[PATCH 00/27] Parallel submission aka multi-bb execbuf

@@ -290,7 +290,8 @@ static void nop_irq_handler(struct intel_engine_cs *engine, u16 iir)
 	GEM_DEBUG_WARN_ON(iir);
 }
 
-static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id)
+static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id,
+			      u8 logical_instance)
 {
 	const struct engine_info *info = &intel_engines[id];
 	struct drm_i915_private *i915 = gt->i915;

@@ -334,6 +335,7 @@ static int intel_engine_setup(struct intel_gt *gt, enum intel_engine_id id)
 
 	engine->class = info->class;
 	engine->instance = info->instance;
+	engine->logical_mask = BIT(logical_instance);
 	__sprint_engine_name(engine);
 
 	engine->props.heartbeat_interval_ms =

@@ -572,6 +574,37 @@ static intel_engine_mask_t init_engine_mask(struct intel_gt *gt)
 	return info->engine_mask;
 }
 
+static void populate_logical_ids(struct intel_gt *gt, u8 *logical_ids,
+				 u8 class, const u8 *map, u8 num_instances)
+{
+	int i, j;
+	u8 current_logical_id = 0;
+
+	for (j = 0; j < num_instances; ++j) {
+		for (i = 0; i < ARRAY_SIZE(intel_engines); ++i) {
+			if (!HAS_ENGINE(gt, i) ||
+			    intel_engines[i].class != class)
+				continue;
+
+			if (intel_engines[i].instance == map[j]) {
+				logical_ids[intel_engines[i].instance] =
+					current_logical_id++;
+				break;
+			}
+		}
+	}
+}
+
+static void setup_logical_ids(struct intel_gt *gt, u8 *logical_ids, u8 class)
+{
+	int i;
+	u8 map[MAX_ENGINE_INSTANCE + 1];
+
+	for (i = 0; i < MAX_ENGINE_INSTANCE + 1; ++i)
+		map[i] = i;
+	populate_logical_ids(gt, logical_ids, class, map, ARRAY_SIZE(map));
+}
+
 /**
  * intel_engines_init_mmio() - allocate and prepare the Engine Command Streamers
  * @gt: pointer to struct intel_gt

@@ -583,7 +616,8 @@ int intel_engines_init_mmio(struct intel_gt *gt)
 	struct drm_i915_private *i915 = gt->i915;
 	const unsigned int engine_mask = init_engine_mask(gt);
 	unsigned int mask = 0;
-	unsigned int i;
+	unsigned int i, class;
+	u8 logical_ids[MAX_ENGINE_INSTANCE + 1];
 	int err;
 
 	drm_WARN_ON(&i915->drm, engine_mask == 0);

@@ -593,15 +627,23 @@ int intel_engines_init_mmio(struct intel_gt *gt)
 	if (i915_inject_probe_failure(i915))
 		return -ENODEV;
 
-	for (i = 0; i < ARRAY_SIZE(intel_engines); i++) {
-		if (!HAS_ENGINE(gt, i))
-			continue;
+	for (class = 0; class < MAX_ENGINE_CLASS + 1; ++class) {
+		setup_logical_ids(gt, logical_ids, class);
 
-		err = intel_engine_setup(gt, i);
-		if (err)
-			goto cleanup;
+		for (i = 0; i < ARRAY_SIZE(intel_engines); ++i) {
+			u8 instance = intel_engines[i].instance;
+
+			if (intel_engines[i].class != class ||
+			    !HAS_ENGINE(gt, i))
+				continue;
 
-		mask |= BIT(i);
+			err = intel_engine_setup(gt, i,
+						 logical_ids[instance]);
+			if (err)
+				goto cleanup;
+
+			mask |= BIT(i);
+		}
 	}
 
 	/*

@@ -266,6 +266,11 @@ struct intel_engine_cs {
 	unsigned int guc_id;
 
 	intel_engine_mask_t mask;
+	/**
+	 * @logical_mask: logical mask of engine, reported to user space via
+	 * query IOCTL and used to communicate with the GuC in logical space
+	 */
+	intel_engine_mask_t logical_mask;
 
 	u8 class;
 	u8 instance;

@@ -3875,6 +3875,7 @@ execlists_create_virtual(struct intel_engine_cs **siblings, unsigned int count)
 
 		ve->siblings[ve->num_siblings++] = sibling;
 		ve->base.mask |= sibling->mask;
+		ve->base.logical_mask |= sibling->logical_mask;
 
 		/*
 		 * All physical engines must be compatible for their emission

@@ -176,7 +176,7 @@ static void guc_mapping_table_init(struct intel_gt *gt,
 	for_each_engine(engine, gt, id) {
 		u8 guc_class = engine_class_to_guc_class(engine->class);
 
-		system_info->mapping_table[guc_class][engine->instance] =
+		system_info->mapping_table[guc_class][ilog2(engine->logical_mask)] =
 			engine->instance;
 	}
 }

@@ -1401,23 +1401,6 @@ static int deregister_context(struct intel_context *ce, u32 guc_id, bool loop)
 	return __guc_action_deregister_context(guc, guc_id, loop);
 }
 
-static intel_engine_mask_t adjust_engine_mask(u8 class, intel_engine_mask_t mask)
-{
-	switch (class) {
-	case RENDER_CLASS:
-		return mask >> RCS0;
-	case VIDEO_ENHANCEMENT_CLASS:
-		return mask >> VECS0;
-	case VIDEO_DECODE_CLASS:
-		return mask >> VCS0;
-	case COPY_ENGINE_CLASS:
-		return mask >> BCS0;
-	default:
-		MISSING_CASE(class);
-		return 0;
-	}
-}
-
 static void guc_context_policy_init(struct intel_engine_cs *engine,
 				    struct guc_lrc_desc *desc)
 {

@@ -1459,8 +1442,7 @@ static int guc_lrc_desc_pin(struct intel_context *ce, bool loop)
 
 	desc = __get_lrc_desc(guc, desc_idx);
 	desc->engine_class = engine_class_to_guc_class(engine->class);
-	desc->engine_submit_mask = adjust_engine_mask(engine->class,
-						      engine->mask);
+	desc->engine_submit_mask = engine->logical_mask;
 	desc->hw_context_desc = ce->lrc.lrca;
 	desc->priority = ce->guc_state.prio;
 	desc->context_flags = CONTEXT_REGISTRATION_FLAG_KMD;

@@ -3260,6 +3242,7 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count)
 		}
 
 		ve->base.mask |= sibling->mask;
+		ve->base.logical_mask |= sibling->logical_mask;
 
 		if (n != 0 && ve->base.class != sibling->class) {
 			DRM_DEBUG("invalid mixing of engine class, sibling %d, already %d\n",

@@ -124,7 +124,9 @@ query_engine_info(struct drm_i915_private *i915,
 	for_each_uabi_engine(engine, i915) {
 		info.engine.engine_class = engine->uabi_class;
 		info.engine.engine_instance = engine->uabi_instance;
+		info.flags = I915_ENGINE_INFO_HAS_LOGICAL_INSTANCE;
 		info.capabilities = engine->uabi_capabilities;
+		info.logical_instance = ilog2(engine->logical_mask);
 
 		if (copy_to_user(info_ptr, &info, sizeof(info)))
 			return -EFAULT;

@@ -2726,14 +2726,20 @@ struct drm_i915_engine_info {
 
 	/** @flags: Engine flags. */
 	__u64 flags;
+#define I915_ENGINE_INFO_HAS_LOGICAL_INSTANCE		(1 << 0)
 
 	/** @capabilities: Capabilities of this engine. */
 	__u64 capabilities;
 #define I915_VIDEO_CLASS_CAPABILITY_HEVC		(1 << 0)
 #define I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC	(1 << 1)
 
+	/** @logical_instance: Logical instance of engine */
+	__u16 logical_instance;
+
 	/** @rsvd1: Reserved fields. */
-	__u64 rsvd1[4];
+	__u16 rsvd1[3];
+	/** @rsvd2: Reserved fields. */
+	__u64 rsvd2[3];
 };
 
 /**

@@ -232,8 +232,20 @@ struct intel_context {
 		/** @parent: pointer to parent if child */
 		struct intel_context *parent;
 
+
+		/** @guc_wqi_head: head pointer in work queue */
+		u16 guc_wqi_head;
+		/** @guc_wqi_tail: tail pointer in work queue */
+		u16 guc_wqi_tail;
+
 		/** @guc_number_children: number of children if parent */
 		u8 guc_number_children;
+
+		/**
+		 * @parent_page: page in context used by parent for work queue,
+		 * work queue descriptor
+		 */
+		u8 parent_page;
 	};
 
 #ifdef CONFIG_DRM_I915_SELFTEST

@@ -861,6 +861,11 @@ __lrc_alloc_state(struct intel_context *ce, struct intel_engine_cs *engine)
 		context_size += PAGE_SIZE;
 	}
 
+	if (intel_context_is_parent(ce)) {
+		ce->parent_page = context_size / PAGE_SIZE;
+		context_size += PAGE_SIZE;
+	}
+
 	obj = i915_gem_object_create_lmem(engine->i915, context_size, 0);
 	if (IS_ERR(obj))
 		obj = i915_gem_object_create_shmem(engine->i915, context_size);

@@ -52,7 +52,7 @@
 
 #define GUC_DOORBELL_INVALID		256
 
-#define GUC_WQ_SIZE			(PAGE_SIZE * 2)
+#define GUC_WQ_SIZE			(PAGE_SIZE / 2)
 
 /* Work queue item header definitions */
 #define WQ_STATUS_ACTIVE		1

@@ -340,6 +340,39 @@ static struct i915_priolist *to_priolist(struct rb_node *rb)
 	return rb_entry(rb, struct i915_priolist, node);
 }
 
+/*
+ * When using multi-lrc submission an extra page in the context state is
+ * reserved for the process descriptor and work queue.
+ *
+ * The layout of this page is below:
+ * 0						guc_process_desc
+ * ...						unused
+ * PAGE_SIZE / 2				work queue start
+ * ...						work queue
+ * PAGE_SIZE - 1				work queue end
+ */
+#define WQ_OFFSET	(PAGE_SIZE / 2)
+static u32 __get_process_desc_offset(struct intel_context *ce)
+{
+	GEM_BUG_ON(!ce->parent_page);
+
+	return ce->parent_page * PAGE_SIZE;
+}
+
+static u32 __get_wq_offset(struct intel_context *ce)
+{
+	return __get_process_desc_offset(ce) + WQ_OFFSET;
+}
+
+static struct guc_process_desc *
+__get_process_desc(struct intel_context *ce)
+{
+	return (struct guc_process_desc *)
+		(ce->lrc_reg_state +
+		 ((__get_process_desc_offset(ce) -
+		   LRC_STATE_OFFSET) / sizeof(u32)));
+}
+
 static struct guc_lrc_desc *__get_lrc_desc(struct intel_guc *guc, u32 index)
 {
 	struct guc_lrc_desc *base = guc->lrc_desc_pool_vaddr;

@@ -1342,6 +1375,30 @@ static void unpin_guc_id(struct intel_guc *guc, struct intel_context *ce)
 	spin_unlock_irqrestore(&guc->submission_state.lock, flags);
 }
 
+static int __guc_action_register_multi_lrc(struct intel_guc *guc,
+					   struct intel_context *ce,
+					   u32 guc_id,
+					   u32 offset,
+					   bool loop)
+{
+	struct intel_context *child;
+	u32 action[4 + MAX_ENGINE_INSTANCE];
+	int len = 0;
+
+	GEM_BUG_ON(ce->guc_number_children > MAX_ENGINE_INSTANCE);
+
+	action[len++] = INTEL_GUC_ACTION_REGISTER_CONTEXT_MULTI_LRC;
+	action[len++] = guc_id;
+	action[len++] = ce->guc_number_children + 1;
+	action[len++] = offset;
+	for_each_child(ce, child) {
+		offset += sizeof(struct guc_lrc_desc);
+		action[len++] = offset;
+	}
+
+	return guc_submission_send_busy_loop(guc, action, len, 0, loop);
+}
+
 static int __guc_action_register_context(struct intel_guc *guc,
 					 u32 guc_id,
 					 u32 offset,

@@ -1364,9 +1421,15 @@ static int register_context(struct intel_context *ce, bool loop)
 		ce->guc_id.id * sizeof(struct guc_lrc_desc);
 	int ret;
 
+	GEM_BUG_ON(intel_context_is_child(ce));
 	trace_intel_context_register(ce);
 
-	ret = __guc_action_register_context(guc, ce->guc_id.id, offset, loop);
+	if (intel_context_is_parent(ce))
+		ret = __guc_action_register_multi_lrc(guc, ce, ce->guc_id.id,
+						      offset, loop);
+	else
+		ret = __guc_action_register_context(guc, ce->guc_id.id, offset,
+						    loop);
 	if (likely(!ret)) {
 		unsigned long flags;
 

@@ -1396,6 +1459,7 @@ static int deregister_context(struct intel_context *ce, u32 guc_id, bool loop)
 {
 	struct intel_guc *guc = ce_to_guc(ce);
 
+	GEM_BUG_ON(intel_context_is_child(ce));
 	trace_intel_context_deregister(ce);
 
 	return __guc_action_deregister_context(guc, guc_id, loop);

@@ -1423,6 +1487,7 @@ static int guc_lrc_desc_pin(struct intel_context *ce, bool loop)
 	struct guc_lrc_desc *desc;
 	bool context_registered;
 	intel_wakeref_t wakeref;
+	struct intel_context *child;
 	int ret = 0;
 
 	GEM_BUG_ON(!engine->mask);

@@ -1448,6 +1513,42 @@ static int guc_lrc_desc_pin(struct intel_context *ce, bool loop)
 	desc->context_flags = CONTEXT_REGISTRATION_FLAG_KMD;
 	guc_context_policy_init(engine, desc);
 
+	/*
+	 * Context is a parent, we need to register a process descriptor
+	 * describing a work queue and register all child contexts.
+	 */
+	if (intel_context_is_parent(ce)) {
+		struct guc_process_desc *pdesc;
+
+		ce->guc_wqi_tail = 0;
+		ce->guc_wqi_head = 0;
+
+		desc->process_desc = i915_ggtt_offset(ce->state) +
+			__get_process_desc_offset(ce);
+		desc->wq_addr = i915_ggtt_offset(ce->state) +
+			__get_wq_offset(ce);
+		desc->wq_size = GUC_WQ_SIZE;
+
+		pdesc = __get_process_desc(ce);
+		memset(pdesc, 0, sizeof(*(pdesc)));
+		pdesc->stage_id = ce->guc_id.id;
+		pdesc->wq_base_addr = desc->wq_addr;
+		pdesc->wq_size_bytes = desc->wq_size;
+		pdesc->priority = GUC_CLIENT_PRIORITY_KMD_NORMAL;
+		pdesc->wq_status = WQ_STATUS_ACTIVE;
+
+		for_each_child(ce, child) {
+			desc = __get_lrc_desc(guc, child->guc_id.id);
+
+			desc->engine_class =
+				engine_class_to_guc_class(engine->class);
+			desc->hw_context_desc = child->lrc.lrca;
+			desc->priority = GUC_CLIENT_PRIORITY_KMD_NORMAL;
+			desc->context_flags = CONTEXT_REGISTRATION_FLAG_KMD;
+			guc_context_policy_init(engine, desc);
+		}
+	}
+
 	/*
 	 * The context_lookup xarray is used to determine if the hardware
 	 * context is currently registered. There are two cases in which it

@@ -2858,6 +2959,12 @@ g2h_context_lookup(struct intel_guc *guc, u32 desc_idx)
 		return NULL;
 	}
 
+	if (unlikely(intel_context_is_child(ce))) {
+		drm_err(&guc_to_gt(guc)->i915->drm,
+			"Context is child, desc_idx %u", desc_idx);
+		return NULL;
+	}
+
 	return ce;
 }
 

@@ -748,3 +748,24 @@ void intel_guc_load_status(struct intel_guc *guc, struct drm_printer *p)
 		}
 	}
 }
+
+void intel_guc_write_barrier(struct intel_guc *guc)
+{
+	struct intel_gt *gt = guc_to_gt(guc);
+
+	if (i915_gem_object_is_lmem(guc->ct.vma->obj)) {
+		GEM_BUG_ON(guc->send_regs.fw_domains);
+		/*
+		 * This register is used by the i915 and GuC for MMIO based
+		 * communication. Once we are in this code CTBs are the only
+		 * method the i915 uses to communicate with the GuC so it is
+		 * safe to write to this register (a value of 0 is NOP for MMIO
+		 * communication). If we ever start mixing CTBs and MMIOs a new
+		 * register will have to be chosen.
+		 */
+		intel_uncore_write_fw(gt->uncore, GEN11_SOFT_SCRATCH(0), 0);
+	} else {
+		/* wmb() sufficient for a barrier if in smem */
+		wmb();
+	}
+}

@@ -37,6 +37,12 @@ struct intel_guc {
 	/* Global engine used to submit requests to GuC */
 	struct i915_sched_engine *sched_engine;
 	struct i915_request *stalled_request;
+	enum {
+		STALL_NONE,
+		STALL_REGISTER_CONTEXT,
+		STALL_MOVE_LRC_TAIL,
+		STALL_ADD_REQUEST,
+	} submission_stall_reason;
 
 	/* intel_guc_recv interrupt related state */
 	spinlock_t irq_lock;

@@ -332,4 +338,6 @@ void intel_guc_submission_cancel_requests(struct intel_guc *guc);
 
 void intel_guc_load_status(struct intel_guc *guc, struct drm_printer *p);
 
+void intel_guc_write_barrier(struct intel_guc *guc);
+
 #endif

@@ -377,28 +377,6 @@ static u32 ct_get_next_fence(struct intel_guc_ct *ct)
 	return ++ct->requests.last_fence;
 }
 
-static void write_barrier(struct intel_guc_ct *ct)
-{
-	struct intel_guc *guc = ct_to_guc(ct);
-	struct intel_gt *gt = guc_to_gt(guc);
-
-	if (i915_gem_object_is_lmem(guc->ct.vma->obj)) {
-		GEM_BUG_ON(guc->send_regs.fw_domains);
-		/*
-		 * This register is used by the i915 and GuC for MMIO based
-		 * communication. Once we are in this code CTBs are the only
-		 * method the i915 uses to communicate with the GuC so it is
-		 * safe to write to this register (a value of 0 is NOP for MMIO
-		 * communication). If we ever start mixing CTBs and MMIOs a new
-		 * register will have to be chosen.
-		 */
-		intel_uncore_write_fw(gt->uncore, GEN11_SOFT_SCRATCH(0), 0);
-	} else {
-		/* wmb() sufficient for a barrier if in smem */
-		wmb();
-	}
-}
-
 static int ct_write(struct intel_guc_ct *ct,
 		    const u32 *action,
 		    u32 len /* in dwords */,

@@ -468,7 +446,7 @@ static int ct_write(struct intel_guc_ct *ct,
 	 * make sure H2G buffer update and LRC tail update (if this triggering a
 	 * submission) are visible before updating the descriptor tail
 	 */
-	write_barrier(ct);
+	intel_guc_write_barrier(ct_to_guc(ct));
 
 	/* update local copies */
 	ctb->tail = tail;

@@ -65,12 +65,14 @@
 #define   WQ_TYPE_PSEUDO		(0x2 << WQ_TYPE_SHIFT)
 #define   WQ_TYPE_INORDER		(0x3 << WQ_TYPE_SHIFT)
 #define   WQ_TYPE_NOOP			(0x4 << WQ_TYPE_SHIFT)
-#define WQ_TARGET_SHIFT			10
+#define   WQ_TYPE_MULTI_LRC		(0x5 << WQ_TYPE_SHIFT)
+#define WQ_TARGET_SHIFT			8
 #define WQ_LEN_SHIFT			16
 #define WQ_NO_WCFLUSH_WAIT		(1 << 27)
 #define WQ_PRESENT_WORKLOAD		(1 << 28)
 
-#define WQ_RING_TAIL_SHIFT		20
+#define WQ_GUC_ID_SHIFT			0
+#define WQ_RING_TAIL_SHIFT		18
 #define WQ_RING_TAIL_MAX		0x7FF	/* 2^11 QWords */
 #define WQ_RING_TAIL_MASK		(WQ_RING_TAIL_MAX << WQ_RING_TAIL_SHIFT)
 

@@ -391,6 +391,29 @@ __get_process_desc(struct intel_context *ce)
 		   LRC_STATE_OFFSET) / sizeof(u32)));
 }
 
+static u32 *get_wq_pointer(struct guc_process_desc *desc,
+			   struct intel_context *ce,
+			   u32 wqi_size)
+{
+	/*
+	 * Check for space in work queue. Caching a value of head pointer in
+	 * intel_context structure in order reduce the number accesses to shared
+	 * GPU memory which may be across a PCIe bus.
+	 */
+#define AVAILABLE_SPACE	\
+	CIRC_SPACE(ce->guc_wqi_tail, ce->guc_wqi_head, GUC_WQ_SIZE)
+	if (wqi_size > AVAILABLE_SPACE) {
+		ce->guc_wqi_head = READ_ONCE(desc->head);
+
+		if (wqi_size > AVAILABLE_SPACE)
+			return NULL;
+	}
+#undef AVAILABLE_SPACE
+
+	return ((u32 *)__get_process_desc(ce)) +
+		((WQ_OFFSET + ce->guc_wqi_tail) / sizeof(u32));
+}
+
 static struct guc_lrc_desc *__get_lrc_desc(struct intel_guc *guc, u32 index)
 {
 	struct guc_lrc_desc *base = guc->lrc_desc_pool_vaddr;

@@ -547,10 +570,10 @@ int intel_guc_wait_for_idle(struct intel_guc *guc, long timeout)
 
 static int guc_lrc_desc_pin(struct intel_context *ce, bool loop);
 
-static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)
+static int __guc_add_request(struct intel_guc *guc, struct i915_request *rq)
 {
 	int err = 0;
-	struct intel_context *ce = rq->context;
+	struct intel_context *ce = request_to_scheduling_context(rq);
 	u32 action[3];
 	int len = 0;
 	u32 g2h_len_dw = 0;

@@ -571,26 +594,17 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)
 	GEM_BUG_ON(!atomic_read(&ce->guc_id.ref));
 	GEM_BUG_ON(context_guc_id_invalid(ce));
 
-	/*
-	 * Corner case where the GuC firmware was blown away and reloaded while
-	 * this context was pinned.
-	 */
-	if (unlikely(!lrc_desc_registered(guc, ce->guc_id.id))) {
-		err = guc_lrc_desc_pin(ce, false);
-		if (unlikely(err))
-			return err;
-	}
-
 	spin_lock(&ce->guc_state.lock);
 
 	/*
 	 * The request / context will be run on the hardware when scheduling
-	 * gets enabled in the unblock.
+	 * gets enabled in the unblock. For multi-lrc we still submit the
+	 * context to move the LRC tails.
 	 */
-	if (unlikely(context_blocked(ce)))
+	if (unlikely(context_blocked(ce) && !intel_context_is_parent(ce)))
 		goto out;
 
-	enabled = context_enabled(ce);
+	enabled = context_enabled(ce) || context_blocked(ce);
 
 	if (!enabled) {
 		action[len++] = INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_SET;

@@ -609,6 +623,18 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)
 		trace_intel_context_sched_enable(ce);
 		atomic_inc(&guc->outstanding_submission_g2h);
 		set_context_enabled(ce);
+
+		/*
+		 * Without multi-lrc KMD does the submission step (moving the
+		 * lrc tail) so enabling scheduling is sufficient to submit the
+		 * context. This isn't the case in multi-lrc submission as the
+		 * GuC needs to move the tails, hence the need for another H2G
+		 * to submit a multi-lrc context after enabling scheduling.
+		 */
+		if (intel_context_is_parent(ce)) {
+			action[0] = INTEL_GUC_ACTION_SCHED_CONTEXT;
+			err = intel_guc_send_nb(guc, action, len - 1, 0);
+		}
 	} else if (!enabled) {
 		clr_context_pending_enable(ce);
 		intel_context_put(ce);

@@ -621,6 +647,18 @@ static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)
 	return err;
 }
 
+static int guc_add_request(struct intel_guc *guc, struct i915_request *rq)
+{
+	int ret = __guc_add_request(guc, rq);
+
+	if (unlikely(ret == -EBUSY)) {
+		guc->stalled_request= rq;
+		guc->submission_stall_reason = STALL_ADD_REQUEST;
+	}
+
+	return ret;
+}
+
 static void guc_set_lrc_tail(struct i915_request *rq)
 {
 	rq->context->lrc_reg_state[CTX_RING_TAIL] =

@@ -632,6 +670,127 @@ static int rq_prio(const struct i915_request *rq)
 	return rq->sched.attr.priority;
 }
 
+static bool is_multi_lrc_rq(struct i915_request *rq)
+{
+	return intel_context_is_child(rq->context) ||
+		intel_context_is_parent(rq->context);
+}
+
+static bool can_merge_rq(struct i915_request *rq,
+			 struct i915_request *last)
+{
+	return request_to_scheduling_context(rq) ==
+		request_to_scheduling_context(last);
+}
+
+static u32 wq_space_until_wrap(struct intel_context *ce)
+{
+	return (GUC_WQ_SIZE - ce->guc_wqi_tail);
+}
+
+static void write_wqi(struct guc_process_desc *desc,
+		      struct intel_context *ce,
+		      u32 wqi_size)
+{
+	/*
+	 * Ensure WQE are visible before updating tail
+	 */
+	intel_guc_write_barrier(ce_to_guc(ce));
+
+	ce->guc_wqi_tail = (ce->guc_wqi_tail + wqi_size) & (GUC_WQ_SIZE - 1);
+	WRITE_ONCE(desc->tail, ce->guc_wqi_tail);
+}
+
+static int guc_wq_noop_append(struct intel_context *ce)
+{
+	struct guc_process_desc *desc = __get_process_desc(ce);
+	u32 *wqi = get_wq_pointer(desc, ce, wq_space_until_wrap(ce));
+
+	if (!wqi)
+		return -EBUSY;
+
+	*wqi = WQ_TYPE_NOOP |
+		((wq_space_until_wrap(ce) / sizeof(u32) - 1) << WQ_LEN_SHIFT);
+	ce->guc_wqi_tail = 0;
+
+	return 0;
+}
+
+static int __guc_wq_item_append(struct i915_request *rq)
+{
+	struct intel_context *ce = request_to_scheduling_context(rq);
+	struct intel_context *child;
+	struct guc_process_desc *desc = __get_process_desc(ce);
+	unsigned int wqi_size = (ce->guc_number_children + 4) *
+		sizeof(u32);
+	u32 *wqi;
+	int ret;
+
+	/* Ensure context is in correct state updating work queue */
+	GEM_BUG_ON(!atomic_read(&ce->guc_id.ref));
+	GEM_BUG_ON(context_guc_id_invalid(ce));
+	GEM_BUG_ON(context_wait_for_deregister_to_register(ce));
+	GEM_BUG_ON(!lrc_desc_registered(ce_to_guc(ce), ce->guc_id.id));
+
+	/* Insert NOOP if this work queue item will wrap the tail pointer. */
+	if (wqi_size > wq_space_until_wrap(ce)) {
+		ret = guc_wq_noop_append(ce);
+		if (ret)
+			return ret;
+	}
+
+	wqi = get_wq_pointer(desc, ce, wqi_size);
+	if (!wqi)
+		return -EBUSY;
+
+	*wqi++ = WQ_TYPE_MULTI_LRC |
+		((wqi_size / sizeof(u32) - 1) << WQ_LEN_SHIFT);
+	*wqi++ = ce->lrc.lrca;
+	*wqi++ = (ce->guc_id.id << WQ_GUC_ID_SHIFT) |
+		 ((ce->ring->tail / sizeof(u64)) << WQ_RING_TAIL_SHIFT);
+	*wqi++ = 0;	/* fence_id */
+	for_each_child(ce, child)
+		*wqi++ = child->ring->tail / sizeof(u64);
+
+	write_wqi(desc, ce, wqi_size);
+
+	return 0;
+}
+
+static int guc_wq_item_append(struct intel_guc *guc,
+			      struct i915_request *rq)
+{
+	struct intel_context *ce = request_to_scheduling_context(rq);
+	int ret = 0;
+
+	if (likely(!intel_context_is_banned(ce))) {
+		ret = __guc_wq_item_append(rq);
+
+		if (unlikely(ret == -EBUSY)) {
+			guc->stalled_request = rq;
+			guc->submission_stall_reason = STALL_MOVE_LRC_TAIL;
+		}
+	}
+
+	return ret;
+}
+
+static bool multi_lrc_submit(struct i915_request *rq)
+{
+	struct intel_context *ce = request_to_scheduling_context(rq);
+
+	intel_ring_set_tail(rq->ring, rq->tail);
+
+	/*
+	 * We expect the front end (execbuf IOCTL) to set this flag on the last
+	 * request generated from a multi-BB submission. This indicates to the
+	 * backend (GuC interface) that we should submit this context thus
+	 * submitting all the requests generated in parallel.
+	 */
+	return test_bit(I915_FENCE_FLAG_SUBMIT_PARALLEL, &rq->fence.flags) ||
+		intel_context_is_banned(ce);
+}
+
 static int guc_dequeue_one_context(struct intel_guc *guc)
 {
 	struct i915_sched_engine * const sched_engine = guc->sched_engine;

@@ -645,7 +804,17 @@ static int guc_dequeue_one_context(struct intel_guc *guc)
 	if (guc->stalled_request) {
 		submit = true;
 		last = guc->stalled_request;
-		goto resubmit;
+
+		switch (guc->submission_stall_reason) {
+		case STALL_REGISTER_CONTEXT:
+			goto register_context;
+		case STALL_MOVE_LRC_TAIL:
+			goto move_lrc_tail;
+		case STALL_ADD_REQUEST:
+			goto add_request;
+		default:
+			MISSING_CASE(guc->submission_stall_reason);
+		}
 	}
 
 	while ((rb = rb_first_cached(&sched_engine->queue))) {

@@ -653,8 +822,8 @@ static int guc_dequeue_one_context(struct intel_guc *guc)
 		struct i915_request *rq, *rn;
 
 		priolist_for_each_request_consume(rq, rn, p) {
-			if (last && rq->context != last->context)
-				goto done;
+			if (last && !can_merge_rq(rq, last))
+				goto register_context;
 
 			list_del_init(&rq->sched.link);
 

@@ -662,33 +831,84 @@ static int guc_dequeue_one_context(struct intel_guc *guc)
 
 			trace_i915_request_in(rq, 0);
 			last = rq;
-			submit = true;
+
+			if (is_multi_lrc_rq(rq)) {
+				/*
+				 * We need to coalesce all multi-lrc requests in
+				 * a relationship into a single H2G. We are
+				 * guaranteed that all of these requests will be
+				 * submitted sequentially.
+				 */
+				if (multi_lrc_submit(rq)) {
+					submit = true;
+					goto register_context;
+				}
+			} else {
+				submit = true;
+			}
 		}
 
 		rb_erase_cached(&p->node, &sched_engine->queue);
 		i915_priolist_free(p);
 	}
-done:
+
+register_context:
 	if (submit) {
-		guc_set_lrc_tail(last);
-resubmit:
+		struct intel_context *ce = request_to_scheduling_context(last);
+
+		if (unlikely(!lrc_desc_registered(guc, ce->guc_id.id) &&
+			     !intel_context_is_banned(ce))) {
+			ret = guc_lrc_desc_pin(ce, false);
+			if (unlikely(ret == -EPIPE)) {
+				goto deadlk;
+			} else if (ret == -EBUSY) {
+				guc->stalled_request = last;
+				guc->submission_stall_reason =
+					STALL_REGISTER_CONTEXT;
+				goto schedule_tasklet;
+			} else if (ret != 0) {
+				GEM_WARN_ON(ret);	/* Unexpected */
+				goto deadlk;
+			}
+		}
+
+move_lrc_tail:
+		if (is_multi_lrc_rq(last)) {
+			ret = guc_wq_item_append(guc, last);
+			if (ret == -EBUSY) {
+				goto schedule_tasklet;
+			} else if (ret != 0) {
+				GEM_WARN_ON(ret);	/* Unexpected */
+				goto deadlk;
+			}
+		} else {
+			guc_set_lrc_tail(last);
+		}
+
+add_request:
 		ret = guc_add_request(guc, last);
-		if (unlikely(ret == -EPIPE))
+		if (unlikely(ret == -EPIPE)) {
+			goto deadlk;
+		} else if (ret == -EBUSY) {
+			goto schedule_tasklet;
+		} else if (ret != 0) {
+			GEM_WARN_ON(ret);	/* Unexpected */
 			goto deadlk;
-		else if (ret == -EBUSY) {
-			tasklet_schedule(&sched_engine->tasklet);
-			guc->stalled_request = last;
-			return false;
 		}
 	}
 
 	guc->stalled_request = NULL;
+	guc->submission_stall_reason = STALL_NONE;
 	return submit;
 
 deadlk:
 	sched_engine->tasklet.callback = NULL;
 	tasklet_disable_nosync(&sched_engine->tasklet);
 	return false;
+
+schedule_tasklet:
+	tasklet_schedule(&sched_engine->tasklet);
+	return false;
 }
 
 static void guc_submission_tasklet(struct tasklet_struct *t)

@@ -1227,10 +1447,16 @@ static int guc_bypass_tasklet_submit(struct intel_guc *guc,
 
 	trace_i915_request_in(rq, 0);
 
-	guc_set_lrc_tail(rq);
-	ret = guc_add_request(guc, rq);
-	if (ret == -EBUSY)
-		guc->stalled_request = rq;
+	if (is_multi_lrc_rq(rq)) {
+		if (multi_lrc_submit(rq)) {
+			ret = guc_wq_item_append(guc, rq);
+			if (!ret)
+				ret = guc_add_request(guc, rq);
+		}
+	} else {
+		guc_set_lrc_tail(rq);
+		ret = guc_add_request(guc, rq);
+	}
 
 	if (unlikely(ret == -EPIPE))
 		disable_submission(guc);

@@ -1238,6 +1464,16 @@ static int guc_bypass_tasklet_submit(struct intel_guc *guc,
 	return ret;
 }
 
+bool need_tasklet(struct intel_guc *guc, struct i915_request *rq)
+{
+	struct i915_sched_engine *sched_engine = rq->engine->sched_engine;
+	struct intel_context *ce = request_to_scheduling_context(rq);
+
+	return submission_disabled(guc) || guc->stalled_request ||
+		!i915_sched_engine_is_empty(sched_engine) ||
+		!lrc_desc_registered(guc, ce->guc_id.id);
+}
+
 static void guc_submit_request(struct i915_request *rq)
 {
 	struct i915_sched_engine *sched_engine = rq->engine->sched_engine;