Re: WARNING: suspicious RCU usage (3)
From: Paul E. McKenney
Date: Wed Nov 29 2017 - 15:51:11 EST
On Wed, Nov 29, 2017 at 07:25:32AM +0100, Dmitry Vyukov wrote:
> On Tue, Nov 28, 2017 at 11:30 PM, Paul E. McKenney
> <paulmck@xxxxxxxxxxxxxxxxxx> wrote:
> > On Tue, Nov 28, 2017 at 01:30:26PM -0800, Andrew Morton wrote:
> >> On Tue, 28 Nov 2017 12:45:01 -0800 syzbot <bot+73a7bec1bc0f4fc0512a246334081f8c671762a8@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
> >>
> >> > Hello,
> >> >
> >> > syzkaller hit the following crash on
> >> > b0a84f19a5161418d4360cd57603e94ed489915e
> >> > git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/master
> >> > compiler: gcc (GCC) 7.1.1 20170620
> >> > .config is attached
> >> > Raw console output is attached.
> >> >
> >> > Unfortunately, I don't have any reproducer for this bug yet.
> >> >
> >> > WARNING: suspicious RCU usage
> >>
> >> There's a bunch of other info which lockdep_rcu_suspicious() should
> >> have printed out, but this trace doesn't have any of it. I wonder why.
> >
> > Yes, there should be more info printed, no idea why it would go missing.
>
> I think that's because while reporting "suspicious RCU usage" kernel hit:
>
> BUG: unable to handle kernel NULL pointer dereference at 0000000000000074
>
> and the rest of the report is actually about the NULL deref.
>
> syzkaller hits too many crashes at the same time. And it's a problem
> for us. We get reports with corrupted/intermixed titles,
> corrupted/intermixed bodies, reports with same titles but about
> different bugs, etc.
Got it, thank you!
Thanx, Paul