block-6.17-20250808

-----BEGIN PGP SIGNATURE-----
 
 iQJEBAABCAAuFiEEwPw5LcreJtl1+l5K99NY+ylx4KYFAmiWLjoQHGF4Ym9lQGtl
 cm5lbC5kawAKCRD301j7KXHgpvveD/9vbvp3XaF0LagRJLH0fcdhcxL7Z+IHD+7U
 v5vICMeoeBhhhOtPJ0y+h/9LMLQWFYDFl6drkY0atSSxp/CK6CB25qFhIDsoA6Qk
 RBM/qZ64z4Uxvlc+VQmCqI2EMc/ZrYtrcr7jsornwORoTSEKXVHdyO5k7Q9002Sw
 XNWc0bZKIibFlgOk12Wnd8ZS5RWHw1uViUcreojcGVZAVR+BuHNGGoa3xq0bLiHU
 ERbQXfjaN28R+eo4E1euCtdf++7tW2kFjClrDmLcszdb27E2+MWMA6AKMiSTBE2k
 2e2TvJUcGZs1s8atqSIIjBtmwQW3rKws33zODLMONzOP8CIErcaniHxyDSaxJIJr
 kjsdKnwlziL3xVnwQcpgnVOPvvDSKZ4OKEqx8rAuYTqiknpz3uhbt/7EqumuPLHr
 e7Rz0MnFolrVN7KZOHQ5CPJIezkEAOAEpItLdfc5cfLS06pbeTN3j+dJZp+tUohi
 WP/K3l2N3C5pkXA0ilAzshRF20Rwv/09M85BoqWocTLBJY7WqyIKXywCNdX81wkv
 tpbQvp2MpPkJXUIbAh5484BOfCfx9vkYVm2cam2UxXJhR6VfrQCjYfXIjfpqF4jp
 q7xxNesUezrOqB2Q/cKxw8dKOaRtO1XzVnmwutBrcKgqqLezMwUTDDjQYe8l6p1Z
 40E74tsJwQ==
 =EQ7g
 -----END PGP SIGNATURE-----

Merge tag 'block-6.17-20250808' of git://git.kernel.dk/linux

Pull more block updates from Jens Axboe:

 - MD pull request via Yu:
      - mddev null-ptr-dereference fix, by Erkun
      - md-cluster fail to remove the faulty disk regression fix, by
        Heming
      - minor cleanup, by Li Nan and Jinchao
      - mdadm lifetime regression fix reported by syzkaller, by Yu Kuai

 - MD pull request via Christoph
      - add support for getting the FDP featuee in fabrics passthru path
        (Nitesh Shetty)
      - add capability to connect to an administrative controller
        (Kamaljit Singh)
      - fix a leak on sgl setup error (Keith Busch)
      - initialize discovery subsys after debugfs is initialized
        (Mohamed Khalfella)
      - fix various comment typos (Bjorn Helgaas)
      - remove unneeded semicolons (Jiapeng Chong)

 - nvmet debugfs ordering issue fix

 - Fix UAF in the tag_set in zloop

 - Ensure sbitmap shallow depth covers entire set

 - Reduce lock roundtrips in io context lookup

 - Move scheduler tags alloc/free out of elevator and freeze lock, to
   fix some lockdep found issues

 - Improve robustness of queue limits checking

 - Fix a regression with IO priorities, if no io context exists

* tag 'block-6.17-20250808' of git://git.kernel.dk/linux: (26 commits)
  lib/sbitmap: make sbitmap_get_shallow() internal
  lib/sbitmap: convert shallow_depth from one word to the whole sbitmap
  nvmet: exit debugfs after discovery subsystem exits
  block, bfq: Reorder struct bfq_iocq_bfqq_data
  md: make rdev_addable usable for rcu mode
  md/raid1: remove struct pool_info and related code
  md/raid1: change r1conf->r1bio_pool to a pointer type
  block: ensure discard_granularity is zero when discard is not supported
  zloop: fix KASAN use-after-free of tag set
  block: Fix default IO priority if there is no IO context
  nvme: fix various comment typos
  nvme-auth: remove unneeded semicolon
  nvme-pci: fix leak on sgl setup error
  nvmet: initialize discovery subsys after debugfs is initialized
  nvme: add capability to connect to an administrative controller
  nvmet: add support for FDP in fabrics passthru path
  md: rename recovery_cp to resync_offset
  md/md-cluster: handle REMOVE message earlier
  md: fix create on open mddev lifetime regression
  block: fix potential deadlock while running nr_hw_queue update
  ...

block/bfq-iosched.c

@@ -454,17 +454,10 @@ static struct bfq_io_cq *icq_to_bic(struct io_cq *icq)
  */
 static struct bfq_io_cq *bfq_bic_lookup(struct request_queue *q)
 {
-	struct bfq_io_cq *icq;
-	unsigned long flags;
-
 	if (!current->io_context)
 		return NULL;
 
-	spin_lock_irqsave(&q->queue_lock, flags);
+	return icq_to_bic(ioc_lookup_icq(q));
-	icq = icq_to_bic(ioc_lookup_icq(q));
-	spin_unlock_irqrestore(&q->queue_lock, flags);
-
-	return icq;
 }
 
 /*
@@ -701,17 +694,13 @@ static void bfq_limit_depth(blk_opf_t opf, struct blk_mq_alloc_data *data)
 {
 	struct bfq_data *bfqd = data->q->elevator->elevator_data;
 	struct bfq_io_cq *bic = bfq_bic_lookup(data->q);
-	int depth;
+	unsigned int limit, act_idx;
-	unsigned limit = data->q->nr_requests;
-	unsigned int act_idx;
 
 	/* Sync reads have full depth available */
-	if (op_is_sync(opf) && !op_is_write(opf)) {
+	if (op_is_sync(opf) && !op_is_write(opf))
-		depth = 0;
+		limit = data->q->nr_requests;
-	} else {
+	else
-		depth = bfqd->word_depths[!!bfqd->wr_busy_queues][op_is_sync(opf)];
+		limit = bfqd->async_depths[!!bfqd->wr_busy_queues][op_is_sync(opf)];
-		limit = (limit * depth) >> bfqd->full_depth_shift;
-	}
 
 	for (act_idx = 0; bic && act_idx < bfqd->num_actuators; act_idx++) {
 		/* Fast path to check if bfqq is already allocated. */
@@ -725,14 +714,16 @@ static void bfq_limit_depth(blk_opf_t opf, struct blk_mq_alloc_data *data)
 		 * available requests and thus starve other entities.
 		 */
 		if (bfqq_request_over_limit(bfqd, bic, opf, act_idx, limit)) {
-			depth = 1;
+			limit = 1;
 			break;
 		}
 	}
+
 	bfq_log(bfqd, "[%s] wr_busy %d sync %d depth %u",
-		__func__, bfqd->wr_busy_queues, op_is_sync(opf), depth);
+		__func__, bfqd->wr_busy_queues, op_is_sync(opf), limit);
-	if (depth)
+
-		data->shallow_depth = depth;
+	if (limit < data->q->nr_requests)
+		data->shallow_depth = limit;
 }
 
 static struct bfq_queue *
@@ -2457,15 +2448,8 @@ static bool bfq_bio_merge(struct request_queue *q, struct bio *bio,
 		unsigned int nr_segs)
 {
 	struct bfq_data *bfqd = q->elevator->elevator_data;
-	struct request *free = NULL;
-	/*
-	 * bfq_bic_lookup grabs the queue_lock: invoke it now and
-	 * store its return value for later use, to avoid nesting
-	 * queue_lock inside the bfqd->lock. We assume that the bic
-	 * returned by bfq_bic_lookup does not go away before
-	 * bfqd->lock is taken.
-	 */
 	struct bfq_io_cq *bic = bfq_bic_lookup(q);
+	struct request *free = NULL;
 	bool ret;
 
 	spin_lock_irq(&bfqd->lock);
@@ -7128,9 +7112,8 @@ void bfq_put_async_queues(struct bfq_data *bfqd, struct bfq_group *bfqg)
  */
 static void bfq_update_depths(struct bfq_data *bfqd, struct sbitmap_queue *bt)
 {
-	unsigned int depth = 1U << bt->sb.shift;
+	unsigned int nr_requests = bfqd->queue->nr_requests;
 
-	bfqd->full_depth_shift = bt->sb.shift;
 	/*
 	 * In-word depths if no bfq_queue is being weight-raised:
 	 * leaving 25% of tags only for sync reads.
@@ -7142,13 +7125,13 @@ static void bfq_update_depths(struct bfq_data *bfqd, struct sbitmap_queue *bt)
 	 * limit 'something'.
 	 */
 	/* no more than 50% of tags for async I/O */
-	bfqd->word_depths[0][0] = max(depth >> 1, 1U);
+	bfqd->async_depths[0][0] = max(nr_requests >> 1, 1U);
 	/*
 	 * no more than 75% of tags for sync writes (25% extra tags
 	 * w.r.t. async I/O, to prevent async I/O from starving sync
 	 * writes)
 	 */
-	bfqd->word_depths[0][1] = max((depth * 3) >> 2, 1U);
+	bfqd->async_depths[0][1] = max((nr_requests * 3) >> 2, 1U);
 
 	/*
 	 * In-word depths in case some bfq_queue is being weight-
@@ -7158,9 +7141,9 @@ static void bfq_update_depths(struct bfq_data *bfqd, struct sbitmap_queue *bt)
 	 * shortage.
 	 */
 	/* no more than ~18% of tags for async I/O */
-	bfqd->word_depths[1][0] = max((depth * 3) >> 4, 1U);
+	bfqd->async_depths[1][0] = max((nr_requests * 3) >> 4, 1U);
 	/* no more than ~37% of tags for sync writes (~20% extra tags) */
-	bfqd->word_depths[1][1] = max((depth * 6) >> 4, 1U);
+	bfqd->async_depths[1][1] = max((nr_requests * 6) >> 4, 1U);
 }
 
 static void bfq_depth_updated(struct blk_mq_hw_ctx *hctx)
@@ -7232,22 +7215,16 @@ static void bfq_init_root_group(struct bfq_group *root_group,
 	root_group->sched_data.bfq_class_idle_last_service = jiffies;
 }
 
-static int bfq_init_queue(struct request_queue *q, struct elevator_type *e)
+static int bfq_init_queue(struct request_queue *q, struct elevator_queue *eq)
 {
 	struct bfq_data *bfqd;
-	struct elevator_queue *eq;
 	unsigned int i;
 	struct blk_independent_access_ranges *ia_ranges = q->disk->ia_ranges;
 
-	eq = elevator_alloc(q, e);
+	bfqd = kzalloc_node(sizeof(*bfqd), GFP_KERNEL, q->node);
-	if (!eq)
+	if (!bfqd)
 		return -ENOMEM;
 
-	bfqd = kzalloc_node(sizeof(*bfqd), GFP_KERNEL, q->node);
-	if (!bfqd) {
-		kobject_put(&eq->kobj);
-		return -ENOMEM;
-	}
 	eq->elevator_data = bfqd;
 
 	spin_lock_irq(&q->queue_lock);
@@ -7405,7 +7382,6 @@ static int bfq_init_queue(struct request_queue *q, struct elevator_type *e)
 
 out_free:
 	kfree(bfqd);
-	kobject_put(&eq->kobj);
 	return -ENOMEM;
 }
 

block/bfq-iosched.h

@@ -427,9 +427,6 @@ struct bfq_iocq_bfqq_data {
 	 */
 	bool saved_IO_bound;
 
-	u64 saved_io_start_time;
-	u64 saved_tot_idle_time;
-
 	/*
 	 * Same purpose as the previous fields for the values of the
 	 * field keeping the queue's belonging to a large burst
@@ -450,6 +447,9 @@ struct bfq_iocq_bfqq_data {
 	 */
 	unsigned int saved_weight;
 
+	u64 saved_io_start_time;
+	u64 saved_tot_idle_time;
+
 	/*
 	 * Similar to previous fields: save wr information.
 	 */
@@ -457,13 +457,13 @@ struct bfq_iocq_bfqq_data {
 	unsigned long saved_last_wr_start_finish;
 	unsigned long saved_service_from_wr;
 	unsigned long saved_wr_start_at_switch_to_srt;
-	unsigned int saved_wr_cur_max_time;
 	struct bfq_ttime saved_ttime;
+	unsigned int saved_wr_cur_max_time;
 
 	/* Save also injection state */
-	u64 saved_last_serv_time_ns;
 	unsigned int saved_inject_limit;
 	unsigned long saved_decrease_time_jif;
+	u64 saved_last_serv_time_ns;
 
 	/* candidate queue for a stable merge (due to close creation time) */
 	struct bfq_queue *stable_merge_bfqq;
@@ -813,8 +813,7 @@ struct bfq_data {
 	 * Depth limits used in bfq_limit_depth (see comments on the
 	 * function)
 	 */
-	unsigned int word_depths[2][2];
+	unsigned int async_depths[2][2];
-	unsigned int full_depth_shift;
 
 	/*
 	 * Number of independent actuators. This is equal to 1 in

block/blk-ioc.c

@@ -308,24 +308,23 @@ int __copy_io(unsigned long clone_flags, struct task_struct *tsk)
 
 #ifdef CONFIG_BLK_ICQ
 /**
- * ioc_lookup_icq - lookup io_cq from ioc
+ * ioc_lookup_icq - lookup io_cq from ioc in io issue path
  * @q: the associated request_queue
  *
  * Look up io_cq associated with @ioc - @q pair from @ioc.  Must be called
- * with @q->queue_lock held.
+ * from io issue path, either return NULL if current issue io to @q for the
+ * first time, or return a valid icq.
  */
 struct io_cq *ioc_lookup_icq(struct request_queue *q)
 {
 	struct io_context *ioc = current->io_context;
 	struct io_cq *icq;
 
-	lockdep_assert_held(&q->queue_lock);
-
 	/*
 	 * icq's are indexed from @ioc using radix tree and hint pointer,
-	 * both of which are protected with RCU.  All removals are done
+	 * both of which are protected with RCU, io issue path ensures that
-	 * holding both q and ioc locks, and we're holding q lock - if we
+	 * both request_queue and current task are valid, the found icq
-	 * find a icq which points to us, it's guaranteed to be valid.
+	 * is guaranteed to be valid until the io is done.
 	 */
 	rcu_read_lock();
 	icq = rcu_dereference(ioc->icq_hint);
@@ -419,10 +418,7 @@ struct io_cq *ioc_find_get_icq(struct request_queue *q)
 		task_unlock(current);
 	} else {
 		get_io_context(ioc);
-
-		spin_lock_irq(&q->queue_lock);
 		icq = ioc_lookup_icq(q);
-		spin_unlock_irq(&q->queue_lock);
 	}
 
 	if (!icq) {

block/blk-mq-sched.c

@@ -374,64 +374,17 @@ bool blk_mq_sched_try_insert_merge(struct request_queue *q, struct request *rq,
 }
 EXPORT_SYMBOL_GPL(blk_mq_sched_try_insert_merge);
 
-static int blk_mq_sched_alloc_map_and_rqs(struct request_queue *q,
-					  struct blk_mq_hw_ctx *hctx,
-					  unsigned int hctx_idx)
-{
-	if (blk_mq_is_shared_tags(q->tag_set->flags)) {
-		hctx->sched_tags = q->sched_shared_tags;
-		return 0;
-	}
-
-	hctx->sched_tags = blk_mq_alloc_map_and_rqs(q->tag_set, hctx_idx,
-						    q->nr_requests);
-
-	if (!hctx->sched_tags)
-		return -ENOMEM;
-	return 0;
-}
-
-static void blk_mq_exit_sched_shared_tags(struct request_queue *queue)
-{
-	blk_mq_free_rq_map(queue->sched_shared_tags);
-	queue->sched_shared_tags = NULL;
-}
-
 /* called in queue's release handler, tagset has gone away */
 static void blk_mq_sched_tags_teardown(struct request_queue *q, unsigned int flags)
 {
 	struct blk_mq_hw_ctx *hctx;
 	unsigned long i;
 
-	queue_for_each_hw_ctx(q, hctx, i) {
+	queue_for_each_hw_ctx(q, hctx, i)
-		if (hctx->sched_tags) {
+		hctx->sched_tags = NULL;
-			if (!blk_mq_is_shared_tags(flags))
-				blk_mq_free_rq_map(hctx->sched_tags);
-			hctx->sched_tags = NULL;
-		}
-	}
 
 	if (blk_mq_is_shared_tags(flags))
-		blk_mq_exit_sched_shared_tags(q);
+		q->sched_shared_tags = NULL;
-}
-
-static int blk_mq_init_sched_shared_tags(struct request_queue *queue)
-{
-	struct blk_mq_tag_set *set = queue->tag_set;
-
-	/*
-	 * Set initial depth at max so that we don't need to reallocate for
-	 * updating nr_requests.
-	 */
-	queue->sched_shared_tags = blk_mq_alloc_map_and_rqs(set,
-						BLK_MQ_NO_HCTX_IDX,
-						MAX_SCHED_RQ);
-	if (!queue->sched_shared_tags)
-		return -ENOMEM;
-
-	blk_mq_tag_update_sched_shared_tags(queue);
-
-	return 0;
 }
 
 void blk_mq_sched_reg_debugfs(struct request_queue *q)
@@ -458,8 +411,140 @@ void blk_mq_sched_unreg_debugfs(struct request_queue *q)
 	mutex_unlock(&q->debugfs_mutex);
 }
 
+void blk_mq_free_sched_tags(struct elevator_tags *et,
+		struct blk_mq_tag_set *set)
+{
+	unsigned long i;
+
+	/* Shared tags are stored at index 0 in @tags. */
+	if (blk_mq_is_shared_tags(set->flags))
+		blk_mq_free_map_and_rqs(set, et->tags[0], BLK_MQ_NO_HCTX_IDX);
+	else {
+		for (i = 0; i < et->nr_hw_queues; i++)
+			blk_mq_free_map_and_rqs(set, et->tags[i], i);
+	}
+
+	kfree(et);
+}
+
+void blk_mq_free_sched_tags_batch(struct xarray *et_table,
+		struct blk_mq_tag_set *set)
+{
+	struct request_queue *q;
+	struct elevator_tags *et;
+
+	lockdep_assert_held_write(&set->update_nr_hwq_lock);
+
+	list_for_each_entry(q, &set->tag_list, tag_set_list) {
+		/*
+		 * Accessing q->elevator without holding q->elevator_lock is
+		 * safe because we're holding here set->update_nr_hwq_lock in
+		 * the writer context. So, scheduler update/switch code (which
+		 * acquires the same lock but in the reader context) can't run
+		 * concurrently.
+		 */
+		if (q->elevator) {
+			et = xa_load(et_table, q->id);
+			if (unlikely(!et))
+				WARN_ON_ONCE(1);
+			else
+				blk_mq_free_sched_tags(et, set);
+		}
+	}
+}
+
+struct elevator_tags *blk_mq_alloc_sched_tags(struct blk_mq_tag_set *set,
+		unsigned int nr_hw_queues)
+{
+	unsigned int nr_tags;
+	int i;
+	struct elevator_tags *et;
+	gfp_t gfp = GFP_NOIO | __GFP_ZERO | __GFP_NOWARN | __GFP_NORETRY;
+
+	if (blk_mq_is_shared_tags(set->flags))
+		nr_tags = 1;
+	else
+		nr_tags = nr_hw_queues;
+
+	et = kmalloc(sizeof(struct elevator_tags) +
+			nr_tags * sizeof(struct blk_mq_tags *), gfp);
+	if (!et)
+		return NULL;
+	/*
+	 * Default to double of smaller one between hw queue_depth and
+	 * 128, since we don't split into sync/async like the old code
+	 * did. Additionally, this is a per-hw queue depth.
+	 */
+	et->nr_requests = 2 * min_t(unsigned int, set->queue_depth,
+			BLKDEV_DEFAULT_RQ);
+	et->nr_hw_queues = nr_hw_queues;
+
+	if (blk_mq_is_shared_tags(set->flags)) {
+		/* Shared tags are stored at index 0 in @tags. */
+		et->tags[0] = blk_mq_alloc_map_and_rqs(set, BLK_MQ_NO_HCTX_IDX,
+					MAX_SCHED_RQ);
+		if (!et->tags[0])
+			goto out;
+	} else {
+		for (i = 0; i < et->nr_hw_queues; i++) {
+			et->tags[i] = blk_mq_alloc_map_and_rqs(set, i,
+					et->nr_requests);
+			if (!et->tags[i])
+				goto out_unwind;
+		}
+	}
+
+	return et;
+out_unwind:
+	while (--i >= 0)
+		blk_mq_free_map_and_rqs(set, et->tags[i], i);
+out:
+	kfree(et);
+	return NULL;
+}
+
+int blk_mq_alloc_sched_tags_batch(struct xarray *et_table,
+		struct blk_mq_tag_set *set, unsigned int nr_hw_queues)
+{
+	struct request_queue *q;
+	struct elevator_tags *et;
+	gfp_t gfp = GFP_NOIO | __GFP_ZERO | __GFP_NOWARN | __GFP_NORETRY;
+
+	lockdep_assert_held_write(&set->update_nr_hwq_lock);
+
+	list_for_each_entry(q, &set->tag_list, tag_set_list) {
+		/*
+		 * Accessing q->elevator without holding q->elevator_lock is
+		 * safe because we're holding here set->update_nr_hwq_lock in
+		 * the writer context. So, scheduler update/switch code (which
+		 * acquires the same lock but in the reader context) can't run
+		 * concurrently.
+		 */
+		if (q->elevator) {
+			et = blk_mq_alloc_sched_tags(set, nr_hw_queues);
+			if (!et)
+				goto out_unwind;
+			if (xa_insert(et_table, q->id, et, gfp))
+				goto out_free_tags;
+		}
+	}
+	return 0;
+out_free_tags:
+	blk_mq_free_sched_tags(et, set);
+out_unwind:
+	list_for_each_entry_continue_reverse(q, &set->tag_list, tag_set_list) {
+		if (q->elevator) {
+			et = xa_load(et_table, q->id);
+			if (et)
+				blk_mq_free_sched_tags(et, set);
+		}
+	}
+	return -ENOMEM;
+}
+
 /* caller must have a reference to @e, will grab another one if successful */
-int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
+int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e,
+		struct elevator_tags *et)
 {
 	unsigned int flags = q->tag_set->flags;
 	struct blk_mq_hw_ctx *hctx;
@@ -467,36 +552,33 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
 	unsigned long i;
 	int ret;
 
-	/*
+	eq = elevator_alloc(q, e, et);
-	 * Default to double of smaller one between hw queue_depth and 128,
+	if (!eq)
-	 * since we don't split into sync/async like the old code did.
+		return -ENOMEM;
-	 * Additionally, this is a per-hw queue depth.
+
-	 */
+	q->nr_requests = et->nr_requests;
-	q->nr_requests = 2 * min_t(unsigned int, q->tag_set->queue_depth,
-				   BLKDEV_DEFAULT_RQ);
 
 	if (blk_mq_is_shared_tags(flags)) {
-		ret = blk_mq_init_sched_shared_tags(q);
+		/* Shared tags are stored at index 0 in @et->tags. */
-		if (ret)
+		q->sched_shared_tags = et->tags[0];
-			return ret;
+		blk_mq_tag_update_sched_shared_tags(q);
 	}
 
 	queue_for_each_hw_ctx(q, hctx, i) {
-		ret = blk_mq_sched_alloc_map_and_rqs(q, hctx, i);
+		if (blk_mq_is_shared_tags(flags))
-		if (ret)
+			hctx->sched_tags = q->sched_shared_tags;
-			goto err_free_map_and_rqs;
+		else
+			hctx->sched_tags = et->tags[i];
 	}
 
-	ret = e->ops.init_sched(q, e);
+	ret = e->ops.init_sched(q, eq);
 	if (ret)
-		goto err_free_map_and_rqs;
+		goto out;
 
 	queue_for_each_hw_ctx(q, hctx, i) {
 		if (e->ops.init_hctx) {
 			ret = e->ops.init_hctx(hctx, i);
 			if (ret) {
-				eq = q->elevator;
-				blk_mq_sched_free_rqs(q);
 				blk_mq_exit_sched(q, eq);
 				kobject_put(&eq->kobj);
 				return ret;
@@ -505,10 +587,9 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
 	}
 	return 0;
 
-err_free_map_and_rqs:
+out:
-	blk_mq_sched_free_rqs(q);
 	blk_mq_sched_tags_teardown(q, flags);
-
+	kobject_put(&eq->kobj);
 	q->elevator = NULL;
 	return ret;
 }

block/blk-mq-sched.h

@@ -18,10 +18,20 @@ void __blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx);
 
 void blk_mq_sched_dispatch_requests(struct blk_mq_hw_ctx *hctx);
 
-int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e);
+int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e,
+		struct elevator_tags *et);
 void blk_mq_exit_sched(struct request_queue *q, struct elevator_queue *e);
 void blk_mq_sched_free_rqs(struct request_queue *q);
 
+struct elevator_tags *blk_mq_alloc_sched_tags(struct blk_mq_tag_set *set,
+		unsigned int nr_hw_queues);
+int blk_mq_alloc_sched_tags_batch(struct xarray *et_table,
+		struct blk_mq_tag_set *set, unsigned int nr_hw_queues);
+void blk_mq_free_sched_tags(struct elevator_tags *et,
+		struct blk_mq_tag_set *set);
+void blk_mq_free_sched_tags_batch(struct xarray *et_table,
+		struct blk_mq_tag_set *set);
+
 static inline void blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx)
 {
 	if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))

block/blk-mq.c

@@ -4974,12 +4974,13 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
  * Switch back to the elevator type stored in the xarray.
  */
 static void blk_mq_elv_switch_back(struct request_queue *q,
-		struct xarray *elv_tbl)
+		struct xarray *elv_tbl, struct xarray *et_tbl)
 {
 	struct elevator_type *e = xa_load(elv_tbl, q->id);
+	struct elevator_tags *t = xa_load(et_tbl, q->id);
 
 	/* The elv_update_nr_hw_queues unfreezes the queue. */
-	elv_update_nr_hw_queues(q, e);
+	elv_update_nr_hw_queues(q, e, t);
 
 	/* Drop the reference acquired in blk_mq_elv_switch_none. */
 	if (e)
@@ -5031,7 +5032,7 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
 	int prev_nr_hw_queues = set->nr_hw_queues;
 	unsigned int memflags;
 	int i;
-	struct xarray elv_tbl;
+	struct xarray elv_tbl, et_tbl;
 
 	lockdep_assert_held(&set->tag_list_lock);
 
@@ -5044,6 +5045,10 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
 
 	memflags = memalloc_noio_save();
 
+	xa_init(&et_tbl);
+	if (blk_mq_alloc_sched_tags_batch(&et_tbl, set, nr_hw_queues) < 0)
+		goto out_memalloc_restore;
+
 	xa_init(&elv_tbl);
 
 	list_for_each_entry(q, &set->tag_list, tag_set_list) {
@@ -5087,7 +5092,7 @@ fallback:
 switch_back:
 	/* The blk_mq_elv_switch_back unfreezes queue for us. */
 	list_for_each_entry(q, &set->tag_list, tag_set_list)
-		blk_mq_elv_switch_back(q, &elv_tbl);
+		blk_mq_elv_switch_back(q, &elv_tbl, &et_tbl);
 
 	list_for_each_entry(q, &set->tag_list, tag_set_list) {
 		blk_mq_sysfs_register_hctxs(q);
@@ -5098,7 +5103,8 @@ switch_back:
 	}
 
 	xa_destroy(&elv_tbl);
-
+	xa_destroy(&et_tbl);
+out_memalloc_restore:
 	memalloc_noio_restore(memflags);
 
 	/* Free the excess tags when nr_hw_queues shrink. */

block/blk-settings.c

@@ -62,16 +62,24 @@ EXPORT_SYMBOL(blk_set_stacking_limits);
 void blk_apply_bdi_limits(struct backing_dev_info *bdi,
 		struct queue_limits *lim)
 {
+	u64 io_opt = lim->io_opt;
+
 	/*
 	 * For read-ahead of large files to be effective, we need to read ahead
-	 * at least twice the optimal I/O size.
+	 * at least twice the optimal I/O size. For rotational devices that do
+	 * not report an optimal I/O size (e.g. ATA HDDs), use the maximum I/O
+	 * size to avoid falling back to the (rather inefficient) small default
+	 * read-ahead size.
 	 *
 	 * There is no hardware limitation for the read-ahead size and the user
 	 * might have increased the read-ahead size through sysfs, so don't ever
 	 * decrease it.
 	 */
+	if (!io_opt && (lim->features & BLK_FEAT_ROTATIONAL))
+		io_opt = (u64)lim->max_sectors << SECTOR_SHIFT;
+
 	bdi->ra_pages = max3(bdi->ra_pages,
-				lim->io_opt * 2 / PAGE_SIZE,
+				io_opt * 2 >> PAGE_SHIFT,
 				VM_READAHEAD_PAGES);
 	bdi->io_pages = lim->max_sectors >> PAGE_SECTORS_SHIFT;
 }
@@ -312,8 +320,12 @@ int blk_validate_limits(struct queue_limits *lim)
 		pr_warn("Invalid logical block size (%d)\n", lim->logical_block_size);
 		return -EINVAL;
 	}
-	if (lim->physical_block_size < lim->logical_block_size)
+	if (lim->physical_block_size < lim->logical_block_size) {
 		lim->physical_block_size = lim->logical_block_size;
+	} else if (!is_power_of_2(lim->physical_block_size)) {
+		pr_warn("Invalid physical block size (%d)\n", lim->physical_block_size);
+		return -EINVAL;
+	}
 
 	/*
 	 * The minimum I/O size defaults to the physical block size unless
@@ -388,12 +400,19 @@ int blk_validate_limits(struct queue_limits *lim)
 	lim->max_discard_sectors =
 		min(lim->max_hw_discard_sectors, lim->max_user_discard_sectors);
 
+	/*
+	 * When discard is not supported, discard_granularity should be reported
+	 * as 0 to userspace.
+	 */
+	if (lim->max_discard_sectors)
+		lim->discard_granularity =
+			max(lim->discard_granularity, lim->physical_block_size);
+	else
+		lim->discard_granularity = 0;
+
 	if (!lim->max_discard_segments)
 		lim->max_discard_segments = 1;
 
-	if (lim->discard_granularity < lim->physical_block_size)
-		lim->discard_granularity = lim->physical_block_size;
-
 	/*
 	 * By default there is no limit on the segment boundary alignment,
 	 * but if there is one it can't be smaller than the page size as
@@ -849,7 +868,7 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
 	}
 
 	/* chunk_sectors a multiple of the physical block size? */
-	if ((t->chunk_sectors << 9) & (t->physical_block_size - 1)) {
+	if (t->chunk_sectors % (t->physical_block_size >> SECTOR_SHIFT)) {
 		t->chunk_sectors = 0;
 		t->flags |= BLK_FLAG_MISALIGNED;
 		ret = -1;

block/blk.h

@@ -12,6 +12,7 @@
 #include "blk-crypto-internal.h"
 
 struct elevator_type;
+struct elevator_tags;
 
 /*
  * Default upper limit for the software max_sectors limit used for regular I/Os.
@@ -330,7 +331,8 @@ bool blk_bio_list_merge(struct request_queue *q, struct list_head *list,
 
 bool blk_insert_flush(struct request *rq);
 
-void elv_update_nr_hw_queues(struct request_queue *q, struct elevator_type *e);
+void elv_update_nr_hw_queues(struct request_queue *q, struct elevator_type *e,
+		struct elevator_tags *t);
 void elevator_set_default(struct request_queue *q);
 void elevator_set_none(struct request_queue *q);
 

block/elevator.c

@@ -54,6 +54,8 @@ struct elv_change_ctx {
 	struct elevator_queue *old;
 	/* for registering new elevator */
 	struct elevator_queue *new;
+	/* holds sched tags data */
+	struct elevator_tags *et;
 };
 
 static DEFINE_SPINLOCK(elv_list_lock);
@@ -132,7 +134,7 @@ static struct elevator_type *elevator_find_get(const char *name)
 static const struct kobj_type elv_ktype;
 
 struct elevator_queue *elevator_alloc(struct request_queue *q,
-				  struct elevator_type *e)
+		struct elevator_type *e, struct elevator_tags *et)
 {
 	struct elevator_queue *eq;
 
@@ -145,10 +147,10 @@ struct elevator_queue *elevator_alloc(struct request_queue *q,
 	kobject_init(&eq->kobj, &elv_ktype);
 	mutex_init(&eq->sysfs_lock);
 	hash_init(eq->hash);
+	eq->et = et;
 
 	return eq;
 }
-EXPORT_SYMBOL(elevator_alloc);
 
 static void elevator_release(struct kobject *kobj)
 {
@@ -166,7 +168,6 @@ static void elevator_exit(struct request_queue *q)
 	lockdep_assert_held(&q->elevator_lock);
 
 	ioc_clear_queue(q);
-	blk_mq_sched_free_rqs(q);
 
 	mutex_lock(&e->sysfs_lock);
 	blk_mq_exit_sched(q, e);
@@ -592,7 +593,7 @@ static int elevator_switch(struct request_queue *q, struct elv_change_ctx *ctx)
 	}
 
 	if (new_e) {
-		ret = blk_mq_init_sched(q, new_e);
+		ret = blk_mq_init_sched(q, new_e, ctx->et);
 		if (ret)
 			goto out_unfreeze;
 		ctx->new = q->elevator;
@@ -627,8 +628,10 @@ static void elv_exit_and_release(struct request_queue *q)
 	elevator_exit(q);
 	mutex_unlock(&q->elevator_lock);
 	blk_mq_unfreeze_queue(q, memflags);
-	if (e)
+	if (e) {
+		blk_mq_free_sched_tags(e->et, q->tag_set);
 		kobject_put(&e->kobj);
+	}
 }
 
 static int elevator_change_done(struct request_queue *q,
@@ -641,6 +644,7 @@ static int elevator_change_done(struct request_queue *q,
 				&ctx->old->flags);
 
 		elv_unregister_queue(q, ctx->old);
+		blk_mq_free_sched_tags(ctx->old->et, q->tag_set);
 		kobject_put(&ctx->old->kobj);
 		if (enable_wbt)
 			wbt_enable_default(q->disk);
@@ -659,9 +663,16 @@ static int elevator_change_done(struct request_queue *q,
 static int elevator_change(struct request_queue *q, struct elv_change_ctx *ctx)
 {
 	unsigned int memflags;
+	struct blk_mq_tag_set *set = q->tag_set;
 	int ret = 0;
 
-	lockdep_assert_held(&q->tag_set->update_nr_hwq_lock);
+	lockdep_assert_held(&set->update_nr_hwq_lock);
+
+	if (strncmp(ctx->name, "none", 4)) {
+		ctx->et = blk_mq_alloc_sched_tags(set, set->nr_hw_queues);
+		if (!ctx->et)
+			return -ENOMEM;
+	}
 
 	memflags = blk_mq_freeze_queue(q);
 	/*
@@ -681,6 +692,11 @@ static int elevator_change(struct request_queue *q, struct elv_change_ctx *ctx)
 	blk_mq_unfreeze_queue(q, memflags);
 	if (!ret)
 		ret = elevator_change_done(q, ctx);
+	/*
+	 * Free sched tags if it's allocated but we couldn't switch elevator.
+	 */
+	if (ctx->et && !ctx->new)
+		blk_mq_free_sched_tags(ctx->et, set);
 
 	return ret;
 }
@@ -689,8 +705,10 @@ static int elevator_change(struct request_queue *q, struct elv_change_ctx *ctx)
  * The I/O scheduler depends on the number of hardware queues, this forces a
  * reattachment when nr_hw_queues changes.
  */
-void elv_update_nr_hw_queues(struct request_queue *q, struct elevator_type *e)
+void elv_update_nr_hw_queues(struct request_queue *q, struct elevator_type *e,
+		struct elevator_tags *t)
 {
+	struct blk_mq_tag_set *set = q->tag_set;
 	struct elv_change_ctx ctx = {};
 	int ret = -ENODEV;
 
@@ -698,6 +716,7 @@ void elv_update_nr_hw_queues(struct request_queue *q, struct elevator_type *e)
 
 	if (e && !blk_queue_dying(q) && blk_queue_registered(q)) {
 		ctx.name = e->elevator_name;
+		ctx.et = t;
 
 		mutex_lock(&q->elevator_lock);
 		/* force to reattach elevator after nr_hw_queue is updated */
@@ -707,6 +726,11 @@ void elv_update_nr_hw_queues(struct request_queue *q, struct elevator_type *e)
 	blk_mq_unfreeze_queue_nomemrestore(q);
 	if (!ret)
 		WARN_ON_ONCE(elevator_change_done(q, &ctx));
+	/*
+	 * Free sched tags if it's allocated but we couldn't switch elevator.
+	 */
+	if (t && !ctx.new)
+		blk_mq_free_sched_tags(t, set);
 }
 
 /*

block/elevator.h

@@ -23,8 +23,17 @@ enum elv_merge {
 struct blk_mq_alloc_data;
 struct blk_mq_hw_ctx;
 
+struct elevator_tags {
+	/* num. of hardware queues for which tags are allocated */
+	unsigned int nr_hw_queues;
+	/* depth used while allocating tags */
+	unsigned int nr_requests;
+	/* shared tag is stored at index 0 */
+	struct blk_mq_tags *tags[];
+};
+
 struct elevator_mq_ops {
-	int (*init_sched)(struct request_queue *, struct elevator_type *);
+	int (*init_sched)(struct request_queue *, struct elevator_queue *);
 	void (*exit_sched)(struct elevator_queue *);
 	int (*init_hctx)(struct blk_mq_hw_ctx *, unsigned int);
 	void (*exit_hctx)(struct blk_mq_hw_ctx *, unsigned int);
@@ -113,6 +122,7 @@ struct request *elv_rqhash_find(struct request_queue *q, sector_t offset);
 struct elevator_queue
 {
 	struct elevator_type *type;
+	struct elevator_tags *et;
 	void *elevator_data;
 	struct kobject kobj;
 	struct mutex sysfs_lock;
@@ -152,8 +162,8 @@ ssize_t elv_iosched_show(struct gendisk *disk, char *page);
 ssize_t elv_iosched_store(struct gendisk *disk, const char *page, size_t count);
 
 extern bool elv_bio_merge_ok(struct request *, struct bio *);
-extern struct elevator_queue *elevator_alloc(struct request_queue *,
+struct elevator_queue *elevator_alloc(struct request_queue *,
-					struct elevator_type *);
+		struct elevator_type *, struct elevator_tags *);
 
 /*
  * Helper functions.

block/kyber-iosched.c

@@ -157,10 +157,7 @@ struct kyber_queue_data {
 	 */
 	struct sbitmap_queue domain_tokens[KYBER_NUM_DOMAINS];
 
-	/*
+	/* Number of allowed async requests. */
-	 * Async request percentage, converted to per-word depth for
-	 * sbitmap_get_shallow().
-	 */
 	unsigned int async_depth;
 
 	struct kyber_cpu_latency __percpu *cpu_latency;
@@ -402,20 +399,13 @@ err:
 	return ERR_PTR(ret);
 }
 
-static int kyber_init_sched(struct request_queue *q, struct elevator_type *e)
+static int kyber_init_sched(struct request_queue *q, struct elevator_queue *eq)
 {
 	struct kyber_queue_data *kqd;
-	struct elevator_queue *eq;
-
-	eq = elevator_alloc(q, e);
-	if (!eq)
-		return -ENOMEM;
 
 	kqd = kyber_queue_data_alloc(q);
-	if (IS_ERR(kqd)) {
+	if (IS_ERR(kqd))
-		kobject_put(&eq->kobj);
 		return PTR_ERR(kqd);
-	}
 
 	blk_stat_enable_accounting(q);
 
@@ -454,10 +444,8 @@ static void kyber_depth_updated(struct blk_mq_hw_ctx *hctx)
 {
 	struct kyber_queue_data *kqd = hctx->queue->elevator->elevator_data;
 	struct blk_mq_tags *tags = hctx->sched_tags;
-	unsigned int shift = tags->bitmap_tags.sb.shift;
-
-	kqd->async_depth = (1U << shift) * KYBER_ASYNC_PERCENT / 100U;
 
+	kqd->async_depth = hctx->queue->nr_requests * KYBER_ASYNC_PERCENT / 100U;
 	sbitmap_queue_min_shallow_depth(&tags->bitmap_tags, kqd->async_depth);
 }
 

block/mq-deadline.c

@@ -487,20 +487,6 @@ unlock:
 	return rq;
 }
 
-/*
- * 'depth' is a number in the range 1..INT_MAX representing a number of
- * requests. Scale it with a factor (1 << bt->sb.shift) / q->nr_requests since
- * 1..(1 << bt->sb.shift) is the range expected by sbitmap_get_shallow().
- * Values larger than q->nr_requests have the same effect as q->nr_requests.
- */
-static int dd_to_word_depth(struct blk_mq_hw_ctx *hctx, unsigned int qdepth)
-{
-	struct sbitmap_queue *bt = &hctx->sched_tags->bitmap_tags;
-	const unsigned int nrr = hctx->queue->nr_requests;
-
-	return ((qdepth << bt->sb.shift) + nrr - 1) / nrr;
-}
-
 /*
  * Called by __blk_mq_alloc_request(). The shallow_depth value set by this
  * function is used by __blk_mq_get_tag().
@@ -517,7 +503,7 @@ static void dd_limit_depth(blk_opf_t opf, struct blk_mq_alloc_data *data)
 	 * Throttle asynchronous requests and writes such that these requests
 	 * do not block the allocation of synchronous requests.
 	 */
-	data->shallow_depth = dd_to_word_depth(data->hctx, dd->async_depth);
+	data->shallow_depth = dd->async_depth;
 }
 
 /* Called by blk_mq_update_nr_requests(). */
@@ -568,20 +554,14 @@ static void dd_exit_sched(struct elevator_queue *e)
 /*
  * initialize elevator private data (deadline_data).
  */
-static int dd_init_sched(struct request_queue *q, struct elevator_type *e)
+static int dd_init_sched(struct request_queue *q, struct elevator_queue *eq)
 {
 	struct deadline_data *dd;
-	struct elevator_queue *eq;
 	enum dd_prio prio;
-	int ret = -ENOMEM;
-
-	eq = elevator_alloc(q, e);
-	if (!eq)
-		return ret;
 
 	dd = kzalloc_node(sizeof(*dd), GFP_KERNEL, q->node);
 	if (!dd)
-		goto put_eq;
+		return -ENOMEM;
 
 	eq->elevator_data = dd;
 
@@ -608,10 +588,6 @@ static int dd_init_sched(struct request_queue *q, struct elevator_type *e)
 
 	q->elevator = eq;
 	return 0;
-
-put_eq:
-	kobject_put(&eq->kobj);
-	return ret;
 }
 
 /*

drivers/block/zloop.c

@@ -700,6 +700,8 @@ static void zloop_free_disk(struct gendisk *disk)
 	struct zloop_device *zlo = disk->private_data;
 	unsigned int i;
 
+	blk_mq_free_tag_set(&zlo->tag_set);
+
 	for (i = 0; i < zlo->nr_zones; i++) {
 		struct zloop_zone *zone = &zlo->zones[i];
 
@@ -1080,7 +1082,6 @@ static int zloop_ctl_remove(struct zloop_options *opts)
 
 	del_gendisk(zlo->disk);
 	put_disk(zlo->disk);
-	blk_mq_free_tag_set(&zlo->tag_set);
 
 	pr_info("Removed device %d\n", opts->id);
 

drivers/md/dm-raid.c

@@ -438,7 +438,7 @@ static bool rs_is_reshapable(struct raid_set *rs)
 /* Return true, if raid set in @rs is recovering */
 static bool rs_is_recovering(struct raid_set *rs)
 {
-	return rs->md.recovery_cp < rs->md.dev_sectors;
+	return rs->md.resync_offset < rs->md.dev_sectors;
 }
 
 /* Return true, if raid set in @rs is reshaping */
@@ -768,7 +768,7 @@ static struct raid_set *raid_set_alloc(struct dm_target *ti, struct raid_type *r
 	rs->md.layout = raid_type->algorithm;
 	rs->md.new_layout = rs->md.layout;
 	rs->md.delta_disks = 0;
-	rs->md.recovery_cp = MaxSector;
+	rs->md.resync_offset = MaxSector;
 
 	for (i = 0; i < raid_devs; i++)
 		md_rdev_init(&rs->dev[i].rdev);
@@ -912,7 +912,7 @@ static int parse_dev_params(struct raid_set *rs, struct dm_arg_set *as)
 		rs->md.external = 0;
 		rs->md.persistent = 1;
 		rs->md.major_version = 2;
-	} else if (rebuild && !rs->md.recovery_cp) {
+	} else if (rebuild && !rs->md.resync_offset) {
 		/*
 		 * Without metadata, we will not be able to tell if the array
 		 * is in-sync or not - we must assume it is not.  Therefore,
@@ -1695,20 +1695,20 @@ static void rs_setup_recovery(struct raid_set *rs, sector_t dev_sectors)
 {
 	/* raid0 does not recover */
 	if (rs_is_raid0(rs))
-		rs->md.recovery_cp = MaxSector;
+		rs->md.resync_offset = MaxSector;
 	/*
 	 * A raid6 set has to be recovered either
 	 * completely or for the grown part to
 	 * ensure proper parity and Q-Syndrome
 	 */
 	else if (rs_is_raid6(rs))
-		rs->md.recovery_cp = dev_sectors;
+		rs->md.resync_offset = dev_sectors;
 	/*
 	 * Other raid set types may skip recovery
 	 * depending on the 'nosync' flag.
 	 */
 	else
-		rs->md.recovery_cp = test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)
+		rs->md.resync_offset = test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)
 				     ? MaxSector : dev_sectors;
 }
 
@@ -2143,7 +2143,7 @@ static void super_sync(struct mddev *mddev, struct md_rdev *rdev)
 	sb->events = cpu_to_le64(mddev->events);
 
 	sb->disk_recovery_offset = cpu_to_le64(rdev->recovery_offset);
-	sb->array_resync_offset = cpu_to_le64(mddev->recovery_cp);
+	sb->array_resync_offset = cpu_to_le64(mddev->resync_offset);
 
 	sb->level = cpu_to_le32(mddev->level);
 	sb->layout = cpu_to_le32(mddev->layout);
@@ -2334,18 +2334,18 @@ static int super_init_validation(struct raid_set *rs, struct md_rdev *rdev)
 	}
 
 	if (!test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags))
-		mddev->recovery_cp = le64_to_cpu(sb->array_resync_offset);
+		mddev->resync_offset = le64_to_cpu(sb->array_resync_offset);
 
 	/*
 	 * During load, we set FirstUse if a new superblock was written.
 	 * There are two reasons we might not have a superblock:
 	 * 1) The raid set is brand new - in which case, all of the
 	 *    devices must have their In_sync bit set.	Also,
-	 *    recovery_cp must be 0, unless forced.
+	 *    resync_offset must be 0, unless forced.
 	 * 2) This is a new device being added to an old raid set
 	 *    and the new device needs to be rebuilt - in which
 	 *    case the In_sync bit will /not/ be set and
-	 *    recovery_cp must be MaxSector.
+	 *    resync_offset must be MaxSector.
 	 * 3) This is/are a new device(s) being added to an old
 	 *    raid set during takeover to a higher raid level
 	 *    to provide capacity for redundancy or during reshape
@@ -2390,8 +2390,8 @@ static int super_init_validation(struct raid_set *rs, struct md_rdev *rdev)
 			      new_devs > 1 ? "s" : "");
 			return -EINVAL;
 		} else if (!test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags) && rs_is_recovering(rs)) {
-			DMERR("'rebuild' specified while raid set is not in-sync (recovery_cp=%llu)",
+			DMERR("'rebuild' specified while raid set is not in-sync (resync_offset=%llu)",
-			      (unsigned long long) mddev->recovery_cp);
+			      (unsigned long long) mddev->resync_offset);
 			return -EINVAL;
 		} else if (rs_is_reshaping(rs)) {
 			DMERR("'rebuild' specified while raid set is being reshaped (reshape_position=%llu)",
@@ -2700,11 +2700,11 @@ static int rs_adjust_data_offsets(struct raid_set *rs)
 	}
 out:
 	/*
-	 * Raise recovery_cp in case data_offset != 0 to
+	 * Raise resync_offset in case data_offset != 0 to
 	 * avoid false recovery positives in the constructor.
 	 */
-	if (rs->md.recovery_cp < rs->md.dev_sectors)
+	if (rs->md.resync_offset < rs->md.dev_sectors)
-		rs->md.recovery_cp += rs->dev[0].rdev.data_offset;
+		rs->md.resync_offset += rs->dev[0].rdev.data_offset;
 
 	/* Adjust data offsets on all rdevs but on any raid4/5/6 journal device */
 	rdev_for_each(rdev, &rs->md) {
@@ -2759,7 +2759,7 @@ static int rs_setup_takeover(struct raid_set *rs)
 	}
 
 	clear_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
-	mddev->recovery_cp = MaxSector;
+	mddev->resync_offset = MaxSector;
 
 	while (d--) {
 		rdev = &rs->dev[d].rdev;
@@ -2767,7 +2767,7 @@ static int rs_setup_takeover(struct raid_set *rs)
 		if (test_bit(d, (void *) rs->rebuild_disks)) {
 			clear_bit(In_sync, &rdev->flags);
 			clear_bit(Faulty, &rdev->flags);
-			mddev->recovery_cp = rdev->recovery_offset = 0;
+			mddev->resync_offset = rdev->recovery_offset = 0;
 			/* Bitmap has to be created when we do an "up" takeover */
 			set_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
 		}
@@ -3225,7 +3225,7 @@ size_check:
 			if (r)
 				goto bad;
 
-			rs_setup_recovery(rs, rs->md.recovery_cp < rs->md.dev_sectors ? rs->md.recovery_cp : rs->md.dev_sectors);
+			rs_setup_recovery(rs, rs->md.resync_offset < rs->md.dev_sectors ? rs->md.resync_offset : rs->md.dev_sectors);
 		} else {
 			/* This is no size change or it is shrinking, update size and record in superblocks */
 			r = rs_set_dev_and_array_sectors(rs, rs->ti->len, false);
@@ -3449,7 +3449,7 @@ static sector_t rs_get_progress(struct raid_set *rs, unsigned long recovery,
 
 	} else {
 		if (state == st_idle && !test_bit(MD_RECOVERY_INTR, &recovery))
-			r = mddev->recovery_cp;
+			r = mddev->resync_offset;
 		else
 			r = mddev->curr_resync_completed;
 
@@ -4077,9 +4077,9 @@ static int raid_preresume(struct dm_target *ti)
 	}
 
 	/* Check for any resize/reshape on @rs and adjust/initiate */
-	if (mddev->recovery_cp && mddev->recovery_cp < MaxSector) {
+	if (mddev->resync_offset && mddev->resync_offset < MaxSector) {
 		set_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
-		mddev->resync_min = mddev->recovery_cp;
+		mddev->resync_min = mddev->resync_offset;
 		if (test_bit(RT_FLAG_RS_GROW, &rs->runtime_flags))
 			mddev->resync_max_sectors = mddev->dev_sectors;
 	}

drivers/md/md-bitmap.c

@@ -1987,12 +1987,12 @@ static void bitmap_dirty_bits(struct mddev *mddev, unsigned long s,
 
 		md_bitmap_set_memory_bits(bitmap, sec, 1);
 		md_bitmap_file_set_bit(bitmap, sec);
-		if (sec < bitmap->mddev->recovery_cp)
+		if (sec < bitmap->mddev->resync_offset)
 			/* We are asserting that the array is dirty,
-			 * so move the recovery_cp address back so
+			 * so move the resync_offset address back so
 			 * that it is obvious that it is dirty
 			 */
-			bitmap->mddev->recovery_cp = sec;
+			bitmap->mddev->resync_offset = sec;
 	}
 }
 
@@ -2258,7 +2258,7 @@ static int bitmap_load(struct mddev *mddev)
 	    || bitmap->events_cleared == mddev->events)
 		/* no need to keep dirty bits to optimise a
 		 * re-add of a missing device */
-		start = mddev->recovery_cp;
+		start = mddev->resync_offset;
 
 	mutex_lock(&mddev->bitmap_info.mutex);
 	err = md_bitmap_init_from_disk(bitmap, start);

drivers/md/md-cluster.c

@@ -337,11 +337,11 @@ static void recover_bitmaps(struct md_thread *thread)
 			md_wakeup_thread(mddev->sync_thread);
 
 		if (hi > 0) {
-			if (lo < mddev->recovery_cp)
+			if (lo < mddev->resync_offset)
-				mddev->recovery_cp = lo;
+				mddev->resync_offset = lo;
 			/* wake up thread to continue resync in case resync
 			 * is not finished */
-			if (mddev->recovery_cp != MaxSector) {
+			if (mddev->resync_offset != MaxSector) {
 				/*
 				 * clear the REMOTE flag since we will launch
 				 * resync thread in current node.
@@ -863,9 +863,9 @@ static int gather_all_resync_info(struct mddev *mddev, int total_slots)
 			lockres_free(bm_lockres);
 			continue;
 		}
-		if ((hi > 0) && (lo < mddev->recovery_cp)) {
+		if ((hi > 0) && (lo < mddev->resync_offset)) {
 			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
-			mddev->recovery_cp = lo;
+			mddev->resync_offset = lo;
 			md_check_recovery(mddev);
 		}
 
@@ -1027,7 +1027,7 @@ static int leave(struct mddev *mddev)
 	 * Also, we should send BITMAP_NEEDS_SYNC message in
 	 * case reshaping is interrupted.
 	 */
-	if ((cinfo->slot_number > 0 && mddev->recovery_cp != MaxSector) ||
+	if ((cinfo->slot_number > 0 && mddev->resync_offset != MaxSector) ||
 	    (mddev->reshape_position != MaxSector &&
 	     test_bit(MD_CLOSING, &mddev->flags)))
 		resync_bitmap(mddev);
@@ -1605,8 +1605,8 @@ static int gather_bitmaps(struct md_rdev *rdev)
 			pr_warn("md-cluster: Could not gather bitmaps from slot %d", sn);
 			goto out;
 		}
-		if ((hi > 0) && (lo < mddev->recovery_cp))
+		if ((hi > 0) && (lo < mddev->resync_offset))
-			mddev->recovery_cp = lo;
+			mddev->resync_offset = lo;
 	}
 out:
 	return err;

drivers/md/md.c

@@ -636,6 +636,12 @@ static void __mddev_put(struct mddev *mddev)
 	    mddev->ctime || mddev->hold_active)
 		return;
 
+	/*
+	 * If array is freed by stopping array, MD_DELETED is set by
+	 * do_md_stop(), MD_DELETED is still set here in case mddev is freed
+	 * directly by closing a mddev that is created by create_on_open.
+	 */
+	set_bit(MD_DELETED, &mddev->flags);
 	/*
 	 * Call queue_work inside the spinlock so that flush_workqueue() after
 	 * mddev_find will succeed in waiting for the work to be done.
@@ -1409,13 +1415,13 @@ static int super_90_validate(struct mddev *mddev, struct md_rdev *freshest, stru
 			mddev->layout = -1;
 
 		if (sb->state & (1<<MD_SB_CLEAN))
-			mddev->recovery_cp = MaxSector;
+			mddev->resync_offset = MaxSector;
 		else {
 			if (sb->events_hi == sb->cp_events_hi &&
 				sb->events_lo == sb->cp_events_lo) {
-				mddev->recovery_cp = sb->recovery_cp;
+				mddev->resync_offset = sb->resync_offset;
 			} else
-				mddev->recovery_cp = 0;
+				mddev->resync_offset = 0;
 		}
 
 		memcpy(mddev->uuid+0, &sb->set_uuid0, 4);
@@ -1541,13 +1547,13 @@ static void super_90_sync(struct mddev *mddev, struct md_rdev *rdev)
 	mddev->minor_version = sb->minor_version;
 	if (mddev->in_sync)
 	{
-		sb->recovery_cp = mddev->recovery_cp;
+		sb->resync_offset = mddev->resync_offset;
 		sb->cp_events_hi = (mddev->events>>32);
 		sb->cp_events_lo = (u32)mddev->events;
-		if (mddev->recovery_cp == MaxSector)
+		if (mddev->resync_offset == MaxSector)
 			sb->state = (1<< MD_SB_CLEAN);
 	} else
-		sb->recovery_cp = 0;
+		sb->resync_offset = 0;
 
 	sb->layout = mddev->layout;
 	sb->chunk_size = mddev->chunk_sectors << 9;
@@ -1895,7 +1901,7 @@ static int super_1_validate(struct mddev *mddev, struct md_rdev *freshest, struc
 		mddev->bitmap_info.default_space = (4096-1024) >> 9;
 		mddev->reshape_backwards = 0;
 
-		mddev->recovery_cp = le64_to_cpu(sb->resync_offset);
+		mddev->resync_offset = le64_to_cpu(sb->resync_offset);
 		memcpy(mddev->uuid, sb->set_uuid, 16);
 
 		mddev->max_disks =  (4096-256)/2;
@@ -2081,7 +2087,7 @@ static void super_1_sync(struct mddev *mddev, struct md_rdev *rdev)
 	sb->utime = cpu_to_le64((__u64)mddev->utime);
 	sb->events = cpu_to_le64(mddev->events);
 	if (mddev->in_sync)
-		sb->resync_offset = cpu_to_le64(mddev->recovery_cp);
+		sb->resync_offset = cpu_to_le64(mddev->resync_offset);
 	else if (test_bit(MD_JOURNAL_CLEAN, &mddev->flags))
 		sb->resync_offset = cpu_to_le64(MaxSector);
 	else
@@ -2761,7 +2767,7 @@ repeat:
 	/* If this is just a dirty<->clean transition, and the array is clean
 	 * and 'events' is odd, we can roll back to the previous clean state */
 	if (nospares
-	    && (mddev->in_sync && mddev->recovery_cp == MaxSector)
+	    && (mddev->in_sync && mddev->resync_offset == MaxSector)
 	    && mddev->can_decrease_events
 	    && mddev->events != 1) {
 		mddev->events--;
@@ -4297,9 +4303,9 @@ __ATTR(chunk_size, S_IRUGO|S_IWUSR, chunk_size_show, chunk_size_store);
 static ssize_t
 resync_start_show(struct mddev *mddev, char *page)
 {
-	if (mddev->recovery_cp == MaxSector)
+	if (mddev->resync_offset == MaxSector)
 		return sprintf(page, "none\n");
-	return sprintf(page, "%llu\n", (unsigned long long)mddev->recovery_cp);
+	return sprintf(page, "%llu\n", (unsigned long long)mddev->resync_offset);
 }
 
 static ssize_t
@@ -4325,7 +4331,7 @@ resync_start_store(struct mddev *mddev, const char *buf, size_t len)
 		err = -EBUSY;
 
 	if (!err) {
-		mddev->recovery_cp = n;
+		mddev->resync_offset = n;
 		if (mddev->pers)
 			set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
 	}
@@ -6417,7 +6423,7 @@ static void md_clean(struct mddev *mddev)
 	mddev->external_size = 0;
 	mddev->dev_sectors = 0;
 	mddev->raid_disks = 0;
-	mddev->recovery_cp = 0;
+	mddev->resync_offset = 0;
 	mddev->resync_min = 0;
 	mddev->resync_max = MaxSector;
 	mddev->reshape_position = MaxSector;
@@ -7362,9 +7368,9 @@ int md_set_array_info(struct mddev *mddev, struct mdu_array_info_s *info)
 	 * openned
 	 */
 	if (info->state & (1<<MD_SB_CLEAN))
-		mddev->recovery_cp = MaxSector;
+		mddev->resync_offset = MaxSector;
 	else
-		mddev->recovery_cp = 0;
+		mddev->resync_offset = 0;
 	mddev->persistent    = ! info->not_persistent;
 	mddev->external	     = 0;
 
@@ -8303,7 +8309,7 @@ static int status_resync(struct seq_file *seq, struct mddev *mddev)
 				seq_printf(seq, "\tresync=REMOTE");
 			return 1;
 		}
-		if (mddev->recovery_cp < MaxSector) {
+		if (mddev->resync_offset < MaxSector) {
 			seq_printf(seq, "\tresync=PENDING");
 			return 1;
 		}
@@ -8946,7 +8952,7 @@ static sector_t md_sync_position(struct mddev *mddev, enum sync_action action)
 		return mddev->resync_min;
 	case ACTION_RESYNC:
 		if (!mddev->bitmap)
-			return mddev->recovery_cp;
+			return mddev->resync_offset;
 		return 0;
 	case ACTION_RESHAPE:
 		/*
@@ -9184,8 +9190,8 @@ void md_do_sync(struct md_thread *thread)
 				   atomic_read(&mddev->recovery_active) == 0);
 			mddev->curr_resync_completed = j;
 			if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) &&
-			    j > mddev->recovery_cp)
+			    j > mddev->resync_offset)
-				mddev->recovery_cp = j;
+				mddev->resync_offset = j;
 			update_time = jiffies;
 			set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
 			sysfs_notify_dirent_safe(mddev->sysfs_completed);
@@ -9305,19 +9311,19 @@ void md_do_sync(struct md_thread *thread)
 	    mddev->curr_resync > MD_RESYNC_ACTIVE) {
 		if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
 			if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
-				if (mddev->curr_resync >= mddev->recovery_cp) {
+				if (mddev->curr_resync >= mddev->resync_offset) {
 					pr_debug("md: checkpointing %s of %s.\n",
 						 desc, mdname(mddev));
 					if (test_bit(MD_RECOVERY_ERROR,
 						&mddev->recovery))
-						mddev->recovery_cp =
+						mddev->resync_offset =
 							mddev->curr_resync_completed;
 					else
-						mddev->recovery_cp =
+						mddev->resync_offset =
 							mddev->curr_resync;
 				}
 			} else
-				mddev->recovery_cp = MaxSector;
+				mddev->resync_offset = MaxSector;
 		} else {
 			if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery))
 				mddev->curr_resync = MaxSector;
@@ -9421,6 +9427,12 @@ static bool rdev_is_spare(struct md_rdev *rdev)
 
 static bool rdev_addable(struct md_rdev *rdev)
 {
+	struct mddev *mddev;
+
+	mddev = READ_ONCE(rdev->mddev);
+	if (!mddev)
+		return false;
+
 	/* rdev is already used, don't add it again. */
 	if (test_bit(Candidate, &rdev->flags) || rdev->raid_disk >= 0 ||
 	    test_bit(Faulty, &rdev->flags))
@@ -9431,7 +9443,7 @@ static bool rdev_addable(struct md_rdev *rdev)
 		return true;
 
 	/* Allow to add if array is read-write. */
-	if (md_is_rdwr(rdev->mddev))
+	if (md_is_rdwr(mddev))
 		return true;
 
 	/*
@@ -9533,7 +9545,7 @@ static bool md_choose_sync_action(struct mddev *mddev, int *spares)
 	}
 
 	/* Check if resync is in progress. */
-	if (mddev->recovery_cp < MaxSector) {
+	if (mddev->resync_offset < MaxSector) {
 		remove_spares(mddev, NULL);
 		set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
 		clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
@@ -9714,7 +9726,7 @@ void md_check_recovery(struct mddev *mddev)
 		test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
 		(mddev->external == 0 && mddev->safemode == 1) ||
 		(mddev->safemode == 2
-		 && !mddev->in_sync && mddev->recovery_cp == MaxSector)
+		 && !mddev->in_sync && mddev->resync_offset == MaxSector)
 		))
 		return;
 
@@ -9771,8 +9783,8 @@ void md_check_recovery(struct mddev *mddev)
 			 * remove disk.
 			 */
 			rdev_for_each_safe(rdev, tmp, mddev) {
-				if (test_and_clear_bit(ClusterRemove, &rdev->flags) &&
+				if (rdev->raid_disk < 0 &&
-						rdev->raid_disk < 0)
+				    test_and_clear_bit(ClusterRemove, &rdev->flags))
 					md_kick_rdev_from_array(rdev);
 			}
 		}
@@ -10078,8 +10090,11 @@ static void check_sb_changes(struct mddev *mddev, struct md_rdev *rdev)
 
 	/* Check for change of roles in the active devices */
 	rdev_for_each_safe(rdev2, tmp, mddev) {
-		if (test_bit(Faulty, &rdev2->flags))
+		if (test_bit(Faulty, &rdev2->flags)) {
+			if (test_bit(ClusterRemove, &rdev2->flags))
+				set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 			continue;
+		}
 
 		/* Check if the roles changed */
 		role = le16_to_cpu(sb->dev_roles[rdev2->desc_nr]);

drivers/md/md.h

@@ -523,7 +523,7 @@ struct mddev {
 	unsigned long			normal_io_events; /* IO event timestamp */
 	atomic_t			recovery_active; /* blocks scheduled, but not written */
 	wait_queue_head_t		recovery_wait;
-	sector_t			recovery_cp;
+	sector_t			resync_offset;
 	sector_t			resync_min;	/* user requested sync
 							 * starts here */
 	sector_t			resync_max;	/* resync should pause

drivers/md/raid0.c

@@ -674,7 +674,7 @@ static void *raid0_takeover_raid45(struct mddev *mddev)
 	mddev->raid_disks--;
 	mddev->delta_disks = -1;
 	/* make sure it will be not marked as dirty */
-	mddev->recovery_cp = MaxSector;
+	mddev->resync_offset = MaxSector;
 	mddev_clear_unsupported_flags(mddev, UNSUPPORTED_MDDEV_FLAGS);
 
 	create_strip_zones(mddev, &priv_conf);
@@ -717,7 +717,7 @@ static void *raid0_takeover_raid10(struct mddev *mddev)
 	mddev->raid_disks += mddev->delta_disks;
 	mddev->degraded = 0;
 	/* make sure it will be not marked as dirty */
-	mddev->recovery_cp = MaxSector;
+	mddev->resync_offset = MaxSector;
 	mddev_clear_unsupported_flags(mddev, UNSUPPORTED_MDDEV_FLAGS);
 
 	create_strip_zones(mddev, &priv_conf);
@@ -760,7 +760,7 @@ static void *raid0_takeover_raid1(struct mddev *mddev)
 	mddev->delta_disks = 1 - mddev->raid_disks;
 	mddev->raid_disks = 1;
 	/* make sure it will be not marked as dirty */
-	mddev->recovery_cp = MaxSector;
+	mddev->resync_offset = MaxSector;
 	mddev_clear_unsupported_flags(mddev, UNSUPPORTED_MDDEV_FLAGS);
 
 	create_strip_zones(mddev, &priv_conf);

drivers/md/raid1-10.c

@@ -283,7 +283,7 @@ static inline int raid1_check_read_range(struct md_rdev *rdev,
 static inline bool raid1_should_read_first(struct mddev *mddev,
 					   sector_t this_sector, int len)
 {
-	if ((mddev->recovery_cp < this_sector + len))
+	if ((mddev->resync_offset < this_sector + len))
 		return true;
 
 	if (mddev_is_clustered(mddev) &&

drivers/md/raid1.c

@@ -127,10 +127,9 @@ static inline struct r1bio *get_resync_r1bio(struct bio *bio)
 	return get_resync_pages(bio)->raid_bio;
 }
 
-static void * r1bio_pool_alloc(gfp_t gfp_flags, void *data)
+static void *r1bio_pool_alloc(gfp_t gfp_flags, struct r1conf *conf)
 {
-	struct pool_info *pi = data;
+	int size = offsetof(struct r1bio, bios[conf->raid_disks * 2]);
-	int size = offsetof(struct r1bio, bios[pi->raid_disks]);
 
 	/* allocate a r1bio with room for raid_disks entries in the bios array */
 	return kzalloc(size, gfp_flags);
@@ -145,18 +144,18 @@ static void * r1bio_pool_alloc(gfp_t gfp_flags, void *data)
 
 static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
 {
-	struct pool_info *pi = data;
+	struct r1conf *conf = data;
 	struct r1bio *r1_bio;
 	struct bio *bio;
 	int need_pages;
 	int j;
 	struct resync_pages *rps;
 
-	r1_bio = r1bio_pool_alloc(gfp_flags, pi);
+	r1_bio = r1bio_pool_alloc(gfp_flags, conf);
 	if (!r1_bio)
 		return NULL;
 
-	rps = kmalloc_array(pi->raid_disks, sizeof(struct resync_pages),
+	rps = kmalloc_array(conf->raid_disks * 2, sizeof(struct resync_pages),
 			    gfp_flags);
 	if (!rps)
 		goto out_free_r1bio;
@@ -164,7 +163,7 @@ static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
 	/*
 	 * Allocate bios : 1 for reading, n-1 for writing
 	 */
-	for (j = pi->raid_disks ; j-- ; ) {
+	for (j = conf->raid_disks * 2; j-- ; ) {
 		bio = bio_kmalloc(RESYNC_PAGES, gfp_flags);
 		if (!bio)
 			goto out_free_bio;
@@ -177,11 +176,11 @@ static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
 	 * If this is a user-requested check/repair, allocate
 	 * RESYNC_PAGES for each bio.
 	 */
-	if (test_bit(MD_RECOVERY_REQUESTED, &pi->mddev->recovery))
+	if (test_bit(MD_RECOVERY_REQUESTED, &conf->mddev->recovery))
-		need_pages = pi->raid_disks;
+		need_pages = conf->raid_disks * 2;
 	else
 		need_pages = 1;
-	for (j = 0; j < pi->raid_disks; j++) {
+	for (j = 0; j < conf->raid_disks * 2; j++) {
 		struct resync_pages *rp = &rps[j];
 
 		bio = r1_bio->bios[j];
@@ -207,7 +206,7 @@ out_free_pages:
 		resync_free_pages(&rps[j]);
 
 out_free_bio:
-	while (++j < pi->raid_disks) {
+	while (++j < conf->raid_disks * 2) {
 		bio_uninit(r1_bio->bios[j]);
 		kfree(r1_bio->bios[j]);
 	}
@@ -220,12 +219,12 @@ out_free_r1bio:
 
 static void r1buf_pool_free(void *__r1_bio, void *data)
 {
-	struct pool_info *pi = data;
+	struct r1conf *conf = data;
 	int i;
 	struct r1bio *r1bio = __r1_bio;
 	struct resync_pages *rp = NULL;
 
-	for (i = pi->raid_disks; i--; ) {
+	for (i = conf->raid_disks * 2; i--; ) {
 		rp = get_resync_pages(r1bio->bios[i]);
 		resync_free_pages(rp);
 		bio_uninit(r1bio->bios[i]);
@@ -255,7 +254,7 @@ static void free_r1bio(struct r1bio *r1_bio)
 	struct r1conf *conf = r1_bio->mddev->private;
 
 	put_all_bios(conf, r1_bio);
-	mempool_free(r1_bio, &conf->r1bio_pool);
+	mempool_free(r1_bio, conf->r1bio_pool);
 }
 
 static void put_buf(struct r1bio *r1_bio)
@@ -1305,9 +1304,8 @@ alloc_r1bio(struct mddev *mddev, struct bio *bio)
 	struct r1conf *conf = mddev->private;
 	struct r1bio *r1_bio;
 
-	r1_bio = mempool_alloc(&conf->r1bio_pool, GFP_NOIO);
+	r1_bio = mempool_alloc(conf->r1bio_pool, GFP_NOIO);
-	/* Ensure no bio records IO_BLOCKED */
+	memset(r1_bio, 0, offsetof(struct r1bio, bios[conf->raid_disks * 2]));
-	memset(r1_bio->bios, 0, conf->raid_disks * sizeof(r1_bio->bios[0]));
 	init_r1bio(r1_bio, mddev, bio);
 	return r1_bio;
 }
@@ -2747,7 +2745,7 @@ static int init_resync(struct r1conf *conf)
 	BUG_ON(mempool_initialized(&conf->r1buf_pool));
 
 	return mempool_init(&conf->r1buf_pool, buffs, r1buf_pool_alloc,
-			    r1buf_pool_free, conf->poolinfo);
+			    r1buf_pool_free, conf);
 }
 
 static struct r1bio *raid1_alloc_init_r1buf(struct r1conf *conf)
@@ -2757,7 +2755,7 @@ static struct r1bio *raid1_alloc_init_r1buf(struct r1conf *conf)
 	struct bio *bio;
 	int i;
 
-	for (i = conf->poolinfo->raid_disks; i--; ) {
+	for (i = conf->raid_disks * 2; i--; ) {
 		bio = r1bio->bios[i];
 		rps = bio->bi_private;
 		bio_reset(bio, NULL, 0);
@@ -2822,7 +2820,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
 	}
 
 	if (mddev->bitmap == NULL &&
-	    mddev->recovery_cp == MaxSector &&
+	    mddev->resync_offset == MaxSector &&
 	    !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery) &&
 	    conf->fullsync == 0) {
 		*skipped = 1;
@@ -3085,6 +3083,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
 	int i;
 	struct raid1_info *disk;
 	struct md_rdev *rdev;
+	size_t r1bio_size;
 	int err = -ENOMEM;
 
 	conf = kzalloc(sizeof(struct r1conf), GFP_KERNEL);
@@ -3121,21 +3120,15 @@ static struct r1conf *setup_conf(struct mddev *mddev)
 	if (!conf->tmppage)
 		goto abort;
 
-	conf->poolinfo = kzalloc(sizeof(*conf->poolinfo), GFP_KERNEL);
+	r1bio_size = offsetof(struct r1bio, bios[mddev->raid_disks * 2]);
-	if (!conf->poolinfo)
+	conf->r1bio_pool = mempool_create_kmalloc_pool(NR_RAID_BIOS, r1bio_size);
-		goto abort;
+	if (!conf->r1bio_pool)
-	conf->poolinfo->raid_disks = mddev->raid_disks * 2;
-	err = mempool_init(&conf->r1bio_pool, NR_RAID_BIOS, r1bio_pool_alloc,
-			   rbio_pool_free, conf->poolinfo);
-	if (err)
 		goto abort;
 
 	err = bioset_init(&conf->bio_split, BIO_POOL_SIZE, 0, 0);
 	if (err)
 		goto abort;
 
-	conf->poolinfo->mddev = mddev;
-
 	err = -EINVAL;
 	spin_lock_init(&conf->device_lock);
 	conf->raid_disks = mddev->raid_disks;
@@ -3198,10 +3191,9 @@ static struct r1conf *setup_conf(struct mddev *mddev)
 
  abort:
 	if (conf) {
-		mempool_exit(&conf->r1bio_pool);
+		mempool_destroy(conf->r1bio_pool);
 		kfree(conf->mirrors);
 		safe_put_page(conf->tmppage);
-		kfree(conf->poolinfo);
 		kfree(conf->nr_pending);
 		kfree(conf->nr_waiting);
 		kfree(conf->nr_queued);
@@ -3282,9 +3274,9 @@ static int raid1_run(struct mddev *mddev)
 	}
 
 	if (conf->raid_disks - mddev->degraded == 1)
-		mddev->recovery_cp = MaxSector;
+		mddev->resync_offset = MaxSector;
 
-	if (mddev->recovery_cp != MaxSector)
+	if (mddev->resync_offset != MaxSector)
 		pr_info("md/raid1:%s: not clean -- starting background reconstruction\n",
 			mdname(mddev));
 	pr_info("md/raid1:%s: active with %d out of %d mirrors\n",
@@ -3311,10 +3303,9 @@ static void raid1_free(struct mddev *mddev, void *priv)
 {
 	struct r1conf *conf = priv;
 
-	mempool_exit(&conf->r1bio_pool);
+	mempool_destroy(conf->r1bio_pool);
 	kfree(conf->mirrors);
 	safe_put_page(conf->tmppage);
-	kfree(conf->poolinfo);
 	kfree(conf->nr_pending);
 	kfree(conf->nr_waiting);
 	kfree(conf->nr_queued);
@@ -3345,8 +3336,8 @@ static int raid1_resize(struct mddev *mddev, sector_t sectors)
 
 	md_set_array_sectors(mddev, newsize);
 	if (sectors > mddev->dev_sectors &&
-	    mddev->recovery_cp > mddev->dev_sectors) {
+	    mddev->resync_offset > mddev->dev_sectors) {
-		mddev->recovery_cp = mddev->dev_sectors;
+		mddev->resync_offset = mddev->dev_sectors;
 		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	}
 	mddev->dev_sectors = sectors;
@@ -3367,17 +3358,13 @@ static int raid1_reshape(struct mddev *mddev)
 	 * At the same time, we "pack" the devices so that all the missing
 	 * devices have the higher raid_disk numbers.
 	 */
-	mempool_t newpool, oldpool;
+	mempool_t *newpool, *oldpool;
-	struct pool_info *newpoolinfo;
+	size_t new_r1bio_size;
 	struct raid1_info *newmirrors;
 	struct r1conf *conf = mddev->private;
 	int cnt, raid_disks;
 	unsigned long flags;
 	int d, d2;
-	int ret;
-
-	memset(&newpool, 0, sizeof(newpool));
-	memset(&oldpool, 0, sizeof(oldpool));
 
 	/* Cannot change chunk_size, layout, or level */
 	if (mddev->chunk_sectors != mddev->new_chunk_sectors ||
@@ -3403,24 +3390,16 @@ static int raid1_reshape(struct mddev *mddev)
 			return -EBUSY;
 	}
 
-	newpoolinfo = kmalloc(sizeof(*newpoolinfo), GFP_KERNEL);
+	new_r1bio_size = offsetof(struct r1bio, bios[raid_disks * 2]);
-	if (!newpoolinfo)
+	newpool = mempool_create_kmalloc_pool(NR_RAID_BIOS, new_r1bio_size);
+	if (!newpool) {
 		return -ENOMEM;
-	newpoolinfo->mddev = mddev;
-	newpoolinfo->raid_disks = raid_disks * 2;
-
-	ret = mempool_init(&newpool, NR_RAID_BIOS, r1bio_pool_alloc,
-			   rbio_pool_free, newpoolinfo);
-	if (ret) {
-		kfree(newpoolinfo);
-		return ret;
 	}
 	newmirrors = kzalloc(array3_size(sizeof(struct raid1_info),
 					 raid_disks, 2),
 			     GFP_KERNEL);
 	if (!newmirrors) {
-		kfree(newpoolinfo);
+		mempool_destroy(newpool);
-		mempool_exit(&newpool);
 		return -ENOMEM;
 	}
 
@@ -3429,7 +3408,6 @@ static int raid1_reshape(struct mddev *mddev)
 	/* ok, everything is stopped */
 	oldpool = conf->r1bio_pool;
 	conf->r1bio_pool = newpool;
-	init_waitqueue_head(&conf->r1bio_pool.wait);
 
 	for (d = d2 = 0; d < conf->raid_disks; d++) {
 		struct md_rdev *rdev = conf->mirrors[d].rdev;
@@ -3446,8 +3424,6 @@ static int raid1_reshape(struct mddev *mddev)
 	}
 	kfree(conf->mirrors);
 	conf->mirrors = newmirrors;
-	kfree(conf->poolinfo);
-	conf->poolinfo = newpoolinfo;
 
 	spin_lock_irqsave(&conf->device_lock, flags);
 	mddev->degraded += (raid_disks - conf->raid_disks);
@@ -3461,7 +3437,7 @@ static int raid1_reshape(struct mddev *mddev)
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
 
-	mempool_exit(&oldpool);
+	mempool_destroy(oldpool);
 	return 0;
 }
 

drivers/md/raid1.h

@@ -49,22 +49,6 @@ struct raid1_info {
 	sector_t	seq_start;
 };
 
-/*
- * memory pools need a pointer to the mddev, so they can force an unplug
- * when memory is tight, and a count of the number of drives that the
- * pool was allocated for, so they know how much to allocate and free.
- * mddev->raid_disks cannot be used, as it can change while a pool is active
- * These two datums are stored in a kmalloced struct.
- * The 'raid_disks' here is twice the raid_disks in r1conf.
- * This allows space for each 'real' device can have a replacement in the
- * second half of the array.
- */
-
-struct pool_info {
-	struct mddev *mddev;
-	int	raid_disks;
-};
-
 struct r1conf {
 	struct mddev		*mddev;
 	struct raid1_info	*mirrors;	/* twice 'raid_disks' to
@@ -114,11 +98,7 @@ struct r1conf {
 	 */
 	int			recovery_disabled;
 
-	/* poolinfo contains information about the content of the
+	mempool_t		*r1bio_pool;
-	 * mempools - it changes when the array grows or shrinks
-	 */
-	struct pool_info	*poolinfo;
-	mempool_t		r1bio_pool;
 	mempool_t		r1buf_pool;
 
 	struct bio_set		bio_split;

drivers/md/raid10.c

@@ -2117,7 +2117,7 @@ static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 	int last = conf->geo.raid_disks - 1;
 	struct raid10_info *p;
 
-	if (mddev->recovery_cp < MaxSector)
+	if (mddev->resync_offset < MaxSector)
 		/* only hot-add to in-sync arrays, as recovery is
 		 * very different from resync
 		 */
@@ -3185,7 +3185,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 	 * of a clean array, like RAID1 does.
 	 */
 	if (mddev->bitmap == NULL &&
-	    mddev->recovery_cp == MaxSector &&
+	    mddev->resync_offset == MaxSector &&
 	    mddev->reshape_position == MaxSector &&
 	    !test_bit(MD_RECOVERY_SYNC, &mddev->recovery) &&
 	    !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery) &&
@@ -4145,7 +4145,7 @@ static int raid10_run(struct mddev *mddev)
 		disk->recovery_disabled = mddev->recovery_disabled - 1;
 	}
 
-	if (mddev->recovery_cp != MaxSector)
+	if (mddev->resync_offset != MaxSector)
 		pr_notice("md/raid10:%s: not clean -- starting background reconstruction\n",
 			  mdname(mddev));
 	pr_info("md/raid10:%s: active with %d out of %d devices\n",
@@ -4245,8 +4245,8 @@ static int raid10_resize(struct mddev *mddev, sector_t sectors)
 
 	md_set_array_sectors(mddev, size);
 	if (sectors > mddev->dev_sectors &&
-	    mddev->recovery_cp > oldsize) {
+	    mddev->resync_offset > oldsize) {
-		mddev->recovery_cp = oldsize;
+		mddev->resync_offset = oldsize;
 		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	}
 	calc_sectors(conf, sectors);
@@ -4275,7 +4275,7 @@ static void *raid10_takeover_raid0(struct mddev *mddev, sector_t size, int devs)
 	mddev->delta_disks = mddev->raid_disks;
 	mddev->raid_disks *= 2;
 	/* make sure it will be not marked as dirty */
-	mddev->recovery_cp = MaxSector;
+	mddev->resync_offset = MaxSector;
 	mddev->dev_sectors = size;
 
 	conf = setup_conf(mddev);
@@ -5087,8 +5087,8 @@ static void raid10_finish_reshape(struct mddev *mddev)
 		return;
 
 	if (mddev->delta_disks > 0) {
-		if (mddev->recovery_cp > mddev->resync_max_sectors) {
+		if (mddev->resync_offset > mddev->resync_max_sectors) {
-			mddev->recovery_cp = mddev->resync_max_sectors;
+			mddev->resync_offset = mddev->resync_max_sectors;
 			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 		}
 		mddev->resync_max_sectors = mddev->array_sectors;

drivers/md/raid5-ppl.c

@@ -1163,7 +1163,7 @@ static int ppl_load_distributed(struct ppl_log *log)
 		    le64_to_cpu(pplhdr->generation));
 
 	/* attempt to recover from log if we are starting a dirty array */
-	if (pplhdr && !mddev->pers && mddev->recovery_cp != MaxSector)
+	if (pplhdr && !mddev->pers && mddev->resync_offset != MaxSector)
 		ret = ppl_recover(log, pplhdr, pplhdr_offset);
 
 	/* write empty header if we are starting the array */
@@ -1422,14 +1422,14 @@ int ppl_init_log(struct r5conf *conf)
 
 	if (ret) {
 		goto err;
-	} else if (!mddev->pers && mddev->recovery_cp == 0 &&
+	} else if (!mddev->pers && mddev->resync_offset == 0 &&
 		   ppl_conf->recovered_entries > 0 &&
 		   ppl_conf->mismatch_count == 0) {
 		/*
 		 * If we are starting a dirty array and the recovery succeeds
 		 * without any issues, set the array as clean.
 		 */
-		mddev->recovery_cp = MaxSector;
+		mddev->resync_offset = MaxSector;
 		set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
 	} else if (mddev->pers && ppl_conf->mismatch_count > 0) {
 		/* no mismatch allowed when enabling PPL for a running array */

drivers/md/raid5.c

@@ -3740,7 +3740,7 @@ static int want_replace(struct stripe_head *sh, int disk_idx)
 	    && !test_bit(Faulty, &rdev->flags)
 	    && !test_bit(In_sync, &rdev->flags)
 	    && (rdev->recovery_offset <= sh->sector
-		|| rdev->mddev->recovery_cp <= sh->sector))
+		|| rdev->mddev->resync_offset <= sh->sector))
 		rv = 1;
 	return rv;
 }
@@ -3832,7 +3832,7 @@ static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s,
 	 * is missing/faulty, then we need to read everything we can.
 	 */
 	if (!force_rcw &&
-	    sh->sector < sh->raid_conf->mddev->recovery_cp)
+	    sh->sector < sh->raid_conf->mddev->resync_offset)
 		/* reconstruct-write isn't being forced */
 		return 0;
 	for (i = 0; i < s->failed && i < 2; i++) {
@@ -4097,7 +4097,7 @@ static int handle_stripe_dirtying(struct r5conf *conf,
 				  int disks)
 {
 	int rmw = 0, rcw = 0, i;
-	sector_t recovery_cp = conf->mddev->recovery_cp;
+	sector_t resync_offset = conf->mddev->resync_offset;
 
 	/* Check whether resync is now happening or should start.
 	 * If yes, then the array is dirty (after unclean shutdown or
@@ -4107,14 +4107,14 @@ static int handle_stripe_dirtying(struct r5conf *conf,
 	 * generate correct data from the parity.
 	 */
 	if (conf->rmw_level == PARITY_DISABLE_RMW ||
-	    (recovery_cp < MaxSector && sh->sector >= recovery_cp &&
+	    (resync_offset < MaxSector && sh->sector >= resync_offset &&
 	     s->failed == 0)) {
 		/* Calculate the real rcw later - for now make it
 		 * look like rcw is cheaper
 		 */
 		rcw = 1; rmw = 2;
-		pr_debug("force RCW rmw_level=%u, recovery_cp=%llu sh->sector=%llu\n",
+		pr_debug("force RCW rmw_level=%u, resync_offset=%llu sh->sector=%llu\n",
-			 conf->rmw_level, (unsigned long long)recovery_cp,
+			 conf->rmw_level, (unsigned long long)resync_offset,
 			 (unsigned long long)sh->sector);
 	} else for (i = disks; i--; ) {
 		/* would I have to read this buffer for read_modify_write */
@@ -4770,14 +4770,14 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s)
 	if (test_bit(STRIPE_SYNCING, &sh->state)) {
 		/* If there is a failed device being replaced,
 		 *     we must be recovering.
-		 * else if we are after recovery_cp, we must be syncing
+		 * else if we are after resync_offset, we must be syncing
 		 * else if MD_RECOVERY_REQUESTED is set, we also are syncing.
 		 * else we can only be replacing
 		 * sync and recovery both need to read all devices, and so
 		 * use the same flag.
 		 */
 		if (do_recovery ||
-		    sh->sector >= conf->mddev->recovery_cp ||
+		    sh->sector >= conf->mddev->resync_offset ||
 		    test_bit(MD_RECOVERY_REQUESTED, &(conf->mddev->recovery)))
 			s->syncing = 1;
 		else
@@ -7780,7 +7780,7 @@ static int raid5_run(struct mddev *mddev)
 	int first = 1;
 	int ret = -EIO;
 
-	if (mddev->recovery_cp != MaxSector)
+	if (mddev->resync_offset != MaxSector)
 		pr_notice("md/raid:%s: not clean -- starting background reconstruction\n",
 			  mdname(mddev));
 
@@ -7921,7 +7921,7 @@ static int raid5_run(struct mddev *mddev)
 				mdname(mddev));
 			mddev->ro = 1;
 			set_disk_ro(mddev->gendisk, 1);
-		} else if (mddev->recovery_cp == MaxSector)
+		} else if (mddev->resync_offset == MaxSector)
 			set_bit(MD_JOURNAL_CLEAN, &mddev->flags);
 	}
 
@@ -7988,7 +7988,7 @@ static int raid5_run(struct mddev *mddev)
 	mddev->resync_max_sectors = mddev->dev_sectors;
 
 	if (mddev->degraded > dirty_parity_disks &&
-	    mddev->recovery_cp != MaxSector) {
+	    mddev->resync_offset != MaxSector) {
 		if (test_bit(MD_HAS_PPL, &mddev->flags))
 			pr_crit("md/raid:%s: starting dirty degraded array with PPL.\n",
 				mdname(mddev));
@@ -8328,8 +8328,8 @@ static int raid5_resize(struct mddev *mddev, sector_t sectors)
 
 	md_set_array_sectors(mddev, newsize);
 	if (sectors > mddev->dev_sectors &&
-	    mddev->recovery_cp > mddev->dev_sectors) {
+	    mddev->resync_offset > mddev->dev_sectors) {
-		mddev->recovery_cp = mddev->dev_sectors;
+		mddev->resync_offset = mddev->dev_sectors;
 		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	}
 	mddev->dev_sectors = sectors;
@@ -8423,7 +8423,7 @@ static int raid5_start_reshape(struct mddev *mddev)
 		return -EINVAL;
 
 	/* raid5 can't handle concurrent reshape and recovery */
-	if (mddev->recovery_cp < MaxSector)
+	if (mddev->resync_offset < MaxSector)
 		return -EBUSY;
 	for (i = 0; i < conf->raid_disks; i++)
 		if (conf->disks[i].replacement)
@@ -8648,7 +8648,7 @@ static void *raid45_takeover_raid0(struct mddev *mddev, int level)
 	mddev->raid_disks += 1;
 	mddev->delta_disks = 1;
 	/* make sure it will be not marked as dirty */
-	mddev->recovery_cp = MaxSector;
+	mddev->resync_offset = MaxSector;
 
 	return setup_conf(mddev);
 }

drivers/nvme/host/auth.c

@@ -742,7 +742,7 @@ static int nvme_auth_secure_concat(struct nvme_ctrl *ctrl,
 			 "%s: qid %d failed to generate digest, error %d\n",
 			 __func__, chap->qid, ret);
 		goto out_free_psk;
-	};
+	}
 	dev_dbg(ctrl->device, "%s: generated digest %s\n",
 		 __func__, digest);
 	ret = nvme_auth_derive_tls_psk(chap->hash_id, psk, psk_len,
@@ -752,7 +752,7 @@ static int nvme_auth_secure_concat(struct nvme_ctrl *ctrl,
 			 "%s: qid %d failed to derive TLS psk, error %d\n",
 			 __func__, chap->qid, ret);
 		goto out_free_digest;
-	};
+	}
 
 	tls_key = nvme_tls_psk_refresh(ctrl->opts->keyring,
 				       ctrl->opts->host->nqn,

drivers/nvme/host/core.c

@@ -3158,6 +3158,11 @@ static inline bool nvme_discovery_ctrl(struct nvme_ctrl *ctrl)
 	return ctrl->opts && ctrl->opts->discovery_nqn;
 }
 
+static inline bool nvme_admin_ctrl(struct nvme_ctrl *ctrl)
+{
+	return ctrl->cntrltype == NVME_CTRL_ADMIN;
+}
+
 static bool nvme_validate_cntlid(struct nvme_subsystem *subsys,
 		struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
 {
@@ -3670,6 +3675,17 @@ int nvme_init_ctrl_finish(struct nvme_ctrl *ctrl, bool was_suspended)
 	if (ret)
 		return ret;
 
+	if (nvme_admin_ctrl(ctrl)) {
+		/*
+		 * An admin controller has one admin queue, but no I/O queues.
+		 * Override queue_count so it only creates an admin queue.
+		 */
+		dev_dbg(ctrl->device,
+			"Subsystem %s is an administrative controller",
+			ctrl->subsys->subnqn);
+		ctrl->queue_count = 1;
+	}
+
 	ret = nvme_configure_apst(ctrl);
 	if (ret < 0)
 		return ret;

drivers/nvme/host/fc.c

@@ -1363,7 +1363,7 @@ nvme_fc_disconnect_assoc_done(struct nvmefc_ls_req *lsreq, int status)
  * down, and the related FC-NVME Association ID and Connection IDs
  * become invalid.
  *
- * The behavior of the fc-nvme initiator is such that it's
+ * The behavior of the fc-nvme initiator is such that its
  * understanding of the association and connections will implicitly
  * be torn down. The action is implicit as it may be due to a loss of
  * connectivity with the fc-nvme target, so you may never get a
@@ -2777,7 +2777,7 @@ nvme_fc_queue_rq(struct blk_mq_hw_ctx *hctx,
 	 * as WRITE ZEROES will return a non-zero rq payload_bytes yet
 	 * there is no actual payload to be transferred.
 	 * To get it right, key data transmission on there being 1 or
-	 * more physical segments in the sg list. If there is no
+	 * more physical segments in the sg list. If there are no
 	 * physical segments, there is no payload.
 	 */
 	if (blk_rq_nr_phys_segments(rq)) {

drivers/nvme/host/pci.c

@@ -935,7 +935,7 @@ static blk_status_t nvme_pci_setup_data_sgl(struct request *req,
 
 	nvme_pci_sgl_set_seg(&iod->cmd.common.dptr.sgl, sgl_dma, mapped);
 	if (unlikely(iter->status))
-		nvme_free_sgls(req);
+		nvme_unmap_data(req);
 	return iter->status;
 }
 

drivers/nvme/host/tcp.c

@@ -2179,7 +2179,7 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new)
 
 	/*
 	 * Only start IO queues for which we have allocated the tagset
-	 * and limitted it to the available queues. On reconnects, the
+	 * and limited it to the available queues. On reconnects, the
 	 * queue number might have changed.
 	 */
 	nr_queues = min(ctrl->tagset->nr_hw_queues + 1, ctrl->queue_count);

drivers/nvme/target/core.c

@@ -1960,24 +1960,24 @@ static int __init nvmet_init(void)
 	if (!nvmet_wq)
 		goto out_free_buffered_work_queue;
 
-	error = nvmet_init_discovery();
+	error = nvmet_init_debugfs();
 	if (error)
 		goto out_free_nvmet_work_queue;
 
-	error = nvmet_init_debugfs();
+	error = nvmet_init_discovery();
-	if (error)
-		goto out_exit_discovery;
-
-	error = nvmet_init_configfs();
 	if (error)
 		goto out_exit_debugfs;
 
+	error = nvmet_init_configfs();
+	if (error)
+		goto out_exit_discovery;
+
 	return 0;
 
-out_exit_debugfs:
-	nvmet_exit_debugfs();
 out_exit_discovery:
 	nvmet_exit_discovery();
+out_exit_debugfs:
+	nvmet_exit_debugfs();
 out_free_nvmet_work_queue:
 	destroy_workqueue(nvmet_wq);
 out_free_buffered_work_queue:
@@ -1992,8 +1992,8 @@ out_destroy_bvec_cache:
 static void __exit nvmet_exit(void)
 {
 	nvmet_exit_configfs();
-	nvmet_exit_debugfs();
 	nvmet_exit_discovery();
+	nvmet_exit_debugfs();
 	ida_destroy(&cntlid_ida);
 	destroy_workqueue(nvmet_wq);
 	destroy_workqueue(buffered_io_wq);

drivers/nvme/target/fc.c

@@ -459,7 +459,7 @@ nvmet_fc_disconnect_assoc_done(struct nvmefc_ls_req *lsreq, int status)
  * down, and the related FC-NVME Association ID and Connection IDs
  * become invalid.
  *
- * The behavior of the fc-nvme target is such that it's
+ * The behavior of the fc-nvme target is such that its
  * understanding of the association and connections will implicitly
  * be torn down. The action is implicit as it may be due to a loss of
  * connectivity with the fc-nvme host, so the target may never get a
@@ -2313,7 +2313,7 @@ nvmet_fc_transfer_fcp_data(struct nvmet_fc_tgtport *tgtport,
 	ret = tgtport->ops->fcp_op(&tgtport->fc_target_port, fod->fcpreq);
 	if (ret) {
 		/*
-		 * should be ok to set w/o lock as its in the thread of
+		 * should be ok to set w/o lock as it's in the thread of
 		 * execution (not an async timer routine) and doesn't
 		 * contend with any clearing action
 		 */
@@ -2629,7 +2629,7 @@ transport_error:
  * and the api of the FC LLDD which may issue a hw command to send the
  * response, but the LLDD may not get the hw completion for that command
  * and upcall the nvmet_fc layer before a new command may be
- * asynchronously received - its possible for a command to be received
+ * asynchronously received - it's possible for a command to be received
  * before the LLDD and nvmet_fc have recycled the job structure. It gives
  * the appearance of more commands received than fits in the sq.
  * To alleviate this scenario, a temporary queue is maintained in the

drivers/nvme/target/passthru.c

@@ -533,6 +533,8 @@ u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req)
 		case NVME_FEAT_HOST_ID:
 			req->execute = nvmet_execute_get_features;
 			return NVME_SC_SUCCESS;
+		case NVME_FEAT_FDP:
+			return nvmet_setup_passthru_command(req);
 		default:
 			return nvmet_passthru_get_set_features(req);
 		}

drivers/nvme/target/rdma.c

@@ -1731,7 +1731,7 @@ static void nvmet_rdma_queue_connect_fail(struct rdma_cm_id *cm_id,
  * We registered an ib_client to handle device removal for queues,
  * so we only need to handle the listening port cm_ids. In this case
  * we nullify the priv to prevent double cm_id destruction and destroying
- * the cm_id implicitely by returning a non-zero rc to the callout.
+ * the cm_id implicitly by returning a non-zero rc to the callout.
  */
 static int nvmet_rdma_device_removal(struct rdma_cm_id *cm_id,
 		struct nvmet_rdma_queue *queue)
@@ -1742,7 +1742,7 @@ static int nvmet_rdma_device_removal(struct rdma_cm_id *cm_id,
 		/*
 		 * This is a queue cm_id. we have registered
 		 * an ib_client to handle queues removal
-		 * so don't interfear and just return.
+		 * so don't interfere and just return.
 		 */
 		return 0;
 	}
@@ -1760,7 +1760,7 @@ static int nvmet_rdma_device_removal(struct rdma_cm_id *cm_id,
 
 	/*
 	 * We need to return 1 so that the core will destroy
-	 * it's own ID.  What a great API design..
+	 * its own ID.  What a great API design..
 	 */
 	return 1;
 }

include/linux/ioprio.h

@@ -60,7 +60,8 @@ static inline int __get_task_ioprio(struct task_struct *p)
 	int prio;
 
 	if (!ioc)
-		return IOPRIO_DEFAULT;
+		return IOPRIO_PRIO_VALUE(task_nice_ioclass(p),
+					 task_nice_ioprio(p));
 
 	if (p != current)
 		lockdep_assert_held(&p->alloc_lock);

include/linux/sbitmap.h

@@ -209,23 +209,6 @@ void sbitmap_resize(struct sbitmap *sb, unsigned int depth);
  */
 int sbitmap_get(struct sbitmap *sb);
 
-/**
- * sbitmap_get_shallow() - Try to allocate a free bit from a &struct sbitmap,
- * limiting the depth used from each word.
- * @sb: Bitmap to allocate from.
- * @shallow_depth: The maximum number of bits to allocate from a single word.
- *
- * This rather specific operation allows for having multiple users with
- * different allocation limits. E.g., there can be a high-priority class that
- * uses sbitmap_get() and a low-priority class that uses sbitmap_get_shallow()
- * with a @shallow_depth of (1 << (@sb->shift - 1)). Then, the low-priority
- * class can only allocate half of the total bits in the bitmap, preventing it
- * from starving out the high-priority class.
- *
- * Return: Non-negative allocated bit number if successful, -1 otherwise.
- */
-int sbitmap_get_shallow(struct sbitmap *sb, unsigned long shallow_depth);
-
 /**
  * sbitmap_any_bit_set() - Check for a set bit in a &struct sbitmap.
  * @sb: Bitmap to check.
@@ -478,7 +461,7 @@ unsigned long __sbitmap_queue_get_batch(struct sbitmap_queue *sbq, int nr_tags,
  * sbitmap_queue, limiting the depth used from each word, with preemption
  * already disabled.
  * @sbq: Bitmap queue to allocate from.
- * @shallow_depth: The maximum number of bits to allocate from a single word.
+ * @shallow_depth: The maximum number of bits to allocate from the queue.
  * See sbitmap_get_shallow().
  *
  * If you call this, make sure to call sbitmap_queue_min_shallow_depth() after

include/uapi/linux/raid/md_p.h

@@ -173,7 +173,7 @@ typedef struct mdp_superblock_s {
 #else
 #error unspecified endianness
 #endif
-	__u32 recovery_cp;	/* 11 recovery checkpoint sector count	      */
+	__u32 resync_offset;	/* 11 resync checkpoint sector count	      */
 	/* There are only valid for minor_version > 90 */
 	__u64 reshape_position;	/* 12,13 next address in array-space for reshape */
 	__u32 new_level;	/* 14 new level we are reshaping to	      */

lib/sbitmap.c

@@ -208,8 +208,28 @@ static int sbitmap_find_bit_in_word(struct sbitmap_word *map,
 	return nr;
 }
 
+static unsigned int __map_depth_with_shallow(const struct sbitmap *sb,
+					     int index,
+					     unsigned int shallow_depth)
+{
+	u64 shallow_word_depth;
+	unsigned int word_depth, reminder;
+
+	word_depth = __map_depth(sb, index);
+	if (shallow_depth >= sb->depth)
+		return word_depth;
+
+	shallow_word_depth = word_depth * shallow_depth;
+	reminder = do_div(shallow_word_depth, sb->depth);
+
+	if (reminder >= (index + 1) * word_depth)
+		shallow_word_depth++;
+
+	return (unsigned int)shallow_word_depth;
+}
+
 static int sbitmap_find_bit(struct sbitmap *sb,
-			    unsigned int depth,
+			    unsigned int shallow_depth,
 			    unsigned int index,
 			    unsigned int alloc_hint,
 			    bool wrap)
@@ -218,12 +238,12 @@ static int sbitmap_find_bit(struct sbitmap *sb,
 	int nr = -1;
 
 	for (i = 0; i < sb->map_nr; i++) {
-		nr = sbitmap_find_bit_in_word(&sb->map[index],
+		unsigned int depth = __map_depth_with_shallow(sb, index,
-					      min_t(unsigned int,
+							      shallow_depth);
-						    __map_depth(sb, index),
-						    depth),
-					      alloc_hint, wrap);
 
+		if (depth)
+			nr = sbitmap_find_bit_in_word(&sb->map[index], depth,
+						      alloc_hint, wrap);
 		if (nr != -1) {
 			nr += index << sb->shift;
 			break;
@@ -287,7 +307,22 @@ static int __sbitmap_get_shallow(struct sbitmap *sb,
 	return sbitmap_find_bit(sb, shallow_depth, index, alloc_hint, true);
 }
 
-int sbitmap_get_shallow(struct sbitmap *sb, unsigned long shallow_depth)
+/**
+ * sbitmap_get_shallow() - Try to allocate a free bit from a &struct sbitmap,
+ * limiting the depth used from each word.
+ * @sb: Bitmap to allocate from.
+ * @shallow_depth: The maximum number of bits to allocate from the bitmap.
+ *
+ * This rather specific operation allows for having multiple users with
+ * different allocation limits. E.g., there can be a high-priority class that
+ * uses sbitmap_get() and a low-priority class that uses sbitmap_get_shallow()
+ * with a @shallow_depth of (sb->depth >> 1). Then, the low-priority
+ * class can only allocate half of the total bits in the bitmap, preventing it
+ * from starving out the high-priority class.
+ *
+ * Return: Non-negative allocated bit number if successful, -1 otherwise.
+ */
+static int sbitmap_get_shallow(struct sbitmap *sb, unsigned long shallow_depth)
 {
 	int nr;
 	unsigned int hint, depth;
@@ -302,7 +337,6 @@ int sbitmap_get_shallow(struct sbitmap *sb, unsigned long shallow_depth)
 
 	return nr;
 }
-EXPORT_SYMBOL_GPL(sbitmap_get_shallow);
 
 bool sbitmap_any_bit_set(const struct sbitmap *sb)
 {
@@ -406,27 +440,9 @@ EXPORT_SYMBOL_GPL(sbitmap_bitmap_show);
 static unsigned int sbq_calc_wake_batch(struct sbitmap_queue *sbq,
 					unsigned int depth)
 {
-	unsigned int wake_batch;
+	return clamp_t(unsigned int,
-	unsigned int shallow_depth;
+		       min(depth, sbq->min_shallow_depth) / SBQ_WAIT_QUEUES,
-
+		       1, SBQ_WAKE_BATCH);
-	/*
-	 * Each full word of the bitmap has bits_per_word bits, and there might
-	 * be a partial word. There are depth / bits_per_word full words and
-	 * depth % bits_per_word bits left over. In bitwise arithmetic:
-	 *
-	 * bits_per_word = 1 << shift
-	 * depth / bits_per_word = depth >> shift
-	 * depth % bits_per_word = depth & ((1 << shift) - 1)
-	 *
-	 * Each word can be limited to sbq->min_shallow_depth bits.
-	 */
-	shallow_depth = min(1U << sbq->sb.shift, sbq->min_shallow_depth);
-	depth = ((depth >> sbq->sb.shift) * shallow_depth +
-		 min(depth & ((1U << sbq->sb.shift) - 1), shallow_depth));
-	wake_batch = clamp_t(unsigned int, depth / SBQ_WAIT_QUEUES, 1,
-			     SBQ_WAKE_BATCH);
-
-	return wake_batch;
 }
 
 int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth,