Message-ID: <c69021e6-2397-4cae-a640-9a3b6ce2e404@gmail.com>
Date: Sun, 28 Apr 2024 11:43:03 +0800
From: Xilin Wu <wuxilin123@...il.com>
To: Bryan O'Donoghue <pure.logic@...us-software.ie>,
 Uwe Kleine-König <u.kleine-koenig@...gutronix.de>,
 Rob Herring <robh@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>,
 Conor Dooley <conor+dt@...nel.org>, Junhao Xie <bigfoot@...ssfun.cn>,
 Neil Armstrong <neil.armstrong@...aro.org>,
 Jessica Zhang <quic_jesszhan@...cinc.com>, Sam Ravnborg <sam@...nborg.org>,
 David Airlie <airlied@...il.com>, Daniel Vetter <daniel@...ll.ch>,
 Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
 Maxime Ripard <mripard@...nel.org>, Thomas Zimmermann <tzimmermann@...e.de>,
 Bjorn Andersson <andersson@...nel.org>,
 Konrad Dybcio <konrad.dybcio@...aro.org>,
 Tengfei Fan <quic_tengfan@...cinc.com>,
 Molly Sophia <mollysophia379@...il.com>
Cc: linux-pwm@...r.kernel.org, devicetree@...r.kernel.org,
 linux-kernel@...r.kernel.org, dri-devel@...ts.freedesktop.org,
 linux-arm-msm@...r.kernel.org
Subject: Re: [PATCH 07/10] arm64: dts: qcom: sm8550: Update EAS properties

On 2024/4/25 6:45, Bryan O'Donoghue wrote:
> On 24/04/2024 16:29, Xilin Wu via B4 Relay wrote:
>> From: Xilin Wu <wuxilin123@...il.com>
>>
>> The original values provided by Qualcomm appear to be quite
>> inaccurate. Specifically, some heavy gaming tasks could be
>> improperly assigned to the A510 cores by the scheduler, resulting
>> in a CPU bottleneck. This update to the EAS properties aims to
>> enhance the user experience across various scenarios.
>>
>> The power numbers were obtained using a Type-C power meter, which
>> was directly connected to the battery connector on the AYN Odin 2
>> motherboard, acting as a fake battery.
>>
>> It should be noted that the A715 cores seem less efficient than the
>> A710 cores. Therefore, an average value has been assigned to them,
>> considering that the A715 and A710 cores share a single cpufreq
>> domain.
>>
>> Cortex-A510 cores:
>> 441 kHz, 564 mV, 43 mW, 350 Cx
>> 556 kHz, 580 mV, 59 mW, 346 Cx
>> 672 kHz, 592 mV, 71 mW, 312 Cx
>> 787 kHz, 604 mV, 83 mW, 290 Cx
>> 902 kHz, 608 mV, 96 mW, 288 Cx
>> 1017 kHz, 624 mV, 107 mW, 264 Cx
>> 1113 kHz, 636 mV, 117 mW, 252 Cx
>> 1228 kHz, 652 mV, 130 mW, 240 Cx
>> 1344 kHz, 668 mV, 146 mW, 235 Cx
>> 1459 kHz, 688 mV, 155 mW, 214 Cx
>> 1555 kHz, 704 mV, 166 mW, 205 Cx
>> 1670 kHz, 724 mV, 178 mW, 192 Cx
>> 1785 kHz, 744 mV, 197 mW, 189 Cx
>> 1900 kHz, 764 mV, 221 mW, 190 Cx
>> 2016 kHz, 784 mV, 243 mW, 188 Cx
>> Your dynamic-power-coefficient for cpu 1: 251
> 
> This looks pretty convincing and like good work.
> 
> A few questions and suggestions for your commit log.
> 
> I'd really love to know more about how you ran this test. What values 
> exactly does your power meter give you?
> 
> How did you lock the core to a specific CPU frequency ?
> 
> Maybe also give the equation to calculate Pdyn in the commit log.
> 
> https://patchwork.kernel.org/project/linux-arm-kernel/patch/1500974575-2244-1-git-send-email-wxt@rock-chips.com/#20763985
> 
> ---
> bod

The power meter accepts a fixed 4 volts input, and outputs to the 
battery connector on the board. It is also connected to a computer for 
data recording, including voltage and current.

The CPU frequency pinning and Pdyn calculation is done by a script on 
the list: [1]. I just removed the power measuring part since it was done 
on the computer with the meter.

I will improve the commit log in v2.

[1] 
https://lore.kernel.org/all/CAD=FV=U1FP0e3_AVHpauUUZtD-5X3XCwh5aT9fH_8S_FFML2Uw@mail.gmail.com/

-- 
Thanks,
Xilin Wu

Powered by blists - more mailing lists