Bug 14669 - [945GM] Sporadic server lockup
Summary: [945GM] Sporadic server lockup
Status: RESOLVED INVALID
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: 7.3 (2007.09)
Hardware: x86 (IA32) Linux (All)
: medium major
Assignee: Jesse Barnes
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords: NEEDINFO
Depends on:
Blocks: 15000
  Show dependency treegraph
 
Reported: 2008-02-25 08:14 UTC by Johannes Engel
Modified: 2008-03-17 01:51 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
Xorg logs (10.04 KB, application/x-bzip)
2008-02-25 08:14 UTC, Johannes Engel
no flags Details
xorg.conf (5.40 KB, text/plain)
2008-02-25 08:14 UTC, Johannes Engel
no flags Details
Log after attempt to restart X server (28.10 KB, text/plain)
2008-02-26 01:53 UTC, Johannes Engel
no flags Details
X log of the lockup (43.77 KB, text/plain)
2008-02-26 01:54 UTC, Johannes Engel
no flags Details

Description Johannes Engel 2008-02-25 08:14:05 UTC
Created attachment 14558 [details]
Xorg logs

Several times per day my X.Org server locks up. I cannot determine the conditions, but working using Firefox, Thunderbird, kile etc. seems to be sufficient.
After locking up the server tries to restart but finds "existing errors in hardware state". Only reboot helps to get it up again.

945GM using driver intel 2.2.1
DRM kernel modules from git as of today
kernel 2.6.24.2

Logs will be attached as well as my xorg.conf
Comment 1 Johannes Engel 2008-02-25 08:14:37 UTC
Created attachment 14559 [details]
xorg.conf
Comment 2 Gordon Jin 2008-02-25 19:32:45 UTC
Thanks for the reporting this. But it's hard to fix without a steady reproducing way. Could you provide more info like dmesg according to http://www.intellinuxgraphics.org/how_to_report_bug.html? And please attach log without zipped.
Comment 3 Wang Zhenyu 2008-02-25 21:18:28 UTC
Pls don't use tarball, and stick with text/plain mime type for all logs and config files.

Could you help to verify does this happen with default options? Or which option
you're using might cause this problem?
Comment 4 Johannes Engel 2008-02-26 01:51:51 UTC
OK, here the logs in plain ascii text. :)
dmesg says exactly nothing at all.
I am sorry that I cannot see an exact way to reproduce the problem. :(
Comment 5 Johannes Engel 2008-02-26 01:53:53 UTC
Created attachment 14575 [details]
Log after attempt to restart X server

X server takes 100% CPU after restart attempt
Comment 6 Johannes Engel 2008-02-26 01:54:20 UTC
Created attachment 14576 [details]
X log of the lockup
Comment 7 Johannes Engel 2008-02-26 01:54:53 UTC
I forgot to mention: The problem also occurs with driver's default options.
Comment 8 Gordon Jin 2008-02-26 18:24:08 UTC
It's working on our side. Without a specific reproducing steps, we can do little. I'm giving this tough bug to jbarnes to see if he happens to have some findings.
Comment 9 Jesse Barnes 2008-03-07 14:52:18 UTC
Hm, backtrace indicates we're coming in through an EXA path (is the backtrace the same whenever you see crashes?); there have been lots of changes to the server in this area, does the problem still occur with older or newer X servers?
Comment 10 Johannes Engel 2008-03-08 09:14:42 UTC
The problem does not occur using the distro's X.org server (1.4 from X.Org 7.3).
I can try a newer one next week.
Comment 11 Wang Zhenyu 2008-03-14 00:34:04 UTC
Any news? Can we close this one?
Comment 12 Michael Fu 2008-03-17 00:53:36 UTC
(In reply to comment #10)
> The problem does not occur using the distro's X.org server (1.4 from X.Org
> 7.3).
> I can try a newer one next week.
> 
Johannes, what xserver version were you using when you met the problem? Isn't it the xserver from your distro's release?
Comment 13 Johannes Engel 2008-03-17 01:51:22 UTC
Using actual X.Org from git master the problem does not occur anymore at the moment. I'll tell you once it happens again. So let's close this one for now.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.