Re: BUG: using smp_processor_id() in preemptible [00000000] code:latencytop/5621

From: Michal Hocko
Date: Tue Nov 18 2008 - 07:37:59 EST


On Sun 16-11-08 00:28:12, Rafael J. Wysocki wrote:
> On Friday, 14 of November 2008, Michal Hocko wrote:
> > Hi Ingo,
> > I am experiencing %subj whenever I run latencytop with the rc4 kernel. I
> > have bisected down to the patch 7cbaef9c83e58bbd4bdd534b09052b6c5ec457d5
> > I am not sure whether this is the same problem as the one reported by
> > Maciej Rutecki (http://lkml.org/lkml/2008/11/12/266), but I have tried
> > your patch (http://lkml.org/lkml/2008/11/12/295) and the problem is
> > still there:
> > BUG: using smp_processor_id() in preemptible [00000000] code:
> > latencytop/5621
> > caller is native_sched_clock+0x47/0x65
> > Pid: 5621, comm: latencytop Not tainted 2.6.28-rc4 #13
> > Call Trace:
> > [<c03f17c0>] ? printk+0xf/0x11
> > [<c022ad36>] debug_smp_processor_id+0xa2/0xb8
> > [<c01080d4>] native_sched_clock+0x47/0x65
> > [<c01259a9>] proc_sched_show_task+0xd72/0xdd0
> > [<c01a73c4>] sched_show+0x39/0x55
> > [<c018cc43>] seq_read+0xee/0x29a
> > [<c018cb55>] ? seq_read+0x0/0x29a
> > [<c017a04c>] vfs_read+0x87/0x12b
> > [<c017a189>] sys_read+0x3b/0x60
> > [<c0103a59>] sysenter_do_call+0x12/0x25
> > [<c03f0000>] ? cpu_callback+0x238/0x25b
> >
> > I receive flood if this messages immediately after latencytop starts.
> > Reverting 7cbaef9c83e58bbd4bdd534b09052b6c5ec457d5 solves the problem.
> > Config is attached.
> >
> > Let me know if you need some more testing (however I will afk until
> > Tuesday).
>
> Please retest with 2.6.28-rc5.

Still problem with rc5 (4e14e833ac3b97a4aa8803eea49f899adc5bb5f4).
Ingo's patch from http://lkml.org/lkml/2008/11/16/26 fixes this problem

>
> Thanks,
> Rafael

--
Michal Hocko
L3 team
SUSE LINUX s.r.o.
Lihovarska 1060/12
190 00 Praha 9
Czech Republic
--
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/