Re: [PATCH v4 2/2] dt-bindings: input: Update dtbinding for adp5588

["Nuno Sá" <noname.nuno@gmail.com>]

On Tue, 2024-07-02 at 16:42 +0100, Conor Dooley wrote:
> On Mon, Jul 01, 2024 at 10:46:09AM -0700, Dmitry Torokhov wrote:
> > On Mon, Jul 01, 2024 at 04:46:12PM +0100, Conor Dooley wrote:
> > > On Mon, Jul 01, 2024 at 04:04:51PM +0100, Utsav Agarwal via B4 Relay
> > > wrote:
> > > > From: Utsav Agarwal <utsav.agarwal@analog.com>
> > > > 
> > > > Updating dt bindings for adp5588. Following properties are now made
> > > > optional:
> > > > 	- interrupts
> > > > 	- keypad,num-rows
> > > > 	- keypad,num-columns
> > > > 	- linux,keymap
> > > > The proposed new property "gpio-only" has been added as an optional
> > > > property with an additional example.
> > > 
> > > I can see that as it is clear in the diff, but this doesn't explain why,
> > > which is what you need to do in your commit message.
> > > 
> > > > 
> > > > Signed-off-by: Utsav Agarwal <utsav.agarwal@analog.com>
> > > > ---
> > > >  .../devicetree/bindings/input/adi,adp5588.yaml     | 28
> > > > ++++++++++++++++++----
> > > >  1 file changed, 24 insertions(+), 4 deletions(-)
> > > > 
> > > > diff --git a/Documentation/devicetree/bindings/input/adi,adp5588.yaml
> > > > b/Documentation/devicetree/bindings/input/adi,adp5588.yaml
> > > > index 26ea66834ae2..158fbf02cc16 100644
> > > > --- a/Documentation/devicetree/bindings/input/adi,adp5588.yaml
> > > > +++ b/Documentation/devicetree/bindings/input/adi,adp5588.yaml
> > > > @@ -46,6 +46,11 @@ properties:
> > > >    '#gpio-cells':
> > > >      const: 2
> > > >  
> > > > +  gpio-only:
> > > > +    description:
> > > > +      This property applies if keypad,num-rows, keypad,num-columns and
> > > > +      linux,keypad are not specified. All keys will be marked as gpio.
> > > 
> > > Why is a property required for this? Is the absence of the 3 keypad
> > > properties not sufficient to determine that you're in this mode?
> > 
> > Yes, I think it should be enough.
> > 
> > > 
> > > 
> > > >    interrupt-controller:
> > > >      description:
> > > >        This property applies if either keypad,num-rows lower than 8 or
> > > > @@ -68,10 +73,6 @@ properties:
> > > >  required:
> > > >    - compatible
> > > >    - reg
> > > > -  - interrupts
> > > 
> > > I don't understand why interrupts is no longer required.
> > 
> > I think it should be possible to use this chip as a GPIO controller but
> > not an interrupt controller, in which case one does not have to wire up
> > the interrupt line from it. However this requires much more elaborate
> > binding description (i.e. no keys and no "interrupt-controller"
> > property).
> 
> Aye. I can totally understand why you might want to make the interrupt
> portion optional - but it seems unrelated to the rest of the changes in
> the patch (use as a keypad without interrupts could be possible, right?)
> and is unexplained.
> 

IMO, it is related as it's the new usecase (of only using the gpios) that
trigger the interrupt not being required anymore. No, I don't think we can use
the keypad without the interrupt line.

I guess (as you suggested before) we should check if one of the rows/columns
property is present and in that case still make 'interrupts' required.

Agree it should be better explained.

- Nuno Sá
>