Re: [PATCH v3 1/2] dt-bindings: media: add ITE IT6625/IT6626 HDMI bridge binding

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [Medium] The HDMI input port (`port@2`) is omitted from the required ports list, allowing the bridge to be defined without an input data source.
- [Medium] The `bus-type` property is not marked as required for the CSI-2 endpoints, leaving the physical layer configuration ambiguous for the IT6626 variant.
- [Medium] Endpoint constraints are tied to the exact node name `endpoint` instead of using `patternProperties`, allowing constraints to be silently bypassed if a device tree uses a numbered endpoint like `endpoint@0`.
--

--- Patch [1]: [PATCH v3 1/2] dt-bindings: media: add ITE IT6625/IT6626 HDMI bridge binding ---
commit 32438c9fff6575c3de96c9e787f4769e322a6f28
Author: Hermes Wu <Hermes.wu@ite.com.tw>

dt-bindings: media: add ITE IT6625/IT6626 HDMI bridge binding

Document the devicetree binding for the ITE IT6625/IT6626 HDMI to
MIPI CSI-2 bridge, supporting single or dual CSI-2 outputs.

> diff --git a/Documentation/devicetree/bindings/media/i2c/ite,it6625.yaml b/Documentation/devicetree/bindings/media/i2c/ite,it6625.yaml
> new file mode 100644
> index 0000000000000..9bc5584b48880
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/media/i2c/ite,it6625.yaml
> @@ -0,0 +1,159 @@

[ ... ]

> +  ports:
> +    $ref: /schemas/graph.yaml#/properties/ports
> +    properties:
> +      port@0:
> +        $ref: /schemas/graph.yaml#/$defs/port-base
> +        unevaluatedProperties: false
> +        description: CSI-2 output port MIPI0
> +
> +        properties:
> +          endpoint:

[Severity: Medium]
Since the constraints are tied to the exact node name "endpoint", will
these restrictions be silently bypassed if a device tree uses a numbered
endpoint like "endpoint@0"? Using patternProperties for 
"^endpoint(@[0-9a-f]+)?$" might be needed here to ensure the constraints
apply to all valid endpoint node names.

> +            $ref: /schemas/media/video-interfaces.yaml#
> +            unevaluatedProperties: false
> +
> +            properties:
> +              data-lanes:
> +                minItems: 1
> +                maxItems: 4
> +
> +              bus-type:
> +                enum:
> +                  - 1 # MEDIA_BUS_TYPE_CSI2_CPHY
> +                  - 4 # MEDIA_BUS_TYPE_CSI2_DPHY
> +
> +              clock-noncontinuous: true
> +              link-frequencies: true
> +
> +            required:
> +              - data-lanes

[Severity: Medium]
Does this allow device trees for the IT6626 variant to omit the bus-type
property? Without it being listed in the required array, the schema
could validate successfully while leaving the driver without the necessary
hardware information to unambiguously configure the physical layer.

[ ... ]

> +      port@2:
> +        $ref: /schemas/graph.yaml#/$defs/port-base
> +        unevaluatedProperties: false
> +        description: HDMI connector input port
> +
> +        properties:
> +          endpoint:
> +            $ref: /schemas/graph.yaml#/$defs/endpoint-base
> +            unevaluatedProperties: false
> +
> +    required:
> +      - port@0

[Severity: Medium]
Is it intentional that the HDMI input port (port@2) is omitted from this
required list? This appears to allow a device tree to pass schema
validation without specifying the video input to the bridge, resulting
in an incomplete graph that will lack a data source at runtime.

> +required:
> +  - compatible
> +  - reg
> +  - ports
> +
> +allOf:
> +  - if:
> +      properties:
> +        compatible:
> +          contains:
> +            const: ite,it6625
> +    then:
> +      properties:
> +        ports:
> +          properties:
> +            port@0:
> +              properties:
> +                endpoint:

[Severity: Medium]
Will this allOf constraint fail to apply if the device tree uses a numbered
endpoint like "endpoint@0"? Similar to the local constraints above, looking
for the exact "endpoint" string allows numbered nodes to bypass the D-PHY
restriction for the IT6625 variant.

> +                  properties:
> +                    bus-type:
> +                      const: 4

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260722-upstream-it6625-it6626-v3-0-947e50f01764@ite.com.tw?part=1