Re: [PATCH v4 5/5] dt-bindings: arm: qcom: Add more sc7180 Chromebook board bindings

[Rob Herring <robh@kernel.org>]

On Mon, May 23, 2022 at 09:19:03AM -0700, Doug Anderson wrote:
> Hi,
> 
> On Sun, May 22, 2022 at 1:01 AM Krzysztof Kozlowski
> <krzysztof.kozlowski@linaro.org> wrote:
> >
> > On 20/05/2022 23:38, Douglas Anderson wrote:
> > > This adds board bindings for boards that are downstream but not quite
> > > upstream yet.
> > >
> > > Signed-off-by: Douglas Anderson <dianders@chromium.org>
> > > Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
> > > ---
> > > Normally this bindings doc would go together in the same series that
> > > adds the device trees. In this case, Joe has been sending patches
> > > supporting these Chromebooks. His most recent posting is:
> > >
> > > https://lore.kernel.org/r/20220510154406.v5.1.Id769ddc5dbf570ccb511db96da59f97d08f75a9c@changeid/
> > >
> > > If he were to add this patch to the end of his v6, however, it would
> > > make things a bit more complicated simply becuase it would cause
> > > conflicts with all the other patches in this series. ...so steady
> > > state it would be correct to keep it in the series with the device
> > > tree files, but for this one time I think it makes sense to keep all
> > > the Chromebook board bindings patches together.
> > >
> > > (no changes since v2)
> > >
> > > Changes in v2:
> > > - Use a "description" instead of a comment for each item.
> > > - Use the marketing name instead of the code name where possible.
> > >
> > >  .../devicetree/bindings/arm/qcom.yaml         | 92 +++++++++++++++++++
> > >  1 file changed, 92 insertions(+)
> > >
> > > diff --git a/Documentation/devicetree/bindings/arm/qcom.yaml b/Documentation/devicetree/bindings/arm/qcom.yaml
> > > index 3d150d1a93cd..277faf59db57 100644
> > > --- a/Documentation/devicetree/bindings/arm/qcom.yaml
> > > +++ b/Documentation/devicetree/bindings/arm/qcom.yaml
> > > @@ -263,6 +263,16 @@ properties:
> > >            - const: google,homestar
> > >            - const: qcom,sc7180
> > >
> > > +      - description: Google Kingoftown (rev0)
> > > +        items:
> > > +          - const: google,kingoftown-rev0
> > > +          - const: qcom,sc7180
> > > +
> > > +      - description: Google Kingoftown (newest rev)
> > > +        items:
> > > +          - const: google,kingoftown
> > > +          - const: qcom,sc7180
> > > +
> > >        - description: Acer Chromebook Spin 513 (rev0)
> > >          items:
> > >            - const: google,lazor-rev0
> > > @@ -364,6 +374,48 @@ properties:
> > >            - const: google,lazor-sku6
> > >            - const: qcom,sc7180
> > >
> > > +      - description: Google Mrbland with AUO panel (rev0)
> > > +        items:
> > > +          - const: google,mrbland-rev0-sku0
> > > +          - const: qcom,sc7180
> > > +
> > > +      - description: Google Mrbland with AUO panel (newest rev)
> > > +        items:
> > > +          - const: google,mrbland-sku1536
> > > +          - const: qcom,sc7180
> > > +
> > > +      - description: Google Mrbland with BOE panel (rev0)
> > > +        items:
> > > +          - const: google,mrbland-rev0-sku16
> >
> > Similar question to patch #3, this could be:
> >
> >
> > +      - description: Google Mrbland
> > +        items:
> > +          - enum:
> > +              - google,mrbland-rev0-sku0     # AUO panel (rev0)
> > +              - google,mrbland-rev0-sku16    # BOE panel (rev0)
> > +          - const: qcom,sc7180
> >
> > and the file gets smaller and easier to read.
> 
> Ah, I guess this answers the question of the description that I asked
> in the previous patch. Of course, this goes opposite of the feedback I
> got from Stephen in previous versions of the patch where he requested
> that I use "description" instead of comments for things. ;-)
> 
> In any case, for now I'll hold off waiting for other opinions here
> since I still feel that the "one entry per dts" is easier to read /
> reason about.

I leave it to the sub-arch maintainers to decide. I somewhat prefer as 
Krzysztof suggested. Some platforms (and most of the ones I converted) 
all the boards for an SoC are one entry (except for the 3 string cases). 
So the above looks like a good middle ground grouping revs or variations 
of boards.

Rob