[hsw] GPU hang in World of Warcraft (wine)
Submitted by Vladimir Voinea
Assigned to Vladimir Voinea
Link to original bug (#102647)
Description
Created attachment 134133 Crash dump from /sys/class/drm/card0/error
Running kernel version 4.13.1 on Debian, once in a while my screen freezes for a few seconds and I get the following messages in dmesg:
[ 212.802305] [drm] GPU HANG: ecode 7:0:0x87d7bffa, in wow-64.exe [2897], reason: Hang on rcs0, action: reset [ 212.802306] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [ 212.802306] [drm] Please file a new bug report on bugs.freedesktop.org against DRI -> DRM/Intel [ 212.802306] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [ 212.802307] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. [ 212.802307] [drm] GPU crash dump saved to /sys/class/drm/card0/error [ 212.802344] drm/i915: Resetting chip after gpu hang [ 216.030367] asynchronous wait on fence i915:[global]:476b timed out [ 220.830428] drm/i915: Resetting chip after gpu hang [ 228.798441] drm/i915: Resetting chip after gpu hang [ 236.802389] drm/i915: Resetting chip after gpu hang [ 244.798335] drm/i915: Resetting chip after gpu hang
When it happens:
- always: When trying to run a game via wine
- sometimes: When browsing Google Maps w/ Google Chrome. Screen freezes, WebGL crashes. Can't reproduce it
Also, I can't reproduce this issue on kernel version 4.7. I'm not sure if this is the latest version without the bug.
Attachment 134133, "Crash dump from /sys/class/drm/card0/error":
error.gz