Received-SPF: pass (google.com: domain of linux-kernel+bounces-187020-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Date: Thu, 23 May 2024 15:54:00 +1000
From: John Watts <contact@jookia.org>
To: =?utf-8?B?44GN44GP44Gh44KD44KT44GV44KT?= <kikuchan98@gmail.com>
Cc: privatesub2@gmail.com, aou@eecs.berkeley.edu, bigunclemax@gmail.com,
	conor+dt@kernel.org, devicetree@vger.kernel.org,
	fusibrandon13@gmail.com, jernej.skrabec@gmail.com,
	krzk+dt@kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, linux-pwm@vger.kernel.org,
	linux-riscv@lists.infradead.org, linux-sunxi@lists.linux.dev,
	mkl@pengutronix.de, p.zabel@pengutronix.de, palmer@dabbelt.com,
	paul.walmsley@sifive.com, robh@kernel.org, samuel@sholland.org,
	ukleinek@kernel.org, wens@csie.org
Subject: Re: [PATCH v9 0/3] Add support for Allwinner PWM on D1/T113s/R329
 SoCs
Message-ID: <Zk7Z-MfdF-YKWeJ6@titan>
References: <CAG40kxGMu-TSchNezkcC_A97hzPnWU3KxeL-X-hJfPhjr_COyQ@mail.gmail.com>
 <Zk6yK3U9tgxOxcBb@titan>
 <CAG40kxFxE_Oj+9aCzGku0a3KFHpuW8ai=gEkV9M8==5gwmjdEA@mail.gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <CAG40kxFxE_Oj+9aCzGku0a3KFHpuW8ai=gEkV9M8==5gwmjdEA@mail.gmail.com>

On Thu, May 23, 2024 at 02:30:32PM +0900, きくちゃんさん wrote:
> Hello John,

Hello,

> 
> Your method is quite impressive and challenging.
> 
> I think PWM is not only for LCD backlighting, but also for signal generation.
> For instance, imagine an application software that sends a stream to
> one PWM channel to synthesize sound by changing its period (plays a
> square wave).
> By your method, if the other PWM channel is used for LCD backlighting,
> it may flicker repeatedly on DIV_M changes.
> (Or simply both channels for streo sound synthesize, you may hear lots
> of pop noise)
> 
> It means the setting of one channel can affect the other, which users
> may not anticipate.

Does Linux guarantee a flicker-free experience with setting up PWM
channels, or that it doesn't affect other channels?
How do other drivers handle this situation?

Maybe out-sourcing the clocking to the device tree and letting people set the
clock and common divisor in the DT would be a good idea, with the HOSC and a
common divisor of 1 by default, then error if the value can't be found.

This would work for periods from 41ns up to 687ms by default. Using APB would
give a better resolution as by default it hangs around 5ns which is nice and
divisible by 10.

> 
> Best regards,
> kikuchan.

John.