Re: [PATCH] dt-bindings: clock: samsung,s3c6400-clock: convert to DT Schema

[Conor Dooley]

On Mon, Mar 18, 2024 at 05:20:50PM +0100, Krzysztof Kozlowski wrote:
> On 17/03/2024 16:49, Conor Dooley wrote:
> > On Sun, Mar 17, 2024 at 04:26:55PM +0100, Krzysztof Kozlowski wrote:
> >> On 17/03/2024 16:23, Conor Dooley wrote:
> >>> On Tue, Mar 12, 2024 at 07:50:35PM +0100, Krzysztof Kozlowski wrote:
> >>>> Convert Samsung S3C6400/S3C6410 SoC clock controller bindings to DT
> >>>> schema.
> >>>
> >>>> +description: |
> >>>> +  There are several clocks that are generated outside the SoC. It is expected
> >>>> +  that they are defined using standard clock bindings with following
> >>>> +  clock-output-names:
> >>>> +   - "fin_pll" - PLL input clock (xtal/extclk) - required,
> >>>> +   - "xusbxti" - USB xtal - required,
> >>>> +   - "iiscdclk0" - I2S0 codec clock - optional,
> >>>> +   - "iiscdclk1" - I2S1 codec clock - optional,
> >>>> +   - "iiscdclk2" - I2S2 codec clock - optional,
> >>>> +   - "pcmcdclk0" - PCM0 codec clock - optional,
> >>>> +   - "pcmcdclk1" - PCM1 codec clock - optional, only S3C6410.
> >>>
> >>> I know you've only transfered this from the text binding, but what is
> >>> the relevance of this to the binding for this clock controller? This
> >>> seems to be describing some ?fixed? clocks that must be provided in
> >>> addition to this controller. I guess there's probably no other suitable
> >>> place to mention these?
> >>
> >> To make it correct, these should be made clock inputs to the clock
> >> controller, even if the driver does not take them, however that's
> >> obsolete platform which might be removed from kernel this or next year,
> >> so I don't want to spend time on it.
> > 
> > I think the comment should probably mention that these are the expected
> > inputs, part of me thought that that was what you were getting at but I
> > wasn't sure if instead they were inputs to some other IP on the SoC.
> 
> I can change it, but just to emphasize: in half a year or next year we
> will probably remove entire platform, thus also this binding.

I know, I saw that. I don't really care what you do given the platform
is being deleted and it is unlikely that anyone is actually going to be
assembling a from-scratch dtsi for this SoC. On the other hand, if
you're doing a conversion, even in this scenario, I think it should be
clear. 
I didn't ack the patch cos I figured you were taking the patch via the
samsung tree (and on to Stephen) yourself, but here:
Acked-by: Conor Dooley <conor.dooley@xxxxxxxxxxxxx>

I'd rather argue about the definition of erratum instead of this :)

Attachment: signature.asc
Description: PGP signature