Re: [PATCH] dt-bindings: mmc: renesas,sdhi: add top-level constraints

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Aug 19, 2024 at 03:38:48PM +0200, Geert Uytterhoeven wrote:
> Hi Krzysztof,
> 
> On Sun, Aug 18, 2024 at 7:29 PM Krzysztof Kozlowski
> <krzysztof.kozlowski@xxxxxxxxxx> wrote:
> > Properties with variable number of items per each device are expected to
> > have widest constraints in top-level "properties:" block and further
> > customized (narrowed) in "if:then:".  Add missing top-level constraints
> > for clocks.
> >
> > Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>
> 
> Thanks for your patch!
> 
> > --- a/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> > +++ b/Documentation/devicetree/bindings/mmc/renesas,sdhi.yaml
> > @@ -77,9 +77,13 @@ properties:
> >      minItems: 1
> >      maxItems: 3
> >
> > -  clocks: true
> > +  clocks:
> > +    minItems: 1
> > +    maxItems: 4
> >
> > -  clock-names: true
> > +  clock-names:
> > +    minItems: 1
> > +    maxItems: 4
> >
> >    dmas:
> >      minItems: 4
> 
> I am a bit puzzled by all these add-top-level-constraint patches.
> E.g. this file already constrains all of them below.
> 
> To me, it feels the same as a patch for driver code that would do:
> 
>     +   if (param < 16 || param > 512)
>     +           return -EINVAL;
>     +
>         if (hw_variant_a) {
>                 if (param < 16 || param > 256)
>                         return -EINVAL;
>                 ...
>         } else if (hw_variant_b) {
>                 if (param < 32 || param > 512)
>                         return -EINVAL;
>                 ...
>         } else /* hw_variant_c */ {
>                 if (param < 32 || param > 384)
>                         return -EINVAL;
>                 ...
>         }
> 
> What's the point?

if/then schemas can be incomplete and we don't enforce they are missing 
constraints. We could change that, but we'd have to do that everywhere. 
It would make the schemas longer. 

If you have a new chip not yet documented, but matches the fallback 
compatible (as many Renesas bindings have), then you at least 
get constraints within the existing bounds.

The keywords didn't exist when we started out. It is somewhat academic 
because we know what the implementation supports, but it is entirely 
possible a json-schema implementation doesn't support if/then schemas. 
The spec says unknown keywords are ignored.

Rob




[Index of Archives]     [Device Tree Compilter]     [Device Tree Spec]     [Linux Driver Backports]     [Video for Linux]     [Linux USB Devel]     [Linux PCI Devel]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Yosemite Backpacking]


  Powered by Linux