Re: [syzbot] possible deadlock in console_lock_spinning_enable (2)

[Dmitry Vyukov]

On Mon, 20 Jun 2022 at 11:24, Sergey Senozhatsky
<senozhatsky@xxxxxxxxxxxx> wrote:
>
> On (22/06/20 10:16), Dmitry Vyukov wrote:
> > On Mon, 20 Jun 2022 at 07:44, Sergey Senozhatsky
> > <senozhatsky@xxxxxxxxxxxx> wrote:
> > >
> > > On (22/06/18 20:47), syzbot wrote:
> > > > syzbot suspects this issue was fixed by commit:
> > > >
> > > > commit faebd693c59387b7b765fab64b543855e15a91b4
> > > > Author: John Ogness <john.ogness@xxxxxxxxxxxxx>
> > > > Date:   Thu Apr 21 21:22:36 2022 +0000
> > > >
> > > >     printk: rename cpulock functions
> > >
> > > I'd rather guess that it was console kthread patch that fixed the problem.
> >
> > Hi Sergey,
> >
> > Do you mean this commit:
> >
> > author: John Ogness 2022-04-21 23:28:48 +0206
> > printk: add kthread console printers
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=09c5ba0aa2fcfdadb17d045c3ee6f86d69270df7
>
> Yes, I'd assume so. The patch removes
>
>         tty -> printk -> tty
>
> which re-enters tty in unsafe manner, and replaces it with
>
>         tty -> printk -> wake_up console printer


Thanks. Let's tell syzbot so that it reports similar issues in future:

#syz fix: printk: add kthread console printers