Re: [PATCH v3 1/4] dt-bindings: phy: qcom,qmp: Mark '#clock-cells' as a 'optional' property

[Bhupesh Sharma]

Hi Bjorn,

On Sat, 23 Apr 2022 at 21:14, Bjorn Andersson
<bjorn.andersson@xxxxxxxxxx> wrote:
>
> On Mon 18 Apr 13:55 PDT 2022, Bhupesh Sharma wrote:
>
> > '#clock-cells' is not a required property for qmp-phy(s) in the
> > '/' node, but it should be is used in 'phy@' subnode (where it is
> > actually a 'required' property). Fix the same.
> >
>
> It's not that #clock-cells is "not a required property", it's that the
> clock comes out of the phy (the child node), so there is no clocks
> provided by the parent device.
>
>
> Please rewrite the commit message.

Ok.

> > This also fixes the following 'make dtbs_check' warning(s):
> >
> > sm8350-microsoft-surface-duo2.dt.yaml: phy@1d87000:
> >   '#clock-cells' is a required property
> >
> > Cc: Bjorn Andersson <bjorn.andersson@xxxxxxxxxx>
> > Cc: Vinod Koul <vkoul@xxxxxxxxxx>
> > Cc: Rob Herring <robh+dt@xxxxxxxxxx>
> > Signed-off-by: Bhupesh Sharma <bhupesh.sharma@xxxxxxxxxx>
> > ---
> >  Documentation/devicetree/bindings/phy/qcom,qmp-phy.yaml | 8 +++-----
> >  1 file changed, 3 insertions(+), 5 deletions(-)
> >
> > diff --git a/Documentation/devicetree/bindings/phy/qcom,qmp-phy.yaml b/Documentation/devicetree/bindings/phy/qcom,qmp-phy.yaml
> > index 8b850c5ab116..c39ead81ecd7 100644
> > --- a/Documentation/devicetree/bindings/phy/qcom,qmp-phy.yaml
> > +++ b/Documentation/devicetree/bindings/phy/qcom,qmp-phy.yaml
> > @@ -66,9 +66,6 @@ properties:
> >        - description: Address and length of PHY's common serdes block.
> >        - description: Address and length of PHY's DP_COM control block.
> >
> > -  "#clock-cells":
> > -    enum: [ 1, 2 ]
> > -
> >    "#address-cells":
> >      enum: [ 1, 2 ]
> >
> > @@ -112,11 +109,13 @@ patternProperties:
> >      description:
> >        Each device node of QMP phy is required to have as many child nodes as
> >        the number of lanes the PHY has.
> > +    properties:
> > +      "#clock-cells":
> > +        enum: [ 0, 1, 2 ]
>
> The commit message doesn't mention the fact that 0 is also a valid
> value. Perhaps just keep it [1, 2] in this patch?

0 is a valid value as mentioned in the example inside the dt-binding
example itself.
For e.g. see the 'sdm845-qmp-pcie-phy' node:

    pcie0_phy: phy@1c06000 {
            compatible = "qcom,sdm845-qmp-pcie-phy";
            <..snip..>

            pcie0_lane: phy@1c06200 {
                    <..snip..>

                #clock-cells = <0>;

So, without [ 0, 1, 2 ] in the yaml bindings we get the following
error while running '$ make dt_binding_check' :
qcom,qmp-phy.example.dtb: phy-wrapper@88eb000:
phy@200:#clock-cells:0:0: 0 is not one of [1, 2]

Thanks,
Bhupesh

> >
> >  required:
> >    - compatible
> >    - reg
> > -  - "#clock-cells"
> >    - "#address-cells"
> >    - "#size-cells"
> >    - ranges
> > @@ -468,7 +467,6 @@ examples:
> >      usb_2_qmpphy: phy-wrapper@88eb000 {
> >          compatible = "qcom,sdm845-qmp-usb3-uni-phy";
> >          reg = <0x088eb000 0x18c>;
> > -        #clock-cells = <1>;
> >          #address-cells = <1>;
> >          #size-cells = <1>;
> >          ranges = <0x0 0x088eb000 0x2000>;
> > --
> > 2.35.1
> >