Re: [syzbot] [bluetooth?] general protection fault in hci_uart_tty_ioctl

Re: [syzbot] [bluetooth?] general protection fault in hci_uart_tty_ioctl

[syzbot]

syzbot suspects this issue was fixed by commit:

commit d772781964415c63759572b917e21c4f7ec08d9f
Author: Jakub Kicinski <kuba@kernel.org>
Date:   Fri Jan 6 06:33:54 2023 +0000

    devlink: bump the instance index directly when iterating

bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=136a0414a80000
start commit:   84368d882b96 Merge tag 'soc-fixes-6.1-3' of git://git.kern..
git tree:       upstream
kernel config:  https://syzkaller.appspot.com/x/.config?x=24d192d47d02d9e1
dashboard link: https://syzkaller.appspot.com/bug?extid=c19afa60d78984711078
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=168fc765880000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=1376e745880000

If the result looks correct, please mark the issue as fixed by replying with:

#syz fix: devlink: bump the instance index directly when iterating

For information about bisection process see: https://goo.gl/tpsmEJ#bisection

Re: [syzbot] [bluetooth?] general protection fault in hci_uart_tty_ioctl

[Jakub Kicinski]

On Mon, 17 Jul 2023 13:22:24 -0700 syzbot wrote:
> syzbot suspects this issue was fixed by commit:
> 
> commit d772781964415c63759572b917e21c4f7ec08d9f
> Author: Jakub Kicinski <kuba@kernel.org>
> Date:   Fri Jan 6 06:33:54 2023 +0000
> 
>     devlink: bump the instance index directly when iterating

Hm, don't think so. It's not the first issue where syzbot decided
this commit was the resolution. I wonder what makes it special.

Re: [syzbot] [bluetooth?] general protection fault in hci_uart_tty_ioctl

[Aleksandr Nogikh]

On Tue, Jul 18, 2023 at 9:45 PM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Mon, 17 Jul 2023 13:22:24 -0700 syzbot wrote:
> > syzbot suspects this issue was fixed by commit:
> >
> > commit d772781964415c63759572b917e21c4f7ec08d9f
> > Author: Jakub Kicinski <kuba@kernel.org>
> > Date:   Fri Jan 6 06:33:54 2023 +0000
> >
> >     devlink: bump the instance index directly when iterating
>
> Hm, don't think so. It's not the first issue where syzbot decided
> this commit was the resolution. I wonder what makes it special.

Judging by the bisection log, the commit fixed some "INFO: rcu
detected stall in" error that was also triggerable by the reproducer.
Though for me it's not clear how exactly -- at least the reproducer
does not seem to be interacting with devlink..

-- 
Aleksandr


>
> --