Bug 104455 - [SNB]GPU hang in "FTL: Faster Than Light"
Summary: [SNB]GPU hang in "FTL: Faster Than Light"
Status: RESOLVED WORKSFORME
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: 2018-01-02 13:35 UTC by Francis Herne
Modified: 2018-03-21 22:35 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
Contents of /sys/class/drm/card0/error (34.13 KB, text/plain)
2018-01-02 13:35 UTC, Francis Herne
Details

Description Francis Herne 2018-01-02 13:35:18 UTC
Created attachment 136492 [details]
Contents of /sys/class/drm/card0/error

Sandybridge laptop (Thinkpad T420s) using mesa and kernel from Arch Linux packages.

Mesa 17.3, kernel 4.14.7, FTL Advanced Edition (standalone, not through Steam).

The display froze repeatedly for about 20s each time, but recovered afterwards.

The problem doesn't seem to be easily reproducible; it only happened after a long period and hasn't recurred.

Dump attached.

(maybe see also https://bugs.freedesktop.org/show_bug.cgi?id=100898 , although that's for Skylake).

Jan 01 23:58:44 tappets kernel: [drm] GPU HANG: ecode 6:0:0x85fffffc, in FTL [23736], reason: Hang on rcs0, action: reset
Jan 01 23:58:44 tappets kernel: [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Jan 01 23:58:44 tappets kernel: [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Jan 01 23:58:44 tappets kernel: [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Jan 01 23:58:44 tappets kernel: [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Jan 01 23:58:44 tappets kernel: [drm] GPU crash dump saved to /sys/class/drm/card0/error
Jan 01 23:58:44 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Jan 01 23:58:52 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Jan 01 23:59:00 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Jan 02 00:00:45 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Jan 02 00:00:53 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Jan 02 00:01:01 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Jan 02 00:01:09 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Jan 02 00:01:17 tappets kernel: i915 0000:00:02.0: Resetting chip after gpu hang
Comment 1 Elizabeth 2018-01-05 22:57:56 UTC
If steam, do you get any trace back when hang happens? If you got any chance, could you try this branch https://cgit.freedesktop.org/mesa/mesa/?
Comment 2 Elizabeth 2018-03-21 20:03:41 UTC
Hello Francis, (In reply to Elizabeth from comment #1)
> If steam, do you get any trace back when hang happens? If you got any
> chance, could you try this branch https://cgit.freedesktop.org/mesa/mesa/?
Sorry, I miss the "standalone" part in your description. It may be beneficial to try the new Mesa release 17.3.6 which includes some fixes for games' hangs if this is still an issue.
Comment 3 Francis Herne 2018-03-21 22:35:44 UTC
I still haven't been able to reproduce the hang, either with the Mesa version I had at the time or with more recent versions up to 17.3.6.

Given the comments on IRC that there were various known hangs for early 17.3 releases I might as well close this.


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.