Bug 101203

Summary:	GPU HANG: ecode 9:0:0x85dffffb, in zoom [1000], reason: Hang on render ring, action: reset
Product:	Mesa	Reporter:	Nick Thomas <fdo>
Component:	Drivers/DRI/i965	Assignee:	Intel 3D Bugs Mailing List <intel-3d-bugs>
Status:	RESOLVED WORKSFORME	QA Contact:	Intel 3D Bugs Mailing List <intel-3d-bugs>
Severity:	normal
Priority:	medium	CC:	intel-gfx-bugs, marcin
Version:	unspecified
Hardware:	x86-64 (AMD64)
OS:	Linux (All)
Whiteboard:
i915 platform:		i915 features:
Attachments:	/sys/class/drm/card0/error another /sys/class/drm/card0/error

Description Nick Thomas 2017-05-26 16:35:38 UTC

Created attachment 131524 [details]
/sys/class/drm/card0/error

dmesg output:

[964958.089992] [drm] GPU HANG: ecode 9:0:0x85dffffb, in zoom [1000], reason: Hang on render ring, action: reset
[964958.089994] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[964958.089995] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[964958.089996] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[964958.089997] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[964958.089998] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[964958.090061] drm/i915: Resetting chip after gpu hang
[964958.090118] [drm] RC6 on
[964958.109035] [drm] GuC firmware load skipped
[964969.033487] drm/i915: Resetting chip after gpu hang
[964969.033558] [drm] RC6 on
[964969.051811] [drm] GuC firmware load skipped
[964969.070939] vrc_d_thread[1817]: segfault at 7ff6c49f7eb8 ip 00007ff6d09843fc sp 00007ff5f75fb790 error 4 in libQt5Core.so.5.6.2[7ff6d0715000+505000]

Error attached

`zoom` is this application: https://www.zoom.us/client/latest/zoom_amd64.deb version: 2.0.89826.0417 . I guess it's proprietary, but based on Qt 5

/sys/class/drm/card0/error attached.

Comment 1 Andy Botting 2017-07-20 04:42:12 UTC

Created attachment 132783 [details]
another /sys/class/drm/card0/error

Also seeing this with Zoom, which I use frequently for work video conf meetings.

No problem with the open source ati drivers, but get this frequently with intel.

Comment 2 Elizabeth 2018-03-06 22:21:46 UTC

Hello, is this still reproducible with latest mesa 17.3 releases and kernel at least on 4.13 version? Thank you.

Comment 3 Denis 2018-07-26 10:01:54 UTC

Hi Nick, Andry. Could you please re-check this issue?
Current app version is 2.3.128305.0716

I checked it on 10.6.0 and 18.2.0-git mesa with 4.9 and 4.17.3 kernel versions (Ubuntu 16.04). My CPU is SKL.

App worked for 2 days in conference mode. I didn't see any hangs. Or, if the issue is still actual for you, provide please additional information, how did you reproduce it.

Comment 4 Denis 2018-10-26 09:11:56 UTC

closing this issue as fixed.
https://bugs.freedesktop.org/show_bug.cgi?id=107968
Here is the same issue (same ecode in hang, same HW) where reporter confirmed that kernel update helped.

Please reopen this issue if it still actual for you.

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.