Re: [PATCH v3 1/2] dt-bindings: watchdog: max63xx: Add GPIO binding

[Rob Herring]

On Thu, Jul 07, 2022 at 02:27:30PM +0200, Pali Rohár wrote:
> On Wednesday 06 July 2022 09:07:26 Rob Herring wrote:
> > On Tue, Jul 05, 2022 at 02:10:22AM +0200, Pali Rohár wrote:
> > > GPIO is optional and used for WDI logic.
> > > 
> > > Signed-off-by: Pali Rohár <pali@xxxxxxxxxx>
> > > ---
> > > Changes in v3:
> > > * Extend description
> > > ---
> > >  Documentation/devicetree/bindings/watchdog/maxim,max63xx.yaml | 4 ++++
> > >  1 file changed, 4 insertions(+)
> > > 
> > > diff --git a/Documentation/devicetree/bindings/watchdog/maxim,max63xx.yaml b/Documentation/devicetree/bindings/watchdog/maxim,max63xx.yaml
> > > index ab9641e845db..a0cf9e6c371d 100644
> > > --- a/Documentation/devicetree/bindings/watchdog/maxim,max63xx.yaml
> > > +++ b/Documentation/devicetree/bindings/watchdog/maxim,max63xx.yaml
> > > @@ -27,6 +27,10 @@ properties:
> > >      description: This is a 1-byte memory-mapped address
> > >      maxItems: 1
> > >  
> > > +  gpios:
> > 
> > As I said before, add a name prefix: wdi-gpios
> 
> So gpio with output direction should be really named that is input?

Names are based on the pins they connect to, so yes. The flags cell will 
make it clear it is a GPIO output.

> I really do not understand this kind of thinking and making every device
> tree description totally illogical and inconsistent with all other.

I don't understand what you mean.

When there is a 2nd GPIO used for this binding, what do we call it? 
'gpios' doesn't scale.

> > > +    description: Optional GPIO used for controlling WDI (watchdog input) when WDI bit is not mapped to memory

Look at it this way, why do I have to find and read the description here 
to know what 'gpios' is for when instead, it can be self-describing when 
I look at the datasheet and the DT to see this is the GPIO connection 
to WDI signal.

Rob