4 files changed, 54 insertions, 20 deletions

diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index d96561d1ce90..6934a5b8708f 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -2017,8 +2017,17 @@ static u64 update_block_group_flags(struct btrfs_fs_info *fs_info, u64 flags)
 	return flags;
 }
 
-int btrfs_inc_block_group_ro(struct btrfs_block_group *cache)
-
+/*
+ * Mark one block group RO, can be called several times for the same block
+ * group.
+ *
+ * @cache:		the destination block group
+ * @do_chunk_alloc:	whether need to do chunk pre-allocation, this is to
+ * 			ensure we still have some free space after marking this
+ * 			block group RO.
+ */
+int btrfs_inc_block_group_ro(struct btrfs_block_group *cache,
+			     bool do_chunk_alloc)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	struct btrfs_trans_handle *trans;
@@ -2048,25 +2057,29 @@ again:
 		goto again;
 	}
 
-	/*
-	 * if we are changing raid levels, try to allocate a corresponding
-	 * block group with the new raid level.
-	 */
-	alloc_flags = update_block_group_flags(fs_info, cache->flags);
-	if (alloc_flags != cache->flags) {
-		ret = btrfs_chunk_alloc(trans, alloc_flags, CHUNK_ALLOC_FORCE);
+	if (do_chunk_alloc) {
 		/*
-		 * ENOSPC is allowed here, we may have enough space
-		 * already allocated at the new raid level to
-		 * carry on
+		 * If we are changing raid levels, try to allocate a
+		 * corresponding block group with the new raid level.
 		 */
-		if (ret == -ENOSPC)
-			ret = 0;
-		if (ret < 0)
-			goto out;
+		alloc_flags = update_block_group_flags(fs_info, cache->flags);
+		if (alloc_flags != cache->flags) {
+			ret = btrfs_chunk_alloc(trans, alloc_flags,
+						CHUNK_ALLOC_FORCE);
+			/*
+			 * ENOSPC is allowed here, we may have enough space
+			 * already allocated at the new raid level to carry on
+			 */
+			if (ret == -ENOSPC)
+				ret = 0;
+			if (ret < 0)
+				goto out;
+		}
 	}
 
-	ret = inc_block_group_ro(cache, 0);
+	ret = inc_block_group_ro(cache, !do_chunk_alloc);
+	if (!do_chunk_alloc)
+		goto unlock_out;
 	if (!ret)
 		goto out;
 	alloc_flags = btrfs_get_alloc_profile(fs_info, cache->space_info->flags);
@@ -2081,6 +2094,7 @@ out:
 		check_system_chunk(trans, alloc_flags);
 		mutex_unlock(&fs_info->chunk_mutex);
 	}
+unlock_out:
 	mutex_unlock(&fs_info->ro_block_group_mutex);
 
 	btrfs_end_transaction(trans);
diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h
index 4e7afc028791..9b409676c4b2 100644
--- a/fs/btrfs/block-group.h
+++ b/fs/btrfs/block-group.h
@@ -206,7 +206,8 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info);
 int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 			   u64 type, u64 chunk_offset, u64 size);
 void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans);
-int btrfs_inc_block_group_ro(struct btrfs_block_group *cache);
+int btrfs_inc_block_group_ro(struct btrfs_block_group *cache,
+			     bool do_chunk_alloc);
 void btrfs_dec_block_group_ro(struct btrfs_block_group *cache);
 int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans);
 int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans);
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index a857fc8271d2..2e16701c6099 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -4325,7 +4325,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 	rc->extent_root = extent_root;
 	rc->block_group = bg;
 
-	ret = btrfs_inc_block_group_ro(rc->block_group);
+	ret = btrfs_inc_block_group_ro(rc->block_group, true);
 	if (ret) {
 		err = ret;
 		goto out;
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index e2c87220600f..bd3f2266c5f4 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -3559,7 +3559,26 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		 * -> btrfs_scrub_pause()
 		 */
 		scrub_pause_on(fs_info);
-		ret = btrfs_inc_block_group_ro(cache);
+
+		/*
+		 * Don't do chunk preallocation for scrub.
+		 *
+		 * This is especially important for SYSTEM bgs, or we can hit
+		 * -EFBIG from btrfs_finish_chunk_alloc() like:
+		 * 1. The only SYSTEM bg is marked RO.
+		 *    Since SYSTEM bg is small, that's pretty common.
+		 * 2. New SYSTEM bg will be allocated
+		 *    Due to regular version will allocate new chunk.
+		 * 3. New SYSTEM bg is empty and will get cleaned up
+		 *    Before cleanup really happens, it's marked RO again.
+		 * 4. Empty SYSTEM bg get scrubbed
+		 *    We go back to 2.
+		 *
+		 * This can easily boost the amount of SYSTEM chunks if cleaner
+		 * thread can't be triggered fast enough, and use up all space
+		 * of btrfs_super_block::sys_chunk_array
+		 */
+		ret = btrfs_inc_block_group_ro(cache, false);
 		if (!ret && sctx->is_dev_replace) {
 			/*
 			 * If we are doing a device replace wait for any tasks