Merge tag 'for-4.17/block-20180402' of git://git.kernel.dk/linux-block

Pull block layer updates from Jens Axboe:
 "It's a pretty quiet round this time, which is nice. This contains:

   - series from Bart, cleaning up the way we set/test/clear atomic
     queue flags.

   - series from Bart, fixing races between gendisk and queue
     registration and removal.

   - set of bcache fixes and improvements from various folks, by way of
     Michael Lyle.

   - set of lightnvm updates from Matias, most of it being the 1.2 to
     2.0 transition.

   - removal of unused DIO flags from Nikolay.

   - blk-mq/sbitmap memory ordering fixes from Omar.

   - divide-by-zero fix for BFQ from Paolo.

   - minor documentation patches from Randy.

   - timeout fix from Tejun.

   - Alpha "can't write a char atomically" fix from Mikulas.

   - set of NVMe fixes by way of Keith.

   - bsg and bsg-lib improvements from Christoph.

   - a few sed-opal fixes from Jonas.

   - cdrom check-disk-change deadlock fix from Maurizio.

   - various little fixes, comment fixes, etc from various folks"

* tag 'for-4.17/block-20180402' of git://git.kernel.dk/linux-block: (139 commits)
  blk-mq: Directly schedule q->timeout_work when aborting a request
  blktrace: fix comment in blktrace_api.h
  lightnvm: remove function name in strings
  lightnvm: pblk: remove some unnecessary NULL checks
  lightnvm: pblk: don't recover unwritten lines
  lightnvm: pblk: implement 2.0 support
  lightnvm: pblk: implement get log report chunk
  lightnvm: pblk: rename ppaf* to addrf*
  lightnvm: pblk: check for supported version
  lightnvm: implement get log report chunk helpers
  lightnvm: make address conversions depend on generic device
  lightnvm: add support for 2.0 address format
  lightnvm: normalize geometry nomenclature
  lightnvm: complete geo structure with maxoc*
  lightnvm: add shorten OCSSD version in geo
  lightnvm: add minor version to generic geometry
  lightnvm: simplify geometry structure
  lightnvm: pblk: refactor init/exit sequences
  lightnvm: Avoid validation of default op value
  lightnvm: centralize permission check for lightnvm ioctl
  ...

18 files changed, 556 insertions, 129 deletions

diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c
index 458e1d38577d..004cc3cc6123 100644
--- a/drivers/md/bcache/alloc.c
+++ b/drivers/md/bcache/alloc.c
@@ -287,7 +287,8 @@ do {									\
 			break;						\
 									\
 		mutex_unlock(&(ca)->set->bucket_lock);			\
-		if (kthread_should_stop()) {				\
+		if (kthread_should_stop() ||				\
+		    test_bit(CACHE_SET_IO_DISABLE, &ca->set->flags)) {	\
 			set_current_state(TASK_RUNNING);		\
 			return 0;					\
 		}							\
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index 12e5197f186c..d338b7086013 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -188,6 +188,7 @@
 #include <linux/refcount.h>
 #include <linux/types.h>
 #include <linux/workqueue.h>
+#include <linux/kthread.h>
 
 #include "bset.h"
 #include "util.h"
@@ -258,10 +259,11 @@ struct bcache_device {
 	struct gendisk		*disk;
 
 	unsigned long		flags;
-#define BCACHE_DEV_CLOSING	0
-#define BCACHE_DEV_DETACHING	1
-#define BCACHE_DEV_UNLINK_DONE	2
-
+#define BCACHE_DEV_CLOSING		0
+#define BCACHE_DEV_DETACHING		1
+#define BCACHE_DEV_UNLINK_DONE		2
+#define BCACHE_DEV_WB_RUNNING		3
+#define BCACHE_DEV_RATE_DW_RUNNING	4
 	unsigned		nr_stripes;
 	unsigned		stripe_size;
 	atomic_t		*stripe_sectors_dirty;
@@ -286,6 +288,12 @@ struct io {
 	sector_t		last;
 };
 
+enum stop_on_failure {
+	BCH_CACHED_DEV_STOP_AUTO = 0,
+	BCH_CACHED_DEV_STOP_ALWAYS,
+	BCH_CACHED_DEV_STOP_MODE_MAX,
+};
+
 struct cached_dev {
 	struct list_head	list;
 	struct bcache_device	disk;
@@ -359,6 +367,7 @@ struct cached_dev {
 	unsigned		sequential_cutoff;
 	unsigned		readahead;
 
+	unsigned		io_disable:1;
 	unsigned		verify:1;
 	unsigned		bypass_torture_test:1;
 
@@ -378,6 +387,11 @@ struct cached_dev {
 	unsigned		writeback_rate_i_term_inverse;
 	unsigned		writeback_rate_p_term_inverse;
 	unsigned		writeback_rate_minimum;
+
+	enum stop_on_failure	stop_when_cache_set_failed;
+#define DEFAULT_CACHED_DEV_ERROR_LIMIT	64
+	atomic_t		io_errors;
+	unsigned		error_limit;
 };
 
 enum alloc_reserve {
@@ -474,10 +488,15 @@ struct gc_stat {
  *
  * CACHE_SET_RUNNING means all cache devices have been registered and journal
  * replay is complete.
+ *
+ * CACHE_SET_IO_DISABLE is set when bcache is stopping the whold cache set, all
+ * external and internal I/O should be denied when this flag is set.
+ *
  */
 #define CACHE_SET_UNREGISTERING		0
 #define	CACHE_SET_STOPPING		1
 #define	CACHE_SET_RUNNING		2
+#define CACHE_SET_IO_DISABLE		3
 
 struct cache_set {
 	struct closure		cl;
@@ -867,8 +886,36 @@ static inline void wake_up_allocators(struct cache_set *c)
 		wake_up_process(ca->alloc_thread);
 }
 
+static inline void closure_bio_submit(struct cache_set *c,
+				      struct bio *bio,
+				      struct closure *cl)
+{
+	closure_get(cl);
+	if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags))) {
+		bio->bi_status = BLK_STS_IOERR;
+		bio_endio(bio);
+		return;
+	}
+	generic_make_request(bio);
+}
+
+/*
+ * Prevent the kthread exits directly, and make sure when kthread_stop()
+ * is called to stop a kthread, it is still alive. If a kthread might be
+ * stopped by CACHE_SET_IO_DISABLE bit set, wait_for_kthread_stop() is
+ * necessary before the kthread returns.
+ */
+static inline void wait_for_kthread_stop(void)
+{
+	while (!kthread_should_stop()) {
+		set_current_state(TASK_INTERRUPTIBLE);
+		schedule();
+	}
+}
+
 /* Forward declarations */
 
+void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio);
 void bch_count_io_errors(struct cache *, blk_status_t, int, const char *);
 void bch_bbio_count_io_errors(struct cache_set *, struct bio *,
 			      blk_status_t, const char *);
@@ -896,6 +943,7 @@ int bch_bucket_alloc_set(struct cache_set *, unsigned,
 			 struct bkey *, int, bool);
 bool bch_alloc_sectors(struct cache_set *, struct bkey *, unsigned,
 		       unsigned, unsigned, bool);
+bool bch_cached_dev_error(struct cached_dev *dc);
 
 __printf(2, 3)
 bool bch_cache_set_error(struct cache_set *, const char *, ...);
@@ -905,6 +953,7 @@ void bch_write_bdev_super(struct cached_dev *, struct closure *);
 
 extern struct workqueue_struct *bcache_wq;
 extern const char * const bch_cache_modes[];
+extern const char * const bch_stop_on_failure_modes[];
 extern struct mutex bch_register_lock;
 extern struct list_head bch_cache_sets;
 
diff --git a/drivers/md/bcache/bset.c b/drivers/md/bcache/bset.c
index e56d3ecdbfcb..579c696a5fe0 100644
--- a/drivers/md/bcache/bset.c
+++ b/drivers/md/bcache/bset.c
@@ -1072,7 +1072,7 @@ EXPORT_SYMBOL(bch_btree_iter_init);
 static inline struct bkey *__bch_btree_iter_next(struct btree_iter *iter,
 						 btree_iter_cmp_fn *cmp)
 {
-	struct btree_iter_set unused;
+	struct btree_iter_set b __maybe_unused;
 	struct bkey *ret = NULL;
 
 	if (!btree_iter_end(iter)) {
@@ -1087,7 +1087,7 @@ static inline struct bkey *__bch_btree_iter_next(struct btree_iter *iter,
 		}
 
 		if (iter->data->k == iter->data->end)
-			heap_pop(iter, unused, cmp);
+			heap_pop(iter, b, cmp);
 		else
 			heap_sift(iter, 0, cmp);
 	}
diff --git a/drivers/md/bcache/bset.h b/drivers/md/bcache/bset.h
index fa506c1aa524..0c24280f3b98 100644
--- a/drivers/md/bcache/bset.h
+++ b/drivers/md/bcache/bset.h
@@ -531,14 +531,15 @@ int __bch_keylist_realloc(struct keylist *, unsigned);
 #ifdef CONFIG_BCACHE_DEBUG
 
 int __bch_count_data(struct btree_keys *);
-void __bch_check_keys(struct btree_keys *, const char *, ...);
+void __printf(2, 3) __bch_check_keys(struct btree_keys *, const char *, ...);
 void bch_dump_bset(struct btree_keys *, struct bset *, unsigned);
 void bch_dump_bucket(struct btree_keys *);
 
 #else
 
 static inline int __bch_count_data(struct btree_keys *b) { return -1; }
-static inline void __bch_check_keys(struct btree_keys *b, const char *fmt, ...) {}
+static inline void __printf(2, 3)
+	__bch_check_keys(struct btree_keys *b, const char *fmt, ...) {}
 static inline void bch_dump_bucket(struct btree_keys *b) {}
 void bch_dump_bset(struct btree_keys *, struct bset *, unsigned);
 
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index fad9fe8817eb..17936b2dc7d6 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -665,6 +665,7 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 	struct btree *b, *t;
 	unsigned long i, nr = sc->nr_to_scan;
 	unsigned long freed = 0;
+	unsigned int btree_cache_used;
 
 	if (c->shrinker_disabled)
 		return SHRINK_STOP;
@@ -689,9 +690,10 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 	nr = min_t(unsigned long, nr, mca_can_free(c));
 
 	i = 0;
+	btree_cache_used = c->btree_cache_used;
 	list_for_each_entry_safe(b, t, &c->btree_cache_freeable, list) {
-		if (freed >= nr)
-			break;
+		if (nr <= 0)
+			goto out;
 
 		if (++i > 3 &&
 		    !mca_reap(b, 0, false)) {
@@ -699,9 +701,10 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 			rw_unlock(true, b);
 			freed++;
 		}
+		nr--;
 	}
 
-	for (i = 0; (nr--) && i < c->btree_cache_used; i++) {
+	for (;  (nr--) && i < btree_cache_used; i++) {
 		if (list_empty(&c->btree_cache))
 			goto out;
 
@@ -719,7 +722,7 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 	}
 out:
 	mutex_unlock(&c->bucket_lock);
-	return freed;
+	return freed * c->btree_pages;
 }
 
 static unsigned long bch_mca_count(struct shrinker *shrink,
@@ -959,7 +962,7 @@ err:
 	return b;
 }
 
-/**
+/*
  * bch_btree_node_get - find a btree node in the cache and lock it, reading it
  * in from disk if necessary.
  *
@@ -1744,6 +1747,7 @@ static void bch_btree_gc(struct cache_set *c)
 
 	btree_gc_start(c);
 
+	/* if CACHE_SET_IO_DISABLE set, gc thread should stop too */
 	do {
 		ret = btree_root(gc_root, c, &op, &writes, &stats);
 		closure_sync(&writes);
@@ -1751,7 +1755,7 @@ static void bch_btree_gc(struct cache_set *c)
 
 		if (ret && ret != -EAGAIN)
 			pr_warn("gc failed!");
-	} while (ret);
+	} while (ret && !test_bit(CACHE_SET_IO_DISABLE, &c->flags));
 
 	bch_btree_gc_finish(c);
 	wake_up_allocators(c);
@@ -1789,15 +1793,19 @@ static int bch_gc_thread(void *arg)
 
 	while (1) {
 		wait_event_interruptible(c->gc_wait,
-			   kthread_should_stop() || gc_should_run(c));
+			   kthread_should_stop() ||
+			   test_bit(CACHE_SET_IO_DISABLE, &c->flags) ||
+			   gc_should_run(c));
 
-		if (kthread_should_stop())
+		if (kthread_should_stop() ||
+		    test_bit(CACHE_SET_IO_DISABLE, &c->flags))
 			break;
 
 		set_gc_sectors(c);
 		bch_btree_gc(c);
 	}
 
+	wait_for_kthread_stop();
 	return 0;
 }
 
@@ -2170,7 +2178,7 @@ int bch_btree_insert_check_key(struct btree *b, struct btree_op *op,
 
 		if (b->key.ptr[0] != btree_ptr ||
                    b->seq != seq + 1) {
-                       op->lock = b->level;
+			op->lock = b->level;
 			goto out;
                }
 	}
diff --git a/drivers/md/bcache/closure.c b/drivers/md/bcache/closure.c
index 7f12920c14f7..0e14969182c6 100644
--- a/drivers/md/bcache/closure.c
+++ b/drivers/md/bcache/closure.c
@@ -46,7 +46,7 @@ void closure_sub(struct closure *cl, int v)
 }
 EXPORT_SYMBOL(closure_sub);
 
-/**
+/*
  * closure_put - decrement a closure's refcount
  */
 void closure_put(struct closure *cl)
@@ -55,7 +55,7 @@ void closure_put(struct closure *cl)
 }
 EXPORT_SYMBOL(closure_put);
 
-/**
+/*
  * closure_wake_up - wake up all closures on a wait list, without memory barrier
  */
 void __closure_wake_up(struct closure_waitlist *wait_list)
@@ -79,9 +79,9 @@ EXPORT_SYMBOL(__closure_wake_up);
 
 /**
  * closure_wait - add a closure to a waitlist
- *
- * @waitlist will own a ref on @cl, which will be released when
+ * @waitlist: will own a ref on @cl, which will be released when
  * closure_wake_up() is called on @waitlist.
+ * @cl: closure pointer.
  *
  */
 bool closure_wait(struct closure_waitlist *waitlist, struct closure *cl)
@@ -157,7 +157,7 @@ void closure_debug_destroy(struct closure *cl)
 }
 EXPORT_SYMBOL(closure_debug_destroy);
 
-static struct dentry *debug;
+static struct dentry *closure_debug;
 
 static int debug_seq_show(struct seq_file *f, void *data)
 {
@@ -199,11 +199,12 @@ static const struct file_operations debug_ops = {
 	.release	= single_release
 };
 
-void __init closure_debug_init(void)
+int __init closure_debug_init(void)
 {
-	debug = debugfs_create_file("closures", 0400, NULL, NULL, &debug_ops);
+	closure_debug = debugfs_create_file("closures",
+				0400, bcache_debug, NULL, &debug_ops);
+	return IS_ERR_OR_NULL(closure_debug);
 }
-
 #endif
 
 MODULE_AUTHOR("Kent Overstreet <koverstreet@google.com>");
diff --git a/drivers/md/bcache/closure.h b/drivers/md/bcache/closure.h
index 3b9dfc9962ad..71427eb5fdae 100644
--- a/drivers/md/bcache/closure.h
+++ b/drivers/md/bcache/closure.h
@@ -105,6 +105,7 @@
 struct closure;
 struct closure_syncer;
 typedef void (closure_fn) (struct closure *);
+extern struct dentry *bcache_debug;
 
 struct closure_waitlist {
 	struct llist_head	list;
@@ -185,13 +186,13 @@ static inline void closure_sync(struct closure *cl)
 
 #ifdef CONFIG_BCACHE_CLOSURES_DEBUG
 
-void closure_debug_init(void);
+int closure_debug_init(void);
 void closure_debug_create(struct closure *cl);
 void closure_debug_destroy(struct closure *cl);
 
 #else
 
-static inline void closure_debug_init(void) {}
+static inline int closure_debug_init(void) { return 0; }
 static inline void closure_debug_create(struct closure *cl) {}
 static inline void closure_debug_destroy(struct closure *cl) {}
 
diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c
index af89408befe8..028f7b386e01 100644
--- a/drivers/md/bcache/debug.c
+++ b/drivers/md/bcache/debug.c
@@ -17,7 +17,7 @@
 #include <linux/random.h>
 #include <linux/seq_file.h>
 
-static struct dentry *debug;
+struct dentry *bcache_debug;
 
 #ifdef CONFIG_BCACHE_DEBUG
 
@@ -232,11 +232,11 @@ static const struct file_operations cache_set_debug_ops = {
 
 void bch_debug_init_cache_set(struct cache_set *c)
 {
-	if (!IS_ERR_OR_NULL(debug)) {
+	if (!IS_ERR_OR_NULL(bcache_debug)) {
 		char name[50];
 		snprintf(name, 50, "bcache-%pU", c->sb.set_uuid);
 
-		c->debug = debugfs_create_file(name, 0400, debug, c,
+		c->debug = debugfs_create_file(name, 0400, bcache_debug, c,
 					       &cache_set_debug_ops);
 	}
 }
@@ -245,13 +245,13 @@ void bch_debug_init_cache_set(struct cache_set *c)
 
 void bch_debug_exit(void)
 {
-	if (!IS_ERR_OR_NULL(debug))
-		debugfs_remove_recursive(debug);
+	if (!IS_ERR_OR_NULL(bcache_debug))
+		debugfs_remove_recursive(bcache_debug);
 }
 
 int __init bch_debug_init(struct kobject *kobj)
 {
-	debug = debugfs_create_dir("bcache", NULL);
+	bcache_debug = debugfs_create_dir("bcache", NULL);
 
-	return IS_ERR_OR_NULL(debug);
+	return IS_ERR_OR_NULL(bcache_debug);
 }
diff --git a/drivers/md/bcache/extents.c b/drivers/md/bcache/extents.c
index f9d391711595..c334e6666461 100644
--- a/drivers/md/bcache/extents.c
+++ b/drivers/md/bcache/extents.c
@@ -534,7 +534,6 @@ err:
 static bool bch_extent_bad(struct btree_keys *bk, const struct bkey *k)
 {
 	struct btree *b = container_of(bk, struct btree, keys);
-	struct bucket *g;
 	unsigned i, stale;
 
 	if (!KEY_PTRS(k) ||
@@ -549,7 +548,6 @@ static bool bch_extent_bad(struct btree_keys *bk, const struct bkey *k)
 		return false;
 
 	for (i = 0; i < KEY_PTRS(k); i++) {
-		g = PTR_BUCKET(b->c, k, i);
 		stale = ptr_stale(b->c, k, i);
 
 		btree_bug_on(stale > 96, b,
diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
index a783c5a41ff1..7fac97ae036e 100644
--- a/drivers/md/bcache/io.c
+++ b/drivers/md/bcache/io.c
@@ -38,7 +38,7 @@ void __bch_submit_bbio(struct bio *bio, struct cache_set *c)
 	bio_set_dev(bio, PTR_CACHE(c, &b->key, 0)->bdev);
 
 	b->submit_time_us = local_clock_us();
-	closure_bio_submit(bio, bio->bi_private);
+	closure_bio_submit(c, bio, bio->bi_private);
 }
 
 void bch_submit_bbio(struct bio *bio, struct cache_set *c,
@@ -50,6 +50,20 @@ void bch_submit_bbio(struct bio *bio, struct cache_set *c,
 }
 
 /* IO errors */
+void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio)
+{
+	char buf[BDEVNAME_SIZE];
+	unsigned errors;
+
+	WARN_ONCE(!dc, "NULL pointer of struct cached_dev");
+
+	errors = atomic_add_return(1, &dc->io_errors);
+	if (errors < dc->error_limit)
+		pr_err("%s: IO error on backing device, unrecoverable",
+			bio_devname(bio, buf));
+	else
+		bch_cached_dev_error(dc);
+}
 
 void bch_count_io_errors(struct cache *ca,
 			 blk_status_t error,
diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c
index 1b736b860739..18f1b5239620 100644
--- a/drivers/md/bcache/journal.c
+++ b/drivers/md/bcache/journal.c
@@ -62,7 +62,7 @@ reread:		left = ca->sb.bucket_size - offset;
 		bio_set_op_attrs(bio, REQ_OP_READ, 0);
 		bch_bio_map(bio, data);
 
-		closure_bio_submit(bio, &cl);
+		closure_bio_submit(ca->set, bio, &cl);
 		closure_sync(&cl);
 
 		/* This function could be simpler now since we no longer write
@@ -493,7 +493,7 @@ static void journal_reclaim(struct cache_set *c)
 	struct cache *ca;
 	uint64_t last_seq;
 	unsigned iter, n = 0;
-	atomic_t p;
+	atomic_t p __maybe_unused;
 
 	atomic_long_inc(&c->reclaim);
 
@@ -594,6 +594,7 @@ static void journal_write_done(struct closure *cl)
 }
 
 static void journal_write_unlock(struct closure *cl)
+	__releases(&c->journal.lock)
 {
 	struct cache_set *c = container_of(cl, struct cache_set, journal.io);
 
@@ -674,7 +675,7 @@ static void journal_write_unlocked(struct closure *cl)
 	spin_unlock(&c->journal.lock);
 
 	while ((bio = bio_list_pop(&list)))
-		closure_bio_submit(bio, cl);
+		closure_bio_submit(c, bio, cl);
 
 	continue_at(cl, journal_write_done, NULL);
 }
@@ -705,6 +706,7 @@ static void journal_try_write(struct cache_set *c)
 
 static struct journal_write *journal_wait_for_write(struct cache_set *c,
 						    unsigned nkeys)
+	__acquires(&c->journal.lock)
 {
 	size_t sectors;
 	struct closure cl;
diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
index 6422846b546e..a65e3365eeb9 100644
--- a/drivers/md/bcache/request.c
+++ b/drivers/md/bcache/request.c
@@ -139,6 +139,7 @@ static void bch_data_invalidate(struct closure *cl)
 	}
 
 	op->insert_data_done = true;
+	/* get in bch_data_insert() */
 	bio_put(bio);
 out:
 	continue_at(cl, bch_data_insert_keys, op->wq);
@@ -295,6 +296,7 @@ err:
 
 /**
  * bch_data_insert - stick some data in the cache
+ * @cl: closure pointer.
  *
  * This is the starting point for any data to end up in a cache device; it could
  * be from a normal write, or a writeback write, or a write to a flash only
@@ -630,6 +632,41 @@ static void request_endio(struct bio *bio)
 	closure_put(cl);
 }
 
+static void backing_request_endio(struct bio *bio)
+{
+	struct closure *cl = bio->bi_private;
+
+	if (bio->bi_status) {
+		struct search *s = container_of(cl, struct search, cl);
+		struct cached_dev *dc = container_of(s->d,
+						     struct cached_dev, disk);
+		/*
+		 * If a bio has REQ_PREFLUSH for writeback mode, it is
+		 * speically assembled in cached_dev_write() for a non-zero
+		 * write request which has REQ_PREFLUSH. we don't set
+		 * s->iop.status by this failure, the status will be decided
+		 * by result of bch_data_insert() operation.
+		 */
+		if (unlikely(s->iop.writeback &&
+			     bio->bi_opf & REQ_PREFLUSH)) {
+			char buf[BDEVNAME_SIZE];
+
+			bio_devname(bio, buf);
+			pr_err("Can't flush %s: returned bi_status %i",
+				buf, bio->bi_status);
+		} else {
+			/* set to orig_bio->bi_status in bio_complete() */
+			s->iop.status = bio->bi_status;
+		}
+		s->recoverable = false;
+		/* should count I/O error for backing device here */
+		bch_count_backing_io_errors(dc, bio);
+	}
+
+	bio_put(bio);
+	closure_put(cl);
+}
+
 static void bio_complete(struct search *s)
 {
 	if (s->orig_bio) {
@@ -644,13 +681,21 @@ static void bio_complete(struct search *s)
 	}
 }
 
-static void do_bio_hook(struct search *s, struct bio *orig_bio)
+static void do_bio_hook(struct search *s,
+			struct bio *orig_bio,
+			bio_end_io_t *end_io_fn)
 {
 	struct bio *bio = &s->bio.bio;
 
 	bio_init(bio, NULL, 0);
 	__bio_clone_fast(bio, orig_bio);
-	bio->bi_end_io		= request_endio;
+	/*
+	 * bi_end_io can be set separately somewhere else, e.g. the
+	 * variants in,
+	 * - cache_bio->bi_end_io from cached_dev_cache_miss()
+	 * - n->bi_end_io from cache_lookup_fn()
+	 */
+	bio->bi_end_io		= end_io_fn;
 	bio->bi_private		= &s->cl;
 
 	bio_cnt_set(bio, 3);
@@ -676,7 +721,7 @@ static inline struct search *search_alloc(struct bio *bio,
 	s = mempool_alloc(d->c->search, GFP_NOIO);
 
 	closure_init(&s->cl, NULL);
-	do_bio_hook(s, bio);
+	do_bio_hook(s, bio, request_endio);
 
 	s->orig_bio		= bio;
 	s->cache_miss		= NULL;
@@ -743,11 +788,12 @@ static void cached_dev_read_error(struct closure *cl)
 		trace_bcache_read_retry(s->orig_bio);
 
 		s->iop.status = 0;
-		do_bio_hook(s, s->orig_bio);
+		do_bio_hook(s, s->orig_bio, backing_request_endio);
 
 		/* XXX: invalidate cache */
 
-		closure_bio_submit(bio, cl);
+		/* I/O request sent to backing device */
+		closure_bio_submit(s->iop.c, bio, cl);
 	}
 
 	continue_at(cl, cached_dev_cache_miss_done, NULL);
@@ -859,7 +905,7 @@ static int cached_dev_cache_miss(struct btree *b, struct search *s,
 	bio_copy_dev(cache_bio, miss);
 	cache_bio->bi_iter.bi_size	= s->insert_bio_sectors << 9;
 
-	cache_bio->bi_end_io	= request_endio;
+	cache_bio->bi_end_io	= backing_request_endio;
 	cache_bio->bi_private	= &s->cl;
 
 	bch_bio_map(cache_bio, NULL);
@@ -872,15 +918,17 @@ static int cached_dev_cache_miss(struct btree *b, struct search *s,
 	s->cache_miss	= miss;
 	s->iop.bio	= cache_bio;
 	bio_get(cache_bio);
-	closure_bio_submit(cache_bio, &s->cl);
+	/* I/O request sent to backing device */
+	closure_bio_submit(s->iop.c, cache_bio, &s->cl);
 
 	return ret;
 out_put:
 	bio_put(cache_bio);
 out_submit:
-	miss->bi_end_io		= request_endio;
+	miss->bi_end_io		= backing_request_endio;
 	miss->bi_private	= &s->cl;
-	closure_bio_submit(miss, &s->cl);
+	/* I/O request sent to backing device */
+	closure_bio_submit(s->iop.c, miss, &s->cl);
 	return ret;
 }
 
@@ -943,31 +991,46 @@ static void cached_dev_write(struct cached_dev *dc, struct search *s)
 		s->iop.bio = s->orig_bio;
 		bio_get(s->iop.bio);
 
-		if ((bio_op(bio) != REQ_OP_DISCARD) ||
-		    blk_queue_discard(bdev_get_queue(dc->bdev)))
-			closure_bio_submit(bio, cl);
+		if (bio_op(bio) == REQ_OP_DISCARD &&
+		    !blk_queue_discard(bdev_get_queue(dc->bdev)))
+			goto insert_data;
+
+		/* I/O request sent to backing device */
+		bio->bi_end_io = backing_request_endio;
+		closure_bio_submit(s->iop.c, bio, cl);
+
 	} else if (s->iop.writeback) {
 		bch_writeback_add(dc);
 		s->iop.bio = bio;
 
 		if (bio->bi_opf & REQ_PREFLUSH) {
-			/* Also need to send a flush to the backing device */
-			struct bio *flush = bio_alloc_bioset(GFP_NOIO, 0,
-							     dc->disk.bio_split);
-
+			/*
+			 * Also need to send a flush to the backing
+			 * device.
+			 */
+			struct bio *flush;
+
+			flush = bio_alloc_bioset(GFP_NOIO, 0,
+						 dc->disk.bio_split);
+			if (!flush) {
+				s->iop.status = BLK_STS_RESOURCE;
+				goto insert_data;
+			}
 			bio_copy_dev(flush, bio);
-			flush->bi_end_io = request_endio;
+			flush->bi_end_io = backing_request_endio;
 			flush->bi_private = cl;
 			flush->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH;
-
-			closure_bio_submit(flush, cl);
+			/* I/O request sent to backing device */
+			closure_bio_submit(s->iop.c, flush, cl);
 		}
 	} else {
 		s->iop.bio = bio_clone_fast(bio, GFP_NOIO, dc->disk.bio_split);
-
-		closure_bio_submit(bio, cl);
+		/* I/O request sent to backing device */
+		bio->bi_end_io = backing_request_endio;
+		closure_bio_submit(s->iop.c, bio, cl);
 	}
 
+insert_data:
 	closure_call(&s->iop.cl, bch_data_insert, NULL, cl);
 	continue_at(cl, cached_dev_write_complete, NULL);
 }
@@ -981,11 +1044,67 @@ static void cached_dev_nodata(struct closure *cl)
 		bch_journal_meta(s->iop.c, cl);
 
 	/* If it's a flush, we send the flush to the backing device too */
-	closure_bio_submit(bio, cl);
+	bio->bi_end_io = backing_request_endio;
+	closure_bio_submit(s->iop.c, bio, cl);
 
 	continue_at(cl, cached_dev_bio_complete, NULL);
 }
 
+struct detached_dev_io_private {
+	struct bcache_device	*d;
+	unsigned long		start_time;
+	bio_end_io_t		*bi_end_io;
+	void			*bi_private;
+};
+
+static void detached_dev_end_io(struct bio *bio)
+{
+	struct detached_dev_io_private *ddip;
+
+	ddip = bio->bi_private;
+	bio->bi_end_io = ddip->bi_end_io;
+	bio->bi_private = ddip->bi_private;
+
+	generic_end_io_acct(ddip->d->disk->queue,
+			    bio_data_dir(bio),
+			    &ddip->d->disk->part0, ddip->start_time);
+
+	if (bio->bi_status) {
+		struct cached_dev *dc = container_of(ddip->d,
+						     struct cached_dev, disk);
+		/* should count I/O error for backing device here */
+		bch_count_backing_io_errors(dc, bio);
+	}
+
+	kfree(ddip);
+	bio->bi_end_io(bio);
+}
+
+static void detached_dev_do_request(struct bcache_device *d, struct bio *bio)
+{
+	struct detached_dev_io_private *ddip;
+	struct cached_dev *dc = container_of(d, struct cached_dev, disk);
+
+	/*
+	 * no need to call closure_get(&dc->disk.cl),
+	 * because upper layer had already opened bcache device,
+	 * which would call closure_get(&dc->disk.cl)
+	 */
+	ddip = kzalloc(sizeof(struct detached_dev_io_private), GFP_NOIO);
+	ddip->d = d;
+	ddip->start_time = jiffies;
+	ddip->bi_end_io = bio->bi_end_io;
+	ddip->bi_private = bio->bi_private;
+	bio->bi_end_io = detached_dev_end_io;
+	bio->bi_private = ddip;
+
+	if ((bio_op(bio) == REQ_OP_DISCARD) &&
+	    !blk_queue_discard(bdev_get_queue(dc->bdev)))
+		bio->bi_end_io(bio);
+	else
+		generic_make_request(bio);
+}
+
 /* Cached devices - read & write stuff */
 
 static blk_qc_t cached_dev_make_request(struct request_queue *q,
@@ -996,6 +1115,13 @@ static blk_qc_t cached_dev_make_request(struct request_queue *q,
 	struct cached_dev *dc = container_of(d, struct cached_dev, disk);
 	int rw = bio_data_dir(bio);
 
+	if (unlikely((d->c && test_bit(CACHE_SET_IO_DISABLE, &d->c->flags)) ||
+		     dc->io_disable)) {
+		bio->bi_status = BLK_STS_IOERR;
+		bio_endio(bio);
+		return BLK_QC_T_NONE;
+	}
+
 	atomic_set(&dc->backing_idle, 0);
 	generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0);
 
@@ -1022,13 +1148,9 @@ static blk_qc_t cached_dev_make_request(struct request_queue *q,
 			else
 				cached_dev_read(dc, s);
 		}
-	} else {
-		if ((bio_op(bio) == REQ_OP_DISCARD) &&
-		    !blk_queue_discard(bdev_get_queue(dc->bdev)))
-			bio_endio(bio);
-		else
-			generic_make_request(bio);
-	}
+	} else
+		/* I/O request sent to backing device */
+		detached_dev_do_request(d, bio);
 
 	return BLK_QC_T_NONE;
 }
@@ -1112,6 +1234,12 @@ static blk_qc_t flash_dev_make_request(struct request_queue *q,
 	struct bcache_device *d = bio->bi_disk->private_data;
 	int rw = bio_data_dir(bio);
 
+	if (unlikely(d->c && test_bit(CACHE_SET_IO_DISABLE, &d->c->flags))) {
+		bio->bi_status = BLK_STS_IOERR;
+		bio_endio(bio);
+		return BLK_QC_T_NONE;
+	}
+
 	generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0);
 
 	s = search_alloc(bio, d);
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index f2273143b3cb..d90d9e59ca00 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -47,6 +47,14 @@ const char * const bch_cache_modes[] = {
 	NULL
 };
 
+/* Default is -1; we skip past it for stop_when_cache_set_failed */
+const char * const bch_stop_on_failure_modes[] = {
+	"default",
+	"auto",
+	"always",
+	NULL
+};
+
 static struct kobject *bcache_kobj;
 struct mutex bch_register_lock;
 LIST_HEAD(bch_cache_sets);
@@ -265,6 +273,7 @@ void bch_write_bdev_super(struct cached_dev *dc, struct closure *parent)
 	bio->bi_private = dc;
 
 	closure_get(cl);
+	/* I/O request sent to backing device */
 	__write_super(&dc->sb, bio);
 
 	closure_return_with_destructor(cl, bch_write_bdev_super_unlock);
@@ -521,7 +530,7 @@ static void prio_io(struct cache *ca, uint64_t bucket, int op,
 	bio_set_op_attrs(bio, op, REQ_SYNC|REQ_META|op_flags);
 	bch_bio_map(bio, ca->disk_buckets);
 
-	closure_bio_submit(bio, &ca->prio);
+	closure_bio_submit(ca->set, bio, &ca->prio);
 	closure_sync(cl);
 }
 
@@ -769,6 +778,8 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size,
 			      sector_t sectors)
 {
 	struct request_queue *q;
+	const size_t max_stripes = min_t(size_t, INT_MAX,
+					 SIZE_MAX / sizeof(atomic_t));
 	size_t n;
 	int idx;
 
@@ -777,9 +788,7 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size,
 
 	d->nr_stripes = DIV_ROUND_UP_ULL(sectors, d->stripe_size);
 
-	if (!d->nr_stripes ||
-	    d->nr_stripes > INT_MAX ||
-	    d->nr_stripes > SIZE_MAX / sizeof(atomic_t)) {
+	if (!d->nr_stripes || d->nr_stripes > max_stripes) {
 		pr_err("nr_stripes too large or invalid: %u (start sector beyond end of disk?)",
 			(unsigned)d->nr_stripes);
 		return -ENOMEM;
@@ -833,9 +842,9 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size,
 	q->limits.io_min		= block_size;
 	q->limits.logical_block_size	= block_size;
 	q->limits.physical_block_size	= block_size;
-	set_bit(QUEUE_FLAG_NONROT,	&d->disk->queue->queue_flags);
-	clear_bit(QUEUE_FLAG_ADD_RANDOM, &d->disk->queue->queue_flags);
-	set_bit(QUEUE_FLAG_DISCARD,	&d->disk->queue->queue_flags);
+	blk_queue_flag_set(QUEUE_FLAG_NONROT, d->disk->queue);
+	blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, d->disk->queue);
+	blk_queue_flag_set(QUEUE_FLAG_DISCARD, d->disk->queue);
 
 	blk_queue_write_cache(q, true, true);
 
@@ -899,6 +908,31 @@ void bch_cached_dev_run(struct cached_dev *dc)
 		pr_debug("error creating sysfs link");
 }
 
+/*
+ * If BCACHE_DEV_RATE_DW_RUNNING is set, it means routine of the delayed
+ * work dc->writeback_rate_update is running. Wait until the routine
+ * quits (BCACHE_DEV_RATE_DW_RUNNING is clear), then continue to
+ * cancel it. If BCACHE_DEV_RATE_DW_RUNNING is not clear after time_out
+ * seconds, give up waiting here and continue to cancel it too.
+ */
+static void cancel_writeback_rate_update_dwork(struct cached_dev *dc)
+{
+	int time_out = WRITEBACK_RATE_UPDATE_SECS_MAX * HZ;
+
+	do {
+		if (!test_bit(BCACHE_DEV_RATE_DW_RUNNING,
+			      &dc->disk.flags))
+			break;
+		time_out--;
+		schedule_timeout_interruptible(1);
+	} while (time_out > 0);
+
+	if (time_out == 0)
+		pr_warn("give up waiting for dc->writeback_write_update to quit");
+
+	cancel_delayed_work_sync(&dc->writeback_rate_update);
+}
+
 static void cached_dev_detach_finish(struct work_struct *w)
 {
 	struct cached_dev *dc = container_of(w, struct cached_dev, detach);
@@ -911,7 +945,9 @@ static void cached_dev_detach_finish(struct work_struct *w)
 
 	mutex_lock(&bch_register_lock);
 
-	cancel_delayed_work_sync(&dc->writeback_rate_update);
+	if (test_and_clear_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags))
+		cancel_writeback_rate_update_dwork(dc);
+
 	if (!IS_ERR_OR_NULL(dc->writeback_thread)) {
 		kthread_stop(dc->writeback_thread);
 		dc->writeback_thread = NULL;
@@ -954,6 +990,7 @@ void bch_cached_dev_detach(struct cached_dev *dc)
 	closure_get(&dc->disk.cl);
 
 	bch_writeback_queue(dc);
+
 	cached_dev_put(dc);
 }
 
@@ -1065,7 +1102,6 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
 	if (BDEV_STATE(&dc->sb) == BDEV_STATE_DIRTY) {
 		bch_sectors_dirty_init(&dc->disk);
 		atomic_set(&dc->has_dirty, 1);
-		refcount_inc(&dc->count);
 		bch_writeback_queue(dc);
 	}
 
@@ -1093,14 +1129,16 @@ static void cached_dev_free(struct closure *cl)
 {
 	struct cached_dev *dc = container_of(cl, struct cached_dev, disk.cl);
 
-	cancel_delayed_work_sync(&dc->writeback_rate_update);
+	mutex_lock(&bch_register_lock);
+
+	if (test_and_clear_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags))
+		cancel_writeback_rate_update_dwork(dc);
+
 	if (!IS_ERR_OR_NULL(dc->writeback_thread))
 		kthread_stop(dc->writeback_thread);
 	if (dc->writeback_write_wq)
 		destroy_workqueue(dc->writeback_write_wq);
 
-	mutex_lock(&bch_register_lock);
-
 	if (atomic_read(&dc->running))
 		bd_unlink_disk_holder(dc->bdev, dc->disk.disk);
 	bcache_device_free(&dc->disk);
@@ -1170,6 +1208,12 @@ static int cached_dev_init(struct cached_dev *dc, unsigned block_size)
 		max(dc->disk.disk->queue->backing_dev_info->ra_pages,
 		    q->backing_dev_info->ra_pages);
 
+	atomic_set(&dc->io_errors, 0);
+	dc->io_disable = false;
+	dc->error_limit = DEFAULT_CACHED_DEV_ERROR_LIMIT;
+	/* default to auto */
+	dc->stop_when_cache_set_failed = BCH_CACHED_DEV_STOP_AUTO;
+
 	bch_cached_dev_request_init(dc);
 	bch_cached_dev_writeback_init(dc);
 	return 0;
@@ -1321,6 +1365,24 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size)
 	return flash_dev_run(c, u);
 }
 
+bool bch_cached_dev_error(struct cached_dev *dc)
+{
+	char name[BDEVNAME_SIZE];
+
+	if (!dc || test_bit(BCACHE_DEV_CLOSING, &dc->disk.flags))
+		return false;
+
+	dc->io_disable = true;
+	/* make others know io_disable is true earlier */
+	smp_mb();
+
+	pr_err("stop %s: too many IO errors on backing device %s\n",
+		dc->disk.disk->disk_name, bdevname(dc->bdev, name));
+
+	bcache_device_stop(&dc->disk);
+	return true;
+}
+
 /* Cache set */
 
 __printf(2, 3)
@@ -1332,6 +1394,9 @@ bool bch_cache_set_error(struct cache_set *c, const char *fmt, ...)
 	    test_bit(CACHE_SET_STOPPING, &c->flags))
 		return false;
 
+	if (test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags))
+		pr_warn("CACHE_SET_IO_DISABLE already set");
+
 	/* XXX: we can be called from atomic context
 	acquire_console_sem();
 	*/
@@ -1443,25 +1508,72 @@ static void cache_set_flush(struct closure *cl)
 	closure_return(cl);
 }
 
+/*
+ * This function is only called when CACHE_SET_IO_DISABLE is set, which means
+ * cache set is unregistering due to too many I/O errors. In this condition,
+ * the bcache device might be stopped, it depends on stop_when_cache_set_failed
+ * value and whether the broken cache has dirty data:
+ *
+ * dc->stop_when_cache_set_failed    dc->has_dirty   stop bcache device
+ *  BCH_CACHED_STOP_AUTO               0               NO
+ *  BCH_CACHED_STOP_AUTO               1               YES
+ *  BCH_CACHED_DEV_STOP_ALWAYS         0               YES
+ *  BCH_CACHED_DEV_STOP_ALWAYS         1               YES
+ *
+ * The expected behavior is, if stop_when_cache_set_failed is configured to
+ * "auto" via sysfs interface, the bcache device will not be stopped if the
+ * backing device is clean on the broken cache device.
+ */
+static void conditional_stop_bcache_device(struct cache_set *c,
+					   struct bcache_device *d,
+					   struct cached_dev *dc)
+{
+	if (dc->stop_when_cache_set_failed == BCH_CACHED_DEV_STOP_ALWAYS) {
+		pr_warn("stop_when_cache_set_failed of %s is \"always\", stop it for failed cache set %pU.",
+			d->disk->disk_name, c->sb.set_uuid);
+		bcache_device_stop(d);
+	} else if (atomic_read(&dc->has_dirty)) {
+		/*
+		 * dc->stop_when_cache_set_failed == BCH_CACHED_STOP_AUTO
+		 * and dc->has_dirty == 1
+		 */
+		pr_warn("stop_when_cache_set_failed of %s is \"auto\" and cache is dirty, stop it to avoid potential data corruption.",
+			d->disk->disk_name);
+			bcache_device_stop(d);
+	} else {
+		/*
+		 * dc->stop_when_cache_set_failed == BCH_CACHED_STOP_AUTO
+		 * and dc->has_dirty == 0
+		 */
+		pr_warn("stop_when_cache_set_failed of %s is \"auto\" and cache is clean, keep it alive.",
+			d->disk->disk_name);
+	}
+}
+
 static void __cache_set_unregister(struct closure *cl)
 {
 	struct cache_set *c = container_of(cl, struct cache_set, caching);
 	struct cached_dev *dc;
+	struct bcache_device *d;
 	size_t i;
 
 	mutex_lock(&bch_register_lock);
 
-	for (i = 0; i < c->devices_max_used; i++)
-		if (c->devices[i]) {
-			if (!UUID_FLASH_ONLY(&c->uuids[i]) &&
-			    test_bit(CACHE_SET_UNREGISTERING, &c->flags)) {
-				dc = container_of(c->devices[i],
-						  struct cached_dev, disk);
-				bch_cached_dev_detach(dc);
-			} else {
-				bcache_device_stop(c->devices[i]);
-			}
+	for (i = 0; i < c->devices_max_used; i++) {
+		d = c->devices[i];
+		if (!d)
+			continue;
+
+		if (!UUID_FLASH_ONLY(&c->uuids[i]) &&
+		    test_bit(CACHE_SET_UNREGISTERING, &c->flags)) {
+			dc = container_of(d, struct cached_dev, disk);
+			bch_cached_dev_detach(dc);
+			if (test_bit(CACHE_SET_IO_DISABLE, &c->flags))
+				conditional_stop_bcache_device(c, d, dc);
+		} else {
+			bcache_device_stop(d);
 		}
+	}
 
 	mutex_unlock(&bch_register_lock);
 
@@ -1567,6 +1679,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
 	c->congested_read_threshold_us	= 2000;
 	c->congested_write_threshold_us	= 20000;
 	c->error_limit	= DEFAULT_IO_ERROR_LIMIT;
+	WARN_ON(test_and_clear_bit(CACHE_SET_IO_DISABLE, &c->flags));
 
 	return c;
 err:
@@ -2148,7 +2261,6 @@ static int __init bcache_init(void)
 	mutex_init(&bch_register_lock);
 	init_waitqueue_head(&unregister_wait);
 	register_reboot_notifier(&reboot);
-	closure_debug_init();
 
 	bcache_major = register_blkdev(0, "bcache");
 	if (bcache_major < 0) {
@@ -2160,7 +2272,7 @@ static int __init bcache_init(void)
 	if (!(bcache_wq = alloc_workqueue("bcache", WQ_MEM_RECLAIM, 0)) ||
 	    !(bcache_kobj = kobject_create_and_add("bcache", fs_kobj)) ||
 	    bch_request_init() ||
-	    bch_debug_init(bcache_kobj) ||
+	    bch_debug_init(bcache_kobj) || closure_debug_init() ||
 	    sysfs_create_files(bcache_kobj, files))
 		goto err;
 
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c
index 78cd7bd50fdd..dfeef583ee50 100644
--- a/drivers/md/bcache/sysfs.c
+++ b/drivers/md/bcache/sysfs.c
@@ -78,6 +78,7 @@ rw_attribute(congested_write_threshold_us);
 rw_attribute(sequential_cutoff);
 rw_attribute(data_csum);
 rw_attribute(cache_mode);
+rw_attribute(stop_when_cache_set_failed);
 rw_attribute(writeback_metadata);
 rw_attribute(writeback_running);
 rw_attribute(writeback_percent);
@@ -95,6 +96,7 @@ read_attribute(partial_stripes_expensive);
 
 rw_attribute(synchronous);
 rw_attribute(journal_delay_ms);
+rw_attribute(io_disable);
 rw_attribute(discard);
 rw_attribute(running);
 rw_attribute(label);
@@ -125,6 +127,12 @@ SHOW(__bch_cached_dev)
 					       bch_cache_modes + 1,
 					       BDEV_CACHE_MODE(&dc->sb));
 
+	if (attr == &sysfs_stop_when_cache_set_failed)
+		return bch_snprint_string_list(buf, PAGE_SIZE,
+					       bch_stop_on_failure_modes + 1,
+					       dc->stop_when_cache_set_failed);
+
+
 	sysfs_printf(data_csum,		"%i", dc->disk.data_csum);
 	var_printf(verify,		"%i");
 	var_printf(bypass_torture_test,	"%i");
@@ -133,7 +141,9 @@ SHOW(__bch_cached_dev)
 	var_print(writeback_delay);
 	var_print(writeback_percent);
 	sysfs_hprint(writeback_rate,	dc->writeback_rate.rate << 9);
-
+	sysfs_hprint(io_errors,		atomic_read(&dc->io_errors));
+	sysfs_printf(io_error_limit,	"%i", dc->error_limit);
+	sysfs_printf(io_disable,	"%i", dc->io_disable);
 	var_print(writeback_rate_update_seconds);
 	var_print(writeback_rate_i_term_inverse);
 	var_print(writeback_rate_p_term_inverse);
@@ -173,7 +183,7 @@ SHOW(__bch_cached_dev)
 	sysfs_hprint(dirty_data,
 		     bcache_dev_sectors_dirty(&dc->disk) << 9);
 
-	sysfs_hprint(stripe_size,	dc->disk.stripe_size << 9);
+	sysfs_hprint(stripe_size,	 ((uint64_t)dc->disk.stripe_size) << 9);
 	var_printf(partial_stripes_expensive,	"%u");
 
 	var_hprint(sequential_cutoff);
@@ -224,6 +234,14 @@ STORE(__cached_dev)
 	d_strtoul(writeback_rate_i_term_inverse);
 	d_strtoul_nonzero(writeback_rate_p_term_inverse);
 
+	sysfs_strtoul_clamp(io_error_limit, dc->error_limit, 0, INT_MAX);
+
+	if (attr == &sysfs_io_disable) {
+		int v = strtoul_or_return(buf);
+
+		dc->io_disable = v ? 1 : 0;
+	}
+
 	d_strtoi_h(sequential_cutoff);
 	d_strtoi_h(readahead);
 
@@ -246,6 +264,15 @@ STORE(__cached_dev)
 		}
 	}
 
+	if (attr == &sysfs_stop_when_cache_set_failed) {
+		v = bch_read_string_list(buf, bch_stop_on_failure_modes + 1);
+
+		if (v < 0)
+			return v;
+
+		dc->stop_when_cache_set_failed = v;
+	}
+
 	if (attr == &sysfs_label) {
 		if (size > SB_LABEL_SIZE)
 			return -EINVAL;
@@ -309,7 +336,8 @@ STORE(bch_cached_dev)
 		bch_writeback_queue(dc);
 
 	if (attr == &sysfs_writeback_percent)
-		schedule_delayed_work(&dc->writeback_rate_update,
+		if (!test_and_set_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags))
+			schedule_delayed_work(&dc->writeback_rate_update,
 				      dc->writeback_rate_update_seconds * HZ);
 
 	mutex_unlock(&bch_register_lock);
@@ -324,6 +352,7 @@ static struct attribute *bch_cached_dev_files[] = {
 	&sysfs_data_csum,
 #endif
 	&sysfs_cache_mode,
+	&sysfs_stop_when_cache_set_failed,
 	&sysfs_writeback_metadata,
 	&sysfs_writeback_running,
 	&sysfs_writeback_delay,
@@ -333,6 +362,9 @@ static struct attribute *bch_cached_dev_files[] = {
 	&sysfs_writeback_rate_i_term_inverse,
 	&sysfs_writeback_rate_p_term_inverse,
 	&sysfs_writeback_rate_debug,
+	&sysfs_errors,
+	&sysfs_io_error_limit,
+	&sysfs_io_disable,
 	&sysfs_dirty_data,
 	&sysfs_stripe_size,
 	&sysfs_partial_stripes_expensive,
@@ -590,6 +622,8 @@ SHOW(__bch_cache_set)
 	sysfs_printf(gc_always_rewrite,		"%i", c->gc_always_rewrite);
 	sysfs_printf(btree_shrinker_disabled,	"%i", c->shrinker_disabled);
 	sysfs_printf(copy_gc_enabled,		"%i", c->copy_gc_enabled);
+	sysfs_printf(io_disable,		"%i",
+		     test_bit(CACHE_SET_IO_DISABLE, &c->flags));
 
 	if (attr == &sysfs_bset_tree_stats)
 		return bch_bset_print_stats(c, buf);
@@ -679,6 +713,20 @@ STORE(__bch_cache_set)
 	if (attr == &sysfs_io_error_halflife)
 		c->error_decay = strtoul_or_return(buf) / 88;
 
+	if (attr == &sysfs_io_disable) {
+		int v = strtoul_or_return(buf);
+
+		if (v) {
+			if (test_and_set_bit(CACHE_SET_IO_DISABLE,
+					     &c->flags))
+				pr_warn("CACHE_SET_IO_DISABLE already set");
+		} else {
+			if (!test_and_clear_bit(CACHE_SET_IO_DISABLE,
+						&c->flags))
+				pr_warn("CACHE_SET_IO_DISABLE already cleared");
+		}
+	}
+
 	sysfs_strtoul(journal_delay_ms,		c->journal_delay_ms);
 	sysfs_strtoul(verify,			c->verify);
 	sysfs_strtoul(key_merging_disabled,	c->key_merging_disabled);
@@ -764,6 +812,7 @@ static struct attribute *bch_cache_set_internal_files[] = {
 	&sysfs_gc_always_rewrite,
 	&sysfs_btree_shrinker_disabled,
 	&sysfs_copy_gc_enabled,
+	&sysfs_io_disable,
 	NULL
 };
 KTYPE(bch_cache_set_internal);
diff --git a/drivers/md/bcache/util.c b/drivers/md/bcache/util.c
index a23cd6a14b74..74febd5230df 100644
--- a/drivers/md/bcache/util.c
+++ b/drivers/md/bcache/util.c
@@ -32,20 +32,27 @@ int bch_ ## name ## _h(const char *cp, type *res)		\
 	case 'y':						\
 	case 'z':						\
 		u++;						\
+		/* fall through */				\
 	case 'e':						\
 		u++;						\
+		/* fall through */				\
 	case 'p':						\
 		u++;						\
+		/* fall through */				\
 	case 't':						\
 		u++;						\
+		/* fall through */				\
 	case 'g':						\
 		u++;						\
+		/* fall through */				\
 	case 'm':						\
 		u++;						\
+		/* fall through */				\
 	case 'k':						\
 		u++;						\
 		if (e++ == cp)					\
 			return -EINVAL;				\
+		/* fall through */				\
 	case '\n':						\
 	case '\0':						\
 		if (*e == '\n')					\
@@ -75,10 +82,9 @@ STRTO_H(strtoll, long long)
 STRTO_H(strtoull, unsigned long long)
 
 /**
- * bch_hprint() - formats @v to human readable string for sysfs.
- *
- * @v - signed 64 bit integer
- * @buf - the (at least 8 byte) buffer to format the result into.
+ * bch_hprint - formats @v to human readable string for sysfs.
+ * @buf: the (at least 8 byte) buffer to format the result into.
+ * @v: signed 64 bit integer
  *
  * Returns the number of bytes used by format.
  */
@@ -218,13 +224,12 @@ void bch_time_stats_update(struct time_stats *stats, uint64_t start_time)
 }
 
 /**
- * bch_next_delay() - increment @d by the amount of work done, and return how
- * long to delay until the next time to do some work.
- *
- * @d - the struct bch_ratelimit to update
- * @done - the amount of work done, in arbitrary units
+ * bch_next_delay() - update ratelimiting statistics and calculate next delay
+ * @d: the struct bch_ratelimit to update
+ * @done: the amount of work done, in arbitrary units
  *
- * Returns the amount of time to delay by, in jiffies
+ * Increment @d by the amount of work done, and return how long to delay in
+ * jiffies until the next time to do some work.
  */
 uint64_t bch_next_delay(struct bch_ratelimit *d, uint64_t done)
 {
diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h
index a6763db7f061..268024529edd 100644
--- a/drivers/md/bcache/util.h
+++ b/drivers/md/bcache/util.h
@@ -567,12 +567,6 @@ static inline sector_t bdev_sectors(struct block_device *bdev)
 	return bdev->bd_inode->i_size >> 9;
 }
 
-#define closure_bio_submit(bio, cl)					\
-do {									\
-	closure_get(cl);						\
-	generic_make_request(bio);					\
-} while (0)
-
 uint64_t bch_crc64_update(uint64_t, const void *, size_t);
 uint64_t bch_crc64(const void *, size_t);
 
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index f1d2fc15abcc..4a9547cdcdc5 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -114,6 +114,27 @@ static void update_writeback_rate(struct work_struct *work)
 	struct cached_dev *dc = container_of(to_delayed_work(work),
 					     struct cached_dev,
 					     writeback_rate_update);
+	struct cache_set *c = dc->disk.c;
+
+	/*
+	 * should check BCACHE_DEV_RATE_DW_RUNNING before calling
+	 * cancel_delayed_work_sync().
+	 */
+	set_bit(BCACHE_DEV_RATE_DW_RUNNING, &dc->disk.flags);
+	/* paired with where BCACHE_DEV_RATE_DW_RUNNING is tested */
+	smp_mb();
+
+	/*
+	 * CACHE_SET_IO_DISABLE might be set via sysfs interface,
+	 * check it here too.
+	 */
+	if (!test_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags) ||
+	    test_bit(CACHE_SET_IO_DISABLE, &c->flags)) {
+		clear_bit(BCACHE_DEV_RATE_DW_RUNNING, &dc->disk.flags);
+		/* paired with where BCACHE_DEV_RATE_DW_RUNNING is tested */
+		smp_mb();
+		return;
+	}
 
 	down_read(&dc->writeback_lock);
 
@@ -123,8 +144,23 @@ static void update_writeback_rate(struct work_struct *work)
 
 	up_read(&dc->writeback_lock);
 
-	schedule_delayed_work(&dc->writeback_rate_update,
+	/*
+	 * CACHE_SET_IO_DISABLE might be set via sysfs interface,
+	 * check it here too.
+	 */
+	if (test_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags) &&
+	    !test_bit(CACHE_SET_IO_DISABLE, &c->flags)) {
+		schedule_delayed_work(&dc->writeback_rate_update,
 			      dc->writeback_rate_update_seconds * HZ);
+	}
+
+	/*
+	 * should check BCACHE_DEV_RATE_DW_RUNNING before calling
+	 * cancel_delayed_work_sync().
+	 */
+	clear_bit(BCACHE_DEV_RATE_DW_RUNNING, &dc->disk.flags);
+	/* paired with where BCACHE_DEV_RATE_DW_RUNNING is tested */
+	smp_mb();
 }
 
 static unsigned writeback_delay(struct cached_dev *dc, unsigned sectors)
@@ -253,7 +289,8 @@ static void write_dirty(struct closure *cl)
 		bio_set_dev(&io->bio, io->dc->bdev);
 		io->bio.bi_end_io	= dirty_endio;
 
-		closure_bio_submit(&io->bio, cl);
+		/* I/O request sent to backing device */
+		closure_bio_submit(io->dc->disk.c, &io->bio, cl);
 	}
 
 	atomic_set(&dc->writeback_sequence_next, next_sequence);
@@ -279,7 +316,7 @@ static void read_dirty_submit(struct closure *cl)
 {
 	struct dirty_io *io = container_of(cl, struct dirty_io, cl);
 
-	closure_bio_submit(&io->bio, cl);
+	closure_bio_submit(io->dc->disk.c, &io->bio, cl);
 
 	continue_at(cl, write_dirty, io->dc->writeback_write_wq);
 }
@@ -305,7 +342,9 @@ static void read_dirty(struct cached_dev *dc)
 
 	next = bch_keybuf_next(&dc->writeback_keys);
 
-	while (!kthread_should_stop() && next) {
+	while (!kthread_should_stop() &&
+	       !test_bit(CACHE_SET_IO_DISABLE, &dc->disk.c->flags) &&
+	       next) {
 		size = 0;
 		nk = 0;
 
@@ -402,7 +441,9 @@ static void read_dirty(struct cached_dev *dc)
 			}
 		}
 
-		while (!kthread_should_stop() && delay) {
+		while (!kthread_should_stop() &&
+		       !test_bit(CACHE_SET_IO_DISABLE, &dc->disk.c->flags) &&
+		       delay) {
 			schedule_timeout_interruptible(delay);
 			delay = writeback_delay(dc, 0);
 		}
@@ -558,21 +599,30 @@ static bool refill_dirty(struct cached_dev *dc)
 static int bch_writeback_thread(void *arg)
 {
 	struct cached_dev *dc = arg;
+	struct cache_set *c = dc->disk.c;
 	bool searched_full_index;
 
 	bch_ratelimit_reset(&dc->writeback_rate);
 
-	while (!kthread_should_stop()) {
+	while (!kthread_should_stop() &&
+	       !test_bit(CACHE_SET_IO_DISABLE, &c->flags)) {
 		down_write(&dc->writeback_lock);
 		set_current_state(TASK_INTERRUPTIBLE);
-		if (!atomic_read(&dc->has_dirty) ||
-		    (!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) &&
-		     !dc->writeback_running)) {
+		/*
+		 * If the bache device is detaching, skip here and continue
+		 * to perform writeback. Otherwise, if no dirty data on cache,
+		 * or there is dirty data on cache but writeback is disabled,
+		 * the writeback thread should sleep here and wait for others
+		 * to wake up it.
+		 */
+		if (!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) &&
+		    (!atomic_read(&dc->has_dirty) || !dc->writeback_running)) {
 			up_write(&dc->writeback_lock);
 
-			if (kthread_should_stop()) {
+			if (kthread_should_stop() ||
+			    test_bit(CACHE_SET_IO_DISABLE, &c->flags)) {
 				set_current_state(TASK_RUNNING);
-				return 0;
+				break;
 			}
 
 			schedule();
@@ -585,9 +635,16 @@ static int bch_writeback_thread(void *arg)
 		if (searched_full_index &&
 		    RB_EMPTY_ROOT(&dc->writeback_keys.keys)) {
 			atomic_set(&dc->has_dirty, 0);
-			cached_dev_put(dc);
 			SET_BDEV_STATE(&dc->sb, BDEV_STATE_CLEAN);
 			bch_write_bdev_super(dc, NULL);
+			/*
+			 * If bcache device is detaching via sysfs interface,
+			 * writeback thread should stop after there is no dirty
+			 * data on cache. BCACHE_DEV_DETACHING flag is set in
+			 * bch_cached_dev_detach().
+			 */
+			if (test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags))
+				break;
 		}
 
 		up_write(&dc->writeback_lock);
@@ -599,6 +656,7 @@ static int bch_writeback_thread(void *arg)
 
 			while (delay &&
 			       !kthread_should_stop() &&
+			       !test_bit(CACHE_SET_IO_DISABLE, &c->flags) &&
 			       !test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags))
 				delay = schedule_timeout_interruptible(delay);
 
@@ -606,6 +664,9 @@ static int bch_writeback_thread(void *arg)
 		}
 	}
 
+	cached_dev_put(dc);
+	wait_for_kthread_stop();
+
 	return 0;
 }
 
@@ -659,6 +720,7 @@ void bch_cached_dev_writeback_init(struct cached_dev *dc)
 	dc->writeback_rate_p_term_inverse = 40;
 	dc->writeback_rate_i_term_inverse = 10000;
 
+	WARN_ON(test_and_clear_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags));
 	INIT_DELAYED_WORK(&dc->writeback_rate_update, update_writeback_rate);
 }
 
@@ -669,11 +731,15 @@ int bch_cached_dev_writeback_start(struct cached_dev *dc)
 	if (!dc->writeback_write_wq)
 		return -ENOMEM;
 
+	cached_dev_get(dc);
 	dc->writeback_thread = kthread_create(bch_writeback_thread, dc,
 					      "bcache_writeback");
-	if (IS_ERR(dc->writeback_thread))
+	if (IS_ERR(dc->writeback_thread)) {
+		cached_dev_put(dc);
 		return PTR_ERR(dc->writeback_thread);
+	}
 
+	WARN_ON(test_and_set_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags));
 	schedule_delayed_work(&dc->writeback_rate_update,
 			      dc->writeback_rate_update_seconds * HZ);
 
diff --git a/drivers/md/bcache/writeback.h b/drivers/md/bcache/writeback.h
index 587b25599856..610fb01de629 100644
--- a/drivers/md/bcache/writeback.h
+++ b/drivers/md/bcache/writeback.h
@@ -39,7 +39,7 @@ static inline uint64_t  bcache_flash_devs_sectors_dirty(struct cache_set *c)
 
 		if (!d || !UUID_FLASH_ONLY(&c->uuids[i]))
 			continue;
-	   ret += bcache_dev_sectors_dirty(d);
+		ret += bcache_dev_sectors_dirty(d);
 	}
 
 	mutex_unlock(&bch_register_lock);
@@ -105,8 +105,6 @@ static inline void bch_writeback_add(struct cached_dev *dc)
 {
 	if (!atomic_read(&dc->has_dirty) &&
 	    !atomic_xchg(&dc->has_dirty, 1)) {
-		refcount_inc(&dc->count);
-
 		if (BDEV_STATE(&dc->sb) != BDEV_STATE_DIRTY) {
 			SET_BDEV_STATE(&dc->sb, BDEV_STATE_DIRTY);
 			/* XXX: should do this synchronously */