Merge tag 'for-5.2/dm-changes-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper updates from Mike Snitzer:

 - Improve DM snapshot target's scalability by using finer grained
   locking. Requires some list_bl interface improvements.

 - Add ability for DM integrity to use a bitmap mode, that tracks
   regions where data and metadata are out of sync, instead of using a
   journal.

 - Improve DM thin provisioning target to not write metadata changes to
   disk if the thin-pool and associated thin devices are merely
   activated but not used. This avoids metadata corruption due to
   concurrent activation of thin devices across different OS instances
   (e.g. split brain scenarios, which ultimately would be avoided if
   proper device filters were used -- but not having proper filtering
   has proven a very common configuration mistake)

 - Fix missing call to path selector type->end_io in DM multipath. This
   fixes reported performance problems due to inaccurate path selector
   IO accounting causing an imbalance of IO (e.g. avoiding issuing IO to
   particular path due to it seemingly being heavily used).

 - Fix bug in DM cache metadata's loading of its discard bitset that
   could lead to all cache blocks being discarded if the very first
   cache block was discarded (thankfully in practice the first cache
   block is generally in use; be it FS superblock, partition table, disk
   label, etc).

 - Add testing-only DM dust target which simulates a device that has
   failing sectors and/or read failures.

 - Fix a DM init error path reference count hang that caused boot hangs
   if user supplied malformed input on kernel commandline.

 - Fix a couple issues with DM crypt target's logging being overly
   verbose or lacking context.

 - Various other small fixes to DM init, DM multipath, DM zoned, and DM
   crypt.

* tag 'for-5.2/dm-changes-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (42 commits)
  dm: fix a couple brace coding style issues
  dm crypt: print device name in integrity error message
  dm crypt: move detailed message into debug level
  dm ioctl: fix hang in early create error condition
  dm integrity: whitespace, coding style and dead code cleanup
  dm integrity: implement synchronous mode for reboot handling
  dm integrity: handle machine reboot in bitmap mode
  dm integrity: add a bitmap mode
  dm integrity: introduce a function add_new_range_and_wait()
  dm integrity: allow large ranges to be described
  dm ingerity: pass size to dm_integrity_alloc_page_list()
  dm integrity: introduce rw_journal_sectors()
  dm integrity: update documentation
  dm integrity: don't report unused options
  dm integrity: don't check null pointer before kvfree and vfree
  dm integrity: correctly calculate the size of metadata area
  dm dust: Make dm_dust_init and dm_dust_exit static
  dm dust: remove redundant unsigned comparison to less than zero
  dm mpath: always free attached_handler_name in parse_path()
  dm init: fix max devices/targets checks
  ...

20 files changed, 1583 insertions, 293 deletions

diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 2557f198e175..db269a348b20 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -436,6 +436,15 @@ config DM_DELAY
 
 	If unsure, say N.
 
+config DM_DUST
+	tristate "Bad sector simulation target"
+	depends on BLK_DEV_DM
+	---help---
+	A target that simulates bad sector behavior.
+	Useful for testing.
+
+	If unsure, say N.
+
 config DM_INIT
 	bool "DM \"dm-mod.create=\" parameter support"
 	depends on BLK_DEV_DM=y
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index a52b703e588e..be7a6eb92abc 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -48,6 +48,7 @@ obj-$(CONFIG_DM_BUFIO)		+= dm-bufio.o
 obj-$(CONFIG_DM_BIO_PRISON)	+= dm-bio-prison.o
 obj-$(CONFIG_DM_CRYPT)		+= dm-crypt.o
 obj-$(CONFIG_DM_DELAY)		+= dm-delay.o
+obj-$(CONFIG_DM_DUST)		+= dm-dust.o
 obj-$(CONFIG_DM_FLAKEY)		+= dm-flakey.o
 obj-$(CONFIG_DM_MULTIPATH)	+= dm-multipath.o dm-round-robin.o
 obj-$(CONFIG_DM_MULTIPATH_QL)	+= dm-queue-length.o
diff --git a/drivers/md/dm-cache-metadata.c b/drivers/md/dm-cache-metadata.c
index 6fc93834da44..151aa95775be 100644
--- a/drivers/md/dm-cache-metadata.c
+++ b/drivers/md/dm-cache-metadata.c
@@ -1167,11 +1167,18 @@ static int __load_discards(struct dm_cache_metadata *cmd,
 		if (r)
 			return r;
 
-		for (b = 0; b < from_dblock(cmd->discard_nr_blocks); b++) {
+		for (b = 0; ; b++) {
 			r = fn(context, cmd->discard_block_size, to_dblock(b),
 			       dm_bitset_cursor_get_value(&c));
 			if (r)
 				break;
+
+			if (b >= (from_dblock(cmd->discard_nr_blocks) - 1))
+				break;
+
+			r = dm_bitset_cursor_next(&c);
+			if (r)
+				break;
 		}
 
 		dm_bitset_cursor_end(&c);
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 7f6462f74ac8..1b16d34bb785 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -946,6 +946,7 @@ static int crypt_integrity_ctr(struct crypt_config *cc, struct dm_target *ti)
 {
 #ifdef CONFIG_BLK_DEV_INTEGRITY
 	struct blk_integrity *bi = blk_get_integrity(cc->dev->bdev->bd_disk);
+	struct mapped_device *md = dm_table_get_md(ti->table);
 
 	/* From now we require underlying device with our integrity profile */
 	if (!bi || strcasecmp(bi->profile->name, "DM-DIF-EXT-TAG")) {
@@ -965,7 +966,7 @@ static int crypt_integrity_ctr(struct crypt_config *cc, struct dm_target *ti)
 
 	if (crypt_integrity_aead(cc)) {
 		cc->integrity_tag_size = cc->on_disk_tag_size - cc->integrity_iv_size;
-		DMINFO("Integrity AEAD, tag size %u, IV size %u.",
+		DMDEBUG("%s: Integrity AEAD, tag size %u, IV size %u.", dm_device_name(md),
 		       cc->integrity_tag_size, cc->integrity_iv_size);
 
 		if (crypto_aead_setauthsize(any_tfm_aead(cc), cc->integrity_tag_size)) {
@@ -973,7 +974,7 @@ static int crypt_integrity_ctr(struct crypt_config *cc, struct dm_target *ti)
 			return -EINVAL;
 		}
 	} else if (cc->integrity_iv_size)
-		DMINFO("Additional per-sector space %u bytes for IV.",
+		DMDEBUG("%s: Additional per-sector space %u bytes for IV.", dm_device_name(md),
 		       cc->integrity_iv_size);
 
 	if ((cc->integrity_tag_size + cc->integrity_iv_size) != bi->tag_size) {
@@ -1031,11 +1032,11 @@ static u8 *org_iv_of_dmreq(struct crypt_config *cc,
 	return iv_of_dmreq(cc, dmreq) + cc->iv_size;
 }
 
-static uint64_t *org_sector_of_dmreq(struct crypt_config *cc,
+static __le64 *org_sector_of_dmreq(struct crypt_config *cc,
 		       struct dm_crypt_request *dmreq)
 {
 	u8 *ptr = iv_of_dmreq(cc, dmreq) + cc->iv_size + cc->iv_size;
-	return (uint64_t*) ptr;
+	return (__le64 *) ptr;
 }
 
 static unsigned int *org_tag_of_dmreq(struct crypt_config *cc,
@@ -1071,7 +1072,7 @@ static int crypt_convert_block_aead(struct crypt_config *cc,
 	struct bio_vec bv_out = bio_iter_iovec(ctx->bio_out, ctx->iter_out);
 	struct dm_crypt_request *dmreq;
 	u8 *iv, *org_iv, *tag_iv, *tag;
-	uint64_t *sector;
+	__le64 *sector;
 	int r = 0;
 
 	BUG_ON(cc->integrity_iv_size && cc->integrity_iv_size != cc->iv_size);
@@ -1143,9 +1144,11 @@ static int crypt_convert_block_aead(struct crypt_config *cc,
 		r = crypto_aead_decrypt(req);
 	}
 
-	if (r == -EBADMSG)
-		DMERR_LIMIT("INTEGRITY AEAD ERROR, sector %llu",
+	if (r == -EBADMSG) {
+		char b[BDEVNAME_SIZE];
+		DMERR_LIMIT("%s: INTEGRITY AEAD ERROR, sector %llu", bio_devname(ctx->bio_in, b),
 			    (unsigned long long)le64_to_cpu(*sector));
+	}
 
 	if (!r && cc->iv_gen_ops && cc->iv_gen_ops->post)
 		r = cc->iv_gen_ops->post(cc, org_iv, dmreq);
@@ -1166,7 +1169,7 @@ static int crypt_convert_block_skcipher(struct crypt_config *cc,
 	struct scatterlist *sg_in, *sg_out;
 	struct dm_crypt_request *dmreq;
 	u8 *iv, *org_iv, *tag_iv;
-	uint64_t *sector;
+	__le64 *sector;
 	int r = 0;
 
 	/* Reject unexpected unaligned bio. */
@@ -1788,7 +1791,8 @@ static void kcryptd_async_done(struct crypto_async_request *async_req,
 		error = cc->iv_gen_ops->post(cc, org_iv_of_dmreq(cc, dmreq), dmreq);
 
 	if (error == -EBADMSG) {
-		DMERR_LIMIT("INTEGRITY AEAD ERROR, sector %llu",
+		char b[BDEVNAME_SIZE];
+		DMERR_LIMIT("%s: INTEGRITY AEAD ERROR, sector %llu", bio_devname(ctx->bio_in, b),
 			    (unsigned long long)le64_to_cpu(*org_sector_of_dmreq(cc, dmreq)));
 		io->error = BLK_STS_PROTECTION;
 	} else if (error < 0)
@@ -1887,7 +1891,7 @@ static int crypt_alloc_tfms_skcipher(struct crypt_config *cc, char *ciphermode)
 	 * algorithm implementation is used.  Help people debug performance
 	 * problems by logging the ->cra_driver_name.
 	 */
-	DMINFO("%s using implementation \"%s\"", ciphermode,
+	DMDEBUG_LIMIT("%s using implementation \"%s\"", ciphermode,
 	       crypto_skcipher_alg(any_tfm(cc))->base.cra_driver_name);
 	return 0;
 }
@@ -1907,7 +1911,7 @@ static int crypt_alloc_tfms_aead(struct crypt_config *cc, char *ciphermode)
 		return err;
 	}
 
-	DMINFO("%s using implementation \"%s\"", ciphermode,
+	DMDEBUG_LIMIT("%s using implementation \"%s\"", ciphermode,
 	       crypto_aead_alg(any_tfm_aead(cc))->base.cra_driver_name);
 	return 0;
 }
diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c
index fddffe251bf6..f496213f8b67 100644
--- a/drivers/md/dm-delay.c
+++ b/drivers/md/dm-delay.c
@@ -121,7 +121,8 @@ static void delay_dtr(struct dm_target *ti)
 {
 	struct delay_c *dc = ti->private;
 
-	destroy_workqueue(dc->kdelayd_wq);
+	if (dc->kdelayd_wq)
+		destroy_workqueue(dc->kdelayd_wq);
 
 	if (dc->read.dev)
 		dm_put_device(ti, dc->read.dev);
diff --git a/drivers/md/dm-dust.c b/drivers/md/dm-dust.c
new file mode 100644
index 000000000000..845f376a72d9
--- /dev/null
+++ b/drivers/md/dm-dust.c
@@ -0,0 +1,515 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2018 Red Hat, Inc.
+ *
+ * This is a test "dust" device, which fails reads on specified
+ * sectors, emulating the behavior of a hard disk drive sending
+ * a "Read Medium Error" sense.
+ *
+ */
+
+#include <linux/device-mapper.h>
+#include <linux/module.h>
+#include <linux/rbtree.h>
+
+#define DM_MSG_PREFIX "dust"
+
+struct badblock {
+	struct rb_node node;
+	sector_t bb;
+};
+
+struct dust_device {
+	struct dm_dev *dev;
+	struct rb_root badblocklist;
+	unsigned long long badblock_count;
+	spinlock_t dust_lock;
+	unsigned int blksz;
+	unsigned int sect_per_block;
+	sector_t start;
+	bool fail_read_on_bb:1;
+	bool quiet_mode:1;
+};
+
+static struct badblock *dust_rb_search(struct rb_root *root, sector_t blk)
+{
+	struct rb_node *node = root->rb_node;
+
+	while (node) {
+		struct badblock *bblk = rb_entry(node, struct badblock, node);
+
+		if (bblk->bb > blk)
+			node = node->rb_left;
+		else if (bblk->bb < blk)
+			node = node->rb_right;
+		else
+			return bblk;
+	}
+
+	return NULL;
+}
+
+static bool dust_rb_insert(struct rb_root *root, struct badblock *new)
+{
+	struct badblock *bblk;
+	struct rb_node **link = &root->rb_node, *parent = NULL;
+	sector_t value = new->bb;
+
+	while (*link) {
+		parent = *link;
+		bblk = rb_entry(parent, struct badblock, node);
+
+		if (bblk->bb > value)
+			link = &(*link)->rb_left;
+		else if (bblk->bb < value)
+			link = &(*link)->rb_right;
+		else
+			return false;
+	}
+
+	rb_link_node(&new->node, parent, link);
+	rb_insert_color(&new->node, root);
+
+	return true;
+}
+
+static int dust_remove_block(struct dust_device *dd, unsigned long long block)
+{
+	struct badblock *bblock;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dd->dust_lock, flags);
+	bblock = dust_rb_search(&dd->badblocklist, block * dd->sect_per_block);
+
+	if (bblock == NULL) {
+		if (!dd->quiet_mode) {
+			DMERR("%s: block %llu not found in badblocklist",
+			      __func__, block);
+		}
+		spin_unlock_irqrestore(&dd->dust_lock, flags);
+		return -EINVAL;
+	}
+
+	rb_erase(&bblock->node, &dd->badblocklist);
+	dd->badblock_count--;
+	if (!dd->quiet_mode)
+		DMINFO("%s: badblock removed at block %llu", __func__, block);
+	kfree(bblock);
+	spin_unlock_irqrestore(&dd->dust_lock, flags);
+
+	return 0;
+}
+
+static int dust_add_block(struct dust_device *dd, unsigned long long block)
+{
+	struct badblock *bblock;
+	unsigned long flags;
+
+	bblock = kmalloc(sizeof(*bblock), GFP_KERNEL);
+	if (bblock == NULL) {
+		if (!dd->quiet_mode)
+			DMERR("%s: badblock allocation failed", __func__);
+		return -ENOMEM;
+	}
+
+	spin_lock_irqsave(&dd->dust_lock, flags);
+	bblock->bb = block * dd->sect_per_block;
+	if (!dust_rb_insert(&dd->badblocklist, bblock)) {
+		if (!dd->quiet_mode) {
+			DMERR("%s: block %llu already in badblocklist",
+			      __func__, block);
+		}
+		spin_unlock_irqrestore(&dd->dust_lock, flags);
+		kfree(bblock);
+		return -EINVAL;
+	}
+
+	dd->badblock_count++;
+	if (!dd->quiet_mode)
+		DMINFO("%s: badblock added at block %llu", __func__, block);
+	spin_unlock_irqrestore(&dd->dust_lock, flags);
+
+	return 0;
+}
+
+static int dust_query_block(struct dust_device *dd, unsigned long long block)
+{
+	struct badblock *bblock;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dd->dust_lock, flags);
+	bblock = dust_rb_search(&dd->badblocklist, block * dd->sect_per_block);
+	if (bblock != NULL)
+		DMINFO("%s: block %llu found in badblocklist", __func__, block);
+	else
+		DMINFO("%s: block %llu not found in badblocklist", __func__, block);
+	spin_unlock_irqrestore(&dd->dust_lock, flags);
+
+	return 0;
+}
+
+static int __dust_map_read(struct dust_device *dd, sector_t thisblock)
+{
+	struct badblock *bblk = dust_rb_search(&dd->badblocklist, thisblock);
+
+	if (bblk)
+		return DM_MAPIO_KILL;
+
+	return DM_MAPIO_REMAPPED;
+}
+
+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;
+
+	if (fail_read_on_bb) {
+		spin_lock_irqsave(&dd->dust_lock, flags);
+		ret = __dust_map_read(dd, thisblock);
+		spin_unlock_irqrestore(&dd->dust_lock, flags);
+	}
+
+	return ret;
+}
+
+static void __dust_map_write(struct dust_device *dd, sector_t thisblock)
+{
+	struct badblock *bblk = dust_rb_search(&dd->badblocklist, thisblock);
+
+	if (bblk) {
+		rb_erase(&bblk->node, &dd->badblocklist);
+		dd->badblock_count--;
+		kfree(bblk);
+		if (!dd->quiet_mode) {
+			sector_div(thisblock, dd->sect_per_block);
+			DMINFO("block %llu removed from badblocklist by write",
+			       (unsigned long long)thisblock);
+		}
+	}
+}
+
+static int dust_map_write(struct dust_device *dd, sector_t thisblock,
+			  bool fail_read_on_bb)
+{
+	unsigned long flags;
+
+	if (fail_read_on_bb) {
+		spin_lock_irqsave(&dd->dust_lock, flags);
+		__dust_map_write(dd, thisblock);
+		spin_unlock_irqrestore(&dd->dust_lock, flags);
+	}
+
+	return DM_MAPIO_REMAPPED;
+}
+
+static int dust_map(struct dm_target *ti, struct bio *bio)
+{
+	struct dust_device *dd = ti->private;
+	int ret;
+
+	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);
+	else
+		ret = dust_map_write(dd, bio->bi_iter.bi_sector, dd->fail_read_on_bb);
+
+	return ret;
+}
+
+static bool __dust_clear_badblocks(struct rb_root *tree,
+				   unsigned long long count)
+{
+	struct rb_node *node = NULL, *nnode = NULL;
+
+	nnode = rb_first(tree);
+	if (nnode == NULL) {
+		BUG_ON(count != 0);
+		return false;
+	}
+
+	while (nnode) {
+		node = nnode;
+		nnode = rb_next(node);
+		rb_erase(node, tree);
+		count--;
+		kfree(node);
+	}
+	BUG_ON(count != 0);
+	BUG_ON(tree->rb_node != NULL);
+
+	return true;
+}
+
+static int dust_clear_badblocks(struct dust_device *dd)
+{
+	unsigned long flags;
+	struct rb_root badblocklist;
+	unsigned long long badblock_count;
+
+	spin_lock_irqsave(&dd->dust_lock, flags);
+	badblocklist = dd->badblocklist;
+	badblock_count = dd->badblock_count;
+	dd->badblocklist = RB_ROOT;
+	dd->badblock_count = 0;
+	spin_unlock_irqrestore(&dd->dust_lock, flags);
+
+	if (!__dust_clear_badblocks(&badblocklist, badblock_count))
+		DMINFO("%s: no badblocks found", __func__);
+	else
+		DMINFO("%s: badblocks cleared", __func__);
+
+	return 0;
+}
+
+/*
+ * Target parameters:
+ *
+ * <device_path> <offset> <blksz>
+ *
+ * device_path: path to the block device
+ * offset: offset to data area from start of device_path
+ * blksz: block size (minimum 512, maximum 1073741824, must be a power of 2)
+ */
+static int dust_ctr(struct dm_target *ti, unsigned int argc, char **argv)
+{
+	struct dust_device *dd;
+	unsigned long long tmp;
+	char dummy;
+	unsigned int blksz;
+	unsigned int sect_per_block;
+	sector_t DUST_MAX_BLKSZ_SECTORS = 2097152;
+	sector_t max_block_sectors = min(ti->len, DUST_MAX_BLKSZ_SECTORS);
+
+	if (argc != 3) {
+		ti->error = "Invalid argument count";
+		return -EINVAL;
+	}
+
+	if (kstrtouint(argv[2], 10, &blksz) || !blksz) {
+		ti->error = "Invalid block size parameter";
+		return -EINVAL;
+	}
+
+	if (blksz < 512) {
+		ti->error = "Block size must be at least 512";
+		return -EINVAL;
+	}
+
+	if (!is_power_of_2(blksz)) {
+		ti->error = "Block size must be a power of 2";
+		return -EINVAL;
+	}
+
+	if (to_sector(blksz) > max_block_sectors) {
+		ti->error = "Block size is too large";
+		return -EINVAL;
+	}
+
+	sect_per_block = (blksz >> SECTOR_SHIFT);
+
+	if (sscanf(argv[1], "%llu%c", &tmp, &dummy) != 1 || tmp != (sector_t)tmp) {
+		ti->error = "Invalid device offset sector";
+		return -EINVAL;
+	}
+
+	dd = kzalloc(sizeof(struct dust_device), GFP_KERNEL);
+	if (dd == NULL) {
+		ti->error = "Cannot allocate context";
+		return -ENOMEM;
+	}
+
+	if (dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &dd->dev)) {
+		ti->error = "Device lookup failed";
+		kfree(dd);
+		return -EINVAL;
+	}
+
+	dd->sect_per_block = sect_per_block;
+	dd->blksz = blksz;
+	dd->start = tmp;
+
+	/*
+	 * Whether to fail a read on a "bad" block.
+	 * Defaults to false; enabled later by message.
+	 */
+	dd->fail_read_on_bb = false;
+
+	/*
+	 * Initialize bad block list rbtree.
+	 */
+	dd->badblocklist = RB_ROOT;
+	dd->badblock_count = 0;
+	spin_lock_init(&dd->dust_lock);
+
+	dd->quiet_mode = false;
+
+	BUG_ON(dm_set_target_max_io_len(ti, dd->sect_per_block) != 0);
+
+	ti->num_discard_bios = 1;
+	ti->num_flush_bios = 1;
+	ti->private = dd;
+
+	return 0;
+}
+
+static void dust_dtr(struct dm_target *ti)
+{
+	struct dust_device *dd = ti->private;
+
+	__dust_clear_badblocks(&dd->badblocklist, dd->badblock_count);
+	dm_put_device(ti, dd->dev);
+	kfree(dd);
+}
+
+static int dust_message(struct dm_target *ti, unsigned int argc, char **argv,
+			char *result_buf, unsigned int maxlen)
+{
+	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;
+	unsigned long long tmp, block;
+	unsigned long flags;
+	char dummy;
+
+	if (argc == 1) {
+		if (!strcasecmp(argv[0], "addbadblock") ||
+		    !strcasecmp(argv[0], "removebadblock") ||
+		    !strcasecmp(argv[0], "queryblock")) {
+			DMERR("%s requires an additional argument", argv[0]);
+		} else if (!strcasecmp(argv[0], "disable")) {
+			DMINFO("disabling read failures on bad sectors");
+			dd->fail_read_on_bb = false;
+			result = 0;
+		} else if (!strcasecmp(argv[0], "enable")) {
+			DMINFO("enabling read failures on bad sectors");
+			dd->fail_read_on_bb = true;
+			result = 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;
+		} else if (!strcasecmp(argv[0], "clearbadblocks")) {
+			result = 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;
+		} else {
+			invalid_msg = true;
+		}
+	} else if (argc == 2) {
+		if (sscanf(argv[1], "%llu%c", &tmp, &dummy) != 1)
+			return result;
+
+		block = tmp;
+		sector_div(size, dd->sect_per_block);
+		if (block > size) {
+			DMERR("selected block value out of range");
+			return result;
+		}
+
+		if (!strcasecmp(argv[0], "addbadblock"))
+			result = dust_add_block(dd, block);
+		else if (!strcasecmp(argv[0], "removebadblock"))
+			result = dust_remove_block(dd, block);
+		else if (!strcasecmp(argv[0], "queryblock"))
+			result = dust_query_block(dd, block);
+		else
+			invalid_msg = true;
+
+	} else
+		DMERR("invalid number of arguments '%d'", argc);
+
+	if (invalid_msg)
+		DMERR("unrecognized message '%s' received", argv[0]);
+
+	return result;
+}
+
+static void dust_status(struct dm_target *ti, status_type_t type,
+			unsigned int status_flags, char *result, unsigned int maxlen)
+{
+	struct dust_device *dd = ti->private;
+	unsigned int sz = 0;
+
+	switch (type) {
+	case STATUSTYPE_INFO:
+		DMEMIT("%s %s %s", dd->dev->name,
+		       dd->fail_read_on_bb ? "fail_read_on_bad_block" : "bypass",
+		       dd->quiet_mode ? "quiet" : "verbose");
+		break;
+
+	case STATUSTYPE_TABLE:
+		DMEMIT("%s %llu %u", dd->dev->name,
+		       (unsigned long long)dd->start, dd->blksz);
+		break;
+	}
+}
+
+static int dust_prepare_ioctl(struct dm_target *ti, struct block_device **bdev)
+{
+	struct dust_device *dd = ti->private;
+	struct dm_dev *dev = dd->dev;
+
+	*bdev = dev->bdev;
+
+	/*
+	 * Only pass ioctls through if the device sizes match exactly.
+	 */
+	if (dd->start ||
+	    ti->len != i_size_read(dev->bdev->bd_inode) >> SECTOR_SHIFT)
+		return 1;
+
+	return 0;
+}
+
+static int dust_iterate_devices(struct dm_target *ti, iterate_devices_callout_fn fn,
+				void *data)
+{
+	struct dust_device *dd = ti->private;
+
+	return fn(ti, dd->dev, dd->start, ti->len, data);
+}
+
+static struct target_type dust_target = {
+	.name = "dust",
+	.version = {1, 0, 0},
+	.module = THIS_MODULE,
+	.ctr = dust_ctr,
+	.dtr = dust_dtr,
+	.iterate_devices = dust_iterate_devices,
+	.map = dust_map,
+	.message = dust_message,
+	.status = dust_status,
+	.prepare_ioctl = dust_prepare_ioctl,
+};
+
+static int __init dm_dust_init(void)
+{
+	int result = dm_register_target(&dust_target);
+
+	if (result < 0)
+		DMERR("dm_register_target failed %d", result);
+
+	return result;
+}
+
+static void __exit dm_dust_exit(void)
+{
+	dm_unregister_target(&dust_target);
+}
+
+module_init(dm_dust_init);
+module_exit(dm_dust_exit);
+
+MODULE_DESCRIPTION(DM_NAME " dust test target");
+MODULE_AUTHOR("Bryan Gurney <dm-devel@redhat.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/md/dm-exception-store.h b/drivers/md/dm-exception-store.h
index 721efc493942..3f4139ac1f60 100644
--- a/drivers/md/dm-exception-store.h
+++ b/drivers/md/dm-exception-store.h
@@ -11,6 +11,7 @@
 #define _LINUX_DM_EXCEPTION_STORE
 
 #include <linux/blkdev.h>
+#include <linux/list_bl.h>
 #include <linux/device-mapper.h>
 
 /*
@@ -27,7 +28,7 @@ typedef sector_t chunk_t;
  * chunk within the device.
  */
 struct dm_exception {
-	struct list_head hash_list;
+	struct hlist_bl_node hash_list;
 
 	chunk_t old_chunk;
 	chunk_t new_chunk;
diff --git a/drivers/md/dm-init.c b/drivers/md/dm-init.c
index 4b76f84424c3..352e803f566e 100644
--- a/drivers/md/dm-init.c
+++ b/drivers/md/dm-init.c
@@ -160,7 +160,7 @@ static int __init dm_parse_table(struct dm_device *dev, char *str)
 
 	while (table_entry) {
 		DMDEBUG("parsing table \"%s\"", str);
-		if (++dev->dmi.target_count >= DM_MAX_TARGETS) {
+		if (++dev->dmi.target_count > DM_MAX_TARGETS) {
 			DMERR("too many targets %u > %d",
 			      dev->dmi.target_count, DM_MAX_TARGETS);
 			return -EINVAL;
@@ -242,9 +242,9 @@ static int __init dm_parse_devices(struct list_head *devices, char *str)
 			return -ENOMEM;
 		list_add_tail(&dev->list, devices);
 
-		if (++ndev >= DM_MAX_DEVICES) {
-			DMERR("too many targets %u > %d",
-			      dev->dmi.target_count, DM_MAX_TARGETS);
+		if (++ndev > DM_MAX_DEVICES) {
+			DMERR("too many devices %lu > %d",
+			      ndev, DM_MAX_DEVICES);
 			return -EINVAL;
 		}
 
diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c
index c27c32cf4a30..44e76cda087a 100644
--- a/drivers/md/dm-integrity.c
+++ b/drivers/md/dm-integrity.c
@@ -15,6 +15,7 @@
 #include <linux/rbtree.h>
 #include <linux/delay.h>
 #include <linux/random.h>
+#include <linux/reboot.h>
 #include <crypto/hash.h>
 #include <crypto/skcipher.h>
 #include <linux/async_tx.h>
@@ -24,6 +25,7 @@
 
 #define DEFAULT_INTERLEAVE_SECTORS	32768
 #define DEFAULT_JOURNAL_SIZE_FACTOR	7
+#define DEFAULT_SECTORS_PER_BITMAP_BIT	32768
 #define DEFAULT_BUFFER_SECTORS		128
 #define DEFAULT_JOURNAL_WATERMARK	50
 #define DEFAULT_SYNC_MSEC		10000
@@ -33,6 +35,8 @@
 #define METADATA_WORKQUEUE_MAX_ACTIVE	16
 #define RECALC_SECTORS			8192
 #define RECALC_WRITE_SUPER		16
+#define BITMAP_BLOCK_SIZE		4096	/* don't change it */
+#define BITMAP_FLUSH_INTERVAL		(10 * HZ)
 
 /*
  * Warning - DEBUG_PRINT prints security-sensitive data to the log,
@@ -48,6 +52,7 @@
 #define SB_MAGIC			"integrt"
 #define SB_VERSION_1			1
 #define SB_VERSION_2			2
+#define SB_VERSION_3			3
 #define SB_SECTORS			8
 #define MAX_SECTORS_PER_BLOCK		8
 
@@ -60,12 +65,14 @@ struct superblock {
 	__u64 provided_data_sectors;	/* userspace uses this value */
 	__u32 flags;
 	__u8 log2_sectors_per_block;
-	__u8 pad[3];
+	__u8 log2_blocks_per_bitmap_bit;
+	__u8 pad[2];
 	__u64 recalc_sector;
 };
 
 #define SB_FLAG_HAVE_JOURNAL_MAC	0x1
 #define SB_FLAG_RECALCULATING		0x2
+#define SB_FLAG_DIRTY_BITMAP		0x4
 
 #define	JOURNAL_ENTRY_ROUNDUP		8
 
@@ -151,9 +158,18 @@ struct dm_integrity_c {
 	struct workqueue_struct *metadata_wq;
 	struct superblock *sb;
 	unsigned journal_pages;
+	unsigned n_bitmap_blocks;
+
 	struct page_list *journal;
 	struct page_list *journal_io;
 	struct page_list *journal_xor;
+	struct page_list *recalc_bitmap;
+	struct page_list *may_write_bitmap;
+	struct bitmap_block_status *bbs;
+	unsigned bitmap_flush_interval;
+	int synchronous_mode;
+	struct bio_list synchronous_bios;
+	struct delayed_work bitmap_flush_work;
 
 	struct crypto_skcipher *journal_crypt;
 	struct scatterlist **journal_scatterlist;
@@ -180,6 +196,7 @@ struct dm_integrity_c {
 	__s8 log2_metadata_run;
 	__u8 log2_buffer_sectors;
 	__u8 sectors_per_block;
+	__u8 log2_blocks_per_bitmap_bit;
 
 	unsigned char mode;
 	int suspending;
@@ -232,17 +249,20 @@ struct dm_integrity_c {
 
 	bool journal_uptodate;
 	bool just_formatted;
+	bool recalculate_flag;
 
 	struct alg_spec internal_hash_alg;
 	struct alg_spec journal_crypt_alg;
 	struct alg_spec journal_mac_alg;
 
 	atomic64_t number_of_mismatches;
+
+	struct notifier_block reboot_notifier;
 };
 
 struct dm_integrity_range {
 	sector_t logical_sector;
-	unsigned n_sectors;
+	sector_t n_sectors;
 	bool waiting;
 	union {
 		struct rb_node node;
@@ -288,6 +308,16 @@ struct journal_io {
 	struct journal_completion *comp;
 };
 
+struct bitmap_block_status {
+	struct work_struct work;
+	struct dm_integrity_c *ic;
+	unsigned idx;
+	unsigned long *bitmap;
+	struct bio_list bio_queue;
+	spinlock_t bio_queue_lock;
+
+};
+
 static struct kmem_cache *journal_io_cache;
 
 #define JOURNAL_IO_MEMPOOL	32
@@ -423,7 +453,9 @@ static void wraparound_section(struct dm_integrity_c *ic, unsigned *sec_ptr)
 
 static void sb_set_version(struct dm_integrity_c *ic)
 {
-	if (ic->meta_dev || ic->sb->flags & cpu_to_le32(SB_FLAG_RECALCULATING))
+	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_FL