Merge tag 'for-5.10/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mike Snitzer:

 - Fix DM's bio splitting changes that were made during v5.9. This
   restores splitting in terms of varied per-target ti->max_io_len
   rather than use block core's single stacked 'chunk_sectors' limit.

 - Like DM crypt, update DM integrity to not use crypto drivers that
   have CRYPTO_ALG_ALLOCATES_MEMORY set.

 - Fix DM writecache target's argument parsing and status display.

 - Remove needless BUG() from dm writecache's persistent_memory_claim()

 - Remove old gcc workaround in DM cache target's block_div() for ARM
   link errors now that gcc >= 4.9 is required.

 - Fix RCU locking in dm_blk_report_zones and dm_dax_zero_page_range.

 - Remove old, and now frowned upon, BUG_ON(in_interrupt()) in
   dm_table_event().

 - Remove invalid sparse annotations from dm_prepare_ioctl() and
   dm_unprepare_ioctl().

* tag 'for-5.10/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm: remove invalid sparse __acquires and __releases annotations
  dm: fix double RCU unlock in dm_dax_zero_page_range() error path
  dm: fix IO splitting
  dm writecache: remove BUG() and fail gracefully instead
  dm table: Remove BUG_ON(in_interrupt())
  dm: fix bug with RCU locking in dm_blk_report_zones
  Revert "dm cache: fix arm link errors with inline"
  dm writecache: fix the maximum number of arguments
  dm writecache: advance the number of arguments when reporting max_age
  dm integrity: don't use drivers that have CRYPTO_ALG_ALLOCATES_MEMORY
diff --git a/block/blk-merge.c b/block/blk-merge.c
index bcf5e45..97b7c28 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -144,7 +144,7 @@ static struct bio *blk_bio_write_same_split(struct request_queue *q,
 static inline unsigned get_max_io_size(struct request_queue *q,
 				       struct bio *bio)
 {
-	unsigned sectors = blk_max_size_offset(q, bio->bi_iter.bi_sector);
+	unsigned sectors = blk_max_size_offset(q, bio->bi_iter.bi_sector, 0);
 	unsigned max_sectors = sectors;
 	unsigned pbs = queue_physical_block_size(q) >> SECTOR_SHIFT;
 	unsigned lbs = queue_logical_block_size(q) >> SECTOR_SHIFT;
diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c
index 9644424..4bc453f 100644
--- a/drivers/md/dm-cache-target.c
+++ b/drivers/md/dm-cache-target.c
@@ -712,10 +712,6 @@ static bool block_size_is_power_of_two(struct cache *cache)
 	return cache->sectors_per_block_shift >= 0;
 }
 
-/* gcc on ARM generates spurious references to __udivdi3 and __umoddi3 */
-#if defined(CONFIG_ARM) && __GNUC__ == 4 && __GNUC_MINOR__ <= 6
-__always_inline
-#endif
 static dm_block_t block_div(dm_block_t b, uint32_t n)
 {
 	do_div(b, n);
diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c
index 3fc3757..5a7a1b9 100644
--- a/drivers/md/dm-integrity.c
+++ b/drivers/md/dm-integrity.c
@@ -3462,7 +3462,7 @@ static int get_mac(struct crypto_shash **hash, struct alg_spec *a, char **error,
 	int r;
 
 	if (a->alg_string) {
-		*hash = crypto_alloc_shash(a->alg_string, 0, 0);
+		*hash = crypto_alloc_shash(a->alg_string, 0, CRYPTO_ALG_ALLOCATES_MEMORY);
 		if (IS_ERR(*hash)) {
 			*error = error_alg;
 			r = PTR_ERR(*hash);
@@ -3519,7 +3519,7 @@ static int create_journal(struct dm_integrity_c *ic, char **error)
 		struct journal_completion comp;
 
 		comp.ic = ic;
-		ic->journal_crypt = crypto_alloc_skcipher(ic->journal_crypt_alg.alg_string, 0, 0);
+		ic->journal_crypt = crypto_alloc_skcipher(ic->journal_crypt_alg.alg_string, 0, CRYPTO_ALG_ALLOCATES_MEMORY);
 		if (IS_ERR(ic->journal_crypt)) {
 			*error = "Invalid journal cipher";
 			r = PTR_ERR(ic->journal_crypt);
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index ce543b7..7eeb7c4 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -18,7 +18,6 @@
 #include <linux/mutex.h>
 #include <linux/delay.h>
 #include <linux/atomic.h>
-#include <linux/lcm.h>
 #include <linux/blk-mq.h>
 #include <linux/mount.h>
 #include <linux/dax.h>
@@ -1247,12 +1246,6 @@ void dm_table_event_callback(struct dm_table *t,
 
 void dm_table_event(struct dm_table *t)
 {
-	/*
-	 * You can no longer call dm_table_event() from interrupt
-	 * context, use a bottom half instead.
-	 */
-	BUG_ON(in_interrupt());
-
 	mutex_lock(&_event_lock);
 	if (t->event_fn)
 		t->event_fn(t->event_context);
@@ -1455,10 +1448,6 @@ int dm_calculate_queue_limits(struct dm_table *table,
 			zone_sectors = ti_limits.chunk_sectors;
 		}
 
-		/* Stack chunk_sectors if target-specific splitting is required */
-		if (ti->max_io_len)
-			ti_limits.chunk_sectors = lcm_not_zero(ti->max_io_len,
-							       ti_limits.chunk_sectors);
 		/* Set I/O hints portion of queue limits */
 		if (ti->type->io_hints)
 			ti->type->io_hints(ti, &ti_limits);
diff --git a/drivers/md/dm-writecache.c b/drivers/md/dm-writecache.c
index 9ae4ce7..d5223a0 100644
--- a/drivers/md/dm-writecache.c
+++ b/drivers/md/dm-writecache.c
@@ -319,7 +319,7 @@ static int persistent_memory_claim(struct dm_writecache *wc)
 #else
 static int persistent_memory_claim(struct dm_writecache *wc)
 {
-	BUG();
+	return -EOPNOTSUPP;
 }
 #endif
 
@@ -2041,7 +2041,7 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	struct wc_memory_superblock s;
 
 	static struct dm_arg _args[] = {
-		{0, 10, "Invalid number of feature args"},
+		{0, 16, "Invalid number of feature args"},
 	};
 
 	as.argc = argc;
@@ -2479,6 +2479,8 @@ static void writecache_status(struct dm_target *ti, status_type_t type,
 			extra_args += 2;
 		if (wc->autocommit_time_set)
 			extra_args += 2;
+		if (wc->max_age != MAX_AGE_UNSPECIFIED)
+			extra_args += 2;
 		if (wc->cleaner)
 			extra_args++;
 		if (wc->writeback_fua_set)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index c18fc25..4e0cbfe 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -476,8 +476,10 @@ static int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
 		return -EAGAIN;
 
 	map = dm_get_live_table(md, &srcu_idx);
-	if (!map)
-		return -EIO;
+	if (!map) {
+		ret = -EIO;
+		goto out;
+	}
 
 	do {
 		struct dm_target *tgt;
@@ -507,7 +509,6 @@ static int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
 
 static int dm_prepare_ioctl(struct mapped_device *md, int *srcu_idx,
 			    struct block_device **bdev)
-	__acquires(md->io_barrier)
 {
 	struct dm_target *tgt;
 	struct dm_table *map;
@@ -541,7 +542,6 @@ static int dm_prepare_ioctl(struct mapped_device *md, int *srcu_idx,
 }
 
 static void dm_unprepare_ioctl(struct mapped_device *md, int srcu_idx)
-	__releases(md->io_barrier)
 {
 	dm_put_live_table(md, srcu_idx);
 }
@@ -1037,15 +1037,18 @@ static sector_t max_io_len(struct dm_target *ti, sector_t sector)
 	sector_t max_len;
 
 	/*
-	 * Does the target need to split even further?
-	 * - q->limits.chunk_sectors reflects ti->max_io_len so
-	 *   blk_max_size_offset() provides required splitting.
-	 * - blk_max_size_offset() also respects q->limits.max_sectors
+	 * Does the target need to split IO even further?
+	 * - varied (per target) IO splitting is a tenet of DM; this
+	 *   explains why stacked chunk_sectors based splitting via
+	 *   blk_max_size_offset() isn't possible here. So pass in
+	 *   ti->max_io_len to override stacked chunk_sectors.
 	 */
-	max_len = blk_max_size_offset(ti->table->md->queue,
-				      target_offset);
-	if (len > max_len)
-		len = max_len;
+	if (ti->max_io_len) {
+		max_len = blk_max_size_offset(ti->table->md->queue,
+					      target_offset, ti->max_io_len);
+		if (len > max_len)
+			len = max_len;
+	}
 
 	return len;
 }
@@ -1196,11 +1199,9 @@ static int dm_dax_zero_page_range(struct dax_device *dax_dev, pgoff_t pgoff,
 		 * ->zero_page_range() is mandatory dax operation. If we are
 		 *  here, something is wrong.
 		 */
-		dm_put_live_table(md, srcu_idx);
 		goto out;
 	}
 	ret = ti->type->dax_zero_page_range(ti, pgoff, nr_pages);
-
  out:
 	dm_put_live_table(md, srcu_idx);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 639cae2..24ae504 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1073,11 +1073,12 @@ static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
  * file system requests.
  */
 static inline unsigned int blk_max_size_offset(struct request_queue *q,
-					       sector_t offset)
+					       sector_t offset,
+					       unsigned int chunk_sectors)
 {
-	unsigned int chunk_sectors = q->limits.chunk_sectors;
-
-	if (!chunk_sectors)
+	if (!chunk_sectors && q->limits.chunk_sectors)
+		chunk_sectors = q->limits.chunk_sectors;
+	else
 		return q->limits.max_sectors;
 
 	if (likely(is_power_of_2(chunk_sectors)))
@@ -1101,7 +1102,7 @@ static inline unsigned int blk_rq_get_max_sectors(struct request *rq,
 	    req_op(rq) == REQ_OP_SECURE_ERASE)
 		return blk_queue_get_max_sectors(q, req_op(rq));
 
-	return min(blk_max_size_offset(q, offset),
+	return min(blk_max_size_offset(q, offset, 0),
 			blk_queue_get_max_sectors(q, req_op(rq)));
 }