https://bugs.freedesktop.org/show_bug.cgi?id=34554
--- Comment #4 from Thibaut GIRKA <[email protected]> 2011-02-23 15:59:15 PST --- Created an attachment (id=43734) --> (https://bugs.freedesktop.org/attachment.cgi?id=43734) Kernel log of the working kernel with drm.debug=1 I'm now sure it's an EDID corruption, and nothing else. The reason why the "old" nouveau works has nothing to do with nouveau, but with drm: as you can see in the kern.log.drm.debug file, the EDID header got corrected by drm. Extracting the EDID dumps from kern.log.unstable reveals some pattern in the "corruption" (see edid-changes). I have to add that the EDID information corrected by DRM (as found in the "edid" sysfs node) is the same as the last dump of the kern.log.unstable logs except for the first byte (corrected to right EDID header). Since the crash occurred as the exact time I've started my game, I guess something in pygame's initialisation might have caused the "corruption". It would partially explain the corruption "cycles". -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. _______________________________________________ Nouveau mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/nouveau
