Summary: | [i845 bisected] Extreme Tux Racer hangs graphics card | ||
---|---|---|---|
Product: | Mesa | Reporter: | Brian Rogers <brian> |
Component: | Drivers/DRI/i915 | Assignee: | Eric Anholt <eric> |
Status: | RESOLVED FIXED | QA Contact: | |
Severity: | critical | ||
Priority: | low | CC: | idr, moikkis |
Version: | git | ||
Hardware: | x86 (IA32) | ||
OS: | Linux (All) | ||
Whiteboard: | |||
i915 platform: | i915 features: | ||
Attachments: |
dmesg with drm.debug=0x06
i915_error_state produced from Extreme Tux Racer freeze |
Description
Brian Rogers
2010-02-13 09:12:25 UTC
Created attachment 33278 [details]
i915_error_state produced from Extreme Tux Racer freeze
If you could bisect when it started hanging, that would help OK, I have started bisecting. So far, the problem is in mesa somewhere between 7.6.0 and 1e414df8. I still have a ways to go... Bisect done. b4a6169412819cc3a027c6a118f0537911145a30 is the first bad commit commit b4a6169412819cc3a027c6a118f0537911145a30 Author: Ian Romanick <ian.d.romanick@intel.com> Date: Tue Dec 8 23:24:58 2009 -0800 intel: Make RGB textures use XRGB8888 I can also confirm that reverting that commit on Ubuntu Lucid's 7.7.0-based Mesa gets Extreme Tux Racer working again. (In reply to comment #5) > I can also confirm that reverting that commit on Ubuntu Lucid's 7.7.0-based > Mesa gets Extreme Tux Racer working again. The X server (and sometimes the kernel) crashes right after I start the game. Also happens with Doomsday and Warzone2100. Reverting this patch doesn't make a difference in my situation. I will try to find a working mesa version (if any) and try to bisect from there. I don't have my 845 here, so I pushed the obvious fix without testing it. Hope it works out. If you could drop a note here that it helps you, I'll get it merged to stable. commit f0ff214bee64a705d3ef6610e9dc25bc1a46a460 Author: Eric Anholt <eric@anholt.net> Date: Fri Jun 4 13:40:48 2010 -0700 i915: Don't use XRGB8888 on 830 and 845. The support for XRGB8888 appeared in the 855 and 865, and this format is reserved on 830/845. This should fix a regression from b4a6169412819cc3a027c6a118f0537911145a30 that caused hangs in etracer on 845s. Bug #26557. I just got a chance to test the affected machine. Yeah, it fixes it. etracer works now both with xorg-edgers on Lucid, and with my own backport of your patch to Mesa 7.7.1. I see this fix is included now in mesa 7.8.2. What about the 7.7 branch? |
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.