[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20171018015902.GA10466@yexl-desktop>
Date: Wed, 18 Oct 2017 09:59:02 +0800
From: Ye Xiaolong <xiaolong.ye@...el.com>
To: Borislav Petkov <bp@...en8.de>
Cc: Markus Trippelsdorf <markus@...ppelsdorf.de>,
Andy Lutomirski <luto@...nel.org>, X86 ML <x86@...nel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Adam Borowski <kilobyte@...band.pl>,
Brian Gerst <brgerst@...il.com>,
Johannes Hirte <johannes.hirte@...enkhaos.de>, LKP <lkp@...org>
Subject: Re: [lkp-robot] [x86/mm] c4c3c3c2d0: will-it-scale.per_process_ops
-61.0% regression
On 10/17, Borislav Petkov wrote:
>On Tue, Oct 17, 2017 at 06:57:43AM +0200, Markus Trippelsdorf wrote:
>> On 2017.10.16 at 18:06 -0700, Andy Lutomirski wrote:
>> > On Mon, Oct 16, 2017 at 3:15 AM, Borislav Petkov <bp@...en8.de> wrote:
>> > > On Mon, Oct 16, 2017 at 10:39:17AM +0800, kernel test robot wrote:
>> > >>
>> > >> Greeting,
>> > >>
>> > >> FYI, we noticed a -61.0% regression of will-it-scale.per_process_ops due to commit:
>>
>> I think you are reading this wrong:
>> -61.0% regression means 61.0% improvement.
>
>Well, it has this:
>
>5b8809deb4b0a77f c4c3c3c2d00826c88b5c02c20e
>---------------- --------------------------
> %stddev change %stddev
> \ | \
> 448554 -61% 174892 will-it-scale.per_process_ops
Here 448554 is the average value of will-it-scale.per_process_ops for commit 5b8809deb4,
while 174892 is the average value of will-it-scale.per_process_ops for commit c4c3c3c2d0,
metric per_process_ops is calculated through raw output of will-it-scale as below
example.
2017-10-14 23:43:02 ./runtest.py context_switch1 295 process 44
tasks,processes,processes_idle,threads,threads_idle,linear
0,0,100,0,100,0
44,7723745,22.45,0,0.00,0
per_process_ops = 7723745 / 44 = 175540
percent -61% is computed by (174892 - 448554) / 448554
Thanks,
Xiaolong
>
>Xiaolong, can you first explain what those numbers mean? And how do you
>compute those 61%?
>
>Thx.
>
>--
>Regards/Gruss,
> Boris.
>
>Good mailing practices for 400: avoid top-posting and trim the reply.
Powered by blists - more mailing lists