Bug 24137

Summary: [855GM, xf86-video-intel-2.8.99.902] "Freeze" - trigger unknown
Product: xorg Reporter: Bruno <bonbons>
Component: Driver/intelAssignee: Carl Worth <cworth>
Status: RESOLVED DUPLICATE QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium CC: andrej, moikkis
Version: unspecified   
Hardware: x86 (IA32)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
GPU dump, bz2 compressed
none
Register state matching with attachment #29830 none

Description Bruno 2009-09-24 10:29:15 UTC
Created attachment 29830 [details]
GPU dump, bz2 compressed

GPU Freeze while reading e-mail.
Regdump and gpudump.

Possibly a cornercase left over after bug #22947

PS: Is there a chance to see support for GPU-reset for i855 similar to what has been posted recently for i965?
Ideally on freeze detection a uevent would be sent to user-space so gpu+regs can be dumped prior to proceeding with reset.
Comment 1 Bruno 2009-09-24 10:30:53 UTC
Created attachment 29831 [details]
Register state matching with attachment #29830 [details]
Comment 2 Eric Anholt 2009-10-09 12:18:11 UTC
Needs information at:

http://intellinuxgraphics.org/how_to_report_bug.html
Comment 3 Bruno 2009-10-09 13:12:29 UTC
(In reply to comment #2)
> Needs information at:
> 
> http://intellinuxgraphics.org/how_to_report_bug.html

I did not repeat everything as it is the state as for my last massage to bug 22947 to which I referred to.

That is:
- distro: Gentoo
- xorg-server-1.6.3
- kernel: 2.6.31
- xf86-video-intel, git 7e7db7ac530b5282b0841585959597b54fcc633b,
  Add new backlight driver "samsung"
- mesa-7.5-r2
- Arch: x86, 32 bit
- System: Acer TM660, LVDS:
  00:02.0 VGA compatible controller [0300]:
    Intel Corporation 82852/855GM Integrated Graphics Device [8086:3582] (rev 02)
  00:02.1 Display controller [0380]:
    Intel Corporation 82852/855GM Integrated Grahics Device [8086:3582] (rev 02)

I've not had a second such freeze since then, as such I hope the cause can be found in the GPU dump or at least search area restricted with that one's help.
Comment 4 Chris Wilson 2010-03-02 08:31:44 UTC
The dump is rather odd as it appears properly constructed. The one bit of information that is missing from the two attachments is IPEHR which would confirm the CPU/GPU incoherency hypothesis... However, judging by the dump that is extremely likely to be the cause...

*** This bug has been marked as a duplicate of bug 26345 ***

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.