https://bugs.freedesktop.org/show_bug.cgi?id=110971

Bug ID: 110971 Summary: GPU HANG: ecode 6:1:0xfffffffe, in spring-main [8656], hang on rcs0 Product: Mesa Version: 19.1 Hardware: x86-64 (AMD64) OS: Linux (All) Status: NEW Severity: critical Priority: medium Component: Drivers/DRI/i915 Assignee: dri-devel@lists.freedesktop.org Reporter: Chris.Rebischke@archlinux.org QA Contact: dri-devel@lists.freedesktop.org

Created attachment 144614 --> https://bugs.freedesktop.org/attachment.cgi?id=144614&action=edit crash dump from /sys/class/drm/card0/error

I get unspecified GPU hangs while playing games. It mostly occurs when my CPU starts throttling:

[ 5838.889104] mce: CPU0: Package temperature above threshold, cpu clock throttled (total events = 100749) [ 5838.889108] mce: CPU2: Package temperature above threshold, cpu clock throttled (total events = 100749) [ 5838.889110] mce: CPU3: Package temperature above threshold, cpu clock throttled (total events = 100749) [ 5838.889112] mce: CPU1: Package temperature above threshold, cpu clock throttled (total events = 100749) [ 5900.819612] mce: CPU2: Core temperature above threshold, cpu clock throttled (total events = 16235) [ 5900.819613] mce: CPU3: Core temperature above threshold, cpu clock throttled (total events = 16235) [ 5900.827685] mce: CPU2: Core temperature/speed normal [ 5900.827686] mce: CPU3: Core temperature/speed normal [ 6138.891505] mce: CPU1: Package temperature above threshold, cpu clock throttled (total events = 171156) [ 6138.891506] mce: CPU0: Package temperature above threshold, cpu clock throttled (total events = 171156) [ 6138.891528] mce: CPU2: Package temperature above threshold, cpu clock throttled (total events = 171156) [ 6138.891529] mce: CPU3: Package temperature above threshold, cpu clock throttled (total events = 171156) [ 6176.479565] i915 0000:00:02.0: GPU HANG: ecode 6:1:0xfffffffe, in spring-main [8656], hang on rcs0 [ 6176.479570] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [ 6176.479572] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel [ 6176.479573] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [ 6176.479574] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. [ 6176.479575] [drm] GPU crash dump saved to /sys/class/drm/card0/error [ 6176.479642] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6422.644956] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6438.888919] mce: CPU3: Package temperature/speed normal [ 6438.888920] mce: CPU2: Package temperature/speed normal [ 6438.888924] mce: CPU1: Package temperature/speed normal [ 6438.888926] mce: CPU0: Package temperature/speed normal