On 02/09/2022 20:51, Andrew Halaney wrote: > The top level RPMh nodes have a supply property, make sure to specify it > so the patternProperties later that are keyed off of the PMIC version > are properly honored. Without this, and the dt-binding containing > additionalProperties: false, you will see the following when running > make dt_binding_check: > > DTEX Documentation/devicetree/bindings/regulator/qcom,rpmh-regulator.example.dts > DTC Documentation/devicetree/bindings/regulator/qcom,rpmh-regulator.example.dtb > CHECK Documentation/devicetree/bindings/regulator/qcom,rpmh-regulator.example.dtb > /mnt/extrassd/git/linux-next/Documentation/devicetree/bindings/regulator/qcom,rpmh-regulator.example.dtb: pm8998-rpmh-regulators: 'vdd-l7-l12-l14-l15-supply' does not match any of the regexes: '^(smps|ldo|lvs)[0-9]+$', 'pinctrl-[0-9]+' > From schema: /mnt/extrassd/git/linux-next/Documentation/devicetree/bindings/regulator/qcom,rpmh-regulator.yaml > > That supply pattern is intended to be considered correct for the > qcom,pm8998-rpmh-regulators compatible, and is no longer complained > about with the supply property described. Which supply pattern? > > Unfortunately this pattern is wide enough that it no longer complains > when you bork the expected supply for a compatible. I.e. for > qcom,pm8998-rpmh-regulators, if I change the example usage in the > binding to: > > vdd-l0-l12-l14-l15-supply = <&pm8998_s5>; > > I get no warning, when really it should be of the pattern: > > vdd-l7-l12-l14-l15-supply = <&pm8998_s5>; > > Signed-off-by: Andrew Halaney <ahalaney@xxxxxxxxxx> No, you basically reverse the change I did, without actually addressing my intentions in that commit. If you want to revert it, please make a proper revert and explain why such revert is needed. Best regards, Krzysztof
