Re: [PATCH v2 04/15] dt-bindings: mfd: ROHM BD72720

[Linus Walleij]

On Wed, Oct 29, 2025 at 1:30 PM Matti Vaittinen
<mazziesaccount@xxxxxxxxx> wrote:
> On 28/10/2025 00:42, Linus Walleij wrote:
> > Hi Matti,
> >
> > thanks for your patch!
> >
> > On Mon, Oct 27, 2025 at 12:45 PM Matti Vaittinen
> > <mazziesaccount@xxxxxxxxx> wrote:
> >
> >> +  rohm,clkout-open-drain:
> >> +    description: clk32kout mode. Set to 1 for "open-drain" or 0 for "cmos".
> >> +    $ref: /schemas/types.yaml#/definitions/uint32
> >> +    minimum: 0
> >> +    maximum: 1
> >
> > I think CMOS is the same as "push-pull" ( I could be wrong, but I think I've
> > seen that before) so I would probably try to use the pin config standard
> > names as strings here but I'm not sure.
> >
> > rohm,clkout-bias-open-drain;
> > rohm,clkout-bias-push-pull;
> >
> > Mutually exclusive.
> >
> > Or maybe use the pattern from rohm,pin-dvs0
> > with string enumerators?
> >
> > rohm,clkout-bias = "open-drain";
> > rohm,clkout-bias = "push-pull";
> >
>
> Hmm. I kind of agree with you. Still, the way it was done in this patch
> is used by the other existing ROHM PMICs (bd71815, bd71828, bd71879). I
> am kind of reluctant to support another way in the same driver - and I
> am also reluctant to change the existing bindings as that sounds a bit
> like asking for a nose-bleed :) (I've in the past worked with some
> devices which didn't update the device-trees when kernel was updated...)
>
> Do you think you could live with using this existing convention? :)

Yeah if there are precedents, either we can reuse that or we need to
change them all, and that invariably involves deprecation and re-implementing
the parsing in several drivers in that case, which is annoying and
takes time.

It's fine with me to keep like this.

Yours,
Linus Walleij