Re: [PATCH v3 3/6] dt-bindings: nvmem: mediatek: efuse: Reuse mt8186-efuse in mt8188

[AngeloGioacchino Del Regno]

Il 03/10/24 10:13, Krzysztof Kozlowski ha scritto:
> On Wed, Oct 02, 2024 at 09:42:32AM +0200, AngeloGioacchino Del Regno wrote:
> > Il 02/10/24 08:11, Krzysztof Kozlowski ha scritto:
> > > On Wed, Oct 02, 2024 at 10:21:35AM +0800, Pablo Sun wrote:
> > > > mt8188 has the same GPU speed binning efuse field just
> > > > like mt8186, which requires post-processing to convert to the
> > > > bit field format specified by OPP table.
> > > >
> > > > Add the binding for the compatible list:
> > > >     "mediatek,mt8188-efuse", "mediatek,mt8186-efuse"
> > > > so mt8188 uses the same conversion.
> > > >
> > > > Suggested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@xxxxxxxxxxxxx>
> > > > Signed-off-by: Pablo Sun <pablo.sun@xxxxxxxxxxxx>
> > > > ---
> > > >    Documentation/devicetree/bindings/nvmem/mediatek,efuse.yaml | 4 ++++
> > > >    1 file changed, 4 insertions(+)
> > > >
> > > > diff --git a/Documentation/devicetree/bindings/nvmem/mediatek,efuse.yaml b/Documentation/devicetree/bindings/nvmem/mediatek,efuse.yaml
> > > > index 32b8c1eb4e80..70815a3329bf 100644
> > > > --- a/Documentation/devicetree/bindings/nvmem/mediatek,efuse.yaml
> > > > +++ b/Documentation/devicetree/bindings/nvmem/mediatek,efuse.yaml
> > > > @@ -39,6 +39,10 @@ properties:
> > > >                  - mediatek,mt8195-efuse
> > > >                  - mediatek,mt8516-efuse
> > > >              - const: mediatek,efuse
> > > > +      - items:
> > > > +          - enum:
> > > > +              - mediatek,mt8188-efuse
> > > > +          - const: mediatek,mt8186-efuse
> > >
> > > And this is not compatible with generic one? This is confusing. Why are
> > > you adding generic fallbacks if they are not valid?
> >
> > It was my suggestion to start dropping the usage of the generic "mediatek,efuse"
> > fallback, as I've seen multiple times feedback saying to not use generic fallbacks.
> >
> > Was that wrong?
>
> No, just nothing provided the background that such change is
> intentional. Please mention in commit msg that the preferred way from
> now on is not using the generic fallback.  Maybe even add it to the
> binding itself as comment, so people won't grow the enum with fallback.

Cool. Thanks for the explanation, krzk.

Pablo, can you please add a comment in the binding saying that no more entries
shall be added to the generic fallback set?

Having a comment there, instead of just into the commit message, would be (imo)
better... as then, anyone trying to add new compatibles will be more likely to
read that.

Cheers,
Angelo