Bug 100038 - [BDW] gpu hangs ecode 8:0:0x84d77c1c
Summary: [BDW] gpu hangs ecode 8:0:0x84d77c1c
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: unspecified
Hardware: Other Linux (All)
: medium normal
Assignee: Intel 3D Bugs Mailing List
QA Contact: Intel 3D Bugs Mailing List
URL:
Whiteboard: ReadyForDev
Keywords:
Depends on:
Blocks:
 
Reported: 2017-03-02 17:27 UTC by Rico
Modified: 2018-03-22 19:37 UTC (History)
1 user (show)

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


Attachments
GPU crash dump (375.65 KB, text/plain)
2017-03-02 17:27 UTC, Rico
Details
dmesg output (71.02 KB, text/plain)
2017-06-28 17:53 UTC, Rico
Details

Description Rico 2017-03-02 17:27:46 UTC
Created attachment 130035 [details]
GPU crash dump

[   40.825253] [drm] stuck on render ring
[   40.826898] [drm] GPU HANG: ecode 8:0:0x84d77c1c, in Xorg [1600], reason: Ring hung, action: reset
[   40.826900] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[   40.826901] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[   40.826902] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[   40.826903] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[   40.826904] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[   40.830690] drm/i915: Resetting chip after gpu hang
[   48.801544] [drm] stuck on render ring
[   48.803213] [drm] GPU HANG: ecode 8:0:0x84d77c1c, in Xorg [1600], reason: Ring hung, action: reset
[   48.805531] [drm:i915_set_reset_status [i915]] *ERROR* gpu hanging too fast, banning!
[   48.807965] drm/i915: Resetting chip after gpu hang
Comment 1 Elizabeth 2017-06-27 21:03:24 UTC
(In reply to Rico from comment #0)
> Created attachment 130035 [details]
> GPU crash dump
> 

Hello Rico,
Is this bug still valid? Is reproducible? If so, could you please add dmesg with drm.debug=0xe on grub? Thank you.
Comment 2 Rico 2017-06-28 17:53:52 UTC
Created attachment 132303 [details]
dmesg output

dmesg attached.
But only reproducible with start splashscreen "mouse", with no/or other xfce-splashscreen no problem
Comment 3 Elizabeth 2017-08-24 15:43:38 UTC
From GPU crash dump:
0xff02d5d8:      0x7b000005: 3DPRIMITIVE: fail sequential
0xff02d5dc:      0x00000000:    vertex count
0xff02d5e0:      0x0000101a:    start vertex
0xff02d5e4:      0x00000fa8:    instance count
0xff02d5e8:      0x00000001:    start instance
0xff02d5ec:      0x00000000:    index bias
0xff02d5f0:      0x00000000: MI_NOOP
0xff02d5f4:      0x05000000: MI_BATCH_BUFFER_END

With last action executed when hang:
0xff02d5d8:      0x7b000005: 3DPRIMITIVE: fail sequential

This seems to be a Mesa problem. Please add your Mesa info and try last version.
Comment 4 Elizabeth 2018-03-21 20:44:03 UTC
Hello Rico, is this still reproducible with latest mesa release 17.3.6?
Comment 5 Rico 2018-03-22 19:37:43 UTC
not reproducible with: 3.0 Mesa 17.2.8


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.