Re: [PATCH 09/10] [RFT] arm64: dts: qcom: sm8250: extend CAMSS with new CSIPHY subdevices

[Krzysztof Kozlowski <krzk@kernel.org>]

On 12/06/2025 03:15, Vladimir Zapolskiy wrote:
> Following the new device tree bindings for CAMSS IPs introduce csiphy2
> device tree node under SM8250 CAMSS, which allows to perform camera
> tests of the model on an RB5 board with an attached vision mezzanine.

How the binding allows to perform camera tests? So camera was not
working here at all? Then this is a fix, no?


> 
> Note that the optional 'phys' property is deliberately not added.

Why? Your commit msg must explain that.

> 
> Signed-off-by: Vladimir Zapolskiy <vladimir.zapolskiy@linaro.org>
> ---
> For testing only, do not merge.
> 
>  arch/arm64/boot/dts/qcom/sm8250.dtsi | 14 ++++++++++++++
>  1 file changed, 14 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/sm8250.dtsi b/arch/arm64/boot/dts/qcom/sm8250.dtsi
> index f0d18fd37aaf..401a32679580 100644
> --- a/arch/arm64/boot/dts/qcom/sm8250.dtsi
> +++ b/arch/arm64/boot/dts/qcom/sm8250.dtsi
> @@ -4613,6 +4613,10 @@ camss: camss@ac6a000 {
>  					     "cam_sf_0_mnoc",
>  					     "cam_sf_icp_mnoc";
>  
> +			#address-cells = <2>;
> +			#size-cells = <2>;
> +			ranges;
> +
>  			ports {
>  				#address-cells = <1>;
>  				#size-cells = <0>;
> @@ -4641,6 +4645,16 @@ port@5 {
>  					reg = <5>;
>  				};
>  			};
> +
> +			csiphy2: phy@ac6e000 {

This will fail checks. You can run them, regardless of "RFT" status.

> +				compatible = "qcom,csiphy";
> +				reg = <0 0x0ac6e000 0 0x1000>;
> +				clocks = <&camcc CAM_CC_CSIPHY2_CLK>,
> +					 <&camcc CAM_CC_CSI2PHYTIMER_CLK>;
> +				clock-names = "csiphy", "csiphy_timer";
> +				interrupts = <GIC_SPI 479 IRQ_TYPE_EDGE_RISING>;
> +				#phy-cells = <0>;

This is also duplicating existing ports thus you have a mixed MMIO and
non-MMIO children which is also issue to fix.

> +			};
>  		};
>  
>  		camcc: clock-controller@ad00000 {


Best regards,
Krzysztof