Merge tag 'for-5.5/dm-changes' of...

Merge tag 'for-5.5/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper updates from Mike Snitzer:

 - Fix DM core to disallow stacking request-based DM on partitions.

 - Fix DM raid target to properly resync raidset even if bitmap needed
   additional pages.

 - Fix DM crypt performance regression due to use of WQ_HIGHPRI for the
   IO and crypt workqueues.

 - Fix DM integrity metadata layout that was aligned on 128K boundary
   rather than the intended 4K boundary (removes 124K of wasted space
   for each metadata block).

 - Improve the DM thin, cache and clone targets to use spin_lock_irq
   rather than spin_lock_irqsave where possible.

 - Fix DM thin single thread performance that was lost due to needless
   workqueue wakeups.

 - Fix DM zoned target performance that was lost due to excessive
   backing device checks.

 - Add ability to trigger write failure with the DM dust test target.

 - Fix whitespace indentation in drivers/md/Kconfig.

 - Various smalls fixes and cleanups (e.g. use struct_size, fix
   uninitialized variable, variable renames, etc).

* tag 'for-5.5/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (22 commits)
  Revert "dm crypt: use WQ_HIGHPRI for the IO and crypt workqueues"
  dm: Fix Kconfig indentation
  dm thin: wakeup worker only when deferred bios exist
  dm integrity: fix excessive alignment of metadata runs
  dm raid: Remove unnecessary negation of a shift in raid10_format_to_md_layout
  dm zoned: reduce overhead of backing device checks
  dm dust: add limited write failure mode
  dm dust: change ret to r in dust_map_read and dust_map
  dm dust: change result vars to r
  dm cache: replace spin_lock_irqsave with spin_lock_irq
  dm bio prison: replace spin_lock_irqsave with spin_lock_irq
  dm thin: replace spin_lock_irqsave with spin_lock_irq
  dm clone: add bucket_lock_irq/bucket_unlock_irq helpers
  dm clone: replace spin_lock_irqsave with spin_lock_irq
  dm writecache: handle REQ_FUA
  dm writecache: fix uninitialized variable warning
  dm stripe: use struct_size() in kmalloc()
  dm raid: streamline rs_get_progress() and its raid_status() caller side
  dm raid: simplify rs_setup_recovery call chain
  dm raid: to ensure resynchronization, perform raid set grow in preresume
  ...

Documentation/admin-guide/device-mapper/dm-integrity.rst

@@ -177,6 +177,11 @@ bitmap_flush_interval:number
 	The bitmap flush interval in milliseconds. The metadata buffers
 	are synchronized when this interval expires.
 
+fix_padding
+	Use a smaller padding of the tag area that is more
+	space-efficient. If this option is not present, large padding is
+	used - that is for compatibility with older kernels.
+
 
 The journal mode (D/J), buffer_sectors, journal_watermark, commit_time can
 be changed when reloading the target (load an inactive table and swap the

Documentation/admin-guide/device-mapper/dm-raid.rst

@@ -417,3 +417,5 @@ Version History
 	deadlock/potential data corruption.  Update superblock when
 	specific devices are requested via rebuild.  Fix RAID leg
 	rebuild errors.
+ 1.15.0 Fix size extensions not being synchronized in case of new MD bitmap
+        pages allocated;  also fix those not occuring after previous reductions

drivers/md/Kconfig

@@ -38,9 +38,9 @@ config MD_AUTODETECT
 	default y
 	---help---
 	  If you say Y here, then the kernel will try to autodetect raid
-	  arrays as part of its boot process. 
+	  arrays as part of its boot process.
 
-	  If you don't use raid and say Y, this autodetection can cause 
+	  If you don't use raid and say Y, this autodetection can cause
 	  a several-second delay in the boot time due to various
 	  synchronisation steps that are part of this step.
 
@@ -290,7 +290,7 @@ config DM_SNAPSHOT
        depends on BLK_DEV_DM
        select DM_BUFIO
        ---help---
-         Allow volume managers to take writable snapshots of a device.
+	 Allow volume managers to take writable snapshots of a device.
 
 config DM_THIN_PROVISIONING
        tristate "Thin provisioning target"
@@ -298,7 +298,7 @@ config DM_THIN_PROVISIONING
        select DM_PERSISTENT_DATA
        select DM_BIO_PRISON
        ---help---
-         Provides thin provisioning and snapshots that share a data store.
+	 Provides thin provisioning and snapshots that share a data store.
 
 config DM_CACHE
        tristate "Cache target (EXPERIMENTAL)"
@@ -307,23 +307,23 @@ config DM_CACHE
        select DM_PERSISTENT_DATA
        select DM_BIO_PRISON
        ---help---
-         dm-cache attempts to improve performance of a block device by
-         moving frequently used data to a smaller, higher performance
-         device.  Different 'policy' plugins can be used to change the
-         algorithms used to select which blocks are promoted, demoted,
-         cleaned etc.  It supports writeback and writethrough modes.
+	 dm-cache attempts to improve performance of a block device by
+	 moving frequently used data to a smaller, higher performance
+	 device.  Different 'policy' plugins can be used to change the
+	 algorithms used to select which blocks are promoted, demoted,
+	 cleaned etc.  It supports writeback and writethrough modes.
 
 config DM_CACHE_SMQ
        tristate "Stochastic MQ Cache Policy (EXPERIMENTAL)"
        depends on DM_CACHE
        default y
        ---help---
-         A cache policy that uses a multiqueue ordered by recent hits
-         to select which blocks should be promoted and demoted.
-         This is meant to be a general purpose policy.  It prioritises
-         reads over writes.  This SMQ policy (vs MQ) offers the promise
-         of less memory utilization, improved performance and increased
-         adaptability in the face of changing workloads.
+	 A cache policy that uses a multiqueue ordered by recent hits
+	 to select which blocks should be promoted and demoted.
+	 This is meant to be a general purpose policy.  It prioritises
+	 reads over writes.  This SMQ policy (vs MQ) offers the promise
+	 of less memory utilization, improved performance and increased
+	 adaptability in the face of changing workloads.
 
 config DM_WRITECACHE
 	tristate "Writecache target"
@@ -343,9 +343,9 @@ config DM_ERA
        select DM_PERSISTENT_DATA
        select DM_BIO_PRISON
        ---help---
-         dm-era tracks which parts of a block device are written to
-         over time.  Useful for maintaining cache coherency when using
-         vendor snapshots.
+	 dm-era tracks which parts of a block device are written to
+	 over time.  Useful for maintaining cache coherency when using
+	 vendor snapshots.
 
 config DM_CLONE
        tristate "Clone target (EXPERIMENTAL)"
@@ -353,20 +353,20 @@ config DM_CLONE
        default n
        select DM_PERSISTENT_DATA
        ---help---
-         dm-clone produces a one-to-one copy of an existing, read-only source
-         device into a writable destination device. The cloned device is
-         visible/mountable immediately and the copy of the source device to the
-         destination device happens in the background, in parallel with user
-         I/O.
+	 dm-clone produces a one-to-one copy of an existing, read-only source
+	 device into a writable destination device. The cloned device is
+	 visible/mountable immediately and the copy of the source device to the
+	 destination device happens in the background, in parallel with user
+	 I/O.
 
-         If unsure, say N.
+	 If unsure, say N.
 
 config DM_MIRROR
        tristate "Mirror target"
        depends on BLK_DEV_DM
        ---help---
-         Allow volume managers to mirror logical volumes, also
-         needed for live data migration tools such as 'pvmove'.
+	 Allow volume managers to mirror logical volumes, also
+	 needed for live data migration tools such as 'pvmove'.
 
 config DM_LOG_USERSPACE
 	tristate "Mirror userspace logging"
@@ -483,7 +483,7 @@ config DM_FLAKEY
        tristate "Flakey target"
        depends on BLK_DEV_DM
        ---help---
-         A target that intermittently fails I/O for debugging purposes.
+	 A target that intermittently fails I/O for debugging purposes.
 
 config DM_VERITY
 	tristate "Verity target support"

drivers/md/dm-bio-prison-v1.c

@@ -150,11 +150,10 @@ static int bio_detain(struct dm_bio_prison *prison,
 		      struct dm_bio_prison_cell **cell_result)
 {
 	int r;
-	unsigned long flags;
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	r = __bio_detain(prison, key, inmate, cell_prealloc, cell_result);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 
 	return r;
 }
@@ -198,11 +197,9 @@ void dm_cell_release(struct dm_bio_prison *prison,
 		     struct dm_bio_prison_cell *cell,
 		     struct bio_list *bios)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	__cell_release(prison, cell, bios);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 }
 EXPORT_SYMBOL_GPL(dm_cell_release);
 
@@ -250,12 +247,10 @@ void dm_cell_visit_release(struct dm_bio_prison *prison,
 			   void *context,
 			   struct dm_bio_prison_cell *cell)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	visit_fn(context, cell);
 	rb_erase(&cell->node, &prison->cells);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 }
 EXPORT_SYMBOL_GPL(dm_cell_visit_release);
 
@@ -275,11 +270,10 @@ int dm_cell_promote_or_release(struct dm_bio_prison *prison,
 			       struct dm_bio_prison_cell *cell)
 {
 	int r;
-	unsigned long flags;
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	r = __promote_or_release(prison, cell);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 
 	return r;
 }
@@ -379,10 +373,9 @@ EXPORT_SYMBOL_GPL(dm_deferred_entry_dec);
 int dm_deferred_set_add_work(struct dm_deferred_set *ds, struct list_head *work)
 {
 	int r = 1;
-	unsigned long flags;
 	unsigned next_entry;
 
-	spin_lock_irqsave(&ds->lock, flags);
+	spin_lock_irq(&ds->lock);
 	if ((ds->sweeper == ds->current_entry) &&
 	    !ds->entries[ds->current_entry].count)
 		r = 0;
@@ -392,7 +385,7 @@ int dm_deferred_set_add_work(struct dm_deferred_set *ds, struct list_head *work)
 		if (!ds->entries[next_entry].count)
 			ds->current_entry = next_entry;
 	}
-	spin_unlock_irqrestore(&ds->lock, flags);
+	spin_unlock_irq(&ds->lock);
 
 	return r;
 }

drivers/md/dm-bio-prison-v2.c

@@ -177,11 +177,10 @@ bool dm_cell_get_v2(struct dm_bio_prison_v2 *prison,
 		    struct dm_bio_prison_cell_v2 **cell_result)
 {
 	int r;
-	unsigned long flags;
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	r = __get(prison, key, lock_level, inmate, cell_prealloc, cell_result);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 
 	return r;
 }
@@ -261,11 +260,10 @@ int dm_cell_lock_v2(struct dm_bio_prison_v2 *prison,
 		    struct dm_bio_prison_cell_v2 **cell_result)
 {
 	int r;
-	unsigned long flags;
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	r = __lock(prison, key, lock_level, cell_prealloc, cell_result);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 
 	return r;
 }
@@ -285,11 +283,9 @@ void dm_cell_quiesce_v2(struct dm_bio_prison_v2 *prison,
 			struct dm_bio_prison_cell_v2 *cell,
 			struct work_struct *continuation)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	__quiesce(prison, cell, continuation);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 }
 EXPORT_SYMBOL_GPL(dm_cell_quiesce_v2);
 
@@ -309,11 +305,10 @@ int dm_cell_lock_promote_v2(struct dm_bio_prison_v2 *prison,
 			    unsigned new_lock_level)
 {
 	int r;
-	unsigned long flags;
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	r = __promote(prison, cell, new_lock_level);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 
 	return r;
 }
@@ -342,11 +337,10 @@ bool dm_cell_unlock_v2(struct dm_bio_prison_v2 *prison,
 		       struct bio_list *bios)
 {
 	bool r;
-	unsigned long flags;
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irq(&prison->lock);
 	r = __unlock(prison, cell, bios);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irq(&prison->lock);
 
 	return r;
 }

drivers/md/dm-cache-target.c

@@ -74,22 +74,19 @@ static bool __iot_idle_for(struct io_tracker *iot, unsigned long jifs)
 static bool iot_idle_for(struct io_tracker *iot, unsigned long jifs)
 {
 	bool r;
-	unsigned long flags;
 
-	spin_lock_irqsave(&iot->lock, flags);
+	spin_lock_irq(&iot->lock);
 	r = __iot_idle_for(iot, jifs);
-	spin_unlock_irqrestore(&iot->lock, flags);
+	spin_unlock_irq(&iot->lock);
 
 	return r;
 }
 
 static void iot_io_begin(struct io_tracker *iot, sector_t len)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&iot->lock, flags);
+	spin_lock_irq(&iot->lock);
 	iot->in_flight += len;
-	spin_unlock_irqrestore(&iot->lock, flags);
+	spin_unlock_irq(&iot->lock);
 }
 
 static void __iot_io_end(struct io_tracker *iot, sector_t len)
@@ -172,7 +169,6 @@ static void __commit(struct work_struct *_ws)
 {
 	struct batcher *b = container_of(_ws, struct batcher, commit_work);
 	blk_status_t r;
-	unsigned long flags;
 	struct list_head work_items;
 	struct work_struct *ws, *tmp;
 	struct continuation *k;
@@ -186,12 +182,12 @@ static void __commit(struct work_struct *_ws)
 	 * We have to grab these before the commit_op to avoid a race
 	 * condition.
 	 */
-	spin_lock_irqsave(&b->lock, flags);
+	spin_lock_irq(&b->lock);
 	list_splice_init(&b->work_items, &work_items);
 	bio_list_merge(&bios, &b->bios);
 	bio_list_init(&b->bios);
 	b->commit_scheduled = false;
-	spin_unlock_irqrestore(&b->lock, flags);
+	spin_unlock_irq(&b->lock);
 
 	r = b->commit_op(b->commit_context);
 
@@ -238,13 +234,12 @@ static void async_commit(struct batcher *b)
 
 static void continue_after_commit(struct batcher *b, struct continuation *k)
 {
-	unsigned long flags;
 	bool commit_scheduled;
 
-	spin_lock_irqsave(&b->lock, flags);
+	spin_lock_irq(&b->lock);
 	commit_scheduled = b->commit_scheduled;
 	list_add_tail(&k->ws.entry, &b->work_items);
-	spin_unlock_irqrestore(&b->lock, flags);
+	spin_unlock_irq(&b->lock);
 
 	if (commit_scheduled)
 		async_commit(b);
@@ -255,13 +250,12 @@ static void continue_after_commit(struct batcher *b, struct continuation *k)
  */
 static void issue_after_commit(struct batcher *b, struct bio *bio)
 {
-       unsigned long flags;
        bool commit_scheduled;
 
-       spin_lock_irqsave(&b->lock, flags);
+       spin_lock_irq(&b->lock);
        commit_scheduled = b->commit_scheduled;
        bio_list_add(&b->bios, bio);
-       spin_unlock_irqrestore(&b->lock, flags);
+       spin_unlock_irq(&b->lock);
 
        if (commit_scheduled)
 	       async_commit(b);
@@ -273,12 +267,11 @@ static void issue_after_commit(struct batcher *b, struct bio *bio)
 static void schedule_commit(struct batcher *b)
 {
 	bool immediate;
-	unsigned long flags;
 
-	spin_lock_irqsave(&b->lock, flags);
+	spin_lock_irq(&b->lock);
 	immediate = !list_empty(&b->work_items) || !bio_list_empty(&b->bios);
 	b->commit_scheduled = true;
-	spin_unlock_irqrestore(&b->lock, flags);
+	spin_unlock_irq(&b->lock);
 
 	if (immediate)
 		async_commit(b);
@@ -630,23 +623,19 @@ static struct per_bio_data *init_per_bio_data(struct bio *bio)
 
 static void defer_bio(struct cache *cache, struct bio *bio)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	bio_list_add(&cache->deferred_bios, bio);
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 
 	wake_deferred_bio_worker(cache);
 }
 
 static void defer_bios(struct cache *cache, struct bio_list *bios)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	bio_list_merge(&cache->deferred_bios, bios);
 	bio_list_init(bios);
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 
 	wake_deferred_bio_worker(cache);
 }
@@ -756,33 +745,27 @@ static dm_dblock_t oblock_to_dblock(struct cache *cache, dm_oblock_t oblock)
 
 static void set_discard(struct cache *cache, dm_dblock_t b)
 {
-	unsigned long flags;
-
 	BUG_ON(from_dblock(b) >= from_dblock(cache->discard_nr_blocks));
 	atomic_inc(&cache->stats.discard_count);
 
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	set_bit(from_dblock(b), cache->discard_bitset);
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 }
 
 static void clear_discard(struct cache *cache, dm_dblock_t b)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	clear_bit(from_dblock(b), cache->discard_bitset);
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 }
 
 static bool is_discarded(struct cache *cache, dm_dblock_t b)
 {
 	int r;
-	unsigned long flags;
-
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	r = test_bit(from_dblock(b), cache->discard_bitset);
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 
 	return r;
 }
@@ -790,12 +773,10 @@ static bool is_discarded(struct cache *cache, dm_dblock_t b)
 static bool is_discarded_oblock(struct cache *cache, dm_oblock_t b)
 {
 	int r;
-	unsigned long flags;
-
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	r = test_bit(from_dblock(oblock_to_dblock(cache, b)),
 		     cache->discard_bitset);
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 
 	return r;
 }
@@ -827,17 +808,16 @@ static void remap_to_cache(struct cache *cache, struct bio *bio,
 
 static void check_if_tick_bio_needed(struct cache *cache, struct bio *bio)
 {
-	unsigned long flags;
 	struct per_bio_data *pb;
 
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	if (cache->need_tick_bio && !op_is_flush(bio->bi_opf) &&
 	    bio_op(bio) != REQ_OP_DISCARD) {
 		pb = get_per_bio_data(bio);
 		pb->tick = true;
 		cache->need_tick_bio = false;
 	}
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 }
 
 static void __remap_to_origin_clear_discard(struct cache *cache, struct bio *bio,
@@ -1889,17 +1869,16 @@ static void process_deferred_bios(struct work_struct *ws)
 {
 	struct cache *cache = container_of(ws, struct cache, deferred_bio_worker);
 
-	unsigned long flags;
 	bool commit_needed = false;
 	struct bio_list bios;
 	struct bio *bio;
 
 	bio_list_init(&bios);
 
-	spin_lock_irqsave(&cache->lock, flags);
+	spin_lock_irq(&cache->lock);
 	bio_list_merge(&bios, &cache->deferred_bios);
 	bio_list_init(&cache->deferred_bios);
-	spin_unlock_irqrestore(&cache->lock, flags);
+	spin_unlock_irq(&cache->lock);
 
 	while ((bio = bio_list_pop(&bios))) {
 		if (bio->bi_opf & REQ_PREFLUSH)

drivers/md/dm-clone-metadata.c

@@ -712,7 +712,7 @@ static int __metadata_commit(struct dm_clone_metadata *cmd)
 static int __flush_dmap(struct dm_clone_metadata *cmd, struct dirty_map *dmap)
 {
 	int r;
-	unsigned long word, flags;
+	unsigned long word;
 
 	word = 0;
 	do {
@@ -736,9 +736,9 @@ static int __flush_dmap(struct dm_clone_metadata *cmd, struct dirty_map *dmap)
 		return r;
 
 	/* Update the changed flag */
-	spin_lock_irqsave(&cmd->bitmap_lock, flags);
+	spin_lock_irq(&cmd->bitmap_lock);
 	dmap->changed = 0;
-	spin_unlock_irqrestore(&cmd->bitmap_lock, flags);
+	spin_unlock_irq(&cmd->bitmap_lock);
 
 	return 0;
 }
@@ -746,7 +746,6 @@ static int __flush_dmap(struct dm_clone_metadata *cmd, struct dirty_map *dmap)
 int dm_clone_metadata_commit(struct dm_clone_metadata *cmd)
 {
 	int r = -EPERM;
-	unsigned long flags;
 	struct dirty_map *dmap, *next_dmap;
 
 	down_write(&cmd->lock);
@@ -770,9 +769,9 @@ int dm_clone_metadata_commit(struct dm_clone_metadata *cmd)
 	}
 
 	/* Swap dirty bitmaps */
-	spin_lock_irqsave(&cmd->bitmap_lock, flags);
+	spin_lock_irq(&cmd->bitmap_lock);
 	cmd->current_dmap = next_dmap;
-	spin_unlock_irqrestore(&cmd->bitmap_lock, flags);
+	spin_unlock_irq(&cmd->bitmap_lock);
 
 	/*
 	 * No one is accessing the old dirty bitmap anymore, so we can flush
@@ -817,9 +816,9 @@ int dm_clone_cond_set_range(struct dm_clone_metadata *cmd, unsigned long start,
 {
 	int r = 0;
 	struct dirty_map *dmap;
-	unsigned long word, region_nr, flags;
+	unsigned long word, region_nr;
 
-	spin_lock_irqsave(&cmd->bitmap_lock, flags);
+	spin_lock_irq(&cmd->bitmap_lock);
 
 	if (cmd->read_only) {
 		r = -EPERM;
@@ -836,7 +835,7 @@ int dm_clone_cond_set_range(struct dm_clone_metadata *cmd, unsigned long start,
 		}
 	}
 out:
-	spin_unlock_irqrestore(&cmd->bitmap_lock, flags);
+	spin_unlock_irq(&cmd->bitmap_lock);
 
 	return r;
 }
@@ -903,13 +902,11 @@ int dm_clone_metadata_abort(struct dm_clone_metadata *cmd)
 
 void dm_clone_metadata_set_read_only(struct dm_clone_metadata *cmd)
 {
-	unsigned long flags;
-
 	down_write(&cmd->lock);
 
-	spin_lock_irqsave(&cmd->bitmap_lock, flags);
+	spin_lock_irq(&cmd->bitmap_lock);
 	cmd->read_only = 1;
-	spin_unlock_irqrestore(&cmd->bitmap_lock, flags);
+	spin_unlock_irq(&cmd->bitmap_lock);
 
 	if (!cmd->fail_io)
 		dm_bm_set_read_only(cmd->bm);
@@ -919,13 +916,11 @@ void dm_clone_metadata_set_read_only(struct dm_clone_metadata *cmd)
 
 void dm_clone_metadata_set_read_write(struct dm_clone_metadata *cmd)
 {
-	unsigned long flags;
-
 	down_write(&cmd->lock);
 
-	spin_lock_irqsave(&cmd->bitmap_lock, flags);
+	spin_lock_irq(&cmd->bitmap_lock);
 	cmd->read_only = 0;
-	spin_unlock_irqrestore(&cmd->bitmap_lock, flags);
+	spin_unlock_irq(&cmd->bitmap_lock);
 
 	if (!cmd->fail_io)
 		dm_bm_set_read_write(cmd->bm);

drivers/md/dm-clone-metadata.h

@@ -44,7 +44,9 @@ int dm_clone_set_region_hydrated(struct dm_clone_metadata *cmd, unsigned long re
  * @start: Starting region number
  * @nr_regions: Number of regions in the range
  *
- * This function doesn't block, so it's safe to call it from interrupt context.
+ * This function doesn't block, but since it uses spin_lock_irq()/spin_unlock_irq()
+ * it's NOT safe to call it from any context where interrupts are disabled, e.g.,
+ * from interrupt context.
  */
 int dm_clone_cond_set_range(struct dm_clone_metadata *cmd, unsigned long start,
 			    unsigned long nr_regions);

drivers/md/dm-clone-target.c

@@ -332,8 +332,6 @@ static void submit_bios(struct bio_list *bios)
  */
 static void issue_bio(struct clone *clone, struct bio *bio)
 {
-	unsigned long flags;
-
 	if (!bio_triggers_commit(clone, bio)) {
 		generic_make_request(bio);
 		return;
@@ -352,9 +350,9 @@ static void issue_bio(struct clone *clone, struct bio *bio)
 	 * Batch together any bios that trigger commits and then issue a single
 	 * commit for them in process_deferred_flush_bios().
 	 */
-	spin_lock_irqsave(&clone->lock, flags);
+	spin_lock_irq(&clone->lock);
 	bio_list_add(&clone->deferred_flush_bios, bio);
-	spin_unlock_irqrestore(&clone->lock, flags);
+	spin_unlock_irq(&clone->lock);
 
 	wake_worker(clone);
 }
@@ -469,7 +467,7 @@ static void complete_discard_bio(struct clone *clone, struct bio *bio, bool succ
 
 static void process_discard_bio(struct clone *clone, struct bio *bio)
 {
-	unsigned long rs, re, flags;
+	unsigned long rs, re;
 
 	bio_region_range(clone, bio, &rs, &re);
 	BUG_ON(re > clone->nr_regions);
@@ -501,9 +499,9 @@ static void process_discard_bio(struct clone *clone, struct bio *bio)
 	/*
 	 * Defer discard processing.
 	 */
-	spin_lock_irqsave(&clone->lock, flags);
+	spin_lock_irq(&clone->lock);
 	bio_list_add(&clone->deferred_discard_bios, bio);
-	spin_unlock_irqrestore(&clone->lock, flags);
+	spin_unlock_irq(&clone->lock);
 
 	wake_worker(clone);
 }
@@ -554,6 +552,12 @@ struct hash_table_bucket {
 #define bucket_unlock_irqrestore(bucket, flags) \
 	spin_unlock_irqrestore(&(bucket)->lock, flags)
 
+#define bucket_lock_irq(bucket) \
+	spin_lock_irq(&(bucket)->lock)
+
+#define bucket_unlock_irq(bucket) \
+	spin_unlock_irq(&(bucket)->lock)
+
 static int hash_table_init(struct clone *clone)
 {
 	unsigned int i, sz;
@@ -851,7 +855,6 @@ static void hydration_overwrite(struct dm_clone_region_hydration *hd, struct bio
  */
 static void hydrate_bio_region(struct clone *clone, struct bio *bio)
 {
-	unsigned long flags;
 	unsigned long region_nr;
 	struct hash_table_bucket *bucket;
 	struct dm_clone_region_hydration *hd, *hd2;
@@ -859,19 +862,19 @@ static void hydrate_bio_region(struct clone *clone, struct bio *bio)
 	region_nr = bio_to_region(clone, bio);
 	bucket = get_hash_table_bucket(clone, region_nr);
 
-	bucket_lock_irqsave(bucket, flags);
+	bucket_lock_irq(bucket);
 
 	hd = __hash_find(bucket, region_nr);
 	if (hd) {
 		/* Someone else is hydrating the region */
 		bio_list_add(&hd->deferred_bios, bio);
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 		return;
 	}
 
 	if (dm_clone_is_region_hydrated(clone->cmd, region_nr)) {
 		/* The region has been hydrated */
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 		issue_bio(clone, bio);
 		return;
 	}
@@ -880,16 +883,16 @@ static void hydrate_bio_region(struct clone *clone, struct bio *bio)
 	 * We must allocate a hydration descriptor and start the hydration of
 	 * the corresponding region.
 	 */
-	bucket_unlock_irqrestore(bucket, flags);
+	bucket_unlock_irq(bucket);
 
 	hd = alloc_hydration(clone);
 	hydration_init(hd, region_nr);
 
-	bucket_lock_irqsave(bucket, flags);
+	bucket_lock_irq(bucket);
 
 	/* Check if the region has been hydrated in the meantime. */
 	if (dm_clone_is_region_hydrated(clone->cmd, region_nr)) {
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 		free_hydration(hd);
 		issue_bio(clone, bio);
 		return;
@@ -899,7 +902,7 @@ static void hydrate_bio_region(struct clone *clone, struct bio *bio)
 	if (hd2 != hd) {
 		/* Someone else started the region's hydration. */
 		bio_list_add(&hd2->deferred_bios, bio);
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 		free_hydration(hd);
 		return;
 	}
@@ -911,7 +914,7 @@ static void hydrate_bio_region(struct clone *clone, struct bio *bio)
 	 */
 	if (unlikely(get_clone_mode(clone) >= CM_READ_ONLY)) {
 		hlist_del(&hd->h);
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 		free_hydration(hd);
 		bio_io_error(bio);
 		return;
@@ -925,11 +928,11 @@ static void hydrate_bio_region(struct clone *clone, struct bio *bio)
 	 * to the destination device.
 	 */
 	if (is_overwrite_bio(clone, bio)) {
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 		hydration_overwrite(hd, bio);
 	} else {
 		bio_list_add(&hd->deferred_bios, bio);
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 		hydration_copy(hd, 1);
 	}
 }
@@ -996,7 +999,6 @@ static unsigned long __start_next_hydration(struct clone *clone,
 					    unsigned long offset,
 					    struct batch_info *batch)
 {
-	unsigned long flags;
 	struct hash_table_bucket *bucket;
 	struct dm_clone_region_hydration *hd;
 	unsigned long nr_regions = clone->nr_regions;
@@ -1010,13 +1012,13 @@ static unsigned long __start_next_hydration(struct clone *clone,
 			break;
 
 		bucket = get_hash_table_bucket(clone, offset);
-		bucket_lock_irqsave(bucket, flags);
+		bucket_lock_irq(bucket);
 
 		if (!dm_clone_is_region_hydrated(clone->cmd, offset) &&
 		    !__hash_find(bucket, offset)) {
 			hydration_init(hd, offset);
 			__insert_region_hydration(bucket, hd);
-			bucket_unlock_irqrestore(bucket, flags);
+			bucket_unlock_irq(bucket);
 
 			/* Batch hydration */
 			__batch_hydration(batch, hd);
@@ -1024,7 +1026,7 @@ static unsigned long __start_next_hydration(struct clone *clone,
 			return (offset + 1);
 		}
 
-		bucket_unlock_irqrestore(bucket, flags);
+		bucket_unlock_irq(bucket);
 
 	} while (++offset < nr_regions);
 
@@ -1140,13 +1142,13 @@ static void process_deferred_discards(struct clone *clone)
 	int r = -EPERM;
 	struct bio *bio;
 	struct blk_plug plug;
-	unsigned long rs, re, flags;
+	unsigned long rs, re;
 	struct bio_list discards = BIO_EMPTY_LIST;
 
-	spin_lock_irqsave(&clone->lock, flags);
+	spin_lock_irq(&clone->lock);
 	bio_list_merge(&discards, &clone->deferred_discard_bios);
 	bio_list_init(&clone->deferred_discard_bios);
-	spin_unlock_irqrestore(&clone->lock, flags);
+	spin_unlock_irq(&clone->lock);
 
 	if (bio_list_empty(&discards))
 		return;
@@ -1176,13 +1178,12 @@ static void process_deferred_discards(struct clone *clone)
 
 static void process_deferred_bios(struct clone *clone)
 {
-	unsigned long flags;
 	struct bio_list bios = BIO_EMPTY_LIST;
 
-	spin_lock_irqsave(&clone->lock, flags);
+	spin_lock_irq(&clone->lock);
 	bio_list_merge(&bios, &clone->deferred_bios);
 	bio_list_init(&clone->deferred_bios);
-	spin_unlock_irqrestore(&clone->lock, flags);
+	spin_unlock_irq(&clone->lock);
 
 	if (bio_list_empty(&bios))
 		return;
@@ -1193,7 +1194,6 @@ static void process_deferred_bios(struct clone *clone)
 static void process_deferred_flush_bios(struct clone *clone)
 {
 	struct bio *bio;
-	unsigned long flags;
 	struct bio_list bios = BIO_EMPTY_LIST;
 	struct bio_list bio_completions = BIO_EMPTY_LIST;
 
@@ -1201,13 +1201,13 @@ static void process_deferred_flush_bios(struct clone *clone)
 	 * If there are any deferred flush bios, we must commit the metadata
 	 * before issuing them or signaling their completion.
 	 */
-	spin_lock_irqsave(&clone->lock, flags);
+	spin_lock_irq(&clone->lock);
 	bio_list_merge(&bios, &clone->deferred_flush_bios);
 	bio_list_init(&clone->deferred_flush_bios);
 
 	bio_list_merge(&bio_completions, &clone->deferred_flush_completions);
 	bio_list_init(&clone->deferred_flush_completions);
-	spin_unlock_irqrestore(&clone->lock, flags);
+	spin_unlock_irq(&clone->lock);
 
 	if (bio_list_empty(&bios) && bio_list_empty(&bio_completions) &&
 	    !(dm_clone_changed_this_transaction(clone->cmd) && need_commit_due_to_time(clone)))

drivers/md/dm-crypt.c

@@ -2700,21 +2700,18 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	}
 
 	ret = -ENOMEM;
-	cc->io_queue = alloc_workqueue("kcryptd_io/%s",
-				       WQ_HIGHPRI | WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM,
-				       1, devname);
+	cc->io_queue = alloc_workqueue("kcryptd_io/%s", WQ_MEM_RECLAIM, 1, devname);
 	if (!cc->io_queue) {
 		ti->error = "Couldn't create kcryptd io queue";
 		goto bad;
 	}
 
 	if (test_bit(DM_CRYPT_SAME_CPU, &cc->flags))
-		cc->crypt_queue = alloc_workqueue("kcryptd/%s",
-						  WQ_HIGHPRI | WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM,
+		cc->crypt_queue = alloc_workqueue("kcryptd/%s", WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM,
 						  1, devname);
 	else
 		cc->crypt_queue = alloc_workqueue("kcryptd/%s",
-						  WQ_HIGHPRI | WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM | WQ_UNBOUND,
+						  WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM | WQ_UNBOUND,
 						  num_online_cpus(), devname);
 	if (!cc->crypt_queue) {
 		ti->error = "Couldn't create kcryptd queue";

drivers/md/dm-dust.c

@@ -17,6 +17,7 @@
 struct badblock {
 	struct rb_node node;
 	sector_t bb;
+	unsigned char wr_fail_cnt;
 };
 
 struct dust_device {
@@ -101,7 +102,8 @@ static int dust_remove_block(struct dust_device *dd, unsigned long long block)
 	return 0;
 }
 
-static int dust_add_block(struct dust_device *dd, unsigned long long block)
+static int dust_add_block(struct dust_device *dd, unsigned long long block,
+			  unsigned char wr_fail_cnt)
 {
 	struct badblock *bblock;
 	unsigned long flags;
@@ -115,6 +117,7 @@ static int dust_add_block(struct dust_device *dd, unsigned long long block)
 
 	spin_lock_irqsave(&dd->dust_lock, flags);
 	bblock->bb = block;
+	bblock->wr_fail_cnt = wr_fail_cnt;
 	if (!dust_rb_insert(&dd->badblocklist, bblock)) {
 		if (!dd->quiet_mode) {
 			DMERR("%s: block %llu already in badblocklist",
@@ -126,8 +129,10 @@ static int dust_add_block(struct dust_device *dd, unsigned long long block)
 	}
 
 	dd->badblock_count++;
-	if (!dd->quiet_mode)
-		DMINFO("%s: badblock added at block %llu", __func__, block);
+	if (!dd->quiet_mode) {
+		DMINFO("%s: badblock added at block %llu with write fail count %hhu",
+		       __func__, block, wr_fail_cnt);
+	}
 	spin_unlock_irqrestore(&dd->dust_lock, flags);
 
 	return 0;
@@ -163,22 +168,27 @@ static int dust_map_read(struct dust_device *dd, sector_t thisblock,
 			 bool fail_read_on_bb)
 {
 	unsigned long flags;
-	int ret = DM_MAPIO_REMAPPED;
+	int r = DM_MAPIO_REMAPPED;
 
 	if (fail_read_on_bb) {
 		thisblock >>= dd->sect_per_block_shift;
 		spin_lock_irqsave(&dd->dust_lock, flags);
-		ret = __dust_map_read(dd, thisblock);
+		r = __dust_map_read(dd, thisblock);
 		spin_unlock_irqrestore(&dd->dust_lock, flags);
 	}
 
-	return ret;
+	return r;
 }
 
-static void __dust_map_write(struct dust_device *dd, sector_t thisblock)
+static int __dust_map_write(struct dust_device *dd, sector_t thisblock)
 {
 	struct badblock *bblk = dust_rb_search(&dd->badblocklist, thisblock);
 
+	if (bblk && bblk->wr_fail_cnt > 0) {
+		bblk->wr_fail_cnt--;
+		return DM_MAPIO_KILL;
+	}
+
 	if (bblk) {
 		rb_erase(&bblk->node, &dd->badblocklist);
 		dd->badblock_count--;
@@ -189,37 +199,40 @@ static void __dust_map_write(struct dust_device *dd, sector_t thisblock)
 			       (unsigned long long)thisblock);
 		}
 	}
+
+	return DM_MAPIO_REMAPPED;
 }
 
 static int dust_map_write(struct dust_device *dd, sector_t thisblock,
 			  bool fail_read_on_bb)
 {
 	unsigned long flags;
+	int ret = DM_MAPIO_REMAPPED;
 
 	if (fail_read_on_bb) {
 		thisblock >>= dd->sect_per_block_shift;
 		spin_lock_irqsave(&dd->dust_lock, flags);
-		__dust_map_write(dd, thisblock);
+		ret = __dust_map_write(dd, thisblock);
 		spin_unlock_irqrestore(&dd->dust_lock, flags);
 	}
 
-	return DM_MAPIO_REMAPPED;
+	return ret;
 }
 
 static int dust_map(struct dm_target *ti, struct bio *bio)
 {
 	struct dust_device *dd = ti->private;
-	int ret;
+	int r;
 
 	bio_set_dev(bio, dd->dev->bdev);
 	bio->bi_iter.bi_sector = dd->start + dm_target_offset(ti, bio->bi_iter.bi_sector);
 
 	if (bio_data_dir(bio) == READ)
-		ret = dust_map_read(dd, bio->bi_iter.bi_sector, dd->fail_read_on_bb);
+		r = dust_map_read(dd, bio->bi_iter.bi_sector, dd->fail_read_on_bb);
 	else
-		ret = dust_map_write(dd, bio->bi_iter.bi_sector, dd->fail_read_on_bb);
+		r = dust_map_write(dd, bio->bi_iter.bi_sector, dd->fail_read_on_bb);
 
-	return ret;
+	return r;
 }
 
 static bool __dust_clear_badblocks(struct rb_root *tree,
@@ -375,8 +388,10 @@ static int dust_message(struct dm_target *ti, unsigned int argc, char **argv,
 	struct dust_device *dd = ti->private;
 	sector_t size = i_size_read(dd->dev->bdev->bd_inode) >> SECTOR_SHIFT;
 	bool invalid_msg = false;
-	int result = -EINVAL;
+	int r = -EINVAL;
 	unsigned long long tmp, block;
+	unsigned char wr_fail_cnt;
+	unsigned int tmp_ui;
 	unsigned long flags;
 	char dummy;
 
@@ -388,45 +403,69 @@ static int dust_message(struct dm_target *ti, unsigned int argc, char **argv,
 		} else if (!strcasecmp(argv[0], "disable")) {
 			DMINFO("disabling read failures on bad sectors");
 			dd->fail_read_on_bb = false;
-			result = 0;
+			r = 0;
 		} else if (!strcasecmp(argv[0], "enable")) {
 			DMINFO("enabling read failures on bad sectors");
 			dd->fail_read_on_bb = true;
-			result = 0;
+			r = 0;
 		} else if (!strcasecmp(argv[0], "countbadblocks")) {
 			spin_lock_irqsave(&dd->dust_lock, flags);
 			DMINFO("countbadblocks: %llu badblock(s) found",
 			       dd->badblock_count);
 			spin_unlock_irqrestore(&dd->dust_lock, flags);
-			result = 0;
+			r = 0;
 		} else if (!strcasecmp(argv[0], "clearbadblocks")) {
-			result = dust_clear_badblocks(dd);
+			r = dust_clear_badblocks(dd);
 		} else if (!strcasecmp(argv[0], "quiet")) {
 			if (!dd->quiet_mode)
 				dd->quiet_mode = true;
 			else
 				dd->quiet_mode = false;
-			result = 0;
+			r = 0;
 		} else {
 			invalid_msg = true;
 		}
 	} else if (argc == 2) {
 		if (sscanf(argv[1], "%llu%c", &tmp, &dummy) != 1)
-			return result;
+			return r;
 
 		block = tmp;
 		sector_div(size, dd->sect_per_block);
 		if (block > size) {
 			DMERR("selected block value out of range");
-			return result;
+			return r;
 		}
 
 		if (!strcasecmp(argv[0], "addbadblock"))
-			result = dust_add_block(dd, block);
+			r = dust_add_block(dd, block, 0);
 		else if (!strcasecmp(argv[0], "removebadblock"))
-			result = dust_remove_block(dd, block);
+			r = dust_remove_block(dd, block);
 		else if (!strcasecmp(argv[0], "queryblock"))
-			result = dust_query_block(dd, block);
+			r = dust_query_block(dd, block);
+		else
+			invalid_msg = true;
+
+	} else if (argc == 3) {
+		if (sscanf(argv[1], "%llu%c", &tmp, &dummy) != 1)
+			return r;
+
+		if (sscanf(argv[2], "%u%c", &tmp_ui, &dummy) != 1)
+			return r;
+
+		block = tmp;
+		if (tmp_ui > 255) {
+			DMERR("selected write fail count out of range");
+			return r;
+		}
+		wr_fail_cnt = tmp_ui;
+		sector_div(size, dd->sect_per_block);
+		if (block > size) {
+			DMERR("selected block value out of range");
+			return r;
+		}
+
+		if (!strcasecmp(argv[0], "addbadblock"))
+			r = dust_add_block(dd, block, wr_fail_cnt);
 		else
 			invalid_msg = true;
 
@@ -436,7 +475,7 @@ static int dust_message(struct dm_target *ti, unsigned int argc, char **argv,
 	if (invalid_msg)
 		DMERR("unrecognized message '%s' received", argv[0]);
 
-	return result;
+	return r;
 }
 
 static void dust_status(struct dm_target *ti, status_type_t type,
@@ -499,12 +538,12 @@ static struct target_type dust_target = {
 
 static int __init dm_dust_init(void)
 {
-	int result = dm_register_target(&dust_target);
+	int r = dm_register_target(&dust_target);
 
-	if (result < 0)
-		DMERR("dm_register_target failed %d", result);
+	if (r < 0)
+		DMERR("dm_register_target failed %d", r);
 
-	return result;
+	return r;
 }
 
 static void __exit dm_dust_exit(void)

drivers/md/dm-integrity.c

@@ -53,6 +53,7 @@
 #define SB_VERSION_1			1
 #define SB_VERSION_2			2
 #define SB_VERSION_3			3
+#define SB_VERSION_4			4
 #define SB_SECTORS			8
 #define MAX_SECTORS_PER_BLOCK		8
 
@@ -73,6 +74,7 @@ struct superblock {
 #define SB_FLAG_HAVE_JOURNAL_MAC	0x1
 #define SB_FLAG_RECALCULATING		0x2
 #define SB_FLAG_DIRTY_BITMAP		0x4
+#define SB_FLAG_FIXED_PADDING		0x8
 
 #define	JOURNAL_ENTRY_ROUNDUP		8
 
@@ -250,6 +252,7 @@ struct dm_integrity_c {
 	bool journal_uptodate;
 	bool just_formatted;
 	bool recalculate_flag;
+	bool fix_padding;
 
 	struct alg_spec internal_hash_alg;
 	struct alg_spec journal_crypt_alg;
@@ -463,7 +466,9 @@ static void wraparound_section(struct dm_integrity_c *ic, unsigned *sec_ptr)
 
 static void sb_set_version(struct dm_integrity_c *ic)
 {
-	if (ic->mode == 'B' || ic->sb->flags & cpu_to_le32(SB_FLAG_DIRTY_BITMAP))
+	if (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING))
+		ic->sb->version = SB_VERSION_4;
+	else if (ic->mode == 'B' || ic->sb->flags & cpu_to_le32(SB_FLAG_DIRTY_BITMAP))
 		ic->sb->version = SB_VERSION_3;
 	else if (ic->meta_dev || ic->sb->flags & cpu_to_le32(SB_FLAG_RECALCULATING))
 		ic->sb->version = SB_VERSION_2;
@@ -2955,6 +2960,7 @@ static void dm_integrity_status(struct dm_target *ti, status_type_t type,
 		arg_count += !!ic->internal_hash_alg.alg_string;
 		arg_count += !!ic->journal_crypt_alg.alg_string;
 		arg_count += !!ic->journal_mac_alg.alg_string;
+		arg_count += (ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING)) != 0;
 		DMEMIT("%s %llu %u %c %u", ic->dev->name, (unsigned long long)ic->start,
 		       ic->tag_size, ic->mode, arg_count);
 		if (ic->meta_dev)
@@ -2974,6 +2980,8 @@ static void dm_integrity_status(struct dm_target *ti, status_type_t type,
 			DMEMIT(" sectors_per_bit:%llu", (unsigned long long)ic->sectors_per_block << ic->log2_blocks_per_bitmap_bit);
 			DMEMIT(" bitmap_flush_interval:%u", jiffies_to_msecs(ic->bitmap_flush_interval));
 		}
+		if ((ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING)) != 0)
+			DMEMIT(" fix_padding");
 
 #define EMIT_ALG(a, n)							\
 		do {							\
@@ -3042,8 +3050,14 @@ static int calculate_device_limits(struct dm_integrity_c *ic)
 	if (!ic->meta_dev) {
 		sector_t last_sector, last_area, last_offset;
 
-		ic->metadata_run = roundup((__u64)ic->tag_size << (ic->sb->log2_interleave_sectors - ic->sb->log2_sectors_per_block),
-					   (__u64)(1 << SECTOR_SHIFT << METADATA_PADDING_SECTORS)) >> SECTOR_SHIFT;
+		/* we have to maintain excessive padding for compatibility with existing volumes */
+		__u64 metadata_run_padding =
+			ic->sb->flags & cpu_to_le32(SB_FLAG_FIXED_PADDING) ?
+			(__u64)(METADATA_PADDING_SECTORS << SECTOR_SHIFT) :
+			(__u64)(1 << SECTOR_SHIFT << METADATA_PADDING_SECTORS);
+
+		ic->metadata_run = round_up((__u64)ic->tag_size << (ic->sb->log2_interleave_sectors - ic->sb->log2_sectors_per_block),
+					    metadata_run_padding) >> SECTOR_SHIFT;
 		if (!(ic->metadata_run & (ic->metadata_run - 1)))
 			ic->log2_metadata_run = __ffs(ic->metadata_run);
 		else
@@ -3086,6 +3100,8 @@ static int initialize_superblock(struct dm_integrity_c *ic, unsigned journal_sec
 		journal_sections = 1;
 
 	if (!ic->meta_dev) {
+		if (ic->fix_padding)
+			ic->sb->flags |= cpu_to_le32(SB_FLAG_FIXED_PADDING);
 		ic->sb->journal_sections = cpu_to_le32(journal_sections);
 		if (!interleave_sectors)
 			interleave_sectors = DEFAULT_INTERLEAVE_SECTORS;
@@ -3725,6 +3741,8 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 				goto bad;
 		} else if (!strcmp(opt_string, "recalculate")) {
 			ic->recalculate_flag = true;
+		} else if (!strcmp(opt_string, "fix_padding")) {
+			ic->fix_padding = true;
 		} else {
 			r = -EINVAL;
 			ti->error = "Invalid argument";
@@ -3867,7 +3885,7 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 			should_write_sb = true;
 	}
 
-	if (!ic->sb->version || ic->sb->version > SB_VERSION_3) {
+	if (!ic->sb->version || ic->sb->version > SB_VERSION_4) {
 		r = -EINVAL;
 		ti->error = "Unknown version";
 		goto bad;
@@ -4182,7 +4200,7 @@ static void dm_integrity_dtr(struct dm_target *ti)
 
 static struct target_type integrity_target = {
 	.name			= "integrity",
-	.version		= {1, 3, 0},
+	.version		= {1, 4, 0},
 	.module			= THIS_MODULE,
 	.features		= DM_TARGET_SINGLETON | DM_TARGET_INTEGRITY,
 	.ctr			= dm_integrity_ctr,

drivers/md/dm-stripe.c

@@ -55,19 +55,6 @@ static void trigger_event(struct work_struct *work)
 	dm_table_event(sc->ti->table);
 }
 
-static inline struct stripe_c *alloc_context(unsigned int stripes)
-{
-	size_t len;
-
-	if (dm_array_too_big(sizeof(struct stripe_c), sizeof(struct stripe),
-			     stripes))
-		return NULL;
-
-	len = sizeof(struct stripe_c) + (sizeof(struct stripe) * stripes);
-
-	return kmalloc(len, GFP_KERNEL);
-}
-
 /*
  * Parse a single <dev> <sector> pair
  */
@@ -142,7 +129,7 @@ static int stripe_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		return -EINVAL;
 	}
 
-	sc = alloc_context(stripes);
+	sc = kmalloc(struct_size(sc, stripe, stripes), GFP_KERNEL);
 	if (!sc) {
 		ti->error = "Memory allocation for striped context "
 		    "failed";

drivers/md/dm-table.c

@@ -918,21 +918,15 @@ bool dm_table_supports_dax(struct dm_table *t,
 
 static bool dm_table_does_not_support_partial_completion(struct dm_table *t);
 
-struct verify_rq_based_data {
-	unsigned sq_count;
-	unsigned mq_count;
-};
-
-static int device_is_rq_based(struct dm_target *ti, struct dm_dev *dev,
-			      sector_t start, sector_t len, void *data)
+static int device_is_rq_stackable(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);
-	struct verify_rq_based_data *v = data;
+	struct block_device *bdev = dev->bdev;
+	struct request_queue *q = bdev_get_queue(bdev);
 
-	if (queue_is_mq(q))
-		v->mq_count++;
-	else
-		v->sq_count++;
+	/* request-based cannot stack on partitions! */
+	if (bdev != bdev->bd_contains)
+		return false;
 
 	return queue_is_mq(q);
 }
@@ -941,7 +935,6 @@ static int dm_table_determine_type(struct dm_table *t)
 {
 	unsigned i;
 	unsigned bio_based = 0, request_based = 0, hybrid = 0;
-	struct verify_rq_based_data v = {.sq_count = 0, .mq_count = 0};
 	struct dm_target *tgt;
 	struct list_head *devices = dm_table_get_devices(t);
 	enum dm_queue_mode live_md_type = dm_get_md_type(t->md);
@@ -1045,14 +1038,10 @@ static int dm_table_determine_type(struct dm_table *t)
 
 	/* Non-request-stackable devices can't be used for request-based dm */
 	if (!tgt->type->iterate_devices ||
-	    !tgt->type->iterate_devices(tgt, device_is_rq_based, &v)) {
+	    !tgt->type->iterate_devices(tgt, device_is_rq_stackable, NULL)) {
 		DMERR("table load rejected: including non-request-stackable devices");
 		return -EINVAL;
 	}
-	if (v.sq_count > 0) {
-		DMERR("table load rejected: not all devices are blk-mq request-stackable");
-		return -EINVAL;
-	}
 
 	return 0;
 }

drivers/md/dm-writecache.c

@@ -1218,7 +1218,8 @@ static int writecache_map(struct dm_target *ti, struct bio *bio)
 			}
 		} while (bio->bi_iter.bi_size);
 
-		if (unlikely(wc->uncommitted_blocks >= wc->autocommit_blocks))
+		if (unlikely(bio->bi_opf & REQ_FUA ||
+			     wc->uncommitted_blocks >= wc->autocommit_blocks))
 			writecache_flush(wc);
 		else
 			writecache_schedule_autocommit(wc);
@@ -1561,7 +1562,7 @@ static void writecache_writeback(struct work_struct *work)
 {
 	struct dm_writecache *wc = container_of(work, struct dm_writecache, writeback_work);
 	struct blk_plug plug;
-	struct wc_entry *f, *g, *e = NULL;
+	struct wc_entry *f, *uninitialized_var(g), *e = NULL;
 	struct rb_node *node, *next_node;
 	struct list_head skipped;
 	struct writeback_list wbl;

drivers/md/dm-zoned-metadata.c

@@ -554,6 +554,7 @@ static struct dmz_mblock *dmz_get_mblock(struct dmz_metadata *zmd,
 		       TASK_UNINTERRUPTIBLE);
 	if (test_bit(DMZ_META_ERROR, &mblk->state)) {
 		dmz_release_mblock(zmd, mblk);
+		dmz_check_bdev(zmd->dev);
 		return ERR_PTR(-EIO);
 	}
 
@@ -625,6 +626,8 @@ static int dmz_rdwr_block(struct dmz_metadata *zmd, int op, sector_t block,
 	ret = submit_bio_wait(bio);
 	bio_put(bio);
 
+	if (ret)
+		dmz_check_bdev(zmd->dev);
 	return ret;
 }
 
@@ -691,6 +694,7 @@ static int dmz_write_dirty_mblocks(struct dmz_metadata *zmd,
 			       TASK_UNINTERRUPTIBLE);
 		if (test_bit(DMZ_META_ERROR, &mblk->state)) {
 			clear_bit(DMZ_META_ERROR, &mblk->state);
+			dmz_check_bdev(zmd->dev);
 			ret = -EIO;
 		}
 		nr_mblks_submitted--;
@@ -768,7 +772,7 @@ int dmz_flush_metadata(struct dmz_metadata *zmd)
 	/* If there are no dirty metadata blocks, just flush the device cache */
 	if (list_empty(&write_list)) {
 		ret = blkdev_issue_flush(zmd->dev->bdev, GFP_NOIO, NULL);
-		goto out;
+		goto err;
 	}
 
 	/*
@@ -778,7 +782,7 @@ int dmz_flush_metadata(struct dmz_metadata *zmd)
 	 */
 	ret = dmz_log_dirty_mblocks(zmd, &write_list);
 	if (ret)
-		goto out;
+		goto err;
 
 	/*
 	 * The log is on disk. It is now safe to update in place
@@ -786,11 +790,11 @@ int dmz_flush_metadata(struct dmz_metadata *zmd)
 	 */
 	ret = dmz_write_dirty_mblocks(zmd, &write_list, zmd->mblk_primary);
 	if (ret)
-		goto out;
+		goto err;
 
 	ret = dmz_write_sb(zmd, zmd->mblk_primary);
 	if (ret)
-		goto out;
+		goto err;
 
 	while (!list_empty(&write_list)) {
 		mblk = list_first_entry(&write_list, struct dmz_mblock, link);
@@ -805,16 +809,20 @@ int dmz_flush_metadata(struct dmz_metadata *zmd)
 
 	zmd->sb_gen++;
 out:
-	if (ret && !list_empty(&write_list)) {
-		spin_lock(&zmd->mblk_lock);
-		list_splice(&write_list, &zmd->mblk_dirty_list);
-		spin_unlock(&zmd->mblk_lock);
-	}
-
 	dmz_unlock_flush(zmd);
 	up_write(&zmd->mblk_sem);
 
 	return ret;
+
+err:
+	if (!list_empty(&write_list)) {
+		spin_lock(&zmd->mblk_lock);
+		list_splice(&write_list, &zmd->mblk_dirty_list);
+		spin_unlock(&zmd->mblk_lock);
+	}
+	if (!dmz_check_bdev(zmd->dev))
+		ret = -EIO;
+	goto out;
 }
 
 /*
@@ -1221,6 +1229,7 @@ static int dmz_update_zone(struct dmz_metadata *zmd, struct dm_zone *zone)
 	if (ret < 0) {
 		dmz_dev_err(zmd->dev, "Get zone %u report failed",
 			    dmz_id(zmd, zone));
+		dmz_check_bdev(zmd->dev);
 		return ret;
 	}
 

drivers/md/dm-zoned-reclaim.c

@@ -82,6 +82,7 @@ static int dmz_reclaim_align_wp(struct dmz_reclaim *zrc, struct dm_zone *zone,
 			    "Align zone %u wp %llu to %llu (wp+%u) blocks failed %d",
 			    dmz_id(zmd, zone), (unsigned long long)wp_block,
 			    (unsigned long long)block, nr_blocks, ret);
+		dmz_check_bdev(zrc->dev);
 		return ret;
 	}
 
@@ -489,12 +490,7 @@ static void dmz_reclaim_work(struct work_struct *work)
 	ret = dmz_do_reclaim(zrc);
 	if (ret) {
 		dmz_dev_debug(zrc->dev, "Reclaim error %d\n", ret);
-		if (ret == -EIO)
-			/*
-			 * LLD might be performing some error handling sequence
-			 * at the underlying device. To not interfere, do not
-			 * attempt to schedule the next reclaim run immediately.
-			 */
+		if (!dmz_check_bdev(zrc->dev))
 			return;
 	}
 

drivers/md/dm-zoned-target.c

@@ -80,6 +80,8 @@ static inline void dmz_bio_endio(struct bio *bio, blk_status_t status)
 
 	if (status != BLK_STS_OK && bio->bi_status == BLK_STS_OK)
 		bio->bi_status = status;
+	if (bio->bi_status != BLK_STS_OK)
+		bioctx->target->dev->flags |= DMZ_CHECK_BDEV;
 
 	if (refcount_dec_and_test(&bioctx->ref)) {
 		struct dm_zone *zone = bioctx->zone;
@@ -565,31 +567,51 @@ static int dmz_queue_chunk_work(struct dmz_target *dmz, struct bio *bio)
 }
 
 /*
- * Check the backing device availability. If it's on the way out,
+ * Check if the backing device is being removed. If it's on the way out,
  * start failing I/O. Reclaim and metadata components also call this
  * function to cleanly abort operation in the event of such failure.
  */
 bool dmz_bdev_is_dying(struct dmz_dev *dmz_dev)
 {
-	struct gendisk *disk;
+	if (dmz_dev->flags & DMZ_BDEV_DYING)
+		return true;
 
-	if (!(dmz_dev->flags & DMZ_BDEV_DYING)) {
-		disk = dmz_dev->bdev->bd_disk;
-		if (blk_queue_dying(bdev_get_queue(dmz_dev->bdev))) {
-			dmz_dev_warn(dmz_dev, "Backing device queue dying");
-			dmz_dev->flags |= DMZ_BDEV_DYING;
-		} else if (disk->fops->check_events) {
-			if (disk->fops->check_events(disk, 0) &
-					DISK_EVENT_MEDIA_CHANGE) {
-				dmz_dev_warn(dmz_dev, "Backing device offline");
-				dmz_dev->flags |= DMZ_BDEV_DYING;
-			}
-		}
+	if (dmz_dev->flags & DMZ_CHECK_BDEV)
+		return !dmz_check_bdev(dmz_dev);
+
+	if (blk_queue_dying(bdev_get_queue(dmz_dev->bdev))) {
+		dmz_dev_warn(dmz_dev, "Backing device queue dying");
+		dmz_dev->flags |= DMZ_BDEV_DYING;
 	}
 
 	return dmz_dev->flags & DMZ_BDEV_DYING;
 }
 
+/*
+ * Check the backing device availability. This detects such events as
+ * backing device going offline due to errors, media removals, etc.
+ * This check is less efficient than dmz_bdev_is_dying() and should
+ * only be performed as a part of error handling.
+ */
+bool dmz_check_bdev(struct dmz_dev *dmz_dev)
+{
+	struct gendisk *disk;
+
+	dmz_dev->flags &= ~DMZ_CHECK_BDEV;
+
+	if (dmz_bdev_is_dying(dmz_dev))
+		return false;
+
+	disk = dmz_dev->bdev->bd_disk;
+	if (disk->fops->check_events &&
+	    disk->fops->check_events(disk, 0) & DISK_EVENT_MEDIA_CHANGE) {
+		dmz_dev_warn(dmz_dev, "Backing device offline");
+		dmz_dev->flags |= DMZ_BDEV_DYING;
+	}
+
+	return !(dmz_dev->flags & DMZ_BDEV_DYING);
+}
+
 /*
  * Process a new BIO.
  */
@@ -902,8 +924,8 @@ static int dmz_prepare_ioctl(struct dm_target *ti, struct block_device **bdev)
 {
 	struct dmz_target *dmz = ti->private;
 
-	if (dmz_bdev_is_dying(dmz->dev))
-		return -ENODEV;
+	if (!dmz_check_bdev(dmz->dev))
+		return -EIO;
 
 	*bdev = dmz->dev->bdev;
 

drivers/md/dm-zoned.h

@@ -72,6 +72,7 @@ struct dmz_dev {
 
 /* Device flags. */
 #define DMZ_BDEV_DYING		(1 << 0)
+#define DMZ_CHECK_BDEV		(2 << 0)
 
 /*
  * Zone descriptor.
@@ -255,5 +256,6 @@ void dmz_schedule_reclaim(struct dmz_reclaim *zrc);
  * Functions defined in dm-zoned-target.c
  */
 bool dmz_bdev_is_dying(struct dmz_dev *dmz_dev);
+bool dmz_check_bdev(struct dmz_dev *dmz_dev);
 
 #endif /* DM_ZONED_H */

include/linux/device-mapper.h

@@ -608,9 +608,6 @@ void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size);
  */
 #define dm_round_up(n, sz) (dm_div_up((n), (sz)) * (sz))
 
-#define dm_array_too_big(fixed, obj, num) \
-	((num) > (UINT_MAX - (fixed)) / (obj))
-
 /*
  * Sector offset taken relative to the start of the target instead of
  * relative to the start of the device.