Re: INFO: rcu_bh detected stall on CPU 3 (t=0 jiffies)

From: Paul E. McKenney
Date: Fri Jan 18 2013 - 12:44:25 EST


On Sun, Jan 13, 2013 at 04:13:51AM +0000, Ben Hutchings wrote:
> On Wed, 2013-01-02 at 19:49 -0500, Steven Rostedt wrote:
> > On Wed, 2013-01-02 at 13:29 -0800, Paul E. McKenney wrote:
> > > On Wed, Jan 02, 2013 at 03:56:30PM -0500, Steven Rostedt wrote:
> > > > On Wed, Jan 02, 2013 at 08:09:20PM +0100, Ralf Hildebrandt wrote:
> > > > > This happened of a virtual guest (Linux mail 3.2.0-35-virtual #55-Ubuntu SMP Wed Dec 5 18:02:05 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux)
> > > >
> > > > 3.2 is rather old and this looks to be a stock Ubuntu kernel. May want
> > > > to file a Bugzilla (or whatever they call it) with them.
> > > >
> > > > >
> > > > > Jan 2 13:24:08 mail kernel: [3531872.484283] INFO: rcu_bh detected stall on CPU 3 (t=0 jiffies)
> > > >
> > > > 0 jiffies?
> > >
> > > One possibility is that they are missing commit a10d206e (rcu: Fix
> > > day-one dyntick-idle stall-warning bug), at least assuming that
> > > CONFIG_PREEMPT=n. This went into 3.6, IIRC.
> >
> > If that does fix the issue, we probably should send that off to Ben for
> > inclusion into 3.2 stable.
>
> If I read the Ubuntu changelog correctly, Ralf's machine is running a
> version based on 3.2.34, which already has this change (I applied it in
> 3.2.32).

OK, another possibility is c8020a67. Ralf, if that does not help, is
it possible for you to run either a 3.7 or a -rcu kernel [*]? There are
some other fixes that could be related, but the probability seems low.

Thanx, Paul

[*] git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git
branch rcu/next

--
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/