On 2.07.20 г. 17:02 ч., David Sterba wrote:
> On Thu, Jul 02, 2020 at 04:46:44PM +0300, Nikolay Borisov wrote:
>> Signed-off-by: Nikolay Borisov <nborisov@xxxxxxxx>
>> ---
>> fs/btrfs/raid56.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
>> index a7ae4d8a47ce..d9415a22617b 100644
>> --- a/fs/btrfs/raid56.c
>> +++ b/fs/btrfs/raid56.c
>> @@ -2093,7 +2093,7 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
>> */
>> if (atomic_read(&rbio->error) <= rbio->bbio->max_errors) {
>> __raid_recover_end_io(rbio);
>> - goto out;
>> + return 0;
>
> No please, when there are labels that do cleanup like the one in the
> context, 'return's make it harder to follow.
>
But I'm not touching the cleanup hand of the if, rather the one which
simply returns 0. SO why jmp + ret when we can straight ret ?
>> } else {
>> goto cleanup;
>> }
>> @@ -2113,7 +2113,7 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
>>
>> submit_bio(bio);
>> }
>> -out:
>> +
>> return 0;
>>
>> cleanup:
>> --
>> 2.17.1
>
