[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJKOXPcZCks1SxwBP-vhpsYzpsrV9MQo91QNU2obgAppxfTiyQ@mail.gmail.com>
Date: Fri, 31 Jan 2020 14:31:26 +0100
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Bartlomiej Zolnierkiewicz <b.zolnierkie@...sung.com>
Cc: lukasz.luba@....com, kgene@...nel.org,
linux-arm-kernel@...ts.infradead.org,
"linux-samsung-soc@...r.kernel.org"
<linux-samsung-soc@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
devicetree@...r.kernel.org, linux-pm@...r.kernel.org,
myungjoo.ham@...sung.com, kyungmin.park@...sung.com,
Chanwoo Choi <cw00.choi@...sung.com>, robh+dt@...nel.org,
mark.rutland@....com, dietmar.eggemann@....com
Subject: Re: [PATCH 3/3] ARM: exynos_defconfig: Enable Energy Model framework
On Fri, 31 Jan 2020 at 14:30, Bartlomiej Zolnierkiewicz
<b.zolnierkie@...sung.com> wrote:
>
>
> Hi,
>
> On 1/27/20 10:54 PM, lukasz.luba@....com wrote:
> > From: Lukasz Luba <lukasz.luba@....com>
> >
> > Enable the Energy Model (EM) brings possibility to use Energy Aware
> > Scheduler (EAS). This compiles the EM but does not enable to run EAS in
> > default. The EAS only works with SchedUtil - a CPUFreq governor which
> > handles direct requests from the scheduler for the frequency change. Thus,
> > to make EAS working in default, the SchedUtil governor should be
> > configured as default CPUFreq governor. Although, the EAS might be enabled
> > in runtime, when the EM is present for CPUs, the SchedUtil is compiled and
> > then set as CPUFreq governor, i.e.:
> >
> > echo schedutil > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
> > echo schedutil > /sys/devices/system/cpu/cpu4/cpufreq/scaling_governor
> >
> > To check if EAS is ready to work, the read output from the command below
> > should show '1':
> > cat /proc/sys/kernel/sched_energy_aware
> >
> > To disable EAS in runtime simply 'echo 0' to the file above.
> >
> > Some test results, which stress the scheduler on Odroid-XU3:
> > hackbench -l 500 -s 4096
> > With mainline code and with this patch set.
> >
> > The tests have been made with and without CONFIG_PROVE_LOCKING (PL)
> > (which is set to =y in default exynos_defconfig)
> >
> > | this patch set | mainline
> > |-----------------------------------------------|---------------
> > | performance | SchedUtil | SchedUtil | performance
> > | governor | governor | governor | governor
> > | | w/o EAS | w/ EAS |
> > ----------------|---------------|---------------|---------------|---------------
> > hackbench w/ PL | 12.7s | 11.7s | 12.0s | 13.0s - 12.2s
> > hackbench w/o PL| 9.2s | 8.1s | 8.2s | 9.2s - 8.4s
>
> Would you happen to have measurements of how much power is
> saved by running hackbench using "SchedUtil governor w/ EAS"
> instead of "SchedUtil governor w/o EAS"?
That's a good point and quite important reason behind enabling (or not) EAS...
Best regards,
Krzysztof
Powered by blists - more mailing lists