Created attachment 96732 [details]
GPU crash dump
[ 4352.772748] [drm] stuck on render ring
[ 4352.772753] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[ 4352.772754] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[ 4352.772755] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[ 4352.772755] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[ 4352.772756] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[ 4352.775394] [drm:i915_set_reset_status] *ERROR* render ring hung inside bo (0x2463000 ctx 1) at 0x2464404
Created attachment 96733 [details]
Please supply the full /sys/class/drm/card0/error, the current attachment is corrupt.
New Attach file
Created attachment 96734 [details]
Context intact, but the wild write clobbered the batchbuffer this time.
Created attachment 97057 [details]
another dump, dmesg output
Created attachment 97058 [details]
new crash dump
using driver intel-linux-graphics-installer_1.0.4-0intel1_amd64.deb
Are you using KDE? If so, this is a duplicate of 77207.
(In reply to comment #8)
> Are you using KDE? If so, this is a duplicate of 77207.
That's right, it happens with KDE, from April 1 to report the bug.
Rather would duplicate the 77207 :)
Fixed in Mesa master. Upgrading to X server 1.15.1 or 1.14.6 will also fix this. Setting KWIN_OPENGL_INTERFACE=egl should also work around the problem without having to upgrade any software.
*** This bug has been marked as a duplicate of bug 77207 ***