1 files changed, 36 insertions, 25 deletions

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 9437a5eb07cf..3ff3d7b49969 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -59,7 +59,8 @@ static int blk_mq_poll_stats_bkt(const struct request *rq)
 }
 
 /*
- * Check if any of the ctx's have pending work in this hardware queue
+ * Check if any of the ctx, dispatch list or elevator
+ * have pending work in this hardware queue.
  */
 static bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
 {
@@ -331,7 +332,6 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data,
 #if defined(CONFIG_BLK_DEV_INTEGRITY)
 	rq->nr_integrity_segments = 0;
 #endif
-	rq->special = NULL;
 	/* tag was already set */
 	rq->extra_len = 0;
 	WRITE_ONCE(rq->deadline, 0);
@@ -340,7 +340,6 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data,
 
 	rq->end_io = NULL;
 	rq->end_io_data = NULL;
-	rq->next_rq = NULL;
 
 	data->ctx->rq_dispatched[op_is_sync(op)]++;
 	refcount_set(&rq->ref, 1);
@@ -364,7 +363,7 @@ static struct request *blk_mq_get_request(struct request_queue *q,
 	}
 	if (likely(!data->hctx))
 		data->hctx = blk_mq_map_queue(q, data->cmd_flags,
-						data->ctx->cpu);
+						data->ctx);
 	if (data->cmd_flags & REQ_NOWAIT)
 		data->flags |= BLK_MQ_REQ_NOWAIT;
 
@@ -550,8 +549,6 @@ inline void __blk_mq_end_request(struct request *rq, blk_status_t error)
 		rq_qos_done(rq->q, rq);
 		rq->end_io(rq, error);
 	} else {
-		if (unlikely(blk_bidi_rq(rq)))
-			blk_mq_free_request(rq->next_rq);
 		blk_mq_free_request(rq);
 	}
 }
@@ -786,7 +783,6 @@ void blk_mq_add_to_requeue_list(struct request *rq, bool at_head,
 	if (kick_requeue_list)
 		blk_mq_kick_requeue_list(q);
 }
-EXPORT_SYMBOL(blk_mq_add_to_requeue_list);
 
 void blk_mq_kick_requeue_list(struct request_queue *q)
 {
@@ -1076,7 +1072,13 @@ static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode,
 	hctx = container_of(wait, struct blk_mq_hw_ctx, dispatch_wait);
 
 	spin_lock(&hctx->dispatch_wait_lock);
-	list_del_init(&wait->entry);
+	if (!list_empty(&wait->entry)) {
+		struct sbitmap_queue *sbq;
+
+		list_del_init(&wait->entry);
+		sbq = &hctx->tags->bitmap_tags;
+		atomic_dec(&sbq->ws_active);
+	}
 	spin_unlock(&hctx->dispatch_wait_lock);
 
 	blk_mq_run_hw_queue(hctx, true);
@@ -1092,13 +1094,13 @@ static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode,
 static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx *hctx,
 				 struct request *rq)
 {
+	struct sbitmap_queue *sbq = &hctx->tags->bitmap_tags;
 	struct wait_queue_head *wq;
 	wait_queue_entry_t *wait;
 	bool ret;
 
 	if (!(hctx->flags & BLK_MQ_F_TAG_SHARED)) {
-		if (!test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
-			set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state);
+		blk_mq_sched_mark_restart_hctx(hctx);
 
 		/*
 		 * It's possible that a tag was freed in the window between the
@@ -1115,7 +1117,7 @@ static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx *hctx,
 	if (!list_empty_careful(&wait->entry))
 		return false;
 
-	wq = &bt_wait_ptr(&hctx->tags->bitmap_tags, hctx)->wait;
+	wq = &bt_wait_ptr(sbq, hctx)->wait;
 
 	spin_lock_irq(&wq->lock);
 	spin_lock(&hctx->dispatch_wait_lock);
@@ -1125,6 +1127,7 @@ static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx *hctx,
 		return false;
 	}
 
+	atomic_inc(&sbq->ws_active);
 	wait->flags &= ~WQ_FLAG_EXCLUSIVE;
 	__add_wait_queue(wq, wait);
 
@@ -1145,6 +1148,7 @@ static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx *hctx,
 	 * someone else gets the wakeup.
 	 */
 	list_del_init(&wait->entry);
+	atomic_dec(&sbq->ws_active);
 	spin_unlock(&hctx->dispatch_wait_lock);
 	spin_unlock_irq(&wq->lock);
 
@@ -2069,7 +2073,7 @@ struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set,
 	struct blk_mq_tags *tags;
 	int node;
 
-	node = blk_mq_hw_queue_to_node(&set->map[0], hctx_idx);
+	node = blk_mq_hw_queue_to_node(&set->map[HCTX_TYPE_DEFAULT], hctx_idx);
 	if (node == NUMA_NO_NODE)
 		node = set->numa_node;
 
@@ -2125,7 +2129,7 @@ int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags,
 	size_t rq_size, left;
 	int node;
 
-	node = blk_mq_hw_queue_to_node(&set->map[0], hctx_idx);
+	node = blk_mq_hw_queue_to_node(&set->map[HCTX_TYPE_DEFAULT], hctx_idx);
 	if (node == NUMA_NO_NODE)
 		node = set->numa_node;
 
@@ -2424,7 +2428,7 @@ static void blk_mq_map_swqueue(struct request_queue *q)
 	 * If the cpu isn't present, the cpu is mapped to first hctx.
 	 */
 	for_each_possible_cpu(i) {
-		hctx_idx = set->map[0].mq_map[i];
+		hctx_idx = set->map[HCTX_TYPE_DEFAULT].mq_map[i];
 		/* unmapped hw queue can be remapped after CPU topo changed */
 		if (!set->tags[hctx_idx] &&
 		    !__blk_mq_alloc_rq_map(set, hctx_idx)) {
@@ -2434,16 +2438,19 @@ static void blk_mq_map_swqueue(struct request_queue *q)
 			 * case, remap the current ctx to hctx[0] which
 			 * is guaranteed to always have tags allocated
 			 */
-			set->map[0].mq_map[i] = 0;
+			set->map[HCTX_TYPE_DEFAULT].mq_map[i] = 0;
 		}
 
 		ctx = per_cpu_ptr(q->queue_ctx, i);
 		for (j = 0; j < set->nr_maps; j++) {
-			if (!set->map[j].nr_queues)
+			if (!set->map[j].nr_queues) {
+				ctx->hctxs[j] = blk_mq_map_queue_type(q,
+						HCTX_TYPE_DEFAULT, i);
 				continue;
+			}
 
 			hctx = blk_mq_map_queue_type(q, j, i);
-
+			ctx->hctxs[j] = hctx;
 			/*
 			 * If the CPU is already set in the mask, then we've
 			 * mapped this one already. This can happen if
@@ -2463,6 +2470,10 @@ static void blk_mq_map_swqueue(struct request_queue *q)
 			 */
 			BUG_ON(!hctx->nr_ctx);
 		}
+
+		for (; j < HCTX_MAX_TYPES; j++)
+			ctx->hctxs[j] = blk_mq_map_queue_type(q,
+					HCTX_TYPE_DEFAULT, i);
 	}
 
 	mutex_unlock(&q->sysfs_lock);
@@ -2734,7 +2745,7 @@ static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 		int node;
 		struct blk_mq_hw_ctx *hctx;
 
-		node = blk_mq_hw_queue_to_node(&set->map[0], i);
+		node = blk_mq_hw_queue_to_node(&set->map[HCTX_TYPE_DEFAULT], i);
 		/*
 		 * If the hw queue has been mapped to another numa node,
 		 * we need to realloc the hctx. If allocation fails, fallback
@@ -2838,9 +2849,6 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	    set->map[HCTX_TYPE_POLL].nr_queues)
 		blk_queue_flag_set(QUEUE_FLAG_POLL, q);
 
-	if (!(set->flags & BLK_MQ_F_SG_MERGE))
-		blk_queue_flag_set(QUEUE_FLAG_NO_SG_MERGE, q);
-
 	q->sg_reserved_size = INT_MAX;
 
 	INIT_DELAYED_WORK(&q->requeue_work, blk_mq_requeue_work);
@@ -2857,7 +2865,7 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	/*
 	 * Default to classic polling
 	 */
-	q->poll_nsec = -1;
+	q->poll_nsec = BLK_MQ_POLL_CLASSIC;
 
 	blk_mq_init_cpu_queues(q, set->nr_hw_queues);
 	blk_mq_add_queue_tag_set(set, q);
@@ -2968,7 +2976,7 @@ static int blk_mq_update_queue_map(struct blk_mq_tag_set *set)
 		return set->ops->map_queues(set);
 	} else {
 		BUG_ON(set->nr_maps > 1);
-		return blk_mq_map_queues(&set->map[0]);
+		return blk_mq_map_queues(&set->map[HCTX_TYPE_DEFAULT]);
 	}
 }
 
@@ -3090,6 +3098,9 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
 	if (!set)
 		return -EINVAL;
 
+	if (q->nr_requests == nr)
+		return 0;
+
 	blk_mq_freeze_queue(q);
 	blk_mq_quiesce_queue(q);
 
@@ -3235,7 +3246,7 @@ fallback:
 			pr_warn("Increasing nr_hw_queues to %d fails, fallback to %d\n",
 					nr_hw_queues, prev_nr_hw_queues);
 			set->nr_hw_queues = prev_nr_hw_queues;
-			blk_mq_map_queues(&set->map[0]);
+			blk_mq_map_queues(&set->map[HCTX_TYPE_DEFAULT]);
 			goto fallback;
 		}
 		blk_mq_map_swqueue(q);
@@ -3389,7 +3400,7 @@ static bool blk_mq_poll_hybrid(struct request_queue *q,
 {
 	struct request *rq;
 
-	if (q->poll_nsec == -1)
+	if (q->poll_nsec == BLK_MQ_POLL_CLASSIC)
 		return false;
 
 	if (!blk_qc_t_is_internal(cookie))