1 files changed, 170 insertions, 235 deletions

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 13f5a6c1de76..ddc2eed64771 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -22,6 +22,7 @@
 #include <linux/sched/sysctl.h>
 #include <linux/delay.h>
 #include <linux/crash_dump.h>
+#include <linux/prefetch.h>
 
 #include <trace/events/block.h>
 
@@ -33,49 +34,28 @@
 static DEFINE_MUTEX(all_q_mutex);
 static LIST_HEAD(all_q_list);
 
-static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx);
-
 /*
  * Check if any of the ctx's have pending work in this hardware queue
  */
 static bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
 {
-	unsigned int i;
-
-	for (i = 0; i < hctx->ctx_map.size; i++)
-		if (hctx->ctx_map.map[i].word)
-			return true;
-
-	return false;
+	return sbitmap_any_bit_set(&hctx->ctx_map);
 }
 
-static inline struct blk_align_bitmap *get_bm(struct blk_mq_hw_ctx *hctx,
-					      struct blk_mq_ctx *ctx)
-{
-	return &hctx->ctx_map.map[ctx->index_hw / hctx->ctx_map.bits_per_word];
-}
-
-#define CTX_TO_BIT(hctx, ctx)	\
-	((ctx)->index_hw & ((hctx)->ctx_map.bits_per_word - 1))
-
 /*
  * Mark this ctx as having pending work in this hardware queue
  */
 static void blk_mq_hctx_mark_pending(struct blk_mq_hw_ctx *hctx,
 				     struct blk_mq_ctx *ctx)
 {
-	struct blk_align_bitmap *bm = get_bm(hctx, ctx);
-
-	if (!test_bit(CTX_TO_BIT(hctx, ctx), &bm->word))
-		set_bit(CTX_TO_BIT(hctx, ctx), &bm->word);
+	if (!sbitmap_test_bit(&hctx->ctx_map, ctx->index_hw))
+		sbitmap_set_bit(&hctx->ctx_map, ctx->index_hw);
 }
 
 static void blk_mq_hctx_clear_pending(struct blk_mq_hw_ctx *hctx,
 				      struct blk_mq_ctx *ctx)
 {
-	struct blk_align_bitmap *bm = get_bm(hctx, ctx);
-
-	clear_bit(CTX_TO_BIT(hctx, ctx), &bm->word);
+	sbitmap_clear_bit(&hctx->ctx_map, ctx->index_hw);
 }
 
 void blk_mq_freeze_queue_start(struct request_queue *q)
@@ -244,21 +224,11 @@ struct request *blk_mq_alloc_request(struct request_queue *q, int rw,
 		return ERR_PTR(ret);
 
 	ctx = blk_mq_get_ctx(q);
-	hctx = q->mq_ops->map_queue(q, ctx->cpu);
+	hctx = blk_mq_map_queue(q, ctx->cpu);
 	blk_mq_set_alloc_data(&alloc_data, q, flags, ctx, hctx);
-
 	rq = __blk_mq_alloc_request(&alloc_data, rw, 0);
-	if (!rq && !(flags & BLK_MQ_REQ_NOWAIT)) {
-		__blk_mq_run_hw_queue(hctx);
-		blk_mq_put_ctx(ctx);
-
-		ctx = blk_mq_get_ctx(q);
-		hctx = q->mq_ops->map_queue(q, ctx->cpu);
-		blk_mq_set_alloc_data(&alloc_data, q, flags, ctx, hctx);
-		rq =  __blk_mq_alloc_request(&alloc_data, rw, 0);
-		ctx = alloc_data.ctx;
-	}
 	blk_mq_put_ctx(ctx);
+
 	if (!rq) {
 		blk_queue_exit(q);
 		return ERR_PTR(-EWOULDBLOCK);
@@ -296,17 +266,29 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q, int rw,
 	if (ret)
 		return ERR_PTR(ret);
 
+	/*
+	 * Check if the hardware context is actually mapped to anything.
+	 * If not tell the caller that it should skip this queue.
+	 */
 	hctx = q->queue_hw_ctx[hctx_idx];
+	if (!blk_mq_hw_queue_mapped(hctx)) {
+		ret = -EXDEV;
+		goto out_queue_exit;
+	}
 	ctx = __blk_mq_get_ctx(q, cpumask_first(hctx->cpumask));
 
 	blk_mq_set_alloc_data(&alloc_data, q, flags, ctx, hctx);
 	rq = __blk_mq_alloc_request(&alloc_data, rw, 0);
 	if (!rq) {
-		blk_queue_exit(q);
-		return ERR_PTR(-EWOULDBLOCK);
+		ret = -EWOULDBLOCK;
+		goto out_queue_exit;
 	}
 
 	return rq;
+
+out_queue_exit:
+	blk_queue_exit(q);
+	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL_GPL(blk_mq_alloc_request_hctx);
 
@@ -321,7 +303,7 @@ static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx,
 	rq->cmd_flags = 0;
 
 	clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
-	blk_mq_put_tag(hctx, tag, &ctx->last_tag);
+	blk_mq_put_tag(hctx, ctx, tag);
 	blk_queue_exit(q);
 }
 
@@ -337,11 +319,7 @@ EXPORT_SYMBOL_GPL(blk_mq_free_hctx_request);
 
 void blk_mq_free_request(struct request *rq)
 {
-	struct blk_mq_hw_ctx *hctx;
-	struct request_queue *q = rq->q;
-
-	hctx = q->mq_ops->map_queue(q, rq->mq_ctx->cpu);
-	blk_mq_free_hctx_request(hctx, rq);
+	blk_mq_free_hctx_request(blk_mq_map_queue(rq->q, rq->mq_ctx->cpu), rq);
 }
 EXPORT_SYMBOL_GPL(blk_mq_free_request);
 
@@ -501,7 +479,7 @@ EXPORT_SYMBOL(blk_mq_requeue_request);
 static void blk_mq_requeue_work(struct work_struct *work)
 {
 	struct request_queue *q =
-		container_of(work, struct request_queue, requeue_work);
+		container_of(work, struct request_queue, requeue_work.work);
 	LIST_HEAD(rq_list);
 	struct request *rq, *next;
 	unsigned long flags;
@@ -556,16 +534,24 @@ EXPORT_SYMBOL(blk_mq_add_to_requeue_list);
 
 void blk_mq_cancel_requeue_work(struct request_queue *q)
 {
-	cancel_work_sync(&q->requeue_work);
+	cancel_delayed_work_sync(&q->requeue_work);
 }
 EXPORT_SYMBOL_GPL(blk_mq_cancel_requeue_work);
 
 void blk_mq_kick_requeue_list(struct request_queue *q)
 {
-	kblockd_schedule_work(&q->requeue_work);
+	kblockd_schedule_delayed_work(&q->requeue_work, 0);
 }
 EXPORT_SYMBOL(blk_mq_kick_requeue_list);
 
+void blk_mq_delay_kick_requeue_list(struct request_queue *q,
+				    unsigned long msecs)
+{
+	kblockd_schedule_delayed_work(&q->requeue_work,
+				      msecs_to_jiffies(msecs));
+}
+EXPORT_SYMBOL(blk_mq_delay_kick_requeue_list);
+
 void blk_mq_abort_requeue_list(struct request_queue *q)
 {
 	unsigned long flags;
@@ -588,8 +574,10 @@ EXPORT_SYMBOL(blk_mq_abort_requeue_list);
 
 struct request *blk_mq_tag_to_rq(struct blk_mq_tags *tags, unsigned int tag)
 {
-	if (tag < tags->nr_tags)
+	if (tag < tags->nr_tags) {
+		prefetch(tags->rqs[tag]);
 		return tags->rqs[tag];
+	}
 
 	return NULL;
 }
@@ -744,38 +732,44 @@ static bool blk_mq_attempt_merge(struct request_queue *q,
 	return false;
 }
 
+struct flush_busy_ctx_data {
+	struct blk_mq_hw_ctx *hctx;
+	struct list_head *list;
+};
+
+static bool flush_busy_ctx(struct sbitmap *sb, unsigned int bitnr, void *data)
+{
+	struct flush_busy_ctx_data *flush_data = data;
+	struct blk_mq_hw_ctx *hctx = flush_data->hctx;
+	struct blk_mq_ctx *ctx = hctx->ctxs[bitnr];
+
+	sbitmap_clear_bit(sb, bitnr);
+	spin_lock(&ctx->lock);
+	list_splice_tail_init(&ctx->rq_list, flush_data->list);
+	spin_unlock(&ctx->lock);
+	return true;
+}
+
 /*
  * Process software queues that have been marked busy, splicing them
  * to the for-dispatch
  */
 static void flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list)
 {
-	struct blk_mq_ctx *ctx;
-	int i;
-
-	for (i = 0; i < hctx->ctx_map.size; i++) {
-		struct blk_align_bitmap *bm = &hctx->ctx_map.map[i];
-		unsigned int off, bit;
-
-		if (!bm->word)
-			continue;
+	struct flush_busy_ctx_data data = {
+		.hctx = hctx,
+		.list = list,
+	};
 
-		bit = 0;
-		off = i * hctx->ctx_map.bits_per_word;
-		do {
-			bit = find_next_bit(&bm->word, bm->depth, bit);
-			if (bit >= bm->depth)
-				break;
+	sbitmap_for_each_set(&hctx->ctx_map, flush_busy_ctx, &data);
+}
 
-			ctx = hctx->ctxs[bit + off];
-			clear_bit(bit, &bm->word);
-			spin_lock(&ctx->lock);
-			list_splice_tail_init(&ctx->rq_list, list);
-			spin_unlock(&ctx->lock);
+static inline unsigned int queued_to_index(unsigned int queued)
+{
+	if (!queued)
+		return 0;
 
-			bit++;
-		} while (1);
-	}
+	return min(BLK_MQ_MAX_DISPATCH_ORDER - 1, ilog2(queued) + 1);
 }
 
 /*
@@ -866,10 +860,7 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 			dptr = &driver_list;
 	}
 
-	if (!queued)
-		hctx->dispatched[0]++;
-	else if (queued < (1 << (BLK_MQ_MAX_DISPATCH_ORDER - 1)))
-		hctx->dispatched[ilog2(queued) + 1]++;
+	hctx->dispatched[queued_to_index(queued)]++;
 
 	/*
 	 * Any items that need requeuing? Stuff them into hctx->dispatch,
@@ -925,7 +916,7 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
 	    !blk_mq_hw_queue_mapped(hctx)))
 		return;
 
-	if (!async) {
+	if (!async && !(hctx->flags & BLK_MQ_F_BLOCKING)) {
 		int cpu = get_cpu();
 		if (cpumask_test_cpu(cpu, hctx->cpumask)) {
 			__blk_mq_run_hw_queue(hctx);
@@ -936,8 +927,7 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
 		put_cpu();
 	}
 
-	kblockd_schedule_delayed_work_on(blk_mq_hctx_next_cpu(hctx),
-			&hctx->run_work, 0);
+	kblockd_schedule_work_on(blk_mq_hctx_next_cpu(hctx), &hctx->run_work);
 }
 
 void blk_mq_run_hw_queues(struct request_queue *q, bool async)
@@ -958,7 +948,7 @@ EXPORT_SYMBOL(blk_mq_run_hw_queues);
 
 void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx)
 {
-	cancel_delayed_work(&hctx->run_work);
+	cancel_work(&hctx->run_work);
 	cancel_delayed_work(&hctx->delay_work);
 	set_bit(BLK_MQ_S_STOPPED, &hctx->state);
 }
@@ -1011,7 +1001,7 @@ static void blk_mq_run_work_fn(struct work_struct *work)
 {
 	struct blk_mq_hw_ctx *hctx;
 
-	hctx = container_of(work, struct blk_mq_hw_ctx, run_work.work);
+	hctx = container_of(work, struct blk_mq_hw_ctx, run_work);
 
 	__blk_mq_run_hw_queue(hctx);
 }
@@ -1064,9 +1054,7 @@ void blk_mq_insert_request(struct request *rq, bool at_head, bool run_queue,
 {
 	struct blk_mq_ctx *ctx = rq->mq_ctx;
 	struct request_queue *q = rq->q;
-	struct blk_mq_hw_ctx *hctx;
-
-	hctx = q->mq_ops->map_queue(q, ctx->cpu);
+	struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, ctx->cpu);
 
 	spin_lock(&ctx->lock);
 	__blk_mq_insert_request(hctx, rq, at_head);
@@ -1083,12 +1071,10 @@ static void blk_mq_insert_requests(struct request_queue *q,
 				     bool from_schedule)
 
 {
-	struct blk_mq_hw_ctx *hctx;
+	struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, ctx->cpu);
 
 	trace_block_unplug(q, depth, !from_schedule);
 
-	hctx = q->mq_ops->map_queue(q, ctx->cpu);
-
 	/*
 	 * preemption doesn't flush plug list, so it's possible ctx->cpu is
 	 * offline now
@@ -1222,26 +1208,14 @@ static struct request *blk_mq_map_request(struct request_queue *q,
 
 	blk_queue_enter_live(q);
 	ctx = blk_mq_get_ctx(q);
-	hctx = q->mq_ops->map_queue(q, ctx->cpu);
+	hctx = blk_mq_map_queue(q, ctx->cpu);
 
 	if (rw_is_sync(bio_op(bio), bio->bi_opf))
 		op_flags |= REQ_SYNC;
 
 	trace_block_getrq(q, bio, op);
-	blk_mq_set_alloc_data(&alloc_data, q, BLK_MQ_REQ_NOWAIT, ctx, hctx);
+	blk_mq_set_alloc_data(&alloc_data, q, 0, ctx, hctx);
 	rq = __blk_mq_alloc_request(&alloc_data, op, op_flags);
-	if (unlikely(!rq)) {
-		__blk_mq_run_hw_queue(hctx);
-		blk_mq_put_ctx(ctx);
-		trace_block_sleeprq(q, bio, op);
-
-		ctx = blk_mq_get_ctx(q);
-		hctx = q->mq_ops->map_queue(q, ctx->cpu);
-		blk_mq_set_alloc_data(&alloc_data, q, 0, ctx, hctx);
-		rq = __blk_mq_alloc_request(&alloc_data, op, op_flags);
-		ctx = alloc_data.ctx;
-		hctx = alloc_data.hctx;
-	}
 
 	hctx->queued++;
 	data->hctx = hctx;
@@ -1253,8 +1227,7 @@ static int blk_mq_direct_issue_request(struct request *rq, blk_qc_t *cookie)
 {
 	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_hw_ctx *hctx = blk_mq_map_queue(q, rq->mq_ctx->cpu);
 	struct blk_mq_queue_data bd = {
 		.rq = rq,
 		.list = NULL,
@@ -1458,15 +1431,6 @@ run_queue:
 	return cookie;
 }
 
-/*
- * Default mapping to a software queue, since we use one per CPU.
- */
-struct blk_mq_hw_ctx *blk_mq_map_queue(struct request_queue *q, const int cpu)
-{
-	return q->queue_hw_ctx[q->mq_map[cpu]];
-}
-EXPORT_SYMBOL(blk_mq_map_queue);
-
 static void blk_mq_free_rq_map(struct blk_mq_tag_set *set,
 		struct blk_mq_tags *tags, unsigned int hctx_idx)
 {
@@ -1594,42 +1558,18 @@ fail:
 	return NULL;
 }
 
-static void blk_mq_free_bitmap(struct blk_mq_ctxmap *bitmap)
-{
-	kfree(bitmap->map);
-}
-
-static int blk_mq_alloc_bitmap(struct blk_mq_ctxmap *bitmap, int node)
-{
-	unsigned int bpw = 8, total, num_maps, i;
-
-	bitmap->bits_per_word = bpw;
-
-	num_maps = ALIGN(nr_cpu_ids, bpw) / bpw;
-	bitmap->map = kzalloc_node(num_maps * sizeof(struct blk_align_bitmap),
-					GFP_KERNEL, node);
-	if (!bitmap->map)
-		return -ENOMEM;
-
-	total = nr_cpu_ids;
-	for (i = 0; i < num_maps; i++) {
-		bitmap->map[i].depth = min(total, bitmap->bits_per_word);
-		total -= bitmap->map[i].depth;
-	}
-
-	return 0;
-}
-
 /*
  * 'cpu' is going away. splice any existing rq_list entries from this
  * software queue to the hw queue dispatch list, and ensure that it
  * gets run.
  */
-static int blk_mq_hctx_cpu_offline(struct blk_mq_hw_ctx *hctx, int cpu)
+static int blk_mq_hctx_notify_dead(unsigned int cpu, struct hlist_node *node)
 {
+	struct blk_mq_hw_ctx *hctx;
 	struct blk_mq_ctx *ctx;
 	LIST_HEAD(tmp);
 
+	hctx = hlist_entry_safe(node, struct blk_mq_hw_ctx, cpuhp_dead);
 	ctx = __blk_mq_get_ctx(hctx->queue, cpu);
 
 	spin_lock(&ctx->lock);
@@ -1640,30 +1580,20 @@ static int blk_mq_hctx_cpu_offline(struct blk_mq_hw_ctx *hctx, int cpu)
 	spin_unlock(&ctx->lock);
 
 	if (list_empty(&tmp))
-		return NOTIFY_OK;
+		return 0;
 
 	spin_lock(&hctx->lock);
 	list_splice_tail_init(&tmp, &hctx->dispatch);
 	spin_unlock(&hctx->lock);
 
 	blk_mq_run_hw_queue(hctx, true);
-	return NOTIFY_OK;
+	return 0;
 }
 
-static int blk_mq_hctx_notify(void *data, unsigned long action,
-			      unsigned int cpu)
+static void blk_mq_remove_cpuhp(struct blk_mq_hw_ctx *hctx)
 {
-	struct blk_mq_hw_ctx *hctx = data;
-
-	if (action == CPU_DEAD || action == CPU_DEAD_FROZEN)
-		return blk_mq_hctx_cpu_offline(hctx, cpu);
-
-	/*
-	 * In case of CPU online, tags may be reallocated
-	 * in blk_mq_map_swqueue() after mapping is updated.
-	 */
-
-	return NOTIFY_OK;
+	cpuhp_state_remove_instance_nocalls(CPUHP_BLK_MQ_DEAD,
+					    &hctx->cpuhp_dead);
 }
 
 /* hctx->ctxs will be freed in queue's release handler */
@@ -1683,9 +1613,9 @@ static void blk_mq_exit_hctx(struct request_queue *q,
 	if (set->ops->exit_hctx)
 		set->ops->exit_hctx(hctx, hctx_idx);
 
-	blk_mq_unregister_cpu_notifier(&hctx->cpu_notifier);
+	blk_mq_remove_cpuhp(hctx);
 	blk_free_flush_queue(hctx->fq);
-	blk_mq_free_bitmap(&hctx->ctx_map);
+	sbitmap_free(&hctx->ctx_map);
 }
 
 static void blk_mq_exit_hw_queues(struct request_queue *q,
@@ -1722,7 +1652,7 @@ static int blk_mq_init_hctx(struct request_queue *q,
 	if (node == NUMA_NO_NODE)
 		node = hctx->numa_node = set->numa_node;
 
-	INIT_DELAYED_WORK(&hctx->run_work, blk_mq_run_work_fn);
+	INIT_WORK(&hctx->run_work, blk_mq_run_work_fn);
 	INIT_DELAYED_WORK(&hctx->delay_work, blk_mq_delay_work_fn);
 	spin_lock_init(&hctx->lock);
 	INIT_LIST_HEAD(&hctx->dispatch);
@@ -1730,9 +1660,7 @@ static int blk_mq_init_hctx(struct request_queue *q,
 	hctx->queue_num = hctx_idx;
 	hctx->flags = set->flags & ~BLK_MQ_F_TAG_SHARED;
 
-	blk_mq_init_cpu_notifier(&hctx->cpu_notifier,
-					blk_mq_hctx_notify, hctx);
-	blk_mq_register_cpu_notifier(&hctx->cpu_notifier);
+	cpuhp_state_add_instance_nocalls(CPUHP_BLK_MQ_DEAD, &hctx->cpuhp_dead);
 
 	hctx->tags = set->tags[hctx_idx];
 
@@ -1745,7 +1673,8 @@ static int blk_mq_init_hctx(struct request_queue *q,
 	if (!hctx->ctxs)
 		goto unregister_cpu_notifier;
 
-	if (blk_mq_alloc_bitmap(&hctx->ctx_map, node))
+	if (sbitmap_init_node(&hctx->ctx_map, nr_cpu_ids, ilog2(8), GFP_KERNEL,
+			      node))
 		goto free_ctxs;
 
 	hctx->nr_ctx = 0;
@@ -1772,12 +1701,11 @@ static int blk_mq_init_hctx(struct request_queue *q,
 	if (set->ops->exit_hctx)
 		set->ops->exit_hctx(hctx, hctx_idx);
  free_bitmap:
-	blk_mq_free_bitmap(&hctx->ctx_map);
+	sbitmap_free(&hctx->ctx_map);
  free_ctxs:
 	kfree(hctx->ctxs);
  unregister_cpu_notifier:
-	blk_mq_unregister_cpu_notifier(&hctx->cpu_notifier);
-
+	blk_mq_remove_cpuhp(hctx);
 	return -1;
 }
 
@@ -1800,7 +1728,7 @@ static void blk_mq_init_cpu_queues(struct request_queue *q,
 		if (!cpu_online(i))
 			continue;
 
-		hctx = q->mq_ops->map_queue(q, i);
+		hctx = blk_mq_map_queue(q, i);
 
 		/*
 		 * Set local node, IFF we have more than one hw queue. If
@@ -1838,7 +1766,7 @@ static void blk_mq_map_swqueue(struct request_queue *q,
 			continue;
 
 		ctx = per_cpu_ptr(q->queue_ctx, i);
-		hctx = q->mq_ops->map_queue(q, i);
+		hctx = blk_mq_map_queue(q, i);
 
 		cpumask_set_cpu(i, hctx->cpumask);
 		ctx->index_hw = hctx->nr_ctx;
@@ -1848,8 +1776,6 @@ static void blk_mq_map_swqueue(struct request_queue *q,
 	mutex_unlock(&q->sysfs_lock);
 
 	queue_for_each_hw_ctx(q, hctx, i) {
-		struct blk_mq_ctxmap *map = &hctx->ctx_map;
-
 		/*
 		 * If no software queues are mapped to this hardware queue,
 		 * disable it and free the request entries.
@@ -1869,13 +1795,12 @@ static void blk_mq_map_swqueue(struct request_queue *q,
 		hctx->tags = set->tags[i];
 		WARN_ON(!hctx->tags);
 
-		cpumask_copy(hctx->tags->cpumask, hctx->cpumask);
 		/*
 		 * Set the map size to the number of mapped software queues.
 		 * This is more accurate and more efficient than looping
 		 * over all possibly mapped software queues.
 		 */
-		map->size = DIV_ROUND_UP(hctx->nr_ctx, map->bits_per_word);
+		sbitmap_resize(&hctx->ctx_map, hctx->nr_ctx);
 
 		/*
 		 * Initialize batch roundrobin counts
@@ -1963,7 +1888,6 @@ void blk_mq_release(struct request_queue *q)
 		kfree(hctx);
 	}
 
-	kfree(q->mq_map);
 	q->mq_map = NULL;
 
 	kfree(q->queue_hw_ctx);
@@ -2062,9 +1986,7 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	if (!q->queue_hw_ctx)
 		goto err_percpu;
 
-	q->mq_map = blk_mq_make_queue_map(set);
-	if (!q->mq_map)
-		goto err_map;
+	q->mq_map = set->mq_map;
 
 	blk_mq_realloc_hw_ctxs(set, q);
 	if (!q->nr_hw_queues)
@@ -2082,7 +2004,7 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 
 	q->sg_reserved_size = INT_MAX;
 
-	INIT_WORK(&q->requeue_work, blk_mq_requeue_work);
+	INIT_DELAYED_WORK(&q->requeue_work, blk_mq_requeue_work);
 	INIT_LIST_HEAD(&q->requeue_list);
 	spin_lock_init(&q->requeue_lock);
 
@@ -2114,8 +2036,6 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	return q;
 
 err_hctxs:
-	kfree(q->mq_map);
-err_map:
 	kfree(q->queue_hw_ctx);
 err_percpu:
 	free_percpu(q->queue_ctx);
@@ -2147,8 +2067,6 @@ static void blk_mq_queue_reinit(struct request_queue *q,
 
 	blk_mq_sysfs_unregister(q);
 
-	blk_mq_update_queue_map(q->mq_map, q->nr_hw_queues, online_mask);
-
 	/*
 	 * redo blk_mq_init_cpu_queues and blk_mq_init_hw_queues. FIXME: maybe
 	 * we should change hctx numa_node according to new topology (this
@@ -2160,50 +2078,18 @@ static void blk_mq_queue_reinit(struct request_queue *q,
 	blk_mq_sysfs_register(q);
 }
 
-static int blk_mq_queue_reinit_notify(struct notifier_block *nb,
-				      unsigned long action, void *hcpu)
+/*
+ * New online cpumask which is going to be set in this hotplug event.
+ * Declare this cpumasks as global as cpu-hotplug operation is invoked
+ * one-by-one and dynamically allocating this could result in a failure.
+ */
+static struct cpumask cpuhp_online_new;
+
+static void blk_mq_queue_reinit_work(void)
 {
 	struct request_queue *q;
-	int cpu = (unsigned long)hcpu;
-	/*
-	 * New online cpumask which is going to be set in this hotplug event.
-	 * Declare this cpumasks as global as cpu-hotplug operation is invoked
-	 * one-by-one and dynamically allocating this could result in a failure.
-	 */
-	static struct cpumask online_new;
-
-	/*
-	 * Before hotadded cpu starts handling requests, new mappings must
-	 * be established.  Otherwise, these requests in hw queue might
-	 * never be dispatched.
-	 *
-	 * For example, there is a single hw queue (hctx) and two CPU queues
-	 * (ctx0 for CPU0, and ctx1 for CPU1).
-	 *
-	 * Now CPU1 is just onlined and a request is inserted into
-	 * ctx1->rq_list and set bit0 in pending bitmap as ctx1->index_hw is
-	 * still zero.
-	 *
-	 * And then while running hw queue, flush_busy_ctxs() finds bit0 is
-	 * set in pending bitmap and tries to retrieve requests in
-	 * hctx->ctxs[0]->rq_list.  But htx->ctxs[0] is a pointer to ctx0,
-	 * so the request in ctx1->rq_list is ignored.
-	 */
-	switch (action & ~CPU_TASKS_FROZEN) {
-	case CPU_DEAD:
-	case CPU_UP_CANCELED:
-		cpumask_copy(&online_new, cpu_online_mask);
-		break;
-	case CPU_UP_PREPARE:
-		cpumask_copy(&online_new, cpu_online_mask);
-		cpumask_set_cpu(cpu, &online_new);
-		break;
-	default:
-		return NOTIFY_OK;
-	}
 
 	mutex_lock(&all_q_mutex);
-
 	/*
 	 * We need to freeze and reinit all existing queues.  Freezing
 	 * involves synchronous wait for an RCU grace period and doing it
@@ -2224,13 +2110,43 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb,
 	}
 
 	list_for_each_entry(q, &all_q_list, all_q_node)
-		blk_mq_queue_reinit(q, &online_new);
+		blk_mq_queue_reinit(q, &cpuhp_online_new);
 
 	list_for_each_entry(q, &all_q_list, all_q_node)
 		blk_mq_unfreeze_queue(q);
 
 	mutex_unlock(&all_q_mutex);
-	return NOTIFY_OK;
+}
+
+static int blk_mq_queue_reinit_dead(unsigned int cpu)
+{
+	cpumask_copy(&cpuhp_online_new, cpu_online_mask);
+	blk_mq_queue_reinit_work();
+	return 0;
+}
+
+/*
+ * Before hotadded cpu starts handling requests, new mappings must be
+ * established.  Otherwise, these requests in hw queue might never be
+ * dispatched.
+ *
+ * For example, there is a single hw queue (hctx) and two CPU queues (ctx0
+ * for CPU0, and ctx1 for CPU1).
+ *
+ * Now CPU1 is just onlined and a request is inserted into ctx1->rq_list
+ * and set bit0 in pending bitmap as ctx1->index_hw is still zero.
+ *
+ * And then while running hw queue, flush_busy_ctxs() finds bit0 is set in
+ * pending bitmap and tries to retrieve requests in hctx->ctxs[0]->rq_list.
+ * But htx->ctxs[0] is a pointer to ctx0, so the request in ctx1->rq_list
+ * is ignored.
+ */
+static int blk_mq_queue_reinit_prepare(unsigned int cpu)
+{
+	cpumask_copy(&cpuhp_online_new, cpu_online_mask);
+	cpumask_set_cpu(cpu, &cpuhp_online_new);
+	blk_mq_queue_reinit_work();
+	return 0;
 }
 
 static int __blk_mq_alloc_rq_maps(struct blk_mq_tag_set *set)
@@ -2287,12 +2203,6 @@ 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
@@ -2301,6 +2211,8 @@ EXPORT_SYMBOL_GPL(blk_mq_tags_cpumask);
  */
 int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
 {
+	int ret;
+
 	BUILD_BUG_ON(BLK_MQ_MAX_DEPTH > 1 << BLK_MQ_UNIQUE_TAG_BITS);
 
 	if (!set->nr_hw_queues)
@@ -2310,7 +2222,7 @@ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
 	if (set->queue_depth < set->reserved_tags + BLK_MQ_TAG_MIN)
 		return -EINVAL;
 
-	if (!set->ops->queue_rq || !set->ops->map_queue)
+	if (!set->ops->queue_rq)
 		return -EINVAL;
 
 	if (set->queue_depth > BLK_MQ_MAX_DEPTH) {
@@ -2339,17 +2251,35 @@ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
 	if (!set->tags)
 		return -ENOMEM;
 
-	if (blk_mq_alloc_rq_maps(set))
-		goto enomem;
+	ret = -ENOMEM;
+	set->mq_map = kzalloc_node(sizeof(*set->mq_map) * nr_cpu_ids,
+			GFP_KERNEL, set->numa_node);
+	if (!set->mq_map)
+		goto out_free_tags;
+
+	if (set->ops->map_queues)
+		ret = set->ops->map_queues(set);
+	else
+		ret = blk_mq_map_queues(set);
+	if (ret)
+		goto out_free_mq_map;
+
+	ret = blk_mq_alloc_rq_maps(set);
+	if (ret)
+		goto out_free_mq_map;
 
 	mutex_init(&set->tag_list_lock);
 	INIT_LIST_HEAD(&set->tag_list);
 
 	return 0;
-enomem:
+
+out_free_mq_map:
+	kfree(set->mq_map);
+	set->mq_map = NULL;
+out_free_tags:
 	kfree(set->tags);
 	set->tags = NULL;
-	return -ENOMEM;
+	return ret;
 }
 EXPORT_SYMBOL(blk_mq_alloc_tag_set);
 
@@ -2362,6 +2292,9 @@ void blk_mq_free_tag_set(struct blk_mq_tag_set *set)
 			blk_mq_free_rq_map(set, set->tags[i], i);
 	}
 
+	kfree(set->mq_map);
+	set->mq_map = NULL;
+
 	kfree(set->tags);
 	set->tags = NULL;
 }
@@ -2432,10 +2365,12 @@ void blk_mq_enable_hotplug(void)
 
 static int __init blk_mq_init(void)
 {
-	blk_mq_cpu_init();
-
-	hotcpu_notifier(blk_mq_queue_reinit_notify, 0);
+	cpuhp_setup_state_multi(CPUHP_BLK_MQ_DEAD, "block/mq:dead", NULL,
+				blk_mq_hctx_notify_dead);
 
+	cpuhp_setup_state_nocalls(CPUHP_BLK_MQ_PREPARE, "block/mq:prepare",
+				  blk_mq_queue_reinit_prepare,
+				  blk_mq_queue_reinit_dead);
 	return 0;
 }
 subsys_initcall(blk_mq_init);