Bug 82167 - [all]igt/gem_userptr_blits/process-exit-gtt doesn't exit /sys/kernel/debug/dri/0/clients
Summary: [all]igt/gem_userptr_blits/process-exit-gtt doesn't exit /sys/kernel/debug/dr...
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: All Linux (All)
: medium normal
Assignee: Paulo Zanoni
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-08-05 03:06 UTC by lu hua
Modified: 2017-07-03 12:23 UTC (History)
2 users (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg (77.31 KB, text/plain)
2014-08-05 03:06 UTC, lu hua
no flags Details

Description lu hua 2014-08-05 03:06:02 UTC
Created attachment 104037 [details]
dmesg

==System Environment==
--------------------------
Regression: No, new case
Non-working platforms: all

==kernel==
--------------------------
(drm-intel-nightly)ddbc288690cd888dd6f5629ccba9a2691a70b7b0

==Bug detailed description==
It happens on all platform with -fixes, -queued and -nightly kernel.
New case gem_userptr_blits/process-exit-gtt's return code is 0, the result shows pass, but the pid still keep in  /sys/kernel/debug/dri/0/clients.

Run this case then run other cases, they will report "other drm clients running!" and fail.

[root@x-sgb4 tests]# cat /sys/kernel/debug/dri/0/clients
a dev   pid    uid      magic

[root@x-sgb4 tests]# ./gem_userptr_blits --run-subtest process-exit-gtt
IGT-Version: 1.7-gac31f19 (i686) (Linux: 3.16.0-rc6_drm-intel-nightly_8980a4_20140805+ i686)
Aperture size is 2048 MiB
Total RAM is 3202 MiB
Testing unsynchronized mappings...
Testing synchronized mappings...
Subtest process-exit-gtt: SUCCESS
[root@x-sgb4 tests]# cat /sys/kernel/debug/dri/0/clients
a dev   pid    uid      magic

y   0  3792     0          0
[root@x-sgb4 tests]# ps -ef |grep 3792 |grep -v grep


==Reproduce steps==
---------------------------- 
1. ./gem_userptr_blits --run-subtest process-exit-gtt
2. cat /sys/kernel/debug/dri/0/clients
Comment 1 Chris Wilson 2014-08-08 10:02:56 UTC
http://patchwork.freedesktop.org/patch/31259/
Comment 2 Guo Jinxian 2014-08-11 06:21:53 UTC
(In reply to comment #1)
> http://patchwork.freedesktop.org/patch/31259/

With this patch on latest -nightly(ee55e72e4941947049780df49f01d320f1b98a4b), the test is able to exit.

root@x-bdw05:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# ./gem_userptr_blits --run-subtest process-exit-gtt
IGT-Version: 1.7-ga17f060 (x86_64) (Linux: 3.16.0_kcloud_59d753_20140811+ x86_64)
Aperture size is 4096 MiB
Total RAM is 3876 MiB
Testing unsynchronized mappings...
Testing synchronized mappings...
Subtest process-exit-gtt: SUCCESS
root@x-bdw05:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# cat /sys/kernel/debug/dri/0/clients
a dev   pid    uid      magic

root@x-bdw05:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests#
Comment 3 Paulo Zanoni 2014-09-03 20:26:36 UTC
Patch was merged. Closing bug.
Comment 4 lu hua 2014-09-04 08:13:26 UTC
Verified.Fixed.
Comment 5 Jari Tahvanainen 2017-07-03 12:23:03 UTC
Closing old verified+fixed.


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.