Bug 72106 - [BDW]igt/gem_render_copy causes gpu hangs
Summary: [BDW]igt/gem_render_copy causes gpu hangs
Status: CLOSED NOTABUG
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: Other All
: high critical
Assignee: Damien Lespiau
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 72107 72108 72110
  Show dependency treegraph
 
Reported: 2013-11-28 09:21 UTC by Guo Jinxian
Modified: 2017-10-06 14:41 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg (93.34 KB, text/plain)
2013-11-28 09:21 UTC, Guo Jinxian
no flags Details
error state (2.65 MB, text/plain)
2013-11-29 03:23 UTC, Guo Jinxian
no flags Details

Description Guo Jinxian 2013-11-28 09:21:33 UTC
Created attachment 89950 [details]
dmesg

System Environment:
--------------------------
Platform: Broadwell
kernel   (drm-intel-nexe-queued)8b4f49e03901e82898540bd1189333edf9588d74
Some additional commit info:
Author: Jesse Barnes <jbarnes@virtuousgeek.org>
Date:   Mon Nov 25 15:51:16 2013 -0800

    drm/i915: split fb allocation and initialization v2

    If we use a stolen buffer, our probe callback shouldn't allocate a new
    buffer; we should re-use the one from the BIOS instead if possible.

    v2: fix locking (Jesse)

Bug detailed description:
-----------------------------
Igt/gem_render_copy causes gpu hangs.

This is the first time to run igt test on Broadwell.

Error log:
[  186.649624] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[  186.649715] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[  186.652134] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[  186.654378] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[  186.656653] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.

Steps:
---------------------------
./gem_render_copy
Comment 1 Daniel Vetter 2013-11-28 13:18:10 UTC
Again, please attach the gpu hang. Also I think Damien is working on patches for the rendercpy code, so assigning to him for now.
Comment 2 Damien Lespiau 2013-11-28 14:43:31 UTC
It'd be interesting to try again with the instruction lengths fixed (intel-gpu-tools as of today). But IIUC what Ben was suggesting, we may have something not quite right in the kernel.
Comment 3 Guo Jinxian 2013-11-29 03:23:19 UTC
Created attachment 89971 [details]
error state

Attached error state.
Comment 4 Guang Yang 2013-12-02 06:10:05 UTC
gem_render_copy testing for gen8_render_copy func blocks many other testcases, so rise its priority.
Comment 5 Ben Widawsky 2013-12-05 23:48:33 UTC
I have disabled rendercopy for now to help QA with their runs. Hangs are annoying.
Comment 6 Damien Lespiau 2013-12-05 23:52:29 UTC
I think it'll be easier to wait for the mesa guys to figure out what's wrong and copy what they do. It could be a null state missing, a magic bit not set in one of the commands. I guess it could also be a missing W/A at this point, we might want to do a pass on them.
Comment 7 Ben Widawsky 2013-12-06 00:39:39 UTC
Closing as not a bug until we have working rendercopy. Please reopen if there is still an issue when we do.
Comment 8 Guo Jinxian 2013-12-10 06:28:49 UTC
(In reply to comment #7)
> Closing as not a bug until we have working rendercopy. Please reopen if
> there is still an issue when we do.

The test had disabled.

[root@x-bdw01 tests]# ./gem_render_copy
Temporarily disabled
Test requirement not met in function main, file gem_render_copy.c:136:
Test requirement: (!(render_copy))
no render-copy function
Comment 9 Elizabeth 2017-10-06 14:41:48 UTC
Closing old verified.


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.