On Sat Jul 11 2020, Andrew Lunn wrote:
> On Sat, Jul 11, 2020 at 01:35:12PM +0200, Kurt Kanzenbach wrote:
>> On Fri Jul 10 2020, Rob Herring wrote:
>> > My bot found errors running 'make dt_binding_check' on your patch:
>> >
>> > /builds/robherring/linux-dt-review/Documentation/devicetree/bindings/net/ti,cpsw-switch.example.dt.yaml: switch@0: 'ports' is a required property
>> > /builds/robherring/linux-dt-review/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.example.dt.yaml: switch@10: 'ports' is a required property
>> 
>> Okay, the requirement for 'ports' has be to removed.
>
> Hummm....
>
> ti.cpsw is not a DSA switch. So this binding should not apply to
> it. It is a plain switchdev switch.
>
> The qcom,ipq806 is just an MDIO bus master. The DSA binding might
> apply, for a specific .dts file, if that dts file has a DSA switch on
> the bus. But in general, it should not apply.
>
> So i actually think you need to work out why this binding is being
> applied when it should not be.
>
> I suspect it is the keyword 'switch'. switch does not imply it is a
> DSA switch. There are other sorts of switches as well.

OK, makes sense. It seems like the nodename is responsible for that.

This fixes the problem:

|diff --git a/Documentation/devicetree/bindings/net/dsa/dsa.yaml b/Documentation/devicetree/bindings/net/dsa/dsa.yaml
|index bec257231bf8..4c360f8b170e 100644
|--- a/Documentation/devicetree/bindings/net/dsa/dsa.yaml
|+++ b/Documentation/devicetree/bindings/net/dsa/dsa.yaml
|@@ -18,9 +18,6 @@ description:
|   properties as required by the device it is embedded within.
| 
| properties:
|-  $nodename:
|-    pattern: "^switch(@.*)?$"
|-
|   dsa,member:
|     minItems: 2
|     maxItems: 2

Thanks,
Kurt