Re: [lkp] [sched/fair] ab522e33f9: [No primary changes] reaim.child_utime -32.9% improvement

[Dietmar Eggemann <dietmar.eggemann@arm.com>]

On 09/10/16 06:59, kernel test robot wrote:
> 
> FYI, we noticed a -32.9% improvement of reaim.child_utime due to commit:
> 
> commit ab522e33f91799661aad47bebb691f241a9f6bb8 ("sched/fair: Fix fixed point arithmetic width for shares and effective load")
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> 
> in testcase: reaim
> on test machine: 28 threads Intel(R) Xeon(R) CPU E5-2695 v3 @ 2.30GHz with 256G memory
> with following parameters:
> 
> 	runtime: 300s
> 	nr_task: 600
> 	test: creat_clo
> 	cpufreq_governor: performance
> 
> REAIM is an updated and improved version of AIM 7 benchmark.
> 
> 
> 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.
> 
> Details are as below:
> -------------------------------------------------------------------------------------------------->
> 
> 
> To reproduce:
> 
>         git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
>         cd lkp-tests
>         bin/lkp install job.yaml  # job file is attached in this email

I always ask myself how could I reproduce this on a slightly different
machine.

This install step didn't work for me on my machine. OTAH, the following
steps are working:

  bin/lkp install ./jobs/reaim-micro-cyclic-lkp-hsw01.yaml
  bin/lkp split-job ./jobs/reaim-micro-cyclic-lkp-hsw01.yaml

  bin/lkp run ./reaim-micro-cyclic-lkp-hsw01-300s-600-creat_clo.yaml

But when I do this, I only get the perf.data file for this specific
testrun for the kernel I run right now.

What are the steps missing to create this list of diffs for the various
counter values like 'reaim.child_utime'?

Thanks,

-- Dietmar

>         bin/lkp run     job.yaml
> 

[...]