Btrfs: traverse and flush the delalloc inodes once

btrfs_start_delalloc_inodes() needn't traverse and flush the delalloc inodes
repeatedly. It is because we can regard the data that the users write after
we start delalloc inodes flush as the one which is after the delalloc inodes
flush is done, and we can flush it next time.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>

fs/btrfs/inode.c

@@ -7614,7 +7614,7 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
 
 	INIT_LIST_HEAD(&works);
 	INIT_LIST_HEAD(&splice);
-again:
+
 	spin_lock(&root->fs_info->delalloc_lock);
 	list_splice_init(&root->fs_info->delalloc_inodes, &splice);
 	while (!list_empty(&splice)) {
@@ -7650,13 +7650,6 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
 		btrfs_wait_and_free_delalloc_work(work);
 	}
 
-	spin_lock(&root->fs_info->delalloc_lock);
-	if (!list_empty(&root->fs_info->delalloc_inodes)) {
-		spin_unlock(&root->fs_info->delalloc_lock);
-		goto again;
-	}
-	spin_unlock(&root->fs_info->delalloc_lock);
-
 	/* the filemap_flush will queue IO into the worker threads, but
 	 * we have to make sure the IO is actually started and that
 	 * ordered extents get created before we return