Bug 91432

Summary:	[drm] GPU HANG: ecode 6:2:0x000097f7 but gpu crash dump file is always empty
Product:	DRI	Reporter:	paul
Component:	DRM/Intel	Assignee:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Status:	CLOSED WORKSFORME	QA Contact:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity:	normal
Priority:	medium	CC:	intel-gfx-bugs
Version:	XOrg git
Hardware:	Other
OS:	All
Whiteboard:
i915 platform:		i915 features:	GPU hang

Description paul 2015-07-22 19:40:05 UTC

Occasionally gpu hangs occur on my system, but the specified gpu crash dump file is always empty.

Linux version: 4.1.2-2, SMP PREEMPT, x86_64 GNU/Linux

$ dmesg | grep drm
[  +0,010497] [drm] Initialized drm 1.1.0 20060810
[  +0,005761] [drm] Memory usable by graphics device = 2048M
[  +0,000001] fb: switching to inteldrmfb from EFI VGA
[  +0,000070] [drm] Replacing VGA console driver
[  +0,006563] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[  +0,000003] [drm] Driver supports precise vblank timestamp query.
[  +0,000039] [drm] Initialized i915 1.6.0 20150327 for 0000:00:02.0 on minor 0
[  +0,010486] fbcon: inteldrmfb (fb0) is primary device
[  +0,004627] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[Jul21 18:13] [drm] stuck on blitter ring
[  +0,000633] [drm] GPU HANG: ecode 6:2:0x000097f7, in Xorg [417], reason: Ring hung, action: reset
[  +0,000001] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[  +0,000001] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[  +0,000001] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[  +0,000001] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[  +0,000001] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[  +0,502197] drm/i915: Resetting chip after gpu hang
[  +0,000014] [drm:i915_reset [i915]] *ERROR* Failed to reset chip: -110

Comment 1 Chris Wilson 2015-07-30 06:39:53 UTC

The error state file is never empty; do not trust ls. Please capture it before you reboot.

Comment 2 yann 2016-04-25 14:44:36 UTC

Paul, is it still reproducible ? 
If this is still the case please attached corresponding /sys/class/drm/card0/error (please refer to https://01.org/linuxgraphics/documentation/how-get-gpu-error-state) and also try v4.6-rc5 or later.

Comment 3 yann 2016-05-04 09:15:34 UTC

timeout. Closing this bug since no activity and not enough data to proceed on investigation. If you encounter same issue, please fill a new issue with proper data.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.