Bug 91052 - [IVB] GPU HANG: ecode 7:0:0x85fffff8, reason: Ring hung
Summary: [IVB] GPU HANG: ecode 7:0:0x85fffff8, reason: Ring hung
Status: CLOSED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) All
: medium normal
Assignee: Intel 3D Bugs Mailing List
QA Contact: Intel 3D Bugs Mailing List
URL:
Whiteboard:
Keywords:
: 93863 (view as bug list)
Depends on:
Blocks:
 
Reported: 2015-06-22 09:06 UTC by Jiri Slaby
Modified: 2016-11-04 13:58 UTC (History)
3 users (show)

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


Attachments
dmesg (110.28 KB, text/plain)
2015-06-22 09:06 UTC, Jiri Slaby
Details
/sys/class/drm/card0/error (2.19 MB, text/plain)
2015-06-22 09:07 UTC, Jiri Slaby
Details
Archlinux dmesg (64.87 KB, text/plain)
2015-06-23 12:06 UTC, Markus Härer
Details
Archlinux /sys/class/drm/card0/error (1.49 MB, text/plain)
2015-06-23 12:07 UTC, Markus Härer
Details

Description Jiri Slaby 2015-06-22 09:06:42 UTC
Created attachment 116644 [details]
dmesg

[drm] GPU HANG: ecode 7:0:0x85fffff8, reason: Ring hung, action: reset
Comment 1 Jiri Slaby 2015-06-22 09:07:17 UTC
Created attachment 116645 [details]
/sys/class/drm/card0/error
Comment 2 Markus Härer 2015-06-23 12:06:06 UTC
Created attachment 116672 [details]
Archlinux dmesg

Same problem here on Archlinux after upgrading mesa and intel driver:

[2015-06-22 23:51] [ALPM] upgraded mesa (10.5.7-1 -> 10.6.0-1)
[2015-06-22 23:51] [ALPM] upgraded mesa-libgl (10.5.7-1 -> 10.6.0-1)
[2015-06-22 23:51] [ALPM] upgraded xf86-video-intel (2.99.917-5 -> 1:2.99.917+364+gb24e758-1)
Comment 3 Markus Härer 2015-06-23 12:07:18 UTC
Created attachment 116673 [details]
Archlinux /sys/class/drm/card0/error
Comment 4 yann 2016-05-20 09:43:31 UTC
*** Bug 93863 has been marked as a duplicate of this bug. ***
Comment 5 yann 2016-09-28 13:03:04 UTC
(In reply to Markus Härer from comment #3)
> Created attachment 116673 [details]
> Archlinux /sys/class/drm/card0/error

Markus, your issue is different from Jiri's one. There were improvements pushed in kernel and Mesa that will benefit to your system, so please re-test with latest kernel & Mesa to see if this issue is still occurring.

From this error dump, hung is happening in render ring batch with active head at 0x037070f8, with 0x60020100 (CONSTANT_BUFFER) as IPEHR, which may require an update of your mesa version.

Batch extract (around 0x037070f8):

0x037070dc:      0x78090003: 3DSTATE_VERTEX_ELEMENTS
0x037070e0:      0x04850000:    buffer 0: valid, type 0x0085, src offset 0x0000 bytes
0x037070e4:      0x11230000:    (X, Y, 0.0, 1.0), dst offset 0x00 bytes
0x037070e8:      0x04850008:    buffer 0: valid, type 0x0085, src offset 0x0008 bytes
0x037070ec:      0x11230000:    (X, Y, 0.0, 1.0), dst offset 0x00 bytes
0x037070f0:      0x60020100: CONSTANT_BUFFER: valid
0x037070f4:      0x042af002:    offset: 0x042af000, length: 192 bytes
0x037070f8:      0x7b009004: 3DPRIMITIVE: tri list random
0x037070fc:      0x00000006:    vertex count
0x03707100:      0x00000000:    start vertex
0x03707104:      0x00000001:    instance count
0x03707108:      0x00000000:    start instance
0x0370710c:      0x00000000:    index bias
Comment 6 yann 2016-09-28 13:05:44 UTC
(In reply to Jiri Slaby from comment #1)
> Created attachment 116645 [details]
> /sys/class/drm/card0/error

We seem to have neglected the bug a bit, apologies.

There were improvements pushed in kernel and Mesa that will benefit to your system, so please re-test with latest kernel & Mesa to see if this issue is still occurring.

In parallel, assigning to Mesa product (please let me know if I am mistaken with this GPU Hang).

Kernel: 4.0.5-1.g0f16252-desktop
Platform: IvyBridge (pci id: 0x0166)
Mesa: [Please confirm your mesa version]

From this error dump, hung is happening in render ring batch with active head at 0x11637e28, with 0x7a000003 (PIPE_CONTROL) as IPEHR.

Batch extract (around 0x11637e28):

0x11637df8:      0x7b000005: 3DPRIMITIVE:
0x11637dfc:      0x00000105:    tri strip random
0x11637e00:      0x0000000c:    vertex count
0x11637e04:      0x00000040:    start vertex
0x11637e08:      0x00000001:    instance count
0x11637e0c:      0x00000000:    start instance
0x11637e10:      0x00000000:    index bias
0x11637e14:      0x7a000003: PIPE_CONTROL
0x11637e18:      0x00106000:    qword write, cs stall, depth stall,
0x11637e1c:      0x0ccb3000:    destination address
0x11637e20:      0x00000000:    immediate dword low
0x11637e24:      0x00000000:    immediate dword high
0x11637e28:      0x78150005: 3DSTATE_CONSTANT_VS
0x11637e2c:      0x00000003:    len 0 = 3, len 1 = 0
0x11637e30:      0x00000000:    len 2 = 0, len 3 = 0
0x11637e34:      0x00007921:    pointer to constbuf 0
0x11637e38:      0x00000000:    pointer to constbuf 1
0x11637e3c:      0x00000000:    pointer to constbuf 2
0x11637e40:      0x00000000:    pointer to constbuf 3
Comment 7 Matt Turner 2016-11-04 00:39:10 UTC
We don't have enough information.


- What kernel version are you using? (uname -a)
- What Mesa version are you using? (glxinfo | grep Mesa)
- What were you doing at the time of the hang? Is it reproducible?


Please test a new version of Mesa (12 or 13) and mark as REOPENED
if you can reproduce and RESOLVED/* if you cannot reproduce.
Comment 8 Jiri Slaby 2016-11-04 09:17:22 UTC
(In reply to Matt Turner from comment #7)
> We don't have enough information.
> 
> 
> - What kernel version are you using? (uname -a)

4.0.5 according to the first attachment.

> - What Mesa version are you using? (glxinfo | grep Mesa)
> - What were you doing at the time of the hang? Is it reproducible?

That, I don't remember after a year and a half.

I don't think I saw it in recently.


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.