Re: [2.6.21.1] soft lockup when removing netconsole module

From: Matt Mackall
Date: Sun May 27 2007 - 16:33:24 EST


On Sat, May 26, 2007 at 06:35:24PM +0200, Thomas Gleixner wrote:
> On Sat, 2007-05-26 at 18:17 +0200, Folkert van Heusden wrote:
> > > > Seems to be hrtimers related - CC'ed Thomas.
> > > I doubt that. The tick interrupt just finds out, that the machine is
> > > stuck in rmmod.
> > > > > Also the rmmod hangs and would not exit even with kill -9. It also
> > > > > sucks up 100% cpu.
> > > Can you please enable CONFIG_PROVE_LOCKING ?
> >
> > Luckily I already had :-)
> > Here is its output:
>
> So there is no output from lockdep, during the rmmod hang ? The rmmod
> sucks 100% cpu, so it busy loops at some place. Looking at the stack
> trace this seems to be in:
>
>
> void cancel_rearming_delayed_workqueue(struct workqueue_struct *wq,
> struct delayed_work *dwork)
> {
> while (!cancel_delayed_work(dwork))
> flush_workqueue(wq);
> }
>
> which is not really related to timers :)

Looks like Steve added the delayed work bits, though I never saw any
of his patches.

--
Mathematics is the supreme nostalgia of our time.
-
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/