This is the upstream report of http://bugs.centos.org/view.php?id=5911 . (See also bug 51929, but apparently not relevant). Copy of the CentOS report: -- I have a brand new Dell Optiplex 7010. It was initially delivered with Ubuntu 11.10, but reinstalled with CentOS. Every other day or so, the screen locks up for a few seconds, then I can see this in the kernel log: [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state Then, the graphics in some applications (including Firefox and Emacs) are distorted. -- Since the original report, I've stopped using xscreensaver. This helps somewhat: It now takes ~1 week before the problem happens, instead of ~1 day. It is still a frustrating bug, though. If there's anything I can try, please let me know.
You need uptodate drivers (both xf86-video-intel and mesa) to avoid this issue on IVB T1 systems. *** This bug has been marked as a duplicate of bug 52473 ***
(Why do you change the Summary? Very annoying, and will make it harder for others to find the bug. The summary I used was from the kernel dmesg - certainly something others will look for.) Do you have any information about which versions that fixes the problem? An approach of "upgrade to the latest drivers and Mesa stack" does not work well for distributions such as CentOS.
Because your summary contained no specific details; that message is printed for every single GPU hang regardless of cause or creed. So it would not have helped anybody looking for the same issue in the future. The upstream versions with the IVB GT1 fix are xf86-video-intel-2.20.3 and mesa-8.0.4.
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.