Re: [PATCH 1/5] arm64: dts: lx2160a: transition to device-specific SerDes compatible strings

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider:
- [High] Unintentional disabling of `serdes_1` on multiple boards.
- [Medium] Consumer phandles for SerDes PHYs were not updated to match the new per-lane PHY providers, contradicting the commit message.
- [Medium] The newly added `serdes_3` node is missing the generic fallback compatible string `fsl,lynx-28g`.
- [Low] Device tree compilation fails due to incorrect node name in `/delete-node/` directive for serdes_3.
--

commit dcf5b2d58e1afb24aa2ae2829db182c735a1dfc1
Author: Vladimir Oltean <vladimir.oltean@nxp.com>

arm64: dts: lx2160a: transition to device-specific SerDes compatible strings

Align to the modern fsl,lynx-28g.yaml binding, where the SoC and SerDes
instance is present in the compatible string, to allow reliable per-lane
capability detection and per-lane customization of electrical properties.

>     These new bindings have #phy-cells = <0> in per-lane PHY providers, so
>     we need to update consumer phandles as well.

[Severity: Medium]
Did we miss updating the consumer phandles in this patch?

The commit message indicates consumer phandles need updating to match
#phy-cells = <0>, but they appear to remain unupdated in the device tree,
and #phy-cells = <1> is still retained on the parent PHY nodes.

> diff --git a/arch/arm64/boot/dts/freescale/fsl-lx2160a.dtsi b/arch/arm64/boot/dts/freescale/fsl-lx2160a.dtsi
> index 1d73abffa6b72..a687eb3e31904 100644
> --- a/arch/arm64/boot/dts/freescale/fsl-lx2160a.dtsi
> +++ b/arch/arm64/boot/dts/freescale/fsl-lx2160a.dtsi
> @@ -621,17 +621,163 @@ soc: soc {
>  		ranges;
>  		dma-ranges = <0x0 0x0 0x0 0x0 0x10000 0x00000000>;
>  
[ ... ]
>  		serdes_1: phy@1ea0000 {
> -			compatible = "fsl,lynx-28g";
> +			compatible = "fsl,lx2160a-serdes1", "fsl,lynx-28g";
>  			reg = <0x0 0x1ea0000 0x0 0x1e30>;
> +			#address-cells = <1>;
> +			#size-cells = <0>;
>  			#phy-cells = <1>;
> +			status = "disabled";

[Severity: High]
Does this default status = "disabled" unintentionally break interfaces
wired to serdes_1 on boards like fsl-lx2162a-clearfog and fsl-lx2160a-half-twins?

These boards consume serdes_1 but do not appear to override the status to
"okay" in this patch.

[ ... ]
> +		serdes_3: phy@1ec0000 {
> +			compatible = "fsl,lx2160a-serdes3";
> +			reg = <0x0 0x1ec0000 0x0 0x1e30>;

[Severity: Medium]
Should serdes_3 include the generic "fsl,lynx-28g" fallback compatible string
here, similar to what was added for serdes_1 and serdes_2?

> +			#address-cells = <1>;
> +			#size-cells = <0>;
> +			status = "disabled";
> +			#phy-cells = <1>;
[ ... ]
> diff --git a/arch/arm64/boot/dts/freescale/fsl-lx2162a.dtsi b/arch/arm64/boot/dts/freescale/fsl-lx2162a.dtsi
> new file mode 100644
> index 0000000000000..b9629e074d943
> --- /dev/null
> +++ b/arch/arm64/boot/dts/freescale/fsl-lx2162a.dtsi
> @@ -0,0 +1,24 @@
[ ... ]
> +&soc {
> +	/delete-node/ serdes@1ec0000;
> +};

[Severity: Low]
Could this /delete-node/ directive fail at build time?

In the base fsl-lx2160a.dtsi, the node is defined as phy@1ec0000 rather
than serdes@1ec0000.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260630110459.516364-1-ioana.ciornei@nxp.com?part=1