Created attachment 100803 [details]
card0-error.dump.xz

Kernel: 3.15.0-rc6+

Noticed following in the kernel log, but no idea what caused this:
[snip]
[27311.226010] [drm] GPU HANG: ecode -1:0x00000000, reason: Command parser error, iir 0x00008000, action: continue
[27311.226010] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[27311.226010] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[27311.226010] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[27311.226010] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[27311.226010] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[27311.226010] i915: render error detected, EIR: 0x00000010
[27311.226010] i915: page table error
[27311.226010] i915:   PGTBL_ER: 0x00000040
[27311.226010] [drm:i915_report_and_clear_eir] *ERROR* EIR stuck: 0x00000010, masking
[27311.226010] [drm] GPU HANG: ecode -1:0x00000000, reason: Command parser error, iir 0x00008000, action: continue
[27311.226010] i915: render error detected, EIR: 0x00000010
[27311.226010] i915: page table error
[27311.226010] i915:   PGTBL_ER: 0x00000040
[/snip]