Btrfs: fix unprotected device->bytes_used update

We should update device->bytes_used in the lock context of
chunk_mutex, or we would get wrong data.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 43530c46cc55cf5a381e788276a93c83cbc95e46)

diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 43ec093e3765a8234984bebc80ae9182a4300699..73295f69bdfac6835fdb4282fc40d38688be7abf 100644 (file)
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -4426,6 +4426,9 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
        if (ret)
                goto error_del_extent;
 
+       for (i = 0; i < map->num_stripes; i++)
+               map->stripes[i].dev->bytes_used += stripe_size;
+
        free_extent_map(em);
        check_raid56_incompat_flag(extent_root->fs_info, type);
 
@@ -4497,7 +4500,6 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
                device = map->stripes[i].dev;
                dev_offset = map->stripes[i].physical;
 
-               device->bytes_used += stripe_size;
                ret = btrfs_update_device(trans, device);
                if (ret)
                        goto out;