Re: [RFC v1 2/4] pinctrl: add polarfire soc mssio pinctrl driver

[Conor Dooley <conor@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 3017 bytes --]

On Wed, Nov 19, 2025 at 10:48:07PM +0100, Linus Walleij wrote:
> On Wed, Nov 19, 2025 at 7:23 PM Conor Dooley <conor@kernel.org> wrote:
> > On Wed, Nov 19, 2025 at 01:08:26PM +0100, Linus Walleij wrote:
> > > On Wed, Nov 12, 2025 at 3:33 PM Conor Dooley <conor@kernel.org> wrote:
> 
> > > > +       //TODO @Linus, it correct to group these 3? There's no control over voltage.
> > > > +       case PIN_CONFIG_INPUT_SCHMITT:
> > > > +       case PIN_CONFIG_INPUT_SCHMITT_ENABLE:
> > > > +       case PIN_CONFIG_INPUT_SCHMITT_UV:
> > >
> > > Consider not supporting some like PIN_CONFIG_INPUT_SCHMITT_UV
> > > in the bindings if they don't make any sense, and let it just return error
> > > if someone tries to do that.
> > >
> > > Isn't PIN_CONFIG_INPUT_SCHMITT_ENABLE the only one that
> > > makes sense for this hardware?
> >
> > Yeah, I just didn't know if having the others was helpful, since they
> > say things like:
> >  * @PIN_CONFIG_INPUT_SCHMITT: this will configure an input pin to run in
> >  *      schmitt-trigger mode. If the schmitt-trigger has adjustable hysteresis,
> >  *      the threshold value is given on a custom format as argument when
> >  *      setting pins to this mode.
> > which implies they should be implemented for systen not permitting
> > hysteresis adjustment.
> 
> I see.
> 
> I looked at the bindings that look like this and are not 1:1 to the
> in-kernel configs:
> 
>   input-schmitt-enable:
>     type: boolean
>     description: enable schmitt-trigger mode
> 
>   input-schmitt-disable:
>     type: boolean
>     description: disable schmitt-trigger mode
> 
>   input-schmitt-microvolt:
>     description: threshold strength for schmitt-trigger
> 
> 1. input-schmitt is missing! But it is right there in
> drivers/pinctrl/pinconf-generic.c ... All DTS files appear to be
> using input-schmitt-enable/disable and -microvolt.
> 
> 2. input-schmitt-microvolt should probably be used separately
> to set the voltage threshold and can be used in conjunction
> with input-schmitt-enable in the same node. In your case
> you probably don't want to use it at all and disallow it.
> 
> They are all treated individually in the parser.
> 
> Maybe we could patch the docs in pinconf-generic.h to make it clear that
> they are all mutually exclusive.
> 
> The DT parser is a bit primitive for these.
> For example right now it is fine with the schema
> to set input-schmitt-enable and input-schmitt-disable at the same time, and
> the result will be enabled because of parse order :/

> The real trick would be to also make the
> schema in Documentation/devicetree/bindings/pinctrl/pincfg-node.yaml
> make them at least mutually exclusive and deprecate the
> input-schmitt that noone is using, maybe that is simpler than I think?

I think that this is probably what to do. Mutual exclusion isn't
difficult to set up there and if there's no property for "input-schmitt"
then deprecating it sounds pretty reasonable?

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]