Merge tag 'for-5.19/block-2022-05-22' of git://git.kernel.dk/linux-block

Pull block updates from Jens Axboe:
 "Here are the core block changes for 5.19. This contains:

   - blk-throttle accounting fix (Laibin)

   - Series removing redundant assignments (Michal)

   - Expose bio cache via the bio_set, so that DM can use it (Mike)

   - Finish off the bio allocation interface cleanups by dealing with
     the weirdest member of the family. bio_kmalloc combines a kmalloc
     for the bio and bio_vecs with a hidden bio_init call and magic
     cleanup semantics (Christoph)

   - Clean up the block layer API so that APIs consumed by file systems
     are (almost) only struct block_device based, so that file systems
     don't have to poke into block layer internals like the
     request_queue (Christoph)

   - Clean up the blk_execute_rq* API (Christoph)

   - Clean up various lose end in the blk-cgroup code to make it easier
     to follow in preparation of reworking the blkcg assignment for bios
     (Christoph)

   - Fix use-after-free issues in BFQ when processes with merged queues
     get moved to different cgroups (Jan)

   - BFQ fixes (Jan)

   - Various fixes and cleanups (Bart, Chengming, Fanjun, Julia, Ming,
     Wolfgang, me)"

* tag 'for-5.19/block-2022-05-22' of git://git.kernel.dk/linux-block: (83 commits)
  blk-mq: fix typo in comment
  bfq: Remove bfq_requeue_request_body()
  bfq: Remove superfluous conversion from RQ_BIC()
  bfq: Allow current waker to defend against a tentative one
  bfq: Relax waker detection for shared queues
  blk-cgroup: delete rcu_read_lock_held() WARN_ON_ONCE()
  blk-throttle: Set BIO_THROTTLED when bio has been throttled
  blk-cgroup: Remove unnecessary rcu_read_lock/unlock()
  blk-cgroup: always terminate io.stat lines
  block, bfq: make bfq_has_work() more accurate
  block, bfq: protect 'bfqd->queued' by 'bfqd->lock'
  block: cleanup the VM accounting in submit_bio
  block: Fix the bio.bi_opf comment
  block: reorder the REQ_ flags
  blk-iocost: combine local_stat and desc_stat to stat
  block: improve the error message from bio_check_eod
  block: allow passing a NULL bdev to bio_alloc_clone/bio_init_clone
  block: remove superfluous calls to blkcg_bio_issue_init
  kthread: unexport kthread_blkcg
  blk-cgroup: cleanup blkcg_maybe_throttle_current
  ...

21 files changed, 71 insertions, 150 deletions

diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c
index 097577ae3c47..ce13c272c387 100644
--- a/drivers/md/bcache/alloc.c
+++ b/drivers/md/bcache/alloc.c
@@ -336,7 +336,7 @@ static int bch_allocator_thread(void *arg)
 				mutex_unlock(&ca->set->bucket_lock);
 				blkdev_issue_discard(ca->bdev,
 					bucket_to_sector(ca->set, bucket),
-					ca->sb.bucket_size, GFP_KERNEL, 0);
+					ca->sb.bucket_size, GFP_KERNEL);
 				mutex_lock(&ca->set->bucket_lock);
 			}
 
diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c
index 6230dfdd9286..7510d1c983a5 100644
--- a/drivers/md/bcache/debug.c
+++ b/drivers/md/bcache/debug.c
@@ -107,15 +107,16 @@ void bch_btree_verify(struct btree *b)
 
 void bch_data_verify(struct cached_dev *dc, struct bio *bio)
 {
+	unsigned int nr_segs = bio_segments(bio);
 	struct bio *check;
 	struct bio_vec bv, cbv;
 	struct bvec_iter iter, citer = { 0 };
 
-	check = bio_kmalloc(GFP_NOIO, bio_segments(bio));
+	check = bio_kmalloc(nr_segs, GFP_NOIO);
 	if (!check)
 		return;
-	bio_set_dev(check, bio->bi_bdev);
-	check->bi_opf = REQ_OP_READ;
+	bio_init(check, bio->bi_bdev, check->bi_inline_vecs, nr_segs,
+		 REQ_OP_READ);
 	check->bi_iter.bi_sector = bio->bi_iter.bi_sector;
 	check->bi_iter.bi_size = bio->bi_iter.bi_size;
 
@@ -146,7 +147,8 @@ void bch_data_verify(struct cached_dev *dc, struct bio *bio)
 
 	bio_free_pages(check);
 out_put:
-	bio_put(check);
+	bio_uninit(check);
+	kfree(check);
 }
 
 #endif
diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
index 320fcdfef48e..9c5dde73da88 100644
--- a/drivers/md/bcache/request.c
+++ b/drivers/md/bcache/request.c
@@ -1005,7 +1005,7 @@ static void cached_dev_write(struct cached_dev *dc, struct search *s)
 		bio_get(s->iop.bio);
 
 		if (bio_op(bio) == REQ_OP_DISCARD &&
-		    !blk_queue_discard(bdev_get_queue(dc->bdev)))
+		    !bdev_max_discard_sectors(dc->bdev))
 			goto insert_data;
 
 		/* I/O request sent to backing device */
@@ -1115,7 +1115,7 @@ static void detached_dev_do_request(struct bcache_device *d, struct bio *bio,
 	bio->bi_private = ddip;
 
 	if ((bio_op(bio) == REQ_OP_DISCARD) &&
-	    !blk_queue_discard(bdev_get_queue(dc->bdev)))
+	    !bdev_max_discard_sectors(dc->bdev))
 		bio->bi_end_io(bio);
 	else
 		submit_bio_noacct(bio);
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index bf3de149d3c9..2f49e31142f6 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -973,7 +973,6 @@ static int bcache_device_init(struct bcache_device *d, unsigned int block_size,
 
 	blk_queue_flag_set(QUEUE_FLAG_NONROT, d->disk->queue);
 	blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, d->disk->queue);
-	blk_queue_flag_set(QUEUE_FLAG_DISCARD, d->disk->queue);
 
 	blk_queue_write_cache(q, true, true);
 
@@ -2350,7 +2349,7 @@ static int register_cache(struct cache_sb *sb, struct cache_sb_disk *sb_disk,
 	ca->bdev->bd_holder = ca;
 	ca->sb_disk = sb_disk;
 
-	if (blk_queue_discard(bdev_get_queue(bdev)))
+	if (bdev_max_discard_sectors((bdev)))
 		ca->discard = CACHE_DISCARD(&ca->sb);
 
 	ret = cache_alloc(ca);
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c
index d1029d71ff3b..c6f677059214 100644
--- a/drivers/md/bcache/sysfs.c
+++ b/drivers/md/bcache/sysfs.c
@@ -1151,7 +1151,7 @@ STORE(__bch_cache)
 	if (attr == &sysfs_discard) {
 		bool v = strtoul_or_return(buf);
 
-		if (blk_queue_discard(bdev_get_queue(ca->bdev)))
+		if (bdev_max_discard_sectors(ca->bdev))
 			ca->discard = v;
 
 		if (v != CACHE_DISCARD(&ca->sb)) {
diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c
index e9cbc70d5a0e..5ffa1dcf84cf 100644
--- a/drivers/md/dm-bufio.c
+++ b/drivers/md/dm-bufio.c
@@ -611,7 +611,8 @@ static void bio_complete(struct bio *bio)
 {
 	struct dm_buffer *b = bio->bi_private;
 	blk_status_t status = bio->bi_status;
-	bio_put(bio);
+	bio_uninit(bio);
+	kfree(bio);
 	b->end_io(b, status);
 }
 
@@ -626,16 +627,14 @@ static void use_bio(struct dm_buffer *b, int rw, sector_t sector,
 	if (unlikely(b->c->sectors_per_block_bits < PAGE_SHIFT - SECTOR_SHIFT))
 		vec_size += 2;
 
-	bio = bio_kmalloc(GFP_NOWAIT | __GFP_NORETRY | __GFP_NOWARN, vec_size);
+	bio = bio_kmalloc(vec_size, GFP_NOWAIT | __GFP_NORETRY | __GFP_NOWARN);
 	if (!bio) {
 dmio:
 		use_dmio(b, rw, sector, n_sectors, offset);
 		return;
 	}
-
+	bio_init(bio, b->c->bdev, bio->bi_inline_vecs, vec_size, rw);
 	bio->bi_iter.bi_sector = sector;
-	bio_set_dev(bio, b->c->bdev);
-	bio_set_op_attrs(bio, rw, 0);
 	bio->bi_end_io = bio_complete;
 	bio->bi_private = b;
 
diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c
index 780a61bc6cc0..28c5de8eca4a 100644
--- a/drivers/md/dm-cache-target.c
+++ b/drivers/md/dm-cache-target.c
@@ -3329,13 +3329,6 @@ static int cache_iterate_devices(struct dm_target *ti,
 	return r;
 }
 
-static bool origin_dev_supports_discard(struct block_device *origin_bdev)
-{
-	struct request_queue *q = bdev_get_queue(origin_bdev);
-
-	return blk_queue_discard(q);
-}
-
 /*
  * If discard_passdown was enabled verify that the origin device
  * supports discards.  Disable discard_passdown if not.
@@ -3349,7 +3342,7 @@ static void disable_passdown_if_not_supported(struct cache *cache)
 	if (!cache->features.discard_passdown)
 		return;
 
-	if (!origin_dev_supports_discard(origin_bdev))
+	if (!bdev_max_discard_sectors(origin_bdev))
 		reason = "discard unsupported";
 
 	else if (origin_limits->max_discard_sectors < cache->sectors_per_block)
diff --git a/drivers/md/dm-clone-target.c b/drivers/md/dm-clone-target.c
index 128316a73d01..811b0a5379d0 100644
--- a/drivers/md/dm-clone-target.c
+++ b/drivers/md/dm-clone-target.c
@@ -2016,13 +2016,6 @@ static void clone_resume(struct dm_target *ti)
 	do_waker(&clone->waker.work);
 }
 
-static bool bdev_supports_discards(struct block_device *bdev)
-{
-	struct request_queue *q = bdev_get_queue(bdev);
-
-	return (q && blk_queue_discard(q));
-}
-
 /*
  * If discard_passdown was enabled verify that the destination device supports
  * discards. Disable discard_passdown if not.
@@ -2036,7 +2029,7 @@ static void disable_passdown_if_not_supported(struct clone *clone)
 	if (!test_bit(DM_CLONE_DISCARD_PASSDOWN, &clone->flags))
 		return;
 
-	if (!bdev_supports_discards(dest_dev))
+	if (!bdev_max_discard_sectors(dest_dev))
 		reason = "discard unsupported";
 	else if (dest_limits->max_discard_sectors < clone->region_size)
 		reason = "max discard sectors smaller than a region";
diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index 5762366333a2..e4b95eaeec8c 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -311,7 +311,7 @@ static void do_region(int op, int op_flags, unsigned region,
 	 * Reject unsupported discard and write same requests.
 	 */
 	if (op == REQ_OP_DISCARD)
-		special_cmd_max_sectors = q->limits.max_discard_sectors;
+		special_cmd_max_sectors = bdev_max_discard_sectors(where->bdev);
 	else if (op == REQ_OP_WRITE_ZEROES)
 		special_cmd_max_sectors = q->limits.max_write_zeroes_sectors;
 	if ((op == REQ_OP_DISCARD || op == REQ_OP_WRITE_ZEROES) &&
diff --git a/drivers/md/dm-log-writes.c b/drivers/md/dm-log-writes.c
index c9d036d6bb2e..e194226c89e5 100644
--- a/drivers/md/dm-log-writes.c
+++ b/drivers/md/dm-log-writes.c
@@ -866,9 +866,8 @@ static int log_writes_message(struct dm_target *ti, unsigned argc, char **argv,
 static void log_writes_io_hints(struct dm_target *ti, struct queue_limits *limits)
 {
 	struct log_writes_c *lc = ti->private;
-	struct request_queue *q = bdev_get_queue(lc->dev->bdev);
 
-	if (!q || !blk_queue_discard(q)) {
+	if (!bdev_max_discard_sectors(lc->dev->bdev)) {
 		lc->device_supports_discard = false;
 		limits->discard_granularity = lc->sectorsize;
 		limits->max_discard_sectors = (UINT_MAX >> SECTOR_SHIFT);
diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
index 2b26435a6946..9526ccbedafb 100644
--- a/drivers/md/dm-raid.c
+++ b/drivers/md/dm-raid.c
@@ -2963,13 +2963,8 @@ static void configure_discard_support(struct raid_set *rs)
 	raid456 = rs_is_raid456(rs);
 
 	for (i = 0; i < rs->raid_disks; i++) {
-		struct request_queue *q;
-
-		if (!rs->dev[i].rdev.bdev)
-			continue;
-
-		q = bdev_get_queue(rs->dev[i].rdev.bdev);
-		if (!q || !blk_queue_discard(q))
+		if (!rs->dev[i].rdev.bdev ||
+		    !bdev_max_discard_sectors(rs->dev[i].rdev.bdev))
 			return;
 
 		if (raid456) {
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 03541cfc2317..e7d42f6335a2 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1820,9 +1820,7 @@ static int device_dax_write_cache_enabled(struct dm_target *ti,
 static int device_is_rotational(struct dm_target *ti, struct dm_dev *dev,
 				sector_t start, sector_t len, void *data)
 {
-	struct request_queue *q = bdev_get_queue(dev->bdev);
-
-	return !blk_queue_nonrot(q);
+	return !bdev_nonrot(dev->bdev);
 }
 
 static int device_is_not_random(struct dm_target *ti, struct dm_dev *dev,
@@ -1890,9 +1888,7 @@ static bool dm_table_supports_nowait(struct dm_table *t)
 static int device_not_discard_capable(struct dm_target *ti, struct dm_dev *dev,
 				      sector_t start, sector_t len, void *data)
 {
-	struct request_queue *q = bdev_get_queue(dev->bdev);
-
-	return !blk_queue_discard(q);
+	return !bdev_max_discard_sectors(dev->bdev);
 }
 
 static bool dm_table_supports_discards(struct dm_table *t)
@@ -1924,9 +1920,7 @@ static int device_not_secure_erase_capable(struct dm_target *ti,
 					   struct dm_dev *dev, sector_t start,
 					   sector_t len, void *data)
 {
-	struct request_queue *q = bdev_get_queue(dev->bdev);
-
-	return !blk_queue_secure_erase(q);
+	return !bdev_max_secure_erase_sectors(dev->bdev);
 }
 
 static bool dm_table_supports_secure_erase(struct dm_table *t)
@@ -1952,9 +1946,7 @@ static int device_requires_stable_pages(struct dm_target *ti,
 					struct dm_dev *dev, sector_t start,
 					sector_t len, void *data)
 {
-	struct request_queue *q = bdev_get_queue(dev->bdev);
-
-	return blk_queue_stable_writes(q);
+	return bdev_stable_writes(dev->bdev);
 }
 
 int dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
@@ -1974,18 +1966,15 @@ int dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 		blk_queue_flag_clear(QUEUE_FLAG_NOWAIT, q);
 
 	if (!dm_table_supports_discards(t)) {
-		blk_queue_flag_clear(QUEUE_FLAG_DISCARD, q);
-		/* Must also clear discard limits... */
 		q->limits.max_discard_sectors = 0;
 		q->limits.max_hw_discard_sectors = 0;
 		q->limits.discard_granularity = 0;
 		q->limits.discard_alignment = 0;
 		q->limits.discard_misaligned = 0;
-	} else
-		blk_queue_flag_set(QUEUE_FLAG_DISCARD, q);
+	}
 
-	if (dm_table_supports_secure_erase(t))
-		blk_queue_flag_set(QUEUE_FLAG_SECERASE, q);
+	if (!dm_table_supports_secure_erase(t))
+		q->limits.max_secure_erase_sectors = 0;
 
 	if (dm_table_supports_flush(t, (1UL << QUEUE_FLAG_WC))) {
 		wc = true;
diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c
index 4d25d0e27031..84c083f76673 100644
--- a/drivers/md/dm-thin.c
+++ b/drivers/md/dm-thin.c
@@ -398,8 +398,8 @@ static int issue_discard(struct discard_op *op, dm_block_t data_b, dm_block_t da
 	sector_t s = block_to_sectors(tc->pool, data_b);
 	sector_t len = block_to_sectors(tc->pool, data_e - data_b);
 
-	return __blkdev_issue_discard(tc->pool_dev->bdev, s, len,
-				      GFP_NOWAIT, 0, &op->bio);
+	return __blkdev_issue_discard(tc->pool_dev->bdev, s, len, GFP_NOWAIT,
+				      &op->bio);
 }
 
 static void end_discard(struct discard_op *op, int r)
@@ -2802,13 +2802,6 @@ static void requeue_bios(struct pool *pool)
 /*----------------------------------------------------------------
  * Binding of control targets to a pool object
  *--------------------------------------------------------------*/
-static bool data_dev_supports_discard(struct pool_c *pt)
-{
-	struct request_queue *q = bdev_get_queue(pt->data_dev->bdev);
-
-	return blk_queue_discard(q);
-}
-
 static bool is_factor(sector_t block_size, uint32_t n)
 {
 	return !sector_div(block_size, n);
@@ -2828,7 +2821,7 @@ static void disable_passdown_if_not_supported(struct pool_c *pt)
 	if (!pt->adjusted_pf.discard_passdown)
 		return;
 
-	if (!data_dev_supports_discard(pt))
+	if (!bdev_max_discard_sectors(pt->data_dev->bdev))
 		reason = "discard unsupported";
 
 	else if (data_limits->max_discard_sectors < pool->sectors_per_block)
@@ -4057,8 +4050,6 @@ static void pool_io_hints(struct dm_target *ti, struct queue_limits *limits)
 		/*
 		 * Must explicitly disallow stacking discard limits otherwise the
 		 * block layer will stack them if pool's data device has support.
-		 * QUEUE_FLAG_DISCARD wouldn't be set but there is no way for the
-		 * user to see that, so make sure to set all discard limits to 0.
 		 */
 		limits->discard_granularity = 0;
 		return;
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 82957bd460e8..39081338ca61 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -955,7 +955,6 @@ void disable_discard(struct mapped_device *md)
 
 	/* device doesn't really support DISCARD, disable it */
 	limits->max_discard_sectors = 0;
-	blk_queue_flag_clear(QUEUE_FLAG_DISCARD, md->queue);
 }
 
 void disable_write_zeroes(struct mapped_device *md)
@@ -982,7 +981,7 @@ static void clone_endio(struct bio *bio)
 
 	if (unlikely(error == BLK_STS_TARGET)) {
 		if (bio_op(bio) == REQ_OP_DISCARD &&
-		    !q->limits.max_discard_sectors)
+		    !bdev_max_discard_sectors(bio->bi_bdev))
 			disable_discard(md);
 		else if (bio_op(bio) == REQ_OP_WRITE_ZEROES &&
 			 !q->limits.max_write_zeroes_sectors)
diff --git a/drivers/md/md-linear.c b/drivers/md/md-linear.c
index 0f55b079371b..138a3b25c5c8 100644
--- a/drivers/md/md-linear.c
+++ b/drivers/md/md-linear.c
@@ -64,7 +64,6 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 	struct linear_conf *conf;
 	struct md_rdev *rdev;
 	int i, cnt;
-	bool discard_supported = false;
 
 	conf = kzalloc(struct_size(conf, disks, raid_disks), GFP_KERNEL);
 	if (!conf)
@@ -96,9 +95,6 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 
 		conf->array_sectors += rdev->sectors;
 		cnt++;
-
-		if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
-			discard_supported = true;
 	}
 	if (cnt != raid_disks) {
 		pr_warn("md/linear:%s: not enough drives present. Aborting!\n",
@@ -106,11 +102,6 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 		goto out;
 	}
 
-	if (!discard_supported)
-		blk_queue_flag_clear(QUEUE_FLAG_DISCARD, mddev->queue);
-	else
-		blk_queue_flag_set(QUEUE_FLAG_DISCARD, mddev->queue);
-
 	/*
 	 * Here we calculate the device offsets.
 	 */
@@ -252,7 +243,7 @@ static bool linear_make_request(struct mddev *mddev, struct bio *bio)
 		start_sector + data_offset;
 
 	if (unlikely((bio_op(bio) == REQ_OP_DISCARD) &&
-		     !blk_queue_discard(bio->bi_bdev->bd_disk->queue))) {
+		     !bdev_max_discard_sectors(bio->bi_bdev))) {
 		/* Just ignore it */
 		bio_endio(bio);
 	} else {
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 309b3af906ad..2587f872c088 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -5991,8 +5991,7 @@ int md_run(struct mddev *mddev)
 		bool nonrot = true;
 
 		rdev_for_each(rdev, mddev) {
-			if (rdev->raid_disk >= 0 &&
-			    !blk_queue_nonrot(bdev_get_queue(rdev->bdev))) {
+			if (rdev->raid_disk >= 0 && !bdev_nonrot(rdev->bdev)) {
 				nonrot = false;
 				break;
 			}
@@ -8585,7 +8584,7 @@ void md_submit_discard_bio(struct mddev *mddev, struct md_rdev *rdev,
 {
 	struct bio *discard_bio = NULL;
 
-	if (__blkdev_issue_discard(rdev->bdev, start, size, GFP_NOIO, 0,
+	if (__blkdev_issue_discard(rdev->bdev, start, size, GFP_NOIO,
 			&discard_bio) || !discard_bio)
 		return;
 
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index b21e101183f4..7231f5e1eaa7 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -399,7 +399,6 @@ static int raid0_run(struct mddev *mddev)
 	conf = mddev->private;
 	if (mddev->queue) {
 		struct md_rdev *rdev;
-		bool discard_supported = false;
 
 		blk_queue_max_hw_sectors(mddev->queue, mddev->chunk_sectors);
 		blk_queue_max_write_zeroes_sectors(mddev->queue, mddev->chunk_sectors);
@@ -412,13 +411,7 @@ static int raid0_run(struct mddev *mddev)
 		rdev_for_each(rdev, mddev) {
 			disk_stack_limits(mddev->gendisk, rdev->bdev,
 					  rdev->data_offset << 9);
-			if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
-				discard_supported = true;
 		}
-		if (!discard_supported)
-			blk_queue_flag_clear(QUEUE_FLAG_DISCARD, mddev->queue);
-		else
-			blk_queue_flag_set(QUEUE_FLAG_DISCARD, mddev->queue);
 	}
 
 	/* calculate array device size */
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 99d5464a51f8..5aed2c8b746e 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -165,9 +165,10 @@ 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-- ; ) {
-		bio = bio_kmalloc(gfp_flags, RESYNC_PAGES);
+		bio = bio_kmalloc(RESYNC_PAGES, gfp_flags);
 		if (!bio)
 			goto out_free_bio;
+		bio_init(bio, NULL, bio->bi_inline_vecs, RESYNC_PAGES, 0);
 		r1_bio->bios[j] = bio;
 	}
 	/*
@@ -206,8 +207,10 @@ out_free_pages:
 		resync_free_pages(&rps[j]);
 
 out_free_bio:
-	while (++j < pi->raid_disks)
-		bio_put(r1_bio->bios[j]);
+	while (++j < pi->raid_disks) {
+		bio_uninit(r1_bio->bios[j]);
+		kfree(r1_bio->bios[j]);
+	}
 	kfree(rps);
 
 out_free_r1bio:
@@ -225,7 +228,8 @@ static void r1buf_pool_free(void *__r1_bio, void *data)
 	for (i = pi->raid_disks; i--; ) {
 		rp = get_resync_pages(r1bio->bios[i]);
 		resync_free_pages(rp);
-		bio_put(r1bio->bios[i]);
+		bio_uninit(r1bio->bios[i]);
+		kfree(r1bio->bios[i]);
 	}
 
 	/* resync pages array stored in the 1st bio's .bi_private */
@@ -704,7 +708,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
 			/* At least two disks to choose from so failfast is OK */
 			set_bit(R1BIO_FailFast, &r1_bio->state);
 
-		nonrot = blk_queue_nonrot(bdev_get_queue(rdev->bdev));
+		nonrot = bdev_nonrot(rdev->bdev);
 		has_nonrot_disk |= nonrot;
 		pending = atomic_read(&rdev->nr_pending);
 		dist = abs(this_sector - conf->mirrors[disk].head_position);
@@ -802,7 +806,7 @@ static void flush_bio_list(struct r1conf *conf, struct bio *bio)
 		if (test_bit(Faulty, &rdev->flags)) {
 			bio_io_error(bio);
 		} else if (unlikely((bio_op(bio) == REQ_OP_DISCARD) &&
-				    !blk_queue_discard(bio->bi_bdev->bd_disk->queue)))
+				    !bdev_max_discard_sectors(bio->bi_bdev)))
 			/* Just ignore it */
 			bio_endio(bio);
 		else
@@ -1826,8 +1830,6 @@ static int raid1_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 			break;
 		}
 	}
-	if (mddev->queue && blk_queue_discard(bdev_get_queue(rdev->bdev)))
-		blk_queue_flag_set(QUEUE_FLAG_DISCARD, mddev->queue);
 	print_conf(conf);
 	return err;
 }
@@ -3106,7 +3108,6 @@ static int raid1_run(struct mddev *mddev)
 	int i;
 	struct md_rdev *rdev;
 	int ret;
-	bool discard_supported = false;
 
 	if (mddev->level != 1) {
 		pr_warn("md/raid1:%s: raid level not set to mirroring (%d)\n",
@@ -3141,8 +3142,6 @@ static int raid1_run(struct mddev *mddev)
 			continue;
 		disk_stack_limits(mddev->gendisk, rdev->bdev,
 				  rdev->data_offset << 9);
-		if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
-			discard_supported = true;
 	}
 
 	mddev->degraded = 0;
@@ -3179,15 +3178,6 @@ static int raid1_run(struct mddev *mddev)
 
 	md_set_array_sectors(mddev, raid1_size(mddev, 0, 0));
 
-	if (mddev->queue) {
-		if (discard_supported)
-			blk_queue_flag_set(QUEUE_FLAG_DISCARD,
-						mddev->queue);
-		else
-			blk_queue_flag_clear(QUEUE_FLAG_DISCARD,
-						  mddev->queue);
-	}
-
 	ret = md_integrity_register(mddev);
 	if (ret) {
 		md_unregister_thread(&mddev->thread);
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index dfe7d62d3fbd..834eb3ba95a6 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -145,15 +145,17 @@ static void * r10buf_pool_alloc(gfp_t gfp_flags, void *data)
 	 * Allocate bios.
 	 */
 	for (j = nalloc ; j-- ; ) {
-		bio = bio_kmalloc(gfp_flags, RESYNC_PAGES);
+		bio = bio_kmalloc(RESYNC_PAGES, gfp_flags);
 		if (!bio)
 			goto out_free_bio;
+		bio_init(bio, NULL, bio->bi_inline_vecs, RESYNC_PAGES, 0);
 		r10_bio->devs[j].bio = bio;
 		if (!conf->have_replacement)
 			continue;
-		bio = bio_kmalloc(gfp_flags, RESYNC_PAGES);
+		bio = bio_kmalloc(RESYNC_PAGES, gfp_flags);
 		if (!bio)
 			goto out_free_bio;
+		bio_init(bio, NULL, bio->bi_inline_vecs, RESYNC_PAGES, 0);
 		r10_bio->devs[j].repl_bio = bio;
 	}
 	/*
@@ -197,9 +199,11 @@ out_free_pages:
 out_free_bio:
 	for ( ; j < nalloc; j++) {
 		if (r10_bio->devs[j].bio)
-			bio_put(r10_bio->devs[j].bio);
+			bio_uninit(r10_bio->devs[j].bio);
+		kfree(r10_bio->devs[j].bio);
 		if (r10_bio->devs[j].repl_bio)
-			bio_put(r10_bio->devs[j].repl_bio);
+			bio_uninit(r10_bio->devs[j].repl_bio);
+		kfree(r10_bio->devs[j].repl_bio);
 	}
 	kfree(rps);
 out_free_r10bio:
@@ -220,12 +224,15 @@ static void r10buf_pool_free(void *__r10_bio, void *data)
 		if (bio) {
 			rp = get_resync_pages(bio);
 			resync_free_pages(rp);
-			bio_put(bio);
+			bio_uninit(bio);
+			kfree(bio);
 		}
 
 		bio = r10bio->devs[j].repl_bio;
-		if (bio)
-			bio_put(bio);
+		if (bio) {
+			bio_uninit(bio);
+			kfree(bio);
+		}
 	}
 
 	/* resync pages array stored in the 1st bio's .bi_private */
@@ -796,7 +803,7 @@ static struct md_rdev *read_balance(struct r10conf *conf,
 		if (!do_balance)
 			break;
 
-		nonrot = blk_queue_nonrot(bdev_get_queue(rdev->bdev));
+		nonrot = bdev_nonrot(rdev->bdev);
 		has_nonrot_disk |= nonrot;
 		pending = atomic_read(&rdev->nr_pending);
 		if (min_pending > pending && nonrot) {
@@ -888,7 +895,7 @@ static void flush_pending_writes(struct r10conf *conf)
 			if (test_bit(Faulty, &rdev->flags)) {
 				bio_io_error(bio);
 			} else if (unlikely((bio_op(bio) ==  REQ_OP_DISCARD) &&
-					    !blk_queue_discard(bio->bi_bdev->bd_disk->queue)))
+					    !bdev_max_discard_sectors(bio->bi_bdev)))
 				/* Just ignore it */
 				bio_endio(bio);
 			else
@@ -1083,7 +1090,7 @@ static void raid10_unplug(struct blk_plug_cb *cb, bool from_schedule)
 		if (test_bit(Faulty, &rdev->flags)) {
 			bio_io_error(bio);
 		} else if (unlikely((bio_op(bio) ==  REQ_OP_DISCARD) &&
-				    !blk_queue_discard(bio->bi_bdev->bd_disk->queue)))
+				    !bdev_max_discard_sectors(bio->bi_bdev)))
 			/* Just ignore it */
 			bio_endio(bio);
 		else
@@ -2144,8 +2151,6 @@ static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 		rcu_assign_pointer(p->rdev, rdev);
 		break;
 	}
-	if (mddev->queue && blk_queue_discard(bdev_get_queue(rdev->bdev)))
-		blk_queue_flag_set(QUEUE_FLAG_DISCARD, mddev->queue);
 
 	print_conf(conf);
 	return err;
@@ -4069,7 +4074,6 @@ static int raid10_run(struct mddev *mddev)
 	sector_t size;
 	sector_t min_offset_diff = 0;
 	int first = 1;
-	bool discard_supported = false;
 
 	if (mddev_init_writes_pending(mddev) < 0)
 		return -ENOMEM;
@@ -4140,20 +4144,9 @@ static int raid10_run(struct mddev *mddev)
 					  rdev->data_offset << 9);
 
 		disk->head_position = 0;
-
-		if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
-			discard_supported = true;
 		first = 0;
 	}
 
-	if (mddev->queue) {
-		if (discard_supported)
-			blk_queue_flag_set(QUEUE_FLAG_DISCARD,
-						mddev->queue);
-		else
-			blk_queue_flag_clear(QUEUE_FLAG_DISCARD,
-						  mddev->queue);
-	}
 	/* need to check that every block has at least one working mirror */
 	if (!enough(conf, -1)) {
 		pr_err("md/raid10:%s: not enough operational mirrors.\n",
diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c
index a7d50ff9020a..094a4042589e 100644
--- a/drivers/md/raid5-cache.c
+++ b/drivers/md/raid5-cache.c
@@ -1318,7 +1318,7 @@ static void r5l_write_super_and_discard_space(struct r5l_log *log,
 
 	r5l_write_super(log, end);
 
-	if (!blk_queue_discard(bdev_get_queue(bdev)))
+	if (!bdev_max_discard_sectors(bdev))
 		return;
 
 	mddev = log->rdev->mddev;
@@ -1344,14 +1344,14 @@ static void r5l_write_super_and_discard_space(struct r5l_log *log,
 	if (log->last_checkpoint < end) {
 		blkdev_issue_discard(bdev,
 				log->last_checkpoint + log->rdev->data_offset,
-				end - log->last_checkpoint, GFP_NOIO, 0);
+				end - log->last_checkpoint, GFP_NOIO);
 	} else {
 		blkdev_issue_discard(bdev,
 				log->last_checkpoint + log->rdev->data_offset,
 				log->device_size - log->last_checkpoint,
-				GFP_NOIO, 0);
+				GFP_NOIO);
 		blkdev_issue_discard(bdev, log->rdev->data_offset, end,
-				GFP_NOIO, 0);
+				GFP_NOIO);
 	}
 }
 
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 351d341a1ffa..59f91e392a2a 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -7242,7 +7242,7 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	rdev_for_each(rdev, mddev) {
 		if (test_bit(Journal, &rdev->flags))
 			continue;
-		if (blk_queue_nonrot(bdev_get_queue(rdev->bdev))) {
+		if (bdev_nonrot(rdev->bdev)) {
 			conf->batch_bio_dispatch = false;
 			break;
 		}
@@ -7776,14 +7776,10 @@ static int raid5_run(struct mddev *mddev)
 		 * A better idea might be to turn DISCARD into WRITE_ZEROES
 		 * requests, as that is required to be safe.
 		 */
-		if (devices_handle_discard_safely &&
-		    mddev->queue->limits.max_discard_sectors >= (stripe >> 9) &&
-		    mddev->queue->limits.discard_granularity >= stripe)
-			blk_queue_flag_set(QUEUE_FLAG_DISCARD,
-						mddev->queue);
-		else
-			blk_queue_flag_clear(QUEUE_FLAG_DISCARD,
-						mddev->queue);
+		if (!devices_handle_discard_safely ||
+		    mddev->queue->limits.max_discard_sectors < (stripe >> 9) ||
+		    mddev->queue->limits.discard_granularity < stripe)
+			blk_queue_max_discard_sectors(mddev->queue, 0);
 
 		blk_queue_max_hw_sectors(mddev->queue, UINT_MAX);
 	}