Bug 100546 - [gen9] GPU HANG: reason: Hang on render ring, action: reset
Summary: [gen9] GPU HANG: reason: Hang on render ring, action: reset
Status: NEEDINFO
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Intel 3D Bugs Mailing List
QA Contact: Intel 3D Bugs Mailing List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-04-04 03:03 UTC by hilbe
Modified: 2018-03-21 20:18 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg (82.47 KB, text/plain)
2017-04-04 03:04 UTC, hilbe
Details
/sys/class/drm/card0/error (26.82 KB, text/plain)
2017-04-04 03:06 UTC, hilbe
Details
lspci -vvv (58.00 KB, text/plain)
2017-04-04 03:06 UTC, hilbe
Details
lshw (26.68 KB, text/plain)
2017-04-04 03:06 UTC, hilbe
Details
rpm -qa (61.93 KB, text/plain)
2017-04-04 03:07 UTC, hilbe
Details

Note You need to log in before you can comment on or make changes to this bug.
Description hilbe 2017-04-04 03:03:36 UTC
Occasionally X crashes when copy/pasting text within LibreOffice Impress.

$ dmesg | grep drm
[    5.974228] [drm] Initialized
[    6.386113] [drm] Memory usable by graphics device = 4096M
[    6.386115] fb: switching to inteldrmfb from VESA VGA
[    7.679582] [drm] Replacing VGA console driver
[    7.686753] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[    7.686754] [drm] Driver supports precise vblank timestamp query.
[    7.702310] [drm] Finished loading i915/skl_dmc_ver1_26.bin (v1.26)
[    7.705321] [drm] GuC firmware load skipped
[    8.068750] [drm] Initialized i915 1.6.0 20161121 for 0000:00:02.0 on minor 0
[    8.446616] fbcon: inteldrmfb (fb0) is primary device
[    8.836647] [drm] RC6 on
[    9.298557] [drm] Initialized nouveau 1.3.1 20120801 for 0000:01:00.0 on minor 1
[   10.133524] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[ 5386.832635] [drm] GPU HANG: ecode 9:0:0x85dffffb, in X [1610], reason: Hang on render ring, action: reset
[ 5386.832637] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[ 5386.832637] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[ 5386.832638] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[ 5386.832638] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[ 5386.832638] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[ 5386.832669] drm/i915: Resetting chip after gpu hang
[ 5386.832778] [drm] RC6 on
[ 5386.843168] [drm] GuC firmware load skipped
[ 5398.793007] drm/i915: Resetting chip after gpu hang
[ 5398.793072] [drm] RC6 on
[ 5398.808162] [drm] GuC firmware load skipped

$ uname -a
Linux linux-wijr 4.10.5-1-default #1 SMP PREEMPT Wed Mar 22 13:15:00 UTC 2017 (f3fbfc6) x86_64 x86_64 x86_64 GNU/Linux

$ less /etc/os-release
OS release:
NAME="openSUSE Tumbleweed"
# VERSION="20170331"
ID=opensuse
ID_LIKE="suse"
VERSION_ID="20170331"
PRETTY_NAME="openSUSE Tumbleweed"
ANSI_COLOR="0;32"
CPE_NAME="cpe:/o:opensuse:tumbleweed:20170331"
BUG_REPORT_URL="https://bugs.opensuse.org"
HOME_URL="https://www.opensuse.org/"
Comment 1 hilbe 2017-04-04 03:04:10 UTC
Created attachment 130661 [details]
dmesg
Comment 2 hilbe 2017-04-04 03:06:04 UTC
Created attachment 130662 [details]
/sys/class/drm/card0/error
Comment 3 hilbe 2017-04-04 03:06:38 UTC
Created attachment 130663 [details]
lspci -vvv
Comment 4 hilbe 2017-04-04 03:06:54 UTC
Created attachment 130664 [details]
lshw
Comment 5 hilbe 2017-04-04 03:07:23 UTC
Created attachment 130665 [details]
rpm -qa
Comment 6 Elizabeth 2018-03-21 20:18:05 UTC
Hello Hilbe, is suspected that the libreoffice issue may be related to a GPU hang for 2D workloads fixed in release 17.3.6, as mentioned here https://bugs.freedesktop.org/show_bug.cgi?id=99720#c33. Could you try it? Thank 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.