Merge branch 'for-linus' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 - fix for a memory leak on certain unplug events
 - a collection of bcache fixes from Kent and Nicolas
 - a few null_blk fixes and updates form Matias
 - a marking of static of functions in the stec pci-e driver

* 'for-linus' of git://git.kernel.dk/linux-block:
  null_blk: support submit_queues on use_per_node_hctx
  null_blk: set use_per_node_hctx param to false
  null_blk: corrections to documentation
  null_blk: warning on ignored submit_queues param
  null_blk: refactor init and init errors code paths
  null_blk: documentation
  null_blk: mem garbage on NUMA systems during init
  drivers: block: Mark the functions as static in skd_main.c
  bcache: New writeback PD controller
  bcache: bugfix for race between moving_gc and bucket_invalidate
  bcache: fix for gc and writeback race
  bcache: bugfix - moving_gc now moves only correct buckets
  bcache: fix for gc crashing when no sectors are used
  bcache: Fix heap_peek() macro
  bcache: Fix for can_attach_cache()
  bcache: Fix dirty_data accounting
  bcache: Use uninterruptible sleep in writeback
  bcache: kthread don't set writeback task to INTERUPTIBLE
  block: fix memory leaks on unplugging block device
  bcache: fix sparse non static symbol warning

9 files changed, 111 insertions, 66 deletions

diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c
index 2b46bf1d7e40..4c9852d92b0a 100644
--- a/drivers/md/bcache/alloc.c
+++ b/drivers/md/bcache/alloc.c
@@ -421,9 +421,11 @@ out:
 
 	if (watermark <= WATERMARK_METADATA) {
 		SET_GC_MARK(b, GC_MARK_METADATA);
+		SET_GC_MOVE(b, 0);
 		b->prio = BTREE_PRIO;
 	} else {
 		SET_GC_MARK(b, GC_MARK_RECLAIMABLE);
+		SET_GC_MOVE(b, 0);
 		b->prio = INITIAL_PRIO;
 	}
 
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index 4beb55a0ff30..754f43177483 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -197,7 +197,7 @@ struct bucket {
 	uint8_t		disk_gen;
 	uint8_t		last_gc; /* Most out of date gen in the btree */
 	uint8_t		gc_gen;
-	uint16_t	gc_mark;
+	uint16_t	gc_mark; /* Bitfield used by GC. See below for field */
 };
 
 /*
@@ -209,7 +209,8 @@ BITMASK(GC_MARK,	 struct bucket, gc_mark, 0, 2);
 #define GC_MARK_RECLAIMABLE	0
 #define GC_MARK_DIRTY		1
 #define GC_MARK_METADATA	2
-BITMASK(GC_SECTORS_USED, struct bucket, gc_mark, 2, 14);
+BITMASK(GC_SECTORS_USED, struct bucket, gc_mark, 2, 13);
+BITMASK(GC_MOVE, struct bucket, gc_mark, 15, 1);
 
 #include "journal.h"
 #include "stats.h"
@@ -372,14 +373,14 @@ struct cached_dev {
 	unsigned char		writeback_percent;
 	unsigned		writeback_delay;
 
-	int			writeback_rate_change;
-	int64_t			writeback_rate_derivative;
 	uint64_t		writeback_rate_target;
+	int64_t			writeback_rate_proportional;
+	int64_t			writeback_rate_derivative;
+	int64_t			writeback_rate_change;
 
 	unsigned		writeback_rate_update_seconds;
 	unsigned		writeback_rate_d_term;
 	unsigned		writeback_rate_p_term_inverse;
-	unsigned		writeback_rate_d_smooth;
 };
 
 enum alloc_watermarks {
@@ -445,7 +446,6 @@ struct cache {
 	 * call prio_write() to keep gens from wrapping.
 	 */
 	uint8_t			need_save_prio;
-	unsigned		gc_move_threshold;
 
 	/*
 	 * If nonzero, we know we aren't going to find any buckets to invalidate
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index 5e2765aadce1..31bb53fcc67a 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -1561,6 +1561,28 @@ size_t bch_btree_gc_finish(struct cache_set *c)
 		SET_GC_MARK(PTR_BUCKET(c, &c->uuid_bucket, i),
 			    GC_MARK_METADATA);
 
+	/* don't reclaim buckets to which writeback keys point */
+	rcu_read_lock();
+	for (i = 0; i < c->nr_uuids; i++) {
+		struct bcache_device *d = c->devices[i];
+		struct cached_dev *dc;
+		struct keybuf_key *w, *n;
+		unsigned j;
+
+		if (!d || UUID_FLASH_ONLY(&c->uuids[i]))
+			continue;
+		dc = container_of(d, struct cached_dev, disk);
+
+		spin_lock(&dc->writeback_keys.lock);
+		rbtree_postorder_for_each_entry_safe(w, n,
+					&dc->writeback_keys.keys, node)
+			for (j = 0; j < KEY_PTRS(&w->key); j++)
+				SET_GC_MARK(PTR_BUCKET(c, &w->key, j),
+					    GC_MARK_DIRTY);
+		spin_unlock(&dc->writeback_keys.lock);
+	}
+	rcu_read_unlock();
+
 	for_each_cache(ca, c, i) {
 		uint64_t *i;
 
@@ -1817,7 +1839,8 @@ static bool fix_overlapping_extents(struct btree *b, struct bkey *insert,
 			if (KEY_START(k) > KEY_START(insert) + sectors_found)
 				goto check_failed;
 
-			if (KEY_PTRS(replace_key) != KEY_PTRS(k))
+			if (KEY_PTRS(k) != KEY_PTRS(replace_key) ||
+			    KEY_DIRTY(k) != KEY_DIRTY(replace_key))
 				goto check_failed;
 
 			/* skip past gen */
@@ -2217,7 +2240,7 @@ struct btree_insert_op {
 	struct bkey	*replace_key;
 };
 
-int btree_insert_fn(struct btree_op *b_op, struct btree *b)
+static int btree_insert_fn(struct btree_op *b_op, struct btree *b)
 {
 	struct btree_insert_op *op = container_of(b_op,
 					struct btree_insert_op, op);
diff --git a/drivers/md/bcache/movinggc.c b/drivers/md/bcache/movinggc.c
index 7c1275e66025..f2f0998c4a91 100644
--- a/drivers/md/bcache/movinggc.c
+++ b/drivers/md/bcache/movinggc.c
@@ -25,10 +25,9 @@ static bool moving_pred(struct keybuf *buf, struct bkey *k)
 	unsigned i;
 
 	for (i = 0; i < KEY_PTRS(k); i++) {
-		struct cache *ca = PTR_CACHE(c, k, i);
 		struct bucket *g = PTR_BUCKET(c, k, i);
 
-		if (GC_SECTORS_USED(g) < ca->gc_move_threshold)
+		if (GC_MOVE(g))
 			return true;
 	}
 
@@ -65,11 +64,16 @@ static void write_moving_finish(struct closure *cl)
 
 static void read_moving_endio(struct bio *bio, int error)
 {
+	struct bbio *b = container_of(bio, struct bbio, bio);
 	struct moving_io *io = container_of(bio->bi_private,
 					    struct moving_io, cl);
 
 	if (error)
 		io->op.error = error;
+	else if (!KEY_DIRTY(&b->key) &&
+		 ptr_stale(io->op.c, &b->key, 0)) {
+		io->op.error = -EINTR;
+	}
 
 	bch_bbio_endio(io->op.c, bio, error, "reading data to move");
 }
@@ -141,6 +145,11 @@ static void read_moving(struct cache_set *c)
 		if (!w)
 			break;
 
+		if (ptr_stale(c, &w->key, 0)) {
+			bch_keybuf_del(&c->moving_gc_keys, w);
+			continue;
+		}
+
 		io = kzalloc(sizeof(struct moving_io) + sizeof(struct bio_vec)
 			     * DIV_ROUND_UP(KEY_SIZE(&w->key), PAGE_SECTORS),
 			     GFP_KERNEL);
@@ -184,7 +193,8 @@ static bool bucket_cmp(struct bucket *l, struct bucket *r)
 
 static unsigned bucket_heap_top(struct cache *ca)
 {
-	return GC_SECTORS_USED(heap_peek(&ca->heap));
+	struct bucket *b;
+	return (b = heap_peek(&ca->heap)) ? GC_SECTORS_USED(b) : 0;
 }
 
 void bch_moving_gc(struct cache_set *c)
@@ -226,9 +236,8 @@ void bch_moving_gc(struct cache_set *c)
 			sectors_to_move -= GC_SECTORS_USED(b);
 		}
 
-		ca->gc_move_threshold = bucket_heap_top(ca);
-
-		pr_debug("threshold %u", ca->gc_move_threshold);
+		while (heap_pop(&ca->heap, b, bucket_cmp))
+			SET_GC_MOVE(b, 1);
 	}
 
 	mutex_unlock(&c->bucket_lock);
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index dec15cd2d797..c57bfa071a57 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -1676,7 +1676,7 @@ err:
 static bool can_attach_cache(struct cache *ca, struct cache_set *c)
 {
 	return ca->sb.block_size	== c->sb.block_size &&
-		ca->sb.bucket_size	== c->sb.block_size &&
+		ca->sb.bucket_size	== c->sb.bucket_size &&
 		ca->sb.nr_in_set	== c->sb.nr_in_set;
 }
 
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c
index 80d4c2bee18a..a1f85612f0b3 100644
--- a/drivers/md/bcache/sysfs.c
+++ b/drivers/md/bcache/sysfs.c
@@ -83,7 +83,6 @@ rw_attribute(writeback_rate);
 rw_attribute(writeback_rate_update_seconds);
 rw_attribute(writeback_rate_d_term);
 rw_attribute(writeback_rate_p_term_inverse);
-rw_attribute(writeback_rate_d_smooth);
 read_attribute(writeback_rate_debug);
 
 read_attribute(stripe_size);
@@ -129,31 +128,41 @@ SHOW(__bch_cached_dev)
 	var_printf(writeback_running,	"%i");
 	var_print(writeback_delay);
 	var_print(writeback_percent);
-	sysfs_print(writeback_rate,	dc->writeback_rate.rate);
+	sysfs_hprint(writeback_rate,	dc->writeback_rate.rate << 9);
 
 	var_print(writeback_rate_update_seconds);
 	var_print(writeback_rate_d_term);
 	var_print(writeback_rate_p_term_inverse);
-	var_print(writeback_rate_d_smooth);
 
 	if (attr == &sysfs_writeback_rate_debug) {
+		char rate[20];
 		char dirty[20];
-		char derivative[20];
 		char target[20];
-		bch_hprint(dirty,
-			   bcache_dev_sectors_dirty(&dc->disk) << 9);
-		bch_hprint(derivative,	dc->writeback_rate_derivative << 9);
+		char proportional[20];
+		char derivative[20];
+		char change[20];
+		s64 next_io;
+
+		bch_hprint(rate,	dc->writeback_rate.rate << 9);
+		bch_hprint(dirty,	bcache_dev_sectors_dirty(&dc->disk) << 9);
 		bch_hprint(target,	dc->writeback_rate_target << 9);
+		bch_hprint(proportional,dc->writeback_rate_proportional << 9);
+		bch_hprint(derivative,	dc->writeback_rate_derivative << 9);
+		bch_hprint(change,	dc->writeback_rate_change << 9);
+
+		next_io = div64_s64(dc->writeback_rate.next - local_clock(),
+				    NSEC_PER_MSEC);
 
 		return sprintf(buf,
-			       "rate:\t\t%u\n"
-			       "change:\t\t%i\n"
+			       "rate:\t\t%s/sec\n"
 			       "dirty:\t\t%s\n"
+			       "target:\t\t%s\n"
+			       "proportional:\t%s\n"
 			       "derivative:\t%s\n"
-			       "target:\t\t%s\n",
-			       dc->writeback_rate.rate,
-			       dc->writeback_rate_change,
-			       dirty, derivative, target);
+			       "change:\t\t%s/sec\n"
+			       "next io:\t%llims\n",
+			       rate, dirty, target, proportional,
+			       derivative, change, next_io);
 	}
 
 	sysfs_hprint(dirty_data,
@@ -189,6 +198,7 @@ STORE(__cached_dev)
 	struct kobj_uevent_env *env;
 
 #define d_strtoul(var)		sysfs_strtoul(var, dc->var)
+#define d_strtoul_nonzero(var)	sysfs_strtoul_clamp(var, dc->var, 1, INT_MAX)
 #define d_strtoi_h(var)		sysfs_hatoi(var, dc->var)
 
 	sysfs_strtoul(data_csum,	dc->disk.data_csum);
@@ -197,16 +207,15 @@ STORE(__cached_dev)
 	d_strtoul(writeback_metadata);
 	d_strtoul(writeback_running);
 	d_strtoul(writeback_delay);
-	sysfs_strtoul_clamp(writeback_rate,
-			    dc->writeback_rate.rate, 1, 1000000);
+
 	sysfs_strtoul_clamp(writeback_percent, dc->writeback_percent, 0, 40);
 
-	d_strtoul(writeback_rate_update_seconds);
+	sysfs_strtoul_clamp(writeback_rate,
+			    dc->writeback_rate.rate, 1, INT_MAX);
+
+	d_strtoul_nonzero(writeback_rate_update_seconds);
 	d_strtoul(writeback_rate_d_term);
-	d_strtoul(writeback_rate_p_term_inverse);
-	sysfs_strtoul_clamp(writeback_rate_p_term_inverse,
-			    dc->writeback_rate_p_term_inverse, 1, INT_MAX);
-	d_strtoul(writeback_rate_d_smooth);
+	d_strtoul_nonzero(writeback_rate_p_term_inverse);
 
 	d_strtoi_h(sequential_cutoff);
 	d_strtoi_h(readahead);
@@ -313,7 +322,6 @@ static struct attribute *bch_cached_dev_files[] = {
 	&sysfs_writeback_rate_update_seconds,
 	&sysfs_writeback_rate_d_term,
 	&sysfs_writeback_rate_p_term_inverse,
-	&sysfs_writeback_rate_d_smooth,
 	&sysfs_writeback_rate_debug,
 	&sysfs_dirty_data,
 	&sysfs_stripe_size,
diff --git a/drivers/md/bcache/util.c b/drivers/md/bcache/util.c
index 462214eeacbe..bb37618e7664 100644
--- a/drivers/md/bcache/util.c
+++ b/drivers/md/bcache/util.c
@@ -209,7 +209,13 @@ uint64_t bch_next_delay(struct bch_ratelimit *d, uint64_t done)
 {
 	uint64_t now = local_clock();
 
-	d->next += div_u64(done, d->rate);
+	d->next += div_u64(done * NSEC_PER_SEC, d->rate);
+
+	if (time_before64(now + NSEC_PER_SEC, d->next))
+		d->next = now + NSEC_PER_SEC;
+
+	if (time_after64(now - NSEC_PER_SEC * 2, d->next))
+		d->next = now - NSEC_PER_SEC * 2;
 
 	return time_after64(d->next, now)
 		? div_u64(d->next - now, NSEC_PER_SEC / HZ)
diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h
index 362c4b3f8b4a..1030c6020e98 100644
--- a/drivers/md/bcache/util.h
+++ b/drivers/md/bcache/util.h
@@ -110,7 +110,7 @@ do {									\
 	_r;								\
 })
 
-#define heap_peek(h)	((h)->size ? (h)->data[0] : NULL)
+#define heap_peek(h)	((h)->used ? (h)->data[0] : NULL)
 
 #define heap_full(h)	((h)->used == (h)->size)
 
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index 99053b1251be..6c44fe059c27 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -30,38 +30,40 @@ static void __update_writeback_rate(struct cached_dev *dc)
 
 	/* PD controller */
 
-	int change = 0;
-	int64_t error;
 	int64_t dirty = bcache_dev_sectors_dirty(&dc->disk);
 	int64_t derivative = dirty - dc->disk.sectors_dirty_last;
+	int64_t proportional = dirty - target;
+	int64_t change;
 
 	dc->disk.sectors_dirty_last = dirty;
 
-	derivative *= dc->writeback_rate_d_term;
-	derivative = clamp(derivative, -dirty, dirty);
+	/* Scale to sectors per second */
 
-	derivative = ewma_add(dc->disk.sectors_dirty_derivative, derivative,
-			      dc->writeback_rate_d_smooth, 0);
+	proportional *= dc->writeback_rate_update_seconds;
+	proportional = div_s64(proportional, dc->writeback_rate_p_term_inverse);
 
-	/* Avoid divide by zero */
-	if (!target)
-		goto out;
+	derivative = div_s64(derivative, dc->writeback_rate_update_seconds);
 
-	error = div64_s64((dirty + derivative - target) << 8, target);
+	derivative = ewma_add(dc->disk.sectors_dirty_derivative, derivative,
+			      (dc->writeback_rate_d_term /
+			       dc->writeback_rate_update_seconds) ?: 1, 0);
+
+	derivative *= dc->writeback_rate_d_term;
+	derivative = div_s64(derivative, dc->writeback_rate_p_term_inverse);
 
-	change = div_s64((dc->writeback_rate.rate * error) >> 8,
-			 dc->writeback_rate_p_term_inverse);
+	change = proportional + derivative;
 
 	/* Don't increase writeback rate if the device isn't keeping up */
 	if (change > 0 &&
 	    time_after64(local_clock(),
-			 dc->writeback_rate.next + 10 * NSEC_PER_MSEC))
+			 dc->writeback_rate.next + NSEC_PER_MSEC))
 		change = 0;
 
 	dc->writeback_rate.rate =
-		clamp_t(int64_t, dc->writeback_rate.rate + change,
+		clamp_t(int64_t, (int64_t) dc->writeback_rate.rate + change,
 			1, NSEC_PER_MSEC);
-out:
+
+	dc->writeback_rate_proportional = proportional;
 	dc->writeback_rate_derivative = derivative;
 	dc->writeback_rate_change = change;
 	dc->writeback_rate_target = target;
@@ -87,15 +89,11 @@ static void update_writeback_rate(struct work_struct *work)
 
 static unsigned writeback_delay(struct cached_dev *dc, unsigned sectors)
 {
-	uint64_t ret;
-
 	if (test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) ||
 	    !dc->writeback_percent)
 		return 0;
 
-	ret = bch_next_delay(&dc->writeback_rate, sectors * 10000000ULL);
-
-	return min_t(uint64_t, ret, HZ);
+	return bch_next_delay(&dc->writeback_rate, sectors);
 }
 
 struct dirty_io {
@@ -241,7 +239,7 @@ static void read_dirty(struct cached_dev *dc)
 		if (KEY_START(&w->key) != dc->last_read ||
 		    jiffies_to_msecs(delay) > 50)
 			while (!kthread_should_stop() && delay)
-				delay = schedule_timeout_interruptible(delay);
+				delay = schedule_timeout_uninterruptible(delay);
 
 		dc->last_read	= KEY_OFFSET(&w->key);
 
@@ -438,7 +436,7 @@ static int bch_writeback_thread(void *arg)
 			while (delay &&
 			       !kthread_should_stop() &&
 			       !test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags))
-				delay = schedule_timeout_interruptible(delay);
+				delay = schedule_timeout_uninterruptible(delay);
 		}
 	}
 
@@ -476,6 +474,8 @@ void bch_sectors_dirty_init(struct cached_dev *dc)
 
 	bch_btree_map_keys(&op.op, dc->disk.c, &KEY(op.inode, 0, 0),
 			   sectors_dirty_init_fn, 0);
+
+	dc->disk.sectors_dirty_last = bcache_dev_sectors_dirty(&dc->disk);
 }
 
 int bch_cached_dev_writeback_init(struct cached_dev *dc)
@@ -490,18 +490,15 @@ int bch_cached_dev_writeback_init(struct cached_dev *dc)
 	dc->writeback_delay		= 30;
 	dc->writeback_rate.rate		= 1024;
 
-	dc->writeback_rate_update_seconds = 30;
-	dc->writeback_rate_d_term	= 16;
-	dc->writeback_rate_p_term_inverse = 64;
-	dc->writeback_rate_d_smooth	= 8;
+	dc->writeback_rate_update_seconds = 5;
+	dc->writeback_rate_d_term	= 30;
+	dc->writeback_rate_p_term_inverse = 6000;
 
 	dc->writeback_thread = kthread_create(bch_writeback_thread, dc,
 					      "bcache_writeback");
 	if (IS_ERR(dc->writeback_thread))
 		return PTR_ERR(dc->writeback_thread);
 
-	set_task_state(dc->writeback_thread, TASK_INTERRUPTIBLE);
-
 	INIT_DELAYED_WORK(&dc->writeback_rate_update, update_writeback_rate);
 	schedule_delayed_work(&dc->writeback_rate_update,
 			      dc->writeback_rate_update_seconds * HZ);