Re: [lkp-robot] [x86/mm/gup] e585513b76: will-it-scale.per_thread_ops -6.9% regression

[Thorsten Leemhuis]

Lo! I added this report to my regression report, but I'm a bit unsure if
that was a good idea.

Is this report still valid (is there a place where I could have
checked)? It's a few weeks old already, but it seems it was ignored. Or
was it discussed (or maybe even fixed) somewhere else?

And is it relevant at all? Or is it considered something like "yes, this
benchmark got slower, but we do not expect to see any problems in the
real world"?

Ciao, Thorsten

On 10.07.2017 04:40, kernel test robot wrote:
> 
> Greeting,
> 
> FYI, we noticed a -6.9% regression of will-it-scale.per_thread_ops due to commit:
> 
> 
> commit: e585513b76f7b05d08ca3fb250fed11f6ba46ee5 ("x86/mm/gup: Switch GUP to the generic get_user_page_fast() implementation")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
> 
> in testcase: will-it-scale
> on test machine: qemu-system-x86_64 -enable-kvm -cpu host -smp 4 -m 5G
> with following parameters:
> 
> 	test: futex2
> 
> test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
> test-url: https://github.com/antonblanchard/will-it-scale
> 
> 
> 
> Details are as below:
> -------------------------------------------------------------------------------------------------->
> 
> 
> To reproduce:
> 
>         git clone https://github.com/01org/lkp-tests.git
>         cd lkp-tests
>         bin/lkp qemu -k <bzImage> job-script  # job-script is attached in this email
> 
> testcase/path_params/tbox_group/run: will-it-scale/futex2/vm-lkp-a03
> 
> 6c690ee1039b251e  e585513b76f7b05d08ca3fb250  
> ----------------  --------------------------  
>          %stddev      change         %stddev
>              \          |                \  
>     882247              -7%     821521        will-it-scale.per_thread_ops
>      33180 Â126%     -3e+04       3507 Â 47%  latency_stats.avg.max
>     171890 Â154%     -1e+05      22345 Â 66%  latency_stats.max.max
> 
> 
> 
>                             will-it-scale.per_thread_ops
> 
>   920000 ++-----------------------------------------------------------------+
>          |                *                                                 |
>   900000 ++ *.      .**. + :  .*      .*.                          .**      |
>          *.*  *.*  *    *  *.*  *.*.**   *.*                     .*   :   *.*
>          |       :+                         :.*                 *     :   : |
>   880000 ++      *                          *  +           *   *       :.*  |
>          |                                      * .*.**.*. :+ +        *    |
>   860000 ++                                      *        *  *              |
>          |                                                                  |
>   840000 ++                                                                 |
>          |              O O                 O                               |
>          | OO              O O OO O O  O   O  O         O      O  O         |
>   820000 O+   O O                    O   O            O   OO O  O           |
>          |           OO                         OO O O                      |
>   800000 ++------O-O--------------------------------------------------------+
> 
> 
> 
>   [*] 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,
> Xiaolong
>