Merge tag 'drivers-5.10-2020-10-12' of git://git.kernel.dk/linux-block

Pull block driver updates from Jens Axboe:
 "Here are the driver updates for 5.10.

  A few SCSI updates in here too, in coordination with Martin as they
  depend on core block changes for the shared tag bitmap.

  This contains:

   - NVMe pull requests via Christoph:
      - fix keep alive timer modification (Amit Engel)
      - order the PCI ID list more sensibly (Andy Shevchenko)
      - cleanup the open by controller helper (Chaitanya Kulkarni)
      - use an xarray for the CSE log lookup (Chaitanya Kulkarni)
      - support ZNS in nvmet passthrough mode (Chaitanya Kulkarni)
      - fix nvme_ns_report_zones (Christoph Hellwig)
      - add a sanity check to nvmet-fc (James Smart)
      - fix interrupt allocation when too many polled queues are
        specified (Jeffle Xu)
      - small nvmet-tcp optimization (Mark Wunderlich)
      - fix a controller refcount leak on init failure (Chaitanya
        Kulkarni)
      - misc cleanups (Chaitanya Kulkarni)
      - major refactoring of the scanning code (Christoph Hellwig)

   - MD updates via Song:
      - Bug fixes in bitmap code, from Zhao Heming
      - Fix a work queue check, from Guoqing Jiang
      - Fix raid5 oops with reshape, from Song Liu
      - Clean up unused code, from Jason Yan
      - Discard improvements, from Xiao Ni
      - raid5/6 page offset support, from Yufen Yu

   - Shared tag bitmap for SCSI/hisi_sas/null_blk (John, Kashyap,
     Hannes)

   - null_blk open/active zone limit support (Niklas)

   - Set of bcache updates (Coly, Dongsheng, Qinglang)"

* tag 'drivers-5.10-2020-10-12' of git://git.kernel.dk/linux-block: (78 commits)
  md/raid5: fix oops during stripe resizing
  md/bitmap: fix memory leak of temporary bitmap
  md: fix the checking of wrong work queue
  md/bitmap: md_bitmap_get_counter returns wrong blocks
  md/bitmap: md_bitmap_read_sb uses wrong bitmap blocks
  md/raid0: remove unused function is_io_in_chunk_boundary()
  nvme-core: remove extra condition for vwc
  nvme-core: remove extra variable
  nvme: remove nvme_identify_ns_list
  nvme: refactor nvme_validate_ns
  nvme: move nvme_validate_ns
  nvme: query namespace identifiers before adding the namespace
  nvme: revalidate zone bitmaps in nvme_update_ns_info
  nvme: remove nvme_update_formats
  nvme: update the known admin effects
  nvme: set the queue limits in nvme_update_ns_info
  nvme: remove the 0 lba_shift check in nvme_update_ns_info
  nvme: clean up the check for too large logic block sizes
  nvme: freeze the queue over ->lba_shift updates
  nvme: factor out a nvme_configure_metadata helper
  ...

24 files changed, 1032 insertions, 620 deletions

diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c
index 52035a78d836..8c371d5eef8e 100644
--- a/drivers/md/bcache/alloc.c
+++ b/drivers/md/bcache/alloc.c
@@ -49,7 +49,7 @@
  *
  * bch_bucket_alloc() allocates a single bucket from a specific cache.
  *
- * bch_bucket_alloc_set() allocates one or more buckets from different caches
+ * bch_bucket_alloc_set() allocates one  bucket from different caches
  * out of a cache set.
  *
  * free_some_buckets() drives all the processes described above. It's called
@@ -87,8 +87,7 @@ void bch_rescale_priorities(struct cache_set *c, int sectors)
 {
 	struct cache *ca;
 	struct bucket *b;
-	unsigned long next = c->nbuckets * c->sb.bucket_size / 1024;
-	unsigned int i;
+	unsigned long next = c->nbuckets * c->cache->sb.bucket_size / 1024;
 	int r;
 
 	atomic_sub(sectors, &c->rescale);
@@ -104,14 +103,14 @@ void bch_rescale_priorities(struct cache_set *c, int sectors)
 
 	c->min_prio = USHRT_MAX;
 
-	for_each_cache(ca, c, i)
-		for_each_bucket(b, ca)
-			if (b->prio &&
-			    b->prio != BTREE_PRIO &&
-			    !atomic_read(&b->pin)) {
-				b->prio--;
-				c->min_prio = min(c->min_prio, b->prio);
-			}
+	ca = c->cache;
+	for_each_bucket(b, ca)
+		if (b->prio &&
+		    b->prio != BTREE_PRIO &&
+		    !atomic_read(&b->pin)) {
+			b->prio--;
+			c->min_prio = min(c->min_prio, b->prio);
+		}
 
 	mutex_unlock(&c->bucket_lock);
 }
@@ -362,7 +361,7 @@ retry_invalidate:
 		 * new stuff to them:
 		 */
 		allocator_wait(ca, !atomic_read(&ca->set->prio_blocked));
-		if (CACHE_SYNC(&ca->set->sb)) {
+		if (CACHE_SYNC(&ca->sb)) {
 			/*
 			 * This could deadlock if an allocation with a btree
 			 * node locked ever blocked - having the btree node
@@ -488,34 +487,29 @@ void bch_bucket_free(struct cache_set *c, struct bkey *k)
 }
 
 int __bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			   struct bkey *k, int n, bool wait)
+			   struct bkey *k, bool wait)
 {
-	int i;
+	struct cache *ca;
+	long b;
 
 	/* No allocation if CACHE_SET_IO_DISABLE bit is set */
 	if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags)))
 		return -1;
 
 	lockdep_assert_held(&c->bucket_lock);
-	BUG_ON(!n || n > c->caches_loaded || n > MAX_CACHES_PER_SET);
 
 	bkey_init(k);
 
-	/* sort by free space/prio of oldest data in caches */
-
-	for (i = 0; i < n; i++) {
-		struct cache *ca = c->cache_by_alloc[i];
-		long b = bch_bucket_alloc(ca, reserve, wait);
+	ca = c->cache;
+	b = bch_bucket_alloc(ca, reserve, wait);
+	if (b == -1)
+		goto err;
 
-		if (b == -1)
-			goto err;
+	k->ptr[0] = MAKE_PTR(ca->buckets[b].gen,
+			     bucket_to_sector(c, b),
+			     ca->sb.nr_this_dev);
 
-		k->ptr[i] = MAKE_PTR(ca->buckets[b].gen,
-				bucket_to_sector(c, b),
-				ca->sb.nr_this_dev);
-
-		SET_KEY_PTRS(k, i + 1);
-	}
+	SET_KEY_PTRS(k, 1);
 
 	return 0;
 err:
@@ -525,12 +519,12 @@ err:
 }
 
 int bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			 struct bkey *k, int n, bool wait)
+			 struct bkey *k, bool wait)
 {
 	int ret;
 
 	mutex_lock(&c->bucket_lock);
-	ret = __bch_bucket_alloc_set(c, reserve, k, n, wait);
+	ret = __bch_bucket_alloc_set(c, reserve, k, wait);
 	mutex_unlock(&c->bucket_lock);
 	return ret;
 }
@@ -589,7 +583,7 @@ static struct open_bucket *pick_data_bucket(struct cache_set *c,
 					   struct open_bucket, list);
 found:
 	if (!ret->sectors_free && KEY_PTRS(alloc)) {
-		ret->sectors_free = c->sb.bucket_size;
+		ret->sectors_free = c->cache->sb.bucket_size;
 		bkey_copy(&ret->key, alloc);
 		bkey_init(alloc);
 	}
@@ -638,7 +632,7 @@ bool bch_alloc_sectors(struct cache_set *c,
 
 		spin_unlock(&c->data_bucket_lock);
 
-		if (bch_bucket_alloc_set(c, watermark, &alloc.key, 1, wait))
+		if (bch_bucket_alloc_set(c, watermark, &alloc.key, wait))
 			return false;
 
 		spin_lock(&c->data_bucket_lock);
@@ -683,7 +677,7 @@ bool bch_alloc_sectors(struct cache_set *c,
 				&PTR_CACHE(c, &b->key, i)->sectors_written);
 	}
 
-	if (b->sectors_free < c->sb.block_size)
+	if (b->sectors_free < c->cache->sb.block_size)
 		b->sectors_free = 0;
 
 	/*
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index 4fd03d2496d8..1d57f48307e6 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -517,11 +517,7 @@ struct cache_set {
 	atomic_t		idle_counter;
 	atomic_t		at_max_writeback_rate;
 
-	struct cache_sb		sb;
-
-	struct cache		*cache[MAX_CACHES_PER_SET];
-	struct cache		*cache_by_alloc[MAX_CACHES_PER_SET];
-	int			caches_loaded;
+	struct cache		*cache;
 
 	struct bcache_device	**devices;
 	unsigned int		devices_max_used;
@@ -670,6 +666,7 @@ struct cache_set {
 	struct mutex		verify_lock;
 #endif
 
+	uint8_t			set_uuid[16];
 	unsigned int		nr_uuids;
 	struct uuid_entry	*uuids;
 	BKEY_PADDED(uuid_bucket);
@@ -758,9 +755,8 @@ struct bbio {
 #define btree_default_blocks(c)						\
 	((unsigned int) ((PAGE_SECTORS * (c)->btree_pages) >> (c)->block_bits))
 
-#define bucket_pages(c)		((c)->sb.bucket_size / PAGE_SECTORS)
-#define bucket_bytes(c)		((c)->sb.bucket_size << 9)
-#define block_bytes(c)		((c)->sb.block_size << 9)
+#define bucket_bytes(ca)	((ca)->sb.bucket_size << 9)
+#define block_bytes(ca)		((ca)->sb.block_size << 9)
 
 static inline unsigned int meta_bucket_pages(struct cache_sb *sb)
 {
@@ -801,14 +797,14 @@ static inline sector_t bucket_to_sector(struct cache_set *c, size_t b)
 
 static inline sector_t bucket_remainder(struct cache_set *c, sector_t s)
 {
-	return s & (c->sb.bucket_size - 1);
+	return s & (c->cache->sb.bucket_size - 1);
 }
 
 static inline struct cache *PTR_CACHE(struct cache_set *c,
 				      const struct bkey *k,
 				      unsigned int ptr)
 {
-	return c->cache[PTR_DEV(k, ptr)];
+	return c->cache;
 }
 
 static inline size_t PTR_BUCKET_NR(struct cache_set *c,
@@ -889,9 +885,6 @@ do {									\
 
 /* Looping macros */
 
-#define for_each_cache(ca, cs, iter)					\
-	for (iter = 0; ca = cs->cache[iter], iter < (cs)->sb.nr_in_set; iter++)
-
 #define for_each_bucket(b, ca)						\
 	for (b = (ca)->buckets + (ca)->sb.first_bucket;			\
 	     b < (ca)->buckets + (ca)->sb.nbuckets; b++)
@@ -933,11 +926,9 @@ static inline uint8_t bucket_gc_gen(struct bucket *b)
 
 static inline void wake_up_allocators(struct cache_set *c)
 {
-	struct cache *ca;
-	unsigned int i;
+	struct cache *ca = c->cache;
 
-	for_each_cache(ca, c, i)
-		wake_up_process(ca->alloc_thread);
+	wake_up_process(ca->alloc_thread);
 }
 
 static inline void closure_bio_submit(struct cache_set *c,
@@ -994,9 +985,9 @@ void bch_bucket_free(struct cache_set *c, struct bkey *k);
 
 long bch_bucket_alloc(struct cache *ca, unsigned int reserve, bool wait);
 int __bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			   struct bkey *k, int n, bool wait);
+			   struct bkey *k, bool wait);
 int bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			 struct bkey *k, int n, bool wait);
+			 struct bkey *k, bool wait);
 bool bch_alloc_sectors(struct cache_set *c, struct bkey *k,
 		       unsigned int sectors, unsigned int write_point,
 		       unsigned int write_prio, bool wait);
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index 3d8bd0692af3..910df242c83d 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -104,7 +104,7 @@
 
 static inline struct bset *write_block(struct btree *b)
 {
-	return ((void *) btree_bset_first(b)) + b->written * block_bytes(b->c);
+	return ((void *) btree_bset_first(b)) + b->written * block_bytes(b->c->cache);
 }
 
 static void bch_btree_init_next(struct btree *b)
@@ -117,7 +117,7 @@ static void bch_btree_init_next(struct btree *b)
 
 	if (b->written < btree_blocks(b))
 		bch_bset_init_next(&b->keys, write_block(b),
-				   bset_magic(&b->c->sb));
+				   bset_magic(&b->c->cache->sb));
 
 }
 
@@ -155,7 +155,7 @@ void bch_btree_node_read_done(struct btree *b)
 	 * See the comment arount cache_set->fill_iter.
 	 */
 	iter = mempool_alloc(&b->c->fill_iter, GFP_NOIO);
-	iter->size = b->c->sb.bucket_size / b->c->sb.block_size;
+	iter->size = b->c->cache->sb.bucket_size / b->c->cache->sb.block_size;
 	iter->used = 0;
 
 #ifdef CONFIG_BCACHE_DEBUG
@@ -173,12 +173,12 @@ void bch_btree_node_read_done(struct btree *b)
 			goto err;
 
 		err = "bad btree header";
-		if (b->written + set_blocks(i, block_bytes(b->c)) >
+		if (b->written + set_blocks(i, block_bytes(b->c->cache)) >
 		    btree_blocks(b))
 			goto err;
 
 		err = "bad magic";
-		if (i->magic != bset_magic(&b->c->sb))
+		if (i->magic != bset_magic(&b->c->cache->sb))
 			goto err;
 
 		err = "bad checksum";
@@ -199,13 +199,13 @@ void bch_btree_node_read_done(struct btree *b)
 
 		bch_btree_iter_push(iter, i->start, bset_bkey_last(i));
 
-		b->written += set_blocks(i, block_bytes(b->c));
+		b->written += set_blocks(i, block_bytes(b->c->cache));
 	}
 
 	err = "corrupted btree";
 	for (i = write_block(b);
 	     bset_sector_offset(&b->keys, i) < KEY_SIZE(&b->key);
-	     i = ((void *) i) + block_bytes(b->c))
+	     i = ((void *) i) + block_bytes(b->c->cache))
 		if (i->seq == b->keys.set[0].data->seq)
 			goto err;
 
@@ -219,7 +219,7 @@ void bch_btree_node_read_done(struct btree *b)
 
 	if (b->written < btree_blocks(b))
 		bch_bset_init_next(&b->keys, write_block(b),
-				   bset_magic(&b->c->sb));
+				   bset_magic(&b->c->cache->sb));
 out:
 	mempool_free(iter, &b->c->fill_iter);
 	return;
@@ -347,7 +347,7 @@ static void do_btree_node_write(struct btree *b)
 
 	b->bio->bi_end_io	= btree_node_write_endio;
 	b->bio->bi_private	= cl;
-	b->bio->bi_iter.bi_size	= roundup(set_bytes(i), block_bytes(b->c));
+	b->bio->bi_iter.bi_size	= roundup(set_bytes(i), block_bytes(b->c->cache));
 	b->bio->bi_opf		= REQ_OP_WRITE | REQ_META | REQ_FUA;
 	bch_bio_map(b->bio, i);
 
@@ -423,10 +423,10 @@ void __bch_btree_node_write(struct btree *b, struct closure *parent)
 
 	do_btree_node_write(b);
 
-	atomic_long_add(set_blocks(i, block_bytes(b->c)) * b->c->sb.block_size,
+	atomic_long_add(set_blocks(i, block_bytes(b->c->cache)) * b->c->cache->sb.block_size,
 			&PTR_CACHE(b->c, &b->key, 0)->btree_sectors_written);
 
-	b->written += set_blocks(i, block_bytes(b->c));
+	b->written += set_blocks(i, block_bytes(b->c->cache));
 }
 
 void bch_btree_node_write(struct btree *b, struct closure *parent)
@@ -514,7 +514,7 @@ static void bch_btree_leaf_dirty(struct btree *b, atomic_t *journal_ref)
  * mca -> memory cache
  */
 
-#define mca_reserve(c)	(((c->root && c->root->level)		\
+#define mca_reserve(c)	(((!IS_ERR_OR_NULL(c->root) && c->root->level) \
 			  ? c->root->level : 1) * 8 + 16)
 #define mca_can_free(c)						\
 	max_t(int, 0, c->btree_cache_used - mca_reserve(c))
@@ -738,7 +738,7 @@ void bch_btree_cache_free(struct cache_set *c)
 	if (c->verify_data)
 		list_move(&c->verify_data->list, &c->btree_cache);
 
-	free_pages((unsigned long) c->verify_ondisk, ilog2(meta_bucket_pages(&c->sb)));
+	free_pages((unsigned long) c->verify_ondisk, ilog2(meta_bucket_pages(&c->cache->sb)));
 #endif
 
 	list_splice(&c->btree_cache_freeable,
@@ -785,7 +785,8 @@ int bch_btree_cache_alloc(struct cache_set *c)
 	mutex_init(&c->verify_lock);
 
 	c->verify_ondisk = (void *)
-		__get_free_pages(GFP_KERNEL|__GFP_COMP, ilog2(meta_bucket_pages(&c->sb)));
+		__get_free_pages(GFP_KERNEL|__GFP_COMP,
+				 ilog2(meta_bucket_pages(&c->cache->sb)));
 	if (!c->verify_ondisk) {
 		/*
 		 * Don't worry about the mca_rereserve buckets
@@ -1091,7 +1092,7 @@ struct btree *__bch_btree_node_alloc(struct cache_set *c, struct btree_op *op,
 
 	mutex_lock(&c->bucket_lock);
 retry:
-	if (__bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, 1, wait))
+	if (__bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, wait))
 		goto err;
 
 	bkey_put(c, &k.key);
@@ -1108,7 +1109,7 @@ retry:
 	}
 
 	b->parent = parent;
-	bch_bset_init_next(&b->keys, b->keys.set->data, bset_magic(&b->c->sb));
+	bch_bset_init_next(&b->keys, b->keys.set->data, bset_magic(&b->c->cache->sb));
 
 	mutex_unlock(&c->bucket_lock);
 
@@ -1167,19 +1168,18 @@ static void make_btree_freeing_key(struct btree *b, struct bkey *k)
 static int btree_check_reserve(struct btree *b, struct btree_op *op)
 {
 	struct cache_set *c = b->c;
-	struct cache *ca;
-	unsigned int i, reserve = (c->root->level - b->level) * 2 + 1;
+	struct cache *ca = c->cache;
+	unsigned int reserve = (c->root->level - b->level) * 2 + 1;
 
 	mutex_lock(&c->bucket_lock);
 
-	for_each_cache(ca, c, i)
-		if (fifo_used(&ca->free[RESERVE_BTREE]) < reserve) {
-			if (op)
-				prepare_to_wait(&c->btree_cache_wait, &op->wait,
-						TASK_UNINTERRUPTIBLE);
-			mutex_unlock(&c->bucket_lock);
-			return -EINTR;
-		}
+	if (fifo_used(&ca->free[RESERVE_BTREE]) < reserve) {
+		if (op)
+			prepare_to_wait(&c->btree_cache_wait, &op->wait,
+					TASK_UNINTERRUPTIBLE);
+		mutex_unlock(&c->bucket_lock);
+		return -EINTR;
+	}
 
 	mutex_unlock(&c->bucket_lock);
 
@@ -1345,7 +1345,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 
 	if (nodes < 2 ||
 	    __set_blocks(b->keys.set[0].data, keys,
-			 block_bytes(b->c)) > blocks * (nodes - 1))
+			 block_bytes(b->c->cache)) > blocks * (nodes - 1))
 		return 0;
 
 	for (i = 0; i < nodes; i++) {
@@ -1379,7 +1379,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 			     k = bkey_next(k)) {
 				if (__set_blocks(n1, n1->keys + keys +
 						 bkey_u64s(k),
-						 block_bytes(b->c)) > blocks)
+						 block_bytes(b->c->cache)) > blocks)
 					break;
 
 				last = k;
@@ -1395,7 +1395,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 			 * though)
 			 */
 			if (__set_blocks(n1, n1->keys + n2->keys,
-					 block_bytes(b->c)) >
+					 block_bytes(b->c->cache)) >
 			    btree_blocks(new_nodes[i]))
 				goto out_unlock_nocoalesce;
 
@@ -1404,7 +1404,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 			last = &r->b->key;
 		}
 
-		BUG_ON(__set_blocks(n1, n1->keys + keys, block_bytes(b->c)) >
+		BUG_ON(__set_blocks(n1, n1->keys + keys, block_bytes(b->c->cache)) >
 		       btree_blocks(new_nodes[i]));
 
 		if (last)
@@ -1695,7 +1695,6 @@ static void btree_gc_start(struct cache_set *c)
 {
 	struct cache *ca;
 	struct bucket *b;
-	unsigned int i;
 
 	if (!c->gc_mark_valid)
 		return;
@@ -1705,14 +1704,14 @@ static void btree_gc_start(struct cache_set *c)
 	c->gc_mark_valid = 0;
 	c->gc_done = ZERO_KEY;
 
-	for_each_cache(ca, c, i)
-		for_each_bucket(b, ca) {
-			b->last_gc = b->gen;
-			if (!atomic_read(&b->pin)) {
-				SET_GC_MARK(b, 0);
-				SET_GC_SECTORS_USED(b, 0);
-			}
+	ca = c->cache;
+	for_each_bucket(b, ca) {
+		b->last_gc = b->gen;
+		if (!atomic_read(&b->pin)) {
+			SET_GC_MARK(b, 0);
+			SET_GC_SECTORS_USED(b, 0);
 		}
+	}
 
 	mutex_unlock(&c->bucket_lock);
 }
@@ -1721,7 +1720,8 @@ static void bch_btree_gc_finish(struct cache_set *c)
 {
 	struct bucket *b;
 	struct cache *ca;
-	unsigned int i;
+	unsigned int i, j;
+	uint64_t *k;
 
 	mutex_lock(&c->bucket_lock);
 
@@ -1739,7 +1739,6 @@ static void bch_btree_gc_finish(struct cache_set *c)
 		struct bcache_device *d = c->devices[i];
 		struct cached_dev *dc;
 		struct keybuf_key *w, *n;
-		unsigned int j;
 
 		if (!d || UUID_FLASH_ONLY(&c->uuids[i]))
 			continue;
@@ -1756,29 +1755,27 @@ static void bch_btree_gc_finish(struct cache_set *c)
 	rcu_read_unlock();
 
 	c->avail_nbuckets = 0;
-	for_each_cache(ca, c, i) {
-		uint64_t *i;
 
-		ca->invalidate_needs_gc = 0;
+	ca = c->cache;
+	ca->invalidate_needs_gc = 0;
 
-		for (i = ca->sb.d; i < ca->sb.d + ca->sb.keys; i++)
-			SET_GC_MARK(ca->buckets + *i, GC_MARK_METADATA);
+	for (k = ca->sb.d; k < ca->sb.d + ca->sb.keys; k++)
+		SET_GC_MARK(ca->buckets + *k, GC_MARK_METADATA);
 
-		for (i = ca->prio_buckets;
-		     i < ca->prio_buckets + prio_buckets(ca) * 2; i++)
-			SET_GC_MARK(ca->buckets + *i, GC_MARK_METADATA);
+	for (k = ca->prio_buckets;
+	     k < ca->prio_buckets + prio_buckets(ca) * 2; k++)
+		SET_GC_MARK(ca->buckets + *k, GC_MARK_METADATA);
 
-		for_each_bucket(b, ca) {
-			c->need_gc	= max(c->need_gc, bucket_gc_gen(b));
+	for_each_bucket(b, ca) {
+		c->need_gc	= max(c->need_gc, bucket_gc_gen(b));
 
-			if (atomic_read(&b->pin))
-				continue;
+		if (atomic_read(&b->pin))
+			continue;
 
-			BUG_ON(!GC_MARK(b) && GC_SECTORS_USED(b));
+		BUG_ON(!GC_MARK(b) && GC_SECTORS_USED(b));
 
-			if (!GC_MARK(b) || GC_MARK(b) == GC_MARK_RECLAIMABLE)
-				c->avail_nbuckets++;
-		}
+		if (!GC_MARK(b) || GC_MARK(b) == GC_MARK_RECLAIMABLE)
+			c->avail_nbuckets++;
 	}
 
 	mutex_unlock(&c->bucket_lock);
@@ -1830,12 +1827,10 @@ static void bch_btree_gc(struct cache_set *c)
 
 static bool gc_should_run(struct cache_set *c)
 {
-	struct cache *ca;
-	unsigned int i;
+	struct cache *ca = c->cache;
 
-	for_each_cache(ca, c, i)
-		if (ca->invalidate_needs_gc)
-			return true;
+	if (ca->invalidate_needs_gc)
+		return true;
 
 	if (atomic_read(&c->sectors_to_gc) < 0)
 		return true;
@@ -2081,9 +2076,8 @@ out:
 
 void bch_initial_gc_finish(struct cache_set *c)
 {
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	struct bucket *b;
-	unsigned int i;
 
 	bch_btree_gc_finish(c);
 
@@ -2098,20 +2092,18 @@ void bch_initial_gc_finish(struct cache_set *c)
 	 * This is only safe for buckets that have no live data in them, which
 	 * there should always be some of.
 	 */
-	for_each_cache(ca, c, i) {
-		for_each_bucket(b, ca) {
-			if (fifo_full(&ca->free[RESERVE_PRIO]) &&
-			    fifo_full(&ca->free[RESERVE_BTREE]))
-				break;
+	for_each_bucket(b, ca) {
+		if (fifo_full(&ca->free[RESERVE_PRIO]) &&
+		    fifo_full(&ca->free[RESERVE_BTREE]))
+			break;
 
-			if (bch_can_invalidate_bucket(ca, b) &&
-			    !GC_MARK(b)) {
-				__bch_invalidate_one_bucket(ca, b);
-				if (!fifo_push(&ca->free[RESERVE_PRIO],
-				   b - ca->buckets))
-					fifo_push(&ca->free[RESERVE_BTREE],
-						  b - ca->buckets);
-			}
+		if (bch_can_invalidate_bucket(ca, b) &&
+		    !GC_MARK(b)) {
+			__bch_invalidate_one_bucket(ca, b);
+			if (!fifo_push(&ca->free[RESERVE_PRIO],
+			   b - ca->buckets))
+				fifo_push(&ca->free[RESERVE_BTREE],
+					  b - ca->buckets);
 		}
 	}
 
@@ -2219,7 +2211,7 @@ static int btree_split(struct btree *b, struct btree_op *op,
 		goto err;
 
 	split = set_blocks(btree_bset_first(n1),
-			   block_bytes(n1->c)) > (btree_blocks(b) * 4) / 5;
+			   block_bytes(n1->c->cache)) > (btree_blocks(b) * 4) / 5;
 
 	if (split) {
 		unsigned int keys = 0;
diff --git a/drivers/md/bcache/btree.h b/drivers/md/bcache/btree.h
index 257969980c49..50482107134f 100644
--- a/drivers/md/bcache/btree.h
+++ b/drivers/md/bcache/btree.h
@@ -194,7 +194,7 @@ static inline unsigned int bset_block_offset(struct btree *b, struct bset *i)
 
 static inline void set_gc_sectors(struct cache_set *c)
 {
-	atomic_set(&c->sectors_to_gc, c->sb.bucket_size * c->nbuckets / 16);
+	atomic_set(&c->sectors_to_gc, c->cache->sb.bucket_size * c->nbuckets / 16);
 }
 
 void bkey_put(struct cache_set *c, struct bkey *k);
diff --git a/drivers/md/bcache/closure.c b/drivers/md/bcache/closure.c
index 0164a1fe94a9..d8d9394a6beb 100644
--- a/drivers/md/bcache/closure.c
+++ b/drivers/md/bcache/closure.c
@@ -159,7 +159,7 @@ void closure_debug_destroy(struct closure *cl)
 
 static struct dentry *closure_debug;
 
-static int debug_seq_show(struct seq_file *f, void *data)
+static int debug_show(struct seq_file *f, void *data)
 {
 	struct closure *cl;
 
@@ -188,17 +188,7 @@ static int debug_seq_show(struct seq_file *f, void *data)
 	return 0;
 }
 
-static int debug_seq_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, debug_seq_show, NULL);
-}
-
-static const struct file_operations debug_ops = {
-	.owner		= THIS_MODULE,
-	.open		= debug_seq_open,
-	.read		= seq_read,
-	.release	= single_release
-};
+DEFINE_SHOW_ATTRIBUTE(debug);
 
 void  __init closure_debug_init(void)
 {
@@ -209,7 +199,7 @@ void  __init closure_debug_init(void)
 		 * about this.
 		 */
 		closure_debug = debugfs_create_file(
-			"closures", 0400, bcache_debug, NULL, &debug_ops);
+			"closures", 0400, bcache_debug, NULL, &debug_fops);
 }
 #endif
 
diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c
index 336f43910383..b00fd08d696b 100644
--- a/drivers/md/bcache/debug.c
+++ b/drivers/md/bcache/debug.c
@@ -25,8 +25,8 @@ struct dentry *bcache_debug;
 	for (i = (start);						\
 	     (void *) i < (void *) (start) + (KEY_SIZE(&b->key) << 9) &&\
 	     i->seq == (start)->seq;					\
-	     i = (void *) i + set_blocks(i, block_bytes(b->c)) *	\
-		 block_bytes(b->c))
+	     i = (void *) i + set_blocks(i, block_bytes(b->c->cache)) *	\
+		 block_bytes(b->c->cache))
 
 void bch_btree_verify(struct btree *b)
 {
@@ -82,14 +82,14 @@ void bch_btree_verify(struct btree *b)
 
 		for_each_written_bset(b, ondisk, i) {
 			unsigned int block = ((void *) i - (void *) ondisk) /
-				block_bytes(b->c);
+				block_bytes(b->c->cache);
 
 			pr_err("*** on disk block %u:\n", block);
 			bch_dump_bset(&b->keys, i, block);
 		}
 
 		pr_err("*** block %zu not written\n",
-		       ((void *) i - (void *) ondisk) / block_bytes(b->c));
+		       ((void *) i - (void *) ondisk) / block_bytes(b->c->cache));
 
 		for (j = 0; j < inmemory->keys; j++)
 			if (inmemory->d[j] != sorted->d[j])
@@ -238,7 +238,7 @@ void bch_debug_init_cache_set(struct cache_set *c)
 	if (!IS_ERR_OR_NULL(bcache_debug)) {
 		char name[50];
 
-		snprintf(name, 50, "bcache-%pU", c->sb.set_uuid);
+		snprintf(name, 50, "bcache-%pU", c->set_uuid);
 		c->debug = debugfs_create_file(name, 0400, bcache_debug, c,
 					       &cache_set_debug_ops);
 	}
diff --git a/drivers/md/bcache/extents.c b/drivers/md/bcache/extents.c
index 9162af5bb6ec..f4658a1f37b8 100644
--- a/drivers/md/bcache/extents.c
+++ b/drivers/md/bcache/extents.c
@@ -54,7 +54,7 @@ static bool __ptr_invalid(struct cache_set *c, const struct bkey *k)
 			size_t bucket = PTR_BUCKET_NR(c, k, i);
 			size_t r = bucket_remainder(c, PTR_OFFSET(k, i));
 
-			if (KEY_SIZE(k) + r > c->sb.bucket_size ||
+			if (KEY_SIZE(k) + r > c->cache->sb.bucket_size ||
 			    bucket <  ca->sb.first_bucket ||
 			    bucket >= ca->sb.nbuckets)
 				return true;
@@ -75,7 +75,7 @@ static const char *bch_ptr_status(struct cache_set *c, const struct bkey *k)
 			size_t bucket = PTR_BUCKET_NR(c, k, i);
 			size_t r = bucket_remainder(c, PTR_OFFSET(k, i));
 
-			if (KEY_SIZE(k) + r > c->sb.bucket_size)
+			if (KEY_SIZE(k) + r > c->cache->sb.bucket_size)
 				return "bad, length too big";
 			if (bucket <  ca->sb.first_bucket)
 				return "bad, short offset";
@@ -136,7 +136,7 @@ static void bch_bkey_dump(struct btree_keys *keys, const struct bkey *k)
 		size_t n = PTR_BUCKET_NR(b->c, k, j);
 
 		pr_cont(" bucket %zu", n);
-		if (n >= b->c->sb.first_bucket && n < b->c->sb.nbuckets)
+		if (n >= b->c->cache->sb.first_bucket && n < b->c->cache->sb.nbuckets)
 			pr_cont(" prio %i",
 				PTR_BUCKET(b->c, k, j)->prio);
 	}
diff --git a/drivers/md/bcache/features.c b/drivers/md/bcache/features.c
index 4442df48d28c..6469223f0b77 100644
--- a/drivers/md/bcache/features.c
+++ b/drivers/md/bcache/features.c
@@ -30,7 +30,7 @@ static struct feature feature_list[] = {
 	for (f = &feature_list[0]; f->compat != 0; f++) {		\
 		if (f->compat != BCH_FEATURE_ ## type)			\
 			continue;					\
-		if (BCH_HAS_ ## type ## _FEATURE(&c->sb, f->mask)) {	\
+		if (BCH_HAS_ ## type ## _FEATURE(&c->cache->sb, f->mask)) {	\
 			if (first) {					\
 				out += snprintf(out, buf + size - out,	\
 						"[");	\
@@ -44,7 +44,7 @@ static struct feature feature_list[] = {
 									\
 		out += snprintf(out, buf + size - out, "%s", f->string);\
 									\
-		if (BCH_HAS_ ## type ## _FEATURE(&c->sb, f->mask))	\
+		if (BCH_HAS_ ## type ## _FEATURE(&c->cache->sb, f->mask))	\
 			out += snprintf(out, buf + size - out, "]");	\
 									\
 		first = false;						\
diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
index a14a445618b4..dad71a6b7889 100644
--- a/drivers/md/bcache/io.c
+++ b/drivers/md/bcache/io.c
@@ -26,7 +26,7 @@ struct bio *bch_bbio_alloc(struct cache_set *c)
 	struct bbio *b = mempool_alloc(&c->bio_meta, GFP_NOIO);
 	struct bio *bio = &b->bio;
 
-	bio_init(bio, bio->bi_inline_vecs, meta_bucket_pages(&c->sb));
+	bio_init(bio, bio->bi_inline_vecs, meta_bucket_pages(&c->cache->sb));
 
 	return bio;
 }
diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c
index c1227bdb57e7..aefbdb7e003b 100644
--- a/drivers/md/bcache/journal.c
+++ b/drivers/md/bcache/journal.c
@@ -98,7 +98,7 @@ reread:		left = ca->sb.bucket_size - offset;
 				return ret;
 			}
 
-			blocks = set_blocks(j, block_bytes(ca->set));
+			blocks = set_blocks(j, block_bytes(ca));
 
 			/*
 			 * Nodes in 'list' are in linear increasing order of
@@ -179,112 +179,109 @@ int bch_journal_read(struct cache_set *c, struct list_head *list)
 		ret;							\
 	})
 
-	struct cache *ca;
-	unsigned int iter;
+	struct cache *ca = c->cache;
 	int ret = 0;
+	struct journal_device *ja = &ca->journal;
+	DECLARE_BITMAP(bitmap, SB_JOURNAL_BUCKETS);
+	unsigned int i, l, r, m;
+	uint64_t seq;
 
-	for_each_cache(ca, c, iter) {
-		struct journal_device *ja = &ca->journal;
-		DECLARE_BITMAP(bitmap, SB_JOURNAL_BUCKETS);
-		unsigned int i, l, r, m;
-		uint64_t seq;
-
-		bitmap_zero(bitmap, SB_JOURNAL_BUCKETS);
-		pr_debug("%u journal buckets\n", ca->sb.njournal_buckets);
+	bitmap_zero(bitmap, SB_JOURNAL_BUCKETS);
+	pr_debug("%u journal buckets\n", ca->sb.njournal_buckets);
 
+	/*
+	 * Read journal buckets ordered by golden ratio hash to quickly
+	 * find a sequence of buckets with valid journal entries
+	 */
+	for (i = 0; i < ca->sb.njournal_buckets; i++) {
 		/*
-		 * Read journal buckets ordered by golden ratio hash to quickly
-		 * find a sequence of buckets with valid journal entries
+		 * We must try the index l with ZERO first for
+		 * correctness due to the scenario that the journal
+		 * bucket is circular buffer which might have wrapped
 		 */
-		for (i = 0; i < ca->sb.njournal_buckets; i++) {
-			/*