Date:   Fri, 7 May 2021 13:07:35 +0000
From:   "Song Bao Hua (Barry Song)" <song.bao.hua@...ilicon.com>
To:     Dietmar Eggemann <dietmar.eggemann@....com>,
        Vincent Guittot <vincent.guittot@...aro.org>
CC:     "tim.c.chen@...ux.intel.com" <tim.c.chen@...ux.intel.com>,
        "catalin.marinas@....com" <catalin.marinas@....com>,
        "will@...nel.org" <will@...nel.org>,
        "rjw@...ysocki.net" <rjw@...ysocki.net>,
        "bp@...en8.de" <bp@...en8.de>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        "mingo@...hat.com" <mingo@...hat.com>,
        "lenb@...nel.org" <lenb@...nel.org>,
        "peterz@...radead.org" <peterz@...radead.org>,
        "rostedt@...dmis.org" <rostedt@...dmis.org>,
        "bsegall@...gle.com" <bsegall@...gle.com>,
        "mgorman@...e.de" <mgorman@...e.de>,
        "msys.mizuma@...il.com" <msys.mizuma@...il.com>,
        "valentin.schneider@....com" <valentin.schneider@....com>,
        "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
        Jonathan Cameron <jonathan.cameron@...wei.com>,
        "juri.lelli@...hat.com" <juri.lelli@...hat.com>,
        "mark.rutland@....com" <mark.rutland@....com>,
        "sudeep.holla@....com" <sudeep.holla@....com>,
        "aubrey.li@...ux.intel.com" <aubrey.li@...ux.intel.com>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>,
        "x86@...nel.org" <x86@...nel.org>, "xuwei (O)" <xuwei5@...wei.com>,
        "Zengtao (B)" <prime.zeng@...ilicon.com>,
        "guodong.xu@...aro.org" <guodong.xu@...aro.org>,
        yangyicong <yangyicong@...wei.com>,
        "Liguozhu (Kenneth)" <liguozhu@...ilicon.com>,
        "linuxarm@...neuler.org" <linuxarm@...neuler.org>,
        "hpa@...or.com" <hpa@...or.com>
Subject: RE: [RFC PATCH v6 3/4] scheduler: scan idle cpu in cluster for tasks
 within one LLC



> -----Original Message-----
> From: Dietmar Eggemann [mailto:dietmar.eggemann@....com]
> Sent: Thursday, May 6, 2021 12:30 AM
> To: Song Bao Hua (Barry Song) <song.bao.hua@...ilicon.com>; Vincent Guittot
> <vincent.guittot@...aro.org>
> Cc: tim.c.chen@...ux.intel.com; catalin.marinas@....com; will@...nel.org;
> rjw@...ysocki.net; bp@...en8.de; tglx@...utronix.de; mingo@...hat.com;
> lenb@...nel.org; peterz@...radead.org; rostedt@...dmis.org;
> bsegall@...gle.com; mgorman@...e.de; msys.mizuma@...il.com;
> valentin.schneider@....com; gregkh@...uxfoundation.org; Jonathan Cameron
> <jonathan.cameron@...wei.com>; juri.lelli@...hat.com; mark.rutland@....com;
> sudeep.holla@....com; aubrey.li@...ux.intel.com;
> linux-arm-kernel@...ts.infradead.org; linux-kernel@...r.kernel.org;
> linux-acpi@...r.kernel.org; x86@...nel.org; xuwei (O) <xuwei5@...wei.com>;
> Zengtao (B) <prime.zeng@...ilicon.com>; guodong.xu@...aro.org; yangyicong
> <yangyicong@...wei.com>; Liguozhu (Kenneth) <liguozhu@...ilicon.com>;
> linuxarm@...neuler.org; hpa@...or.com
> Subject: Re: [RFC PATCH v6 3/4] scheduler: scan idle cpu in cluster for tasks
> within one LLC
> 
> On 03/05/2021 13:35, Song Bao Hua (Barry Song) wrote:
> 
> [...]
> 
> >> From: Song Bao Hua (Barry Song)
> 
> [...]
> 
> >>> From: Dietmar Eggemann [mailto:dietmar.eggemann@....com]
> 
> [...]
> 
> >>> On 29/04/2021 00:41, Song Bao Hua (Barry Song) wrote:
> >>>>
> >>>>
> >>>>> -----Original Message-----
> >>>>> From: Dietmar Eggemann [mailto:dietmar.eggemann@....com]
> >>>
> >>> [...]
> >>>
> >>>>>>>> From: Dietmar Eggemann [mailto:dietmar.eggemann@....com]
> >>>>>
> >>>>> [...]
> >>>>>
> >>>>>>>> On 20/04/2021 02:18, Barry Song wrote:
> 
> [...]
> 
> >
> > On the other hand, according to "sched: Implement smarter wake-affine logic"
> >
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/
> ?id=62470419
> >
> > Proper factor in wake_wide is mainly beneficial of 1:n tasks like
> postgresql/pgbench.
> > So using the smaller cluster size as factor might help make wake_affine false
> so
> > improve pgbench.
> >
> > From the commit log, while clients =  2*cpus, the commit made the biggest
> > improvement. In my case, It should be clients=48 for a machine whose LLC
> > size is 24.
> >
> > In Linux, I created a 240MB database and ran "pgbench -c 48 -S -T 20 pgbench"
> > under two different scenarios:
> > 1. page cache always hit, so no real I/O for database read
> > 2. echo 3 > /proc/sys/vm/drop_caches
> >
> > For case 1, using cluster_size and using llc_size will result in similar
> > tps= ~108000, all of 24 cpus have 100% cpu utilization.
> >
> > For case 2, using llc_size still shows better performance.
> >
> > tps for each test round(cluster size as factor in wake_wide):
> > 1398.450887 1275.020401 1632.542437 1412.241627 1611.095692 1381.354294
> 1539.877146
> > avg tps = 1464
> >
> > tps for each test round(llc size as factor in wake_wide):
> > 1718.402983 1443.169823 1502.353823 1607.415861 1597.396924 1745.651814
> 1876.802168
> > avg tps = 1641  (+12%)
> >
> > so it seems using cluster_size as factor in "slave >= factor && master >=
> slave *
> > factor" isn't a good choice for my machine at least.
> 
> So SD size = 4 (instead of 24) seems to be too small for `-c 48`.
> 
> Just curious, have you seen the benefit of using wake wide on SD size =
> 24 (LLC) compared to not using it at all?

At least in my benchmark made today, I have not seen any benefit to use
llc_size. Always returning 0 in wake_wide() seems to be much better.

postgres@...ntu:$pgbench -i pgbench
postgres@...ench:$ pgbench -T 120 -c 48 pgbench

using llc_size, it got to 123tps
always returning 0 in wake_wide(), it got to 158tps

actually, I really couldn't reproduce the performance improvement
the commit "sched: Implement smarter wake-affine logic" mentioned.
on the other hand, the commit log didn't present the pgbench command
parameter used. I guess the benchmark result will highly depend on
the command parameter and disk I/O speed.

Thanks
Barry

Powered by blists - more mailing lists