Re: [syzbot] [io-uring?] general protection fault in io_sqe_buffer_register

[Jens Axboe <axboe@kernel.dk>]

On 12/4/24 10:34 AM, Aleksandr Nogikh wrote:
> On Wed, Dec 4, 2024 at 6:14 PM Jens Axboe <axboe@kernel.dk> wrote:
>>
>> On 12/4/24 10:11 AM, Aleksandr Nogikh wrote:
>>> Hi Jens,
>>>
>>> Just in case:
>>>
>>> Syzbot reported this commit as the result of the cause (bug origin)
>>> bisection, not as the commit after which the problem was gone. So
>>> (unless it actually is a fixing commit) reporting it back via #syz fix
>>> is not correct.
>>
>> The commit got fixed, and hence there isn't a good way to convey this
>> to syzbot as far as I can tell. Just marking the updated one as the
>> fixer seems to be the best/closest option.
>>
>> Other option is to mark it as invalid, but that also doesn't seem right.
>>
>> I'm fine doing whatever to get issues like this closed, but it's not
>> an uncommon thing to have a buggy commit that's not upstream yet be
>> fixed up and hence not have the issue anymore.
> 
> I see. You are right, thanks for the explanation!
> 
> There's indeed no better way to convey this at the moment. I've filed
> https://github.com/google/syzkaller/issues/5567 to discuss what can be
> done.

Thanks! Guess I wasn't totally blind, I did check to see if there was
a better way to do this currently and didn't spot it.

-- 
Jens Axboe