Merge tag 'drm-intel-next-2016-08-08' of git://anongit.freedesktop.org/drm-intel into drm-next

- refactor ddi buffer programming a bit (Ville)
- large-scale renaming to untangle naming in the gem code (Chris)
- rework vma/active tracking for accurately reaping idle mappings of shared
  objects (Chris)
- misc dp sst/mst probing corner case fixes (Ville)
- tons of cleanup&tunings all around in gem
- lockless (rcu-protected) request lookup, plus use it everywhere for
  non(b)locking waits (Chris)
- pipe crc debugfs fixes (Rodrigo)
- random fixes all over

* tag 'drm-intel-next-2016-08-08' of git://anongit.freedesktop.org/drm-intel: (222 commits)
  drm/i915: Update DRIVER_DATE to 20160808
  drm/i915: fix aliasing_ppgtt leak
  drm/i915: Update comment before i915_spin_request
  drm/i915: Use drm official vblank_no_hw_counter callback.
  drm/i915: Fix copy_to_user usage for pipe_crc
  Revert "drm/i915: Track active streams also for DP SST"
  drm/i915: fix WaInsertDummyPushConstPs
  drm/i915: Assert that the request hasn't been retired
  drm/i915: Repack fence tiling mode and stride into a single integer
  drm/i915: Document and reject invalid tiling modes
  drm/i915: Remove locking for get_tiling
  drm/i915: Remove pinned check from madvise ioctl
  drm/i915: Reduce locking inside swfinish ioctl
  drm/i915: Remove (struct_mutex) locking for busy-ioctl
  drm/i915: Remove (struct_mutex) locking for wait-ioctl
  drm/i915: Do a nonblocking wait first in pread/pwrite
  drm/i915: Remove unused no-shrinker-steal
  drm/i915: Tidy generation of the GTT mmap offset
  drm/i915/shrinker: Wait before acquiring struct_mutex under oom
  drm/i915: Simplify do_idling() (Ironlake vt-d w/a)
  ...

Documentation/gpu/i915.rst

@@ -70,6 +70,9 @@ Frontbuffer Tracking
 .. kernel-doc:: drivers/gpu/drm/i915/intel_frontbuffer.c
    :doc: frontbuffer tracking
 
+.. kernel-doc:: drivers/gpu/drm/i915/intel_frontbuffer.h
+   :internal:
+
 .. kernel-doc:: drivers/gpu/drm/i915/intel_frontbuffer.c
    :internal:
 

drivers/gpu/drm/i915/Makefile

@@ -25,7 +25,6 @@ i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o
 i915-y += i915_cmd_parser.o \
 	  i915_gem_batch_pool.o \
 	  i915_gem_context.o \
-	  i915_gem_debug.o \
 	  i915_gem_dmabuf.o \
 	  i915_gem_evict.o \
 	  i915_gem_execbuffer.o \
@@ -33,6 +32,7 @@ i915-y += i915_cmd_parser.o \
 	  i915_gem_gtt.o \
 	  i915_gem.o \
 	  i915_gem_render_state.o \
+	  i915_gem_request.o \
 	  i915_gem_shrinker.o \
 	  i915_gem_stolen.o \
 	  i915_gem_tiling.o \
@@ -40,6 +40,7 @@ i915-y += i915_cmd_parser.o \
 	  i915_gpu_error.o \
 	  i915_trace_points.o \
 	  intel_breadcrumbs.o \
+	  intel_engine_cs.o \
 	  intel_lrc.o \
 	  intel_mocs.o \
 	  intel_ringbuffer.o \

drivers/gpu/drm/i915/i915_cmd_parser.c

@@ -62,23 +62,23 @@
  * The parser always rejects such commands.
  *
  * The majority of the problematic commands fall in the MI_* range, with only a
- * few specific commands on each ring (e.g. PIPE_CONTROL and MI_FLUSH_DW).
+ * few specific commands on each engine (e.g. PIPE_CONTROL and MI_FLUSH_DW).
  *
  * Implementation:
- * Each ring maintains tables of commands and registers which the parser uses in
- * scanning batch buffers submitted to that ring.
+ * Each engine maintains tables of commands and registers which the parser
+ * uses in scanning batch buffers submitted to that engine.
  *
  * Since the set of commands that the parser must check for is significantly
  * smaller than the number of commands supported, the parser tables contain only
  * those commands required by the parser. This generally works because command
  * opcode ranges have standard command length encodings. So for commands that
  * the parser does not need to check, it can easily skip them. This is
- * implemented via a per-ring length decoding vfunc.
+ * implemented via a per-engine length decoding vfunc.
  *
  * Unfortunately, there are a number of commands that do not follow the standard
  * length encoding for their opcode range, primarily amongst the MI_* commands.
  * To handle this, the parser provides a way to define explicit "skip" entries
- * in the per-ring command tables.
+ * in the per-engine command tables.
  *
  * Other command table entries map fairly directly to high level categories
  * mentioned above: rejected, master-only, register whitelist. The parser
@@ -603,7 +603,7 @@ static u32 gen7_blt_get_cmd_length_mask(u32 cmd_header)
 	return 0;
 }
 
-static bool validate_cmds_sorted(struct intel_engine_cs *engine,
+static bool validate_cmds_sorted(const struct intel_engine_cs *engine,
 				 const struct drm_i915_cmd_table *cmd_tables,
 				 int cmd_table_count)
 {
@@ -624,8 +624,10 @@ static bool validate_cmds_sorted(struct intel_engine_cs *engine,
 			u32 curr = desc->cmd.value & desc->cmd.mask;
 
 			if (curr < previous) {
-				DRM_ERROR("CMD: table not sorted ring=%d table=%d entry=%d cmd=0x%08X prev=0x%08X\n",
-					  engine->id, i, j, curr, previous);
+				DRM_ERROR("CMD: %s [%d] command table not sorted: "
+					  "table=%d entry=%d cmd=0x%08X prev=0x%08X\n",
+					  engine->name, engine->id,
+					  i, j, curr, previous);
 				ret = false;
 			}
 
@@ -636,7 +638,7 @@ static bool validate_cmds_sorted(struct intel_engine_cs *engine,
 	return ret;
 }
 
-static bool check_sorted(int ring_id,
+static bool check_sorted(const struct intel_engine_cs *engine,
 			 const struct drm_i915_reg_descriptor *reg_table,
 			 int reg_count)
 {
@@ -648,8 +650,10 @@ static bool check_sorted(int ring_id,
 		u32 curr = i915_mmio_reg_offset(reg_table[i].addr);
 
 		if (curr < previous) {
-			DRM_ERROR("CMD: table not sorted ring=%d entry=%d reg=0x%08X prev=0x%08X\n",
-				  ring_id, i, curr, previous);
+			DRM_ERROR("CMD: %s [%d] register table not sorted: "
+				  "entry=%d reg=0x%08X prev=0x%08X\n",
+				  engine->name, engine->id,
+				  i, curr, previous);
 			ret = false;
 		}
 
@@ -666,7 +670,7 @@ static bool validate_regs_sorted(struct intel_engine_cs *engine)
 
 	for (i = 0; i < engine->reg_table_count; i++) {
 		table = &engine->reg_tables[i];
-		if (!check_sorted(engine->id, table->regs, table->num_regs))
+		if (!check_sorted(engine, table->regs, table->num_regs))
 			return false;
 	}
 
@@ -736,7 +740,7 @@ static void fini_hash_table(struct intel_engine_cs *engine)
 }
 
 /**
- * i915_cmd_parser_init_ring() - set cmd parser related fields for a ringbuffer
+ * intel_engine_init_cmd_parser() - set cmd parser related fields for an engine
  * @engine: the engine to initialize
  *
  * Optionally initializes fields related to batch buffer command parsing in the
@@ -745,7 +749,7 @@ static void fini_hash_table(struct intel_engine_cs *engine)
  *
  * Return: non-zero if initialization fails
  */
-int i915_cmd_parser_init_ring(struct intel_engine_cs *engine)
+int intel_engine_init_cmd_parser(struct intel_engine_cs *engine)
 {
 	const struct drm_i915_cmd_table *cmd_tables;
 	int cmd_table_count;
@@ -806,8 +810,7 @@ int i915_cmd_parser_init_ring(struct intel_engine_cs *engine)
 		engine->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
 	default:
-		DRM_ERROR("CMD: cmd_parser_init with unknown ring: %d\n",
-			  engine->id);
+		MISSING_CASE(engine->id);
 		BUG();
 	}
 
@@ -829,13 +832,13 @@ int i915_cmd_parser_init_ring(struct intel_engine_cs *engine)
 }
 
 /**
- * i915_cmd_parser_fini_ring() - clean up cmd parser related fields
+ * intel_engine_cleanup_cmd_parser() - clean up cmd parser related fields
  * @engine: the engine to clean up
  *
  * Releases any resources related to command parsing that may have been
- * initialized for the specified ring.
+ * initialized for the specified engine.
  */
-void i915_cmd_parser_fini_ring(struct intel_engine_cs *engine)
+void intel_engine_cleanup_cmd_parser(struct intel_engine_cs *engine)
 {
 	if (!engine->needs_cmd_parser)
 		return;
@@ -866,9 +869,9 @@ find_cmd_in_table(struct intel_engine_cs *engine,
  * Returns a pointer to a descriptor for the command specified by cmd_header.
  *
  * The caller must supply space for a default descriptor via the default_desc
- * parameter. If no descriptor for the specified command exists in the ring's
+ * parameter. If no descriptor for the specified command exists in the engine's
  * command parser tables, this function fills in default_desc based on the
- * ring's default length encoding and returns default_desc.
+ * engine's default length encoding and returns default_desc.
  */
 static const struct drm_i915_cmd_descriptor*
 find_cmd(struct intel_engine_cs *engine,
@@ -1023,15 +1026,16 @@ unpin_src:
 }
 
 /**
- * i915_needs_cmd_parser() - should a given ring use software command parsing?
+ * intel_engine_needs_cmd_parser() - should a given engine use software
+ *                                   command parsing?
  * @engine: the engine in question
  *
  * Only certain platforms require software batch buffer command parsing, and
  * only when enabled via module parameter.
  *
- * Return: true if the ring requires software command parsing
+ * Return: true if the engine requires software command parsing
  */
-bool i915_needs_cmd_parser(struct intel_engine_cs *engine)
+bool intel_engine_needs_cmd_parser(struct intel_engine_cs *engine)
 {
 	if (!engine->needs_cmd_parser)
 		return false;
@@ -1078,8 +1082,8 @@ static bool check_cmd(const struct intel_engine_cs *engine,
 						   reg_addr);
 
 			if (!reg) {
-				DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
-						 reg_addr, *cmd, engine->id);
+				DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (exec_id=%d)\n",
+						 reg_addr, *cmd, engine->exec_id);
 				return false;
 			}
 
@@ -1159,11 +1163,11 @@ static bool check_cmd(const struct intel_engine_cs *engine,
 				desc->bits[i].mask;
 
 			if (dword != desc->bits[i].expected) {
-				DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X for bitmask 0x%08X (exp=0x%08X act=0x%08X) (ring=%d)\n",
+				DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X for bitmask 0x%08X (exp=0x%08X act=0x%08X) (exec_id=%d)\n",
 						 *cmd,
 						 desc->bits[i].mask,
 						 desc->bits[i].expected,
-						 dword, engine->id);
+						 dword, engine->exec_id);
 				return false;
 			}
 		}
@@ -1189,12 +1193,12 @@ static bool check_cmd(const struct intel_engine_cs *engine,
  * Return: non-zero if the parser finds violations or otherwise fails; -EACCES
  * if the batch appears legal but should use hardware parsing
  */
-int i915_parse_cmds(struct intel_engine_cs *engine,
-		    struct drm_i915_gem_object *batch_obj,
-		    struct drm_i915_gem_object *shadow_batch_obj,
-		    u32 batch_start_offset,
-		    u32 batch_len,
-		    bool is_master)
+int intel_engine_cmd_parser(struct intel_engine_cs *engine,
+			    struct drm_i915_gem_object *batch_obj,
+			    struct drm_i915_gem_object *shadow_batch_obj,
+			    u32 batch_start_offset,
+			    u32 batch_len,
+			    bool is_master)
 {
 	u32 *cmd, *batch_base, *batch_end;
 	struct drm_i915_cmd_descriptor default_desc = { 0 };
@@ -1295,7 +1299,7 @@ int i915_cmd_parser_get_version(struct drm_i915_private *dev_priv)
 
 	/* If the command parser is not enabled, report 0 - unsupported */
 	for_each_engine(engine, dev_priv) {
-		if (i915_needs_cmd_parser(engine)) {
+		if (intel_engine_needs_cmd_parser(engine)) {
 			active = true;
 			break;
 		}

drivers/gpu/drm/i915/i915_debugfs.c

@@ -91,7 +91,7 @@ static int i915_capabilities(struct seq_file *m, void *data)
 
 static char get_active_flag(struct drm_i915_gem_object *obj)
 {
-	return obj->active ? '*' : ' ';
+	return i915_gem_object_is_active(obj) ? '*' : ' ';
 }
 
 static char get_pin_flag(struct drm_i915_gem_object *obj)
@@ -101,7 +101,7 @@ static char get_pin_flag(struct drm_i915_gem_object *obj)
 
 static char get_tiling_flag(struct drm_i915_gem_object *obj)
 {
-	switch (obj->tiling_mode) {
+	switch (i915_gem_object_get_tiling(obj)) {
 	default:
 	case I915_TILING_NONE: return ' ';
 	case I915_TILING_X: return 'X';
@@ -125,7 +125,7 @@ static u64 i915_gem_obj_total_ggtt_size(struct drm_i915_gem_object *obj)
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, &obj->vma_list, obj_link) {
-		if (vma->is_ggtt && drm_mm_node_allocated(&vma->node))
+		if (i915_vma_is_ggtt(vma) && drm_mm_node_allocated(&vma->node))
 			size += vma->node.size;
 	}
 
@@ -138,6 +138,7 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 	struct intel_engine_cs *engine;
 	struct i915_vma *vma;
+	unsigned int frontbuffer_bits;
 	int pin_count = 0;
 	enum intel_engine_id id;
 
@@ -155,17 +156,20 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		   obj->base.write_domain);
 	for_each_engine_id(engine, dev_priv, id)
 		seq_printf(m, "%x ",
-				i915_gem_request_get_seqno(obj->last_read_req[id]));
+			   i915_gem_active_get_seqno(&obj->last_read[id],
+						     &obj->base.dev->struct_mutex));
 	seq_printf(m, "] %x %x%s%s%s",
-		   i915_gem_request_get_seqno(obj->last_write_req),
-		   i915_gem_request_get_seqno(obj->last_fenced_req),
+		   i915_gem_active_get_seqno(&obj->last_write,
+					     &obj->base.dev->struct_mutex),
+		   i915_gem_active_get_seqno(&obj->last_fence,
+					     &obj->base.dev->struct_mutex),
 		   i915_cache_level_str(to_i915(obj->base.dev), obj->cache_level),
 		   obj->dirty ? " dirty" : "",
 		   obj->madv == I915_MADV_DONTNEED ? " purgeable" : "");
 	if (obj->base.name)
 		seq_printf(m, " (name: %d)", obj->base.name);
 	list_for_each_entry(vma, &obj->vma_list, obj_link) {
-		if (vma->pin_count > 0)
+		if (i915_vma_is_pinned(vma))
 			pin_count++;
 	}
 	seq_printf(m, " (pinned x %d)", pin_count);
@@ -174,10 +178,13 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 	if (obj->fence_reg != I915_FENCE_REG_NONE)
 		seq_printf(m, " (fence: %d)", obj->fence_reg);
 	list_for_each_entry(vma, &obj->vma_list, obj_link) {
+		if (!drm_mm_node_allocated(&vma->node))
+			continue;
+
 		seq_printf(m, " (%sgtt offset: %08llx, size: %08llx",
-			   vma->is_ggtt ? "g" : "pp",
+			   i915_vma_is_ggtt(vma) ? "g" : "pp",
 			   vma->node.start, vma->node.size);
-		if (vma->is_ggtt)
+		if (i915_vma_is_ggtt(vma))
 			seq_printf(m, ", type: %u", vma->ggtt_view.type);
 		seq_puts(m, ")");
 	}
@@ -192,11 +199,15 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		*t = '\0';
 		seq_printf(m, " (%s mappable)", s);
 	}
-	if (obj->last_write_req != NULL)
-		seq_printf(m, " (%s)",
-			   i915_gem_request_get_engine(obj->last_write_req)->name);
-	if (obj->frontbuffer_bits)
-		seq_printf(m, " (frontbuffer: 0x%03x)", obj->frontbuffer_bits);
+
+	engine = i915_gem_active_get_engine(&obj->last_write,
+					    &obj->base.dev->struct_mutex);
+	if (engine)
+		seq_printf(m, " (%s)", engine->name);
+
+	frontbuffer_bits = atomic_read(&obj->frontbuffer_bits);
+	if (frontbuffer_bits)
+		seq_printf(m, " (frontbuffer: 0x%03x)", frontbuffer_bits);
 }
 
 static int i915_gem_object_list_info(struct seq_file *m, void *data)
@@ -338,46 +349,29 @@ static int per_file_stats(int id, void *ptr, void *data)
 
 	stats->count++;
 	stats->total += obj->base.size;
-
+	if (!obj->bind_count)
+		stats->unbound += obj->base.size;
 	if (obj->base.name || obj->base.dma_buf)
 		stats->shared += obj->base.size;
 
-	if (USES_FULL_PPGTT(obj->base.dev)) {
-		list_for_each_entry(vma, &obj->vma_list, obj_link) {
-			struct i915_hw_ppgtt *ppgtt;
-
-			if (!drm_mm_node_allocated(&vma->node))
-				continue;
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
+		if (!drm_mm_node_allocated(&vma->node))
+			continue;
 
-			if (vma->is_ggtt) {
-				stats->global += obj->base.size;
-				continue;
-			}
+		if (i915_vma_is_ggtt(vma)) {
+			stats->global += vma->node.size;
+		} else {
+			struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vma->vm);
 
-			ppgtt = container_of(vma->vm, struct i915_hw_ppgtt, base);
-			if (ppgtt->file_priv != stats->file_priv)
+			if (ppgtt->base.file != stats->file_priv)
 				continue;
-
-			if (obj->active) /* XXX per-vma statistic */
-				stats->active += obj->base.size;
-			else
-				stats->inactive += obj->base.size;
-
-			return 0;
-		}
-	} else {
-		if (i915_gem_obj_ggtt_bound(obj)) {
-			stats->global += obj->base.size;
-			if (obj->active)
-				stats->active += obj->base.size;
-			else
-				stats->inactive += obj->base.size;
-			return 0;
 		}
-	}
 
-	if (!list_empty(&obj->global_list))
-		stats->unbound += obj->base.size;
+		if (i915_vma_is_active(vma))
+			stats->active += vma->node.size;
+		else
+			stats->inactive += vma->node.size;
+	}
 
 	return 0;
 }
@@ -425,8 +419,8 @@ static int per_file_ctx_stats(int id, void *ptr, void *data)
 	for (n = 0; n < ARRAY_SIZE(ctx->engine); n++) {
 		if (ctx->engine[n].state)
 			per_file_stats(0, ctx->engine[n].state, data);
-		if (ctx->engine[n].ringbuf)
-			per_file_stats(0, ctx->engine[n].ringbuf->obj, data);
+		if (ctx->engine[n].ring)
+			per_file_stats(0, ctx->engine[n].ring->obj, data);
 	}
 
 	return 0;
@@ -754,13 +748,13 @@ static int i915_gem_request_info(struct seq_file *m, void *data)
 		int count;
 
 		count = 0;
-		list_for_each_entry(req, &engine->request_list, list)
+		list_for_each_entry(req, &engine->request_list, link)
 			count++;
 		if (count == 0)
 			continue;
 
 		seq_printf(m, "%s requests: %d\n", engine->name, count);
-		list_for_each_entry(req, &engine->request_list, list) {
+		list_for_each_entry(req, &engine->request_list, link) {
 			struct task_struct *task;
 
 			rcu_read_lock();
@@ -768,7 +762,7 @@ static int i915_gem_request_info(struct seq_file *m, void *data)
 			if (req->pid)
 				task = pid_task(req->pid, PIDTYPE_PID);
 			seq_printf(m, "    %x @ %d: %s [%d]\n",
-				   req->seqno,
+				   req->fence.seqno,
 				   (int) (jiffies - req->emitted_jiffies),
 				   task ? task->comm : "<unknown>",
 				   task ? task->pid : -1);
@@ -1205,8 +1199,6 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 
 	intel_runtime_pm_get(dev_priv);
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
 	if (IS_GEN5(dev)) {
 		u16 rgvswctl = I915_READ16(MEMSWCTL);
 		u16 rgvstat = I915_READ16(MEMSTAT_ILK);
@@ -1381,6 +1373,8 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 			   intel_gpu_freq(dev_priv, dev_priv->rps.idle_freq));
 		seq_printf(m, "Min freq: %d MHz\n",
 			   intel_gpu_freq(dev_priv, dev_priv->rps.min_freq));
+		seq_printf(m, "Boost freq: %d MHz\n",
+			   intel_gpu_freq(dev_priv, dev_priv->rps.boost_freq));
 		seq_printf(m, "Max freq: %d MHz\n",
 			   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq));
 		seq_printf(m,
@@ -1419,7 +1413,7 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 	intel_runtime_pm_get(dev_priv);
 
 	for_each_engine_id(engine, dev_priv, id) {
-		acthd[id] = intel_ring_get_active_head(engine);
+		acthd[id] = intel_engine_get_active_head(engine);
 		seqno[id] = intel_engine_get_seqno(engine);
 	}
 
@@ -1602,6 +1596,7 @@ static int gen6_drpc_info(struct seq_file *m)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	u32 rpmodectl1, gt_core_status, rcctl1, rc6vids = 0;
+	u32 gen9_powergate_enable = 0, gen9_powergate_status = 0;
 	unsigned forcewake_count;
 	int count = 0, ret;
 
@@ -1629,6 +1624,10 @@ static int gen6_drpc_info(struct seq_file *m)
 
 	rpmodectl1 = I915_READ(GEN6_RP_CONTROL);
 	rcctl1 = I915_READ(GEN6_RC_CONTROL);
+	if (INTEL_INFO(dev)->gen >= 9) {
+		gen9_powergate_enable = I915_READ(GEN9_PG_ENABLE);
+		gen9_powergate_status = I915_READ(GEN9_PWRGT_DOMAIN_STATUS);
+	}
 	mutex_unlock(&dev->struct_mutex);
 	mutex_lock(&dev_priv->rps.hw_lock);
 	sandybridge_pcode_read(dev_priv, GEN6_PCODE_READ_RC6VIDS, &rc6vids);
@@ -1647,6 +1646,12 @@ static int gen6_drpc_info(struct seq_file *m)
 		   yesno(rcctl1 & GEN6_RC_CTL_RC1e_ENABLE));
 	seq_printf(m, "RC6 Enabled: %s\n",
 		   yesno(rcctl1 & GEN6_RC_CTL_RC6_ENABLE));
+	if (INTEL_INFO(dev)->gen >= 9) {
+		seq_printf(m, "Render Well Gating Enabled: %s\n",
+			yesno(gen9_powergate_enable & GEN9_RENDER_PG_ENABLE));
+		seq_printf(m, "Media Well Gating Enabled: %s\n",
+			yesno(gen9_powergate_enable & GEN9_MEDIA_PG_ENABLE));
+	}
 	seq_printf(m, "Deep RC6 Enabled: %s\n",
 		   yesno(rcctl1 & GEN6_RC_CTL_RC6p_ENABLE));
 	seq_printf(m, "Deepest RC6 Enabled: %s\n",
@@ -1675,6 +1680,14 @@ static int gen6_drpc_info(struct seq_file *m)
 
 	seq_printf(m, "Core Power Down: %s\n",
 		   yesno(gt_core_status & GEN6_CORE_CPD_STATE_MASK));
+	if (INTEL_INFO(dev)->gen >= 9) {
+		seq_printf(m, "Render Power Well: %s\n",
+			(gen9_powergate_status &
+			 GEN9_PWRGT_RENDER_STATUS_MASK) ? "Up" : "Down");
+		seq_printf(m, "Media Power Well: %s\n",
+			(gen9_powergate_status &
+			 GEN9_PWRGT_MEDIA_STATUS_MASK) ? "Up" : "Down");
+	}
 
 	/* Not exactly sure what this is */
 	seq_printf(m, "RC6 \"Locked to RPn\" residency since boot: %u\n",
@@ -1692,7 +1705,7 @@ static int gen6_drpc_info(struct seq_file *m)
 		   GEN6_DECODE_RC6_VID(((rc6vids >> 8) & 0xff)));
 	seq_printf(m, "RC6++ voltage: %dmV\n",
 		   GEN6_DECODE_RC6_VID(((rc6vids >> 16) & 0xff)));
-	return 0;
+	return i915_forcewake_domains(m, NULL);
 }
 
 static int i915_drpc_info(struct seq_file *m, void *unused)
@@ -1896,8 +1909,6 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused)
 
 	intel_runtime_pm_get(dev_priv);
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
 	if (ret)
 		goto out;
@@ -2019,12 +2030,11 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 	return 0;
 }
 
-static void describe_ctx_ringbuf(struct seq_file *m,
-				 struct intel_ringbuffer *ringbuf)
+static void describe_ctx_ring(struct seq_file *m, struct intel_ring *ring)
 {
 	seq_printf(m, " (ringbuffer, space: %d, head: %u, tail: %u, last head: %d)",
-		   ringbuf->space, ringbuf->head, ringbuf->tail,
-		   ringbuf->last_retired_head);
+		   ring->space, ring->head, ring->tail,
+		   ring->last_retired_head);
 }
 
 static int i915_context_status(struct seq_file *m, void *unused)
@@ -2068,8 +2078,8 @@ static int i915_context_status(struct seq_file *m, void *unused)
 			seq_putc(m, ce->initialised ? 'I' : 'i');
 			if (ce->state)
 				describe_obj(m, ce->state);
-			if (ce->ringbuf)
-				describe_ctx_ringbuf(m, ce->ringbuf);
+			if (ce->ring)
+				describe_ctx_ring(m, ce->ring);
 			seq_putc(m, '\n');
 		}
 
@@ -2467,13 +2477,7 @@ static int i915_rps_boost_info(struct seq_file *m, void *data)
 			   list_empty(&file_priv->rps.link) ? "" : ", active");
 		rcu_read_unlock();
 	}
-	seq_printf(m, "Semaphore boosts: %d%s\n",
-		   dev_priv->rps.semaphores.boosts,
-		   list_empty(&dev_priv->rps.semaphores.link) ? "" : ", active");
-	seq_printf(m, "MMIO flip boosts: %d%s\n",
-		   dev_priv->rps.mmioflips.boosts,
-		   list_empty(&dev_priv->rps.mmioflips.link) ? "" : ", active");
-	seq_printf(m, "Kernel boosts: %d\n", dev_priv->rps.boosts);
+	seq_printf(m, "Kernel (anonymous) boosts: %d\n", dev_priv->rps.boosts);
 	spin_unlock(&dev_priv->rps.client_lock);
 	mutex_unlock(&dev->filelist_mutex);
 
@@ -3228,7 +3232,7 @@ static int i915_semaphore_status(struct seq_file *m, void *unused)
 	enum intel_engine_id id;
 	int j, ret;
 
-	if (!i915_semaphore_is_enabled(dev_priv)) {
+	if (!i915.semaphores) {
 		seq_puts(m, "Semaphores are disabled\n");
 		return 0;
 	}
@@ -3621,7 +3625,6 @@ i915_pipe_crc_read(struct file *filep, char __user *user_buf, size_t count,
 	while (n_entries > 0) {
 		struct intel_pipe_crc_entry *entry =
 			&pipe_crc->entries[pipe_crc->tail];
-		int ret;
 
 		if (CIRC_CNT(pipe_crc->head, pipe_crc->tail,
 			     INTEL_PIPE_CRC_ENTRIES_NR) < 1)
@@ -3638,8 +3641,7 @@ i915_pipe_crc_read(struct file *filep, char __user *user_buf, size_t count,
 
 		spin_unlock_irq(&pipe_crc->lock);
 
-		ret = copy_to_user(user_buf, buf, PIPE_CRC_LINE_LEN);
-		if (ret == PIPE_CRC_LINE_LEN)
+		if (copy_to_user(user_buf, buf, PIPE_CRC_LINE_LEN))
 			return -EFAULT;
 
 		user_buf += PIPE_CRC_LINE_LEN;
@@ -4921,7 +4923,7 @@ i915_drop_caches_set(void *data, u64 val)
 		return ret;
 
 	if (val & DROP_ACTIVE) {
-		ret = i915_gem_wait_for_idle(dev_priv);
+		ret = i915_gem_wait_for_idle(dev_priv, true);
 		if (ret)
 			goto unlock;
 	}
@@ -4950,20 +4952,11 @@ i915_max_freq_get(void *data, u64 *val)
 {
 	struct drm_device *dev = data;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
 
 	if (INTEL_INFO(dev)->gen < 6)
 		return -ENODEV;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
-	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
-	if (ret)
-		return ret;
-
 	*val = intel_gpu_freq(dev_priv, dev_priv->rps.max_freq_softlimit);
-	mutex_unlock(&dev_priv->rps.hw_lock);
-
 	return 0;
 }
 
@@ -4978,8 +4971,6 @@ i915_max_freq_set(void *data, u64 val)
 	if (INTEL_INFO(dev)->gen < 6)
 		return -ENODEV;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
 	DRM_DEBUG_DRIVER("Manually setting max freq to %llu\n", val);
 
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
@@ -5017,20 +5008,11 @@ i915_min_freq_get(void *data, u64 *val)
 {
 	struct drm_device *dev = data;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
 
-	if (INTEL_INFO(dev)->gen < 6)
+	if (INTEL_GEN(dev_priv) < 6)
 		return -ENODEV;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
-	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
-	if (ret)
-		return ret;
-
 	*val = intel_gpu_freq(dev_priv, dev_priv->rps.min_freq_softlimit);
-	mutex_unlock(&dev_priv->rps.hw_lock);
-
 	return 0;
 }
 
@@ -5042,11 +5024,9 @@ i915_min_freq_set(void *data, u64 val)
 	u32 hw_max, hw_min;
 	int ret;
 
-	if (INTEL_INFO(dev)->gen < 6)
+	if (INTEL_GEN(dev_priv) < 6)
 		return -ENODEV;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
 	DRM_DEBUG_DRIVER("Manually setting min freq to %llu\n", val);
 
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
@@ -5268,7 +5248,8 @@ static void broadwell_sseu_device_status(struct drm_device *dev,
 static int i915_sseu_status(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
-	struct drm_device *dev = node->minor->dev;
+	struct drm_i915_private *dev_priv = to_i915(node->minor->dev);
+	struct drm_device *dev = &dev_priv->drm;
 	struct sseu_dev_status stat;
 
 	if (INTEL_INFO(dev)->gen < 8)
@@ -5298,6 +5279,9 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
 
 	seq_puts(m, "SSEU Device Status\n");
 	memset(&stat, 0, sizeof(stat));
+
+	intel_runtime_pm_get(dev_priv);
+
 	if (IS_CHERRYVIEW(dev)) {
 		cherryview_sseu_device_status(dev, &stat);
 	} else if (IS_BROADWELL(dev)) {
@@ -5305,6 +5289,9 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
 	} else if (INTEL_INFO(dev)->gen >= 9) {
 		gen9_sseu_device_status(dev, &stat);
 	}
+
+	intel_runtime_pm_put(dev_priv);
+
 	seq_printf(m, "  Enabled Slice Total: %u\n",
 		   stat.slice_total);
 	seq_printf(m, "  Enabled Subslice Total: %u\n",

drivers/gpu/drm/i915/i915_drv.c

@@ -228,27 +228,6 @@ static void intel_detect_pch(struct drm_device *dev)
 	pci_dev_put(pch);
 }
 
-bool i915_semaphore_is_enabled(struct drm_i915_private *dev_priv)
-{
-	if (INTEL_GEN(dev_priv) < 6)
-		return false;
-
-	if (i915.semaphores >= 0)
-		return i915.semaphores;
-
-	/* TODO: make semaphores and Execlists play nicely together */
-	if (i915.enable_execlists)
-		return false;
-
-#ifdef CONFIG_INTEL_IOMMU
-	/* Enable semaphores on SNB when IO remapping is off */
-	if (IS_GEN6(dev_priv) && intel_iommu_gfx_mapped)
-		return false;
-#endif
-
-	return true;
-}
-
 static int i915_getparam(struct drm_device *dev, void *data,
 			 struct drm_file *file_priv)
 {
@@ -324,7 +303,7 @@ static int i915_getparam(struct drm_device *dev, void *data,
 		value = 1;
 		break;
 	case I915_PARAM_HAS_SEMAPHORES:
-		value = i915_semaphore_is_enabled(dev_priv);
+		value = i915.semaphores;
 		break;
 	case I915_PARAM_HAS_PRIME_VMAP_FLUSH:
 		value = 1;
@@ -999,6 +978,9 @@ static void intel_sanitize_options(struct drm_i915_private *dev_priv)
 	i915.enable_ppgtt =
 		intel_sanitize_enable_ppgtt(dev_priv, i915.enable_ppgtt);
 	DRM_DEBUG_DRIVER("ppgtt mode: %i\n", i915.enable_ppgtt);
+
+	i915.semaphores = intel_sanitize_semaphores(dev_priv, i915.semaphores);
+	DRM_DEBUG_DRIVER("use GPU sempahores? %s\n", yesno(i915.semaphores));
 }
 
 /**
@@ -1011,8 +993,6 @@ static void intel_sanitize_options(struct drm_i915_private *dev_priv)
 static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = &dev_priv->drm;
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-	uint32_t aperture_size;
 	int ret;
 
 	if (i915_inject_load_failure())
@@ -1022,16 +1002,10 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 
 	intel_sanitize_options(dev_priv);
 
-	ret = i915_ggtt_init_hw(dev);
+	ret = i915_ggtt_probe_hw(dev_priv);
 	if (ret)
 		return ret;
 
-	ret = i915_ggtt_enable_hw(dev);
-	if (ret) {
-		DRM_ERROR("failed to enable GGTT\n");
-		goto out_ggtt;
-	}
-
 	/* WARNING: Apparently we must kick fbdev drivers before vgacon,
 	 * otherwise the vga fbdev driver falls over. */
 	ret = i915_kick_out_firmware_fb(dev_priv);
@@ -1046,6 +1020,16 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 		goto out_ggtt;
 	}
 
+	ret = i915_ggtt_init_hw(dev_priv);
+	if (ret)
+		return ret;
+
+	ret = i915_ggtt_enable_hw(dev_priv);
+	if (ret) {
+		DRM_ERROR("failed to enable GGTT\n");
+		goto out_ggtt;
+	}
+
 	pci_set_master(dev->pdev);
 
 	/* overlay on gen2 is broken and can't address above 1G */
@@ -1058,7 +1042,6 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 		}
 	}
 
-
 	/* 965GM sometimes incorrectly writes to hardware status page (HWS)
 	 * using 32bit addressing, overwriting memory if HWS is located
 	 * above 4GB.
@@ -1077,19 +1060,6 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 		}
 	}
 
-	aperture_size = ggtt->mappable_end;
-
-	ggtt->mappable =
-		io_mapping_create_wc(ggtt->mappable_base,
-				     aperture_size);
-	if (!ggtt->mappable) {
-		ret = -EIO;
-		goto out_ggtt;
-	}
-
-	ggtt->mtrr = arch_phys_wc_add(ggtt->mappable_base,
-					      aperture_size);
-
 	pm_qos_add_request(&dev_priv->pm_qos, PM_QOS_CPU_DMA_LATENCY,
 			   PM_QOS_DEFAULT_VALUE);
 
@@ -1118,7 +1088,7 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 	return 0;
 
 out_ggtt:
-	i915_ggtt_cleanup_hw(dev);
+	i915_ggtt_cleanup_hw(dev_priv);
 
 	return ret;
 }
@@ -1130,15 +1100,12 @@ out_ggtt:
 static void i915_driver_cleanup_hw(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = &dev_priv->drm;
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 
 	if (dev->pdev->msi_enabled)
 		pci_disable_msi(dev->pdev);
 
 	pm_qos_remove_request(&dev_priv->pm_qos);
-	arch_phys_wc_del(ggtt->mtrr);
-	io_mapping_free(ggtt->mappable);
-	i915_ggtt_cleanup_hw(dev);
+	i915_ggtt_cleanup_hw(dev_priv);
 }
 
 /**
@@ -1343,7 +1310,7 @@ void i915_driver_unload(struct drm_device *dev)
 	i915_destroy_error_state(dev);
 
 	/* Flush any outstanding unpin_work. */
-	flush_workqueue(dev_priv->wq);
+	drain_workqueue(dev_priv->wq);
 
 	intel_guc_fini(dev);
 	i915_gem_fini(dev);
@@ -1458,8 +1425,6 @@ static int i915_drm_suspend(struct drm_device *dev)
 
 	intel_guc_suspend(dev);
 
-	intel_suspend_gt_powersave(dev_priv);
-
 	intel_display_suspend(dev);
 
 	intel_dp_mst_suspend(dev);
@@ -1586,15 +1551,13 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	disable_rpm_wakeref_asserts(dev_priv);
 
-	ret = i915_ggtt_enable_hw(dev);
+	ret = i915_ggtt_enable_hw(dev_priv);
 	if (ret)
 		DRM_ERROR("failed to re-enable GGTT\n");
 
 	intel_csr_ucode_resume(dev_priv);
 
-	mutex_lock(&dev->struct_mutex);
-	i915_gem_restore_gtt_mappings(dev);
-	mutex_unlock(&dev->struct_mutex);
+	i915_gem_resume(dev);
 
 	i915_restore_state(dev);
 	intel_opregion_setup(dev_priv);
@@ -1652,6 +1615,7 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	intel_opregion_notify_adapter(dev_priv, PCI_D0);
 
+	intel_autoenable_gt_powersave(dev_priv);
 	drm_kms_helper_poll_enable(dev);
 
 	enable_rpm_wakeref_asserts(dev_priv);
@@ -1778,8 +1742,6 @@ int i915_reset(struct drm_i915_private *dev_priv)
 	unsigned reset_counter;
 	int ret;
 
-	intel_reset_gt_powersave(dev_priv);
-
 	mutex_lock(&dev->struct_mutex);
 
 	/* Clear any previous failed attempts at recovery. Time to try again. */
@@ -1835,8 +1797,7 @@ int i915_reset(struct drm_i915_private *dev_priv)
 	 * previous concerns that it doesn't respond well to some forms
 	 * of re-init after reset.
 	 */
-	if (INTEL_INFO(dev)->gen > 5)
-		intel_enable_gt_powersave(dev_priv);
+	intel_autoenable_gt_powersave(dev_priv);
 
 	return 0;
 
@@ -2462,7 +2423,6 @@ static int intel_runtime_resume(struct device *device)
 	 * we can do is to hope that things will still work (and disable RPM).
 	 */
 	i915_gem_init_swizzling(dev);
-	gen6_update_ring_freq(dev_priv);
 
 	intel_runtime_pm_enable_interrupts(dev_priv);
 
@@ -2618,6 +2578,7 @@ static struct drm_driver driver = {
 	.postclose = i915_driver_postclose,
 	.set_busid = drm_pci_set_busid,
 
+	.gem_close_object = i915_gem_close_object,
 	.gem_free_object = i915_gem_free_object,
 	.gem_vm_ops = &i915_gem_vm_ops,
 

drivers/gpu/drm/i915/i915_drv.h

@@ -61,6 +61,7 @@
 #include "i915_gem.h"
 #include "i915_gem_gtt.h"
 #include "i915_gem_render_state.h"
+#include "i915_gem_request.h"
 
 #include "intel_gvt.h"
 
@@ -69,7 +70,7 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20160711"
+#define DRIVER_DATE		"20160808"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
@@ -401,7 +402,7 @@ struct drm_i915_file_private {
 		unsigned boosts;
 	} rps;
 
-	unsigned int bsd_ring;
+	unsigned int bsd_engine;
 };
 
 /* Used by dp and fdi links */
@@ -431,8 +432,6 @@ void intel_link_compute_m_n(int bpp, int nlanes,
 #define DRIVER_MINOR		6
 #define DRIVER_PATCHLEVEL	0
 
-#define WATCH_LISTS	0
-
 struct opregion_header;
 struct opregion_acpi;
 struct opregion_swsci;
@@ -511,13 +510,13 @@ struct drm_i915_error_state {
 	struct intel_display_error_state *display;
 	struct drm_i915_error_object *semaphore_obj;
 
-	struct drm_i915_error_ring {
-		bool valid;
+	struct drm_i915_error_engine {
+		int engine_id;
 		/* Software tracked state */
 		bool waiting;
 		int num_waiters;
 		int hangcheck_score;
-		enum intel_ring_hangcheck_action hangcheck_action;
+		enum intel_engine_hangcheck_action hangcheck_action;
 		int num_requests;
 
 		/* our own tracking of ring head and tail */
@@ -577,7 +576,7 @@ struct drm_i915_error_state {
 
 		pid_t pid;
 		char comm[TASK_COMM_LEN];
-	} ring[I915_NUM_ENGINES];
+	} engine[I915_NUM_ENGINES];
 
 	struct drm_i915_error_buffer {
 		u32 size;
@@ -592,7 +591,7 @@ struct drm_i915_error_state {
 		u32 dirty:1;
 		u32 purgeable:1;
 		u32 userptr:1;
-		s32 ring:4;
+		s32 engine:4;
 		u32 cache_level:3;
 	} **active_bo, **pinned_bo;
 
@@ -893,7 +892,7 @@ struct i915_gem_context {
 
 	struct intel_context {
 		struct drm_i915_gem_object *state;
-		struct intel_ringbuffer *ringbuf;
+		struct intel_ring *ring;
 		struct i915_vma *lrc_vma;
 		uint32_t *lrc_reg_state;
 		u64 lrc_desc;
@@ -908,6 +907,7 @@ struct i915_gem_context {
 	struct list_head link;
 
 	u8 remap_slice;
+	bool closed:1;
 };
 
 enum fb_op_origin {
@@ -1173,6 +1173,7 @@ struct intel_gen6_power_mgmt {
 	u8 max_freq_softlimit;	/* Max frequency permitted by the driver */
 	u8 max_freq;		/* Maximum frequency, RP0 if not overclocking */
 	u8 min_freq;		/* AKA RPn. Minimum frequency */
+	u8 boost_freq;		/* Frequency to request when wait boosting */
 	u8 idle_freq;		/* Frequency to request when we are idle */
 	u8 efficient_freq;	/* AKA RPe. Pre-determined balanced frequency */
 	u8 rp1_freq;		/* "less than" RP0 power/freqency */
@@ -1190,11 +1191,9 @@ struct intel_gen6_power_mgmt {
 	bool client_boost;
 
 	bool enabled;
-	struct delayed_work delayed_resume_work;
+	struct delayed_work autoenable_work;
 	unsigned boosts;
 
-	struct intel_rps_client semaphores, mmioflips;
-
 	/* manual wa residency calculations */
 	struct intel_rps_ei up_ei, down_ei;
 
@@ -1319,7 +1318,6 @@ struct i915_gem_mm {
 	struct notifier_block oom_notifier;
 	struct notifier_block vmap_notifier;
 	struct shrinker shrinker;
-	bool shrinker_no_lock_stealing;
 
 	/** LRU list of objects with fence regs on them. */
 	struct list_head fence_list;
@@ -1331,7 +1329,7 @@ struct i915_gem_mm {
 	bool interruptible;
 
 	/* the indicator for dispatch video commands on two BSD rings */
-	unsigned int bsd_ring_dispatch_index;
+	unsigned int bsd_engine_dispatch_index;
 
 	/** Bit 6 swizzling required for X tiling */
 	uint32_t bit_6_swizzle_x;
@@ -1670,7 +1668,7 @@ struct intel_pipe_crc {
 };
 
 struct i915_frontbuffer_tracking {
-	struct mutex lock;
+	spinlock_t lock;
 
 	/*
 	 * Tracking bits for delayed frontbuffer flushing du to gpu activity or
@@ -1705,18 +1703,6 @@ struct i915_virtual_gpu {
 	bool active;
 };
 
-struct i915_execbuffer_params {
-	struct drm_device               *dev;
-	struct drm_file                 *file;
-	uint32_t                        dispatch_flags;
-	uint32_t                        args_batch_start_offset;
-	uint64_t                        batch_obj_vm_offset;
-	struct intel_engine_cs *engine;
-	struct drm_i915_gem_object      *batch_obj;
-	struct i915_gem_context            *ctx;
-	struct drm_i915_gem_request     *request;
-};
-
 /* used in computing the new watermarks state */
 struct intel_wm_config {
 	unsigned int num_pipes_active;
@@ -1769,7 +1755,7 @@ struct drm_i915_private {
 	struct i915_gem_context *kernel_context;
 	struct intel_engine_cs engine[I915_NUM_ENGINES];
 	struct drm_i915_gem_object *semaphore_obj;
-	uint32_t last_seqno, next_seqno;
+	u32 next_seqno;
 
 	struct drm_dma_handle *status_page_dmah;
 	struct resource mch_res;
@@ -2016,12 +2002,7 @@ struct drm_i915_private {
 
 	/* Abstract the submission mechanism (legacy ringbuffer or execlists) away */
 	struct {
-		int (*execbuf_submit)(struct i915_execbuffer_params *params,
-				      struct drm_i915_gem_execbuffer2 *args,
-				      struct list_head *vmas);
-		int (*init_engines)(struct drm_device *dev);
 		void (*cleanup_engine)(struct intel_engine_cs *engine);
-		void (*stop_engine)(struct intel_engine_cs *engine);
 
 		/**
 		 * Is the GPU currently considered idle, or busy executing
@@ -2144,8 +2125,6 @@ struct drm_i915_gem_object_ops {
  */
 #define INTEL_MAX_SPRITE_BITS_PER_PIPE 5
 #define INTEL_FRONTBUFFER_BITS_PER_PIPE 8
-#define INTEL_FRONTBUFFER_BITS \
-	(INTEL_FRONTBUFFER_BITS_PER_PIPE * I915_MAX_PIPES)
 #define INTEL_FRONTBUFFER_PRIMARY(pipe) \
 	(1 << (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe)))
 #define INTEL_FRONTBUFFER_CURSOR(pipe) \
@@ -2169,18 +2148,21 @@ struct drm_i915_gem_object {
 	struct drm_mm_node *stolen;
 	struct list_head global_list;
 
-	struct list_head engine_list[I915_NUM_ENGINES];
 	/** Used in execbuf to temporarily hold a ref */
 	struct list_head obj_exec_link;
 
 	struct list_head batch_pool_link;
 
+	unsigned long flags;
 	/**
 	 * This is set if the object is on the active lists (has pending
 	 * rendering and so a non-zero seqno), and is not set if it i s on
 	 * inactive (ready to be unbound) list.
 	 */
-	unsigned int active:I915_NUM_ENGINES;
+#define I915_BO_ACTIVE_SHIFT 0
+#define I915_BO_ACTIVE_MASK ((1 << I915_NUM_ENGINES) - 1)
+#define __I915_BO_ACTIVE(bo) \
+	((READ_ONCE((bo)->flags) >> I915_BO_ACTIVE_SHIFT) & I915_BO_ACTIVE_MASK)
 
 	/**
 	 * This is set if the object has been written to since last bound
@@ -2200,10 +2182,6 @@ struct drm_i915_gem_object {
 	 */
 	unsigned int madv:2;
 
-	/**
-	 * Current tiling mode for the object.
-	 */
-	unsigned int tiling_mode:2;
 	/**
 	 * Whether the tiling parameters for the currently associated fence
 	 * register have changed. Note that for the purposes of tracking
@@ -2234,9 +2212,17 @@ struct drm_i915_gem_object {
 	unsigned int cache_level:3;
 	unsigned int cache_dirty:1;
 
-	unsigned int frontbuffer_bits:INTEL_FRONTBUFFER_BITS;
+	atomic_t frontbuffer_bits;
+
+	/** Current tiling stride for the object, if it's tiled. */
+	unsigned int tiling_and_stride;
+#define FENCE_MINIMUM_STRIDE 128 /* See i915_tiling_ok() */
+#define TILING_MASK (FENCE_MINIMUM_STRIDE-1)
+#define STRIDE_MASK (~TILING_MASK)
 
 	unsigned int has_wc_mmap;
+	/** Count of VMA actually bound by this object */
+	unsigned int bind_count;
 	unsigned int pin_display;
 
 	struct sg_table *pages;
@@ -2256,14 +2242,10 @@ struct drm_i915_gem_object {
 	 * requests on one ring where the write request is older than the
 	 * read request. This allows for the CPU to read from an active
 	 * buffer by only waiting for the write to complete.
-	 * */
-	struct drm_i915_gem_request *last_read_req[I915_NUM_ENGINES];
-	struct drm_i915_gem_request *last_write_req;
-	/** Breadcrumb of last fenced GPU access to the buffer. */
-	struct drm_i915_gem_request *last_fenced_req;
-
-	/** Current tiling stride for the object, if it's tiled. */
-	uint32_t stride;
+	 */
+	struct i915_gem_active last_read[I915_NUM_ENGINES];
+	struct i915_gem_active last_write;
+	struct i915_gem_active last_fence;
 
 	/** References from framebuffers, locks out tiling changes. */
 	unsigned long framebuffer_references;
@@ -2287,7 +2269,56 @@ struct drm_i915_gem_object {
 		} userptr;
 	};
 };
-#define to_intel_bo(x) container_of(x, struct drm_i915_gem_object, base)
+
+static inline struct drm_i915_gem_object *
+to_intel_bo(struct drm_gem_object *gem)
+{
+	/* Assert that to_intel_bo(NULL) == NULL */
+	BUILD_BUG_ON(offsetof(struct drm_i915_gem_object, base));
+
+	return container_of(gem, struct drm_i915_gem_object, base);
+}
+
+static inline struct drm_i915_gem_object *
+i915_gem_object_lookup(struct drm_file *file, u32 handle)
+{
+	return to_intel_bo(drm_gem_object_lookup(file, handle));
+}
+
+__deprecated
+extern struct drm_gem_object *
+drm_gem_object_lookup(struct drm_file *file, u32 handle);
+
+__attribute__((nonnull))
+static inline struct drm_i915_gem_object *
+i915_gem_object_get(struct drm_i915_gem_object *obj)
+{
+	drm_gem_object_reference(&obj->base);
+	return obj;
+}
+
+__deprecated
+extern void drm_gem_object_reference(struct drm_gem_object *);
+
+__attribute__((nonnull))
+static inline void
+i915_gem_object_put(struct drm_i915_gem_object *obj)
+{
+	drm_gem_object_unreference(&obj->base);
+}
+
+__deprecated
+extern void drm_gem_object_unreference(struct drm_gem_object *);
+
+__attribute__((nonnull))
+static inline void
+i915_gem_object_put_unlocked(struct drm_i915_gem_object *obj)
+{
+	drm_gem_object_unreference_unlocked(&obj->base);
+}
+
+__deprecated
+extern void drm_gem_object_unreference_unlocked(struct drm_gem_object *);
 
 static inline bool
 i915_gem_object_has_struct_page(const struct drm_i915_gem_object *obj)
@@ -2295,6 +2326,55 @@ i915_gem_object_has_struct_page(const struct drm_i915_gem_object *obj)
 	return obj->ops->flags & I915_GEM_OBJECT_HAS_STRUCT_PAGE;
 }
 
+static inline unsigned long
+i915_gem_object_get_active(const struct drm_i915_gem_object *obj)
+{
+	return (obj->flags >> I915_BO_ACTIVE_SHIFT) & I915_BO_ACTIVE_MASK;
+}
+
+static inline bool
+i915_gem_object_is_active(const struct drm_i915_gem_object *obj)
+{
+	return i915_gem_object_get_active(obj);
+}
+
+static inline void
+i915_gem_object_set_active(struct drm_i915_gem_object *obj, int engine)
+{
+	obj->flags |= BIT(engine + I915_BO_ACTIVE_SHIFT);
+}
+
+static inline void
+i915_gem_object_clear_active(struct drm_i915_gem_object *obj, int engine)
+{
+	obj->flags &= ~BIT(engine + I915_BO_ACTIVE_SHIFT);
+}
+
+static inline bool
+i915_gem_object_has_active_engine(const struct drm_i915_gem_object *obj,
+				  int engine)
+{
+	return obj->flags & BIT(engine + I915_BO_ACTIVE_SHIFT);
+}
+
+static inline unsigned int
+i915_gem_object_get_tiling(struct drm_i915_gem_object *obj)
+{
+	return obj->tiling_and_stride & TILING_MASK;
+}
+
+static inline bool
+i915_gem_object_is_tiled(struct drm_i915_gem_object *obj)
+{
+	return i915_gem_object_get_tiling(obj) != I915_TILING_NONE;
+}
+
+static inline unsigned int
+i915_gem_object_get_stride(struct drm_i915_gem_object *obj)
+{
+	return obj->tiling_and_stride & STRIDE_MASK;
+}
+
 /*
  * Optimised SGL iterator for GEM objects
  */
@@ -2365,171 +2445,6 @@ static inline struct scatterlist *__sg_next(struct scatterlist *sg)
 	     (((__iter).curr += PAGE_SIZE) < (__iter).max) ||		\
 	     ((__iter) = __sgt_iter(__sg_next((__iter).sgp), false), 0))
 
-/**
- * Request queue structure.
- *
- * The request queue allows us to note sequence numbers that have been emitted
- * and may be associated with active buffers to be retired.
- *
- * By keeping this list, we can avoid having to do questionable sequence
- * number comparisons on buffer last_read|write_seqno. It also allows an
- * emission time to be associated with the request for tracking how far ahead
- * of the GPU the submission is.
- *
- * The requests are reference counted, so upon creation they should have an
- * initial reference taken using kref_init
- */
-struct drm_i915_gem_request {
-	struct kref ref;
-
-	/** On Which ring this request was generated */
-	struct drm_i915_private *i915;
-	struct intel_engine_cs *engine;
-	struct intel_signal_node signaling;
-
-	 /** GEM sequence number associated with the previous request,
-	  * when the HWS breadcrumb is equal to this the GPU is processing
-	  * this request.
-	  */
-	u32 previous_seqno;
-
-	 /** GEM sequence number associated with this request,
-	  * when the HWS breadcrumb is equal or greater than this the GPU
-	  * has finished processing this request.
-	  */
-	u32 seqno;
-
-	/** Position in the ringbuffer of the start of the request */
-	u32 head;
-
-	/**
-	 * Position in the ringbuffer of the start of the postfix.
-	 * This is required to calculate the maximum available ringbuffer
-	 * space without overwriting the postfix.
-	 */
-	 u32 postfix;
-
-	/** Position in the ringbuffer of the end of the whole request */
-	u32 tail;
-
-	/** Preallocate space in the ringbuffer for the emitting the request */
-	u32 reserved_space;
-
-	/**
-	 * Context and ring buffer related to this request
-	 * Contexts are refcounted, so when this request is associated with a
-	 * context, we must increment the context's refcount, to guarantee that
-	 * it persists while any request is linked to it. Requests themselves
-	 * are also refcounted, so the request will only be freed when the last
-	 * reference to it is dismissed, and the code in
-	 * i915_gem_request_free() will then decrement the refcount on the
-	 * context.
-	 */
-	struct i915_gem_context *ctx;
-	struct intel_ringbuffer *ringbuf;
-
-	/**
-	 * Context related to the previous request.
-	 * As the contexts are accessed by the hardware until the switch is
-	 * completed to a new context, the hardware may still be writing
-	 * to the context object after the breadcrumb is visible. We must
-	 * not unpin/unbind/prune that object whilst still active and so
-	 * we keep the previous context pinned until the following (this)
-	 * request is retired.
-	 */
-	struct i915_gem_context *previous_context;
-
-	/** Batch buffer related to this request if any (used for
-	    error state dump only) */
-	struct drm_i915_gem_object *batch_obj;
-
-	/** Time at which this request was emitted, in jiffies. */
-	unsigned long emitted_jiffies;
-
-	/** global list entry for this request */
-	struct list_head list;
-
-	struct drm_i915_file_private *file_priv;
-	/** file_priv list entry for this request */
-	struct list_head client_list;
-
-	/** process identifier submitting this request */
-	struct pid *pid;
-
-	/**
-	 * The ELSP only accepts two elements at a time, so we queue
-	 * context/tail pairs on a given queue (ring->execlist_queue) until the
-	 * hardware is available. The queue serves a double purpose: we also use
-	 * it to keep track of the up to 2 contexts currently in the hardware
-	 * (usually one in execution and the other queued up by the GPU): We
-	 * only remove elements from the head of the queue when the hardware
-	 * informs us that an element has been completed.
-	 *
-	 * All accesses to the queue are mediated by a spinlock
-	 * (ring->execlist_lock).
-	 */
-
-	/** Execlist link in the submission queue.*/
-	struct list_head execlist_link;
-
-	/** Execlists no. of times this request has been sent to the ELSP */
-	int elsp_submitted;
-
-	/** Execlists context hardware id. */
-	unsigned ctx_hw_id;
-};
-
-struct drm_i915_gem_request * __must_check
-i915_gem_request_alloc(struct intel_engine_cs *engine,
-		       struct i915_gem_context *ctx);
-void i915_gem_request_free(struct kref *req_ref);
-int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
-				   struct drm_file *file);
-
-static inline uint32_t
-i915_gem_request_get_seqno(struct drm_i915_gem_request *req)
-{
-	return req ? req->seqno : 0;
-}
-
-static inline struct intel_engine_cs *
-i915_gem_request_get_engine(struct drm_i915_gem_request *req)
-{
-	return req ? req->engine : NULL;
-}
-
-static inline struct drm_i915_gem_request *
-i915_gem_request_reference(struct drm_i915_gem_request *req)
-{
-	if (req)
-		kref_get(&req->ref);
-	return req;
-}
-
-static inline void
-i915_gem_request_unreference(struct drm_i915_gem_request *req)
-{
-	kref_put(&req->ref, i915_gem_request_free);
-}
-
-static inline void i915_gem_request_assign(struct drm_i915_gem_request **pdst,
-					   struct drm_i915_gem_request *src)
-{
-	if (src)
-		i915_gem_request_reference(src);
-
-	if (*pdst)
-		i915_gem_request_unreference(*pdst);
-
-	*pdst = src;
-}
-
-/*
- * XXX: i915_gem_request_completed should be here but currently needs the
- * definition of i915_seqno_passed() which is below. It will be moved in
- * a later patch when the call to i915_seqno_passed() is obsoleted...
- */
-
 /*
  * A command that requires special handling by the command parser.
  */
@@ -2617,8 +2532,9 @@ struct drm_i915_cmd_descriptor {
 /*
  * A table of commands requiring special handling by the command parser.
  *
- * Each ring has an array of tables. Each table consists of an array of command
- * descriptors, which must be sorted with command opcodes in ascending order.
+ * Each engine has an array of tables. Each table consists of an array of
+ * command descriptors, which must be sorted with command opcodes in
+ * ascending order.
  */
 struct drm_i915_cmd_table {
 	const struct drm_i915_cmd_descriptor *table;
@@ -2932,6 +2848,8 @@ extern int i915_resume_switcheroo(struct drm_device *dev);
 int intel_sanitize_enable_ppgtt(struct drm_i915_private *dev_priv,
 			       	int enable_ppgtt);
 
+bool intel_sanitize_semaphores(struct drm_i915_private *dev_priv, int value);
+
 /* i915_drv.c */
 void __printf(3, 4)
 __i915_printk(struct drm_i915_private *dev_priv, const char *level,
@@ -3107,11 +3025,6 @@ int i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 			      struct drm_file *file_priv);
 int i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
 			     struct drm_file *file_priv);
-void i915_gem_execbuffer_move_to_active(struct list_head *vmas,
-					struct drm_i915_gem_request *req);
-int i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
-				   struct drm_i915_gem_execbuffer2 *args,
-				   struct list_head *vmas);
 int i915_gem_execbuffer(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 int i915_gem_execbuffer2(struct drm_device *dev, void *data,
@@ -3150,40 +3063,24 @@ struct drm_i915_gem_object *i915_gem_object_create(struct drm_device *dev,
 						  size_t size);
 struct drm_i915_gem_object *i915_gem_object_create_from_data(
 		struct drm_device *dev, const void *data, size_t size);
+void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file);
 void i915_gem_free_object(struct drm_gem_object *obj);
-void i915_gem_vma_destroy(struct i915_vma *vma);
-
-/* Flags used by pin/bind&friends. */
-#define PIN_MAPPABLE	(1<<0)
-#define PIN_NONBLOCK	(1<<1)
-#define PIN_GLOBAL	(1<<2)
-#define PIN_OFFSET_BIAS	(1<<3)
-#define PIN_USER	(1<<4)
-#define PIN_UPDATE	(1<<5)
-#define PIN_ZONE_4G	(1<<6)
-#define PIN_HIGH	(1<<7)
-#define PIN_OFFSET_FIXED	(1<<8)
-#define PIN_OFFSET_MASK (~4095)
-int __must_check
-i915_gem_object_pin(struct drm_i915_gem_object *obj,
-		    struct i915_address_space *vm,
-		    uint32_t alignment,
-		    uint64_t flags);
+
 int __must_check
 i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
 			 const struct i915_ggtt_view *view,
-			 uint32_t alignment,
-			 uint64_t flags);
+			 u64 size,
+			 u64 alignment,
+			 u64 flags);
 
 int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 		  u32 flags);
 void __i915_vma_set_map_and_fenceable(struct i915_vma *vma);
 int __must_check i915_vma_unbind(struct i915_vma *vma);
-/*
- * BEWARE: Do not use the function below unless you can _absolutely_
- * _guarantee_ VMA in question is _not in use_ anywhere.
- */
-int __must_check __i915_vma_unbind_no_wait(struct i915_vma *vma);
+void i915_vma_close(struct i915_vma *vma);
+void i915_vma_destroy(struct i915_vma *vma);
+
+int i915_gem_object_unbind(struct drm_i915_gem_object *obj);
 int i915_gem_object_put_pages(struct drm_i915_gem_object *obj);
 void i915_gem_release_all_mmaps(struct drm_i915_private *dev_priv);
 void i915_gem_release_mmap(struct drm_i915_gem_object *obj);
@@ -3285,10 +3182,10 @@ static inline void i915_gem_object_unpin_map(struct drm_i915_gem_object *obj)
 
 int __must_check i915_mutex_lock_interruptible(struct drm_device *dev);
 int i915_gem_object_sync(struct drm_i915_gem_object *obj,
-			 struct intel_engine_cs *to,
-			 struct drm_i915_gem_request **to_req);
+			 struct drm_i915_gem_request *to);
 void i915_vma_move_to_active(struct i915_vma *vma,
-			     struct drm_i915_gem_request *req);
+			     struct drm_i915_gem_request *req,
+			     unsigned int flags);
 int i915_gem_dumb_create(struct drm_file *file_priv,
 			 struct drm_device *dev,
 			 struct drm_mode_create_dumb *args);
@@ -3299,44 +3196,12 @@ void i915_gem_track_fb(struct drm_i915_gem_object *old,
 		       struct drm_i915_gem_object *new,
 		       unsigned frontbuffer_bits);
 
-/**
- * Returns true if seq1 is later than seq2.
- */
-static inline bool
-i915_seqno_passed(uint32_t seq1, uint32_t seq2)
-{
-	return (int32_t)(seq1 - seq2) >= 0;
-}
-
-static inline bool i915_gem_request_started(const struct drm_i915_gem_request *req)
-{
-	return i915_seqno_passed(intel_engine_get_seqno(req->engine),
-				 req->previous_seqno);
-}
-
-static inline bool i915_gem_request_completed(const struct drm_i915_gem_request *req)
-{
-	return i915_seqno_passed(intel_engine_get_seqno(req->engine),
-				 req->seqno);
-}
-
-bool __i915_spin_request(const struct drm_i915_gem_request *request,
-			 int state, unsigned long timeout_us);
-static inline bool i915_spin_request(const struct drm_i915_gem_request *request,
-				     int state, unsigned long timeout_us)
-{
-	return (i915_gem_request_started(request) &&
-		__i915_spin_request(request, state, timeout_us));
-}
-
-int __must_check i915_gem_get_seqno(struct drm_i915_private *dev_priv, u32 *seqno);
 int __must_check i915_gem_set_seqno(struct drm_device *dev, u32 seqno);
 
 struct drm_i915_gem_request *
 i915_gem_find_active_request(struct intel_engine_cs *engine);
 
 void i915_gem_retire_requests(struct drm_i915_private *dev_priv);
-void i915_gem_retire_requests_ring(struct intel_engine_cs *engine);
 
 static inline u32 i915_reset_counter(struct i915_gpu_error *error)
 {
@@ -3381,24 +3246,13 @@ static inline u32 i915_reset_count(struct i915_gpu_error *error)
 void i915_gem_reset(struct drm_device *dev);
 bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, bool force);
 int __must_check i915_gem_init(struct drm_device *dev);
-int i915_gem_init_engines(struct drm_device *dev);
 int __must_check i915_gem_init_hw(struct drm_device *dev);
 void i915_gem_init_swizzling(struct drm_device *dev);
 void i915_gem_cleanup_engines(struct drm_device *dev);
-int __must_check i915_gem_wait_for_idle(struct drm_i915_private *dev_priv);
+int __must_check i915_gem_wait_for_idle(struct drm_i915_private *dev_priv,
+					bool interruptible);
 int __must_check i915_gem_suspend(struct drm_device *dev);
-void __i915_add_request(struct drm_i915_gem_request *req,
-			struct drm_i915_gem_object *batch_obj,
-			bool flush_caches);
-#define i915_add_request(req) \
-	__i915_add_request(req, NULL, true)
-#define i915_add_request_no_flush(req) \
-	__i915_add_request(req, NULL, false)
-int __i915_wait_request(struct drm_i915_gem_request *req,
-			bool interruptible,
-			s64 *timeout,
-			struct intel_rps_client *rps);
-int __must_check i915_wait_request(struct drm_i915_gem_request *req);
+void i915_gem_resume(struct drm_device *dev);
 int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
 int __must_check
 i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
@@ -3419,11 +3273,10 @@ int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj,
 int i915_gem_open(struct drm_device *dev, struct drm_file *file);
 void i915_gem_release(struct drm_device *dev, struct drm_file *file);
 
-uint32_t
-i915_gem_get_gtt_size(struct drm_device *dev, uint32_t size, int tiling_mode);
-uint32_t
-i915_gem_get_gtt_alignment(struct drm_device *dev, uint32_t size,
-			    int tiling_mode, bool fenced);
+u64 i915_gem_get_ggtt_size(struct drm_i915_private *dev_priv, u64 size,
+			   int tiling_mode);
+u64 i915_gem_get_ggtt_alignment(struct drm_i915_private *dev_priv, u64 size,
+				int tiling_mode, bool fenced);
 
 int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 				    enum i915_cache_level cache_level);
@@ -3444,7 +3297,6 @@ i915_gem_obj_ggtt_offset(struct drm_i915_gem_object *o)
 	return i915_gem_obj_ggtt_offset_view(o, &i915_ggtt_view_normal);
 }
 
-bool i915_gem_obj_bound_any(struct drm_i915_gem_object *o);
 bool i915_gem_obj_ggtt_bound_view(struct drm_i915_gem_object *o,
 				  const struct i915_ggtt_view *view);
 bool i915_gem_obj_bound(struct drm_i915_gem_object *o,
@@ -3478,7 +3330,6 @@ i915_vm_to_ppgtt(struct i915_address_space *vm)
 	return container_of(vm, struct i915_hw_ppgtt, base);
 }
 
-
 static inline bool i915_gem_obj_ggtt_bound(struct drm_i915_gem_object *obj)
 {
 	return i915_gem_obj_ggtt_bound_view(obj, &i915_ggtt_view_normal);
@@ -3487,18 +3338,6 @@ static inline bool i915_gem_obj_ggtt_bound(struct drm_i915_gem_object *obj)
 unsigned long
 i915_gem_obj_ggtt_size(struct drm_i915_gem_object *obj);
 
-static inline int __must_check
-i915_gem_obj_ggtt_pin(struct drm_i915_gem_object *obj,
-		      uint32_t alignment,
-		      unsigned flags)
-{
-	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-
-	return i915_gem_object_pin(obj, &ggtt->base,
-				   alignment, flags | PIN_GLOBAL);
-}
-
 void i915_gem_object_ggtt_unpin_view(struct drm_i915_gem_object *obj,
 				     const struct i915_ggtt_view *view);
 static inline void
@@ -3528,6 +3367,7 @@ void i915_gem_context_reset(struct drm_device *dev);
 int i915_gem_context_open(struct drm_device *dev, struct drm_file *file);
 void i915_gem_context_close(struct drm_device *dev, struct drm_file *file);
 int i915_switch_context(struct drm_i915_gem_request *req);
+int i915_gem_switch_to_kernel_context(struct drm_i915_private *dev_priv);
 void i915_gem_context_free(struct kref *ctx_ref);
 struct drm_i915_gem_object *
 i915_gem_alloc_context_obj(struct drm_device *dev, size_t size);
@@ -3548,12 +3388,14 @@ i915_gem_context_lookup(struct drm_i915_file_private *file_priv, u32 id)
 	return ctx;
 }
 
-static inline void i915_gem_context_reference(struct i915_gem_context *ctx)
+static inline struct i915_gem_context *
+i915_gem_context_get(struct i915_gem_context *ctx)
 {
 	kref_get(&ctx->ref);
+	return ctx;
 }
 
-static inline void i915_gem_context_unreference(struct i915_gem_context *ctx)
+static inline void i915_gem_context_put(struct i915_gem_context *ctx)
 {
 	lockdep_assert_held(&ctx->i915->drm.struct_mutex);
 	kref_put(&ctx->ref, i915_gem_context_free);
@@ -3576,13 +3418,10 @@ int i915_gem_context_reset_stats_ioctl(struct drm_device *dev, void *data,
 				       struct drm_file *file);
 
 /* i915_gem_evict.c */
-int __must_check i915_gem_evict_something(struct drm_device *dev,
-					  struct i915_address_space *vm,
-					  int min_size,
-					  unsigned alignment,
+int __must_check i915_gem_evict_something(struct i915_address_space *vm,
+					  u64 min_size, u64 alignment,
 					  unsigned cache_level,
-					  unsigned long start,
-					  unsigned long end,
+					  u64 start, u64 end,
 					  unsigned flags);
 int __must_check i915_gem_evict_for_vma(struct i915_vma *target);
 int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle);
@@ -3634,16 +3473,9 @@ static inline bool i915_gem_object_needs_bit17_swizzle(struct drm_i915_gem_objec
 	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 
 	return dev_priv->mm.bit_6_swizzle_x == I915_BIT_6_SWIZZLE_9_10_17 &&
-		obj->tiling_mode != I915_TILING_NONE;
+		i915_gem_object_is_tiled(obj);
 }
 
-/* i915_gem_debug.c */
-#if WATCH_LISTS
-int i915_verify_lists(struct drm_device *dev);
-#else
-#define i915_verify_lists(dev) 0
-#endif
-
 /* i915_debugfs.c */
 #ifdef CONFIG_DEBUG_FS
 int i915_debugfs_register(struct drm_i915_private *dev_priv);
@@ -3684,15 +3516,15 @@ const char *i915_cache_level_str(struct drm_i915_private *i915, int type);
 
 /* i915_cmd_parser.c */
 int i915_cmd_parser_get_version(struct drm_i915_private *dev_priv);
-int i915_cmd_parser_init_ring(struct intel_engine_cs *engine);
-void i915_cmd_parser_fini_ring(struct intel_engine_cs *engine);
-bool i915_needs_cmd_parser(struct intel_engine_cs *engine);
-int i915_parse_cmds(struct intel_engine_cs *engine,
-		    struct drm_i915_gem_object *batch_obj,
-		    struct drm_i915_gem_object *shadow_batch_obj,
-		    u32 batch_start_offset,
-		    u32 batch_len,
-		    bool is_master);
+int intel_engine_init_cmd_parser(struct intel_engine_cs *engine);
+void intel_engine_cleanup_cmd_parser(struct intel_engine_cs *engine);
+bool intel_engine_needs_cmd_parser(struct intel_engine_cs *engine);
+int intel_engine_cmd_parser(struct intel_engine_cs *engine,
+			    struct drm_i915_gem_object *batch_obj,
+			    struct drm_i915_gem_object *shadow_batch_obj,
+			    u32 batch_start_offset,
+			    u32 batch_len,
+			    bool is_master);
 
 /* i915_suspend.c */
 extern int i915_save_state(struct drm_device *dev);
@@ -3800,7 +3632,6 @@ extern void intel_set_rps(struct drm_i915_private *dev_priv, u8 val);
 extern void intel_set_memory_cxsr(struct drm_i915_private *dev_priv,
 				  bool enable);
 
-extern bool i915_semaphore_is_enabled(struct drm_i915_private *dev_priv);
 int i915_reg_read_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file);
 

drivers/gpu/drm/i915/i915_gem.c

@@ -29,10 +29,13 @@
 #include <drm/drm_vma_manager.h>
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
+#include "i915_gem_dmabuf.h"
 #include "i915_vgpu.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include "intel_frontbuffer.h"
 #include "intel_mocs.h"
+#include <linux/reservation.h>
 #include <linux/shmem_fs.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
@@ -41,10 +44,6 @@
 
 static void i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj);
 static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj);
-static void
-i915_gem_object_retire__write(struct drm_i915_gem_object *obj);
-static void
-i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring);
 
 static bool cpu_cache_is_coherent(struct drm_device *dev,
 				  enum i915_cache_level level)
@@ -139,7 +138,6 @@ int i915_mutex_lock_interruptible(struct drm_device *dev)
 	if (ret)
 		return ret;
 
-	WARN_ON(i915_verify_lists(dev));
 	return 0;
 }
 
@@ -156,10 +154,10 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 	pinned = 0;
 	mutex_lock(&dev->struct_mutex);
 	list_for_each_entry(vma, &ggtt->base.active_list, vm_link)
-		if (vma->pin_count)
+		if (i915_vma_is_pinned(vma))
 			pinned += vma->node.size;
 	list_for_each_entry(vma, &ggtt->base.inactive_list, vm_link)
-		if (vma->pin_count)
+		if (i915_vma_is_pinned(vma))
 			pinned += vma->node.size;
 	mutex_unlock(&dev->struct_mutex);
 
@@ -281,23 +279,119 @@ static const struct drm_i915_gem_object_ops i915_gem_phys_ops = {
 	.release = i915_gem_object_release_phys,
 };
 
-static int
-drop_pages(struct drm_i915_gem_object *obj)
+int
+i915_gem_object_unbind(struct drm_i915_gem_object *obj)
 {
-	struct i915_vma *vma, *next;
+	struct i915_vma *vma;
+	LIST_HEAD(still_in_list);
 	int ret;
 
-	drm_gem_object_reference(&obj->base);
-	list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link)
-		if (i915_vma_unbind(vma))
+	/* The vma will only be freed if it is marked as closed, and if we wait
+	 * upon rendering to the vma, we may unbind anything in the list.
+	 */
+	while ((vma = list_first_entry_or_null(&obj->vma_list,
+					       struct i915_vma,
+					       obj_link))) {
+		list_move_tail(&vma->obj_link, &still_in_list);
+		ret = i915_vma_unbind(vma);
+		if (ret)
 			break;
-
-	ret = i915_gem_object_put_pages(obj);
-	drm_gem_object_unreference(&obj->base);
+	}
+	list_splice(&still_in_list, &obj->vma_list);
 
 	return ret;
 }
 
+/**
+ * Ensures that all rendering to the object has completed and the object is
+ * safe to unbind from the GTT or access from the CPU.
+ * @obj: i915 gem object
+ * @readonly: waiting for just read access or read-write access
+ */
+int
+i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
+			       bool readonly)
+{
+	struct reservation_object *resv;
+	struct i915_gem_active *active;
+	unsigned long active_mask;
+	int idx;
+
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
+
+	if (!readonly) {
+		active = obj->last_read;
+		active_mask = i915_gem_object_get_active(obj);
+	} else {
+		active_mask = 1;
+		active = &obj->last_write;
+	}
+
+	for_each_active(active_mask, idx) {
+		int ret;
+
+		ret = i915_gem_active_wait(&active[idx],
+					   &obj->base.dev->struct_mutex);
+		if (ret)
+			return ret;
+	}
+
+	resv = i915_gem_object_get_dmabuf_resv(obj);
+	if (resv) {
+		long err;
+
+		err = reservation_object_wait_timeout_rcu(resv, !readonly, true,
+							  MAX_SCHEDULE_TIMEOUT);
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
+
+/* A nonblocking variant of the above wait. Must be called prior to
+ * acquiring the mutex for the object, as the object state may change
+ * during this call. A reference must be held by the caller for the object.
+ */
+static __must_check int
+__unsafe_wait_rendering(struct drm_i915_gem_object *obj,
+			struct intel_rps_client *rps,
+			bool readonly)
+{
+	struct i915_gem_active *active;
+	unsigned long active_mask;
+	int idx;
+
+	active_mask = __I915_BO_ACTIVE(obj);
+	if (!active_mask)
+		return 0;
+
+	if (!readonly) {
+		active = obj->last_read;
+	} else {
+		active_mask = 1;
+		active = &obj->last_write;
+	}
+
+	for_each_active(active_mask, idx) {
+		int ret;
+
+		ret = i915_gem_active_wait_unlocked(&active[idx],
+						    true, NULL, rps);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static struct intel_rps_client *to_rps_client(struct drm_file *file)
+{
+	struct drm_i915_file_private *fpriv = file->driver_priv;
+
+	return &fpriv->rps;
+}
+
 int
 i915_gem_object_attach_phys(struct drm_i915_gem_object *obj,
 			    int align)
@@ -318,7 +412,11 @@ i915_gem_object_attach_phys(struct drm_i915_gem_object *obj,
 	if (obj->base.filp == NULL)
 		return -EINVAL;
 
-	ret = drop_pages(obj);
+	ret = i915_gem_object_unbind(obj);
+	if (ret)
+		return ret;
+
+	ret = i915_gem_object_put_pages(obj);
 	if (ret)
 		return ret;
 
@@ -408,7 +506,7 @@ i915_gem_create(struct drm_file *file,
 
 	ret = drm_gem_handle_create(file, &obj->base, &handle);
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(&obj->base);
+	i915_gem_object_put_unlocked(obj);
 	if (ret)
 		return ret;
 
@@ -511,6 +609,10 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
 	if (WARN_ON(!i915_gem_object_has_struct_page(obj)))
 		return -EINVAL;
 
+	ret = i915_gem_object_wait_rendering(obj, true);
+	if (ret)
+		return ret;
+
 	if (!(obj->base.read_domains & I915_GEM_DOMAIN_CPU)) {
 		/* If we're not in the cpu read domain, set ourself into the gtt
 		 * read domain and manually flush cachelines (if required). This
@@ -518,9 +620,6 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
 		 * anyway again before the next pread happens. */
 		*needs_clflush = !cpu_cache_is_coherent(obj->base.dev,
 							obj->cache_level);
-		ret = i915_gem_object_wait_rendering(obj, true);
-		if (ret)
-			return ret;
 	}
 
 	ret = i915_gem_object_get_pages(obj);
@@ -644,7 +743,7 @@ i915_gem_gtt_pread(struct drm_device *dev,
 	uint64_t offset;
 	int ret;
 
-	ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_MAPPABLE);
+	ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 0, PIN_MAPPABLE);
 	if (ret) {
 		ret = insert_mappable_node(dev_priv, &node, PAGE_SIZE);
 		if (ret)
@@ -857,36 +956,44 @@ i915_gem_pread_ioctl(struct drm_device *dev, void *data,
 		       args->size))
 		return -EFAULT;
 
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
-
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL) {
-		ret = -ENOENT;
-		goto unlock;
-	}
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
+		return -ENOENT;
 
 	/* Bounds check source.  */
 	if (args->offset > obj->base.size ||
 	    args->size > obj->base.size - args->offset) {
 		ret = -EINVAL;
-		goto out;
+		goto err;
 	}
 
 	trace_i915_gem_object_pread(obj, args->offset, args->size);
 
+	ret = __unsafe_wait_rendering(obj, to_rps_client(file), true);
+	if (ret)
+		goto err;
+
+	ret = i915_mutex_lock_interruptible(dev);
+	if (ret)
+		goto err;
+
 	ret = i915_gem_shmem_pread(dev, obj, args, file);
 
 	/* pread for non shmem backed objects */
-	if (ret == -EFAULT || ret == -ENODEV)
+	if (ret == -EFAULT || ret == -ENODEV) {
+		intel_runtime_pm_get(to_i915(dev));
 		ret = i915_gem_gtt_pread(dev, obj, args->size,
 					args->offset, args->data_ptr);
+		intel_runtime_pm_put(to_i915(dev));
+	}
 
-out:
-	drm_gem_object_unreference(&obj->base);
-unlock:
+	i915_gem_object_put(obj);
 	mutex_unlock(&dev->struct_mutex);
+
+	return ret;
+
+err:
+	i915_gem_object_put_unlocked(obj);
 	return ret;
 }
 
@@ -916,7 +1023,7 @@ fast_user_write(struct io_mapping *mapping,
 /**
  * This is the fast pwrite path, where we copy the data directly from the
  * user into the GTT, uncached.
- * @dev: drm device pointer
+ * @i915: i915 device private data
  * @obj: i915 gem object
  * @args: pwrite arguments structure
  * @file: drm file pointer
@@ -935,10 +1042,11 @@ i915_gem_gtt_pwrite_fast(struct drm_i915_private *i915,
 	int ret;
 	bool hit_slow_path = false;
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		return -EFAULT;
 
-	ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_MAPPABLE | PIN_NONBLOCK);
+	ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 0,
+				       PIN_MAPPABLE | PIN_NONBLOCK);
 	if (ret) {
 		ret = insert_mappable_node(i915, &node, PAGE_SIZE);
 		if (ret)
@@ -1132,15 +1240,16 @@ i915_gem_shmem_pwrite(struct drm_device *dev,
 
 	obj_do_bit17_swizzling = i915_gem_object_needs_bit17_swizzle(obj);
 
+	ret = i915_gem_object_wait_rendering(obj, false);
+	if (ret)
+		return ret;
+
 	if (obj->base.write_domain != I915_GEM_DOMAIN_CPU) {
 		/* If we're not in the cpu write domain, set ourself into the gtt
 		 * write domain and manually flush cachelines (if required). This
 		 * optimizes for the case when the gpu will use the data
 		 * right away and we therefore have to clflush anyway. */
 		needs_clflush_after = cpu_write_needs_clflush(obj);
-		ret = i915_gem_object_wait_rendering(obj, false);
-		if (ret)
-			return ret;
 	}
 	/* Same trick applies to invalidate partially written cachelines read
 	 * before writing. */
@@ -1270,27 +1379,29 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
 			return -EFAULT;
 	}
 
-	intel_runtime_pm_get(dev_priv);
-
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		goto put_rpm;
-
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL) {
-		ret = -ENOENT;
-		goto unlock;
-	}
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
+		return -ENOENT;
 
 	/* Bounds check destination. */
 	if (args->offset > obj->base.size ||
 	    args->size > obj->base.size - args->offset) {
 		ret = -EINVAL;
-		goto out;
+		goto err;
 	}
 
 	trace_i915_gem_object_pwrite(obj, args->offset, args->size);
 
+	ret = __unsafe_wait_rendering(obj, to_rps_client(file), false);
+	if (ret)
+		goto err;
+
+	intel_runtime_pm_get(dev_priv);
+
+	ret = i915_mutex_lock_interruptible(dev);
+	if (ret)
+		goto err_rpm;
+
 	ret = -EFAULT;
 	/* We can only do the GTT pwrite on untiled buffers, as otherwise
 	 * it would end up going through the fenced access, and we'll get
@@ -1306,7 +1417,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
 		 * textures). Fallback to the shmem path in that case. */
 	}
 
-	if (ret == -EFAULT) {
+	if (ret == -EFAULT || ret == -ENOSPC) {
 		if (obj->phys_handle)
 			ret = i915_gem_phys_pwrite(obj, args, file);
 		else if (i915_gem_object_has_struct_page(obj))
@@ -1315,647 +1426,164 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
 			ret = -ENODEV;
 	}
 
-out:
-	drm_gem_object_unreference(&obj->base);
-unlock:
+	i915_gem_object_put(obj);
 	mutex_unlock(&dev->struct_mutex);
-put_rpm:
 	intel_runtime_pm_put(dev_priv);
 
 	return ret;
-}
-
-static int
-i915_gem_check_wedge(unsigned reset_counter, bool interruptible)
-{
-	if (__i915_terminally_wedged(reset_counter))
-		return -EIO;
-
-	if (__i915_reset_in_progress(reset_counter)) {
-		/* Non-interruptible callers can't handle -EAGAIN, hence return
-		 * -EIO unconditionally for these. */
-		if (!interruptible)
-			return -EIO;
 
-		return -EAGAIN;
-	}
-
-	return 0;
+err_rpm:
+	intel_runtime_pm_put(dev_priv);
+err:
+	i915_gem_object_put_unlocked(obj);
+	return ret;
 }
 
-static unsigned long local_clock_us(unsigned *cpu)
+static enum fb_op_origin
+write_origin(struct drm_i915_gem_object *obj, unsigned domain)
 {
-	unsigned long t;
-
-	/* Cheaply and approximately convert from nanoseconds to microseconds.
-	 * The result and subsequent calculations are also defined in the same
-	 * approximate microseconds units. The principal source of timing
-	 * error here is from the simple truncation.
-	 *
-	 * Note that local_clock() is only defined wrt to the current CPU;
-	 * the comparisons are no longer valid if we switch CPUs. Instead of
-	 * blocking preemption for the entire busywait, we can detect the CPU
-	 * switch and use that as indicator of system load and a reason to
-	 * stop busywaiting, see busywait_stop().
-	 */
-	*cpu = get_cpu();
-	t = local_clock() >> 10;
-	put_cpu();
-
-	return t;
+	return domain == I915_GEM_DOMAIN_GTT && !obj->has_wc_mmap ?
+	       ORIGIN_GTT : ORIGIN_CPU;
 }
 
-static bool busywait_stop(unsigned long timeout, unsigned cpu)
+/**
+ * Called when user space prepares to use an object with the CPU, either
+ * through the mmap ioctl's mapping or a GTT mapping.
+ * @dev: drm device
+ * @data: ioctl data blob
+ * @file: drm file
+ */
+int
+i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
+			  struct drm_file *file)
 {
-	unsigned this_cpu;
+	struct drm_i915_gem_set_domain *args = data;
+	struct drm_i915_gem_object *obj;
+	uint32_t read_domains = args->read_domains;
+	uint32_t write_domain = args->write_domain;
+	int ret;
 
-	if (time_after(local_clock_us(&this_cpu), timeout))
-		return true;
+	/* Only handle setting domains to types used by the CPU. */
+	if ((write_domain | read_domains) & I915_GEM_GPU_DOMAINS)
+		return -EINVAL;
 
-	return this_cpu != cpu;
-}
+	/* Having something in the write domain implies it's in the read
+	 * domain, and only that read domain.  Enforce that in the request.
+	 */
+	if (write_domain != 0 && read_domains != write_domain)
+		return -EINVAL;
 
-bool __i915_spin_request(const struct drm_i915_gem_request *req,
-			 int state, unsigned long timeout_us)
-{
-	unsigned cpu;
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
+		return -ENOENT;
 
-	/* When waiting for high frequency requests, e.g. during synchronous
-	 * rendering split between the CPU and GPU, the finite amount of time
-	 * required to set up the irq and wait upon it limits the response
-	 * rate. By busywaiting on the request completion for a short while we
-	 * can service the high frequency waits as quick as possible. However,
-	 * if it is a slow request, we want to sleep as quickly as possible.
-	 * The tradeoff between waiting and sleeping is roughly the time it
-	 * takes to sleep on a request, on the order of a microsecond.
+	/* Try to flush the object off the GPU without holding the lock.
+	 * We will repeat the flush holding the lock in the normal manner
+	 * to catch cases where we are gazumped.
 	 */
+	ret = __unsafe_wait_rendering(obj, to_rps_client(file), !write_domain);
+	if (ret)
+		goto err;
 
-	timeout_us += local_clock_us(&cpu);
-	do {
-		if (i915_gem_request_completed(req))
-			return true;
+	ret = i915_mutex_lock_interruptible(dev);
+	if (ret)
+		goto err;
 
-		if (signal_pending_state(state, current))
-			break;
+	if (read_domains & I915_GEM_DOMAIN_GTT)
+		ret = i915_gem_object_set_to_gtt_domain(obj, write_domain != 0);
+	else
+		ret = i915_gem_object_set_to_cpu_domain(obj, write_domain != 0);
 
-		if (busywait_stop(timeout_us, cpu))
-			break;
+	if (write_domain != 0)
+		intel_fb_obj_invalidate(obj, write_origin(obj, write_domain));
 
-		cpu_relax_lowlatency();
-	} while (!need_resched());
+	i915_gem_object_put(obj);
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
 
-	return false;
+err:
+	i915_gem_object_put_unlocked(obj);
+	return ret;
 }
 
 /**
- * __i915_wait_request - wait until execution of request has finished
- * @req: duh!
- * @interruptible: do an interruptible wait (normally yes)
- * @timeout: in - how long to wait (NULL forever); out - how much time remaining
- * @rps: RPS client
- *
- * Note: It is of utmost importance that the passed in seqno and reset_counter
- * values have been read by the caller in an smp safe manner. Where read-side
- * locks are involved, it is sufficient to read the reset_counter before
- * unlocking the lock that protects the seqno. For lockless tricks, the
- * reset_counter _must_ be read before, and an appropriate smp_rmb must be
- * inserted.
- *
- * Returns 0 if the request was found within the alloted time. Else returns the
- * errno with remaining time filled in timeout argument.
+ * Called when user space has done writes to this buffer
+ * @dev: drm device
+ * @data: ioctl data blob
+ * @file: drm file
  */
-int __i915_wait_request(struct drm_i915_gem_request *req,
-			bool interruptible,
-			s64 *timeout,
-			struct intel_rps_client *rps)
-{
-	int state = interruptible ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE;
-	DEFINE_WAIT(reset);
-	struct intel_wait wait;
-	unsigned long timeout_remain;
-	s64 before = 0; /* Only to silence a compiler warning. */
-	int ret = 0;
+int
+i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
+			 struct drm_file *file)
+{
+	struct drm_i915_gem_sw_finish *args = data;
+	struct drm_i915_gem_object *obj;
+	int err = 0;
 
-	might_sleep();
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
+		return -ENOENT;
 
-	if (list_empty(&req->list))
-		return 0;
+	/* Pinned buffers may be scanout, so flush the cache */
+	if (READ_ONCE(obj->pin_display)) {
+		err = i915_mutex_lock_interruptible(dev);
+		if (!err) {
+			i915_gem_object_flush_cpu_write_domain(obj);
+			mutex_unlock(&dev->struct_mutex);
+		}
+	}
 
-	if (i915_gem_request_completed(req))
-		return 0;
+	i915_gem_object_put_unlocked(obj);
+	return err;
+}
 
-	timeout_remain = MAX_SCHEDULE_TIMEOUT;
-	if (timeout) {
-		if (WARN_ON(*timeout < 0))
-			return -EINVAL;
+/**
+ * i915_gem_mmap_ioctl - Maps the contents of an object, returning the address
+ *			 it is mapped to.
+ * @dev: drm device
+ * @data: ioctl data blob
+ * @file: drm file
+ *
+ * While the mapping holds a reference on the contents of the object, it doesn't
+ * imply a ref on the object itself.
+ *
+ * IMPORTANT:
+ *
+ * DRM driver writers who look a this function as an example for how to do GEM
+ * mmap support, please don't implement mmap support like here. The modern way
+ * to implement DRM mmap support is with an mmap offset ioctl (like
+ * i915_gem_mmap_gtt) and then using the mmap syscall on the DRM fd directly.
+ * That way debug tooling like valgrind will understand what's going on, hiding
+ * the mmap call in a driver private ioctl will break that. The i915 driver only
+ * does cpu mmaps this way because we didn't know better.
+ */
+int
+i915_gem_mmap_ioctl(struct drm_device *dev, void *data,
+		    struct drm_file *file)
+{
+	struct drm_i915_gem_mmap *args = data;
+	struct drm_i915_gem_object *obj;
+	unsigned long addr;
 
-		if (*timeout == 0)
-			return -ETIME;
+	if (args->flags & ~(I915_MMAP_WC))
+		return -EINVAL;
 
-		timeout_remain = nsecs_to_jiffies_timeout(*timeout);
+	if (args->flags & I915_MMAP_WC && !boot_cpu_has(X86_FEATURE_PAT))
+		return -ENODEV;
 
-		/*
-		 * Record current time in case interrupted by signal, or wedged.
-		 */
-		before = ktime_get_raw_ns();
-	}
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
+		return -ENOENT;
 
-	trace_i915_gem_request_wait_begin(req);
-
-	/* This client is about to stall waiting for the GPU. In many cases
-	 * this is undesirable and limits the throughput of the system, as
-	 * many clients cannot continue processing user input/output whilst
-	 * blocked. RPS autotuning may take tens of milliseconds to respond
-	 * to the GPU load and thus incurs additional latency for the client.
-	 * We can circumvent that by promoting the GPU frequency to maximum
-	 * before we wait. This makes the GPU throttle up much more quickly
-	 * (good for benchmarks and user experience, e.g. window animations),
-	 * but at a cost of spending more power processing the workload
-	 * (bad for battery). Not all clients even want their results
-	 * immediately and for them we should just let the GPU select its own
-	 * frequency to maximise efficiency. To prevent a single client from
-	 * forcing the clocks too high for the whole system, we only allow
-	 * each client to waitboost once in a busy period.
+	/* prime objects have no backing filp to GEM mmap
+	 * pages from.
 	 */
-	if (INTEL_INFO(req->i915)->gen >= 6)
-		gen6_rps_boost(req->i915, rps, req->emitted_jiffies);
-
-	/* Optimistic spin for the next ~jiffie before touching IRQs */
-	if (i915_spin_request(req, state, 5))
-		goto complete;
+	if (!obj->base.filp) {
+		i915_gem_object_put_unlocked(obj);
+		return -EINVAL;
+	}
 
-	set_current_state(state);
-	add_wait_queue(&req->i915->gpu_error.wait_queue, &reset);
-
-	intel_wait_init(&wait, req->seqno);
-	if (intel_engine_add_wait(req->engine, &wait))
-		/* In order to check that we haven't missed the interrupt
-		 * as we enabled it, we need to kick ourselves to do a
-		 * coherent check on the seqno before we sleep.
-		 */
-		goto wakeup;
-
-	for (;;) {
-		if (signal_pending_state(state, current)) {
-			ret = -ERESTARTSYS;
-			break;
-		}
-
-		timeout_remain = io_schedule_timeout(timeout_remain);
-		if (timeout_remain == 0) {
-			ret = -ETIME;
-			break;
-		}
-
-		if (intel_wait_complete(&wait))
-			break;
-
-		set_current_state(state);
-
-wakeup:
-		/* Carefully check if the request is complete, giving time
-		 * for the seqno to be visible following the interrupt.
-		 * We also have to check in case we are kicked by the GPU
-		 * reset in order to drop the struct_mutex.
-		 */
-		if (__i915_request_irq_complete(req))
-			break;
-
-		/* Only spin if we know the GPU is processing this request */
-		if (i915_spin_request(req, state, 2))
-			break;
-	}
-	remove_wait_queue(&req->i915->gpu_error.wait_queue, &reset);
-
-	intel_engine_remove_wait(req->engine, &wait);
-	__set_current_state(TASK_RUNNING);
-complete:
-	trace_i915_gem_request_wait_end(req);
-
-	if (timeout) {
-		s64 tres = *timeout - (ktime_get_raw_ns() - before);
-
-		*timeout = tres < 0 ? 0 : tres;
-
-		/*
-		 * Apparently ktime isn't accurate enough and occasionally has a
-		 * bit of mismatch in the jiffies<->nsecs<->ktime loop. So patch
-		 * things up to make the test happy. We allow up to 1 jiffy.
-		 *
-		 * This is a regrssion from the timespec->ktime conversion.
-		 */
-		if (ret == -ETIME && *timeout < jiffies_to_usecs(1)*1000)
-			*timeout = 0;
-	}
-
-	if (rps && req->seqno == req->engine->last_submitted_seqno) {
-		/* The GPU is now idle and this client has stalled.
-		 * Since no other client has submitted a request in the
-		 * meantime, assume that this client is the only one
-		 * supplying work to the GPU but is unable to keep that
-		 * work supplied because it is waiting. Since the GPU is
-		 * then never kept fully busy, RPS autoclocking will
-		 * keep the clocks relatively low, causing further delays.
-		 * Compensate by giving the synchronous client credit for
-		 * a waitboost next time.
-		 */
-		spin_lock(&req->i915->rps.client_lock);
-		list_del_init(&rps->link);
-		spin_unlock(&req->i915->rps.client_lock);
-	}
-
-	return ret;
-}
-
-int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
-				   struct drm_file *file)
-{
-	struct drm_i915_file_private *file_priv;
-
-	WARN_ON(!req || !file || req->file_priv);
-
-	if (!req || !file)
-		return -EINVAL;
-
-	if (req->file_priv)
-		return -EINVAL;
-
-	file_priv = file->driver_priv;
-
-	spin_lock(&file_priv->mm.lock);
-	req->file_priv = file_priv;
-	list_add_tail(&req->client_list, &file_priv->mm.request_list);
-	spin_unlock(&file_priv->mm.lock);
-
-	req->pid = get_pid(task_pid(current));
-
-	return 0;
-}
-
-static inline void
-i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
-{
-	struct drm_i915_file_private *file_priv = request->file_priv;
-
-	if (!file_priv)
-		return;
-
-	spin_lock(&file_priv->mm.lock);
-	list_del(&request->client_list);
-	request->file_priv = NULL;
-	spin_unlock(&file_priv->mm.lock);
-
-	put_pid(request->pid);
-	request->pid = NULL;
-}
-
-static void i915_gem_request_retire(struct drm_i915_gem_request *request)
-{
-	trace_i915_gem_request_retire(request);
-
-	/* We know the GPU must have read the request to have
-	 * sent us the seqno + interrupt, so use the position
-	 * of tail of the request to update the last known position
-	 * of the GPU head.
-	 *
-	 * Note this requires that we are always called in request
-	 * completion order.
-	 */
-	request->ringbuf->last_retired_head = request->postfix;
-
-	list_del_init(&request->list);
-	i915_gem_request_remove_from_client(request);
-
-	if (request->previous_context) {
-		if (i915.enable_execlists)
-			intel_lr_context_unpin(request->previous_context,
-					       request->engine);
-	}
-
-	i915_gem_context_unreference(request->ctx);
-	i915_gem_request_unreference(request);
-}
-
-static void
-__i915_gem_request_retire__upto(struct drm_i915_gem_request *req)
-{
-	struct intel_engine_cs *engine = req->engine;
-	struct drm_i915_gem_request *tmp;
-
-	lockdep_assert_held(&engine->i915->drm.struct_mutex);
-
-	if (list_empty(&req->list))
-		return;
-
-	do {
-		tmp = list_first_entry(&engine->request_list,
-				       typeof(*tmp), list);
-
-		i915_gem_request_retire(tmp);
-	} while (tmp != req);
-
-	WARN_ON(i915_verify_lists(engine->dev));
-}
-
-/**
- * Waits for a request to be signaled, and cleans up the
- * request and object lists appropriately for that event.
- * @req: request to wait on
- */
-int
-i915_wait_request(struct drm_i915_gem_request *req)
-{
-	struct drm_i915_private *dev_priv = req->i915;
-	bool interruptible;
-	int ret;
-
-	interruptible = dev_priv->mm.interruptible;
-
-	BUG_ON(!mutex_is_locked(&dev_priv->drm.struct_mutex));
-
-	ret = __i915_wait_request(req, interruptible, NULL, NULL);
-	if (ret)
-		return ret;
-
-	/* If the GPU hung, we want to keep the requests to find the guilty. */
-	if (!i915_reset_in_progress(&dev_priv->gpu_error))
-		__i915_gem_request_retire__upto(req);
-
-	return 0;
-}
-
-/**
- * Ensures that all rendering to the object has completed and the object is
- * safe to unbind from the GTT or access from the CPU.
- * @obj: i915 gem object
- * @readonly: waiting for read access or write
- */
-int
-i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
-			       bool readonly)
-{
-	int ret, i;
-
-	if (!obj->active)
-		return 0;
-
-	if (readonly) {
-		if (obj->last_write_req != NULL) {
-			ret = i915_wait_request(obj->last_write_req);
-			if (ret)
-				return ret;
-
-			i = obj->last_write_req->engine->id;
-			if (obj->last_read_req[i] == obj->last_write_req)
-				i915_gem_object_retire__read(obj, i);
-			else
-				i915_gem_object_retire__write(obj);
-		}
-	} else {
-		for (i = 0; i < I915_NUM_ENGINES; i++) {
-			if (obj->last_read_req[i] == NULL)
-				continue;
-
-			ret = i915_wait_request(obj->last_read_req[i]);
-			if (ret)
-				return ret;
-
-			i915_gem_object_retire__read(obj, i);
-		}
-		GEM_BUG_ON(obj->active);
-	}
-
-	return 0;
-}
-
-static void
-i915_gem_object_retire_request(struct drm_i915_gem_object *obj,
-			       struct drm_i915_gem_request *req)
-{
-	int ring = req->engine->id;
-
-	if (obj->last_read_req[ring] == req)
-		i915_gem_object_retire__read(obj, ring);
-	else if (obj->last_write_req == req)
-		i915_gem_object_retire__write(obj);
-
-	if (!i915_reset_in_progress(&req->i915->gpu_error))
-		__i915_gem_request_retire__upto(req);
-}
-
-/* A nonblocking variant of the above wait. This is a highly dangerous routine
- * as the object state may change during this call.
- */
-static __must_check int
-i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
-					    struct intel_rps_client *rps,
-					    bool readonly)
-{
-	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_i915_gem_request *requests[I915_NUM_ENGINES];
-	int ret, i, n = 0;
-
-	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-	BUG_ON(!dev_priv->mm.interruptible);
-
-	if (!obj->active)
-		return 0;
-
-	if (readonly) {
-		struct drm_i915_gem_request *req;
-
-		req = obj->last_write_req;
-		if (req == NULL)
-			return 0;
-
-		requests[n++] = i915_gem_request_reference(req);
-	} else {
-		for (i = 0; i < I915_NUM_ENGINES; i++) {
-			struct drm_i915_gem_request *req;
-
-			req = obj->last_read_req[i];
-			if (req == NULL)
-				continue;
-
-			requests[n++] = i915_gem_request_reference(req);
-		}
-	}
-
-	mutex_unlock(&dev->struct_mutex);
-	ret = 0;
-	for (i = 0; ret == 0 && i < n; i++)
-		ret = __i915_wait_request(requests[i], true, NULL, rps);
-	mutex_lock(&dev->struct_mutex);
-
-	for (i = 0; i < n; i++) {
-		if (ret == 0)
-			i915_gem_object_retire_request(obj, requests[i]);
-		i915_gem_request_unreference(requests[i]);
-	}
-
-	return ret;
-}
-
-static struct intel_rps_client *to_rps_client(struct drm_file *file)
-{
-	struct drm_i915_file_private *fpriv = file->driver_priv;
-	return &fpriv->rps;
-}
-
-static enum fb_op_origin
-write_origin(struct drm_i915_gem_object *obj, unsigned domain)
-{
-	return domain == I915_GEM_DOMAIN_GTT && !obj->has_wc_mmap ?
-	       ORIGIN_GTT : ORIGIN_CPU;
-}
-
-/**
- * Called when user space prepares to use an object with the CPU, either
- * through the mmap ioctl's mapping or a GTT mapping.
- * @dev: drm device
- * @data: ioctl data blob
- * @file: drm file
- */
-int
-i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
-			  struct drm_file *file)
-{
-	struct drm_i915_gem_set_domain *args = data;
-	struct drm_i915_gem_object *obj;
-	uint32_t read_domains = args->read_domains;
-	uint32_t write_domain = args->write_domain;
-	int ret;
-
-	/* Only handle setting domains to types used by the CPU. */
-	if (write_domain & I915_GEM_GPU_DOMAINS)
-		return -EINVAL;
-
-	if (read_domains & I915_GEM_GPU_DOMAINS)
-		return -EINVAL;
-
-	/* Having something in the write domain implies it's in the read
-	 * domain, and only that read domain.  Enforce that in the request.
-	 */
-	if (write_domain != 0 && read_domains != write_domain)
-		return -EINVAL;
-
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
-
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL) {
-		ret = -ENOENT;
-		goto unlock;
-	}
-
-	/* Try to flush the object off the GPU without holding the lock.
-	 * We will repeat the flush holding the lock in the normal manner
-	 * to catch cases where we are gazumped.
-	 */
-	ret = i915_gem_object_wait_rendering__nonblocking(obj,
-							  to_rps_client(file),
-							  !write_domain);
-	if (ret)
-		goto unref;
-
-	if (read_domains & I915_GEM_DOMAIN_GTT)
-		ret = i915_gem_object_set_to_gtt_domain(obj, write_domain != 0);
-	else
-		ret = i915_gem_object_set_to_cpu_domain(obj, write_domain != 0);
-
-	if (write_domain != 0)
-		intel_fb_obj_invalidate(obj, write_origin(obj, write_domain));
-
-unref:
-	drm_gem_object_unreference(&obj->base);
-unlock:
-	mutex_unlock(&dev->struct_mutex);
-	return ret;
-}
-
-/**
- * Called when user space has done writes to this buffer
- * @dev: drm device
- * @data: ioctl data blob
- * @file: drm file
- */
-int
-i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
-			 struct drm_file *file)
-{
-	struct drm_i915_gem_sw_finish *args = data;
-	struct drm_i915_gem_object *obj;
-	int ret = 0;
-
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
-
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL) {
-		ret = -ENOENT;
-		goto unlock;
-	}
-
-	/* Pinned buffers may be scanout, so flush the cache */
-	if (obj->pin_display)
-		i915_gem_object_flush_cpu_write_domain(obj);
-
-	drm_gem_object_unreference(&obj->base);
-unlock:
-	mutex_unlock(&dev->struct_mutex);
-	return ret;
-}
-
-/**
- * i915_gem_mmap_ioctl - Maps the contents of an object, returning the address
- *			 it is mapped to.
- * @dev: drm device
- * @data: ioctl data blob
- * @file: drm file
- *
- * While the mapping holds a reference on the contents of the object, it doesn't
- * imply a ref on the object itself.
- *
- * IMPORTANT:
- *
- * DRM driver writers who look a this function as an example for how to do GEM
- * mmap support, please don't implement mmap support like here. The modern way
- * to implement DRM mmap support is with an mmap offset ioctl (like
- * i915_gem_mmap_gtt) and then using the mmap syscall on the DRM fd directly.
- * That way debug tooling like valgrind will understand what's going on, hiding
- * the mmap call in a driver private ioctl will break that. The i915 driver only
- * does cpu mmaps this way because we didn't know better.
- */
-int
-i915_gem_mmap_ioctl(struct drm_device *dev, void *data,
-		    struct drm_file *file)
-{
-	struct drm_i915_gem_mmap *args = data;
-	struct drm_gem_object *obj;
-	unsigned long addr;
-
-	if (args->flags & ~(I915_MMAP_WC))
-		return -EINVAL;
-
-	if (args->flags & I915_MMAP_WC && !boot_cpu_has(X86_FEATURE_PAT))
-		return -ENODEV;
-
-	obj = drm_gem_object_lookup(file, args->handle);
-	if (obj == NULL)
-		return -ENOENT;
-
-	/* prime objects have no backing filp to GEM mmap
-	 * pages from.
-	 */
-	if (!obj->filp) {
-		drm_gem_object_unreference_unlocked(obj);
-		return -EINVAL;
-	}
-
-	addr = vm_mmap(obj->filp, 0, args->size,
+	addr = vm_mmap(obj->base.filp, 0, args->size,
 		       PROT_READ | PROT_WRITE, MAP_SHARED,
 		       args->offset);
 	if (args->flags & I915_MMAP_WC) {
@@ -1963,7 +1591,7 @@ i915_gem_mmap_ioctl(struct drm_device *dev, void *data,
 		struct vm_area_struct *vma;
 
 		if (down_write_killable(&mm->mmap_sem)) {
-			drm_gem_object_unreference_unlocked(obj);
+			i915_gem_object_put_unlocked(obj);
 			return -EINTR;
 		}
 		vma = find_vma(mm, addr);
@@ -1975,9 +1603,9 @@ i915_gem_mmap_ioctl(struct drm_device *dev, void *data,
 		up_write(&mm->mmap_sem);
 
 		/* This may race, but that's ok, it only gets set */
-		WRITE_ONCE(to_intel_bo(obj)->has_wc_mmap, true);
+		WRITE_ONCE(obj->has_wc_mmap, true);
 	}
-	drm_gem_object_unreference_unlocked(obj);
+	i915_gem_object_put_unlocked(obj);
 	if (IS_ERR((void *)addr))
 		return addr;
 
@@ -2009,41 +1637,41 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_ggtt_view view = i915_ggtt_view_normal;
+	bool write = !!(vmf->flags & FAULT_FLAG_WRITE);
 	pgoff_t page_offset;
 	unsigned long pfn;
-	int ret = 0;
-	bool write = !!(vmf->flags & FAULT_FLAG_WRITE);
-
-	intel_runtime_pm_get(dev_priv);
+	int ret;
 
 	/* We don't use vmf->pgoff since that has the fake offset */
 	page_offset = ((unsigned long)vmf->virtual_address - vma->vm_start) >>
 		PAGE_SHIFT;
 
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		goto out;
-
 	trace_i915_gem_object_fault(obj, page_offset, true, write);
 
 	/* Try to flush the object off the GPU first without holding the lock.
-	 * Upon reacquiring the lock, we will perform our sanity checks and then
+	 * Upon acquiring the lock, we will perform our sanity checks and then
 	 * repeat the flush holding the lock in the normal manner to catch cases
 	 * where we are gazumped.
 	 */
-	ret = i915_gem_object_wait_rendering__nonblocking(obj, NULL, !write);
+	ret = __unsafe_wait_rendering(obj, NULL, !write);
 	if (ret)
-		goto unlock;
+		goto err;
+
+	intel_runtime_pm_get(dev_priv);
+
+	ret = i915_mutex_lock_interruptible(dev);
+	if (ret)
+		goto err_rpm;
 
 	/* Access to snoopable pages through the GTT is incoherent. */
 	if (obj->cache_level != I915_CACHE_NONE && !HAS_LLC(dev)) {
 		ret = -EFAULT;
-		goto unlock;
+		goto err_unlock;
 	}
 
 	/* Use a partial view if the object is bigger than the aperture. */
 	if (obj->base.size >= ggtt->mappable_end &&
-	    obj->tiling_mode == I915_TILING_NONE) {
+	    !i915_gem_object_is_tiled(obj)) {
 		static const unsigned int chunk_size = 256; // 1 MiB
 
 		memset(&view, 0, sizeof(view));
@@ -2057,17 +1685,17 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	}
 
 	/* Now pin it into the GTT if needed */
-	ret = i915_gem_object_ggtt_pin(obj, &view, 0, PIN_MAPPABLE);
+	ret = i915_gem_object_ggtt_pin(obj, &view, 0, 0, PIN_MAPPABLE);
 	if (ret)
-		goto unlock;
+		goto err_unlock;
 
 	ret = i915_gem_object_set_to_gtt_domain(obj, write);
 	if (ret)
-		goto unpin;
+		goto err_unpin;
 
 	ret = i915_gem_object_get_fence(obj);
 	if (ret)
-		goto unpin;
+		goto err_unpin;
 
 	/* Finally, remap it using the new GTT offset */
 	pfn = ggtt->mappable_base +
@@ -2112,11 +1740,13 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 					    (unsigned long)vmf->virtual_address,
 					    pfn + page_offset);
 	}
-unpin:
+err_unpin:
 	i915_gem_object_ggtt_unpin_view(obj, &view);
-unlock:
+err_unlock:
 	mutex_unlock(&dev->struct_mutex);
-out:
+err_rpm:
+	intel_runtime_pm_put(dev_priv);
+err:
 	switch (ret) {
 	case -EIO:
 		/*
@@ -2157,8 +1787,6 @@ out:
 		ret = VM_FAULT_SIGBUS;
 		break;
 	}
-
-	intel_runtime_pm_put(dev_priv);
 	return ret;
 }
 
@@ -2212,46 +1840,58 @@ i915_gem_release_all_mmaps(struct drm_i915_private *dev_priv)
 		i915_gem_release_mmap(obj);
 }
 
-uint32_t
-i915_gem_get_gtt_size(struct drm_device *dev, uint32_t size, int tiling_mode)
+/**
+ * i915_gem_get_ggtt_size - return required global GTT size for an object
+ * @dev_priv: i915 device
+ * @size: object size
+ * @tiling_mode: tiling mode
+ *
+ * Return the required global GTT size for an object, taking into account
+ * potential fence register mapping.
+ */
+u64 i915_gem_get_ggtt_size(struct drm_i915_private *dev_priv,
+			   u64 size, int tiling_mode)
 {
-	uint32_t gtt_size;
+	u64 ggtt_size;
+
+	GEM_BUG_ON(size == 0);
 
-	if (INTEL_INFO(dev)->gen >= 4 ||
+	if (INTEL_GEN(dev_priv) >= 4 ||
 	    tiling_mode == I915_TILING_NONE)
 		return size;
 
 	/* Previous chips need a power-of-two fence region when tiling */
-	if (IS_GEN3(dev))
-		gtt_size = 1024*1024;
+	if (IS_GEN3(dev_priv))
+		ggtt_size = 1024*1024;
 	else
-		gtt_size = 512*1024;
+		ggtt_size = 512*1024;
 
-	while (gtt_size < size)
-		gtt_size <<= 1;
+	while (ggtt_size < size)
+		ggtt_size <<= 1;
 
-	return gtt_size;
+	return ggtt_size;
 }
 
 /**
- * i915_gem_get_gtt_alignment - return required GTT alignment for an object
- * @dev: drm device
+ * i915_gem_get_ggtt_alignment - return required global GTT alignment
+ * @dev_priv: i915 device
  * @size: object size
  * @tiling_mode: tiling mode
- * @fenced: is fenced alignemned required or not
+ * @fenced: is fenced alignment required or not
  *
- * Return the required GTT alignment for an object, taking into account
+ * Return the required global GTT alignment for an object, taking into account
  * potential fence register mapping.
  */
-uint32_t
-i915_gem_get_gtt_alignment(struct drm_device *dev, uint32_t size,
-			   int tiling_mode, bool fenced)
+u64 i915_gem_get_ggtt_alignment(struct drm_i915_private *dev_priv, u64 size,
+				int tiling_mode, bool fenced)
 {
+	GEM_BUG_ON(size == 0);
+
 	/*
 	 * Minimum alignment is 4k (GTT page size), but might be greater
 	 * if a fence register is needed for the object.
 	 */
-	if (INTEL_INFO(dev)->gen >= 4 || (!fenced && IS_G33(dev)) ||
+	if (INTEL_GEN(dev_priv) >= 4 || (!fenced && IS_G33(dev_priv)) ||
 	    tiling_mode == I915_TILING_NONE)
 		return 4096;
 
@@ -2259,42 +1899,34 @@ i915_gem_get_gtt_alignment(struct drm_device *dev, uint32_t size,
 	 * Previous chips need to be aligned to the size of the smallest
 	 * fence register that can contain the object.
 	 */
-	return i915_gem_get_gtt_size(dev, size, tiling_mode);
+	return i915_gem_get_ggtt_size(dev_priv, size, tiling_mode);
 }
 
 static int i915_gem_object_create_mmap_offset(struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
-	int ret;
-
-	dev_priv->mm.shrinker_no_lock_stealing = true;
+	int err;
 
-	ret = drm_gem_create_mmap_offset(&obj->base);
-	if (ret != -ENOSPC)
-		goto out;
+	err = drm_gem_create_mmap_offset(&obj->base);
+	if (!err)
+		return 0;
 
-	/* Badly fragmented mmap space? The only way we can recover
-	 * space is by destroying unwanted objects. We can't randomly release
-	 * mmap_offsets as userspace expects them to be persistent for the
-	 * lifetime of the objects. The closest we can is to release the
-	 * offsets on purgeable objects by truncating it and marking it purged,
-	 * which prevents userspace from ever using that object again.
+	/* We can idle the GPU locklessly to flush stale objects, but in order
+	 * to claim that space for ourselves, we need to take the big
+	 * struct_mutex to free the requests+objects and allocate our slot.
 	 */
-	i915_gem_shrink(dev_priv,
-			obj->base.size >> PAGE_SHIFT,
-			I915_SHRINK_BOUND |
-			I915_SHRINK_UNBOUND |
-			I915_SHRINK_PURGEABLE);
-	ret = drm_gem_create_mmap_offset(&obj->base);
-	if (ret != -ENOSPC)
-		goto out;
+	err = i915_gem_wait_for_idle(dev_priv, true);
+	if (err)
+		return err;
 
-	i915_gem_shrink_all(dev_priv);
-	ret = drm_gem_create_mmap_offset(&obj->base);
-out:
-	dev_priv->mm.shrinker_no_lock_stealing = false;
+	err = i915_mutex_lock_interruptible(&dev_priv->drm);
+	if (!err) {
+		i915_gem_retire_requests(dev_priv);
+		err = drm_gem_create_mmap_offset(&obj->base);
+		mutex_unlock(&dev_priv->drm.struct_mutex);
+	}
 
-	return ret;
+	return err;
 }
 
 static void i915_gem_object_free_mmap_offset(struct drm_i915_gem_object *obj)
@@ -2311,32 +1943,15 @@ i915_gem_mmap_gtt(struct drm_file *file,
 	struct drm_i915_gem_object *obj;
 	int ret;
 
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
-
-	obj = to_intel_bo(drm_gem_object_lookup(file, handle));
-	if (&obj->base == NULL) {
-		ret = -ENOENT;
-		goto unlock;
-	}
-
-	if (obj->madv != I915_MADV_WILLNEED) {
-		DRM_DEBUG("Attempting to mmap a purgeable buffer\n");
-		ret = -EFAULT;
-		goto out;
-	}
+	obj = i915_gem_object_lookup(file, handle);
+	if (!obj)
+		return -ENOENT;
 
 	ret = i915_gem_object_create_mmap_offset(obj);
-	if (ret)
-		goto out;
+	if (ret == 0)
+		*offset = drm_vma_node_offset_addr(&obj->base.vma_node);
 
-	*offset = drm_vma_node_offset_addr(&obj->base.vma_node);
-
-out:
-	drm_gem_object_unreference(&obj->base);
-unlock:
-	mutex_unlock(&dev->struct_mutex);
+	i915_gem_object_put_unlocked(obj);
 	return ret;
 }
 
@@ -2454,7 +2069,7 @@ i915_gem_object_put_pages(struct drm_i915_gem_object *obj)
 	if (obj->pages_pin_count)
 		return -EBUSY;
 
-	BUG_ON(i915_gem_obj_bound_any(obj));
+	GEM_BUG_ON(obj->bind_count);
 
 	/* ->put_pages might need to allocate memory for the bit17 swizzle
 	 * array, hence protect them from being reaped by removing them from gtt
@@ -2574,7 +2189,7 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	if (i915_gem_object_needs_bit17_swizzle(obj))
 		i915_gem_object_do_bit_17_swizzle(obj);
 
-	if (obj->tiling_mode != I915_TILING_NONE &&
+	if (i915_gem_object_is_tiled(obj) &&
 	    dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES)
 		i915_gem_object_pin_pages(obj);
 
@@ -2650,301 +2265,87 @@ static void *i915_gem_object_map(const struct drm_i915_gem_object *obj)
 	void *addr;
 
 	/* A single page can always be kmapped */
-	if (n_pages == 1)
-		return kmap(sg_page(sgt->sgl));
-
-	if (n_pages > ARRAY_SIZE(stack_pages)) {
-		/* Too big for stack -- allocate temporary array instead */
-		pages = drm_malloc_gfp(n_pages, sizeof(*pages), GFP_TEMPORARY);
-		if (!pages)
-			return NULL;
-	}
-
-	for_each_sgt_page(page, sgt_iter, sgt)
-		pages[i++] = page;
-
-	/* Check that we have the expected number of pages */
-	GEM_BUG_ON(i != n_pages);
-
-	addr = vmap(pages, n_pages, 0, PAGE_KERNEL);
-
-	if (pages != stack_pages)
-		drm_free_large(pages);
-
-	return addr;
-}
-
-/* get, pin, and map the pages of the object into kernel space */
-void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj)
-{
-	int ret;
-
-	lockdep_assert_held(&obj->base.dev->struct_mutex);
-
-	ret = i915_gem_object_get_pages(obj);
-	if (ret)
-		return ERR_PTR(ret);
-
-	i915_gem_object_pin_pages(obj);
-
-	if (!obj->mapping) {
-		obj->mapping = i915_gem_object_map(obj);
-		if (!obj->mapping) {
-			i915_gem_object_unpin_pages(obj);
-			return ERR_PTR(-ENOMEM);
-		}
-	}
-
-	return obj->mapping;
-}
-
-void i915_vma_move_to_active(struct i915_vma *vma,
-			     struct drm_i915_gem_request *req)
-{
-	struct drm_i915_gem_object *obj = vma->obj;
-	struct intel_engine_cs *engine;
-
-	engine = i915_gem_request_get_engine(req);
-
-	/* Add a reference if we're newly entering the active list. */
-	if (obj->active == 0)
-		drm_gem_object_reference(&obj->base);
-	obj->active |= intel_engine_flag(engine);
-
-	list_move_tail(&obj->engine_list[engine->id], &engine->active_list);
-	i915_gem_request_assign(&obj->last_read_req[engine->id], req);
-
-	list_move_tail(&vma->vm_link, &vma->vm->active_list);
-}
-
-static void
-i915_gem_object_retire__write(struct drm_i915_gem_object *obj)
-{
-	GEM_BUG_ON(obj->last_write_req == NULL);
-	GEM_BUG_ON(!(obj->active & intel_engine_flag(obj->last_write_req->engine)));
-
-	i915_gem_request_assign(&obj->last_write_req, NULL);
-	intel_fb_obj_flush(obj, true, ORIGIN_CS);
-}
-
-static void
-i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring)
-{
-	struct i915_vma *vma;
-
-	GEM_BUG_ON(obj->last_read_req[ring] == NULL);
-	GEM_BUG_ON(!(obj->active & (1 << ring)));
-
-	list_del_init(&obj->engine_list[ring]);
-	i915_gem_request_assign(&obj->last_read_req[ring], NULL);
-
-	if (obj->last_write_req && obj->last_write_req->engine->id == ring)
-		i915_gem_object_retire__write(obj);
-
-	obj->active &= ~(1 << ring);
-	if (obj->active)
-		return;
-
-	/* Bump our place on the bound list to keep it roughly in LRU order
-	 * so that we don't steal from recently used but inactive objects
-	 * (unless we are forced to ofc!)
-	 */
-	list_move_tail(&obj->global_list,
-		       &to_i915(obj->base.dev)->mm.bound_list);
-
-	list_for_each_entry(vma, &obj->vma_list, obj_link) {
-		if (!list_empty(&vma->vm_link))
-			list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
-	}
-
-	i915_gem_request_assign(&obj->last_fenced_req, NULL);
-	drm_gem_object_unreference(&obj->base);
-}
-
-static int
-i915_gem_init_seqno(struct drm_i915_private *dev_priv, u32 seqno)
-{
-	struct intel_engine_cs *engine;
-	int ret;
-
-	/* Carefully retire all requests without writing to the rings */
-	for_each_engine(engine, dev_priv) {
-		ret = intel_engine_idle(engine);
-		if (ret)
-			return ret;
-	}
-	i915_gem_retire_requests(dev_priv);
-
-	/* If the seqno wraps around, we need to clear the breadcrumb rbtree */
-	if (!i915_seqno_passed(seqno, dev_priv->next_seqno)) {
-		while (intel_kick_waiters(dev_priv) ||
-		       intel_kick_signalers(dev_priv))
-			yield();
-	}
-
-	/* Finally reset hw state */
-	for_each_engine(engine, dev_priv)
-		intel_ring_init_seqno(engine, seqno);
-
-	return 0;
-}
-
-int i915_gem_set_seqno(struct drm_device *dev, u32 seqno)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
-
-	if (seqno == 0)
-		return -EINVAL;
-
-	/* HWS page needs to be set less than what we
-	 * will inject to ring
-	 */
-	ret = i915_gem_init_seqno(dev_priv, seqno - 1);
-	if (ret)
-		return ret;
-
-	/* Carefully set the last_seqno value so that wrap
-	 * detection still works
-	 */
-	dev_priv->next_seqno = seqno;
-	dev_priv->last_seqno = seqno - 1;
-	if (dev_priv->last_seqno == 0)
-		dev_priv->last_seqno--;
-
-	return 0;
-}
-
-int
-i915_gem_get_seqno(struct drm_i915_private *dev_priv, u32 *seqno)
-{
-	/* reserve 0 for non-seqno */
-	if (dev_priv->next_seqno == 0) {
-		int ret = i915_gem_init_seqno(dev_priv, 0);
-		if (ret)
-			return ret;
+	if (n_pages == 1)
+		return kmap(sg_page(sgt->sgl));
 
-		dev_priv->next_seqno = 1;
+	if (n_pages > ARRAY_SIZE(stack_pages)) {
+		/* Too big for stack -- allocate temporary array instead */
+		pages = drm_malloc_gfp(n_pages, sizeof(*pages), GFP_TEMPORARY);
+		if (!pages)
+			return NULL;
 	}
 
-	*seqno = dev_priv->last_seqno = dev_priv->next_seqno++;
-	return 0;
-}
-
-static void i915_gem_mark_busy(const struct intel_engine_cs *engine)
-{
-	struct drm_i915_private *dev_priv = engine->i915;
+	for_each_sgt_page(page, sgt_iter, sgt)
+		pages[i++] = page;
 
-	dev_priv->gt.active_engines |= intel_engine_flag(engine);
-	if (dev_priv->gt.awake)
-		return;
+	/* Check that we have the expected number of pages */
+	GEM_BUG_ON(i != n_pages);
 
-	intel_runtime_pm_get_noresume(dev_priv);
-	dev_priv->gt.awake = true;
+	addr = vmap(pages, n_pages, 0, PAGE_KERNEL);
 
-	i915_update_gfx_val(dev_priv);
-	if (INTEL_GEN(dev_priv) >= 6)
-		gen6_rps_busy(dev_priv);
+	if (pages != stack_pages)
+		drm_free_large(pages);
 
-	queue_delayed_work(dev_priv->wq,
-			   &dev_priv->gt.retire_work,
-			   round_jiffies_up_relative(HZ));
+	return addr;
 }
 
-/*
- * NB: This function is not allowed to fail. Doing so would mean the the
- * request is not being tracked for completion but the work itself is
- * going to happen on the hardware. This would be a Bad Thing(tm).
- */
-void __i915_add_request(struct drm_i915_gem_request *request,
-			struct drm_i915_gem_object *obj,
-			bool flush_caches)
+/* get, pin, and map the pages of the object into kernel space */
+void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj)
 {
-	struct intel_engine_cs *engine;
-	struct intel_ringbuffer *ringbuf;
-	u32 request_start;
-	u32 reserved_tail;
 	int ret;
 
-	if (WARN_ON(request == NULL))
-		return;
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
 
-	engine = request->engine;
-	ringbuf = request->ringbuf;
+	ret = i915_gem_object_get_pages(obj);
+	if (ret)
+		return ERR_PTR(ret);
 
-	/*
-	 * To ensure that this call will not fail, space for its emissions
-	 * should already have been reserved in the ring buffer. Let the ring
-	 * know that it is time to use that space up.
-	 */
-	request_start = intel_ring_get_tail(ringbuf);
-	reserved_tail = request->reserved_space;
-	request->reserved_space = 0;
+	i915_gem_object_pin_pages(obj);
 
-	/*
-	 * Emit any outstanding flushes - execbuf can fail to emit the flush
-	 * after having emitted the batchbuffer command. Hence we need to fix
-	 * things up similar to emitting the lazy request. The difference here
-	 * is that the flush _must_ happen before the next request, no matter
-	 * what.
-	 */
-	if (flush_caches) {
-		if (i915.enable_execlists)
-			ret = logical_ring_flush_all_caches(request);
-		else
-			ret = intel_ring_flush_all_caches(request);
-		/* Not allowed to fail! */
-		WARN(ret, "*_ring_flush_all_caches failed: %d!\n", ret);
+	if (!obj->mapping) {
+		obj->mapping = i915_gem_object_map(obj);
+		if (!obj->mapping) {
+			i915_gem_object_unpin_pages(obj);
+			return ERR_PTR(-ENOMEM);
+		}
 	}
 
-	trace_i915_gem_request_add(request);
+	return obj->mapping;
+}
 
-	request->head = request_start;
+static void
+i915_gem_object_retire__write(struct i915_gem_active *active,
+			      struct drm_i915_gem_request *request)
+{
+	struct drm_i915_gem_object *obj =
+		container_of(active, struct drm_i915_gem_object, last_write);
 
-	/* Whilst this request exists, batch_obj will be on the
-	 * active_list, and so will hold the active reference. Only when this
-	 * request is retired will the the batch_obj be moved onto the
-	 * inactive_list and lose its active reference. Hence we do not need
-	 * to explicitly hold another reference here.
-	 */
-	request->batch_obj = obj;
+	intel_fb_obj_flush(obj, true, ORIGIN_CS);
+}
 
-	/* Seal the request and mark it as pending execution. Note that
-	 * we may inspect this state, without holding any locks, during
-	 * hangcheck. Hence we apply the barrier to ensure that we do not
-	 * see a more recent value in the hws than we are tracking.
-	 */
-	request->emitted_jiffies = jiffies;
-	request->previous_seqno = engine->last_submitted_seqno;
-	smp_store_mb(engine->last_submitted_seqno, request->seqno);
-	list_add_tail(&request->list, &engine->request_list);
-
-	/* Record the position of the start of the request so that
-	 * should we detect the updated seqno part-way through the
-	 * GPU processing the request, we never over-estimate the
-	 * position of the head.
-	 */
-	request->postfix = intel_ring_get_tail(ringbuf);
+static void
+i915_gem_object_retire__read(struct i915_gem_active *active,
+			     struct drm_i915_gem_request *request)
+{
+	int idx = request->engine->id;
+	struct drm_i915_gem_object *obj =
+		container_of(active, struct drm_i915_gem_object, last_read[idx]);
 
-	if (i915.enable_execlists)
-		ret = engine->emit_request(request);
-	else {
-		ret = engine->add_request(request);
+	GEM_BUG_ON(!i915_gem_object_has_active_engine(obj, idx));
 
-		request->tail = intel_ring_get_tail(ringbuf);
-	}
-	/* Not allowed to fail! */
-	WARN(ret, "emit|add_request failed: %d!\n", ret);
-	/* Sanity check that the reserved size was large enough. */
-	ret = intel_ring_get_tail(ringbuf) - request_start;
-	if (ret < 0)
-		ret += ringbuf->size;
-	WARN_ONCE(ret > reserved_tail,
-		  "Not enough space reserved (%d bytes) "
-		  "for adding the request (%d bytes)\n",
-		  reserved_tail, ret);
+	i915_gem_object_clear_active(obj, idx);
+	if (i915_gem_object_is_active(obj))
+		return;
+
+	/* Bump our place on the bound list to keep it roughly in LRU order
+	 * so that we don't steal from recently used but inactive objects
+	 * (unless we are forced to ofc!)
+	 */
+	if (obj->bind_count)
+		list_move_tail(&obj->global_list,
+			       &request->i915->mm.bound_list);
 
-	i915_gem_mark_busy(engine);
+	i915_gem_object_put(obj);
 }
 
 static bool i915_context_is_banned(const struct i915_gem_context *ctx)
@@ -2978,101 +2379,6 @@ static void i915_set_reset_status(struct i915_gem_context *ctx,
 	}
 }
 
-void i915_gem_request_free(struct kref *req_ref)
-{
-	struct drm_i915_gem_request *req = container_of(req_ref,
-						 typeof(*req), ref);
-	kmem_cache_free(req->i915->requests, req);
-}
-
-static inline int
-__i915_gem_request_alloc(struct intel_engine_cs *engine,
-			 struct i915_gem_context *ctx,
-			 struct drm_i915_gem_request **req_out)
-{
-	struct drm_i915_private *dev_priv = engine->i915;
-	unsigned reset_counter = i915_reset_counter(&dev_priv->gpu_error);
-	struct drm_i915_gem_request *req;
-	int ret;
-
-	if (!req_out)
-		return -EINVAL;
-
-	*req_out = NULL;
-
-	/* ABI: Before userspace accesses the GPU (e.g. execbuffer), report
-	 * EIO if the GPU is already wedged, or EAGAIN to drop the struct_mutex
-	 * and restart.
-	 */
-	ret = i915_gem_check_wedge(reset_counter, dev_priv->mm.interruptible);
-	if (ret)
-		return ret;
-
-	req = kmem_cache_zalloc(dev_priv->requests, GFP_KERNEL);
-	if (req == NULL)
-		return -ENOMEM;
-
-	ret = i915_gem_get_seqno(engine->i915, &req->seqno);
-	if (ret)
-		goto err;
-
-	kref_init(&req->ref);
-	req->i915 = dev_priv;
-	req->engine = engine;
-	req->ctx  = ctx;
-	i915_gem_context_reference(req->ctx);
-
-	/*
-	 * Reserve space in the ring buffer for all the commands required to
-	 * eventually emit this request. This is to guarantee that the
-	 * i915_add_request() call can't fail. Note that the reserve may need
-	 * to be redone if the request is not actually submitted straight
-	 * away, e.g. because a GPU scheduler has deferred it.
-	 */
-	req->reserved_space = MIN_SPACE_FOR_ADD_REQUEST;
-
-	if (i915.enable_execlists)
-		ret = intel_logical_ring_alloc_request_extras(req);
-	else
-		ret = intel_ring_alloc_request_extras(req);
-	if (ret)
-		goto err_ctx;
-
-	*req_out = req;
-	return 0;
-
-err_ctx:
-	i915_gem_context_unreference(ctx);
-err:
-	kmem_cache_free(dev_priv->requests, req);
-	return ret;
-}
-
-/**
- * i915_gem_request_alloc - allocate a request structure
- *
- * @engine: engine that we wish to issue the request on.
- * @ctx: context that the request will be associated with.
- *       This can be NULL if the request is not directly related to
- *       any specific user context, in which case this function will
- *       choose an appropriate context to use.
- *
- * Returns a pointer to the allocated request if successful,
- * or an error code if not.
- */
-struct drm_i915_gem_request *
-i915_gem_request_alloc(struct intel_engine_cs *engine,
-		       struct i915_gem_context *ctx)
-{
-	struct drm_i915_gem_request *req;
-	int err;
-
-	if (ctx == NULL)
-		ctx = engine->i915->kernel_context;
-	err = __i915_gem_request_alloc(engine, ctx, &req);
-	return err ? ERR_PTR(err) : req;
-}
-
 struct drm_i915_gem_request *
 i915_gem_find_active_request(struct intel_engine_cs *engine)
 {
@@ -3086,7 +2392,7 @@ i915_gem_find_active_request(struct intel_engine_cs *engine)
 	 * extra delay for a recent interrupt is pointless. Hence, we do
 	 * not need an engine->irq_seqno_barrier() before the seqno reads.
 	 */
-	list_for_each_entry(request, &engine->request_list, list) {
+	list_for_each_entry(request, &engine->request_list, link) {
 		if (i915_gem_request_completed(request))
 			continue;
 
@@ -3108,23 +2414,24 @@ static void i915_gem_reset_engine_status(struct intel_engine_cs *engine)
 	ring_hung = engine->hangcheck.score >= HANGCHECK_SCORE_RING_HUNG;
 
 	i915_set_reset_status(request->ctx, ring_hung);
-	list_for_each_entry_continue(request, &engine->request_list, list)
+	list_for_each_entry_continue(request, &engine->request_list, link)
 		i915_set_reset_status(request->ctx, false);
 }
 
 static void i915_gem_reset_engine_cleanup(struct intel_engine_cs *engine)
 {
-	struct intel_ringbuffer *buffer;
-
-	while (!list_empty(&engine->active_list)) {
-		struct drm_i915_gem_object *obj;
+	struct drm_i915_gem_request *request;
+	struct intel_ring *ring;
 
-		obj = list_first_entry(&engine->active_list,
-				       struct drm_i915_gem_object,
-				       engine_list[engine->id]);
+	request = i915_gem_active_peek(&engine->last_request,
+				       &engine->i915->drm.struct_mutex);
 
-		i915_gem_object_retire__read(obj, engine->id);
-	}
+	/* Mark all pending requests as complete so that any concurrent
+	 * (lockless) lookup doesn't try and wait upon the request as we
+	 * reset it.
+	 */
+	if (request)
+		intel_engine_init_seqno(engine, request->fence.seqno);
 
 	/*
 	 * Clear the execlists queue up before freeing the requests, as those
@@ -3146,15 +2453,9 @@ static void i915_gem_reset_engine_cleanup(struct intel_engine_cs *engine)
 	 * implicit references on things like e.g. ppgtt address spaces through
 	 * the request.
 	 */
-	while (!list_empty(&engine->request_list)) {
-		struct drm_i915_gem_request *request;
-
-		request = list_first_entry(&engine->request_list,
-					   struct drm_i915_gem_request,
-					   list);
-
-		i915_gem_request_retire(request);
-	}
+	if (request)
+		i915_gem_request_retire_upto(request);
+	GEM_BUG_ON(intel_engine_is_active(engine));
 
 	/* Having flushed all requests from all queues, we know that all
 	 * ringbuffers must now be empty. However, since we do not reclaim
@@ -3163,12 +2464,12 @@ static void i915_gem_reset_engine_cleanup(struct intel_engine_cs *engine)
 	 * upon reset is less than when we start. Do one more pass over
 	 * all the ringbuffers to reset last_retired_head.
 	 */
-	list_for_each_entry(buffer, &engine->buffers, link) {
-		buffer->last_retired_head = buffer->tail;
-		intel_ring_update_space(buffer);
+	list_for_each_entry(ring, &engine->buffers, link) {
+		ring->last_retired_head = ring->tail;
+		intel_ring_update_space(ring);
 	}
 
-	intel_ring_init_seqno(engine, engine->last_submitted_seqno);
+	engine->i915->gt.active_engines &= ~intel_engine_flag(engine);
 }
 
 void i915_gem_reset(struct drm_device *dev)
@@ -3186,82 +2487,11 @@ void i915_gem_reset(struct drm_device *dev)
 
 	for_each_engine(engine, dev_priv)
 		i915_gem_reset_engine_cleanup(engine);
+	mod_delayed_work(dev_priv->wq, &dev_priv->gt.idle_work, 0);
 
 	i915_gem_context_reset(dev);
 
 	i915_gem_restore_fences(dev);
-
-	WARN_ON(i915_verify_lists(dev));
-}
-
-/**
- * This function clears the request list as sequence numbers are passed.
- * @engine: engine to retire requests on
- */
-void
-i915_gem_retire_requests_ring(struct intel_engine_cs *engine)
-{
-	WARN_ON(i915_verify_lists(engine->dev));
-
-	/* Retire requests first as we use it above for the early return.
-	 * If we retire requests last, we may use a later seqno and so clear
-	 * the requests lists without clearing the active list, leading to
-	 * confusion.
-	 */
-	while (!list_empty(&engine->request_list)) {
-		struct drm_i915_gem_request *request;
-
-		request = list_first_entry(&engine->request_list,
-					   struct drm_i915_gem_request,
-					   list);
-
-		if (!i915_gem_request_completed(request))
-			break;
-
-		i915_gem_request_retire(request);
-	}
-
-	/* Move any buffers on the active list that are no longer referenced
-	 * by the ringbuffer to the flushing/inactive lists as appropriate,
-	 * before we free the context associated with the requests.
-	 */
-	while (!list_empty(&engine->active_list)) {
-		struct drm_i915_gem_object *obj;
-
-		obj = list_first_entry(&engine->active_list,
-				       struct drm_i915_gem_object,
-				       engine_list[engine->id]);
-
-		if (!list_empty(&obj->last_read_req[engine->id]->list))
-			break;
-
-		i915_gem_object_retire__read(obj, engine->id);
-	}
-
-	WARN_ON(i915_verify_lists(engine->dev));
-}
-
-void i915_gem_retire_requests(struct drm_i915_private *dev_priv)
-{
-	struct intel_engine_cs *engine;
-
-	lockdep_assert_held(&dev_priv->drm.struct_mutex);
-
-	if (dev_priv->gt.active_engines == 0)
-		return;
-
-	GEM_BUG_ON(!dev_priv->gt.awake);
-
-	for_each_engine(engine, dev_priv) {
-		i915_gem_retire_requests_ring(engine);
-		if (list_empty(&engine->request_list))
-			dev_priv->gt.active_engines &= ~intel_engine_flag(engine);
-	}
-
-	if (dev_priv->gt.active_engines == 0)
-		queue_delayed_work(dev_priv->wq,
-				   &dev_priv->gt.idle_work,
-				   msecs_to_jiffies(100));
 }
 
 static void
@@ -3281,10 +2511,12 @@ i915_gem_retire_work_handler(struct work_struct *work)
 	 * We do not need to do this test under locking as in the worst-case
 	 * we queue the retire worker once too often.
 	 */
-	if (READ_ONCE(dev_priv->gt.awake))
+	if (READ_ONCE(dev_priv->gt.awake)) {
+		i915_queue_hangcheck(dev_priv);
 		queue_delayed_work(dev_priv->wq,
 				   &dev_priv->gt.retire_work,
 				   round_jiffies_up_relative(HZ));
+	}
 }
 
 static void
@@ -3324,11 +2556,14 @@ i915_gem_idle_work_handler(struct work_struct *work)
 	dev_priv->gt.awake = false;
 	rearm_hangcheck = false;
 
+	/* As we have disabled hangcheck, we need to unstick any waiters still
+	 * hanging around. However, as we may be racing against the interrupt
+	 * handler or the waiters themselves, we skip enabling the fake-irq.
+	 */
 	stuck_engines = intel_kick_waiters(dev_priv);
-	if (unlikely(stuck_engines)) {
-		DRM_DEBUG_DRIVER("kicked stuck waiters...missed irq\n");
-		dev_priv->gpu_error.missed_irq_rings |= stuck_engines;
-	}
+	if (unlikely(stuck_engines))
+		DRM_DEBUG_DRIVER("kicked stuck waiters (%x)...missed irq?\n",
+				 stuck_engines);
 
 	if (INTEL_GEN(dev_priv) >= 6)
 		gen6_rps_idle(dev_priv);
@@ -3343,32 +2578,17 @@ out_rearm:
 	}
 }
 
-/**
- * Ensures that an object will eventually get non-busy by flushing any required
- * write domains, emitting any outstanding lazy request and retiring and
- * completed requests.
- * @obj: object to flush
- */
-static int
-i915_gem_object_flush_active(struct drm_i915_gem_object *obj)
+void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file)
 {
-	int i;
-
-	if (!obj->active)
-		return 0;
-
-	for (i = 0; i < I915_NUM_ENGINES; i++) {
-		struct drm_i915_gem_request *req;
-
-		req = obj->last_read_req[i];
-		if (req == NULL)
-			continue;
-
-		if (i915_gem_request_completed(req))
-			i915_gem_object_retire__read(obj, i);
-	}
+	struct drm_i915_gem_object *obj = to_intel_bo(gem);
+	struct drm_i915_file_private *fpriv = file->driver_priv;
+	struct i915_vma *vma, *vn;
 
-	return 0;
+	mutex_lock(&obj->base.dev->struct_mutex);
+	list_for_each_entry_safe(vma, vn, &obj->vma_list, obj_link)
+		if (vma->vm->file == fpriv)
+			i915_vma_close(vma);
+	mutex_unlock(&obj->base.dev->struct_mutex);
 }
 
 /**
@@ -3399,122 +2619,58 @@ int
 i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 {
 	struct drm_i915_gem_wait *args = data;
+	struct intel_rps_client *rps = to_rps_client(file);
 	struct drm_i915_gem_object *obj;
-	struct drm_i915_gem_request *req[I915_NUM_ENGINES];
-	int i, n = 0;
-	int ret;
+	unsigned long active;
+	int idx, ret = 0;
 
 	if (args->flags != 0)
 		return -EINVAL;
 
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
-
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->bo_handle));
-	if (&obj->base == NULL) {
-		mutex_unlock(&dev->struct_mutex);
+	obj = i915_gem_object_lookup(file, args->bo_handle);
+	if (!obj)
 		return -ENOENT;
-	}
-
-	/* Need to make sure the object gets inactive eventually. */
-	ret = i915_gem_object_flush_active(obj);
-	if (ret)
-		goto out;
-
-	if (!obj->active)
-		goto out;
-
-	/* Do this after OLR check to make sure we make forward progress polling
-	 * on this IOCTL with a timeout == 0 (like busy ioctl)
-	 */
-	if (args->timeout_ns == 0) {
-		ret = -ETIME;
-		goto out;
-	}
-
-	drm_gem_object_unreference(&obj->base);
 
-	for (i = 0; i < I915_NUM_ENGINES; i++) {
-		if (obj->last_read_req[i] == NULL)
-			continue;
-
-		req[n++] = i915_gem_request_reference(obj->last_read_req[i]);
-	}
-
-	mutex_unlock(&dev->struct_mutex);
-
-	for (i = 0; i < n; i++) {
-		if (ret == 0)
-			ret = __i915_wait_request(req[i], true,
-						  args->timeout_ns > 0 ? &args->timeout_ns : NULL,
-						  to_rps_client(file));
-		i915_gem_request_unreference(req[i]);
+	active = __I915_BO_ACTIVE(obj);
+	for_each_active(active, idx) {
+		s64 *timeout = args->timeout_ns >= 0 ? &args->timeout_ns : NULL;
+		ret = i915_gem_active_wait_unlocked(&obj->last_read[idx], true,
+						    timeout, rps);
+		if (ret)
+			break;
 	}
-	return ret;
 
-out:
-	drm_gem_object_unreference(&obj->base);
-	mutex_unlock(&dev->struct_mutex);
+	i915_gem_object_put_unlocked(obj);
 	return ret;
 }
 
 static int
-__i915_gem_object_sync(struct drm_i915_gem_object *obj,
-		       struct intel_engine_cs *to,
-		       struct drm_i915_gem_request *from_req,
-		       struct drm_i915_gem_request **to_req)
+__i915_gem_object_sync(struct drm_i915_gem_request *to,
+		       struct drm_i915_gem_request *from)
 {
-	struct intel_engine_cs *from;
 	int ret;
 
-	from = i915_gem_request_get_engine(from_req);
-	if (to == from)
+	if (to->engine == from->engine)
 		return 0;
 
-	if (i915_gem_request_completed(from_req))
-		return 0;
-
-	if (!i915_semaphore_is_enabled(to_i915(obj->base.dev))) {
-		struct drm_i915_private *i915 = to_i915(obj->base.dev);
-		ret = __i915_wait_request(from_req,
-					  i915->mm.interruptible,
-					  NULL,
-					  &i915->rps.semaphores);
+	if (!i915.semaphores) {
+		ret = i915_wait_request(from,
+					from->i915->mm.interruptible,
+					NULL,
+					NO_WAITBOOST);
 		if (ret)
 			return ret;
-
-		i915_gem_object_retire_request(obj, from_req);
 	} else {
-		int idx = intel_ring_sync_index(from, to);
-		u32 seqno = i915_gem_request_get_seqno(from_req);
-
-		WARN_ON(!to_req);
-
-		if (seqno <= from->semaphore.sync_seqno[idx])
+		int idx = intel_engine_sync_index(from->engine, to->engine);
+		if (from->fence.seqno <= from->engine->semaphore.sync_seqno[idx])
 			return 0;
 
-		if (*to_req == NULL) {
-			struct drm_i915_gem_request *req;
-
-			req = i915_gem_request_alloc(to, NULL);
-			if (IS_ERR(req))
-				return PTR_ERR(req);
-
-			*to_req = req;
-		}
-
-		trace_i915_gem_ring_sync_to(*to_req, from, from_req);
-		ret = to->semaphore.sync_to(*to_req, from, seqno);
+		trace_i915_gem_ring_sync_to(to, from);
+		ret = to->engine->semaphore.sync_to(to, from);
 		if (ret)
 			return ret;
 
-		/* We use last_read_req because sync_to()
-		 * might have just caused seqno wrap under
-		 * the radar.
-		 */
-		from->semaphore.sync_seqno[idx] =
-			i915_gem_request_get_seqno(obj->last_read_req[from->id]);
+		from->engine->semaphore.sync_seqno[idx] = from->fence.seqno;
 	}
 
 	return 0;
@@ -3524,17 +2680,12 @@ __i915_gem_object_sync(struct drm_i915_gem_object *obj,
  * i915_gem_object_sync - sync an object to a ring.
  *
  * @obj: object which may be in use on another ring.
- * @to: ring we wish to use the object on. May be NULL.
- * @to_req: request we wish to use the object for. See below.
- *          This will be allocated and returned if a request is
- *          required but not passed in.
+ * @to: request we are wishing to use
  *
  * This code is meant to abstract object synchronization with the GPU.
- * Calling with NULL implies synchronizing the object with the CPU
- * rather than a particular GPU ring. Conceptually we serialise writes
- * between engines inside the GPU. We only allow one engine to write
- * into a buffer at any time, but multiple readers. To ensure each has
- * a coherent view of memory, we must:
+ * Conceptually we serialise writes between engines inside the GPU.
+ * We only allow one engine to write into a buffer at any time, but
+ * multiple readers. To ensure each has a coherent view of memory, we must:
  *
  * - If there is an outstanding write request to the object, the new
  *   request must wait for it to complete (either CPU or in hw, requests
@@ -3543,44 +2694,39 @@ __i915_gem_object_sync(struct drm_i915_gem_object *obj,
  * - If we are a write request (pending_write_domain is set), the new
  *   request must wait for outstanding read requests to complete.
  *
- * For CPU synchronisation (NULL to) no request is required. For syncing with
- * rings to_req must be non-NULL. However, a request does not have to be
- * pre-allocated. If *to_req is NULL and sync commands will be emitted then a
- * request will be allocated automatically and returned through *to_req. Note
- * that it is not guaranteed that commands will be emitted (because the system
- * might already be idle). Hence there is no need to create a request that
- * might never have any work submitted. Note further that if a request is
- * returned in *to_req, it is the responsibility of the caller to submit
- * that request (after potentially adding more work to it).
- *
  * Returns 0 if successful, else propagates up the lower layer error.
  */
 int
 i915_gem_object_sync(struct drm_i915_gem_object *obj,
-		     struct intel_engine_cs *to,
-		     struct drm_i915_gem_request **to_req)
+		     struct drm_i915_gem_request *to)
 {
-	const bool readonly = obj->base.pending_write_domain == 0;
-	struct drm_i915_gem_request *req[I915_NUM_ENGINES];
-	int ret, i, n;
+	struct i915_gem_active *active;
+	unsigned long active_mask;
+	int idx;
 
-	if (!obj->active)
-		return 0;
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
 
-	if (to == NULL)
-		return i915_gem_object_wait_rendering(obj, readonly);
+	active_mask = i915_gem_object_get_active(obj);
+	if (!active_mask)
+		return 0;
 
-	n = 0;
-	if (readonly) {
-		if (obj->last_write_req)
-			req[n++] = obj->last_write_req;
+	if (obj->base.pending_write_domain) {
+		active = obj->last_read;
 	} else {
-		for (i = 0; i < I915_NUM_ENGINES; i++)
-			if (obj->last_read_req[i])
-				req[n++] = obj->last_read_req[i];
+		active_mask = 1;
+		active = &obj->last_write;
 	}
-	for (i = 0; i < n; i++) {
-		ret = __i915_gem_object_sync(obj, to, req[i], to_req);
+
+	for_each_active(active_mask, idx) {
+		struct drm_i915_gem_request *request;
+		int ret;
+
+		request = i915_gem_active_peek(&active[idx],
+					       &obj->base.dev->struct_mutex);
+		if (!request)
+			continue;
+
+		ret = __i915_gem_object_sync(to, request);
 		if (ret)
 			return ret;
 	}
@@ -3611,7 +2757,7 @@ static void i915_gem_object_finish_gtt(struct drm_i915_gem_object *obj)
 
 static void __i915_vma_iounmap(struct i915_vma *vma)
 {
-	GEM_BUG_ON(vma->pin_count);
+	GEM_BUG_ON(i915_vma_is_pinned(vma));
 
 	if (vma->iomap == NULL)
 		return;
@@ -3620,32 +2766,51 @@ static void __i915_vma_iounmap(struct i915_vma *vma)
 	vma->iomap = NULL;
 }
 
-static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
+int i915_vma_unbind(struct i915_vma *vma)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
-	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
+	unsigned long active;
 	int ret;
 
-	if (list_empty(&vma->obj_link))
-		return 0;
+	/* First wait upon any activity as retiring the request may
+	 * have side-effects such as unpinning or even unbinding this vma.
+	 */
+	active = i915_vma_get_active(vma);
+	if (active) {
+		int idx;
+
+		/* When a closed VMA is retired, it is unbound - eek.
+		 * In order to prevent it from being recursively closed,
+		 * take a pin on the vma so that the second unbind is
+		 * aborted.
+		 */
+		__i915_vma_pin(vma);
 
-	if (!drm_mm_node_allocated(&vma->node)) {
-		i915_gem_vma_destroy(vma);
-		return 0;
+		for_each_active(active, idx) {
+			ret = i915_gem_active_retire(&vma->last_read[idx],
+						   &vma->vm->dev->struct_mutex);
+			if (ret)
+				break;
+		}
+
+		__i915_vma_unpin(vma);
+		if (ret)
+			return ret;
+
+		GEM_BUG_ON(i915_vma_is_active(vma));
 	}
 
-	if (vma->pin_count)
+	if (i915_vma_is_pinned(vma))
 		return -EBUSY;
 
-	BUG_ON(obj->pages == NULL);
+	if (!drm_mm_node_allocated(&vma->node))
+		goto destroy;
 
-	if (wait) {
-		ret = i915_gem_object_wait_rendering(obj, false);
-		if (ret)
-			return ret;
-	}
+	GEM_BUG_ON(obj->bind_count == 0);
+	GEM_BUG_ON(!obj->pages);
 
-	if (vma->is_ggtt && vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
+	if (i915_vma_is_ggtt(vma) &&
+	    vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
 		i915_gem_object_finish_gtt(obj);
 
 		/* release the fence reg _after_ flushing */
@@ -3656,13 +2821,16 @@ static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
 		__i915_vma_iounmap(vma);
 	}
 
-	trace_i915_vma_unbind(vma);
+	if (likely(!vma->vm->closed)) {
+		trace_i915_vma_unbind(vma);
+		vma->vm->unbind_vma(vma);
+	}
+	vma->flags &= ~(I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND);
 
-	vma->vm->unbind_vma(vma);
-	vma->bound = 0;
+	drm_mm_remove_node(&vma->node);
+	list_move_tail(&vma->vm_link, &vma->vm->unbound_list);
 
-	list_del_init(&vma->vm_link);
-	if (vma->is_ggtt) {
+	if (i915_vma_is_ggtt(vma)) {
 		if (vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
 			obj->map_and_fenceable = false;
 		} else if (vma->ggtt_view.pages) {
@@ -3672,13 +2840,11 @@ static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
 		vma->ggtt_view.pages = NULL;
 	}
 
-	drm_mm_remove_node(&vma->node);
-	i915_gem_vma_destroy(vma);
-
 	/* Since the unbound list is global, only move to that list if
 	 * no more VMAs exist. */
-	if (list_empty(&obj->vma_list))
-		list_move_tail(&obj->global_list, &dev_priv->mm.unbound_list);
+	if (--obj->bind_count == 0)
+		list_move_tail(&obj->global_list,
+			       &to_i915(obj->base.dev)->mm.unbound_list);
 
 	/* And finally now the object is completely decoupled from this vma,
 	 * we can drop its hold on the backing storage and allow it to be
@@ -3686,36 +2852,28 @@ static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
 	 */
 	i915_gem_object_unpin_pages(obj);
 
-	return 0;
-}
-
-int i915_vma_unbind(struct i915_vma *vma)
-{
-	return __i915_vma_unbind(vma, true);
-}
+destroy:
+	if (unlikely(i915_vma_is_closed(vma)))
+		i915_vma_destroy(vma);
 
-int __i915_vma_unbind_no_wait(struct i915_vma *vma)
-{
-	return __i915_vma_unbind(vma, false);
+	return 0;
 }
 
-int i915_gem_wait_for_idle(struct drm_i915_private *dev_priv)
+int i915_gem_wait_for_idle(struct drm_i915_private *dev_priv,
+			   bool interruptible)
 {
 	struct intel_engine_cs *engine;
 	int ret;
 
-	lockdep_assert_held(&dev_priv->drm.struct_mutex);
-
 	for_each_engine(engine, dev_priv) {
 		if (engine->last_context == NULL)
 			continue;
 
-		ret = intel_engine_idle(engine);
+		ret = intel_engine_idle(engine, interruptible);
 		if (ret)
 			return ret;
 	}
 
-	WARN_ON(i915_verify_lists(dev));
 	return 0;
 }
 
@@ -3753,128 +2911,95 @@ static bool i915_gem_valid_gtt_space(struct i915_vma *vma,
 }
 
 /**
- * Finds free space in the GTT aperture and binds the object or a view of it
- * there.
- * @obj: object to bind
- * @vm: address space to bind into
- * @ggtt_view: global gtt view if applicable
- * @alignment: requested alignment
+ * i915_vma_insert - finds a slot for the vma in its address space
+ * @vma: the vma
+ * @size: requested size in bytes (can be larger than the VMA)
+ * @alignment: required alignment
  * @flags: mask of PIN_* flags to use
+ *
+ * First we try to allocate some free space that meets the requirements for
+ * the VMA. Failiing that, if the flags permit, it will evict an old VMA,
+ * preferrably the oldest idle entry to make room for the new VMA.
+ *
+ * Returns:
+ * 0 on success, negative error code otherwise.
  */
-static struct i915_vma *
-i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
-			   struct i915_address_space *vm,
-			   const struct i915_ggtt_view *ggtt_view,
-			   unsigned alignment,
-			   uint64_t flags)
+static int
+i915_vma_insert(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
 {
-	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-	u32 fence_alignment, unfenced_alignment;
-	u32 search_flag, alloc_flag;
+	struct drm_i915_private *dev_priv = to_i915(vma->vm->dev);
+	struct drm_i915_gem_object *obj = vma->obj;
 	u64 start, end;
-	u64 size, fence_size;
-	struct i915_vma *vma;
+	u64 min_alignment;
 	int ret;
 
-	if (i915_is_ggtt(vm)) {
-		u32 view_size;
-
-		if (WARN_ON(!ggtt_view))
-			return ERR_PTR(-EINVAL);
-
-		view_size = i915_ggtt_view_size(obj, ggtt_view);
-
-		fence_size = i915_gem_get_gtt_size(dev,
-						   view_size,
-						   obj->tiling_mode);
-		fence_alignment = i915_gem_get_gtt_alignment(dev,
-							     view_size,
-							     obj->tiling_mode,
-							     true);
-		unfenced_alignment = i915_gem_get_gtt_alignment(dev,
-								view_size,
-								obj->tiling_mode,
-								false);
-		size = flags & PIN_MAPPABLE ? fence_size : view_size;
-	} else {
-		fence_size = i915_gem_get_gtt_size(dev,
-						   obj->base.size,
-						   obj->tiling_mode);
-		fence_alignment = i915_gem_get_gtt_alignment(dev,
-							     obj->base.size,
-							     obj->tiling_mode,
-							     true);
-		unfenced_alignment =
-			i915_gem_get_gtt_alignment(dev,
-						   obj->base.size,
-						   obj->tiling_mode,
-						   false);
-		size = flags & PIN_MAPPABLE ? fence_size : obj->base.size;
+	GEM_BUG_ON(vma->flags & (I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND));
+	GEM_BUG_ON(drm_mm_node_allocated(&vma->node));
+
+	size = max(size, vma->size);
+	if (flags & PIN_MAPPABLE)
+		size = i915_gem_get_ggtt_size(dev_priv, size,
+					      i915_gem_object_get_tiling(obj));
+
+	min_alignment =
+		i915_gem_get_ggtt_alignment(dev_priv, size,
+					    i915_gem_object_get_tiling(obj),
+					    flags & PIN_MAPPABLE);
+	if (alignment == 0)
+		alignment = min_alignment;
+	if (alignment & (min_alignment - 1)) {
+		DRM_DEBUG("Invalid object alignment requested %llu, minimum %llu\n",
+			  alignment, min_alignment);
+		return -EINVAL;
 	}
 
 	start = flags & PIN_OFFSET_BIAS ? flags & PIN_OFFSET_MASK : 0;
-	end = vm->total;
+
+	end = vma->vm->total;
 	if (flags & PIN_MAPPABLE)
-		end = min_t(u64, end, ggtt->mappable_end);
+		end = min_t(u64, end, dev_priv->ggtt.mappable_end);
 	if (flags & PIN_ZONE_4G)
 		end = min_t(u64, end, (1ULL << 32) - PAGE_SIZE);
 
-	if (alignment == 0)
-		alignment = flags & PIN_MAPPABLE ? fence_alignment :
-						unfenced_alignment;
-	if (flags & PIN_MAPPABLE && alignment & (fence_alignment - 1)) {
-		DRM_DEBUG("Invalid object (view type=%u) alignment requested %u\n",
-			  ggtt_view ? ggtt_view->type : 0,
-			  alignment);
-		return ERR_PTR(-EINVAL);
-	}
-
 	/* If binding the object/GGTT view requires more space than the entire
 	 * aperture has, reject it early before evicting everything in a vain
 	 * attempt to find space.
 	 */
 	if (size > end) {
-		DRM_DEBUG("Attempting to bind an object (view type=%u) larger than the aperture: size=%llu > %s aperture=%llu\n",
-			  ggtt_view ? ggtt_view->type : 0,
-			  size,
+		DRM_DEBUG("Attempting to bind an object larger than the aperture: request=%llu [object=%zd] > %s aperture=%llu\n",
+			  size, obj->base.size,
 			  flags & PIN_MAPPABLE ? "mappable" : "total",
 			  end);
-		return ERR_PTR(-E2BIG);
+		return -E2BIG;
 	}
 
 	ret = i915_gem_object_get_pages(obj);
 	if (ret)
-		return ERR_PTR(ret);
+		return ret;
 
 	i915_gem_object_pin_pages(obj);
 
-	vma = ggtt_view ? i915_gem_obj_lookup_or_create_ggtt_vma(obj, ggtt_view) :
-			  i915_gem_obj_lookup_or_create_vma(obj, vm);
-
-	if (IS_ERR(vma))
-		goto err_unpin;
-
 	if (flags & PIN_OFFSET_FIXED) {
-		uint64_t offset = flags & PIN_OFFSET_MASK;
-
-		if (offset & (alignment - 1) || offset + size > end) {
+		u64 offset = flags & PIN_OFFSET_MASK;
+		if (offset & (alignment - 1) || offset > end - size) {
 			ret = -EINVAL;
-			goto err_free_vma;
+			goto err_unpin;
 		}
+
 		vma->node.start = offset;
 		vma->node.size = size;
 		vma->node.color = obj->cache_level;
-		ret = drm_mm_reserve_node(&vm->mm, &vma->node);
+		ret = drm_mm_reserve_node(&vma->vm->mm, &vma->node);
 		if (ret) {
 			ret = i915_gem_evict_for_vma(vma);
 			if (ret == 0)
-				ret = drm_mm_reserve_node(&vm->mm, &vma->node);
+				ret = drm_mm_reserve_node(&vma->vm->mm, &vma->node);
+			if (ret)
+				goto err_unpin;
 		}
-		if (ret)
-			goto err_free_vma;
 	} else {
+		u32 search_flag, alloc_flag;
+
 		if (flags & PIN_HIGH) {
 			search_flag = DRM_MM_SEARCH_BELOW;
 			alloc_flag = DRM_MM_CREATE_TOP;
@@ -3883,47 +3008,45 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 			alloc_flag = DRM_MM_CREATE_DEFAULT;
 		}
 
+		/* We only allocate in PAGE_SIZE/GTT_PAGE_SIZE (4096) chunks,
+		 * so we know that we always have a minimum alignment of 4096.
+		 * The drm_mm range manager is optimised to return results
+		 * with zero alignment, so where possible use the optimal
+		 * path.
+		 */
+		if (alignment <= 4096)
+			alignment = 0;
+
 search_free:
-		ret = drm_mm_insert_node_in_range_generic(&vm->mm, &vma->node,
+		ret = drm_mm_insert_node_in_range_generic(&vma->vm->mm,
+							  &vma->node,
 							  size, alignment,
 							  obj->cache_level,
 							  start, end,
 							  search_flag,
 							  alloc_flag);
 		if (ret) {
-			ret = i915_gem_evict_something(dev, vm, size, alignment,
+			ret = i915_gem_evict_something(vma->vm, size, alignment,
 						       obj->cache_level,
 						       start, end,
 						       flags);
 			if (ret == 0)
 				goto search_free;
 
-			goto err_free_vma;
+			goto err_unpin;
 		}
 	}
-	if (WARN_ON(!i915_gem_valid_gtt_space(vma, obj->cache_level))) {
-		ret = -EINVAL;
-		goto err_remove_node;
-	}
-
-	trace_i915_vma_bind(vma, flags);
-	ret = i915_vma_bind(vma, obj->cache_level, flags);
-	if (ret)
-		goto err_remove_node;
+	GEM_BUG_ON(!i915_gem_valid_gtt_space(vma, obj->cache_level));
 
 	list_move_tail(&obj->global_list, &dev_priv->mm.bound_list);
-	list_add_tail(&vma->vm_link, &vm->inactive_list);
+	list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
+	obj->bind_count++;
 
-	return vma;
+	return 0;
 
-err_remove_node:
-	drm_mm_remove_node(&vma->node);
-err_free_vma:
-	i915_gem_vma_destroy(vma);
-	vma = ERR_PTR(ret);
 err_unpin:
 	i915_gem_object_unpin_pages(obj);
-	return vma;
+	return ret;
 }
 
 bool
@@ -4026,20 +3149,17 @@ i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj)
 int
 i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 {
-	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	uint32_t old_write_domain, old_read_domains;
 	struct i915_vma *vma;
 	int ret;
 
-	if (obj->base.write_domain == I915_GEM_DOMAIN_GTT)
-		return 0;
-
 	ret = i915_gem_object_wait_rendering(obj, !write);
 	if (ret)
 		return ret;
 
+	if (obj->base.write_domain == I915_GEM_DOMAIN_GTT)
+		return 0;
+
 	/* Flush and acquire obj->pages so that we are coherent through
 	 * direct access in memory with previous cached writes through
 	 * shmemfs and that our cache domain tracking remains valid.
@@ -4081,9 +3201,10 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 
 	/* And bump the LRU for this access */
 	vma = i915_gem_obj_to_ggtt(obj);
-	if (vma && drm_mm_node_allocated(&vma->node) && !obj->active)
-		list_move_tail(&vma->vm_link,
-			       &ggtt->base.inactive_list);
+	if (vma &&
+	    drm_mm_node_allocated(&vma->node) &&
+	    !i915_vma_is_active(vma))
+		list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
 
 	return 0;
 }
@@ -4106,9 +3227,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 				    enum i915_cache_level cache_level)
 {
-	struct drm_device *dev = obj->base.dev;
-	struct i915_vma *vma, *next;
-	bool bound = false;
+	struct i915_vma *vma;
 	int ret = 0;
 
 	if (obj->cache_level == cache_level)
@@ -4119,21 +3238,28 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 	 * catch the issue of the CS prefetch crossing page boundaries and
 	 * reading an invalid PTE on older architectures.
 	 */
-	list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link) {
+restart:
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
 		if (!drm_mm_node_allocated(&vma->node))
 			continue;
 
-		if (vma->pin_count) {
+		if (i915_vma_is_pinned(vma)) {
 			DRM_DEBUG("can not change the cache level of pinned objects\n");
 			return -EBUSY;
 		}
 
-		if (!i915_gem_valid_gtt_space(vma, cache_level)) {
-			ret = i915_vma_unbind(vma);
-			if (ret)
-				return ret;
-		} else
-			bound = true;
+		if (i915_gem_valid_gtt_space(vma, cache_level))
+			continue;
+
+		ret = i915_vma_unbind(vma);
+		if (ret)
+			return ret;
+
+		/* As unbinding may affect other elements in the
+		 * obj->vma_list (due to side-effects from retiring
+		 * an active vma), play safe and restart the iterator.
+		 */
+		goto restart;
 	}
 
 	/* We can reuse the existing drm_mm nodes but need to change the
@@ -4143,7 +3269,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 	 * rewrite the PTE in the belief that doing so tramples upon less
 	 * state and so involves less work.
 	 */
-	if (bound) {
+	if (obj->bind_count) {
 		/* Before we change the PTE, the GPU must not be accessing it.
 		 * If we wait upon the object, we know that all the bound
 		 * VMA are no longer active.
@@ -4152,7 +3278,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 		if (ret)
 			return ret;
 
-		if (!HAS_LLC(dev) && cache_level != I915_CACHE_NONE) {
+		if (!HAS_LLC(obj->base.dev) && cache_level != I915_CACHE_NONE) {
 			/* Access to snoopable pages through the GTT is
 			 * incoherent and on some machines causes a hard
 			 * lockup. Relinquish the CPU mmaping to force
@@ -4215,8 +3341,8 @@ int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data,
 	struct drm_i915_gem_caching *args = data;
 	struct drm_i915_gem_object *obj;
 
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL)
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
 		return -ENOENT;
 
 	switch (obj->cache_level) {
@@ -4234,7 +3360,7 @@ int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data,
 		break;
 	}
 
-	drm_gem_object_unreference_unlocked(&obj->base);
+	i915_gem_object_put_unlocked(obj);
 	return 0;
 }
 
@@ -4276,15 +3402,15 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto rpm_put;
 
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL) {
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj) {
 		ret = -ENOENT;
 		goto unlock;
 	}
 
 	ret = i915_gem_object_set_cache_level(obj, level);
 
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 unlock:
 	mutex_unlock(&dev->struct_mutex);
 rpm_put:
@@ -4329,7 +3455,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	 * (e.g. libkms for the bootup splash), we have to ensure that we
 	 * always use map_and_fenceable for all scanout buffers.
 	 */
-	ret = i915_gem_object_ggtt_pin(obj, view, alignment,
+	ret = i915_gem_object_ggtt_pin(obj, view, 0, alignment,
 				       view->type == I915_GGTT_VIEW_NORMAL ?
 				       PIN_MAPPABLE : 0);
 	if (ret)
@@ -4383,13 +3509,13 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write)
 	uint32_t old_write_domain, old_read_domains;
 	int ret;
 
-	if (obj->base.write_domain == I915_GEM_DOMAIN_CPU)
-		return 0;
-
 	ret = i915_gem_object_wait_rendering(obj, !write);
 	if (ret)
 		return ret;
 
+	if (obj->base.write_domain == I915_GEM_DOMAIN_CPU)
+		return 0;
+
 	i915_gem_object_flush_gtt_write_domain(obj);
 
 	old_write_domain = obj->base.write_domain;
@@ -4464,25 +3590,30 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 		target = request;
 	}
 	if (target)
-		i915_gem_request_reference(target);
+		i915_gem_request_get(target);
 	spin_unlock(&file_priv->mm.lock);
 
 	if (target == NULL)
 		return 0;
 
-	ret = __i915_wait_request(target, true, NULL, NULL);
-	i915_gem_request_unreference(target);
+	ret = i915_wait_request(target, true, NULL, NULL);
+	i915_gem_request_put(target);
 
 	return ret;
 }
 
 static bool
-i915_vma_misplaced(struct i915_vma *vma, uint32_t alignment, uint64_t flags)
+i915_vma_misplaced(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
 
-	if (alignment &&
-	    vma->node.start & (alignment - 1))
+	if (!drm_mm_node_allocated(&vma->node))
+		return false;
+
+	if (vma->node.size < size)
+		return true;
+
+	if (alignment && vma->node.start & (alignment - 1))
 		return true;
 
 	if (flags & PIN_MAPPABLE && !obj->map_and_fenceable)
@@ -4502,135 +3633,159 @@ i915_vma_misplaced(struct i915_vma *vma, uint32_t alignment, uint64_t flags)
 void __i915_vma_set_map_and_fenceable(struct i915_vma *vma)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 	bool mappable, fenceable;
 	u32 fence_size, fence_alignment;
 
-	fence_size = i915_gem_get_gtt_size(obj->base.dev,
-					   obj->base.size,
-					   obj->tiling_mode);
-	fence_alignment = i915_gem_get_gtt_alignment(obj->base.dev,
-						     obj->base.size,
-						     obj->tiling_mode,
-						     true);
+	fence_size = i915_gem_get_ggtt_size(dev_priv,
+					    obj->base.size,
+					    i915_gem_object_get_tiling(obj));
+	fence_alignment = i915_gem_get_ggtt_alignment(dev_priv,
+						      obj->base.size,
+						      i915_gem_object_get_tiling(obj),
+						      true);
 
 	fenceable = (vma->node.size == fence_size &&
 		     (vma->node.start & (fence_alignment - 1)) == 0);
 
 	mappable = (vma->node.start + fence_size <=
-		    to_i915(obj->base.dev)->ggtt.mappable_end);
+		    dev_priv->ggtt.mappable_end);
 
 	obj->map_and_fenceable = mappable && fenceable;
 }
 
-static int
-i915_gem_object_do_pin(struct drm_i915_gem_object *obj,
-		       struct i915_address_space *vm,
-		       const struct i915_ggtt_view *ggtt_view,
-		       uint32_t alignment,
-		       uint64_t flags)
+int __i915_vma_do_pin(struct i915_vma *vma,
+		      u64 size, u64 alignment, u64 flags)
 {
-	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
-	struct i915_vma *vma;
-	unsigned bound;
+	unsigned int bound = vma->flags;
 	int ret;
 
-	if (WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base))
-		return -ENODEV;
-
-	if (WARN_ON(flags & (PIN_GLOBAL | PIN_MAPPABLE) && !i915_is_ggtt(vm)))
-		return -EINVAL;
-
-	if (WARN_ON((flags & (PIN_MAPPABLE | PIN_GLOBAL)) == PIN_MAPPABLE))
-		return -EINVAL;
-
-	if (WARN_ON(i915_is_ggtt(vm) != !!ggtt_view))
-		return -EINVAL;
-
-	vma = ggtt_view ? i915_gem_obj_to_ggtt_view(obj, ggtt_view) :
-			  i915_gem_obj_to_vma(obj, vm);
-
-	if (vma) {
-		if (WARN_ON(vma->pin_count == DRM_I915_GEM_OBJECT_MAX_PIN_COUNT))
-			return -EBUSY;
+	GEM_BUG_ON((flags & (PIN_GLOBAL | PIN_USER)) == 0);
+	GEM_BUG_ON((flags & PIN_GLOBAL) && !i915_vma_is_ggtt(vma));
 
-		if (i915_vma_misplaced(vma, alignment, flags)) {
-			WARN(vma->pin_count,
-			     "bo is already pinned in %s with incorrect alignment:"
-			     " offset=%08x %08x, req.alignment=%x, req.map_and_fenceable=%d,"
-			     " obj->map_and_fenceable=%d\n",
-			     ggtt_view ? "ggtt" : "ppgtt",
-			     upper_32_bits(vma->node.start),
-			     lower_32_bits(vma->node.start),
-			     alignment,
-			     !!(flags & PIN_MAPPABLE),
-			     obj->map_and_fenceable);
-			ret = i915_vma_unbind(vma);
-			if (ret)
-				return ret;
-
-			vma = NULL;
-		}
+	if (WARN_ON(bound & I915_VMA_PIN_OVERFLOW)) {
+		ret = -EBUSY;
+		goto err;
 	}
 
-	bound = vma ? vma->bound : 0;
-	if (vma == NULL || !drm_mm_node_allocated(&vma->node)) {
-		vma = i915_gem_object_bind_to_vm(obj, vm, ggtt_view, alignment,
-						 flags);
-		if (IS_ERR(vma))
-			return PTR_ERR(vma);
-	} else {
-		ret = i915_vma_bind(vma, obj->cache_level, flags);
+	if ((bound & I915_VMA_BIND_MASK) == 0) {
+		ret = i915_vma_insert(vma, size, alignment, flags);
 		if (ret)
-			return ret;
+			goto err;
 	}
 
-	if (ggtt_view && ggtt_view->type == I915_GGTT_VIEW_NORMAL &&
-	    (bound ^ vma->bound) & GLOBAL_BIND) {
+	ret = i915_vma_bind(vma, vma->obj->cache_level, flags);
+	if (ret)
+		goto err;
+
+	if ((bound ^ vma->flags) & I915_VMA_GLOBAL_BIND)
 		__i915_vma_set_map_and_fenceable(vma);
-		WARN_ON(flags & PIN_MAPPABLE && !obj->map_and_fenceable);
-	}
 
-	vma->pin_count++;
+	GEM_BUG_ON(i915_vma_misplaced(vma, size, alignment, flags));
 	return 0;
-}
 
-int
-i915_gem_object_pin(struct drm_i915_gem_object *obj,
-		    struct i915_address_space *vm,
-		    uint32_t alignment,
-		    uint64_t flags)
-{
-	return i915_gem_object_do_pin(obj, vm,
-				      i915_is_ggtt(vm) ? &i915_ggtt_view_normal : NULL,
-				      alignment, flags);
+err:
+	__i915_vma_unpin(vma);
+	return ret;
 }
 
 int
 i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
 			 const struct i915_ggtt_view *view,
-			 uint32_t alignment,
-			 uint64_t flags)
+			 u64 size,
+			 u64 alignment,
+			 u64 flags)
 {
-	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+	struct i915_vma *vma;
+	int ret;
+
+	if (!view)
+		view = &i915_ggtt_view_normal;
 
-	BUG_ON(!view);
+	vma = i915_gem_obj_lookup_or_create_ggtt_vma(obj, view);
+	if (IS_ERR(vma))
+		return PTR_ERR(vma);
+
+	if (i915_vma_misplaced(vma, size, alignment, flags)) {
+		if (flags & PIN_NONBLOCK &&
+		    (i915_vma_is_pinned(vma) || i915_vma_is_active(vma)))
+			return -ENOSPC;
+
+		WARN(i915_vma_is_pinned(vma),
+		     "bo is already pinned in ggtt with incorrect alignment:"
+		     " offset=%08x %08x, req.alignment=%llx, req.map_and_fenceable=%d,"
+		     " obj->map_and_fenceable=%d\n",
+		     upper_32_bits(vma->node.start),
+		     lower_32_bits(vma->node.start),
+		     alignment,
+		     !!(flags & PIN_MAPPABLE),
+		     obj->map_and_fenceable);
+		ret = i915_vma_unbind(vma);
+		if (ret)
+			return ret;
+	}
 
-	return i915_gem_object_do_pin(obj, &ggtt->base, view,
-				      alignment, flags | PIN_GLOBAL);
+	return i915_vma_pin(vma, size, alignment, flags | PIN_GLOBAL);
 }
 
 void
 i915_gem_object_ggtt_unpin_view(struct drm_i915_gem_object *obj,
 				const struct i915_ggtt_view *view)
 {
-	struct i915_vma *vma = i915_gem_obj_to_ggtt_view(obj, view);
+	i915_vma_unpin(i915_gem_obj_to_ggtt_view(obj, view));
+}
+
+static __always_inline unsigned __busy_read_flag(unsigned int id)
+{
+	/* Note that we could alias engines in the execbuf API, but
+	 * that would be very unwise as it prevents userspace from
+	 * fine control over engine selection. Ahem.
+	 *
+	 * This should be something like EXEC_MAX_ENGINE instead of
+	 * I915_NUM_ENGINES.
+	 */
+	BUILD_BUG_ON(I915_NUM_ENGINES > 16);
+	return 0x10000 << id;
+}
+
+static __always_inline unsigned int __busy_write_id(unsigned int id)
+{
+	return id;
+}
+
+static __always_inline unsigned
+__busy_set_if_active(const struct i915_gem_active *active,
+		     unsigned int (*flag)(unsigned int id))
+{
+	/* For more discussion about the barriers and locking concerns,
+	 * see __i915_gem_active_get_rcu().
+	 */
+	do {
+		struct drm_i915_gem_request *request;
+		unsigned int id;
+
+		request = rcu_dereference(active->request);
+		if (!request || i915_gem_request_completed(request))
+			return 0;
+
+		id = request->engine->exec_id;
 
-	WARN_ON(vma->pin_count == 0);
-	WARN_ON(!i915_gem_obj_ggtt_bound_view(obj, view));
+		/* Check that the pointer wasn't reassigned and overwritten. */
+		if (request == rcu_access_pointer(active->request))
+			return flag(id);
+	} while (1);
+}
+
+static inline unsigned
+busy_check_reader(const struct i915_gem_active *active)
+{
+	return __busy_set_if_active(active, __busy_read_flag);
+}
 
-	--vma->pin_count;
+static inline unsigned
+busy_check_writer(const struct i915_gem_active *active)
+{
+	return __busy_set_if_active(active, __busy_write_id);
 }
 
 int
@@ -4639,47 +3794,61 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_i915_gem_busy *args = data;
 	struct drm_i915_gem_object *obj;
-	int ret;
+	unsigned long active;
 
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
+		return -ENOENT;
 
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL) {
-		ret = -ENOENT;
-		goto unlock;
-	}
+	args->busy = 0;
+	active = __I915_BO_ACTIVE(obj);
+	if (active) {
+		int idx;
 
-	/* Count all active objects as busy, even if they are currently not used
-	 * by the gpu. Users of this interface expect objects to eventually
-	 * become non-busy without any further actions, therefore emit any
-	 * necessary flushes here.
-	 */
-	ret = i915_gem_object_flush_active(obj);
-	if (ret)
-		goto unref;
+		/* Yes, the lookups are intentionally racy.
+		 *
+		 * First, we cannot simply rely on __I915_BO_ACTIVE. We have
+		 * to regard the value as stale and as our ABI guarantees
+		 * forward progress, we confirm the status of each active
+		 * request with the hardware.
+		 *
+		 * Even though we guard the pointer lookup by RCU, that only
+		 * guarantees that the pointer and its contents remain
+		 * dereferencable and does *not* mean that the request we
+		 * have is the same as the one being tracked by the object.
+		 *
+		 * Consider that we lookup the request just as it is being
+		 * retired and freed. We take a local copy of the pointer,
+		 * but before we add its engine into the busy set, the other
+		 * thread reallocates it and assigns it to a task on another
+		 * engine with a fresh and incomplete seqno.
+		 *
+		 * So after we lookup the engine's id, we double check that
+		 * the active request is the same and only then do we add it
+		 * into the busy set.
+		 */
+		rcu_read_lock();
 
-	args->busy = 0;
-	if (obj->active) {
-		int i;
+		for_each_active(active, idx)
+			args->busy |= busy_check_reader(&obj->last_read[idx]);
 
-		for (i = 0; i < I915_NUM_ENGINES; i++) {
-			struct drm_i915_gem_request *req;
+		/* For ABI sanity, we only care that the write engine is in
+		 * the set of read engines. This is ensured by the ordering
+		 * of setting last_read/last_write in i915_vma_move_to_active,
+		 * and then in reverse in retire.
+		 *
+		 * We don't care that the set of active read/write engines
+		 * may change during construction of the result, as it is
+		 * equally liable to change before userspace can inspect
+		 * the result.
+		 */
+		args->busy |= busy_check_writer(&obj->last_write);
 
-			req = obj->last_read_req[i];
-			if (req)
-				args->busy |= 1 << (16 + req->engine->exec_id);
-		}
-		if (obj->last_write_req)
-			args->busy |= obj->last_write_req->engine->exec_id;
+		rcu_read_unlock();
 	}
 
-unref:
-	drm_gem_object_unreference(&obj->base);
-unlock:
-	mutex_unlock(&dev->struct_mutex);
-	return ret;
+	i915_gem_object_put_unlocked(obj);
+	return 0;
 }
 
 int
@@ -4710,19 +3879,14 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	obj = to_intel_bo(drm_gem_object_lookup(file_priv, args->handle));
-	if (&obj->base == NULL) {
+	obj = i915_gem_object_lookup(file_priv, args->handle);
+	if (!obj) {
 		ret = -ENOENT;
 		goto unlock;
 	}
 
-	if (i915_gem_obj_is_pinned(obj)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
 	if (obj->pages &&
-	    obj->tiling_mode != I915_TILING_NONE &&
+	    i915_gem_object_is_tiled(obj) &&
 	    dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
 		if (obj->madv == I915_MADV_WILLNEED)
 			i915_gem_object_unpin_pages(obj);
@@ -4739,8 +3903,7 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
 
 	args->retained = obj->madv != __I915_MADV_PURGED;
 
-out:
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 unlock:
 	mutex_unlock(&dev->struct_mutex);
 	return ret;
@@ -4753,7 +3916,11 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj,
 
 	INIT_LIST_HEAD(&obj->global_list);
 	for (i = 0; i < I915_NUM_ENGINES; i++)
-		INIT_LIST_HEAD(&obj->engine_list[i]);
+		init_request_active(&obj->last_read[i],
+				    i915_gem_object_retire__read);
+	init_request_active(&obj->last_write,
+			    i915_gem_object_retire__write);
+	init_request_active(&obj->last_fence, NULL);
 	INIT_LIST_HEAD(&obj->obj_exec_link);
 	INIT_LIST_HEAD(&obj->vma_list);
 	INIT_LIST_HEAD(&obj->batch_pool_link);
@@ -4865,33 +4032,31 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 
 	trace_i915_gem_object_destroy(obj);
 
+	/* All file-owned VMA should have been released by this point through
+	 * i915_gem_close_object(), or earlier by i915_gem_context_close().
+	 * However, the object may also be bound into the global GTT (e.g.
+	 * older GPUs without per-process support, or for direct access through
+	 * the GTT either for the user or for scanout). Those VMA still need to
+	 * unbound now.
+	 */
 	list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link) {
-		int ret;
-
-		vma->pin_count = 0;
-		ret = i915_vma_unbind(vma);
-		if (WARN_ON(ret == -ERESTARTSYS)) {
-			bool was_interruptible;
-
-			was_interruptible = dev_priv->mm.interruptible;
-			dev_priv->mm.interruptible = false;
-
-			WARN_ON(i915_vma_unbind(vma));
-
-			dev_priv->mm.interruptible = was_interruptible;
-		}
+		GEM_BUG_ON(!i915_vma_is_ggtt(vma));
+		GEM_BUG_ON(i915_vma_is_active(vma));
+		vma->flags &= ~I915_VMA_PIN_MASK;
+		i915_vma_close(vma);
 	}
+	GEM_BUG_ON(obj->bind_count);
 
 	/* Stolen objects don't hold a ref, but do hold pin count. Fix that up
 	 * before progressing. */
 	if (obj->stolen)
 		i915_gem_object_unpin_pages(obj);
 
-	WARN_ON(obj->frontbuffer_bits);
+	WARN_ON(atomic_read(&obj->frontbuffer_bits));
 
 	if (obj->pages && obj->madv == I915_MADV_WILLNEED &&
 	    dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES &&
-	    obj->tiling_mode != I915_TILING_NONE)
+	    i915_gem_object_is_tiled(obj))
 		i915_gem_object_unpin_pages(obj);
 
 	if (WARN_ON(obj->pages_pin_count))
@@ -4899,7 +4064,6 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 	if (discard_backing_storage(obj))
 		obj->madv = I915_MADV_DONTNEED;
 	i915_gem_object_put_pages(obj);
-	i915_gem_object_free_mmap_offset(obj);
 
 	BUG_ON(obj->pages);
 
@@ -4938,51 +4102,39 @@ struct i915_vma *i915_gem_obj_to_ggtt_view(struct drm_i915_gem_object *obj,
 	GEM_BUG_ON(!view);
 
 	list_for_each_entry(vma, &obj->vma_list, obj_link)
-		if (vma->is_ggtt && i915_ggtt_view_equal(&vma->ggtt_view, view))
+		if (i915_vma_is_ggtt(vma) &&
+		    i915_ggtt_view_equal(&vma->ggtt_view, view))
 			return vma;
 	return NULL;
 }
 
-void i915_gem_vma_destroy(struct i915_vma *vma)
-{
-	WARN_ON(vma->node.allocated);
-
-	/* Keep the vma as a placeholder in the execbuffer reservation lists */
-	if (!list_empty(&vma->exec_list))
-		return;
-
-	if (!vma->is_ggtt)
-		i915_ppgtt_put(i915_vm_to_ppgtt(vma->vm));
-
-	list_del(&vma->obj_link);
-
-	kmem_cache_free(to_i915(vma->obj->base.dev)->vmas, vma);
-}
-
-static void
-i915_gem_stop_engines(struct drm_device *dev)
+int i915_gem_suspend(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_engine_cs *engine;
-
-	for_each_engine(engine, dev_priv)
-		dev_priv->gt.stop_engine(engine);
-}
+	int ret;
 
-int
-i915_gem_suspend(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret = 0;
+	intel_suspend_gt_powersave(dev_priv);
 
 	mutex_lock(&dev->struct_mutex);
-	ret = i915_gem_wait_for_idle(dev_priv);
+
+	/* We have to flush all the executing contexts to main memory so
+	 * that they can saved in the hibernation image. To ensure the last
+	 * context image is coherent, we have to switch away from it. That
+	 * leaves the dev_priv->kernel_context still active when
+	 * we actually suspend, and its image in memory may not match the GPU
+	 * state. Fortunately, the kernel_context is disposable and we do
+	 * not rely on its state.
+	 */
+	ret = i915_gem_switch_to_kernel_context(dev_priv);
+	if (ret)
+		goto err;
+
+	ret = i915_gem_wait_for_idle(dev_priv, true);
 	if (ret)
 		goto err;
 
 	i915_gem_retire_requests(dev_priv);
 
-	i915_gem_stop_engines(dev);
 	i915_gem_context_lost(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
 
@@ -5002,6 +4154,23 @@ err:
 	return ret;
 }
 
+void i915_gem_resume(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+
+	mutex_lock(&dev->struct_mutex);
+	i915_gem_restore_gtt_mappings(dev);
+
+	/* As we didn't flush the kernel context before suspend, we cannot
+	 * guarantee that the context image is complete. So let's just reset
+	 * it and start again.
+	 */
+	if (i915.enable_execlists)
+		intel_lr_context_reset(dev_priv, dev_priv->kernel_context);
+
+	mutex_unlock(&dev->struct_mutex);
+}
+
 void i915_gem_init_swizzling(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -5054,53 +4223,6 @@ static void init_unused_rings(struct drm_device *dev)
 	}
 }
 
-int i915_gem_init_engines(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
-
-	ret = intel_init_render_ring_buffer(dev);
-	if (ret)
-		return ret;
-
-	if (HAS_BSD(dev)) {
-		ret = intel_init_bsd_ring_buffer(dev);
-		if (ret)
-			goto cleanup_render_ring;
-	}
-
-	if (HAS_BLT(dev)) {
-		ret = intel_init_blt_ring_buffer(dev);
-		if (ret)
-			goto cleanup_bsd_ring;
-	}
-
-	if (HAS_VEBOX(dev)) {
-		ret = intel_init_vebox_ring_buffer(dev);
-		if (ret)
-			goto cleanup_blt_ring;
-	}
-
-	if (HAS_BSD2(dev)) {
-		ret = intel_init_bsd2_ring_buffer(dev);
-		if (ret)
-			goto cleanup_vebox_ring;
-	}
-
-	return 0;
-
-cleanup_vebox_ring:
-	intel_cleanup_engine(&dev_priv->engine[VECS]);
-cleanup_blt_ring:
-	intel_cleanup_engine(&dev_priv->engine[BCS]);
-cleanup_bsd_ring:
-	intel_cleanup_engine(&dev_priv->engine[VCS]);
-cleanup_render_ring:
-	intel_cleanup_engine(&dev_priv->engine[RCS]);
-
-	return ret;
-}
-
 int
 i915_gem_init_hw(struct drm_device *dev)
 {
@@ -5167,6 +4289,27 @@ out:
 	return ret;
 }
 
+bool intel_sanitize_semaphores(struct drm_i915_private *dev_priv, int value)
+{
+	if (INTEL_INFO(dev_priv)->gen < 6)
+		return false;
+
+	/* TODO: make semaphores and Execlists play nicely together */
+	if (i915.enable_execlists)
+		return false;
+
+	if (value >= 0)
+		return value;
+
+#ifdef CONFIG_INTEL_IOMMU
+	/* Enable semaphores on SNB when IO remapping is off */
+	if (INTEL_INFO(dev_priv)->gen == 6 && intel_iommu_gfx_mapped)
+		return false;
+#endif
+
+	return true;
+}
+
 int i915_gem_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -5175,15 +4318,9 @@ int i915_gem_init(struct drm_device *dev)
 	mutex_lock(&dev->struct_mutex);
 
 	if (!i915.enable_execlists) {
-		dev_priv->gt.execbuf_submit = i915_gem_ringbuffer_submission;
-		dev_priv->gt.init_engines = i915_gem_init_engines;
-		dev_priv->gt.cleanup_engine = intel_cleanup_engine;
-		dev_priv->gt.stop_engine = intel_stop_engine;
+		dev_priv->gt.cleanup_engine = intel_engine_cleanup;
 	} else {
-		dev_priv->gt.execbuf_submit = intel_execlists_submission;
-		dev_priv->gt.init_engines = intel_logical_rings_init;
 		dev_priv->gt.cleanup_engine = intel_logical_ring_cleanup;
-		dev_priv->gt.stop_engine = intel_logical_ring_stop;
 	}
 
 	/* This is just a security blanket to placate dragons.
@@ -5195,19 +4332,22 @@ int i915_gem_init(struct drm_device *dev)
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
 	i915_gem_init_userptr(dev_priv);
-	i915_gem_init_ggtt(dev);
+
+	ret = i915_gem_init_ggtt(dev_priv);
+	if (ret)
+		goto out_unlock;
 
 	ret = i915_gem_context_init(dev);
 	if (ret)
 		goto out_unlock;
 
-	ret = dev_priv->gt.init_engines(dev);
+	ret = intel_engines_init(dev);
 	if (ret)
 		goto out_unlock;
 
 	ret = i915_gem_init_hw(dev);
 	if (ret == -EIO) {
-		/* Allow ring initialisation to fail by marking the GPU as
+		/* Allow engine initialisation to fail by marking the GPU as
 		 * wedged. But we only want to do this where the GPU is angry,
 		 * for all other failure, such as an allocation failure, bail.
 		 */
@@ -5236,7 +4376,6 @@ i915_gem_cleanup_engines(struct drm_device *dev)
 static void
 init_engine_lists(struct intel_engine_cs *engine)
 {
-	INIT_LIST_HEAD(&engine->active_list);
 	INIT_LIST_HEAD(&engine->request_list);
 }
 
@@ -5283,10 +4422,11 @@ i915_gem_load_init(struct drm_device *dev)
 	dev_priv->requests =
 		kmem_cache_create("i915_gem_request",
 				  sizeof(struct drm_i915_gem_request), 0,
-				  SLAB_HWCACHE_ALIGN,
+				  SLAB_HWCACHE_ALIGN |
+				  SLAB_RECLAIM_ACCOUNT |
+				  SLAB_DESTROY_BY_RCU,
 				  NULL);
 
-	INIT_LIST_HEAD(&dev_priv->vm_list);
 	INIT_LIST_HEAD(&dev_priv->context_list);
 	INIT_LIST_HEAD(&dev_priv->mm.unbound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.bound_list);
@@ -5310,7 +4450,7 @@ i915_gem_load_init(struct drm_device *dev)
 
 	dev_priv->mm.interruptible = true;
 
-	mutex_init(&dev_priv->fb_tracking.lock);
+	spin_lock_init(&dev_priv->fb_tracking.lock);
 }
 
 void i915_gem_load_cleanup(struct drm_device *dev)
@@ -5320,6 +4460,9 @@ void i915_gem_load_cleanup(struct drm_device *dev)
 	kmem_cache_destroy(dev_priv->requests);
 	kmem_cache_destroy(dev_priv->vmas);
 	kmem_cache_destroy(dev_priv->objects);
+
+	/* And ensure that our DESTROY_BY_RCU slabs are truly destroyed */
+	rcu_barrier();
 }
 
 int i915_gem_freeze_late(struct drm_i915_private *dev_priv)
@@ -5353,21 +4496,15 @@ int i915_gem_freeze_late(struct drm_i915_private *dev_priv)
 void i915_gem_release(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct drm_i915_gem_request *request;
 
 	/* Clean up our request list when the client is going away, so that
 	 * later retire_requests won't dereference our soon-to-be-gone
 	 * file_priv.
 	 */
 	spin_lock(&file_priv->mm.lock);
-	while (!list_empty(&file_priv->mm.request_list)) {
-		struct drm_i915_gem_request *request;
-
-		request = list_first_entry(&file_priv->mm.request_list,
-					   struct drm_i915_gem_request,
-					   client_list);
-		list_del(&request->client_list);
+	list_for_each_entry(request, &file_priv->mm.request_list, client_list)
 		request->file_priv = NULL;
-	}
 	spin_unlock(&file_priv->mm.lock);
 
 	if (!list_empty(&file_priv->rps.link)) {
@@ -5396,7 +4533,7 @@ int i915_gem_open(struct drm_device *dev, struct drm_file *file)
 	spin_lock_init(&file_priv->mm.lock);
 	INIT_LIST_HEAD(&file_priv->mm.request_list);
 
-	file_priv->bsd_ring = -1;
+	file_priv->bsd_engine = -1;
 
 	ret = i915_gem_context_open(dev, file);
 	if (ret)
@@ -5418,16 +4555,23 @@ void i915_gem_track_fb(struct drm_i915_gem_object *old,
 		       struct drm_i915_gem_object *new,
 		       unsigned frontbuffer_bits)
 {
+	/* Control of individual bits within the mask are guarded by
+	 * the owning plane->mutex, i.e. we can never see concurrent
+	 * manipulation of individual bits. But since the bitfield as a whole
+	 * is updated using RMW, we need to use atomics in order to update
+	 * the bits.
+	 */
+	BUILD_BUG_ON(INTEL_FRONTBUFFER_BITS_PER_PIPE * I915_MAX_PIPES >
+		     sizeof(atomic_t) * BITS_PER_BYTE);
+
 	if (old) {
-		WARN_ON(!mutex_is_locked(&old->base.dev->struct_mutex));
-		WARN_ON(!(old->frontbuffer_bits & frontbuffer_bits));
-		old->frontbuffer_bits &= ~frontbuffer_bits;
+		WARN_ON(!(atomic_read(&old->frontbuffer_bits) & frontbuffer_bits));
+		atomic_andnot(frontbuffer_bits, &old->frontbuffer_bits);
 	}
 
 	if (new) {
-		WARN_ON(!mutex_is_locked(&new->base.dev->struct_mutex));
-		WARN_ON(new->frontbuffer_bits & frontbuffer_bits);
-		new->frontbuffer_bits |= frontbuffer_bits;
+		WARN_ON(atomic_read(&new->frontbuffer_bits) & frontbuffer_bits);
+		atomic_or(frontbuffer_bits, &new->frontbuffer_bits);
 	}
 }
 
@@ -5441,7 +4585,7 @@ u64 i915_gem_obj_offset(struct drm_i915_gem_object *o,
 	WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base);
 
 	list_for_each_entry(vma, &o->vma_list, obj_link) {
-		if (vma->is_ggtt &&
+		if (i915_vma_is_ggtt(vma) &&
 		    vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
 			continue;
 		if (vma->vm == vm)
@@ -5459,7 +4603,8 @@ u64 i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, &o->vma_list, obj_link)
-		if (vma->is_ggtt && i915_ggtt_view_equal(&vma->ggtt_view, view))
+		if (i915_vma_is_ggtt(vma) &&
+		    i915_ggtt_view_equal(&vma->ggtt_view, view))
 			return vma->node.start;
 
 	WARN(1, "global vma for this object not found. (view=%u)\n", view->type);
@@ -5472,7 +4617,7 @@ bool i915_gem_obj_bound(struct drm_i915_gem_object *o,
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, &o->vma_list, obj_link) {
-		if (vma->is_ggtt &&
+		if (i915_vma_is_ggtt(vma) &&
 		    vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
 			continue;
 		if (vma->vm == vm && drm_mm_node_allocated(&vma->node))
@@ -5488,7 +4633,7 @@ bool i915_gem_obj_ggtt_bound_view(struct drm_i915_gem_object *o,
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, &o->vma_list, obj_link)
-		if (vma->is_ggtt &&
+		if (i915_vma_is_ggtt(vma) &&
 		    i915_ggtt_view_equal(&vma->ggtt_view, view) &&
 		    drm_mm_node_allocated(&vma->node))
 			return true;
@@ -5496,17 +4641,6 @@ bool i915_gem_obj_ggtt_bound_view(struct drm_i915_gem_object *o,
 	return false;
 }
 
-bool i915_gem_obj_bound_any(struct drm_i915_gem_object *o)
-{
-	struct i915_vma *vma;
-
-	list_for_each_entry(vma, &o->vma_list, obj_link)
-		if (drm_mm_node_allocated(&vma->node))
-			return true;
-
-	return false;
-}
-
 unsigned long i915_gem_obj_ggtt_size(struct drm_i915_gem_object *o)
 {
 	struct i915_vma *vma;
@@ -5514,7 +4648,7 @@ unsigned long i915_gem_obj_ggtt_size(struct drm_i915_gem_object *o)
 	GEM_BUG_ON(list_empty(&o->vma_list));
 
 	list_for_each_entry(vma, &o->vma_list, obj_link) {
-		if (vma->is_ggtt &&
+		if (i915_vma_is_ggtt(vma) &&
 		    vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL)
 			return vma->node.size;
 	}
@@ -5526,7 +4660,7 @@ bool i915_gem_obj_is_pinned(struct drm_i915_gem_object *obj)
 {
 	struct i915_vma *vma;
 	list_for_each_entry(vma, &obj->vma_list, obj_link)
-		if (vma->pin_count > 0)
+		if (i915_vma_is_pinned(vma))
 			return true;
 
 	return false;
@@ -5584,6 +4718,6 @@ i915_gem_object_create_from_data(struct drm_device *dev,
 	return obj;
 
 fail:
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	return ERR_PTR(ret);
 }

drivers/gpu/drm/i915/i915_gem_batch_pool.c

@@ -41,15 +41,15 @@
 
 /**
  * i915_gem_batch_pool_init() - initialize a batch buffer pool
- * @dev: the drm device
+ * @engine: the associated request submission engine
  * @pool: the batch buffer pool
  */
-void i915_gem_batch_pool_init(struct drm_device *dev,
+void i915_gem_batch_pool_init(struct intel_engine_cs *engine,
 			      struct i915_gem_batch_pool *pool)
 {
 	int n;
 
-	pool->dev = dev;
+	pool->engine = engine;
 
 	for (n = 0; n < ARRAY_SIZE(pool->cache_list); n++)
 		INIT_LIST_HEAD(&pool->cache_list[n]);
@@ -65,18 +65,17 @@ void i915_gem_batch_pool_fini(struct i915_gem_batch_pool *pool)
 {
 	int n;
 
-	WARN_ON(!mutex_is_locked(&pool->dev->struct_mutex));
+	lockdep_assert_held(&pool->engine->i915->drm.struct_mutex);
 
 	for (n = 0; n < ARRAY_SIZE(pool->cache_list); n++) {
-		while (!list_empty(&pool->cache_list[n])) {
-			struct drm_i915_gem_object *obj =
-				list_first_entry(&pool->cache_list[n],
-						 struct drm_i915_gem_object,
-						 batch_pool_link);
-
-			list_del(&obj->batch_pool_link);
-			drm_gem_object_unreference(&obj->base);
-		}
+		struct drm_i915_gem_object *obj, *next;
+
+		list_for_each_entry_safe(obj, next,
+					 &pool->cache_list[n],
+					 batch_pool_link)
+			i915_gem_object_put(obj);
+
+		INIT_LIST_HEAD(&pool->cache_list[n]);
 	}
 }
 
@@ -102,7 +101,7 @@ i915_gem_batch_pool_get(struct i915_gem_batch_pool *pool,
 	struct list_head *list;
 	int n;
 
-	WARN_ON(!mutex_is_locked(&pool->dev->struct_mutex));
+	lockdep_assert_held(&pool->engine->i915->drm.struct_mutex);
 
 	/* Compute a power-of-two bucket, but throw everything greater than
 	 * 16KiB into the same bucket: i.e. the the buckets hold objects of
@@ -115,13 +114,14 @@ i915_gem_batch_pool_get(struct i915_gem_batch_pool *pool,
 
 	list_for_each_entry_safe(tmp, next, list, batch_pool_link) {
 		/* The batches are strictly LRU ordered */
-		if (tmp->active)
+		if (!i915_gem_active_is_idle(&tmp->last_read[pool->engine->id],
+					     &tmp->base.dev->struct_mutex))
 			break;
 
 		/* While we're looping, do some clean up */
 		if (tmp->madv == __I915_MADV_PURGED) {
 			list_del(&tmp->batch_pool_link);
-			drm_gem_object_unreference(&tmp->base);
+			i915_gem_object_put(tmp);
 			continue;
 		}
 
@@ -134,7 +134,7 @@ i915_gem_batch_pool_get(struct i915_gem_batch_pool *pool,
 	if (obj == NULL) {
 		int ret;
 
-		obj = i915_gem_object_create(pool->dev, size);
+		obj = i915_gem_object_create(&pool->engine->i915->drm, size);
 		if (IS_ERR(obj))
 			return obj;
 

drivers/gpu/drm/i915/i915_gem_batch_pool.h

@@ -27,13 +27,15 @@
 
 #include "i915_drv.h"
 
+struct intel_engine_cs;
+
 struct i915_gem_batch_pool {
-	struct drm_device *dev;
+	struct intel_engine_cs *engine;
 	struct list_head cache_list[4];
 };
 
 /* i915_gem_batch_pool.c */
-void i915_gem_batch_pool_init(struct drm_device *dev,
+void i915_gem_batch_pool_init(struct intel_engine_cs *engine,
 			      struct i915_gem_batch_pool *pool);
 void i915_gem_batch_pool_fini(struct i915_gem_batch_pool *pool);
 struct drm_i915_gem_object*

drivers/gpu/drm/i915/i915_gem_context.c

@@ -134,21 +134,6 @@ static int get_context_size(struct drm_i915_private *dev_priv)
 	return ret;
 }
 
-static void i915_gem_context_clean(struct i915_gem_context *ctx)
-{
-	struct i915_hw_ppgtt *ppgtt = ctx->ppgtt;
-	struct i915_vma *vma, *next;
-
-	if (!ppgtt)
-		return;
-
-	list_for_each_entry_safe(vma, next, &ppgtt->base.inactive_list,
-				 vm_link) {
-		if (WARN_ON(__i915_vma_unbind_no_wait(vma)))
-			break;
-	}
-}
-
 void i915_gem_context_free(struct kref *ctx_ref)
 {
 	struct i915_gem_context *ctx = container_of(ctx_ref, typeof(*ctx), ref);
@@ -156,13 +141,7 @@ void i915_gem_context_free(struct kref *ctx_ref)
 
 	lockdep_assert_held(&ctx->i915->drm.struct_mutex);
 	trace_i915_context_free(ctx);
-
-	/*
-	 * This context is going away and we need to remove all VMAs still
-	 * around. This is to handle imported shared objects for which
-	 * destructor did not run when their handles were closed.
-	 */
-	i915_gem_context_clean(ctx);
+	GEM_BUG_ON(!ctx->closed);
 
 	i915_ppgtt_put(ctx->ppgtt);
 
@@ -173,10 +152,10 @@ void i915_gem_context_free(struct kref *ctx_ref)
 			continue;
 
 		WARN_ON(ce->pin_count);
-		if (ce->ringbuf)
-			intel_ringbuffer_free(ce->ringbuf);
+		if (ce->ring)
+			intel_ring_free(ce->ring);
 
-		drm_gem_object_unreference(&ce->state->base);
+		i915_gem_object_put(ce->state);
 	}
 
 	list_del(&ctx->link);
@@ -216,7 +195,7 @@ i915_gem_alloc_context_obj(struct drm_device *dev, size_t size)
 		ret = i915_gem_object_set_cache_level(obj, I915_CACHE_L3_LLC);
 		/* Failure shouldn't ever happen this early */
 		if (WARN_ON(ret)) {
-			drm_gem_object_unreference(&obj->base);
+			i915_gem_object_put(obj);
 			return ERR_PTR(ret);
 		}
 	}
@@ -224,6 +203,37 @@ i915_gem_alloc_context_obj(struct drm_device *dev, size_t size)
 	return obj;
 }
 
+static void i915_ppgtt_close(struct i915_address_space *vm)
+{
+	struct list_head *phases[] = {
+		&vm->active_list,
+		&vm->inactive_list,
+		&vm->unbound_list,
+		NULL,
+	}, **phase;
+
+	GEM_BUG_ON(vm->closed);
+	vm->closed = true;
+
+	for (phase = phases; *phase; phase++) {
+		struct i915_vma *vma, *vn;
+
+		list_for_each_entry_safe(vma, vn, *phase, vm_link)
+			if (!i915_vma_is_closed(vma))
+				i915_vma_close(vma);
+	}
+}
+
+static void context_close(struct i915_gem_context *ctx)
+{
+	GEM_BUG_ON(ctx->closed);
+	ctx->closed = true;
+	if (ctx->ppgtt)
+		i915_ppgtt_close(&ctx->ppgtt->base);
+	ctx->file_priv = ERR_PTR(-EBADF);
+	i915_gem_context_put(ctx);
+}
+
 static int assign_hw_id(struct drm_i915_private *dev_priv, unsigned *out)
 {
 	int ret;
@@ -305,7 +315,7 @@ __create_hw_context(struct drm_device *dev,
 	return ctx;
 
 err_out:
-	i915_gem_context_unreference(ctx);
+	context_close(ctx);
 	return ERR_PTR(ret);
 }
 
@@ -327,13 +337,14 @@ i915_gem_create_context(struct drm_device *dev,
 		return ctx;
 
 	if (USES_FULL_PPGTT(dev)) {
-		struct i915_hw_ppgtt *ppgtt = i915_ppgtt_create(dev, file_priv);
+		struct i915_hw_ppgtt *ppgtt =
+			i915_ppgtt_create(to_i915(dev), file_priv);
 
 		if (IS_ERR(ppgtt)) {
 			DRM_DEBUG_DRIVER("PPGTT setup failed (%ld)\n",
 					 PTR_ERR(ppgtt));
 			idr_remove(&file_priv->context_idr, ctx->user_handle);
-			i915_gem_context_unreference(ctx);
+			context_close(ctx);
 			return ERR_CAST(ppgtt);
 		}
 
@@ -390,7 +401,7 @@ static void i915_gem_context_unpin(struct i915_gem_context *ctx,
 		if (ce->state)
 			i915_gem_object_ggtt_unpin(ce->state);
 
-		i915_gem_context_unreference(ctx);
+		i915_gem_context_put(ctx);
 	}
 }
 
@@ -504,7 +515,7 @@ void i915_gem_context_fini(struct drm_device *dev)
 
 	lockdep_assert_held(&dev->struct_mutex);
 
-	i915_gem_context_unreference(dctx);
+	context_close(dctx);
 	dev_priv->kernel_context = NULL;
 
 	ida_destroy(&dev_priv->context_hw_ida);
@@ -514,8 +525,7 @@ static int context_idr_cleanup(int id, void *p, void *data)
 {
 	struct i915_gem_context *ctx = p;
 
-	ctx->file_priv = ERR_PTR(-EBADF);
-	i915_gem_context_unreference(ctx);
+	context_close(ctx);
 	return 0;
 }
 
@@ -552,11 +562,12 @@ static inline int
 mi_set_context(struct drm_i915_gem_request *req, u32 hw_flags)
 {
 	struct drm_i915_private *dev_priv = req->i915;
+	struct intel_ring *ring = req->ring;
 	struct intel_engine_cs *engine = req->engine;
 	u32 flags = hw_flags | MI_MM_SPACE_GTT;
 	const int num_rings =
 		/* Use an extended w/a on ivb+ if signalling from other rings */
-		i915_semaphore_is_enabled(dev_priv) ?
+		i915.semaphores ?
 		hweight32(INTEL_INFO(dev_priv)->ring_mask) - 1 :
 		0;
 	int len, ret;
@@ -567,7 +578,7 @@ mi_set_context(struct drm_i915_gem_request *req, u32 hw_flags)
 	 * itlb_before_ctx_switch.
 	 */
 	if (IS_GEN6(dev_priv)) {
-		ret = engine->flush(req, I915_GEM_GPU_DOMAINS, 0);
+		ret = engine->emit_flush(req, EMIT_INVALIDATE);
 		if (ret)
 			return ret;
 	}
@@ -589,64 +600,64 @@ mi_set_context(struct drm_i915_gem_request *req, u32 hw_flags)
 
 	/* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw,bdw,chv */
 	if (INTEL_GEN(dev_priv) >= 7) {
-		intel_ring_emit(engine, MI_ARB_ON_OFF | MI_ARB_DISABLE);
+		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE);
 		if (num_rings) {
 			struct intel_engine_cs *signaller;
 
-			intel_ring_emit(engine,
+			intel_ring_emit(ring,
 					MI_LOAD_REGISTER_IMM(num_rings));
 			for_each_engine(signaller, dev_priv) {
 				if (signaller == engine)
 					continue;
 
-				intel_ring_emit_reg(engine,
+				intel_ring_emit_reg(ring,
 						    RING_PSMI_CTL(signaller->mmio_base));
-				intel_ring_emit(engine,
+				intel_ring_emit(ring,
 						_MASKED_BIT_ENABLE(GEN6_PSMI_SLEEP_MSG_DISABLE));
 			}
 		}
 	}
 
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_emit(engine, MI_SET_CONTEXT);
-	intel_ring_emit(engine,
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_emit(ring, MI_SET_CONTEXT);
+	intel_ring_emit(ring,
 			i915_gem_obj_ggtt_offset(req->ctx->engine[RCS].state) |
 			flags);
 	/*
 	 * w/a: MI_SET_CONTEXT must always be followed by MI_NOOP
 	 * WaMiSetContext_Hang:snb,ivb,vlv
 	 */
-	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_emit(ring, MI_NOOP);
 
 	if (INTEL_GEN(dev_priv) >= 7) {
 		if (num_rings) {
 			struct intel_engine_cs *signaller;
 			i915_reg_t last_reg = {}; /* keep gcc quiet */
 
-			intel_ring_emit(engine,
+			intel_ring_emit(ring,
 					MI_LOAD_REGISTER_IMM(num_rings));
 			for_each_engine(signaller, dev_priv) {
 				if (signaller == engine)
 					continue;
 
 				last_reg = RING_PSMI_CTL(signaller->mmio_base);
-				intel_ring_emit_reg(engine, last_reg);
-				intel_ring_emit(engine,
+				intel_ring_emit_reg(ring, last_reg);
+				intel_ring_emit(ring,
 						_MASKED_BIT_DISABLE(GEN6_PSMI_SLEEP_MSG_DISABLE));
 			}
 
 			/* Insert a delay before the next switch! */
-			intel_ring_emit(engine,
+			intel_ring_emit(ring,
 					MI_STORE_REGISTER_MEM |
 					MI_SRM_LRM_GLOBAL_GTT);
-			intel_ring_emit_reg(engine, last_reg);
-			intel_ring_emit(engine, engine->scratch.gtt_offset);
-			intel_ring_emit(engine, MI_NOOP);
+			intel_ring_emit_reg(ring, last_reg);
+			intel_ring_emit(ring, engine->scratch.gtt_offset);
+			intel_ring_emit(ring, MI_NOOP);
 		}
-		intel_ring_emit(engine, MI_ARB_ON_OFF | MI_ARB_ENABLE);
+		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE);
 	}
 
-	intel_ring_advance(engine);
+	intel_ring_advance(ring);
 
 	return ret;
 }
@@ -654,7 +665,7 @@ mi_set_context(struct drm_i915_gem_request *req, u32 hw_flags)
 static int remap_l3(struct drm_i915_gem_request *req, int slice)
 {
 	u32 *remap_info = req->i915->l3_parity.remap_info[slice];
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int i, ret;
 
 	if (!remap_info)
@@ -669,13 +680,13 @@ static int remap_l3(struct drm_i915_gem_request *req, int slice)
 	 * here because no other code should access these registers other than
 	 * at initialization time.
 	 */
-	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(GEN7_L3LOG_SIZE/4));
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(GEN7_L3LOG_SIZE/4));
 	for (i = 0; i < GEN7_L3LOG_SIZE/4; i++) {
-		intel_ring_emit_reg(engine, GEN7_L3LOG(slice, i));
-		intel_ring_emit(engine, remap_info[i]);
+		intel_ring_emit_reg(ring, GEN7_L3LOG(slice, i));
+		intel_ring_emit(ring, remap_info[i]);
 	}
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -752,9 +763,8 @@ static int do_rcs_switch(struct drm_i915_gem_request *req)
 		return 0;
 
 	/* Trying to pin first makes error handling easier. */
-	ret = i915_gem_obj_ggtt_pin(to->engine[RCS].state,
-				    to->ggtt_alignment,
-				    0);
+	ret = i915_gem_object_ggtt_pin(to->engine[RCS].state, NULL, 0,
+				       to->ggtt_alignment, 0);
 	if (ret)
 		return ret;
 
@@ -814,8 +824,8 @@ static int do_rcs_switch(struct drm_i915_gem_request *req)
 	 * MI_SET_CONTEXT instead of when the next seqno has completed.
 	 */
 	if (from != NULL) {
-		from->engine[RCS].state->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION;
-		i915_vma_move_to_active(i915_gem_obj_to_ggtt(from->engine[RCS].state), req);
+		struct drm_i915_gem_object *obj = from->engine[RCS].state;
+
 		/* As long as MI_SET_CONTEXT is serializing, ie. it flushes the
 		 * whole damn pipeline, we don't need to explicitly mark the
 		 * object dirty. The only exception is that the context must be
@@ -823,14 +833,14 @@ static int do_rcs_switch(struct drm_i915_gem_request *req)
 		 * able to defer doing this until we know the object would be
 		 * swapped, but there is no way to do that yet.
 		 */
-		from->engine[RCS].state->dirty = 1;
+		obj->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION;
+		i915_vma_move_to_active(i915_gem_obj_to_ggtt(obj), req, 0);
 
 		/* obj is kept alive until the next request by its active ref */
-		i915_gem_object_ggtt_unpin(from->engine[RCS].state);
-		i915_gem_context_unreference(from);
+		i915_gem_object_ggtt_unpin(obj);
+		i915_gem_context_put(from);
 	}
-	i915_gem_context_reference(to);
-	engine->last_context = to;
+	engine->last_context = i915_gem_context_get(to);
 
 	/* GEN8 does *not* require an explicit reload if the PDPs have been
 	 * setup, and we do not wish to move them.
@@ -894,8 +904,9 @@ int i915_switch_context(struct drm_i915_gem_request *req)
 {
 	struct intel_engine_cs *engine = req->engine;
 
-	WARN_ON(i915.enable_execlists);
 	lockdep_assert_held(&req->i915->drm.struct_mutex);
+	if (i915.enable_execlists)
+		return 0;
 
 	if (!req->ctx->engine[engine->id].state) {
 		struct i915_gem_context *to = req->ctx;
@@ -914,10 +925,9 @@ int i915_switch_context(struct drm_i915_gem_request *req)
 		}
 
 		if (to != engine->last_context) {
-			i915_gem_context_reference(to);
 			if (engine->last_context)
-				i915_gem_context_unreference(engine->last_context);
-			engine->last_context = to;
+				i915_gem_context_put(engine->last_context);
+			engine->last_context = i915_gem_context_get(to);
 		}
 
 		return 0;
@@ -926,6 +936,33 @@ int i915_switch_context(struct drm_i915_gem_request *req)
 	return do_rcs_switch(req);
 }
 
+int i915_gem_switch_to_kernel_context(struct drm_i915_private *dev_priv)
+{
+	struct intel_engine_cs *engine;
+
+	for_each_engine(engine, dev_priv) {
+		struct drm_i915_gem_request *req;
+		int ret;
+
+		if (engine->last_context == NULL)
+			continue;
+
+		if (engine->last_context == dev_priv->kernel_context)
+			continue;
+
+		req = i915_gem_request_alloc(engine, dev_priv->kernel_context);
+		if (IS_ERR(req))
+			return PTR_ERR(req);
+
+		ret = i915_switch_context(req);
+		i915_add_request_no_flush(req);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static bool contexts_enabled(struct drm_device *dev)
 {
 	return i915.enable_execlists || to_i915(dev)->hw_context_size;
@@ -985,7 +1022,7 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 	}
 
 	idr_remove(&file_priv->context_idr, ctx->user_handle);
-	i915_gem_context_unreference(ctx);
+	context_close(ctx);
 	mutex_unlock(&dev->struct_mutex);
 
 	DRM_DEBUG_DRIVER("HW context %d destroyed\n", args->ctx_id);

drivers/gpu/drm/i915/i915_gem_debug.c

@@ -1,70 +0,0 @@
-/*
- * Copyright © 2008 Intel Corporation
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice (including the next
- * paragraph) shall be included in all copies or substantial portions of the
- * Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
- * IN THE SOFTWARE.
- *
- * Authors:
- *    Keith Packard <keithp@keithp.com>
- *
- */
-
-#include <drm/drmP.h>
-#include <drm/i915_drm.h>
-#include "i915_drv.h"
-
-#if WATCH_LISTS
-int
-i915_verify_lists(struct drm_device *dev)
-{
-	static int warned;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_i915_gem_object *obj;
-	struct intel_engine_cs *engine;
-	int err = 0;
-
-	if (warned)
-		return 0;
-
-	for_each_engine(engine, dev_priv) {
-		list_for_each_entry(obj, &engine->active_list,
-				    engine_list[engine->id]) {
-			if (obj->base.dev != dev ||
-			    !atomic_read(&obj->base.refcount.refcount)) {
-				DRM_ERROR("%s: freed active obj %p\n",
-					  engine->name, obj);
-				err++;
-				break;
-			} else if (!obj->active ||
-				   obj->last_read_req[engine->id] == NULL) {
-				DRM_ERROR("%s: invalid active obj %p\n",
-					  engine->name, obj);
-				err++;
-			} else if (obj->base.write_domain) {
-				DRM_ERROR("%s: invalid write obj %p (w %x)\n",
-					  engine->name,
-					  obj, obj->base.write_domain);
-				err++;
-			}
-		}
-	}
-
-	return warned = err;
-}
-#endif /* WATCH_LIST */

drivers/gpu/drm/i915/i915_gem_dmabuf.c

@@ -23,9 +23,13 @@
  * Authors:
  *	Dave Airlie <airlied@redhat.com>
  */
+
+#include <linux/dma-buf.h>
+#include <linux/reservation.h>
+
 #include <drm/drmP.h>
+
 #include "i915_drv.h"
-#include <linux/dma-buf.h>
 
 static struct drm_i915_gem_object *dma_buf_to_obj(struct dma_buf *buf)
 {
@@ -218,25 +222,73 @@ static const struct dma_buf_ops i915_dmabuf_ops =  {
 	.end_cpu_access = i915_gem_end_cpu_access,
 };
 
+static void export_fences(struct drm_i915_gem_object *obj,
+			  struct dma_buf *dma_buf)
+{
+	struct reservation_object *resv = dma_buf->resv;
+	struct drm_i915_gem_request *req;
+	unsigned long active;
+	int idx;
+
+	active = __I915_BO_ACTIVE(obj);
+	if (!active)
+		return;
+
+	/* Serialise with execbuf to prevent concurrent fence-loops */
+	mutex_lock(&obj->base.dev->struct_mutex);
+
+	/* Mark the object for future fences before racily adding old fences */
+	obj->base.dma_buf = dma_buf;
+
+	ww_mutex_lock(&resv->lock, NULL);
+
+	for_each_active(active, idx) {
+		req = i915_gem_active_get(&obj->last_read[idx],
+					  &obj->base.dev->struct_mutex);
+		if (!req)
+			continue;
+
+		if (reservation_object_reserve_shared(resv) == 0)
+			reservation_object_add_shared_fence(resv, &req->fence);
+
+		i915_gem_request_put(req);
+	}
+
+	req = i915_gem_active_get(&obj->last_write,
+				  &obj->base.dev->struct_mutex);
+	if (req) {
+		reservation_object_add_excl_fence(resv, &req->fence);
+		i915_gem_request_put(req);
+	}
+
+	ww_mutex_unlock(&resv->lock);
+	mutex_unlock(&obj->base.dev->struct_mutex);
+}
+
 struct dma_buf *i915_gem_prime_export(struct drm_device *dev,
 				      struct drm_gem_object *gem_obj, int flags)
 {
 	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
 	DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
+	struct dma_buf *dma_buf;
 
 	exp_info.ops = &i915_dmabuf_ops;
 	exp_info.size = gem_obj->size;
 	exp_info.flags = flags;
 	exp_info.priv = gem_obj;
 
-
 	if (obj->ops->dmabuf_export) {
 		int ret = obj->ops->dmabuf_export(obj);
 		if (ret)
 			return ERR_PTR(ret);
 	}
 
-	return dma_buf_export(&exp_info);
+	dma_buf = dma_buf_export(&exp_info);
+	if (IS_ERR(dma_buf))
+		return dma_buf;
+
+	export_fences(obj, dma_buf);
+	return dma_buf;
 }
 
 static int i915_gem_object_get_pages_dmabuf(struct drm_i915_gem_object *obj)
@@ -278,8 +330,7 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev,
 			 * Importing dmabuf exported from out own gem increases
 			 * refcount on gem itself instead of f_count of dmabuf.
 			 */
-			drm_gem_object_reference(&obj->base);
-			return &obj->base;
+			return &i915_gem_object_get(obj)->base;
 		}
 	}
 
@@ -300,6 +351,16 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev,
 	i915_gem_object_init(obj, &i915_gem_object_dmabuf_ops);
 	obj->base.import_attach = attach;
 
+	/* We use GTT as shorthand for a coherent domain, one that is
+	 * neither in the GPU cache nor in the CPU cache, where all
+	 * writes are immediately visible in memory. (That's not strictly
+	 * true, but it's close! There are internal buffers such as the
+	 * write-combined buffer or a delay through the chipset for GTT
+	 * writes that do require us to treat GTT as a separate cache domain.)
+	 */
+	obj->base.read_domains = I915_GEM_DOMAIN_GTT;
+	obj->base.write_domain = 0;
+
 	return &obj->base;
 
 fail_detach:

drivers/gpu/drm/i915/i915_gem_evict.c

@@ -33,41 +33,23 @@
 #include "intel_drv.h"
 #include "i915_trace.h"
 
-static int switch_to_pinned_context(struct drm_i915_private *dev_priv)
+static bool
+gpu_is_idle(struct drm_i915_private *dev_priv)
 {
 	struct intel_engine_cs *engine;
 
-	if (i915.enable_execlists)
-		return 0;
-
 	for_each_engine(engine, dev_priv) {
-		struct drm_i915_gem_request *req;
-		int ret;
-
-		if (engine->last_context == NULL)
-			continue;
-
-		if (engine->last_context == dev_priv->kernel_context)
-			continue;
-
-		req = i915_gem_request_alloc(engine, dev_priv->kernel_context);
-		if (IS_ERR(req))
-			return PTR_ERR(req);
-
-		ret = i915_switch_context(req);
-		i915_add_request_no_flush(req);
-		if (ret)
-			return ret;
+		if (intel_engine_is_active(engine))
+			return false;
 	}
 
-	return 0;
+	return true;
 }
 
-
 static bool
 mark_free(struct i915_vma *vma, struct list_head *unwind)
 {
-	if (vma->pin_count)
+	if (i915_vma_is_pinned(vma))
 		return false;
 
 	if (WARN_ON(!list_empty(&vma->exec_list)))
@@ -79,7 +61,6 @@ mark_free(struct i915_vma *vma, struct list_head *unwind)
 
 /**
  * i915_gem_evict_something - Evict vmas to make room for binding a new one
- * @dev: drm_device
  * @vm: address space to evict from
  * @min_size: size of the desired free space
  * @alignment: alignment constraint of the desired free space
@@ -102,42 +83,37 @@ mark_free(struct i915_vma *vma, struct list_head *unwind)
  * memory in e.g. the shrinker.
  */
 int
-i915_gem_evict_something(struct drm_device *dev, struct i915_address_space *vm,
-			 int min_size, unsigned alignment, unsigned cache_level,
-			 unsigned long start, unsigned long end,
+i915_gem_evict_something(struct i915_address_space *vm,
+			 u64 min_size, u64 alignment,
+			 unsigned cache_level,
+			 u64 start, u64 end,
 			 unsigned flags)
 {
-	struct list_head eviction_list, unwind_list;
-	struct i915_vma *vma;
-	int ret = 0;
-	int pass = 0;
+	struct drm_i915_private *dev_priv = to_i915(vm->dev);
+	struct list_head eviction_list;
+	struct list_head *phases[] = {
+		&vm->inactive_list,
+		&vm->active_list,
+		NULL,
+	}, **phase;
+	struct i915_vma *vma, *next;
+	int ret;
 
-	trace_i915_gem_evict(dev, min_size, alignment, flags);
+	trace_i915_gem_evict(vm, min_size, alignment, flags);
 
 	/*
 	 * The goal is to evict objects and amalgamate space in LRU order.
 	 * The oldest idle objects reside on the inactive list, which is in
-	 * retirement order. The next objects to retire are those on the (per
-	 * ring) active list that do not have an outstanding flush. Once the
-	 * hardware reports completion (the seqno is updated after the
-	 * batchbuffer has been finished) the clean buffer objects would
-	 * be retired to the inactive list. Any dirty objects would be added
-	 * to the tail of the flushing list. So after processing the clean
-	 * active objects we need to emit a MI_FLUSH to retire the flushing
-	 * list, hence the retirement order of the flushing list is in
-	 * advance of the dirty objects on the active lists.
+	 * retirement order. The next objects to retire are those in flight,
+	 * on the active list, again in retirement order.
 	 *
 	 * The retirement sequence is thus:
 	 *   1. Inactive objects (already retired)
-	 *   2. Clean active objects
-	 *   3. Flushing list
-	 *   4. Dirty active objects.
+	 *   2. Active objects (will stall on unbinding)
 	 *
 	 * On each list, the oldest objects lie at the HEAD with the freshest
 	 * object on the TAIL.
 	 */
-
-	INIT_LIST_HEAD(&unwind_list);
 	if (start != 0 || end != vm->total) {
 		drm_mm_init_scan_with_range(&vm->mm, min_size,
 					    alignment, cache_level,
@@ -145,96 +121,84 @@ i915_gem_evict_something(struct drm_device *dev, struct i915_address_space *vm,
 	} else
 		drm_mm_init_scan(&vm->mm, min_size, alignment, cache_level);
 
-search_again:
-	/* First see if there is a large enough contiguous idle region... */
-	list_for_each_entry(vma, &vm->inactive_list, vm_link) {
-		if (mark_free(vma, &unwind_list))
-			goto found;
-	}
-
 	if (flags & PIN_NONBLOCK)
-		goto none;
+		phases[1] = NULL;
 
-	/* Now merge in the soon-to-be-expired objects... */
-	list_for_each_entry(vma, &vm->active_list, vm_link) {
-		if (mark_free(vma, &unwind_list))
-			goto found;
-	}
+search_again:
+	INIT_LIST_HEAD(&eviction_list);
+	phase = phases;
+	do {
+		list_for_each_entry(vma, *phase, vm_link)
+			if (mark_free(vma, &eviction_list))
+				goto found;
+	} while (*++phase);
 
-none:
 	/* Nothing found, clean up and bail out! */
-	while (!list_empty(&unwind_list)) {
-		vma = list_first_entry(&unwind_list,
-				       struct i915_vma,
-				       exec_list);
+	list_for_each_entry_safe(vma, next, &eviction_list, exec_list) {
 		ret = drm_mm_scan_remove_block(&vma->node);
 		BUG_ON(ret);
 
-		list_del_init(&vma->exec_list);
+		INIT_LIST_HEAD(&vma->exec_list);
 	}
 
 	/* Can we unpin some objects such as idle hw contents,
-	 * or pending flips?
+	 * or pending flips? But since only the GGTT has global entries
+	 * such as scanouts, rinbuffers and contexts, we can skip the
+	 * purge when inspecting per-process local address spaces.
 	 */
-	if (flags & PIN_NONBLOCK)
+	if (!i915_is_ggtt(vm) || flags & PIN_NONBLOCK)
 		return -ENOSPC;
 
-	/* Only idle the GPU and repeat the search once */
-	if (pass++ == 0) {
-		struct drm_i915_private *dev_priv = to_i915(dev);
-
-		if (i915_is_ggtt(vm)) {
-			ret = switch_to_pinned_context(dev_priv);
-			if (ret)
-				return ret;
-		}
-
-		ret = i915_gem_wait_for_idle(dev_priv);
-		if (ret)
-			return ret;
-
-		i915_gem_retire_requests(dev_priv);
-		goto search_again;
+	if (gpu_is_idle(dev_priv)) {
+		/* If we still have pending pageflip completions, drop
+		 * back to userspace to give our workqueues time to
+		 * acquire our locks and unpin the old scanouts.
+		 */
+		return intel_has_pending_fb_unpin(vm->dev) ? -EAGAIN : -ENOSPC;
 	}
 
-	/* If we still have pending pageflip completions, drop
-	 * back to userspace to give our workqueues time to
-	 * acquire our locks and unpin the old scanouts.
+	/* Not everything in the GGTT is tracked via vma (otherwise we
+	 * could evict as required with minimal stalling) so we are forced
+	 * to idle the GPU and explicitly retire outstanding requests in
+	 * the hopes that we can then remove contexts and the like only
+	 * bound by their active reference.
 	 */
-	return intel_has_pending_fb_unpin(dev) ? -EAGAIN : -ENOSPC;
+	ret = i915_gem_switch_to_kernel_context(dev_priv);
+	if (ret)
+		return ret;
+
+	ret = i915_gem_wait_for_idle(dev_priv, true);
+	if (ret)
+		return ret;
+
+	i915_gem_retire_requests(dev_priv);
+	goto search_again;
 
 found:
 	/* drm_mm doesn't allow any other other operations while
-	 * scanning, therefore store to be evicted objects on a
-	 * temporary list. */
-	INIT_LIST_HEAD(&eviction_list);
-	while (!list_empty(&unwind_list)) {
-		vma = list_first_entry(&unwind_list,
-				       struct i915_vma,
-				       exec_list);
-		if (drm_mm_scan_remove_block(&vma->node)) {
-			list_move(&vma->exec_list, &eviction_list);
-			drm_gem_object_reference(&vma->obj->base);
-			continue;
-		}
-		list_del_init(&vma->exec_list);
+	 * scanning, therefore store to-be-evicted objects on a
+	 * temporary list and take a reference for all before
+	 * calling unbind (which may remove the active reference
+	 * of any of our objects, thus corrupting the list).
+	 */
+	list_for_each_entry_safe(vma, next, &eviction_list, exec_list) {
+		if (drm_mm_scan_remove_block(&vma->node))
+			__i915_vma_pin(vma);
+		else
+			list_del_init(&vma->exec_list);
 	}
 
 	/* Unbinding will emit any required flushes */
 	while (!list_empty(&eviction_list)) {
-		struct drm_gem_object *obj;
 		vma = list_first_entry(&eviction_list,
 				       struct i915_vma,
 				       exec_list);
 
-		obj =  &vma->obj->base;
 		list_del_init(&vma->exec_list);
+		__i915_vma_unpin(vma);
 		if (ret == 0)
 			ret = i915_vma_unbind(vma);
-
-		drm_gem_object_unreference(obj);
 	}
-
 	return ret;
 }
 
@@ -256,8 +220,8 @@ i915_gem_evict_for_vma(struct i915_vma *target)
 
 		vma = container_of(node, typeof(*vma), node);
 
-		if (vma->pin_count) {
-			if (!vma->exec_entry || (vma->pin_count > 1))
+		if (i915_vma_is_pinned(vma)) {
+			if (!vma->exec_entry || i915_vma_pin_count(vma) > 1)
 				/* Object is pinned for some other use */
 				return -EBUSY;
 
@@ -303,22 +267,21 @@ int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle)
 		struct drm_i915_private *dev_priv = to_i915(vm->dev);
 
 		if (i915_is_ggtt(vm)) {
-			ret = switch_to_pinned_context(dev_priv);
+			ret = i915_gem_switch_to_kernel_context(dev_priv);
 			if (ret)
 				return ret;
 		}
 
-		ret = i915_gem_wait_for_idle(dev_priv);
+		ret = i915_gem_wait_for_idle(dev_priv, true);
 		if (ret)
 			return ret;
 
 		i915_gem_retire_requests(dev_priv);
-
 		WARN_ON(!list_empty(&vm->active_list));
 	}
 
 	list_for_each_entry_safe(vma, next, &vm->inactive_list, vm_link)
-		if (vma->pin_count == 0)
+		if (!i915_vma_is_pinned(vma))
 			WARN_ON(i915_vma_unbind(vma));
 
 	return 0;

drivers/gpu/drm/i915/i915_gem_execbuffer.c

@@ -26,21 +26,38 @@
  *
  */
 
+#include <linux/dma_remapping.h>
+#include <linux/reservation.h>
+#include <linux/uaccess.h>
+
 #include <drm/drmP.h>
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
+#include "i915_gem_dmabuf.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
-#include <linux/dma_remapping.h>
-#include <linux/uaccess.h>
+#include "intel_frontbuffer.h"
 
-#define  __EXEC_OBJECT_HAS_PIN (1<<31)
-#define  __EXEC_OBJECT_HAS_FENCE (1<<30)
-#define  __EXEC_OBJECT_NEEDS_MAP (1<<29)
-#define  __EXEC_OBJECT_NEEDS_BIAS (1<<28)
+#define  __EXEC_OBJECT_HAS_PIN		(1<<31)
+#define  __EXEC_OBJECT_HAS_FENCE	(1<<30)
+#define  __EXEC_OBJECT_NEEDS_MAP	(1<<29)
+#define  __EXEC_OBJECT_NEEDS_BIAS	(1<<28)
+#define  __EXEC_OBJECT_INTERNAL_FLAGS (0xf<<28) /* all of the above */
 
 #define BATCH_OFFSET_BIAS (256*1024)
 
+struct i915_execbuffer_params {
+	struct drm_device               *dev;
+	struct drm_file                 *file;
+	struct i915_vma			*batch;
+	u32				dispatch_flags;
+	u32				args_batch_start_offset;
+	struct intel_engine_cs          *engine;
+	struct i915_gem_context         *ctx;
+	struct drm_i915_gem_request     *request;
+};
+
 struct eb_vmas {
 	struct list_head vmas;
 	int and;
@@ -89,6 +106,26 @@ eb_reset(struct eb_vmas *eb)
 		memset(eb->buckets, 0, (eb->and+1)*sizeof(struct hlist_head));
 }
 
+static struct i915_vma *
+eb_get_batch(struct eb_vmas *eb)
+{
+	struct i915_vma *vma = list_entry(eb->vmas.prev, typeof(*vma), exec_list);
+
+	/*
+	 * SNA is doing fancy tricks with compressing batch buffers, which leads
+	 * to negative relocation deltas. Usually that works out ok since the
+	 * relocate address is still positive, except when the batch is placed
+	 * very low in the GTT. Ensure this doesn't happen.
+	 *
+	 * Note that actual hangs have only been observed on gen7, but for
+	 * paranoia do it everywhere.
+	 */
+	if ((vma->exec_entry->flags & EXEC_OBJECT_PINNED) == 0)
+		vma->exec_entry->flags |= __EXEC_OBJECT_NEEDS_BIAS;
+
+	return vma;
+}
+
 static int
 eb_lookup_vmas(struct eb_vmas *eb,
 	       struct drm_i915_gem_exec_object2 *exec,
@@ -122,7 +159,7 @@ eb_lookup_vmas(struct eb_vmas *eb,
 			goto err;
 		}
 
-		drm_gem_object_reference(&obj->base);
+		i915_gem_object_get(obj);
 		list_add_tail(&obj->obj_exec_link, &objects);
 	}
 	spin_unlock(&file->table_lock);
@@ -175,7 +212,7 @@ err:
 				       struct drm_i915_gem_object,
 				       obj_exec_link);
 		list_del_init(&obj->obj_exec_link);
-		drm_gem_object_unreference(&obj->base);
+		i915_gem_object_put(obj);
 	}
 	/*
 	 * Objects already transfered to the vmas list will be unreferenced by
@@ -219,7 +256,7 @@ i915_gem_execbuffer_unreserve_vma(struct i915_vma *vma)
 		i915_gem_object_unpin_fence(obj);
 
 	if (entry->flags & __EXEC_OBJECT_HAS_PIN)
-		vma->pin_count--;
+		__i915_vma_unpin(vma);
 
 	entry->flags &= ~(__EXEC_OBJECT_HAS_FENCE | __EXEC_OBJECT_HAS_PIN);
 }
@@ -234,7 +271,7 @@ static void eb_destroy(struct eb_vmas *eb)
 				       exec_list);
 		list_del_init(&vma->exec_list);
 		i915_gem_execbuffer_unreserve_vma(vma);
-		drm_gem_object_unreference(&vma->obj->base);
+		i915_gem_object_put(vma->obj);
 	}
 	kfree(eb);
 }
@@ -399,6 +436,20 @@ relocate_entry_clflush(struct drm_i915_gem_object *obj,
 	return 0;
 }
 
+static bool object_is_idle(struct drm_i915_gem_object *obj)
+{
+	unsigned long active = i915_gem_object_get_active(obj);
+	int idx;
+
+	for_each_active(active, idx) {
+		if (!i915_gem_active_is_idle(&obj->last_read[idx],
+					     &obj->base.dev->struct_mutex))
+			return false;
+	}
+
+	return true;
+}
+
 static int
 i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 				   struct eb_vmas *eb,
@@ -482,7 +533,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 	}
 
 	/* We can't wait for rendering with pagefaults disabled */
-	if (obj->active && pagefault_disabled())
+	if (pagefault_disabled() && !object_is_idle(obj))
 		return -EFAULT;
 
 	if (use_cpu_reloc(obj))
@@ -626,12 +677,16 @@ i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
 			flags |= PIN_HIGH;
 	}
 
-	ret = i915_gem_object_pin(obj, vma->vm, entry->alignment, flags);
-	if ((ret == -ENOSPC  || ret == -E2BIG) &&
+	ret = i915_vma_pin(vma,
+			   entry->pad_to_size,
+			   entry->alignment,
+			   flags);
+	if ((ret == -ENOSPC || ret == -E2BIG) &&
 	    only_mappable_for_reloc(entry->flags))
-		ret = i915_gem_object_pin(obj, vma->vm,
-					  entry->alignment,
-					  flags & ~PIN_MAPPABLE);
+		ret = i915_vma_pin(vma,
+				   entry->pad_to_size,
+				   entry->alignment,
+				   flags & ~PIN_MAPPABLE);
 	if (ret)
 		return ret;
 
@@ -667,7 +722,7 @@ need_reloc_mappable(struct i915_vma *vma)
 	if (entry->relocation_count == 0)
 		return false;
 
-	if (!vma->is_ggtt)
+	if (!i915_vma_is_ggtt(vma))
 		return false;
 
 	/* See also use_cpu_reloc() */
@@ -686,12 +741,16 @@ eb_vma_misplaced(struct i915_vma *vma)
 	struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
 	struct drm_i915_gem_object *obj = vma->obj;
 
-	WARN_ON(entry->flags & __EXEC_OBJECT_NEEDS_MAP && !vma->is_ggtt);
+	WARN_ON(entry->flags & __EXEC_OBJECT_NEEDS_MAP &&
+		!i915_vma_is_ggtt(vma));
 
 	if (entry->alignment &&
 	    vma->node.start & (entry->alignment - 1))
 		return true;
 
+	if (vma->node.size < entry->pad_to_size)
+		return true;
+
 	if (entry->flags & EXEC_OBJECT_PINNED &&
 	    vma->node.start != entry->offset)
 		return true;
@@ -725,8 +784,6 @@ i915_gem_execbuffer_reserve(struct intel_engine_cs *engine,
 	bool has_fenced_gpu_access = INTEL_GEN(engine->i915) < 4;
 	int retry;
 
-	i915_gem_retire_requests_ring(engine);
-
 	vm = list_first_entry(vmas, struct i915_vma, exec_list)->vm;
 
 	INIT_LIST_HEAD(&ordered_vmas);
@@ -746,7 +803,7 @@ i915_gem_execbuffer_reserve(struct intel_engine_cs *engine,
 			entry->flags &= ~EXEC_OBJECT_NEEDS_FENCE;
 		need_fence =
 			entry->flags & EXEC_OBJECT_NEEDS_FENCE &&
-			obj->tiling_mode != I915_TILING_NONE;
+			i915_gem_object_is_tiled(obj);
 		need_mappable = need_fence || need_reloc_mappable(vma);
 
 		if (entry->flags & EXEC_OBJECT_PINNED)
@@ -843,7 +900,7 @@ i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 		vma = list_first_entry(&eb->vmas, struct i915_vma, exec_list);
 		list_del_init(&vma->exec_list);
 		i915_gem_execbuffer_unreserve_vma(vma);
-		drm_gem_object_unreference(&vma->obj->base);
+		i915_gem_object_put(vma->obj);
 	}
 
 	mutex_unlock(&dev->struct_mutex);
@@ -937,11 +994,21 @@ err:
 	return ret;
 }
 
+static unsigned int eb_other_engines(struct drm_i915_gem_request *req)
+{
+	unsigned int mask;
+
+	mask = ~intel_engine_flag(req->engine) & I915_BO_ACTIVE_MASK;
+	mask <<= I915_BO_ACTIVE_SHIFT;
+
+	return mask;
+}
+
 static int
 i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 				struct list_head *vmas)
 {
-	const unsigned other_rings = ~intel_engine_flag(req->engine);
+	const unsigned int other_rings = eb_other_engines(req);
 	struct i915_vma *vma;
 	uint32_t flush_domains = 0;
 	bool flush_chipset = false;
@@ -950,8 +1017,8 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 	list_for_each_entry(vma, vmas, exec_list) {
 		struct drm_i915_gem_object *obj = vma->obj;
 
-		if (obj->active & other_rings) {
-			ret = i915_gem_object_sync(obj, req->engine, &req);
+		if (obj->flags & other_rings) {
+			ret = i915_gem_object_sync(obj, req);
 			if (ret)
 				return ret;
 		}
@@ -968,10 +1035,8 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 	if (flush_domains & I915_GEM_DOMAIN_GTT)
 		wmb();
 
-	/* Unconditionally invalidate gpu caches and ensure that we do flush
-	 * any residual writes from the previous batch.
-	 */
-	return intel_ring_invalidate_all_caches(req);
+	/* Unconditionally invalidate GPU caches and TLBs. */
+	return req->engine->emit_flush(req, EMIT_INVALIDATE);
 }
 
 static bool
@@ -1007,6 +1072,9 @@ validate_exec_list(struct drm_device *dev,
 	unsigned invalid_flags;
 	int i;
 
+	/* INTERNAL flags must not overlap with external ones */
+	BUILD_BUG_ON(__EXEC_OBJECT_INTERNAL_FLAGS & ~__EXEC_OBJECT_UNKNOWN_FLAGS);
+
 	invalid_flags = __EXEC_OBJECT_UNKNOWN_FLAGS;
 	if (USES_FULL_PPGTT(dev))
 		invalid_flags |= EXEC_OBJECT_NEEDS_GTT;
@@ -1036,6 +1104,14 @@ validate_exec_list(struct drm_device *dev,
 		if (exec[i].alignment && !is_power_of_2(exec[i].alignment))
 			return -EINVAL;
 
+		/* pad_to_size was once a reserved field, so sanitize it */
+		if (exec[i].flags & EXEC_OBJECT_PAD_TO_SIZE) {
+			if (offset_in_page(exec[i].pad_to_size))
+				return -EINVAL;
+		} else {
+			exec[i].pad_to_size = 0;
+		}
+
 		/* First check for malicious input causing overflow in
 		 * the worst case where we need to allocate the entire
 		 * relocation tree as a single array.
@@ -1086,66 +1162,106 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
 	return ctx;
 }
 
-void
+void i915_vma_move_to_active(struct i915_vma *vma,
+			     struct drm_i915_gem_request *req,
+			     unsigned int flags)
+{
+	struct drm_i915_gem_object *obj = vma->obj;
+	const unsigned int idx = req->engine->id;
+
+	GEM_BUG_ON(!drm_mm_node_allocated(&vma->node));
+
+	obj->dirty = 1; /* be paranoid  */
+
+	/* Add a reference if we're newly entering the active list.
+	 * The order in which we add operations to the retirement queue is
+	 * vital here: mark_active adds to the start of the callback list,
+	 * such that subsequent callbacks are called first. Therefore we
+	 * add the active reference first and queue for it to be dropped
+	 * *last*.
+	 */
+	if (!i915_gem_object_is_active(obj))
+		i915_gem_object_get(obj);
+	i915_gem_object_set_active(obj, idx);
+	i915_gem_active_set(&obj->last_read[idx], req);
+
+	if (flags & EXEC_OBJECT_WRITE) {
+		i915_gem_active_set(&obj->last_write, req);
+
+		intel_fb_obj_invalidate(obj, ORIGIN_CS);
+
+		/* update for the implicit flush after a batch */
+		obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;
+	}
+
+	if (flags & EXEC_OBJECT_NEEDS_FENCE) {
+		i915_gem_active_set(&obj->last_fence, req);
+		if (flags & __EXEC_OBJECT_HAS_FENCE) {
+			struct drm_i915_private *dev_priv = req->i915;
+
+			list_move_tail(&dev_priv->fence_regs[obj->fence_reg].lru_list,
+				       &dev_priv->mm.fence_list);
+		}
+	}
+
+	i915_vma_set_active(vma, idx);
+	i915_gem_active_set(&vma->last_read[idx], req);
+	list_move_tail(&vma->vm_link, &vma->vm->active_list);
+}
+
+static void eb_export_fence(struct drm_i915_gem_object *obj,
+			    struct drm_i915_gem_request *req,
+			    unsigned int flags)
+{
+	struct reservation_object *resv;
+
+	resv = i915_gem_object_get_dmabuf_resv(obj);
+	if (!resv)
+		return;
+
+	/* Ignore errors from failing to allocate the new fence, we can't
+	 * handle an error right now. Worst case should be missed
+	 * synchronisation leading to rendering corruption.
+	 */
+	ww_mutex_lock(&resv->lock, NULL);
+	if (flags & EXEC_OBJECT_WRITE)
+		reservation_object_add_excl_fence(resv, &req->fence);
+	else if (reservation_object_reserve_shared(resv) == 0)
+		reservation_object_add_shared_fence(resv, &req->fence);
+	ww_mutex_unlock(&resv->lock);
+}
+
+static void
 i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 				   struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = i915_gem_request_get_engine(req);
 	struct i915_vma *vma;
 
 	list_for_each_entry(vma, vmas, exec_list) {
-		struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
 		struct drm_i915_gem_object *obj = vma->obj;
 		u32 old_read = obj->base.read_domains;
 		u32 old_write = obj->base.write_domain;
 
-		obj->dirty = 1; /* be paranoid  */
 		obj->base.write_domain = obj->base.pending_write_domain;
-		if (obj->base.write_domain == 0)
+		if (obj->base.write_domain)
+			vma->exec_entry->flags |= EXEC_OBJECT_WRITE;
+		else
 			obj->base.pending_read_domains |= obj->base.read_domains;
 		obj->base.read_domains = obj->base.pending_read_domains;
 
-		i915_vma_move_to_active(vma, req);
-		if (obj->base.write_domain) {
-			i915_gem_request_assign(&obj->last_write_req, req);
-
-			intel_fb_obj_invalidate(obj, ORIGIN_CS);
-
-			/* update for the implicit flush after a batch */
-			obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;
-		}
-		if (entry->flags & EXEC_OBJECT_NEEDS_FENCE) {
-			i915_gem_request_assign(&obj->last_fenced_req, req);
-			if (entry->flags & __EXEC_OBJECT_HAS_FENCE) {
-				struct drm_i915_private *dev_priv = engine->i915;
-				list_move_tail(&dev_priv->fence_regs[obj->fence_reg].lru_list,
-					       &dev_priv->mm.fence_list);
-			}
-		}
-
+		i915_vma_move_to_active(vma, req, vma->exec_entry->flags);
+		eb_export_fence(obj, req, vma->exec_entry->flags);
 		trace_i915_gem_object_change_domain(obj, old_read, old_write);
 	}
 }
 
-static void
-i915_gem_execbuffer_retire_commands(struct i915_execbuffer_params *params)
-{
-	/* Unconditionally force add_request to emit a full flush. */
-	params->engine->gpu_caches_dirty = true;
-
-	/* Add a breadcrumb for the completion of the batch buffer */
-	__i915_add_request(params->request, params->batch_obj, true);
-}
-
 static int
-i915_reset_gen7_sol_offsets(struct drm_device *dev,
-			    struct drm_i915_gem_request *req)
+i915_reset_gen7_sol_offsets(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = req->engine;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_ring *ring = req->ring;
 	int ret, i;
 
-	if (!IS_GEN7(dev) || engine != &dev_priv->engine[RCS]) {
+	if (!IS_GEN7(req->i915) || req->engine->id != RCS) {
 		DRM_DEBUG("sol reset is gen7/rcs only\n");
 		return -EINVAL;
 	}
@@ -1155,21 +1271,21 @@ i915_reset_gen7_sol_offsets(struct drm_device *dev,
 		return ret;
 
 	for (i = 0; i < 4; i++) {
-		intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit_reg(engine, GEN7_SO_WRITE_OFFSET(i));
-		intel_ring_emit(engine, 0);
+		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit_reg(ring, GEN7_SO_WRITE_OFFSET(i));
+		intel_ring_emit(ring, 0);
 	}
 
-	intel_ring_advance(engine);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
-static struct drm_i915_gem_object*
+static struct i915_vma*
 i915_gem_execbuffer_parse(struct intel_engine_cs *engine,
 			  struct drm_i915_gem_exec_object2 *shadow_exec_entry,
-			  struct eb_vmas *eb,
 			  struct drm_i915_gem_object *batch_obj,
+			  struct eb_vmas *eb,
 			  u32 batch_start_offset,
 			  u32 batch_len,
 			  bool is_master)
@@ -1181,18 +1297,18 @@ i915_gem_execbuffer_parse(struct intel_engine_cs *engine,
 	shadow_batch_obj = i915_gem_batch_pool_get(&engine->batch_pool,
 						   PAGE_ALIGN(batch_len));
 	if (IS_ERR(shadow_batch_obj))
-		return shadow_batch_obj;
-
-	ret = i915_parse_cmds(engine,
-			      batch_obj,
-			      shadow_batch_obj,
-			      batch_start_offset,
-			      batch_len,
-			      is_master);
+		return ERR_CAST(shadow_batch_obj);
+
+	ret = intel_engine_cmd_parser(engine,
+				      batch_obj,
+				      shadow_batch_obj,
+				      batch_start_offset,
+				      batch_len,
+				      is_master);
 	if (ret)
 		goto err;
 
-	ret = i915_gem_obj_ggtt_pin(shadow_batch_obj, 0, 0);
+	ret = i915_gem_object_ggtt_pin(shadow_batch_obj, NULL, 0, 0, 0);
 	if (ret)
 		goto err;
 
@@ -1203,29 +1319,25 @@ i915_gem_execbuffer_parse(struct intel_engine_cs *engine,
 	vma = i915_gem_obj_to_ggtt(shadow_batch_obj);
 	vma->exec_entry = shadow_exec_entry;
 	vma->exec_entry->flags = __EXEC_OBJECT_HAS_PIN;
-	drm_gem_object_reference(&shadow_batch_obj->base);
+	i915_gem_object_get(shadow_batch_obj);
 	list_add_tail(&vma->exec_list, &eb->vmas);
 
-	shadow_batch_obj->base.pending_read_domains = I915_GEM_DOMAIN_COMMAND;
-
-	return shadow_batch_obj;
+	return vma;
 
 err:
 	i915_gem_object_unpin_pages(shadow_batch_obj);
 	if (ret == -EACCES) /* unhandled chained batch */
-		return batch_obj;
+		return NULL;
 	else
 		return ERR_PTR(ret);
 }
 
-int
-i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
-			       struct drm_i915_gem_execbuffer2 *args,
-			       struct list_head *vmas)
+static int
+execbuf_submit(struct i915_execbuffer_params *params,
+	       struct drm_i915_gem_execbuffer2 *args,
+	       struct list_head *vmas)
 {
-	struct drm_device *dev = params->dev;
-	struct intel_engine_cs *engine = params->engine;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_private *dev_priv = params->request->i915;
 	u64 exec_start, exec_len;
 	int instp_mode;
 	u32 instp_mask;
@@ -1239,34 +1351,31 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	if (ret)
 		return ret;
 
-	WARN(params->ctx->ppgtt && params->ctx->ppgtt->pd_dirty_rings & (1<<engine->id),
-	     "%s didn't clear reload\n", engine->name);
-
 	instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
 	instp_mask = I915_EXEC_CONSTANTS_MASK;
 	switch (instp_mode) {
 	case I915_EXEC_CONSTANTS_REL_GENERAL:
 	case I915_EXEC_CONSTANTS_ABSOLUTE:
 	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (instp_mode != 0 && engine != &dev_priv->engine[RCS]) {
+		if (instp_mode != 0 && params->engine->id != RCS) {
 			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
 			return -EINVAL;
 		}
 
 		if (instp_mode != dev_priv->relative_constants_mode) {
-			if (INTEL_INFO(dev)->gen < 4) {
+			if (INTEL_INFO(dev_priv)->gen < 4) {
 				DRM_DEBUG("no rel constants on pre-gen4\n");
 				return -EINVAL;
 			}
 
-			if (INTEL_INFO(dev)->gen > 5 &&
+			if (INTEL_INFO(dev_priv)->gen > 5 &&
 			    instp_mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
 				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
 				return -EINVAL;
 			}
 
 			/* The HW changed the meaning on this bit on gen6 */
-			if (INTEL_INFO(dev)->gen >= 6)
+			if (INTEL_INFO(dev_priv)->gen >= 6)
 				instp_mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
 		}
 		break;
@@ -1275,37 +1384,39 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 		return -EINVAL;
 	}
 
-	if (engine == &dev_priv->engine[RCS] &&
+	if (params->engine->id == RCS &&
 	    instp_mode != dev_priv->relative_constants_mode) {
+		struct intel_ring *ring = params->request->ring;
+
 		ret = intel_ring_begin(params->request, 4);
 		if (ret)
 			return ret;
 
-		intel_ring_emit(engine, MI_NOOP);
-		intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit_reg(engine, INSTPM);
-		intel_ring_emit(engine, instp_mask << 16 | instp_mode);
-		intel_ring_advance(engine);
+		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit_reg(ring, INSTPM);
+		intel_ring_emit(ring, instp_mask << 16 | instp_mode);
+		intel_ring_advance(ring);
 
 		dev_priv->relative_constants_mode = instp_mode;
 	}
 
 	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
-		ret = i915_reset_gen7_sol_offsets(dev, params->request);
+		ret = i915_reset_gen7_sol_offsets(params->request);
 		if (ret)
 			return ret;
 	}
 
 	exec_len   = args->batch_len;
-	exec_start = params->batch_obj_vm_offset +
+	exec_start = params->batch->node.start +
 		     params->args_batch_start_offset;
 
 	if (exec_len == 0)
-		exec_len = params->batch_obj->base.size;
+		exec_len = params->batch->size;
 
-	ret = engine->dispatch_execbuffer(params->request,
-					exec_start, exec_len,
-					params->dispatch_flags);
+	ret = params->engine->emit_bb_start(params->request,
+					    exec_start, exec_len,
+					    params->dispatch_flags);
 	if (ret)
 		return ret;
 
@@ -1318,43 +1429,24 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 
 /**
  * Find one BSD ring to dispatch the corresponding BSD command.
- * The ring index is returned.
+ * The engine index is returned.
  */
 static unsigned int
-gen8_dispatch_bsd_ring(struct drm_i915_private *dev_priv, struct drm_file *file)
+gen8_dispatch_bsd_engine(struct drm_i915_private *dev_priv,
+			 struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
 	/* Check whether the file_priv has already selected one ring. */
-	if ((int)file_priv->bsd_ring < 0) {
+	if ((int)file_priv->bsd_engine < 0) {
 		/* If not, use the ping-pong mechanism to select one. */
 		mutex_lock(&dev_priv->drm.struct_mutex);
-		file_priv->bsd_ring = dev_priv->mm.bsd_ring_dispatch_index;
-		dev_priv->mm.bsd_ring_dispatch_index ^= 1;
+		file_priv->bsd_engine = dev_priv->mm.bsd_engine_dispatch_index;
+		dev_priv->mm.bsd_engine_dispatch_index ^= 1;
 		mutex_unlock(&dev_priv->drm.struct_mutex);
 	}
 
-	return file_priv->bsd_ring;
-}
-
-static struct drm_i915_gem_object *
-eb_get_batch(struct eb_vmas *eb)
-{
-	struct i915_vma *vma = list_entry(eb->vmas.prev, typeof(*vma), exec_list);
-
-	/*
-	 * SNA is doing fancy tricks with compressing batch buffers, which leads
-	 * to negative relocation deltas. Usually that works out ok since the
-	 * relocate address is still positive, except when the batch is placed
-	 * very low in the GTT. Ensure this doesn't happen.
-	 *
-	 * Note that actual hangs have only been observed on gen7, but for
-	 * paranoia do it everywhere.
-	 */
-	if ((vma->exec_entry->flags & EXEC_OBJECT_PINNED) == 0)
-		vma->exec_entry->flags |= __EXEC_OBJECT_NEEDS_BIAS;
-
-	return vma->obj;
+	return file_priv->bsd_engine;
 }
 
 #define I915_USER_RINGS (4)
@@ -1367,31 +1459,31 @@ static const enum intel_engine_id user_ring_map[I915_USER_RINGS + 1] = {
 	[I915_EXEC_VEBOX]	= VECS
 };
 
-static int
-eb_select_ring(struct drm_i915_private *dev_priv,
-	       struct drm_file *file,
-	       struct drm_i915_gem_execbuffer2 *args,
-	       struct intel_engine_cs **ring)
+static struct intel_engine_cs *
+eb_select_engine(struct drm_i915_private *dev_priv,
+		 struct drm_file *file,
+		 struct drm_i915_gem_execbuffer2 *args)
 {
 	unsigned int user_ring_id = args->flags & I915_EXEC_RING_MASK;
+	struct intel_engine_cs *engine;
 
 	if (user_ring_id > I915_USER_RINGS) {
 		DRM_DEBUG("execbuf with unknown ring: %u\n", user_ring_id);
-		return -EINVAL;
+		return NULL;
 	}
 
 	if ((user_ring_id != I915_EXEC_BSD) &&
 	    ((args->flags & I915_EXEC_BSD_MASK) != 0)) {
 		DRM_DEBUG("execbuf with non bsd ring but with invalid "
 			  "bsd dispatch flags: %d\n", (int)(args->flags));
-		return -EINVAL;
+		return NULL;
 	}
 
 	if (user_ring_id == I915_EXEC_BSD && HAS_BSD2(dev_priv)) {
 		unsigned int bsd_idx = args->flags & I915_EXEC_BSD_MASK;
 
 		if (bsd_idx == I915_EXEC_BSD_DEFAULT) {
-			bsd_idx = gen8_dispatch_bsd_ring(dev_priv, file);
+			bsd_idx = gen8_dispatch_bsd_engine(dev_priv, file);
 		} else if (bsd_idx >= I915_EXEC_BSD_RING1 &&
 			   bsd_idx <= I915_EXEC_BSD_RING2) {
 			bsd_idx >>= I915_EXEC_BSD_SHIFT;
@@ -1399,20 +1491,20 @@ eb_select_ring(struct drm_i915_private *dev_priv,
 		} else {
 			DRM_DEBUG("execbuf with unknown bsd ring: %u\n",
 				  bsd_idx);
-			return -EINVAL;
+			return NULL;
 		}
 
-		*ring = &dev_priv->engine[_VCS(bsd_idx)];
+		engine = &dev_priv->engine[_VCS(bsd_idx)];
 	} else {
-		*ring = &dev_priv->engine[user_ring_map[user_ring_id]];
+		engine = &dev_priv->engine[user_ring_map[user_ring_id]];
 	}
 
-	if (!intel_engine_initialized(*ring)) {
+	if (!intel_engine_initialized(engine)) {
 		DRM_DEBUG("execbuf with invalid ring: %u\n", user_ring_id);
-		return -EINVAL;
+		return NULL;
 	}
 
-	return 0;
+	return engine;
 }
 
 static int
@@ -1423,9 +1515,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-	struct drm_i915_gem_request *req = NULL;
 	struct eb_vmas *eb;
-	struct drm_i915_gem_object *batch_obj;
 	struct drm_i915_gem_exec_object2 shadow_exec_entry;
 	struct intel_engine_cs *engine;
 	struct i915_gem_context *ctx;
@@ -1454,9 +1544,9 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (args->flags & I915_EXEC_IS_PINNED)
 		dispatch_flags |= I915_DISPATCH_PINNED;
 
-	ret = eb_select_ring(dev_priv, file, args, &engine);
-	if (ret)
-		return ret;
+	engine = eb_select_engine(dev_priv, file, args);
+	if (!engine)
+		return -EINVAL;
 
 	if (args->buffer_count < 1) {
 		DRM_DEBUG("execbuf with %d buffers\n", args->buffer_count);
@@ -1496,7 +1586,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		goto pre_mutex_err;
 	}
 
-	i915_gem_context_reference(ctx);
+	i915_gem_context_get(ctx);
 
 	if (ctx->ppgtt)
 		vm = &ctx->ppgtt->base;
@@ -1507,7 +1597,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	eb = eb_create(args);
 	if (eb == NULL) {
-		i915_gem_context_unreference(ctx);
+		i915_gem_context_put(ctx);
 		mutex_unlock(&dev->struct_mutex);
 		ret = -ENOMEM;
 		goto pre_mutex_err;
@@ -1519,7 +1609,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		goto err;
 
 	/* take note of the batch buffer before we might reorder the lists */
-	batch_obj = eb_get_batch(eb);
+	params->batch = eb_get_batch(eb);
 
 	/* Move the objects en-masse into the GTT, evicting if necessary. */
 	need_relocs = (args->flags & I915_EXEC_NO_RELOC) == 0;
@@ -1543,34 +1633,28 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	}
 
 	/* Set the pending read domains for the batch buffer to COMMAND */
-	if (batch_obj->base.pending_write_domain) {
+	if (params->batch->obj->base.pending_write_domain) {
 		DRM_DEBUG("Attempting to use self-modifying batch buffer\n");
 		ret = -EINVAL;
 		goto err;
 	}
 
 	params->args_batch_start_offset = args->batch_start_offset;
-	if (i915_needs_cmd_parser(engine) && args->batch_len) {
-		struct drm_i915_gem_object *parsed_batch_obj;
-
-		parsed_batch_obj = i915_gem_execbuffer_parse(engine,
-							     &shadow_exec_entry,
-							     eb,
-							     batch_obj,
-							     args->batch_start_offset,
-							     args->batch_len,
-							     drm_is_current_master(file));
-		if (IS_ERR(parsed_batch_obj)) {
-			ret = PTR_ERR(parsed_batch_obj);
+	if (intel_engine_needs_cmd_parser(engine) && args->batch_len) {
+		struct i915_vma *vma;
+
+		vma = i915_gem_execbuffer_parse(engine, &shadow_exec_entry,
+						params->batch->obj,
+						eb,
+						args->batch_start_offset,
+						args->batch_len,
+						drm_is_current_master(file));
+		if (IS_ERR(vma)) {
+			ret = PTR_ERR(vma);
 			goto err;
 		}
 
-		/*
-		 * parsed_batch_obj == batch_obj means batch not fully parsed:
-		 * Accept, but don't promote to secure.
-		 */
-
-		if (parsed_batch_obj != batch_obj) {
+		if (vma) {
 			/*
 			 * Batch parsed and accepted:
 			 *
@@ -1582,16 +1666,18 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 			 */
 			dispatch_flags |= I915_DISPATCH_SECURE;
 			params->args_batch_start_offset = 0;
-			batch_obj = parsed_batch_obj;
+			params->batch = vma;
 		}
 	}
 
-	batch_obj->base.pending_read_domains |= I915_GEM_DOMAIN_COMMAND;
+	params->batch->obj->base.pending_read_domains |= I915_GEM_DOMAIN_COMMAND;
 
 	/* snb/ivb/vlv conflate the "batch in ppgtt" bit with the "non-secure
 	 * batch" bit. Hence we need to pin secure batches into the global gtt.
 	 * hsw should have this fixed, but bdw mucks it up again. */
 	if (dispatch_flags & I915_DISPATCH_SECURE) {
+		struct drm_i915_gem_object *obj = params->batch->obj;
+
 		/*
 		 * So on first glance it looks freaky that we pin the batch here
 		 * outside of the reservation loop. But:
@@ -1602,22 +1688,21 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		 *   fitting due to fragmentation.
 		 * So this is actually safe.
 		 */
-		ret = i915_gem_obj_ggtt_pin(batch_obj, 0, 0);
+		ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 0, 0);
 		if (ret)
 			goto err;
 
-		params->batch_obj_vm_offset = i915_gem_obj_ggtt_offset(batch_obj);
-	} else
-		params->batch_obj_vm_offset = i915_gem_obj_offset(batch_obj, vm);
+		params->batch = i915_gem_obj_to_ggtt(obj);
+	}
 
 	/* Allocate a request for this batch buffer nice and early. */
-	req = i915_gem_request_alloc(engine, ctx);
-	if (IS_ERR(req)) {
-		ret = PTR_ERR(req);
+	params->request = i915_gem_request_alloc(engine, ctx);
+	if (IS_ERR(params->request)) {
+		ret = PTR_ERR(params->request);
 		goto err_batch_unpin;
 	}
 
-	ret = i915_gem_request_add_to_client(req, file);
+	ret = i915_gem_request_add_to_client(params->request, file);
 	if (ret)
 		goto err_request;
 
@@ -1631,13 +1716,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	params->file                    = file;
 	params->engine                    = engine;
 	params->dispatch_flags          = dispatch_flags;
-	params->batch_obj               = batch_obj;
 	params->ctx                     = ctx;
-	params->request                 = req;
 
-	ret = dev_priv->gt.execbuf_submit(params, args, &eb->vmas);
+	ret = execbuf_submit(params, args, &eb->vmas);
 err_request:
-	i915_gem_execbuffer_retire_commands(params);
+	__i915_add_request(params->request, params->batch->obj, ret == 0);
 
 err_batch_unpin:
 	/*
@@ -1647,11 +1730,10 @@ err_batch_unpin:
 	 * active.
 	 */
 	if (dispatch_flags & I915_DISPATCH_SECURE)
-		i915_gem_object_ggtt_unpin(batch_obj);
-
+		i915_vma_unpin(params->batch);
 err:
 	/* the request owns the ref now */
-	i915_gem_context_unreference(ctx);
+	i915_gem_context_put(ctx);
 	eb_destroy(eb);
 
 	mutex_unlock(&dev->struct_mutex);

drivers/gpu/drm/i915/i915_gem_fence.c

@@ -86,20 +86,22 @@ static void i965_write_fence_reg(struct drm_device *dev, int reg,
 
 	if (obj) {
 		u32 size = i915_gem_obj_ggtt_size(obj);
+		unsigned int tiling = i915_gem_object_get_tiling(obj);
+		unsigned int stride = i915_gem_object_get_stride(obj);
 		uint64_t val;
 
 		/* Adjust fence size to match tiled area */
-		if (obj->tiling_mode != I915_TILING_NONE) {
-			uint32_t row_size = obj->stride *
-				(obj->tiling_mode == I915_TILING_Y ? 32 : 8);
+		if (tiling != I915_TILING_NONE) {
+			uint32_t row_size = stride *
+				(tiling == I915_TILING_Y ? 32 : 8);
 			size = (size / row_size) * row_size;
 		}
 
 		val = (uint64_t)((i915_gem_obj_ggtt_offset(obj) + size - 4096) &
 				 0xfffff000) << 32;
 		val |= i915_gem_obj_ggtt_offset(obj) & 0xfffff000;
-		val |= (uint64_t)((obj->stride / 128) - 1) << fence_pitch_shift;
-		if (obj->tiling_mode == I915_TILING_Y)
+		val |= (uint64_t)((stride / 128) - 1) << fence_pitch_shift;
+		if (tiling == I915_TILING_Y)
 			val |= 1 << I965_FENCE_TILING_Y_SHIFT;
 		val |= I965_FENCE_REG_VALID;
 
@@ -122,6 +124,8 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg,
 
 	if (obj) {
 		u32 size = i915_gem_obj_ggtt_size(obj);
+		unsigned int tiling = i915_gem_object_get_tiling(obj);
+		unsigned int stride = i915_gem_object_get_stride(obj);
 		int pitch_val;
 		int tile_width;
 
@@ -131,17 +135,17 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg,
 		     "object 0x%08llx [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n",
 		     i915_gem_obj_ggtt_offset(obj), obj->map_and_fenceable, size);
 
-		if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev))
+		if (tiling == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev))
 			tile_width = 128;
 		else
 			tile_width = 512;
 
 		/* Note: pitch better be a power of two tile widths */
-		pitch_val = obj->stride / tile_width;
+		pitch_val = stride / tile_width;
 		pitch_val = ffs(pitch_val) - 1;
 
 		val = i915_gem_obj_ggtt_offset(obj);
-		if (obj->tiling_mode == I915_TILING_Y)
+		if (tiling == I915_TILING_Y)
 			val |= 1 << I830_FENCE_TILING_Y_SHIFT;
 		val |= I915_FENCE_SIZE_BITS(size);
 		val |= pitch_val << I830_FENCE_PITCH_SHIFT;
@@ -161,6 +165,8 @@ static void i830_write_fence_reg(struct drm_device *dev, int reg,
 
 	if (obj) {
 		u32 size = i915_gem_obj_ggtt_size(obj);
+		unsigned int tiling = i915_gem_object_get_tiling(obj);
+		unsigned int stride = i915_gem_object_get_stride(obj);
 		uint32_t pitch_val;
 
 		WARN((i915_gem_obj_ggtt_offset(obj) & ~I830_FENCE_START_MASK) ||
@@ -169,11 +175,11 @@ static void i830_write_fence_reg(struct drm_device *dev, int reg,
 		     "object 0x%08llx not 512K or pot-size 0x%08x aligned\n",
 		     i915_gem_obj_ggtt_offset(obj), size);
 
-		pitch_val = obj->stride / 128;
+		pitch_val = stride / 128;
 		pitch_val = ffs(pitch_val) - 1;
 
 		val = i915_gem_obj_ggtt_offset(obj);
-		if (obj->tiling_mode == I915_TILING_Y)
+		if (tiling == I915_TILING_Y)
 			val |= 1 << I830_FENCE_TILING_Y_SHIFT;
 		val |= I830_FENCE_SIZE_BITS(size);
 		val |= pitch_val << I830_FENCE_PITCH_SHIFT;
@@ -201,9 +207,12 @@ static void i915_gem_write_fence(struct drm_device *dev, int reg,
 	if (i915_gem_object_needs_mb(dev_priv->fence_regs[reg].obj))
 		mb();
 
-	WARN(obj && (!obj->stride || !obj->tiling_mode),
+	WARN(obj &&
+	     (!i915_gem_object_get_stride(obj) ||
+	      !i915_gem_object_get_tiling(obj)),
 	     "bogus fence setup with stride: 0x%x, tiling mode: %i\n",
-	     obj->stride, obj->tiling_mode);
+	     i915_gem_object_get_stride(obj),
+	     i915_gem_object_get_tiling(obj));
 
 	if (IS_GEN2(dev))
 		i830_write_fence_reg(dev, reg, obj);
@@ -248,7 +257,7 @@ static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj,
 
 static inline void i915_gem_object_fence_lost(struct drm_i915_gem_object *obj)
 {
-	if (obj->tiling_mode)
+	if (i915_gem_object_is_tiled(obj))
 		i915_gem_release_mmap(obj);
 
 	/* As we do not have an associated fence register, we will force
@@ -261,15 +270,8 @@ static inline void i915_gem_object_fence_lost(struct drm_i915_gem_object *obj)
 static int
 i915_gem_object_wait_fence(struct drm_i915_gem_object *obj)
 {
-	if (obj->last_fenced_req) {
-		int ret = i915_wait_request(obj->last_fenced_req);
-		if (ret)
-			return ret;
-
-		i915_gem_request_assign(&obj->last_fenced_req, NULL);
-	}
-
-	return 0;
+	return i915_gem_active_retire(&obj->last_fence,
+				      &obj->base.dev->struct_mutex);
 }
 
 /**
@@ -368,7 +370,7 @@ i915_gem_object_get_fence(struct drm_i915_gem_object *obj)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	bool enable = obj->tiling_mode != I915_TILING_NONE;
+	bool enable = i915_gem_object_is_tiled(obj);
 	struct drm_i915_fence_reg *reg;
 	int ret;
 
@@ -438,7 +440,7 @@ i915_gem_object_pin_fence(struct drm_i915_gem_object *obj)
 
 		WARN_ON(!ggtt_vma ||
 			dev_priv->fence_regs[obj->fence_reg].pin_count >
-			ggtt_vma->pin_count);
+			i915_vma_pin_count(ggtt_vma));
 		dev_priv->fence_regs[obj->fence_reg].pin_count++;
 		return true;
 	} else
@@ -484,7 +486,7 @@ void i915_gem_restore_fences(struct drm_device *dev)
 		 */
 		if (reg->obj) {
 			i915_gem_object_update_fence(reg->obj, reg,
-						     reg->obj->tiling_mode);
+						     i915_gem_object_get_tiling(reg->obj));
 		} else {
 			i915_gem_write_fence(dev, i, NULL);
 		}

drivers/gpu/drm/i915/i915_gem_gtt.c

@@ -184,7 +184,7 @@ static void ppgtt_unbind_vma(struct i915_vma *vma)
 {
 	vma->vm->clear_range(vma->vm,
 			     vma->node.start,
-			     vma->obj->base.size,
+			     vma->size,
 			     true);
 }
 
@@ -669,6 +669,7 @@ static int gen8_write_pdp(struct drm_i915_gem_request *req,
 			  unsigned entry,
 			  dma_addr_t addr)
 {
+	struct intel_ring *ring = req->ring;
 	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
@@ -678,13 +679,13 @@ static int gen8_write_pdp(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
-	intel_ring_emit_reg(engine, GEN8_RING_PDP_UDW(engine, entry));
-	intel_ring_emit(engine, upper_32_bits(addr));
-	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
-	intel_ring_emit_reg(engine, GEN8_RING_PDP_LDW(engine, entry));
-	intel_ring_emit(engine, lower_32_bits(addr));
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+	intel_ring_emit_reg(ring, GEN8_RING_PDP_UDW(engine, entry));
+	intel_ring_emit(ring, upper_32_bits(addr));
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+	intel_ring_emit_reg(ring, GEN8_RING_PDP_LDW(engine, entry));
+	intel_ring_emit(ring, lower_32_bits(addr));
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -1660,11 +1661,12 @@ static uint32_t get_pd_offset(struct i915_hw_ppgtt *ppgtt)
 static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 			 struct drm_i915_gem_request *req)
 {
+	struct intel_ring *ring = req->ring;
 	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	/* NB: TLBs must be flushed and invalidated before a switch */
-	ret = engine->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
+	ret = engine->emit_flush(req, EMIT_INVALIDATE | EMIT_FLUSH);
 	if (ret)
 		return ret;
 
@@ -1672,13 +1674,13 @@ static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(2));
-	intel_ring_emit_reg(engine, RING_PP_DIR_DCLV(engine));
-	intel_ring_emit(engine, PP_DIR_DCLV_2G);
-	intel_ring_emit_reg(engine, RING_PP_DIR_BASE(engine));
-	intel_ring_emit(engine, get_pd_offset(ppgtt));
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(2));
+	intel_ring_emit_reg(ring, RING_PP_DIR_DCLV(engine));
+	intel_ring_emit(ring, PP_DIR_DCLV_2G);
+	intel_ring_emit_reg(ring, RING_PP_DIR_BASE(engine));
+	intel_ring_emit(ring, get_pd_offset(ppgtt));
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -1686,11 +1688,12 @@ static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
 			  struct drm_i915_gem_request *req)
 {
+	struct intel_ring *ring = req->ring;
 	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
 	/* NB: TLBs must be flushed and invalidated before a switch */
-	ret = engine->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
+	ret = engine->emit_flush(req, EMIT_INVALIDATE | EMIT_FLUSH);
 	if (ret)
 		return ret;
 
@@ -1698,17 +1701,17 @@ static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(2));
-	intel_ring_emit_reg(engine, RING_PP_DIR_DCLV(engine));
-	intel_ring_emit(engine, PP_DIR_DCLV_2G);
-	intel_ring_emit_reg(engine, RING_PP_DIR_BASE(engine));
-	intel_ring_emit(engine, get_pd_offset(ppgtt));
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(2));
+	intel_ring_emit_reg(ring, RING_PP_DIR_DCLV(engine));
+	intel_ring_emit(ring, PP_DIR_DCLV_2G);
+	intel_ring_emit_reg(ring, RING_PP_DIR_BASE(engine));
+	intel_ring_emit(ring, get_pd_offset(ppgtt));
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	/* XXX: RCS is the only one to auto invalidate the TLBs? */
 	if (engine->id != RCS) {
-		ret = engine->flush(req, I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS);
+		ret = engine->emit_flush(req, EMIT_INVALIDATE | EMIT_FLUSH);
 		if (ret)
 			return ret;
 	}
@@ -2009,7 +2012,7 @@ alloc:
 						  0, ggtt->base.total,
 						  DRM_MM_TOPDOWN);
 	if (ret == -ENOSPC && !retried) {
-		ret = i915_gem_evict_something(dev, &ggtt->base,
+		ret = i915_gem_evict_something(&ggtt->base,
 					       GEN6_PD_SIZE, GEN6_PD_ALIGN,
 					       I915_CACHE_NONE,
 					       0, ggtt->base.total,
@@ -2101,11 +2104,12 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 	return 0;
 }
 
-static int __hw_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
+static int __hw_ppgtt_init(struct i915_hw_ppgtt *ppgtt,
+			   struct drm_i915_private *dev_priv)
 {
-	ppgtt->base.dev = dev;
+	ppgtt->base.dev = &dev_priv->drm;
 
-	if (INTEL_INFO(dev)->gen < 8)
+	if (INTEL_INFO(dev_priv)->gen < 8)
 		return gen6_ppgtt_init(ppgtt);
 	else
 		return gen8_ppgtt_init(ppgtt);
@@ -2115,9 +2119,9 @@ static void i915_address_space_init(struct i915_address_space *vm,
 				    struct drm_i915_private *dev_priv)
 {
 	drm_mm_init(&vm->mm, vm->start, vm->total);
-	vm->dev = &dev_priv->drm;
 	INIT_LIST_HEAD(&vm->active_list);
 	INIT_LIST_HEAD(&vm->inactive_list);
+	INIT_LIST_HEAD(&vm->unbound_list);
 	list_add_tail(&vm->global_link, &dev_priv->vm_list);
 }
 
@@ -2140,15 +2144,17 @@ static void gtt_write_workarounds(struct drm_device *dev)
 		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT);
 }
 
-static int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
+static int i915_ppgtt_init(struct i915_hw_ppgtt *ppgtt,
+			   struct drm_i915_private *dev_priv,
+			   struct drm_i915_file_private *file_priv)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret = 0;
+	int ret;
 
-	ret = __hw_ppgtt_init(dev, ppgtt);
+	ret = __hw_ppgtt_init(ppgtt, dev_priv);
 	if (ret == 0) {
 		kref_init(&ppgtt->ref);
 		i915_address_space_init(&ppgtt->base, dev_priv);
+		ppgtt->base.file = file_priv;
 	}
 
 	return ret;
@@ -2180,7 +2186,8 @@ int i915_ppgtt_init_hw(struct drm_device *dev)
 }
 
 struct i915_hw_ppgtt *
-i915_ppgtt_create(struct drm_device *dev, struct drm_i915_file_private *fpriv)
+i915_ppgtt_create(struct drm_i915_private *dev_priv,
+		  struct drm_i915_file_private *fpriv)
 {
 	struct i915_hw_ppgtt *ppgtt;
 	int ret;
@@ -2189,14 +2196,12 @@ i915_ppgtt_create(struct drm_device *dev, struct drm_i915_file_private *fpriv)
 	if (!ppgtt)
 		return ERR_PTR(-ENOMEM);
 
-	ret = i915_ppgtt_init(dev, ppgtt);
+	ret = i915_ppgtt_init(ppgtt, dev_priv, fpriv);
 	if (ret) {
 		kfree(ppgtt);
 		return ERR_PTR(ret);
 	}
 
-	ppgtt->file_priv = fpriv;
-
 	trace_i915_ppgtt_create(&ppgtt->base);
 
 	return ppgtt;
@@ -2209,9 +2214,10 @@ void  i915_ppgtt_release(struct kref *kref)
 
 	trace_i915_ppgtt_release(&ppgtt->base);
 
-	/* vmas should already be unbound */
+	/* vmas should already be unbound and destroyed */
 	WARN_ON(!list_empty(&ppgtt->base.active_list));
 	WARN_ON(!list_empty(&ppgtt->base.inactive_list));
+	WARN_ON(!list_empty(&ppgtt->base.unbound_list));
 
 	list_del(&ppgtt->base.global_link);
 	drm_mm_takedown(&ppgtt->base.mm);
@@ -2220,47 +2226,21 @@ void  i915_ppgtt_release(struct kref *kref)
 	kfree(ppgtt);
 }
 
-extern int intel_iommu_gfx_mapped;
 /* Certain Gen5 chipsets require require idling the GPU before
  * unmapping anything from the GTT when VT-d is enabled.
  */
-static bool needs_idle_maps(struct drm_device *dev)
+static bool needs_idle_maps(struct drm_i915_private *dev_priv)
 {
 #ifdef CONFIG_INTEL_IOMMU
 	/* Query intel_iommu to see if we need the workaround. Presumably that
 	 * was loaded first.
 	 */
-	if (IS_GEN5(dev) && IS_MOBILE(dev) && intel_iommu_gfx_mapped)
+	if (IS_GEN5(dev_priv) && IS_MOBILE(dev_priv) && intel_iommu_gfx_mapped)
 		return true;
 #endif
 	return false;
 }
 
-static bool do_idling(struct drm_i915_private *dev_priv)
-{
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-	bool ret = dev_priv->mm.interruptible;
-
-	if (unlikely(ggtt->do_idle_maps)) {
-		dev_priv->mm.interruptible = false;
-		if (i915_gem_wait_for_idle(dev_priv)) {
-			DRM_ERROR("Failed to wait for idle; VT'd may hang.\n");
-			/* Wait a bit, in hopes it avoids the hang */
-			udelay(10);
-		}
-	}
-
-	return ret;
-}
-
-static void undo_idling(struct drm_i915_private *dev_priv, bool interruptible)
-{
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-
-	if (unlikely(ggtt->do_idle_maps))
-		dev_priv->mm.interruptible = interruptible;
-}
-
 void i915_check_and_clear_faults(struct drm_i915_private *dev_priv)
 {
 	struct intel_engine_cs *engine;
@@ -2647,7 +2627,7 @@ static int ggtt_bind_vma(struct i915_vma *vma,
 	 * GLOBAL/LOCAL_BIND, it's all the same ptes. Hence unconditionally
 	 * upgrade to both bound if we bind either to avoid double-binding.
 	 */
-	vma->bound |= GLOBAL_BIND | LOCAL_BIND;
+	vma->flags |= I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND;
 
 	return 0;
 }
@@ -2669,14 +2649,14 @@ static int aliasing_gtt_bind_vma(struct i915_vma *vma,
 		pte_flags |= PTE_READ_ONLY;
 
 
-	if (flags & GLOBAL_BIND) {
+	if (flags & I915_VMA_GLOBAL_BIND) {
 		vma->vm->insert_entries(vma->vm,
 					vma->ggtt_view.pages,
 					vma->node.start,
 					cache_level, pte_flags);
 	}
 
-	if (flags & LOCAL_BIND) {
+	if (flags & I915_VMA_LOCAL_BIND) {
 		struct i915_hw_ppgtt *appgtt =
 			to_i915(vma->vm->dev)->mm.aliasing_ppgtt;
 		appgtt->base.insert_entries(&appgtt->base,
@@ -2690,42 +2670,36 @@ static int aliasing_gtt_bind_vma(struct i915_vma *vma,
 
 static void ggtt_unbind_vma(struct i915_vma *vma)
 {
-	struct drm_device *dev = vma->vm->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_i915_gem_object *obj = vma->obj;
-	const uint64_t size = min_t(uint64_t,
-				    obj->base.size,
-				    vma->node.size);
+	struct i915_hw_ppgtt *appgtt = to_i915(vma->vm->dev)->mm.aliasing_ppgtt;
+	const u64 size = min(vma->size, vma->node.size);
 
-	if (vma->bound & GLOBAL_BIND) {
+	if (vma->flags & I915_VMA_GLOBAL_BIND)
 		vma->vm->clear_range(vma->vm,
-				     vma->node.start,
-				     size,
+				     vma->node.start, size,
 				     true);
-	}
-
-	if (dev_priv->mm.aliasing_ppgtt && vma->bound & LOCAL_BIND) {
-		struct i915_hw_ppgtt *appgtt = dev_priv->mm.aliasing_ppgtt;
 
+	if (vma->flags & I915_VMA_LOCAL_BIND && appgtt)
 		appgtt->base.clear_range(&appgtt->base,
-					 vma->node.start,
-					 size,
+					 vma->node.start, size,
 					 true);
-	}
 }
 
 void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	bool interruptible;
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 
-	interruptible = do_idling(dev_priv);
+	if (unlikely(ggtt->do_idle_maps)) {
+		if (i915_gem_wait_for_idle(dev_priv, false)) {
+			DRM_ERROR("Failed to wait for idle; VT'd may hang.\n");
+			/* Wait a bit, in hopes it avoids the hang */
+			udelay(10);
+		}
+	}
 
 	dma_unmap_sg(&dev->pdev->dev, obj->pages->sgl, obj->pages->nents,
 		     PCI_DMA_BIDIRECTIONAL);
-
-	undo_idling(dev_priv, interruptible);
 }
 
 static void i915_gtt_color_adjust(struct drm_mm_node *node,
@@ -2736,19 +2710,14 @@ static void i915_gtt_color_adjust(struct drm_mm_node *node,
 	if (node->color != color)
 		*start += 4096;
 
-	if (!list_empty(&node->node_list)) {
-		node = list_entry(node->node_list.next,
-				  struct drm_mm_node,
-				  node_list);
-		if (node->allocated && node->color != color)
-			*end -= 4096;
-	}
+	node = list_first_entry_or_null(&node->node_list,
+					struct drm_mm_node,
+					node_list);
+	if (node && node->allocated && node->color != color)
+		*end -= 4096;
 }
 
-static int i915_gem_setup_global_gtt(struct drm_device *dev,
-				     u64 start,
-				     u64 mappable_end,
-				     u64 end)
+int i915_gem_init_ggtt(struct drm_i915_private *dev_priv)
 {
 	/* Let GEM Manage all of the aperture.
 	 *
@@ -2759,48 +2728,15 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 	 * aperture.  One page should be enough to keep any prefetching inside
 	 * of the aperture.
 	 */
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-	struct drm_mm_node *entry;
-	struct drm_i915_gem_object *obj;
 	unsigned long hole_start, hole_end;
+	struct drm_mm_node *entry;
 	int ret;
 
-	BUG_ON(mappable_end > end);
-
-	ggtt->base.start = start;
-
-	/* Subtract the guard page before address space initialization to
-	 * shrink the range used by drm_mm */
-	ggtt->base.total = end - start - PAGE_SIZE;
-	i915_address_space_init(&ggtt->base, dev_priv);
-	ggtt->base.total += PAGE_SIZE;
-
 	ret = intel_vgt_balloon(dev_priv);
 	if (ret)
 		return ret;
 
-	if (!HAS_LLC(dev))
-		ggtt->base.mm.color_adjust = i915_gtt_color_adjust;
-
-	/* Mark any preallocated objects as occupied */
-	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		struct i915_vma *vma = i915_gem_obj_to_vma(obj, &ggtt->base);
-
-		DRM_DEBUG_KMS("reserving preallocated space: %llx + %zx\n",
-			      i915_gem_obj_ggtt_offset(obj), obj->base.size);
-
-		WARN_ON(i915_gem_obj_ggtt_bound(obj));
-		ret = drm_mm_reserve_node(&ggtt->base.mm, &vma->node);
-		if (ret) {
-			DRM_DEBUG_KMS("Reservation failed: %i\n", ret);
-			return ret;
-		}
-		vma->bound |= GLOBAL_BIND;
-		__i915_vma_set_map_and_fenceable(vma);
-		list_add_tail(&vma->vm_link, &ggtt->base.inactive_list);
-	}
-
 	/* Clear any non-preallocated blocks */
 	drm_mm_for_each_hole(entry, &ggtt->base.mm, hole_start, hole_end) {
 		DRM_DEBUG_KMS("clearing unused GTT space: [%lx, %lx]\n",
@@ -2810,18 +2746,19 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 	}
 
 	/* And finally clear the reserved guard page */
-	ggtt->base.clear_range(&ggtt->base, end - PAGE_SIZE, PAGE_SIZE, true);
+	ggtt->base.clear_range(&ggtt->base,
+			       ggtt->base.total - PAGE_SIZE, PAGE_SIZE,
+			       true);
 
-	if (USES_PPGTT(dev) && !USES_FULL_PPGTT(dev)) {
+	if (USES_PPGTT(dev_priv) && !USES_FULL_PPGTT(dev_priv)) {
 		struct i915_hw_ppgtt *ppgtt;
 
 		ppgtt = kzalloc(sizeof(*ppgtt), GFP_KERNEL);
 		if (!ppgtt)
 			return -ENOMEM;
 
-		ret = __hw_ppgtt_init(dev, ppgtt);
+		ret = __hw_ppgtt_init(ppgtt, dev_priv);
 		if (ret) {
-			ppgtt->base.cleanup(&ppgtt->base);
 			kfree(ppgtt);
 			return ret;
 		}
@@ -2848,34 +2785,22 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 	return 0;
 }
 
-/**
- * i915_gem_init_ggtt - Initialize GEM for Global GTT
- * @dev: DRM device
- */
-void i915_gem_init_ggtt(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
-
-	i915_gem_setup_global_gtt(dev, 0, ggtt->mappable_end, ggtt->base.total);
-}
-
 /**
  * i915_ggtt_cleanup_hw - Clean up GGTT hardware initialization
- * @dev: DRM device
+ * @dev_priv: i915 device
  */
-void i915_ggtt_cleanup_hw(struct drm_device *dev)
+void i915_ggtt_cleanup_hw(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 
 	if (dev_priv->mm.aliasing_ppgtt) {
 		struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
 
 		ppgtt->base.cleanup(&ppgtt->base);
+		kfree(ppgtt);
 	}
 
-	i915_gem_cleanup_stolen(dev);
+	i915_gem_cleanup_stolen(&dev_priv->drm);
 
 	if (drm_mm_initialized(&ggtt->base.mm)) {
 		intel_vgt_deballoon(dev_priv);
@@ -2885,6 +2810,9 @@ void i915_ggtt_cleanup_hw(struct drm_device *dev)
 	}
 
 	ggtt->base.cleanup(&ggtt->base);
+
+	arch_phys_wc_del(ggtt->mtrr);
+	io_mapping_free(ggtt->mappable);
 }
 
 static unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
@@ -2965,17 +2893,14 @@ static size_t gen9_get_stolen_size(u16 gen9_gmch_ctl)
 		return (gen9_gmch_ctl - 0xf0 + 1) << 22;
 }
 
-static int ggtt_probe_common(struct drm_device *dev,
-			     size_t gtt_size)
+static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+	struct pci_dev *pdev = ggtt->base.dev->pdev;
 	struct i915_page_scratch *scratch_page;
-	phys_addr_t ggtt_phys_addr;
+	phys_addr_t phys_addr;
 
 	/* For Modern GENs the PTEs and register space are split in the BAR */
-	ggtt_phys_addr = pci_resource_start(dev->pdev, 0) +
-			 (pci_resource_len(dev->pdev, 0) / 2);
+	phys_addr = pci_resource_start(pdev, 0) + pci_resource_len(pdev, 0) / 2;
 
 	/*
 	 * On BXT writes larger than 64 bit to the GTT pagetable range will be
@@ -2984,16 +2909,16 @@ static int ggtt_probe_common(struct drm_device *dev,
 	 * resort to an uncached mapping. The WC issue is easily caught by the
 	 * readback check when writing GTT PTE entries.
 	 */
-	if (IS_BROXTON(dev))
-		ggtt->gsm = ioremap_nocache(ggtt_phys_addr, gtt_size);
+	if (IS_BROXTON(ggtt->base.dev))
+		ggtt->gsm = ioremap_nocache(phys_addr, size);
 	else
-		ggtt->gsm = ioremap_wc(ggtt_phys_addr, gtt_size);
+		ggtt->gsm = ioremap_wc(phys_addr, size);
 	if (!ggtt->gsm) {
-		DRM_ERROR("Failed to map the gtt page table\n");
+		DRM_ERROR("Failed to map the ggtt page table\n");
 		return -ENOMEM;
 	}
 
-	scratch_page = alloc_scratch_page(dev);
+	scratch_page = alloc_scratch_page(ggtt->base.dev);
 	if (IS_ERR(scratch_page)) {
 		DRM_ERROR("Scratch setup failed\n");
 		/* iounmap will also get called at remove, but meh */
@@ -3079,42 +3004,49 @@ static void chv_setup_private_ppat(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN8_PRIVATE_PAT_HI, pat >> 32);
 }
 
+static void gen6_gmch_remove(struct i915_address_space *vm)
+{
+	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
+
+	iounmap(ggtt->gsm);
+	free_scratch_page(vm->dev, vm->scratch_page);
+}
+
 static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 {
-	struct drm_device *dev = ggtt->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_private *dev_priv = to_i915(ggtt->base.dev);
+	struct pci_dev *pdev = dev_priv->drm.pdev;
+	unsigned int size;
 	u16 snb_gmch_ctl;
-	int ret;
 
 	/* TODO: We're not aware of mappable constraints on gen8 yet */
-	ggtt->mappable_base = pci_resource_start(dev->pdev, 2);
-	ggtt->mappable_end = pci_resource_len(dev->pdev, 2);
+	ggtt->mappable_base = pci_resource_start(pdev, 2);
+	ggtt->mappable_end = pci_resource_len(pdev, 2);
 
-	if (!pci_set_dma_mask(dev->pdev, DMA_BIT_MASK(39)))
-		pci_set_consistent_dma_mask(dev->pdev, DMA_BIT_MASK(39));
+	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(39)))
+		pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(39));
 
-	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
+	pci_read_config_word(pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
 
-	if (INTEL_INFO(dev)->gen >= 9) {
+	if (INTEL_GEN(dev_priv) >= 9) {
 		ggtt->stolen_size = gen9_get_stolen_size(snb_gmch_ctl);
-		ggtt->size = gen8_get_total_gtt_size(snb_gmch_ctl);
-	} else if (IS_CHERRYVIEW(dev)) {
+		size = gen8_get_total_gtt_size(snb_gmch_ctl);
+	} else if (IS_CHERRYVIEW(dev_priv)) {
 		ggtt->stolen_size = chv_get_stolen_size(snb_gmch_ctl);
-		ggtt->size = chv_get_total_gtt_size(snb_gmch_ctl);
+		size = chv_get_total_gtt_size(snb_gmch_ctl);
 	} else {
 		ggtt->stolen_size = gen8_get_stolen_size(snb_gmch_ctl);
-		ggtt->size = gen8_get_total_gtt_size(snb_gmch_ctl);
+		size = gen8_get_total_gtt_size(snb_gmch_ctl);
 	}
 
-	ggtt->base.total = (ggtt->size / sizeof(gen8_pte_t)) << PAGE_SHIFT;
+	ggtt->base.total = (size / sizeof(gen8_pte_t)) << PAGE_SHIFT;
 
-	if (IS_CHERRYVIEW(dev) || IS_BROXTON(dev))
+	if (IS_CHERRYVIEW(dev_priv) || IS_BROXTON(dev_priv))
 		chv_setup_private_ppat(dev_priv);
 	else
 		bdw_setup_private_ppat(dev_priv);
 
-	ret = ggtt_probe_common(dev, ggtt->size);
-
+	ggtt->base.cleanup = gen6_gmch_remove;
 	ggtt->base.bind_vma = ggtt_bind_vma;
 	ggtt->base.unbind_vma = ggtt_unbind_vma;
 	ggtt->base.insert_page = gen8_ggtt_insert_page;
@@ -3126,57 +3058,65 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 	if (IS_CHERRYVIEW(dev_priv))
 		ggtt->base.insert_entries = gen8_ggtt_insert_entries__BKL;
 
-	return ret;
+	return ggtt_probe_common(ggtt, size);
 }
 
 static int gen6_gmch_probe(struct i915_ggtt *ggtt)
 {
-	struct drm_device *dev = ggtt->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(ggtt->base.dev);
+	struct pci_dev *pdev = dev_priv->drm.pdev;
+	unsigned int size;
 	u16 snb_gmch_ctl;
-	int ret;
 
-	ggtt->mappable_base = pci_resource_start(dev->pdev, 2);
-	ggtt->mappable_end = pci_resource_len(dev->pdev, 2);
+	ggtt->mappable_base = pci_resource_start(pdev, 2);
+	ggtt->mappable_end = pci_resource_len(pdev, 2);
 
 	/* 64/512MB is the current min/max we actually know of, but this is just
 	 * a coarse sanity check.
 	 */
-	if ((ggtt->mappable_end < (64<<20) || (ggtt->mappable_end > (512<<20)))) {
+	if (ggtt->mappable_end < (64<<20) || ggtt->mappable_end > (512<<20)) {
 		DRM_ERROR("Unknown GMADR size (%llx)\n", ggtt->mappable_end);
 		return -ENXIO;
 	}
 
-	if (!pci_set_dma_mask(dev->pdev, DMA_BIT_MASK(40)))
-		pci_set_consistent_dma_mask(dev->pdev, DMA_BIT_MASK(40));
-	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
+	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(40)))
+		pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(40));
+	pci_read_config_word(pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
 
 	ggtt->stolen_size = gen6_get_stolen_size(snb_gmch_ctl);
-	ggtt->size = gen6_get_total_gtt_size(snb_gmch_ctl);
-	ggtt->base.total = (ggtt->size / sizeof(gen6_pte_t)) << PAGE_SHIFT;
 
-	ret = ggtt_probe_common(dev, ggtt->size);
+	size = gen6_get_total_gtt_size(snb_gmch_ctl);
+	ggtt->base.total = (size / sizeof(gen6_pte_t)) << PAGE_SHIFT;
 
 	ggtt->base.clear_range = gen6_ggtt_clear_range;
 	ggtt->base.insert_page = gen6_ggtt_insert_page;
 	ggtt->base.insert_entries = gen6_ggtt_insert_entries;
 	ggtt->base.bind_vma = ggtt_bind_vma;
 	ggtt->base.unbind_vma = ggtt_unbind_vma;
+	ggtt->base.cleanup = gen6_gmch_remove;
+
+	if (HAS_EDRAM(dev_priv))
+		ggtt->base.pte_encode = iris_pte_encode;
+	else if (IS_HASWELL(dev_priv))
+		ggtt->base.pte_encode = hsw_pte_encode;
+	else if (IS_VALLEYVIEW(dev_priv))
+		ggtt->base.pte_encode = byt_pte_encode;
+	else if (INTEL_GEN(dev_priv) >= 7)
+		ggtt->base.pte_encode = ivb_pte_encode;
+	else
+		ggtt->base.pte_encode = snb_pte_encode;
 
-	return ret;
+	return ggtt_probe_common(ggtt, size);
 }
 
-static void gen6_gmch_remove(struct i915_address_space *vm)
+static void i915_gmch_remove(struct i915_address_space *vm)
 {
-	struct i915_ggtt *ggtt = container_of(vm, struct i915_ggtt, base);
-
-	iounmap(ggtt->gsm);
-	free_scratch_page(vm->dev, vm->scratch_page);
+	intel_gmch_remove();
 }
 
 static int i915_gmch_probe(struct i915_ggtt *ggtt)
 {
-	struct drm_device *dev = ggtt->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_private *dev_priv = to_i915(ggtt->base.dev);
 	int ret;
 
 	ret = intel_gmch_probe(dev_priv->bridge_dev, dev_priv->drm.pdev, NULL);
@@ -3188,12 +3128,13 @@ static int i915_gmch_probe(struct i915_ggtt *ggtt)
 	intel_gtt_get(&ggtt->base.total, &ggtt->stolen_size,
 		      &ggtt->mappable_base, &ggtt->mappable_end);
 
-	ggtt->do_idle_maps = needs_idle_maps(&dev_priv->drm);
+	ggtt->do_idle_maps = needs_idle_maps(dev_priv);
 	ggtt->base.insert_page = i915_ggtt_insert_page;
 	ggtt->base.insert_entries = i915_ggtt_insert_entries;
 	ggtt->base.clear_range = i915_ggtt_clear_range;
 	ggtt->base.bind_vma = ggtt_bind_vma;
 	ggtt->base.unbind_vma = ggtt_unbind_vma;
+	ggtt->base.cleanup = i915_gmch_remove;
 
 	if (unlikely(ggtt->do_idle_maps))
 		DRM_INFO("applying Ironlake quirks for intel_iommu\n");
@@ -3201,65 +3142,40 @@ static int i915_gmch_probe(struct i915_ggtt *ggtt)
 	return 0;
 }
 
-static void i915_gmch_remove(struct i915_address_space *vm)
-{
-	intel_gmch_remove();
-}
-
 /**
- * i915_ggtt_init_hw - Initialize GGTT hardware
- * @dev: DRM device
+ * i915_ggtt_probe_hw - Probe GGTT hardware location
+ * @dev_priv: i915 device
  */
-int i915_ggtt_init_hw(struct drm_device *dev)
+int i915_ggtt_probe_hw(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	int ret;
 
-	if (INTEL_INFO(dev)->gen <= 5) {
-		ggtt->probe = i915_gmch_probe;
-		ggtt->base.cleanup = i915_gmch_remove;
-	} else if (INTEL_INFO(dev)->gen < 8) {
-		ggtt->probe = gen6_gmch_probe;
-		ggtt->base.cleanup = gen6_gmch_remove;
-
-		if (HAS_EDRAM(dev))
-			ggtt->base.pte_encode = iris_pte_encode;
-		else if (IS_HASWELL(dev))
-			ggtt->base.pte_encode = hsw_pte_encode;
-		else if (IS_VALLEYVIEW(dev))
-			ggtt->base.pte_encode = byt_pte_encode;
-		else if (INTEL_INFO(dev)->gen >= 7)
-			ggtt->base.pte_encode = ivb_pte_encode;
-		else
-			ggtt->base.pte_encode = snb_pte_encode;
-	} else {
-		ggtt->probe = gen8_gmch_probe;
-		ggtt->base.cleanup = gen6_gmch_remove;
-	}
+	ggtt->base.dev = &dev_priv->drm;
 
-	ggtt->base.dev = dev;
-	ggtt->base.is_ggtt = true;
-
-	ret = ggtt->probe(ggtt);
+	if (INTEL_GEN(dev_priv) <= 5)
+		ret = i915_gmch_probe(ggtt);
+	else if (INTEL_GEN(dev_priv) < 8)
+		ret = gen6_gmch_probe(ggtt);
+	else
+		ret = gen8_gmch_probe(ggtt);
 	if (ret)
 		return ret;
 
 	if ((ggtt->base.total - 1) >> 32) {
 		DRM_ERROR("We never expected a Global GTT with more than 32bits"
-			  "of address space! Found %lldM!\n",
+			  " of address space! Found %lldM!\n",
 			  ggtt->base.total >> 20);
 		ggtt->base.total = 1ULL << 32;
 		ggtt->mappable_end = min(ggtt->mappable_end, ggtt->base.total);
 	}
 
-	/*
-	 * Initialise stolen early so that we may reserve preallocated
-	 * objects for the BIOS to KMS transition.
-	 */
-	ret = i915_gem_init_stolen(dev);
-	if (ret)
-		goto out_gtt_cleanup;
+	if (ggtt->mappable_end > ggtt->base.total) {
+		DRM_ERROR("mappable aperture extends past end of GGTT,"
+			  " aperture=%llx, total=%llx\n",
+			  ggtt->mappable_end, ggtt->base.total);
+		ggtt->mappable_end = ggtt->base.total;
+	}
 
 	/* GMADR is the PCI mmio aperture into the global GTT. */
 	DRM_INFO("Memory usable by graphics device = %lluM\n",
@@ -3272,16 +3188,55 @@ int i915_ggtt_init_hw(struct drm_device *dev)
 #endif
 
 	return 0;
+}
+
+/**
+ * i915_ggtt_init_hw - Initialize GGTT hardware
+ * @dev_priv: i915 device
+ */
+int i915_ggtt_init_hw(struct drm_i915_private *dev_priv)
+{
+	struct i915_ggtt *ggtt = &dev_priv->ggtt;
+	int ret;
+
+	INIT_LIST_HEAD(&dev_priv->vm_list);
+
+	/* Subtract the guard page before address space initialization to
+	 * shrink the range used by drm_mm.
+	 */
+	ggtt->base.total -= PAGE_SIZE;
+	i915_address_space_init(&ggtt->base, dev_priv);
+	ggtt->base.total += PAGE_SIZE;
+	if (!HAS_LLC(dev_priv))
+		ggtt->base.mm.color_adjust = i915_gtt_color_adjust;
+
+	ggtt->mappable =
+		io_mapping_create_wc(ggtt->mappable_base, ggtt->mappable_end);
+	if (!ggtt->mappable) {
+		ret = -EIO;
+		goto out_gtt_cleanup;
+	}
+
+	ggtt->mtrr = arch_phys_wc_add(ggtt->mappable_base, ggtt->mappable_end);
+
+	/*
+	 * Initialise stolen early so that we may reserve preallocated
+	 * objects for the BIOS to KMS transition.
+	 */
+	ret = i915_gem_init_stolen(&dev_priv->drm);
+	if (ret)
+		goto out_gtt_cleanup;
+
+	return 0;
 
 out_gtt_cleanup:
 	ggtt->base.cleanup(&ggtt->base);
-
 	return ret;
 }
 
-int i915_ggtt_enable_hw(struct drm_device *dev)
+int i915_ggtt_enable_hw(struct drm_i915_private *dev_priv)
 {
-	if (INTEL_INFO(dev)->gen < 6 && !intel_enable_gtt())
+	if (INTEL_GEN(dev_priv) < 6 && !intel_enable_gtt())
 		return -EIO;
 
 	return 0;
@@ -3331,7 +3286,7 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 
 			struct i915_hw_ppgtt *ppgtt;
 
-			if (vm->is_ggtt)
+			if (i915_is_ggtt(vm))
 				ppgtt = dev_priv->mm.aliasing_ppgtt;
 			else
 				ppgtt = i915_vm_to_ppgtt(vm);
@@ -3344,31 +3299,88 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 	i915_ggtt_flush(dev_priv);
 }
 
+static void
+i915_vma_retire(struct i915_gem_active *active,
+		struct drm_i915_gem_request *rq)
+{
+	const unsigned int idx = rq->engine->id;
+	struct i915_vma *vma =
+		container_of(active, struct i915_vma, last_read[idx]);
+
+	GEM_BUG_ON(!i915_vma_has_active_engine(vma, idx));
+
+	i915_vma_clear_active(vma, idx);
+	if (i915_vma_is_active(vma))
+		return;
+
+	list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
+	if (unlikely(i915_vma_is_closed(vma) && !i915_vma_is_pinned(vma)))
+		WARN_ON(i915_vma_unbind(vma));
+}
+
+void i915_vma_destroy(struct i915_vma *vma)
+{
+	GEM_BUG_ON(vma->node.allocated);
+	GEM_BUG_ON(i915_vma_is_active(vma));
+	GEM_BUG_ON(!i915_vma_is_closed(vma));
+
+	list_del(&vma->vm_link);
+	if (!i915_vma_is_ggtt(vma))
+		i915_ppgtt_put(i915_vm_to_ppgtt(vma->vm));
+
+	kmem_cache_free(to_i915(vma->obj->base.dev)->vmas, vma);
+}
+
+void i915_vma_close(struct i915_vma *vma)
+{
+	GEM_BUG_ON(i915_vma_is_closed(vma));
+	vma->flags |= I915_VMA_CLOSED;
+
+	list_del_init(&vma->obj_link);
+	if (!i915_vma_is_active(vma) && !i915_vma_is_pinned(vma))
+		WARN_ON(i915_vma_unbind(vma));
+}
+
 static struct i915_vma *
 __i915_gem_vma_create(struct drm_i915_gem_object *obj,
 		      struct i915_address_space *vm,
-		      const struct i915_ggtt_view *ggtt_view)
+		      const struct i915_ggtt_view *view)
 {
 	struct i915_vma *vma;
+	int i;
+
+	GEM_BUG_ON(vm->closed);
 
-	if (WARN_ON(i915_is_ggtt(vm) != !!ggtt_view))
+	if (WARN_ON(i915_is_ggtt(vm) != !!view))
 		return ERR_PTR(-EINVAL);
 
 	vma = kmem_cache_zalloc(to_i915(obj->base.dev)->vmas, GFP_KERNEL);
 	if (vma == NULL)
 		return ERR_PTR(-ENOMEM);
 
-	INIT_LIST_HEAD(&vma->vm_link);
 	INIT_LIST_HEAD(&vma->obj_link);
 	INIT_LIST_HEAD(&vma->exec_list);
+	for (i = 0; i < ARRAY_SIZE(vma->last_read); i++)
+		init_request_active(&vma->last_read[i], i915_vma_retire);
+	list_add(&vma->vm_link, &vm->unbound_list);
 	vma->vm = vm;
 	vma->obj = obj;
-	vma->is_ggtt = i915_is_ggtt(vm);
-
-	if (i915_is_ggtt(vm))
-		vma->ggtt_view = *ggtt_view;
-	else
+	vma->size = obj->base.size;
+
+	if (i915_is_ggtt(vm)) {
+		vma->flags |= I915_VMA_GGTT;
+		vma->ggtt_view = *view;
+		if (view->type == I915_GGTT_VIEW_PARTIAL) {
+			vma->size = view->params.partial.size;
+			vma->size <<= PAGE_SHIFT;
+		} else if (view->type == I915_GGTT_VIEW_ROTATED) {
+			vma->size =
+				intel_rotation_info_size(&view->params.rotated);
+			vma->size <<= PAGE_SHIFT;
+		}
+	} else {
 		i915_ppgtt_get(i915_vm_to_ppgtt(vm));
+	}
 
 	list_add_tail(&vma->obj_link, &obj->vma_list);
 
@@ -3398,9 +3410,12 @@ i915_gem_obj_lookup_or_create_ggtt_vma(struct drm_i915_gem_object *obj,
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct i915_vma *vma = i915_gem_obj_to_ggtt_view(obj, view);
 
+	GEM_BUG_ON(!view);
+
 	if (!vma)
 		vma = __i915_gem_vma_create(obj, &ggtt->base, view);
 
+	GEM_BUG_ON(i915_vma_is_closed(vma));
 	return vma;
 
 }
@@ -3611,34 +3626,32 @@ i915_get_ggtt_vma_pages(struct i915_vma *vma)
 int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 		  u32 flags)
 {
-	int ret;
 	u32 bind_flags;
+	u32 vma_flags;
+	int ret;
 
 	if (WARN_ON(flags == 0))
 		return -EINVAL;
 
 	bind_flags = 0;
 	if (flags & PIN_GLOBAL)
-		bind_flags |= GLOBAL_BIND;
+		bind_flags |= I915_VMA_GLOBAL_BIND;
 	if (flags & PIN_USER)
-		bind_flags |= LOCAL_BIND;
+		bind_flags |= I915_VMA_LOCAL_BIND;
 
+	vma_flags = vma->flags & (I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND);
 	if (flags & PIN_UPDATE)
-		bind_flags |= vma->bound;
+		bind_flags |= vma_flags;
 	else
-		bind_flags &= ~vma->bound;
-
+		bind_flags &= ~vma_flags;
 	if (bind_flags == 0)
 		return 0;
 
-	if (vma->bound == 0 && vma->vm->allocate_va_range) {
-		/* XXX: i915_vma_pin() will fix this +- hack */
-		vma->pin_count++;
+	if (vma_flags == 0 && vma->vm->allocate_va_range) {
 		trace_i915_va_alloc(vma);
 		ret = vma->vm->allocate_va_range(vma->vm,
 						 vma->node.start,
 						 vma->node.size);
-		vma->pin_count--;
 		if (ret)
 			return ret;
 	}
@@ -3647,44 +3660,20 @@ int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 	if (ret)
 		return ret;
 
-	vma->bound |= bind_flags;
-
+	vma->flags |= bind_flags;
 	return 0;
 }
 
-/**
- * i915_ggtt_view_size - Get the size of a GGTT view.
- * @obj: Object the view is of.
- * @view: The view in question.
- *
- * @return The size of the GGTT view in bytes.
- */
-size_t
-i915_ggtt_view_size(struct drm_i915_gem_object *obj,
-		    const struct i915_ggtt_view *view)
-{
-	if (view->type == I915_GGTT_VIEW_NORMAL) {
-		return obj->base.size;
-	} else if (view->type == I915_GGTT_VIEW_ROTATED) {
-		return intel_rotation_info_size(&view->params.rotated) << PAGE_SHIFT;
-	} else if (view->type == I915_GGTT_VIEW_PARTIAL) {
-		return view->params.partial.size << PAGE_SHIFT;
-	} else {
-		WARN_ONCE(1, "GGTT view %u not implemented!\n", view->type);
-		return obj->base.size;
-	}
-}
-
 void __iomem *i915_vma_pin_iomap(struct i915_vma *vma)
 {
 	void __iomem *ptr;
 
 	lockdep_assert_held(&vma->vm->dev->struct_mutex);
 	if (WARN_ON(!vma->obj->map_and_fenceable))
-		return ERR_PTR(-ENODEV);
+		return IO_ERR_PTR(-ENODEV);
 
-	GEM_BUG_ON(!vma->is_ggtt);
-	GEM_BUG_ON((vma->bound & GLOBAL_BIND) == 0);
+	GEM_BUG_ON(!i915_vma_is_ggtt(vma));
+	GEM_BUG_ON((vma->flags & I915_VMA_GLOBAL_BIND) == 0);
 
 	ptr = vma->iomap;
 	if (ptr == NULL) {
@@ -3692,11 +3681,11 @@ void __iomem *i915_vma_pin_iomap(struct i915_vma *vma)
 					vma->node.start,
 					vma->node.size);
 		if (ptr == NULL)
-			return ERR_PTR(-ENOMEM);
+			return IO_ERR_PTR(-ENOMEM);
 
 		vma->iomap = ptr;
 	}
 
-	vma->pin_count++;
+	__i915_vma_pin(vma);
 	return ptr;
 }

drivers/gpu/drm/i915/i915_gem_gtt.h

@@ -36,6 +36,8 @@
 
 #include <linux/io-mapping.h>
 
+#include "i915_gem_request.h"
+
 struct drm_i915_file_private;
 
 typedef uint32_t gen6_pte_t;
@@ -178,12 +180,32 @@ struct i915_vma {
 	struct drm_i915_gem_object *obj;
 	struct i915_address_space *vm;
 	void __iomem *iomap;
+	u64 size;
+
+	unsigned int flags;
+	/**
+	 * How many users have pinned this object in GTT space. The following
+	 * users can each hold at most one reference: pwrite/pread, execbuffer
+	 * (objects are not allowed multiple times for the same batchbuffer),
+	 * and the framebuffer code. When switching/pageflipping, the
+	 * framebuffer code has at most two buffers pinned per crtc.
+	 *
+	 * In the worst case this is 1 + 1 + 1 + 2*2 = 7. That would fit into 3
+	 * bits with absolutely no headroom. So use 4 bits.
+	 */
+#define I915_VMA_PIN_MASK 0xf
+#define I915_VMA_PIN_OVERFLOW	BIT(5)
 
 	/** Flags and address space this VMA is bound to */
-#define GLOBAL_BIND	(1<<0)
-#define LOCAL_BIND	(1<<1)
-	unsigned int bound : 4;
-	bool is_ggtt : 1;
+#define I915_VMA_GLOBAL_BIND	BIT(6)
+#define I915_VMA_LOCAL_BIND	BIT(7)
+#define I915_VMA_BIND_MASK (I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND | I915_VMA_PIN_OVERFLOW)
+
+#define I915_VMA_GGTT	BIT(8)
+#define I915_VMA_CLOSED BIT(9)
+
+	unsigned int active;
+	struct i915_gem_active last_read[I915_NUM_ENGINES];
 
 	/**
 	 * Support different GGTT views into the same object.
@@ -208,20 +230,46 @@ struct i915_vma {
 	struct hlist_node exec_node;
 	unsigned long exec_handle;
 	struct drm_i915_gem_exec_object2 *exec_entry;
-
-	/**
-	 * How many users have pinned this object in GTT space. The following
-	 * users can each hold at most one reference: pwrite/pread, execbuffer
-	 * (objects are not allowed multiple times for the same batchbuffer),
-	 * and the framebuffer code. When switching/pageflipping, the
-	 * framebuffer code has at most two buffers pinned per crtc.
-	 *
-	 * In the worst case this is 1 + 1 + 1 + 2*2 = 7. That would fit into 3
-	 * bits with absolutely no headroom. So use 4 bits. */
-	unsigned int pin_count:4;
-#define DRM_I915_GEM_OBJECT_MAX_PIN_COUNT 0xf
 };
 
+static inline bool i915_vma_is_ggtt(const struct i915_vma *vma)
+{
+	return vma->flags & I915_VMA_GGTT;
+}
+
+static inline bool i915_vma_is_closed(const struct i915_vma *vma)
+{
+	return vma->flags & I915_VMA_CLOSED;
+}
+
+static inline unsigned int i915_vma_get_active(const struct i915_vma *vma)
+{
+	return vma->active;
+}
+
+static inline bool i915_vma_is_active(const struct i915_vma *vma)
+{
+	return i915_vma_get_active(vma);
+}
+
+static inline void i915_vma_set_active(struct i915_vma *vma,
+				       unsigned int engine)
+{
+	vma->active |= BIT(engine);
+}
+
+static inline void i915_vma_clear_active(struct i915_vma *vma,
+					 unsigned int engine)
+{
+	vma->active &= ~BIT(engine);
+}
+
+static inline bool i915_vma_has_active_engine(const struct i915_vma *vma,
+					      unsigned int engine)
+{
+	return vma->active & BIT(engine);
+}
+
 struct i915_page_dma {
 	struct page *page;
 	union {
@@ -272,11 +320,20 @@ struct i915_pml4 {
 struct i915_address_space {
 	struct drm_mm mm;
 	struct drm_device *dev;
+	/* Every address space belongs to a struct file - except for the global
+	 * GTT that is owned by the driver (and so @file is set to NULL). In
+	 * principle, no information should leak from one context to another
+	 * (or between files/processes etc) unless explicitly shared by the
+	 * owner. Tracking the owner is important in order to free up per-file
+	 * objects along with the file, to aide resource tracking, and to
+	 * assign blame.
+	 */
+	struct drm_i915_file_private *file;
 	struct list_head global_link;
 	u64 start;		/* Start offset always 0 for dri2 */
 	u64 total;		/* size addr space maps (ex. 2GB for ggtt) */
 
-	bool is_ggtt;
+	bool closed;
 
 	struct i915_page_scratch *scratch_page;
 	struct i915_page_table *scratch_pt;
@@ -306,6 +363,13 @@ struct i915_address_space {
 	 */
 	struct list_head inactive_list;
 
+	/**
+	 * List of vma that have been unbound.
+	 *
+	 * A reference is not held on the buffer while on this list.
+	 */
+	struct list_head unbound_list;
+
 	/* FIXME: Need a more generic return type */
 	gen6_pte_t (*pte_encode)(dma_addr_t addr,
 				 enum i915_cache_level level,
@@ -338,7 +402,7 @@ struct i915_address_space {
 			u32 flags);
 };
 
-#define i915_is_ggtt(V) ((V)->is_ggtt)
+#define i915_is_ggtt(V) (!(V)->file)
 
 /* 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
@@ -354,7 +418,6 @@ struct i915_ggtt {
 	size_t stolen_usable_size;	/* Total size minus BIOS reserved */
 	size_t stolen_reserved_base;
 	size_t stolen_reserved_size;
-	size_t size;			/* Total size of Global GTT */
 	u64 mappable_end;		/* End offset that we can CPU map */
 	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
 	phys_addr_t mappable_base;	/* PA of our GMADR */
@@ -365,8 +428,6 @@ struct i915_ggtt {
 	bool do_idle_maps;
 
 	int mtrr;
-
-	int (*probe)(struct i915_ggtt *ggtt);
 };
 
 struct i915_hw_ppgtt {
@@ -380,8 +441,6 @@ struct i915_hw_ppgtt {
 		struct i915_page_directory pd;		/* GEN6-7 */
 	};
 
-	struct drm_i915_file_private *file_priv;
-
 	gen6_pte_t __iomem *pd_addr;
 
 	int (*enable)(struct i915_hw_ppgtt *ppgtt);
@@ -521,14 +580,15 @@ i915_page_dir_dma_addr(const struct i915_hw_ppgtt *ppgtt, const unsigned n)
 		px_dma(ppgtt->base.scratch_pd);
 }
 
-int i915_ggtt_init_hw(struct drm_device *dev);
-int i915_ggtt_enable_hw(struct drm_device *dev);
-void i915_gem_init_ggtt(struct drm_device *dev);
-void i915_ggtt_cleanup_hw(struct drm_device *dev);
+int i915_ggtt_probe_hw(struct drm_i915_private *dev_priv);
+int i915_ggtt_init_hw(struct drm_i915_private *dev_priv);
+int i915_ggtt_enable_hw(struct drm_i915_private *dev_priv);
+int i915_gem_init_ggtt(struct drm_i915_private *dev_priv);
+void i915_ggtt_cleanup_hw(struct drm_i915_private *dev_priv);
 
 int i915_ppgtt_init_hw(struct drm_device *dev);
 void i915_ppgtt_release(struct kref *kref);
-struct i915_hw_ppgtt *i915_ppgtt_create(struct drm_device *dev,
+struct i915_hw_ppgtt *i915_ppgtt_create(struct drm_i915_private *dev_priv,
 					struct drm_i915_file_private *fpriv);
 static inline void i915_ppgtt_get(struct i915_hw_ppgtt *ppgtt)
 {
@@ -562,9 +622,66 @@ i915_ggtt_view_equal(const struct i915_ggtt_view *a,
 	return true;
 }
 
-size_t
-i915_ggtt_view_size(struct drm_i915_gem_object *obj,
-		    const struct i915_ggtt_view *view);
+/* Flags used by pin/bind&friends. */
+#define PIN_NONBLOCK		BIT(0)
+#define PIN_MAPPABLE		BIT(1)
+#define PIN_ZONE_4G		BIT(2)
+
+#define PIN_MBZ			BIT(5) /* I915_VMA_PIN_OVERFLOW */
+#define PIN_GLOBAL		BIT(6) /* I915_VMA_GLOBAL_BIND */
+#define PIN_USER		BIT(7) /* I915_VMA_LOCAL_BIND */
+#define PIN_UPDATE		BIT(8)
+
+#define PIN_HIGH		BIT(9)
+#define PIN_OFFSET_BIAS		BIT(10)
+#define PIN_OFFSET_FIXED	BIT(11)
+#define PIN_OFFSET_MASK		(~4095)
+
+int __i915_vma_do_pin(struct i915_vma *vma,
+		      u64 size, u64 alignment, u64 flags);
+static inline int __must_check
+i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
+{
+	BUILD_BUG_ON(PIN_MBZ != I915_VMA_PIN_OVERFLOW);
+	BUILD_BUG_ON(PIN_GLOBAL != I915_VMA_GLOBAL_BIND);
+	BUILD_BUG_ON(PIN_USER != I915_VMA_LOCAL_BIND);
+
+	/* Pin early to prevent the shrinker/eviction logic from destroying
+	 * our vma as we insert and bind.
+	 */
+	if (likely(((++vma->flags ^ flags) & I915_VMA_BIND_MASK) == 0))
+		return 0;
+
+	return __i915_vma_do_pin(vma, size, alignment, flags);
+}
+
+static inline int i915_vma_pin_count(const struct i915_vma *vma)
+{
+	return vma->flags & I915_VMA_PIN_MASK;
+}
+
+static inline bool i915_vma_is_pinned(const struct i915_vma *vma)
+{
+	return i915_vma_pin_count(vma);
+}
+
+static inline void __i915_vma_pin(struct i915_vma *vma)
+{
+	vma->flags++;
+	GEM_BUG_ON(vma->flags & I915_VMA_PIN_OVERFLOW);
+}
+
+static inline void __i915_vma_unpin(struct i915_vma *vma)
+{
+	GEM_BUG_ON(!i915_vma_is_pinned(vma));
+	vma->flags--;
+}
+
+static inline void i915_vma_unpin(struct i915_vma *vma)
+{
+	GEM_BUG_ON(!drm_mm_node_allocated(&vma->node));
+	__i915_vma_unpin(vma);
+}
 
 /**
  * i915_vma_pin_iomap - calls ioremap_wc to map the GGTT VMA via the aperture
@@ -580,6 +697,7 @@ i915_ggtt_view_size(struct drm_i915_gem_object *obj,
  * Returns a valid iomapped pointer or ERR_PTR.
  */
 void __iomem *i915_vma_pin_iomap(struct i915_vma *vma);
+#define IO_ERR_PTR(x) ((void __iomem *)ERR_PTR(x))
 
 /**
  * i915_vma_unpin_iomap - unpins the mapping returned from i915_vma_iomap
@@ -593,9 +711,8 @@ void __iomem *i915_vma_pin_iomap(struct i915_vma *vma);
 static inline void i915_vma_unpin_iomap(struct i915_vma *vma)
 {
 	lockdep_assert_held(&vma->vm->dev->struct_mutex);
-	GEM_BUG_ON(vma->pin_count == 0);
 	GEM_BUG_ON(vma->iomap == NULL);
-	vma->pin_count--;
+	i915_vma_unpin(vma);
 }
 
 #endif

drivers/gpu/drm/i915/i915_gem_render_state.c

@@ -28,10 +28,18 @@
 #include "i915_drv.h"
 #include "intel_renderstate.h"
 
+struct render_state {
+	const struct intel_renderstate_rodata *rodata;
+	struct drm_i915_gem_object *obj;
+	u64 ggtt_offset;
+	u32 aux_batch_size;
+	u32 aux_batch_offset;
+};
+
 static const struct intel_renderstate_rodata *
-render_state_get_rodata(const int gen)
+render_state_get_rodata(const struct drm_i915_gem_request *req)
 {
-	switch (gen) {
+	switch (INTEL_GEN(req->i915)) {
 	case 6:
 		return &gen6_null_state;
 	case 7:
@@ -45,35 +53,6 @@ render_state_get_rodata(const int gen)
 	return NULL;
 }
 
-static int render_state_init(struct render_state *so,
-			     struct drm_i915_private *dev_priv)
-{
-	int ret;
-
-	so->gen = INTEL_GEN(dev_priv);
-	so->rodata = render_state_get_rodata(so->gen);
-	if (so->rodata == NULL)
-		return 0;
-
-	if (so->rodata->batch_items * 4 > 4096)
-		return -EINVAL;
-
-	so->obj = i915_gem_object_create(&dev_priv->drm, 4096);
-	if (IS_ERR(so->obj))
-		return PTR_ERR(so->obj);
-
-	ret = i915_gem_obj_ggtt_pin(so->obj, 4096, 0);
-	if (ret)
-		goto free_gem;
-
-	so->ggtt_offset = i915_gem_obj_ggtt_offset(so->obj);
-	return 0;
-
-free_gem:
-	drm_gem_object_unreference(&so->obj->base);
-	return ret;
-}
-
 /*
  * Macro to add commands to auxiliary batch.
  * This macro only checks for page overflow before inserting the commands,
@@ -96,6 +75,7 @@ static int render_state_setup(struct render_state *so)
 {
 	struct drm_device *dev = so->obj->base.dev;
 	const struct intel_renderstate_rodata *rodata = so->rodata;
+	const bool has_64bit_reloc = INTEL_GEN(dev) >= 8;
 	unsigned int i = 0, reloc_index = 0;
 	struct page *page;
 	u32 *d;
@@ -114,7 +94,7 @@ static int render_state_setup(struct render_state *so)
 		if (i * 4  == rodata->reloc[reloc_index]) {
 			u64 r = s + so->ggtt_offset;
 			s = lower_32_bits(r);
-			if (so->gen >= 8) {
+			if (has_64bit_reloc) {
 				if (i + 1 >= rodata->batch_items ||
 				    rodata->batch[i + 1] != 0) {
 					ret = -EINVAL;
@@ -192,67 +172,55 @@ err_out:
 
 #undef OUT_BATCH
 
-void i915_gem_render_state_fini(struct render_state *so)
-{
-	i915_gem_object_ggtt_unpin(so->obj);
-	drm_gem_object_unreference(&so->obj->base);
-}
-
-int i915_gem_render_state_prepare(struct intel_engine_cs *engine,
-				  struct render_state *so)
+int i915_gem_render_state_init(struct drm_i915_gem_request *req)
 {
+	struct render_state so;
 	int ret;
 
-	if (WARN_ON(engine->id != RCS))
+	if (WARN_ON(req->engine->id != RCS))
 		return -ENOENT;
 
-	ret = render_state_init(so, engine->i915);
-	if (ret)
-		return ret;
-
-	if (so->rodata == NULL)
+	so.rodata = render_state_get_rodata(req);
+	if (!so.rodata)
 		return 0;
 
-	ret = render_state_setup(so);
-	if (ret) {
-		i915_gem_render_state_fini(so);
-		return ret;
-	}
+	if (so.rodata->batch_items * 4 > 4096)
+		return -EINVAL;
 
-	return 0;
-}
+	so.obj = i915_gem_object_create(&req->i915->drm, 4096);
+	if (IS_ERR(so.obj))
+		return PTR_ERR(so.obj);
 
-int i915_gem_render_state_init(struct drm_i915_gem_request *req)
-{
-	struct render_state so;
-	int ret;
-
-	ret = i915_gem_render_state_prepare(req->engine, &so);
+	ret = i915_gem_object_ggtt_pin(so.obj, NULL, 0, 0, 0);
 	if (ret)
-		return ret;
+		goto err_obj;
 
-	if (so.rodata == NULL)
-		return 0;
+	so.ggtt_offset = i915_gem_obj_ggtt_offset(so.obj);
+
+	ret = render_state_setup(&so);
+	if (ret)
+		goto err_unpin;
 
-	ret = req->engine->dispatch_execbuffer(req, so.ggtt_offset,
-					     so.rodata->batch_items * 4,
-					     I915_DISPATCH_SECURE);
+	ret = req->engine->emit_bb_start(req, so.ggtt_offset,
+					 so.rodata->batch_items * 4,
+					 I915_DISPATCH_SECURE);
 	if (ret)
-		goto out;
+		goto err_unpin;
 
 	if (so.aux_batch_size > 8) {
-		ret = req->engine->dispatch_execbuffer(req,
-						     (so.ggtt_offset +
-						      so.aux_batch_offset),
-						     so.aux_batch_size,
-						     I915_DISPATCH_SECURE);
+		ret = req->engine->emit_bb_start(req,
+						 (so.ggtt_offset +
+						  so.aux_batch_offset),
+						 so.aux_batch_size,
+						 I915_DISPATCH_SECURE);
 		if (ret)
-			goto out;
+			goto err_unpin;
 	}
 
-	i915_vma_move_to_active(i915_gem_obj_to_ggtt(so.obj), req);
-
-out:
-	i915_gem_render_state_fini(&so);
+	i915_vma_move_to_active(i915_gem_obj_to_ggtt(so.obj), req, 0);
+err_unpin:
+	i915_gem_object_ggtt_unpin(so.obj);
+err_obj:
+	i915_gem_object_put(so.obj);
 	return ret;
 }

drivers/gpu/drm/i915/i915_gem_render_state.h

@@ -26,24 +26,6 @@
 
 #include <linux/types.h>
 
-struct intel_renderstate_rodata {
-	const u32 *reloc;
-	const u32 *batch;
-	const u32 batch_items;
-};
-
-struct render_state {
-	const struct intel_renderstate_rodata *rodata;
-	struct drm_i915_gem_object *obj;
-	u64 ggtt_offset;
-	int gen;
-	u32 aux_batch_size;
-	u32 aux_batch_offset;
-};
-
 int i915_gem_render_state_init(struct drm_i915_gem_request *req);
-void i915_gem_render_state_fini(struct render_state *so);
-int i915_gem_render_state_prepare(struct intel_engine_cs *engine,
-				  struct render_state *so);
 
 #endif /* _I915_GEM_RENDER_STATE_H_ */

drivers/gpu/drm/i915/i915_gem_request.c

@@ -0,0 +1,767 @@
+/*
+ * Copyright © 2008-2015 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/prefetch.h>
+
+#include "i915_drv.h"
+
+static const char *i915_fence_get_driver_name(struct fence *fence)
+{
+	return "i915";
+}
+
+static const char *i915_fence_get_timeline_name(struct fence *fence)
+{
+	/* Timelines are bound by eviction to a VM. However, since
+	 * we only have a global seqno at the moment, we only have
+	 * a single timeline. Note that each timeline will have
+	 * multiple execution contexts (fence contexts) as we allow
+	 * engines within a single timeline to execute in parallel.
+	 */
+	return "global";
+}
+
+static bool i915_fence_signaled(struct fence *fence)
+{
+	return i915_gem_request_completed(to_request(fence));
+}
+
+static bool i915_fence_enable_signaling(struct fence *fence)
+{
+	if (i915_fence_signaled(fence))
+		return false;
+
+	intel_engine_enable_signaling(to_request(fence));
+	return true;
+}
+
+static signed long i915_fence_wait(struct fence *fence,
+				   bool interruptible,
+				   signed long timeout_jiffies)
+{
+	s64 timeout_ns, *timeout;
+	int ret;
+
+	if (timeout_jiffies != MAX_SCHEDULE_TIMEOUT) {
+		timeout_ns = jiffies_to_nsecs(timeout_jiffies);
+		timeout = &timeout_ns;
+	} else {
+		timeout = NULL;
+	}
+
+	ret = i915_wait_request(to_request(fence),
+				interruptible, timeout,
+				NO_WAITBOOST);
+	if (ret == -ETIME)
+		return 0;
+
+	if (ret < 0)
+		return ret;
+
+	if (timeout_jiffies != MAX_SCHEDULE_TIMEOUT)
+		timeout_jiffies = nsecs_to_jiffies(timeout_ns);
+
+	return timeout_jiffies;
+}
+
+static void i915_fence_value_str(struct fence *fence, char *str, int size)
+{
+	snprintf(str, size, "%u", fence->seqno);
+}
+
+static void i915_fence_timeline_value_str(struct fence *fence, char *str,
+					  int size)
+{
+	snprintf(str, size, "%u",
+		 intel_engine_get_seqno(to_request(fence)->engine));
+}
+
+static void i915_fence_release(struct fence *fence)
+{
+	struct drm_i915_gem_request *req = to_request(fence);
+
+	kmem_cache_free(req->i915->requests, req);
+}
+
+const struct fence_ops i915_fence_ops = {
+	.get_driver_name = i915_fence_get_driver_name,
+	.get_timeline_name = i915_fence_get_timeline_name,
+	.enable_signaling = i915_fence_enable_signaling,
+	.signaled = i915_fence_signaled,
+	.wait = i915_fence_wait,
+	.release = i915_fence_release,
+	.fence_value_str = i915_fence_value_str,
+	.timeline_value_str = i915_fence_timeline_value_str,
+};
+
+int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
+				   struct drm_file *file)
+{
+	struct drm_i915_private *dev_private;
+	struct drm_i915_file_private *file_priv;
+
+	WARN_ON(!req || !file || req->file_priv);
+
+	if (!req || !file)
+		return -EINVAL;
+
+	if (req->file_priv)
+		return -EINVAL;
+
+	dev_private = req->i915;
+	file_priv = file->driver_priv;
+
+	spin_lock(&file_priv->mm.lock);
+	req->file_priv = file_priv;
+	list_add_tail(&req->client_list, &file_priv->mm.request_list);
+	spin_unlock(&file_priv->mm.lock);
+
+	req->pid = get_pid(task_pid(current));
+
+	return 0;
+}
+
+static inline void
+i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
+{
+	struct drm_i915_file_private *file_priv = request->file_priv;
+
+	if (!file_priv)
+		return;
+
+	spin_lock(&file_priv->mm.lock);
+	list_del(&request->client_list);
+	request->file_priv = NULL;
+	spin_unlock(&file_priv->mm.lock);
+
+	put_pid(request->pid);
+	request->pid = NULL;
+}
+
+void i915_gem_retire_noop(struct i915_gem_active *active,
+			  struct drm_i915_gem_request *request)
+{
+	/* Space left intentionally blank */
+}
+
+static void i915_gem_request_retire(struct drm_i915_gem_request *request)
+{
+	struct i915_gem_active *active, *next;
+
+	trace_i915_gem_request_retire(request);
+	list_del(&request->link);
+
+	/* We know the GPU must have read the request to have
+	 * sent us the seqno + interrupt, so use the position
+	 * of tail of the request to update the last known position
+	 * of the GPU head.
+	 *
+	 * Note this requires that we are always called in request
+	 * completion order.
+	 */
+	list_del(&request->ring_link);
+	request->ring->last_retired_head = request->postfix;
+
+	/* Walk through the active list, calling retire on each. This allows
+	 * objects to track their GPU activity and mark themselves as idle
+	 * when their *last* active request is completed (updating state
+	 * tracking lists for eviction, active references for GEM, etc).
+	 *
+	 * As the ->retire() may free the node, we decouple it first and
+	 * pass along the auxiliary information (to avoid dereferencing
+	 * the node after the callback).
+	 */
+	list_for_each_entry_safe(active, next, &request->active_list, link) {
+		/* In microbenchmarks or focusing upon time inside the kernel,
+		 * we may spend an inordinate amount of time simply handling
+		 * the retirement of requests and processing their callbacks.
+		 * Of which, this loop itself is particularly hot due to the
+		 * cache misses when jumping around the list of i915_gem_active.
+		 * So we try to keep this loop as streamlined as possible and
+		 * also prefetch the next i915_gem_active to try and hide
+		 * the likely cache miss.
+		 */
+		prefetchw(next);
+
+		INIT_LIST_HEAD(&active->link);
+		RCU_INIT_POINTER(active->request, NULL);
+
+		active->retire(active, request);
+	}
+
+	i915_gem_request_remove_from_client(request);
+
+	if (request->previous_context) {
+		if (i915.enable_execlists)
+			intel_lr_context_unpin(request->previous_context,
+					       request->engine);
+	}
+
+	i915_gem_context_put(request->ctx);
+	i915_gem_request_put(request);
+}
+
+void i915_gem_request_retire_upto(struct drm_i915_gem_request *req)
+{
+	struct intel_engine_cs *engine = req->engine;
+	struct drm_i915_gem_request *tmp;
+
+	lockdep_assert_held(&req->i915->drm.struct_mutex);
+	GEM_BUG_ON(list_empty(&req->link));
+
+	do {
+		tmp = list_first_entry(&engine->request_list,
+				       typeof(*tmp), link);
+
+		i915_gem_request_retire(tmp);
+	} while (tmp != req);
+}
+
+static int i915_gem_check_wedge(unsigned int reset_counter, bool interruptible)
+{
+	if (__i915_terminally_wedged(reset_counter))
+		return -EIO;
+
+	if (__i915_reset_in_progress(reset_counter)) {
+		/* Non-interruptible callers can't handle -EAGAIN, hence return
+		 * -EIO unconditionally for these.
+		 */
+		if (!interruptible)
+			return -EIO;
+
+		return -EAGAIN;
+	}
+
+	return 0;
+}
+
+static int i915_gem_init_seqno(struct drm_i915_private *dev_priv, u32 seqno)
+{
+	struct intel_engine_cs *engine;
+	int ret;
+
+	/* Carefully retire all requests without writing to the rings */
+	for_each_engine(engine, dev_priv) {
+		ret = intel_engine_idle(engine, true);
+		if (ret)
+			return ret;
+	}
+	i915_gem_retire_requests(dev_priv);
+
+	/* If the seqno wraps around, we need to clear the breadcrumb rbtree */
+	if (!i915_seqno_passed(seqno, dev_priv->next_seqno)) {
+		while (intel_kick_waiters(dev_priv) ||
+		       intel_kick_signalers(dev_priv))
+			yield();
+	}
+
+	/* Finally reset hw state */
+	for_each_engine(engine, dev_priv)
+		intel_engine_init_seqno(engine, seqno);
+
+	return 0;
+}
+
+int i915_gem_set_seqno(struct drm_device *dev, u32 seqno)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	int ret;
+
+	if (seqno == 0)
+		return -EINVAL;
+
+	/* HWS page needs to be set less than what we
+	 * will inject to ring
+	 */
+	ret = i915_gem_init_seqno(dev_priv, seqno - 1);
+	if (ret)
+		return ret;
+
+	dev_priv->next_seqno = seqno;
+	return 0;
+}
+
+static int i915_gem_get_seqno(struct drm_i915_private *dev_priv, u32 *seqno)
+{
+	/* reserve 0 for non-seqno */
+	if (unlikely(dev_priv->next_seqno == 0)) {
+		int ret;
+
+		ret = i915_gem_init_seqno(dev_priv, 0);
+		if (ret)
+			return ret;
+
+		dev_priv->next_seqno = 1;
+	}
+
+	*seqno = dev_priv->next_seqno++;
+	return 0;
+}
+
+/**
+ * i915_gem_request_alloc - allocate a request structure
+ *
+ * @engine: engine that we wish to issue the request on.
+ * @ctx: context that the request will be associated with.
+ *       This can be NULL if the request is not directly related to
+ *       any specific user context, in which case this function will
+ *       choose an appropriate context to use.
+ *
+ * Returns a pointer to the allocated request if successful,
+ * or an error code if not.
+ */
+struct drm_i915_gem_request *
+i915_gem_request_alloc(struct intel_engine_cs *engine,
+		       struct i915_gem_context *ctx)
+{
+	struct drm_i915_private *dev_priv = engine->i915;
+	unsigned int reset_counter = i915_reset_counter(&dev_priv->gpu_error);
+	struct drm_i915_gem_request *req;
+	u32 seqno;
+	int ret;
+
+	/* ABI: Before userspace accesses the GPU (e.g. execbuffer), report
+	 * EIO if the GPU is already wedged, or EAGAIN to drop the struct_mutex
+	 * and restart.
+	 */
+	ret = i915_gem_check_wedge(reset_counter, dev_priv->mm.interruptible);
+	if (ret)
+		return ERR_PTR(ret);
+
+	/* Move the oldest request to the slab-cache (if not in use!) */
+	req = list_first_entry_or_null(&engine->request_list,
+				       typeof(*req), link);
+	if (req && i915_gem_request_completed(req))
+		i915_gem_request_retire(req);
+
+	req = kmem_cache_zalloc(dev_priv->requests, GFP_KERNEL);
+	if (!req)
+		return ERR_PTR(-ENOMEM);
+
+	ret = i915_gem_get_seqno(dev_priv, &seqno);
+	if (ret)
+		goto err;
+
+	spin_lock_init(&req->lock);
+	fence_init(&req->fence,
+		   &i915_fence_ops,
+		   &req->lock,
+		   engine->fence_context,
+		   seqno);
+
+	INIT_LIST_HEAD(&req->active_list);
+	req->i915 = dev_priv;
+	req->engine = engine;
+	req->ctx = i915_gem_context_get(ctx);
+
+	/*
+	 * Reserve space in the ring buffer for all the commands required to
+	 * eventually emit this request. This is to guarantee that the
+	 * i915_add_request() call can't fail. Note that the reserve may need
+	 * to be redone if the request is not actually submitted straight
+	 * away, e.g. because a GPU scheduler has deferred it.
+	 */
+	req->reserved_space = MIN_SPACE_FOR_ADD_REQUEST;
+
+	if (i915.enable_execlists)
+		ret = intel_logical_ring_alloc_request_extras(req);
+	else
+		ret = intel_ring_alloc_request_extras(req);
+	if (ret)
+		goto err_ctx;
+
+	return req;
+
+err_ctx:
+	i915_gem_context_put(ctx);
+err:
+	kmem_cache_free(dev_priv->requests, req);
+	return ERR_PTR(ret);
+}
+
+static void i915_gem_mark_busy(const struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *dev_priv = engine->i915;
+
+	dev_priv->gt.active_engines |= intel_engine_flag(engine);
+	if (dev_priv->gt.awake)
+		return;
+
+	intel_runtime_pm_get_noresume(dev_priv);
+	dev_priv->gt.awake = true;
+
+	intel_enable_gt_powersave(dev_priv);
+	i915_update_gfx_val(dev_priv);
+	if (INTEL_GEN(dev_priv) >= 6)
+		gen6_rps_busy(dev_priv);
+
+	queue_delayed_work(dev_priv->wq,
+			   &dev_priv->gt.retire_work,
+			   round_jiffies_up_relative(HZ));
+}
+
+/*
+ * NB: This function is not allowed to fail. Doing so would mean the the
+ * request is not being tracked for completion but the work itself is
+ * going to happen on the hardware. This would be a Bad Thing(tm).
+ */
+void __i915_add_request(struct drm_i915_gem_request *request,
+			struct drm_i915_gem_object *obj,
+			bool flush_caches)
+{
+	struct intel_engine_cs *engine;
+	struct intel_ring *ring;
+	u32 request_start;
+	u32 reserved_tail;
+	int ret;
+
+	if (WARN_ON(!request))
+		return;
+
+	engine = request->engine;
+	ring = request->ring;
+
+	/*
+	 * To ensure that this call will not fail, space for its emissions
+	 * should already have been reserved in the ring buffer. Let the ring
+	 * know that it is time to use that space up.
+	 */
+	request_start = ring->tail;
+	reserved_tail = request->reserved_space;
+	request->reserved_space = 0;
+
+	/*
+	 * Emit any outstanding flushes - execbuf can fail to emit the flush
+	 * after having emitted the batchbuffer command. Hence we need to fix
+	 * things up similar to emitting the lazy request. The difference here
+	 * is that the flush _must_ happen before the next request, no matter
+	 * what.
+	 */
+	if (flush_caches) {
+		ret = engine->emit_flush(request, EMIT_FLUSH);
+
+		/* Not allowed to fail! */
+		WARN(ret, "engine->emit_flush() failed: %d!\n", ret);
+	}
+
+	trace_i915_gem_request_add(request);
+
+	request->head = request_start;
+
+	/* Whilst this request exists, batch_obj will be on the
+	 * active_list, and so will hold the active reference. Only when this
+	 * request is retired will the the batch_obj be moved onto the
+	 * inactive_list and lose its active reference. Hence we do not need
+	 * to explicitly hold another reference here.
+	 */
+	request->batch_obj = obj;
+
+	/* Seal the request and mark it as pending execution. Note that
+	 * we may inspect this state, without holding any locks, during
+	 * hangcheck. Hence we apply the barrier to ensure that we do not
+	 * see a more recent value in the hws than we are tracking.
+	 */
+	request->emitted_jiffies = jiffies;
+	request->previous_seqno = engine->last_submitted_seqno;
+	engine->last_submitted_seqno = request->fence.seqno;
+	i915_gem_active_set(&engine->last_request, request);
+	list_add_tail(&request->link, &engine->request_list);
+	list_add_tail(&request->ring_link, &ring->request_list);
+
+	/* Record the position of the start of the request so that
+	 * should we detect the updated seqno part-way through the
+	 * GPU processing the request, we never over-estimate the
+	 * position of the head.
+	 */
+	request->postfix = ring->tail;
+
+	/* Not allowed to fail! */
+	ret = engine->emit_request(request);
+	WARN(ret, "(%s)->emit_request failed: %d!\n", engine->name, ret);
+
+	/* Sanity check that the reserved size was large enough. */
+	ret = ring->tail - request_start;
+	if (ret < 0)
+		ret += ring->size;
+	WARN_ONCE(ret > reserved_tail,
+		  "Not enough space reserved (%d bytes) "
+		  "for adding the request (%d bytes)\n",
+		  reserved_tail, ret);
+
+	i915_gem_mark_busy(engine);
+	engine->submit_request(request);
+}
+
+static unsigned long local_clock_us(unsigned int *cpu)
+{
+	unsigned long t;
+
+	/* Cheaply and approximately convert from nanoseconds to microseconds.
+	 * The result and subsequent calculations are also defined in the same
+	 * approximate microseconds units. The principal source of timing
+	 * error here is from the simple truncation.
+	 *
+	 * Note that local_clock() is only defined wrt to the current CPU;
+	 * the comparisons are no longer valid if we switch CPUs. Instead of
+	 * blocking preemption for the entire busywait, we can detect the CPU
+	 * switch and use that as indicator of system load and a reason to
+	 * stop busywaiting, see busywait_stop().
+	 */
+	*cpu = get_cpu();
+	t = local_clock() >> 10;
+	put_cpu();
+
+	return t;
+}
+
+static bool busywait_stop(unsigned long timeout, unsigned int cpu)
+{
+	unsigned int this_cpu;
+
+	if (time_after(local_clock_us(&this_cpu), timeout))
+		return true;
+
+	return this_cpu != cpu;
+}
+
+bool __i915_spin_request(const struct drm_i915_gem_request *req,
+			 int state, unsigned long timeout_us)
+{
+	unsigned int cpu;
+
+	/* When waiting for high frequency requests, e.g. during synchronous
+	 * rendering split between the CPU and GPU, the finite amount of time
+	 * required to set up the irq and wait upon it limits the response
+	 * rate. By busywaiting on the request completion for a short while we
+	 * can service the high frequency waits as quick as possible. However,
+	 * if it is a slow request, we want to sleep as quickly as possible.
+	 * The tradeoff between waiting and sleeping is roughly the time it
+	 * takes to sleep on a request, on the order of a microsecond.
+	 */
+
+	timeout_us += local_clock_us(&cpu);
+	do {
+		if (i915_gem_request_completed(req))
+			return true;
+
+		if (signal_pending_state(state, current))
+			break;
+
+		if (busywait_stop(timeout_us, cpu))
+			break;
+
+		cpu_relax_lowlatency();
+	} while (!need_resched());
+
+	return false;
+}
+
+/**
+ * i915_wait_request - wait until execution of request has finished
+ * @req: duh!
+ * @interruptible: do an interruptible wait (normally yes)
+ * @timeout: in - how long to wait (NULL forever); out - how much time remaining
+ * @rps: client to charge for RPS boosting
+ *
+ * Note: It is of utmost importance that the passed in seqno and reset_counter
+ * values have been read by the caller in an smp safe manner. Where read-side
+ * locks are involved, it is sufficient to read the reset_counter before
+ * unlocking the lock that protects the seqno. For lockless tricks, the
+ * reset_counter _must_ be read before, and an appropriate smp_rmb must be
+ * inserted.
+ *
+ * Returns 0 if the request was found within the alloted time. Else returns the
+ * errno with remaining time filled in timeout argument.
+ */
+int i915_wait_request(struct drm_i915_gem_request *req,
+		      bool interruptible,
+		      s64 *timeout,
+		      struct intel_rps_client *rps)
+{
+	int state = interruptible ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE;
+	DEFINE_WAIT(reset);
+	struct intel_wait wait;
+	unsigned long timeout_remain;
+	int ret = 0;
+
+	might_sleep();
+
+	if (i915_gem_request_completed(req))
+		return 0;
+
+	timeout_remain = MAX_SCHEDULE_TIMEOUT;
+	if (timeout) {
+		if (WARN_ON(*timeout < 0))
+			return -EINVAL;
+
+		if (*timeout == 0)
+			return -ETIME;
+
+		/* Record current time in case interrupted, or wedged */
+		timeout_remain = nsecs_to_jiffies_timeout(*timeout);
+		*timeout += ktime_get_raw_ns();
+	}
+
+	trace_i915_gem_request_wait_begin(req);
+
+	/* This client is about to stall waiting for the GPU. In many cases
+	 * this is undesirable and limits the throughput of the system, as
+	 * many clients cannot continue processing user input/output whilst
+	 * blocked. RPS autotuning may take tens of milliseconds to respond
+	 * to the GPU load and thus incurs additional latency for the client.
+	 * We can circumvent that by promoting the GPU frequency to maximum
+	 * before we wait. This makes the GPU throttle up much more quickly
+	 * (good for benchmarks and user experience, e.g. window animations),
+	 * but at a cost of spending more power processing the workload
+	 * (bad for battery). Not all clients even want their results
+	 * immediately and for them we should just let the GPU select its own
+	 * frequency to maximise efficiency. To prevent a single client from
+	 * forcing the clocks too high for the whole system, we only allow
+	 * each client to waitboost once in a busy period.
+	 */
+	if (IS_RPS_CLIENT(rps) && INTEL_GEN(req->i915) >= 6)
+		gen6_rps_boost(req->i915, rps, req->emitted_jiffies);
+
+	/* Optimistic short spin before touching IRQs */
+	if (i915_spin_request(req, state, 5))
+		goto complete;
+
+	set_current_state(state);
+	add_wait_queue(&req->i915->gpu_error.wait_queue, &reset);
+
+	intel_wait_init(&wait, req->fence.seqno);
+	if (intel_engine_add_wait(req->engine, &wait))
+		/* In order to check that we haven't missed the interrupt
+		 * as we enabled it, we need to kick ourselves to do a
+		 * coherent check on the seqno before we sleep.
+		 */
+		goto wakeup;
+
+	for (;;) {
+		if (signal_pending_state(state, current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+
+		timeout_remain = io_schedule_timeout(timeout_remain);
+		if (timeout_remain == 0) {
+			ret = -ETIME;
+			break;
+		}
+
+		if (intel_wait_complete(&wait))
+			break;
+
+		set_current_state(state);
+
+wakeup:
+		/* Carefully check if the request is complete, giving time
+		 * for the seqno to be visible following the interrupt.
+		 * We also have to check in case we are kicked by the GPU
+		 * reset in order to drop the struct_mutex.
+		 */
+		if (__i915_request_irq_complete(req))
+			break;
+
+		/* Only spin if we know the GPU is processing this request */
+		if (i915_spin_request(req, state, 2))
+			break;
+	}
+	remove_wait_queue(&req->i915->gpu_error.wait_queue, &reset);
+
+	intel_engine_remove_wait(req->engine, &wait);
+	__set_current_state(TASK_RUNNING);
+complete:
+	trace_i915_gem_request_wait_end(req);
+
+	if (timeout) {
+		*timeout -= ktime_get_raw_ns();
+		if (*timeout < 0)
+			*timeout = 0;
+
+		/*
+		 * Apparently ktime isn't accurate enough and occasionally has a
+		 * bit of mismatch in the jiffies<->nsecs<->ktime loop. So patch
+		 * things up to make the test happy. We allow up to 1 jiffy.
+		 *
+		 * This is a regrssion from the timespec->ktime conversion.
+		 */
+		if (ret == -ETIME && *timeout < jiffies_to_usecs(1)*1000)
+			*timeout = 0;
+	}
+
+	if (IS_RPS_USER(rps) &&
+	    req->fence.seqno == req->engine->last_submitted_seqno) {
+		/* The GPU is now idle and this client has stalled.
+		 * Since no other client has submitted a request in the
+		 * meantime, assume that this client is the only one
+		 * supplying work to the GPU but is unable to keep that
+		 * work supplied because it is waiting. Since the GPU is
+		 * then never kept fully busy, RPS autoclocking will
+		 * keep the clocks relatively low, causing further delays.
+		 * Compensate by giving the synchronous client credit for
+		 * a waitboost next time.
+		 */
+		spin_lock(&req->i915->rps.client_lock);
+		list_del_init(&rps->link);
+		spin_unlock(&req->i915->rps.client_lock);
+	}
+
+	return ret;
+}
+
+static void engine_retire_requests(struct intel_engine_cs *engine)
+{
+	struct drm_i915_gem_request *request, *next;
+
+	list_for_each_entry_safe(request, next, &engine->request_list, link) {
+		if (!i915_gem_request_completed(request))
+			break;
+
+		i915_gem_request_retire(request);
+	}
+}
+
+void i915_gem_retire_requests(struct drm_i915_private *dev_priv)
+{
+	struct intel_engine_cs *engine;
+
+	lockdep_assert_held(&dev_priv->drm.struct_mutex);
+
+	if (dev_priv->gt.active_engines == 0)
+		return;
+
+	GEM_BUG_ON(!dev_priv->gt.awake);
+
+	for_each_engine(engine, dev_priv) {
+		engine_retire_requests(engine);
+		if (!intel_engine_is_active(engine))
+			dev_priv->gt.active_engines &= ~intel_engine_flag(engine);
+	}
+
+	if (dev_priv->gt.active_engines == 0)
+		queue_delayed_work(dev_priv->wq,
+				   &dev_priv->gt.idle_work,
+				   msecs_to_jiffies(100));
+}

drivers/gpu/drm/i915/i915_gem_request.h

@@ -0,0 +1,676 @@
+/*
+ * Copyright © 2008-2015 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#ifndef I915_GEM_REQUEST_H
+#define I915_GEM_REQUEST_H
+
+#include <linux/fence.h>
+
+#include "i915_gem.h"
+
+struct intel_wait {
+	struct rb_node node;
+	struct task_struct *tsk;
+	u32 seqno;
+};
+
+struct intel_signal_node {
+	struct rb_node node;
+	struct intel_wait wait;
+};
+
+/**
+ * Request queue structure.
+ *
+ * The request queue allows us to note sequence numbers that have been emitted
+ * and may be associated with active buffers to be retired.
+ *
+ * By keeping this list, we can avoid having to do questionable sequence
+ * number comparisons on buffer last_read|write_seqno. It also allows an
+ * emission time to be associated with the request for tracking how far ahead
+ * of the GPU the submission is.
+ *
+ * The requests are reference counted.
+ */
+struct drm_i915_gem_request {
+	struct fence fence;
+	spinlock_t lock;
+
+	/** On Which ring this request was generated */
+	struct drm_i915_private *i915;
+
+	/**
+	 * Context and ring buffer related to this request
+	 * Contexts are refcounted, so when this request is associated with a
+	 * context, we must increment the context's refcount, to guarantee that
+	 * it persists while any request is linked to it. Requests themselves
+	 * are also refcounted, so the request will only be freed when the last
+	 * reference to it is dismissed, and the code in
+	 * i915_gem_request_free() will then decrement the refcount on the
+	 * context.
+	 */
+	struct i915_gem_context *ctx;
+	struct intel_engine_cs *engine;
+	struct intel_ring *ring;
+	struct intel_signal_node signaling;
+
+	/** GEM sequence number associated with the previous request,
+	 * when the HWS breadcrumb is equal to this the GPU is processing
+	 * this request.
+	 */
+	u32 previous_seqno;
+
+	/** Position in the ringbuffer of the start of the request */
+	u32 head;
+
+	/**
+	 * Position in the ringbuffer of the start of the postfix.
+	 * This is required to calculate the maximum available ringbuffer
+	 * space without overwriting the postfix.
+	 */
+	u32 postfix;
+
+	/** Position in the ringbuffer of the end of the whole request */
+	u32 tail;
+
+	/** Preallocate space in the ringbuffer for the emitting the request */
+	u32 reserved_space;
+
+	/**
+	 * Context related to the previous request.
+	 * As the contexts are accessed by the hardware until the switch is
+	 * completed to a new context, the hardware may still be writing
+	 * to the context object after the breadcrumb is visible. We must
+	 * not unpin/unbind/prune that object whilst still active and so
+	 * we keep the previous context pinned until the following (this)
+	 * request is retired.
+	 */
+	struct i915_gem_context *previous_context;
+
+	/** Batch buffer related to this request if any (used for
+	 * error state dump only).
+	 */
+	struct drm_i915_gem_object *batch_obj;
+	struct list_head active_list;
+
+	/** Time at which this request was emitted, in jiffies. */
+	unsigned long emitted_jiffies;
+
+	/** engine->request_list entry for this request */
+	struct list_head link;
+
+	/** ring->request_list entry for this request */
+	struct list_head ring_link;
+
+	struct drm_i915_file_private *file_priv;
+	/** file_priv list entry for this request */
+	struct list_head client_list;
+
+	/** process identifier submitting this request */
+	struct pid *pid;
+
+	/**
+	 * The ELSP only accepts two elements at a time, so we queue
+	 * context/tail pairs on a given queue (ring->execlist_queue) until the
+	 * hardware is available. The queue serves a double purpose: we also use
+	 * it to keep track of the up to 2 contexts currently in the hardware
+	 * (usually one in execution and the other queued up by the GPU): We
+	 * only remove elements from the head of the queue when the hardware
+	 * informs us that an element has been completed.
+	 *
+	 * All accesses to the queue are mediated by a spinlock
+	 * (ring->execlist_lock).
+	 */
+
+	/** Execlist link in the submission queue.*/
+	struct list_head execlist_link;
+
+	/** Execlists no. of times this request has been sent to the ELSP */
+	int elsp_submitted;
+
+	/** Execlists context hardware id. */
+	unsigned int ctx_hw_id;
+};
+
+extern const struct fence_ops i915_fence_ops;
+
+static inline bool fence_is_i915(struct fence *fence)
+{
+	return fence->ops == &i915_fence_ops;
+}
+
+struct drm_i915_gem_request * __must_check
+i915_gem_request_alloc(struct intel_engine_cs *engine,
+		       struct i915_gem_context *ctx);
+int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
+				   struct drm_file *file);
+void i915_gem_request_retire_upto(struct drm_i915_gem_request *req);
+
+static inline u32
+i915_gem_request_get_seqno(struct drm_i915_gem_request *req)
+{
+	return req ? req->fence.seqno : 0;
+}
+
+static inline struct intel_engine_cs *
+i915_gem_request_get_engine(struct drm_i915_gem_request *req)
+{
+	return req ? req->engine : NULL;
+}
+
+static inline struct drm_i915_gem_request *
+to_request(struct fence *fence)
+{
+	/* We assume that NULL fence/request are interoperable */
+	BUILD_BUG_ON(offsetof(struct drm_i915_gem_request, fence) != 0);
+	GEM_BUG_ON(fence && !fence_is_i915(fence));
+	return container_of(fence, struct drm_i915_gem_request, fence);
+}
+
+static inline struct drm_i915_gem_request *
+i915_gem_request_get(struct drm_i915_gem_request *req)
+{
+	return to_request(fence_get(&req->fence));
+}
+
+static inline struct drm_i915_gem_request *
+i915_gem_request_get_rcu(struct drm_i915_gem_request *req)
+{
+	return to_request(fence_get_rcu(&req->fence));
+}
+
+static inline void
+i915_gem_request_put(struct drm_i915_gem_request *req)
+{
+	fence_put(&req->fence);
+}
+
+static inline void i915_gem_request_assign(struct drm_i915_gem_request **pdst,
+					   struct drm_i915_gem_request *src)
+{
+	if (src)
+		i915_gem_request_get(src);
+
+	if (*pdst)
+		i915_gem_request_put(*pdst);
+
+	*pdst = src;
+}
+
+void __i915_add_request(struct drm_i915_gem_request *req,
+			struct drm_i915_gem_object *batch_obj,
+			bool flush_caches);
+#define i915_add_request(req) \
+	__i915_add_request(req, NULL, true)
+#define i915_add_request_no_flush(req) \
+	__i915_add_request(req, NULL, false)
+
+struct intel_rps_client;
+#define NO_WAITBOOST ERR_PTR(-1)
+#define IS_RPS_CLIENT(p) (!IS_ERR(p))
+#define IS_RPS_USER(p) (!IS_ERR_OR_NULL(p))
+
+int i915_wait_request(struct drm_i915_gem_request *req,
+		      bool interruptible,
+		      s64 *timeout,
+		      struct intel_rps_client *rps)
+	__attribute__((nonnull(1)));
+
+static inline u32 intel_engine_get_seqno(struct intel_engine_cs *engine);
+
+/**
+ * Returns true if seq1 is later than seq2.
+ */
+static inline bool i915_seqno_passed(u32 seq1, u32 seq2)
+{
+	return (s32)(seq1 - seq2) >= 0;
+}
+
+static inline bool
+i915_gem_request_started(const struct drm_i915_gem_request *req)
+{
+	return i915_seqno_passed(intel_engine_get_seqno(req->engine),
+				 req->previous_seqno);
+}
+
+static inline bool
+i915_gem_request_completed(const struct drm_i915_gem_request *req)
+{
+	return i915_seqno_passed(intel_engine_get_seqno(req->engine),
+				 req->fence.seqno);
+}
+
+bool __i915_spin_request(const struct drm_i915_gem_request *request,
+			 int state, unsigned long timeout_us);
+static inline bool i915_spin_request(const struct drm_i915_gem_request *request,
+				     int state, unsigned long timeout_us)
+{
+	return (i915_gem_request_started(request) &&
+		__i915_spin_request(request, state, timeout_us));
+}
+
+/* We treat requests as fences. This is not be to confused with our
+ * "fence registers" but pipeline synchronisation objects ala GL_ARB_sync.
+ * We use the fences to synchronize access from the CPU with activity on the
+ * GPU, for example, we should not rewrite an object's PTE whilst the GPU
+ * is reading them. We also track fences at a higher level to provide
+ * implicit synchronisation around GEM objects, e.g. set-domain will wait
+ * for outstanding GPU rendering before marking the object ready for CPU
+ * access, or a pageflip will wait until the GPU is complete before showing
+ * the frame on the scanout.
+ *
+ * In order to use a fence, the object must track the fence it needs to
+ * serialise with. For example, GEM objects want to track both read and
+ * write access so that we can perform concurrent read operations between
+ * the CPU and GPU engines, as well as waiting for all rendering to
+ * complete, or waiting for the last GPU user of a "fence register". The
+ * object then embeds a #i915_gem_active to track the most recent (in
+ * retirement order) request relevant for the desired mode of access.
+ * The #i915_gem_active is updated with i915_gem_active_set() to track the
+ * most recent fence request, typically this is done as part of
+ * i915_vma_move_to_active().
+ *
+ * When the #i915_gem_active completes (is retired), it will
+ * signal its completion to the owner through a callback as well as mark
+ * itself as idle (i915_gem_active.request == NULL). The owner
+ * can then perform any action, such as delayed freeing of an active
+ * resource including itself.
+ */
+struct i915_gem_active;
+
+typedef void (*i915_gem_retire_fn)(struct i915_gem_active *,
+				   struct drm_i915_gem_request *);
+
+struct i915_gem_active {
+	struct drm_i915_gem_request __rcu *request;
+	struct list_head link;
+	i915_gem_retire_fn retire;
+};
+
+void i915_gem_retire_noop(struct i915_gem_active *,
+			  struct drm_i915_gem_request *request);
+
+/**
+ * init_request_active - prepares the activity tracker for use
+ * @active - the active tracker
+ * @func - a callback when then the tracker is retired (becomes idle),
+ *         can be NULL
+ *
+ * init_request_active() prepares the embedded @active struct for use as
+ * an activity tracker, that is for tracking the last known active request
+ * associated with it. When the last request becomes idle, when it is retired
+ * after completion, the optional callback @func is invoked.
+ */
+static inline void
+init_request_active(struct i915_gem_active *active,
+		    i915_gem_retire_fn retire)
+{
+	INIT_LIST_HEAD(&active->link);
+	active->retire = retire ?: i915_gem_retire_noop;
+}
+
+/**
+ * i915_gem_active_set - updates the tracker to watch the current request
+ * @active - the active tracker
+ * @request - the request to watch
+ *
+ * i915_gem_active_set() watches the given @request for completion. Whilst
+ * that @request is busy, the @active reports busy. When that @request is
+ * retired, the @active tracker is updated to report idle.
+ */
+static inline void
+i915_gem_active_set(struct i915_gem_active *active,
+		    struct drm_i915_gem_request *request)
+{
+	list_move(&active->link, &request->active_list);
+	rcu_assign_pointer(active->request, request);
+}
+
+static inline struct drm_i915_gem_request *
+__i915_gem_active_peek(const struct i915_gem_active *active)
+{
+	/* Inside the error capture (running with the driver in an unknown
+	 * state), we want to bend the rules slightly (a lot).
+	 *
+	 * Work is in progress to make it safer, in the meantime this keeps
+	 * the known issue from spamming the logs.
+	 */
+	return rcu_dereference_protected(active->request, 1);
+}
+
+/**
+ * i915_gem_active_peek - report the active request being monitored
+ * @active - the active tracker
+ *
+ * i915_gem_active_peek() returns the current request being tracked if
+ * still active, or NULL. It does not obtain a reference on the request
+ * for the caller, so the caller must hold struct_mutex.
+ */
+static inline struct drm_i915_gem_request *
+i915_gem_active_peek(const struct i915_gem_active *active, struct mutex *mutex)
+{
+	struct drm_i915_gem_request *request;
+
+	request = rcu_dereference_protected(active->request,
+					    lockdep_is_held(mutex));
+	if (!request || i915_gem_request_completed(request))
+		return NULL;
+
+	return request;
+}
+
+/**
+ * i915_gem_active_peek_rcu - report the active request being monitored
+ * @active - the active tracker
+ *
+ * i915_gem_active_peek_rcu() returns the current request being tracked if
+ * still active, or NULL. It does not obtain a reference on the request
+ * for the caller, and inspection of the request is only valid under
+ * the RCU lock.
+ */
+static inline struct drm_i915_gem_request *
+i915_gem_active_peek_rcu(const struct i915_gem_active *active)
+{
+	struct drm_i915_gem_request *request;
+
+	request = rcu_dereference(active->request);
+	if (!request || i915_gem_request_completed(request))
+		return NULL;
+
+	return request;
+}
+
+/**
+ * i915_gem_active_get - return a reference to the active request
+ * @active - the active tracker
+ *
+ * i915_gem_active_get() returns a reference to the active request, or NULL
+ * if the active tracker is idle. The caller must hold struct_mutex.
+ */
+static inline struct drm_i915_gem_request *
+i915_gem_active_get(const struct i915_gem_active *active, struct mutex *mutex)
+{
+	return i915_gem_request_get(i915_gem_active_peek(active, mutex));
+}
+
+/**
+ * __i915_gem_active_get_rcu - return a reference to the active request
+ * @active - the active tracker
+ *
+ * __i915_gem_active_get() returns a reference to the active request, or NULL
+ * if the active tracker is idle. The caller must hold the RCU read lock, but
+ * the returned pointer is safe to use outside of RCU.
+ */
+static inline struct drm_i915_gem_request *
+__i915_gem_active_get_rcu(const struct i915_gem_active *active)
+{
+	/* Performing a lockless retrieval of the active request is super
+	 * tricky. SLAB_DESTROY_BY_RCU merely guarantees that the backing
+	 * slab of request objects will not be freed whilst we hold the
+	 * RCU read lock. It does not guarantee that the request itself
+	 * will not be freed and then *reused*. Viz,
+	 *
+	 * Thread A			Thread B
+	 *
+	 * req = active.request
+	 *				retire(req) -> free(req);
+	 *				(req is now first on the slab freelist)
+	 *				active.request = NULL
+	 *
+	 *				req = new submission on a new object
+	 * ref(req)
+	 *
+	 * To prevent the request from being reused whilst the caller
+	 * uses it, we take a reference like normal. Whilst acquiring
+	 * the reference we check that it is not in a destroyed state
+	 * (refcnt == 0). That prevents the request being reallocated
+	 * whilst the caller holds on to it. To check that the request
+	 * was not reallocated as we acquired the reference we have to
+	 * check that our request remains the active request across
+	 * the lookup, in the same manner as a seqlock. The visibility
+	 * of the pointer versus the reference counting is controlled
+	 * by using RCU barriers (rcu_dereference and rcu_assign_pointer).
+	 *
+	 * In the middle of all that, we inspect whether the request is
+	 * complete. Retiring is lazy so the request may be completed long
+	 * before the active tracker is updated. Querying whether the
+	 * request is complete is far cheaper (as it involves no locked
+	 * instructions setting cachelines to exclusive) than acquiring
+	 * the reference, so we do it first. The RCU read lock ensures the
+	 * pointer dereference is valid, but does not ensure that the
+	 * seqno nor HWS is the right one! However, if the request was
+	 * reallocated, that means the active tracker's request was complete.
+	 * If the new request is also complete, then both are and we can
+	 * just report the active tracker is idle. If the new request is
+	 * incomplete, then we acquire a reference on it and check that
+	 * it remained the active request.
+	 */
+	do {
+		struct drm_i915_gem_request *request;
+
+		request = rcu_dereference(active->request);
+		if (!request || i915_gem_request_completed(request))
+			return NULL;
+
+		request = i915_gem_request_get_rcu(request);
+
+		/* What stops the following rcu_access_pointer() from occurring
+		 * before the above i915_gem_request_get_rcu()? If we were
+		 * to read the value before pausing to get the reference to
+		 * the request, we may not notice a change in the active
+		 * tracker.
+		 *
+		 * The rcu_access_pointer() is a mere compiler barrier, which
+		 * means both the CPU and compiler are free to perform the
+		 * memory read without constraint. The compiler only has to
+		 * ensure that any operations after the rcu_access_pointer()
+		 * occur afterwards in program order. This means the read may
+		 * be performed earlier by an out-of-order CPU, or adventurous
+		 * compiler.
+		 *
+		 * The atomic operation at the heart of
+		 * i915_gem_request_get_rcu(), see fence_get_rcu(), is
+		 * atomic_inc_not_zero() which is only a full memory barrier
+		 * when successful. That is, if i915_gem_request_get_rcu()
+		 * returns the request (and so with the reference counted
+		 * incremented) then the following read for rcu_access_pointer()
+		 * must occur after the atomic operation and so confirm
+		 * that this request is the one currently being tracked.
+		 */
+		if (!request || request == rcu_access_pointer(active->request))
+			return rcu_pointer_handoff(request);
+
+		i915_gem_request_put(request);
+	} while (1);
+}
+
+/**
+ * i915_gem_active_get_unlocked - return a reference to the active request
+ * @active - the active tracker
+ *
+ * i915_gem_active_get_unlocked() returns a reference to the active request,
+ * or NULL if the active tracker is idle. The reference is obtained under RCU,
+ * so no locking is required by the caller.
+ *
+ * The reference should be freed with i915_gem_request_put().
+ */
+static inline struct drm_i915_gem_request *
+i915_gem_active_get_unlocked(const struct i915_gem_active *active)
+{
+	struct drm_i915_gem_request *request;
+
+	rcu_read_lock();
+	request = __i915_gem_active_get_rcu(active);
+	rcu_read_unlock();
+
+	return request;
+}
+
+/**
+ * i915_gem_active_isset - report whether the active tracker is assigned
+ * @active - the active tracker
+ *
+ * i915_gem_active_isset() returns true if the active tracker is currently
+ * assigned to a request. Due to the lazy retiring, that request may be idle
+ * and this may report stale information.
+ */
+static inline bool
+i915_gem_active_isset(const struct i915_gem_active *active)
+{
+	return rcu_access_pointer(active->request);
+}
+
+/**
+ * i915_gem_active_is_idle - report whether the active tracker is idle
+ * @active - the active tracker
+ *
+ * i915_gem_active_is_idle() returns true if the active tracker is currently
+ * unassigned or if the request is complete (but not yet retired). Requires
+ * the caller to hold struct_mutex (but that can be relaxed if desired).
+ */
+static inline bool
+i915_gem_active_is_idle(const struct i915_gem_active *active,
+			struct mutex *mutex)
+{
+	return !i915_gem_active_peek(active, mutex);
+}
+
+/**
+ * i915_gem_active_wait - waits until the request is completed
+ * @active - the active request on which to wait
+ *
+ * i915_gem_active_wait() waits until the request is completed before
+ * returning. Note that it does not guarantee that the request is
+ * retired first, see i915_gem_active_retire().
+ *
+ * i915_gem_active_wait() returns immediately if the active
+ * request is already complete.
+ */
+static inline int __must_check
+i915_gem_active_wait(const struct i915_gem_active *active, struct mutex *mutex)
+{
+	struct drm_i915_gem_request *request;
+
+	request = i915_gem_active_peek(active, mutex);
+	if (!request)
+		return 0;
+
+	return i915_wait_request(request, true, NULL, NULL);
+}
+
+/**
+ * i915_gem_active_wait_unlocked - waits until the request is completed
+ * @active - the active request on which to wait
+ * @interruptible - whether the wait can be woken by a userspace signal
+ * @timeout - how long to wait at most
+ * @rps - userspace client to charge for a waitboost
+ *
+ * i915_gem_active_wait_unlocked() waits until the request is completed before
+ * returning, without requiring any locks to be held. Note that it does not
+ * retire any requests before returning.
+ *
+ * This function relies on RCU in order to acquire the reference to the active
+ * request without holding any locks. See __i915_gem_active_get_rcu() for the
+ * glory details on how that is managed. Once the reference is acquired, we
+ * can then wait upon the request, and afterwards release our reference,
+ * free of any locking.
+ *
+ * This function wraps i915_wait_request(), see it for the full details on
+ * the arguments.
+ *
+ * Returns 0 if successful, or a negative error code.
+ */
+static inline int
+i915_gem_active_wait_unlocked(const struct i915_gem_active *active,
+			      bool interruptible,
+			      s64 *timeout,
+			      struct intel_rps_client *rps)
+{
+	struct drm_i915_gem_request *request;
+	int ret = 0;
+
+	request = i915_gem_active_get_unlocked(active);
+	if (request) {
+		ret = i915_wait_request(request, interruptible, timeout, rps);
+		i915_gem_request_put(request);
+	}
+
+	return ret;
+}
+
+/**
+ * i915_gem_active_retire - waits until the request is retired
+ * @active - the active request on which to wait
+ *
+ * i915_gem_active_retire() waits until the request is completed,
+ * and then ensures that at least the retirement handler for this
+ * @active tracker is called before returning. If the @active
+ * tracker is idle, the function returns immediately.
+ */
+static inline int __must_check
+i915_gem_active_retire(struct i915_gem_active *active,
+		       struct mutex *mutex)
+{
+	struct drm_i915_gem_request *request;
+	int ret;
+
+	request = rcu_dereference_protected(active->request,
+					    lockdep_is_held(mutex));
+	if (!request)
+		return 0;
+
+	ret = i915_wait_request(request, true, NULL, NULL);
+	if (ret)
+		return ret;
+
+	list_del_init(&active->link);
+	RCU_INIT_POINTER(active->request, NULL);
+
+	active->retire(active, request);
+
+	return 0;
+}
+
+/* Convenience functions for peeking at state inside active's request whilst
+ * guarded by the struct_mutex.
+ */
+
+static inline uint32_t
+i915_gem_active_get_seqno(const struct i915_gem_active *active,
+			  struct mutex *mutex)
+{
+	return i915_gem_request_get_seqno(i915_gem_active_peek(active, mutex));
+}
+
+static inline struct intel_engine_cs *
+i915_gem_active_get_engine(const struct i915_gem_active *active,
+			   struct mutex *mutex)
+{
+	return i915_gem_request_get_engine(i915_gem_active_peek(active, mutex));
+}
+
+#define for_each_active(mask, idx) \
+	for (; mask ? idx = ffs(mask) - 1, 1 : 0; mask &= ~BIT(idx))
+
+#endif /* I915_GEM_REQUEST_H */

drivers/gpu/drm/i915/i915_gem_shrinker.c

@@ -48,19 +48,15 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task)
 #endif
 }
 
-static int num_vma_bound(struct drm_i915_gem_object *obj)
+static bool any_vma_pinned(struct drm_i915_gem_object *obj)
 {
 	struct i915_vma *vma;
-	int count = 0;
 
-	list_for_each_entry(vma, &obj->vma_list, obj_link) {
-		if (drm_mm_node_allocated(&vma->node))
-			count++;
-		if (vma->pin_count)
-			count++;
-	}
+	list_for_each_entry(vma, &obj->vma_list, obj_link)
+		if (i915_vma_is_pinned(vma))
+			return true;
 
-	return count;
+	return false;
 }
 
 static bool swap_available(void)
@@ -82,7 +78,10 @@ static bool can_release_pages(struct drm_i915_gem_object *obj)
 	 * to the GPU, simply unbinding from the GPU is not going to succeed
 	 * in releasing our pin count on the pages themselves.
 	 */
-	if (obj->pages_pin_count != num_vma_bound(obj))
+	if (obj->pages_pin_count > obj->bind_count)
+		return false;
+
+	if (any_vma_pinned(obj))
 		return false;
 
 	/* We can only return physical pages to the system if we can either
@@ -163,17 +162,16 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 	 */
 	for (phase = phases; phase->list; phase++) {
 		struct list_head still_in_list;
+		struct drm_i915_gem_object *obj;
 
 		if ((flags & phase->bit) == 0)
 			continue;
 
 		INIT_LIST_HEAD(&still_in_list);
-		while (count < target && !list_empty(phase->list)) {
-			struct drm_i915_gem_object *obj;
-			struct i915_vma *vma, *v;
-
-			obj = list_first_entry(phase->list,
-					       typeof(*obj), global_list);
+		while (count < target &&
+		       (obj = list_first_entry_or_null(phase->list,
+						       typeof(*obj),
+						       global_list))) {
 			list_move_tail(&obj->global_list, &still_in_list);
 
 			if (flags & I915_SHRINK_PURGEABLE &&
@@ -184,24 +182,21 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 			    !is_vmalloc_addr(obj->mapping))
 				continue;
 
-			if ((flags & I915_SHRINK_ACTIVE) == 0 && obj->active)
+			if ((flags & I915_SHRINK_ACTIVE) == 0 &&
+			    i915_gem_object_is_active(obj))
 				continue;
 
 			if (!can_release_pages(obj))
 				continue;
 
-			drm_gem_object_reference(&obj->base);
+			i915_gem_object_get(obj);
 
 			/* For the unbound phase, this should be a no-op! */
-			list_for_each_entry_safe(vma, v,
-						 &obj->vma_list, obj_link)
-				if (i915_vma_unbind(vma))
-					break;
-
+			i915_gem_object_unbind(obj);
 			if (i915_gem_object_put_pages(obj) == 0)
 				count += obj->base.size >> PAGE_SHIFT;
 
-			drm_gem_object_unreference(&obj->base);
+			i915_gem_object_put(obj);
 		}
 		list_splice(&still_in_list, phase->list);
 	}
@@ -210,6 +205,8 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 		intel_runtime_pm_put(dev_priv);
 
 	i915_gem_retire_requests(dev_priv);
+	/* expedite the RCU grace period to free some request slabs */
+	synchronize_rcu_expedited();
 
 	return count;
 }
@@ -230,10 +227,15 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
  */
 unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 {
-	return i915_gem_shrink(dev_priv, -1UL,
-			       I915_SHRINK_BOUND |
-			       I915_SHRINK_UNBOUND |
-			       I915_SHRINK_ACTIVE);
+	unsigned long freed;
+
+	freed = i915_gem_shrink(dev_priv, -1UL,
+				I915_SHRINK_BOUND |
+				I915_SHRINK_UNBOUND |
+				I915_SHRINK_ACTIVE);
+	rcu_barrier(); /* wait until our RCU delayed slab frees are completed */
+
+	return freed;
 }
 
 static bool i915_gem_shrinker_lock(struct drm_device *dev, bool *unlock)
@@ -242,9 +244,6 @@ static bool i915_gem_shrinker_lock(struct drm_device *dev, bool *unlock)
 		if (!mutex_is_locked_by(&dev->struct_mutex, current))
 			return false;
 
-		if (to_i915(dev)->mm.shrinker_no_lock_stealing)
-			return false;
-
 		*unlock = false;
 	} else
 		*unlock = true;
@@ -273,7 +272,7 @@ i915_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 			count += obj->base.size >> PAGE_SHIFT;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		if (!obj->active && can_release_pages(obj))
+		if (!i915_gem_object_is_active(obj) && can_release_pages(obj))
 			count += obj->base.size >> PAGE_SHIFT;
 	}
 
@@ -321,17 +320,22 @@ i915_gem_shrinker_lock_uninterruptible(struct drm_i915_private *dev_priv,
 				       struct shrinker_lock_uninterruptible *slu,
 				       int timeout_ms)
 {
-	unsigned long timeout = msecs_to_jiffies(timeout_ms) + 1;
+	unsigned long timeout = jiffies + msecs_to_jiffies_timeout(timeout_ms);
+
+	do {
+		if (i915_gem_wait_for_idle(dev_priv, false) == 0 &&
+		    i915_gem_shrinker_lock(&dev_priv->drm, &slu->unlock))
+			break;
 
-	while (!i915_gem_shrinker_lock(&dev_priv->drm, &slu->unlock)) {
 		schedule_timeout_killable(1);
 		if (fatal_signal_pending(current))
 			return false;
-		if (--timeout == 0) {
+
+		if (time_after(jiffies, timeout)) {
 			pr_err("Unable to lock GPU to purge memory.\n");
 			return false;
 		}
-	}
+	} while (1);
 
 	slu->was_interruptible = dev_priv->mm.interruptible;
 	dev_priv->mm.interruptible = false;
@@ -410,7 +414,7 @@ i915_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr
 		return NOTIFY_DONE;
 
 	/* Force everything onto the inactive lists */
-	ret = i915_gem_wait_for_idle(dev_priv);
+	ret = i915_gem_wait_for_idle(dev_priv, false);
 	if (ret)
 		goto out;
 

drivers/gpu/drm/i915/i915_gem_stolen.c

@@ -698,24 +698,24 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 	 */
 	vma->node.start = gtt_offset;
 	vma->node.size = size;
-	if (drm_mm_initialized(&ggtt->base.mm)) {
-		ret = drm_mm_reserve_node(&ggtt->base.mm, &vma->node);
-		if (ret) {
-			DRM_DEBUG_KMS("failed to allocate stolen GTT space\n");
-			goto err;
-		}
 
-		vma->bound |= GLOBAL_BIND;
-		__i915_vma_set_map_and_fenceable(vma);
-		list_add_tail(&vma->vm_link, &ggtt->base.inactive_list);
+	ret = drm_mm_reserve_node(&ggtt->base.mm, &vma->node);
+	if (ret) {
+		DRM_DEBUG_KMS("failed to allocate stolen GTT space\n");
+		goto err;
 	}
 
+	vma->flags |= I915_VMA_GLOBAL_BIND;
+	__i915_vma_set_map_and_fenceable(vma);
+	list_move_tail(&vma->vm_link, &ggtt->base.inactive_list);
+	obj->bind_count++;
+
 	list_add_tail(&obj->global_list, &dev_priv->mm.bound_list);
 	i915_gem_object_pin_pages(obj);
 
 	return obj;
 
 err:
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	return NULL;
 }

drivers/gpu/drm/i915/i915_gem_tiling.c

@@ -68,6 +68,9 @@ i915_tiling_ok(struct drm_device *dev, int stride, int size, int tiling_mode)
 	if (tiling_mode == I915_TILING_NONE)
 		return true;
 
+	if (tiling_mode > I915_TILING_LAST)
+		return false;
+
 	if (IS_GEN2(dev) ||
 	    (tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)))
 		tile_width = 128;
@@ -117,15 +120,16 @@ i915_tiling_ok(struct drm_device *dev, int stride, int size, int tiling_mode)
 static bool
 i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode)
 {
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 	u32 size;
 
 	if (tiling_mode == I915_TILING_NONE)
 		return true;
 
-	if (INTEL_INFO(obj->base.dev)->gen >= 4)
+	if (INTEL_GEN(dev_priv) >= 4)
 		return true;
 
-	if (IS_GEN3(obj->base.dev)) {
+	if (IS_GEN3(dev_priv)) {
 		if (i915_gem_obj_ggtt_offset(obj) & ~I915_FENCE_START_MASK)
 			return false;
 	} else {
@@ -133,7 +137,7 @@ i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode)
 			return false;
 	}
 
-	size = i915_gem_get_gtt_size(obj->base.dev, obj->base.size, tiling_mode);
+	size = i915_gem_get_ggtt_size(dev_priv, obj->base.size, tiling_mode);
 	if (i915_gem_obj_ggtt_size(obj) != size)
 		return false;
 
@@ -166,13 +170,16 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 	struct drm_i915_gem_object *obj;
 	int ret = 0;
 
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL)
+	/* Make sure we don't cross-contaminate obj->tiling_and_stride */
+	BUILD_BUG_ON(I915_TILING_LAST & STRIDE_MASK);
+
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
 		return -ENOENT;
 
 	if (!i915_tiling_ok(dev,
 			    args->stride, obj->base.size, args->tiling_mode)) {
-		drm_gem_object_unreference_unlocked(&obj->base);
+		i915_gem_object_put_unlocked(obj);
 		return -EINVAL;
 	}
 
@@ -213,8 +220,8 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 		}
 	}
 
-	if (args->tiling_mode != obj->tiling_mode ||
-	    args->stride != obj->stride) {
+	if (args->tiling_mode != i915_gem_object_get_tiling(obj) ||
+	    args->stride != i915_gem_object_get_stride(obj)) {
 		/* We need to rebind the object if its current allocation
 		 * no longer meets the alignment restrictions for its new
 		 * tiling mode. Otherwise we can just leave it alone, but
@@ -237,24 +244,25 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 			    dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
 				if (args->tiling_mode == I915_TILING_NONE)
 					i915_gem_object_unpin_pages(obj);
-				if (obj->tiling_mode == I915_TILING_NONE)
+				if (!i915_gem_object_is_tiled(obj))
 					i915_gem_object_pin_pages(obj);
 			}
 
 			obj->fence_dirty =
-				obj->last_fenced_req ||
+				!i915_gem_active_is_idle(&obj->last_fence,
+							 &dev->struct_mutex) ||
 				obj->fence_reg != I915_FENCE_REG_NONE;
 
-			obj->tiling_mode = args->tiling_mode;
-			obj->stride = args->stride;
+			obj->tiling_and_stride =
+				args->stride | args->tiling_mode;
 
 			/* Force the fence to be reacquired for GTT access */
 			i915_gem_release_mmap(obj);
 		}
 	}
 	/* we have to maintain this existing ABI... */
-	args->stride = obj->stride;
-	args->tiling_mode = obj->tiling_mode;
+	args->stride = i915_gem_object_get_stride(obj);
+	args->tiling_mode = i915_gem_object_get_tiling(obj);
 
 	/* Try to preallocate memory required to save swizzling on put-pages */
 	if (i915_gem_object_needs_bit17_swizzle(obj)) {
@@ -268,7 +276,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 	}
 
 err:
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	mutex_unlock(&dev->struct_mutex);
 
 	intel_runtime_pm_put(dev_priv);
@@ -297,14 +305,12 @@ i915_gem_get_tiling(struct drm_device *dev, void *data,
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_i915_gem_object *obj;
 
-	obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
-	if (&obj->base == NULL)
+	obj = i915_gem_object_lookup(file, args->handle);
+	if (!obj)
 		return -ENOENT;
 
-	mutex_lock(&dev->struct_mutex);
-
-	args->tiling_mode = obj->tiling_mode;
-	switch (obj->tiling_mode) {
+	args->tiling_mode = READ_ONCE(obj->tiling_and_stride) & TILING_MASK;
+	switch (args->tiling_mode) {
 	case I915_TILING_X:
 		args->swizzle_mode = dev_priv->mm.bit_6_swizzle_x;
 		break;
@@ -328,8 +334,6 @@ i915_gem_get_tiling(struct drm_device *dev, void *data,
 	if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17)
 		args->swizzle_mode = I915_BIT_6_SWIZZLE_9_10;
 
-	drm_gem_object_unreference(&obj->base);
-	mutex_unlock(&dev->struct_mutex);
-
+	i915_gem_object_put_unlocked(obj);
 	return 0;
 }

drivers/gpu/drm/i915/i915_gem_userptr.c

@@ -63,33 +63,12 @@ struct i915_mmu_object {
 
 static void wait_rendering(struct drm_i915_gem_object *obj)
 {
-	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_gem_request *requests[I915_NUM_ENGINES];
-	int i, n;
-
-	if (!obj->active)
-		return;
-
-	n = 0;
-	for (i = 0; i < I915_NUM_ENGINES; i++) {
-		struct drm_i915_gem_request *req;
-
-		req = obj->last_read_req[i];
-		if (req == NULL)
-			continue;
-
-		requests[n++] = i915_gem_request_reference(req);
-	}
-
-	mutex_unlock(&dev->struct_mutex);
+	unsigned long active = __I915_BO_ACTIVE(obj);
+	int idx;
 
-	for (i = 0; i < n; i++)
-		__i915_wait_request(requests[i], false, NULL, NULL);
-
-	mutex_lock(&dev->struct_mutex);
-
-	for (i = 0; i < n; i++)
-		i915_gem_request_unreference(requests[i]);
+	for_each_active(active, idx)
+		i915_gem_active_wait_unlocked(&obj->last_read[idx],
+					      false, NULL, NULL);
 }
 
 static void cancel_userptr(struct work_struct *work)
@@ -98,28 +77,19 @@ static void cancel_userptr(struct work_struct *work)
 	struct drm_i915_gem_object *obj = mo->obj;
 	struct drm_device *dev = obj->base.dev;
 
+	wait_rendering(obj);
+
 	mutex_lock(&dev->struct_mutex);
 	/* Cancel any active worker and force us to re-evaluate gup */
 	obj->userptr.work = NULL;
 
 	if (obj->pages != NULL) {
-		struct drm_i915_private *dev_priv = to_i915(dev);
-		struct i915_vma *vma, *tmp;
-		bool was_interruptible;
-
-		wait_rendering(obj);
-
-		was_interruptible = dev_priv->mm.interruptible;
-		dev_priv->mm.interruptible = false;
-
-		list_for_each_entry_safe(vma, tmp, &obj->vma_list, obj_link)
-			WARN_ON(i915_vma_unbind(vma));
+		/* We are inside a kthread context and can't be interrupted */
+		WARN_ON(i915_gem_object_unbind(obj));
 		WARN_ON(i915_gem_object_put_pages(obj));
-
-		dev_priv->mm.interruptible = was_interruptible;
 	}
 
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	mutex_unlock(&dev->struct_mutex);
 }
 
@@ -577,7 +547,7 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 	}
 
 	obj->userptr.workers--;
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	mutex_unlock(&dev->struct_mutex);
 
 	release_pages(pvec, pinned, 0);
@@ -622,8 +592,7 @@ __i915_gem_userptr_get_pages_schedule(struct drm_i915_gem_object *obj,
 	obj->userptr.work = &work->work;
 	obj->userptr.workers++;
 
-	work->obj = obj;
-	drm_gem_object_reference(&obj->base);
+	work->obj = i915_gem_object_get(obj);
 
 	work->task = current;
 	get_task_struct(work->task);
@@ -846,7 +815,7 @@ i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file
 		ret = drm_gem_handle_create(file, &obj->base, &handle);
 
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(&obj->base);
+	i915_gem_object_put_unlocked(obj);
 	if (ret)
 		return ret;
 

drivers/gpu/drm/i915/i915_gpu_error.c

@@ -30,9 +30,9 @@
 #include <generated/utsrelease.h>
 #include "i915_drv.h"
 
-static const char *ring_str(int ring)
+static const char *engine_str(int engine)
 {
-	switch (ring) {
+	switch (engine) {
 	case RCS: return "render";
 	case VCS: return "bsd";
 	case BCS: return "blt";
@@ -207,8 +207,8 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 		err_puts(m, dirty_flag(err->dirty));
 		err_puts(m, purgeable_flag(err->purgeable));
 		err_puts(m, err->userptr ? " userptr" : "");
-		err_puts(m, err->ring != -1 ? " " : "");
-		err_puts(m, ring_str(err->ring));
+		err_puts(m, err->engine != -1 ? " " : "");
+		err_puts(m, engine_str(err->engine));
 		err_puts(m, i915_cache_level_str(m->i915, err->cache_level));
 
 		if (err->name)
@@ -221,7 +221,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 	}
 }
 
-static const char *hangcheck_action_to_str(enum intel_ring_hangcheck_action a)
+static const char *hangcheck_action_to_str(enum intel_engine_hangcheck_action a)
 {
 	switch (a) {
 	case HANGCHECK_IDLE:
@@ -239,70 +239,65 @@ static const char *hangcheck_action_to_str(enum intel_ring_hangcheck_action a)
 	return "unknown";
 }
 
-static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
-				  struct drm_device *dev,
-				  struct drm_i915_error_state *error,
-				  int ring_idx)
+static void error_print_engine(struct drm_i915_error_state_buf *m,
+			       struct drm_i915_error_engine *ee)
 {
-	struct drm_i915_error_ring *ring = &error->ring[ring_idx];
-
-	if (!ring->valid)
-		return;
-
-	err_printf(m, "%s command stream:\n", ring_str(ring_idx));
-	err_printf(m, "  START: 0x%08x\n", ring->start);
-	err_printf(m, "  HEAD:  0x%08x\n", ring->head);
-	err_printf(m, "  TAIL:  0x%08x\n", ring->tail);
-	err_printf(m, "  CTL:   0x%08x\n", ring->ctl);
-	err_printf(m, "  HWS:   0x%08x\n", ring->hws);
-	err_printf(m, "  ACTHD: 0x%08x %08x\n", (u32)(ring->acthd>>32), (u32)ring->acthd);
-	err_printf(m, "  IPEIR: 0x%08x\n", ring->ipeir);
-	err_printf(m, "  IPEHR: 0x%08x\n", ring->ipehr);
-	err_printf(m, "  INSTDONE: 0x%08x\n", ring->instdone);
-	if (INTEL_INFO(dev)->gen >= 4) {
-		err_printf(m, "  BBADDR: 0x%08x %08x\n", (u32)(ring->bbaddr>>32), (u32)ring->bbaddr);
-		err_printf(m, "  BB_STATE: 0x%08x\n", ring->bbstate);
-		err_printf(m, "  INSTPS: 0x%08x\n", ring->instps);
+	err_printf(m, "%s command stream:\n", engine_str(ee->engine_id));
+	err_printf(m, "  START: 0x%08x\n", ee->start);
+	err_printf(m, "  HEAD:  0x%08x\n", ee->head);
+	err_printf(m, "  TAIL:  0x%08x\n", ee->tail);
+	err_printf(m, "  CTL:   0x%08x\n", ee->ctl);
+	err_printf(m, "  HWS:   0x%08x\n", ee->hws);
+	err_printf(m, "  ACTHD: 0x%08x %08x\n",
+		   (u32)(ee->acthd>>32), (u32)ee->acthd);
+	err_printf(m, "  IPEIR: 0x%08x\n", ee->ipeir);
+	err_printf(m, "  IPEHR: 0x%08x\n", ee->ipehr);
+	err_printf(m, "  INSTDONE: 0x%08x\n", ee->instdone);
+	if (INTEL_GEN(m->i915) >= 4) {
+		err_printf(m, "  BBADDR: 0x%08x %08x\n",
+			   (u32)(ee->bbaddr>>32), (u32)ee->bbaddr);
+		err_printf(m, "  BB_STATE: 0x%08x\n", ee->bbstate);
+		err_printf(m, "  INSTPS: 0x%08x\n", ee->instps);
 	}
-	err_printf(m, "  INSTPM: 0x%08x\n", ring->instpm);
-	err_printf(m, "  FADDR: 0x%08x %08x\n", upper_32_bits(ring->faddr),
-		   lower_32_bits(ring->faddr));
-	if (INTEL_INFO(dev)->gen >= 6) {
-		err_printf(m, "  RC PSMI: 0x%08x\n", ring->rc_psmi);
-		err_printf(m, "  FAULT_REG: 0x%08x\n", ring->fault_reg);
+	err_printf(m, "  INSTPM: 0x%08x\n", ee->instpm);
+	err_printf(m, "  FADDR: 0x%08x %08x\n", upper_32_bits(ee->faddr),
+		   lower_32_bits(ee->faddr));
+	if (INTEL_GEN(m->i915) >= 6) {
+		err_printf(m, "  RC PSMI: 0x%08x\n", ee->rc_psmi);
+		err_printf(m, "  FAULT_REG: 0x%08x\n", ee->fault_reg);
 		err_printf(m, "  SYNC_0: 0x%08x [last synced 0x%08x]\n",
-			   ring->semaphore_mboxes[0],
-			   ring->semaphore_seqno[0]);
+			   ee->semaphore_mboxes[0],
+			   ee->semaphore_seqno[0]);
 		err_printf(m, "  SYNC_1: 0x%08x [last synced 0x%08x]\n",
-			   ring->semaphore_mboxes[1],
-			   ring->semaphore_seqno[1]);
-		if (HAS_VEBOX(dev)) {
+			   ee->semaphore_mboxes[1],
+			   ee->semaphore_seqno[1]);
+		if (HAS_VEBOX(m->i915)) {
 			err_printf(m, "  SYNC_2: 0x%08x [last synced 0x%08x]\n",
-				   ring->semaphore_mboxes[2],
-				   ring->semaphore_seqno[2]);
+				   ee->semaphore_mboxes[2],
+				   ee->semaphore_seqno[2]);
 		}
 	}
-	if (USES_PPGTT(dev)) {
-		err_printf(m, "  GFX_MODE: 0x%08x\n", ring->vm_info.gfx_mode);
+	if (USES_PPGTT(m->i915)) {
+		err_printf(m, "  GFX_MODE: 0x%08x\n", ee->vm_info.gfx_mode);
 
-		if (INTEL_INFO(dev)->gen >= 8) {
+		if (INTEL_GEN(m->i915) >= 8) {
 			int i;
 			for (i = 0; i < 4; i++)
 				err_printf(m, "  PDP%d: 0x%016llx\n",
-					   i, ring->vm_info.pdp[i]);
+					   i, ee->vm_info.pdp[i]);
 		} else {
 			err_printf(m, "  PP_DIR_BASE: 0x%08x\n",
-				   ring->vm_info.pp_dir_base);
+				   ee->vm_info.pp_dir_base);
 		}
 	}
-	err_printf(m, "  seqno: 0x%08x\n", ring->seqno);
-	err_printf(m, "  last_seqno: 0x%08x\n", ring->last_seqno);
-	err_printf(m, "  waiting: %s\n", yesno(ring->waiting));
-	err_printf(m, "  ring->head: 0x%08x\n", ring->cpu_ring_head);
-	err_printf(m, "  ring->tail: 0x%08x\n", ring->cpu_ring_tail);
+	err_printf(m, "  seqno: 0x%08x\n", ee->seqno);
+	err_printf(m, "  last_seqno: 0x%08x\n", ee->last_seqno);
+	err_printf(m, "  waiting: %s\n", yesno(ee->waiting));
+	err_printf(m, "  ring->head: 0x%08x\n", ee->cpu_ring_head);
+	err_printf(m, "  ring->tail: 0x%08x\n", ee->cpu_ring_tail);
 	err_printf(m, "  hangcheck: %s [%d]\n",
-		   hangcheck_action_to_str(ring->hangcheck_action),
-		   ring->hangcheck_score);
+		   hangcheck_action_to_str(ee->hangcheck_action),
+		   ee->hangcheck_score);
 }
 
 void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...)
@@ -348,17 +343,17 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 		   error->time.tv_usec);
 	err_printf(m, "Kernel: " UTS_RELEASE "\n");
 	max_hangcheck_score = 0;
-	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
-		if (error->ring[i].hangcheck_score > max_hangcheck_score)
-			max_hangcheck_score = error->ring[i].hangcheck_score;
+	for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
+		if (error->engine[i].hangcheck_score > max_hangcheck_score)
+			max_hangcheck_score = error->engine[i].hangcheck_score;
 	}
-	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
-		if (error->ring[i].hangcheck_score == max_hangcheck_score &&
-		    error->ring[i].pid != -1) {
+	for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
+		if (error->engine[i].hangcheck_score == max_hangcheck_score &&
+		    error->engine[i].pid != -1) {
 			err_printf(m, "Active process (on ring %s): %s [%d]\n",
-				   ring_str(i),
-				   error->ring[i].comm,
-				   error->ring[i].pid);
+				   engine_str(i),
+				   error->engine[i].comm,
+				   error->engine[i].pid);
 		}
 	}
 	err_printf(m, "Reset count: %u\n", error->reset_count);
@@ -414,8 +409,10 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 	if (IS_GEN7(dev))
 		err_printf(m, "ERR_INT: 0x%08x\n", error->err_int);
 
-	for (i = 0; i < ARRAY_SIZE(error->ring); i++)
-		i915_ring_error_state(m, dev, error, i);
+	for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
+		if (error->engine[i].engine_id != -1)
+			error_print_engine(m, &error->engine[i]);
+	}
 
 	for (i = 0; i < error->vm_count; i++) {
 		err_printf(m, "vm[%d]\n", i);
@@ -429,21 +426,23 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 				    error->pinned_bo_count[i]);
 	}
 
-	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
-		obj = error->ring[i].batchbuffer;
+	for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
+		struct drm_i915_error_engine *ee = &error->engine[i];
+
+		obj = ee->batchbuffer;
 		if (obj) {
 			err_puts(m, dev_priv->engine[i].name);
-			if (error->ring[i].pid != -1)
+			if (ee->pid != -1)
 				err_printf(m, " (submitted by %s [%d])",
-					   error->ring[i].comm,
-					   error->ring[i].pid);
+					   ee->comm,
+					   ee->pid);
 			err_printf(m, " --- gtt_offset = 0x%08x %08x\n",
 				   upper_32_bits(obj->gtt_offset),
 				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
 
-		obj = error->ring[i].wa_batchbuffer;
+		obj = ee->wa_batchbuffer;
 		if (obj) {
 			err_printf(m, "%s (w/a) --- gtt_offset = 0x%08x\n",
 				   dev_priv->engine[i].name,
@@ -451,38 +450,38 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			print_error_obj(m, obj);
 		}
 
-		if (error->ring[i].num_requests) {
+		if (ee->num_requests) {
 			err_printf(m, "%s --- %d requests\n",
 				   dev_priv->engine[i].name,
-				   error->ring[i].num_requests);
-			for (j = 0; j < error->ring[i].num_requests; j++) {
+				   ee->num_requests);
+			for (j = 0; j < ee->num_requests; j++) {
 				err_printf(m, "  seqno 0x%08x, emitted %ld, tail 0x%08x\n",
-					   error->ring[i].requests[j].seqno,
-					   error->ring[i].requests[j].jiffies,
-					   error->ring[i].requests[j].tail);
+					   ee->requests[j].seqno,
+					   ee->requests[j].jiffies,
+					   ee->requests[j].tail);
 			}
 		}
 
-		if (error->ring[i].num_waiters) {
+		if (ee->num_waiters) {
 			err_printf(m, "%s --- %d waiters\n",
 				   dev_priv->engine[i].name,
-				   error->ring[i].num_waiters);
-			for (j = 0; j < error->ring[i].num_waiters; j++) {
+				   ee->num_waiters);
+			for (j = 0; j < ee->num_waiters; j++) {
 				err_printf(m, " seqno 0x%08x for %s [%d]\n",
-					   error->ring[i].waiters[j].seqno,
-					   error->ring[i].waiters[j].comm,
-					   error->ring[i].waiters[j].pid);
+					   ee->waiters[j].seqno,
+					   ee->waiters[j].comm,
+					   ee->waiters[j].pid);
 			}
 		}
 
-		if ((obj = error->ring[i].ringbuffer)) {
+		if ((obj = ee->ringbuffer)) {
 			err_printf(m, "%s --- ringbuffer = 0x%08x\n",
 				   dev_priv->engine[i].name,
 				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
 
-		if ((obj = error->ring[i].hws_page)) {
+		if ((obj = ee->hws_page)) {
 			u64 hws_offset = obj->gtt_offset;
 			u32 *hws_page = &obj->pages[0][0];
 
@@ -504,7 +503,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			}
 		}
 
-		obj = error->ring[i].wa_ctx;
+		obj = ee->wa_ctx;
 		if (obj) {
 			u64 wa_ctx_offset = obj->gtt_offset;
 			u32 *wa_ctx_page = &obj->pages[0][0];
@@ -526,7 +525,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			}
 		}
 
-		if ((obj = error->ring[i].ctx)) {
+		if ((obj = ee->ctx)) {
 			err_printf(m, "%s --- HW Context = 0x%08x\n",
 				   dev_priv->engine[i].name,
 				   lower_32_bits(obj->gtt_offset));
@@ -611,15 +610,18 @@ static void i915_error_state_free(struct kref *error_ref)
 							  typeof(*error), ref);
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
-		i915_error_object_free(error->ring[i].batchbuffer);
-		i915_error_object_free(error->ring[i].wa_batchbuffer);
-		i915_error_object_free(error->ring[i].ringbuffer);
-		i915_error_object_free(error->ring[i].hws_page);
-		i915_error_object_free(error->ring[i].ctx);
-		i915_error_object_free(error->ring[i].wa_ctx);
-		kfree(error->ring[i].requests);
-		kfree(error->ring[i].waiters);
+	for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
+		struct drm_i915_error_engine *ee = &error->engine[i];
+
+		i915_error_object_free(ee->batchbuffer);
+		i915_error_object_free(ee->wa_batchbuffer);
+		i915_error_object_free(ee->ringbuffer);
+		i915_error_object_free(ee->hws_page);
+		i915_error_object_free(ee->ctx);
+		i915_error_object_free(ee->wa_ctx);
+
+		kfree(ee->requests);
+		kfree(ee->waiters);
 	}
 
 	i915_error_object_free(error->semaphore_obj);
@@ -667,14 +669,14 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 	if (i915_is_ggtt(vm))
 		vma = i915_gem_obj_to_ggtt(src);
 	use_ggtt = (src->cache_level == I915_CACHE_NONE &&
-		   vma && (vma->bound & GLOBAL_BIND) &&
+		   vma && (vma->flags & I915_VMA_GLOBAL_BIND) &&
 		   reloc_offset + num_pages * PAGE_SIZE <= ggtt->mappable_end);
 
 	/* Cannot access stolen address directly, try to use the aperture */
 	if (src->stolen) {
 		use_ggtt = true;
 
-		if (!(vma && vma->bound & GLOBAL_BIND))
+		if (!(vma && vma->flags & I915_VMA_GLOBAL_BIND))
 			goto unwind;
 
 		reloc_offset = i915_gem_obj_ggtt_offset(src);
@@ -740,6 +742,24 @@ unwind:
 #define i915_error_ggtt_object_create(dev_priv, src) \
 	i915_error_object_create((dev_priv), (src), &(dev_priv)->ggtt.base)
 
+/* The error capture is special as tries to run underneath the normal
+ * locking rules - so we use the raw version of the i915_gem_active lookup.
+ */
+static inline uint32_t
+__active_get_seqno(struct i915_gem_active *active)
+{
+	return i915_gem_request_get_seqno(__i915_gem_active_peek(active));
+}
+
+static inline int
+__active_get_engine_id(struct i915_gem_active *active)
+{
+	struct intel_engine_cs *engine;
+
+	engine = i915_gem_request_get_engine(__i915_gem_active_peek(active));
+	return engine ? engine->id : -1;
+}
+
 static void capture_bo(struct drm_i915_error_buffer *err,
 		       struct i915_vma *vma)
 {
@@ -748,9 +768,12 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 
 	err->size = obj->base.size;
 	err->name = obj->base.name;
+
 	for (i = 0; i < I915_NUM_ENGINES; i++)
-		err->rseqno[i] = i915_gem_request_get_seqno(obj->last_read_req[i]);
-	err->wseqno = i915_gem_request_get_seqno(obj->last_write_req);
+		err->rseqno[i] = __active_get_seqno(&obj->last_read[i]);
+	err->wseqno = __active_get_seqno(&obj->last_write);
+	err->engine = __active_get_engine_id(&obj->last_write);
+
 	err->gtt_offset = vma->node.start;
 	err->read_domains = obj->base.read_domains;
 	err->write_domain = obj->base.write_domain;
@@ -758,12 +781,10 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 	err->pinned = 0;
 	if (i915_gem_obj_is_pinned(obj))
 		err->pinned = 1;
-	err->tiling = obj->tiling_mode;
+	err->tiling = i915_gem_object_get_tiling(obj);
 	err->dirty = obj->dirty;
 	err->purgeable = obj->madv != I915_MADV_WILLNEED;
 	err->userptr = obj->userptr.mm != NULL;
-	err->ring = obj->last_write_req ?
-			i915_gem_request_get_engine(obj->last_write_req)->id : -1;
 	err->cache_level = obj->cache_level;
 }
 
@@ -797,7 +818,7 @@ static u32 capture_pinned_bo(struct drm_i915_error_buffer *err,
 			break;
 
 		list_for_each_entry(vma, &obj->vma_list, obj_link)
-			if (vma->vm == vm && vma->pin_count > 0)
+			if (vma->vm == vm && i915_vma_is_pinned(vma))
 				capture_bo(err++, vma);
 	}
 
@@ -815,7 +836,7 @@ static u32 capture_pinned_bo(struct drm_i915_error_buffer *err,
  */
 static uint32_t i915_error_generate_code(struct drm_i915_private *dev_priv,
 					 struct drm_i915_error_state *error,
-					 int *ring_id)
+					 int *engine_id)
 {
 	uint32_t error_code = 0;
 	int i;
@@ -826,11 +847,11 @@ static uint32_t i915_error_generate_code(struct drm_i915_private *dev_priv,
 	 * strictly a client bug. Use instdone to differentiate those some.
 	 */
 	for (i = 0; i < I915_NUM_ENGINES; i++) {
-		if (error->ring[i].hangcheck_action == HANGCHECK_HUNG) {
-			if (ring_id)
-				*ring_id = i;
+		if (error->engine[i].hangcheck_action == HANGCHECK_HUNG) {
+			if (engine_id)
+				*engine_id = i;
 
-			return error->ring[i].ipehr ^ error->ring[i].instdone;
+			return error->engine[i].ipehr ^ error->engine[i].instdone;
 		}
 	}
 
@@ -855,21 +876,16 @@ static void i915_gem_record_fences(struct drm_i915_private *dev_priv,
 }
 
 
-static void gen8_record_semaphore_state(struct drm_i915_private *dev_priv,
-					struct drm_i915_error_state *error,
+static void gen8_record_semaphore_state(struct drm_i915_error_state *error,
 					struct intel_engine_cs *engine,
-					struct drm_i915_error_ring *ering)
+					struct drm_i915_error_engine *ee)
 {
+	struct drm_i915_private *dev_priv = engine->i915;
 	struct intel_engine_cs *to;
 	enum intel_engine_id id;
 
-	if (!i915_semaphore_is_enabled(dev_priv))
-		return;
-
 	if (!error->semaphore_obj)
-		error->semaphore_obj =
-			i915_error_ggtt_object_create(dev_priv,
-						      dev_priv->semaphore_obj);
+		return;
 
 	for_each_engine_id(to, dev_priv, id) {
 		int idx;
@@ -879,42 +895,43 @@ static void gen8_record_semaphore_state(struct drm_i915_private *dev_priv,
 		if (engine == to)
 			continue;
 
-		signal_offset = (GEN8_SIGNAL_OFFSET(engine, id) & (PAGE_SIZE - 1))
-				/ 4;
+		signal_offset =
+			(GEN8_SIGNAL_OFFSET(engine, id) & (PAGE_SIZE - 1)) / 4;
 		tmp = error->semaphore_obj->pages[0];
-		idx = intel_ring_sync_index(engine, to);
+		idx = intel_engine_sync_index(engine, to);
 
-		ering->semaphore_mboxes[idx] = tmp[signal_offset];
-		ering->semaphore_seqno[idx] = engine->semaphore.sync_seqno[idx];
+		ee->semaphore_mboxes[idx] = tmp[signal_offset];
+		ee->semaphore_seqno[idx] = engine->semaphore.sync_seqno[idx];
 	}
 }
 
-static void gen6_record_semaphore_state(struct drm_i915_private *dev_priv,
-					struct intel_engine_cs *engine,
-					struct drm_i915_error_ring *ering)
+static void gen6_record_semaphore_state(struct intel_engine_cs *engine,
+					struct drm_i915_error_engine *ee)
 {
-	ering->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(engine->mmio_base));
-	ering->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(engine->mmio_base));
-	ering->semaphore_seqno[0] = engine->semaphore.sync_seqno[0];
-	ering->semaphore_seqno[1] = engine->semaphore.sync_seqno[1];
+	struct drm_i915_private *dev_priv = engine->i915;
+
+	ee->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(engine->mmio_base));
+	ee->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(engine->mmio_base));
+	ee->semaphore_seqno[0] = engine->semaphore.sync_seqno[0];
+	ee->semaphore_seqno[1] = engine->semaphore.sync_seqno[1];
 
 	if (HAS_VEBOX(dev_priv)) {
-		ering->semaphore_mboxes[2] =
+		ee->semaphore_mboxes[2] =
 			I915_READ(RING_SYNC_2(engine->mmio_base));
-		ering->semaphore_seqno[2] = engine->semaphore.sync_seqno[2];
+		ee->semaphore_seqno[2] = engine->semaphore.sync_seqno[2];
 	}
 }
 
-static void engine_record_waiters(struct intel_engine_cs *engine,
-				  struct drm_i915_error_ring *ering)
+static void error_record_engine_waiters(struct intel_engine_cs *engine,
+					struct drm_i915_error_engine *ee)
 {
 	struct intel_breadcrumbs *b = &engine->breadcrumbs;
 	struct drm_i915_error_waiter *waiter;
 	struct rb_node *rb;
 	int count;
 
-	ering->num_waiters = 0;
-	ering->waiters = NULL;
+	ee->num_waiters = 0;
+	ee->waiters = NULL;
 
 	spin_lock(&b->lock);
 	count = 0;
@@ -930,7 +947,7 @@ static void engine_record_waiters(struct intel_engine_cs *engine,
 	if (!waiter)
 		return;
 
-	ering->waiters = waiter;
+	ee->waiters = waiter;
 
 	spin_lock(&b->lock);
 	for (rb = rb_first(&b->waiters); rb; rb = rb_next(rb)) {
@@ -941,55 +958,55 @@ static void engine_record_waiters(struct intel_engine_cs *engine,
 		waiter->seqno = w->seqno;
 		waiter++;
 
-		if (++ering->num_waiters == count)
+		if (++ee->num_waiters == count)
 			break;
 	}
 	spin_unlock(&b->lock);
 }
 
-static void i915_record_ring_state(struct drm_i915_private *dev_priv,
-				   struct drm_i915_error_state *error,
-				   struct intel_engine_cs *engine,
-				   struct drm_i915_error_ring *ering)
+static void error_record_engine_registers(struct drm_i915_error_state *error,
+					  struct intel_engine_cs *engine,
+					  struct drm_i915_error_engine *ee)
 {
+	struct drm_i915_private *dev_priv = engine->i915;
+
 	if (INTEL_GEN(dev_priv) >= 6) {
-		ering->rc_psmi = I915_READ(RING_PSMI_CTL(engine->mmio_base));
-		ering->fault_reg = I915_READ(RING_FAULT_REG(engine));
+		ee->rc_psmi = I915_READ(RING_PSMI_CTL(engine->mmio_base));
+		ee->fault_reg = I915_READ(RING_FAULT_REG(engine));
 		if (INTEL_GEN(dev_priv) >= 8)
-			gen8_record_semaphore_state(dev_priv, error, engine,
-						    ering);
+			gen8_record_semaphore_state(error, engine, ee);
 		else
-			gen6_record_semaphore_state(dev_priv, engine, ering);
+			gen6_record_semaphore_state(engine, ee);
 	}
 
 	if (INTEL_GEN(dev_priv) >= 4) {
-		ering->faddr = I915_READ(RING_DMA_FADD(engine->mmio_base));
-		ering->ipeir = I915_READ(RING_IPEIR(engine->mmio_base));
-		ering->ipehr = I915_READ(RING_IPEHR(engine->mmio_base));
-		ering->instdone = I915_READ(RING_INSTDONE(engine->mmio_base));
-		ering->instps = I915_READ(RING_INSTPS(engine->mmio_base));
-		ering->bbaddr = I915_READ(RING_BBADDR(engine->mmio_base));
+		ee->faddr = I915_READ(RING_DMA_FADD(engine->mmio_base));
+		ee->ipeir = I915_READ(RING_IPEIR(engine->mmio_base));
+		ee->ipehr = I915_READ(RING_IPEHR(engine->mmio_base));
+		ee->instdone = I915_READ(RING_INSTDONE(engine->mmio_base));
+		ee->instps = I915_READ(RING_INSTPS(engine->mmio_base));
+		ee->bbaddr = I915_READ(RING_BBADDR(engine->mmio_base));
 		if (INTEL_GEN(dev_priv) >= 8) {
-			ering->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(engine->mmio_base)) << 32;
-			ering->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(engine->mmio_base)) << 32;
+			ee->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(engine->mmio_base)) << 32;
+			ee->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(engine->mmio_base)) << 32;
 		}
-		ering->bbstate = I915_READ(RING_BBSTATE(engine->mmio_base));
+		ee->bbstate = I915_READ(RING_BBSTATE(engine->mmio_base));
 	} else {
-		ering->faddr = I915_READ(DMA_FADD_I8XX);
-		ering->ipeir = I915_READ(IPEIR);
-		ering->ipehr = I915_READ(IPEHR);
-		ering->instdone = I915_READ(GEN2_INSTDONE);
+		ee->faddr = I915_READ(DMA_FADD_I8XX);
+		ee->ipeir = I915_READ(IPEIR);
+		ee->ipehr = I915_READ(IPEHR);
+		ee->instdone = I915_READ(GEN2_INSTDONE);
 	}
 
-	ering->waiting = intel_engine_has_waiter(engine);
-	ering->instpm = I915_READ(RING_INSTPM(engine->mmio_base));
-	ering->acthd = intel_ring_get_active_head(engine);
-	ering->seqno = intel_engine_get_seqno(engine);
-	ering->last_seqno = engine->last_submitted_seqno;
-	ering->start = I915_READ_START(engine);
-	ering->head = I915_READ_HEAD(engine);
-	ering->tail = I915_READ_TAIL(engine);
-	ering->ctl = I915_READ_CTL(engine);
+	ee->waiting = intel_engine_has_waiter(engine);
+	ee->instpm = I915_READ(RING_INSTPM(engine->mmio_base));
+	ee->acthd = intel_engine_get_active_head(engine);
+	ee->seqno = intel_engine_get_seqno(engine);
+	ee->last_seqno = engine->last_submitted_seqno;
+	ee->start = I915_READ_START(engine);
+	ee->head = I915_READ_HEAD(engine);
+	ee->tail = I915_READ_TAIL(engine);
+	ee->ctl = I915_READ_CTL(engine);
 
 	if (I915_NEED_GFX_HWS(dev_priv)) {
 		i915_reg_t mmio;
@@ -1017,29 +1034,29 @@ static void i915_record_ring_state(struct drm_i915_private *dev_priv,
 			mmio = RING_HWS_PGA(engine->mmio_base);
 		}
 
-		ering->hws = I915_READ(mmio);
+		ee->hws = I915_READ(mmio);
 	}
 
-	ering->hangcheck_score = engine->hangcheck.score;
-	ering->hangcheck_action = engine->hangcheck.action;
+	ee->hangcheck_score = engine->hangcheck.score;
+	ee->hangcheck_action = engine->hangcheck.action;
 
 	if (USES_PPGTT(dev_priv)) {
 		int i;
 
-		ering->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(engine));
+		ee->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(engine));
 
 		if (IS_GEN6(dev_priv))
-			ering->vm_info.pp_dir_base =
+			ee->vm_info.pp_dir_base =
 				I915_READ(RING_PP_DIR_BASE_READ(engine));
 		else if (IS_GEN7(dev_priv))
-			ering->vm_info.pp_dir_base =
+			ee->vm_info.pp_dir_base =
 				I915_READ(RING_PP_DIR_BASE(engine));
 		else if (INTEL_GEN(dev_priv) >= 8)
 			for (i = 0; i < 4; i++) {
-				ering->vm_info.pdp[i] =
+				ee->vm_info.pdp[i] =
 					I915_READ(GEN8_RING_PDP_UDW(engine, i));
-				ering->vm_info.pdp[i] <<= 32;
-				ering->vm_info.pdp[i] |=
+				ee->vm_info.pdp[i] <<= 32;
+				ee->vm_info.pdp[i] |=
 					I915_READ(GEN8_RING_PDP_LDW(engine, i));
 			}
 	}
@@ -1048,7 +1065,7 @@ static void i915_record_ring_state(struct drm_i915_private *dev_priv,
 
 static void i915_gem_record_active_context(struct intel_engine_cs *engine,
 					   struct drm_i915_error_state *error,
-					   struct drm_i915_error_ring *ering)
+					   struct drm_i915_error_engine *ee)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
 	struct drm_i915_gem_object *obj;
@@ -1062,7 +1079,7 @@ static void i915_gem_record_active_context(struct intel_engine_cs *engine,
 			continue;
 
 		if ((error->ccid & PAGE_MASK) == i915_gem_obj_ggtt_offset(obj)) {
-			ering->ctx = i915_error_ggtt_object_create(dev_priv, obj);
+			ee->ctx = i915_error_ggtt_object_create(dev_priv, obj);
 			break;
 		}
 	}
@@ -1075,23 +1092,31 @@ static void i915_gem_record_rings(struct drm_i915_private *dev_priv,
 	struct drm_i915_gem_request *request;
 	int i, count;
 
+	if (dev_priv->semaphore_obj) {
+		error->semaphore_obj =
+			i915_error_ggtt_object_create(dev_priv,
+						      dev_priv->semaphore_obj);
+	}
+
 	for (i = 0; i < I915_NUM_ENGINES; i++) {
 		struct intel_engine_cs *engine = &dev_priv->engine[i];
+		struct drm_i915_error_engine *ee = &error->engine[i];
 
-		error->ring[i].pid = -1;
+		ee->pid = -1;
+		ee->engine_id = -1;
 
 		if (!intel_engine_initialized(engine))
 			continue;
 
-		error->ring[i].valid = true;
+		ee->engine_id = i;
 
-		i915_record_ring_state(dev_priv, error, engine, &error->ring[i]);
-		engine_record_waiters(engine, &error->ring[i]);
+		error_record_engine_registers(error, engine, ee);
+		error_record_engine_waiters(engine, ee);
 
 		request = i915_gem_find_active_request(engine);
 		if (request) {
 			struct i915_address_space *vm;
-			struct intel_ringbuffer *rb;
+			struct intel_ring *ring;
 
 			vm = request->ctx->ppgtt ?
 				&request->ctx->ppgtt->base : &ggtt->base;
@@ -1100,15 +1125,15 @@ static void i915_gem_record_rings(struct drm_i915_private *dev_priv,
 			 * as the simplest method to avoid being overwritten
 			 * by userspace.
 			 */
-			error->ring[i].batchbuffer =
+			ee->batchbuffer =
 				i915_error_object_create(dev_priv,
 							 request->batch_obj,
 							 vm);
 
 			if (HAS_BROKEN_CS_TLB(dev_priv))
-				error->ring[i].wa_batchbuffer =
+				ee->wa_batchbuffer =
 					i915_error_ggtt_object_create(dev_priv,
-							     engine->scratch.obj);
+								      engine->scratch.obj);
 
 			if (request->pid) {
 				struct task_struct *task;
@@ -1116,8 +1141,8 @@ static void i915_gem_record_rings(struct drm_i915_private *dev_priv,
 				rcu_read_lock();
 				task = pid_task(request->pid, PIDTYPE_PID);
 				if (task) {
-					strcpy(error->ring[i].comm, task->comm);
-					error->ring[i].pid = task->pid;
+					strcpy(ee->comm, task->comm);
+					ee->pid = task->pid;
 				}
 				rcu_read_unlock();
 			}
@@ -1125,44 +1150,40 @@ static void i915_gem_record_rings(struct drm_i915_private *dev_priv,
 			error->simulated |=
 				request->ctx->flags & CONTEXT_NO_ERROR_CAPTURE;
 
-			rb = request->ringbuf;
-			error->ring[i].cpu_ring_head = rb->head;
-			error->ring[i].cpu_ring_tail = rb->tail;
-			error->ring[i].ringbuffer =
+			ring = request->ring;
+			ee->cpu_ring_head = ring->head;
+			ee->cpu_ring_tail = ring->tail;
+			ee->ringbuffer =
 				i915_error_ggtt_object_create(dev_priv,
-							      rb->obj);
+							      ring->obj);
 		}
 
-		error->ring[i].hws_page =
+		ee->hws_page =
 			i915_error_ggtt_object_create(dev_priv,
 						      engine->status_page.obj);
 
-		if (engine->wa_ctx.obj) {
-			error->ring[i].wa_ctx =
-				i915_error_ggtt_object_create(dev_priv,
-							      engine->wa_ctx.obj);
-		}
+		ee->wa_ctx = i915_error_ggtt_object_create(dev_priv,
+							   engine->wa_ctx.obj);
 
-		i915_gem_record_active_context(engine, error, &error->ring[i]);
+		i915_gem_record_active_context(engine, error, ee);
 
 		count = 0;
-		list_for_each_entry(request, &engine->request_list, list)
+		list_for_each_entry(request, &engine->request_list, link)
 			count++;
 
-		error->ring[i].num_requests = count;
-		error->ring[i].requests =
-			kcalloc(count, sizeof(*error->ring[i].requests),
-				GFP_ATOMIC);
-		if (error->ring[i].requests == NULL) {
-			error->ring[i].num_requests = 0;
+		ee->num_requests = count;
+		ee->requests =
+			kcalloc(count, sizeof(*ee->requests), GFP_ATOMIC);
+		if (!ee->requests) {
+			ee->num_requests = 0;
 			continue;
 		}
 
 		count = 0;
-		list_for_each_entry(request, &engine->request_list, list) {
+		list_for_each_entry(request, &engine->request_list, link) {
 			struct drm_i915_error_request *erq;
 
-			if (count >= error->ring[i].num_requests) {
+			if (count >= ee->num_requests) {
 				/*
 				 * If the ring request list was changed in
 				 * between the point where the error request
@@ -1181,8 +1202,8 @@ static void i915_gem_record_rings(struct drm_i915_private *dev_priv,
 				break;
 			}
 
-			erq = &error->ring[i].requests[count++];
-			erq->seqno = request->seqno;
+			erq = &ee->requests[count++];
+			erq->seqno = request->fence.seqno;
 			erq->jiffies = request->emitted_jiffies;
 			erq->tail = request->postfix;
 		}
@@ -1209,7 +1230,7 @@ static void i915_gem_capture_vm(struct drm_i915_private *dev_priv,
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
 		list_for_each_entry(vma, &obj->vma_list, obj_link)
-			if (vma->vm == vm && vma->pin_count > 0)
+			if (vma->vm == vm && i915_vma_is_pinned(vma))
 				i++;
 	}
 	error->pinned_bo_count[ndx] = i - error->active_bo_count[ndx];
@@ -1352,20 +1373,20 @@ static void i915_error_capture_msg(struct drm_i915_private *dev_priv,
 				   const char *error_msg)
 {
 	u32 ecode;
-	int ring_id = -1, len;
+	int engine_id = -1, len;
 
-	ecode = i915_error_generate_code(dev_priv, error, &ring_id);
+	ecode = i915_error_generate_code(dev_priv, error, &engine_id);
 
 	len = scnprintf(error->error_msg, sizeof(error->error_msg),
 			"GPU HANG: ecode %d:%d:0x%08x",
-			INTEL_GEN(dev_priv), ring_id, ecode);
+			INTEL_GEN(dev_priv), engine_id, ecode);
 
-	if (ring_id != -1 && error->ring[ring_id].pid != -1)
+	if (engine_id != -1 && error->engine[engine_id].pid != -1)
 		len += scnprintf(error->error_msg + len,
 				 sizeof(error->error_msg) - len,
 				 ", in %s [%d]",
-				 error->ring[ring_id].comm,
-				 error->ring[ring_id].pid);
+				 error->engine[engine_id].comm,
+				 error->engine[engine_id].pid);
 
 	scnprintf(error->error_msg + len, sizeof(error->error_msg) - len,
 		  ", reason: %s, action: %s",

drivers/gpu/drm/i915/i915_guc_submission.c

@@ -363,7 +363,7 @@ static void guc_init_ctx_desc(struct intel_guc *guc,
 		lrc->context_id = (client->ctx_index << GUC_ELC_CTXID_OFFSET) |
 				(engine->guc_id << GUC_ELC_ENGINE_OFFSET);
 
-		obj = ce->ringbuf->obj;
+		obj = ce->ring->obj;
 		gfx_addr = i915_gem_obj_ggtt_offset(obj);
 
 		lrc->ring_begin = gfx_addr;
@@ -506,7 +506,7 @@ static void guc_add_workqueue_item(struct i915_guc_client *gc,
 							     rq->engine);
 
 	wqi->ring_tail = tail << WQ_RING_TAIL_SHIFT;
-	wqi->fence_id = rq->seqno;
+	wqi->fence_id = rq->fence.seqno;
 
 	kunmap_atomic(base);
 }
@@ -585,7 +585,7 @@ static int guc_ring_doorbell(struct i915_guc_client *gc)
  * The only error here arises if the doorbell hardware isn't functioning
  * as expected, which really shouln't happen.
  */
-int i915_guc_submit(struct drm_i915_gem_request *rq)
+static void i915_guc_submit(struct drm_i915_gem_request *rq)
 {
 	unsigned int engine_id = rq->engine->id;
 	struct intel_guc *guc = &rq->i915->guc;
@@ -601,9 +601,7 @@ int i915_guc_submit(struct drm_i915_gem_request *rq)
 		client->b_fail += 1;
 
 	guc->submissions[engine_id] += 1;
-	guc->last_seqno[engine_id] = rq->seqno;
-
-	return b_ret;
+	guc->last_seqno[engine_id] = rq->fence.seqno;
 }
 
 /*
@@ -633,13 +631,13 @@ gem_allocate_guc_obj(struct drm_i915_private *dev_priv, u32 size)
 		return NULL;
 
 	if (i915_gem_object_get_pages(obj)) {
-		drm_gem_object_unreference(&obj->base);
+		i915_gem_object_put(obj);
 		return NULL;
 	}
 
-	if (i915_gem_obj_ggtt_pin(obj, PAGE_SIZE,
-			PIN_OFFSET_BIAS | GUC_WOPCM_TOP)) {
-		drm_gem_object_unreference(&obj->base);
+	if (i915_gem_object_ggtt_pin(obj, NULL, 0, PAGE_SIZE,
+				     PIN_OFFSET_BIAS | GUC_WOPCM_TOP)) {
+		i915_gem_object_put(obj);
 		return NULL;
 	}
 
@@ -661,7 +659,7 @@ static void gem_release_guc_obj(struct drm_i915_gem_object *obj)
 	if (i915_gem_obj_is_pinned(obj))
 		i915_gem_object_ggtt_unpin(obj);
 
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 }
 
 static void
@@ -992,6 +990,7 @@ int i915_guc_submission_enable(struct drm_i915_private *dev_priv)
 {
 	struct intel_guc *guc = &dev_priv->guc;
 	struct i915_guc_client *client;
+	struct intel_engine_cs *engine;
 
 	/* client for execbuf submission */
 	client = guc_client_alloc(dev_priv,
@@ -1006,6 +1005,10 @@ int i915_guc_submission_enable(struct drm_i915_private *dev_priv)
 	host2guc_sample_forcewake(guc, client);
 	guc_init_doorbell_hw(guc);
 
+	/* Take over from manual control of ELSP (execlists) */
+	for_each_engine(engine, dev_priv)
+		engine->submit_request = i915_guc_submit;
+
 	return 0;
 }
 
@@ -1013,6 +1016,12 @@ void i915_guc_submission_disable(struct drm_i915_private *dev_priv)
 {
 	struct intel_guc *guc = &dev_priv->guc;
 
+	if (!guc->execbuf_client)
+		return;
+
+	/* Revert back to manual ELSP submission */
+	intel_execlists_enable_submission(dev_priv);
+
 	guc_client_free(dev_priv, guc->execbuf_client);
 	guc->execbuf_client = NULL;
 }

drivers/gpu/drm/i915/i915_irq.c

@@ -656,12 +656,6 @@ static void i915_enable_asle_pipestat(struct drm_i915_private *dev_priv)
  *   of horizontal active on the first line of vertical active
  */
 
-static u32 i8xx_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
-{
-	/* Gen2 doesn't have a hardware frame counter */
-	return 0;
-}
-
 /* Called from drm generic code, passed a 'crtc', which
  * we use as a pipe index
  */
@@ -1105,9 +1099,10 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	new_delay = dev_priv->rps.cur_freq;
 	min = dev_priv->rps.min_freq_softlimit;
 	max = dev_priv->rps.max_freq_softlimit;
-
-	if (client_boost) {
-		new_delay = dev_priv->rps.max_freq_softlimit;
+	if (client_boost || any_waiters(dev_priv))
+		max = dev_priv->rps.max_freq;
+	if (client_boost && new_delay < dev_priv->rps.boost_freq) {
+		new_delay = dev_priv->rps.boost_freq;
 		adj = 0;
 	} else if (pm_iir & GEN6_PM_RP_UP_THRESHOLD) {
 		if (adj > 0)
@@ -1122,7 +1117,7 @@ static void gen6_pm_rps_work(struct work_struct *work)
 			new_delay = dev_priv->rps.efficient_freq;
 			adj = 0;
 		}
-	} else if (any_waiters(dev_priv)) {
+	} else if (client_boost || any_waiters(dev_priv)) {
 		adj = 0;
 	} else if (pm_iir & GEN6_PM_RP_DOWN_TIMEOUT) {
 		if (dev_priv->rps.cur_freq > dev_priv->rps.efficient_freq)
@@ -2803,13 +2798,6 @@ static void gen8_disable_vblank(struct drm_device *dev, unsigned int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
-static bool
-ring_idle(struct intel_engine_cs *engine, u32 seqno)
-{
-	return i915_seqno_passed(seqno,
-				 READ_ONCE(engine->last_submitted_seqno));
-}
-
 static bool
 ipehr_is_semaphore_wait(struct intel_engine_cs *engine, u32 ipehr)
 {
@@ -2859,6 +2847,7 @@ static struct intel_engine_cs *
 semaphore_waits_for(struct intel_engine_cs *engine, u32 *seqno)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
+	void __iomem *vaddr;
 	u32 cmd, ipehr, head;
 	u64 offset = 0;
 	int i, backwards;
@@ -2897,6 +2886,7 @@ semaphore_waits_for(struct intel_engine_cs *engine, u32 *seqno)
 	 */
 	head = I915_READ_HEAD(engine) & HEAD_ADDR;
 	backwards = (INTEL_GEN(dev_priv) >= 8) ? 5 : 4;
+	vaddr = (void __iomem *)engine->buffer->vaddr;
 
 	for (i = backwards; i; --i) {
 		/*
@@ -2907,7 +2897,7 @@ semaphore_waits_for(struct intel_engine_cs *engine, u32 *seqno)
 		head &= engine->buffer->size - 1;
 
 		/* This here seems to blow up */
-		cmd = ioread32(engine->buffer->virtual_start + head);
+		cmd = ioread32(vaddr + head);
 		if (cmd == ipehr)
 			break;
 
@@ -2917,11 +2907,11 @@ semaphore_waits_for(struct intel_engine_cs *engine, u32 *seqno)
 	if (!i)
 		return NULL;
 
-	*seqno = ioread32(engine->buffer->virtual_start + head + 4) + 1;
+	*seqno = ioread32(vaddr + head + 4) + 1;
 	if (INTEL_GEN(dev_priv) >= 8) {
-		offset = ioread32(engine->buffer->virtual_start + head + 12);
+		offset = ioread32(vaddr + head + 12);
 		offset <<= 32;
-		offset = ioread32(engine->buffer->virtual_start + head + 8);
+		offset |= ioread32(vaddr + head + 8);
 	}
 	return semaphore_wait_to_signaller_ring(engine, ipehr, offset);
 }
@@ -2990,7 +2980,7 @@ static bool subunits_stuck(struct intel_engine_cs *engine)
 	return stuck;
 }
 
-static enum intel_ring_hangcheck_action
+static enum intel_engine_hangcheck_action
 head_stuck(struct intel_engine_cs *engine, u64 acthd)
 {
 	if (acthd != engine->hangcheck.acthd) {
@@ -3008,11 +2998,11 @@ head_stuck(struct intel_engine_cs *engine, u64 acthd)
 	return HANGCHECK_HUNG;
 }
 
-static enum intel_ring_hangcheck_action
-ring_stuck(struct intel_engine_cs *engine, u64 acthd)
+static enum intel_engine_hangcheck_action
+engine_stuck(struct intel_engine_cs *engine, u64 acthd)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
-	enum intel_ring_hangcheck_action ha;
+	enum intel_engine_hangcheck_action ha;
 	u32 tmp;
 
 	ha = head_stuck(engine, acthd);
@@ -3121,14 +3111,14 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 		if (engine->irq_seqno_barrier)
 			engine->irq_seqno_barrier(engine);
 
-		acthd = intel_ring_get_active_head(engine);
+		acthd = intel_engine_get_active_head(engine);
 		seqno = intel_engine_get_seqno(engine);
 
 		/* Reset stuck interrupts between batch advances */
 		user_interrupts = 0;
 
 		if (engine->hangcheck.seqno == seqno) {
-			if (ring_idle(engine, seqno)) {
+			if (!intel_engine_is_active(engine)) {
 				engine->hangcheck.action = HANGCHECK_IDLE;
 				if (busy) {
 					/* Safeguard against driver failure */
@@ -3137,13 +3127,13 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 				}
 			} else {
 				/* We always increment the hangcheck score
-				 * if the ring is busy and still processing
+				 * if the engine is busy and still processing
 				 * the same request, so that no single request
 				 * can run indefinitely (such as a chain of
 				 * batches). The only time we do not increment
 				 * the hangcheck score on this ring, if this
-				 * ring is in a legitimate wait for another
-				 * ring. In that case the waiting ring is a
+				 * engine is in a legitimate wait for another
+				 * engine. In that case the waiting engine is a
 				 * victim and we want to be sure we catch the
 				 * right culprit. Then every time we do kick
 				 * the ring, add a small increment to the
@@ -3151,8 +3141,8 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 				 * being repeatedly kicked and so responsible
 				 * for stalling the machine.
 				 */
-				engine->hangcheck.action = ring_stuck(engine,
-								      acthd);
+				engine->hangcheck.action =
+					engine_stuck(engine, acthd);
 
 				switch (engine->hangcheck.action) {
 				case HANGCHECK_IDLE:
@@ -4542,8 +4532,9 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 			  i915_hangcheck_elapsed);
 
 	if (IS_GEN2(dev_priv)) {
+		/* Gen2 doesn't have a hardware frame counter */
 		dev->max_vblank_count = 0;
-		dev->driver->get_vblank_counter = i8xx_get_vblank_counter;
+		dev->driver->get_vblank_counter = drm_vblank_no_hw_counter;
 	} else if (IS_G4X(dev_priv) || INTEL_INFO(dev_priv)->gen >= 5) {
 		dev->max_vblank_count = 0xffffffff; /* full 32 bit counter */
 		dev->driver->get_vblank_counter = g4x_get_vblank_counter;

drivers/gpu/drm/i915/i915_pci.c

@@ -173,6 +173,7 @@ static const struct intel_device_info intel_pineview_info = {
 	.gen = 3, .is_g33 = 1, .is_pineview = 1, .is_mobile = 1, .num_pipes = 2,
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.has_overlay = 1,
+	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
 	CURSOR_OFFSETS,
 };

drivers/gpu/drm/i915/i915_reg.h

@@ -186,13 +186,13 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define  GEN9_GRDOM_GUC			(1 << 5)
 #define  GEN8_GRDOM_MEDIA2		(1 << 7)
 
-#define RING_PP_DIR_BASE(ring)		_MMIO((ring)->mmio_base+0x228)
-#define RING_PP_DIR_BASE_READ(ring)	_MMIO((ring)->mmio_base+0x518)
-#define RING_PP_DIR_DCLV(ring)		_MMIO((ring)->mmio_base+0x220)
+#define RING_PP_DIR_BASE(engine)	_MMIO((engine)->mmio_base+0x228)
+#define RING_PP_DIR_BASE_READ(engine)	_MMIO((engine)->mmio_base+0x518)
+#define RING_PP_DIR_DCLV(engine)	_MMIO((engine)->mmio_base+0x220)
 #define   PP_DIR_DCLV_2G		0xffffffff
 
-#define GEN8_RING_PDP_UDW(ring, n)	_MMIO((ring)->mmio_base+0x270 + (n) * 8 + 4)
-#define GEN8_RING_PDP_LDW(ring, n)	_MMIO((ring)->mmio_base+0x270 + (n) * 8)
+#define GEN8_RING_PDP_UDW(engine, n)	_MMIO((engine)->mmio_base+0x270 + (n) * 8 + 4)
+#define GEN8_RING_PDP_LDW(engine, n)	_MMIO((engine)->mmio_base+0x270 + (n) * 8)
 
 #define GEN8_R_PWR_CLK_STATE		_MMIO(0x20C8)
 #define   GEN8_RPCS_ENABLE		(1 << 31)
@@ -1536,6 +1536,7 @@ enum skl_disp_power_wells {
 #define BALANCE_LEG_MASK(port)		(7<<(8+3*(port)))
 /* Balance leg disable bits */
 #define BALANCE_LEG_DISABLE_SHIFT	23
+#define BALANCE_LEG_DISABLE(port)	(1 << (23 + (port)))
 
 /*
  * Fence registers
@@ -1647,7 +1648,7 @@ enum skl_disp_power_wells {
 #define   ARB_MODE_BWGTLB_DISABLE (1<<9)
 #define   ARB_MODE_SWIZZLE_BDW	(1<<1)
 #define RENDER_HWS_PGA_GEN7	_MMIO(0x04080)
-#define RING_FAULT_REG(ring)	_MMIO(0x4094 + 0x100*(ring)->id)
+#define RING_FAULT_REG(engine)	_MMIO(0x4094 + 0x100*(engine)->hw_id)
 #define   RING_FAULT_GTTSEL_MASK (1<<11)
 #define   RING_FAULT_SRCID(x)	(((x) >> 3) & 0xff)
 #define   RING_FAULT_FAULT_TYPE(x) (((x) >> 1) & 0x3)
@@ -1845,7 +1846,7 @@ enum skl_disp_power_wells {
 
 #define GFX_MODE	_MMIO(0x2520)
 #define GFX_MODE_GEN7	_MMIO(0x229c)
-#define RING_MODE_GEN7(ring)	_MMIO((ring)->mmio_base+0x29c)
+#define RING_MODE_GEN7(engine)	_MMIO((engine)->mmio_base+0x29c)
 #define   GFX_RUN_LIST_ENABLE		(1<<15)
 #define   GFX_INTERRUPT_STEERING	(1<<14)
 #define   GFX_TLB_INVALIDATE_EXPLICIT	(1<<13)
@@ -6132,6 +6133,7 @@ enum {
 # define GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC	((1<<10) | (1<<26))
 # define GEN9_RHWO_OPTIMIZATION_DISABLE		(1<<14)
 #define COMMON_SLICE_CHICKEN2			_MMIO(0x7014)
+# define GEN9_DISABLE_GATHER_AT_SET_SHADER_COMMON_SLICE (1<<12)
 # define GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION (1<<8)
 # define GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE	(1<<0)
 
@@ -6958,6 +6960,9 @@ enum {
 #define  ECOBUS					_MMIO(0xa180)
 #define    FORCEWAKE_MT_ENABLE			(1<<5)
 #define  VLV_SPAREG2H				_MMIO(0xA194)
+#define  GEN9_PWRGT_DOMAIN_STATUS		_MMIO(0xA2A0)
+#define   GEN9_PWRGT_MEDIA_STATUS_MASK		(1 << 0)
+#define   GEN9_PWRGT_RENDER_STATUS_MASK		(1 << 1)
 
 #define  GTFIFODBG				_MMIO(0x120000)
 #define    GT_FIFO_SBDEDICATE_FREE_ENTRY_CHV	(0x1f << 20)
@@ -7485,6 +7490,7 @@ enum {
 #define _DDI_BUF_TRANS_A		0x64E00
 #define _DDI_BUF_TRANS_B		0x64E60
 #define DDI_BUF_TRANS_LO(port, i)	_MMIO(_PORT(port, _DDI_BUF_TRANS_A, _DDI_BUF_TRANS_B) + (i) * 8)
+#define  DDI_BUF_BALANCE_LEG_ENABLE	(1 << 31)
 #define DDI_BUF_TRANS_HI(port, i)	_MMIO(_PORT(port, _DDI_BUF_TRANS_A, _DDI_BUF_TRANS_B) + (i) * 8 + 4)
 
 /* Sideband Interface (SBI) is programmed indirectly, via

drivers/gpu/drm/i915/i915_sysfs.c

@@ -271,8 +271,6 @@ static ssize_t gt_act_freq_mhz_show(struct device *kdev,
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	int ret;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
 	intel_runtime_pm_get(dev_priv);
 
 	mutex_lock(&dev_priv->rps.hw_lock);
@@ -303,19 +301,46 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+			intel_gpu_freq(dev_priv,
+				       dev_priv->rps.cur_freq));
+}
 
-	intel_runtime_pm_get(dev_priv);
+static ssize_t gt_boost_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
+{
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
+	struct drm_i915_private *dev_priv = to_i915(minor->dev);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+			intel_gpu_freq(dev_priv,
+				       dev_priv->rps.boost_freq));
+}
+
+static ssize_t gt_boost_freq_mhz_store(struct device *kdev,
+				       struct device_attribute *attr,
+				       const char *buf, size_t count)
+{
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
+	struct drm_device *dev = minor->dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	u32 val;
+	ssize_t ret;
+
+	ret = kstrtou32(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	/* 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)
+		return -EINVAL;
 
 	mutex_lock(&dev_priv->rps.hw_lock);
-	ret = intel_gpu_freq(dev_priv, dev_priv->rps.cur_freq);
+	dev_priv->rps.boost_freq = val;
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
-	intel_runtime_pm_put(dev_priv);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n", ret);
+	return count;
 }
 
 static ssize_t vlv_rpe_freq_mhz_show(struct device *kdev,
@@ -325,9 +350,9 @@ static ssize_t vlv_rpe_freq_mhz_show(struct device *kdev,
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 
-	return snprintf(buf, PAGE_SIZE,
-			"%d\n",
-			intel_gpu_freq(dev_priv, dev_priv->rps.efficient_freq));
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+			intel_gpu_freq(dev_priv,
+				       dev_priv->rps.efficient_freq));
 }
 
 static ssize_t gt_max_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
@@ -335,15 +360,10 @@ static ssize_t gt_max_freq_mhz_show(struct device *kdev, struct device_attribute
 	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
-
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
-	mutex_lock(&dev_priv->rps.hw_lock);
-	ret = intel_gpu_freq(dev_priv, dev_priv->rps.max_freq_softlimit);
-	mutex_unlock(&dev_priv->rps.hw_lock);
 
-	return snprintf(buf, PAGE_SIZE, "%d\n", ret);
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+			intel_gpu_freq(dev_priv,
+				       dev_priv->rps.max_freq_softlimit));
 }
 
 static ssize_t gt_max_freq_mhz_store(struct device *kdev,
@@ -360,8 +380,6 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 	if (ret)
 		return ret;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
 	intel_runtime_pm_get(dev_priv);
 
 	mutex_lock(&dev_priv->rps.hw_lock);
@@ -403,15 +421,10 @@ static ssize_t gt_min_freq_mhz_show(struct device *kdev, struct device_attribute
 	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
-
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
 
-	mutex_lock(&dev_priv->rps.hw_lock);
-	ret = intel_gpu_freq(dev_priv, dev_priv->rps.min_freq_softlimit);
-	mutex_unlock(&dev_priv->rps.hw_lock);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n", ret);
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+			intel_gpu_freq(dev_priv,
+				       dev_priv->rps.min_freq_softlimit));
 }
 
 static ssize_t gt_min_freq_mhz_store(struct device *kdev,
@@ -428,8 +441,6 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 	if (ret)
 		return ret;
 
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
 	intel_runtime_pm_get(dev_priv);
 
 	mutex_lock(&dev_priv->rps.hw_lock);
@@ -465,6 +476,7 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 
 static DEVICE_ATTR(gt_act_freq_mhz, S_IRUGO, gt_act_freq_mhz_show, NULL);
 static DEVICE_ATTR(gt_cur_freq_mhz, S_IRUGO, gt_cur_freq_mhz_show, NULL);
+static DEVICE_ATTR(gt_boost_freq_mhz, S_IRUGO, gt_boost_freq_mhz_show, gt_boost_freq_mhz_store);
 static DEVICE_ATTR(gt_max_freq_mhz, S_IRUGO | S_IWUSR, gt_max_freq_mhz_show, gt_max_freq_mhz_store);
 static DEVICE_ATTR(gt_min_freq_mhz, S_IRUGO | S_IWUSR, gt_min_freq_mhz_show, gt_min_freq_mhz_store);
 
@@ -498,6 +510,7 @@ static ssize_t gt_rp_mhz_show(struct device *kdev, struct device_attribute *attr
 static const struct attribute *gen6_attrs[] = {
 	&dev_attr_gt_act_freq_mhz.attr,
 	&dev_attr_gt_cur_freq_mhz.attr,
+	&dev_attr_gt_boost_freq_mhz.attr,
 	&dev_attr_gt_max_freq_mhz.attr,
 	&dev_attr_gt_min_freq_mhz.attr,
 	&dev_attr_gt_RP0_freq_mhz.attr,
@@ -509,6 +522,7 @@ static const struct attribute *gen6_attrs[] = {
 static const struct attribute *vlv_attrs[] = {
 	&dev_attr_gt_act_freq_mhz.attr,
 	&dev_attr_gt_cur_freq_mhz.attr,
+	&dev_attr_gt_boost_freq_mhz.attr,
 	&dev_attr_gt_max_freq_mhz.attr,
 	&dev_attr_gt_min_freq_mhz.attr,
 	&dev_attr_gt_RP0_freq_mhz.attr,

drivers/gpu/drm/i915/i915_trace.h

@@ -394,25 +394,27 @@ DEFINE_EVENT(i915_gem_object, i915_gem_object_destroy,
 );
 
 TRACE_EVENT(i915_gem_evict,
-	    TP_PROTO(struct drm_device *dev, u32 size, u32 align, unsigned flags),
-	    TP_ARGS(dev, size, align, flags),
+	    TP_PROTO(struct i915_address_space *vm, u32 size, u32 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(unsigned, flags)
+			     __field(unsigned int, flags)
 			    ),
 
 	    TP_fast_assign(
-			   __entry->dev = dev->primary->index;
+			   __entry->dev = vm->dev->primary->index;
+			   __entry->vm = vm;
 			   __entry->size = size;
 			   __entry->align = align;
 			   __entry->flags = flags;
 			  ),
 
-	    TP_printk("dev=%d, size=%d, align=%d %s",
-		      __entry->dev, __entry->size, __entry->align,
+	    TP_printk("dev=%d, vm=%p, size=%d, align=%d %s",
+		      __entry->dev, __entry->vm, __entry->size, __entry->align,
 		      __entry->flags & PIN_MAPPABLE ? ", mappable" : "")
 );
 
@@ -449,10 +451,9 @@ TRACE_EVENT(i915_gem_evict_vm,
 );
 
 TRACE_EVENT(i915_gem_ring_sync_to,
-	    TP_PROTO(struct drm_i915_gem_request *to_req,
-		     struct intel_engine_cs *from,
-		     struct drm_i915_gem_request *req),
-	    TP_ARGS(to_req, from, req),
+	    TP_PROTO(struct drm_i915_gem_request *to,
+		     struct drm_i915_gem_request *from),
+	    TP_ARGS(to, from),
 
 	    TP_STRUCT__entry(
 			     __field(u32, dev)
@@ -463,9 +464,9 @@ TRACE_EVENT(i915_gem_ring_sync_to,
 
 	    TP_fast_assign(
 			   __entry->dev = from->i915->drm.primary->index;
-			   __entry->sync_from = from->id;
-			   __entry->sync_to = to_req->engine->id;
-			   __entry->seqno = i915_gem_request_get_seqno(req);
+			   __entry->sync_from = from->engine->id;
+			   __entry->sync_to = to->engine->id;
+			   __entry->seqno = from->fence.seqno;
 			   ),
 
 	    TP_printk("dev=%u, sync-from=%u, sync-to=%u, seqno=%u",
@@ -488,9 +489,9 @@ TRACE_EVENT(i915_gem_ring_dispatch,
 	    TP_fast_assign(
 			   __entry->dev = req->i915->drm.primary->index;
 			   __entry->ring = req->engine->id;
-			   __entry->seqno = req->seqno;
+			   __entry->seqno = req->fence.seqno;
 			   __entry->flags = flags;
-			   intel_engine_enable_signaling(req);
+			   fence_enable_sw_signaling(&req->fence);
 			   ),
 
 	    TP_printk("dev=%u, ring=%u, seqno=%u, flags=%x",
@@ -533,7 +534,7 @@ DECLARE_EVENT_CLASS(i915_gem_request,
 	    TP_fast_assign(
 			   __entry->dev = req->i915->drm.primary->index;
 			   __entry->ring = req->engine->id;
-			   __entry->seqno = req->seqno;
+			   __entry->seqno = req->fence.seqno;
 			   ),
 
 	    TP_printk("dev=%u, ring=%u, seqno=%u",
@@ -595,7 +596,7 @@ TRACE_EVENT(i915_gem_request_wait_begin,
 	    TP_fast_assign(
 			   __entry->dev = req->i915->drm.primary->index;
 			   __entry->ring = req->engine->id;
-			   __entry->seqno = req->seqno;
+			   __entry->seqno = req->fence.seqno;
 			   __entry->blocking =
 				     mutex_is_locked(&req->i915->drm.struct_mutex);
 			   ),

drivers/gpu/drm/i915/i915_vgpu.c

@@ -97,6 +97,7 @@ static struct _balloon_info_ bl_info;
 
 /**
  * intel_vgt_deballoon - deballoon reserved graphics address trunks
+ * @dev_priv: i915 device private data
  *
  * This function is called to deallocate the ballooned-out graphic memory, when
  * driver is unloaded or when ballooning fails.
@@ -138,7 +139,7 @@ static int vgt_balloon_space(struct drm_mm *mm,
 
 /**
  * intel_vgt_balloon - balloon out reserved graphics address trunks
- * @dev: drm device
+ * @dev_priv: i915 device private data
  *
  * This function is called at the initialization stage, to balloon out the
  * graphic address space allocated to other vGPUs, by marking these spaces as

drivers/gpu/drm/i915/intel_audio.c

@@ -600,6 +600,8 @@ static void i915_audio_component_codec_wake_override(struct device *dev,
 	if (!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv))
 		return;
 
+	i915_audio_component_get_power(dev);
+
 	/*
 	 * Enable/disable generating the codec wake signal, overriding the
 	 * internal logic to generate the codec wake to controller.
@@ -615,6 +617,8 @@ static void i915_audio_component_codec_wake_override(struct device *dev,
 		I915_WRITE(HSW_AUD_CHICKENBIT, tmp);
 		usleep_range(1000, 1500);
 	}
+
+	i915_audio_component_put_power(dev);
 }
 
 /* Get CDCLK in kHz  */
@@ -648,6 +652,7 @@ static int i915_audio_component_sync_audio_rate(struct device *dev,
 	    !IS_HASWELL(dev_priv))
 		return 0;
 
+	i915_audio_component_get_power(dev);
 	mutex_lock(&dev_priv->av_mutex);
 	/* 1. get the pipe */
 	intel_encoder = dev_priv->dig_port_map[port];
@@ -698,6 +703,7 @@ static int i915_audio_component_sync_audio_rate(struct device *dev,
 
  unlock:
 	mutex_unlock(&dev_priv->av_mutex);
+	i915_audio_component_put_power(dev);
 	return err;
 }
 

drivers/gpu/drm/i915/intel_breadcrumbs.c

@@ -51,6 +51,13 @@ static void irq_enable(struct intel_engine_cs *engine)
 	 */
 	engine->breadcrumbs.irq_posted = true;
 
+	/* Make sure the current hangcheck doesn't falsely accuse a just
+	 * started irq handler from missing an interrupt (because the
+	 * interrupt count still matches the stale value from when
+	 * the irq handler was disabled, many hangchecks ago).
+	 */
+	engine->breadcrumbs.irq_wakeups++;
+
 	spin_lock_irq(&engine->i915->irq_lock);
 	engine->irq_enable(engine);
 	spin_unlock_irq(&engine->i915->irq_lock);
@@ -436,6 +443,7 @@ static int intel_breadcrumbs_signaler(void *arg)
 			 */
 			intel_engine_remove_wait(engine,
 						 &request->signaling.wait);
+			fence_signal(&request->fence);
 
 			/* Find the next oldest signal. Note that as we have
 			 * not been holding the lock, another client may
@@ -452,7 +460,7 @@ static int intel_breadcrumbs_signaler(void *arg)
 			rb_erase(&request->signaling.node, &b->signals);
 			spin_unlock(&b->lock);
 
-			i915_gem_request_unreference(request);
+			i915_gem_request_put(request);
 		} else {
 			if (kthread_should_stop())
 				break;
@@ -472,18 +480,14 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request)
 	struct rb_node *parent, **p;
 	bool first, wakeup;
 
-	if (unlikely(READ_ONCE(request->signaling.wait.tsk)))
-		return;
-
-	spin_lock(&b->lock);
-	if (unlikely(request->signaling.wait.tsk)) {
-		wakeup = false;
-		goto unlock;
-	}
+	/* locked by fence_enable_sw_signaling() */
+	assert_spin_locked(&request->lock);
 
 	request->signaling.wait.tsk = b->signaler;
-	request->signaling.wait.seqno = request->seqno;
-	i915_gem_request_reference(request);
+	request->signaling.wait.seqno = request->fence.seqno;
+	i915_gem_request_get(request);
+
+	spin_lock(&b->lock);
 
 	/* First add ourselves into the list of waiters, but register our
 	 * bottom-half as the signaller thread. As per usual, only the oldest
@@ -504,8 +508,8 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request)
 	p = &b->signals.rb_node;
 	while (*p) {
 		parent = *p;
-		if (i915_seqno_passed(request->seqno,
-				      to_signaler(parent)->seqno)) {
+		if (i915_seqno_passed(request->fence.seqno,
+				      to_signaler(parent)->fence.seqno)) {
 			p = &parent->rb_right;
 			first = false;
 		} else {
@@ -517,7 +521,6 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request)
 	if (first)
 		smp_store_mb(b->first_signal, request);
 
-unlock:
 	spin_unlock(&b->lock);
 
 	if (wakeup)

drivers/gpu/drm/i915/intel_csr.c

@@ -32,13 +32,6 @@
  * onwards to drive newly added DMC (Display microcontroller) in display
  * engine to save and restore the state of display engine when it enter into
  * low-power state and comes back to normal.
- *
- * Firmware loading status will be one of the below states: FW_UNINITIALIZED,
- * FW_LOADED, FW_FAILED.
- *
- * Once the firmware is written into the registers status will be moved from
- * FW_UNINITIALIZED to FW_LOADED and for any erroneous condition status will
- * be moved to FW_FAILED.
  */
 
 #define I915_CSR_KBL "i915/kbl_dmc_ver1.bin"

drivers/gpu/drm/i915/intel_ddi.c

@@ -145,7 +145,7 @@ static const struct ddi_buf_trans skl_ddi_translations_dp[] = {
 static const struct ddi_buf_trans skl_u_ddi_translations_dp[] = {
 	{ 0x0000201B, 0x000000A2, 0x0 },
 	{ 0x00005012, 0x00000088, 0x0 },
-	{ 0x80007011, 0x000000CD, 0x0 },
+	{ 0x80007011, 0x000000CD, 0x1 },
 	{ 0x80009010, 0x000000C0, 0x1 },
 	{ 0x0000201B, 0x0000009D, 0x0 },
 	{ 0x80005012, 0x000000C0, 0x1 },
@@ -158,7 +158,7 @@ static const struct ddi_buf_trans skl_u_ddi_translations_dp[] = {
 static const struct ddi_buf_trans skl_y_ddi_translations_dp[] = {
 	{ 0x00000018, 0x000000A2, 0x0 },
 	{ 0x00005012, 0x00000088, 0x0 },
-	{ 0x80007011, 0x000000CD, 0x0 },
+	{ 0x80007011, 0x000000CD, 0x3 },
 	{ 0x80009010, 0x000000C0, 0x3 },
 	{ 0x00000018, 0x0000009D, 0x0 },
 	{ 0x80005012, 0x000000C0, 0x3 },
@@ -301,45 +301,34 @@ static const struct bxt_ddi_buf_trans bxt_ddi_translations_hdmi[] = {
 	{ 154, 0x9A, 1, 128, true },	/* 9:	1200		0   */
 };
 
-static void bxt_ddi_vswing_sequence(struct drm_i915_private *dev_priv,
-				    u32 level, enum port port, int type);
-
-static void ddi_get_encoder_port(struct intel_encoder *intel_encoder,
-				 struct intel_digital_port **dig_port,
-				 enum port *port)
+enum port intel_ddi_get_encoder_port(struct intel_encoder *encoder)
 {
-	struct drm_encoder *encoder = &intel_encoder->base;
-
-	switch (intel_encoder->type) {
+	switch (encoder->type) {
 	case INTEL_OUTPUT_DP_MST:
-		*dig_port = enc_to_mst(encoder)->primary;
-		*port = (*dig_port)->port;
-		break;
-	default:
-		WARN(1, "Invalid DDI encoder type %d\n", intel_encoder->type);
-		/* fallthrough and treat as unknown */
+		return enc_to_mst(&encoder->base)->primary->port;
 	case INTEL_OUTPUT_DP:
 	case INTEL_OUTPUT_EDP:
 	case INTEL_OUTPUT_HDMI:
 	case INTEL_OUTPUT_UNKNOWN:
-		*dig_port = enc_to_dig_port(encoder);
-		*port = (*dig_port)->port;
-		break;
+		return enc_to_dig_port(&encoder->base)->port;
 	case INTEL_OUTPUT_ANALOG:
-		*dig_port = NULL;
-		*port = PORT_E;
-		break;
+		return PORT_E;
+	default:
+		MISSING_CASE(encoder->type);
+		return PORT_A;
 	}
 }
 
-enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
+static const struct ddi_buf_trans *
+bdw_get_buf_trans_edp(struct drm_i915_private *dev_priv, int *n_entries)
 {
-	struct intel_digital_port *dig_port;
-	enum port port;
-
-	ddi_get_encoder_port(intel_encoder, &dig_port, &port);
-
-	return port;
+	if (dev_priv->vbt.edp.low_vswing) {
+		*n_entries = ARRAY_SIZE(bdw_ddi_translations_edp);
+		return bdw_ddi_translations_edp;
+	} else {
+		*n_entries = ARRAY_SIZE(bdw_ddi_translations_dp);
+		return bdw_ddi_translations_dp;
+	}
 }
 
 static const struct ddi_buf_trans *
@@ -388,39 +377,58 @@ skl_get_buf_trans_hdmi(struct drm_i915_private *dev_priv, int *n_entries)
 	}
 }
 
+static int intel_ddi_hdmi_level(struct drm_i915_private *dev_priv, enum port port)
+{
+	int n_hdmi_entries;
+	int hdmi_level;
+	int hdmi_default_entry;
+
+	hdmi_level = dev_priv->vbt.ddi_port_info[port].hdmi_level_shift;
+
+	if (IS_BROXTON(dev_priv))
+		return hdmi_level;
+
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
+		skl_get_buf_trans_hdmi(dev_priv, &n_hdmi_entries);
+		hdmi_default_entry = 8;
+	} else if (IS_BROADWELL(dev_priv)) {
+		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
+		hdmi_default_entry = 7;
+	} else if (IS_HASWELL(dev_priv)) {
+		n_hdmi_entries = ARRAY_SIZE(hsw_ddi_translations_hdmi);
+		hdmi_default_entry = 6;
+	} else {
+		WARN(1, "ddi translation table missing\n");
+		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
+		hdmi_default_entry = 7;
+	}
+
+	/* Choose a good default if VBT is badly populated */
+	if (hdmi_level == HDMI_LEVEL_SHIFT_UNKNOWN ||
+	    hdmi_level >= n_hdmi_entries)
+		hdmi_level = hdmi_default_entry;
+
+	return hdmi_level;
+}
+
 /*
  * Starting with Haswell, DDI port buffers must be programmed with correct
- * values in advance. The buffer values are different for FDI and DP modes,
- * but the HDMI/DVI fields are shared among those. So we program the DDI
- * in either FDI or DP modes only, as HDMI connections will work with both
- * of those
+ * values in advance. This function programs the correct values for
+ * DP/eDP/FDI use cases.
  */
-void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
+void intel_prepare_dp_ddi_buffers(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 iboost_bit = 0;
-	int i, n_hdmi_entries, n_dp_entries, n_edp_entries, hdmi_default_entry,
-	    size;
-	int hdmi_level;
-	enum port port;
+	int i, n_dp_entries, n_edp_entries, size;
+	enum port port = intel_ddi_get_encoder_port(encoder);
 	const struct ddi_buf_trans *ddi_translations_fdi;
 	const struct ddi_buf_trans *ddi_translations_dp;
 	const struct ddi_buf_trans *ddi_translations_edp;
-	const struct ddi_buf_trans *ddi_translations_hdmi;
 	const struct ddi_buf_trans *ddi_translations;
 
-	port = intel_ddi_get_encoder_port(encoder);
-	hdmi_level = dev_priv->vbt.ddi_port_info[port].hdmi_level_shift;
-
-	if (IS_BROXTON(dev_priv)) {
-		if (encoder->type != INTEL_OUTPUT_HDMI)
-			return;
-
-		/* Vswing programming for HDMI */
-		bxt_ddi_vswing_sequence(dev_priv, hdmi_level, port,
-					INTEL_OUTPUT_HDMI);
+	if (IS_BROXTON(dev_priv))
 		return;
-	}
 
 	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
 		ddi_translations_fdi = NULL;
@@ -428,13 +436,10 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 				skl_get_buf_trans_dp(dev_priv, &n_dp_entries);
 		ddi_translations_edp =
 				skl_get_buf_trans_edp(dev_priv, &n_edp_entries);
-		ddi_translations_hdmi =
-				skl_get_buf_trans_hdmi(dev_priv, &n_hdmi_entries);
-		hdmi_default_entry = 8;
+
 		/* If we're boosting the current, set bit 31 of trans1 */
-		if (dev_priv->vbt.ddi_port_info[port].hdmi_boost_level ||
-		    dev_priv->vbt.ddi_port_info[port].dp_boost_level)
-			iboost_bit = 1<<31;
+		if (dev_priv->vbt.ddi_port_info[port].dp_boost_level)
+			iboost_bit = DDI_BUF_BALANCE_LEG_ENABLE;
 
 		if (WARN_ON(encoder->type == INTEL_OUTPUT_EDP &&
 			    port != PORT_A && port != PORT_E &&
@@ -443,38 +448,20 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 	} else if (IS_BROADWELL(dev_priv)) {
 		ddi_translations_fdi = bdw_ddi_translations_fdi;
 		ddi_translations_dp = bdw_ddi_translations_dp;
-
-		if (dev_priv->vbt.edp.low_vswing) {
-			ddi_translations_edp = bdw_ddi_translations_edp;
-			n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_edp);
-		} else {
-			ddi_translations_edp = bdw_ddi_translations_dp;
-			n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_dp);
-		}
-
-		ddi_translations_hdmi = bdw_ddi_translations_hdmi;
-
+		ddi_translations_edp = bdw_get_buf_trans_edp(dev_priv, &n_edp_entries);
 		n_dp_entries = ARRAY_SIZE(bdw_ddi_translations_dp);
-		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
-		hdmi_default_entry = 7;
 	} else if (IS_HASWELL(dev_priv)) {
 		ddi_translations_fdi = hsw_ddi_translations_fdi;
 		ddi_translations_dp = hsw_ddi_translations_dp;
 		ddi_translations_edp = hsw_ddi_translations_dp;
-		ddi_translations_hdmi = hsw_ddi_translations_hdmi;
 		n_dp_entries = n_edp_entries = ARRAY_SIZE(hsw_ddi_translations_dp);
-		n_hdmi_entries = ARRAY_SIZE(hsw_ddi_translations_hdmi);
-		hdmi_default_entry = 6;
 	} else {
 		WARN(1, "ddi translation table missing\n");
 		ddi_translations_edp = bdw_ddi_translations_dp;
 		ddi_translations_fdi = bdw_ddi_translations_fdi;
 		ddi_translations_dp = bdw_ddi_translations_dp;
-		ddi_translations_hdmi = bdw_ddi_translations_hdmi;
 		n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_edp);
 		n_dp_entries = ARRAY_SIZE(bdw_ddi_translations_dp);
-		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
-		hdmi_default_entry = 7;
 	}
 
 	switch (encoder->type) {
@@ -483,7 +470,6 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 		size = n_edp_entries;
 		break;
 	case INTEL_OUTPUT_DP:
-	case INTEL_OUTPUT_HDMI:
 		ddi_translations = ddi_translations_dp;
 		size = n_dp_entries;
 		break;
@@ -501,19 +487,48 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 		I915_WRITE(DDI_BUF_TRANS_HI(port, i),
 			   ddi_translations[i].trans2);
 	}
+}
+
+/*
+ * Starting with Haswell, DDI port buffers must be programmed with correct
+ * values in advance. This function programs the correct values for
+ * HDMI/DVI use cases.
+ */
+static void intel_prepare_hdmi_ddi_buffers(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	u32 iboost_bit = 0;
+	int n_hdmi_entries, hdmi_level;
+	enum port port = intel_ddi_get_encoder_port(encoder);
+	const struct ddi_buf_trans *ddi_translations_hdmi;
 
-	if (encoder->type != INTEL_OUTPUT_HDMI)
+	if (IS_BROXTON(dev_priv))
 		return;
 
-	/* Choose a good default if VBT is badly populated */
-	if (hdmi_level == HDMI_LEVEL_SHIFT_UNKNOWN ||
-	    hdmi_level >= n_hdmi_entries)
-		hdmi_level = hdmi_default_entry;
+	hdmi_level = intel_ddi_hdmi_level(dev_priv, port);
+
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
+		ddi_translations_hdmi = skl_get_buf_trans_hdmi(dev_priv, &n_hdmi_entries);
+
+		/* If we're boosting the current, set bit 31 of trans1 */
+		if (dev_priv->vbt.ddi_port_info[port].hdmi_boost_level)
+			iboost_bit = DDI_BUF_BALANCE_LEG_ENABLE;
+	} else if (IS_BROADWELL(dev_priv)) {
+		ddi_translations_hdmi = bdw_ddi_translations_hdmi;
+		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
+	} else if (IS_HASWELL(dev_priv)) {
+		ddi_translations_hdmi = hsw_ddi_translations_hdmi;
+		n_hdmi_entries = ARRAY_SIZE(hsw_ddi_translations_hdmi);
+	} else {
+		WARN(1, "ddi translation table missing\n");
+		ddi_translations_hdmi = bdw_ddi_translations_hdmi;
+		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
+	}
 
 	/* Entry 9 is for HDMI: */
-	I915_WRITE(DDI_BUF_TRANS_LO(port, i),
+	I915_WRITE(DDI_BUF_TRANS_LO(port, 9),
 		   ddi_translations_hdmi[hdmi_level].trans1 | iboost_bit);
-	I915_WRITE(DDI_BUF_TRANS_HI(port, i),
+	I915_WRITE(DDI_BUF_TRANS_HI(port, 9),
 		   ddi_translations_hdmi[hdmi_level].trans2);
 }
 
@@ -550,7 +565,7 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		WARN_ON(encoder->type != INTEL_OUTPUT_ANALOG);
-		intel_prepare_ddi_buffer(encoder);
+		intel_prepare_dp_ddi_buffers(encoder);
 	}
 
 	/* Set the FDI_RX_MISC pwrdn lanes and the 2 workarounds listed at the
@@ -1111,7 +1126,6 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *intel_encoder = intel_ddi_get_crtc_encoder(crtc);
-	struct drm_encoder *encoder = &intel_encoder->base;
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	enum pipe pipe = intel_crtc->pipe;
@@ -1177,29 +1191,15 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 			temp |= TRANS_DDI_MODE_SELECT_HDMI;
 		else
 			temp |= TRANS_DDI_MODE_SELECT_DVI;
-
 	} else if (type == INTEL_OUTPUT_ANALOG) {
 		temp |= TRANS_DDI_MODE_SELECT_FDI;
 		temp |= (intel_crtc->config->fdi_lanes - 1) << 1;
-
 	} else if (type == INTEL_OUTPUT_DP ||
 		   type == INTEL_OUTPUT_EDP) {
-		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-
-		if (intel_dp->is_mst) {
-			temp |= TRANS_DDI_MODE_SELECT_DP_MST;
-		} else
-			temp |= TRANS_DDI_MODE_SELECT_DP_SST;
-
+		temp |= TRANS_DDI_MODE_SELECT_DP_SST;
 		temp |= DDI_PORT_WIDTH(intel_crtc->config->lane_count);
 	} else if (type == INTEL_OUTPUT_DP_MST) {
-		struct intel_dp *intel_dp = &enc_to_mst(encoder)->primary->dp;
-
-		if (intel_dp->is_mst) {
-			temp |= TRANS_DDI_MODE_SELECT_DP_MST;
-		} else
-			temp |= TRANS_DDI_MODE_SELECT_DP_SST;
-
+		temp |= TRANS_DDI_MODE_SELECT_DP_MST;
 		temp |= DDI_PORT_WIDTH(intel_crtc->config->lane_count);
 	} else {
 		WARN(1, "Invalid encoder type %d for pipe %c\n",
@@ -1379,14 +1379,30 @@ void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc)
 			   TRANS_CLK_SEL_DISABLED);
 }
 
-static void skl_ddi_set_iboost(struct drm_i915_private *dev_priv,
-			       u32 level, enum port port, int type)
+static void _skl_ddi_set_iboost(struct drm_i915_private *dev_priv,
+				enum port port, uint8_t iboost)
 {
+	u32 tmp;
+
+	tmp = I915_READ(DISPIO_CR_TX_BMU_CR0);
+	tmp &= ~(BALANCE_LEG_MASK(port) | BALANCE_LEG_DISABLE(port));
+	if (iboost)
+		tmp |= iboost << BALANCE_LEG_SHIFT(port);
+	else
+		tmp |= BALANCE_LEG_DISABLE(port);
+	I915_WRITE(DISPIO_CR_TX_BMU_CR0, tmp);
+}
+
+static void skl_ddi_set_iboost(struct intel_encoder *encoder, u32 level)
+{
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(&encoder->base);
+	struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
+	enum port port = intel_dig_port->port;
+	int type = encoder->type;
 	const struct ddi_buf_trans *ddi_translations;
 	uint8_t iboost;
 	uint8_t dp_iboost, hdmi_iboost;
 	int n_entries;
-	u32 reg;
 
 	/* VBT may override standard boost values */
 	dp_iboost = dev_priv->vbt.ddi_port_info[port].dp_boost_level;
@@ -1428,16 +1444,10 @@ static void skl_ddi_set_iboost(struct drm_i915_private *dev_priv,
 		return;
 	}
 
-	reg = I915_READ(DISPIO_CR_TX_BMU_CR0);
-	reg &= ~BALANCE_LEG_MASK(port);
-	reg &= ~(1 << (BALANCE_LEG_DISABLE_SHIFT + port));
+	_skl_ddi_set_iboost(dev_priv, port, iboost);
 
-	if (iboost)
-		reg |= iboost << BALANCE_LEG_SHIFT(port);
-	else
-		reg |= 1 << (BALANCE_LEG_DISABLE_SHIFT + port);
-
-	I915_WRITE(DISPIO_CR_TX_BMU_CR0, reg);
+	if (port == PORT_A && intel_dig_port->max_lanes == 4)
+		_skl_ddi_set_iboost(dev_priv, PORT_E, iboost);
 }
 
 static void bxt_ddi_vswing_sequence(struct drm_i915_private *dev_priv,
@@ -1568,7 +1578,7 @@ uint32_t ddi_signal_levels(struct intel_dp *intel_dp)
 	level = translate_signal_level(signal_levels);
 
 	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
-		skl_ddi_set_iboost(dev_priv, level, port, encoder->type);
+		skl_ddi_set_iboost(encoder, level);
 	else if (IS_BROXTON(dev_priv))
 		bxt_ddi_vswing_sequence(dev_priv, level, port, encoder->type);
 
@@ -1615,8 +1625,6 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 		intel_dp_dual_mode_set_tmds_output(intel_hdmi, true);
 	}
 
-	intel_prepare_ddi_buffer(intel_encoder);
-
 	if (type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 		intel_edp_panel_on(intel_dp);
@@ -1627,6 +1635,8 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 	if (type == INTEL_OUTPUT_DP || type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
+		intel_prepare_dp_ddi_buffers(intel_encoder);
+
 		intel_dp_set_link_params(intel_dp, crtc->config);
 
 		intel_ddi_init_dp_buf_reg(intel_encoder);
@@ -1637,6 +1647,15 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 			intel_dp_stop_link_train(intel_dp);
 	} else if (type == INTEL_OUTPUT_HDMI) {
 		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+		int level = intel_ddi_hdmi_level(dev_priv, port);
+
+		intel_prepare_hdmi_ddi_buffers(intel_encoder);
+
+		if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
+			skl_ddi_set_iboost(intel_encoder, level);
+		else if (IS_BROXTON(dev_priv))
+			bxt_ddi_vswing_sequence(dev_priv, level, port,
+						INTEL_OUTPUT_HDMI);
 
 		intel_hdmi->set_infoframes(encoder,
 					   crtc->config->has_hdmi_sink,
@@ -2105,7 +2124,7 @@ void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp)
 
 	val = DP_TP_CTL_ENABLE |
 	      DP_TP_CTL_LINK_TRAIN_PAT1 | DP_TP_CTL_SCRAMBLE_DISABLE;
-	if (intel_dp->is_mst)
+	if (intel_dp->link_mst)
 		val |= DP_TP_CTL_MODE_MST;
 	else {
 		val |= DP_TP_CTL_MODE_SST;

drivers/gpu/drm/i915/intel_display.c

@@ -34,6 +34,7 @@
 #include <drm/drm_edid.h>
 #include <drm/drmP.h>
 #include "intel_drv.h"
+#include "intel_frontbuffer.h"
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 #include "i915_gem_dmabuf.h"
@@ -2465,9 +2466,8 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 		return false;
 	}
 
-	obj->tiling_mode = plane_config->tiling;
-	if (obj->tiling_mode == I915_TILING_X)
-		obj->stride = fb->pitches[0];
+	if (plane_config->tiling == I915_TILING_X)
+		obj->tiling_and_stride = fb->pitches[0] | I915_TILING_X;
 
 	mode_cmd.pixel_format = fb->pixel_format;
 	mode_cmd.width = fb->width;
@@ -2488,7 +2488,7 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 	return true;
 
 out_unref_obj:
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	mutex_unlock(&dev->struct_mutex);
 	return false;
 }
@@ -2593,14 +2593,15 @@ valid_fb:
 	intel_state->base.dst.y2 = plane_state->crtc_y + plane_state->crtc_h;
 
 	obj = intel_fb_obj(fb);
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		dev_priv->preserve_bios_swizzle = true;
 
 	drm_framebuffer_reference(fb);
 	primary->fb = primary->state->fb = fb;
 	primary->crtc = primary->state->crtc = &intel_crtc->base;
 	intel_crtc->base.state->plane_mask |= (1 << drm_plane_index(primary));
-	obj->frontbuffer_bits |= to_intel_plane(primary)->frontbuffer_bit;
+	atomic_or(to_intel_plane(primary)->frontbuffer_bit,
+		  &obj->frontbuffer_bits);
 }
 
 static void i9xx_update_primary_plane(struct drm_plane *primary,
@@ -2670,8 +2671,7 @@ static void i9xx_update_primary_plane(struct drm_plane *primary,
 		BUG();
 	}
 
-	if (INTEL_INFO(dev)->gen >= 4 &&
-	    obj->tiling_mode != I915_TILING_NONE)
+	if (INTEL_INFO(dev)->gen >= 4 && i915_gem_object_is_tiled(obj))
 		dspcntr |= DISPPLANE_TILED;
 
 	if (IS_G4X(dev))
@@ -2780,7 +2780,7 @@ static void ironlake_update_primary_plane(struct drm_plane *primary,
 		BUG();
 	}
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		dspcntr |= DISPPLANE_TILED;
 
 	if (!IS_HASWELL(dev) && !IS_BROADWELL(dev))
@@ -4564,12 +4564,11 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state)
 	struct drm_atomic_state *old_state = old_crtc_state->base.state;
 	struct intel_crtc_state *pipe_config =
 		to_intel_crtc_state(crtc->base.state);
-	struct drm_device *dev = crtc->base.dev;
 	struct drm_plane *primary = crtc->base.primary;
 	struct drm_plane_state *old_pri_state =
 		drm_atomic_get_existing_plane_state(old_state, primary);
 
-	intel_frontbuffer_flip(dev, pipe_config->fb_bits);
+	intel_frontbuffer_flip(to_i915(crtc->base.dev), pipe_config->fb_bits);
 
 	crtc->wm.cxsr_allowed = true;
 
@@ -4692,7 +4691,7 @@ static void intel_crtc_disable_planes(struct drm_crtc *crtc, unsigned plane_mask
 	 * to compute the mask of flip planes precisely. For the time being
 	 * consider this a flip to a NULL plane.
 	 */
-	intel_frontbuffer_flip(dev, INTEL_FRONTBUFFER_ALL_MASK(pipe));
+	intel_frontbuffer_flip(to_i915(dev), INTEL_FRONTBUFFER_ALL_MASK(pipe));
 }
 
 static void ironlake_crtc_enable(struct drm_crtc *crtc)
@@ -10434,7 +10433,7 @@ intel_framebuffer_create_for_mode(struct drm_device *dev,
 
 	fb = intel_framebuffer_create(dev, &mode_cmd, obj);
 	if (IS_ERR(fb))
-		drm_gem_object_unreference_unlocked(&obj->base);
+		i915_gem_object_put_unlocked(obj);
 
 	return fb;
 }
@@ -10945,13 +10944,13 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 
 	mutex_lock(&dev->struct_mutex);
 	intel_unpin_fb_obj(work->old_fb, primary->state->rotation);
-	drm_gem_object_unreference(&work->pending_flip_obj->base);
-
-	if (work->flip_queued_req)
-		i915_gem_request_assign(&work->flip_queued_req, NULL);
+	i915_gem_object_put(work->pending_flip_obj);
 	mutex_unlock(&dev->struct_mutex);
 
-	intel_frontbuffer_flip_complete(dev, to_intel_plane(primary)->frontbuffer_bit);
+	i915_gem_request_put(work->flip_queued_req);
+
+	intel_frontbuffer_flip_complete(to_i915(dev),
+					to_intel_plane(primary)->frontbuffer_bit);
 	intel_fbc_post_update(crtc);
 	drm_framebuffer_unreference(work->old_fb);
 
@@ -11116,7 +11115,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
 	int ret;
@@ -11132,13 +11131,13 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
 	else
 		flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
-	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | flip_mask);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_emit(engine, MI_DISPLAY_FLIP |
+	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | flip_mask);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(engine, fb->pitches[0]);
-	intel_ring_emit(engine, intel_crtc->flip_work->gtt_offset);
-	intel_ring_emit(engine, 0); /* aux display base address, unused */
+	intel_ring_emit(ring, fb->pitches[0]);
+	intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset);
+	intel_ring_emit(ring, 0); /* aux display base address, unused */
 
 	return 0;
 }
@@ -11150,7 +11149,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
 	int ret;
@@ -11163,13 +11162,13 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
 	else
 		flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
-	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | flip_mask);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_emit(engine, MI_DISPLAY_FLIP_I915 |
+	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | flip_mask);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(engine, fb->pitches[0]);
-	intel_ring_emit(engine, intel_crtc->flip_work->gtt_offset);
-	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_emit(ring, fb->pitches[0]);
+	intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset);
+	intel_ring_emit(ring, MI_NOOP);
 
 	return 0;
 }
@@ -11181,7 +11180,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t pf, pipesrc;
@@ -11195,11 +11194,11 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	 * Display Registers (which do not change across a page-flip)
 	 * so we need only reprogram the base address.
 	 */
-	intel_ring_emit(engine, MI_DISPLAY_FLIP |
+	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(engine, fb->pitches[0]);
-	intel_ring_emit(engine, intel_crtc->flip_work->gtt_offset |
-			obj->tiling_mode);
+	intel_ring_emit(ring, fb->pitches[0]);
+	intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset |
+			i915_gem_object_get_tiling(obj));
 
 	/* XXX Enabling the panel-fitter across page-flip is so far
 	 * untested on non-native modes, so ignore it for now.
@@ -11207,7 +11206,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	 */
 	pf = 0;
 	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
-	intel_ring_emit(engine, pf | pipesrc);
+	intel_ring_emit(ring, pf | pipesrc);
 
 	return 0;
 }
@@ -11219,7 +11218,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t pf, pipesrc;
@@ -11229,10 +11228,10 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_DISPLAY_FLIP |
+	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	intel_ring_emit(engine, fb->pitches[0] | obj->tiling_mode);
-	intel_ring_emit(engine, intel_crtc->flip_work->gtt_offset);
+	intel_ring_emit(ring, fb->pitches[0] | i915_gem_object_get_tiling(obj));
+	intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset);
 
 	/* Contrary to the suggestions in the documentation,
 	 * "Enable Panel Fitter" does not seem to be required when page
@@ -11242,7 +11241,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	 */
 	pf = 0;
 	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
-	intel_ring_emit(engine, pf | pipesrc);
+	intel_ring_emit(ring, pf | pipesrc);
 
 	return 0;
 }
@@ -11254,7 +11253,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 				 struct drm_i915_gem_request *req,
 				 uint32_t flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t plane_bit = 0;
 	int len, ret;
@@ -11275,7 +11274,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	}
 
 	len = 4;
-	if (engine->id == RCS) {
+	if (req->engine->id == RCS) {
 		len += 6;
 		/*
 		 * On Gen 8, SRM is now taking an extra dword to accommodate
@@ -11313,30 +11312,30 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	 * for the RCS also doesn't appear to drop events. Setting the DERRMR
 	 * to zero does lead to lockups within MI_DISPLAY_FLIP.
 	 */
-	if (engine->id == RCS) {
-		intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit_reg(engine, DERRMR);
-		intel_ring_emit(engine, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
+	if (req->engine->id == RCS) {
+		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit_reg(ring, DERRMR);
+		intel_ring_emit(ring, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
 					  DERRMR_PIPEB_PRI_FLIP_DONE |
 					  DERRMR_PIPEC_PRI_FLIP_DONE));
 		if (IS_GEN8(dev))
-			intel_ring_emit(engine, MI_STORE_REGISTER_MEM_GEN8 |
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM_GEN8 |
 					      MI_SRM_LRM_GLOBAL_GTT);
 		else
-			intel_ring_emit(engine, MI_STORE_REGISTER_MEM |
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM |
 					      MI_SRM_LRM_GLOBAL_GTT);
-		intel_ring_emit_reg(engine, DERRMR);
-		intel_ring_emit(engine, engine->scratch.gtt_offset + 256);
+		intel_ring_emit_reg(ring, DERRMR);
+		intel_ring_emit(ring, req->engine->scratch.gtt_offset + 256);
 		if (IS_GEN8(dev)) {
-			intel_ring_emit(engine, 0);
-			intel_ring_emit(engine, MI_NOOP);
+			intel_ring_emit(ring, 0);
+			intel_ring_emit(ring, MI_NOOP);
 		}
 	}
 
-	intel_ring_emit(engine, MI_DISPLAY_FLIP_I915 | plane_bit);
-	intel_ring_emit(engine, (fb->pitches[0] | obj->tiling_mode));
-	intel_ring_emit(engine, intel_crtc->flip_work->gtt_offset);
-	intel_ring_emit(engine, (MI_NOOP));
+	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit);
+	intel_ring_emit(ring, fb->pitches[0] | i915_gem_object_get_tiling(obj));
+	intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset);
+	intel_ring_emit(ring, (MI_NOOP));
 
 	return 0;
 }
@@ -11371,7 +11370,8 @@ static bool use_mmio_flip(struct intel_engine_cs *engine,
 	if (resv && !reservation_object_test_signaled_rcu(resv, false))
 		return true;
 
-	return engine != i915_gem_request_get_engine(obj->last_write_req);
+	return engine != i915_gem_active_get_engine(&obj->last_write,
+						    &obj->base.dev->struct_mutex);
 }
 
 static void skl_do_mmio_flip(struct intel_crtc *intel_crtc,
@@ -11440,7 +11440,7 @@ static void ilk_do_mmio_flip(struct intel_crtc *intel_crtc,
 
 	dspcntr = I915_READ(reg);
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		dspcntr |= DISPPLANE_TILED;
 	else
 		dspcntr &= ~DISPPLANE_TILED;
@@ -11463,9 +11463,9 @@ static void intel_mmio_flip_work_func(struct work_struct *w)
 	struct reservation_object *resv;
 
 	if (work->flip_queued_req)
-		WARN_ON(__i915_wait_request(work->flip_queued_req,
-					    false, NULL,
-					    &dev_priv->rps.mmioflips));
+		WARN_ON(i915_wait_request(work->flip_queued_req,
+					  false, NULL,
+					  NO_WAITBOOST));
 
 	/* For framebuffer backed by dmabuf, wait for fence */
 	resv = i915_gem_object_get_dmabuf_resv(obj);
@@ -11576,7 +11576,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	struct intel_flip_work *work;
 	struct intel_engine_cs *engine;
 	bool mmio_flip;
-	struct drm_i915_gem_request *request = NULL;
+	struct drm_i915_gem_request *request;
 	int ret;
 
 	/*
@@ -11642,7 +11642,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	/* Reference the objects for the scheduled work. */
 	drm_framebuffer_reference(work->old_fb);
-	drm_gem_object_reference(&obj->base);
 
 	crtc->primary->fb = fb;
 	update_state_fb(crtc->primary);
@@ -11650,7 +11649,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	intel_fbc_pre_update(intel_crtc, intel_crtc->config,
 			     to_intel_plane_state(primary->state));
 
-	work->pending_flip_obj = obj;
+	work->pending_flip_obj = i915_gem_object_get(obj);
 
 	ret = i915_mutex_lock_interruptible(dev);
 	if (ret)
@@ -11669,13 +11668,15 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
 		engine = &dev_priv->engine[BCS];
-		if (obj->tiling_mode != intel_fb_obj(work->old_fb)->tiling_mode)
+		if (i915_gem_object_get_tiling(obj) !=
+		    i915_gem_object_get_tiling(intel_fb_obj(work->old_fb)))
 			/* vlv: DISPLAY_FLIP fails to change tiling */
 			engine = NULL;
 	} else if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
 		engine = &dev_priv->engine[BCS];
 	} else if (INTEL_INFO(dev)->gen >= 7) {
-		engine = i915_gem_request_get_engine(obj->last_write_req);
+		engine = i915_gem_active_get_engine(&obj->last_write,
+						    &obj->base.dev->struct_mutex);
 		if (engine == NULL || engine->id != RCS)
 			engine = &dev_priv->engine[BCS];
 	} else {
@@ -11684,22 +11685,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	mmio_flip = use_mmio_flip(engine, obj);
 
-	/* When using CS flips, we want to emit semaphores between rings.
-	 * However, when using mmio flips we will create a task to do the
-	 * synchronisation, so all we want here is to pin the framebuffer
-	 * into the display plane and skip any waits.
-	 */
-	if (!mmio_flip) {
-		ret = i915_gem_object_sync(obj, engine, &request);
-		if (!ret && !request) {
-			request = i915_gem_request_alloc(engine, NULL);
-			ret = PTR_ERR_OR_ZERO(request);
-		}
-
-		if (ret)
-			goto cleanup_pending;
-	}
-
 	ret = intel_pin_and_fence_fb_obj(fb, primary->state->rotation);
 	if (ret)
 		goto cleanup_pending;
@@ -11712,19 +11697,28 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (mmio_flip) {
 		INIT_WORK(&work->mmio_work, intel_mmio_flip_work_func);
 
-		i915_gem_request_assign(&work->flip_queued_req,
-					obj->last_write_req);
-
+		work->flip_queued_req = i915_gem_active_get(&obj->last_write,
+							    &obj->base.dev->struct_mutex);
 		schedule_work(&work->mmio_work);
 	} else {
-		i915_gem_request_assign(&work->flip_queued_req, request);
+		request = i915_gem_request_alloc(engine, engine->last_context);
+		if (IS_ERR(request)) {
+			ret = PTR_ERR(request);
+			goto cleanup_unpin;
+		}
+
+		ret = i915_gem_object_sync(obj, request);
+		if (ret)
+			goto cleanup_request;
+
 		ret = dev_priv->display.queue_flip(dev, crtc, fb, obj, request,
 						   page_flip_flags);
 		if (ret)
-			goto cleanup_unpin;
+			goto cleanup_request;
 
 		intel_mark_page_flip_active(intel_crtc, work);
 
+		work->flip_queued_req = i915_gem_request_get(request);
 		i915_add_request_no_flush(request);
 	}
 
@@ -11732,25 +11726,25 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 			  to_intel_plane(primary)->frontbuffer_bit);
 	mutex_unlock(&dev->struct_mutex);
 
-	intel_frontbuffer_flip_prepare(dev,
+	intel_frontbuffer_flip_prepare(to_i915(dev),
 				       to_intel_plane(primary)->frontbuffer_bit);
 
 	trace_i915_flip_request(intel_crtc->plane, obj);
 
 	return 0;
 
+cleanup_request:
+	i915_add_request_no_flush(request);
 cleanup_unpin:
 	intel_unpin_fb_obj(fb, crtc->primary->state->rotation);
 cleanup_pending:
-	if (!IS_ERR_OR_NULL(request))
-		i915_add_request_no_flush(request);
 	atomic_dec(&intel_crtc->unpin_work_count);
 	mutex_unlock(&dev->struct_mutex);
 cleanup:
 	crtc->primary->fb = old_fb;
 	update_state_fb(crtc->primary);
 
-	drm_gem_object_unreference_unlocked(&obj->base);
+	i915_gem_object_put_unlocked(obj);
 	drm_framebuffer_unreference(work->old_fb);
 
 	spin_lock_irq(&dev->event_lock);
@@ -12298,6 +12292,7 @@ static bool check_digital_port_conflicts(struct drm_atomic_state *state)
 	struct drm_device *dev = state->dev;
 	struct drm_connector *connector;
 	unsigned int used_ports = 0;
+	unsigned int used_mst_ports = 0;
 
 	/*
 	 * Walk the connector list instead of the encoder
@@ -12334,11 +12329,20 @@ static bool check_digital_port_conflicts(struct drm_atomic_state *state)
 				return false;
 
 			used_ports |= port_mask;
+			break;
+		case INTEL_OUTPUT_DP_MST:
+			used_mst_ports |=
+				1 << enc_to_mst(&encoder->base)->primary->port;
+			break;
 		default:
 			break;
 		}
 	}
 
+	/* can't mix MST and SST/HDMI on the same port */
+	if (used_ports & used_mst_ports)
+		return false;
+
 	return true;
 }
 
@@ -13506,8 +13510,8 @@ static int intel_atomic_prepare_commit(struct drm_device *dev,
 			if (!intel_plane_state->wait_req)
 				continue;
 
-			ret = __i915_wait_request(intel_plane_state->wait_req,
-						  true, NULL, NULL);
+			ret = i915_wait_request(intel_plane_state->wait_req,
+						true, NULL, NULL);
 			if (ret) {
 				/* Any hang should be swallowed by the wait */
 				WARN_ON(ret == -EIO);
@@ -13619,8 +13623,8 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state)
 		if (!intel_plane_state->wait_req)
 			continue;
 
-		ret = __i915_wait_request(intel_plane_state->wait_req,
-					  true, NULL, NULL);
+		ret = i915_wait_request(intel_plane_state->wait_req,
+					true, NULL, NULL);
 		/* EIO should be eaten, and we can't get interrupted in the
 		 * worker, and blocking commits have waited already. */
 		WARN_ON(ret);
@@ -13797,19 +13801,12 @@ static void intel_atomic_track_fbs(struct drm_atomic_state *state)
 {
 	struct drm_plane_state *old_plane_state;
 	struct drm_plane *plane;
-	struct drm_i915_gem_object *obj, *old_obj;
-	struct intel_plane *intel_plane;
 	int i;
 
-	mutex_lock(&state->dev->struct_mutex);
-	for_each_plane_in_state(state, plane, old_plane_state, i) {
-		obj = intel_fb_obj(plane->state->fb);
-		old_obj = intel_fb_obj(old_plane_state->fb);
-		intel_plane = to_intel_plane(plane);
-
-		i915_gem_track_fb(old_obj, obj, intel_plane->frontbuffer_bit);
-	}
-	mutex_unlock(&state->dev->struct_mutex);
+	for_each_plane_in_state(state, plane, old_plane_state, i)
+		i915_gem_track_fb(intel_fb_obj(old_plane_state->fb),
+				  intel_fb_obj(plane->state->fb),
+				  to_intel_plane(plane)->frontbuffer_bit);
 }
 
 /**
@@ -14038,11 +14035,9 @@ intel_prepare_plane_fb(struct drm_plane *plane,
 	}
 
 	if (ret == 0) {
-		struct intel_plane_state *plane_state =
-			to_intel_plane_state(new_state);
-
-		i915_gem_request_assign(&plane_state->wait_req,
-					obj->last_write_req);
+		to_intel_plane_state(new_state)->wait_req =
+			i915_gem_active_get(&obj->last_write,
+					    &obj->base.dev->struct_mutex);
 	}
 
 	return ret;
@@ -14063,6 +14058,7 @@ intel_cleanup_plane_fb(struct drm_plane *plane,
 {
 	struct drm_device *dev = plane->dev;
 	struct intel_plane_state *old_intel_state;
+	struct intel_plane_state *intel_state = to_intel_plane_state(plane->state);
 	struct drm_i915_gem_object *old_obj = intel_fb_obj(old_state->fb);
 	struct drm_i915_gem_object *obj = intel_fb_obj(plane->state->fb);
 
@@ -14075,6 +14071,7 @@ intel_cleanup_plane_fb(struct drm_plane *plane,
 	    !INTEL_INFO(dev)->cursor_needs_physical))
 		intel_unpin_fb_obj(old_state->fb, old_state->rotation);
 
+	i915_gem_request_assign(&intel_state->wait_req, NULL);
 	i915_gem_request_assign(&old_intel_state->wait_req, NULL);
 }
 
@@ -14831,7 +14828,7 @@ static void intel_user_framebuffer_destroy(struct drm_framebuffer *fb)
 	drm_framebuffer_cleanup(fb);
 	mutex_lock(&dev->struct_mutex);
 	WARN_ON(!intel_fb->obj->framebuffer_references--);
-	drm_gem_object_unreference(&intel_fb->obj->base);
+	i915_gem_object_put(intel_fb->obj);
 	mutex_unlock(&dev->struct_mutex);
 	kfree(intel_fb);
 }
@@ -14920,15 +14917,15 @@ static int intel_framebuffer_init(struct drm_device *dev,
 	if (mode_cmd->flags & DRM_MODE_FB_MODIFIERS) {
 		/* Enforce that fb modifier and tiling mode match, but only for
 		 * X-tiled. This is needed for FBC. */
-		if (!!(obj->tiling_mode == I915_TILING_X) !=
+		if (!!(i915_gem_object_get_tiling(obj) == I915_TILING_X) !=
 		    !!(mode_cmd->modifier[0] == I915_FORMAT_MOD_X_TILED)) {
 			DRM_DEBUG("tiling_mode doesn't match fb modifier\n");
 			return -EINVAL;
 		}
 	} else {
-		if (obj->tiling_mode == I915_TILING_X)
+		if (i915_gem_object_get_tiling(obj) == I915_TILING_X)
 			mode_cmd->modifier[0] = I915_FORMAT_MOD_X_TILED;
-		else if (obj->tiling_mode == I915_TILING_Y) {
+		else if (i915_gem_object_get_tiling(obj) == I915_TILING_Y) {
 			DRM_DEBUG("No Y tiling for legacy addfb\n");
 			return -EINVAL;
 		}
@@ -14972,9 +14969,10 @@ static int intel_framebuffer_init(struct drm_device *dev,
 	}
 
 	if (mode_cmd->modifier[0] == I915_FORMAT_MOD_X_TILED &&
-	    mode_cmd->pitches[0] != obj->stride) {
+	    mode_cmd->pitches[0] != i915_gem_object_get_stride(obj)) {
 		DRM_DEBUG("pitch (%d) must match tiling stride (%d)\n",
-			  mode_cmd->pitches[0], obj->stride);
+			  mode_cmd->pitches[0],
+			  i915_gem_object_get_stride(obj));
 		return -EINVAL;
 	}
 
@@ -15068,13 +15066,13 @@ intel_user_framebuffer_create(struct drm_device *dev,
 	struct drm_i915_gem_object *obj;
 	struct drm_mode_fb_cmd2 mode_cmd = *user_mode_cmd;
 
-	obj = to_intel_bo(drm_gem_object_lookup(filp, mode_cmd.handles[0]));
-	if (&obj->base == NULL)
+	obj = i915_gem_object_lookup(filp, mode_cmd.handles[0]);
+	if (!obj)
 		return ERR_PTR(-ENOENT);
 
 	fb = intel_framebuffer_create(dev, &mode_cmd, obj);
 	if (IS_ERR(fb))
-		drm_gem_object_unreference_unlocked(&obj->base);
+		i915_gem_object_put_unlocked(obj);
 
 	return fb;
 }
@@ -15482,7 +15480,6 @@ void intel_modeset_init_hw(struct drm_device *dev)
 	dev_priv->atomic_cdclk_freq = dev_priv->cdclk_freq;
 
 	intel_init_clock_gating(dev);
-	intel_enable_gt_powersave(dev_priv);
 }
 
 /*

drivers/gpu/drm/i915/intel_dp.c

@@ -1041,10 +1041,10 @@ intel_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 		if (WARN_ON(txsize > 20))
 			return -E2BIG;
 
+		WARN_ON(!msg->buffer != !msg->size);
+
 		if (msg->buffer)
 			memcpy(txbuf + HEADER_SIZE, msg->buffer, msg->size);
-		else
-			WARN_ON(msg->size);
 
 		ret = intel_dp_aux_ch(intel_dp, txbuf, txsize, rxbuf, rxsize);
 		if (ret > 0) {
@@ -1447,7 +1447,7 @@ intel_dp_max_link_rate(struct intel_dp *intel_dp)
 	if (WARN_ON(len <= 0))
 		return 162000;
 
-	return rates[rate_to_index(0, rates) - 1];
+	return rates[len - 1];
 }
 
 int intel_dp_rate_select(struct intel_dp *intel_dp, int rate)
@@ -1651,6 +1651,7 @@ void intel_dp_set_link_params(struct intel_dp *intel_dp,
 {
 	intel_dp->link_rate = pipe_config->port_clock;
 	intel_dp->lane_count = pipe_config->lane_count;
+	intel_dp->link_mst = intel_crtc_has_type(pipe_config, INTEL_OUTPUT_DP_MST);
 }
 
 static void intel_dp_prepare(struct intel_encoder *encoder)
@@ -3395,84 +3396,67 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 }
 
 static bool
-intel_dp_get_dpcd(struct intel_dp *intel_dp)
+intel_dp_read_dpcd(struct intel_dp *intel_dp)
 {
-	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = dig_port->base.base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-
 	if (drm_dp_dpcd_read(&intel_dp->aux, 0x000, intel_dp->dpcd,
 			     sizeof(intel_dp->dpcd)) < 0)
 		return false; /* aux transfer failed */
 
 	DRM_DEBUG_KMS("DPCD: %*ph\n", (int) sizeof(intel_dp->dpcd), intel_dp->dpcd);
 
-	if (intel_dp->dpcd[DP_DPCD_REV] == 0)
-		return false; /* DPCD not present */
+	return intel_dp->dpcd[DP_DPCD_REV] != 0;
+}
 
-	if (drm_dp_dpcd_read(&intel_dp->aux, DP_SINK_COUNT,
-			     &intel_dp->sink_count, 1) < 0)
-		return false;
+static bool
+intel_edp_init_dpcd(struct intel_dp *intel_dp)
+{
+	struct drm_i915_private *dev_priv =
+		to_i915(dp_to_dig_port(intel_dp)->base.base.dev);
 
-	/*
-	 * Sink count can change between short pulse hpd hence
-	 * a member variable in intel_dp will track any changes
-	 * between short pulse interrupts.
-	 */
-	intel_dp->sink_count = DP_GET_SINK_COUNT(intel_dp->sink_count);
+	/* this function is meant to be called only once */
+	WARN_ON(intel_dp->dpcd[DP_DPCD_REV] != 0);
 
-	/*
-	 * SINK_COUNT == 0 and DOWNSTREAM_PORT_PRESENT == 1 implies that
-	 * a dongle is present but no display. Unless we require to know
-	 * if a dongle is present or not, we don't need to update
-	 * downstream port information. So, an early return here saves
-	 * time from performing other operations which are not required.
-	 */
-	if (!is_edp(intel_dp) && !intel_dp->sink_count)
+	if (!intel_dp_read_dpcd(intel_dp))
 		return false;
 
-	/* Check if the panel supports PSR */
-	memset(intel_dp->psr_dpcd, 0, sizeof(intel_dp->psr_dpcd));
-	if (is_edp(intel_dp)) {
-		drm_dp_dpcd_read(&intel_dp->aux, DP_PSR_SUPPORT,
-				 intel_dp->psr_dpcd,
-				 sizeof(intel_dp->psr_dpcd));
-		if (intel_dp->psr_dpcd[0] & DP_PSR_IS_SUPPORTED) {
-			dev_priv->psr.sink_support = true;
-			DRM_DEBUG_KMS("Detected EDP PSR Panel.\n");
-		}
-
-		if (INTEL_INFO(dev)->gen >= 9 &&
-			(intel_dp->psr_dpcd[0] & DP_PSR2_IS_SUPPORTED)) {
-			uint8_t frame_sync_cap;
-
-			dev_priv->psr.sink_support = true;
-			drm_dp_dpcd_read(&intel_dp->aux,
-					 DP_SINK_DEVICE_AUX_FRAME_SYNC_CAP,
-					 &frame_sync_cap, 1);
-			dev_priv->psr.aux_frame_sync = frame_sync_cap ? true : false;
-			/* PSR2 needs frame sync as well */
-			dev_priv->psr.psr2_support = dev_priv->psr.aux_frame_sync;
-			DRM_DEBUG_KMS("PSR2 %s on sink",
-				dev_priv->psr.psr2_support ? "supported" : "not supported");
-		}
+	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
+		dev_priv->no_aux_handshake = intel_dp->dpcd[DP_MAX_DOWNSPREAD] &
+			DP_NO_AUX_HANDSHAKE_LINK_TRAINING;
 
-		/* Read the eDP Display control capabilities registers */
-		memset(intel_dp->edp_dpcd, 0, sizeof(intel_dp->edp_dpcd));
-		if ((intel_dp->dpcd[DP_EDP_CONFIGURATION_CAP] & DP_DPCD_DISPLAY_CONTROL_CAPABLE) &&
-				(drm_dp_dpcd_read(&intel_dp->aux, DP_EDP_DPCD_REV,
-						intel_dp->edp_dpcd, sizeof(intel_dp->edp_dpcd)) ==
-								sizeof(intel_dp->edp_dpcd)))
-			DRM_DEBUG_KMS("EDP DPCD : %*ph\n", (int) sizeof(intel_dp->edp_dpcd),
-					intel_dp->edp_dpcd);
-	}
-
-	DRM_DEBUG_KMS("Display Port TPS3 support: source %s, sink %s\n",
-		      yesno(intel_dp_source_supports_hbr2(intel_dp)),
-		      yesno(drm_dp_tps3_supported(intel_dp->dpcd)));
+	/* Check if the panel supports PSR */
+	drm_dp_dpcd_read(&intel_dp->aux, DP_PSR_SUPPORT,
+			 intel_dp->psr_dpcd,
+			 sizeof(intel_dp->psr_dpcd));
+	if (intel_dp->psr_dpcd[0] & DP_PSR_IS_SUPPORTED) {
+		dev_priv->psr.sink_support = true;
+		DRM_DEBUG_KMS("Detected EDP PSR Panel.\n");
+	}
+
+	if (INTEL_GEN(dev_priv) >= 9 &&
+	    (intel_dp->psr_dpcd[0] & DP_PSR2_IS_SUPPORTED)) {
+		uint8_t frame_sync_cap;
+
+		dev_priv->psr.sink_support = true;
+		drm_dp_dpcd_read(&intel_dp->aux,
+				 DP_SINK_DEVICE_AUX_FRAME_SYNC_CAP,
+				 &frame_sync_cap, 1);
+		dev_priv->psr.aux_frame_sync = frame_sync_cap ? true : false;
+		/* PSR2 needs frame sync as well */
+		dev_priv->psr.psr2_support = dev_priv->psr.aux_frame_sync;
+		DRM_DEBUG_KMS("PSR2 %s on sink",
+			      dev_priv->psr.psr2_support ? "supported" : "not supported");
+	}
+
+	/* Read the eDP Display control capabilities registers */
+	if ((intel_dp->dpcd[DP_EDP_CONFIGURATION_CAP] & DP_DPCD_DISPLAY_CONTROL_CAPABLE) &&
+	    drm_dp_dpcd_read(&intel_dp->aux, DP_EDP_DPCD_REV,
+			     intel_dp->edp_dpcd, sizeof(intel_dp->edp_dpcd) ==
+			     sizeof(intel_dp->edp_dpcd)))
+		DRM_DEBUG_KMS("EDP DPCD : %*ph\n", (int) sizeof(intel_dp->edp_dpcd),
+			      intel_dp->edp_dpcd);
 
 	/* Intermediate frequency support */
-	if (is_edp(intel_dp) && (intel_dp->edp_dpcd[0] >= 0x03)) { /* eDp v1.4 or higher */
+	if (intel_dp->edp_dpcd[0] >= 0x03) { /* eDp v1.4 or higher */
 		__le16 sink_rates[DP_MAX_SUPPORTED_RATES];
 		int i;
 
@@ -3491,7 +3475,36 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 		intel_dp->num_sink_rates = i;
 	}
 
-	intel_dp_print_rates(intel_dp);
+	return true;
+}
+
+
+static bool
+intel_dp_get_dpcd(struct intel_dp *intel_dp)
+{
+	if (!intel_dp_read_dpcd(intel_dp))
+		return false;
+
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_SINK_COUNT,
+			     &intel_dp->sink_count, 1) < 0)
+		return false;
+
+	/*
+	 * Sink count can change between short pulse hpd hence
+	 * a member variable in intel_dp will track any changes
+	 * between short pulse interrupts.
+	 */
+	intel_dp->sink_count = DP_GET_SINK_COUNT(intel_dp->sink_count);
+
+	/*
+	 * SINK_COUNT == 0 and DOWNSTREAM_PORT_PRESENT == 1 implies that
+	 * a dongle is present but no display. Unless we require to know
+	 * if a dongle is present or not, we don't need to update
+	 * downstream port information. So, an early return here saves
+	 * time from performing other operations which are not required.
+	 */
+	if (!is_edp(intel_dp) && !intel_dp->sink_count)
+		return false;
 
 	if (!(intel_dp->dpcd[DP_DOWNSTREAMPORT_PRESENT] &
 	      DP_DWN_STRM_PORT_PRESENT))
@@ -3526,7 +3539,7 @@ intel_dp_probe_oui(struct intel_dp *intel_dp)
 }
 
 static bool
-intel_dp_probe_mst(struct intel_dp *intel_dp)
+intel_dp_can_mst(struct intel_dp *intel_dp)
 {
 	u8 buf[1];
 
@@ -3539,18 +3552,30 @@ intel_dp_probe_mst(struct intel_dp *intel_dp)
 	if (intel_dp->dpcd[DP_DPCD_REV] < 0x12)
 		return false;
 
-	if (drm_dp_dpcd_read(&intel_dp->aux, DP_MSTM_CAP, buf, 1)) {
-		if (buf[0] & DP_MST_CAP) {
-			DRM_DEBUG_KMS("Sink is MST capable\n");
-			intel_dp->is_mst = true;
-		} else {
-			DRM_DEBUG_KMS("Sink is not MST capable\n");
-			intel_dp->is_mst = false;
-		}
-	}
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_MSTM_CAP, buf, 1) != 1)
+		return false;
 
-	drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr, intel_dp->is_mst);
-	return intel_dp->is_mst;
+	return buf[0] & DP_MST_CAP;
+}
+
+static void
+intel_dp_configure_mst(struct intel_dp *intel_dp)
+{
+	if (!i915.enable_dp_mst)
+		return;
+
+	if (!intel_dp->can_mst)
+		return;
+
+	intel_dp->is_mst = intel_dp_can_mst(intel_dp);
+
+	if (intel_dp->is_mst)
+		DRM_DEBUG_KMS("Sink is MST capable\n");
+	else
+		DRM_DEBUG_KMS("Sink is not MST capable\n");
+
+	drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr,
+					intel_dp->is_mst);
 }
 
 static int intel_dp_sink_crc_stop(struct intel_dp *intel_dp)
@@ -3909,7 +3934,7 @@ static bool
 intel_dp_short_pulse(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
-	u8 sink_irq_vector;
+	u8 sink_irq_vector = 0;
 	u8 old_sink_count = intel_dp->sink_count;
 	bool ret;
 
@@ -3936,7 +3961,8 @@ intel_dp_short_pulse(struct intel_dp *intel_dp)
 
 	/* Try to read the source of the interrupt */
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
-	    intel_dp_get_sink_irq(intel_dp, &sink_irq_vector)) {
+	    intel_dp_get_sink_irq(intel_dp, &sink_irq_vector) &&
+	    sink_irq_vector != 0) {
 		/* Clear interrupt source */
 		drm_dp_dpcd_writeb(&intel_dp->aux,
 				   DP_DEVICE_SERVICE_IRQ_VECTOR,
@@ -3980,6 +4006,9 @@ intel_dp_detect_dpcd(struct intel_dp *intel_dp)
 		connector_status_connected : connector_status_disconnected;
 	}
 
+	if (intel_dp_can_mst(intel_dp))
+		return connector_status_connected;
+
 	/* If no HPD, poke DDC gently */
 	if (drm_probe_ddc(&intel_dp->aux.ddc))
 		return connector_status_connected;
@@ -4217,8 +4246,7 @@ intel_dp_long_pulse(struct intel_connector *intel_connector)
 	struct drm_device *dev = connector->dev;
 	enum drm_connector_status status;
 	enum intel_display_power_domain power_domain;
-	bool ret;
-	u8 sink_irq_vector;
+	u8 sink_irq_vector = 0;
 
 	power_domain = intel_display_port_aux_power_domain(intel_encoder);
 	intel_display_power_get(to_i915(dev), power_domain);
@@ -4252,10 +4280,17 @@ intel_dp_long_pulse(struct intel_connector *intel_connector)
 	if (intel_encoder->type != INTEL_OUTPUT_EDP)
 		intel_encoder->type = INTEL_OUTPUT_DP;
 
+	DRM_DEBUG_KMS("Display Port TPS3 support: source %s, sink %s\n",
+		      yesno(intel_dp_source_supports_hbr2(intel_dp)),
+		      yesno(drm_dp_tps3_supported(intel_dp->dpcd)));
+
+	intel_dp_print_rates(intel_dp);
+
 	intel_dp_probe_oui(intel_dp);
 
-	ret = intel_dp_probe_mst(intel_dp);
-	if (ret) {
+	intel_dp_configure_mst(intel_dp);
+
+	if (intel_dp->is_mst) {
 		/*
 		 * If we are in MST mode then this connector
 		 * won't appear connected or have anything
@@ -4290,7 +4325,8 @@ intel_dp_long_pulse(struct intel_connector *intel_connector)
 
 	/* Try to read the source of the interrupt */
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
-	    intel_dp_get_sink_irq(intel_dp, &sink_irq_vector)) {
+	    intel_dp_get_sink_irq(intel_dp, &sink_irq_vector) &&
+	    sink_irq_vector != 0) {
 		/* Clear interrupt source */
 		drm_dp_dpcd_writeb(&intel_dp->aux,
 				   DP_DEVICE_SERVICE_IRQ_VECTOR,
@@ -5186,7 +5222,7 @@ unlock:
 
 /**
  * intel_edp_drrs_invalidate - Disable Idleness DRRS
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called everytime rendering on the given planes start.
@@ -5194,10 +5230,9 @@ unlock:
  *
  * Dirty frontbuffers relevant to DRRS are tracked in busy_frontbuffer_bits.
  */
-void intel_edp_drrs_invalidate(struct drm_device *dev,
-		unsigned frontbuffer_bits)
+void intel_edp_drrs_invalidate(struct drm_i915_private *dev_priv,
+			       unsigned int frontbuffer_bits)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_crtc *crtc;
 	enum pipe pipe;
 
@@ -5229,7 +5264,7 @@ void intel_edp_drrs_invalidate(struct drm_device *dev,
 
 /**
  * intel_edp_drrs_flush - Restart Idleness DRRS
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called every time rendering on the given planes has
@@ -5239,10 +5274,9 @@ void intel_edp_drrs_invalidate(struct drm_device *dev,
  *
  * Dirty frontbuffers relevant to DRRS are tracked in busy_frontbuffer_bits.
  */
-void intel_edp_drrs_flush(struct drm_device *dev,
-		unsigned frontbuffer_bits)
+void intel_edp_drrs_flush(struct drm_i915_private *dev_priv,
+			  unsigned int frontbuffer_bits)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_crtc *crtc;
 	enum pipe pipe;
 
@@ -5413,14 +5447,9 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	pps_unlock(intel_dp);
 
 	/* Cache DPCD and EDID for edp. */
-	has_dpcd = intel_dp_get_dpcd(intel_dp);
+	has_dpcd = intel_edp_init_dpcd(intel_dp);
 
-	if (has_dpcd) {
-		if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
-			dev_priv->no_aux_handshake =
-				intel_dp->dpcd[DP_MAX_DOWNSPREAD] &
-				DP_NO_AUX_HANDSHAKE_LINK_TRAINING;
-	} else {
+	if (!has_dpcd) {
 		/* if this fails, presume the device is a ghost */
 		DRM_INFO("failed to retrieve link info, disabling eDP\n");
 		goto out_vdd_off;

drivers/gpu/drm/i915/intel_dp_mst.c

@@ -170,10 +170,10 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder)
 	intel_mst->connector = found;
 
 	if (intel_dp->active_mst_links == 0) {
-		intel_prepare_ddi_buffer(&intel_dig_port->base);
-
 		intel_ddi_clk_select(&intel_dig_port->base, intel_crtc->config);
 
+		intel_prepare_dp_ddi_buffers(&intel_dig_port->base);
+
 		intel_dp_set_link_params(intel_dp, intel_crtc->config);
 
 		intel_ddi_init_dp_buf_reg(&intel_dig_port->base);

drivers/gpu/drm/i915/intel_drv.h

@@ -849,6 +849,7 @@ struct intel_dp {
 	int link_rate;
 	uint8_t lane_count;
 	uint8_t sink_count;
+	bool link_mst;
 	bool has_audio;
 	bool detect_done;
 	enum hdmi_force_audio force_audio;
@@ -1104,7 +1105,7 @@ void intel_crt_reset(struct drm_encoder *encoder);
 /* intel_ddi.c */
 void intel_ddi_clk_select(struct intel_encoder *encoder,
 			  const struct intel_crtc_state *pipe_config);
-void intel_prepare_ddi_buffer(struct intel_encoder *encoder);
+void intel_prepare_dp_ddi_buffers(struct intel_encoder *encoder);
 void hsw_fdi_link_train(struct drm_crtc *crtc);
 void intel_ddi_init(struct drm_device *dev, enum port port);
 enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder);
@@ -1131,21 +1132,10 @@ void intel_ddi_clock_get(struct intel_encoder *encoder,
 void intel_ddi_set_vc_payload_alloc(struct drm_crtc *crtc, bool state);
 uint32_t ddi_signal_levels(struct intel_dp *intel_dp);
 
-/* intel_frontbuffer.c */
-void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
-			     enum fb_op_origin origin);
-void intel_frontbuffer_flip_prepare(struct drm_device *dev,
-				    unsigned frontbuffer_bits);
-void intel_frontbuffer_flip_complete(struct drm_device *dev,
-				     unsigned frontbuffer_bits);
-void intel_frontbuffer_flip(struct drm_device *dev,
-			    unsigned frontbuffer_bits);
 unsigned int intel_fb_align_height(struct drm_device *dev,
 				   unsigned int height,
 				   uint32_t pixel_format,
 				   uint64_t fb_format_modifier);
-void intel_fb_obj_flush(struct drm_i915_gem_object *obj, bool retire,
-			enum fb_op_origin origin);
 u32 intel_fb_stride_alignment(const struct drm_i915_private *dev_priv,
 			      uint64_t fb_modifier, uint32_t pixel_format);
 
@@ -1381,11 +1371,12 @@ uint32_t intel_dp_pack_aux(const uint8_t *src, int src_bytes);
 void intel_plane_destroy(struct drm_plane *plane);
 void intel_edp_drrs_enable(struct intel_dp *intel_dp);
 void intel_edp_drrs_disable(struct intel_dp *intel_dp);
-void intel_edp_drrs_invalidate(struct drm_device *dev,
-		unsigned frontbuffer_bits);
-void intel_edp_drrs_flush(struct drm_device *dev, unsigned frontbuffer_bits);
+void intel_edp_drrs_invalidate(struct drm_i915_private *dev_priv,
+			       unsigned int frontbuffer_bits);
+void intel_edp_drrs_flush(struct drm_i915_private *dev_priv,
+			  unsigned int frontbuffer_bits);
 bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
-					 struct intel_digital_port *port);
+				  struct intel_digital_port *port);
 
 void
 intel_dp_program_link_training_pattern(struct intel_dp *intel_dp,
@@ -1558,13 +1549,13 @@ static inline void intel_backlight_device_unregister(struct intel_connector *con
 /* intel_psr.c */
 void intel_psr_enable(struct intel_dp *intel_dp);
 void intel_psr_disable(struct intel_dp *intel_dp);
-void intel_psr_invalidate(struct drm_device *dev,
+void intel_psr_invalidate(struct drm_i915_private *dev_priv,
 			  unsigned frontbuffer_bits);
-void intel_psr_flush(struct drm_device *dev,
+void intel_psr_flush(struct drm_i915_private *dev_priv,
 		     unsigned frontbuffer_bits,
 		     enum fb_op_origin origin);
 void intel_psr_init(struct drm_device *dev);
-void intel_psr_single_frame_update(struct drm_device *dev,
+void intel_psr_single_frame_update(struct drm_i915_private *dev_priv,
 				   unsigned frontbuffer_bits);
 
 /* intel_runtime_pm.c */
@@ -1664,13 +1655,6 @@ enable_rpm_wakeref_asserts(struct drm_i915_private *dev_priv)
 	atomic_dec(&dev_priv->pm.wakeref_count);
 }
 
-/* TODO: convert users of these to rely instead on proper RPM refcounting */
-#define DISABLE_RPM_WAKEREF_ASSERTS(dev_priv)	\
-	disable_rpm_wakeref_asserts(dev_priv)
-
-#define ENABLE_RPM_WAKEREF_ASSERTS(dev_priv)	\
-	enable_rpm_wakeref_asserts(dev_priv)
-
 void intel_runtime_pm_get(struct drm_i915_private *dev_priv);
 bool intel_runtime_pm_get_if_in_use(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_get_noresume(struct drm_i915_private *dev_priv);
@@ -1696,11 +1680,11 @@ void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
 void intel_gpu_ips_teardown(void);
 void intel_init_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv);
+void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_enable_gt_powersave(struct drm_i915_private *dev_priv);
+void intel_autoenable_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_disable_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv);
-void intel_reset_gt_powersave(struct drm_i915_private *dev_priv);
-void gen6_update_ring_freq(struct drm_i915_private *dev_priv);
 void gen6_rps_busy(struct drm_i915_private *dev_priv);
 void gen6_rps_reset_ei(struct drm_i915_private *dev_priv);
 void gen6_rps_idle(struct drm_i915_private *dev_priv);

drivers/gpu/drm/i915/intel_engine_cs.c

@@ -0,0 +1,231 @@
+/*
+ * Copyright © 2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#include "i915_drv.h"
+#include "intel_ringbuffer.h"
+#include "intel_lrc.h"
+
+static const struct engine_info {
+	const char *name;
+	unsigned exec_id;
+	unsigned guc_id;
+	u32 mmio_base;
+	unsigned irq_shift;
+	int (*init_legacy)(struct intel_engine_cs *engine);
+	int (*init_execlists)(struct intel_engine_cs *engine);
+} intel_engines[] = {
+	[RCS] = {
+		.name = "render ring",
+		.exec_id = I915_EXEC_RENDER,
+		.guc_id = GUC_RENDER_ENGINE,
+		.mmio_base = RENDER_RING_BASE,
+		.irq_shift = GEN8_RCS_IRQ_SHIFT,
+		.init_execlists = logical_render_ring_init,
+		.init_legacy = intel_init_render_ring_buffer,
+	},
+	[BCS] = {
+		.name = "blitter ring",
+		.exec_id = I915_EXEC_BLT,
+		.guc_id = GUC_BLITTER_ENGINE,
+		.mmio_base = BLT_RING_BASE,
+		.irq_shift = GEN8_BCS_IRQ_SHIFT,
+		.init_execlists = logical_xcs_ring_init,
+		.init_legacy = intel_init_blt_ring_buffer,
+	},
+	[VCS] = {
+		.name = "bsd ring",
+		.exec_id = I915_EXEC_BSD,
+		.guc_id = GUC_VIDEO_ENGINE,
+		.mmio_base = GEN6_BSD_RING_BASE,
+		.irq_shift = GEN8_VCS1_IRQ_SHIFT,
+		.init_execlists = logical_xcs_ring_init,
+		.init_legacy = intel_init_bsd_ring_buffer,
+	},
+	[VCS2] = {
+		.name = "bsd2 ring",
+		.exec_id = I915_EXEC_BSD,
+		.guc_id = GUC_VIDEO_ENGINE2,
+		.mmio_base = GEN8_BSD2_RING_BASE,
+		.irq_shift = GEN8_VCS2_IRQ_SHIFT,
+		.init_execlists = logical_xcs_ring_init,
+		.init_legacy = intel_init_bsd2_ring_buffer,
+	},
+	[VECS] = {
+		.name = "video enhancement ring",
+		.exec_id = I915_EXEC_VEBOX,
+		.guc_id = GUC_VIDEOENHANCE_ENGINE,
+		.mmio_base = VEBOX_RING_BASE,
+		.irq_shift = GEN8_VECS_IRQ_SHIFT,
+		.init_execlists = logical_xcs_ring_init,
+		.init_legacy = intel_init_vebox_ring_buffer,
+	},
+};
+
+static struct intel_engine_cs *
+intel_engine_setup(struct drm_i915_private *dev_priv,
+		   enum intel_engine_id id)
+{
+	const struct engine_info *info = &intel_engines[id];
+	struct intel_engine_cs *engine = &dev_priv->engine[id];
+
+	engine->id = id;
+	engine->i915 = dev_priv;
+	engine->name = info->name;
+	engine->exec_id = info->exec_id;
+	engine->hw_id = engine->guc_id = info->guc_id;
+	engine->mmio_base = info->mmio_base;
+	engine->irq_shift = info->irq_shift;
+
+	return engine;
+}
+
+/**
+ * intel_engines_init() - allocate, populate and init the Engine Command Streamers
+ * @dev: DRM device.
+ *
+ * Return: non-zero if the initialization failed.
+ */
+int intel_engines_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	unsigned int mask = 0;
+	int (*init)(struct intel_engine_cs *engine);
+	unsigned int i;
+	int ret;
+
+	WARN_ON(INTEL_INFO(dev_priv)->ring_mask == 0);
+	WARN_ON(INTEL_INFO(dev_priv)->ring_mask &
+		GENMASK(sizeof(mask) * BITS_PER_BYTE - 1, I915_NUM_ENGINES));
+
+	for (i = 0; i < ARRAY_SIZE(intel_engines); i++) {
+		if (!HAS_ENGINE(dev_priv, i))
+			continue;
+
+		if (i915.enable_execlists)
+			init = intel_engines[i].init_execlists;
+		else
+			init = intel_engines[i].init_legacy;
+
+		if (!init)
+			continue;
+
+		ret = init(intel_engine_setup(dev_priv, i));
+		if (ret)
+			goto cleanup;
+
+		mask |= ENGINE_MASK(i);
+	}
+
+	/*
+	 * Catch failures to update intel_engines table when the new engines
+	 * are added to the driver by a warning and disabling the forgotten
+	 * engines.
+	 */
+	if (WARN_ON(mask != INTEL_INFO(dev_priv)->ring_mask)) {
+		struct intel_device_info *info =
+			(struct intel_device_info *)&dev_priv->info;
+		info->ring_mask = mask;
+	}
+
+	return 0;
+
+cleanup:
+	for (i = 0; i < I915_NUM_ENGINES; i++) {
+		if (i915.enable_execlists)
+			intel_logical_ring_cleanup(&dev_priv->engine[i]);
+		else
+			intel_engine_cleanup(&dev_priv->engine[i]);
+	}
+
+	return ret;
+}
+
+void intel_engine_init_hangcheck(struct intel_engine_cs *engine)
+{
+	memset(&engine->hangcheck, 0, sizeof(engine->hangcheck));
+}
+
+static void intel_engine_init_requests(struct intel_engine_cs *engine)
+{
+	init_request_active(&engine->last_request, NULL);
+	INIT_LIST_HEAD(&engine->request_list);
+}
+
+/**
+ * intel_engines_setup_common - setup engine state not requiring hw access
+ * @engine: Engine to setup.
+ *
+ * Initializes @engine@ structure members shared between legacy and execlists
+ * submission modes which do not require hardware access.
+ *
+ * Typically done early in the submission mode specific engine setup stage.
+ */
+void intel_engine_setup_common(struct intel_engine_cs *engine)
+{
+	INIT_LIST_HEAD(&engine->buffers);
+	INIT_LIST_HEAD(&engine->execlist_queue);
+	spin_lock_init(&engine->execlist_lock);
+
+	engine->fence_context = fence_context_alloc(1);
+
+	intel_engine_init_requests(engine);
+	intel_engine_init_hangcheck(engine);
+	i915_gem_batch_pool_init(engine, &engine->batch_pool);
+}
+
+/**
+ * intel_engines_init_common - initialize cengine state which might require hw access
+ * @engine: Engine to initialize.
+ *
+ * Initializes @engine@ structure members shared between legacy and execlists
+ * submission modes which do require hardware access.
+ *
+ * Typcally done at later stages of submission mode specific engine setup.
+ *
+ * Returns zero on success or an error code on failure.
+ */
+int intel_engine_init_common(struct intel_engine_cs *engine)
+{
+	int ret;
+
+	ret = intel_engine_init_breadcrumbs(engine);
+	if (ret)
+		return ret;
+
+	return intel_engine_init_cmd_parser(engine);
+}
+
+/**
+ * intel_engines_cleanup_common - cleans up the engine state created by
+ *                                the common initiailizers.
+ * @engine: Engine to cleanup.
+ *
+ * This cleans up everything created by the common helpers.
+ */
+void intel_engine_cleanup_common(struct intel_engine_cs *engine)
+{
+	intel_engine_cleanup_cmd_parser(engine);
+	intel_engine_fini_breadcrumbs(engine);
+	i915_gem_batch_pool_fini(&engine->batch_pool);
+}

drivers/gpu/drm/i915/intel_fbc.c

@@ -741,7 +741,7 @@ static void intel_fbc_update_state_cache(struct intel_crtc *crtc,
 	cache->fb.pixel_format = fb->pixel_format;
 	cache->fb.stride = fb->pitches[0];
 	cache->fb.fence_reg = obj->fence_reg;
-	cache->fb.tiling_mode = obj->tiling_mode;
+	cache->fb.tiling_mode = i915_gem_object_get_tiling(obj);
 }
 
 static bool intel_fbc_can_activate(struct intel_crtc *crtc)
@@ -1075,6 +1075,8 @@ out:
 /**
  * intel_fbc_enable: tries to enable FBC on the CRTC
  * @crtc: the CRTC
+ * @crtc_state: corresponding &drm_crtc_state for @crtc
+ * @plane_state: corresponding &drm_plane_state for the primary plane of @crtc
  *
  * This function checks if the given CRTC was chosen for FBC, then enables it if
  * possible. Notice that it doesn't activate FBC. It is valid to call
@@ -1163,11 +1165,8 @@ void intel_fbc_disable(struct intel_crtc *crtc)
 		return;
 
 	mutex_lock(&fbc->lock);
-	if (fbc->crtc == crtc) {
-		WARN_ON(!fbc->enabled);
-		WARN_ON(fbc->active);
+	if (fbc->crtc == crtc)
 		__intel_fbc_disable(dev_priv);
-	}
 	mutex_unlock(&fbc->lock);
 
 	cancel_work_sync(&fbc->work.work);
@@ -1230,12 +1229,29 @@ static int intel_sanitize_fbc_option(struct drm_i915_private *dev_priv)
 	if (i915.enable_fbc >= 0)
 		return !!i915.enable_fbc;
 
+	if (!HAS_FBC(dev_priv))
+		return 0;
+
 	if (IS_BROADWELL(dev_priv))
 		return 1;
 
 	return 0;
 }
 
+static bool need_fbc_vtd_wa(struct drm_i915_private *dev_priv)
+{
+#ifdef CONFIG_INTEL_IOMMU
+	/* WaFbcTurnOffFbcWhenHyperVisorIsUsed:skl,bxt */
+	if (intel_iommu_gfx_mapped &&
+	    (IS_SKYLAKE(dev_priv) || IS_BROXTON(dev_priv))) {
+		DRM_INFO("Disabling framebuffer compression (FBC) to prevent screen flicker with VT-d enabled\n");
+		return true;
+	}
+#endif
+
+	return false;
+}
+
 /**
  * intel_fbc_init - Initialize FBC
  * @dev_priv: the i915 device
@@ -1253,6 +1269,9 @@ void intel_fbc_init(struct drm_i915_private *dev_priv)
 	fbc->active = false;
 	fbc->work.scheduled = false;
 
+	if (need_fbc_vtd_wa(dev_priv))
+		mkwrite_device_info(dev_priv)->has_fbc = false;
+
 	i915.enable_fbc = intel_sanitize_fbc_option(dev_priv);
 	DRM_DEBUG_KMS("Sanitized enable_fbc value: %d\n", i915.enable_fbc);
 

drivers/gpu/drm/i915/intel_fbdev.c

@@ -41,6 +41,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
 #include "intel_drv.h"
+#include "intel_frontbuffer.h"
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
@@ -158,7 +159,7 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 
 	fb = __intel_framebuffer_create(dev, &mode_cmd, obj);
 	if (IS_ERR(fb)) {
-		drm_gem_object_unreference(&obj->base);
+		i915_gem_object_put(obj);
 		ret = PTR_ERR(fb);
 		goto out;
 	}
@@ -188,7 +189,7 @@ static int intelfb_create(struct drm_fb_helper *helper,
 	struct i915_vma *vma;
 	struct drm_i915_gem_object *obj;
 	bool prealloc = false;
-	void *vaddr;
+	void __iomem *vaddr;
 	int ret;
 
 	if (intel_fb &&
@@ -767,7 +768,7 @@ void intel_fbdev_fini(struct drm_device *dev)
 	if (!ifbdev)
 		return;
 
-	flush_work(&dev_priv->fbdev_suspend_work);
+	cancel_work_sync(&dev_priv->fbdev_suspend_work);
 	if (!current_is_async())
 		intel_fbdev_sync(ifbdev);
 

drivers/gpu/drm/i915/intel_frontbuffer.c

@@ -63,47 +63,30 @@
 #include <drm/drmP.h>
 
 #include "intel_drv.h"
+#include "intel_frontbuffer.h"
 #include "i915_drv.h"
 
-/**
- * intel_fb_obj_invalidate - invalidate frontbuffer object
- * @obj: GEM object to invalidate
- * @origin: which operation caused the invalidation
- *
- * This function gets called every time rendering on the given object starts and
- * frontbuffer caching (fbc, low refresh rate for DRRS, panel self refresh) must
- * be invalidated. For ORIGIN_CS any subsequent invalidation will be delayed
- * until the rendering completes or a flip on this frontbuffer plane is
- * scheduled.
- */
-void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
-			     enum fb_op_origin origin)
+void __intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
+			       enum fb_op_origin origin,
+			       unsigned int frontbuffer_bits)
 {
-	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-
-	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
-
-	if (!obj->frontbuffer_bits)
-		return;
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 
 	if (origin == ORIGIN_CS) {
-		mutex_lock(&dev_priv->fb_tracking.lock);
-		dev_priv->fb_tracking.busy_bits
-			|= obj->frontbuffer_bits;
-		dev_priv->fb_tracking.flip_bits
-			&= ~obj->frontbuffer_bits;
-		mutex_unlock(&dev_priv->fb_tracking.lock);
+		spin_lock(&dev_priv->fb_tracking.lock);
+		dev_priv->fb_tracking.busy_bits |= frontbuffer_bits;
+		dev_priv->fb_tracking.flip_bits &= ~frontbuffer_bits;
+		spin_unlock(&dev_priv->fb_tracking.lock);
 	}
 
-	intel_psr_invalidate(dev, obj->frontbuffer_bits);
-	intel_edp_drrs_invalidate(dev, obj->frontbuffer_bits);
-	intel_fbc_invalidate(dev_priv, obj->frontbuffer_bits, origin);
+	intel_psr_invalidate(dev_priv, frontbuffer_bits);
+	intel_edp_drrs_invalidate(dev_priv, frontbuffer_bits);
+	intel_fbc_invalidate(dev_priv, frontbuffer_bits, origin);
 }
 
 /**
  * intel_frontbuffer_flush - flush frontbuffer
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  * @origin: which operation caused the flush
  *
@@ -113,64 +96,45 @@ void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
  *
  * Can be called without any locks held.
  */
-static void intel_frontbuffer_flush(struct drm_device *dev,
+static void intel_frontbuffer_flush(struct drm_i915_private *dev_priv,
 				    unsigned frontbuffer_bits,
 				    enum fb_op_origin origin)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-
 	/* Delay flushing when rings are still busy.*/
-	mutex_lock(&dev_priv->fb_tracking.lock);
+	spin_lock(&dev_priv->fb_tracking.lock);
 	frontbuffer_bits &= ~dev_priv->fb_tracking.busy_bits;
-	mutex_unlock(&dev_priv->fb_tracking.lock);
+	spin_unlock(&dev_priv->fb_tracking.lock);
 
 	if (!frontbuffer_bits)
 		return;
 
-	intel_edp_drrs_flush(dev, frontbuffer_bits);
-	intel_psr_flush(dev, frontbuffer_bits, origin);
+	intel_edp_drrs_flush(dev_priv, frontbuffer_bits);
+	intel_psr_flush(dev_priv, frontbuffer_bits, origin);
 	intel_fbc_flush(dev_priv, frontbuffer_bits, origin);
 }
 
-/**
- * intel_fb_obj_flush - flush frontbuffer object
- * @obj: GEM object to flush
- * @retire: set when retiring asynchronous rendering
- * @origin: which operation caused the flush
- *
- * This function gets called every time rendering on the given object has
- * completed and frontbuffer caching can be started again. If @retire is true
- * then any delayed flushes will be unblocked.
- */
-void intel_fb_obj_flush(struct drm_i915_gem_object *obj,
-			bool retire, enum fb_op_origin origin)
+void __intel_fb_obj_flush(struct drm_i915_gem_object *obj,
+			  bool retire,
+			  enum fb_op_origin origin,
+			  unsigned int frontbuffer_bits)
 {
-	struct drm_device *dev = obj->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	unsigned frontbuffer_bits;
-
-	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
-
-	if (!obj->frontbuffer_bits)
-		return;
-
-	frontbuffer_bits = obj->frontbuffer_bits;
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 
 	if (retire) {
-		mutex_lock(&dev_priv->fb_tracking.lock);
+		spin_lock(&dev_priv->fb_tracking.lock);
 		/* Filter out new bits since rendering started. */
 		frontbuffer_bits &= dev_priv->fb_tracking.busy_bits;
-
 		dev_priv->fb_tracking.busy_bits &= ~frontbuffer_bits;
-		mutex_unlock(&dev_priv->fb_tracking.lock);
+		spin_unlock(&dev_priv->fb_tracking.lock);
 	}
 
-	intel_frontbuffer_flush(dev, frontbuffer_bits, origin);
+	if (frontbuffer_bits)
+		intel_frontbuffer_flush(dev_priv, frontbuffer_bits, origin);
 }
 
 /**
  * intel_frontbuffer_flip_prepare - prepare asynchronous frontbuffer flip
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called after scheduling a flip on @obj. The actual
@@ -180,23 +144,21 @@ void intel_fb_obj_flush(struct drm_i915_gem_object *obj,
  *
  * Can be called without any locks held.
  */
-void intel_frontbuffer_flip_prepare(struct drm_device *dev,
+void intel_frontbuffer_flip_prepare(struct drm_i915_private *dev_priv,
 				    unsigned frontbuffer_bits)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-
-	mutex_lock(&dev_priv->fb_tracking.lock);
+	spin_lock(&dev_priv->fb_tracking.lock);
 	dev_priv->fb_tracking.flip_bits |= frontbuffer_bits;
 	/* Remove stale busy bits due to the old buffer. */
 	dev_priv->fb_tracking.busy_bits &= ~frontbuffer_bits;
-	mutex_unlock(&dev_priv->fb_tracking.lock);
+	spin_unlock(&dev_priv->fb_tracking.lock);
 
-	intel_psr_single_frame_update(dev, frontbuffer_bits);
+	intel_psr_single_frame_update(dev_priv, frontbuffer_bits);
 }
 
 /**
  * intel_frontbuffer_flip_complete - complete asynchronous frontbuffer flip
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called after the flip has been latched and will complete
@@ -204,23 +166,23 @@ void intel_frontbuffer_flip_prepare(struct drm_device *dev,
  *
  * Can be called without any locks held.
  */
-void intel_frontbuffer_flip_complete(struct drm_device *dev,
+void intel_frontbuffer_flip_complete(struct drm_i915_private *dev_priv,
 				     unsigned frontbuffer_bits)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-
-	mutex_lock(&dev_priv->fb_tracking.lock);
+	spin_lock(&dev_priv->fb_tracking.lock);
 	/* Mask any cancelled flips. */
 	frontbuffer_bits &= dev_priv->fb_tracking.flip_bits;
 	dev_priv->fb_tracking.flip_bits &= ~frontbuffer_bits;
-	mutex_unlock(&dev_priv->fb_tracking.lock);
+	spin_unlock(&dev_priv->fb_tracking.lock);
 
-	intel_frontbuffer_flush(dev, frontbuffer_bits, ORIGIN_FLIP);
+	if (frontbuffer_bits)
+		intel_frontbuffer_flush(dev_priv,
+					frontbuffer_bits, ORIGIN_FLIP);
 }
 
 /**
  * intel_frontbuffer_flip - synchronous frontbuffer flip
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called after scheduling a flip on @obj. This is for
@@ -229,15 +191,13 @@ void intel_frontbuffer_flip_complete(struct drm_device *dev,
  *
  * Can be called without any locks held.
  */
-void intel_frontbuffer_flip(struct drm_device *dev,
+void intel_frontbuffer_flip(struct drm_i915_private *dev_priv,
 			    unsigned frontbuffer_bits)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-
-	mutex_lock(&dev_priv->fb_tracking.lock);
+	spin_lock(&dev_priv->fb_tracking.lock);
 	/* Remove stale busy bits due to the old buffer. */
 	dev_priv->fb_tracking.busy_bits &= ~frontbuffer_bits;
-	mutex_unlock(&dev_priv->fb_tracking.lock);
+	spin_unlock(&dev_priv->fb_tracking.lock);
 
-	intel_frontbuffer_flush(dev, frontbuffer_bits, ORIGIN_FLIP);
+	intel_frontbuffer_flush(dev_priv, frontbuffer_bits, ORIGIN_FLIP);
 }

drivers/gpu/drm/i915/intel_frontbuffer.h

@@ -0,0 +1,91 @@
+/*
+ * Copyright (c) 2014-2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __INTEL_FRONTBUFFER_H__
+#define __INTEL_FRONTBUFFER_H__
+
+struct drm_i915_private;
+struct drm_i915_gem_object;
+
+void intel_frontbuffer_flip_prepare(struct drm_i915_private *dev_priv,
+				    unsigned frontbuffer_bits);
+void intel_frontbuffer_flip_complete(struct drm_i915_private *dev_priv,
+				     unsigned frontbuffer_bits);
+void intel_frontbuffer_flip(struct drm_i915_private *dev_priv,
+			    unsigned frontbuffer_bits);
+
+void __intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
+			       enum fb_op_origin origin,
+			       unsigned int frontbuffer_bits);
+void __intel_fb_obj_flush(struct drm_i915_gem_object *obj,
+			  bool retire,
+			  enum fb_op_origin origin,
+			  unsigned int frontbuffer_bits);
+
+/**
+ * intel_fb_obj_invalidate - invalidate frontbuffer object
+ * @obj: GEM object to invalidate
+ * @origin: which operation caused the invalidation
+ *
+ * This function gets called every time rendering on the given object starts and
+ * frontbuffer caching (fbc, low refresh rate for DRRS, panel self refresh) must
+ * be invalidated. For ORIGIN_CS any subsequent invalidation will be delayed
+ * until the rendering completes or a flip on this frontbuffer plane is
+ * scheduled.
+ */
+static inline void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
+					   enum fb_op_origin origin)
+{
+	unsigned int frontbuffer_bits;
+
+	frontbuffer_bits = atomic_read(&obj->frontbuffer_bits);
+	if (!frontbuffer_bits)
+		return;
+
+	__intel_fb_obj_invalidate(obj, origin, frontbuffer_bits);
+}
+
+/**
+ * intel_fb_obj_flush - flush frontbuffer object
+ * @obj: GEM object to flush
+ * @retire: set when retiring asynchronous rendering
+ * @origin: which operation caused the flush
+ *
+ * This function gets called every time rendering on the given object has
+ * completed and frontbuffer caching can be started again. If @retire is true
+ * then any delayed flushes will be unblocked.
+ */
+static inline void intel_fb_obj_flush(struct drm_i915_gem_object *obj,
+				      bool retire,
+				      enum fb_op_origin origin)
+{
+	unsigned int frontbuffer_bits;
+
+	frontbuffer_bits = atomic_read(&obj->frontbuffer_bits);
+	if (!frontbuffer_bits)
+		return;
+
+	__intel_fb_obj_flush(obj, retire, origin, frontbuffer_bits);
+}
+
+#endif /* __INTEL_FRONTBUFFER_H__ */

drivers/gpu/drm/i915/intel_guc.h

@@ -160,7 +160,6 @@ extern int intel_guc_resume(struct drm_device *dev);
 int i915_guc_submission_init(struct drm_i915_private *dev_priv);
 int i915_guc_submission_enable(struct drm_i915_private *dev_priv);
 int i915_guc_wq_check_space(struct drm_i915_gem_request *rq);
-int i915_guc_submit(struct drm_i915_gem_request *rq);
 void i915_guc_submission_disable(struct drm_i915_private *dev_priv);
 void i915_guc_submission_fini(struct drm_i915_private *dev_priv);
 

drivers/gpu/drm/i915/intel_guc_loader.c

@@ -323,7 +323,7 @@ static int guc_ucode_xfer(struct drm_i915_private *dev_priv)
 		return ret;
 	}
 
-	ret = i915_gem_obj_ggtt_pin(guc_fw->guc_fw_obj, 0, 0);
+	ret = i915_gem_object_ggtt_pin(guc_fw->guc_fw_obj, NULL, 0, 0, 0);
 	if (ret) {
 		DRM_DEBUG_DRIVER("pin failed %d\n", ret);
 		return ret;
@@ -349,7 +349,9 @@ static int guc_ucode_xfer(struct drm_i915_private *dev_priv)
 	}
 
 	/* WaC6DisallowByGfxPause*/
-	I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
+	if (IS_SKL_REVID(dev, 0, SKL_REVID_C0) ||
+	    IS_BXT_REVID(dev, 0, BXT_REVID_B0))
+		I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
 
 	if (IS_BROXTON(dev))
 		I915_WRITE(GEN9LP_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
@@ -662,7 +664,7 @@ fail:
 	mutex_lock(&dev->struct_mutex);
 	obj = guc_fw->guc_fw_obj;
 	if (obj)
-		drm_gem_object_unreference(&obj->base);
+		i915_gem_object_put(obj);
 	guc_fw->guc_fw_obj = NULL;
 	mutex_unlock(&dev->struct_mutex);
 
@@ -743,7 +745,7 @@ void intel_guc_fini(struct drm_device *dev)
 	i915_guc_submission_fini(dev_priv);
 
 	if (guc_fw->guc_fw_obj)
-		drm_gem_object_unreference(&guc_fw->guc_fw_obj->base);
+		i915_gem_object_put(guc_fw->guc_fw_obj);
 	guc_fw->guc_fw_obj = NULL;
 	mutex_unlock(&dev->struct_mutex);
 

drivers/gpu/drm/i915/intel_hotplug.c

@@ -525,7 +525,6 @@ void i915_hpd_poll_init_work(struct work_struct *work) {
 /**
  * intel_hpd_poll_init - enables/disables polling for connectors with hpd
  * @dev_priv: i915 device instance
- * @enabled: Whether to enable or disable polling
  *
  * This function enables polling for all connectors, regardless of whether or
  * not they support hotplug detection. Under certain conditions HPD may not be

drivers/gpu/drm/i915/intel_lrc.c

@@ -288,7 +288,6 @@ logical_ring_init_platform_invariants(struct intel_engine_cs *engine)
 /**
  * intel_lr_context_descriptor_update() - calculate & cache the descriptor
  * 					  descriptor for a pinned context
- *
  * @ctx: Context to work on
  * @engine: Engine the descriptor will be used with
  *
@@ -297,12 +296,13 @@ logical_ring_init_platform_invariants(struct intel_engine_cs *engine)
  * expensive to calculate, we'll just do it once and cache the result,
  * which remains valid until the context is unpinned.
  *
- * This is what a descriptor looks like, from LSB to MSB:
- *    bits  0-11:    flags, GEN8_CTX_* (cached in ctx_desc_template)
- *    bits 12-31:    LRCA, GTT address of (the HWSP of) this context
- *    bits 32-52:    ctx ID, a globally unique tag
- *    bits 53-54:    mbz, reserved for use by hardware
- *    bits 55-63:    group ID, currently unused and set to 0
+ * This is what a descriptor looks like, from LSB to MSB::
+ *
+ *      bits  0-11:    flags, GEN8_CTX_* (cached in ctx_desc_template)
+ *      bits 12-31:    LRCA, GTT address of (the HWSP of) this context
+ *      bits 32-52:    ctx ID, a globally unique tag
+ *      bits 53-54:    mbz, reserved for use by hardware
+ *      bits 55-63:    group ID, currently unused and set to 0
  */
 static void
 intel_lr_context_descriptor_update(struct i915_gem_context *ctx,
@@ -373,7 +373,7 @@ static void execlists_update_context(struct drm_i915_gem_request *rq)
 	struct i915_hw_ppgtt *ppgtt = rq->ctx->ppgtt;
 	uint32_t *reg_state = rq->ctx->engine[engine->id].lrc_reg_state;
 
-	reg_state[CTX_RING_TAIL+1] = rq->tail;
+	reg_state[CTX_RING_TAIL+1] = intel_ring_offset(rq->ring, rq->tail);
 
 	/* True 32b PPGTT with dynamic page allocation: update PDP
 	 * registers and point the unallocated PDPs to scratch page.
@@ -384,8 +384,8 @@ static void execlists_update_context(struct drm_i915_gem_request *rq)
 		execlists_update_context_pdps(ppgtt, reg_state);
 }
 
-static void execlists_submit_requests(struct drm_i915_gem_request *rq0,
-				      struct drm_i915_gem_request *rq1)
+static void execlists_elsp_submit_contexts(struct drm_i915_gem_request *rq0,
+					   struct drm_i915_gem_request *rq1)
 {
 	struct drm_i915_private *dev_priv = rq0->i915;
 	unsigned int fw_domains = rq0->engine->fw_domains;
@@ -418,7 +418,7 @@ static inline void execlists_context_status_change(
 	atomic_notifier_call_chain(&rq->ctx->status_notifier, status, rq);
 }
 
-static void execlists_context_unqueue(struct intel_engine_cs *engine)
+static void execlists_unqueue(struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_request *req0 = NULL, *req1 = NULL;
 	struct drm_i915_gem_request *cursor, *tmp;
@@ -441,7 +441,7 @@ static void execlists_context_unqueue(struct intel_engine_cs *engine)
 			 * will update tail past first request's workload */
 			cursor->elsp_submitted = req0->elsp_submitted;
 			list_del(&req0->execlist_link);
-			i915_gem_request_unreference(req0);
+			i915_gem_request_put(req0);
 			req0 = cursor;
 		} else {
 			if (IS_ENABLED(CONFIG_DRM_I915_GVT)) {
@@ -482,14 +482,11 @@ static void execlists_context_unqueue(struct intel_engine_cs *engine)
 		 * resubmit the request. See gen8_emit_request() for where we
 		 * prepare the padding after the end of the request.
 		 */
-		struct intel_ringbuffer *ringbuf;
-
-		ringbuf = req0->ctx->engine[engine->id].ringbuf;
 		req0->tail += 8;
-		req0->tail &= ringbuf->size - 1;
+		req0->tail &= req0->ring->size - 1;
 	}
 
-	execlists_submit_requests(req0, req1);
+	execlists_elsp_submit_contexts(req0, req1);
 }
 
 static unsigned int
@@ -514,7 +511,7 @@ execlists_check_remove_request(struct intel_engine_cs *engine, u32 ctx_id)
 	execlists_context_status_change(head_req, INTEL_CONTEXT_SCHEDULE_OUT);
 
 	list_del(&head_req->execlist_link);
-	i915_gem_request_unreference(head_req);
+	i915_gem_request_put(head_req);
 
 	return 1;
 }
@@ -539,10 +536,7 @@ get_context_status(struct intel_engine_cs *engine, unsigned int read_pointer,
 	return status;
 }
 
-/**
- * intel_lrc_irq_handler() - handle Context Switch interrupts
- * @data: tasklet handler passed in unsigned long
- *
+/*
  * Check the unread Context Status Buffers and manage the submission of new
  * contexts to the ELSP accordingly.
  */
@@ -603,7 +597,7 @@ static void intel_lrc_irq_handler(unsigned long data)
 	if (submit_contexts) {
 		if (!engine->disable_lite_restore_wa ||
 		    (csb[i][0] & GEN8_CTX_STATUS_ACTIVE_IDLE))
-			execlists_context_unqueue(engine);
+			execlists_unqueue(engine);
 	}
 
 	spin_unlock(&engine->execlist_lock);
@@ -612,7 +606,7 @@ static void intel_lrc_irq_handler(unsigned long data)
 		DRM_ERROR("More than two context complete events?\n");
 }
 
-static void execlists_context_queue(struct drm_i915_gem_request *request)
+static void execlists_submit_request(struct drm_i915_gem_request *request)
 {
 	struct intel_engine_cs *engine = request->engine;
 	struct drm_i915_gem_request *cursor;
@@ -635,70 +629,19 @@ static void execlists_context_queue(struct drm_i915_gem_request *request)
 			WARN(tail_req->elsp_submitted != 0,
 				"More than 2 already-submitted reqs queued\n");
 			list_del(&tail_req->execlist_link);
-			i915_gem_request_unreference(tail_req);
+			i915_gem_request_put(tail_req);
 		}
 	}
 
-	i915_gem_request_reference(request);
+	i915_gem_request_get(request);
 	list_add_tail(&request->execlist_link, &engine->execlist_queue);
 	request->ctx_hw_id = request->ctx->hw_id;
 	if (num_elements == 0)
-		execlists_context_unqueue(engine);
+		execlists_unqueue(engine);
 
 	spin_unlock_bh(&engine->execlist_lock);
 }
 
-static int logical_ring_invalidate_all_caches(struct drm_i915_gem_request *req)
-{
-	struct intel_engine_cs *engine = req->engine;
-	uint32_t flush_domains;
-	int ret;
-
-	flush_domains = 0;
-	if (engine->gpu_caches_dirty)
-		flush_domains = I915_GEM_GPU_DOMAINS;
-
-	ret = engine->emit_flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
-	if (ret)
-		return ret;
-
-	engine->gpu_caches_dirty = false;
-	return 0;
-}
-
-static int execlists_move_to_gpu(struct drm_i915_gem_request *req,
-				 struct list_head *vmas)
-{
-	const unsigned other_rings = ~intel_engine_flag(req->engine);
-	struct i915_vma *vma;
-	uint32_t flush_domains = 0;
-	bool flush_chipset = false;
-	int ret;
-
-	list_for_each_entry(vma, vmas, exec_list) {
-		struct drm_i915_gem_object *obj = vma->obj;
-
-		if (obj->active & other_rings) {
-			ret = i915_gem_object_sync(obj, req->engine, &req);
-			if (ret)
-				return ret;
-		}
-
-		if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
-			flush_chipset |= i915_gem_clflush_object(obj, false);
-
-		flush_domains |= obj->base.write_domain;
-	}
-
-	if (flush_domains & I915_GEM_DOMAIN_GTT)
-		wmb();
-
-	/* Unconditionally invalidate gpu caches and ensure that we do flush
-	 * any residual writes from the previous batch.
-	 */
-	return logical_ring_invalidate_all_caches(req);
-}
-
 int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request)
 {
 	struct intel_engine_cs *engine = request->engine;
@@ -717,7 +660,7 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request
 			return ret;
 	}
 
-	request->ringbuf = ce->ringbuf;
+	request->ring = ce->ring;
 
 	if (i915.enable_guc_submission) {
 		/*
@@ -762,7 +705,7 @@ err_unpin:
 }
 
 /*
- * intel_logical_ring_advance_and_submit() - advance the tail and submit the workload
+ * intel_logical_ring_advance() - advance the tail and prepare for submission
  * @request: Request to advance the logical ringbuffer of.
  *
  * The tail is updated in our logical ringbuffer struct, not in the actual context. What
@@ -771,13 +714,13 @@ err_unpin:
  * point, the tail *inside* the context is updated and the ELSP written to.
  */
 static int
-intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
+intel_logical_ring_advance(struct drm_i915_gem_request *request)
 {
-	struct intel_ringbuffer *ringbuf = request->ringbuf;
+	struct intel_ring *ring = request->ring;
 	struct intel_engine_cs *engine = request->engine;
 
-	intel_logical_ring_advance(ringbuf);
-	request->tail = ringbuf->tail;
+	intel_ring_advance(ring);
+	request->tail = ring->tail;
 
 	/*
 	 * Here we add two extra NOOPs as padding to avoid
@@ -785,9 +728,9 @@ intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 	 *
 	 * Caller must reserve WA_TAIL_DWORDS for us!
 	 */
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	/* We keep the previous context alive until we retire the following
 	 * request. This ensures that any the context object is still pinned
@@ -797,100 +740,6 @@ intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 	 */
 	request->previous_context = engine->last_context;
 	engine->last_context = request->ctx;
-
-	if (i915.enable_guc_submission)
-		i915_guc_submit(request);
-	else
-		execlists_context_queue(request);
-
-	return 0;
-}
-
-/**
- * execlists_submission() - submit a batchbuffer for execution, Execlists style
- * @params: execbuffer call parameters.
- * @args: execbuffer call arguments.
- * @vmas: list of vmas.
- *
- * This is the evil twin version of i915_gem_ringbuffer_submission. It abstracts
- * away the submission details of the execbuffer ioctl call.
- *
- * Return: non-zero if the submission fails.
- */
-int intel_execlists_submission(struct i915_execbuffer_params *params,
-			       struct drm_i915_gem_execbuffer2 *args,
-			       struct list_head *vmas)
-{
-	struct drm_device       *dev = params->dev;
-	struct intel_engine_cs *engine = params->engine;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_ringbuffer *ringbuf = params->ctx->engine[engine->id].ringbuf;
-	u64 exec_start;
-	int instp_mode;
-	u32 instp_mask;
-	int ret;
-
-	instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
-	instp_mask = I915_EXEC_CONSTANTS_MASK;
-	switch (instp_mode) {
-	case I915_EXEC_CONSTANTS_REL_GENERAL:
-	case I915_EXEC_CONSTANTS_ABSOLUTE:
-	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (instp_mode != 0 && engine != &dev_priv->engine[RCS]) {
-			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
-			return -EINVAL;
-		}
-
-		if (instp_mode != dev_priv->relative_constants_mode) {
-			if (instp_mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
-				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
-				return -EINVAL;
-			}
-
-			/* The HW changed the meaning on this bit on gen6 */
-			instp_mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
-		}
-		break;
-	default:
-		DRM_DEBUG("execbuf with unknown constants: %d\n", instp_mode);
-		return -EINVAL;
-	}
-
-	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
-		DRM_DEBUG("sol reset is gen7 only\n");
-		return -EINVAL;
-	}
-
-	ret = execlists_move_to_gpu(params->request, vmas);
-	if (ret)
-		return ret;
-
-	if (engine == &dev_priv->engine[RCS] &&
-	    instp_mode != dev_priv->relative_constants_mode) {
-		ret = intel_ring_begin(params->request, 4);
-		if (ret)
-			return ret;
-
-		intel_logical_ring_emit(ringbuf, MI_NOOP);
-		intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(1));
-		intel_logical_ring_emit_reg(ringbuf, INSTPM);
-		intel_logical_ring_emit(ringbuf, instp_mask << 16 | instp_mode);
-		intel_logical_ring_advance(ringbuf);
-
-		dev_priv->relative_constants_mode = instp_mode;
-	}
-
-	exec_start = params->batch_obj_vm_offset +
-		     args->batch_start_offset;
-
-	ret = engine->emit_bb_start(params->request, exec_start, params->dispatch_flags);
-	if (ret)
-		return ret;
-
-	trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
-
-	i915_gem_execbuffer_move_to_active(vmas, params->request);
-
 	return 0;
 }
 
@@ -907,51 +756,10 @@ void intel_execlists_cancel_requests(struct intel_engine_cs *engine)
 
 	list_for_each_entry_safe(req, tmp, &cancel_list, execlist_link) {
 		list_del(&req->execlist_link);
-		i915_gem_request_unreference(req);
+		i915_gem_request_put(req);
 	}
 }
 
-void intel_logical_ring_stop(struct intel_engine_cs *engine)
-{
-	struct drm_i915_private *dev_priv = engine->i915;
-	int ret;
-
-	if (!intel_engine_initialized(engine))
-		return;
-
-	ret = intel_engine_idle(engine);
-	if (ret)
-		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
-			  engine->name, ret);
-
-	/* TODO: Is this correct with Execlists enabled? */
-	I915_WRITE_MODE(engine, _MASKED_BIT_ENABLE(STOP_RING));
-	if (intel_wait_for_register(dev_priv,
-				    RING_MI_MODE(engine->mmio_base),
-				    MODE_IDLE, MODE_IDLE,
-				    1000)) {
-		DRM_ERROR("%s :timed out trying to stop ring\n", engine->name);
-		return;
-	}
-	I915_WRITE_MODE(engine, _MASKED_BIT_DISABLE(STOP_RING));
-}
-
-int logical_ring_flush_all_caches(struct drm_i915_gem_request *req)
-{
-	struct intel_engine_cs *engine = req->engine;
-	int ret;
-
-	if (!engine->gpu_caches_dirty)
-		return 0;
-
-	ret = engine->emit_flush(req, 0, I915_GEM_GPU_DOMAINS);
-	if (ret)
-		return ret;
-
-	engine->gpu_caches_dirty = false;
-	return 0;
-}
-
 static int intel_lr_context_pin(struct i915_gem_context *ctx,
 				struct intel_engine_cs *engine)
 {
@@ -966,8 +774,9 @@ static int intel_lr_context_pin(struct i915_gem_context *ctx,
 	if (ce->pin_count++)
 		return 0;
 
-	ret = i915_gem_obj_ggtt_pin(ce->state, GEN8_LR_CONTEXT_ALIGN,
-				    PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
+	ret = i915_gem_object_ggtt_pin(ce->state, NULL,
+				       0, GEN8_LR_CONTEXT_ALIGN,
+				       PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
 	if (ret)
 		goto err;
 
@@ -979,15 +788,14 @@ static int intel_lr_context_pin(struct i915_gem_context *ctx,
 
 	lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
 
-	ret = intel_pin_and_map_ringbuffer_obj(dev_priv, ce->ringbuf);
+	ret = intel_ring_pin(ce->ring);
 	if (ret)
 		goto unpin_map;
 
-	i915_gem_context_reference(ctx);
 	ce->lrc_vma = i915_gem_obj_to_ggtt(ce->state);
 	intel_lr_context_descriptor_update(ctx, engine);
 
-	lrc_reg_state[CTX_RING_BUFFER_START+1] = ce->ringbuf->vma->node.start;
+	lrc_reg_state[CTX_RING_BUFFER_START+1] = ce->ring->vma->node.start;
 	ce->lrc_reg_state = lrc_reg_state;
 	ce->state->dirty = true;
 
@@ -995,6 +803,7 @@ static int intel_lr_context_pin(struct i915_gem_context *ctx,
 	if (i915.enable_guc_submission)
 		I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
 
+	i915_gem_context_get(ctx);
 	return 0;
 
 unpin_map:
@@ -1017,7 +826,7 @@ void intel_lr_context_unpin(struct i915_gem_context *ctx,
 	if (--ce->pin_count)
 		return;
 
-	intel_unpin_ringbuffer_obj(ce->ringbuf);
+	intel_ring_unpin(ce->ring);
 
 	i915_gem_object_unpin_map(ce->state);
 	i915_gem_object_ggtt_unpin(ce->state);
@@ -1026,21 +835,19 @@ void intel_lr_context_unpin(struct i915_gem_context *ctx,
 	ce->lrc_desc = 0;
 	ce->lrc_reg_state = NULL;
 
-	i915_gem_context_unreference(ctx);
+	i915_gem_context_put(ctx);
 }
 
 static int intel_logical_ring_workarounds_emit(struct drm_i915_gem_request *req)
 {
 	int ret, i;
-	struct intel_engine_cs *engine = req->engine;
-	struct intel_ringbuffer *ringbuf = req->ringbuf;
+	struct intel_ring *ring = req->ring;
 	struct i915_workarounds *w = &req->i915->workarounds;
 
 	if (w->count == 0)
 		return 0;
 
-	engine->gpu_caches_dirty = true;
-	ret = logical_ring_flush_all_caches(req);
+	ret = req->engine->emit_flush(req, EMIT_BARRIER);
 	if (ret)
 		return ret;
 
@@ -1048,17 +855,16 @@ static int intel_logical_ring_workarounds_emit(struct drm_i915_gem_request *req)
 	if (ret)
 		return ret;
 
-	intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(w->count));
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(w->count));
 	for (i = 0; i < w->count; i++) {
-		intel_logical_ring_emit_reg(ringbuf, w->reg[i].addr);
-		intel_logical_ring_emit(ringbuf, w->reg[i].value);
+		intel_ring_emit_reg(ring, w->reg[i].addr);
+		intel_ring_emit(ring, w->reg[i].value);
 	}
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
+	intel_ring_emit(ring, MI_NOOP);
 
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_advance(ring);
 
-	engine->gpu_caches_dirty = true;
-	ret = logical_ring_flush_all_caches(req);
+	ret = req->engine->emit_flush(req, EMIT_BARRIER);
 	if (ret)
 		return ret;
 
@@ -1094,7 +900,7 @@ static int intel_logical_ring_workarounds_emit(struct drm_i915_gem_request *req)
  * code duplication.
  */
 static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine,
-						uint32_t *const batch,
+						uint32_t *batch,
 						uint32_t index)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
@@ -1156,37 +962,24 @@ static inline int wa_ctx_end(struct i915_wa_ctx_bb *wa_ctx,
 	return 0;
 }
 
-/**
- * gen8_init_indirectctx_bb() - initialize indirect ctx batch with WA
- *
- * @engine: only applicable for RCS
- * @wa_ctx: structure representing wa_ctx
- *  offset: specifies start of the batch, should be cache-aligned. This is updated
- *    with the offset value received as input.
- *  size: size of the batch in DWORDS but HW expects in terms of cachelines
- * @batch: page in which WA are loaded
- * @offset: This field specifies the start of the batch, it should be
- *  cache-aligned otherwise it is adjusted accordingly.
- *  Typically we only have one indirect_ctx and per_ctx batch buffer which are
- *  initialized at the beginning and shared across all contexts but this field
- *  helps us to have multiple batches at different offsets and select them based
- *  on a criteria. At the moment this batch always start at the beginning of the page
- *  and at this point we don't have multiple wa_ctx batch buffers.
- *
- *  The number of WA applied are not known at the beginning; we use this field
- *  to return the no of DWORDS written.
+/*
+ * Typically we only have one indirect_ctx and per_ctx batch buffer which are
+ * initialized at the beginning and shared across all contexts but this field
+ * helps us to have multiple batches at different offsets and select them based
+ * on a criteria. At the moment this batch always start at the beginning of the page
+ * and at this point we don't have multiple wa_ctx batch buffers.
  *
- *  It is to be noted that this batch does not contain MI_BATCH_BUFFER_END
- *  so it adds NOOPs as padding to make it cacheline aligned.
- *  MI_BATCH_BUFFER_END will be added to perctx batch and both of them together
- *  makes a complete batch buffer.
+ * The number of WA applied are not known at the beginning; we use this field
+ * to return the no of DWORDS written.
  *
- * Return: non-zero if we exceed the PAGE_SIZE limit.
+ * It is to be noted that this batch does not contain MI_BATCH_BUFFER_END
+ * so it adds NOOPs as padding to make it cacheline aligned.
+ * MI_BATCH_BUFFER_END will be added to perctx batch and both of them together
+ * makes a complete batch buffer.
  */
-
 static int gen8_init_indirectctx_bb(struct intel_engine_cs *engine,
 				    struct i915_wa_ctx_bb *wa_ctx,
-				    uint32_t *const batch,
+				    uint32_t *batch,
 				    uint32_t *offset)
 {
 	uint32_t scratch_addr;
@@ -1230,26 +1023,18 @@ static int gen8_init_indirectctx_bb(struct intel_engine_cs *engine,
 	return wa_ctx_end(wa_ctx, *offset = index, CACHELINE_DWORDS);
 }
 
-/**
- * gen8_init_perctx_bb() - initialize per ctx batch with WA
- *
- * @engine: only applicable for RCS
- * @wa_ctx: structure representing wa_ctx
- *  offset: specifies start of the batch, should be cache-aligned.
- *  size: size of the batch in DWORDS but HW expects in terms of cachelines
- * @batch: page in which WA are loaded
- * @offset: This field specifies the start of this batch.
- *   This batch is started immediately after indirect_ctx batch. Since we ensure
- *   that indirect_ctx ends on a cacheline this batch is aligned automatically.
+/*
+ *  This batch is started immediately after indirect_ctx batch. Since we ensure
+ *  that indirect_ctx ends on a cacheline this batch is aligned automatically.
  *
- *   The number of DWORDS written are returned using this field.
+ *  The number of DWORDS written are returned using this field.
  *
  *  This batch is terminated with MI_BATCH_BUFFER_END and so we need not add padding
  *  to align it with cacheline as padding after MI_BATCH_BUFFER_END is redundant.
  */
 static int gen8_init_perctx_bb(struct intel_engine_cs *engine,
 			       struct i915_wa_ctx_bb *wa_ctx,
-			       uint32_t *const batch,
+			       uint32_t *batch,
 			       uint32_t *offset)
 {
 	uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
@@ -1264,7 +1049,7 @@ static int gen8_init_perctx_bb(struct intel_engine_cs *engine,
 
 static int gen9_init_indirectctx_bb(struct intel_engine_cs *engine,
 				    struct i915_wa_ctx_bb *wa_ctx,
-				    uint32_t *const batch,
+				    uint32_t *batch,
 				    uint32_t *offset)
 {
 	int ret;
@@ -1282,6 +1067,13 @@ static int gen9_init_indirectctx_bb(struct intel_engine_cs *engine,
 		return ret;
 	index = ret;
 
+	/* WaDisableGatherAtSetShaderCommonSlice:skl,bxt,kbl */
+	wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(1));
+	wa_ctx_emit_reg(batch, index, COMMON_SLICE_CHICKEN2);
+	wa_ctx_emit(batch, index, _MASKED_BIT_DISABLE(
+			    GEN9_DISABLE_GATHER_AT_SET_SHADER_COMMON_SLICE));
+	wa_ctx_emit(batch, index, MI_NOOP);
+
 	/* WaClearSlmSpaceAtContextSwitch:kbl */
 	/* Actual scratch location is at 128 bytes offset */
 	if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_A0)) {
@@ -1332,7 +1124,7 @@ static int gen9_init_indirectctx_bb(struct intel_engine_cs *engine,
 
 static int gen9_init_perctx_bb(struct intel_engine_cs *engine,
 			       struct i915_wa_ctx_bb *wa_ctx,
-			       uint32_t *const batch,
+			       uint32_t *batch,
 			       uint32_t *offset)
 {
 	uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
@@ -1389,11 +1181,12 @@ static int lrc_setup_wa_ctx_obj(struct intel_engine_cs *engine, u32 size)
 		return ret;
 	}
 
-	ret = i915_gem_obj_ggtt_pin(engine->wa_ctx.obj, PAGE_SIZE, 0);
+	ret = i915_gem_object_ggtt_pin(engine->wa_ctx.obj, NULL,
+				       0, PAGE_SIZE, 0);
 	if (ret) {
 		DRM_DEBUG_DRIVER("pin LRC WA ctx backing obj failed: %d\n",
 				 ret);
-		drm_gem_object_unreference(&engine->wa_ctx.obj->base);
+		i915_gem_object_put(engine->wa_ctx.obj);
 		return ret;
 	}
 
@@ -1404,7 +1197,7 @@ static void lrc_destroy_wa_ctx_obj(struct intel_engine_cs *engine)
 {
 	if (engine->wa_ctx.obj) {
 		i915_gem_object_ggtt_unpin(engine->wa_ctx.obj);
-		drm_gem_object_unreference(&engine->wa_ctx.obj->base);
+		i915_gem_object_put(engine->wa_ctx.obj);
 		engine->wa_ctx.obj = NULL;
 	}
 }
@@ -1572,8 +1365,8 @@ static int gen9_init_render_ring(struct intel_engine_cs *engine)
 static int intel_logical_ring_emit_pdps(struct drm_i915_gem_request *req)
 {
 	struct i915_hw_ppgtt *ppgtt = req->ctx->ppgtt;
+	struct intel_ring *ring = req->ring;
 	struct intel_engine_cs *engine = req->engine;
-	struct intel_ringbuffer *ringbuf = req->ringbuf;
 	const int num_lri_cmds = GEN8_LEGACY_PDPES * 2;
 	int i, ret;
 
@@ -1581,28 +1374,27 @@ static int intel_logical_ring_emit_pdps(struct drm_i915_gem_request *req)
 	if (ret)
 		return ret;
 
-	intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(num_lri_cmds));
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_lri_cmds));
 	for (i = GEN8_LEGACY_PDPES - 1; i >= 0; i--) {
 		const dma_addr_t pd_daddr = i915_page_dir_dma_addr(ppgtt, i);
 
-		intel_logical_ring_emit_reg(ringbuf,
-					    GEN8_RING_PDP_UDW(engine, i));
-		intel_logical_ring_emit(ringbuf, upper_32_bits(pd_daddr));
-		intel_logical_ring_emit_reg(ringbuf,
-					    GEN8_RING_PDP_LDW(engine, i));
-		intel_logical_ring_emit(ringbuf, lower_32_bits(pd_daddr));
+		intel_ring_emit_reg(ring, GEN8_RING_PDP_UDW(engine, i));
+		intel_ring_emit(ring, upper_32_bits(pd_daddr));
+		intel_ring_emit_reg(ring, GEN8_RING_PDP_LDW(engine, i));
+		intel_ring_emit(ring, lower_32_bits(pd_daddr));
 	}
 
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
-			      u64 offset, unsigned dispatch_flags)
+			      u64 offset, u32 len,
+			      unsigned int dispatch_flags)
 {
-	struct intel_ringbuffer *ringbuf = req->ringbuf;
+	struct intel_ring *ring = req->ring;
 	bool ppgtt = !(dispatch_flags & I915_DISPATCH_SECURE);
 	int ret;
 
@@ -1629,14 +1421,14 @@ static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
 		return ret;
 
 	/* FIXME(BDW): Address space and security selectors. */
-	intel_logical_ring_emit(ringbuf, MI_BATCH_BUFFER_START_GEN8 |
-				(ppgtt<<8) |
-				(dispatch_flags & I915_DISPATCH_RS ?
-				 MI_BATCH_RESOURCE_STREAMER : 0));
-	intel_logical_ring_emit(ringbuf, lower_32_bits(offset));
-	intel_logical_ring_emit(ringbuf, upper_32_bits(offset));
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 |
+			(ppgtt<<8) |
+			(dispatch_flags & I915_DISPATCH_RS ?
+			 MI_BATCH_RESOURCE_STREAMER : 0));
+	intel_ring_emit(ring, lower_32_bits(offset));
+	intel_ring_emit(ring, upper_32_bits(offset));
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -1655,14 +1447,10 @@ static void gen8_logical_ring_disable_irq(struct intel_engine_cs *engine)
 	I915_WRITE_IMR(engine, ~engine->irq_keep_mask);
 }
 
-static int gen8_emit_flush(struct drm_i915_gem_request *request,
-			   u32 invalidate_domains,
-			   u32 unused)
+static int gen8_emit_flush(struct drm_i915_gem_request *request, u32 mode)
 {
-	struct intel_ringbuffer *ringbuf = request->ringbuf;
-	struct intel_engine_cs *engine = ringbuf->engine;
-	struct drm_i915_private *dev_priv = request->i915;
-	uint32_t cmd;
+	struct intel_ring *ring = request->ring;
+	u32 cmd;
 	int ret;
 
 	ret = intel_ring_begin(request, 4);
@@ -1678,29 +1466,28 @@ static int gen8_emit_flush(struct drm_i915_gem_request *request,
 	 */
 	cmd |= MI_FLUSH_DW_STORE_INDEX | MI_FLUSH_DW_OP_STOREDW;
 
-	if (invalidate_domains & I915_GEM_GPU_DOMAINS) {
+	if (mode & EMIT_INVALIDATE) {
 		cmd |= MI_INVALIDATE_TLB;
-		if (engine == &dev_priv->engine[VCS])
+		if (request->engine->id == VCS)
 			cmd |= MI_INVALIDATE_BSD;
 	}
 
-	intel_logical_ring_emit(ringbuf, cmd);
-	intel_logical_ring_emit(ringbuf,
-				I915_GEM_HWS_SCRATCH_ADDR |
-				MI_FLUSH_DW_USE_GTT);
-	intel_logical_ring_emit(ringbuf, 0); /* upper addr */
-	intel_logical_ring_emit(ringbuf, 0); /* value */
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_emit(ring, cmd);
+	intel_ring_emit(ring,
+			I915_GEM_HWS_SCRATCH_ADDR |
+			MI_FLUSH_DW_USE_GTT);
+	intel_ring_emit(ring, 0); /* upper addr */
+	intel_ring_emit(ring, 0); /* value */
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int gen8_emit_flush_render(struct drm_i915_gem_request *request,
-				  u32 invalidate_domains,
-				  u32 flush_domains)
+				  u32 mode)
 {
-	struct intel_ringbuffer *ringbuf = request->ringbuf;
-	struct intel_engine_cs *engine = ringbuf->engine;
+	struct intel_ring *ring = request->ring;
+	struct intel_engine_cs *engine = request->engine;
 	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	bool vf_flush_wa = false, dc_flush_wa = false;
 	u32 flags = 0;
@@ -1709,14 +1496,14 @@ static int gen8_emit_flush_render(struct drm_i915_gem_request *request,
 
 	flags |= PIPE_CONTROL_CS_STALL;
 
-	if (flush_domains) {
+	if (mode & EMIT_FLUSH) {
 		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DC_FLUSH_ENABLE;
 		flags |= PIPE_CONTROL_FLUSH_ENABLE;
 	}
 
-	if (invalidate_domains) {
+	if (mode & EMIT_INVALIDATE) {
 		flags |= PIPE_CONTROL_TLB_INVALIDATE;
 		flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
@@ -1751,40 +1538,40 @@ static int gen8_emit_flush_render(struct drm_i915_gem_request *request,
 		return ret;
 
 	if (vf_flush_wa) {
-		intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
+		intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
 	}
 
 	if (dc_flush_wa) {
-		intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
-		intel_logical_ring_emit(ringbuf, PIPE_CONTROL_DC_FLUSH_ENABLE);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
+		intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+		intel_ring_emit(ring, PIPE_CONTROL_DC_FLUSH_ENABLE);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
 	}
 
-	intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
-	intel_logical_ring_emit(ringbuf, flags);
-	intel_logical_ring_emit(ringbuf, scratch_addr);
-	intel_logical_ring_emit(ringbuf, 0);
-	intel_logical_ring_emit(ringbuf, 0);
-	intel_logical_ring_emit(ringbuf, 0);
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+	intel_ring_emit(ring, flags);
+	intel_ring_emit(ring, scratch_addr);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
 
 	if (dc_flush_wa) {
-		intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
-		intel_logical_ring_emit(ringbuf, PIPE_CONTROL_CS_STALL);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
-		intel_logical_ring_emit(ringbuf, 0);
+		intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+		intel_ring_emit(ring, PIPE_CONTROL_CS_STALL);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, 0);
 	}
 
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -1813,7 +1600,7 @@ static void bxt_a_seqno_barrier(struct intel_engine_cs *engine)
 
 static int gen8_emit_request(struct drm_i915_gem_request *request)
 {
-	struct intel_ringbuffer *ringbuf = request->ringbuf;
+	struct intel_ring *ring = request->ring;
 	int ret;
 
 	ret = intel_ring_begin(request, 6 + WA_TAIL_DWORDS);
@@ -1823,21 +1610,20 @@ static int gen8_emit_request(struct drm_i915_gem_request *request)
 	/* w/a: bit 5 needs to be zero for MI_FLUSH_DW address. */
 	BUILD_BUG_ON(I915_GEM_HWS_INDEX_ADDR & (1 << 5));
 
-	intel_logical_ring_emit(ringbuf,
-				(MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW);
-	intel_logical_ring_emit(ringbuf,
-				intel_hws_seqno_address(request->engine) |
-				MI_FLUSH_DW_USE_GTT);
-	intel_logical_ring_emit(ringbuf, 0);
-	intel_logical_ring_emit(ringbuf, request->seqno);
-	intel_logical_ring_emit(ringbuf, MI_USER_INTERRUPT);
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	return intel_logical_ring_advance_and_submit(request);
+	intel_ring_emit(ring, (MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW);
+	intel_ring_emit(ring,
+			intel_hws_seqno_address(request->engine) |
+			MI_FLUSH_DW_USE_GTT);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, request->fence.seqno);
+	intel_ring_emit(ring, MI_USER_INTERRUPT);
+	intel_ring_emit(ring, MI_NOOP);
+	return intel_logical_ring_advance(request);
 }
 
 static int gen8_emit_request_render(struct drm_i915_gem_request *request)
 {
-	struct intel_ringbuffer *ringbuf = request->ringbuf;
+	struct intel_ring *ring = request->ring;
 	int ret;
 
 	ret = intel_ring_begin(request, 8 + WA_TAIL_DWORDS);
@@ -1851,50 +1637,19 @@ static int gen8_emit_request_render(struct drm_i915_gem_request *request)
 	 * need a prior CS_STALL, which is emitted by the flush
 	 * following the batch.
 	 */
-	intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
-	intel_logical_ring_emit(ringbuf,
-				(PIPE_CONTROL_GLOBAL_GTT_IVB |
-				 PIPE_CONTROL_CS_STALL |
-				 PIPE_CONTROL_QW_WRITE));
-	intel_logical_ring_emit(ringbuf,
-				intel_hws_seqno_address(request->engine));
-	intel_logical_ring_emit(ringbuf, 0);
-	intel_logical_ring_emit(ringbuf, i915_gem_request_get_seqno(request));
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+	intel_ring_emit(ring,
+			(PIPE_CONTROL_GLOBAL_GTT_IVB |
+			 PIPE_CONTROL_CS_STALL |
+			 PIPE_CONTROL_QW_WRITE));
+	intel_ring_emit(ring, intel_hws_seqno_address(request->engine));
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, i915_gem_request_get_seqno(request));
 	/* We're thrashing one dword of HWS. */
-	intel_logical_ring_emit(ringbuf, 0);
-	intel_logical_ring_emit(ringbuf, MI_USER_INTERRUPT);
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	return intel_logical_ring_advance_and_submit(request);
-}
-
-static int intel_lr_context_render_state_init(struct drm_i915_gem_request *req)
-{
-	struct render_state so;
-	int ret;
-
-	ret = i915_gem_render_state_prepare(req->engine, &so);
-	if (ret)
-		return ret;
-
-	if (so.rodata == NULL)
-		return 0;
-
-	ret = req->engine->emit_bb_start(req, so.ggtt_offset,
-				       I915_DISPATCH_SECURE);
-	if (ret)
-		goto out;
-
-	ret = req->engine->emit_bb_start(req,
-				       (so.ggtt_offset + so.aux_batch_offset),
-				       I915_DISPATCH_SECURE);
-	if (ret)
-		goto out;
-
-	i915_vma_move_to_active(i915_gem_obj_to_ggtt(so.obj), req);
-
-out:
-	i915_gem_render_state_fini(&so);
-	return ret;
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, MI_USER_INTERRUPT);
+	intel_ring_emit(ring, MI_NOOP);
+	return intel_logical_ring_advance(request);
 }
 
 static int gen8_init_rcs_context(struct drm_i915_gem_request *req)
@@ -1913,14 +1668,12 @@ static int gen8_init_rcs_context(struct drm_i915_gem_request *req)
 	if (ret)
 		DRM_ERROR("MOCS failed to program: expect performance issues.\n");
 
-	return intel_lr_context_render_state_init(req);
+	return i915_gem_render_state_init(req);
 }
 
 /**
  * intel_logical_ring_cleanup() - deallocate the Engine Command Streamer
- *
  * @engine: Engine Command Streamer.
- *
  */
 void intel_logical_ring_cleanup(struct intel_engine_cs *engine)
 {
@@ -1939,17 +1692,13 @@ void intel_logical_ring_cleanup(struct intel_engine_cs *engine)
 	dev_priv = engine->i915;
 
 	if (engine->buffer) {
-		intel_logical_ring_stop(engine);
 		WARN_ON((I915_READ_MODE(engine) & MODE_IDLE) == 0);
 	}
 
 	if (engine->cleanup)
 		engine->cleanup(engine);
 
-	i915_cmd_parser_fini_ring(engine);
-	i915_gem_batch_pool_fini(&engine->batch_pool);
-
-	intel_engine_fini_breadcrumbs(engine);
+	intel_engine_cleanup_common(engine);
 
 	if (engine->status_page.obj) {
 		i915_gem_object_unpin_map(engine->status_page.obj);
@@ -1965,13 +1714,23 @@ void intel_logical_ring_cleanup(struct intel_engine_cs *engine)
 	engine->i915 = NULL;
 }
 
+void intel_execlists_enable_submission(struct drm_i915_private *dev_priv)
+{
+	struct intel_engine_cs *engine;
+
+	for_each_engine(engine, dev_priv)
+		engine->submit_request = execlists_submit_request;
+}
+
 static void
 logical_ring_default_vfuncs(struct intel_engine_cs *engine)
 {
 	/* Default vfuncs which can be overriden by each engine. */
 	engine->init_hw = gen8_init_common_ring;
-	engine->emit_request = gen8_emit_request;
 	engine->emit_flush = gen8_emit_flush;
+	engine->emit_request = gen8_emit_request;
+	engine->submit_request = execlists_submit_request;
+
 	engine->irq_enable = gen8_logical_ring_enable_irq;
 	engine->irq_disable = gen8_logical_ring_disable_irq;
 	engine->emit_bb_start = gen8_emit_bb_start;
@@ -1980,8 +1739,9 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine)
 }
 
 static inline void
-logical_ring_default_irqs(struct intel_engine_cs *engine, unsigned shift)
+logical_ring_default_irqs(struct intel_engine_cs *engine)
 {
+	unsigned shift = engine->irq_shift;
 	engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT << shift;
 	engine->irq_keep_mask = GT_CONTEXT_SWITCH_INTERRUPT << shift;
 }
@@ -2004,17 +1764,46 @@ lrc_setup_hws(struct intel_engine_cs *engine,
 	return 0;
 }
 
+static void
+logical_ring_setup(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *dev_priv = engine->i915;
+	enum forcewake_domains fw_domains;
+
+	intel_engine_setup_common(engine);
+
+	/* Intentionally left blank. */
+	engine->buffer = NULL;
+
+	fw_domains = intel_uncore_forcewake_for_reg(dev_priv,
+						    RING_ELSP(engine),
+						    FW_REG_WRITE);
+
+	fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
+						     RING_CONTEXT_STATUS_PTR(engine),
+						     FW_REG_READ | FW_REG_WRITE);
+
+	fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
+						     RING_CONTEXT_STATUS_BUF_BASE(engine),
+						     FW_REG_READ);
+
+	engine->fw_domains = fw_domains;
+
+	tasklet_init(&engine->irq_tasklet,
+		     intel_lrc_irq_handler, (unsigned long)engine);
+
+	logical_ring_init_platform_invariants(engine);
+	logical_ring_default_vfuncs(engine);
+	logical_ring_default_irqs(engine);
+}
+
 static int
 logical_ring_init(struct intel_engine_cs *engine)
 {
 	struct i915_gem_context *dctx = engine->i915->kernel_context;
 	int ret;
 
-	ret = intel_engine_init_breadcrumbs(engine);
-	if (ret)
-		goto error;
-
-	ret = i915_cmd_parser_init_ring(engine);
+	ret = intel_engine_init_common(engine);
 	if (ret)
 		goto error;
 
@@ -2044,11 +1833,13 @@ error:
 	return ret;
 }
 
-static int logical_render_ring_init(struct intel_engine_cs *engine)
+int logical_render_ring_init(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
 	int ret;
 
+	logical_ring_setup(engine);
+
 	if (HAS_L3_DPF(dev_priv))
 		engine->irq_keep_mask |= GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
 
@@ -2085,160 +1876,11 @@ static int logical_render_ring_init(struct intel_engine_cs *engine)
 	return ret;
 }
 
-static const struct logical_ring_info {
-	const char *name;
-	unsigned exec_id;
-	unsigned guc_id;
-	u32 mmio_base;
-	unsigned irq_shift;
-	int (*init)(struct intel_engine_cs *engine);
-} logical_rings[] = {
-	[RCS] = {
-		.name = "render ring",
-		.exec_id = I915_EXEC_RENDER,
-		.guc_id = GUC_RENDER_ENGINE,
-		.mmio_base = RENDER_RING_BASE,
-		.irq_shift = GEN8_RCS_IRQ_SHIFT,
-		.init = logical_render_ring_init,
-	},
-	[BCS] = {
-		.name = "blitter ring",
-		.exec_id = I915_EXEC_BLT,
-		.guc_id = GUC_BLITTER_ENGINE,
-		.mmio_base = BLT_RING_BASE,
-		.irq_shift = GEN8_BCS_IRQ_SHIFT,
-		.init = logical_ring_init,
-	},
-	[VCS] = {
-		.name = "bsd ring",
-		.exec_id = I915_EXEC_BSD,
-		.guc_id = GUC_VIDEO_ENGINE,
-		.mmio_base = GEN6_BSD_RING_BASE,
-		.irq_shift = GEN8_VCS1_IRQ_SHIFT,
-		.init = logical_ring_init,
-	},
-	[VCS2] = {
-		.name = "bsd2 ring",
-		.exec_id = I915_EXEC_BSD,
-		.guc_id = GUC_VIDEO_ENGINE2,
-		.mmio_base = GEN8_BSD2_RING_BASE,
-		.irq_shift = GEN8_VCS2_IRQ_SHIFT,
-		.init = logical_ring_init,
-	},
-	[VECS] = {
-		.name = "video enhancement ring",
-		.exec_id = I915_EXEC_VEBOX,
-		.guc_id = GUC_VIDEOENHANCE_ENGINE,
-		.mmio_base = VEBOX_RING_BASE,
-		.irq_shift = GEN8_VECS_IRQ_SHIFT,
-		.init = logical_ring_init,
-	},
-};
-
-static struct intel_engine_cs *
-logical_ring_setup(struct drm_i915_private *dev_priv, enum intel_engine_id id)
+int logical_xcs_ring_init(struct intel_engine_cs *engine)
 {
-	const struct logical_ring_info *info = &logical_rings[id];
-	struct intel_engine_cs *engine = &dev_priv->engine[id];
-	enum forcewake_domains fw_domains;
+	logical_ring_setup(engine);
 
-	engine->id = id;
-	engine->name = info->name;
-	engine->exec_id = info->exec_id;
-	engine->guc_id = info->guc_id;
-	engine->mmio_base = info->mmio_base;
-
-	engine->i915 = dev_priv;
-
-	/* Intentionally left blank. */
-	engine->buffer = NULL;
-
-	fw_domains = intel_uncore_forcewake_for_reg(dev_priv,
-						    RING_ELSP(engine),
-						    FW_REG_WRITE);
-
-	fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
-						     RING_CONTEXT_STATUS_PTR(engine),
-						     FW_REG_READ | FW_REG_WRITE);
-
-	fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
-						     RING_CONTEXT_STATUS_BUF_BASE(engine),
-						     FW_REG_READ);
-
-	engine->fw_domains = fw_domains;
-
-	INIT_LIST_HEAD(&engine->active_list);
-	INIT_LIST_HEAD(&engine->request_list);
-	INIT_LIST_HEAD(&engine->buffers);
-	INIT_LIST_HEAD(&engine->execlist_queue);
-	spin_lock_init(&engine->execlist_lock);
-
-	tasklet_init(&engine->irq_tasklet,
-		     intel_lrc_irq_handler, (unsigned long)engine);
-
-	logical_ring_init_platform_invariants(engine);
-	logical_ring_default_vfuncs(engine);
-	logical_ring_default_irqs(engine, info->irq_shift);
-
-	intel_engine_init_hangcheck(engine);
-	i915_gem_batch_pool_init(&dev_priv->drm, &engine->batch_pool);
-
-	return engine;
-}
-
-/**
- * intel_logical_rings_init() - allocate, populate and init the Engine Command Streamers
- * @dev: DRM device.
- *
- * This function inits the engines for an Execlists submission style (the
- * equivalent in the legacy ringbuffer submission world would be
- * i915_gem_init_engines). It does it only for those engines that are present in
- * the hardware.
- *
- * Return: non-zero if the initialization failed.
- */
-int intel_logical_rings_init(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	unsigned int mask = 0;
-	unsigned int i;
-	int ret;
-
-	WARN_ON(INTEL_INFO(dev_priv)->ring_mask &
-		GENMASK(sizeof(mask) * BITS_PER_BYTE - 1, I915_NUM_ENGINES));
-
-	for (i = 0; i < ARRAY_SIZE(logical_rings); i++) {
-		if (!HAS_ENGINE(dev_priv, i))
-			continue;
-
-		if (!logical_rings[i].init)
-			continue;
-
-		ret = logical_rings[i].init(logical_ring_setup(dev_priv, i));
-		if (ret)
-			goto cleanup;
-
-		mask |= ENGINE_MASK(i);
-	}
-
-	/*
-	 * Catch failures to update logical_rings table when the new engines
-	 * are added to the driver by a warning and disabling the forgotten
-	 * engines.
-	 */
-	if (WARN_ON(mask != INTEL_INFO(dev_priv)->ring_mask)) {
-		struct intel_device_info *info =
-			(struct intel_device_info *)&dev_priv->info;
-		info->ring_mask = mask;
-	}
-
-	return 0;
-
-cleanup:
-	for (i = 0; i < I915_NUM_ENGINES; i++)
-		intel_logical_ring_cleanup(&dev_priv->engine[i]);
-
-	return ret;
+	return logical_ring_init(engine);
 }
 
 static u32
@@ -2309,7 +1951,7 @@ static int
 populate_lr_context(struct i915_gem_context *ctx,
 		    struct drm_i915_gem_object *ctx_obj,
 		    struct intel_engine_cs *engine,
-		    struct intel_ringbuffer *ringbuf)
+		    struct intel_ring *ring)
 {
 	struct drm_i915_private *dev_priv = ctx->i915;
 	struct i915_hw_ppgtt *ppgtt = ctx->ppgtt;
@@ -2362,7 +2004,7 @@ populate_lr_context(struct i915_gem_context *ctx,
 		       RING_START(engine->mmio_base), 0);
 	ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_CONTROL,
 		       RING_CTL(engine->mmio_base),
-		       ((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES) | RING_VALID);
+		       ((ring->size - PAGE_SIZE) & RING_NR_PAGES) | RING_VALID);
 	ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_U,
 		       RING_BBADDR_UDW(engine->mmio_base), 0);
 	ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_L,
@@ -2484,26 +2126,13 @@ uint32_t intel_lr_context_size(struct intel_engine_cs *engine)
 	return ret;
 }
 
-/**
- * execlists_context_deferred_alloc() - create the LRC specific bits of a context
- * @ctx: LR context to create.
- * @engine: engine to be used with the context.
- *
- * This function can be called more than once, with different engines, if we plan
- * to use the context with them. The context backing objects and the ringbuffers
- * (specially the ringbuffer backing objects) suck a lot of memory up, and that's why
- * the creation is a deferred call: it's better to make sure first that we need to use
- * a given ring with the context.
- *
- * Return: non-zero on error.
- */
 static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
 					    struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_object *ctx_obj;
 	struct intel_context *ce = &ctx->engine[engine->id];
 	uint32_t context_size;
-	struct intel_ringbuffer *ringbuf;
+	struct intel_ring *ring;
 	int ret;
 
 	WARN_ON(ce->state);
@@ -2519,29 +2148,29 @@ static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
 		return PTR_ERR(ctx_obj);
 	}
 
-	ringbuf = intel_engine_create_ringbuffer(engine, ctx->ring_size);
-	if (IS_ERR(ringbuf)) {
-		ret = PTR_ERR(ringbuf);
+	ring = intel_engine_create_ring(engine, ctx->ring_size);
+	if (IS_ERR(ring)) {
+		ret = PTR_ERR(ring);
 		goto error_deref_obj;
 	}
 
-	ret = populate_lr_context(ctx, ctx_obj, engine, ringbuf);
+	ret = populate_lr_context(ctx, ctx_obj, engine, ring);
 	if (ret) {
 		DRM_DEBUG_DRIVER("Failed to populate LRC: %d\n", ret);
-		goto error_ringbuf;
+		goto error_ring_free;
 	}
 
-	ce->ringbuf = ringbuf;
+	ce->ring = ring;
 	ce->state = ctx_obj;
 	ce->initialised = engine->init_context == NULL;
 
 	return 0;
 
-error_ringbuf:
-	intel_ringbuffer_free(ringbuf);
+error_ring_free:
+	intel_ring_free(ring);
 error_deref_obj:
-	drm_gem_object_unreference(&ctx_obj->base);
-	ce->ringbuf = NULL;
+	i915_gem_object_put(ctx_obj);
+	ce->ring = NULL;
 	ce->state = NULL;
 	return ret;
 }
@@ -2572,7 +2201,7 @@ void intel_lr_context_reset(struct drm_i915_private *dev_priv,
 
 		i915_gem_object_unpin_map(ctx_obj);
 
-		ce->ringbuf->head = 0;
-		ce->ringbuf->tail = 0;
+		ce->ring->head = 0;
+		ce->ring->tail = 0;
 	}
 }

drivers/gpu/drm/i915/intel_lrc.h

@@ -29,17 +29,17 @@
 #define GEN8_LR_CONTEXT_ALIGN 4096
 
 /* Execlists regs */
-#define RING_ELSP(ring)				_MMIO((ring)->mmio_base + 0x230)
-#define RING_EXECLIST_STATUS_LO(ring)		_MMIO((ring)->mmio_base + 0x234)
-#define RING_EXECLIST_STATUS_HI(ring)		_MMIO((ring)->mmio_base + 0x234 + 4)
-#define RING_CONTEXT_CONTROL(ring)		_MMIO((ring)->mmio_base + 0x244)
+#define RING_ELSP(engine)			_MMIO((engine)->mmio_base + 0x230)
+#define RING_EXECLIST_STATUS_LO(engine)		_MMIO((engine)->mmio_base + 0x234)
+#define RING_EXECLIST_STATUS_HI(engine)		_MMIO((engine)->mmio_base + 0x234 + 4)
+#define RING_CONTEXT_CONTROL(engine)		_MMIO((engine)->mmio_base + 0x244)
 #define	  CTX_CTRL_INHIBIT_SYN_CTX_SWITCH	(1 << 3)
 #define	  CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT	(1 << 0)
 #define   CTX_CTRL_RS_CTX_ENABLE                (1 << 1)
-#define RING_CONTEXT_STATUS_BUF_BASE(ring)	_MMIO((ring)->mmio_base + 0x370)
-#define RING_CONTEXT_STATUS_BUF_LO(ring, i)	_MMIO((ring)->mmio_base + 0x370 + (i) * 8)
-#define RING_CONTEXT_STATUS_BUF_HI(ring, i)	_MMIO((ring)->mmio_base + 0x370 + (i) * 8 + 4)
-#define RING_CONTEXT_STATUS_PTR(ring)		_MMIO((ring)->mmio_base + 0x3a0)
+#define RING_CONTEXT_STATUS_BUF_BASE(engine)	_MMIO((engine)->mmio_base + 0x370)
+#define RING_CONTEXT_STATUS_BUF_LO(engine, i)	_MMIO((engine)->mmio_base + 0x370 + (i) * 8)
+#define RING_CONTEXT_STATUS_BUF_HI(engine, i)	_MMIO((engine)->mmio_base + 0x370 + (i) * 8 + 4)
+#define RING_CONTEXT_STATUS_PTR(engine)		_MMIO((engine)->mmio_base + 0x3a0)
 
 /* The docs specify that the write pointer wraps around after 5h, "After status
  * is written out to the last available status QW at offset 5h, this pointer
@@ -67,35 +67,10 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request
 int intel_logical_ring_reserve_space(struct drm_i915_gem_request *request);
 void intel_logical_ring_stop(struct intel_engine_cs *engine);
 void intel_logical_ring_cleanup(struct intel_engine_cs *engine);
-int intel_logical_rings_init(struct drm_device *dev);
+int logical_render_ring_init(struct intel_engine_cs *engine);
+int logical_xcs_ring_init(struct intel_engine_cs *engine);
 
-int logical_ring_flush_all_caches(struct drm_i915_gem_request *req);
-/**
- * intel_logical_ring_advance() - advance the ringbuffer tail
- * @ringbuf: Ringbuffer to advance.
- *
- * The tail is only updated in our logical ringbuffer struct.
- */
-static inline void intel_logical_ring_advance(struct intel_ringbuffer *ringbuf)
-{
-	ringbuf->tail &= ringbuf->size - 1;
-}
-/**
- * intel_logical_ring_emit() - write a DWORD to the ringbuffer.
- * @ringbuf: Ringbuffer to write to.
- * @data: DWORD to write.
- */
-static inline void intel_logical_ring_emit(struct intel_ringbuffer *ringbuf,
-					   u32 data)
-{
-	iowrite32(data, ringbuf->virtual_start + ringbuf->tail);
-	ringbuf->tail += 4;
-}
-static inline void intel_logical_ring_emit_reg(struct intel_ringbuffer *ringbuf,
-					       i915_reg_t reg)
-{
-	intel_logical_ring_emit(ringbuf, i915_mmio_reg_offset(reg));
-}
+int intel_engines_init(struct drm_device *dev);
 
 /* Logical Ring Contexts */
 
@@ -120,10 +95,7 @@ uint64_t intel_lr_context_descriptor(struct i915_gem_context *ctx,
 /* Execlists */
 int intel_sanitize_enable_execlists(struct drm_i915_private *dev_priv,
 				    int enable_execlists);
-struct i915_execbuffer_params;
-int intel_execlists_submission(struct i915_execbuffer_params *params,
-			       struct drm_i915_gem_execbuffer2 *args,
-			       struct list_head *vmas);
+void intel_execlists_enable_submission(struct drm_i915_private *dev_priv);
 
 void intel_execlists_cancel_requests(struct intel_engine_cs *engine);
 

drivers/gpu/drm/i915/intel_mocs.c

@@ -97,7 +97,8 @@ struct drm_i915_mocs_table {
  *       end.
  */
 static const struct drm_i915_mocs_entry skylake_mocs_table[] = {
-	{ /* 0x00000009 */
+	[I915_MOCS_UNCACHED] = {
+	  /* 0x00000009 */
 	  .control_value = LE_CACHEABILITY(LE_UC) |
 			   LE_TGT_CACHE(LE_TC_LLC_ELLC) |
 			   LE_LRUM(0) | LE_AOM(0) | LE_RSC(0) | LE_SCC(0) |
@@ -106,7 +107,7 @@ static const struct drm_i915_mocs_entry skylake_mocs_table[] = {
 	  /* 0x0010 */
 	  .l3cc_value =    L3_ESC(0) | L3_SCC(0) | L3_CACHEABILITY(L3_UC),
 	},
-	{
+	[I915_MOCS_PTE] = {
 	  /* 0x00000038 */
 	  .control_value = LE_CACHEABILITY(LE_PAGETABLE) |
 			   LE_TGT_CACHE(LE_TC_LLC_ELLC) |
@@ -115,7 +116,7 @@ static const struct drm_i915_mocs_entry skylake_mocs_table[] = {
 	  /* 0x0030 */
 	  .l3cc_value =    L3_ESC(0) | L3_SCC(0) | L3_CACHEABILITY(L3_WB),
 	},
-	{
+	[I915_MOCS_CACHED] = {
 	  /* 0x0000003b */
 	  .control_value = LE_CACHEABILITY(LE_WB) |
 			   LE_TGT_CACHE(LE_TC_LLC_ELLC) |
@@ -128,7 +129,7 @@ static const struct drm_i915_mocs_entry skylake_mocs_table[] = {
 
 /* NOTE: the LE_TGT_CACHE is not used on Broxton */
 static const struct drm_i915_mocs_entry broxton_mocs_table[] = {
-	{
+	[I915_MOCS_UNCACHED] = {
 	  /* 0x00000009 */
 	  .control_value = LE_CACHEABILITY(LE_UC) |
 			   LE_TGT_CACHE(LE_TC_LLC_ELLC) |
@@ -138,7 +139,7 @@ static const struct drm_i915_mocs_entry broxton_mocs_table[] = {
 	  /* 0x0010 */
 	  .l3cc_value =    L3_ESC(0) | L3_SCC(0) | L3_CACHEABILITY(L3_UC),
 	},
-	{
+	[I915_MOCS_PTE] = {
 	  /* 0x00000038 */
 	  .control_value = LE_CACHEABILITY(LE_PAGETABLE) |
 			   LE_TGT_CACHE(LE_TC_LLC_ELLC) |
@@ -148,7 +149,7 @@ static const struct drm_i915_mocs_entry broxton_mocs_table[] = {
 	  /* 0x0030 */
 	  .l3cc_value =    L3_ESC(0) | L3_SCC(0) | L3_CACHEABILITY(L3_WB),
 	},
-	{
+	[I915_MOCS_CACHED] = {
 	  /* 0x00000039 */
 	  .control_value = LE_CACHEABILITY(LE_UC) |
 			   LE_TGT_CACHE(LE_TC_LLC_ELLC) |
@@ -203,9 +204,9 @@ static bool get_mocs_settings(struct drm_i915_private *dev_priv,
 	return result;
 }
 
-static i915_reg_t mocs_register(enum intel_engine_id ring, int index)
+static i915_reg_t mocs_register(enum intel_engine_id engine_id, int index)
 {
-	switch (ring) {
+	switch (engine_id) {
 	case RCS:
 		return GEN9_GFX_MOCS(index);
 	case VCS:
@@ -217,7 +218,7 @@ static i915_reg_t mocs_register(enum intel_engine_id ring, int index)
 	case VCS2:
 		return GEN9_MFX1_MOCS(index);
 	default:
-		MISSING_CASE(ring);
+		MISSING_CASE(engine_id);
 		return INVALID_MMIO_REG;
 	}
 }
@@ -275,7 +276,7 @@ int intel_mocs_init_engine(struct intel_engine_cs *engine)
 static int emit_mocs_control_table(struct drm_i915_gem_request *req,
 				   const struct drm_i915_mocs_table *table)
 {
-	struct intel_ringbuffer *ringbuf = req->ringbuf;
+	struct intel_ring *ring = req->ring;
 	enum intel_engine_id engine = req->engine->id;
 	unsigned int index;
 	int ret;
@@ -287,14 +288,11 @@ static int emit_mocs_control_table(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_logical_ring_emit(ringbuf,
-				MI_LOAD_REGISTER_IMM(GEN9_NUM_MOCS_ENTRIES));
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(GEN9_NUM_MOCS_ENTRIES));
 
 	for (index = 0; index < table->size; index++) {
-		intel_logical_ring_emit_reg(ringbuf,
-					    mocs_register(engine, index));
-		intel_logical_ring_emit(ringbuf,
-					table->table[index].control_value);
+		intel_ring_emit_reg(ring, mocs_register(engine, index));
+		intel_ring_emit(ring, table->table[index].control_value);
 	}
 
 	/*
@@ -306,14 +304,12 @@ static int emit_mocs_control_table(struct drm_i915_gem_request *req,
 	 * that value to all the used entries.
 	 */
 	for (; index < GEN9_NUM_MOCS_ENTRIES; index++) {
-		intel_logical_ring_emit_reg(ringbuf,
-					    mocs_register(engine, index));
-		intel_logical_ring_emit(ringbuf,
-					table->table[0].control_value);
+		intel_ring_emit_reg(ring, mocs_register(engine, index));
+		intel_ring_emit(ring, table->table[0].control_value);
 	}
 
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -340,7 +336,7 @@ static inline u32 l3cc_combine(const struct drm_i915_mocs_table *table,
 static int emit_mocs_l3cc_table(struct drm_i915_gem_request *req,
 				const struct drm_i915_mocs_table *table)
 {
-	struct intel_ringbuffer *ringbuf = req->ringbuf;
+	struct intel_ring *ring = req->ring;
 	unsigned int i;
 	int ret;
 
@@ -351,19 +347,18 @@ static int emit_mocs_l3cc_table(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_logical_ring_emit(ringbuf,
+	intel_ring_emit(ring,
 			MI_LOAD_REGISTER_IMM(GEN9_NUM_MOCS_ENTRIES / 2));
 
 	for (i = 0; i < table->size/2; i++) {
-		intel_logical_ring_emit_reg(ringbuf, GEN9_LNCFCMOCS(i));
-		intel_logical_ring_emit(ringbuf,
-					l3cc_combine(table, 2*i, 2*i+1));
+		intel_ring_emit_reg(ring, GEN9_LNCFCMOCS(i));
+		intel_ring_emit(ring, l3cc_combine(table, 2*i, 2*i+1));
 	}
 
 	if (table->size & 0x01) {
 		/* Odd table size - 1 left over */
-		intel_logical_ring_emit_reg(ringbuf, GEN9_LNCFCMOCS(i));
-		intel_logical_ring_emit(ringbuf, l3cc_combine(table, 2*i, 0));
+		intel_ring_emit_reg(ring, GEN9_LNCFCMOCS(i));
+		intel_ring_emit(ring, l3cc_combine(table, 2*i, 0));
 		i++;
 	}
 
@@ -373,12 +368,12 @@ static int emit_mocs_l3cc_table(struct drm_i915_gem_request *req,
 	 * they are reserved by the hardware.
 	 */
 	for (; i < GEN9_NUM_MOCS_ENTRIES / 2; i++) {
-		intel_logical_ring_emit_reg(ringbuf, GEN9_LNCFCMOCS(i));
-		intel_logical_ring_emit(ringbuf, l3cc_combine(table, 0, 0));
+		intel_ring_emit_reg(ring, GEN9_LNCFCMOCS(i));
+		intel_ring_emit(ring, l3cc_combine(table, 0, 0));
 	}
 
-	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	intel_logical_ring_advance(ringbuf);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }

drivers/gpu/drm/i915/intel_mocs.h

@@ -54,6 +54,6 @@
 
 int intel_rcs_context_init_mocs(struct drm_i915_gem_request *req);
 void intel_mocs_init_l3cc_table(struct drm_device *dev);
-int intel_mocs_init_engine(struct intel_engine_cs *ring);
+int intel_mocs_init_engine(struct intel_engine_cs *engine);
 
 #endif

drivers/gpu/drm/i915/intel_overlay.c

@@ -30,6 +30,7 @@
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_drv.h"
+#include "intel_frontbuffer.h"
 
 /* Limits for overlay size. According to intel doc, the real limits are:
  * Y width: 4095, UV width (planar): 2047, Y height: 2047,
@@ -183,8 +184,7 @@ struct intel_overlay {
 	u32 flip_addr;
 	struct drm_i915_gem_object *reg_bo;
 	/* flip handling */
-	struct drm_i915_gem_request *last_flip_req;
-	void (*flip_tail)(struct intel_overlay *);
+	struct i915_gem_active last_flip;
 };
 
 static struct overlay_registers __iomem *
@@ -210,37 +210,46 @@ static void intel_overlay_unmap_regs(struct intel_overlay *overlay,
 		io_mapping_unmap(regs);
 }
 
-static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
+static void intel_overlay_submit_request(struct intel_overlay *overlay,
 					 struct drm_i915_gem_request *req,
-					 void (*tail)(struct intel_overlay *))
+					 i915_gem_retire_fn retire)
 {
-	int ret;
-
-	WARN_ON(overlay->last_flip_req);
-	i915_gem_request_assign(&overlay->last_flip_req, req);
+	GEM_BUG_ON(i915_gem_active_peek(&overlay->last_flip,
+					&overlay->i915->drm.struct_mutex));
+	overlay->last_flip.retire = retire;
+	i915_gem_active_set(&overlay->last_flip, req);
 	i915_add_request(req);
+}
 
-	overlay->flip_tail = tail;
-	ret = i915_wait_request(overlay->last_flip_req);
-	if (ret)
-		return ret;
+static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
+					 struct drm_i915_gem_request *req,
+					 i915_gem_retire_fn retire)
+{
+	intel_overlay_submit_request(overlay, req, retire);
+	return i915_gem_active_retire(&overlay->last_flip,
+				      &overlay->i915->drm.struct_mutex);
+}
 
-	i915_gem_request_assign(&overlay->last_flip_req, NULL);
-	return 0;
+static struct drm_i915_gem_request *alloc_request(struct intel_overlay *overlay)
+{
+	struct drm_i915_private *dev_priv = overlay->i915;
+	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
+
+	return i915_gem_request_alloc(engine, dev_priv->kernel_context);
 }
 
 /* overlay needs to be disable in OCMD reg */
 static int intel_overlay_on(struct intel_overlay *overlay)
 {
 	struct drm_i915_private *dev_priv = overlay->i915;
-	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	struct drm_i915_gem_request *req;
+	struct intel_ring *ring;
 	int ret;
 
 	WARN_ON(overlay->active);
 	WARN_ON(IS_I830(dev_priv) && !(dev_priv->quirks & QUIRK_PIPEA_FORCE));
 
-	req = i915_gem_request_alloc(engine, NULL);
+	req = alloc_request(overlay);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
@@ -252,11 +261,12 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 
 	overlay->active = true;
 
-	intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_ON);
-	intel_ring_emit(engine, overlay->flip_addr | OFC_UPDATE);
-	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	ring = req->ring;
+	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_ON);
+	intel_ring_emit(ring, overlay->flip_addr | OFC_UPDATE);
+	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return intel_overlay_do_wait_request(overlay, req, NULL);
 }
@@ -266,8 +276,8 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 				  bool load_polyphase_filter)
 {
 	struct drm_i915_private *dev_priv = overlay->i915;
-	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	struct drm_i915_gem_request *req;
+	struct intel_ring *ring;
 	u32 flip_addr = overlay->flip_addr;
 	u32 tmp;
 	int ret;
@@ -282,7 +292,7 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 	if (tmp & (1 << 17))
 		DRM_DEBUG("overlay underrun, DOVSTA: %x\n", tmp);
 
-	req = i915_gem_request_alloc(engine, NULL);
+	req = alloc_request(overlay);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
@@ -292,29 +302,37 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 		return ret;
 	}
 
-	intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
-	intel_ring_emit(engine, flip_addr);
-	intel_ring_advance(engine);
+	ring = req->ring;
+	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
+	intel_ring_emit(ring, flip_addr);
+	intel_ring_advance(ring);
 
-	WARN_ON(overlay->last_flip_req);
-	i915_gem_request_assign(&overlay->last_flip_req, req);
-	i915_add_request(req);
+	intel_overlay_submit_request(overlay, req, NULL);
 
 	return 0;
 }
 
-static void intel_overlay_release_old_vid_tail(struct intel_overlay *overlay)
+static void intel_overlay_release_old_vid_tail(struct i915_gem_active *active,
+					       struct drm_i915_gem_request *req)
 {
+	struct intel_overlay *overlay =
+		container_of(active, typeof(*overlay), last_flip);
 	struct drm_i915_gem_object *obj = overlay->old_vid_bo;
 
+	i915_gem_track_fb(obj, NULL,
+			  INTEL_FRONTBUFFER_OVERLAY(overlay->crtc->pipe));
+
 	i915_gem_object_ggtt_unpin(obj);
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 
 	overlay->old_vid_bo = NULL;
 }
 
-static void intel_overlay_off_tail(struct intel_overlay *overlay)
+static void intel_overlay_off_tail(struct i915_gem_active *active,
+				   struct drm_i915_gem_request *req)
 {
+	struct intel_overlay *overlay =
+		container_of(active, typeof(*overlay), last_flip);
 	struct drm_i915_gem_object *obj = overlay->vid_bo;
 
 	/* never have the overlay hw on without showing a frame */
@@ -322,7 +340,7 @@ static void intel_overlay_off_tail(struct intel_overlay *overlay)
 		return;
 
 	i915_gem_object_ggtt_unpin(obj);
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	overlay->vid_bo = NULL;
 
 	overlay->crtc->overlay = NULL;
@@ -334,8 +352,8 @@ static void intel_overlay_off_tail(struct intel_overlay *overlay)
 static int intel_overlay_off(struct intel_overlay *overlay)
 {
 	struct drm_i915_private *dev_priv = overlay->i915;
-	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	struct drm_i915_gem_request *req;
+	struct intel_ring *ring;
 	u32 flip_addr = overlay->flip_addr;
 	int ret;
 
@@ -347,7 +365,7 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 	 * of the hw. Do it in both cases */
 	flip_addr |= OFC_UPDATE;
 
-	req = i915_gem_request_alloc(engine, NULL);
+	req = alloc_request(overlay);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
@@ -357,46 +375,36 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 		return ret;
 	}
 
+	ring = req->ring;
 	/* wait for overlay to go idle */
-	intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
-	intel_ring_emit(engine, flip_addr);
-	intel_ring_emit(engine, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
+	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
+	intel_ring_emit(ring, flip_addr);
+	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
 	/* turn overlay off */
 	if (IS_I830(dev_priv)) {
 		/* Workaround: Don't disable the overlay fully, since otherwise
 		 * it dies on the next OVERLAY_ON cmd. */
-		intel_ring_emit(engine, MI_NOOP);
-		intel_ring_emit(engine, MI_NOOP);
-		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(ring, MI_NOOP);
 	} else {
-		intel_ring_emit(engine, MI_OVERLAY_FLIP | MI_OVERLAY_OFF);
-		intel_ring_emit(engine, flip_addr);
-		intel_ring_emit(engine,
+		intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_OFF);
+		intel_ring_emit(ring, flip_addr);
+		intel_ring_emit(ring,
 				MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
 	}
-	intel_ring_advance(engine);
+	intel_ring_advance(ring);
 
-	return intel_overlay_do_wait_request(overlay, req, intel_overlay_off_tail);
+	return intel_overlay_do_wait_request(overlay, req,
+					     intel_overlay_off_tail);
 }
 
 /* recover from an interruption due to a signal
  * We have to be careful not to repeat work forever an make forward progess. */
 static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay)
 {
-	int ret;
-
-	if (overlay->last_flip_req == NULL)
-		return 0;
-
-	ret = i915_wait_request(overlay->last_flip_req);
-	if (ret)
-		return ret;
-
-	if (overlay->flip_tail)
-		overlay->flip_tail(overlay);
-
-	i915_gem_request_assign(&overlay->last_flip_req, NULL);
-	return 0;
+	return i915_gem_active_retire(&overlay->last_flip,
+				      &overlay->i915->drm.struct_mutex);
 }
 
 /* Wait for pending overlay flip and release old frame.
@@ -406,7 +414,6 @@ static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay)
 static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 {
 	struct drm_i915_private *dev_priv = overlay->i915;
-	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
 	int ret;
 
 	lockdep_assert_held(&dev_priv->drm.struct_mutex);
@@ -420,8 +427,9 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 	if (I915_READ(ISR) & I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT) {
 		/* synchronous slowpath */
 		struct drm_i915_gem_request *req;
+		struct intel_ring *ring;
 
-		req = i915_gem_request_alloc(engine, NULL);
+		req = alloc_request(overlay);
 		if (IS_ERR(req))
 			return PTR_ERR(req);
 
@@ -431,22 +439,19 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 			return ret;
 		}
 
-		intel_ring_emit(engine,
+		ring = req->ring;
+		intel_ring_emit(ring,
 				MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
-		intel_ring_emit(engine, MI_NOOP);
-		intel_ring_advance(engine);
+		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_advance(ring);
 
 		ret = intel_overlay_do_wait_request(overlay, req,
 						    intel_overlay_release_old_vid_tail);
 		if (ret)
 			return ret;
-	}
-
-	intel_overlay_release_old_vid_tail(overlay);
+	} else
+		intel_overlay_release_old_vid_tail(&overlay->last_flip, NULL);
 
-
-	i915_gem_track_fb(overlay->old_vid_bo, NULL,
-			  INTEL_FRONTBUFFER_OVERLAY(overlay->crtc->pipe));
 	return 0;
 }
 
@@ -459,7 +464,6 @@ void intel_overlay_reset(struct drm_i915_private *dev_priv)
 
 	intel_overlay_release_old_vid(overlay);
 
-	overlay->last_flip_req = NULL;
 	overlay->old_xscale = 0;
 	overlay->old_yscale = 0;
 	overlay->crtc = NULL;
@@ -836,8 +840,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
 	overlay->old_vid_bo = overlay->vid_bo;
 	overlay->vid_bo = new_bo;
 
-	intel_frontbuffer_flip(&dev_priv->drm,
-			       INTEL_FRONTBUFFER_OVERLAY(pipe));
+	intel_frontbuffer_flip(dev_priv, INTEL_FRONTBUFFER_OVERLAY(pipe));
 
 	return 0;
 
@@ -870,12 +873,7 @@ int intel_overlay_switch_off(struct intel_overlay *overlay)
 	iowrite32(0, &regs->OCMD);
 	intel_overlay_unmap_regs(overlay, regs);
 
-	ret = intel_overlay_off(overlay);
-	if (ret != 0)
-		return ret;
-
-	intel_overlay_off_tail(overlay);
-	return 0;
+	return intel_overlay_off(overlay);
 }
 
 static int check_overlay_possible_on_crtc(struct intel_overlay *overlay,
@@ -1122,9 +1120,8 @@ int intel_overlay_put_image_ioctl(struct drm_device *dev, void *data,
 	}
 	crtc = to_intel_crtc(drmmode_crtc);
 
-	new_bo = to_intel_bo(drm_gem_object_lookup(file_priv,
-						   put_image_rec->bo_handle));
-	if (&new_bo->base == NULL) {
+	new_bo = i915_gem_object_lookup(file_priv, put_image_rec->bo_handle);
+	if (!new_bo) {
 		ret = -ENOENT;
 		goto out_free;
 	}
@@ -1132,7 +1129,7 @@ int intel_overlay_put_image_ioctl(struct drm_device *dev, void *data,
 	drm_modeset_lock_all(dev);
 	mutex_lock(&dev->struct_mutex);
 
-	if (new_bo->tiling_mode) {
+	if (i915_gem_object_is_tiled(new_bo)) {
 		DRM_DEBUG_KMS("buffer used for overlay image can not be tiled\n");
 		ret = -EINVAL;
 		goto out_unlock;
@@ -1220,7 +1217,7 @@ int intel_overlay_put_image_ioctl(struct drm_device *dev, void *data,
 out_unlock:
 	mutex_unlock(&dev->struct_mutex);
 	drm_modeset_unlock_all(dev);
-	drm_gem_object_unreference_unlocked(&new_bo->base);
+	i915_gem_object_put_unlocked(new_bo);
 out_free:
 	kfree(params);
 
@@ -1404,7 +1401,8 @@ void intel_setup_overlay(struct drm_i915_private *dev_priv)
 		}
 		overlay->flip_addr = reg_bo->phys_handle->busaddr;
 	} else {
-		ret = i915_gem_obj_ggtt_pin(reg_bo, PAGE_SIZE, PIN_MAPPABLE);
+		ret = i915_gem_object_ggtt_pin(reg_bo, NULL,
+					       0, PAGE_SIZE, PIN_MAPPABLE);
 		if (ret) {
 			DRM_ERROR("failed to pin overlay register bo\n");
 			goto out_free_bo;
@@ -1444,7 +1442,7 @@ out_unpin_bo:
 	if (!OVERLAY_NEEDS_PHYSICAL(dev_priv))
 		i915_gem_object_ggtt_unpin(reg_bo);
 out_free_bo:
-	drm_gem_object_unreference(&reg_bo->base);
+	i915_gem_object_put(reg_bo);
 out_free:
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 	kfree(overlay);
@@ -1461,7 +1459,7 @@ void intel_cleanup_overlay(struct drm_i915_private *dev_priv)
 	 * hardware should be off already */
 	WARN_ON(dev_priv->overlay->active);
 
-	drm_gem_object_unreference_unlocked(&dev_priv->overlay->reg_bo->base);
+	i915_gem_object_put_unlocked(dev_priv->overlay->reg_bo);
 	kfree(dev_priv->overlay);
 }
 

drivers/gpu/drm/i915/intel_pm.c

@@ -340,6 +340,11 @@ void intel_set_memory_cxsr(struct drm_i915_private *dev_priv, bool enable)
 		I915_WRITE(FW_BLC_SELF, val);
 		POSTING_READ(FW_BLC_SELF);
 	} else if (IS_I915GM(dev)) {
+		/*
+		 * FIXME can't find a bit like this for 915G, and
+		 * and yet it does have the related watermark in
+		 * FW_BLC_SELF. What's going on?
+		 */
 		val = enable ? _MASKED_BIT_ENABLE(INSTPM_SELF_EN) :
 			       _MASKED_BIT_DISABLE(INSTPM_SELF_EN);
 		I915_WRITE(INSTPM, val);
@@ -1580,7 +1585,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 		obj = intel_fb_obj(enabled->primary->state->fb);
 
 		/* self-refresh seems busted with untiled */
-		if (obj->tiling_mode == I915_TILING_NONE)
+		if (!i915_gem_object_is_tiled(obj))
 			enabled = NULL;
 	}
 
@@ -1604,6 +1609,9 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 		unsigned long line_time_us;
 		int entries;
 
+		if (IS_I915GM(dev) || IS_I945GM(dev))
+			cpp = 4;
+
 		line_time_us = max(htotal * 1000 / clock, 1);
 
 		/* Use ns/us then divide to preserve precision */
@@ -1618,7 +1626,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 		if (IS_I945G(dev) || IS_I945GM(dev))
 			I915_WRITE(FW_BLC_SELF,
 				   FW_BLC_SELF_FIFO_MASK | (srwm & 0xff));
-		else if (IS_I915GM(dev))
+		else
 			I915_WRITE(FW_BLC_SELF, srwm & 0x3f);
 	}
 
@@ -3344,6 +3352,8 @@ static uint32_t skl_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal,
 		plane_bytes_per_line *= 4;
 		plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 		plane_blocks_per_line /= 4;
+	} else if (tiling == DRM_FORMAT_MOD_NONE) {
+		plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512) + 1;
 	} else {
 		plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 	}
@@ -4912,7 +4922,7 @@ void gen6_rps_boost(struct drm_i915_private *dev_priv,
 	 */
 	if (!(dev_priv->gt.awake &&
 	      dev_priv->rps.enabled &&
-	      dev_priv->rps.cur_freq < dev_priv->rps.max_freq_softlimit))
+	      dev_priv->rps.cur_freq < dev_priv->rps.boost_freq))
 		return;
 
 	/* Force a RPS boost (and don't count it against the client) if
@@ -5103,35 +5113,31 @@ int sanitize_rc6_option(struct drm_i915_private *dev_priv, int enable_rc6)
 
 static void gen6_init_rps_frequencies(struct drm_i915_private *dev_priv)
 {
-	uint32_t rp_state_cap;
-	u32 ddcc_status = 0;
-	int ret;
-
 	/* All of these values are in units of 50MHz */
-	dev_priv->rps.cur_freq		= 0;
+
 	/* static values from HW: RP0 > RP1 > RPn (min_freq) */
 	if (IS_BROXTON(dev_priv)) {
-		rp_state_cap = I915_READ(BXT_RP_STATE_CAP);
+		u32 rp_state_cap = I915_READ(BXT_RP_STATE_CAP);
 		dev_priv->rps.rp0_freq = (rp_state_cap >> 16) & 0xff;
 		dev_priv->rps.rp1_freq = (rp_state_cap >>  8) & 0xff;
 		dev_priv->rps.min_freq = (rp_state_cap >>  0) & 0xff;
 	} else {
-		rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
+		u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
 		dev_priv->rps.rp0_freq = (rp_state_cap >>  0) & 0xff;
 		dev_priv->rps.rp1_freq = (rp_state_cap >>  8) & 0xff;
 		dev_priv->rps.min_freq = (rp_state_cap >> 16) & 0xff;
 	}
-
 	/* hw_max = RP0 until we check for overclocking */
-	dev_priv->rps.max_freq		= dev_priv->rps.rp0_freq;
+	dev_priv->rps.max_freq = dev_priv->rps.rp0_freq;
 
 	dev_priv->rps.efficient_freq = dev_priv->rps.rp1_freq;
 	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv) ||
 	    IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
-		ret = sandybridge_pcode_read(dev_priv,
-					HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL,
-					&ddcc_status);
-		if (0 == ret)
+		u32 ddcc_status = 0;
+
+		if (sandybridge_pcode_read(dev_priv,
+					   HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL,
+					   &ddcc_status) == 0)
 			dev_priv->rps.efficient_freq =
 				clamp_t(u8,
 					((ddcc_status >> 8) & 0xff),
@@ -5141,29 +5147,26 @@ static void gen6_init_rps_frequencies(struct drm_i915_private *dev_priv)
 
 	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
 		/* Store the frequency values in 16.66 MHZ units, which is
-		   the natural hardware unit for SKL */
+		 * the natural hardware unit for SKL
+		 */
 		dev_priv->rps.rp0_freq *= GEN9_FREQ_SCALER;
 		dev_priv->rps.rp1_freq *= GEN9_FREQ_SCALER;
 		dev_priv->rps.min_freq *= GEN9_FREQ_SCALER;
 		dev_priv->rps.max_freq *= GEN9_FREQ_SCALER;
 		dev_priv->rps.efficient_freq *= GEN9_FREQ_SCALER;
 	}
+}
 
-	dev_priv->rps.idle_freq = dev_priv->rps.min_freq;
+static void reset_rps(struct drm_i915_private *dev_priv,
+		      void (*set)(struct drm_i915_private *, u8))
+{
+	u8 freq = dev_priv->rps.cur_freq;
 
-	/* Preserve min/max settings in case of re-init */
-	if (dev_priv->rps.max_freq_softlimit == 0)
-		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
+	/* force a reset */
+	dev_priv->rps.power = -1;
+	dev_priv->rps.cur_freq = -1;
 
-	if (dev_priv->rps.min_freq_softlimit == 0) {
-		if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
-			dev_priv->rps.min_freq_softlimit =
-				max_t(int, dev_priv->rps.efficient_freq,
-				      intel_freq_opcode(dev_priv, 450));
-		else
-			dev_priv->rps.min_freq_softlimit =
-				dev_priv->rps.min_freq;
-	}
+	set(dev_priv, freq);
 }
 
 /* See the Gen9_GT_PM_Programming_Guide doc for the below */
@@ -5171,8 +5174,6 @@ static void gen9_enable_rps(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
-	gen6_init_rps_frequencies(dev_priv);
-
 	/* WaGsvDisableTurbo: Workaround to disable turbo on BXT A* */
 	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
 		/*
@@ -5202,8 +5203,7 @@ static void gen9_enable_rps(struct drm_i915_private *dev_priv)
 	/* Leaning on the below call to gen6_set_rps to program/setup the
 	 * Up/Down EI & threshold registers, as well as the RP_CONTROL,
 	 * RP_INTERRUPT_LIMITS & RPNSWREQ registers */
-	dev_priv->rps.power = HIGH_POWER; /* force a reset */
-	gen6_set_rps(dev_priv, dev_priv->rps.idle_freq);
+	reset_rps(dev_priv, gen6_set_rps);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -5290,9 +5290,6 @@ static void gen8_enable_rps(struct drm_i915_private *dev_priv)
 	/* 2a: Disable RC states. */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 
-	/* Initialize rps frequencies */
-	gen6_init_rps_frequencies(dev_priv);
-
 	/* 2b: Program RC6 thresholds.*/
 	I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
 	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
@@ -5349,8 +5346,7 @@ static void gen8_enable_rps(struct drm_i915_private *dev_priv)
 
 	/* 6: Ring frequency + overclocking (our driver does this later */
 
-	dev_priv->rps.power = HIGH_POWER; /* force a reset */
-	gen6_set_rps(dev_priv, dev_priv->rps.idle_freq);
+	reset_rps(dev_priv, gen6_set_rps);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -5358,7 +5354,7 @@ static void gen8_enable_rps(struct drm_i915_private *dev_priv)
 static void gen6_enable_rps(struct drm_i915_private *dev_priv)
 {
 	struct intel_engine_cs *engine;
-	u32 rc6vids, pcu_mbox = 0, rc6_mask = 0;
+	u32 rc6vids, rc6_mask = 0;
 	u32 gtfifodbg;
 	int rc6_mode;
 	int ret;
@@ -5382,9 +5378,6 @@ static void gen6_enable_rps(struct drm_i915_private *dev_priv)
 
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
-	/* Initialize rps frequencies */
-	gen6_init_rps_frequencies(dev_priv);
-
 	/* disable the counters and set deterministic thresholds */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 
@@ -5435,16 +5428,7 @@ static void gen6_enable_rps(struct drm_i915_private *dev_priv)
 	if (ret)
 		DRM_DEBUG_DRIVER("Failed to set the min frequency\n");
 
-	ret = sandybridge_pcode_read(dev_priv, GEN6_READ_OC_PARAMS, &pcu_mbox);
-	if (!ret && (pcu_mbox & (1<<31))) { /* OC supported */
-		DRM_DEBUG_DRIVER("Overclocking supported. Max: %dMHz, Overclock max: %dMHz\n",
-				 (dev_priv->rps.max_freq_softlimit & 0xff) * 50,
-				 (pcu_mbox & 0xff) * 50);
-		dev_priv->rps.max_freq = pcu_mbox & 0xff;
-	}
-
-	dev_priv->rps.power = HIGH_POWER; /* force a reset */
-	gen6_set_rps(dev_priv, dev_priv->rps.idle_freq);
+	reset_rps(dev_priv, gen6_set_rps);
 
 	rc6vids = 0;
 	ret = sandybridge_pcode_read(dev_priv, GEN6_PCODE_READ_RC6VIDS, &rc6vids);
@@ -5463,7 +5447,7 @@ static void gen6_enable_rps(struct drm_i915_private *dev_priv)
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
 
-static void __gen6_update_ring_freq(struct drm_i915_private *dev_priv)
+static void gen6_update_ring_freq(struct drm_i915_private *dev_priv)
 {
 	int min_freq = 15;
 	unsigned int gpu_freq;
@@ -5547,16 +5531,6 @@ static void __gen6_update_ring_freq(struct drm_i915_private *dev_priv)
 	}
 }
 
-void gen6_update_ring_freq(struct drm_i915_private *dev_priv)
-{
-	if (!HAS_CORE_RING_FREQ(dev_priv))
-		return;
-
-	mutex_lock(&dev_priv->rps.hw_lock);
-	__gen6_update_ring_freq(dev_priv);
-	mutex_unlock(&dev_priv->rps.hw_lock);
-}
-
 static int cherryview_rps_max_freq(struct drm_i915_private *dev_priv)
 {
 	u32 val, rp0;
@@ -5746,7 +5720,7 @@ static void valleyview_cleanup_pctx(struct drm_i915_private *dev_priv)
 	if (WARN_ON(!dev_priv->vlv_pctx))
 		return;
 
-	drm_gem_object_unreference_unlocked(&dev_priv->vlv_pctx->base);
+	i915_gem_object_put_unlocked(dev_priv->vlv_pctx);
 	dev_priv->vlv_pctx = NULL;
 }
 
@@ -5769,8 +5743,6 @@ static void valleyview_init_gt_powersave(struct drm_i915_private *dev_priv)
 
 	vlv_init_gpll_ref_freq(dev_priv);
 
-	mutex_lock(&dev_priv->rps.hw_lock);
-
 	val = vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS);
 	switch ((val >> 6) & 3) {
 	case 0:
@@ -5806,17 +5778,6 @@ static void valleyview_init_gt_powersave(struct drm_i915_private *dev_priv)
 	DRM_DEBUG_DRIVER("min GPU freq: %d MHz (%u)\n",
 			 intel_gpu_freq(dev_priv, dev_priv->rps.min_freq),
 			 dev_priv->rps.min_freq);
-
-	dev_priv->rps.idle_freq = dev_priv->rps.min_freq;
-
-	/* Preserve min/max settings in case of re-init */
-	if (dev_priv->rps.max_freq_softlimit == 0)
-		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
-
-	if (dev_priv->rps.min_freq_softlimit == 0)
-		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
-
-	mutex_unlock(&dev_priv->rps.hw_lock);
 }
 
 static void cherryview_init_gt_powersave(struct drm_i915_private *dev_priv)
@@ -5827,8 +5788,6 @@ static void cherryview_init_gt_powersave(struct drm_i915_private *dev_priv)
 
 	vlv_init_gpll_ref_freq(dev_priv);
 
-	mutex_lock(&dev_priv->rps.hw_lock);
-
 	mutex_lock(&dev_priv->sb_lock);
 	val = vlv_cck_read(dev_priv, CCK_FUSE_REG);
 	mutex_unlock(&dev_priv->sb_lock);
@@ -5870,17 +5829,6 @@ static void cherryview_init_gt_powersave(struct drm_i915_private *dev_priv)
 		   dev_priv->rps.rp1_freq |
 		   dev_priv->rps.min_freq) & 1,
 		  "Odd GPU freq values\n");
-
-	dev_priv->rps.idle_freq = dev_priv->rps.min_freq;
-
-	/* Preserve min/max settings in case of re-init */
-	if (dev_priv->rps.max_freq_softlimit == 0)
-		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
-
-	if (dev_priv->rps.min_freq_softlimit == 0)
-		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
-
-	mutex_unlock(&dev_priv->rps.hw_lock);
 }
 
 static void valleyview_cleanup_gt_powersave(struct drm_i915_private *dev_priv)
@@ -5971,16 +5919,7 @@ static void cherryview_enable_rps(struct drm_i915_private *dev_priv)
 	DRM_DEBUG_DRIVER("GPLL enabled? %s\n", yesno(val & GPLLENABLE));
 	DRM_DEBUG_DRIVER("GPU status: 0x%08x\n", val);
 
-	dev_priv->rps.cur_freq = (val >> 8) & 0xff;
-	DRM_DEBUG_DRIVER("current GPU freq: %d MHz (%u)\n",
-			 intel_gpu_freq(dev_priv, dev_priv->rps.cur_freq),
-			 dev_priv->rps.cur_freq);
-
-	DRM_DEBUG_DRIVER("setting GPU freq to %d MHz (%u)\n",
-			 intel_gpu_freq(dev_priv, dev_priv->rps.idle_freq),
-			 dev_priv->rps.idle_freq);
-
-	valleyview_set_rps(dev_priv, dev_priv->rps.idle_freq);
+	reset_rps(dev_priv, valleyview_set_rps);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -6060,16 +5999,7 @@ static void valleyview_enable_rps(struct drm_i915_private *dev_priv)
 	DRM_DEBUG_DRIVER("GPLL enabled? %s\n", yesno(val & GPLLENABLE));
 	DRM_DEBUG_DRIVER("GPU status: 0x%08x\n", val);
 
-	dev_priv->rps.cur_freq = (val >> 8) & 0xff;
-	DRM_DEBUG_DRIVER("current GPU freq: %d MHz (%u)\n",
-			 intel_gpu_freq(dev_priv, dev_priv->rps.cur_freq),
-			 dev_priv->rps.cur_freq);
-
-	DRM_DEBUG_DRIVER("setting GPU freq to %d MHz (%u)\n",
-			 intel_gpu_freq(dev_priv, dev_priv->rps.idle_freq),
-			 dev_priv->rps.idle_freq);
-
-	valleyview_set_rps(dev_priv, dev_priv->rps.idle_freq);
+	reset_rps(dev_priv, valleyview_set_rps);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -6398,19 +6328,11 @@ EXPORT_SYMBOL_GPL(i915_gpu_lower);
  */
 bool i915_gpu_busy(void)
 {
-	struct drm_i915_private *dev_priv;
-	struct intel_engine_cs *engine;
 	bool ret = false;
 
 	spin_lock_irq(&mchdev_lock);
-	if (!i915_mch_dev)
-		goto out_unlock;
-	dev_priv = i915_mch_dev;
-
-	for_each_engine(engine, dev_priv)
-		ret |= !list_empty(&engine->request_list);
-
-out_unlock:
+	if (i915_mch_dev)
+		ret = i915_mch_dev->gt.awake;
 	spin_unlock_irq(&mchdev_lock);
 
 	return ret;
@@ -6566,30 +6488,60 @@ void intel_init_gt_powersave(struct drm_i915_private *dev_priv)
 		intel_runtime_pm_get(dev_priv);
 	}
 
+	mutex_lock(&dev_priv->rps.hw_lock);
+
+	/* Initialize RPS limits (for userspace) */
 	if (IS_CHERRYVIEW(dev_priv))
 		cherryview_init_gt_powersave(dev_priv);
 	else if (IS_VALLEYVIEW(dev_priv))
 		valleyview_init_gt_powersave(dev_priv);
+	else if (INTEL_GEN(dev_priv) >= 6)
+		gen6_init_rps_frequencies(dev_priv);
+
+	/* Derive initial user preferences/limits from the hardware limits */
+	dev_priv->rps.idle_freq = dev_priv->rps.min_freq;
+	dev_priv->rps.cur_freq = dev_priv->rps.idle_freq;
+
+	dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
+	dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
+
+	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+		dev_priv->rps.min_freq_softlimit =
+			max_t(int,
+			      dev_priv->rps.efficient_freq,
+			      intel_freq_opcode(dev_priv, 450));
+
+	/* After setting max-softlimit, find the overclock max freq */
+	if (IS_GEN6(dev_priv) ||
+	    IS_IVYBRIDGE(dev_priv) || IS_HASWELL(dev_priv)) {
+		u32 params = 0;
+
+		sandybridge_pcode_read(dev_priv, GEN6_READ_OC_PARAMS, &params);
+		if (params & BIT(31)) { /* OC supported */
+			DRM_DEBUG_DRIVER("Overclocking supported, max: %dMHz, overclock: %dMHz\n",
+					 (dev_priv->rps.max_freq & 0xff) * 50,
+					 (params & 0xff) * 50);
+			dev_priv->rps.max_freq = params & 0xff;
+		}
+	}
+
+	/* Finally allow us to boost to max by default */
+	dev_priv->rps.boost_freq = dev_priv->rps.max_freq;
+
+	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	intel_autoenable_gt_powersave(dev_priv);
 }
 
 void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv)
 {
-	if (IS_CHERRYVIEW(dev_priv))
-		return;
-	else if (IS_VALLEYVIEW(dev_priv))
+	if (IS_VALLEYVIEW(dev_priv))
 		valleyview_cleanup_gt_powersave(dev_priv);
 
 	if (!i915.enable_rc6)
 		intel_runtime_pm_put(dev_priv);
 }
 
-static void gen6_suspend_rps(struct drm_i915_private *dev_priv)
-{
-	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
-
-	gen6_disable_rps_interrupts(dev_priv);
-}
-
 /**
  * intel_suspend_gt_powersave - suspend PM work and helper threads
  * @dev_priv: i915 device
@@ -6603,60 +6555,76 @@ void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv)
 	if (INTEL_GEN(dev_priv) < 6)
 		return;
 
-	gen6_suspend_rps(dev_priv);
+	if (cancel_delayed_work_sync(&dev_priv->rps.autoenable_work))
+		intel_runtime_pm_put(dev_priv);
 
-	/* Force GPU to min freq during suspend */
-	gen6_rps_idle(dev_priv);
+	/* gen6_rps_idle() will be called later to disable interrupts */
+}
+
+void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv)
+{
+	dev_priv->rps.enabled = true; /* force disabling */
+	intel_disable_gt_powersave(dev_priv);
+
+	gen6_reset_rps_interrupts(dev_priv);
 }
 
 void intel_disable_gt_powersave(struct drm_i915_private *dev_priv)
 {
-	if (IS_IRONLAKE_M(dev_priv)) {
-		ironlake_disable_drps(dev_priv);
-	} else if (INTEL_INFO(dev_priv)->gen >= 6) {
-		intel_suspend_gt_powersave(dev_priv);
+	if (!READ_ONCE(dev_priv->rps.enabled))
+		return;
 
-		mutex_lock(&dev_priv->rps.hw_lock);
-		if (INTEL_INFO(dev_priv)->gen >= 9) {
-			gen9_disable_rc6(dev_priv);
-			gen9_disable_rps(dev_priv);
-		} else if (IS_CHERRYVIEW(dev_priv))
-			cherryview_disable_rps(dev_priv);
-		else if (IS_VALLEYVIEW(dev_priv))
-			valleyview_disable_rps(dev_priv);
-		else
-			gen6_disable_rps(dev_priv);
+	mutex_lock(&dev_priv->rps.hw_lock);
 
-		dev_priv->rps.enabled = false;
-		mutex_unlock(&dev_priv->rps.hw_lock);
+	if (INTEL_GEN(dev_priv) >= 9) {
+		gen9_disable_rc6(dev_priv);
+		gen9_disable_rps(dev_priv);
+	} else if (IS_CHERRYVIEW(dev_priv)) {
+		cherryview_disable_rps(dev_priv);
+	} else if (IS_VALLEYVIEW(dev_priv)) {
+		valleyview_disable_rps(dev_priv);
+	} else if (INTEL_GEN(dev_priv) >= 6) {
+		gen6_disable_rps(dev_priv);
+	}  else if (IS_IRONLAKE_M(dev_priv)) {
+		ironlake_disable_drps(dev_priv);
 	}
+
+	dev_priv->rps.enabled = false;
+	mutex_unlock(&dev_priv->rps.hw_lock);
 }
 
-static void intel_gen6_powersave_work(struct work_struct *work)
+void intel_enable_gt_powersave(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv =
-		container_of(work, struct drm_i915_private,
-			     rps.delayed_resume_work.work);
+	/* We shouldn't be disabling as we submit, so this should be less
+	 * racy than it appears!
+	 */
+	if (READ_ONCE(dev_priv->rps.enabled))
+		return;
 
-	mutex_lock(&dev_priv->rps.hw_lock);
+	/* Powersaving is controlled by the host when inside a VM */
+	if (intel_vgpu_active(dev_priv))
+		return;
 
-	gen6_reset_rps_interrupts(dev_priv);
+	mutex_lock(&dev_priv->rps.hw_lock);
 
 	if (IS_CHERRYVIEW(dev_priv)) {
 		cherryview_enable_rps(dev_priv);
 	} else if (IS_VALLEYVIEW(dev_priv)) {
 		valleyview_enable_rps(dev_priv);
-	} else if (INTEL_INFO(dev_priv)->gen >= 9) {
+	} else if (INTEL_GEN(dev_priv) >= 9) {
 		gen9_enable_rc6(dev_priv);
 		gen9_enable_rps(dev_priv);
 		if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
-			__gen6_update_ring_freq(dev_priv);
+			gen6_update_ring_freq(dev_priv);
 	} else if (IS_BROADWELL(dev_priv)) {
 		gen8_enable_rps(dev_priv);
-		__gen6_update_ring_freq(dev_priv);
-	} else {
+		gen6_update_ring_freq(dev_priv);
+	} else if (INTEL_GEN(dev_priv) >= 6) {
 		gen6_enable_rps(dev_priv);
-		__gen6_update_ring_freq(dev_priv);
+		gen6_update_ring_freq(dev_priv);
+	} else if (IS_IRONLAKE_M(dev_priv)) {
+		ironlake_enable_drps(dev_priv);
+		intel_init_emon(dev_priv);
 	}
 
 	WARN_ON(dev_priv->rps.max_freq < dev_priv->rps.min_freq);
@@ -6666,18 +6634,47 @@ static void intel_gen6_powersave_work(struct work_struct *work)
 	WARN_ON(dev_priv->rps.efficient_freq > dev_priv->rps.max_freq);
 
 	dev_priv->rps.enabled = true;
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
 
-	gen6_enable_rps_interrupts(dev_priv);
+static void __intel_autoenable_gt_powersave(struct work_struct *work)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(work, typeof(*dev_priv), rps.autoenable_work.work);
+	struct intel_engine_cs *rcs;
+	struct drm_i915_gem_request *req;
 
-	mutex_unlock(&dev_priv->rps.hw_lock);
+	if (READ_ONCE(dev_priv->rps.enabled))
+		goto out;
+
+	rcs = &dev_priv->engine[RCS];
+	if (rcs->last_context)
+		goto out;
+
+	if (!rcs->init_context)
+		goto out;
 
+	mutex_lock(&dev_priv->drm.struct_mutex);
+
+	req = i915_gem_request_alloc(rcs, dev_priv->kernel_context);
+	if (IS_ERR(req))
+		goto unlock;
+
+	if (!i915.enable_execlists && i915_switch_context(req) == 0)
+		rcs->init_context(req);
+
+	/* Mark the device busy, calling intel_enable_gt_powersave() */
+	i915_add_request_no_flush(req);
+
+unlock:
+	mutex_unlock(&dev_priv->drm.struct_mutex);
+out:
 	intel_runtime_pm_put(dev_priv);
 }
 
-void intel_enable_gt_powersave(struct drm_i915_private *dev_priv)
+void intel_autoenable_gt_powersave(struct drm_i915_private *dev_priv)
 {
-	/* Powersaving is controlled by the host when inside a VM */
-	if (intel_vgpu_active(dev_priv))
+	if (READ_ONCE(dev_priv->rps.enabled))
 		return;
 
 	if (IS_IRONLAKE_M(dev_priv)) {
@@ -6698,21 +6695,13 @@ void intel_enable_gt_powersave(struct drm_i915_private *dev_priv)
 		 * paths, so the _noresume version is enough (and in case of
 		 * runtime resume it's necessary).
 		 */
-		if (schedule_delayed_work(&dev_priv->rps.delayed_resume_work,
-					   round_jiffies_up_relative(HZ)))
+		if (queue_delayed_work(dev_priv->wq,
+				       &dev_priv->rps.autoenable_work,
+				       round_jiffies_up_relative(HZ)))
 			intel_runtime_pm_get_noresume(dev_priv);
 	}
 }
 
-void intel_reset_gt_powersave(struct drm_i915_private *dev_priv)
-{
-	if (INTEL_INFO(dev_priv)->gen < 6)
-		return;
-
-	gen6_suspend_rps(dev_priv);
-	dev_priv->rps.enabled = false;
-}
-
 static void ibx_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -7787,7 +7776,7 @@ static void __intel_rps_boost_work(struct work_struct *work)
 	if (!i915_gem_request_completed(req))
 		gen6_rps_boost(req->i915, NULL, req->emitted_jiffies);
 
-	i915_gem_request_unreference(req);
+	i915_gem_request_put(req);
 	kfree(boost);
 }
 
@@ -7805,8 +7794,7 @@ void intel_queue_rps_boost_for_request(struct drm_i915_gem_request *req)
 	if (boost == NULL)
 		return;
 
-	i915_gem_request_reference(req);
-	boost->req = req;
+	boost->req = i915_gem_request_get(req);
 
 	INIT_WORK(&boost->work, __intel_rps_boost_work);
 	queue_work(req->i915->wq, &boost->work);
@@ -7819,11 +7807,9 @@ void intel_pm_setup(struct drm_device *dev)
 	mutex_init(&dev_priv->rps.hw_lock);
 	spin_lock_init(&dev_priv->rps.client_lock);
 
-	INIT_DELAYED_WORK(&dev_priv->rps.delayed_resume_work,
-			  intel_gen6_powersave_work);
+	INIT_DELAYED_WORK(&dev_priv->rps.autoenable_work,
+			  __intel_autoenable_gt_powersave);
 	INIT_LIST_HEAD(&dev_priv->rps.clients);
-	INIT_LIST_HEAD(&dev_priv->rps.semaphores.link);
-	INIT_LIST_HEAD(&dev_priv->rps.mmioflips.link);
 
 	dev_priv->pm.suspended = false;
 	atomic_set(&dev_priv->pm.wakeref_count, 0);

drivers/gpu/drm/i915/intel_psr.c

@@ -645,9 +645,8 @@ unlock:
 	mutex_unlock(&dev_priv->psr.lock);
 }
 
-static void intel_psr_exit(struct drm_device *dev)
+static void intel_psr_exit(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_dp *intel_dp = dev_priv->psr.enabled;
 	struct drm_crtc *crtc = dp_to_dig_port(intel_dp)->base.base.crtc;
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
@@ -656,7 +655,7 @@ static void intel_psr_exit(struct drm_device *dev)
 	if (!dev_priv->psr.active)
 		return;
 
-	if (HAS_DDI(dev)) {
+	if (HAS_DDI(dev_priv)) {
 		val = I915_READ(EDP_PSR_CTL);
 
 		WARN_ON(!(val & EDP_PSR_ENABLE));
@@ -691,7 +690,7 @@ static void intel_psr_exit(struct drm_device *dev)
 
 /**
  * intel_psr_single_frame_update - Single Frame Update
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * Some platforms support a single frame update feature that is used to
@@ -699,10 +698,9 @@ static void intel_psr_exit(struct drm_device *dev)
  * So far it is only implemented for Valleyview and Cherryview because
  * hardware requires this to be done before a page flip.
  */
-void intel_psr_single_frame_update(struct drm_device *dev,
+void intel_psr_single_frame_update(struct drm_i915_private *dev_priv,
 				   unsigned frontbuffer_bits)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_crtc *crtc;
 	enum pipe pipe;
 	u32 val;
@@ -711,7 +709,7 @@ void intel_psr_single_frame_update(struct drm_device *dev,
 	 * Single frame update is already supported on BDW+ but it requires
 	 * many W/A and it isn't really needed.
 	 */
-	if (!IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev))
+	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv))
 		return;
 
 	mutex_lock(&dev_priv->psr.lock);
@@ -737,7 +735,7 @@ void intel_psr_single_frame_update(struct drm_device *dev,
 
 /**
  * intel_psr_invalidate - Invalidade PSR
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * Since the hardware frontbuffer tracking has gaps we need to integrate
@@ -747,10 +745,9 @@ void intel_psr_single_frame_update(struct drm_device *dev,
  *
  * Dirty frontbuffers relevant to PSR are tracked in busy_frontbuffer_bits."
  */
-void intel_psr_invalidate(struct drm_device *dev,
+void intel_psr_invalidate(struct drm_i915_private *dev_priv,
 			  unsigned frontbuffer_bits)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_crtc *crtc;
 	enum pipe pipe;
 
@@ -767,14 +764,14 @@ void intel_psr_invalidate(struct drm_device *dev,
 	dev_priv->psr.busy_frontbuffer_bits |= frontbuffer_bits;
 
 	if (frontbuffer_bits)
-		intel_psr_exit(dev);
+		intel_psr_exit(dev_priv);
 
 	mutex_unlock(&dev_priv->psr.lock);
 }
 
 /**
  * intel_psr_flush - Flush PSR
- * @dev: DRM device
+ * @dev_priv: i915 device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  * @origin: which operation caused the flush
  *
@@ -785,10 +782,9 @@ void intel_psr_invalidate(struct drm_device *dev,
  *
  * Dirty frontbuffers relevant to PSR are tracked in busy_frontbuffer_bits.
  */
-void intel_psr_flush(struct drm_device *dev,
+void intel_psr_flush(struct drm_i915_private *dev_priv,
 		     unsigned frontbuffer_bits, enum fb_op_origin origin)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_crtc *crtc;
 	enum pipe pipe;
 
@@ -806,7 +802,7 @@ void intel_psr_flush(struct drm_device *dev,
 
 	/* By definition flush = invalidate + flush */
 	if (frontbuffer_bits)
-		intel_psr_exit(dev);
+		intel_psr_exit(dev_priv);
 
 	if (!dev_priv->psr.active && !dev_priv->psr.busy_frontbuffer_bits)
 		if (!work_busy(&dev_priv->psr.work.work))

drivers/gpu/drm/i915/intel_renderstate.h

@@ -24,12 +24,13 @@
 #ifndef _INTEL_RENDERSTATE_H
 #define _INTEL_RENDERSTATE_H
 
-#include "i915_drv.h"
+#include <linux/types.h>
 
-extern const struct intel_renderstate_rodata gen6_null_state;
-extern const struct intel_renderstate_rodata gen7_null_state;
-extern const struct intel_renderstate_rodata gen8_null_state;
-extern const struct intel_renderstate_rodata gen9_null_state;
+struct intel_renderstate_rodata {
+	const u32 *reloc;
+	const u32 *batch;
+	const u32 batch_items;
+};
 
 #define RO_RENDERSTATE(_g)						\
 	const struct intel_renderstate_rodata gen ## _g ## _null_state = { \
@@ -38,4 +39,9 @@ extern const struct intel_renderstate_rodata gen9_null_state;
 		.batch_items = sizeof(gen ## _g ## _null_state_batch)/4, \
 	}
 
+extern const struct intel_renderstate_rodata gen6_null_state;
+extern const struct intel_renderstate_rodata gen7_null_state;
+extern const struct intel_renderstate_rodata gen8_null_state;
+extern const struct intel_renderstate_rodata gen9_null_state;
+
 #endif /* INTEL_RENDERSTATE_H */

drivers/gpu/drm/i915/intel_ringbuffer.c

@@ -47,57 +47,44 @@ int __intel_ring_space(int head, int tail, int size)
 	return space - I915_RING_FREE_SPACE;
 }
 
-void intel_ring_update_space(struct intel_ringbuffer *ringbuf)
+void intel_ring_update_space(struct intel_ring *ring)
 {
-	if (ringbuf->last_retired_head != -1) {
-		ringbuf->head = ringbuf->last_retired_head;
-		ringbuf->last_retired_head = -1;
+	if (ring->last_retired_head != -1) {
+		ring->head = ring->last_retired_head;
+		ring->last_retired_head = -1;
 	}
 
-	ringbuf->space = __intel_ring_space(ringbuf->head & HEAD_ADDR,
-					    ringbuf->tail, ringbuf->size);
-}
-
-static void __intel_ring_advance(struct intel_engine_cs *engine)
-{
-	struct intel_ringbuffer *ringbuf = engine->buffer;
-	ringbuf->tail &= ringbuf->size - 1;
-	engine->write_tail(engine, ringbuf->tail);
+	ring->space = __intel_ring_space(ring->head & HEAD_ADDR,
+					 ring->tail, ring->size);
 }
 
 static int
-gen2_render_ring_flush(struct drm_i915_gem_request *req,
-		       u32	invalidate_domains,
-		       u32	flush_domains)
+gen2_render_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	u32 cmd;
 	int ret;
 
 	cmd = MI_FLUSH;
-	if (((invalidate_domains|flush_domains) & I915_GEM_DOMAIN_RENDER) == 0)
-		cmd |= MI_NO_WRITE_FLUSH;
 
-	if (invalidate_domains & I915_GEM_DOMAIN_SAMPLER)
+	if (mode & EMIT_INVALIDATE)
 		cmd |= MI_READ_FLUSH;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, cmd);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, cmd);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int
-gen4_render_ring_flush(struct drm_i915_gem_request *req,
-		       u32	invalidate_domains,
-		       u32	flush_domains)
+gen4_render_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	u32 cmd;
 	int ret;
 
@@ -129,23 +116,20 @@ gen4_render_ring_flush(struct drm_i915_gem_request *req,
 	 * are flushed at any MI_FLUSH.
 	 */
 
-	cmd = MI_FLUSH | MI_NO_WRITE_FLUSH;
-	if ((invalidate_domains|flush_domains) & I915_GEM_DOMAIN_RENDER)
-		cmd &= ~MI_NO_WRITE_FLUSH;
-	if (invalidate_domains & I915_GEM_DOMAIN_INSTRUCTION)
+	cmd = MI_FLUSH;
+	if (mode & EMIT_INVALIDATE) {
 		cmd |= MI_EXE_FLUSH;
-
-	if (invalidate_domains & I915_GEM_DOMAIN_COMMAND &&
-	    (IS_G4X(req->i915) || IS_GEN5(req->i915)))
-		cmd |= MI_INVALIDATE_ISP;
+		if (IS_G4X(req->i915) || IS_GEN5(req->i915))
+			cmd |= MI_INVALIDATE_ISP;
+	}
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, cmd);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, cmd);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -190,45 +174,46 @@ gen4_render_ring_flush(struct drm_i915_gem_request *req,
 static int
 intel_emit_post_sync_nonzero_flush(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = req->engine;
-	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	struct intel_ring *ring = req->ring;
+	u32 scratch_addr =
+		req->engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	ret = intel_ring_begin(req, 6);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(5));
-	intel_ring_emit(engine, PIPE_CONTROL_CS_STALL |
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(5));
+	intel_ring_emit(ring, PIPE_CONTROL_CS_STALL |
 			PIPE_CONTROL_STALL_AT_SCOREBOARD);
-	intel_ring_emit(engine, scratch_addr | PIPE_CONTROL_GLOBAL_GTT); /* address */
-	intel_ring_emit(engine, 0); /* low dword */
-	intel_ring_emit(engine, 0); /* high dword */
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
+	intel_ring_emit(ring, 0); /* low dword */
+	intel_ring_emit(ring, 0); /* high dword */
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	ret = intel_ring_begin(req, 6);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(5));
-	intel_ring_emit(engine, PIPE_CONTROL_QW_WRITE);
-	intel_ring_emit(engine, scratch_addr | PIPE_CONTROL_GLOBAL_GTT); /* address */
-	intel_ring_emit(engine, 0);
-	intel_ring_emit(engine, 0);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(5));
+	intel_ring_emit(ring, PIPE_CONTROL_QW_WRITE);
+	intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int
-gen6_render_ring_flush(struct drm_i915_gem_request *req,
-		       u32 invalidate_domains, u32 flush_domains)
+gen6_render_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
+	u32 scratch_addr =
+		req->engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	u32 flags = 0;
-	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/* Force SNB workarounds for PIPE_CONTROL flushes */
@@ -240,7 +225,7 @@ gen6_render_ring_flush(struct drm_i915_gem_request *req,
 	 * number of bits based on the write domains has little performance
 	 * impact.
 	 */
-	if (flush_domains) {
+	if (mode & EMIT_FLUSH) {
 		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
 		/*
@@ -249,7 +234,7 @@ gen6_render_ring_flush(struct drm_i915_gem_request *req,
 		 */
 		flags |= PIPE_CONTROL_CS_STALL;
 	}
-	if (invalidate_domains) {
+	if (mode & EMIT_INVALIDATE) {
 		flags |= PIPE_CONTROL_TLB_INVALIDATE;
 		flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
@@ -266,11 +251,11 @@ gen6_render_ring_flush(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(4));
-	intel_ring_emit(engine, flags);
-	intel_ring_emit(engine, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
-	intel_ring_emit(engine, 0);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
+	intel_ring_emit(ring, flags);
+	intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
+	intel_ring_emit(ring, 0);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -278,30 +263,31 @@ gen6_render_ring_flush(struct drm_i915_gem_request *req,
 static int
 gen7_render_ring_cs_stall_wa(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
 	ret = intel_ring_begin(req, 4);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(4));
-	intel_ring_emit(engine, PIPE_CONTROL_CS_STALL |
-			      PIPE_CONTROL_STALL_AT_SCOREBOARD);
-	intel_ring_emit(engine, 0);
-	intel_ring_emit(engine, 0);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
+	intel_ring_emit(ring,
+			PIPE_CONTROL_CS_STALL |
+			PIPE_CONTROL_STALL_AT_SCOREBOARD);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int
-gen7_render_ring_flush(struct drm_i915_gem_request *req,
-		       u32 invalidate_domains, u32 flush_domains)
+gen7_render_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
+	u32 scratch_addr =
+		req->engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	u32 flags = 0;
-	u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/*
@@ -318,13 +304,13 @@ gen7_render_ring_flush(struct drm_i915_gem_request *req,
 	 * number of bits based on the write domains has little performance
 	 * impact.
 	 */
-	if (flush_domains) {
+	if (mode & EMIT_FLUSH) {
 		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DC_FLUSH_ENABLE;
 		flags |= PIPE_CONTROL_FLUSH_ENABLE;
 	}
-	if (invalidate_domains) {
+	if (mode & EMIT_INVALIDATE) {
 		flags |= PIPE_CONTROL_TLB_INVALIDATE;
 		flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
@@ -350,11 +336,11 @@ gen7_render_ring_flush(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(4));
-	intel_ring_emit(engine, flags);
-	intel_ring_emit(engine, scratch_addr);
-	intel_ring_emit(engine, 0);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
+	intel_ring_emit(ring, flags);
+	intel_ring_emit(ring, scratch_addr);
+	intel_ring_emit(ring, 0);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -363,41 +349,40 @@ static int
 gen8_emit_pipe_control(struct drm_i915_gem_request *req,
 		       u32 flags, u32 scratch_addr)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
 	ret = intel_ring_begin(req, 6);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(6));
-	intel_ring_emit(engine, flags);
-	intel_ring_emit(engine, scratch_addr);
-	intel_ring_emit(engine, 0);
-	intel_ring_emit(engine, 0);
-	intel_ring_emit(engine, 0);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+	intel_ring_emit(ring, flags);
+	intel_ring_emit(ring, scratch_addr);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int
-gen8_render_ring_flush(struct drm_i915_gem_request *req,
-		       u32 invalidate_domains, u32 flush_domains)
+gen8_render_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	u32 flags = 0;
 	u32 scratch_addr = req->engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	u32 flags = 0;
 	int ret;
 
 	flags |= PIPE_CONTROL_CS_STALL;
 
-	if (flush_domains) {
+	if (mode & EMIT_FLUSH) {
 		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DC_FLUSH_ENABLE;
 		flags |= PIPE_CONTROL_FLUSH_ENABLE;
 	}
-	if (invalidate_domains) {
+	if (mode & EMIT_INVALIDATE) {
 		flags |= PIPE_CONTROL_TLB_INVALIDATE;
 		flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
@@ -419,14 +404,7 @@ gen8_render_ring_flush(struct drm_i915_gem_request *req,
 	return gen8_emit_pipe_control(req, flags, scratch_addr);
 }
 
-static void ring_write_tail(struct intel_engine_cs *engine,
-			    u32 value)
-{
-	struct drm_i915_private *dev_priv = engine->i915;
-	I915_WRITE_TAIL(engine, value);
-}
-
-u64 intel_ring_get_active_head(struct intel_engine_cs *engine)
+u64 intel_engine_get_active_head(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
 	u64 acthd;
@@ -539,7 +517,7 @@ static bool stop_ring(struct intel_engine_cs *engine)
 
 	I915_WRITE_CTL(engine, 0);
 	I915_WRITE_HEAD(engine, 0);
-	engine->write_tail(engine, 0);
+	I915_WRITE_TAIL(engine, 0);
 
 	if (!IS_GEN2(dev_priv)) {
 		(void)I915_READ_CTL(engine);
@@ -549,16 +527,11 @@ static bool stop_ring(struct intel_engine_cs *engine)
 	return (I915_READ_HEAD(engine) & HEAD_ADDR) == 0;
 }
 
-void intel_engine_init_hangcheck(struct intel_engine_cs *engine)
-{
-	memset(&engine->hangcheck, 0, sizeof(engine->hangcheck));
-}
-
 static int init_ring_common(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
-	struct intel_ringbuffer *ringbuf = engine->buffer;
-	struct drm_i915_gem_object *obj = ringbuf->obj;
+	struct intel_ring *ring = engine->buffer;
+	struct drm_i915_gem_object *obj = ring->obj;
 	int ret = 0;
 
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
@@ -608,7 +581,7 @@ static int init_ring_common(struct intel_engine_cs *engine)
 	(void)I915_READ_HEAD(engine);
 
 	I915_WRITE_CTL(engine,
-			((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES)
+			((ring->size - PAGE_SIZE) & RING_NR_PAGES)
 			| RING_VALID);
 
 	/* If the head is still not zero, the ring is dead */
@@ -627,10 +600,10 @@ static int init_ring_common(struct intel_engine_cs *engine)
 		goto out;
 	}
 
-	ringbuf->last_retired_head = -1;
-	ringbuf->head = I915_READ_HEAD(engine);
-	ringbuf->tail = I915_READ_TAIL(engine) & TAIL_ADDR;
-	intel_ring_update_space(ringbuf);
+	ring->last_retired_head = -1;
+	ring->head = I915_READ_HEAD(engine);
+	ring->tail = I915_READ_TAIL(engine) & TAIL_ADDR;
+	intel_ring_update_space(ring);
 
 	intel_engine_init_hangcheck(engine);
 
@@ -646,7 +619,7 @@ void intel_fini_pipe_control(struct intel_engine_cs *engine)
 		return;
 
 	i915_gem_object_ggtt_unpin(engine->scratch.obj);
-	drm_gem_object_unreference(&engine->scratch.obj->base);
+	i915_gem_object_put(engine->scratch.obj);
 	engine->scratch.obj = NULL;
 }
 
@@ -666,7 +639,7 @@ int intel_init_pipe_control(struct intel_engine_cs *engine, int size)
 		goto err;
 	}
 
-	ret = i915_gem_obj_ggtt_pin(obj, 4096, PIN_HIGH);
+	ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 4096, PIN_HIGH);
 	if (ret)
 		goto err_unref;
 
@@ -677,22 +650,21 @@ int intel_init_pipe_control(struct intel_engine_cs *engine, int size)
 	return 0;
 
 err_unref:
-	drm_gem_object_unreference(&engine->scratch.obj->base);
+	i915_gem_object_put(engine->scratch.obj);
 err:
 	return ret;
 }
 
 static int intel_ring_workarounds_emit(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	struct i915_workarounds *w = &req->i915->workarounds;
 	int ret, i;
 
 	if (w->count == 0)
 		return 0;
 
-	engine->gpu_caches_dirty = true;
-	ret = intel_ring_flush_all_caches(req);
+	ret = req->engine->emit_flush(req, EMIT_BARRIER);
 	if (ret)
 		return ret;
 
@@ -700,17 +672,16 @@ static int intel_ring_workarounds_emit(struct drm_i915_gem_request *req)
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(w->count));
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(w->count));
 	for (i = 0; i < w->count; i++) {
-		intel_ring_emit_reg(engine, w->reg[i].addr);
-		intel_ring_emit(engine, w->reg[i].value);
+		intel_ring_emit_reg(ring, w->reg[i].addr);
+		intel_ring_emit(ring, w->reg[i].value);
 	}
-	intel_ring_emit(engine, MI_NOOP);
+	intel_ring_emit(ring, MI_NOOP);
 
-	intel_ring_advance(engine);
+	intel_ring_advance(ring);
 
-	engine->gpu_caches_dirty = true;
-	ret = intel_ring_flush_all_caches(req);
+	ret = req->engine->emit_flush(req, EMIT_BARRIER);
 	if (ret)
 		return ret;
 
@@ -1178,8 +1149,8 @@ static int bxt_init_workarounds(struct intel_engine_cs *engine)
 		I915_WRITE(GEN8_L3SQCREG1, L3_GENERAL_PRIO_CREDITS(62) |
 					   L3_HIGH_PRIO_CREDITS(2));
 
-	/* WaInsertDummyPushConstPs:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))
+	/* WaToEnableHwFixForPushConstHWBug:bxt */
+	if (IS_BXT_REVID(dev_priv, BXT_REVID_C0, REVID_FOREVER))
 		WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2,
 				  GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION);
 
@@ -1222,8 +1193,8 @@ static int kbl_init_workarounds(struct intel_engine_cs *engine)
 		I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) |
 			   GEN8_LQSC_RO_PERF_DIS);
 
-	/* WaInsertDummyPushConstPs:kbl */
-	if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_B0))
+	/* WaToEnableHwFixForPushConstHWBug:kbl */
+	if (IS_KBL_REVID(dev_priv, KBL_REVID_C0, REVID_FOREVER))
 		WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2,
 				  GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION);
 
@@ -1331,189 +1302,195 @@ static void render_ring_cleanup(struct intel_engine_cs *engine)
 
 	if (dev_priv->semaphore_obj) {
 		i915_gem_object_ggtt_unpin(dev_priv->semaphore_obj);
-		drm_gem_object_unreference(&dev_priv->semaphore_obj->base);
+		i915_gem_object_put(dev_priv->semaphore_obj);
 		dev_priv->semaphore_obj = NULL;
 	}
 
 	intel_fini_pipe_control(engine);
 }
 
-static int gen8_rcs_signal(struct drm_i915_gem_request *signaller_req,
-			   unsigned int num_dwords)
+static int gen8_rcs_signal(struct drm_i915_gem_request *req)
 {
-#define MBOX_UPDATE_DWORDS 8
-	struct intel_engine_cs *signaller = signaller_req->engine;
-	struct drm_i915_private *dev_priv = signaller_req->i915;
+	struct intel_ring *ring = req->ring;
+	struct drm_i915_private *dev_priv = req->i915;
 	struct intel_engine_cs *waiter;
 	enum intel_engine_id id;
 	int ret, num_rings;
 
 	num_rings = hweight32(INTEL_INFO(dev_priv)->ring_mask);
-	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
-#undef MBOX_UPDATE_DWORDS
-
-	ret = intel_ring_begin(signaller_req, num_dwords);
+	ret = intel_ring_begin(req, (num_rings-1) * 8);
 	if (ret)
 		return ret;
 
 	for_each_engine_id(waiter, dev_priv, id) {
-		u64 gtt_offset = signaller->semaphore.signal_ggtt[id];
+		u64 gtt_offset = req->engine->semaphore.signal_ggtt[id];
 		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
 			continue;
 
-		intel_ring_emit(signaller, GFX_OP_PIPE_CONTROL(6));
-		intel_ring_emit(signaller, PIPE_CONTROL_GLOBAL_GTT_IVB |
-					   PIPE_CONTROL_QW_WRITE |
-					   PIPE_CONTROL_CS_STALL);
-		intel_ring_emit(signaller, lower_32_bits(gtt_offset));
-		intel_ring_emit(signaller, upper_32_bits(gtt_offset));
-		intel_ring_emit(signaller, signaller_req->seqno);
-		intel_ring_emit(signaller, 0);
-		intel_ring_emit(signaller, MI_SEMAPHORE_SIGNAL |
-					   MI_SEMAPHORE_TARGET(waiter->hw_id));
-		intel_ring_emit(signaller, 0);
+		intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+		intel_ring_emit(ring,
+				PIPE_CONTROL_GLOBAL_GTT_IVB |
+				PIPE_CONTROL_QW_WRITE |
+				PIPE_CONTROL_CS_STALL);
+		intel_ring_emit(ring, lower_32_bits(gtt_offset));
+		intel_ring_emit(ring, upper_32_bits(gtt_offset));
+		intel_ring_emit(ring, req->fence.seqno);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring,
+				MI_SEMAPHORE_SIGNAL |
+				MI_SEMAPHORE_TARGET(waiter->hw_id));
+		intel_ring_emit(ring, 0);
 	}
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
-static int gen8_xcs_signal(struct drm_i915_gem_request *signaller_req,
-			   unsigned int num_dwords)
+static int gen8_xcs_signal(struct drm_i915_gem_request *req)
 {
-#define MBOX_UPDATE_DWORDS 6
-	struct intel_engine_cs *signaller = signaller_req->engine;
-	struct drm_i915_private *dev_priv = signaller_req->i915;
+	struct intel_ring *ring = req->ring;
+	struct drm_i915_private *dev_priv = req->i915;
 	struct intel_engine_cs *waiter;
 	enum intel_engine_id id;
 	int ret, num_rings;
 
 	num_rings = hweight32(INTEL_INFO(dev_priv)->ring_mask);
-	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
-#undef MBOX_UPDATE_DWORDS
-
-	ret = intel_ring_begin(signaller_req, num_dwords);
+	ret = intel_ring_begin(req, (num_rings-1) * 6);
 	if (ret)
 		return ret;
 
 	for_each_engine_id(waiter, dev_priv, id) {
-		u64 gtt_offset = signaller->semaphore.signal_ggtt[id];
+		u64 gtt_offset = req->engine->semaphore.signal_ggtt[id];
 		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
 			continue;
 
-		intel_ring_emit(signaller, (MI_FLUSH_DW + 1) |
-					   MI_FLUSH_DW_OP_STOREDW);
-		intel_ring_emit(signaller, lower_32_bits(gtt_offset) |
-					   MI_FLUSH_DW_USE_GTT);
-		intel_ring_emit(signaller, upper_32_bits(gtt_offset));
-		intel_ring_emit(signaller, signaller_req->seqno);
-		intel_ring_emit(signaller, MI_SEMAPHORE_SIGNAL |
-					   MI_SEMAPHORE_TARGET(waiter->hw_id));
-		intel_ring_emit(signaller, 0);
+		intel_ring_emit(ring,
+				(MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW);
+		intel_ring_emit(ring,
+				lower_32_bits(gtt_offset) |
+				MI_FLUSH_DW_USE_GTT);
+		intel_ring_emit(ring, upper_32_bits(gtt_offset));
+		intel_ring_emit(ring, req->fence.seqno);
+		intel_ring_emit(ring,
+				MI_SEMAPHORE_SIGNAL |
+				MI_SEMAPHORE_TARGET(waiter->hw_id));
+		intel_ring_emit(ring, 0);
 	}
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
-static int gen6_signal(struct drm_i915_gem_request *signaller_req,
-		       unsigned int num_dwords)
+static int gen6_signal(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *signaller = signaller_req->engine;
-	struct drm_i915_private *dev_priv = signaller_req->i915;
+	struct intel_ring *ring = req->ring;
+	struct drm_i915_private *dev_priv = req->i915;
 	struct intel_engine_cs *useless;
 	enum intel_engine_id id;
 	int ret, num_rings;
 
-#define MBOX_UPDATE_DWORDS 3
 	num_rings = hweight32(INTEL_INFO(dev_priv)->ring_mask);
-	num_dwords += round_up((num_rings-1) * MBOX_UPDATE_DWORDS, 2);
-#undef MBOX_UPDATE_DWORDS
-
-	ret = intel_ring_begin(signaller_req, num_dwords);
+	ret = intel_ring_begin(req, round_up((num_rings-1) * 3, 2));
 	if (ret)
 		return ret;
 
 	for_each_engine_id(useless, dev_priv, id) {
-		i915_reg_t mbox_reg = signaller->semaphore.mbox.signal[id];
+		i915_reg_t mbox_reg = req->engine->semaphore.mbox.signal[id];
 
 		if (i915_mmio_reg_valid(mbox_reg)) {
-			intel_ring_emit(signaller, MI_LOAD_REGISTER_IMM(1));
-			intel_ring_emit_reg(signaller, mbox_reg);
-			intel_ring_emit(signaller, signaller_req->seqno);
+			intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+			intel_ring_emit_reg(ring, mbox_reg);
+			intel_ring_emit(ring, req->fence.seqno);
 		}
 	}
 
 	/* If num_dwords was rounded, make sure the tail pointer is correct */
 	if (num_rings % 2 == 0)
-		intel_ring_emit(signaller, MI_NOOP);
+		intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
+
+	return 0;
+}
+
+static void i9xx_submit_request(struct drm_i915_gem_request *request)
+{
+	struct drm_i915_private *dev_priv = request->i915;
+
+	I915_WRITE_TAIL(request->engine,
+			intel_ring_offset(request->ring, request->tail));
+}
+
+static int i9xx_emit_request(struct drm_i915_gem_request *req)
+{
+	struct intel_ring *ring = req->ring;
+	int ret;
+
+	ret = intel_ring_begin(req, 4);
+	if (ret)
+		return ret;
+
+	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
+	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
+	intel_ring_emit(ring, req->fence.seqno);
+	intel_ring_emit(ring, MI_USER_INTERRUPT);
+	intel_ring_advance(ring);
+
+	req->tail = ring->tail;
 
 	return 0;
 }
 
 /**
- * gen6_add_request - Update the semaphore mailbox registers
+ * gen6_sema_emit_request - Update the semaphore mailbox registers
  *
  * @request - request to write to the ring
  *
  * Update the mailbox registers in the *other* rings with the current seqno.
  * This acts like a signal in the canonical semaphore.
  */
-static int
-gen6_add_request(struct drm_i915_gem_request *req)
+static int gen6_sema_emit_request(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = req->engine;
 	int ret;
 
-	if (engine->semaphore.signal)
-		ret = engine->semaphore.signal(req, 4);
-	else
-		ret = intel_ring_begin(req, 4);
-
+	ret = req->engine->semaphore.signal(req);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_STORE_DWORD_INDEX);
-	intel_ring_emit(engine,
-			I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
-	intel_ring_emit(engine, req->seqno);
-	intel_ring_emit(engine, MI_USER_INTERRUPT);
-	__intel_ring_advance(engine);
-
-	return 0;
+	return i9xx_emit_request(req);
 }
 
-static int
-gen8_render_add_request(struct drm_i915_gem_request *req)
+static int gen8_render_emit_request(struct drm_i915_gem_request *req)
 {
 	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
-	if (engine->semaphore.signal)
-		ret = engine->semaphore.signal(req, 8);
-	else
-		ret = intel_ring_begin(req, 8);
+	if (engine->semaphore.signal) {
+		ret = engine->semaphore.signal(req);
+		if (ret)
+			return ret;
+	}
+
+	ret = intel_ring_begin(req, 8);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, GFX_OP_PIPE_CONTROL(6));
-	intel_ring_emit(engine, (PIPE_CONTROL_GLOBAL_GTT_IVB |
-				 PIPE_CONTROL_CS_STALL |
-				 PIPE_CONTROL_QW_WRITE));
-	intel_ring_emit(engine, intel_hws_seqno_address(req->engine));
-	intel_ring_emit(engine, 0);
-	intel_ring_emit(engine, i915_gem_request_get_seqno(req));
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+	intel_ring_emit(ring, (PIPE_CONTROL_GLOBAL_GTT_IVB |
+			       PIPE_CONTROL_CS_STALL |
+			       PIPE_CONTROL_QW_WRITE));
+	intel_ring_emit(ring, intel_hws_seqno_address(engine));
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, i915_gem_request_get_seqno(req));
 	/* We're thrashing one dword of HWS. */
-	intel_ring_emit(engine, 0);
-	intel_ring_emit(engine, MI_USER_INTERRUPT);
-	intel_ring_emit(engine, MI_NOOP);
-	__intel_ring_advance(engine);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, MI_USER_INTERRUPT);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
-	return 0;
-}
+	req->tail = ring->tail;
 
-static inline bool i915_gem_has_seqno_wrapped(struct drm_i915_private *dev_priv,
-					      u32 seqno)
-{
-	return dev_priv->last_seqno < seqno;
+	return 0;
 }
 
 /**
@@ -1525,82 +1502,71 @@ static inline bool i915_gem_has_seqno_wrapped(struct drm_i915_private *dev_priv,
  */
 
 static int
-gen8_ring_sync(struct drm_i915_gem_request *waiter_req,
-	       struct intel_engine_cs *signaller,
-	       u32 seqno)
+gen8_ring_sync_to(struct drm_i915_gem_request *req,
+		  struct drm_i915_gem_request *signal)
 {
-	struct intel_engine_cs *waiter = waiter_req->engine;
-	struct drm_i915_private *dev_priv = waiter_req->i915;
-	u64 offset = GEN8_WAIT_OFFSET(waiter, signaller->id);
+	struct intel_ring *ring = req->ring;
+	struct drm_i915_private *dev_priv = req->i915;
+	u64 offset = GEN8_WAIT_OFFSET(req->engine, signal->engine->id);
 	struct i915_hw_ppgtt *ppgtt;
 	int ret;
 
-	ret = intel_ring_begin(waiter_req, 4);
+	ret = intel_ring_begin(req, 4);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(waiter, MI_SEMAPHORE_WAIT |
-				MI_SEMAPHORE_GLOBAL_GTT |
-				MI_SEMAPHORE_SAD_GTE_SDD);
-	intel_ring_emit(waiter, seqno);
-	intel_ring_emit(waiter, lower_32_bits(offset));
-	intel_ring_emit(waiter, upper_32_bits(offset));
-	intel_ring_advance(waiter);
+	intel_ring_emit(ring,
+			MI_SEMAPHORE_WAIT |
+			MI_SEMAPHORE_GLOBAL_GTT |
+			MI_SEMAPHORE_SAD_GTE_SDD);
+	intel_ring_emit(ring, signal->fence.seqno);
+	intel_ring_emit(ring, lower_32_bits(offset));
+	intel_ring_emit(ring, upper_32_bits(offset));
+	intel_ring_advance(ring);
 
 	/* When the !RCS engines idle waiting upon a semaphore, they lose their
 	 * pagetables and we must reload them before executing the batch.
 	 * We do this on the i915_switch_context() following the wait and
 	 * before the dispatch.
 	 */
-	ppgtt = waiter_req->ctx->ppgtt;
-	if (ppgtt && waiter_req->engine->id != RCS)
-		ppgtt->pd_dirty_rings |= intel_engine_flag(waiter_req->engine);
+	ppgtt = req->ctx->ppgtt;
+	if (ppgtt && req->engine->id != RCS)
+		ppgtt->pd_dirty_rings |= intel_engine_flag(req->engine);
 	return 0;
 }
 
 static int
-gen6_ring_sync(struct drm_i915_gem_request *waiter_req,
-	       struct intel_engine_cs *signaller,
-	       u32 seqno)
+gen6_ring_sync_to(struct drm_i915_gem_request *req,
+		  struct drm_i915_gem_request *signal)
 {
-	struct intel_engine_cs *waiter = waiter_req->engine;
+	struct intel_ring *ring = req->ring;
 	u32 dw1 = MI_SEMAPHORE_MBOX |
 		  MI_SEMAPHORE_COMPARE |
 		  MI_SEMAPHORE_REGISTER;
-	u32 wait_mbox = signaller->semaphore.mbox.wait[waiter->id];
+	u32 wait_mbox = signal->engine->semaphore.mbox.wait[req->engine->id];
 	int ret;
 
-	/* Throughout all of the GEM code, seqno passed implies our current
-	 * seqno is >= the last seqno executed. However for hardware the
-	 * comparison is strictly greater than.
-	 */
-	seqno -= 1;
-
 	WARN_ON(wait_mbox == MI_SEMAPHORE_SYNC_INVALID);
 
-	ret = intel_ring_begin(waiter_req, 4);
+	ret = intel_ring_begin(req, 4);
 	if (ret)
 		return ret;
 
-	/* If seqno wrap happened, omit the wait with no-ops */
-	if (likely(!i915_gem_has_seqno_wrapped(waiter_req->i915, seqno))) {
-		intel_ring_emit(waiter, dw1 | wait_mbox);
-		intel_ring_emit(waiter, seqno);
-		intel_ring_emit(waiter, 0);
-		intel_ring_emit(waiter, MI_NOOP);
-	} else {
-		intel_ring_emit(waiter, MI_NOOP);
-		intel_ring_emit(waiter, MI_NOOP);
-		intel_ring_emit(waiter, MI_NOOP);
-		intel_ring_emit(waiter, MI_NOOP);
-	}
-	intel_ring_advance(waiter);
+	intel_ring_emit(ring, dw1 | wait_mbox);
+	/* Throughout all of the GEM code, seqno passed implies our current
+	 * seqno is >= the last seqno executed. However for hardware the
+	 * comparison is strictly greater than.
+	 */
+	intel_ring_emit(ring, signal->fence.seqno - 1);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static void
-gen5_seqno_barrier(struct intel_engine_cs *ring)
+gen5_seqno_barrier(struct intel_engine_cs *engine)
 {
 	/* MI_STORE are internally buffered by the GPU and not flushed
 	 * either by MI_FLUSH or SyncFlush or any other combination of
@@ -1693,40 +1659,18 @@ i8xx_irq_disable(struct intel_engine_cs *engine)
 }
 
 static int
-bsd_ring_flush(struct drm_i915_gem_request *req,
-	       u32     invalidate_domains,
-	       u32     flush_domains)
+bsd_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_FLUSH);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
-	return 0;
-}
-
-static int
-i9xx_add_request(struct drm_i915_gem_request *req)
-{
-	struct intel_engine_cs *engine = req->engine;
-	int ret;
-
-	ret = intel_ring_begin(req, 4);
-	if (ret)
-		return ret;
-
-	intel_ring_emit(engine, MI_STORE_DWORD_INDEX);
-	intel_ring_emit(engine,
-			I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
-	intel_ring_emit(engine, req->seqno);
-	intel_ring_emit(engine, MI_USER_INTERRUPT);
-	__intel_ring_advance(engine);
-
+	intel_ring_emit(ring, MI_FLUSH);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 	return 0;
 }
 
@@ -1788,24 +1732,24 @@ gen8_irq_disable(struct intel_engine_cs *engine)
 }
 
 static int
-i965_dispatch_execbuffer(struct drm_i915_gem_request *req,
-			 u64 offset, u32 length,
-			 unsigned dispatch_flags)
+i965_emit_bb_start(struct drm_i915_gem_request *req,
+		   u64 offset, u32 length,
+		   unsigned int dispatch_flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine,
+	intel_ring_emit(ring,
 			MI_BATCH_BUFFER_START |
 			MI_BATCH_GTT |
 			(dispatch_flags & I915_DISPATCH_SECURE ?
 			 0 : MI_BATCH_NON_SECURE_I965));
-	intel_ring_emit(engine, offset);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, offset);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -1815,12 +1759,12 @@ i965_dispatch_execbuffer(struct drm_i915_gem_request *req,
 #define I830_TLB_ENTRIES (2)
 #define I830_WA_SIZE max(I830_TLB_ENTRIES*4096, I830_BATCH_LIMIT)
 static int
-i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
-			 u64 offset, u32 len,
-			 unsigned dispatch_flags)
+i830_emit_bb_start(struct drm_i915_gem_request *req,
+		   u64 offset, u32 len,
+		   unsigned int dispatch_flags)
 {
-	struct intel_engine_cs *engine = req->engine;
-	u32 cs_offset = engine->scratch.gtt_offset;
+	struct intel_ring *ring = req->ring;
+	u32 cs_offset = req->engine->scratch.gtt_offset;
 	int ret;
 
 	ret = intel_ring_begin(req, 6);
@@ -1828,13 +1772,13 @@ i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
 		return ret;
 
 	/* Evict the invalid PTE TLBs */
-	intel_ring_emit(engine, COLOR_BLT_CMD | BLT_WRITE_RGBA);
-	intel_ring_emit(engine, BLT_DEPTH_32 | BLT_ROP_COLOR_COPY | 4096);
-	intel_ring_emit(engine, I830_TLB_ENTRIES << 16 | 4); /* load each page */
-	intel_ring_emit(engine, cs_offset);
-	intel_ring_emit(engine, 0xdeadbeef);
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, COLOR_BLT_CMD | BLT_WRITE_RGBA);
+	intel_ring_emit(ring, BLT_DEPTH_32 | BLT_ROP_COLOR_COPY | 4096);
+	intel_ring_emit(ring, I830_TLB_ENTRIES << 16 | 4); /* load each page */
+	intel_ring_emit(ring, cs_offset);
+	intel_ring_emit(ring, 0xdeadbeef);
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	if ((dispatch_flags & I915_DISPATCH_PINNED) == 0) {
 		if (len > I830_BATCH_LIMIT)
@@ -1848,17 +1792,17 @@ i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
 		 * stable batch scratch bo area (so that the CS never
 		 * stumbles over its tlb invalidation bug) ...
 		 */
-		intel_ring_emit(engine, SRC_COPY_BLT_CMD | BLT_WRITE_RGBA);
-		intel_ring_emit(engine,
+		intel_ring_emit(ring, SRC_COPY_BLT_CMD | BLT_WRITE_RGBA);
+		intel_ring_emit(ring,
 				BLT_DEPTH_32 | BLT_ROP_SRC_COPY | 4096);
-		intel_ring_emit(engine, DIV_ROUND_UP(len, 4096) << 16 | 4096);
-		intel_ring_emit(engine, cs_offset);
-		intel_ring_emit(engine, 4096);
-		intel_ring_emit(engine, offset);
+		intel_ring_emit(ring, DIV_ROUND_UP(len, 4096) << 16 | 4096);
+		intel_ring_emit(ring, cs_offset);
+		intel_ring_emit(ring, 4096);
+		intel_ring_emit(ring, offset);
 
-		intel_ring_emit(engine, MI_FLUSH);
-		intel_ring_emit(engine, MI_NOOP);
-		intel_ring_advance(engine);
+		intel_ring_emit(ring, MI_FLUSH);
+		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_advance(ring);
 
 		/* ... and execute it. */
 		offset = cs_offset;
@@ -1868,30 +1812,30 @@ i830_dispatch_execbuffer(struct drm_i915_gem_request *req,
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
-	intel_ring_emit(engine, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
-					  0 : MI_BATCH_NON_SECURE));
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
+	intel_ring_emit(ring, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
+					0 : MI_BATCH_NON_SECURE));
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int
-i915_dispatch_execbuffer(struct drm_i915_gem_request *req,
-			 u64 offset, u32 len,
-			 unsigned dispatch_flags)
+i915_emit_bb_start(struct drm_i915_gem_request *req,
+		   u64 offset, u32 len,
+		   unsigned int dispatch_flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
-	intel_ring_emit(engine, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
-					  0 : MI_BATCH_NON_SECURE));
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, MI_BATCH_BUFFER_START | MI_BATCH_GTT);
+	intel_ring_emit(ring, offset | (dispatch_flags & I915_DISPATCH_SECURE ?
+					0 : MI_BATCH_NON_SECURE));
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -1917,7 +1861,7 @@ static void cleanup_status_page(struct intel_engine_cs *engine)
 
 	kunmap(sg_page(obj->pages->sgl));
 	i915_gem_object_ggtt_unpin(obj);
-	drm_gem_object_unreference(&obj->base);
+	i915_gem_object_put(obj);
 	engine->status_page.obj = NULL;
 }
 
@@ -1952,10 +1896,10 @@ static int init_status_page(struct intel_engine_cs *engine)
 			 * actualy map it).
 			 */
 			flags |= PIN_MAPPABLE;
-		ret = i915_gem_obj_ggtt_pin(obj, 4096, flags);
+		ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 4096, flags);
 		if (ret) {
 err_unref:
-			drm_gem_object_unreference(&obj->base);
+			i915_gem_object_put(obj);
 			return ret;
 		}
 
@@ -1989,32 +1933,17 @@ static int init_phys_status_page(struct intel_engine_cs *engine)
 	return 0;
 }
 
-void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
-{
-	GEM_BUG_ON(ringbuf->vma == NULL);
-	GEM_BUG_ON(ringbuf->virtual_start == NULL);
-
-	if (HAS_LLC(ringbuf->obj->base.dev) && !ringbuf->obj->stolen)
-		i915_gem_object_unpin_map(ringbuf->obj);
-	else
-		i915_vma_unpin_iomap(ringbuf->vma);
-	ringbuf->virtual_start = NULL;
-
-	i915_gem_object_ggtt_unpin(ringbuf->obj);
-	ringbuf->vma = NULL;
-}
-
-int intel_pin_and_map_ringbuffer_obj(struct drm_i915_private *dev_priv,
-				     struct intel_ringbuffer *ringbuf)
+int intel_ring_pin(struct intel_ring *ring)
 {
-	struct drm_i915_gem_object *obj = ringbuf->obj;
+	struct drm_i915_private *dev_priv = ring->engine->i915;
+	struct drm_i915_gem_object *obj = ring->obj;
 	/* Ring wraparound at offset 0 sometimes hangs. No idea why. */
 	unsigned flags = PIN_OFFSET_BIAS | 4096;
 	void *addr;
 	int ret;
 
 	if (HAS_LLC(dev_priv) && !obj->stolen) {
-		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, flags);
+		ret = i915_gem_object_ggtt_pin(obj, NULL, 0, PAGE_SIZE, flags);
 		if (ret)
 			return ret;
 
@@ -2028,8 +1957,8 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_i915_private *dev_priv,
 			goto err_unpin;
 		}
 	} else {
-		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE,
-					    flags | PIN_MAPPABLE);
+		ret = i915_gem_object_ggtt_pin(obj, NULL, 0, PAGE_SIZE,
+					       flags | PIN_MAPPABLE);
 		if (ret)
 			return ret;
 
@@ -2040,15 +1969,16 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_i915_private *dev_priv,
 		/* Access through the GTT requires the device to be awake. */
 		assert_rpm_wakelock_held(dev_priv);
 
-		addr = i915_vma_pin_iomap(i915_gem_obj_to_ggtt(obj));
+		addr = (void __force *)
+			i915_vma_pin_iomap(i915_gem_obj_to_ggtt(obj));
 		if (IS_ERR(addr)) {
 			ret = PTR_ERR(addr);
 			goto err_unpin;
 		}
 	}
 
-	ringbuf->virtual_start = addr;
-	ringbuf->vma = i915_gem_obj_to_ggtt(obj);
+	ring->vaddr = addr;
+	ring->vma = i915_gem_obj_to_ggtt(obj);
 	return 0;
 
 err_unpin:
@@ -2056,39 +1986,56 @@ err_unpin:
 	return ret;
 }
 
-static void intel_destroy_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
+void intel_ring_unpin(struct intel_ring *ring)
 {
-	drm_gem_object_unreference(&ringbuf->obj->base);
-	ringbuf->obj = NULL;
+	GEM_BUG_ON(!ring->vma);
+	GEM_BUG_ON(!ring->vaddr);
+
+	if (HAS_LLC(ring->engine->i915) && !ring->obj->stolen)
+		i915_gem_object_unpin_map(ring->obj);
+	else
+		i915_vma_unpin_iomap(ring->vma);
+	ring->vaddr = NULL;
+
+	i915_gem_object_ggtt_unpin(ring->obj);
+	ring->vma = NULL;
+}
+
+static void intel_destroy_ringbuffer_obj(struct intel_ring *ring)
+{
+	i915_gem_object_put(ring->obj);
+	ring->obj = NULL;
 }
 
 static int intel_alloc_ringbuffer_obj(struct drm_device *dev,
-				      struct intel_ringbuffer *ringbuf)
+				      struct intel_ring *ring)
 {
 	struct drm_i915_gem_object *obj;
 
 	obj = NULL;
 	if (!HAS_LLC(dev))
-		obj = i915_gem_object_create_stolen(dev, ringbuf->size);
+		obj = i915_gem_object_create_stolen(dev, ring->size);
 	if (obj == NULL)
-		obj = i915_gem_object_create(dev, ringbuf->size);
+		obj = i915_gem_object_create(dev, ring->size);
 	if (IS_ERR(obj))
 		return PTR_ERR(obj);
 
 	/* mark ring buffers as read-only from GPU side by default */
 	obj->gt_ro = 1;
 
-	ringbuf->obj = obj;
+	ring->obj = obj;
 
 	return 0;
 }
 
-struct intel_ringbuffer *
-intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size)
+struct intel_ring *
+intel_engine_create_ring(struct intel_engine_cs *engine, int size)
 {
-	struct intel_ringbuffer *ring;
+	struct intel_ring *ring;
 	int ret;
 
+	GEM_BUG_ON(!is_power_of_2(size));
+
 	ring = kzalloc(sizeof(*ring), GFP_KERNEL);
 	if (ring == NULL) {
 		DRM_DEBUG_DRIVER("Failed to allocate ringbuffer %s\n",
@@ -2099,6 +2046,8 @@ intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size)
 	ring->engine = engine;
 	list_add(&ring->link, &engine->buffers);
 
+	INIT_LIST_HEAD(&ring->request_list);
+
 	ring->size = size;
 	/* Workaround an erratum on the i830 which causes a hang if
 	 * the TAIL pointer points to within the last 2 cachelines
@@ -2124,7 +2073,7 @@ intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size)
 }
 
 void
-intel_ringbuffer_free(struct intel_ringbuffer *ring)
+intel_ring_free(struct intel_ring *ring)
 {
 	intel_destroy_ringbuffer_obj(ring);
 	list_del(&ring->link);
@@ -2143,7 +2092,8 @@ static int intel_ring_context_pin(struct i915_gem_context *ctx,
 		return 0;
 
 	if (ce->state) {
-		ret = i915_gem_obj_ggtt_pin(ce->state, ctx->ggtt_alignment, 0);
+		ret = i915_gem_object_ggtt_pin(ce->state, NULL, 0,
+					       ctx->ggtt_alignment, 0);
 		if (ret)
 			goto error;
 	}
@@ -2158,7 +2108,7 @@ static int intel_ring_context_pin(struct i915_gem_context *ctx,
 	if (ctx == ctx->i915->kernel_context)
 		ce->initialised = true;
 
-	i915_gem_context_reference(ctx);
+	i915_gem_context_get(ctx);
 	return 0;
 
 error:
@@ -2179,28 +2129,23 @@ static void intel_ring_context_unpin(struct i915_gem_context *ctx,
 	if (ce->state)
 		i915_gem_object_ggtt_unpin(ce->state);
 
-	i915_gem_context_unreference(ctx);
+	i915_gem_context_put(ctx);
 }
 
-static int intel_init_ring_buffer(struct drm_device *dev,
-				  struct intel_engine_cs *engine)
+static int intel_init_ring_buffer(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_ringbuffer *ringbuf;
+	struct drm_i915_private *dev_priv = engine->i915;
+	struct intel_ring *ring;
 	int ret;
 
 	WARN_ON(engine->buffer);
 
-	engine->i915 = dev_priv;
-	INIT_LIST_HEAD(&engine->active_list);
-	INIT_LIST_HEAD(&engine->request_list);
-	INIT_LIST_HEAD(&engine->execlist_queue);
-	INIT_LIST_HEAD(&engine->buffers);
-	i915_gem_batch_pool_init(dev, &engine->batch_pool);
+	intel_engine_setup_common(engine);
+
 	memset(engine->semaphore.sync_seqno, 0,
 	       sizeof(engine->semaphore.sync_seqno));
 
-	ret = intel_engine_init_breadcrumbs(engine);
+	ret = intel_engine_init_common(engine);
 	if (ret)
 		goto error;
 
@@ -2215,12 +2160,12 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	if (ret)
 		goto error;
 
-	ringbuf = intel_engine_create_ringbuffer(engine, 32 * PAGE_SIZE);
-	if (IS_ERR(ringbuf)) {
-		ret = PTR_ERR(ringbuf);
+	ring = intel_engine_create_ring(engine, 32 * PAGE_SIZE);
+	if (IS_ERR(ring)) {
+		ret = PTR_ERR(ring);
 		goto error;
 	}
-	engine->buffer = ringbuf;
+	engine->buffer = ring;
 
 	if (I915_NEED_GFX_HWS(dev_priv)) {
 		ret = init_status_page(engine);
@@ -2233,26 +2178,22 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 			goto error;
 	}
 
-	ret = intel_pin_and_map_ringbuffer_obj(dev_priv, ringbuf);
+	ret = intel_ring_pin(ring);
 	if (ret) {
 		DRM_ERROR("Failed to pin and map ringbuffer %s: %d\n",
 				engine->name, ret);
-		intel_destroy_ringbuffer_obj(ringbuf);
+		intel_destroy_ringbuffer_obj(ring);
 		goto error;
 	}
 
-	ret = i915_cmd_parser_init_ring(engine);
-	if (ret)
-		goto error;
-
 	return 0;
 
 error:
-	intel_cleanup_engine(engine);
+	intel_engine_cleanup(engine);
 	return ret;
 }
 
-void intel_cleanup_engine(struct intel_engine_cs *engine)
+void intel_engine_cleanup(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv;
 
@@ -2262,11 +2203,10 @@ void intel_cleanup_engine(struct intel_engine_cs *engine)
 	dev_priv = engine->i915;
 
 	if (engine->buffer) {
-		intel_stop_engine(engine);
 		WARN_ON(!IS_GEN2(dev_priv) && (I915_READ_MODE(engine) & MODE_IDLE) == 0);
 
-		intel_unpin_ringbuffer_obj(engine->buffer);
-		intel_ringbuffer_free(engine->buffer);
+		intel_ring_unpin(engine->buffer);
+		intel_ring_free(engine->buffer);
 		engine->buffer = NULL;
 	}
 
@@ -2280,33 +2220,13 @@ void intel_cleanup_engine(struct intel_engine_cs *engine)
 		cleanup_phys_status_page(engine);
 	}
 
-	i915_cmd_parser_fini_ring(engine);
-	i915_gem_batch_pool_fini(&engine->batch_pool);
-	intel_engine_fini_breadcrumbs(engine);
+	intel_engine_cleanup_common(engine);
 
 	intel_ring_context_unpin(dev_priv->kernel_context, engine);
 
 	engine->i915 = NULL;
 }
 
-int intel_engine_idle(struct intel_engine_cs *engine)
-{
-	struct drm_i915_gem_request *req;
-
-	/* Wait upon the last request to be completed */
-	if (list_empty(&engine->request_list))
-		return 0;
-
-	req = list_entry(engine->request_list.prev,
-			 struct drm_i915_gem_request,
-			 list);
-
-	/* Make sure we do not trigger any retires */
-	return __i915_wait_request(req,
-				   req->i915->mm.interruptible,
-				   NULL, NULL);
-}
-
 int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request)
 {
 	int ret;
@@ -2317,7 +2237,7 @@ int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request)
 	 */
 	request->reserved_space += LEGACY_REQUEST_SIZE;
 
-	request->ringbuf = request->engine->buffer;
+	request->ring = request->engine->buffer;
 
 	ret = intel_ring_begin(request, 0);
 	if (ret)
@@ -2329,12 +2249,12 @@ int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request)
 
 static int wait_for_space(struct drm_i915_gem_request *req, int bytes)
 {
-	struct intel_ringbuffer *ringbuf = req->ringbuf;
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	struct drm_i915_gem_request *target;
+	int ret;
 
-	intel_ring_update_space(ringbuf);
-	if (ringbuf->space >= bytes)
+	intel_ring_update_space(ring);
+	if (ring->space >= bytes)
 		return 0;
 
 	/*
@@ -2348,35 +2268,38 @@ static int wait_for_space(struct drm_i915_gem_request *req, int bytes)
 	 */
 	GEM_BUG_ON(!req->reserved_space);
 
-	list_for_each_entry(target, &engine->request_list, list) {
+	list_for_each_entry(target, &ring->request_list, ring_link) {
 		unsigned space;
 
-		/*
-		 * The request queue is per-engine, so can contain requests
-		 * from multiple ringbuffers. Here, we must ignore any that
-		 * aren't from the ringbuffer we're considering.
-		 */
-		if (target->ringbuf != ringbuf)
-			continue;
-
 		/* Would completion of this request free enough space? */
-		space = __intel_ring_space(target->postfix, ringbuf->tail,
-					   ringbuf->size);
+		space = __intel_ring_space(target->postfix, ring->tail,
+					   ring->size);
 		if (space >= bytes)
 			break;
 	}
 
-	if (WARN_ON(&target->list == &engine->request_list))
+	if (WARN_ON(&target->ring_link == &ring->request_list))
 		return -ENOSPC;
 
-	return i915_wait_request(target);
+	ret = i915_wait_request(target, true, NULL, NO_WAITBOOST);
+	if (ret)
+		return ret;
+
+	if (i915_reset_in_progress(&target->i915->gpu_error))
+		return -EAGAIN;
+
+	i915_gem_request_retire_upto(target);
+
+	intel_ring_update_space(ring);
+	GEM_BUG_ON(ring->space < bytes);
+	return 0;
 }
 
 int intel_ring_begin(struct drm_i915_gem_request *req, int num_dwords)
 {
-	struct intel_ringbuffer *ringbuf = req->ringbuf;
-	int remain_actual = ringbuf->size - ringbuf->tail;
-	int remain_usable = ringbuf->effective_size - ringbuf->tail;
+	struct intel_ring *ring = req->ring;
+	int remain_actual = ring->size - ring->tail;
+	int remain_usable = ring->effective_size - ring->tail;
 	int bytes = num_dwords * sizeof(u32);
 	int total_bytes, wait_bytes;
 	bool need_wrap = false;
@@ -2403,37 +2326,33 @@ int intel_ring_begin(struct drm_i915_gem_request *req, int num_dwords)
 		wait_bytes = total_bytes;
 	}
 
-	if (wait_bytes > ringbuf->space) {
+	if (wait_bytes > ring->space) {
 		int ret = wait_for_space(req, wait_bytes);
 		if (unlikely(ret))
 			return ret;
-
-		intel_ring_update_space(ringbuf);
-		if (unlikely(ringbuf->space < wait_bytes))
-			return -EAGAIN;
 	}
 
 	if (unlikely(need_wrap)) {
-		GEM_BUG_ON(remain_actual > ringbuf->space);
-		GEM_BUG_ON(ringbuf->tail + remain_actual > ringbuf->size);
+		GEM_BUG_ON(remain_actual > ring->space);
+		GEM_BUG_ON(ring->tail + remain_actual > ring->size);
 
 		/* Fill the tail with MI_NOOP */
-		memset(ringbuf->virtual_start + ringbuf->tail,
-		       0, remain_actual);
-		ringbuf->tail = 0;
-		ringbuf->space -= remain_actual;
+		memset(ring->vaddr + ring->tail, 0, remain_actual);
+		ring->tail = 0;
+		ring->space -= remain_actual;
 	}
 
-	ringbuf->space -= bytes;
-	GEM_BUG_ON(ringbuf->space < 0);
+	ring->space -= bytes;
+	GEM_BUG_ON(ring->space < 0);
 	return 0;
 }
 
 /* Align the ring tail to a cacheline boundary */
 int intel_ring_cacheline_align(struct drm_i915_gem_request *req)
 {
-	struct intel_engine_cs *engine = req->engine;
-	int num_dwords = (engine->buffer->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
+	struct intel_ring *ring = req->ring;
+	int num_dwords =
+		(ring->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
 	int ret;
 
 	if (num_dwords == 0)
@@ -2445,14 +2364,14 @@ int intel_ring_cacheline_align(struct drm_i915_gem_request *req)
 		return ret;
 
 	while (num_dwords--)
-		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_emit(ring, MI_NOOP);
 
-	intel_ring_advance(engine);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
-void intel_ring_init_seqno(struct intel_engine_cs *engine, u32 seqno)
+void intel_engine_init_seqno(struct intel_engine_cs *engine, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
 
@@ -2496,10 +2415,9 @@ void intel_ring_init_seqno(struct intel_engine_cs *engine, u32 seqno)
 	rcu_read_unlock();
 }
 
-static void gen6_bsd_ring_write_tail(struct intel_engine_cs *engine,
-				     u32 value)
+static void gen6_bsd_submit_request(struct drm_i915_gem_request *request)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
+	struct drm_i915_private *dev_priv = request->i915;
 
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
@@ -2523,8 +2441,7 @@ static void gen6_bsd_ring_write_tail(struct intel_engine_cs *engine,
 		DRM_ERROR("timed out waiting for the BSD ring to wake up\n");
 
 	/* Now that the ring is fully powered up, update the tail */
-	I915_WRITE_FW(RING_TAIL(engine->mmio_base), value);
-	POSTING_READ_FW(RING_TAIL(engine->mmio_base));
+	i9xx_submit_request(request);
 
 	/* Let the ring send IDLE messages to the GT again,
 	 * and so let it sleep to conserve power when idle.
@@ -2535,10 +2452,9 @@ static void gen6_bsd_ring_write_tail(struct intel_engine_cs *engine,
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 }
 
-static int gen6_bsd_ring_flush(struct drm_i915_gem_request *req,
-			       u32 invalidate, u32 flush)
+static int gen6_bsd_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	uint32_t cmd;
 	int ret;
 
@@ -2563,30 +2479,29 @@ static int gen6_bsd_ring_flush(struct drm_i915_gem_request *req,
 	 * operation is complete. This bit is only valid when the
 	 * Post-Sync Operation field is a value of 1h or 3h."
 	 */
-	if (invalidate & I915_GEM_GPU_DOMAINS)
+	if (mode & EMIT_INVALIDATE)
 		cmd |= MI_INVALIDATE_TLB | MI_INVALIDATE_BSD;
 
-	intel_ring_emit(engine, cmd);
-	intel_ring_emit(engine,
-			I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
+	intel_ring_emit(ring, cmd);
+	intel_ring_emit(ring, I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
 	if (INTEL_GEN(req->i915) >= 8) {
-		intel_ring_emit(engine, 0); /* upper addr */
-		intel_ring_emit(engine, 0); /* value */
+		intel_ring_emit(ring, 0); /* upper addr */
+		intel_ring_emit(ring, 0); /* value */
 	} else  {
-		intel_ring_emit(engine, 0);
-		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, MI_NOOP);
 	}
-	intel_ring_advance(engine);
+	intel_ring_advance(ring);
 	return 0;
 }
 
 static int
-gen8_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
-			      u64 offset, u32 len,
-			      unsigned dispatch_flags)
+gen8_emit_bb_start(struct drm_i915_gem_request *req,
+		   u64 offset, u32 len,
+		   unsigned int dispatch_flags)
 {
-	struct intel_engine_cs *engine = req->engine;
-	bool ppgtt = USES_PPGTT(engine->dev) &&
+	struct intel_ring *ring = req->ring;
+	bool ppgtt = USES_PPGTT(req->i915) &&
 			!(dispatch_flags & I915_DISPATCH_SECURE);
 	int ret;
 
@@ -2595,71 +2510,70 @@ gen8_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
 		return ret;
 
 	/* FIXME(BDW): Address space and security selectors. */
-	intel_ring_emit(engine, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8) |
+	intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8) |
 			(dispatch_flags & I915_DISPATCH_RS ?
 			 MI_BATCH_RESOURCE_STREAMER : 0));
-	intel_ring_emit(engine, lower_32_bits(offset));
-	intel_ring_emit(engine, upper_32_bits(offset));
-	intel_ring_emit(engine, MI_NOOP);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, lower_32_bits(offset));
+	intel_ring_emit(ring, upper_32_bits(offset));
+	intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int
-hsw_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
-			     u64 offset, u32 len,
-			     unsigned dispatch_flags)
+hsw_emit_bb_start(struct drm_i915_gem_request *req,
+		  u64 offset, u32 len,
+		  unsigned int dispatch_flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine,
+	intel_ring_emit(ring,
 			MI_BATCH_BUFFER_START |
 			(dispatch_flags & I915_DISPATCH_SECURE ?
 			 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW) |
 			(dispatch_flags & I915_DISPATCH_RS ?
 			 MI_BATCH_RESOURCE_STREAMER : 0));
 	/* bit0-7 is the length on GEN6+ */
-	intel_ring_emit(engine, offset);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, offset);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 static int
-gen6_ring_dispatch_execbuffer(struct drm_i915_gem_request *req,
-			      u64 offset, u32 len,
-			      unsigned dispatch_flags)
+gen6_emit_bb_start(struct drm_i915_gem_request *req,
+		   u64 offset, u32 len,
+		   unsigned int dispatch_flags)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	int ret;
 
 	ret = intel_ring_begin(req, 2);
 	if (ret)
 		return ret;
 
-	intel_ring_emit(engine,
+	intel_ring_emit(ring,
 			MI_BATCH_BUFFER_START |
 			(dispatch_flags & I915_DISPATCH_SECURE ?
 			 0 : MI_BATCH_NON_SECURE_I965));
 	/* bit0-7 is the length on GEN6+ */
-	intel_ring_emit(engine, offset);
-	intel_ring_advance(engine);
+	intel_ring_emit(ring, offset);
+	intel_ring_advance(ring);
 
 	return 0;
 }
 
 /* Blitter support (SandyBridge+) */
 
-static int gen6_ring_flush(struct drm_i915_gem_request *req,
-			   u32 invalidate, u32 flush)
+static int gen6_ring_flush(struct drm_i915_gem_request *req, u32 mode)
 {
-	struct intel_engine_cs *engine = req->engine;
+	struct intel_ring *ring = req->ring;
 	uint32_t cmd;
 	int ret;
 
@@ -2684,19 +2598,19 @@ static int gen6_ring_flush(struct drm_i915_gem_request *req,
 	 * operation is complete. This bit is only valid when the
 	 * Post-Sync Operation field is a value of 1h or 3h."
 	 */
-	if (invalidate & I915_GEM_DOMAIN_RENDER)
+	if (mode & EMIT_INVALIDATE)
 		cmd |= MI_INVALIDATE_TLB;
-	intel_ring_emit(engine, cmd);
-	intel_ring_emit(engine,
+	intel_ring_emit(ring, cmd);
+	intel_ring_emit(ring,
 			I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
 	if (INTEL_GEN(req->i915) >= 8) {
-		intel_ring_emit(engine, 0); /* upper addr */
-		intel_ring_emit(engine, 0); /* value */
+		intel_ring_emit(ring, 0); /* upper addr */
+		intel_ring_emit(ring, 0); /* value */
 	} else  {
-		intel_ring_emit(engine, 0);
-		intel_ring_emit(engine, MI_NOOP);
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, MI_NOOP);
 	}
-	intel_ring_advance(engine);
+	intel_ring_advance(ring);
 
 	return 0;
 }
@@ -2707,7 +2621,7 @@ static void intel_ring_init_semaphores(struct drm_i915_private *dev_priv,
 	struct drm_i915_gem_object *obj;
 	int ret, i;
 
-	if (!i915_semaphore_is_enabled(dev_priv))
+	if (!i915.semaphores)
 		return;
 
 	if (INTEL_GEN(dev_priv) >= 8 && !dev_priv->semaphore_obj) {
@@ -2717,9 +2631,9 @@ static void intel_ring_init_semaphores(struct drm_i915_private *dev_priv,
 			i915.semaphores = 0;
 		} else {
 			i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
-			ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_NONBLOCK);
+			ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 0, 0);
 			if (ret != 0) {
-				drm_gem_object_unreference(&obj->base);
+				i915_gem_object_put(obj);
 				DRM_ERROR("Failed to pin semaphore bo. Disabling semaphores\n");
 				i915.semaphores = 0;
 			} else {
@@ -2728,13 +2642,13 @@ static void intel_ring_init_semaphores(struct drm_i915_private *dev_priv,
 		}
 	}
 
-	if (!i915_semaphore_is_enabled(dev_priv))
+	if (!i915.semaphores)
 		return;
 
 	if (INTEL_GEN(dev_priv) >= 8) {
 		u64 offset = i915_gem_obj_ggtt_offset(dev_priv->semaphore_obj);
 
-		engine->semaphore.sync_to = gen8_ring_sync;
+		engine->semaphore.sync_to = gen8_ring_sync_to;
 		engine->semaphore.signal = gen8_xcs_signal;
 
 		for (i = 0; i < I915_NUM_ENGINES; i++) {
@@ -2748,7 +2662,7 @@ static void intel_ring_init_semaphores(struct drm_i915_private *dev_priv,
 			engine->semaphore.signal_ggtt[i] = ring_offset;
 		}
 	} else if (INTEL_GEN(dev_priv) >= 6) {
-		engine->semaphore.sync_to = gen6_ring_sync;
+		engine->semaphore.sync_to = gen6_ring_sync_to;
 		engine->semaphore.signal = gen6_signal;
 
 		/*
@@ -2804,6 +2718,8 @@ static void intel_ring_init_semaphores(struct drm_i915_private *dev_priv,
 static void intel_ring_init_irq(struct drm_i915_private *dev_priv,
 				struct intel_engine_cs *engine)
 {
+	engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT << engine->irq_shift;
+
 	if (INTEL_GEN(dev_priv) >= 8) {
 		engine->irq_enable = gen8_irq_enable;
 		engine->irq_disable = gen8_irq_disable;
@@ -2828,74 +2744,66 @@ static void intel_ring_init_irq(struct drm_i915_private *dev_priv,
 static void intel_ring_default_vfuncs(struct drm_i915_private *dev_priv,
 				      struct intel_engine_cs *engine)
 {
+	intel_ring_init_irq(dev_priv, engine);
+	intel_ring_init_semaphores(dev_priv, engine);
+
 	engine->init_hw = init_ring_common;
-	engine->write_tail = ring_write_tail;
 
-	engine->add_request = i9xx_add_request;
-	if (INTEL_GEN(dev_priv) >= 6)
-		engine->add_request = gen6_add_request;
+	engine->emit_request = i9xx_emit_request;
+	if (i915.semaphores)
+		engine->emit_request = gen6_sema_emit_request;
+	engine->submit_request = i9xx_submit_request;
 
 	if (INTEL_GEN(dev_priv) >= 8)
-		engine->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		engine->emit_bb_start = gen8_emit_bb_start;
 	else if (INTEL_GEN(dev_priv) >= 6)
-		engine->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		engine->emit_bb_start = gen6_emit_bb_start;
 	else if (INTEL_GEN(dev_priv) >= 4)
-		engine->dispatch_execbuffer = i965_dispatch_execbuffer;
+		engine->emit_bb_start = i965_emit_bb_start;
 	else if (IS_I830(dev_priv) || IS_845G(dev_priv))
-		engine->dispatch_execbuffer = i830_dispatch_execbuffer;
+		engine->emit_bb_start = i830_emit_bb_start;
 	else
-		engine->dispatch_execbuffer = i915_dispatch_execbuffer;
-
-	intel_ring_init_irq(dev_priv, engine);
-	intel_ring_init_semaphores(dev_priv, engine);
+		engine->emit_bb_start = i915_emit_bb_start;
 }
 
-int intel_init_render_ring_buffer(struct drm_device *dev)
+int intel_init_render_ring_buffer(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_engine_cs *engine = &dev_priv->engine[RCS];
+	struct drm_i915_private *dev_priv = engine->i915;
 	int ret;
 
-	engine->name = "render ring";
-	engine->id = RCS;
-	engine->exec_id = I915_EXEC_RENDER;
-	engine->hw_id = 0;
-	engine->mmio_base = RENDER_RING_BASE;
-
 	intel_ring_default_vfuncs(dev_priv, engine);
 
-	engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
 	if (HAS_L3_DPF(dev_priv))
 		engine->irq_keep_mask = GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
 
 	if (INTEL_GEN(dev_priv) >= 8) {
 		engine->init_context = intel_rcs_ctx_init;
-		engine->add_request = gen8_render_add_request;
-		engine->flush = gen8_render_ring_flush;
-		if (i915_semaphore_is_enabled(dev_priv))
+		engine->emit_request = gen8_render_emit_request;
+		engine->emit_flush = gen8_render_ring_flush;
+		if (i915.semaphores)
 			engine->semaphore.signal = gen8_rcs_signal;
 	} else if (INTEL_GEN(dev_priv) >= 6) {
 		engine->init_context = intel_rcs_ctx_init;
-		engine->flush = gen7_render_ring_flush;
+		engine->emit_flush = gen7_render_ring_flush;
 		if (IS_GEN6(dev_priv))
-			engine->flush = gen6_render_ring_flush;
+			engine->emit_flush = gen6_render_ring_flush;
 	} else if (IS_GEN5(dev_priv)) {
-		engine->flush = gen4_render_ring_flush;
+		engine->emit_flush = gen4_render_ring_flush;
 	} else {
 		if (INTEL_GEN(dev_priv) < 4)
-			engine->flush = gen2_render_ring_flush;
+			engine->emit_flush = gen2_render_ring_flush;
 		else
-			engine->flush = gen4_render_ring_flush;
+			engine->emit_flush = gen4_render_ring_flush;
 		engine->irq_enable_mask = I915_USER_INTERRUPT;
 	}
 
 	if (IS_HASWELL(dev_priv))
-		engine->dispatch_execbuffer = hsw_ring_dispatch_execbuffer;
+		engine->emit_bb_start = hsw_emit_bb_start;
 
 	engine->init_hw = init_render_ring;
 	engine->cleanup = render_ring_cleanup;
 
-	ret = intel_init_ring_buffer(dev, engine);
+	ret = intel_init_ring_buffer(engine);
 	if (ret)
 		return ret;
 
@@ -2912,166 +2820,71 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 	return 0;
 }
 
-int intel_init_bsd_ring_buffer(struct drm_device *dev)
+int intel_init_bsd_ring_buffer(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_engine_cs *engine = &dev_priv->engine[VCS];
-
-	engine->name = "bsd ring";
-	engine->id = VCS;
-	engine->exec_id = I915_EXEC_BSD;
-	engine->hw_id = 1;
+	struct drm_i915_private *dev_priv = engine->i915;
 
 	intel_ring_default_vfuncs(dev_priv, engine);
 
 	if (INTEL_GEN(dev_priv) >= 6) {
-		engine->mmio_base = GEN6_BSD_RING_BASE;
 		/* gen6 bsd needs a special wa for tail updates */
 		if (IS_GEN6(dev_priv))
-			engine->write_tail = gen6_bsd_ring_write_tail;
-		engine->flush = gen6_bsd_ring_flush;
-		if (INTEL_GEN(dev_priv) >= 8)
-			engine->irq_enable_mask =
-				GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT;
-		else
+			engine->submit_request = gen6_bsd_submit_request;
+		engine->emit_flush = gen6_bsd_ring_flush;
+		if (INTEL_GEN(dev_priv) < 8)
 			engine->irq_enable_mask = GT_BSD_USER_INTERRUPT;
 	} else {
 		engine->mmio_base = BSD_RING_BASE;
-		engine->flush = bsd_ring_flush;
+		engine->emit_flush = bsd_ring_flush;
 		if (IS_GEN5(dev_priv))
 			engine->irq_enable_mask = ILK_BSD_USER_INTERRUPT;
 		else
 			engine->irq_enable_mask = I915_BSD_USER_INTERRUPT;
 	}
 
-	return intel_init_ring_buffer(dev, engine);
+	return intel_init_ring_buffer(engine);
 }
 
 /**
  * Initialize the second BSD ring (eg. Broadwell GT3, Skylake GT3)
  */
-int intel_init_bsd2_ring_buffer(struct drm_device *dev)
+int intel_init_bsd2_ring_buffer(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_engine_cs *engine = &dev_priv->engine[VCS2];
-
-	engine->name = "bsd2 ring";
-	engine->id = VCS2;
-	engine->exec_id = I915_EXEC_BSD;
-	engine->hw_id = 4;
-	engine->mmio_base = GEN8_BSD2_RING_BASE;
+	struct drm_i915_private *dev_priv = engine->i915;
 
 	intel_ring_default_vfuncs(dev_priv, engine);
 
-	engine->flush = gen6_bsd_ring_flush;
-	engine->irq_enable_mask =
-			GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT;
+	engine->emit_flush = gen6_bsd_ring_flush;
 
-	return intel_init_ring_buffer(dev, engine);
+	return intel_init_ring_buffer(engine);
 }
 
-int intel_init_blt_ring_buffer(struct drm_device *dev)
+int intel_init_blt_ring_buffer(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_engine_cs *engine = &dev_priv->engine[BCS];
-
-	engine->name = "blitter ring";
-	engine->id = BCS;
-	engine->exec_id = I915_EXEC_BLT;
-	engine->hw_id = 2;
-	engine->mmio_base = BLT_RING_BASE;
+	struct drm_i915_private *dev_priv = engine->i915;
 
 	intel_ring_default_vfuncs(dev_priv, engine);
 
-	engine->flush = gen6_ring_flush;
-	if (INTEL_GEN(dev_priv) >= 8)
-		engine->irq_enable_mask =
-			GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT;
-	else
+	engine->emit_flush = gen6_ring_flush;
+	if (INTEL_GEN(dev_priv) < 8)
 		engine->irq_enable_mask = GT_BLT_USER_INTERRUPT;
 
-	return intel_init_ring_buffer(dev, engine);
+	return intel_init_ring_buffer(engine);
 }
 
-int intel_init_vebox_ring_buffer(struct drm_device *dev)
+int intel_init_vebox_ring_buffer(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_engine_cs *engine = &dev_priv->engine[VECS];
-
-	engine->name = "video enhancement ring";
-	engine->id = VECS;
-	engine->exec_id = I915_EXEC_VEBOX;
-	engine->hw_id = 3;
-	engine->mmio_base = VEBOX_RING_BASE;
+	struct drm_i915_private *dev_priv = engine->i915;
 
 	intel_ring_default_vfuncs(dev_priv, engine);
 
-	engine->flush = gen6_ring_flush;
+	engine->emit_flush = gen6_ring_flush;
 
-	if (INTEL_GEN(dev_priv) >= 8) {
-		engine->irq_enable_mask =
-			GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT;
-	} else {
+	if (INTEL_GEN(dev_priv) < 8) {
 		engine->irq_enable_mask = PM_VEBOX_USER_INTERRUPT;
 		engine->irq_enable = hsw_vebox_irq_enable;
 		engine->irq_disable = hsw_vebox_irq_disable;
 	}
 
-	return intel_init_ring_buffer(dev, engine);
-}
-
-int
-intel_ring_flush_all_caches(struct drm_i915_gem_request *req)
-{
-	struct intel_engine_cs *engine = req->engine;
-	int ret;
-
-	if (!engine->gpu_caches_dirty)
-		return 0;
-
-	ret = engine->flush(req, 0, I915_GEM_GPU_DOMAINS);
-	if (ret)
-		return ret;
-
-	trace_i915_gem_ring_flush(req, 0, I915_GEM_GPU_DOMAINS);
-
-	engine->gpu_caches_dirty = false;
-	return 0;
-}
-
-int
-intel_ring_invalidate_all_caches(struct drm_i915_gem_request *req)
-{
-	struct intel_engine_cs *engine = req->engine;
-	uint32_t flush_domains;
-	int ret;
-
-	flush_domains = 0;
-	if (engine->gpu_caches_dirty)
-		flush_domains = I915_GEM_GPU_DOMAINS;
-
-	ret = engine->flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
-	if (ret)
-		return ret;
-
-	trace_i915_gem_ring_flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
-
-	engine->gpu_caches_dirty = false;
-	return 0;
-}
-
-void
-intel_stop_engine(struct intel_engine_cs *engine)
-{
-	int ret;
-
-	if (!intel_engine_initialized(engine))
-		return;
-
-	ret = intel_engine_idle(engine);
-	if (ret)
-		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
-			  engine->name, ret);
-
-	stop_ring(engine);
+	return intel_init_ring_buffer(engine);
 }

drivers/gpu/drm/i915/intel_ringbuffer.h

@@ -3,6 +3,7 @@
 
 #include <linux/hashtable.h>
 #include "i915_gem_batch_pool.h"
+#include "i915_gem_request.h"
 
 #define I915_CMD_HASH_ORDER 9
 
@@ -31,23 +32,23 @@ struct  intel_hw_status_page {
 	struct		drm_i915_gem_object *obj;
 };
 
-#define I915_READ_TAIL(ring) I915_READ(RING_TAIL((ring)->mmio_base))
-#define I915_WRITE_TAIL(ring, val) I915_WRITE(RING_TAIL((ring)->mmio_base), val)
+#define I915_READ_TAIL(engine) I915_READ(RING_TAIL((engine)->mmio_base))
+#define I915_WRITE_TAIL(engine, val) I915_WRITE(RING_TAIL((engine)->mmio_base), val)
 
-#define I915_READ_START(ring) I915_READ(RING_START((ring)->mmio_base))
-#define I915_WRITE_START(ring, val) I915_WRITE(RING_START((ring)->mmio_base), val)
+#define I915_READ_START(engine) I915_READ(RING_START((engine)->mmio_base))
+#define I915_WRITE_START(engine, val) I915_WRITE(RING_START((engine)->mmio_base), val)
 
-#define I915_READ_HEAD(ring)  I915_READ(RING_HEAD((ring)->mmio_base))
-#define I915_WRITE_HEAD(ring, val) I915_WRITE(RING_HEAD((ring)->mmio_base), val)
+#define I915_READ_HEAD(engine)  I915_READ(RING_HEAD((engine)->mmio_base))
+#define I915_WRITE_HEAD(engine, val) I915_WRITE(RING_HEAD((engine)->mmio_base), val)
 
-#define I915_READ_CTL(ring) I915_READ(RING_CTL((ring)->mmio_base))
-#define I915_WRITE_CTL(ring, val) I915_WRITE(RING_CTL((ring)->mmio_base), val)
+#define I915_READ_CTL(engine) I915_READ(RING_CTL((engine)->mmio_base))
+#define I915_WRITE_CTL(engine, val) I915_WRITE(RING_CTL((engine)->mmio_base), val)
 
-#define I915_READ_IMR(ring) I915_READ(RING_IMR((ring)->mmio_base))
-#define I915_WRITE_IMR(ring, val) I915_WRITE(RING_IMR((ring)->mmio_base), val)
+#define I915_READ_IMR(engine) I915_READ(RING_IMR((engine)->mmio_base))
+#define I915_WRITE_IMR(engine, val) I915_WRITE(RING_IMR((engine)->mmio_base), val)
 
-#define I915_READ_MODE(ring) I915_READ(RING_MI_MODE((ring)->mmio_base))
-#define I915_WRITE_MODE(ring, val) I915_WRITE(RING_MI_MODE((ring)->mmio_base), val)
+#define I915_READ_MODE(engine) I915_READ(RING_MI_MODE((engine)->mmio_base))
+#define I915_WRITE_MODE(engine, val) I915_WRITE(RING_MI_MODE((engine)->mmio_base), val)
 
 /* seqno size is actually only a uint32, but since we plan to use MI_FLUSH_DW to
  * do the writes, and that must have qw aligned offsets, simply pretend it's 8b.
@@ -62,7 +63,7 @@ struct  intel_hw_status_page {
 	(i915_gem_obj_ggtt_offset(dev_priv->semaphore_obj) + \
 	 GEN8_SEMAPHORE_OFFSET(from, (__ring)->id))
 
-enum intel_ring_hangcheck_action {
+enum intel_engine_hangcheck_action {
 	HANGCHECK_IDLE = 0,
 	HANGCHECK_WAIT,
 	HANGCHECK_ACTIVE,
@@ -72,24 +73,26 @@ enum intel_ring_hangcheck_action {
 
 #define HANGCHECK_SCORE_RING_HUNG 31
 
-struct intel_ring_hangcheck {
+struct intel_engine_hangcheck {
 	u64 acthd;
 	unsigned long user_interrupts;
 	u32 seqno;
 	int score;
-	enum intel_ring_hangcheck_action action;
+	enum intel_engine_hangcheck_action action;
 	int deadlock;
 	u32 instdone[I915_NUM_INSTDONE_REG];
 };
 
-struct intel_ringbuffer {
+struct intel_ring {
 	struct drm_i915_gem_object *obj;
-	void __iomem *virtual_start;
+	void *vaddr;
 	struct i915_vma *vma;
 
 	struct intel_engine_cs *engine;
 	struct list_head link;
 
+	struct list_head request_list;
+
 	u32 head;
 	u32 tail;
 	int space;
@@ -146,8 +149,10 @@ struct intel_engine_cs {
 	unsigned int exec_id;
 	unsigned int hw_id;
 	unsigned int guc_id; /* XXX same as hw_id? */
+	u64 fence_context;
 	u32		mmio_base;
-	struct intel_ringbuffer *buffer;
+	unsigned int irq_shift;
+	struct intel_ring *buffer;
 	struct list_head buffers;
 
 	/* Rather than have every client wait upon all user interrupts,
@@ -195,33 +200,34 @@ struct intel_engine_cs {
 
 	u32             irq_keep_mask; /* always keep these interrupts */
 	u32		irq_enable_mask; /* bitmask to enable ring interrupt */
-	void		(*irq_enable)(struct intel_engine_cs *ring);
-	void		(*irq_disable)(struct intel_engine_cs *ring);
+	void		(*irq_enable)(struct intel_engine_cs *engine);
+	void		(*irq_disable)(struct intel_engine_cs *engine);
 
-	int		(*init_hw)(struct intel_engine_cs *ring);
+	int		(*init_hw)(struct intel_engine_cs *engine);
 
 	int		(*init_context)(struct drm_i915_gem_request *req);
 
-	void		(*write_tail)(struct intel_engine_cs *ring,
-				      u32 value);
-	int __must_check (*flush)(struct drm_i915_gem_request *req,
-				  u32	invalidate_domains,
-				  u32	flush_domains);
-	int		(*add_request)(struct drm_i915_gem_request *req);
+	int		(*emit_flush)(struct drm_i915_gem_request *request,
+				      u32 mode);
+#define EMIT_INVALIDATE	BIT(0)
+#define EMIT_FLUSH	BIT(1)
+#define EMIT_BARRIER	(EMIT_INVALIDATE | EMIT_FLUSH)
+	int		(*emit_bb_start)(struct drm_i915_gem_request *req,
+					 u64 offset, u32 length,
+					 unsigned int dispatch_flags);
+#define I915_DISPATCH_SECURE BIT(0)
+#define I915_DISPATCH_PINNED BIT(1)
+#define I915_DISPATCH_RS     BIT(2)
+	int		(*emit_request)(struct drm_i915_gem_request *req);
+	void		(*submit_request)(struct drm_i915_gem_request *req);
 	/* Some chipsets are not quite as coherent as advertised and need
 	 * an expensive kick to force a true read of the up-to-date seqno.
 	 * However, the up-to-date seqno is not always required and the last
 	 * seen value is good enough. Note that the seqno will always be
 	 * monotonic, even if not coherent.
 	 */
-	void		(*irq_seqno_barrier)(struct intel_engine_cs *ring);
-	int		(*dispatch_execbuffer)(struct drm_i915_gem_request *req,
-					       u64 offset, u32 length,
-					       unsigned dispatch_flags);
-#define I915_DISPATCH_SECURE 0x1
-#define I915_DISPATCH_PINNED 0x2
-#define I915_DISPATCH_RS     0x4
-	void		(*cleanup)(struct intel_engine_cs *ring);
+	void		(*irq_seqno_barrier)(struct intel_engine_cs *engine);
+	void		(*cleanup)(struct intel_engine_cs *engine);
 
 	/* GEN8 signal/wait table - never trust comments!
 	 *	  signal to	signal to    signal to   signal to      signal to
@@ -274,12 +280,9 @@ struct intel_engine_cs {
 		};
 
 		/* AKA wait() */
-		int	(*sync_to)(struct drm_i915_gem_request *to_req,
-				   struct intel_engine_cs *from,
-				   u32 seqno);
-		int	(*signal)(struct drm_i915_gem_request *signaller_req,
-				  /* num_dwords needed by caller */
-				  unsigned int num_dwords);
+		int	(*sync_to)(struct drm_i915_gem_request *req,
+				   struct drm_i915_gem_request *signal);
+		int	(*signal)(struct drm_i915_gem_request *req);
 	} semaphore;
 
 	/* Execlists */
@@ -291,24 +294,6 @@ struct intel_engine_cs {
 	unsigned int idle_lite_restore_wa;
 	bool disable_lite_restore_wa;
 	u32 ctx_desc_template;
-	int		(*emit_request)(struct drm_i915_gem_request *request);
-	int		(*emit_flush)(struct drm_i915_gem_request *request,
-				      u32 invalidate_domains,
-				      u32 flush_domains);
-	int		(*emit_bb_start)(struct drm_i915_gem_request *req,
-					 u64 offset, unsigned dispatch_flags);
-
-	/**
-	 * List of objects currently involved in rendering from the
-	 * ringbuffer.
-	 *
-	 * Includes buffers having the contents of their GPU caches
-	 * flushed, not necessarily primitives.  last_read_req
-	 * represents when the rendering involved will be completed.
-	 *
-	 * A reference is held on the buffer while on this list.
-	 */
-	struct list_head active_list;
 
 	/**
 	 * List of breadcrumbs associated with GPU requests currently
@@ -323,11 +308,16 @@ struct intel_engine_cs {
 	 */
 	u32 last_submitted_seqno;
 
-	bool gpu_caches_dirty;
+	/* An RCU guarded pointer to the last request. No reference is
+	 * held to the request, users must carefully acquire a reference to
+	 * the request using i915_gem_active_get_request_rcu(), or hold the
+	 * struct_mutex.
+	 */
+	struct i915_gem_active last_request;
 
 	struct i915_gem_context *last_context;
 
-	struct intel_ring_hangcheck hangcheck;
+	struct intel_engine_hangcheck hangcheck;
 
 	struct {
 		struct drm_i915_gem_object *obj;
@@ -338,7 +328,7 @@ struct intel_engine_cs {
 
 	/*
 	 * Table of commands the command parser needs to know about
-	 * for this ring.
+	 * for this engine.
 	 */
 	DECLARE_HASHTABLE(cmd_hash, I915_CMD_HASH_ORDER);
 
@@ -352,11 +342,11 @@ struct intel_engine_cs {
 	 * Returns the bitmask for the length field of the specified command.
 	 * Return 0 for an unrecognized/invalid command.
 	 *
-	 * If the command parser finds an entry for a command in the ring's
+	 * If the command parser finds an entry for a command in the engine's
 	 * cmd_tables, it gets the command's length based on the table entry.
-	 * If not, it calls this function to determine the per-ring length field
-	 * encoding for the command (i.e. certain opcode ranges use certain bits
-	 * to encode the command length in the header).
+	 * If not, it calls this function to determine the per-engine length
+	 * field encoding for the command (i.e. different opcode ranges use
+	 * certain bits to encode the command length in the header).
 	 */
 	u32 (*get_cmd_length_mask)(u32 cmd_header);
 };
@@ -374,8 +364,8 @@ intel_engine_flag(const struct intel_engine_cs *engine)
 }
 
 static inline u32
-intel_ring_sync_index(struct intel_engine_cs *engine,
-		      struct intel_engine_cs *other)
+intel_engine_sync_index(struct intel_engine_cs *engine,
+			struct intel_engine_cs *other)
 {
 	int idx;
 
@@ -437,67 +427,83 @@ intel_write_status_page(struct intel_engine_cs *engine,
 #define I915_GEM_HWS_SCRATCH_INDEX	0x40
 #define I915_GEM_HWS_SCRATCH_ADDR (I915_GEM_HWS_SCRATCH_INDEX << MI_STORE_DWORD_INDEX_SHIFT)
 
-struct intel_ringbuffer *
-intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size);
-int intel_pin_and_map_ringbuffer_obj(struct drm_i915_private *dev_priv,
-				     struct intel_ringbuffer *ringbuf);
-void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf);
-void intel_ringbuffer_free(struct intel_ringbuffer *ring);
+struct intel_ring *
+intel_engine_create_ring(struct intel_engine_cs *engine, int size);
+int intel_ring_pin(struct intel_ring *ring);
+void intel_ring_unpin(struct intel_ring *ring);
+void intel_ring_free(struct intel_ring *ring);
 
-void intel_stop_engine(struct intel_engine_cs *engine);
-void intel_cleanup_engine(struct intel_engine_cs *engine);
+void intel_engine_stop(struct intel_engine_cs *engine);
+void intel_engine_cleanup(struct intel_engine_cs *engine);
 
 int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request);
 
 int __must_check intel_ring_begin(struct drm_i915_gem_request *req, int n);
 int __must_check intel_ring_cacheline_align(struct drm_i915_gem_request *req);
-static inline void intel_ring_emit(struct intel_engine_cs *engine,
-				   u32 data)
+
+static inline void intel_ring_emit(struct intel_ring *ring, u32 data)
 {
-	struct intel_ringbuffer *ringbuf = engine->buffer;
-	iowrite32(data, ringbuf->virtual_start + ringbuf->tail);
-	ringbuf->tail += 4;
+	*(uint32_t *)(ring->vaddr + ring->tail) = data;
+	ring->tail += 4;
 }
-static inline void intel_ring_emit_reg(struct intel_engine_cs *engine,
-				       i915_reg_t reg)
+
+static inline void intel_ring_emit_reg(struct intel_ring *ring, i915_reg_t reg)
 {
-	intel_ring_emit(engine, i915_mmio_reg_offset(reg));
+	intel_ring_emit(ring, i915_mmio_reg_offset(reg));
+}
+
+static inline void intel_ring_advance(struct intel_ring *ring)
+{
+	/* Dummy function.
+	 *
+	 * This serves as a placeholder in the code so that the reader
+	 * can compare against the preceding intel_ring_begin() and
+	 * check that the number of dwords emitted matches the space
+	 * reserved for the command packet (i.e. the value passed to
+	 * intel_ring_begin()).
+	 */
 }
-static inline void intel_ring_advance(struct intel_engine_cs *engine)
+
+static inline u32 intel_ring_offset(struct intel_ring *ring, u32 value)
 {
-	struct intel_ringbuffer *ringbuf = engine->buffer;
-	ringbuf->tail &= ringbuf->size - 1;
+	/* Don't write ring->size (equivalent to 0) as that hangs some GPUs. */
+	return value & (ring->size - 1);
 }
+
 int __intel_ring_space(int head, int tail, int size);
-void intel_ring_update_space(struct intel_ringbuffer *ringbuf);
+void intel_ring_update_space(struct intel_ring *ring);
 
-int __must_check intel_engine_idle(struct intel_engine_cs *engine);
-void intel_ring_init_seqno(struct intel_engine_cs *engine, u32 seqno);
-int intel_ring_flush_all_caches(struct drm_i915_gem_request *req);
-int intel_ring_invalidate_all_caches(struct drm_i915_gem_request *req);
+void intel_engine_init_seqno(struct intel_engine_cs *engine, u32 seqno);
 
 int intel_init_pipe_control(struct intel_engine_cs *engine, int size);
 void intel_fini_pipe_control(struct intel_engine_cs *engine);
 
-int intel_init_render_ring_buffer(struct drm_device *dev);
-int intel_init_bsd_ring_buffer(struct drm_device *dev);
-int intel_init_bsd2_ring_buffer(struct drm_device *dev);
-int intel_init_blt_ring_buffer(struct drm_device *dev);
-int intel_init_vebox_ring_buffer(struct drm_device *dev);
+void intel_engine_setup_common(struct intel_engine_cs *engine);
+int intel_engine_init_common(struct intel_engine_cs *engine);
+void intel_engine_cleanup_common(struct intel_engine_cs *engine);
 
-u64 intel_ring_get_active_head(struct intel_engine_cs *engine);
-static inline u32 intel_engine_get_seqno(struct intel_engine_cs *engine)
+static inline int intel_engine_idle(struct intel_engine_cs *engine,
+				    bool interruptible)
 {
-	return intel_read_status_page(engine, I915_GEM_HWS_INDEX);
+	/* Wait upon the last request to be completed */
+	return i915_gem_active_wait_unlocked(&engine->last_request,
+					     interruptible, NULL, NULL);
 }
 
-int init_workarounds_ring(struct intel_engine_cs *engine);
+int intel_init_render_ring_buffer(struct intel_engine_cs *engine);
+int intel_init_bsd_ring_buffer(struct intel_engine_cs *engine);
+int intel_init_bsd2_ring_buffer(struct intel_engine_cs *engine);
+int intel_init_blt_ring_buffer(struct intel_engine_cs *engine);
+int intel_init_vebox_ring_buffer(struct intel_engine_cs *engine);
 
-static inline u32 intel_ring_get_tail(struct intel_ringbuffer *ringbuf)
+u64 intel_engine_get_active_head(struct intel_engine_cs *engine);
+static inline u32 intel_engine_get_seqno(struct intel_engine_cs *engine)
 {
-	return ringbuf->tail;
+	return intel_read_status_page(engine, I915_GEM_HWS_INDEX);
 }
 
+int init_workarounds_ring(struct intel_engine_cs *engine);
+
 /*
  * Arbitrary size for largest possible 'add request' sequence. The code paths
  * are complex and variable. Empirical measurement shows that the worst case
@@ -513,17 +519,6 @@ static inline u32 intel_hws_seqno_address(struct intel_engine_cs *engine)
 }
 
 /* intel_breadcrumbs.c -- user interrupt bottom-half for waiters */
-struct intel_wait {
-	struct rb_node node;
-	struct task_struct *tsk;
-	u32 seqno;
-};
-
-struct intel_signal_node {
-	struct rb_node node;
-	struct intel_wait wait;
-};
-
 int intel_engine_init_breadcrumbs(struct intel_engine_cs *engine);
 
 static inline void intel_wait_init(struct intel_wait *wait, u32 seqno)
@@ -570,4 +565,9 @@ void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine);
 unsigned int intel_kick_waiters(struct drm_i915_private *i915);
 unsigned int intel_kick_signalers(struct drm_i915_private *i915);
 
+static inline bool intel_engine_is_active(struct intel_engine_cs *engine)
+{
+	return i915_gem_active_isset(&engine->last_request);
+}
+
 #endif /* _INTEL_RINGBUFFER_H_ */

drivers/gpu/drm/i915/intel_sprite.c

@@ -36,6 +36,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_plane_helper.h>
 #include "intel_drv.h"
+#include "intel_frontbuffer.h"
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
@@ -430,7 +431,7 @@ vlv_update_plane(struct drm_plane *dplane,
 	 */
 	sprctl |= SP_GAMMA_ENABLE;
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		sprctl |= SP_TILED;
 
 	/* Sizes are 0 based */
@@ -467,7 +468,7 @@ vlv_update_plane(struct drm_plane *dplane,
 	I915_WRITE(SPSTRIDE(pipe, plane), fb->pitches[0]);
 	I915_WRITE(SPPOS(pipe, plane), (crtc_y << 16) | crtc_x);
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		I915_WRITE(SPTILEOFF(pipe, plane), (y << 16) | x);
 	else
 		I915_WRITE(SPLINOFF(pipe, plane), linear_offset);
@@ -552,7 +553,7 @@ ivb_update_plane(struct drm_plane *plane,
 	 */
 	sprctl |= SPRITE_GAMMA_ENABLE;
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		sprctl |= SPRITE_TILED;
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
@@ -606,7 +607,7 @@ ivb_update_plane(struct drm_plane *plane,
 	 * register */
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		I915_WRITE(SPROFFSET(pipe), (y << 16) | x);
-	else if (obj->tiling_mode != I915_TILING_NONE)
+	else if (i915_gem_object_is_tiled(obj))
 		I915_WRITE(SPRTILEOFF(pipe), (y << 16) | x);
 	else
 		I915_WRITE(SPRLINOFF(pipe), linear_offset);
@@ -693,7 +694,7 @@ ilk_update_plane(struct drm_plane *plane,
 	 */
 	dvscntr |= DVS_GAMMA_ENABLE;
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		dvscntr |= DVS_TILED;
 
 	if (IS_GEN6(dev))
@@ -736,7 +737,7 @@ ilk_update_plane(struct drm_plane *plane,
 	I915_WRITE(DVSSTRIDE(pipe), fb->pitches[0]);
 	I915_WRITE(DVSPOS(pipe), (crtc_y << 16) | crtc_x);
 
-	if (obj->tiling_mode != I915_TILING_NONE)
+	if (i915_gem_object_is_tiled(obj))
 		I915_WRITE(DVSTILEOFF(pipe), (y << 16) | x);
 	else
 		I915_WRITE(DVSLINOFF(pipe), linear_offset);

drivers/gpu/drm/i915/intel_uncore.c

@@ -435,7 +435,7 @@ void intel_uncore_sanitize(struct drm_i915_private *dev_priv)
 	i915.enable_rc6 = sanitize_rc6_option(dev_priv, i915.enable_rc6);
 
 	/* BIOS often leaves RC6 enabled, but disable it for hw init */
-	intel_disable_gt_powersave(dev_priv);
+	intel_sanitize_gt_powersave(dev_priv);
 }
 
 static void __intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
@@ -1618,8 +1618,10 @@ static int gen6_reset_engines(struct drm_i915_private *dev_priv,
  * @timeout_ms: timeout in millisecond
  *
  * This routine waits until the target register @reg contains the expected
- * @value after applying the @mask, i.e. it waits until
- *   (I915_READ_FW(@reg) & @mask) == @value
+ * @value after applying the @mask, i.e. it waits until ::
+ *
+ *     (I915_READ_FW(reg) & mask) == value
+ *
  * Otherwise, the wait will timeout after @timeout_ms milliseconds.
  *
  * Note that this routine assumes the caller holds forcewake asserted, it is
@@ -1652,8 +1654,10 @@ int intel_wait_for_register_fw(struct drm_i915_private *dev_priv,
  * @timeout_ms: timeout in millisecond
  *
  * This routine waits until the target register @reg contains the expected
- * @value after applying the @mask, i.e. it waits until
- *   (I915_READ(@reg) & @mask) == @value
+ * @value after applying the @mask, i.e. it waits until ::
+ *
+ *     (I915_READ(reg) & mask) == value
+ *
  * Otherwise, the wait will timeout after @timeout_ms milliseconds.
  *
  * Returns 0 if the register matches the desired condition, or -ETIMEOUT.

include/uapi/drm/i915_drm.h

@@ -62,6 +62,30 @@ extern "C" {
 #define I915_ERROR_UEVENT		"ERROR"
 #define I915_RESET_UEVENT		"RESET"
 
+/*
+ * MOCS indexes used for GPU surfaces, defining the cacheability of the
+ * surface data and the coherency for this data wrt. CPU vs. GPU accesses.
+ */
+enum i915_mocs_table_index {
+	/*
+	 * Not cached anywhere, coherency between CPU and GPU accesses is
+	 * guaranteed.
+	 */
+	I915_MOCS_UNCACHED,
+	/*
+	 * Cacheability and coherency controlled by the kernel automatically
+	 * based on the DRM_I915_GEM_SET_CACHING IOCTL setting and the current
+	 * usage of the surface (used for display scanout or not).
+	 */
+	I915_MOCS_PTE,
+	/*
+	 * Cached in all GPU caches available on the platform.
+	 * Coherency between CPU and GPU accesses to the surface is not
+	 * guaranteed without extra synchronization.
+	 */
+	I915_MOCS_CACHED,
+};
+
 /* Each region is a minimum of 16k, and there are at most 255 of them.
  */
 #define I915_NR_TEX_REGIONS 255	/* table size 2k - maximum due to use
@@ -698,15 +722,20 @@ struct drm_i915_gem_exec_object2 {
 	 */
 	__u64 offset;
 
-#define EXEC_OBJECT_NEEDS_FENCE (1<<0)
-#define EXEC_OBJECT_NEEDS_GTT	(1<<1)
-#define EXEC_OBJECT_WRITE	(1<<2)
+#define EXEC_OBJECT_NEEDS_FENCE		 (1<<0)
+#define EXEC_OBJECT_NEEDS_GTT		 (1<<1)
+#define EXEC_OBJECT_WRITE		 (1<<2)
 #define EXEC_OBJECT_SUPPORTS_48B_ADDRESS (1<<3)
-#define EXEC_OBJECT_PINNED	(1<<4)
-#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_PINNED<<1)
+#define EXEC_OBJECT_PINNED		 (1<<4)
+#define EXEC_OBJECT_PAD_TO_SIZE		 (1<<5)
+/* All remaining bits are MBZ and RESERVED FOR FUTURE USE */
+#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_PAD_TO_SIZE<<1)
 	__u64 flags;
 
-	__u64 rsvd1;
+	union {
+		__u64 rsvd1;
+		__u64 pad_to_size;
+	};
 	__u64 rsvd2;
 };
 
@@ -897,6 +926,7 @@ struct drm_i915_gem_caching {
 #define I915_TILING_NONE	0
 #define I915_TILING_X		1
 #define I915_TILING_Y		2
+#define I915_TILING_LAST	I915_TILING_Y
 
 #define I915_BIT_6_SWIZZLE_NONE		0
 #define I915_BIT_6_SWIZZLE_9		1