[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CANp29Y7O8tgw=NLydjYq=Wa4CrfT1NsidhrOByW6He9qS4LpmQ@mail.gmail.com>
Date: Wed, 13 Nov 2024 12:04:25 +0100
From: Aleksandr Nogikh <nogikh@...gle.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: syzbot <syzbot+b506de56cbbb63148c33@...kaller.appspotmail.com>,
andrii@...nel.org, ast@...nel.org, bpf@...r.kernel.org, daniel@...earbox.net,
eddyz87@...il.com, frederic@...nel.org, haoluo@...gle.com,
houtao@...weicloud.com, john.fastabend@...il.com, jolsa@...nel.org,
kpsingh@...nel.org, linux-kernel@...r.kernel.org, martin.lau@...ux.dev,
peterz@...radead.org, sdf@...ichev.me, song@...nel.org,
syzkaller-bugs@...glegroups.com, yonghong.song@...ux.dev
Subject: Re: [syzbot] [bpf?] WARNING: locking bug in trie_delete_elem
On Tue, Nov 12, 2024 at 3:24 AM Thomas Gleixner <tglx@...utronix.de> wrote:
>
> On Thu, Nov 07 2024 at 13:00, syzbot wrote:
> > syzbot has bisected this issue to:
> >
> > commit 4febce44cfebcb490b196d5d10ae9f403ca4c956
> > Author: Thomas Gleixner <tglx@...utronix.de>
> > Date: Tue Oct 1 08:42:03 2024 +0000
> >
> > posix-timers: Cure si_sys_private race
> >
> > bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=129f2d87980000
> > start commit: f9f24ca362a4 Add linux-next specific files for 20241031
> > git tree: linux-next
> > console output: https://syzkaller.appspot.com/x/log.txt?x=169f2d87980000
> > kernel config: https://syzkaller.appspot.com/x/.config?x=328572ed4d152be9
> > dashboard link: https://syzkaller.appspot.com/bug?extid=b506de56cbbb63148c33
> > syz repro: https://syzkaller.appspot.com/x/repro.syz?x=1387655f980000
> > C reproducer: https://syzkaller.appspot.com/x/repro.c?x=11ac5540580000
> >
> > Reported-by: syzbot+b506de56cbbb63148c33@...kaller.appspotmail.com
> > Fixes: 4febce44cfeb ("posix-timers: Cure si_sys_private race")
> >
> > For information about bisection process see: https://goo.gl/tpsmEJ#bisection
>
> I seriously doubt that this bisection is even remotely correct.
>
> This commit has absolutely nothing to do with the lockdep splat and
> trie_delete_elem().
Yes, the bisection is wrong, please ignore it.
I've added this case to the issue that tracks the underlying problem:
https://github.com/google/syzkaller/issues/5414
--
Aleksandr
>
> Thanks,
>
> tglx
>
> --
Powered by blists - more mailing lists