Bug 82878 - [drm] GPU HANG: ecode 1:0x8ea5c276, in multiqueue0:src [20766], reason: Ring hung, action: reset
Summary: [drm] GPU HANG: ecode 1:0x8ea5c276, in multiqueue0:src [20766], reason: Ring ...
Status: RESOLVED WORKSFORME
Alias: None
Product: libva
Classification: Unclassified
Component: intel (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: ykzhao
QA Contact: Sean V Kelley
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-08-20 21:48 UTC by Anton Kochkov
Modified: 2016-12-07 03:14 UTC (History)
4 users (show)

See Also:
i915 platform:
i915 features:

Attachments
GPU crash dump (125.36 KB, application/octet-stream)
2014-08-20 21:50 UTC, Anton Kochkov 		Details
lspci -vvvnxx output (39.88 KB, text/plain)
2014-08-21 07:53 UTC, Anton Kochkov 		Details
full dmesg (137.80 KB, text/plain)
2014-08-21 07:54 UTC, Anton Kochkov 		Details
View All

Description Anton Kochkov 2014-08-20 21:48:31 UTC 
[1750628.426877] [drm] stuck on bsd ring
[1750628.427956] [drm] GPU HANG: ecode 1:0x8ea5c276, in multiqueue0:src [20766], reason: Ring hung, action: reset
[1750628.427958] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[1750628.427960] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[1750628.427961] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[1750628.427962] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[1750628.427963] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[1750630.429300] [drm] Enabling RC6 states: RC6 on, RC6p off, RC6pp off
Comment 1 Anton Kochkov 2014-08-20 21:50:04 UTC 
Created attachment 104994 [details]
GPU crash dump
Comment 2 Focus.Luo 2014-08-21 07:16:32 UTC 
Can you help to share the detail steps what you run?
Comment 3 Focus.Luo 2014-08-21 07:22:22 UTC 
can you also help to update the environment including the Hardware platform info, linux kernel info and the media driver info?
Comment 4 Anton Kochkov 2014-08-21 07:52:45 UTC 
(In reply to comment #3)
> can you also help to update the environment including the Hardware platform
> info, linux kernel info and the media driver info?

I don't remember what it was, but it can possibly be:
1) firefox
2) chromium
3) vlc
4) mpv

Attaching lspci and full dmesg

xf86-video-intel version 2.99.914
libva version 1.3.1
libva-intel-driver version 1.3.2
libdrm version 2.4.54
mesa version 10.2.4
Comment 5 Anton Kochkov 2014-08-21 07:53:38 UTC 
Created attachment 105017 [details]
lspci -vvvnxx output

lspci -vvvnxx output
Comment 6 Anton Kochkov 2014-08-21 07:54:09 UTC 
Created attachment 105018 [details]
full dmesg

dmesg output
Comment 7 ykzhao 2015-11-26 04:17:58 UTC 
Hi, Anton

    Is the issue still reproduced on the latest intel-driver?
    
    If the issue is reproduced, will you please attach the log of /sys/class/drm/card0/error?

 BTW: Can you try to disable the PPGTT and see whether it is helpful?
(The PPGTT can be disabled by adding the kernel option of "i915.enable_ppgtt=0").

Thanks
Comment 8 haihao 2016-12-07 03:14:01 UTC 
Closed as worksforme because of no response over 1 year. Please feel free to reopen the bug if you still experience this issue on SNB.

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.