Re: [syzbot] WARNING in __percpu_ref_exit (2)

[Pavel Begunkov]

On 9/16/21 8:59 AM, Dmitry Vyukov wrote:
> On Mon, 13 Sept 2021 at 11:22, syzbot
> <syzbot+d6218cb2fae0b2411e9d@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>>
>> syzbot suspects this issue was fixed by commit:
>>
>> commit 43016d02cf6e46edfc4696452251d34bba0c0435
>> Author: Florian Westphal <fw@xxxxxxxxx>
>> Date:   Mon May 3 11:51:15 2021 +0000
>>
>>     netfilter: arptables: use pernet ops struct during unregister
>>
>> bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=10acd273300000
>> start commit:   c98ff1d013d2 Merge tag 'scsi-fixes' of git://git.kernel.or..
>> git tree:       upstream
>> kernel config:  https://syzkaller.appspot.com/x/.config?x=1c70e618af4c2e92
>> dashboard link: https://syzkaller.appspot.com/bug?extid=d6218cb2fae0b2411e9d
>> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=145cb2b6d00000
>> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=157b72b1d00000
>>
>> If the result looks correct, please mark the issue as fixed by replying with:
>>
>> #syz fix: netfilter: arptables: use pernet ops struct during unregister
>>
>> For information about bisection process see: https://goo.gl/tpsmEJ#bisection
> 
> I guess this is a wrong commit and it was fixed by something in io_uring.
> Searching for refcount fixes I see
> a298232ee6b9a1d5d732aa497ff8be0d45b5bd82 "io_uring: fix link timeout
> refs".
> Pavel, does it look right to you?

I don't remember to be honest, if the dates fit, it can pretty well be it.
Let's test one thing to be sure it hasn't been shut just by coincidence.

#syz test: https://github.com/isilence/linux.git syz_test_quiesce_files


-- 
Pavel Begunkov