[Bug 80578] New: [All Regression]igt/gem_fence_upload/thread-contention costs long time to execute sporadically

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Thu Jun 26 19:53:36 PDT 2014


https://bugs.freedesktop.org/show_bug.cgi?id=80578

          Priority: medium
            Bug ID: 80578
                CC: intel-gfx-bugs at lists.freedesktop.org
          Assignee: intel-gfx-bugs at lists.freedesktop.org
           Summary: [All Regression]igt/gem_fence_upload/thread-contention
                    costs long time to execute sporadically
        QA Contact: intel-gfx-bugs at lists.freedesktop.org
          Severity: normal
    Classification: Unclassified
                OS: All
          Reporter: jinxianx.guo at intel.com
          Hardware: Other
            Status: NEW
           Version: unspecified
         Component: DRM/Intel
           Product: DRI

Created attachment 101834
  --> https://bugs.freedesktop.org/attachment.cgi?id=101834&action=edit
dmesg

==System Environment==
--------------------------
Regression: Yes. 

The result was failed before Bug 80079

Non-working platforms: All

==kernel==
--------------------------
origin/drm-intel-nightly: 1087d4bf01e79523898c6c31615bf0c369e0039a(fails)
    drm-intel-nightly: 2014y-06m-25d-13h-11m-05s integration manifest
origin/drm-intel-next-queued: 91565c85b66db820f01894a971d39aaef60c4325(fails)
    drm/i915: Don't try to look up object for non-existent fb
origin/drm-intel-fixes: 8525a235c96a548873c6c5644f50df32b31f04c6(fails)
    drm/i915: vlv_prepare_pll is only needed in case of non DSI interfaces

==Bug detailed description==
-----------------------------
igt/gem_fence_upload/thread-contention costs long time to execute sporadically

Output:
root at x-bdw05:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# time
./gem_fence_upload --run-subtest thread-contention
IGT-Version: 1.7-g7ef5372 (x86_64) (Linux:
3.16.0-rc2_drm-intel-nightly_1087d4_20140626+ x86_64)
Contended upload rate for 1 linear threads:     866.974MiB/s
Contended upload rate for 1 tiled threads:      888.099MiB/s
Contended upload rate for 2 linear threads:     587.835MiB/s
Contended upload rate for 2 tiled threads:      513.421MiB/s
Contended upload rate for 4 linear threads:     438.407MiB/s
Contended upload rate for 4 tiled threads:      422.278MiB/s
Contended upload rate for 8 linear threads:       2.486MiB/s
Contended upload rate for 8 tiled threads:        2.449MiB/s
Contended upload rate for 16 linear threads:      2.313MiB/s
Contended upload rate for 16 tiled threads:       2.328MiB/s
Contended upload rate for 32 linear threads:      3.088MiB/s
Contended upload rate for 32 tiled threads:       3.089MiB/s
Contended upload rate for 64 linear threads:      3.837MiB/s
Contended upload rate for 64 tiled threads:       3.838MiB/s
Test assertion failure function thread_contention, file gem_fence_upload.c:321:
Last errno: 0, Success
Failed assertion: linear[1] > 0.75 * linear[0]
Subtest thread-contention: FAIL

real    19m50.070s
user    0m0.142s
sys     78m47.751s

==Reproduce steps==
---------------------------- 
1. ./gem_fence_upload --run-subtest thread-contention

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20140627/2bad0af1/attachment.html>


More information about the intel-gfx-bugs mailing list