone[i] |= tmp;
+		engine->hangcheck.instdone[i] |= tmp;
 	}
 
 	return stuck;
 }
 
 static enum intel_ring_hangcheck_action
-head_stuck(struct intel_engine_cs *ring, u64 acthd)
+head_stuck(struct intel_engine_cs *engine, u64 acthd)
 {
-	if (acthd != ring->hangcheck.acthd) {
+	if (acthd != engine->hangcheck.acthd) {
 
 		/* Clear subunit states on head movement */
-		memset(ring->hangcheck.instdone, 0,
-		       sizeof(ring->hangcheck.instdone));
+		memset(engine->hangcheck.instdone, 0,
+		       sizeof(engine->hangcheck.instdone));
 
-		if (acthd > ring->hangcheck.max_acthd) {
-			ring->hangcheck.max_acthd = acthd;
-			return HANGCHECK_ACTIVE;
-		}
-
-		return HANGCHECK_ACTIVE_LOOP;
+		return HANGCHECK_ACTIVE;
 	}
 
-	if (!subunits_stuck(ring))
+	if (!subunits_stuck(engine))
 		return HANGCHECK_ACTIVE;
 
 	return HANGCHECK_HUNG;
 }
 
 static enum intel_ring_hangcheck_action
-ring_stuck(struct intel_engine_cs *ring, u64 acthd)
+ring_stuck(struct intel_engine_cs *engine, u64 acthd)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum intel_ring_hangcheck_action ha;
 	u32 tmp;
 
-	ha = head_stuck(ring, acthd);
+	ha = head_stuck(engine, acthd);
 	if (ha != HANGCHECK_HUNG)
 		return ha;
 
@@ -3035,24 +3086,24 @@ ring_stuck(struct intel_engine_cs *ring, u64 acthd)
 	 * and break the hang. This should work on
 	 * all but the second generation chipsets.
 	 */
-	tmp = I915_READ_CTL(ring);
+	tmp = I915_READ_CTL(engine);
 	if (tmp & RING_WAIT) {
-		i915_handle_error(dev, false,
+		i915_handle_error(dev, 0,
 				  "Kicking stuck wait on %s",
-				  ring->name);
-		I915_WRITE_CTL(ring, tmp);
+				  engine->name);
+		I915_WRITE_CTL(engine, tmp);
 		return HANGCHECK_KICK;
 	}
 
 	if (INTEL_INFO(dev)->gen >= 6 && tmp & RING_WAIT_SEMAPHORE) {
-		switch (semaphore_passed(ring)) {
+		switch (semaphore_passed(engine)) {
 		default:
 			return HANGCHECK_HUNG;
 		case 1:
-			i915_handle_error(dev, false,
+			i915_handle_error(dev, 0,
 					  "Kicking stuck semaphore on %s",
-					  ring->name);
-			I915_WRITE_CTL(ring, tmp);
+					  engine->name);
+			I915_WRITE_CTL(engine, tmp);
 			return HANGCHECK_KICK;
 		case 0:
 			return HANGCHECK_WAIT;
@@ -3062,6 +3113,24 @@ ring_stuck(struct intel_engine_cs *ring, u64 acthd)
 	return HANGCHECK_HUNG;
 }
 
+static unsigned kick_waiters(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *i915 = to_i915(engine->dev);
+	unsigned user_interrupts = READ_ONCE(engine->user_interrupts);
+
+	if (engine->hangcheck.user_interrupts == user_interrupts &&
+	    !test_and_set_bit(engine->id, &i915->gpu_error.missed_irq_rings)) {
+		if (!(i915->gpu_error.test_irq_rings & intel_engine_flag(engine)))
+			DRM_ERROR("Hangcheck timer elapsed... %s idle\n",
+				  engine->name);
+		else
+			DRM_INFO("Fake missed irq on %s\n",
+				 engine->name);
+		wake_up_all(&engine->irq_queue);
+	}
+
+	return user_interrupts;
+}
 /*
  * This is called when the chip hasn't reported back with completed
  * batchbuffers in a long time. We keep track per ring seqno progress and
@@ -3076,13 +3145,14 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 		container_of(work, typeof(*dev_priv),
 			     gpu_error.hangcheck_work.work);
 	struct drm_device *dev = dev_priv->dev;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
 	int busy_count = 0, rings_hung = 0;
-	bool stuck[I915_NUM_RINGS] = { 0 };
+	bool stuck[I915_NUM_ENGINES] = { 0 };
 #define BUSY 1
 #define KICK 5
 #define HUNG 20
+#define ACTIVE_DECAY 15
 
 	if (!i915.enable_hangcheck)
 		return;
@@ -3100,33 +3170,37 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 	 */
 	intel_uncore_arm_unclaimed_mmio_detection(dev_priv);
 
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine_id(engine, dev_priv, id) {
 		u64 acthd;
 		u32 seqno;
+		unsigned user_interrupts;
 		bool busy = true;
 
 		semaphore_clear_deadlocks(dev_priv);
 
-		seqno = ring->get_seqno(ring, false);
-		acthd = intel_ring_get_active_head(ring);
-
-		if (ring->hangcheck.seqno == seqno) {
-			if (ring_idle(ring, seqno)) {
-				ring->hangcheck.action = HANGCHECK_IDLE;
-
-				if (waitqueue_active(&ring->irq_queue)) {
-					/* Issue a wake-up to catch stuck h/w. */
-					if (!test_and_set_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings)) {
-						if (!(dev_priv->gpu_error.test_irq_rings & intel_ring_flag(ring)))
-							DRM_ERROR("Hangcheck timer elapsed... %s idle\n",
-								  ring->name);
-						else
-							DRM_INFO("Fake missed irq on %s\n",
-								 ring->name);
-						wake_up_all(&ring->irq_queue);
-					}
+		/* We don't strictly need an irq-barrier here, as we are not
+		 * serving an interrupt request, be paranoid in case the
+		 * barrier has side-effects (such as preventing a broken
+		 * cacheline snoop) and so be sure that we can see the seqno
+		 * advance. If the seqno should stick, due to a stale
+		 * cacheline, we would erroneously declare the GPU hung.
+		 */
+		if (engine->irq_seqno_barrier)
+			engine->irq_seqno_barrier(engine);
+
+		acthd = intel_ring_get_active_head(engine);
+		seqno = engine->get_seqno(engine);
+
+		/* Reset stuck interrupts between batch advances */
+		user_interrupts = 0;
+
+		if (engine->hangcheck.seqno == seqno) {
+			if (ring_idle(engine, seqno)) {
+				engine->hangcheck.action = HANGCHECK_IDLE;
+				if (waitqueue_active(&engine->irq_queue)) {
 					/* Safeguard against driver failure */
-					ring->hangcheck.score += BUSY;
+					user_interrupts = kick_waiters(engine);
+					engine->hangcheck.score += BUSY;
 				} else
 					busy = false;
 			} else {
@@ -3145,58 +3219,60 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 				 * being repeatedly kicked and so responsible
 				 * for stalling the machine.
 				 */
-				ring->hangcheck.action = ring_stuck(ring,
-								    acthd);
+				engine->hangcheck.action = ring_stuck(engine,
+								      acthd);
 
-				switch (ring->hangcheck.action) {
+				switch (engine->hangcheck.action) {
 				case HANGCHECK_IDLE:
 				case HANGCHECK_WAIT:
-				case HANGCHECK_ACTIVE:
 					break;
-				case HANGCHECK_ACTIVE_LOOP:
-					ring->hangcheck.score += BUSY;
+				case HANGCHECK_ACTIVE:
+					engine->hangcheck.score += BUSY;
 					break;
 				case HANGCHECK_KICK:
-					ring->hangcheck.score += KICK;
+					engine->hangcheck.score += KICK;
 					break;
 				case HANGCHECK_HUNG:
-					ring->hangcheck.score += HUNG;
-					stuck[i] = true;
+					engine->hangcheck.score += HUNG;
+					stuck[id] = true;
 					break;
 				}
 			}
 		} else {
-			ring->hangcheck.action = HANGCHECK_ACTIVE;
+			engine->hangcheck.action = HANGCHECK_ACTIVE;
 
 			/* Gradually reduce the count so that we catch DoS
 			 * attempts across multiple batches.
 			 */
-			if (ring->hangcheck.score > 0)
-				ring->hangcheck.score--;
+			if (engine->hangcheck.score > 0)
+				engine->hangcheck.score -= ACTIVE_DECAY;
+			if (engine->hangcheck.score < 0)
+				engine->hangcheck.score = 0;
 
 			/* Clear head and subunit states on seqno movement */
-			ring->hangcheck.acthd = ring->hangcheck.max_acthd = 0;
+			acthd = 0;
 
-			memset(ring->hangcheck.instdone, 0,
-			       sizeof(ring->hangcheck.instdone));
+			memset(engine->hangcheck.instdone, 0,
+			       sizeof(engine->hangcheck.instdone));
 		}
 
-		ring->hangcheck.seqno = seqno;
-		ring->hangcheck.acthd = acthd;
+		engine->hangcheck.seqno = seqno;
+		engine->hangcheck.acthd = acthd;
+		engine->hangcheck.user_interrupts = user_interrupts;
 		busy_count += busy;
 	}
 
-	for_each_ring(ring, dev_priv, i) {
-		if (ring->hangcheck.score >= HANGCHECK_SCORE_RING_HUNG) {
+	for_each_engine_id(engine, dev_priv, id) {
+		if (engine->hangcheck.score >= HANGCHECK_SCORE_RING_HUNG) {
 			DRM_INFO("%s on %s\n",
-				 stuck[i] ? "stuck" : "no progress",
-				 ring->name);
-			rings_hung++;
+				 stuck[id] ? "stuck" : "no progress",
+				 engine->name);
+			rings_hung |= intel_engine_flag(engine);
 		}
 	}
 
 	if (rings_hung) {
-		i915_handle_error(dev, true, "Ring hung");
+		i915_handle_error(dev, rings_hung, "Engine(s) hung");
 		goto out;
 	}
 
@@ -3267,6 +3343,55 @@ static void gen5_gt_irq_reset(struct drm_device *dev)
 		GEN5_IRQ_RESET(GEN6_PM);
 }
 
+static void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
+{
+	enum pipe pipe;
+
+	if (IS_CHERRYVIEW(dev_priv))
+		I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
+	else
+		I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK);
+
+	i915_hotplug_interrupt_update_locked(dev_priv, 0xffffffff, 0);
+	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
+
+	for_each_pipe(dev_priv, pipe) {
+		I915_WRITE(PIPESTAT(pipe),
+			   PIPE_FIFO_UNDERRUN_STATUS |
+			   PIPESTAT_INT_STATUS_MASK);
+		dev_priv->pipestat_irq_mask[pipe] = 0;
+	}
+
+	GEN5_IRQ_RESET(VLV_);
+	dev_priv->irq_mask = ~0;
+}
+
+static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	u32 pipestat_mask;
+	u32 enable_mask;
+	enum pipe pipe;
+
+	pipestat_mask = PLANE_FLIP_DONE_INT_STATUS_VLV |
+			PIPE_CRC_DONE_INTERRUPT_STATUS;
+
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
+	for_each_pipe(dev_priv, pipe)
+		i915_enable_pipestat(dev_priv, pipe, pipestat_mask);
+
+	enable_mask = I915_DISPLAY_PORT_INTERRUPT |
+		I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
+		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
+	if (IS_CHERRYVIEW(dev_priv))
+		enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
+
+	WARN_ON(dev_priv->irq_mask != ~0);
+
+	dev_priv->irq_mask = ~enable_mask;
+
+	GEN5_IRQ_INIT(VLV_, dev_priv->irq_mask, enable_mask);
+}
+
 /* drm_dma.h hooks
 */
 static void ironlake_irq_reset(struct drm_device *dev)
@@ -3284,34 +3409,19 @@ static void ironlake_irq_reset(struct drm_device *dev)
 	ibx_irq_reset(dev);
 }
 
-static void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
-{
-	enum pipe pipe;
-
-	i915_hotplug_interrupt_update(dev_priv, 0xFFFFFFFF, 0);
-	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
-
-	for_each_pipe(dev_priv, pipe)
-		I915_WRITE(PIPESTAT(pipe), 0xffff);
-
-	GEN5_IRQ_RESET(VLV_);
-}
-
 static void valleyview_irq_preinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	/* VLV magic */
-	I915_WRITE(VLV_IMR, 0);
-	I915_WRITE(RING_IMR(RENDER_RING_BASE), 0);
-	I915_WRITE(RING_IMR(GEN6_BSD_RING_BASE), 0);
-	I915_WRITE(RING_IMR(BLT_RING_BASE), 0);
+	I915_WRITE(VLV_MASTER_IER, 0);
+	POSTING_READ(VLV_MASTER_IER);
 
 	gen5_gt_irq_reset(dev);
 
-	I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK);
-
-	vlv_display_irq_reset(dev_priv);
+	spin_lock_irq(&dev_priv->irq_lock);
+	if (dev_priv->display_irqs_enabled)
+		vlv_display_irq_reset(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
 static void gen8_gt_irq_reset(struct drm_i915_private *dev_priv)
@@ -3384,9 +3494,10 @@ static void cherryview_irq_preinstall(struct drm_device *dev)
 
 	GEN5_IRQ_RESET(GEN8_PCU_);
 
-	I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
-
-	vlv_display_irq_reset(dev_priv);
+	spin_lock_irq(&dev_priv->irq_lock);
+	if (dev_priv->display_irqs_enabled)
+		vlv_display_irq_reset(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
 static u32 intel_hpd_enabled_irqs(struct drm_device *dev,
@@ -3506,6 +3617,26 @@ static void bxt_hpd_irq_setup(struct drm_device *dev)
 	hotplug = I915_READ(PCH_PORT_HOTPLUG);
 	hotplug |= PORTC_HOTPLUG_ENABLE | PORTB_HOTPLUG_ENABLE |
 		PORTA_HOTPLUG_ENABLE;
+
+	DRM_DEBUG_KMS("Invert bit setting: hp_ctl:%x hp_port:%x\n",
+		      hotplug, enabled_irqs);
+	hotplug &= ~BXT_DDI_HPD_INVERT_MASK;
+
+	/*
+	 * For BXT invert bit has to be set based on AOB design
+	 * for HPD detection logic, update it based on VBT fields.
+	 */
+
+	if ((enabled_irqs & BXT_DE_PORT_HP_DDIA) &&
+	    intel_bios_is_port_hpd_inverted(dev_priv, PORT_A))
+		hotplug |= BXT_DDIA_HPD_INVERT;
+	if ((enabled_irqs & BXT_DE_PORT_HP_DDIB) &&
+	    intel_bios_is_port_hpd_inverted(dev_priv, PORT_B))
+		hotplug |= BXT_DDIB_HPD_INVERT;
+	if ((enabled_irqs & BXT_DE_PORT_HP_DDIC) &&
+	    intel_bios_is_port_hpd_inverted(dev_priv, PORT_C))
+		hotplug |= BXT_DDIC_HPD_INVERT;
+
 	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
 }
 
@@ -3613,74 +3744,6 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
-static void valleyview_display_irqs_install(struct drm_i915_private *dev_priv)
-{
-	u32 pipestat_mask;
-	u32 iir_mask;
-	enum pipe pipe;
-
-	pipestat_mask = PIPESTAT_INT_STATUS_MASK |
-			PIPE_FIFO_UNDERRUN_STATUS;
-
-	for_each_pipe(dev_priv, pipe)
-		I915_WRITE(PIPESTAT(pipe), pipestat_mask);
-	POSTING_READ(PIPESTAT(PIPE_A));
-
-	pipestat_mask = PLANE_FLIP_DONE_INT_STATUS_VLV |
-			PIPE_CRC_DONE_INTERRUPT_STATUS;
-
-	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
-	for_each_pipe(dev_priv, pipe)
-		      i915_enable_pipestat(dev_priv, pipe, pipestat_mask);
-
-	iir_mask = I915_DISPLAY_PORT_INTERRUPT |
-		   I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
-		   I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
-	if (IS_CHERRYVIEW(dev_priv))
-		iir_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
-	dev_priv->irq_mask &= ~iir_mask;
-
-	I915_WRITE(VLV_IIR, iir_mask);
-	I915_WRITE(VLV_IIR, iir_mask);
-	I915_WRITE(VLV_IER, ~dev_priv->irq_mask);
-	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
-	POSTING_READ(VLV_IMR);
-}
-
-static void valleyview_display_irqs_uninstall(struct drm_i915_private *dev_priv)
-{
-	u32 pipestat_mask;
-	u32 iir_mask;
-	enum pipe pipe;
-
-	iir_mask = I915_DISPLAY_PORT_INTERRUPT |
-		   I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
-		   I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
-	if (IS_CHERRYVIEW(dev_priv))
-		iir_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
-
-	dev_priv->irq_mask |= iir_mask;
-	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
-	I915_WRITE(VLV_IER, ~dev_priv->irq_mask);
-	I915_WRITE(VLV_IIR, iir_mask);
-	I915_WRITE(VLV_IIR, iir_mask);
-	POSTING_READ(VLV_IIR);
-
-	pipestat_mask = PLANE_FLIP_DONE_INT_STATUS_VLV |
-			PIPE_CRC_DONE_INTERRUPT_STATUS;
-
-	i915_disable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
-	for_each_pipe(dev_priv, pipe)
-		i915_disable_pipestat(dev_priv, pipe, pipestat_mask);
-
-	pipestat_mask = PIPESTAT_INT_STATUS_MASK |
-			PIPE_FIFO_UNDERRUN_STATUS;
-
-	for_each_pipe(dev_priv, pipe)
-		I915_WRITE(PIPESTAT(pipe), pipestat_mask);
-	POSTING_READ(PIPESTAT(PIPE_A));
-}
-
 void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv)
 {
 	assert_spin_locked(&dev_priv->irq_lock);
@@ -3690,8 +3753,10 @@ void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv)
 
 	dev_priv->display_irqs_enabled = true;
 
-	if (intel_irqs_enabled(dev_priv))
-		valleyview_display_irqs_install(dev_priv);
+	if (intel_irqs_enabled(dev_priv)) {
+		vlv_display_irq_reset(dev_priv);
+		vlv_display_irq_postinstall(dev_priv);
+	}
 }
 
 void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv)
@@ -3704,45 +3769,23 @@ void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv)
 	dev_priv->display_irqs_enabled = false;
 
 	if (intel_irqs_enabled(dev_priv))
-		valleyview_display_irqs_uninstall(dev_priv);
+		vlv_display_irq_reset(dev_priv);
 }
 
-static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	dev_priv->irq_mask = ~0;
-
-	i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
-	POSTING_READ(PORT_HOTPLUG_EN);
-
-	I915_WRITE(VLV_IIR, 0xffffffff);
-	I915_WRITE(VLV_IIR, 0xffffffff);
-	I915_WRITE(VLV_IER, ~dev_priv->irq_mask);
-	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
-	POSTING_READ(VLV_IMR);
-
-	/* Interrupt setup is already guaranteed to be single-threaded, this is
-	 * just to make the assert_spin_locked check happy. */
-	spin_lock_irq(&dev_priv->irq_lock);
-	if (dev_priv->display_irqs_enabled)
-		valleyview_display_irqs_install(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
-}
 
 static int valleyview_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	vlv_display_irq_postinstall(dev_priv);
-
 	gen5_gt_irq_postinstall(dev);
 
-	/* ack & enable invalid PTE error interrupts */
-#if 0 /* FIXME: add support to irq handler for checking these bits */
-	I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK);
-	I915_WRITE(DPINVGTT, DPINVGTT_EN_MASK);
-#endif
+	spin_lock_irq(&dev_priv->irq_lock);
+	if (dev_priv->display_irqs_enabled)
+		vlv_display_irq_postinstall(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
 
 	I915_WRITE(VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE);
+	POSTING_READ(VLV_MASTER_IER);
 
 	return 0;
 }
@@ -3753,7 +3796,6 @@ static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv)
 	uint32_t gt_interrupts[] = {
 		GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
 			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
-			GT_RENDER_L3_PARITY_ERROR_INTERRUPT |
 			GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT |
 			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_BCS_IRQ_SHIFT,
 		GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT |
@@ -3765,6 +3807,9 @@ static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv)
 			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VECS_IRQ_SHIFT
 		};
 
+	if (HAS_L3_DPF(dev_priv))
+		gt_interrupts[0] |= GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
+
 	dev_priv->pm_irq_mask = 0xffffffff;
 	GEN8_IRQ_INIT_NDX(GT, 0, ~gt_interrupts[0], gt_interrupts[0]);
 	GEN8_IRQ_INIT_NDX(GT, 1, ~gt_interrupts[1], gt_interrupts[1]);
@@ -3832,7 +3877,7 @@ static int gen8_irq_postinstall(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev))
 		ibx_irq_postinstall(dev);
 
-	I915_WRITE(GEN8_MASTER_IRQ, DE_MASTER_IRQ_CONTROL);
+	I915_WRITE(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
 	POSTING_READ(GEN8_MASTER_IRQ);
 
 	return 0;
@@ -3842,11 +3887,14 @@ static int cherryview_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	vlv_display_irq_postinstall(dev_priv);
-
 	gen8_gt_irq_postinstall(dev_priv);
 
-	I915_WRITE(GEN8_MASTER_IRQ, MASTER_INTERRUPT_ENABLE);
+	spin_lock_irq(&dev_priv->irq_lock);
+	if (dev_priv->display_irqs_enabled)
+		vlv_display_irq_postinstall(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
+
+	I915_WRITE(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
 	POSTING_READ(GEN8_MASTER_IRQ);
 
 	return 0;
@@ -3862,20 +3910,6 @@ static void gen8_irq_uninstall(struct drm_device *dev)
 	gen8_irq_reset(dev);
 }
 
-static void vlv_display_irq_uninstall(struct drm_i915_private *dev_priv)
-{
-	/* Interrupt setup is already guaranteed to be single-threaded, this is
-	 * just to make the assert_spin_locked check happy. */
-	spin_lock_irq(&dev_priv->irq_lock);
-	if (dev_priv->display_irqs_enabled)
-		valleyview_display_irqs_uninstall(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
-
-	vlv_display_irq_reset(dev_priv);
-
-	dev_priv->irq_mask = ~0;
-}
-
 static void valleyview_irq_uninstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3884,12 +3918,16 @@ static void valleyview_irq_uninstall(struct drm_device *dev)
 		return;
 
 	I915_WRITE(VLV_MASTER_IER, 0);
+	POSTING_READ(VLV_MASTER_IER);
 
 	gen5_gt_irq_reset(dev);
 
 	I915_WRITE(HWSTAM, 0xffffffff);
 
-	vlv_display_irq_uninstall(dev_priv);
+	spin_lock_irq(&dev_priv->irq_lock);
+	if (dev_priv->display_irqs_enabled)
+		vlv_display_irq_reset(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
 static void cherryview_irq_uninstall(struct drm_device *dev)
@@ -3906,7 +3944,10 @@ static void cherryview_irq_uninstall(struct drm_device *dev)
 
 	GEN5_IRQ_RESET(GEN8_PCU_);
 
-	vlv_display_irq_uninstall(dev_priv);
+	spin_lock_irq(&dev_priv->irq_lock);
+	if (dev_priv->display_irqs_enabled)
+		vlv_display_irq_reset(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
 static void ironlake_irq_uninstall(struct drm_device *dev)
@@ -4044,7 +4085,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 		new_iir = I915_READ16(IIR); /* Flush posted writes */
 
 		if (iir & I915_USER_INTERRUPT)
-			notify_ring(&dev_priv->ring[RCS]);
+			notify_ring(&dev_priv->engine[RCS]);
 
 		for_each_pipe(dev_priv, pipe) {
 			int plane = pipe;
@@ -4233,14 +4274,17 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 
 		/* Consume port.  Then clear IIR or we'll miss events */
 		if (I915_HAS_HOTPLUG(dev) &&
-		    iir & I915_DISPLAY_PORT_INTERRUPT)
-			i9xx_hpd_irq_handler(dev);
+		    iir & I915_DISPLAY_PORT_INTERRUPT) {
+			u32 hotplug_status = i9xx_hpd_irq_ack(dev_priv);
+			if (hotplug_status)
+				i9xx_hpd_irq_handler(dev, hotplug_status);
+		}
 
 		I915_WRITE(IIR, iir & ~flip_mask);
 		new_iir = I915_READ(IIR); /* Flush posted writes */
 
 		if (iir & I915_USER_INTERRUPT)
-			notify_ring(&dev_priv->ring[RCS]);
+			notify_ring(&dev_priv->engine[RCS]);
 
 		for_each_pipe(dev_priv, pipe) {
 			int plane = pipe;
@@ -4463,16 +4507,19 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		ret = IRQ_HANDLED;
 
 		/* Consume port.  Then clear IIR or we'll miss events */
-		if (iir & I915_DISPLAY_PORT_INTERRUPT)
-			i9xx_hpd_irq_handler(dev);
+		if (iir & I915_DISPLAY_PORT_INTERRUPT) {
+			u32 hotplug_status = i9xx_hpd_irq_ack(dev_priv);
+			if (hotplug_status)
+				i9xx_hpd_irq_handler(dev, hotplug_status);
+		}
 
 		I915_WRITE(IIR, iir & ~flip_mask);
 		new_iir = I915_READ(IIR); /* Flush posted writes */
 
 		if (iir & I915_USER_INTERRUPT)
-			notify_ring(&dev_priv->ring[RCS]);
+			notify_ring(&dev_priv->engine[RCS]);
 		if (iir & I915_BSD_USER_INTERRUPT)
-			notify_ring(&dev_priv->ring[VCS]);
+			notify_ring(&dev_priv->engine[VCS]);
 
 		for_each_pipe(dev_priv, pipe) {
 			if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS &&
@@ -4567,8 +4614,6 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 	INIT_DELAYED_WORK(&dev_priv->gpu_error.hangcheck_work,
 			  i915_hangcheck_elapsed);
 
-	pm_qos_add_request(&dev_priv->pm_qos, PM_QOS_CPU_DMA_LATENCY, PM_QOS_DEFAULT_VALUE);
-
 	if (IS_GEN2(dev_priv)) {
 		dev->max_vblank_count = 0;
 		dev->driver->get_vblank_counter = i8xx_get_vblank_counter;
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index 278c9c40c2e0..1779f02e6df8 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -56,6 +56,8 @@ struct i915_params i915 __read_mostly = {
 	.edp_vswing = 0,
 	.enable_guc_submission = false,
 	.guc_log_level = -1,
+	.enable_dp_mst = true,
+	.inject_load_failure = 0,
 };
 
 module_param_named(modeset, i915.modeset, int, 0400);
@@ -201,3 +203,10 @@ MODULE_PARM_DESC(enable_guc_submission, "Enable GuC submission (default:false)")
 module_param_named(guc_log_level, i915.guc_log_level, int, 0400);
 MODULE_PARM_DESC(guc_log_level,
 	"GuC firmware logging level (-1:disabled (default), 0-3:enabled)");
+
+module_param_named_unsafe(enable_dp_mst, i915.enable_dp_mst, bool, 0600);
+MODULE_PARM_DESC(enable_dp_mst,
+	"Enable multi-stream transport (MST) for new DisplayPort sinks. (default: true)");
+module_param_named_unsafe(inject_load_failure, i915.inject_load_failure, uint, 0400);
+MODULE_PARM_DESC(inject_load_failure,
+	"Force an error after a number of failure check points (0:disabled (default), N:force failure at the Nth failure check point)");
diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h
index bd5026b15d3e..02bc27804291 100644
--- a/drivers/gpu/drm/i915/i915_params.h
+++ b/drivers/gpu/drm/i915/i915_params.h
@@ -49,6 +49,7 @@ struct i915_params {
 	int use_mmio_flip;
 	int mmio_debug;
 	int edp_vswing;
+	unsigned int inject_load_failure;
 	/* leave bools at the end to not create holes */
 	bool enable_hangcheck;
 	bool fastboot;
@@ -59,6 +60,7 @@ struct i915_params {
 	bool enable_guc_submission;
 	bool verbose_state_checks;
 	bool nuclear_pageflip;
+	bool enable_dp_mst;
 };
 
 extern struct i915_params i915 __read_mostly;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 363bd79dea2e..b407411e31ba 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -79,6 +79,16 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 
 /* PCI config space */
 
+#define MCHBAR_I915 0x44
+#define MCHBAR_I965 0x48
+#define MCHBAR_SIZE (4 * 4096)
+
+#define DEVEN 0x54
+#define   DEVEN_MCHBAR_EN (1 << 28)
+
+#define BSM 0x5c
+#define   BSM_MASK (0xFFFF << 20)
+
 #define HPLLCC	0xc0 /* 85x only */
 #define   GC_CLOCK_CONTROL_MASK		(0x7 << 0)
 #define   GC_CLOCK_133_200		(0 << 0)
@@ -90,6 +100,16 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define   GC_CLOCK_166_266		(6 << 0)
 #define   GC_CLOCK_166_250		(7 << 0)
 
+#define I915_GDRST 0xc0 /* PCI config register */
+#define   GRDOM_FULL		(0 << 2)
+#define   GRDOM_RENDER		(1 << 2)
+#define   GRDOM_MEDIA		(3 << 2)
+#define   GRDOM_MASK		(3 << 2)
+#define   GRDOM_RESET_STATUS	(1 << 1)
+#define   GRDOM_RESET_ENABLE	(1 << 0)
+
+#define GCDGMBUS 0xcc
+
 #define GCFGC2	0xda
 #define GCFGC	0xf0 /* 915+ only */
 #define   GC_LOW_FREQUENCY_ENABLE	(1 << 7)
@@ -121,18 +141,16 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define   I915_GC_RENDER_CLOCK_166_MHZ	(0 << 0)
 #define   I915_GC_RENDER_CLOCK_200_MHZ	(1 << 0)
 #define   I915_GC_RENDER_CLOCK_333_MHZ	(4 << 0)
-#define GCDGMBUS 0xcc
-#define PCI_LBPC 0xf4 /* legacy/combination backlight modes, also called LBB */
 
+#define ASLE	0xe4
+#define ASLS	0xfc
+
+#define SWSCI	0xe8
+#define   SWSCI_SCISEL	(1 << 15)
+#define   SWSCI_GSSCIE	(1 << 0)
+
+#define LBPC 0xf4 /* legacy/combination backlight modes, also called LBB */
 
-/* Graphics reset regs */
-#define I915_GDRST 0xc0 /* PCI config register */
-#define  GRDOM_FULL	(0<<2)
-#define  GRDOM_RENDER	(1<<2)
-#define  GRDOM_MEDIA	(3<<2)
-#define  GRDOM_MASK	(3<<2)
-#define  GRDOM_RESET_STATUS (1<<1)
-#define  GRDOM_RESET_ENABLE (1<<0)
 
 #define ILK_GDSR _MMIO(MCHBAR_MIRROR_BASE + 0x2ca4)
 #define  ILK_GRDOM_FULL		(0<<1)
@@ -164,6 +182,9 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define  GEN6_GRDOM_RENDER		(1 << 1)
 #define  GEN6_GRDOM_MEDIA		(1 << 2)
 #define  GEN6_GRDOM_BLT			(1 << 3)
+#define  GEN6_GRDOM_VECS		(1 << 4)
+#define  GEN9_GRDOM_GUC			(1 << 5)
+#define  GEN8_GRDOM_MEDIA2		(1 << 7)
 
 #define RING_PP_DIR_BASE(ring)		_MMIO((ring)->mmio_base+0x228)
 #define RING_PP_DIR_BASE_READ(ring)	_MMIO((ring)->mmio_base+0x518)
@@ -586,6 +607,10 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define GEN7_GPGPU_DISPATCHDIMY         _MMIO(0x2504)
 #define GEN7_GPGPU_DISPATCHDIMZ         _MMIO(0x2508)
 
+/* There are the 16 64-bit CS General Purpose Registers */
+#define HSW_CS_GPR(n)                   _MMIO(0x2600 + (n) * 8)
+#define HSW_CS_GPR_UDW(n)               _MMIO(0x2600 + (n) * 8 + 4)
+
 #define OACONTROL _MMIO(0x2360)
 
 #define _GEN7_PIPEA_DE_LOAD_SL	0x70068
@@ -621,6 +646,10 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define   IOSF_PORT_GPIO_SC			0x48
 #define   IOSF_PORT_GPIO_SUS			0xa8
 #define   IOSF_PORT_CCU				0xa9
+#define   CHV_IOSF_PORT_GPIO_N			0x13
+#define   CHV_IOSF_PORT_GPIO_SE			0x48
+#define   CHV_IOSF_PORT_GPIO_E			0xa8
+#define   CHV_IOSF_PORT_GPIO_SW			0xb2
 #define VLV_IOSF_DATA				_MMIO(VLV_DISPLAY_BASE + 0x2104)
 #define VLV_IOSF_ADDR				_MMIO(VLV_DISPLAY_BASE + 0x2108)
 
@@ -785,7 +814,9 @@ enum skl_disp_power_wells {
 #define  DSI_PLL_M1_DIV_SHIFT			0
 #define  DSI_PLL_M1_DIV_MASK			(0x1ff << 0)
 #define CCK_CZ_CLOCK_CONTROL			0x62
+#define CCK_GPLL_CLOCK_CONTROL			0x67
 #define CCK_DISPLAY_CLOCK_CONTROL		0x6b
+#define CCK_DISPLAY_REF_CLOCK_CONTROL		0x6c
 #define  CCK_TRUNK_FORCE_ON			(1 << 17)
 #define  CCK_TRUNK_FORCE_OFF			(1 << 16)
 #define  CCK_FREQUENCY_STATUS			(0x1f << 8)
@@ -1317,6 +1348,7 @@ enum skl_disp_power_wells {
 #define _PORT_CL1CM_DW0_A		0x162000
 #define _PORT_CL1CM_DW0_BC		0x6C000
 #define   PHY_POWER_GOOD		(1 << 16)
+#define   PHY_RESERVED			(1 << 7)
 #define BXT_PORT_CL1CM_DW0(phy)		_BXT_PHY((phy), _PORT_CL1CM_DW0_BC, \
 							_PORT_CL1CM_DW0_A)
 
@@ -1361,14 +1393,10 @@ enum skl_disp_power_wells {
 
 #define _PORT_REF_DW6_A			0x162198
 #define _PORT_REF_DW6_BC		0x6C198
-/*
- * FIXME: BSpec/CHV ConfigDB disagrees on the following two fields, fix them
- * after testing.
- */
-#define   GRC_CODE_SHIFT		23
-#define   GRC_CODE_MASK			(0x1FF << GRC_CODE_SHIFT)
+#define   GRC_CODE_SHIFT		24
+#define   GRC_CODE_MASK			(0xFF << GRC_CODE_SHIFT)
 #define   GRC_CODE_FAST_SHIFT		16
-#define   GRC_CODE_FAST_MASK		(0x7F << GRC_CODE_FAST_SHIFT)
+#define   GRC_CODE_FAST_MASK		(0xFF << GRC_CODE_FAST_SHIFT)
 #define   GRC_CODE_SLOW_SHIFT		8
 #define   GRC_CODE_SLOW_MASK		(0xFF << GRC_CODE_SLOW_SHIFT)
 #define   GRC_CODE_NOM_MASK		0xFF
@@ -1776,6 +1804,18 @@ enum skl_disp_power_wells {
 #define   GEN9_IZ_HASHING_MASK(slice)			(0x3 << ((slice) * 2))
 #define   GEN9_IZ_HASHING(slice, val)			((val) << ((slice) * 2))
 
+/* WaClearTdlStateAckDirtyBits */
+#define GEN8_STATE_ACK		_MMIO(0x20F0)
+#define GEN9_STATE_ACK_SLICE1	_MMIO(0x20F8)
+#define GEN9_STATE_ACK_SLICE2	_MMIO(0x2100)
+#define   GEN9_STATE_ACK_TDL0 (1 << 12)
+#define   GEN9_STATE_ACK_TDL1 (1 << 13)
+#define   GEN9_STATE_ACK_TDL2 (1 << 14)
+#define   GEN9_STATE_ACK_TDL3 (1 << 15)
+#define   GEN9_SUBSLICE_TDL_ACK_BITS \
+	(GEN9_STATE_ACK_TDL3 | GEN9_STATE_ACK_TDL2 | \
+	 GEN9_STATE_ACK_TDL1 | GEN9_STATE_ACK_TDL0)
+
 #define GFX_MODE	_MMIO(0x2520)
 #define GFX_MODE_GEN7	_MMIO(0x229c)
 #define RING_MODE_GEN7(ring)	_MMIO((ring)->mmio_base+0x29c)
@@ -1795,6 +1835,7 @@ enum skl_disp_power_wells {
 
 #define VLV_DISPLAY_BASE 0x180000
 #define VLV_MIPI_BASE VLV_DISPLAY_BASE
+#define BXT_MIPI_BASE 0x60000
 
 #define VLV_GU_CTL0	_MMIO(VLV_DISPLAY_BASE + 0x2030)
 #define VLV_GU_CTL1	_MMIO(VLV_DISPLAY_BASE + 0x2034)
@@ -2923,6 +2964,15 @@ enum skl_disp_power_wells {
 				INTERVAL_1_33_US(us)) : \
 				INTERVAL_1_28_US(us))
 
+#define INTERVAL_1_28_TO_US(interval)  (((interval) << 7) / 100)
+#define INTERVAL_1_33_TO_US(interval)  (((interval) << 2) / 3)
+#define INTERVAL_0_833_TO_US(interval) (((interval) * 5)  / 6)
+#define GT_PM_INTERVAL_TO_US(dev_priv, interval) (IS_GEN9(dev_priv) ? \
+                           (IS_BROXTON(dev_priv) ? \
+                           INTERVAL_0_833_TO_US(interval) : \
+                           INTERVAL_1_33_TO_US(interval)) : \
+                           INTERVAL_1_28_TO_US(interval))
+
 /*
  * Logical Context regs
  */
@@ -4784,6 +4834,10 @@ enum skl_disp_power_wells {
 #define  CBR_PND_DEADLINE_DISABLE	(1<<31)
 #define  CBR_PWM_CLOCK_MUX_SELECT	(1<<30)
 
+#define CBR4_VLV			_MMIO(VLV_DISPLAY_BASE + 0x70450)
+#define  CBR_DPLLBMD_PIPE_C		(1<<29)
+#define  CBR_DPLLBMD_PIPE_B		(1<<18)
+
 /* FIFO watermark sizes etc */
 #define G4X_FIFO_LINE_SIZE	64
 #define I915_FIFO_LINE_SIZE	64
@@ -6184,6 +6238,7 @@ enum skl_disp_power_wells {
 /* digital port hotplug */
 #define PCH_PORT_HOTPLUG		_MMIO(0xc4030)	/* SHOTPLUG_CTL */
 #define  PORTA_HOTPLUG_ENABLE		(1 << 28) /* LPT:LP+ & BXT */
+#define  BXT_DDIA_HPD_INVERT            (1 << 27)
 #define  PORTA_HOTPLUG_STATUS_MASK	(3 << 24) /* SPT+ & BXT */
 #define  PORTA_HOTPLUG_NO_DETECT	(0 << 24) /* SPT+ & BXT */
 #define  PORTA_HOTPLUG_SHORT_DETECT	(1 << 24) /* SPT+ & BXT */
@@ -6199,6 +6254,7 @@ enum skl_disp_power_wells {
 #define  PORTD_HOTPLUG_SHORT_DETECT	(1 << 16)
 #define  PORTD_HOTPLUG_LONG_DETECT	(2 << 16)
 #define  PORTC_HOTPLUG_ENABLE		(1 << 12)
+#define  BXT_DDIC_HPD_INVERT            (1 << 11)
 #define  PORTC_PULSE_DURATION_2ms	(0 << 10) /* pre-LPT */
 #define  PORTC_PULSE_DURATION_4_5ms	(1 << 10) /* pre-LPT */
 #define  PORTC_PULSE_DURATION_6ms	(2 << 10) /* pre-LPT */
@@ -6209,6 +6265,7 @@ enum skl_disp_power_wells {
 #define  PORTC_HOTPLUG_SHORT_DETECT	(1 << 8)
 #define  PORTC_HOTPLUG_LONG_DETECT	(2 << 8)
 #define  PORTB_HOTPLUG_ENABLE		(1 << 4)
+#define  BXT_DDIB_HPD_INVERT            (1 << 3)
 #define  PORTB_PULSE_DURATION_2ms	(0 << 2) /* pre-LPT */
 #define  PORTB_PULSE_DURATION_4_5ms	(1 << 2) /* pre-LPT */
 #define  PORTB_PULSE_DURATION_6ms	(2 << 2) /* pre-LPT */
@@ -6218,6 +6275,9 @@ enum skl_disp_power_wells {
 #define  PORTB_HOTPLUG_NO_DETECT	(0 << 0)
 #define  PORTB_HOTPLUG_SHORT_DETECT	(1 << 0)
 #define  PORTB_HOTPLUG_LONG_DETECT	(2 << 0)
+#define  BXT_DDI_HPD_INVERT_MASK	(BXT_DDIA_HPD_INVERT | \
+					BXT_DDIB_HPD_INVERT | \
+					BXT_DDIC_HPD_INVERT)
 
 #define PCH_PORT_HOTPLUG2		_MMIO(0xc403C)	/* SHOTPLUG_CTL2 SPT+ */
 #define  PORTE_HOTPLUG_ENABLE		(1 << 4)
@@ -6836,6 +6896,8 @@ enum skl_disp_power_wells {
 #define  VLV_SPAREG2H				_MMIO(0xA194)
 
 #define  GTFIFODBG				_MMIO(0x120000)
+#define    GT_FIFO_SBDEDICATE_FREE_ENTRY_CHV	(0x1f << 20)
+#define    GT_FIFO_FREE_ENTRIES_CHV		(0x7f << 13)
 #define    GT_FIFO_SBDROPERR			(1<<6)
 #define    GT_FIFO_BLOBDROPERR			(1<<5)
 #define    GT_FIFO_SB_READ_ABORTERR		(1<<4)
@@ -6852,8 +6914,11 @@ enum skl_disp_power_wells {
 
 #define  HSW_IDICR				_MMIO(0x9008)
 #define    IDIHASHMSK(x)			(((x) & 0x3f) << 16)
-#define  HSW_EDRAM_PRESENT			_MMIO(0x120010)
+#define  HSW_EDRAM_CAP				_MMIO(0x120010)
 #define    EDRAM_ENABLED			0x1
+#define    EDRAM_NUM_BANKS(cap)			(((cap) >> 1) & 0xf)
+#define    EDRAM_WAYS_IDX(cap)			(((cap) >> 5) & 0x7)
+#define    EDRAM_SETS_IDX(cap)			(((cap) >> 8) & 0x3)
 
 #define GEN6_UCGCTL1				_MMIO(0x9400)
 # define GEN6_EU_TCUNIT_CLOCK_GATE_DISABLE		(1 << 16)
@@ -7109,6 +7174,7 @@ enum skl_disp_power_wells {
 #define   GEN9_CCS_TLB_PREFETCH_ENABLE	(1<<3)
 
 #define GEN8_ROW_CHICKEN		_MMIO(0xe4f0)
+#define   FLOW_CONTROL_ENABLE		(1<<15)
 #define   PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE	(1<<8)
 #define   STALL_DOP_GATING_DISABLE		(1<<5)
 
@@ -7130,6 +7196,7 @@ enum skl_disp_power_wells {
 
 #define GEN9_HALF_SLICE_CHICKEN7	_MMIO(0xe194)
 #define   GEN9_ENABLE_YV12_BUGFIX	(1<<4)
+#define   GEN9_ENABLE_GPGPU_PREEMPTION	(1<<2)
 
 /* Audio */
 #define G4X_AUD_VID_DID			_MMIO(dev_priv->info.display_mmio_offset + 0x62020)
@@ -7369,9 +7436,11 @@ enum skl_disp_power_wells {
 /* SBI offsets */
 #define  SBI_SSCDIVINTPHASE			0x0200
 #define  SBI_SSCDIVINTPHASE6			0x0600
-#define   SBI_SSCDIVINTPHASE_DIVSEL_MASK	((0x7f)<<1)
+#define   SBI_SSCDIVINTPHASE_DIVSEL_SHIFT	1
+#define   SBI_SSCDIVINTPHASE_DIVSEL_MASK	(0x7f<<1)
 #define   SBI_SSCDIVINTPHASE_DIVSEL(x)		((x)<<1)
-#define   SBI_SSCDIVINTPHASE_INCVAL_MASK	((0x7f)<<8)
+#define   SBI_SSCDIVINTPHASE_INCVAL_SHIFT	8
+#define   SBI_SSCDIVINTPHASE_INCVAL_MASK	(0x7f<<8)
 #define   SBI_SSCDIVINTPHASE_INCVAL(x)		((x)<<8)
 #define   SBI_SSCDIVINTPHASE_DIR(x)		((x)<<15)
 #define   SBI_SSCDIVINTPHASE_PROPAGATE		(1<<0)
@@ -7381,6 +7450,8 @@ enum skl_disp_power_wells {
 #define   SBI_SSCCTL_PATHALT			(1<<3)
 #define   SBI_SSCCTL_DISABLE			(1<<0)
 #define  SBI_SSCAUXDIV6				0x0610
+#define   SBI_SSCAUXDIV_FINALDIV2SEL_SHIFT	4
+#define   SBI_SSCAUXDIV_FINALDIV2SEL_MASK	(1<<4)
 #define   SBI_SSCAUXDIV_FINALDIV2SEL(x)		((x)<<4)
 #define  SBI_DBUFF0				0x2a00
 #define  SBI_GEN0				0x1f00
@@ -7660,6 +7731,59 @@ enum skl_disp_power_wells {
 #define PIPE_CSC_POSTOFF_ME(pipe)	_MMIO_PIPE(pipe, _PIPE_A_CSC_POSTOFF_ME, _PIPE_B_CSC_POSTOFF_ME)
 #define PIPE_CSC_POSTOFF_LO(pipe)	_MMIO_PIPE(pipe, _PIPE_A_CSC_POSTOFF_LO, _PIPE_B_CSC_POSTOFF_LO)
 
+/* pipe degamma/gamma LUTs on IVB+ */
+#define _PAL_PREC_INDEX_A	0x4A400
+#define _PAL_PREC_INDEX_B	0x4AC00
+#define _PAL_PREC_INDEX_C	0x4B400
+#define   PAL_PREC_10_12_BIT		(0 << 31)
+#define   PAL_PREC_SPLIT_MODE		(1 << 31)
+#define   PAL_PREC_AUTO_INCREMENT	(1 << 15)
+#define _PAL_PREC_DATA_A	0x4A404
+#define _PAL_PREC_DATA_B	0x4AC04
+#define _PAL_PREC_DATA_C	0x4B404
+#define _PAL_PREC_GC_MAX_A	0x4A410
+#define _PAL_PREC_GC_MAX_B	0x4AC10
+#define _PAL_PREC_GC_MAX_C	0x4B410
+#define _PAL_PREC_EXT_GC_MAX_A	0x4A420
+#define _PAL_PREC_EXT_GC_MAX_B	0x4AC20
+#define _PAL_PREC_EXT_GC_MAX_C	0x4B420
+
+#define PREC_PAL_INDEX(pipe)		_MMIO_PIPE(pipe, _PAL_PREC_INDEX_A, _PAL_PREC_INDEX_B)
+#define PREC_PAL_DATA(pipe)		_MMIO_PIPE(pipe, _PAL_PREC_DATA_A, _PAL_PREC_DATA_B)
+#define PREC_PAL_GC_MAX(pipe, i)	_MMIO(_PIPE(pipe, _PAL_PREC_GC_MAX_A, _PAL_PREC_GC_MAX_B) + (i) * 4)
+#define PREC_PAL_EXT_GC_MAX(pipe, i)	_MMIO(_PIPE(pipe, _PAL_PREC_EXT_GC_MAX_A, _PAL_PREC_EXT_GC_MAX_B) + (i) * 4)
+
+/* pipe CSC & degamma/gamma LUTs on CHV */
+#define _CGM_PIPE_A_CSC_COEFF01	(VLV_DISPLAY_BASE + 0x67900)
+#define _CGM_PIPE_A_CSC_COEFF23	(VLV_DISPLAY_BASE + 0x67904)
+#define _CGM_PIPE_A_CSC_COEFF45	(VLV_DISPLAY_BASE + 0x67908)
+#define _CGM_PIPE_A_CSC_COEFF67	(VLV_DISPLAY_BASE + 0x6790C)
+#define _CGM_PIPE_A_CSC_COEFF8	(VLV_DISPLAY_BASE + 0x67910)
+#define _CGM_PIPE_A_DEGAMMA	(VLV_DISPLAY_BASE + 0x66000)
+#define _CGM_PIPE_A_GAMMA	(VLV_DISPLAY_BASE + 0x67000)
+#define _CGM_PIPE_A_MODE	(VLV_DISPLAY_BASE + 0x67A00)
+#define   CGM_PIPE_MODE_GAMMA	(1 << 2)
+#define   CGM_PIPE_MODE_CSC	(1 << 1)
+#define   CGM_PIPE_MODE_DEGAMMA	(1 << 0)
+
+#define _CGM_PIPE_B_CSC_COEFF01	(VLV_DISPLAY_BASE + 0x69900)
+#define _CGM_PIPE_B_CSC_COEFF23	(VLV_DISPLAY_BASE + 0x69904)
+#define _CGM_PIPE_B_CSC_COEFF45	(VLV_DISPLAY_BASE + 0x69908)
+#define _CGM_PIPE_B_CSC_COEFF67	(VLV_DISPLAY_BASE + 0x6990C)
+#define _CGM_PIPE_B_CSC_COEFF8	(VLV_DISPLAY_BASE + 0x69910)
+#define _CGM_PIPE_B_DEGAMMA	(VLV_DISPLAY_BASE + 0x68000)
+#define _CGM_PIPE_B_GAMMA	(VLV_DISPLAY_BASE + 0x69000)
+#define _CGM_PIPE_B_MODE	(VLV_DISPLAY_BASE + 0x69A00)
+
+#define CGM_PIPE_CSC_COEFF01(pipe)	_MMIO_PIPE(pipe, _CGM_PIPE_A_CSC_COEFF01, _CGM_PIPE_B_CSC_COEFF01)
+#define CGM_PIPE_CSC_COEFF23(pipe)	_MMIO_PIPE(pipe, _CGM_PIPE_A_CSC_COEFF23, _CGM_PIPE_B_CSC_COEFF23)
+#define CGM_PIPE_CSC_COEFF45(pipe)	_MMIO_PIPE(pipe, _CGM_PIPE_A_CSC_COEFF45, _CGM_PIPE_B_CSC_COEFF45)
+#define CGM_PIPE_CSC_COEFF67(pipe)	_MMIO_PIPE(pipe, _CGM_PIPE_A_CSC_COEFF67, _CGM_PIPE_B_CSC_COEFF67)
+#define CGM_PIPE_CSC_COEFF8(pipe)	_MMIO_PIPE(pipe, _CGM_PIPE_A_CSC_COEFF8, _CGM_PIPE_B_CSC_COEFF8)
+#define CGM_PIPE_DEGAMMA(pipe, i, w)	_MMIO(_PIPE(pipe, _CGM_PIPE_A_DEGAMMA, _CGM_PIPE_B_DEGAMMA) + (i) * 8 + (w) * 4)
+#define CGM_PIPE_GAMMA(pipe, i, w)	_MMIO(_PIPE(pipe, _CGM_PIPE_A_GAMMA, _CGM_PIPE_B_GAMMA) + (i) * 8 + (w) * 4)
+#define CGM_PIPE_MODE(pipe)		_MMIO_PIPE(pipe, _CGM_PIPE_A_MODE, _CGM_PIPE_B_MODE)
+
 /* MIPI DSI registers */
 
 #define _MIPI_PORT(port, a, c)	_PORT3(port, a, 0, c)	/* ports A and C only */
@@ -7674,58 +7798,62 @@ enum skl_disp_power_wells {
 #define  BXT_MIPI_DIV_SHIFT(port)		\
 			_MIPI_PORT(port, BXT_MIPI1_DIV_SHIFT, \
 					BXT_MIPI2_DIV_SHIFT)
-/* Var clock divider to generate TX source. Result must be < 39.5 M */
-#define  BXT_MIPI1_ESCLK_VAR_DIV_MASK		(0x3F << 26)
-#define  BXT_MIPI2_ESCLK_VAR_DIV_MASK		(0x3F << 10)
-#define  BXT_MIPI_ESCLK_VAR_DIV_MASK(port)	\
-			_MIPI_PORT(port, BXT_MIPI1_ESCLK_VAR_DIV_MASK, \
-						BXT_MIPI2_ESCLK_VAR_DIV_MASK)
-
-#define  BXT_MIPI_ESCLK_VAR_DIV(port, val)	\
-			(val << BXT_MIPI_DIV_SHIFT(port))
+
 /* TX control divider to select actual TX clock output from (8x/var) */
-#define  BXT_MIPI1_TX_ESCLK_SHIFT		21
-#define  BXT_MIPI2_TX_ESCLK_SHIFT		5
+#define  BXT_MIPI1_TX_ESCLK_SHIFT		26
+#define  BXT_MIPI2_TX_ESCLK_SHIFT		10
 #define  BXT_MIPI_TX_ESCLK_SHIFT(port)		\
 			_MIPI_PORT(port, BXT_MIPI1_TX_ESCLK_SHIFT, \
 					BXT_MIPI2_TX_ESCLK_SHIFT)
-#define  BXT_MIPI1_TX_ESCLK_FIXDIV_MASK		(3 << 21)
-#define  BXT_MIPI2_TX_ESCLK_FIXDIV_MASK		(3 << 5)
+#define  BXT_MIPI1_TX_ESCLK_FIXDIV_MASK		(0x3F << 26)
+#define  BXT_MIPI2_TX_ESCLK_FIXDIV_MASK		(0x3F << 10)
 #define  BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port)	\
 			_MIPI_PORT(port, BXT_MIPI1_TX_ESCLK_FIXDIV_MASK, \
-						BXT_MIPI2_TX_ESCLK_FIXDIV_MASK)
-#define  BXT_MIPI_TX_ESCLK_8XDIV_BY2(port)	\
-		(0x0 << BXT_MIPI_TX_ESCLK_SHIFT(port))
-#define  BXT_MIPI_TX_ESCLK_8XDIV_BY4(port)	\
-		(0x1 << BXT_MIPI_TX_ESCLK_SHIFT(port))
-#define  BXT_MIPI_TX_ESCLK_8XDIV_BY8(port)	\
-		(0x2 << BXT_MIPI_TX_ESCLK_SHIFT(port))
-/* RX control divider to select actual RX clock output from 8x*/
-#define  BXT_MIPI1_RX_ESCLK_SHIFT		19
-#define  BXT_MIPI2_RX_ESCLK_SHIFT		3
-#define  BXT_MIPI_RX_ESCLK_SHIFT(port)		\
-			_MIPI_PORT(port, BXT_MIPI1_RX_ESCLK_SHIFT, \
-					BXT_MIPI2_RX_ESCLK_SHIFT)
-#define  BXT_MIPI1_RX_ESCLK_FIXDIV_MASK		(3 << 19)
-#define  BXT_MIPI2_RX_ESCLK_FIXDIV_MASK		(3 << 3)
-#define  BXT_MIPI_RX_ESCLK_FIXDIV_MASK(port)	\
-		(3 << BXT_MIPI_RX_ESCLK_SHIFT(port))
-#define  BXT_MIPI_RX_ESCLK_8X_BY2(port)	\
-		(1 << BXT_MIPI_RX_ESCLK_SHIFT(port))
-#define  BXT_MIPI_RX_ESCLK_8X_BY3(port)	\
-		(2 << BXT_MIPI_RX_ESCLK_SHIFT(port))
-#define  BXT_MIPI_RX_ESCLK_8X_BY4(port)	\
-		(3 << BXT_MIPI_RX_ESCLK_SHIFT(port))
-/* BXT-A WA: Always prog DPHY dividers to 00 */
-#define  BXT_MIPI1_DPHY_DIV_SHIFT		16
-#define  BXT_MIPI2_DPHY_DIV_SHIFT		0
-#define  BXT_MIPI_DPHY_DIV_SHIFT(port)		\
-			_MIPI_PORT(port, BXT_MIPI1_DPHY_DIV_SHIFT, \
-					BXT_MIPI2_DPHY_DIV_SHIFT)
-#define  BXT_MIPI_1_DPHY_DIVIDER_MASK		(3 << 16)
-#define  BXT_MIPI_2_DPHY_DIVIDER_MASK		(3 << 0)
-#define  BXT_MIPI_DPHY_DIVIDER_MASK(port)	\
-		(3 << BXT_MIPI_DPHY_DIV_SHIFT(port))
+					BXT_MIPI2_TX_ESCLK_FIXDIV_MASK)
+#define  BXT_MIPI_TX_ESCLK_DIVIDER(port, val)	\
+		((val & 0x3F) << BXT_MIPI_TX_ESCLK_SHIFT(port))
+/* RX upper control divider to select actual RX clock output from 8x */
+#define  BXT_MIPI1_RX_ESCLK_UPPER_SHIFT		21
+#define  BXT_MIPI2_RX_ESCLK_UPPER_SHIFT		5
+#define  BXT_MIPI_RX_ESCLK_UPPER_SHIFT(port)		\
+			_MIPI_PORT(port, BXT_MIPI1_RX_ESCLK_UPPER_SHIFT, \
+					BXT_MIPI2_RX_ESCLK_UPPER_SHIFT)
+#define  BXT_MIPI1_RX_ESCLK_UPPER_FIXDIV_MASK		(3 << 21)
+#define  BXT_MIPI2_RX_ESCLK_UPPER_FIXDIV_MASK		(3 << 5)
+#define  BXT_MIPI_RX_ESCLK_UPPER_FIXDIV_MASK(port)	\
+			_MIPI_PORT(port, BXT_MIPI1_RX_ESCLK_UPPER_FIXDIV_MASK, \
+					BXT_MIPI2_RX_ESCLK_UPPER_FIXDIV_MASK)
+#define  BXT_MIPI_RX_ESCLK_UPPER_DIVIDER(port, val)	\
+		((val & 3) << BXT_MIPI_RX_ESCLK_UPPER_SHIFT(port))
+/* 8/3X divider to select the actual 8/3X clock output from 8x */
+#define  BXT_MIPI1_8X_BY3_SHIFT                19
+#define  BXT_MIPI2_8X_BY3_SHIFT                3
+#define  BXT_MIPI_8X_BY3_SHIFT(port)          \
+			_MIPI_PORT(port, BXT_MIPI1_8X_BY3_SHIFT, \
+					BXT_MIPI2_8X_BY3_SHIFT)
+#define  BXT_MIPI1_8X_BY3_DIVIDER_MASK         (3 << 19)
+#define  BXT_MIPI2_8X_BY3_DIVIDER_MASK         (3 << 3)
+#define  BXT_MIPI_8X_BY3_DIVIDER_MASK(port)    \
+			_MIPI_PORT(port, BXT_MIPI1_8X_BY3_DIVIDER_MASK, \
+						BXT_MIPI2_8X_BY3_DIVIDER_MASK)
+#define  BXT_MIPI_8X_BY3_DIVIDER(port, val)    \
+			((val & 3) << BXT_MIPI_8X_BY3_SHIFT(port))
+/* RX lower control divider to select actual RX clock output from 8x */
+#define  BXT_MIPI1_RX_ESCLK_LOWER_SHIFT		16
+#define  BXT_MIPI2_RX_ESCLK_LOWER_SHIFT		0
+#define  BXT_MIPI_RX_ESCLK_LOWER_SHIFT(port)		\
+			_MIPI_PORT(port, BXT_MIPI1_RX_ESCLK_LOWER_SHIFT, \
+					BXT_MIPI2_RX_ESCLK_LOWER_SHIFT)
+#define  BXT_MIPI1_RX_ESCLK_LOWER_FIXDIV_MASK		(3 << 16)
+#define  BXT_MIPI2_RX_ESCLK_LOWER_FIXDIV_MASK		(3 << 0)
+#define  BXT_MIPI_RX_ESCLK_LOWER_FIXDIV_MASK(port)	\
+			_MIPI_PORT(port, BXT_MIPI1_RX_ESCLK_LOWER_FIXDIV_MASK, \
+					BXT_MIPI2_RX_ESCLK_LOWER_FIXDIV_MASK)
+#define  BXT_MIPI_RX_ESCLK_LOWER_DIVIDER(port, val)	\
+		((val & 3) << BXT_MIPI_RX_ESCLK_LOWER_SHIFT(port))
+
+#define RX_DIVIDER_BIT_1_2                     0x3
+#define RX_DIVIDER_BIT_3_4                     0xC
 
 /* BXT MIPI mode configure */
 #define  _BXT_MIPIA_TRANS_HACTIVE			0x6B0F8
@@ -7750,9 +7878,11 @@ enum skl_disp_power_wells {
 #define  BXT_DSIC_16X_BY2		(1 << 10)
 #define  BXT_DSIC_16X_BY3		(2 << 10)
 #define  BXT_DSIC_16X_BY4		(3 << 10)
+#define  BXT_DSIC_16X_MASK		(3 << 10)
 #define  BXT_DSIA_16X_BY2		(1 << 8)
 #define  BXT_DSIA_16X_BY3		(2 << 8)
 #define  BXT_DSIA_16X_BY4		(3 << 8)
+#define  BXT_DSIA_16X_MASK		(3 << 8)
 #define  BXT_DSI_FREQ_SEL_SHIFT		8
 #define  BXT_DSI_FREQ_SEL_MASK		(0xF << BXT_DSI_FREQ_SEL_SHIFT)
 
@@ -7887,8 +8017,8 @@ enum skl_disp_power_wells {
 #define  VID_MODE_FORMAT_MASK				(0xf << 7)
 #define  VID_MODE_NOT_SUPPORTED				(0 << 7)
 #define  VID_MODE_FORMAT_RGB565				(1 << 7)
-#define  VID_MODE_FORMAT_RGB666				(2 << 7)
-#define  VID_MODE_FORMAT_RGB666_LOOSE			(3 << 7)
+#define  VID_MODE_FORMAT_RGB666_PACKED			(2 << 7)
+#define  VID_MODE_FORMAT_RGB666				(3 << 7)
 #define  VID_MODE_FORMAT_RGB888				(4 << 7)
 #define  CMD_MODE_CHANNEL_NUMBER_SHIFT			5
 #define  CMD_MODE_CHANNEL_NUMBER_MASK			(3 << 5)
@@ -8144,6 +8274,7 @@ enum skl_disp_power_wells {
 #define  READ_REQUEST_PRIORITY_HIGH			(3 << 3)
 #define  RGB_FLIP_TO_BGR				(1 << 2)
 
+#define  BXT_PIPE_SELECT_SHIFT				7
 #define  BXT_PIPE_SELECT_MASK				(7 << 7)
 #define  BXT_PIPE_SELECT(pipe)				((pipe) << 7)
 
diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index c6188dddb341..2d576b7ff299 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -370,6 +370,8 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 
 	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
 
+	intel_runtime_pm_get(dev_priv);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 
 	val = intel_freq_opcode(dev_priv, val);
@@ -378,6 +380,7 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 	    val > dev_priv->rps.max_freq ||
 	    val < dev_priv->rps.min_freq_softlimit) {
 		mutex_unlock(&dev_priv->rps.hw_lock);
+		intel_runtime_pm_put(dev_priv);
 		return -EINVAL;
 	}
 
@@ -398,6 +401,8 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
+	intel_runtime_pm_put(dev_priv);
+
 	return count;
 }
 
@@ -433,6 +438,8 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 
 	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
 
+	intel_runtime_pm_get(dev_priv);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 
 	val = intel_freq_opcode(dev_priv, val);
@@ -441,6 +448,7 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 	    val > dev_priv->rps.max_freq ||
 	    val > dev_priv->rps.max_freq_softlimit) {
 		mutex_unlock(&dev_priv->rps.hw_lock);
+		intel_runtime_pm_put(dev_priv);
 		return -EINVAL;
 	}
 
@@ -457,6 +465,8 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
+	intel_runtime_pm_put(dev_priv);
+
 	return count;
 
 }
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index fa09e5581137..dc0def210097 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -464,7 +464,7 @@ TRACE_EVENT(i915_gem_ring_sync_to,
 	    TP_fast_assign(
 			   __entry->dev = from->dev->primary->index;
 			   __entry->sync_from = from->id;
-			   __entry->sync_to = to_req->ring->id;
+			   __entry->sync_to = to_req->engine->id;
 			   __entry->seqno = i915_gem_request_get_seqno(req);
 			   ),
 
@@ -486,13 +486,13 @@ TRACE_EVENT(i915_gem_ring_dispatch,
 			     ),
 
 	    TP_fast_assign(
-			   struct intel_engine_cs *ring =
-						i915_gem_request_get_ring(req);
-			   __entry->dev = ring->dev->primary->index;
-			   __entry->ring = ring->id;
+			   struct intel_engine_cs *engine =
+						i915_gem_request_get_engine(req);
+			   __entry->dev = engine->dev->primary->index;
+			   __entry->ring = engine->id;
 			   __entry->seqno = i915_gem_request_get_seqno(req);
 			   __entry->flags = flags;
-			   i915_trace_irq_get(ring, req);
+			   i915_trace_irq_get(engine, req);
 			   ),
 
 	    TP_printk("dev=%u, ring=%u, seqno=%u, flags=%x",
@@ -511,8 +511,8 @@ TRACE_EVENT(i915_gem_ring_flush,
 			     ),
 
 	    TP_fast_assign(
-			   __entry->dev = req->ring->dev->primary->index;
-			   __entry->ring = req->ring->id;
+			   __entry->dev = req->engine->dev->primary->index;
+			   __entry->ring = req->engine->id;
 			   __entry->invalidate = invalidate;
 			   __entry->flush = flush;
 			   ),
@@ -533,10 +533,10 @@ DECLARE_EVENT_CLASS(i915_gem_request,
 			     ),
 
 	    TP_fast_assign(
-			   struct intel_engine_cs *ring =
-						i915_gem_request_get_ring(req);
-			   __entry->dev = ring->dev->primary->index;
-			   __entry->ring = ring->id;
+			   struct intel_engine_cs *engine =
+						i915_gem_request_get_engine(req);
+			   __entry->dev = engine->dev->primary->index;
+			   __entry->ring = engine->id;
 			   __entry->seqno = i915_gem_request_get_seqno(req);
 			   ),
 
@@ -550,8 +550,8 @@ DEFINE_EVENT(i915_gem_request, i915_gem_request_add,
 );
 
 TRACE_EVENT(i915_gem_request_notify,
-	    TP_PROTO(struct intel_engine_cs *ring),
-	    TP_ARGS(ring),
+	    TP_PROTO(struct intel_engine_cs *engine),
+	    TP_ARGS(engine),
 
 	    TP_STRUCT__entry(
 			     __field(u32, dev)
@@ -560,9 +560,9 @@ TRACE_EVENT(i915_gem_request_notify,
 			     ),
 
 	    TP_fast_assign(
-			   __entry->dev = ring->dev->primary->index;
-			   __entry->ring = ring->id;
-			   __entry->seqno = ring->get_seqno(ring, false);
+			   __entry->dev = engine->dev->primary->index;
+			   __entry->ring = engine->id;
+			   __entry->seqno = engine->get_seqno(engine);
 			   ),
 
 	    TP_printk("dev=%u, ring=%u, seqno=%u",
@@ -597,13 +597,13 @@ TRACE_EVENT(i915_gem_request_wait_begin,
 	     * less desirable.
 	     */
 	    TP_fast_assign(
-			   struct intel_engine_cs *ring =
-						i915_gem_request_get_ring(req);
-			   __entry->dev = ring->dev->primary->index;
-			   __entry->ring = ring->id;
+			   struct intel_engine_cs *engine =
+						i915_gem_request_get_engine(req);
+			   __entry->dev = engine->dev->primary->index;
+			   __entry->ring = engine->id;
 			   __entry->seqno = i915_gem_request_get_seqno(req);
 			   __entry->blocking =
-				     mutex_is_locked(&ring->dev->struct_mutex);
+				     mutex_is_locked(&engine->dev->struct_mutex);
 			   ),
 
 	    TP_printk("dev=%u, ring=%u, seqno=%u, blocking=%s",
@@ -777,9 +777,9 @@ DEFINE_EVENT(i915_context, i915_context_free,
  * called only if full ppgtt is enabled.
  */
 TRACE_EVENT(switch_mm,
-	TP_PROTO(struct intel_engine_cs *ring, struct intel_context *to),
+	TP_PROTO(struct intel_engine_cs *engine, struct intel_context *to),
 
-	TP_ARGS(ring, to),
+	TP_ARGS(engine, to),
 
 	TP_STRUCT__entry(
 			__field(u32, ring)
@@ -789,10 +789,10 @@ TRACE_EVENT(switch_mm,
 	),
 
 	TP_fast_assign(
-			__entry->ring = ring->id;
+			__entry->ring = engine->id;
 			__entry->to = to;
 			__entry->vm = to->ppgtt? &to->ppgtt->base : NULL;
-			__entry->dev = ring->dev->primary->index;
+			__entry->dev = engine->dev->primary->index;
 	),
 
 	TP_printk("dev=%u, ring=%u, ctx=%p, ctx_vm=%p",
diff --git a/drivers/gpu/drm/i915/i915_vgpu.c b/drivers/gpu/drm/i915/i915_vgpu.c
index dea7429be4d0..d02efb8cad4d 100644
--- a/drivers/gpu/drm/i915/i915_vgpu.c
+++ b/drivers/gpu/drm/i915/i915_vgpu.c
@@ -181,8 +181,8 @@ static int vgt_balloon_space(struct drm_mm *mm,
 int intel_vgt_balloon(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct i915_address_space *ggtt_vm = &dev_priv->gtt.base;
-	unsigned long ggtt_vm_end = ggtt_vm->start + ggtt_vm->total;
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+	unsigned long ggtt_end = ggtt->base.start + ggtt->base.total;
 
 	unsigned long mappable_base, mappable_size, mappable_end;
 	unsigned long unmappable_base, unmappable_size, unmappable_end;
@@ -202,19 +202,19 @@ int intel_vgt_balloon(struct drm_device *dev)
 	DRM_INFO("Unmappable graphic memory: base 0x%lx size %ldKiB\n",
 		 unmappable_base, unmappable_size / 1024);
 
-	if (mappable_base < ggtt_vm->start ||
-	    mappable_end > dev_priv->gtt.mappable_end ||
-	    unmappable_base < dev_priv->gtt.mappable_end ||
-	    unmappable_end > ggtt_vm_end) {
+	if (mappable_base < ggtt->base.start ||
+	    mappable_end > ggtt->mappable_end ||
+	    unmappable_base < ggtt->mappable_end ||
+	    unmappable_end > ggtt_end) {
 		DRM_ERROR("Invalid ballooning configuration!\n");
 		return -EINVAL;
 	}
 
 	/* Unmappable graphic memory ballooning */
-	if (unmappable_base > dev_priv->gtt.mappable_end) {
-		ret = vgt_balloon_space(&ggtt_vm->mm,
+	if (unmappable_base > ggtt->mappable_end) {
+		ret = vgt_balloon_space(&ggtt->base.mm,
 					&bl_info.space[2],
-					dev_priv->gtt.mappable_end,
+					ggtt->mappable_end,
 					unmappable_base);
 
 		if (ret)
@@ -225,30 +225,30 @@ int intel_vgt_balloon(struct drm_device *dev)
 	 * No need to partition out the last physical page,
 	 * because it is reserved to the guard page.
 	 */
-	if (unmappable_end < ggtt_vm_end - PAGE_SIZE) {
-		ret = vgt_balloon_space(&ggtt_vm->mm,
+	if (unmappable_end < ggtt_end - PAGE_SIZE) {
+		ret = vgt_balloon_space(&ggtt->base.mm,
 					&bl_info.space[3],
 					unmappable_end,
-					ggtt_vm_end - PAGE_SIZE);
+					ggtt_end - PAGE_SIZE);
 		if (ret)
 			goto err;
 	}
 
 	/* Mappable graphic memory ballooning */
-	if (mappable_base > ggtt_vm->start) {
-		ret = vgt_balloon_space(&ggtt_vm->mm,
+	if (mappable_base > ggtt->base.start) {
+		ret = vgt_balloon_space(&ggtt->base.mm,
 					&bl_info.space[0],
-					ggtt_vm->start, mappable_base);
+					ggtt->base.start, mappable_base);
 
 		if (ret)
 			goto err;
 	}
 
-	if (mappable_end < dev_priv->gtt.mappable_end) {
-		ret = vgt_balloon_space(&ggtt_vm->mm,
+	if (mappable_end < ggtt->mappable_end) {
+		ret = vgt_balloon_space(&ggtt->base.mm,
 					&bl_info.space[1],
 					mappable_end,
-					dev_priv->gtt.mappable_end);
+					ggtt->mappable_end);
 
 		if (ret)
 			goto err;
diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 8e579a8505ac..50ff90aea721 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -96,8 +96,11 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
 	crtc_state->update_pipe = false;
 	crtc_state->disable_lp_wm = false;
 	crtc_state->disable_cxsr = false;
-	crtc_state->wm_changed = false;
+	crtc_state->update_wm_pre = false;
+	crtc_state->update_wm_post = false;
 	crtc_state->fb_changed = false;
+	crtc_state->wm.need_postvbl_update = false;
+	crtc_state->fb_bits = 0;
 
 	return &crtc_state->base;
 }
diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c
index e0b851a0004a..7de7721f65bc 100644
--- a/drivers/gpu/drm/i915/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/intel_atomic_plane.c
@@ -195,12 +195,10 @@ static void intel_plane_atomic_update(struct drm_plane *plane,
 	struct intel_plane_state *intel_state =
 		to_intel_plane_state(plane->state);
 	struct drm_crtc *crtc = plane->state->crtc ?: old_state->crtc;
-	struct drm_crtc_state *crtc_state =
-		drm_atomic_get_existing_crtc_state(old_state->state, crtc);
 
 	if (intel_state->visible)
 		intel_plane->update_plane(plane,
-					  to_intel_crtc_state(crtc_state),
+					  to_intel_crtc_state(crtc->state),
 					  intel_state);
 	else
 		intel_plane->disable_plane(plane, crtc);
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 7d281b40064a..02a7527ce7bb 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -372,7 +372,7 @@ static void ilk_audio_codec_disable(struct intel_encoder *encoder)
 	if (WARN_ON(port == PORT_A))
 		return;
 
-	if (HAS_PCH_IBX(dev_priv->dev)) {
+	if (HAS_PCH_IBX(dev_priv)) {
 		aud_config = IBX_AUD_CFG(pipe);
 		aud_cntrl_st2 = IBX_AUD_CNTL_ST2;
 	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
@@ -561,23 +561,21 @@ void intel_audio_codec_disable(struct intel_encoder *intel_encoder)
 }
 
 /**
- * intel_init_audio - Set up chip specific audio functions
- * @dev: drm device
+ * intel_init_audio_hooks - Set up chip specific audio hooks
+ * @dev_priv: device private
  */
-void intel_init_audio(struct drm_device *dev)
+void intel_init_audio_hooks(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (IS_G4X(dev)) {
+	if (IS_G4X(dev_priv)) {
 		dev_priv->display.audio_codec_enable = g4x_audio_codec_enable;
 		dev_priv->display.audio_codec_disable = g4x_audio_codec_disable;
-	} else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
+	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
 		dev_priv->display.audio_codec_enable = ilk_audio_codec_enable;
 		dev_priv->display.audio_codec_disable = ilk_audio_codec_disable;
-	} else if (IS_HASWELL(dev) || INTEL_INFO(dev)->gen >= 8) {
+	} else if (IS_HASWELL(dev_priv) || INTEL_INFO(dev_priv)->gen >= 8) {
 		dev_priv->display.audio_codec_enable = hsw_audio_codec_enable;
 		dev_priv->display.audio_codec_disable = hsw_audio_codec_disable;
-	} else if (HAS_PCH_SPLIT(dev)) {
+	} else if (HAS_PCH_SPLIT(dev_priv)) {
 		dev_priv->display.audio_codec_enable = ilk_audio_codec_enable;
 		dev_priv->display.audio_codec_disable = ilk_audio_codec_disable;
 	}
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index bf62a19c8f69..e72dd9a8d6bf 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -29,7 +29,9 @@
 #include <drm/drmP.h>
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
-#include "intel_bios.h"
+
+#define _INTEL_BIOS_PRIVATE
+#include "intel_vbt_defs.h"
 
 /**
  * DOC: Video BIOS Table (VBT)
@@ -56,8 +58,6 @@
 #define	SLAVE_ADDR1	0x70
 #define	SLAVE_ADDR2	0x72
 
-static int panel_type;
-
 /* Get BDB block size given a pointer to Block ID. */
 static u32 _get_blocksize(const u8 *block_base)
 {
@@ -203,17 +203,32 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 	const struct lvds_dvo_timing *panel_dvo_timing;
 	const struct lvds_fp_timing *fp_timing;
 	struct drm_display_mode *panel_fixed_mode;
+	int panel_type;
 	int drrs_mode;
+	int ret;
 
 	lvds_options = find_section(bdb, BDB_LVDS_OPTIONS);
 	if (!lvds_options)
 		return;
 
 	dev_priv->vbt.lvds_dither = lvds_options->pixel_dither;
-	if (lvds_options->panel_type == 0xff)
-		return;
 
-	panel_type = lvds_options->panel_type;
+	ret = intel_opregion_get_panel_type(dev_priv->dev);
+	if (ret >= 0) {
+		WARN_ON(ret > 0xf);
+		panel_type = ret;
+		DRM_DEBUG_KMS("Panel type: %d (OpRegion)\n", panel_type);
+	} else {
+		if (lvds_options->panel_type > 0xf) {
+			DRM_DEBUG_KMS("Invalid VBT panel type 0x%x\n",
+				      lvds_options->panel_type);
+			return;
+		}
+		panel_type = lvds_options->panel_type;
+		DRM_DEBUG_KMS("Panel type: %d (VBT)\n", panel_type);
+	}
+
+	dev_priv->vbt.panel_type = panel_type;
 
 	drrs_mode = (lvds_options->dps_panel_type_bits
 				>> (panel_type * 2)) & MODE_MASK;
@@ -249,7 +264,7 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 
 	panel_dvo_timing = get_lvds_dvo_timing(lvds_lfp_data,
 					       lvds_lfp_data_ptrs,
-					       lvds_options->panel_type);
+					       panel_type);
 
 	panel_fixed_mode = kzalloc(sizeof(*panel_fixed_mode), GFP_KERNEL);
 	if (!panel_fixed_mode)
@@ -264,7 +279,7 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 
 	fp_timing = get_lvds_fp_timing(bdb, lvds_lfp_data,
 				       lvds_lfp_data_ptrs,
-				       lvds_options->panel_type);
+				       panel_type);
 	if (fp_timing) {
 		/* check the resolution, just to be sure */
 		if (fp_timing->x_res == panel_fixed_mode->hdisplay &&
@@ -282,6 +297,7 @@ parse_lfp_backlight(struct drm_i915_private *dev_priv,
 {
 	const struct bdb_lfp_backlight_data *backlight_data;
 	const struct bdb_lfp_backlight_data_entry *entry;
+	int panel_type = dev_priv->vbt.panel_type;
 
 	backlight_data = find_section(bdb, BDB_LVDS_BACKLIGHT);
 	if (!backlight_data)
@@ -480,7 +496,7 @@ parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
 			      child->slave_addr,
 			      (child->dvo_port == DEVICE_PORT_DVOB) ?
 			      "SDVOB" : "SDVOC");
-		p_mapping = &(dev_priv->sdvo_mappings[child->dvo_port - 1]);
+		p_mapping = &dev_priv->vbt.sdvo_mappings[child->dvo_port - 1];
 		if (!p_mapping->initialized) {
 			p_mapping->dvo_port = child->dvo_port;
 			p_mapping->slave_addr = child->slave_addr;
@@ -525,10 +541,7 @@ parse_driver_features(struct drm_i915_private *dev_priv,
 		return;
 
 	if (driver->lvds_config == BDB_DRIVER_FEATURE_EDP)
-		dev_priv->vbt.edp_support = 1;
-
-	if (driver->dual_frequency)
-		dev_priv->render_reclock_avail = true;
+		dev_priv->vbt.edp.support = 1;
 
 	DRM_DEBUG_KMS("DRRS State Enabled:%d\n", driver->drrs_enabled);
 	/*
@@ -547,23 +560,24 @@ parse_edp(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 	const struct bdb_edp *edp;
 	const struct edp_power_seq *edp_pps;
 	const struct edp_link_params *edp_link_params;
+	int panel_type = dev_priv->vbt.panel_type;
 
 	edp = find_section(bdb, BDB_EDP);
 	if (!edp) {
-		if (dev_priv->vbt.edp_support)
+		if (dev_priv->vbt.edp.support)
 			DRM_DEBUG_KMS("No eDP BDB found but eDP panel supported.\n");
 		return;
 	}
 
 	switch ((edp->color_depth >> (panel_type * 2)) & 3) {
 	case EDP_18BPP:
-		dev_priv->vbt.edp_bpp = 18;
+		dev_priv->vbt.edp.bpp = 18;
 		break;
 	case EDP_24BPP:
-		dev_priv->vbt.edp_bpp = 24;
+		dev_priv->vbt.edp.bpp = 24;
 		break;
 	case EDP_30BPP:
-		dev_priv->vbt.edp_bpp = 30;
+		dev_priv->vbt.edp.bpp = 30;
 		break;
 	}
 
@@ -571,14 +585,14 @@ parse_edp(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 	edp_pps = &edp->power_seqs[panel_type];
 	edp_link_params = &edp->link_params[panel_type];
 
-	dev_priv->vbt.edp_pps = *edp_pps;
+	dev_priv->vbt.edp.pps = *edp_pps;
 
 	switch (edp_link_params->rate) {
 	case EDP_RATE_1_62:
-		dev_priv->vbt.edp_rate = DP_LINK_BW_1_62;
+		dev_priv->vbt.edp.rate = DP_LINK_BW_1_62;
 		break;
 	case EDP_RATE_2_7:
-		dev_priv->vbt.edp_rate = DP_LINK_BW_2_7;
+		dev_priv->vbt.edp.rate = DP_LINK_BW_2_7;
 		break;
 	default:
 		DRM_DEBUG_KMS("VBT has unknown eDP link rate value %u\n",
@@ -588,13 +602,13 @@ parse_edp(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 
 	switch (edp_link_params->lanes) {
 	case EDP_LANE_1:
-		dev_priv->vbt.edp_lanes = 1;
+		dev_priv->vbt.edp.lanes = 1;
 		break;
 	case EDP_LANE_2:
-		dev_priv->vbt.edp_lanes = 2;
+		dev_priv->vbt.edp.lanes = 2;
 		break;
 	case EDP_LANE_4:
-		dev_priv->vbt.edp_lanes = 4;
+		dev_priv->vbt.edp.lanes = 4;
 		break;
 	default:
 		DRM_DEBUG_KMS("VBT has unknown eDP lane count value %u\n",
@@ -604,16 +618,16 @@ parse_edp(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 
 	switch (edp_link_params->preemphasis) {
 	case EDP_PREEMPHASIS_NONE:
-		dev_priv->vbt.edp_preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_0;
+		dev_priv->vbt.edp.preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_0;
 		break;
 	case EDP_PREEMPHASIS_3_5dB:
-		dev_priv->vbt.edp_preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_1;
+		dev_priv->vbt.edp.preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_1;
 		break;
 	case EDP_PREEMPHASIS_6dB:
-		dev_priv->vbt.edp_preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_2;
+		dev_priv->vbt.edp.preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_2;
 		break;
 	case EDP_PREEMPHASIS_9_5dB:
-		dev_priv->vbt.edp_preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_3;
+		dev_priv->vbt.edp.preemphasis = DP_TRAIN_PRE_EMPH_LEVEL_3;
 		break;
 	default:
 		DRM_DEBUG_KMS("VBT has unknown eDP pre-emphasis value %u\n",
@@ -623,16 +637,16 @@ parse_edp(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 
 	switch (edp_link_params->vswing) {
 	case EDP_VSWING_0_4V:
-		dev_priv->vbt.edp_vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_0;
+		dev_priv->vbt.edp.vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_0;
 		break;
 	case EDP_VSWING_0_6V:
-		dev_priv->vbt.edp_vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_1;
+		dev_priv->vbt.edp.vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_1;
 		break;
 	case EDP_VSWING_0_8V:
-		dev_priv->vbt.edp_vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_2;
+		dev_priv->vbt.edp.vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_2;
 		break;
 	case EDP_VSWING_1_2V:
-		dev_priv->vbt.edp_vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
+		dev_priv->vbt.edp.vswing = DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
 		break;
 	default:
 		DRM_DEBUG_KMS("VBT has unknown eDP voltage swing value %u\n",
@@ -645,10 +659,10 @@ parse_edp(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 
 		/* Don't read from VBT if module parameter has valid value*/
 		if (i915.edp_vswing) {
-			dev_priv->edp_low_vswing = i915.edp_vswing == 1;
+			dev_priv->vbt.edp.low_vswing = i915.edp_vswing == 1;
 		} else {
 			vswing = (edp->edp_vswing_preemph >> (panel_type * 4)) & 0xF;
-			dev_priv->edp_low_vswing = vswing == 0;
+			dev_priv->vbt.edp.low_vswing = vswing == 0;
 		}
 	}
 }
@@ -658,6 +672,7 @@ parse_psr(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 {
 	const struct bdb_psr *psr;
 	const struct psr_table *psr_table;
+	int panel_type = dev_priv->vbt.panel_type;
 
 	psr = find_section(bdb, BDB_PSR);
 	if (!psr) {
@@ -704,9 +719,10 @@ parse_mipi_config(struct drm_i915_private *dev_priv,
 	const struct bdb_mipi_config *start;
 	const struct mipi_config *config;
 	const struct mipi_pps_data *pps;
+	int panel_type = dev_priv->vbt.panel_type;
 
 	/* parse MIPI blocks only if LFP type is MIPI */
-	if (!dev_priv->vbt.has_mipi)
+	if (!intel_bios_is_dsi_present(dev_priv, NULL))
 		return;
 
 	/* Initialize this to undefined indicating no generic MIPI support */
@@ -911,6 +927,7 @@ static void
 parse_mipi_sequence(struct drm_i915_private *dev_priv,
 		    const struct bdb_header *bdb)
 {
+	int panel_type = dev_priv->vbt.panel_type;
 	const struct bdb_mipi_sequence *sequence;
 	const u8 *seq_data;
 	u32 seq_size;
@@ -1124,7 +1141,7 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
 	}
 
 	/* Parse the I_boost config for SKL and above */
-	if (bdb->version >= 196 && (child->common.flags_1 & IBOOST_ENABLE)) {
+	if (bdb->version >= 196 && child->common.iboost) {
 		info->dp_boost_level = translate_iboost(child->common.iboost_level & 0xF);
 		DRM_DEBUG_KMS("VBT (e)DP boost level for port %c: %d\n",
 			      port_name(port), info->dp_boost_level);
@@ -1232,14 +1249,6 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 			continue;
 		}
 
-		if (p_child->common.dvo_port >= DVO_PORT_MIPIA
-		    && p_child->common.dvo_port <= DVO_PORT_MIPID
-		    &&p_child->common.device_type & DEVICE_TYPE_MIPI_OUTPUT) {
-			DRM_DEBUG_KMS("Found MIPI as LFP\n");
-			dev_priv->vbt.has_mipi = 1;
-			dev_priv->vbt.dsi.port = p_child->common.dvo_port;
-		}
-
 		child_dev_ptr = dev_priv->vbt.child_dev + count;
 		count++;
 
@@ -1250,6 +1259,19 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 		 */
 		memcpy(child_dev_ptr, p_child,
 		       min_t(size_t, p_defs->child_dev_size, sizeof(*p_child)));
+
+		/*
+		 * copied full block, now init values when they are not
+		 * available in current version
+		 */
+		if (bdb->version < 196) {
+			/* Set default values for bits added from v196 */
+			child_dev_ptr->common.iboost = 0;
+			child_dev_ptr->common.hpd_invert = 0;
+		}
+
+		if (bdb->version < 192)
+			child_dev_ptr->common.lspcon = 0;
 	}
 	return;
 }
@@ -1431,3 +1453,210 @@ intel_bios_init(struct drm_i915_private *dev_priv)
 
 	return 0;
 }
+
+/**
+ * intel_bios_is_tv_present - is integrated TV present in VBT
+ * @dev_priv:	i915 device instance
+ *
+ * Return true if TV is present. If no child devices were parsed from VBT,
+ * assume TV is present.
+ */
+bool intel_bios_is_tv_present(struct drm_i915_private *dev_priv)
+{
+	union child_device_config *p_child;
+	int i;
+
+	if (!dev_priv->vbt.int_tv_support)
+		return false;
+
+	if (!dev_priv->vbt.child_dev_num)
+		return true;
+
+	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		p_child = dev_priv->vbt.child_dev + i;
+		/*
+		 * If the device type is not TV, continue.
+		 */
+		switch (p_child->old.device_type) {
+		case DEVICE_TYPE_INT_TV:
+		case DEVICE_TYPE_TV:
+		case DEVICE_TYPE_TV_SVIDEO_COMPOSITE:
+			break;
+		default:
+			continue;
+		}
+		/* Only when the addin_offset is non-zero, it is regarded
+		 * as present.
+		 */
+		if (p_child->old.addin_offset)
+			return true;
+	}
+
+	return false;
+}
+
+/**
+ * intel_bios_is_lvds_present - is LVDS present in VBT
+ * @dev_priv:	i915 device instance
+ * @i2c_pin:	i2c pin for LVDS if present
+ *
+ * Return true if LVDS is present. If no child devices were parsed from VBT,
+ * assume LVDS is present.
+ */
+bool intel_bios_is_lvds_present(struct drm_i915_private *dev_priv, u8 *i2c_pin)
+{
+	int i;
+
+	if (!dev_priv->vbt.child_dev_num)
+		return true;
+
+	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		union child_device_config *uchild = dev_priv->vbt.child_dev + i;
+		struct old_child_dev_config *child = &uchild->old;
+
+		/* If the device type is not LFP, continue.
+		 * We have to check both the new identifiers as well as the
+		 * old for compatibility with some BIOSes.
+		 */
+		if (child->device_type != DEVICE_TYPE_INT_LFP &&
+		    child->device_type != DEVICE_TYPE_LFP)
+			continue;
+
+		if (intel_gmbus_is_valid_pin(dev_priv, child->i2c_pin))
+			*i2c_pin = child->i2c_pin;
+
+		/* However, we cannot trust the BIOS writers to populate
+		 * the VBT correctly.  Since LVDS requires additional
+		 * information from AIM blocks, a non-zero addin offset is
+		 * a good indicator that the LVDS is actually present.
+		 */
+		if (child->addin_offset)
+			return true;
+
+		/* But even then some BIOS writers perform some black magic
+		 * and instantiate the device without reference to any
+		 * additional data.  Trust that if the VBT was written into
+		 * the OpRegion then they have validated the LVDS's existence.
+		 */
+		if (dev_priv->opregion.vbt)
+			return true;
+	}
+
+	return false;
+}
+
+/**
+ * intel_bios_is_port_edp - is the device in given port eDP
+ * @dev_priv:	i915 device instance
+ * @port:	port to check
+ *
+ * Return true if the device in %port is eDP.
+ */
+bool intel_bios_is_port_edp(struct drm_i915_private *dev_priv, enum port port)
+{
+	union child_device_config *p_child;
+	static const short port_mapping[] = {
+		[PORT_B] = DVO_PORT_DPB,
+		[PORT_C] = DVO_PORT_DPC,
+		[PORT_D] = DVO_PORT_DPD,
+		[PORT_E] = DVO_PORT_DPE,
+	};
+	int i;
+
+	if (!dev_priv->vbt.child_dev_num)
+		return false;
+
+	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		p_child = dev_priv->vbt.child_dev + i;
+
+		if (p_child->common.dvo_port == port_mapping[port] &&
+		    (p_child->common.device_type & DEVICE_TYPE_eDP_BITS) ==
+		    (DEVICE_TYPE_eDP & DEVICE_TYPE_eDP_BITS))
+			return true;
+	}
+
+	return false;
+}
+
+/**
+ * intel_bios_is_dsi_present - is DSI present in VBT
+ * @dev_priv:	i915 device instance
+ * @port:	port for DSI if present
+ *
+ * Return true if DSI is present, and return the port in %port.
+ */
+bool intel_bios_is_dsi_present(struct drm_i915_private *dev_priv,
+			       enum port *port)
+{
+	union child_device_config *p_child;
+	u8 dvo_port;
+	int i;
+
+	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		p_child = dev_priv->vbt.child_dev + i;
+
+		if (!(p_child->common.device_type & DEVICE_TYPE_MIPI_OUTPUT))
+			continue;
+
+		dvo_port = p_child->common.dvo_port;
+
+		switch (dvo_port) {
+		case DVO_PORT_MIPIA:
+		case DVO_PORT_MIPIC:
+			if (port)
+				*port = dvo_port - DVO_PORT_MIPIA;
+			return true;
+		case DVO_PORT_MIPIB:
+		case DVO_PORT_MIPID:
+			DRM_DEBUG_KMS("VBT has unsupported DSI port %c\n",
+				      port_name(dvo_port - DVO_PORT_MIPIA));
+			break;
+		}
+	}
+
+	return false;
+}
+
+/**
+ * intel_bios_is_port_hpd_inverted - is HPD inverted for %port
+ * @dev_priv:	i915 device instance
+ * @port:	port to check
+ *
+ * Return true if HPD should be inverted for %port.
+ */
+bool
+intel_bios_is_port_hpd_inverted(struct drm_i915_private *dev_priv,
+				enum port port)
+{
+	int i;
+
+	if (WARN_ON_ONCE(!IS_BROXTON(dev_priv)))
+		return false;
+
+	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		if (!dev_priv->vbt.child_dev[i].common.hpd_invert)
+			continue;
+
+		switch (dev_priv->vbt.child_dev[i].common.dvo_port) {
+		case DVO_PORT_DPA:
+		case DVO_PORT_HDMIA:
+			if (port == PORT_A)
+				return true;
+			break;
+		case DVO_PORT_DPB:
+		case DVO_PORT_HDMIB:
+			if (port == PORT_B)
+				return true;
+			break;
+		case DVO_PORT_DPC:
+		case DVO_PORT_HDMIC:
+			if (port == PORT_C)
+				return true;
+			break;
+		default:
+			break;
+		}
+	}
+
+	return false;
+}
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index 350d4e0f75a4..ab0ea315eddb 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -1,5 +1,5 @@
 /*
- * Copyright © 2006 Intel Corporation
+ * Copyright © 2016 Intel Corporation
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -19,543 +19,16 @@
  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
- *
- * Authors:
- *    Eric Anholt <eric@anholt.net>
- *
- */
-
-#ifndef _INTEL_BIOS_H_
-#define _INTEL_BIOS_H_
-
-/**
- * struct vbt_header - VBT Header structure
- * @signature:		VBT signature, always starts with "$VBT"
- * @version:		Version of this structure
- * @header_size:	Size of this structure
- * @vbt_size:		Size of VBT (VBT Header, BDB Header and data blocks)
- * @vbt_checksum:	Checksum
- * @reserved0:		Reserved
- * @bdb_offset:		Offset of &struct bdb_header from beginning of VBT
- * @aim_offset:		Offsets of add-in data blocks from beginning of VBT
- */
-struct vbt_header {
-	u8 signature[20];
-	u16 version;
-	u16 header_size;
-	u16 vbt_size;
-	u8 vbt_checksum;
-	u8 reserved0;
-	u32 bdb_offset;
-	u32 aim_offset[4];
-} __packed;
-
-/**
- * struct bdb_header - BDB Header structure
- * @signature:		BDB signature "BIOS_DATA_BLOCK"
- * @version:		Version of the data block definitions
- * @header_size:	Size of this structure
- * @bdb_size:		Size of BDB (BDB Header and data blocks)
- */
-struct bdb_header {
-	u8 signature[16];
-	u16 version;
-	u16 header_size;
-	u16 bdb_size;
-} __packed;
-
-/* strictly speaking, this is a "skip" block, but it has interesting info */
-struct vbios_data {
-	u8 type; /* 0 == desktop, 1 == mobile */
-	u8 relstage;
-	u8 chipset;
-	u8 lvds_present:1;
-	u8 tv_present:1;
-	u8 rsvd2:6; /* finish byte */
-	u8 rsvd3[4];
-	u8 signon[155];
-	u8 copyright[61];
-	u16 code_segment;
-	u8 dos_boot_mode;
-	u8 bandwidth_percent;
-	u8 rsvd4; /* popup memory size */
-	u8 resize_pci_bios;
-	u8 rsvd5; /* is crt already on ddc2 */
-} __packed;
-
-/*
- * There are several types of BIOS data blocks (BDBs), each block has
- * an ID and size in the first 3 bytes (ID in first, size in next 2).
- * Known types are listed below.
  */
-#define BDB_GENERAL_FEATURES	  1
-#define BDB_GENERAL_DEFINITIONS	  2
-#define BDB_OLD_TOGGLE_LIST	  3
-#define BDB_MODE_SUPPORT_LIST	  4
-#define BDB_GENERIC_MODE_TABLE	  5
-#define BDB_EXT_MMIO_REGS	  6
-#define BDB_SWF_IO		  7
-#define BDB_SWF_MMIO		  8
-#define BDB_PSR			  9
-#define BDB_MODE_REMOVAL_TABLE	 10
-#define BDB_CHILD_DEVICE_TABLE	 11
-#define BDB_DRIVER_FEATURES	 12
-#define BDB_DRIVER_PERSISTENCE	 13
-#define BDB_EXT_TABLE_PTRS	 14
-#define BDB_DOT_CLOCK_OVERRIDE	 15
-#define BDB_DISPLAY_SELECT	 16
-/* 17 rsvd */
-#define BDB_DRIVER_ROTATION	 18
-#define BDB_DISPLAY_REMOVE	 19
-#define BDB_OEM_CUSTOM		 20
-#define BDB_EFP_LIST		 21 /* workarounds for VGA hsync/vsync */
-#define BDB_SDVO_LVDS_OPTIONS	 22
-#define BDB_SDVO_PANEL_DTDS	 23
-#define BDB_SDVO_LVDS_PNP_IDS	 24
-#define BDB_SDVO_LVDS_POWER_SEQ	 25
-#define BDB_TV_OPTIONS		 26
-#define BDB_EDP			 27
-#define BDB_LVDS_OPTIONS	 40
-#define BDB_LVDS_LFP_DATA_PTRS	 41
-#define BDB_LVDS_LFP_DATA	 42
-#define BDB_LVDS_BACKLIGHT	 43
-#define BDB_LVDS_POWER		 44
-#define BDB_MIPI_CONFIG		 52
-#define BDB_MIPI_SEQUENCE	 53
-#define BDB_SKIP		254 /* VBIOS private block, ignore */
-
-struct bdb_general_features {
-        /* bits 1 */
-	u8 panel_fitting:2;
-	u8 flexaim:1;
-	u8 msg_enable:1;
-	u8 clear_screen:3;
-	u8 color_flip:1;
-
-        /* bits 2 */
-	u8 download_ext_vbt:1;
-	u8 enable_ssc:1;
-	u8 ssc_freq:1;
-	u8 enable_lfp_on_override:1;
-	u8 disable_ssc_ddt:1;
-	u8 rsvd7:1;
-	u8 display_clock_mode:1;
-	u8 rsvd8:1; /* finish byte */
-
-        /* bits 3 */
-	u8 disable_smooth_vision:1;
-	u8 single_dvi:1;
-	u8 rsvd9:1;
-	u8 fdi_rx_polarity_inverted:1;
-	u8 rsvd10:4; /* finish byte */
-
-        /* bits 4 */
-	u8 legacy_monitor_detect;
-
-        /* bits 5 */
-	u8 int_crt_support:1;
-	u8 int_tv_support:1;
-	u8 int_efp_support:1;
-	u8 dp_ssc_enb:1;	/* PCH attached eDP supports SSC */
-	u8 dp_ssc_freq:1;	/* SSC freq for PCH attached eDP */
-	u8 rsvd11:3; /* finish byte */
-} __packed;
-
-/* pre-915 */
-#define GPIO_PIN_DVI_LVDS	0x03 /* "DVI/LVDS DDC GPIO pins" */
-#define GPIO_PIN_ADD_I2C	0x05 /* "ADDCARD I2C GPIO pins" */
-#define GPIO_PIN_ADD_DDC	0x04 /* "ADDCARD DDC GPIO pins" */
-#define GPIO_PIN_ADD_DDC_I2C	0x06 /* "ADDCARD DDC/I2C GPIO pins" */
-
-/* Pre 915 */
-#define DEVICE_TYPE_NONE	0x00
-#define DEVICE_TYPE_CRT		0x01
-#define DEVICE_TYPE_TV		0x09
-#define DEVICE_TYPE_EFP		0x12
-#define DEVICE_TYPE_LFP		0x22
-/* On 915+ */
-#define DEVICE_TYPE_CRT_DPMS		0x6001
-#define DEVICE_TYPE_CRT_DPMS_HOTPLUG	0x4001
-#define DEVICE_TYPE_TV_COMPOSITE	0x0209
-#define DEVICE_TYPE_TV_MACROVISION	0x0289
-#define DEVICE_TYPE_TV_RF_COMPOSITE	0x020c
-#define DEVICE_TYPE_TV_SVIDEO_COMPOSITE	0x0609
-#define DEVICE_TYPE_TV_SCART		0x0209
-#define DEVICE_TYPE_TV_CODEC_HOTPLUG_PWR 0x6009
-#define DEVICE_TYPE_EFP_HOTPLUG_PWR	0x6012
-#define DEVICE_TYPE_EFP_DVI_HOTPLUG_PWR	0x6052
-#define DEVICE_TYPE_EFP_DVI_I		0x6053
-#define DEVICE_TYPE_EFP_DVI_D_DUAL	0x6152
-#define DEVICE_TYPE_EFP_DVI_D_HDCP	0x60d2
-#define DEVICE_TYPE_OPENLDI_HOTPLUG_PWR	0x6062
-#define DEVICE_TYPE_OPENLDI_DUALPIX	0x6162
-#define DEVICE_TYPE_LFP_PANELLINK	0x5012
-#define DEVICE_TYPE_LFP_CMOS_PWR	0x5042
-#define DEVICE_TYPE_LFP_LVDS_PWR	0x5062
-#define DEVICE_TYPE_LFP_LVDS_DUAL	0x5162
-#define DEVICE_TYPE_LFP_LVDS_DUAL_HDCP	0x51e2
-
-#define DEVICE_CFG_NONE		0x00
-#define DEVICE_CFG_12BIT_DVOB	0x01
-#define DEVICE_CFG_12BIT_DVOC	0x02
-#define DEVICE_CFG_24BIT_DVOBC	0x09
-#define DEVICE_CFG_24BIT_DVOCB	0x0a
-#define DEVICE_CFG_DUAL_DVOB	0x11
-#define DEVICE_CFG_DUAL_DVOC	0x12
-#define DEVICE_CFG_DUAL_DVOBC	0x13
-#define DEVICE_CFG_DUAL_LINK_DVOBC	0x19
-#define DEVICE_CFG_DUAL_LINK_DVOCB	0x1a
-
-#define DEVICE_WIRE_NONE	0x00
-#define DEVICE_WIRE_DVOB	0x01
-#define DEVICE_WIRE_DVOC	0x02
-#define DEVICE_WIRE_DVOBC	0x03
-#define DEVICE_WIRE_DVOBB	0x05
-#define DEVICE_WIRE_DVOCC	0x06
-#define DEVICE_WIRE_DVOB_MASTER 0x0d
-#define DEVICE_WIRE_DVOC_MASTER 0x0e
-
-#define DEVICE_PORT_DVOA	0x00 /* none on 845+ */
-#define DEVICE_PORT_DVOB	0x01
-#define DEVICE_PORT_DVOC	0x02
 
 /*
- * We used to keep this struct but without any version control. We should avoid
- * using it in the future, but it should be safe to keep using it in the old
- * code. Do not change; we rely on its size.
+ * Please use intel_vbt_defs.h for VBT private data, to hide and abstract away
+ * the VBT from the rest of the driver. Add the parsed, clean data to struct
+ * intel_vbt_data within struct drm_i915_private.
  */
-struct old_child_dev_config {
-	u16 handle;
-	u16 device_type;
-	u8  device_id[10]; /* ascii string */
-	u16 addin_offset;
-	u8  dvo_port; /* See Device_PORT_* above */
-	u8  i2c_pin;
-	u8  slave_addr;
-	u8  ddc_pin;
-	u16 edid_ptr;
-	u8  dvo_cfg; /* See DEVICE_CFG_* above */
-	u8  dvo2_port;
-	u8  i2c2_pin;
-	u8  slave2_addr;
-	u8  ddc2_pin;
-	u8  capabilities;
-	u8  dvo_wiring;/* See DEVICE_WIRE_* above */
-	u8  dvo2_wiring;
-	u16 extended_type;
-	u8  dvo_function;
-} __packed;
-
-/* This one contains field offsets that are known to be common for all BDB
- * versions. Notice that the meaning of the contents contents may still change,
- * but at least the offsets are consistent. */
-
-/* Definitions for flags_1 */
-#define IBOOST_ENABLE (1<<3)
-
-struct common_child_dev_config {
-	u16 handle;
-	u16 device_type;
-	u8 not_common1[12];
-	u8 dvo_port;
-	u8 not_common2[2];
-	u8 ddc_pin;
-	u16 edid_ptr;
-	u8 obsolete;
-	u8 flags_1;
-	u8 not_common3[13];
-	u8 iboost_level;
-} __packed;
-
-
-/* This field changes depending on the BDB version, so the most reliable way to
- * read it is by checking the BDB version and reading the raw pointer. */
-union child_device_config {
-	/* This one is safe to be used anywhere, but the code should still check
-	 * the BDB version. */
-	u8 raw[33];
-	/* This one should only be kept for legacy code. */
-	struct old_child_dev_config old;
-	/* This one should also be safe to use anywhere, even without version
-	 * checks. */
-	struct common_child_dev_config common;
-} __packed;
-
-struct bdb_general_definitions {
-	/* DDC GPIO */
-	u8 crt_ddc_gmbus_pin;
-
-	/* DPMS bits */
-	u8 dpms_acpi:1;
-	u8 skip_boot_crt_detect:1;
-	u8 dpms_aim:1;
-	u8 rsvd1:5; /* finish byte */
-
-	/* boot device bits */
-	u8 boot_display[2];
-	u8 child_dev_size;
-
-	/*
-	 * Device info:
-	 * If TV is present, it'll be at devices[0].
-	 * LVDS will be next, either devices[0] or [1], if present.
-	 * On some platforms the number of device is 6. But could be as few as
-	 * 4 if both TV and LVDS are missing.
-	 * And the device num is related with the size of general definition
-	 * block. It is obtained by using the following formula:
-	 * number = (block_size - sizeof(bdb_general_definitions))/
-	 *	     defs->child_dev_size;
-	 */
-	uint8_t devices[0];
-} __packed;
-
-/* Mask for DRRS / Panel Channel / SSC / BLT control bits extraction */
-#define MODE_MASK		0x3
-
-struct bdb_lvds_options {
-	u8 panel_type;
-	u8 rsvd1;
-	/* LVDS capabilities, stored in a dword */
-	u8 pfit_mode:2;
-	u8 pfit_text_mode_enhanced:1;
-	u8 pfit_gfx_mode_enhanced:1;
-	u8 pfit_ratio_auto:1;
-	u8 pixel_dither:1;
-	u8 lvds_edid:1;
-	u8 rsvd2:1;
-	u8 rsvd4;
-	/* LVDS Panel channel bits stored here */
-	u32 lvds_panel_channel_bits;
-	/* LVDS SSC (Spread Spectrum Clock) bits stored here. */
-	u16 ssc_bits;
-	u16 ssc_freq;
-	u16 ssc_ddt;
-	/* Panel color depth defined here */
-	u16 panel_color_depth;
-	/* LVDS panel type bits stored here */
-	u32 dps_panel_type_bits;
-	/* LVDS backlight control type bits stored here */
-	u32 blt_control_type_bits;
-} __packed;
-
-/* LFP pointer table contains entries to the struct below */
-struct bdb_lvds_lfp_data_ptr {
-	u16 fp_timing_offset; /* offsets are from start of bdb */
-	u8 fp_table_size;
-	u16 dvo_timing_offset;
-	u8 dvo_table_size;
-	u16 panel_pnp_id_offset;
-	u8 pnp_table_size;
-} __packed;
-
-struct bdb_lvds_lfp_data_ptrs {
-	u8 lvds_entries; /* followed by one or more lvds_data_ptr structs */
-	struct bdb_lvds_lfp_data_ptr ptr[16];
-} __packed;
-
-/* LFP data has 3 blocks per entry */
-struct lvds_fp_timing {
-	u16 x_res;
-	u16 y_res;
-	u32 lvds_reg;
-	u32 lvds_reg_val;
-	u32 pp_on_reg;
-	u32 pp_on_reg_val;
-	u32 pp_off_reg;
-	u32 pp_off_reg_val;
-	u32 pp_cycle_reg;
-	u32 pp_cycle_reg_val;
-	u32 pfit_reg;
-	u32 pfit_reg_val;
-	u16 terminator;
-} __packed;
-
-struct lvds_dvo_timing {
-	u16 clock;		/**< In 10khz */
-	u8 hactive_lo;
-	u8 hblank_lo;
-	u8 hblank_hi:4;
-	u8 hactive_hi:4;
-	u8 vactive_lo;
-	u8 vblank_lo;
-	u8 vblank_hi:4;
-	u8 vactive_hi:4;
-	u8 hsync_off_lo;
-	u8 hsync_pulse_width;
-	u8 vsync_pulse_width:4;
-	u8 vsync_off:4;
-	u8 rsvd0:6;
-	u8 hsync_off_hi:2;
-	u8 h_image;
-	u8 v_image;
-	u8 max_hv;
-	u8 h_border;
-	u8 v_border;
-	u8 rsvd1:3;
-	u8 digital:2;
-	u8 vsync_positive:1;
-	u8 hsync_positive:1;
-	u8 rsvd2:1;
-} __packed;
-
-struct lvds_pnp_id {
-	u16 mfg_name;
-	u16 product_code;
-	u32 serial;
-	u8 mfg_week;
-	u8 mfg_year;
-} __packed;
-
-struct bdb_lvds_lfp_data_entry {
-	struct lvds_fp_timing fp_timing;
-	struct lvds_dvo_timing dvo_timing;
-	struct lvds_pnp_id pnp_id;
-} __packed;
-
-struct bdb_lvds_lfp_data {
-	struct bdb_lvds_lfp_data_entry data[16];
-} __packed;
-
-#define BDB_BACKLIGHT_TYPE_NONE	0
-#define BDB_BACKLIGHT_TYPE_PWM	2
-
-struct bdb_lfp_backlight_data_entry {
-	u8 type:2;
-	u8 active_low_pwm:1;
-	u8 obsolete1:5;
-	u16 pwm_freq_hz;
-	u8 min_brightness;
-	u8 obsolete2;
-	u8 obsolete3;
-} __packed;
-
-struct bdb_lfp_backlight_data {
-	u8 entry_size;
-	struct bdb_lfp_backlight_data_entry data[16];
-	u8 level[16];
-} __packed;
-
-struct aimdb_header {
-	char signature[16];
-	char oem_device[20];
-	u16 aimdb_version;
-	u16 aimdb_header_size;
-	u16 aimdb_size;
-} __packed;
-
-struct aimdb_block {
-	u8 aimdb_id;
-	u16 aimdb_size;
-} __packed;
 
-struct vch_panel_data {
-	u16 fp_timing_offset;
-	u8 fp_timing_size;
-	u16 dvo_timing_offset;
-	u8 dvo_timing_size;
-	u16 text_fitting_offset;
-	u8 text_fitting_size;
-	u16 graphics_fitting_offset;
-	u8 graphics_fitting_size;
-} __packed;
-
-struct vch_bdb_22 {
-	struct aimdb_block aimdb_block;
-	struct vch_panel_data panels[16];
-} __packed;
-
-struct bdb_sdvo_lvds_options {
-	u8 panel_backlight;
-	u8 h40_set_panel_type;
-	u8 panel_type;
-	u8 ssc_clk_freq;
-	u16 als_low_trip;
-	u16 als_high_trip;
-	u8 sclalarcoeff_tab_row_num;
-	u8 sclalarcoeff_tab_row_size;
-	u8 coefficient[8];
-	u8 panel_misc_bits_1;
-	u8 panel_misc_bits_2;
-	u8 panel_misc_bits_3;
-	u8 panel_misc_bits_4;
-} __packed;
-
-
-#define BDB_DRIVER_FEATURE_NO_LVDS		0
-#define BDB_DRIVER_FEATURE_INT_LVDS		1
-#define BDB_DRIVER_FEATURE_SDVO_LVDS		2
-#define BDB_DRIVER_FEATURE_EDP			3
-
-struct bdb_driver_features {
-	u8 boot_dev_algorithm:1;
-	u8 block_display_switch:1;
-	u8 allow_display_switch:1;
-	u8 hotplug_dvo:1;
-	u8 dual_view_zoom:1;
-	u8 int15h_hook:1;
-	u8 sprite_in_clone:1;
-	u8 primary_lfp_id:1;
-
-	u16 boot_mode_x;
-	u16 boot_mode_y;
-	u8 boot_mode_bpp;
-	u8 boot_mode_refresh;
-
-	u16 enable_lfp_primary:1;
-	u16 selective_mode_pruning:1;
-	u16 dual_frequency:1;
-	u16 render_clock_freq:1; /* 0: high freq; 1: low freq */
-	u16 nt_clone_support:1;
-	u16 power_scheme_ui:1; /* 0: CUI; 1: 3rd party */
-	u16 sprite_display_assign:1; /* 0: secondary; 1: primary */
-	u16 cui_aspect_scaling:1;
-	u16 preserve_aspect_ratio:1;
-	u16 sdvo_device_power_down:1;
-	u16 crt_hotplug:1;
-	u16 lvds_config:2;
-	u16 tv_hotplug:1;
-	u16 hdmi_config:2;
-
-	u8 static_display:1;
-	u8 reserved2:7;
-	u16 legacy_crt_max_x;
-	u16 legacy_crt_max_y;
-	u8 legacy_crt_max_refresh;
-
-	u8 hdmi_termination;
-	u8 custom_vbt_version;
-	/* Driver features data block */
-	u16 rmpm_enabled:1;
-	u16 s2ddt_enabled:1;
-	u16 dpst_enabled:1;
-	u16 bltclt_enabled:1;
-	u16 adb_enabled:1;
-	u16 drrs_enabled:1;
-	u16 grs_enabled:1;
-	u16 gpmt_enabled:1;
-	u16 tbt_enabled:1;
-	u16 psr_enabled:1;
-	u16 ips_enabled:1;
-	u16 reserved3:4;
-	u16 pc_feature_valid:1;
-} __packed;
-
-#define EDP_18BPP	0
-#define EDP_24BPP	1
-#define EDP_30BPP	2
-#define EDP_RATE_1_62	0
-#define EDP_RATE_2_7	1
-#define EDP_LANE_1	0
-#define EDP_LANE_2	1
-#define EDP_LANE_4	3
-#define EDP_PREEMPHASIS_NONE	0
-#define EDP_PREEMPHASIS_3_5dB	1
-#define EDP_PREEMPHASIS_6dB	2
-#define EDP_PREEMPHASIS_9_5dB	3
-#define EDP_VSWING_0_4V		0
-#define EDP_VSWING_0_6V		1
-#define EDP_VSWING_0_8V		2
-#define EDP_VSWING_1_2V		3
+#ifndef _INTEL_BIOS_H_
+#define _INTEL_BIOS_H_
 
 struct edp_power_seq {
 	u16 t1_t3;
@@ -565,245 +38,37 @@ struct edp_power_seq {
 	u16 t11_t12;
 } __packed;
 
-struct edp_link_params {
-	u8 rate:4;
-	u8 lanes:4;
-	u8 preemphasis:4;
-	u8 vswing:4;
-} __packed;
-
-struct bdb_edp {
-	struct edp_power_seq power_seqs[16];
-	u32 color_depth;
-	struct edp_link_params link_params[16];
-	u32 sdrrs_msa_timing_delay;
-
-	/* ith bit indicates enabled/disabled for (i+1)th panel */
-	u16 edp_s3d_feature;
-	u16 edp_t3_optimization;
-	u64 edp_vswing_preemph;		/* v173 */
-} __packed;
-
-struct psr_table {
-	/* Feature bits */
-	u8 full_link:1;
-	u8 require_aux_to_wakeup:1;
-	u8 feature_bits_rsvd:6;
-
-	/* Wait times */
-	u8 idle_frames:4;
-	u8 lines_to_wait:3;
-	u8 wait_times_rsvd:1;
-
-	/* TP wake up time in multiple of 100 */
-	u16 tp1_wakeup_time;
-	u16 tp2_tp3_wakeup_time;
-} __packed;
-
-struct bdb_psr {
-	struct psr_table psr_table[16];
-} __packed;
-
-/*
- * Driver<->VBIOS interaction occurs through scratch bits in
- * GR18 & SWF*.
- */
-
-/* GR18 bits are set on display switch and hotkey events */
-#define GR18_DRIVER_SWITCH_EN	(1<<7) /* 0: VBIOS control, 1: driver control */
-#define GR18_HOTKEY_MASK	0x78 /* See also SWF4 15:0 */
-#define   GR18_HK_NONE		(0x0<<3)
-#define   GR18_HK_LFP_STRETCH	(0x1<<3)
-#define   GR18_HK_TOGGLE_DISP	(0x2<<3)
-#define   GR18_HK_DISP_SWITCH	(0x4<<3) /* see SWF14 15:0 for what to enable */
-#define   GR18_HK_POPUP_DISABLED (0x6<<3)
-#define   GR18_HK_POPUP_ENABLED	(0x7<<3)
-#define   GR18_HK_PFIT		(0x8<<3)
-#define   GR18_HK_APM_CHANGE	(0xa<<3)
-#define   GR18_HK_MULTIPLE	(0xc<<3)
-#define GR18_USER_INT_EN	(1<<2)
-#define GR18_A0000_FLUSH_EN	(1<<1)
-#define GR18_SMM_EN		(1<<0)
-
-/* Set by driver, cleared by VBIOS */
-#define SWF00_YRES_SHIFT	16
-#define SWF00_XRES_SHIFT	0
-#define SWF00_RES_MASK		0xffff
-
-/* Set by VBIOS at boot time and driver at runtime */
-#define SWF01_TV2_FORMAT_SHIFT	8
-#define SWF01_TV1_FORMAT_SHIFT	0
-#define SWF01_TV_FORMAT_MASK	0xffff
-
-#define SWF10_VBIOS_BLC_I2C_EN	(1<<29)
-#define SWF10_GTT_OVERRIDE_EN	(1<<28)
-#define SWF10_LFP_DPMS_OVR	(1<<27) /* override DPMS on display switch */
-#define SWF10_ACTIVE_TOGGLE_LIST_MASK (7<<24)
-#define   SWF10_OLD_TOGGLE	0x0
-#define   SWF10_TOGGLE_LIST_1	0x1
-#define   SWF10_TOGGLE_LIST_2	0x2
-#define   SWF10_TOGGLE_LIST_3	0x3
-#define   SWF10_TOGGLE_LIST_4	0x4
-#define SWF10_PANNING_EN	(1<<23)
-#define SWF10_DRIVER_LOADED	(1<<22)
-#define SWF10_EXTENDED_DESKTOP	(1<<21)
-#define SWF10_EXCLUSIVE_MODE	(1<<20)
-#define SWF10_OVERLAY_EN	(1<<19)
-#define SWF10_PLANEB_HOLDOFF	(1<<18)
-#define SWF10_PLANEA_HOLDOFF	(1<<17)
-#define SWF10_VGA_HOLDOFF	(1<<16)
-#define SWF10_ACTIVE_DISP_MASK	0xffff
-#define   SWF10_PIPEB_LFP2	(1<<15)
-#define   SWF10_PIPEB_EFP2	(1<<14)
-#define   SWF10_PIPEB_TV2	(1<<13)
-#define   SWF10_PIPEB_CRT2	(1<<12)
-#define   SWF10_PIPEB_LFP	(1<<11)
-#define   SWF10_PIPEB_EFP	(1<<10)
-#define   SWF10_PIPEB_TV	(1<<9)
-#define   SWF10_PIPEB_CRT	(1<<8)
-#define   SWF10_PIPEA_LFP2	(1<<7)
-#define   SWF10_PIPEA_EFP2	(1<<6)
-#define   SWF10_PIPEA_TV2	(1<<5)
-#define   SWF10_PIPEA_CRT2	(1<<4)
-#define   SWF10_PIPEA_LFP	(1<<3)
-#define   SWF10_PIPEA_EFP	(1<<2)
-#define   SWF10_PIPEA_TV	(1<<1)
-#define   SWF10_PIPEA_CRT	(1<<0)
-
-#define SWF11_MEMORY_SIZE_SHIFT	16
-#define SWF11_SV_TEST_EN	(1<<15)
-#define SWF11_IS_AGP		(1<<14)
-#define SWF11_DISPLAY_HOLDOFF	(1<<13)
-#define SWF11_DPMS_REDUCED	(1<<12)
-#define SWF11_IS_VBE_MODE	(1<<11)
-#define SWF11_PIPEB_ACCESS	(1<<10) /* 0 here means pipe a */
-#define SWF11_DPMS_MASK		0x07
-#define   SWF11_DPMS_OFF	(1<<2)
-#define   SWF11_DPMS_SUSPEND	(1<<1)
-#define   SWF11_DPMS_STANDBY	(1<<0)
-#define   SWF11_DPMS_ON		0
-
-#define SWF14_GFX_PFIT_EN	(1<<31)
-#define SWF14_TEXT_PFIT_EN	(1<<30)
-#define SWF14_LID_STATUS_CLOSED	(1<<29) /* 0 here means open */
-#define SWF14_POPUP_EN		(1<<28)
-#define SWF14_DISPLAY_HOLDOFF	(1<<27)
-#define SWF14_DISP_DETECT_EN	(1<<26)
-#define SWF14_DOCKING_STATUS_DOCKED (1<<25) /* 0 here means undocked */
-#define SWF14_DRIVER_STATUS	(1<<24)
-#define SWF14_OS_TYPE_WIN9X	(1<<23)
-#define SWF14_OS_TYPE_WINNT	(1<<22)
-/* 21:19 rsvd */
-#define SWF14_PM_TYPE_MASK	0x00070000
-#define   SWF14_PM_ACPI_VIDEO	(0x4 << 16)
-#define   SWF14_PM_ACPI		(0x3 << 16)
-#define   SWF14_PM_APM_12	(0x2 << 16)
-#define   SWF14_PM_APM_11	(0x1 << 16)
-#define SWF14_HK_REQUEST_MASK	0x0000ffff /* see GR18 6:3 for event type */
-          /* if GR18 indicates a display switch */
-#define   SWF14_DS_PIPEB_LFP2_EN (1<<15)
-#define   SWF14_DS_PIPEB_EFP2_EN (1<<14)
-#define   SWF14_DS_PIPEB_TV2_EN  (1<<13)
-#define   SWF14_DS_PIPEB_CRT2_EN (1<<12)
-#define   SWF14_DS_PIPEB_LFP_EN  (1<<11)
-#define   SWF14_DS_PIPEB_EFP_EN  (1<<10)
-#define   SWF14_DS_PIPEB_TV_EN   (1<<9)
-#define   SWF14_DS_PIPEB_CRT_EN  (1<<8)
-#define   SWF14_DS_PIPEA_LFP2_EN (1<<7)
-#define   SWF14_DS_PIPEA_EFP2_EN (1<<6)
-#define   SWF14_DS_PIPEA_TV2_EN  (1<<5)
-#define   SWF14_DS_PIPEA_CRT2_EN (1<<4)
-#define   SWF14_DS_PIPEA_LFP_EN  (1<<3)
-#define   SWF14_DS_PIPEA_EFP_EN  (1<<2)
-#define   SWF14_DS_PIPEA_TV_EN   (1<<1)
-#define   SWF14_DS_PIPEA_CRT_EN  (1<<0)
-          /* if GR18 indicates a panel fitting request */
-#define   SWF14_PFIT_EN		(1<<0) /* 0 means disable */
-          /* if GR18 indicates an APM change request */
-#define   SWF14_APM_HIBERNATE	0x4
-#define   SWF14_APM_SUSPEND	0x3
-#define   SWF14_APM_STANDBY	0x1
-#define   SWF14_APM_RESTORE	0x0
-
-/* Add the device class for LFP, TV, HDMI */
-#define	 DEVICE_TYPE_INT_LFP	0x1022
-#define	 DEVICE_TYPE_INT_TV	0x1009
-#define	 DEVICE_TYPE_HDMI	0x60D2
-#define	 DEVICE_TYPE_DP		0x68C6
-#define	 DEVICE_TYPE_eDP	0x78C6
-
-#define  DEVICE_TYPE_CLASS_EXTENSION	(1 << 15)
-#define  DEVICE_TYPE_POWER_MANAGEMENT	(1 << 14)
-#define  DEVICE_TYPE_HOTPLUG_SIGNALING	(1 << 13)
-#define  DEVICE_TYPE_INTERNAL_CONNECTOR	(1 << 12)
-#define  DEVICE_TYPE_NOT_HDMI_OUTPUT	(1 << 11)
-#define  DEVICE_TYPE_MIPI_OUTPUT	(1 << 10)
-#define  DEVICE_TYPE_COMPOSITE_OUTPUT	(1 << 9)
-#define  DEVICE_TYPE_DUAL_CHANNEL	(1 << 8)
-#define  DEVICE_TYPE_HIGH_SPEED_LINK	(1 << 6)
-#define  DEVICE_TYPE_LVDS_SINGALING	(1 << 5)
-#define  DEVICE_TYPE_TMDS_DVI_SIGNALING	(1 << 4)
-#define  DEVICE_TYPE_VIDEO_SIGNALING	(1 << 3)
-#define  DEVICE_TYPE_DISPLAYPORT_OUTPUT	(1 << 2)
-#define  DEVICE_TYPE_DIGITAL_OUTPUT	(1 << 1)
-#define  DEVICE_TYPE_ANALOG_OUTPUT	(1 << 0)
-
-/*
- * Bits we care about when checking for DEVICE_TYPE_eDP
- * Depending on the system, the other bits may or may not
- * be set for eDP outputs.
- */
-#define DEVICE_TYPE_eDP_BITS \
-	(DEVICE_TYPE_INTERNAL_CONNECTOR | \
-	 DEVICE_TYPE_MIPI_OUTPUT | \
-	 DEVICE_TYPE_COMPOSITE_OUTPUT | \
-	 DEVICE_TYPE_DUAL_CHANNEL | \
-	 DEVICE_TYPE_LVDS_SINGALING | \
-	 DEVICE_TYPE_TMDS_DVI_SIGNALING | \
-	 DEVICE_TYPE_VIDEO_SIGNALING | \
-	 DEVICE_TYPE_DISPLAYPORT_OUTPUT | \
-	 DEVICE_TYPE_ANALOG_OUTPUT)
-
-/* define the DVO port for HDMI output type */
-#define		DVO_B		1
-#define		DVO_C		2
-#define		DVO_D		3
-
-/* Possible values for the "DVO Port" field for versions >= 155: */
-#define DVO_PORT_HDMIA	0
-#define DVO_PORT_HDMIB	1
-#define DVO_PORT_HDMIC	2
-#define DVO_PORT_HDMID	3
-#define DVO_PORT_LVDS	4
-#define DVO_PORT_TV	5
-#define DVO_PORT_CRT	6
-#define DVO_PORT_DPB	7
-#define DVO_PORT_DPC	8
-#define DVO_PORT_DPD	9
-#define DVO_PORT_DPA	10
-#define DVO_PORT_DPE	11
-#define DVO_PORT_HDMIE	12
-#define DVO_PORT_MIPIA	21
-#define DVO_PORT_MIPIB	22
-#define DVO_PORT_MIPIC	23
-#define DVO_PORT_MIPID	24
+/* MIPI Sequence Block definitions */
+enum mipi_seq {
+	MIPI_SEQ_END = 0,
+	MIPI_SEQ_ASSERT_RESET,
+	MIPI_SEQ_INIT_OTP,
+	MIPI_SEQ_DISPLAY_ON,
+	MIPI_SEQ_DISPLAY_OFF,
+	MIPI_SEQ_DEASSERT_RESET,
+	MIPI_SEQ_BACKLIGHT_ON,		/* sequence block v2+ */
+	MIPI_SEQ_BACKLIGHT_OFF,		/* sequence block v2+ */
+	MIPI_SEQ_TEAR_ON,		/* sequence block v2+ */
+	MIPI_SEQ_TEAR_OFF,		/* sequence block v3+ */
+	MIPI_SEQ_POWER_ON,		/* sequence block v3+ */
+	MIPI_SEQ_POWER_OFF,		/* sequence block v3+ */
+	MIPI_SEQ_MAX
+};
 
-/* Block 52 contains MIPI Panel info
- * 6 such enteries will there. Index into correct
- * entery is based on the panel_index in #40 LFP
- */
-#define MAX_MIPI_CONFIGURATIONS	6
+enum mipi_seq_element {
+	MIPI_SEQ_ELEM_END = 0,
+	MIPI_SEQ_ELEM_SEND_PKT,
+	MIPI_SEQ_ELEM_DELAY,
+	MIPI_SEQ_ELEM_GPIO,
+	MIPI_SEQ_ELEM_I2C,		/* sequence block v2+ */
+	MIPI_SEQ_ELEM_SPI,		/* sequence block v3+ */
+	MIPI_SEQ_ELEM_PMIC,		/* sequence block v3+ */
+	MIPI_SEQ_ELEM_MAX
+};
 
 #define MIPI_DSI_UNDEFINED_PANEL_ID	0
 #define MIPI_DSI_GENERIC_PANEL_ID	1
 
-/*
- * PMIC vs SoC Backlight support specified in pwm_blc
- * field in mipi_config block below.
-*/
-#define PPS_BLC_PMIC   0
-#define PPS_BLC_SOC    1
-
 struct mipi_config {
 	u16 panel_id;
 
@@ -821,6 +86,8 @@ struct mipi_config {
 	u32 video_transfer_mode:2;
 
 	u32 cabc_supported:1;
+#define PPS_BLC_PMIC   0
+#define PPS_BLC_SOC    1
 	u32 pwm_blc:1;
 
 	/* Bit 13:10 */
@@ -924,12 +191,7 @@ struct mipi_config {
 
 } __packed;
 
-/* Block 52 contains MIPI configuration block
- * 6 * bdb_mipi_config, followed by 6 pps data
- * block below
- *
- * all delays has a unit of 100us
- */
+/* all delays have a unit of 100us */
 struct mipi_pps_data {
 	u16 panel_on_delay;
 	u16 bl_enable_delay;
@@ -938,57 +200,4 @@ struct mipi_pps_data {
 	u16 panel_power_cycle_delay;
 } __packed;
 
-struct bdb_mipi_config {
-	struct mipi_config config[MAX_MIPI_CONFIGURATIONS];
-	struct mipi_pps_data pps[MAX_MIPI_CONFIGURATIONS];
-} __packed;
-
-/* Block 53 contains MIPI sequences as needed by the panel
- * for enabling it. This block can be variable in size and
- * can be maximum of 6 blocks
- */
-struct bdb_mipi_sequence {
-	u8 version;
-	u8 data[0];
-} __packed;
-
-/* MIPI Sequnece Block definitions */
-enum mipi_seq {
-	MIPI_SEQ_END = 0,
-	MIPI_SEQ_ASSERT_RESET,
-	MIPI_SEQ_INIT_OTP,
-	MIPI_SEQ_DISPLAY_ON,
-	MIPI_SEQ_DISPLAY_OFF,
-	MIPI_SEQ_DEASSERT_RESET,
-	MIPI_SEQ_BACKLIGHT_ON,		/* sequence block v2+ */
-	MIPI_SEQ_BACKLIGHT_OFF,		/* sequence block v2+ */
-	MIPI_SEQ_TEAR_ON,		/* sequence block v2+ */
-	MIPI_SEQ_TEAR_OFF,		/* sequence block v3+ */
-	MIPI_SEQ_POWER_ON,		/* sequence block v3+ */
-	MIPI_SEQ_POWER_OFF,		/* sequence block v3+ */
-	MIPI_SEQ_MAX
-};
-
-enum mipi_seq_element {
-	MIPI_SEQ_ELEM_END = 0,
-	MIPI_SEQ_ELEM_SEND_PKT,
-	MIPI_SEQ_ELEM_DELAY,
-	MIPI_SEQ_ELEM_GPIO,
-	MIPI_SEQ_ELEM_I2C,		/* sequence block v2+ */
-	MIPI_SEQ_ELEM_SPI,		/* sequence block v3+ */
-	MIPI_SEQ_ELEM_PMIC,		/* sequence block v3+ */
-	MIPI_SEQ_ELEM_MAX
-};
-
-enum mipi_gpio_pin_index {
-	MIPI_GPIO_UNDEFINED = 0,
-	MIPI_GPIO_PANEL_ENABLE,
-	MIPI_GPIO_BL_ENABLE,
-	MIPI_GPIO_PWM_ENABLE,
-	MIPI_GPIO_RESET_N,
-	MIPI_GPIO_PWR_DOWN_R,
-	MIPI_GPIO_STDBY_RST_N,
-	MIPI_GPIO_MAX
-};
-
 #endif /* _INTEL_BIOS_H_ */
diff --git a/drivers/gpu/drm/i915/intel_color.c b/drivers/gpu/drm/i915/intel_color.c
new file mode 100644
index 000000000000..1b3f97449395
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_color.c
@@ -0,0 +1,553 @@
+/*
+ * Copyright © 2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "intel_drv.h"
+
+#define CTM_COEFF_SIGN	(1ULL << 63)
+
+#define CTM_COEFF_1_0	(1ULL << 32)
+#define CTM_COEFF_2_0	(CTM_COEFF_1_0 << 1)
+#define CTM_COEFF_4_0	(CTM_COEFF_2_0 << 1)
+#define CTM_COEFF_8_0	(CTM_COEFF_4_0 << 1)
+#define CTM_COEFF_0_5	(CTM_COEFF_1_0 >> 1)
+#define CTM_COEFF_0_25	(CTM_COEFF_0_5 >> 1)
+#define CTM_COEFF_0_125	(CTM_COEFF_0_25 >> 1)
+
+#define CTM_COEFF_LIMITED_RANGE ((235ULL - 16ULL) * CTM_COEFF_1_0 / 255)
+
+#define CTM_COEFF_NEGATIVE(coeff)	(((coeff) & CTM_COEFF_SIGN) != 0)
+#define CTM_COEFF_ABS(coeff)		((coeff) & (CTM_COEFF_SIGN - 1))
+
+#define LEGACY_LUT_LENGTH		(sizeof(struct drm_color_lut) * 256)
+
+/*
+ * Extract the CSC coefficient from a CTM coefficient (in U32.32 fixed point
+ * format). This macro takes the coefficient we want transformed and the
+ * number of fractional bits.
+ *
+ * We only have a 9 bits precision window which slides depending on the value
+ * of the CTM coefficient and we write the value from bit 3. We also round the
+ * value.
+ */
+#define I9XX_CSC_COEFF_FP(coeff, fbits)	\
+	(clamp_val(((coeff) >> (32 - (fbits) - 3)) + 4, 0, 0xfff) & 0xff8)
+
+#define I9XX_CSC_COEFF_LIMITED_RANGE	\
+	I9XX_CSC_COEFF_FP(CTM_COEFF_LIMITED_RANGE, 9)
+#define I9XX_CSC_COEFF_1_0		\
+	((7 << 12) | I9XX_CSC_COEFF_FP(CTM_COEFF_1_0, 8))
+
+static bool crtc_state_is_legacy(struct drm_crtc_state *state)
+{
+	return !state->degamma_lut &&
+		!state->ctm &&
+		state->gamma_lut &&
+		state->gamma_lut->length == LEGACY_LUT_LENGTH;
+}
+
+/*
+ * When using limited range, multiply the matrix given by userspace by
+ * the matrix that we would use for the limited range. We do the
+ * multiplication in U2.30 format.
+ */
+static void ctm_mult_by_limited(uint64_t *result, int64_t *input)
+{
+	int i;
+
+	for (i = 0; i < 9; i++)
+		result[i] = 0;
+
+	for (i = 0; i < 3; i++) {
+		int64_t user_coeff = input[i * 3 + i];
+		uint64_t limited_coeff = CTM_COEFF_LIMITED_RANGE >> 2;
+		uint64_t abs_coeff = clamp_val(CTM_COEFF_ABS(user_coeff),
+					       0,
+					       CTM_COEFF_4_0 - 1) >> 2;
+
+		result[i * 3 + i] = (limited_coeff * abs_coeff) >> 27;
+		if (CTM_COEFF_NEGATIVE(user_coeff))
+			result[i * 3 + i] |= CTM_COEFF_SIGN;
+	}
+}
+
+/* Set up the pipe CSC unit. */
+static void i9xx_load_csc_matrix(struct drm_crtc_state *crtc_state)
+{
+	struct drm_crtc *crtc = crtc_state->crtc;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int i, pipe = intel_crtc->pipe;
+	uint16_t coeffs[9] = { 0, };
+
+	if (crtc_state->ctm) {
+		struct drm_color_ctm *ctm =
+			(struct drm_color_ctm *)crtc_state->ctm->data;
+		uint64_t input[9] = { 0, };
+
+		if (intel_crtc->config->limited_color_range) {
+			ctm_mult_by_limited(input, ctm->matrix);
+		} else {
+			for (i = 0; i < ARRAY_SIZE(input); i++)
+				input[i] = ctm->matrix[i];
+		}
+
+		/*
+		 * Convert fixed point S31.32 input to format supported by the
+		 * hardware.
+		 */
+		for (i = 0; i < ARRAY_SIZE(coeffs); i++) {
+			uint64_t abs_coeff = ((1ULL << 63) - 1) & input[i];
+
+			/*
+			 * Clamp input value to min/max supported by
+			 * hardware.
+			 */
+			abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_4_0 - 1);
+
+			/* sign bit */
+			if (CTM_COEFF_NEGATIVE(input[i]))
+				coeffs[i] |= 1 << 15;
+
+			if (abs_coeff < CTM_COEFF_0_125)
+				coeffs[i] |= (3 << 12) |
+					I9XX_CSC_COEFF_FP(abs_coeff, 12);
+			else if (abs_coeff < CTM_COEFF_0_25)
+				coeffs[i] |= (2 << 12) |
+					I9XX_CSC_COEFF_FP(abs_coeff, 11);
+			else if (abs_coeff < CTM_COEFF_0_5)
+				coeffs[i] |= (1 << 12) |
+					I9XX_CSC_COEFF_FP(abs_coeff, 10);
+			else if (abs_coeff < CTM_COEFF_1_0)
+				coeffs[i] |= I9XX_CSC_COEFF_FP(abs_coeff, 9);
+			else if (abs_coeff < CTM_COEFF_2_0)
+				coeffs[i] |= (7 << 12) |
+					I9XX_CSC_COEFF_FP(abs_coeff, 8);
+			else
+				coeffs[i] |= (6 << 12) |
+					I9XX_CSC_COEFF_FP(abs_coeff, 7);
+		}
+	} else {
+		/*
+		 * Load an identity matrix if no coefficients are provided.
+		 *
+		 * TODO: Check what kind of values actually come out of the
+		 * pipe with these coeff/postoff values and adjust to get the
+		 * best accuracy. Perhaps we even need to take the bpc value
+		 * into consideration.
+		 */
+		for (i = 0; i < 3; i++) {
+			if (intel_crtc->config->limited_color_range)
+				coeffs[i * 3 + i] =
+					I9XX_CSC_COEFF_LIMITED_RANGE;
+			else
+				coeffs[i * 3 + i] = I9XX_CSC_COEFF_1_0;
+		}
+	}
+
+	I915_WRITE(PIPE_CSC_COEFF_RY_GY(pipe), coeffs[0] << 16 | coeffs[1]);
+	I915_WRITE(PIPE_CSC_COEFF_BY(pipe), coeffs[2] << 16);
+
+	I915_WRITE(PIPE_CSC_COEFF_RU_GU(pipe), coeffs[3] << 16 | coeffs[4]);
+	I915_WRITE(PIPE_CSC_COEFF_BU(pipe), coeffs[5] << 16);
+
+	I915_WRITE(PIPE_CSC_COEFF_RV_GV(pipe), coeffs[6] << 16 | coeffs[7]);
+	I915_WRITE(PIPE_CSC_COEFF_BV(pipe), coeffs[8] << 16);
+
+	I915_WRITE(PIPE_CSC_PREOFF_HI(pipe), 0);
+	I915_WRITE(PIPE_CSC_PREOFF_ME(pipe), 0);
+	I915_WRITE(PIPE_CSC_PREOFF_LO(pipe), 0);
+
+	if (INTEL_INFO(dev)->gen > 6) {
+		uint16_t postoff = 0;
+
+		if (intel_crtc->config->limited_color_range)
+			postoff = (16 * (1 << 12) / 255) & 0x1fff;
+
+		I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), postoff);
+		I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), postoff);
+		I915_WRITE(PIPE_CSC_POSTOFF_LO(pipe), postoff);
+
+		I915_WRITE(PIPE_CSC_MODE(pipe), 0);
+	} else {
+		uint32_t mode = CSC_MODE_YUV_TO_RGB;
+
+		if (intel_crtc->config->limited_color_range)
+			mode |= CSC_BLACK_SCREEN_OFFSET;
+
+		I915_WRITE(PIPE_CSC_MODE(pipe), mode);
+	}
+}
+
+/*
+ * Set up the pipe CSC unit on CherryView.
+ */
+static void cherryview_load_csc_matrix(struct drm_crtc_state *state)
+{
+	struct drm_crtc *crtc = state->crtc;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = to_intel_crtc(crtc)->pipe;
+	uint32_t mode;
+
+	if (state->ctm) {
+		struct drm_color_ctm *ctm =
+			(struct drm_color_ctm *) state->ctm->data;
+		uint16_t coeffs[9] = { 0, };
+		int i;
+
+		for (i = 0; i < ARRAY_SIZE(coeffs); i++) {
+			uint64_t abs_coeff =
+				((1ULL << 63) - 1) & ctm->matrix[i];
+
+			/* Round coefficient. */
+			abs_coeff += 1 << (32 - 13);
+			/* Clamp to hardware limits. */
+			abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_8_0 - 1);
+
+			/* Write coefficients in S3.12 format. */
+			if (ctm->matrix[i] & (1ULL << 63))
+				coeffs[i] = 1 << 15;
+			coeffs[i] |= ((abs_coeff >> 32) & 7) << 12;
+			coeffs[i] |= (abs_coeff >> 20) & 0xfff;
+		}
+
+		I915_WRITE(CGM_PIPE_CSC_COEFF01(pipe),
+			   coeffs[1] << 16 | coeffs[0]);
+		I915_WRITE(CGM_PIPE_CSC_COEFF23(pipe),
+			   coeffs[3] << 16 | coeffs[2]);
+		I915_WRITE(CGM_PIPE_CSC_COEFF45(pipe),
+			   coeffs[5] << 16 | coeffs[4]);
+		I915_WRITE(CGM_PIPE_CSC_COEFF67(pipe),
+			   coeffs[7] << 16 | coeffs[6]);
+		I915_WRITE(CGM_PIPE_CSC_COEFF8(pipe), coeffs[8]);
+	}
+
+	mode = (state->ctm ? CGM_PIPE_MODE_CSC : 0);
+	if (!crtc_state_is_legacy(state)) {
+		mode |= (state->degamma_lut ? CGM_PIPE_MODE_DEGAMMA : 0) |
+			(state->gamma_lut ? CGM_PIPE_MODE_GAMMA : 0);
+	}
+	I915_WRITE(CGM_PIPE_MODE(pipe), mode);
+}
+
+void intel_color_set_csc(struct drm_crtc_state *crtc_state)
+{
+	struct drm_device *dev = crtc_state->crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (dev_priv->display.load_csc_matrix)
+		dev_priv->display.load_csc_matrix(crtc_state);
+}
+
+/* Loads the legacy palette/gamma unit for the CRTC. */
+static void i9xx_load_luts_internal(struct drm_crtc *crtc,
+				    struct drm_property_blob *blob)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	int i;
+
+	if (HAS_GMCH_DISPLAY(dev)) {
+		if (intel_crtc->config->has_dsi_encoder)
+			assert_dsi_pll_enabled(dev_priv);
+		else
+			assert_pll_enabled(dev_priv, pipe);
+	}
+
+	if (blob) {
+		struct drm_color_lut *lut = (struct drm_color_lut *) blob->data;
+		for (i = 0; i < 256; i++) {
+			uint32_t word =
+				(drm_color_lut_extract(lut[i].red, 8) << 16) |
+				(drm_color_lut_extract(lut[i].green, 8) << 8) |
+				drm_color_lut_extract(lut[i].blue, 8);
+
+			if (HAS_GMCH_DISPLAY(dev))
+				I915_WRITE(PALETTE(pipe, i), word);
+			else
+				I915_WRITE(LGC_PALETTE(pipe, i), word);
+		}
+	} else {
+		for (i = 0; i < 256; i++) {
+			uint32_t word = (i << 16) | (i << 8) | i;
+
+			if (HAS_GMCH_DISPLAY(dev))
+				I915_WRITE(PALETTE(pipe, i), word);
+			else
+				I915_WRITE(LGC_PALETTE(pipe, i), word);
+		}
+	}
+}
+
+static void i9xx_load_luts(struct drm_crtc_state *crtc_state)
+{
+	i9xx_load_luts_internal(crtc_state->crtc, crtc_state->gamma_lut);
+}
+
+/* Loads the legacy palette/gamma unit for the CRTC on Haswell. */
+static void haswell_load_luts(struct drm_crtc_state *crtc_state)
+{
+	struct drm_crtc *crtc = crtc_state->crtc;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *intel_crtc_state =
+		to_intel_crtc_state(crtc_state);
+	bool reenable_ips = false;
+
+	/*
+	 * Workaround : Do not read or write the pipe palette/gamma data while
+	 * GAMMA_MODE is configured for split gamma and IPS_CTL has IPS enabled.
+	 */
+	if (IS_HASWELL(dev) && intel_crtc->config->ips_enabled &&
+	    (intel_crtc_state->gamma_mode == GAMMA_MODE_MODE_SPLIT)) {
+		hsw_disable_ips(intel_crtc);
+		reenable_ips = true;
+	}
+
+	intel_crtc_state->gamma_mode = GAMMA_MODE_MODE_8BIT;
+	I915_WRITE(GAMMA_MODE(intel_crtc->pipe), GAMMA_MODE_MODE_8BIT);
+
+	i9xx_load_luts(crtc_state);
+
+	if (reenable_ips)
+		hsw_enable_ips(intel_crtc);
+}
+
+/* Loads the palette/gamma unit for the CRTC on Broadwell+. */
+static void broadwell_load_luts(struct drm_crtc_state *state)
+{
+	struct drm_crtc *crtc = state->crtc;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc_state *intel_state = to_intel_crtc_state(state);
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	uint32_t i, lut_size = INTEL_INFO(dev)->color.degamma_lut_size;
+
+	if (crtc_state_is_legacy(state)) {
+		haswell_load_luts(state);
+		return;
+	}
+
+	I915_WRITE(PREC_PAL_INDEX(pipe),
+		   PAL_PREC_SPLIT_MODE | PAL_PREC_AUTO_INCREMENT);
+
+	if (state->degamma_lut) {
+		struct drm_color_lut *lut =
+			(struct drm_color_lut *) state->degamma_lut->data;
+
+		for (i = 0; i < lut_size; i++) {
+			uint32_t word =
+			drm_color_lut_extract(lut[i].red, 10) << 20 |
+			drm_color_lut_extract(lut[i].green, 10) << 10 |
+			drm_color_lut_extract(lut[i].blue, 10);
+
+			I915_WRITE(PREC_PAL_DATA(pipe), word);
+		}
+	} else {
+		for (i = 0; i < lut_size; i++) {
+			uint32_t v = (i * ((1 << 10) - 1)) / (lut_size - 1);
+
+			I915_WRITE(PREC_PAL_DATA(pipe),
+				   (v << 20) | (v << 10) | v);
+		}
+	}
+
+	if (state->gamma_lut) {
+		struct drm_color_lut *lut =
+			(struct drm_color_lut *) state->gamma_lut->data;
+
+		for (i = 0; i < lut_size; i++) {
+			uint32_t word =
+			(drm_color_lut_extract(lut[i].red, 10) << 20) |
+			(drm_color_lut_extract(lut[i].green, 10) << 10) |
+			drm_color_lut_extract(lut[i].blue, 10);
+
+			I915_WRITE(PREC_PAL_DATA(pipe), word);
+		}
+
+		/* Program the max register to clamp values > 1.0. */
+		I915_WRITE(PREC_PAL_GC_MAX(pipe, 0),
+			   drm_color_lut_extract(lut[i].red, 16));
+		I915_WRITE(PREC_PAL_GC_MAX(pipe, 1),
+			   drm_color_lut_extract(lut[i].green, 16));
+		I915_WRITE(PREC_PAL_GC_MAX(pipe, 2),
+			   drm_color_lut_extract(lut[i].blue, 16));
+	} else {
+		for (i = 0; i < lut_size; i++) {
+			uint32_t v = (i * ((1 << 10) - 1)) / (lut_size - 1);
+
+			I915_WRITE(PREC_PAL_DATA(pipe),
+				   (v << 20) | (v << 10) | v);
+		}
+
+		I915_WRITE(PREC_PAL_GC_MAX(pipe, 0), (1 << 16) - 1);
+		I915_WRITE(PREC_PAL_GC_MAX(pipe, 1), (1 << 16) - 1);
+		I915_WRITE(PREC_PAL_GC_MAX(pipe, 2), (1 << 16) - 1);
+	}
+
+	intel_state->gamma_mode = GAMMA_MODE_MODE_SPLIT;
+	I915_WRITE(GAMMA_MODE(pipe), GAMMA_MODE_MODE_SPLIT);
+	POSTING_READ(GAMMA_MODE(pipe));
+
+	/*
+	 * Reset the index, otherwise it prevents the legacy palette to be
+	 * written properly.
+	 */
+	I915_WRITE(PREC_PAL_INDEX(pipe), 0);
+}
+
+/* Loads the palette/gamma unit for the CRTC on CherryView. */
+static void cherryview_load_luts(struct drm_crtc_state *state)
+{
+	struct drm_crtc *crtc = state->crtc;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	struct drm_color_lut *lut;
+	uint32_t i, lut_size;
+	uint32_t word0, word1;
+
+	if (crtc_state_is_legacy(state)) {
+		/* Turn off degamma/gamma on CGM block. */
+		I915_WRITE(CGM_PIPE_MODE(pipe),
+			   (state->ctm ? CGM_PIPE_MODE_CSC : 0));
+		i9xx_load_luts_internal(crtc, state->gamma_lut);
+		return;
+	}
+
+	if (state->degamma_lut) {
+		lut = (struct drm_color_lut *) state->degamma_lut->data;
+		lut_size = INTEL_INFO(dev)->color.degamma_lut_size;
+		for (i = 0; i < lut_size; i++) {
+			/* Write LUT in U0.14 format. */
+			word0 =
+			(drm_color_lut_extract(lut[i].green, 14) << 16) |
+			drm_color_lut_extract(lut[i].blue, 14);
+			word1 = drm_color_lut_extract(lut[i].red, 14);
+
+			I915_WRITE(CGM_PIPE_DEGAMMA(pipe, i, 0), word0);
+			I915_WRITE(CGM_PIPE_DEGAMMA(pipe, i, 1), word1);
+		}
+	}
+
+	if (state->gamma_lut) {
+		lut = (struct drm_color_lut *) state->gamma_lut->data;
+		lut_size = INTEL_INFO(dev)->color.gamma_lut_size;
+		for (i = 0; i < lut_size; i++) {
+			/* Write LUT in U0.10 format. */
+			word0 =
+			(drm_color_lut_extract(lut[i].green, 10) << 16) |
+			drm_color_lut_extract(lut[i].blue, 10);
+			word1 = drm_color_lut_extract(lut[i].red, 10);
+
+			I915_WRITE(CGM_PIPE_GAMMA(pipe, i, 0), word0);
+			I915_WRITE(CGM_PIPE_GAMMA(pipe, i, 1), word1);
+		}
+	}
+
+	I915_WRITE(CGM_PIPE_MODE(pipe),
+		   (state->ctm ? CGM_PIPE_MODE_CSC : 0) |
+		   (state->degamma_lut ? CGM_PIPE_MODE_DEGAMMA : 0) |
+		   (state->gamma_lut ? CGM_PIPE_MODE_GAMMA : 0));
+
+	/*
+	 * Also program a linear LUT in the legacy block (behind the
+	 * CGM block).
+	 */
+	i9xx_load_luts_internal(crtc, NULL);
+}
+
+void intel_color_load_luts(struct drm_crtc_state *crtc_state)
+{
+	struct drm_device *dev = crtc_state->crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->display.load_luts(crtc_state);
+}
+
+int intel_color_check(struct drm_crtc *crtc,
+		      struct drm_crtc_state *crtc_state)
+{
+	struct drm_device *dev = crtc->dev;
+	size_t gamma_length, degamma_length;
+
+	degamma_length = INTEL_INFO(dev)->color.degamma_lut_size *
+		sizeof(struct drm_color_lut);
+	gamma_length = INTEL_INFO(dev)->color.gamma_lut_size *
+		sizeof(struct drm_color_lut);
+
+	/*
+	 * We allow both degamma & gamma luts at the right size or
+	 * NULL.
+	 */
+	if ((!crtc_state->degamma_lut ||
+	     crtc_state->degamma_lut->length == degamma_length) &&
+	    (!crtc_state->gamma_lut ||
+	     crtc_state->gamma_lut->length == gamma_length))
+		return 0;
+
+	/*
+	 * We also allow no degamma lut and a gamma lut at the legacy
+	 * size (256 entries).
+	 */
+	if (!crtc_state->degamma_lut &&
+	    crtc_state->gamma_lut &&
+	    crtc_state->gamma_lut->length == LEGACY_LUT_LENGTH)
+		return 0;
+
+	return -EINVAL;
+}
+
+void intel_color_init(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	drm_mode_crtc_set_gamma_size(crtc, 256);
+
+	if (IS_CHERRYVIEW(dev)) {
+		dev_priv->display.load_csc_matrix = cherryview_load_csc_matrix;
+		dev_priv->display.load_luts = cherryview_load_luts;
+	} else if (IS_HASWELL(dev)) {
+		dev_priv->display.load_csc_matrix = i9xx_load_csc_matrix;
+		dev_priv->display.load_luts = haswell_load_luts;
+	} else if (IS_BROADWELL(dev) || IS_SKYLAKE(dev) ||
+		   IS_BROXTON(dev) || IS_KABYLAKE(dev)) {
+		dev_priv->display.load_csc_matrix = i9xx_load_csc_matrix;
+		dev_priv->display.load_luts = broadwell_load_luts;
+	} else {
+		dev_priv->display.load_luts = i9xx_load_luts;
+	}
+
+	/* Enable color management support when we have degamma & gamma LUTs. */
+	if (INTEL_INFO(dev)->color.degamma_lut_size != 0 &&
+	    INTEL_INFO(dev)->color.gamma_lut_size != 0)
+		drm_helper_crtc_enable_color_mgmt(crtc,
+					INTEL_INFO(dev)->color.degamma_lut_size,
+					INTEL_INFO(dev)->color.gamma_lut_size);
+}
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 0364292367b1..3fbb6fc66451 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -120,22 +120,16 @@ static unsigned int intel_crt_get_flags(struct intel_encoder *encoder)
 static void intel_crt_get_config(struct intel_encoder *encoder,
 				 struct intel_crtc_state *pipe_config)
 {
-	struct drm_device *dev = encoder->base.dev;
-	int dotclock;
-
 	pipe_config->base.adjusted_mode.flags |= intel_crt_get_flags(encoder);
 
-	dotclock = pipe_config->port_clock;
-
-	if (HAS_PCH_SPLIT(dev))
-		ironlake_check_encoder_dotclock(pipe_config, dotclock);
-
-	pipe_config->base.adjusted_mode.crtc_clock = dotclock;
+	pipe_config->base.adjusted_mode.crtc_clock = pipe_config->port_clock;
 }
 
 static void hsw_crt_get_config(struct intel_encoder *encoder,
 			       struct intel_crtc_state *pipe_config)
 {
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+
 	intel_ddi_get_config(encoder, pipe_config);
 
 	pipe_config->base.adjusted_mode.flags &= ~(DRM_MODE_FLAG_PHSYNC |
@@ -143,6 +137,8 @@ static void hsw_crt_get_config(struct intel_encoder *encoder,
 					      DRM_MODE_FLAG_PVSYNC |
 					      DRM_MODE_FLAG_NVSYNC);
 	pipe_config->base.adjusted_mode.flags |= intel_crt_get_flags(encoder);
+
+	pipe_config->base.adjusted_mode.crtc_clock = lpt_get_iclkip(dev_priv);
 }
 
 /* Note: The caller is required to filter out dpms modes not supported by the
@@ -222,18 +218,26 @@ intel_crt_mode_valid(struct drm_connector *connector,
 {
 	struct drm_device *dev = connector->dev;
 	int max_dotclk = to_i915(dev)->max_dotclk_freq;
+	int max_clock;
 
-	int max_clock = 0;
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return MODE_NO_DBLESCAN;
 
 	if (mode->clock < 25000)
 		return MODE_CLOCK_LOW;
 
-	if (IS_GEN2(dev))
-		max_clock = 350000;
-	else
+	if (HAS_PCH_LPT(dev))
+		max_clock = 180000;
+	else if (IS_VALLEYVIEW(dev))
+		/*
+		 * 270 MHz due to current DPLL limits,
+		 * DAC limit supposedly 355 MHz.
+		 */
+		max_clock = 270000;
+	else if (IS_GEN3(dev) || IS_GEN4(dev))
 		max_clock = 400000;
+	else
+		max_clock = 350000;
 	if (mode->clock > max_clock)
 		return MODE_CLOCK_HIGH;
 
@@ -267,15 +271,9 @@ static bool intel_crt_compute_config(struct intel_encoder *encoder,
 	}
 
 	/* FDI must always be 2.7 GHz */
-	if (HAS_DDI(dev)) {
-		pipe_config->ddi_pll_sel = PORT_CLK_SEL_SPLL;
+	if (HAS_DDI(dev))
 		pipe_config->port_clock = 135000 * 2;
 
-		pipe_config->dpll_hw_state.wrpll = 0;
-		pipe_config->dpll_hw_state.spll =
-			SPLL_PLL_ENABLE | SPLL_PLL_FREQ_1350MHz | SPLL_PLL_SSC;
-	}
-
 	return true;
 }
 
@@ -658,6 +656,8 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 		else if (INTEL_INFO(dev)->gen < 4)
 			status = intel_crt_load_detect(crt,
 				to_intel_crtc(connector->state->crtc)->pipe);
+		else if (i915.load_detect_test)
+			status = connector_status_disconnected;
 		else
 			status = connector_status_unknown;
 		intel_release_load_detect_pipe(connector, &tmp, &ctx);
diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index 902054efb902..a34c23eceba0 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -50,6 +50,7 @@ MODULE_FIRMWARE(I915_CSR_SKL);
 MODULE_FIRMWARE(I915_CSR_BXT);
 
 #define SKL_CSR_VERSION_REQUIRED	CSR_VERSION(1, 23)
+#define BXT_CSR_VERSION_REQUIRED	CSR_VERSION(1, 7)
 
 #define CSR_MAX_FW_SIZE			0x2FFF
 #define CSR_DEFAULT_FW_OFFSET		0xFFFFFFFF
@@ -188,28 +189,49 @@ static const struct stepping_info bxt_stepping_info[] = {
 	{'B', '0'}, {'B', '1'}, {'B', '2'}
 };
 
-static const struct stepping_info *intel_get_stepping_info(struct drm_device *dev)
+static const struct stepping_info no_stepping_info = { '*', '*' };
+
+static const struct stepping_info *
+intel_get_stepping_info(struct drm_i915_private *dev_priv)
 {
 	const struct stepping_info *si;
 	unsigned int size;
 
-	if (IS_KABYLAKE(dev)) {
+	if (IS_KABYLAKE(dev_priv)) {
 		size = ARRAY_SIZE(kbl_stepping_info);
 		si = kbl_stepping_info;
-	} else if (IS_SKYLAKE(dev)) {
+	} else if (IS_SKYLAKE(dev_priv)) {
 		size = ARRAY_SIZE(skl_stepping_info);
 		si = skl_stepping_info;
-	} else if (IS_BROXTON(dev)) {
+	} else if (IS_BROXTON(dev_priv)) {
 		size = ARRAY_SIZE(bxt_stepping_info);
 		si = bxt_stepping_info;
 	} else {
-		return NULL;
+		size = 0;
 	}
 
-	if (INTEL_REVID(dev) < size)
-		return si + INTEL_REVID(dev);
+	if (INTEL_REVID(dev_priv) < size)
+		return si + INTEL_REVID(dev_priv);
 
-	return NULL;
+	return &no_stepping_info;
+}
+
+static void gen9_set_dc_state_debugmask(struct drm_i915_private *dev_priv)
+{
+	uint32_t val, mask;
+
+	mask = DC_STATE_DEBUG_MASK_MEMORY_UP;
+
+	if (IS_BROXTON(dev_priv))
+		mask |= DC_STATE_DEBUG_MASK_CORES;
+
+	/* The below bit doesn't need to be cleared ever afterwards */
+	val = I915_READ(DC_STATE_DEBUG);
+	if ((val & mask) != mask) {
+		val |= mask;
+		I915_WRITE(DC_STATE_DEBUG, val);
+		POSTING_READ(DC_STATE_DEBUG);
+	}
 }
 
 /**
@@ -220,19 +242,19 @@ static const struct stepping_info *intel_get_stepping_info(struct drm_device *de
  * Everytime display comes back from low power state this function is called to
  * copy the firmware from internal memory to registers.
  */
-bool intel_csr_load_program(struct drm_i915_private *dev_priv)
+void intel_csr_load_program(struct drm_i915_private *dev_priv)
 {
 	u32 *payload = dev_priv->csr.dmc_payload;
 	uint32_t i, fw_size;
 
 	if (!IS_GEN9(dev_priv)) {
 		DRM_ERROR("No CSR support available for this platform\n");
-		return false;
+		return;
 	}
 
 	if (!dev_priv->csr.dmc_payload) {
 		DRM_ERROR("Tried to program CSR with empty payload\n");
-		return false;
+		return;
 	}
 
 	fw_size = dev_priv->csr.dmc_fw_size;
@@ -246,34 +268,25 @@ bool intel_csr_load_program(struct drm_i915_private *dev_priv)
 
 	dev_priv->csr.dc_state = 0;
 
-	return true;
+	gen9_set_dc_state_debugmask(dev_priv);
 }
 
 static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv,
 			      const struct firmware *fw)
 {
-	struct drm_device *dev = dev_priv->dev;
 	struct intel_css_header *css_header;
 	struct intel_package_header *package_header;
 	struct intel_dmc_header *dmc_header;
 	struct intel_csr *csr = &dev_priv->csr;
-	const struct stepping_info *stepping_info = intel_get_stepping_info(dev);
-	char stepping, substepping;
+	const struct stepping_info *si = intel_get_stepping_info(dev_priv);
 	uint32_t dmc_offset = CSR_DEFAULT_FW_OFFSET, readcount = 0, nbytes;
 	uint32_t i;
 	uint32_t *dmc_payload;
+	uint32_t required_min_version;
 
 	if (!fw)
 		return NULL;
 
-	if (!stepping_info) {
-		DRM_ERROR("Unknown stepping info, firmware loading failed\n");
-		return NULL;
-	}
-
-	stepping = stepping_info->stepping;
-	substepping = stepping_info->substepping;
-
 	/* Extract CSS Header information*/
 	css_header = (struct intel_css_header *)fw->data;
 	if (sizeof(struct intel_css_header) !=
@@ -285,15 +298,23 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv,
 
 	csr->version = css_header->version;
 
-	if ((IS_SKYLAKE(dev) || IS_KABYLAKE(dev)) &&
-	    csr->version < SKL_CSR_VERSION_REQUIRED) {
-		DRM_INFO("Refusing to load old Skylake DMC firmware v%u.%u,"
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
+		required_min_version = SKL_CSR_VERSION_REQUIRED;
+	} else if (IS_BROXTON(dev_priv)) {
+		required_min_version = BXT_CSR_VERSION_REQUIRED;
+	} else {
+		MISSING_CASE(INTEL_REVID(dev_priv));
+		required_min_version = 0;
+	}
+
+	if (csr->version < required_min_version) {
+		DRM_INFO("Refusing to load old DMC firmware v%u.%u,"
 			 " please upgrade to v%u.%u or later"
 			   " [" FIRMWARE_URL "].\n",
 			 CSR_VERSION_MAJOR(csr->version),
 			 CSR_VERSION_MINOR(csr->version),
-			 CSR_VERSION_MAJOR(SKL_CSR_VERSION_REQUIRED),
-			 CSR_VERSION_MINOR(SKL_CSR_VERSION_REQUIRED));
+			 CSR_VERSION_MAJOR(required_min_version),
+			 CSR_VERSION_MINOR(required_min_version));
 		return NULL;
 	}
 
@@ -313,11 +334,11 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv,
 	/* Search for dmc_offset to find firware binary. */
 	for (i = 0; i < package_header->num_entries; i++) {
 		if (package_header->fw_info[i].substepping == '*' &&
-		    stepping == package_header->fw_info[i].stepping) {
+		    si->stepping == package_header->fw_info[i].stepping) {
 			dmc_offset = package_header->fw_info[i].offset;
 			break;
-		} else if (stepping == package_header->fw_info[i].stepping &&
-			substepping == package_header->fw_info[i].substepping) {
+		} else if (si->stepping == package_header->fw_info[i].stepping &&
+			   si->substepping == package_header->fw_info[i].substepping) {
 			dmc_offset = package_header->fw_info[i].offset;
 			break;
 		} else if (package_header->fw_info[i].stepping == '*' &&
@@ -325,7 +346,8 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv,
 			dmc_offset = package_header->fw_info[i].offset;
 	}
 	if (dmc_offset == CSR_DEFAULT_FW_OFFSET) {
-		DRM_ERROR("Firmware not supported for %c stepping\n", stepping);
+		DRM_ERROR("Firmware not supported for %c stepping\n",
+			  si->stepping);
 		return NULL;
 	}
 	readcount += dmc_offset;
@@ -371,9 +393,7 @@ static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv,
 		return NULL;
 	}
 
-	memcpy(dmc_payload, &fw->data[readcount], nbytes);
-
-	return dmc_payload;
+	return memcpy(dmc_payload, &fw->data[readcount], nbytes);
 }
 
 static void csr_load_work_fn(struct work_struct *work)
@@ -388,18 +408,12 @@ static void csr_load_work_fn(struct work_struct *work)
 
 	ret = request_firmware(&fw, dev_priv->csr.fw_path,
 			       &dev_priv->dev->pdev->dev);
-	if (!fw)
-		goto out;
-
-	dev_priv->csr.dmc_payload = parse_csr_fw(dev_priv, fw);
-	if (!dev_priv->csr.dmc_payload)
-		goto out;
+	if (fw)
+		dev_priv->csr.dmc_payload = parse_csr_fw(dev_priv, fw);
 
-	/* load csr program during system boot, as needed for DC states */
-	intel_csr_load_program(dev_priv);
-
-out:
 	if (dev_priv->csr.dmc_payload) {
+		intel_csr_load_program(dev_priv);
+
 		intel_display_power_put(dev_priv, POWER_DOMAIN_INIT);
 
 		DRM_INFO("Finished loading %s (v%u.%u)\n",
@@ -453,10 +467,50 @@ void intel_csr_ucode_init(struct drm_i915_private *dev_priv)
 }
 
 /**
+ * intel_csr_ucode_suspend() - prepare CSR firmware before system suspend
+ * @dev_priv: i915 drm device
+ *
+ * Prepare the DMC firmware before entering system suspend. This includes
+ * flushing pending work items and releasing any resources acquired during
+ * init.
+ */
+void intel_csr_ucode_suspend(struct drm_i915_private *dev_priv)
+{
+	if (!HAS_CSR(dev_priv))
+		return;
+
+	flush_work(&dev_priv->csr.work);
+
+	/* Drop the reference held in case DMC isn't loaded. */
+	if (!dev_priv->csr.dmc_payload)
+		intel_display_power_put(dev_priv, POWER_DOMAIN_INIT);
+}
+
+/**
+ * intel_csr_ucode_resume() - init CSR firmware during system resume
+ * @dev_priv: i915 drm device
+ *
+ * Reinitialize the DMC firmware during system resume, reacquiring any
+ * resources released in intel_csr_ucode_suspend().
+ */
+void intel_csr_ucode_resume(struct drm_i915_private *dev_priv)
+{
+	if (!HAS_CSR(dev_priv))
+		return;
+
+	/*
+	 * Reacquire the reference to keep RPM disabled in case DMC isn't
+	 * loaded.
+	 */
+	if (!dev_priv->csr.dmc_payload)
+		intel_display_power_get(dev_priv, POWER_DOMAIN_INIT);
+}
+
+/**
  * intel_csr_ucode_fini() - unload the CSR firmware.
  * @dev_priv: i915 drm device.
  *
- * Firmmware unloading includes freeing the internal momory and reset the
+ * Firmmware unloading includes freeing the internal memory and reset the
  * firmware loading status.
  */
 void intel_csr_ucode_fini(struct drm_i915_private *dev_priv)
@@ -464,7 +518,7 @@ void intel_csr_ucode_fini(struct drm_i915_private *dev_priv)
 	if (!HAS_CSR(dev_priv))
 		return;
 
-	flush_work(&dev_priv->csr.work);
+	intel_csr_ucode_suspend(dev_priv);
 
 	kfree(dev_priv->csr.dmc_payload);
 }
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 96ffcc541e17..3fac04602a25 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -315,6 +315,9 @@ static void ddi_get_encoder_port(struct intel_encoder *intel_encoder,
 		*dig_port = enc_to_mst(encoder)->primary;
 		*port = (*dig_port)->port;
 		break;
+	default:
+		WARN(1, "Invalid DDI encoder type %d\n", intel_encoder->type);
+		/* fallthrough and treat as unknown */
 	case INTEL_OUTPUT_DISPLAYPORT:
 	case INTEL_OUTPUT_EDP:
 	case INTEL_OUTPUT_HDMI:
@@ -326,9 +329,6 @@ static void ddi_get_encoder_port(struct intel_encoder *intel_encoder,
 		*dig_port = NULL;
 		*port = PORT_E;
 		break;
-	default:
-		WARN(1, "Invalid DDI encoder type %d\n", intel_encoder->type);
-		break;
 	}
 }
 
@@ -360,7 +360,7 @@ skl_get_buf_trans_dp(struct drm_i915_private *dev_priv, int *n_entries)
 static const struct ddi_buf_trans *
 skl_get_buf_trans_edp(struct drm_i915_private *dev_priv, int *n_entries)
 {
-	if (dev_priv->edp_low_vswing) {
+	if (dev_priv->vbt.edp.low_vswing) {
 		if (IS_SKL_ULX(dev_priv) || IS_KBL_ULX(dev_priv)) {
 			*n_entries = ARRAY_SIZE(skl_y_ddi_translations_edp);
 			return skl_y_ddi_translations_edp;
@@ -444,7 +444,7 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 		ddi_translations_fdi = bdw_ddi_translations_fdi;
 		ddi_translations_dp = bdw_ddi_translations_dp;
 
-		if (dev_priv->edp_low_vswing) {
+		if (dev_priv->vbt.edp.low_vswing) {
 			ddi_translations_edp = bdw_ddi_translations_edp;
 			n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_edp);
 		} else {
@@ -637,6 +637,10 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 			break;
 		}
 
+		rx_ctl_val &= ~FDI_RX_ENABLE;
+		I915_WRITE(FDI_RX_CTL(PIPE_A), rx_ctl_val);
+		POSTING_READ(FDI_RX_CTL(PIPE_A));
+
 		temp = I915_READ(DDI_BUF_CTL(PORT_E));
 		temp &= ~DDI_BUF_CTL_ENABLE;
 		I915_WRITE(DDI_BUF_CTL(PORT_E), temp);
@@ -651,10 +655,6 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 
 		intel_wait_ddi_buf_idle(dev_priv, PORT_E);
 
-		rx_ctl_val &= ~FDI_RX_ENABLE;
-		I915_WRITE(FDI_RX_CTL(PIPE_A), rx_ctl_val);
-		POSTING_READ(FDI_RX_CTL(PIPE_A));
-
 		/* Reset FDI_RX_MISC pwrdn lanes */
 		temp = I915_READ(FDI_RX_MISC(PIPE_A));
 		temp &= ~(FDI_RX_PWRDN_LANE1_MASK | FDI_RX_PWRDN_LANE0_MASK);
@@ -732,160 +732,6 @@ intel_ddi_get_crtc_new_encoder(struct intel_crtc_state *crtc_state)
 }
 
 #define LC_FREQ 2700
-#define LC_FREQ_2K U64_C(LC_FREQ * 2000)
-
-#define P_MIN 2
-#define P_MAX 64
-#define P_INC 2
-
-/* Constraints for PLL good behavior */
-#define REF_MIN 48
-#define REF_MAX 400
-#define VCO_MIN 2400
-#define VCO_MAX 4800
-
-#define abs_diff(a, b) ({			\
-	typeof(a) __a = (a);			\
-	typeof(b) __b = (b);			\
-	(void) (&__a == &__b);			\
-	__a > __b ? (__a - __b) : (__b - __a); })
-
-struct hsw_wrpll_rnp {
-	unsigned p, n2, r2;
-};
-
-static unsigned hsw_wrpll_get_budget_for_freq(int clock)
-{
-	unsigned budget;
-
-	switch (clock) {
-	case 25175000:
-	case 25200000:
-	case 27000000:
-	case 27027000:
-	case 37762500:
-	case 37800000:
-	case 40500000:
-	case 40541000:
-	case 54000000:
-	case 54054000:
-	case 59341000:
-	case 59400000:
-	case 72000000:
-	case 74176000:
-	case 74250000:
-	case 81000000:
-	case 81081000:
-	case 89012000:
-	case 89100000:
-	case 108000000:
-	case 108108000:
-	case 111264000:
-	case 111375000:
-	case 148352000:
-	case 148500000:
-	case 162000000:
-	case 162162000:
-	case 222525000:
-	case 222750000:
-	case 296703000:
-	case 297000000:
-		budget = 0;
-		break;
-	case 233500000:
-	case 245250000:
-	case 247750000:
-	case 253250000:
-	case 298000000:
-		budget = 1500;
-		break;
-	case 169128000:
-	case 169500000:
-	case 179500000:
-	case 202000000:
-		budget = 2000;
-		break;
-	case 256250000:
-	case 262500000:
-	case 270000000:
-	case 272500000:
-	case 273750000:
-	case 280750000:
-	case 281250000:
-	case 286000000:
-	case 291750000:
-		budget = 4000;
-		break;
-	case 267250000:
-	case 268500000:
-		budget = 5000;
-		break;
-	default:
-		budget = 1000;
-		break;
-	}
-
-	return budget;
-}
-
-static void hsw_wrpll_update_rnp(uint64_t freq2k, unsigned budget,
-				 unsigned r2, unsigned n2, unsigned p,
-				 struct hsw_wrpll_rnp *best)
-{
-	uint64_t a, b, c, d, diff, diff_best;
-
-	/* No best (r,n,p) yet */
-	if (best->p == 0) {
-		best->p = p;
-		best->n2 = n2;
-		best->r2 = r2;
-		return;
-	}
-
-	/*
-	 * Output clock is (LC_FREQ_2K / 2000) * N / (P * R), which compares to
-	 * freq2k.
-	 *
-	 * delta = 1e6 *
-	 *	   abs(freq2k - (LC_FREQ_2K * n2/(p * r2))) /
-	 *	   freq2k;
-	 *
-	 * and we would like delta <= budget.
-	 *
-	 * If the discrepancy is above the PPM-based budget, always prefer to
-	 * improve upon the previous solution.  However, if you're within the
-	 * budget, try to maximize Ref * VCO, that is N / (P * R^2).
-	 */
-	a = freq2k * budget * p * r2;
-	b = freq2k * budget * best->p * best->r2;
-	diff = abs_diff(freq2k * p * r2, LC_FREQ_2K * n2);
-	diff_best = abs_diff(freq2k * best->p * best->r2,
-			     LC_FREQ_2K * best->n2);
-	c = 1000000 * diff;
-	d = 1000000 * diff_best;
-
-	if (a < c && b < d) {
-		/* If both are above the budget, pick the closer */
-		if (best->p * best->r2 * diff < p * r2 * diff_best) {
-			best->p = p;
-			best->n2 = n2;
-			best->r2 = r2;
-		}
-	} else if (a >= c && b < d) {
-		/* If A is below the threshold but B is above it?  Update. */
-		best->p = p;
-		best->n2 = n2;
-		best->r2 = r2;
-	} else if (a >= c && b >= d) {
-		/* Both are below the limit, so pick the higher n2/(r2*r2) */
-		if (n2 * best->r2 * best->r2 > best->n2 * r2 * r2) {
-			best->p = p;
-			best->n2 = n2;
-			best->r2 = r2;
-		}
-	}
-	/* Otherwise a < c && b >= d, do nothing */
-}
 
 static int hsw_ddi_calc_wrpll_link(struct drm_i915_private *dev_priv,
 				   i915_reg_t reg)
@@ -1147,363 +993,20 @@ void intel_ddi_clock_get(struct intel_encoder *encoder,
 		bxt_ddi_clock_get(encoder, pipe_config);
 }
 
-static void
-hsw_ddi_calculate_wrpll(int clock /* in Hz */,
-			unsigned *r2_out, unsigned *n2_out, unsigned *p_out)
-{
-	uint64_t freq2k;
-	unsigned p, n2, r2;
-	struct hsw_wrpll_rnp best = { 0, 0, 0 };
-	unsigned budget;
-
-	freq2k = clock / 100;
-
-	budget = hsw_wrpll_get_budget_for_freq(clock);
-
-	/* Special case handling for 540 pixel clock: bypass WR PLL entirely
-	 * and directly pass the LC PLL to it. */
-	if (freq2k == 5400000) {
-		*n2_out = 2;
-		*p_out = 1;
-		*r2_out = 2;
-		return;
-	}
-
-	/*
-	 * Ref = LC_FREQ / R, where Ref is the actual reference input seen by
-	 * the WR PLL.
-	 *
-	 * We want R so that REF_MIN <= Ref <= REF_MAX.
-	 * Injecting R2 = 2 * R gives:
-	 *   REF_MAX * r2 > LC_FREQ * 2 and
-	 *   REF_MIN * r2 < LC_FREQ * 2
-	 *
-	 * Which means the desired boundaries for r2 are:
-	 *  LC_FREQ * 2 / REF_MAX < r2 < LC_FREQ * 2 / REF_MIN
-	 *
-	 */
-	for (r2 = LC_FREQ * 2 / REF_MAX + 1;
-	     r2 <= LC_FREQ * 2 / REF_MIN;
-	     r2++) {
-
-		/*
-		 * VCO = N * Ref, that is: VCO = N * LC_FREQ / R
-		 *
-		 * Once again we want VCO_MIN <= VCO <= VCO_MAX.
-		 * Injecting R2 = 2 * R and N2 = 2 * N, we get:
-		 *   VCO_MAX * r2 > n2 * LC_FREQ and
-		 *   VCO_MIN * r2 < n2 * LC_FREQ)
-		 *
-		 * Which means the desired boundaries for n2 are:
-		 * VCO_MIN * r2 / LC_FREQ < n2 < VCO_MAX * r2 / LC_FREQ
-		 */
-		for (n2 = VCO_MIN * r2 / LC_FREQ + 1;
-		     n2 <= VCO_MAX * r2 / LC_FREQ;
-		     n2++) {
-
-			for (p = P_MIN; p <= P_MAX; p += P_INC)
-				hsw_wrpll_update_rnp(freq2k, budget,
-						     r2, n2, p, &best);
-		}
-	}
-
-	*n2_out = best.n2;
-	*p_out = best.p;
-	*r2_out = best.r2;
-}
-
 static bool
 hsw_ddi_pll_select(struct intel_crtc *intel_crtc,
 		   struct intel_crtc_state *crtc_state,
 		   struct intel_encoder *intel_encoder)
 {
-	int clock = crtc_state->port_clock;
-
-	if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
-		struct intel_shared_dpll *pll;
-		uint32_t val;
-		unsigned p, n2, r2;
-
-		hsw_ddi_calculate_wrpll(clock * 1000, &r2, &n2, &p);
-
-		val = WRPLL_PLL_ENABLE | WRPLL_PLL_LCPLL |
-		      WRPLL_DIVIDER_REFERENCE(r2) | WRPLL_DIVIDER_FEEDBACK(n2) |
-		      WRPLL_DIVIDER_POST(p);
-
-		memset(&crtc_state->dpll_hw_state, 0,
-		       sizeof(crtc_state->dpll_hw_state));
-
-		crtc_state->dpll_hw_state.wrpll = val;
-
-		pll = intel_get_shared_dpll(intel_crtc, crtc_state);
-		if (pll == NULL) {
-			DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
-					 pipe_name(intel_crtc->pipe));
-			return false;
-		}
-
-		crtc_state->ddi_pll_sel = PORT_CLK_SEL_WRPLL(pll->id);
-	} else if (crtc_state->ddi_pll_sel == PORT_CLK_SEL_SPLL) {
-		struct drm_atomic_state *state = crtc_state->base.state;
-		struct intel_shared_dpll_config *spll =
-			&intel_atomic_get_shared_dpll_state(state)[DPLL_ID_SPLL];
-
-		if (spll->crtc_mask &&
-		    WARN_ON(spll->hw_state.spll != crtc_state->dpll_hw_state.spll))
-			return false;
-
-		crtc_state->shared_dpll = DPLL_ID_SPLL;
-		spll->hw_state.spll = crtc_state->dpll_hw_state.spll;
-		spll->crtc_mask |= 1 << intel_crtc->pipe;
-	}
-
-	return true;
-}
-
-struct skl_wrpll_context {
-	uint64_t min_deviation;		/* current minimal deviation */
-	uint64_t central_freq;		/* chosen central freq */
-	uint64_t dco_freq;		/* chosen dco freq */
-	unsigned int p;			/* chosen divider */
-};
-
-static void skl_wrpll_context_init(struct skl_wrpll_context *ctx)
-{
-	memset(ctx, 0, sizeof(*ctx));
-
-	ctx->min_deviation = U64_MAX;
-}
-
-/* DCO freq must be within +1%/-6%  of the DCO central freq */
-#define SKL_DCO_MAX_PDEVIATION	100
-#define SKL_DCO_MAX_NDEVIATION	600
-
-static void skl_wrpll_try_divider(struct skl_wrpll_context *ctx,
-				  uint64_t central_freq,
-				  uint64_t dco_freq,
-				  unsigned int divider)
-{
-	uint64_t deviation;
-
-	deviation = div64_u64(10000 * abs_diff(dco_freq, central_freq),
-			      central_freq);
-
-	/* positive deviation */
-	if (dco_freq >= central_freq) {
-		if (deviation < SKL_DCO_MAX_PDEVIATION &&
-		    deviation < ctx->min_deviation) {
-			ctx->min_deviation = deviation;
-			ctx->central_freq = central_freq;
-			ctx->dco_freq = dco_freq;
-			ctx->p = divider;
-		}
-	/* negative deviation */
-	} else if (deviation < SKL_DCO_MAX_NDEVIATION &&
-		   deviation < ctx->min_deviation) {
-		ctx->min_deviation = deviation;
-		ctx->central_freq = central_freq;
-		ctx->dco_freq = dco_freq;
-		ctx->p = divider;
-	}
-}
-
-static void skl_wrpll_get_multipliers(unsigned int p,
-				      unsigned int *p0 /* out */,
-				      unsigned int *p1 /* out */,
-				      unsigned int *p2 /* out */)
-{
-	/* even dividers */
-	if (p % 2 == 0) {
-		unsigned int half = p / 2;
-
-		if (half == 1 || half == 2 || half == 3 || half == 5) {
-			*p0 = 2;
-			*p1 = 1;
-			*p2 = half;
-		} else if (half % 2 == 0) {
-			*p0 = 2;
-			*p1 = half / 2;
-			*p2 = 2;
-		} else if (half % 3 == 0) {
-			*p0 = 3;
-			*p1 = half / 3;
-			*p2 = 2;
-		} else if (half % 7 == 0) {
-			*p0 = 7;
-			*p1 = half / 7;
-			*p2 = 2;
-		}
-	} else if (p == 3 || p == 9) {  /* 3, 5, 7, 9, 15, 21, 35 */
-		*p0 = 3;
-		*p1 = 1;
-		*p2 = p / 3;
-	} else if (p == 5 || p == 7) {
-		*p0 = p;
-		*p1 = 1;
-		*p2 = 1;
-	} else if (p == 15) {
-		*p0 = 3;
-		*p1 = 1;
-		*p2 = 5;
-	} else if (p == 21) {
-		*p0 = 7;
-		*p1 = 1;
-		*p2 = 3;
-	} else if (p == 35) {
-		*p0 = 7;
-		*p1 = 1;
-		*p2 = 5;
-	}
-}
-
-struct skl_wrpll_params {
-	uint32_t        dco_fraction;
-	uint32_t        dco_integer;
-	uint32_t        qdiv_ratio;
-	uint32_t        qdiv_mode;
-	uint32_t        kdiv;
-	uint32_t        pdiv;
-	uint32_t        central_freq;
-};
-
-static void skl_wrpll_params_populate(struct skl_wrpll_params *params,
-				      uint64_t afe_clock,
-				      uint64_t central_freq,
-				      uint32_t p0, uint32_t p1, uint32_t p2)
-{
-	uint64_t dco_freq;
-
-	switch (central_freq) {
-	case 9600000000ULL:
-		params->central_freq = 0;
-		break;
-	case 9000000000ULL:
-		params->central_freq = 1;
-		break;
-	case 8400000000ULL:
-		params->central_freq = 3;
-	}
-
-	switch (p0) {
-	case 1:
-		params->pdiv = 0;
-		break;
-	case 2:
-		params->pdiv = 1;
-		break;
-	case 3:
-		params->pdiv = 2;
-		break;
-	case 7:
-		params->pdiv = 4;
-		break;
-	default:
-		WARN(1, "Incorrect PDiv\n");
-	}
-
-	switch (p2) {
-	case 5:
-		params->kdiv = 0;
-		break;
-	case 2:
-		params->kdiv = 1;
-		break;
-	case 3:
-		params->kdiv = 2;
-		break;
-	case 1:
-		params->kdiv = 3;
-		break;
-	default:
-		WARN(1, "Incorrect KDiv\n");
-	}
-
-	params->qdiv_ratio = p1;
-	params->qdiv_mode = (params->qdiv_ratio == 1) ? 0 : 1;
-
-	dco_freq = p0 * p1 * p2 * afe_clock;
-
-	/*
-	 * Intermediate values are in Hz.
-	 * Divide by MHz to match bsepc
-	 */
-	params->dco_integer = div_u64(dco_freq, 24 * MHz(1));
-	params->dco_fraction =
-		div_u64((div_u64(dco_freq, 24) -
-			 params->dco_integer * MHz(1)) * 0x8000, MHz(1));
-}
-
-static bool
-skl_ddi_calculate_wrpll(int clock /* in Hz */,
-			struct skl_wrpll_params *wrpll_params)
-{
-	uint64_t afe_clock = clock * 5; /* AFE Clock is 5x Pixel clock */
-	uint64_t dco_central_freq[3] = {8400000000ULL,
-					9000000000ULL,
-					9600000000ULL};
-	static const int even_dividers[] = {  4,  6,  8, 10, 12, 14, 16, 18, 20,
-					     24, 28, 30, 32, 36, 40, 42, 44,
-					     48, 52, 54, 56, 60, 64, 66, 68,
-					     70, 72, 76, 78, 80, 84, 88, 90,
-					     92, 96, 98 };
-	static const int odd_dividers[] = { 3, 5, 7, 9, 15, 21, 35 };
-	static const struct {
-		const int *list;
-		int n_dividers;
-	} dividers[] = {
-		{ even_dividers, ARRAY_SIZE(even_dividers) },
-		{ odd_dividers, ARRAY_SIZE(odd_dividers) },
-	};
-	struct skl_wrpll_context ctx;
-	unsigned int dco, d, i;
-	unsigned int p0, p1, p2;
-
-	skl_wrpll_context_init(&ctx);
-
-	for (d = 0; d < ARRAY_SIZE(dividers); d++) {
-		for (dco = 0; dco < ARRAY_SIZE(dco_central_freq); dco++) {
-			for (i = 0; i < dividers[d].n_dividers; i++) {
-				unsigned int p = dividers[d].list[i];
-				uint64_t dco_freq = p * afe_clock;
-
-				skl_wrpll_try_divider(&ctx,
-						      dco_central_freq[dco],
-						      dco_freq,
-						      p);
-				/*
-				 * Skip the remaining dividers if we're sure to
-				 * have found the definitive divider, we can't
-				 * improve a 0 deviation.
-				 */
-				if (ctx.min_deviation == 0)
-					goto skip_remaining_dividers;
-			}
-		}
-
-skip_remaining_dividers:
-		/*
-		 * If a solution is found with an even divider, prefer
-		 * this one.
-		 */
-		if (d == 0 && ctx.p)
-			break;
-	}
-
-	if (!ctx.p) {
-		DRM_DEBUG_DRIVER("No valid divider found for %dHz\n", clock);
-		return false;
-	}
+	struct intel_shared_dpll *pll;
 
-	/*
-	 * gcc incorrectly analyses that these can be used without being
-	 * initialized. To be fair, it's hard to guess.
-	 */
-	p0 = p1 = p2 = 0;
-	skl_wrpll_get_multipliers(ctx.p, &p0, &p1, &p2);
-	skl_wrpll_params_populate(wrpll_params, afe_clock, ctx.central_freq,
-				  p0, p1, p2);
+	pll = intel_get_shared_dpll(intel_crtc, crtc_state,
+				    intel_encoder);
+	if (!pll)
+		DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
+				 pipe_name(intel_crtc->pipe));
 
-	return true;
+	return pll;
 }
 
 static bool
@@ -1512,218 +1015,23 @@ skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 		   struct intel_encoder *intel_encoder)
 {
 	struct intel_shared_dpll *pll;
-	uint32_t ctrl1, cfgcr1, cfgcr2;
-	int clock = crtc_state->port_clock;
-
-	/*
-	 * See comment in intel_dpll_hw_state to understand why we always use 0
-	 * as the DPLL id in this function.
-	 */
-
-	ctrl1 = DPLL_CTRL1_OVERRIDE(0);
-
-	if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
-		struct skl_wrpll_params wrpll_params = { 0, };
-
-		ctrl1 |= DPLL_CTRL1_HDMI_MODE(0);
-
-		if (!skl_ddi_calculate_wrpll(clock * 1000, &wrpll_params))
-			return false;
-
-		cfgcr1 = DPLL_CFGCR1_FREQ_ENABLE |
-			 DPLL_CFGCR1_DCO_FRACTION(wrpll_params.dco_fraction) |
-			 wrpll_params.dco_integer;
-
-		cfgcr2 = DPLL_CFGCR2_QDIV_RATIO(wrpll_params.qdiv_ratio) |
-			 DPLL_CFGCR2_QDIV_MODE(wrpll_params.qdiv_mode) |
-			 DPLL_CFGCR2_KDIV(wrpll_params.kdiv) |
-			 DPLL_CFGCR2_PDIV(wrpll_params.pdiv) |
-			 wrpll_params.central_freq;
-	} else if (intel_encoder->type == INTEL_OUTPUT_DISPLAYPORT ||
-		   intel_encoder->type == INTEL_OUTPUT_DP_MST) {
-		switch (crtc_state->port_clock / 2) {
-		case 81000:
-			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810, 0);
-			break;
-		case 135000:
-			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1350, 0);
-			break;
-		case 270000:
-			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2700, 0);
-			break;
-		}
-
-		cfgcr1 = cfgcr2 = 0;
-	} else if (intel_encoder->type == INTEL_OUTPUT_EDP) {
-		return true;
-	} else
-		return false;
 
-	memset(&crtc_state->dpll_hw_state, 0,
-	       sizeof(crtc_state->dpll_hw_state));
-
-	crtc_state->dpll_hw_state.ctrl1 = ctrl1;
-	crtc_state->dpll_hw_state.cfgcr1 = cfgcr1;
-	crtc_state->dpll_hw_state.cfgcr2 = cfgcr2;
-
-	pll = intel_get_shared_dpll(intel_crtc, crtc_state);
+	pll = intel_get_shared_dpll(intel_crtc, crtc_state, intel_encoder);
 	if (pll == NULL) {
 		DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
 				 pipe_name(intel_crtc->pipe));
 		return false;
 	}
 
-	/* shared DPLL id 0 is DPLL 1 */
-	crtc_state->ddi_pll_sel = pll->id + 1;
-
 	return true;
 }
 
-/* bxt clock parameters */
-struct bxt_clk_div {
-	int clock;
-	uint32_t p1;
-	uint32_t p2;
-	uint32_t m2_int;
-	uint32_t m2_frac;
-	bool m2_frac_en;
-	uint32_t n;
-};
-
-/* pre-calculated values for DP linkrates */
-static const struct bxt_clk_div bxt_dp_clk_val[] = {
-	{162000, 4, 2, 32, 1677722, 1, 1},
-	{270000, 4, 1, 27,       0, 0, 1},
-	{540000, 2, 1, 27,       0, 0, 1},
-	{216000, 3, 2, 32, 1677722, 1, 1},
-	{243000, 4, 1, 24, 1258291, 1, 1},
-	{324000, 4, 1, 32, 1677722, 1, 1},
-	{432000, 3, 1, 32, 1677722, 1, 1}
-};
-
 static bool
 bxt_ddi_pll_select(struct intel_crtc *intel_crtc,
 		   struct intel_crtc_state *crtc_state,
 		   struct intel_encoder *intel_encoder)
 {
-	struct intel_shared_dpll *pll;
-	struct bxt_clk_div clk_div = {0};
-	int vco = 0;
-	uint32_t prop_coef, int_coef, gain_ctl, targ_cnt;
-	uint32_t lanestagger;
-	int clock = crtc_state->port_clock;
-
-	if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
-		intel_clock_t best_clock;
-
-		/* Calculate HDMI div */
-		/*
-		 * FIXME: tie the following calculation into
-		 * i9xx_crtc_compute_clock
-		 */
-		if (!bxt_find_best_dpll(crtc_state, clock, &best_clock)) {
-			DRM_DEBUG_DRIVER("no PLL dividers found for clock %d pipe %c\n",
-					 clock, pipe_name(intel_crtc->pipe));
-			return false;
-		}
-
-		clk_div.p1 = best_clock.p1;
-		clk_div.p2 = best_clock.p2;
-		WARN_ON(best_clock.m1 != 2);
-		clk_div.n = best_clock.n;
-		clk_div.m2_int = best_clock.m2 >> 22;
-		clk_div.m2_frac = best_clock.m2 & ((1 << 22) - 1);
-		clk_div.m2_frac_en = clk_div.m2_frac != 0;
-
-		vco = best_clock.vco;
-	} else if (intel_encoder->type == INTEL_OUTPUT_DISPLAYPORT ||
-			intel_encoder->type == INTEL_OUTPUT_EDP) {
-		int i;
-
-		clk_div = bxt_dp_clk_val[0];
-		for (i = 0; i < ARRAY_SIZE(bxt_dp_clk_val); ++i) {
-			if (bxt_dp_clk_val[i].clock == clock) {
-				clk_div = bxt_dp_clk_val[i];
-				break;
-			}
-		}
-		vco = clock * 10 / 2 * clk_div.p1 * clk_div.p2;
-	}
-
-	if (vco >= 6200000 && vco <= 6700000) {
-		prop_coef = 4;
-		int_coef = 9;
-		gain_ctl = 3;
-		targ_cnt = 8;
-	} else if ((vco > 5400000 && vco < 6200000) ||
-			(vco >= 4800000 && vco < 5400000)) {
-		prop_coef = 5;
-		int_coef = 11;
-		gain_ctl = 3;
-		targ_cnt = 9;
-	} else if (vco == 5400000) {
-		prop_coef = 3;
-		int_coef = 8;
-		gain_ctl = 1;
-		targ_cnt = 9;
-	} else {
-		DRM_ERROR("Invalid VCO\n");
-		return false;
-	}
-
-	memset(&crtc_state->dpll_hw_state, 0,
-	       sizeof(crtc_state->dpll_hw_state));
-
-	if (clock > 270000)
-		lanestagger = 0x18;
-	else if (clock > 135000)
-		lanestagger = 0x0d;
-	else if (clock > 67000)
-		lanestagger = 0x07;
-	else if (clock > 33000)
-		lanestagger = 0x04;
-	else
-		lanestagger = 0x02;
-
-	crtc_state->dpll_hw_state.ebb0 =
-		PORT_PLL_P1(clk_div.p1) | PORT_PLL_P2(clk_div.p2);
-	crtc_state->dpll_hw_state.pll0 = clk_div.m2_int;
-	crtc_state->dpll_hw_state.pll1 = PORT_PLL_N(clk_div.n);
-	crtc_state->dpll_hw_state.pll2 = clk_div.m2_frac;
-
-	if (clk_div.m2_frac_en)
-		crtc_state->dpll_hw_state.pll3 =
-			PORT_PLL_M2_FRAC_ENABLE;
-
-	crtc_state->dpll_hw_state.pll6 =
-		prop_coef | PORT_PLL_INT_COEFF(int_coef);
-	crtc_state->dpll_hw_state.pll6 |=
-		PORT_PLL_GAIN_CTL(gain_ctl);
-
-	crtc_state->dpll_hw_state.pll8 = targ_cnt;
-
-	crtc_state->dpll_hw_state.pll9 = 5 << PORT_PLL_LOCK_THRESHOLD_SHIFT;
-
-	crtc_state->dpll_hw_state.pll10 =
-		PORT_PLL_DCO_AMP(PORT_PLL_DCO_AMP_DEFAULT)
-		| PORT_PLL_DCO_AMP_OVR_EN_H;
-
-	crtc_state->dpll_hw_state.ebb4 = PORT_PLL_10BIT_CLK_ENABLE;
-
-	crtc_state->dpll_hw_state.pcsdw12 =
-		LANESTAGGER_STRAP_OVRD | lanestagger;
-
-	pll = intel_get_shared_dpll(intel_crtc, crtc_state);
-	if (pll == NULL) {
-		DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
-			pipe_name(intel_crtc->pipe));
-		return false;
-	}
-
-	/* shared DPLL id 0 is DPLL A */
-	crtc_state->ddi_pll_sel = pll->id;
-
-	return true;
+	return !!intel_get_shared_dpll(intel_crtc, crtc_state, intel_encoder);
 }
 
 /*
@@ -1761,6 +1069,8 @@ void intel_ddi_set_pipe_settings(struct drm_crtc *crtc)
 	uint32_t temp;
 
 	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP || type == INTEL_OUTPUT_DP_MST) {
+		WARN_ON(transcoder_is_dsi(cpu_transcoder));
+
 		temp = TRANS_MSA_SYNC_CLK;
 		switch (intel_crtc->config->pipe_bpp) {
 		case 18:
@@ -2129,7 +1439,7 @@ static void bxt_ddi_vswing_sequence(struct drm_i915_private *dev_priv,
 	u32 n_entries, i;
 	uint32_t val;
 
-	if (type == INTEL_OUTPUT_EDP && dev_priv->edp_low_vswing) {
+	if (type == INTEL_OUTPUT_EDP && dev_priv->vbt.edp.low_vswing) {
 		n_entries = ARRAY_SIZE(bxt_ddi_translations_edp);
 		ddi_translations = bxt_ddi_translations_edp;
 	} else if (type == INTEL_OUTPUT_DISPLAYPORT
@@ -2267,24 +1577,6 @@ void intel_ddi_clk_select(struct intel_encoder *encoder,
 		uint32_t dpll = pipe_config->ddi_pll_sel;
 		uint32_t val;
 
-		/*
-		 * DPLL0 is used for eDP and is the only "private" DPLL (as
-		 * opposed to shared) on SKL
-		 */
-		if (encoder->type == INTEL_OUTPUT_EDP) {
-			WARN_ON(dpll != SKL_DPLL0);
-
-			val = I915_READ(DPLL_CTRL1);
-
-			val &= ~(DPLL_CTRL1_HDMI_MODE(dpll) |
-				 DPLL_CTRL1_SSC(dpll) |
-				 DPLL_CTRL1_LINK_RATE_MASK(dpll));
-			val |= pipe_config->dpll_hw_state.ctrl1 << (dpll * 6);
-
-			I915_WRITE(DPLL_CTRL1, val);
-			POSTING_READ(DPLL_CTRL1);
-		}
-
 		/* DDI -> PLL mapping  */
 		val = I915_READ(DPLL_CTRL2);
 
@@ -2438,251 +1730,101 @@ static void intel_disable_ddi(struct intel_encoder *intel_encoder)
 	}
 }
 
-static void hsw_ddi_wrpll_enable(struct drm_i915_private *dev_priv,
-			       struct intel_shared_dpll *pll)
-{
-	I915_WRITE(WRPLL_CTL(pll->id), pll->config.hw_state.wrpll);
-	POSTING_READ(WRPLL_CTL(pll->id));
-	udelay(20);
-}
-
-static void hsw_ddi_spll_enable(struct drm_i915_private *dev_priv,
-				struct intel_shared_dpll *pll)
-{
-	I915_WRITE(SPLL_CTL, pll->config.hw_state.spll);
-	POSTING_READ(SPLL_CTL);
-	udelay(20);
-}
-
-static void hsw_ddi_wrpll_disable(struct drm_i915_private *dev_priv,
-				  struct intel_shared_dpll *pll)
+static bool broxton_phy_is_enabled(struct drm_i915_private *dev_priv,
+				   enum dpio_phy phy)
 {
-	uint32_t val;
-
-	val = I915_READ(WRPLL_CTL(pll->id));
-	I915_WRITE(WRPLL_CTL(pll->id), val & ~WRPLL_PLL_ENABLE);
-	POSTING_READ(WRPLL_CTL(pll->id));
-}
-
-static void hsw_ddi_spll_disable(struct drm_i915_private *dev_priv,
-				 struct intel_shared_dpll *pll)
-{
-	uint32_t val;
-
-	val = I915_READ(SPLL_CTL);
-	I915_WRITE(SPLL_CTL, val & ~SPLL_PLL_ENABLE);
-	POSTING_READ(SPLL_CTL);
-}
-
-static bool hsw_ddi_wrpll_get_hw_state(struct drm_i915_private *dev_priv,
-				       struct intel_shared_dpll *pll,
-				       struct intel_dpll_hw_state *hw_state)
-{
-	uint32_t val;
-
-	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
+	if (!(I915_READ(BXT_P_CR_GT_DISP_PWRON) & GT_DISPLAY_POWER_ON(phy)))
 		return false;
 
-	val = I915_READ(WRPLL_CTL(pll->id));
-	hw_state->wrpll = val;
-
-	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
-
-	return val & WRPLL_PLL_ENABLE;
-}
-
-static bool hsw_ddi_spll_get_hw_state(struct drm_i915_private *dev_priv,
-				      struct intel_shared_dpll *pll,
-				      struct intel_dpll_hw_state *hw_state)
-{
-	uint32_t val;
+	if ((I915_READ(BXT_PORT_CL1CM_DW0(phy)) &
+	     (PHY_POWER_GOOD | PHY_RESERVED)) != PHY_POWER_GOOD) {
+		DRM_DEBUG_DRIVER("DDI PHY %d powered, but power hasn't settled\n",
+				 phy);
 
-	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
 		return false;
+	}
 
-	val = I915_READ(SPLL_CTL);
-	hw_state->spll = val;
-
-	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
-
-	return val & SPLL_PLL_ENABLE;
-}
-
-
-static const char * const hsw_ddi_pll_names[] = {
-	"WRPLL 1",
-	"WRPLL 2",
-	"SPLL"
-};
+	if (phy == DPIO_PHY1 &&
+	    !(I915_READ(BXT_PORT_REF_DW3(DPIO_PHY1)) & GRC_DONE)) {
+		DRM_DEBUG_DRIVER("DDI PHY 1 powered, but GRC isn't done\n");
 
-static void hsw_shared_dplls_init(struct drm_i915_private *dev_priv)
-{
-	int i;
+		return false;
+	}
 
-	dev_priv->num_shared_dpll = 3;
+	if (!(I915_READ(BXT_PHY_CTL_FAMILY(phy)) & COMMON_RESET_DIS)) {
+		DRM_DEBUG_DRIVER("DDI PHY %d powered, but still in reset\n",
+				 phy);
 
-	for (i = 0; i < 2; i++) {
-		dev_priv->shared_dplls[i].id = i;
-		dev_priv->shared_dplls[i].name = hsw_ddi_pll_names[i];
-		dev_priv->shared_dplls[i].disable = hsw_ddi_wrpll_disable;
-		dev_priv->shared_dplls[i].enable = hsw_ddi_wrpll_enable;
-		dev_priv->shared_dplls[i].get_hw_state =
-			hsw_ddi_wrpll_get_hw_state;
+		return false;
 	}
 
-	/* SPLL is special, but needs to be initialized anyway.. */
-	dev_priv->shared_dplls[i].id = i;
-	dev_priv->shared_dplls[i].name = hsw_ddi_pll_names[i];
-	dev_priv->shared_dplls[i].disable = hsw_ddi_spll_disable;
-	dev_priv->shared_dplls[i].enable = hsw_ddi_spll_enable;
-	dev_priv->shared_dplls[i].get_hw_state = hsw_ddi_spll_get_hw_state;
-
+	return true;
 }
 
-static const char * const skl_ddi_pll_names[] = {
-	"DPLL 1",
-	"DPLL 2",
-	"DPLL 3",
-};
-
-struct skl_dpll_regs {
-	i915_reg_t ctl, cfgcr1, cfgcr2;
-};
-
-/* this array is indexed by the *shared* pll id */
-static const struct skl_dpll_regs skl_dpll_regs[3] = {
-	{
-		/* DPLL 1 */
-		.ctl = LCPLL2_CTL,
-		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL1),
-		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL1),
-	},
-	{
-		/* DPLL 2 */
-		.ctl = WRPLL_CTL(0),
-		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL2),
-		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL2),
-	},
-	{
-		/* DPLL 3 */
-		.ctl = WRPLL_CTL(1),
-		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL3),
-		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL3),
-	},
-};
-
-static void skl_ddi_pll_enable(struct drm_i915_private *dev_priv,
-			       struct intel_shared_dpll *pll)
+static u32 broxton_get_grc(struct drm_i915_private *dev_priv, enum dpio_phy phy)
 {
-	uint32_t val;
-	unsigned int dpll;
-	const struct skl_dpll_regs *regs = skl_dpll_regs;
-
-	/* DPLL0 is not part of the shared DPLLs, so pll->id is 0 for DPLL1 */
-	dpll = pll->id + 1;
-
-	val = I915_READ(DPLL_CTRL1);
-
-	val &= ~(DPLL_CTRL1_HDMI_MODE(dpll) | DPLL_CTRL1_SSC(dpll) |
-		 DPLL_CTRL1_LINK_RATE_MASK(dpll));
-	val |= pll->config.hw_state.ctrl1 << (dpll * 6);
-
-	I915_WRITE(DPLL_CTRL1, val);
-	POSTING_READ(DPLL_CTRL1);
-
-	I915_WRITE(regs[pll->id].cfgcr1, pll->config.hw_state.cfgcr1);
-	I915_WRITE(regs[pll->id].cfgcr2, pll->config.hw_state.cfgcr2);
-	POSTING_READ(regs[pll->id].cfgcr1);
-	POSTING_READ(regs[pll->id].cfgcr2);
+	u32 val = I915_READ(BXT_PORT_REF_DW6(phy));
 
-	/* the enable bit is always bit 31 */
-	I915_WRITE(regs[pll->id].ctl,
-		   I915_READ(regs[pll->id].ctl) | LCPLL_PLL_ENABLE);
-
-	if (wait_for(I915_READ(DPLL_STATUS) & DPLL_LOCK(dpll), 5))
-		DRM_ERROR("DPLL %d not locked\n", dpll);
+	return (val & GRC_CODE_MASK) >> GRC_CODE_SHIFT;
 }
 
-static void skl_ddi_pll_disable(struct drm_i915_private *dev_priv,
-				struct intel_shared_dpll *pll)
+static void broxton_phy_wait_grc_done(struct drm_i915_private *dev_priv,
+				      enum dpio_phy phy)
 {
-	const struct skl_dpll_regs *regs = skl_dpll_regs;
-
-	/* the enable bit is always bit 31 */
-	I915_WRITE(regs[pll->id].ctl,
-		   I915_READ(regs[pll->id].ctl) & ~LCPLL_PLL_ENABLE);
-	POSTING_READ(regs[pll->id].ctl);
+	if (wait_for(I915_READ(BXT_PORT_REF_DW3(phy)) & GRC_DONE, 10))
+		DRM_ERROR("timeout waiting for PHY%d GRC\n", phy);
 }
 
-static bool skl_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
-				     struct intel_shared_dpll *pll,
-				     struct intel_dpll_hw_state *hw_state)
-{
-	uint32_t val;
-	unsigned int dpll;
-	const struct skl_dpll_regs *regs = skl_dpll_regs;
-	bool ret;
+static bool broxton_phy_verify_state(struct drm_i915_private *dev_priv,
+				     enum dpio_phy phy);
 
-	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
-		return false;
-
-	ret = false;
-
-	/* DPLL0 is not part of the shared DPLLs, so pll->id is 0 for DPLL1 */
-	dpll = pll->id + 1;
-
-	val = I915_READ(regs[pll->id].ctl);
-	if (!(val & LCPLL_PLL_ENABLE))
-		goto out;
-
-	val = I915_READ(DPLL_CTRL1);
-	hw_state->ctrl1 = (val >> (dpll * 6)) & 0x3f;
-
-	/* avoid reading back stale values if HDMI mode is not enabled */
-	if (val & DPLL_CTRL1_HDMI_MODE(dpll)) {
-		hw_state->cfgcr1 = I915_READ(regs[pll->id].cfgcr1);
-		hw_state->cfgcr2 = I915_READ(regs[pll->id].cfgcr2);
-	}
-	ret = true;
-
-out:
-	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+static void broxton_phy_init(struct drm_i915_private *dev_priv,
+			     enum dpio_phy phy)
+{
+	enum port port;
+	u32 ports, val;
 
-	return ret;
-}
+	if (broxton_phy_is_enabled(dev_priv, phy)) {
+		/* Still read out the GRC value for state verification */
+		if (phy == DPIO_PHY0)
+			dev_priv->bxt_phy_grc = broxton_get_grc(dev_priv, phy);
 
-static void skl_shared_dplls_init(struct drm_i915_private *dev_priv)
-{
-	int i;
+		if (broxton_phy_verify_state(dev_priv, phy)) {
+			DRM_DEBUG_DRIVER("DDI PHY %d already enabled, "
+					 "won't reprogram it\n", phy);
 
-	dev_priv->num_shared_dpll = 3;
+			return;
+		}
 
-	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
-		dev_priv->shared_dplls[i].id = i;
-		dev_priv->shared_dplls[i].name = skl_ddi_pll_names[i];
-		dev_priv->shared_dplls[i].disable = skl_ddi_pll_disable;
-		dev_priv->shared_dplls[i].enable = skl_ddi_pll_enable;
-		dev_priv->shared_dplls[i].get_hw_state =
-			skl_ddi_pll_get_hw_state;
+		DRM_DEBUG_DRIVER("DDI PHY %d enabled with invalid state, "
+				 "force reprogramming it\n", phy);
+	} else {
+		DRM_DEBUG_DRIVER("DDI PHY %d not enabled, enabling it\n", phy);
 	}
-}
-
-static void broxton_phy_init(struct drm_i915_private *dev_priv,
-			     enum dpio_phy phy)
-{
-	enum port port;
-	uint32_t val;
 
 	val = I915_READ(BXT_P_CR_GT_DISP_PWRON);
 	val |= GT_DISPLAY_POWER_ON(phy);
 	I915_WRITE(BXT_P_CR_GT_DISP_PWRON, val);
 
-	/* Considering 10ms timeout until BSpec is updated */
-	if (wait_for(I915_READ(BXT_PORT_CL1CM_DW0(phy)) & PHY_POWER_GOOD, 10))
+	/*
+	 * The PHY registers start out inaccessible and respond to reads with
+	 * all 1s.  Eventually they become accessible as they power up, then
+	 * the reserved bit will give the default 0.  Poll on the reserved bit
+	 * becoming 0 to find when the PHY is accessible.
+	 * HW team confirmed that the time to reach phypowergood status is
+	 * anywhere between 50 us and 100us.
+	 */
+	if (wait_for_us(((I915_READ(BXT_PORT_CL1CM_DW0(phy)) &
+		(PHY_RESERVED | PHY_POWER_GOOD)) == PHY_POWER_GOOD), 100)) {
 		DRM_ERROR("timeout during PHY%d power on\n", phy);
+	}
 
-	for (port =  (phy == DPIO_PHY0 ? PORT_B : PORT_A);
-	     port <= (phy == DPIO_PHY0 ? PORT_C : PORT_A); port++) {
+	if (phy == DPIO_PHY0)
+		ports = BIT(PORT_B) | BIT(PORT_C);
+	else
+		ports = BIT(PORT_A);
+
+	for_each_port_masked(port, ports) {
 		int lane;
 
 		for (lane = 0; lane < 4; lane++) {
@@ -2730,6 +1872,9 @@ static void broxton_phy_init(struct drm_i915_private *dev_priv,
 	 * enabled.
 	 * TODO: port C is only connected on BXT-P, so on BXT0/1 we should
 	 * power down the second channel on PHY0 as well.
+	 *
+	 * FIXME: Clarify programming of the following, the register is
+	 * read-only with bit 6 fixed at 0 at least in stepping A.
 	 */
 	if (phy == DPIO_PHY1)
 		val |= OCL2_LDOFUSE_PWR_DIS;
@@ -2742,12 +1887,10 @@ static void broxton_phy_init(struct drm_i915_private *dev_priv,
 		 * the corresponding calibrated value from PHY1, and disable
 		 * the automatic calibration on PHY0.
 		 */
-		if (wait_for(I915_READ(BXT_PORT_REF_DW3(DPIO_PHY1)) & GRC_DONE,
-			     10))
-			DRM_ERROR("timeout waiting for PHY1 GRC\n");
+		broxton_phy_wait_grc_done(dev_priv, DPIO_PHY1);
 
-		val = I915_READ(BXT_PORT_REF_DW6(DPIO_PHY1));
-		val = (val & GRC_CODE_MASK) >> GRC_CODE_SHIFT;
+		val = dev_priv->bxt_phy_grc = broxton_get_grc(dev_priv,
+							      DPIO_PHY1);
 		grc_code = val << GRC_CODE_FAST_SHIFT |
 			   val << GRC_CODE_SLOW_SHIFT |
 			   val;
@@ -2757,17 +1900,27 @@ static void broxton_phy_init(struct drm_i915_private *dev_priv,
 		val |= GRC_DIS | GRC_RDY_OVRD;
 		I915_WRITE(BXT_PORT_REF_DW8(DPIO_PHY0), val);
 	}
+	/*
+	 * During PHY1 init delay waiting for GRC calibration to finish, since
+	 * it can happen in parallel with the subsequent PHY0 init.
+	 */
 
 	val = I915_READ(BXT_PHY_CTL_FAMILY(phy));
 	val |= COMMON_RESET_DIS;
 	I915_WRITE(BXT_PHY_CTL_FAMILY(phy), val);
 }
 
-void broxton_ddi_phy_init(struct drm_device *dev)
+void broxton_ddi_phy_init(struct drm_i915_private *dev_priv)
 {
 	/* Enable PHY1 first since it provides Rcomp for PHY0 */
-	broxton_phy_init(dev->dev_private, DPIO_PHY1);
-	broxton_phy_init(dev->dev_private, DPIO_PHY0);
+	broxton_phy_init(dev_priv, DPIO_PHY1);
+	broxton_phy_init(dev_priv, DPIO_PHY0);
+
+	/*
+	 * If BIOS enabled only PHY0 and not PHY1, we skipped waiting for the
+	 * PHY1 GRC calibration to finish, so wait for it here.
+	 */
+	broxton_phy_wait_grc_done(dev_priv, DPIO_PHY1);
 }
 
 static void broxton_phy_uninit(struct drm_i915_private *dev_priv,
@@ -2778,260 +1931,126 @@ static void broxton_phy_uninit(struct drm_i915_private *dev_priv,
 	val = I915_READ(BXT_PHY_CTL_FAMILY(phy));
 	val &= ~COMMON_RESET_DIS;
 	I915_WRITE(BXT_PHY_CTL_FAMILY(phy), val);
+
+	val = I915_READ(BXT_P_CR_GT_DISP_PWRON);
+	val &= ~GT_DISPLAY_POWER_ON(phy);
+	I915_WRITE(BXT_P_CR_GT_DISP_PWRON, val);
 }
 
-void broxton_ddi_phy_uninit(struct drm_device *dev)
+void broxton_ddi_phy_uninit(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
 	broxton_phy_uninit(dev_priv, DPIO_PHY1);
 	broxton_phy_uninit(dev_priv, DPIO_PHY0);
-
-	/* FIXME: do this in broxton_phy_uninit per phy */
-	I915_WRITE(BXT_P_CR_GT_DISP_PWRON, 0);
 }
 
-static const char * const bxt_ddi_pll_names[] = {
-	"PORT PLL A",
-	"PORT PLL B",
-	"PORT PLL C",
-};
-
-static void bxt_ddi_pll_enable(struct drm_i915_private *dev_priv,
-				struct intel_shared_dpll *pll)
+static bool __printf(6, 7)
+__phy_reg_verify_state(struct drm_i915_private *dev_priv, enum dpio_phy phy,
+		       i915_reg_t reg, u32 mask, u32 expected,
+		       const char *reg_fmt, ...)
 {
-	uint32_t temp;
-	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
-
-	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
-	temp &= ~PORT_PLL_REF_SEL;
-	/* Non-SSC reference */
-	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
-
-	/* Disable 10 bit clock */
-	temp = I915_READ(BXT_PORT_PLL_EBB_4(port));
-	temp &= ~PORT_PLL_10BIT_CLK_ENABLE;
-	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
-
-	/* Write P1 & P2 */
-	temp = I915_READ(BXT_PORT_PLL_EBB_0(port));
-	temp &= ~(PORT_PLL_P1_MASK | PORT_PLL_P2_MASK);
-	temp |= pll->config.hw_state.ebb0;
-	I915_WRITE(BXT_PORT_PLL_EBB_0(port), temp);
-
-	/* Write M2 integer */
-	temp = I915_READ(BXT_PORT_PLL(port, 0));
-	temp &= ~PORT_PLL_M2_MASK;
-	temp |= pll->config.hw_state.pll0;
-	I915_WRITE(BXT_PORT_PLL(port, 0), temp);
-
-	/* Write N */
-	temp = I915_READ(BXT_PORT_PLL(port, 1));
-	temp &= ~PORT_PLL_N_MASK;
-	temp |= pll->config.hw_state.pll1;
-	I915_WRITE(BXT_PORT_PLL(port, 1), temp);
-
-	/* Write M2 fraction */
-	temp = I915_READ(BXT_PORT_PLL(port, 2));
-	temp &= ~PORT_PLL_M2_FRAC_MASK;
-	temp |= pll->config.hw_state.pll2;
-	I915_WRITE(BXT_PORT_PLL(port, 2), temp);
-
-	/* Write M2 fraction enable */
-	temp = I915_READ(BXT_PORT_PLL(port, 3));
-	temp &= ~PORT_PLL_M2_FRAC_ENABLE;
-	temp |= pll->config.hw_state.pll3;
-	I915_WRITE(BXT_PORT_PLL(port, 3), temp);
-
-	/* Write coeff */
-	temp = I915_READ(BXT_PORT_PLL(port, 6));
-	temp &= ~PORT_PLL_PROP_COEFF_MASK;
-	temp &= ~PORT_PLL_INT_COEFF_MASK;
-	temp &= ~PORT_PLL_GAIN_CTL_MASK;
-	temp |= pll->config.hw_state.pll6;
-	I915_WRITE(BXT_PORT_PLL(port, 6), temp);
-
-	/* Write calibration val */
-	temp = I915_READ(BXT_PORT_PLL(port, 8));
-	temp &= ~PORT_PLL_TARGET_CNT_MASK;
-	temp |= pll->config.hw_state.pll8;
-	I915_WRITE(BXT_PORT_PLL(port, 8), temp);
-
-	temp = I915_READ(BXT_PORT_PLL(port, 9));
-	temp &= ~PORT_PLL_LOCK_THRESHOLD_MASK;
-	temp |= pll->config.hw_state.pll9;
-	I915_WRITE(BXT_PORT_PLL(port, 9), temp);
-
-	temp = I915_READ(BXT_PORT_PLL(port, 10));
-	temp &= ~PORT_PLL_DCO_AMP_OVR_EN_H;
-	temp &= ~PORT_PLL_DCO_AMP_MASK;
-	temp |= pll->config.hw_state.pll10;
-	I915_WRITE(BXT_PORT_PLL(port, 10), temp);
-
-	/* Recalibrate with new settings */
-	temp = I915_READ(BXT_PORT_PLL_EBB_4(port));
-	temp |= PORT_PLL_RECALIBRATE;
-	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
-	temp &= ~PORT_PLL_10BIT_CLK_ENABLE;
-	temp |= pll->config.hw_state.ebb4;
-	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
-
-	/* Enable PLL */
-	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
-	temp |= PORT_PLL_ENABLE;
-	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
-	POSTING_READ(BXT_PORT_PLL_ENABLE(port));
-
-	if (wait_for_atomic_us((I915_READ(BXT_PORT_PLL_ENABLE(port)) &
-			PORT_PLL_LOCK), 200))
-		DRM_ERROR("PLL %d not locked\n", port);
-
-	/*
-	 * While we write to the group register to program all lanes at once we
-	 * can read only lane registers and we pick lanes 0/1 for that.
-	 */
-	temp = I915_READ(BXT_PORT_PCS_DW12_LN01(port));
-	temp &= ~LANE_STAGGER_MASK;
-	temp &= ~LANESTAGGER_STRAP_OVRD;
-	temp |= pll->config.hw_state.pcsdw12;
-	I915_WRITE(BXT_PORT_PCS_DW12_GRP(port), temp);
-}
-
-static void bxt_ddi_pll_disable(struct drm_i915_private *dev_priv,
-					struct intel_shared_dpll *pll)
-{
-	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
-	uint32_t temp;
+	struct va_format vaf;
+	va_list args;
+	u32 val;
 
-	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
-	temp &= ~PORT_PLL_ENABLE;
-	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
-	POSTING_READ(BXT_PORT_PLL_ENABLE(port));
-}
+	val = I915_READ(reg);
+	if ((val & mask) == expected)
+		return true;
 
-static bool bxt_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
-					struct intel_shared_dpll *pll,
-					struct intel_dpll_hw_state *hw_state)
-{
-	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
-	uint32_t val;
-	bool ret;
+	va_start(args, reg_fmt);
+	vaf.fmt = reg_fmt;
+	vaf.va = &args;
 
-	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
-		return false;
+	DRM_DEBUG_DRIVER("DDI PHY %d reg %pV [%08x] state mismatch: "
+			 "current %08x, expected %08x (mask %08x)\n",
+			 phy, &vaf, reg.reg, val, (val & ~mask) | expected,
+			 mask);
 
-	ret = false;
+	va_end(args);
 
-	val = I915_READ(BXT_PORT_PLL_ENABLE(port));
-	if (!(val & PORT_PLL_ENABLE))
-		goto out;
+	return false;
+}
 
-	hw_state->ebb0 = I915_READ(BXT_PORT_PLL_EBB_0(port));
-	hw_state->ebb0 &= PORT_PLL_P1_MASK | PORT_PLL_P2_MASK;
+static bool broxton_phy_verify_state(struct drm_i915_private *dev_priv,
+				     enum dpio_phy phy)
+{
+	enum port port;
+	u32 ports;
+	uint32_t mask;
+	bool ok;
 
-	hw_state->ebb4 = I915_READ(BXT_PORT_PLL_EBB_4(port));
-	hw_state->ebb4 &= PORT_PLL_10BIT_CLK_ENABLE;
+#define _CHK(reg, mask, exp, fmt, ...)					\
+	__phy_reg_verify_state(dev_priv, phy, reg, mask, exp, fmt,	\
+			       ## __VA_ARGS__)
 
-	hw_state->pll0 = I915_READ(BXT_PORT_PLL(port, 0));
-	hw_state->pll0 &= PORT_PLL_M2_MASK;
+	/* We expect the PHY to be always enabled */
+	if (!broxton_phy_is_enabled(dev_priv, phy))
+		return false;
 
-	hw_state->pll1 = I915_READ(BXT_PORT_PLL(port, 1));
-	hw_state->pll1 &= PORT_PLL_N_MASK;
+	ok = true;
 
-	hw_state->pll2 = I915_READ(BXT_PORT_PLL(port, 2));
-	hw_state->pll2 &= PORT_PLL_M2_FRAC_MASK;
+	if (phy == DPIO_PHY0)
+		ports = BIT(PORT_B) | BIT(PORT_C);
+	else
+		ports = BIT(PORT_A);
 
-	hw_state->pll3 = I915_READ(BXT_PORT_PLL(port, 3));
-	hw_state->pll3 &= PORT_PLL_M2_FRAC_ENABLE;
+	for_each_port_masked(port, ports) {
+		int lane;
 
-	hw_state->pll6 = I915_READ(BXT_PORT_PLL(port, 6));
-	hw_state->pll6 &= PORT_PLL_PROP_COEFF_MASK |
-			  PORT_PLL_INT_COEFF_MASK |
-			  PORT_PLL_GAIN_CTL_MASK;
+		for (lane = 0; lane < 4; lane++)
+			ok &= _CHK(BXT_PORT_TX_DW14_LN(port, lane),
+				    LATENCY_OPTIM,
+				    lane != 1 ? LATENCY_OPTIM : 0,
+				    "BXT_PORT_TX_DW14_LN(%d, %d)", port, lane);
+	}
 
-	hw_state->pll8 = I915_READ(BXT_PORT_PLL(port, 8));
-	hw_state->pll8 &= PORT_PLL_TARGET_CNT_MASK;
+	/* PLL Rcomp code offset */
+	ok &= _CHK(BXT_PORT_CL1CM_DW9(phy),
+		    IREF0RC_OFFSET_MASK, 0xe4 << IREF0RC_OFFSET_SHIFT,
+		    "BXT_PORT_CL1CM_DW9(%d)", phy);
+	ok &= _CHK(BXT_PORT_CL1CM_DW10(phy),
+		    IREF1RC_OFFSET_MASK, 0xe4 << IREF1RC_OFFSET_SHIFT,
+		    "BXT_PORT_CL1CM_DW10(%d)", phy);
 
-	hw_state->pll9 = I915_READ(BXT_PORT_PLL(port, 9));
-	hw_state->pll9 &= PORT_PLL_LOCK_THRESHOLD_MASK;
+	/* Power gating */
+	mask = OCL1_POWER_DOWN_EN | DW28_OLDO_DYN_PWR_DOWN_EN | SUS_CLK_CONFIG;
+	ok &= _CHK(BXT_PORT_CL1CM_DW28(phy), mask, mask,
+		    "BXT_PORT_CL1CM_DW28(%d)", phy);
 
-	hw_state->pll10 = I915_READ(BXT_PORT_PLL(port, 10));
-	hw_state->pll10 &= PORT_PLL_DCO_AMP_OVR_EN_H |
-			   PORT_PLL_DCO_AMP_MASK;
+	if (phy == DPIO_PHY0)
+		ok &= _CHK(BXT_PORT_CL2CM_DW6_BC,
+			   DW6_OLDO_DYN_PWR_DOWN_EN, DW6_OLDO_DYN_PWR_DOWN_EN,
+			   "BXT_PORT_CL2CM_DW6_BC");
 
 	/*
-	 * While we write to the group register to program all lanes at once we
-	 * can read only lane registers. We configure all lanes the same way, so
-	 * here just read out lanes 0/1 and output a note if lanes 2/3 differ.
+	 * TODO: Verify BXT_PORT_CL1CM_DW30 bit OCL2_LDOFUSE_PWR_DIS,
+	 * at least on stepping A this bit is read-only and fixed at 0.
 	 */
-	hw_state->pcsdw12 = I915_READ(BXT_PORT_PCS_DW12_LN01(port));
-	if (I915_READ(BXT_PORT_PCS_DW12_LN23(port)) != hw_state->pcsdw12)
-		DRM_DEBUG_DRIVER("lane stagger config different for lane 01 (%08x) and 23 (%08x)\n",
-				 hw_state->pcsdw12,
-				 I915_READ(BXT_PORT_PCS_DW12_LN23(port)));
-	hw_state->pcsdw12 &= LANE_STAGGER_MASK | LANESTAGGER_STRAP_OVRD;
-
-	ret = true;
 
-out:
-	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
-
-	return ret;
-}
-
-static void bxt_shared_dplls_init(struct drm_i915_private *dev_priv)
-{
-	int i;
+	if (phy == DPIO_PHY0) {
+		u32 grc_code = dev_priv->bxt_phy_grc;
 
-	dev_priv->num_shared_dpll = 3;
+		grc_code = grc_code << GRC_CODE_FAST_SHIFT |
+			   grc_code << GRC_CODE_SLOW_SHIFT |
+			   grc_code;
+		mask = GRC_CODE_FAST_MASK | GRC_CODE_SLOW_MASK |
+		       GRC_CODE_NOM_MASK;
+		ok &= _CHK(BXT_PORT_REF_DW6(DPIO_PHY0), mask, grc_code,
+			    "BXT_PORT_REF_DW6(%d)", DPIO_PHY0);
 
-	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
-		dev_priv->shared_dplls[i].id = i;
-		dev_priv->shared_dplls[i].name = bxt_ddi_pll_names[i];
-		dev_priv->shared_dplls[i].disable = bxt_ddi_pll_disable;
-		dev_priv->shared_dplls[i].enable = bxt_ddi_pll_enable;
-		dev_priv->shared_dplls[i].get_hw_state =
-			bxt_ddi_pll_get_hw_state;
+		mask = GRC_DIS | GRC_RDY_OVRD;
+		ok &= _CHK(BXT_PORT_REF_DW8(DPIO_PHY0), mask, mask,
+			    "BXT_PORT_REF_DW8(%d)", DPIO_PHY0);
 	}
+
+	return ok;
+#undef _CHK
 }
 
-void intel_ddi_pll_init(struct drm_device *dev)
+void broxton_ddi_phy_verify_state(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t val = I915_READ(LCPLL_CTL);
-
-	if (IS_SKYLAKE(dev) || IS_KABYLAKE(dev))
-		skl_shared_dplls_init(dev_priv);
-	else if (IS_BROXTON(dev))
-		bxt_shared_dplls_init(dev_priv);
-	else
-		hsw_shared_dplls_init(dev_priv);
-
-	if (IS_SKYLAKE(dev) || IS_KABYLAKE(dev)) {
-		int cdclk_freq;
-
-		cdclk_freq = dev_priv->display.get_display_clock_speed(dev);
-		dev_priv->skl_boot_cdclk = cdclk_freq;
-		if (skl_sanitize_cdclk(dev_priv))
-			DRM_DEBUG_KMS("Sanitized cdclk programmed by pre-os\n");
-		if (!(I915_READ(LCPLL1_CTL) & LCPLL_PLL_ENABLE))
-			DRM_ERROR("LCPLL1 is disabled\n");
-	} else if (IS_BROXTON(dev)) {
-		broxton_init_cdclk(dev);
-		broxton_ddi_phy_init(dev);
-	} else {
-		/*
-		 * The LCPLL register should be turned on by the BIOS. For now
-		 * let's just check its state and print errors in case
-		 * something is wrong.  Don't even try to turn it on.
-		 */
-
-		if (val & LCPLL_CD_SOURCE_FCLK)
-			DRM_ERROR("CDCLK source is not LCPLL\n");
-
-		if (val & LCPLL_PLL_DISABLE)
-			DRM_ERROR("LCPLL is disabled\n");
-	}
+	if (!broxton_phy_verify_state(dev_priv, DPIO_PHY0) ||
+	    !broxton_phy_verify_state(dev_priv, DPIO_PHY1))
+		i915_report_error(dev_priv, "DDI PHY state mismatch\n");
 }
 
 void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp)
@@ -3086,12 +2105,18 @@ void intel_ddi_fdi_disable(struct drm_crtc *crtc)
 	struct intel_encoder *intel_encoder = intel_ddi_get_crtc_encoder(crtc);
 	uint32_t val;
 
-	intel_ddi_post_disable(intel_encoder);
-
+	/*
+	 * Bspec lists this as both step 13 (before DDI_BUF_CTL disable)
+	 * and step 18 (after clearing PORT_CLK_SEL). Based on a BUN,
+	 * step 13 is the correct place for it. Step 18 is where it was
+	 * originally before the BUN.
+	 */
 	val = I915_READ(FDI_RX_CTL(PIPE_A));
 	val &= ~FDI_RX_ENABLE;
 	I915_WRITE(FDI_RX_CTL(PIPE_A), val);
 
+	intel_ddi_post_disable(intel_encoder);
+
 	val = I915_READ(FDI_RX_MISC(PIPE_A));
 	val &= ~(FDI_RX_PWRDN_LANE1_MASK | FDI_RX_PWRDN_LANE0_MASK);
 	val |= FDI_RX_PWRDN_LANE1_VAL(2) | FDI_RX_PWRDN_LANE0_VAL(2);
@@ -3115,6 +2140,10 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 	struct intel_hdmi *intel_hdmi;
 	u32 temp, flags = 0;
 
+	/* XXX: DSI transcoder paranoia */
+	if (WARN_ON(transcoder_is_dsi(cpu_transcoder)))
+		return;
+
 	temp = I915_READ(TRANS_DDI_FUNC_CTL(cpu_transcoder));
 	if (temp & TRANS_DDI_PHSYNC)
 		flags |= DRM_MODE_FLAG_PHSYNC;
@@ -3172,8 +2201,8 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 			pipe_config->has_audio = true;
 	}
 
-	if (encoder->type == INTEL_OUTPUT_EDP && dev_priv->vbt.edp_bpp &&
-	    pipe_config->pipe_bpp > dev_priv->vbt.edp_bpp) {
+	if (encoder->type == INTEL_OUTPUT_EDP && dev_priv->vbt.edp.bpp &&
+	    pipe_config->pipe_bpp > dev_priv->vbt.edp.bpp) {
 		/*
 		 * This is a big fat ugly hack.
 		 *
@@ -3188,8 +2217,8 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 		 * load.
 		 */
 		DRM_DEBUG_KMS("pipe has %d bpp for eDP panel, overriding BIOS-provided max %d bpp\n",
-			      pipe_config->pipe_bpp, dev_priv->vbt.edp_bpp);
-		dev_priv->vbt.edp_bpp = pipe_config->pipe_bpp;
+			      pipe_config->pipe_bpp, dev_priv->vbt.edp.bpp);
+		dev_priv->vbt.edp.bpp = pipe_config->pipe_bpp;
 	}
 
 	intel_ddi_clock_get(encoder, pipe_config);
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 0104a06d01fd..46f9be3ad5a2 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -36,6 +36,7 @@
 #include "intel_drv.h"
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
+#include "intel_dsi.h"
 #include "i915_trace.h"
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
@@ -96,12 +97,13 @@ static int intel_framebuffer_init(struct drm_device *dev,
 				  struct drm_i915_gem_object *obj);
 static void i9xx_set_pipeconf(struct intel_crtc *intel_crtc);
 static void intel_set_pipe_timings(struct intel_crtc *intel_crtc);
+static void intel_set_pipe_src_size(struct intel_crtc *intel_crtc);
 static void intel_cpu_transcoder_set_m_n(struct intel_crtc *crtc,
 					 struct intel_link_m_n *m_n,
 					 struct intel_link_m_n *m2_n2);
 static void ironlake_set_pipeconf(struct drm_crtc *crtc);
 static void haswell_set_pipeconf(struct drm_crtc *crtc);
-static void intel_set_pipe_csc(struct drm_crtc *crtc);
+static void haswell_set_pipemisc(struct drm_crtc *crtc);
 static void vlv_prepare_pll(struct intel_crtc *crtc,
 			    const struct intel_crtc_state *pipe_config);
 static void chv_prepare_pll(struct intel_crtc *crtc,
@@ -110,13 +112,11 @@ static void intel_begin_crtc_commit(struct drm_crtc *, struct drm_crtc_state *);
 static void intel_finish_crtc_commit(struct drm_crtc *, struct drm_crtc_state *);
 static void skl_init_scalers(struct drm_device *dev, struct intel_crtc *intel_crtc,
 	struct intel_crtc_state *crtc_state);
-static int i9xx_get_refclk(const struct intel_crtc_state *crtc_state,
-			   int num_connectors);
 static void skylake_pfit_enable(struct intel_crtc *crtc);
 static void ironlake_pfit_disable(struct intel_crtc *crtc, bool force);
 static void ironlake_pfit_enable(struct intel_crtc *crtc);
 static void intel_modeset_setup_hw_state(struct drm_device *dev);
-static void intel_pre_disable_primary(struct drm_crtc *crtc);
+static void intel_pre_disable_primary_noatomic(struct drm_crtc *crtc);
 
 typedef struct {
 	int	min, max;
@@ -147,15 +147,12 @@ static int valleyview_get_vco(struct drm_i915_private *dev_priv)
 	return vco_freq[hpll_freq] * 1000;
 }
 
-static int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv,
-				  const char *name, u32 reg)
+int vlv_get_cck_clock(struct drm_i915_private *dev_priv,
+		      const char *name, u32 reg, int ref_freq)
 {
 	u32 val;
 	int divider;
 
-	if (dev_priv->hpll_freq == 0)
-		dev_priv->hpll_freq = valleyview_get_vco(dev_priv);
-
 	mutex_lock(&dev_priv->sb_lock);
 	val = vlv_cck_read(dev_priv, reg);
 	mutex_unlock(&dev_priv->sb_lock);
@@ -166,52 +163,75 @@ static int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv,
 	     (divider << CCK_FREQUENCY_STATUS_SHIFT),
 	     "%s change in progress\n", name);
 
-	return DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, divider + 1);
+	return DIV_ROUND_CLOSEST(ref_freq << 1, divider + 1);
 }
 
-int
-intel_pch_rawclk(struct drm_device *dev)
+static int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv,
+				  const char *name, u32 reg)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	if (dev_priv->hpll_freq == 0)
+		dev_priv->hpll_freq = valleyview_get_vco(dev_priv);
 
-	WARN_ON(!HAS_PCH_SPLIT(dev));
+	return vlv_get_cck_clock(dev_priv, name, reg,
+				 dev_priv->hpll_freq);
+}
 
-	return I915_READ(PCH_RAWCLK_FREQ) & RAWCLK_FREQ_MASK;
+static int
+intel_pch_rawclk(struct drm_i915_private *dev_priv)
+{
+	return (I915_READ(PCH_RAWCLK_FREQ) & RAWCLK_FREQ_MASK) * 1000;
 }
 
-/* hrawclock is 1/4 the FSB frequency */
-int intel_hrawclk(struct drm_device *dev)
+static int
+intel_vlv_hrawclk(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t clkcfg;
+	return vlv_get_cck_clock_hpll(dev_priv, "hrawclk",
+				      CCK_DISPLAY_REF_CLOCK_CONTROL);
+}
 
-	/* There is no CLKCFG reg in Valleyview. VLV hrawclk is 200 MHz */
-	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
-		return 200;
+static int
+intel_g4x_hrawclk(struct drm_i915_private *dev_priv)
+{
+	uint32_t clkcfg;
 
+	/* hrawclock is 1/4 the FSB frequency */
 	clkcfg = I915_READ(CLKCFG);
 	switch (clkcfg & CLKCFG_FSB_MASK) {
 	case CLKCFG_FSB_400:
-		return 100;
+		return 100000;
 	case CLKCFG_FSB_533:
-		return 133;
+		return 133333;
 	case CLKCFG_FSB_667:
-		return 166;
+		return 166667;
 	case CLKCFG_FSB_800:
-		return 200;
+		return 200000;
 	case CLKCFG_FSB_1067:
-		return 266;
+		return 266667;
 	case CLKCFG_FSB_1333:
-		return 333;
+		return 333333;
 	/* these two are just a guess; one of them might be right */
 	case CLKCFG_FSB_1600:
 	case CLKCFG_FSB_1600_ALT:
-		return 400;
+		return 400000;
 	default:
-		return 133;
+		return 133333;
 	}
 }
 
+static void intel_update_rawclk(struct drm_i915_private *dev_priv)
+{
+	if (HAS_PCH_SPLIT(dev_priv))
+		dev_priv->rawclk_freq = intel_pch_rawclk(dev_priv);
+	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+		dev_priv->rawclk_freq = intel_vlv_hrawclk(dev_priv);
+	else if (IS_G4X(dev_priv) || IS_PINEVIEW(dev_priv))
+		dev_priv->rawclk_freq = intel_g4x_hrawclk(dev_priv);
+	else
+		return; /* no rawclk on other platforms, or no need to know it */
+
+	DRM_DEBUG_DRIVER("rawclk rate: %d kHz\n", dev_priv->rawclk_freq);
+}
+
 static void intel_update_czclk(struct drm_i915_private *dev_priv)
 {
 	if (!(IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)))
@@ -224,13 +244,15 @@ static void intel_update_czclk(struct drm_i915_private *dev_priv)
 }
 
 static inline u32 /* units of 100MHz */
-intel_fdi_link_freq(struct drm_device *dev)
+intel_fdi_link_freq(struct drm_i915_private *dev_priv,
+		    const struct intel_crtc_state *pipe_config)
 {
-	if (IS_GEN5(dev)) {
-		struct drm_i915_private *dev_priv = dev->dev_private;
-		return (I915_READ(FDI_PLL_BIOS_0) & FDI_PLL_FB_CLOCK_MASK) + 2;
-	} else
-		return 27;
+	if (HAS_DDI(dev_priv))
+		return pipe_config->port_clock; /* SPLL */
+	else if (IS_GEN5(dev_priv))
+		return ((I915_READ(FDI_PLL_BIOS_0) & FDI_PLL_FB_CLOCK_MASK) + 2) * 10000;
+	else
+		return 270000;
 }
 
 static const intel_limit_t intel_limits_i8xx_dac = {
@@ -550,89 +572,6 @@ static bool intel_pipe_will_have_type(const struct intel_crtc_state *crtc_state,
 	return false;
 }
 
-static const intel_limit_t *
-intel_ironlake_limit(struct intel_crtc_state *crtc_state, int refclk)
-{
-	struct drm_device *dev = crtc_state->base.crtc->dev;
-	const intel_limit_t *limit;
-
-	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS)) {
-		if (intel_is_dual_link_lvds(dev)) {
-			if (refclk == 100000)
-				limit = &intel_limits_ironlake_dual_lvds_100m;
-			else
-				limit = &intel_limits_ironlake_dual_lvds;
-		} else {
-			if (refclk == 100000)
-				limit = &intel_limits_ironlake_single_lvds_100m;
-			else
-				limit = &intel_limits_ironlake_single_lvds;
-		}
-	} else
-		limit = &intel_limits_ironlake_dac;
-
-	return limit;
-}
-
-static const intel_limit_t *
-intel_g4x_limit(struct intel_crtc_state *crtc_state)
-{
-	struct drm_device *dev = crtc_state->base.crtc->dev;
-	const intel_limit_t *limit;
-
-	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS)) {
-		if (intel_is_dual_link_lvds(dev))
-			limit = &intel_limits_g4x_dual_channel_lvds;
-		else
-			limit = &intel_limits_g4x_single_channel_lvds;
-	} else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_HDMI) ||
-		   intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_ANALOG)) {
-		limit = &intel_limits_g4x_hdmi;
-	} else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_SDVO)) {
-		limit = &intel_limits_g4x_sdvo;
-	} else /* The option is for other outputs */
-		limit = &intel_limits_i9xx_sdvo;
-
-	return limit;
-}
-
-static const intel_limit_t *
-intel_limit(struct intel_crtc_state *crtc_state, int refclk)
-{
-	struct drm_device *dev = crtc_state->base.crtc->dev;
-	const intel_limit_t *limit;
-
-	if (IS_BROXTON(dev))
-		limit = &intel_limits_bxt;
-	else if (HAS_PCH_SPLIT(dev))
-		limit = intel_ironlake_limit(crtc_state, refclk);
-	else if (IS_G4X(dev)) {
-		limit = intel_g4x_limit(crtc_state);
-	} else if (IS_PINEVIEW(dev)) {
-		if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS))
-			limit = &intel_limits_pineview_lvds;
-		else
-			limit = &intel_limits_pineview_sdvo;
-	} else if (IS_CHERRYVIEW(dev)) {
-		limit = &intel_limits_chv;
-	} else if (IS_VALLEYVIEW(dev)) {
-		limit = &intel_limits_vlv;
-	} else if (!IS_GEN2(dev)) {
-		if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS))
-			limit = &intel_limits_i9xx_lvds;
-		else
-			limit = &intel_limits_i9xx_sdvo;
-	} else {
-		if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS))
-			limit = &intel_limits_i8xx_lvds;
-		else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_DVO))
-			limit = &intel_limits_i8xx_dvo;
-		else
-			limit = &intel_limits_i8xx_dac;
-	}
-	return limit;
-}
-
 /*
  * Platform specific helpers to calculate the port PLL loopback- (clock.m),
  * and post-divider (clock.p) values, pre- (clock.vco) and post-divided fast
@@ -763,6 +702,16 @@ i9xx_select_p2_div(const intel_limit_t *limit,
 	}
 }
 
+/*
+ * Returns a set of divisors for the desired target clock with the given
+ * refclk, or FALSE.  The returned values represent the clock equation:
+ * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2.
+ *
+ * Target and reference clocks are specified in kHz.
+ *
+ * If match_clock is provided, then best_clock P divider must match the P
+ * divider from @match_clock used for LVDS downclocking.
+ */
 static bool
 i9xx_find_best_dpll(const intel_limit_t *limit,
 		    struct intel_crtc_state *crtc_state,
@@ -810,6 +759,16 @@ i9xx_find_best_dpll(const intel_limit_t *limit,
 	return (err != target);
 }
 
+/*
+ * Returns a set of divisors for the desired target clock with the given
+ * refclk, or FALSE.  The returned values represent the clock equation:
+ * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2.
+ *
+ * Target and reference clocks are specified in kHz.
+ *
+ * If match_clock is provided, then best_clock P divider must match the P
+ * divider from @match_clock used for LVDS downclocking.
+ */
 static bool
 pnv_find_best_dpll(const intel_limit_t *limit,
 		   struct intel_crtc_state *crtc_state,
@@ -855,6 +814,16 @@ pnv_find_best_dpll(const intel_limit_t *limit,
 	return (err != target);
 }
 
+/*
+ * Returns a set of divisors for the desired target clock with the given
+ * refclk, or FALSE.  The returned values represent the clock equation:
+ * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2.
+ *
+ * Target and reference clocks are specified in kHz.
+ *
+ * If match_clock is provided, then best_clock P divider must match the P
+ * divider from @match_clock used for LVDS downclocking.
+ */
 static bool
 g4x_find_best_dpll(const intel_limit_t *limit,
 		   struct intel_crtc_state *crtc_state,
@@ -943,6 +912,11 @@ static bool vlv_PLL_is_optimal(struct drm_device *dev, int target_freq,
 	return *error_ppm + 10 < best_error_ppm;
 }
 
+/*
+ * Returns a set of divisors for the desired target clock with the given
+ * refclk, or FALSE.  The returned values represent the clock equation:
+ * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2.
+ */
 static bool
 vlv_find_best_dpll(const intel_limit_t *limit,
 		   struct intel_crtc_state *crtc_state,
@@ -997,6 +971,11 @@ vlv_find_best_dpll(const intel_limit_t *limit,
 	return found;
 }
 
+/*
+ * Returns a set of divisors for the desired target clock with the given
+ * refclk, or FALSE.  The returned values represent the clock equation:
+ * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2.
+ */
 static bool
 chv_find_best_dpll(const intel_limit_t *limit,
 		   struct intel_crtc_state *crtc_state,
@@ -1058,9 +1037,10 @@ chv_find_best_dpll(const intel_limit_t *limit,
 bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state, int target_clock,
 			intel_clock_t *best_clock)
 {
-	int refclk = i9xx_get_refclk(crtc_state, 0);
+	int refclk = 100000;
+	const intel_limit_t *limit = &intel_limits_bxt;
 
-	return chv_find_best_dpll(intel_limit(crtc_state, refclk), crtc_state,
+	return chv_find_best_dpll(limit, crtc_state,
 				  target_clock, refclk, NULL, best_clock);
 }
 
@@ -1165,7 +1145,7 @@ void assert_pll(struct drm_i915_private *dev_priv,
 }
 
 /* XXX: the dsi pll is shared between MIPI DSI ports */
-static void assert_dsi_pll(struct drm_i915_private *dev_priv, bool state)
+void assert_dsi_pll(struct drm_i915_private *dev_priv, bool state)
 {
 	u32 val;
 	bool cur_state;
@@ -1179,36 +1159,6 @@ static void assert_dsi_pll(struct drm_i915_private *dev_priv, bool state)
 	     "DSI PLL state assertion failure (expected %s, current %s)\n",
 			onoff(state), onoff(cur_state));
 }
-#define assert_dsi_pll_enabled(d) assert_dsi_pll(d, true)
-#define assert_dsi_pll_disabled(d) assert_dsi_pll(d, false)
-
-struct intel_shared_dpll *
-intel_crtc_to_shared_dpll(struct intel_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-
-	if (crtc->config->shared_dpll < 0)
-		return NULL;
-
-	return &dev_priv->shared_dplls[crtc->config->shared_dpll];
-}
-
-/* For ILK+ */
-void assert_shared_dpll(struct drm_i915_private *dev_priv,
-			struct intel_shared_dpll *pll,
-			bool state)
-{
-	bool cur_state;
-	struct intel_dpll_hw_state hw_state;
-
-	if (WARN(!pll, "asserting DPLL %s with no DPLL\n", onoff(state)))
-		return;
-
-	cur_state = pll->get_hw_state(dev_priv, pll, &hw_state);
-	I915_STATE_WARN(cur_state != state,
-	     "%s assertion failure (expected %s, current %s)\n",
-			pll->name, onoff(state), onoff(cur_state));
-}
 
 static void assert_fdi_tx(struct drm_i915_private *dev_priv,
 			  enum pipe pipe, bool state)
@@ -1217,7 +1167,7 @@ static void assert_fdi_tx(struct drm_i915_private *dev_priv,
 	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
 								      pipe);
 
-	if (HAS_DDI(dev_priv->dev)) {
+	if (HAS_DDI(dev_priv)) {
 		/* DDI does not have a specific FDI_TX register */
 		u32 val = I915_READ(TRANS_DDI_FUNC_CTL(cpu_transcoder));
 		cur_state = !!(val & TRANS_DDI_FUNC_ENABLE);
@@ -1253,11 +1203,11 @@ static void assert_fdi_tx_pll_enabled(struct drm_i915_private *dev_priv,
 	u32 val;
 
 	/* ILK FDI PLL is always enabled */
-	if (INTEL_INFO(dev_priv->dev)->gen == 5)
+	if (INTEL_INFO(dev_priv)->gen == 5)
 		return;
 
 	/* On Haswell, DDI ports are responsible for the FDI PLL setup */
-	if (HAS_DDI(dev_priv->dev))
+	if (HAS_DDI(dev_priv))
 		return;
 
 	val = I915_READ(FDI_TX_CTL(pipe));
@@ -1446,21 +1396,8 @@ static void assert_vblank_disabled(struct drm_crtc *crtc)
 		drm_crtc_vblank_put(crtc);
 }
 
-static void ibx_assert_pch_refclk_enabled(struct drm_i915_private *dev_priv)
-{
-	u32 val;
-	bool enabled;
-
-	I915_STATE_WARN_ON(!(HAS_PCH_IBX(dev_priv->dev) || HAS_PCH_CPT(dev_priv->dev)));
-
-	val = I915_READ(PCH_DREF_CONTROL);
-	enabled = !!(val & (DREF_SSC_SOURCE_MASK | DREF_NONSPREAD_SOURCE_MASK |
-			    DREF_SUPERSPREAD_SOURCE_MASK));
-	I915_STATE_WARN(!enabled, "PCH refclk assertion failure, should be active but is disabled\n");
-}
-
-static void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv,
-					   enum pipe pipe)
+void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv,
+				    enum pipe pipe)
 {
 	u32 val;
 	bool enabled;
@@ -1478,11 +1415,11 @@ static bool dp_pipe_enabled(struct drm_i915_private *dev_priv,
 	if ((val & DP_PORT_EN) == 0)
 		return false;
 
-	if (HAS_PCH_CPT(dev_priv->dev)) {
+	if (HAS_PCH_CPT(dev_priv)) {
 		u32 trans_dp_ctl = I915_READ(TRANS_DP_CTL(pipe));
 		if ((trans_dp_ctl & TRANS_DP_PORT_SEL_MASK) != port_sel)
 			return false;
-	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
+	} else if (IS_CHERRYVIEW(dev_priv)) {
 		if ((val & DP_PIPE_MASK_CHV) != DP_PIPE_SELECT_CHV(pipe))
 			return false;
 	} else {
@@ -1498,10 +1435,10 @@ static bool hdmi_pipe_enabled(struct drm_i915_private *dev_priv,
 	if ((val & SDVO_ENABLE) == 0)
 		return false;
 
-	if (HAS_PCH_CPT(dev_priv->dev)) {
+	if (HAS_PCH_CPT(dev_priv)) {
 		if ((val & SDVO_PIPE_SEL_MASK_CPT) != SDVO_PIPE_SEL_CPT(pipe))
 			return false;
-	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
+	} else if (IS_CHERRYVIEW(dev_priv)) {
 		if ((val & SDVO_PIPE_SEL_MASK_CHV) != SDVO_PIPE_SEL_CHV(pipe))
 			return false;
 	} else {
@@ -1517,7 +1454,7 @@ static bool lvds_pipe_enabled(struct drm_i915_private *dev_priv,
 	if ((val & LVDS_PORT_EN) == 0)
 		return false;
 
-	if (HAS_PCH_CPT(dev_priv->dev)) {
+	if (HAS_PCH_CPT(dev_priv)) {
 		if ((val & PORT_TRANS_SEL_MASK) != PORT_TRANS_SEL_CPT(pipe))
 			return false;
 	} else {
@@ -1532,7 +1469,7 @@ static bool adpa_pipe_enabled(struct drm_i915_private *dev_priv,
 {
 	if ((val & ADPA_DAC_ENABLE) == 0)
 		return false;
-	if (HAS_PCH_CPT(dev_priv->dev)) {
+	if (HAS_PCH_CPT(dev_priv)) {
 		if ((val & PORT_TRANS_SEL_MASK) != PORT_TRANS_SEL_CPT(pipe))
 			return false;
 	} else {
@@ -1551,7 +1488,7 @@ static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv,
 	     "PCH DP (0x%08x) enabled on transcoder %c, should be disabled\n",
 	     i915_mmio_reg_offset(reg), pipe_name(pipe));
 
-	I915_STATE_WARN(HAS_PCH_IBX(dev_priv->dev) && (val & DP_PORT_EN) == 0
+	I915_STATE_WARN(HAS_PCH_IBX(dev_priv) && (val & DP_PORT_EN) == 0
 	     && (val & DP_PIPEB_SELECT),
 	     "IBX PCH dp port still using transcoder B\n");
 }
@@ -1564,7 +1501,7 @@ static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv,
 	     "PCH HDMI (0x%08x) enabled on transcoder %c, should be disabled\n",
 	     i915_mmio_reg_offset(reg), pipe_name(pipe));
 
-	I915_STATE_WARN(HAS_PCH_IBX(dev_priv->dev) && (val & SDVO_ENABLE) == 0
+	I915_STATE_WARN(HAS_PCH_IBX(dev_priv) && (val & SDVO_ENABLE) == 0
 	     && (val & SDVO_PIPE_B_SELECT),
 	     "IBX PCH hdmi port still using transcoder B\n");
 }
@@ -1593,53 +1530,47 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
 	assert_pch_hdmi_disabled(dev_priv, pipe, PCH_HDMID);
 }
 
+static void _vlv_enable_pll(struct intel_crtc *crtc,
+			    const struct intel_crtc_state *pipe_config)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+
+	I915_WRITE(DPLL(pipe), pipe_config->dpll_hw_state.dpll);
+	POSTING_READ(DPLL(pipe));
+	udelay(150);
+
+	if (wait_for(((I915_READ(DPLL(pipe)) & DPLL_LOCK_VLV) == DPLL_LOCK_VLV), 1))
+		DRM_ERROR("DPLL %d failed to lock\n", pipe);
+}
+
 static void vlv_enable_pll(struct intel_crtc *crtc,
 			   const struct intel_crtc_state *pipe_config)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	i915_reg_t reg = DPLL(crtc->pipe);
-	u32 dpll = pipe_config->dpll_hw_state.dpll;
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
 
-	assert_pipe_disabled(dev_priv, crtc->pipe);
+	assert_pipe_disabled(dev_priv, pipe);
 
 	/* PLL is protected by panel, make sure we can write it */
-	if (IS_MOBILE(dev_priv->dev))
-		assert_panel_unlocked(dev_priv, crtc->pipe);
-
-	I915_WRITE(reg, dpll);
-	POSTING_READ(reg);
-	udelay(150);
-
-	if (wait_for(((I915_READ(reg) & DPLL_LOCK_VLV) == DPLL_LOCK_VLV), 1))
-		DRM_ERROR("DPLL %d failed to lock\n", crtc->pipe);
+	assert_panel_unlocked(dev_priv, pipe);
 
-	I915_WRITE(DPLL_MD(crtc->pipe), pipe_config->dpll_hw_state.dpll_md);
-	POSTING_READ(DPLL_MD(crtc->pipe));
+	if (pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE)
+		_vlv_enable_pll(crtc, pipe_config);
 
-	/* We do this three times for luck */
-	I915_WRITE(reg, dpll);
-	POSTING_READ(reg);
-	udelay(150); /* wait for warmup */
-	I915_WRITE(reg, dpll);
-	POSTING_READ(reg);
-	udelay(150); /* wait for warmup */
-	I915_WRITE(reg, dpll);
-	POSTING_READ(reg);
-	udelay(150); /* wait for warmup */
+	I915_WRITE(DPLL_MD(pipe), pipe_config->dpll_hw_state.dpll_md);
+	POSTING_READ(DPLL_MD(pipe));
 }
 
-static void chv_enable_pll(struct intel_crtc *crtc,
-			   const struct intel_crtc_state *pipe_config)
+
+static void _chv_enable_pll(struct intel_crtc *crtc,
+			    const struct intel_crtc_state *pipe_config)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe = crtc->pipe;
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
 	enum dpio_channel port = vlv_pipe_to_channel(pipe);
 	u32 tmp;
 
-	assert_pipe_disabled(dev_priv, crtc->pipe);
-
 	mutex_lock(&dev_priv->sb_lock);
 
 	/* Enable back the 10bit clock to display controller */
@@ -1660,10 +1591,43 @@ static void chv_enable_pll(struct intel_crtc *crtc,
 	/* Check PLL is locked */
 	if (wait_for(((I915_READ(DPLL(pipe)) & DPLL_LOCK_VLV) == DPLL_LOCK_VLV), 1))
 		DRM_ERROR("PLL %d failed to lock\n", pipe);
+}
 
-	/* not sure when this should be written */
-	I915_WRITE(DPLL_MD(pipe), pipe_config->dpll_hw_state.dpll_md);
-	POSTING_READ(DPLL_MD(pipe));
+static void chv_enable_pll(struct intel_crtc *crtc,
+			   const struct intel_crtc_state *pipe_config)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+
+	assert_pipe_disabled(dev_priv, pipe);
+
+	/* PLL is protected by panel, make sure we can write it */
+	assert_panel_unlocked(dev_priv, pipe);
+
+	if (pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE)
+		_chv_enable_pll(crtc, pipe_config);
+
+	if (pipe != PIPE_A) {
+		/*
+		 * WaPixelRepeatModeFixForC0:chv
+		 *
+		 * DPLLCMD is AWOL. Use chicken bits to propagate
+		 * the value from DPLLBMD to either pipe B or C.
+		 */
+		I915_WRITE(CBR4_VLV, pipe == PIPE_B ? CBR_DPLLBMD_PIPE_B : CBR_DPLLBMD_PIPE_C);
+		I915_WRITE(DPLL_MD(PIPE_B), pipe_config->dpll_hw_state.dpll_md);
+		I915_WRITE(CBR4_VLV, 0);
+		dev_priv->chv_dpll_md[pipe] = pipe_config->dpll_hw_state.dpll_md;
+
+		/*
+		 * DPLLB VGA mode also seems to cause problems.
+		 * We should always have it disabled.
+		 */
+		WARN_ON((I915_READ(DPLL(PIPE_B)) & DPLL_VGA_MODE_DIS) == 0);
+	} else {
+		I915_WRITE(DPLL_MD(pipe), pipe_config->dpll_hw_state.dpll_md);
+		POSTING_READ(DPLL_MD(pipe));
+	}
 }
 
 static int intel_num_dvo_pipes(struct drm_device *dev)
@@ -1687,9 +1651,6 @@ static void i9xx_enable_pll(struct intel_crtc *crtc)
 
 	assert_pipe_disabled(dev_priv, crtc->pipe);
 
-	/* No really, not for ILK+ */
-	BUG_ON(INTEL_INFO(dev)->gen >= 5);
-
 	/* PLL is protected by panel, make sure we can write it */
 	if (IS_MOBILE(dev) && !IS_I830(dev))
 		assert_panel_unlocked(dev_priv, crtc->pipe);
@@ -1788,16 +1749,13 @@ static void vlv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 	/* Make sure the pipe isn't still relying on us */
 	assert_pipe_disabled(dev_priv, pipe);
 
-	/*
-	 * Leave integrated clock source and reference clock enabled for pipe B.
-	 * The latter is needed for VGA hotplug / manual detection.
-	 */
-	val = DPLL_VGA_MODE_DIS;
-	if (pipe == PIPE_B)
-		val = DPLL_INTEGRATED_CRI_CLK_VLV | DPLL_REF_CLK_ENABLE_VLV;
+	val = DPLL_INTEGRATED_REF_CLK_VLV |
+		DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
+	if (pipe != PIPE_A)
+		val |= DPLL_INTEGRATED_CRI_CLK_VLV;
+
 	I915_WRITE(DPLL(pipe), val);
 	POSTING_READ(DPLL(pipe));
-
 }
 
 static void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
@@ -1808,11 +1766,11 @@ static void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 	/* Make sure the pipe isn't still relying on us */
 	assert_pipe_disabled(dev_priv, pipe);
 
-	/* Set PLL en = 0 */
 	val = DPLL_SSC_REF_CLK_CHV |
 		DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
 	if (pipe != PIPE_A)
 		val |= DPLL_INTEGRATED_CRI_CLK_VLV;
+
 	I915_WRITE(DPLL(pipe), val);
 	POSTING_READ(DPLL(pipe));
 
@@ -1856,100 +1814,6 @@ void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 		     port_name(dport->port), I915_READ(dpll_reg) & port_mask, expected_mask);
 }
 
-static void intel_prepare_shared_dpll(struct intel_crtc *crtc)
-{
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
-
-	if (WARN_ON(pll == NULL))
-		return;
-
-	WARN_ON(!pll->config.crtc_mask);
-	if (pll->active == 0) {
-		DRM_DEBUG_DRIVER("setting up %s\n", pll->name);
-		WARN_ON(pll->on);
-		assert_shared_dpll_disabled(dev_priv, pll);
-
-		pll->mode_set(dev_priv, pll);
-	}
-}
-
-/**
- * intel_enable_shared_dpll - enable PCH PLL
- * @dev_priv: i915 private structure
- * @pipe: pipe PLL to enable
- *
- * The PCH PLL needs to be enabled before the PCH transcoder, since it
- * drives the transcoder clock.
- */
-static void intel_enable_shared_dpll(struct intel_crtc *crtc)
-{
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
-
-	if (WARN_ON(pll == NULL))
-		return;
-
-	if (WARN_ON(pll->config.crtc_mask == 0))
-		return;
-
-	DRM_DEBUG_KMS("enable %s (active %d, on? %d) for crtc %d\n",
-		      pll->name, pll->active, pll->on,
-		      crtc->base.base.id);
-
-	if (pll->active++) {
-		WARN_ON(!pll->on);
-		assert_shared_dpll_enabled(dev_priv, pll);
-		return;
-	}
-	WARN_ON(pll->on);
-
-	intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
-
-	DRM_DEBUG_KMS("enabling %s\n", pll->name);
-	pll->enable(dev_priv, pll);
-	pll->on = true;
-}
-
-static void intel_disable_shared_dpll(struct intel_crtc *crtc)
-{
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
-
-	/* PCH only available on ILK+ */
-	if (INTEL_INFO(dev)->gen < 5)
-		return;
-
-	if (pll == NULL)
-		return;
-
-	if (WARN_ON(!(pll->config.crtc_mask & (1 << drm_crtc_index(&crtc->base)))))
-		return;
-
-	DRM_DEBUG_KMS("disable %s (active %d, on? %d) for crtc %d\n",
-		      pll->name, pll->active, pll->on,
-		      crtc->base.base.id);
-
-	if (WARN_ON(pll->active == 0)) {
-		assert_shared_dpll_disabled(dev_priv, pll);
-		return;
-	}
-
-	assert_shared_dpll_enabled(dev_priv, pll);
-	WARN_ON(!pll->on);
-	if (--pll->active)
-		return;
-
-	DRM_DEBUG_KMS("disabling %s\n", pll->name);
-	pll->disable(dev_priv, pll);
-	pll->on = false;
-
-	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
-}
-
 static void ironlake_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 					   enum pipe pipe)
 {
@@ -1959,12 +1823,8 @@ static void ironlake_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 	i915_reg_t reg;
 	uint32_t val, pipeconf_val;
 
-	/* PCH only available on ILK+ */
-	BUG_ON(!HAS_PCH_SPLIT(dev));
-
 	/* Make sure PCH DPLL is enabled */
-	assert_shared_dpll_enabled(dev_priv,
-				   intel_crtc_to_shared_dpll(intel_crtc));
+	assert_shared_dpll_enabled(dev_priv, intel_crtc->config->shared_dpll);
 
 	/* FDI must be feeding us bits for PCH ports */
 	assert_fdi_tx_enabled(dev_priv, pipe);
@@ -1983,7 +1843,7 @@ static void ironlake_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 	val = I915_READ(reg);
 	pipeconf_val = I915_READ(PIPECONF(pipe));
 
-	if (HAS_PCH_IBX(dev_priv->dev)) {
+	if (HAS_PCH_IBX(dev_priv)) {
 		/*
 		 * Make the BPC in transcoder be consistent with
 		 * that in pipeconf reg. For HDMI we must use 8bpc
@@ -1998,7 +1858,7 @@ static void ironlake_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 
 	val &= ~TRANS_INTERLACE_MASK;
 	if ((pipeconf_val & PIPECONF_INTERLACE_MASK) == PIPECONF_INTERLACED_ILK)
-		if (HAS_PCH_IBX(dev_priv->dev) &&
+		if (HAS_PCH_IBX(dev_priv) &&
 		    intel_pipe_has_type(intel_crtc, INTEL_OUTPUT_SDVO))
 			val |= TRANS_LEGACY_INTERLACED_ILK;
 		else
@@ -2016,9 +1876,6 @@ static void lpt_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 {
 	u32 val, pipeconf_val;
 
-	/* PCH only available on ILK+ */
-	BUG_ON(!HAS_PCH_SPLIT(dev_priv->dev));
-
 	/* FDI must be feeding us bits for PCH ports */
 	assert_fdi_tx_enabled(dev_priv, (enum pipe) cpu_transcoder);
 	assert_fdi_rx_enabled(dev_priv, TRANSCODER_A);
@@ -2113,7 +1970,7 @@ static void intel_enable_pipe(struct intel_crtc *crtc)
 	assert_cursor_disabled(dev_priv, pipe);
 	assert_sprites_disabled(dev_priv, pipe);
 
-	if (HAS_PCH_LPT(dev_priv->dev))
+	if (HAS_PCH_LPT(dev_priv))
 		pch_transcoder = TRANSCODER_A;
 	else
 		pch_transcoder = pipe;
@@ -2123,7 +1980,7 @@ static void intel_enable_pipe(struct intel_crtc *crtc)
 	 * a plane.  On ILK+ the pipe PLLs are integrated, so we don't
 	 * need the check.
 	 */
-	if (HAS_GMCH_DISPLAY(dev_priv->dev))
+	if (HAS_GMCH_DISPLAY(dev_priv))
 		if (crtc->config->has_dsi_encoder)
 			assert_dsi_pll_enabled(dev_priv);
 		else
@@ -2225,8 +2082,8 @@ static unsigned int intel_tile_size(const struct drm_i915_private *dev_priv)
 	return IS_GEN2(dev_priv) ? 2048 : 4096;
 }
 
-static unsigned int intel_tile_width(const struct drm_i915_private *dev_priv,
-				     uint64_t fb_modifier, unsigned int cpp)
+static unsigned int intel_tile_width_bytes(const struct drm_i915_private *dev_priv,
+					   uint64_t fb_modifier, unsigned int cpp)
 {
 	switch (fb_modifier) {
 	case DRM_FORMAT_MOD_NONE:
@@ -2269,7 +2126,21 @@ unsigned int intel_tile_height(const struct drm_i915_private *dev_priv,
 		return 1;
 	else
 		return intel_tile_size(dev_priv) /
-			intel_tile_width(dev_priv, fb_modifier, cpp);
+			intel_tile_width_bytes(dev_priv, fb_modifier, cpp);
+}
+
+/* Return the tile dimensions in pixel units */
+static void intel_tile_dims(const struct drm_i915_private *dev_priv,
+			    unsigned int *tile_width,
+			    unsigned int *tile_height,
+			    uint64_t fb_modifier,
+			    unsigned int cpp)
+{
+	unsigned int tile_width_bytes =
+		intel_tile_width_bytes(dev_priv, fb_modifier, cpp);
+
+	*tile_width = tile_width_bytes / cpp;
+	*tile_height = intel_tile_size(dev_priv) / tile_width_bytes;
 }
 
 unsigned int
@@ -2282,48 +2153,54 @@ intel_fb_align_height(struct drm_device *dev, unsigned int height,
 	return ALIGN(height, tile_height);
 }
 
-static void
-intel_fill_fb_ggtt_view(struct i915_ggtt_view *view, struct drm_framebuffer *fb,
-			const struct drm_plane_state *plane_state)
+unsigned int intel_rotation_info_size(const struct intel_rotation_info *rot_info)
 {
-	struct drm_i915_private *dev_priv = to_i915(fb->dev);
-	struct intel_rotation_info *info = &view->params.rotated;
-	unsigned int tile_size, tile_width, tile_height, cpp;
-
-	*view = i915_ggtt_view_normal;
+	unsigned int size = 0;
+	int i;
 
-	if (!plane_state)
-		return;
+	for (i = 0 ; i < ARRAY_SIZE(rot_info->plane); i++)
+		size += rot_info->plane[i].width * rot_info->plane[i].height;
 
-	if (!intel_rotation_90_or_270(plane_state->rotation))
-		return;
+	return size;
+}
 
-	*view = i915_ggtt_view_rotated;
+static void
+intel_fill_fb_ggtt_view(struct i915_ggtt_view *view,
+			const struct drm_framebuffer *fb,
+			unsigned int rotation)
+{
+	if (intel_rotation_90_or_270(rotation)) {
+		*view = i915_ggtt_view_rotated;
+		view->params.rotated = to_intel_framebuffer(fb)->rot_info;
+	} else {
+		*view = i915_ggtt_view_normal;
+	}
+}
 
-	info->height = fb->height;
-	info->pixel_format = fb->pixel_format;
-	info->pitch = fb->pitches[0];
-	info->uv_offset = fb->offsets[1];
-	info->fb_modifier = fb->modifier[0];
+static void
+intel_fill_fb_info(struct drm_i915_private *dev_priv,
+		   struct drm_framebuffer *fb)
+{
+	struct intel_rotation_info *info = &to_intel_framebuffer(fb)->rot_info;
+	unsigned int tile_size, tile_width, tile_height, cpp;
 
 	tile_size = intel_tile_size(dev_priv);
 
 	cpp = drm_format_plane_cpp(fb->pixel_format, 0);
-	tile_width = intel_tile_width(dev_priv, fb->modifier[0], cpp);
-	tile_height = tile_size / tile_width;
+	intel_tile_dims(dev_priv, &tile_width, &tile_height,
+			fb->modifier[0], cpp);
 
-	info->width_pages = DIV_ROUND_UP(fb->pitches[0], tile_width);
-	info->height_pages = DIV_ROUND_UP(fb->height, tile_height);
-	info->size = info->width_pages * info->height_pages * tile_size;
+	info->plane[0].width = DIV_ROUND_UP(fb->pitches[0], tile_width * cpp);
+	info->plane[0].height = DIV_ROUND_UP(fb->height, tile_height);
 
 	if (info->pixel_format == DRM_FORMAT_NV12) {
 		cpp = drm_format_plane_cpp(fb->pixel_format, 1);
-		tile_width = intel_tile_width(dev_priv, fb->modifier[1], cpp);
-		tile_height = tile_size / tile_width;
+		intel_tile_dims(dev_priv, &tile_width, &tile_height,
+				fb->modifier[1], cpp);
 
-		info->width_pages_uv = DIV_ROUND_UP(fb->pitches[1], tile_width);
-		info->height_pages_uv = DIV_ROUND_UP(fb->height / 2, tile_height);
-		info->size_uv = info->width_pages_uv * info->height_pages_uv * tile_size;
+		info->uv_offset = fb->offsets[1];
+		info->plane[1].width = DIV_ROUND_UP(fb->pitches[1], tile_width * cpp);
+		info->plane[1].height = DIV_ROUND_UP(fb->height / 2, tile_height);
 	}
 }
 
@@ -2360,9 +2237,8 @@ static unsigned int intel_surf_alignment(const struct drm_i915_private *dev_priv
 }
 
 int
-intel_pin_and_fence_fb_obj(struct drm_plane *plane,
-			   struct drm_framebuffer *fb,
-			   const struct drm_plane_state *plane_state)
+intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb,
+			   unsigned int rotation)
 {
 	struct drm_device *dev = fb->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2375,7 +2251,7 @@ intel_pin_and_fence_fb_obj(struct drm_plane *plane,
 
 	alignment = intel_surf_alignment(dev_priv, fb->modifier[0]);
 
-	intel_fill_fb_ggtt_view(&view, fb, plane_state);
+	intel_fill_fb_ggtt_view(&view, fb, rotation);
 
 	/* Note that the w/a also requires 64 PTE of padding following the
 	 * bo. We currently fill all unused PTE with the shadow page and so
@@ -2433,15 +2309,14 @@ err_pm:
 	return ret;
 }
 
-static void intel_unpin_fb_obj(struct drm_framebuffer *fb,
-			       const struct drm_plane_state *plane_state)
+static void intel_unpin_fb_obj(struct drm_framebuffer *fb, unsigned int rotation)
 {
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	struct i915_ggtt_view view;
 
 	WARN_ON(!mutex_is_locked(&obj->base.dev->struct_mutex));
 
-	intel_fill_fb_ggtt_view(&view, fb, plane_state);
+	intel_fill_fb_ggtt_view(&view, fb, rotation);
 
 	if (view.type == I915_GGTT_VIEW_NORMAL)
 		i915_gem_object_unpin_fence(obj);
@@ -2449,38 +2324,93 @@ static void intel_unpin_fb_obj(struct drm_framebuffer *fb,
 	i915_gem_object_unpin_from_display_plane(obj, &view);
 }
 
-/* Computes the linear offset to the base tile and adjusts x, y. bytes per pixel
- * is assumed to be a power-of-two. */
-u32 intel_compute_tile_offset(struct drm_i915_private *dev_priv,
-			      int *x, int *y,
-			      uint64_t fb_modifier,
-			      unsigned int cpp,
-			      unsigned int pitch)
+/*
+ * Adjust the tile offset by moving the difference into
+ * the x/y offsets.
+ *
+ * Input tile dimensions and pitch must already be
+ * rotated to match x and y, and in pixel units.
+ */
+static u32 intel_adjust_tile_offset(int *x, int *y,
+				    unsigned int tile_width,
+				    unsigned int tile_height,
+				    unsigned int tile_size,
+				    unsigned int pitch_tiles,
+				    u32 old_offset,
+				    u32 new_offset)
 {
+	unsigned int tiles;
+
+	WARN_ON(old_offset & (tile_size - 1));
+	WARN_ON(new_offset & (tile_size - 1));
+	WARN_ON(new_offset > old_offset);
+
+	tiles = (old_offset - new_offset) / tile_size;
+
+	*y += tiles / pitch_tiles * tile_height;
+	*x += tiles % pitch_tiles * tile_width;
+
+	return new_offset;
+}
+
+/*
+ * Computes the linear offset to the base tile and adjusts
+ * x, y. bytes per pixel is assumed to be a power-of-two.
+ *
+ * In the 90/270 rotated case, x and y are assumed
+ * to be already rotated to match the rotated GTT view, and
+ * pitch is the tile_height aligned framebuffer height.
+ */
+u32 intel_compute_tile_offset(int *x, int *y,
+			      const struct drm_framebuffer *fb, int plane,
+			      unsigned int pitch,
+			      unsigned int rotation)
+{
+	const struct drm_i915_private *dev_priv = to_i915(fb->dev);
+	uint64_t fb_modifier = fb->modifier[plane];
+	unsigned int cpp = drm_format_plane_cpp(fb->pixel_format, plane);
+	u32 offset, offset_aligned, alignment;
+
+	alignment = intel_surf_alignment(dev_priv, fb_modifier);
+	if (alignment)
+		alignment--;
+
 	if (fb_modifier != DRM_FORMAT_MOD_NONE) {
 		unsigned int tile_size, tile_width, tile_height;
-		unsigned int tile_rows, tiles;
+		unsigned int tile_rows, tiles, pitch_tiles;
 
 		tile_size = intel_tile_size(dev_priv);
-		tile_width = intel_tile_width(dev_priv, fb_modifier, cpp);
-		tile_height = tile_size / tile_width;
+		intel_tile_dims(dev_priv, &tile_width, &tile_height,
+				fb_modifier, cpp);
+
+		if (intel_rotation_90_or_270(rotation)) {
+			pitch_tiles = pitch / tile_height;
+			swap(tile_width, tile_height);
+		} else {
+			pitch_tiles = pitch / (tile_width * cpp);
+		}
 
 		tile_rows = *y / tile_height;
 		*y %= tile_height;
 
-		tiles = *x / (tile_width/cpp);
-		*x %= tile_width/cpp;
+		tiles = *x / tile_width;
+		*x %= tile_width;
 
-		return tile_rows * pitch * tile_height + tiles * tile_size;
-	} else {
-		unsigned int alignment = intel_linear_alignment(dev_priv) - 1;
-		unsigned int offset;
+		offset = (tile_rows * pitch_tiles + tiles) * tile_size;
+		offset_aligned = offset & ~alignment;
 
+		intel_adjust_tile_offset(x, y, tile_width, tile_height,
+					 tile_size, pitch_tiles,
+					 offset, offset_aligned);
+	} else {
 		offset = *y * pitch + *x * cpp;
+		offset_aligned = offset & ~alignment;
+
 		*y = (offset & alignment) / pitch;
 		*x = ((offset & alignment) - *y * pitch) / cpp;
-		return offset & ~alignment;
 	}
+
+	return offset_aligned;
 }
 
 static int i9xx_format_to_fourcc(int format)
@@ -2536,6 +2466,7 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_object *obj = NULL;
 	struct drm_mode_fb_cmd2 mode_cmd = { 0 };
 	struct drm_framebuffer *fb = &plane_config->fb->base;
@@ -2551,7 +2482,7 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 	/* If the FB is too big, just don't use it since fbdev is not very
 	 * important and we should probably use that space with FBC or other
 	 * features. */
-	if (size_aligned * 2 > dev_priv->gtt.stolen_usable_size)
+	if (size_aligned * 2 > ggtt->stolen_usable_size)
 		return false;
 
 	mutex_lock(&dev->struct_mutex);
@@ -2667,7 +2598,7 @@ intel_find_initial_plane_obj(struct intel_crtc *intel_crtc,
 	 */
 	to_intel_plane_state(plane_state)->visible = false;
 	crtc_state->plane_mask &= ~(1 << drm_plane_index(primary));
-	intel_pre_disable_primary(&intel_crtc->base);
+	intel_pre_disable_primary_noatomic(&intel_crtc->base);
 	intel_plane->disable_plane(primary, &intel_crtc->base);
 
 	return;
@@ -2716,6 +2647,7 @@ static void i9xx_update_primary_plane(struct drm_plane *primary,
 	u32 linear_offset;
 	u32 dspcntr;
 	i915_reg_t reg = DSPCNTR(plane);
+	unsigned int rotation = plane_state->base.rotation;
 	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	int x = plane_state->src.x1 >> 16;
 	int y = plane_state->src.y1 >> 16;
@@ -2780,15 +2712,14 @@ static void i9xx_update_primary_plane(struct drm_plane *primary,
 
 	if (INTEL_INFO(dev)->gen >= 4) {
 		intel_crtc->dspaddr_offset =
-			intel_compute_tile_offset(dev_priv, &x, &y,
-						  fb->modifier[0], cpp,
-						  fb->pitches[0]);
+			intel_compute_tile_offset(&x, &y, fb, 0,
+						  fb->pitches[0], rotation);
 		linear_offset -= intel_crtc->dspaddr_offset;
 	} else {
 		intel_crtc->dspaddr_offset = linear_offset;
 	}
 
-	if (plane_state->base.rotation == BIT(DRM_ROTATE_180)) {
+	if (rotation == BIT(DRM_ROTATE_180)) {
 		dspcntr |= DISPPLANE_ROTATE_180;
 
 		x += (crtc_state->pipe_src_w - 1);
@@ -2846,6 +2777,7 @@ static void ironlake_update_primary_plane(struct drm_plane *primary,
 	u32 linear_offset;
 	u32 dspcntr;
 	i915_reg_t reg = DSPCNTR(plane);
+	unsigned int rotation = plane_state->base.rotation;
 	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	int x = plane_state->src.x1 >> 16;
 	int y = plane_state->src.y1 >> 16;
@@ -2887,11 +2819,10 @@ static void ironlake_update_primary_plane(struct drm_plane *primary,
 
 	linear_offset = y * fb->pitches[0] + x * cpp;
 	intel_crtc->dspaddr_offset =
-		intel_compute_tile_offset(dev_priv, &x, &y,
-					  fb->modifier[0], cpp,
-					  fb->pitches[0]);
+		intel_compute_tile_offset(&x, &y, fb, 0,
+					  fb->pitches[0], rotation);
 	linear_offset -= intel_crtc->dspaddr_offset;
-	if (plane_state->base.rotation == BIT(DRM_ROTATE_180)) {
+	if (rotation == BIT(DRM_ROTATE_180)) {
 		dspcntr |= DISPPLANE_ROTATE_180;
 
 		if (!IS_HASWELL(dev) && !IS_BROADWELL(dev)) {
@@ -2931,7 +2862,7 @@ u32 intel_fb_stride_alignment(const struct drm_i915_private *dev_priv,
 	} else {
 		int cpp = drm_format_plane_cpp(pixel_format, 0);
 
-		return intel_tile_width(dev_priv, fb_modifier, cpp);
+		return intel_tile_width_bytes(dev_priv, fb_modifier, cpp);
 	}
 }
 
@@ -2944,7 +2875,7 @@ u32 intel_plane_obj_offset(struct intel_plane *intel_plane,
 	u64 offset;
 
 	intel_fill_fb_ggtt_view(&view, intel_plane->base.state->fb,
-				intel_plane->base.state);
+				intel_plane->base.state->rotation);
 
 	vma = i915_gem_obj_to_ggtt_view(obj, &view);
 	if (WARN(!vma, "ggtt vma for display object not found! (view=%u)\n",
@@ -3284,12 +3215,12 @@ void intel_finish_reset(struct drm_device *dev)
 static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	unsigned reset_counter;
 	bool pending;
 
-	if (i915_reset_in_progress(&dev_priv->gpu_error) ||
-	    intel_crtc->reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
+	reset_counter = i915_reset_counter(&to_i915(dev)->gpu_error);
+	if (intel_crtc->reset_counter != reset_counter)
 		return false;
 
 	spin_lock_irq(&dev->event_lock);
@@ -3314,9 +3245,6 @@ static void intel_update_pipe_config(struct intel_crtc *crtc,
 		      old_crtc_state->pipe_src_w, old_crtc_state->pipe_src_h,
 		      pipe_config->pipe_src_w, pipe_config->pipe_src_h);
 
-	if (HAS_DDI(dev))
-		intel_set_pipe_csc(&crtc->base);
-
 	/*
 	 * Update pipe size and adjust fitter if needed: the reason for this is
 	 * that in compute_mode_changes we check the native mode (not the pfit
@@ -3894,9 +3822,7 @@ static void page_flip_completed(struct intel_crtc *intel_crtc)
 	intel_crtc->unpin_work = NULL;
 
 	if (work->event)
-		drm_send_vblank_event(intel_crtc->base.dev,
-				      intel_crtc->pipe,
-				      work->event);
+		drm_crtc_send_vblank_event(&intel_crtc->base, work->event);
 
 	drm_crtc_vblank_put(&intel_crtc->base);
 
@@ -3955,37 +3881,35 @@ static void lpt_disable_iclkip(struct drm_i915_private *dev_priv)
 /* Program iCLKIP clock to the desired frequency */
 static void lpt_program_iclkip(struct drm_crtc *crtc)
 {
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
 	int clock = to_intel_crtc(crtc)->config->base.adjusted_mode.crtc_clock;
 	u32 divsel, phaseinc, auxdiv, phasedir = 0;
 	u32 temp;
 
 	lpt_disable_iclkip(dev_priv);
 
-	/* 20MHz is a corner case which is out of range for the 7-bit divisor */
-	if (clock == 20000) {
-		auxdiv = 1;
-		divsel = 0x41;
-		phaseinc = 0x20;
-	} else {
-		/* The iCLK virtual clock root frequency is in MHz,
-		 * but the adjusted_mode->crtc_clock in in KHz. To get the
-		 * divisors, it is necessary to divide one by another, so we
-		 * convert the virtual clock precision to KHz here for higher
-		 * precision.
-		 */
+	/* The iCLK virtual clock root frequency is in MHz,
+	 * but the adjusted_mode->crtc_clock in in KHz. To get the
+	 * divisors, it is necessary to divide one by another, so we
+	 * convert the virtual clock precision to KHz here for higher
+	 * precision.
+	 */
+	for (auxdiv = 0; auxdiv < 2; auxdiv++) {
 		u32 iclk_virtual_root_freq = 172800 * 1000;
 		u32 iclk_pi_range = 64;
-		u32 desired_divisor, msb_divisor_value, pi_value;
+		u32 desired_divisor;
 
-		desired_divisor = DIV_ROUND_CLOSEST(iclk_virtual_root_freq, clock);
-		msb_divisor_value = desired_divisor / iclk_pi_range;
-		pi_value = desired_divisor % iclk_pi_range;
+		desired_divisor = DIV_ROUND_CLOSEST(iclk_virtual_root_freq,
+						    clock << auxdiv);
+		divsel = (desired_divisor / iclk_pi_range) - 2;
+		phaseinc = desired_divisor % iclk_pi_range;
 
-		auxdiv = 0;
-		divsel = msb_divisor_value - 2;
-		phaseinc = pi_value;
+		/*
+		 * Near 20MHz is a corner case which is
+		 * out of range for the 7-bit divisor
+		 */
+		if (divsel <= 0x7f)
+			break;
 	}
 
 	/* This should not happen with any sane values */
@@ -4032,6 +3956,43 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 	I915_WRITE(PIXCLK_GATE, PIXCLK_GATE_UNGATE);
 }
 
+int lpt_get_iclkip(struct drm_i915_private *dev_priv)
+{
+	u32 divsel, phaseinc, auxdiv;
+	u32 iclk_virtual_root_freq = 172800 * 1000;
+	u32 iclk_pi_range = 64;
+	u32 desired_divisor;
+	u32 temp;
+
+	if ((I915_READ(PIXCLK_GATE) & PIXCLK_GATE_UNGATE) == 0)
+		return 0;
+
+	mutex_lock(&dev_priv->sb_lock);
+
+	temp = intel_sbi_read(dev_priv, SBI_SSCCTL6, SBI_ICLK);
+	if (temp & SBI_SSCCTL_DISABLE) {
+		mutex_unlock(&dev_priv->sb_lock);
+		return 0;
+	}
+
+	temp = intel_sbi_read(dev_priv, SBI_SSCDIVINTPHASE6, SBI_ICLK);
+	divsel = (temp & SBI_SSCDIVINTPHASE_DIVSEL_MASK) >>
+		SBI_SSCDIVINTPHASE_DIVSEL_SHIFT;
+	phaseinc = (temp & SBI_SSCDIVINTPHASE_INCVAL_MASK) >>
+		SBI_SSCDIVINTPHASE_INCVAL_SHIFT;
+
+	temp = intel_sbi_read(dev_priv, SBI_SSCAUXDIV6, SBI_ICLK);
+	auxdiv = (temp & SBI_SSCAUXDIV_FINALDIV2SEL_MASK) >>
+		SBI_SSCAUXDIV_FINALDIV2SEL_SHIFT;
+
+	mutex_unlock(&dev_priv->sb_lock);
+
+	desired_divisor = (divsel + 2) * iclk_pi_range + phaseinc;
+
+	return DIV_ROUND_CLOSEST(iclk_virtual_root_freq,
+				 desired_divisor << auxdiv);
+}
+
 static void ironlake_pch_transcoder_set_timings(struct intel_crtc *crtc,
 						enum pipe pch_transcoder)
 {
@@ -4142,12 +4103,6 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 	I915_WRITE(FDI_RX_TUSIZE1(pipe),
 		   I915_READ(PIPE_DATA_M1(pipe)) & TU_SIZE_MASK);
 
-	/*
-	 * Sometimes spurious CPU pipe underruns happen during FDI
-	 * training, at least with VGA+HDMI cloning. Suppress them.
-	 */
-	intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
-
 	/* For PCH output, training FDI link */
 	dev_priv->display.fdi_link_train(crtc);
 
@@ -4159,7 +4114,8 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 		temp = I915_READ(PCH_DPLL_SEL);
 		temp |= TRANS_DPLL_ENABLE(pipe);
 		sel = TRANS_DPLLB_SEL(pipe);
-		if (intel_crtc->config->shared_dpll == DPLL_ID_PCH_PLL_B)
+		if (intel_crtc->config->shared_dpll ==
+		    intel_get_shared_dpll_by_id(dev_priv, DPLL_ID_PCH_PLL_B))
 			temp |= sel;
 		else
 			temp &= ~sel;
@@ -4181,8 +4137,6 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 
 	intel_fdi_normal_train(crtc);
 
-	intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
-
 	/* For PCH DP, enable TRANS_DP_CTL */
 	if (HAS_PCH_CPT(dev) && intel_crtc->config->has_dp_encoder) {
 		const struct drm_display_mode *adjusted_mode =
@@ -4238,113 +4192,6 @@ static void lpt_pch_enable(struct drm_crtc *crtc)
 	lpt_enable_pch_transcoder(dev_priv, cpu_transcoder);
 }
 
-struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc,
-						struct intel_crtc_state *crtc_state)
-{
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct intel_shared_dpll *pll;
-	struct intel_shared_dpll_config *shared_dpll;
-	enum intel_dpll_id i;
-	int max = dev_priv->num_shared_dpll;
-
-	shared_dpll = intel_atomic_get_shared_dpll_state(crtc_state->base.state);
-
-	if (HAS_PCH_IBX(dev_priv->dev)) {
-		/* Ironlake PCH has a fixed PLL->PCH pipe mapping. */
-		i = (enum intel_dpll_id) crtc->pipe;
-		pll = &dev_priv->shared_dplls[i];
-
-		DRM_DEBUG_KMS("CRTC:%d using pre-allocated %s\n",
-			      crtc->base.base.id, pll->name);
-
-		WARN_ON(shared_dpll[i].crtc_mask);
-
-		goto found;
-	}
-
-	if (IS_BROXTON(dev_priv->dev)) {
-		/* PLL is attached to port in bxt */
-		struct intel_encoder *encoder;
-		struct intel_digital_port *intel_dig_port;
-
-		encoder = intel_ddi_get_crtc_new_encoder(crtc_state);
-		if (WARN_ON(!encoder))
-			return NULL;
-
-		intel_dig_port = enc_to_dig_port(&encoder->base);
-		/* 1:1 mapping between ports and PLLs */
-		i = (enum intel_dpll_id)intel_dig_port->port;
-		pll = &dev_priv->shared_dplls[i];
-		DRM_DEBUG_KMS("CRTC:%d using pre-allocated %s\n",
-			crtc->base.base.id, pll->name);
-		WARN_ON(shared_dpll[i].crtc_mask);
-
-		goto found;
-	} else if (INTEL_INFO(dev_priv)->gen < 9 && HAS_DDI(dev_priv))
-		/* Do not consider SPLL */
-		max = 2;
-
-	for (i = 0; i < max; i++) {
-		pll = &dev_priv->shared_dplls[i];
-
-		/* Only want to check enabled timings first */
-		if (shared_dpll[i].crtc_mask == 0)
-			continue;
-
-		if (memcmp(&crtc_state->dpll_hw_state,
-			   &shared_dpll[i].hw_state,
-			   sizeof(crtc_state->dpll_hw_state)) == 0) {
-			DRM_DEBUG_KMS("CRTC:%d sharing existing %s (crtc mask 0x%08x, ative %d)\n",
-				      crtc->base.base.id, pll->name,
-				      shared_dpll[i].crtc_mask,
-				      pll->active);
-			goto found;
-		}
-	}
-
-	/* Ok no matching timings, maybe there's a free one? */
-	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
-		pll = &dev_priv->shared_dplls[i];
-		if (shared_dpll[i].crtc_mask == 0) {
-			DRM_DEBUG_KMS("CRTC:%d allocated %s\n",
-				      crtc->base.base.id, pll->name);
-			goto found;
-		}
-	}
-
-	return NULL;
-
-found:
-	if (shared_dpll[i].crtc_mask == 0)
-		shared_dpll[i].hw_state =
-			crtc_state->dpll_hw_state;
-
-	crtc_state->shared_dpll = i;
-	DRM_DEBUG_DRIVER("using %s for pipe %c\n", pll->name,
-			 pipe_name(crtc->pipe));
-
-	shared_dpll[i].crtc_mask |= 1 << crtc->pipe;
-
-	return pll;
-}
-
-static void intel_shared_dpll_commit(struct drm_atomic_state *state)
-{
-	struct drm_i915_private *dev_priv = to_i915(state->dev);
-	struct intel_shared_dpll_config *shared_dpll;
-	struct intel_shared_dpll *pll;
-	enum intel_dpll_id i;
-
-	if (!to_intel_atomic_state(state)->dpll_set)
-		return;
-
-	shared_dpll = to_intel_atomic_state(state)->shared_dpll;
-	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
-		pll = &dev_priv->shared_dplls[i];
-		pll->config = shared_dpll[i];
-	}
-}
-
 static void cpt_verify_modeset(struct drm_device *dev, int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4576,8 +4423,11 @@ void hsw_enable_ips(struct intel_crtc *crtc)
 	if (!crtc->config->ips_enabled)
 		return;
 
-	/* We can only enable IPS after we enable a plane and wait for a vblank */
-	intel_wait_for_vblank(dev, crtc->pipe);
+	/*
+	 * We can only enable IPS after we enable a plane and wait for a vblank
+	 * This function is called from post_plane_update, which is run after
+	 * a vblank wait.
+	 */
 
 	assert_plane_enabled(dev_priv, crtc->plane);
 	if (IS_BROADWELL(dev)) {
@@ -4626,55 +4476,6 @@ void hsw_disable_ips(struct intel_crtc *crtc)
 	intel_wait_for_vblank(dev, crtc->pipe);
 }
 
-/** Loads the palette/gamma unit for the CRTC with the prepared values */
-static void intel_crtc_load_lut(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	enum pipe pipe = intel_crtc->pipe;
-	int i;
-	bool reenable_ips = false;
-
-	/* The clocks have to be on to load the palette. */
-	if (!crtc->state->active)
-		return;
-
-	if (HAS_GMCH_DISPLAY(dev_priv->dev)) {
-		if (intel_crtc->config->has_dsi_encoder)
-			assert_dsi_pll_enabled(dev_priv);
-		else
-			assert_pll_enabled(dev_priv, pipe);
-	}
-
-	/* Workaround : Do not read or write the pipe palette/gamma data while
-	 * GAMMA_MODE is configured for split gamma and IPS_CTL has IPS enabled.
-	 */
-	if (IS_HASWELL(dev) && intel_crtc->config->ips_enabled &&
-	    ((I915_READ(GAMMA_MODE(pipe)) & GAMMA_MODE_MODE_MASK) ==
-	     GAMMA_MODE_MODE_SPLIT)) {
-		hsw_disable_ips(intel_crtc);
-		reenable_ips = true;
-	}
-
-	for (i = 0; i < 256; i++) {
-		i915_reg_t palreg;
-
-		if (HAS_GMCH_DISPLAY(dev))
-			palreg = PALETTE(pipe, i);
-		else
-			palreg = LGC_PALETTE(pipe, i);
-
-		I915_WRITE(palreg,
-			   (intel_crtc->lut_r[i] << 16) |
-			   (intel_crtc->lut_g[i] << 8) |
-			   intel_crtc->lut_b[i]);
-	}
-
-	if (reenable_ips)
-		hsw_enable_ips(intel_crtc);
-}
-
 static void intel_crtc_dpms_overlay_disable(struct intel_crtc *intel_crtc)
 {
 	if (intel_crtc->overlay) {
@@ -4734,16 +4535,7 @@ intel_post_enable_primary(struct drm_crtc *crtc)
 	intel_check_pch_fifo_underruns(dev_priv);
 }
 
-/**
- * intel_pre_disable_primary - Perform operations before disabling primary plane
- * @crtc: the CRTC whose primary plane is to be disabled
- *
- * Performs potentially sleeping operations that must be done before the
- * primary plane is disabled, such as updating FBC and IPS.  Note that this may
- * be called due to an explicit primary plane update, or due to an implicit
- * disable that is caused when a sprite plane completely hides the primary
- * plane.
- */
+/* FIXME move all this to pre_plane_update() with proper state tracking */
 static void
 intel_pre_disable_primary(struct drm_crtc *crtc)
 {
@@ -4762,6 +4554,26 @@ intel_pre_disable_primary(struct drm_crtc *crtc)
 		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
 
 	/*
+	 * FIXME IPS should be fine as long as one plane is
+	 * enabled, but in practice it seems to have problems
+	 * when going from primary only to sprite only and vice
+	 * versa.
+	 */
+	hsw_disable_ips(intel_crtc);
+}
+
+/* FIXME get rid of this and use pre_plane_update */
+static void
+intel_pre_disable_primary_noatomic(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+
+	intel_pre_disable_primary(crtc);
+
+	/*
 	 * Vblank time updates from the shadow to live plane control register
 	 * are blocked if the memory self-refresh mode is active at that
 	 * moment. So to make sure the plane gets truly disabled, disable
@@ -4775,37 +4587,39 @@ intel_pre_disable_primary(struct drm_crtc *crtc)
 		dev_priv->wm.vlv.cxsr = false;
 		intel_wait_for_vblank(dev, pipe);
 	}
-
-	/*
-	 * FIXME IPS should be fine as long as one plane is
-	 * enabled, but in practice it seems to have problems
-	 * when going from primary only to sprite only and vice
-	 * versa.
-	 */
-	hsw_disable_ips(intel_crtc);
 }
 
-static void intel_post_plane_update(struct intel_crtc *crtc)
+static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state)
 {
-	struct intel_crtc_atomic_commit *atomic = &crtc->atomic;
+	struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->base.crtc);
+	struct drm_atomic_state *old_state = old_crtc_state->base.state;
 	struct intel_crtc_state *pipe_config =
 		to_intel_crtc_state(crtc->base.state);
 	struct drm_device *dev = crtc->base.dev;
+	struct drm_plane *primary = crtc->base.primary;
+	struct drm_plane_state *old_pri_state =
+		drm_atomic_get_existing_plane_state(old_state, primary);
 
-	intel_frontbuffer_flip(dev, atomic->fb_bits);
+	intel_frontbuffer_flip(dev, pipe_config->fb_bits);
 
 	crtc->wm.cxsr_allowed = true;
 
-	if (pipe_config->wm_changed && pipe_config->base.active)
+	if (pipe_config->update_wm_post && pipe_config->base.active)
 		intel_update_watermarks(&crtc->base);
 
-	if (atomic->update_fbc)
-		intel_fbc_post_update(crtc);
+	if (old_pri_state) {
+		struct intel_plane_state *primary_state =
+			to_intel_plane_state(primary->state);
+		struct intel_plane_state *old_primary_state =
+			to_intel_plane_state(old_pri_state);
 
-	if (atomic->post_enable_primary)
-		intel_post_enable_primary(&crtc->base);
+		intel_fbc_post_update(crtc);
 
-	memset(atomic, 0, sizeof(*atomic));
+		if (primary_state->visible &&
+		    (needs_modeset(&pipe_config->base) ||
+		     !old_primary_state->visible))
+			intel_post_enable_primary(&crtc->base);
+	}
 }
 
 static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state)
@@ -4813,7 +4627,6 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state)
 	struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->base.crtc);
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc_atomic_commit *atomic = &crtc->atomic;
 	struct intel_crtc_state *pipe_config =
 		to_intel_crtc_state(crtc->base.state);
 	struct drm_atomic_state *old_state = old_crtc_state->base.state;
@@ -4822,15 +4635,14 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state)
 		drm_atomic_get_existing_plane_state(old_state, primary);
 	bool modeset = needs_modeset(&pipe_config->base);
 
-	if (atomic->update_fbc)
-		intel_fbc_pre_update(crtc);
-
 	if (old_pri_state) {
 		struct intel_plane_state *primary_state =
 			to_intel_plane_state(primary->state);
 		struct intel_plane_state *old_primary_state =
 			to_intel_plane_state(old_pri_state);
 
+		intel_fbc_pre_update(crtc);
+
 		if (old_primary_state->visible &&
 		    (modeset || !primary_state->visible))
 			intel_pre_disable_primary(&crtc->base);
@@ -4839,11 +4651,58 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state)
 	if (pipe_config->disable_cxsr) {
 		crtc->wm.cxsr_allowed = false;
 
-		if (old_crtc_state->base.active)
+		/*
+		 * Vblank time updates from the shadow to live plane control register
+		 * are blocked if the memory self-refresh mode is active at that
+		 * moment. So to make sure the plane gets truly disabled, disable
+		 * first the self-refresh mode. The self-refresh enable bit in turn
+		 * will be checked/applied by the HW only at the next frame start
+		 * event which is after the vblank start event, so we need to have a
+		 * wait-for-vblank between disabling the plane and the pipe.
+		 */
+		if (old_crtc_state->base.active) {
 			intel_set_memory_cxsr(dev_priv, false);
+			dev_priv->wm.vlv.cxsr = false;
+			intel_wait_for_vblank(dev, crtc->pipe);
+		}
 	}
 
-	if (!needs_modeset(&pipe_config->base) && pipe_config->wm_changed)
+	/*
+	 * IVB workaround: must disable low power watermarks for at least
+	 * one frame before enabling scaling.  LP watermarks can be re-enabled
+	 * when scaling is disabled.
+	 *
+	 * WaCxSRDisabledForSpriteScaling:ivb
+	 */
+	if (pipe_config->disable_lp_wm) {
+		ilk_disable_lp_wm(dev);
+		intel_wait_for_vblank(dev, crtc->pipe);
+	}
+
+	/*
+	 * If we're doing a modeset, we're done.  No need to do any pre-vblank
+	 * watermark programming here.
+	 */
+	if (needs_modeset(&pipe_config->base))
+		return;
+
+	/*
+	 * For platforms that support atomic watermarks, program the
+	 * 'intermediate' watermarks immediately.  On pre-gen9 platforms, these
+	 * will be the intermediate values that are safe for both pre- and
+	 * post- vblank; when vblank happens, the 'active' values will be set
+	 * to the final 'target' values and we'll do this again to get the
+	 * optimal watermarks.  For gen9+ platforms, the values we program here
+	 * will be the final target values which will get automatically latched
+	 * at vblank time; no further programming will be necessary.
+	 *
+	 * If a platform hasn't been transitioned to atomic watermarks yet,
+	 * we'll continue to update watermarks the old way, if flags tell
+	 * us to.
+	 */
+	if (dev_priv->display.initial_watermarks != NULL)
+		dev_priv->display.initial_watermarks(pipe_config);
+	else if (pipe_config->update_wm_pre)
 		intel_update_watermarks(&crtc->base);
 }
 
@@ -4874,10 +4733,24 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
+	struct intel_crtc_state *pipe_config =
+		to_intel_crtc_state(crtc->state);
 
 	if (WARN_ON(intel_crtc->active))
 		return;
 
+	/*
+	 * Sometimes spurious CPU pipe underruns happen during FDI
+	 * training, at least with VGA+HDMI cloning. Suppress them.
+	 *
+	 * On ILK we get an occasional spurious CPU pipe underruns
+	 * between eDP port A enable and vdd enable. Also PCH port
+	 * enable seems to result in the occasional CPU pipe underrun.
+	 *
+	 * Spurious PCH underruns also occur during PCH enabling.
+	 */
+	if (intel_crtc->config->has_pch_encoder || IS_GEN5(dev_priv))
+		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
 	if (intel_crtc->config->has_pch_encoder)
 		intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, false);
 
@@ -4888,6 +4761,7 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 		intel_dp_set_m_n(intel_crtc, M1_N1);
 
 	intel_set_pipe_timings(intel_crtc);
+	intel_set_pipe_src_size(intel_crtc);
 
 	if (intel_crtc->config->has_pch_encoder) {
 		intel_cpu_transcoder_set_m_n(intel_crtc,
@@ -4898,8 +4772,6 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 
 	intel_crtc->active = true;
 
-	intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
-
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
 			encoder->pre_enable(encoder);
@@ -4920,9 +4792,10 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	 * On ILK+ LUT must be loaded before the pipe is running but with
 	 * clocks enabled
 	 */
-	intel_crtc_load_lut(crtc);
+	intel_color_load_luts(&pipe_config->base);
 
-	intel_update_watermarks(crtc);
+	if (dev_priv->display.initial_watermarks != NULL)
+		dev_priv->display.initial_watermarks(intel_crtc->config);
 	intel_enable_pipe(intel_crtc);
 
 	if (intel_crtc->config->has_pch_encoder)
@@ -4940,6 +4813,7 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	/* Must wait for vblank to avoid spurious PCH FIFO underruns */
 	if (intel_crtc->config->has_pch_encoder)
 		intel_wait_for_vblank(dev, pipe);
+	intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
 	intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, true);
 }
 
@@ -4956,6 +4830,7 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe, hsw_workaround_pipe;
+	enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder;
 	struct intel_crtc_state *pipe_config =
 		to_intel_crtc_state(crtc->state);
 
@@ -4966,16 +4841,20 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 		intel_set_pch_fifo_underrun_reporting(dev_priv, TRANSCODER_A,
 						      false);
 
-	if (intel_crtc_to_shared_dpll(intel_crtc))
+	if (intel_crtc->config->shared_dpll)
 		intel_enable_shared_dpll(intel_crtc);
 
 	if (intel_crtc->config->has_dp_encoder)
 		intel_dp_set_m_n(intel_crtc, M1_N1);
 
-	intel_set_pipe_timings(intel_crtc);
+	if (!intel_crtc->config->has_dsi_encoder)
+		intel_set_pipe_timings(intel_crtc);
+
+	intel_set_pipe_src_size(intel_crtc);
 
-	if (intel_crtc->config->cpu_transcoder != TRANSCODER_EDP) {
-		I915_WRITE(PIPE_MULT(intel_crtc->config->cpu_transcoder),
+	if (cpu_transcoder != TRANSCODER_EDP &&
+	    !transcoder_is_dsi(cpu_transcoder)) {
+		I915_WRITE(PIPE_MULT(cpu_transcoder),
 			   intel_crtc->config->pixel_multiplier - 1);
 	}
 
@@ -4984,9 +4863,12 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 				     &intel_crtc->config->fdi_m_n, NULL);
 	}
 
-	haswell_set_pipeconf(crtc);
+	if (!intel_crtc->config->has_dsi_encoder)
+		haswell_set_pipeconf(crtc);
+
+	haswell_set_pipemisc(crtc);
 
-	intel_set_pipe_csc(crtc);
+	intel_color_set_csc(&pipe_config->base);
 
 	intel_crtc->active = true;
 
@@ -5015,14 +4897,20 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	 * On ILK+ LUT must be loaded before the pipe is running but with
 	 * clocks enabled
 	 */
-	intel_crtc_load_lut(crtc);
+	intel_color_load_luts(&pipe_config->base);
 
 	intel_ddi_set_pipe_settings(crtc);
 	if (!intel_crtc->config->has_dsi_encoder)
 		intel_ddi_enable_transcoder_func(crtc);
 
-	intel_update_watermarks(crtc);
-	intel_enable_pipe(intel_crtc);
+	if (dev_priv->display.initial_watermarks != NULL)
+		dev_priv->display.initial_watermarks(pipe_config);
+	else
+		intel_update_watermarks(crtc);
+
+	/* XXX: Do the pipe assertions at the right place for BXT DSI. */
+	if (!intel_crtc->config->has_dsi_encoder)
+		intel_enable_pipe(intel_crtc);
 
 	if (intel_crtc->config->has_pch_encoder)
 		lpt_pch_enable(crtc);
@@ -5078,8 +4966,15 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
 
-	if (intel_crtc->config->has_pch_encoder)
+	/*
+	 * Sometimes spurious CPU pipe underruns happen when the
+	 * pipe is already disabled, but FDI RX/TX is still enabled.
+	 * Happens at least with VGA+HDMI cloning. Suppress them.
+	 */
+	if (intel_crtc->config->has_pch_encoder) {
+		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
 		intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, false);
+	}
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
@@ -5087,22 +4982,12 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 	drm_crtc_vblank_off(crtc);
 	assert_vblank_disabled(crtc);
 
-	/*
-	 * Sometimes spurious CPU pipe underruns happen when the
-	 * pipe is already disabled, but FDI RX/TX is still enabled.
-	 * Happens at least with VGA+HDMI cloning. Suppress them.
-	 */
-	if (intel_crtc->config->has_pch_encoder)
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
-
 	intel_disable_pipe(intel_crtc);
 
 	ironlake_pfit_disable(intel_crtc, false);
 
-	if (intel_crtc->config->has_pch_encoder) {
+	if (intel_crtc->config->has_pch_encoder)
 		ironlake_fdi_disable(crtc);
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
-	}
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->post_disable)
@@ -5132,6 +5017,7 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 		ironlake_fdi_pll_disable(intel_crtc);
 	}
 
+	intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
 	intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, true);
 }
 
@@ -5155,7 +5041,9 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	drm_crtc_vblank_off(crtc);
 	assert_vblank_disabled(crtc);
 
-	intel_disable_pipe(intel_crtc);
+	/* XXX: Do the pipe assertions at the right place for BXT DSI. */
+	if (!intel_crtc->config->has_dsi_encoder)
+		intel_disable_pipe(intel_crtc);
 
 	if (intel_crtc->config->dp_encoder_is_mst)
 		intel_ddi_set_vc_payload_alloc(crtc, false);
@@ -5330,6 +5218,9 @@ static unsigned long get_crtc_power_domains(struct drm_crtc *crtc,
 		mask |= BIT(intel_display_port_power_domain(intel_encoder));
 	}
 
+	if (crtc_state->shared_dpll)
+		mask |= BIT(POWER_DOMAIN_PLLS);
+
 	return mask;
 }
 
@@ -5393,6 +5284,8 @@ static void intel_update_max_cdclk(struct drm_device *dev)
 			dev_priv->max_cdclk_freq = 450000;
 		else
 			dev_priv->max_cdclk_freq = 337500;
+	} else if (IS_BROXTON(dev)) {
+		dev_priv->max_cdclk_freq = 624000;
 	} else if (IS_BROADWELL(dev))  {
 		/*
 		 * FIXME with extra cooling we can allow
@@ -5452,9 +5345,8 @@ static void intel_update_cdclk(struct drm_device *dev)
 		intel_update_max_cdclk(dev);
 }
 
-static void broxton_set_cdclk(struct drm_device *dev, int frequency)
+static void broxton_set_cdclk(struct drm_i915_private *dev_priv, int frequency)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t divider;
 	uint32_t ratio;
 	uint32_t current_freq;
@@ -5568,33 +5460,46 @@ static void broxton_set_cdclk(struct drm_device *dev, int frequency)
 		return;
 	}
 
-	intel_update_cdclk(dev);
+	intel_update_cdclk(dev_priv->dev);
 }
 
-void broxton_init_cdclk(struct drm_device *dev)
+static bool broxton_cdclk_is_enabled(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t val;
+	if (!(I915_READ(BXT_DE_PLL_ENABLE) & BXT_DE_PLL_PLL_ENABLE))
+		return false;
 
-	/*
-	 * NDE_RSTWRN_OPT RST PCH Handshake En must always be 0b on BXT
-	 * or else the reset will hang because there is no PCH to respond.
-	 * Move the handshake programming to initialization sequence.
-	 * Previously was left up to BIOS.
-	 */
-	val = I915_READ(HSW_NDE_RSTWRN_OPT);
-	val &= ~RESET_PCH_HANDSHAKE_ENABLE;
-	I915_WRITE(HSW_NDE_RSTWRN_OPT, val);
+	/* TODO: Check for a valid CDCLK rate */
+
+	if (!(I915_READ(DBUF_CTL) & DBUF_POWER_REQUEST)) {
+		DRM_DEBUG_DRIVER("CDCLK enabled, but DBUF power not requested\n");
 
-	/* Enable PG1 for cdclk */
-	intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
+		return false;
+	}
+
+	if (!(I915_READ(DBUF_CTL) & DBUF_POWER_STATE)) {
+		DRM_DEBUG_DRIVER("CDCLK enabled, but DBUF power hasn't settled\n");
+
+		return false;
+	}
 
+	return true;
+}
+
+bool broxton_cdclk_verify_state(struct drm_i915_private *dev_priv)
+{
+	return broxton_cdclk_is_enabled(dev_priv);
+}
+
+void broxton_init_cdclk(struct drm_i915_private *dev_priv)
+{
 	/* check if cd clock is enabled */
-	if (I915_READ(BXT_DE_PLL_ENABLE) & BXT_DE_PLL_PLL_ENABLE) {
-		DRM_DEBUG_KMS("Display already initialized\n");
+	if (broxton_cdclk_is_enabled(dev_priv)) {
+		DRM_DEBUG_KMS("CDCLK already enabled, won't reprogram it\n");
 		return;
 	}
 
+	DRM_DEBUG_KMS("CDCLK not enabled, enabling it\n");
+
 	/*
 	 * FIXME:
 	 * - The initial CDCLK needs to be read from VBT.
@@ -5602,7 +5507,7 @@ void broxton_init_cdclk(struct drm_device *dev)
 	 * - check if setting the max (or any) cdclk freq is really necessary
 	 *   here, it belongs to modeset time
 	 */
-	broxton_set_cdclk(dev, 624000);
+	broxton_set_cdclk(dev_priv, 624000);
 
 	I915_WRITE(DBUF_CTL, I915_READ(DBUF_CTL) | DBUF_POWER_REQUEST);
 	POSTING_READ(DBUF_CTL);
@@ -5613,10 +5518,8 @@ void broxton_init_cdclk(struct drm_device *dev)
 		DRM_ERROR("DBuf power enable timeout!\n");
 }
 
-void broxton_uninit_cdclk(struct drm_device *dev)
+void broxton_uninit_cdclk(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
 	I915_WRITE(DBUF_CTL, I915_READ(DBUF_CTL) & ~DBUF_POWER_REQUEST);
 	POSTING_READ(DBUF_CTL);
 
@@ -5626,9 +5529,7 @@ void broxton_uninit_cdclk(struct drm_device *dev)
 		DRM_ERROR("DBuf power disable timeout!\n");
 
 	/* Set minimum (bypass) frequency, in effect turning off the DE PLL */
-	broxton_set_cdclk(dev, 19200);
-
-	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+	broxton_set_cdclk(dev_priv, 19200);
 }
 
 static const struct skl_cdclk_entry {
@@ -6165,6 +6066,8 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
+	struct intel_crtc_state *pipe_config =
+		to_intel_crtc_state(crtc->state);
 	int pipe = intel_crtc->pipe;
 
 	if (WARN_ON(intel_crtc->active))
@@ -6174,6 +6077,7 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 		intel_dp_set_m_n(intel_crtc, M1_N1);
 
 	intel_set_pipe_timings(intel_crtc);
+	intel_set_pipe_src_size(intel_crtc);
 
 	if (IS_CHERRYVIEW(dev) && pipe == PIPE_B) {
 		struct drm_i915_private *dev_priv = dev->dev_private;
@@ -6192,14 +6096,12 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 		if (encoder->pre_pll_enable)
 			encoder->pre_pll_enable(encoder);
 
-	if (!intel_crtc->config->has_dsi_encoder) {
-		if (IS_CHERRYVIEW(dev)) {
-			chv_prepare_pll(intel_crtc, intel_crtc->config);
-			chv_enable_pll(intel_crtc, intel_crtc->config);
-		} else {
-			vlv_prepare_pll(intel_crtc, intel_crtc->config);
-			vlv_enable_pll(intel_crtc, intel_crtc->config);
-		}
+	if (IS_CHERRYVIEW(dev)) {
+		chv_prepare_pll(intel_crtc, intel_crtc->config);
+		chv_enable_pll(intel_crtc, intel_crtc->config);
+	} else {
+		vlv_prepare_pll(intel_crtc, intel_crtc->config);
+		vlv_enable_pll(intel_crtc, intel_crtc->config);
 	}
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
@@ -6208,8 +6110,9 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 
 	i9xx_pfit_enable(intel_crtc);
 
-	intel_crtc_load_lut(crtc);
+	intel_color_load_luts(&pipe_config->base);
 
+	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
 
 	assert_vblank_disabled(crtc);
@@ -6234,7 +6137,9 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
-	int pipe = intel_crtc->pipe;
+	struct intel_crtc_state *pipe_config =
+		to_intel_crtc_state(crtc->state);
+	enum pipe pipe = intel_crtc->pipe;
 
 	if (WARN_ON(intel_crtc->active))
 		return;
@@ -6245,6 +6150,7 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 		intel_dp_set_m_n(intel_crtc, M1_N1);
 
 	intel_set_pipe_timings(intel_crtc);
+	intel_set_pipe_src_size(intel_crtc);
 
 	i9xx_set_pipeconf(intel_crtc);
 
@@ -6261,7 +6167,7 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 
 	i9xx_pfit_enable(intel_crtc);
 
-	intel_crtc_load_lut(crtc);
+	intel_color_load_luts(&pipe_config->base);
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
@@ -6299,10 +6205,9 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
-	 * We also need to wait on all gmch platforms because of the
-	 * self-refresh mode constraint explained above.
 	 */
-	intel_wait_for_vblank(dev, pipe);
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
@@ -6337,6 +6242,7 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 
 static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
 {
+	struct intel_encoder *encoder;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
 	enum intel_display_power_domain domain;
@@ -6348,14 +6254,27 @@ static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
 	if (to_intel_plane_state(crtc->primary->state)->visible) {
 		WARN_ON(intel_crtc->unpin_work);
 
-		intel_pre_disable_primary(crtc);
+		intel_pre_disable_primary_noatomic(crtc);
 
 		intel_crtc_disable_planes(crtc, 1 << drm_plane_index(crtc->primary));
 		to_intel_plane_state(crtc->primary->state)->visible = false;
 	}
 
 	dev_priv->display.crtc_disable(crtc);
+
+	DRM_DEBUG_KMS("[CRTC:%d] hw state adjusted, was enabled, now disabled\n",
+		      crtc->base.id);
+
+	WARN_ON(drm_atomic_set_mode_for_crtc(crtc->state, NULL) < 0);
+	crtc->state->active = false;
 	intel_crtc->active = false;
+	crtc->enabled = false;
+	crtc->state->connector_mask = 0;
+	crtc->state->encoder_mask = 0;
+
+	for_each_encoder_on_crtc(crtc->dev, crtc, encoder)
+		encoder->base.crtc = NULL;
+
 	intel_fbc_disable(intel_crtc);
 	intel_update_watermarks(crtc);
 	intel_disable_shared_dpll(intel_crtc);
@@ -6398,7 +6317,7 @@ void intel_encoder_destroy(struct drm_encoder *encoder)
 
 /* Cross check the actual hw state with our own modeset state tracking (and it's
  * internal consistency). */
-static void intel_connector_check_state(struct intel_connector *connector)
+static void intel_connector_verify_state(struct intel_connector *connector)
 {
 	struct drm_crtc *crtc = connector->base.state->crtc;
 
@@ -6568,7 +6487,7 @@ retry:
 	 * Hence the bw of each lane in terms of the mode signal
 	 * is:
 	 */
-	link_bw = intel_fdi_link_freq(dev) * MHz(100)/KHz(1)/10;
+	link_bw = intel_fdi_link_freq(to_i915(dev), pipe_config);
 
 	fdi_dotclock = adjusted_mode->crtc_clock;
 
@@ -6580,8 +6499,7 @@ retry:
 	intel_link_compute_m_n(pipe_config->pipe_bpp, lane, fdi_dotclock,
 			       link_bw, &pipe_config->fdi_m_n);
 
-	ret = ironlake_check_fdi_lanes(intel_crtc->base.dev,
-				       intel_crtc->pipe, pipe_config);
+	ret = ironlake_check_fdi_lanes(dev, intel_crtc->pipe, pipe_config);
 	if (ret == -EINVAL && pipe_config->pipe_bpp > 6*3) {
 		pipe_config->pipe_bpp -= 2*3;
 		DRM_DEBUG_KMS("fdi link bw constraint, reducing pipe bpp to %i\n",
@@ -6605,7 +6523,7 @@ static bool pipe_config_supports_ips(struct drm_i915_private *dev_priv,
 		return false;
 
 	/* HSW can handle pixel rate up to cdclk? */
-	if (IS_HASWELL(dev_priv->dev))
+	if (IS_HASWELL(dev_priv))
 		return true;
 
 	/*
@@ -7133,30 +7051,6 @@ static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
 		&& !(dev_priv->quirks & QUIRK_LVDS_SSC_DISABLE);
 }
 
-static int i9xx_get_refclk(const struct intel_crtc_state *crtc_state,
-			   int num_connectors)
-{
-	struct drm_device *dev = crtc_state->base.crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int refclk;
-
-	WARN_ON(!crtc_state->base.state);
-
-	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev) || IS_BROXTON(dev)) {
-		refclk = 100000;
-	} else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS) &&
-	    intel_panel_use_ssc(dev_priv) && num_connectors < 2) {
-		refclk = dev_priv->vbt.lvds_ssc_freq;
-		DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n", refclk);
-	} else if (!IS_GEN2(dev)) {
-		refclk = 96000;
-	} else {
-		refclk = 48000;
-	}
-
-	return refclk;
-}
-
 static uint32_t pnv_dpll_compute_fp(struct dpll *dpll)
 {
 	return (1 << dpll->n) << 16 | dpll->m2;
@@ -7300,24 +7194,34 @@ void intel_dp_set_m_n(struct intel_crtc *crtc, enum link_m_n_set m_n)
 static void vlv_compute_dpll(struct intel_crtc *crtc,
 			     struct intel_crtc_state *pipe_config)
 {
-	u32 dpll, dpll_md;
+	pipe_config->dpll_hw_state.dpll = DPLL_INTEGRATED_REF_CLK_VLV |
+		DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
+	if (crtc->pipe != PIPE_A)
+		pipe_config->dpll_hw_state.dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
 
-	/*
-	 * Enable DPIO clock input. We should never disable the reference
-	 * clock for pipe B, since VGA hotplug / manual detection depends
-	 * on it.
-	 */
-	dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REF_CLK_ENABLE_VLV |
-		DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_REF_CLK_VLV;
-	/* We should never disable this, set it here for state tracking */
-	if (crtc->pipe == PIPE_B)
-		dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
-	dpll |= DPLL_VCO_ENABLE;
-	pipe_config->dpll_hw_state.dpll = dpll;
+	/* DPLL not used with DSI, but still need the rest set up */
+	if (!pipe_config->has_dsi_encoder)
+		pipe_config->dpll_hw_state.dpll |= DPLL_VCO_ENABLE |
+			DPLL_EXT_BUFFER_ENABLE_VLV;
 
-	dpll_md = (pipe_config->pixel_multiplier - 1)
-		<< DPLL_MD_UDI_MULTIPLIER_SHIFT;
-	pipe_config->dpll_hw_state.dpll_md = dpll_md;
+	pipe_config->dpll_hw_state.dpll_md =
+		(pipe_config->pixel_multiplier - 1) << DPLL_MD_UDI_MULTIPLIER_SHIFT;
+}
+
+static void chv_compute_dpll(struct intel_crtc *crtc,
+			     struct intel_crtc_state *pipe_config)
+{
+	pipe_config->dpll_hw_state.dpll = DPLL_SSC_REF_CLK_CHV |
+		DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
+	if (crtc->pipe != PIPE_A)
+		pipe_config->dpll_hw_state.dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
+
+	/* DPLL not used with DSI, but still need the rest set up */
+	if (!pipe_config->has_dsi_encoder)
+		pipe_config->dpll_hw_state.dpll |= DPLL_VCO_ENABLE;
+
+	pipe_config->dpll_hw_state.dpll_md =
+		(pipe_config->pixel_multiplier - 1) << DPLL_MD_UDI_MULTIPLIER_SHIFT;
 }
 
 static void vlv_prepare_pll(struct intel_crtc *crtc,
@@ -7325,11 +7229,20 @@ static void vlv_prepare_pll(struct intel_crtc *crtc,
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe = crtc->pipe;
+	enum pipe pipe = crtc->pipe;
 	u32 mdiv;
 	u32 bestn, bestm1, bestm2, bestp1, bestp2;
 	u32 coreclk, reg_val;
 
+	/* Enable Refclk */
+	I915_WRITE(DPLL(pipe),
+		   pipe_config->dpll_hw_state.dpll &
+		   ~(DPLL_VCO_ENABLE | DPLL_EXT_BUFFER_ENABLE_VLV));
+
+	/* No need to actually set up the DPLL with DSI */
+	if ((pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE) == 0)
+		return;
+
 	mutex_lock(&dev_priv->sb_lock);
 
 	bestn = pipe_config->dpll.n;
@@ -7411,32 +7324,26 @@ static void vlv_prepare_pll(struct intel_crtc *crtc,
 	mutex_unlock(&dev_priv->sb_lock);
 }
 
-static void chv_compute_dpll(struct intel_crtc *crtc,
-			     struct intel_crtc_state *pipe_config)
-{
-	pipe_config->dpll_hw_state.dpll = DPLL_SSC_REF_CLK_CHV |
-		DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS |
-		DPLL_VCO_ENABLE;
-	if (crtc->pipe != PIPE_A)
-		pipe_config->dpll_hw_state.dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
-
-	pipe_config->dpll_hw_state.dpll_md =
-		(pipe_config->pixel_multiplier - 1) << DPLL_MD_UDI_MULTIPLIER_SHIFT;
-}
-
 static void chv_prepare_pll(struct intel_crtc *crtc,
 			    const struct intel_crtc_state *pipe_config)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe = crtc->pipe;
-	i915_reg_t dpll_reg = DPLL(crtc->pipe);
+	enum pipe pipe = crtc->pipe;
 	enum dpio_channel port = vlv_pipe_to_channel(pipe);
 	u32 loopfilter, tribuf_calcntr;
 	u32 bestn, bestm1, bestm2, bestp1, bestp2, bestm2_frac;
 	u32 dpio_val;
 	int vco;
 
+	/* Enable Refclk and SSC */
+	I915_WRITE(DPLL(pipe),
+		   pipe_config->dpll_hw_state.dpll & ~DPLL_VCO_ENABLE);
+
+	/* No need to actually set up the DPLL with DSI */
+	if ((pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE) == 0)
+		return;
+
 	bestn = pipe_config->dpll.n;
 	bestm2_frac = pipe_config->dpll.m2 & 0x3fffff;
 	bestm1 = pipe_config->dpll.m1;
@@ -7447,12 +7354,6 @@ static void chv_prepare_pll(struct intel_crtc *crtc,
 	dpio_val = 0;
 	loopfilter = 0;
 
-	/*
-	 * Enable Refclk and SSC
-	 */
-	I915_WRITE(dpll_reg,
-		   pipe_config->dpll_hw_state.dpll & ~DPLL_VCO_ENABLE);
-
 	mutex_lock(&dev_priv->sb_lock);
 
 	/* p1 and p2 divider */
@@ -7586,8 +7487,7 @@ void vlv_force_pll_off(struct drm_device *dev, enum pipe pipe)
 
 static void i9xx_compute_dpll(struct intel_crtc *crtc,
 			      struct intel_crtc_state *crtc_state,
-			      intel_clock_t *reduced_clock,
-			      int num_connectors)
+			      intel_clock_t *reduced_clock)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -7646,7 +7546,7 @@ static void i9xx_compute_dpll(struct intel_crtc *crtc,
 	if (crtc_state->sdvo_tv_clock)
 		dpll |= PLL_REF_INPUT_TVCLKINBC;
 	else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS) &&
-		 intel_panel_use_ssc(dev_priv) && num_connectors < 2)
+		 intel_panel_use_ssc(dev_priv))
 		dpll |= PLLB_REF_INPUT_SPREADSPECTRUMIN;
 	else
 		dpll |= PLL_REF_INPUT_DREFCLK;
@@ -7663,8 +7563,7 @@ static void i9xx_compute_dpll(struct intel_crtc *crtc,
 
 static void i8xx_compute_dpll(struct intel_crtc *crtc,
 			      struct intel_crtc_state *crtc_state,
-			      intel_clock_t *reduced_clock,
-			      int num_connectors)
+			      intel_clock_t *reduced_clock)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -7690,7 +7589,7 @@ static void i8xx_compute_dpll(struct intel_crtc *crtc,
 		dpll |= DPLL_DVO_2X_MODE;
 
 	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS) &&
-		 intel_panel_use_ssc(dev_priv) && num_connectors < 2)
+	    intel_panel_use_ssc(dev_priv))
 		dpll |= PLLB_REF_INPUT_SPREADSPECTRUMIN;
 	else
 		dpll |= PLL_REF_INPUT_DREFCLK;
@@ -7759,6 +7658,14 @@ static void intel_set_pipe_timings(struct intel_crtc *intel_crtc)
 	    (pipe == PIPE_B || pipe == PIPE_C))
 		I915_WRITE(VTOTAL(pipe), I915_READ(VTOTAL(cpu_transcoder)));
 
+}
+
+static void intel_set_pipe_src_size(struct intel_crtc *intel_crtc)
+{
+	struct drm_device *dev = intel_crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe pipe = intel_crtc->pipe;
+
 	/* pipesrc controls the size that is scaled from, which should
 	 * always be the user's requested size.
 	 */
@@ -7800,6 +7707,14 @@ static void intel_get_pipe_timings(struct intel_crtc *crtc,
 		pipe_config->base.adjusted_mode.crtc_vtotal += 1;
 		pipe_config->base.adjusted_mode.crtc_vblank_end += 1;
 	}
+}
+
+static void intel_get_pipe_src_size(struct intel_crtc *crtc,
+				    struct intel_crtc_state *pipe_config)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 tmp;
 
 	tmp = I915_READ(PIPESRC(crtc->pipe));
 	pipe_config->pipe_src_h = (tmp & 0xffff) + 1;
@@ -7897,69 +7812,192 @@ static void i9xx_set_pipeconf(struct intel_crtc *intel_crtc)
 	POSTING_READ(PIPECONF(intel_crtc->pipe));
 }
 
-static int i9xx_crtc_compute_clock(struct intel_crtc *crtc,
+static int i8xx_crtc_compute_clock(struct intel_crtc *crtc,
 				   struct intel_crtc_state *crtc_state)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int refclk, num_connectors = 0;
-	intel_clock_t clock;
-	bool ok;
 	const intel_limit_t *limit;
-	struct drm_atomic_state *state = crtc_state->base.state;
-	struct drm_connector *connector;
-	struct drm_connector_state *connector_state;
-	int i;
+	int refclk = 48000;
 
 	memset(&crtc_state->dpll_hw_state, 0,
 	       sizeof(crtc_state->dpll_hw_state));
 
-	if (crtc_state->has_dsi_encoder)
-		return 0;
+	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS)) {
+		if (intel_panel_use_ssc(dev_priv)) {
+			refclk = dev_priv->vbt.lvds_ssc_freq;
+			DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n", refclk);
+		}
 
-	for_each_connector_in_state(state, connector, connector_state, i) {
-		if (connector_state->crtc == &crtc->base)
-			num_connectors++;
+		limit = &intel_limits_i8xx_lvds;
+	} else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_DVO)) {
+		limit = &intel_limits_i8xx_dvo;
+	} else {
+		limit = &intel_limits_i8xx_dac;
 	}
 
-	if (!crtc_state->clock_set) {
-		refclk = i9xx_get_refclk(crtc_state, num_connectors);
+	if (!crtc_state->clock_set &&
+	    !i9xx_find_best_dpll(limit, crtc_state, crtc_state->port_clock,
+				 refclk, NULL, &crtc_state->dpll)) {
+		DRM_ERROR("Couldn't find PLL settings for mode!\n");
+		return -EINVAL;
+	}
 
-		/*
-		 * Returns a set of divisors for the desired target clock with
-		 * the given refclk, or FALSE.  The returned values represent
-		 * the clock equation: reflck * (5 * (m1 + 2) + (m2 + 2)) / (n +
-		 * 2) / p1 / p2.
-		 */
-		limit = intel_limit(crtc_state, refclk);
-		ok = dev_priv->display.find_dpll(limit, crtc_state,
-						 crtc_state->port_clock,
-						 refclk, NULL, &clock);
-		if (!ok) {
-			DRM_ERROR("Couldn't find PLL settings for mode!\n");
-			return -EINVAL;
+	i8xx_compute_dpll(crtc, crtc_state, NULL);
+
+	return 0;
+}
+
+static int g4x_crtc_compute_clock(struct intel_crtc *crtc,
+				  struct intel_crtc_state *crtc_state)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	const intel_limit_t *limit;
+	int refclk = 96000;
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS)) {
+		if (intel_panel_use_ssc(dev_priv)) {
+			refclk = dev_priv->vbt.lvds_ssc_freq;
+			DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n", refclk);
+		}
+
+		if (intel_is_dual_link_lvds(dev))
+			limit = &intel_limits_g4x_dual_channel_lvds;
+		else
+			limit = &intel_limits_g4x_single_channel_lvds;
+	} else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_HDMI) ||
+		   intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_ANALOG)) {
+		limit = &intel_limits_g4x_hdmi;
+	} else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_SDVO)) {
+		limit = &intel_limits_g4x_sdvo;
+	} else {
+		/* The option is for other outputs */
+		limit = &intel_limits_i9xx_sdvo;
+	}
+
+	if (!crtc_state->clock_set &&
+	    !g4x_find_best_dpll(limit, crtc_state, crtc_state->port_clock,
+				refclk, NULL, &crtc_state->dpll)) {
+		DRM_ERROR("Couldn't find PLL settings for mode!\n");
+		return -EINVAL;
+	}
+
+	i9xx_compute_dpll(crtc, crtc_state, NULL);
+
+	return 0;
+}
+
+static int pnv_crtc_compute_clock(struct intel_crtc *crtc,
+				  struct intel_crtc_state *crtc_state)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	const intel_limit_t *limit;
+	int refclk = 96000;
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS)) {
+		if (intel_panel_use_ssc(dev_priv)) {
+			refclk = dev_priv->vbt.lvds_ssc_freq;
+			DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n", refclk);
 		}
 
-		/* Compat-code for transition, will disappear. */
-		crtc_state->dpll.n = clock.n;
-		crtc_state->dpll.m1 = clock.m1;
-		crtc_state->dpll.m2 = clock.m2;
-		crtc_state->dpll.p1 = clock.p1;
-		crtc_state->dpll.p2 = clock.p2;
+		limit = &intel_limits_pineview_lvds;
+	} else {
+		limit = &intel_limits_pineview_sdvo;
 	}
 
-	if (IS_GEN2(dev)) {
-		i8xx_compute_dpll(crtc, crtc_state, NULL,
-				  num_connectors);
-	} else if (IS_CHERRYVIEW(dev)) {
-		chv_compute_dpll(crtc, crtc_state);
-	} else if (IS_VALLEYVIEW(dev)) {
-		vlv_compute_dpll(crtc, crtc_state);
+	if (!crtc_state->clock_set &&
+	    !pnv_find_best_dpll(limit, crtc_state, crtc_state->port_clock,
+				refclk, NULL, &crtc_state->dpll)) {
+		DRM_ERROR("Couldn't find PLL settings for mode!\n");
+		return -EINVAL;
+	}
+
+	i9xx_compute_dpll(crtc, crtc_state, NULL);
+
+	return 0;
+}
+
+static int i9xx_crtc_compute_clock(struct intel_crtc *crtc,
+				   struct intel_crtc_state *crtc_state)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	const intel_limit_t *limit;
+	int refclk = 96000;
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS)) {
+		if (intel_panel_use_ssc(dev_priv)) {
+			refclk = dev_priv->vbt.lvds_ssc_freq;
+			DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n", refclk);
+		}
+
+		limit = &intel_limits_i9xx_lvds;
 	} else {
-		i9xx_compute_dpll(crtc, crtc_state, NULL,
-				  num_connectors);
+		limit = &intel_limits_i9xx_sdvo;
+	}
+
+	if (!crtc_state->clock_set &&
+	    !i9xx_find_best_dpll(limit, crtc_state, crtc_state->port_clock,
+				 refclk, NULL, &crtc_state->dpll)) {
+		DRM_ERROR("Couldn't find PLL settings for mode!\n");
+		return -EINVAL;
+	}
+
+	i9xx_compute_dpll(crtc, crtc_state, NULL);
+
+	return 0;
+}
+
+static int chv_crtc_compute_clock(struct intel_crtc *crtc,
+				  struct intel_crtc_state *crtc_state)
+{
+	int refclk = 100000;
+	const intel_limit_t *limit = &intel_limits_chv;
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (!crtc_state->clock_set &&
+	    !chv_find_best_dpll(limit, crtc_state, crtc_state->port_clock,
+				refclk, NULL, &crtc_state->dpll)) {
+		DRM_ERROR("Couldn't find PLL settings for mode!\n");
+		return -EINVAL;
+	}
+
+	chv_compute_dpll(crtc, crtc_state);
+
+	return 0;
+}
+
+static int vlv_crtc_compute_clock(struct intel_crtc *crtc,
+				  struct intel_crtc_state *crtc_state)
+{
+	int refclk = 100000;
+	const intel_limit_t *limit = &intel_limits_vlv;
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (!crtc_state->clock_set &&
+	    !vlv_find_best_dpll(limit, crtc_state, crtc_state->port_clock,
+				refclk, NULL, &crtc_state->dpll)) {
+		DRM_ERROR("Couldn't find PLL settings for mode!\n");
+		return -EINVAL;
 	}
 
+	vlv_compute_dpll(crtc, crtc_state);
+
 	return 0;
 }
 
@@ -8000,8 +8038,8 @@ static void vlv_crtc_clock_get(struct intel_crtc *crtc,
 	u32 mdiv;
 	int refclk = 100000;
 
-	/* In case of MIPI DPLL will not even be used */
-	if (!(pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE))
+	/* In case of DSI, DPLL will not be used */
+	if ((pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE) == 0)
 		return;
 
 	mutex_lock(&dev_priv->sb_lock);
@@ -8097,6 +8135,10 @@ static void chv_crtc_clock_get(struct intel_crtc *crtc,
 	u32 cmn_dw13, pll_dw0, pll_dw1, pll_dw2, pll_dw3;
 	int refclk = 100000;
 
+	/* In case of DSI, DPLL will not be used */
+	if ((pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE) == 0)
+		return;
+
 	mutex_lock(&dev_priv->sb_lock);
 	cmn_dw13 = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW13(port));
 	pll_dw0 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW0(port));
@@ -8130,7 +8172,7 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 		return false;
 
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
-	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
+	pipe_config->shared_dpll = NULL;
 
 	ret = false;
 
@@ -8162,11 +8204,16 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 		pipe_config->double_wide = tmp & PIPECONF_DOUBLE_WIDE;
 
 	intel_get_pipe_timings(crtc, pipe_config);
+	intel_get_pipe_src_size(crtc, pipe_config);
 
 	i9xx_get_pfit_config(crtc, pipe_config);
 
 	if (INTEL_INFO(dev)->gen >= 4) {
-		tmp = I915_READ(DPLL_MD(crtc->pipe));
+		/* No way to read it out on pipes B and C */
+		if (IS_CHERRYVIEW(dev) && crtc->pipe != PIPE_A)
+			tmp = dev_priv->chv_dpll_md[crtc->pipe];
+		else
+			tmp = I915_READ(DPLL_MD(crtc->pipe));
 		pipe_config->pixel_multiplier =
 			((tmp & DPLL_MD_UDI_MULTIPLIER_MASK)
 			 >> DPLL_MD_UDI_MULTIPLIER_SHIFT) + 1;
@@ -8635,42 +8682,6 @@ void intel_init_pch_refclk(struct drm_device *dev)
 		lpt_init_pch_refclk(dev);
 }
 
-static int ironlake_get_refclk(struct intel_crtc_state *crtc_state)
-{
-	struct drm_device *dev = crtc_state->base.crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_atomic_state *state = crtc_state->base.state;
-	struct drm_connector *connector;
-	struct drm_connector_state *connector_state;
-	struct intel_encoder *encoder;
-	int num_connectors = 0, i;
-	bool is_lvds = false;
-
-	for_each_connector_in_state(state, connector, connector_state, i) {
-		if (connector_state->crtc != crtc_state->base.crtc)
-			continue;
-
-		encoder = to_intel_encoder(connector_state->best_encoder);
-
-		switch (encoder->type) {
-		case INTEL_OUTPUT_LVDS:
-			is_lvds = true;
-			break;
-		default:
-			break;
-		}
-		num_connectors++;
-	}
-
-	if (is_lvds && intel_panel_use_ssc(dev_priv) && num_connectors < 2) {
-		DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n",
-			      dev_priv->vbt.lvds_ssc_freq);
-		return dev_priv->vbt.lvds_ssc_freq;
-	}
-
-	return 120000;
-}
-
 static void ironlake_set_pipeconf(struct drm_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
@@ -8713,82 +8724,14 @@ static void ironlake_set_pipeconf(struct drm_crtc *crtc)
 	POSTING_READ(PIPECONF(pipe));
 }
 
-/*
- * Set up the pipe CSC unit.
- *
- * Currently only full range RGB to limited range RGB conversion
- * is supported, but eventually this should handle various
- * RGB<->YCbCr scenarios as well.
- */
-static void intel_set_pipe_csc(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	uint16_t coeff = 0x7800; /* 1.0 */
-
-	/*
-	 * TODO: Check what kind of values actually come out of the pipe
-	 * with these coeff/postoff values and adjust to get the best
-	 * accuracy. Perhaps we even need to take the bpc value into
-	 * consideration.
-	 */
-
-	if (intel_crtc->config->limited_color_range)
-		coeff = ((235 - 16) * (1 << 12) / 255) & 0xff8; /* 0.xxx... */
-
-	/*
-	 * GY/GU and RY/RU should be the other way around according
-	 * to BSpec, but reality doesn't agree. Just set them up in
-	 * a way that results in the correct picture.
-	 */
-	I915_WRITE(PIPE_CSC_COEFF_RY_GY(pipe), coeff << 16);
-	I915_WRITE(PIPE_CSC_COEFF_BY(pipe), 0);
-
-	I915_WRITE(PIPE_CSC_COEFF_RU_GU(pipe), coeff);
-	I915_WRITE(PIPE_CSC_COEFF_BU(pipe), 0);
-
-	I915_WRITE(PIPE_CSC_COEFF_RV_GV(pipe), 0);
-	I915_WRITE(PIPE_CSC_COEFF_BV(pipe), coeff << 16);
-
-	I915_WRITE(PIPE_CSC_PREOFF_HI(pipe), 0);
-	I915_WRITE(PIPE_CSC_PREOFF_ME(pipe), 0);
-	I915_WRITE(PIPE_CSC_PREOFF_LO(pipe), 0);
-
-	if (INTEL_INFO(dev)->gen > 6) {
-		uint16_t postoff = 0;
-
-		if (intel_crtc->config->limited_color_range)
-			postoff = (16 * (1 << 12) / 255) & 0x1fff;
-
-		I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), postoff);
-		I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), postoff);
-		I915_WRITE(PIPE_CSC_POSTOFF_LO(pipe), postoff);
-
-		I915_WRITE(PIPE_CSC_MODE(pipe), 0);
-	} else {
-		uint32_t mode = CSC_MODE_YUV_TO_RGB;
-
-		if (intel_crtc->config->limited_color_range)
-			mode |= CSC_BLACK_SCREEN_OFFSET;
-
-		I915_WRITE(PIPE_CSC_MODE(pipe), mode);
-	}
-}
-
 static void haswell_set_pipeconf(struct drm_crtc *crtc)
 {
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	enum pipe pipe = intel_crtc->pipe;
 	enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder;
-	uint32_t val;
+	u32 val = 0;
 
-	val = 0;
-
-	if (IS_HASWELL(dev) && intel_crtc->config->dither)
+	if (IS_HASWELL(dev_priv) && intel_crtc->config->dither)
 		val |= (PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_SP);
 
 	if (intel_crtc->config->base.adjusted_mode.flags & DRM_MODE_FLAG_INTERLACE)
@@ -8798,12 +8741,15 @@ static void haswell_set_pipeconf(struct drm_crtc *crtc)
 
 	I915_WRITE(PIPECONF(cpu_transcoder), val);
 	POSTING_READ(PIPECONF(cpu_transcoder));
+}
 
-	I915_WRITE(GAMMA_MODE(intel_crtc->pipe), GAMMA_MODE_MODE_8BIT);
-	POSTING_READ(GAMMA_MODE(intel_crtc->pipe));
+static void haswell_set_pipemisc(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
-	if (IS_BROADWELL(dev) || INTEL_INFO(dev)->gen >= 9) {
-		val = 0;
+	if (IS_BROADWELL(dev_priv) || INTEL_INFO(dev_priv)->gen >= 9) {
+		u32 val = 0;
 
 		switch (intel_crtc->config->pipe_bpp) {
 		case 18:
@@ -8826,39 +8772,10 @@ static void haswell_set_pipeconf(struct drm_crtc *crtc)
 		if (intel_crtc->config->dither)
 			val |= PIPEMISC_DITHER_ENABLE | PIPEMISC_DITHER_TYPE_SP;
 
-		I915_WRITE(PIPEMISC(pipe), val);
+		I915_WRITE(PIPEMISC(intel_crtc->pipe), val);
 	}
 }
 
-static bool ironlake_compute_clocks(struct drm_crtc *crtc,
-				    struct intel_crtc_state *crtc_state,
-				    intel_clock_t *clock,
-				    bool *has_reduced_clock,
-				    intel_clock_t *reduced_clock)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int refclk;
-	const intel_limit_t *limit;
-	bool ret;
-
-	refclk = ironlake_get_refclk(crtc_state);
-
-	/*
-	 * Returns a set of divisors for the desired target clock with the given
-	 * refclk, or FALSE.  The returned values represent the clock equation:
-	 * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2.
-	 */
-	limit = intel_limit(crtc_state, refclk);
-	ret = dev_priv->display.find_dpll(limit, crtc_state,
-					  crtc_state->port_clock,
-					  refclk, NULL, clock);
-	if (!ret)
-		return false;
-
-	return true;
-}
-
 int ironlake_get_lanes_required(int target_clock, int link_bw, int bpp)
 {
 	/*
@@ -8875,10 +8792,9 @@ static bool ironlake_needs_fb_cb_tune(struct dpll *dpll, int factor)
 	return i9xx_dpll_compute_m(dpll) < factor * dpll->n;
 }
 
-static uint32_t ironlake_compute_dpll(struct intel_crtc *intel_crtc,
-				      struct intel_crtc_state *crtc_state,
-				      u32 *fp,
-				      intel_clock_t *reduced_clock, u32 *fp2)
+static void ironlake_compute_dpll(struct intel_crtc *intel_crtc,
+				  struct intel_crtc_state *crtc_state,
+				  intel_clock_t *reduced_clock)
 {
 	struct drm_crtc *crtc = &intel_crtc->base;
 	struct drm_device *dev = crtc->dev;
@@ -8887,8 +8803,8 @@ static uint32_t ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 	struct drm_connector *connector;
 	struct drm_connector_state *connector_state;
 	struct intel_encoder *encoder;
-	uint32_t dpll;
-	int factor, num_connectors = 0, i;
+	u32 dpll, fp, fp2;
+	int factor, i;
 	bool is_lvds = false, is_sdvo = false;
 
 	for_each_connector_in_state(state, connector, connector_state, i) {
@@ -8908,8 +8824,6 @@ static uint32_t ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 		default:
 			break;
 		}
-
-		num_connectors++;
 	}
 
 	/* Enable autotuning of the PLL clock (if permissible) */
@@ -8922,11 +8836,19 @@ static uint32_t ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 	} else if (crtc_state->sdvo_tv_clock)
 		factor = 20;
 
+	fp = i9xx_dpll_compute_fp(&crtc_state->dpll);
+
 	if (ironlake_needs_fb_cb_tune(&crtc_state->dpll, factor))
-		*fp |= FP_CB_TUNE;
+		fp |= FP_CB_TUNE;
 
-	if (fp2 && (reduced_clock->m < factor * reduced_clock->n))
-		*fp2 |= FP_CB_TUNE;
+	if (reduced_clock) {
+		fp2 = i9xx_dpll_compute_fp(reduced_clock);
+
+		if (reduced_clock->m < factor * reduced_clock->n)
+			fp2 |= FP_CB_TUNE;
+	} else {
+		fp2 = fp;
+	}
 
 	dpll = 0;
 
@@ -8963,76 +8885,80 @@ static uint32_t ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 		break;
 	}
 
-	if (is_lvds && intel_panel_use_ssc(dev_priv) && num_connectors < 2)
+	if (is_lvds && intel_panel_use_ssc(dev_priv))
 		dpll |= PLLB_REF_INPUT_SPREADSPECTRUMIN;
 	else
 		dpll |= PLL_REF_INPUT_DREFCLK;
 
-	return dpll | DPLL_VCO_ENABLE;
+	dpll |= DPLL_VCO_ENABLE;
+
+	crtc_state->dpll_hw_state.dpll = dpll;
+	crtc_state->dpll_hw_state.fp0 = fp;
+	crtc_state->dpll_hw_state.fp1 = fp2;
 }
 
 static int ironlake_crtc_compute_clock(struct intel_crtc *crtc,
 				       struct intel_crtc_state *crtc_state)
 {
 	struct drm_device *dev = crtc->base.dev;
-	intel_clock_t clock, reduced_clock;
-	u32 dpll = 0, fp = 0, fp2 = 0;
-	bool ok, has_reduced_clock = false;
-	bool is_lvds = false;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	intel_clock_t reduced_clock;
+	bool has_reduced_clock = false;
 	struct intel_shared_dpll *pll;
+	const intel_limit_t *limit;
+	int refclk = 120000;
 
 	memset(&crtc_state->dpll_hw_state, 0,
 	       sizeof(crtc_state->dpll_hw_state));
 
-	is_lvds = intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS);
+	crtc->lowfreq_avail = false;
 
-	WARN(!(HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)),
-	     "Unexpected PCH type %d\n", INTEL_PCH_TYPE(dev));
+	/* CPU eDP is the only output that doesn't need a PCH PLL of its own. */
+	if (!crtc_state->has_pch_encoder)
+		return 0;
 
-	ok = ironlake_compute_clocks(&crtc->base, crtc_state, &clock,
-				     &has_reduced_clock, &reduced_clock);
-	if (!ok && !crtc_state->clock_set) {
+	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS)) {
+		if (intel_panel_use_ssc(dev_priv)) {
+			DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n",
+				      dev_priv->vbt.lvds_ssc_freq);
+			refclk = dev_priv->vbt.lvds_ssc_freq;
+		}
+
+		if (intel_is_dual_link_lvds(dev)) {
+			if (refclk == 100000)
+				limit = &intel_limits_ironlake_dual_lvds_100m;
+			else
+				limit = &intel_limits_ironlake_dual_lvds;
+		} else {
+			if (refclk == 100000)
+				limit = &intel_limits_ironlake_single_lvds_100m;
+			else
+				limit = &intel_limits_ironlake_single_lvds;
+		}
+	} else {
+		limit = &intel_limits_ironlake_dac;
+	}
+
+	if (!crtc_state->clock_set &&
+	    !g4x_find_best_dpll(limit, crtc_state, crtc_state->port_clock,
+				refclk, NULL, &crtc_state->dpll)) {
 		DRM_ERROR("Couldn't find PLL settings for mode!\n");
 		return -EINVAL;
 	}
-	/* Compat-code for transition, will disappear. */
-	if (!crtc_state->clock_set) {
-		crtc_state->dpll.n = clock.n;
-		crtc_state->dpll.m1 = clock.m1;
-		crtc_state->dpll.m2 = clock.m2;
-		crtc_state->dpll.p1 = clock.p1;
-		crtc_state->dpll.p2 = clock.p2;
-	}
-
-	/* CPU eDP is the only output that doesn't need a PCH PLL of its own. */
-	if (crtc_state->has_pch_encoder) {
-		fp = i9xx_dpll_compute_fp(&crtc_state->dpll);
-		if (has_reduced_clock)
-			fp2 = i9xx_dpll_compute_fp(&reduced_clock);
 
-		dpll = ironlake_compute_dpll(crtc, crtc_state,
-					     &fp, &reduced_clock,
-					     has_reduced_clock ? &fp2 : NULL);
+	ironlake_compute_dpll(crtc, crtc_state,
+			      has_reduced_clock ? &reduced_clock : NULL);
 
-		crtc_state->dpll_hw_state.dpll = dpll;
-		crtc_state->dpll_hw_state.fp0 = fp;
-		if (has_reduced_clock)
-			crtc_state->dpll_hw_state.fp1 = fp2;
-		else
-			crtc_state->dpll_hw_state.fp1 = fp;
-
-		pll = intel_get_shared_dpll(crtc, crtc_state);
-		if (pll == NULL) {
-			DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
-					 pipe_name(crtc->pipe));
-			return -EINVAL;
-		}
+	pll = intel_get_shared_dpll(crtc, crtc_state, NULL);
+	if (pll == NULL) {
+		DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
+				 pipe_name(crtc->pipe));
+		return -EINVAL;
 	}
 
-	if (is_lvds && has_reduced_clock)
+	if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS) &&
+	    has_reduced_clock)
 		crtc->lowfreq_avail = true;
-	else
-		crtc->lowfreq_avail = false;
 
 	return 0;
 }
@@ -9334,7 +9260,7 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 		return false;
 
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
-	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
+	pipe_config->shared_dpll = NULL;
 
 	ret = false;
 	tmp = I915_READ(PIPECONF(crtc->pipe));
@@ -9363,6 +9289,7 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 
 	if (I915_READ(PCH_TRANSCONF(crtc->pipe)) & TRANS_ENABLE) {
 		struct intel_shared_dpll *pll;
+		enum intel_dpll_id pll_id;
 
 		pipe_config->has_pch_encoder = true;
 
@@ -9372,21 +9299,22 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 
 		ironlake_get_fdi_m_n_config(crtc, pipe_config);
 
-		if (HAS_PCH_IBX(dev_priv->dev)) {
-			pipe_config->shared_dpll =
-				(enum intel_dpll_id) crtc->pipe;
+		if (HAS_PCH_IBX(dev_priv)) {
+			pll_id = (enum intel_dpll_id) crtc->pipe;
 		} else {
 			tmp = I915_READ(PCH_DPLL_SEL);
 			if (tmp & TRANS_DPLLB_SEL(crtc->pipe))
-				pipe_config->shared_dpll = DPLL_ID_PCH_PLL_B;
+				pll_id = DPLL_ID_PCH_PLL_B;
 			else
-				pipe_config->shared_dpll = DPLL_ID_PCH_PLL_A;
+				pll_id= DPLL_ID_PCH_PLL_A;
 		}
 
-		pll = &dev_priv->shared_dplls[pipe_config->shared_dpll];
+		pipe_config->shared_dpll =
+			intel_get_shared_dpll_by_id(dev_priv, pll_id);
+		pll = pipe_config->shared_dpll;
 
-		WARN_ON(!pll->get_hw_state(dev_priv, pll,
-					   &pipe_config->dpll_hw_state));
+		WARN_ON(!pll->funcs.get_hw_state(dev_priv, pll,
+						 &pipe_config->dpll_hw_state));
 
 		tmp = pipe_config->dpll_hw_state.dpll;
 		pipe_config->pixel_multiplier =
@@ -9399,6 +9327,7 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 	}
 
 	intel_get_pipe_timings(crtc, pipe_config);
+	intel_get_pipe_src_size(crtc, pipe_config);
 
 	ironlake_get_pfit_config(crtc, pipe_config);
 
@@ -9638,7 +9567,7 @@ static void broxton_modeset_commit_cdclk(struct drm_atomic_state *old_state)
 		to_intel_atomic_state(old_state);
 	unsigned int req_cdclk = old_intel_state->dev_cdclk;
 
-	broxton_set_cdclk(dev, req_cdclk);
+	broxton_set_cdclk(to_i915(dev), req_cdclk);
 }
 
 /* compute the max rate for new configuration */
@@ -9706,8 +9635,8 @@ static void broadwell_set_cdclk(struct drm_device *dev, int cdclk)
 	val |= LCPLL_CD_SOURCE_FCLK;
 	I915_WRITE(LCPLL_CTL, val);
 
-	if (wait_for_atomic_us(I915_READ(LCPLL_CTL) &
-			       LCPLL_CD_SOURCE_FCLK_DONE, 1))
+	if (wait_for_us(I915_READ(LCPLL_CTL) &
+			LCPLL_CD_SOURCE_FCLK_DONE, 1))
 		DRM_ERROR("Switching to FCLK failed\n");
 
 	val = I915_READ(LCPLL_CTL);
@@ -9741,8 +9670,8 @@ static void broadwell_set_cdclk(struct drm_device *dev, int cdclk)
 	val &= ~LCPLL_CD_SOURCE_FCLK;
 	I915_WRITE(LCPLL_CTL, val);
 
-	if (wait_for_atomic_us((I915_READ(LCPLL_CTL) &
-				LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1))
+	if (wait_for_us((I915_READ(LCPLL_CTL) &
+			LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1))
 		DRM_ERROR("Switching back to LCPLL failed\n");
 
 	mutex_lock(&dev_priv->rps.hw_lock);
@@ -9821,72 +9750,193 @@ static void bxt_get_ddi_pll(struct drm_i915_private *dev_priv,
 				enum port port,
 				struct intel_crtc_state *pipe_config)
 {
+	enum intel_dpll_id id;
+
 	switch (port) {
 	case PORT_A:
 		pipe_config->ddi_pll_sel = SKL_DPLL0;
-		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL1;
+		id = DPLL_ID_SKL_DPLL0;
 		break;
 	case PORT_B:
 		pipe_config->ddi_pll_sel = SKL_DPLL1;
-		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL2;
+		id = DPLL_ID_SKL_DPLL1;
 		break;
 	case PORT_C:
 		pipe_config->ddi_pll_sel = SKL_DPLL2;
-		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL3;
+		id = DPLL_ID_SKL_DPLL2;
 		break;
 	default:
 		DRM_ERROR("Incorrect port type\n");
+		return;
 	}
+
+	pipe_config->shared_dpll = intel_get_shared_dpll_by_id(dev_priv, id);
 }
 
 static void skylake_get_ddi_pll(struct drm_i915_private *dev_priv,
 				enum port port,
 				struct intel_crtc_state *pipe_config)
 {
-	u32 temp, dpll_ctl1;
+	enum intel_dpll_id id;
+	u32 temp;
 
 	temp = I915_READ(DPLL_CTRL2) & DPLL_CTRL2_DDI_CLK_SEL_MASK(port);
 	pipe_config->ddi_pll_sel = temp >> (port * 3 + 1);
 
 	switch (pipe_config->ddi_pll_sel) {
 	case SKL_DPLL0:
-		/*
-		 * On SKL the eDP DPLL (DPLL0 as we don't use SSC) is not part
-		 * of the shared DPLL framework and thus needs to be read out
-		 * separately
-		 */
-		dpll_ctl1 = I915_READ(DPLL_CTRL1);
-		pipe_config->dpll_hw_state.ctrl1 = dpll_ctl1 & 0x3f;
+		id = DPLL_ID_SKL_DPLL0;
 		break;
 	case SKL_DPLL1:
-		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL1;
+		id = DPLL_ID_SKL_DPLL1;
 		break;
 	case SKL_DPLL2:
-		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL2;
+		id = DPLL_ID_SKL_DPLL2;
 		break;
 	case SKL_DPLL3:
-		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL3;
+		id = DPLL_ID_SKL_DPLL3;
 		break;
+	default:
+		MISSING_CASE(pipe_config->ddi_pll_sel);
+		return;
 	}
+
+	pipe_config->shared_dpll = intel_get_shared_dpll_by_id(dev_priv, id);
 }
 
 static void haswell_get_ddi_pll(struct drm_i915_private *dev_priv,
 				enum port port,
 				struct intel_crtc_state *pipe_config)
 {
+	enum intel_dpll_id id;
+
 	pipe_config->ddi_pll_sel = I915_READ(PORT_CLK_SEL(port));
 
 	switch (pipe_config->ddi_pll_sel) {
 	case PORT_CLK_SEL_WRPLL1:
-		pipe_config->shared_dpll = DPLL_ID_WRPLL1;
+		id = DPLL_ID_WRPLL1;
 		break;
 	case PORT_CLK_SEL_WRPLL2:
-		pipe_config->shared_dpll = DPLL_ID_WRPLL2;
+		id = DPLL_ID_WRPLL2;
 		break;
 	case PORT_CLK_SEL_SPLL:
-		pipe_config->shared_dpll = DPLL_ID_SPLL;
+		id = DPLL_ID_SPLL;
+		break;
+	case PORT_CLK_SEL_LCPLL_810:
+		id = DPLL_ID_LCPLL_810;
+		break;
+	case PORT_CLK_SEL_LCPLL_1350:
+		id = DPLL_ID_LCPLL_1350;
+		break;
+	case PORT_CLK_SEL_LCPLL_2700:
+		id = DPLL_ID_LCPLL_2700;
+		break;
+	default:
+		MISSING_CASE(pipe_config->ddi_pll_sel);
+		/* fall through */
+	case PORT_CLK_SEL_NONE:
+		return;
+	}
+
+	pipe_config->shared_dpll = intel_get_shared_dpll_by_id(dev_priv, id);
+}
+
+static bool hsw_get_transcoder_state(struct intel_crtc *crtc,
+				     struct intel_crtc_state *pipe_config,
+				     unsigned long *power_domain_mask)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
+	u32 tmp;
+
+	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
+
+	/*
+	 * XXX: Do intel_display_power_get_if_enabled before reading this (for
+	 * consistency and less surprising code; it's in always on power).
+	 */
+	tmp = I915_READ(TRANS_DDI_FUNC_CTL(TRANSCODER_EDP));
+	if (tmp & TRANS_DDI_FUNC_ENABLE) {
+		enum pipe trans_edp_pipe;
+		switch (tmp & TRANS_DDI_EDP_INPUT_MASK) {
+		default:
+			WARN(1, "unknown pipe linked to edp transcoder\n");
+		case TRANS_DDI_EDP_INPUT_A_ONOFF:
+		case TRANS_DDI_EDP_INPUT_A_ON:
+			trans_edp_pipe = PIPE_A;
+			break;
+		case TRANS_DDI_EDP_INPUT_B_ONOFF:
+			trans_edp_pipe = PIPE_B;
+			break;
+		case TRANS_DDI_EDP_INPUT_C_ONOFF:
+			trans_edp_pipe = PIPE_C;
+			break;
+		}
+
+		if (trans_edp_pipe == crtc->pipe)
+			pipe_config->cpu_transcoder = TRANSCODER_EDP;
+	}
+
+	power_domain = POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder);
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
+		return false;
+	*power_domain_mask |= BIT(power_domain);
+
+	tmp = I915_READ(PIPECONF(pipe_config->cpu_transcoder));
+
+	return tmp & PIPECONF_ENABLE;
+}
+
+static bool bxt_get_dsi_transcoder_state(struct intel_crtc *crtc,
+					 struct intel_crtc_state *pipe_config,
+					 unsigned long *power_domain_mask)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
+	enum port port;
+	enum transcoder cpu_transcoder;
+	u32 tmp;
+
+	pipe_config->has_dsi_encoder = false;
+
+	for_each_port_masked(port, BIT(PORT_A) | BIT(PORT_C)) {
+		if (port == PORT_A)
+			cpu_transcoder = TRANSCODER_DSI_A;
+		else
+			cpu_transcoder = TRANSCODER_DSI_C;
+
+		power_domain = POWER_DOMAIN_TRANSCODER(cpu_transcoder);
+		if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
+			continue;
+		*power_domain_mask |= BIT(power_domain);
+
+		/*
+		 * The PLL needs to be enabled with a valid divider
+		 * configuration, otherwise accessing DSI registers will hang
+		 * the machine. See BSpec North Display Engine
+		 * registers/MIPI[BXT]. We can break out here early, since we
+		 * need the same DSI PLL to be enabled for both DSI ports.
+		 */
+		if (!intel_dsi_pll_is_enabled(dev_priv))
+			break;
+
+		/* XXX: this works for video mode only */
+		tmp = I915_READ(BXT_MIPI_PORT_CTRL(port));
+		if (!(tmp & DPI_ENABLE))
+			continue;
+
+		tmp = I915_READ(MIPI_CTRL(port));
+		if ((tmp & BXT_PIPE_SELECT_MASK) != BXT_PIPE_SELECT(crtc->pipe))
+			continue;
+
+		pipe_config->cpu_transcoder = cpu_transcoder;
+		pipe_config->has_dsi_encoder = true;
 		break;
 	}
+
+	return pipe_config->has_dsi_encoder;
 }
 
 static void haswell_get_ddi_port_state(struct intel_crtc *crtc,
@@ -9909,11 +9959,10 @@ static void haswell_get_ddi_port_state(struct intel_crtc *crtc,
 	else
 		haswell_get_ddi_pll(dev_priv, port, pipe_config);
 
-	if (pipe_config->shared_dpll >= 0) {
-		pll = &dev_priv->shared_dplls[pipe_config->shared_dpll];
-
-		WARN_ON(!pll->get_hw_state(dev_priv, pll,
-					   &pipe_config->dpll_hw_state));
+	pll = pipe_config->shared_dpll;
+	if (pll) {
+		WARN_ON(!pll->funcs.get_hw_state(dev_priv, pll,
+						 &pipe_config->dpll_hw_state));
 	}
 
 	/*
@@ -9940,53 +9989,37 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum intel_display_power_domain power_domain;
 	unsigned long power_domain_mask;
-	uint32_t tmp;
-	bool ret;
+	bool active;
 
 	power_domain = POWER_DOMAIN_PIPE(crtc->pipe);
 	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 	power_domain_mask = BIT(power_domain);
 
-	ret = false;
-
-	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
-	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
+	pipe_config->shared_dpll = NULL;
 
-	tmp = I915_READ(TRANS_DDI_FUNC_CTL(TRANSCODER_EDP));
-	if (tmp & TRANS_DDI_FUNC_ENABLE) {
-		enum pipe trans_edp_pipe;
-		switch (tmp & TRANS_DDI_EDP_INPUT_MASK) {
-		default:
-			WARN(1, "unknown pipe linked to edp transcoder\n");
-		case TRANS_DDI_EDP_INPUT_A_ONOFF:
-		case TRANS_DDI_EDP_INPUT_A_ON:
-			trans_edp_pipe = PIPE_A;
-			break;
-		case TRANS_DDI_EDP_INPUT_B_ONOFF:
-			trans_edp_pipe = PIPE_B;
-			break;
-		case TRANS_DDI_EDP_INPUT_C_ONOFF:
-			trans_edp_pipe = PIPE_C;
-			break;
-		}
+	active = hsw_get_transcoder_state(crtc, pipe_config, &power_domain_mask);
 
-		if (trans_edp_pipe == crtc->pipe)
-			pipe_config->cpu_transcoder = TRANSCODER_EDP;
+	if (IS_BROXTON(dev_priv)) {
+		bxt_get_dsi_transcoder_state(crtc, pipe_config,
+					     &power_domain_mask);
+		WARN_ON(active && pipe_config->has_dsi_encoder);
+		if (pipe_config->has_dsi_encoder)
+			active = true;
 	}
 
-	power_domain = POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder);
-	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
+	if (!active)
 		goto out;
-	power_domain_mask |= BIT(power_domain);
 
-	tmp = I915_READ(PIPECONF(pipe_config->cpu_transcoder));
-	if (!(tmp & PIPECONF_ENABLE))
-		goto out;
+	if (!pipe_config->has_dsi_encoder) {
+		haswell_get_ddi_port_state(crtc, pipe_config);
+		intel_get_pipe_timings(crtc, pipe_config);
+	}
 
-	haswell_get_ddi_port_state(crtc, pipe_config);
+	intel_get_pipe_src_size(crtc, pipe_config);
 
-	intel_get_pipe_timings(crtc, pipe_config);
+	pipe_config->gamma_mode =
+		I915_READ(GAMMA_MODE(crtc->pipe)) & GAMMA_MODE_MODE_MASK;
 
 	if (INTEL_INFO(dev)->gen >= 9) {
 		skl_init_scalers(dev, crtc, pipe_config);
@@ -10010,20 +10043,19 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 		pipe_config->ips_enabled = hsw_crtc_supports_ips(crtc) &&
 			(I915_READ(IPS_CTL) & IPS_ENABLE);
 
-	if (pipe_config->cpu_transcoder != TRANSCODER_EDP) {
+	if (pipe_config->cpu_transcoder != TRANSCODER_EDP &&
+	    !transcoder_is_dsi(pipe_config->cpu_transcoder)) {
 		pipe_config->pixel_multiplier =
 			I915_READ(PIPE_MULT(pipe_config->cpu_transcoder)) + 1;
 	} else {
 		pipe_config->pixel_multiplier = 1;
 	}
 
-	ret = true;
-
 out:
 	for_each_power_domain(power_domain, power_domain_mask)
 		intel_display_power_put(dev_priv, power_domain);
 
-	return ret;
+	return active;
 }
 
 static void i845_update_cursor(struct drm_crtc *crtc, u32 base,
@@ -10216,21 +10248,6 @@ static bool cursor_size_ok(struct drm_device *dev,
 	return true;
 }
 
-static void intel_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green,
-				 u16 *blue, uint32_t start, uint32_t size)
-{
-	int end = (start + size > 256) ? 256 : start + size, i;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-
-	for (i = start; i < end; i++) {
-		intel_crtc->lut_r[i] = red[i] >> 8;
-		intel_crtc->lut_g[i] = green[i] >> 8;
-		intel_crtc->lut_b[i] = blue[i] >> 8;
-	}
-
-	intel_crtc_load_lut(crtc);
-}
-
 /* VESA 640x480x72Hz mode to set on the pipe */
 static struct drm_display_mode load_detect_mode = {
 	DRM_MODE("640x480", DRM_MODE_TYPE_DEFAULT, 31500, 640, 664,
@@ -10718,19 +10735,18 @@ int intel_dotclock_calculate(int link_freq,
 static void ironlake_pch_clock_get(struct intel_crtc *crtc,
 				   struct intel_crtc_state *pipe_config)
 {
-	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 
 	/* read out port_clock from the DPLL */
 	i9xx_crtc_clock_get(crtc, pipe_config);
 
 	/*
-	 * This value does not include pixel_multiplier.
-	 * We will check that port_clock and adjusted_mode.crtc_clock
-	 * agree once we know their relationship in the encoder's
-	 * get_config() function.
+	 * In case there is an active pipe without active ports,
+	 * we may need some idea for the dotclock anyway.
+	 * Calculate one based on the FDI configuration.
 	 */
 	pipe_config->base.adjusted_mode.crtc_clock =
-		intel_dotclock_calculate(intel_fdi_link_freq(dev) * 10000,
+		intel_dotclock_calculate(intel_fdi_link_freq(dev_priv, pipe_config),
 					 &pipe_config->fdi_m_n);
 }
 
@@ -10849,7 +10865,7 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 	struct drm_plane *primary = crtc->base.primary;
 
 	mutex_lock(&dev->struct_mutex);
-	intel_unpin_fb_obj(work->old_fb, primary->state);
+	intel_unpin_fb_obj(work->old_fb, primary->state->rotation);
 	drm_gem_object_unreference(&work->pending_flip_obj->base);
 
 	if (work->flip_queued_req)
@@ -10923,9 +10939,10 @@ static bool page_flip_finished(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned reset_counter;
 
-	if (i915_reset_in_progress(&dev_priv->gpu_error) ||
-	    crtc->reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
+	reset_counter = i915_reset_counter(&dev_priv->gpu_error);
+	if (crtc->reset_counter != reset_counter)
 		return true;
 
 	/*
@@ -11003,7 +11020,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
 	int ret;
@@ -11019,13 +11036,13 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
 	else
 		flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
-	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | flip_mask);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_emit(ring, MI_DISPLAY_FLIP |
+	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | flip_mask);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_emit(engine, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
-	intel_ring_emit(ring, 0); /* aux display base address, unused */
+	intel_ring_emit(engine, fb->pitches[0]);
+	intel_ring_emit(engine, intel_crtc->unpin_work->gtt_offset);
+	intel_ring_emit(engine, 0); /* aux display base address, unused */
 
 	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
@@ -11038,7 +11055,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
 	int ret;
@@ -11051,13 +11068,13 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
 	else
 		flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
-	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | flip_mask);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 |
+	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | flip_mask);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_emit(engine, MI_DISPLAY_FLIP_I915 |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
-	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_emit(engine, fb->pitches[0]);
+	intel_ring_emit(engine, intel_crtc->unpin_work->gtt_offset);
+	intel_ring_emit(engine, MI_NOOP);
 
 	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
@@ -11070,7 +11087,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t pf, pipesrc;
@@ -11084,10 +11101,10 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	 * Display Registers (which do not change across a page-flip)
 	 * so we need only reprogram the base address.
 	 */
-	intel_ring_emit(ring, MI_DISPLAY_FLIP |
+	intel_ring_emit(engine, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset |
+	intel_ring_emit(engine, fb->pitches[0]);
+	intel_ring_emit(engine, intel_crtc->unpin_work->gtt_offset |
 			obj->tiling_mode);
 
 	/* XXX Enabling the panel-fitter across page-flip is so far
@@ -11096,7 +11113,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	 */
 	pf = 0;
 	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
-	intel_ring_emit(ring, pf | pipesrc);
+	intel_ring_emit(engine, pf | pipesrc);
 
 	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
@@ -11109,7 +11126,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t pf, pipesrc;
@@ -11119,10 +11136,10 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_DISPLAY_FLIP |
+	intel_ring_emit(engine, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(ring, fb->pitches[0] | obj->tiling_mode);
-	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
+	intel_ring_emit(engine, fb->pitches[0] | obj->tiling_mode);
+	intel_ring_emit(engine, intel_crtc->unpin_work->gtt_offset);
 
 	/* Contrary to the suggestions in the documentation,
 	 * "Enable Panel Fitter" does not seem to be required when page
@@ -11132,7 +11149,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	 */
 	pf = 0;
 	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
-	intel_ring_emit(ring, pf | pipesrc);
+	intel_ring_emit(engine, pf | pipesrc);
 
 	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
@@ -11145,7 +11162,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t plane_bit = 0;
 	int len, ret;
@@ -11166,7 +11183,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	}
 
 	len = 4;
-	if (ring->id == RCS) {
+	if (engine->id == RCS) {
 		len += 6;
 		/*
 		 * On Gen 8, SRM is now taking an extra dword to accommodate
@@ -11204,36 +11221,36 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	 * for the RCS also doesn't appear to drop events. Setting the DERRMR
 	 * to zero does lead to lockups within MI_DISPLAY_FLIP.
 	 */
-	if (ring->id == RCS) {
-		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit_reg(ring, DERRMR);
-		intel_ring_emit(ring, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
-					DERRMR_PIPEB_PRI_FLIP_DONE |
-					DERRMR_PIPEC_PRI_FLIP_DONE));
+	if (engine->id == RCS) {
+		intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit_reg(engine, DERRMR);
+		intel_ring_emit(engine, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
+					  DERRMR_PIPEB_PRI_FLIP_DONE |
+					  DERRMR_PIPEC_PRI_FLIP_DONE));
 		if (IS_GEN8(dev))
-			intel_ring_emit(ring, MI_STORE_REGISTER_MEM_GEN8 |
+			intel_ring_emit(engine, MI_STORE_REGISTER_MEM_GEN8 |
 					      MI_SRM_LRM_GLOBAL_GTT);
 		else
-			intel_ring_emit(ring, MI_STORE_REGISTER_MEM |
+			intel_ring_emit(engine, MI_STORE_REGISTER_MEM |
 					      MI_SRM_LRM_GLOBAL_GTT);
-		intel_ring_emit_reg(ring, DERRMR);
-		intel_ring_emit(ring, ring->scratch.gtt_offset + 256);
+		intel_ring_emit_reg(engine, DERRMR);
+		intel_ring_emit(engine, engine->scratch.gtt_offset + 256);
 		if (IS_GEN8(dev)) {
-			intel_ring_emit(ring, 0);
-			intel_ring_emit(ring, MI_NOOP);
+			intel_ring_emit(engine, 0);
+			intel_ring_emit(engine, MI_NOOP);
 		}
 	}
 
-	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit);
-	intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode));
-	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
-	intel_ring_emit(ring, (MI_NOOP));
+	intel_ring_emit(engine, MI_DISPLAY_FLIP_I915 | plane_bit);
+	intel_ring_emit(engine, (fb->pitches[0] | obj->tiling_mode));
+	intel_ring_emit(engine, intel_crtc->unpin_work->gtt_offset);
+	intel_ring_emit(engine, (MI_NOOP));
 
 	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
 }
 
-static bool use_mmio_flip(struct intel_engine_cs *ring,
+static bool use_mmio_flip(struct intel_engine_cs *engine,
 			  struct drm_i915_gem_object *obj)
 {
 	/*
@@ -11244,10 +11261,10 @@ static bool use_mmio_flip(struct intel_engine_cs *ring,
 	 * So using MMIO flips there would disrupt this mechanism.
 	 */
 
-	if (ring == NULL)
+	if (engine == NULL)
 		return true;
 
-	if (INTEL_INFO(ring->dev)->gen < 5)
+	if (INTEL_INFO(engine->dev)->gen < 5)
 		return false;
 
 	if (i915.use_mmio_flip < 0)
@@ -11261,7 +11278,7 @@ static bool use_mmio_flip(struct intel_engine_cs *ring,
 						       false))
 		return true;
 	else
-		return ring != i915_gem_request_get_ring(obj->last_write_req);
+		return engine != i915_gem_request_get_engine(obj->last_write_req);
 }
 
 static void skl_do_mmio_flip(struct intel_crtc *intel_crtc,
@@ -11379,7 +11396,6 @@ static void intel_mmio_flip_work_func(struct work_struct *work)
 
 	if (mmio_flip->req) {
 		WARN_ON(__i915_wait_request(mmio_flip->req,
-					    mmio_flip->crtc->reset_counter,
 					    false, NULL,
 					    &mmio_flip->i915->rps.mmioflips));
 		i915_gem_request_unreference__unlocked(mmio_flip->req);
@@ -11507,7 +11523,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	struct drm_plane *primary = crtc->primary;
 	enum pipe pipe = intel_crtc->pipe;
 	struct intel_unpin_work *work;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	bool mmio_flip;
 	struct drm_i915_gem_request *request = NULL;
 	int ret;
@@ -11587,28 +11603,33 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (ret)
 		goto cleanup;
 
+	intel_crtc->reset_counter = i915_reset_counter(&dev_priv->gpu_error);
+	if (__i915_reset_in_progress_or_wedged(intel_crtc->reset_counter)) {
+		ret = -EIO;
+		goto cleanup;
+	}
+
 	atomic_inc(&intel_crtc->unpin_work_count);
-	intel_crtc->reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 
 	if (INTEL_INFO(dev)->gen >= 5 || IS_G4X(dev))
 		work->flip_count = I915_READ(PIPE_FLIPCOUNT_G4X(pipe)) + 1;
 
 	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
-		ring = &dev_priv->ring[BCS];
+		engine = &dev_priv->engine[BCS];
 		if (obj->tiling_mode != intel_fb_obj(work->old_fb)->tiling_mode)
 			/* vlv: DISPLAY_FLIP fails to change tiling */
-			ring = NULL;
+			engine = NULL;
 	} else if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
-		ring = &dev_priv->ring[BCS];
+		engine = &dev_priv->engine[BCS];
 	} else if (INTEL_INFO(dev)->gen >= 7) {
-		ring = i915_gem_request_get_ring(obj->last_write_req);
-		if (ring == NULL || ring->id != RCS)
-			ring = &dev_priv->ring[BCS];
+		engine = i915_gem_request_get_engine(obj->last_write_req);
+		if (engine == NULL || engine->id != RCS)
+			engine = &dev_priv->engine[BCS];
 	} else {
-		ring = &dev_priv->ring[RCS];
+		engine = &dev_priv->engine[RCS];
 	}
 
-	mmio_flip = use_mmio_flip(ring, obj);
+	mmio_flip = use_mmio_flip(engine, obj);
 
 	/* When using CS flips, we want to emit semaphores between rings.
 	 * However, when using mmio flips we will create a task to do the
@@ -11616,13 +11637,12 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	 * into the display plane and skip any waits.
 	 */
 	if (!mmio_flip) {
-		ret = i915_gem_object_sync(obj, ring, &request);
+		ret = i915_gem_object_sync(obj, engine, &request);
 		if (ret)
 			goto cleanup_pending;
 	}
 
-	ret = intel_pin_and_fence_fb_obj(crtc->primary, fb,
-					 crtc->primary->state);
+	ret = intel_pin_and_fence_fb_obj(fb, primary->state->rotation);
 	if (ret)
 		goto cleanup_pending;
 
@@ -11639,7 +11659,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 					obj->last_write_req);
 	} else {
 		if (!request) {
-			request = i915_gem_request_alloc(ring, NULL);
+			request = i915_gem_request_alloc(engine, NULL);
 			if (IS_ERR(request)) {
 				ret = PTR_ERR(request);
 				goto cleanup_unpin;
@@ -11672,10 +11692,10 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	return 0;
 
 cleanup_unpin:
-	intel_unpin_fb_obj(fb, crtc->primary->state);
+	intel_unpin_fb_obj(fb, crtc->primary->state->rotation);
 cleanup_pending:
 	if (!IS_ERR_OR_NULL(request))
-		i915_gem_request_cancel(request);
+		i915_add_request_no_flush(request);
 	atomic_dec(&intel_crtc->unpin_work_count);
 	mutex_unlock(&dev->struct_mutex);
 cleanup:
@@ -11725,7 +11745,7 @@ retry:
 
 		if (ret == 0 && event) {
 			spin_lock_irq(&dev->event_lock);
-			drm_send_vblank_event(dev, pipe, event);
+			drm_crtc_send_vblank_event(crtc, event);
 			spin_unlock_irq(&dev->event_lock);
 		}
 	}
@@ -11785,6 +11805,7 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_plane *plane = plane_state->plane;
 	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_plane_state *old_plane_state =
 		to_intel_plane_state(plane->state);
 	int idx = intel_crtc->base.base.id, ret;
@@ -11833,42 +11854,43 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 			 plane->base.id, was_visible, visible,
 			 turn_off, turn_on, mode_changed);
 
-	if (turn_on || turn_off) {
-		pipe_config->wm_changed = true;
+	if (turn_on) {
+		pipe_config->update_wm_pre = true;
+
+		/* must disable cxsr around plane enable/disable */
+		if (plane->type != DRM_PLANE_TYPE_CURSOR)
+			pipe_config->disable_cxsr = true;
+	} else if (turn_off) {
+		pipe_config->update_wm_post = true;
 
 		/* must disable cxsr around plane enable/disable */
 		if (plane->type != DRM_PLANE_TYPE_CURSOR)
 			pipe_config->disable_cxsr = true;
 	} else if (intel_wm_need_update(plane, plane_state)) {
-		pipe_config->wm_changed = true;
+		/* FIXME bollocks */
+		pipe_config->update_wm_pre = true;
+		pipe_config->update_wm_post = true;
 	}
 
-	if (visible || was_visible)
-		intel_crtc->atomic.fb_bits |=
-			to_intel_plane(plane)->frontbuffer_bit;
+	/* Pre-gen9 platforms need two-step watermark updates */
+	if ((pipe_config->update_wm_pre || pipe_config->update_wm_post) &&
+	    INTEL_INFO(dev)->gen < 9 && dev_priv->display.optimize_watermarks)
+		to_intel_crtc_state(crtc_state)->wm.need_postvbl_update = true;
 
-	switch (plane->type) {
-	case DRM_PLANE_TYPE_PRIMARY:
-		intel_crtc->atomic.post_enable_primary = turn_on;
-		intel_crtc->atomic.update_fbc = true;
+	if (visible || was_visible)
+		pipe_config->fb_bits |= to_intel_plane(plane)->frontbuffer_bit;
 
-		break;
-	case DRM_PLANE_TYPE_CURSOR:
-		break;
-	case DRM_PLANE_TYPE_OVERLAY:
-		/*
-		 * WaCxSRDisabledForSpriteScaling:ivb
-		 *
-		 * cstate->update_wm was already set above, so this flag will
-		 * take effect when we commit and program watermarks.
-		 */
-		if (IS_IVYBRIDGE(dev) &&
-		    needs_scaling(to_intel_plane_state(plane_state)) &&
-		    !needs_scaling(old_plane_state))
-			pipe_config->disable_lp_wm = true;
+	/*
+	 * WaCxSRDisabledForSpriteScaling:ivb
+	 *
+	 * cstate->update_wm was already set above, so this flag will
+	 * take effect when we commit and program watermarks.
+	 */
+	if (plane->type == DRM_PLANE_TYPE_OVERLAY && IS_IVYBRIDGE(dev) &&
+	    needs_scaling(to_intel_plane_state(plane_state)) &&
+	    !needs_scaling(old_plane_state))
+		pipe_config->disable_lp_wm = true;
 
-		break;
-	}
 	return 0;
 }
 
@@ -11940,22 +11962,49 @@ static int intel_crtc_atomic_check(struct drm_crtc *crtc,
 	}
 
 	if (mode_changed && !crtc_state->active)
-		pipe_config->wm_changed = true;
+		pipe_config->update_wm_post = true;
 
 	if (mode_changed && crtc_state->enable &&
 	    dev_priv->display.crtc_compute_clock &&
-	    !WARN_ON(pipe_config->shared_dpll != DPLL_ID_PRIVATE)) {
+	    !WARN_ON(pipe_config->shared_dpll)) {
 		ret = dev_priv->display.crtc_compute_clock(intel_crtc,
 							   pipe_config);
 		if (ret)
 			return ret;
 	}
 
+	if (crtc_state->color_mgmt_changed) {
+		ret = intel_color_check(crtc, crtc_state);
+		if (ret)
+			return ret;
+	}
+
 	ret = 0;
 	if (dev_priv->display.compute_pipe_wm) {
-		ret = dev_priv->display.compute_pipe_wm(intel_crtc, state);
-		if (ret)
+		ret = dev_priv->display.compute_pipe_wm(pipe_config);
+		if (ret) {
+			DRM_DEBUG_KMS("Target pipe watermarks are invalid\n");
 			return ret;
+		}
+	}
+
+	if (dev_priv->display.compute_intermediate_wm &&
+	    !to_intel_atomic_state(state)->skip_intermediate_wm) {
+		if (WARN_ON(!dev_priv->display.compute_pipe_wm))
+			return 0;
+
+		/*
+		 * Calculate 'intermediate' watermarks that satisfy both the
+		 * old state and the new state.  We can program these
+		 * immediately.
+		 */
+		ret = dev_priv->display.compute_intermediate_wm(crtc->dev,
+								intel_crtc,
+								pipe_config);
+		if (ret) {
+			DRM_DEBUG_KMS("No valid intermediate pipe watermarks are possible\n");
+			return ret;
+		}
 	}
 
 	if (INTEL_INFO(dev)->gen >= 9) {
@@ -11972,7 +12021,6 @@ static int intel_crtc_atomic_check(struct drm_crtc *crtc,
 
 static const struct drm_crtc_helper_funcs intel_helper_funcs = {
 	.mode_set_base_atomic = intel_pipe_set_base_atomic,
-	.load_lut = intel_crtc_load_lut,
 	.atomic_begin = intel_begin_crtc_commit,
 	.atomic_flush = intel_finish_crtc_commit,
 	.atomic_check = intel_crtc_atomic_check,
@@ -11983,11 +12031,16 @@ static void intel_modeset_update_connector_atomic_state(struct drm_device *dev)
 	struct intel_connector *connector;
 
 	for_each_intel_connector(dev, connector) {
+		if (connector->base.state->crtc)
+			drm_connector_unreference(&connector->base);
+
 		if (connector->base.encoder) {
 			connector->base.state->best_encoder =
 				connector->base.encoder;
 			connector->base.state->crtc =
 				connector->base.encoder->crtc;
+
+			drm_connector_reference(&connector->base);
 		} else {
 			connector->base.state->best_encoder = NULL;
 			connector->base.state->crtc = NULL;
@@ -12089,7 +12142,7 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 	DRM_DEBUG_KMS("[CRTC:%d]%s config %p for pipe %c\n", crtc->base.base.id,
 		      context, pipe_config, pipe_name(crtc->pipe));
 
-	DRM_DEBUG_KMS("cpu_transcoder: %c\n", transcoder_name(pipe_config->cpu_transcoder));
+	DRM_DEBUG_KMS("cpu_transcoder: %s\n", transcoder_name(pipe_config->cpu_transcoder));
 	DRM_DEBUG_KMS("pipe bpp: %i, dithering: %i\n",
 		      pipe_config->pipe_bpp, pipe_config->dither);
 	DRM_DEBUG_KMS("fdi/pch: %i, lanes: %i, gmch_m: %u, gmch_n: %u, link_m: %u, link_n: %u, tu: %u\n",
@@ -12165,7 +12218,7 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 			      pipe_config->dpll_hw_state.cfgcr1,
 			      pipe_config->dpll_hw_state.cfgcr2);
 	} else if (HAS_DDI(dev)) {
-		DRM_DEBUG_KMS("ddi_pll_sel: %u; dpll_hw_state: wrpll: 0x%x spll: 0x%x\n",
+		DRM_DEBUG_KMS("ddi_pll_sel: 0x%x; dpll_hw_state: wrpll: 0x%x spll: 0x%x\n",
 			      pipe_config->ddi_pll_sel,
 			      pipe_config->dpll_hw_state.wrpll,
 			      pipe_config->dpll_hw_state.spll);
@@ -12268,7 +12321,7 @@ clear_intel_crtc_state(struct intel_crtc_state *crtc_state)
 	struct drm_crtc_state tmp_state;
 	struct intel_crtc_scaler_state scaler_state;
 	struct intel_dpll_hw_state dpll_hw_state;
-	enum intel_dpll_id shared_dpll;
+	struct intel_shared_dpll *shared_dpll;
 	uint32_t ddi_pll_sel;
 	bool force_thru;
 
@@ -12538,6 +12591,15 @@ intel_pipe_config_compare(struct drm_device *dev,
 		ret = false; \
 	}
 
+#define PIPE_CONF_CHECK_P(name)	\
+	if (current_config->name != pipe_config->name) { \
+		INTEL_ERR_OR_DBG_KMS("mismatch in " #name " " \
+			  "(expected %p, found %p)\n", \
+			  current_config->name, \
+			  pipe_config->name); \
+		ret = false; \
+	}
+
 #define PIPE_CONF_CHECK_M_N(name) \
 	if (!intel_compare_link_m_n(&current_config->name, \
 				    &pipe_config->name,\
@@ -12558,6 +12620,11 @@ intel_pipe_config_compare(struct drm_device *dev,
 		ret = false; \
 	}
 
+/* This is required for BDW+ where there is only one set of registers for
+ * switching between high and low RR.
+ * This macro can be used whenever a comparison has to be made between one
+ * hw state and multiple sw state variables.
+ */
 #define PIPE_CONF_CHECK_M_N_ALT(name, alt_name) \
 	if (!intel_compare_link_m_n(&current_config->name, \
 				    &pipe_config->name, adjust) && \
@@ -12585,22 +12652,6 @@ intel_pipe_config_compare(struct drm_device *dev,
 		ret = false; \
 	}
 
-/* This is required for BDW+ where there is only one set of registers for
- * switching between high and low RR.
- * This macro can be used whenever a comparison has to be made between one
- * hw state and multiple sw state variables.
- */
-#define PIPE_CONF_CHECK_I_ALT(name, alt_name) \
-	if ((current_config->name != pipe_config->name) && \
-		(current_config->alt_name != pipe_config->name)) { \
-			INTEL_ERR_OR_DBG_KMS("mismatch in " #name " " \
-				  "(expected %i or %i, found %i)\n", \
-				  current_config->name, \
-				  current_config->alt_name, \
-				  pipe_config->name); \
-			ret = false; \
-	}
-
 #define PIPE_CONF_CHECK_FLAGS(name, mask)	\
 	if ((current_config->name ^ pipe_config->name) & (mask)) { \
 		INTEL_ERR_OR_DBG_KMS("mismatch in " #name "(" #mask ") " \
@@ -12681,7 +12732,7 @@ intel_pipe_config_compare(struct drm_device *dev,
 	PIPE_CONF_CHECK_X(gmch_pfit.control);
 	/* pfit ratios are autocomputed by the hw on gen4+ */
 	if (INTEL_INFO(dev)->gen < 4)
-		PIPE_CONF_CHECK_I(gmch_pfit.pgm_ratios);
+		PIPE_CONF_CHECK_X(gmch_pfit.pgm_ratios);
 	PIPE_CONF_CHECK_X(gmch_pfit.lvds_border_bits);
 
 	if (!adjust) {
@@ -12705,7 +12756,7 @@ intel_pipe_config_compare(struct drm_device *dev,
 
 	PIPE_CONF_CHECK_X(ddi_pll_sel);
 
-	PIPE_CONF_CHECK_I(shared_dpll);
+	PIPE_CONF_CHECK_P(shared_dpll);
 	PIPE_CONF_CHECK_X(dpll_hw_state.dpll);
 	PIPE_CONF_CHECK_X(dpll_hw_state.dpll_md);
 	PIPE_CONF_CHECK_X(dpll_hw_state.fp0);
@@ -12716,6 +12767,9 @@ intel_pipe_config_compare(struct drm_device *dev,
 	PIPE_CONF_CHECK_X(dpll_hw_state.cfgcr1);
 	PIPE_CONF_CHECK_X(dpll_hw_state.cfgcr2);
 
+	PIPE_CONF_CHECK_X(dsi_pll.ctrl);
+	PIPE_CONF_CHECK_X(dsi_pll.div);
+
 	if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5)
 		PIPE_CONF_CHECK_I(pipe_bpp);
 
@@ -12724,7 +12778,7 @@ intel_pipe_config_compare(struct drm_device *dev,
 
 #undef PIPE_CONF_CHECK_X
 #undef PIPE_CONF_CHECK_I
-#undef PIPE_CONF_CHECK_I_ALT
+#undef PIPE_CONF_CHECK_P
 #undef PIPE_CONF_CHECK_FLAGS
 #undef PIPE_CONF_CHECK_CLOCK_FUZZY
 #undef PIPE_CONF_QUIRK
@@ -12733,48 +12787,61 @@ intel_pipe_config_compare(struct drm_device *dev,
 	return ret;
 }
 
-static void check_wm_state(struct drm_device *dev)
+static void intel_pipe_config_sanity_check(struct drm_i915_private *dev_priv,
+					   const struct intel_crtc_state *pipe_config)
+{
+	if (pipe_config->has_pch_encoder) {
+		int fdi_dotclock = intel_dotclock_calculate(intel_fdi_link_freq(dev_priv, pipe_config),
+							    &pipe_config->fdi_m_n);
+		int dotclock = pipe_config->base.adjusted_mode.crtc_clock;
+
+		/*
+		 * FDI already provided one idea for the dotclock.
+		 * Yell if the encoder disagrees.
+		 */
+		WARN(!intel_fuzzy_clock_check(fdi_dotclock, dotclock),
+		     "FDI dotclock and encoder dotclock mismatch, fdi: %i, encoder: %i\n",
+		     fdi_dotclock, dotclock);
+	}
+}
+
+static void verify_wm_state(struct drm_crtc *crtc,
+			    struct drm_crtc_state *new_state)
 {
+	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct skl_ddb_allocation hw_ddb, *sw_ddb;
-	struct intel_crtc *intel_crtc;
+	struct skl_ddb_entry *hw_entry, *sw_entry;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	const enum pipe pipe = intel_crtc->pipe;
 	int plane;
 
-	if (INTEL_INFO(dev)->gen < 9)
+	if (INTEL_INFO(dev)->gen < 9 || !new_state->active)
 		return;
 
 	skl_ddb_get_hw_state(dev_priv, &hw_ddb);
 	sw_ddb = &dev_priv->wm.skl_hw.ddb;
 
-	for_each_intel_crtc(dev, intel_crtc) {
-		struct skl_ddb_entry *hw_entry, *sw_entry;
-		const enum pipe pipe = intel_crtc->pipe;
+	/* planes */
+	for_each_plane(dev_priv, pipe, plane) {
+		hw_entry = &hw_ddb.plane[pipe][plane];
+		sw_entry = &sw_ddb->plane[pipe][plane];
 
-		if (!intel_crtc->active)
+		if (skl_ddb_entry_equal(hw_entry, sw_entry))
 			continue;
 
-		/* planes */
-		for_each_plane(dev_priv, pipe, plane) {
-			hw_entry = &hw_ddb.plane[pipe][plane];
-			sw_entry = &sw_ddb->plane[pipe][plane];
-
-			if (skl_ddb_entry_equal(hw_entry, sw_entry))
-				continue;
-
-			DRM_ERROR("mismatch in DDB state pipe %c plane %d "
-				  "(expected (%u,%u), found (%u,%u))\n",
-				  pipe_name(pipe), plane + 1,
-				  sw_entry->start, sw_entry->end,
-				  hw_entry->start, hw_entry->end);
-		}
-
-		/* cursor */
-		hw_entry = &hw_ddb.plane[pipe][PLANE_CURSOR];
-		sw_entry = &sw_ddb->plane[pipe][PLANE_CURSOR];
+		DRM_ERROR("mismatch in DDB state pipe %c plane %d "
+			  "(expected (%u,%u), found (%u,%u))\n",
+			  pipe_name(pipe), plane + 1,
+			  sw_entry->start, sw_entry->end,
+			  hw_entry->start, hw_entry->end);
+	}
 
-		if (skl_ddb_entry_equal(hw_entry, sw_entry))
-			continue;
+	/* cursor */
+	hw_entry = &hw_ddb.plane[pipe][PLANE_CURSOR];
+	sw_entry = &sw_ddb->plane[pipe][PLANE_CURSOR];
 
+	if (!skl_ddb_entry_equal(hw_entry, sw_entry)) {
 		DRM_ERROR("mismatch in DDB state pipe %c cursor "
 			  "(expected (%u,%u), found (%u,%u))\n",
 			  pipe_name(pipe),
@@ -12784,20 +12851,18 @@ static void check_wm_state(struct drm_device *dev)
 }
 
 static void
-check_connector_state(struct drm_device *dev,
-		      struct drm_atomic_state *old_state)
+verify_connector_state(struct drm_device *dev, struct drm_crtc *crtc)
 {
-	struct drm_connector_state *old_conn_state;
 	struct drm_connector *connector;
-	int i;
 
-	for_each_connector_in_state(old_state, connector, old_conn_state, i) {
+	drm_for_each_connector(connector, dev) {
 		struct drm_encoder *encoder = connector->encoder;
 		struct drm_connector_state *state = connector->state;
 
-		/* This also checks the encoder/connector hw state with the
-		 * ->get_hw_state callbacks. */
-		intel_connector_check_state(to_intel_connector(connector));
+		if (state->crtc != crtc)
+			continue;
+
+		intel_connector_verify_state(to_intel_connector(connector));
 
 		I915_STATE_WARN(state->best_encoder != encoder,
 		     "connector's atomic encoder doesn't match legacy encoder\n");
@@ -12805,7 +12870,7 @@ check_connector_state(struct drm_device *dev,
 }
 
 static void
-check_encoder_state(struct drm_device *dev)
+verify_encoder_state(struct drm_device *dev)
 {
 	struct intel_encoder *encoder;
 	struct intel_connector *connector;
@@ -12845,149 +12910,186 @@ check_encoder_state(struct drm_device *dev)
 }
 
 static void
-check_crtc_state(struct drm_device *dev, struct drm_atomic_state *old_state)
+verify_crtc_state(struct drm_crtc *crtc,
+		  struct drm_crtc_state *old_crtc_state,
+		  struct drm_crtc_state *new_crtc_state)
 {
+	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_encoder *encoder;
-	struct drm_crtc_state *old_crtc_state;
-	struct drm_crtc *crtc;
-	int i;
-
-	for_each_crtc_in_state(old_state, crtc, old_crtc_state, i) {
-		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-		struct intel_crtc_state *pipe_config, *sw_config;
-		bool active;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *pipe_config, *sw_config;
+	struct drm_atomic_state *old_state;
+	bool active;
 
-		if (!needs_modeset(crtc->state) &&
-		    !to_intel_crtc_state(crtc->state)->update_pipe)
-			continue;
+	old_state = old_crtc_state->state;
+	__drm_atomic_helper_crtc_destroy_state(old_crtc_state);
+	pipe_config = to_intel_crtc_state(old_crtc_state);
+	memset(pipe_config, 0, sizeof(*pipe_config));
+	pipe_config->base.crtc = crtc;
+	pipe_config->base.state = old_state;
 
-		__drm_atomic_helper_crtc_destroy_state(crtc, old_crtc_state);
-		pipe_config = to_intel_crtc_state(old_crtc_state);
-		memset(pipe_config, 0, sizeof(*pipe_config));
-		pipe_config->base.crtc = crtc;
-		pipe_config->base.state = old_state;
+	DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id);
 
-		DRM_DEBUG_KMS("[CRTC:%d]\n",
-			      crtc->base.id);
+	active = dev_priv->display.get_pipe_config(intel_crtc, pipe_config);
 
-		active = dev_priv->display.get_pipe_config(intel_crtc,
-							   pipe_config);
+	/* hw state is inconsistent with the pipe quirk */
+	if ((intel_crtc->pipe == PIPE_A && dev_priv->quirks & QUIRK_PIPEA_FORCE) ||
+	    (intel_crtc->pipe == PIPE_B && dev_priv->quirks & QUIRK_PIPEB_FORCE))
+		active = new_crtc_state->active;
 
-		/* hw state is inconsistent with the pipe quirk */
-		if ((intel_crtc->pipe == PIPE_A && dev_priv->quirks & QUIRK_PIPEA_FORCE) ||
-		    (intel_crtc->pipe == PIPE_B && dev_priv->quirks & QUIRK_PIPEB_FORCE))
-			active = crtc->state->active;
+	I915_STATE_WARN(new_crtc_state->active != active,
+	     "crtc active state doesn't match with hw state "
+	     "(expected %i, found %i)\n", new_crtc_state->active, active);
 
-		I915_STATE_WARN(crtc->state->active != active,
-		     "crtc active state doesn't match with hw state "
-		     "(expected %i, found %i)\n", crtc->state->active, active);
+	I915_STATE_WARN(intel_crtc->active != new_crtc_state->active,
+	     "transitional active state does not match atomic hw state "
+	     "(expected %i, found %i)\n", new_crtc_state->active, intel_crtc->active);
 
-		I915_STATE_WARN(intel_crtc->active != crtc->state->active,
-		     "transitional active state does not match atomic hw state "
-		     "(expected %i, found %i)\n", crtc->state->active, intel_crtc->active);
+	for_each_encoder_on_crtc(dev, crtc, encoder) {
+		enum pipe pipe;
 
-		for_each_encoder_on_crtc(dev, crtc, encoder) {
-			enum pipe pipe;
+		active = encoder->get_hw_state(encoder, &pipe);
+		I915_STATE_WARN(active != new_crtc_state->active,
+			"[ENCODER:%i] active %i with crtc active %i\n",
+			encoder->base.base.id, active, new_crtc_state->active);
 
-			active = encoder->get_hw_state(encoder, &pipe);
-			I915_STATE_WARN(active != crtc->state->active,
-				"[ENCODER:%i] active %i with crtc active %i\n",
-				encoder->base.base.id, active, crtc->state->active);
+		I915_STATE_WARN(active && intel_crtc->pipe != pipe,
+				"Encoder connected to wrong pipe %c\n",
+				pipe_name(pipe));
 
-			I915_STATE_WARN(active && intel_crtc->pipe != pipe,
-					"Encoder connected to wrong pipe %c\n",
-					pipe_name(pipe));
+		if (active)
+			encoder->get_config(encoder, pipe_config);
+	}
 
-			if (active)
-				encoder->get_config(encoder, pipe_config);
-		}
+	if (!new_crtc_state->active)
+		return;
 
-		if (!crtc->state->active)
-			continue;
+	intel_pipe_config_sanity_check(dev_priv, pipe_config);
 
-		sw_config = to_intel_crtc_state(crtc->state);
-		if (!intel_pipe_config_compare(dev, sw_config,
-					       pipe_config, false)) {
-			I915_STATE_WARN(1, "pipe state doesn't match!\n");
-			intel_dump_pipe_config(intel_crtc, pipe_config,
-					       "[hw state]");
-			intel_dump_pipe_config(intel_crtc, sw_config,
-					       "[sw state]");
-		}
+	sw_config = to_intel_crtc_state(crtc->state);
+	if (!intel_pipe_config_compare(dev, sw_config,
+				       pipe_config, false)) {
+		I915_STATE_WARN(1, "pipe state doesn't match!\n");
+		intel_dump_pipe_config(intel_crtc, pipe_config,
+				       "[hw state]");
+		intel_dump_pipe_config(intel_crtc, sw_config,
+				       "[sw state]");
 	}
 }
 
 static void
-check_shared_dpll_state(struct drm_device *dev)
+verify_single_dpll_state(struct drm_i915_private *dev_priv,
+			 struct intel_shared_dpll *pll,
+			 struct drm_crtc *crtc,
+			 struct drm_crtc_state *new_state)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *crtc;
 	struct intel_dpll_hw_state dpll_hw_state;
-	int i;
+	unsigned crtc_mask;
+	bool active;
 
-	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
-		struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i];
-		int enabled_crtcs = 0, active_crtcs = 0;
-		bool active;
-
-		memset(&dpll_hw_state, 0, sizeof(dpll_hw_state));
+	memset(&dpll_hw_state, 0, sizeof(dpll_hw_state));
 
-		DRM_DEBUG_KMS("%s\n", pll->name);
+	DRM_DEBUG_KMS("%s\n", pll->name);
 
-		active = pll->get_hw_state(dev_priv, pll, &dpll_hw_state);
+	active = pll->funcs.get_hw_state(dev_priv, pll, &dpll_hw_state);
 
-		I915_STATE_WARN(pll->active > hweight32(pll->config.crtc_mask),
-		     "more active pll users than references: %i vs %i\n",
-		     pll->active, hweight32(pll->config.crtc_mask));
-		I915_STATE_WARN(pll->active && !pll->on,
+	if (!(pll->flags & INTEL_DPLL_ALWAYS_ON)) {
+		I915_STATE_WARN(!pll->on && pll->active_mask,
 		     "pll in active use but not on in sw tracking\n");
-		I915_STATE_WARN(pll->on && !pll->active,
-		     "pll in on but not on in use in sw tracking\n");
+		I915_STATE_WARN(pll->on && !pll->active_mask,
+		     "pll is on but not used by any active crtc\n");
 		I915_STATE_WARN(pll->on != active,
 		     "pll on state mismatch (expected %i, found %i)\n",
 		     pll->on, active);
+	}
 
-		for_each_intel_crtc(dev, crtc) {
-			if (crtc->base.state->enable && intel_crtc_to_shared_dpll(crtc) == pll)
-				enabled_crtcs++;
-			if (crtc->active && intel_crtc_to_shared_dpll(crtc) == pll)
-				active_crtcs++;
-		}
-		I915_STATE_WARN(pll->active != active_crtcs,
-		     "pll active crtcs mismatch (expected %i, found %i)\n",
-		     pll->active, active_crtcs);
-		I915_STATE_WARN(hweight32(pll->config.crtc_mask) != enabled_crtcs,
-		     "pll enabled crtcs mismatch (expected %i, found %i)\n",
-		     hweight32(pll->config.crtc_mask), enabled_crtcs);
+	if (!crtc) {
+		I915_STATE_WARN(pll->active_mask & ~pll->config.crtc_mask,
+				"more active pll users than references: %x vs %x\n",
+				pll->active_mask, pll->config.crtc_mask);
 
-		I915_STATE_WARN(pll->on && memcmp(&pll->config.hw_state, &dpll_hw_state,
-				       sizeof(dpll_hw_state)),
-		     "pll hw state mismatch\n");
+		return;
 	}
+
+	crtc_mask = 1 << drm_crtc_index(crtc);
+
+	if (new_state->active)
+		I915_STATE_WARN(!(pll->active_mask & crtc_mask),
+				"pll active mismatch (expected pipe %c in active mask 0x%02x)\n",
+				pipe_name(drm_crtc_index(crtc)), pll->active_mask);
+	else
+		I915_STATE_WARN(pll->active_mask & crtc_mask,
+				"pll active mismatch (didn't expect pipe %c in active mask 0x%02x)\n",
+				pipe_name(drm_crtc_index(crtc)), pll->active_mask);
+
+	I915_STATE_WARN(!(pll->config.crtc_mask & crtc_mask),
+			"pll enabled crtcs mismatch (expected 0x%x in 0x%02x)\n",
+			crtc_mask, pll->config.crtc_mask);
+
+	I915_STATE_WARN(pll->on && memcmp(&pll->config.hw_state,
+					  &dpll_hw_state,
+					  sizeof(dpll_hw_state)),
+			"pll hw state mismatch\n");
 }
 
 static void
-intel_modeset_check_state(struct drm_device *dev,
-			  struct drm_atomic_state *old_state)
+verify_shared_dpll_state(struct drm_device *dev, struct drm_crtc *crtc,
+			 struct drm_crtc_state *old_crtc_state,
+			 struct drm_crtc_state *new_crtc_state)
 {
-	check_wm_state(dev);
-	check_connector_state(dev, old_state);
-	check_encoder_state(dev);
-	check_crtc_state(dev, old_state);
-	check_shared_dpll_state(dev);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc_state *old_state = to_intel_crtc_state(old_crtc_state);
+	struct intel_crtc_state *new_state = to_intel_crtc_state(new_crtc_state);
+
+	if (new_state->shared_dpll)
+		verify_single_dpll_state(dev_priv, new_state->shared_dpll, crtc, new_crtc_state);
+
+	if (old_state->shared_dpll &&
+	    old_state->shared_dpll != new_state->shared_dpll) {
+		unsigned crtc_mask = 1 << drm_crtc_index(crtc);
+		struct intel_shared_dpll *pll = old_state->shared_dpll;
+
+		I915_STATE_WARN(pll->active_mask & crtc_mask,
+				"pll active mismatch (didn't expect pipe %c in active mask)\n",
+				pipe_name(drm_crtc_index(crtc)));
+		I915_STATE_WARN(pll->config.crtc_mask & crtc_mask,
+				"pll enabled crtcs mismatch (found %x in enabled mask)\n",
+				pipe_name(drm_crtc_index(crtc)));
+	}
 }
 
-void ironlake_check_encoder_dotclock(const struct intel_crtc_state *pipe_config,
-				     int dotclock)
+static void
+intel_modeset_verify_crtc(struct drm_crtc *crtc,
+			 struct drm_crtc_state *old_state,
+			 struct drm_crtc_state *new_state)
 {
-	/*
-	 * FDI already provided one idea for the dotclock.
-	 * Yell if the encoder disagrees.
-	 */
-	WARN(!intel_fuzzy_clock_check(pipe_config->base.adjusted_mode.crtc_clock, dotclock),
-	     "FDI dotclock and encoder dotclock mismatch, fdi: %i, encoder: %i\n",
-	     pipe_config->base.adjusted_mode.crtc_clock, dotclock);
+	if (!needs_modeset(new_state) &&
+	    !to_intel_crtc_state(new_state)->update_pipe)
+		return;
+
+	verify_wm_state(crtc, new_state);
+	verify_connector_state(crtc->dev, crtc);
+	verify_crtc_state(crtc, old_state, new_state);
+	verify_shared_dpll_state(crtc->dev, crtc, old_state, new_state);
+}
+
+static void
+verify_disabled_dpll_state(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int i;
+
+	for (i = 0; i < dev_priv->num_shared_dpll; i++)
+		verify_single_dpll_state(dev_priv, &dev_priv->shared_dplls[i], NULL, NULL);
+}
+
+static void
+intel_modeset_verify_disabled(struct drm_device *dev)
+{
+	verify_encoder_state(dev);
+	verify_connector_state(dev, NULL);
+	verify_disabled_dpll_state(dev);
 }
 
 static void update_scanline_offset(struct intel_crtc *crtc)
@@ -13042,20 +13144,21 @@ static void intel_modeset_clear_plls(struct drm_atomic_state *state)
 
 	for_each_crtc_in_state(state, crtc, crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-		int old_dpll = to_intel_crtc_state(crtc->state)->shared_dpll;
+		struct intel_shared_dpll *old_dpll =
+			to_intel_crtc_state(crtc->state)->shared_dpll;
 
 		if (!needs_modeset(crtc_state))
 			continue;
 
-		to_intel_crtc_state(crtc_state)->shared_dpll = DPLL_ID_PRIVATE;
+		to_intel_crtc_state(crtc_state)->shared_dpll = NULL;
 
-		if (old_dpll == DPLL_ID_PRIVATE)
+		if (!old_dpll)
 			continue;
 
 		if (!shared_dpll)
 			shared_dpll = intel_atomic_get_shared_dpll_state(state);
 
-		shared_dpll[old_dpll].crtc_mask &= ~(1 << intel_crtc->pipe);
+		intel_shared_dpll_config_put(shared_dpll, old_dpll, intel_crtc);
 	}
 }
 
@@ -13267,9 +13370,6 @@ static int intel_atomic_check(struct drm_device *dev,
 		struct intel_crtc_state *pipe_config =
 			to_intel_crtc_state(crtc_state);
 
-		memset(&to_intel_crtc(crtc)->atomic, 0,
-		       sizeof(struct intel_crtc_atomic_commit));
-
 		/* Catch I915_MODE_FLAG_INHERITED */
 		if (crtc_state->mode.private_flags != crtc->state->mode.private_flags)
 			crtc_state->mode_changed = true;
@@ -13335,7 +13435,7 @@ static int intel_atomic_check(struct drm_device *dev,
 
 static int intel_atomic_prepare_commit(struct drm_device *dev,
 				       struct drm_atomic_state *state,
-				       bool async)
+				       bool nonblock)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_plane_state *plane_state;
@@ -13344,8 +13444,8 @@ static int intel_atomic_prepare_commit(struct drm_device *dev,
 	struct drm_crtc *crtc;
 	int i, ret;
 
-	if (async) {
-		DRM_DEBUG_KMS("i915 does not yet support async commit\n");
+	if (nonblock) {
+		DRM_DEBUG_KMS("i915 does not yet support nonblocking commit\n");
 		return -EINVAL;
 	}
 
@@ -13366,12 +13466,9 @@ static int intel_atomic_prepare_commit(struct drm_device *dev,
 		return ret;
 
 	ret = drm_atomic_helper_prepare_planes(dev, state);
-	if (!ret && !async && !i915_reset_in_progress(&dev_priv->gpu_error)) {
-		u32 reset_counter;
-
-		reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
-		mutex_unlock(&dev->struct_mutex);
+	mutex_unlock(&dev->struct_mutex);
 
+	if (!ret && !nonblock) {
 		for_each_plane_in_state(state, plane, plane_state, i) {
 			struct intel_plane_state *intel_plane_state =
 				to_intel_plane_state(plane_state);
@@ -13380,25 +13477,18 @@ static int intel_atomic_prepare_commit(struct drm_device *dev,
 				continue;
 
 			ret = __i915_wait_request(intel_plane_state->wait_req,
-						  reset_counter, true,
-						  NULL, NULL);
-
-			/* Swallow -EIO errors to allow updates during hw lockup. */
-			if (ret == -EIO)
-				ret = 0;
-
-			if (ret)
+						  true, NULL, NULL);
+			if (ret) {
+				/* Any hang should be swallowed by the wait */
+				WARN_ON(ret == -EIO);
+				mutex_lock(&dev->struct_mutex);
+				drm_atomic_helper_cleanup_planes(dev, state);
+				mutex_unlock(&dev->struct_mutex);
 				break;
+			}
 		}
-
-		if (!ret)
-			return 0;
-
-		mutex_lock(&dev->struct_mutex);
-		drm_atomic_helper_cleanup_planes(dev, state);
 	}
 
-	mutex_unlock(&dev->struct_mutex);
 	return ret;
 }
 
@@ -13440,7 +13530,7 @@ static void intel_atomic_wait_for_vblanks(struct drm_device *dev,
 					drm_crtc_vblank_count(crtc),
 				msecs_to_jiffies(50));
 
-		WARN_ON(!lret);
+		WARN(!lret, "pipe %c vblank wait timed out\n", pipe_name(pipe));
 
 		drm_crtc_vblank_put(crtc);
 	}
@@ -13453,12 +13543,12 @@ static bool needs_vblank_wait(struct intel_crtc_state *crtc_state)
 		return true;
 
 	/* wm changes, need vblank before final wm's */
-	if (crtc_state->wm_changed)
+	if (crtc_state->update_wm_post)
 		return true;
 
 	/*
 	 * cxsr is re-enabled after vblank.
-	 * This is already handled by crtc_state->wm_changed,
+	 * This is already handled by crtc_state->update_wm_post,
 	 * but added for clarity.
 	 */
 	if (crtc_state->disable_cxsr)
@@ -13471,39 +13561,41 @@ static bool needs_vblank_wait(struct intel_crtc_state *crtc_state)
  * intel_atomic_commit - commit validated state object
  * @dev: DRM device
  * @state: the top-level driver state object
- * @async: asynchronous commit
+ * @nonblock: nonblocking commit
  *
  * This function commits a top-level state object that has been validated
  * with drm_atomic_helper_check().
  *
  * FIXME:  Atomic modeset support for i915 is not yet complete.  At the moment
  * we can only handle plane-related operations and do not yet support
- * asynchronous commit.
+ * nonblocking commit.
  *
  * RETURNS
  * Zero for success or -errno.
  */
 static int intel_atomic_commit(struct drm_device *dev,
 			       struct drm_atomic_state *state,
-			       bool async)
+			       bool nonblock)
 {
 	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_crtc_state *crtc_state;
+	struct drm_crtc_state *old_crtc_state;
 	struct drm_crtc *crtc;
+	struct intel_crtc_state *intel_cstate;
 	int ret = 0, i;
 	bool hw_check = intel_state->modeset;
 	unsigned long put_domains[I915_MAX_PIPES] = {};
 	unsigned crtc_vblank_mask = 0;
 
-	ret = intel_atomic_prepare_commit(dev, state, async);
+	ret = intel_atomic_prepare_commit(dev, state, nonblock);
 	if (ret) {
 		DRM_DEBUG_ATOMIC("Preparing state failed with %i\n", ret);
 		return ret;
 	}
 
 	drm_atomic_helper_swap_state(dev, state);
-	dev_priv->wm.config = to_intel_atomic_state(state)->wm_config;
+	dev_priv->wm.config = intel_state->wm_config;
+	intel_shared_dpll_commit(state);
 
 	if (intel_state->modeset) {
 		memcpy(dev_priv->min_pixclk, intel_state->min_pixclk,
@@ -13514,7 +13606,7 @@ static int intel_atomic_commit(struct drm_device *dev,
 		intel_display_power_get(dev_priv, POWER_DOMAIN_MODESET);
 	}
 
-	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+	for_each_crtc_in_state(state, crtc, old_crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
 		if (needs_modeset(crtc->state) ||
@@ -13529,10 +13621,10 @@ static int intel_atomic_commit(struct drm_device *dev,
 		if (!needs_modeset(crtc->state))
 			continue;
 
-		intel_pre_plane_update(to_intel_crtc_state(crtc_state));
+		intel_pre_plane_update(to_intel_crtc_state(old_crtc_state));
 
-		if (crtc_state->active) {
-			intel_crtc_disable_planes(crtc, crtc_state->plane_mask);
+		if (old_crtc_state->active) {
+			intel_crtc_disable_planes(crtc, old_crtc_state->plane_mask);
 			dev_priv->display.crtc_disable(crtc);
 			intel_crtc->active = false;
 			intel_fbc_disable(intel_crtc);
@@ -13555,17 +13647,17 @@ static int intel_atomic_commit(struct drm_device *dev,
 	intel_modeset_update_crtc_state(state);
 
 	if (intel_state->modeset) {
-		intel_shared_dpll_commit(state);
-
 		drm_atomic_helper_update_legacy_modeset_state(state->dev, state);
 
 		if (dev_priv->display.modeset_commit_cdclk &&
 		    intel_state->dev_cdclk != dev_priv->cdclk_freq)
 			dev_priv->display.modeset_commit_cdclk(state);
+
+		intel_modeset_verify_disabled(dev);
 	}
 
 	/* Now enable the clocks, plane, pipe, and connectors that we set up. */
-	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+	for_each_crtc_in_state(state, crtc, old_crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 		bool modeset = needs_modeset(crtc->state);
 		struct intel_crtc_state *pipe_config =
@@ -13578,14 +13670,15 @@ static int intel_atomic_commit(struct drm_device *dev,
 		}
 
 		if (!modeset)
-			intel_pre_plane_update(to_intel_crtc_state(crtc_state));
+			intel_pre_plane_update(to_intel_crtc_state(old_crtc_state));
 
-		if (crtc->state->active && intel_crtc->atomic.update_fbc)
+		if (crtc->state->active &&
+		    drm_atomic_get_existing_plane_state(state, crtc->primary))
 			intel_fbc_enable(intel_crtc);
 
 		if (crtc->state->active &&
 		    (crtc->state->planes_changed || update_pipe))
-			drm_atomic_helper_commit_planes_on_crtc(crtc_state);
+			drm_atomic_helper_commit_planes_on_crtc(old_crtc_state);
 
 		if (pipe_config->base.active && needs_vblank_wait(pipe_config))
 			crtc_vblank_mask |= 1 << i;
@@ -13596,11 +13689,27 @@ static int intel_atomic_commit(struct drm_device *dev,
 	if (!state->legacy_cursor_update)
 		intel_atomic_wait_for_vblanks(dev, dev_priv, crtc_vblank_mask);
 
-	for_each_crtc_in_state(state, crtc, crtc_state, i) {
-		intel_post_plane_update(to_intel_crtc(crtc));
+	/*
+	 * Now that the vblank has passed, we can go ahead and program the
+	 * optimal watermarks on platforms that need two-step watermark
+	 * programming.
+	 *
+	 * TODO: Move this (and other cleanup) to an async worker eventually.
+	 */
+	for_each_crtc_in_state(state, crtc, old_crtc_state, i) {
+		intel_cstate = to_intel_crtc_state(crtc->state);
+
+		if (dev_priv->display.optimize_watermarks)
+			dev_priv->display.optimize_watermarks(intel_cstate);
+	}
+
+	for_each_crtc_in_state(state, crtc, old_crtc_state, i) {
+		intel_post_plane_update(to_intel_crtc_state(old_crtc_state));
 
 		if (put_domains[i])
 			modeset_put_power_domains(dev_priv, put_domains[i]);
+
+		intel_modeset_verify_crtc(crtc, old_crtc_state, crtc->state);
 	}
 
 	if (intel_state->modeset)
@@ -13610,9 +13719,6 @@ static int intel_atomic_commit(struct drm_device *dev,
 	drm_atomic_helper_cleanup_planes(dev, state);
 	mutex_unlock(&dev->struct_mutex);
 
-	if (hw_check)
-		intel_modeset_check_state(dev, state);
-
 	drm_atomic_state_free(state);
 
 	/* As one of the primary mmio accessors, KMS has a high likelihood
@@ -13672,116 +13778,15 @@ out:
 #undef for_each_intel_crtc_masked
 
 static const struct drm_crtc_funcs intel_crtc_funcs = {
-	.gamma_set = intel_crtc_gamma_set,
+	.gamma_set = drm_atomic_helper_legacy_gamma_set,
 	.set_config = drm_atomic_helper_set_config,
+	.set_property = drm_atomic_helper_crtc_set_property,
 	.destroy = intel_crtc_destroy,
 	.page_flip = intel_crtc_page_flip,
 	.atomic_duplicate_state = intel_crtc_duplicate_state,
 	.atomic_destroy_state = intel_crtc_destroy_state,
 };
 
-static bool ibx_pch_dpll_get_hw_state(struct drm_i915_private *dev_priv,
-				      struct intel_shared_dpll *pll,
-				      struct intel_dpll_hw_state *hw_state)
-{
-	uint32_t val;
-
-	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
-		return false;
-
-	val = I915_READ(PCH_DPLL(pll->id));
-	hw_state->dpll = val;
-	hw_state->fp0 = I915_READ(PCH_FP0(pll->id));
-	hw_state->fp1 = I915_READ(PCH_FP1(pll->id));
-
-	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
-
-	return val & DPLL_VCO_ENABLE;
-}
-
-static void ibx_pch_dpll_mode_set(struct drm_i915_private *dev_priv,
-				  struct intel_shared_dpll *pll)
-{
-	I915_WRITE(PCH_FP0(pll->id), pll->config.hw_state.fp0);
-	I915_WRITE(PCH_FP1(pll->id), pll->config.hw_state.fp1);
-}
-
-static void ibx_pch_dpll_enable(struct drm_i915_private *dev_priv,
-				struct intel_shared_dpll *pll)
-{
-	/* PCH refclock must be enabled first */
-	ibx_assert_pch_refclk_enabled(dev_priv);
-
-	I915_WRITE(PCH_DPLL(pll->id), pll->config.hw_state.dpll);
-
-	/* Wait for the clocks to stabilize. */
-	POSTING_READ(PCH_DPLL(pll->id));
-	udelay(150);
-
-	/* The pixel multiplier can only be updated once the
-	 * DPLL is enabled and the clocks are stable.
-	 *
-	 * So write it again.
-	 */
-	I915_WRITE(PCH_DPLL(pll->id), pll->config.hw_state.dpll);
-	POSTING_READ(PCH_DPLL(pll->id));
-	udelay(200);
-}
-
-static void ibx_pch_dpll_disable(struct drm_i915_private *dev_priv,
-				 struct intel_shared_dpll *pll)
-{
-	struct drm_device *dev = dev_priv->dev;
-	struct intel_crtc *crtc;
-
-	/* Make sure no transcoder isn't still depending on us. */
-	for_each_intel_crtc(dev, crtc) {
-		if (intel_crtc_to_shared_dpll(crtc) == pll)
-			assert_pch_transcoder_disabled(dev_priv, crtc->pipe);
-	}
-
-	I915_WRITE(PCH_DPLL(pll->id), 0);
-	POSTING_READ(PCH_DPLL(pll->id));
-	udelay(200);
-}
-
-static char *ibx_pch_dpll_names[] = {
-	"PCH DPLL A",
-	"PCH DPLL B",
-};
-
-static void ibx_pch_dpll_init(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int i;
-
-	dev_priv->num_shared_dpll = 2;
-
-	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
-		dev_priv->shared_dplls[i].id = i;
-		dev_priv->shared_dplls[i].name = ibx_pch_dpll_names[i];
-		dev_priv->shared_dplls[i].mode_set = ibx_pch_dpll_mode_set;
-		dev_priv->shared_dplls[i].enable = ibx_pch_dpll_enable;
-		dev_priv->shared_dplls[i].disable = ibx_pch_dpll_disable;
-		dev_priv->shared_dplls[i].get_hw_state =
-			ibx_pch_dpll_get_hw_state;
-	}
-}
-
-static void intel_shared_dpll_init(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (HAS_DDI(dev))
-		intel_ddi_pll_init(dev);
-	else if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
-		ibx_pch_dpll_init(dev);
-	else
-		dev_priv->num_shared_dpll = 0;
-
-	BUG_ON(dev_priv->num_shared_dpll > I915_NUM_PLLS);
-}
-
 /**
  * intel_prepare_plane_fb - Prepare fb for usage on plane
  * @plane: drm plane to prepare for
@@ -13827,10 +13832,11 @@ intel_prepare_plane_fb(struct drm_plane *plane,
 		 */
 		if (needs_modeset(crtc_state))
 			ret = i915_gem_object_wait_rendering(old_obj, true);
-
-		/* Swallow -EIO errors to allow updates during hw lockup. */
-		if (ret && ret != -EIO)
+		if (ret) {
+			/* GPU hangs should have been swallowed by the wait */
+			WARN_ON(ret == -EIO);
 			return ret;
+		}
 	}
 
 	/* For framebuffer backed by dmabuf, wait for fence */
@@ -13855,7 +13861,7 @@ intel_prepare_plane_fb(struct drm_plane *plane,
 		if (ret)
 			DRM_DEBUG_KMS("failed to attach phys object\n");
 	} else {
-		ret = intel_pin_and_fence_fb_obj(plane, fb, new_state);
+		ret = intel_pin_and_fence_fb_obj(fb, new_state->rotation);
 	}
 
 	if (ret == 0) {
@@ -13899,7 +13905,7 @@ intel_cleanup_plane_fb(struct drm_plane *plane,
 
 	if (old_obj && (plane->type != DRM_PLANE_TYPE_CURSOR ||
 	    !INTEL_INFO(dev)->cursor_needs_physical))
-		intel_unpin_fb_obj(old_state->fb, old_state);
+		intel_unpin_fb_obj(old_state->fb, old_state->rotation);
 
 	/* prepare_fb aborted? */
 	if ((old_obj && (old_obj->frontbuffer_bits & intel_plane->frontbuffer_bit)) ||
@@ -13907,7 +13913,6 @@ intel_cleanup_plane_fb(struct drm_plane *plane,
 		i915_gem_track_fb(old_obj, obj, intel_plane->frontbuffer_bit);
 
 	i915_gem_request_assign(&old_intel_state->wait_req, NULL);
-
 }
 
 int
@@ -13982,6 +13987,11 @@ static void intel_begin_crtc_commit(struct drm_crtc *crtc,
 	if (modeset)
 		return;
 
+	if (crtc->state->color_mgmt_changed || to_intel_crtc_state(crtc->state)->update_pipe) {
+		intel_color_set_csc(crtc->state);
+		intel_color_load_luts(crtc->state);
+	}
+
 	if (to_intel_crtc_state(crtc->state)->update_pipe)
 		intel_update_pipe_config(intel_crtc, old_intel_state);
 	else if (INTEL_INFO(dev)->gen >= 9)
@@ -14025,20 +14035,19 @@ const struct drm_plane_funcs intel_plane_funcs = {
 static struct drm_plane *intel_primary_plane_create(struct drm_device *dev,
 						    int pipe)
 {
-	struct intel_plane *primary;
-	struct intel_plane_state *state;
+	struct intel_plane *primary = NULL;
+	struct intel_plane_state *state = NULL;
 	const uint32_t *intel_primary_formats;
 	unsigned int num_formats;
+	int ret;
 
 	primary = kzalloc(sizeof(*primary), GFP_KERNEL);
-	if (primary == NULL)
-		return NULL;
+	if (!primary)
+		goto fail;
 
 	state = intel_create_plane_state(&primary->base);
-	if (!state) {
-		kfree(primary);
-		return NULL;
-	}
+	if (!state)
+		goto fail;
 	primary->base.state = &state->base;
 
 	primary->can_scale = false;
@@ -14080,10 +14089,12 @@ static struct drm_plane *intel_primary_plane_create(struct drm_device *dev,
 		primary->disable_plane = i9xx_disable_primary_plane;
 	}
 
-	drm_universal_plane_init(dev, &primary->base, 0,
-				 &intel_plane_funcs,
-				 intel_primary_formats, num_formats,
-				 DRM_PLANE_TYPE_PRIMARY, NULL);
+	ret = drm_universal_plane_init(dev, &primary->base, 0,
+				       &intel_plane_funcs,
+				       intel_primary_formats, num_formats,
+				       DRM_PLANE_TYPE_PRIMARY, NULL);
+	if (ret)
+		goto fail;
 
 	if (INTEL_INFO(dev)->gen >= 4)
 		intel_create_rotation_property(dev, primary);
@@ -14091,6 +14102,12 @@ static struct drm_plane *intel_primary_plane_create(struct drm_device *dev,
 	drm_plane_helper_add(&primary->base, &intel_plane_helper_funcs);
 
 	return &primary->base;
+
+fail:
+	kfree(state);
+	kfree(primary);
+
+	return NULL;
 }
 
 void intel_create_rotation_property(struct drm_device *dev, struct intel_plane *plane)
@@ -14207,18 +14224,17 @@ intel_update_cursor_plane(struct drm_plane *plane,
 static struct drm_plane *intel_cursor_plane_create(struct drm_device *dev,
 						   int pipe)
 {
-	struct intel_plane *cursor;
-	struct intel_plane_state *state;
+	struct intel_plane *cursor = NULL;
+	struct intel_plane_state *state = NULL;
+	int ret;
 
 	cursor = kzalloc(sizeof(*cursor), GFP_KERNEL);
-	if (cursor == NULL)
-		return NULL;
+	if (!cursor)
+		goto fail;
 
 	state = intel_create_plane_state(&cursor->base);
-	if (!state) {
-		kfree(cursor);
-		return NULL;
-	}
+	if (!state)
+		goto fail;
 	cursor->base.state = &state->base;
 
 	cursor->can_scale = false;
@@ -14230,11 +14246,13 @@ static struct drm_plane *intel_cursor_plane_create(struct drm_device *dev,
 	cursor->update_plane = intel_update_cursor_plane;
 	cursor->disable_plane = intel_disable_cursor_plane;
 
-	drm_universal_plane_init(dev, &cursor->base, 0,
-				 &intel_plane_funcs,
-				 intel_cursor_formats,
-				 ARRAY_SIZE(intel_cursor_formats),
-				 DRM_PLANE_TYPE_CURSOR, NULL);
+	ret = drm_universal_plane_init(dev, &cursor->base, 0,
+				       &intel_plane_funcs,
+				       intel_cursor_formats,
+				       ARRAY_SIZE(intel_cursor_formats),
+				       DRM_PLANE_TYPE_CURSOR, NULL);
+	if (ret)
+		goto fail;
 
 	if (INTEL_INFO(dev)->gen >= 4) {
 		if (!dev->mode_config.rotation_property)
@@ -14254,6 +14272,12 @@ static struct drm_plane *intel_cursor_plane_create(struct drm_device *dev,
 	drm_plane_helper_add(&cursor->base, &intel_plane_helper_funcs);
 
 	return &cursor->base;
+
+fail:
+	kfree(state);
+	kfree(cursor);
+
+	return NULL;
 }
 
 static void skl_init_scalers(struct drm_device *dev, struct intel_crtc *intel_crtc,
@@ -14279,7 +14303,7 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 	struct intel_crtc_state *crtc_state = NULL;
 	struct drm_plane *primary = NULL;
 	struct drm_plane *cursor = NULL;
-	int i, ret;
+	int ret;
 
 	intel_crtc = kzalloc(sizeof(*intel_crtc), GFP_KERNEL);
 	if (intel_crtc == NULL)
@@ -14315,13 +14339,6 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 	if (ret)
 		goto fail;
 
-	drm_mode_crtc_set_gamma_size(&intel_crtc->base, 256);
-	for (i = 0; i < 256; i++) {
-		intel_crtc->lut_r[i] = i;
-		intel_crtc->lut_g[i] = i;
-		intel_crtc->lut_b[i] = i;
-	}
-
 	/*
 	 * On gen2/3 only plane A can do fbc, but the panel fitter and lvds port
 	 * is hooked to pipe B. Hence we want plane A feeding pipe B.
@@ -14346,6 +14363,8 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 
 	drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs);
 
+	intel_color_init(&intel_crtc->base);
+
 	WARN_ON(drm_crtc_index(&intel_crtc->base) != intel_crtc->pipe);
 	return;
 
@@ -14470,6 +14489,8 @@ static void intel_setup_outputs(struct drm_device *dev)
 		intel_ddi_init(dev, PORT_A);
 		intel_ddi_init(dev, PORT_B);
 		intel_ddi_init(dev, PORT_C);
+
+		intel_dsi_init(dev);
 	} else if (HAS_DDI(dev)) {
 		int found;
 
@@ -14839,6 +14860,8 @@ static int intel_framebuffer_init(struct drm_device *dev,
 	drm_helper_mode_fill_fb_struct(&intel_fb->base, mode_cmd);
 	intel_fb->obj = obj;
 
+	intel_fill_fb_info(dev_priv, &intel_fb->base);
+
 	ret = drm_framebuffer_init(dev, &intel_fb->base, &intel_fb_funcs);
 	if (ret) {
 		DRM_ERROR("framebuffer init failed %d\n", ret);
@@ -14859,8 +14882,7 @@ intel_user_framebuffer_create(struct drm_device *dev,
 	struct drm_i915_gem_object *obj;
 	struct drm_mode_fb_cmd2 mode_cmd = *user_mode_cmd;
 
-	obj = to_intel_bo(drm_gem_object_lookup(dev, filp,
-						mode_cmd.handles[0]));
+	obj = to_intel_bo(drm_gem_object_lookup(filp, mode_cmd.handles[0]));
 	if (&obj->base == NULL)
 		return ERR_PTR(-ENOENT);
 
@@ -14886,23 +14908,13 @@ static const struct drm_mode_config_funcs intel_mode_funcs = {
 	.atomic_state_clear = intel_atomic_state_clear,
 };
 
-/* Set up chip specific display functions */
-static void intel_init_display(struct drm_device *dev)
+/**
+ * intel_init_display_hooks - initialize the display modesetting hooks
+ * @dev_priv: device private
+ */
+void intel_init_display_hooks(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (HAS_PCH_SPLIT(dev) || IS_G4X(dev))
-		dev_priv->display.find_dpll = g4x_find_best_dpll;
-	else if (IS_CHERRYVIEW(dev))
-		dev_priv->display.find_dpll = chv_find_best_dpll;
-	else if (IS_VALLEYVIEW(dev))
-		dev_priv->display.find_dpll = vlv_find_best_dpll;
-	else if (IS_PINEVIEW(dev))
-		dev_priv->display.find_dpll = pnv_find_best_dpll;
-	else
-		dev_priv->display.find_dpll = i9xx_find_best_dpll;
-
-	if (INTEL_INFO(dev)->gen >= 9) {
+	if (INTEL_INFO(dev_priv)->gen >= 9) {
 		dev_priv->display.get_pipe_config = haswell_get_pipe_config;
 		dev_priv->display.get_initial_plane_config =
 			skylake_get_initial_plane_config;
@@ -14910,7 +14922,7 @@ static void intel_init_display(struct drm_device *dev)
 			haswell_crtc_compute_clock;
 		dev_priv->display.crtc_enable = haswell_crtc_enable;
 		dev_priv->display.crtc_disable = haswell_crtc_disable;
-	} else if (HAS_DDI(dev)) {
+	} else if (HAS_DDI(dev_priv)) {
 		dev_priv->display.get_pipe_config = haswell_get_pipe_config;
 		dev_priv->display.get_initial_plane_config =
 			ironlake_get_initial_plane_config;
@@ -14918,7 +14930,7 @@ static void intel_init_display(struct drm_device *dev)
 			haswell_crtc_compute_clock;
 		dev_priv->display.crtc_enable = haswell_crtc_enable;
 		dev_priv->display.crtc_disable = haswell_crtc_disable;
-	} else if (HAS_PCH_SPLIT(dev)) {
+	} else if (HAS_PCH_SPLIT(dev_priv)) {
 		dev_priv->display.get_pipe_config = ironlake_get_pipe_config;
 		dev_priv->display.get_initial_plane_config =
 			ironlake_get_initial_plane_config;
@@ -14926,106 +14938,134 @@ static void intel_init_display(struct drm_device *dev)
 			ironlake_crtc_compute_clock;
 		dev_priv->display.crtc_enable = ironlake_crtc_enable;
 		dev_priv->display.crtc_disable = ironlake_crtc_disable;
-	} else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
+	} else if (IS_CHERRYVIEW(dev_priv)) {
 		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
 		dev_priv->display.get_initial_plane_config =
 			i9xx_get_initial_plane_config;
-		dev_priv->display.crtc_compute_clock = i9xx_crtc_compute_clock;
+		dev_priv->display.crtc_compute_clock = chv_crtc_compute_clock;
 		dev_priv->display.crtc_enable = valleyview_crtc_enable;
 		dev_priv->display.crtc_disable = i9xx_crtc_disable;
-	} else {
+	} else if (IS_VALLEYVIEW(dev_priv)) {
+		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
+		dev_priv->display.get_initial_plane_config =
+			i9xx_get_initial_plane_config;
+		dev_priv->display.crtc_compute_clock = vlv_crtc_compute_clock;
+		dev_priv->display.crtc_enable = valleyview_crtc_enable;
+		dev_priv->display.crtc_disable = i9xx_crtc_disable;
+	} else if (IS_G4X(dev_priv)) {
+		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
+		dev_priv->display.get_initial_plane_config =
+			i9xx_get_initial_plane_config;
+		dev_priv->display.crtc_compute_clock = g4x_crtc_compute_clock;
+		dev_priv->display.crtc_enable = i9xx_crtc_enable;
+		dev_priv->display.crtc_disable = i9xx_crtc_disable;
+	} else if (IS_PINEVIEW(dev_priv)) {
+		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
+		dev_priv->display.get_initial_plane_config =
+			i9xx_get_initial_plane_config;
+		dev_priv->display.crtc_compute_clock = pnv_crtc_compute_clock;
+		dev_priv->display.crtc_enable = i9xx_crtc_enable;
+		dev_priv->display.crtc_disable = i9xx_crtc_disable;
+	} else if (!IS_GEN2(dev_priv)) {
 		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
 		dev_priv->display.get_initial_plane_config =
 			i9xx_get_initial_plane_config;
 		dev_priv->display.crtc_compute_clock = i9xx_crtc_compute_clock;
 		dev_priv->display.crtc_enable = i9xx_crtc_enable;
 		dev_priv->display.crtc_disable = i9xx_crtc_disable;
+	} else {
+		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
+		dev_priv->display.get_initial_plane_config =
+			i9xx_get_initial_plane_config;
+		dev_priv->display.crtc_compute_clock = i8xx_crtc_compute_clock;
+		dev_priv->display.crtc_enable = i9xx_crtc_enable;
+		dev_priv->display.crtc_disable = i9xx_crtc_disable;
 	}
 
 	/* Returns the core display clock speed */
-	if (IS_SKYLAKE(dev) || IS_KABYLAKE(dev))
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			skylake_get_display_clock_speed;
-	else if (IS_BROXTON(dev))
+	else if (IS_BROXTON(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			broxton_get_display_clock_speed;
-	else if (IS_BROADWELL(dev))
+	else if (IS_BROADWELL(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			broadwell_get_display_clock_speed;
-	else if (IS_HASWELL(dev))
+	else if (IS_HASWELL(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			haswell_get_display_clock_speed;
-	else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
+	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			valleyview_get_display_clock_speed;
-	else if (IS_GEN5(dev))
+	else if (IS_GEN5(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			ilk_get_display_clock_speed;
-	else if (IS_I945G(dev) || IS_BROADWATER(dev) ||
-		 IS_GEN6(dev) || IS_IVYBRIDGE(dev))
+	else if (IS_I945G(dev_priv) || IS_BROADWATER(dev_priv) ||
+		 IS_GEN6(dev_priv) || IS_IVYBRIDGE(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			i945_get_display_clock_speed;
-	else if (IS_GM45(dev))
+	else if (IS_GM45(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			gm45_get_display_clock_speed;
-	else if (IS_CRESTLINE(dev))
+	else if (IS_CRESTLINE(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			i965gm_get_display_clock_speed;
-	else if (IS_PINEVIEW(dev))
+	else if (IS_PINEVIEW(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			pnv_get_display_clock_speed;
-	else if (IS_G33(dev) || IS_G4X(dev))
+	else if (IS_G33(dev_priv) || IS_G4X(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			g33_get_display_clock_speed;
-	else if (IS_I915G(dev))
+	else if (IS_I915G(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			i915_get_display_clock_speed;
-	else if (IS_I945GM(dev) || IS_845G(dev))
+	else if (IS_I945GM(dev_priv) || IS_845G(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			i9xx_misc_get_display_clock_speed;
-	else if (IS_I915GM(dev))
+	else if (IS_I915GM(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			i915gm_get_display_clock_speed;
-	else if (IS_I865G(dev))
+	else if (IS_I865G(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			i865_get_display_clock_speed;
-	else if (IS_I85X(dev))
+	else if (IS_I85X(dev_priv))
 		dev_priv->display.get_display_clock_speed =
 			i85x_get_display_clock_speed;
 	else { /* 830 */
-		WARN(!IS_I830(dev), "Unknown platform. Assuming 133 MHz CDCLK\n");
+		WARN(!IS_I830(dev_priv), "Unknown platform. Assuming 133 MHz CDCLK\n");
 		dev_priv->display.get_display_clock_speed =
 			i830_get_display_clock_speed;
 	}
 
-	if (IS_GEN5(dev)) {
+	if (IS_GEN5(dev_priv)) {
 		dev_priv->display.fdi_link_train = ironlake_fdi_link_train;
-	} else if (IS_GEN6(dev)) {
+	} else if (IS_GEN6(dev_priv)) {
 		dev_priv->display.fdi_link_train = gen6_fdi_link_train;
-	} else if (IS_IVYBRIDGE(dev)) {
+	} else if (IS_IVYBRIDGE(dev_priv)) {
 		/* FIXME: detect B0+ stepping and use auto training */
 		dev_priv->display.fdi_link_train = ivb_manual_fdi_link_train;
-	} else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
+	} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
 		dev_priv->display.fdi_link_train = hsw_fdi_link_train;
-		if (IS_BROADWELL(dev)) {
+		if (IS_BROADWELL(dev_priv)) {
 			dev_priv->display.modeset_commit_cdclk =
 				broadwell_modeset_commit_cdclk;
 			dev_priv->display.modeset_calc_cdclk =
 				broadwell_modeset_calc_cdclk;
 		}
-	} else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
+	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
 		dev_priv->display.modeset_commit_cdclk =
 			valleyview_modeset_commit_cdclk;
 		dev_priv->display.modeset_calc_cdclk =
 			valleyview_modeset_calc_cdclk;
-	} else if (IS_BROXTON(dev)) {
+	} else if (IS_BROXTON(dev_priv)) {
 		dev_priv->display.modeset_commit_cdclk =
 			broxton_modeset_commit_cdclk;
 		dev_priv->display.modeset_calc_cdclk =
 			broxton_modeset_calc_cdclk;
 	}
 
-	switch (INTEL_INFO(dev)->gen) {
+	switch (INTEL_INFO(dev_priv)->gen) {
 	case 2:
 		dev_priv->display.queue_flip = intel_gen2_queue_flip;
 		break;
@@ -15052,8 +15092,6 @@ static void intel_init_display(struct drm_device *dev)
 		/* Default just returns -ENODEV to indicate unsupported */
 		dev_priv->display.queue_flip = intel_default_queue_flip;
 	}
-
-	mutex_init(&dev_priv->pps_mutex);
 }
 
 /*
@@ -15276,7 +15314,7 @@ static void sanitize_watermarks(struct drm_device *dev)
 	int i;
 
 	/* Only supported on platforms that use atomic watermark design */
-	if (!dev_priv->display.program_watermarks)
+	if (!dev_priv->display.optimize_watermarks)
 		return;
 
 	/*
@@ -15297,6 +15335,13 @@ retry:
 	if (WARN_ON(IS_ERR(state)))
 		goto fail;
 
+	/*
+	 * Hardware readout is the only time we don't want to calculate
+	 * intermediate watermarks (since we don't trust the current
+	 * watermarks).
+	 */
+	to_intel_atomic_state(state)->skip_intermediate_wm = true;
+
 	ret = intel_atomic_check(dev, state);
 	if (ret) {
 		/*
@@ -15319,7 +15364,8 @@ retry:
 	for_each_crtc_in_state(state, crtc, cstate, i) {
 		struct intel_crtc_state *cs = to_intel_crtc_state(cstate);
 
-		dev_priv->display.program_watermarks(cs);
+		cs->wm.need_postvbl_update = true;
+		dev_priv->display.optimize_watermarks(cs);
 	}
 
 	drm_atomic_state_free(state);
@@ -15330,7 +15376,8 @@ fail:
 
 void intel_modeset_init(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	int sprite, ret;
 	enum pipe pipe;
 	struct intel_crtc *crtc;
@@ -15372,9 +15419,6 @@ void intel_modeset_init(struct drm_device *dev)
 		}
 	}
 
-	intel_init_display(dev);
-	intel_init_audio(dev);
-
 	if (IS_GEN2(dev)) {
 		dev->mode_config.max_width = 2048;
 		dev->mode_config.max_height = 2048;
@@ -15397,7 +15441,7 @@ void intel_modeset_init(struct drm_device *dev)
 		dev->mode_config.cursor_height = MAX_CURSOR_HEIGHT;
 	}
 
-	dev->mode_config.fb_base = dev_priv->gtt.mappable_base;
+	dev->mode_config.fb_base = ggtt->mappable_base;
 
 	DRM_DEBUG_KMS("%d display pipe%s available.\n",
 		      INTEL_INFO(dev)->num_pipes,
@@ -15414,6 +15458,7 @@ void intel_modeset_init(struct drm_device *dev)
 	}
 
 	intel_update_czclk(dev_priv);
+	intel_update_rawclk(dev_priv);
 	intel_update_cdclk(dev);
 
 	intel_shared_dpll_init(dev);
@@ -15526,10 +15571,15 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	i915_reg_t reg = PIPECONF(crtc->config->cpu_transcoder);
+	enum transcoder cpu_transcoder = crtc->config->cpu_transcoder;
 
 	/* Clear any frame start delays used for debugging left by the BIOS */
-	I915_WRITE(reg, I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK);
+	if (!transcoder_is_dsi(cpu_transcoder)) {
+		i915_reg_t reg = PIPECONF(cpu_transcoder);
+
+		I915_WRITE(reg,
+			   I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK);
+	}
 
 	/* restore vblank interrupts to correct state */
 	drm_crtc_vblank_reset(&crtc->base);
@@ -15577,38 +15627,9 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 
 	/* Adjust the state of the output pipe according to whether we
 	 * have active connectors/encoders. */
-	if (!intel_crtc_has_encoders(crtc))
+	if (crtc->active && !intel_crtc_has_encoders(crtc))
 		intel_crtc_disable_noatomic(&crtc->base);
 
-	if (crtc->active != crtc->base.state->active) {
-		struct intel_encoder *encoder;
-
-		/* This can happen either due to bugs in the get_hw_state
-		 * functions or because of calls to intel_crtc_disable_noatomic,
-		 * or because the pipe is force-enabled due to the
-		 * pipe A quirk. */
-		DRM_DEBUG_KMS("[CRTC:%d] hw state adjusted, was %s, now %s\n",
-			      crtc->base.base.id,
-			      crtc->base.state->enable ? "enabled" : "disabled",
-			      crtc->active ? "enabled" : "disabled");
-
-		WARN_ON(drm_atomic_set_mode_for_crtc(crtc->base.state, NULL) < 0);
-		crtc->base.state->active = crtc->active;
-		crtc->base.enabled = crtc->active;
-		crtc->base.state->connector_mask = 0;
-		crtc->base.state->encoder_mask = 0;
-
-		/* Because we only establish the connector -> encoder ->
-		 * crtc links if something is active, this means the
-		 * crtc is now deactivated. Break the links. connector
-		 * -> encoder links are only establish when things are
-		 *  actually up, hence no need to break them. */
-		WARN_ON(crtc->active);
-
-		for_each_encoder_on_crtc(dev, &crtc->base, encoder)
-			encoder->base.crtc = NULL;
-	}
-
 	if (crtc->active || HAS_GMCH_DISPLAY(dev)) {
 		/*
 		 * We start out with underrun reporting disabled to avoid races.
@@ -15738,7 +15759,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		struct intel_crtc_state *crtc_state = crtc->config;
 		int pixclk = 0;
 
-		__drm_atomic_helper_crtc_destroy_state(&crtc->base, &crtc_state->base);
+		__drm_atomic_helper_crtc_destroy_state(&crtc_state->base);
 		memset(crtc_state, 0, sizeof(*crtc_state));
 		crtc_state->base.crtc = &crtc->base;
 
@@ -15777,22 +15798,17 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
 		struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i];
 
-		pll->on = pll->get_hw_state(dev_priv, pll,
-					    &pll->config.hw_state);
-		pll->active = 0;
+		pll->on = pll->funcs.get_hw_state(dev_priv, pll,
+						  &pll->config.hw_state);
 		pll->config.crtc_mask = 0;
 		for_each_intel_crtc(dev, crtc) {
-			if (crtc->active && intel_crtc_to_shared_dpll(crtc) == pll) {
-				pll->active++;
+			if (crtc->active && crtc->config->shared_dpll == pll)
 				pll->config.crtc_mask |= 1 << crtc->pipe;
-			}
 		}
+		pll->active_mask = pll->config.crtc_mask;
 
 		DRM_DEBUG_KMS("%s hw state readout: crtc_mask 0x%08x, on %i\n",
 			      pll->name, pll->config.crtc_mask, pll->on);
-
-		if (pll->config.crtc_mask)
-			intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
 	}
 
 	for_each_intel_encoder(dev, encoder) {
@@ -15874,6 +15890,8 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 			drm_calc_timestamping_constants(&crtc->base, &crtc->base.hwmode);
 			update_scanline_offset(crtc);
 		}
+
+		intel_pipe_config_sanity_check(dev_priv, crtc->config);
 	}
 }
 
@@ -15908,12 +15926,12 @@ intel_modeset_setup_hw_state(struct drm_device *dev)
 	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
 		struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i];
 
-		if (!pll->on || pll->active)
+		if (!pll->on || pll->active_mask)
 			continue;
 
 		DRM_DEBUG_KMS("%s enabled but not in use, disabling\n", pll->name);
 
-		pll->disable(dev_priv, pll);
+		pll->funcs.disable(dev_priv, pll);
 		pll->on = false;
 	}
 
@@ -16022,9 +16040,8 @@ void intel_modeset_gem_init(struct drm_device *dev)
 			continue;
 
 		mutex_lock(&dev->struct_mutex);
-		ret = intel_pin_and_fence_fb_obj(c->primary,
-						 c->primary->fb,
-						 c->primary->state);
+		ret = intel_pin_and_fence_fb_obj(c->primary->fb,
+						 c->primary->state->rotation);
 		mutex_unlock(&dev->struct_mutex);
 		if (ret) {
 			DRM_ERROR("failed to pin boot fb on pipe %d\n",
@@ -16233,8 +16250,9 @@ intel_display_capture_error_state(struct drm_device *dev)
 			error->pipe[i].stat = I915_READ(PIPESTAT(i));
 	}
 
+	/* Note: this does not include DSI transcoders. */
 	error->num_transcoders = INTEL_INFO(dev)->num_pipes;
-	if (HAS_DDI(dev_priv->dev))
+	if (HAS_DDI(dev_priv))
 		error->num_transcoders++; /* Account for eDP. */
 
 	for (i = 0; i < error->num_transcoders; i++) {
@@ -16305,7 +16323,7 @@ intel_display_print_error_state(struct drm_i915_error_state_buf *m,
 	}
 
 	for (i = 0; i < error->num_transcoders; i++) {
-		err_printf(m, "CPU transcoder: %c\n",
+		err_printf(m, "CPU transcoder: %s\n",
 			   transcoder_name(error->transcoder[i].cpu_transcoder));
 		err_printf(m, "  Power: %s\n",
 			   onoff(error->transcoder[i].power_domain_on));
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 412a34c39522..f192f58708c2 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -129,6 +129,7 @@ static void edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync);
 static void vlv_init_panel_power_sequencer(struct intel_dp *intel_dp);
 static void vlv_steal_power_sequencer(struct drm_device *dev,
 				      enum pipe pipe);
+static void intel_dp_unset_edid(struct intel_dp *intel_dp);
 
 static unsigned int intel_dp_unused_lane_mask(int lane_count)
 {
@@ -671,60 +672,55 @@ intel_dp_aux_wait_done(struct intel_dp *intel_dp, bool has_aux_irq)
 	return status;
 }
 
-static uint32_t i9xx_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
+static uint32_t g4x_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = intel_dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
+
+	if (index)
+		return 0;
 
 	/*
 	 * The clock divider is based off the hrawclk, and would like to run at
-	 * 2MHz.  So, take the hrawclk value and divide by 2 and use that
+	 * 2MHz.  So, take the hrawclk value and divide by 2000 and use that
 	 */
-	return index ? 0 : DIV_ROUND_CLOSEST(intel_hrawclk(dev), 2);
+	return DIV_ROUND_CLOSEST(dev_priv->rawclk_freq, 2000);
 }
 
 static uint32_t ilk_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = intel_dig_port->base.base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
 
 	if (index)
 		return 0;
 
-	if (intel_dig_port->port == PORT_A) {
+	/*
+	 * The clock divider is based off the cdclk or PCH rawclk, and would
+	 * like to run at 2MHz.  So, take the cdclk or PCH rawclk value and
+	 * divide by 2000 and use that
+	 */
+	if (intel_dig_port->port == PORT_A)
 		return DIV_ROUND_CLOSEST(dev_priv->cdclk_freq, 2000);
-
-	} else {
-		return DIV_ROUND_CLOSEST(intel_pch_rawclk(dev), 2);
-	}
+	else
+		return DIV_ROUND_CLOSEST(dev_priv->rawclk_freq, 2000);
 }
 
 static uint32_t hsw_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = intel_dig_port->base.base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
 
-	if (intel_dig_port->port == PORT_A) {
-		if (index)
-			return 0;
-		return DIV_ROUND_CLOSEST(dev_priv->cdclk_freq, 2000);
-	} else if (HAS_PCH_LPT_H(dev_priv)) {
+	if (intel_dig_port->port != PORT_A && HAS_PCH_LPT_H(dev_priv)) {
 		/* Workaround for non-ULT HSW */
 		switch (index) {
 		case 0: return 63;
 		case 1: return 72;
 		default: return 0;
 		}
-	} else  {
-		return index ? 0 : DIV_ROUND_CLOSEST(intel_pch_rawclk(dev), 2);
 	}
-}
 
-static uint32_t vlv_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
-{
-	return index ? 0 : 100;
+	return ilk_get_aux_clock_divider(intel_dp, index);
 }
 
 static uint32_t skl_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
@@ -737,10 +733,10 @@ static uint32_t skl_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 	return index ? 0 : 1;
 }
 
-static uint32_t i9xx_get_aux_send_ctl(struct intel_dp *intel_dp,
-				      bool has_aux_irq,
-				      int send_bytes,
-				      uint32_t aux_clock_divider)
+static uint32_t g4x_get_aux_send_ctl(struct intel_dp *intel_dp,
+				     bool has_aux_irq,
+				     int send_bytes,
+				     uint32_t aux_clock_divider)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
@@ -1229,71 +1225,6 @@ intel_dp_connector_unregister(struct intel_connector *intel_connector)
 	intel_connector_unregister(intel_connector);
 }
 
-static void
-skl_edp_set_pll_config(struct intel_crtc_state *pipe_config)
-{
-	u32 ctrl1;
-
-	memset(&pipe_config->dpll_hw_state, 0,
-	       sizeof(pipe_config->dpll_hw_state));
-
-	pipe_config->ddi_pll_sel = SKL_DPLL0;
-	pipe_config->dpll_hw_state.cfgcr1 = 0;
-	pipe_config->dpll_hw_state.cfgcr2 = 0;
-
-	ctrl1 = DPLL_CTRL1_OVERRIDE(SKL_DPLL0);
-	switch (pipe_config->port_clock / 2) {
-	case 81000:
-		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810,
-					      SKL_DPLL0);
-		break;
-	case 135000:
-		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1350,
-					      SKL_DPLL0);
-		break;
-	case 270000:
-		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2700,
-					      SKL_DPLL0);
-		break;
-	case 162000:
-		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1620,
-					      SKL_DPLL0);
-		break;
-	/* TBD: For DP link rates 2.16 GHz and 4.32 GHz, VCO is 8640 which
-	results in CDCLK change. Need to handle the change of CDCLK by
-	disabling pipes and re-enabling them */
-	case 108000:
-		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1080,
-					      SKL_DPLL0);
-		break;
-	case 216000:
-		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2160,
-					      SKL_DPLL0);
-		break;
-
-	}
-	pipe_config->dpll_hw_state.ctrl1 = ctrl1;
-}
-
-void
-hsw_dp_set_ddi_pll_sel(struct intel_crtc_state *pipe_config)
-{
-	memset(&pipe_config->dpll_hw_state, 0,
-	       sizeof(pipe_config->dpll_hw_state));
-
-	switch (pipe_config->port_clock / 2) {
-	case 81000:
-		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_810;
-		break;
-	case 135000:
-		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_1350;
-		break;
-	case 270000:
-		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_2700;
-		break;
-	}
-}
-
 static int
 intel_dp_sink_rates(struct intel_dp *intel_dp, const int **sink_rates)
 {
@@ -1570,10 +1501,10 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 
 		/* Get bpp from vbt only for panels that dont have bpp in edid */
 		if (intel_connector->base.display_info.bpc == 0 &&
-			(dev_priv->vbt.edp_bpp && dev_priv->vbt.edp_bpp < bpp)) {
+			(dev_priv->vbt.edp.bpp && dev_priv->vbt.edp.bpp < bpp)) {
 			DRM_DEBUG_KMS("clamping bpp for eDP panel to BIOS-provided %i\n",
-				      dev_priv->vbt.edp_bpp);
-			bpp = dev_priv->vbt.edp_bpp;
+				      dev_priv->vbt.edp.bpp);
+			bpp = dev_priv->vbt.edp.bpp;
 		}
 
 		/*
@@ -1651,13 +1582,7 @@ found:
 				&pipe_config->dp_m2_n2);
 	}
 
-	if ((IS_SKYLAKE(dev)  || IS_KABYLAKE(dev)) && is_edp(intel_dp))
-		skl_edp_set_pll_config(pipe_config);
-	else if (IS_BROXTON(dev))
-		/* handled in ddi */;
-	else if (IS_HASWELL(dev) || IS_BROADWELL(dev))
-		hsw_dp_set_ddi_pll_sel(pipe_config);
-	else
+	if (!HAS_DDI(dev))
 		intel_dp_set_clock(encoder, pipe_config);
 
 	return true;
@@ -1779,11 +1704,11 @@ static void wait_panel_status(struct intel_dp *intel_dp,
 			I915_READ(pp_stat_reg),
 			I915_READ(pp_ctrl_reg));
 
-	if (_wait_for((I915_READ(pp_stat_reg) & mask) == value, 5000, 10)) {
+	if (_wait_for((I915_READ(pp_stat_reg) & mask) == value,
+		      5 * USEC_PER_SEC, 10 * USEC_PER_MSEC))
 		DRM_ERROR("Panel status timeout: status %08x control %08x\n",
 				I915_READ(pp_stat_reg),
 				I915_READ(pp_ctrl_reg));
-	}
 
 	DRM_DEBUG_KMS("Wait complete\n");
 }
@@ -2290,6 +2215,15 @@ static void ironlake_edp_pll_on(struct intel_dp *intel_dp)
 	POSTING_READ(DP_A);
 	udelay(500);
 
+	/*
+	 * [DevILK] Work around required when enabling DP PLL
+	 * while a pipe is enabled going to FDI:
+	 * 1. Wait for the start of vertical blank on the enabled pipe going to FDI
+	 * 2. Program DP PLL enable
+	 */
+	if (IS_GEN5(dev_priv))
+		intel_wait_for_vblank_if_active(dev_priv->dev, !crtc->pipe);
+
 	intel_dp->DP |= DP_PLL_ENABLE;
 
 	I915_WRITE(DP_A, intel_dp->DP);
@@ -2409,7 +2343,6 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum port port = dp_to_dig_port(intel_dp)->port;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	int dotclock;
 
 	tmp = I915_READ(intel_dp->output_reg);
 
@@ -2459,16 +2392,12 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 			pipe_config->port_clock = 270000;
 	}
 
-	dotclock = intel_dotclock_calculate(pipe_config->port_clock,
-					    &pipe_config->dp_m_n);
-
-	if (HAS_PCH_SPLIT(dev_priv->dev) && port != PORT_A)
-		ironlake_check_encoder_dotclock(pipe_config, dotclock);
-
-	pipe_config->base.adjusted_mode.crtc_clock = dotclock;
+	pipe_config->base.adjusted_mode.crtc_clock =
+		intel_dotclock_calculate(pipe_config->port_clock,
+					 &pipe_config->dp_m_n);
 
-	if (is_edp(intel_dp) && dev_priv->vbt.edp_bpp &&
-	    pipe_config->pipe_bpp > dev_priv->vbt.edp_bpp) {
+	if (is_edp(intel_dp) && dev_priv->vbt.edp.bpp &&
+	    pipe_config->pipe_bpp > dev_priv->vbt.edp.bpp) {
 		/*
 		 * This is a big fat ugly hack.
 		 *
@@ -2483,8 +2412,8 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 		 * load.
 		 */
 		DRM_DEBUG_KMS("pipe has %d bpp for eDP panel, overriding BIOS-provided max %d bpp\n",
-			      pipe_config->pipe_bpp, dev_priv->vbt.edp_bpp);
-		dev_priv->vbt.edp_bpp = pipe_config->pipe_bpp;
+			      pipe_config->pipe_bpp, dev_priv->vbt.edp.bpp);
+		dev_priv->vbt.edp.bpp = pipe_config->pipe_bpp;
 	}
 }
 
@@ -2710,7 +2639,6 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	uint32_t dp_reg = I915_READ(intel_dp->output_reg);
-	enum port port = dp_to_dig_port(intel_dp)->port;
 	enum pipe pipe = crtc->pipe;
 
 	if (WARN_ON(dp_reg & DP_PORT_EN))
@@ -2721,35 +2649,12 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
 		vlv_init_panel_power_sequencer(intel_dp);
 
-	/*
-	 * We get an occasional spurious underrun between the port
-	 * enable and vdd enable, when enabling port A eDP.
-	 *
-	 * FIXME: Not sure if this applies to (PCH) port D eDP as well
-	 */
-	if (port == PORT_A)
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
-
 	intel_dp_enable_port(intel_dp);
 
-	if (port == PORT_A && IS_GEN5(dev_priv)) {
-		/*
-		 * Underrun reporting for the other pipe was disabled in
-		 * g4x_pre_enable_dp(). The eDP PLL and port have now been
-		 * enabled, so it's now safe to re-enable underrun reporting.
-		 */
-		intel_wait_for_vblank_if_active(dev_priv->dev, !pipe);
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, !pipe, true);
-		intel_set_pch_fifo_underrun_reporting(dev_priv, !pipe, true);
-	}
-
 	edp_panel_vdd_on(intel_dp);
 	edp_panel_on(intel_dp);
 	edp_panel_vdd_off(intel_dp, true);
 
-	if (port == PORT_A)
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
-
 	pps_unlock(intel_dp);
 
 	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
@@ -2791,26 +2696,11 @@ static void vlv_enable_dp(struct intel_encoder *encoder)
 
 static void g4x_pre_enable_dp(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	enum port port = dp_to_dig_port(intel_dp)->port;
-	enum pipe pipe = to_intel_crtc(encoder->base.crtc)->pipe;
 
 	intel_dp_prepare(encoder);
 
-	if (port == PORT_A && IS_GEN5(dev_priv)) {
-		/*
-		 * We get FIFO underruns on the other pipe when
-		 * enabling the CPU eDP PLL, and when enabling CPU
-		 * eDP port. We could potentially avoid the PLL
-		 * underrun with a vblank wait just prior to enabling
-		 * the PLL, but that doesn't appear to help the port
-		 * enable case. Just sweep it all under the rug.
-		 */
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, !pipe, false);
-		intel_set_pch_fifo_underrun_reporting(dev_priv, !pipe, false);
-	}
-
 	/* Only ilk+ has port A */
 	if (port == PORT_A)
 		ironlake_edp_pll_on(intel_dp);
@@ -3184,47 +3074,14 @@ static void chv_dp_post_pll_disable(struct intel_encoder *encoder)
 }
 
 /*
- * Native read with retry for link status and receiver capability reads for
- * cases where the sink may still be asleep.
- *
- * Sinks are *supposed* to come up within 1ms from an off state, but we're also
- * supposed to retry 3 times per the spec.
- */
-static ssize_t
-intel_dp_dpcd_read_wake(struct drm_dp_aux *aux, unsigned int offset,
-			void *buffer, size_t size)
-{
-	ssize_t ret;
-	int i;
-
-	/*
-	 * Sometime we just get the same incorrect byte repeated
-	 * over the entire buffer. Doing just one throw away read
-	 * initially seems to "solve" it.
-	 */
-	drm_dp_dpcd_read(aux, DP_DPCD_REV, buffer, 1);
-
-	for (i = 0; i < 3; i++) {
-		ret = drm_dp_dpcd_read(aux, offset, buffer, size);
-		if (ret == size)
-			return ret;
-		msleep(1);
-	}
-
-	return ret;
-}
-
-/*
  * Fetch AUX CH registers 0x202 - 0x207 which contain
  * link status information
  */
 bool
 intel_dp_get_link_status(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
-	return intel_dp_dpcd_read_wake(&intel_dp->aux,
-				       DP_LANE0_1_STATUS,
-				       link_status,
-				       DP_LINK_STATUS_SIZE) == DP_LINK_STATUS_SIZE;
+	return drm_dp_dpcd_read(&intel_dp->aux, DP_LANE0_1_STATUS, link_status,
+				DP_LINK_STATUS_SIZE) == DP_LINK_STATUS_SIZE;
 }
 
 /* These are source-specific values. */
@@ -3238,7 +3095,7 @@ intel_dp_voltage_max(struct intel_dp *intel_dp)
 	if (IS_BROXTON(dev))
 		return DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
 	else if (INTEL_INFO(dev)->gen >= 9) {
-		if (dev_priv->edp_low_vswing && port == PORT_A)
+		if (dev_priv->vbt.edp.low_vswing && port == PORT_A)
 			return DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
 		return DP_TRAIN_VOLTAGE_SWING_LEVEL_2;
 	} else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
@@ -3859,8 +3716,8 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint8_t rev;
 
-	if (intel_dp_dpcd_read_wake(&intel_dp->aux, 0x000, intel_dp->dpcd,
-				    sizeof(intel_dp->dpcd)) < 0)
+	if (drm_dp_dpcd_read(&intel_dp->aux, 0x000, intel_dp->dpcd,
+			     sizeof(intel_dp->dpcd)) < 0)
 		return false; /* aux transfer failed */
 
 	DRM_DEBUG_KMS("DPCD: %*ph\n", (int) sizeof(intel_dp->dpcd), intel_dp->dpcd);
@@ -3868,12 +3725,33 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 	if (intel_dp->dpcd[DP_DPCD_REV] == 0)
 		return false; /* DPCD not present */
 
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_SINK_COUNT,
+			     &intel_dp->sink_count, 1) < 0)
+		return false;
+
+	/*
+	 * Sink count can change between short pulse hpd hence
+	 * a member variable in intel_dp will track any changes
+	 * between short pulse interrupts.
+	 */
+	intel_dp->sink_count = DP_GET_SINK_COUNT(intel_dp->sink_count);
+
+	/*
+	 * SINK_COUNT == 0 and DOWNSTREAM_PORT_PRESENT == 1 implies that
+	 * a dongle is present but no display. Unless we require to know
+	 * if a dongle is present or not, we don't need to update
+	 * downstream port information. So, an early return here saves
+	 * time from performing other operations which are not required.
+	 */
+	if (!is_edp(intel_dp) && !intel_dp->sink_count)
+		return false;
+
 	/* Check if the panel supports PSR */
 	memset(intel_dp->psr_dpcd, 0, sizeof(intel_dp->psr_dpcd));
 	if (is_edp(intel_dp)) {
-		intel_dp_dpcd_read_wake(&intel_dp->aux, DP_PSR_SUPPORT,
-					intel_dp->psr_dpcd,
-					sizeof(intel_dp->psr_dpcd));
+		drm_dp_dpcd_read(&intel_dp->aux, DP_PSR_SUPPORT,
+				 intel_dp->psr_dpcd,
+				 sizeof(intel_dp->psr_dpcd));
 		if (intel_dp->psr_dpcd[0] & DP_PSR_IS_SUPPORTED) {
 			dev_priv->psr.sink_support = true;
 			DRM_DEBUG_KMS("Detected EDP PSR Panel.\n");
@@ -3884,9 +3762,9 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 			uint8_t frame_sync_cap;
 
 			dev_priv->psr.sink_support = true;
-			intel_dp_dpcd_read_wake(&intel_dp->aux,
-					DP_SINK_DEVICE_AUX_FRAME_SYNC_CAP,
-					&frame_sync_cap, 1);
+			drm_dp_dpcd_read(&intel_dp->aux,
+					 DP_SINK_DEVICE_AUX_FRAME_SYNC_CAP,
+					 &frame_sync_cap, 1);
 			dev_priv->psr.aux_frame_sync = frame_sync_cap ? true : false;
 			/* PSR2 needs frame sync as well */
 			dev_priv->psr.psr2_support = dev_priv->psr.aux_frame_sync;
@@ -3902,15 +3780,13 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 	/* Intermediate frequency support */
 	if (is_edp(intel_dp) &&
 	    (intel_dp->dpcd[DP_EDP_CONFIGURATION_CAP] &	DP_DPCD_DISPLAY_CONTROL_CAPABLE) &&
-	    (intel_dp_dpcd_read_wake(&intel_dp->aux, DP_EDP_DPCD_REV, &rev, 1) == 1) &&
+	    (drm_dp_dpcd_read(&intel_dp->aux, DP_EDP_DPCD_REV, &rev, 1) == 1) &&
 	    (rev >= 0x03)) { /* eDp v1.4 or higher */
 		__le16 sink_rates[DP_MAX_SUPPORTED_RATES];
 		int i;
 
-		intel_dp_dpcd_read_wake(&intel_dp->aux,
-				DP_SUPPORTED_LINK_RATES,
-				sink_rates,
-				sizeof(sink_rates));
+		drm_dp_dpcd_read(&intel_dp->aux, DP_SUPPORTED_LINK_RATES,
+				sink_rates, sizeof(sink_rates));
 
 		for (i = 0; i < ARRAY_SIZE(sink_rates); i++) {
 			int val = le16_to_cpu(sink_rates[i]);
@@ -3933,9 +3809,9 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 	if (intel_dp->dpcd[DP_DPCD_REV] == 0x10)
 		return true; /* no per-port downstream info */
 
-	if (intel_dp_dpcd_read_wake(&intel_dp->aux, DP_DOWNSTREAM_PORT_0,
-				    intel_dp->downstream_ports,
-				    DP_MAX_DOWNSTREAM_PORTS) < 0)
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_DOWNSTREAM_PORT_0,
+			     intel_dp->downstream_ports,
+			     DP_MAX_DOWNSTREAM_PORTS) < 0)
 		return false; /* downstream port status fetch failed */
 
 	return true;
@@ -3949,11 +3825,11 @@ intel_dp_probe_oui(struct intel_dp *intel_dp)
 	if (!(intel_dp->dpcd[DP_DOWN_STREAM_PORT_COUNT] & DP_OUI_SUPPORT))
 		return;
 
-	if (intel_dp_dpcd_read_wake(&intel_dp->aux, DP_SINK_OUI, buf, 3) == 3)
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_SINK_OUI, buf, 3) == 3)
 		DRM_DEBUG_KMS("Sink OUI: %02hx%02hx%02hx\n",
 			      buf[0], buf[1], buf[2]);
 
-	if (intel_dp_dpcd_read_wake(&intel_dp->aux, DP_BRANCH_OUI, buf, 3) == 3)
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_BRANCH_OUI, buf, 3) == 3)
 		DRM_DEBUG_KMS("Branch OUI: %02hx%02hx%02hx\n",
 			      buf[0], buf[1], buf[2]);
 }
@@ -3963,13 +3839,16 @@ intel_dp_probe_mst(struct intel_dp *intel_dp)
 {
 	u8 buf[1];
 
+	if (!i915.enable_dp_mst)
+		return false;
+
 	if (!intel_dp->can_mst)
 		return false;
 
 	if (intel_dp->dpcd[DP_DPCD_REV] < 0x12)
 		return false;
 
-	if (intel_dp_dpcd_read_wake(&intel_dp->aux, DP_MSTM_CAP, buf, 1)) {
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_MSTM_CAP, buf, 1)) {
 		if (buf[0] & DP_MST_CAP) {
 			DRM_DEBUG_KMS("Sink is MST capable\n");
 			intel_dp->is_mst = true;
@@ -4106,7 +3985,7 @@ stop:
 static bool
 intel_dp_get_sink_irq(struct intel_dp *intel_dp, u8 *sink_irq_vector)
 {
-	return intel_dp_dpcd_read_wake(&intel_dp->aux,
+	return drm_dp_dpcd_read(&intel_dp->aux,
 				       DP_DEVICE_SERVICE_IRQ_VECTOR,
 				       sink_irq_vector, 1) == 1;
 }
@@ -4116,7 +3995,7 @@ intel_dp_get_sink_irq_esi(struct intel_dp *intel_dp, u8 *sink_irq_vector)
 {
 	int ret;
 
-	ret = intel_dp_dpcd_read_wake(&intel_dp->aux,
+	ret = drm_dp_dpcd_read(&intel_dp->aux,
 					     DP_SINK_COUNT_ESI,
 					     sink_irq_vector, 14);
 	if (ret != 14)
@@ -4292,6 +4171,36 @@ go_again:
 	return -EINVAL;
 }
 
+static void
+intel_dp_check_link_status(struct intel_dp *intel_dp)
+{
+	struct intel_encoder *intel_encoder = &dp_to_dig_port(intel_dp)->base;
+	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	u8 link_status[DP_LINK_STATUS_SIZE];
+
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+
+	if (!intel_dp_get_link_status(intel_dp, link_status)) {
+		DRM_ERROR("Failed to get link status\n");
+		return;
+	}
+
+	if (!intel_encoder->base.crtc)
+		return;
+
+	if (!to_intel_crtc(intel_encoder->base.crtc)->active)
+		return;
+
+	/* if link training is requested we should perform it always */
+	if ((intel_dp->compliance_test_type == DP_TEST_LINK_TRAINING) ||
+	    (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count))) {
+		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
+			      intel_encoder->base.name);
+		intel_dp_start_link_train(intel_dp);
+		intel_dp_stop_link_train(intel_dp);
+	}
+}
+
 /*
  * According to DP spec
  * 5.1.2:
@@ -4299,16 +4208,19 @@ go_again:
  *  2. Configure link according to Receiver Capabilities
  *  3. Use Link Training from 2.5.3.3 and 3.5.1.3
  *  4. Check link status on receipt of hot-plug interrupt
+ *
+ * intel_dp_short_pulse -  handles short pulse interrupts
+ * when full detection is not required.
+ * Returns %true if short pulse is handled and full detection
+ * is NOT required and %false otherwise.
  */
-static void
-intel_dp_check_link_status(struct intel_dp *intel_dp)
+static bool
+intel_dp_short_pulse(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
-	struct intel_encoder *intel_encoder = &dp_to_dig_port(intel_dp)->base;
 	u8 sink_irq_vector;
-	u8 link_status[DP_LINK_STATUS_SIZE];
-
-	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+	u8 old_sink_count = intel_dp->sink_count;
+	bool ret;
 
 	/*
 	 * Clearing compliance test variables to allow capturing
@@ -4318,20 +4230,17 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 	intel_dp->compliance_test_type = 0;
 	intel_dp->compliance_test_data = 0;
 
-	if (!intel_encoder->base.crtc)
-		return;
-
-	if (!to_intel_crtc(intel_encoder->base.crtc)->active)
-		return;
-
-	/* Try to read receiver status if the link appears to be up */
-	if (!intel_dp_get_link_status(intel_dp, link_status)) {
-		return;
-	}
+	/*
+	 * Now read the DPCD to see if it's actually running
+	 * If the current value of sink count doesn't match with
+	 * the value that was stored earlier or dpcd read failed
+	 * we need to do full detection
+	 */
+	ret = intel_dp_get_dpcd(intel_dp);
 
-	/* Now read the DPCD to see if it's actually running */
-	if (!intel_dp_get_dpcd(intel_dp)) {
-		return;
+	if ((old_sink_count != intel_dp->sink_count) || !ret) {
+		/* No need to proceed if we are going to do full detect */
+		return false;
 	}
 
 	/* Try to read the source of the interrupt */
@@ -4348,14 +4257,11 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 			DRM_DEBUG_DRIVER("CP or sink specific irq unhandled\n");
 	}
 
-	/* if link training is requested we should perform it always */
-	if ((intel_dp->compliance_test_type == DP_TEST_LINK_TRAINING) ||
-		(!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count))) {
-		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
-			      intel_encoder->base.name);
-		intel_dp_start_link_train(intel_dp);
-		intel_dp_stop_link_train(intel_dp);
-	}
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
+	intel_dp_check_link_status(intel_dp);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
+
+	return true;
 }
 
 /* XXX this is probably wrong for multiple downstream ports */
@@ -4368,6 +4274,9 @@ intel_dp_detect_dpcd(struct intel_dp *intel_dp)
 	if (!intel_dp_get_dpcd(intel_dp))
 		return connector_status_disconnected;
 
+	if (is_edp(intel_dp))
+		return connector_status_connected;
+
 	/* if there's no downstream port, we're done */
 	if (!(dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DWN_STRM_PORT_PRESENT))
 		return connector_status_connected;
@@ -4375,14 +4284,9 @@ intel_dp_detect_dpcd(struct intel_dp *intel_dp)
 	/* If we're HPD-aware, SINK_COUNT changes dynamically */
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
 	    intel_dp->downstream_ports[0] & DP_DS_PORT_HPD) {
-		uint8_t reg;
 
-		if (intel_dp_dpcd_read_wake(&intel_dp->aux, DP_SINK_COUNT,
-					    &reg, 1) < 0)
-			return connector_status_unknown;
-
-		return DP_GET_SINK_COUNT(reg) ? connector_status_connected
-					      : connector_status_disconnected;
+		return intel_dp->sink_count ?
+		connector_status_connected : connector_status_disconnected;
 	}
 
 	/* If no HPD, poke DDC gently */
@@ -4591,6 +4495,7 @@ intel_dp_set_edid(struct intel_dp *intel_dp)
 	struct intel_connector *intel_connector = intel_dp->attached_connector;
 	struct edid *edid;
 
+	intel_dp_unset_edid(intel_dp);
 	edid = intel_dp_get_edid(intel_dp);
 	intel_connector->detect_edid = edid;
 
@@ -4611,9 +4516,10 @@ intel_dp_unset_edid(struct intel_dp *intel_dp)
 	intel_dp->has_audio = false;
 }
 
-static enum drm_connector_status
-intel_dp_detect(struct drm_connector *connector, bool force)
+static void
+intel_dp_long_pulse(struct intel_connector *intel_connector)
 {
+	struct drm_connector *connector = &intel_connector->base;
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct intel_encoder *intel_encoder = &intel_dig_port->base;
@@ -4623,17 +4529,6 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	bool ret;
 	u8 sink_irq_vector;
 
-	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, connector->name);
-	intel_dp_unset_edid(intel_dp);
-
-	if (intel_dp->is_mst) {
-		/* MST devices are disconnected from a monitor POV */
-		if (intel_encoder->type != INTEL_OUTPUT_EDP)
-			intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
-		return connector_status_disconnected;
-	}
-
 	power_domain = intel_display_port_aux_power_domain(intel_encoder);
 	intel_display_power_get(to_i915(dev), power_domain);
 
@@ -4651,19 +4546,42 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 		intel_dp->compliance_test_type = 0;
 		intel_dp->compliance_test_data = 0;
 
+		if (intel_dp->is_mst) {
+			DRM_DEBUG_KMS("MST device may have disappeared %d vs %d\n",
+				      intel_dp->is_mst,
+				      intel_dp->mst_mgr.mst_state);
+			intel_dp->is_mst = false;
+			drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr,
+							intel_dp->is_mst);
+		}
+
 		goto out;
 	}
 
+	if (intel_encoder->type != INTEL_OUTPUT_EDP)
+		intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
+
 	intel_dp_probe_oui(intel_dp);
 
 	ret = intel_dp_probe_mst(intel_dp);
 	if (ret) {
-		/* if we are in MST mode then this connector
-		   won't appear connected or have anything with EDID on it */
-		if (intel_encoder->type != INTEL_OUTPUT_EDP)
-			intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
+		/*
+		 * If we are in MST mode then this connector
+		 * won't appear connected or have anything
+		 * with EDID on it
+		 */
 		status = connector_status_disconnected;
 		goto out;
+	} else if (connector->status == connector_status_connected) {
+		/*
+		 * If display was connected already and is still connected
+		 * check links status, there has been known issues of
+		 * link loss triggerring long pulse!!!!
+		 */
+		drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
+		intel_dp_check_link_status(intel_dp);
+		drm_modeset_unlock(&dev->mode_config.connection_mutex);
+		goto out;
 	}
 
 	/*
@@ -4676,9 +4594,8 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 
 	intel_dp_set_edid(intel_dp);
 
-	if (intel_encoder->type != INTEL_OUTPUT_EDP)
-		intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
 	status = connector_status_connected;
+	intel_dp->detect_done = true;
 
 	/* Try to read the source of the interrupt */
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
@@ -4695,8 +4612,43 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	}
 
 out:
+	if ((status != connector_status_connected) &&
+	    (intel_dp->is_mst == false))
+		intel_dp_unset_edid(intel_dp);
+
 	intel_display_power_put(to_i915(dev), power_domain);
-	return status;
+	return;
+}
+
+static enum drm_connector_status
+intel_dp_detect(struct drm_connector *connector, bool force)
+{
+	struct intel_dp *intel_dp = intel_attached_dp(connector);
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
+	struct intel_connector *intel_connector = to_intel_connector(connector);
+
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
+		      connector->base.id, connector->name);
+
+	if (intel_dp->is_mst) {
+		/* MST devices are disconnected from a monitor POV */
+		intel_dp_unset_edid(intel_dp);
+		if (intel_encoder->type != INTEL_OUTPUT_EDP)
+			intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
+		return connector_status_disconnected;
+	}
+
+	/* If full detect is not performed yet, do a full detect */
+	if (!intel_dp->detect_done)
+		intel_dp_long_pulse(intel_dp->attached_connector);
+
+	intel_dp->detect_done = false;
+
+	if (intel_connector->detect_edid)
+		return connector_status_connected;
+	else
+		return connector_status_disconnected;
 }
 
 static void
@@ -4835,6 +4787,11 @@ intel_dp_set_property(struct drm_connector *connector,
 			DRM_DEBUG_KMS("no scaling not supported\n");
 			return -EINVAL;
 		}
+		if (HAS_GMCH_DISPLAY(dev_priv) &&
+		    val == DRM_MODE_SCALE_CENTER) {
+			DRM_DEBUG_KMS("centering not supported\n");
+			return -EINVAL;
+		}
 
 		if (intel_connector->panel.fitting_mode == val) {
 			/* the eDP scaling property is not changed */
@@ -5023,44 +4980,37 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
 		/* indicate that we need to restart link training */
 		intel_dp->train_set_valid = false;
 
-		if (!intel_digital_port_connected(dev_priv, intel_dig_port))
-			goto mst_fail;
+		intel_dp_long_pulse(intel_dp->attached_connector);
+		if (intel_dp->is_mst)
+			ret = IRQ_HANDLED;
+		goto put_power;
 
-		if (!intel_dp_get_dpcd(intel_dp)) {
-			goto mst_fail;
-		}
-
-		intel_dp_probe_oui(intel_dp);
-
-		if (!intel_dp_probe_mst(intel_dp)) {
-			drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
-			intel_dp_check_link_status(intel_dp);
-			drm_modeset_unlock(&dev->mode_config.connection_mutex);
-			goto mst_fail;
-		}
 	} else {
 		if (intel_dp->is_mst) {
-			if (intel_dp_check_mst_status(intel_dp) == -EINVAL)
-				goto mst_fail;
+			if (intel_dp_check_mst_status(intel_dp) == -EINVAL) {
+				/*
+				 * If we were in MST mode, and device is not
+				 * there, get out of MST mode
+				 */
+				DRM_DEBUG_KMS("MST device may have disappeared %d vs %d\n",
+					      intel_dp->is_mst, intel_dp->mst_mgr.mst_state);
+				intel_dp->is_mst = false;
+				drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr,
+								intel_dp->is_mst);
+				goto put_power;
+			}
 		}
 
 		if (!intel_dp->is_mst) {
-			drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
-			intel_dp_check_link_status(intel_dp);
-			drm_modeset_unlock(&dev->mode_config.connection_mutex);
+			if (!intel_dp_short_pulse(intel_dp)) {
+				intel_dp_long_pulse(intel_dp->attached_connector);
+				goto put_power;
+			}
 		}
 	}
 
 	ret = IRQ_HANDLED;
 
-	goto put_power;
-mst_fail:
-	/* if we were in MST mode, and device is not there get out of MST mode */
-	if (intel_dp->is_mst) {
-		DRM_DEBUG_KMS("MST device may have disappeared %d vs %d\n", intel_dp->is_mst, intel_dp->mst_mgr.mst_state);
-		intel_dp->is_mst = false;
-		drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr, intel_dp->is_mst);
-	}
 put_power:
 	intel_display_power_put(dev_priv, power_domain);
 
@@ -5071,14 +5021,6 @@ put_power:
 bool intel_dp_is_edp(struct drm_device *dev, enum port port)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	union child_device_config *p_child;
-	int i;
-	static const short port_mapping[] = {
-		[PORT_B] = DVO_PORT_DPB,
-		[PORT_C] = DVO_PORT_DPC,
-		[PORT_D] = DVO_PORT_DPD,
-		[PORT_E] = DVO_PORT_DPE,
-	};
 
 	/*
 	 * eDP not supported on g4x. so bail out early just
@@ -5090,18 +5032,7 @@ bool intel_dp_is_edp(struct drm_device *dev, enum port port)
 	if (port == PORT_A)
 		return true;
 
-	if (!dev_priv->vbt.child_dev_num)
-		return false;
-
-	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
-		p_child = dev_priv->vbt.child_dev + i;
-
-		if (p_child->common.dvo_port == port_mapping[port] &&
-		    (p_child->common.device_type & DEVICE_TYPE_eDP_BITS) ==
-		    (DEVICE_TYPE_eDP & DEVICE_TYPE_eDP_BITS))
-			return true;
-	}
-	return false;
+	return intel_bios_is_port_edp(dev_priv, port);
 }
 
 void
@@ -5208,7 +5139,7 @@ intel_dp_init_panel_power_sequencer(struct drm_device *dev,
 	DRM_DEBUG_KMS("cur t1_t3 %d t8 %d t9 %d t10 %d t11_t12 %d\n",
 		      cur.t1_t3, cur.t8, cur.t9, cur.t10, cur.t11_t12);
 
-	vbt = dev_priv->vbt.edp_pps;
+	vbt = dev_priv->vbt.edp.pps;
 
 	/* Upper limits from eDP 1.3 spec. Note that we use the clunky units of
 	 * our hw here, which are all in 100usec. */
@@ -5259,7 +5190,7 @@ intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 pp_on, pp_off, pp_div, port_sel = 0;
-	int div = HAS_PCH_SPLIT(dev) ? intel_pch_rawclk(dev) : intel_hrawclk(dev);
+	int div = dev_priv->rawclk_freq / 1000;
 	i915_reg_t pp_on_reg, pp_off_reg, pp_div_reg, pp_ctrl_reg;
 	enum port port = dp_to_dig_port(intel_dp)->port;
 	const struct edp_power_seq *seq = &intel_dp->pps_delays;
@@ -5852,19 +5783,17 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 	/* intel_dp vfuncs */
 	if (INTEL_INFO(dev)->gen >= 9)
 		intel_dp->get_aux_clock_divider = skl_get_aux_clock_divider;
-	else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
-		intel_dp->get_aux_clock_divider = vlv_get_aux_clock_divider;
 	else if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		intel_dp->get_aux_clock_divider = hsw_get_aux_clock_divider;
 	else if (HAS_PCH_SPLIT(dev))
 		intel_dp->get_aux_clock_divider = ilk_get_aux_clock_divider;
 	else
-		intel_dp->get_aux_clock_divider = i9xx_get_aux_clock_divider;
+		intel_dp->get_aux_clock_divider = g4x_get_aux_clock_divider;
 
 	if (INTEL_INFO(dev)->gen >= 9)
 		intel_dp->get_aux_send_ctl = skl_get_aux_send_ctl;
 	else
-		intel_dp->get_aux_send_ctl = i9xx_get_aux_send_ctl;
+		intel_dp->get_aux_send_ctl = g4x_get_aux_send_ctl;
 
 	if (HAS_DDI(dev))
 		intel_dp->prepare_link_retrain = intel_ddi_prepare_link_retrain;
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index 2c999725b3d4..7a34090cef34 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -33,7 +33,6 @@
 static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 					struct intel_crtc_state *pipe_config)
 {
-	struct drm_device *dev = encoder->base.dev;
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
 	struct intel_digital_port *intel_dig_port = intel_mst->primary;
 	struct intel_dp *intel_dp = &intel_dig_port->dp;
@@ -90,9 +89,6 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 
 	pipe_config->dp_m_n.tu = slots;
 
-	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
-		hsw_dp_set_ddi_pll_sel(pipe_config);
-
 	return true;
 
 }
@@ -106,7 +102,7 @@ static void intel_mst_disable_dp(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("%d\n", intel_dp->active_mst_links);
 
-	drm_dp_mst_reset_vcpi_slots(&intel_dp->mst_mgr, intel_mst->port);
+	drm_dp_mst_reset_vcpi_slots(&intel_dp->mst_mgr, intel_mst->connector->port);
 
 	ret = drm_dp_update_payload_part1(&intel_dp->mst_mgr);
 	if (ret) {
@@ -127,10 +123,11 @@ static void intel_mst_post_disable_dp(struct intel_encoder *encoder)
 	/* and this can also fail */
 	drm_dp_update_payload_part2(&intel_dp->mst_mgr);
 
-	drm_dp_mst_deallocate_vcpi(&intel_dp->mst_mgr, intel_mst->port);
+	drm_dp_mst_deallocate_vcpi(&intel_dp->mst_mgr, intel_mst->connector->port);
 
 	intel_dp->active_mst_links--;
-	intel_mst->port = NULL;
+
+	intel_mst->connector = NULL;
 	if (intel_dp->active_mst_links == 0) {
 		intel_dig_port->base.post_disable(&intel_dig_port->base);
 		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_OFF);
@@ -170,7 +167,8 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder)
 	found->encoder = encoder;
 
 	DRM_DEBUG_KMS("%d\n", intel_dp->active_mst_links);
-	intel_mst->port = found->port;
+
+	intel_mst->connector = found;
 
 	if (intel_dp->active_mst_links == 0) {
 		intel_prepare_ddi_buffer(&intel_dig_port->base);
@@ -188,7 +186,7 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder)
 	}
 
 	ret = drm_dp_mst_allocate_vcpi(&intel_dp->mst_mgr,
-				       intel_mst->port,
+				       intel_mst->connector->port,
 				       intel_crtc->config->pbn, &slots);
 	if (ret == false) {
 		DRM_ERROR("failed to allocate vcpi\n");
@@ -229,7 +227,7 @@ static bool intel_dp_mst_enc_get_hw_state(struct intel_encoder *encoder,
 {
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
 	*pipe = intel_mst->pipe;
-	if (intel_mst->port)
+	if (intel_mst->connector)
 		return true;
 	return false;
 }
@@ -290,10 +288,11 @@ static int intel_dp_mst_get_ddc_modes(struct drm_connector *connector)
 	struct edid *edid;
 	int ret;
 
-	edid = drm_dp_mst_get_edid(connector, &intel_dp->mst_mgr, intel_connector->port);
-	if (!edid)
-		return 0;
+	if (!intel_dp) {
+		return intel_connector_update_modes(connector, NULL);
+	}
 
+	edid = drm_dp_mst_get_edid(connector, &intel_dp->mst_mgr, intel_connector->port);
 	ret = intel_connector_update_modes(connector, edid);
 	kfree(edid);
 
@@ -306,6 +305,8 @@ intel_dp_mst_detect(struct drm_connector *connector, bool force)
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct intel_dp *intel_dp = intel_connector->mst_port;
 
+	if (!intel_dp)
+		return connector_status_disconnected;
 	return drm_dp_mst_detect_port(connector, &intel_dp->mst_mgr, intel_connector->port);
 }
 
@@ -371,6 +372,8 @@ static struct drm_encoder *intel_mst_atomic_best_encoder(struct drm_connector *c
 	struct intel_dp *intel_dp = intel_connector->mst_port;
 	struct intel_crtc *crtc = to_intel_crtc(state->crtc);
 
+	if (!intel_dp)
+		return NULL;
 	return &intel_dp->mst_encoders[crtc->pipe]->base.base;
 }
 
@@ -378,6 +381,8 @@ static struct drm_encoder *intel_mst_best_encoder(struct drm_connector *connecto
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct intel_dp *intel_dp = intel_connector->mst_port;
+	if (!intel_dp)
+		return NULL;
 	return &intel_dp->mst_encoders[0]->base.base;
 }
 
@@ -488,23 +493,11 @@ static void intel_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
 
 	/* need to nuke the connector */
 	drm_modeset_lock_all(dev);
-	if (connector->state->crtc) {
-		struct drm_mode_set set;
-		int ret;
-
-		memset(&set, 0, sizeof(set));
-		set.crtc = connector->state->crtc,
-
-		ret = drm_atomic_helper_set_config(&set);
-
-		WARN(ret, "Disabling mst crtc failed with %i\n", ret);
-	}
-
 	intel_connector_remove_from_fbdev(intel_connector);
-	drm_connector_cleanup(connector);
+	intel_connector->mst_port = NULL;
 	drm_modeset_unlock_all(dev);
 
-	kfree(intel_connector);
+	drm_connector_unreference(&intel_connector->base);
 	DRM_DEBUG_KMS("\n");
 }
 
diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.c b/drivers/gpu/drm/i915/intel_dpll_mgr.c
new file mode 100644
index 000000000000..639bf0209c15
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dpll_mgr.c
@@ -0,0 +1,1783 @@
+/*
+ * Copyright © 2006-2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "intel_drv.h"
+
+struct intel_shared_dpll *
+intel_get_shared_dpll_by_id(struct drm_i915_private *dev_priv,
+			    enum intel_dpll_id id)
+{
+	return &dev_priv->shared_dplls[id];
+}
+
+enum intel_dpll_id
+intel_get_shared_dpll_id(struct drm_i915_private *dev_priv,
+			 struct intel_shared_dpll *pll)
+{
+	if (WARN_ON(pll < dev_priv->shared_dplls||
+		    pll > &dev_priv->shared_dplls[dev_priv->num_shared_dpll]))
+		return -1;
+
+	return (enum intel_dpll_id) (pll - dev_priv->shared_dplls);
+}
+
+void
+intel_shared_dpll_config_get(struct intel_shared_dpll_config *config,
+			     struct intel_shared_dpll *pll,
+			     struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum intel_dpll_id id = intel_get_shared_dpll_id(dev_priv, pll);
+
+	config[id].crtc_mask |= 1 << crtc->pipe;
+}
+
+void
+intel_shared_dpll_config_put(struct intel_shared_dpll_config *config,
+			     struct intel_shared_dpll *pll,
+			     struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum intel_dpll_id id = intel_get_shared_dpll_id(dev_priv, pll);
+
+	config[id].crtc_mask &= ~(1 << crtc->pipe);
+}
+
+/* For ILK+ */
+void assert_shared_dpll(struct drm_i915_private *dev_priv,
+			struct intel_shared_dpll *pll,
+			bool state)
+{
+	bool cur_state;
+	struct intel_dpll_hw_state hw_state;
+
+	if (WARN(!pll, "asserting DPLL %s with no DPLL\n", onoff(state)))
+		return;
+
+	cur_state = pll->funcs.get_hw_state(dev_priv, pll, &hw_state);
+	I915_STATE_WARN(cur_state != state,
+	     "%s assertion failure (expected %s, current %s)\n",
+			pll->name, onoff(state), onoff(cur_state));
+}
+
+void intel_prepare_shared_dpll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_shared_dpll *pll = crtc->config->shared_dpll;
+
+	if (WARN_ON(pll == NULL))
+		return;
+
+	mutex_lock(&dev_priv->dpll_lock);
+	WARN_ON(!pll->config.crtc_mask);
+	if (!pll->active_mask) {
+		DRM_DEBUG_DRIVER("setting up %s\n", pll->name);
+		WARN_ON(pll->on);
+		assert_shared_dpll_disabled(dev_priv, pll);
+
+		pll->funcs.mode_set(dev_priv, pll);
+	}
+	mutex_unlock(&dev_priv->dpll_lock);
+}
+
+/**
+ * intel_enable_shared_dpll - enable PCH PLL
+ * @dev_priv: i915 private structure
+ * @pipe: pipe PLL to enable
+ *
+ * The PCH PLL needs to be enabled before the PCH transcoder, since it
+ * drives the transcoder clock.
+ */
+void intel_enable_shared_dpll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_shared_dpll *pll = crtc->config->shared_dpll;
+	unsigned crtc_mask = 1 << drm_crtc_index(&crtc->base);
+	unsigned old_mask;
+
+	if (WARN_ON(pll == NULL))
+		return;
+
+	mutex_lock(&dev_priv->dpll_lock);
+	old_mask = pll->active_mask;
+
+	if (WARN_ON(!(pll->config.crtc_mask & crtc_mask)) ||
+	    WARN_ON(pll->active_mask & crtc_mask))
+		goto out;
+
+	pll->active_mask |= crtc_mask;
+
+	DRM_DEBUG_KMS("enable %s (active %x, on? %d) for crtc %d\n",
+		      pll->name, pll->active_mask, pll->on,
+		      crtc->base.base.id);
+
+	if (old_mask) {
+		WARN_ON(!pll->on);
+		assert_shared_dpll_enabled(dev_priv, pll);
+		goto out;
+	}
+	WARN_ON(pll->on);
+
+	DRM_DEBUG_KMS("enabling %s\n", pll->name);
+	pll->funcs.enable(dev_priv, pll);
+	pll->on = true;
+
+out:
+	mutex_unlock(&dev_priv->dpll_lock);
+}
+
+void intel_disable_shared_dpll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_shared_dpll *pll = crtc->config->shared_dpll;
+	unsigned crtc_mask = 1 << drm_crtc_index(&crtc->base);
+
+	/* PCH only available on ILK+ */
+	if (INTEL_INFO(dev)->gen < 5)
+		return;
+
+	if (pll == NULL)
+		return;
+
+	mutex_lock(&dev_priv->dpll_lock);
+	if (WARN_ON(!(pll->active_mask & crtc_mask)))
+		goto out;
+
+	DRM_DEBUG_KMS("disable %s (active %x, on? %d) for crtc %d\n",
+		      pll->name, pll->active_mask, pll->on,
+		      crtc->base.base.id);
+
+	assert_shared_dpll_enabled(dev_priv, pll);
+	WARN_ON(!pll->on);
+
+	pll->active_mask &= ~crtc_mask;
+	if (pll->active_mask)
+		goto out;
+
+	DRM_DEBUG_KMS("disabling %s\n", pll->name);
+	pll->funcs.disable(dev_priv, pll);
+	pll->on = false;
+
+out:
+	mutex_unlock(&dev_priv->dpll_lock);
+}
+
+static struct intel_shared_dpll *
+intel_find_shared_dpll(struct intel_crtc *crtc,
+		       struct intel_crtc_state *crtc_state,
+		       enum intel_dpll_id range_min,
+		       enum intel_dpll_id range_max)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_shared_dpll *pll;
+	struct intel_shared_dpll_config *shared_dpll;
+	enum intel_dpll_id i;
+
+	shared_dpll = intel_atomic_get_shared_dpll_state(crtc_state->base.state);
+
+	for (i = range_min; i <= range_max; i++) {
+		pll = &dev_priv->shared_dplls[i];
+
+		/* Only want to check enabled timings first */
+		if (shared_dpll[i].crtc_mask == 0)
+			continue;
+
+		if (memcmp(&crtc_state->dpll_hw_state,
+			   &shared_dpll[i].hw_state,
+			   sizeof(crtc_state->dpll_hw_state)) == 0) {
+			DRM_DEBUG_KMS("CRTC:%d sharing existing %s (crtc mask 0x%08x, active %x)\n",
+				      crtc->base.base.id, pll->name,
+				      shared_dpll[i].crtc_mask,
+				      pll->active_mask);
+			return pll;
+		}
+	}
+
+	/* Ok no matching timings, maybe there's a free one? */
+	for (i = range_min; i <= range_max; i++) {
+		pll = &dev_priv->shared_dplls[i];
+		if (shared_dpll[i].crtc_mask == 0) {
+			DRM_DEBUG_KMS("CRTC:%d allocated %s\n",
+				      crtc->base.base.id, pll->name);
+			return pll;
+		}
+	}
+
+	return NULL;
+}
+
+static void
+intel_reference_shared_dpll(struct intel_shared_dpll *pll,
+			    struct intel_crtc_state *crtc_state)
+{
+	struct intel_shared_dpll_config *shared_dpll;
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	enum intel_dpll_id i = pll->id;
+
+	shared_dpll = intel_atomic_get_shared_dpll_state(crtc_state->base.state);
+
+	if (shared_dpll[i].crtc_mask == 0)
+		shared_dpll[i].hw_state =
+			crtc_state->dpll_hw_state;
+
+	crtc_state->shared_dpll = pll;
+	DRM_DEBUG_DRIVER("using %s for pipe %c\n", pll->name,
+			 pipe_name(crtc->pipe));
+
+	intel_shared_dpll_config_get(shared_dpll, pll, crtc);
+}
+
+void intel_shared_dpll_commit(struct drm_atomic_state *state)
+{
+	struct drm_i915_private *dev_priv = to_i915(state->dev);
+	struct intel_shared_dpll_config *shared_dpll;
+	struct intel_shared_dpll *pll;
+	enum intel_dpll_id i;
+
+	if (!to_intel_atomic_state(state)->dpll_set)
+		return;
+
+	shared_dpll = to_intel_atomic_state(state)->shared_dpll;
+	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
+		pll = &dev_priv->shared_dplls[i];
+		pll->config = shared_dpll[i];
+	}
+}
+
+static bool ibx_pch_dpll_get_hw_state(struct drm_i915_private *dev_priv,
+				      struct intel_shared_dpll *pll,
+				      struct intel_dpll_hw_state *hw_state)
+{
+	uint32_t val;
+
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	val = I915_READ(PCH_DPLL(pll->id));
+	hw_state->dpll = val;
+	hw_state->fp0 = I915_READ(PCH_FP0(pll->id));
+	hw_state->fp1 = I915_READ(PCH_FP1(pll->id));
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return val & DPLL_VCO_ENABLE;
+}
+
+static void ibx_pch_dpll_mode_set(struct drm_i915_private *dev_priv,
+				  struct intel_shared_dpll *pll)
+{
+	I915_WRITE(PCH_FP0(pll->id), pll->config.hw_state.fp0);
+	I915_WRITE(PCH_FP1(pll->id), pll->config.hw_state.fp1);
+}
+
+static void ibx_assert_pch_refclk_enabled(struct drm_i915_private *dev_priv)
+{
+	u32 val;
+	bool enabled;
+
+	I915_STATE_WARN_ON(!(HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv)));
+
+	val = I915_READ(PCH_DREF_CONTROL);
+	enabled = !!(val & (DREF_SSC_SOURCE_MASK | DREF_NONSPREAD_SOURCE_MASK |
+			    DREF_SUPERSPREAD_SOURCE_MASK));
+	I915_STATE_WARN(!enabled, "PCH refclk assertion failure, should be active but is disabled\n");
+}
+
+static void ibx_pch_dpll_enable(struct drm_i915_private *dev_priv,
+				struct intel_shared_dpll *pll)
+{
+	/* PCH refclock must be enabled first */
+	ibx_assert_pch_refclk_enabled(dev_priv);
+
+	I915_WRITE(PCH_DPLL(pll->id), pll->config.hw_state.dpll);
+
+	/* Wait for the clocks to stabilize. */
+	POSTING_READ(PCH_DPLL(pll->id));
+	udelay(150);
+
+	/* The pixel multiplier can only be updated once the
+	 * DPLL is enabled and the clocks are stable.
+	 *
+	 * So write it again.
+	 */
+	I915_WRITE(PCH_DPLL(pll->id), pll->config.hw_state.dpll);
+	POSTING_READ(PCH_DPLL(pll->id));
+	udelay(200);
+}
+
+static void ibx_pch_dpll_disable(struct drm_i915_private *dev_priv,
+				 struct intel_shared_dpll *pll)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct intel_crtc *crtc;
+
+	/* Make sure no transcoder isn't still depending on us. */
+	for_each_intel_crtc(dev, crtc) {
+		if (crtc->config->shared_dpll == pll)
+			assert_pch_transcoder_disabled(dev_priv, crtc->pipe);
+	}
+
+	I915_WRITE(PCH_DPLL(pll->id), 0);
+	POSTING_READ(PCH_DPLL(pll->id));
+	udelay(200);
+}
+
+static struct intel_shared_dpll *
+ibx_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+	     struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_shared_dpll *pll;
+	enum intel_dpll_id i;
+
+	if (HAS_PCH_IBX(dev_priv)) {
+		/* Ironlake PCH has a fixed PLL->PCH pipe mapping. */
+		i = (enum intel_dpll_id) crtc->pipe;
+		pll = &dev_priv->shared_dplls[i];
+
+		DRM_DEBUG_KMS("CRTC:%d using pre-allocated %s\n",
+			      crtc->base.base.id, pll->name);
+	} else {
+		pll = intel_find_shared_dpll(crtc, crtc_state,
+					     DPLL_ID_PCH_PLL_A,
+					     DPLL_ID_PCH_PLL_B);
+	}
+
+	/* reference the pll */
+	intel_reference_shared_dpll(pll, crtc_state);
+
+	return pll;
+}
+
+static const struct intel_shared_dpll_funcs ibx_pch_dpll_funcs = {
+	.mode_set = ibx_pch_dpll_mode_set,
+	.enable = ibx_pch_dpll_enable,
+	.disable = ibx_pch_dpll_disable,
+	.get_hw_state = ibx_pch_dpll_get_hw_state,
+};
+
+static void hsw_ddi_wrpll_enable(struct drm_i915_private *dev_priv,
+			       struct intel_shared_dpll *pll)
+{
+	I915_WRITE(WRPLL_CTL(pll->id), pll->config.hw_state.wrpll);
+	POSTING_READ(WRPLL_CTL(pll->id));
+	udelay(20);
+}
+
+static void hsw_ddi_spll_enable(struct drm_i915_private *dev_priv,
+				struct intel_shared_dpll *pll)
+{
+	I915_WRITE(SPLL_CTL, pll->config.hw_state.spll);
+	POSTING_READ(SPLL_CTL);
+	udelay(20);
+}
+
+static void hsw_ddi_wrpll_disable(struct drm_i915_private *dev_priv,
+				  struct intel_shared_dpll *pll)
+{
+	uint32_t val;
+
+	val = I915_READ(WRPLL_CTL(pll->id));
+	I915_WRITE(WRPLL_CTL(pll->id), val & ~WRPLL_PLL_ENABLE);
+	POSTING_READ(WRPLL_CTL(pll->id));
+}
+
+static void hsw_ddi_spll_disable(struct drm_i915_private *dev_priv,
+				 struct intel_shared_dpll *pll)
+{
+	uint32_t val;
+
+	val = I915_READ(SPLL_CTL);
+	I915_WRITE(SPLL_CTL, val & ~SPLL_PLL_ENABLE);
+	POSTING_READ(SPLL_CTL);
+}
+
+static bool hsw_ddi_wrpll_get_hw_state(struct drm_i915_private *dev_priv,
+				       struct intel_shared_dpll *pll,
+				       struct intel_dpll_hw_state *hw_state)
+{
+	uint32_t val;
+
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	val = I915_READ(WRPLL_CTL(pll->id));
+	hw_state->wrpll = val;
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return val & WRPLL_PLL_ENABLE;
+}
+
+static bool hsw_ddi_spll_get_hw_state(struct drm_i915_private *dev_priv,
+				      struct intel_shared_dpll *pll,
+				      struct intel_dpll_hw_state *hw_state)
+{
+	uint32_t val;
+
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	val = I915_READ(SPLL_CTL);
+	hw_state->spll = val;
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return val & SPLL_PLL_ENABLE;
+}
+
+static uint32_t hsw_pll_to_ddi_pll_sel(struct intel_shared_dpll *pll)
+{
+	switch (pll->id) {
+	case DPLL_ID_WRPLL1:
+		return PORT_CLK_SEL_WRPLL1;
+	case DPLL_ID_WRPLL2:
+		return PORT_CLK_SEL_WRPLL2;
+	case DPLL_ID_SPLL:
+		return PORT_CLK_SEL_SPLL;
+	case DPLL_ID_LCPLL_810:
+		return PORT_CLK_SEL_LCPLL_810;
+	case DPLL_ID_LCPLL_1350:
+		return PORT_CLK_SEL_LCPLL_1350;
+	case DPLL_ID_LCPLL_2700:
+		return PORT_CLK_SEL_LCPLL_2700;
+	default:
+		return PORT_CLK_SEL_NONE;
+	}
+}
+
+#define LC_FREQ 2700
+#define LC_FREQ_2K U64_C(LC_FREQ * 2000)
+
+#define P_MIN 2
+#define P_MAX 64
+#define P_INC 2
+
+/* Constraints for PLL good behavior */
+#define REF_MIN 48
+#define REF_MAX 400
+#define VCO_MIN 2400
+#define VCO_MAX 4800
+
+struct hsw_wrpll_rnp {
+	unsigned p, n2, r2;
+};
+
+static unsigned hsw_wrpll_get_budget_for_freq(int clock)
+{
+	unsigned budget;
+
+	switch (clock) {
+	case 25175000:
+	case 25200000:
+	case 27000000:
+	case 27027000:
+	case 37762500:
+	case 37800000:
+	case 40500000:
+	case 40541000:
+	case 54000000:
+	case 54054000:
+	case 59341000:
+	case 59400000:
+	case 72000000:
+	case 74176000:
+	case 74250000:
+	case 81000000:
+	case 81081000:
+	case 89012000:
+	case 89100000:
+	case 108000000:
+	case 108108000:
+	case 111264000:
+	case 111375000:
+	case 148352000:
+	case 148500000:
+	case 162000000:
+	case 162162000:
+	case 222525000:
+	case 222750000:
+	case 296703000:
+	case 297000000:
+		budget = 0;
+		break;
+	case 233500000:
+	case 245250000:
+	case 247750000:
+	case 253250000:
+	case 298000000:
+		budget = 1500;
+		break;
+	case 169128000:
+	case 169500000:
+	case 179500000:
+	case 202000000:
+		budget = 2000;
+		break;
+	case 256250000:
+	case 262500000:
+	case 270000000:
+	case 272500000:
+	case 273750000:
+	case 280750000:
+	case 281250000:
+	case 286000000:
+	case 291750000:
+		budget = 4000;
+		break;
+	case 267250000:
+	case 268500000:
+		budget = 5000;
+		break;
+	default:
+		budget = 1000;
+		break;
+	}
+
+	return budget;
+}
+
+static void hsw_wrpll_update_rnp(uint64_t freq2k, unsigned budget,
+				 unsigned r2, unsigned n2, unsigned p,
+				 struct hsw_wrpll_rnp *best)
+{
+	uint64_t a, b, c, d, diff, diff_best;
+
+	/* No best (r,n,p) yet */
+	if (best->p == 0) {
+		best->p = p;
+		best->n2 = n2;
+		best->r2 = r2;
+		return;
+	}
+
+	/*
+	 * Output clock is (LC_FREQ_2K / 2000) * N / (P * R), which compares to
+	 * freq2k.
+	 *
+	 * delta = 1e6 *
+	 *	   abs(freq2k - (LC_FREQ_2K * n2/(p * r2))) /
+	 *	   freq2k;
+	 *
+	 * and we would like delta <= budget.
+	 *
+	 * If the discrepancy is above the PPM-based budget, always prefer to
+	 * improve upon the previous solution.  However, if you're within the
+	 * budget, try to maximize Ref * VCO, that is N / (P * R^2).
+	 */
+	a = freq2k * budget * p * r2;
+	b = freq2k * budget * best->p * best->r2;
+	diff = abs_diff(freq2k * p * r2, LC_FREQ_2K * n2);
+	diff_best = abs_diff(freq2k * best->p * best->r2,
+			     LC_FREQ_2K * best->n2);
+	c = 1000000 * diff;
+	d = 1000000 * diff_best;
+
+	if (a < c && b < d) {
+		/* If both are above the budget, pick the closer */
+		if (best->p * best->r2 * diff < p * r2 * diff_best) {
+			best->p = p;
+			best->n2 = n2;
+			best->r2 = r2;
+		}
+	} else if (a >= c && b < d) {
+		/* If A is below the threshold but B is above it?  Update. */
+		best->p = p;
+		best->n2 = n2;
+		best->r2 = r2;
+	} else if (a >= c && b >= d) {
+		/* Both are below the limit, so pick the higher n2/(r2*r2) */
+		if (n2 * best->r2 * best->r2 > best->n2 * r2 * r2) {
+			best->p = p;
+			best->n2 = n2;
+			best->r2 = r2;
+		}
+	}
+	/* Otherwise a < c && b >= d, do nothing */
+}
+
+static void
+hsw_ddi_calculate_wrpll(int clock /* in Hz */,
+			unsigned *r2_out, unsigned *n2_out, unsigned *p_out)
+{
+	uint64_t freq2k;
+	unsigned p, n2, r2;
+	struct hsw_wrpll_rnp best = { 0, 0, 0 };
+	unsigned budget;
+
+	freq2k = clock / 100;
+
+	budget = hsw_wrpll_get_budget_for_freq(clock);
+
+	/* Special case handling for 540 pixel clock: bypass WR PLL entirely
+	 * and directly pass the LC PLL to it. */
+	if (freq2k == 5400000) {
+		*n2_out = 2;
+		*p_out = 1;
+		*r2_out = 2;
+		return;
+	}
+
+	/*
+	 * Ref = LC_FREQ / R, where Ref is the actual reference input seen by
+	 * the WR PLL.
+	 *
+	 * We want R so that REF_MIN <= Ref <= REF_MAX.
+	 * Injecting R2 = 2 * R gives:
+	 *   REF_MAX * r2 > LC_FREQ * 2 and
+	 *   REF_MIN * r2 < LC_FREQ * 2
+	 *
+	 * Which means the desired boundaries for r2 are:
+	 *  LC_FREQ * 2 / REF_MAX < r2 < LC_FREQ * 2 / REF_MIN
+	 *
+	 */
+	for (r2 = LC_FREQ * 2 / REF_MAX + 1;
+	     r2 <= LC_FREQ * 2 / REF_MIN;
+	     r2++) {
+
+		/*
+		 * VCO = N * Ref, that is: VCO = N * LC_FREQ / R
+		 *
+		 * Once again we want VCO_MIN <= VCO <= VCO_MAX.
+		 * Injecting R2 = 2 * R and N2 = 2 * N, we get:
+		 *   VCO_MAX * r2 > n2 * LC_FREQ and
+		 *   VCO_MIN * r2 < n2 * LC_FREQ)
+		 *
+		 * Which means the desired boundaries for n2 are:
+		 * VCO_MIN * r2 / LC_FREQ < n2 < VCO_MAX * r2 / LC_FREQ
+		 */
+		for (n2 = VCO_MIN * r2 / LC_FREQ + 1;
+		     n2 <= VCO_MAX * r2 / LC_FREQ;
+		     n2++) {
+
+			for (p = P_MIN; p <= P_MAX; p += P_INC)
+				hsw_wrpll_update_rnp(freq2k, budget,
+						     r2, n2, p, &best);
+		}
+	}
+
+	*n2_out = best.n2;
+	*p_out = best.p;
+	*r2_out = best.r2;
+}
+
+static struct intel_shared_dpll *
+hsw_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+	     struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_shared_dpll *pll;
+	int clock = crtc_state->port_clock;
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (encoder->type == INTEL_OUTPUT_HDMI) {
+		uint32_t val;
+		unsigned p, n2, r2;
+
+		hsw_ddi_calculate_wrpll(clock * 1000, &r2, &n2, &p);
+
+		val = WRPLL_PLL_ENABLE | WRPLL_PLL_LCPLL |
+		      WRPLL_DIVIDER_REFERENCE(r2) | WRPLL_DIVIDER_FEEDBACK(n2) |
+		      WRPLL_DIVIDER_POST(p);
+
+		crtc_state->dpll_hw_state.wrpll = val;
+
+		pll = intel_find_shared_dpll(crtc, crtc_state,
+					     DPLL_ID_WRPLL1, DPLL_ID_WRPLL2);
+
+	} else if (encoder->type == INTEL_OUTPUT_DISPLAYPORT ||
+		   encoder->type == INTEL_OUTPUT_DP_MST ||
+		   encoder->type == INTEL_OUTPUT_EDP) {
+		enum intel_dpll_id pll_id;
+
+		switch (clock / 2) {
+		case 81000:
+			pll_id = DPLL_ID_LCPLL_810;
+			break;
+		case 135000:
+			pll_id = DPLL_ID_LCPLL_1350;
+			break;
+		case 270000:
+			pll_id = DPLL_ID_LCPLL_2700;
+			break;
+		default:
+			DRM_DEBUG_KMS("Invalid clock for DP: %d\n", clock);
+			return NULL;
+		}
+
+		pll = intel_get_shared_dpll_by_id(dev_priv, pll_id);
+
+	} else if (encoder->type == INTEL_OUTPUT_ANALOG) {
+		if (WARN_ON(crtc_state->port_clock / 2 != 135000))
+			return NULL;
+
+		crtc_state->dpll_hw_state.spll =
+			SPLL_PLL_ENABLE | SPLL_PLL_FREQ_1350MHz | SPLL_PLL_SSC;
+
+		pll = intel_find_shared_dpll(crtc, crtc_state,
+					     DPLL_ID_SPLL, DPLL_ID_SPLL);
+	} else {
+		return NULL;
+	}
+
+	if (!pll)
+		return NULL;
+
+	crtc_state->ddi_pll_sel = hsw_pll_to_ddi_pll_sel(pll);
+
+	intel_reference_shared_dpll(pll, crtc_state);
+
+	return pll;
+}
+
+
+static const struct intel_shared_dpll_funcs hsw_ddi_wrpll_funcs = {
+	.enable = hsw_ddi_wrpll_enable,
+	.disable = hsw_ddi_wrpll_disable,
+	.get_hw_state = hsw_ddi_wrpll_get_hw_state,
+};
+
+static const struct intel_shared_dpll_funcs hsw_ddi_spll_funcs = {
+	.enable = hsw_ddi_spll_enable,
+	.disable = hsw_ddi_spll_disable,
+	.get_hw_state = hsw_ddi_spll_get_hw_state,
+};
+
+static void hsw_ddi_lcpll_enable(struct drm_i915_private *dev_priv,
+				 struct intel_shared_dpll *pll)
+{
+}
+
+static void hsw_ddi_lcpll_disable(struct drm_i915_private *dev_priv,
+				  struct intel_shared_dpll *pll)
+{
+}
+
+static bool hsw_ddi_lcpll_get_hw_state(struct drm_i915_private *dev_priv,
+				       struct intel_shared_dpll *pll,
+				       struct intel_dpll_hw_state *hw_state)
+{
+	return true;
+}
+
+static const struct intel_shared_dpll_funcs hsw_ddi_lcpll_funcs = {
+	.enable = hsw_ddi_lcpll_enable,
+	.disable = hsw_ddi_lcpll_disable,
+	.get_hw_state = hsw_ddi_lcpll_get_hw_state,
+};
+
+struct skl_dpll_regs {
+	i915_reg_t ctl, cfgcr1, cfgcr2;
+};
+
+/* this array is indexed by the *shared* pll id */
+static const struct skl_dpll_regs skl_dpll_regs[4] = {
+	{
+		/* DPLL 0 */
+		.ctl = LCPLL1_CTL,
+		/* DPLL 0 doesn't support HDMI mode */
+	},
+	{
+		/* DPLL 1 */
+		.ctl = LCPLL2_CTL,
+		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL1),
+		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL1),
+	},
+	{
+		/* DPLL 2 */
+		.ctl = WRPLL_CTL(0),
+		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL2),
+		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL2),
+	},
+	{
+		/* DPLL 3 */
+		.ctl = WRPLL_CTL(1),
+		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL3),
+		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL3),
+	},
+};
+
+static void skl_ddi_pll_write_ctrl1(struct drm_i915_private *dev_priv,
+				    struct intel_shared_dpll *pll)
+{
+	uint32_t val;
+
+	val = I915_READ(DPLL_CTRL1);
+
+	val &= ~(DPLL_CTRL1_HDMI_MODE(pll->id) | DPLL_CTRL1_SSC(pll->id) |
+		 DPLL_CTRL1_LINK_RATE_MASK(pll->id));
+	val |= pll->config.hw_state.ctrl1 << (pll->id * 6);
+
+	I915_WRITE(DPLL_CTRL1, val);
+	POSTING_READ(DPLL_CTRL1);
+}
+
+static void skl_ddi_pll_enable(struct drm_i915_private *dev_priv,
+			       struct intel_shared_dpll *pll)
+{
+	const struct skl_dpll_regs *regs = skl_dpll_regs;
+
+	skl_ddi_pll_write_ctrl1(dev_priv, pll);
+
+	I915_WRITE(regs[pll->id].cfgcr1, pll->config.hw_state.cfgcr1);
+	I915_WRITE(regs[pll->id].cfgcr2, pll->config.hw_state.cfgcr2);
+	POSTING_READ(regs[pll->id].cfgcr1);
+	POSTING_READ(regs[pll->id].cfgcr2);
+
+	/* the enable bit is always bit 31 */
+	I915_WRITE(regs[pll->id].ctl,
+		   I915_READ(regs[pll->id].ctl) | LCPLL_PLL_ENABLE);
+
+	if (wait_for(I915_READ(DPLL_STATUS) & DPLL_LOCK(pll->id), 5))
+		DRM_ERROR("DPLL %d not locked\n", pll->id);
+}
+
+static void skl_ddi_dpll0_enable(struct drm_i915_private *dev_priv,
+				 struct intel_shared_dpll *pll)
+{
+	skl_ddi_pll_write_ctrl1(dev_priv, pll);
+}
+
+static void skl_ddi_pll_disable(struct drm_i915_private *dev_priv,
+				struct intel_shared_dpll *pll)
+{
+	const struct skl_dpll_regs *regs = skl_dpll_regs;
+
+	/* the enable bit is always bit 31 */
+	I915_WRITE(regs[pll->id].ctl,
+		   I915_READ(regs[pll->id].ctl) & ~LCPLL_PLL_ENABLE);
+	POSTING_READ(regs[pll->id].ctl);
+}
+
+static void skl_ddi_dpll0_disable(struct drm_i915_private *dev_priv,
+				  struct intel_shared_dpll *pll)
+{
+}
+
+static bool skl_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
+				     struct intel_shared_dpll *pll,
+				     struct intel_dpll_hw_state *hw_state)
+{
+	uint32_t val;
+	const struct skl_dpll_regs *regs = skl_dpll_regs;
+	bool ret;
+
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	ret = false;
+
+	val = I915_READ(regs[pll->id].ctl);
+	if (!(val & LCPLL_PLL_ENABLE))
+		goto out;
+
+	val = I915_READ(DPLL_CTRL1);
+	hw_state->ctrl1 = (val >> (pll->id * 6)) & 0x3f;
+
+	/* avoid reading back stale values if HDMI mode is not enabled */
+	if (val & DPLL_CTRL1_HDMI_MODE(pll->id)) {
+		hw_state->cfgcr1 = I915_READ(regs[pll->id].cfgcr1);
+		hw_state->cfgcr2 = I915_READ(regs[pll->id].cfgcr2);
+	}
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return ret;
+}
+
+static bool skl_ddi_dpll0_get_hw_state(struct drm_i915_private *dev_priv,
+				       struct intel_shared_dpll *pll,
+				       struct intel_dpll_hw_state *hw_state)
+{
+	uint32_t val;
+	const struct skl_dpll_regs *regs = skl_dpll_regs;
+	bool ret;
+
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	ret = false;
+
+	/* DPLL0 is always enabled since it drives CDCLK */
+	val = I915_READ(regs[pll->id].ctl);
+	if (WARN_ON(!(val & LCPLL_PLL_ENABLE)))
+		goto out;
+
+	val = I915_READ(DPLL_CTRL1);
+	hw_state->ctrl1 = (val >> (pll->id * 6)) & 0x3f;
+
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return ret;
+}
+
+struct skl_wrpll_context {
+	uint64_t min_deviation;		/* current minimal deviation */
+	uint64_t central_freq;		/* chosen central freq */
+	uint64_t dco_freq;		/* chosen dco freq */
+	unsigned int p;			/* chosen divider */
+};
+
+static void skl_wrpll_context_init(struct skl_wrpll_context *ctx)
+{
+	memset(ctx, 0, sizeof(*ctx));
+
+	ctx->min_deviation = U64_MAX;
+}
+
+/* DCO freq must be within +1%/-6%  of the DCO central freq */
+#define SKL_DCO_MAX_PDEVIATION	100
+#define SKL_DCO_MAX_NDEVIATION	600
+
+static void skl_wrpll_try_divider(struct skl_wrpll_context *ctx,
+				  uint64_t central_freq,
+				  uint64_t dco_freq,
+				  unsigned int divider)
+{
+	uint64_t deviation;
+
+	deviation = div64_u64(10000 * abs_diff(dco_freq, central_freq),
+			      central_freq);
+
+	/* positive deviation */
+	if (dco_freq >= central_freq) {
+		if (deviation < SKL_DCO_MAX_PDEVIATION &&
+		    deviation < ctx->min_deviation) {
+			ctx->min_deviation = deviation;
+			ctx->central_freq = central_freq;
+			ctx->dco_freq = dco_freq;
+			ctx->p = divider;
+		}
+	/* negative deviation */
+	} else if (deviation < SKL_DCO_MAX_NDEVIATION &&
+		   deviation < ctx->min_deviation) {
+		ctx->min_deviation = deviation;
+		ctx->central_freq = central_freq;
+		ctx->dco_freq = dco_freq;
+		ctx->p = divider;
+	}
+}
+
+static void skl_wrpll_get_multipliers(unsigned int p,
+				      unsigned int *p0 /* out */,
+				      unsigned int *p1 /* out */,
+				      unsigned int *p2 /* out */)
+{
+	/* even dividers */
+	if (p % 2 == 0) {
+		unsigned int half = p / 2;
+
+		if (half == 1 || half == 2 || half == 3 || half == 5) {
+			*p0 = 2;
+			*p1 = 1;
+			*p2 = half;
+		} else if (half % 2 == 0) {
+			*p0 = 2;
+			*p1 = half / 2;
+			*p2 = 2;
+		} else if (half % 3 == 0) {
+			*p0 = 3;
+			*p1 = half / 3;
+			*p2 = 2;
+		} else if (half % 7 == 0) {
+			*p0 = 7;
+			*p1 = half / 7;
+			*p2 = 2;
+		}
+	} else if (p == 3 || p == 9) {  /* 3, 5, 7, 9, 15, 21, 35 */
+		*p0 = 3;
+		*p1 = 1;
+		*p2 = p / 3;
+	} else if (p == 5 || p == 7) {
+		*p0 = p;
+		*p1 = 1;
+		*p2 = 1;
+	} else if (p == 15) {
+		*p0 = 3;
+		*p1 = 1;
+		*p2 = 5;
+	} else if (p == 21) {
+		*p0 = 7;
+		*p1 = 1;
+		*p2 = 3;
+	} else if (p == 35) {
+		*p0 = 7;
+		*p1 = 1;
+		*p2 = 5;
+	}
+}
+
+struct skl_wrpll_params {
+	uint32_t        dco_fraction;
+	uint32_t        dco_integer;
+	uint32_t        qdiv_ratio;
+	uint32_t        qdiv_mode;
+	uint32_t        kdiv;
+	uint32_t        pdiv;
+	uint32_t        central_freq;
+};
+
+static void skl_wrpll_params_populate(struct skl_wrpll_params *params,
+				      uint64_t afe_clock,
+				      uint64_t central_freq,
+				      uint32_t p0, uint32_t p1, uint32_t p2)
+{
+	uint64_t dco_freq;
+
+	switch (central_freq) {
+	case 9600000000ULL:
+		params->central_freq = 0;
+		break;
+	case 9000000000ULL:
+		params->central_freq = 1;
+		break;
+	case 8400000000ULL:
+		params->central_freq = 3;
+	}
+
+	switch (p0) {
+	case 1:
+		params->pdiv = 0;
+		break;
+	case 2:
+		params->pdiv = 1;
+		break;
+	case 3:
+		params->pdiv = 2;
+		break;
+	case 7:
+		params->pdiv = 4;
+		break;
+	default:
+		WARN(1, "Incorrect PDiv\n");
+	}
+
+	switch (p2) {
+	case 5:
+		params->kdiv = 0;
+		break;
+	case 2:
+		params->kdiv = 1;
+		break;
+	case 3:
+		params->kdiv = 2;
+		break;
+	case 1:
+		params->kdiv = 3;
+		break;
+	default:
+		WARN(1, "Incorrect KDiv\n");
+	}
+
+	params->qdiv_ratio = p1;
+	params->qdiv_mode = (params->qdiv_ratio == 1) ? 0 : 1;
+
+	dco_freq = p0 * p1 * p2 * afe_clock;
+
+	/*
+	 * Intermediate values are in Hz.
+	 * Divide by MHz to match bsepc
+	 */
+	params->dco_integer = div_u64(dco_freq, 24 * MHz(1));
+	params->dco_fraction =
+		div_u64((div_u64(dco_freq, 24) -
+			 params->dco_integer * MHz(1)) * 0x8000, MHz(1));
+}
+
+static bool
+skl_ddi_calculate_wrpll(int clock /* in Hz */,
+			struct skl_wrpll_params *wrpll_params)
+{
+	uint64_t afe_clock = clock * 5; /* AFE Clock is 5x Pixel clock */
+	uint64_t dco_central_freq[3] = {8400000000ULL,
+					9000000000ULL,
+					9600000000ULL};
+	static const int even_dividers[] = {  4,  6,  8, 10, 12, 14, 16, 18, 20,
+					     24, 28, 30, 32, 36, 40, 42, 44,
+					     48, 52, 54, 56, 60, 64, 66, 68,
+					     70, 72, 76, 78, 80, 84, 88, 90,
+					     92, 96, 98 };
+	static const int odd_dividers[] = { 3, 5, 7, 9, 15, 21, 35 };
+	static const struct {
+		const int *list;
+		int n_dividers;
+	} dividers[] = {
+		{ even_dividers, ARRAY_SIZE(even_dividers) },
+		{ odd_dividers, ARRAY_SIZE(odd_dividers) },
+	};
+	struct skl_wrpll_context ctx;
+	unsigned int dco, d, i;
+	unsigned int p0, p1, p2;
+
+	skl_wrpll_context_init(&ctx);
+
+	for (d = 0; d < ARRAY_SIZE(dividers); d++) {
+		for (dco = 0; dco < ARRAY_SIZE(dco_central_freq); dco++) {
+			for (i = 0; i < dividers[d].n_dividers; i++) {
+				unsigned int p = dividers[d].list[i];
+				uint64_t dco_freq = p * afe_clock;
+
+				skl_wrpll_try_divider(&ctx,
+						      dco_central_freq[dco],
+						      dco_freq,
+						      p);
+				/*
+				 * Skip the remaining dividers if we're sure to
+				 * have found the definitive divider, we can't
+				 * improve a 0 deviation.
+				 */
+				if (ctx.min_deviation == 0)
+					goto skip_remaining_dividers;
+			}
+		}
+
+skip_remaining_dividers:
+		/*
+		 * If a solution is found with an even divider, prefer
+		 * this one.
+		 */
+		if (d == 0 && ctx.p)
+			break;
+	}
+
+	if (!ctx.p) {
+		DRM_DEBUG_DRIVER("No valid divider found for %dHz\n", clock);
+		return false;
+	}
+
+	/*
+	 * gcc incorrectly analyses that these can be used without being
+	 * initialized. To be fair, it's hard to guess.
+	 */
+	p0 = p1 = p2 = 0;
+	skl_wrpll_get_multipliers(ctx.p, &p0, &p1, &p2);
+	skl_wrpll_params_populate(wrpll_params, afe_clock, ctx.central_freq,
+				  p0, p1, p2);
+
+	return true;
+}
+
+static struct intel_shared_dpll *
+skl_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+	     struct intel_encoder *encoder)
+{
+	struct intel_shared_dpll *pll;
+	uint32_t ctrl1, cfgcr1, cfgcr2;
+	int clock = crtc_state->port_clock;
+
+	/*
+	 * See comment in intel_dpll_hw_state to understand why we always use 0
+	 * as the DPLL id in this function.
+	 */
+
+	ctrl1 = DPLL_CTRL1_OVERRIDE(0);
+
+	if (encoder->type == INTEL_OUTPUT_HDMI) {
+		struct skl_wrpll_params wrpll_params = { 0, };
+
+		ctrl1 |= DPLL_CTRL1_HDMI_MODE(0);
+
+		if (!skl_ddi_calculate_wrpll(clock * 1000, &wrpll_params))
+			return NULL;
+
+		cfgcr1 = DPLL_CFGCR1_FREQ_ENABLE |
+			 DPLL_CFGCR1_DCO_FRACTION(wrpll_params.dco_fraction) |
+			 wrpll_params.dco_integer;
+
+		cfgcr2 = DPLL_CFGCR2_QDIV_RATIO(wrpll_params.qdiv_ratio) |
+			 DPLL_CFGCR2_QDIV_MODE(wrpll_params.qdiv_mode) |
+			 DPLL_CFGCR2_KDIV(wrpll_params.kdiv) |
+			 DPLL_CFGCR2_PDIV(wrpll_params.pdiv) |
+			 wrpll_params.central_freq;
+	} else if (encoder->type == INTEL_OUTPUT_DISPLAYPORT ||
+		   encoder->type == INTEL_OUTPUT_DP_MST ||
+		   encoder->type == INTEL_OUTPUT_EDP) {
+		switch (crtc_state->port_clock / 2) {
+		case 81000:
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810, 0);
+			break;
+		case 135000:
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1350, 0);
+			break;
+		case 270000:
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2700, 0);
+			break;
+		/* eDP 1.4 rates */
+		case 162000:
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1620, 0);
+			break;
+		/* TBD: For DP link rates 2.16 GHz and 4.32 GHz, VCO is 8640 which
+		results in CDCLK change. Need to handle the change of CDCLK by
+		disabling pipes and re-enabling them */
+		case 108000:
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1080, 0);
+			break;
+		case 216000:
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2160, 0);
+			break;
+		}
+
+		cfgcr1 = cfgcr2 = 0;
+	} else {
+		return NULL;
+	}
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	crtc_state->dpll_hw_state.ctrl1 = ctrl1;
+	crtc_state->dpll_hw_state.cfgcr1 = cfgcr1;
+	crtc_state->dpll_hw_state.cfgcr2 = cfgcr2;
+
+	if (encoder->type == INTEL_OUTPUT_EDP)
+		pll = intel_find_shared_dpll(crtc, crtc_state,
+					     DPLL_ID_SKL_DPLL0,
+					     DPLL_ID_SKL_DPLL0);
+	else
+		pll = intel_find_shared_dpll(crtc, crtc_state,
+					     DPLL_ID_SKL_DPLL1,
+					     DPLL_ID_SKL_DPLL3);
+	if (!pll)
+		return NULL;
+
+	crtc_state->ddi_pll_sel = pll->id;
+
+	intel_reference_shared_dpll(pll, crtc_state);
+
+	return pll;
+}
+
+static const struct intel_shared_dpll_funcs skl_ddi_pll_funcs = {
+	.enable = skl_ddi_pll_enable,
+	.disable = skl_ddi_pll_disable,
+	.get_hw_state = skl_ddi_pll_get_hw_state,
+};
+
+static const struct intel_shared_dpll_funcs skl_ddi_dpll0_funcs = {
+	.enable = skl_ddi_dpll0_enable,
+	.disable = skl_ddi_dpll0_disable,
+	.get_hw_state = skl_ddi_dpll0_get_hw_state,
+};
+
+static void bxt_ddi_pll_enable(struct drm_i915_private *dev_priv,
+				struct intel_shared_dpll *pll)
+{
+	uint32_t temp;
+	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
+
+	/* Non-SSC reference */
+	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	temp |= PORT_PLL_REF_SEL;
+	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
+
+	/* Disable 10 bit clock */
+	temp = I915_READ(BXT_PORT_PLL_EBB_4(port));
+	temp &= ~PORT_PLL_10BIT_CLK_ENABLE;
+	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
+
+	/* Write P1 & P2 */
+	temp = I915_READ(BXT_PORT_PLL_EBB_0(port));
+	temp &= ~(PORT_PLL_P1_MASK | PORT_PLL_P2_MASK);
+	temp |= pll->config.hw_state.ebb0;
+	I915_WRITE(BXT_PORT_PLL_EBB_0(port), temp);
+
+	/* Write M2 integer */
+	temp = I915_READ(BXT_PORT_PLL(port, 0));
+	temp &= ~PORT_PLL_M2_MASK;
+	temp |= pll->config.hw_state.pll0;
+	I915_WRITE(BXT_PORT_PLL(port, 0), temp);
+
+	/* Write N */
+	temp = I915_READ(BXT_PORT_PLL(port, 1));
+	temp &= ~PORT_PLL_N_MASK;
+	temp |= pll->config.hw_state.pll1;
+	I915_WRITE(BXT_PORT_PLL(port, 1), temp);
+
+	/* Write M2 fraction */
+	temp = I915_READ(BXT_PORT_PLL(port, 2));
+	temp &= ~PORT_PLL_M2_FRAC_MASK;
+	temp |= pll->config.hw_state.pll2;
+	I915_WRITE(BXT_PORT_PLL(port, 2), temp);
+
+	/* Write M2 fraction enable */
+	temp = I915_READ(BXT_PORT_PLL(port, 3));
+	temp &= ~PORT_PLL_M2_FRAC_ENABLE;
+	temp |= pll->config.hw_state.pll3;
+	I915_WRITE(BXT_PORT_PLL(port, 3), temp);
+
+	/* Write coeff */
+	temp = I915_READ(BXT_PORT_PLL(port, 6));
+	temp &= ~PORT_PLL_PROP_COEFF_MASK;
+	temp &= ~PORT_PLL_INT_COEFF_MASK;
+	temp &= ~PORT_PLL_GAIN_CTL_MASK;
+	temp |= pll->config.hw_state.pll6;
+	I915_WRITE(BXT_PORT_PLL(port, 6), temp);
+
+	/* Write calibration val */
+	temp = I915_READ(BXT_PORT_PLL(port, 8));
+	temp &= ~PORT_PLL_TARGET_CNT_MASK;
+	temp |= pll->config.hw_state.pll8;
+	I915_WRITE(BXT_PORT_PLL(port, 8), temp);
+
+	temp = I915_READ(BXT_PORT_PLL(port, 9));
+	temp &= ~PORT_PLL_LOCK_THRESHOLD_MASK;
+	temp |= pll->config.hw_state.pll9;
+	I915_WRITE(BXT_PORT_PLL(port, 9), temp);
+
+	temp = I915_READ(BXT_PORT_PLL(port, 10));
+	temp &= ~PORT_PLL_DCO_AMP_OVR_EN_H;
+	temp &= ~PORT_PLL_DCO_AMP_MASK;
+	temp |= pll->config.hw_state.pll10;
+	I915_WRITE(BXT_PORT_PLL(port, 10), temp);
+
+	/* Recalibrate with new settings */
+	temp = I915_READ(BXT_PORT_PLL_EBB_4(port));
+	temp |= PORT_PLL_RECALIBRATE;
+	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
+	temp &= ~PORT_PLL_10BIT_CLK_ENABLE;
+	temp |= pll->config.hw_state.ebb4;
+	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
+
+	/* Enable PLL */
+	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	temp |= PORT_PLL_ENABLE;
+	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
+	POSTING_READ(BXT_PORT_PLL_ENABLE(port));
+
+	if (wait_for_atomic_us((I915_READ(BXT_PORT_PLL_ENABLE(port)) &
+			PORT_PLL_LOCK), 200))
+		DRM_ERROR("PLL %d not locked\n", port);
+
+	/*
+	 * While we write to the group register to program all lanes at once we
+	 * can read only lane registers and we pick lanes 0/1 for that.
+	 */
+	temp = I915_READ(BXT_PORT_PCS_DW12_LN01(port));
+	temp &= ~LANE_STAGGER_MASK;
+	temp &= ~LANESTAGGER_STRAP_OVRD;
+	temp |= pll->config.hw_state.pcsdw12;
+	I915_WRITE(BXT_PORT_PCS_DW12_GRP(port), temp);
+}
+
+static void bxt_ddi_pll_disable(struct drm_i915_private *dev_priv,
+					struct intel_shared_dpll *pll)
+{
+	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
+	uint32_t temp;
+
+	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	temp &= ~PORT_PLL_ENABLE;
+	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
+	POSTING_READ(BXT_PORT_PLL_ENABLE(port));
+}
+
+static bool bxt_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
+					struct intel_shared_dpll *pll,
+					struct intel_dpll_hw_state *hw_state)
+{
+	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
+	uint32_t val;
+	bool ret;
+
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	ret = false;
+
+	val = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	if (!(val & PORT_PLL_ENABLE))
+		goto out;
+
+	hw_state->ebb0 = I915_READ(BXT_PORT_PLL_EBB_0(port));
+	hw_state->ebb0 &= PORT_PLL_P1_MASK | PORT_PLL_P2_MASK;
+
+	hw_state->ebb4 = I915_READ(BXT_PORT_PLL_EBB_4(port));
+	hw_state->ebb4 &= PORT_PLL_10BIT_CLK_ENABLE;
+
+	hw_state->pll0 = I915_READ(BXT_PORT_PLL(port, 0));
+	hw_state->pll0 &= PORT_PLL_M2_MASK;
+
+	hw_state->pll1 = I915_READ(BXT_PORT_PLL(port, 1));
+	hw_state->pll1 &= PORT_PLL_N_MASK;
+
+	hw_state->pll2 = I915_READ(BXT_PORT_PLL(port, 2));
+	hw_state->pll2 &= PORT_PLL_M2_FRAC_MASK;
+
+	hw_state->pll3 = I915_READ(BXT_PORT_PLL(port, 3));
+	hw_state->pll3 &= PORT_PLL_M2_FRAC_ENABLE;
+
+	hw_state->pll6 = I915_READ(BXT_PORT_PLL(port, 6));
+	hw_state->pll6 &= PORT_PLL_PROP_COEFF_MASK |
+			  PORT_PLL_INT_COEFF_MASK |
+			  PORT_PLL_GAIN_CTL_MASK;
+
+	hw_state->pll8 = I915_READ(BXT_PORT_PLL(port, 8));
+	hw_state->pll8 &= PORT_PLL_TARGET_CNT_MASK;
+
+	hw_state->pll9 = I915_READ(BXT_PORT_PLL(port, 9));
+	hw_state->pll9 &= PORT_PLL_LOCK_THRESHOLD_MASK;
+
+	hw_state->pll10 = I915_READ(BXT_PORT_PLL(port, 10));
+	hw_state->pll10 &= PORT_PLL_DCO_AMP_OVR_EN_H |
+			   PORT_PLL_DCO_AMP_MASK;
+
+	/*
+	 * While we write to the group register to program all lanes at once we
+	 * can read only lane registers. We configure all lanes the same way, so
+	 * here just read out lanes 0/1 and output a note if lanes 2/3 differ.
+	 */
+	hw_state->pcsdw12 = I915_READ(BXT_PORT_PCS_DW12_LN01(port));
+	if (I915_READ(BXT_PORT_PCS_DW12_LN23(port)) != hw_state->pcsdw12)
+		DRM_DEBUG_DRIVER("lane stagger config different for lane 01 (%08x) and 23 (%08x)\n",
+				 hw_state->pcsdw12,
+				 I915_READ(BXT_PORT_PCS_DW12_LN23(port)));
+	hw_state->pcsdw12 &= LANE_STAGGER_MASK | LANESTAGGER_STRAP_OVRD;
+
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return ret;
+}
+
+/* bxt clock parameters */
+struct bxt_clk_div {
+	int clock;
+	uint32_t p1;
+	uint32_t p2;
+	uint32_t m2_int;
+	uint32_t m2_frac;
+	bool m2_frac_en;
+	uint32_t n;
+};
+
+/* pre-calculated values for DP linkrates */
+static const struct bxt_clk_div bxt_dp_clk_val[] = {
+	{162000, 4, 2, 32, 1677722, 1, 1},
+	{270000, 4, 1, 27,       0, 0, 1},
+	{540000, 2, 1, 27,       0, 0, 1},
+	{216000, 3, 2, 32, 1677722, 1, 1},
+	{243000, 4, 1, 24, 1258291, 1, 1},
+	{324000, 4, 1, 32, 1677722, 1, 1},
+	{432000, 3, 1, 32, 1677722, 1, 1}
+};
+
+static struct intel_shared_dpll *
+bxt_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+	     struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_shared_dpll *pll;
+	enum intel_dpll_id i;
+	struct intel_digital_port *intel_dig_port;
+	struct bxt_clk_div clk_div = {0};
+	int vco = 0;
+	uint32_t prop_coef, int_coef, gain_ctl, targ_cnt;
+	uint32_t lanestagger;
+	int clock = crtc_state->port_clock;
+
+	if (encoder->type == INTEL_OUTPUT_HDMI) {
+		intel_clock_t best_clock;
+
+		/* Calculate HDMI div */
+		/*
+		 * FIXME: tie the following calculation into
+		 * i9xx_crtc_compute_clock
+		 */
+		if (!bxt_find_best_dpll(crtc_state, clock, &best_clock)) {
+			DRM_DEBUG_DRIVER("no PLL dividers found for clock %d pipe %c\n",
+					 clock, pipe_name(crtc->pipe));
+			return NULL;
+		}
+
+		clk_div.p1 = best_clock.p1;
+		clk_div.p2 = best_clock.p2;
+		WARN_ON(best_clock.m1 != 2);
+		clk_div.n = best_clock.n;
+		clk_div.m2_int = best_clock.m2 >> 22;
+		clk_div.m2_frac = best_clock.m2 & ((1 << 22) - 1);
+		clk_div.m2_frac_en = clk_div.m2_frac != 0;
+
+		vco = best_clock.vco;
+	} else if (encoder->type == INTEL_OUTPUT_DISPLAYPORT ||
+		   encoder->type == INTEL_OUTPUT_EDP) {
+		int i;
+
+		clk_div = bxt_dp_clk_val[0];
+		for (i = 0; i < ARRAY_SIZE(bxt_dp_clk_val); ++i) {
+			if (bxt_dp_clk_val[i].clock == clock) {
+				clk_div = bxt_dp_clk_val[i];
+				break;
+			}
+		}
+		vco = clock * 10 / 2 * clk_div.p1 * clk_div.p2;
+	}
+
+	if (vco >= 6200000 && vco <= 6700000) {
+		prop_coef = 4;
+		int_coef = 9;
+		gain_ctl = 3;
+		targ_cnt = 8;
+	} else if ((vco > 5400000 && vco < 6200000) ||
+			(vco >= 4800000 && vco < 5400000)) {
+		prop_coef = 5;
+		int_coef = 11;
+		gain_ctl = 3;
+		targ_cnt = 9;
+	} else if (vco == 5400000) {
+		prop_coef = 3;
+		int_coef = 8;
+		gain_ctl = 1;
+		targ_cnt = 9;
+	} else {
+		DRM_ERROR("Invalid VCO\n");
+		return NULL;
+	}
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (clock > 270000)
+		lanestagger = 0x18;
+	else if (clock > 135000)
+		lanestagger = 0x0d;
+	else if (clock > 67000)
+		lanestagger = 0x07;
+	else if (clock > 33000)
+		lanestagger = 0x04;
+	else
+		lanestagger = 0x02;
+
+	crtc_state->dpll_hw_state.ebb0 =
+		PORT_PLL_P1(clk_div.p1) | PORT_PLL_P2(clk_div.p2);
+	crtc_state->dpll_hw_state.pll0 = clk_div.m2_int;
+	crtc_state->dpll_hw_state.pll1 = PORT_PLL_N(clk_div.n);
+	crtc_state->dpll_hw_state.pll2 = clk_div.m2_frac;
+
+	if (clk_div.m2_frac_en)
+		crtc_state->dpll_hw_state.pll3 =
+			PORT_PLL_M2_FRAC_ENABLE;
+
+	crtc_state->dpll_hw_state.pll6 =
+		prop_coef | PORT_PLL_INT_COEFF(int_coef);
+	crtc_state->dpll_hw_state.pll6 |=
+		PORT_PLL_GAIN_CTL(gain_ctl);
+
+	crtc_state->dpll_hw_state.pll8 = targ_cnt;
+
+	crtc_state->dpll_hw_state.pll9 = 5 << PORT_PLL_LOCK_THRESHOLD_SHIFT;
+
+	crtc_state->dpll_hw_state.pll10 =
+		PORT_PLL_DCO_AMP(PORT_PLL_DCO_AMP_DEFAULT)
+		| PORT_PLL_DCO_AMP_OVR_EN_H;
+
+	crtc_state->dpll_hw_state.ebb4 = PORT_PLL_10BIT_CLK_ENABLE;
+
+	crtc_state->dpll_hw_state.pcsdw12 =
+		LANESTAGGER_STRAP_OVRD | lanestagger;
+
+	intel_dig_port = enc_to_dig_port(&encoder->base);
+
+	/* 1:1 mapping between ports and PLLs */
+	i = (enum intel_dpll_id) intel_dig_port->port;
+	pll = intel_get_shared_dpll_by_id(dev_priv, i);
+
+	DRM_DEBUG_KMS("CRTC:%d using pre-allocated %s\n",
+		crtc->base.base.id, pll->name);
+
+	intel_reference_shared_dpll(pll, crtc_state);
+
+	/* shared DPLL id 0 is DPLL A */
+	crtc_state->ddi_pll_sel = pll->id;
+
+	return pll;
+}
+
+static const struct intel_shared_dpll_funcs bxt_ddi_pll_funcs = {
+	.enable = bxt_ddi_pll_enable,
+	.disable = bxt_ddi_pll_disable,
+	.get_hw_state = bxt_ddi_pll_get_hw_state,
+};
+
+static void intel_ddi_pll_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t val = I915_READ(LCPLL_CTL);
+
+	if (IS_SKYLAKE(dev) || IS_KABYLAKE(dev)) {
+		int cdclk_freq;
+
+		cdclk_freq = dev_priv->display.get_display_clock_speed(dev);
+		dev_priv->skl_boot_cdclk = cdclk_freq;
+		if (skl_sanitize_cdclk(dev_priv))
+			DRM_DEBUG_KMS("Sanitized cdclk programmed by pre-os\n");
+		if (!(I915_READ(LCPLL1_CTL) & LCPLL_PLL_ENABLE))
+			DRM_ERROR("LCPLL1 is disabled\n");
+	} else if (!IS_BROXTON(dev_priv)) {
+		/*
+		 * The LCPLL register should be turned on by the BIOS. For now
+		 * let's just check its state and print errors in case
+		 * something is wrong.  Don't even try to turn it on.
+		 */
+
+		if (val & LCPLL_CD_SOURCE_FCLK)
+			DRM_ERROR("CDCLK source is not LCPLL\n");
+
+		if (val & LCPLL_PLL_DISABLE)
+			DRM_ERROR("LCPLL is disabled\n");
+	}
+}
+
+struct dpll_info {
+	const char *name;
+	const int id;
+	const struct intel_shared_dpll_funcs *funcs;
+	uint32_t flags;
+};
+
+struct intel_dpll_mgr {
+	const struct dpll_info *dpll_info;
+
+	struct intel_shared_dpll *(*get_dpll)(struct intel_crtc *crtc,
+					      struct intel_crtc_state *crtc_state,
+					      struct intel_encoder *encoder);
+};
+
+static const struct dpll_info pch_plls[] = {
+	{ "PCH DPLL A", DPLL_ID_PCH_PLL_A, &ibx_pch_dpll_funcs, 0 },
+	{ "PCH DPLL B", DPLL_ID_PCH_PLL_B, &ibx_pch_dpll_funcs, 0 },
+	{ NULL, -1, NULL, 0 },
+};
+
+static const struct intel_dpll_mgr pch_pll_mgr = {
+	.dpll_info = pch_plls,
+	.get_dpll = ibx_get_dpll,
+};
+
+static const struct dpll_info hsw_plls[] = {
+	{ "WRPLL 1",    DPLL_ID_WRPLL1,     &hsw_ddi_wrpll_funcs, 0 },
+	{ "WRPLL 2",    DPLL_ID_WRPLL2,     &hsw_ddi_wrpll_funcs, 0 },
+	{ "SPLL",       DPLL_ID_SPLL,       &hsw_ddi_spll_funcs,  0 },
+	{ "LCPLL 810",  DPLL_ID_LCPLL_810,  &hsw_ddi_lcpll_funcs, INTEL_DPLL_ALWAYS_ON },
+	{ "LCPLL 1350", DPLL_ID_LCPLL_1350, &hsw_ddi_lcpll_funcs, INTEL_DPLL_ALWAYS_ON },
+	{ "LCPLL 2700", DPLL_ID_LCPLL_2700, &hsw_ddi_lcpll_funcs, INTEL_DPLL_ALWAYS_ON },
+	{ NULL, -1, NULL, },
+};
+
+static const struct intel_dpll_mgr hsw_pll_mgr = {
+	.dpll_info = hsw_plls,
+	.get_dpll = hsw_get_dpll,
+};
+
+static const struct dpll_info skl_plls[] = {
+	{ "DPLL 0", DPLL_ID_SKL_DPLL0, &skl_ddi_dpll0_funcs, INTEL_DPLL_ALWAYS_ON },
+	{ "DPPL 1", DPLL_ID_SKL_DPLL1, &skl_ddi_pll_funcs,   0 },
+	{ "DPPL 2", DPLL_ID_SKL_DPLL2, &skl_ddi_pll_funcs,   0 },
+	{ "DPPL 3", DPLL_ID_SKL_DPLL3, &skl_ddi_pll_funcs,   0 },
+	{ NULL, -1, NULL, },
+};
+
+static const struct intel_dpll_mgr skl_pll_mgr = {
+	.dpll_info = skl_plls,
+	.get_dpll = skl_get_dpll,
+};
+
+static const struct dpll_info bxt_plls[] = {
+	{ "PORT PLL A", DPLL_ID_SKL_DPLL0, &bxt_ddi_pll_funcs, 0 },
+	{ "PORT PLL B", DPLL_ID_SKL_DPLL1, &bxt_ddi_pll_funcs, 0 },
+	{ "PORT PLL C", DPLL_ID_SKL_DPLL2, &bxt_ddi_pll_funcs, 0 },
+	{ NULL, -1, NULL, },
+};
+
+static const struct intel_dpll_mgr bxt_pll_mgr = {
+	.dpll_info = bxt_plls,
+	.get_dpll = bxt_get_dpll,
+};
+
+void intel_shared_dpll_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	const struct intel_dpll_mgr *dpll_mgr = NULL;
+	const struct dpll_info *dpll_info;
+	int i;
+
+	if (IS_SKYLAKE(dev) || IS_KABYLAKE(dev))
+		dpll_mgr = &skl_pll_mgr;
+	else if (IS_BROXTON(dev))
+		dpll_mgr = &bxt_pll_mgr;
+	else if (HAS_DDI(dev))
+		dpll_mgr = &hsw_pll_mgr;
+	else if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
+		dpll_mgr = &pch_pll_mgr;
+
+	if (!dpll_mgr) {
+		dev_priv->num_shared_dpll = 0;
+		return;
+	}
+
+	dpll_info = dpll_mgr->dpll_info;
+
+	for (i = 0; dpll_info[i].id >= 0; i++) {
+		WARN_ON(i != dpll_info[i].id);
+
+		dev_priv->shared_dplls[i].id = dpll_info[i].id;
+		dev_priv->shared_dplls[i].name = dpll_info[i].name;
+		dev_priv->shared_dplls[i].funcs = *dpll_info[i].funcs;
+		dev_priv->shared_dplls[i].flags = dpll_info[i].flags;
+	}
+
+	dev_priv->dpll_mgr = dpll_mgr;
+	dev_priv->num_shared_dpll = i;
+	mutex_init(&dev_priv->dpll_lock);
+
+	BUG_ON(dev_priv->num_shared_dpll > I915_NUM_PLLS);
+
+	/* FIXME: Move this to a more suitable place */
+	if (HAS_DDI(dev))
+		intel_ddi_pll_init(dev);
+}
+
+struct intel_shared_dpll *
+intel_get_shared_dpll(struct intel_crtc *crtc,
+		      struct intel_crtc_state *crtc_state,
+		      struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll_mgr;
+
+	if (WARN_ON(!dpll_mgr))
+		return NULL;
+
+	return dpll_mgr->get_dpll(crtc, crtc_state, encoder);
+}
diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.h b/drivers/gpu/drm/i915/intel_dpll_mgr.h
new file mode 100644
index 000000000000..89c5ada1a315
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dpll_mgr.h
@@ -0,0 +1,164 @@
+/*
+ * Copyright © 2012-2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#ifndef _INTEL_DPLL_MGR_H_
+#define _INTEL_DPLL_MGR_H_
+
+/*FIXME: Move this to a more appropriate place. */
+#define abs_diff(a, b) ({			\
+	typeof(a) __a = (a);			\
+	typeof(b) __b = (b);			\
+	(void) (&__a == &__b);			\
+	__a > __b ? (__a - __b) : (__b - __a); })
+
+struct drm_i915_private;
+struct intel_crtc;
+struct intel_crtc_state;
+struct intel_encoder;
+
+struct intel_shared_dpll;
+struct intel_dpll_mgr;
+
+enum intel_dpll_id {
+	DPLL_ID_PRIVATE = -1, /* non-shared dpll in use */
+	/* real shared dpll ids must be >= 0 */
+	DPLL_ID_PCH_PLL_A = 0,
+	DPLL_ID_PCH_PLL_B = 1,
+	/* hsw/bdw */
+	DPLL_ID_WRPLL1 = 0,
+	DPLL_ID_WRPLL2 = 1,
+	DPLL_ID_SPLL = 2,
+	DPLL_ID_LCPLL_810 = 3,
+	DPLL_ID_LCPLL_1350 = 4,
+	DPLL_ID_LCPLL_2700 = 5,
+
+	/* skl */
+	DPLL_ID_SKL_DPLL0 = 0,
+	DPLL_ID_SKL_DPLL1 = 1,
+	DPLL_ID_SKL_DPLL2 = 2,
+	DPLL_ID_SKL_DPLL3 = 3,
+};
+#define I915_NUM_PLLS 6
+
+/** Inform the state checker that the DPLL is kept enabled even if not
+ * in use by any crtc.
+ */
+#define INTEL_DPLL_ALWAYS_ON	(1 << 0)
+
+struct intel_dpll_hw_state {
+	/* i9xx, pch plls */
+	uint32_t dpll;
+	uint32_t dpll_md;
+	uint32_t fp0;
+	uint32_t fp1;
+
+	/* hsw, bdw */
+	uint32_t wrpll;
+	uint32_t spll;
+
+	/* skl */
+	/*
+	 * DPLL_CTRL1 has 6 bits for each each this DPLL. We store those in
+	 * lower part of ctrl1 and they get shifted into position when writing
+	 * the register.  This allows us to easily compare the state to share
+	 * the DPLL.
+	 */
+	uint32_t ctrl1;
+	/* HDMI only, 0 when used for DP */
+	uint32_t cfgcr1, cfgcr2;
+
+	/* bxt */
+	uint32_t ebb0, ebb4, pll0, pll1, pll2, pll3, pll6, pll8, pll9, pll10,
+		 pcsdw12;
+};
+
+struct intel_shared_dpll_config {
+	unsigned crtc_mask; /* mask of CRTCs sharing this PLL */
+	struct intel_dpll_hw_state hw_state;
+};
+
+struct intel_shared_dpll_funcs {
+	/* The mode_set hook is optional and should be used together with the
+	 * intel_prepare_shared_dpll function. */
+	void (*mode_set)(struct drm_i915_private *dev_priv,
+			 struct intel_shared_dpll *pll);
+	void (*enable)(struct drm_i915_private *dev_priv,
+		       struct intel_shared_dpll *pll);
+	void (*disable)(struct drm_i915_private *dev_priv,
+			struct intel_shared_dpll *pll);
+	bool (*get_hw_state)(struct drm_i915_private *dev_priv,
+			     struct intel_shared_dpll *pll,
+			     struct intel_dpll_hw_state *hw_state);
+};
+
+struct intel_shared_dpll {
+	struct intel_shared_dpll_config config;
+
+	unsigned active_mask; /* mask of active CRTCs (i.e. DPMS on) */
+	bool on; /* is the PLL actually active? Disabled during modeset */
+	const char *name;
+	/* should match the index in the dev_priv->shared_dplls array */
+	enum intel_dpll_id id;
+
+	struct intel_shared_dpll_funcs funcs;
+
+	uint32_t flags;
+};
+
+#define SKL_DPLL0 0
+#define SKL_DPLL1 1
+#define SKL_DPLL2 2
+#define SKL_DPLL3 3
+
+/* shared dpll functions */
+struct intel_shared_dpll *
+intel_get_shared_dpll_by_id(struct drm_i915_private *dev_priv,
+			    enum intel_dpll_id id);
+enum intel_dpll_id
+intel_get_shared_dpll_id(struct drm_i915_private *dev_priv,
+			 struct intel_shared_dpll *pll);
+void
+intel_shared_dpll_config_get(struct intel_shared_dpll_config *config,
+			     struct intel_shared_dpll *pll,
+			     struct intel_crtc *crtc);
+void
+intel_shared_dpll_config_put(struct intel_shared_dpll_config *config,
+			     struct intel_shared_dpll *pll,
+			     struct intel_crtc *crtc);
+void assert_shared_dpll(struct drm_i915_private *dev_priv,
+			struct intel_shared_dpll *pll,
+			bool state);
+#define assert_shared_dpll_enabled(d, p) assert_shared_dpll(d, p, true)
+#define assert_shared_dpll_disabled(d, p) assert_shared_dpll(d, p, false)
+struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc,
+						struct intel_crtc_state *state,
+						struct intel_encoder *encoder);
+void intel_prepare_shared_dpll(struct intel_crtc *crtc);
+void intel_enable_shared_dpll(struct intel_crtc *crtc);
+void intel_disable_shared_dpll(struct intel_crtc *crtc);
+void intel_shared_dpll_commit(struct drm_atomic_state *state);
+void intel_shared_dpll_init(struct drm_device *dev);
+
+
+#endif /* _INTEL_DPLL_MGR_H_ */
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 9d0770c23fde..5da29a02b9e3 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -44,9 +44,13 @@
  * contexts. Note that it's important that we check the condition again after
  * having timed out, since the timeout could be due to preemption or similar and
  * we've never had a chance to check the condition before the timeout.
+ *
+ * TODO: When modesetting has fully transitioned to atomic, the below
+ * drm_can_sleep() can be removed and in_atomic()/!in_atomic() asserts
+ * added.
  */
-#define _wait_for(COND, MS, W) ({ \
-	unsigned long timeout__ = jiffies + msecs_to_jiffies(MS) + 1;	\
+#define _wait_for(COND, US, W) ({ \
+	unsigned long timeout__ = jiffies + usecs_to_jiffies(US) + 1;	\
 	int ret__ = 0;							\
 	while (!(COND)) {						\
 		if (time_after(jiffies, timeout__)) {			\
@@ -55,7 +59,7 @@
 			break;						\
 		}							\
 		if ((W) && drm_can_sleep()) {				\
-			usleep_range((W)*1000, (W)*2000);		\
+			usleep_range((W), (W)*2);			\
 		} else {						\
 			cpu_relax();					\
 		}							\
@@ -63,10 +67,40 @@
 	ret__;								\
 })
 
-#define wait_for(COND, MS) _wait_for(COND, MS, 1)
-#define wait_for_atomic(COND, MS) _wait_for(COND, MS, 0)
-#define wait_for_atomic_us(COND, US) _wait_for((COND), \
-					       DIV_ROUND_UP((US), 1000), 0)
+#define wait_for(COND, MS)	  	_wait_for((COND), (MS) * 1000, 1000)
+#define wait_for_us(COND, US)	  	_wait_for((COND), (US), 1)
+
+/* If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false. */
+#if defined(CONFIG_DRM_I915_DEBUG) && defined(CONFIG_PREEMPT_COUNT)
+# define _WAIT_FOR_ATOMIC_CHECK WARN_ON_ONCE(!in_atomic())
+#else
+# define _WAIT_FOR_ATOMIC_CHECK do { } while (0)
+#endif
+
+#define _wait_for_atomic(COND, US) ({ \
+	unsigned long end__; \
+	int ret__ = 0; \
+	_WAIT_FOR_ATOMIC_CHECK; \
+	BUILD_BUG_ON((US) > 50000); \
+	end__ = (local_clock() >> 10) + (US) + 1; \
+	while (!(COND)) { \
+		if (time_after((unsigned long)(local_clock() >> 10), end__)) { \
+			/* Unlike the regular wait_for(), this atomic variant \
+			 * cannot be preempted (and we'll just ignore the issue\
+			 * of irq interruptions) and so we know that no time \
+			 * has passed since the last check of COND and can \
+			 * immediately report the timeout. \
+			 */ \
+			ret__ = -ETIMEDOUT; \
+			break; \
+		} \
+		cpu_relax(); \
+	} \
+	ret__; \
+})
+
+#define wait_for_atomic(COND, MS)	_wait_for_atomic((COND), (MS) * 1000)
+#define wait_for_atomic_us(COND, US)	_wait_for_atomic((COND), (US))
 
 #define KHz(x) (1000 * (x))
 #define MHz(x) KHz(1000 * (x))
@@ -118,6 +152,7 @@ enum intel_output_type {
 struct intel_framebuffer {
 	struct drm_framebuffer base;
 	struct drm_i915_gem_object *obj;
+	struct intel_rotation_info rot_info;
 };
 
 struct intel_fbdev {
@@ -260,6 +295,12 @@ struct intel_atomic_state {
 
 	struct intel_shared_dpll_config shared_dpll[I915_NUM_PLLS];
 	struct intel_wm_config wm_config;
+
+	/*
+	 * Current watermarks can't be trusted during hardware readout, so
+	 * don't bother calculating intermediate watermarks.
+	 */
+	bool skip_intermediate_wm;
 };
 
 struct intel_plane_state {
@@ -349,6 +390,7 @@ struct intel_crtc_scaler_state {
 
 struct intel_pipe_wm {
 	struct intel_wm_level wm[5];
+	struct intel_wm_level raw_wm[5];
 	uint32_t linetime;
 	bool fbc_wm_enabled;
 	bool pipe_enabled;
@@ -376,9 +418,10 @@ struct intel_crtc_state {
 #define PIPE_CONFIG_QUIRK_MODE_SYNC_FLAGS	(1<<0) /* unreliable sync mode.flags */
 	unsigned long quirks;
 
+	unsigned fb_bits; /* framebuffers to flip */
 	bool update_pipe; /* can a fast modeset be performed? */
 	bool disable_cxsr;
-	bool wm_changed; /* watermarks are updated */
+	bool update_wm_pre, update_wm_post; /* watermarks are updated */
 	bool fb_changed; /* fb on any of the planes is changed */
 
 	/* Pipe source size (ie. panel fitter input size)
@@ -394,7 +437,8 @@ struct intel_crtc_state {
 	bool has_infoframe;
 
 	/* CPU Transcoder for the pipe. Currently this can only differ from the
-	 * pipe on Haswell (where we have a special eDP transcoder). */
+	 * pipe on Haswell and later (where we have a special eDP transcoder)
+	 * and Broxton (where we have special DSI transcoders). */
 	enum transcoder cpu_transcoder;
 
 	/*
@@ -441,8 +485,8 @@ struct intel_crtc_state {
 	 * haswell. */
 	struct dpll dpll;
 
-	/* Selected dpll when shared or DPLL_ID_PRIVATE. */
-	enum intel_dpll_id shared_dpll;
+	/* Selected dpll when shared or NULL. */
+	struct intel_shared_dpll *shared_dpll;
 
 	/*
 	 * - PORT_CLK_SEL for DDI ports on HSW/BDW.
@@ -453,6 +497,11 @@ struct intel_crtc_state {
 	/* Actual register state of the dpll, for shared dpll cross-checking. */
 	struct intel_dpll_hw_state dpll_hw_state;
 
+	/* DSI PLL registers */
+	struct {
+		u32 ctrl, div;
+	} dsi_pll;
+
 	int pipe_bpp;
 	struct intel_link_m_n dp_m_n;
 
@@ -510,14 +559,33 @@ struct intel_crtc_state {
 
 	struct {
 		/*
-		 * optimal watermarks, programmed post-vblank when this state
-		 * is committed
+		 * Optimal watermarks, programmed post-vblank when this state
+		 * is committed.
 		 */
 		union {
 			struct intel_pipe_wm ilk;
 			struct skl_pipe_wm skl;
 		} optimal;
+
+		/*
+		 * Intermediate watermarks; these can be programmed immediately
+		 * since they satisfy both the current configuration we're
+		 * switching away from and the new configuration we're switching
+		 * to.
+		 */
+		struct intel_pipe_wm intermediate;
+
+		/*
+		 * Platforms with two-step watermark programming will need to
+		 * update watermark programming post-vblank to switch from the
+		 * safe intermediate watermarks to the optimal final
+		 * watermarks.
+		 */
+		bool need_postvbl_update;
 	} wm;
+
+	/* Gamma mode programmed on the pipe */
+	uint32_t gamma_mode;
 };
 
 struct vlv_wm_state {
@@ -537,23 +605,6 @@ struct intel_mmio_flip {
 	unsigned int rotation;
 };
 
-/*
- * Tracking of operations that need to be performed at the beginning/end of an
- * atomic commit, outside the atomic section where interrupts are disabled.
- * These are generally operations that grab mutexes or might otherwise sleep
- * and thus can't be run with interrupts disabled.
- */
-struct intel_crtc_atomic_commit {
-	/* Sleepable operations to perform before commit */
-
-	/* Sleepable operations to perform after commit */
-	unsigned fb_bits;
-	bool post_enable_primary;
-
-	/* Sleepable operations to perform before and after commit */
-	bool update_fbc;
-};
-
 struct intel_crtc {
 	struct drm_crtc base;
 	enum pipe pipe;
@@ -600,6 +651,7 @@ struct intel_crtc {
 			struct intel_pipe_wm ilk;
 			struct skl_pipe_wm skl;
 		} active;
+
 		/* allow CxSR on this pipe */
 		bool cxsr_allowed;
 	} wm;
@@ -613,8 +665,6 @@ struct intel_crtc {
 		int scanline_start;
 	} debug;
 
-	struct intel_crtc_atomic_commit atomic;
-
 	/* scalers available on this crtc */
 	int num_scalers;
 
@@ -751,7 +801,9 @@ struct intel_dp {
 	uint32_t DP;
 	int link_rate;
 	uint8_t lane_count;
+	uint8_t sink_count;
 	bool has_audio;
+	bool detect_done;
 	enum hdmi_force_audio force_audio;
 	bool limited_color_range;
 	bool color_range_auto;
@@ -831,7 +883,7 @@ struct intel_dp_mst_encoder {
 	struct intel_encoder base;
 	enum pipe pipe;
 	struct intel_digital_port *primary;
-	void *port; /* store this opaque as its illegal to dereference it */
+	struct intel_connector *connector;
 };
 
 static inline enum dpio_channel
@@ -1007,7 +1059,6 @@ void hsw_fdi_link_train(struct drm_crtc *crtc);
 void intel_ddi_init(struct drm_device *dev, enum port port);
 enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder);
 bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe);
-void intel_ddi_pll_init(struct drm_device *dev);
 void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc);
 void intel_ddi_disable_transcoder_func(struct drm_i915_private *dev_priv,
 				       enum transcoder cpu_transcoder);
@@ -1049,17 +1100,19 @@ u32 intel_fb_stride_alignment(const struct drm_i915_private *dev_priv,
 			      uint64_t fb_modifier, uint32_t pixel_format);
 
 /* intel_audio.c */
-void intel_init_audio(struct drm_device *dev);
+void intel_init_audio_hooks(struct drm_i915_private *dev_priv);
 void intel_audio_codec_enable(struct intel_encoder *encoder);
 void intel_audio_codec_disable(struct intel_encoder *encoder);
 void i915_audio_component_init(struct drm_i915_private *dev_priv);
 void i915_audio_component_cleanup(struct drm_i915_private *dev_priv);
 
 /* intel_display.c */
+int vlv_get_cck_clock(struct drm_i915_private *dev_priv,
+		      const char *name, u32 reg, int ref_freq);
 extern const struct drm_plane_funcs intel_plane_funcs;
+void intel_init_display_hooks(struct drm_i915_private *dev_priv);
+unsigned int intel_rotation_info_size(const struct intel_rotation_info *rot_info);
 bool intel_has_pending_fb_unpin(struct drm_device *dev);
-int intel_pch_rawclk(struct drm_device *dev);
-int intel_hrawclk(struct drm_device *dev);
 void intel_mark_busy(struct drm_device *dev);
 void intel_mark_idle(struct drm_device *dev);
 void intel_crtc_restore_mode(struct drm_crtc *crtc);
@@ -1104,9 +1157,8 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 void intel_release_load_detect_pipe(struct drm_connector *connector,
 				    struct intel_load_detect_pipe *old,
 				    struct drm_modeset_acquire_ctx *ctx);
-int intel_pin_and_fence_fb_obj(struct drm_plane *plane,
-			       struct drm_framebuffer *fb,
-			       const struct drm_plane_state *plane_state);
+int intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb,
+			       unsigned int rotation);
 struct drm_framebuffer *
 __intel_framebuffer_create(struct drm_device *dev,
 			   struct drm_mode_fb_cmd2 *mode_cmd,
@@ -1142,19 +1194,13 @@ intel_rotation_90_or_270(unsigned int rotation)
 void intel_create_rotation_property(struct drm_device *dev,
 					struct intel_plane *plane);
 
-/* shared dpll functions */
-struct intel_shared_dpll *intel_crtc_to_shared_dpll(struct intel_crtc *crtc);
-void assert_shared_dpll(struct drm_i915_private *dev_priv,
-			struct intel_shared_dpll *pll,
-			bool state);
-#define assert_shared_dpll_enabled(d, p) assert_shared_dpll(d, p, true)
-#define assert_shared_dpll_disabled(d, p) assert_shared_dpll(d, p, false)
-struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc,
-						struct intel_crtc_state *state);
+void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv,
+				    enum pipe pipe);
 
 int vlv_force_pll_on(struct drm_device *dev, enum pipe pipe,
 		     const struct dpll *dpll);
 void vlv_force_pll_off(struct drm_device *dev, enum pipe pipe);
+int lpt_get_iclkip(struct drm_i915_private *dev_priv);
 
 /* modesetting asserts */
 void assert_panel_unlocked(struct drm_i915_private *dev_priv,
@@ -1163,6 +1209,9 @@ void assert_pll(struct drm_i915_private *dev_priv,
 		enum pipe pipe, bool state);
 #define assert_pll_enabled(d, p) assert_pll(d, p, true)
 #define assert_pll_disabled(d, p) assert_pll(d, p, false)
+void assert_dsi_pll(struct drm_i915_private *dev_priv, bool state);
+#define assert_dsi_pll_enabled(d) assert_dsi_pll(d, true)
+#define assert_dsi_pll_disabled(d) assert_dsi_pll(d, false)
 void assert_fdi_rx_pll(struct drm_i915_private *dev_priv,
 		       enum pipe pipe, bool state);
 #define assert_fdi_rx_pll_enabled(d, p) assert_fdi_rx_pll(d, p, true)
@@ -1170,21 +1219,24 @@ void assert_fdi_rx_pll(struct drm_i915_private *dev_priv,
 void assert_pipe(struct drm_i915_private *dev_priv, enum pipe pipe, bool state);
 #define assert_pipe_enabled(d, p) assert_pipe(d, p, true)
 #define assert_pipe_disabled(d, p) assert_pipe(d, p, false)
-u32 intel_compute_tile_offset(struct drm_i915_private *dev_priv,
-			      int *x, int *y,
-			      uint64_t fb_modifier,
-			      unsigned int cpp,
-			      unsigned int pitch);
+u32 intel_compute_tile_offset(int *x, int *y,
+			      const struct drm_framebuffer *fb, int plane,
+			      unsigned int pitch,
+			      unsigned int rotation);
 void intel_prepare_reset(struct drm_device *dev);
 void intel_finish_reset(struct drm_device *dev);
 void hsw_enable_pc8(struct drm_i915_private *dev_priv);
 void hsw_disable_pc8(struct drm_i915_private *dev_priv);
-void broxton_init_cdclk(struct drm_device *dev);
-void broxton_uninit_cdclk(struct drm_device *dev);
-void broxton_ddi_phy_init(struct drm_device *dev);
-void broxton_ddi_phy_uninit(struct drm_device *dev);
+void broxton_init_cdclk(struct drm_i915_private *dev_priv);
+void broxton_uninit_cdclk(struct drm_i915_private *dev_priv);
+bool broxton_cdclk_verify_state(struct drm_i915_private *dev_priv);
+void broxton_ddi_phy_init(struct drm_i915_private *dev_priv);
+void broxton_ddi_phy_uninit(struct drm_i915_private *dev_priv);
+void broxton_ddi_phy_verify_state(struct drm_i915_private *dev_priv);
+void gen9_sanitize_dc_state(struct drm_i915_private *dev_priv);
 void bxt_enable_dc9(struct drm_i915_private *dev_priv);
 void bxt_disable_dc9(struct drm_i915_private *dev_priv);
+void gen9_enable_dc5(struct drm_i915_private *dev_priv);
 void skl_init_cdclk(struct drm_i915_private *dev_priv);
 int skl_sanitize_cdclk(struct drm_i915_private *dev_priv);
 void skl_uninit_cdclk(struct drm_i915_private *dev_priv);
@@ -1194,9 +1246,6 @@ void intel_dp_get_m_n(struct intel_crtc *crtc,
 		      struct intel_crtc_state *pipe_config);
 void intel_dp_set_m_n(struct intel_crtc *crtc, enum link_m_n_set m_n);
 int intel_dotclock_calculate(int link_freq, const struct intel_link_m_n *m_n);
-void
-ironlake_check_encoder_dotclock(const struct intel_crtc_state *pipe_config,
-				int dotclock);
 bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state, int target_clock,
 			intel_clock_t *best_clock);
 int chv_calc_dpll_params(int refclk, intel_clock_t *pll_clock);
@@ -1224,8 +1273,10 @@ u32 skl_plane_ctl_rotation(unsigned int rotation);
 
 /* intel_csr.c */
 void intel_csr_ucode_init(struct drm_i915_private *);
-bool intel_csr_load_program(struct drm_i915_private *);
+void intel_csr_load_program(struct drm_i915_private *);
 void intel_csr_ucode_fini(struct drm_i915_private *);
+void intel_csr_ucode_suspend(struct drm_i915_private *);
+void intel_csr_ucode_resume(struct drm_i915_private *);
 
 /* intel_dp.c */
 void intel_dp_init(struct drm_device *dev, i915_reg_t output_reg, enum port port);
@@ -1266,7 +1317,6 @@ void intel_edp_drrs_invalidate(struct drm_device *dev,
 void intel_edp_drrs_flush(struct drm_device *dev, unsigned frontbuffer_bits);
 bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
 					 struct intel_digital_port *port);
-void hsw_dp_set_ddi_pll_sel(struct intel_crtc_state *pipe_config);
 
 void
 intel_dp_program_link_training_pattern(struct intel_dp *intel_dp,
@@ -1423,8 +1473,8 @@ int intel_power_domains_init(struct drm_i915_private *);
 void intel_power_domains_fini(struct drm_i915_private *);
 void intel_power_domains_init_hw(struct drm_i915_private *dev_priv, bool resume);
 void intel_power_domains_suspend(struct drm_i915_private *dev_priv);
-void skl_pw1_misc_io_init(struct drm_i915_private *dev_priv);
-void skl_pw1_misc_io_fini(struct drm_i915_private *dev_priv);
+void bxt_display_core_init(struct drm_i915_private *dev_priv, bool resume);
+void bxt_display_core_uninit(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_enable(struct drm_i915_private *dev_priv);
 const char *
 intel_display_power_domain_str(enum intel_display_power_domain domain);
@@ -1541,6 +1591,7 @@ void intel_suspend_hw(struct drm_device *dev);
 int ilk_wm_max_level(const struct drm_device *dev);
 void intel_update_watermarks(struct drm_crtc *crtc);
 void intel_init_pm(struct drm_device *dev);
+void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv);
 void intel_pm_setup(struct drm_device *dev);
 void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
 void intel_gpu_ips_teardown(void);
@@ -1565,6 +1616,7 @@ void skl_wm_get_hw_state(struct drm_device *dev);
 void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 			  struct skl_ddb_allocation *ddb /* out */);
 uint32_t ilk_pipe_pixel_rate(const struct intel_crtc_state *pipe_config);
+bool ilk_disable_lp_wm(struct drm_device *dev);
 int sanitize_rc6_option(const struct drm_device *dev, int enable_rc6);
 
 /* intel_sdvo.c */
@@ -1606,6 +1658,18 @@ intel_atomic_get_crtc_state(struct drm_atomic_state *state,
 
 	return to_intel_crtc_state(crtc_state);
 }
+
+static inline struct intel_plane_state *
+intel_atomic_get_existing_plane_state(struct drm_atomic_state *state,
+				      struct intel_plane *plane)
+{
+	struct drm_plane_state *plane_state;
+
+	plane_state = drm_atomic_get_existing_plane_state(state, &plane->base);
+
+	return to_intel_plane_state(plane_state);
+}
+
 int intel_atomic_setup_scalers(struct drm_device *dev,
 	struct intel_crtc *intel_crtc,
 	struct intel_crtc_state *crtc_state);
@@ -1617,4 +1681,10 @@ void intel_plane_destroy_state(struct drm_plane *plane,
 			       struct drm_plane_state *state);
 extern const struct drm_plane_helper_funcs intel_plane_helper_funcs;
 
+/* intel_color.c */
+void intel_color_init(struct drm_crtc *crtc);
+int intel_color_check(struct drm_crtc *crtc, struct drm_crtc_state *state);
+void intel_color_set_csc(struct drm_crtc_state *crtc_state);
+void intel_color_load_luts(struct drm_crtc_state *crtc_state);
+
 #endif /* __INTEL_DRV_H__ */
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 01b8e9f4c272..2b22bb9bb86f 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -46,6 +46,24 @@ static const struct {
 	},
 };
 
+enum mipi_dsi_pixel_format pixel_format_from_register_bits(u32 fmt)
+{
+	/* It just so happens the VBT matches register contents. */
+	switch (fmt) {
+	case VID_MODE_FORMAT_RGB888:
+		return MIPI_DSI_FMT_RGB888;
+	case VID_MODE_FORMAT_RGB666:
+		return MIPI_DSI_FMT_RGB666;
+	case VID_MODE_FORMAT_RGB666_PACKED:
+		return MIPI_DSI_FMT_RGB666_PACKED;
+	case VID_MODE_FORMAT_RGB565:
+		return MIPI_DSI_FMT_RGB565;
+	default:
+		MISSING_CASE(fmt);
+		return MIPI_DSI_FMT_RGB666;
+	}
+}
+
 static void wait_for_dsi_fifo_empty(struct intel_dsi *intel_dsi, enum port port)
 {
 	struct drm_encoder *encoder = &intel_dsi->base.base;
@@ -268,22 +286,47 @@ static inline bool is_cmd_mode(struct intel_dsi *intel_dsi)
 static bool intel_dsi_compute_config(struct intel_encoder *encoder,
 				     struct intel_crtc_state *pipe_config)
 {
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = container_of(encoder, struct intel_dsi,
 						   base);
 	struct intel_connector *intel_connector = intel_dsi->attached_connector;
-	struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
+	struct intel_crtc *crtc = to_intel_crtc(pipe_config->base.crtc);
+	const struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
+	int ret;
 
 	DRM_DEBUG_KMS("\n");
 
 	pipe_config->has_dsi_encoder = true;
 
-	if (fixed_mode)
+	if (fixed_mode) {
 		intel_fixed_panel_mode(fixed_mode, adjusted_mode);
 
+		if (HAS_GMCH_DISPLAY(dev_priv))
+			intel_gmch_panel_fitting(crtc, pipe_config,
+						 intel_connector->panel.fitting_mode);
+		else
+			intel_pch_panel_fitting(crtc, pipe_config,
+						intel_connector->panel.fitting_mode);
+	}
+
 	/* DSI uses short packets for sync events, so clear mode flags for DSI */
 	adjusted_mode->flags = 0;
 
+	if (IS_BROXTON(dev_priv)) {
+		/* Dual link goes to DSI transcoder A. */
+		if (intel_dsi->ports == BIT(PORT_C))
+			pipe_config->cpu_transcoder = TRANSCODER_DSI_C;
+		else
+			pipe_config->cpu_transcoder = TRANSCODER_DSI_A;
+	}
+
+	ret = intel_compute_dsi_pll(encoder, pipe_config);
+	if (ret)
+		return false;
+
+	pipe_config->clock_set = true;
+
 	return true;
 }
 
@@ -403,7 +446,7 @@ static void intel_dsi_port_enable(struct intel_encoder *encoder)
 		temp &= ~LANE_CONFIGURATION_MASK;
 		temp &= ~DUAL_LINK_MODE_MASK;
 
-		if (intel_dsi->ports == ((1 << PORT_A) | (1 << PORT_C))) {
+		if (intel_dsi->ports == (BIT(PORT_A) | BIT(PORT_C))) {
 			temp |= (intel_dsi->dual_link - 1)
 						<< DUAL_LINK_MODE_SHIFT;
 			temp |= intel_crtc->pipe ?
@@ -471,14 +514,19 @@ static void intel_dsi_pre_enable(struct intel_encoder *encoder)
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
-	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
-	enum pipe pipe = intel_crtc->pipe;
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	enum port port;
 	u32 tmp;
 
 	DRM_DEBUG_KMS("\n");
 
-	intel_enable_dsi_pll(encoder);
+	/*
+	 * The BIOS may leave the PLL in a wonky state where it doesn't
+	 * lock. It needs to be fully powered down to fix it.
+	 */
+	intel_disable_dsi_pll(encoder);
+	intel_enable_dsi_pll(encoder, crtc->config);
+
 	intel_dsi_prepare(encoder);
 
 	/* Panel Enable over CRC PMIC */
@@ -488,19 +536,7 @@ static void intel_dsi_pre_enable(struct intel_encoder *encoder)
 	msleep(intel_dsi->panel_on_delay);
 
 	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
-		/*
-		 * Disable DPOunit clock gating, can stall pipe
-		 * and we need DPLL REFA always enabled
-		 */
-		tmp = I915_READ(DPLL(pipe));
-		tmp |= DPLL_REF_CLK_ENABLE_VLV;
-		I915_WRITE(DPLL(pipe), tmp);
-
-		/* update the hw state for DPLL */
-		intel_crtc->config->dpll_hw_state.dpll =
-				DPLL_INTEGRATED_REF_CLK_VLV |
-					DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
-
+		/* Disable DPOunit clock gating, can stall pipe */
 		tmp = I915_READ(DSPCLK_GATE_D);
 		tmp |= DPOUNIT_CLOCK_GATE_DISABLE;
 		I915_WRITE(DSPCLK_GATE_D, tmp);
@@ -652,11 +688,16 @@ static void intel_dsi_post_disable(struct intel_encoder *encoder)
 	drm_panel_unprepare(intel_dsi->panel);
 
 	msleep(intel_dsi->panel_off_delay);
-	msleep(intel_dsi->panel_pwr_cycle_delay);
 
 	/* Panel Disable over CRC PMIC */
 	if (intel_dsi->gpio_panel)
 		gpiod_set_value_cansleep(intel_dsi->gpio_panel, 0);
+
+	/*
+	 * FIXME As we do with eDP, just make a note of the time here
+	 * and perform the wait before the next panel power on.
+	 */
+	msleep(intel_dsi->panel_pwr_cycle_delay);
 }
 
 static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
@@ -667,7 +708,7 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 	struct drm_device *dev = encoder->base.dev;
 	enum intel_display_power_domain power_domain;
 	enum port port;
-	bool ret;
+	bool active = false;
 
 	DRM_DEBUG_KMS("\n");
 
@@ -675,55 +716,133 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
-	ret = false;
+	/*
+	 * On Broxton the PLL needs to be enabled with a valid divider
+	 * configuration, otherwise accessing DSI registers will hang the
+	 * machine. See BSpec North Display Engine registers/MIPI[BXT].
+	 */
+	if (IS_BROXTON(dev_priv) && !intel_dsi_pll_is_enabled(dev_priv))
+		goto out_put_power;
 
 	/* XXX: this only works for one DSI output */
 	for_each_dsi_port(port, intel_dsi->ports) {
 		i915_reg_t ctrl_reg = IS_BROXTON(dev) ?
 			BXT_MIPI_PORT_CTRL(port) : MIPI_PORT_CTRL(port);
-		u32 dpi_enabled, func;
-
-		func = I915_READ(MIPI_DSI_FUNC_PRG(port));
-		dpi_enabled = I915_READ(ctrl_reg) & DPI_ENABLE;
+		bool enabled = I915_READ(ctrl_reg) & DPI_ENABLE;
 
-		/* Due to some hardware limitations on BYT, MIPI Port C DPI
-		 * Enable bit does not get set. To check whether DSI Port C
-		 * was enabled in BIOS, check the Pipe B enable bit
+		/*
+		 * Due to some hardware limitations on VLV/CHV, the DPI enable
+		 * bit in port C control register does not get set. As a
+		 * workaround, check pipe B conf instead.
 		 */
-		if (IS_VALLEYVIEW(dev) && port == PORT_C)
-			dpi_enabled = I915_READ(PIPECONF(PIPE_B)) &
-							PIPECONF_ENABLE;
+		if ((IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) && port == PORT_C)
+			enabled = I915_READ(PIPECONF(PIPE_B)) & PIPECONF_ENABLE;
+
+		/* Try command mode if video mode not enabled */
+		if (!enabled) {
+			u32 tmp = I915_READ(MIPI_DSI_FUNC_PRG(port));
+			enabled = tmp & CMD_MODE_DATA_WIDTH_MASK;
+		}
 
-		if (dpi_enabled || (func & CMD_MODE_DATA_WIDTH_MASK)) {
-			if (I915_READ(MIPI_DEVICE_READY(port)) & DEVICE_READY) {
-				*pipe = port == PORT_A ? PIPE_A : PIPE_B;
-				ret = true;
+		if (!enabled)
+			continue;
 
-				goto out;
-			}
+		if (!(I915_READ(MIPI_DEVICE_READY(port)) & DEVICE_READY))
+			continue;
+
+		if (IS_BROXTON(dev_priv)) {
+			u32 tmp = I915_READ(MIPI_CTRL(port));
+			tmp &= BXT_PIPE_SELECT_MASK;
+			tmp >>= BXT_PIPE_SELECT_SHIFT;
+
+			if (WARN_ON(tmp > PIPE_C))
+				continue;
+
+			*pipe = tmp;
+		} else {
+			*pipe = port == PORT_A ? PIPE_A : PIPE_B;
 		}
+
+		active = true;
+		break;
 	}
-out:
+
+out_put_power:
 	intel_display_power_put(dev_priv, power_domain);
 
-	return ret;
+	return active;
 }
 
+static void bxt_dsi_get_pipe_config(struct intel_encoder *encoder,
+				 struct intel_crtc_state *pipe_config)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_display_mode *adjusted_mode =
+					&pipe_config->base.adjusted_mode;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	unsigned int bpp, fmt;
+	enum port port;
+	u16 vfp, vsync, vbp;
+
+	/*
+	 * Atleast one port is active as encoder->get_config called only if
+	 * encoder->get_hw_state() returns true.
+	 */
+	for_each_dsi_port(port, intel_dsi->ports) {
+		if (I915_READ(BXT_MIPI_PORT_CTRL(port)) & DPI_ENABLE)
+			break;
+	}
+
+	fmt = I915_READ(MIPI_DSI_FUNC_PRG(port)) & VID_MODE_FORMAT_MASK;
+	pipe_config->pipe_bpp =
+			mipi_dsi_pixel_format_to_bpp(
+				pixel_format_from_register_bits(fmt));
+	bpp = pipe_config->pipe_bpp;
+
+	/* In terms of pixels */
+	adjusted_mode->crtc_hdisplay =
+				I915_READ(BXT_MIPI_TRANS_HACTIVE(port));
+	adjusted_mode->crtc_vdisplay =
+				I915_READ(BXT_MIPI_TRANS_VACTIVE(port));
+	adjusted_mode->crtc_vtotal =
+				I915_READ(BXT_MIPI_TRANS_VTOTAL(port));
+
+	/*
+	 * TODO: Retrieve hfp, hsync and hbp. Adjust them for dual link and
+	 * calculate hsync_start, hsync_end, htotal and hblank_end
+	 */
+
+	/* vertical values are in terms of lines */
+	vfp = I915_READ(MIPI_VFP_COUNT(port));
+	vsync = I915_READ(MIPI_VSYNC_PADDING_COUNT(port));
+	vbp = I915_READ(MIPI_VBP_COUNT(port));
+
+	adjusted_mode->crtc_hblank_start = adjusted_mode->crtc_hdisplay;
+
+	adjusted_mode->crtc_vsync_start =
+				vfp + adjusted_mode->crtc_vdisplay;
+	adjusted_mode->crtc_vsync_end =
+				vsync + adjusted_mode->crtc_vsync_start;
+	adjusted_mode->crtc_vblank_start = adjusted_mode->crtc_vdisplay;
+	adjusted_mode->crtc_vblank_end = adjusted_mode->crtc_vtotal;
+}
+
+
 static void intel_dsi_get_config(struct intel_encoder *encoder,
 				 struct intel_crtc_state *pipe_config)
 {
+	struct drm_device *dev = encoder->base.dev;
 	u32 pclk;
 	DRM_DEBUG_KMS("\n");
 
 	pipe_config->has_dsi_encoder = true;
 
-	/*
-	 * DPLL_MD is not used in case of DSI, reading will get some default value
-	 * set dpll_md = 0
-	 */
-	pipe_config->dpll_hw_state.dpll_md = 0;
+	if (IS_BROXTON(dev))
+		bxt_dsi_get_pipe_config(encoder, pipe_config);
 
-	pclk = intel_dsi_get_pclk(encoder, pipe_config->pipe_bpp);
+	pclk = intel_dsi_get_pclk(encoder, pipe_config->pipe_bpp,
+				  pipe_config);
 	if (!pclk)
 		return;
 
@@ -736,7 +855,7 @@ intel_dsi_mode_valid(struct drm_connector *connector,
 		     struct drm_display_mode *mode)
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
-	struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
+	const struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
 	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
 
 	DRM_DEBUG_KMS("\n");
@@ -787,7 +906,7 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
-	unsigned int bpp = dsi_pixel_format_bpp(intel_dsi->pixel_format);
+	unsigned int bpp = mipi_dsi_pixel_format_to_bpp(intel_dsi->pixel_format);
 	unsigned int lane_count = intel_dsi->lane_count;
 
 	u16 hactive, hfp, hsync, hbp, vfp, vsync, vbp;
@@ -849,6 +968,23 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 	}
 }
 
+static u32 pixel_format_to_reg(enum mipi_dsi_pixel_format fmt)
+{
+	switch (fmt) {
+	case MIPI_DSI_FMT_RGB888:
+		return VID_MODE_FORMAT_RGB888;
+	case MIPI_DSI_FMT_RGB666:
+		return VID_MODE_FORMAT_RGB666;
+	case MIPI_DSI_FMT_RGB666_PACKED:
+		return VID_MODE_FORMAT_RGB666_PACKED;
+	case MIPI_DSI_FMT_RGB565:
+		return VID_MODE_FORMAT_RGB565;
+	default:
+		MISSING_CASE(fmt);
+		return VID_MODE_FORMAT_RGB666;
+	}
+}
+
 static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
@@ -858,7 +994,7 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum port port;
-	unsigned int bpp = dsi_pixel_format_bpp(intel_dsi->pixel_format);
+	unsigned int bpp = mipi_dsi_pixel_format_to_bpp(intel_dsi->pixel_format);
 	u32 val, tmp;
 	u16 mode_hdisplay;
 
@@ -917,9 +1053,7 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 		val |= CMD_MODE_DATA_WIDTH_8_BIT; /* XXX */
 	} else {
 		val |= intel_dsi->channel << VID_MODE_CHANNEL_NUMBER_SHIFT;
-
-		/* XXX: cross-check bpp vs. pixel format? */
-		val |= intel_dsi->pixel_format;
+		val |= pixel_format_to_reg(intel_dsi->pixel_format);
 	}
 
 	tmp = 0;
@@ -1059,6 +1193,48 @@ static int intel_dsi_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
+static int intel_dsi_set_property(struct drm_connector *connector,
+				  struct drm_property *property,
+				  uint64_t val)
+{
+	struct drm_device *dev = connector->dev;
+	struct intel_connector *intel_connector = to_intel_connector(connector);
+	struct drm_crtc *crtc;
+	int ret;
+
+	ret = drm_object_property_set_value(&connector->base, property, val);
+	if (ret)
+		return ret;
+
+	if (property == dev->mode_config.scaling_mode_property) {
+		if (val == DRM_MODE_SCALE_NONE) {
+			DRM_DEBUG_KMS("no scaling not supported\n");
+			return -EINVAL;
+		}
+		if (HAS_GMCH_DISPLAY(dev) &&
+		    val == DRM_MODE_SCALE_CENTER) {
+			DRM_DEBUG_KMS("centering not supported\n");
+			return -EINVAL;
+		}
+
+		if (intel_connector->panel.fitting_mode == val)
+			return 0;
+
+		intel_connector->panel.fitting_mode = val;
+	}
+
+	crtc = intel_attached_encoder(connector)->base.crtc;
+	if (crtc && crtc->state->enable) {
+		/*
+		 * If the CRTC is enabled, the display will be changed
+		 * according to the new panel fitting mode.
+		 */
+		intel_crtc_restore_mode(crtc);
+	}
+
+	return 0;
+}
+
 static void intel_dsi_connector_destroy(struct drm_connector *connector)
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
@@ -1101,11 +1277,25 @@ static const struct drm_connector_funcs intel_dsi_connector_funcs = {
 	.detect = intel_dsi_detect,
 	.destroy = intel_dsi_connector_destroy,
 	.fill_modes = drm_helper_probe_single_connector_modes,
+	.set_property = intel_dsi_set_property,
 	.atomic_get_property = intel_connector_atomic_get_property,
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
 };
 
+static void intel_dsi_add_properties(struct intel_connector *connector)
+{
+	struct drm_device *dev = connector->base.dev;
+
+	if (connector->panel.fixed_mode) {
+		drm_mode_create_scaling_mode_property(dev);
+		drm_object_attach_property(&connector->base.base,
+					   dev->mode_config.scaling_mode_property,
+					   DRM_MODE_SCALE_ASPECT);
+		connector->panel.fitting_mode = DRM_MODE_SCALE_ASPECT;
+	}
+}
+
 void intel_dsi_init(struct drm_device *dev)
 {
 	struct intel_dsi *intel_dsi;
@@ -1121,11 +1311,13 @@ void intel_dsi_init(struct drm_device *dev)
 	DRM_DEBUG_KMS("\n");
 
 	/* There is no detection method for MIPI so rely on VBT */
-	if (!dev_priv->vbt.has_mipi)
+	if (!intel_bios_is_dsi_present(dev_priv, &port))
 		return;
 
 	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
 		dev_priv->mipi_mmio_base = VLV_MIPI_BASE;
+	} else if (IS_BROXTON(dev)) {
+		dev_priv->mipi_mmio_base = BXT_MIPI_BASE;
 	} else {
 		DRM_ERROR("Unsupported Mipi device to reg base");
 		return;
@@ -1161,17 +1353,21 @@ void intel_dsi_init(struct drm_device *dev)
 	intel_connector->get_hw_state = intel_connector_get_hw_state;
 	intel_connector->unregister = intel_connector_unregister;
 
-	/* Pipe A maps to MIPI DSI port A, pipe B maps to MIPI DSI port C */
-	if (dev_priv->vbt.dsi.port == DVO_PORT_MIPIA) {
-		intel_encoder->crtc_mask = (1 << PIPE_A);
-		intel_dsi->ports = (1 << PORT_A);
-	} else if (dev_priv->vbt.dsi.port == DVO_PORT_MIPIC) {
-		intel_encoder->crtc_mask = (1 << PIPE_B);
-		intel_dsi->ports = (1 << PORT_C);
-	}
+	/*
+	 * On BYT/CHV, pipe A maps to MIPI DSI port A, pipe B maps to MIPI DSI
+	 * port C. BXT isn't limited like this.
+	 */
+	if (IS_BROXTON(dev_priv))
+		intel_encoder->crtc_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C);
+	else if (port == PORT_A)
+		intel_encoder->crtc_mask = BIT(PIPE_A);
+	else
+		intel_encoder->crtc_mask = BIT(PIPE_B);
 
 	if (dev_priv->vbt.dsi.config->dual_link)
-		intel_dsi->ports = ((1 << PORT_A) | (1 << PORT_C));
+		intel_dsi->ports = BIT(PORT_A) | BIT(PORT_C);
+	else
+		intel_dsi->ports = BIT(port);
 
 	/* Create a DSI host (and a device) for each port. */
 	for_each_dsi_port(port, intel_dsi->ports) {
@@ -1223,8 +1419,6 @@ void intel_dsi_init(struct drm_device *dev)
 
 	intel_connector_attach_encoder(intel_connector, intel_encoder);
 
-	drm_connector_register(connector);
-
 	drm_panel_attach(intel_dsi->panel, connector);
 
 	mutex_lock(&dev->mode_config.mutex);
@@ -1243,6 +1437,11 @@ void intel_dsi_init(struct drm_device *dev)
 	}
 
 	intel_panel_init(&intel_connector->panel, fixed_mode, NULL);
+
+	intel_dsi_add_properties(intel_connector);
+
+	drm_connector_register(connector);
+
 	intel_panel_setup_backlight(connector, INVALID_PIPE);
 
 	return;
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index 92f39227b361..61a6957fc6c2 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -34,8 +34,6 @@
 #define DSI_DUAL_LINK_FRONT_BACK	1
 #define DSI_DUAL_LINK_PIXEL_ALT		2
 
-int dsi_pixel_format_bpp(int pixel_format);
-
 struct intel_dsi_host;
 
 struct intel_dsi {
@@ -64,8 +62,12 @@ struct intel_dsi {
 	/* number of DSI lanes */
 	unsigned int lane_count;
 
-	/* video mode pixel format for MIPI_DSI_FUNC_PRG register */
-	u32 pixel_format;
+	/*
+	 * video mode pixel format
+	 *
+	 * XXX: consolidate on .format in struct mipi_dsi_device.
+	 */
+	enum mipi_dsi_pixel_format pixel_format;
 
 	/* video mode format for MIPI_VIDEO_MODE_FORMAT register */
 	u32 video_mode_format;
@@ -117,21 +119,25 @@ static inline struct intel_dsi_host *to_intel_dsi_host(struct mipi_dsi_host *h)
 	return container_of(h, struct intel_dsi_host, base);
 }
 
-#define for_each_dsi_port(__port, __ports_mask) \
-	for ((__port) = PORT_A; (__port) < I915_MAX_PORTS; (__port)++)	\
-		for_each_if ((__ports_mask) & (1 << (__port)))
+#define for_each_dsi_port(__port, __ports_mask) for_each_port_masked(__port, __ports_mask)
 
 static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
 {
 	return container_of(encoder, struct intel_dsi, base.base);
 }
 
-extern void intel_enable_dsi_pll(struct intel_encoder *encoder);
-extern void intel_disable_dsi_pll(struct intel_encoder *encoder);
-extern u32 intel_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp);
-extern void intel_dsi_reset_clocks(struct intel_encoder *encoder,
-							enum port port);
+bool intel_dsi_pll_is_enabled(struct drm_i915_private *dev_priv);
+int intel_compute_dsi_pll(struct intel_encoder *encoder,
+			  struct intel_crtc_state *config);
+void intel_enable_dsi_pll(struct intel_encoder *encoder,
+			  const struct intel_crtc_state *config);
+void intel_disable_dsi_pll(struct intel_encoder *encoder);
+u32 intel_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp,
+		       struct intel_crtc_state *config);
+void intel_dsi_reset_clocks(struct intel_encoder *encoder,
+			    enum port port);
 
 struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id);
+enum mipi_dsi_pixel_format pixel_format_from_register_bits(u32 fmt);
 
 #endif /* _INTEL_DSI_H */
diff --git a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
index 7f145b4fec6a..e498f1c3221e 100644
--- a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
+++ b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
@@ -58,50 +58,41 @@ static inline struct vbt_panel *to_vbt_panel(struct drm_panel *panel)
 
 #define NS_KHZ_RATIO 1000000
 
-#define GPI0_NC_0_HV_DDI0_HPD           0x4130
-#define GPIO_NC_0_HV_DDI0_PAD           0x4138
-#define GPIO_NC_1_HV_DDI0_DDC_SDA       0x4120
-#define GPIO_NC_1_HV_DDI0_DDC_SDA_PAD   0x4128
-#define GPIO_NC_2_HV_DDI0_DDC_SCL       0x4110
-#define GPIO_NC_2_HV_DDI0_DDC_SCL_PAD   0x4118
-#define GPIO_NC_3_PANEL0_VDDEN          0x4140
-#define GPIO_NC_3_PANEL0_VDDEN_PAD      0x4148
-#define GPIO_NC_4_PANEL0_BLKEN          0x4150
-#define GPIO_NC_4_PANEL0_BLKEN_PAD      0x4158
-#define GPIO_NC_5_PANEL0_BLKCTL         0x4160
-#define GPIO_NC_5_PANEL0_BLKCTL_PAD     0x4168
-#define GPIO_NC_6_PCONF0                0x4180
-#define GPIO_NC_6_PAD                   0x4188
-#define GPIO_NC_7_PCONF0                0x4190
-#define GPIO_NC_7_PAD                   0x4198
-#define GPIO_NC_8_PCONF0                0x4170
-#define GPIO_NC_8_PAD                   0x4178
-#define GPIO_NC_9_PCONF0                0x4100
-#define GPIO_NC_9_PAD                   0x4108
-#define GPIO_NC_10_PCONF0               0x40E0
-#define GPIO_NC_10_PAD                  0x40E8
-#define GPIO_NC_11_PCONF0               0x40F0
-#define GPIO_NC_11_PAD                  0x40F8
-
-struct gpio_table {
-	u16 function_reg;
-	u16 pad_reg;
-	u8 init;
+/* base offsets for gpio pads */
+#define VLV_GPIO_NC_0_HV_DDI0_HPD	0x4130
+#define VLV_GPIO_NC_1_HV_DDI0_DDC_SDA	0x4120
+#define VLV_GPIO_NC_2_HV_DDI0_DDC_SCL	0x4110
+#define VLV_GPIO_NC_3_PANEL0_VDDEN	0x4140
+#define VLV_GPIO_NC_4_PANEL0_BKLTEN	0x4150
+#define VLV_GPIO_NC_5_PANEL0_BKLTCTL	0x4160
+#define VLV_GPIO_NC_6_HV_DDI1_HPD	0x4180
+#define VLV_GPIO_NC_7_HV_DDI1_DDC_SDA	0x4190
+#define VLV_GPIO_NC_8_HV_DDI1_DDC_SCL	0x4170
+#define VLV_GPIO_NC_9_PANEL1_VDDEN	0x4100
+#define VLV_GPIO_NC_10_PANEL1_BKLTEN	0x40E0
+#define VLV_GPIO_NC_11_PANEL1_BKLTCTL	0x40F0
+
+#define VLV_GPIO_PCONF0(base_offset)	(base_offset)
+#define VLV_GPIO_PAD_VAL(base_offset)	((base_offset) + 8)
+
+struct gpio_map {
+	u16 base_offset;
+	bool init;
 };
 
-static struct gpio_table gtable[] = {
-	{ GPI0_NC_0_HV_DDI0_HPD, GPIO_NC_0_HV_DDI0_PAD, 0 },
-	{ GPIO_NC_1_HV_DDI0_DDC_SDA, GPIO_NC_1_HV_DDI0_DDC_SDA_PAD, 0 },
-	{ GPIO_NC_2_HV_DDI0_DDC_SCL, GPIO_NC_2_HV_DDI0_DDC_SCL_PAD, 0 },
-	{ GPIO_NC_3_PANEL0_VDDEN, GPIO_NC_3_PANEL0_VDDEN_PAD, 0 },
-	{ GPIO_NC_4_PANEL0_BLKEN, GPIO_NC_4_PANEL0_BLKEN_PAD, 0 },
-	{ GPIO_NC_5_PANEL0_BLKCTL, GPIO_NC_5_PANEL0_BLKCTL_PAD, 0 },
-	{ GPIO_NC_6_PCONF0, GPIO_NC_6_PAD, 0 },
-	{ GPIO_NC_7_PCONF0, GPIO_NC_7_PAD, 0 },
-	{ GPIO_NC_8_PCONF0, GPIO_NC_8_PAD, 0 },
-	{ GPIO_NC_9_PCONF0, GPIO_NC_9_PAD, 0 },
-	{ GPIO_NC_10_PCONF0, GPIO_NC_10_PAD, 0},
-	{ GPIO_NC_11_PCONF0, GPIO_NC_11_PAD, 0}
+static struct gpio_map vlv_gpio_table[] = {
+	{ VLV_GPIO_NC_0_HV_DDI0_HPD },
+	{ VLV_GPIO_NC_1_HV_DDI0_DDC_SDA },
+	{ VLV_GPIO_NC_2_HV_DDI0_DDC_SCL },
+	{ VLV_GPIO_NC_3_PANEL0_VDDEN },
+	{ VLV_GPIO_NC_4_PANEL0_BKLTEN },
+	{ VLV_GPIO_NC_5_PANEL0_BKLTCTL },
+	{ VLV_GPIO_NC_6_HV_DDI1_HPD },
+	{ VLV_GPIO_NC_7_HV_DDI1_DDC_SDA },
+	{ VLV_GPIO_NC_8_HV_DDI1_DDC_SCL },
+	{ VLV_GPIO_NC_9_PANEL1_VDDEN },
+	{ VLV_GPIO_NC_10_PANEL1_BKLTEN },
+	{ VLV_GPIO_NC_11_PANEL1_BKLTCTL },
 };
 
 static inline enum port intel_dsi_seq_port_to_port(u8 port)
@@ -196,56 +187,76 @@ static const u8 *mipi_exec_delay(struct intel_dsi *intel_dsi, const u8 *data)
 	return data;
 }
 
-static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
+static void vlv_exec_gpio(struct drm_i915_private *dev_priv,
+			  u8 gpio_source, u8 gpio_index, bool value)
 {
-	u8 gpio, action;
-	u16 function, pad;
-	u32 val;
-	struct drm_device *dev = intel_dsi->base.base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (dev_priv->vbt.dsi.seq_version >= 3)
-		data++;
-
-	gpio = *data++;
+	struct gpio_map *map;
+	u16 pconf0, padval;
+	u32 tmp;
+	u8 port;
 
-	/* pull up/down */
-	action = *data++ & 1;
-
-	if (gpio >= ARRAY_SIZE(gtable)) {
-		DRM_DEBUG_KMS("unknown gpio %u\n", gpio);
-		goto out;
+	if (gpio_index >= ARRAY_SIZE(vlv_gpio_table)) {
+		DRM_DEBUG_KMS("unknown gpio index %u\n", gpio_index);
+		return;
 	}
 
-	if (!IS_VALLEYVIEW(dev_priv)) {
-		DRM_DEBUG_KMS("GPIO element not supported on this platform\n");
-		goto out;
-	}
+	map = &vlv_gpio_table[gpio_index];
 
 	if (dev_priv->vbt.dsi.seq_version >= 3) {
 		DRM_DEBUG_KMS("GPIO element v3 not supported\n");
-		goto out;
+		return;
+	} else {
+		if (gpio_source == 0) {
+			port = IOSF_PORT_GPIO_NC;
+		} else if (gpio_source == 1) {
+			port = IOSF_PORT_GPIO_SC;
+		} else {
+			DRM_DEBUG_KMS("unknown gpio source %u\n", gpio_source);
+			return;
+		}
 	}
 
-	function = gtable[gpio].function_reg;
-	pad = gtable[gpio].pad_reg;
+	pconf0 = VLV_GPIO_PCONF0(map->base_offset);
+	padval = VLV_GPIO_PAD_VAL(map->base_offset);
 
 	mutex_lock(&dev_priv->sb_lock);
-	if (!gtable[gpio].init) {
-		/* program the function */
+	if (!map->init) {
 		/* FIXME: remove constant below */
-		vlv_iosf_sb_write(dev_priv, IOSF_PORT_GPIO_NC, function,
-				  0x2000CC00);
-		gtable[gpio].init = 1;
+		vlv_iosf_sb_write(dev_priv, port, pconf0, 0x2000CC00);
+		map->init = true;
 	}
 
-	val = 0x4 | action;
+	tmp = 0x4 | value;
+	vlv_iosf_sb_write(dev_priv, port, padval, tmp);
+	mutex_unlock(&dev_priv->sb_lock);
+}
+
+static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
+{
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u8 gpio_source, gpio_index;
+	bool value;
+
+	if (dev_priv->vbt.dsi.seq_version >= 3)
+		data++;
+
+	gpio_index = *data++;
+
+	/* gpio source in sequence v2 only */
+	if (dev_priv->vbt.dsi.seq_version == 2)
+		gpio_source = (*data >> 1) & 3;
+	else
+		gpio_source = 0;
 
 	/* pull up/down */
-	vlv_iosf_sb_write(dev_priv, IOSF_PORT_GPIO_NC, pad, val);
-	mutex_unlock(&dev_priv->sb_lock);
+	value = *data++ & 1;
+
+	if (IS_VALLEYVIEW(dev_priv))
+		vlv_exec_gpio(dev_priv, gpio_source, gpio_index, value);
+	else
+		DRM_DEBUG_KMS("GPIO element not supported on this platform\n");
 
-out:
 	return data;
 }
 
@@ -420,7 +431,7 @@ struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id)
 	struct mipi_pps_data *pps = dev_priv->vbt.dsi.pps;
 	struct drm_display_mode *mode = dev_priv->vbt.lfp_lvds_vbt_mode;
 	struct vbt_panel *vbt_panel;
-	u32 bits_per_pixel = 24;
+	u32 bpp;
 	u32 tlpx_ns, extra_byte_count, bitrate, tlpx_ui;
 	u32 ui_num, ui_den;
 	u32 prepare_cnt, exit_zero_cnt, clk_zero_cnt, trail_cnt;
@@ -436,12 +447,13 @@ struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id)
 	intel_dsi->eotp_pkt = mipi_config->eot_pkt_disabled ? 0 : 1;
 	intel_dsi->clock_stop = mipi_config->enable_clk_stop ? 1 : 0;
 	intel_dsi->lane_count = mipi_config->lane_cnt + 1;
-	intel_dsi->pixel_format = mipi_config->videomode_color_format << 7;
+	intel_dsi->pixel_format =
+			pixel_format_from_register_bits(
+				mipi_config->videomode_color_format << 7);
+	bpp = mipi_dsi_pixel_format_to_bpp(intel_dsi->pixel_format);
+
 	intel_dsi->dual_link = mipi_config->dual_link;
 	intel_dsi->pixel_overlap = mipi_config->pixel_overlap;
-
-	bits_per_pixel = dsi_pixel_format_bpp(intel_dsi->pixel_format);
-
 	intel_dsi->operation_mode = mipi_config->is_cmd_mode;
 	intel_dsi->video_mode_format = mipi_config->video_transfer_mode;
 	intel_dsi->escape_clk_div = mipi_config->byte_clk_sel;
@@ -475,8 +487,7 @@ struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id)
 	 */
 	if (intel_dsi->video_mode_format == VIDEO_MODE_BURST) {
 		if (mipi_config->target_burst_mode_freq) {
-			computed_ddr =
-				(pclk * bits_per_pixel) / intel_dsi->lane_count;
+			computed_ddr = (pclk * bpp) / intel_dsi->lane_count;
 
 			if (mipi_config->target_burst_mode_freq <
 								computed_ddr) {
@@ -499,7 +510,7 @@ struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id)
 	intel_dsi->burst_mode_ratio = burst_mode_ratio;
 	intel_dsi->pclk = pclk;
 
-	bitrate = (pclk * bits_per_pixel) / intel_dsi->lane_count;
+	bitrate = (pclk * bpp) / intel_dsi->lane_count;
 
 	switch (intel_dsi->escape_clk_div) {
 	case 0:
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index 70883c54cb0a..1765e6e18f2c 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -30,33 +30,7 @@
 #include "i915_drv.h"
 #include "intel_dsi.h"
 
-int dsi_pixel_format_bpp(int pixel_format)
-{
-	int bpp;
-
-	switch (pixel_format) {
-	default:
-	case VID_MODE_FORMAT_RGB888:
-	case VID_MODE_FORMAT_RGB666_LOOSE:
-		bpp = 24;
-		break;
-	case VID_MODE_FORMAT_RGB666:
-		bpp = 18;
-		break;
-	case VID_MODE_FORMAT_RGB565:
-		bpp = 16;
-		break;
-	}
-
-	return bpp;
-}
-
-struct dsi_mnp {
-	u32 dsi_pll_ctrl;
-	u32 dsi_pll_div;
-};
-
-static const u32 lfsr_converts[] = {
+static const u16 lfsr_converts[] = {
 	426, 469, 234, 373, 442, 221, 110, 311, 411,		/* 62 - 70 */
 	461, 486, 243, 377, 188, 350, 175, 343, 427, 213,	/* 71 - 80 */
 	106, 53, 282, 397, 454, 227, 113, 56, 284, 142,		/* 81 - 90 */
@@ -64,10 +38,11 @@ static const u32 lfsr_converts[] = {
 };
 
 /* Get DSI clock from pixel clock */
-static u32 dsi_clk_from_pclk(u32 pclk, int pixel_format, int lane_count)
+static u32 dsi_clk_from_pclk(u32 pclk, enum mipi_dsi_pixel_format fmt,
+			     int lane_count)
 {
 	u32 dsi_clk_khz;
-	u32 bpp = dsi_pixel_format_bpp(pixel_format);
+	u32 bpp = mipi_dsi_pixel_format_to_bpp(fmt);
 
 	/* DSI data rate = pixel clock * bits per pixel / lane count
 	   pixel clock is converted from KHz to Hz */
@@ -77,7 +52,8 @@ static u32 dsi_clk_from_pclk(u32 pclk, int pixel_format, int lane_count)
 }
 
 static int dsi_calc_mnp(struct drm_i915_private *dev_priv,
-			struct dsi_mnp *dsi_mnp, int target_dsi_clk)
+			struct intel_crtc_state *config,
+			int target_dsi_clk)
 {
 	unsigned int calc_m = 0, calc_p = 0;
 	unsigned int m_min, m_max, p_min = 2, p_max = 6;
@@ -123,8 +99,8 @@ static int dsi_calc_mnp(struct drm_i915_private *dev_priv,
 	/* register has log2(N1), this works fine for powers of two */
 	n = ffs(n) - 1;
 	m_seed = lfsr_converts[calc_m - 62];
-	dsi_mnp->dsi_pll_ctrl = 1 << (DSI_PLL_P1_POST_DIV_SHIFT + calc_p - 2);
-	dsi_mnp->dsi_pll_div = n << DSI_PLL_N1_DIV_SHIFT |
+	config->dsi_pll.ctrl = 1 << (DSI_PLL_P1_POST_DIV_SHIFT + calc_p - 2);
+	config->dsi_pll.div = n << DSI_PLL_N1_DIV_SHIFT |
 		m_seed << DSI_PLL_M1_DIV_SHIFT;
 
 	return 0;
@@ -134,54 +110,55 @@ static int dsi_calc_mnp(struct drm_i915_private *dev_priv,
  * XXX: The muxing and gating is hard coded for now. Need to add support for
  * sharing PLLs with two DSI outputs.
  */
-static void vlv_configure_dsi_pll(struct intel_encoder *encoder)
+static int vlv_compute_dsi_pll(struct intel_encoder *encoder,
+			       struct intel_crtc_state *config)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	int ret;
-	struct dsi_mnp dsi_mnp;
 	u32 dsi_clk;
 
 	dsi_clk = dsi_clk_from_pclk(intel_dsi->pclk, intel_dsi->pixel_format,
 				    intel_dsi->lane_count);
 
-	ret = dsi_calc_mnp(dev_priv, &dsi_mnp, dsi_clk);
+	ret = dsi_calc_mnp(dev_priv, config, dsi_clk);
 	if (ret) {
 		DRM_DEBUG_KMS("dsi_calc_mnp failed\n");
-		return;
+		return ret;
 	}
 
 	if (intel_dsi->ports & (1 << PORT_A))
-		dsi_mnp.dsi_pll_ctrl |= DSI_PLL_CLK_GATE_DSI0_DSIPLL;
+		config->dsi_pll.ctrl |= DSI_PLL_CLK_GATE_DSI0_DSIPLL;
 
 	if (intel_dsi->ports & (1 << PORT_C))
-		dsi_mnp.dsi_pll_ctrl |= DSI_PLL_CLK_GATE_DSI1_DSIPLL;
+		config->dsi_pll.ctrl |= DSI_PLL_CLK_GATE_DSI1_DSIPLL;
+
+	config->dsi_pll.ctrl |= DSI_PLL_VCO_EN;
 
 	DRM_DEBUG_KMS("dsi pll div %08x, ctrl %08x\n",
-		      dsi_mnp.dsi_pll_div, dsi_mnp.dsi_pll_ctrl);
+		      config->dsi_pll.div, config->dsi_pll.ctrl);
 
-	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, 0);
-	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_DIVIDER, dsi_mnp.dsi_pll_div);
-	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, dsi_mnp.dsi_pll_ctrl);
+	return 0;
 }
 
-static void vlv_enable_dsi_pll(struct intel_encoder *encoder)
+static void vlv_enable_dsi_pll(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
-	u32 tmp;
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
 	DRM_DEBUG_KMS("\n");
 
 	mutex_lock(&dev_priv->sb_lock);
 
-	vlv_configure_dsi_pll(encoder);
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, 0);
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_DIVIDER, config->dsi_pll.div);
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL,
+		      config->dsi_pll.ctrl & ~DSI_PLL_VCO_EN);
 
 	/* wait at least 0.5 us after ungating before enabling VCO */
 	usleep_range(1, 10);
 
-	tmp = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
-	tmp |= DSI_PLL_VCO_EN;
-	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, tmp);
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, config->dsi_pll.ctrl);
 
 	if (wait_for(vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL) &
 						DSI_PLL_LOCK, 20)) {
@@ -197,7 +174,7 @@ static void vlv_enable_dsi_pll(struct intel_encoder *encoder)
 
 static void vlv_disable_dsi_pll(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 tmp;
 
 	DRM_DEBUG_KMS("\n");
@@ -212,9 +189,39 @@ static void vlv_disable_dsi_pll(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->sb_lock);
 }
 
+static bool bxt_dsi_pll_is_enabled(struct drm_i915_private *dev_priv)
+{
+	bool enabled;
+	u32 val;
+	u32 mask;
+
+	mask = BXT_DSI_PLL_DO_ENABLE | BXT_DSI_PLL_LOCKED;
+	val = I915_READ(BXT_DSI_PLL_ENABLE);
+	enabled = (val & mask) == mask;
+
+	if (!enabled)
+		return false;
+
+	/*
+	 * Both dividers must be programmed with valid values even if only one
+	 * of the PLL is used, see BSpec/Broxton Clocks. Check this here for
+	 * paranoia, since BIOS is known to misconfigure PLLs in this way at
+	 * times, and since accessing DSI registers with invalid dividers
+	 * causes a system hang.
+	 */
+	val = I915_READ(BXT_DSI_PLL_CTL);
+	if (!(val & BXT_DSIA_16X_MASK) || !(val & BXT_DSIC_16X_MASK)) {
+		DRM_DEBUG_DRIVER("PLL is enabled with invalid divider settings (%08x)\n",
+				 val);
+		enabled = false;
+	}
+
+	return enabled;
+}
+
 static void bxt_disable_dsi_pll(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 val;
 
 	DRM_DEBUG_KMS("\n");
@@ -232,23 +239,24 @@ static void bxt_disable_dsi_pll(struct intel_encoder *encoder)
 		DRM_ERROR("Timeout waiting for PLL lock deassertion\n");
 }
 
-static void assert_bpp_mismatch(int pixel_format, int pipe_bpp)
+static void assert_bpp_mismatch(enum mipi_dsi_pixel_format fmt, int pipe_bpp)
 {
-	int bpp = dsi_pixel_format_bpp(pixel_format);
+	int bpp = mipi_dsi_pixel_format_to_bpp(fmt);
 
 	WARN(bpp != pipe_bpp,
 	     "bpp match assertion failure (expected %d, current %d)\n",
 	     bpp, pipe_bpp);
 }
 
-static u32 vlv_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp)
+static u32 vlv_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp,
+			    struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	u32 dsi_clock, pclk;
 	u32 pll_ctl, pll_div;
 	u32 m = 0, p = 0, n;
-	int refclk = 25000;
+	int refclk = IS_CHERRYVIEW(dev_priv) ? 100000 : 25000;
 	int i;
 
 	DRM_DEBUG_KMS("\n");
@@ -258,6 +266,9 @@ static u32 vlv_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	pll_div = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_DIVIDER);
 	mutex_unlock(&dev_priv->sb_lock);
 
+	config->dsi_pll.ctrl = pll_ctl & ~DSI_PLL_LOCK;
+	config->dsi_pll.div = pll_div;
+
 	/* mask out other bits and extract the P1 divisor */
 	pll_ctl &= DSI_PLL_P1_POST_DIV_MASK;
 	pll_ctl = pll_ctl >> (DSI_PLL_P1_POST_DIV_SHIFT - 2);
@@ -303,7 +314,8 @@ static u32 vlv_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	return pclk;
 }
 
-static u32 bxt_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp)
+static u32 bxt_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp,
+			    struct intel_crtc_state *config)
 {
 	u32 pclk;
 	u32 dsi_clk;
@@ -317,15 +329,9 @@ static u32 bxt_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp)
 		return 0;
 	}
 
-	dsi_ratio = I915_READ(BXT_DSI_PLL_CTL) &
-				BXT_DSI_PLL_RATIO_MASK;
+	config->dsi_pll.ctrl = I915_READ(BXT_DSI_PLL_CTL);
 
-	/* Invalid DSI ratio ? */
-	if (dsi_ratio < BXT_DSI_PLL_RATIO_MIN ||
-			dsi_ratio > BXT_DSI_PLL_RATIO_MAX) {
-		DRM_ERROR("Invalid DSI pll ratio(%u) programmed\n", dsi_ratio);
-		return 0;
-	}
+	dsi_ratio = config->dsi_pll.ctrl & BXT_DSI_PLL_RATIO_MASK;
 
 	dsi_clk = (dsi_ratio * BXT_REF_CLOCK_KHZ) / 2;
 
@@ -338,12 +344,13 @@ static u32 bxt_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	return pclk;
 }
 
-u32 intel_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp)
+u32 intel_dsi_get_pclk(struct intel_encoder *encoder, int pipe_bpp,
+		       struct intel_crtc_state *config)
 {
 	if (IS_BROXTON(encoder->base.dev))
-		return bxt_dsi_get_pclk(encoder, pipe_bpp);
+		return bxt_dsi_get_pclk(encoder, pipe_bpp, config);
 	else
-		return vlv_dsi_get_pclk(encoder, pipe_bpp);
+		return vlv_dsi_get_pclk(encoder, pipe_bpp, config);
 }
 
 static void vlv_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
@@ -360,51 +367,72 @@ static void vlv_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
 }
 
 /* Program BXT Mipi clocks and dividers */
-static void bxt_dsi_program_clocks(struct drm_device *dev, enum port port)
+static void bxt_dsi_program_clocks(struct drm_device *dev, enum port port,
+				   const struct intel_crtc_state *config)
 {
-	u32 tmp;
-	u32 divider;
-	u32 dsi_rate;
-	u32 pll_ratio;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 tmp;
+	u32 dsi_rate = 0;
+	u32 pll_ratio = 0;
+	u32 rx_div;
+	u32 tx_div;
+	u32 rx_div_upper;
+	u32 rx_div_lower;
+	u32 mipi_8by3_divider;
 
 	/* Clear old configurations */
 	tmp = I915_READ(BXT_MIPI_CLOCK_CTL);
 	tmp &= ~(BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port));
-	tmp &= ~(BXT_MIPI_RX_ESCLK_FIXDIV_MASK(port));
-	tmp &= ~(BXT_MIPI_ESCLK_VAR_DIV_MASK(port));
-	tmp &= ~(BXT_MIPI_DPHY_DIVIDER_MASK(port));
+	tmp &= ~(BXT_MIPI_RX_ESCLK_UPPER_FIXDIV_MASK(port));
+	tmp &= ~(BXT_MIPI_8X_BY3_DIVIDER_MASK(port));
+	tmp &= ~(BXT_MIPI_RX_ESCLK_LOWER_FIXDIV_MASK(port));
 
 	/* Get the current DSI rate(actual) */
-	pll_ratio = I915_READ(BXT_DSI_PLL_CTL) &
-				BXT_DSI_PLL_RATIO_MASK;
+	pll_ratio = config->dsi_pll.ctrl & BXT_DSI_PLL_RATIO_MASK;
 	dsi_rate = (BXT_REF_CLOCK_KHZ * pll_ratio) / 2;
 
-	/* Max possible output of clock is 39.5 MHz, program value -1 */
-	divider = (dsi_rate / BXT_MAX_VAR_OUTPUT_KHZ) - 1;
-	tmp |= BXT_MIPI_ESCLK_VAR_DIV(port, divider);
+	/*
+	 * tx clock should be <= 20MHz and the div value must be
+	 * subtracted by 1 as per bspec
+	 */
+	tx_div = DIV_ROUND_UP(dsi_rate, 20000) - 1;
+	/*
+	 * rx clock should be <= 150MHz and the div value must be
+	 * subtracted by 1 as per bspec
+	 */
+	rx_div = DIV_ROUND_UP(dsi_rate, 150000) - 1;
 
 	/*
-	 * Tx escape clock must be as close to 20MHz possible, but should
-	 * not exceed it. Hence select divide by 2
+	 * rx divider value needs to be updated in the
+	 * two differnt bit fields in the register hence splitting the
+	 * rx divider value accordingly
 	 */
-	tmp |= BXT_MIPI_TX_ESCLK_8XDIV_BY2(port);
+	rx_div_lower = rx_div & RX_DIVIDER_BIT_1_2;
+	rx_div_upper = (rx_div & RX_DIVIDER_BIT_3_4) >> 2;
 
-	tmp |= BXT_MIPI_RX_ESCLK_8X_BY3(port);
+	/* As per bpsec program the 8/3X clock divider to the below value */
+	if (dev_priv->vbt.dsi.config->is_cmd_mode)
+		mipi_8by3_divider = 0x2;
+	else
+		mipi_8by3_divider = 0x3;
+
+	tmp |= BXT_MIPI_8X_BY3_DIVIDER(port, mipi_8by3_divider);
+	tmp |= BXT_MIPI_TX_ESCLK_DIVIDER(port, tx_div);
+	tmp |= BXT_MIPI_RX_ESCLK_LOWER_DIVIDER(port, rx_div_lower);
+	tmp |= BXT_MIPI_RX_ESCLK_UPPER_DIVIDER(port, rx_div_upper);
 
 	I915_WRITE(BXT_MIPI_CLOCK_CTL, tmp);
 }
 
-static bool bxt_configure_dsi_pll(struct intel_encoder *encoder)
+static int bxt_compute_dsi_pll(struct intel_encoder *encoder,
+			       struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	u8 dsi_ratio;
 	u32 dsi_clk;
-	u32 val;
 
 	dsi_clk = dsi_clk_from_pclk(intel_dsi->pclk, intel_dsi->pixel_format,
-			intel_dsi->lane_count);
+				    intel_dsi->lane_count);
 
 	/*
 	 * From clock diagram, to get PLL ratio divider, divide double of DSI
@@ -413,9 +441,9 @@ static bool bxt_configure_dsi_pll(struct intel_encoder *encoder)
 	 */
 	dsi_ratio = DIV_ROUND_UP(dsi_clk * 2, BXT_REF_CLOCK_KHZ);
 	if (dsi_ratio < BXT_DSI_PLL_RATIO_MIN ||
-			dsi_ratio > BXT_DSI_PLL_RATIO_MAX) {
+	    dsi_ratio > BXT_DSI_PLL_RATIO_MAX) {
 		DRM_ERROR("Cant get a suitable ratio from DSI PLL ratios\n");
-		return false;
+		return -ECHRNG;
 	}
 
 	/*
@@ -423,27 +451,19 @@ static bool bxt_configure_dsi_pll(struct intel_encoder *encoder)
 	 * Spec says both have to be programmed, even if one is not getting
 	 * used. Configure MIPI_CLOCK_CTL dividers in modeset
 	 */
-	val = I915_READ(BXT_DSI_PLL_CTL);
-	val &= ~BXT_DSI_PLL_PVD_RATIO_MASK;
-	val &= ~BXT_DSI_FREQ_SEL_MASK;
-	val &= ~BXT_DSI_PLL_RATIO_MASK;
-	val |= (dsi_ratio | BXT_DSIA_16X_BY2 | BXT_DSIC_16X_BY2);
+	config->dsi_pll.ctrl = dsi_ratio | BXT_DSIA_16X_BY2 | BXT_DSIC_16X_BY2;
 
 	/* As per recommendation from hardware team,
 	 * Prog PVD ratio =1 if dsi ratio <= 50
 	 */
-	if (dsi_ratio <= 50) {
-		val &= ~BXT_DSI_PLL_PVD_RATIO_MASK;
-		val |= BXT_DSI_PLL_PVD_RATIO_1;
-	}
+	if (dsi_ratio <= 50)
+		config->dsi_pll.ctrl |= BXT_DSI_PLL_PVD_RATIO_1;
 
-	I915_WRITE(BXT_DSI_PLL_CTL, val);
-	POSTING_READ(BXT_DSI_PLL_CTL);
-
-	return true;
+	return 0;
 }
 
-static void bxt_enable_dsi_pll(struct intel_encoder *encoder)
+static void bxt_enable_dsi_pll(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *config)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
@@ -452,23 +472,13 @@ static void bxt_enable_dsi_pll(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
-	val = I915_READ(BXT_DSI_PLL_ENABLE);
-
-	if (val & BXT_DSI_PLL_DO_ENABLE) {
-		WARN(1, "DSI PLL already enabled. Disabling it.\n");
-		val &= ~BXT_DSI_PLL_DO_ENABLE;
-		I915_WRITE(BXT_DSI_PLL_ENABLE, val);
-	}
-
 	/* Configure PLL vales */
-	if (!bxt_configure_dsi_pll(encoder)) {
-		DRM_ERROR("Configure DSI PLL failed, abort PLL enable\n");
-		return;
-	}
+	I915_WRITE(BXT_DSI_PLL_CTL, config->dsi_pll.ctrl);
+	POSTING_READ(BXT_DSI_PLL_CTL);
 
 	/* Program TX, RX, Dphy clocks */
 	for_each_dsi_port(port, intel_dsi->ports)
-		bxt_dsi_program_clocks(encoder->base.dev, port);
+		bxt_dsi_program_clocks(encoder->base.dev, port, config);
 
 	/* Enable DSI PLL */
 	val = I915_READ(BXT_DSI_PLL_ENABLE);
@@ -484,14 +494,38 @@ static void bxt_enable_dsi_pll(struct intel_encoder *encoder)
 	DRM_DEBUG_KMS("DSI PLL locked\n");
 }
 
-void intel_enable_dsi_pll(struct intel_encoder *encoder)
+bool intel_dsi_pll_is_enabled(struct drm_i915_private *dev_priv)
+{
+	if (IS_BROXTON(dev_priv))
+		return bxt_dsi_pll_is_enabled(dev_priv);
+
+	MISSING_CASE(INTEL_DEVID(dev_priv));
+
+	return false;
+}
+
+int intel_compute_dsi_pll(struct intel_encoder *encoder,
+			  struct intel_crtc_state *config)
+{
+	struct drm_device *dev = encoder->base.dev;
+
+	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
+		return vlv_compute_dsi_pll(encoder, config);
+	else if (IS_BROXTON(dev))
+		return bxt_compute_dsi_pll(encoder, config);
+
+	return -ENODEV;
+}
+
+void intel_enable_dsi_pll(struct intel_encoder *encoder,
+			  const struct intel_crtc_state *config)
 {
 	struct drm_device *dev = encoder->base.dev;
 
 	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
-		vlv_enable_dsi_pll(encoder);
+		vlv_enable_dsi_pll(encoder, config);
 	else if (IS_BROXTON(dev))
-		bxt_enable_dsi_pll(encoder);
+		bxt_enable_dsi_pll(encoder, config);
 }
 
 void intel_disable_dsi_pll(struct intel_encoder *encoder)
@@ -513,9 +547,9 @@ static void bxt_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
 	/* Clear old configurations */
 	tmp = I915_READ(BXT_MIPI_CLOCK_CTL);
 	tmp &= ~(BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port));
-	tmp &= ~(BXT_MIPI_RX_ESCLK_FIXDIV_MASK(port));
-	tmp &= ~(BXT_MIPI_ESCLK_VAR_DIV_MASK(port));
-	tmp &= ~(BXT_MIPI_DPHY_DIVIDER_MASK(port));
+	tmp &= ~(BXT_MIPI_RX_ESCLK_UPPER_FIXDIV_MASK(port));
+	tmp &= ~(BXT_MIPI_8X_BY3_DIVIDER_MASK(port));
+	tmp &= ~(BXT_MIPI_RX_ESCLK_LOWER_FIXDIV_MASK(port));
 	I915_WRITE(BXT_MIPI_CLOCK_CTL, tmp);
 	I915_WRITE(MIPI_EOT_DISABLE(port), CLOCKSTOP);
 }
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 0f0492f4a357..d5a7cfec589b 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -506,6 +506,7 @@ static int find_compression_threshold(struct drm_i915_private *dev_priv,
 				      int size,
 				      int fb_cpp)
 {
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	int compression_threshold = 1;
 	int ret;
 	u64 end;
@@ -516,9 +517,9 @@ static int find_compression_threshold(struct drm_i915_private *dev_priv,
 	 * underruns, even if that range is not reserved by the BIOS. */
 	if (IS_BROADWELL(dev_priv) ||
 	    IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
-		end = dev_priv->gtt.stolen_size - 8 * 1024 * 1024;
+		end = ggtt->stolen_size - 8 * 1024 * 1024;
 	else
-		end = dev_priv->gtt.stolen_usable_size;
+		end = ggtt->stolen_usable_size;
 
 	/* HACK: This code depends on what we will do in *_enable_fbc. If that
 	 * code changes, this code needs to change as well.
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 97a91e631915..ab8d09a81f14 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -122,6 +122,7 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 	struct drm_framebuffer *fb;
 	struct drm_device *dev = helper->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_mode_fb_cmd2 mode_cmd = {};
 	struct drm_i915_gem_object *obj = NULL;
 	int size, ret;
@@ -146,7 +147,7 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 	/* If the FB is too big, just don't use it since fbdev is not very
 	 * important and we should probably use that space with FBC or other
 	 * features. */
-	if (size * 2 < dev_priv->gtt.stolen_usable_size)
+	if (size * 2 < ggtt->stolen_usable_size)
 		obj = i915_gem_object_create_stolen(dev, size);
 	if (obj == NULL)
 		obj = i915_gem_alloc_object(dev, size);
@@ -181,7 +182,8 @@ static int intelfb_create(struct drm_fb_helper *helper,
 		container_of(helper, struct intel_fbdev, helper);
 	struct intel_framebuffer *intel_fb = ifbdev->fb;
 	struct drm_device *dev = helper->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct fb_info *info;
 	struct drm_framebuffer *fb;
 	struct drm_i915_gem_object *obj;
@@ -220,7 +222,7 @@ static int intelfb_create(struct drm_fb_helper *helper,
 	 * This also validates that any existing fb inherited from the
 	 * BIOS is suitable for own access.
 	 */
-	ret = intel_pin_and_fence_fb_obj(NULL, &ifbdev->fb->base, NULL);
+	ret = intel_pin_and_fence_fb_obj(&ifbdev->fb->base, BIT(DRM_ROTATE_0));
 	if (ret)
 		goto out_unlock;
 
@@ -244,13 +246,13 @@ static int intelfb_create(struct drm_fb_helper *helper,
 
 	/* setup aperture base/size for vesafb takeover */
 	info->apertures->ranges[0].base = dev->mode_config.fb_base;
-	info->apertures->ranges[0].size = dev_priv->gtt.mappable_end;
+	info->apertures->ranges[0].size = ggtt->mappable_end;
 
 	info->fix.smem_start = dev->mode_config.fb_base + i915_gem_obj_ggtt_offset(obj);
 	info->fix.smem_len = size;
 
 	info->screen_base =
-		ioremap_wc(dev_priv->gtt.mappable_base + i915_gem_obj_ggtt_offset(obj),
+		ioremap_wc(ggtt->mappable_base + i915_gem_obj_ggtt_offset(obj),
 			   size);
 	if (!info->screen_base) {
 		DRM_ERROR("Failed to remap framebuffer into virtual memory\n");
@@ -366,12 +368,12 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 	uint64_t conn_configured = 0, mask;
 	int pass = 0;
 
-	save_enabled = kcalloc(dev->mode_config.num_connector, sizeof(bool),
+	save_enabled = kcalloc(fb_helper->connector_count, sizeof(bool),
 			       GFP_KERNEL);
 	if (!save_enabled)
 		return false;
 
-	memcpy(save_enabled, enabled, dev->mode_config.num_connector);
+	memcpy(save_enabled, enabled, fb_helper->connector_count);
 	mask = (1 << fb_helper->connector_count) - 1;
 retry:
 	for (i = 0; i < fb_helper->connector_count; i++) {
@@ -379,6 +381,7 @@ retry:
 		struct drm_connector *connector;
 		struct drm_encoder *encoder;
 		struct drm_fb_helper_crtc *new_crtc;
+		struct intel_crtc *intel_crtc;
 
 		fb_conn = fb_helper->connector_info[i];
 		connector = fb_conn->connector;
@@ -420,6 +423,13 @@ retry:
 
 		num_connectors_enabled++;
 
+		intel_crtc = to_intel_crtc(connector->state->crtc);
+		for (j = 0; j < 256; j++) {
+			intel_crtc->lut_r[j] = j;
+			intel_crtc->lut_g[j] = j;
+			intel_crtc->lut_b[j] = j;
+		}
+
 		new_crtc = intel_fb_helper_crtc(fb_helper, connector->state->crtc);
 
 		/*
@@ -510,7 +520,7 @@ retry:
 	if (fallback) {
 bail:
 		DRM_DEBUG_KMS("Not using firmware configuration\n");
-		memcpy(enabled, save_enabled, dev->mode_config.num_connector);
+		memcpy(enabled, save_enabled, fb_helper->connector_count);
 		kfree(save_enabled);
 		return false;
 	}
diff --git a/drivers/gpu/drm/i915/intel_fifo_underrun.c b/drivers/gpu/drm/i915/intel_fifo_underrun.c
index bda526660e20..9be839a242f9 100644
--- a/drivers/gpu/drm/i915/intel_fifo_underrun.c
+++ b/drivers/gpu/drm/i915/intel_fifo_underrun.c
@@ -212,7 +212,7 @@ static void cpt_check_pch_fifo_underruns(struct intel_crtc *crtc)
 	I915_WRITE(SERR_INT, SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder));
 	POSTING_READ(SERR_INT);
 
-	DRM_ERROR("pch fifo underrun on pch transcoder %c\n",
+	DRM_ERROR("pch fifo underrun on pch transcoder %s\n",
 		  transcoder_name(pch_transcoder));
 }
 
@@ -235,7 +235,7 @@ static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
 
 		if (old && I915_READ(SERR_INT) &
 		    SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder)) {
-			DRM_ERROR("uncleared pch fifo underrun on pch transcoder %c\n",
+			DRM_ERROR("uncleared pch fifo underrun on pch transcoder %s\n",
 				  transcoder_name(pch_transcoder));
 		}
 	}
@@ -333,7 +333,7 @@ bool intel_set_pch_fifo_underrun_reporting(struct drm_i915_private *dev_priv,
 	old = !intel_crtc->pch_fifo_underrun_disabled;
 	intel_crtc->pch_fifo_underrun_disabled = !enable;
 
-	if (HAS_PCH_IBX(dev_priv->dev))
+	if (HAS_PCH_IBX(dev_priv))
 		ibx_set_fifo_underrun_reporting(dev_priv->dev, pch_transcoder,
 						enable);
 	else
@@ -363,7 +363,7 @@ void intel_cpu_fifo_underrun_irq_handler(struct drm_i915_private *dev_priv,
 		return;
 
 	/* GMCH can't disable fifo underruns, filter them. */
-	if (HAS_GMCH_DISPLAY(dev_priv->dev) &&
+	if (HAS_GMCH_DISPLAY(dev_priv) &&
 	    to_intel_crtc(crtc)->cpu_fifo_underrun_disabled)
 		return;
 
@@ -386,7 +386,7 @@ void intel_pch_fifo_underrun_irq_handler(struct drm_i915_private *dev_priv,
 {
 	if (intel_set_pch_fifo_underrun_reporting(dev_priv, pch_transcoder,
 						  false))
-		DRM_ERROR("PCH transcoder %c FIFO underrun\n",
+		DRM_ERROR("PCH transcoder %s FIFO underrun\n",
 			  transcoder_name(pch_transcoder));
 }
 
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index 73002e901ff2..9d79c4c3e256 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -27,8 +27,34 @@
 #include "intel_guc_fwif.h"
 #include "i915_guc_reg.h"
 
+struct drm_i915_gem_request;
+
+/*
+ * This structure primarily describes the GEM object shared with the GuC.
+ * The GEM object is held for the entire lifetime of our interaction with
+ * the GuC, being allocated before the GuC is loaded with its firmware.
+ * Because there's no way to update the address used by the GuC after
+ * initialisation, the shared object must stay pinned into the GGTT as
+ * long as the GuC is in use. We also keep the first page (only) mapped
+ * into kernel address space, as it includes shared data that must be
+ * updated on every request submission.
+ *
+ * The single GEM object described here is actually made up of several
+ * separate areas, as far as the GuC is concerned. The first page (kept
+ * kmap'd) includes the "process decriptor" which holds sequence data for
+ * the doorbell, and one cacheline which actually *is* the doorbell; a
+ * write to this will "ring the doorbell" (i.e. send an interrupt to the
+ * GuC). The subsequent  pages of the client object constitute the work
+ * queue (a circular array of work items), again described in the process
+ * descriptor. Work queue pages are mapped momentarily as required.
+ *
+ * Finally, we also keep a few statistics here, including the number of
+ * submissions to each engine, and a record of the last submission failure
+ * (if any).
+ */
 struct i915_guc_client {
 	struct drm_i915_gem_object *client_obj;
+	void *client_base;		/* first page (only) of above	*/
 	struct intel_context *owner;
 	struct intel_guc *guc;
 	uint32_t priority;
@@ -43,13 +69,14 @@ struct i915_guc_client {
 	uint32_t wq_offset;
 	uint32_t wq_size;
 	uint32_t wq_tail;
-	uint32_t wq_head;
+	uint32_t unused;		/* Was 'wq_head'		*/
 
 	/* GuC submission statistics & status */
 	uint64_t submissions[GUC_MAX_ENGINES_NUM];
 	uint32_t q_fail;
 	uint32_t b_fail;
 	int retcode;
+	int spare;			/* pad to 32 DWords		*/
 };
 
 enum intel_guc_fw_status {
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 82a3c03fbc0e..876e5da44c4e 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -59,7 +59,7 @@
  *
  */
 
-#define I915_SKL_GUC_UCODE "i915/skl_guc_ver4.bin"
+#define I915_SKL_GUC_UCODE "i915/skl_guc_ver6.bin"
 MODULE_FIRMWARE(I915_SKL_GUC_UCODE);
 
 /* User-friendly representation of an enum */
@@ -81,14 +81,14 @@ const char *intel_guc_fw_status_repr(enum intel_guc_fw_status status)
 
 static void direct_interrupts_to_host(struct drm_i915_private *dev_priv)
 {
-	struct intel_engine_cs *ring;
-	int i, irqs;
+	struct intel_engine_cs *engine;
+	int irqs;
 
 	/* tell all command streamers NOT to forward interrupts and vblank to GuC */
 	irqs = _MASKED_FIELD(GFX_FORWARD_VBLANK_MASK, GFX_FORWARD_VBLANK_NEVER);
 	irqs |= _MASKED_BIT_DISABLE(GFX_INTERRUPT_STEERING);
-	for_each_ring(ring, dev_priv, i)
-		I915_WRITE(RING_MODE_GEN7(ring), irqs);
+	for_each_engine(engine, dev_priv)
+		I915_WRITE(RING_MODE_GEN7(engine), irqs);
 
 	/* route all GT interrupts to the host */
 	I915_WRITE(GUC_BCS_RCS_IER, 0);
@@ -98,14 +98,14 @@ static void direct_interrupts_to_host(struct drm_i915_private *dev_priv)
 
 static void direct_interrupts_to_guc(struct drm_i915_private *dev_priv)
 {
-	struct intel_engine_cs *ring;
-	int i, irqs;
+	struct intel_engine_cs *engine;
+	int irqs;
 
 	/* tell all command streamers to forward interrupts and vblank to GuC */
 	irqs = _MASKED_FIELD(GFX_FORWARD_VBLANK_MASK, GFX_FORWARD_VBLANK_ALWAYS);
 	irqs |= _MASKED_BIT_ENABLE(GFX_INTERRUPT_STEERING);
-	for_each_ring(ring, dev_priv, i)
-		I915_WRITE(RING_MODE_GEN7(ring), irqs);
+	for_each_engine(engine, dev_priv)
+		I915_WRITE(RING_MODE_GEN7(engine), irqs);
 
 	/* route USER_INTERRUPT to Host, all others are sent to GuC. */
 	irqs = GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
@@ -353,6 +353,24 @@ static int guc_ucode_xfer(struct drm_i915_private *dev_priv)
 	return ret;
 }
 
+static int i915_reset_guc(struct drm_i915_private *dev_priv)
+{
+	int ret;
+	u32 guc_status;
+
+	ret = intel_guc_reset(dev_priv);
+	if (ret) {
+		DRM_ERROR("GuC reset failed, ret = %d\n", ret);
+		return ret;
+	}
+
+	guc_status = I915_READ(GUC_STATUS);
+	WARN(!(guc_status & GS_MIA_IN_RESET),
+	     "GuC status: 0x%x, MIA core expected to be in reset\n", guc_status);
+
+	return ret;
+}
+
 /**
  * intel_guc_ucode_load() - load GuC uCode into the device
  * @dev:	drm device
@@ -369,7 +387,7 @@ int intel_guc_ucode_load(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
-	int err = 0;
+	int retries, err = 0;
 
 	if (!i915.enable_guc_submission)
 		return 0;
@@ -417,9 +435,33 @@ int intel_guc_ucode_load(struct drm_device *dev)
 	if (err)
 		goto fail;
 
-	err = guc_ucode_xfer(dev_priv);
-	if (err)
-		goto fail;
+	/*
+	 * WaEnableuKernelHeaderValidFix:skl,bxt
+	 * For BXT, this is only upto B0 but below WA is required for later
+	 * steppings also so this is extended as well.
+	 */
+	/* WaEnableGuCBootHashCheckNotSet:skl,bxt */
+	for (retries = 3; ; ) {
+		/*
+		 * Always reset the GuC just before (re)loading, so
+		 * that the state and timing are fairly predictable
+		 */
+		err = i915_reset_guc(dev_priv);
+		if (err) {
+			DRM_ERROR("GuC reset failed, err %d\n", err);
+			goto fail;
+		}
+
+		err = guc_ucode_xfer(dev_priv);
+		if (!err)
+			break;
+
+		if (--retries == 0)
+			goto fail;
+
+		DRM_INFO("GuC fw load failed, err %d; will reset and "
+			"retry %d more time(s)\n", err, retries);
+	}
 
 	guc_fw->guc_fw_load_status = GUC_FIRMWARE_SUCCESS;
 
@@ -440,6 +482,7 @@ int intel_guc_ucode_load(struct drm_device *dev)
 	return 0;
 
 fail:
+	DRM_ERROR("GuC firmware load failed, err %d\n", err);
 	if (guc_fw->guc_fw_load_status == GUC_FIRMWARE_PENDING)
 		guc_fw->guc_fw_load_status = GUC_FIRMWARE_FAIL;
 
@@ -595,8 +638,8 @@ void intel_guc_ucode_init(struct drm_device *dev)
 		fw_path = NULL;
 	} else if (IS_SKYLAKE(dev)) {
 		fw_path = I915_SKL_GUC_UCODE;
-		guc_fw->guc_fw_major_wanted = 4;
-		guc_fw->guc_fw_minor_wanted = 3;
+		guc_fw->guc_fw_major_wanted = 6;
+		guc_fw->guc_fw_minor_wanted = 1;
 	} else {
 		i915.enable_guc_submission = false;
 		fw_path = "";	/* unknown device */
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 1ab6f687f640..2cdab73046f8 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -638,7 +638,7 @@ static bool intel_hdmi_set_gcp_infoframe(struct drm_encoder *encoder)
 		reg = HSW_TVIDEO_DIP_GCP(crtc->config->cpu_transcoder);
 	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 		reg = VLV_TVIDEO_DIP_GCP(crtc->pipe);
-	else if (HAS_PCH_SPLIT(dev_priv->dev))
+	else if (HAS_PCH_SPLIT(dev_priv))
 		reg = TVIDEO_DIP_GCP(crtc->pipe);
 	else
 		return false;
@@ -952,9 +952,6 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 	if (pipe_config->pixel_multiplier)
 		dotclock /= pipe_config->pixel_multiplier;
 
-	if (HAS_PCH_SPLIT(dev_priv->dev))
-		ironlake_check_encoder_dotclock(pipe_config, dotclock);
-
 	pipe_config->base.adjusted_mode.crtc_clock = dotclock;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index 52fbe530fc9e..81de23098be7 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -124,7 +124,7 @@ static void intel_i2c_quirk_set(struct drm_i915_private *dev_priv, bool enable)
 	u32 val;
 
 	/* When using bit bashing for I2C, this bit needs to be set to 1 */
-	if (!IS_PINEVIEW(dev_priv->dev))
+	if (!IS_PINEVIEW(dev_priv))
 		return;
 
 	val = I915_READ(DSPCLK_GATE_D);
@@ -264,7 +264,7 @@ gmbus_wait_hw_status(struct drm_i915_private *dev_priv,
 	u32 gmbus2 = 0;
 	DEFINE_WAIT(wait);
 
-	if (!HAS_GMBUS_IRQ(dev_priv->dev))
+	if (!HAS_GMBUS_IRQ(dev_priv))
 		gmbus4_irq_en = 0;
 
 	/* Important: The hw handles only the first bit, so set only one! Since
@@ -300,7 +300,7 @@ gmbus_wait_idle(struct drm_i915_private *dev_priv)
 
 #define C ((I915_READ_NOTRACE(GMBUS2) & GMBUS_ACTIVE) == 0)
 
-	if (!HAS_GMBUS_IRQ(dev_priv->dev))
+	if (!HAS_GMBUS_IRQ(dev_priv))
 		return wait_for(C, 10);
 
 	/* Important: The hw handles only the first bit, so set only one! */
@@ -571,15 +571,14 @@ clear_err:
 	goto out;
 
 timeout:
-	DRM_INFO("GMBUS [%s] timed out, falling back to bit banging on pin %d\n",
-		 bus->adapter.name, bus->reg0 & 0xff);
+	DRM_DEBUG_KMS("GMBUS [%s] timed out, falling back to bit banging on pin %d\n",
+		      bus->adapter.name, bus->reg0 & 0xff);
 	I915_WRITE(GMBUS0, 0);
 
 	/*
 	 * Hardware may not support GMBUS over these pins? Try GPIO bitbanging
 	 * instead. Use EAGAIN to have i2c core retry.
 	 */
-	bus->force_bit = 1;
 	ret = -EAGAIN;
 
 out:
@@ -597,10 +596,15 @@ gmbus_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs, int num)
 	intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 	mutex_lock(&dev_priv->gmbus_mutex);
 
-	if (bus->force_bit)
+	if (bus->force_bit) {
 		ret = i2c_bit_algo.master_xfer(adapter, msgs, num);
-	else
+		if (ret < 0)
+			bus->force_bit &= ~GMBUS_FORCE_BIT_RETRY;
+	} else {
 		ret = do_gmbus_xfer(adapter, msgs, num);
+		if (ret == -EAGAIN)
+			bus->force_bit |= GMBUS_FORCE_BIT_RETRY;
+	}
 
 	mutex_unlock(&dev_priv->gmbus_mutex);
 	intel_display_power_put(dev_priv, POWER_DOMAIN_GMBUS);
@@ -718,11 +722,16 @@ void intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed)
 void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit)
 {
 	struct intel_gmbus *bus = to_intel_gmbus(adapter);
+	struct drm_i915_private *dev_priv = bus->dev_priv;
+
+	mutex_lock(&dev_priv->gmbus_mutex);
 
 	bus->force_bit += force_bit ? 1 : -1;
 	DRM_DEBUG_KMS("%sabling bit-banging on %s. force bit now %d\n",
 		      force_bit ? "en" : "dis", adapter->name,
 		      bus->force_bit);
+
+	mutex_unlock(&dev_priv->gmbus_mutex);
 }
 
 void intel_teardown_gmbus(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 5c6080fd0968..6179b591ee84 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -131,6 +131,7 @@
  * preemption, but just sampling the new tail pointer).
  *
  */
+#include <linux/interrupt.h>
 
 #include <drm/drmP.h>
 #include <drm/i915_drm.h>
@@ -228,9 +229,6 @@ enum {
 
 static int intel_lr_context_pin(struct intel_context *ctx,
 				struct intel_engine_cs *engine);
-static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
-		struct drm_i915_gem_object *default_ctx_obj);
-
 
 /**
  * intel_sanitize_enable_execlists() - sanitize i915.enable_execlists
@@ -266,20 +264,23 @@ int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists
 }
 
 static void
-logical_ring_init_platform_invariants(struct intel_engine_cs *ring)
+logical_ring_init_platform_invariants(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
+
+	if (IS_GEN8(dev) || IS_GEN9(dev))
+		engine->idle_lite_restore_wa = ~0;
 
-	ring->disable_lite_restore_wa = (IS_SKL_REVID(dev, 0, SKL_REVID_B0) ||
+	engine->disable_lite_restore_wa = (IS_SKL_REVID(dev, 0, SKL_REVID_B0) ||
 					IS_BXT_REVID(dev, 0, BXT_REVID_A1)) &&
-					(ring->id == VCS || ring->id == VCS2);
+					(engine->id == VCS || engine->id == VCS2);
 
-	ring->ctx_desc_template = GEN8_CTX_VALID;
-	ring->ctx_desc_template |= GEN8_CTX_ADDRESSING_MODE(dev) <<
+	engine->ctx_desc_template = GEN8_CTX_VALID;
+	engine->ctx_desc_template |= GEN8_CTX_ADDRESSING_MODE(dev) <<
 				   GEN8_CTX_ADDRESSING_MODE_SHIFT;
 	if (IS_GEN8(dev))
-		ring->ctx_desc_template |= GEN8_CTX_L3LLC_COHERENT;
-	ring->ctx_desc_template |= GEN8_CTX_PRIVILEGE;
+		engine->ctx_desc_template |= GEN8_CTX_L3LLC_COHERENT;
+	engine->ctx_desc_template |= GEN8_CTX_PRIVILEGE;
 
 	/* TODO: WaDisableLiteRestore when we start using semaphore
 	 * signalling between Command Streamers */
@@ -287,8 +288,8 @@ logical_ring_init_platform_invariants(struct intel_engine_cs *ring)
 
 	/* WaEnableForceRestoreInCtxtDescForVCS:skl */
 	/* WaEnableForceRestoreInCtxtDescForVCS:bxt */
-	if (ring->disable_lite_restore_wa)
-		ring->ctx_desc_template |= GEN8_CTX_FORCE_RESTORE;
+	if (engine->disable_lite_restore_wa)
+		engine->ctx_desc_template |= GEN8_CTX_FORCE_RESTORE;
 }
 
 /**
@@ -311,24 +312,24 @@ logical_ring_init_platform_invariants(struct intel_engine_cs *ring)
  */
 static void
 intel_lr_context_descriptor_update(struct intel_context *ctx,
-				   struct intel_engine_cs *ring)
+				   struct intel_engine_cs *engine)
 {
 	uint64_t lrca, desc;
 
-	lrca = ctx->engine[ring->id].lrc_vma->node.start +
+	lrca = ctx->engine[engine->id].lrc_vma->node.start +
 	       LRC_PPHWSP_PN * PAGE_SIZE;
 
-	desc = ring->ctx_desc_template;			   /* bits  0-11 */
+	desc = engine->ctx_desc_template;			   /* bits  0-11 */
 	desc |= lrca;					   /* bits 12-31 */
 	desc |= (lrca >> PAGE_SHIFT) << GEN8_CTX_ID_SHIFT; /* bits 32-51 */
 
-	ctx->engine[ring->id].lrc_desc = desc;
+	ctx->engine[engine->id].lrc_desc = desc;
 }
 
 uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
-				     struct intel_engine_cs *ring)
+				     struct intel_engine_cs *engine)
 {
-	return ctx->engine[ring->id].lrc_desc;
+	return ctx->engine[engine->id].lrc_desc;
 }
 
 /**
@@ -348,98 +349,103 @@ uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
  * Return: 20-bits globally unique context ID.
  */
 u32 intel_execlists_ctx_id(struct intel_context *ctx,
-			   struct intel_engine_cs *ring)
+			   struct intel_engine_cs *engine)
 {
-	return intel_lr_context_descriptor(ctx, ring) >> GEN8_CTX_ID_SHIFT;
+	return intel_lr_context_descriptor(ctx, engine) >> GEN8_CTX_ID_SHIFT;
 }
 
 static void execlists_elsp_write(struct drm_i915_gem_request *rq0,
 				 struct drm_i915_gem_request *rq1)
 {
 
-	struct intel_engine_cs *ring = rq0->ring;
-	struct drm_device *dev = ring->dev;
+	struct intel_engine_cs *engine = rq0->engine;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint64_t desc[2];
 
 	if (rq1) {
-		desc[1] = intel_lr_context_descriptor(rq1->ctx, rq1->ring);
+		desc[1] = intel_lr_context_descriptor(rq1->ctx, rq1->engine);
 		rq1->elsp_submitted++;
 	} else {
 		desc[1] = 0;
 	}
 
-	desc[0] = intel_lr_context_descriptor(rq0->ctx, rq0->ring);
+	desc[0] = intel_lr_context_descriptor(rq0->ctx, rq0->engine);
 	rq0->elsp_submitted++;
 
 	/* You must always write both descriptors in the order below. */
-	spin_lock(&dev_priv->uncore.lock);
-	intel_uncore_forcewake_get__locked(dev_priv, FORCEWAKE_ALL);
-	I915_WRITE_FW(RING_ELSP(ring), upper_32_bits(desc[1]));
-	I915_WRITE_FW(RING_ELSP(ring), lower_32_bits(desc[1]));
+	I915_WRITE_FW(RING_ELSP(engine), upper_32_bits(desc[1]));
+	I915_WRITE_FW(RING_ELSP(engine), lower_32_bits(desc[1]));
 
-	I915_WRITE_FW(RING_ELSP(ring), upper_32_bits(desc[0]));
+	I915_WRITE_FW(RING_ELSP(engine), upper_32_bits(desc[0]));
 	/* The context is automatically loaded after the following */
-	I915_WRITE_FW(RING_ELSP(ring), lower_32_bits(desc[0]));
+	I915_WRITE_FW(RING_ELSP(engine), lower_32_bits(desc[0]));
 
 	/* ELSP is a wo register, use another nearby reg for posting */
-	POSTING_READ_FW(RING_EXECLIST_STATUS_LO(ring));
-	intel_uncore_forcewake_put__locked(dev_priv, FORCEWAKE_ALL);
-	spin_unlock(&dev_priv->uncore.lock);
+	POSTING_READ_FW(RING_EXECLIST_STATUS_LO(engine));
+}
+
+static void
+execlists_update_context_pdps(struct i915_hw_ppgtt *ppgtt, u32 *reg_state)
+{
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 0);
 }
 
-static int execlists_update_context(struct drm_i915_gem_request *rq)
+static void execlists_update_context(struct drm_i915_gem_request *rq)
 {
-	struct intel_engine_cs *ring = rq->ring;
+	struct intel_engine_cs *engine = rq->engine;
 	struct i915_hw_ppgtt *ppgtt = rq->ctx->ppgtt;
-	uint32_t *reg_state = rq->ctx->engine[ring->id].lrc_reg_state;
+	uint32_t *reg_state = rq->ctx->engine[engine->id].lrc_reg_state;
 
 	reg_state[CTX_RING_TAIL+1] = rq->tail;
 
-	if (ppgtt && !USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
-		/* True 32b PPGTT with dynamic page allocation: update PDP
-		 * registers and point the unallocated PDPs to scratch page.
-		 * PML4 is allocated during ppgtt init, so this is not needed
-		 * in 48-bit mode.
-		 */
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 0);
-	}
-
-	return 0;
+	/* True 32b PPGTT with dynamic page allocation: update PDP
+	 * registers and point the unallocated PDPs to scratch page.
+	 * PML4 is allocated during ppgtt init, so this is not needed
+	 * in 48-bit mode.
+	 */
+	if (ppgtt && !USES_FULL_48BIT_PPGTT(ppgtt->base.dev))
+		execlists_update_context_pdps(ppgtt, reg_state);
 }
 
 static void execlists_submit_requests(struct drm_i915_gem_request *rq0,
 				      struct drm_i915_gem_request *rq1)
 {
+	struct drm_i915_private *dev_priv = rq0->i915;
+	unsigned int fw_domains = rq0->engine->fw_domains;
+
 	execlists_update_context(rq0);
 
 	if (rq1)
 		execlists_update_context(rq1);
 
+	spin_lock_irq(&dev_priv->uncore.lock);
+	intel_uncore_forcewake_get__locked(dev_priv, fw_domains);
+
 	execlists_elsp_write(rq0, rq1);
+
+	intel_uncore_forcewake_put__locked(dev_priv, fw_domains);
+	spin_unlock_irq(&dev_priv->uncore.lock);
 }
 
-static void execlists_context_unqueue(struct intel_engine_cs *ring)
+static void execlists_context_unqueue(struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_request *req0 = NULL, *req1 = NULL;
-	struct drm_i915_gem_request *cursor = NULL, *tmp = NULL;
+	struct drm_i915_gem_request *cursor, *tmp;
 
-	assert_spin_locked(&ring->execlist_lock);
+	assert_spin_locked(&engine->execlist_lock);
 
 	/*
 	 * If irqs are not active generate a warning as batches that finish
 	 * without the irqs may get lost and a GPU Hang may occur.
 	 */
-	WARN_ON(!intel_irqs_enabled(ring->dev->dev_private));
-
-	if (list_empty(&ring->execlist_queue))
-		return;
+	WARN_ON(!intel_irqs_enabled(engine->dev->dev_private));
 
 	/* Try to read in pairs */
-	list_for_each_entry_safe(cursor, tmp, &ring->execlist_queue,
+	list_for_each_entry_safe(cursor, tmp, &engine->execlist_queue,
 				 execlist_link) {
 		if (!req0) {
 			req0 = cursor;
@@ -448,172 +454,179 @@ static void execlists_context_unqueue(struct intel_engine_cs *ring)
 			 * will update tail past first request's workload */
 			cursor->elsp_submitted = req0->elsp_submitted;
 			list_move_tail(&req0->execlist_link,
-				       &ring->execlist_retired_req_list);
+				       &engine->execlist_retired_req_list);
 			req0 = cursor;
 		} else {
 			req1 = cursor;
+			WARN_ON(req1->elsp_submitted);
 			break;
 		}
 	}
 
-	if (IS_GEN8(ring->dev) || IS_GEN9(ring->dev)) {
+	if (unlikely(!req0))
+		return;
+
+	if (req0->elsp_submitted & engine->idle_lite_restore_wa) {
 		/*
-		 * WaIdleLiteRestore: make sure we never cause a lite
-		 * restore with HEAD==TAIL
+		 * WaIdleLiteRestore: make sure we never cause a lite restore
+		 * with HEAD==TAIL.
+		 *
+		 * Apply the wa NOOPS to prevent ring:HEAD == req:TAIL as we
+		 * resubmit the request. See gen8_emit_request() for where we
+		 * prepare the padding after the end of the request.
 		 */
-		if (req0->elsp_submitted) {
-			/*
-			 * Apply the wa NOOPS to prevent ring:HEAD == req:TAIL
-			 * as we resubmit the request. See gen8_emit_request()
-			 * for where we prepare the padding after the end of the
-			 * request.
-			 */
-			struct intel_ringbuffer *ringbuf;
+		struct intel_ringbuffer *ringbuf;
 
-			ringbuf = req0->ctx->engine[ring->id].ringbuf;
-			req0->tail += 8;
-			req0->tail &= ringbuf->size - 1;
-		}
+		ringbuf = req0->ctx->engine[engine->id].ringbuf;
+		req0->tail += 8;
+		req0->tail &= ringbuf->size - 1;
 	}
 
-	WARN_ON(req1 && req1->elsp_submitted);
-
 	execlists_submit_requests(req0, req1);
 }
 
-static bool execlists_check_remove_request(struct intel_engine_cs *ring,
-					   u32 request_id)
+static unsigned int
+execlists_check_remove_request(struct intel_engine_cs *engine, u32 request_id)
 {
 	struct drm_i915_gem_request *head_req;
 
-	assert_spin_locked(&ring->execlist_lock);
+	assert_spin_locked(&engine->execlist_lock);
 
-	head_req = list_first_entry_or_null(&ring->execlist_queue,
+	head_req = list_first_entry_or_null(&engine->execlist_queue,
 					    struct drm_i915_gem_request,
 					    execlist_link);
 
-	if (head_req != NULL) {
-		if (intel_execlists_ctx_id(head_req->ctx, ring) == request_id) {
-			WARN(head_req->elsp_submitted == 0,
-			     "Never submitted head request\n");
+	if (!head_req)
+		return 0;
 
-			if (--head_req->elsp_submitted <= 0) {
-				list_move_tail(&head_req->execlist_link,
-					       &ring->execlist_retired_req_list);
-				return true;
-			}
-		}
-	}
+	if (unlikely(intel_execlists_ctx_id(head_req->ctx, engine) != request_id))
+		return 0;
+
+	WARN(head_req->elsp_submitted == 0, "Never submitted head request\n");
+
+	if (--head_req->elsp_submitted > 0)
+		return 0;
+
+	list_move_tail(&head_req->execlist_link,
+		       &engine->execlist_retired_req_list);
 
-	return false;
+	return 1;
 }
 
-static void get_context_status(struct intel_engine_cs *ring,
-			       u8 read_pointer,
-			       u32 *status, u32 *context_id)
+static u32
+get_context_status(struct intel_engine_cs *engine, unsigned int read_pointer,
+		   u32 *context_id)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
+	u32 status;
 
-	if (WARN_ON(read_pointer >= GEN8_CSB_ENTRIES))
-		return;
+	read_pointer %= GEN8_CSB_ENTRIES;
+
+	status = I915_READ_FW(RING_CONTEXT_STATUS_BUF_LO(engine, read_pointer));
+
+	if (status & GEN8_CTX_STATUS_IDLE_ACTIVE)
+		return 0;
+
+	*context_id = I915_READ_FW(RING_CONTEXT_STATUS_BUF_HI(engine,
+							      read_pointer));
 
-	*status = I915_READ(RING_CONTEXT_STATUS_BUF_LO(ring, read_pointer));
-	*context_id = I915_READ(RING_CONTEXT_STATUS_BUF_HI(ring, read_pointer));
+	return status;
 }
 
 /**
  * intel_lrc_irq_handler() - handle Context Switch interrupts
- * @ring: Engine Command Streamer to handle.
+ * @engine: Engine Command Streamer to handle.
  *
  * Check the unread Context Status Buffers and manage the submission of new
  * contexts to the ELSP accordingly.
  */
-void intel_lrc_irq_handler(struct intel_engine_cs *ring)
+static void intel_lrc_irq_handler(unsigned long data)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct intel_engine_cs *engine = (struct intel_engine_cs *)data;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	u32 status_pointer;
-	u8 read_pointer;
-	u8 write_pointer;
-	u32 status = 0;
-	u32 status_id;
-	u32 submit_contexts = 0;
+	unsigned int read_pointer, write_pointer;
+	u32 csb[GEN8_CSB_ENTRIES][2];
+	unsigned int csb_read = 0, i;
+	unsigned int submit_contexts = 0;
 
-	status_pointer = I915_READ(RING_CONTEXT_STATUS_PTR(ring));
+	intel_uncore_forcewake_get(dev_priv, engine->fw_domains);
 
-	read_pointer = ring->next_context_status_buffer;
+	status_pointer = I915_READ_FW(RING_CONTEXT_STATUS_PTR(engine));
+
+	read_pointer = engine->next_context_status_buffer;
 	write_pointer = GEN8_CSB_WRITE_PTR(status_pointer);
 	if (read_pointer > write_pointer)
 		write_pointer += GEN8_CSB_ENTRIES;
 
-	spin_lock(&ring->execlist_lock);
-
 	while (read_pointer < write_pointer) {
+		if (WARN_ON_ONCE(csb_read == GEN8_CSB_ENTRIES))
+			break;
+		csb[csb_read][0] = get_context_status(engine, ++read_pointer,
+						      &csb[csb_read][1]);
+		csb_read++;
+	}
 
-		get_context_status(ring, ++read_pointer % GEN8_CSB_ENTRIES,
-				   &status, &status_id);
+	engine->next_context_status_buffer = write_pointer % GEN8_CSB_ENTRIES;
 
-		if (status & GEN8_CTX_STATUS_IDLE_ACTIVE)
-			continue;
+	/* Update the read pointer to the old write pointer. Manual ringbuffer
+	 * management ftw </sarcasm> */
+	I915_WRITE_FW(RING_CONTEXT_STATUS_PTR(engine),
+		      _MASKED_FIELD(GEN8_CSB_READ_PTR_MASK,
+				    engine->next_context_status_buffer << 8));
 
-		if (status & GEN8_CTX_STATUS_PREEMPTED) {
-			if (status & GEN8_CTX_STATUS_LITE_RESTORE) {
-				if (execlists_check_remove_request(ring, status_id))
+	intel_uncore_forcewake_put(dev_priv, engine->fw_domains);
+
+	spin_lock(&engine->execlist_lock);
+
+	for (i = 0; i < csb_read; i++) {
+		if (unlikely(csb[i][0] & GEN8_CTX_STATUS_PREEMPTED)) {
+			if (csb[i][0] & GEN8_CTX_STATUS_LITE_RESTORE) {
+				if (execlists_check_remove_request(engine, csb[i][1]))
 					WARN(1, "Lite Restored request removed from queue\n");
 			} else
 				WARN(1, "Preemption without Lite Restore\n");
 		}
 
-		if ((status & GEN8_CTX_STATUS_ACTIVE_IDLE) ||
-		    (status & GEN8_CTX_STATUS_ELEMENT_SWITCH)) {
-			if (execlists_check_remove_request(ring, status_id))
-				submit_contexts++;
-		}
+		if (csb[i][0] & (GEN8_CTX_STATUS_ACTIVE_IDLE |
+		    GEN8_CTX_STATUS_ELEMENT_SWITCH))
+			submit_contexts +=
+				execlists_check_remove_request(engine, csb[i][1]);
 	}
 
-	if (ring->disable_lite_restore_wa) {
-		/* Prevent a ctx to preempt itself */
-		if ((status & GEN8_CTX_STATUS_ACTIVE_IDLE) &&
-		    (submit_contexts != 0))
-			execlists_context_unqueue(ring);
-	} else if (submit_contexts != 0) {
-		execlists_context_unqueue(ring);
+	if (submit_contexts) {
+		if (!engine->disable_lite_restore_wa ||
+		    (csb[i][0] & GEN8_CTX_STATUS_ACTIVE_IDLE))
+			execlists_context_unqueue(engine);
 	}
 
-	spin_unlock(&ring->execlist_lock);
+	spin_unlock(&engine->execlist_lock);
 
 	if (unlikely(submit_contexts > 2))
 		DRM_ERROR("More than two context complete events?\n");
-
-	ring->next_context_status_buffer = write_pointer % GEN8_CSB_ENTRIES;
-
-	/* Update the read pointer to the old write pointer. Manual ringbuffer
-	 * management ftw </sarcasm> */
-	I915_WRITE(RING_CONTEXT_STATUS_PTR(ring),
-		   _MASKED_FIELD(GEN8_CSB_READ_PTR_MASK,
-				 ring->next_context_status_buffer << 8));
 }
 
-static int execlists_context_queue(struct drm_i915_gem_request *request)
+static void execlists_context_queue(struct drm_i915_gem_request *request)
 {
-	struct intel_engine_cs *ring = request->ring;
+	struct intel_engine_cs *engine = request->engine;
 	struct drm_i915_gem_request *cursor;
 	int num_elements = 0;
 
 	if (request->ctx != request->i915->kernel_context)
-		intel_lr_context_pin(request->ctx, ring);
+		intel_lr_context_pin(request->ctx, engine);
 
 	i915_gem_request_reference(request);
 
-	spin_lock_irq(&ring->execlist_lock);
+	spin_lock_bh(&engine->execlist_lock);
 
-	list_for_each_entry(cursor, &ring->execlist_queue, execlist_link)
+	list_for_each_entry(cursor, &engine->execlist_queue, execlist_link)
 		if (++num_elements > 2)
 			break;
 
 	if (num_elements > 2) {
 		struct drm_i915_gem_request *tail_req;
 
-		tail_req = list_last_entry(&ring->execlist_queue,
+		tail_req = list_last_entry(&engine->execlist_queue,
 					   struct drm_i915_gem_request,
 					   execlist_link);
 
@@ -621,41 +634,39 @@ static int execlists_context_queue(struct drm_i915_gem_request *request)
 			WARN(tail_req->elsp_submitted != 0,
 				"More than 2 already-submitted reqs queued\n");
 			list_move_tail(&tail_req->execlist_link,
-				       &ring->execlist_retired_req_list);
+				       &engine->execlist_retired_req_list);
 		}
 	}
 
-	list_add_tail(&request->execlist_link, &ring->execlist_queue);
+	list_add_tail(&request->execlist_link, &engine->execlist_queue);
 	if (num_elements == 0)
-		execlists_context_unqueue(ring);
+		execlists_context_unqueue(engine);
 
-	spin_unlock_irq(&ring->execlist_lock);
-
-	return 0;
+	spin_unlock_bh(&engine->execlist_lock);
 }
 
 static int logical_ring_invalidate_all_caches(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	uint32_t flush_domains;
 	int ret;
 
 	flush_domains = 0;
-	if (ring->gpu_caches_dirty)
+	if (engine->gpu_caches_dirty)
 		flush_domains = I915_GEM_GPU_DOMAINS;
 
-	ret = ring->emit_flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
+	ret = engine->emit_flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
 	if (ret)
 		return ret;
 
-	ring->gpu_caches_dirty = false;
+	engine->gpu_caches_dirty = false;
 	return 0;
 }
 
 static int execlists_move_to_gpu(struct drm_i915_gem_request *req,
 				 struct list_head *vmas)
 {
-	const unsigned other_rings = ~intel_ring_flag(req->ring);
+	const unsigned other_rings = ~intel_engine_flag(req->engine);
 	struct i915_vma *vma;
 	uint32_t flush_domains = 0;
 	bool flush_chipset = false;
@@ -665,7 +676,7 @@ static int execlists_move_to_gpu(struct drm_i915_gem_request *req,
 		struct drm_i915_gem_object *obj = vma->obj;
 
 		if (obj->active & other_rings) {
-			ret = i915_gem_object_sync(obj, req->ring, &req);
+			ret = i915_gem_object_sync(obj, req->engine, &req);
 			if (ret)
 				return ret;
 		}
@@ -689,7 +700,7 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request
 {
 	int ret = 0;
 
-	request->ringbuf = request->ctx->engine[request->ring->id].ringbuf;
+	request->ringbuf = request->ctx->engine[request->engine->id].ringbuf;
 
 	if (i915.enable_guc_submission) {
 		/*
@@ -705,7 +716,7 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request
 	}
 
 	if (request->ctx != request->i915->kernel_context)
-		ret = intel_lr_context_pin(request->ctx, request->ring);
+		ret = intel_lr_context_pin(request->ctx, request->engine);
 
 	return ret;
 }
@@ -714,7 +725,7 @@ static int logical_ring_wait_for_space(struct drm_i915_gem_request *req,
 				       int bytes)
 {
 	struct intel_ringbuffer *ringbuf = req->ringbuf;
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct drm_i915_gem_request *target;
 	unsigned space;
 	int ret;
@@ -725,7 +736,7 @@ static int logical_ring_wait_for_space(struct drm_i915_gem_request *req,
 	/* The whole point of reserving space is to not wait! */
 	WARN_ON(ringbuf->reserved_in_use);
 
-	list_for_each_entry(target, &ring->request_list, list) {
+	list_for_each_entry(target, &engine->request_list, list) {
 		/*
 		 * The request queue is per-engine, so can contain requests
 		 * from multiple ringbuffers. Here, we must ignore any that
@@ -741,7 +752,7 @@ static int logical_ring_wait_for_space(struct drm_i915_gem_request *req,
 			break;
 	}
 
-	if (WARN_ON(&target->list == &ring->request_list))
+	if (WARN_ON(&target->list == &engine->request_list))
 		return -ENOSPC;
 
 	ret = i915_wait_request(target);
@@ -766,7 +777,7 @@ intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 {
 	struct intel_ringbuffer *ringbuf = request->ringbuf;
 	struct drm_i915_private *dev_priv = request->i915;
-	struct intel_engine_cs *engine = request->ring;
+	struct intel_engine_cs *engine = request->engine;
 
 	intel_logical_ring_advance(ringbuf);
 	request->tail = ringbuf->tail;
@@ -781,7 +792,7 @@ intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 	intel_logical_ring_emit(ringbuf, MI_NOOP);
 	intel_logical_ring_advance(ringbuf);
 
-	if (intel_ring_stopped(engine))
+	if (intel_engine_stopped(engine))
 		return 0;
 
 	if (engine->last_context != request->ctx) {
@@ -879,17 +890,8 @@ static int logical_ring_prepare(struct drm_i915_gem_request *req, int bytes)
  */
 int intel_logical_ring_begin(struct drm_i915_gem_request *req, int num_dwords)
 {
-	struct drm_i915_private *dev_priv;
 	int ret;
 
-	WARN_ON(req == NULL);
-	dev_priv = req->ring->dev->dev_private;
-
-	ret = i915_gem_check_wedge(&dev_priv->gpu_error,
-				   dev_priv->mm.interruptible);
-	if (ret)
-		return ret;
-
 	ret = logical_ring_prepare(req, num_dwords * sizeof(uint32_t));
 	if (ret)
 		return ret;
@@ -935,9 +937,9 @@ int intel_execlists_submission(struct i915_execbuffer_params *params,
 			       struct list_head *vmas)
 {
 	struct drm_device       *dev = params->dev;
-	struct intel_engine_cs  *ring = params->ring;
+	struct intel_engine_cs *engine = params->engine;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ringbuffer *ringbuf = params->ctx->engine[ring->id].ringbuf;
+	struct intel_ringbuffer *ringbuf = params->ctx->engine[engine->id].ringbuf;
 	u64 exec_start;
 	int instp_mode;
 	u32 instp_mask;
@@ -949,7 +951,7 @@ int intel_execlists_submission(struct i915_execbuffer_params *params,
 	case I915_EXEC_CONSTANTS_REL_GENERAL:
 	case I915_EXEC_CONSTANTS_ABSOLUTE:
 	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (instp_mode != 0 && ring != &dev_priv->ring[RCS]) {
+		if (instp_mode != 0 && engine != &dev_priv->engine[RCS]) {
 			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
 			return -EINVAL;
 		}
@@ -978,7 +980,7 @@ int intel_execlists_submission(struct i915_execbuffer_params *params,
 	if (ret)
 		return ret;
 
-	if (ring == &dev_priv->ring[RCS] &&
+	if (engine == &dev_priv->engine[RCS] &&
 	    instp_mode != dev_priv->relative_constants_mode) {
 		ret = intel_logical_ring_begin(params->request, 4);
 		if (ret)
@@ -996,116 +998,116 @@ int intel_execlists_submission(struct i915_execbuffer_params *params,
 	exec_start = params->batch_obj_vm_offset +
 		     args->batch_start_offset;
 
-	ret = ring->emit_bb_start(params->request, exec_start, params->dispatch_flags);
+	ret = engine->emit_bb_start(params->request, exec_start, params->dispatch_flags);
 	if (ret)
 		return ret;
 
 	trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
 
 	i915_gem_execbuffer_move_to_active(vmas, params->request);
-	i915_gem_execbuffer_retire_commands(params);
 
 	return 0;
 }
 
-void intel_execlists_retire_requests(struct intel_engine_cs *ring)
+void intel_execlists_retire_requests(struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_request *req, *tmp;
 	struct list_head retired_list;
 
-	WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
-	if (list_empty(&ring->execlist_retired_req_list))
+	WARN_ON(!mutex_is_locked(&engine->dev->struct_mutex));
+	if (list_empty(&engine->execlist_retired_req_list))
 		return;
 
 	INIT_LIST_HEAD(&retired_list);
-	spin_lock_irq(&ring->execlist_lock);
-	list_replace_init(&ring->execlist_retired_req_list, &retired_list);
-	spin_unlock_irq(&ring->execlist_lock);
+	spin_lock_bh(&engine->execlist_lock);
+	list_replace_init(&engine->execlist_retired_req_list, &retired_list);
+	spin_unlock_bh(&engine->execlist_lock);
 
 	list_for_each_entry_safe(req, tmp, &retired_list, execlist_link) {
 		struct intel_context *ctx = req->ctx;
 		struct drm_i915_gem_object *ctx_obj =
-				ctx->engine[ring->id].state;
+				ctx->engine[engine->id].state;
 
 		if (ctx_obj && (ctx != req->i915->kernel_context))
-			intel_lr_context_unpin(ctx, ring);
+			intel_lr_context_unpin(ctx, engine);
 
 		list_del(&req->execlist_link);
 		i915_gem_request_unreference(req);
 	}
 }
 
-void intel_logical_ring_stop(struct intel_engine_cs *ring)
+void intel_logical_ring_stop(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	int ret;
 
-	if (!intel_ring_initialized(ring))
+	if (!intel_engine_initialized(engine))
 		return;
 
-	ret = intel_ring_idle(ring);
-	if (ret && !i915_reset_in_progress(&to_i915(ring->dev)->gpu_error))
+	ret = intel_engine_idle(engine);
+	if (ret)
 		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
-			  ring->name, ret);
+			  engine->name, ret);
 
 	/* TODO: Is this correct with Execlists enabled? */
-	I915_WRITE_MODE(ring, _MASKED_BIT_ENABLE(STOP_RING));
-	if (wait_for_atomic((I915_READ_MODE(ring) & MODE_IDLE) != 0, 1000)) {
-		DRM_ERROR("%s :timed out trying to stop ring\n", ring->name);
+	I915_WRITE_MODE(engine, _MASKED_BIT_ENABLE(STOP_RING));
+	if (wait_for((I915_READ_MODE(engine) & MODE_IDLE) != 0, 1000)) {
+		DRM_ERROR("%s :timed out trying to stop ring\n", engine->name);
 		return;
 	}
-	I915_WRITE_MODE(ring, _MASKED_BIT_DISABLE(STOP_RING));
+	I915_WRITE_MODE(engine, _MASKED_BIT_DISABLE(STOP_RING));
 }
 
 int logical_ring_flush_all_caches(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
-	if (!ring->gpu_caches_dirty)
+	if (!engine->gpu_caches_dirty)
 		return 0;
 
-	ret = ring->emit_flush(req, 0, I915_GEM_GPU_DOMAINS);
+	ret = engine->emit_flush(req, 0, I915_GEM_GPU_DOMAINS);
 	if (ret)
 		return ret;
 
-	ring->gpu_caches_dirty = false;
+	engine->gpu_caches_dirty = false;
 	return 0;
 }
 
 static int intel_lr_context_do_pin(struct intel_context *ctx,
-				   struct intel_engine_cs *ring)
+				   struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
-	struct intel_ringbuffer *ringbuf = ctx->engine[ring->id].ringbuf;
-	struct page *lrc_state_page;
-	uint32_t *lrc_reg_state;
+	struct drm_i915_gem_object *ctx_obj = ctx->engine[engine->id].state;
+	struct intel_ringbuffer *ringbuf = ctx->engine[engine->id].ringbuf;
+	void *vaddr;
+	u32 *lrc_reg_state;
 	int ret;
 
-	WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
+	WARN_ON(!mutex_is_locked(&engine->dev->struct_mutex));
 
 	ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN,
 			PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
 	if (ret)
 		return ret;
 
-	lrc_state_page = i915_gem_object_get_dirty_page(ctx_obj, LRC_STATE_PN);
-	if (WARN_ON(!lrc_state_page)) {
-		ret = -ENODEV;
+	vaddr = i915_gem_object_pin_map(ctx_obj);
+	if (IS_ERR(vaddr)) {
+		ret = PTR_ERR(vaddr);
 		goto unpin_ctx_obj;
 	}
 
-	ret = intel_pin_and_map_ringbuffer_obj(ring->dev, ringbuf);
+	lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
+
+	ret = intel_pin_and_map_ringbuffer_obj(engine->dev, ringbuf);
 	if (ret)
-		goto unpin_ctx_obj;
+		goto unpin_map;
 
-	ctx->engine[ring->id].lrc_vma = i915_gem_obj_to_ggtt(ctx_obj);
-	intel_lr_context_descriptor_update(ctx, ring);
-	lrc_reg_state = kmap(lrc_state_page);
+	ctx->engine[engine->id].lrc_vma = i915_gem_obj_to_ggtt(ctx_obj);
+	intel_lr_context_descriptor_update(ctx, engine);
 	lrc_reg_state[CTX_RING_BUFFER_START+1] = ringbuf->vma->node.start;
-	ctx->engine[ring->id].lrc_reg_state = lrc_reg_state;
+	ctx->engine[engine->id].lrc_reg_state = lrc_reg_state;
 	ctx_obj->dirty = true;
 
 	/* Invalidate GuC TLB. */
@@ -1114,6 +1116,8 @@ static int intel_lr_context_do_pin(struct intel_context *ctx,
 
 	return ret;
 
+unpin_map:
+	i915_gem_object_unpin_map(ctx_obj);
 unpin_ctx_obj:
 	i915_gem_object_ggtt_unpin(ctx_obj);
 
@@ -1146,7 +1150,7 @@ void intel_lr_context_unpin(struct intel_context *ctx,
 
 	WARN_ON(!mutex_is_locked(&ctx->i915->dev->struct_mutex));
 	if (--ctx->engine[engine->id].pin_count == 0) {
-		kunmap(kmap_to_page(ctx->engine[engine->id].lrc_reg_state));
+		i915_gem_object_unpin_map(ctx_obj);
 		intel_unpin_ringbuffer_obj(ctx->engine[engine->id].ringbuf);
 		i915_gem_object_ggtt_unpin(ctx_obj);
 		ctx->engine[engine->id].lrc_vma = NULL;
@@ -1160,16 +1164,16 @@ void intel_lr_context_unpin(struct intel_context *ctx,
 static int intel_logical_ring_workarounds_emit(struct drm_i915_gem_request *req)
 {
 	int ret, i;
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct intel_ringbuffer *ringbuf = req->ringbuf;
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_workarounds *w = &dev_priv->workarounds;
 
 	if (w->count == 0)
 		return 0;
 
-	ring->gpu_caches_dirty = true;
+	engine->gpu_caches_dirty = true;
 	ret = logical_ring_flush_all_caches(req);
 	if (ret)
 		return ret;
@@ -1187,7 +1191,7 @@ static int intel_logical_ring_workarounds_emit(struct drm_i915_gem_request *req)
 
 	intel_logical_ring_advance(ringbuf);
 
-	ring->gpu_caches_dirty = true;
+	engine->gpu_caches_dirty = true;
 	ret = logical_ring_flush_all_caches(req);
 	if (ret)
 		return ret;
@@ -1223,7 +1227,7 @@ static int intel_logical_ring_workarounds_emit(struct drm_i915_gem_request *req)
  * This WA is also required for Gen9 so extracting as a function avoids
  * code duplication.
  */
-static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *ring,
+static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine,
 						uint32_t *const batch,
 						uint32_t index)
 {
@@ -1235,13 +1239,13 @@ static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *ring,
 	 * this batch updates GEN8_L3SQCREG4 with default value we need to
 	 * set this bit here to retain the WA during flush.
 	 */
-	if (IS_SKL_REVID(ring->dev, 0, SKL_REVID_E0))
+	if (IS_SKL_REVID(engine->dev, 0, SKL_REVID_E0))
 		l3sqc4_flush |= GEN8_LQSC_RO_PERF_DIS;
 
 	wa_ctx_emit(batch, index, (MI_STORE_REGISTER_MEM_GEN8 |
 				   MI_SRM_LRM_GLOBAL_GTT));
 	wa_ctx_emit_reg(batch, index, GEN8_L3SQCREG4);
-	wa_ctx_emit(batch, index, ring->scratch.gtt_offset + 256);
+	wa_ctx_emit(batch, index, engine->scratch.gtt_offset + 256);
 	wa_ctx_emit(batch, index, 0);
 
 	wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(1));
@@ -1259,7 +1263,7 @@ static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *ring,
 	wa_ctx_emit(batch, index, (MI_LOAD_REGISTER_MEM_GEN8 |
 				   MI_SRM_LRM_GLOBAL_GTT));
 	wa_ctx_emit_reg(batch, index, GEN8_L3SQCREG4);
-	wa_ctx_emit(batch, index, ring->scratch.gtt_offset + 256);
+	wa_ctx_emit(batch, index, engine->scratch.gtt_offset + 256);
 	wa_ctx_emit(batch, index, 0);
 
 	return index;
@@ -1312,7 +1316,7 @@ static inline int wa_ctx_end(struct i915_wa_ctx_bb *wa_ctx,
  * Return: non-zero if we exceed the PAGE_SIZE limit.
  */
 
-static int gen8_init_indirectctx_bb(struct intel_engine_cs *ring,
+static int gen8_init_indirectctx_bb(struct intel_engine_cs *engine,
 				    struct i915_wa_ctx_bb *wa_ctx,
 				    uint32_t *const batch,
 				    uint32_t *offset)
@@ -1324,8 +1328,8 @@ static int gen8_init_indirectctx_bb(struct intel_engine_cs *ring,
 	wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_DISABLE);
 
 	/* WaFlushCoherentL3CacheLinesAtContextSwitch:bdw */
-	if (IS_BROADWELL(ring->dev)) {
-		int rc = gen8_emit_flush_coherentl3_wa(ring, batch, index);
+	if (IS_BROADWELL(engine->dev)) {
+		int rc = gen8_emit_flush_coherentl3_wa(engine, batch, index);
 		if (rc < 0)
 			return rc;
 		index = rc;
@@ -1333,7 +1337,7 @@ static int gen8_init_indirectctx_bb(struct intel_engine_cs *ring,
 
 	/* WaClearSlmSpaceAtContextSwitch:bdw,chv */
 	/* Actual scratch location is at 128 bytes offset */
-	scratch_addr = ring->scratch.gtt_offset + 2*CACHELINE_BYTES;
+	scratch_addr = engine->scratch.gtt_offset + 2*CACHELINE_BYTES;
 
 	wa_ctx_emit(batch, index, GFX_OP_PIPE_CONTROL(6));
 	wa_ctx_emit(batch, index, (PIPE_CONTROL_FLUSH_L3 |
@@ -1375,7 +1379,7 @@ static int gen8_init_indirectctx_bb(struct intel_engine_cs *ring,
  *  This batch is terminated with MI_BATCH_BUFFER_END and so we need not add padding
  *  to align it with cacheline as padding after MI_BATCH_BUFFER_END is redundant.
  */
-static int gen8_init_perctx_bb(struct intel_engine_cs *ring,
+static int gen8_init_perctx_bb(struct intel_engine_cs *engine,
 			       struct i915_wa_ctx_bb *wa_ctx,
 			       uint32_t *const batch,
 			       uint32_t *offset)
@@ -1390,13 +1394,13 @@ static int gen8_init_perctx_bb(struct intel_engine_cs *ring,
 	return wa_ctx_end(wa_ctx, *offset = index, 1);
 }
 
-static int gen9_init_indirectctx_bb(struct intel_engine_cs *ring,
+static int gen9_init_indirectctx_bb(struct intel_engine_cs *engine,
 				    struct i915_wa_ctx_bb *wa_ctx,
 				    uint32_t *const batch,
 				    uint32_t *offset)
 {
 	int ret;
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
 
 	/* WaDisableCtxRestoreArbitration:skl,bxt */
@@ -1405,7 +1409,7 @@ static int gen9_init_indirectctx_bb(struct intel_engine_cs *ring,
 		wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_DISABLE);
 
 	/* WaFlushCoherentL3CacheLinesAtContextSwitch:skl,bxt */
-	ret = gen8_emit_flush_coherentl3_wa(ring, batch, index);
+	ret = gen8_emit_flush_coherentl3_wa(engine, batch, index);
 	if (ret < 0)
 		return ret;
 	index = ret;
@@ -1417,12 +1421,12 @@ static int gen9_init_indirectctx_bb(struct intel_engine_cs *ring,
 	return wa_ctx_end(wa_ctx, *offset = index, CACHELINE_DWORDS);
 }
 
-static int gen9_init_perctx_bb(struct intel_engine_cs *ring,
+static int gen9_init_perctx_bb(struct intel_engine_cs *engine,
 			       struct i915_wa_ctx_bb *wa_ctx,
 			       uint32_t *const batch,
 			       uint32_t *offset)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
 
 	/* WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken:skl,bxt */
@@ -1435,6 +1439,25 @@ static int gen9_init_perctx_bb(struct intel_engine_cs *ring,
 		wa_ctx_emit(batch, index, MI_NOOP);
 	}
 
+	/* WaClearTdlStateAckDirtyBits:bxt */
+	if (IS_BXT_REVID(dev, 0, BXT_REVID_B0)) {
+		wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(4));
+
+		wa_ctx_emit_reg(batch, index, GEN8_STATE_ACK);
+		wa_ctx_emit(batch, index, _MASKED_BIT_DISABLE(GEN9_SUBSLICE_TDL_ACK_BITS));
+
+		wa_ctx_emit_reg(batch, index, GEN9_STATE_ACK_SLICE1);
+		wa_ctx_emit(batch, index, _MASKED_BIT_DISABLE(GEN9_SUBSLICE_TDL_ACK_BITS));
+
+		wa_ctx_emit_reg(batch, index, GEN9_STATE_ACK_SLICE2);
+		wa_ctx_emit(batch, index, _MASKED_BIT_DISABLE(GEN9_SUBSLICE_TDL_ACK_BITS));
+
+		wa_ctx_emit_reg(batch, index, GEN7_ROW_CHICKEN2);
+		/* dummy write to CS, mask bits are 0 to ensure the register is not modified */
+		wa_ctx_emit(batch, index, 0x0);
+		wa_ctx_emit(batch, index, MI_NOOP);
+	}
+
 	/* WaDisableCtxRestoreArbitration:skl,bxt */
 	if (IS_SKL_REVID(dev, 0, SKL_REVID_D0) ||
 	    IS_BXT_REVID(dev, 0, BXT_REVID_A1))
@@ -1445,60 +1468,61 @@ static int gen9_init_perctx_bb(struct intel_engine_cs *ring,
 	return wa_ctx_end(wa_ctx, *offset = index, 1);
 }
 
-static int lrc_setup_wa_ctx_obj(struct intel_engine_cs *ring, u32 size)
+static int lrc_setup_wa_ctx_obj(struct intel_engine_cs *engine, u32 size)
 {
 	int ret;
 
-	ring->wa_ctx.obj = i915_gem_alloc_object(ring->dev, PAGE_ALIGN(size));
-	if (!ring->wa_ctx.obj) {
+	engine->wa_ctx.obj = i915_gem_alloc_object(engine->dev,
+						   PAGE_ALIGN(size));
+	if (!engine->wa_ctx.obj) {
 		DRM_DEBUG_DRIVER("alloc LRC WA ctx backing obj failed.\n");
 		return -ENOMEM;
 	}
 
-	ret = i915_gem_obj_ggtt_pin(ring->wa_ctx.obj, PAGE_SIZE, 0);
+	ret = i915_gem_obj_ggtt_pin(engine->wa_ctx.obj, PAGE_SIZE, 0);
 	if (ret) {
 		DRM_DEBUG_DRIVER("pin LRC WA ctx backing obj failed: %d\n",
 				 ret);
-		drm_gem_object_unreference(&ring->wa_ctx.obj->base);
+		drm_gem_object_unreference(&engine->wa_ctx.obj->base);
 		return ret;
 	}
 
 	return 0;
 }
 
-static void lrc_destroy_wa_ctx_obj(struct intel_engine_cs *ring)
+static void lrc_destroy_wa_ctx_obj(struct intel_engine_cs *engine)
 {
-	if (ring->wa_ctx.obj) {
-		i915_gem_object_ggtt_unpin(ring->wa_ctx.obj);
-		drm_gem_object_unreference(&ring->wa_ctx.obj->base);
-		ring->wa_ctx.obj = NULL;
+	if (engine->wa_ctx.obj) {
+		i915_gem_object_ggtt_unpin(engine->wa_ctx.obj);
+		drm_gem_object_unreference(&engine->wa_ctx.obj->base);
+		engine->wa_ctx.obj = NULL;
 	}
 }
 
-static int intel_init_workaround_bb(struct intel_engine_cs *ring)
+static int intel_init_workaround_bb(struct intel_engine_cs *engine)
 {
 	int ret;
 	uint32_t *batch;
 	uint32_t offset;
 	struct page *page;
-	struct i915_ctx_workarounds *wa_ctx = &ring->wa_ctx;
+	struct i915_ctx_workarounds *wa_ctx = &engine->wa_ctx;
 
-	WARN_ON(ring->id != RCS);
+	WARN_ON(engine->id != RCS);
 
 	/* update this when WA for higher Gen are added */
-	if (INTEL_INFO(ring->dev)->gen > 9) {
+	if (INTEL_INFO(engine->dev)->gen > 9) {
 		DRM_ERROR("WA batch buffer is not initialized for Gen%d\n",
-			  INTEL_INFO(ring->dev)->gen);
+			  INTEL_INFO(engine->dev)->gen);
 		return 0;
 	}
 
 	/* some WA perform writes to scratch page, ensure it is valid */
-	if (ring->scratch.obj == NULL) {
-		DRM_ERROR("scratch page not allocated for %s\n", ring->name);
+	if (engine->scratch.obj == NULL) {
+		DRM_ERROR("scratch page not allocated for %s\n", engine->name);
 		return -EINVAL;
 	}
 
-	ret = lrc_setup_wa_ctx_obj(ring, PAGE_SIZE);
+	ret = lrc_setup_wa_ctx_obj(engine, PAGE_SIZE);
 	if (ret) {
 		DRM_DEBUG_DRIVER("Failed to setup context WA page: %d\n", ret);
 		return ret;
@@ -1508,29 +1532,29 @@ static int intel_init_workaround_bb(struct intel_engine_cs *ring)
 	batch = kmap_atomic(page);
 	offset = 0;
 
-	if (INTEL_INFO(ring->dev)->gen == 8) {
-		ret = gen8_init_indirectctx_bb(ring,
+	if (INTEL_INFO(engine->dev)->gen == 8) {
+		ret = gen8_init_indirectctx_bb(engine,
 					       &wa_ctx->indirect_ctx,
 					       batch,
 					       &offset);
 		if (ret)
 			goto out;
 
-		ret = gen8_init_perctx_bb(ring,
+		ret = gen8_init_perctx_bb(engine,
 					  &wa_ctx->per_ctx,
 					  batch,
 					  &offset);
 		if (ret)
 			goto out;
-	} else if (INTEL_INFO(ring->dev)->gen == 9) {
-		ret = gen9_init_indirectctx_bb(ring,
+	} else if (INTEL_INFO(engine->dev)->gen == 9) {
+		ret = gen9_init_indirectctx_bb(engine,
 					       &wa_ctx->indirect_ctx,
 					       batch,
 					       &offset);
 		if (ret)
 			goto out;
 
-		ret = gen9_init_perctx_bb(ring,
+		ret = gen9_init_perctx_bb(engine,
 					  &wa_ctx->per_ctx,
 					  batch,
 					  &offset);
@@ -1541,27 +1565,36 @@ static int intel_init_workaround_bb(struct intel_engine_cs *ring)
 out:
 	kunmap_atomic(batch);
 	if (ret)
-		lrc_destroy_wa_ctx_obj(ring);
+		lrc_destroy_wa_ctx_obj(engine);
 
 	return ret;
 }
 
-static int gen8_init_common_ring(struct intel_engine_cs *ring)
+static void lrc_init_hws(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
+
+	I915_WRITE(RING_HWS_PGA(engine->mmio_base),
+		   (u32)engine->status_page.gfx_addr);
+	POSTING_READ(RING_HWS_PGA(engine->mmio_base));
+}
+
+static int gen8_init_common_ring(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u8 next_context_status_buffer_hw;
+	unsigned int next_context_status_buffer_hw;
 
-	lrc_setup_hardware_status_page(ring,
-				dev_priv->kernel_context->engine[ring->id].state);
+	lrc_init_hws(engine);
 
-	I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | ring->irq_keep_mask));
-	I915_WRITE(RING_HWSTAM(ring->mmio_base), 0xffffffff);
+	I915_WRITE_IMR(engine,
+		       ~(engine->irq_enable_mask | engine->irq_keep_mask));
+	I915_WRITE(RING_HWSTAM(engine->mmio_base), 0xffffffff);
 
-	I915_WRITE(RING_MODE_GEN7(ring),
+	I915_WRITE(RING_MODE_GEN7(engine),
 		   _MASKED_BIT_DISABLE(GFX_REPLAY_MODE) |
 		   _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE));
-	POSTING_READ(RING_MODE_GEN7(ring));
+	POSTING_READ(RING_MODE_GEN7(engine));
 
 	/*
 	 * Instead of resetting the Context Status Buffer (CSB) read pointer to
@@ -1576,7 +1609,7 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring)
 	 * BXT  |         ?                |         ?            |
 	 */
 	next_context_status_buffer_hw =
-		GEN8_CSB_WRITE_PTR(I915_READ(RING_CONTEXT_STATUS_PTR(ring)));
+		GEN8_CSB_WRITE_PTR(I915_READ(RING_CONTEXT_STATUS_PTR(engine)));
 
 	/*
 	 * When the CSB registers are reset (also after power-up / gpu reset),
@@ -1586,21 +1619,21 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring)
 	if (next_context_status_buffer_hw == GEN8_CSB_PTR_MASK)
 		next_context_status_buffer_hw = (GEN8_CSB_ENTRIES - 1);
 
-	ring->next_context_status_buffer = next_context_status_buffer_hw;
-	DRM_DEBUG_DRIVER("Execlists enabled for %s\n", ring->name);
+	engine->next_context_status_buffer = next_context_status_buffer_hw;
+	DRM_DEBUG_DRIVER("Execlists enabled for %s\n", engine->name);
 
-	memset(&ring->hangcheck, 0, sizeof(ring->hangcheck));
+	intel_engine_init_hangcheck(engine);
 
-	return 0;
+	return intel_mocs_init_engine(engine);
 }
 
-static int gen8_init_render_ring(struct intel_engine_cs *ring)
+static int gen8_init_render_ring(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
-	ret = gen8_init_common_ring(ring);
+	ret = gen8_init_common_ring(engine);
 	if (ret)
 		return ret;
 
@@ -1614,24 +1647,24 @@ static int gen8_init_render_ring(struct intel_engine_cs *ring)
 
 	I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING));
 
-	return init_workarounds_ring(ring);
+	return init_workarounds_ring(engine);
 }
 
-static int gen9_init_render_ring(struct intel_engine_cs *ring)
+static int gen9_init_render_ring(struct intel_engine_cs *engine)
 {
 	int ret;
 
-	ret = gen8_init_common_ring(ring);
+	ret = gen8_init_common_ring(engine);
 	if (ret)
 		return ret;
 
-	return init_workarounds_ring(ring);
+	return init_workarounds_ring(engine);
 }
 
 static int intel_logical_ring_emit_pdps(struct drm_i915_gem_request *req)
 {
 	struct i915_hw_ppgtt *ppgtt = req->ctx->ppgtt;
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	struct intel_ringbuffer *ringbuf = req->ringbuf;
 	const int num_lri_cmds = GEN8_LEGACY_PDPES * 2;
 	int i, ret;
@@ -1644,9 +1677,11 @@ static int intel_logical_ring_emit_pdps(struct drm_i915_gem_request *req)
 	for (i = GEN8_LEGACY_PDPES - 1; i >= 0; i--) {
 		const dma_addr_t pd_daddr = i915_page_dir_dma_addr(ppgtt, i);
 
-		intel_logical_ring_emit_reg(ringbuf, GEN8_RING_PDP_UDW(ring, i));
+		intel_logical_ring_emit_reg(ringbuf,
+					    GEN8_RING_PDP_UDW(engine, i));
 		intel_logical_ring_emit(ringbuf, upper_32_bits(pd_daddr));
-		intel_logical_ring_emit_reg(ringbuf, GEN8_RING_PDP_LDW(ring, i));
+		intel_logical_ring_emit_reg(ringbuf,
+					    GEN8_RING_PDP_LDW(engine, i));
 		intel_logical_ring_emit(ringbuf, lower_32_bits(pd_daddr));
 	}
 
@@ -1670,7 +1705,7 @@ static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
 	 * not idle). PML4 is allocated during ppgtt init so this is
 	 * not needed in 48-bit.*/
 	if (req->ctx->ppgtt &&
-	    (intel_ring_flag(req->ring) & req->ctx->ppgtt->pd_dirty_rings)) {
+	    (intel_engine_flag(req->engine) & req->ctx->ppgtt->pd_dirty_rings)) {
 		if (!USES_FULL_48BIT_PPGTT(req->i915) &&
 		    !intel_vgpu_active(req->i915->dev)) {
 			ret = intel_logical_ring_emit_pdps(req);
@@ -1678,7 +1713,7 @@ static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
 				return ret;
 		}
 
-		req->ctx->ppgtt->pd_dirty_rings &= ~intel_ring_flag(req->ring);
+		req->ctx->ppgtt->pd_dirty_rings &= ~intel_engine_flag(req->engine);
 	}
 
 	ret = intel_logical_ring_begin(req, 4);
@@ -1698,9 +1733,9 @@ static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
 	return 0;
 }
 
-static bool gen8_logical_ring_get_irq(struct intel_engine_cs *ring)
+static bool gen8_logical_ring_get_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
@@ -1708,25 +1743,26 @@ static bool gen8_logical_ring_get_irq(struct intel_engine_cs *ring)
 		return false;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (ring->irq_refcount++ == 0) {
-		I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | ring->irq_keep_mask));
-		POSTING_READ(RING_IMR(ring->mmio_base));
+	if (engine->irq_refcount++ == 0) {
+		I915_WRITE_IMR(engine,
+			       ~(engine->irq_enable_mask | engine->irq_keep_mask));
+		POSTING_READ(RING_IMR(engine->mmio_base));
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 
 	return true;
 }
 
-static void gen8_logical_ring_put_irq(struct intel_engine_cs *ring)
+static void gen8_logical_ring_put_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (--ring->irq_refcount == 0) {
-		I915_WRITE_IMR(ring, ~ring->irq_keep_mask);
-		POSTING_READ(RING_IMR(ring->mmio_base));
+	if (--engine->irq_refcount == 0) {
+		I915_WRITE_IMR(engine, ~engine->irq_keep_mask);
+		POSTING_READ(RING_IMR(engine->mmio_base));
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
@@ -1736,8 +1772,8 @@ static int gen8_emit_flush(struct drm_i915_gem_request *request,
 			   u32 unused)
 {
 	struct intel_ringbuffer *ringbuf = request->ringbuf;
-	struct intel_engine_cs *ring = ringbuf->ring;
-	struct drm_device *dev = ring->dev;
+	struct intel_engine_cs *engine = ringbuf->engine;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t cmd;
 	int ret;
@@ -1757,7 +1793,7 @@ static int gen8_emit_flush(struct drm_i915_gem_request *request,
 
 	if (invalidate_domains & I915_GEM_GPU_DOMAINS) {
 		cmd |= MI_INVALIDATE_TLB;
-		if (ring == &dev_priv->ring[VCS])
+		if (engine == &dev_priv->engine[VCS])
 			cmd |= MI_INVALIDATE_BSD;
 	}
 
@@ -1777,8 +1813,8 @@ static int gen8_emit_flush_render(struct drm_i915_gem_request *request,
 				  u32 flush_domains)
 {
 	struct intel_ringbuffer *ringbuf = request->ringbuf;
-	struct intel_engine_cs *ring = ringbuf->ring;
-	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	struct intel_engine_cs *engine = ringbuf->engine;
+	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	bool vf_flush_wa = false;
 	u32 flags = 0;
 	int ret;
@@ -1806,7 +1842,7 @@ static int gen8_emit_flush_render(struct drm_i915_gem_request *request,
 		 * On GEN9: before VF_CACHE_INVALIDATE we need to emit a NULL
 		 * pipe control.
 		 */
-		if (IS_GEN9(ring->dev))
+		if (IS_GEN9(engine->dev))
 			vf_flush_wa = true;
 	}
 
@@ -1834,19 +1870,18 @@ static int gen8_emit_flush_render(struct drm_i915_gem_request *request,
 	return 0;
 }
 
-static u32 gen8_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+static u32 gen8_get_seqno(struct intel_engine_cs *engine)
 {
-	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
+	return intel_read_status_page(engine, I915_GEM_HWS_INDEX);
 }
 
-static void gen8_set_seqno(struct intel_engine_cs *ring, u32 seqno)
+static void gen8_set_seqno(struct intel_engine_cs *engine, u32 seqno)
 {
-	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
+	intel_write_status_page(engine, I915_GEM_HWS_INDEX, seqno);
 }
 
-static u32 bxt_a_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+static void bxt_a_seqno_barrier(struct intel_engine_cs *engine)
 {
-
 	/*
 	 * On BXT A steppings there is a HW coherency issue whereby the
 	 * MI_STORE_DATA_IMM storing the completed request's seqno
@@ -1857,19 +1892,15 @@ static u32 bxt_a_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
 	 * bxt_a_set_seqno(), where we also do a clflush after the write. So
 	 * this clflush in practice becomes an invalidate operation.
 	 */
-
-	if (!lazy_coherency)
-		intel_flush_status_page(ring, I915_GEM_HWS_INDEX);
-
-	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
+	intel_flush_status_page(engine, I915_GEM_HWS_INDEX);
 }
 
-static void bxt_a_set_seqno(struct intel_engine_cs *ring, u32 seqno)
+static void bxt_a_set_seqno(struct intel_engine_cs *engine, u32 seqno)
 {
-	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
+	intel_write_status_page(engine, I915_GEM_HWS_INDEX, seqno);
 
 	/* See bxt_a_get_seqno() explaining the reason for the clflush. */
-	intel_flush_status_page(ring, I915_GEM_HWS_INDEX);
+	intel_flush_status_page(engine, I915_GEM_HWS_INDEX);
 }
 
 /*
@@ -1899,7 +1930,7 @@ static int gen8_emit_request(struct drm_i915_gem_request *request)
 	intel_logical_ring_emit(ringbuf,
 				(MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW);
 	intel_logical_ring_emit(ringbuf,
-				hws_seqno_address(request->ring) |
+				hws_seqno_address(request->engine) |
 				MI_FLUSH_DW_USE_GTT);
 	intel_logical_ring_emit(ringbuf, 0);
 	intel_logical_ring_emit(ringbuf, i915_gem_request_get_seqno(request));
@@ -1929,7 +1960,7 @@ static int gen8_emit_request_render(struct drm_i915_gem_request *request)
 				(PIPE_CONTROL_GLOBAL_GTT_IVB |
 				 PIPE_CONTROL_CS_STALL |
 				 PIPE_CONTROL_QW_WRITE));
-	intel_logical_ring_emit(ringbuf, hws_seqno_address(request->ring));
+	intel_logical_ring_emit(ringbuf, hws_seqno_address(request->engine));
 	intel_logical_ring_emit(ringbuf, 0);
 	intel_logical_ring_emit(ringbuf, i915_gem_request_get_seqno(request));
 	/* We're thrashing one dword of HWS. */
@@ -1944,19 +1975,19 @@ static int intel_lr_context_render_state_init(struct drm_i915_gem_request *req)
 	struct render_state so;
 	int ret;
 
-	ret = i915_gem_render_state_prepare(req->ring, &so);
+	ret = i915_gem_render_state_prepare(req->engine, &so);
 	if (ret)
 		return ret;
 
 	if (so.rodata == NULL)
 		return 0;
 
-	ret = req->ring->emit_bb_start(req, so.ggtt_offset,
+	ret = req->engine->emit_bb_start(req, so.ggtt_offset,
 				       I915_DISPATCH_SECURE);
 	if (ret)
 		goto out;
 
-	ret = req->ring->emit_bb_start(req,
+	ret = req->engine->emit_bb_start(req,
 				       (so.ggtt_offset + so.aux_batch_offset),
 				       I915_DISPATCH_SECURE);
 	if (ret)
@@ -1994,146 +2025,197 @@ static int gen8_init_rcs_context(struct drm_i915_gem_request *req)
  * @ring: Engine Command Streamer.
  *
  */
-void intel_logical_ring_cleanup(struct intel_engine_cs *ring)
+void intel_logical_ring_cleanup(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv;
 
-	if (!intel_ring_initialized(ring))
+	if (!intel_engine_initialized(engine))
 		return;
 
-	dev_priv = ring->dev->dev_private;
+	/*
+	 * Tasklet cannot be active at this point due intel_mark_active/idle
+	 * so this is just for documentation.
+	 */
+	if (WARN_ON(test_bit(TASKLET_STATE_SCHED, &engine->irq_tasklet.state)))
+		tasklet_kill(&engine->irq_tasklet);
+
+	dev_priv = engine->dev->dev_private;
 
-	if (ring->buffer) {
-		intel_logical_ring_stop(ring);
-		WARN_ON((I915_READ_MODE(ring) & MODE_IDLE) == 0);
+	if (engine->buffer) {
+		intel_logical_ring_stop(engine);
+		WARN_ON((I915_READ_MODE(engine) & MODE_IDLE) == 0);
 	}
 
-	if (ring->cleanup)
-		ring->cleanup(ring);
+	if (engine->cleanup)
+		engine->cleanup(engine);
 
-	i915_cmd_parser_fini_ring(ring);
-	i915_gem_batch_pool_fini(&ring->batch_pool);
+	i915_cmd_parser_fini_ring(engine);
+	i915_gem_batch_pool_fini(&engine->batch_pool);
 
-	if (ring->status_page.obj) {
-		kunmap(sg_page(ring->status_page.obj->pages->sgl));
-		ring->status_page.obj = NULL;
+	if (engine->status_page.obj) {
+		i915_gem_object_unpin_map(engine->status_page.obj);
+		engine->status_page.obj = NULL;
 	}
 
-	ring->disable_lite_restore_wa = false;
-	ring->ctx_desc_template = 0;
+	engine->idle_lite_restore_wa = 0;
+	engine->disable_lite_restore_wa = false;
+	engine->ctx_desc_template = 0;
 
-	lrc_destroy_wa_ctx_obj(ring);
-	ring->dev = NULL;
+	lrc_destroy_wa_ctx_obj(engine);
+	engine->dev = NULL;
 }
 
 static void
 logical_ring_default_vfuncs(struct drm_device *dev,
-			    struct intel_engine_cs *ring)
+			    struct intel_engine_cs *engine)
 {
 	/* Default vfuncs which can be overriden by each engine. */
-	ring->init_hw = gen8_init_common_ring;
-	ring->emit_request = gen8_emit_request;
-	ring->emit_flush = gen8_emit_flush;
-	ring->irq_get = gen8_logical_ring_get_irq;
-	ring->irq_put = gen8_logical_ring_put_irq;
-	ring->emit_bb_start = gen8_emit_bb_start;
+	engine->init_hw = gen8_init_common_ring;
+	engine->emit_request = gen8_emit_request;
+	engine->emit_flush = gen8_emit_flush;
+	engine->irq_get = gen8_logical_ring_get_irq;
+	engine->irq_put = gen8_logical_ring_put_irq;
+	engine->emit_bb_start = gen8_emit_bb_start;
+	engine->get_seqno = gen8_get_seqno;
+	engine->set_seqno = gen8_set_seqno;
 	if (IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
-		ring->get_seqno = bxt_a_get_seqno;
-		ring->set_seqno = bxt_a_set_seqno;
-	} else {
-		ring->get_seqno = gen8_get_seqno;
-		ring->set_seqno = gen8_set_seqno;
+		engine->irq_seqno_barrier = bxt_a_seqno_barrier;
+		engine->set_seqno = bxt_a_set_seqno;
 	}
 }
 
 static inline void
-logical_ring_default_irqs(struct intel_engine_cs *ring, unsigned shift)
+logical_ring_default_irqs(struct intel_engine_cs *engine, unsigned shift)
 {
-	ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT << shift;
-	ring->irq_keep_mask = GT_CONTEXT_SWITCH_INTERRUPT << shift;
+	engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT << shift;
+	engine->irq_keep_mask = GT_CONTEXT_SWITCH_INTERRUPT << shift;
+}
+
+static int
+lrc_setup_hws(struct intel_engine_cs *engine,
+	      struct drm_i915_gem_object *dctx_obj)
+{
+	void *hws;
+
+	/* The HWSP is part of the default context object in LRC mode. */
+	engine->status_page.gfx_addr = i915_gem_obj_ggtt_offset(dctx_obj) +
+				       LRC_PPHWSP_PN * PAGE_SIZE;
+	hws = i915_gem_object_pin_map(dctx_obj);
+	if (IS_ERR(hws))
+		return PTR_ERR(hws);
+	engine->status_page.page_addr = hws + LRC_PPHWSP_PN * PAGE_SIZE;
+	engine->status_page.obj = dctx_obj;
+
+	return 0;
 }
 
 static int
-logical_ring_init(struct drm_device *dev, struct intel_engine_cs *ring)
+logical_ring_init(struct drm_device *dev, struct intel_engine_cs *engine)
 {
-	struct intel_context *dctx = to_i915(dev)->kernel_context;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_context *dctx = dev_priv->kernel_context;
+	enum forcewake_domains fw_domains;
 	int ret;
 
 	/* Intentionally left blank. */
-	ring->buffer = NULL;
+	engine->buffer = NULL;
+
+	engine->dev = dev;
+	INIT_LIST_HEAD(&engine->active_list);
+	INIT_LIST_HEAD(&engine->request_list);
+	i915_gem_batch_pool_init(dev, &engine->batch_pool);
+	init_waitqueue_head(&engine->irq_queue);
+
+	INIT_LIST_HEAD(&engine->buffers);
+	INIT_LIST_HEAD(&engine->execlist_queue);
+	INIT_LIST_HEAD(&engine->execlist_retired_req_list);
+	spin_lock_init(&engine->execlist_lock);
+
+	tasklet_init(&engine->irq_tasklet,
+		     intel_lrc_irq_handler, (unsigned long)engine);
+
+	logical_ring_init_platform_invariants(engine);
+
+	fw_domains = intel_uncore_forcewake_for_reg(dev_priv,
+						    RING_ELSP(engine),
+						    FW_REG_WRITE);
 
-	ring->dev = dev;
-	INIT_LIST_HEAD(&ring->active_list);
-	INIT_LIST_HEAD(&ring->request_list);
-	i915_gem_batch_pool_init(dev, &ring->batch_pool);
-	init_waitqueue_head(&ring->irq_queue);
+	fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
+						     RING_CONTEXT_STATUS_PTR(engine),
+						     FW_REG_READ | FW_REG_WRITE);
 
-	INIT_LIST_HEAD(&ring->buffers);
-	INIT_LIST_HEAD(&ring->execlist_queue);
-	INIT_LIST_HEAD(&ring->execlist_retired_req_list);
-	spin_lock_init(&ring->execlist_lock);
+	fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
+						     RING_CONTEXT_STATUS_BUF_BASE(engine),
+						     FW_REG_READ);
 
-	logical_ring_init_platform_invariants(ring);
+	engine->fw_domains = fw_domains;
 
-	ret = i915_cmd_parser_init_ring(ring);
+	ret = i915_cmd_parser_init_ring(engine);
 	if (ret)
 		goto error;
 
-	ret = intel_lr_context_deferred_alloc(dctx, ring);
+	ret = intel_lr_context_deferred_alloc(dctx, engine);
 	if (ret)
 		goto error;
 
 	/* As this is the default context, always pin it */
-	ret = intel_lr_context_do_pin(dctx, ring);
+	ret = intel_lr_context_do_pin(dctx, engine);
 	if (ret) {
 		DRM_ERROR(
 			"Failed to pin and map ringbuffer %s: %d\n",
-			ring->name, ret);
+			engine->name, ret);
+		goto error;
+	}
+
+	/* And setup the hardware status page. */
+	ret = lrc_setup_hws(engine, dctx->engine[engine->id].state);
+	if (ret) {
+		DRM_ERROR("Failed to set up hws %s: %d\n", engine->name, ret);
 		goto error;
 	}
 
 	return 0;
 
 error:
-	intel_logical_ring_cleanup(ring);
+	intel_logical_ring_cleanup(engine);
 	return ret;
 }
 
 static int logical_render_ring_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	int ret;
 
-	ring->name = "render ring";
-	ring->id = RCS;
-	ring->exec_id = I915_EXEC_RENDER;
-	ring->guc_id = GUC_RENDER_ENGINE;
-	ring->mmio_base = RENDER_RING_BASE;
+	engine->name = "render ring";
+	engine->id = RCS;
+	engine->exec_id = I915_EXEC_RENDER;
+	engine->guc_id = GUC_RENDER_ENGINE;
+	engine->mmio_base = RENDER_RING_BASE;
 
-	logical_ring_default_irqs(ring, GEN8_RCS_IRQ_SHIFT);
+	logical_ring_default_irqs(engine, GEN8_RCS_IRQ_SHIFT);
 	if (HAS_L3_DPF(dev))
-		ring->irq_keep_mask |= GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
+		engine->irq_keep_mask |= GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
 
-	logical_ring_default_vfuncs(dev, ring);
+	logical_ring_default_vfuncs(dev, engine);
 
 	/* Override some for render ring. */
 	if (INTEL_INFO(dev)->gen >= 9)
-		ring->init_hw = gen9_init_render_ring;
+		engine->init_hw = gen9_init_render_ring;
 	else
-		ring->init_hw = gen8_init_render_ring;
-	ring->init_context = gen8_init_rcs_context;
-	ring->cleanup = intel_fini_pipe_control;
-	ring->emit_flush = gen8_emit_flush_render;
-	ring->emit_request = gen8_emit_request_render;
+		engine->init_hw = gen8_init_render_ring;
+	engine->init_context = gen8_init_rcs_context;
+	engine->cleanup = intel_fini_pipe_control;
+	engine->emit_flush = gen8_emit_flush_render;
+	engine->emit_request = gen8_emit_request_render;
 
-	ring->dev = dev;
+	engine->dev = dev;
 
-	ret = intel_init_pipe_control(ring);
+	ret = intel_init_pipe_control(engine);
 	if (ret)
 		return ret;
 
-	ret = intel_init_workaround_bb(ring);
+	ret = intel_init_workaround_bb(engine);
 	if (ret) {
 		/*
 		 * We continue even if we fail to initialize WA batch
@@ -2144,9 +2226,9 @@ static int logical_render_ring_init(struct drm_device *dev)
 			  ret);
 	}
 
-	ret = logical_ring_init(dev, ring);
+	ret = logical_ring_init(dev, engine);
 	if (ret) {
-		lrc_destroy_wa_ctx_obj(ring);
+		lrc_destroy_wa_ctx_obj(engine);
 	}
 
 	return ret;
@@ -2155,69 +2237,69 @@ static int logical_render_ring_init(struct drm_device *dev)
 static int logical_bsd_ring_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[VCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[VCS];
 
-	ring->name = "bsd ring";
-	ring->id = VCS;
-	ring->exec_id = I915_EXEC_BSD;
-	ring->guc_id = GUC_VIDEO_ENGINE;
-	ring->mmio_base = GEN6_BSD_RING_BASE;
+	engine->name = "bsd ring";
+	engine->id = VCS;
+	engine->exec_id = I915_EXEC_BSD;
+	engine->guc_id = GUC_VIDEO_ENGINE;
+	engine->mmio_base = GEN6_BSD_RING_BASE;
 
-	logical_ring_default_irqs(ring, GEN8_VCS1_IRQ_SHIFT);
-	logical_ring_default_vfuncs(dev, ring);
+	logical_ring_default_irqs(engine, GEN8_VCS1_IRQ_SHIFT);
+	logical_ring_default_vfuncs(dev, engine);
 
-	return logical_ring_init(dev, ring);
+	return logical_ring_init(dev, engine);
 }
 
 static int logical_bsd2_ring_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[VCS2];
+	struct intel_engine_cs *engine = &dev_priv->engine[VCS2];
 
-	ring->name = "bsd2 ring";
-	ring->id = VCS2;
-	ring->exec_id = I915_EXEC_BSD;
-	ring->guc_id = GUC_VIDEO_ENGINE2;
-	ring->mmio_base = GEN8_BSD2_RING_BASE;
+	engine->name = "bsd2 ring";
+	engine->id = VCS2;
+	engine->exec_id = I915_EXEC_BSD;
+	engine->guc_id = GUC_VIDEO_ENGINE2;
+	engine->mmio_base = GEN8_BSD2_RING_BASE;
 
-	logical_ring_default_irqs(ring, GEN8_VCS2_IRQ_SHIFT);
-	logical_ring_default_vfuncs(dev, ring);
+	logical_ring_default_irqs(engine, GEN8_VCS2_IRQ_SHIFT);
+	logical_ring_default_vfuncs(dev, engine);
 
-	return logical_ring_init(dev, ring);
+	return logical_ring_init(dev, engine);
 }
 
 static int logical_blt_ring_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[BCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[BCS];
 
-	ring->name = "blitter ring";
-	ring->id = BCS;
-	ring->exec_id = I915_EXEC_BLT;
-	ring->guc_id = GUC_BLITTER_ENGINE;
-	ring->mmio_base = BLT_RING_BASE;
+	engine->name = "blitter ring";
+	engine->id = BCS;
+	engine->exec_id = I915_EXEC_BLT;
+	engine->guc_id = GUC_BLITTER_ENGINE;
+	engine->mmio_base = BLT_RING_BASE;
 
-	logical_ring_default_irqs(ring, GEN8_BCS_IRQ_SHIFT);
-	logical_ring_default_vfuncs(dev, ring);
+	logical_ring_default_irqs(engine, GEN8_BCS_IRQ_SHIFT);
+	logical_ring_default_vfuncs(dev, engine);
 
-	return logical_ring_init(dev, ring);
+	return logical_ring_init(dev, engine);
 }
 
 static int logical_vebox_ring_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[VECS];
+	struct intel_engine_cs *engine = &dev_priv->engine[VECS];
 
-	ring->name = "video enhancement ring";
-	ring->id = VECS;
-	ring->exec_id = I915_EXEC_VEBOX;
-	ring->guc_id = GUC_VIDEOENHANCE_ENGINE;
-	ring->mmio_base = VEBOX_RING_BASE;
+	engine->name = "video enhancement ring";
+	engine->id = VECS;
+	engine->exec_id = I915_EXEC_VEBOX;
+	engine->guc_id = GUC_VIDEOENHANCE_ENGINE;
+	engine->mmio_base = VEBOX_RING_BASE;
 
-	logical_ring_default_irqs(ring, GEN8_VECS_IRQ_SHIFT);
-	logical_ring_default_vfuncs(dev, ring);
+	logical_ring_default_irqs(engine, GEN8_VECS_IRQ_SHIFT);
+	logical_ring_default_vfuncs(dev, engine);
 
-	return logical_ring_init(dev, ring);
+	return logical_ring_init(dev, engine);
 }
 
 /**
@@ -2225,7 +2307,7 @@ static int logical_vebox_ring_init(struct drm_device *dev)
  * @dev: DRM device.
  *
  * This function inits the engines for an Execlists submission style (the equivalent in the
- * legacy ringbuffer submission world would be i915_gem_init_rings). It does it only for
+ * legacy ringbuffer submission world would be i915_gem_init_engines). It does it only for
  * those engines that are present in the hardware.
  *
  * Return: non-zero if the initialization failed.
@@ -2266,13 +2348,13 @@ int intel_logical_rings_init(struct drm_device *dev)
 	return 0;
 
 cleanup_vebox_ring:
-	intel_logical_ring_cleanup(&dev_priv->ring[VECS]);
+	intel_logical_ring_cleanup(&dev_priv->engine[VECS]);
 cleanup_blt_ring:
-	intel_logical_ring_cleanup(&dev_priv->ring[BCS]);
+	intel_logical_ring_cleanup(&dev_priv->engine[BCS]);
 cleanup_bsd_ring:
-	intel_logical_ring_cleanup(&dev_priv->ring[VCS]);
+	intel_logical_ring_cleanup(&dev_priv->engine[VCS]);
 cleanup_render_ring:
-	intel_logical_ring_cleanup(&dev_priv->ring[RCS]);
+	intel_logical_ring_cleanup(&dev_priv->engine[RCS]);
 
 	return ret;
 }
@@ -2320,13 +2402,13 @@ make_rpcs(struct drm_device *dev)
 	return rpcs;
 }
 
-static u32 intel_lr_indirect_ctx_offset(struct intel_engine_cs *ring)
+static u32 intel_lr_indirect_ctx_offset(struct intel_engine_cs *engine)
 {
 	u32 indirect_ctx_offset;
 
-	switch (INTEL_INFO(ring->dev)->gen) {
+	switch (INTEL_INFO(engine->dev)->gen) {
 	default:
-		MISSING_CASE(INTEL_INFO(ring->dev)->gen);
+		MISSING_CASE(INTEL_INFO(engine->dev)->gen);
 		/* fall through */
 	case 9:
 		indirect_ctx_offset =
@@ -2342,14 +2424,16 @@ static u32 intel_lr_indirect_ctx_offset(struct intel_engine_cs *ring)
 }
 
 static int
-populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_obj,
-		    struct intel_engine_cs *ring, struct intel_ringbuffer *ringbuf)
+populate_lr_context(struct intel_context *ctx,
+		    struct drm_i915_gem_object *ctx_obj,
+		    struct intel_engine_cs *engine,
+		    struct intel_ringbuffer *ringbuf)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_hw_ppgtt *ppgtt = ctx->ppgtt;
-	struct page *page;
-	uint32_t *reg_state;
+	void *vaddr;
+	u32 *reg_state;
 	int ret;
 
 	if (!ppgtt)
@@ -2361,18 +2445,17 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 		return ret;
 	}
 
-	ret = i915_gem_object_get_pages(ctx_obj);
-	if (ret) {
-		DRM_DEBUG_DRIVER("Could not get object pages\n");
+	vaddr = i915_gem_object_pin_map(ctx_obj);
+	if (IS_ERR(vaddr)) {
+		ret = PTR_ERR(vaddr);
+		DRM_DEBUG_DRIVER("Could not map object pages! (%d)\n", ret);
 		return ret;
 	}
-
-	i915_gem_object_pin_pages(ctx_obj);
+	ctx_obj->dirty = true;
 
 	/* The second page of the context object contains some fields which must
 	 * be set up prior to the first execution. */
-	page = i915_gem_object_get_dirty_page(ctx_obj, LRC_STATE_PN);
-	reg_state = kmap_atomic(page);
+	reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
 
 	/* A context is actually a big batch buffer with several MI_LOAD_REGISTER_IMM
 	 * commands followed by (reg, value) pairs. The values we are setting here are
@@ -2380,33 +2463,47 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 	 * recreate this batchbuffer with new values (including all the missing
 	 * MI_LOAD_REGISTER_IMM commands that we are not initializing here). */
 	reg_state[CTX_LRI_HEADER_0] =
-		MI_LOAD_REGISTER_IMM(ring->id == RCS ? 14 : 11) | MI_LRI_FORCE_POSTED;
-	ASSIGN_CTX_REG(reg_state, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(ring),
+		MI_LOAD_REGISTER_IMM(engine->id == RCS ? 14 : 11) | MI_LRI_FORCE_POSTED;
+	ASSIGN_CTX_REG(reg_state, CTX_CONTEXT_CONTROL,
+		       RING_CONTEXT_CONTROL(engine),
 		       _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH |
 					  CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |
 					  (HAS_RESOURCE_STREAMER(dev) ?
 					    CTX_CTRL_RS_CTX_ENABLE : 0)));
-	ASSIGN_CTX_REG(reg_state, CTX_RING_HEAD, RING_HEAD(ring->mmio_base), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_RING_TAIL, RING_TAIL(ring->mmio_base), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_RING_HEAD, RING_HEAD(engine->mmio_base),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_RING_TAIL, RING_TAIL(engine->mmio_base),
+		       0);
 	/* Ring buffer start address is not known until the buffer is pinned.
 	 * It is written to the context image in execlists_update_context()
 	 */
-	ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_START, RING_START(ring->mmio_base), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_CONTROL, RING_CTL(ring->mmio_base),
+	ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_START,
+		       RING_START(engine->mmio_base), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_CONTROL,
+		       RING_CTL(engine->mmio_base),
 		       ((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES) | RING_VALID);
-	ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_U, RING_BBADDR_UDW(ring->mmio_base), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_L, RING_BBADDR(ring->mmio_base), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_BB_STATE, RING_BBSTATE(ring->mmio_base),
+	ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_U,
+		       RING_BBADDR_UDW(engine->mmio_base), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_L,
+		       RING_BBADDR(engine->mmio_base), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_BB_STATE,
+		       RING_BBSTATE(engine->mmio_base),
 		       RING_BB_PPGTT);
-	ASSIGN_CTX_REG(reg_state, CTX_SECOND_BB_HEAD_U, RING_SBBADDR_UDW(ring->mmio_base), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_SECOND_BB_HEAD_L, RING_SBBADDR(ring->mmio_base), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_SECOND_BB_STATE, RING_SBBSTATE(ring->mmio_base), 0);
-	if (ring->id == RCS) {
-		ASSIGN_CTX_REG(reg_state, CTX_BB_PER_CTX_PTR, RING_BB_PER_CTX_PTR(ring->mmio_base), 0);
-		ASSIGN_CTX_REG(reg_state, CTX_RCS_INDIRECT_CTX, RING_INDIRECT_CTX(ring->mmio_base), 0);
-		ASSIGN_CTX_REG(reg_state, CTX_RCS_INDIRECT_CTX_OFFSET, RING_INDIRECT_CTX_OFFSET(ring->mmio_base), 0);
-		if (ring->wa_ctx.obj) {
-			struct i915_ctx_workarounds *wa_ctx = &ring->wa_ctx;
+	ASSIGN_CTX_REG(reg_state, CTX_SECOND_BB_HEAD_U,
+		       RING_SBBADDR_UDW(engine->mmio_base), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_SECOND_BB_HEAD_L,
+		       RING_SBBADDR(engine->mmio_base), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_SECOND_BB_STATE,
+		       RING_SBBSTATE(engine->mmio_base), 0);
+	if (engine->id == RCS) {
+		ASSIGN_CTX_REG(reg_state, CTX_BB_PER_CTX_PTR,
+			       RING_BB_PER_CTX_PTR(engine->mmio_base), 0);
+		ASSIGN_CTX_REG(reg_state, CTX_RCS_INDIRECT_CTX,
+			       RING_INDIRECT_CTX(engine->mmio_base), 0);
+		ASSIGN_CTX_REG(reg_state, CTX_RCS_INDIRECT_CTX_OFFSET,
+			       RING_INDIRECT_CTX_OFFSET(engine->mmio_base), 0);
+		if (engine->wa_ctx.obj) {
+			struct i915_ctx_workarounds *wa_ctx = &engine->wa_ctx;
 			uint32_t ggtt_offset = i915_gem_obj_ggtt_offset(wa_ctx->obj);
 
 			reg_state[CTX_RCS_INDIRECT_CTX+1] =
@@ -2414,7 +2511,7 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 				(wa_ctx->indirect_ctx.size / CACHELINE_DWORDS);
 
 			reg_state[CTX_RCS_INDIRECT_CTX_OFFSET+1] =
-				intel_lr_indirect_ctx_offset(ring) << 6;
+				intel_lr_indirect_ctx_offset(engine) << 6;
 
 			reg_state[CTX_BB_PER_CTX_PTR+1] =
 				(ggtt_offset + wa_ctx->per_ctx.offset * sizeof(uint32_t)) |
@@ -2422,16 +2519,25 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 		}
 	}
 	reg_state[CTX_LRI_HEADER_1] = MI_LOAD_REGISTER_IMM(9) | MI_LRI_FORCE_POSTED;
-	ASSIGN_CTX_REG(reg_state, CTX_CTX_TIMESTAMP, RING_CTX_TIMESTAMP(ring->mmio_base), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_CTX_TIMESTAMP,
+		       RING_CTX_TIMESTAMP(engine->mmio_base), 0);
 	/* PDP values well be assigned later if needed */
-	ASSIGN_CTX_REG(reg_state, CTX_PDP3_UDW, GEN8_RING_PDP_UDW(ring, 3), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_PDP3_LDW, GEN8_RING_PDP_LDW(ring, 3), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_PDP2_UDW, GEN8_RING_PDP_UDW(ring, 2), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_PDP2_LDW, GEN8_RING_PDP_LDW(ring, 2), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_PDP1_UDW, GEN8_RING_PDP_UDW(ring, 1), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_PDP1_LDW, GEN8_RING_PDP_LDW(ring, 1), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_PDP0_UDW, GEN8_RING_PDP_UDW(ring, 0), 0);
-	ASSIGN_CTX_REG(reg_state, CTX_PDP0_LDW, GEN8_RING_PDP_LDW(ring, 0), 0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP3_UDW, GEN8_RING_PDP_UDW(engine, 3),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP3_LDW, GEN8_RING_PDP_LDW(engine, 3),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP2_UDW, GEN8_RING_PDP_UDW(engine, 2),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP2_LDW, GEN8_RING_PDP_LDW(engine, 2),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP1_UDW, GEN8_RING_PDP_UDW(engine, 1),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP1_LDW, GEN8_RING_PDP_LDW(engine, 1),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP0_UDW, GEN8_RING_PDP_UDW(engine, 0),
+		       0);
+	ASSIGN_CTX_REG(reg_state, CTX_PDP0_LDW, GEN8_RING_PDP_LDW(engine, 0),
+		       0);
 
 	if (USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
 		/* 64b PPGTT (48bit canonical)
@@ -2445,20 +2551,16 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 		 * With dynamic page allocation, PDPs may not be allocated at
 		 * this point. Point the unallocated PDPs to the scratch page
 		 */
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
-		ASSIGN_CTX_PDP(ppgtt, reg_state, 0);
+		execlists_update_context_pdps(ppgtt, reg_state);
 	}
 
-	if (ring->id == RCS) {
+	if (engine->id == RCS) {
 		reg_state[CTX_LRI_HEADER_2] = MI_LOAD_REGISTER_IMM(1);
 		ASSIGN_CTX_REG(reg_state, CTX_R_PWR_CLK_STATE, GEN8_R_PWR_CLK_STATE,
 			       make_rpcs(dev));
 	}
 
-	kunmap_atomic(reg_state);
-	i915_gem_object_unpin_pages(ctx_obj);
+	i915_gem_object_unpin_map(ctx_obj);
 
 	return 0;
 }
@@ -2475,7 +2577,7 @@ void intel_lr_context_free(struct intel_context *ctx)
 {
 	int i;
 
-	for (i = I915_NUM_RINGS; --i >= 0; ) {
+	for (i = I915_NUM_ENGINES; --i >= 0; ) {
 		struct intel_ringbuffer *ringbuf = ctx->engine[i].ringbuf;
 		struct drm_i915_gem_object *ctx_obj = ctx->engine[i].state;
 
@@ -2485,6 +2587,7 @@ void intel_lr_context_free(struct intel_context *ctx)
 		if (ctx == ctx->i915->kernel_context) {
 			intel_unpin_ringbuffer_obj(ringbuf);
 			i915_gem_object_ggtt_unpin(ctx_obj);
+			i915_gem_object_unpin_map(ctx_obj);
 		}
 
 		WARN_ON(ctx->engine[i].pin_count);
@@ -2507,15 +2610,15 @@ void intel_lr_context_free(struct intel_context *ctx)
  * in LRC mode, but does not include the "shared data page" used with
  * GuC submission. The caller should account for this if using the GuC.
  */
-uint32_t intel_lr_context_size(struct intel_engine_cs *ring)
+uint32_t intel_lr_context_size(struct intel_engine_cs *engine)
 {
 	int ret = 0;
 
-	WARN_ON(INTEL_INFO(ring->dev)->gen < 8);
+	WARN_ON(INTEL_INFO(engine->dev)->gen < 8);
 
-	switch (ring->id) {
+	switch (engine->id) {
 	case RCS:
-		if (INTEL_INFO(ring->dev)->gen >= 9)
+		if (INTEL_INFO(engine->dev)->gen >= 9)
 			ret = GEN9_LR_CONTEXT_RENDER_SIZE;
 		else
 			ret = GEN8_LR_CONTEXT_RENDER_SIZE;
@@ -2531,24 +2634,6 @@ uint32_t intel_lr_context_size(struct intel_engine_cs *ring)
 	return ret;
 }
 
-static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
-		struct drm_i915_gem_object *default_ctx_obj)
-{
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	struct page *page;
-
-	/* The HWSP is part of the default context object in LRC mode. */
-	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(default_ctx_obj)
-			+ LRC_PPHWSP_PN * PAGE_SIZE;
-	page = i915_gem_object_get_page(default_ctx_obj, LRC_PPHWSP_PN);
-	ring->status_page.page_addr = kmap(page);
-	ring->status_page.obj = default_ctx_obj;
-
-	I915_WRITE(RING_HWS_PGA(ring->mmio_base),
-			(u32)ring->status_page.gfx_addr);
-	POSTING_READ(RING_HWS_PGA(ring->mmio_base));
-}
-
 /**
  * intel_lr_context_deferred_alloc() - create the LRC specific bits of a context
  * @ctx: LR context to create.
@@ -2564,18 +2649,18 @@ static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
  */
 
 int intel_lr_context_deferred_alloc(struct intel_context *ctx,
-				    struct intel_engine_cs *ring)
+				    struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_gem_object *ctx_obj;
 	uint32_t context_size;
 	struct intel_ringbuffer *ringbuf;
 	int ret;
 
 	WARN_ON(ctx->legacy_hw_ctx.rcs_state != NULL);
-	WARN_ON(ctx->engine[ring->id].state);
+	WARN_ON(ctx->engine[engine->id].state);
 
-	context_size = round_up(intel_lr_context_size(ring), 4096);
+	context_size = round_up(intel_lr_context_size(engine), 4096);
 
 	/* One extra page as the sharing data between driver and GuC */
 	context_size += PAGE_SIZE * LRC_PPHWSP_PN;
@@ -2586,39 +2671,38 @@ int intel_lr_context_deferred_alloc(struct intel_context *ctx,
 		return -ENOMEM;
 	}
 
-	ringbuf = intel_engine_create_ringbuffer(ring, 4 * PAGE_SIZE);
+	ringbuf = intel_engine_create_ringbuffer(engine, 4 * PAGE_SIZE);
 	if (IS_ERR(ringbuf)) {
 		ret = PTR_ERR(ringbuf);
 		goto error_deref_obj;
 	}
 
-	ret = populate_lr_context(ctx, ctx_obj, ring, ringbuf);
+	ret = populate_lr_context(ctx, ctx_obj, engine, ringbuf);
 	if (ret) {
 		DRM_DEBUG_DRIVER("Failed to populate LRC: %d\n", ret);
 		goto error_ringbuf;
 	}
 
-	ctx->engine[ring->id].ringbuf = ringbuf;
-	ctx->engine[ring->id].state = ctx_obj;
+	ctx->engine[engine->id].ringbuf = ringbuf;
+	ctx->engine[engine->id].state = ctx_obj;
 
-	if (ctx != ctx->i915->kernel_context && ring->init_context) {
+	if (ctx != ctx->i915->kernel_context && engine->init_context) {
 		struct drm_i915_gem_request *req;
 
-		req = i915_gem_request_alloc(ring, ctx);
+		req = i915_gem_request_alloc(engine, ctx);
 		if (IS_ERR(req)) {
 			ret = PTR_ERR(req);
 			DRM_ERROR("ring create req: %d\n", ret);
 			goto error_ringbuf;
 		}
 
-		ret = ring->init_context(req);
+		ret = engine->init_context(req);
+		i915_add_request_no_flush(req);
 		if (ret) {
 			DRM_ERROR("ring init context: %d\n",
 				ret);
-			i915_gem_request_cancel(req);
 			goto error_ringbuf;
 		}
-		i915_add_request_no_flush(req);
 	}
 	return 0;
 
@@ -2626,40 +2710,38 @@ error_ringbuf:
 	intel_ringbuffer_free(ringbuf);
 error_deref_obj:
 	drm_gem_object_unreference(&ctx_obj->base);
-	ctx->engine[ring->id].ringbuf = NULL;
-	ctx->engine[ring->id].state = NULL;
+	ctx->engine[engine->id].ringbuf = NULL;
+	ctx->engine[engine->id].state = NULL;
 	return ret;
 }
 
-void intel_lr_context_reset(struct drm_device *dev,
-			struct intel_context *ctx)
+void intel_lr_context_reset(struct drm_i915_private *dev_priv,
+			    struct intel_context *ctx)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
-	int i;
+	struct intel_engine_cs *engine;
 
-	for_each_ring(ring, dev_priv, i) {
+	for_each_engine(engine, dev_priv) {
 		struct drm_i915_gem_object *ctx_obj =
-				ctx->engine[ring->id].state;
+				ctx->engine[engine->id].state;
 		struct intel_ringbuffer *ringbuf =
-				ctx->engine[ring->id].ringbuf;
+				ctx->engine[engine->id].ringbuf;
+		void *vaddr;
 		uint32_t *reg_state;
-		struct page *page;
 
 		if (!ctx_obj)
 			continue;
 
-		if (i915_gem_object_get_pages(ctx_obj)) {
-			WARN(1, "Failed get_pages for context obj\n");
+		vaddr = i915_gem_object_pin_map(ctx_obj);
+		if (WARN_ON(IS_ERR(vaddr)))
 			continue;
-		}
-		page = i915_gem_object_get_dirty_page(ctx_obj, LRC_STATE_PN);
-		reg_state = kmap_atomic(page);
+
+		reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
+		ctx_obj->dirty = true;
 
 		reg_state[CTX_RING_HEAD+1] = 0;
 		reg_state[CTX_RING_TAIL+1] = 0;
 
-		kunmap_atomic(reg_state);
+		i915_gem_object_unpin_map(ctx_obj);
 
 		ringbuf->head = 0;
 		ringbuf->tail = 0;
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index e6cda3e225d0..461f1ef9b5c1 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -24,6 +24,8 @@
 #ifndef _INTEL_LRC_H_
 #define _INTEL_LRC_H_
 
+#include "intel_ringbuffer.h"
+
 #define GEN8_LR_CONTEXT_ALIGN 4096
 
 /* Execlists regs */
@@ -34,6 +36,7 @@
 #define	  CTX_CTRL_INHIBIT_SYN_CTX_SWITCH	(1 << 3)
 #define	  CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT	(1 << 0)
 #define   CTX_CTRL_RS_CTX_ENABLE                (1 << 1)
+#define RING_CONTEXT_STATUS_BUF_BASE(ring)	_MMIO((ring)->mmio_base + 0x370)
 #define RING_CONTEXT_STATUS_BUF_LO(ring, i)	_MMIO((ring)->mmio_base + 0x370 + (i) * 8)
 #define RING_CONTEXT_STATUS_BUF_HI(ring, i)	_MMIO((ring)->mmio_base + 0x370 + (i) * 8 + 4)
 #define RING_CONTEXT_STATUS_PTR(ring)		_MMIO((ring)->mmio_base + 0x3a0)
@@ -57,8 +60,8 @@
 /* Logical Rings */
 int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request);
 int intel_logical_ring_reserve_space(struct drm_i915_gem_request *request);
-void intel_logical_ring_stop(struct intel_engine_cs *ring);
-void intel_logical_ring_cleanup(struct intel_engine_cs *ring);
+void intel_logical_ring_stop(struct intel_engine_cs *engine);
+void intel_logical_ring_cleanup(struct intel_engine_cs *engine);
 int intel_logical_rings_init(struct drm_device *dev);
 int intel_logical_ring_begin(struct drm_i915_gem_request *req, int num_dwords);
 
@@ -98,18 +101,21 @@ static inline void intel_logical_ring_emit_reg(struct intel_ringbuffer *ringbuf,
 #define LRC_STATE_PN	(LRC_PPHWSP_PN + 1)
 
 void intel_lr_context_free(struct intel_context *ctx);
-uint32_t intel_lr_context_size(struct intel_engine_cs *ring);
+uint32_t intel_lr_context_size(struct intel_engine_cs *engine);
 int intel_lr_context_deferred_alloc(struct intel_context *ctx,
-				    struct intel_engine_cs *ring);
+				    struct intel_engine_cs *engine);
 void intel_lr_context_unpin(struct intel_context *ctx,
 			    struct intel_engine_cs *engine);
-void intel_lr_context_reset(struct drm_device *dev,
-			struct intel_context *ctx);
+
+struct drm_i915_private;
+
+void intel_lr_context_reset(struct drm_i915_private *dev_priv,
+			    struct intel_context *ctx);
 uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
-				     struct intel_engine_cs *ring);
+				     struct intel_engine_cs *engine);
 
 u32 intel_execlists_ctx_id(struct intel_context *ctx,
-			   struct intel_engine_cs *ring);
+			   struct intel_engine_cs *engine);
 
 /* Execlists */
 int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists);
@@ -118,7 +124,6 @@ int intel_execlists_submission(struct i915_execbuffer_params *params,
 			       struct drm_i915_gem_execbuffer2 *args,
 			       struct list_head *vmas);
 
-void intel_lrc_irq_handler(struct intel_engine_cs *ring);
-void intel_execlists_retire_requests(struct intel_engine_cs *ring);
+void intel_execlists_retire_requests(struct intel_engine_cs *engine);
 
 #endif /* _INTEL_LRC_H_ */
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 10dc3517b63b..bc53c0dd34d0 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -109,7 +109,6 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
 	u32 tmp, flags = 0;
-	int dotclock;
 
 	tmp = I915_READ(lvds_encoder->reg);
 	if (tmp & LVDS_HSYNC_POLARITY)
@@ -134,12 +133,7 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 		pipe_config->gmch_pfit.control |= tmp & PANEL_8TO6_DITHER_ENABLE;
 	}
 
-	dotclock = pipe_config->port_clock;
-
-	if (HAS_PCH_SPLIT(dev_priv->dev))
-		ironlake_check_encoder_dotclock(pipe_config, dotclock);
-
-	pipe_config->base.adjusted_mode.crtc_clock = dotclock;
+	pipe_config->base.adjusted_mode.crtc_clock = pipe_config->port_clock;
 }
 
 static void intel_pre_enable_lvds(struct intel_encoder *encoder)
@@ -155,7 +149,7 @@ static void intel_pre_enable_lvds(struct intel_encoder *encoder)
 	if (HAS_PCH_SPLIT(dev)) {
 		assert_fdi_rx_pll_disabled(dev_priv, pipe);
 		assert_shared_dpll_disabled(dev_priv,
-					    intel_crtc_to_shared_dpll(crtc));
+					    crtc->config->shared_dpll);
 	} else {
 		assert_pll_disabled(dev_priv, pipe);
 	}
@@ -782,57 +776,6 @@ static const struct dmi_system_id intel_no_lvds[] = {
 	{ }	/* terminating entry */
 };
 
-/*
- * Enumerate the child dev array parsed from VBT to check whether
- * the LVDS is present.
- * If it is present, return 1.
- * If it is not present, return false.
- * If no child dev is parsed from VBT, it assumes that the LVDS is present.
- */
-static bool lvds_is_present_in_vbt(struct drm_device *dev,
-				   u8 *i2c_pin)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int i;
-
-	if (!dev_priv->vbt.child_dev_num)
-		return true;
-
-	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
-		union child_device_config *uchild = dev_priv->vbt.child_dev + i;
-		struct old_child_dev_config *child = &uchild->old;
-
-		/* If the device type is not LFP, continue.
-		 * We have to check both the new identifiers as well as the
-		 * old for compatibility with some BIOSes.
-		 */
-		if (child->device_type != DEVICE_TYPE_INT_LFP &&
-		    child->device_type != DEVICE_TYPE_LFP)
-			continue;
-
-		if (intel_gmbus_is_valid_pin(dev_priv, child->i2c_pin))
-			*i2c_pin = child->i2c_pin;
-
-		/* However, we cannot trust the BIOS writers to populate
-		 * the VBT correctly.  Since LVDS requires additional
-		 * information from AIM blocks, a non-zero addin offset is
-		 * a good indicator that the LVDS is actually present.
-		 */
-		if (child->addin_offset)
-			return true;
-
-		/* But even then some BIOS writers perform some black magic
-		 * and instantiate the device without reference to any
-		 * additional data.  Trust that if the VBT was written into
-		 * the OpRegion then they have validated the LVDS's existence.
-		 */
-		if (dev_priv->opregion.vbt)
-			return true;
-	}
-
-	return false;
-}
-
 static int intel_dual_link_lvds_callback(const struct dmi_system_id *id)
 {
 	DRM_INFO("Forcing lvds to dual link mode on %s\n", id->ident);
@@ -982,14 +925,14 @@ void intel_lvds_init(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		if ((lvds & LVDS_DETECTED) == 0)
 			return;
-		if (dev_priv->vbt.edp_support) {
+		if (dev_priv->vbt.edp.support) {
 			DRM_DEBUG_KMS("disable LVDS for eDP support\n");
 			return;
 		}
 	}
 
 	pin = GMBUS_PIN_PANEL;
-	if (!lvds_is_present_in_vbt(dev, &pin)) {
+	if (!intel_bios_is_lvds_present(dev_priv, &pin)) {
 		if ((lvds & LVDS_PORT_EN) == 0) {
 			DRM_DEBUG_KMS("LVDS is not present in VBT\n");
 			return;
diff --git a/drivers/gpu/drm/i915/intel_mocs.c b/drivers/gpu/drm/i915/intel_mocs.c
index fed7bea19cc9..23b8545ad6b0 100644
--- a/drivers/gpu/drm/i915/intel_mocs.c
+++ b/drivers/gpu/drm/i915/intel_mocs.c
@@ -128,9 +128,9 @@ static const struct drm_i915_mocs_entry broxton_mocs_table[] = {
 
 /**
  * get_mocs_settings()
- * @dev:        DRM device.
+ * @dev_priv:	i915 device.
  * @table:      Output table that will be made to point at appropriate
- *              MOCS values for the device.
+ *	      MOCS values for the device.
  *
  * This function will return the values of the MOCS table that needs to
  * be programmed for the platform. It will return the values that need
@@ -138,28 +138,28 @@ static const struct drm_i915_mocs_entry broxton_mocs_table[] = {
  *
  * Return: true if there are applicable MOCS settings for the device.
  */
-static bool get_mocs_settings(struct drm_device *dev,
+static bool get_mocs_settings(struct drm_i915_private *dev_priv,
 			      struct drm_i915_mocs_table *table)
 {
 	bool result = false;
 
-	if (IS_SKYLAKE(dev) || IS_KABYLAKE(dev)) {
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
 		table->size  = ARRAY_SIZE(skylake_mocs_table);
 		table->table = skylake_mocs_table;
 		result = true;
-	} else if (IS_BROXTON(dev)) {
+	} else if (IS_BROXTON(dev_priv)) {
 		table->size  = ARRAY_SIZE(broxton_mocs_table);
 		table->table = broxton_mocs_table;
 		result = true;
 	} else {
-		WARN_ONCE(INTEL_INFO(dev)->gen >= 9,
+		WARN_ONCE(INTEL_INFO(dev_priv)->gen >= 9,
 			  "Platform that should have a MOCS table does not.\n");
 	}
 
 	return result;
 }
 
-static i915_reg_t mocs_register(enum intel_ring_id ring, int index)
+static i915_reg_t mocs_register(enum intel_engine_id ring, int index)
 {
 	switch (ring) {
 	case RCS:
@@ -179,10 +179,49 @@ static i915_reg_t mocs_register(enum intel_ring_id ring, int index)
 }
 
 /**
+ * intel_mocs_init_engine() - emit the mocs control table
+ * @engine:	The engine for whom to emit the registers.
+ *
+ * This function simply emits a MI_LOAD_REGISTER_IMM command for the
+ * given table starting at the given address.
+ *
+ * Return: 0 on success, otherwise the error status.
+ */
+int intel_mocs_init_engine(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *dev_priv = to_i915(engine->dev);
+	struct drm_i915_mocs_table table;
+	unsigned int index;
+
+	if (!get_mocs_settings(dev_priv, &table))
+		return 0;
+
+	if (WARN_ON(table.size > GEN9_NUM_MOCS_ENTRIES))
+		return -ENODEV;
+
+	for (index = 0; index < table.size; index++)
+		I915_WRITE(mocs_register(engine->id, index),
+			   table.table[index].control_value);
+
+	/*
+	 * Ok, now set the unused entries to uncached. These entries
+	 * are officially undefined and no contract for the contents
+	 * and settings is given for these entries.
+	 *
+	 * Entry 0 in the table is uncached - so we are just writing
+	 * that value to all the used entries.
+	 */
+	for (; index < GEN9_NUM_MOCS_ENTRIES; index++)
+		I915_WRITE(mocs_register(engine->id, index),
+			   table.table[0].control_value);
+
+	return 0;
+}
+
+/**
  * emit_mocs_control_table() - emit the mocs control table
  * @req:	Request to set up the MOCS table for.
  * @table:	The values to program into the control regs.
- * @ring:	The engine for whom to emit the registers.
  *
  * This function simply emits a MI_LOAD_REGISTER_IMM command for the
  * given table starting at the given address.
@@ -190,10 +229,10 @@ static i915_reg_t mocs_register(enum intel_ring_id ring, int index)
  * Return: 0 on success, otherwise the error status.
  */
 static int emit_mocs_control_table(struct drm_i915_gem_request *req,
-				   const struct drm_i915_mocs_table *table,
-				   enum intel_ring_id ring)
+				   const struct drm_i915_mocs_table *table)
 {
 	struct intel_ringbuffer *ringbuf = req->ringbuf;
+	enum intel_engine_id engine = req->engine->id;
 	unsigned int index;
 	int ret;
 
@@ -210,7 +249,8 @@ static int emit_mocs_control_table(struct drm_i915_gem_request *req,
 				MI_LOAD_REGISTER_IMM(GEN9_NUM_MOCS_ENTRIES));
 
 	for (index = 0; index < table->size; index++) {
-		intel_logical_ring_emit_reg(ringbuf, mocs_register(ring, index));
+		intel_logical_ring_emit_reg(ringbuf,
+					    mocs_register(engine, index));
 		intel_logical_ring_emit(ringbuf,
 					table->table[index].control_value);
 	}
@@ -224,8 +264,10 @@ static int emit_mocs_control_table(struct drm_i915_gem_request *req,
 	 * that value to all the used entries.
 	 */
 	for (; index < GEN9_NUM_MOCS_ENTRIES; index++) {
-		intel_logical_ring_emit_reg(ringbuf, mocs_register(ring, index));
-		intel_logical_ring_emit(ringbuf, table->table[0].control_value);
+		intel_logical_ring_emit_reg(ringbuf,
+					    mocs_register(engine, index));
+		intel_logical_ring_emit(ringbuf,
+					table->table[0].control_value);
 	}
 
 	intel_logical_ring_emit(ringbuf, MI_NOOP);
@@ -234,6 +276,14 @@ static int emit_mocs_control_table(struct drm_i915_gem_request *req,
 	return 0;
 }
 
+static inline u32 l3cc_combine(const struct drm_i915_mocs_table *table,
+			       u16 low,
+			       u16 high)
+{
+	return table->table[low].l3cc_value |
+	       table->table[high].l3cc_value << 16;
+}
+
 /**
  * emit_mocs_l3cc_table() - emit the mocs control table
  * @req:	Request to set up the MOCS table for.
@@ -249,11 +299,7 @@ static int emit_mocs_l3cc_table(struct drm_i915_gem_request *req,
 				const struct drm_i915_mocs_table *table)
 {
 	struct intel_ringbuffer *ringbuf = req->ringbuf;
-	unsigned int count;
 	unsigned int i;
-	u32 value;
-	u32 filler = (table->table[0].l3cc_value & 0xffff) |
-			((table->table[0].l3cc_value & 0xffff) << 16);
 	int ret;
 
 	if (WARN_ON(table->size > GEN9_NUM_MOCS_ENTRIES))
@@ -268,20 +314,18 @@ static int emit_mocs_l3cc_table(struct drm_i915_gem_request *req,
 	intel_logical_ring_emit(ringbuf,
 			MI_LOAD_REGISTER_IMM(GEN9_NUM_MOCS_ENTRIES / 2));
 
-	for (i = 0, count = 0; i < table->size / 2; i++, count += 2) {
-		value = (table->table[count].l3cc_value & 0xffff) |
-			((table->table[count + 1].l3cc_value & 0xffff) << 16);
-
+	for (i = 0; i < table->size/2; i++) {
 		intel_logical_ring_emit_reg(ringbuf, GEN9_LNCFCMOCS(i));
-		intel_logical_ring_emit(ringbuf, value);
+		intel_logical_ring_emit(ringbuf,
+					l3cc_combine(table, 2*i, 2*i+1));
 	}
 
 	if (table->size & 0x01) {
 		/* Odd table size - 1 left over */
-		value = (table->table[count].l3cc_value & 0xffff) |
-			((table->table[0].l3cc_value & 0xffff) << 16);
-	} else
-		value = filler;
+		intel_logical_ring_emit_reg(ringbuf, GEN9_LNCFCMOCS(i));
+		intel_logical_ring_emit(ringbuf, l3cc_combine(table, 2*i, 0));
+		i++;
+	}
 
 	/*
 	 * Now set the rest of the table to uncached - use entry 0 as
@@ -290,9 +334,7 @@ static int emit_mocs_l3cc_table(struct drm_i915_gem_request *req,
 	 */
 	for (; i < GEN9_NUM_MOCS_ENTRIES / 2; i++) {
 		intel_logical_ring_emit_reg(ringbuf, GEN9_LNCFCMOCS(i));
-		intel_logical_ring_emit(ringbuf, value);
-
-		value = filler;
+		intel_logical_ring_emit(ringbuf, l3cc_combine(table, 0, 0));
 	}
 
 	intel_logical_ring_emit(ringbuf, MI_NOOP);
@@ -302,6 +344,47 @@ static int emit_mocs_l3cc_table(struct drm_i915_gem_request *req,
 }
 
 /**
+ * intel_mocs_init_l3cc_table() - program the mocs control table
+ * @dev:      The the device to be programmed.
+ *
+ * This function simply programs the mocs registers for the given table
+ * starting at the given address. This register set is  programmed in pairs.
+ *
+ * These registers may get programmed more than once, it is simpler to
+ * re-program 32 registers than maintain the state of when they were programmed.
+ * We are always reprogramming with the same values and this only on context
+ * start.
+ *
+ * Return: Nothing.
+ */
+void intel_mocs_init_l3cc_table(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_mocs_table table;
+	unsigned int i;
+
+	if (!get_mocs_settings(dev_priv, &table))
+		return;
+
+	for (i = 0; i < table.size/2; i++)
+		I915_WRITE(GEN9_LNCFCMOCS(i), l3cc_combine(&table, 2*i, 2*i+1));
+
+	/* Odd table size - 1 left over */
+	if (table.size & 0x01) {
+		I915_WRITE(GEN9_LNCFCMOCS(i), l3cc_combine(&table, 2*i, 0));
+		i++;
+	}
+
+	/*
+	 * Now set the rest of the table to uncached - use entry 0 as
+	 * this will be uncached. Leave the last pair as initialised as
+	 * they are reserved by the hardware.
+	 */
+	for (; i < (GEN9_NUM_MOCS_ENTRIES / 2); i++)
+		I915_WRITE(GEN9_LNCFCMOCS(i), l3cc_combine(&table, 0, 0));
+}
+
+/**
  * intel_rcs_context_init_mocs() - program the MOCS register.
  * @req:	Request to set up the MOCS tables for.
  *
@@ -322,17 +405,11 @@ int intel_rcs_context_init_mocs(struct drm_i915_gem_request *req)
 	struct drm_i915_mocs_table t;
 	int ret;
 
-	if (get_mocs_settings(req->ring->dev, &t)) {
-		struct drm_i915_private *dev_priv = req->i915;
-		struct intel_engine_cs *ring;
-		enum intel_ring_id ring_id;
-
-		/* Program the control registers */
-		for_each_ring(ring, dev_priv, ring_id) {
-			ret = emit_mocs_control_table(req, &t, ring_id);
-			if (ret)
-				return ret;
-		}
+	if (get_mocs_settings(req->i915, &t)) {
+		/* Program the RCS control registers */
+		ret = emit_mocs_control_table(req, &t);
+		if (ret)
+			return ret;
 
 		/* Now program the l3cc registers */
 		ret = emit_mocs_l3cc_table(req, &t);
diff --git a/drivers/gpu/drm/i915/intel_mocs.h b/drivers/gpu/drm/i915/intel_mocs.h
index 76e45b1748b3..4640299e04ec 100644
--- a/drivers/gpu/drm/i915/intel_mocs.h
+++ b/drivers/gpu/drm/i915/intel_mocs.h
@@ -53,5 +53,7 @@
 #include "i915_drv.h"
 
 int intel_rcs_context_init_mocs(struct drm_i915_gem_request *req);
+void intel_mocs_init_l3cc_table(struct drm_device *dev);
+int intel_mocs_init_engine(struct intel_engine_cs *ring);
 
 #endif
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index c15718b4862a..99e26034ae8d 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -34,12 +34,6 @@
 #include "i915_drv.h"
 #include "intel_drv.h"
 
-#define PCI_ASLE		0xe4
-#define PCI_ASLS		0xfc
-#define PCI_SWSCI		0xe8
-#define PCI_SWSCI_SCISEL	(1 << 15)
-#define PCI_SWSCI_GSSCIE	(1 << 0)
-
 #define OPREGION_HEADER_OFFSET 0
 #define OPREGION_ACPI_OFFSET   0x100
 #define   ACPI_CLID 0x01ac /* current lid state indicator */
@@ -246,13 +240,12 @@ struct opregion_asle_ext {
 
 #define MAX_DSLP	1500
 
-#ifdef CONFIG_ACPI
 static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct opregion_swsci *swsci = dev_priv->opregion.swsci;
 	u32 main_function, sub_function, scic;
-	u16 pci_swsci;
+	u16 swsci_val;
 	u32 dslp;
 
 	if (!swsci)
@@ -300,16 +293,16 @@ static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 	swsci->scic = scic;
 
 	/* Ensure SCI event is selected and event trigger is cleared. */
-	pci_read_config_word(dev->pdev, PCI_SWSCI, &pci_swsci);
-	if (!(pci_swsci & PCI_SWSCI_SCISEL) || (pci_swsci & PCI_SWSCI_GSSCIE)) {
-		pci_swsci |= PCI_SWSCI_SCISEL;
-		pci_swsci &= ~PCI_SWSCI_GSSCIE;
-		pci_write_config_word(dev->pdev, PCI_SWSCI, pci_swsci);
+	pci_read_config_word(dev->pdev, SWSCI, &swsci_val);
+	if (!(swsci_val & SWSCI_SCISEL) || (swsci_val & SWSCI_GSSCIE)) {
+		swsci_val |= SWSCI_SCISEL;
+		swsci_val &= ~SWSCI_GSSCIE;
+		pci_write_config_word(dev->pdev, SWSCI, swsci_val);
 	}
 
 	/* Use event trigger to tell bios to check the mail. */
-	pci_swsci |= PCI_SWSCI_GSSCIE;
-	pci_write_config_word(dev->pdev, PCI_SWSCI, pci_swsci);
+	swsci_val |= SWSCI_GSSCIE;
+	pci_write_config_word(dev->pdev, SWSCI, swsci_val);
 
 	/* Poll for the result. */
 #define C (((scic = swsci->scic) & SWSCI_SCIC_INDICATOR) == 0)
@@ -905,9 +898,6 @@ static void swsci_setup(struct drm_device *dev)
 			 opregion->swsci_gbda_sub_functions,
 			 opregion->swsci_sbcb_sub_functions);
 }
-#else /* CONFIG_ACPI */
-static inline void swsci_setup(struct drm_device *dev) {}
-#endif  /* CONFIG_ACPI */
 
 static int intel_no_opregion_vbt_callback(const struct dmi_system_id *id)
 {
@@ -943,16 +933,14 @@ int intel_opregion_setup(struct drm_device *dev)
 	BUILD_BUG_ON(sizeof(struct opregion_asle) != 0x100);
 	BUILD_BUG_ON(sizeof(struct opregion_asle_ext) != 0x400);
 
-	pci_read_config_dword(dev->pdev, PCI_ASLS, &asls);
+	pci_read_config_dword(dev->pdev, ASLS, &asls);
 	DRM_DEBUG_DRIVER("graphic opregion physical addr: 0x%x\n", asls);
 	if (asls == 0) {
 		DRM_DEBUG_DRIVER("ACPI OpRegion not supported!\n");
 		return -ENOTSUPP;
 	}
 
-#ifdef CONFIG_ACPI
 	INIT_WORK(&opregion->asle_work, asle_work);
-#endif
 
 	base = memremap(asls, OPREGION_SIZE, MEMREMAP_WB);
 	if (!base)
@@ -1024,3 +1012,31 @@ err_out:
 	memunmap(base);
 	return err;
 }
+
+int
+intel_opregion_get_panel_type(struct drm_device *dev)
+{
+	u32 panel_details;
+	int ret;
+
+	ret = swsci(dev, SWSCI_GBDA_PANEL_DETAILS, 0x0, &panel_details);
+	if (ret) {
+		DRM_DEBUG_KMS("Failed to get panel details from OpRegion (%d)\n",
+			      ret);
+		return ret;
+	}
+
+	ret = (panel_details >> 8) & 0xff;
+	if (ret > 0x10) {
+		DRM_DEBUG_KMS("Invalid OpRegion panel type 0x%x\n", ret);
+		return -EINVAL;
+	}
+
+	/* fall back to VBT panel type? */
+	if (ret == 0x0) {
+		DRM_DEBUG_KMS("No panel type in OpRegion\n");
+		return -ENODEV;
+	}
+
+	return ret - 1;
+}
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 9168413fe204..bd38e49f7334 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -190,13 +190,14 @@ struct intel_overlay {
 static struct overlay_registers __iomem *
 intel_overlay_map_regs(struct intel_overlay *overlay)
 {
-	struct drm_i915_private *dev_priv = overlay->dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(overlay->dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct overlay_registers __iomem *regs;
 
 	if (OVERLAY_NEEDS_PHYSICAL(overlay->dev))
 		regs = (struct overlay_registers __iomem *)overlay->reg_bo->phys_handle->vaddr;
 	else
-		regs = io_mapping_map_wc(dev_priv->gtt.mappable,
+		regs = io_mapping_map_wc(ggtt->mappable,
 					 i915_gem_obj_ggtt_offset(overlay->reg_bo));
 
 	return regs;
@@ -233,30 +234,30 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	struct drm_i915_gem_request *req;
 	int ret;
 
 	WARN_ON(overlay->active);
 	WARN_ON(IS_I830(dev) && !(dev_priv->quirks & QUIRK_PIPEA_FORCE));
 
-	req = i915_gem_request_alloc(ring, NULL);
+	req = i915_gem_request_alloc(engine, NULL);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
 	ret = intel_ring_begin(req, 4);
 	if (ret) {
-		i915_gem_request_cancel(req);
+		i915_add_request_no_flush(req);
 		return ret;
 	}
 
 	overlay->active = true;
 
-	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_ON);
-	intel_ring_emit(ring, overlay->flip_addr | OFC_UPDATE);
-	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_ON);
+	intel_ring_emit(engine, overlay->flip_addr | OFC_UPDATE);
+	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	return intel_overlay_do_wait_request(overlay, req, NULL);
 }
@@ -267,7 +268,7 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	struct drm_i915_gem_request *req;
 	u32 flip_addr = overlay->flip_addr;
 	u32 tmp;
@@ -283,19 +284,19 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 	if (tmp & (1 << 17))
 		DRM_DEBUG("overlay underrun, DOVSTA: %x\n", tmp);
 
-	req = i915_gem_request_alloc(ring, NULL);
+	req = i915_gem_request_alloc(engine, NULL);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
 	ret = intel_ring_begin(req, 2);
 	if (ret) {
-		i915_gem_request_cancel(req);
+		i915_add_request_no_flush(req);
 		return ret;
 	}
 
-	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
-	intel_ring_emit(ring, flip_addr);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
+	intel_ring_emit(engine, flip_addr);
+	intel_ring_advance(engine);
 
 	WARN_ON(overlay->last_flip_req);
 	i915_gem_request_assign(&overlay->last_flip_req, req);
@@ -336,7 +337,7 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	struct drm_i915_gem_request *req;
 	u32 flip_addr = overlay->flip_addr;
 	int ret;
@@ -349,33 +350,34 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 	 * of the hw. Do it in both cases */
 	flip_addr |= OFC_UPDATE;
 
-	req = i915_gem_request_alloc(ring, NULL);
+	req = i915_gem_request_alloc(engine, NULL);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
 	ret = intel_ring_begin(req, 6);
 	if (ret) {
-		i915_gem_request_cancel(req);
+		i915_add_request_no_flush(req);
 		return ret;
 	}
 
 	/* wait for overlay to go idle */
-	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
-	intel_ring_emit(ring, flip_addr);
-	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
+	intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
+	intel_ring_emit(engine, flip_addr);
+	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
 	/* turn overlay off */
 	if (IS_I830(dev)) {
 		/* Workaround: Don't disable the overlay fully, since otherwise
 		 * it dies on the next OVERLAY_ON cmd. */
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_emit(engine, MI_NOOP);
 	} else {
-		intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_OFF);
-		intel_ring_emit(ring, flip_addr);
-		intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
+		intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_OFF);
+		intel_ring_emit(engine, flip_addr);
+		intel_ring_emit(engine,
+				MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
 	}
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 
 	return intel_overlay_do_wait_request(overlay, req, intel_overlay_off_tail);
 }
@@ -408,7 +410,7 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	int ret;
 
 	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
@@ -423,19 +425,20 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 		/* synchronous slowpath */
 		struct drm_i915_gem_request *req;
 
-		req = i915_gem_request_alloc(ring, NULL);
+		req = i915_gem_request_alloc(engine, NULL);
 		if (IS_ERR(req))
 			return PTR_ERR(req);
 
 		ret = intel_ring_begin(req, 2);
 		if (ret) {
-			i915_gem_request_cancel(req);
+			i915_add_request_no_flush(req);
 			return ret;
 		}
 
-		intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_advance(ring);
+		intel_ring_emit(engine,
+				MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
+		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_advance(engine);
 
 		ret = intel_overlay_do_wait_request(overlay, req,
 						    intel_overlay_release_old_vid_tail);
@@ -1124,7 +1127,7 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 	}
 	crtc = to_intel_crtc(drmmode_crtc);
 
-	new_bo = to_intel_bo(drm_gem_object_lookup(dev, file_priv,
+	new_bo = to_intel_bo(drm_gem_object_lookup(file_priv,
 						   put_image_rec->bo_handle));
 	if (&new_bo->base == NULL) {
 		ret = -ENOENT;
@@ -1479,7 +1482,8 @@ struct intel_overlay_error_state {
 static struct overlay_registers __iomem *
 intel_overlay_map_regs_atomic(struct intel_overlay *overlay)
 {
-	struct drm_i915_private *dev_priv = overlay->dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(overlay->dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct overlay_registers __iomem *regs;
 
 	if (OVERLAY_NEEDS_PHYSICAL(overlay->dev))
@@ -1488,7 +1492,7 @@ intel_overlay_map_regs_atomic(struct intel_overlay *overlay)
 		regs = (struct overlay_registers __iomem *)
 			overlay->reg_bo->phys_handle->vaddr;
 	else
-		regs = io_mapping_map_atomic_wc(dev_priv->gtt.mappable,
+		regs = io_mapping_map_atomic_wc(ggtt->mappable,
 						i915_gem_obj_ggtt_offset(overlay->reg_bo));
 
 	return regs;
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 21ee6477bf98..a0788763757b 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -504,7 +504,7 @@ static u32 i9xx_get_backlight(struct intel_connector *connector)
 	if (panel->backlight.combination_mode) {
 		u8 lbpc;
 
-		pci_read_config_byte(dev_priv->dev->pdev, PCI_LBPC, &lbpc);
+		pci_read_config_byte(dev_priv->dev->pdev, LBPC, &lbpc);
 		val *= lbpc;
 	}
 
@@ -592,7 +592,7 @@ static void i9xx_set_backlight(struct intel_connector *connector, u32 level)
 
 		lbpc = level * 0xfe / panel->backlight.max + 1;
 		level /= lbpc;
-		pci_write_config_byte(dev_priv->dev->pdev, PCI_LBPC, lbpc);
+		pci_write_config_byte(dev_priv->dev->pdev, LBPC, lbpc);
 	}
 
 	if (IS_GEN4(dev_priv)) {
@@ -1240,7 +1240,7 @@ static void intel_backlight_device_unregister(struct intel_connector *connector)
  */
 static u32 bxt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 {
-	return KHz(19200) / pwm_freq_hz;
+	return DIV_ROUND_CLOSEST(KHz(19200), pwm_freq_hz);
 }
 
 /*
@@ -1251,16 +1251,14 @@ static u32 bxt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 static u32 spt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
-	u32 mul, clock;
+	u32 mul;
 
 	if (I915_READ(SOUTH_CHICKEN1) & SPT_PWM_GRANULARITY)
 		mul = 128;
 	else
 		mul = 16;
 
-	clock = MHz(24);
-
-	return clock / (pwm_freq_hz * mul);
+	return DIV_ROUND_CLOSEST(MHz(24), pwm_freq_hz * mul);
 }
 
 /*
@@ -1283,7 +1281,7 @@ static u32 lpt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 	else
 		clock = MHz(24); /* LPT:LP */
 
-	return clock / (pwm_freq_hz * mul);
+	return DIV_ROUND_CLOSEST(clock, pwm_freq_hz * mul);
 }
 
 /*
@@ -1292,10 +1290,9 @@ static u32 lpt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
  */
 static u32 pch_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 {
-	struct drm_device *dev = connector->base.dev;
-	int clock = MHz(intel_pch_rawclk(dev));
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 
-	return clock / (pwm_freq_hz * 128);
+	return DIV_ROUND_CLOSEST(KHz(dev_priv->rawclk_freq), pwm_freq_hz * 128);
 }
 
 /*
@@ -1308,16 +1305,15 @@ static u32 pch_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
  */
 static u32 i9xx_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 {
-	struct drm_device *dev = connector->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 	int clock;
 
-	if (IS_PINEVIEW(dev))
-		clock = MHz(intel_hrawclk(dev));
+	if (IS_PINEVIEW(dev_priv))
+		clock = KHz(dev_priv->rawclk_freq);
 	else
-		clock = 1000 * dev_priv->cdclk_freq;
+		clock = KHz(dev_priv->cdclk_freq);
 
-	return clock / (pwm_freq_hz * 32);
+	return DIV_ROUND_CLOSEST(clock, pwm_freq_hz * 32);
 }
 
 /*
@@ -1332,11 +1328,11 @@ static u32 i965_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 	int clock;
 
 	if (IS_G4X(dev_priv))
-		clock = MHz(intel_hrawclk(dev));
+		clock = KHz(dev_priv->rawclk_freq);
 	else
-		clock = 1000 * dev_priv->cdclk_freq;
+		clock = KHz(dev_priv->cdclk_freq);
 
-	return clock / (pwm_freq_hz * 128);
+	return DIV_ROUND_CLOSEST(clock, pwm_freq_hz * 128);
 }
 
 /*
@@ -1346,19 +1342,21 @@ static u32 i965_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
  */
 static u32 vlv_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 {
-	struct drm_device *dev = connector->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int clock;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	int mul, clock;
 
 	if ((I915_READ(CBR1_VLV) & CBR_PWM_CLOCK_MUX_SELECT) == 0) {
-		if (IS_CHERRYVIEW(dev))
-			return KHz(19200) / (pwm_freq_hz * 16);
+		if (IS_CHERRYVIEW(dev_priv))
+			clock = KHz(19200);
 		else
-			return MHz(25) / (pwm_freq_hz * 16);
+			clock = MHz(25);
+		mul = 16;
 	} else {
-		clock = intel_hrawclk(dev);
-		return MHz(clock) / (pwm_freq_hz * 128);
+		clock = KHz(dev_priv->rawclk_freq);
+		mul = 128;
 	}
+
+	return DIV_ROUND_CLOSEST(clock, pwm_freq_hz * mul);
 }
 
 static u32 get_backlight_max_vbt(struct intel_connector *connector)
@@ -1745,7 +1743,7 @@ intel_panel_init_backlight_funcs(struct intel_panel *panel)
 		panel->backlight.get = pch_get_backlight;
 		panel->backlight.hz_to_pwm = pch_hz_to_pwm;
 	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
-		if (dev_priv->vbt.has_mipi) {
+		if (connector->base.connector_type == DRM_MODE_CONNECTOR_DSI) {
 			panel->backlight.setup = pwm_setup_backlight;
 			panel->backlight.enable = pwm_enable_backlight;
 			panel->backlight.disable = pwm_disable_backlight;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 3425d8e737b3..4b60005cda37 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -487,20 +487,6 @@ static const struct intel_watermark_params g4x_cursor_wm_info = {
 	.guard_size = 2,
 	.cacheline_size = G4X_FIFO_LINE_SIZE,
 };
-static const struct intel_watermark_params valleyview_wm_info = {
-	.fifo_size = VALLEYVIEW_FIFO_SIZE,
-	.max_wm = VALLEYVIEW_MAX_WM,
-	.default_wm = VALLEYVIEW_MAX_WM,
-	.guard_size = 2,
-	.cacheline_size = G4X_FIFO_LINE_SIZE,
-};
-static const struct intel_watermark_params valleyview_cursor_wm_info = {
-	.fifo_size = I965_CURSOR_FIFO,
-	.max_wm = VALLEYVIEW_CURSOR_MAX_WM,
-	.default_wm = I965_CURSOR_DFT_WM,
-	.guard_size = 2,
-	.cacheline_size = G4X_FIFO_LINE_SIZE,
-};
 static const struct intel_watermark_params i965_cursor_wm_info = {
 	.fifo_size = I965_CURSOR_FIFO,
 	.max_wm = I965_CURSOR_MAX_WM,
@@ -2010,11 +1996,18 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 		cur_latency *= 5;
 	}
 
-	result->pri_val = ilk_compute_pri_wm(cstate, pristate,
-					     pri_latency, level);
-	result->spr_val = ilk_compute_spr_wm(cstate, sprstate, spr_latency);
-	result->cur_val = ilk_compute_cur_wm(cstate, curstate, cur_latency);
-	result->fbc_val = ilk_compute_fbc_wm(cstate, pristate, result->pri_val);
+	if (pristate) {
+		result->pri_val = ilk_compute_pri_wm(cstate, pristate,
+						     pri_latency, level);
+		result->fbc_val = ilk_compute_fbc_wm(cstate, pristate, result->pri_val);
+	}
+
+	if (sprstate)
+		result->spr_val = ilk_compute_spr_wm(cstate, sprstate, spr_latency);
+
+	if (curstate)
+		result->cur_val = ilk_compute_cur_wm(cstate, curstate, cur_latency);
+
 	result->enable = true;
 }
 
@@ -2278,100 +2271,171 @@ static void skl_setup_wm_latency(struct drm_device *dev)
 	intel_print_wm_latency(dev, "Gen9 Plane", dev_priv->wm.skl_latency);
 }
 
+static bool ilk_validate_pipe_wm(struct drm_device *dev,
+				 struct intel_pipe_wm *pipe_wm)
+{
+	/* LP0 watermark maximums depend on this pipe alone */
+	const struct intel_wm_config config = {
+		.num_pipes_active = 1,
+		.sprites_enabled = pipe_wm->sprites_enabled,
+		.sprites_scaled = pipe_wm->sprites_scaled,
+	};
+	struct ilk_wm_maximums max;
+
+	/* LP0 watermarks always use 1/2 DDB partitioning */
+	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
+
+	/* At least LP0 must be valid */
+	if (!ilk_validate_wm_level(0, &max, &pipe_wm->wm[0])) {
+		DRM_DEBUG_KMS("LP0 watermark invalid\n");
+		return false;
+	}
+
+	return true;
+}
+
 /* Compute new watermarks for the pipe */
-static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
-			       struct drm_atomic_state *state)
+static int ilk_compute_pipe_wm(struct intel_crtc_state *cstate)
 {
+	struct drm_atomic_state *state = cstate->base.state;
+	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
 	struct intel_pipe_wm *pipe_wm;
-	struct drm_device *dev = intel_crtc->base.dev;
+	struct drm_device *dev = state->dev;
 	const struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc_state *cstate = NULL;
 	struct intel_plane *intel_plane;
-	struct drm_plane_state *ps;
 	struct intel_plane_state *pristate = NULL;
 	struct intel_plane_state *sprstate = NULL;
 	struct intel_plane_state *curstate = NULL;
-	int level, max_level = ilk_wm_max_level(dev);
-	/* LP0 watermark maximums depend on this pipe alone */
-	struct intel_wm_config config = {
-		.num_pipes_active = 1,
-	};
+	int level, max_level = ilk_wm_max_level(dev), usable_level;
 	struct ilk_wm_maximums max;
 
-	cstate = intel_atomic_get_crtc_state(state, intel_crtc);
-	if (IS_ERR(cstate))
-		return PTR_ERR(cstate);
-
 	pipe_wm = &cstate->wm.optimal.ilk;
-	memset(pipe_wm, 0, sizeof(*pipe_wm));
 
 	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
-		ps = drm_atomic_get_plane_state(state,
-						&intel_plane->base);
-		if (IS_ERR(ps))
-			return PTR_ERR(ps);
+		struct intel_plane_state *ps;
+
+		ps = intel_atomic_get_existing_plane_state(state,
+							   intel_plane);
+		if (!ps)
+			continue;
 
 		if (intel_plane->base.type == DRM_PLANE_TYPE_PRIMARY)
-			pristate = to_intel_plane_state(ps);
+			pristate = ps;
 		else if (intel_plane->base.type == DRM_PLANE_TYPE_OVERLAY)
-			sprstate = to_intel_plane_state(ps);
+			sprstate = ps;
 		else if (intel_plane->base.type == DRM_PLANE_TYPE_CURSOR)
-			curstate = to_intel_plane_state(ps);
+			curstate = ps;
 	}
 
-	config.sprites_enabled = sprstate->visible;
-	config.sprites_scaled = sprstate->visible &&
-		(drm_rect_width(&sprstate->dst) != drm_rect_width(&sprstate->src) >> 16 ||
-		drm_rect_height(&sprstate->dst) != drm_rect_height(&sprstate->src) >> 16);
-
 	pipe_wm->pipe_enabled = cstate->base.active;
-	pipe_wm->sprites_enabled = config.sprites_enabled;
-	pipe_wm->sprites_scaled = config.sprites_scaled;
+	if (sprstate) {
+		pipe_wm->sprites_enabled = sprstate->visible;
+		pipe_wm->sprites_scaled = sprstate->visible &&
+			(drm_rect_width(&sprstate->dst) != drm_rect_width(&sprstate->src) >> 16 ||
+			 drm_rect_height(&sprstate->dst) != drm_rect_height(&sprstate->src) >> 16);
+	}
+
+	usable_level = max_level;
 
 	/* ILK/SNB: LP2+ watermarks only w/o sprites */
-	if (INTEL_INFO(dev)->gen <= 6 && sprstate->visible)
-		max_level = 1;
+	if (INTEL_INFO(dev)->gen <= 6 && pipe_wm->sprites_enabled)
+		usable_level = 1;
 
 	/* ILK/SNB/IVB: LP1+ watermarks only w/o scaling */
-	if (config.sprites_scaled)
-		max_level = 0;
+	if (pipe_wm->sprites_scaled)
+		usable_level = 0;
 
 	ilk_compute_wm_level(dev_priv, intel_crtc, 0, cstate,
-			     pristate, sprstate, curstate, &pipe_wm->wm[0]);
+			     pristate, sprstate, curstate, &pipe_wm->raw_wm[0]);
+
+	memset(&pipe_wm->wm, 0, sizeof(pipe_wm->wm));
+	pipe_wm->wm[0] = pipe_wm->raw_wm[0];
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		pipe_wm->linetime = hsw_compute_linetime_wm(dev, cstate);
 
-	/* LP0 watermarks always use 1/2 DDB partitioning */
-	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
-
-	/* At least LP0 must be valid */
-	if (!ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]))
+	if (!ilk_validate_pipe_wm(dev, pipe_wm))
 		return -EINVAL;
 
 	ilk_compute_wm_reg_maximums(dev, 1, &max);
 
 	for (level = 1; level <= max_level; level++) {
-		struct intel_wm_level wm = {};
+		struct intel_wm_level *wm = &pipe_wm->raw_wm[level];
 
 		ilk_compute_wm_level(dev_priv, intel_crtc, level, cstate,
-				     pristate, sprstate, curstate, &wm);
+				     pristate, sprstate, curstate, wm);
 
 		/*
 		 * Disable any watermark level that exceeds the
 		 * register maximums since such watermarks are
 		 * always invalid.
 		 */
-		if (!ilk_validate_wm_level(level, &max, &wm))
-			break;
+		if (level > usable_level)
+			continue;
 
-		pipe_wm->wm[level] = wm;
+		if (ilk_validate_wm_level(level, &max, wm))
+			pipe_wm->wm[level] = *wm;
+		else
+			usable_level = level;
 	}
 
 	return 0;
 }
 
 /*
+ * Build a set of 'intermediate' watermark values that satisfy both the old
+ * state and the new state.  These can be programmed to the hardware
+ * immediately.
+ */
+static int ilk_compute_intermediate_wm(struct drm_device *dev,
+				       struct intel_crtc *intel_crtc,
+				       struct intel_crtc_state *newstate)
+{
+	struct intel_pipe_wm *a = &newstate->wm.intermediate;
+	struct intel_pipe_wm *b = &intel_crtc->wm.active.ilk;
+	int level, max_level = ilk_wm_max_level(dev);
+
+	/*
+	 * Start with the final, target watermarks, then combine with the
+	 * currently active watermarks to get values that are safe both before
+	 * and after the vblank.
+	 */
+	*a = newstate->wm.optimal.ilk;
+	a->pipe_enabled |= b->pipe_enabled;
+	a->sprites_enabled |= b->sprites_enabled;
+	a->sprites_scaled |= b->sprites_scaled;
+
+	for (level = 0; level <= max_level; level++) {
+		struct intel_wm_level *a_wm = &a->wm[level];
+		const struct intel_wm_level *b_wm = &b->wm[level];
+
+		a_wm->enable &= b_wm->enable;
+		a_wm->pri_val = max(a_wm->pri_val, b_wm->pri_val);
+		a_wm->spr_val = max(a_wm->spr_val, b_wm->spr_val);
+		a_wm->cur_val = max(a_wm->cur_val, b_wm->cur_val);
+		a_wm->fbc_val = max(a_wm->fbc_val, b_wm->fbc_val);
+	}
+
+	/*
+	 * We need to make sure that these merged watermark values are
+	 * actually a valid configuration themselves.  If they're not,
+	 * there's no safe way to transition from the old state to
+	 * the new state, so we need to fail the atomic transaction.
+	 */
+	if (!ilk_validate_pipe_wm(dev, a))
+		return -EINVAL;
+
+	/*
+	 * If our intermediate WM are identical to the final WM, then we can
+	 * omit the post-vblank programming; only update if it's different.
+	 */
+	if (memcmp(a, &newstate->wm.optimal.ilk, sizeof(*a)) == 0)
+		newstate->wm.need_postvbl_update = false;
+
+	return 0;
+}
+
+/*
  * Merge the watermarks from all active pipes for a specific level.
  */
 static void ilk_merge_wm_level(struct drm_device *dev,
@@ -2383,9 +2447,7 @@ static void ilk_merge_wm_level(struct drm_device *dev,
 	ret_wm->enable = true;
 
 	for_each_intel_crtc(dev, intel_crtc) {
-		const struct intel_crtc_state *cstate =
-			to_intel_crtc_state(intel_crtc->base.state);
-		const struct intel_pipe_wm *active = &cstate->wm.optimal.ilk;
+		const struct intel_pipe_wm *active = &intel_crtc->wm.active.ilk;
 		const struct intel_wm_level *wm = &active->wm[level];
 
 		if (!active->pipe_enabled)
@@ -2421,7 +2483,7 @@ static void ilk_wm_merge(struct drm_device *dev,
 	/* ILK/SNB/IVB: LP1+ watermarks only w/ single pipe */
 	if ((INTEL_INFO(dev)->gen <= 6 || IS_IVYBRIDGE(dev)) &&
 	    config->num_pipes_active > 1)
-		return;
+		last_enabled_level = 0;
 
 	/* ILK: FBC WM must be disabled always */
 	merged->fbc_wm_enabled = INTEL_INFO(dev)->gen >= 6;
@@ -2533,15 +2595,14 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 
 	/* LP0 register values */
 	for_each_intel_crtc(dev, intel_crtc) {
-		const struct intel_crtc_state *cstate =
-			to_intel_crtc_state(intel_crtc->base.state);
 		enum pipe pipe = intel_crtc->pipe;
-		const struct intel_wm_level *r = &cstate->wm.optimal.ilk.wm[0];
+		const struct intel_wm_level *r =
+			&intel_crtc->wm.active.ilk.wm[0];
 
 		if (WARN_ON(!r->enable))
 			continue;
 
-		results->wm_linetime[pipe] = cstate->wm.optimal.ilk.linetime;
+		results->wm_linetime[pipe] = intel_crtc->wm.active.ilk.linetime;
 
 		results->wm_pipe[pipe] =
 			(r->pri_val << WM0_PIPE_PLANE_SHIFT) |
@@ -2748,7 +2809,7 @@ static void ilk_write_wm_values(struct drm_i915_private *dev_priv,
 	dev_priv->wm.hw = *results;
 }
 
-static bool ilk_disable_lp_wm(struct drm_device *dev)
+bool ilk_disable_lp_wm(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -3657,11 +3718,9 @@ static void ilk_compute_wm_config(struct drm_device *dev,
 	}
 }
 
-static void ilk_program_watermarks(struct intel_crtc_state *cstate)
+static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
 {
-	struct drm_crtc *crtc = cstate->base.crtc;
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_device *dev = dev_priv->dev;
 	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
 	struct ilk_wm_maximums max;
 	struct intel_wm_config config = {};
@@ -3692,28 +3751,28 @@ static void ilk_program_watermarks(struct intel_crtc_state *cstate)
 	ilk_write_wm_values(dev_priv, &results);
 }
 
-static void ilk_update_wm(struct drm_crtc *crtc)
+static void ilk_initial_watermarks(struct intel_crtc_state *cstate)
 {
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
-
-	WARN_ON(cstate->base.active != intel_crtc->active);
+	struct drm_i915_private *dev_priv = to_i915(cstate->base.crtc->dev);
+	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
 
-	/*
-	 * IVB workaround: must disable low power watermarks for at least
-	 * one frame before enabling scaling.  LP watermarks can be re-enabled
-	 * when scaling is disabled.
-	 *
-	 * WaCxSRDisabledForSpriteScaling:ivb
-	 */
-	if (cstate->disable_lp_wm) {
-		ilk_disable_lp_wm(crtc->dev);
-		intel_wait_for_vblank(crtc->dev, intel_crtc->pipe);
-	}
+	mutex_lock(&dev_priv->wm.wm_mutex);
+	intel_crtc->wm.active.ilk = cstate->wm.intermediate;
+	ilk_program_watermarks(dev_priv);
+	mutex_unlock(&dev_priv->wm.wm_mutex);
+}
 
-	intel_crtc->wm.active.ilk = cstate->wm.optimal.ilk;
+static void ilk_optimize_watermarks(struct intel_crtc_state *cstate)
+{
+	struct drm_i915_private *dev_priv = to_i915(cstate->base.crtc->dev);
+	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
 
-	ilk_program_watermarks(cstate);
+	mutex_lock(&dev_priv->wm.wm_mutex);
+	if (cstate->wm.need_postvbl_update) {
+		intel_crtc->wm.active.ilk = cstate->wm.optimal.ilk;
+		ilk_program_watermarks(dev_priv);
+	}
+	mutex_unlock(&dev_priv->wm.wm_mutex);
 }
 
 static void skl_pipe_wm_active_state(uint32_t val,
@@ -4243,7 +4302,7 @@ static u32 intel_rps_limits(struct drm_i915_private *dev_priv, u8 val)
 	 * the hw runs at the minimal clock before selecting the desired
 	 * frequency, if the down threshold expires in that window we will not
 	 * receive a down interrupt. */
-	if (IS_GEN9(dev_priv->dev)) {
+	if (IS_GEN9(dev_priv)) {
 		limits = (dev_priv->rps.max_freq_softlimit) << 23;
 		if (val <= dev_priv->rps.min_freq_softlimit)
 			limits |= (dev_priv->rps.min_freq_softlimit) << 14;
@@ -4528,7 +4587,7 @@ void intel_set_rps(struct drm_device *dev, u8 val)
 		gen6_set_rps(dev, val);
 }
 
-static void gen9_disable_rps(struct drm_device *dev)
+static void gen9_disable_rc6(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -4536,12 +4595,20 @@ static void gen9_disable_rps(struct drm_device *dev)
 	I915_WRITE(GEN9_PG_ENABLE, 0);
 }
 
+static void gen9_disable_rps(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(GEN6_RP_CONTROL, 0);
+}
+
 static void gen6_disable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 	I915_WRITE(GEN6_RPNSWREQ, 1 << 31);
+	I915_WRITE(GEN6_RP_CONTROL, 0);
 }
 
 static void cherryview_disable_rps(struct drm_device *dev)
@@ -4585,7 +4652,8 @@ static void intel_print_rc6_info(struct drm_device *dev, u32 mode)
 
 static bool bxt_check_bios_rc6_setup(const struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	bool enable_rc6 = true;
 	unsigned long rc6_ctx_base;
 
@@ -4599,9 +4667,9 @@ static bool bxt_check_bios_rc6_setup(const struct drm_device *dev)
 	 * for this check.
 	 */
 	rc6_ctx_base = I915_READ(RC6_CTX_BASE) & RC6_CTX_BASE_MASK;
-	if (!((rc6_ctx_base >= dev_priv->gtt.stolen_reserved_base) &&
-	      (rc6_ctx_base + PAGE_SIZE <= dev_priv->gtt.stolen_reserved_base +
-					dev_priv->gtt.stolen_reserved_size))) {
+	if (!((rc6_ctx_base >= ggtt->stolen_reserved_base) &&
+	      (rc6_ctx_base + PAGE_SIZE <= ggtt->stolen_reserved_base +
+					ggtt->stolen_reserved_size))) {
 		DRM_DEBUG_KMS("RC6 Base address not as expected.\n");
 		enable_rc6 = false;
 	}
@@ -4744,6 +4812,16 @@ static void gen9_enable_rps(struct drm_device *dev)
 
 	/* WaGsvDisableTurbo: Workaround to disable turbo on BXT A* */
 	if (IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
+		/*
+		 * BIOS could leave the Hw Turbo enabled, so need to explicitly
+		 * clear out the Control register just to avoid inconsitency
+		 * with debugfs interface, which will show  Turbo as enabled
+		 * only and that is not expected by the User after adding the
+		 * WaGsvDisableTurbo. Apart from this there is no problem even
+		 * if the Turbo is left enabled in the Control register, as the
+		 * Up/Down interrupts would remain masked.
+		 */
+		gen9_disable_rps(dev);
 		intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 		return;
 	}
@@ -4762,7 +4840,7 @@ static void gen9_enable_rps(struct drm_device *dev)
 	 * Up/Down EI & threshold registers, as well as the RP_CONTROL,
 	 * RP_INTERRUPT_LIMITS & RPNSWREQ registers */
 	dev_priv->rps.power = HIGH_POWER; /* force a reset */
-	gen6_set_rps(dev_priv->dev, dev_priv->rps.min_freq_softlimit);
+	gen6_set_rps(dev_priv->dev, dev_priv->rps.idle_freq);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -4770,9 +4848,8 @@ static void gen9_enable_rps(struct drm_device *dev)
 static void gen9_enable_rc6(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	uint32_t rc6_mask = 0;
-	int unused;
 
 	/* 1a: Software RC state - RC0 */
 	I915_WRITE(GEN6_RC_STATE, 0);
@@ -4793,8 +4870,8 @@ static void gen9_enable_rc6(struct drm_device *dev)
 		I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16);
 	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
 	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
-	for_each_ring(ring, dev_priv, unused)
-		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
+	for_each_engine(engine, dev_priv)
+		I915_WRITE(RING_MAX_IDLE(engine->mmio_base), 10);
 
 	if (HAS_GUC_UCODE(dev))
 		I915_WRITE(GUC_MAX_IDLE_COUNT, 0xA);
@@ -4840,9 +4917,8 @@ static void gen9_enable_rc6(struct drm_device *dev)
 static void gen8_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	uint32_t rc6_mask = 0;
-	int unused;
 
 	/* 1a: Software RC state - RC0 */
 	I915_WRITE(GEN6_RC_STATE, 0);
@@ -4861,8 +4937,8 @@ static void gen8_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
 	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
 	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
-	for_each_ring(ring, dev_priv, unused)
-		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
+	for_each_engine(engine, dev_priv)
+		I915_WRITE(RING_MAX_IDLE(engine->mmio_base), 10);
 	I915_WRITE(GEN6_RC_SLEEP, 0);
 	if (IS_BROADWELL(dev))
 		I915_WRITE(GEN6_RC6_THRESHOLD, 625); /* 800us/1.28 for TO */
@@ -4922,11 +4998,11 @@ static void gen8_enable_rps(struct drm_device *dev)
 static void gen6_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	u32 rc6vids, pcu_mbox = 0, rc6_mask = 0;
 	u32 gtfifodbg;
 	int rc6_mode;
-	int i, ret;
+	int ret;
 
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
@@ -4939,7 +5015,8 @@ static void gen6_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_STATE, 0);
 
 	/* Clear the DBG now so we don't confuse earlier errors */
-	if ((gtfifodbg = I915_READ(GTFIFODBG))) {
+	gtfifodbg = I915_READ(GTFIFODBG);
+	if (gtfifodbg) {
 		DRM_ERROR("GT fifo had a previous error %x\n", gtfifodbg);
 		I915_WRITE(GTFIFODBG, gtfifodbg);
 	}
@@ -4958,8 +5035,8 @@ static void gen6_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000);
 	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25);
 
-	for_each_ring(ring, dev_priv, i)
-		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
+	for_each_engine(engine, dev_priv)
+		I915_WRITE(RING_MAX_IDLE(engine->mmio_base), 10);
 
 	I915_WRITE(GEN6_RC_SLEEP, 0);
 	I915_WRITE(GEN6_RC1e_THRESHOLD, 1000);
@@ -5244,9 +5321,9 @@ static void cherryview_check_pctx(struct drm_i915_private *dev_priv)
 
 static void cherryview_setup_pctx(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	unsigned long pctx_paddr, paddr;
-	struct i915_gtt *gtt = &dev_priv->gtt;
 	u32 pcbr;
 	int pctx_size = 32*1024;
 
@@ -5254,7 +5331,7 @@ static void cherryview_setup_pctx(struct drm_device *dev)
 	if ((pcbr >> VLV_PCBR_ADDR_SHIFT) == 0) {
 		DRM_DEBUG_DRIVER("BIOS didn't set up PCBR, fixing up\n");
 		paddr = (dev_priv->mm.stolen_base +
-			 (gtt->stolen_size - pctx_size));
+			 (ggtt->stolen_size - pctx_size));
 
 		pctx_paddr = (paddr & (~4095));
 		I915_WRITE(VLV_PCBR, pctx_paddr);
@@ -5322,6 +5399,17 @@ static void valleyview_cleanup_pctx(struct drm_device *dev)
 	dev_priv->vlv_pctx = NULL;
 }
 
+static void vlv_init_gpll_ref_freq(struct drm_i915_private *dev_priv)
+{
+	dev_priv->rps.gpll_ref_freq =
+		vlv_get_cck_clock(dev_priv, "GPLL ref",
+				  CCK_GPLL_CLOCK_CONTROL,
+				  dev_priv->czclk_freq);
+
+	DRM_DEBUG_DRIVER("GPLL reference freq: %d kHz\n",
+			 dev_priv->rps.gpll_ref_freq);
+}
+
 static void valleyview_init_gt_powersave(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -5329,6 +5417,8 @@ static void valleyview_init_gt_powersave(struct drm_device *dev)
 
 	valleyview_setup_pctx(dev);
 
+	vlv_init_gpll_ref_freq(dev_priv);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 
 	val = vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS);
@@ -5386,6 +5476,8 @@ static void cherryview_init_gt_powersave(struct drm_device *dev)
 
 	cherryview_setup_pctx(dev);
 
+	vlv_init_gpll_ref_freq(dev_priv);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 
 	mutex_lock(&dev_priv->sb_lock);
@@ -5450,13 +5542,13 @@ static void valleyview_cleanup_gt_powersave(struct drm_device *dev)
 static void cherryview_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	u32 gtfifodbg, val, rc6_mode = 0, pcbr;
-	int i;
 
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
-	gtfifodbg = I915_READ(GTFIFODBG);
+	gtfifodbg = I915_READ(GTFIFODBG) & ~(GT_FIFO_SBDEDICATE_FREE_ENTRY_CHV |
+					     GT_FIFO_FREE_ENTRIES_CHV);
 	if (gtfifodbg) {
 		DRM_DEBUG_DRIVER("GT fifo had a previous error %x\n",
 				 gtfifodbg);
@@ -5477,8 +5569,8 @@ static void cherryview_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
 	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
 
-	for_each_ring(ring, dev_priv, i)
-		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
+	for_each_engine(engine, dev_priv)
+		I915_WRITE(RING_MAX_IDLE(engine->mmio_base), 10);
 	I915_WRITE(GEN6_RC_SLEEP, 0);
 
 	/* TO threshold set to 500 us ( 0x186 * 1.28 us) */
@@ -5537,10 +5629,10 @@ static void cherryview_enable_rps(struct drm_device *dev)
 			 dev_priv->rps.cur_freq);
 
 	DRM_DEBUG_DRIVER("setting GPU freq to %d MHz (%u)\n",
-			 intel_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
-			 dev_priv->rps.efficient_freq);
+			 intel_gpu_freq(dev_priv, dev_priv->rps.idle_freq),
+			 dev_priv->rps.idle_freq);
 
-	valleyview_set_rps(dev_priv->dev, dev_priv->rps.efficient_freq);
+	valleyview_set_rps(dev_priv->dev, dev_priv->rps.idle_freq);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -5548,15 +5640,15 @@ static void cherryview_enable_rps(struct drm_device *dev)
 static void valleyview_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	u32 gtfifodbg, val, rc6_mode = 0;
-	int i;
 
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
 	valleyview_check_pctx(dev_priv);
 
-	if ((gtfifodbg = I915_READ(GTFIFODBG))) {
+	gtfifodbg = I915_READ(GTFIFODBG);
+	if (gtfifodbg) {
 		DRM_DEBUG_DRIVER("GT fifo had a previous error %x\n",
 				 gtfifodbg);
 		I915_WRITE(GTFIFODBG, gtfifodbg);
@@ -5588,8 +5680,8 @@ static void valleyview_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000);
 	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25);
 
-	for_each_ring(ring, dev_priv, i)
-		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
+	for_each_engine(engine, dev_priv)
+		I915_WRITE(RING_MAX_IDLE(engine->mmio_base), 10);
 
 	I915_WRITE(GEN6_RC6_THRESHOLD, 0x557);
 
@@ -5627,10 +5719,10 @@ static void valleyview_enable_rps(struct drm_device *dev)
 			 dev_priv->rps.cur_freq);
 
 	DRM_DEBUG_DRIVER("setting GPU freq to %d MHz (%u)\n",
-			 intel_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
-			 dev_priv->rps.efficient_freq);
+			 intel_gpu_freq(dev_priv, dev_priv->rps.idle_freq),
+			 dev_priv->rps.idle_freq);
 
-	valleyview_set_rps(dev_priv->dev, dev_priv->rps.efficient_freq);
+	valleyview_set_rps(dev_priv->dev, dev_priv->rps.idle_freq);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -5965,17 +6057,16 @@ EXPORT_SYMBOL_GPL(i915_gpu_lower);
 bool i915_gpu_busy(void)
 {
 	struct drm_i915_private *dev_priv;
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	bool ret = false;
-	int i;
 
 	spin_lock_irq(&mchdev_lock);
 	if (!i915_mch_dev)
 		goto out_unlock;
 	dev_priv = i915_mch_dev;
 
-	for_each_ring(ring, dev_priv, i)
-		ret |= !list_empty(&ring->request_list);
+	for_each_engine(engine, dev_priv)
+		ret |= !list_empty(&engine->request_list);
 
 out_unlock:
 	spin_unlock_irq(&mchdev_lock);
@@ -6195,9 +6286,10 @@ void intel_disable_gt_powersave(struct drm_device *dev)
 		intel_suspend_gt_powersave(dev);
 
 		mutex_lock(&dev_priv->rps.hw_lock);
-		if (INTEL_INFO(dev)->gen >= 9)
+		if (INTEL_INFO(dev)->gen >= 9) {
+			gen9_disable_rc6(dev);
 			gen9_disable_rps(dev);
-		else if (IS_CHERRYVIEW(dev))
+		} else if (IS_CHERRYVIEW(dev))
 			cherryview_disable_rps(dev);
 		else if (IS_VALLEYVIEW(dev))
 			valleyview_disable_rps(dev);
@@ -6818,23 +6910,10 @@ static void ivybridge_init_clock_gating(struct drm_device *dev)
 	gen6_check_mch_setup(dev);
 }
 
-static void vlv_init_display_clock_gating(struct drm_i915_private *dev_priv)
-{
-	I915_WRITE(DSPCLK_GATE_D, VRHUNIT_CLOCK_GATE_DISABLE);
-
-	/*
-	 * Disable trickle feed and enable pnd deadline calculation
-	 */
-	I915_WRITE(MI_ARB_VLV, MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE);
-	I915_WRITE(CBR1_VLV, 0);
-}
-
 static void valleyview_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	vlv_init_display_clock_gating(dev_priv);
-
 	/* WaDisableEarlyCull:vlv */
 	I915_WRITE(_3D_CHICKEN3,
 		   _MASKED_BIT_ENABLE(_3D_CHICKEN_SF_DISABLE_OBJEND_CULL));
@@ -6917,8 +6996,6 @@ static void cherryview_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	vlv_init_display_clock_gating(dev_priv);
-
 	/* WaVSRefCountFullforceMissDisable:chv */
 	/* WaDSRefCountFullforceMissDisable:chv */
 	I915_WRITE(GEN7_FF_THREAD_MODE,
@@ -7058,8 +7135,7 @@ void intel_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (dev_priv->display.init_clock_gating)
-		dev_priv->display.init_clock_gating(dev);
+	dev_priv->display.init_clock_gating(dev);
 }
 
 void intel_suspend_hw(struct drm_device *dev)
@@ -7068,6 +7144,60 @@ void intel_suspend_hw(struct drm_device *dev)
 		lpt_suspend_hw(dev);
 }
 
+static void nop_init_clock_gating(struct drm_device *dev)
+{
+	DRM_DEBUG_KMS("No clock gating settings or workarounds applied.\n");
+}
+
+/**
+ * intel_init_clock_gating_hooks - setup the clock gating hooks
+ * @dev_priv: device private
+ *
+ * Setup the hooks that configure which clocks of a given platform can be
+ * gated and also apply various GT and display specific workarounds for these
+ * platforms. Note that some GT specific workarounds are applied separately
+ * when GPU contexts or batchbuffers start their execution.
+ */
+void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv)
+{
+	if (IS_SKYLAKE(dev_priv))
+		dev_priv->display.init_clock_gating = nop_init_clock_gating;
+	else if (IS_KABYLAKE(dev_priv))
+		dev_priv->display.init_clock_gating = nop_init_clock_gating;
+	else if (IS_BROXTON(dev_priv))
+		dev_priv->display.init_clock_gating = bxt_init_clock_gating;
+	else if (IS_BROADWELL(dev_priv))
+		dev_priv->display.init_clock_gating = broadwell_init_clock_gating;
+	else if (IS_CHERRYVIEW(dev_priv))
+		dev_priv->display.init_clock_gating = cherryview_init_clock_gating;
+	else if (IS_HASWELL(dev_priv))
+		dev_priv->display.init_clock_gating = haswell_init_clock_gating;
+	else if (IS_IVYBRIDGE(dev_priv))
+		dev_priv->display.init_clock_gating = ivybridge_init_clock_gating;
+	else if (IS_VALLEYVIEW(dev_priv))
+		dev_priv->display.init_clock_gating = valleyview_init_clock_gating;
+	else if (IS_GEN6(dev_priv))
+		dev_priv->display.init_clock_gating = gen6_init_clock_gating;
+	else if (IS_GEN5(dev_priv))
+		dev_priv->display.init_clock_gating = ironlake_init_clock_gating;
+	else if (IS_G4X(dev_priv))
+		dev_priv->display.init_clock_gating = g4x_init_clock_gating;
+	else if (IS_CRESTLINE(dev_priv))
+		dev_priv->display.init_clock_gating = crestline_init_clock_gating;
+	else if (IS_BROADWATER(dev_priv))
+		dev_priv->display.init_clock_gating = broadwater_init_clock_gating;
+	else if (IS_GEN3(dev_priv))
+		dev_priv->display.init_clock_gating = gen3_init_clock_gating;
+	else if (IS_I85X(dev_priv) || IS_I865G(dev_priv))
+		dev_priv->display.init_clock_gating = i85x_init_clock_gating;
+	else if (IS_GEN2(dev_priv))
+		dev_priv->display.init_clock_gating = i830_init_clock_gating;
+	else {
+		MISSING_CASE(INTEL_DEVID(dev_priv));
+		dev_priv->display.init_clock_gating = nop_init_clock_gating;
+	}
+}
+
 /* Set up chip specific power management-related functions */
 void intel_init_pm(struct drm_device *dev)
 {
@@ -7084,10 +7214,6 @@ void intel_init_pm(struct drm_device *dev)
 	/* For FIFO watermark updates */
 	if (INTEL_INFO(dev)->gen >= 9) {
 		skl_setup_wm_latency(dev);
-
-		if (IS_BROXTON(dev))
-			dev_priv->display.init_clock_gating =
-				bxt_init_clock_gating;
 		dev_priv->display.update_wm = skl_update_wm;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		ilk_setup_wm_latency(dev);
@@ -7096,36 +7222,23 @@ void intel_init_pm(struct drm_device *dev)
 		     dev_priv->wm.spr_latency[1] && dev_priv->wm.cur_latency[1]) ||
 		    (!IS_GEN5(dev) && dev_priv->wm.pri_latency[0] &&
 		     dev_priv->wm.spr_latency[0] && dev_priv->wm.cur_latency[0])) {
-			dev_priv->display.update_wm = ilk_update_wm;
 			dev_priv->display.compute_pipe_wm = ilk_compute_pipe_wm;
-			dev_priv->display.program_watermarks = ilk_program_watermarks;
+			dev_priv->display.compute_intermediate_wm =
+				ilk_compute_intermediate_wm;
+			dev_priv->display.initial_watermarks =
+				ilk_initial_watermarks;
+			dev_priv->display.optimize_watermarks =
+				ilk_optimize_watermarks;
 		} else {
 			DRM_DEBUG_KMS("Failed to read display plane latency. "
 				      "Disable CxSR\n");
 		}
-
-		if (IS_GEN5(dev))
-			dev_priv->display.init_clock_gating = ironlake_init_clock_gating;
-		else if (IS_GEN6(dev))
-			dev_priv->display.init_clock_gating = gen6_init_clock_gating;
-		else if (IS_IVYBRIDGE(dev))
-			dev_priv->display.init_clock_gating = ivybridge_init_clock_gating;
-		else if (IS_HASWELL(dev))
-			dev_priv->display.init_clock_gating = haswell_init_clock_gating;
-		else if (INTEL_INFO(dev)->gen == 8)
-			dev_priv->display.init_clock_gating = broadwell_init_clock_gating;
 	} else if (IS_CHERRYVIEW(dev)) {
 		vlv_setup_wm_latency(dev);
-
 		dev_priv->display.update_wm = vlv_update_wm;
-		dev_priv->display.init_clock_gating =
-			cherryview_init_clock_gating;
 	} else if (IS_VALLEYVIEW(dev)) {
 		vlv_setup_wm_latency(dev);
-
 		dev_priv->display.update_wm = vlv_update_wm;
-		dev_priv->display.init_clock_gating =
-			valleyview_init_clock_gating;
 	} else if (IS_PINEVIEW(dev)) {
 		if (!intel_get_cxsr_latency(IS_PINEVIEW_G(dev),
 					    dev_priv->is_ddr3,
@@ -7141,20 +7254,13 @@ void intel_init_pm(struct drm_device *dev)
 			dev_priv->display.update_wm = NULL;
 		} else
 			dev_priv->display.update_wm = pineview_update_wm;
-		dev_priv->display.init_clock_gating = gen3_init_clock_gating;
 	} else if (IS_G4X(dev)) {
 		dev_priv->display.update_wm = g4x_update_wm;
-		dev_priv->display.init_clock_gating = g4x_init_clock_gating;
 	} else if (IS_GEN4(dev)) {
 		dev_priv->display.update_wm = i965_update_wm;
-		if (IS_CRESTLINE(dev))
-			dev_priv->display.init_clock_gating = crestline_init_clock_gating;
-		else if (IS_BROADWATER(dev))
-			dev_priv->display.init_clock_gating = broadwater_init_clock_gating;
 	} else if (IS_GEN3(dev)) {
 		dev_priv->display.update_wm = i9xx_update_wm;
 		dev_priv->display.get_fifo_size = i9xx_get_fifo_size;
-		dev_priv->display.init_clock_gating = gen3_init_clock_gating;
 	} else if (IS_GEN2(dev)) {
 		if (INTEL_INFO(dev)->num_pipes == 1) {
 			dev_priv->display.update_wm = i845_update_wm;
@@ -7163,11 +7269,6 @@ void intel_init_pm(struct drm_device *dev)
 			dev_priv->display.update_wm = i9xx_update_wm;
 			dev_priv->display.get_fifo_size = i830_get_fifo_size;
 		}
-
-		if (IS_I85X(dev) || IS_I865G(dev))
-			dev_priv->display.init_clock_gating = i85x_init_clock_gating;
-		else
-			dev_priv->display.init_clock_gating = i830_init_clock_gating;
 	} else {
 		DRM_ERROR("unexpected fall-through in intel_init_pm\n");
 	}
@@ -7221,78 +7322,43 @@ int sandybridge_pcode_write(struct drm_i915_private *dev_priv, u32 mbox, u32 val
 	return 0;
 }
 
-static int vlv_gpu_freq_div(unsigned int czclk_freq)
-{
-	switch (czclk_freq) {
-	case 200:
-		return 10;
-	case 267:
-		return 12;
-	case 320:
-	case 333:
-		return 16;
-	case 400:
-		return 20;
-	default:
-		return -1;
-	}
-}
-
 static int byt_gpu_freq(struct drm_i915_private *dev_priv, int val)
 {
-	int div, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
-
-	div = vlv_gpu_freq_div(czclk_freq);
-	if (div < 0)
-		return div;
-
-	return DIV_ROUND_CLOSEST(czclk_freq * (val + 6 - 0xbd), div);
+	/*
+	 * N = val - 0xb7
+	 * Slow = Fast = GPLL ref * N
+	 */
+	return DIV_ROUND_CLOSEST(dev_priv->rps.gpll_ref_freq * (val - 0xb7), 1000);
 }
 
 static int byt_freq_opcode(struct drm_i915_private *dev_priv, int val)
 {
-	int mul, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
-
-	mul = vlv_gpu_freq_div(czclk_freq);
-	if (mul < 0)
-		return mul;
-
-	return DIV_ROUND_CLOSEST(mul * val, czclk_freq) + 0xbd - 6;
+	return DIV_ROUND_CLOSEST(1000 * val, dev_priv->rps.gpll_ref_freq) + 0xb7;
 }
 
 static int chv_gpu_freq(struct drm_i915_private *dev_priv, int val)
 {
-	int div, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
-
-	div = vlv_gpu_freq_div(czclk_freq);
-	if (div < 0)
-		return div;
-	div /= 2;
-
-	return DIV_ROUND_CLOSEST(czclk_freq * val, 2 * div) / 2;
+	/*
+	 * N = val / 2
+	 * CU (slow) = CU2x (fast) / 2 = GPLL ref * N / 2
+	 */
+	return DIV_ROUND_CLOSEST(dev_priv->rps.gpll_ref_freq * val, 2 * 2 * 1000);
 }
 
 static int chv_freq_opcode(struct drm_i915_private *dev_priv, int val)
 {
-	int mul, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
-
-	mul = vlv_gpu_freq_div(czclk_freq);
-	if (mul < 0)
-		return mul;
-	mul /= 2;
-
 	/* CHV needs even values */
-	return DIV_ROUND_CLOSEST(val * 2 * mul, czclk_freq) * 2;
+	return DIV_ROUND_CLOSEST(2 * 1000 * val, dev_priv->rps.gpll_ref_freq) * 2;
 }
 
 int intel_gpu_freq(struct drm_i915_private *dev_priv, int val)
 {
-	if (IS_GEN9(dev_priv->dev))
+	if (IS_GEN9(dev_priv))
 		return DIV_ROUND_CLOSEST(val * GT_FREQUENCY_MULTIPLIER,
 					 GEN9_FREQ_SCALER);
-	else if (IS_CHERRYVIEW(dev_priv->dev))
+	else if (IS_CHERRYVIEW(dev_priv))
 		return chv_gpu_freq(dev_priv, val);
-	else if (IS_VALLEYVIEW(dev_priv->dev))
+	else if (IS_VALLEYVIEW(dev_priv))
 		return byt_gpu_freq(dev_priv, val);
 	else
 		return val * GT_FREQUENCY_MULTIPLIER;
@@ -7300,12 +7366,12 @@ int intel_gpu_freq(struct drm_i915_private *dev_priv, int val)
 
 int intel_freq_opcode(struct drm_i915_private *dev_priv, int val)
 {
-	if (IS_GEN9(dev_priv->dev))
+	if (IS_GEN9(dev_priv))
 		return DIV_ROUND_CLOSEST(val * GEN9_FREQ_SCALER,
 					 GT_FREQUENCY_MULTIPLIER);
-	else if (IS_CHERRYVIEW(dev_priv->dev))
+	else if (IS_CHERRYVIEW(dev_priv))
 		return chv_freq_opcode(dev_priv, val);
-	else if (IS_VALLEYVIEW(dev_priv->dev))
+	else if (IS_VALLEYVIEW(dev_priv))
 		return byt_freq_opcode(dev_priv, val);
 	else
 		return DIV_ROUND_CLOSEST(val, GT_FREQUENCY_MULTIPLIER);
@@ -7322,7 +7388,7 @@ static void __intel_rps_boost_work(struct work_struct *work)
 	struct drm_i915_gem_request *req = boost->req;
 
 	if (!i915_gem_request_completed(req, true))
-		gen6_rps_boost(to_i915(req->ring->dev), NULL,
+		gen6_rps_boost(to_i915(req->engine->dev), NULL,
 			       req->emitted_jiffies);
 
 	i915_gem_request_unreference__unlocked(req);
diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c
index 0b42ada338c8..c3abae4bc596 100644
--- a/drivers/gpu/drm/i915/intel_psr.c
+++ b/drivers/gpu/drm/i915/intel_psr.c
@@ -507,7 +507,8 @@ static void hsw_psr_disable(struct intel_dp *intel_dp)
 
 		/* Wait till PSR is idle */
 		if (_wait_for((I915_READ(EDP_PSR_STATUS_CTL) &
-			       EDP_PSR_STATUS_STATE_MASK) == 0, 2000, 10))
+			       EDP_PSR_STATUS_STATE_MASK) == 0,
+			       2 * USEC_PER_SEC, 10 * USEC_PER_MSEC))
 			DRM_ERROR("Timed out waiting for PSR Idle State\n");
 
 		dev_priv->psr.active = false;
@@ -562,7 +563,7 @@ static void intel_psr_work(struct work_struct *work)
 	 * PSR might take some time to get fully disabled
 	 * and be ready for re-enable.
 	 */
-	if (HAS_DDI(dev_priv->dev)) {
+	if (HAS_DDI(dev_priv)) {
 		if (wait_for((I915_READ(EDP_PSR_STATUS_CTL) &
 			      EDP_PSR_STATUS_STATE_MASK) == 0, 50)) {
 			DRM_ERROR("Timed out waiting for PSR Idle for re-enable\n");
@@ -780,8 +781,7 @@ void intel_psr_init(struct drm_device *dev)
 
 	/* Per platform default */
 	if (i915.enable_psr == -1) {
-		if (IS_HASWELL(dev) || IS_BROADWELL(dev) ||
-		    IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
+		if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 			i915.enable_psr = 1;
 		else
 			i915.enable_psr = 0;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 9121646d7c4d..245386e20c52 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -59,19 +59,19 @@ int intel_ring_space(struct intel_ringbuffer *ringbuf)
 	return ringbuf->space;
 }
 
-bool intel_ring_stopped(struct intel_engine_cs *ring)
+bool intel_engine_stopped(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	return dev_priv->gpu_error.stop_rings & intel_ring_flag(ring);
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
+	return dev_priv->gpu_error.stop_rings & intel_engine_flag(engine);
 }
 
-static void __intel_ring_advance(struct intel_engine_cs *ring)
+static void __intel_ring_advance(struct intel_engine_cs *engine)
 {
-	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct intel_ringbuffer *ringbuf = engine->buffer;
 	ringbuf->tail &= ringbuf->size - 1;
-	if (intel_ring_stopped(ring))
+	if (intel_engine_stopped(engine))
 		return;
-	ring->write_tail(ring, ringbuf->tail);
+	engine->write_tail(engine, ringbuf->tail);
 }
 
 static int
@@ -79,7 +79,7 @@ gen2_render_ring_flush(struct drm_i915_gem_request *req,
 		       u32	invalidate_domains,
 		       u32	flush_domains)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	u32 cmd;
 	int ret;
 
@@ -94,9 +94,9 @@ gen2_render_ring_flush(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, cmd);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, cmd);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -106,8 +106,8 @@ gen4_render_ring_flush(struct drm_i915_gem_request *req,
 		       u32	invalidate_domains,
 		       u32	flush_domains)
 {
-	struct intel_engine_cs *ring = req->ring;
-	struct drm_device *dev = ring->dev;
+	struct intel_engine_cs *engine = req->engine;
+	struct drm_device *dev = engine->dev;
 	u32 cmd;
 	int ret;
 
@@ -153,9 +153,9 @@ gen4_render_ring_flush(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, cmd);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, cmd);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -200,34 +200,34 @@ gen4_render_ring_flush(struct drm_i915_gem_request *req,
 static int
 intel_emit_post_sync_nonzero_flush(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
-	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	struct intel_engine_cs *engine = req->engine;
+	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	ret = intel_ring_begin(req, 6);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(5));
-	intel_ring_emit(ring, PIPE_CONTROL_CS_STALL |
+	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(5));
+	intel_ring_emit(engine, PIPE_CONTROL_CS_STALL |
 			PIPE_CONTROL_STALL_AT_SCOREBOARD);
-	intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT); /* address */
-	intel_ring_emit(ring, 0); /* low dword */
-	intel_ring_emit(ring, 0); /* high dword */
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, scratch_addr | PIPE_CONTROL_GLOBAL_GTT); /* address */
+	intel_ring_emit(engine, 0); /* low dword */
+	intel_ring_emit(engine, 0); /* high dword */
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	ret = intel_ring_begin(req, 6);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(5));
-	intel_ring_emit(ring, PIPE_CONTROL_QW_WRITE);
-	intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT); /* address */
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(5));
+	intel_ring_emit(engine, PIPE_CONTROL_QW_WRITE);
+	intel_ring_emit(engine, scratch_addr | PIPE_CONTROL_GLOBAL_GTT); /* address */
+	intel_ring_emit(engine, 0);
+	intel_ring_emit(engine, 0);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -236,9 +236,9 @@ static int
 gen6_render_ring_flush(struct drm_i915_gem_request *req,
 		       u32 invalidate_domains, u32 flush_domains)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/* Force SNB workarounds for PIPE_CONTROL flushes */
@@ -276,11 +276,11 @@ gen6_render_ring_flush(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
-	intel_ring_emit(ring, flags);
-	intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
-	intel_ring_emit(ring, 0);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(4));
+	intel_ring_emit(engine, flags);
+	intel_ring_emit(engine, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
+	intel_ring_emit(engine, 0);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -288,19 +288,19 @@ gen6_render_ring_flush(struct drm_i915_gem_request *req,
 static int
 gen7_render_ring_cs_stall_wa(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 4);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
-	intel_ring_emit(ring, PIPE_CONTROL_CS_STALL |
+	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(4));
+	intel_ring_emit(engine, PIPE_CONTROL_CS_STALL |
 			      PIPE_CONTROL_STALL_AT_SCOREBOARD);
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, 0);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, 0);
+	intel_ring_emit(engine, 0);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -309,9 +309,9 @@ static int
 gen7_render_ring_flush(struct drm_i915_gem_request *req,
 		       u32 invalidate_domains, u32 flush_domains)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/*
@@ -360,11 +360,11 @@ gen7_render_ring_flush(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
-	intel_ring_emit(ring, flags);
-	intel_ring_emit(ring, scratch_addr);
-	intel_ring_emit(ring, 0);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(4));
+	intel_ring_emit(engine, flags);
+	intel_ring_emit(engine, scratch_addr);
+	intel_ring_emit(engine, 0);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -373,20 +373,20 @@ static int
 gen8_emit_pipe_control(struct drm_i915_gem_request *req,
 		       u32 flags, u32 scratch_addr)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 6);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
-	intel_ring_emit(ring, flags);
-	intel_ring_emit(ring, scratch_addr);
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, 0);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(6));
+	intel_ring_emit(engine, flags);
+	intel_ring_emit(engine, scratch_addr);
+	intel_ring_emit(engine, 0);
+	intel_ring_emit(engine, 0);
+	intel_ring_emit(engine, 0);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -396,7 +396,7 @@ gen8_render_ring_flush(struct drm_i915_gem_request *req,
 		       u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = req->ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	u32 scratch_addr = req->engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	flags |= PIPE_CONTROL_CS_STALL;
@@ -429,51 +429,51 @@ gen8_render_ring_flush(struct drm_i915_gem_request *req,
 	return gen8_emit_pipe_control(req, flags, scratch_addr);
 }
 
-static void ring_write_tail(struct intel_engine_cs *ring,
+static void ring_write_tail(struct intel_engine_cs *engine,
 			    u32 value)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	I915_WRITE_TAIL(ring, value);
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
+	I915_WRITE_TAIL(engine, value);
 }
 
-u64 intel_ring_get_active_head(struct intel_engine_cs *ring)
+u64 intel_ring_get_active_head(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	u64 acthd;
 
-	if (INTEL_INFO(ring->dev)->gen >= 8)
-		acthd = I915_READ64_2x32(RING_ACTHD(ring->mmio_base),
-					 RING_ACTHD_UDW(ring->mmio_base));
-	else if (INTEL_INFO(ring->dev)->gen >= 4)
-		acthd = I915_READ(RING_ACTHD(ring->mmio_base));
+	if (INTEL_INFO(engine->dev)->gen >= 8)
+		acthd = I915_READ64_2x32(RING_ACTHD(engine->mmio_base),
+					 RING_ACTHD_UDW(engine->mmio_base));
+	else if (INTEL_INFO(engine->dev)->gen >= 4)
+		acthd = I915_READ(RING_ACTHD(engine->mmio_base));
 	else
 		acthd = I915_READ(ACTHD);
 
 	return acthd;
 }
 
-static void ring_setup_phys_status_page(struct intel_engine_cs *ring)
+static void ring_setup_phys_status_page(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	u32 addr;
 
 	addr = dev_priv->status_page_dmah->busaddr;
-	if (INTEL_INFO(ring->dev)->gen >= 4)
+	if (INTEL_INFO(engine->dev)->gen >= 4)
 		addr |= (dev_priv->status_page_dmah->busaddr >> 28) & 0xf0;
 	I915_WRITE(HWS_PGA, addr);
 }
 
-static void intel_ring_setup_status_page(struct intel_engine_cs *ring)
+static void intel_ring_setup_status_page(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_device *dev = engine->dev;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	i915_reg_t mmio;
 
 	/* The ring status page addresses are no longer next to the rest of
 	 * the ring registers as of gen7.
 	 */
 	if (IS_GEN7(dev)) {
-		switch (ring->id) {
+		switch (engine->id) {
 		case RCS:
 			mmio = RENDER_HWS_PGA_GEN7;
 			break;
@@ -492,14 +492,14 @@ static void intel_ring_setup_status_page(struct intel_engine_cs *ring)
 			mmio = VEBOX_HWS_PGA_GEN7;
 			break;
 		}
-	} else if (IS_GEN6(ring->dev)) {
-		mmio = RING_HWS_PGA_GEN6(ring->mmio_base);
+	} else if (IS_GEN6(engine->dev)) {
+		mmio = RING_HWS_PGA_GEN6(engine->mmio_base);
 	} else {
 		/* XXX: gen8 returns to sanity */
-		mmio = RING_HWS_PGA(ring->mmio_base);
+		mmio = RING_HWS_PGA(engine->mmio_base);
 	}
 
-	I915_WRITE(mmio, (u32)ring->status_page.gfx_addr);
+	I915_WRITE(mmio, (u32)engine->status_page.gfx_addr);
 	POSTING_READ(mmio);
 
 	/*
@@ -510,10 +510,10 @@ static void intel_ring_setup_status_page(struct intel_engine_cs *ring)
 	 * invalidating the TLB?
 	 */
 	if (INTEL_INFO(dev)->gen >= 6 && INTEL_INFO(dev)->gen < 8) {
-		i915_reg_t reg = RING_INSTPM(ring->mmio_base);
+		i915_reg_t reg = RING_INSTPM(engine->mmio_base);
 
 		/* ring should be idle before issuing a sync flush*/
-		WARN_ON((I915_READ_MODE(ring) & MODE_IDLE) == 0);
+		WARN_ON((I915_READ_MODE(engine) & MODE_IDLE) == 0);
 
 		I915_WRITE(reg,
 			   _MASKED_BIT_ENABLE(INSTPM_TLB_INVALIDATE |
@@ -521,117 +521,125 @@ static void intel_ring_setup_status_page(struct intel_engine_cs *ring)
 		if (wait_for((I915_READ(reg) & INSTPM_SYNC_FLUSH) == 0,
 			     1000))
 			DRM_ERROR("%s: wait for SyncFlush to complete for TLB invalidation timed out\n",
-				  ring->name);
+				  engine->name);
 	}
 }
 
-static bool stop_ring(struct intel_engine_cs *ring)
+static bool stop_ring(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(ring->dev);
+	struct drm_i915_private *dev_priv = to_i915(engine->dev);
 
-	if (!IS_GEN2(ring->dev)) {
-		I915_WRITE_MODE(ring, _MASKED_BIT_ENABLE(STOP_RING));
-		if (wait_for((I915_READ_MODE(ring) & MODE_IDLE) != 0, 1000)) {
-			DRM_ERROR("%s : timed out trying to stop ring\n", ring->name);
+	if (!IS_GEN2(engine->dev)) {
+		I915_WRITE_MODE(engine, _MASKED_BIT_ENABLE(STOP_RING));
+		if (wait_for((I915_READ_MODE(engine) & MODE_IDLE) != 0, 1000)) {
+			DRM_ERROR("%s : timed out trying to stop ring\n",
+				  engine->name);
 			/* Sometimes we observe that the idle flag is not
 			 * set even though the ring is empty. So double
 			 * check before giving up.
 			 */
-			if (I915_READ_HEAD(ring) != I915_READ_TAIL(ring))
+			if (I915_READ_HEAD(engine) != I915_READ_TAIL(engine))
 				return false;
 		}
 	}
 
-	I915_WRITE_CTL(ring, 0);
-	I915_WRITE_HEAD(ring, 0);
-	ring->write_tail(ring, 0);
+	I915_WRITE_CTL(engine, 0);
+	I915_WRITE_HEAD(engine, 0);
+	engine->write_tail(engine, 0);
 
-	if (!IS_GEN2(ring->dev)) {
-		(void)I915_READ_CTL(ring);
-		I915_WRITE_MODE(ring, _MASKED_BIT_DISABLE(STOP_RING));
+	if (!IS_GEN2(engine->dev)) {
+		(void)I915_READ_CTL(engine);
+		I915_WRITE_MODE(engine, _MASKED_BIT_DISABLE(STOP_RING));
 	}
 
-	return (I915_READ_HEAD(ring) & HEAD_ADDR) == 0;
+	return (I915_READ_HEAD(engine) & HEAD_ADDR) == 0;
 }
 
-static int init_ring_common(struct intel_engine_cs *ring)
+void intel_engine_init_hangcheck(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	memset(&engine->hangcheck, 0, sizeof(engine->hangcheck));
+}
+
+static int init_ring_common(struct intel_engine_cs *engine)
+{
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct intel_ringbuffer *ringbuf = engine->buffer;
 	struct drm_i915_gem_object *obj = ringbuf->obj;
 	int ret = 0;
 
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
-	if (!stop_ring(ring)) {
+	if (!stop_ring(engine)) {
 		/* G45 ring initialization often fails to reset head to zero */
 		DRM_DEBUG_KMS("%s head not reset to zero "
 			      "ctl %08x head %08x tail %08x start %08x\n",
-			      ring->name,
-			      I915_READ_CTL(ring),
-			      I915_READ_HEAD(ring),
-			      I915_READ_TAIL(ring),
-			      I915_READ_START(ring));
+			      engine->name,
+			      I915_READ_CTL(engine),
+			      I915_READ_HEAD(engine),
+			      I915_READ_TAIL(engine),
+			      I915_READ_START(engine));
 
-		if (!stop_ring(ring)) {
+		if (!stop_ring(engine)) {
 			DRM_ERROR("failed to set %s head to zero "
 				  "ctl %08x head %08x tail %08x start %08x\n",
-				  ring->name,
-				  I915_READ_CTL(ring),
-				  I915_READ_HEAD(ring),
-				  I915_READ_TAIL(ring),
-				  I915_READ_START(ring));
+				  engine->name,
+				  I915_READ_CTL(engine),
+				  I915_READ_HEAD(engine),
+				  I915_READ_TAIL(engine),
+				  I915_READ_START(engine));
 			ret = -EIO;
 			goto out;
 		}
 	}
 
 	if (I915_NEED_GFX_HWS(dev))
-		intel_ring_setup_status_page(ring);
+		intel_ring_setup_status_page(engine);
 	else
-		ring_setup_phys_status_page(ring);
+		ring_setup_phys_status_page(engine);
 
 	/* Enforce ordering by reading HEAD register back */
-	I915_READ_HEAD(ring);
+	I915_READ_HEAD(engine);
 
 	/* Initialize the ring. This must happen _after_ we've cleared the ring
 	 * registers with the above sequence (the readback of the HEAD registers
 	 * also enforces ordering), otherwise the hw might lose the new ring
 	 * register values. */
-	I915_WRITE_START(ring, i915_gem_obj_ggtt_offset(obj));
+	I915_WRITE_START(engine, i915_gem_obj_ggtt_offset(obj));
 
 	/* WaClearRingBufHeadRegAtInit:ctg,elk */
-	if (I915_READ_HEAD(ring))
+	if (I915_READ_HEAD(engine))
 		DRM_DEBUG("%s initialization failed [head=%08x], fudging\n",
-			  ring->name, I915_READ_HEAD(ring));
-	I915_WRITE_HEAD(ring, 0);
-	(void)I915_READ_HEAD(ring);
+			  engine->name, I915_READ_HEAD(engine));
+	I915_WRITE_HEAD(engine, 0);
+	(void)I915_READ_HEAD(engine);
 
-	I915_WRITE_CTL(ring,
+	I915_WRITE_CTL(engine,
 			((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES)
 			| RING_VALID);
 
 	/* If the head is still not zero, the ring is dead */
-	if (wait_for((I915_READ_CTL(ring) & RING_VALID) != 0 &&
-		     I915_READ_START(ring) == i915_gem_obj_ggtt_offset(obj) &&
-		     (I915_READ_HEAD(ring) & HEAD_ADDR) == 0, 50)) {
+	if (wait_for((I915_READ_CTL(engine) & RING_VALID) != 0 &&
+		     I915_READ_START(engine) == i915_gem_obj_ggtt_offset(obj) &&
+		     (I915_READ_HEAD(engine) & HEAD_ADDR) == 0, 50)) {
 		DRM_ERROR("%s initialization failed "
 			  "ctl %08x (valid? %d) head %08x tail %08x start %08x [expected %08lx]\n",
-			  ring->name,
-			  I915_READ_CTL(ring), I915_READ_CTL(ring) & RING_VALID,
-			  I915_READ_HEAD(ring), I915_READ_TAIL(ring),
-			  I915_READ_START(ring), (unsigned long)i915_gem_obj_ggtt_offset(obj));
+			  engine->name,
+			  I915_READ_CTL(engine),
+			  I915_READ_CTL(engine) & RING_VALID,
+			  I915_READ_HEAD(engine), I915_READ_TAIL(engine),
+			  I915_READ_START(engine),
+			  (unsigned long)i915_gem_obj_ggtt_offset(obj));
 		ret = -EIO;
 		goto out;
 	}
 
 	ringbuf->last_retired_head = -1;
-	ringbuf->head = I915_READ_HEAD(ring);
-	ringbuf->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
+	ringbuf->head = I915_READ_HEAD(engine);
+	ringbuf->tail = I915_READ_TAIL(engine) & TAIL_ADDR;
 	intel_ring_update_space(ringbuf);
 
-	memset(&ring->hangcheck, 0, sizeof(ring->hangcheck));
+	intel_engine_init_hangcheck(engine);
 
 out:
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
@@ -640,59 +648,60 @@ out:
 }
 
 void
-intel_fini_pipe_control(struct intel_engine_cs *ring)
+intel_fini_pipe_control(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 
-	if (ring->scratch.obj == NULL)
+	if (engine->scratch.obj == NULL)
 		return;
 
 	if (INTEL_INFO(dev)->gen >= 5) {
-		kunmap(sg_page(ring->scratch.obj->pages->sgl));
-		i915_gem_object_ggtt_unpin(ring->scratch.obj);
+		kunmap(sg_page(engine->scratch.obj->pages->sgl));
+		i915_gem_object_ggtt_unpin(engine->scratch.obj);
 	}
 
-	drm_gem_object_unreference(&ring->scratch.obj->base);
-	ring->scratch.obj = NULL;
+	drm_gem_object_unreference(&engine->scratch.obj->base);
+	engine->scratch.obj = NULL;
 }
 
 int
-intel_init_pipe_control(struct intel_engine_cs *ring)
+intel_init_pipe_control(struct intel_engine_cs *engine)
 {
 	int ret;
 
-	WARN_ON(ring->scratch.obj);
+	WARN_ON(engine->scratch.obj);
 
-	ring->scratch.obj = i915_gem_alloc_object(ring->dev, 4096);
-	if (ring->scratch.obj == NULL) {
+	engine->scratch.obj = i915_gem_alloc_object(engine->dev, 4096);
+	if (engine->scratch.obj == NULL) {
 		DRM_ERROR("Failed to allocate seqno page\n");
 		ret = -ENOMEM;
 		goto err;
 	}
 
-	ret = i915_gem_object_set_cache_level(ring->scratch.obj, I915_CACHE_LLC);
+	ret = i915_gem_object_set_cache_level(engine->scratch.obj,
+					      I915_CACHE_LLC);
 	if (ret)
 		goto err_unref;
 
-	ret = i915_gem_obj_ggtt_pin(ring->scratch.obj, 4096, 0);
+	ret = i915_gem_obj_ggtt_pin(engine->scratch.obj, 4096, 0);
 	if (ret)
 		goto err_unref;
 
-	ring->scratch.gtt_offset = i915_gem_obj_ggtt_offset(ring->scratch.obj);
-	ring->scratch.cpu_page = kmap(sg_page(ring->scratch.obj->pages->sgl));
-	if (ring->scratch.cpu_page == NULL) {
+	engine->scratch.gtt_offset = i915_gem_obj_ggtt_offset(engine->scratch.obj);
+	engine->scratch.cpu_page = kmap(sg_page(engine->scratch.obj->pages->sgl));
+	if (engine->scratch.cpu_page == NULL) {
 		ret = -ENOMEM;
 		goto err_unpin;
 	}
 
 	DRM_DEBUG_DRIVER("%s pipe control offset: 0x%08x\n",
-			 ring->name, ring->scratch.gtt_offset);
+			 engine->name, engine->scratch.gtt_offset);
 	return 0;
 
 err_unpin:
-	i915_gem_object_ggtt_unpin(ring->scratch.obj);
+	i915_gem_object_ggtt_unpin(engine->scratch.obj);
 err_unref:
-	drm_gem_object_unreference(&ring->scratch.obj->base);
+	drm_gem_object_unreference(&engine->scratch.obj->base);
 err:
 	return ret;
 }
@@ -700,15 +709,15 @@ err:
 static int intel_ring_workarounds_emit(struct drm_i915_gem_request *req)
 {
 	int ret, i;
-	struct intel_engine_cs *ring = req->ring;
-	struct drm_device *dev = ring->dev;
+	struct intel_engine_cs *engine = req->engine;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_workarounds *w = &dev_priv->workarounds;
 
 	if (w->count == 0)
 		return 0;
 
-	ring->gpu_caches_dirty = true;
+	engine->gpu_caches_dirty = true;
 	ret = intel_ring_flush_all_caches(req);
 	if (ret)
 		return ret;
@@ -717,16 +726,16 @@ static int intel_ring_workarounds_emit(struct drm_i915_gem_request *req)
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(w->count));
+	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(w->count));
 	for (i = 0; i < w->count; i++) {
-		intel_ring_emit_reg(ring, w->reg[i].addr);
-		intel_ring_emit(ring, w->reg[i].value);
+		intel_ring_emit_reg(engine, w->reg[i].addr);
+		intel_ring_emit(engine, w->reg[i].value);
 	}
-	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_emit(engine, MI_NOOP);
 
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 
-	ring->gpu_caches_dirty = true;
+	engine->gpu_caches_dirty = true;
 	ret = intel_ring_flush_all_caches(req);
 	if (ret)
 		return ret;
@@ -789,25 +798,26 @@ static int wa_add(struct drm_i915_private *dev_priv,
 
 #define WA_WRITE(addr, val) WA_REG(addr, 0xffffffff, val)
 
-static int wa_ring_whitelist_reg(struct intel_engine_cs *ring, i915_reg_t reg)
+static int wa_ring_whitelist_reg(struct intel_engine_cs *engine,
+				 i915_reg_t reg)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 	struct i915_workarounds *wa = &dev_priv->workarounds;
-	const uint32_t index = wa->hw_whitelist_count[ring->id];
+	const uint32_t index = wa->hw_whitelist_count[engine->id];
 
 	if (WARN_ON(index >= RING_MAX_NONPRIV_SLOTS))
 		return -EINVAL;
 
-	WA_WRITE(RING_FORCE_TO_NONPRIV(ring->mmio_base, index),
+	WA_WRITE(RING_FORCE_TO_NONPRIV(engine->mmio_base, index),
 		 i915_mmio_reg_offset(reg));
-	wa->hw_whitelist_count[ring->id]++;
+	wa->hw_whitelist_count[engine->id]++;
 
 	return 0;
 }
 
-static int gen8_init_workarounds(struct intel_engine_cs *ring)
+static int gen8_init_workarounds(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING);
@@ -857,13 +867,13 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 	return 0;
 }
 
-static int bdw_init_workarounds(struct intel_engine_cs *ring)
+static int bdw_init_workarounds(struct intel_engine_cs *engine)
 {
 	int ret;
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	ret = gen8_init_workarounds(ring);
+	ret = gen8_init_workarounds(engine);
 	if (ret)
 		return ret;
 
@@ -886,13 +896,13 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 	return 0;
 }
 
-static int chv_init_workarounds(struct intel_engine_cs *ring)
+static int chv_init_workarounds(struct intel_engine_cs *engine)
 {
 	int ret;
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	ret = gen8_init_workarounds(ring);
+	ret = gen8_init_workarounds(engine);
 	if (ret)
 		return ret;
 
@@ -905,9 +915,9 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 	return 0;
 }
 
-static int gen9_init_workarounds(struct intel_engine_cs *ring)
+static int gen9_init_workarounds(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t tmp;
 	int ret;
@@ -920,8 +930,10 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
 		   ECOCHK_DIS_TLB);
 
+	/* WaClearFlowControlGpgpuContextSave:skl,bxt */
 	/* WaDisablePartialInstShootdown:skl,bxt */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
+			  FLOW_CONTROL_ENABLE |
 			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE);
 
 	/* Syncing dependencies between camera and graphics:skl,bxt */
@@ -947,9 +959,10 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 	}
 
 	/* WaEnableYV12BugFixInHalfSliceChicken7:skl,bxt */
-	if (IS_SKL_REVID(dev, SKL_REVID_C0, REVID_FOREVER) || IS_BROXTON(dev))
-		WA_SET_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN7,
-				  GEN9_ENABLE_YV12_BUGFIX);
+	/* WaEnableSamplerGPGPUPreemptionSupport:skl,bxt */
+	WA_SET_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN7,
+			  GEN9_ENABLE_YV12_BUGFIX |
+			  GEN9_ENABLE_GPGPU_PREEMPTION);
 
 	/* Wa4x4STCOptimizationDisable:skl,bxt */
 	/* WaDisablePartialResolveInVc:skl,bxt */
@@ -986,21 +999,21 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 				    GEN8_LQSC_FLUSH_COHERENT_LINES));
 
 	/* WaEnablePreemptionGranularityControlByUMD:skl,bxt */
-	ret= wa_ring_whitelist_reg(ring, GEN8_CS_CHICKEN1);
+	ret= wa_ring_whitelist_reg(engine, GEN8_CS_CHICKEN1);
 	if (ret)
 		return ret;
 
 	/* WaAllowUMDToModifyHDCChicken1:skl,bxt */
-	ret = wa_ring_whitelist_reg(ring, GEN8_HDC_CHICKEN1);
+	ret = wa_ring_whitelist_reg(engine, GEN8_HDC_CHICKEN1);
 	if (ret)
 		return ret;
 
 	return 0;
 }
 
-static int skl_tune_iz_hashing(struct intel_engine_cs *ring)
+static int skl_tune_iz_hashing(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u8 vals[3] = { 0, 0, 0 };
 	unsigned int i;
@@ -1040,13 +1053,13 @@ static int skl_tune_iz_hashing(struct intel_engine_cs *ring)
 	return 0;
 }
 
-static int skl_init_workarounds(struct intel_engine_cs *ring)
+static int skl_init_workarounds(struct intel_engine_cs *engine)
 {
 	int ret;
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	ret = gen9_init_workarounds(ring);
+	ret = gen9_init_workarounds(engine);
 	if (ret)
 		return ret;
 
@@ -1114,20 +1127,20 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 			GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE);
 
 	/* WaDisableLSQCROPERFforOCL:skl */
-	ret = wa_ring_whitelist_reg(ring, GEN8_L3SQCREG4);
+	ret = wa_ring_whitelist_reg(engine, GEN8_L3SQCREG4);
 	if (ret)
 		return ret;
 
-	return skl_tune_iz_hashing(ring);
+	return skl_tune_iz_hashing(engine);
 }
 
-static int bxt_init_workarounds(struct intel_engine_cs *ring)
+static int bxt_init_workarounds(struct intel_engine_cs *engine)
 {
 	int ret;
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	ret = gen9_init_workarounds(ring);
+	ret = gen9_init_workarounds(engine);
 	if (ret)
 		return ret;
 
@@ -1158,11 +1171,11 @@ static int bxt_init_workarounds(struct intel_engine_cs *ring)
 	/* WaDisableObjectLevelPreemtionForInstanceId:bxt */
 	/* WaDisableLSQCROPERFforOCL:bxt */
 	if (IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
-		ret = wa_ring_whitelist_reg(ring, GEN9_CS_DEBUG_MODE1);
+		ret = wa_ring_whitelist_reg(engine, GEN9_CS_DEBUG_MODE1);
 		if (ret)
 			return ret;
 
-		ret = wa_ring_whitelist_reg(ring, GEN8_L3SQCREG4);
+		ret = wa_ring_whitelist_reg(engine, GEN8_L3SQCREG4);
 		if (ret)
 			return ret;
 	}
@@ -1170,36 +1183,36 @@ static int bxt_init_workarounds(struct intel_engine_cs *ring)
 	return 0;
 }
 
-int init_workarounds_ring(struct intel_engine_cs *ring)
+int init_workarounds_ring(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	WARN_ON(ring->id != RCS);
+	WARN_ON(engine->id != RCS);
 
 	dev_priv->workarounds.count = 0;
 	dev_priv->workarounds.hw_whitelist_count[RCS] = 0;
 
 	if (IS_BROADWELL(dev))
-		return bdw_init_workarounds(ring);
+		return bdw_init_workarounds(engine);
 
 	if (IS_CHERRYVIEW(dev))
-		return chv_init_workarounds(ring);
+		return chv_init_workarounds(engine);
 
 	if (IS_SKYLAKE(dev))
-		return skl_init_workarounds(ring);
+		return skl_init_workarounds(engine);
 
 	if (IS_BROXTON(dev))
-		return bxt_init_workarounds(ring);
+		return bxt_init_workarounds(engine);
 
 	return 0;
 }
 
-static int init_render_ring(struct intel_engine_cs *ring)
+static int init_render_ring(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret = init_ring_common(ring);
+	int ret = init_ring_common(engine);
 	if (ret)
 		return ret;
 
@@ -1242,14 +1255,14 @@ static int init_render_ring(struct intel_engine_cs *ring)
 		I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING));
 
 	if (HAS_L3_DPF(dev))
-		I915_WRITE_IMR(ring, ~GT_PARITY_ERROR(dev));
+		I915_WRITE_IMR(engine, ~GT_PARITY_ERROR(dev));
 
-	return init_workarounds_ring(ring);
+	return init_workarounds_ring(engine);
 }
 
-static void render_ring_cleanup(struct intel_engine_cs *ring)
+static void render_ring_cleanup(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (dev_priv->semaphore_obj) {
@@ -1258,18 +1271,19 @@ static void render_ring_cleanup(struct intel_engine_cs *ring)
 		dev_priv->semaphore_obj = NULL;
 	}
 
-	intel_fini_pipe_control(ring);
+	intel_fini_pipe_control(engine);
 }
 
 static int gen8_rcs_signal(struct drm_i915_gem_request *signaller_req,
 			   unsigned int num_dwords)
 {
 #define MBOX_UPDATE_DWORDS 8
-	struct intel_engine_cs *signaller = signaller_req->ring;
+	struct intel_engine_cs *signaller = signaller_req->engine;
 	struct drm_device *dev = signaller->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *waiter;
-	int i, ret, num_rings;
+	enum intel_engine_id id;
+	int ret, num_rings;
 
 	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
 	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
@@ -1279,9 +1293,9 @@ static int gen8_rcs_signal(struct drm_i915_gem_request *signaller_req,
 	if (ret)
 		return ret;
 
-	for_each_ring(waiter, dev_priv, i) {
+	for_each_engine_id(waiter, dev_priv, id) {
 		u32 seqno;
-		u64 gtt_offset = signaller->semaphore.signal_ggtt[i];
+		u64 gtt_offset = signaller->semaphore.signal_ggtt[id];
 		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
 			continue;
 
@@ -1306,11 +1320,12 @@ static int gen8_xcs_signal(struct drm_i915_gem_request *signaller_req,
 			   unsigned int num_dwords)
 {
 #define MBOX_UPDATE_DWORDS 6
-	struct intel_engine_cs *signaller = signaller_req->ring;
+	struct intel_engine_cs *signaller = signaller_req->engine;
 	struct drm_device *dev = signaller->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *waiter;
-	int i, ret, num_rings;
+	enum intel_engine_id id;
+	int ret, num_rings;
 
 	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
 	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
@@ -1320,9 +1335,9 @@ static int gen8_xcs_signal(struct drm_i915_gem_request *signaller_req,
 	if (ret)
 		return ret;
 
-	for_each_ring(waiter, dev_priv, i) {
+	for_each_engine_id(waiter, dev_priv, id) {
 		u32 seqno;
-		u64 gtt_offset = signaller->semaphore.signal_ggtt[i];
+		u64 gtt_offset = signaller->semaphore.signal_ggtt[id];
 		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
 			continue;
 
@@ -1344,11 +1359,12 @@ static int gen8_xcs_signal(struct drm_i915_gem_request *signaller_req,
 static int gen6_signal(struct drm_i915_gem_request *signaller_req,
 		       unsigned int num_dwords)
 {
-	struct intel_engine_cs *signaller = signaller_req->ring;
+	struct intel_engine_cs *signaller = signaller_req->engine;
 	struct drm_device *dev = signaller->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *useless;
-	int i, ret, num_rings;
+	enum intel_engine_id id;
+	int ret, num_rings;
 
 #define MBOX_UPDATE_DWORDS 3
 	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
@@ -1359,8 +1375,8 @@ static int gen6_signal(struct drm_i915_gem_request *signaller_req,
 	if (ret)
 		return ret;
 
-	for_each_ring(useless, dev_priv, i) {
-		i915_reg_t mbox_reg = signaller->semaphore.mbox.signal[i];
+	for_each_engine_id(useless, dev_priv, id) {
+		i915_reg_t mbox_reg = signaller->semaphore.mbox.signal[id];
 
 		if (i915_mmio_reg_valid(mbox_reg)) {
 			u32 seqno = i915_gem_request_get_seqno(signaller_req);
@@ -1389,22 +1405,23 @@ static int gen6_signal(struct drm_i915_gem_request *signaller_req,
 static int
 gen6_add_request(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
-	if (ring->semaphore.signal)
-		ret = ring->semaphore.signal(req, 4);
+	if (engine->semaphore.signal)
+		ret = engine->semaphore.signal(req, 4);
 	else
 		ret = intel_ring_begin(req, 4);
 
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
-	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
-	intel_ring_emit(ring, i915_gem_request_get_seqno(req));
-	intel_ring_emit(ring, MI_USER_INTERRUPT);
-	__intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_STORE_DWORD_INDEX);
+	intel_ring_emit(engine,
+			I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
+	intel_ring_emit(engine, i915_gem_request_get_seqno(req));
+	intel_ring_emit(engine, MI_USER_INTERRUPT);
+	__intel_ring_advance(engine);
 
 	return 0;
 }
@@ -1429,7 +1446,7 @@ gen8_ring_sync(struct drm_i915_gem_request *waiter_req,
 	       struct intel_engine_cs *signaller,
 	       u32 seqno)
 {
-	struct intel_engine_cs *waiter = waiter_req->ring;
+	struct intel_engine_cs *waiter = waiter_req->engine;
 	struct drm_i915_private *dev_priv = waiter->dev->dev_private;
 	int ret;
 
@@ -1455,7 +1472,7 @@ gen6_ring_sync(struct drm_i915_gem_request *waiter_req,
 	       struct intel_engine_cs *signaller,
 	       u32 seqno)
 {
-	struct intel_engine_cs *waiter = waiter_req->ring;
+	struct intel_engine_cs *waiter = waiter_req->engine;
 	u32 dw1 = MI_SEMAPHORE_MBOX |
 		  MI_SEMAPHORE_COMPARE |
 		  MI_SEMAPHORE_REGISTER;
@@ -1503,8 +1520,8 @@ do {									\
 static int
 pc_render_add_request(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
-	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	struct intel_engine_cs *engine = req->engine;
+	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/* For Ironlake, MI_USER_INTERRUPT was deprecated and apparently
@@ -1519,78 +1536,87 @@ pc_render_add_request(struct drm_i915_gem_request *req)
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE |
+	intel_ring_emit(engine,
+			GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE |
 			PIPE_CONTROL_WRITE_FLUSH |
 			PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE);
-	intel_ring_emit(ring, ring->scratch.gtt_offset | PIPE_CONTROL_GLOBAL_GTT);
-	intel_ring_emit(ring, i915_gem_request_get_seqno(req));
-	intel_ring_emit(ring, 0);
-	PIPE_CONTROL_FLUSH(ring, scratch_addr);
+	intel_ring_emit(engine,
+			engine->scratch.gtt_offset | PIPE_CONTROL_GLOBAL_GTT);
+	intel_ring_emit(engine, i915_gem_request_get_seqno(req));
+	intel_ring_emit(engine, 0);
+	PIPE_CONTROL_FLUSH(engine, scratch_addr);
 	scratch_addr += 2 * CACHELINE_BYTES; /* write to separate cachelines */
-	PIPE_CONTROL_FLUSH(ring, scratch_addr);
+	PIPE_CONTROL_FLUSH(engine, scratch_addr);
 	scratch_addr += 2 * CACHELINE_BYTES;
-	PIPE_CONTROL_FLUSH(ring, scratch_addr);
+	PIPE_CONTROL_FLUSH(engine, scratch_addr);
 	scratch_addr += 2 * CACHELINE_BYTES;
-	PIPE_CONTROL_FLUSH(ring, scratch_addr);
+	PIPE_CONTROL_FLUSH(engine, scratch_addr);
 	scratch_addr += 2 * CACHELINE_BYTES;
-	PIPE_CONTROL_FLUSH(ring, scratch_addr);
+	PIPE_CONTROL_FLUSH(engine, scratch_addr);
 	scratch_addr += 2 * CACHELINE_BYTES;
-	PIPE_CONTROL_FLUSH(ring, scratch_addr);
+	PIPE_CONTROL_FLUSH(engine, scratch_addr);
 
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE |
+	intel_ring_emit(engine,
+			GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE |
 			PIPE_CONTROL_WRITE_FLUSH |
 			PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE |
 			PIPE_CONTROL_NOTIFY);
-	intel_ring_emit(ring, ring->scratch.gtt_offset | PIPE_CONTROL_GLOBAL_GTT);
-	intel_ring_emit(ring, i915_gem_request_get_seqno(req));
-	intel_ring_emit(ring, 0);
-	__intel_ring_advance(ring);
+	intel_ring_emit(engine,
+			engine->scratch.gtt_offset | PIPE_CONTROL_GLOBAL_GTT);
+	intel_ring_emit(engine, i915_gem_request_get_seqno(req));
+	intel_ring_emit(engine, 0);
+	__intel_ring_advance(engine);
 
 	return 0;
 }
 
-static u32
-gen6_ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+static void
+gen6_seqno_barrier(struct intel_engine_cs *engine)
 {
 	/* Workaround to force correct ordering between irq and seqno writes on
 	 * ivb (and maybe also on snb) by reading from a CS register (like
-	 * ACTHD) before reading the status page. */
-	if (!lazy_coherency) {
-		struct drm_i915_private *dev_priv = ring->dev->dev_private;
-		POSTING_READ(RING_ACTHD(ring->mmio_base));
-	}
-
-	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
+	 * ACTHD) before reading the status page.
+	 *
+	 * Note that this effectively stalls the read by the time it takes to
+	 * do a memory transaction, which more or less ensures that the write
+	 * from the GPU has sufficient time to invalidate the CPU cacheline.
+	 * Alternatively we could delay the interrupt from the CS ring to give
+	 * the write time to land, but that would incur a delay after every
+	 * batch i.e. much more frequent than a delay when waiting for the
+	 * interrupt (with the same net latency).
+	 */
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
+	POSTING_READ_FW(RING_ACTHD(engine->mmio_base));
 }
 
 static u32
-ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+ring_get_seqno(struct intel_engine_cs *engine)
 {
-	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
+	return intel_read_status_page(engine, I915_GEM_HWS_INDEX);
 }
 
 static void
-ring_set_seqno(struct intel_engine_cs *ring, u32 seqno)
+ring_set_seqno(struct intel_engine_cs *engine, u32 seqno)
 {
-	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
+	intel_write_status_page(engine, I915_GEM_HWS_INDEX, seqno);
 }
 
 static u32
-pc_render_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+pc_render_get_seqno(struct intel_engine_cs *engine)
 {
-	return ring->scratch.cpu_page[0];
+	return engine->scratch.cpu_page[0];
 }
 
 static void
-pc_render_set_seqno(struct intel_engine_cs *ring, u32 seqno)
+pc_render_set_seqno(struct intel_engine_cs *engine, u32 seqno)
 {
-	ring->scratch.cpu_page[0] = seqno;
+	engine->scratch.cpu_page[0] = seqno;
 }
 
 static bool
-gen5_ring_get_irq(struct intel_engine_cs *ring)
+gen5_ring_get_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
@@ -1598,30 +1624,30 @@ gen5_ring_get_irq(struct intel_engine_cs *ring)
 		return false;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (ring->irq_refcount++ == 0)
-		gen5_enable_gt_irq(dev_priv, ring->irq_enable_mask);
+	if (engine->irq_refcount++ == 0)
+		gen5_enable_gt_irq(dev_priv, engine->irq_enable_mask);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 
 	return true;
 }
 
 static void
-gen5_ring_put_irq(struct intel_engine_cs *ring)
+gen5_ring_put_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (--ring->irq_refcount == 0)
-		gen5_disable_gt_irq(dev_priv, ring->irq_enable_mask);
+	if (--engine->irq_refcount == 0)
+		gen5_disable_gt_irq(dev_priv, engine->irq_enable_mask);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
 
 static bool
-i9xx_ring_get_irq(struct intel_engine_cs *ring)
+i9xx_ring_get_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
@@ -1629,8 +1655,8 @@ i9xx_ring_get_irq(struct intel_engine_cs *ring)
 		return false;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (ring->irq_refcount++ == 0) {
-		dev_priv->irq_mask &= ~ring->irq_enable_mask;
+	if (engine->irq_refcount++ == 0) {
+		dev_priv->irq_mask &= ~engine->irq_enable_mask;
 		I915_WRITE(IMR, dev_priv->irq_mask);
 		POSTING_READ(IMR);
 	}
@@ -1640,15 +1666,15 @@ i9xx_ring_get_irq(struct intel_engine_cs *ring)
 }
 
 static void
-i9xx_ring_put_irq(struct intel_engine_cs *ring)
+i9xx_ring_put_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (--ring->irq_refcount == 0) {
-		dev_priv->irq_mask |= ring->irq_enable_mask;
+	if (--engine->irq_refcount == 0) {
+		dev_priv->irq_mask |= engine->irq_enable_mask;
 		I915_WRITE(IMR, dev_priv->irq_mask);
 		POSTING_READ(IMR);
 	}
@@ -1656,9 +1682,9 @@ i9xx_ring_put_irq(struct intel_engine_cs *ring)
 }
 
 static bool
-i8xx_ring_get_irq(struct intel_engine_cs *ring)
+i8xx_ring_get_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
@@ -1666,8 +1692,8 @@ i8xx_ring_get_irq(struct intel_engine_cs *ring)
 		return false;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (ring->irq_refcount++ == 0) {
-		dev_priv->irq_mask &= ~ring->irq_enable_mask;
+	if (engine->irq_refcount++ == 0) {
+		dev_priv->irq_mask &= ~engine->irq_enable_mask;
 		I915_WRITE16(IMR, dev_priv->irq_mask);
 		POSTING_READ16(IMR);
 	}
@@ -1677,15 +1703,15 @@ i8xx_ring_get_irq(struct intel_engine_cs *ring)
 }
 
 static void
-i8xx_ring_put_irq(struct intel_engine_cs *ring)
+i8xx_ring_put_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (--ring->irq_refcount == 0) {
-		dev_priv->irq_mask |= ring->irq_enable_mask;
+	if (--engine->irq_refcount == 0) {
+		dev_priv->irq_mask |= engine->irq_enable_mask;
 		I915_WRITE16(IMR, dev_priv->irq_mask);
 		POSTING_READ16(IMR);
 	}
@@ -1697,42 +1723,43 @@ bsd_ring_flush(struct drm_i915_gem_request *req,
 	       u32     invalidate_domains,
 	       u32     flush_domains)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_FLUSH);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_FLUSH);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 	return 0;
 }
 
 static int
 i9xx_add_request(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 4);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
-	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
-	intel_ring_emit(ring, i915_gem_request_get_seqno(req));
-	intel_ring_emit(ring, MI_USER_INTERRUPT);
-	__intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_STORE_DWORD_INDEX);
+	intel_ring_emit(engine,
+			I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
+	intel_ring_emit(engine, i915_gem_request_get_seqno(req));
+	intel_ring_emit(engine, MI_USER_INTERRUPT);
+	__intel_ring_advance(engine);
 
 	return 0;
 }
 
 static bool
-gen6_ring_get_irq(struct intel_engine_cs *ring)
+gen6_ring_get_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
@@ -1740,14 +1767,14 @@ gen6_ring_get_irq(struct intel_engine_cs *ring)
 		return false;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (ring->irq_refcount++ == 0) {
-		if (HAS_L3_DPF(dev) && ring->id == RCS)
-			I915_WRITE_IMR(ring,
-				       ~(ring->irq_enable_mask |
+	if (engine->irq_refcount++ == 0) {
+		if (HAS_L3_DPF(dev) && engine->id == RCS)
+			I915_WRITE_IMR(engine,
+				       ~(engine->irq_enable_mask |
 					 GT_PARITY_ERROR(dev)));
 		else
-			I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
-		gen5_enable_gt_irq(dev_priv, ring->irq_enable_mask);
+			I915_WRITE_IMR(engine, ~engine->irq_enable_mask);
+		gen5_enable_gt_irq(dev_priv, engine->irq_enable_mask);
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 
@@ -1755,27 +1782,27 @@ gen6_ring_get_irq(struct intel_engine_cs *ring)
 }
 
 static void
-gen6_ring_put_irq(struct intel_engine_cs *ring)
+gen6_ring_put_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (--ring->irq_refcount == 0) {
-		if (HAS_L3_DPF(dev) && ring->id == RCS)
-			I915_WRITE_IMR(ring, ~GT_PARITY_ERROR(dev));
+	if (--engine->irq_refcount == 0) {
+		if (HAS_L3_DPF(dev) && engine->id == RCS)
+			I915_WRITE_IMR(engine, ~GT_PARITY_ERROR(dev));
 		else
-			I915_WRITE_IMR(ring, ~0);
-		gen5_disable_gt_irq(dev_priv, ring->irq_enable_mask);
+			I915_WRITE_IMR(engine, ~0);
+		gen5_disable_gt_irq(dev_priv, engine->irq_enable_mask);
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
 
 static bool
-hsw_vebox_get_irq(struct intel_engine_cs *ring)
+hsw_vebox_get_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
@@ -1783,9 +1810,9 @@ hsw_vebox_get_irq(struct intel_engine_cs *ring)
 		return false;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (ring->irq_refcount++ == 0) {
-		I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
-		gen6_enable_pm_irq(dev_priv, ring->irq_enable_mask);
+	if (engine->irq_refcount++ == 0) {
+		I915_WRITE_IMR(engine, ~engine->irq_enable_mask);
+		gen6_enable_pm_irq(dev_priv, engine->irq_enable_mask);
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 
@@ -1793,24 +1820,24 @@ hsw_vebox_get_irq(struct intel_engine_cs *ring)
 }
 
 static void
-hsw_vebox_put_irq(struct intel_engine_cs *ring)
+hsw_vebox_put_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (--ring->irq_refcount == 0) {
-		I915_WRITE_IMR(ring, ~0);
-		gen6_disable_pm_irq(dev_priv, ring->irq_enable_mask);
+	if (--engine->irq_refcount == 0) {
+		I915_WRITE_IMR(engine, ~0);
+		gen6_disable_pm_irq(dev_priv, engine->irq_enable_mask);
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
 
 static bool
-gen8_ring_get_irq(struct intel_engine_cs *ring)
+gen8_ring_get_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
@@ -1818,15 +1845,15 @@ gen8_ring_get_irq(struct intel_engine_cs *ring)
 		return false;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (ring->irq_refcount++ == 0) {
-		if (HAS_L3_DPF(dev) && ring->id == RCS) {
-			I915_WRITE_IMR(ring,
-				       ~(ring->irq_enable_mask |
+	if (engine->irq_refcount++ == 0) {
+		if (HAS_L3_DPF(dev) && engine->id == RCS) {
+			I915_WRITE_IMR(engine,
+				       ~(engine->irq_enable_mask |
 					 GT_RENDER_L3_PARITY_ERROR_INTERRUPT));
 		} else {
-			I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
+			I915_WRITE_IMR(engine, ~engine->irq_enable_mask);
 		}
-		POSTING_READ(RING_IMR(ring->mmio_base));
+		POSTING_READ(RING_IMR(engine->mmio_base));
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 
@@ -1834,21 +1861,21 @@ gen8_ring_get_irq(struct intel_engine_cs *ring)
 }
 
 static void
-gen8_ring_put_irq(struct intel_engine_cs *ring)
+gen8_ring_put_irq(struct intel_engine_cs *engine)
 {
-	struct drm_device *dev = ring->dev;
+	struct drm_device *dev = engine->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	if (--ring->irq_refcount == 0) {
-		if (HAS_L3_DPF(dev) && ring->id == RCS) {
-			I915_WRITE_IMR(ring,
+	if (--engine->irq_refcount == 0) {
+		if (HAS_L3_DPF(dev) && engine->id == RCS) {
+			I915_WRITE_IMR(engine,
 				       ~GT_RENDER_L3_PARITY_ERROR_INTERRUPT);
 		} else {
-			I915_WRITE_IMR(ring, ~0);
+			I915_WRITE_IMR(engine, ~0);
 		}
-		POSTING_READ(RING_IMR(ring->mmio_base));
+		POSTING_READ(RING_IMR(engine->mmio_base));
 	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
@@ -1858,20 +1885,20 @@ i965_dispatch_execbuffer(struct drm_i915_gem_request *req,
 			 u64 offset, u32 length,
 			 unsigned dispatch_flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring,
+	intel_ring_emit(engine,
 			MI_BATCH_BUFFER_START |
 			MI_BATCH_GTT |
 			(dispatch_flags & I915_DISPATCH_SECURE ?
 			 0 : MI_BATCH_NON_SECURE_I965));
-	intel_ring_emit(ring, offset);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, offset);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -1885,8 +1912,8 @@ i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
 			 u64 offset, u32 len,
 			 unsigned dispatch_flags)
 {
-	struct intel_engine_cs *ring = req->ring;
-	u32 cs_offset = ring->scratch.gtt_offset;
+	struct intel_engine_cs *engine = req->engine;
+	u32 cs_offset = engine->scratch.gtt_offset;
 	int ret;
 
 	ret = intel_ring_begin(req, 6);
@@ -1894,13 +1921,13 @@ i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
 		return ret;
 
 	/* Evict the invalid PTE TLBs */
-	intel_ring_emit(ring, COLOR_BLT_CMD | BLT_WRITE_RGBA);
-	intel_ring_emit(ring, BLT_DEPTH_32 | BLT_ROP_COLOR_COPY | 4096);
-	intel_ring_emit(ring, I830_TLB_ENTRIES << 16 | 4); /* load each page */
-	intel_ring_emit(ring, cs_offset);
-	intel_ring_emit(ring, 0xdeadbeef);
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, COLOR_BLT_CMD | BLT_WRITE_RGBA);
+	intel_ring_emit(engine, BLT_DEPTH_32 | BLT_ROP_COLOR_COPY | 4096);
+	intel_ring_emit(engine, I830_TLB_ENTRIES << 16 | 4); /* load each page */
+	intel_ring_emit(engine, cs_offset);
+	intel_ring_emit(engine, 0xdeadbeef);
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	if ((dispatch_flags & I915_DISPATCH_PINNED) == 0) {
 		if (len > I830_BATCH_LIMIT)
@@ -1914,16 +1941,17 @@ i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
 		 * stable batch scratch bo area (so that the CS never
 		 * stumbles over its tlb invalidation bug) ...
 		 */
-		intel_ring_emit(ring, SRC_COPY_BLT_CMD | BLT_WRITE_RGBA);
-		intel_ring_emit(ring, BLT_DEPTH_32 | BLT_ROP_SRC_COPY | 4096);
-		intel_ring_emit(ring, DIV_ROUND_UP(len, 4096) << 16 | 4096);
-		intel_ring_emit(ring, cs_offset);
-		intel_ring_emit(ring, 4096);
-		intel_ring_emit(ring, offset);
-
-		intel_ring_emit(ring, MI_FLUSH);
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_advance(ring);
+		intel_ring_emit(engine, SRC_COPY_BLT_CMD | BLT_WRITE_RGBA);
+		intel_ring_emit(engine,
+				BLT_DEPTH_32 | BLT_ROP_SRC_COPY | 4096);
+		intel_ring_emit(engine, DIV_ROUND_UP(len, 4096) << 16 | 4096);
+		intel_ring_emit(engine, cs_offset);
+		intel_ring_emit(engine, 4096);
+		intel_ring_emit(engine, offset);
+
+		intel_ring_emit(engine, MI_FLUSH);
+		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_advance(engine);
 
 		/* ... and execute it. */
 		offset = cs_offset;
@@ -1933,10 +1961,10 @@ i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
-	intel_ring_emit(ring, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
-					0 : MI_BATCH_NON_SECURE));
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
+	intel_ring_emit(engine, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
+					  0 : MI_BATCH_NON_SECURE));
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -1946,55 +1974,55 @@ i915_dispatch_execbuffer(struct drm_i915_gem_request *req,
 			 u64 offset, u32 len,
 			 unsigned dispatch_flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
-	intel_ring_emit(ring, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
-					0 : MI_BATCH_NON_SECURE));
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
+	intel_ring_emit(engine, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
+					  0 : MI_BATCH_NON_SECURE));
+	intel_ring_advance(engine);
 
 	return 0;
 }
 
-static void cleanup_phys_status_page(struct intel_engine_cs *ring)
+static void cleanup_phys_status_page(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(ring->dev);
+	struct drm_i915_private *dev_priv = to_i915(engine->dev);
 
 	if (!dev_priv->status_page_dmah)
 		return;
 
-	drm_pci_free(ring->dev, dev_priv->status_page_dmah);
-	ring->status_page.page_addr = NULL;
+	drm_pci_free(engine->dev, dev_priv->status_page_dmah);
+	engine->status_page.page_addr = NULL;
 }
 
-static void cleanup_status_page(struct intel_engine_cs *ring)
+static void cleanup_status_page(struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_object *obj;
 
-	obj = ring->status_page.obj;
+	obj = engine->status_page.obj;
 	if (obj == NULL)
 		return;
 
 	kunmap(sg_page(obj->pages->sgl));
 	i915_gem_object_ggtt_unpin(obj);
 	drm_gem_object_unreference(&obj->base);
-	ring->status_page.obj = NULL;
+	engine->status_page.obj = NULL;
 }
 
-static int init_status_page(struct intel_engine_cs *ring)
+static int init_status_page(struct intel_engine_cs *engine)
 {
-	struct drm_i915_gem_object *obj = ring->status_page.obj;
+	struct drm_i915_gem_object *obj = engine->status_page.obj;
 
 	if (obj == NULL) {
 		unsigned flags;
 		int ret;
 
-		obj = i915_gem_alloc_object(ring->dev, 4096);
+		obj = i915_gem_alloc_object(engine->dev, 4096);
 		if (obj == NULL) {
 			DRM_ERROR("Failed to allocate status page\n");
 			return -ENOMEM;
@@ -2005,7 +2033,7 @@ static int init_status_page(struct intel_engine_cs *ring)
 			goto err_unref;
 
 		flags = 0;
-		if (!HAS_LLC(ring->dev))
+		if (!HAS_LLC(engine->dev))
 			/* On g33, we cannot place HWS above 256MiB, so
 			 * restrict its pinning to the low mappable arena.
 			 * Though this restriction is not documented for
@@ -2024,32 +2052,32 @@ err_unref:
 			return ret;
 		}
 
-		ring->status_page.obj = obj;
+		engine->status_page.obj = obj;
 	}
 
-	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(obj);
-	ring->status_page.page_addr = kmap(sg_page(obj->pages->sgl));
-	memset(ring->status_page.page_addr, 0, PAGE_SIZE);
+	engine->status_page.gfx_addr = i915_gem_obj_ggtt_offset(obj);
+	engine->status_page.page_addr = kmap(sg_page(obj->pages->sgl));
+	memset(engine->status_page.page_addr, 0, PAGE_SIZE);
 
 	DRM_DEBUG_DRIVER("%s hws offset: 0x%08x\n",
-			ring->name, ring->status_page.gfx_addr);
+			engine->name, engine->status_page.gfx_addr);
 
 	return 0;
 }
 
-static int init_phys_status_page(struct intel_engine_cs *ring)
+static int init_phys_status_page(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 
 	if (!dev_priv->status_page_dmah) {
 		dev_priv->status_page_dmah =
-			drm_pci_alloc(ring->dev, PAGE_SIZE, PAGE_SIZE);
+			drm_pci_alloc(engine->dev, PAGE_SIZE, PAGE_SIZE);
 		if (!dev_priv->status_page_dmah)
 			return -ENOMEM;
 	}
 
-	ring->status_page.page_addr = dev_priv->status_page_dmah->vaddr;
-	memset(ring->status_page.page_addr, 0, PAGE_SIZE);
+	engine->status_page.page_addr = dev_priv->status_page_dmah->vaddr;
+	memset(engine->status_page.page_addr, 0, PAGE_SIZE);
 
 	return 0;
 }
@@ -2057,7 +2085,7 @@ static int init_phys_status_page(struct intel_engine_cs *ring)
 void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
 {
 	if (HAS_LLC(ringbuf->obj->base.dev) && !ringbuf->obj->stolen)
-		vunmap(ringbuf->virtual_start);
+		i915_gem_object_unpin_map(ringbuf->obj);
 	else
 		iounmap(ringbuf->virtual_start);
 	ringbuf->virtual_start = NULL;
@@ -2065,34 +2093,15 @@ void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
 	i915_gem_object_ggtt_unpin(ringbuf->obj);
 }
 
-static u32 *vmap_obj(struct drm_i915_gem_object *obj)
-{
-	struct sg_page_iter sg_iter;
-	struct page **pages;
-	void *addr;
-	int i;
-
-	pages = drm_malloc_ab(obj->base.size >> PAGE_SHIFT, sizeof(*pages));
-	if (pages == NULL)
-		return NULL;
-
-	i = 0;
-	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0)
-		pages[i++] = sg_page_iter_page(&sg_iter);
-
-	addr = vmap(pages, i, 0, PAGE_KERNEL);
-	drm_free_large(pages);
-
-	return addr;
-}
-
 int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 				     struct intel_ringbuffer *ringbuf)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_object *obj = ringbuf->obj;
 	/* Ring wraparound at offset 0 sometimes hangs. No idea why. */
 	unsigned flags = PIN_OFFSET_BIAS | 4096;
+	void *addr;
 	int ret;
 
 	if (HAS_LLC(dev_priv) && !obj->stolen) {
@@ -2101,15 +2110,13 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 			return ret;
 
 		ret = i915_gem_object_set_to_cpu_domain(obj, true);
-		if (ret) {
-			i915_gem_object_ggtt_unpin(obj);
-			return ret;
-		}
+		if (ret)
+			goto err_unpin;
 
-		ringbuf->virtual_start = vmap_obj(obj);
-		if (ringbuf->virtual_start == NULL) {
-			i915_gem_object_ggtt_unpin(obj);
-			return -ENOMEM;
+		addr = i915_gem_object_pin_map(obj);
+		if (IS_ERR(addr)) {
+			ret = PTR_ERR(addr);
+			goto err_unpin;
 		}
 	} else {
 		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE,
@@ -2118,25 +2125,27 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 			return ret;
 
 		ret = i915_gem_object_set_to_gtt_domain(obj, true);
-		if (ret) {
-			i915_gem_object_ggtt_unpin(obj);
-			return ret;
-		}
+		if (ret)
+			goto err_unpin;
 
 		/* Access through the GTT requires the device to be awake. */
 		assert_rpm_wakelock_held(dev_priv);
 
-		ringbuf->virtual_start = ioremap_wc(dev_priv->gtt.mappable_base +
-						    i915_gem_obj_ggtt_offset(obj), ringbuf->size);
-		if (ringbuf->virtual_start == NULL) {
-			i915_gem_object_ggtt_unpin(obj);
-			return -EINVAL;
+		addr = ioremap_wc(ggtt->mappable_base +
+				  i915_gem_obj_ggtt_offset(obj), ringbuf->size);
+		if (addr == NULL) {
+			ret = -ENOMEM;
+			goto err_unpin;
 		}
 	}
 
+	ringbuf->virtual_start = addr;
 	ringbuf->vma = i915_gem_obj_to_ggtt(obj);
-
 	return 0;
+
+err_unpin:
+	i915_gem_object_ggtt_unpin(obj);
+	return ret;
 }
 
 static void intel_destroy_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
@@ -2179,7 +2188,7 @@ intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size)
 		return ERR_PTR(-ENOMEM);
 	}
 
-	ring->ring = engine;
+	ring->engine = engine;
 	list_add(&ring->link, &engine->buffers);
 
 	ring->size = size;
@@ -2215,37 +2224,38 @@ intel_ringbuffer_free(struct intel_ringbuffer *ring)
 }
 
 static int intel_init_ring_buffer(struct drm_device *dev,
-				  struct intel_engine_cs *ring)
+				  struct intel_engine_cs *engine)
 {
 	struct intel_ringbuffer *ringbuf;
 	int ret;
 
-	WARN_ON(ring->buffer);
+	WARN_ON(engine->buffer);
 
-	ring->dev = dev;
-	INIT_LIST_HEAD(&ring->active_list);
-	INIT_LIST_HEAD(&ring->request_list);
-	INIT_LIST_HEAD(&ring->execlist_queue);
-	INIT_LIST_HEAD(&ring->buffers);
-	i915_gem_batch_pool_init(dev, &ring->batch_pool);
-	memset(ring->semaphore.sync_seqno, 0, sizeof(ring->semaphore.sync_seqno));
+	engine->dev = dev;
+	INIT_LIST_HEAD(&engine->active_list);
+	INIT_LIST_HEAD(&engine->request_list);
+	INIT_LIST_HEAD(&engine->execlist_queue);
+	INIT_LIST_HEAD(&engine->buffers);
+	i915_gem_batch_pool_init(dev, &engine->batch_pool);
+	memset(engine->semaphore.sync_seqno, 0,
+	       sizeof(engine->semaphore.sync_seqno));
 
-	init_waitqueue_head(&ring->irq_queue);
+	init_waitqueue_head(&engine->irq_queue);
 
-	ringbuf = intel_engine_create_ringbuffer(ring, 32 * PAGE_SIZE);
+	ringbuf = intel_engine_create_ringbuffer(engine, 32 * PAGE_SIZE);
 	if (IS_ERR(ringbuf)) {
 		ret = PTR_ERR(ringbuf);
 		goto error;
 	}
-	ring->buffer = ringbuf;
+	engine->buffer = ringbuf;
 
 	if (I915_NEED_GFX_HWS(dev)) {
-		ret = init_status_page(ring);
+		ret = init_status_page(engine);
 		if (ret)
 			goto error;
 	} else {
-		WARN_ON(ring->id != RCS);
-		ret = init_phys_status_page(ring);
+		WARN_ON(engine->id != RCS);
+		ret = init_phys_status_page(engine);
 		if (ret)
 			goto error;
 	}
@@ -2253,58 +2263,58 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	ret = intel_pin_and_map_ringbuffer_obj(dev, ringbuf);
 	if (ret) {
 		DRM_ERROR("Failed to pin and map ringbuffer %s: %d\n",
-				ring->name, ret);
+				engine->name, ret);
 		intel_destroy_ringbuffer_obj(ringbuf);
 		goto error;
 	}
 
-	ret = i915_cmd_parser_init_ring(ring);
+	ret = i915_cmd_parser_init_ring(engine);
 	if (ret)
 		goto error;
 
 	return 0;
 
 error:
-	intel_cleanup_ring_buffer(ring);
+	intel_cleanup_engine(engine);
 	return ret;
 }
 
-void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
+void intel_cleanup_engine(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv;
 
-	if (!intel_ring_initialized(ring))
+	if (!intel_engine_initialized(engine))
 		return;
 
-	dev_priv = to_i915(ring->dev);
+	dev_priv = to_i915(engine->dev);
 
-	if (ring->buffer) {
-		intel_stop_ring_buffer(ring);
-		WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
+	if (engine->buffer) {
+		intel_stop_engine(engine);
+		WARN_ON(!IS_GEN2(engine->dev) && (I915_READ_MODE(engine) & MODE_IDLE) == 0);
 
-		intel_unpin_ringbuffer_obj(ring->buffer);
-		intel_ringbuffer_free(ring->buffer);
-		ring->buffer = NULL;
+		intel_unpin_ringbuffer_obj(engine->buffer);
+		intel_ringbuffer_free(engine->buffer);
+		engine->buffer = NULL;
 	}
 
-	if (ring->cleanup)
-		ring->cleanup(ring);
+	if (engine->cleanup)
+		engine->cleanup(engine);
 
-	if (I915_NEED_GFX_HWS(ring->dev)) {
-		cleanup_status_page(ring);
+	if (I915_NEED_GFX_HWS(engine->dev)) {
+		cleanup_status_page(engine);
 	} else {
-		WARN_ON(ring->id != RCS);
-		cleanup_phys_status_page(ring);
+		WARN_ON(engine->id != RCS);
+		cleanup_phys_status_page(engine);
 	}
 
-	i915_cmd_parser_fini_ring(ring);
-	i915_gem_batch_pool_fini(&ring->batch_pool);
-	ring->dev = NULL;
+	i915_cmd_parser_fini_ring(engine);
+	i915_gem_batch_pool_fini(&engine->batch_pool);
+	engine->dev = NULL;
 }
 
-static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
+static int ring_wait_for_space(struct intel_engine_cs *engine, int n)
 {
-	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct intel_ringbuffer *ringbuf = engine->buffer;
 	struct drm_i915_gem_request *request;
 	unsigned space;
 	int ret;
@@ -2315,14 +2325,14 @@ static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
 	/* The whole point of reserving space is to not wait! */
 	WARN_ON(ringbuf->reserved_in_use);
 
-	list_for_each_entry(request, &ring->request_list, list) {
+	list_for_each_entry(request, &engine->request_list, list) {
 		space = __intel_ring_space(request->postfix, ringbuf->tail,
 					   ringbuf->size);
 		if (space >= n)
 			break;
 	}
 
-	if (WARN_ON(&request->list == &ring->request_list))
+	if (WARN_ON(&request->list == &engine->request_list))
 		return -ENOSPC;
 
 	ret = i915_wait_request(request);
@@ -2347,28 +2357,27 @@ static void __wrap_ring_buffer(struct intel_ringbuffer *ringbuf)
 	intel_ring_update_space(ringbuf);
 }
 
-int intel_ring_idle(struct intel_engine_cs *ring)
+int intel_engine_idle(struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_request *req;
 
 	/* Wait upon the last request to be completed */
-	if (list_empty(&ring->request_list))
+	if (list_empty(&engine->request_list))
 		return 0;
 
-	req = list_entry(ring->request_list.prev,
-			struct drm_i915_gem_request,
-			list);
+	req = list_entry(engine->request_list.prev,
+			 struct drm_i915_gem_request,
+			 list);
 
 	/* Make sure we do not trigger any retires */
 	return __i915_wait_request(req,
-				   atomic_read(&to_i915(ring->dev)->gpu_error.reset_counter),
-				   to_i915(ring->dev)->mm.interruptible,
+				   req->i915->mm.interruptible,
 				   NULL, NULL);
 }
 
 int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request)
 {
-	request->ringbuf = request->ring->buffer;
+	request->ringbuf = request->engine->buffer;
 	return 0;
 }
 
@@ -2434,9 +2443,9 @@ void intel_ring_reserved_space_end(struct intel_ringbuffer *ringbuf)
 	ringbuf->reserved_in_use = false;
 }
 
-static int __intel_ring_prepare(struct intel_engine_cs *ring, int bytes)
+static int __intel_ring_prepare(struct intel_engine_cs *engine, int bytes)
 {
-	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct intel_ringbuffer *ringbuf = engine->buffer;
 	int remain_usable = ringbuf->effective_size - ringbuf->tail;
 	int remain_actual = ringbuf->size - ringbuf->tail;
 	int ret, total_bytes, wait_bytes = 0;
@@ -2470,7 +2479,7 @@ static int __intel_ring_prepare(struct intel_engine_cs *ring, int bytes)
 	}
 
 	if (wait_bytes) {
-		ret = ring_wait_for_space(ring, wait_bytes);
+		ret = ring_wait_for_space(engine, wait_bytes);
 		if (unlikely(ret))
 			return ret;
 
@@ -2484,32 +2493,22 @@ static int __intel_ring_prepare(struct intel_engine_cs *ring, int bytes)
 int intel_ring_begin(struct drm_i915_gem_request *req,
 		     int num_dwords)
 {
-	struct intel_engine_cs *ring;
-	struct drm_i915_private *dev_priv;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
-	WARN_ON(req == NULL);
-	ring = req->ring;
-	dev_priv = ring->dev->dev_private;
-
-	ret = i915_gem_check_wedge(&dev_priv->gpu_error,
-				   dev_priv->mm.interruptible);
-	if (ret)
-		return ret;
-
-	ret = __intel_ring_prepare(ring, num_dwords * sizeof(uint32_t));
+	ret = __intel_ring_prepare(engine, num_dwords * sizeof(uint32_t));
 	if (ret)
 		return ret;
 
-	ring->buffer->space -= num_dwords * sizeof(uint32_t);
+	engine->buffer->space -= num_dwords * sizeof(uint32_t);
 	return 0;
 }
 
 /* Align the ring tail to a cacheline boundary */
 int intel_ring_cacheline_align(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
-	int num_dwords = (ring->buffer->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
+	struct intel_engine_cs *engine = req->engine;
+	int num_dwords = (engine->buffer->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
 	int ret;
 
 	if (num_dwords == 0)
@@ -2521,33 +2520,52 @@ int intel_ring_cacheline_align(struct drm_i915_gem_request *req)
 		return ret;
 
 	while (num_dwords--)
-		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(engine, MI_NOOP);
 
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 
 	return 0;
 }
 
-void intel_ring_init_seqno(struct intel_engine_cs *ring, u32 seqno)
+void intel_ring_init_seqno(struct intel_engine_cs *engine, u32 seqno)
 {
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(engine->dev);
 
-	if (INTEL_INFO(dev)->gen == 6 || INTEL_INFO(dev)->gen == 7) {
-		I915_WRITE(RING_SYNC_0(ring->mmio_base), 0);
-		I915_WRITE(RING_SYNC_1(ring->mmio_base), 0);
-		if (HAS_VEBOX(dev))
-			I915_WRITE(RING_SYNC_2(ring->mmio_base), 0);
+	/* Our semaphore implementation is strictly monotonic (i.e. we proceed
+	 * so long as the semaphore value in the register/page is greater
+	 * than the sync value), so whenever we reset the seqno,
+	 * so long as we reset the tracking semaphore value to 0, it will
+	 * always be before the next request's seqno. If we don't reset
+	 * the semaphore value, then when the seqno moves backwards all
+	 * future waits will complete instantly (causing rendering corruption).
+	 */
+	if (INTEL_INFO(dev_priv)->gen == 6 || INTEL_INFO(dev_priv)->gen == 7) {
+		I915_WRITE(RING_SYNC_0(engine->mmio_base), 0);
+		I915_WRITE(RING_SYNC_1(engine->mmio_base), 0);
+		if (HAS_VEBOX(dev_priv))
+			I915_WRITE(RING_SYNC_2(engine->mmio_base), 0);
+	}
+	if (dev_priv->semaphore_obj) {
+		struct drm_i915_gem_object *obj = dev_priv->semaphore_obj;
+		struct page *page = i915_gem_object_get_dirty_page(obj, 0);
+		void *semaphores = kmap(page);
+		memset(semaphores + GEN8_SEMAPHORE_OFFSET(engine->id, 0),
+		       0, I915_NUM_ENGINES * gen8_semaphore_seqno_size);
+		kunmap(page);
 	}
+	memset(engine->semaphore.sync_seqno, 0,
+	       sizeof(engine->semaphore.sync_seqno));
 
-	ring->set_seqno(ring, seqno);
-	ring->hangcheck.seqno = seqno;
+	engine->set_seqno(engine, seqno);
+	engine->last_submitted_seqno = seqno;
+
+	engine->hangcheck.seqno = seqno;
 }
 
-static void gen6_bsd_ring_write_tail(struct intel_engine_cs *ring,
+static void gen6_bsd_ring_write_tail(struct intel_engine_cs *engine,
 				     u32 value)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
 
        /* Every tail move must follow the sequence below */
 
@@ -2567,8 +2585,8 @@ static void gen6_bsd_ring_write_tail(struct intel_engine_cs *ring,
 		DRM_ERROR("timed out waiting for the BSD ring to wake up\n");
 
 	/* Now that the ring is fully powered up, update the tail */
-	I915_WRITE_TAIL(ring, value);
-	POSTING_READ(RING_TAIL(ring->mmio_base));
+	I915_WRITE_TAIL(engine, value);
+	POSTING_READ(RING_TAIL(engine->mmio_base));
 
 	/* Let the ring send IDLE messages to the GT again,
 	 * and so let it sleep to conserve power when idle.
@@ -2580,7 +2598,7 @@ static void gen6_bsd_ring_write_tail(struct intel_engine_cs *ring,
 static int gen6_bsd_ring_flush(struct drm_i915_gem_request *req,
 			       u32 invalidate, u32 flush)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	uint32_t cmd;
 	int ret;
 
@@ -2589,7 +2607,7 @@ static int gen6_bsd_ring_flush(struct drm_i915_gem_request *req,
 		return ret;
 
 	cmd = MI_FLUSH_DW;
-	if (INTEL_INFO(ring->dev)->gen >= 8)
+	if (INTEL_INFO(engine->dev)->gen >= 8)
 		cmd += 1;
 
 	/* We always require a command barrier so that subsequent
@@ -2608,16 +2626,17 @@ static int gen6_bsd_ring_flush(struct drm_i915_gem_request *req,
 	if (invalidate & I915_GEM_GPU_DOMAINS)
 		cmd |= MI_INVALIDATE_TLB | MI_INVALIDATE_BSD;
 
-	intel_ring_emit(ring, cmd);
-	intel_ring_emit(ring, I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
-	if (INTEL_INFO(ring->dev)->gen >= 8) {
-		intel_ring_emit(ring, 0); /* upper addr */
-		intel_ring_emit(ring, 0); /* value */
+	intel_ring_emit(engine, cmd);
+	intel_ring_emit(engine,
+			I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
+	if (INTEL_INFO(engine->dev)->gen >= 8) {
+		intel_ring_emit(engine, 0); /* upper addr */
+		intel_ring_emit(engine, 0); /* value */
 	} else  {
-		intel_ring_emit(ring, 0);
-		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(engine, 0);
+		intel_ring_emit(engine, MI_NOOP);
 	}
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 	return 0;
 }
 
@@ -2626,8 +2645,8 @@ gen8_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
 			      u64 offset, u32 len,
 			      unsigned dispatch_flags)
 {
-	struct intel_engine_cs *ring = req->ring;
-	bool ppgtt = USES_PPGTT(ring->dev) &&
+	struct intel_engine_cs *engine = req->engine;
+	bool ppgtt = USES_PPGTT(engine->dev) &&
 			!(dispatch_flags & I915_DISPATCH_SECURE);
 	int ret;
 
@@ -2636,13 +2655,13 @@ gen8_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
 		return ret;
 
 	/* FIXME(BDW): Address space and security selectors. */
-	intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8) |
+	intel_ring_emit(engine, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8) |
 			(dispatch_flags & I915_DISPATCH_RS ?
 			 MI_BATCH_RESOURCE_STREAMER : 0));
-	intel_ring_emit(ring, lower_32_bits(offset));
-	intel_ring_emit(ring, upper_32_bits(offset));
-	intel_ring_emit(ring, MI_NOOP);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, lower_32_bits(offset));
+	intel_ring_emit(engine, upper_32_bits(offset));
+	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -2652,22 +2671,22 @@ hsw_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
 			     u64 offset, u32 len,
 			     unsigned dispatch_flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring,
+	intel_ring_emit(engine,
 			MI_BATCH_BUFFER_START |
 			(dispatch_flags & I915_DISPATCH_SECURE ?
 			 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW) |
 			(dispatch_flags & I915_DISPATCH_RS ?
 			 MI_BATCH_RESOURCE_STREAMER : 0));
 	/* bit0-7 is the length on GEN6+ */
-	intel_ring_emit(ring, offset);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, offset);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -2677,20 +2696,20 @@ gen6_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
 			      u64 offset, u32 len,
 			      unsigned dispatch_flags)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(ring,
+	intel_ring_emit(engine,
 			MI_BATCH_BUFFER_START |
 			(dispatch_flags & I915_DISPATCH_SECURE ?
 			 0 : MI_BATCH_NON_SECURE_I965));
 	/* bit0-7 is the length on GEN6+ */
-	intel_ring_emit(ring, offset);
-	intel_ring_advance(ring);
+	intel_ring_emit(engine, offset);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -2700,8 +2719,8 @@ gen6_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
 static int gen6_ring_flush(struct drm_i915_gem_request *req,
 			   u32 invalidate, u32 flush)
 {
-	struct intel_engine_cs *ring = req->ring;
-	struct drm_device *dev = ring->dev;
+	struct intel_engine_cs *engine = req->engine;
+	struct drm_device *dev = engine->dev;
 	uint32_t cmd;
 	int ret;
 
@@ -2728,16 +2747,17 @@ static int gen6_ring_flush(struct drm_i915_gem_request *req,
 	 */
 	if (invalidate & I915_GEM_DOMAIN_RENDER)
 		cmd |= MI_INVALIDATE_TLB;
-	intel_ring_emit(ring, cmd);
-	intel_ring_emit(ring, I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
+	intel_ring_emit(engine, cmd);
+	intel_ring_emit(engine,
+			I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
 	if (INTEL_INFO(dev)->gen >= 8) {
-		intel_ring_emit(ring, 0); /* upper addr */
-		intel_ring_emit(ring, 0); /* value */
+		intel_ring_emit(engine, 0); /* upper addr */
+		intel_ring_emit(engine, 0); /* value */
 	} else  {
-		intel_ring_emit(ring, 0);
-		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(engine, 0);
+		intel_ring_emit(engine, MI_NOOP);
 	}
-	intel_ring_advance(ring);
+	intel_ring_advance(engine);
 
 	return 0;
 }
@@ -2745,14 +2765,14 @@ static int gen6_ring_flush(struct drm_i915_gem_request *req,
 int intel_init_render_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	struct drm_i915_gem_object *obj;
 	int ret;
 
-	ring->name = "render ring";
-	ring->id = RCS;
-	ring->exec_id = I915_EXEC_RENDER;
-	ring->mmio_base = RENDER_RING_BASE;
+	engine->name = "render ring";
+	engine->id = RCS;
+	engine->exec_id = I915_EXEC_RENDER;
+	engine->mmio_base = RENDER_RING_BASE;
 
 	if (INTEL_INFO(dev)->gen >= 8) {
 		if (i915_semaphore_is_enabled(dev)) {
@@ -2772,34 +2792,36 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 			}
 		}
 
-		ring->init_context = intel_rcs_ctx_init;
-		ring->add_request = gen6_add_request;
-		ring->flush = gen8_render_ring_flush;
-		ring->irq_get = gen8_ring_get_irq;
-		ring->irq_put = gen8_ring_put_irq;
-		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
-		ring->get_seqno = gen6_ring_get_seqno;
-		ring->set_seqno = ring_set_seqno;
+		engine->init_context = intel_rcs_ctx_init;
+		engine->add_request = gen6_add_request;
+		engine->flush = gen8_render_ring_flush;
+		engine->irq_get = gen8_ring_get_irq;
+		engine->irq_put = gen8_ring_put_irq;
+		engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
+		engine->irq_seqno_barrier = gen6_seqno_barrier;
+		engine->get_seqno = ring_get_seqno;
+		engine->set_seqno = ring_set_seqno;
 		if (i915_semaphore_is_enabled(dev)) {
 			WARN_ON(!dev_priv->semaphore_obj);
-			ring->semaphore.sync_to = gen8_ring_sync;
-			ring->semaphore.signal = gen8_rcs_signal;
-			GEN8_RING_SEMAPHORE_INIT;
+			engine->semaphore.sync_to = gen8_ring_sync;
+			engine->semaphore.signal = gen8_rcs_signal;
+			GEN8_RING_SEMAPHORE_INIT(engine);
 		}
 	} else if (INTEL_INFO(dev)->gen >= 6) {
-		ring->init_context = intel_rcs_ctx_init;
-		ring->add_request = gen6_add_request;
-		ring->flush = gen7_render_ring_flush;
+		engine->init_context = intel_rcs_ctx_init;
+		engine->add_request = gen6_add_request;
+		engine->flush = gen7_render_ring_flush;
 		if (INTEL_INFO(dev)->gen == 6)
-			ring->flush = gen6_render_ring_flush;
-		ring->irq_get = gen6_ring_get_irq;
-		ring->irq_put = gen6_ring_put_irq;
-		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
-		ring->get_seqno = gen6_ring_get_seqno;
-		ring->set_seqno = ring_set_seqno;
+			engine->flush = gen6_render_ring_flush;
+		engine->irq_get = gen6_ring_get_irq;
+		engine->irq_put = gen6_ring_put_irq;
+		engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
+		engine->irq_seqno_barrier = gen6_seqno_barrier;
+		engine->get_seqno = ring_get_seqno;
+		engine->set_seqno = ring_set_seqno;
 		if (i915_semaphore_is_enabled(dev)) {
-			ring->semaphore.sync_to = gen6_ring_sync;
-			ring->semaphore.signal = gen6_signal;
+			engine->semaphore.sync_to = gen6_ring_sync;
+			engine->semaphore.signal = gen6_signal;
 			/*
 			 * The current semaphore is only applied on pre-gen8
 			 * platform.  And there is no VCS2 ring on the pre-gen8
@@ -2807,59 +2829,59 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 			 * initialized as INVALID.  Gen8 will initialize the
 			 * sema between VCS2 and RCS later.
 			 */
-			ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
-			ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_RV;
-			ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_RB;
-			ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_RVE;
-			ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-			ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
-			ring->semaphore.mbox.signal[VCS] = GEN6_VRSYNC;
-			ring->semaphore.mbox.signal[BCS] = GEN6_BRSYNC;
-			ring->semaphore.mbox.signal[VECS] = GEN6_VERSYNC;
-			ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+			engine->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+			engine->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_RV;
+			engine->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_RB;
+			engine->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_RVE;
+			engine->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+			engine->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
+			engine->semaphore.mbox.signal[VCS] = GEN6_VRSYNC;
+			engine->semaphore.mbox.signal[BCS] = GEN6_BRSYNC;
+			engine->semaphore.mbox.signal[VECS] = GEN6_VERSYNC;
+			engine->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 		}
 	} else if (IS_GEN5(dev)) {
-		ring->add_request = pc_render_add_request;
-		ring->flush = gen4_render_ring_flush;
-		ring->get_seqno = pc_render_get_seqno;
-		ring->set_seqno = pc_render_set_seqno;
-		ring->irq_get = gen5_ring_get_irq;
-		ring->irq_put = gen5_ring_put_irq;
-		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT |
+		engine->add_request = pc_render_add_request;
+		engine->flush = gen4_render_ring_flush;
+		engine->get_seqno = pc_render_get_seqno;
+		engine->set_seqno = pc_render_set_seqno;
+		engine->irq_get = gen5_ring_get_irq;
+		engine->irq_put = gen5_ring_put_irq;
+		engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT |
 					GT_RENDER_PIPECTL_NOTIFY_INTERRUPT;
 	} else {
-		ring->add_request = i9xx_add_request;
+		engine->add_request = i9xx_add_request;
 		if (INTEL_INFO(dev)->gen < 4)
-			ring->flush = gen2_render_ring_flush;
+			engine->flush = gen2_render_ring_flush;
 		else
-			ring->flush = gen4_render_ring_flush;
-		ring->get_seqno = ring_get_seqno;
-		ring->set_seqno = ring_set_seqno;
+			engine->flush = gen4_render_ring_flush;
+		engine->get_seqno = ring_get_seqno;
+		engine->set_seqno = ring_set_seqno;
 		if (IS_GEN2(dev)) {
-			ring->irq_get = i8xx_ring_get_irq;
-			ring->irq_put = i8xx_ring_put_irq;
+			engine->irq_get = i8xx_ring_get_irq;
+			engine->irq_put = i8xx_ring_put_irq;
 		} else {
-			ring->irq_get = i9xx_ring_get_irq;
-			ring->irq_put = i9xx_ring_put_irq;
+			engine->irq_get = i9xx_ring_get_irq;
+			engine->irq_put = i9xx_ring_put_irq;
 		}
-		ring->irq_enable_mask = I915_USER_INTERRUPT;
+		engine->irq_enable_mask = I915_USER_INTERRUPT;
 	}
-	ring->write_tail = ring_write_tail;
+	engine->write_tail = ring_write_tail;
 
 	if (IS_HASWELL(dev))
-		ring->dispatch_execbuffer = hsw_ring_dispatch_execbuffer;
+		engine->dispatch_execbuffer = hsw_ring_dispatch_execbuffer;
 	else if (IS_GEN8(dev))
-		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		engine->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
 	else if (INTEL_INFO(dev)->gen >= 6)
-		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		engine->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 	else if (INTEL_INFO(dev)->gen >= 4)
-		ring->dispatch_execbuffer = i965_dispatch_execbuffer;
+		engine->dispatch_execbuffer = i965_dispatch_execbuffer;
 	else if (IS_I830(dev) || IS_845G(dev))
-		ring->dispatch_execbuffer = i830_dispatch_execbuffer;
+		engine->dispatch_execbuffer = i830_dispatch_execbuffer;
 	else
-		ring->dispatch_execbuffer = i915_dispatch_execbuffer;
-	ring->init_hw = init_render_ring;
-	ring->cleanup = render_ring_cleanup;
+		engine->dispatch_execbuffer = i915_dispatch_execbuffer;
+	engine->init_hw = init_render_ring;
+	engine->cleanup = render_ring_cleanup;
 
 	/* Workaround batchbuffer to combat CS tlb bug. */
 	if (HAS_BROKEN_CS_TLB(dev)) {
@@ -2876,16 +2898,16 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 			return ret;
 		}
 
-		ring->scratch.obj = obj;
-		ring->scratch.gtt_offset = i915_gem_obj_ggtt_offset(obj);
+		engine->scratch.obj = obj;
+		engine->scratch.gtt_offset = i915_gem_obj_ggtt_offset(obj);
 	}
 
-	ret = intel_init_ring_buffer(dev, ring);
+	ret = intel_init_ring_buffer(dev, engine);
 	if (ret)
 		return ret;
 
 	if (INTEL_INFO(dev)->gen >= 5) {
-		ret = intel_init_pipe_control(ring);
+		ret = intel_init_pipe_control(engine);
 		if (ret)
 			return ret;
 	}
@@ -2896,75 +2918,76 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 int intel_init_bsd_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[VCS];
+	struct intel_engine_cs *engine = &dev_priv->engine[VCS];
 
-	ring->name = "bsd ring";
-	ring->id = VCS;
-	ring->exec_id = I915_EXEC_BSD;
+	engine->name = "bsd ring";
+	engine->id = VCS;
+	engine->exec_id = I915_EXEC_BSD;
 
-	ring->write_tail = ring_write_tail;
+	engine->write_tail = ring_write_tail;
 	if (INTEL_INFO(dev)->gen >= 6) {
-		ring->mmio_base = GEN6_BSD_RING_BASE;
+		engine->mmio_base = GEN6_BSD_RING_BASE;
 		/* gen6 bsd needs a special wa for tail updates */
 		if (IS_GEN6(dev))
-			ring->write_tail = gen6_bsd_ring_write_tail;
-		ring->flush = gen6_bsd_ring_flush;
-		ring->add_request = gen6_add_request;
-		ring->get_seqno = gen6_ring_get_seqno;
-		ring->set_seqno = ring_set_seqno;
+			engine->write_tail = gen6_bsd_ring_write_tail;
+		engine->flush = gen6_bsd_ring_flush;
+		engine->add_request = gen6_add_request;
+		engine->irq_seqno_barrier = gen6_seqno_barrier;
+		engine->get_seqno = ring_get_seqno;
+		engine->set_seqno = ring_set_seqno;
 		if (INTEL_INFO(dev)->gen >= 8) {
-			ring->irq_enable_mask =
+			engine->irq_enable_mask =
 				GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT;
-			ring->irq_get = gen8_ring_get_irq;
-			ring->irq_put = gen8_ring_put_irq;
-			ring->dispatch_execbuffer =
+			engine->irq_get = gen8_ring_get_irq;
+			engine->irq_put = gen8_ring_put_irq;
+			engine->dispatch_execbuffer =
 				gen8_ring_dispatch_execbuffer;
 			if (i915_semaphore_is_enabled(dev)) {
-				ring->semaphore.sync_to = gen8_ring_sync;
-				ring->semaphore.signal = gen8_xcs_signal;
-				GEN8_RING_SEMAPHORE_INIT;
+				engine->semaphore.sync_to = gen8_ring_sync;
+				engine->semaphore.signal = gen8_xcs_signal;
+				GEN8_RING_SEMAPHORE_INIT(engine);
 			}
 		} else {
-			ring->irq_enable_mask = GT_BSD_USER_INTERRUPT;
-			ring->irq_get = gen6_ring_get_irq;
-			ring->irq_put = gen6_ring_put_irq;
-			ring->dispatch_execbuffer =
+			engine->irq_enable_mask = GT_BSD_USER_INTERRUPT;
+			engine->irq_get = gen6_ring_get_irq;
+			engine->irq_put = gen6_ring_put_irq;
+			engine->dispatch_execbuffer =
 				gen6_ring_dispatch_execbuffer;
 			if (i915_semaphore_is_enabled(dev)) {
-				ring->semaphore.sync_to = gen6_ring_sync;
-				ring->semaphore.signal = gen6_signal;
-				ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VR;
-				ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
-				ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VB;
-				ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_VVE;
-				ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-				ring->semaphore.mbox.signal[RCS] = GEN6_RVSYNC;
-				ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
-				ring->semaphore.mbox.signal[BCS] = GEN6_BVSYNC;
-				ring->semaphore.mbox.signal[VECS] = GEN6_VEVSYNC;
-				ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+				engine->semaphore.sync_to = gen6_ring_sync;
+				engine->semaphore.signal = gen6_signal;
+				engine->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VR;
+				engine->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
+				engine->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VB;
+				engine->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_VVE;
+				engine->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+				engine->semaphore.mbox.signal[RCS] = GEN6_RVSYNC;
+				engine->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
+				engine->semaphore.mbox.signal[BCS] = GEN6_BVSYNC;
+				engine->semaphore.mbox.signal[VECS] = GEN6_VEVSYNC;
+				engine->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 			}
 		}
 	} else {
-		ring->mmio_base = BSD_RING_BASE;
-		ring->flush = bsd_ring_flush;
-		ring->add_request = i9xx_add_request;
-		ring->get_seqno = ring_get_seqno;
-		ring->set_seqno = ring_set_seqno;
+		engine->mmio_base = BSD_RING_BASE;
+		engine->flush = bsd_ring_flush;
+		engine->add_request = i9xx_add_request;
+		engine->get_seqno = ring_get_seqno;
+		engine->set_seqno = ring_set_seqno;
 		if (IS_GEN5(dev)) {
-			ring->irq_enable_mask = ILK_BSD_USER_INTERRUPT;
-			ring->irq_get = gen5_ring_get_irq;
-			ring->irq_put = gen5_ring_put_irq;
+			engine->irq_enable_mask = ILK_BSD_USER_INTERRUPT;
+			engine->irq_get = gen5_ring_get_irq;
+			engine->irq_put = gen5_ring_put_irq;
 		} else {
-			ring->irq_enable_mask = I915_BSD_USER_INTERRUPT;
-			ring->irq_get = i9xx_ring_get_irq;
-			ring->irq_put = i9xx_ring_put_irq;
+			engine->irq_enable_mask = I915_BSD_USER_INTERRUPT;
+			engine->irq_get = i9xx_ring_get_irq;
+			engine->irq_put = i9xx_ring_put_irq;
 		}
-		ring->dispatch_execbuffer = i965_dispatch_execbuffer;
+		engine->dispatch_execbuffer = i965_dispatch_execbuffer;
 	}
-	ring->init_hw = init_ring_common;
+	engine->init_hw = init_ring_common;
 
-	return intel_init_ring_buffer(dev, ring);
+	return intel_init_ring_buffer(dev, engine);
 }
 
 /**
@@ -2973,68 +2996,70 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 int intel_init_bsd2_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[VCS2];
-
-	ring->name = "bsd2 ring";
-	ring->id = VCS2;
-	ring->exec_id = I915_EXEC_BSD;
-
-	ring->write_tail = ring_write_tail;
-	ring->mmio_base = GEN8_BSD2_RING_BASE;
-	ring->flush = gen6_bsd_ring_flush;
-	ring->add_request = gen6_add_request;
-	ring->get_seqno = gen6_ring_get_seqno;
-	ring->set_seqno = ring_set_seqno;
-	ring->irq_enable_mask =
+	struct intel_engine_cs *engine = &dev_priv->engine[VCS2];
+
+	engine->name = "bsd2 ring";
+	engine->id = VCS2;
+	engine->exec_id = I915_EXEC_BSD;
+
+	engine->write_tail = ring_write_tail;
+	engine->mmio_base = GEN8_BSD2_RING_BASE;
+	engine->flush = gen6_bsd_ring_flush;
+	engine->add_request = gen6_add_request;
+	engine->irq_seqno_barrier = gen6_seqno_barrier;
+	engine->get_seqno = ring_get_seqno;
+	engine->set_seqno = ring_set_seqno;
+	engine->irq_enable_mask =
 			GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT;
-	ring->irq_get = gen8_ring_get_irq;
-	ring->irq_put = gen8_ring_put_irq;
-	ring->dispatch_execbuffer =
+	engine->irq_get = gen8_ring_get_irq;
+	engine->irq_put = gen8_ring_put_irq;
+	engine->dispatch_execbuffer =
 			gen8_ring_dispatch_execbuffer;
 	if (i915_semaphore_is_enabled(dev)) {
-		ring->semaphore.sync_to = gen8_ring_sync;
-		ring->semaphore.signal = gen8_xcs_signal;
-		GEN8_RING_SEMAPHORE_INIT;
+		engine->semaphore.sync_to = gen8_ring_sync;
+		engine->semaphore.signal = gen8_xcs_signal;
+		GEN8_RING_SEMAPHORE_INIT(engine);
 	}
-	ring->init_hw = init_ring_common;
+	engine->init_hw = init_ring_common;
 
-	return intel_init_ring_buffer(dev, ring);
+	return intel_init_ring_buffer(dev, engine);
 }
 
 int intel_init_blt_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[BCS];
-
-	ring->name = "blitter ring";
-	ring->id = BCS;
-	ring->exec_id = I915_EXEC_BLT;
-
-	ring->mmio_base = BLT_RING_BASE;
-	ring->write_tail = ring_write_tail;
-	ring->flush = gen6_ring_flush;
-	ring->add_request = gen6_add_request;
-	ring->get_seqno = gen6_ring_get_seqno;
-	ring->set_seqno = ring_set_seqno;
+	struct intel_engine_cs *engine = &dev_priv->engine[BCS];
+
+	engine->name = "blitter ring";
+	engine->id = BCS;
+	engine->exec_id = I915_EXEC_BLT;
+
+	engine->mmio_base = BLT_RING_BASE;
+	engine->write_tail = ring_write_tail;
+	engine->flush = gen6_ring_flush;
+	engine->add_request = gen6_add_request;
+	engine->irq_seqno_barrier = gen6_seqno_barrier;
+	engine->get_seqno = ring_get_seqno;
+	engine->set_seqno = ring_set_seqno;
 	if (INTEL_INFO(dev)->gen >= 8) {
-		ring->irq_enable_mask =
+		engine->irq_enable_mask =
 			GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT;
-		ring->irq_get = gen8_ring_get_irq;
-		ring->irq_put = gen8_ring_put_irq;
-		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		engine->irq_get = gen8_ring_get_irq;
+		engine->irq_put = gen8_ring_put_irq;
+		engine->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
 		if (i915_semaphore_is_enabled(dev)) {
-			ring->semaphore.sync_to = gen8_ring_sync;
-			ring->semaphore.signal = gen8_xcs_signal;
-			GEN8_RING_SEMAPHORE_INIT;
+			engine->semaphore.sync_to = gen8_ring_sync;
+			engine->semaphore.signal = gen8_xcs_signal;
+			GEN8_RING_SEMAPHORE_INIT(engine);
 		}
 	} else {
-		ring->irq_enable_mask = GT_BLT_USER_INTERRUPT;
-		ring->irq_get = gen6_ring_get_irq;
-		ring->irq_put = gen6_ring_put_irq;
-		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		engine->irq_enable_mask = GT_BLT_USER_INTERRUPT;
+		engine->irq_get = gen6_ring_get_irq;
+		engine->irq_put = gen6_ring_put_irq;
+		engine->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 		if (i915_semaphore_is_enabled(dev)) {
-			ring->semaphore.signal = gen6_signal;
-			ring->semaphore.sync_to = gen6_ring_sync;
+			engine->semaphore.signal = gen6_signal;
+			engine->semaphore.sync_to = gen6_ring_sync;
 			/*
 			 * The current semaphore is only applied on pre-gen8
 			 * platform.  And there is no VCS2 ring on the pre-gen8
@@ -3042,127 +3067,128 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 			 * initialized as INVALID.  Gen8 will initialize the
 			 * sema between BCS and VCS2 later.
 			 */
-			ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_BR;
-			ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_BV;
-			ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
-			ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_BVE;
-			ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-			ring->semaphore.mbox.signal[RCS] = GEN6_RBSYNC;
-			ring->semaphore.mbox.signal[VCS] = GEN6_VBSYNC;
-			ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
-			ring->semaphore.mbox.signal[VECS] = GEN6_VEBSYNC;
-			ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+			engine->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_BR;
+			engine->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_BV;
+			engine->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
+			engine->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_BVE;
+			engine->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+			engine->semaphore.mbox.signal[RCS] = GEN6_RBSYNC;
+			engine->semaphore.mbox.signal[VCS] = GEN6_VBSYNC;
+			engine->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
+			engine->semaphore.mbox.signal[VECS] = GEN6_VEBSYNC;
+			engine->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 		}
 	}
-	ring->init_hw = init_ring_common;
+	engine->init_hw = init_ring_common;
 
-	return intel_init_ring_buffer(dev, ring);
+	return intel_init_ring_buffer(dev, engine);
 }
 
 int intel_init_vebox_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_engine_cs *ring = &dev_priv->ring[VECS];
+	struct intel_engine_cs *engine = &dev_priv->engine[VECS];
 
-	ring->name = "video enhancement ring";
-	ring->id = VECS;
-	ring->exec_id = I915_EXEC_VEBOX;
+	engine->name = "video enhancement ring";
+	engine->id = VECS;
+	engine->exec_id = I915_EXEC_VEBOX;
 
-	ring->mmio_base = VEBOX_RING_BASE;
-	ring->write_tail = ring_write_tail;
-	ring->flush = gen6_ring_flush;
-	ring->add_request = gen6_add_request;
-	ring->get_seqno = gen6_ring_get_seqno;
-	ring->set_seqno = ring_set_seqno;
+	engine->mmio_base = VEBOX_RING_BASE;
+	engine->write_tail = ring_write_tail;
+	engine->flush = gen6_ring_flush;
+	engine->add_request = gen6_add_request;
+	engine->irq_seqno_barrier = gen6_seqno_barrier;
+	engine->get_seqno = ring_get_seqno;
+	engine->set_seqno = ring_set_seqno;
 
 	if (INTEL_INFO(dev)->gen >= 8) {
-		ring->irq_enable_mask =
+		engine->irq_enable_mask =
 			GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT;
-		ring->irq_get = gen8_ring_get_irq;
-		ring->irq_put = gen8_ring_put_irq;
-		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		engine->irq_get = gen8_ring_get_irq;
+		engine->irq_put = gen8_ring_put_irq;
+		engine->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
 		if (i915_semaphore_is_enabled(dev)) {
-			ring->semaphore.sync_to = gen8_ring_sync;
-			ring->semaphore.signal = gen8_xcs_signal;
-			GEN8_RING_SEMAPHORE_INIT;
+			engine->semaphore.sync_to = gen8_ring_sync;
+			engine->semaphore.signal = gen8_xcs_signal;
+			GEN8_RING_SEMAPHORE_INIT(engine);
 		}
 	} else {
-		ring->irq_enable_mask = PM_VEBOX_USER_INTERRUPT;
-		ring->irq_get = hsw_vebox_get_irq;
-		ring->irq_put = hsw_vebox_put_irq;
-		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		engine->irq_enable_mask = PM_VEBOX_USER_INTERRUPT;
+		engine->irq_get = hsw_vebox_get_irq;
+		engine->irq_put = hsw_vebox_put_irq;
+		engine->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 		if (i915_semaphore_is_enabled(dev)) {
-			ring->semaphore.sync_to = gen6_ring_sync;
-			ring->semaphore.signal = gen6_signal;
-			ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VER;
-			ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_VEV;
-			ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VEB;
-			ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
-			ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-			ring->semaphore.mbox.signal[RCS] = GEN6_RVESYNC;
-			ring->semaphore.mbox.signal[VCS] = GEN6_VVESYNC;
-			ring->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
-			ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
-			ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+			engine->semaphore.sync_to = gen6_ring_sync;
+			engine->semaphore.signal = gen6_signal;
+			engine->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VER;
+			engine->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_VEV;
+			engine->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VEB;
+			engine->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
+			engine->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+			engine->semaphore.mbox.signal[RCS] = GEN6_RVESYNC;
+			engine->semaphore.mbox.signal[VCS] = GEN6_VVESYNC;
+			engine->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
+			engine->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
+			engine->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 		}
 	}
-	ring->init_hw = init_ring_common;
+	engine->init_hw = init_ring_common;
 
-	return intel_init_ring_buffer(dev, ring);
+	return intel_init_ring_buffer(dev, engine);
 }
 
 int
 intel_ring_flush_all_caches(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
-	if (!ring->gpu_caches_dirty)
+	if (!engine->gpu_caches_dirty)
 		return 0;
 
-	ret = ring->flush(req, 0, I915_GEM_GPU_DOMAINS);
+	ret = engine->flush(req, 0, I915_GEM_GPU_DOMAINS);
 	if (ret)
 		return ret;
 
 	trace_i915_gem_ring_flush(req, 0, I915_GEM_GPU_DOMAINS);
 
-	ring->gpu_caches_dirty = false;
+	engine->gpu_caches_dirty = false;
 	return 0;
 }
 
 int
 intel_ring_invalidate_all_caches(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *ring = req->ring;
+	struct intel_engine_cs *engine = req->engine;
 	uint32_t flush_domains;
 	int ret;
 
 	flush_domains = 0;
-	if (ring->gpu_caches_dirty)
+	if (engine->gpu_caches_dirty)
 		flush_domains = I915_GEM_GPU_DOMAINS;
 
-	ret = ring->flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
+	ret = engine->flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
 	if (ret)
 		return ret;
 
 	trace_i915_gem_ring_flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
 
-	ring->gpu_caches_dirty = false;
+	engine->gpu_caches_dirty = false;
 	return 0;
 }
 
 void
-intel_stop_ring_buffer(struct intel_engine_cs *ring)
+intel_stop_engine(struct intel_engine_cs *engine)
 {
 	int ret;
 
-	if (!intel_ring_initialized(ring))
+	if (!intel_engine_initialized(engine))
 		return;
 
-	ret = intel_ring_idle(ring);
-	if (ret && !i915_reset_in_progress(&to_i915(ring->dev)->gpu_error))
+	ret = intel_engine_idle(engine);
+	if (ret)
 		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
-			  ring->name, ret);
+			  engine->name, ret);
 
-	stop_ring(ring);
+	stop_ring(engine);
 }
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 566b0ae10ce0..2ade194bbea9 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -52,34 +52,32 @@ struct  intel_hw_status_page {
 /* seqno size is actually only a uint32, but since we plan to use MI_FLUSH_DW to
  * do the writes, and that must have qw aligned offsets, simply pretend it's 8b.
  */
-#define i915_semaphore_seqno_size sizeof(uint64_t)
+#define gen8_semaphore_seqno_size sizeof(uint64_t)
+#define GEN8_SEMAPHORE_OFFSET(__from, __to)			     \
+	(((__from) * I915_NUM_ENGINES  + (__to)) * gen8_semaphore_seqno_size)
 #define GEN8_SIGNAL_OFFSET(__ring, to)			     \
 	(i915_gem_obj_ggtt_offset(dev_priv->semaphore_obj) + \
-	((__ring)->id * I915_NUM_RINGS * i915_semaphore_seqno_size) +	\
-	(i915_semaphore_seqno_size * (to)))
-
+	 GEN8_SEMAPHORE_OFFSET((__ring)->id, (to)))
 #define GEN8_WAIT_OFFSET(__ring, from)			     \
 	(i915_gem_obj_ggtt_offset(dev_priv->semaphore_obj) + \
-	((from) * I915_NUM_RINGS * i915_semaphore_seqno_size) + \
-	(i915_semaphore_seqno_size * (__ring)->id))
+	 GEN8_SEMAPHORE_OFFSET(from, (__ring)->id))
 
-#define GEN8_RING_SEMAPHORE_INIT do { \
+#define GEN8_RING_SEMAPHORE_INIT(e) do { \
 	if (!dev_priv->semaphore_obj) { \
 		break; \
 	} \
-	ring->semaphore.signal_ggtt[RCS] = GEN8_SIGNAL_OFFSET(ring, RCS); \
-	ring->semaphore.signal_ggtt[VCS] = GEN8_SIGNAL_OFFSET(ring, VCS); \
-	ring->semaphore.signal_ggtt[BCS] = GEN8_SIGNAL_OFFSET(ring, BCS); \
-	ring->semaphore.signal_ggtt[VECS] = GEN8_SIGNAL_OFFSET(ring, VECS); \
-	ring->semaphore.signal_ggtt[VCS2] = GEN8_SIGNAL_OFFSET(ring, VCS2); \
-	ring->semaphore.signal_ggtt[ring->id] = MI_SEMAPHORE_SYNC_INVALID; \
+	(e)->semaphore.signal_ggtt[RCS] = GEN8_SIGNAL_OFFSET((e), RCS); \
+	(e)->semaphore.signal_ggtt[VCS] = GEN8_SIGNAL_OFFSET((e), VCS); \
+	(e)->semaphore.signal_ggtt[BCS] = GEN8_SIGNAL_OFFSET((e), BCS); \
+	(e)->semaphore.signal_ggtt[VECS] = GEN8_SIGNAL_OFFSET((e), VECS); \
+	(e)->semaphore.signal_ggtt[VCS2] = GEN8_SIGNAL_OFFSET((e), VCS2); \
+	(e)->semaphore.signal_ggtt[(e)->id] = MI_SEMAPHORE_SYNC_INVALID; \
 	} while(0)
 
 enum intel_ring_hangcheck_action {
 	HANGCHECK_IDLE = 0,
 	HANGCHECK_WAIT,
 	HANGCHECK_ACTIVE,
-	HANGCHECK_ACTIVE_LOOP,
 	HANGCHECK_KICK,
 	HANGCHECK_HUNG,
 };
@@ -88,8 +86,8 @@ enum intel_ring_hangcheck_action {
 
 struct intel_ring_hangcheck {
 	u64 acthd;
-	u64 max_acthd;
 	u32 seqno;
+	unsigned user_interrupts;
 	int score;
 	enum intel_ring_hangcheck_action action;
 	int deadlock;
@@ -101,7 +99,7 @@ struct intel_ringbuffer {
 	void __iomem *virtual_start;
 	struct i915_vma *vma;
 
-	struct intel_engine_cs *ring;
+	struct intel_engine_cs *engine;
 	struct list_head link;
 
 	u32 head;
@@ -125,7 +123,7 @@ struct intel_ringbuffer {
 };
 
 struct	intel_context;
-struct drm_i915_reg_descriptor;
+struct drm_i915_reg_table;
 
 /*
  * we use a single page to load ctx workarounds so all of these
@@ -148,14 +146,14 @@ struct  i915_ctx_workarounds {
 
 struct  intel_engine_cs {
 	const char	*name;
-	enum intel_ring_id {
+	enum intel_engine_id {
 		RCS = 0,
 		BCS,
 		VCS,
 		VCS2,	/* Keep instances of the same type engine together. */
 		VECS
 	} id;
-#define I915_NUM_RINGS 5
+#define I915_NUM_ENGINES 5
 #define _VCS(n) (VCS + (n))
 	unsigned int exec_id;
 	unsigned int guc_id;
@@ -196,8 +194,8 @@ struct  intel_engine_cs {
 	 * seen value is good enough. Note that the seqno will always be
 	 * monotonic, even if not coherent.
 	 */
-	u32		(*get_seqno)(struct intel_engine_cs *ring,
-				     bool lazy_coherency);
+	void		(*irq_seqno_barrier)(struct intel_engine_cs *ring);
+	u32		(*get_seqno)(struct intel_engine_cs *ring);
 	void		(*set_seqno)(struct intel_engine_cs *ring,
 				     u32 seqno);
 	int		(*dispatch_execbuffer)(struct drm_i915_gem_request *req,
@@ -246,16 +244,16 @@ struct  intel_engine_cs {
 	 *  ie. transpose of f(x, y)
 	 */
 	struct {
-		u32	sync_seqno[I915_NUM_RINGS-1];
+		u32	sync_seqno[I915_NUM_ENGINES-1];
 
 		union {
 			struct {
 				/* our mbox written by others */
-				u32		wait[I915_NUM_RINGS];
+				u32		wait[I915_NUM_ENGINES];
 				/* mboxes this ring signals to */
-				i915_reg_t	signal[I915_NUM_RINGS];
+				i915_reg_t	signal[I915_NUM_ENGINES];
 			} mbox;
-			u64		signal_ggtt[I915_NUM_RINGS];
+			u64		signal_ggtt[I915_NUM_ENGINES];
 		};
 
 		/* AKA wait() */
@@ -268,10 +266,13 @@ struct  intel_engine_cs {
 	} semaphore;
 
 	/* Execlists */
-	spinlock_t execlist_lock;
+	struct tasklet_struct irq_tasklet;
+	spinlock_t execlist_lock; /* used inside tasklet, use spin_lock_bh */
 	struct list_head execlist_queue;
 	struct list_head execlist_retired_req_list;
-	u8 next_context_status_buffer;
+	unsigned int fw_domains;
+	unsigned int next_context_status_buffer;
+	unsigned int idle_lite_restore_wa;
 	bool disable_lite_restore_wa;
 	u32 ctx_desc_template;
 	u32             irq_keep_mask; /* bitmask for interrupts that should not be masked */
@@ -306,6 +307,7 @@ struct  intel_engine_cs {
 	 * inspecting request list.
 	 */
 	u32 last_submitted_seqno;
+	unsigned user_interrupts;
 
 	bool gpu_caches_dirty;
 
@@ -332,15 +334,8 @@ struct  intel_engine_cs {
 	/*
 	 * Table of registers allowed in commands that read/write registers.
 	 */
-	const struct drm_i915_reg_descriptor *reg_table;
-	int reg_count;
-
-	/*
-	 * Table of registers allowed in commands that read/write registers, but
-	 * only from the DRM master.
-	 */
-	const struct drm_i915_reg_descriptor *master_reg_table;
-	int master_reg_count;
+	const struct drm_i915_reg_table *reg_tables;
+	int reg_table_count;
 
 	/*
 	 * Returns the bitmask for the length field of the specified command.
@@ -356,19 +351,19 @@ struct  intel_engine_cs {
 };
 
 static inline bool
-intel_ring_initialized(struct intel_engine_cs *ring)
+intel_engine_initialized(struct intel_engine_cs *engine)
 {
-	return ring->dev != NULL;
+	return engine->dev != NULL;
 }
 
 static inline unsigned
-intel_ring_flag(struct intel_engine_cs *ring)
+intel_engine_flag(struct intel_engine_cs *engine)
 {
-	return 1 << ring->id;
+	return 1 << engine->id;
 }
 
 static inline u32
-intel_ring_sync_index(struct intel_engine_cs *ring,
+intel_ring_sync_index(struct intel_engine_cs *engine,
 		      struct intel_engine_cs *other)
 {
 	int idx;
@@ -381,34 +376,33 @@ intel_ring_sync_index(struct intel_engine_cs *ring,
 	 * vcs2 -> 0 = rcs, 1 = vcs, 2 = bcs, 3 = vecs;
 	 */
 
-	idx = (other - ring) - 1;
+	idx = (other - engine) - 1;
 	if (idx < 0)
-		idx += I915_NUM_RINGS;
+		idx += I915_NUM_ENGINES;
 
 	return idx;
 }
 
 static inline void
-intel_flush_status_page(struct intel_engine_cs *ring, int reg)
+intel_flush_status_page(struct intel_engine_cs *engine, int reg)
 {
-	drm_clflush_virt_range(&ring->status_page.page_addr[reg],
-			       sizeof(uint32_t));
+	mb();
+	clflush(&engine->status_page.page_addr[reg]);
+	mb();
 }
 
 static inline u32
-intel_read_status_page(struct intel_engine_cs *ring,
-		       int reg)
+intel_read_status_page(struct intel_engine_cs *engine, int reg)
 {
 	/* Ensure that the compiler doesn't optimize away the load. */
-	barrier();
-	return ring->status_page.page_addr[reg];
+	return READ_ONCE(engine->status_page.page_addr[reg]);
 }
 
 static inline void
-intel_write_status_page(struct intel_engine_cs *ring,
+intel_write_status_page(struct intel_engine_cs *engine,
 			int reg, u32 value)
 {
-	ring->status_page.page_addr[reg] = value;
+	engine->status_page.page_addr[reg] = value;
 }
 
 /*
@@ -439,42 +433,42 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf);
 void intel_ringbuffer_free(struct intel_ringbuffer *ring);
 
-void intel_stop_ring_buffer(struct intel_engine_cs *ring);
-void intel_cleanup_ring_buffer(struct intel_engine_cs *ring);
+void intel_stop_engine(struct intel_engine_cs *engine);
+void intel_cleanup_engine(struct intel_engine_cs *engine);
 
 int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request);
 
 int __must_check intel_ring_begin(struct drm_i915_gem_request *req, int n);
 int __must_check intel_ring_cacheline_align(struct drm_i915_gem_request *req);
-static inline void intel_ring_emit(struct intel_engine_cs *ring,
+static inline void intel_ring_emit(struct intel_engine_cs *engine,
 				   u32 data)
 {
-	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct intel_ringbuffer *ringbuf = engine->buffer;
 	iowrite32(data, ringbuf->virtual_start + ringbuf->tail);
 	ringbuf->tail += 4;
 }
-static inline void intel_ring_emit_reg(struct intel_engine_cs *ring,
+static inline void intel_ring_emit_reg(struct intel_engine_cs *engine,
 				       i915_reg_t reg)
 {
-	intel_ring_emit(ring, i915_mmio_reg_offset(reg));
+	intel_ring_emit(engine, i915_mmio_reg_offset(reg));
 }
-static inline void intel_ring_advance(struct intel_engine_cs *ring)
+static inline void intel_ring_advance(struct intel_engine_cs *engine)
 {
-	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct intel_ringbuffer *ringbuf = engine->buffer;
 	ringbuf->tail &= ringbuf->size - 1;
 }
 int __intel_ring_space(int head, int tail, int size);
 void intel_ring_update_space(struct intel_ringbuffer *ringbuf);
 int intel_ring_space(struct intel_ringbuffer *ringbuf);
-bool intel_ring_stopped(struct intel_engine_cs *ring);
+bool intel_engine_stopped(struct intel_engine_cs *engine);
 
-int __must_check intel_ring_idle(struct intel_engine_cs *ring);
-void intel_ring_init_seqno(struct intel_engine_cs *ring, u32 seqno);
+int __must_check intel_engine_idle(struct intel_engine_cs *engine);
+void intel_ring_init_seqno(struct intel_engine_cs *engine, u32 seqno);
 int intel_ring_flush_all_caches(struct drm_i915_gem_request *req);
 int intel_ring_invalidate_all_caches(struct drm_i915_gem_request *req);
 
-void intel_fini_pipe_control(struct intel_engine_cs *ring);
-int intel_init_pipe_control(struct intel_engine_cs *ring);
+void intel_fini_pipe_control(struct intel_engine_cs *engine);
+int intel_init_pipe_control(struct intel_engine_cs *engine);
 
 int intel_init_render_ring_buffer(struct drm_device *dev);
 int intel_init_bsd_ring_buffer(struct drm_device *dev);
@@ -482,9 +476,9 @@ int intel_init_bsd2_ring_buffer(struct drm_device *dev);
 int intel_init_blt_ring_buffer(struct drm_device *dev);
 int intel_init_vebox_ring_buffer(struct drm_device *dev);
 
-u64 intel_ring_get_active_head(struct intel_engine_cs *ring);
+u64 intel_ring_get_active_head(struct intel_engine_cs *engine);
 
-int init_workarounds_ring(struct intel_engine_cs *ring);
+int init_workarounds_ring(struct intel_engine_cs *engine);
 
 static inline u32 intel_ring_get_tail(struct intel_ringbuffer *ringbuf)
 {
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 6e54d978d9d4..7fb1da4e7fc3 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -89,6 +89,10 @@ intel_display_power_domain_str(enum intel_display_power_domain domain)
 		return "TRANSCODER_C";
 	case POWER_DOMAIN_TRANSCODER_EDP:
 		return "TRANSCODER_EDP";
+	case POWER_DOMAIN_TRANSCODER_DSI_A:
+		return "TRANSCODER_DSI_A";
+	case POWER_DOMAIN_TRANSCODER_DSI_C:
+		return "TRANSCODER_DSI_C";
 	case POWER_DOMAIN_PORT_DDI_A_LANES:
 		return "PORT_DDI_A_LANES";
 	case POWER_DOMAIN_PORT_DDI_B_LANES:
@@ -393,11 +397,6 @@ static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 	BIT(POWER_DOMAIN_MODESET) |			\
 	BIT(POWER_DOMAIN_AUX_A) |			\
 	BIT(POWER_DOMAIN_INIT))
-#define SKL_DISPLAY_ALWAYS_ON_POWER_DOMAINS (		\
-	(POWER_DOMAIN_MASK & ~(				\
-	SKL_DISPLAY_POWERWELL_2_POWER_DOMAINS |		\
-	SKL_DISPLAY_DC_OFF_POWER_DOMAINS)) |		\
-	BIT(POWER_DOMAIN_INIT))
 
 #define BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS (		\
 	BIT(POWER_DOMAIN_TRANSCODER_A) |		\
@@ -415,36 +414,21 @@ static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 	BIT(POWER_DOMAIN_VGA) |				\
 	BIT(POWER_DOMAIN_GMBUS) |			\
 	BIT(POWER_DOMAIN_INIT))
-#define BXT_DISPLAY_POWERWELL_1_POWER_DOMAINS (		\
-	BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS |		\
-	BIT(POWER_DOMAIN_PIPE_A) |			\
-	BIT(POWER_DOMAIN_TRANSCODER_EDP) |		\
-	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER) |		\
-	BIT(POWER_DOMAIN_PORT_DDI_A_LANES) |		\
-	BIT(POWER_DOMAIN_AUX_A) |			\
-	BIT(POWER_DOMAIN_PLLS) |			\
-	BIT(POWER_DOMAIN_INIT))
 #define BXT_DISPLAY_DC_OFF_POWER_DOMAINS (		\
 	BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS |		\
 	BIT(POWER_DOMAIN_MODESET) |			\
 	BIT(POWER_DOMAIN_AUX_A) |			\
 	BIT(POWER_DOMAIN_INIT))
-#define BXT_DISPLAY_ALWAYS_ON_POWER_DOMAINS (		\
-	(POWER_DOMAIN_MASK & ~(BXT_DISPLAY_POWERWELL_1_POWER_DOMAINS |	\
-	BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS)) |	\
-	BIT(POWER_DOMAIN_INIT))
 
 static void assert_can_enable_dc9(struct drm_i915_private *dev_priv)
 {
-	struct drm_device *dev = dev_priv->dev;
-
-	WARN(!IS_BROXTON(dev), "Platform doesn't support DC9.\n");
-	WARN((I915_READ(DC_STATE_EN) & DC_STATE_EN_DC9),
-		"DC9 already programmed to be enabled.\n");
-	WARN(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5,
-		"DC5 still not disabled to enable DC9.\n");
-	WARN(I915_READ(HSW_PWR_WELL_DRIVER), "Power well on.\n");
-	WARN(intel_irqs_enabled(dev_priv), "Interrupts not disabled yet.\n");
+	WARN_ONCE((I915_READ(DC_STATE_EN) & DC_STATE_EN_DC9),
+		  "DC9 already programmed to be enabled.\n");
+	WARN_ONCE(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5,
+		  "DC5 still not disabled to enable DC9.\n");
+	WARN_ONCE(I915_READ(HSW_PWR_WELL_DRIVER), "Power well on.\n");
+	WARN_ONCE(intel_irqs_enabled(dev_priv),
+		  "Interrupts not disabled yet.\n");
 
 	 /*
 	  * TODO: check for the following to verify the conditions to enter DC9
@@ -457,11 +441,10 @@ static void assert_can_enable_dc9(struct drm_i915_private *dev_priv)
 
 static void assert_can_disable_dc9(struct drm_i915_private *dev_priv)
 {
-	WARN(intel_irqs_enabled(dev_priv), "Interrupts not disabled yet.\n");
-	WARN(!(I915_READ(DC_STATE_EN) & DC_STATE_EN_DC9),
-		"DC9 already programmed to be disabled.\n");
-	WARN(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5,
-		"DC5 still not disabled.\n");
+	WARN_ONCE(intel_irqs_enabled(dev_priv),
+		  "Interrupts not disabled yet.\n");
+	WARN_ONCE(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5,
+		  "DC5 still not disabled.\n");
 
 	 /*
 	  * TODO: check for the following to verify DC9 state was indeed
@@ -472,24 +455,6 @@ static void assert_can_disable_dc9(struct drm_i915_private *dev_priv)
 	  */
 }
 
-static void gen9_set_dc_state_debugmask(struct drm_i915_private *dev_priv)
-{
-	uint32_t val, mask;
-
-	mask = DC_STATE_DEBUG_MASK_MEMORY_UP;
-
-	if (IS_BROXTON(dev_priv))
-		mask |= DC_STATE_DEBUG_MASK_CORES;
-
-	/* The below bit doesn't need to be cleared ever afterwards */
-	val = I915_READ(DC_STATE_DEBUG);
-	if ((val & mask) != mask) {
-		val |= mask;
-		I915_WRITE(DC_STATE_DEBUG, val);
-		POSTING_READ(DC_STATE_DEBUG);
-	}
-}
-
 static void gen9_write_dc_state(struct drm_i915_private *dev_priv,
 				u32 state)
 {
@@ -527,10 +492,9 @@ static void gen9_write_dc_state(struct drm_i915_private *dev_priv,
 			      state, rewrites);
 }
 
-static void gen9_set_dc_state(struct drm_i915_private *dev_priv, uint32_t state)
+static u32 gen9_dc_mask(struct drm_i915_private *dev_priv)
 {
-	uint32_t val;
-	uint32_t mask;
+	u32 mask;
 
 	mask = DC_STATE_EN_UPTO_DC5;
 	if (IS_BROXTON(dev_priv))
@@ -538,14 +502,30 @@ static void gen9_set_dc_state(struct drm_i915_private *dev_priv, uint32_t state)
 	else
 		mask |= DC_STATE_EN_UPTO_DC6;
 
-	WARN_ON_ONCE(state & ~mask);
+	return mask;
+}
+
+void gen9_sanitize_dc_state(struct drm_i915_private *dev_priv)
+{
+	u32 val;
 
-	if (i915.enable_dc == 0)
-		state = DC_STATE_DISABLE;
-	else if (i915.enable_dc == 1 && state > DC_STATE_EN_UPTO_DC5)
-		state = DC_STATE_EN_UPTO_DC5;
+	val = I915_READ(DC_STATE_EN) & gen9_dc_mask(dev_priv);
+
+	DRM_DEBUG_KMS("Resetting DC state tracking from %02x to %02x\n",
+		      dev_priv->csr.dc_state, val);
+	dev_priv->csr.dc_state = val;
+}
+
+static void gen9_set_dc_state(struct drm_i915_private *dev_priv, uint32_t state)
+{
+	uint32_t val;
+	uint32_t mask;
+
+	if (WARN_ON_ONCE(state & ~dev_priv->csr.allowed_dc_mask))
+		state &= dev_priv->csr.allowed_dc_mask;
 
 	val = I915_READ(DC_STATE_EN);
+	mask = gen9_dc_mask(dev_priv);
 	DRM_DEBUG_KMS("Setting DC state from %02x to %02x\n",
 		      val & mask, state);
 
@@ -590,13 +570,9 @@ static void assert_csr_loaded(struct drm_i915_private *dev_priv)
 
 static void assert_can_enable_dc5(struct drm_i915_private *dev_priv)
 {
-	struct drm_device *dev = dev_priv->dev;
 	bool pg2_enabled = intel_display_power_well_is_enabled(dev_priv,
 					SKL_DISP_PW_2);
 
-	WARN_ONCE(!IS_SKYLAKE(dev) && !IS_KABYLAKE(dev),
-		  "Platform doesn't support DC5.\n");
-	WARN_ONCE(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
 	WARN_ONCE(pg2_enabled, "PG2 not disabled to enable DC5.\n");
 
 	WARN_ONCE((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5),
@@ -606,19 +582,7 @@ static void assert_can_enable_dc5(struct drm_i915_private *dev_priv)
 	assert_csr_loaded(dev_priv);
 }
 
-static void assert_can_disable_dc5(struct drm_i915_private *dev_priv)
-{
-	/*
-	 * During initialization, the firmware may not be loaded yet.
-	 * We still want to make sure that the DC enabling flag is cleared.
-	 */
-	if (dev_priv->power_domains.initializing)
-		return;
-
-	assert_rpm_wakelock_held(dev_priv);
-}
-
-static void gen9_enable_dc5(struct drm_i915_private *dev_priv)
+void gen9_enable_dc5(struct drm_i915_private *dev_priv)
 {
 	assert_can_enable_dc5(dev_priv);
 
@@ -629,11 +593,6 @@ static void gen9_enable_dc5(struct drm_i915_private *dev_priv)
 
 static void assert_can_enable_dc6(struct drm_i915_private *dev_priv)
 {
-	struct drm_device *dev = dev_priv->dev;
-
-	WARN_ONCE(!IS_SKYLAKE(dev) && !IS_KABYLAKE(dev),
-		  "Platform doesn't support DC6.\n");
-	WARN_ONCE(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
 	WARN_ONCE(I915_READ(UTIL_PIN_CTL) & UTIL_PIN_ENABLE,
 		  "Backlight is not disabled.\n");
 	WARN_ONCE((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
@@ -642,47 +601,60 @@ static void assert_can_enable_dc6(struct drm_i915_private *dev_priv)
 	assert_csr_loaded(dev_priv);
 }
 
-static void assert_can_disable_dc6(struct drm_i915_private *dev_priv)
+void skl_enable_dc6(struct drm_i915_private *dev_priv)
 {
-	/*
-	 * During initialization, the firmware may not be loaded yet.
-	 * We still want to make sure that the DC enabling flag is cleared.
-	 */
-	if (dev_priv->power_domains.initializing)
-		return;
+	assert_can_enable_dc6(dev_priv);
+
+	DRM_DEBUG_KMS("Enabling DC6\n");
+
+	gen9_set_dc_state(dev_priv, DC_STATE_EN_UPTO_DC6);
 
-	WARN_ONCE(!(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
-		  "DC6 already programmed to be disabled.\n");
 }
 
-static void gen9_disable_dc5_dc6(struct drm_i915_private *dev_priv)
+void skl_disable_dc6(struct drm_i915_private *dev_priv)
 {
-	assert_can_disable_dc5(dev_priv);
-
-	if ((IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) &&
-	    i915.enable_dc != 0 && i915.enable_dc != 1)
-		assert_can_disable_dc6(dev_priv);
+	DRM_DEBUG_KMS("Disabling DC6\n");
 
 	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
 }
 
-void skl_enable_dc6(struct drm_i915_private *dev_priv)
+static void
+gen9_sanitize_power_well_requests(struct drm_i915_private *dev_priv,
+				  struct i915_power_well *power_well)
 {
-	assert_can_enable_dc6(dev_priv);
+	enum skl_disp_power_wells power_well_id = power_well->data;
+	u32 val;
+	u32 mask;
 
-	DRM_DEBUG_KMS("Enabling DC6\n");
+	mask = SKL_POWER_WELL_REQ(power_well_id);
 
-	gen9_set_dc_state(dev_priv, DC_STATE_EN_UPTO_DC6);
+	val = I915_READ(HSW_PWR_WELL_KVMR);
+	if (WARN_ONCE(val & mask, "Clearing unexpected KVMR request for %s\n",
+		      power_well->name))
+		I915_WRITE(HSW_PWR_WELL_KVMR, val & ~mask);
 
-}
+	val = I915_READ(HSW_PWR_WELL_BIOS);
+	val |= I915_READ(HSW_PWR_WELL_DEBUG);
 
-void skl_disable_dc6(struct drm_i915_private *dev_priv)
-{
-	assert_can_disable_dc6(dev_priv);
+	if (!(val & mask))
+		return;
 
-	DRM_DEBUG_KMS("Disabling DC6\n");
+	/*
+	 * DMC is known to force on the request bits for power well 1 on SKL
+	 * and BXT and the misc IO power well on SKL but we don't expect any
+	 * other request bits to be set, so WARN for those.
+	 */
+	if (power_well_id == SKL_DISP_PW_1 ||
+	    ((IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) &&
+	     power_well_id == SKL_DISP_PW_MISC_IO))
+		DRM_DEBUG_DRIVER("Clearing auxiliary requests for %s forced on "
+				 "by DMC\n", power_well->name);
+	else
+		WARN_ONCE(1, "Clearing unexpected auxiliary requests for %s\n",
+			  power_well->name);
 
-	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
+	I915_WRITE(HSW_PWR_WELL_BIOS, val & ~mask);
+	I915_WRITE(HSW_PWR_WELL_DEBUG, val & ~mask);
 }
 
 static void skl_set_power_well(struct drm_i915_private *dev_priv,
@@ -739,10 +711,6 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv,
 
 		if (!is_enabled) {
 			DRM_DEBUG_KMS("Enabling %s\n", power_well->name);
-			if (wait_for((I915_READ(HSW_PWR_WELL_DRIVER) &
-				state_mask), 1))
-				DRM_ERROR("%s enable timeout\n",
-					power_well->name);
 			check_fuse_status = true;
 		}
 	} else {
@@ -751,8 +719,16 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv,
 			POSTING_READ(HSW_PWR_WELL_DRIVER);
 			DRM_DEBUG_KMS("Disabling %s\n", power_well->name);
 		}
+
+		if (IS_GEN9(dev_priv))
+			gen9_sanitize_power_well_requests(dev_priv, power_well);
 	}
 
+	if (wait_for(!!(I915_READ(HSW_PWR_WELL_DRIVER) & state_mask) == enable,
+		     1))
+		DRM_ERROR("%s %s timeout\n",
+			  power_well->name, enable ? "enable" : "disable");
+
 	if (check_fuse_status) {
 		if (power_well->data == SKL_DISP_PW_1) {
 			if (wait_for((I915_READ(SKL_FUSE_STATUS) &
@@ -833,32 +809,33 @@ static bool gen9_dc_off_power_well_enabled(struct drm_i915_private *dev_priv,
 static void gen9_dc_off_power_well_enable(struct drm_i915_private *dev_priv,
 					  struct i915_power_well *power_well)
 {
-	gen9_disable_dc5_dc6(dev_priv);
+	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
+
+	if (IS_BROXTON(dev_priv)) {
+		broxton_cdclk_verify_state(dev_priv);
+		broxton_ddi_phy_verify_state(dev_priv);
+	}
 }
 
 static void gen9_dc_off_power_well_disable(struct drm_i915_private *dev_priv,
 					   struct i915_power_well *power_well)
 {
-	if ((IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) &&
-	    i915.enable_dc != 0 && i915.enable_dc != 1)
+	if (!dev_priv->csr.dmc_payload)
+		return;
+
+	if (dev_priv->csr.allowed_dc_mask & DC_STATE_EN_UPTO_DC6)
 		skl_enable_dc6(dev_priv);
-	else
+	else if (dev_priv->csr.allowed_dc_mask & DC_STATE_EN_UPTO_DC5)
 		gen9_enable_dc5(dev_priv);
 }
 
 static void gen9_dc_off_power_well_sync_hw(struct drm_i915_private *dev_priv,
 					   struct i915_power_well *power_well)
 {
-	if (power_well->count > 0) {
-		gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
-	} else {
-		if ((IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) &&
-		    i915.enable_dc != 0 &&
-		    i915.enable_dc != 1)
-			gen9_set_dc_state(dev_priv, DC_STATE_EN_UPTO_DC6);
-		else
-			gen9_set_dc_state(dev_priv, DC_STATE_EN_UPTO_DC5);
-	}
+	if (power_well->count > 0)
+		gen9_dc_off_power_well_enable(dev_priv, power_well);
+	else
+		gen9_dc_off_power_well_disable(dev_priv, power_well);
 }
 
 static void i9xx_always_on_power_well_noop(struct drm_i915_private *dev_priv,
@@ -962,6 +939,17 @@ static bool vlv_power_well_enabled(struct drm_i915_private *dev_priv,
 	return enabled;
 }
 
+static void vlv_init_display_clock_gating(struct drm_i915_private *dev_priv)
+{
+	I915_WRITE(DSPCLK_GATE_D, VRHUNIT_CLOCK_GATE_DISABLE);
+
+	/*
+	 * Disable trickle feed and enable pnd deadline calculation
+	 */
+	I915_WRITE(MI_ARB_VLV, MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE);
+	I915_WRITE(CBR1_VLV, 0);
+}
+
 static void vlv_display_power_well_init(struct drm_i915_private *dev_priv)
 {
 	enum pipe pipe;
@@ -984,6 +972,8 @@ static void vlv_display_power_well_init(struct drm_i915_private *dev_priv)
 		I915_WRITE(DPLL(pipe), val);
 	}
 
+	vlv_init_display_clock_gating(dev_priv);
+
 	spin_lock_irq(&dev_priv->irq_lock);
 	valleyview_enable_display_irqs(dev_priv);
 	spin_unlock_irq(&dev_priv->irq_lock);
@@ -1622,34 +1612,56 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
 	intel_runtime_pm_put(dev_priv);
 }
 
-#define HSW_ALWAYS_ON_POWER_DOMAINS (			\
-	BIT(POWER_DOMAIN_PIPE_A) |			\
-	BIT(POWER_DOMAIN_TRANSCODER_EDP) |		\
-	BIT(POWER_DOMAIN_PORT_DDI_A_LANES) |		\
+#define HSW_DISPLAY_POWER_DOMAINS (			\
+	BIT(POWER_DOMAIN_PIPE_B) |			\
+	BIT(POWER_DOMAIN_PIPE_C) |			\
+	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER) |		\
+	BIT(POWER_DOMAIN_PIPE_B_PANEL_FITTER) |		\
+	BIT(POWER_DOMAIN_PIPE_C_PANEL_FITTER) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_A) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_B) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_C) |		\
 	BIT(POWER_DOMAIN_PORT_DDI_B_LANES) |		\
 	BIT(POWER_DOMAIN_PORT_DDI_C_LANES) |		\
 	BIT(POWER_DOMAIN_PORT_DDI_D_LANES) |		\
-	BIT(POWER_DOMAIN_PORT_CRT) |			\
-	BIT(POWER_DOMAIN_PLLS) |			\
-	BIT(POWER_DOMAIN_AUX_A) |			\
-	BIT(POWER_DOMAIN_AUX_B) |			\
-	BIT(POWER_DOMAIN_AUX_C) |			\
-	BIT(POWER_DOMAIN_AUX_D) |			\
-	BIT(POWER_DOMAIN_GMBUS) |			\
-	BIT(POWER_DOMAIN_INIT))
-#define HSW_DISPLAY_POWER_DOMAINS (				\
-	(POWER_DOMAIN_MASK & ~HSW_ALWAYS_ON_POWER_DOMAINS) |	\
+	BIT(POWER_DOMAIN_PORT_CRT) | /* DDI E */	\
+	BIT(POWER_DOMAIN_VGA) |				\
+	BIT(POWER_DOMAIN_AUDIO) |			\
 	BIT(POWER_DOMAIN_INIT))
 
-#define BDW_ALWAYS_ON_POWER_DOMAINS (			\
-	HSW_ALWAYS_ON_POWER_DOMAINS |			\
-	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER))
-#define BDW_DISPLAY_POWER_DOMAINS (				\
-	(POWER_DOMAIN_MASK & ~BDW_ALWAYS_ON_POWER_DOMAINS) |	\
+#define BDW_DISPLAY_POWER_DOMAINS (			\
+	BIT(POWER_DOMAIN_PIPE_B) |			\
+	BIT(POWER_DOMAIN_PIPE_C) |			\
+	BIT(POWER_DOMAIN_PIPE_B_PANEL_FITTER) |		\
+	BIT(POWER_DOMAIN_PIPE_C_PANEL_FITTER) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_A) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_B) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_C) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_B_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_C_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_D_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_CRT) | /* DDI E */	\
+	BIT(POWER_DOMAIN_VGA) |				\
+	BIT(POWER_DOMAIN_AUDIO) |			\
 	BIT(POWER_DOMAIN_INIT))
 
-#define VLV_ALWAYS_ON_POWER_DOMAINS	BIT(POWER_DOMAIN_INIT)
-#define VLV_DISPLAY_POWER_DOMAINS	POWER_DOMAIN_MASK
+#define VLV_DISPLAY_POWER_DOMAINS (		\
+	BIT(POWER_DOMAIN_PIPE_A) |		\
+	BIT(POWER_DOMAIN_PIPE_B) |		\
+	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER) |	\
+	BIT(POWER_DOMAIN_PIPE_B_PANEL_FITTER) |	\
+	BIT(POWER_DOMAIN_TRANSCODER_A) |	\
+	BIT(POWER_DOMAIN_TRANSCODER_B) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_B_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_C_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DSI) |		\
+	BIT(POWER_DOMAIN_PORT_CRT) |		\
+	BIT(POWER_DOMAIN_VGA) |			\
+	BIT(POWER_DOMAIN_AUDIO) |		\
+	BIT(POWER_DOMAIN_AUX_B) |		\
+	BIT(POWER_DOMAIN_AUX_C) |		\
+	BIT(POWER_DOMAIN_GMBUS) |		\
+	BIT(POWER_DOMAIN_INIT))
 
 #define VLV_DPIO_CMN_BC_POWER_DOMAINS (		\
 	BIT(POWER_DOMAIN_PORT_DDI_B_LANES) |	\
@@ -1679,6 +1691,28 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
 	BIT(POWER_DOMAIN_AUX_C) |		\
 	BIT(POWER_DOMAIN_INIT))
 
+#define CHV_DISPLAY_POWER_DOMAINS (		\
+	BIT(POWER_DOMAIN_PIPE_A) |		\
+	BIT(POWER_DOMAIN_PIPE_B) |		\
+	BIT(POWER_DOMAIN_PIPE_C) |		\
+	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER) |	\
+	BIT(POWER_DOMAIN_PIPE_B_PANEL_FITTER) |	\
+	BIT(POWER_DOMAIN_PIPE_C_PANEL_FITTER) |	\
+	BIT(POWER_DOMAIN_TRANSCODER_A) |	\
+	BIT(POWER_DOMAIN_TRANSCODER_B) |	\
+	BIT(POWER_DOMAIN_TRANSCODER_C) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_B_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_C_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_D_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DSI) |		\
+	BIT(POWER_DOMAIN_VGA) |			\
+	BIT(POWER_DOMAIN_AUDIO) |		\
+	BIT(POWER_DOMAIN_AUX_B) |		\
+	BIT(POWER_DOMAIN_AUX_C) |		\
+	BIT(POWER_DOMAIN_AUX_D) |		\
+	BIT(POWER_DOMAIN_GMBUS) |		\
+	BIT(POWER_DOMAIN_INIT))
+
 #define CHV_DPIO_CMN_BC_POWER_DOMAINS (		\
 	BIT(POWER_DOMAIN_PORT_DDI_B_LANES) |	\
 	BIT(POWER_DOMAIN_PORT_DDI_C_LANES) |	\
@@ -1746,7 +1780,7 @@ static struct i915_power_well hsw_power_wells[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
-		.domains = HSW_ALWAYS_ON_POWER_DOMAINS,
+		.domains = POWER_DOMAIN_MASK,
 		.ops = &i9xx_always_on_power_well_ops,
 	},
 	{
@@ -1760,7 +1794,7 @@ static struct i915_power_well bdw_power_wells[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
-		.domains = BDW_ALWAYS_ON_POWER_DOMAINS,
+		.domains = POWER_DOMAIN_MASK,
 		.ops = &i9xx_always_on_power_well_ops,
 	},
 	{
@@ -1795,7 +1829,7 @@ static struct i915_power_well vlv_power_wells[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
-		.domains = VLV_ALWAYS_ON_POWER_DOMAINS,
+		.domains = POWER_DOMAIN_MASK,
 		.ops = &i9xx_always_on_power_well_ops,
 		.data = PUNIT_POWER_WELL_ALWAYS_ON,
 	},
@@ -1853,7 +1887,7 @@ static struct i915_power_well chv_power_wells[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
-		.domains = VLV_ALWAYS_ON_POWER_DOMAINS,
+		.domains = POWER_DOMAIN_MASK,
 		.ops = &i9xx_always_on_power_well_ops,
 	},
 	{
@@ -1863,7 +1897,7 @@ static struct i915_power_well chv_power_wells[] = {
 		 * power wells don't actually exist. Pipe A power well is
 		 * required for any pipe to work.
 		 */
-		.domains = VLV_DISPLAY_POWER_DOMAINS,
+		.domains = CHV_DISPLAY_POWER_DOMAINS,
 		.data = PIPE_A,
 		.ops = &chv_pipe_power_well_ops,
 	},
@@ -1897,7 +1931,7 @@ static struct i915_power_well skl_power_wells[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
-		.domains = SKL_DISPLAY_ALWAYS_ON_POWER_DOMAINS,
+		.domains = POWER_DOMAIN_MASK,
 		.ops = &i9xx_always_on_power_well_ops,
 		.data = SKL_DISP_PW_ALWAYS_ON,
 	},
@@ -1953,44 +1987,16 @@ static struct i915_power_well skl_power_wells[] = {
 	},
 };
 
-void skl_pw1_misc_io_init(struct drm_i915_private *dev_priv)
-{
-	struct i915_power_well *well;
-
-	if (!(IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)))
-		return;
-
-	well = lookup_power_well(dev_priv, SKL_DISP_PW_1);
-	intel_power_well_enable(dev_priv, well);
-
-	well = lookup_power_well(dev_priv, SKL_DISP_PW_MISC_IO);
-	intel_power_well_enable(dev_priv, well);
-}
-
-void skl_pw1_misc_io_fini(struct drm_i915_private *dev_priv)
-{
-	struct i915_power_well *well;
-
-	if (!(IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)))
-		return;
-
-	well = lookup_power_well(dev_priv, SKL_DISP_PW_1);
-	intel_power_well_disable(dev_priv, well);
-
-	well = lookup_power_well(dev_priv, SKL_DISP_PW_MISC_IO);
-	intel_power_well_disable(dev_priv, well);
-}
-
 static struct i915_power_well bxt_power_wells[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
-		.domains = BXT_DISPLAY_ALWAYS_ON_POWER_DOMAINS,
+		.domains = POWER_DOMAIN_MASK,
 		.ops = &i9xx_always_on_power_well_ops,
 	},
 	{
 		.name = "power well 1",
-		.domains = BXT_DISPLAY_POWERWELL_1_POWER_DOMAINS,
+		.domains = 0,
 		.ops = &skl_power_well_ops,
 		.data = SKL_DISP_PW_1,
 	},
@@ -2015,12 +2021,56 @@ sanitize_disable_power_well_option(const struct drm_i915_private *dev_priv,
 	if (disable_power_well >= 0)
 		return !!disable_power_well;
 
-	if (IS_BROXTON(dev_priv)) {
-		DRM_DEBUG_KMS("Disabling display power well support\n");
-		return 0;
+	return 1;
+}
+
+static uint32_t get_allowed_dc_mask(const struct drm_i915_private *dev_priv,
+				    int enable_dc)
+{
+	uint32_t mask;
+	int requested_dc;
+	int max_dc;
+
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
+		max_dc = 2;
+		mask = 0;
+	} else if (IS_BROXTON(dev_priv)) {
+		max_dc = 1;
+		/*
+		 * DC9 has a separate HW flow from the rest of the DC states,
+		 * not depending on the DMC firmware. It's needed by system
+		 * suspend/resume, so allow it unconditionally.
+		 */
+		mask = DC_STATE_EN_DC9;
+	} else {
+		max_dc = 0;
+		mask = 0;
 	}
 
-	return 1;
+	if (!i915.disable_power_well)
+		max_dc = 0;
+
+	if (enable_dc >= 0 && enable_dc <= max_dc) {
+		requested_dc = enable_dc;
+	} else if (enable_dc == -1) {
+		requested_dc = max_dc;
+	} else if (enable_dc > max_dc && enable_dc <= 2) {
+		DRM_DEBUG_KMS("Adjusting requested max DC state (%d->%d)\n",
+			      enable_dc, max_dc);
+		requested_dc = max_dc;
+	} else {
+		DRM_ERROR("Unexpected value for enable_dc (%d)\n", enable_dc);
+		requested_dc = max_dc;
+	}
+
+	if (requested_dc > 1)
+		mask |= DC_STATE_EN_UPTO_DC6;
+	if (requested_dc > 0)
+		mask |= DC_STATE_EN_UPTO_DC5;
+
+	DRM_DEBUG_KMS("Allowed DC state mask %02x\n", mask);
+
+	return mask;
 }
 
 #define set_power_wells(power_domains, __power_wells) ({		\
@@ -2041,6 +2091,8 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv)
 
 	i915.disable_power_well = sanitize_disable_power_well_option(dev_priv,
 						     i915.disable_power_well);
+	dev_priv->csr.allowed_dc_mask = get_allowed_dc_mask(dev_priv,
+							    i915.enable_dc);
 
 	BUILD_BUG_ON(POWER_DOMAIN_NUM > 31);
 
@@ -2050,17 +2102,17 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv)
 	 * The enabling order will be from lower to higher indexed wells,
 	 * the disabling order is reversed.
 	 */
-	if (IS_HASWELL(dev_priv->dev)) {
+	if (IS_HASWELL(dev_priv)) {
 		set_power_wells(power_domains, hsw_power_wells);
-	} else if (IS_BROADWELL(dev_priv->dev)) {
+	} else if (IS_BROADWELL(dev_priv)) {
 		set_power_wells(power_domains, bdw_power_wells);
-	} else if (IS_SKYLAKE(dev_priv->dev) || IS_KABYLAKE(dev_priv->dev)) {
+	} else if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
 		set_power_wells(power_domains, skl_power_wells);
-	} else if (IS_BROXTON(dev_priv->dev)) {
+	} else if (IS_BROXTON(dev_priv)) {
 		set_power_wells(power_domains, bxt_power_wells);
-	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
+	} else if (IS_CHERRYVIEW(dev_priv)) {
 		set_power_wells(power_domains, chv_power_wells);
-	} else if (IS_VALLEYVIEW(dev_priv->dev)) {
+	} else if (IS_VALLEYVIEW(dev_priv)) {
 		set_power_wells(power_domains, vlv_power_wells);
 	} else {
 		set_power_wells(power_domains, i9xx_always_on_power_well);
@@ -2120,9 +2172,10 @@ static void intel_power_domains_sync_hw(struct drm_i915_private *dev_priv)
 }
 
 static void skl_display_core_init(struct drm_i915_private *dev_priv,
-				  bool resume)
+				   bool resume)
 {
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *well;
 	uint32_t val;
 
 	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
@@ -2133,7 +2186,13 @@ static void skl_display_core_init(struct drm_i915_private *dev_priv,
 
 	/* enable PG1 and Misc I/O */
 	mutex_lock(&power_domains->lock);
-	skl_pw1_misc_io_init(dev_priv);
+
+	well = lookup_power_well(dev_priv, SKL_DISP_PW_1);
+	intel_power_well_enable(dev_priv, well);
+
+	well = lookup_power_well(dev_priv, SKL_DISP_PW_MISC_IO);
+	intel_power_well_enable(dev_priv, well);
+
 	mutex_unlock(&power_domains->lock);
 
 	if (!resume)
@@ -2141,13 +2200,14 @@ static void skl_display_core_init(struct drm_i915_private *dev_priv,
 
 	skl_init_cdclk(dev_priv);
 
-	if (dev_priv->csr.dmc_payload && intel_csr_load_program(dev_priv))
-		gen9_set_dc_state_debugmask(dev_priv);
+	if (dev_priv->csr.dmc_payload)
+		intel_csr_load_program(dev_priv);
 }
 
 static void skl_display_core_uninit(struct drm_i915_private *dev_priv)
 {
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *well;
 
 	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
 
@@ -2155,8 +2215,73 @@ static void skl_display_core_uninit(struct drm_i915_private *dev_priv)
 
 	/* The spec doesn't call for removing the reset handshake flag */
 	/* disable PG1 and Misc I/O */
+
+	mutex_lock(&power_domains->lock);
+
+	well = lookup_power_well(dev_priv, SKL_DISP_PW_MISC_IO);
+	intel_power_well_disable(dev_priv, well);
+
+	well = lookup_power_well(dev_priv, SKL_DISP_PW_1);
+	intel_power_well_disable(dev_priv, well);
+
+	mutex_unlock(&power_domains->lock);
+}
+
+void bxt_display_core_init(struct drm_i915_private *dev_priv,
+			   bool resume)
+{
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *well;
+	uint32_t val;
+
+	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
+
+	/*
+	 * NDE_RSTWRN_OPT RST PCH Handshake En must always be 0b on BXT
+	 * or else the reset will hang because there is no PCH to respond.
+	 * Move the handshake programming to initialization sequence.
+	 * Previously was left up to BIOS.
+	 */
+	val = I915_READ(HSW_NDE_RSTWRN_OPT);
+	val &= ~RESET_PCH_HANDSHAKE_ENABLE;
+	I915_WRITE(HSW_NDE_RSTWRN_OPT, val);
+
+	/* Enable PG1 */
+	mutex_lock(&power_domains->lock);
+
+	well = lookup_power_well(dev_priv, SKL_DISP_PW_1);
+	intel_power_well_enable(dev_priv, well);
+
+	mutex_unlock(&power_domains->lock);
+
+	broxton_init_cdclk(dev_priv);
+	broxton_ddi_phy_init(dev_priv);
+
+	broxton_cdclk_verify_state(dev_priv);
+	broxton_ddi_phy_verify_state(dev_priv);
+
+	if (resume && dev_priv->csr.dmc_payload)
+		intel_csr_load_program(dev_priv);
+}
+
+void bxt_display_core_uninit(struct drm_i915_private *dev_priv)
+{
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *well;
+
+	gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
+
+	broxton_ddi_phy_uninit(dev_priv);
+	broxton_uninit_cdclk(dev_priv);
+
+	/* The spec doesn't call for removing the reset handshake flag */
+
+	/* Disable PG1 */
 	mutex_lock(&power_domains->lock);
-	skl_pw1_misc_io_fini(dev_priv);
+
+	well = lookup_power_well(dev_priv, SKL_DISP_PW_1);
+	intel_power_well_disable(dev_priv, well);
+
 	mutex_unlock(&power_domains->lock);
 }
 
@@ -2291,6 +2416,8 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv, bool resume)
 
 	if (IS_SKYLAKE(dev) || IS_KABYLAKE(dev)) {
 		skl_display_core_init(dev_priv, resume);
+	} else if (IS_BROXTON(dev)) {
+		bxt_display_core_init(dev_priv, resume);
 	} else if (IS_CHERRYVIEW(dev)) {
 		mutex_lock(&power_domains->lock);
 		chv_phy_control_init(dev_priv);
@@ -2328,6 +2455,8 @@ void intel_power_domains_suspend(struct drm_i915_private *dev_priv)
 
 	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
 		skl_display_core_uninit(dev_priv);
+	else if (IS_BROXTON(dev_priv))
+		bxt_display_core_uninit(dev_priv);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 4ecc076c4041..2128fae5687d 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -1398,12 +1398,10 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 	}
 
 	dotclock = pipe_config->port_clock;
+
 	if (pipe_config->pixel_multiplier)
 		dotclock /= pipe_config->pixel_multiplier;
 
-	if (HAS_PCH_SPLIT(dev))
-		ironlake_check_encoder_dotclock(pipe_config, dotclock);
-
 	pipe_config->base.adjusted_mode.crtc_clock = dotclock;
 
 	/* Cross check the port pixel multiplier with the sdvo encoder state. */
@@ -2262,9 +2260,9 @@ intel_sdvo_select_ddc_bus(struct drm_i915_private *dev_priv,
 	struct sdvo_device_mapping *mapping;
 
 	if (sdvo->port == PORT_B)
-		mapping = &(dev_priv->sdvo_mappings[0]);
+		mapping = &dev_priv->vbt.sdvo_mappings[0];
 	else
-		mapping = &(dev_priv->sdvo_mappings[1]);
+		mapping = &dev_priv->vbt.sdvo_mappings[1];
 
 	if (mapping->initialized)
 		sdvo->ddc_bus = 1 << ((mapping->ddc_pin & 0xf0) >> 4);
@@ -2280,9 +2278,9 @@ intel_sdvo_select_i2c_bus(struct drm_i915_private *dev_priv,
 	u8 pin;
 
 	if (sdvo->port == PORT_B)
-		mapping = &dev_priv->sdvo_mappings[0];
+		mapping = &dev_priv->vbt.sdvo_mappings[0];
 	else
-		mapping = &dev_priv->sdvo_mappings[1];
+		mapping = &dev_priv->vbt.sdvo_mappings[1];
 
 	if (mapping->initialized &&
 	    intel_gmbus_is_valid_pin(dev_priv, mapping->i2c_pin))
@@ -2318,11 +2316,11 @@ intel_sdvo_get_slave_addr(struct drm_device *dev, struct intel_sdvo *sdvo)
 	struct sdvo_device_mapping *my_mapping, *other_mapping;
 
 	if (sdvo->port == PORT_B) {
-		my_mapping = &dev_priv->sdvo_mappings[0];
-		other_mapping = &dev_priv->sdvo_mappings[1];
+		my_mapping = &dev_priv->vbt.sdvo_mappings[0];
+		other_mapping = &dev_priv->vbt.sdvo_mappings[1];
 	} else {
-		my_mapping = &dev_priv->sdvo_mappings[1];
-		other_mapping = &dev_priv->sdvo_mappings[0];
+		my_mapping = &dev_priv->vbt.sdvo_mappings[1];
+		other_mapping = &dev_priv->vbt.sdvo_mappings[0];
 	}
 
 	/* If the BIOS described our SDVO device, take advantage of it. */
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index a2582c455b36..0f3e2303e0e9 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -193,7 +193,7 @@ skl_update_plane(struct drm_plane *drm_plane,
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 	u32 surf_addr;
 	u32 tile_height, plane_offset, plane_size;
-	unsigned int rotation;
+	unsigned int rotation = plane_state->base.rotation;
 	int x_offset, y_offset;
 	int crtc_x = plane_state->dst.x1;
 	int crtc_y = plane_state->dst.y1;
@@ -213,7 +213,6 @@ skl_update_plane(struct drm_plane *drm_plane,
 	plane_ctl |= skl_plane_ctl_format(fb->pixel_format);
 	plane_ctl |= skl_plane_ctl_tiling(fb->modifier[0]);
 
-	rotation = plane_state->base.rotation;
 	plane_ctl |= skl_plane_ctl_rotation(rotation);
 
 	stride_div = intel_fb_stride_alignment(dev_priv, fb->modifier[0],
@@ -351,6 +350,7 @@ vlv_update_plane(struct drm_plane *dplane,
 	int plane = intel_plane->plane;
 	u32 sprctl;
 	u32 sprsurf_offset, linear_offset;
+	unsigned int rotation = dplane->state->rotation;
 	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 	int crtc_x = plane_state->dst.x1;
@@ -423,12 +423,11 @@ vlv_update_plane(struct drm_plane *dplane,
 	crtc_h--;
 
 	linear_offset = y * fb->pitches[0] + x * cpp;
-	sprsurf_offset = intel_compute_tile_offset(dev_priv, &x, &y,
-						   fb->modifier[0], cpp,
-						   fb->pitches[0]);
+	sprsurf_offset = intel_compute_tile_offset(&x, &y, fb, 0,
+						   fb->pitches[0], rotation);
 	linear_offset -= sprsurf_offset;
 
-	if (plane_state->base.rotation == BIT(DRM_ROTATE_180)) {
+	if (rotation == BIT(DRM_ROTATE_180)) {
 		sprctl |= SP_ROTATE_180;
 
 		x += src_w;
@@ -493,6 +492,7 @@ ivb_update_plane(struct drm_plane *plane,
 	enum pipe pipe = intel_plane->pipe;
 	u32 sprctl, sprscale = 0;
 	u32 sprsurf_offset, linear_offset;
+	unsigned int rotation = plane_state->base.rotation;
 	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 	int crtc_x = plane_state->dst.x1;
@@ -556,12 +556,11 @@ ivb_update_plane(struct drm_plane *plane,
 		sprscale = SPRITE_SCALE_ENABLE | (src_w << 16) | src_h;
 
 	linear_offset = y * fb->pitches[0] + x * cpp;
-	sprsurf_offset = intel_compute_tile_offset(dev_priv, &x, &y,
-						   fb->modifier[0], cpp,
-						   fb->pitches[0]);
+	sprsurf_offset = intel_compute_tile_offset(&x, &y, fb, 0,
+						   fb->pitches[0], rotation);
 	linear_offset -= sprsurf_offset;
 
-	if (plane_state->base.rotation == BIT(DRM_ROTATE_180)) {
+	if (rotation == BIT(DRM_ROTATE_180)) {
 		sprctl |= SPRITE_ROTATE_180;
 
 		/* HSW and BDW does this automagically in hardware */
@@ -634,6 +633,7 @@ ilk_update_plane(struct drm_plane *plane,
 	int pipe = intel_plane->pipe;
 	u32 dvscntr, dvsscale;
 	u32 dvssurf_offset, linear_offset;
+	unsigned int rotation = plane_state->base.rotation;
 	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 	int crtc_x = plane_state->dst.x1;
@@ -693,12 +693,11 @@ ilk_update_plane(struct drm_plane *plane,
 		dvsscale = DVS_SCALE_ENABLE | (src_w << 16) | src_h;
 
 	linear_offset = y * fb->pitches[0] + x * cpp;
-	dvssurf_offset = intel_compute_tile_offset(dev_priv, &x, &y,
-						   fb->modifier[0], cpp,
-						   fb->pitches[0]);
+	dvssurf_offset = intel_compute_tile_offset(&x, &y, fb, 0,
+						   fb->pitches[0], rotation);
 	linear_offset -= dvssurf_offset;
 
-	if (plane_state->base.rotation == BIT(DRM_ROTATE_180)) {
+	if (rotation == BIT(DRM_ROTATE_180)) {
 		dvscntr |= DVS_ROTATE_180;
 
 		x += src_w;
@@ -1026,8 +1025,8 @@ static uint32_t skl_plane_formats[] = {
 int
 intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 {
-	struct intel_plane *intel_plane;
-	struct intel_plane_state *state;
+	struct intel_plane *intel_plane = NULL;
+	struct intel_plane_state *state = NULL;
 	unsigned long possible_crtcs;
 	const uint32_t *plane_formats;
 	int num_plane_formats;
@@ -1037,13 +1036,15 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 		return -ENODEV;
 
 	intel_plane = kzalloc(sizeof(*intel_plane), GFP_KERNEL);
-	if (!intel_plane)
-		return -ENOMEM;
+	if (!intel_plane) {
+		ret = -ENOMEM;
+		goto fail;
+	}
 
 	state = intel_create_plane_state(&intel_plane->base);
 	if (!state) {
-		kfree(intel_plane);
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto fail;
 	}
 	intel_plane->base.state = &state->base;
 
@@ -1098,28 +1099,34 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 		num_plane_formats = ARRAY_SIZE(skl_plane_formats);
 		break;
 	default:
-		kfree(intel_plane);
-		return -ENODEV;
+		MISSING_CASE(INTEL_INFO(dev)->gen);
+		ret = -ENODEV;
+		goto fail;
 	}
 
 	intel_plane->pipe = pipe;
 	intel_plane->plane = plane;
 	intel_plane->frontbuffer_bit = INTEL_FRONTBUFFER_SPRITE(pipe, plane);
 	intel_plane->check_plane = intel_check_sprite_plane;
+
 	possible_crtcs = (1 << pipe);
+
 	ret = drm_universal_plane_init(dev, &intel_plane->base, possible_crtcs,
 				       &intel_plane_funcs,
 				       plane_formats, num_plane_formats,
 				       DRM_PLANE_TYPE_OVERLAY, NULL);
-	if (ret) {
-		kfree(intel_plane);
-		goto out;
-	}
+	if (ret)
+		goto fail;
 
 	intel_create_rotation_property(dev, intel_plane);
 
 	drm_plane_helper_add(&intel_plane->base, &intel_plane_helper_funcs);
 
-out:
+	return 0;
+
+fail:
+	kfree(state);
+	kfree(intel_plane);
+
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 6745bad5bff0..223129d3c765 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -326,24 +326,12 @@ static const struct color_conversion sdtv_csc_yprpb = {
 	.rv = 0x0100, .gv = 0x03ad, .bv = 0x074d, .av = 0x0200,
 };
 
-static const struct color_conversion sdtv_csc_rgb = {
-	.ry = 0x0000, .gy = 0x0f00, .by = 0x0000, .ay = 0x0166,
-	.ru = 0x0000, .gu = 0x0000, .bu = 0x0f00, .au = 0x0166,
-	.rv = 0x0f00, .gv = 0x0000, .bv = 0x0000, .av = 0x0166,
-};
-
 static const struct color_conversion hdtv_csc_yprpb = {
 	.ry = 0x05b3, .gy = 0x016e, .by = 0x0728, .ay = 0x0145,
 	.ru = 0x07d5, .gu = 0x038b, .bu = 0x0100, .au = 0x0200,
 	.rv = 0x0100, .gv = 0x03d1, .bv = 0x06bc, .av = 0x0200,
 };
 
-static const struct color_conversion hdtv_csc_rgb = {
-	.ry = 0x0000, .gy = 0x0f00, .by = 0x0000, .ay = 0x0166,
-	.ru = 0x0000, .gu = 0x0000, .bu = 0x0f00, .au = 0x0166,
-	.rv = 0x0f00, .gv = 0x0000, .bv = 0x0000, .av = 0x0166,
-};
-
 static const struct video_levels component_levels = {
 	.blank = 279, .black = 279, .burst = 0,
 };
@@ -1531,47 +1519,6 @@ static const struct drm_encoder_funcs intel_tv_enc_funcs = {
 	.destroy = intel_encoder_destroy,
 };
 
-/*
- * Enumerate the child dev array parsed from VBT to check whether
- * the integrated TV is present.
- * If it is present, return 1.
- * If it is not present, return false.
- * If no child dev is parsed from VBT, it assumes that the TV is present.
- */
-static int tv_is_present_in_vbt(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	union child_device_config *p_child;
-	int i, ret;
-
-	if (!dev_priv->vbt.child_dev_num)
-		return 1;
-
-	ret = 0;
-	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
-		p_child = dev_priv->vbt.child_dev + i;
-		/*
-		 * If the device type is not TV, continue.
-		 */
-		switch (p_child->old.device_type) {
-		case DEVICE_TYPE_INT_TV:
-		case DEVICE_TYPE_TV:
-		case DEVICE_TYPE_TV_SVIDEO_COMPOSITE:
-			break;
-		default:
-			continue;
-		}
-		/* Only when the addin_offset is non-zero, it is regarded
-		 * as present.
-		 */
-		if (p_child->old.addin_offset) {
-			ret = 1;
-			break;
-		}
-	}
-	return ret;
-}
-
 void
 intel_tv_init(struct drm_device *dev)
 {
@@ -1587,13 +1534,10 @@ intel_tv_init(struct drm_device *dev)
 	if ((I915_READ(TV_CTL) & TV_FUSE_STATE_MASK) == TV_FUSE_STATE_DISABLED)
 		return;
 
-	if (!tv_is_present_in_vbt(dev)) {
+	if (!intel_bios_is_tv_present(dev_priv)) {
 		DRM_DEBUG_KMS("Integrated TV is not present.\n");
 		return;
 	}
-	/* Even if we have an encoder we may not have a connector */
-	if (!dev_priv->vbt.int_tv_support)
-		return;
 
 	/*
 	 * Sanity check the TV output by checking to see if the
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 68b6f69aa682..4f1dfe616856 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -60,7 +60,11 @@ fw_domain_reset(const struct intel_uncore_forcewake_domain *d)
 static inline void
 fw_domain_arm_timer(struct intel_uncore_forcewake_domain *d)
 {
-	mod_timer_pinned(&d->timer, jiffies + 1);
+	d->wake_count++;
+	hrtimer_start_range_ns(&d->timer,
+			       ktime_set(0, NSEC_PER_MSEC),
+			       NSEC_PER_MSEC,
+			       HRTIMER_MODE_REL);
 }
 
 static inline void
@@ -107,22 +111,22 @@ static void
 fw_domains_get(struct drm_i915_private *dev_priv, enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *d;
-	enum forcewake_domain_id id;
 
-	for_each_fw_domain_mask(d, fw_domains, dev_priv, id) {
+	for_each_fw_domain_masked(d, fw_domains, dev_priv) {
 		fw_domain_wait_ack_clear(d);
 		fw_domain_get(d);
-		fw_domain_wait_ack(d);
 	}
+
+	for_each_fw_domain_masked(d, fw_domains, dev_priv)
+		fw_domain_wait_ack(d);
 }
 
 static void
 fw_domains_put(struct drm_i915_private *dev_priv, enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *d;
-	enum forcewake_domain_id id;
 
-	for_each_fw_domain_mask(d, fw_domains, dev_priv, id) {
+	for_each_fw_domain_masked(d, fw_domains, dev_priv) {
 		fw_domain_put(d);
 		fw_domain_posting_read(d);
 	}
@@ -132,10 +136,9 @@ static void
 fw_domains_posting_read(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore_forcewake_domain *d;
-	enum forcewake_domain_id id;
 
 	/* No need to do for all, just do for first found */
-	for_each_fw_domain(d, dev_priv, id) {
+	for_each_fw_domain(d, dev_priv) {
 		fw_domain_posting_read(d);
 		break;
 	}
@@ -145,12 +148,11 @@ static void
 fw_domains_reset(struct drm_i915_private *dev_priv, enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *d;
-	enum forcewake_domain_id id;
 
 	if (dev_priv->uncore.fw_domains == 0)
 		return;
 
-	for_each_fw_domain_mask(d, fw_domains, dev_priv, id)
+	for_each_fw_domain_masked(d, fw_domains, dev_priv)
 		fw_domain_reset(d);
 
 	fw_domains_posting_read(dev_priv);
@@ -204,7 +206,7 @@ static int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
 
 	/* On VLV, FIFO will be shared by both SW and HW.
 	 * So, we need to read the FREE_ENTRIES everytime */
-	if (IS_VALLEYVIEW(dev_priv->dev))
+	if (IS_VALLEYVIEW(dev_priv))
 		dev_priv->uncore.fifo_count = fifo_free_entries(dev_priv);
 
 	if (dev_priv->uncore.fifo_count < GT_FIFO_NUM_RESERVED_ENTRIES) {
@@ -224,9 +226,11 @@ static int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
 	return ret;
 }
 
-static void intel_uncore_fw_release_timer(unsigned long arg)
+static enum hrtimer_restart
+intel_uncore_fw_release_timer(struct hrtimer *timer)
 {
-	struct intel_uncore_forcewake_domain *domain = (void *)arg;
+	struct intel_uncore_forcewake_domain *domain =
+	       container_of(timer, struct intel_uncore_forcewake_domain, timer);
 	unsigned long irqflags;
 
 	assert_rpm_device_not_suspended(domain->i915);
@@ -240,6 +244,8 @@ static void intel_uncore_fw_release_timer(unsigned long arg)
 							  1 << domain->id);
 
 	spin_unlock_irqrestore(&domain->i915->uncore.lock, irqflags);
+
+	return HRTIMER_NORESTART;
 }
 
 void intel_uncore_forcewake_reset(struct drm_device *dev, bool restore)
@@ -248,7 +254,6 @@ void intel_uncore_forcewake_reset(struct drm_device *dev, bool restore)
 	unsigned long irqflags;
 	struct intel_uncore_forcewake_domain *domain;
 	int retry_count = 100;
-	enum forcewake_domain_id id;
 	enum forcewake_domains fw = 0, active_domains;
 
 	/* Hold uncore.lock across reset to prevent any register access
@@ -258,18 +263,18 @@ void intel_uncore_forcewake_reset(struct drm_device *dev, bool restore)
 	while (1) {
 		active_domains = 0;
 
-		for_each_fw_domain(domain, dev_priv, id) {
-			if (del_timer_sync(&domain->timer) == 0)
+		for_each_fw_domain(domain, dev_priv) {
+			if (hrtimer_cancel(&domain->timer) == 0)
 				continue;
 
-			intel_uncore_fw_release_timer((unsigned long)domain);
+			intel_uncore_fw_release_timer(&domain->timer);
 		}
 
 		spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
 
-		for_each_fw_domain(domain, dev_priv, id) {
-			if (timer_pending(&domain->timer))
-				active_domains |= (1 << id);
+		for_each_fw_domain(domain, dev_priv) {
+			if (hrtimer_active(&domain->timer))
+				active_domains |= domain->mask;
 		}
 
 		if (active_domains == 0)
@@ -286,9 +291,9 @@ void intel_uncore_forcewake_reset(struct drm_device *dev, bool restore)
 
 	WARN_ON(active_domains);
 
-	for_each_fw_domain(domain, dev_priv, id)
+	for_each_fw_domain(domain, dev_priv)
 		if (domain->wake_count)
-			fw |= 1 << id;
+			fw |= domain->mask;
 
 	if (fw)
 		dev_priv->uncore.funcs.force_wake_put(dev_priv, fw);
@@ -310,21 +315,49 @@ void intel_uncore_forcewake_reset(struct drm_device *dev, bool restore)
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
-static void intel_uncore_ellc_detect(struct drm_device *dev)
+static u64 gen9_edram_size(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	const unsigned int ways[8] = { 4, 8, 12, 16, 16, 16, 16, 16 };
+	const unsigned int sets[4] = { 1, 1, 2, 2 };
+	const u32 cap = dev_priv->edram_cap;
+
+	return EDRAM_NUM_BANKS(cap) *
+		ways[EDRAM_WAYS_IDX(cap)] *
+		sets[EDRAM_SETS_IDX(cap)] *
+		1024 * 1024;
+}
+
+u64 intel_uncore_edram_size(struct drm_i915_private *dev_priv)
+{
+	if (!HAS_EDRAM(dev_priv))
+		return 0;
 
-	if ((IS_HASWELL(dev) || IS_BROADWELL(dev) ||
-	     INTEL_INFO(dev)->gen >= 9) &&
-	    (__raw_i915_read32(dev_priv, HSW_EDRAM_PRESENT) & EDRAM_ENABLED)) {
-		/* The docs do not explain exactly how the calculation can be
-		 * made. It is somewhat guessable, but for now, it's always
-		 * 128MB.
-		 * NB: We can't write IDICR yet because we do not have gt funcs
+	/* The needed capability bits for size calculation
+	 * are not there with pre gen9 so return 128MB always.
+	 */
+	if (INTEL_GEN(dev_priv) < 9)
+		return 128 * 1024 * 1024;
+
+	return gen9_edram_size(dev_priv);
+}
+
+static void intel_uncore_edram_detect(struct drm_i915_private *dev_priv)
+{
+	if (IS_HASWELL(dev_priv) ||
+	    IS_BROADWELL(dev_priv) ||
+	    INTEL_GEN(dev_priv) >= 9) {
+		dev_priv->edram_cap = __raw_i915_read32(dev_priv,
+							HSW_EDRAM_CAP);
+
+		/* NB: We can't write IDICR yet because we do not have gt funcs
 		 * set up */
-		dev_priv->ellc_size = 128;
-		DRM_INFO("Found %zuMB of eLLC\n", dev_priv->ellc_size);
+	} else {
+		dev_priv->edram_cap = 0;
 	}
+
+	if (HAS_EDRAM(dev_priv))
+		DRM_INFO("Found %lluMB of eDRAM\n",
+			 intel_uncore_edram_size(dev_priv) / (1024 * 1024));
 }
 
 static bool
@@ -410,16 +443,15 @@ static void __intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
 					 enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *domain;
-	enum forcewake_domain_id id;
 
 	if (!dev_priv->uncore.funcs.force_wake_get)
 		return;
 
 	fw_domains &= dev_priv->uncore.fw_domains;
 
-	for_each_fw_domain_mask(domain, fw_domains, dev_priv, id) {
+	for_each_fw_domain_masked(domain, fw_domains, dev_priv) {
 		if (domain->wake_count++)
-			fw_domains &= ~(1 << id);
+			fw_domains &= ~domain->mask;
 	}
 
 	if (fw_domains)
@@ -477,21 +509,19 @@ static void __intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
 					 enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *domain;
-	enum forcewake_domain_id id;
 
 	if (!dev_priv->uncore.funcs.force_wake_put)
 		return;
 
 	fw_domains &= dev_priv->uncore.fw_domains;
 
-	for_each_fw_domain_mask(domain, fw_domains, dev_priv, id) {
+	for_each_fw_domain_masked(domain, fw_domains, dev_priv) {
 		if (WARN_ON(domain->wake_count == 0))
 			continue;
 
 		if (--domain->wake_count)
 			continue;
 
-		domain->wake_count++;
 		fw_domain_arm_timer(domain);
 	}
 }
@@ -539,18 +569,27 @@ void intel_uncore_forcewake_put__locked(struct drm_i915_private *dev_priv,
 void assert_forcewakes_inactive(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore_forcewake_domain *domain;
-	enum forcewake_domain_id id;
 
 	if (!dev_priv->uncore.funcs.force_wake_get)
 		return;
 
-	for_each_fw_domain(domain, dev_priv, id)
+	for_each_fw_domain(domain, dev_priv)
 		WARN_ON(domain->wake_count);
 }
 
 /* We give fast paths for the really cool registers */
 #define NEEDS_FORCE_WAKE(reg) ((reg) < 0x40000)
 
+#define __gen6_reg_read_fw_domains(offset) \
+({ \
+	enum forcewake_domains __fwd; \
+	if (NEEDS_FORCE_WAKE(offset)) \
+		__fwd = FORCEWAKE_RENDER; \
+	else \
+		__fwd = 0; \
+	__fwd; \
+})
+
 #define REG_RANGE(reg, start, end) ((reg) >= (start) && (reg) < (end))
 
 #define FORCEWAKE_VLV_RENDER_RANGE_OFFSET(reg) \
@@ -564,6 +603,48 @@ void assert_forcewakes_inactive(struct drm_i915_private *dev_priv)
 	 REG_RANGE((reg), 0x22000, 0x24000) || \
 	 REG_RANGE((reg), 0x30000, 0x40000))
 
+#define __vlv_reg_read_fw_domains(offset) \
+({ \
+	enum forcewake_domains __fwd = 0; \
+	if (!NEEDS_FORCE_WAKE(offset)) \
+		__fwd = 0; \
+	else if (FORCEWAKE_VLV_RENDER_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER; \
+	else if (FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_MEDIA; \
+	__fwd; \
+})
+
+static const i915_reg_t gen8_shadowed_regs[] = {
+	GEN6_RPNSWREQ,
+	GEN6_RC_VIDEO_FREQ,
+	RING_TAIL(RENDER_RING_BASE),
+	RING_TAIL(GEN6_BSD_RING_BASE),
+	RING_TAIL(VEBOX_RING_BASE),
+	RING_TAIL(BLT_RING_BASE),
+	/* TODO: Other registers are not yet used */
+};
+
+static bool is_gen8_shadowed(u32 offset)
+{
+	int i;
+	for (i = 0; i < ARRAY_SIZE(gen8_shadowed_regs); i++)
+		if (offset == gen8_shadowed_regs[i].reg)
+			return true;
+
+	return false;
+}
+
+#define __gen8_reg_write_fw_domains(offset) \
+({ \
+	enum forcewake_domains __fwd; \
+	if (NEEDS_FORCE_WAKE(offset) && !is_gen8_shadowed(offset)) \
+		__fwd = FORCEWAKE_RENDER; \
+	else \
+		__fwd = 0; \
+	__fwd; \
+})
+
 #define FORCEWAKE_CHV_RENDER_RANGE_OFFSET(reg) \
 	(REG_RANGE((reg), 0x2000, 0x4000) || \
 	 REG_RANGE((reg), 0x5200, 0x8000) || \
@@ -586,6 +667,34 @@ void assert_forcewakes_inactive(struct drm_i915_private *dev_priv)
 	 REG_RANGE((reg), 0x9000, 0xB000) || \
 	 REG_RANGE((reg), 0xF000, 0x10000))
 
+#define __chv_reg_read_fw_domains(offset) \
+({ \
+	enum forcewake_domains __fwd = 0; \
+	if (!NEEDS_FORCE_WAKE(offset)) \
+		__fwd = 0; \
+	else if (FORCEWAKE_CHV_RENDER_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER; \
+	else if (FORCEWAKE_CHV_MEDIA_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_MEDIA; \
+	else if (FORCEWAKE_CHV_COMMON_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
+	__fwd; \
+})
+
+#define __chv_reg_write_fw_domains(offset) \
+({ \
+	enum forcewake_domains __fwd = 0; \
+	if (!NEEDS_FORCE_WAKE(offset) || is_gen8_shadowed(offset)) \
+		__fwd = 0; \
+	else if (FORCEWAKE_CHV_RENDER_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER; \
+	else if (FORCEWAKE_CHV_MEDIA_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_MEDIA; \
+	else if (FORCEWAKE_CHV_COMMON_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
+	__fwd; \
+})
+
 #define FORCEWAKE_GEN9_UNCORE_RANGE_OFFSET(reg) \
 	REG_RANGE((reg), 0xB00,  0x2000)
 
@@ -618,6 +727,61 @@ void assert_forcewakes_inactive(struct drm_i915_private *dev_priv)
 	 !FORCEWAKE_GEN9_MEDIA_RANGE_OFFSET(reg) && \
 	 !FORCEWAKE_GEN9_COMMON_RANGE_OFFSET(reg))
 
+#define SKL_NEEDS_FORCE_WAKE(reg) \
+	((reg) < 0x40000 && !FORCEWAKE_GEN9_UNCORE_RANGE_OFFSET(reg))
+
+#define __gen9_reg_read_fw_domains(offset) \
+({ \
+	enum forcewake_domains __fwd; \
+	if (!SKL_NEEDS_FORCE_WAKE(offset)) \
+		__fwd = 0; \
+	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER; \
+	else if (FORCEWAKE_GEN9_MEDIA_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_MEDIA; \
+	else if (FORCEWAKE_GEN9_COMMON_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
+	else \
+		__fwd = FORCEWAKE_BLITTER; \
+	__fwd; \
+})
+
+static const i915_reg_t gen9_shadowed_regs[] = {
+	RING_TAIL(RENDER_RING_BASE),
+	RING_TAIL(GEN6_BSD_RING_BASE),
+	RING_TAIL(VEBOX_RING_BASE),
+	RING_TAIL(BLT_RING_BASE),
+	GEN6_RPNSWREQ,
+	GEN6_RC_VIDEO_FREQ,
+	/* TODO: Other registers are not yet used */
+};
+
+static bool is_gen9_shadowed(u32 offset)
+{
+	int i;
+	for (i = 0; i < ARRAY_SIZE(gen9_shadowed_regs); i++)
+		if (offset == gen9_shadowed_regs[i].reg)
+			return true;
+
+	return false;
+}
+
+#define __gen9_reg_write_fw_domains(offset) \
+({ \
+	enum forcewake_domains __fwd; \
+	if (!SKL_NEEDS_FORCE_WAKE(offset) || is_gen9_shadowed(offset)) \
+		__fwd = 0; \
+	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER; \
+	else if (FORCEWAKE_GEN9_MEDIA_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_MEDIA; \
+	else if (FORCEWAKE_GEN9_COMMON_RANGE_OFFSET(offset)) \
+		__fwd = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
+	else \
+		__fwd = FORCEWAKE_BLITTER; \
+	__fwd; \
+})
+
 static void
 ilk_dummy_write(struct drm_i915_private *dev_priv)
 {
@@ -633,15 +797,6 @@ __unclaimed_reg_debug(struct drm_i915_private *dev_priv,
 		      const bool read,
 		      const bool before)
 {
-	/* XXX. We limit the auto arming traces for mmio
-	 * debugs on these platforms. There are just too many
-	 * revealed by these and CI/Bat suffers from the noise.
-	 * Please fix and then re-enable the automatic traces.
-	 */
-	if (i915.mmio_debug < 2 &&
-	    (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)))
-		return;
-
 	if (WARN(check_for_unclaimed_mmio(dev_priv),
 		 "Unclaimed register detected %s %s register 0x%x\n",
 		 before ? "before" : "after",
@@ -716,23 +871,21 @@ __gen2_read(64)
 	trace_i915_reg_rw(false, reg, val, sizeof(val), trace); \
 	return val
 
-static inline void __force_wake_get(struct drm_i915_private *dev_priv,
-				    enum forcewake_domains fw_domains)
+static inline void __force_wake_auto(struct drm_i915_private *dev_priv,
+				     enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *domain;
-	enum forcewake_domain_id id;
 
 	if (WARN_ON(!fw_domains))
 		return;
 
 	/* Ideally GCC would be constant-fold and eliminate this loop */
-	for_each_fw_domain_mask(domain, fw_domains, dev_priv, id) {
+	for_each_fw_domain_masked(domain, fw_domains, dev_priv) {
 		if (domain->wake_count) {
-			fw_domains &= ~(1 << id);
+			fw_domains &= ~domain->mask;
 			continue;
 		}
 
-		domain->wake_count++;
 		fw_domain_arm_timer(domain);
 	}
 
@@ -743,9 +896,11 @@ static inline void __force_wake_get(struct drm_i915_private *dev_priv,
 #define __gen6_read(x) \
 static u##x \
 gen6_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
+	enum forcewake_domains fw_engine; \
 	GEN6_READ_HEADER(x); \
-	if (NEEDS_FORCE_WAKE(offset)) \
-		__force_wake_get(dev_priv, FORCEWAKE_RENDER); \
+	fw_engine = __gen6_reg_read_fw_domains(offset); \
+	if (fw_engine) \
+		__force_wake_auto(dev_priv, fw_engine); \
 	val = __raw_i915_read##x(dev_priv, reg); \
 	GEN6_READ_FOOTER; \
 }
@@ -753,16 +908,11 @@ gen6_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
 #define __vlv_read(x) \
 static u##x \
 vlv_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
-	enum forcewake_domains fw_engine = 0; \
+	enum forcewake_domains fw_engine; \
 	GEN6_READ_HEADER(x); \
-	if (!NEEDS_FORCE_WAKE(offset)) \
-		fw_engine = 0; \
-	else if (FORCEWAKE_VLV_RENDER_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER; \
-	else if (FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_MEDIA; \
+	fw_engine = __vlv_reg_read_fw_domains(offset); \
 	if (fw_engine) \
-		__force_wake_get(dev_priv, fw_engine); \
+		__force_wake_auto(dev_priv, fw_engine); \
 	val = __raw_i915_read##x(dev_priv, reg); \
 	GEN6_READ_FOOTER; \
 }
@@ -770,42 +920,23 @@ vlv_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
 #define __chv_read(x) \
 static u##x \
 chv_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
-	enum forcewake_domains fw_engine = 0; \
+	enum forcewake_domains fw_engine; \
 	GEN6_READ_HEADER(x); \
-	if (!NEEDS_FORCE_WAKE(offset)) \
-		fw_engine = 0; \
-	else if (FORCEWAKE_CHV_RENDER_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER; \
-	else if (FORCEWAKE_CHV_MEDIA_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_MEDIA; \
-	else if (FORCEWAKE_CHV_COMMON_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
+	fw_engine = __chv_reg_read_fw_domains(offset); \
 	if (fw_engine) \
-		__force_wake_get(dev_priv, fw_engine); \
+		__force_wake_auto(dev_priv, fw_engine); \
 	val = __raw_i915_read##x(dev_priv, reg); \
 	GEN6_READ_FOOTER; \
 }
 
-#define SKL_NEEDS_FORCE_WAKE(reg) \
-	((reg) < 0x40000 && !FORCEWAKE_GEN9_UNCORE_RANGE_OFFSET(reg))
-
 #define __gen9_read(x) \
 static u##x \
 gen9_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
 	enum forcewake_domains fw_engine; \
 	GEN6_READ_HEADER(x); \
-	if (!SKL_NEEDS_FORCE_WAKE(offset)) \
-		fw_engine = 0; \
-	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER; \
-	else if (FORCEWAKE_GEN9_MEDIA_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_MEDIA; \
-	else if (FORCEWAKE_GEN9_COMMON_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
-	else \
-		fw_engine = FORCEWAKE_BLITTER; \
+	fw_engine = __gen9_reg_read_fw_domains(offset); \
 	if (fw_engine) \
-		__force_wake_get(dev_priv, fw_engine); \
+		__force_wake_auto(dev_priv, fw_engine); \
 	val = __raw_i915_read##x(dev_priv, reg); \
 	GEN6_READ_FOOTER; \
 }
@@ -942,34 +1073,14 @@ hsw_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool t
 	GEN6_WRITE_FOOTER; \
 }
 
-static const i915_reg_t gen8_shadowed_regs[] = {
-	FORCEWAKE_MT,
-	GEN6_RPNSWREQ,
-	GEN6_RC_VIDEO_FREQ,
-	RING_TAIL(RENDER_RING_BASE),
-	RING_TAIL(GEN6_BSD_RING_BASE),
-	RING_TAIL(VEBOX_RING_BASE),
-	RING_TAIL(BLT_RING_BASE),
-	/* TODO: Other registers are not yet used */
-};
-
-static bool is_gen8_shadowed(struct drm_i915_private *dev_priv,
-			     i915_reg_t reg)
-{
-	int i;
-	for (i = 0; i < ARRAY_SIZE(gen8_shadowed_regs); i++)
-		if (i915_mmio_reg_equal(reg, gen8_shadowed_regs[i]))
-			return true;
-
-	return false;
-}
-
 #define __gen8_write(x) \
 static void \
 gen8_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool trace) { \
+	enum forcewake_domains fw_engine; \
 	GEN6_WRITE_HEADER; \
-	if (NEEDS_FORCE_WAKE(offset) && !is_gen8_shadowed(dev_priv, reg)) \
-		__force_wake_get(dev_priv, FORCEWAKE_RENDER); \
+	fw_engine = __gen8_reg_write_fw_domains(offset); \
+	if (fw_engine) \
+		__force_wake_auto(dev_priv, fw_engine); \
 	__raw_i915_write##x(dev_priv, reg, val); \
 	GEN6_WRITE_FOOTER; \
 }
@@ -977,66 +1088,24 @@ gen8_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool
 #define __chv_write(x) \
 static void \
 chv_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool trace) { \
-	enum forcewake_domains fw_engine = 0; \
+	enum forcewake_domains fw_engine; \
 	GEN6_WRITE_HEADER; \
-	if (!NEEDS_FORCE_WAKE(offset) || \
-	    is_gen8_shadowed(dev_priv, reg)) \
-		fw_engine = 0; \
-	else if (FORCEWAKE_CHV_RENDER_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER; \
-	else if (FORCEWAKE_CHV_MEDIA_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_MEDIA; \
-	else if (FORCEWAKE_CHV_COMMON_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
+	fw_engine = __chv_reg_write_fw_domains(offset); \
 	if (fw_engine) \
-		__force_wake_get(dev_priv, fw_engine); \
+		__force_wake_auto(dev_priv, fw_engine); \
 	__raw_i915_write##x(dev_priv, reg, val); \
 	GEN6_WRITE_FOOTER; \
 }
 
-static const i915_reg_t gen9_shadowed_regs[] = {
-	RING_TAIL(RENDER_RING_BASE),
-	RING_TAIL(GEN6_BSD_RING_BASE),
-	RING_TAIL(VEBOX_RING_BASE),
-	RING_TAIL(BLT_RING_BASE),
-	FORCEWAKE_BLITTER_GEN9,
-	FORCEWAKE_RENDER_GEN9,
-	FORCEWAKE_MEDIA_GEN9,
-	GEN6_RPNSWREQ,
-	GEN6_RC_VIDEO_FREQ,
-	/* TODO: Other registers are not yet used */
-};
-
-static bool is_gen9_shadowed(struct drm_i915_private *dev_priv,
-			     i915_reg_t reg)
-{
-	int i;
-	for (i = 0; i < ARRAY_SIZE(gen9_shadowed_regs); i++)
-		if (i915_mmio_reg_equal(reg, gen9_shadowed_regs[i]))
-			return true;
-
-	return false;
-}
-
 #define __gen9_write(x) \
 static void \
 gen9_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, \
 		bool trace) { \
 	enum forcewake_domains fw_engine; \
 	GEN6_WRITE_HEADER; \
-	if (!SKL_NEEDS_FORCE_WAKE(offset) || \
-	    is_gen9_shadowed(dev_priv, reg)) \
-		fw_engine = 0; \
-	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER; \
-	else if (FORCEWAKE_GEN9_MEDIA_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_MEDIA; \
-	else if (FORCEWAKE_GEN9_COMMON_RANGE_OFFSET(offset)) \
-		fw_engine = FORCEWAKE_RENDER | FORCEWAKE_MEDIA; \
-	else \
-		fw_engine = FORCEWAKE_BLITTER; \
+	fw_engine = __gen9_reg_write_fw_domains(offset); \
 	if (fw_engine) \
-		__force_wake_get(dev_priv, fw_engine); \
+		__force_wake_auto(dev_priv, fw_engine); \
 	__raw_i915_write##x(dev_priv, reg, val); \
 	GEN6_WRITE_FOOTER; \
 }
@@ -1150,7 +1219,14 @@ static void fw_domain_init(struct drm_i915_private *dev_priv,
 	d->i915 = dev_priv;
 	d->id = domain_id;
 
-	setup_timer(&d->timer, intel_uncore_fw_release_timer, (unsigned long)d);
+	BUILD_BUG_ON(FORCEWAKE_RENDER != (1 << FW_DOMAIN_ID_RENDER));
+	BUILD_BUG_ON(FORCEWAKE_BLITTER != (1 << FW_DOMAIN_ID_BLITTER));
+	BUILD_BUG_ON(FORCEWAKE_MEDIA != (1 << FW_DOMAIN_ID_MEDIA));
+
+	d->mask = 1 << domain_id;
+
+	hrtimer_init(&d->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	d->timer.function = intel_uncore_fw_release_timer;
 
 	dev_priv->uncore.fw_domains |= (1 << domain_id);
 
@@ -1161,7 +1237,7 @@ static void intel_uncore_fw_domains_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (INTEL_INFO(dev_priv->dev)->gen <= 5)
+	if (INTEL_INFO(dev_priv)->gen <= 5)
 		return;
 
 	if (IS_GEN9(dev)) {
@@ -1257,7 +1333,7 @@ void intel_uncore_init(struct drm_device *dev)
 
 	i915_check_vgpu(dev);
 
-	intel_uncore_ellc_detect(dev);
+	intel_uncore_edram_detect(dev_priv);
 	intel_uncore_fw_domains_init(dev);
 	__intel_uncore_early_sanitize(dev, false);
 
@@ -1437,7 +1513,7 @@ static int i915_reset_complete(struct drm_device *dev)
 	return (gdrst & GRDOM_RESET_STATUS) == 0;
 }
 
-static int i915_do_reset(struct drm_device *dev)
+static int i915_do_reset(struct drm_device *dev, unsigned engine_mask)
 {
 	/* assert reset for at least 20 usec */
 	pci_write_config_byte(dev->pdev, I915_GDRST, GRDOM_RESET_ENABLE);
@@ -1454,13 +1530,13 @@ static int g4x_reset_complete(struct drm_device *dev)
 	return (gdrst & GRDOM_RESET_ENABLE) == 0;
 }
 
-static int g33_do_reset(struct drm_device *dev)
+static int g33_do_reset(struct drm_device *dev, unsigned engine_mask)
 {
 	pci_write_config_byte(dev->pdev, I915_GDRST, GRDOM_RESET_ENABLE);
 	return wait_for(g4x_reset_complete(dev), 500);
 }
 
-static int g4x_do_reset(struct drm_device *dev)
+static int g4x_do_reset(struct drm_device *dev, unsigned engine_mask)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
@@ -1490,7 +1566,7 @@ static int g4x_do_reset(struct drm_device *dev)
 	return 0;
 }
 
-static int ironlake_do_reset(struct drm_device *dev)
+static int ironlake_do_reset(struct drm_device *dev, unsigned engine_mask)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
@@ -1514,75 +1590,132 @@ static int ironlake_do_reset(struct drm_device *dev)
 	return 0;
 }
 
-static int gen6_do_reset(struct drm_device *dev)
+/* Reset the hardware domains (GENX_GRDOM_*) specified by mask */
+static int gen6_hw_domain_reset(struct drm_i915_private *dev_priv,
+				u32 hw_domain_mask)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int	ret;
-
-	/* Reset the chip */
+	int ret;
 
 	/* GEN6_GDRST is not in the gt power well, no need to check
 	 * for fifo space for the write or forcewake the chip for
 	 * the read
 	 */
-	__raw_i915_write32(dev_priv, GEN6_GDRST, GEN6_GRDOM_FULL);
+	__raw_i915_write32(dev_priv, GEN6_GDRST, hw_domain_mask);
+
+#define ACKED ((__raw_i915_read32(dev_priv, GEN6_GDRST) & hw_domain_mask) == 0)
+	/* Spin waiting for the device to ack the reset requests */
+	ret = wait_for(ACKED, 500);
+#undef ACKED
+
+	return ret;
+}
 
-	/* Spin waiting for the device to ack the reset request */
-	ret = wait_for((__raw_i915_read32(dev_priv, GEN6_GDRST) & GEN6_GRDOM_FULL) == 0, 500);
+/**
+ * gen6_reset_engines - reset individual engines
+ * @dev: DRM device
+ * @engine_mask: mask of intel_ring_flag() engines or ALL_ENGINES for full reset
+ *
+ * This function will reset the individual engines that are set in engine_mask.
+ * If you provide ALL_ENGINES as mask, full global domain reset will be issued.
+ *
+ * Note: It is responsibility of the caller to handle the difference between
+ * asking full domain reset versus reset for all available individual engines.
+ *
+ * Returns 0 on success, nonzero on error.
+ */
+static int gen6_reset_engines(struct drm_device *dev, unsigned engine_mask)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *engine;
+	const u32 hw_engine_mask[I915_NUM_ENGINES] = {
+		[RCS] = GEN6_GRDOM_RENDER,
+		[BCS] = GEN6_GRDOM_BLT,
+		[VCS] = GEN6_GRDOM_MEDIA,
+		[VCS2] = GEN8_GRDOM_MEDIA2,
+		[VECS] = GEN6_GRDOM_VECS,
+	};
+	u32 hw_mask;
+	int ret;
+
+	if (engine_mask == ALL_ENGINES) {
+		hw_mask = GEN6_GRDOM_FULL;
+	} else {
+		hw_mask = 0;
+		for_each_engine_masked(engine, dev_priv, engine_mask)
+			hw_mask |= hw_engine_mask[engine->id];
+	}
+
+	ret = gen6_hw_domain_reset(dev_priv, hw_mask);
 
 	intel_uncore_forcewake_reset(dev, true);
 
 	return ret;
 }
 
-static int wait_for_register(struct drm_i915_private *dev_priv,
-			     i915_reg_t reg,
-			     const u32 mask,
-			     const u32 value,
-			     const unsigned long timeout_ms)
+static int wait_for_register_fw(struct drm_i915_private *dev_priv,
+				i915_reg_t reg,
+				const u32 mask,
+				const u32 value,
+				const unsigned long timeout_ms)
 {
-	return wait_for((I915_READ(reg) & mask) == value, timeout_ms);
+	return wait_for((I915_READ_FW(reg) & mask) == value, timeout_ms);
 }
 
-static int gen8_do_reset(struct drm_device *dev)
+static int gen8_request_engine_reset(struct intel_engine_cs *engine)
+{
+	int ret;
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
+
+	I915_WRITE_FW(RING_RESET_CTL(engine->mmio_base),
+		      _MASKED_BIT_ENABLE(RESET_CTL_REQUEST_RESET));
+
+	ret = wait_for_register_fw(dev_priv,
+				   RING_RESET_CTL(engine->mmio_base),
+				   RESET_CTL_READY_TO_RESET,
+				   RESET_CTL_READY_TO_RESET,
+				   700);
+	if (ret)
+		DRM_ERROR("%s: reset request timeout\n", engine->name);
+
+	return ret;
+}
+
+static void gen8_unrequest_engine_reset(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *dev_priv = engine->dev->dev_private;
+
+	I915_WRITE_FW(RING_RESET_CTL(engine->mmio_base),
+		      _MASKED_BIT_DISABLE(RESET_CTL_REQUEST_RESET));
+}
+
+static int gen8_reset_engines(struct drm_device *dev, unsigned engine_mask)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *engine;
-	int i;
 
-	for_each_ring(engine, dev_priv, i) {
-		I915_WRITE(RING_RESET_CTL(engine->mmio_base),
-			   _MASKED_BIT_ENABLE(RESET_CTL_REQUEST_RESET));
-
-		if (wait_for_register(dev_priv,
-				      RING_RESET_CTL(engine->mmio_base),
-				      RESET_CTL_READY_TO_RESET,
-				      RESET_CTL_READY_TO_RESET,
-				      700)) {
-			DRM_ERROR("%s: reset request timeout\n", engine->name);
+	for_each_engine_masked(engine, dev_priv, engine_mask)
+		if (gen8_request_engine_reset(engine))
 			goto not_ready;
-		}
-	}
 
-	return gen6_do_reset(dev);
+	return gen6_reset_engines(dev, engine_mask);
 
 not_ready:
-	for_each_ring(engine, dev_priv, i)
-		I915_WRITE(RING_RESET_CTL(engine->mmio_base),
-			   _MASKED_BIT_DISABLE(RESET_CTL_REQUEST_RESET));
+	for_each_engine_masked(engine, dev_priv, engine_mask)
+		gen8_unrequest_engine_reset(engine);
 
 	return -EIO;
 }
 
-static int (*intel_get_gpu_reset(struct drm_device *dev))(struct drm_device *)
+static int (*intel_get_gpu_reset(struct drm_device *dev))(struct drm_device *,
+							  unsigned engine_mask)
 {
 	if (!i915.reset)
 		return NULL;
 
 	if (INTEL_INFO(dev)->gen >= 8)
-		return gen8_do_reset;
+		return gen8_reset_engines;
 	else if (INTEL_INFO(dev)->gen >= 6)
-		return gen6_do_reset;
+		return gen6_reset_engines;
 	else if (IS_GEN5(dev))
 		return ironlake_do_reset;
 	else if (IS_G4X(dev))
@@ -1595,10 +1728,10 @@ static int (*intel_get_gpu_reset(struct drm_device *dev))(struct drm_device *)
 		return NULL;
 }
 
-int intel_gpu_reset(struct drm_device *dev)
+int intel_gpu_reset(struct drm_device *dev, unsigned engine_mask)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	int (*reset)(struct drm_device *);
+	int (*reset)(struct drm_device *, unsigned);
 	int ret;
 
 	reset = intel_get_gpu_reset(dev);
@@ -1609,7 +1742,7 @@ int intel_gpu_reset(struct drm_device *dev)
 	 * request may be dropped and never completes (causing -EIO).
 	 */
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
-	ret = reset(dev);
+	ret = reset(dev, engine_mask);
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 
 	return ret;
@@ -1620,6 +1753,25 @@ bool intel_has_gpu_reset(struct drm_device *dev)
 	return intel_get_gpu_reset(dev) != NULL;
 }
 
+int intel_guc_reset(struct drm_i915_private *dev_priv)
+{
+	int ret;
+	unsigned long irqflags;
+
+	if (!i915.enable_guc_submission)
+		return -EINVAL;
+
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+
+	ret = gen6_hw_domain_reset(dev_priv, GEN9_GRDOM_GUC);
+
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+
+	return ret;
+}
+
 bool intel_uncore_unclaimed_mmio(struct drm_i915_private *dev_priv)
 {
 	return check_for_unclaimed_mmio(dev_priv);
@@ -1643,3 +1795,111 @@ intel_uncore_arm_unclaimed_mmio_detection(struct drm_i915_private *dev_priv)
 
 	return false;
 }
+
+static enum forcewake_domains
+intel_uncore_forcewake_for_read(struct drm_i915_private *dev_priv,
+				i915_reg_t reg)
+{
+	enum forcewake_domains fw_domains;
+
+	if (intel_vgpu_active(dev_priv->dev))
+		return 0;
+
+	switch (INTEL_INFO(dev_priv)->gen) {
+	case 9:
+		fw_domains = __gen9_reg_read_fw_domains(i915_mmio_reg_offset(reg));
+		break;
+	case 8:
+		if (IS_CHERRYVIEW(dev_priv))
+			fw_domains = __chv_reg_read_fw_domains(i915_mmio_reg_offset(reg));
+		else
+			fw_domains = __gen6_reg_read_fw_domains(i915_mmio_reg_offset(reg));
+		break;
+	case 7:
+	case 6:
+		if (IS_VALLEYVIEW(dev_priv))
+			fw_domains = __vlv_reg_read_fw_domains(i915_mmio_reg_offset(reg));
+		else
+			fw_domains = __gen6_reg_read_fw_domains(i915_mmio_reg_offset(reg));
+		break;
+	default:
+		MISSING_CASE(INTEL_INFO(dev_priv)->gen);
+	case 5: /* forcewake was introduced with gen6 */
+	case 4:
+	case 3:
+	case 2:
+		return 0;
+	}
+
+	WARN_ON(fw_domains & ~dev_priv->uncore.fw_domains);
+
+	return fw_domains;
+}
+
+static enum forcewake_domains
+intel_uncore_forcewake_for_write(struct drm_i915_private *dev_priv,
+				 i915_reg_t reg)
+{
+	enum forcewake_domains fw_domains;
+
+	if (intel_vgpu_active(dev_priv->dev))
+		return 0;
+
+	switch (INTEL_INFO(dev_priv)->gen) {
+	case 9:
+		fw_domains = __gen9_reg_write_fw_domains(i915_mmio_reg_offset(reg));
+		break;
+	case 8:
+		if (IS_CHERRYVIEW(dev_priv))
+			fw_domains = __chv_reg_write_fw_domains(i915_mmio_reg_offset(reg));
+		else
+			fw_domains = __gen8_reg_write_fw_domains(i915_mmio_reg_offset(reg));
+		break;
+	case 7:
+	case 6:
+		fw_domains = FORCEWAKE_RENDER;
+		break;
+	default:
+		MISSING_CASE(INTEL_INFO(dev_priv)->gen);
+	case 5:
+	case 4:
+	case 3:
+	case 2:
+		return 0;
+	}
+
+	WARN_ON(fw_domains & ~dev_priv->uncore.fw_domains);
+
+	return fw_domains;
+}
+
+/**
+ * intel_uncore_forcewake_for_reg - which forcewake domains are needed to access
+ * 				    a register
+ * @dev_priv: pointer to struct drm_i915_private
+ * @reg: register in question
+ * @op: operation bitmask of FW_REG_READ and/or FW_REG_WRITE
+ *
+ * Returns a set of forcewake domains required to be taken with for example
+ * intel_uncore_forcewake_get for the specified register to be accessible in the
+ * specified mode (read, write or read/write) with raw mmio accessors.
+ *
+ * NOTE: On Gen6 and Gen7 write forcewake domain (FORCEWAKE_RENDER) requires the
+ * callers to do FIFO management on their own or risk losing writes.
+ */
+enum forcewake_domains
+intel_uncore_forcewake_for_reg(struct drm_i915_private *dev_priv,
+			       i915_reg_t reg, unsigned int op)
+{
+	enum forcewake_domains fw_domains = 0;
+
+	WARN_ON(!op);
+
+	if (op & FW_REG_READ)
+		fw_domains = intel_uncore_forcewake_for_read(dev_priv, reg);
+
+	if (op & FW_REG_WRITE)
+		fw_domains |= intel_uncore_forcewake_for_write(dev_priv, reg);
+
+	return fw_domains;
+}
diff --git a/drivers/gpu/drm/i915/intel_vbt_defs.h b/drivers/gpu/drm/i915/intel_vbt_defs.h
new file mode 100644
index 000000000000..9ff1e960d617
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_vbt_defs.h
@@ -0,0 +1,832 @@
+/*
+ * Copyright © 2006-2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ * Authors:
+ *    Eric Anholt <eric@anholt.net>
+ *
+ */
+
+/*
+ * This information is private to VBT parsing in intel_bios.c.
+ *
+ * Please do NOT include anywhere else.
+ */
+#ifndef _INTEL_BIOS_PRIVATE
+#error "intel_vbt_defs.h is private to intel_bios.c"
+#endif
+
+#ifndef _INTEL_VBT_DEFS_H_
+#define _INTEL_VBT_DEFS_H_
+
+#include "intel_bios.h"
+
+/**
+ * struct vbt_header - VBT Header structure
+ * @signature:		VBT signature, always starts with "$VBT"
+ * @version:		Version of this structure
+ * @header_size:	Size of this structure
+ * @vbt_size:		Size of VBT (VBT Header, BDB Header and data blocks)
+ * @vbt_checksum:	Checksum
+ * @reserved0:		Reserved
+ * @bdb_offset:		Offset of &struct bdb_header from beginning of VBT
+ * @aim_offset:		Offsets of add-in data blocks from beginning of VBT
+ */
+struct vbt_header {
+	u8 signature[20];
+	u16 version;
+	u16 header_size;
+	u16 vbt_size;
+	u8 vbt_checksum;
+	u8 reserved0;
+	u32 bdb_offset;
+	u32 aim_offset[4];
+} __packed;
+
+/**
+ * struct bdb_header - BDB Header structure
+ * @signature:		BDB signature "BIOS_DATA_BLOCK"
+ * @version:		Version of the data block definitions
+ * @header_size:	Size of this structure
+ * @bdb_size:		Size of BDB (BDB Header and data blocks)
+ */
+struct bdb_header {
+	u8 signature[16];
+	u16 version;
+	u16 header_size;
+	u16 bdb_size;
+} __packed;
+
+/* strictly speaking, this is a "skip" block, but it has interesting info */
+struct vbios_data {
+	u8 type; /* 0 == desktop, 1 == mobile */
+	u8 relstage;
+	u8 chipset;
+	u8 lvds_present:1;
+	u8 tv_present:1;
+	u8 rsvd2:6; /* finish byte */
+	u8 rsvd3[4];
+	u8 signon[155];
+	u8 copyright[61];
+	u16 code_segment;
+	u8 dos_boot_mode;
+	u8 bandwidth_percent;
+	u8 rsvd4; /* popup memory size */
+	u8 resize_pci_bios;
+	u8 rsvd5; /* is crt already on ddc2 */
+} __packed;
+
+/*
+ * There are several types of BIOS data blocks (BDBs), each block has
+ * an ID and size in the first 3 bytes (ID in first, size in next 2).
+ * Known types are listed below.
+ */
+#define BDB_GENERAL_FEATURES	  1
+#define BDB_GENERAL_DEFINITIONS	  2
+#define BDB_OLD_TOGGLE_LIST	  3
+#define BDB_MODE_SUPPORT_LIST	  4
+#define BDB_GENERIC_MODE_TABLE	  5
+#define BDB_EXT_MMIO_REGS	  6
+#define BDB_SWF_IO		  7
+#define BDB_SWF_MMIO		  8
+#define BDB_PSR			  9
+#define BDB_MODE_REMOVAL_TABLE	 10
+#define BDB_CHILD_DEVICE_TABLE	 11
+#define BDB_DRIVER_FEATURES	 12
+#define BDB_DRIVER_PERSISTENCE	 13
+#define BDB_EXT_TABLE_PTRS	 14
+#define BDB_DOT_CLOCK_OVERRIDE	 15
+#define BDB_DISPLAY_SELECT	 16
+/* 17 rsvd */
+#define BDB_DRIVER_ROTATION	 18
+#define BDB_DISPLAY_REMOVE	 19
+#define BDB_OEM_CUSTOM		 20
+#define BDB_EFP_LIST		 21 /* workarounds for VGA hsync/vsync */
+#define BDB_SDVO_LVDS_OPTIONS	 22
+#define BDB_SDVO_PANEL_DTDS	 23
+#define BDB_SDVO_LVDS_PNP_IDS	 24
+#define BDB_SDVO_LVDS_POWER_SEQ	 25
+#define BDB_TV_OPTIONS		 26
+#define BDB_EDP			 27
+#define BDB_LVDS_OPTIONS	 40
+#define BDB_LVDS_LFP_DATA_PTRS	 41
+#define BDB_LVDS_LFP_DATA	 42
+#define BDB_LVDS_BACKLIGHT	 43
+#define BDB_LVDS_POWER		 44
+#define BDB_MIPI_CONFIG		 52
+#define BDB_MIPI_SEQUENCE	 53
+#define BDB_SKIP		254 /* VBIOS private block, ignore */
+
+struct bdb_general_features {
+        /* bits 1 */
+	u8 panel_fitting:2;
+	u8 flexaim:1;
+	u8 msg_enable:1;
+	u8 clear_screen:3;
+	u8 color_flip:1;
+
+        /* bits 2 */
+	u8 download_ext_vbt:1;
+	u8 enable_ssc:1;
+	u8 ssc_freq:1;
+	u8 enable_lfp_on_override:1;
+	u8 disable_ssc_ddt:1;
+	u8 rsvd7:1;
+	u8 display_clock_mode:1;
+	u8 rsvd8:1; /* finish byte */
+
+        /* bits 3 */
+	u8 disable_smooth_vision:1;
+	u8 single_dvi:1;
+	u8 rsvd9:1;
+	u8 fdi_rx_polarity_inverted:1;
+	u8 rsvd10:4; /* finish byte */
+
+        /* bits 4 */
+	u8 legacy_monitor_detect;
+
+        /* bits 5 */
+	u8 int_crt_support:1;
+	u8 int_tv_support:1;
+	u8 int_efp_support:1;
+	u8 dp_ssc_enb:1;	/* PCH attached eDP supports SSC */
+	u8 dp_ssc_freq:1;	/* SSC freq for PCH attached eDP */
+	u8 rsvd11:3; /* finish byte */
+} __packed;
+
+/* pre-915 */
+#define GPIO_PIN_DVI_LVDS	0x03 /* "DVI/LVDS DDC GPIO pins" */
+#define GPIO_PIN_ADD_I2C	0x05 /* "ADDCARD I2C GPIO pins" */
+#define GPIO_PIN_ADD_DDC	0x04 /* "ADDCARD DDC GPIO pins" */
+#define GPIO_PIN_ADD_DDC_I2C	0x06 /* "ADDCARD DDC/I2C GPIO pins" */
+
+/* Pre 915 */
+#define DEVICE_TYPE_NONE	0x00
+#define DEVICE_TYPE_CRT		0x01
+#define DEVICE_TYPE_TV		0x09
+#define DEVICE_TYPE_EFP		0x12
+#define DEVICE_TYPE_LFP		0x22
+/* On 915+ */
+#define DEVICE_TYPE_CRT_DPMS		0x6001
+#define DEVICE_TYPE_CRT_DPMS_HOTPLUG	0x4001
+#define DEVICE_TYPE_TV_COMPOSITE	0x0209
+#define DEVICE_TYPE_TV_MACROVISION	0x0289
+#define DEVICE_TYPE_TV_RF_COMPOSITE	0x020c
+#define DEVICE_TYPE_TV_SVIDEO_COMPOSITE	0x0609
+#define DEVICE_TYPE_TV_SCART		0x0209
+#define DEVICE_TYPE_TV_CODEC_HOTPLUG_PWR 0x6009
+#define DEVICE_TYPE_EFP_HOTPLUG_PWR	0x6012
+#define DEVICE_TYPE_EFP_DVI_HOTPLUG_PWR	0x6052
+#define DEVICE_TYPE_EFP_DVI_I		0x6053
+#define DEVICE_TYPE_EFP_DVI_D_DUAL	0x6152
+#define DEVICE_TYPE_EFP_DVI_D_HDCP	0x60d2
+#define DEVICE_TYPE_OPENLDI_HOTPLUG_PWR	0x6062
+#define DEVICE_TYPE_OPENLDI_DUALPIX	0x6162
+#define DEVICE_TYPE_LFP_PANELLINK	0x5012
+#define DEVICE_TYPE_LFP_CMOS_PWR	0x5042
+#define DEVICE_TYPE_LFP_LVDS_PWR	0x5062
+#define DEVICE_TYPE_LFP_LVDS_DUAL	0x5162
+#define DEVICE_TYPE_LFP_LVDS_DUAL_HDCP	0x51e2
+
+#define DEVICE_CFG_NONE		0x00
+#define DEVICE_CFG_12BIT_DVOB	0x01
+#define DEVICE_CFG_12BIT_DVOC	0x02
+#define DEVICE_CFG_24BIT_DVOBC	0x09
+#define DEVICE_CFG_24BIT_DVOCB	0x0a
+#define DEVICE_CFG_DUAL_DVOB	0x11
+#define DEVICE_CFG_DUAL_DVOC	0x12
+#define DEVICE_CFG_DUAL_DVOBC	0x13
+#define DEVICE_CFG_DUAL_LINK_DVOBC	0x19
+#define DEVICE_CFG_DUAL_LINK_DVOCB	0x1a
+
+#define DEVICE_WIRE_NONE	0x00
+#define DEVICE_WIRE_DVOB	0x01
+#define DEVICE_WIRE_DVOC	0x02
+#define DEVICE_WIRE_DVOBC	0x03
+#define DEVICE_WIRE_DVOBB	0x05
+#define DEVICE_WIRE_DVOCC	0x06
+#define DEVICE_WIRE_DVOB_MASTER 0x0d
+#define DEVICE_WIRE_DVOC_MASTER 0x0e
+
+#define DEVICE_PORT_DVOA	0x00 /* none on 845+ */
+#define DEVICE_PORT_DVOB	0x01
+#define DEVICE_PORT_DVOC	0x02
+
+/*
+ * We used to keep this struct but without any version control. We should avoid
+ * using it in the future, but it should be safe to keep using it in the old
+ * code. Do not change; we rely on its size.
+ */
+struct old_child_dev_config {
+	u16 handle;
+	u16 device_type;
+	u8  device_id[10]; /* ascii string */
+	u16 addin_offset;
+	u8  dvo_port; /* See Device_PORT_* above */
+	u8  i2c_pin;
+	u8  slave_addr;
+	u8  ddc_pin;
+	u16 edid_ptr;
+	u8  dvo_cfg; /* See DEVICE_CFG_* above */
+	u8  dvo2_port;
+	u8  i2c2_pin;
+	u8  slave2_addr;
+	u8  ddc2_pin;
+	u8  capabilities;
+	u8  dvo_wiring;/* See DEVICE_WIRE_* above */
+	u8  dvo2_wiring;
+	u16 extended_type;
+	u8  dvo_function;
+} __packed;
+
+/* This one contains field offsets that are known to be common for all BDB
+ * versions. Notice that the meaning of the contents contents may still change,
+ * but at least the offsets are consistent. */
+
+struct common_child_dev_config {
+	u16 handle;
+	u16 device_type;
+	u8 not_common1[12];
+	u8 dvo_port;
+	u8 not_common2[2];
+	u8 ddc_pin;
+	u16 edid_ptr;
+	u8 dvo_cfg; /* See DEVICE_CFG_* above */
+	u8 efp_routed:1;
+	u8 lane_reversal:1;
+	u8 lspcon:1;
+	u8 iboost:1;
+	u8 hpd_invert:1;
+	u8 flag_reserved:3;
+	u8 hdmi_support:1;
+	u8 dp_support:1;
+	u8 tmds_support:1;
+	u8 support_reserved:5;
+	u8 not_common3[12];
+	u8 iboost_level;
+} __packed;
+
+
+/* This field changes depending on the BDB version, so the most reliable way to
+ * read it is by checking the BDB version and reading the raw pointer. */
+union child_device_config {
+	/* This one is safe to be used anywhere, but the code should still check
+	 * the BDB version. */
+	u8 raw[33];
+	/* This one should only be kept for legacy code. */
+	struct old_child_dev_config old;
+	/* This one should also be safe to use anywhere, even without version
+	 * checks. */
+	struct common_child_dev_config common;
+} __packed;
+
+struct bdb_general_definitions {
+	/* DDC GPIO */
+	u8 crt_ddc_gmbus_pin;
+
+	/* DPMS bits */
+	u8 dpms_acpi:1;
+	u8 skip_boot_crt_detect:1;
+	u8 dpms_aim:1;
+	u8 rsvd1:5; /* finish byte */
+
+	/* boot device bits */
+	u8 boot_display[2];
+	u8 child_dev_size;
+
+	/*
+	 * Device info:
+	 * If TV is present, it'll be at devices[0].
+	 * LVDS will be next, either devices[0] or [1], if present.
+	 * On some platforms the number of device is 6. But could be as few as
+	 * 4 if both TV and LVDS are missing.
+	 * And the device num is related with the size of general definition
+	 * block. It is obtained by using the following formula:
+	 * number = (block_size - sizeof(bdb_general_definitions))/
+	 *	     defs->child_dev_size;
+	 */
+	uint8_t devices[0];
+} __packed;
+
+/* Mask for DRRS / Panel Channel / SSC / BLT control bits extraction */
+#define MODE_MASK		0x3
+
+struct bdb_lvds_options {
+	u8 panel_type;
+	u8 rsvd1;
+	/* LVDS capabilities, stored in a dword */
+	u8 pfit_mode:2;
+	u8 pfit_text_mode_enhanced:1;
+	u8 pfit_gfx_mode_enhanced:1;
+	u8 pfit_ratio_auto:1;
+	u8 pixel_dither:1;
+	u8 lvds_edid:1;
+	u8 rsvd2:1;
+	u8 rsvd4;
+	/* LVDS Panel channel bits stored here */
+	u32 lvds_panel_channel_bits;
+	/* LVDS SSC (Spread Spectrum Clock) bits stored here. */
+	u16 ssc_bits;
+	u16 ssc_freq;
+	u16 ssc_ddt;
+	/* Panel color depth defined here */
+	u16 panel_color_depth;
+	/* LVDS panel type bits stored here */
+	u32 dps_panel_type_bits;
+	/* LVDS backlight control type bits stored here */
+	u32 blt_control_type_bits;
+} __packed;
+
+/* LFP pointer table contains entries to the struct below */
+struct bdb_lvds_lfp_data_ptr {
+	u16 fp_timing_offset; /* offsets are from start of bdb */
+	u8 fp_table_size;
+	u16 dvo_timing_offset;
+	u8 dvo_table_size;
+	u16 panel_pnp_id_offset;
+	u8 pnp_table_size;
+} __packed;
+
+struct bdb_lvds_lfp_data_ptrs {
+	u8 lvds_entries; /* followed by one or more lvds_data_ptr structs */
+	struct bdb_lvds_lfp_data_ptr ptr[16];
+} __packed;
+
+/* LFP data has 3 blocks per entry */
+struct lvds_fp_timing {
+	u16 x_res;
+	u16 y_res;
+	u32 lvds_reg;
+	u32 lvds_reg_val;
+	u32 pp_on_reg;
+	u32 pp_on_reg_val;
+	u32 pp_off_reg;
+	u32 pp_off_reg_val;
+	u32 pp_cycle_reg;
+	u32 pp_cycle_reg_val;
+	u32 pfit_reg;
+	u32 pfit_reg_val;
+	u16 terminator;
+} __packed;
+
+struct lvds_dvo_timing {
+	u16 clock;		/**< In 10khz */
+	u8 hactive_lo;
+	u8 hblank_lo;
+	u8 hblank_hi:4;
+	u8 hactive_hi:4;
+	u8 vactive_lo;
+	u8 vblank_lo;
+	u8 vblank_hi:4;
+	u8 vactive_hi:4;
+	u8 hsync_off_lo;
+	u8 hsync_pulse_width;
+	u8 vsync_pulse_width:4;
+	u8 vsync_off:4;
+	u8 rsvd0:6;
+	u8 hsync_off_hi:2;
+	u8 h_image;
+	u8 v_image;
+	u8 max_hv;
+	u8 h_border;
+	u8 v_border;
+	u8 rsvd1:3;
+	u8 digital:2;
+	u8 vsync_positive:1;
+	u8 hsync_positive:1;
+	u8 rsvd2:1;
+} __packed;
+
+struct lvds_pnp_id {
+	u16 mfg_name;
+	u16 product_code;
+	u32 serial;
+	u8 mfg_week;
+	u8 mfg_year;
+} __packed;
+
+struct bdb_lvds_lfp_data_entry {
+	struct lvds_fp_timing fp_timing;
+	struct lvds_dvo_timing dvo_timing;
+	struct lvds_pnp_id pnp_id;
+} __packed;
+
+struct bdb_lvds_lfp_data {
+	struct bdb_lvds_lfp_data_entry data[16];
+} __packed;
+
+#define BDB_BACKLIGHT_TYPE_NONE	0
+#define BDB_BACKLIGHT_TYPE_PWM	2
+
+struct bdb_lfp_backlight_data_entry {
+	u8 type:2;
+	u8 active_low_pwm:1;
+	u8 obsolete1:5;
+	u16 pwm_freq_hz;
+	u8 min_brightness;
+	u8 obsolete2;
+	u8 obsolete3;
+} __packed;
+
+struct bdb_lfp_backlight_data {
+	u8 entry_size;
+	struct bdb_lfp_backlight_data_entry data[16];
+	u8 level[16];
+} __packed;
+
+struct aimdb_header {
+	char signature[16];
+	char oem_device[20];
+	u16 aimdb_version;
+	u16 aimdb_header_size;
+	u16 aimdb_size;
+} __packed;
+
+struct aimdb_block {
+	u8 aimdb_id;
+	u16 aimdb_size;
+} __packed;
+
+struct vch_panel_data {
+	u16 fp_timing_offset;
+	u8 fp_timing_size;
+	u16 dvo_timing_offset;
+	u8 dvo_timing_size;
+	u16 text_fitting_offset;
+	u8 text_fitting_size;
+	u16 graphics_fitting_offset;
+	u8 graphics_fitting_size;
+} __packed;
+
+struct vch_bdb_22 {
+	struct aimdb_block aimdb_block;
+	struct vch_panel_data panels[16];
+} __packed;
+
+struct bdb_sdvo_lvds_options {
+	u8 panel_backlight;
+	u8 h40_set_panel_type;
+	u8 panel_type;
+	u8 ssc_clk_freq;
+	u16 als_low_trip;
+	u16 als_high_trip;
+	u8 sclalarcoeff_tab_row_num;
+	u8 sclalarcoeff_tab_row_size;
+	u8 coefficient[8];
+	u8 panel_misc_bits_1;
+	u8 panel_misc_bits_2;
+	u8 panel_misc_bits_3;
+	u8 panel_misc_bits_4;
+} __packed;
+
+
+#define BDB_DRIVER_FEATURE_NO_LVDS		0
+#define BDB_DRIVER_FEATURE_INT_LVDS		1
+#define BDB_DRIVER_FEATURE_SDVO_LVDS		2
+#define BDB_DRIVER_FEATURE_EDP			3
+
+struct bdb_driver_features {
+	u8 boot_dev_algorithm:1;
+	u8 block_display_switch:1;
+	u8 allow_display_switch:1;
+	u8 hotplug_dvo:1;
+	u8 dual_view_zoom:1;
+	u8 int15h_hook:1;
+	u8 sprite_in_clone:1;
+	u8 primary_lfp_id:1;
+
+	u16 boot_mode_x;
+	u16 boot_mode_y;
+	u8 boot_mode_bpp;
+	u8 boot_mode_refresh;
+
+	u16 enable_lfp_primary:1;
+	u16 selective_mode_pruning:1;
+	u16 dual_frequency:1;
+	u16 render_clock_freq:1; /* 0: high freq; 1: low freq */
+	u16 nt_clone_support:1;
+	u16 power_scheme_ui:1; /* 0: CUI; 1: 3rd party */
+	u16 sprite_display_assign:1; /* 0: secondary; 1: primary */
+	u16 cui_aspect_scaling:1;
+	u16 preserve_aspect_ratio:1;
+	u16 sdvo_device_power_down:1;
+	u16 crt_hotplug:1;
+	u16 lvds_config:2;
+	u16 tv_hotplug:1;
+	u16 hdmi_config:2;
+
+	u8 static_display:1;
+	u8 reserved2:7;
+	u16 legacy_crt_max_x;
+	u16 legacy_crt_max_y;
+	u8 legacy_crt_max_refresh;
+
+	u8 hdmi_termination;
+	u8 custom_vbt_version;
+	/* Driver features data block */
+	u16 rmpm_enabled:1;
+	u16 s2ddt_enabled:1;
+	u16 dpst_enabled:1;
+	u16 bltclt_enabled:1;
+	u16 adb_enabled:1;
+	u16 drrs_enabled:1;
+	u16 grs_enabled:1;
+	u16 gpmt_enabled:1;
+	u16 tbt_enabled:1;
+	u16 psr_enabled:1;
+	u16 ips_enabled:1;
+	u16 reserved3:4;
+	u16 pc_feature_valid:1;
+} __packed;
+
+#define EDP_18BPP	0
+#define EDP_24BPP	1
+#define EDP_30BPP	2
+#define EDP_RATE_1_62	0
+#define EDP_RATE_2_7	1
+#define EDP_LANE_1	0
+#define EDP_LANE_2	1
+#define EDP_LANE_4	3
+#define EDP_PREEMPHASIS_NONE	0
+#define EDP_PREEMPHASIS_3_5dB	1
+#define EDP_PREEMPHASIS_6dB	2
+#define EDP_PREEMPHASIS_9_5dB	3
+#define EDP_VSWING_0_4V		0
+#define EDP_VSWING_0_6V		1
+#define EDP_VSWING_0_8V		2
+#define EDP_VSWING_1_2V		3
+
+
+struct edp_link_params {
+	u8 rate:4;
+	u8 lanes:4;
+	u8 preemphasis:4;
+	u8 vswing:4;
+} __packed;
+
+struct bdb_edp {
+	struct edp_power_seq power_seqs[16];
+	u32 color_depth;
+	struct edp_link_params link_params[16];
+	u32 sdrrs_msa_timing_delay;
+
+	/* ith bit indicates enabled/disabled for (i+1)th panel */
+	u16 edp_s3d_feature;
+	u16 edp_t3_optimization;
+	u64 edp_vswing_preemph;		/* v173 */
+} __packed;
+
+struct psr_table {
+	/* Feature bits */
+	u8 full_link:1;
+	u8 require_aux_to_wakeup:1;
+	u8 feature_bits_rsvd:6;
+
+	/* Wait times */
+	u8 idle_frames:4;
+	u8 lines_to_wait:3;
+	u8 wait_times_rsvd:1;
+
+	/* TP wake up time in multiple of 100 */
+	u16 tp1_wakeup_time;
+	u16 tp2_tp3_wakeup_time;
+} __packed;
+
+struct bdb_psr {
+	struct psr_table psr_table[16];
+} __packed;
+
+/*
+ * Driver<->VBIOS interaction occurs through scratch bits in
+ * GR18 & SWF*.
+ */
+
+/* GR18 bits are set on display switch and hotkey events */
+#define GR18_DRIVER_SWITCH_EN	(1<<7) /* 0: VBIOS control, 1: driver control */
+#define GR18_HOTKEY_MASK	0x78 /* See also SWF4 15:0 */
+#define   GR18_HK_NONE		(0x0<<3)
+#define   GR18_HK_LFP_STRETCH	(0x1<<3)
+#define   GR18_HK_TOGGLE_DISP	(0x2<<3)
+#define   GR18_HK_DISP_SWITCH	(0x4<<3) /* see SWF14 15:0 for what to enable */
+#define   GR18_HK_POPUP_DISABLED (0x6<<3)
+#define   GR18_HK_POPUP_ENABLED	(0x7<<3)
+#define   GR18_HK_PFIT		(0x8<<3)
+#define   GR18_HK_APM_CHANGE	(0xa<<3)
+#define   GR18_HK_MULTIPLE	(0xc<<3)
+#define GR18_USER_INT_EN	(1<<2)
+#define GR18_A0000_FLUSH_EN	(1<<1)
+#define GR18_SMM_EN		(1<<0)
+
+/* Set by driver, cleared by VBIOS */
+#define SWF00_YRES_SHIFT	16
+#define SWF00_XRES_SHIFT	0
+#define SWF00_RES_MASK		0xffff
+
+/* Set by VBIOS at boot time and driver at runtime */
+#define SWF01_TV2_FORMAT_SHIFT	8
+#define SWF01_TV1_FORMAT_SHIFT	0
+#define SWF01_TV_FORMAT_MASK	0xffff
+
+#define SWF10_VBIOS_BLC_I2C_EN	(1<<29)
+#define SWF10_GTT_OVERRIDE_EN	(1<<28)
+#define SWF10_LFP_DPMS_OVR	(1<<27) /* override DPMS on display switch */
+#define SWF10_ACTIVE_TOGGLE_LIST_MASK (7<<24)
+#define   SWF10_OLD_TOGGLE	0x0
+#define   SWF10_TOGGLE_LIST_1	0x1
+#define   SWF10_TOGGLE_LIST_2	0x2
+#define   SWF10_TOGGLE_LIST_3	0x3
+#define   SWF10_TOGGLE_LIST_4	0x4
+#define SWF10_PANNING_EN	(1<<23)
+#define SWF10_DRIVER_LOADED	(1<<22)
+#define SWF10_EXTENDED_DESKTOP	(1<<21)
+#define SWF10_EXCLUSIVE_MODE	(1<<20)
+#define SWF10_OVERLAY_EN	(1<<19)
+#define SWF10_PLANEB_HOLDOFF	(1<<18)
+#define SWF10_PLANEA_HOLDOFF	(1<<17)
+#define SWF10_VGA_HOLDOFF	(1<<16)
+#define SWF10_ACTIVE_DISP_MASK	0xffff
+#define   SWF10_PIPEB_LFP2	(1<<15)
+#define   SWF10_PIPEB_EFP2	(1<<14)
+#define   SWF10_PIPEB_TV2	(1<<13)
+#define   SWF10_PIPEB_CRT2	(1<<12)
+#define   SWF10_PIPEB_LFP	(1<<11)
+#define   SWF10_PIPEB_EFP	(1<<10)
+#define   SWF10_PIPEB_TV	(1<<9)
+#define   SWF10_PIPEB_CRT	(1<<8)
+#define   SWF10_PIPEA_LFP2	(1<<7)
+#define   SWF10_PIPEA_EFP2	(1<<6)
+#define   SWF10_PIPEA_TV2	(1<<5)
+#define   SWF10_PIPEA_CRT2	(1<<4)
+#define   SWF10_PIPEA_LFP	(1<<3)
+#define   SWF10_PIPEA_EFP	(1<<2)
+#define   SWF10_PIPEA_TV	(1<<1)
+#define   SWF10_PIPEA_CRT	(1<<0)
+
+#define SWF11_MEMORY_SIZE_SHIFT	16
+#define SWF11_SV_TEST_EN	(1<<15)
+#define SWF11_IS_AGP		(1<<14)
+#define SWF11_DISPLAY_HOLDOFF	(1<<13)
+#define SWF11_DPMS_REDUCED	(1<<12)
+#define SWF11_IS_VBE_MODE	(1<<11)
+#define SWF11_PIPEB_ACCESS	(1<<10) /* 0 here means pipe a */
+#define SWF11_DPMS_MASK		0x07
+#define   SWF11_DPMS_OFF	(1<<2)
+#define   SWF11_DPMS_SUSPEND	(1<<1)
+#define   SWF11_DPMS_STANDBY	(1<<0)
+#define   SWF11_DPMS_ON		0
+
+#define SWF14_GFX_PFIT_EN	(1<<31)
+#define SWF14_TEXT_PFIT_EN	(1<<30)
+#define SWF14_LID_STATUS_CLOSED	(1<<29) /* 0 here means open */
+#define SWF14_POPUP_EN		(1<<28)
+#define SWF14_DISPLAY_HOLDOFF	(1<<27)
+#define SWF14_DISP_DETECT_EN	(1<<26)
+#define SWF14_DOCKING_STATUS_DOCKED (1<<25) /* 0 here means undocked */
+#define SWF14_DRIVER_STATUS	(1<<24)
+#define SWF14_OS_TYPE_WIN9X	(1<<23)
+#define SWF14_OS_TYPE_WINNT	(1<<22)
+/* 21:19 rsvd */
+#define SWF14_PM_TYPE_MASK	0x00070000
+#define   SWF14_PM_ACPI_VIDEO	(0x4 << 16)
+#define   SWF14_PM_ACPI		(0x3 << 16)
+#define   SWF14_PM_APM_12	(0x2 << 16)
+#define   SWF14_PM_APM_11	(0x1 << 16)
+#define SWF14_HK_REQUEST_MASK	0x0000ffff /* see GR18 6:3 for event type */
+          /* if GR18 indicates a display switch */
+#define   SWF14_DS_PIPEB_LFP2_EN (1<<15)
+#define   SWF14_DS_PIPEB_EFP2_EN (1<<14)
+#define   SWF14_DS_PIPEB_TV2_EN  (1<<13)
+#define   SWF14_DS_PIPEB_CRT2_EN (1<<12)
+#define   SWF14_DS_PIPEB_LFP_EN  (1<<11)
+#define   SWF14_DS_PIPEB_EFP_EN  (1<<10)
+#define   SWF14_DS_PIPEB_TV_EN   (1<<9)
+#define   SWF14_DS_PIPEB_CRT_EN  (1<<8)
+#define   SWF14_DS_PIPEA_LFP2_EN (1<<7)
+#define   SWF14_DS_PIPEA_EFP2_EN (1<<6)
+#define   SWF14_DS_PIPEA_TV2_EN  (1<<5)
+#define   SWF14_DS_PIPEA_CRT2_EN (1<<4)
+#define   SWF14_DS_PIPEA_LFP_EN  (1<<3)
+#define   SWF14_DS_PIPEA_EFP_EN  (1<<2)
+#define   SWF14_DS_PIPEA_TV_EN   (1<<1)
+#define   SWF14_DS_PIPEA_CRT_EN  (1<<0)
+          /* if GR18 indicates a panel fitting request */
+#define   SWF14_PFIT_EN		(1<<0) /* 0 means disable */
+          /* if GR18 indicates an APM change request */
+#define   SWF14_APM_HIBERNATE	0x4
+#define   SWF14_APM_SUSPEND	0x3
+#define   SWF14_APM_STANDBY	0x1
+#define   SWF14_APM_RESTORE	0x0
+
+/* Add the device class for LFP, TV, HDMI */
+#define	 DEVICE_TYPE_INT_LFP	0x1022
+#define	 DEVICE_TYPE_INT_TV	0x1009
+#define	 DEVICE_TYPE_HDMI	0x60D2
+#define	 DEVICE_TYPE_DP		0x68C6
+#define	 DEVICE_TYPE_eDP	0x78C6
+
+#define  DEVICE_TYPE_CLASS_EXTENSION	(1 << 15)
+#define  DEVICE_TYPE_POWER_MANAGEMENT	(1 << 14)
+#define  DEVICE_TYPE_HOTPLUG_SIGNALING	(1 << 13)
+#define  DEVICE_TYPE_INTERNAL_CONNECTOR	(1 << 12)
+#define  DEVICE_TYPE_NOT_HDMI_OUTPUT	(1 << 11)
+#define  DEVICE_TYPE_MIPI_OUTPUT	(1 << 10)
+#define  DEVICE_TYPE_COMPOSITE_OUTPUT	(1 << 9)
+#define  DEVICE_TYPE_DUAL_CHANNEL	(1 << 8)
+#define  DEVICE_TYPE_HIGH_SPEED_LINK	(1 << 6)
+#define  DEVICE_TYPE_LVDS_SINGALING	(1 << 5)
+#define  DEVICE_TYPE_TMDS_DVI_SIGNALING	(1 << 4)
+#define  DEVICE_TYPE_VIDEO_SIGNALING	(1 << 3)
+#define  DEVICE_TYPE_DISPLAYPORT_OUTPUT	(1 << 2)
+#define  DEVICE_TYPE_DIGITAL_OUTPUT	(1 << 1)
+#define  DEVICE_TYPE_ANALOG_OUTPUT	(1 << 0)
+
+/*
+ * Bits we care about when checking for DEVICE_TYPE_eDP
+ * Depending on the system, the other bits may or may not
+ * be set for eDP outputs.
+ */
+#define DEVICE_TYPE_eDP_BITS \
+	(DEVICE_TYPE_INTERNAL_CONNECTOR | \
+	 DEVICE_TYPE_MIPI_OUTPUT | \
+	 DEVICE_TYPE_COMPOSITE_OUTPUT | \
+	 DEVICE_TYPE_DUAL_CHANNEL | \
+	 DEVICE_TYPE_LVDS_SINGALING | \
+	 DEVICE_TYPE_TMDS_DVI_SIGNALING | \
+	 DEVICE_TYPE_VIDEO_SIGNALING | \
+	 DEVICE_TYPE_DISPLAYPORT_OUTPUT | \
+	 DEVICE_TYPE_ANALOG_OUTPUT)
+
+/* define the DVO port for HDMI output type */
+#define		DVO_B		1
+#define		DVO_C		2
+#define		DVO_D		3
+
+/* Possible values for the "DVO Port" field for versions >= 155: */
+#define DVO_PORT_HDMIA	0
+#define DVO_PORT_HDMIB	1
+#define DVO_PORT_HDMIC	2
+#define DVO_PORT_HDMID	3
+#define DVO_PORT_LVDS	4
+#define DVO_PORT_TV	5
+#define DVO_PORT_CRT	6
+#define DVO_PORT_DPB	7
+#define DVO_PORT_DPC	8
+#define DVO_PORT_DPD	9
+#define DVO_PORT_DPA	10
+#define DVO_PORT_DPE	11
+#define DVO_PORT_HDMIE	12
+#define DVO_PORT_MIPIA	21
+#define DVO_PORT_MIPIB	22
+#define DVO_PORT_MIPIC	23
+#define DVO_PORT_MIPID	24
+
+/* Block 52 contains MIPI configuration block
+ * 6 * bdb_mipi_config, followed by 6 pps data block
+ * block below
+ */
+#define MAX_MIPI_CONFIGURATIONS	6
+
+struct bdb_mipi_config {
+	struct mipi_config config[MAX_MIPI_CONFIGURATIONS];
+	struct mipi_pps_data pps[MAX_MIPI_CONFIGURATIONS];
+} __packed;
+
+/* Block 53 contains MIPI sequences as needed by the panel
+ * for enabling it. This block can be variable in size and
+ * can be maximum of 6 blocks
+ */
+struct bdb_mipi_sequence {
+	u8 version;
+	u8 data[0];
+} __packed;
+
+enum mipi_gpio_pin_index {
+	MIPI_GPIO_UNDEFINED = 0,
+	MIPI_GPIO_PANEL_ENABLE,
+	MIPI_GPIO_BL_ENABLE,
+	MIPI_GPIO_PWM_ENABLE,
+	MIPI_GPIO_RESET_N,
+	MIPI_GPIO_PWR_DOWN_R,
+	MIPI_GPIO_STDBY_RST_N,
+	MIPI_GPIO_MAX
+};
+
+#endif /* _INTEL_VBT_DEFS_H_ */