Merge branch 'bcache-for-upstream' of...

Merge branch 'bcache-for-upstream' of http://evilpiepirate.org/git/linux-bcache into for-3.10/drivers

Kent writes:

Hey Jens, this is everything I've got ready for 3.10 - there's _still_
one more bug I'm trying to track down.

Andrew - I've got patches that rip out the pkey() and pbtree() macros,
but they're somewhat tied up with some other nontrivial refactorings so
I think I'm going to wait a bit on those.

drivers/md/bcache/alloc.c

@@ -243,31 +243,37 @@ static void invalidate_buckets_lru(struct cache *ca)
 	ca->heap.used = 0;
 
 	for_each_bucket(b, ca) {
+		/*
+		 * If we fill up the unused list, if we then return before
+		 * adding anything to the free_inc list we'll skip writing
+		 * prios/gens and just go back to allocating from the unused
+		 * list:
+		 */
+		if (fifo_full(&ca->unused))
+			return;
+
 		if (!can_invalidate_bucket(ca, b))
 			continue;
 
-		if (!GC_SECTORS_USED(b)) {
-			if (!bch_bucket_add_unused(ca, b))
-				return;
-		} else {
-			if (!heap_full(&ca->heap))
-				heap_add(&ca->heap, b, bucket_max_cmp);
-			else if (bucket_max_cmp(b, heap_peek(&ca->heap))) {
-				ca->heap.data[0] = b;
-				heap_sift(&ca->heap, 0, bucket_max_cmp);
-			}
+		if (!GC_SECTORS_USED(b) &&
+		    bch_bucket_add_unused(ca, b))
+			continue;
+
+		if (!heap_full(&ca->heap))
+			heap_add(&ca->heap, b, bucket_max_cmp);
+		else if (bucket_max_cmp(b, heap_peek(&ca->heap))) {
+			ca->heap.data[0] = b;
+			heap_sift(&ca->heap, 0, bucket_max_cmp);
 		}
 	}
 
-	if (ca->heap.used * 2 < ca->heap.size)
-		bch_queue_gc(ca->set);
-
 	for (i = ca->heap.used / 2 - 1; i >= 0; --i)
 		heap_sift(&ca->heap, i, bucket_min_cmp);
 
 	while (!fifo_full(&ca->free_inc)) {
 		if (!heap_pop(&ca->heap, b, bucket_min_cmp)) {
-			/* We don't want to be calling invalidate_buckets()
+			/*
+			 * We don't want to be calling invalidate_buckets()
 			 * multiple times when it can't do anything
 			 */
 			ca->invalidate_needs_gc = 1;
@@ -343,15 +349,22 @@ static void invalidate_buckets(struct cache *ca)
 		invalidate_buckets_random(ca);
 		break;
 	}
+
+	pr_debug("free %zu/%zu free_inc %zu/%zu unused %zu/%zu",
+		 fifo_used(&ca->free), ca->free.size,
+		 fifo_used(&ca->free_inc), ca->free_inc.size,
+		 fifo_used(&ca->unused), ca->unused.size);
 }
 
 #define allocator_wait(ca, cond)					\
 do {									\
 	DEFINE_WAIT(__wait);						\
 									\
-	while (!(cond)) {						\
+	while (1) {							\
 		prepare_to_wait(&ca->set->alloc_wait,			\
 				&__wait, TASK_INTERRUPTIBLE);		\
+		if (cond)						\
+			break;						\
 									\
 		mutex_unlock(&(ca)->set->bucket_lock);			\
 		if (test_bit(CACHE_SET_STOPPING_2, &ca->set->flags)) {	\
@@ -360,7 +373,6 @@ do {									\
 		}							\
 									\
 		schedule();						\
-		__set_current_state(TASK_RUNNING);			\
 		mutex_lock(&(ca)->set->bucket_lock);			\
 	}								\
 									\
@@ -374,6 +386,11 @@ void bch_allocator_thread(struct closure *cl)
 	mutex_lock(&ca->set->bucket_lock);
 
 	while (1) {
+		/*
+		 * First, we pull buckets off of the unused and free_inc lists,
+		 * possibly issue discards to them, then we add the bucket to
+		 * the free list:
+		 */
 		while (1) {
 			long bucket;
 
@@ -398,17 +415,26 @@ void bch_allocator_thread(struct closure *cl)
 			}
 		}
 
-		allocator_wait(ca, ca->set->gc_mark_valid);
-		invalidate_buckets(ca);
+		/*
+		 * We've run out of free buckets, we need to find some buckets
+		 * we can invalidate. First, invalidate them in memory and add
+		 * them to the free_inc list:
+		 */
 
-		allocator_wait(ca, !atomic_read(&ca->set->prio_blocked) ||
-			       !CACHE_SYNC(&ca->set->sb));
+		allocator_wait(ca, ca->set->gc_mark_valid &&
+			       (ca->need_save_prio > 64 ||
+				!ca->invalidate_needs_gc));
+		invalidate_buckets(ca);
 
+		/*
+		 * Now, we write their new gens to disk so we can start writing
+		 * new stuff to them:
+		 */
+		allocator_wait(ca, !atomic_read(&ca->set->prio_blocked));
 		if (CACHE_SYNC(&ca->set->sb) &&
 		    (!fifo_empty(&ca->free_inc) ||
-		     ca->need_save_prio > 64)) {
+		     ca->need_save_prio > 64))
 			bch_prio_write(ca);
-		}
 	}
 }
 
@@ -475,7 +501,7 @@ void bch_bucket_free(struct cache_set *c, struct bkey *k)
 	for (i = 0; i < KEY_PTRS(k); i++) {
 		struct bucket *b = PTR_BUCKET(c, k, i);
 
-		SET_GC_MARK(b, 0);
+		SET_GC_MARK(b, GC_MARK_RECLAIMABLE);
 		SET_GC_SECTORS_USED(b, 0);
 		bch_bucket_add_unused(PTR_CACHE(c, k, i), b);
 	}

drivers/md/bcache/bcache.h

@@ -223,11 +223,17 @@ struct bkey {
 #define BKEY_PADDED(key)					\
 	union { struct bkey key; uint64_t key ## _pad[BKEY_PAD]; }
 
-/* Version 1: Backing device
+/* Version 0: Cache device
+ * Version 1: Backing device
  * Version 2: Seed pointer into btree node checksum
- * Version 3: New UUID format
+ * Version 3: Cache device with new UUID format
+ * Version 4: Backing device with data offset
  */
-#define BCACHE_SB_VERSION	3
+#define BCACHE_SB_VERSION_CDEV			0
+#define BCACHE_SB_VERSION_BDEV			1
+#define BCACHE_SB_VERSION_CDEV_WITH_UUID	3
+#define BCACHE_SB_VERSION_BDEV_WITH_OFFSET	4
+#define BCACHE_SB_MAX_VERSION			4
 
 #define SB_SECTOR		8
 #define SB_SIZE			4096
@@ -236,13 +242,12 @@ struct bkey {
 /* SB_JOURNAL_BUCKETS must be divisible by BITS_PER_LONG */
 #define MAX_CACHES_PER_SET	8
 
-#define BDEV_DATA_START		16	/* sectors */
+#define BDEV_DATA_START_DEFAULT	16	/* sectors */
 
 struct cache_sb {
 	uint64_t		csum;
 	uint64_t		offset;	/* sector where this sb was written */
 	uint64_t		version;
-#define CACHE_BACKING_DEV	1
 
 	uint8_t			magic[16];
 
@@ -257,12 +262,28 @@ struct cache_sb {
 	uint64_t		seq;
 	uint64_t		pad[8];
 
-	uint64_t		nbuckets;	/* device size */
-	uint16_t		block_size;	/* sectors */
-	uint16_t		bucket_size;	/* sectors */
+	union {
+	struct {
+		/* Cache devices */
+		uint64_t	nbuckets;	/* device size */
+
+		uint16_t	block_size;	/* sectors */
+		uint16_t	bucket_size;	/* sectors */
 
-	uint16_t		nr_in_set;
-	uint16_t		nr_this_dev;
+		uint16_t	nr_in_set;
+		uint16_t	nr_this_dev;
+	};
+	struct {
+		/* Backing devices */
+		uint64_t	data_offset;
+
+		/*
+		 * block_size from the cache device section is still used by
+		 * backing devices, so don't add anything here until we fix
+		 * things to not need it for backing devices anymore
+		 */
+	};
+	};
 
 	uint32_t		last_mount;	/* time_t */
 
@@ -861,6 +882,12 @@ static inline bool key_merging_disabled(struct cache_set *c)
 #endif
 }
 
+static inline bool SB_IS_BDEV(const struct cache_sb *sb)
+{
+	return sb->version == BCACHE_SB_VERSION_BDEV
+		|| sb->version == BCACHE_SB_VERSION_BDEV_WITH_OFFSET;
+}
+
 struct bbio {
 	unsigned		submit_time_us;
 	union {

drivers/md/bcache/btree.c

@@ -984,7 +984,7 @@ static void btree_node_free(struct btree *b, struct btree_op *op)
 
 	if (b->prio_blocked &&
 	    !atomic_sub_return(b->prio_blocked, &b->c->prio_blocked))
-		closure_wake_up(&b->c->bucket_wait);
+		wake_up(&b->c->alloc_wait);
 
 	b->prio_blocked = 0;
 
@@ -1548,7 +1548,6 @@ static void bch_btree_gc(struct closure *cl)
 
 	trace_bcache_gc_end(c->sb.set_uuid);
 	wake_up(&c->alloc_wait);
-	closure_wake_up(&c->bucket_wait);
 
 	continue_at(cl, bch_moving_gc, bch_gc_wq);
 }

drivers/md/bcache/io.c

@@ -38,6 +38,15 @@ static void bch_generic_make_request_hack(struct bio *bio)
 		bio = clone;
 	}
 
+	/*
+	 * Hack, since drivers that clone bios clone up to bi_max_vecs, but our
+	 * bios might have had more than that (before we split them per device
+	 * limitations).
+	 *
+	 * To be taken out once immutable bvec stuff is in.
+	 */
+	bio->bi_max_vecs = bio->bi_vcnt;
+
 	generic_make_request(bio);
 }
 
@@ -149,34 +158,32 @@ static unsigned bch_bio_max_sectors(struct bio *bio)
 {
 	unsigned ret = bio_sectors(bio);
 	struct request_queue *q = bdev_get_queue(bio->bi_bdev);
+	unsigned max_segments = min_t(unsigned, BIO_MAX_PAGES,
+				      queue_max_segments(q));
 	struct bio_vec *bv, *end = bio_iovec(bio) +
-		min_t(int, bio_segments(bio), queue_max_segments(q));
-
-	struct bvec_merge_data bvm = {
-		.bi_bdev	= bio->bi_bdev,
-		.bi_sector	= bio->bi_sector,
-		.bi_size	= 0,
-		.bi_rw		= bio->bi_rw,
-	};
+		min_t(int, bio_segments(bio), max_segments);
 
 	if (bio->bi_rw & REQ_DISCARD)
 		return min(ret, q->limits.max_discard_sectors);
 
-	if (bio_segments(bio) > queue_max_segments(q) ||
+	if (bio_segments(bio) > max_segments ||
 	    q->merge_bvec_fn) {
 		ret = 0;
 
 		for (bv = bio_iovec(bio); bv < end; bv++) {
+			struct bvec_merge_data bvm = {
+				.bi_bdev	= bio->bi_bdev,
+				.bi_sector	= bio->bi_sector,
+				.bi_size	= ret << 9,
+				.bi_rw		= bio->bi_rw,
+			};
+
 			if (q->merge_bvec_fn &&
 			    q->merge_bvec_fn(q, &bvm, bv) < (int) bv->bv_len)
 				break;
 
-			ret		+= bv->bv_len >> 9;
-			bvm.bi_size	+= bv->bv_len;
+			ret += bv->bv_len >> 9;
 		}
-
-		if (ret >= (BIO_MAX_PAGES * PAGE_SIZE) >> 9)
-			return (BIO_MAX_PAGES * PAGE_SIZE) >> 9;
 	}
 
 	ret = min(ret, queue_max_sectors(q));

drivers/md/bcache/request.c

@@ -1220,7 +1220,7 @@ static void cached_dev_make_request(struct request_queue *q, struct bio *bio)
 	part_stat_unlock();
 
 	bio->bi_bdev = dc->bdev;
-	bio->bi_sector += BDEV_DATA_START;
+	bio->bi_sector += dc->sb.data_offset;
 
 	if (cached_dev_get(dc)) {
 		s = search_alloc(bio, d);

drivers/md/bcache/super.c

@@ -110,15 +110,7 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev,
 
 	sb->flags		= le64_to_cpu(s->flags);
 	sb->seq			= le64_to_cpu(s->seq);
-
-	sb->nbuckets		= le64_to_cpu(s->nbuckets);
-	sb->block_size		= le16_to_cpu(s->block_size);
-	sb->bucket_size		= le16_to_cpu(s->bucket_size);
-
-	sb->nr_in_set		= le16_to_cpu(s->nr_in_set);
-	sb->nr_this_dev		= le16_to_cpu(s->nr_this_dev);
 	sb->last_mount		= le32_to_cpu(s->last_mount);
-
 	sb->first_bucket	= le16_to_cpu(s->first_bucket);
 	sb->keys		= le16_to_cpu(s->keys);
 
@@ -147,53 +139,81 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev,
 	if (bch_is_zero(sb->uuid, 16))
 		goto err;
 
-	err = "Unsupported superblock version";
-	if (sb->version > BCACHE_SB_VERSION)
-		goto err;
+	sb->block_size	= le16_to_cpu(s->block_size);
 
-	err = "Bad block/bucket size";
-	if (!is_power_of_2(sb->block_size) || sb->block_size > PAGE_SECTORS ||
-	    !is_power_of_2(sb->bucket_size) || sb->bucket_size < PAGE_SECTORS)
+	err = "Superblock block size smaller than device block size";
+	if (sb->block_size << 9 < bdev_logical_block_size(bdev))
 		goto err;
 
-	err = "Too many buckets";
-	if (sb->nbuckets > LONG_MAX)
-		goto err;
+	switch (sb->version) {
+	case BCACHE_SB_VERSION_BDEV:
+		sb->data_offset	= BDEV_DATA_START_DEFAULT;
+		break;
+	case BCACHE_SB_VERSION_BDEV_WITH_OFFSET:
+		sb->data_offset	= le64_to_cpu(s->data_offset);
 
-	err = "Not enough buckets";
-	if (sb->nbuckets < 1 << 7)
-		goto err;
+		err = "Bad data offset";
+		if (sb->data_offset < BDEV_DATA_START_DEFAULT)
+			goto err;
 
-	err = "Invalid superblock: device too small";
-	if (get_capacity(bdev->bd_disk) < sb->bucket_size * sb->nbuckets)
-		goto err;
+		break;
+	case BCACHE_SB_VERSION_CDEV:
+	case BCACHE_SB_VERSION_CDEV_WITH_UUID:
+		sb->nbuckets	= le64_to_cpu(s->nbuckets);
+		sb->block_size	= le16_to_cpu(s->block_size);
+		sb->bucket_size	= le16_to_cpu(s->bucket_size);
 
-	if (sb->version == CACHE_BACKING_DEV)
-		goto out;
+		sb->nr_in_set	= le16_to_cpu(s->nr_in_set);
+		sb->nr_this_dev	= le16_to_cpu(s->nr_this_dev);
 
-	err = "Bad UUID";
-	if (bch_is_zero(sb->set_uuid, 16))
-		goto err;
+		err = "Too many buckets";
+		if (sb->nbuckets > LONG_MAX)
+			goto err;
 
-	err = "Bad cache device number in set";
-	if (!sb->nr_in_set ||
-	    sb->nr_in_set <= sb->nr_this_dev ||
-	    sb->nr_in_set > MAX_CACHES_PER_SET)
-		goto err;
+		err = "Not enough buckets";
+		if (sb->nbuckets < 1 << 7)
+			goto err;
 
-	err = "Journal buckets not sequential";
-	for (i = 0; i < sb->keys; i++)
-		if (sb->d[i] != sb->first_bucket + i)
+		err = "Bad block/bucket size";
+		if (!is_power_of_2(sb->block_size) ||
+		    sb->block_size > PAGE_SECTORS ||
+		    !is_power_of_2(sb->bucket_size) ||
+		    sb->bucket_size < PAGE_SECTORS)
 			goto err;
 
-	err = "Too many journal buckets";
-	if (sb->first_bucket + sb->keys > sb->nbuckets)
-		goto err;
+		err = "Invalid superblock: device too small";
+		if (get_capacity(bdev->bd_disk) < sb->bucket_size * sb->nbuckets)
+			goto err;
+
+		err = "Bad UUID";
+		if (bch_is_zero(sb->set_uuid, 16))
+			goto err;
+
+		err = "Bad cache device number in set";
+		if (!sb->nr_in_set ||
+		    sb->nr_in_set <= sb->nr_this_dev ||
+		    sb->nr_in_set > MAX_CACHES_PER_SET)
+			goto err;
 
-	err = "Invalid superblock: first bucket comes before end of super";
-	if (sb->first_bucket * sb->bucket_size < 16)
+		err = "Journal buckets not sequential";
+		for (i = 0; i < sb->keys; i++)
+			if (sb->d[i] != sb->first_bucket + i)
+				goto err;
+
+		err = "Too many journal buckets";
+		if (sb->first_bucket + sb->keys > sb->nbuckets)
+			goto err;
+
+		err = "Invalid superblock: first bucket comes before end of super";
+		if (sb->first_bucket * sb->bucket_size < 16)
+			goto err;
+
+		break;
+	default:
+		err = "Unsupported superblock version";
 		goto err;
-out:
+	}
+
 	sb->last_mount = get_seconds();
 	err = NULL;
 
@@ -286,7 +306,7 @@ void bcache_write_super(struct cache_set *c)
 	for_each_cache(ca, c, i) {
 		struct bio *bio = &ca->sb_bio;
 
-		ca->sb.version		= BCACHE_SB_VERSION;
+		ca->sb.version		= BCACHE_SB_VERSION_CDEV_WITH_UUID;
 		ca->sb.seq		= c->sb.seq;
 		ca->sb.last_mount	= c->sb.last_mount;
 
@@ -641,6 +661,35 @@ void bcache_device_stop(struct bcache_device *d)
 		closure_queue(&d->cl);
 }
 
+static void bcache_device_unlink(struct bcache_device *d)
+{
+	unsigned i;
+	struct cache *ca;
+
+	sysfs_remove_link(&d->c->kobj, d->name);
+	sysfs_remove_link(&d->kobj, "cache");
+
+	for_each_cache(ca, d->c, i)
+		bd_unlink_disk_holder(ca->bdev, d->disk);
+}
+
+static void bcache_device_link(struct bcache_device *d, struct cache_set *c,
+			       const char *name)
+{
+	unsigned i;
+	struct cache *ca;
+
+	for_each_cache(ca, d->c, i)
+		bd_link_disk_holder(ca->bdev, d->disk);
+
+	snprintf(d->name, BCACHEDEVNAME_SIZE,
+		 "%s%u", name, d->id);
+
+	WARN(sysfs_create_link(&d->kobj, &c->kobj, "cache") ||
+	     sysfs_create_link(&c->kobj, &d->kobj, d->name),
+	     "Couldn't create device <-> cache set symlinks");
+}
+
 static void bcache_device_detach(struct bcache_device *d)
 {
 	lockdep_assert_held(&bch_register_lock);
@@ -656,6 +705,8 @@ static void bcache_device_detach(struct bcache_device *d)
 		atomic_set(&d->detaching, 0);
 	}
 
+	bcache_device_unlink(d);
+
 	d->c->devices[d->id] = NULL;
 	closure_put(&d->c->caching);
 	d->c = NULL;
@@ -673,17 +724,6 @@ static void bcache_device_attach(struct bcache_device *d, struct cache_set *c,
 	closure_get(&c->caching);
 }
 
-static void bcache_device_link(struct bcache_device *d, struct cache_set *c,
-			       const char *name)
-{
-	snprintf(d->name, BCACHEDEVNAME_SIZE,
-		 "%s%u", name, d->id);
-
-	WARN(sysfs_create_link(&d->kobj, &c->kobj, "cache") ||
-	     sysfs_create_link(&c->kobj, &d->kobj, d->name),
-	     "Couldn't create device <-> cache set symlinks");
-}
-
 static void bcache_device_free(struct bcache_device *d)
 {
 	lockdep_assert_held(&bch_register_lock);
@@ -784,6 +824,7 @@ void bch_cached_dev_run(struct cached_dev *dc)
 	}
 
 	add_disk(d->disk);
+	bd_link_disk_holder(dc->bdev, dc->disk.disk);
 #if 0
 	char *env[] = { "SYMLINK=label" , NULL };
 	kobject_uevent_env(&disk_to_dev(d->disk)->kobj, KOBJ_CHANGE, env);
@@ -803,9 +844,6 @@ static void cached_dev_detach_finish(struct work_struct *w)
 	BUG_ON(!atomic_read(&dc->disk.detaching));
 	BUG_ON(atomic_read(&dc->count));
 
-	sysfs_remove_link(&dc->disk.c->kobj, dc->disk.name);
-	sysfs_remove_link(&dc->disk.kobj, "cache");
-
 	mutex_lock(&bch_register_lock);
 
 	memset(&dc->sb.set_uuid, 0, 16);
@@ -920,7 +958,6 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c)
 	}
 
 	bcache_device_attach(&dc->disk, c, u - c->uuids);
-	bcache_device_link(&dc->disk, c, "bdev");
 	list_move(&dc->list, &c->cached_devs);
 	calc_cached_dev_sectors(c);
 
@@ -938,6 +975,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c)
 	}
 
 	bch_cached_dev_run(dc);
+	bcache_device_link(&dc->disk, c, "bdev");
 
 	pr_info("Caching %s as %s on set %pU",
 		bdevname(dc->bdev, buf), dc->disk.disk->disk_name,
@@ -961,6 +999,7 @@ static void cached_dev_free(struct closure *cl)
 
 	mutex_lock(&bch_register_lock);
 
+	bd_unlink_disk_holder(dc->bdev, dc->disk.disk);
 	bcache_device_free(&dc->disk);
 	list_del(&dc->list);
 
@@ -1049,7 +1088,11 @@ static const char *register_bdev(struct cache_sb *sb, struct page *sb_page,
 
 	g = dc->disk.disk;
 
-	set_capacity(g, dc->bdev->bd_part->nr_sects - 16);
+	set_capacity(g, dc->bdev->bd_part->nr_sects - dc->sb.data_offset);
+
+	g->queue->backing_dev_info.ra_pages =
+		max(g->queue->backing_dev_info.ra_pages,
+		    bdev->bd_queue->backing_dev_info.ra_pages);
 
 	bch_cached_dev_request_init(dc);
 
@@ -1099,8 +1142,7 @@ static void flash_dev_flush(struct closure *cl)
 {
 	struct bcache_device *d = container_of(cl, struct bcache_device, cl);
 
-	sysfs_remove_link(&d->c->kobj, d->name);
-	sysfs_remove_link(&d->kobj, "cache");
+	bcache_device_unlink(d);
 	kobject_del(&d->kobj);
 	continue_at(cl, flash_dev_free, system_wq);
 }
@@ -1802,7 +1844,7 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr,
 	if (err)
 		goto err_close;
 
-	if (sb->version == CACHE_BACKING_DEV) {
+	if (SB_IS_BDEV(sb)) {
 		struct cached_dev *dc = kzalloc(sizeof(*dc), GFP_KERNEL);
 
 		err = register_bdev(sb, sb_page, bdev, dc);