Re: [PATCH 1/2] arm64: dts: qcom: x1e80100-crd: Enable external DP support
From: Dmitry Baryshkov
Date: Tue Sep 24 2024 - 08:03:35 EST
On Tue, 24 Sept 2024 at 13:08, Abel Vesa <abel.vesa@xxxxxxxxxx> wrote:
>
> On 24-09-02 21:54:58, Dmitry Baryshkov wrote:
> > On Mon, Sep 02, 2024 at 06:01:35PM GMT, Abel Vesa wrote:
> > > The Qualcomm Snapdragon X Elite CRD board has 3 USB Type-C ports,
> > > all of them supporting external DP altmode. Between each QMP
> > > combo PHY and the corresponding Type-C port, sits one Parade PS8830
> > > retimer which handles both orientation and SBU muxing. Add nodes for
> > > each retimer, fix the graphs between connectors and the PHYs accordingly,
> > > add the voltage regulators needed by each retimer and then enable all
> > > 3 remaining DPUs.
> > >
> > > Signed-off-by: Abel Vesa <abel.vesa@xxxxxxxxxx>
> > > ---
> > > arch/arm64/boot/dts/qcom/x1e80100-crd.dts | 414 +++++++++++++++++++++++++++++-
> > > 1 file changed, 408 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/arch/arm64/boot/dts/qcom/x1e80100-crd.dts b/arch/arm64/boot/dts/qcom/x1e80100-crd.dts
> > > index 10b28d870f08..6dfc85eda354 100644
> > > --- a/arch/arm64/boot/dts/qcom/x1e80100-crd.dts
> > > +++ b/arch/arm64/boot/dts/qcom/x1e80100-crd.dts
> >
> > [skipped]
> >
> >
> > > @@ -709,6 +861,163 @@ keyboard@3a {
> > > };
> > > };
> > >
> > > +&i2c1 {
> > > + clock-frequency = <400000>;
> > > +
> > > + status = "okay";
> > > +
> > > + typec-mux@8 {
> > > + compatible = "parade,ps8830";
> > > + reg = <0x08>;
> > > +
> > > + clocks = <&rpmhcc RPMH_RF_CLK5>;
> > > + clock-names = "xo";
> > > +
> > > + vdd15-supply = <&vreg_rtmr2_1p15>;
> > > + vdd18-supply = <&vreg_rtmr2_1p8>;
> > > + vdd33-supply = <&vreg_rtmr2_3p3>;
> > > +
> > > + reset-gpios = <&tlmm 185 GPIO_ACTIVE_HIGH>;
> > > +
> > > + orientation-switch;
> > > + retimer-switch;
> >
> > Doesn't it need to listen to mode switching events? 4-lane DP vs
> > 2/2-lane DP + USB3 requires propagating of the altmode events to the QMP
> > PHY, see the original 4-lane series.
>
> The mode is passed through the retimer state as well, isn't it?
ack.
>
> >
> > > +
> > > + ports {
> > > + #address-cells = <1>;
> > > + #size-cells = <0>;
> > > +
> > > + port@0 {
> > > + reg = <0>;
> > > +
> > > + retimer_ss2_ss_out: endpoint {
> > > + remote-endpoint = <&pmic_glink_ss2_ss_in>;
> > > + };
> > > + };
> > > +
> > > + port@1 {
> > > + reg = <1>;
> > > +
> > > + retimer_ss2_ss_in: endpoint {
> > > + remote-endpoint = <&usb_1_ss2_qmpphy_out>;
> > > + };
> > > + };
> > > +
> > > + port@2 {
> > > + reg = <2>;
> > > +
> > > + retimer_ss2_con_sbu_out: endpoint {
> > > + remote-endpoint = <&pmic_glink_ss2_con_sbu_in>;
> > > + };
> > > + };
> > > + };
> > > + };
> > > +};
> > > +
> >
> >
> > > @@ -1164,7 +1566,7 @@ &usb_1_ss0_dwc3_hs {
> > > };
> > >
> > > &usb_1_ss0_qmpphy_out {
> > > - remote-endpoint = <&pmic_glink_ss0_ss_in>;
> > > + remote-endpoint = <&retimer_ss0_ss_in>;
> > > };
> >
> > orientation-switch and mode-switch for the QMP PHY?
>
> orientation-switch is already in the SoC dtsi nodes.
>
> But the qmp phy doesn't currently support mode switching. Right?
Yeah, we didn't land that patchset :-(
>
> >
> > >
> > > &usb_1_ss1_hsphy {
> > > @@ -1196,7 +1598,7 @@ &usb_1_ss1_dwc3_hs {
> > > };
> > >
> > > &usb_1_ss1_qmpphy_out {
> > > - remote-endpoint = <&pmic_glink_ss1_ss_in>;
> > > + remote-endpoint = <&retimer_ss1_ss_in>;
> > > };
> > >
> > > &usb_1_ss2_hsphy {
> > > @@ -1228,5 +1630,5 @@ &usb_1_ss2_dwc3_hs {
> > > };
> > >
> > > &usb_1_ss2_qmpphy_out {
> > > - remote-endpoint = <&pmic_glink_ss2_ss_in>;
> > > + remote-endpoint = <&retimer_ss2_ss_in>;
> > > };
> > >
> > > --
> > > 2.34.1
> > >
> >
> > --
> > With best wishes
> > Dmitry
>
> Thanks for reviewing.
>
> Abel
--
With best wishes
Dmitry