Re: [PATCH 2/3] dt-bindings: iio: adc: Add GPADC for Allwinner H616

[Andre Przywara <andre.przywara@arm.com>]

On Wed, 17 Apr 2024 12:04:22 -0500
Chris Morgan <macroalpha82@gmail.com> wrote:

Hi,

> From: Chris Morgan <macromorgan@hotmail.com>
> 
> Add support for the GPADC for the Allwinner H616. It is identical to
> the existing ADC for the D1/T113s/R329/T507 SoCs.

The H616 is using the same die as the T507 and the H700, and since the
T507 is already mentioned in the commit message for the original
binding, I wonder if we actually need a new compatible?
I guess we follow the usual approach and provide this new
per-SoC compatible string, since the D1 and H616 SoCs are quite
different? Just wanted to point this out...

> Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
> ---
>  .../bindings/iio/adc/allwinner,sun20i-d1-gpadc.yaml      | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/iio/adc/allwinner,sun20i-d1-gpadc.yaml b/Documentation/devicetree/bindings/iio/adc/allwinner,sun20i-d1-gpadc.yaml
> index 7ef46c90ebc8..da605a051b94 100644
> --- a/Documentation/devicetree/bindings/iio/adc/allwinner,sun20i-d1-gpadc.yaml
> +++ b/Documentation/devicetree/bindings/iio/adc/allwinner,sun20i-d1-gpadc.yaml
> @@ -11,8 +11,13 @@ maintainers:
>  
>  properties:
>    compatible:
> -    enum:
> -      - allwinner,sun20i-d1-gpadc
> +    oneOf:
> +      - enum:
> +          - allwinner,sun20i-d1-gpadc
> +      - items:
> +          - enum:
> +              - allwinner,sun50i-h616-gpadc
> +          - const: allwinner,sun20i-d1-gpadc

I think a more compact way to write this would be:
	oneOf:
	  - const: allwinner,sun20i-d1-gpadc
	  - items:
	      - const: allwinner,sun50i-h616-gpadc
	      - const: allwinner,sun20i-d1-gpadc

In general: the description in the T507 manual looks the same as in the
D1 manual, just with the former having 4, and the latter 2 channels. If
I understand correctly, this difference is not modelled in the binding
(or the Linux driver, fwiw), so using the compatible fallback looks
good.

Cheers,
Andre

>  
>    "#io-channel-cells":
>      const: 1