On 20/06/2022 13:27, Dmitry Baryshkov wrote:
> On Mon, 20 Jun 2022 at 14:24, Krzysztof Kozlowski
> <krzysztof.kozlowski@xxxxxxxxxx> wrote:
>>
>> On 20/06/2022 09:19, Dmitry Baryshkov wrote:
>>> Add missing GCC clock sources. This includes PCIe and USB PIPE and UFS
>>> symbol clocks.
>>>
>>> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
>>> ---
>>> arch/arm64/boot/dts/qcom/msm8996.dtsi | 18 ++++++++++++++++--
>>> 1 file changed, 16 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/arch/arm64/boot/dts/qcom/msm8996.dtsi b/arch/arm64/boot/dts/qcom/msm8996.dtsi
>>> index e97f193aefd3..6c7380f86383 100644
>>> --- a/arch/arm64/boot/dts/qcom/msm8996.dtsi
>>> +++ b/arch/arm64/boot/dts/qcom/msm8996.dtsi
>>> @@ -689,8 +689,22 @@ gcc: clock-controller@300000 {
>>>
>>> clocks = <&rpmcc RPM_SMD_BB_CLK1>,
>>> <&rpmcc RPM_SMD_LN_BB_CLK>,
>>> - <&sleep_clk>;
>>> - clock-names = "cxo", "cxo2", "sleep_clk";
>>> + <&sleep_clk>,
>>> + <&pciephy_0>,
>>> + <&pciephy_1>,
>>> + <&pciephy_2>,
>>> + <&ssusb_phy_0>,
>>> + <0>, <0>, <0>;
>>
>> Since the clocks are optional, there is no need to pass <0> to them. I
>> think it does not bring any benefits.
>
> It serves as a reminder that they should be filled with the proper
> sources. We do the same thing for DSI/DP clocks where the PHY is not
> yet supported, but the GCC lists them.
>
> I had to mark them as optional so that the existing schema files pass
> validation. Otherwise they would be in the 'required' part.
>
Hm, ok. Let's hope someone will actually remember to fill these in once
there is proper clock.
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>
Best regards,
Krzysztof
