Reset to 3.12.37

29 files changed, 487 insertions, 314 deletions

diff --git a/drivers/md/bcache/Kconfig b/drivers/md/bcache/Kconfig
index 5eb76dd..f950c9d 100644
--- a/drivers/md/bcache/Kconfig
+++ b/drivers/md/bcache/Kconfig
@@ -1,7 +1,6 @@
 
 config BCACHE
 	tristate "Block device as cache"
-	depends on !PREEMPT_RT_FULL
 	---help---
 	Allows a block device to be used as cache for other devices; uses
 	a btree for indexing and the layout is optimized for SSDs.
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index 0f12382..7552207 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -663,9 +663,13 @@ struct gc_stat {
  * CACHE_SET_STOPPING always gets set first when we're closing down a cache set;
  * we'll continue to run normally for awhile with CACHE_SET_STOPPING set (i.e.
  * flushing dirty data).
+ *
+ * CACHE_SET_RUNNING means all cache devices have been registered and journal
+ * replay is complete.
  */
 #define CACHE_SET_UNREGISTERING		0
 #define	CACHE_SET_STOPPING		1
+#define	CACHE_SET_RUNNING		2
 
 struct cache_set {
 	struct closure		cl;
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index d1734d9..26ca4db 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -141,7 +141,7 @@ static void bch_btree_node_read_done(struct btree *b)
 	struct bset *i = b->sets[0].data;
 	struct btree_iter *iter;
 
-	iter = mempool_alloc(b->c->fill_iter, GFP_NOWAIT);
+	iter = mempool_alloc(b->c->fill_iter, GFP_NOIO);
 	iter->size = b->c->sb.bucket_size / b->c->sb.block_size;
 	iter->used = 0;
 
diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c
index 8435f81..c494379 100644
--- a/drivers/md/bcache/journal.c
+++ b/drivers/md/bcache/journal.c
@@ -42,11 +42,11 @@ static int journal_read_bucket(struct cache *ca, struct list_head *list,
 	int ret = 0;
 	sector_t bucket = bucket_to_sector(ca->set, ca->sb.d[bucket_index]);
 
-	pr_debug("reading %llu", (uint64_t) bucket);
+	pr_debug("reading %u", bucket_index);
 
 	while (offset < ca->sb.bucket_size) {
 reread:		left = ca->sb.bucket_size - offset;
-		len = min_t(unsigned, left, PAGE_SECTORS * 8);
+		len = min_t(unsigned, left, PAGE_SECTORS << JSET_BITS);
 
 		bio_reset(bio);
 		bio->bi_sector	= bucket + offset;
@@ -72,17 +72,26 @@ reread:		left = ca->sb.bucket_size - offset;
 			struct list_head *where;
 			size_t blocks, bytes = set_bytes(j);
 
-			if (j->magic != jset_magic(ca->set))
+			if (j->magic != jset_magic(ca->set)) {
+				pr_debug("%u: bad magic", bucket_index);
 				return ret;
+			}
 
-			if (bytes > left << 9)
+			if (bytes > left << 9 ||
+			    bytes > PAGE_SIZE << JSET_BITS) {
+				pr_info("%u: too big, %zu bytes, offset %u",
+					bucket_index, bytes, offset);
 				return ret;
+			}
 
 			if (bytes > len << 9)
 				goto reread;
 
-			if (j->csum != csum_set(j))
+			if (j->csum != csum_set(j)) {
+				pr_info("%u: bad csum, %zu bytes, offset %u",
+					bucket_index, bytes, offset);
 				return ret;
+			}
 
 			blocks = set_blocks(j, ca->set);
 
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 547c4c5..f5004c5 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -1235,6 +1235,9 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size)
 	if (test_bit(CACHE_SET_STOPPING, &c->flags))
 		return -EINTR;
 
+	if (!test_bit(CACHE_SET_RUNNING, &c->flags))
+		return -EPERM;
+
 	u = uuid_find_empty(c);
 	if (!u) {
 		pr_err("Can't create volume, no room for UUID");
@@ -1300,8 +1303,11 @@ static void cache_set_free(struct closure *cl)
 	bch_journal_free(c);
 
 	for_each_cache(ca, c, i)
-		if (ca)
+		if (ca) {
+			ca->set = NULL;
+			c->cache[ca->sb.nr_this_dev] = NULL;
 			kobject_put(&ca->kobj);
+		}
 
 	free_pages((unsigned long) c->uuids, ilog2(bucket_pages(c)));
 	free_pages((unsigned long) c->sort, ilog2(bucket_pages(c)));
@@ -1637,6 +1643,7 @@ static void run_cache_set(struct cache_set *c)
 
 	flash_devs_run(c);
 
+	set_bit(CACHE_SET_RUNNING, &c->flags);
 	return;
 err_unlock_gc:
 	closure_set_stopped(&c->gc.cl);
@@ -1722,8 +1729,10 @@ void bch_cache_release(struct kobject *kobj)
 {
 	struct cache *ca = container_of(kobj, struct cache, kobj);
 
-	if (ca->set)
+	if (ca->set) {
+		BUG_ON(ca->set->cache[ca->sb.nr_this_dev] != ca);
 		ca->set->cache[ca->sb.nr_this_dev] = NULL;
+	}
 
 	bch_cache_allocator_exit(ca);
 
@@ -1794,7 +1803,7 @@ err:
 }
 
 static void register_cache(struct cache_sb *sb, struct page *sb_page,
-				  struct block_device *bdev, struct cache *ca)
+				struct block_device *bdev, struct cache *ca)
 {
 	char name[BDEVNAME_SIZE];
 	const char *err = "cannot allocate memory";
diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h
index ea345c6..4dd1e2c 100644
--- a/drivers/md/bcache/util.h
+++ b/drivers/md/bcache/util.h
@@ -417,8 +417,8 @@ do {									\
 			  average_frequency,	frequency_units);	\
 	__print_time_stat(stats, name,					\
 			  average_duration,	duration_units);	\
-	__print_time_stat(stats, name,					\
-			  max_duration,		duration_units);	\
+	sysfs_print(name ## _ ##max_duration ## _ ## duration_units,	\
+			div_u64((stats)->max_duration, NSEC_PER_ ## duration_units));\
 									\
 	sysfs_print(name ## _last_ ## frequency_units, (stats)->last	\
 		    ? div_s64(local_clock() - (stats)->last,		\
diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c
index 54bdd923..93edd89 100644
--- a/drivers/md/dm-bufio.c
+++ b/drivers/md/dm-bufio.c
@@ -463,6 +463,7 @@ static void __relink_lru(struct dm_buffer *b, int dirty)
 	c->n_buffers[dirty]++;
 	b->list_mode = dirty;
 	list_move(&b->lru_list, &c->lru[dirty]);
+	b->last_accessed = jiffies;
 }
 
 /*----------------------------------------------------------------
@@ -529,6 +530,19 @@ static void use_dmio(struct dm_buffer *b, int rw, sector_t block,
 		end_io(&b->bio, r);
 }
 
+static void inline_endio(struct bio *bio, int error)
+{
+	bio_end_io_t *end_fn = bio->bi_private;
+
+	/*
+	 * Reset the bio to free any attached resources
+	 * (e.g. bio integrity profiles).
+	 */
+	bio_reset(bio);
+
+	end_fn(bio, error);
+}
+
 static void use_inline_bio(struct dm_buffer *b, int rw, sector_t block,
 			   bio_end_io_t *end_io)
 {
@@ -540,7 +554,12 @@ static void use_inline_bio(struct dm_buffer *b, int rw, sector_t block,
 	b->bio.bi_max_vecs = DM_BUFIO_INLINE_VECS;
 	b->bio.bi_sector = block << b->c->sectors_per_block_bits;
 	b->bio.bi_bdev = b->c->bdev;
-	b->bio.bi_end_io = end_io;
+	b->bio.bi_end_io = inline_endio;
+	/*
+	 * Use of .bi_private isn't a problem here because
+	 * the dm_buffer's inline bio is local to bufio.
+	 */
+	b->bio.bi_private = end_io;
 
 	/*
 	 * We assume that if len >= PAGE_SIZE ptr is page-aligned.
@@ -1417,9 +1436,9 @@ static void drop_buffers(struct dm_bufio_client *c)
 
 /*
  * Test if the buffer is unused and too old, and commit it.
- * At if noio is set, we must not do any I/O because we hold
- * dm_bufio_clients_lock and we would risk deadlock if the I/O gets rerouted to
- * different bufio client.
+ * And if GFP_NOFS is used, we must not do any I/O because we hold
+ * dm_bufio_clients_lock and we would risk deadlock if the I/O gets
+ * rerouted to different bufio client.
  */
 static int __cleanup_old_buffer(struct dm_buffer *b, gfp_t gfp,
 				unsigned long max_jiffies)
@@ -1427,7 +1446,7 @@ static int __cleanup_old_buffer(struct dm_buffer *b, gfp_t gfp,
 	if (jiffies - b->last_accessed < max_jiffies)
 		return 0;
 
-	if (!(gfp & __GFP_IO)) {
+	if (!(gfp & __GFP_FS)) {
 		if (test_bit(B_READING, &b->state) ||
 		    test_bit(B_WRITING, &b->state) ||
 		    test_bit(B_DIRTY, &b->state))
@@ -1455,9 +1474,9 @@ static long __scan(struct dm_bufio_client *c, unsigned long nr_to_scan,
 		list_for_each_entry_safe_reverse(b, tmp, &c->lru[l], lru_list) {
 			freed += __cleanup_old_buffer(b, gfp_mask, 0);
 			if (!--nr_to_scan)
-				break;
+				return freed;
+			dm_bufio_cond_resched();
 		}
-		dm_bufio_cond_resched();
 	}
 	return freed;
 }
@@ -1469,7 +1488,7 @@ dm_bufio_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
 	unsigned long freed;
 
 	c = container_of(shrink, struct dm_bufio_client, shrinker);
-	if (sc->gfp_mask & __GFP_IO)
+	if (sc->gfp_mask & __GFP_FS)
 		dm_bufio_lock(c);
 	else if (!dm_bufio_trylock(c))
 		return SHRINK_STOP;
@@ -1486,7 +1505,7 @@ dm_bufio_shrink_count(struct shrinker *shrink, struct shrink_control *sc)
 	unsigned long count;
 
 	c = container_of(shrink, struct dm_bufio_client, shrinker);
-	if (sc->gfp_mask & __GFP_IO)
+	if (sc->gfp_mask & __GFP_FS)
 		dm_bufio_lock(c);
 	else if (!dm_bufio_trylock(c))
 		return 0;
@@ -1511,7 +1530,7 @@ struct dm_bufio_client *dm_bufio_client_create(struct block_device *bdev, unsign
 	BUG_ON(block_size < 1 << SECTOR_SHIFT ||
 	       (block_size & (block_size - 1)));
 
-	c = kmalloc(sizeof(*c), GFP_KERNEL);
+	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c) {
 		r = -ENOMEM;
 		goto bad_client;
diff --git a/drivers/md/dm-cache-metadata.c b/drivers/md/dm-cache-metadata.c
index 1af7255..0bfd9c0 100644
--- a/drivers/md/dm-cache-metadata.c
+++ b/drivers/md/dm-cache-metadata.c
@@ -88,6 +88,9 @@ struct cache_disk_superblock {
 } __packed;
 
 struct dm_cache_metadata {
+	atomic_t ref_count;
+	struct list_head list;
+
 	struct block_device *bdev;
 	struct dm_block_manager *bm;
 	struct dm_space_map *metadata_sm;
@@ -114,6 +117,12 @@ struct dm_cache_metadata {
 	unsigned policy_version[CACHE_POLICY_VERSION_SIZE];
 	size_t policy_hint_size;
 	struct dm_cache_statistics stats;
+
+	/*
+	 * Reading the space map root can fail, so we read it into this
+	 * buffer before the superblock is locked and updated.
+	 */
+	__u8 metadata_space_map_root[SPACE_MAP_ROOT_SIZE];
 };
 
 /*-------------------------------------------------------------------
@@ -242,11 +251,31 @@ static void __setup_mapping_info(struct dm_cache_metadata *cmd)
 	}
 }
 
+static int __save_sm_root(struct dm_cache_metadata *cmd)
+{
+	int r;
+	size_t metadata_len;
+
+	r = dm_sm_root_size(cmd->metadata_sm, &metadata_len);
+	if (r < 0)
+		return r;
+
+	return dm_sm_copy_root(cmd->metadata_sm, &cmd->metadata_space_map_root,
+			       metadata_len);
+}
+
+static void __copy_sm_root(struct dm_cache_metadata *cmd,
+			   struct cache_disk_superblock *disk_super)
+{
+	memcpy(&disk_super->metadata_space_map_root,
+	       &cmd->metadata_space_map_root,
+	       sizeof(cmd->metadata_space_map_root));
+}
+
 static int __write_initial_superblock(struct dm_cache_metadata *cmd)
 {
 	int r;
 	struct dm_block *sblock;
-	size_t metadata_len;
 	struct cache_disk_superblock *disk_super;
 	sector_t bdev_size = i_size_read(cmd->bdev->bd_inode) >> SECTOR_SHIFT;
 
@@ -254,12 +283,16 @@ static int __write_initial_superblock(struct dm_cache_metadata *cmd)
 	if (bdev_size > DM_CACHE_METADATA_MAX_SECTORS)
 		bdev_size = DM_CACHE_METADATA_MAX_SECTORS;
 
-	r = dm_sm_root_size(cmd->metadata_sm, &metadata_len);
+	r = dm_tm_pre_commit(cmd->tm);
 	if (r < 0)
 		return r;
 
-	r = dm_tm_pre_commit(cmd->tm);
-	if (r < 0)
+	/*
+	 * dm_sm_copy_root() can fail.  So we need to do it before we start
+	 * updating the superblock.
+	 */
+	r = __save_sm_root(cmd);
+	if (r)
 		return r;
 
 	r = superblock_lock_zero(cmd, &sblock);
@@ -275,10 +308,7 @@ static int __write_initial_superblock(struct dm_cache_metadata *cmd)
 	memset(disk_super->policy_version, 0, sizeof(disk_super->policy_version));
 	disk_super->policy_hint_size = 0;
 
-	r = dm_sm_copy_root(cmd->metadata_sm, &disk_super->metadata_space_map_root,
-			    metadata_len);
-	if (r < 0)
-		goto bad_locked;
+	__copy_sm_root(cmd, disk_super);
 
 	disk_super->mapping_root = cpu_to_le64(cmd->root);
 	disk_super->hint_root = cpu_to_le64(cmd->hint_root);
@@ -295,10 +325,6 @@ static int __write_initial_superblock(struct dm_cache_metadata *cmd)
 	disk_super->write_misses = cpu_to_le32(0);
 
 	return dm_tm_commit(cmd->tm, sblock);
-
-bad_locked:
-	dm_bm_unlock(sblock);
-	return r;
 }
 
 static int __format_metadata(struct dm_cache_metadata *cmd)
@@ -384,6 +410,15 @@ static int __open_metadata(struct dm_cache_metadata *cmd)
 
 	disk_super = dm_block_data(sblock);
 
+	/* Verify the data block size hasn't changed */
+	if (le32_to_cpu(disk_super->data_block_size) != cmd->data_block_size) {
+		DMERR("changing the data block size (from %u to %llu) is not supported",
+		      le32_to_cpu(disk_super->data_block_size),
+		      (unsigned long long)cmd->data_block_size);
+		r = -EINVAL;
+		goto bad;
+	}
+
 	r = __check_incompat_features(disk_super, cmd);
 	if (r < 0)
 		goto bad;
@@ -511,8 +546,9 @@ static int __begin_transaction_flags(struct dm_cache_metadata *cmd,
 	disk_super = dm_block_data(sblock);
 	update_flags(disk_super, mutator);
 	read_superblock_fields(cmd, disk_super);
+	dm_bm_unlock(sblock);
 
-	return dm_bm_flush_and_unlock(cmd->bm, sblock);
+	return dm_bm_flush(cmd->bm);
 }
 
 static int __begin_transaction(struct dm_cache_metadata *cmd)
@@ -540,7 +576,6 @@ static int __commit_transaction(struct dm_cache_metadata *cmd,
 				flags_mutator mutator)
 {
 	int r;
-	size_t metadata_len;
 	struct cache_disk_superblock *disk_super;
 	struct dm_block *sblock;
 
@@ -558,8 +593,8 @@ static int __commit_transaction(struct dm_cache_metadata *cmd,
 	if (r < 0)
 		return r;
 
-	r = dm_sm_root_size(cmd->metadata_sm, &metadata_len);
-	if (r < 0)
+	r = __save_sm_root(cmd);
+	if (r)
 		return r;
 
 	r = superblock_lock(cmd, &sblock);
@@ -586,13 +621,7 @@ static int __commit_transaction(struct dm_cache_metadata *cmd,
 	disk_super->read_misses = cpu_to_le32(cmd->stats.read_misses);
 	disk_super->write_hits = cpu_to_le32(cmd->stats.write_hits);
 	disk_super->write_misses = cpu_to_le32(cmd->stats.write_misses);
-
-	r = dm_sm_copy_root(cmd->metadata_sm, &disk_super->metadata_space_map_root,
-			    metadata_len);
-	if (r < 0) {
-		dm_bm_unlock(sblock);
-		return r;
-	}
+	__copy_sm_root(cmd, disk_super);
 
 	return dm_tm_commit(cmd->tm, sblock);
 }
@@ -624,10 +653,10 @@ static void unpack_value(__le64 value_le, dm_oblock_t *block, unsigned *flags)
 
 /*----------------------------------------------------------------*/
 
-struct dm_cache_metadata *dm_cache_metadata_open(struct block_device *bdev,
-						 sector_t data_block_size,
-						 bool may_format_device,
-						 size_t policy_hint_size)
+static struct dm_cache_metadata *metadata_open(struct block_device *bdev,
+					       sector_t data_block_size,
+					       bool may_format_device,
+					       size_t policy_hint_size)
 {
 	int r;
 	struct dm_cache_metadata *cmd;
@@ -638,6 +667,7 @@ struct dm_cache_metadata *dm_cache_metadata_open(struct block_device *bdev,
 		return NULL;
 	}
 
+	atomic_set(&cmd->ref_count, 1);
 	init_rwsem(&cmd->root_lock);
 	cmd->bdev = bdev;
 	cmd->data_block_size = data_block_size;
@@ -660,10 +690,95 @@ struct dm_cache_metadata *dm_cache_metadata_open(struct block_device *bdev,
 	return cmd;
 }
 
+/*
+ * We keep a little list of ref counted metadata objects to prevent two
+ * different target instances creating separate bufio instances.  This is
+ * an issue if a table is reloaded before the suspend.
+ */
+static DEFINE_MUTEX(table_lock);
+static LIST_HEAD(table);
+
+static struct dm_cache_metadata *lookup(struct block_device *bdev)
+{
+	struct dm_cache_metadata *cmd;
+
+	list_for_each_entry(cmd, &table, list)
+		if (cmd->bdev == bdev) {
+			atomic_inc(&cmd->ref_count);
+			return cmd;
+		}
+
+	return NULL;
+}
+
+static struct dm_cache_metadata *lookup_or_open(struct block_device *bdev,
+						sector_t data_block_size,
+						bool may_format_device,
+						size_t policy_hint_size)
+{
+	struct dm_cache_metadata *cmd, *cmd2;
+
+	mutex_lock(&table_lock);
+	cmd = lookup(bdev);
+	mutex_unlock(&table_lock);
+
+	if (cmd)
+		return cmd;
+
+	cmd = metadata_open(bdev, data_block_size, may_format_device, policy_hint_size);
+	if (cmd) {
+		mutex_lock(&table_lock);
+		cmd2 = lookup(bdev);
+		if (cmd2) {
+			mutex_unlock(&table_lock);
+			__destroy_persistent_data_objects(cmd);
+			kfree(cmd);
+			return cmd2;
+		}
+		list_add(&cmd->list, &table);
+		mutex_unlock(&table_lock);
+	}
+
+	return cmd;
+}
+
+static bool same_params(struct dm_cache_metadata *cmd, sector_t data_block_size)
+{
+	if (cmd->data_block_size != data_block_size) {
+		DMERR("data_block_size (%llu) different from that in metadata (%llu)\n",
+		      (unsigned long long) data_block_size,
+		      (unsigned long long) cmd->data_block_size);
+		return false;
+	}
+
+	return true;
+}
+
+struct dm_cache_metadata *dm_cache_metadata_open(struct block_device *bdev,
+						 sector_t data_block_size,
+						 bool may_format_device,
+						 size_t policy_hint_size)
+{
+	struct dm_cache_metadata *cmd = lookup_or_open(bdev, data_block_size,
+						       may_format_device, policy_hint_size);
+	if (cmd && !same_params(cmd, data_block_size)) {
+		dm_cache_metadata_close(cmd);
+		return NULL;
+	}
+
+	return cmd;
+}
+
 void dm_cache_metadata_close(struct dm_cache_metadata *cmd)
 {
-	__destroy_persistent_data_objects(cmd);
-	kfree(cmd);
+	if (atomic_dec_and_test(&cmd->ref_count)) {
+		mutex_lock(&table_lock);
+		list_del(&cmd->list);
+		mutex_unlock(&table_lock);
+
+		__destroy_persistent_data_objects(cmd);
+		kfree(cmd);
+	}
 }
 
 int dm_cache_resize(struct dm_cache_metadata *cmd, dm_cblock_t new_cache_size)
diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c
index 6ab68e0..1771845 100644
--- a/drivers/md/dm-cache-target.c
+++ b/drivers/md/dm-cache-target.c
@@ -146,7 +146,13 @@ struct cache {
 	struct list_head need_commit_migrations;
 	sector_t migration_threshold;
 	wait_queue_head_t migration_wait;
-	atomic_t nr_migrations;
+	atomic_t nr_allocated_migrations;
+
+	/*
+	 * The number of in flight migrations that are performing
+	 * background io. eg, promotion, writeback.
+	 */
+	atomic_t nr_io_migrations;
 
 	wait_queue_head_t quiescing_wait;
 	atomic_t quiescing_ack;
@@ -154,7 +160,7 @@ struct cache {
 	/*
 	 * cache_size entries, dirty if set
 	 */
-	dm_cblock_t nr_dirty;
+	atomic_t nr_dirty;
 	unsigned long *dirty_bitset;
 
 	/*
@@ -162,7 +168,7 @@ struct cache {
 	 */
 	dm_dblock_t discard_nr_blocks;
 	unsigned long *discard_bitset;
-	uint32_t discard_block_size; /* a power of 2 times sectors per block */
+	uint32_t discard_block_size;
 
 	/*
 	 * Rather than reconstructing the table line for the status we just
@@ -182,7 +188,6 @@ struct cache {
 	struct dm_deferred_set *all_io_ds;
 
 	mempool_t *migration_pool;
-	struct dm_cache_migration *next_migration;
 
 	struct dm_cache_policy *policy;
 	unsigned policy_nr_args;
@@ -265,10 +270,31 @@ static void free_prison_cell(struct cache *cache, struct dm_bio_prison_cell *cel
 	dm_bio_prison_free_cell(cache->prison, cell);
 }
 
+static struct dm_cache_migration *alloc_migration(struct cache *cache)
+{
+	struct dm_cache_migration *mg;
+
+	mg = mempool_alloc(cache->migration_pool, GFP_NOWAIT);
+	if (mg) {
+		mg->cache = cache;
+		atomic_inc(&mg->cache->nr_allocated_migrations);
+	}
+
+	return mg;
+}
+
+static void free_migration(struct dm_cache_migration *mg)
+{
+	if (atomic_dec_and_test(&mg->cache->nr_allocated_migrations))
+		wake_up(&mg->cache->migration_wait);
+
+	mempool_free(mg, mg->cache->migration_pool);
+}
+
 static int prealloc_data_structs(struct cache *cache, struct prealloc *p)
 {
 	if (!p->mg) {
-		p->mg = mempool_alloc(cache->migration_pool, GFP_NOWAIT);
+		p->mg = alloc_migration(cache);
 		if (!p->mg)
 			return -ENOMEM;
 	}
@@ -297,7 +323,7 @@ static void prealloc_free_structs(struct cache *cache, struct prealloc *p)
 		free_prison_cell(cache, p->cell1);
 
 	if (p->mg)
-		mempool_free(p->mg, cache->migration_pool);
+		free_migration(p->mg);
 }
 
 static struct dm_cache_migration *prealloc_get_migration(struct prealloc *p)
@@ -408,7 +434,7 @@ static bool is_dirty(struct cache *cache, dm_cblock_t b)
 static void set_dirty(struct cache *cache, dm_oblock_t oblock, dm_cblock_t cblock)
 {
 	if (!test_and_set_bit(from_cblock(cblock), cache->dirty_bitset)) {
-		cache->nr_dirty = to_cblock(from_cblock(cache->nr_dirty) + 1);
+		atomic_inc(&cache->nr_dirty);
 		policy_set_dirty(cache->policy, oblock);
 	}
 }
@@ -417,8 +443,7 @@ static void clear_dirty(struct cache *cache, dm_oblock_t oblock, dm_cblock_t cbl
 {
 	if (test_and_clear_bit(from_cblock(cblock), cache->dirty_bitset)) {
 		policy_clear_dirty(cache->policy, oblock);
-		cache->nr_dirty = to_cblock(from_cblock(cache->nr_dirty) - 1);
-		if (!from_cblock(cache->nr_dirty))
+		if (atomic_dec_return(&cache->nr_dirty) == 0)
 			dm_table_event(cache->ti->table);
 	}
 }
@@ -709,24 +734,14 @@ static void remap_to_origin_then_cache(struct cache *cache, struct bio *bio,
  * Migration covers moving data from the origin device to the cache, or
  * vice versa.
  *--------------------------------------------------------------*/
-static void free_migration(struct dm_cache_migration *mg)
+static void inc_io_migrations(struct cache *cache)
 {
-	mempool_free(mg, mg->cache->migration_pool);
+	atomic_inc(&cache->nr_io_migrations);
 }
 
-static void inc_nr_migrations(struct cache *cache)
+static void dec_io_migrations(struct cache *cache)
 {
-	atomic_inc(&cache->nr_migrations);
-}
-
-static void dec_nr_migrations(struct cache *cache)
-{
-	atomic_dec(&cache->nr_migrations);
-
-	/*
-	 * Wake the worker in case we're suspending the target.
-	 */
-	wake_up(&cache->migration_wait);
+	atomic_dec(&cache->nr_io_migrations);
 }
 
 static void __cell_defer(struct cache *cache, struct dm_bio_prison_cell *cell,
@@ -749,11 +764,10 @@ static void cell_defer(struct cache *cache, struct dm_bio_prison_cell *cell,
 	wake_worker(cache);
 }
 
-static void cleanup_migration(struct dm_cache_migration *mg)
+static void free_io_migration(struct dm_cache_migration *mg)
 {
-	struct cache *cache = mg->cache;
+	dec_io_migrations(mg->cache);
 	free_migration(mg);
-	dec_nr_migrations(cache);
 }
 
 static void migration_failure(struct dm_cache_migration *mg)
@@ -778,7 +792,7 @@ static void migration_failure(struct dm_cache_migration *mg)
 		cell_defer(cache, mg->new_ocell, 1);
 	}
 
-	cleanup_migration(mg);
+	free_io_migration(mg);
 }
 
 static void migration_success_pre_commit(struct dm_cache_migration *mg)
@@ -787,9 +801,9 @@ static void migration_success_pre_commit(struct dm_cache_migration *mg)
 	struct cache *cache = mg->cache;
 
 	if (mg->writeback) {
-		cell_defer(cache, mg->old_ocell, false);
 		clear_dirty(cache, mg->old_oblock, mg->cblock);
-		cleanup_migration(mg);
+		cell_defer(cache, mg->old_ocell, false);
+		free_io_migration(mg);
 		return;
 
 	} else if (mg->demote) {
@@ -799,14 +813,14 @@ static void migration_success_pre_commit(struct dm_cache_migration *mg)
 					     mg->old_oblock);
 			if (mg->promote)
 				cell_defer(cache, mg->new_ocell, true);
-			cleanup_migration(mg);
+			free_io_migration(mg);
 			return;
 		}
 	} else {
 		if (dm_cache_insert_mapping(cache->cmd, mg->cblock, mg->new_oblock)) {
 			DMWARN_LIMIT("promotion failed; couldn't update on disk metadata");
 			policy_remove_mapping(cache->policy, mg->new_oblock);
-			cleanup_migration(mg);
+			free_io_migration(mg);
 			return;
 		}
 	}
@@ -837,12 +851,12 @@ static void migration_success_post_commit(struct dm_cache_migration *mg)
 			spin_unlock_irqrestore(&cache->lock, flags);
 
 		} else
-			cleanup_migration(mg);
+			free_io_migration(mg);
 
 	} else {
-		cell_defer(cache, mg->new_ocell, true);
 		clear_dirty(cache, mg->new_oblock, mg->cblock);
-		cleanup_migration(mg);
+		cell_defer(cache, mg->new_ocell, true);
+		free_io_migration(mg);
 	}
 }
 
@@ -1003,7 +1017,7 @@ static void promote(struct cache *cache, struct prealloc *structs,
 	mg->new_ocell = cell;
 	mg->start_jiffies = jiffies;
 
-	inc_nr_migrations(cache);
+	inc_io_migrations(cache);
 	quiesce_migration(mg);
 }
 
@@ -1024,7 +1038,7 @@ static void writeback(struct cache *cache, struct prealloc *structs,
 	mg->new_ocell = NULL;
 	mg->start_jiffies = jiffies;
 
-	inc_nr_migrations(cache);
+	inc_io_migrations(cache);
 	quiesce_migration(mg);
 }
 
@@ -1048,7 +1062,7 @@ static void demote_then_promote(struct cache *cache, struct prealloc *structs,
 	mg->new_ocell = new_ocell;
 	mg->start_jiffies = jiffies;
 
-	inc_nr_migrations(cache);
+	inc_io_migrations(cache);
 	quiesce_migration(mg);
 }
 
@@ -1109,7 +1123,7 @@ static void process_discard_bio(struct cache *cache, struct bio *bio)
 
 static bool spare_migration_bandwidth(struct cache *cache)
 {
-	sector_t current_volume = (atomic_read(&cache->nr_migrations) + 1) *
+	sector_t current_volume = (atomic_read(&cache->nr_io_migrations) + 1) *
 		cache->sectors_per_block;
 	return current_volume < cache->migration_threshold;
 }
@@ -1400,7 +1414,7 @@ static void stop_quiescing(struct cache *cache)
 
 static void wait_for_migrations(struct cache *cache)
 {
-	wait_event(cache->migration_wait, !atomic_read(&cache->nr_migrations));
+	wait_event(cache->migration_wait, !atomic_read(&cache->nr_allocated_migrations));
 }
 
 static void stop_worker(struct cache *cache)
@@ -1509,9 +1523,6 @@ static void destroy(struct cache *cache)
 {
 	unsigned i;
 
-	if (cache->next_migration)
-		mempool_free(cache->next_migration, cache->migration_pool);
-
 	if (cache->migration_pool)
 		mempool_destroy(cache->migration_pool);
 
@@ -1908,35 +1919,6 @@ static int create_cache_policy(struct cache *cache, struct cache_args *ca,
 	return 0;
 }
 
-/*
- * We want the discard block size to be a power of two, at least the size
- * of the cache block size, and have no more than 2^14 discard blocks
- * across the origin.
- */
-#define MAX_DISCARD_BLOCKS (1 << 14)
-
-static bool too_many_discard_blocks(sector_t discard_block_size,
-				    sector_t origin_size)
-{
-	(void) sector_div(origin_size, discard_block_size);
-
-	return origin_size > MAX_DISCARD_BLOCKS;
-}
-
-static sector_t calculate_discard_block_size(sector_t cache_block_size,
-					     sector_t origin_size)
-{
-	sector_t discard_block_size;
-
-	discard_block_size = roundup_pow_of_two(cache_block_size);
-
-	if (origin_size)
-		while (too_many_discard_blocks(discard_block_size, origin_size))
-			discard_block_size *= 2;
-
-	return discard_block_size;
-}
-
 #define DEFAULT_MIGRATION_THRESHOLD 2048
 
 static int cache_create(struct cache_args *ca, struct cache **result)
@@ -1961,6 +1943,8 @@ static int cache_create(struct cache_args *ca, struct cache **result)
 	ti->num_discard_bios = 1;
 	ti->discards_supported = true;
 	ti->discard_zeroes_data_unsupported = true;
+	/* Discard bios must be split on a block boundary */
+	ti->split_discard_bios = true;
 
 	cache->features = ca->features;
 	ti->per_bio_data_size = get_per_bio_data_size(cache);
@@ -2026,14 +2010,15 @@ static int cache_create(struct cache_args *ca, struct cache **result)
 	INIT_LIST_HEAD(&cache->quiesced_migrations);
 	INIT_LIST_HEAD(&cache->completed_migrations);
 	INIT_LIST_HEAD(&cache->need_commit_migrations);
-	atomic_set(&cache->nr_migrations, 0);
+	atomic_set(&cache->nr_allocated_migrations, 0);
+	atomic_set(&cache->nr_io_migrations, 0);
 	init_waitqueue_head(&cache->migration_wait);
 
 	init_waitqueue_head(&cache->quiescing_wait);
 	atomic_set(&cache->quiescing_ack, 0);
 
 	r = -ENOMEM;
-	cache->nr_dirty = 0;
+	atomic_set(&cache->nr_dirty, 0);
 	cache->dirty_bitset = alloc_bitset(from_cblock(cache->cache_size));
 	if (!cache->dirty_bitset) {
 		*error = "could not allocate dirty bitset";
@@ -2041,9 +2026,7 @@ static int cache_create(struct cache_args *ca, struct cache **result)
 	}
 	clear_bitset(cache->dirty_bitset, from_cblock(cache->cache_size));
 
-	cache->discard_block_size =
-		calculate_discard_block_size(cache->sectors_per_block,
-					     cache->origin_sectors);
+	cache->discard_block_size = cache->sectors_per_block;
 	cache->discard_nr_blocks = oblock_to_dblock(cache, cache->origin_blocks);
 	cache->discard_bitset = alloc_bitset(from_dblock(cache->discard_nr_blocks));
 	if (!cache->discard_bitset) {
@@ -2087,8 +2070,6 @@ static int cache_create(struct cache_args *ca, struct cache **result)
 		goto bad;
 	}
 
-	cache->next_migration = NULL;
-
 	cache->need_tick_bio = true;
 	cache->sized = false;
 	cache->quiescing = false;
@@ -2531,7 +2512,7 @@ static void cache_status(struct dm_target *ti, status_type_t type,
 
 		residency = policy_residency(cache->policy);
 
-		DMEMIT("%llu/%llu %u %u %u %u %u %u %llu %u ",
+		DMEMIT("%llu/%llu %u %u %u %u %u %u %llu %lu ",
 		       (unsigned long long)(nr_blocks_metadata - nr_free_blocks_metadata),
 		       (unsigned long long)nr_blocks_metadata,
 		       (unsigned) atomic_read(&cache->stats.read_hit),
@@ -2541,7 +2522,7 @@ static void cache_status(struct dm_target *ti, status_type_t type,
 		       (unsigned) atomic_read(&cache->stats.demotion),
 		       (unsigned) atomic_read(&cache->stats.promotion),
 		       (unsigned long long) from_cblock(residency),
-		       cache->nr_dirty);
+		       (unsigned long) atomic_read(&cache->nr_dirty));
 
 		if (cache->features.write_through)
 			DMEMIT("1 writethrough ");
@@ -2630,7 +2611,7 @@ static void set_discard_limits(struct cache *cache, struct queue_limits *limits)
 	/*
 	 * FIXME: these limits may be incompatible with the cache device
 	 */
-	limits->max_discard_sectors = cache->discard_block_size * 1024;
+	limits->max_discard_sectors = cache->discard_block_size;
 	limits->discard_granularity = cache->discard_block_size << SECTOR_SHIFT;
 }
 
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 0fce0bc..0f64dc5 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -18,7 +18,6 @@
 #include <linux/crypto.h>
 #include <linux/workqueue.h>
 #include <linux/backing-dev.h>
-#include <linux/percpu.h>
 #include <linux/atomic.h>
 #include <linux/scatterlist.h>
 #include <asm/page.h>
@@ -44,6 +43,7 @@ struct convert_context {
 	unsigned int idx_out;
 	sector_t cc_sector;
 	atomic_t cc_pending;
+	struct ablkcipher_request *req;
 };
 
 /*
@@ -105,15 +105,7 @@ struct iv_lmk_private {
 enum flags { DM_CRYPT_SUSPENDED, DM_CRYPT_KEY_VALID };
 
 /*
- * Duplicated per-CPU state for cipher.
- */
-struct crypt_cpu {
-	struct ablkcipher_request *req;
-};
-
-/*
- * The fields in here must be read only after initialization,
- * changing state should be in crypt_cpu.
+ * The fields in here must be read only after initialization.
  */
 struct crypt_config {
 	struct dm_dev *dev;
@@ -143,12 +135,6 @@ struct crypt_config {
 	sector_t iv_offset;
 	unsigned int iv_size;
 
-	/*
-	 * Duplicated per cpu state. Access through
-	 * per_cpu_ptr() only.
-	 */
-	struct crypt_cpu __percpu *cpu;
-
 	/* ESSIV: struct crypto_cipher *essiv_tfm */
 	void *iv_private;
 	struct crypto_ablkcipher **tfms;
@@ -184,11 +170,6 @@ static void clone_init(struct dm_crypt_io *, struct bio *);
 static void kcryptd_queue_crypt(struct dm_crypt_io *io);
 static u8 *iv_of_dmreq(struct crypt_config *cc, struct dm_crypt_request *dmreq);
 
-static struct crypt_cpu *this_crypt_config(struct crypt_config *cc)
-{
-	return this_cpu_ptr(cc->cpu);
-}
-
 /*
  * Use this to access cipher attributes that are the same for each CPU.
  */
@@ -738,16 +719,15 @@ static void kcryptd_async_done(struct crypto_async_request *async_req,
 static void crypt_alloc_req(struct crypt_config *cc,
 			    struct convert_context *ctx)
 {
-	struct crypt_cpu *this_cc = this_crypt_config(cc);
 	unsigned key_index = ctx->cc_sector & (cc->tfms_count - 1);
 
-	if (!this_cc->req)
-		this_cc->req = mempool_alloc(cc->req_pool, GFP_NOIO);
+	if (!ctx->req)
+		ctx->req = mempool_alloc(cc->req_pool, GFP_NOIO);
 
-	ablkcipher_request_set_tfm(this_cc->req, cc->tfms[key_index]);
-	ablkcipher_request_set_callback(this_cc->req,
+	ablkcipher_request_set_tfm(ctx->req, cc->tfms[key_index]);
+	ablkcipher_request_set_callback(ctx->req,
 	    CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
-	    kcryptd_async_done, dmreq_of_req(cc, this_cc->req));
+	    kcryptd_async_done, dmreq_of_req(cc, ctx->req));
 }
 
 /*
@@ -756,7 +736,6 @@ static void crypt_alloc_req(struct crypt_config *cc,
 static int crypt_convert(struct crypt_config *cc,
 			 struct convert_context *ctx)
 {
-	struct crypt_cpu *this_cc = this_crypt_config(cc);
 	int r;
 
 	atomic_set(&ctx->cc_pending, 1);
@@ -768,7 +747,7 @@ static int crypt_convert(struct crypt_config *cc,
 
 		atomic_inc(&ctx->cc_pending);
 
-		r = crypt_convert_block(cc, ctx, this_cc->req);
+		r = crypt_convert_block(cc, ctx, ctx->req);
 
 		switch (r) {
 		/* async */
@@ -777,7 +756,7 @@ static int crypt_convert(struct crypt_config *cc,
 			INIT_COMPLETION(ctx->restart);
 			/* fall through*/
 		case -EINPROGRESS:
-			this_cc->req = NULL;
+			ctx->req = NULL;
 			ctx->cc_sector++;
 			continue;
 
@@ -876,6 +855,7 @@ static struct dm_crypt_io *crypt_io_alloc(struct crypt_config *cc,
 	io->sector = sector;
 	io->error = 0;
 	io->base_io = NULL;
+	io->ctx.req = NULL;
 	atomic_set(&io->io_pending, 0);
 
 	return io;
@@ -901,6 +881,8 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
 	if (!atomic_dec_and_test(&io->io_pending))
 		return;
 
+	if (io->ctx.req)
+		mempool_free(io->ctx.req, cc->req_pool);
 	mempool_free(io, cc->io_pool);
 
 	if (likely(!base_io))
@@ -1326,8 +1308,6 @@ static int crypt_wipe_key(struct crypt_config *cc)
 static void crypt_dtr(struct dm_target *ti)
 {
 	struct crypt_config *cc = ti->private;
-	struct crypt_cpu *cpu_cc;
-	int cpu;
 
 	ti->private = NULL;
 
@@ -1339,13 +1319,6 @@ static void crypt_dtr(struct dm_target *ti)
 	if (cc->crypt_queue)
 		destroy_workqueue(cc->crypt_queue);
 
-	if (cc->cpu)
-		for_each_possible_cpu(cpu) {
-			cpu_cc = per_cpu_ptr(cc->cpu, cpu);
-			if (cpu_cc->req)
-				mempool_free(cpu_cc->req, cc->req_pool);
-		}
-
 	crypt_free_tfms(cc);
 
 	if (cc->bs)
@@ -1364,9 +1337,6 @@ static void crypt_dtr(struct dm_target *ti)
 	if (cc->dev)
 		dm_put_device(ti, cc->dev);
 
-	if (cc->cpu)
-		free_percpu(cc->cpu);
-
 	kzfree(cc->cipher);
 	kzfree(cc->cipher_string);
 
@@ -1421,13 +1391,6 @@ static int crypt_ctr_cipher(struct dm_target *ti,
 	if (tmp)
 		DMWARN("Ignoring unexpected additional cipher options");
 
-	cc->cpu = __alloc_percpu(sizeof(*(cc->cpu)),
-				 __alignof__(struct crypt_cpu));
-	if (!cc->cpu) {
-		ti->error = "Cannot allocate per cpu state";
-		goto bad_mem;
-	}
-
 	/*
 	 * For compatibility with the original dm-crypt mapping format, if
 	 * only the cipher name is supplied, use cbc-plain.
@@ -1543,6 +1506,7 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	unsigned int key_size, opt_params;
 	unsigned long long tmpll;
 	int ret;
+	size_t iv_size_padding;
 	struct dm_arg_set as;
 	const char *opt_string;
 	char dummy;
@@ -1579,12 +1543,23 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 
 	cc->dmreq_start = sizeof(struct ablkcipher_request);
 	cc->dmreq_start += crypto_ablkcipher_reqsize(any_tfm(cc));
-	cc->dmreq_start = ALIGN(cc->dmreq_start, crypto_tfm_ctx_alignment());
-	cc->dmreq_start += crypto_ablkcipher_alignmask(any_tfm(cc)) &
-			   ~(crypto_tfm_ctx_alignment() - 1);
+	cc->dmreq_start = ALIGN(cc->dmreq_start, __alignof__(struct dm_crypt_request));
+
+	if (crypto_ablkcipher_alignmask(any_tfm(cc)) < CRYPTO_MINALIGN) {
+		/* Allocate the padding exactly */
+		iv_size_padding = -(cc->dmreq_start + sizeof(struct dm_crypt_request))
+				& crypto_ablkcipher_alignmask(any_tfm(cc));
+	} else {
+		/*
+		 * If the cipher requires greater alignment than kmalloc
+		 * alignment, we don't know the exact position of the
+		 * initialization vector. We must assume worst case.
+		 */
+		iv_size_padding = crypto_ablkcipher_alignmask(any_tfm(cc));
+	}
 
 	cc->req_pool = mempool_create_kmalloc_pool(MIN_IOS, cc->dmreq_start +
-			sizeof(struct dm_crypt_request) + cc->iv_size);
+			sizeof(struct dm_crypt_request) + iv_size_padding + cc->iv_size);
 	if (!cc->req_pool) {
 		ti->error = "Cannot allocate crypt request mempool";
 		goto bad;
diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index 2a20986..e60c2ea 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -10,6 +10,7 @@
 #include <linux/device-mapper.h>
 
 #include <linux/bio.h>
+#include <linux/completion.h>
 #include <linux/mempool.h>
 #include <linux/module.h>
 #include <linux/sched.h>
@@ -32,7 +33,7 @@ struct dm_io_client {
 struct io {
 	unsigned long error_bits;
 	atomic_t count;
-	struct task_struct *sleeper;
+	struct completion *wait;
 	struct dm_io_client *client;
 	io_notify_fn callback;
 	void *context;
@@ -121,8 +122,8 @@ static void dec_count(struct io *io, unsigned int region, int error)
 			invalidate_kernel_vmap_range(io->vma_invalidate_address,
 						     io->vma_invalidate_size);
 
-		if (io->sleeper)
-			wake_up_process(io->sleeper);
+		if (io->wait)
+			complete(io->wait);
 
 		else {
 			unsigned long r = io->error_bits;
@@ -385,6 +386,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 	 */
 	volatile char io_[sizeof(struct io) + __alignof__(struct io) - 1];
 	struct io *io = (struct io *)PTR_ALIGN(&io_, __alignof__(struct io));
+	DECLARE_COMPLETION_ONSTACK(wait);
 
 	if (num_regions > 1 && (rw & RW_MASK) != WRITE) {
 		WARN_ON(1);
@@ -393,7 +395,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 
 	io->error_bits = 0;
 	atomic_set(&io->count, 1); /* see dispatch_io() */
-	io->sleeper = current;
+	io->wait = &wait;
 	io->client = client;
 
 	io->vma_invalidate_address = dp->vma_invalidate_address;
@@ -401,15 +403,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 
 	dispatch_io(rw, num_regions, where, dp, io, 1);
 
-	while (1) {
-		set_current_state(TASK_UNINTERRUPTIBLE);
-
-		if (!atomic_read(&io->count))
-			break;
-
-		io_schedule();
-	}
-	set_current_state(TASK_RUNNING);
+	wait_for_completion_io(&wait);
 
 	if (error_bits)
 		*error_bits = io->error_bits;
@@ -432,7 +426,7 @@ static int async_io(struct dm_io_client *client, unsigned int num_regions,
 	io = mempool_alloc(client->pool, GFP_NOIO);
 	io->error_bits = 0;
 	atomic_set(&io->count, 1); /* see dispatch_io() */
-	io->sleeper = NULL;
+	io->wait = NULL;
 	io->client = client;
 	io->callback = fn;
 	io->context = context;
diff --git a/drivers/md/dm-log-userspace-transfer.c b/drivers/md/dm-log-userspace-transfer.c
index 08d9a20..c69d0b7 100644
--- a/drivers/md/dm-log-userspace-transfer.c
+++ b/drivers/md/dm-log-userspace-transfer.c
@@ -272,7 +272,7 @@ int dm_ulog_tfr_init(void)
 
 	r = cn_add_callback(&ulog_cn_id, "dmlogusr", cn_ulog_callback);
 	if (r) {
-		cn_del_callback(&ulog_cn_id);
+		kfree(prealloced_cn_msg);
 		return r;
 	}
 
diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
index 4880b69..5971538 100644
--- a/drivers/md/dm-raid.c
+++ b/drivers/md/dm-raid.c
@@ -785,8 +785,7 @@ struct dm_raid_superblock {
 	__le32 layout;
 	__le32 stripe_sectors;
 
-	__u8 pad[452];		/* Round struct to 512 bytes. */
-				/* Always set to 0 when writing. */
+	/* Remainder of a logical block is zero-filled when writing (see super_sync()). */
 } __packed;
 
 static int read_disk_sb(struct md_rdev *rdev, int size)
@@ -823,7 +822,7 @@ static void super_sync(struct mddev *mddev, struct md_rdev *rdev)
 		    test_bit(Faulty, &(rs->dev[i].rdev.flags)))
 			failed_devices |= (1ULL << i);
 
-	memset(sb, 0, sizeof(*sb));
+	memset(sb + 1, 0, rdev->sb_size - sizeof(*sb));
 
 	sb->magic = cpu_to_le32(DM_RAID_MAGIC);
 	sb->features = cpu_to_le32(0);	/* No features yet */
@@ -858,7 +857,11 @@ static int super_load(struct md_rdev *rdev, struct md_rdev *refdev)
 	uint64_t events_sb, events_refsb;
 
 	rdev->sb_start = 0;
-	rdev->sb_size = sizeof(*sb);
+	rdev->sb_size = bdev_logical_block_size(rdev->meta_bdev);
+	if (rdev->sb_size < sizeof(*sb) || rdev->sb_size > PAGE_SIZE) {
+		DMERR("superblock size of a logical block is no longer valid");
+		return -EINVAL;
+	}
 
 	ret = read_disk_sb(rdev, rdev->sb_size);
 	if (ret)
diff --git a/drivers/md/dm-thin-metadata.c b/drivers/md/dm-thin-metadata.c
index 3bb4506..b63095c 100644
--- a/drivers/md/dm-thin-metadata.c
+++ b/drivers/md/dm-thin-metadata.c
@@ -192,6 +192,13 @@ struct dm_pool_metadata {
 	 * operation possible in this state is the closing of the device.
 	 */
 	bool fail_io:1;
+
+	/*
+	 * Reading the space map roots can fail, so we read it into these
+	 * buffers before the superblock is locked and updated.
+	 */
+	__u8 data_space_map_root[SPACE_MAP_ROOT_SIZE];
+	__u8 metadata_space_map_root[SPACE_MAP_ROOT_SIZE];
 };
 
 struct dm_thin_device {
@@ -431,26 +438,53 @@ static void __setup_btree_details(struct dm_pool_metadata *pmd)
 	pmd->details_info.value_type.equal = NULL;
 }
 
+static int save_sm_roots(struct dm_pool_metadata *pmd)
+{
+	int r;
+	size_t len;
+
+	r = dm_sm_root_size(pmd->metadata_sm, &len);
+	if (r < 0)
+		return r;
+
+	r = dm_sm_copy_root(pmd->metadata_sm, &pmd->metadata_space_map_root, len);
+	if (r < 0)
+		return r;
+
+	r = dm_sm_root_size(pmd->data_sm, &len);
+	if (r < 0)
+		return r;
+
+	return dm_sm_copy_root(pmd->data_sm, &pmd->data_space_map_root, len);
+}
+
+static void copy_sm_roots(struct dm_pool_metadata *pmd,
+			  struct thin_disk_superblock *disk)
+{
+	memcpy(&disk->metadata_space_map_root,
+	       &pmd->metadata_space_map_root,
+	       sizeof(pmd->metadata_space_map_root));
+
+	memcpy(&disk->data_space_map_root,
+	       &pmd->data_space_map_root,
+	       sizeof(pmd->data_space_map_root));
+}
+
 static int __write_initial_superblock(struct dm_pool_metadata *pmd)
 {
 	int r;
 	struct dm_block *sblock;
-	size_t metadata_len, data_len;
 	struct thin_disk_superblock *disk_super;
 	sector_t bdev_size = i_size_read(pmd->bdev->bd_inode) >> SECTOR_SHIFT;
 
 	if (bdev_size > THIN_METADATA_MAX_SECTORS)
 		bdev_size = THIN_METADATA_MAX_SECTORS;
 
-	r = dm_sm_root_size(pmd->metadata_sm, &metadata_len);
-	if (r < 0)
-		return r;
-
-	r = dm_sm_root_size(pmd->data_sm, &data_len);
+	r = dm_sm_commit(pmd->data_sm);
 	if (r < 0)
 		return r;
 
-	r = dm_sm_commit(pmd->data_sm);
+	r = save_sm_roots(pmd);
 	if (r < 0)
 		return r;
 
@@ -471,15 +505,7 @@ static int __write_initial_superblock(struct dm_pool_metadata *pmd)
 	disk_super->trans_id = 0;
 	disk_super->held_root = 0;
 
-	r = dm_sm_copy_root(pmd->metadata_sm, &disk_super->metadata_space_map_root,
-			    metadata_len);
-	if (r < 0)
-		goto bad_locked;
-
-	r = dm_sm_copy_root(pmd->data_sm, &disk_super->data_space_map_root,
-			    data_len);
-	if (r < 0)
-		goto bad_locked;
+	copy_sm_roots(pmd, disk_super);
 
 	disk_super->data_mapping_root = cpu_to_le64(pmd->root);
 	disk_super->device_details_root = cpu_to_le64(pmd->details_root);
@@ -488,10 +514,6 @@ static int __write_initial_superblock(struct dm_pool_metadata *pmd)
 	disk_super->data_block_size = cpu_to_le32(pmd->data_block_size);
 
 	return dm_tm_commit(pmd->tm, sblock);
-
-bad_locked:
-	dm_bm_unlock(sblock);
-	return r;
 }
 
 static int __format_metadata(struct dm_pool_metadata *pmd)
@@ -591,6 +613,15 @@ static int __open_metadata(struct dm_pool_metadata *pmd)
 
 	disk_super = dm_block_data(sblock);
 
+	/* Verify the data block size hasn't changed */
+	if (le32_to_cpu(disk_super->data_block_size) != pmd->data_block_size) {
+		DMERR("changing the data block size (from %u to %llu) is not supported",
+		      le32_to_cpu(disk_super->data_block_size),
+		      (unsigned long long)pmd->data_block_size);
+		r = -EINVAL;
+		goto bad_unlock_sblock;
+	}
+
 	r = __check_incompat_features(disk_super, pmd);
 	if (r < 0)
 		goto bad_unlock_sblock;
@@ -769,6 +800,10 @@ static int __commit_transaction(struct dm_pool_metadata *pmd)
 	if (r < 0)
 		return r;
 
+	r = save_sm_roots(pmd);
+	if (r < 0)
+		return r;
+
 	r = superblock_lock(pmd, &sblock);
 	if (r)
 		return r;
@@ -780,21 +815,9 @@ static int __commit_transaction(struct dm_pool_metadata *pmd)
 	disk_super->trans_id = cpu_to_le64(pmd->trans_id);
 	disk_super->flags = cpu_to_le32(pmd->flags);
 
-	r = dm_sm_copy_root(pmd->metadata_sm, &disk_super->metadata_space_map_root,
-			    metadata_len);
-	if (r < 0)
-		goto out_locked;
-
-	r = dm_sm_copy_root(pmd->data_sm, &disk_super->data_space_map_root,
-			    data_len);
-	if (r < 0)
-		goto out_locked;
+	copy_sm_roots(pmd, disk_super);
 
 	return dm_tm_commit(pmd->tm, sblock);
-
-out_locked:
-	dm_bm_unlock(sblock);
-	return r;
 }
 
 struct dm_pool_metadata *dm_pool_metadata_open(struct block_device *bdev,
diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c
index e9587101..0396d7f 100644
--- a/drivers/md/dm-thin.c
+++ b/drivers/md/dm-thin.c
@@ -1332,9 +1332,9 @@ static void process_deferred_bios(struct pool *pool)
 		 */
 		if (ensure_next_mapping(pool)) {
 			spin_lock_irqsave(&pool->lock, flags);
+			bio_list_add(&pool->deferred_bios, bio);
 			bio_list_merge(&pool->deferred_bios, &bios);
 			spin_unlock_irqrestore(&pool->lock, flags);
-
 			break;
 		}
 
@@ -1504,6 +1504,14 @@ static int thin_bio_map(struct dm_target *ti, struct bio *bio)
 		return DM_MAPIO_SUBMITTED;
 	}
 
+	/*
+	 * We must hold the virtual cell before doing the lookup, otherwise
+	 * there's a race with discard.
+	 */
+	build_virtual_key(tc->td, block, &key);
+	if (dm_bio_detain(tc->pool->prison, &key, bio, &cell1, &cell_result))
+		return DM_MAPIO_SUBMITTED;
+
 	r = dm_thin_find_block(td, block, 0, &result);
 
 	/*
@@ -1527,13 +1535,10 @@ static int thin_bio_map(struct dm_target *ti, struct bio *bio)
 			 * shared flag will be set in their case.
 			 */
 			thin_defer_bio(tc, bio);
+			cell_defer_no_holder_no_free(tc, &cell1);
 			return DM_MAPIO_SUBMITTED;
 		}
 
-		build_virtual_key(tc->td, block, &key);
-		if (dm_bio_detain(tc->pool->prison, &key, bio, &cell1, &cell_result))
-			return DM_MAPIO_SUBMITTED;
-
 		build_data_key(tc->td, result.block, &key);
 		if (dm_bio_detain(tc->pool->prison, &key, bio, &cell2, &cell_result)) {
 			cell_defer_no_holder_no_free(tc, &cell1);
@@ -1554,6 +1559,7 @@ static int thin_bio_map(struct dm_target *ti, struct bio *bio)
 			 * of doing so.  Just error it.
 			 */
 			bio_io_error(bio);
+			cell_defer_no_holder_no_free(tc, &cell1);
 			return DM_MAPIO_SUBMITTED;
 		}
 		/* fall through */
@@ -1564,6 +1570,7 @@ static int thin_bio_map(struct dm_target *ti, struct bio *bio)
 		 * provide the hint to load the metadata into cache.
 		 */
 		thin_defer_bio(tc, bio);
+		cell_defer_no_holder_no_free(tc, &cell1);
 		return DM_MAPIO_SUBMITTED;
 
 	default:
@@ -1573,6 +1580,7 @@ static int thin_bio_map(struct dm_target *ti, struct bio *bio)
 		 * pool is switched to fail-io mode.
 		 */
 		bio_io_error(bio);
+		cell_defer_no_holder_no_free(tc, &cell1);
 		return DM_MAPIO_SUBMITTED;
 	}
 }
@@ -2695,7 +2703,8 @@ static void set_discard_limits(struct pool_c *pt, struct queue_limits *limits)
 	 */
 	if (pt->adjusted_pf.discard_passdown) {
 		data_limits = &bdev_get_queue(pt->data_dev->bdev)->limits;
-		limits->discard_granularity = data_limits->discard_granularity;
+		limits->discard_granularity = max(data_limits->discard_granularity,
+						  pool->sectors_per_block << SECTOR_SHIFT);
 	} else
 		limits->discard_granularity = pool->sectors_per_block << SECTOR_SHIFT;
 }
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index c147f06..a562d5a 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1813,14 +1813,14 @@ static void dm_request_fn(struct request_queue *q)
 		if (map_request(ti, clone, md))
 			goto requeued;
 
-		BUG_ON_NONRT(!irqs_disabled());
+		BUG_ON(!irqs_disabled());
 		spin_lock(q->queue_lock);
 	}
 
 	goto out;
 
 requeued:
-	BUG_ON_NONRT(!irqs_disabled());
+	BUG_ON(!irqs_disabled());
 	spin_lock(q->queue_lock);
 
 delay_and_out:
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 015bc45..bf030d4 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -7371,8 +7371,10 @@ void md_do_sync(struct md_thread *thread)
 	/* just incase thread restarts... */
 	if (test_bit(MD_RECOVERY_DONE, &mddev->recovery))
 		return;
-	if (mddev->ro) /* never try to sync a read-only array */
+	if (mddev->ro) {/* never try to sync a read-only array */
+		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 		return;
+	}
 
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
 		if (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)) {
@@ -7482,6 +7484,19 @@ void md_do_sync(struct md_thread *thread)
 			    rdev->recovery_offset < j)
 				j = rdev->recovery_offset;
 		rcu_read_unlock();
+
+		/* If there is a bitmap, we need to make sure all
+		 * writes that started before we added a spare
+		 * complete before we start doing a recovery.
+		 * Otherwise the write might complete and (via
+		 * bitmap_endwrite) set a bit in the bitmap after the
+		 * recovery has checked that bit and skipped that
+		 * region.
+		 */
+		if (mddev->bitmap) {
+			mddev->pers->quiesce(mddev, 1);
+			mddev->pers->quiesce(mddev, 0);
+		}
 	}
 
 	printk(KERN_INFO "md: %s of RAID array %s\n", desc, mdname(mddev));
@@ -7825,6 +7840,7 @@ void md_check_recovery(struct mddev *mddev)
 			/* There is no thread, but we need to call
 			 * ->spare_active and clear saved_raid_disk
 			 */
+			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 			md_reap_sync_thread(mddev);
 			clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 			goto unlock;
@@ -8520,7 +8536,8 @@ static int md_notify_reboot(struct notifier_block *this,
 		if (mddev_trylock(mddev)) {
 			if (mddev->pers)
 				__md_stop_writes(mddev);
-			mddev->safemode = 2;
+			if (mddev->persistent)
+				mddev->safemode = 2;
 			mddev_unlock(mddev);
 		}
 		need_delay = 1;
diff --git a/drivers/md/persistent-data/dm-block-manager.c b/drivers/md/persistent-data/dm-block-manager.c
index 064a3c2..30597f3 100644
--- a/drivers/md/persistent-data/dm-block-manager.c
+++ b/drivers/md/persistent-data/dm-block-manager.c
@@ -595,25 +595,14 @@ int dm_bm_unlock(struct dm_block *b)
 }
 EXPORT_SYMBOL_GPL(dm_bm_unlock);
 
-int dm_bm_flush_and_unlock(struct dm_block_manager *bm,
-			   struct dm_block *superblock)
+int dm_bm_flush(struct dm_block_manager *bm)
 {
-	int r;
-
 	if (bm->read_only)
 		return -EPERM;
 
-	r = dm_bufio_write_dirty_buffers(bm->bufio);
-	if (unlikely(r)) {
-		dm_bm_unlock(superblock);
-		return r;
-	}
-
-	dm_bm_unlock(superblock);
-
 	return dm_bufio_write_dirty_buffers(bm->bufio);
 }
-EXPORT_SYMBOL_GPL(dm_bm_flush_and_unlock);
+EXPORT_SYMBOL_GPL(dm_bm_flush);
 
 void dm_bm_prefetch(struct dm_block_manager *bm, dm_block_t b)
 {
diff --git a/drivers/md/persistent-data/dm-block-manager.h b/drivers/md/persistent-data/dm-block-manager.h
index 13cd58e..1b95dfc 100644
--- a/drivers/md/persistent-data/dm-block-manager.h
+++ b/drivers/md/persistent-data/dm-block-manager.h
@@ -105,8 +105,7 @@ int dm_bm_unlock(struct dm_block *b);
  *
  * This method always blocks.
  */
-int dm_bm_flush_and_unlock(struct dm_block_manager *bm,
-			   struct dm_block *superblock);
+int dm_bm_flush(struct dm_block_manager *bm);
 
 /*
  * Request data is prefetched into the cache.
diff --git a/drivers/md/persistent-data/dm-btree-internal.h b/drivers/md/persistent-data/dm-btree-internal.h
index 37d367b..bf2b80d 100644
--- a/drivers/md/persistent-data/dm-btree-internal.h
+++ b/drivers/md/persistent-data/dm-btree-internal.h
@@ -42,6 +42,12 @@ struct btree_node {
 } __packed;
 
 
+/*
+ * Locks a block using the btree node validator.
+ */
+int bn_read_lock(struct dm_btree_info *info, dm_block_t b,
+		 struct dm_block **result);
+
 void inc_children(struct dm_transaction_manager *tm, struct btree_node *n,
 		  struct dm_btree_value_type *vt);
 
diff --git a/drivers/md/persistent-data/dm-btree-spine.c b/drivers/md/persistent-data/dm-btree-spine.c
index cf9fd67..1b5e13e 100644
--- a/drivers/md/persistent-data/dm-btree-spine.c
+++ b/drivers/md/persistent-data/dm-btree-spine.c
@@ -92,7 +92,7 @@ struct dm_block_validator btree_node_validator = {
 
 /*----------------------------------------------------------------*/
 
-static int bn_read_lock(struct dm_btree_info *info, dm_block_t b,
+int bn_read_lock(struct dm_btree_info *info, dm_block_t b,
 		 struct dm_block **result)
 {
 	return dm_tm_read_lock(info->tm, b, &btree_node_validator, result);
diff --git a/drivers/md/persistent-data/dm-btree.c b/drivers/md/persistent-data/dm-btree.c
index 468e371..9701d29 100644
--- a/drivers/md/persistent-data/dm-btree.c
+++ b/drivers/md/persistent-data/dm-btree.c
@@ -828,22 +828,26 @@ EXPORT_SYMBOL_GPL(dm_btree_find_highest_key);
  * FIXME: We shouldn't use a recursive algorithm when we have limited stack
  * space.  Also this only works for single level trees.
  */
-static int walk_node(struct ro_spine *s, dm_block_t block,
+static int walk_node(struct dm_btree_info *info, dm_block_t block,
 		     int (*fn)(void *context, uint64_t *keys, void *leaf),
 		     void *context)
 {
 	int r;
 	unsigned i, nr;
+	struct dm_block *node;
 	struct btree_node *n;
 	uint64_t keys;
 
-	r = ro_step(s, block);
-	n = ro_node(s);
+	r = bn_read_lock(info, block, &node);
+	if (r)
+		return r;
+
+	n = dm_block_data(node);
 
 	nr = le32_to_cpu(n->header.nr_entries);
 	for (i = 0; i < nr; i++) {
 		if (le32_to_cpu(n->header.flags) & INTERNAL_NODE) {
-			r = walk_node(s, value64(n, i), fn, context);
+			r = walk_node(info, value64(n, i), fn, context);
 			if (r)
 				goto out;
 		} else {
@@ -855,7 +859,7 @@ static int walk_node(struct ro_spine *s, dm_block_t block,
 	}
 
 out:
-	ro_pop(s);
+	dm_tm_unlock(info->tm, node);
 	return r;
 }
 
@@ -863,15 +867,7 @@ int dm_btree_walk(struct dm_btree_info *info, dm_block_t root,
 		  int (*fn)(void *context, uint64_t *keys, void *leaf),
 		  void *context)
 {
-	int r;
-	struct ro_spine spine;
-
 	BUG_ON(info->levels > 1);
-
-	init_ro_spine(&spine, info);
-	r = walk_node(&spine, root, fn, context);
-	exit_ro_spine(&spine);
-
-	return r;
+	return walk_node(info, root, fn, context);
 }
 EXPORT_SYMBOL_GPL(dm_btree_walk);
diff --git a/drivers/md/persistent-data/dm-space-map-metadata.c b/drivers/md/persistent-data/dm-space-map-metadata.c
index 579b582..d9a5aa5 100644
--- a/drivers/md/persistent-data/dm-space-map-metadata.c
+++ b/drivers/md/persistent-data/dm-space-map-metadata.c
@@ -564,7 +564,9 @@ static int sm_bootstrap_get_nr_blocks(struct dm_space_map *sm, dm_block_t *count
 {
 	struct sm_metadata *smm = container_of(sm, struct sm_metadata, sm);
 
-	return smm->ll.nr_blocks;
+	*count = smm->ll.nr_blocks;
+
+	return 0;
 }
 
 static int sm_bootstrap_get_nr_free(struct dm_space_map *sm, dm_block_t *count)
diff --git a/drivers/md/persistent-data/dm-transaction-manager.c b/drivers/md/persistent-data/dm-transaction-manager.c
index 81da1a2..3bc30a0 100644
--- a/drivers/md/persistent-data/dm-transaction-manager.c
+++ b/drivers/md/persistent-data/dm-transaction-manager.c
@@ -154,7 +154,7 @@ int dm_tm_pre_commit(struct dm_transaction_manager *tm)
 	if (r < 0)
 		return r;
 
-	return 0;
+	return dm_bm_flush(tm->bm);
 }
 EXPORT_SYMBOL_GPL(dm_tm_pre_commit);
 
@@ -164,8 +164,9 @@ int dm_tm_commit(struct dm_transaction_manager *tm, struct dm_block *root)
 		return -EWOULDBLOCK;
 
 	wipe_shadow_table(tm);
+	dm_bm_unlock(root);
 
-	return dm_bm_flush_and_unlock(tm->bm, root);
+	return dm_bm_flush(tm->bm);
 }
 EXPORT_SYMBOL_GPL(dm_tm_commit);
 
diff --git a/drivers/md/persistent-data/dm-transaction-manager.h b/drivers/md/persistent-data/dm-transaction-manager.h
index b5b1390..2772ed2 100644
--- a/drivers/md/persistent-data/dm-transaction-manager.h
+++ b/drivers/md/persistent-data/dm-transaction-manager.h
@@ -38,18 +38,17 @@ struct dm_transaction_manager *dm_tm_create_non_blocking_clone(struct dm_transac
 /*
  * We use a 2-phase commit here.
  *
- * i) In the first phase the block manager is told to start flushing, and
- * the changes to the space map are written to disk.  You should interrogate
- * your particular space map to get detail of its root node etc. to be
- * included in your superblock.
+ * i) Make all changes for the transaction *except* for the superblock.
+ * Then call dm_tm_pre_commit() to flush them to disk.
  *
- * ii) @root will be committed last.  You shouldn't use more than the
- * first 512 bytes of @root if you wish the transaction to survive a power
- * failure.  You *must* have a write lock held on @root for both stage (i)
- * and (ii).  The commit will drop the write lock.
+ * ii) Lock your superblock.  Update.  Then call dm_tm_commit() which will
+ * unlock the superblock and flush it.  No other blocks should be updated
+ * during this period.  Care should be taken to never unlock a partially
+ * updated superblock; perform any operations that could fail *before* you
+ * take the superblock lock.
  */
 int dm_tm_pre_commit(struct dm_transaction_manager *tm);
-int dm_tm_commit(struct dm_transaction_manager *tm, struct dm_block *root);
+int dm_tm_commit(struct dm_transaction_manager *tm, struct dm_block *superblock);
 
 /*
  * These methods are the only way to get hold of a writeable block.
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 6edc2db..6564eeb 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -94,6 +94,7 @@ static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
 	struct pool_info *pi = data;
 	struct r1bio *r1_bio;
 	struct bio *bio;
+	int need_pages;
 	int i, j;
 
 	r1_bio = r1bio_pool_alloc(gfp_flags, pi);
@@ -116,15 +117,15 @@ static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
 	 * RESYNC_PAGES for each bio.
 	 */
 	if (test_bit(MD_RECOVERY_REQUESTED, &pi->mddev->recovery))
-		j = pi->raid_disks;
+		need_pages = pi->raid_disks;
 	else
-		j = 1;
-	while(j--) {
+		need_pages = 1;
+	for (j = 0; j < need_pages; j++) {
 		bio = r1_bio->bios[j];
 		bio->bi_vcnt = RESYNC_PAGES;
 
 		if (bio_alloc_pages(bio, gfp_flags))
-			goto out_free_bio;
+			goto out_free_pages;
 	}
 	/* If not user-requests, copy the page pointers to all bios */
 	if (!test_bit(MD_RECOVERY_REQUESTED, &pi->mddev->recovery)) {
@@ -138,6 +139,14 @@ static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
 
 	return r1_bio;
 
+out_free_pages:
+	while (--j >= 0) {
+		struct bio_vec *bv;
+
+		bio_for_each_segment_all(bv, r1_bio->bios[j], i)
+			__free_page(bv->bv_page);
+	}
+
 out_free_bio:
 	while (++j < pi->raid_disks)
 		bio_put(r1_bio->bios[j]);
@@ -1397,12 +1406,12 @@ static void error(struct mddev *mddev, struct md_rdev *rdev)
 		mddev->degraded++;
 		set_bit(Faulty, &rdev->flags);
 		spin_unlock_irqrestore(&conf->device_lock, flags);
-		/*
-		 * if recovery is running, make sure it aborts.
-		 */
-		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	} else
 		set_bit(Faulty, &rdev->flags);
+	/*
+	 * if recovery is running, make sure it aborts.
+	 */
+	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
 	printk(KERN_ALERT
 	       "md/raid1:%s: Disk failure on %s, disabling device.\n"
@@ -2043,7 +2052,7 @@ static void fix_read_error(struct r1conf *conf, int read_disk,
 			d--;
 			rdev = conf->mirrors[d].rdev;
 			if (rdev &&
-			    test_bit(In_sync, &rdev->flags))
+			    !test_bit(Faulty, &rdev->flags))
 				r1_sync_page_io(rdev, sect, s,
 						conf->tmppage, WRITE);
 		}
@@ -2055,7 +2064,7 @@ static void fix_read_error(struct r1conf *conf, int read_disk,
 			d--;
 			rdev = conf->mirrors[d].rdev;
 			if (rdev &&
-			    test_bit(In_sync, &rdev->flags)) {
+			    !test_bit(Faulty, &rdev->flags)) {
 				if (r1_sync_page_io(rdev, sect, s,
 						    conf->tmppage, READ)) {
 					atomic_add(s, &rdev->corrected_errors);
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 308575d..9ccb107 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1698,13 +1698,12 @@ static void error(struct mddev *mddev, struct md_rdev *rdev)
 		spin_unlock_irqrestore(&conf->device_lock, flags);
 		return;
 	}
-	if (test_and_clear_bit(In_sync, &rdev->flags)) {
+	if (test_and_clear_bit(In_sync, &rdev->flags))
 		mddev->degraded++;
-			/*
-		 * if recovery is running, make sure it aborts.
-		 */
-		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
-	}
+	/*
+	 * If recovery is running, make sure it aborts.
+	 */
+	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	set_bit(Blocked, &rdev->flags);
 	set_bit(Faulty, &rdev->flags);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
@@ -2970,6 +2969,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr,
 		 */
 		if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) {
 			end_reshape(conf);
+			close_sync(conf);
 			return 0;
 		}
 
@@ -4420,7 +4420,7 @@ read_more:
 	read_bio->bi_private = r10_bio;
 	read_bio->bi_end_io = end_sync_read;
 	read_bio->bi_rw = READ;
-	read_bio->bi_flags &= ~(BIO_POOL_MASK - 1);
+	read_bio->bi_flags &= (~0UL << BIO_RESET_BITS);
 	read_bio->bi_flags |= 1 << BIO_UPTODATE;
 	read_bio->bi_vcnt = 0;
 	read_bio->bi_size = 0;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index ab00c1e..7b54c3b 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -64,6 +64,10 @@
 #define cpu_to_group(cpu) cpu_to_node(cpu)
 #define ANY_GROUP NUMA_NO_NODE
 
+static bool devices_handle_discard_safely = false;
+module_param(devices_handle_discard_safely, bool, 0644);
+MODULE_PARM_DESC(devices_handle_discard_safely,
+		 "Set to Y if all devices in each array reliably return zeroes on reads from discarded regions");
 static struct workqueue_struct *raid5_wq;
 /*
  * Stripe cache
@@ -1541,9 +1545,8 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
 	struct raid5_percpu *percpu;
 	unsigned long cpu;
 
-	cpu = get_cpu_light();
+	cpu = get_cpu();
 	percpu = per_cpu_ptr(conf->percpu, cpu);
-	spin_lock(&percpu->lock);
 	if (test_bit(STRIPE_OP_BIOFILL, &ops_request)) {
 		ops_run_biofill(sh);
 		overlap_clear++;
@@ -1595,8 +1598,7 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
 			if (test_and_clear_bit(R5_Overlap, &dev->flags))
 				wake_up(&sh->raid_conf->wait_for_overlap);
 		}
-	spin_unlock(&percpu->lock);
-	put_cpu_light();
+	put_cpu();
 }
 
 static int grow_one_stripe(struct r5conf *conf)
@@ -2787,7 +2789,8 @@ static int fetch_block(struct stripe_head *sh, struct stripe_head_state *s,
 	     (s->failed >= 2 && fdev[1]->toread) ||
 	     (sh->raid_conf->level <= 5 && s->failed && fdev[0]->towrite &&
 	      !test_bit(R5_OVERWRITE, &fdev[0]->flags)) ||
-	     (sh->raid_conf->level == 6 && s->failed && s->to_write))) {
+	     ((sh->raid_conf->level == 6 || sh->sector >= sh->raid_conf->mddev->recovery_cp)
+	      && s->failed && s->to_write))) {
 		/* we would like to get this block, possibly by computing it,
 		 * otherwise read it if the backing disk is insync
 		 */
@@ -3674,6 +3677,8 @@ static void handle_stripe(struct stripe_head *sh)
 				set_bit(R5_Wantwrite, &dev->flags);
 				if (prexor)
 					continue;
+				if (s.failed > 1)
+					continue;
 				if (!test_bit(R5_Insync, &dev->flags) ||
 				    ((i == sh->pd_idx || i == sh->qd_idx)  &&
 				     s.failed == 0))
@@ -5458,7 +5463,6 @@ static int raid5_alloc_percpu(struct r5conf *conf)
 			       __func__, cpu);
 			break;
 		}
-		spin_lock_init(&per_cpu_ptr(conf->percpu, cpu)->lock);
 	}
 	put_online_cpus();
 
@@ -5939,7 +5943,7 @@ static int run(struct mddev *mddev)
 		mddev->queue->limits.discard_granularity = stripe;
 		/*
 		 * unaligned part of discard request will be ignored, so can't
-		 * guarantee discard_zerors_data
+		 * guarantee discard_zeroes_data
 		 */
 		mddev->queue->limits.discard_zeroes_data = 0;
 
@@ -5964,6 +5968,18 @@ static int run(struct mddev *mddev)
 			    !bdev_get_queue(rdev->bdev)->
 						limits.discard_zeroes_data)
 				discard_supported = false;
+			/* Unfortunately, discard_zeroes_data is not currently
+			 * a guarantee - just a hint.  So we only allow DISCARD
+			 * if the sysadmin has confirmed that only safe devices
+			 * are in use by setting a module parameter.
+			 */
+			if (!devices_handle_discard_safely) {
+				if (discard_supported) {
+					pr_info("md/raid456: discard support disabled due to uncertainty.\n");
+					pr_info("Set raid456.devices_handle_discard_safely=Y to override.\n");
+				}
+				discard_supported = false;
+			}
 		}
 
 		if (discard_supported &&
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index 0a5e1e1..2113ffa 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -444,7 +444,6 @@ struct r5conf {
 	int			recovery_disabled;
 	/* per cpu variables */
 	struct raid5_percpu {
-		spinlock_t	lock;	     /* Protection for -RT */
 		struct page	*spare_page; /* Used when checking P/Q in raid6 */
 		void		*scribble;   /* space for constructing buffer
 					      * lists and performing address