Re: [Announce] 2.6.29-rc8-rt2

From: Fernando Lopez-Lezcano
Date: Mon Mar 23 2009 - 23:13:00 EST


On Mon, 2009-03-23 at 09:07 +0100, Ingo Molnar wrote:
> * Fernando Lopez-Lezcano <nando@xxxxxxxxxxxxxxxxxx> wrote:
> > On Sun, 2009-03-22 at 14:09 +0100, Ingo Molnar wrote:
> > > * Fernando Lopez-Lezcano <nando@xxxxxxxxxxxxxxxxxx> wrote:
> > I hit another problem. On a quad code intel machine here at home there
> > is a problem during shutdown. The final shutdown dies with a traceback
> > (something like "rc0 was killed by SIGSEGV" or similar words at the very
> > end). Then I have to push the power button for x+ seconds to force a
> > power off.
> >
> > Should I post a picture of the screen as it is at that point, or just
> > send that to you and Thomas?
>
> any kernel stack dump in that?

The tail end of it.
(still happening in rc8-rt4)

> You could try to do this before shutting down:
>
> echo 1 > /proc/sys/kernel/print-fatal-signals
> dmesg -n 8
>
> that way there's a (hopefully helpful) stack dump done of the
> SIGSEGV itself.

without all the funny numbers this is what remains on screen:

native_cpu_die
_cpu_down
disable_nonboot_cpus
kernel_power_off
sys_reboot
__hrtimer_start_range_ns
default_spin_lock_flags
__spin_lock_irqsave
__spin_lock_irqrestore
hrtimer_try_to_cancel
hrtimer_cancel
do_nanosleep
hrtimer_nanosleep
hrtimer_wakeup
path_put
audit_syscall_entry
audit_syscall_exit
sysenter_do_call
Code: 00 00 6a 00 ff 15 08 49 85 c0 58 etc etc etc
EIP [<c06f2549>] text_poke+0x163/0x181 SS:ESP 0068:f4567ddc
-- [end trace db3e80122adba82c ]---
init: rc0 main process (3047) killed by SEGV signal

(I can take a picture if that would help - I don't have a second machine
with a serial port to try to capture the whole thing)

BTW, the hard disk powers down and everything else seems to complete.

-- Fernando


--
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/