Re: [PATCH 2/5] dt-bindings: phy: rockchip-inno-csi-dphy: add rk3588 variant

[Michael Riesch <michael.riesch@collabora.com>]

Hi Diederik,

Thanks for your comments!

On 6/17/25 16:12, Diederik de Haas wrote:
> Hi,
> 
> I'm (unfortunately) not seeing any @rock-chips.com recipients ...

Oops, I meant to include at least Kever, but forgot to do it. Will do in v2.

Cc: Kever

> 
> On Tue Jun 17, 2025 at 10:54 AM CEST, Michael Riesch via B4 Relay wrote:
>> From: Michael Riesch <michael.riesch@collabora.com>
>>
>> The Rockchip RK3588 variant of the CSI-2 DPHY features two reset lines.
>> Add the variant and allow for the additional reset.
>>
>> Signed-off-by: Michael Riesch <michael.riesch@collabora.com>
>> ---
>>  .../bindings/phy/rockchip-inno-csi-dphy.yaml       | 60 ++++++++++++++++++++--
>>  1 file changed, 55 insertions(+), 5 deletions(-)
>>
>> diff --git a/Documentation/devicetree/bindings/phy/rockchip-inno-csi-dphy.yaml b/Documentation/devicetree/bindings/phy/rockchip-inno-csi-dphy.yaml
>> index 5ac994b3c0aa..6755738b13ee 100644
>> --- a/Documentation/devicetree/bindings/phy/rockchip-inno-csi-dphy.yaml
>> +++ b/Documentation/devicetree/bindings/phy/rockchip-inno-csi-dphy.yaml
>> @@ -21,6 +21,7 @@ properties:
>>        - rockchip,rk3326-csi-dphy
>>        - rockchip,rk3368-csi-dphy
>>        - rockchip,rk3568-csi-dphy
>> +      - rockchip,rk3588-csi-dphy
>>  
>>    reg:
>>      maxItems: 1
>> @@ -39,18 +40,49 @@ properties:
>>      maxItems: 1
>>  
>>    resets:
>> -    items:
>> -      - description: exclusive PHY reset line
>> +    minItems: 1
>> +    maxItems: 2
>>  
>>    reset-names:
>> -    items:
>> -      - const: apb
>> +    minItems: 1
>> +    maxItems: 2
>>  
>>    rockchip,grf:
>>      $ref: /schemas/types.yaml#/definitions/phandle
>>      description:
>>        Some additional phy settings are access through GRF regs.
>>  
>> +allOf:
>> +  - if:
>> +      properties:
>> +        compatible:
>> +          contains:
>> +            enum:
>> +              - rockchip,px30-csi-dphy
>> +              - rockchip,rk1808-csi-dphy
>> +              - rockchip,rk3326-csi-dphy
>> +              - rockchip,rk3368-csi-dphy
>> +              - rockchip,rk3568-csi-dphy
>> +    then:
>> +      properties:
>> +        resets:
>> +          items:
>> +            - description: exclusive PHY reset line
>> +
>> +        reset-names:
>> +          items:
>> +            - const: apb
>> +
>> +      required:
>> +        - reset-names
>> +    else:
>> +      properties:
>> +        resets:
>> +          minItems: 2
>> +
>> +        reset-names:
>> +          minItems: 2
>> +
>>  required:
>>    - compatible
>>    - reg
>> @@ -59,7 +91,6 @@ required:
>>    - '#phy-cells'
>>    - power-domains
>>    - resets
>> -  - reset-names
>>    - rockchip,grf
>>  
>>  additionalProperties: false
>> @@ -78,3 +109,22 @@ examples:
>>          reset-names = "apb";
>>          rockchip,grf = <&grf>;
>>      };
>> +  - |
>> +    #include <dt-bindings/clock/rockchip,rk3588-cru.h>
>> +    #include <dt-bindings/reset/rockchip,rk3588-cru.h>
>> +
>> +    soc {
>> +        #address-cells = <2>;
>> +        #size-cells = <2>;
>> +
>> +        csi_dphy0: phy@fedc0000 {
>> +            compatible = "rockchip,rk3588-csi-dphy";
>> +            reg = <0x0 0xfedc0000 0x0 0x8000>;
>> +            clocks = <&cru PCLK_CSIPHY0>;
>> +            clock-names = "pclk";
>> +            #phy-cells = <0>;
>> +            resets = <&cru SRST_CSIPHY0>, <&cru SRST_P_CSIPHY0>;
>> +            rockchip,grf = <&csidphy0_grf>;
>> +            status = "disabled";
>> +        };
>> +    };
> 
> ... which could hopefully tell us what the value is/should be for the
> *required* 'power-domains' property, which is missing in this example.
> IOW: the binding example is invalid according to its own binding.

Huh, indeed. Hm. Why didn't make dt_binding_check warn me about that?!

TRM Part 1, p. 1097 states that HDMI_CSI_DPHY is in the ALIVE(PD_BUS)
power domain. With some creativity one can interpret that the CSI DPHY
is always on anyways. @Kever: Could you please elaborate on that?

> (btw: you can drop the 'csi_dphy0' label)

Will do.

> 
> And hopefully also for rk3568 so we can add it to rk356x-base.dtsi and
> you can add it in patch 5 where it's also missing.

I recall a similar discussion [0]. In the RK3568 the CSIPHY is in the
ALIVE power domain.

Note that the PHY must not be confused with the HOST blocks, which are
the MIPI CSI-2 receivers.

I guess the correct solution is to make power-domains optional. Further
input welcome, though.

Best regards,
Michael

> 
> Grepping for "csi-dphy" in arch/arm*/boot/dts/rockchip returns:
> - px30.dtsi
> - rk356x-base.dtsi
> 
> With this patch set applied, we'd have a 3rd result: rk3588-base.dtsi
> 
> For all the listed compatibles, it's only actually defined in px30.dtsi.
> 
> Cheers (and sorry),
>   Diederik

[0] https://lore.kernel.org/all/D4QNJ85V43NU.YD01E8AB4116@cknow.org/