On Thu, Jun 04, 2026 at 09:06:37AM +0000, [email protected] wrote: > Thank you for your contribution! Sashiko AI review found 1 potential issue(s) > to consider: > - [Medium] The `data-lanes` property is defined strictly for the `endpoint` > node name, causing schema validation to fail or miss constraints if applied > to valid unit-addressed nodes like `endpoint@0`. > -- > > commit c6792a12ea4c785878c85982c6a4537b6a1901d6 > Author: Damon Ding <[email protected]> > > dt-bindings: display: bridge: analogix-dp: Add data-lanes support for endpoint > > This commit adds data-lanes property support to the port@1 endpoint in > the analogix-dp bridge binding to allow physical lane mapping configuration. > It also updates the Rockchip eDP DT node example to demonstrate its usage. > > > diff --git > > a/Documentation/devicetree/bindings/display/bridge/analogix,dp.yaml > > b/Documentation/devicetree/bindings/display/bridge/analogix,dp.yaml > > index 62f0521b09245..ecf206871cdde 100644 > > --- a/Documentation/devicetree/bindings/display/bridge/analogix,dp.yaml > > +++ b/Documentation/devicetree/bindings/display/bridge/analogix,dp.yaml > > @@ -42,13 +42,22 @@ properties: > > properties: > > port@0: > > [ ... ] > > > port@1: > > - $ref: /schemas/graph.yaml#/properties/port > > - description: > > - Port node with one endpoint connected to a dp-connector node. > > + $ref: /schemas/graph.yaml#/$defs/port-base > > + unevaluatedProperties: false > > + description: Port node with one endpoint connected to sink device > > node. > > + properties: > > + endpoint: > ^^^^^^^^ > [Severity: Medium] > Does restricting this to the exact node name 'endpoint' cause validation > issues for unit-addressed nodes? > > Because JSON Schema strictly matches object keys, a node named endpoint@0 > under port@1 will bypass this block, meaning the video-interfaces $ref > and data-lanes validation constraints are never applied.
That could be an issue... > Should this use patternProperties: "^endpoint(@[0-9a-f]+)?$": instead > of properties: endpoint: to ensure unit-addressed endpoint nodes don't > bypass these constraints? But no, this is fine as-is because it defines that only a single endpoint is valid and we'd need to fix everywhere as this is a common pattern. Rob
