Bug 102550 - [ilk] GPU hang in plasmashell
Summary: [ilk] GPU hang in plasmashell
Status: REOPENED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: 17.0
Hardware: x86-64 (AMD64) All
: medium normal
Assignee: Intel 3D Bugs Mailing List
QA Contact: Intel 3D Bugs Mailing List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-09-05 18:24 UTC by in+fd
Modified: 2018-04-26 19:36 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features: display/LVDS


Attachments
GPU crash dump (119.77 KB, application/gzip)
2017-09-05 18:24 UTC, in+fd
Details

Note You need to log in before you can comment on or make changes to this bug.
Description in+fd 2017-09-05 18:24:45 UTC
Created attachment 133973 [details]
GPU crash dump

Bug description:

Hello,

My logs told me to open this ticket:

Sep 05 19:18:59 asdf kernel: [drm] GPU HANG: ecode 5:0:0x97cafffb, in plasmashell [3636], reason: Hang on render ring, action: reset
Sep 05 19:18:59 asdf kernel: [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Sep 05 19:18:59 asdf kernel: [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Sep 05 19:18:59 asdf kernel: [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Sep 05 19:18:59 asdf kernel: [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Sep 05 19:18:59 asdf kernel: [drm] GPU crash dump saved to /sys/class/drm/card0/error
Sep 05 19:18:59 asdf kernel: drm/i915: Resetting chip after gpu hang
Sep 05 19:19:09 asdf kernel: drm/i915: Resetting chip after gpu hang
Sep 05 19:19:20 asdf kernel: drm/i915: Resetting chip after gpu hang


System environment:
-- chipset: i915
-- system architecture: 64-bit
-- xf86-video-intel: 2.99.917.770_gcb6ba2da-1.3
-- xserver: 7.6_1.18.3-22.1
-- mesa: 17.0.5-174.2
-- libdrm: 2.4.76-1.2
-- kernel: 4.4.79-19.1
-- Linux distribution: OpenSUSE Leap 42.3
-- Machine or mobo model: TOSHIBA PORTEGE R700 PT311E-018019EN
-- Display connector: LVDS


Reproducing steps:
The bug occurs on a daily basis, there's no obvious trigger.


Additional info:
The log points to a crash dump which is attach to this bug report.



If more information is needed, please let me know.

Best Regards
SH
Comment 1 Elizabeth 2018-03-06 21:01:42 UTC
Hi! If this is still reproducible, could you try latest mesa release 17.3.6? It had fixed various GPU hangs including some DE ones.
Comment 2 in+fd 2018-04-23 19:19:30 UTC
Hello, Mesa 17.3.6 is not available for my distro, but 18.0.1. Does it help if I try to reproduce the issue with 18.0.1 ?
Comment 3 Mark Janes 2018-04-23 20:07:46 UTC
18.0.1 is even better, thanks.
Comment 4 in+fd 2018-04-26 19:36:25 UTC
Since I didn't use the system for some time, there were many outdated packages.

At first I updated Mesa 18.0.1 and it's dependencies only and not any other packages. After the installation I rebooted and logged in to KDE. After a few seconds of doing nothing, the issue reappeared and the whole system froze. I had to hard reset the system.

Afterwards I retried after installing all outdated packages. This time the system froze immediately after the login to KDE: Black screen, no cursor, no tty switch (ALT + F-Keys).


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.