[Intel-gfx] [drm/i915/gem] 59dd13ad31: phoronix-test-suite.jxrendermark.RadialGradientPaint.1024x1024.operations_per_second -54.0% regression

[Oliver Sang]

On Fri, Nov 13, 2020 at 04:27:13PM +0200, Joonas Lahtinen wrote:
> Hi,
> 
> Could you add intel-gfx at lists.freedesktop.org into reports going
> forward.
> 
> Quoting kernel test robot (2020-11-11 17:58:11)
> > 
> > Greeting,
> > 
> > FYI, we noticed a -54.0% regression of phoronix-test-suite.jxrendermark.RadialGradientPaint.1024x1024.operations_per_second due to commit:
> 
> How many runs are there on the bad version to ensure the bisect is
> repeatable?

test 4 times.
zxing at inn:/result/phoronix-test-suite/performance-true-Radial_Gradient_Paint-1024x1024-jxrendermark-1.2.4-ucode=0xd6-monitor=da39a3ee/lkp-cfl-d1/debian-x86_64-phoronix/x86_64-rhel-8.3/gcc-9/59dd13ad310793757e34afa489dd6fc8544fc3da$ grep -r "operations_per_second" */stats.json
0/stats.json: "phoronix-test-suite.jxrendermark.RadialGradientPaint.1024x1024.operations_per_second": 4133.487932,
1/stats.json: "phoronix-test-suite.jxrendermark.RadialGradientPaint.1024x1024.operations_per_second": 4120.421503,
2/stats.json: "phoronix-test-suite.jxrendermark.RadialGradientPaint.1024x1024.operations_per_second": 4188.414835,
3/stats.json: "phoronix-test-suite.jxrendermark.RadialGradientPaint.1024x1024.operations_per_second": 4068.549514,

> 
> According to Chris test has various factors affecting why the result
> could fluctuate and has been known. Reverting the patch did not have
> an effect on the benchmark and was not expected to do so, either.
> 
> Is there some mechanism to queue a re-run?
> 
> Would it make sense to do further runs before sending out the e-mail
> to avoid false positives.
> 
> It could of course be also solved by sticking to tests that have less
> fluctuation in them.
> 
> Regards, Joonas
> 
> > 
> > 
> > commit: 59dd13ad310793757e34afa489dd6fc8544fc3da ("drm/i915/gem: Flush coherency domains on first set-domain-ioctl")
> > https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
> > 
> > 
> > in testcase: phoronix-test-suite
> > on test machine: 12 threads Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz with 8G memory
> > with following parameters:
> > 
> >         need_x: true
> >         test: jxrendermark-1.2.4
> >         option_a: Radial Gradient Paint
> >         option_b: 1024x1024
> >         cpufreq_governor: performance
> >         ucode: 0xd6
> > 
> > test-description: The Phoronix Test Suite is the most comprehensive testing and benchmarking platform available that provides an extensible framework for which new tests can be easily added.
> > test-url: http://www.phoronix-test-suite.com/
> > 
> > 
> > 
> > If you fix the issue, kindly add following tag
> > Reported-by: kernel test robot <oliver.sang at intel.com>
> > 
> > 
> > Details are as below:
> > -------------------------------------------------------------------------------------------------->
> > 
> > 
> > To reproduce:
> > 
> >         git clone https://github.com/intel/lkp-tests.git
> >         cd lkp-tests
> >         bin/lkp install job.yaml  # job file is attached in this email
> >         bin/lkp run     job.yaml
> > 
> > =========================================================================================
> > compiler/cpufreq_governor/kconfig/need_x/option_a/option_b/rootfs/tbox_group/test/testcase/ucode:
> >   gcc-9/performance/x86_64-rhel-8.3/true/Radial Gradient Paint/1024x1024/debian-x86_64-phoronix/lkp-cfl-d1/jxrendermark-1.2.4/phoronix-test-suite/0xd6
> > 
> > commit: 
> >   0dccdba51e ("Merge tag 'gvt-fixes-2020-10-30' of https://github.com/intel/gvt-linux into drm-intel-fixes")
> >   59dd13ad31 ("drm/i915/gem: Flush coherency domains on first set-domain-ioctl")
> > 
> > 0dccdba51e852271 59dd13ad310793757e34afa489d 
> > ---------------- --------------------------- 
> >          %stddev     %change         %stddev
> >              \          |                \  
> >       8980 ±  2%     -54.0%       4127        phoronix-test-suite.jxrendermark.RadialGradientPaint.1024x1024.operations_per_second
> >       9.00           +13.9%      10.25 ±  4%  phoronix-test-suite.time.percent_of_cpu_this_job_got
> > 
> > 
> >                                                                                 
> >                                                                                 
> >                                                                                 
> >   10000 +-------------------------------------------------------------------+   
> >         |                                                                   |   
> >    9000 |-+.+. .+.+.+.+.+.   .+. .+.   .+. .+.+.     .+. .+.   .+. .+.+.   .|   
> >         |.+   +           +.+   +   +.+   +     +.+.+   +   +.+   +     +.+ |   
> >         |                                                                   |   
> >    8000 |-+                                                                 |   
> >         |                                                                   |   
> >    7000 |-+                                                                 |   
> >         |                                                                   |   
> >    6000 |-+                                                                 |   
> >         |                                                                   |   
> >         |                                                                   |   
> >    5000 |-+                                                                 |   
> >         |                                 O                                 |   
> >    4000 +-------------------------------------------------------------------+   
> >                                                                                 
> >                                                                                 
> > [*] bisect-good sample
> > [O] bisect-bad  sample
> > 
> > 
> > 
> > Disclaimer:
> > Results have been estimated based on internal Intel analysis and are provided
> > for informational purposes only. Any difference in system hardware or software
> > design or configuration may affect actual performance.
> > 
> > 
> > Thanks,
> > Oliver Sang
> >