Merge tag 'drm-intel-next-2017-10-12' of git://anongit.freedesktop.org/drm/drm-intel into drm-next

Last batch of drm/i915 features for v4.15:

- transparent huge pages support (Matthew)
- uapi: I915_PARAM_HAS_SCHEDULER into a capability bitmask (Chris)
- execlists: preemption (Chris)
- scheduler: user defined priorities (Chris)
- execlists optimization (Michał)
- plenty of display fixes (Imre)
- has_ipc fix (Rodrigo)
- platform features definition refactoring (Rodrigo)
- legacy cursor update fix (Maarten)
- fix vblank waits for cursor updates (Maarten)
- reprogram dmc firmware on resume, dmc state fix (Imre)
- remove use_mmio_flip module parameter (Maarten)
- wa fixes (Oscar)
- huc/guc firmware refacoring (Sagar, Michal)
- push encoder specific code to encoder hooks (Jani)
- DP MST fixes (Dhinakaran)
- eDP power sequencing fixes (Manasi)
- selftest updates (Chris, Matthew)
- mmu notifier cpu hotplug deadlock fix (Daniel)
- more VBT parser refactoring (Jani)
- max pipe refactoring (Mika Kahola)
- rc6/rps refactoring and separation (Sagar)
- userptr lockdep fix (Chris)
- tracepoint fixes and defunct tracepoint removal (Chris)
- use rcu instead of abusing stop_machine (Daniel)
- plenty of other fixes all around (Everyone)

* tag 'drm-intel-next-2017-10-12' of git://anongit.freedesktop.org/drm/drm-intel: (145 commits)
  drm/i915: Update DRIVER_DATE to 20171012
  drm/i915: Simplify intel_sanitize_enable_ppgtt
  drm/i915/userptr: Drop struct_mutex before cleanup
  drm/i915/dp: limit sink rates based on rate
  drm/i915/dp: centralize max source rate conditions more
  drm/i915: Allow PCH platforms fall back to BIOS LVDS mode
  drm/i915: Reuse normal state readout for LVDS/DVO fixed mode
  drm/i915: Use rcu instead of stop_machine in set_wedged
  drm/i915: Introduce separate status variable for RC6 and LLC ring frequency setup
  drm/i915: Create generic functions to control RC6, RPS
  drm/i915: Create generic function to setup LLC ring frequency table
  drm/i915: Rename intel_enable_rc6 to intel_rc6_enabled
  drm/i915: Name structure in dev_priv that contains RPS/RC6 state as "gt_pm"
  drm/i915: Move rps.hw_lock to dev_priv and s/hw_lock/pcu_lock
  drm/i915: Name i915_runtime_pm structure in dev_priv as "runtime_pm"
  drm/i915: Separate RPS and RC6 handling for CHV
  drm/i915: Separate RPS and RC6 handling for VLV
  drm/i915: Separate RPS and RC6 handling for BDW
  drm/i915: Remove superfluous IS_BDW checks and non-BDW changes from gen8_enable_rps
  drm/i915: Separate RPS and RC6 handling for gen6+
  ...

drivers/gpu/drm/i915/Makefile

@@ -47,6 +47,7 @@ i915-y += i915_cmd_parser.o \
 	  i915_gem_tiling.o \
 	  i915_gem_timeline.o \
 	  i915_gem_userptr.o \
+	  i915_gemfs.o \
 	  i915_trace_points.o \
 	  i915_vma.o \
 	  intel_breadcrumbs.o \
@@ -59,6 +60,8 @@ i915-y += i915_cmd_parser.o \
 
 # general-purpose microcontroller (GuC) support
 i915-y += intel_uc.o \
+	  intel_uc_fw.o \
+	  intel_guc.o \
 	  intel_guc_ct.o \
 	  intel_guc_log.o \
 	  intel_guc_loader.o \

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

@@ -174,6 +174,7 @@ static int shadow_context_status_change(struct notifier_block *nb,
 		atomic_set(&workload->shadow_ctx_active, 1);
 		break;
 	case INTEL_CONTEXT_SCHEDULE_OUT:
+	case INTEL_CONTEXT_SCHEDULE_PREEMPTED:
 		atomic_set(&workload->shadow_ctx_active, 0);
 		break;
 	default:

drivers/gpu/drm/i915/i915_drv.c

@@ -367,9 +367,18 @@ static int i915_getparam(struct drm_device *dev, void *data,
 		value = i915_gem_mmap_gtt_version();
 		break;
 	case I915_PARAM_HAS_SCHEDULER:
-		value = dev_priv->engine[RCS] &&
-			dev_priv->engine[RCS]->schedule;
+		value = 0;
+		if (dev_priv->engine[RCS] && dev_priv->engine[RCS]->schedule) {
+			value |= I915_SCHEDULER_CAP_ENABLED;
+			value |= I915_SCHEDULER_CAP_PRIORITY;
+
+			if (INTEL_INFO(dev_priv)->has_logical_ring_preemption &&
+			    i915_modparams.enable_execlists &&
+			    !i915_modparams.enable_guc_submission)
+				value |= I915_SCHEDULER_CAP_PREEMPTION;
+		}
 		break;
+
 	case I915_PARAM_MMAP_VERSION:
 		/* Remember to bump this if the version changes! */
 	case I915_PARAM_HAS_GEM:
@@ -606,9 +615,10 @@ static void i915_gem_fini(struct drm_i915_private *dev_priv)
 	intel_uc_fini_hw(dev_priv);
 	i915_gem_cleanup_engines(dev_priv);
 	i915_gem_contexts_fini(dev_priv);
-	i915_gem_cleanup_userptr(dev_priv);
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 
+	i915_gem_cleanup_userptr(dev_priv);
+
 	i915_gem_drain_freed_objects(dev_priv);
 
 	WARN_ON(!list_empty(&dev_priv->contexts.list));
@@ -1007,6 +1017,8 @@ static int i915_driver_init_mmio(struct drm_i915_private *dev_priv)
 
 	intel_uncore_init(dev_priv);
 
+	intel_uc_init_mmio(dev_priv);
+
 	ret = intel_engines_init_mmio(dev_priv);
 	if (ret)
 		goto err_uncore;
@@ -1580,7 +1592,7 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation)
 
 	intel_display_set_init_power(dev_priv, false);
 
-	fw_csr = !IS_GEN9_LP(dev_priv) &&
+	fw_csr = !IS_GEN9_LP(dev_priv) && !hibernation &&
 		suspend_to_idle(dev_priv) && dev_priv->csr.dmc_payload;
 	/*
 	 * In case of firmware assisted context save/restore don't manually
@@ -2070,11 +2082,14 @@ static int i915_pm_resume(struct device *kdev)
 /* freeze: before creating the hibernation_image */
 static int i915_pm_freeze(struct device *kdev)
 {
+	struct drm_device *dev = &kdev_to_i915(kdev)->drm;
 	int ret;
 
-	ret = i915_pm_suspend(kdev);
-	if (ret)
-		return ret;
+	if (dev->switch_power_state != DRM_SWITCH_POWER_OFF) {
+		ret = i915_drm_suspend(dev);
+		if (ret)
+			return ret;
+	}
 
 	ret = i915_gem_freeze(kdev_to_i915(kdev));
 	if (ret)
@@ -2085,11 +2100,14 @@ static int i915_pm_freeze(struct device *kdev)
 
 static int i915_pm_freeze_late(struct device *kdev)
 {
+	struct drm_device *dev = &kdev_to_i915(kdev)->drm;
 	int ret;
 
-	ret = i915_pm_suspend_late(kdev);
-	if (ret)
-		return ret;
+	if (dev->switch_power_state != DRM_SWITCH_POWER_OFF) {
+		ret = i915_drm_suspend_late(dev, true);
+		if (ret)
+			return ret;
+	}
 
 	ret = i915_gem_freeze_late(kdev_to_i915(kdev));
 	if (ret)
@@ -2485,7 +2503,7 @@ static int intel_runtime_suspend(struct device *kdev)
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	int ret;
 
-	if (WARN_ON_ONCE(!(dev_priv->rps.enabled && intel_enable_rc6())))
+	if (WARN_ON_ONCE(!(dev_priv->gt_pm.rc6.enabled && intel_rc6_enabled())))
 		return -ENODEV;
 
 	if (WARN_ON_ONCE(!HAS_RUNTIME_PM(dev_priv)))
@@ -2527,12 +2545,12 @@ static int intel_runtime_suspend(struct device *kdev)
 	intel_uncore_suspend(dev_priv);
 
 	enable_rpm_wakeref_asserts(dev_priv);
-	WARN_ON_ONCE(atomic_read(&dev_priv->pm.wakeref_count));
+	WARN_ON_ONCE(atomic_read(&dev_priv->runtime_pm.wakeref_count));
 
 	if (intel_uncore_arm_unclaimed_mmio_detection(dev_priv))
 		DRM_ERROR("Unclaimed access detected prior to suspending\n");
 
-	dev_priv->pm.suspended = true;
+	dev_priv->runtime_pm.suspended = true;
 
 	/*
 	 * FIXME: We really should find a document that references the arguments
@@ -2578,11 +2596,11 @@ static int intel_runtime_resume(struct device *kdev)
 
 	DRM_DEBUG_KMS("Resuming device\n");
 
-	WARN_ON_ONCE(atomic_read(&dev_priv->pm.wakeref_count));
+	WARN_ON_ONCE(atomic_read(&dev_priv->runtime_pm.wakeref_count));
 	disable_rpm_wakeref_asserts(dev_priv);
 
 	intel_opregion_notify_adapter(dev_priv, PCI_D0);
-	dev_priv->pm.suspended = false;
+	dev_priv->runtime_pm.suspended = false;
 	if (intel_uncore_unclaimed_mmio(dev_priv))
 		DRM_DEBUG_DRIVER("Unclaimed access during suspend, bios?\n");
 

drivers/gpu/drm/i915/i915_drv.h

@@ -80,8 +80,8 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20170929"
-#define DRIVER_TIMESTAMP	1506682238
+#define DRIVER_DATE		"20171012"
+#define DRIVER_TIMESTAMP	1507831511
 
 /* 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
@@ -609,7 +609,7 @@ struct drm_i915_file_private {
 
 	struct intel_rps_client {
 		atomic_t boosts;
-	} rps;
+	} rps_client;
 
 	unsigned int bsd_engine;
 
@@ -783,6 +783,7 @@ struct intel_csr {
 	func(has_l3_dpf); \
 	func(has_llc); \
 	func(has_logical_ring_contexts); \
+	func(has_logical_ring_preemption); \
 	func(has_overlay); \
 	func(has_pipe_cxsr); \
 	func(has_pooled_eu); \
@@ -868,6 +869,8 @@ struct intel_device_info {
 	u8 num_sprites[I915_MAX_PIPES];
 	u8 num_scalers[I915_MAX_PIPES];
 
+	unsigned int page_sizes; /* page sizes supported by the HW */
+
 #define DEFINE_FLAG(name) u8 name:1
 	DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG);
 #undef DEFINE_FLAG
@@ -981,6 +984,7 @@ struct i915_gpu_state {
 			pid_t pid;
 			u32 handle;
 			u32 hw_id;
+			int priority;
 			int ban_score;
 			int active;
 			int guilty;
@@ -1003,6 +1007,7 @@ struct i915_gpu_state {
 			long jiffies;
 			pid_t pid;
 			u32 context;
+			int priority;
 			int ban_score;
 			u32 seqno;
 			u32 head;
@@ -1312,7 +1317,7 @@ struct intel_rps_ei {
 	u32 media_c0;
 };
 
-struct intel_gen6_power_mgmt {
+struct intel_rps {
 	/*
 	 * work, interrupts_enabled and pm_iir are protected by
 	 * dev_priv->irq_lock
@@ -1353,20 +1358,26 @@ struct intel_gen6_power_mgmt {
 	enum { LOW_POWER, BETWEEN, HIGH_POWER } power;
 
 	bool enabled;
-	struct delayed_work autoenable_work;
 	atomic_t num_waiters;
 	atomic_t boosts;
 
 	/* manual wa residency calculations */
 	struct intel_rps_ei ei;
+};
 
-	/*
-	 * Protects RPS/RC6 register access and PCU communication.
-	 * Must be taken after struct_mutex if nested. Note that
-	 * this lock may be held for long periods of time when
-	 * talking to hw - so only take it when talking to hw!
-	 */
-	struct mutex hw_lock;
+struct intel_rc6 {
+	bool enabled;
+};
+
+struct intel_llc_pstate {
+	bool enabled;
+};
+
+struct intel_gen6_power_mgmt {
+	struct intel_rps rps;
+	struct intel_rc6 rc6;
+	struct intel_llc_pstate llc_pstate;
+	struct delayed_work autoenable_work;
 };
 
 /* defined intel_pm.c */
@@ -1508,6 +1519,11 @@ struct i915_gem_mm {
 	/** Usable portion of the GTT for GEM */
 	dma_addr_t stolen_base; /* limited to low memory (32-bit) */
 
+	/**
+	 * tmpfs instance used for shmem backed objects
+	 */
+	struct vfsmount *gemfs;
+
 	/** PPGTT used for aliasing the PPGTT with the GTT */
 	struct i915_hw_ppgtt *aliasing_ppgtt;
 
@@ -2251,8 +2267,11 @@ struct drm_i915_private {
 	wait_queue_head_t gmbus_wait_queue;
 
 	struct pci_dev *bridge_dev;
-	struct i915_gem_context *kernel_context;
 	struct intel_engine_cs *engine[I915_NUM_ENGINES];
+	/* Context used internally to idle the GPU and setup initial state */
+	struct i915_gem_context *kernel_context;
+	/* Context only to be used for injecting preemption commands */
+	struct i915_gem_context *preempt_context;
 	struct i915_vma *semaphore;
 
 	struct drm_dma_handle *status_page_dmah;
@@ -2408,8 +2427,16 @@ struct drm_i915_private {
 	/* Cannot be determined by PCIID. You must always read a register. */
 	u32 edram_cap;
 
-	/* gen6+ rps state */
-	struct intel_gen6_power_mgmt rps;
+	/*
+	 * Protects RPS/RC6 register access and PCU communication.
+	 * Must be taken after struct_mutex if nested. Note that
+	 * this lock may be held for long periods of time when
+	 * talking to hw - so only take it when talking to hw!
+	 */
+	struct mutex pcu_lock;
+
+	/* gen6+ GT PM state */
+	struct intel_gen6_power_mgmt gt_pm;
 
 	/* ilk-only ips/rps state. Everything in here is protected by the global
 	 * mchdev_lock in intel_pm.c */
@@ -2520,7 +2547,7 @@ struct drm_i915_private {
 		bool distrust_bios_wm;
 	} wm;
 
-	struct i915_runtime_pm pm;
+	struct i915_runtime_pm runtime_pm;
 
 	struct {
 		bool initialized;
@@ -2859,6 +2886,21 @@ static inline struct scatterlist *__sg_next(struct scatterlist *sg)
 	     (((__iter).curr += PAGE_SIZE) >= (__iter).max) ?		\
 	     (__iter) = __sgt_iter(__sg_next((__iter).sgp), false), 0 : 0)
 
+static inline unsigned int i915_sg_page_sizes(struct scatterlist *sg)
+{
+	unsigned int page_sizes;
+
+	page_sizes = 0;
+	while (sg) {
+		GEM_BUG_ON(sg->offset);
+		GEM_BUG_ON(!IS_ALIGNED(sg->length, PAGE_SIZE));
+		page_sizes |= sg->length;
+		sg = __sg_next(sg);
+	}
+
+	return page_sizes;
+}
+
 static inline unsigned int i915_sg_segment_size(void)
 {
 	unsigned int size = swiotlb_max_segment();
@@ -3088,6 +3130,10 @@ intel_info(const struct drm_i915_private *dev_priv)
 #define USES_PPGTT(dev_priv)		(i915_modparams.enable_ppgtt)
 #define USES_FULL_PPGTT(dev_priv)	(i915_modparams.enable_ppgtt >= 2)
 #define USES_FULL_48BIT_PPGTT(dev_priv)	(i915_modparams.enable_ppgtt == 3)
+#define HAS_PAGE_SIZES(dev_priv, sizes) ({ \
+	GEM_BUG_ON((sizes) == 0); \
+	((sizes) & ~(dev_priv)->info.page_sizes) == 0; \
+})
 
 #define HAS_OVERLAY(dev_priv)		 ((dev_priv)->info.has_overlay)
 #define OVERLAY_NEEDS_PHYSICAL(dev_priv) \
@@ -3504,7 +3550,8 @@ i915_gem_object_get_dma_address(struct drm_i915_gem_object *obj,
 				unsigned long n);
 
 void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
-				 struct sg_table *pages);
+				 struct sg_table *pages,
+				 unsigned int sg_page_sizes);
 int __i915_gem_object_get_pages(struct drm_i915_gem_object *obj);
 
 static inline int __must_check
@@ -3726,8 +3773,6 @@ 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);

drivers/gpu/drm/i915/i915_gem_context.c

@@ -416,14 +416,43 @@ i915_gem_context_create_gvt(struct drm_device *dev)
 	return ctx;
 }
 
+static struct i915_gem_context *
+create_kernel_context(struct drm_i915_private *i915, int prio)
+{
+	struct i915_gem_context *ctx;
+
+	ctx = i915_gem_create_context(i915, NULL);
+	if (IS_ERR(ctx))
+		return ctx;
+
+	i915_gem_context_clear_bannable(ctx);
+	ctx->priority = prio;
+	ctx->ring_size = PAGE_SIZE;
+
+	GEM_BUG_ON(!i915_gem_context_is_kernel(ctx));
+
+	return ctx;
+}
+
+static void
+destroy_kernel_context(struct i915_gem_context **ctxp)
+{
+	struct i915_gem_context *ctx;
+
+	/* Keep the context ref so that we can free it immediately ourselves */
+	ctx = i915_gem_context_get(fetch_and_zero(ctxp));
+	GEM_BUG_ON(!i915_gem_context_is_kernel(ctx));
+
+	context_close(ctx);
+	i915_gem_context_free(ctx);
+}
+
 int i915_gem_contexts_init(struct drm_i915_private *dev_priv)
 {
 	struct i915_gem_context *ctx;
+	int err;
 
-	/* Init should only be called once per module load. Eventually the
-	 * restriction on the context_disabled check can be loosened. */
-	if (WARN_ON(dev_priv->kernel_context))
-		return 0;
+	GEM_BUG_ON(dev_priv->kernel_context);
 
 	INIT_LIST_HEAD(&dev_priv->contexts.list);
 	INIT_WORK(&dev_priv->contexts.free_work, contexts_free_worker);
@@ -441,28 +470,38 @@ int i915_gem_contexts_init(struct drm_i915_private *dev_priv)
 	BUILD_BUG_ON(MAX_CONTEXT_HW_ID > INT_MAX);
 	ida_init(&dev_priv->contexts.hw_ida);
 
-	ctx = i915_gem_create_context(dev_priv, NULL);
+	/* lowest priority; idle task */
+	ctx = create_kernel_context(dev_priv, I915_PRIORITY_MIN);
 	if (IS_ERR(ctx)) {
-		DRM_ERROR("Failed to create default global context (error %ld)\n",
-			  PTR_ERR(ctx));
-		return PTR_ERR(ctx);
+		DRM_ERROR("Failed to create default global context\n");
+		err = PTR_ERR(ctx);
+		goto err;
 	}
-
-	/* For easy recognisablity, we want the kernel context to be 0 and then
+	/*
+	 * For easy recognisablity, we want the kernel context to be 0 and then
 	 * all user contexts will have non-zero hw_id.
 	 */
 	GEM_BUG_ON(ctx->hw_id);
-
-	i915_gem_context_clear_bannable(ctx);
-	ctx->priority = I915_PRIORITY_MIN; /* lowest priority; idle task */
 	dev_priv->kernel_context = ctx;
 
-	GEM_BUG_ON(!i915_gem_context_is_kernel(ctx));
+	/* highest priority; preempting task */
+	ctx = create_kernel_context(dev_priv, INT_MAX);
+	if (IS_ERR(ctx)) {
+		DRM_ERROR("Failed to create default preempt context\n");
+		err = PTR_ERR(ctx);
+		goto err_kernel_context;
+	}
+	dev_priv->preempt_context = ctx;
 
 	DRM_DEBUG_DRIVER("%s context support initialized\n",
 			 dev_priv->engine[RCS]->context_size ? "logical" :
 			 "fake");
 	return 0;
+
+err_kernel_context:
+	destroy_kernel_context(&dev_priv->kernel_context);
+err:
+	return err;
 }
 
 void i915_gem_contexts_lost(struct drm_i915_private *dev_priv)
@@ -507,15 +546,10 @@ void i915_gem_contexts_lost(struct drm_i915_private *dev_priv)
 
 void i915_gem_contexts_fini(struct drm_i915_private *i915)
 {
-	struct i915_gem_context *ctx;
-
 	lockdep_assert_held(&i915->drm.struct_mutex);
 
-	/* Keep the context so that we can free it immediately ourselves */
-	ctx = i915_gem_context_get(fetch_and_zero(&i915->kernel_context));
-	GEM_BUG_ON(!i915_gem_context_is_kernel(ctx));
-	context_close(ctx);
-	i915_gem_context_free(ctx);
+	destroy_kernel_context(&i915->preempt_context);
+	destroy_kernel_context(&i915->kernel_context);
 
 	/* Must free all deferred contexts (via flush_workqueue) first */
 	ida_destroy(&i915->contexts.hw_ida);
@@ -1036,6 +1070,9 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data,
 	case I915_CONTEXT_PARAM_BANNABLE:
 		args->value = i915_gem_context_is_bannable(ctx);
 		break;
+	case I915_CONTEXT_PARAM_PRIORITY:
+		args->value = ctx->priority;
+		break;
 	default:
 		ret = -EINVAL;
 		break;
@@ -1091,6 +1128,26 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
 		else
 			i915_gem_context_clear_bannable(ctx);
 		break;
+
+	case I915_CONTEXT_PARAM_PRIORITY:
+		{
+			int priority = args->value;
+
+			if (args->size)
+				ret = -EINVAL;
+			else if (!to_i915(dev)->engine[RCS]->schedule)
+				ret = -ENODEV;
+			else if (priority > I915_CONTEXT_MAX_USER_PRIORITY ||
+				 priority < I915_CONTEXT_MIN_USER_PRIORITY)
+				ret = -EINVAL;
+			else if (priority > I915_CONTEXT_DEFAULT_PRIORITY &&
+				 !capable(CAP_SYS_NICE))
+				ret = -EPERM;
+			else
+				ctx->priority = priority;
+		}
+		break;
+
 	default:
 		ret = -EINVAL;
 		break;

drivers/gpu/drm/i915/i915_gem_dmabuf.c

@@ -256,11 +256,21 @@ struct dma_buf *i915_gem_prime_export(struct drm_device *dev,
 	return drm_gem_dmabuf_export(dev, &exp_info);
 }
 
-static struct sg_table *
-i915_gem_object_get_pages_dmabuf(struct drm_i915_gem_object *obj)
+static int i915_gem_object_get_pages_dmabuf(struct drm_i915_gem_object *obj)
 {
-	return dma_buf_map_attachment(obj->base.import_attach,
-				      DMA_BIDIRECTIONAL);
+	struct sg_table *pages;
+	unsigned int sg_page_sizes;
+
+	pages = dma_buf_map_attachment(obj->base.import_attach,
+				       DMA_BIDIRECTIONAL);
+	if (IS_ERR(pages))
+		return PTR_ERR(pages);
+
+	sg_page_sizes = i915_sg_page_sizes(pages->sgl);
+
+	__i915_gem_object_set_pages(obj, pages, sg_page_sizes);
+
+	return 0;
 }
 
 static void i915_gem_object_put_pages_dmabuf(struct drm_i915_gem_object *obj,

drivers/gpu/drm/i915/i915_gem_evict.c

@@ -82,7 +82,7 @@ mark_free(struct drm_mm_scan *scan,
 	if (i915_vma_is_pinned(vma))
 		return false;
 
-	if (flags & PIN_NONFAULT && !list_empty(&vma->obj->userfault_link))
+	if (flags & PIN_NONFAULT && i915_vma_has_userfault(vma))
 		return false;
 
 	list_add(&vma->evict_link, unwind);
@@ -315,6 +315,11 @@ int i915_gem_evict_for_node(struct i915_address_space *vm,
 			break;
 		}
 
+		if (flags & PIN_NONFAULT && i915_vma_has_userfault(vma)) {
+			ret = -ENOSPC;
+			break;
+		}
+
 		/* Overlap of objects in the same batch? */
 		if (i915_vma_is_pinned(vma)) {
 			ret = -ENOSPC;

drivers/gpu/drm/i915/i915_gem_execbuffer.c

@@ -367,12 +367,12 @@ eb_pin_vma(struct i915_execbuffer *eb,
 		return false;
 
 	if (unlikely(exec_flags & EXEC_OBJECT_NEEDS_FENCE)) {
-		if (unlikely(i915_vma_get_fence(vma))) {
+		if (unlikely(i915_vma_pin_fence(vma))) {
 			i915_vma_unpin(vma);
 			return false;
 		}
 
-		if (i915_vma_pin_fence(vma))
+		if (vma->fence)
 			exec_flags |= __EXEC_OBJECT_HAS_FENCE;
 	}
 
@@ -385,7 +385,7 @@ static inline void __eb_unreserve_vma(struct i915_vma *vma, unsigned int flags)
 	GEM_BUG_ON(!(flags & __EXEC_OBJECT_HAS_PIN));
 
 	if (unlikely(flags & __EXEC_OBJECT_HAS_FENCE))
-		i915_vma_unpin_fence(vma);
+		__i915_vma_unpin_fence(vma);
 
 	__i915_vma_unpin(vma);
 }
@@ -563,13 +563,13 @@ static int eb_reserve_vma(const struct i915_execbuffer *eb,
 	}
 
 	if (unlikely(exec_flags & EXEC_OBJECT_NEEDS_FENCE)) {
-		err = i915_vma_get_fence(vma);
+		err = i915_vma_pin_fence(vma);
 		if (unlikely(err)) {
 			i915_vma_unpin(vma);
 			return err;
 		}
 
-		if (i915_vma_pin_fence(vma))
+		if (vma->fence)
 			exec_flags |= __EXEC_OBJECT_HAS_FENCE;
 	}
 
@@ -974,7 +974,9 @@ static void *reloc_iomap(struct drm_i915_gem_object *obj,
 			return ERR_PTR(err);
 
 		vma = i915_gem_object_ggtt_pin(obj, NULL, 0, 0,
-					       PIN_MAPPABLE | PIN_NONBLOCK);
+					       PIN_MAPPABLE |
+					       PIN_NONBLOCK |
+					       PIN_NONFAULT);
 		if (IS_ERR(vma)) {
 			memset(&cache->node, 0, sizeof(cache->node));
 			err = drm_mm_insert_node_in_range

drivers/gpu/drm/i915/i915_gem_fence_reg.c

@@ -240,7 +240,8 @@ static int fence_update(struct drm_i915_fence_reg *fence,
 		/* Ensure that all userspace CPU access is completed before
 		 * stealing the fence.
 		 */
-		i915_gem_release_mmap(fence->vma->obj);
+		GEM_BUG_ON(fence->vma->fence != fence);
+		i915_vma_revoke_mmap(fence->vma);
 
 		fence->vma->fence = NULL;
 		fence->vma = NULL;
@@ -280,8 +281,7 @@ static int fence_update(struct drm_i915_fence_reg *fence,
  *
  * 0 on success, negative error code on failure.
  */
-int
-i915_vma_put_fence(struct i915_vma *vma)
+int i915_vma_put_fence(struct i915_vma *vma)
 {
 	struct drm_i915_fence_reg *fence = vma->fence;
 
@@ -299,6 +299,8 @@ static struct drm_i915_fence_reg *fence_find(struct drm_i915_private *dev_priv)
 	struct drm_i915_fence_reg *fence;
 
 	list_for_each_entry(fence, &dev_priv->mm.fence_list, link) {
+		GEM_BUG_ON(fence->vma && fence->vma->fence != fence);
+
 		if (fence->pin_count)
 			continue;
 
@@ -313,7 +315,7 @@ static struct drm_i915_fence_reg *fence_find(struct drm_i915_private *dev_priv)
 }
 
 /**
- * i915_vma_get_fence - set up fencing for a vma
+ * i915_vma_pin_fence - set up fencing for a vma
  * @vma: vma to map through a fence reg
  *
  * When mapping objects through the GTT, userspace wants to be able to write
@@ -331,10 +333,11 @@ static struct drm_i915_fence_reg *fence_find(struct drm_i915_private *dev_priv)
  * 0 on success, negative error code on failure.
  */
 int
-i915_vma_get_fence(struct i915_vma *vma)
+i915_vma_pin_fence(struct i915_vma *vma)
 {
 	struct drm_i915_fence_reg *fence;
 	struct i915_vma *set = i915_gem_object_is_tiled(vma->obj) ? vma : NULL;
+	int err;
 
 	/* Note that we revoke fences on runtime suspend. Therefore the user
 	 * must keep the device awake whilst using the fence.
@@ -344,6 +347,8 @@ i915_vma_get_fence(struct i915_vma *vma)
 	/* Just update our place in the LRU if our fence is getting reused. */
 	if (vma->fence) {
 		fence = vma->fence;
+		GEM_BUG_ON(fence->vma != vma);
+		fence->pin_count++;
 		if (!fence->dirty) {
 			list_move_tail(&fence->link,
 				       &fence->i915->mm.fence_list);
@@ -353,10 +358,25 @@ i915_vma_get_fence(struct i915_vma *vma)
 		fence = fence_find(vma->vm->i915);
 		if (IS_ERR(fence))
 			return PTR_ERR(fence);
+
+		GEM_BUG_ON(fence->pin_count);
+		fence->pin_count++;
 	} else
 		return 0;
 
-	return fence_update(fence, set);
+	err = fence_update(fence, set);
+	if (err)
+		goto out_unpin;
+
+	GEM_BUG_ON(fence->vma != set);
+	GEM_BUG_ON(vma->fence != (set ? fence : NULL));
+
+	if (set)
+		return 0;
+
+out_unpin:
+	fence->pin_count--;
+	return err;
 }
 
 /**
@@ -429,8 +449,10 @@ void i915_gem_revoke_fences(struct drm_i915_private *dev_priv)
 	for (i = 0; i < dev_priv->num_fence_regs; i++) {
 		struct drm_i915_fence_reg *fence = &dev_priv->fence_regs[i];
 
+		GEM_BUG_ON(fence->vma && fence->vma->fence != fence);
+
 		if (fence->vma)
-			i915_gem_release_mmap(fence->vma->obj);
+			i915_vma_revoke_mmap(fence->vma);
 	}
 }
 
@@ -450,13 +472,15 @@ void i915_gem_restore_fences(struct drm_i915_private *dev_priv)
 		struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i];
 		struct i915_vma *vma = reg->vma;
 
+		GEM_BUG_ON(vma && vma->fence != reg);
+
 		/*
 		 * Commit delayed tiling changes if we have an object still
 		 * attached to the fence, otherwise just clear the fence.
 		 */
 		if (vma && !i915_gem_object_is_tiled(vma->obj)) {
 			GEM_BUG_ON(!reg->dirty);
-			GEM_BUG_ON(!list_empty(&vma->obj->userfault_link));
+			GEM_BUG_ON(i915_vma_has_userfault(vma));
 
 			list_move(&reg->link, &dev_priv->mm.fence_list);
 			vma->fence = NULL;

drivers/gpu/drm/i915/i915_gem_gtt.h

@@ -42,7 +42,13 @@
 #include "i915_gem_request.h"
 #include "i915_selftest.h"
 
-#define I915_GTT_PAGE_SIZE 4096UL
+#define I915_GTT_PAGE_SIZE_4K BIT(12)
+#define I915_GTT_PAGE_SIZE_64K BIT(16)
+#define I915_GTT_PAGE_SIZE_2M BIT(21)
+
+#define I915_GTT_PAGE_SIZE I915_GTT_PAGE_SIZE_4K
+#define I915_GTT_MAX_PAGE_SIZE I915_GTT_PAGE_SIZE_2M
+
 #define I915_GTT_MIN_ALIGNMENT I915_GTT_PAGE_SIZE
 
 #define I915_FENCE_REG_NONE -1
@@ -148,6 +154,9 @@ typedef u64 gen8_ppgtt_pml4e_t;
 #define GEN8_PPAT_GET_AGE(x) ((x) & (3 << 4))
 #define CHV_PPAT_GET_SNOOP(x) ((x) & (1 << 6))
 
+#define GEN8_PDE_IPS_64K BIT(11)
+#define GEN8_PDE_PS_2M   BIT(7)
+
 struct sg_table;
 
 struct intel_rotation_info {
@@ -207,6 +216,7 @@ struct i915_vma;
 
 struct i915_page_dma {
 	struct page *page;
+	int order;
 	union {
 		dma_addr_t daddr;
 
@@ -329,6 +339,8 @@ struct i915_address_space {
 	int (*bind_vma)(struct i915_vma *vma,
 			enum i915_cache_level cache_level,
 			u32 flags);
+	int (*set_pages)(struct i915_vma *vma);
+	void (*clear_pages)(struct i915_vma *vma);
 
 	I915_SELFTEST_DECLARE(struct fault_attr fault_attr);
 };
@@ -341,6 +353,12 @@ i915_vm_is_48bit(const struct i915_address_space *vm)
 	return (vm->total - 1) >> 32;
 }
 
+static inline bool
+i915_vm_has_scratch_64K(struct i915_address_space *vm)
+{
+	return vm->scratch_page.order == get_order(I915_GTT_PAGE_SIZE_64K);
+}
+
 /* The Graphics Translation Table is the way in which GEN hardware translates a
  * Graphics Virtual Address into a Physical Address. In addition to the normal
  * collateral associated with any va->pa translations GEN hardware also has a

drivers/gpu/drm/i915/i915_gem_internal.c

@@ -44,12 +44,12 @@ static void internal_free_pages(struct sg_table *st)
 	kfree(st);
 }
 
-static struct sg_table *
-i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj)
+static int i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
 	struct sg_table *st;
 	struct scatterlist *sg;
+	unsigned int sg_page_sizes;
 	unsigned int npages;
 	int max_order;
 	gfp_t gfp;
@@ -78,16 +78,17 @@ i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj)
 create_st:
 	st = kmalloc(sizeof(*st), GFP_KERNEL);
 	if (!st)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 
 	npages = obj->base.size / PAGE_SIZE;
 	if (sg_alloc_table(st, npages, GFP_KERNEL)) {
 		kfree(st);
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 	}
 
 	sg = st->sgl;
 	st->nents = 0;
+	sg_page_sizes = 0;
 
 	do {
 		int order = min(fls(npages) - 1, max_order);
@@ -105,6 +106,7 @@ i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj)
 		} while (1);
 
 		sg_set_page(sg, page, PAGE_SIZE << order, 0);
+		sg_page_sizes |= PAGE_SIZE << order;
 		st->nents++;
 
 		npages -= 1 << order;
@@ -132,13 +134,17 @@ i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj)
 	 * object are only valid whilst active and pinned.
 	 */
 	obj->mm.madv = I915_MADV_DONTNEED;
-	return st;
+
+	__i915_gem_object_set_pages(obj, st, sg_page_sizes);
+
+	return 0;
 
 err:
 	sg_set_page(sg, NULL, 0, 0);
 	sg_mark_end(sg);
 	internal_free_pages(st);
-	return ERR_PTR(-ENOMEM);
+
+	return -ENOMEM;
 }
 
 static void i915_gem_object_put_pages_internal(struct drm_i915_gem_object *obj,

drivers/gpu/drm/i915/i915_gem_object.h

@@ -69,7 +69,7 @@ struct drm_i915_gem_object_ops {
 	 * being released or under memory pressure (where we attempt to
 	 * reap pages for the shrinker).
 	 */
-	struct sg_table *(*get_pages)(struct drm_i915_gem_object *);
+	int (*get_pages)(struct drm_i915_gem_object *);
 	void (*put_pages)(struct drm_i915_gem_object *, struct sg_table *);
 
 	int (*pwrite)(struct drm_i915_gem_object *,
@@ -123,6 +123,7 @@ struct drm_i915_gem_object {
 	/**
 	 * Whether the object is currently in the GGTT mmap.
 	 */
+	unsigned int userfault_count;
 	struct list_head userfault_link;
 
 	struct list_head batch_pool_link;
@@ -169,6 +170,35 @@ struct drm_i915_gem_object {
 		struct sg_table *pages;
 		void *mapping;
 
+		/* TODO: whack some of this into the error state */
+		struct i915_page_sizes {
+			/**
+			 * The sg mask of the pages sg_table. i.e the mask of
+			 * of the lengths for each sg entry.
+			 */
+			unsigned int phys;
+
+			/**
+			 * The gtt page sizes we are allowed to use given the
+			 * sg mask and the supported page sizes. This will
+			 * express the smallest unit we can use for the whole
+			 * object, as well as the larger sizes we may be able
+			 * to use opportunistically.
+			 */
+			unsigned int sg;
+
+			/**
+			 * The actual gtt page size usage. Since we can have
+			 * multiple vma associated with this object we need to
+			 * prevent any trampling of state, hence a copy of this
+			 * struct also lives in each vma, therefore the gtt
+			 * value here should only be read/write through the vma.
+			 */
+			unsigned int gtt;
+		} page_sizes;
+
+		I915_SELFTEST_DECLARE(unsigned int page_mask);
+
 		struct i915_gem_object_page_iter {
 			struct scatterlist *sg_pos;
 			unsigned int sg_idx; /* in pages, but 32bit eek! */

drivers/gpu/drm/i915/i915_gem_request.c

@@ -186,7 +186,7 @@ i915_priotree_init(struct i915_priotree *pt)
 	INIT_LIST_HEAD(&pt->signalers_list);
 	INIT_LIST_HEAD(&pt->waiters_list);
 	INIT_LIST_HEAD(&pt->link);
-	pt->priority = INT_MIN;
+	pt->priority = I915_PRIORITY_INVALID;
 }
 
 static int reset_all_global_seqno(struct drm_i915_private *i915, u32 seqno)
@@ -416,7 +416,7 @@ static void i915_gem_request_retire(struct drm_i915_gem_request *request)
 
 	spin_lock_irq(&request->lock);
 	if (request->waitboost)
-		atomic_dec(&request->i915->rps.num_waiters);
+		atomic_dec(&request->i915->gt_pm.rps.num_waiters);
 	dma_fence_signal_locked(&request->fence);
 	spin_unlock_irq(&request->lock);
 
@@ -556,7 +556,16 @@ submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
 	switch (state) {
 	case FENCE_COMPLETE:
 		trace_i915_gem_request_submit(request);
+		/*
+		 * We need to serialize use of the submit_request() callback with its
+		 * hotplugging performed during an emergency i915_gem_set_wedged().
+		 * We use the RCU mechanism to mark the critical section in order to
+		 * force i915_gem_set_wedged() to wait until the submit_request() is
+		 * completed before proceeding.
+		 */
+		rcu_read_lock();
 		request->engine->submit_request(request);
+		rcu_read_unlock();
 		break;
 
 	case FENCE_FREE:
@@ -587,6 +596,13 @@ i915_gem_request_alloc(struct intel_engine_cs *engine,
 
 	lockdep_assert_held(&dev_priv->drm.struct_mutex);
 
+	/*
+	 * Preempt contexts are reserved for exclusive use to inject a
+	 * preemption context switch. They are never to be used for any trivial
+	 * request!
+	 */
+	GEM_BUG_ON(ctx == dev_priv->preempt_context);
+
 	/* ABI: Before userspace accesses the GPU (e.g. execbuffer), report
 	 * EIO if the GPU is already wedged.
 	 */

drivers/gpu/drm/i915/i915_gem_request.h

@@ -30,6 +30,8 @@
 #include "i915_gem.h"
 #include "i915_sw_fence.h"
 
+#include <uapi/drm/i915_drm.h>
+
 struct drm_file;
 struct drm_i915_gem_object;
 struct drm_i915_gem_request;
@@ -69,9 +71,14 @@ struct i915_priotree {
 	struct list_head waiters_list; /* those after us, they depend upon us */
 	struct list_head link;
 	int priority;
-#define I915_PRIORITY_MAX 1024
-#define I915_PRIORITY_NORMAL 0
-#define I915_PRIORITY_MIN (-I915_PRIORITY_MAX)
+};
+
+enum {
+	I915_PRIORITY_MIN = I915_CONTEXT_MIN_USER_PRIORITY - 1,
+	I915_PRIORITY_NORMAL = I915_CONTEXT_DEFAULT_PRIORITY,
+	I915_PRIORITY_MAX = I915_CONTEXT_MAX_USER_PRIORITY + 1,
+
+	I915_PRIORITY_INVALID = INT_MIN
 };
 
 struct i915_gem_capture_list {

drivers/gpu/drm/i915/i915_gem_stolen.c

@@ -539,12 +539,18 @@ i915_pages_create_for_stolen(struct drm_device *dev,
 	return st;
 }
 
-static struct sg_table *
-i915_gem_object_get_pages_stolen(struct drm_i915_gem_object *obj)
+static int i915_gem_object_get_pages_stolen(struct drm_i915_gem_object *obj)
 {
-	return i915_pages_create_for_stolen(obj->base.dev,
-					    obj->stolen->start,
-					    obj->stolen->size);
+	struct sg_table *pages =
+		i915_pages_create_for_stolen(obj->base.dev,
+					     obj->stolen->start,
+					     obj->stolen->size);
+	if (IS_ERR(pages))
+		return PTR_ERR(pages);
+
+	__i915_gem_object_set_pages(obj, pages, obj->stolen->size);
+
+	return 0;
 }
 
 static void i915_gem_object_put_pages_stolen(struct drm_i915_gem_object *obj,

drivers/gpu/drm/i915/i915_gem_userptr.c

@@ -164,7 +164,6 @@ static struct i915_mmu_notifier *
 i915_mmu_notifier_create(struct mm_struct *mm)
 {
 	struct i915_mmu_notifier *mn;
-	int ret;
 
 	mn = kmalloc(sizeof(*mn), GFP_KERNEL);
 	if (mn == NULL)
@@ -179,14 +178,6 @@ i915_mmu_notifier_create(struct mm_struct *mm)
 		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);
-	}
-
 	return mn;
 }
 
@@ -210,23 +201,40 @@ i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj)
 static struct i915_mmu_notifier *
 i915_mmu_notifier_find(struct i915_mm_struct *mm)
 {
-	struct i915_mmu_notifier *mn = mm->mn;
+	struct i915_mmu_notifier *mn;
+	int err = 0;
 
 	mn = mm->mn;
 	if (mn)
 		return mn;
 
+	mn = i915_mmu_notifier_create(mm->mm);
+	if (IS_ERR(mn))
+		err = PTR_ERR(mn);
+
 	down_write(&mm->mm->mmap_sem);
 	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;
+	if (mm->mn == NULL && !err) {
+		/* Protected by mmap_sem (write-lock) */
+		err = __mmu_notifier_register(&mn->mn, mm->mm);
+		if (!err) {
+			/* Protected by mm_lock */
+			mm->mn = fetch_and_zero(&mn);
+		}
+	} else {
+		/* someone else raced and successfully installed the mmu
+		 * notifier, we can cancel our own errors */
+		err = 0;
 	}
 	mutex_unlock(&mm->i915->mm_lock);
 	up_write(&mm->mm->mmap_sem);
 
-	return mn;
+	if (mn) {
+		destroy_workqueue(mn->wq);
+		kfree(mn);
+	}
+
+	return err ? ERR_PTR(err) : mm->mn;
 }
 
 static int
@@ -405,6 +413,7 @@ __i915_gem_userptr_alloc_pages(struct drm_i915_gem_object *obj,
 {
 	unsigned int max_segment = i915_sg_segment_size();
 	struct sg_table *st;
+	unsigned int sg_page_sizes;
 	int ret;
 
 	st = kmalloc(sizeof(*st), GFP_KERNEL);
@@ -434,6 +443,10 @@ __i915_gem_userptr_alloc_pages(struct drm_i915_gem_object *obj,
 		return ERR_PTR(ret);
 	}
 
+	sg_page_sizes = i915_sg_page_sizes(st->sgl);
+
+	__i915_gem_object_set_pages(obj, st, sg_page_sizes);
+
 	return st;
 }
 
@@ -521,7 +534,6 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 			pages = __i915_gem_userptr_alloc_pages(obj, pvec,
 							       npages);
 			if (!IS_ERR(pages)) {
-				__i915_gem_object_set_pages(obj, pages);
 				pinned = 0;
 				pages = NULL;
 			}
@@ -582,8 +594,7 @@ __i915_gem_userptr_get_pages_schedule(struct drm_i915_gem_object *obj)
 	return ERR_PTR(-EAGAIN);
 }
 
-static struct sg_table *
-i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
+static int i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 {
 	const int num_pages = obj->base.size >> PAGE_SHIFT;
 	struct mm_struct *mm = obj->userptr.mm->mm;
@@ -612,9 +623,9 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 	if (obj->userptr.work) {
 		/* active flag should still be held for the pending work */
 		if (IS_ERR(obj->userptr.work))
-			return ERR_CAST(obj->userptr.work);
+			return PTR_ERR(obj->userptr.work);
 		else
-			return ERR_PTR(-EAGAIN);
+			return -EAGAIN;
 	}
 
 	pvec = NULL;
@@ -650,7 +661,7 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 		release_pages(pvec, pinned, 0);
 	kvfree(pvec);
 
-	return pages;
+	return PTR_ERR_OR_ZERO(pages);
 }
 
 static void

drivers/gpu/drm/i915/i915_gemfs.c

+/*
+ * Copyright © 2017 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 <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/pagemap.h>
+
+#include "i915_drv.h"
+#include "i915_gemfs.h"
+
+int i915_gemfs_init(struct drm_i915_private *i915)
+{
+	struct file_system_type *type;
+	struct vfsmount *gemfs;
+
+	type = get_fs_type("tmpfs");
+	if (!type)
+		return -ENODEV;
+
+	gemfs = kern_mount(type);
+	if (IS_ERR(gemfs))
+		return PTR_ERR(gemfs);
+
+	/*
+	 * Enable huge-pages for objects that are at least HPAGE_PMD_SIZE, most
+	 * likely 2M. Note that within_size may overallocate huge-pages, if say
+	 * we allocate an object of size 2M + 4K, we may get 2M + 2M, but under
+	 * memory pressure shmem should split any huge-pages which can be
+	 * shrunk.
+	 */
+
+	if (has_transparent_hugepage()) {
+		struct super_block *sb = gemfs->mnt_sb;
+		char options[] = "huge=within_size";
+		int flags = 0;
+		int err;
+
+		err = sb->s_op->remount_fs(sb, &flags, options);
+		if (err) {
+			kern_unmount(gemfs);
+			return err;
+		}
+	}
+
+	i915->mm.gemfs = gemfs;
+
+	return 0;
+}
+
+void i915_gemfs_fini(struct drm_i915_private *i915)
+{
+	kern_unmount(i915->mm.gemfs);
+}

drivers/gpu/drm/i915/i915_gemfs.h

+/*
+ * Copyright © 2017 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_GEMFS_H__
+#define __I915_GEMFS_H__
+
+struct drm_i915_private;
+
+int i915_gemfs_init(struct drm_i915_private *i915);
+
+void i915_gemfs_fini(struct drm_i915_private *i915);
+
+#endif

drivers/gpu/drm/i915/i915_gpu_error.c

@@ -377,9 +377,9 @@ static void error_print_request(struct drm_i915_error_state_buf *m,
 	if (!erq->seqno)
 		return;
 
-	err_printf(m, "%s pid %d, ban score %d, seqno %8x:%08x, emitted %dms ago, head %08x, tail %08x\n",
+	err_printf(m, "%s pid %d, ban score %d, seqno %8x:%08x, prio %d, emitted %dms ago, head %08x, tail %08x\n",
 		   prefix, erq->pid, erq->ban_score,
-		   erq->context, erq->seqno,
+		   erq->context, erq->seqno, erq->priority,
 		   jiffies_to_msecs(jiffies - erq->jiffies),
 		   erq->head, erq->tail);
 }
@@ -388,9 +388,9 @@ static void error_print_context(struct drm_i915_error_state_buf *m,
 				const char *header,
 				const struct drm_i915_error_context *ctx)
 {
-	err_printf(m, "%s%s[%d] user_handle %d hw_id %d, ban score %d guilty %d active %d\n",
+	err_printf(m, "%s%s[%d] user_handle %d hw_id %d, prio %d, ban score %d guilty %d active %d\n",
 		   header, ctx->comm, ctx->pid, ctx->handle, ctx->hw_id,
-		   ctx->ban_score, ctx->guilty, ctx->active);
+		   ctx->priority, ctx->ban_score, ctx->guilty, ctx->active);
 }
 
 static void error_print_engine(struct drm_i915_error_state_buf *m,
@@ -1271,6 +1271,7 @@ static void record_request(struct drm_i915_gem_request *request,
 			   struct drm_i915_error_request *erq)
 {
 	erq->context = request->ctx->hw_id;
+	erq->priority = request->priotree.priority;
 	erq->ban_score = atomic_read(&request->ctx->ban_score);
 	erq->seqno = request->global_seqno;
 	erq->jiffies = request->emitted_jiffies;
@@ -1364,6 +1365,7 @@ static void record_context(struct drm_i915_error_context *e,
 
 	e->handle = ctx->user_handle;
 	e->hw_id = ctx->hw_id;
+	e->priority = ctx->priority;
 	e->ban_score = atomic_read(&ctx->ban_score);
 	e->guilty = atomic_read(&ctx->guilty_count);
 	e->active = atomic_read(&ctx->active_count);
@@ -1672,8 +1674,8 @@ static void i915_capture_gen_state(struct drm_i915_private *dev_priv,
 				   struct i915_gpu_state *error)
 {
 	error->awake = dev_priv->gt.awake;
-	error->wakelock = atomic_read(&dev_priv->pm.wakeref_count);
-	error->suspended = dev_priv->pm.suspended;
+	error->wakelock = atomic_read(&dev_priv->runtime_pm.wakeref_count);
+	error->suspended = dev_priv->runtime_pm.suspended;
 
 	error->iommu = -1;
 #ifdef CONFIG_INTEL_IOMMU

drivers/gpu/drm/i915/i915_guc_submission.c

@@ -21,12 +21,13 @@
  * IN THE SOFTWARE.
  *
  */
-#include <linux/circ_buf.h>
-#include "i915_drv.h"
-#include "intel_uc.h"
 
+#include <linux/circ_buf.h>
 #include <trace/events/dma_fence.h>
 
+#include "i915_guc_submission.h"
+#include "i915_drv.h"
+
 /**
  * DOC: GuC-based command submission
  *
@@ -337,7 +338,7 @@ static void guc_stage_desc_init(struct intel_guc *guc,
 
 	for_each_engine_masked(engine, dev_priv, client->engines, tmp) {
 		struct intel_context *ce = &ctx->engine[engine->id];
-		uint32_t guc_engine_id = engine->guc_id;
+		u32 guc_engine_id = engine->guc_id;
 		struct guc_execlist_context *lrc = &desc->lrc[guc_engine_id];
 
 		/* TODO: We have a design issue to be solved here. Only when we
@@ -387,13 +388,13 @@ static void guc_stage_desc_init(struct intel_guc *guc,
 	gfx_addr = guc_ggtt_offset(client->vma);
 	desc->db_trigger_phy = sg_dma_address(client->vma->pages->sgl) +
 				client->doorbell_offset;
-	desc->db_trigger_cpu = (uintptr_t)__get_doorbell(client);
+	desc->db_trigger_cpu = ptr_to_u64(__get_doorbell(client));
 	desc->db_trigger_uk = gfx_addr + client->doorbell_offset;
 	desc->process_desc = gfx_addr + client->proc_desc_offset;
 	desc->wq_addr = gfx_addr + GUC_DB_SIZE;
 	desc->wq_size = GUC_WQ_SIZE;
 
-	desc->desc_private = (uintptr_t)client;
+	desc->desc_private = ptr_to_u64(client);
 }
 
 static void guc_stage_desc_fini(struct intel_guc *guc,
@@ -499,7 +500,7 @@ static void i915_guc_submit(struct intel_engine_cs *engine)
 	const unsigned int engine_id = engine->id;
 	unsigned int n;
 
-	for (n = 0; n < ARRAY_SIZE(execlists->port); n++) {
+	for (n = 0; n < execlists_num_ports(execlists); n++) {
 		struct drm_i915_gem_request *rq;
 		unsigned int count;
 
@@ -643,48 +644,6 @@ static void i915_guc_irq_handler(unsigned long data)
  * path of i915_guc_submit() above.
  */
 
-/**
- * intel_guc_allocate_vma() - Allocate a GGTT VMA for GuC usage
- * @guc:	the guc
- * @size:	size of area to allocate (both virtual space and memory)
- *
- * This is a wrapper to create an object for use with the GuC. In order to
- * use it inside the GuC, an object needs to be pinned lifetime, so we allocate
- * both some backing storage and a range inside the Global GTT. We must pin
- * it in the GGTT somewhere other than than [0, GUC_WOPCM_TOP) because that
- * range is reserved inside GuC.
- *
- * Return:	A i915_vma if successful, otherwise an ERR_PTR.
- */
-struct i915_vma *intel_guc_allocate_vma(struct intel_guc *guc, u32 size)
-{
-	struct drm_i915_private *dev_priv = guc_to_i915(guc);
-	struct drm_i915_gem_object *obj;
-	struct i915_vma *vma;
-	int ret;
-
-	obj = i915_gem_object_create(dev_priv, size);
-	if (IS_ERR(obj))
-		return ERR_CAST(obj);
-
-	vma = i915_vma_instance(obj, &dev_priv->ggtt.base, NULL);
-	if (IS_ERR(vma))
-		goto err;
-
-	ret = i915_vma_pin(vma, 0, PAGE_SIZE,
-			   PIN_GLOBAL | PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
-	if (ret) {
-		vma = ERR_PTR(ret);
-		goto err;
-	}
-
-	return vma;
-
-err:
-	i915_gem_object_put(obj);
-	return vma;
-}
-
 /* Check that a doorbell register is in the expected state */
 static bool doorbell_ok(struct intel_guc *guc, u16 db_id)
 {
@@ -796,8 +755,8 @@ static int guc_init_doorbell_hw(struct intel_guc *guc)
  */
 static struct i915_guc_client *
 guc_client_alloc(struct drm_i915_private *dev_priv,
-		 uint32_t engines,
-		 uint32_t priority,
+		 u32 engines,
+		 u32 priority,
 		 struct i915_gem_context *ctx)
 {
 	struct i915_guc_client *client;
@@ -1069,6 +1028,7 @@ void i915_guc_submission_fini(struct drm_i915_private *dev_priv)
 
 static void guc_interrupts_capture(struct drm_i915_private *dev_priv)
 {
+	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 	int irqs;
@@ -1105,12 +1065,13 @@ static void guc_interrupts_capture(struct drm_i915_private *dev_priv)
 	 * Here we CLEAR REDIRECT_TO_GUC bit in pm_intrmsk_mbz, which will
 	 * result in the register bit being left SET!
 	 */
-	dev_priv->rps.pm_intrmsk_mbz |= ARAT_EXPIRED_INTRMSK;
-	dev_priv->rps.pm_intrmsk_mbz &= ~GEN8_PMINTR_DISABLE_REDIRECT_TO_GUC;
+	rps->pm_intrmsk_mbz |= ARAT_EXPIRED_INTRMSK;
+	rps->pm_intrmsk_mbz &= ~GEN8_PMINTR_DISABLE_REDIRECT_TO_GUC;
 }
 
 static void guc_interrupts_release(struct drm_i915_private *dev_priv)
 {
+	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 	int irqs;
@@ -1129,8 +1090,8 @@ static void guc_interrupts_release(struct drm_i915_private *dev_priv)
 	I915_WRITE(GUC_VCS2_VCS1_IER, 0);
 	I915_WRITE(GUC_WD_VECS_IER, 0);
 
-	dev_priv->rps.pm_intrmsk_mbz |= GEN8_PMINTR_DISABLE_REDIRECT_TO_GUC;
-	dev_priv->rps.pm_intrmsk_mbz &= ~ARAT_EXPIRED_INTRMSK;
+	rps->pm_intrmsk_mbz |= GEN8_PMINTR_DISABLE_REDIRECT_TO_GUC;
+	rps->pm_intrmsk_mbz &= ~ARAT_EXPIRED_INTRMSK;
 }
 
 int i915_guc_submission_enable(struct drm_i915_private *dev_priv)
@@ -1212,55 +1173,3 @@ void i915_guc_submission_disable(struct drm_i915_private *dev_priv)
 	guc_client_free(guc->execbuf_client);
 	guc->execbuf_client = NULL;
 }
-
-/**
- * intel_guc_suspend() - notify GuC entering suspend state
- * @dev_priv:	i915 device private
- */
-int intel_guc_suspend(struct drm_i915_private *dev_priv)
-{
-	struct intel_guc *guc = &dev_priv->guc;
-	struct i915_gem_context *ctx;
-	u32 data[3];
-
-	if (guc->fw.load_status != INTEL_UC_FIRMWARE_SUCCESS)
-		return 0;
-
-	gen9_disable_guc_interrupts(dev_priv);
-
-	ctx = dev_priv->kernel_context;
-
-	data[0] = INTEL_GUC_ACTION_ENTER_S_STATE;
-	/* any value greater than GUC_POWER_D0 */
-	data[1] = GUC_POWER_D1;
-	/* first page is shared data with GuC */
-	data[2] = guc_ggtt_offset(ctx->engine[RCS].state) + LRC_GUCSHR_PN * PAGE_SIZE;
-
-	return intel_guc_send(guc, data, ARRAY_SIZE(data));
-}
-
-/**
- * intel_guc_resume() - notify GuC resuming from suspend state
- * @dev_priv:	i915 device private
- */
-int intel_guc_resume(struct drm_i915_private *dev_priv)
-{
-	struct intel_guc *guc = &dev_priv->guc;
-	struct i915_gem_context *ctx;
-	u32 data[3];
-
-	if (guc->fw.load_status != INTEL_UC_FIRMWARE_SUCCESS)
-		return 0;
-
-	if (i915_modparams.guc_log_level >= 0)
-		gen9_enable_guc_interrupts(dev_priv);
-
-	ctx = dev_priv->kernel_context;
-
-	data[0] = INTEL_GUC_ACTION_EXIT_S_STATE;
-	data[1] = GUC_POWER_D0;
-	/* first page is shared data with GuC */
-	data[2] = guc_ggtt_offset(ctx->engine[RCS].state) + LRC_GUCSHR_PN * PAGE_SIZE;
-
-	return intel_guc_send(guc, data, ARRAY_SIZE(data));
-}

drivers/gpu/drm/i915/i915_guc_submission.h

+/*
+ * Copyright © 2014-2017 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_GUC_SUBMISSION_H_
+#define _I915_GUC_SUBMISSION_H_
+
+#include <linux/spinlock.h>
+
+#include "i915_gem.h"
+
+struct drm_i915_private;
+
+/*
+ * This structure primarily describes the GEM object shared with the GuC.
+ * The specs sometimes refer to this object as a "GuC context", but we use
+ * the term "client" to avoid confusion with hardware contexts. This
+ * 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 descriptor" 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.
+ */
+struct i915_guc_client {
+	struct i915_vma *vma;
+	void *vaddr;
+	struct i915_gem_context *owner;
+	struct intel_guc *guc;
+
+	/* bitmap of (host) engine ids */
+	u32 engines;
+	u32 priority;
+	u32 stage_id;
+	u32 proc_desc_offset;
+
+	u16 doorbell_id;
+	unsigned long doorbell_offset;
+
+	spinlock_t wq_lock;
+	/* Per-engine counts of GuC submissions */
+	u64 submissions[I915_NUM_ENGINES];
+};
+
+int i915_guc_submission_init(struct drm_i915_private *dev_priv);
+int i915_guc_submission_enable(struct drm_i915_private *dev_priv);
+void i915_guc_submission_disable(struct drm_i915_private *dev_priv);
+void i915_guc_submission_fini(struct drm_i915_private *dev_priv);
+
+#endif

drivers/gpu/drm/i915/i915_params.c

@@ -146,9 +146,6 @@ i915_param_named(disable_display, bool, 0400,
 i915_param_named_unsafe(enable_cmd_parser, bool, 0400,
 	"Enable command parsing (true=enabled [default], false=disabled)");
 
-i915_param_named_unsafe(use_mmio_flip, int, 0600,
-	"use MMIO flips (-1=never, 0=driver discretion [default], 1=always)");
-
 i915_param_named(mmio_debug, int, 0600,
 	"Enable the MMIO debug code for the first N failures (default: off). "
 	"This may negatively affect performance.");

drivers/gpu/drm/i915/i915_params.h

@@ -49,7 +49,6 @@
 	param(int, guc_log_level, -1) \
 	param(char *, guc_firmware_path, NULL) \
 	param(char *, huc_firmware_path, NULL) \
-	param(int, use_mmio_flip, 0) \
 	param(int, mmio_debug, 0) \
 	param(int, edp_vswing, 0) \
 	param(int, reset, 2) \

drivers/gpu/drm/i915/i915_pci.c

@@ -54,8 +54,14 @@
 	.color = { .degamma_lut_size = 512, .gamma_lut_size = 512 }
 #define CHV_COLORS \
 	.color = { .degamma_lut_size = 65, .gamma_lut_size = 257 }
+#define GLK_COLORS \
+	.color = { .degamma_lut_size = 0, .gamma_lut_size = 1024 }
 
 /* Keep in gen based order, and chronological order within a gen */
+
+#define GEN_DEFAULT_PAGE_SIZES \
+	.page_sizes = I915_GTT_PAGE_SIZE_4K
+
 #define GEN2_FEATURES \
 	.gen = 2, .num_pipes = 1, \
 	.has_overlay = 1, .overlay_needs_physical = 1, \
@@ -65,6 +71,7 @@
 	.ring_mask = RENDER_RING, \
 	.has_snoop = true, \
 	GEN_DEFAULT_PIPEOFFSETS, \
+	GEN_DEFAULT_PAGE_SIZES, \
 	CURSOR_OFFSETS
 
 static const struct intel_device_info intel_i830_info __initconst = {
@@ -98,6 +105,7 @@ static const struct intel_device_info intel_i865g_info __initconst = {
 	.ring_mask = RENDER_RING, \
 	.has_snoop = true, \
 	GEN_DEFAULT_PIPEOFFSETS, \
+	GEN_DEFAULT_PAGE_SIZES, \
 	CURSOR_OFFSETS
 
 static const struct intel_device_info intel_i915g_info __initconst = {
@@ -161,6 +169,7 @@ static const struct intel_device_info intel_pineview_info __initconst = {
 	.ring_mask = RENDER_RING, \
 	.has_snoop = true, \
 	GEN_DEFAULT_PIPEOFFSETS, \
+	GEN_DEFAULT_PAGE_SIZES, \
 	CURSOR_OFFSETS
 
 static const struct intel_device_info intel_i965g_info __initconst = {
@@ -203,6 +212,7 @@ static const struct intel_device_info intel_gm45_info __initconst = {
 	.ring_mask = RENDER_RING | BSD_RING, \
 	.has_snoop = true, \
 	GEN_DEFAULT_PIPEOFFSETS, \
+	GEN_DEFAULT_PAGE_SIZES, \
 	CURSOR_OFFSETS
 
 static const struct intel_device_info intel_ironlake_d_info __initconst = {
@@ -226,6 +236,7 @@ static const struct intel_device_info intel_ironlake_m_info __initconst = {
 	.has_rc6p = 1, \
 	.has_aliasing_ppgtt = 1, \
 	GEN_DEFAULT_PIPEOFFSETS, \
+	GEN_DEFAULT_PAGE_SIZES, \
 	CURSOR_OFFSETS
 
 #define SNB_D_PLATFORM \
@@ -269,6 +280,7 @@ static const struct intel_device_info intel_sandybridge_m_gt2_info __initconst =
 	.has_aliasing_ppgtt = 1, \
 	.has_full_ppgtt = 1, \
 	GEN_DEFAULT_PIPEOFFSETS, \
+	GEN_DEFAULT_PAGE_SIZES, \
 	IVB_CURSOR_OFFSETS
 
 #define IVB_D_PLATFORM \
@@ -325,11 +337,12 @@ static const struct intel_device_info intel_valleyview_info __initconst = {
 	.has_snoop = true,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
 	.display_mmio_offset = VLV_DISPLAY_BASE,
+	GEN_DEFAULT_PAGE_SIZES,
 	GEN_DEFAULT_PIPEOFFSETS,
 	CURSOR_OFFSETS
 };
 
-#define HSW_FEATURES  \
+#define G75_FEATURES  \
 	GEN7_FEATURES, \
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, \
 	.has_ddi = 1, \
@@ -341,7 +354,7 @@ static const struct intel_device_info intel_valleyview_info __initconst = {
 	.has_runtime_pm = 1
 
 #define HSW_PLATFORM \
-	HSW_FEATURES, \
+	G75_FEATURES, \
 	.platform = INTEL_HASWELL, \
 	.has_l3_dpf = 1
 
@@ -360,16 +373,18 @@ static const struct intel_device_info intel_haswell_gt3_info __initconst = {
 	.gt = 3,
 };
 
-#define BDW_FEATURES \
-	HSW_FEATURES, \
+#define GEN8_FEATURES \
+	G75_FEATURES, \
 	BDW_COLORS, \
+	.page_sizes = I915_GTT_PAGE_SIZE_4K | \
+		      I915_GTT_PAGE_SIZE_2M, \
 	.has_logical_ring_contexts = 1, \
 	.has_full_48bit_ppgtt = 1, \
 	.has_64bit_reloc = 1, \
 	.has_reset_engine = 1
 
 #define BDW_PLATFORM \
-	BDW_FEATURES, \
+	GEN8_FEATURES, \
 	.gen = 8, \
 	.platform = INTEL_BROADWELL
 
@@ -415,19 +430,31 @@ static const struct intel_device_info intel_cherryview_info __initconst = {
 	.has_reset_engine = 1,
 	.has_snoop = true,
 	.display_mmio_offset = VLV_DISPLAY_BASE,
+	GEN_DEFAULT_PAGE_SIZES,
 	GEN_CHV_PIPEOFFSETS,
 	CURSOR_OFFSETS,
 	CHV_COLORS,
 };
 
-#define SKL_PLATFORM \
-	BDW_FEATURES, \
-	.gen = 9, \
-	.platform = INTEL_SKYLAKE, \
+#define GEN9_DEFAULT_PAGE_SIZES \
+	.page_sizes = I915_GTT_PAGE_SIZE_4K | \
+		      I915_GTT_PAGE_SIZE_64K | \
+		      I915_GTT_PAGE_SIZE_2M
+
+#define GEN9_FEATURES \
+	GEN8_FEATURES, \
+	GEN9_DEFAULT_PAGE_SIZES, \
+	.has_logical_ring_preemption = 1, \
 	.has_csr = 1, \
 	.has_guc = 1, \
+	.has_ipc = 1, \
 	.ddb_size = 896
 
+#define SKL_PLATFORM \
+	GEN9_FEATURES, \
+	.gen = 9, \
+	.platform = INTEL_SKYLAKE
+
 static const struct intel_device_info intel_skylake_gt1_info __initconst = {
 	SKL_PLATFORM,
 	.gt = 1,
@@ -463,6 +490,7 @@ static const struct intel_device_info intel_skylake_gt4_info __initconst = {
 	.has_ddi = 1, \
 	.has_fpga_dbg = 1, \
 	.has_fbc = 1, \
+	.has_psr = 1, \
 	.has_runtime_pm = 1, \
 	.has_pooled_eu = 0, \
 	.has_csr = 1, \
@@ -470,6 +498,7 @@ static const struct intel_device_info intel_skylake_gt4_info __initconst = {
 	.has_rc6 = 1, \
 	.has_dp_mst = 1, \
 	.has_logical_ring_contexts = 1, \
+	.has_logical_ring_preemption = 1, \
 	.has_guc = 1, \
 	.has_aliasing_ppgtt = 1, \
 	.has_full_ppgtt = 1, \
@@ -477,6 +506,7 @@ static const struct intel_device_info intel_skylake_gt4_info __initconst = {
 	.has_reset_engine = 1, \
 	.has_snoop = true, \
 	.has_ipc = 1, \
+	GEN9_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_PIPEOFFSETS, \
 	IVB_CURSOR_OFFSETS, \
 	BDW_COLORS
@@ -491,17 +521,13 @@ static const struct intel_device_info intel_geminilake_info __initconst = {
 	GEN9_LP_FEATURES,
 	.platform = INTEL_GEMINILAKE,
 	.ddb_size = 1024,
-	.color = { .degamma_lut_size = 0, .gamma_lut_size = 1024 }
+	GLK_COLORS,
 };
 
 #define KBL_PLATFORM \
-	BDW_FEATURES, \
+	GEN9_FEATURES, \
 	.gen = 9, \
-	.platform = INTEL_KABYLAKE, \
-	.has_csr = 1, \
-	.has_guc = 1, \
-	.has_ipc = 1, \
-	.ddb_size = 896
+	.platform = INTEL_KABYLAKE
 
 static const struct intel_device_info intel_kabylake_gt1_info __initconst = {
 	KBL_PLATFORM,
@@ -520,13 +546,9 @@ static const struct intel_device_info intel_kabylake_gt3_info __initconst = {
 };
 
 #define CFL_PLATFORM \
-	BDW_FEATURES, \
+	GEN9_FEATURES, \
 	.gen = 9, \
-	.platform = INTEL_COFFEELAKE, \
-	.has_csr = 1, \
-	.has_guc = 1, \
-	.has_ipc = 1, \
-	.ddb_size = 896
+	.platform = INTEL_COFFEELAKE
 
 static const struct intel_device_info intel_coffeelake_gt1_info __initconst = {
 	CFL_PLATFORM,
@@ -544,16 +566,17 @@ static const struct intel_device_info intel_coffeelake_gt3_info __initconst = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
 };
 
+#define GEN10_FEATURES \
+	GEN9_FEATURES, \
+	.ddb_size = 1024, \
+	GLK_COLORS
+
 static const struct intel_device_info intel_cannonlake_gt2_info __initconst = {
-	BDW_FEATURES,
+	GEN10_FEATURES,
 	.is_alpha_support = 1,
 	.platform = INTEL_CANNONLAKE,
 	.gen = 10,
 	.gt = 2,
-	.ddb_size = 1024,
-	.has_csr = 1,
-	.has_ipc = 1,
-	.color = { .degamma_lut_size = 0, .gamma_lut_size = 1024 }
 };
 
 /*

drivers/gpu/drm/i915/i915_reg.h

@@ -2371,6 +2371,9 @@ enum i915_power_well_id {
 #define GEN9_GAMT_ECO_REG_RW_IA _MMIO(0x4ab0)
 #define   GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS	(1<<18)
 
+#define GEN8_GAMW_ECO_DEV_RW_IA _MMIO(0x4080)
+#define   GAMW_ECO_ENABLE_64K_IPS_FIELD 0xF
+
 #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)
@@ -3819,6 +3822,16 @@ enum {
 #define   PWM2_GATING_DIS		(1 << 14)
 #define   PWM1_GATING_DIS		(1 << 13)
 
+#define _CLKGATE_DIS_PSL_A		0x46520
+#define _CLKGATE_DIS_PSL_B		0x46524
+#define _CLKGATE_DIS_PSL_C		0x46528
+#define   DPF_GATING_DIS		(1 << 10)
+#define   DPF_RAM_GATING_DIS		(1 << 9)
+#define   DPFR_GATING_DIS		(1 << 8)
+
+#define CLKGATE_DIS_PSL(pipe) \
+	_MMIO_PIPE(pipe, _CLKGATE_DIS_PSL_A, _CLKGATE_DIS_PSL_B)
+
 /*
  * GEN10 clock gating regs
  */
@@ -5671,8 +5684,7 @@ enum {
 #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)
+#define  CBR_DPLLBMD_PIPE(pipe)		(1<<(7+(pipe)*11)) /* pipes B and C */
 
 /* FIFO watermark sizes etc */
 #define G4X_FIFO_LINE_SIZE	64
@@ -6993,6 +7005,12 @@ enum {
 #define GEN9_CS_DEBUG_MODE1		_MMIO(0x20ec)
 #define GEN9_CTX_PREEMPT_REG		_MMIO(0x2248)
 #define GEN8_CS_CHICKEN1		_MMIO(0x2580)
+#define GEN9_PREEMPT_3D_OBJECT_LEVEL		(1<<0)
+#define GEN9_PREEMPT_GPGPU_LEVEL(hi, lo)	(((hi) << 2) | ((lo) << 1))
+#define GEN9_PREEMPT_GPGPU_MID_THREAD_LEVEL	GEN9_PREEMPT_GPGPU_LEVEL(0, 0)
+#define GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL	GEN9_PREEMPT_GPGPU_LEVEL(0, 1)
+#define GEN9_PREEMPT_GPGPU_COMMAND_LEVEL	GEN9_PREEMPT_GPGPU_LEVEL(1, 0)
+#define GEN9_PREEMPT_GPGPU_LEVEL_MASK		GEN9_PREEMPT_GPGPU_LEVEL(1, 1)
 
 /* GEN7 chicken */
 #define GEN7_COMMON_SLICE_CHICKEN1		_MMIO(0x7010)
@@ -7164,9 +7182,6 @@ enum {
 
 #define SERR_INT			_MMIO(0xc4040)
 #define  SERR_INT_POISON		(1<<31)
-#define  SERR_INT_TRANS_C_FIFO_UNDERRUN	(1<<6)
-#define  SERR_INT_TRANS_B_FIFO_UNDERRUN	(1<<3)
-#define  SERR_INT_TRANS_A_FIFO_UNDERRUN	(1<<0)
 #define  SERR_INT_TRANS_FIFO_UNDERRUN(pipe)	(1<<((pipe)*3))
 
 /* digital port hotplug */

drivers/gpu/drm/i915/i915_suspend.c

@@ -108,8 +108,6 @@ int i915_restore_state(struct drm_i915_private *dev_priv)
 
 	mutex_lock(&dev_priv->drm.struct_mutex);
 
-	i915_gem_restore_fences(dev_priv);
-
 	if (IS_GEN4(dev_priv))
 		pci_write_config_word(pdev, GCDGMBUS,
 				      dev_priv->regfile.saveGCDGMBUS);

drivers/gpu/drm/i915/i915_sysfs.c

@@ -49,7 +49,7 @@ static u32 calc_residency(struct drm_i915_private *dev_priv,
 static ssize_t
 show_rc6_mask(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	return snprintf(buf, PAGE_SIZE, "%x\n", intel_enable_rc6());
+	return snprintf(buf, PAGE_SIZE, "%x\n", intel_rc6_enabled());
 }
 
 static ssize_t
@@ -246,7 +246,7 @@ static ssize_t gt_act_freq_mhz_show(struct device *kdev,
 
 	intel_runtime_pm_get(dev_priv);
 
-	mutex_lock(&dev_priv->rps.hw_lock);
+	mutex_lock(&dev_priv->pcu_lock);
 	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
 		u32 freq;
 		freq = vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS);
@@ -261,7 +261,7 @@ static ssize_t gt_act_freq_mhz_show(struct device *kdev,
 			ret = (rpstat & GEN6_CAGF_MASK) >> GEN6_CAGF_SHIFT;
 		ret = intel_gpu_freq(dev_priv, ret);
 	}
-	mutex_unlock(&dev_priv->rps.hw_lock);
+	mutex_unlock(&dev_priv->pcu_lock);
 
 	intel_runtime_pm_put(dev_priv);
 
@@ -275,7 +275,7 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 
 	return snprintf(buf, PAGE_SIZE, "%d\n",
 			intel_gpu_freq(dev_priv,
-				       dev_priv->rps.cur_freq));
+				       dev_priv->gt_pm.rps.cur_freq));
 }
 
 static ssize_t gt_boost_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
@@ -284,7 +284,7 @@ static ssize_t gt_boost_freq_mhz_show(struct device *kdev, struct device_attribu
 
 	return snprintf(buf, PAGE_SIZE, "%d\n",
 			intel_gpu_freq(dev_priv,
-				       dev_priv->rps.boost_freq));
+				       dev_priv->gt_pm.rps.boost_freq));
 }
 
 static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
@@ -292,6 +292,7 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
 				       const char *buf, size_t count)
 {
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
+	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	u32 val;
 	ssize_t ret;
 
@@ -301,12 +302,12 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
 
 	/* Validate against (static) hardware limits */
 	val = intel_freq_opcode(dev_priv, val);
-	if (val < dev_priv->rps.min_freq || val > dev_priv->rps.max_freq)
+	if (val < rps->min_freq || val > rps->max_freq)
 		return -EINVAL;
 
-	mutex_lock(&dev_priv->rps.hw_lock);
-	dev_priv->rps.boost_freq = val;
-	mutex_unlock(&dev_priv->rps.hw_lock);
+	mutex_lock(&dev_priv->pcu_lock);
+	rps->boost_freq = val;
+	mutex_unlock(&dev_priv->pcu_lock);
 
 	return count;
 }
@@ -318,7 +319,7 @@ static ssize_t vlv_rpe_freq_mhz_show(struct device *kdev,
 
 	return snprintf(buf, PAGE_SIZE, "%d\n",
 			intel_gpu_freq(dev_priv,
-				       dev_priv->rps.efficient_freq));
+				       dev_priv->gt_pm.rps.efficient_freq));
 }
 
 static ssize_t gt_max_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
@@ -327,7 +328,7 @@ static ssize_t gt_max_freq_mhz_show(struct device *kdev, struct device_attribute
 
 	return snprintf(buf, PAGE_SIZE, "%d\n",
 			intel_gpu_freq(dev_priv,
-				       dev_priv->rps.max_freq_softlimit));
+				       dev_priv->gt_pm.rps.max_freq_softlimit));
 }
 
 static ssize_t gt_max_freq_mhz_store(struct device *kdev,
@@ -335,6 +336,7 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 				     const char *buf, size_t count)
 {
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
+	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	u32 val;
 	ssize_t ret;
 
@@ -344,34 +346,34 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 
 	intel_runtime_pm_get(dev_priv);
 
-	mutex_lock(&dev_priv->rps.hw_lock);
+	mutex_lock(&dev_priv->pcu_lock);
 
 	val = intel_freq_opcode(dev_priv, val);
 
-	if (val < dev_priv->rps.min_freq ||
-	    val > dev_priv->rps.max_freq ||
-	    val < dev_priv->rps.min_freq_softlimit) {
-		mutex_unlock(&dev_priv->rps.hw_lock);
+	if (val < rps->min_freq ||
+	    val > rps->max_freq ||
+	    val < rps->min_freq_softlimit) {
+		mutex_unlock(&dev_priv->pcu_lock);
 		intel_runtime_pm_put(dev_priv);
 		return -EINVAL;
 	}
 
-	if (val > dev_priv->rps.rp0_freq)
+	if (val > rps->rp0_freq)
 		DRM_DEBUG("User requested overclocking to %d\n",
 			  intel_gpu_freq(dev_priv, val));
 
-	dev_priv->rps.max_freq_softlimit = val;
+	rps->max_freq_softlimit = val;
 
-	val = clamp_t(int, dev_priv->rps.cur_freq,
-		      dev_priv->rps.min_freq_softlimit,
-		      dev_priv->rps.max_freq_softlimit);
+	val = clamp_t(int, rps->cur_freq,
+		      rps->min_freq_softlimit,
+		      rps->max_freq_softlimit);
 
 	/* We still need *_set_rps to process the new max_delay and
 	 * update the interrupt limits and PMINTRMSK even though
 	 * frequency request may be unchanged. */
 	ret = intel_set_rps(dev_priv, val);
 
-	mutex_unlock(&dev_priv->rps.hw_lock);
+	mutex_unlock(&dev_priv->pcu_lock);
 
 	intel_runtime_pm_put(dev_priv);
 
@@ -384,7 +386,7 @@ static ssize_t gt_min_freq_mhz_show(struct device *kdev, struct device_attribute
 
 	return snprintf(buf, PAGE_SIZE, "%d\n",
 			intel_gpu_freq(dev_priv,
-				       dev_priv->rps.min_freq_softlimit));
+				       dev_priv->gt_pm.rps.min_freq_softlimit));
 }
 
 static ssize_t gt_min_freq_mhz_store(struct device *kdev,
@@ -392,6 +394,7 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 				     const char *buf, size_t count)
 {
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
+	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	u32 val;
 	ssize_t ret;
 
@@ -401,30 +404,30 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 
 	intel_runtime_pm_get(dev_priv);
 
-	mutex_lock(&dev_priv->rps.hw_lock);
+	mutex_lock(&dev_priv->pcu_lock);
 
 	val = intel_freq_opcode(dev_priv, val);
 
-	if (val < dev_priv->rps.min_freq ||
-	    val > dev_priv->rps.max_freq ||
-	    val > dev_priv->rps.max_freq_softlimit) {
-		mutex_unlock(&dev_priv->rps.hw_lock);
+	if (val < rps->min_freq ||
+	    val > rps->max_freq ||
+	    val > rps->max_freq_softlimit) {
+		mutex_unlock(&dev_priv->pcu_lock);
 		intel_runtime_pm_put(dev_priv);
 		return -EINVAL;
 	}
 
-	dev_priv->rps.min_freq_softlimit = val;
+	rps->min_freq_softlimit = val;
 
-	val = clamp_t(int, dev_priv->rps.cur_freq,
-		      dev_priv->rps.min_freq_softlimit,
-		      dev_priv->rps.max_freq_softlimit);
+	val = clamp_t(int, rps->cur_freq,
+		      rps->min_freq_softlimit,
+		      rps->max_freq_softlimit);
 
 	/* We still need *_set_rps to process the new min_delay and
 	 * update the interrupt limits and PMINTRMSK even though
 	 * frequency request may be unchanged. */
 	ret = intel_set_rps(dev_priv, val);
 
-	mutex_unlock(&dev_priv->rps.hw_lock);
+	mutex_unlock(&dev_priv->pcu_lock);
 
 	intel_runtime_pm_put(dev_priv);
 
@@ -448,14 +451,15 @@ static DEVICE_ATTR(gt_RPn_freq_mhz, S_IRUGO, gt_rp_mhz_show, NULL);
 static ssize_t gt_rp_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
 {
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
+	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	u32 val;
 
 	if (attr == &dev_attr_gt_RP0_freq_mhz)
-		val = intel_gpu_freq(dev_priv, dev_priv->rps.rp0_freq);
+		val = intel_gpu_freq(dev_priv, rps->rp0_freq);
 	else if (attr == &dev_attr_gt_RP1_freq_mhz)
-		val = intel_gpu_freq(dev_priv, dev_priv->rps.rp1_freq);
+		val = intel_gpu_freq(dev_priv, rps->rp1_freq);
 	else if (attr == &dev_attr_gt_RPn_freq_mhz)
-		val = intel_gpu_freq(dev_priv, dev_priv->rps.min_freq);
+		val = intel_gpu_freq(dev_priv, rps->min_freq);
 	else
 		BUG();
 

drivers/gpu/drm/i915/i915_trace.h

@@ -345,7 +345,7 @@ TRACE_EVENT(i915_gem_object_create,
 
 	    TP_STRUCT__entry(
 			     __field(struct drm_i915_gem_object *, obj)
-			     __field(u32, size)
+			     __field(u64, size)
 			     ),
 
 	    TP_fast_assign(
@@ -353,7 +353,7 @@ TRACE_EVENT(i915_gem_object_create,
 			   __entry->size = obj->base.size;
 			   ),
 
-	    TP_printk("obj=%p, size=%u", __entry->obj, __entry->size)
+	    TP_printk("obj=%p, size=0x%llx", __entry->obj, __entry->size)
 );
 
 TRACE_EVENT(i915_gem_shrink,
@@ -384,7 +384,7 @@ TRACE_EVENT(i915_vma_bind,
 			     __field(struct drm_i915_gem_object *, obj)
 			     __field(struct i915_address_space *, vm)
 			     __field(u64, offset)
-			     __field(u32, size)
+			     __field(u64, size)
 			     __field(unsigned, flags)
 			     ),
 
@@ -396,7 +396,7 @@ TRACE_EVENT(i915_vma_bind,
 			   __entry->flags = flags;
 			   ),
 
-	    TP_printk("obj=%p, offset=%016llx size=%x%s vm=%p",
+	    TP_printk("obj=%p, offset=0x%016llx size=0x%llx%s vm=%p",
 		      __entry->obj, __entry->offset, __entry->size,
 		      __entry->flags & PIN_MAPPABLE ? ", mappable" : "",
 		      __entry->vm)
@@ -410,7 +410,7 @@ TRACE_EVENT(i915_vma_unbind,
 			     __field(struct drm_i915_gem_object *, obj)
 			     __field(struct i915_address_space *, vm)
 			     __field(u64, offset)
-			     __field(u32, size)
+			     __field(u64, size)
 			     ),
 
 	    TP_fast_assign(
@@ -420,18 +420,18 @@ TRACE_EVENT(i915_vma_unbind,
 			   __entry->size = vma->node.size;
 			   ),
 
-	    TP_printk("obj=%p, offset=%016llx size=%x vm=%p",
+	    TP_printk("obj=%p, offset=0x%016llx size=0x%llx vm=%p",
 		      __entry->obj, __entry->offset, __entry->size, __entry->vm)
 );
 
 TRACE_EVENT(i915_gem_object_pwrite,
-	    TP_PROTO(struct drm_i915_gem_object *obj, u32 offset, u32 len),
+	    TP_PROTO(struct drm_i915_gem_object *obj, u64 offset, u64 len),
 	    TP_ARGS(obj, offset, len),
 
 	    TP_STRUCT__entry(
 			     __field(struct drm_i915_gem_object *, obj)
-			     __field(u32, offset)
-			     __field(u32, len)
+			     __field(u64, offset)
+			     __field(u64, len)
 			     ),
 
 	    TP_fast_assign(
@@ -440,18 +440,18 @@ TRACE_EVENT(i915_gem_object_pwrite,
 			   __entry->len = len;
 			   ),
 
-	    TP_printk("obj=%p, offset=%u, len=%u",
+	    TP_printk("obj=%p, offset=0x%llx, len=0x%llx",
 		      __entry->obj, __entry->offset, __entry->len)
 );
 
 TRACE_EVENT(i915_gem_object_pread,
-	    TP_PROTO(struct drm_i915_gem_object *obj, u32 offset, u32 len),
+	    TP_PROTO(struct drm_i915_gem_object *obj, u64 offset, u64 len),
 	    TP_ARGS(obj, offset, len),
 
 	    TP_STRUCT__entry(
 			     __field(struct drm_i915_gem_object *, obj)
-			     __field(u32, offset)
-			     __field(u32, len)
+			     __field(u64, offset)
+			     __field(u64, len)
 			     ),
 
 	    TP_fast_assign(
@@ -460,17 +460,17 @@ TRACE_EVENT(i915_gem_object_pread,
 			   __entry->len = len;
 			   ),
 
-	    TP_printk("obj=%p, offset=%u, len=%u",
+	    TP_printk("obj=%p, offset=0x%llx, len=0x%llx",
 		      __entry->obj, __entry->offset, __entry->len)
 );
 
 TRACE_EVENT(i915_gem_object_fault,
-	    TP_PROTO(struct drm_i915_gem_object *obj, u32 index, bool gtt, bool write),
+	    TP_PROTO(struct drm_i915_gem_object *obj, u64 index, bool gtt, bool write),
 	    TP_ARGS(obj, index, gtt, write),
 
 	    TP_STRUCT__entry(
 			     __field(struct drm_i915_gem_object *, obj)
-			     __field(u32, index)
+			     __field(u64, index)
 			     __field(bool, gtt)
 			     __field(bool, write)
 			     ),
@@ -482,7 +482,7 @@ TRACE_EVENT(i915_gem_object_fault,
 			   __entry->write = write;
 			   ),
 
-	    TP_printk("obj=%p, %s index=%u %s",
+	    TP_printk("obj=%p, %s index=%llu %s",
 		      __entry->obj,
 		      __entry->gtt ? "GTT" : "CPU",
 		      __entry->index,
@@ -515,14 +515,14 @@ DEFINE_EVENT(i915_gem_object, i915_gem_object_destroy,
 );
 
 TRACE_EVENT(i915_gem_evict,
-	    TP_PROTO(struct i915_address_space *vm, u32 size, u32 align, unsigned int flags),
+	    TP_PROTO(struct i915_address_space *vm, u64 size, u64 align, unsigned int flags),
 	    TP_ARGS(vm, size, align, flags),
 
 	    TP_STRUCT__entry(
 			     __field(u32, dev)
 			     __field(struct i915_address_space *, vm)
-			     __field(u32, size)
-			     __field(u32, align)
+			     __field(u64, size)
+			     __field(u64, align)
 			     __field(unsigned int, flags)
 			    ),
 
@@ -534,43 +534,11 @@ TRACE_EVENT(i915_gem_evict,
 			   __entry->flags = flags;
 			  ),
 
-	    TP_printk("dev=%d, vm=%p, size=%d, align=%d %s",
+	    TP_printk("dev=%d, vm=%p, size=0x%llx, align=0x%llx %s",
 		      __entry->dev, __entry->vm, __entry->size, __entry->align,
 		      __entry->flags & PIN_MAPPABLE ? ", mappable" : "")
 );
 
-TRACE_EVENT(i915_gem_evict_everything,
-	    TP_PROTO(struct drm_device *dev),
-	    TP_ARGS(dev),
-
-	    TP_STRUCT__entry(
-			     __field(u32, dev)
-			    ),
-
-	    TP_fast_assign(
-			   __entry->dev = dev->primary->index;
-			  ),
-
-	    TP_printk("dev=%d", __entry->dev)
-);
-
-TRACE_EVENT(i915_gem_evict_vm,
-	    TP_PROTO(struct i915_address_space *vm),
-	    TP_ARGS(vm),
-
-	    TP_STRUCT__entry(
-			     __field(u32, dev)
-			     __field(struct i915_address_space *, vm)
-			    ),
-
-	    TP_fast_assign(
-			   __entry->dev = vm->i915->drm.primary->index;
-			   __entry->vm = vm;
-			  ),
-
-	    TP_printk("dev=%d, vm=%p", __entry->dev, __entry->vm)
-);
-
 TRACE_EVENT(i915_gem_evict_node,
 	    TP_PROTO(struct i915_address_space *vm, struct drm_mm_node *node, unsigned int flags),
 	    TP_ARGS(vm, node, flags),
@@ -593,12 +561,29 @@ TRACE_EVENT(i915_gem_evict_node,
 			   __entry->flags = flags;
 			  ),
 
-	    TP_printk("dev=%d, vm=%p, start=%llx size=%llx, color=%lx, flags=%x",
+	    TP_printk("dev=%d, vm=%p, start=0x%llx size=0x%llx, color=0x%lx, flags=%x",
 		      __entry->dev, __entry->vm,
 		      __entry->start, __entry->size,
 		      __entry->color, __entry->flags)
 );
 
+TRACE_EVENT(i915_gem_evict_vm,
+	    TP_PROTO(struct i915_address_space *vm),
+	    TP_ARGS(vm),
+
+	    TP_STRUCT__entry(
+			     __field(u32, dev)
+			     __field(struct i915_address_space *, vm)
+			    ),
+
+	    TP_fast_assign(
+			   __entry->dev = vm->i915->drm.primary->index;
+			   __entry->vm = vm;
+			  ),
+
+	    TP_printk("dev=%d, vm=%p", __entry->dev, __entry->vm)
+);
+
 TRACE_EVENT(i915_gem_ring_sync_to,
 	    TP_PROTO(struct drm_i915_gem_request *to,
 		     struct drm_i915_gem_request *from),
@@ -649,29 +634,6 @@ TRACE_EVENT(i915_gem_request_queue,
 		      __entry->flags)
 );
 
-TRACE_EVENT(i915_gem_ring_flush,
-	    TP_PROTO(struct drm_i915_gem_request *req, u32 invalidate, u32 flush),
-	    TP_ARGS(req, invalidate, flush),
-
-	    TP_STRUCT__entry(
-			     __field(u32, dev)
-			     __field(u32, ring)
-			     __field(u32, invalidate)
-			     __field(u32, flush)
-			     ),
-
-	    TP_fast_assign(
-			   __entry->dev = req->i915->drm.primary->index;
-			   __entry->ring = req->engine->id;
-			   __entry->invalidate = invalidate;
-			   __entry->flush = flush;
-			   ),
-
-	    TP_printk("dev=%u, ring=%x, invalidate=%04x, flush=%04x",
-		      __entry->dev, __entry->ring,
-		      __entry->invalidate, __entry->flush)
-);
-
 DECLARE_EVENT_CLASS(i915_gem_request,
 	    TP_PROTO(struct drm_i915_gem_request *req),
 	    TP_ARGS(req),

drivers/gpu/drm/i915/i915_utils.h

@@ -99,6 +99,11 @@
 	__T;								\
 })
 
+static inline u64 ptr_to_u64(const void *ptr)
+{
+	return (uintptr_t)ptr;
+}
+
 #define u64_to_ptr(T, x) ({						\
 	typecheck(u64, x);						\
 	(T *)(uintptr_t)(x);						\
@@ -119,4 +124,17 @@ static inline void __list_del_many(struct list_head *head,
 	WRITE_ONCE(head->next, first);
 }
 
+/*
+ * Wait until the work is finally complete, even if it tries to postpone
+ * by requeueing itself. Note, that if the worker never cancels itself,
+ * we will spin forever.
+ */
+static inline void drain_delayed_work(struct delayed_work *dw)
+{
+	do {
+		while (flush_delayed_work(dw))
+			;
+	} while (delayed_work_pending(dw));
+}
+
 #endif /* !__I915_UTILS_H */

drivers/gpu/drm/i915/i915_vma.c

@@ -266,6 +266,8 @@ int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 	if (bind_flags == 0)
 		return 0;
 
+	GEM_BUG_ON(!vma->pages);
+
 	trace_i915_vma_bind(vma, bind_flags);
 	ret = vma->vm->bind_vma(vma, cache_level, bind_flags);
 	if (ret)
@@ -278,13 +280,16 @@ int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 void __iomem *i915_vma_pin_iomap(struct i915_vma *vma)
 {
 	void __iomem *ptr;
+	int err;
 
 	/* Access through the GTT requires the device to be awake. */
 	assert_rpm_wakelock_held(vma->vm->i915);
 
 	lockdep_assert_held(&vma->vm->i915->drm.struct_mutex);
-	if (WARN_ON(!i915_vma_is_map_and_fenceable(vma)))
-		return IO_ERR_PTR(-ENODEV);
+	if (WARN_ON(!i915_vma_is_map_and_fenceable(vma))) {
+		err = -ENODEV;
+		goto err;
+	}
 
 	GEM_BUG_ON(!i915_vma_is_ggtt(vma));
 	GEM_BUG_ON((vma->flags & I915_VMA_GLOBAL_BIND) == 0);
@@ -294,14 +299,36 @@ void __iomem *i915_vma_pin_iomap(struct i915_vma *vma)
 		ptr = io_mapping_map_wc(&i915_vm_to_ggtt(vma->vm)->mappable,
 					vma->node.start,
 					vma->node.size);
-		if (ptr == NULL)
-			return IO_ERR_PTR(-ENOMEM);
+		if (ptr == NULL) {
+			err = -ENOMEM;
+			goto err;
+		}
 
 		vma->iomap = ptr;
 	}
 
 	__i915_vma_pin(vma);
+
+	err = i915_vma_pin_fence(vma);
+	if (err)
+		goto err_unpin;
+
 	return ptr;
+
+err_unpin:
+	__i915_vma_unpin(vma);
+err:
+	return IO_ERR_PTR(err);
+}
+
+void i915_vma_unpin_iomap(struct i915_vma *vma)
+{
+	lockdep_assert_held(&vma->obj->base.dev->struct_mutex);
+
+	GEM_BUG_ON(vma->iomap == NULL);
+
+	i915_vma_unpin_fence(vma);
+	i915_vma_unpin(vma);
 }
 
 void i915_vma_unpin_and_release(struct i915_vma **p_vma)
@@ -471,25 +498,64 @@ i915_vma_insert(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
 	if (ret)
 		return ret;
 
+	GEM_BUG_ON(vma->pages);
+
+	ret = vma->vm->set_pages(vma);
+	if (ret)
+		goto err_unpin;
+
 	if (flags & PIN_OFFSET_FIXED) {
 		u64 offset = flags & PIN_OFFSET_MASK;
 		if (!IS_ALIGNED(offset, alignment) ||
 		    range_overflows(offset, size, end)) {
 			ret = -EINVAL;
-			goto err_unpin;
+			goto err_clear;
 		}
 
 		ret = i915_gem_gtt_reserve(vma->vm, &vma->node,
 					   size, offset, obj->cache_level,
 					   flags);
 		if (ret)
-			goto err_unpin;
+			goto err_clear;
 	} else {
+		/*
+		 * We only support huge gtt pages through the 48b PPGTT,
+		 * however we also don't want to force any alignment for
+		 * objects which need to be tightly packed into the low 32bits.
+		 *
+		 * Note that we assume that GGTT are limited to 4GiB for the
+		 * forseeable future. See also i915_ggtt_offset().
+		 */
+		if (upper_32_bits(end - 1) &&
+		    vma->page_sizes.sg > I915_GTT_PAGE_SIZE) {
+			/*
+			 * We can't mix 64K and 4K PTEs in the same page-table
+			 * (2M block), and so to avoid the ugliness and
+			 * complexity of coloring we opt for just aligning 64K
+			 * objects to 2M.
+			 */
+			u64 page_alignment =
+				rounddown_pow_of_two(vma->page_sizes.sg |
+						     I915_GTT_PAGE_SIZE_2M);
+
+			/*
+			 * Check we don't expand for the limited Global GTT
+			 * (mappable aperture is even more precious!). This
+			 * also checks that we exclude the aliasing-ppgtt.
+			 */
+			GEM_BUG_ON(i915_vma_is_ggtt(vma));
+
+			alignment = max(alignment, page_alignment);
+
+			if (vma->page_sizes.sg & I915_GTT_PAGE_SIZE_64K)
+				size = round_up(size, I915_GTT_PAGE_SIZE_2M);
+		}
+
 		ret = i915_gem_gtt_insert(vma->vm, &vma->node,
 					  size, alignment, obj->cache_level,
 					  start, end, flags);
 		if (ret)
-			goto err_unpin;
+			goto err_clear;
 
 		GEM_BUG_ON(vma->node.start < start);
 		GEM_BUG_ON(vma->node.start + vma->node.size > end);
@@ -504,6 +570,8 @@ i915_vma_insert(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
 
 	return 0;
 
+err_clear:
+	vma->vm->clear_pages(vma);
 err_unpin:
 	i915_gem_object_unpin_pages(obj);
 	return ret;
@@ -517,6 +585,8 @@ i915_vma_remove(struct i915_vma *vma)
 	GEM_BUG_ON(!drm_mm_node_allocated(&vma->node));
 	GEM_BUG_ON(vma->flags & (I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND));
 
+	vma->vm->clear_pages(vma);
+
 	drm_mm_remove_node(&vma->node);
 	list_move_tail(&vma->vm_link, &vma->vm->unbound_list);
 
@@ -569,8 +639,8 @@ int __i915_vma_do_pin(struct i915_vma *vma,
 
 err_remove:
 	if ((bound & I915_VMA_BIND_MASK) == 0) {
-		GEM_BUG_ON(vma->pages);
 		i915_vma_remove(vma);
+		GEM_BUG_ON(vma->pages);
 	}
 err_unpin:
 	__i915_vma_unpin(vma);
@@ -620,6 +690,30 @@ static void __i915_vma_iounmap(struct i915_vma *vma)
 	vma->iomap = NULL;
 }
 
+void i915_vma_revoke_mmap(struct i915_vma *vma)
+{
+	struct drm_vma_offset_node *node = &vma->obj->base.vma_node;
+	u64 vma_offset;
+
+	lockdep_assert_held(&vma->vm->i915->drm.struct_mutex);
+
+	if (!i915_vma_has_userfault(vma))
+		return;
+
+	GEM_BUG_ON(!i915_vma_is_map_and_fenceable(vma));
+	GEM_BUG_ON(!vma->obj->userfault_count);
+
+	vma_offset = vma->ggtt_view.partial.offset << PAGE_SHIFT;
+	unmap_mapping_range(vma->vm->i915->drm.anon_inode->i_mapping,
+			    drm_vma_node_offset_addr(node) + vma_offset,
+			    vma->size,
+			    1);
+
+	i915_vma_unset_userfault(vma);
+	if (!--vma->obj->userfault_count)
+		list_del(&vma->obj->userfault_link);
+}
+
 int i915_vma_unbind(struct i915_vma *vma)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
@@ -683,11 +777,13 @@ int i915_vma_unbind(struct i915_vma *vma)
 			return ret;
 
 		/* Force a pagefault for domain tracking on next user access */
-		i915_gem_release_mmap(obj);
+		i915_vma_revoke_mmap(vma);
 
 		__i915_vma_iounmap(vma);
 		vma->flags &= ~I915_VMA_CAN_FENCE;
 	}
+	GEM_BUG_ON(vma->fence);
+	GEM_BUG_ON(i915_vma_has_userfault(vma));
 
 	if (likely(!vma->vm->closed)) {
 		trace_i915_vma_unbind(vma);
@@ -695,13 +791,6 @@ int i915_vma_unbind(struct i915_vma *vma)
 	}
 	vma->flags &= ~(I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND);
 
-	if (vma->pages != obj->mm.pages) {
-		GEM_BUG_ON(!vma->pages);
-		sg_free_table(vma->pages);
-		kfree(vma->pages);
-	}
-	vma->pages = NULL;
-
 	i915_vma_remove(vma);
 
 destroy:

drivers/gpu/drm/i915/i915_vma.h

@@ -55,6 +55,7 @@ struct i915_vma {
 	void __iomem *iomap;
 	u64 size;
 	u64 display_alignment;
+	struct i915_page_sizes page_sizes;
 
 	u32 fence_size;
 	u32 fence_alignment;
@@ -65,7 +66,7 @@ struct i915_vma {
 	 * that exist in the ctx->handle_vmas LUT for this vma.
 	 */
 	unsigned int open_count;
-	unsigned int flags;
+	unsigned long flags;
 	/**
 	 * How many users have pinned this object in GTT space. The following
 	 * users can each hold at most one reference: pwrite/pread, execbuffer
@@ -87,6 +88,8 @@ struct i915_vma {
 #define I915_VMA_GGTT		BIT(8)
 #define I915_VMA_CAN_FENCE	BIT(9)
 #define I915_VMA_CLOSED		BIT(10)
+#define I915_VMA_USERFAULT_BIT	11
+#define I915_VMA_USERFAULT	BIT(I915_VMA_USERFAULT_BIT)
 
 	unsigned int active;
 	struct i915_gem_active last_read[I915_NUM_ENGINES];
@@ -145,6 +148,22 @@ static inline bool i915_vma_is_closed(const struct i915_vma *vma)
 	return vma->flags & I915_VMA_CLOSED;
 }
 
+static inline bool i915_vma_set_userfault(struct i915_vma *vma)
+{
+	GEM_BUG_ON(!i915_vma_is_map_and_fenceable(vma));
+	return __test_and_set_bit(I915_VMA_USERFAULT_BIT, &vma->flags);
+}
+
+static inline void i915_vma_unset_userfault(struct i915_vma *vma)
+{
+	return __clear_bit(I915_VMA_USERFAULT_BIT, &vma->flags);
+}
+
+static inline bool i915_vma_has_userfault(const struct i915_vma *vma)
+{
+	return test_bit(I915_VMA_USERFAULT_BIT, &vma->flags);
+}
+
 static inline unsigned int i915_vma_get_active(const struct i915_vma *vma)
 {
 	return vma->active;
@@ -243,6 +262,7 @@ bool i915_gem_valid_gtt_space(struct i915_vma *vma, unsigned long cache_level);
 bool i915_vma_misplaced(const struct i915_vma *vma,
 			u64 size, u64 alignment, u64 flags);
 void __i915_vma_set_map_and_fenceable(struct i915_vma *vma);
+void i915_vma_revoke_mmap(struct i915_vma *vma);
 int __must_check i915_vma_unbind(struct i915_vma *vma);
 void i915_vma_unlink_ctx(struct i915_vma *vma);
 void i915_vma_close(struct i915_vma *vma);
@@ -321,12 +341,7 @@ void __iomem *i915_vma_pin_iomap(struct i915_vma *vma);
  * Callers must hold the struct_mutex. This function is only valid to be
  * called on a VMA previously iomapped by the caller with i915_vma_pin_iomap().
  */
-static inline void i915_vma_unpin_iomap(struct i915_vma *vma)
-{
-	lockdep_assert_held(&vma->obj->base.dev->struct_mutex);
-	GEM_BUG_ON(vma->iomap == NULL);
-	i915_vma_unpin(vma);
-}
+void i915_vma_unpin_iomap(struct i915_vma *vma);
 
 static inline struct page *i915_vma_first_page(struct i915_vma *vma)
 {
@@ -349,15 +364,13 @@ static inline struct page *i915_vma_first_page(struct i915_vma *vma)
  *
  * True if the vma has a fence, false otherwise.
  */
-static inline bool
-i915_vma_pin_fence(struct i915_vma *vma)
+int i915_vma_pin_fence(struct i915_vma *vma);
+int __must_check i915_vma_put_fence(struct i915_vma *vma);
+
+static inline void __i915_vma_unpin_fence(struct i915_vma *vma)
 {
-	lockdep_assert_held(&vma->obj->base.dev->struct_mutex);
-	if (vma->fence) {
-		vma->fence->pin_count++;
-		return true;
-	} else
-		return false;
+	GEM_BUG_ON(vma->fence->pin_count <= 0);
+	vma->fence->pin_count--;
 }
 
 /**
@@ -372,10 +385,8 @@ static inline void
 i915_vma_unpin_fence(struct i915_vma *vma)
 {
 	lockdep_assert_held(&vma->obj->base.dev->struct_mutex);
-	if (vma->fence) {
-		GEM_BUG_ON(vma->fence->pin_count <= 0);
-		vma->fence->pin_count--;
-	}
+	if (vma->fence)
+		__i915_vma_unpin_fence(vma);
 }
 
 #endif

drivers/gpu/drm/i915/intel_audio.c

@@ -754,7 +754,7 @@ static struct intel_encoder *get_saved_enc(struct drm_i915_private *dev_priv,
 {
 	struct intel_encoder *encoder;
 
-	if (WARN_ON(pipe >= I915_MAX_PIPES))
+	if (WARN_ON(pipe >= INTEL_INFO(dev_priv)->num_pipes))
 		return NULL;
 
 	/* MST */

drivers/gpu/drm/i915/intel_bios.c

@@ -431,27 +431,6 @@ parse_general_features(struct drm_i915_private *dev_priv,
 		      dev_priv->vbt.fdi_rx_polarity_inverted);
 }
 
-static void
-parse_general_definitions(struct drm_i915_private *dev_priv,
-			  const struct bdb_header *bdb)
-{
-	const struct bdb_general_definitions *general;
-
-	general = find_section(bdb, BDB_GENERAL_DEFINITIONS);
-	if (general) {
-		u16 block_size = get_blocksize(general);
-		if (block_size >= sizeof(*general)) {
-			int bus_pin = general->crt_ddc_gmbus_pin;
-			DRM_DEBUG_KMS("crt_ddc_bus_pin: %d\n", bus_pin);
-			if (intel_gmbus_is_valid_pin(dev_priv, bus_pin))
-				dev_priv->vbt.crt_ddc_pin = bus_pin;
-		} else {
-			DRM_DEBUG_KMS("BDB_GD too small (%d). Invalid.\n",
-				      block_size);
-		}
-	}
-}
-
 static const struct child_device_config *
 child_device_ptr(const struct bdb_general_definitions *defs, int i)
 {
@@ -459,41 +438,24 @@ child_device_ptr(const struct bdb_general_definitions *defs, int i)
 }
 
 static void
-parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
-			  const struct bdb_header *bdb)
+parse_sdvo_device_mapping(struct drm_i915_private *dev_priv, u8 bdb_version)
 {
 	struct sdvo_device_mapping *mapping;
-	const struct bdb_general_definitions *defs;
 	const struct child_device_config *child;
-	int i, child_device_num, count;
-	u16	block_size;
-
-	defs = find_section(bdb, BDB_GENERAL_DEFINITIONS);
-	if (!defs) {
-		DRM_DEBUG_KMS("No general definition block is found, unable to construct sdvo mapping.\n");
-		return;
-	}
+	int i, count = 0;
 
 	/*
-	 * Only parse SDVO mappings when the general definitions block child
-	 * device size matches that of the *legacy* child device config
-	 * struct. Thus, SDVO mapping will be skipped for newer VBT.
+	 * Only parse SDVO mappings on gens that could have SDVO. This isn't
+	 * accurate and doesn't have to be, as long as it's not too strict.
 	 */
-	if (defs->child_dev_size != LEGACY_CHILD_DEVICE_CONFIG_SIZE) {
-		DRM_DEBUG_KMS("Unsupported child device size for SDVO mapping.\n");
+	if (!IS_GEN(dev_priv, 3, 7)) {
+		DRM_DEBUG_KMS("Skipping SDVO device mapping\n");
 		return;
 	}
-	/* get the block size of general definitions */
-	block_size = get_blocksize(defs);
-	/* get the number of child device */
-	child_device_num = (block_size - sizeof(*defs)) / defs->child_dev_size;
-	count = 0;
-	for (i = 0; i < child_device_num; i++) {
-		child = child_device_ptr(defs, i);
-		if (!child->device_type) {
-			/* skip the device block if device type is invalid */
-			continue;
-		}
+
+	for (i = 0, count = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		child = dev_priv->vbt.child_dev + i;
+
 		if (child->slave_addr != SLAVE_ADDR1 &&
 		    child->slave_addr != SLAVE_ADDR2) {
 			/*
@@ -544,7 +506,6 @@ parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
 		/* No SDVO device info is found */
 		DRM_DEBUG_KMS("No SDVO device info is found in VBT\n");
 	}
-	return;
 }
 
 static void
@@ -1111,7 +1072,7 @@ static void sanitize_aux_ch(struct drm_i915_private *dev_priv,
 }
 
 static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
-			   const struct bdb_header *bdb)
+			   u8 bdb_version)
 {
 	struct child_device_config *it, *child = NULL;
 	struct ddi_vbt_port_info *info = &dev_priv->vbt.ddi_port_info[port];
@@ -1215,7 +1176,7 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
 		sanitize_aux_ch(dev_priv, port);
 	}
 
-	if (bdb->version >= 158) {
+	if (bdb_version >= 158) {
 		/* The VBT HDMI level shift values match the table we have. */
 		hdmi_level_shift = child->hdmi_level_shifter_value;
 		DRM_DEBUG_KMS("VBT HDMI level shift for port %c: %d\n",
@@ -1225,7 +1186,7 @@ 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->iboost) {
+	if (bdb_version >= 196 && child->iboost) {
 		info->dp_boost_level = translate_iboost(child->dp_iboost_level);
 		DRM_DEBUG_KMS("VBT (e)DP boost level for port %c: %d\n",
 			      port_name(port), info->dp_boost_level);
@@ -1235,40 +1196,52 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
 	}
 }
 
-static void parse_ddi_ports(struct drm_i915_private *dev_priv,
-			    const struct bdb_header *bdb)
+static void parse_ddi_ports(struct drm_i915_private *dev_priv, u8 bdb_version)
 {
 	enum port port;
 
-	if (!HAS_DDI(dev_priv))
+	if (!HAS_DDI(dev_priv) && !IS_CHERRYVIEW(dev_priv))
 		return;
 
 	if (!dev_priv->vbt.child_dev_num)
 		return;
 
-	if (bdb->version < 155)
+	if (bdb_version < 155)
 		return;
 
 	for (port = PORT_A; port < I915_MAX_PORTS; port++)
-		parse_ddi_port(dev_priv, port, bdb);
+		parse_ddi_port(dev_priv, port, bdb_version);
 }
 
 static void
-parse_device_mapping(struct drm_i915_private *dev_priv,
-		     const struct bdb_header *bdb)
+parse_general_definitions(struct drm_i915_private *dev_priv,
+			  const struct bdb_header *bdb)
 {
 	const struct bdb_general_definitions *defs;
 	const struct child_device_config *child;
-	struct child_device_config *child_dev_ptr;
 	int i, child_device_num, count;
 	u8 expected_size;
 	u16 block_size;
+	int bus_pin;
 
 	defs = find_section(bdb, BDB_GENERAL_DEFINITIONS);
 	if (!defs) {
 		DRM_DEBUG_KMS("No general definition block is found, no devices defined.\n");
 		return;
 	}
+
+	block_size = get_blocksize(defs);
+	if (block_size < sizeof(*defs)) {
+		DRM_DEBUG_KMS("General definitions block too small (%u)\n",
+			      block_size);
+		return;
+	}
+
+	bus_pin = defs->crt_ddc_gmbus_pin;
+	DRM_DEBUG_KMS("crt_ddc_bus_pin: %d\n", bus_pin);
+	if (intel_gmbus_is_valid_pin(dev_priv, bus_pin))
+		dev_priv->vbt.crt_ddc_pin = bus_pin;
+
 	if (bdb->version < 106) {
 		expected_size = 22;
 	} else if (bdb->version < 111) {
@@ -1298,18 +1271,14 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 		return;
 	}
 
-	/* get the block size of general definitions */
-	block_size = get_blocksize(defs);
 	/* get the number of child device */
 	child_device_num = (block_size - sizeof(*defs)) / defs->child_dev_size;
 	count = 0;
 	/* get the number of child device that is present */
 	for (i = 0; i < child_device_num; i++) {
 		child = child_device_ptr(defs, i);
-		if (!child->device_type) {
-			/* skip the device block if device type is invalid */
+		if (!child->device_type)
 			continue;
-		}
 		count++;
 	}
 	if (!count) {
@@ -1326,36 +1295,18 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 	count = 0;
 	for (i = 0; i < child_device_num; i++) {
 		child = child_device_ptr(defs, i);
-		if (!child->device_type) {
-			/* skip the device block if device type is invalid */
+		if (!child->device_type)
 			continue;
-		}
-
-		child_dev_ptr = dev_priv->vbt.child_dev + count;
-		count++;
 
 		/*
 		 * Copy as much as we know (sizeof) and is available
 		 * (child_dev_size) of the child device. Accessing the data must
 		 * depend on VBT version.
 		 */
-		memcpy(child_dev_ptr, child,
+		memcpy(dev_priv->vbt.child_dev + count, child,
 		       min_t(size_t, defs->child_dev_size, sizeof(*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->iboost = 0;
-			child_dev_ptr->hpd_invert = 0;
-		}
-
-		if (bdb->version < 192)
-			child_dev_ptr->lspcon = 0;
+		count++;
 	}
-	return;
 }
 
 /* Common defaults which may be overridden by VBT. */
@@ -1536,14 +1487,15 @@ void intel_bios_init(struct drm_i915_private *dev_priv)
 	parse_lfp_panel_data(dev_priv, bdb);
 	parse_lfp_backlight(dev_priv, bdb);
 	parse_sdvo_panel_data(dev_priv, bdb);
-	parse_sdvo_device_mapping(dev_priv, bdb);
-	parse_device_mapping(dev_priv, bdb);
 	parse_driver_features(dev_priv, bdb);
 	parse_edp(dev_priv, bdb);
 	parse_psr(dev_priv, bdb);
 	parse_mipi_config(dev_priv, bdb);
 	parse_mipi_sequence(dev_priv, bdb);
-	parse_ddi_ports(dev_priv, bdb);
+
+	/* Further processing on pre-parsed data */
+	parse_sdvo_device_mapping(dev_priv, bdb->version);
+	parse_ddi_ports(dev_priv, bdb->version);
 
 out:
 	if (!vbt) {

drivers/gpu/drm/i915/intel_csr.c

@@ -216,7 +216,7 @@ static void gen9_set_dc_state_debugmask(struct drm_i915_private *dev_priv)
 
 	mask = DC_STATE_DEBUG_MASK_MEMORY_UP;
 
-	if (IS_BROXTON(dev_priv))
+	if (IS_GEN9_LP(dev_priv))
 		mask |= DC_STATE_DEBUG_MASK_CORES;
 
 	/* The below bit doesn't need to be cleared ever afterwards */

drivers/gpu/drm/i915/intel_dp_mst.c

@@ -162,14 +162,19 @@ static void intel_mst_post_disable_dp(struct intel_encoder *encoder,
 
 	drm_dp_mst_deallocate_vcpi(&intel_dp->mst_mgr, connector->port);
 
+	/*
+	 * Power down mst path before disabling the port, otherwise we end
+	 * up getting interrupts from the sink upon detecting link loss.
+	 */
+	drm_dp_send_power_updown_phy(&intel_dp->mst_mgr, connector->port,
+				     false);
+
 	intel_dp->active_mst_links--;
 
 	intel_mst->connector = NULL;
 	if (intel_dp->active_mst_links == 0) {
 		intel_dig_port->base.post_disable(&intel_dig_port->base,
 						  NULL, NULL);
-
-		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_OFF);
 	}
 	DRM_DEBUG_KMS("active links %d\n", intel_dp->active_mst_links);
 }
@@ -196,6 +201,7 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder,
 
 	DRM_DEBUG_KMS("active links %d\n", intel_dp->active_mst_links);
 
+	drm_dp_send_power_updown_phy(&intel_dp->mst_mgr, connector->port, true);
 	if (intel_dp->active_mst_links == 0)
 		intel_dig_port->base.pre_enable(&intel_dig_port->base,
 						pipe_config, NULL);