Re: mm snapshot broken-out-2006-08-24-00-22.tar.gz uploaded

From: Andrew Morton
Date: Thu Aug 24 2006 - 10:35:37 EST


On Thu, 24 Aug 2006 13:52:28 +0200
"Michal Piotrowski" <michal.k.k.piotrowski@xxxxxxxxx> wrote:

> Hi,
>
> On 24/08/06, akpm@xxxxxxxx <akpm@xxxxxxxx> wrote:
> > The mm snapshot broken-out-2006-08-24-00-22.tar.gz has been uploaded to
> >
> > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/mm/broken-out-2006-08-24-00-22.tar.gz
> >
> > It contains the following patches against 2.6.18-rc4:
>
> I haven't seen this on 2.6.18-rc4-mm2. Any idea which patch is causing
> this soft lockup?
>
> BUG: soft lockup detected on CPU#0!
> [<c01041b5>] dump_trace+0x64/0x1b2
> [<c0104315>] show_trace_log_lvl+0x12/0x25
> [<c0104985>] show_trace+0xd/0x10
> [<c0104a4d>] dump_stack+0x19/0x1b
> [<c014f40d>] softlockup_tick+0xc5/0xd9
> [<c012a0c7>] run_local_timers+0x12/0x14
> [<c012a329>] update_process_times+0x40/0x65
> [<c0113cc7>] smp_apic_timer_interrupt+0x6e/0x77
> [<c0103ce2>] apic_timer_interrupt+0x2a/0x30
> DWARF2 unwinder stuck at apic_timer_interrupt+0x2a/0x30
> Leftover inexact backtrace:

Sigh, useless.

> [<c0104315>] show_trace_log_lvl+0x12/0x25
> [<c0104985>] show_trace+0xd/0x10
> [<c0104a4d>] dump_stack+0x19/0x1b
> [<c014f40d>] softlockup_tick+0xc5/0xd9
> [<c012a0c7>] run_local_timers+0x12/0x14
> [<c012a329>] update_process_times+0x40/0x65
> [<c0113cc7>] smp_apic_timer_interrupt+0x6e/0x77
> [<c0103ce2>] apic_timer_interrupt+0x2a/0x30
> [<c01f3010>] __delay+0x9/0xb
> [<c01ffc47>] _raw_spin_lock+0xca/0x11d
> [<c02f86a8>] _spin_lock+0x2a/0x32
> [<c013cc4e>] do_futex+0xee/0xd9a
> [<c013d9d4>] sys_futex+0xda/0xf0
> [<c01031b5>] sysenter_past_esp+0x56/0x8d


There are a couple of recent futex changes, but they're fairly
innocent-looking. What is the workload?

-
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/