Message-ID: <1569251515.4102.31.camel@mtksdccf07>
Date:   Mon, 23 Sep 2019 23:11:55 +0800
From:   Sam Shih <sam.shih@...iatek.com>
To:     Thierry Reding <thierry.reding@...il.com>
CC:     Mark Rutland <mark.rutland@....com>, <linux-pwm@...r.kernel.org>,
        "Ryder Lee" <ryder.lee@...iatek.com>, <devicetree@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, Rob Herring <robh+dt@...nel.org>,
        <linux-mediatek@...ts.infradead.org>,
        John Crispin <john@...ozen.org>,
        Matthias Brugger <matthias.bgg@...il.com>
Subject: Re: [PATCH v9 07/11] dt-bindings: pwm: pwm-mediatek: add a property
 "num-pwms"

On Mon, 2019-09-23 at 15:36 +0200, Thierry Reding wrote:
> On Mon, Sep 23, 2019 at 11:20:57AM +0800, Sam Shih wrote:
> > On Sat, 2019-09-21 at 02:21 +0200, Thierry Reding wrote:
> > > On Fri, Sep 20, 2019 at 06:49:07AM +0800, Sam Shih wrote:
> > > > From: Ryder Lee <ryder.lee@...iatek.com>
> > > > 
> > > > This adds a property "num-pwms" in example so that we could
> > > > specify the number of PWM channels via device tree.
> > > > 
> > > > Signed-off-by: Ryder Lee <ryder.lee@...iatek.com>
> > > > Signed-off-by: Sam Shih <sam.shih@...iatek.com>
> > > > Reviewed-by: Matthias Brugger <matthias.bgg@...il.com>
> > > > Acked-by: Uwe Kleine-König <u.kleine-koenig@...gutronix.de>
> > > > ---
> > > > Changes since v6:
> > > > Follow reviewers's comments:
> > > > - The subject should indicate this is for Mediatek
> > > > 
> > > > Changes since v5:
> > > > - Add an Acked-by tag
> > > > - This file is original v4 patch 5/10
> > > > (https://patchwork.kernel.org/patch/11102577/)
> > > > 
> > > > ---
> > > >  Documentation/devicetree/bindings/pwm/pwm-mediatek.txt | 7 ++++---
> > > >  1 file changed, 4 insertions(+), 3 deletions(-)
> > > 
> > > You failed to address Rob's questions repeatedly and I agree with him
> > > that you can just as easily derive the number of PWMs from the specific
> > > compatible string. I won't be applying this and none of the patches that
> > > depend on it.
> > > 
> > 
> > Hi, 
> > 
> > Thanks for getting back to me.
> > 
> > New pwm driver (patch 04/11 : "pwm: mediatek: allocate the clks array
> > dynamically") can support different variants with different number of
> > PWMs by the new property <num-pwms>
> > 
> > For example:
> > 1. Use "num-pwms" = <2> and assign clocks pwm1, pwm2 for mt7622
> > 2. Use "num-pwms" = <6> and assign clocks pwm1, pwm2, pwm3, pwm4, pwm5,
> > pwm6 for mt7622.
> > 
> > If we just as easily derive the number of PWMs from the specific
> > compatible string in this document:
> > 
> >    - "pwm1-6": the six per PWM clocks for mt7622
> >    
> > This looks like all "pwm1", "pwm2", "pwm3", "pwm4", "pwm5", "pwm6" is
> > required property in DT, It doesn't make sense.
> 
> I don't understand. Why doesn't that make sense? If your hardware block
> has 6 PWMs and each can be driven by its own clock, then you need to
> provide references for each of those clocks, otherwise you won't be able
> to use them.

Thank you for your instruction,
I will add all clock-names and clocks according to
hardware blocks instead of value of <num-pwms> in DT.

> >    
> > So we removed those descriptions and added  
> > 
> >    - "pwm1-N": the PWM clocks for each channel 
> >    
> >    
> > But the max number of clocks from the compatible string are still
> > important information that should be provide in this document.
> > 
> > 
> > What do you think of this?
> > 
> >    - "pwm1-N": per PWM clocks for mt2712, the max number of PWM channels
> > is 8
> > 
> >    - "pwm1-N": per PWM clocks for mt7622, the max number of PWM channels
> > is 6
> > 
> >    - "pwm1-N": per PWM clocks for mt7623, the max number of PWM channels
> > is 5
> 
> That's what's in the bindings already, isn't it?
> 
> 	 - clocks: phandle and clock specifier of the PWM reference clock.
> 	 - clock-names: must contain the following, except for MT7628 which
> 			has no clocks
> 	   - "top": the top clock generator
> 	   - "main": clock used by the PWM core
> 	   - "pwm1-8": the eight per PWM clocks for mt2712
> 	   - "pwm1-6": the six per PWM clocks for mt7622
> 	   - "pwm1-5": the five per PWM clocks for mt7623

Yes, You are right, 
I will keep original description and remove "pwm1-N" from this patch.
	   - "top": the top clock generator
	   - "main": clock used by the PWM core
	   - "pwm1-8": the eight per PWM clocks for mt2712
	   - "pwm1-6": the six per PWM clocks for mt7622
	   - "pwm1-5": the five per PWM clocks for mt7623       

Actually, MT7629 also use "mediatek,mt7622-pwm" as compatible string,
but it's hardware only support one pwm, so I was wrongly stick by
expecting "pwm1-N" in clock-names based on "num-pwms" in DT.
(that we can assign num-pwms to 1 and only provide pwm1 as clock-names)

Maybe added mt7629 description to this document can solve this simply.
	   - "pwm1": the PWM1 clock for mt7629    

> Note that the description of the "clocks" property isn't quite accurate.
> It should be something like:
> 
> 	- clocks: One phandle and clock specifier for each entry in the
> 	          "clock-names" property.
> 
> In the above you clearly describe which PWMs you have to specify for
> each generation of the hardware block.

Thanks for your advise, I will update this description
	- clocks: One phandle and clock specifier for each entry in the
	          "clock-names" property.

> > 
> >    where N starting from 1 to the maximum number of PWM channels
> >    - num-pwms: the number of PWM channels.
> 
> That's redundant information. The specific number of PWMs in already
> implied by the compatible string, so you don't need to duplicate that
> information here.

Okay, I will remove it.



Thanks,
Best Regards,
Sam 

Powered by blists - more mailing lists