Bug 22210 - [GM965 GEM KMS] GPU wedge while running compiz
Summary: [GM965 GEM KMS] GPU wedge while running compiz
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: git
Hardware: Other All
: low critical
Assignee: Eric Anholt
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords: NEEDINFO
Depends on:
Blocks:
 
Reported: 2009-06-10 08:57 UTC by Ben Gamari
Modified: 2009-10-09 15:11 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:

Attachments
dmesg log (54.15 KB, text/plain)
2009-06-10 08:57 UTC, Ben Gamari 		no flags Details
Xorg log (17.35 KB, text/plain)
2009-06-10 08:57 UTC, Ben Gamari 		no flags Details
GPU dump (172.85 KB, application/x-bzip)
2009-06-10 08:58 UTC, Ben Gamari 		no flags Details
View All

Description Ben Gamari 2009-06-10 08:57:10 UTC 
Created attachment 26639 [details]
dmesg log

GPU wedged up while drawing menu (I believe).
$ xorg-versions.sh 
Xorg components as of Wed Jun 10 11:46:17 EDT 2009
drm: 	6e88027eb5ae669cbe9710bca5309b3a06b0adc5
xf86-video-intel: 	63e6538fb8d6124612e2ecfdf9651261a28bec3b
mesa: 	305ada97f0be62d4024f5c0cde1c9ebc47bb753a
xserver: 	65c706829c2497a8a718b399cd5d4f9bbb452b1e
Comment 1 Ben Gamari 2009-06-10 08:57:36 UTC 
Created attachment 26640 [details]
Xorg log
Comment 2 Ben Gamari 2009-06-10 08:58:55 UTC 
Created attachment 26641 [details]
GPU dump
Comment 3 Ben Gamari 2009-06-10 08:59:33 UTC 
Backtrace:

(gdb) bt
#0  0x00007f777f64dec7 in ioctl () from /lib/libc.so.6
#1  0x00007f777e3ec6a3 in drmIoctl (fd=8, request=25688, arg=0x0)
    at xf86drm.c:187
#2  0x00007f777e3ec9a6 in drmCommandNone (fd=8, 
    drmCommandIndex=<value optimized out>) at xf86drm.c:2313
#3  0x00007f777df70810 in I830BlockHandler (i=0, 
    blockData=<value optimized out>, pTimeout=0x7fff3413ad50, 
    pReadmask=0x87d8c0) at i830_driver.c:2281
#4  0x00000000004e790d in AnimCurScreenBlockHandler (screenNum=0, 
    blockData=0x0, pTimeout=0x7fff3413ad50, pReadmask=0x87d8c0)
    at animcur.c:222
#5  0x00000000005d1c6a in compBlockHandler (i=0, blockData=0x0, 
    pTimeout=0x7fff3413ad50, pReadmask=0x87d8c0) at compinit.c:166
#6  0x000000000042c390 in BlockHandler (pTimeout=0x7fff3413ad50, 
    pReadmask=0x87d8c0) at dixutils.c:379
#7  0x000000000047da9b in WaitForSomething (pClientsReady=0x4711e50)
    at WaitFor.c:215
#8  0x0000000000446850 in Dispatch () at dispatch.c:362
#9  0x000000000042680e in main (argc=10, argv=0x7fff3413af48, 
    envp=0x7fff3413afa0) at main.c:283
Comment 4 Eric Anholt 2009-08-03 11:07:29 UTC 
(low priority unless we can get a reliable way to reproduce it)
Comment 5 Ben Gamari 2009-08-03 12:22:36 UTC 
In fact, I don't believe I have seen another crash similar to this. It's tough saying since I'm not exactly sure what triggered this one, but I'd be comfortable marking it as INVALID.
Comment 6 Eric Anholt 2009-08-07 19:27:03 UTC 
Would you have been running any other 3D stuff while this happened?  Wondering if it's the fix series I just pushed, but just compiz wouldn't do it I don't think, since you're running at 1920x1200.
Comment 7 Ben Gamari 2009-08-08 07:49:48 UTC 
Hmm, I doubt it. Compiz is generally the only OpenGL application that I use on a day-to-day basis. Unfortunately, compiz hasn't been starting recently so I've been largely unable to test with it. I'm going to try figuring this out today and hopefully I'll have some feedback (although this bug definitely takes quite a while to reproduce)
Comment 8 Eric Anholt 2009-10-09 15:11:34 UTC 
OK, I'll mark it fixed for now, assuming that:

commit 5604b27b9326ac542069a49ed9650c4b0d3e939a
Author: Eric Anholt <eric@anholt.net>
Date:   Wed Sep 9 12:35:30 2009 -0700

    i965: Fix relocation delta for WM surfaces.
    
    This was a regression in 0f328c90dbc893e15005f2ab441d309c1c176245.

fixed it.

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.