On Tue, Sep 02, 2025 at 08:07:22AM +0200, Krzysztof Kozlowski wrote:
> On 02/09/2025 06:04, Dmitry Baryshkov wrote:
> >>>
> >>> diff --git a/arch/arm64/boot/dts/qcom/sm6350.dtsi
> >>> b/arch/arm64/boot/dts/qcom/sm6350.dtsi
> >>> index
> >>> 2493b9611dcb675f4c33794ecc0ee9e8823e24d4..8459b27cacc72a4827a2e289e669163ad6250059
> >>> 100644
> >>> --- a/arch/arm64/boot/dts/qcom/sm6350.dtsi
> >>> +++ b/arch/arm64/boot/dts/qcom/sm6350.dtsi
> >>> @@ -2249,7 +2249,7 @@ opp-560000000 {
> >>> };
> >>>
> >>> mdss_dp: displayport-controller@ae90000 {
> >>> - compatible = "qcom,sm6350-dp", "qcom,sm8350-dp";
> >>> + compatible = "qcom,sm6350-dp", "qcom,sc7180-dp";
> >>
> >> No, that's breaking all the users.
> >
> > WHy though? Both old and new lines are using fallbacks to bind the
> > driver to the device.
>
> Kernel has sc7180 fallback, but what if other DTS user does not and that
> other user was relying on sm8350 fallback compatible? That other user
> won't have sm6350 dedicated handling as well.
Oh, a user which has SM8350 support, wants to support SM6350, but
doesn't support SC7180 DP? How hypothetical should be our users?
>
> That breaking of users I meant.
>
> With the kernel it should work, assuming SC7180-dp was introduced
> similar time as 8350-dp.
SC7180 DP was introduced several years ahead of SM8350, if my memory
doesn't deceive me.
>
> Best regards,
> Krzysztof
--
With best wishes
Dmitry
