Re: [PATCH v8 8/8] arm64: dts: qcom: kodiak: Desribe EUD UTMI routing and connector mapping

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Enabling EUD by default causes endless probe deferral on boards where usb_1 is disabled
--

commit f996caeceb74900c278431f664f8ba097d1904a7
Author: Elson Serrao <elson.serrao@oss.qualcomm.com>

arm64: dts: qcom: kodiak: Desribe EUD UTMI routing and connector mapping

This commit updates the EUD description in Kodiak boards to explicitly model
the topology using an EUD path child node and adds a role-switch declaration.
It enables the EUD node by default and maps the USB connector endpoint to
the EUD instead of directly to the USB controller.

> diff --git a/arch/arm64/boot/dts/qcom/kodiak.dtsi b/arch/arm64/boot/dts/qcom/kodiak.dtsi
> index f2da3706d5c88..6c64dc9902e47 100644
> --- a/arch/arm64/boot/dts/qcom/kodiak.dtsi
> +++ b/arch/arm64/boot/dts/qcom/kodiak.dtsi
[ ... ]
> @@ -4753,16 +4747,30 @@ eud: eud@88e0000 {
>  			      <0 0x88e2000 0 0x1000>;
>  			interrupts-extended = <&pdc 11 IRQ_TYPE_LEVEL_HIGH>;
>  
> -			status = "disabled";
> +			#address-cells = <1>;
> +			#size-cells = <0>;

[Severity: Medium]
Does removing status = "disabled" here cause endless probe deferral on boards
where usb_1 is left disabled?

By enabling the eud node by default for all kodiak-based boards, the EUD
driver will attempt to probe and find the role switch for the downstream
usb_1 controller. On boards that do not enable usb_1 (such as
qcs6490-vicharak-axon-mini.dts and qcs6490-radxa-dragon-q6a.dts), the
controller driver never probes and does not register a role switch.

This causes the EUD driver to explicitly return -EPROBE_DEFER. Since the
dependency is disabled, the EUD driver will defer forever and linger in the
deferred probe list.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260912225301.3396556-1-elson.serrao@oss.qualcomm.com?part=8