On 13/11/2019 16:00, David Sterba wrote:
> On Wed, Nov 13, 2019 at 11:27:25AM +0100, Johannes Thumshirn wrote:
>> close_fs_devices() will be able to return an error instead of crashing
>> after the following patch.
>>
>> Prepare btrfs_close_devices() for this.
>>
>> Signed-off-by: Johannes Thumshirn <jthumshirn@xxxxxxx>
>> ---
>> fs/btrfs/volumes.c | 11 +++++++----
>> 1 file changed, 7 insertions(+), 4 deletions(-)
>>
>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>> index e5864ca3bb3b..be1fd935edf7 100644
>> --- a/fs/btrfs/volumes.c
>> +++ b/fs/btrfs/volumes.c
>> @@ -1143,10 +1143,10 @@ static int close_fs_devices(struct btrfs_fs_devices *fs_devices)
>> int btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
>> {
>> struct btrfs_fs_devices *seed_devices = NULL;
>> - int ret;
>> + int err, err2 = 0;
>
> Please use ret and ret2.
Sure.
>> mutex_lock(&uuid_mutex);
>> - ret = close_fs_devices(fs_devices);
>> + err = close_fs_devices(fs_devices);
>> if (!fs_devices->opened) {
>> seed_devices = fs_devices->seed;
>> fs_devices->seed = NULL;
>> @@ -1156,10 +1156,13 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
>> while (seed_devices) {
>> fs_devices = seed_devices;
>> seed_devices = fs_devices->seed;
>> - close_fs_devices(fs_devices);
>> + err2 = close_fs_devices(fs_devices);
>
> So only the last error value gets propagated to the return statements.
> Is that intentional?
Yes it was. Even if the first close_fs_devices() call fails, we can
still try to close eventual seed devices. If this fails as well, we
return the second error. If it succeeds we'll return the first error (or 0).
Byte,
Johannes
--
Johannes Thumshirn SUSE Labs Filesystems
jthumshirn@xxxxxxx +49 911 74053 689
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5
90409 Nürnberg
Germany
(HRB 36809, AG Nürnberg)
Geschäftsführer: Felix Imendörffer
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
