All of lore.kernel.org
 help / color / mirror / Atom feed
From: Bernd Zeimetz <bernd@bzed.de>
To: sparclinux@vger.kernel.org
Subject: Re: unkillable dpkg-query processes
Date: Sun, 04 Nov 2007 19:55:20 +0000	[thread overview]
Message-ID: <472E23A8.90203@bzed.de> (raw)
In-Reply-To: <20071024223332.GA24557@keid.carnet.hr>

[-- Attachment #1: Type: text/plain, Size: 2046 bytes --]


> Ok, the key in the trace is:
> 
> Nov  2 16:25:30 titan kernel: [  978.134874]   CPU[  1]: TSTATE[0000000080009603] TPC[000000000067d2e0] TNPC[000000000067d2d4] TASK[aptitude:3204]
> Nov  2 16:25:30 titan kernel: [  978.257809]              TPC[_write_unlock_irq+0x20/0x110]
>  ...
> Nov  2 16:25:30 titan kernel: [  978.507778]   CPU[  3]: TSTATE[0000000011009605] TPC[00000000004419f8] TNPC[00000000004419fc] TASK[aptitude:3203]
> Nov  2 16:25:30 titan kernel: [  978.630707]              TPC[cheetah_xcall_deliver+0x174/0x23c]
> 
> The first symbol is misleading, it says _write_unlock_irq but actually
> in the assembler the PC is in the spinlock read spinning loop
> section.  So actually it's hanging in _spin_lock().
> 
> CPU #3 is trying to send a cross-call message interrupt, but for
> some reason that isn't making forward progress.
> 
> Let's see what's calling these things by adding some more debugging
> information.  Please retry the test with the following patch on
> top of the original sysrq-g debugging patch and please get new
> logs when it hangs.


Today I was a bit out of luck, either the machine crashed so badly that
it just didn't react on anything anymore, or it didn't crash.
The machine went amok a bit slower when I did the following things,
which also resulted in the attached sysrq output.
- run stress -c 2 to get the load up, didn't need that the last time...
- run something like `while true; do echo g > /proc/sysrg-trigger; sleep
0.5; done`
- run aptitude -u several times until the machine died.

So I'm not sure if the result is really useful for you - if not just let
me know. I've attached the last ~10-20 sysrq-g outputs - as it was
running in a loop I have a ton of them. In case you're wondering: http
is aptitude's http method.

We'll also run the patched Kernel on a US II machine form tomorrow on -
but it always took a longer time until it crashed, so we'll see if it
happens at all.

Thanks for your work,


Bernd


-- 
Bernd Zeimetz
<bernd@bzed.de>                         <http://bzed.de/>

[-- Attachment #2: sysrq2.txt --]
[-- Type: application/pgp-keys, Size: 25048 bytes --]

  parent reply	other threads:[~2007-11-04 19:55 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-10-24 22:33 unkillable dpkg-query processes Josip Rodin
2007-10-24 22:58 ` David Miller
2007-10-24 23:28 ` Josip Rodin
2007-10-24 23:36 ` David Miller
2007-10-25  6:41 ` David Miller
2007-10-25 14:58 ` Josip Rodin
2007-10-25 15:07 ` Josip Rodin
2007-10-25 16:43 ` Josip Rodin
2007-10-26 11:30 ` David Miller
2007-10-26 12:23 ` Bernd Zeimetz
2007-10-26 12:30 ` Bernd Zeimetz
2007-10-26 22:01 ` David Miller
2007-10-26 22:30 ` Bernd Zeimetz
2007-10-26 22:32 ` Josip Rodin
2007-10-26 22:33 ` Josip Rodin
2007-10-26 22:37 ` Bernd Zeimetz
2007-10-26 22:40 ` Bernd Zeimetz
2007-10-27 18:09 ` Bernd Zeimetz
2007-10-27 18:46 ` Bernd Zeimetz
2007-10-28  3:03 ` Bernd Zeimetz
2007-10-28  3:40 ` David Miller
2007-10-28  4:24 ` David Miller
2007-10-28 12:20 ` Bernd Zeimetz
2007-10-28 12:58 ` Bernd Zeimetz
2007-10-28 13:02 ` Bernd Zeimetz
2007-10-28 13:38 ` Sébastien Bernard
2007-10-29  0:01 ` Bernd Zeimetz
2007-10-29  0:19 ` Bernd Zeimetz
2007-10-29  0:41 ` David Miller
2007-10-29  1:18 ` Bernd Zeimetz
2007-10-29  1:39 ` David Miller
2007-10-29  2:06 ` Bernd Zeimetz
2007-10-29  2:18 ` David Miller
2007-10-29 12:06 ` seb
2007-10-29 23:37 ` Josip Rodin
2007-10-29 23:45 ` David Miller
2007-10-30  0:30 ` Josip Rodin
2007-10-30  0:47 ` Bernd Zeimetz
2007-10-30  0:50 ` Bernd Zeimetz
2007-10-30  0:51 ` Bernd Zeimetz
2007-10-30  1:24 ` David Miller
2007-10-30  1:35 ` David Miller
2007-10-30  1:54 ` Bernd Zeimetz
2007-10-30  2:11 ` David Miller
2007-10-30  2:43 ` Bernd Zeimetz
2007-10-30  2:49 ` David Miller
2007-11-01 21:40 ` Josip Rodin
2007-11-01 22:01 ` David Miller
2007-11-01 22:47 ` Bernd Zeimetz
2007-11-02  4:55 ` David Miller
2007-11-02 15:37 ` Bernd Zeimetz
2007-11-02 16:21 ` Josip Rodin
2007-11-04  5:10 ` David Miller
2007-11-04  8:07 ` David Miller
2007-11-04 19:55 ` Bernd Zeimetz [this message]
2007-11-04 21:51 ` David Miller
2007-11-04 23:14 ` Bernd Zeimetz
2007-11-05  0:03 ` Bernd Zeimetz
2007-11-06  3:51 ` Bernd Zeimetz
2007-11-06 11:13 ` David Miller
2007-11-06 13:30 ` Bernd Zeimetz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=472E23A8.90203@bzed.de \
    --to=bernd@bzed.de \
    --cc=sparclinux@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.