RE: [PATCH v6 net-next 02/15] dt-bindings: net: dsa: add NETC switch

> Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to
> consider:
> - [Low] The `dsa,member` property lacks validation constraints for the switch
> index.
> - [Low] The `patternProperties` regex for ethernet ports is overly restrictive and
> diverges from standard DSA bindings.
> --
> 
> commit 8c698895982ff77d4bd027af7fa3d21f57f09eba
> Author: Wei Fang <[email protected]>
> 
> dt-bindings: net: dsa: add NETC switch
> 
> This commit adds the device tree bindings for the NETC switch. It is a PCIe
> function of the NETC IP that supports advanced QoS and TSN capabilities. The
> switch connects to an internal ENETC port via a light-weight pseudo MAC.
> 
> > diff --git a/Documentation/devicetree/bindings/net/dsa/nxp,netc-switch.yaml
> b/Documentation/devicetree/bindings/net/dsa/nxp,netc-switch.yaml
> > new file mode 100644
> > index 0000000000000..988688bf4467d
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/net/dsa/nxp,netc-switch.yaml
> 
> [ ... ]
> 
> > +  reg:
> > +    maxItems: 1
> > +
> > +  dsa,member:
> > +    description: >
> > +      The property indicates DSA cluster and switch index. For NETC switch,
> > +      the valid range of the switch index is 1 ~ 7, the index is reflected
> > +      in the switch tag as an indication of the switch ID where the frame
> > +      originated. The value 0 is reserved for ENETC VEPA switch, whose ID
> > +      is hardwired to zero.
> 
> The description mentions that the valid range of the switch index is 1 to 7
> and that 0 is reserved, but there do not appear to be any constraints
> enforcing this restriction in the schema definition.
> 
> Should an items array constraint be added here with a minimum of 1 and
> maximum of 7 for the second element to ensure invalid configurations like
> <0 0> are caught during schema validation?

I have added the check in the driver, I can add the following the check in
next version if no more comments.

  dsa,member:
    description: >
      The property indicates DSA cluster and switch index. For NETC switch,
      the valid range of the switch index is 1 ~ 7, the index is reflected
      in the switch tag as an indication of the switch ID where the frame
      originated. The value 0 is reserved for ENETC VEPA switch, whose ID
      is hardwired to zero.
    items:
      - true
      - minimum: 1
       maximum: 7

> 
> > +
> > +  ethernet-ports:
> > +    type: object
> > +    patternProperties:
> > +      "^ethernet-port@[0-9a-f]$":
> 
> Is this regex intentionally more restrictive than the standard DSA bindings?
> 

'ethernet-ports' for new bindings. See https://lore.kernel.org/imx/[email protected]/

> The standard DSA schema in dsa.yaml allows ^(ethernet-)?port@[0-9a-f]+$.
> Because this binding sets unevaluatedProperties to false for the
> ethernet-ports node, if a device tree author uses the standard port@1
> naming convention, the binding will reject it.
> 
> Additionally, the single hex digit restriction would prevent port numbers
> from exceeding 15. Could this be aligned with the standard pattern?

Currently, the i.MX94 NETC switch only has 4 ports, a single hex digit is
enough, and this is consistent with the definition in dsa.yaml. please
ignore this warning.