Re: frequent lockups in 3.18rc4

From: Dave Jones
Date: Sat Dec 13 2014 - 15:42:46 EST


On Sat, Dec 13, 2014 at 10:04:08AM -0800, Paul E. McKenney wrote:
> On Sat, Dec 13, 2014 at 11:59:15AM -0500, Dave Jones wrote:
> > On Fri, Dec 12, 2014 at 11:14:06AM -0800, Linus Torvalds wrote:
> > > On Fri, Dec 12, 2014 at 10:54 AM, Dave Jones <davej@xxxxxxxxxx> wrote:
> > > >
> > > > Something that's still making me wonder if it's some kind of hardware
> > > > problem is the non-deterministic nature of this bug.
> > >
> > > I'd expect it to be a race condition, though. Which can easily cause
> > > these kinds of issues, and the timing will be pretty random even if
> > > the load is very regular.
> > >
> > > And we know that the scheduler has an integer overflow under Sasha's
> > > loads, although I didn't hear anything from Ingo and friends about it.
> > > Ingo/Peter, you were cc'd on that report, where at least one of the
> > > multiplcations in wake_affine() ended up overflowing..
> > >
> > > Some scheduler thing that overflows only under heavy load, and screws
> > > up scheduling could easily account for the RCU thread thing. I see it
> > > *less* easily accounting for DaveJ's case, though, because the
> > > watchdog is running at RT priority, and the scheduler would have to
> > > screw up much more to then not schedule an RT task, but..
> > >
> > > I'm also not sure if the bug ever happens with preemption disabled.
> >
> > Bah, so I see some watchdog traces with preemption off, and that then
> > taints the kernel, and the fuzzing stops. I'll hack something up
> > so it ignores the taint and keeps going. All I really care about here
> > is the "machine hangs completely" case, which the trace below didn't
> > hit..
> >
> > (back to fuzzing almost everything, not just lsetxattr btw)
>
> Hmmm... This one looks like the RCU grace-period kthread is getting
> starved: "idle=b4c/0/0". Is this running with the "dangerous" patch
> that sets these kthreads to RT priority?

sorry, no. Ran out of time yesterday. I'll try and get to applying that
later this evening if I get chance.

Dave
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/