Re: [RFC PATCH 6/6] arm64: dts: qcom: sc7180: Add SoC specific compatible to soc node

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

On 10.01.2025 12:45 AM, Dmitry Baryshkov wrote:
> On Thu, Jan 09, 2025 at 03:10:33PM +0100, Konrad Dybcio wrote:
>> On 8.01.2025 2:02 PM, Dmitry Baryshkov wrote:
>>> On Tue, Jan 07, 2025 at 05:28:43PM -0800, Stephen Boyd wrote:
>>>> Allow an SoC driver to probe for these devices. Add the SoC specific
>>>> compatible to the soc node. Leave the original simple-bus compatible in
>>>> place so that everything keeps working.
>>>>
>>>> Cc: Rob Herring <robh@kernel.org>
>>>> Cc: Bjorn Andersson <andersson@kernel.org>
>>>> Cc: Konrad Dybcio <konradybcio@kernel.org>
>>>> Cc: <linux-arm-msm@vger.kernel.org>
>>>> Signed-off-by: Stephen Boyd <swboyd@chromium.org>
>>>> ---
>>>>  arch/arm64/boot/dts/qcom/sc7180.dtsi | 2 +-
>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/arch/arm64/boot/dts/qcom/sc7180.dtsi b/arch/arm64/boot/dts/qcom/sc7180.dtsi
>>>> index 76fe314d2ad5..257890a193e6 100644
>>>> --- a/arch/arm64/boot/dts/qcom/sc7180.dtsi
>>>> +++ b/arch/arm64/boot/dts/qcom/sc7180.dtsi
>>>> @@ -782,7 +782,7 @@ soc: soc@0 {
>>>>  		#size-cells = <2>;
>>>>  		ranges = <0 0 0 0 0x10 0>;
>>>>  		dma-ranges = <0 0 0 0 0x10 0>;
>>>> -		compatible = "simple-bus";
>>>> +		compatible = "qcom,soc-sc7180", "simple-bus";
>>>
>>> If the new driver requires this compatible, it will break compatibility
>>> with older DT files (and it should be avoided).
>>
>> IIUC the intent here is to provide backwards compatibility through checking
>> for sth like IS_SOCPM_MANAGED(), sorta like HAS_ACPI_COMPANION(). In that
>> case, power sequencing would be done by the socpm driver, whereas if it
>> doesn't hold, the resources would be toggled by the device driver
> 
> I think that this way we end up having PM code both in the device driver
> and in the socpm. Ideally in my opinion we should be able to migrate all
> pm code to socpm, keeping compat with old DT files. In the end, if this
> is the only change to the SoC.dtsi, then we should be able to live
> without this compat change.

We should be able to do that with a dynamic overlay, I suppose.. which we
could drop after some time (probably a rather large amount of time)

Konrad