GPU HANG: ecode 9:0:0x85dffffb, in X [2869], reason: Hang on render ring, action: reset
Submitted by Jan Kundrát
Assigned to Intel 3D Bugs Mailing List
Link to original bug (#100459)
Description
Created attachment 130556 /sys/class/drm/card0/error
Kernel 4.9.18, Gentoo, Skylake i5-6300U, Thinkpad T460s, xorg-drivers-1.19, xorg-server-1.19.0, xorg-x11-7.4-r2, mesa oldish git 5f0e4c7c (built on 2016-12-05), KDE Plasma, kwin df306a4062ca946de34507a3e2be1d7178dfdf38 (built on 2017-03-06).
Here's what dmesg said:
[50141.741666] [drm] GPU HANG: ecode 9:0:0x85dffffb, in X [2869], reason: Hang on render ring, action: reset [50141.741670] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [50141.741671] [drm] Please file a new bug report on bugs.freedesktop.org against DRI -> DRM/Intel [50141.741673] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [50141.741674] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. [50141.741676] [drm] GPU crash dump saved to /sys/class/drm/card0/error [50141.741796] drm/i915: Resetting chip after gpu hang [50141.742030] [drm] RC6 on [50141.758485] [drm] GuC firmware load skipped [50151.789804] drm/i915: Resetting chip after gpu hang [50151.790021] [drm] RC6 on [50151.805713] [drm] GuC firmware load skipped [50151.828349] kscreen_backend[3160]: segfault at 10 ip 00007fd5e834944c sp 00007ffc6cf8b680 error 4 in KSC_XRandR.so[7fd5e8330000+22000] [50153.951117] wlp4s0: deauthenticating from f4:f2:6d:18:f3:bd by local choice (Reason: 3=DEAUTH_LEAVING)
I'll be attaching the crash dump, what I have in the dmesg, and the Xorg log.
Attachment 130556, "/sys/class/drm/card0/error":
2017-03-29-drm-error