Created attachment 142023 [details]
sudo cat /sys/class/drm/card0/error > crashdump
Following error appeared in dmesg
[ 8969.075804] [drm] GPU HANG: ecode 7:0:0x85dffffc, in kwin_x11 , reason: hang on rcs0, action: reset
[ 8969.075805] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[ 8969.075805] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[ 8969.075806] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[ 8969.075806] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[ 8969.075807] [drm] GPU crash dump saved to /sys/class/drm/card0/error
crash dump attached
Could tell us what version of Mesa you're running?
We appear to be missing an MI_BATCH_BUFFER_END here too (like https://bugs.freedesktop.org/show_bug.cgi?id=108358)
This is mesa 18.0.5 for Ubuntu 18.04 amd64.
Any news on this?