Re: NOHZ: WARNING: at arch/x86/kernel/smp.c:123native_smp_send_reschedule
From: Borislav Petkov
Date: Thu May 09 2013 - 08:47:23 EST
On Thu, May 09, 2013 at 02:29:18PM +0200, Jiri Kosina wrote:
> Hi,
>
> I just got the warning below when resuming from hibernation with kernel
> that has NO_HZ_FULL_ALL=y. This is with topmost commit e0fd9affeb640.
Did you boot with any of the NO_HZ_FULL options on the command line,
i.e. rcu_nocbs?
Because I saw the same issue during boot when I was testing the
NO_HZ_FULL stuff previously:
Feb 21 11:13:17 gondor kernel: [ 0.093417] #3
Feb 21 11:13:17 gondor kernel: [ 0.106844] SMP alternatives: lockdep: fixing up alternatives
Feb 21 11:13:17 gondor kernel: [ 0.118060] ------------[ cut here ]------------
Feb 21 11:13:17 gondor kernel: [ 0.118069] WARNING: at arch/x86/kernel/smp.c:123 native_smp_send_reschedule+0x58/0x60()
Feb 21 11:13:17 gondor kernel: [ 0.118069] Hardware name: Precision T3600
Feb 21 11:13:17 gondor kernel: [ 0.106955] #4
Feb 21 11:13:17 gondor kernel: [ 0.118070] Modules linked in:
Feb 21 11:13:17 gondor kernel: [ 0.118073] Pid: 0, comm: swapper/4 Not tainted 3.8.0-rc6+ #4
Feb 21 11:13:17 gondor kernel: [ 0.118074] Call Trace:
Feb 21 11:13:17 gondor kernel: [ 0.118076] [<ffffffff8103d35f>] warn_slowpath_common+0x7f/0xc0
Feb 21 11:13:17 gondor kernel: [ 0.118079] [<ffffffff8103d3ba>] warn_slowpath_null+0x1a/0x20
Feb 21 11:13:17 gondor kernel: [ 0.118081] [<ffffffff81024a58>] native_smp_send_reschedule+0x58/0x60
Feb 21 11:13:17 gondor kernel: [ 0.118083] [<ffffffff81071ba0>] wake_up_nohz_cpu+0x80/0x90
Feb 21 11:13:17 gondor kernel: [ 0.118086] [<ffffffff8104e611>] add_timer_on+0x91/0x110
Feb 21 11:13:17 gondor kernel: [ 0.118090] [<ffffffff8101ccab>] mce_start_timer.isra.12+0x6b/0x80
Feb 21 11:13:17 gondor kernel: [ 0.118092] [<ffffffff8101cd1d>] __mcheck_cpu_init_timer+0x5d/0x70
Feb 21 11:13:17 gondor kernel: [ 0.118094] [<ffffffff815cc466>] mcheck_cpu_init+0x36b/0x400
Feb 21 11:13:17 gondor kernel: [ 0.118097] [<ffffffff815ca15b>] identify_cpu+0x39d/0x3d0
Feb 21 11:13:17 gondor kernel: [ 0.118099] [<ffffffff815ca1a2>] identify_secondary_cpu+0x14/0x1b
Feb 21 11:13:17 gondor kernel: [ 0.118101] [<ffffffff815cd008>] smp_store_cpu_info+0x38/0x3a
Feb 21 11:13:17 gondor kernel: [ 0.118103] [<ffffffff815cd550>] start_secondary+0xfb/0x1f7
Feb 21 11:13:17 gondor kernel: [ 0.118121] ---[ end trace 8ffeaaf1f7048154 ]---
And I *think* I saw this yesterday too on a randbuild kernel.
Looks like we're sending a resched IPI to a cpu which is not online yet
in order to start the MCE polling timer. So the rcu* options are kinda
unlikely to be related, AFAICT.
Frederic?
Thanks.
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
--
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/