Summary: | Radeon XPRESS 200M (RC410) - xorg crash when enabling compiz (regression) | ||
---|---|---|---|
Product: | Mesa | Reporter: | Alex Villacís Lasso <avillaci> |
Component: | Mesa core | Assignee: | mesa-dev |
Status: | RESOLVED FIXED | QA Contact: | |
Severity: | major | ||
Priority: | medium | Keywords: | regression |
Version: | unspecified | ||
Hardware: | x86 (IA32) | ||
OS: | Linux (All) | ||
Whiteboard: | |||
i915 platform: | i915 features: | ||
Attachments: |
Configuration file used for Xorg
Xorg.0.log after X server crashes dmesg output right after crash Console output with assertion commented out and replaced with message |
Description
Alex Villacís Lasso
2008-11-27 08:54:46 UTC
Created attachment 20631 [details]
Configuration file used for Xorg
Created attachment 20632 [details]
Xorg.0.log after X server crashes
Created attachment 20633 [details]
dmesg output right after crash
System is still responsive, and can be ssh'd into. That is how I got these files. ps aux shows no trace of the xserver process, so it has crashed, but I see no trace at all on the log file.
The key line is X: shader/prog_statevars.c:783: append_token: Assertion `0' failed. Looks like the failing assertion was introduced in mesa commit 11701b4c151de450e0b80d9e61266ff178b4bd60 ('mesa: add some missing switch cases for generating state var strings'). I'm removing the assertion. It's harmless. But it would be nice to know what the switch value is that's not handled. (In reply to comment #5) > I'm removing the assertion. It's harmless. But it would be nice to know what > the switch value is that's not handled. > I commented out the assertion and put in its place a call to _mesa_problem. I get the following: Mesa 7.3-devel implementation error: unhandled state index 0x0000009e while appe nding to 'state.(internal)' Please report at bugzilla.freedesktop.org ...but otherwise the problem is 'fixed'. Created attachment 20638 [details]
Console output with assertion commented out and replaced with message
OK, 0x0000009e corresponds to driver-private state. I'm committing a minor change/fix. Closing this bug. |
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.