Re: [PATCH] Btrfs: don't miss roots in deletion during walking backrefs

[Wang Shilong <wangsl.fnst@cn.fujitsu.com>]

Please ignore this patch, there is still a race condition between
root deletion and walking backrefs, i will send them together later!

Thanks,
Wang
On 01/14/2014 08:36 PM, Wang Shilong wrote:
> btrfs_read_fs_root_no_name() will check its root refs, and will
> return us ENOENT if this root in deletion.
>
> However, in some case, for example qgroup, we still need to track
> those roots, fix it.
>
> Signed-off-by: Wang Shilong<wangsl.fnst@cn.fujitsu.com>
> ---
>   fs/btrfs/backref.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
> index 835b6c9..ec4a1d2 100644
> --- a/fs/btrfs/backref.c
> +++ b/fs/btrfs/backref.c
> @@ -303,7 +303,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
>   	root_key.objectid = ref->root_id;
>   	root_key.type = BTRFS_ROOT_ITEM_KEY;
>   	root_key.offset = (u64)-1;
> -	root = btrfs_read_fs_root_no_name(fs_info, &root_key);
> +	root = btrfs_get_fs_root(fs_info, &root_key, false);
>   	if (IS_ERR(root)) {
>   		ret = PTR_ERR(root);
>   		goto out;