Created attachment 108238 [details]
Contents of /sys/class/drm/card0/error

Kernel occasionally reports GPU HANG. I noticed no user-visible effects correlated to this message. This is GPU integrated on Sandy Bridge Core i7-2600K.

Kernel log:

[  518.149655] [drm] GPU HANG: ecode -1:0x00000000, reason: Kicking stuck semaphore on render ring, action: continue
[  518.149658] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[  518.149658] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[  518.149659] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[  518.149660] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[  518.149660] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[  518.150077] [drm] GPU HANG: ecode -1:0x00000000, reason: Kicking stuck semaphore on blitter ring, action: continue

xdpyinfo excerpt:

vendor string:    The X.Org Foundation
vendor release number:    11601000
X.Org version: 1.16.1

OS: Debian jessie.

I can provide more info when you need it.