Merge tag 'drm-intel-next-2017-09-07' of git://anongit.freedesktop.org/git/drm-intel into drm-next

Getting started with v4.15 features:

- Cannonlake workarounds (Rodrigo, Oscar)
- Infoframe refactoring and fixes to enable infoframes for DP (Ville)
- VBT definition updates (Jani)
- Sparse warning fixes (Ville, Chris)
- Crtc state usage fixes and cleanups (Ville)
- DP vswing, pre-emph and buffer translation refactoring and fixes (Rodrigo)
- Prevent IPS from interfering with CRC capture (Ville, Marta)
- Enable Mesa to advertise ARB_timer_query (Nanley)
- Refactor GT number into intel_device_info (Lionel)
- Avoid eDP DP AUX CH timeouts harder (Manasi)
- CDCLK check improvements (Ville)
- Restore GPU clock boost on missed pageflip vblanks (Chris)
- Fence register reservation API for vGPU (Changbin)
- First batch of CCS fixes (Ville)
- Finally, numerous GEM fixes, cleanups and improvements (Chris)

* tag 'drm-intel-next-2017-09-07' of git://anongit.freedesktop.org/git/drm-intel: (100 commits)
  drm/i915: Update DRIVER_DATE to 20170907
  drm/i915/cnl: WaThrottleEUPerfToAvoidTDBackPressure:cnl(pre-prod)
  drm/i915: Lift has-pinned-pages assert to caller of ____i915_gem_object_get_pages
  drm/i915: Display WA #1133 WaFbcSkipSegments:cnl, glk
  drm/i915/cnl: Allow the reg_read ioctl to read the RCS TIMESTAMP register
  drm/i915: Move device_info.has_snoop into the static tables
  drm/i915: Disable MI_STORE_DATA_IMM for i915g/i915gm
  drm/i915: Re-enable GTT following a device reset
  drm/i915/cnp: Wa 1181: Fix Backlight issue
  drm/i915: Annotate user relocs with __user
  drm/i915: Constify load detect mode
  drm/i915/perf: Remove __user from u64 in drm_i915_perf_oa_config
  drm/i915: Silence sparse by using gfp_t
  drm/i915: io unmap functions want __iomem
  drm/i915: Add __rcu to radix tree slot pointer
  drm/i915: Wake up the device for the fbdev setup
  drm/i915: Add interface to reserve fence registers for vGPU
  drm/i915: Use correct path to trace include
  drm/i915: Fix the missing PPAT cache attributes on CNL
  drm/i915: Fix enum pipe vs. enum transcoder for the PCH transcoder
  ...

drivers/gpu/drm/i915/Makefile

@@ -150,5 +150,3 @@ endif
 i915-y += intel_lpe_audio.o
 
 obj-$(CONFIG_DRM_I915) += i915.o
-
-CFLAGS_i915_trace_points.o := -I$(src)

drivers/gpu/drm/i915/gvt/aperture_gm.c

@@ -173,8 +173,8 @@ static void free_vgpu_fence(struct intel_vgpu *vgpu)
 	_clear_vgpu_fence(vgpu);
 	for (i = 0; i < vgpu_fence_sz(vgpu); i++) {
 		reg = vgpu->fence.regs[i];
-		list_add_tail(&reg->link,
-			      &dev_priv->mm.fence_list);
+		i915_unreserve_fence(reg);
+		vgpu->fence.regs[i] = NULL;
 	}
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 
@@ -187,24 +187,19 @@ static int alloc_vgpu_fence(struct intel_vgpu *vgpu)
 	struct drm_i915_private *dev_priv = gvt->dev_priv;
 	struct drm_i915_fence_reg *reg;
 	int i;
-	struct list_head *pos, *q;
 
 	intel_runtime_pm_get(dev_priv);
 
 	/* Request fences from host */
 	mutex_lock(&dev_priv->drm.struct_mutex);
-	i = 0;
-	list_for_each_safe(pos, q, &dev_priv->mm.fence_list) {
-		reg = list_entry(pos, struct drm_i915_fence_reg, link);
-		if (reg->pin_count || reg->vma)
-			continue;
-		list_del(pos);
+
+	for (i = 0; i < vgpu_fence_sz(vgpu); i++) {
+		reg = i915_reserve_fence(dev_priv);
+		if (IS_ERR(reg))
+			goto out_free_fence;
+
 		vgpu->fence.regs[i] = reg;
-		if (++i == vgpu_fence_sz(vgpu))
-			break;
 	}
-	if (i != vgpu_fence_sz(vgpu))
-		goto out_free_fence;
 
 	_clear_vgpu_fence(vgpu);
 
@@ -212,13 +207,14 @@ static int alloc_vgpu_fence(struct intel_vgpu *vgpu)
 	intel_runtime_pm_put(dev_priv);
 	return 0;
 out_free_fence:
+	gvt_vgpu_err("Failed to alloc fences\n");
 	/* Return fences to host, if fail */
 	for (i = 0; i < vgpu_fence_sz(vgpu); i++) {
 		reg = vgpu->fence.regs[i];
 		if (!reg)
 			continue;
-		list_add_tail(&reg->link,
-			      &dev_priv->mm.fence_list);
+		i915_unreserve_fence(reg);
+		vgpu->fence.regs[i] = NULL;
 	}
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 	intel_runtime_pm_put(dev_priv);

drivers/gpu/drm/i915/i915_drv.c

@@ -239,7 +239,8 @@ static void intel_detect_pch(struct drm_i915_private *dev_priv)
 				dev_priv->pch_type = PCH_KBP;
 				DRM_DEBUG_KMS("Found Kaby Lake PCH (KBP)\n");
 				WARN_ON(!IS_SKYLAKE(dev_priv) &&
-					!IS_KABYLAKE(dev_priv));
+					!IS_KABYLAKE(dev_priv) &&
+					!IS_COFFEELAKE(dev_priv));
 			} else if (id == INTEL_PCH_CNP_DEVICE_ID_TYPE) {
 				dev_priv->pch_type = PCH_CNP;
 				DRM_DEBUG_KMS("Found Cannon Lake PCH (CNP)\n");

drivers/gpu/drm/i915/i915_drv.h

@@ -80,8 +80,8 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20170818"
-#define DRIVER_TIMESTAMP	1503088845
+#define DRIVER_DATE		"20170907"
+#define DRIVER_TIMESTAMP	1504772900
 
 /* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and
  * WARN_ON()) for hw state sanity checks to check for unexpected conditions
@@ -569,6 +569,24 @@ struct i915_hotplug {
 	     (__i)++) \
 		for_each_if (plane_state)
 
+#define for_each_new_intel_crtc_in_state(__state, crtc, new_crtc_state, __i) \
+	for ((__i) = 0; \
+	     (__i) < (__state)->base.dev->mode_config.num_crtc && \
+		     ((crtc) = to_intel_crtc((__state)->base.crtcs[__i].ptr), \
+		      (new_crtc_state) = to_intel_crtc_state((__state)->base.crtcs[__i].new_state), 1); \
+	     (__i)++) \
+		for_each_if (crtc)
+
+
+#define for_each_oldnew_intel_plane_in_state(__state, plane, old_plane_state, new_plane_state, __i) \
+	for ((__i) = 0; \
+	     (__i) < (__state)->base.dev->mode_config.num_total_plane && \
+		     ((plane) = to_intel_plane((__state)->base.planes[__i].ptr), \
+		      (old_plane_state) = to_intel_plane_state((__state)->base.planes[__i].old_state), \
+		      (new_plane_state) = to_intel_plane_state((__state)->base.planes[__i].new_state), 1); \
+	     (__i)++) \
+		for_each_if (plane)
+
 struct drm_i915_private;
 struct i915_mm_struct;
 struct i915_mmu_object;
@@ -841,6 +859,7 @@ struct intel_device_info {
 	u8 gen;
 	u16 gen_mask;
 	enum intel_platform platform;
+	u8 gt; /* GT number, 0 if undefined */
 	u8 ring_mask; /* Rings supported by the HW */
 	u8 num_rings;
 #define DEFINE_FLAG(name) u8 name:1
@@ -1106,6 +1125,7 @@ struct intel_fbc {
 		} fb;
 
 		int cfb_size;
+		unsigned int gen9_wa_cfb_stride;
 	} params;
 
 	struct intel_fbc_work {
@@ -1464,6 +1484,11 @@ struct i915_gem_mm {
 	struct llist_head free_list;
 	struct work_struct free_work;
 
+	/**
+	 * Small stash of WC pages
+	 */
+	struct pagevec wc_stash;
+
 	/** Usable portion of the GTT for GEM */
 	dma_addr_t stolen_base; /* limited to low memory (32-bit) */
 
@@ -1717,7 +1742,7 @@ struct intel_vbt_data {
 	int crt_ddc_pin;
 
 	int child_dev_num;
-	union child_device_config *child_dev;
+	struct child_device_config *child_dev;
 
 	struct ddi_vbt_port_info ddi_port_info[I915_MAX_PORTS];
 	struct sdvo_device_mapping sdvo_mappings[2];
@@ -2328,7 +2353,8 @@ struct drm_i915_private {
 	struct mutex dpll_lock;
 
 	unsigned int active_crtcs;
-	unsigned int min_pixclk[I915_MAX_PIPES];
+	/* minimum acceptable cdclk for each pipe */
+	int min_cdclk[I915_MAX_PIPES];
 
 	int dpio_phy_iosf_port[I915_NUM_PHYS_VLV];
 
@@ -2861,9 +2887,8 @@ intel_info(const struct drm_i915_private *dev_priv)
 #define IS_G33(dev_priv)	((dev_priv)->info.platform == INTEL_G33)
 #define IS_IRONLAKE_M(dev_priv)	(INTEL_DEVID(dev_priv) == 0x0046)
 #define IS_IVYBRIDGE(dev_priv)	((dev_priv)->info.platform == INTEL_IVYBRIDGE)
-#define IS_IVB_GT1(dev_priv)	(INTEL_DEVID(dev_priv) == 0x0156 || \
-				 INTEL_DEVID(dev_priv) == 0x0152 || \
-				 INTEL_DEVID(dev_priv) == 0x015a)
+#define IS_IVB_GT1(dev_priv)	(IS_IVYBRIDGE(dev_priv) && \
+				 (dev_priv)->info.gt == 1)
 #define IS_VALLEYVIEW(dev_priv)	((dev_priv)->info.platform == INTEL_VALLEYVIEW)
 #define IS_CHERRYVIEW(dev_priv)	((dev_priv)->info.platform == INTEL_CHERRYVIEW)
 #define IS_HASWELL(dev_priv)	((dev_priv)->info.platform == INTEL_HASWELL)
@@ -2885,11 +2910,11 @@ intel_info(const struct drm_i915_private *dev_priv)
 #define IS_BDW_ULX(dev_priv)	(IS_BROADWELL(dev_priv) && \
 				 (INTEL_DEVID(dev_priv) & 0xf) == 0xe)
 #define IS_BDW_GT3(dev_priv)	(IS_BROADWELL(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x0020)
+				 (dev_priv)->info.gt == 3)
 #define IS_HSW_ULT(dev_priv)	(IS_HASWELL(dev_priv) && \
 				 (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0A00)
 #define IS_HSW_GT3(dev_priv)	(IS_HASWELL(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x0020)
+				 (dev_priv)->info.gt == 3)
 /* ULX machines are also considered ULT. */
 #define IS_HSW_ULX(dev_priv)	(INTEL_DEVID(dev_priv) == 0x0A0E || \
 				 INTEL_DEVID(dev_priv) == 0x0A1E)
@@ -2910,15 +2935,15 @@ intel_info(const struct drm_i915_private *dev_priv)
 				 INTEL_DEVID(dev_priv) == 0x5915 || \
 				 INTEL_DEVID(dev_priv) == 0x591E)
 #define IS_SKL_GT2(dev_priv)	(IS_SKYLAKE(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x0010)
+				 (dev_priv)->info.gt == 2)
 #define IS_SKL_GT3(dev_priv)	(IS_SKYLAKE(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x0020)
+				 (dev_priv)->info.gt == 3)
 #define IS_SKL_GT4(dev_priv)	(IS_SKYLAKE(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x0030)
+				 (dev_priv)->info.gt == 4)
 #define IS_KBL_GT2(dev_priv)	(IS_KABYLAKE(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x0010)
+				 (dev_priv)->info.gt == 2)
 #define IS_KBL_GT3(dev_priv)	(IS_KABYLAKE(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x0020)
+				 (dev_priv)->info.gt == 3)
 #define IS_CFL_ULT(dev_priv)	(IS_COFFEELAKE(dev_priv) && \
 				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x00A0)
 
@@ -3647,6 +3672,9 @@ i915_vm_to_ppgtt(struct i915_address_space *vm)
 /* i915_gem_fence_reg.c */
 int __must_check i915_vma_get_fence(struct i915_vma *vma);
 int __must_check i915_vma_put_fence(struct i915_vma *vma);
+struct drm_i915_fence_reg *
+i915_reserve_fence(struct drm_i915_private *dev_priv);
+void i915_unreserve_fence(struct drm_i915_fence_reg *fence);
 
 void i915_gem_revoke_fences(struct drm_i915_private *dev_priv);
 void i915_gem_restore_fences(struct drm_i915_private *dev_priv);
@@ -4332,11 +4360,4 @@ int remap_io_mapping(struct vm_area_struct *vma,
 		     unsigned long addr, unsigned long pfn, unsigned long size,
 		     struct io_mapping *iomap);
 
-static inline bool
-intel_engine_can_store_dword(struct intel_engine_cs *engine)
-{
-	return __intel_engine_can_store_dword(INTEL_GEN(engine->i915),
-					      engine->class);
-}
-
 #endif

drivers/gpu/drm/i915/i915_gem.c

@@ -1013,17 +1013,20 @@ gtt_user_read(struct io_mapping *mapping,
 	      loff_t base, int offset,
 	      char __user *user_data, int length)
 {
-	void *vaddr;
+	void __iomem *vaddr;
 	unsigned long unwritten;
 
 	/* We can use the cpu mem copy function because this is X86. */
-	vaddr = (void __force *)io_mapping_map_atomic_wc(mapping, base);
-	unwritten = __copy_to_user_inatomic(user_data, vaddr + offset, length);
+	vaddr = io_mapping_map_atomic_wc(mapping, base);
+	unwritten = __copy_to_user_inatomic(user_data,
+					    (void __force *)vaddr + offset,
+					    length);
 	io_mapping_unmap_atomic(vaddr);
 	if (unwritten) {
-		vaddr = (void __force *)
-			io_mapping_map_wc(mapping, base, PAGE_SIZE);
-		unwritten = copy_to_user(user_data, vaddr + offset, length);
+		vaddr = io_mapping_map_wc(mapping, base, PAGE_SIZE);
+		unwritten = copy_to_user(user_data,
+					 (void __force *)vaddr + offset,
+					 length);
 		io_mapping_unmap(vaddr);
 	}
 	return unwritten;
@@ -1189,18 +1192,18 @@ ggtt_write(struct io_mapping *mapping,
 	   loff_t base, int offset,
 	   char __user *user_data, int length)
 {
-	void *vaddr;
+	void __iomem *vaddr;
 	unsigned long unwritten;
 
 	/* We can use the cpu mem copy function because this is X86. */
-	vaddr = (void __force *)io_mapping_map_atomic_wc(mapping, base);
-	unwritten = __copy_from_user_inatomic_nocache(vaddr + offset,
+	vaddr = io_mapping_map_atomic_wc(mapping, base);
+	unwritten = __copy_from_user_inatomic_nocache((void __force *)vaddr + offset,
 						      user_data, length);
 	io_mapping_unmap_atomic(vaddr);
 	if (unwritten) {
-		vaddr = (void __force *)
-			io_mapping_map_wc(mapping, base, PAGE_SIZE);
-		unwritten = copy_from_user(vaddr + offset, user_data, length);
+		vaddr = io_mapping_map_wc(mapping, base, PAGE_SIZE);
+		unwritten = copy_from_user((void __force *)vaddr + offset,
+					   user_data, length);
 		io_mapping_unmap(vaddr);
 	}
 
@@ -2476,8 +2479,6 @@ static int ____i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 {
 	struct sg_table *pages;
 
-	GEM_BUG_ON(i915_gem_object_has_pinned_pages(obj));
-
 	if (unlikely(obj->mm.madv != I915_MADV_WILLNEED)) {
 		DRM_DEBUG("Attempting to obtain a purgeable object\n");
 		return -EFAULT;
@@ -2507,6 +2508,8 @@ int __i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 		return err;
 
 	if (unlikely(IS_ERR_OR_NULL(obj->mm.pages))) {
+		GEM_BUG_ON(i915_gem_object_has_pinned_pages(obj));
+
 		err = ____i915_gem_object_get_pages(obj);
 		if (err)
 			goto unlock;
@@ -2590,6 +2593,8 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj,
 
 	if (!atomic_inc_not_zero(&obj->mm.pages_pin_count)) {
 		if (unlikely(IS_ERR_OR_NULL(obj->mm.pages))) {
+			GEM_BUG_ON(i915_gem_object_has_pinned_pages(obj));
+
 			ret = ____i915_gem_object_get_pages(obj);
 			if (ret)
 				goto err_unlock;
@@ -3257,11 +3262,11 @@ void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file)
 		struct i915_gem_context *ctx = lut->ctx;
 		struct i915_vma *vma;
 
+		GEM_BUG_ON(ctx->file_priv == ERR_PTR(-EBADF));
 		if (ctx->file_priv != fpriv)
 			continue;
 
 		vma = radix_tree_delete(&ctx->handles_vma, lut->handle);
-
 		GEM_BUG_ON(vma->obj != obj);
 
 		/* We allow the process to have multiple handles to the same
@@ -3375,26 +3380,14 @@ static int wait_for_timeline(struct i915_gem_timeline *tl, unsigned int flags)
 	return 0;
 }
 
-static int wait_for_engine(struct intel_engine_cs *engine, int timeout_ms)
-{
-	return wait_for(intel_engine_is_idle(engine), timeout_ms);
-}
-
 static int wait_for_engines(struct drm_i915_private *i915)
 {
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
-
-	for_each_engine(engine, i915, id) {
-		if (GEM_WARN_ON(wait_for_engine(engine, 50))) {
+	if (wait_for(intel_engines_are_idle(i915), 50)) {
+		DRM_ERROR("Failed to idle engines, declaring wedged!\n");
 		i915_gem_set_wedged(i915);
 		return -EIO;
 	}
 
-		GEM_BUG_ON(intel_engine_get_seqno(engine) !=
-			   intel_engine_last_submit(engine));
-	}
-
 	return 0;
 }
 
@@ -4426,6 +4419,7 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915,
 	llist_for_each_entry_safe(obj, on, freed, freed) {
 		GEM_BUG_ON(obj->bind_count);
 		GEM_BUG_ON(atomic_read(&obj->frontbuffer_bits));
+		GEM_BUG_ON(!list_empty(&obj->lut_list));
 
 		if (obj->ops->release)
 			obj->ops->release(obj);
@@ -4533,6 +4527,12 @@ static void assert_kernel_context_is_current(struct drm_i915_private *dev_priv)
 
 void i915_gem_sanitize(struct drm_i915_private *i915)
 {
+	if (i915_terminally_wedged(&i915->gpu_error)) {
+		mutex_lock(&i915->drm.struct_mutex);
+		i915_gem_unset_wedged(i915);
+		mutex_unlock(&i915->drm.struct_mutex);
+	}
+
 	/*
 	 * If we inherit context state from the BIOS or earlier occupants
 	 * of the GPU, the GPU may be in an inconsistent state when we
@@ -4572,7 +4572,7 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv)
 	ret = i915_gem_wait_for_idle(dev_priv,
 				     I915_WAIT_INTERRUPTIBLE |
 				     I915_WAIT_LOCKED);
-	if (ret)
+	if (ret && ret != -EIO)
 		goto err_unlock;
 
 	assert_kernel_context_is_current(dev_priv);
@@ -4594,7 +4594,8 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv)
 	 * reset the GPU back to its idle, low power state.
 	 */
 	WARN_ON(dev_priv->gt.awake);
-	WARN_ON(!intel_engines_are_idle(dev_priv));
+	if (WARN_ON(!intel_engines_are_idle(dev_priv)))
+		i915_gem_set_wedged(dev_priv); /* no hope, discard everything */
 
 	/*
 	 * Neither the BIOS, ourselves or any other kernel
@@ -4616,11 +4617,12 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv)
 	 * machine in an unusable condition.
 	 */
 	i915_gem_sanitize(dev_priv);
-	goto out_rpm_put;
+
+	intel_runtime_pm_put(dev_priv);
+	return 0;
 
 err_unlock:
 	mutex_unlock(&dev->struct_mutex);
-out_rpm_put:
 	intel_runtime_pm_put(dev_priv);
 	return ret;
 }

drivers/gpu/drm/i915/i915_gem_execbuffer.c

@@ -268,6 +268,11 @@ static inline u64 gen8_noncanonical_addr(u64 address)
 	return address & GENMASK_ULL(GEN8_HIGH_ADDRESS_BIT, 0);
 }
 
+static inline bool eb_use_cmdparser(const struct i915_execbuffer *eb)
+{
+	return eb->engine->needs_cmd_parser && eb->batch_len;
+}
+
 static int eb_create(struct i915_execbuffer *eb)
 {
 	if (!(eb->args->flags & I915_EXEC_HANDLE_LUT)) {
@@ -1159,6 +1164,13 @@ static u32 *reloc_gpu(struct i915_execbuffer *eb,
 	if (unlikely(!cache->rq)) {
 		int err;
 
+		/* If we need to copy for the cmdparser, we will stall anyway */
+		if (eb_use_cmdparser(eb))
+			return ERR_PTR(-EWOULDBLOCK);
+
+		if (!intel_engine_can_store_dword(eb->engine))
+			return ERR_PTR(-ENODEV);
+
 		err = __reloc_gpu_alloc(eb, vma, len);
 		if (unlikely(err))
 			return ERR_PTR(err);
@@ -1183,9 +1195,7 @@ relocate_entry(struct i915_vma *vma,
 
 	if (!eb->reloc_cache.vaddr &&
 	    (DBG_FORCE_RELOC == FORCE_GPU_RELOC ||
-	     !reservation_object_test_signaled_rcu(vma->resv, true)) &&
-	    __intel_engine_can_store_dword(eb->reloc_cache.gen,
-					   eb->engine->class)) {
+	     !reservation_object_test_signaled_rcu(vma->resv, true))) {
 		const unsigned int gen = eb->reloc_cache.gen;
 		unsigned int len;
 		u32 *batch;
@@ -2291,7 +2301,7 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 		goto err_vma;
 	}
 
-	if (eb.engine->needs_cmd_parser && eb.batch_len) {
+	if (eb_use_cmdparser(&eb)) {
 		struct i915_vma *vma;
 
 		vma = eb_parse(&eb, drm_is_current_master(file));

drivers/gpu/drm/i915/i915_gem_fence_reg.c

@@ -359,6 +359,57 @@ i915_vma_get_fence(struct i915_vma *vma)
 	return fence_update(fence, set);
 }
 
+/**
+ * i915_reserve_fence - Reserve a fence for vGPU
+ * @dev_priv: i915 device private
+ *
+ * This function walks the fence regs looking for a free one and remove
+ * it from the fence_list. It is used to reserve fence for vGPU to use.
+ */
+struct drm_i915_fence_reg *
+i915_reserve_fence(struct drm_i915_private *dev_priv)
+{
+	struct drm_i915_fence_reg *fence;
+	int count;
+	int ret;
+
+	lockdep_assert_held(&dev_priv->drm.struct_mutex);
+
+	/* Keep at least one fence available for the display engine. */
+	count = 0;
+	list_for_each_entry(fence, &dev_priv->mm.fence_list, link)
+		count += !fence->pin_count;
+	if (count <= 1)
+		return ERR_PTR(-ENOSPC);
+
+	fence = fence_find(dev_priv);
+	if (IS_ERR(fence))
+		return fence;
+
+	if (fence->vma) {
+		/* Force-remove fence from VMA */
+		ret = fence_update(fence, NULL);
+		if (ret)
+			return ERR_PTR(ret);
+	}
+
+	list_del(&fence->link);
+	return fence;
+}
+
+/**
+ * i915_unreserve_fence - Reclaim a reserved fence
+ * @fence: the fence reg
+ *
+ * This function add a reserved fence register from vGPU to the fence_list.
+ */
+void i915_unreserve_fence(struct drm_i915_fence_reg *fence)
+{
+	lockdep_assert_held(&fence->i915->drm.struct_mutex);
+
+	list_add(&fence->link, &fence->i915->mm.fence_list);
+}
+
 /**
  * i915_gem_revoke_fences - revoke fence state
  * @dev_priv: i915 device private

drivers/gpu/drm/i915/i915_gem_gtt.c

@@ -356,39 +356,86 @@ static gen6_pte_t iris_pte_encode(dma_addr_t addr,
 
 static struct page *vm_alloc_page(struct i915_address_space *vm, gfp_t gfp)
 {
-	struct page *page;
+	struct pagevec *pvec = &vm->free_pages;
 
 	if (I915_SELFTEST_ONLY(should_fail(&vm->fault_attr, 1)))
 		i915_gem_shrink_all(vm->i915);
 
-	if (vm->free_pages.nr)
-		return vm->free_pages.pages[--vm->free_pages.nr];
+	if (likely(pvec->nr))
+		return pvec->pages[--pvec->nr];
+
+	if (!vm->pt_kmap_wc)
+		return alloc_page(gfp);
+
+	/* A placeholder for a specific mutex to guard the WC stash */
+	lockdep_assert_held(&vm->i915->drm.struct_mutex);
+
+	/* Look in our global stash of WC pages... */
+	pvec = &vm->i915->mm.wc_stash;
+	if (likely(pvec->nr))
+		return pvec->pages[--pvec->nr];
+
+	/* Otherwise batch allocate pages to amoritize cost of set_pages_wc. */
+	do {
+		struct page *page;
 
 		page = alloc_page(gfp);
-	if (!page)
+		if (unlikely(!page))
+			break;
+
+		pvec->pages[pvec->nr++] = page;
+	} while (pagevec_space(pvec));
+
+	if (unlikely(!pvec->nr))
 		return NULL;
 
-	if (vm->pt_kmap_wc)
-		set_pages_array_wc(&page, 1);
+	set_pages_array_wc(pvec->pages, pvec->nr);
 
-	return page;
+	return pvec->pages[--pvec->nr];
 }
 
-static void vm_free_pages_release(struct i915_address_space *vm)
+static void vm_free_pages_release(struct i915_address_space *vm,
+				  bool immediate)
 {
-	GEM_BUG_ON(!pagevec_count(&vm->free_pages));
+	struct pagevec *pvec = &vm->free_pages;
+
+	GEM_BUG_ON(!pagevec_count(pvec));
+
+	if (vm->pt_kmap_wc) {
+		struct pagevec *stash = &vm->i915->mm.wc_stash;
 
-	if (vm->pt_kmap_wc)
-		set_pages_array_wb(vm->free_pages.pages,
-				   pagevec_count(&vm->free_pages));
+		/* When we use WC, first fill up the global stash and then
+		 * only if full immediately free the overflow.
+		 */
 
-	__pagevec_release(&vm->free_pages);
+		lockdep_assert_held(&vm->i915->drm.struct_mutex);
+		if (pagevec_space(stash)) {
+			do {
+				stash->pages[stash->nr++] =
+					pvec->pages[--pvec->nr];
+				if (!pvec->nr)
+					return;
+			} while (pagevec_space(stash));
+
+			/* As we have made some room in the VM's free_pages,
+			 * we can wait for it to fill again. Unless we are
+			 * inside i915_address_space_fini() and must
+			 * immediately release the pages!
+			 */
+			if (!immediate)
+				return;
+		}
+
+		set_pages_array_wb(pvec->pages, pvec->nr);
+	}
+
+	__pagevec_release(pvec);
 }
 
 static void vm_free_page(struct i915_address_space *vm, struct page *page)
 {
 	if (!pagevec_add(&vm->free_pages, page))
-		vm_free_pages_release(vm);
+		vm_free_pages_release(vm, false);
 }
 
 static int __setup_page_dma(struct i915_address_space *vm,
@@ -452,12 +499,31 @@ static void fill_page_dma_32(struct i915_address_space *vm,
 static int
 setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
 {
-	return __setup_page_dma(vm, &vm->scratch_page, gfp | __GFP_ZERO);
+	struct page *page;
+	dma_addr_t addr;
+
+	page = alloc_page(gfp | __GFP_ZERO);
+	if (unlikely(!page))
+		return -ENOMEM;
+
+	addr = dma_map_page(vm->dma, page, 0, PAGE_SIZE,
+			    PCI_DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(vm->dma, addr))) {
+		__free_page(page);
+		return -ENOMEM;
+	}
+
+	vm->scratch_page.page = page;
+	vm->scratch_page.daddr = addr;
+	return 0;
 }
 
 static void cleanup_scratch_page(struct i915_address_space *vm)
 {
-	cleanup_page_dma(vm, &vm->scratch_page);
+	struct i915_page_dma *p = &vm->scratch_page;
+
+	dma_unmap_page(vm->dma, p->daddr, PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+	__free_page(p->page);
 }
 
 static struct i915_page_table *alloc_pt(struct i915_address_space *vm)
@@ -1337,18 +1403,18 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 		1ULL << 48 :
 		1ULL << 32;
 
-	ret = gen8_init_scratch(&ppgtt->base);
-	if (ret) {
-		ppgtt->base.total = 0;
-		return ret;
-	}
-
 	/* There are only few exceptions for gen >=6. chv and bxt.
 	 * And we are not sure about the latter so play safe for now.
 	 */
 	if (IS_CHERRYVIEW(dev_priv) || IS_BROXTON(dev_priv))
 		ppgtt->base.pt_kmap_wc = true;
 
+	ret = gen8_init_scratch(&ppgtt->base);
+	if (ret) {
+		ppgtt->base.total = 0;
+		return ret;
+	}
+
 	if (use_4lvl(vm)) {
 		ret = setup_px(&ppgtt->base, &ppgtt->pml4);
 		if (ret)
@@ -1872,7 +1938,7 @@ static void i915_address_space_init(struct i915_address_space *vm,
 static void i915_address_space_fini(struct i915_address_space *vm)
 {
 	if (pagevec_count(&vm->free_pages))
-		vm_free_pages_release(vm);
+		vm_free_pages_release(vm, true);
 
 	i915_gem_timeline_fini(&vm->timeline);
 	drm_mm_takedown(&vm->mm);
@@ -1885,12 +1951,12 @@ static void gtt_write_workarounds(struct drm_i915_private *dev_priv)
 	 * called on driver load and after a GPU reset, so you can place
 	 * workarounds here even if they get overwritten by GPU reset.
 	 */
-	/* WaIncreaseDefaultTLBEntries:chv,bdw,skl,bxt,kbl,glk,cfl */
+	/* WaIncreaseDefaultTLBEntries:chv,bdw,skl,bxt,kbl,glk,cfl,cnl */
 	if (IS_BROADWELL(dev_priv))
 		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_BDW);
 	else if (IS_CHERRYVIEW(dev_priv))
 		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_CHV);
-	else if (IS_GEN9_BC(dev_priv))
+	else if (IS_GEN9_BC(dev_priv) || IS_GEN10(dev_priv))
 		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_SKL);
 	else if (IS_GEN9_LP(dev_priv))
 		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT);
@@ -2598,6 +2664,7 @@ void i915_ggtt_cleanup_hw(struct drm_i915_private *dev_priv)
 {
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_vma *vma, *vn;
+	struct pagevec *pvec;
 
 	ggtt->base.closed = true;
 
@@ -2621,6 +2688,13 @@ void i915_ggtt_cleanup_hw(struct drm_i915_private *dev_priv)
 	}
 
 	ggtt->base.cleanup(&ggtt->base);
+
+	pvec = &dev_priv->mm.wc_stash;
+	if (pvec->nr) {
+		set_pages_array_wb(pvec->pages, pvec->nr);
+		__pagevec_release(pvec);
+	}
+
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 
 	arch_phys_wc_del(ggtt->mtrr);
@@ -2716,13 +2790,13 @@ static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size)
 	phys_addr = pci_resource_start(pdev, 0) + pci_resource_len(pdev, 0) / 2;
 
 	/*
-	 * On BXT writes larger than 64 bit to the GTT pagetable range will be
-	 * dropped. For WC mappings in general we have 64 byte burst writes
-	 * when the WC buffer is flushed, so we can't use it, but have to
+	 * On BXT+/CNL+ writes larger than 64 bit to the GTT pagetable range
+	 * will be dropped. For WC mappings in general we have 64 byte burst
+	 * writes when the WC buffer is flushed, so we can't use it, but have to
 	 * resort to an uncached mapping. The WC issue is easily caught by the
 	 * readback check when writing GTT PTE entries.
 	 */
-	if (IS_GEN9_LP(dev_priv))
+	if (IS_GEN9_LP(dev_priv) || INTEL_GEN(dev_priv) >= 10)
 		ggtt->gsm = ioremap_nocache(phys_addr, size);
 	else
 		ggtt->gsm = ioremap_wc(phys_addr, size);

drivers/gpu/drm/i915/i915_irq.c

@@ -336,7 +336,7 @@ void gen6_mask_pm_irq(struct drm_i915_private *dev_priv, u32 mask)
 	__gen6_mask_pm_irq(dev_priv, mask);
 }
 
-void gen6_reset_pm_iir(struct drm_i915_private *dev_priv, u32 reset_mask)
+static void gen6_reset_pm_iir(struct drm_i915_private *dev_priv, u32 reset_mask)
 {
 	i915_reg_t reg = gen6_pm_iir(dev_priv);
 
@@ -347,7 +347,7 @@ void gen6_reset_pm_iir(struct drm_i915_private *dev_priv, u32 reset_mask)
 	POSTING_READ(reg);
 }
 
-void gen6_enable_pm_irq(struct drm_i915_private *dev_priv, u32 enable_mask)
+static void gen6_enable_pm_irq(struct drm_i915_private *dev_priv, u32 enable_mask)
 {
 	lockdep_assert_held(&dev_priv->irq_lock);
 
@@ -357,7 +357,7 @@ void gen6_enable_pm_irq(struct drm_i915_private *dev_priv, u32 enable_mask)
 	/* unmask_pm_irq provides an implicit barrier (POSTING_READ) */
 }
 
-void gen6_disable_pm_irq(struct drm_i915_private *dev_priv, u32 disable_mask)
+static void gen6_disable_pm_irq(struct drm_i915_private *dev_priv, u32 disable_mask)
 {
 	lockdep_assert_held(&dev_priv->irq_lock);
 
@@ -405,7 +405,7 @@ void gen6_disable_rps_interrupts(struct drm_i915_private *dev_priv)
 	synchronize_irq(dev_priv->drm.irq);
 
 	/* Now that we will not be generating any more work, flush any
-	 * outsanding tasks. As we are called on the RPS idle path,
+	 * outstanding tasks. As we are called on the RPS idle path,
 	 * we will reset the GPU to minimum frequencies, so the current
 	 * state of the worker can be discarded.
 	 */

drivers/gpu/drm/i915/i915_reg.h

@@ -2373,6 +2373,7 @@ enum i915_power_well_id {
 
 #define GAMT_CHKN_BIT_REG	_MMIO(0x4ab8)
 #define   GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING	(1<<28)
+#define   GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT	(1<<24)
 
 #if 0
 #define PRB0_TAIL	_MMIO(0x2030)
@@ -2491,6 +2492,7 @@ enum i915_power_well_id {
 # define _3D_CHICKEN2_WM_READ_PIPELINED			(1 << 14)
 #define _3D_CHICKEN3	_MMIO(0x2090)
 #define  _3D_CHICKEN_SF_DISABLE_OBJEND_CULL		(1 << 10)
+#define  _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE	(1 << 5)
 #define  _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL		(1 << 5)
 #define  _3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(x)	((x)<<1) /* gen8+ */
 #define  _3D_CHICKEN3_SF_DISABLE_PIPELINED_ATTR_FETCH	(1 << 1) /* gen6 */
@@ -2938,6 +2940,9 @@ enum i915_power_well_id {
 #define ILK_DPFC_CHICKEN	_MMIO(0x43224)
 #define   ILK_DPFC_DISABLE_DUMMY0 (1<<8)
 #define   ILK_DPFC_NUKE_ON_ANY_MODIFICATION	(1<<23)
+#define   GLK_SKIP_SEG_EN		(1<<12)
+#define   GLK_SKIP_SEG_COUNT_MASK	(3<<10)
+#define   GLK_SKIP_SEG_COUNT(x)		((x)<<10)
 #define ILK_FBC_RT_BASE		_MMIO(0x2128)
 #define   ILK_FBC_RT_VALID	(1<<0)
 #define   SNB_FBC_FRONT_BUFFER	(1<<1)
@@ -3806,6 +3811,12 @@ enum {
 #define   PWM2_GATING_DIS		(1 << 14)
 #define   PWM1_GATING_DIS		(1 << 13)
 
+/*
+ * GEN10 clock gating regs
+ */
+#define SLICE_UNIT_LEVEL_CLKGATE	_MMIO(0x94d4)
+#define  SARBUNIT_CLKGATE_DIS		(1 << 5)
+
 /*
  * Display engine regs
  */
@@ -6916,6 +6927,10 @@ enum {
 #define  GLK_CL1_PWR_DOWN	(1 << 11)
 #define  GLK_CL0_PWR_DOWN	(1 << 10)
 
+#define CHICKEN_MISC_4		_MMIO(0x4208c)
+#define   FBC_STRIDE_OVERRIDE	(1 << 13)
+#define   FBC_STRIDE_MASK	0x1FFF
+
 #define _CHICKEN_PIPESL_1_A	0x420b0
 #define _CHICKEN_PIPESL_1_B	0x420b4
 #define  HSW_FBCQ_DIS			(1 << 22)
@@ -7017,6 +7032,7 @@ enum {
 
 /* GEN8 chicken */
 #define HDC_CHICKEN0				_MMIO(0x7300)
+#define CNL_HDC_CHICKEN0			_MMIO(0xE5F0)
 #define  HDC_FORCE_CSR_NON_COHERENT_OVR_DISABLE	(1<<15)
 #define  HDC_FENCE_DEST_SLM_DISABLE		(1<<14)
 #define  HDC_DONOT_FETCH_MEM_WHEN_MASKED	(1<<11)
@@ -7470,6 +7486,7 @@ enum {
 #define  PCH_DPLUNIT_CLOCK_GATE_DISABLE (1<<30)
 #define  PCH_DPLSUNIT_CLOCK_GATE_DISABLE (1<<29)
 #define  PCH_CPUNIT_CLOCK_GATE_DISABLE (1<<14)
+#define  CNP_PWM_CGE_GATING_DISABLE (1<<13)
 #define  PCH_LP_PARTITION_LEVEL_DISABLE  (1<<12)
 
 /* CPU: FDI_TX */
@@ -8044,10 +8061,12 @@ enum {
 #define   FLOW_CONTROL_ENABLE		(1<<15)
 #define   PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE	(1<<8)
 #define   STALL_DOP_GATING_DISABLE		(1<<5)
+#define   THROTTLE_12_5				(7<<2)
 
 #define GEN7_ROW_CHICKEN2		_MMIO(0xe4f4)
 #define GEN7_ROW_CHICKEN2_GT2		_MMIO(0xf4f4)
 #define   DOP_CLOCK_GATING_DISABLE	(1<<0)
+#define   PUSH_CONSTANT_DEREF_DISABLE	(1<<8)
 
 #define HSW_ROW_CHICKEN3		_MMIO(0xe49c)
 #define  HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE    (1 << 6)
@@ -8059,6 +8078,7 @@ enum {
 #define   HSW_SAMPLE_C_PERFORMANCE	(1<<9)
 #define   GEN8_CENTROID_PIXEL_OPT_DIS	(1<<8)
 #define   GEN9_DISABLE_OCL_OOB_SUPPRESS_LOGIC	(1<<5)
+#define   CNL_FAST_ANISO_L1_BANKING_FIX	(1<<4)
 #define   GEN8_SAMPLER_POWER_BYPASS_DIS	(1<<1)
 
 #define GEN9_HALF_SLICE_CHICKEN7	_MMIO(0xe194)

drivers/gpu/drm/i915/i915_trace.h

@@ -1031,5 +1031,5 @@ TRACE_EVENT(switch_mm,
 
 /* This part must be outside protection */
 #undef TRACE_INCLUDE_PATH
-#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/i915
 #include <trace/define_trace.h>

drivers/gpu/drm/i915/intel_atomic_plane.c

@@ -107,7 +107,9 @@ intel_plane_destroy_state(struct drm_plane *plane,
 	drm_atomic_helper_plane_destroy_state(plane, state);
 }
 
-int intel_plane_atomic_check_with_state(struct intel_crtc_state *crtc_state,
+int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_state,
+					struct intel_crtc_state *crtc_state,
+					const struct intel_plane_state *old_plane_state,
 					struct intel_plane_state *intel_state)
 {
 	struct drm_plane *plane = intel_state->base.plane;
@@ -124,7 +126,7 @@ int intel_plane_atomic_check_with_state(struct intel_crtc_state *crtc_state,
 	 * anything driver-specific we need to test in that case, so
 	 * just return success.
 	 */
-	if (!intel_state->base.crtc && !plane->state->crtc)
+	if (!intel_state->base.crtc && !old_plane_state->base.crtc)
 		return 0;
 
 	/* Clip all planes to CRTC size, or 0x0 if CRTC is disabled */
@@ -194,16 +196,21 @@ int intel_plane_atomic_check_with_state(struct intel_crtc_state *crtc_state,
 	else
 		crtc_state->active_planes &= ~BIT(intel_plane->id);
 
-	return intel_plane_atomic_calc_changes(&crtc_state->base, state);
+	return intel_plane_atomic_calc_changes(old_crtc_state,
+					       &crtc_state->base,
+					       old_plane_state,
+					       state);
 }
 
 static int intel_plane_atomic_check(struct drm_plane *plane,
-				    struct drm_plane_state *state)
+				    struct drm_plane_state *new_plane_state)
 {
-	struct drm_crtc *crtc = state->crtc;
-	struct drm_crtc_state *drm_crtc_state;
-
-	crtc = crtc ? crtc : plane->state->crtc;
+	struct drm_atomic_state *state = new_plane_state->state;
+	const struct drm_plane_state *old_plane_state =
+		drm_atomic_get_old_plane_state(state, plane);
+	struct drm_crtc *crtc = new_plane_state->crtc ?: old_plane_state->crtc;
+	const struct drm_crtc_state *old_crtc_state;
+	struct drm_crtc_state *new_crtc_state;
 
 	/*
 	 * Both crtc and plane->crtc could be NULL if we're updating a
@@ -214,29 +221,33 @@ static int intel_plane_atomic_check(struct drm_plane *plane,
 	if (!crtc)
 		return 0;
 
-	drm_crtc_state = drm_atomic_get_existing_crtc_state(state->state, crtc);
-	if (WARN_ON(!drm_crtc_state))
-		return -EINVAL;
+	old_crtc_state = drm_atomic_get_old_crtc_state(state, crtc);
+	new_crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
 
-	return intel_plane_atomic_check_with_state(to_intel_crtc_state(drm_crtc_state),
-						   to_intel_plane_state(state));
+	return intel_plane_atomic_check_with_state(to_intel_crtc_state(old_crtc_state),
+						   to_intel_crtc_state(new_crtc_state),
+						   to_intel_plane_state(old_plane_state),
+						   to_intel_plane_state(new_plane_state));
 }
 
 static void intel_plane_atomic_update(struct drm_plane *plane,
 				      struct drm_plane_state *old_state)
 {
+	struct intel_atomic_state *state = to_intel_atomic_state(old_state->state);
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct intel_plane_state *intel_state =
-		to_intel_plane_state(plane->state);
-	struct drm_crtc *crtc = plane->state->crtc ?: old_state->crtc;
+	const struct intel_plane_state *new_plane_state =
+		intel_atomic_get_new_plane_state(state, intel_plane);
+	struct drm_crtc *crtc = new_plane_state->base.crtc ?: old_state->crtc;
+
+	if (new_plane_state->base.visible) {
+		const struct intel_crtc_state *new_crtc_state =
+			intel_atomic_get_new_crtc_state(state, to_intel_crtc(crtc));
 
-	if (intel_state->base.visible) {
 		trace_intel_update_plane(plane,
 					 to_intel_crtc(crtc));
 
 		intel_plane->update_plane(intel_plane,
-					  to_intel_crtc_state(crtc->state),
-					  intel_state);
+					  new_crtc_state, new_plane_state);
 	} else {
 		trace_intel_disable_plane(plane,
 					  to_intel_crtc(crtc));

drivers/gpu/drm/i915/intel_crt.c

@@ -143,7 +143,7 @@ static void hsw_crt_get_config(struct intel_encoder *encoder,
 /* Note: The caller is required to filter out dpms modes not supported by the
  * platform. */
 static void intel_crt_set_dpms(struct intel_encoder *encoder,
-			       struct intel_crtc_state *crtc_state,
+			       const struct intel_crtc_state *crtc_state,
 			       int mode)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
@@ -194,28 +194,28 @@ static void intel_crt_set_dpms(struct intel_encoder *encoder,
 }
 
 static void intel_disable_crt(struct intel_encoder *encoder,
-			      struct intel_crtc_state *old_crtc_state,
-			      struct drm_connector_state *old_conn_state)
+			      const struct intel_crtc_state *old_crtc_state,
+			      const struct drm_connector_state *old_conn_state)
 {
 	intel_crt_set_dpms(encoder, old_crtc_state, DRM_MODE_DPMS_OFF);
 }
 
 static void pch_disable_crt(struct intel_encoder *encoder,
-			    struct intel_crtc_state *old_crtc_state,
-			    struct drm_connector_state *old_conn_state)
+			    const struct intel_crtc_state *old_crtc_state,
+			    const struct drm_connector_state *old_conn_state)
 {
 }
 
 static void pch_post_disable_crt(struct intel_encoder *encoder,
-				 struct intel_crtc_state *old_crtc_state,
-				 struct drm_connector_state *old_conn_state)
+				 const struct intel_crtc_state *old_crtc_state,
+				 const struct drm_connector_state *old_conn_state)
 {
 	intel_disable_crt(encoder, old_crtc_state, old_conn_state);
 }
 
 static void hsw_post_disable_crt(struct intel_encoder *encoder,
-				 struct intel_crtc_state *old_crtc_state,
-				 struct drm_connector_state *old_conn_state)
+				 const struct intel_crtc_state *old_crtc_state,
+				 const struct drm_connector_state *old_conn_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
@@ -228,8 +228,8 @@ static void hsw_post_disable_crt(struct intel_encoder *encoder,
 }
 
 static void intel_enable_crt(struct intel_encoder *encoder,
-			     struct intel_crtc_state *pipe_config,
-			     struct drm_connector_state *conn_state)
+			     const struct intel_crtc_state *pipe_config,
+			     const struct drm_connector_state *conn_state)
 {
 	intel_crt_set_dpms(encoder, pipe_config, DRM_MODE_DPMS_ON);
 }

drivers/gpu/drm/i915/intel_device_info.c

@@ -412,7 +412,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 	else if (INTEL_INFO(dev_priv)->gen >= 9)
 		gen9_sseu_info_init(dev_priv);
 
-	info->has_snoop = !info->has_llc;
+	WARN_ON(info->has_snoop != !info->has_llc);
 
 	DRM_DEBUG_DRIVER("slice mask: %04x\n", info->sseu.slice_mask);
 	DRM_DEBUG_DRIVER("slice total: %u\n", hweight8(info->sseu.slice_mask));

drivers/gpu/drm/i915/intel_dp_mst.c

@@ -123,8 +123,8 @@ static int intel_dp_mst_atomic_check(struct drm_connector *connector,
 }
 
 static void intel_mst_disable_dp(struct intel_encoder *encoder,
-				 struct intel_crtc_state *old_crtc_state,
-				 struct drm_connector_state *old_conn_state)
+				 const struct intel_crtc_state *old_crtc_state,
+				 const struct drm_connector_state *old_conn_state)
 {
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
 	struct intel_digital_port *intel_dig_port = intel_mst->primary;
@@ -146,8 +146,8 @@ static void intel_mst_disable_dp(struct intel_encoder *encoder,
 }
 
 static void intel_mst_post_disable_dp(struct intel_encoder *encoder,
-				      struct intel_crtc_state *old_crtc_state,
-				      struct drm_connector_state *old_conn_state)
+				      const struct intel_crtc_state *old_crtc_state,
+				      const struct drm_connector_state *old_conn_state)
 {
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
 	struct intel_digital_port *intel_dig_port = intel_mst->primary;
@@ -176,8 +176,8 @@ static void intel_mst_post_disable_dp(struct intel_encoder *encoder,
 }
 
 static void intel_mst_pre_enable_dp(struct intel_encoder *encoder,
-				    struct intel_crtc_state *pipe_config,
-				    struct drm_connector_state *conn_state)
+				    const struct intel_crtc_state *pipe_config,
+				    const struct drm_connector_state *conn_state)
 {
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
 	struct intel_digital_port *intel_dig_port = intel_mst->primary;
@@ -219,8 +219,8 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder,
 }
 
 static void intel_mst_enable_dp(struct intel_encoder *encoder,
-				struct intel_crtc_state *pipe_config,
-				struct drm_connector_state *conn_state)
+				const struct intel_crtc_state *pipe_config,
+				const struct drm_connector_state *conn_state)
 {
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
 	struct intel_digital_port *intel_dig_port = intel_mst->primary;

drivers/gpu/drm/i915/intel_dsi.c

@@ -731,7 +731,7 @@ static void intel_dsi_port_disable(struct intel_encoder *encoder)
 }
 
 static void intel_dsi_prepare(struct intel_encoder *intel_encoder,
-			      struct intel_crtc_state *pipe_config);
+			      const struct intel_crtc_state *pipe_config);
 static void intel_dsi_unprepare(struct intel_encoder *encoder);
 
 static void intel_dsi_msleep(struct intel_dsi *intel_dsi, int msec)
@@ -783,8 +783,8 @@ static void intel_dsi_msleep(struct intel_dsi *intel_dsi, int msec)
  */
 
 static void intel_dsi_pre_enable(struct intel_encoder *encoder,
-				 struct intel_crtc_state *pipe_config,
-				 struct drm_connector_state *conn_state)
+				 const struct intel_crtc_state *pipe_config,
+				 const struct drm_connector_state *conn_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
@@ -878,8 +878,8 @@ static void intel_dsi_pre_enable(struct intel_encoder *encoder,
  * the pre_enable hook.
  */
 static void intel_dsi_enable_nop(struct intel_encoder *encoder,
-				 struct intel_crtc_state *pipe_config,
-				 struct drm_connector_state *conn_state)
+				 const struct intel_crtc_state *pipe_config,
+				 const struct drm_connector_state *conn_state)
 {
 	DRM_DEBUG_KMS("\n");
 }
@@ -889,8 +889,8 @@ static void intel_dsi_enable_nop(struct intel_encoder *encoder,
  * the post_disable hook.
  */
 static void intel_dsi_disable(struct intel_encoder *encoder,
-			      struct intel_crtc_state *old_crtc_state,
-			      struct drm_connector_state *old_conn_state)
+			      const struct intel_crtc_state *old_crtc_state,
+			      const struct drm_connector_state *old_conn_state)
 {
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	enum port port;
@@ -925,8 +925,8 @@ static void intel_dsi_clear_device_ready(struct intel_encoder *encoder)
 }
 
 static void intel_dsi_post_disable(struct intel_encoder *encoder,
-				   struct intel_crtc_state *pipe_config,
-				   struct drm_connector_state *conn_state)
+				   const struct intel_crtc_state *pipe_config,
+				   const struct drm_connector_state *conn_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
@@ -1370,7 +1370,7 @@ static u32 pixel_format_to_reg(enum mipi_dsi_pixel_format fmt)
 }
 
 static void intel_dsi_prepare(struct intel_encoder *intel_encoder,
-			      struct intel_crtc_state *pipe_config)
+			      const struct intel_crtc_state *pipe_config)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
 	struct drm_device *dev = encoder->dev;

drivers/gpu/drm/i915/intel_dvo.c

@@ -175,8 +175,8 @@ static void intel_dvo_get_config(struct intel_encoder *encoder,
 }
 
 static void intel_disable_dvo(struct intel_encoder *encoder,
-			      struct intel_crtc_state *old_crtc_state,
-			      struct drm_connector_state *old_conn_state)
+			      const struct intel_crtc_state *old_crtc_state,
+			      const struct drm_connector_state *old_conn_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
@@ -189,8 +189,8 @@ static void intel_disable_dvo(struct intel_encoder *encoder,
 }
 
 static void intel_enable_dvo(struct intel_encoder *encoder,
-			     struct intel_crtc_state *pipe_config,
-			     struct drm_connector_state *conn_state)
+			     const struct intel_crtc_state *pipe_config,
+			     const struct drm_connector_state *conn_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
@@ -258,8 +258,8 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder,
 }
 
 static void intel_dvo_pre_enable(struct intel_encoder *encoder,
-				 struct intel_crtc_state *pipe_config,
-				 struct drm_connector_state *conn_state)
+				 const struct intel_crtc_state *pipe_config,
+				 const struct drm_connector_state *conn_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->base.crtc);

drivers/gpu/drm/i915/intel_engine_cs.c

@@ -1065,6 +1065,51 @@ static int bxt_init_workarounds(struct intel_engine_cs *engine)
 	return 0;
 }
 
+static int cnl_init_workarounds(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *dev_priv = engine->i915;
+	int ret;
+
+	/* WaDisableI2mCycleOnWRPort: cnl (pre-prod) */
+	if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0))
+		WA_SET_BIT(GAMT_CHKN_BIT_REG,
+			   GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT);
+
+	/* WaForceContextSaveRestoreNonCoherent:cnl */
+	WA_SET_BIT_MASKED(CNL_HDC_CHICKEN0,
+			  HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT);
+
+	/* WaThrottleEUPerfToAvoidTDBackPressure:cnl(pre-prod) */
+	if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0))
+		WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, THROTTLE_12_5);
+
+	/* WaDisableReplayBufferBankArbitrationOptimization:cnl */
+	WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2,
+			  GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION);
+
+	/* WaDisableEnhancedSBEVertexCaching:cnl (pre-prod) */
+	if (IS_CNL_REVID(dev_priv, 0, CNL_REVID_B0))
+		WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2,
+				  GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE);
+
+	/* WaInPlaceDecompressionHang:cnl */
+	WA_SET_BIT(GEN9_GAMT_ECO_REG_RW_IA,
+		   GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS);
+
+	/* WaPushConstantDereferenceHoldDisable:cnl */
+	WA_SET_BIT(GEN7_ROW_CHICKEN2, PUSH_CONSTANT_DEREF_DISABLE);
+
+	/* FtrEnableFastAnisoL1BankingFix: cnl */
+	WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3, CNL_FAST_ANISO_L1_BANKING_FIX);
+
+	/* WaEnablePreemptionGranularityControlByUMD:cnl */
+	ret= wa_ring_whitelist_reg(engine, GEN8_CS_CHICKEN1);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
 static int kbl_init_workarounds(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
@@ -1185,6 +1230,8 @@ int init_workarounds_ring(struct intel_engine_cs *engine)
 		err =  glk_init_workarounds(engine);
 	else if (IS_COFFEELAKE(dev_priv))
 		err = cfl_init_workarounds(engine);
+	else if (IS_CANNONLAKE(dev_priv))
+		err = cnl_init_workarounds(engine);
 	else
 		err = 0;
 	if (err)
@@ -1335,6 +1382,21 @@ void intel_engines_mark_idle(struct drm_i915_private *i915)
 	}
 }
 
+bool intel_engine_can_store_dword(struct intel_engine_cs *engine)
+{
+	switch (INTEL_GEN(engine->i915)) {
+	case 2:
+		return false; /* uses physical not virtual addresses */
+	case 3:
+		/* maybe only uses physical not virtual addresses */
+		return !(IS_I915G(engine->i915) || IS_I915GM(engine->i915));
+	case 6:
+		return engine->class != VIDEO_DECODE_CLASS; /* b0rked */
+	default:
+		return true;
+	}
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/mock_engine.c"
 #endif

drivers/gpu/drm/i915/intel_fbc.c

@@ -291,6 +291,19 @@ static void gen7_fbc_activate(struct drm_i915_private *dev_priv)
 	u32 dpfc_ctl;
 	int threshold = dev_priv->fbc.threshold;
 
+	/* Display WA #0529: skl, kbl, bxt. */
+	if (IS_GEN9(dev_priv) && !IS_GEMINILAKE(dev_priv)) {
+		u32 val = I915_READ(CHICKEN_MISC_4);
+
+		val &= ~(FBC_STRIDE_OVERRIDE | FBC_STRIDE_MASK);
+
+		if (i915_gem_object_get_tiling(params->vma->obj) !=
+		    I915_TILING_X)
+			val |= FBC_STRIDE_OVERRIDE | params->gen9_wa_cfb_stride;
+
+		I915_WRITE(CHICKEN_MISC_4, val);
+	}
+
 	dpfc_ctl = 0;
 	if (IS_IVYBRIDGE(dev_priv))
 		dpfc_ctl |= IVB_DPFC_CTL_PLANE(params->crtc.plane);
@@ -881,6 +894,10 @@ static void intel_fbc_get_reg_params(struct intel_crtc *crtc,
 	params->fb.stride = cache->fb.stride;
 
 	params->cfb_size = intel_fbc_calculate_cfb_size(dev_priv, cache);
+
+	if (IS_GEN9(dev_priv) && !IS_GEMINILAKE(dev_priv))
+		params->gen9_wa_cfb_stride = DIV_ROUND_UP(cache->plane.src_w,
+						32 * fbc->threshold) * 8;
 }
 
 static bool intel_fbc_reg_params_equal(struct intel_fbc_reg_params *params1,

drivers/gpu/drm/i915/intel_fbdev.c

@@ -206,6 +206,7 @@ static int intelfb_create(struct drm_fb_helper *helper,
 	}
 
 	mutex_lock(&dev->struct_mutex);
+	intel_runtime_pm_get(dev_priv);
 
 	/* Pin the GGTT vma for our access via info->screen_base.
 	 * This also validates that any existing fb inherited from the
@@ -269,6 +270,7 @@ static int intelfb_create(struct drm_fb_helper *helper,
 		      fb->width, fb->height, i915_ggtt_offset(vma));
 	ifbdev->vma = vma;
 
+	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
 	vga_switcheroo_client_fb_set(pdev, info);
 	return 0;
@@ -276,6 +278,7 @@ static int intelfb_create(struct drm_fb_helper *helper,
 out_unpin:
 	intel_unpin_fb_vma(vma);
 out_unlock:
+	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
 	return ret;
 }

drivers/gpu/drm/i915/intel_lrc.c

@@ -1175,6 +1175,8 @@ static int intel_init_workaround_bb(struct intel_engine_cs *engine)
 		return -EINVAL;
 
 	switch (INTEL_GEN(engine->i915)) {
+	case 10:
+		return 0;
 	case 9:
 		wa_bb_fn[0] = gen9_init_indirectctx_bb;
 		wa_bb_fn[1] = gen9_init_perctx_bb;

drivers/gpu/drm/i915/intel_lvds.c

@@ -229,8 +229,8 @@ static void intel_lvds_pps_init_hw(struct drm_i915_private *dev_priv,
 }
 
 static void intel_pre_enable_lvds(struct intel_encoder *encoder,
-				  struct intel_crtc_state *pipe_config,
-				  struct drm_connector_state *conn_state)
+				  const struct intel_crtc_state *pipe_config,
+				  const struct drm_connector_state *conn_state)
 {
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
@@ -306,8 +306,8 @@ static void intel_pre_enable_lvds(struct intel_encoder *encoder,
  * Sets the power state for the panel.
  */
 static void intel_enable_lvds(struct intel_encoder *encoder,
-			      struct intel_crtc_state *pipe_config,
-			      struct drm_connector_state *conn_state)
+			      const struct intel_crtc_state *pipe_config,
+			      const struct drm_connector_state *conn_state)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
@@ -324,8 +324,8 @@ static void intel_enable_lvds(struct intel_encoder *encoder,
 }
 
 static void intel_disable_lvds(struct intel_encoder *encoder,
-			       struct intel_crtc_state *old_crtc_state,
-			       struct drm_connector_state *old_conn_state)
+			       const struct intel_crtc_state *old_crtc_state,
+			       const struct drm_connector_state *old_conn_state)
 {
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
@@ -339,8 +339,8 @@ static void intel_disable_lvds(struct intel_encoder *encoder,
 }
 
 static void gmch_disable_lvds(struct intel_encoder *encoder,
-			      struct intel_crtc_state *old_crtc_state,
-			      struct drm_connector_state *old_conn_state)
+			      const struct intel_crtc_state *old_crtc_state,
+			      const struct drm_connector_state *old_conn_state)
 
 {
 	intel_panel_disable_backlight(old_conn_state);
@@ -349,15 +349,15 @@ static void gmch_disable_lvds(struct intel_encoder *encoder,
 }
 
 static void pch_disable_lvds(struct intel_encoder *encoder,
-			     struct intel_crtc_state *old_crtc_state,
-			     struct drm_connector_state *old_conn_state)
+			     const struct intel_crtc_state *old_crtc_state,
+			     const struct drm_connector_state *old_conn_state)
 {
 	intel_panel_disable_backlight(old_conn_state);
 }
 
 static void pch_post_disable_lvds(struct intel_encoder *encoder,
-				  struct intel_crtc_state *old_crtc_state,
-				  struct drm_connector_state *old_conn_state)
+				  const struct intel_crtc_state *old_crtc_state,
+				  const struct drm_connector_state *old_conn_state)
 {
 	intel_disable_lvds(encoder, old_crtc_state, old_conn_state);
 }

drivers/gpu/drm/i915/intel_pipe_crc.c

@@ -506,7 +506,7 @@ static int ilk_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
 	return 0;
 }
 
-static void hsw_trans_edp_pipe_A_crc_wa(struct drm_i915_private *dev_priv,
+static void hsw_pipe_A_crc_wa(struct drm_i915_private *dev_priv,
 			      bool enable)
 {
 	struct drm_device *dev = &dev_priv->drm;
@@ -533,10 +533,24 @@ static void hsw_trans_edp_pipe_A_crc_wa(struct drm_i915_private *dev_priv,
 		goto put_state;
 	}
 
+	if (HAS_IPS(dev_priv)) {
+		/*
+		 * When IPS gets enabled, the pipe CRC changes. Since IPS gets
+		 * enabled and disabled dynamically based on package C states,
+		 * user space can't make reliable use of the CRCs, so let's just
+		 * completely disable it.
+		 */
+		pipe_config->ips_force_disable = enable;
+		if (pipe_config->ips_enabled == enable)
+			pipe_config->base.connectors_changed = true;
+	}
+
+	if (IS_HASWELL(dev_priv)) {
 		pipe_config->pch_pfit.force_thru = enable;
 		if (pipe_config->cpu_transcoder == TRANSCODER_EDP &&
 		    pipe_config->pch_pfit.enabled != enable)
 			pipe_config->base.connectors_changed = true;
+	}
 
 	ret = drm_atomic_commit(state);
 
@@ -570,8 +584,9 @@ static int ivb_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_SPRITE_IVB;
 		break;
 	case INTEL_PIPE_CRC_SOURCE_PF:
-		if (IS_HASWELL(dev_priv) && pipe == PIPE_A)
-			hsw_trans_edp_pipe_A_crc_wa(dev_priv, true);
+		if ((IS_HASWELL(dev_priv) ||
+		     IS_BROADWELL(dev_priv)) && pipe == PIPE_A)
+			hsw_pipe_A_crc_wa(dev_priv, true);
 
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PF_IVB;
 		break;
@@ -606,7 +621,6 @@ static int pipe_crc_set_source(struct drm_i915_private *dev_priv,
 			       enum intel_pipe_crc_source source)
 {
 	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[pipe];
-	struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe);
 	enum intel_display_power_domain power_domain;
 	u32 val = 0; /* shut up gcc */
 	int ret;
@@ -643,14 +657,6 @@ static int pipe_crc_set_source(struct drm_i915_private *dev_priv,
 			goto out;
 		}
 
-		/*
-		 * When IPS gets enabled, the pipe CRC changes. Since IPS gets
-		 * enabled and disabled dynamically based on package C states,
-		 * user space can't make reliable use of the CRCs, so let's just
-		 * completely disable it.
-		 */
-		hsw_disable_ips(crtc);
-
 		spin_lock_irq(&pipe_crc->lock);
 		kfree(pipe_crc->entries);
 		pipe_crc->entries = entries;
@@ -691,10 +697,9 @@ static int pipe_crc_set_source(struct drm_i915_private *dev_priv,
 			g4x_undo_pipe_scramble_reset(dev_priv, pipe);
 		else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 			vlv_undo_pipe_scramble_reset(dev_priv, pipe);
-		else if (IS_HASWELL(dev_priv) && pipe == PIPE_A)
-			hsw_trans_edp_pipe_A_crc_wa(dev_priv, false);
-
-		hsw_enable_ips(crtc);
+		else if ((IS_HASWELL(dev_priv) ||
+			  IS_BROADWELL(dev_priv)) && pipe == PIPE_A)
+			hsw_pipe_A_crc_wa(dev_priv, false);
 	}
 
 	ret = 0;
@@ -914,7 +919,6 @@ int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name,
 {
 	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
 	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[crtc->index];
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum intel_display_power_domain power_domain;
 	enum intel_pipe_crc_source source;
 	u32 val = 0; /* shut up gcc */
@@ -935,16 +939,6 @@ int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name,
 	if (ret != 0)
 		goto out;
 
-	if (source) {
-		/*
-		 * When IPS gets enabled, the pipe CRC changes. Since IPS gets
-		 * enabled and disabled dynamically based on package C states,
-		 * user space can't make reliable use of the CRCs, so let's just
-		 * completely disable it.
-		 */
-		hsw_disable_ips(intel_crtc);
-	}
-
 	I915_WRITE(PIPE_CRC_CTL(crtc->index), val);
 	POSTING_READ(PIPE_CRC_CTL(crtc->index));
 
@@ -953,10 +947,9 @@ int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name,
 			g4x_undo_pipe_scramble_reset(dev_priv, crtc->index);
 		else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 			vlv_undo_pipe_scramble_reset(dev_priv, crtc->index);
-		else if (IS_HASWELL(dev_priv) && crtc->index == PIPE_A)
-			hsw_trans_edp_pipe_A_crc_wa(dev_priv, false);
-
-		hsw_enable_ips(intel_crtc);
+		else if ((IS_HASWELL(dev_priv) ||
+			  IS_BROADWELL(dev_priv)) && crtc->index == PIPE_A)
+			hsw_pipe_A_crc_wa(dev_priv, false);
 	}
 
 	pipe_crc->skipped = 0;

drivers/gpu/drm/i915/intel_psr.c

@@ -103,28 +103,26 @@ static void intel_psr_write_vsc(struct intel_dp *intel_dp,
 	POSTING_READ(ctl_reg);
 }
 
-static void vlv_psr_setup_vsc(struct intel_dp *intel_dp)
+static void vlv_psr_setup_vsc(struct intel_dp *intel_dp,
+			      const struct intel_crtc_state *crtc_state)
 {
-	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(dev);
-	struct drm_crtc *crtc = intel_dig_port->base.base.crtc;
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	uint32_t val;
 
 	/* VLV auto-generate VSC package as per EDP 1.3 spec, Table 3.10 */
-	val  = I915_READ(VLV_VSCSDP(pipe));
+	val  = I915_READ(VLV_VSCSDP(crtc->pipe));
 	val &= ~VLV_EDP_PSR_SDP_FREQ_MASK;
 	val |= VLV_EDP_PSR_SDP_FREQ_EVFRAME;
-	I915_WRITE(VLV_VSCSDP(pipe), val);
+	I915_WRITE(VLV_VSCSDP(crtc->pipe), val);
 }
 
-static void skl_psr_setup_su_vsc(struct intel_dp *intel_dp)
+static void skl_psr_setup_su_vsc(struct intel_dp *intel_dp,
+				 const struct intel_crtc_state *crtc_state)
 {
-	struct edp_vsc_psr psr_vsc;
 	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(dev);
+	struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
+	struct edp_vsc_psr psr_vsc;
 
 	/* Prepare VSC Header for SU as per EDP 1.4 spec, Table 6.11 */
 	memset(&psr_vsc, 0, sizeof(psr_vsc));
@@ -145,7 +143,8 @@ static void skl_psr_setup_su_vsc(struct intel_dp *intel_dp)
 	intel_psr_write_vsc(intel_dp, &psr_vsc);
 }
 
-static void hsw_psr_setup_vsc(struct intel_dp *intel_dp)
+static void hsw_psr_setup_vsc(struct intel_dp *intel_dp,
+			      const struct intel_crtc_state *crtc_state)
 {
 	struct edp_vsc_psr psr_vsc;
 
@@ -233,16 +232,15 @@ static void hsw_psr_enable_sink(struct intel_dp *intel_dp)
 	I915_WRITE(aux_ctl_reg, aux_ctl);
 }
 
-static void vlv_psr_enable_source(struct intel_dp *intel_dp)
+static void vlv_psr_enable_source(struct intel_dp *intel_dp,
+				  const struct intel_crtc_state *crtc_state)
 {
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = dig_port->base.base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_crtc *crtc = dig_port->base.base.crtc;
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev);
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 
 	/* Transition from PSR_state 0 to PSR_state 1, i.e. PSR Inactive */
-	I915_WRITE(VLV_PSRCTL(pipe),
+	I915_WRITE(VLV_PSRCTL(crtc->pipe),
 		   VLV_EDP_PSR_MODE_SW_TIMER |
 		   VLV_EDP_PSR_SRC_TRANSMITTER_STATE |
 		   VLV_EDP_PSR_ENABLE);
@@ -485,16 +483,17 @@ static void intel_psr_activate(struct intel_dp *intel_dp)
 /**
  * intel_psr_enable - Enable PSR
  * @intel_dp: Intel DP
+ * @crtc_state: new CRTC state
  *
  * This function can only be called after the pipe is fully trained and enabled.
  */
-void intel_psr_enable(struct intel_dp *intel_dp)
+void intel_psr_enable(struct intel_dp *intel_dp,
+		      const struct intel_crtc_state *crtc_state)
 {
 	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(dev);
-	struct intel_crtc *crtc = to_intel_crtc(intel_dig_port->base.base.crtc);
-	enum transcoder cpu_transcoder = crtc->config->cpu_transcoder;
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 chicken;
 
 	if (!HAS_PSR(dev_priv)) {
@@ -520,11 +519,13 @@ void intel_psr_enable(struct intel_dp *intel_dp)
 
 	if (HAS_DDI(dev_priv)) {
 		if (dev_priv->psr.psr2_support) {
-			skl_psr_setup_su_vsc(intel_dp);
+			skl_psr_setup_su_vsc(intel_dp, crtc_state);
+
 			chicken = PSR2_VSC_ENABLE_PROG_HEADER;
 			if (dev_priv->psr.y_cord_support)
 				chicken |= PSR2_ADD_VERTICAL_LINE_COUNT;
 			I915_WRITE(CHICKEN_TRANS(cpu_transcoder), chicken);
+
 			I915_WRITE(EDP_PSR_DEBUG_CTL,
 				   EDP_PSR_DEBUG_MASK_MEMUP |
 				   EDP_PSR_DEBUG_MASK_HPD |
@@ -533,7 +534,8 @@ void intel_psr_enable(struct intel_dp *intel_dp)
 				   EDP_PSR_DEBUG_MASK_DISP_REG_WRITE);
 		} else {
 			/* set up vsc header for psr1 */
-			hsw_psr_setup_vsc(intel_dp);
+			hsw_psr_setup_vsc(intel_dp, crtc_state);
+
 			/*
 			 * Per Spec: Avoid continuous PSR exit by masking MEMUP
 			 * and HPD. also mask LPSP to avoid dependency on other
@@ -553,7 +555,7 @@ void intel_psr_enable(struct intel_dp *intel_dp)
 		if (INTEL_GEN(dev_priv) >= 9)
 			intel_psr_activate(intel_dp);
 	} else {
-		vlv_psr_setup_vsc(intel_dp);
+		vlv_psr_setup_vsc(intel_dp, crtc_state);
 
 		/* Enable PSR on the panel */
 		vlv_psr_enable_sink(intel_dp);
@@ -564,7 +566,7 @@ void intel_psr_enable(struct intel_dp *intel_dp)
 		 * but let it on inactive state. So we might do this prior
 		 * to active transition, i.e. here.
 		 */
-		vlv_psr_enable_source(intel_dp);
+		vlv_psr_enable_source(intel_dp, crtc_state);
 	}
 
 	/*
@@ -585,37 +587,38 @@ void intel_psr_enable(struct intel_dp *intel_dp)
 	mutex_unlock(&dev_priv->psr.lock);
 }
 
-static void vlv_psr_disable(struct intel_dp *intel_dp)
+static void vlv_psr_disable(struct intel_dp *intel_dp,
+			    const struct intel_crtc_state *old_crtc_state)
 {
 	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(dev);
-	struct intel_crtc *intel_crtc =
-		to_intel_crtc(intel_dig_port->base.base.crtc);
+	struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->base.crtc);
 	uint32_t val;
 
 	if (dev_priv->psr.active) {
 		/* Put VLV PSR back to PSR_state 0 that is PSR Disabled. */
 		if (intel_wait_for_register(dev_priv,
-					    VLV_PSRSTAT(intel_crtc->pipe),
+					    VLV_PSRSTAT(crtc->pipe),
 					    VLV_EDP_PSR_IN_TRANS,
 					    0,
 					    1))
 			WARN(1, "PSR transition took longer than expected\n");
 
-		val = I915_READ(VLV_PSRCTL(intel_crtc->pipe));
+		val = I915_READ(VLV_PSRCTL(crtc->pipe));
 		val &= ~VLV_EDP_PSR_ACTIVE_ENTRY;
 		val &= ~VLV_EDP_PSR_ENABLE;
 		val &= ~VLV_EDP_PSR_MODE_MASK;
-		I915_WRITE(VLV_PSRCTL(intel_crtc->pipe), val);
+		I915_WRITE(VLV_PSRCTL(crtc->pipe), val);
 
 		dev_priv->psr.active = false;
 	} else {
-		WARN_ON(vlv_is_psr_active_on_pipe(dev, intel_crtc->pipe));
+		WARN_ON(vlv_is_psr_active_on_pipe(dev, crtc->pipe));
 	}
 }
 
-static void hsw_psr_disable(struct intel_dp *intel_dp)
+static void hsw_psr_disable(struct intel_dp *intel_dp,
+			    const struct intel_crtc_state *old_crtc_state)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
@@ -664,10 +667,12 @@ static void hsw_psr_disable(struct intel_dp *intel_dp)
 /**
  * intel_psr_disable - Disable PSR
  * @intel_dp: Intel DP
+ * @old_crtc_state: old CRTC state
  *
  * This function needs to be called before disabling pipe.
  */
-void intel_psr_disable(struct intel_dp *intel_dp)
+void intel_psr_disable(struct intel_dp *intel_dp,
+		       const struct intel_crtc_state *old_crtc_state)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
@@ -681,9 +686,9 @@ void intel_psr_disable(struct intel_dp *intel_dp)
 
 	/* Disable PSR on Source */
 	if (HAS_DDI(dev_priv))
-		hsw_psr_disable(intel_dp);
+		hsw_psr_disable(intel_dp, old_crtc_state);
 	else
-		vlv_psr_disable(intel_dp);
+		vlv_psr_disable(intel_dp, old_crtc_state);
 
 	/* Disable PSR on Sink */
 	drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG, 0);

drivers/gpu/drm/i915/intel_ringbuffer.h

@@ -735,16 +735,6 @@ bool intel_engines_are_idle(struct drm_i915_private *dev_priv);
 void intel_engines_mark_idle(struct drm_i915_private *i915);
 void intel_engines_reset_default_submission(struct drm_i915_private *i915);
 
-static inline bool
-__intel_engine_can_store_dword(unsigned int gen, unsigned int class)
-{
-	if (gen <= 2)
-		return false; /* uses physical not virtual addresses */
-
-	if (gen == 6 && class == VIDEO_DECODE_CLASS)
-		return false; /* b0rked */
-
-	return true;
-}
+bool intel_engine_can_store_dword(struct intel_engine_cs *engine);
 
 #endif /* _INTEL_RINGBUFFER_H_ */

drivers/gpu/drm/i915/intel_uncore.c

@@ -1251,7 +1251,7 @@ static const struct register_whitelist {
 } whitelist[] = {
 	{ .offset_ldw = RING_TIMESTAMP(RENDER_RING_BASE),
 	  .offset_udw = RING_TIMESTAMP_UDW(RENDER_RING_BASE),
-	  .size = 8, .gen_bitmask = GEN_RANGE(4, 9) },
+	  .size = 8, .gen_bitmask = GEN_RANGE(4, 10) },
 };
 
 int i915_reg_read_ioctl(struct drm_device *dev,

include/uapi/drm/i915_drm.h

@@ -1509,9 +1509,9 @@ struct drm_i915_perf_oa_config {
 	__u32 n_boolean_regs;
 	__u32 n_flex_regs;
 
-	__u64 __user mux_regs_ptr;
-	__u64 __user boolean_regs_ptr;
-	__u64 __user flex_regs_ptr;
+	__u64 mux_regs_ptr;
+	__u64 boolean_regs_ptr;
+	__u64 flex_regs_ptr;
 };
 
 #if defined(__cplusplus)