On Tue, Nov 12, 2019 at 08:41:50PM +0800, Qu Wenruo wrote:
>
>
> On 2019/11/12 下午8:24, Johannes Thumshirn wrote:
> > Now that the preparation work is done, remove the temporary BUG_ON() in
> > close_fs_devices() and return an error instead.
> >
> > Signed-off-by: Johannes Thumshirn <jthumshirn@xxxxxxx>
> > ---
> > fs/btrfs/volumes.c | 7 ++++++-
> > 1 file changed, 6 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> > index be1fd935edf7..844333b96075 100644
> > --- a/fs/btrfs/volumes.c
> > +++ b/fs/btrfs/volumes.c
> > @@ -1128,7 +1128,12 @@ static int close_fs_devices(struct btrfs_fs_devices *fs_devices)
> > mutex_lock(&fs_devices->device_list_mutex);
> > list_for_each_entry_safe(device, tmp, &fs_devices->devices, dev_list) {
> > ret = btrfs_close_one_device(device);
> > - BUG_ON(ret); /* -ENOMEM */
> > + if (ret) {
> > + mutex_unlock(&fs_devices->device_list_mutex);
> > + return ret;
> > + }
> > + fs_devices->opened--;
> > + fs_devices->seeding--;
>
> This seeding-- doesn't look safe to me.
Yeah, same here, it could be correct in the sense that it's 1 -> 0
exactly once, but otherwise its a bool, and handled in a special way.
