Bug 100792 - [SKL] GPU HANG: ecode 9:1:0xedfffff5, in Xorg [1076], reason: Hang on blitter ring, action: reset
Summary: [SKL] GPU HANG: ecode 9:1:0xedfffff5, in Xorg [1076], reason: Hang on blitter...
Status: CLOSED WORKSFORME
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-04-26 00:49 UTC by Daniel J Blueman
Modified: 2017-06-27 14:09 UTC (History)
1 user (show)

See Also:
i915 platform: SKL
i915 features: GPU hang


Attachments
Kernel log (71.42 KB, text/plain)
2017-04-26 00:49 UTC, Daniel J Blueman
no flags Details
GPU error state (74.71 KB, application/x-bzip)
2017-04-26 00:50 UTC, Daniel J Blueman
no flags Details
Xorg log (31.56 KB, text/x-log)
2017-04-26 00:50 UTC, Daniel J Blueman
no flags Details

Description Daniel J Blueman 2017-04-26 00:49:49 UTC
Created attachment 131028 [details]
Kernel log

Bug description:
Display freeze for ~60s.

System environment:
-- chipset: Skylake
-- system architecture: 64-bit
-- xf86-video-intel: 2:2.99.917+git20160706-1ubuntu1~elementary0.4.1
-- xserver: 1:7.7+13ubuntu3
-- mesa: 12.0.6-0ubuntu0.16.04.1
-- libdrm: 2.4.70-1~ubuntu16.04.1
-- kernel: 4.9.24-040924-generic
-- Linux distribution: Elementary OS based on Ubuntu 16.04.
-- Machine or mobo model: Dell XPS 15 9550
-- Display connector:  Internal eDP1

Reproducing steps:
I experienced this while running a complicated legacy X application.

Additional info:
In another X hang which didn't recover (no extra kernel messages as not a GPU hang), I saw "blocked rendering commands" in the Xorg log, so could be related.
Comment 1 Daniel J Blueman 2017-04-26 00:50:39 UTC
Created attachment 131029 [details]
GPU error state
Comment 2 Daniel J Blueman 2017-04-26 00:50:54 UTC
Created attachment 131030 [details]
Xorg log
Comment 3 Elizabeth 2017-06-21 15:49:53 UTC
(In reply to Daniel J Blueman from comment #1)
> Created attachment 131029 [details]
> GPU error state

Hello, if it's possible, could you attach the error state uncompressed, and the dmesg log with the following parameter: drm.debug=0xe
Thank you.
Comment 4 Elizabeth 2017-06-22 16:07:17 UTC
(In reply to Daniel J Blueman from comment #0)
> Created attachment 131028 [details]
> Kernel log
> 
> Bug description:
> Display freeze for ~60s.
> 
> System environment:
> -- chipset: Skylake
> -- system architecture: 64-bit
> -- xf86-video-intel: 2:2.99.917+git20160706-1ubuntu1~elementary0.4.1
> -- xserver: 1:7.7+13ubuntu3
> -- mesa: 12.0.6-0ubuntu0.16.04.1
> -- libdrm: 2.4.70-1~ubuntu16.04.1
> -- kernel: 4.9.24-040924-generic
> -- Linux distribution: Elementary OS based on Ubuntu 16.04.
> -- Machine or mobo model: Dell XPS 15 9550
> -- Display connector:  Internal eDP1
> 
> Reproducing steps:
> I experienced this while running a complicated legacy X application.
> 
> Additional info:
> In another X hang which didn't recover (no extra kernel messages as not a
> GPU hang), I saw "blocked rendering commands" in the Xorg log, so could be
> related.

Hello Daniel, it's this bug still reproducible with last kernel update??
Comment 5 Daniel J Blueman 2017-06-24 10:40:13 UTC
It's trivial to decompress the error state if you are using Windows eg using http://7-zip.org.

Since I can't easily reproduce this bug at present, you can close this report, and I'll open it if I hit the issue again. Thanks!


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.