1 files changed, 353 insertions, 203 deletions

diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 0afcd452fcb..b6d198f5181 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -158,12 +158,20 @@ struct scrub_fixup_nodatasum {
 	int			mirror_num;
 };
 
+struct scrub_nocow_inode {
+	u64			inum;
+	u64			offset;
+	u64			root;
+	struct list_head	list;
+};
+
 struct scrub_copy_nocow_ctx {
 	struct scrub_ctx	*sctx;
 	u64			logical;
 	u64			len;
 	int			mirror_num;
 	u64			physical_for_dev_replace;
+	struct list_head	inodes;
 	struct btrfs_work	work;
 };
 
@@ -200,7 +208,6 @@ static void scrub_recheck_block_checksum(struct btrfs_fs_info *fs_info,
 					 int is_metadata, int have_csum,
 					 const u8 *csum, u64 generation,
 					 u16 csum_size);
-static void scrub_complete_bio_end_io(struct bio *bio, int err);
 static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad,
 					     struct scrub_block *sblock_good,
 					     int force_write);
@@ -245,10 +252,12 @@ static void scrub_wr_bio_end_io_worker(struct btrfs_work *work);
 static int write_page_nocow(struct scrub_ctx *sctx,
 			    u64 physical_for_dev_replace, struct page *page);
 static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root,
-				      void *ctx);
+				      struct scrub_copy_nocow_ctx *ctx);
 static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
 			    int mirror_num, u64 physical_for_dev_replace);
 static void copy_nocow_pages_worker(struct btrfs_work *work);
+static void __scrub_blocked_if_needed(struct btrfs_fs_info *fs_info);
+static void scrub_blocked_if_needed(struct btrfs_fs_info *fs_info);
 
 
 static void scrub_pending_bio_inc(struct scrub_ctx *sctx)
@@ -262,6 +271,29 @@ static void scrub_pending_bio_dec(struct scrub_ctx *sctx)
 	wake_up(&sctx->list_wait);
 }
 
+static void __scrub_blocked_if_needed(struct btrfs_fs_info *fs_info)
+{
+	while (atomic_read(&fs_info->scrub_pause_req)) {
+		mutex_unlock(&fs_info->scrub_lock);
+		wait_event(fs_info->scrub_pause_wait,
+		   atomic_read(&fs_info->scrub_pause_req) == 0);
+		mutex_lock(&fs_info->scrub_lock);
+	}
+}
+
+static void scrub_blocked_if_needed(struct btrfs_fs_info *fs_info)
+{
+	atomic_inc(&fs_info->scrubs_paused);
+	wake_up(&fs_info->scrub_pause_wait);
+
+	mutex_lock(&fs_info->scrub_lock);
+	__scrub_blocked_if_needed(fs_info);
+	atomic_dec(&fs_info->scrubs_paused);
+	mutex_unlock(&fs_info->scrub_lock);
+
+	wake_up(&fs_info->scrub_pause_wait);
+}
+
 /*
  * used for workers that require transaction commits (i.e., for the
  * NOCOW case)
@@ -283,6 +315,16 @@ static void scrub_pending_trans_workers_inc(struct scrub_ctx *sctx)
 	atomic_inc(&fs_info->scrubs_running);
 	atomic_inc(&fs_info->scrubs_paused);
 	mutex_unlock(&fs_info->scrub_lock);
+
+	/*
+	 * check if @scrubs_running=@scrubs_paused condition
+	 * inside wait_event() is not an atomic operation.
+	 * which means we may inc/dec @scrub_running/paused
+	 * at any time. Let's wake up @scrub_pause_wait as
+	 * much as we can to let commit transaction blocked less.
+	 */
+	wake_up(&fs_info->scrub_pause_wait);
+
 	atomic_inc(&sctx->workers_pending);
 }
 
@@ -386,7 +428,8 @@ struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace)
 		sbio->index = i;
 		sbio->sctx = sctx;
 		sbio->page_count = 0;
-		sbio->work.func = scrub_bio_end_io_worker;
+		btrfs_init_work(&sbio->work, scrub_bio_end_io_worker,
+				NULL, NULL);
 
 		if (i != SCRUB_BIOS_PER_SCTX - 1)
 			sctx->bios[i]->next_free = i + 1;
@@ -473,7 +516,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 	 * hold all of the paths here
 	 */
 	for (i = 0; i < ipath->fspath->elem_cnt; ++i)
-		printk_in_rcu(KERN_WARNING "btrfs: %s at logical %llu on dev "
+		printk_in_rcu(KERN_WARNING "BTRFS: %s at logical %llu on dev "
 			"%s, sector %llu, root %llu, inode %llu, offset %llu, "
 			"length %llu, links %u (path: %s)\n", swarn->errstr,
 			swarn->logical, rcu_str_deref(swarn->dev->name),
@@ -485,7 +528,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 	return 0;
 
 err:
-	printk_in_rcu(KERN_WARNING "btrfs: %s at logical %llu on dev "
+	printk_in_rcu(KERN_WARNING "BTRFS: %s at logical %llu on dev "
 		"%s, sector %llu, root %llu, inode %llu, offset %llu: path "
 		"resolving failed with ret=%d\n", swarn->errstr,
 		swarn->logical, rcu_str_deref(swarn->dev->name),
@@ -545,10 +588,11 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
 
 	if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
 		do {
-			ret = tree_backref_for_extent(&ptr, eb, ei, item_size,
-							&ref_root, &ref_level);
+			ret = tree_backref_for_extent(&ptr, eb, &found_key, ei,
+						      item_size, &ref_root,
+						      &ref_level);
 			printk_in_rcu(KERN_WARNING
-				"btrfs: %s at logical %llu on dev %s, "
+				"BTRFS: %s at logical %llu on dev %s, "
 				"sector %llu: metadata %s (level %d) in tree "
 				"%llu\n", errstr, swarn.logical,
 				rcu_str_deref(dev->name),
@@ -674,8 +718,8 @@ static int scrub_fixup_readpage(u64 inum, u64 offset, u64 root, void *fixup_ctx)
 out:
 	if (page)
 		put_page(page);
-	if (inode)
-		iput(inode);
+
+	iput(inode);
 
 	if (ret < 0)
 		return ret;
@@ -697,13 +741,11 @@ static void scrub_fixup_nodatasum(struct btrfs_work *work)
 	struct scrub_fixup_nodatasum *fixup;
 	struct scrub_ctx *sctx;
 	struct btrfs_trans_handle *trans = NULL;
-	struct btrfs_fs_info *fs_info;
 	struct btrfs_path *path;
 	int uncorrectable = 0;
 
 	fixup = container_of(work, struct scrub_fixup_nodatasum, work);
 	sctx = fixup->sctx;
-	fs_info = fixup->root->fs_info;
 
 	path = btrfs_alloc_path();
 	if (!path) {
@@ -752,8 +794,8 @@ out:
 		btrfs_dev_replace_stats_inc(
 			&sctx->dev_root->fs_info->dev_replace.
 			num_uncorrectable_read_errors);
-		printk_ratelimited_in_rcu(KERN_ERR
-			"btrfs: unable to fixup (nodatasum) error at logical %llu on dev %s\n",
+		printk_ratelimited_in_rcu(KERN_ERR "BTRFS: "
+		    "unable to fixup (nodatasum) error at logical %llu on dev %s\n",
 			fixup->logical, rcu_str_deref(fixup->dev->name));
 	}
 
@@ -930,8 +972,10 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 				BTRFS_DEV_STAT_CORRUPTION_ERRS);
 	}
 
-	if (sctx->readonly && !sctx->is_dev_replace)
-		goto did_not_correct_error;
+	if (sctx->readonly) {
+		ASSERT(!sctx->is_dev_replace);
+		goto out;
+	}
 
 	if (!is_metadata && !have_csum) {
 		struct scrub_fixup_nodatasum *fixup_nodatasum;
@@ -955,9 +999,10 @@ nodatasum_case:
 		fixup_nodatasum->root = fs_info->extent_root;
 		fixup_nodatasum->mirror_num = failed_mirror_index + 1;
 		scrub_pending_trans_workers_inc(sctx);
-		fixup_nodatasum->work.func = scrub_fixup_nodatasum;
-		btrfs_queue_worker(&fs_info->scrub_workers,
-				   &fixup_nodatasum->work);
+		btrfs_init_work(&fixup_nodatasum->work, scrub_fixup_nodatasum,
+				NULL, NULL);
+		btrfs_queue_work(fs_info->scrub_workers,
+				 &fixup_nodatasum->work);
 		goto out;
 	}
 
@@ -1152,7 +1197,7 @@ corrected_error:
 			sctx->stat.corrected_errors++;
 			spin_unlock(&sctx->stat_lock);
 			printk_ratelimited_in_rcu(KERN_ERR
-				"btrfs: fixed up error at logical %llu on dev %s\n",
+				"BTRFS: fixed up error at logical %llu on dev %s\n",
 				logical, rcu_str_deref(dev->name));
 		}
 	} else {
@@ -1161,7 +1206,7 @@ did_not_correct_error:
 		sctx->stat.uncorrectable_errors++;
 		spin_unlock(&sctx->stat_lock);
 		printk_ratelimited_in_rcu(KERN_ERR
-			"btrfs: unable to fixup (regular) error at logical %llu on dev %s\n",
+			"BTRFS: unable to fixup (regular) error at logical %llu on dev %s\n",
 			logical, rcu_str_deref(dev->name));
 	}
 
@@ -1284,7 +1329,6 @@ static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
 	for (page_num = 0; page_num < sblock->page_count; page_num++) {
 		struct bio *bio;
 		struct scrub_page *page = sblock->pagev[page_num];
-		DECLARE_COMPLETION_ONSTACK(complete);
 
 		if (page->dev->bdev == NULL) {
 			page->io_error = 1;
@@ -1300,19 +1344,12 @@ static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
 			continue;
 		}
 		bio->bi_bdev = page->dev->bdev;
-		bio->bi_sector = page->physical >> 9;
-		bio->bi_end_io = scrub_complete_bio_end_io;
-		bio->bi_private = &complete;
+		bio->bi_iter.bi_sector = page->physical >> 9;
 
 		bio_add_page(bio, page->page, PAGE_SIZE, 0);
-		btrfsic_submit_bio(READ, bio);
-
-		/* this will also unplug the queue */
-		wait_for_completion(&complete);
-
-		page->io_error = !test_bit(BIO_UPTODATE, &bio->bi_flags);
-		if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
+		if (btrfsic_submit_bio_wait(READ, bio))
 			sblock->no_io_error_seen = 0;
+
 		bio_put(bio);
 	}
 
@@ -1381,11 +1418,6 @@ static void scrub_recheck_block_checksum(struct btrfs_fs_info *fs_info,
 		sblock->checksum_error = 1;
 }
 
-static void scrub_complete_bio_end_io(struct bio *bio, int err)
-{
-	complete((struct completion *)bio->bi_private);
-}
-
 static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad,
 					     struct scrub_block *sblock_good,
 					     int force_write)
@@ -1420,11 +1452,11 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 	    sblock_bad->checksum_error || page_bad->io_error) {
 		struct bio *bio;
 		int ret;
-		DECLARE_COMPLETION_ONSTACK(complete);
 
 		if (!page_bad->dev->bdev) {
-			printk_ratelimited(KERN_WARNING
-				"btrfs: scrub_repair_page_from_good_copy(bdev == NULL) is unexpected!\n");
+			printk_ratelimited(KERN_WARNING "BTRFS: "
+				"scrub_repair_page_from_good_copy(bdev == NULL) "
+				"is unexpected!\n");
 			return -EIO;
 		}
 
@@ -1432,20 +1464,15 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 		if (!bio)
 			return -EIO;
 		bio->bi_bdev = page_bad->dev->bdev;
-		bio->bi_sector = page_bad->physical >> 9;
-		bio->bi_end_io = scrub_complete_bio_end_io;
-		bio->bi_private = &complete;
+		bio->bi_iter.bi_sector = page_bad->physical >> 9;
 
 		ret = bio_add_page(bio, page_good->page, PAGE_SIZE, 0);
 		if (PAGE_SIZE != ret) {
 			bio_put(bio);
 			return -EIO;
 		}
-		btrfsic_submit_bio(WRITE, bio);
 
-		/* this will also unplug the queue */
-		wait_for_completion(&complete);
-		if (!bio_flagged(bio, BIO_UPTODATE)) {
+		if (btrfsic_submit_bio_wait(WRITE, bio)) {
 			btrfs_dev_stat_inc_and_print(page_bad->dev,
 				BTRFS_DEV_STAT_WRITE_ERRS);
 			btrfs_dev_replace_stats_inc(
@@ -1530,7 +1557,7 @@ again:
 		bio->bi_private = sbio;
 		bio->bi_end_io = scrub_wr_bio_end_io;
 		bio->bi_bdev = sbio->dev->bdev;
-		bio->bi_sector = sbio->physical >> 9;
+		bio->bi_iter.bi_sector = sbio->physical >> 9;
 		sbio->err = 0;
 	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
 		   spage->physical_for_dev_replace ||
@@ -1589,8 +1616,8 @@ static void scrub_wr_bio_end_io(struct bio *bio, int err)
 	sbio->err = err;
 	sbio->bio = bio;
 
-	sbio->work.func = scrub_wr_bio_end_io_worker;
-	btrfs_queue_worker(&fs_info->scrub_wr_completion_workers, &sbio->work);
+	btrfs_init_work(&sbio->work, scrub_wr_bio_end_io_worker, NULL, NULL);
+	btrfs_queue_work(fs_info->scrub_wr_completion_workers, &sbio->work);
 }
 
 static void scrub_wr_bio_end_io_worker(struct btrfs_work *work)
@@ -1887,7 +1914,7 @@ static void scrub_submit(struct scrub_ctx *sctx)
 		 * This case is handled correctly (but _very_ slowly).
 		 */
 		printk_ratelimited(KERN_WARNING
-			"btrfs: scrub_submit(bio bdev == NULL) is unexpected!\n");
+			"BTRFS: scrub_submit(bio bdev == NULL) is unexpected!\n");
 		bio_endio(sbio->bio, -EIO);
 	} else {
 		btrfsic_submit_bio(READ, sbio->bio);
@@ -1936,7 +1963,7 @@ again:
 		bio->bi_private = sbio;
 		bio->bi_end_io = scrub_bio_end_io;
 		bio->bi_bdev = sbio->dev->bdev;
-		bio->bi_sector = sbio->physical >> 9;
+		bio->bi_iter.bi_sector = sbio->physical >> 9;
 		sbio->err = 0;
 	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
 		   spage->physical ||
@@ -2058,7 +2085,7 @@ static void scrub_bio_end_io(struct bio *bio, int err)
 	sbio->err = err;
 	sbio->bio = bio;
 
-	btrfs_queue_worker(&fs_info->scrub_workers, &sbio->work);
+	btrfs_queue_work(fs_info->scrub_workers, &sbio->work);
 }
 
 static void scrub_bio_end_io_worker(struct btrfs_work *work)
@@ -2209,6 +2236,47 @@ behind_scrub_pages:
 	return 0;
 }
 
+/*
+ * Given a physical address, this will calculate it's
+ * logical offset. if this is a parity stripe, it will return
+ * the most left data stripe's logical offset.
+ *
+ * return 0 if it is a data stripe, 1 means parity stripe.
+ */
+static int get_raid56_logic_offset(u64 physical, int num,
+				   struct map_lookup *map, u64 *offset)
+{
+	int i;
+	int j = 0;
+	u64 stripe_nr;
+	u64 last_offset;
+	int stripe_index;
+	int rot;
+
+	last_offset = (physical - map->stripes[num].physical) *
+		      nr_data_stripes(map);
+	*offset = last_offset;
+	for (i = 0; i < nr_data_stripes(map); i++) {
+		*offset = last_offset + i * map->stripe_len;
+
+		stripe_nr = *offset;
+		do_div(stripe_nr, map->stripe_len);
+		do_div(stripe_nr, nr_data_stripes(map));
+
+		/* Work out the disk rotation on this stripe-set */
+		rot = do_div(stripe_nr, map->num_stripes);
+		/* calculate which stripe this data locates */
+		rot += i;
+		stripe_index = rot % map->num_stripes;
+		if (stripe_index == num)
+			return 0;
+		if (stripe_index < num)
+			j++;
+	}
+	*offset = last_offset + j * map->stripe_len;
+	return 1;
+}
+
 static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 					   struct map_lookup *map,
 					   struct btrfs_device *scrub_dev,
@@ -2230,6 +2298,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	u64 physical;
 	u64 logical;
 	u64 logic_end;
+	u64 physical_end;
 	u64 generation;
 	int mirror_num;
 	struct reada_control *reada1;
@@ -2243,16 +2312,10 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	u64 extent_len;
 	struct btrfs_device *extent_dev;
 	int extent_mirror_num;
-	int stop_loop;
-
-	if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
-			 BTRFS_BLOCK_GROUP_RAID6)) {
-		if (num >= nr_data_stripes(map)) {
-			return 0;
-		}
-	}
+	int stop_loop = 0;
 
 	nstripes = length;
+	physical = map->stripes[num].physical;
 	offset = 0;
 	do_div(nstripes, map->stripe_len);
 	if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
@@ -2270,6 +2333,11 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	} else if (map->type & BTRFS_BLOCK_GROUP_DUP) {
 		increment = map->stripe_len;
 		mirror_num = num % map->num_stripes + 1;
+	} else if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+				BTRFS_BLOCK_GROUP_RAID6)) {
+		get_raid56_logic_offset(physical, num, map, &offset);
+		increment = map->stripe_len * nr_data_stripes(map);
+		mirror_num = 1;
 	} else {
 		increment = map->stripe_len;
 		mirror_num = 1;
@@ -2293,17 +2361,24 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	 * to not hold off transaction commits
 	 */
 	logical = base + offset;
-
+	physical_end = physical + nstripes * map->stripe_len;
+	if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+			 BTRFS_BLOCK_GROUP_RAID6)) {
+		get_raid56_logic_offset(physical_end, num,
+					map, &logic_end);
+		logic_end += base;
+	} else {
+		logic_end = logical + increment * nstripes;
+	}
 	wait_event(sctx->list_wait,
 		   atomic_read(&sctx->bios_in_flight) == 0);
-	atomic_inc(&fs_info->scrubs_paused);
-	wake_up(&fs_info->scrub_pause_wait);
+	scrub_blocked_if_needed(fs_info);
 
 	/* FIXME it might be better to start readahead at commit root */
 	key_start.objectid = logical;
 	key_start.type = BTRFS_EXTENT_ITEM_KEY;
 	key_start.offset = (u64)0;
-	key_end.objectid = base + offset + nstripes * increment;
+	key_end.objectid = logic_end;
 	key_end.type = BTRFS_METADATA_ITEM_KEY;
 	key_end.offset = (u64)-1;
 	reada1 = btrfs_reada_add(root, &key_start, &key_end);
@@ -2313,7 +2388,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	key_start.offset = logical;
 	key_end.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
 	key_end.type = BTRFS_EXTENT_CSUM_KEY;
-	key_end.offset = base + offset + nstripes * increment;
+	key_end.offset = logic_end;
 	reada2 = btrfs_reada_add(csum_root, &key_start, &key_end);
 
 	if (!IS_ERR(reada1))
@@ -2321,16 +2396,6 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	if (!IS_ERR(reada2))
 		btrfs_reada_wait(reada2);
 
-	mutex_lock(&fs_info->scrub_lock);
-	while (atomic_read(&fs_info->scrub_pause_req)) {
-		mutex_unlock(&fs_info->scrub_lock);
-		wait_event(fs_info->scrub_pause_wait,
-		   atomic_read(&fs_info->scrub_pause_req) == 0);
-		mutex_lock(&fs_info->scrub_lock);
-	}
-	atomic_dec(&fs_info->scrubs_paused);
-	mutex_unlock(&fs_info->scrub_lock);
-	wake_up(&fs_info->scrub_pause_wait);
 
 	/*
 	 * collect all data csums for the stripe to avoid seeking during
@@ -2341,11 +2406,17 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	/*
 	 * now find all extents for each stripe and scrub them
 	 */
-	logical = base + offset;
-	physical = map->stripes[num].physical;
-	logic_end = logical + increment * nstripes;
 	ret = 0;
-	while (logical < logic_end) {
+	while (physical < physical_end) {
+		/* for raid56, we skip parity stripe */
+		if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+				BTRFS_BLOCK_GROUP_RAID6)) {
+			ret = get_raid56_logic_offset(physical, num,
+					map, &logical);
+			logical += base;
+			if (ret)
+				goto skip;
+		}
 		/*
 		 * canceled?
 		 */
@@ -2367,22 +2438,14 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 			wait_event(sctx->list_wait,
 				   atomic_read(&sctx->bios_in_flight) == 0);
 			atomic_set(&sctx->wr_ctx.flush_all_writes, 0);
-			atomic_inc(&fs_info->scrubs_paused);
-			wake_up(&fs_info->scrub_pause_wait);
-			mutex_lock(&fs_info->scrub_lock);
-			while (atomic_read(&fs_info->scrub_pause_req)) {
-				mutex_unlock(&fs_info->scrub_lock);
-				wait_event(fs_info->scrub_pause_wait,
-				   atomic_read(&fs_info->scrub_pause_req) == 0);
-				mutex_lock(&fs_info->scrub_lock);
-			}
-			atomic_dec(&fs_info->scrubs_paused);
-			mutex_unlock(&fs_info->scrub_lock);
-			wake_up(&fs_info->scrub_pause_wait);
+			scrub_blocked_if_needed(fs_info);
 		}
 
+		if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
+			key.type = BTRFS_METADATA_ITEM_KEY;
+		else
+			key.type = BTRFS_EXTENT_ITEM_KEY;
 		key.objectid = logical;
-		key.type = BTRFS_EXTENT_ITEM_KEY;
 		key.offset = (u64)-1;
 
 		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
@@ -2390,8 +2453,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 			goto out;
 
 		if (ret > 0) {
-			ret = btrfs_previous_item(root, path, 0,
-						  BTRFS_EXTENT_ITEM_KEY);
+			ret = btrfs_previous_extent_item(root, path, 0);
 			if (ret < 0)
 				goto out;
 			if (ret > 0) {
@@ -2449,9 +2511,9 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 
 			if (key.objectid < logical &&
 			    (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK)) {
-				printk(KERN_ERR
-				       "btrfs scrub: tree block %llu spanning "
-				       "stripes, ignored. logical=%llu\n",
+				btrfs_err(fs_info,
+					   "scrub: tree block %llu spanning "
+					   "stripes, ignored. logical=%llu",
 				       key.objectid, logical);
 				goto next;
 			}
@@ -2498,15 +2560,29 @@ again:
 			scrub_free_csums(sctx);
 			if (extent_logical + extent_len <
 			    key.objectid + bytes) {
-				logical += increment;
-				physical += map->stripe_len;
-
+				if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+					BTRFS_BLOCK_GROUP_RAID6)) {
+					/*
+					 * loop until we find next data stripe
+					 * or we have finished all stripes.
+					 */
+					do {
+						physical += map->stripe_len;
+						ret = get_raid56_logic_offset(
+								physical, num,
+								map, &logical);
+						logical += base;
+					} while (physical < physical_end && ret);
+				} else {
+					physical += map->stripe_len;
+					logical += increment;
+				}
 				if (logical < key.objectid + bytes) {
 					cond_resched();
 					goto again;
 				}
 
-				if (logical >= logic_end) {
+				if (physical >= physical_end) {
 					stop_loop = 1;
 					break;
 				}
@@ -2515,6 +2591,7 @@ next:
 			path->slots[0]++;
 		}
 		btrfs_release_path(path);
+skip:
 		logical += increment;
 		physical += map->stripe_len;
 		spin_lock(&sctx->stat_lock);
@@ -2648,11 +2725,8 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		dev_extent = btrfs_item_ptr(l, slot, struct btrfs_dev_extent);
 		length = btrfs_dev_extent_length(l, dev_extent);
 
-		if (found_key.offset + length <= start) {
-			key.offset = found_key.offset + length;
-			btrfs_release_path(path);
-			continue;
-		}
+		if (found_key.offset + length <= start)
+			goto skip;
 
 		chunk_tree = btrfs_dev_extent_chunk_tree(l, dev_extent);
 		chunk_objectid = btrfs_dev_extent_chunk_objectid(l, dev_extent);
@@ -2663,10 +2737,12 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		 * the chunk from going away while we scrub it
 		 */
 		cache = btrfs_lookup_block_group(fs_info, chunk_offset);
-		if (!cache) {
-			ret = -ENOENT;
-			break;
-		}
+
+		/* some chunks are removed but not committed to disk yet,
+		 * continue scrubbing */
+		if (!cache)
+			goto skip;
+
 		dev_replace->cursor_right = found_key.offset + length;
 		dev_replace->cursor_left = found_key.offset;
 		dev_replace->item_needs_writeback = 1;
@@ -2692,25 +2768,24 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 
 		wait_event(sctx->list_wait,
 			   atomic_read(&sctx->bios_in_flight) == 0);
-		atomic_set(&sctx->wr_ctx.flush_all_writes, 0);
 		atomic_inc(&fs_info->scrubs_paused);
 		wake_up(&fs_info->scrub_pause_wait);
+
+		/*
+		 * must be called before we decrease @scrub_paused.
+		 * make sure we don't block transaction commit while
+		 * we are waiting pending workers finished.
+		 */
 		wait_event(sctx->list_wait,
 			   atomic_read(&sctx->workers_pending) == 0);
+		atomic_set(&sctx->wr_ctx.flush_all_writes, 0);
 
 		mutex_lock(&fs_info->scrub_lock);
-		while (atomic_read(&fs_info->scrub_pause_req)) {
-			mutex_unlock(&fs_info->scrub_lock);
-			wait_event(fs_info->scrub_pause_wait,
-			   atomic_read(&fs_info->scrub_pause_req) == 0);
-			mutex_lock(&fs_info->scrub_lock);
-		}
+		__scrub_blocked_if_needed(fs_info);
 		atomic_dec(&fs_info->scrubs_paused);
 		mutex_unlock(&fs_info->scrub_lock);
 		wake_up(&fs_info->scrub_pause_wait);
 
-		dev_replace->cursor_left = dev_replace->cursor_right;
-		dev_replace->item_needs_writeback = 1;
 		btrfs_put_block_group(cache);
 		if (ret)
 			break;
@@ -2724,6 +2799,9 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			break;
 		}
 
+		dev_replace->cursor_left = dev_replace->cursor_right;
+		dev_replace->item_needs_writeback = 1;
+skip:
 		key.offset = found_key.offset + length;
 		btrfs_release_path(path);
 	}
@@ -2774,52 +2852,49 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 						int is_dev_replace)
 {
 	int ret = 0;
+	int flags = WQ_FREEZABLE | WQ_UNBOUND;
+	int max_active = fs_info->thread_pool_size;
 
-	mutex_lock(&fs_info->scrub_lock);
 	if (fs_info->scrub_workers_refcnt == 0) {
 		if (is_dev_replace)
-			btrfs_init_workers(&fs_info->scrub_workers, "scrub", 1,
-					&fs_info->generic_worker);
+			fs_info->scrub_workers =
+				btrfs_alloc_workqueue("btrfs-scrub", flags,
+						      1, 4);
 		else
-			btrfs_init_workers(&fs_info->scrub_workers, "scrub",
-					fs_info->thread_pool_size,
-					&fs_info->generic_worker);
-		fs_info->scrub_workers.idle_thresh = 4;
-		ret = btrfs_start_workers(&fs_info->scrub_workers);
-		if (ret)
+			fs_info->scrub_workers =
+				btrfs_alloc_workqueue("btrfs-scrub", flags,
+						      max_active, 4);
+		if (!fs_info->scrub_workers) {
+			ret = -ENOMEM;
 			goto out;
-		btrfs_init_workers(&fs_info->scrub_wr_completion_workers,
-				   "scrubwrc",
-				   fs_info->thread_pool_size,
-				   &fs_info->generic_worker);
-		fs_info->scrub_wr_completion_workers.idle_thresh = 2;
-		ret = btrfs_start_workers(
-				&fs_info->scrub_wr_completion_workers);
-		if (ret)
+		}
+		fs_info->scrub_wr_completion_workers =
+			btrfs_alloc_workqueue("btrfs-scrubwrc", flags,
+					      max_active, 2);
+		if (!fs_info->scrub_wr_completion_workers) {
+			ret = -ENOMEM;
 			goto out;
-		btrfs_init_workers(&fs_info->scrub_nocow_workers, "scrubnc", 1,
-				   &fs_info->generic_worker);
-		ret = btrfs_start_workers(&fs_info->scrub_nocow_workers);
-		if (ret)
+		}
+		fs_info->scrub_nocow_workers =
+			btrfs_alloc_workqueue("btrfs-scrubnc", flags, 1, 0);
+		if (!fs_info->scrub_nocow_workers) {
+			ret = -ENOMEM;
 			goto out;
+		}
 	}
 	++fs_info->scrub_workers_refcnt;
 out:
-	mutex_unlock(&fs_info->scrub_lock);
-
 	return ret;
 }
 
 static noinline_for_stack void scrub_workers_put(struct btrfs_fs_info *fs_info)
 {
-	mutex_lock(&fs_info->scrub_lock);
 	if (--fs_info->scrub_workers_refcnt == 0) {
-		btrfs_stop_workers(&fs_info->scrub_workers);
-		btrfs_stop_workers(&fs_info->scrub_wr_completion_workers);
-		btrfs_stop_workers(&fs_info->scrub_nocow_workers);
+		btrfs_destroy_workqueue(fs_info->scrub_workers);
+		btrfs_destroy_workqueue(fs_info->scrub_wr_completion_workers);
+		btrfs_destroy_workqueue(fs_info->scrub_nocow_workers);
 	}
 	WARN_ON(fs_info->scrub_workers_refcnt < 0);
-	mutex_unlock(&fs_info->scrub_lock);
 }
 
 int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
@@ -2837,8 +2912,8 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	 * check some assumptions
 	 */
 	if (fs_info->chunk_root->nodesize != fs_info->chunk_root->leafsize) {
-		printk(KERN_ERR
-		       "btrfs_scrub: size assumption nodesize == leafsize (%d == %d) fails\n",
+		btrfs_err(fs_info,
+			   "scrub: size assumption nodesize == leafsize (%d == %d) fails",
 		       fs_info->chunk_root->nodesize,
 		       fs_info->chunk_root->leafsize);
 		return -EINVAL;
@@ -2850,16 +2925,17 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		 * the way scrub is implemented. Do not handle this
 		 * situation at all because it won't ever happen.
 		 */
-		printk(KERN_ERR
-		       "btrfs_scrub: size assumption nodesize <= BTRFS_STRIPE_LEN (%d <= %d) fails\n",
+		btrfs_err(fs_info,
+			   "scrub: size assumption nodesize <= BTRFS_STRIPE_LEN (%d <= %d) fails",
 		       fs_info->chunk_root->nodesize, BTRFS_STRIPE_LEN);
 		return -EINVAL;
 	}
 
 	if (fs_info->chunk_root->sectorsize != PAGE_SIZE) {
 		/* not supported for data w/o checksums */
-		printk(KERN_ERR
-		       "btrfs_scrub: size assumption sectorsize != PAGE_SIZE (%d != %lu) fails\n",
+		btrfs_err(fs_info,
+			   "scrub: size assumption sectorsize != PAGE_SIZE "
+			   "(%d != %lu) fails",
 		       fs_info->chunk_root->sectorsize, PAGE_SIZE);
 		return -EINVAL;
 	}
@@ -2872,7 +2948,8 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		 * would exhaust the array bounds of pagev member in
 		 * struct scrub_block
 		 */
-		pr_err("btrfs_scrub: size assumption nodesize and sectorsize <= SCRUB_MAX_PAGES_PER_BLOCK (%d <= %d && %d <= %d) fails\n",
+		btrfs_err(fs_info, "scrub: size assumption nodesize and sectorsize "
+			   "<= SCRUB_MAX_PAGES_PER_BLOCK (%d <= %d && %d <= %d) fails",
 		       fs_info->chunk_root->nodesize,
 		       SCRUB_MAX_PAGES_PER_BLOCK,
 		       fs_info->chunk_root->sectorsize,
@@ -2880,23 +2957,18 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		return -EINVAL;
 	}
 
-	ret = scrub_workers_get(fs_info, is_dev_replace);
-	if (ret)
-		return ret;
 
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	dev = btrfs_find_device(fs_info, devid, NULL, NULL);
 	if (!dev || (dev->missing && !is_dev_replace)) {
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		scrub_workers_put(fs_info);
 		return -ENODEV;
 	}
-	mutex_lock(&fs_info->scrub_lock);
 
+	mutex_lock(&fs_info->scrub_lock);
 	if (!dev->in_fs_metadata || dev->is_tgtdev_for_dev_replace) {
 		mutex_unlock(&fs_info->scrub_lock);
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		scrub_workers_put(fs_info);
 		return -EIO;
 	}
 
@@ -2907,10 +2979,17 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		btrfs_dev_replace_unlock(&fs_info->dev_replace);
 		mutex_unlock(&fs_info->scrub_lock);
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		scrub_workers_put(fs_info);
 		return -EINPROGRESS;
 	}
 	btrfs_dev_replace_unlock(&fs_info->dev_replace);
+
+	ret = scrub_workers_get(fs_info, is_dev_replace);
+	if (ret) {
+		mutex_unlock(&fs_info->scrub_lock);
+		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+		return ret;
+	}
+
 	sctx = scrub_setup_ctx(dev, is_dev_replace);
 	if (IS_ERR(sctx)) {
 		mutex_unlock(&fs_info->scrub_lock);
@@ -2920,15 +2999,24 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	}
 	sctx->readonly = readonly;
 	dev->scrub_device = sctx;
+	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
+	/*
+	 * checking @scrub_pause_req here, we can avoid
+	 * race between committing transaction and scrubbing.
+	 */
+	__scrub_blocked_if_needed(fs_info);
 	atomic_inc(&fs_info->scrubs_running);
 	mutex_unlock(&fs_info->scrub_lock);
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
 	if (!is_dev_replace) {
-		down_read(&fs_info->scrub_super_lock);
+		/*
+		 * by holding device list mutex, we can
+		 * kick off writing super in log tree sync.
+		 */
+		mutex_lock(&fs_info->fs_devices->device_list_mutex);
 		ret = scrub_supers(sctx, dev);
-		up_read(&fs_info->scrub_super_lock);
+		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 	}
 
 	if (!ret)
@@ -2946,10 +3034,10 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 
 	mutex_lock(&fs_info->scrub_lock);
 	dev->scrub_device = NULL;
+	scrub_workers_put(fs_info);
 	mutex_unlock(&fs_info->scrub_lock);
 
 	scrub_free_ctx(sctx);
-	scrub_workers_put(fs_info);
 
 	return ret;
 }
@@ -2979,16 +3067,6 @@ void btrfs_scrub_continue(struct btrfs_root *root)
 	wake_up(&fs_info->scrub_pause_wait);
 }
 
-void btrfs_scrub_pause_super(struct btrfs_root *root)
-{
-	down_write(&root->fs_info->scrub_super_lock);
-}
-
-void btrfs_scrub_continue_super(struct btrfs_root *root)
-{
-	up_write(&root->fs_info->scrub_super_lock);
-}
-
 int btrfs_scrub_cancel(struct btrfs_fs_info *fs_info)
 {
 	mutex_lock(&fs_info->scrub_lock);
@@ -3125,13 +3203,31 @@ static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
 	nocow_ctx->len = len;
 	nocow_ctx->mirror_num = mirror_num;
 	nocow_ctx->physical_for_dev_replace = physical_for_dev_replace;
-	nocow_ctx->work.func = copy_nocow_pages_worker;
-	btrfs_queue_worker(&fs_info->scrub_nocow_workers,
-			   &nocow_ctx->work);
+	btrfs_init_work(&nocow_ctx->work, copy_nocow_pages_worker, NULL, NULL);
+	INIT_LIST_HEAD(&nocow_ctx->inodes);
+	btrfs_queue_work(fs_info->scrub_nocow_workers,
+			 &nocow_ctx->work);
 
 	return 0;
 }
 
+static int record_inode_for_nocow(u64 inum, u64 offset, u64 root, void *ctx)
+{
+	struct scrub_copy_nocow_ctx *nocow_ctx = ctx;
+	struct scrub_nocow_inode *nocow_inode;
+
+	nocow_inode = kzalloc(sizeof(*nocow_inode), GFP_NOFS);
+	if (!nocow_inode)
+		return -ENOMEM;
+	nocow_inode->inum = inum;
+	nocow_inode->offset = offset;
+	nocow_inode->root = root;
+	list_add_tail(&nocow_inode->list, &nocow_ctx->inodes);
+	return 0;
+}
+
+#define COPY_COMPLETE 1
+
 static void copy_nocow_pages_worker(struct btrfs_work *work)
 {
 	struct scrub_copy_nocow_ctx *nocow_ctx =
@@ -3167,17 +3263,43 @@ static void copy_nocow_pages_worker(struct btrfs_work *work)
 	}
 
 	ret = iterate_inodes_from_logical(logical, fs_info, path,
-					  copy_nocow_pages_for_inode,
-					  nocow_ctx);
+					  record_inode_for_nocow, nocow_ctx);
 	if (ret != 0 && ret != -ENOENT) {
-		pr_warn("iterate_inodes_from_logical() failed: log %llu, phys %llu, len %llu, mir %u, ret %d\n",
+		btrfs_warn(fs_info, "iterate_inodes_from_logical() failed: log %llu, "
+			"phys %llu, len %llu, mir %u, ret %d",
 			logical, physical_for_dev_replace, len, mirror_num,
 			ret);
 		not_written = 1;
 		goto out;
 	}
 
+	btrfs_end_transaction(trans, root);
+	trans = NULL;
+	while (!list_empty(&nocow_ctx->inodes)) {
+		struct scrub_nocow_inode *entry;
+		entry = list_first_entry(&nocow_ctx->inodes,
+					 struct scrub_nocow_inode,
+					 list);
+		list_del_init(&entry->list);
+		ret = copy_nocow_pages_for_inode(entry->inum, entry->offset,
+						 entry->root, nocow_ctx);
+		kfree(entry);
+		if (ret == COPY_COMPLETE) {
+			ret = 0;
+			break;
+		} else if (ret) {
+			break;
+		}
+	}
 out:
+	while (!list_empty(&nocow_ctx->inodes)) {
+		struct scrub_nocow_inode *entry;
+		entry = list_first_entry(&nocow_ctx->inodes,
+					 struct scrub_nocow_inode,
+					 list);
+		list_del_init(&entry->list);
+		kfree(entry);
+	}
 	if (trans && !IS_ERR(trans))
 		btrfs_end_transaction(trans, root);
 	if (not_written)
@@ -3190,20 +3312,25 @@ out:
 	scrub_pending_trans_workers_dec(sctx);
 }
 
-static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root, void *ctx)
+static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root,
+				      struct scrub_copy_nocow_ctx *nocow_ctx)
 {
-	struct scrub_copy_nocow_ctx *nocow_ctx = ctx;
 	struct btrfs_fs_info *fs_info = nocow_ctx->sctx->dev_root->fs_info;
 	struct btrfs_key key;
 	struct inode *inode;
 	struct page *page;
 	struct btrfs_root *local_root;
+	struct btrfs_ordered_extent *ordered;
+	struct extent_map *em;
+	struct extent_state *cached_state = NULL;
+	struct extent_io_tree *io_tree;
 	u64 physical_for_dev_replace;
-	u64 len;
+	u64 len = nocow_ctx->len;
+	u64 lockstart = offset, lockend = offset + len - 1;
 	unsigned long index;
 	int srcu_index;
-	int ret;
-	int err;
+	int ret = 0;
+	int err = 0;
 
 	key.objectid = root;
 	key.type = BTRFS_ROOT_ITEM_KEY;
@@ -3229,15 +3356,39 @@ static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root, void *ctx)
 	mutex_lock(&inode->i_mutex);
 	inode_dio_wait(inode);
 
-	ret = 0;
 	physical_for_dev_replace = nocow_ctx->physical_for_dev_replace;
-	len = nocow_ctx->len;
+	io_tree = &BTRFS_I(inode)->io_tree;
+
+	lock_extent_bits(io_tree, lockstart, lockend, 0, &cached_state);
+	ordered = btrfs_lookup_ordered_range(inode, lockstart, len);
+	if (ordered) {
+		btrfs_put_ordered_extent(ordered);
+		goto out_unlock;
+	}
+
+	em = btrfs_get_extent(inode, NULL, 0, lockstart, len, 0);
+	if (IS_ERR(em)) {
+		ret = PTR_ERR(em);
+		goto out_unlock;
+	}
+
+	/*
+	 * This extent does not actually cover the logical extent anymore,
+	 * move on to the next inode.
+	 */
+	if (em->block_start > nocow_ctx->logical ||
+	    em->block_start + em->block_len < nocow_ctx->logical + len) {
+		free_extent_map(em);
+		goto out_unlock;
+	}
+	free_extent_map(em);
+
 	while (len >= PAGE_CACHE_SIZE) {
 		index = offset >> PAGE_CACHE_SHIFT;
 again:
 		page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
 		if (!page) {
-			pr_err("find_or_create_page() failed\n");
+			btrfs_err(fs_info, "find_or_create_page() failed");
 			ret = -ENOMEM;
 			goto out;
 		}
@@ -3247,10 +3398,9 @@ again:
 				goto next_page;
 		} else {
 			ClearPageError(page);
-			err = extent_read_full_page(&BTRFS_I(inode)->
-							 io_tree,
-							page, btrfs_get_extent,
-							nocow_ctx->mirror_num);
+			err = extent_read_full_page_nolock(io_tree, page,
+							   btrfs_get_extent,
+							   nocow_ctx->mirror_num);
 			if (err) {
 				ret = err;
 				goto next_page;
@@ -3264,6 +3414,7 @@ again:
 			 * page in the page cache.
 			 */
 			if (page->mapping != inode->i_mapping) {
+				unlock_page(page);
 				page_cache_release(page);
 				goto again;
 			}
@@ -3287,6 +3438,10 @@ next_page:
 		physical_for_dev_replace += PAGE_CACHE_SIZE;
 		len -= PAGE_CACHE_SIZE;
 	}
+	ret = COPY_COMPLETE;
+out_unlock:
+	unlock_extent_cached(io_tree, lockstart, lockend, &cached_state,
+			     GFP_NOFS);
 out:
 	mutex_unlock(&inode->i_mutex);
 	iput(inode);
@@ -3299,14 +3454,13 @@ static int write_page_nocow(struct scrub_ctx *sctx,
 	struct bio *bio;
 	struct btrfs_device *dev;
 	int ret;
-	DECLARE_COMPLETION_ONSTACK(compl);
 
 	dev = sctx->wr_ctx.tgtdev;
 	if (!dev)
 		return -EIO;
 	if (!dev->bdev) {
 		printk_ratelimited(KERN_WARNING
-			"btrfs: scrub write_page_nocow(bdev == NULL) is unexpected!\n");
+			"BTRFS: scrub write_page_nocow(bdev == NULL) is unexpected!\n");
 		return -EIO;
 	}
 	bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
@@ -3316,10 +3470,8 @@ static int write_page_nocow(struct scrub_ctx *sctx,
 		spin_unlock(&sctx->stat_lock);
 		return -ENOMEM;
 	}
-	bio->bi_private = &compl;
-	bio->bi_end_io = scrub_complete_bio_end_io;
-	bio->bi_size = 0;
-	bio->bi_sector = physical_for_dev_replace >> 9;
+	bio->bi_iter.bi_size = 0;
+	bio->bi_iter.bi_sector = physical_for_dev_replace >> 9;
 	bio->bi_bdev = dev->bdev;
 	ret = bio_add_page(bio, page, PAGE_CACHE_SIZE, 0);
 	if (ret != PAGE_CACHE_SIZE) {
@@ -3328,10 +3480,8 @@ leave_with_eio:
 		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS);
 		return -EIO;
 	}
-	btrfsic_submit_bio(WRITE_SYNC, bio);
-	wait_for_completion(&compl);
 
-	if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
+	if (btrfsic_submit_bio_wait(WRITE_SYNC, bio))
 		goto leave_with_eio;
 
 	bio_put(bio);