* Date/UnixTime of SysRq state dump
@ 2003-10-01 18:28 Konstantin Kletschke
2003-10-01 18:50 ` Konstantin Kletschke
` (3 more replies)
0 siblings, 4 replies; 9+ messages in thread
From: Konstantin Kletschke @ 2003-10-01 18:28 UTC (permalink / raw)
To: linux-kernel
Hi there!
I know my machine as a rockstable P4 Computer, since
linux-2.6.0-test5-mm4 it freezes once a day. Not here and then, it
happened after long usage from 17:00 or 18:00 around midnight two times
yesterday and before yesterday. Don't laugh, is it possible that a day
change triggers a bug which freezes system? The second time I looked at
my screens at realized, that I saw this picture the day before: Irssi's
last message was "Day changed to ...". I was in heavy chatting, there
was no message even a second after that. Reading that I would laugh, but
it happened two times...
linux-2.6.0-test5-mm3 and older never freezed.
I have the state dump, catched via serial line at
http://ludenkalle.de/capture-2003-09-30-linux-2.6.0-test5-mm4.log
for who is interested, but the Kernel ist tainted by the nvidia module
:(
But this time it is different. Normally, ehen this module is involved,
sound hangs, but keeps on playing in a half a second long loop.
This times all was silent...
Well, if the tainting is not acceptable tell me even if an clock/time
bug in kernel on day change could cause a complete system freeze :)
Konsti
--
2.6.0-test6-mm1
Konstantin Kletschke <konsti@ludenkalle.de>, <konsti@ku-gbr.de>
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E A080 1E69 3FDA EF62 FCEF
keulator.homelinux.org up 1:16, 14 users
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-01 18:28 Date/UnixTime of SysRq state dump Konstantin Kletschke
@ 2003-10-01 18:50 ` Konstantin Kletschke
2003-10-01 19:09 ` Guennadi Liakhovetski
` (2 subsequent siblings)
3 siblings, 0 replies; 9+ messages in thread
From: Konstantin Kletschke @ 2003-10-01 18:50 UTC (permalink / raw)
To: linux-kernel
Err, what I wanted also to point out, can the freeze time decoded out of
the state dump (Was smoking a cig outside when it happened)?
Konsti
--
2.6.0-test6-mm1
Konstantin Kletschke <konsti@ludenkalle.de>, <konsti@ku-gbr.de>
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E A080 1E69 3FDA EF62 FCEF
keulator.homelinux.org up 1:46, 17 users
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-01 18:28 Date/UnixTime of SysRq state dump Konstantin Kletschke
2003-10-01 18:50 ` Konstantin Kletschke
@ 2003-10-01 19:09 ` Guennadi Liakhovetski
2003-10-01 21:32 ` Konstantin Kletschke
2003-10-01 21:48 ` Mike Fedyk
2003-10-01 23:21 ` Andrew Morton
3 siblings, 1 reply; 9+ messages in thread
From: Guennadi Liakhovetski @ 2003-10-01 19:09 UTC (permalink / raw)
To: Konstantin Kletschke; +Cc: linux-kernel
On Wed, 1 Oct 2003, Konstantin Kletschke wrote:
> I know my machine as a rockstable P4 Computer, since
> linux-2.6.0-test5-mm4 it freezes once a day. Not here and then, it
> happened after long usage from 17:00 or 18:00 around midnight two times
> yesterday and before yesterday. Don't laugh, is it possible that a day
> change triggers a bug which freezes system? The second time I looked at
cronjob?
Guennadi
---
Guennadi Liakhovetski
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-01 19:09 ` Guennadi Liakhovetski
@ 2003-10-01 21:32 ` Konstantin Kletschke
0 siblings, 0 replies; 9+ messages in thread
From: Konstantin Kletschke @ 2003-10-01 21:32 UTC (permalink / raw)
To: linux-kernel
* Guennadi Liakhovetski <g.liakhovetski@gmx.de> [Wed, Oct 01, 2003 at 09:09:30PM +0200]:
>
> cronjob?
Hm...
There are only updatedb in cron.daily and rdate -s 192.168.0.194 in
cron.hourly...
Konsti
--
2.6.0-test6-mm1
Konstantin Kletschke <konsti@ludenkalle.de>, <konsti@ku-gbr.de>
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E A080 1E69 3FDA EF62 FCEF
keulator.homelinux.org up 4:28, 26 users
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-01 18:28 Date/UnixTime of SysRq state dump Konstantin Kletschke
2003-10-01 18:50 ` Konstantin Kletschke
2003-10-01 19:09 ` Guennadi Liakhovetski
@ 2003-10-01 21:48 ` Mike Fedyk
2003-10-01 22:44 ` Konstantin Kletschke
2003-10-01 23:21 ` Andrew Morton
3 siblings, 1 reply; 9+ messages in thread
From: Mike Fedyk @ 2003-10-01 21:48 UTC (permalink / raw)
To: linux-kernel
On Wed, Oct 01, 2003 at 08:28:59PM +0200, Konstantin Kletschke wrote:
> for who is interested, but the Kernel ist tainted by the nvidia module
> :(
Reproduce without tainting please...
Really that should have been the first thing to try...
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-01 21:48 ` Mike Fedyk
@ 2003-10-01 22:44 ` Konstantin Kletschke
0 siblings, 0 replies; 9+ messages in thread
From: Konstantin Kletschke @ 2003-10-01 22:44 UTC (permalink / raw)
To: linux-kernel
* Mike Fedyk <mfedyk@matchmail.com> [Wed, Oct 01, 2003 at 02:48:46PM -0700]:
>
> Reproduce without tainting please...
>
> Really that should have been the first thing to try...
Ya, OK. I understand. It is really acquired to reproduce this, when
possible, with an untainted kernel... Otherwise it is not possible to
work, you really do not know what this _huge_ module does...
I updated to 2.6.0-test6-mm1 yesterday, which lifes still after day
change :) and when this freezes I will get rid of the nvidia module to
reproduce that with the nv driver (problem is, only one monitor then!)
and only then I will report.
Really, I only wanted to know if a day change can trigger a kernel
freeze, but, I realize now, one does not know what goes on with that:
nvidia 1709612 10
o_O
Regards, Konsti
--
2.6.0-test6-mm1
Konstantin Kletschke <konsti@ludenkalle.de>, <konsti@ku-gbr.de>
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E A080 1E69 3FDA EF62 FCEF
keulator.homelinux.org up 5:35, 29 users
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-01 18:28 Date/UnixTime of SysRq state dump Konstantin Kletschke
` (2 preceding siblings ...)
2003-10-01 21:48 ` Mike Fedyk
@ 2003-10-01 23:21 ` Andrew Morton
2003-10-02 9:16 ` Konstantin Kletschke
3 siblings, 1 reply; 9+ messages in thread
From: Andrew Morton @ 2003-10-01 23:21 UTC (permalink / raw)
To: Konstantin Kletschke; +Cc: linux-kernel
Konstantin Kletschke <konsti@ludenkalle.de> wrote:
>
> I know my machine as a rockstable P4 Computer, since
> linux-2.6.0-test5-mm4 it freezes once a day.
> ...
>
> I have the state dump, catched via serial line at
> http://ludenkalle.de/capture-2003-09-30-linux-2.6.0-test5-mm4.log
Well it is stuck here:
mpd R current 17897 17786 17898 (NOTLB)
c010de4a 00000000 c0375000 00000001 00000001 00000000 c011de24 eaea56a0
00000040 00000000 c011de24 eaea56a0 c02eebbb cdd7be20 00000000 00000040
00000010 4037c480 4037c440 eaea56a0 eaea56a0 00000000 cdd7007b 0000007b
Call Trace:
[<c010de4a>] do_IRQ+0x24a/0x393
[<c011de24>] do_page_fault+0x0/0x4fc
[<c011de24>] do_page_fault+0x0/0x4fc
[<c02eebbb>] error_code+0x2f/0x38
[<c01f28fc>] __copy_from_user_ll+0x44/0x6c
[<f09debdf>] snd_pcm_lib_write_transfer+0x93/0x117 [snd_pcm]
[<f09df1c1>] snd_pcm_lib_write1+0x55e/0xa42 [snd_pcm]
[<c025a281>] freed_request+0xa1/0xa9
[<c0120cac>] default_wake_function+0x0/0x2e
[<f09df757>] snd_pcm_lib_write+0xb2/0x173 [snd_pcm]
[<f09deb4c>] snd_pcm_lib_write_transfer+0x0/0x117 [snd_pcm]
[<f09d8b4a>] snd_pcm_playback_ioctl1+0x585/0x674 [snd_pcm]
[<c01321e2>] do_timer+0xdf/0xe4
[<c02eeb1e>] apic_timer_interrupt+0x1a/0x20
[<f09d90e8>] snd_pcm_playback_ioctl+0x0/0x8d [snd_pcm]
[<c018dce2>] sys_ioctl+0x214/0x406
[<c02ee18f>] syscall_call+0x7/0xb
it might be coincidence, it might not be. Please gather another sysrq
trace next time it happens, just like that one.
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-01 23:21 ` Andrew Morton
@ 2003-10-02 9:16 ` Konstantin Kletschke
2003-10-03 15:54 ` Konstantin Kletschke
0 siblings, 1 reply; 9+ messages in thread
From: Konstantin Kletschke @ 2003-10-02 9:16 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel
On Wed, Oct 01, 2003 at 04:21:41PM -0700, Andrew Morton wrote:
> > http://ludenkalle.de/capture-2003-09-30-linux-2.6.0-test5-mm4.log
>
> mpd R current 17897 17786 17898 (NOTLB)
>
> it might be coincidence, it might not be. Please gather another sysrq
> trace next time it happens, just like that one.
OK, I will keep an eye on it. Poorly I upgraded Kernel and mpd after
that. But if it happens again I will get another sysrq and after that
try that with same versions but untainted, lets see :)
Regards, Konsti
--
2.6.0-test1-mm2
Konstantin Kletschke <konsti@ludenkalle.de>, <konsti@ku-gbr.de>
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E A080 1E69 3FDA EF62 FCEF
keulator.homelinux.org up 3 days, 2
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Date/UnixTime of SysRq state dump
2003-10-02 9:16 ` Konstantin Kletschke
@ 2003-10-03 15:54 ` Konstantin Kletschke
0 siblings, 0 replies; 9+ messages in thread
From: Konstantin Kletschke @ 2003-10-03 15:54 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel
* Konstantin Kletschke <konsti@ludenkalle.de> [Thu, Oct 02, 2003 at 11:16:38AM +0200]:
> OK, I will keep an eye on it. Poorly I upgraded Kernel and mpd after
Ok, the freeze at day change was random, a pure hoax of my PC! This time
it happened at 17:35 somehow.
Sadly this fｧ"$% minicom trunkated the state dump at top while my pc
rebootet :-( STOP, it is not trunkated, one sees the Show State command
at the first line :)
Andrew, the mpd tasks are still in there, may be you recognize a
similair pattern in its state or even not, one knows more then...
Again, this one
http://ludenkalle.de/capture-2003-10-03.log
is tainted. _Now_ I am back at nv driver and one screen instead of too
:(. Is there any XFree86 os driver supporting two screen output on a
Graphics Card which has two Outputs?
Regards, Konsti
--
2.6.0-test6-mm1
Konstantin Kletschke <konsti@ludenkalle.de>, <konsti@ku-gbr.de>
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E A080 1E69 3FDA EF62 FCEF
keulator.homelinux.org up 12 min,
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2003-10-03 15:54 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-10-01 18:28 Date/UnixTime of SysRq state dump Konstantin Kletschke
2003-10-01 18:50 ` Konstantin Kletschke
2003-10-01 19:09 ` Guennadi Liakhovetski
2003-10-01 21:32 ` Konstantin Kletschke
2003-10-01 21:48 ` Mike Fedyk
2003-10-01 22:44 ` Konstantin Kletschke
2003-10-01 23:21 ` Andrew Morton
2003-10-02 9:16 ` Konstantin Kletschke
2003-10-03 15:54 ` Konstantin Kletschke
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox