Date:   Tue, 4 Aug 2020 09:47:41 +0200
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Dongdong Yang <contribute.kernel@...il.com>
Cc:     Viresh Kumar <viresh.kumar@...aro.org>, rjw@...ysocki.net,
        mingo@...hat.com, peterz@...radead.org, juri.lelli@...hat.com,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Benjamin Segall <bsegall@...gle.com>, mgorman@...e.de,
        linux-kernel@...r.kernel.org, devel@...verdev.osuosl.org,
        linux-pm@...r.kernel.org, yangdongdong@...omi.com,
        tanggeliang@...omi.com, taojun@...omi.com, huangqiwu@...omi.com,
        rocking@...ux.alibaba.com, fengwei@...omi.com,
        zhangguoquan@...omi.com, gulinghua@...omi.com, duhui@...omi.com
Subject: Re: [PATCH v3] Provide USF for the portable equipment.

A: http://en.wikipedia.org/wiki/Top_post
Q: Were do I find info about this thing called top-posting?
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

A: No.
Q: Should I include quotations after my reply?

http://daringfireball.net/2007/07/on_top

On Tue, Aug 04, 2020 at 03:34:25PM +0800, Dongdong Yang wrote:
> Appreciate Viresh for your help. I thought Peter's NAK was only for the
> initial version. I am going to upload the verified version 4. Could you
> please kindly help to further review?
> 
> 1. Motivation
> =============
> 
> The power consumption and UI response are more cared for by the portable
> equipment users.

That's not true, everyone cares about this.

> USF(User Sensitive Feedback factor) auxiliary cpufreq
> governor
> is providing more utils adjustment settings to the high level by scenario
> identification.

Odd line-wrapping :(

And what do you mean by "more utils adjustment settings to the high
level by scenario identification"?  I can not parse that at all.

> 2. Introduction
> ===============
> 
> The USF auxiliary scheduling is based on FrameBuffer and schdeutil cpufreq
> governor to adjust utils by the identificated scenario from User Space.

What is "adjust utils"?

And why is "User Space" in caps?

> It is for portable equipment which "screen off" status stands for no request
> from the user, however, the kernel is still expected to notify the user for
> UI in
> time on modem, network or powerkey events occur. In order to save power, the
> sysfs inode nonux is provided to set the utils down level on userspace
> tasks.

Having custom sysfs apis is almost never a good idea.  Do other cpufreq
governers do this?

> In addition, the portable equipment users usually heavily interact with
> devices
> by touch, and other peripherals. On "screen on" status, The boost preemptive
> counts are marking the load requirement urgent, vice versa. USF provides up
> and
> down sysfs inodes to adjust utils according to such feedback factor and the
> level setting from the user space identified scenario.
> 
> adjust_task_pred_set is as the switch to enable or disable the adjustment.
> If no USF sysfs inodes is set and no screen on or off event be received,
> adjust_task_pred_demand shall not be executed.
> 
> 3. System wide settings
> =======================
> 
> sched_usf_non_ux_r:
>         The ratio of utils is cut down on screen off. The default value is
> 0,

The line-wrapping makes it almost impossible to read here, can you fix
that up?

>         which no util be adjusted on sugov calculating utils to select

"sugov"?

> cpufreq.
>         Its range is [-100 , 0]. If its value falls into [-50, 0), the half
> of
>         utils, which calculates cpufreq, shall be  cut down. If its value
> falls
>         into [-100, -50), only a quarter of utils be left to continue to
>         calculates cpufreq.
>         It is expected to be set [-100, 0) once enter into the identificated
>         scenario, such as listen to music on screen off, and recover to 0 on
>         out of the scenario, such as screen on.

sysfs files are "one value per file", please do not parse such complex
things in the kernel.

> 
> sched_usf_up_l0_r:
>         The ratio of utils is boosted up on screen on. The default value is
> 0,
>         which no util be adjusted on sugov calculates utils to select
> cpufreq.
>         Its range is [0 , 100]. If its value falls into (0, 50], a quarter
> of
>         extra utils, which calculates cpufreq, shall be added. If its value
>         falls into (50, 100], the half of extra utils be added to continue
> to
>         calculates cpufreq.
>         It is expected to be set (0, 100] once enter into the identificated
>         scenario, such as browsing videolet on screen on, and recover to 0
> on
>         out of the scenario, such as screen off or videolet into background.
> 
> sched_usf_down_r:
>         The ratio of utils is cut down on screen on. The default value is 0,
>         which no util be adjusted on sugov calculating utils to select
> cpufreq.
>         Its range is [-100 , 0]. If its value falls into [-50, 0), the half
> of
>         utils, which calculates cpufreq, shall be  cut down. If its value
> falls
>         into [-100, -50), only a quarter of utils be left to continue to
>         calculates cpufreq.
>         It is expected to be set [-100, 0) once enter into the identificated
>         scenario, such as browsing videolet on screen on, and recover to 0
> on
>         out of the scenario, such as screen off or vidolet into background.

Why can't all of these work automatically?  Why do you need userspace
interaction here?

thanks,

greg k-h

Powered by blists - more mailing lists