Re: [PATCH 1/2] dt-bindings: soc: qcom: Document CDSP Power Management

[Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>]

On 5/19/26 9:05 PM, Vignesh Viswanathan wrote:
> Add documentation for the CDSP Power Management driver, which handles
> Dynamic Clock and Voltage Scaling (DCVS) requests via SMEM, manages Low
> Power Mode (LPM) transitions via MPM handshake, and provides virtual
> regulators for the remoteproc driver to control CDSP power rails.
> 
> Signed-off-by: Vignesh Viswanathan <vignesh.viswanathan@oss.qualcomm.com>
> ---

[...]

> +maintainers:
> +  - Vignesh Viswanathan <vignesh.viswanathan@oss.qualcomm.com>
> +
> +description:
> +  The CDSP Power Management driver provides power management services for the
> +  Qualcomm Compute DSP (CDSP) subsystem. It handles Dynamic Clock and Voltage
> +  Scaling (DCVS) requests via SMEM, manages Low Power Mode (LPM) transitions
> +  via MPM handshake, and provides virtual regulators that are consumed by the
> +  CDSP remoteproc driver.
> +
> +properties:
> +  compatible:
> +    const: qcom,cdsp-power

This really needs to be SoC-specific

> +
> +  reg:
> +    items:
> +      - description: MPM (Modem Power Manager) register region
> +      - description: RSCC (RSC Configuration) register region

I was under the impression that these two belong to separate
universes..

> +  vdd-cx-supply:
> +    description:
> +      Phandle to the CX voltage regulator. This is the actual hardware regulator
> +      (e.g., from MP8899 PMIC) that supplies power to the CDSP CX rail.

I believe the second sentence is wholly redundant

[...]

> +    #include <dt-bindings/interrupt-controller/arm-gic.h>
> +    #include <dt-bindings/interrupt-controller/irq.h>
> +    #include <dt-bindings/mailbox/qcom-ipcc.h>
> +
> +    soc {
> +        #address-cells = <2>;
> +        #size-cells = <2>;
> +
> +        cdsp_power: cdsp-power@4ae000 {
> +            compatible = "qcom,cdsp-power";
> +            reg = <0x0 0x004ae000 0x0 0x1000>,

MPM is a top-level peripheral, but this is a slice dedicated to the
CDSP so maybe it'd pass, but..

> +                  <0x0 0x26018018 0x0 0x4>;

This is a single random register within the CDSP's register space, so
we should definitely be able to describe this better..

Konrad