1 files changed, 141 insertions, 87 deletions

diff --git a/block/blk-mq.c b/block/blk-mq.c
index ade8a2d1b0aa..f53779692c77 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -89,7 +89,8 @@ static int blk_mq_queue_enter(struct request_queue *q, gfp_t gfp)
 			return -EBUSY;
 
 		ret = wait_event_interruptible(q->mq_freeze_wq,
-				!q->mq_freeze_depth || blk_queue_dying(q));
+				!atomic_read(&q->mq_freeze_depth) ||
+				blk_queue_dying(q));
 		if (blk_queue_dying(q))
 			return -ENODEV;
 		if (ret)
@@ -112,13 +113,10 @@ static void blk_mq_usage_counter_release(struct percpu_ref *ref)
 
 void blk_mq_freeze_queue_start(struct request_queue *q)
 {
-	bool freeze;
+	int freeze_depth;
 
-	spin_lock_irq(q->queue_lock);
-	freeze = !q->mq_freeze_depth++;
-	spin_unlock_irq(q->queue_lock);
-
-	if (freeze) {
+	freeze_depth = atomic_inc_return(&q->mq_freeze_depth);
+	if (freeze_depth == 1) {
 		percpu_ref_kill(&q->mq_usage_counter);
 		blk_mq_run_hw_queues(q, false);
 	}
@@ -143,13 +141,11 @@ EXPORT_SYMBOL_GPL(blk_mq_freeze_queue);
 
 void blk_mq_unfreeze_queue(struct request_queue *q)
 {
-	bool wake;
+	int freeze_depth;
 
-	spin_lock_irq(q->queue_lock);
-	wake = !--q->mq_freeze_depth;
-	WARN_ON_ONCE(q->mq_freeze_depth < 0);
-	spin_unlock_irq(q->queue_lock);
-	if (wake) {
+	freeze_depth = atomic_dec_return(&q->mq_freeze_depth);
+	WARN_ON_ONCE(freeze_depth < 0);
+	if (!freeze_depth) {
 		percpu_ref_reinit(&q->mq_usage_counter);
 		wake_up_all(&q->mq_freeze_wq);
 	}
@@ -677,8 +673,11 @@ static void blk_mq_rq_timer(unsigned long priv)
 		data.next = blk_rq_timeout(round_jiffies_up(data.next));
 		mod_timer(&q->timeout, data.next);
 	} else {
-		queue_for_each_hw_ctx(q, hctx, i)
-			blk_mq_tag_idle(hctx);
+		queue_for_each_hw_ctx(q, hctx, i) {
+			/* the hctx may be unmapped, so check it here */
+			if (blk_mq_hw_queue_mapped(hctx))
+				blk_mq_tag_idle(hctx);
+		}
 	}
 }
 
@@ -855,6 +854,16 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 		spin_lock(&hctx->lock);
 		list_splice(&rq_list, &hctx->dispatch);
 		spin_unlock(&hctx->lock);
+		/*
+		 * the queue is expected stopped with BLK_MQ_RQ_QUEUE_BUSY, but
+		 * it's possible the queue is stopped and restarted again
+		 * before this. Queue restart will dispatch requests. And since
+		 * requests in rq_list aren't added into hctx->dispatch yet,
+		 * the requests in rq_list might get lost.
+		 *
+		 * blk_mq_run_hw_queue() already checks the STOPPED bit
+		 **/
+		blk_mq_run_hw_queue(hctx, true);
 	}
 }
 
@@ -1224,6 +1233,38 @@ static struct request *blk_mq_map_request(struct request_queue *q,
 	return rq;
 }
 
+static int blk_mq_direct_issue_request(struct request *rq)
+{
+	int ret;
+	struct request_queue *q = rq->q;
+	struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q,
+			rq->mq_ctx->cpu);
+	struct blk_mq_queue_data bd = {
+		.rq = rq,
+		.list = NULL,
+		.last = 1
+	};
+
+	/*
+	 * For OK queue, we are done. For error, kill it. Any other
+	 * error (busy), just add it to our list as we previously
+	 * would have done
+	 */
+	ret = q->mq_ops->queue_rq(hctx, &bd);
+	if (ret == BLK_MQ_RQ_QUEUE_OK)
+		return 0;
+	else {
+		__blk_mq_requeue_request(rq);
+
+		if (ret == BLK_MQ_RQ_QUEUE_ERROR) {
+			rq->errors = -EIO;
+			blk_mq_end_request(rq, rq->errors);
+			return 0;
+		}
+		return -1;
+	}
+}
+
 /*
  * Multiple hardware queue variant. This will not use per-process plugs,
  * but will attempt to bypass the hctx queueing if we can go straight to
@@ -1235,6 +1276,9 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio)
 	const int is_flush_fua = bio->bi_rw & (REQ_FLUSH | REQ_FUA);
 	struct blk_map_ctx data;
 	struct request *rq;
+	unsigned int request_count = 0;
+	struct blk_plug *plug;
+	struct request *same_queue_rq = NULL;
 
 	blk_queue_bounce(q, &bio);
 
@@ -1243,6 +1287,10 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio)
 		return;
 	}
 
+	if (!is_flush_fua && !blk_queue_nomerges(q) &&
+	    blk_attempt_plug_merge(q, bio, &request_count, &same_queue_rq))
+		return;
+
 	rq = blk_mq_map_request(q, bio, &data);
 	if (unlikely(!rq))
 		return;
@@ -1253,38 +1301,42 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio)
 		goto run_queue;
 	}
 
+	plug = current->plug;
 	/*
 	 * If the driver supports defer issued based on 'last', then
 	 * queue it up like normal since we can potentially save some
 	 * CPU this way.
 	 */
-	if (is_sync && !(data.hctx->flags & BLK_MQ_F_DEFER_ISSUE)) {
-		struct blk_mq_queue_data bd = {
-			.rq = rq,
-			.list = NULL,
-			.last = 1
-		};
-		int ret;
+	if (((plug && !blk_queue_nomerges(q)) || is_sync) &&
+	    !(data.hctx->flags & BLK_MQ_F_DEFER_ISSUE)) {
+		struct request *old_rq = NULL;
 
 		blk_mq_bio_to_request(rq, bio);
 
 		/*
-		 * For OK queue, we are done. For error, kill it. Any other
-		 * error (busy), just add it to our list as we previously
-		 * would have done
+		 * we do limited pluging. If bio can be merged, do merge.
+		 * Otherwise the existing request in the plug list will be
+		 * issued. So the plug list will have one request at most
 		 */
-		ret = q->mq_ops->queue_rq(data.hctx, &bd);
-		if (ret == BLK_MQ_RQ_QUEUE_OK)
-			goto done;
-		else {
-			__blk_mq_requeue_request(rq);
-
-			if (ret == BLK_MQ_RQ_QUEUE_ERROR) {
-				rq->errors = -EIO;
-				blk_mq_end_request(rq, rq->errors);
-				goto done;
+		if (plug) {
+			/*
+			 * The plug list might get flushed before this. If that
+			 * happens, same_queue_rq is invalid and plug list is empty
+			 **/
+			if (same_queue_rq && !list_empty(&plug->mq_list)) {
+				old_rq = same_queue_rq;
+				list_del_init(&old_rq->queuelist);
 			}
-		}
+			list_add_tail(&rq->queuelist, &plug->mq_list);
+		} else /* is_sync */
+			old_rq = rq;
+		blk_mq_put_ctx(data.ctx);
+		if (!old_rq)
+			return;
+		if (!blk_mq_direct_issue_request(old_rq))
+			return;
+		blk_mq_insert_request(old_rq, false, true, true);
+		return;
 	}
 
 	if (!blk_mq_merge_queue_io(data.hctx, data.ctx, rq, bio)) {
@@ -1297,7 +1349,6 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio)
 run_queue:
 		blk_mq_run_hw_queue(data.hctx, !is_sync || is_flush_fua);
 	}
-done:
 	blk_mq_put_ctx(data.ctx);
 }
 
@@ -1309,16 +1360,11 @@ static void blk_sq_make_request(struct request_queue *q, struct bio *bio)
 {
 	const int is_sync = rw_is_sync(bio->bi_rw);
 	const int is_flush_fua = bio->bi_rw & (REQ_FLUSH | REQ_FUA);
-	unsigned int use_plug, request_count = 0;
+	struct blk_plug *plug;
+	unsigned int request_count = 0;
 	struct blk_map_ctx data;
 	struct request *rq;
 
-	/*
-	 * If we have multiple hardware queues, just go directly to
-	 * one of those for sync IO.
-	 */
-	use_plug = !is_flush_fua && !is_sync;
-
 	blk_queue_bounce(q, &bio);
 
 	if (bio_integrity_enabled(bio) && bio_integrity_prep(bio)) {
@@ -1326,8 +1372,8 @@ static void blk_sq_make_request(struct request_queue *q, struct bio *bio)
 		return;
 	}
 
-	if (use_plug && !blk_queue_nomerges(q) &&
-	    blk_attempt_plug_merge(q, bio, &request_count))
+	if (!is_flush_fua && !blk_queue_nomerges(q) &&
+	    blk_attempt_plug_merge(q, bio, &request_count, NULL))
 		return;
 
 	rq = blk_mq_map_request(q, bio, &data);
@@ -1345,21 +1391,18 @@ static void blk_sq_make_request(struct request_queue *q, struct bio *bio)
 	 * utilize that to temporarily store requests until the task is
 	 * either done or scheduled away.
 	 */
-	if (use_plug) {
-		struct blk_plug *plug = current->plug;
-
-		if (plug) {
-			blk_mq_bio_to_request(rq, bio);
-			if (list_empty(&plug->mq_list))
-				trace_block_plug(q);
-			else if (request_count >= BLK_MAX_REQUEST_COUNT) {
-				blk_flush_plug_list(plug, false);
-				trace_block_plug(q);
-			}
-			list_add_tail(&rq->queuelist, &plug->mq_list);
-			blk_mq_put_ctx(data.ctx);
-			return;
+	plug = current->plug;
+	if (plug) {
+		blk_mq_bio_to_request(rq, bio);
+		if (list_empty(&plug->mq_list))
+			trace_block_plug(q);
+		else if (request_count >= BLK_MAX_REQUEST_COUNT) {
+			blk_flush_plug_list(plug, false);
+			trace_block_plug(q);
 		}
+		list_add_tail(&rq->queuelist, &plug->mq_list);
+		blk_mq_put_ctx(data.ctx);
+		return;
 	}
 
 	if (!blk_mq_merge_queue_io(data.hctx, data.ctx, rq, bio)) {
@@ -1495,7 +1538,6 @@ static struct blk_mq_tags *blk_mq_init_rq_map(struct blk_mq_tag_set *set,
 			i++;
 		}
 	}
-
 	return tags;
 
 fail:
@@ -1571,22 +1613,6 @@ static int blk_mq_hctx_cpu_offline(struct blk_mq_hw_ctx *hctx, int cpu)
 	return NOTIFY_OK;
 }
 
-static int blk_mq_hctx_cpu_online(struct blk_mq_hw_ctx *hctx, int cpu)
-{
-	struct request_queue *q = hctx->queue;
-	struct blk_mq_tag_set *set = q->tag_set;
-
-	if (set->tags[hctx->queue_num])
-		return NOTIFY_OK;
-
-	set->tags[hctx->queue_num] = blk_mq_init_rq_map(set, hctx->queue_num);
-	if (!set->tags[hctx->queue_num])
-		return NOTIFY_STOP;
-
-	hctx->tags = set->tags[hctx->queue_num];
-	return NOTIFY_OK;
-}
-
 static int blk_mq_hctx_notify(void *data, unsigned long action,
 			      unsigned int cpu)
 {
@@ -1594,12 +1620,16 @@ static int blk_mq_hctx_notify(void *data, unsigned long action,
 
 	if (action == CPU_DEAD || action == CPU_DEAD_FROZEN)
 		return blk_mq_hctx_cpu_offline(hctx, cpu);
-	else if (action == CPU_ONLINE || action == CPU_ONLINE_FROZEN)
-		return blk_mq_hctx_cpu_online(hctx, cpu);
+
+	/*
+	 * In case of CPU online, tags may be reallocated
+	 * in blk_mq_map_swqueue() after mapping is updated.
+	 */
 
 	return NOTIFY_OK;
 }
 
+/* hctx->ctxs will be freed in queue's release handler */
 static void blk_mq_exit_hctx(struct request_queue *q,
 		struct blk_mq_tag_set *set,
 		struct blk_mq_hw_ctx *hctx, unsigned int hctx_idx)
@@ -1618,7 +1648,6 @@ static void blk_mq_exit_hctx(struct request_queue *q,
 
 	blk_mq_unregister_cpu_notifier(&hctx->cpu_notifier);
 	blk_free_flush_queue(hctx->fq);
-	kfree(hctx->ctxs);
 	blk_mq_free_bitmap(&hctx->ctx_map);
 }
 
@@ -1775,6 +1804,7 @@ static void blk_mq_map_swqueue(struct request_queue *q)
 	unsigned int i;
 	struct blk_mq_hw_ctx *hctx;
 	struct blk_mq_ctx *ctx;
+	struct blk_mq_tag_set *set = q->tag_set;
 
 	queue_for_each_hw_ctx(q, hctx, i) {
 		cpumask_clear(hctx->cpumask);
@@ -1791,6 +1821,7 @@ static void blk_mq_map_swqueue(struct request_queue *q)
 
 		hctx = q->mq_ops->map_queue(q, i);
 		cpumask_set_cpu(i, hctx->cpumask);
+		cpumask_set_cpu(i, hctx->tags->cpumask);
 		ctx->index_hw = hctx->nr_ctx;
 		hctx->ctxs[hctx->nr_ctx++] = ctx;
 	}
@@ -1803,16 +1834,20 @@ static void blk_mq_map_swqueue(struct request_queue *q)
 		 * disable it and free the request entries.
 		 */
 		if (!hctx->nr_ctx) {
-			struct blk_mq_tag_set *set = q->tag_set;
-
 			if (set->tags[i]) {
 				blk_mq_free_rq_map(set, set->tags[i], i);
 				set->tags[i] = NULL;
-				hctx->tags = NULL;
 			}
+			hctx->tags = NULL;
 			continue;
 		}
 
+		/* unmapped hw queue can be remapped after CPU topo changed */
+		if (!set->tags[i])
+			set->tags[i] = blk_mq_init_rq_map(set, i);
+		hctx->tags = set->tags[i];
+		WARN_ON(!hctx->tags);
+
 		/*
 		 * Set the map size to the number of mapped software queues.
 		 * This is more accurate and more efficient than looping
@@ -1886,8 +1921,12 @@ void blk_mq_release(struct request_queue *q)
 	unsigned int i;
 
 	/* hctx kobj stays in hctx */
-	queue_for_each_hw_ctx(q, hctx, i)
+	queue_for_each_hw_ctx(q, hctx, i) {
+		if (!hctx)
+			continue;
+		kfree(hctx->ctxs);
 		kfree(hctx);
+	}
 
 	kfree(q->queue_hw_ctx);
 
@@ -2047,7 +2086,7 @@ void blk_mq_free_queue(struct request_queue *q)
 /* Basically redo blk_mq_init_queue with queue frozen */
 static void blk_mq_queue_reinit(struct request_queue *q)
 {
-	WARN_ON_ONCE(!q->mq_freeze_depth);
+	WARN_ON_ONCE(!atomic_read(&q->mq_freeze_depth));
 
 	blk_mq_sysfs_unregister(q);
 
@@ -2090,9 +2129,16 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb,
 	 */
 	list_for_each_entry(q, &all_q_list, all_q_node)
 		blk_mq_freeze_queue_start(q);
-	list_for_each_entry(q, &all_q_list, all_q_node)
+	list_for_each_entry(q, &all_q_list, all_q_node) {
 		blk_mq_freeze_queue_wait(q);
 
+		/*
+		 * timeout handler can't touch hw queue during the
+		 * reinitialization
+		 */
+		del_timer_sync(&q->timeout);
+	}
+
 	list_for_each_entry(q, &all_q_list, all_q_node)
 		blk_mq_queue_reinit(q);
 
@@ -2157,6 +2203,12 @@ static int blk_mq_alloc_rq_maps(struct blk_mq_tag_set *set)
 	return 0;
 }
 
+struct cpumask *blk_mq_tags_cpumask(struct blk_mq_tags *tags)
+{
+	return tags->cpumask;
+}
+EXPORT_SYMBOL_GPL(blk_mq_tags_cpumask);
+
 /*
  * Alloc a tag set to be associated with one or more request queues.
  * May fail with EINVAL for various error conditions. May adjust the
@@ -2218,8 +2270,10 @@ void blk_mq_free_tag_set(struct blk_mq_tag_set *set)
 	int i;
 
 	for (i = 0; i < set->nr_hw_queues; i++) {
-		if (set->tags[i])
+		if (set->tags[i]) {
 			blk_mq_free_rq_map(set, set->tags[i], i);
+			free_cpumask_var(set->tags[i]->cpumask);
+		}
 	}
 
 	kfree(set->tags);