| Message ID | 20230324215550.1966809-4-dmitry.baryshkov@linaro.org |
|---|---|
| State | Superseded |
| Headers | show |
| Series | phy: qcom-qmp-usb: split away legacy USB+DP code | expand |
On 24/03/2023 22:55, Dmitry Baryshkov wrote: > The first USB PHY on the sm8150 platform is really the USB+DP combo > PHY. Add the DP part of the PHY. > > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> > --- > arch/arm64/boot/dts/qcom/sm8150.dtsi | 17 +++++++++++++---- > 1 file changed, 13 insertions(+), 4 deletions(-) > > diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi > index 9491be4a6bf0..a618218f7b68 100644 > --- a/arch/arm64/boot/dts/qcom/sm8150.dtsi > +++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi > @@ -3392,20 +3392,19 @@ usb_2_hsphy: phy@88e3000 { > }; > > usb_1_qmpphy: phy@88e9000 { > - compatible = "qcom,sm8150-qmp-usb3-phy"; > + compatible = "qcom,sm8150-qmp-usb3-dp-phy"; > reg = <0 0x088e9000 0 0x18c>, > - <0 0x088e8000 0 0x10>; > + <0 0x088e8000 0 0x38>, > + <0 0x088ea000 0 0x40>; > status = "disabled"; > #address-cells = <2>; > #size-cells = <2>; > ranges; > - > clocks = <&gcc GCC_USB3_PRIM_PHY_AUX_CLK>, > <&rpmhcc RPMH_CXO_CLK>, > <&gcc GCC_USB3_PRIM_CLKREF_CLK>, > <&gcc GCC_USB3_PRIM_PHY_COM_AUX_CLK>; > clock-names = "aux", "ref_clk_src", "ref", "com_aux"; > - > resets = <&gcc GCC_USB3_DP_PHY_PRIM_BCR>, > <&gcc GCC_USB3_PHY_PRIM_BCR>; > reset-names = "phy", "common"; > @@ -3423,6 +3422,16 @@ usb_1_ssphy: phy@88e9200 { > clock-names = "pipe0"; > clock-output-names = "usb3_phy_pipe_clk_src"; > }; > + > + usb_1_dpphy: phy@88ea200 { > + reg = <0 0x088ea200 0 0x200>, > + <0 0x088ea400 0 0x200>, > + <0 0x088eaa00 0 0x200>, > + <0 0x088ea600 0 0x200>, > + <0 0x088ea800 0 0x200>; > + #clock-cells = <1>; > + #phy-cells = <0>; > + }; Is there a reason why the new flat bindings from qcom,sc8280xp-qmp-usb43dp-phy.yaml are not used instead ? Neil > }; > > usb_2_qmpphy: phy@88eb000 {
On 27/03/2023 09:59, Neil Armstrong wrote: > On 24/03/2023 22:55, Dmitry Baryshkov wrote: >> The first USB PHY on the sm8150 platform is really the USB+DP combo >> PHY. Add the DP part of the PHY. >> >> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> >> --- >> arch/arm64/boot/dts/qcom/sm8150.dtsi | 17 +++++++++++++---- >> 1 file changed, 13 insertions(+), 4 deletions(-) >> >> diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi >> index 9491be4a6bf0..a618218f7b68 100644 >> --- a/arch/arm64/boot/dts/qcom/sm8150.dtsi >> +++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi >> @@ -3392,20 +3392,19 @@ usb_2_hsphy: phy@88e3000 { >> }; >> usb_1_qmpphy: phy@88e9000 { >> - compatible = "qcom,sm8150-qmp-usb3-phy"; >> + compatible = "qcom,sm8150-qmp-usb3-dp-phy"; >> reg = <0 0x088e9000 0 0x18c>, >> - <0 0x088e8000 0 0x10>; >> + <0 0x088e8000 0 0x38>, >> + <0 0x088ea000 0 0x40>; >> status = "disabled"; >> #address-cells = <2>; >> #size-cells = <2>; >> ranges; >> - >> clocks = <&gcc GCC_USB3_PRIM_PHY_AUX_CLK>, >> <&rpmhcc RPMH_CXO_CLK>, >> <&gcc GCC_USB3_PRIM_CLKREF_CLK>, >> <&gcc GCC_USB3_PRIM_PHY_COM_AUX_CLK>; >> clock-names = "aux", "ref_clk_src", "ref", "com_aux"; >> - >> resets = <&gcc GCC_USB3_DP_PHY_PRIM_BCR>, >> <&gcc GCC_USB3_PHY_PRIM_BCR>; >> reset-names = "phy", "common"; >> @@ -3423,6 +3422,16 @@ usb_1_ssphy: phy@88e9200 { >> clock-names = "pipe0"; >> clock-output-names = "usb3_phy_pipe_clk_src"; >> }; >> + >> + usb_1_dpphy: phy@88ea200 { >> + reg = <0 0x088ea200 0 0x200>, >> + <0 0x088ea400 0 0x200>, >> + <0 0x088eaa00 0 0x200>, >> + <0 0x088ea600 0 0x200>, >> + <0 0x088ea800 0 0x200>; >> + #clock-cells = <1>; >> + #phy-cells = <0>; >> + }; > > Is there a reason why the new flat bindings from qcom,sc8280xp-qmp-usb43dp-phy.yaml are not used instead ? > Oh ok I see "phy: qcom-qmp-combo: convert to newer style of bindings" is the followup of this serie, please specify it because it wasn't obvious... Neil > Neil > >> }; >> usb_2_qmpphy: phy@88eb000 { >
On 27/03/2023 11:05, Neil Armstrong wrote: > On 27/03/2023 09:59, Neil Armstrong wrote: >> On 24/03/2023 22:55, Dmitry Baryshkov wrote: >>> The first USB PHY on the sm8150 platform is really the USB+DP combo >>> PHY. Add the DP part of the PHY. >>> >>> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> >>> --- >>> arch/arm64/boot/dts/qcom/sm8150.dtsi | 17 +++++++++++++---- >>> 1 file changed, 13 insertions(+), 4 deletions(-) >>> >>> diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi >>> b/arch/arm64/boot/dts/qcom/sm8150.dtsi >>> index 9491be4a6bf0..a618218f7b68 100644 >>> --- a/arch/arm64/boot/dts/qcom/sm8150.dtsi >>> +++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi >>> @@ -3392,20 +3392,19 @@ usb_2_hsphy: phy@88e3000 { >>> }; >>> usb_1_qmpphy: phy@88e9000 { >>> - compatible = "qcom,sm8150-qmp-usb3-phy"; >>> + compatible = "qcom,sm8150-qmp-usb3-dp-phy"; >>> reg = <0 0x088e9000 0 0x18c>, >>> - <0 0x088e8000 0 0x10>; >>> + <0 0x088e8000 0 0x38>, >>> + <0 0x088ea000 0 0x40>; >>> status = "disabled"; >>> #address-cells = <2>; >>> #size-cells = <2>; >>> ranges; >>> - >>> clocks = <&gcc GCC_USB3_PRIM_PHY_AUX_CLK>, >>> <&rpmhcc RPMH_CXO_CLK>, >>> <&gcc GCC_USB3_PRIM_CLKREF_CLK>, >>> <&gcc GCC_USB3_PRIM_PHY_COM_AUX_CLK>; >>> clock-names = "aux", "ref_clk_src", "ref", "com_aux"; >>> - >>> resets = <&gcc GCC_USB3_DP_PHY_PRIM_BCR>, >>> <&gcc GCC_USB3_PHY_PRIM_BCR>; >>> reset-names = "phy", "common"; >>> @@ -3423,6 +3422,16 @@ usb_1_ssphy: phy@88e9200 { >>> clock-names = "pipe0"; >>> clock-output-names = "usb3_phy_pipe_clk_src"; >>> }; >>> + >>> + usb_1_dpphy: phy@88ea200 { >>> + reg = <0 0x088ea200 0 0x200>, >>> + <0 0x088ea400 0 0x200>, >>> + <0 0x088eaa00 0 0x200>, >>> + <0 0x088ea600 0 0x200>, >>> + <0 0x088ea800 0 0x200>; >>> + #clock-cells = <1>; >>> + #phy-cells = <0>; >>> + }; >> >> Is there a reason why the new flat bindings from >> qcom,sc8280xp-qmp-usb43dp-phy.yaml are not used instead ? >> > > Oh ok I see "phy: qcom-qmp-combo: convert to newer style of bindings" is > the followup of this serie, > please specify it because it wasn't obvious... I thought that a note in the cover letter was good enough, but yeah, maybe it should be more explicit. Do you think it warrants v2? I can send one.
On 27.03.2023 12:14, Dmitry Baryshkov wrote: > On 27/03/2023 11:05, Neil Armstrong wrote: >> On 27/03/2023 09:59, Neil Armstrong wrote: >>> On 24/03/2023 22:55, Dmitry Baryshkov wrote: >>>> The first USB PHY on the sm8150 platform is really the USB+DP combo >>>> PHY. Add the DP part of the PHY. >>>> >>>> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> >>>> --- >>>> arch/arm64/boot/dts/qcom/sm8150.dtsi | 17 +++++++++++++---- >>>> 1 file changed, 13 insertions(+), 4 deletions(-) >>>> >>>> diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi >>>> index 9491be4a6bf0..a618218f7b68 100644 >>>> --- a/arch/arm64/boot/dts/qcom/sm8150.dtsi >>>> +++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi >>>> @@ -3392,20 +3392,19 @@ usb_2_hsphy: phy@88e3000 { >>>> }; >>>> usb_1_qmpphy: phy@88e9000 { >>>> - compatible = "qcom,sm8150-qmp-usb3-phy"; >>>> + compatible = "qcom,sm8150-qmp-usb3-dp-phy"; >>>> reg = <0 0x088e9000 0 0x18c>, >>>> - <0 0x088e8000 0 0x10>; >>>> + <0 0x088e8000 0 0x38>, >>>> + <0 0x088ea000 0 0x40>; >>>> status = "disabled"; >>>> #address-cells = <2>; >>>> #size-cells = <2>; >>>> ranges; >>>> - >>>> clocks = <&gcc GCC_USB3_PRIM_PHY_AUX_CLK>, >>>> <&rpmhcc RPMH_CXO_CLK>, >>>> <&gcc GCC_USB3_PRIM_CLKREF_CLK>, >>>> <&gcc GCC_USB3_PRIM_PHY_COM_AUX_CLK>; >>>> clock-names = "aux", "ref_clk_src", "ref", "com_aux"; >>>> - >>>> resets = <&gcc GCC_USB3_DP_PHY_PRIM_BCR>, >>>> <&gcc GCC_USB3_PHY_PRIM_BCR>; >>>> reset-names = "phy", "common"; >>>> @@ -3423,6 +3422,16 @@ usb_1_ssphy: phy@88e9200 { >>>> clock-names = "pipe0"; >>>> clock-output-names = "usb3_phy_pipe_clk_src"; >>>> }; >>>> + >>>> + usb_1_dpphy: phy@88ea200 { >>>> + reg = <0 0x088ea200 0 0x200>, >>>> + <0 0x088ea400 0 0x200>, >>>> + <0 0x088eaa00 0 0x200>, >>>> + <0 0x088ea600 0 0x200>, >>>> + <0 0x088ea800 0 0x200>; >>>> + #clock-cells = <1>; >>>> + #phy-cells = <0>; >>>> + }; >>> >>> Is there a reason why the new flat bindings from qcom,sc8280xp-qmp-usb43dp-phy.yaml are not used instead ? >>> >> >> Oh ok I see "phy: qcom-qmp-combo: convert to newer style of bindings" is the followup of this serie, >> please specify it because it wasn't obvious... > > I thought that a note in the cover letter was good enough, but yeah, maybe it should be more explicit. Do you think it warrants v2? I can send one. IMO it's unnecessary so long as both get in. Konrad >
diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi index 9491be4a6bf0..a618218f7b68 100644 --- a/arch/arm64/boot/dts/qcom/sm8150.dtsi +++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi @@ -3392,20 +3392,19 @@ usb_2_hsphy: phy@88e3000 { }; usb_1_qmpphy: phy@88e9000 { - compatible = "qcom,sm8150-qmp-usb3-phy"; + compatible = "qcom,sm8150-qmp-usb3-dp-phy"; reg = <0 0x088e9000 0 0x18c>, - <0 0x088e8000 0 0x10>; + <0 0x088e8000 0 0x38>, + <0 0x088ea000 0 0x40>; status = "disabled"; #address-cells = <2>; #size-cells = <2>; ranges; - clocks = <&gcc GCC_USB3_PRIM_PHY_AUX_CLK>, <&rpmhcc RPMH_CXO_CLK>, <&gcc GCC_USB3_PRIM_CLKREF_CLK>, <&gcc GCC_USB3_PRIM_PHY_COM_AUX_CLK>; clock-names = "aux", "ref_clk_src", "ref", "com_aux"; - resets = <&gcc GCC_USB3_DP_PHY_PRIM_BCR>, <&gcc GCC_USB3_PHY_PRIM_BCR>; reset-names = "phy", "common"; @@ -3423,6 +3422,16 @@ usb_1_ssphy: phy@88e9200 { clock-names = "pipe0"; clock-output-names = "usb3_phy_pipe_clk_src"; }; + + usb_1_dpphy: phy@88ea200 { + reg = <0 0x088ea200 0 0x200>, + <0 0x088ea400 0 0x200>, + <0 0x088eaa00 0 0x200>, + <0 0x088ea600 0 0x200>, + <0 0x088ea800 0 0x200>; + #clock-cells = <1>; + #phy-cells = <0>; + }; }; usb_2_qmpphy: phy@88eb000 {
The first USB PHY on the sm8150 platform is really the USB+DP combo PHY. Add the DP part of the PHY. Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> --- arch/arm64/boot/dts/qcom/sm8150.dtsi | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-)