arm64: dts: rockchip: rk3588-nanopc-t6: fix usb-c port functionality
Subsystems (2)
PATCH 1/1 arm64: dts: rockchip: rk3588-nanopc-t6: fix usb-c port functionality
The USB-C port on the NanoPC-T6 was not providing VBUS (vbus5v0_typec regulator disabled, gpio-58 out lo) due to misconfiguration. The original setup with regulator-always-on and regulator-boot-on forced the port on, masking the issue, but removing these properties revealed that the fusb302 driver was not enabling the regulator dynamically.
Changes:
- Removed regulator-always-on and regulator-boot-on from vbus5v0_typec and vbus5v0_usb to allow driver control.
- Changed power-role from "source" to "dual" in the usb-c-connector to support OTG functionality.
- Add pd-revision = /bits/ 8 <0x2 0x0 0x1 0x2>; to the FUSB302MPX node to specify USB Power Delivery (PD) Revision 2.0, Version 1.2, ensuring the driver correctly advertises PD capabilities and negotiates power roles (source/sink) per the FUSB302MPX’s supported PD protocol.
- Added op-sink-microwatt and sink-pdos for proper sink mode configuration (1w min, 15w max).
- Add typec-power-opmode = "3.0A"; to enable 3.0A (15W) fallback for non-PD USB-C devices with the FUSB302MPX.
- Set try-power-role to "source" to prioritize VBUS enablement.
- Adjusted usb_host0_xhci dr_mode from "host" to "otg" and added usb-role-switch for dual-role support.
Testing:
- Verified VBUS (5V) delivery to a sink device (USB thumb drive).
- Confirmed USB host mode with lsusb detecting connected devices.
- Validated USB device mode with adb devices when connected to a PC.
- Tested dual-role (OTG) functionality with try-power-role set to "source" and "sink"; "source" prioritizes faster VBUS activation.
- Validated functionality with a mobile device, including USB Power Delivery, file transfer, USB tethering, MIDI, and image transfer.
- Tested USB-C Ethernet adapter compatibility, ensuring proper operation in host mode.
- Tested USB-C hub compatibility, ensuring proper operation in host mode.
--- .../boot/dts/rockchip/rk3588-nanopc-t6.dtsi | 21 ++++++++++--------- 1 file changed, 11 insertions(+), 10 deletions(-)
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +11 -10
@@ -174,8 +174,6 @@ vbus5v0_typec: regulator-vbus5v0-typec { gpio = <&gpio1 RK_PD2 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&typec5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_typec"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>;
@@ -188,8 +186,6 @@ vbus5v0_usb: regulator-vbus5v0-usb { gpio = <&gpio4 RK_PB0 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&usb5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_usb"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>;
@@ -465,24 +461,30 @@ regulator-state-mem { }; &i2c6 { - clock-frequency = <200000>; status = "okay"; - fusb302: typec-portc@22 { + usbc0: usb-typec@22 { compatible = "fcs,fusb302"; reg = <0x22>; interrupt-parent = <&gpio0>; interrupts = <RK_PD3 IRQ_TYPE_LEVEL_LOW>; - pinctrl-0 = <&usbc0_int>; pinctrl-names = "default"; + pinctrl-0 = <&usbc0_int>; vbus-supply = <&vbus5v0_typec>; + status = "okay"; connector { compatible = "usb-c-connector"; data-role = "dual"; label = "USB-C"; - power-role = "source"; + /* fusb302 supports PD Rev 2.0 Ver 1.2 */ + pd-revision = /bits/ 8 <0x2 0x0 0x1 0x2>; + power-role = "dual"; + op-sink-microwatt = <1000000>; + sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)>; source-pdos = <PDO_FIXED(5000, 2000, PDO_FIXED_USB_COMM)>; + try-power-role = "source"; + typec-power-opmode = "3.0A"; ports { #address-cells = <1>;
@@ -1135,9 +1137,8 @@ &usb_host0_ohci { }; &usb_host0_xhci { - dr_mode = "host"; - status = "okay"; usb-role-switch; + status = "okay"; port { usb_host0_xhci_drd_sw: endpoint {
Signature · 1 line
2.39.5
Hi John,
Quoted 85 lines · John Clark
The USB-C port on the NanoPC-T6 was not providing VBUS (
vbus5v0_typecregulator disabled, gpio-58 out lo) due to misconfiguration. The original setup with regulator-always-on and regulator-boot-on forced the port on, masking the issue, but removing these properties revealed that the fusb302 driver was not enabling the regulator dynamically.Changes:
- Removed regulator-always-on and regulator-boot-on fromvbus5v0_typecandvbus5v0_usbto allow driver control.
- Changed power-role from "source" to "dual" in the usb-c-connector to support OTG functionality.
- Add pd-revision = /bits/ 8 <0x20x00x10x2>; to the FUSB302MPX node to specify USB Power Delivery (PD) Revision 2.0, Version 1.2, ensuring the driver correctly advertises PD capabilities and negotiates power roles (source/sink) per the FUSB302MPX’s supported PD protocol.
- Added op-sink-microwatt and sink-pdos for proper sink mode configuration (1w min, 15w max).
- Add typec-power-opmode = "3.0A"; to enable 3.0A (15W) fallback for non-PD USB-C devices with the FUSB302MPX.
- Set try-power-role to "source" to prioritize VBUS enablement.
- Adjustedusb_host0_xhcidr_modefrom "host" to "otg" and added usb-role-switch for dual-role support.Testing:
- Verified VBUS (5V) delivery to a sink device (USB thumb drive).
- Confirmed USB host mode with lsusb detecting connected devices.
- Validated USB device mode with adb devices when connected to a PC.
- Tested dual-role (OTG) functionality with try-power-role set to "source" and "sink"; "source" prioritizes faster VBUS activation.
- Validated functionality with a mobile device, including USB Power Delivery, file transfer, USB tethering, MIDI, and image transfer.
- Tested USB-C Ethernet adapter compatibility, ensuring proper operation in host mode.
- Tested USB-C hub compatibility, ensuring proper operation in host mode.Signed-off-by John Clark <[email protected]>--- .../boot/dts/rockchip/rk3588-nanopc-t6.dtsi | 21 ++++++++++--------- 1 file changed, 11 insertions(+), 10 deletions(-)arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +3
-7@@ -174,8 +174,6 @@ vbus5v0_typec: regulator-vbus5v0-typec { gpio = <&gpio1 RK_PD2 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&typec5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_typec"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>;@@ -188,8 +186,6 @@ vbus5v0_usb: regulator-vbus5v0-usb { gpio = <&gpio4 RK_PB0 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&usb5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_usb"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>;@@ -465,24 +461,30 @@ regulator-state-mem { }; &i2c6 { - clock-frequency = <200000>; status = "okay"; - fusb302: typec-portc@22 { + usbc0: usb-typec@22 { compatible = "fcs,fusb302"; reg = <0x22>; interrupt-parent = <&gpio0>; interrupts = <RK_PD3 IRQ_TYPE_LEVEL_LOW>; - pinctrl-0 = <&usbc0_int>; pinctrl-names = "default"; + pinctrl-0 = <&usbc0_int>; vbus-supply = <&vbus5v0_typec>; + status = "okay"; connector { compatible = "usb-c-connector"; data-role = "dual";
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi regulator-state-mem { · line 487
Quoted 10 lines
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +7
-1label = "USB-C"; - power-role = "source"; + /* fusb302 supports PD Rev 2.0 Ver 1.2 */ + pd-revision = /bits/ 8 <0x2 0x0 0x1 0x2>; + power-role = "dual"; + op-sink-microwatt = <1000000>; + sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)>; source-pdos = <PDO_FIXED(5000, 2000, PDO_FIXED_USB_COMM)>; + try-power-role = "source"; + typec-power-opmode = "3.0A";
According to the manufacturer wiki [1] "Power Output Capacity" table, the USB-C port maximum output is 5V/2A. So I think "1.5A" would be a better value here.
Quoted 4 lines
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +0
-0ports { #address-cells = <1>;
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi &usb_host0_ohci { · line 1144
Quoted 10 lines
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +1
-2@@ -1135,9 +1137,8 @@ &usb_host0_ohci { }; &usb_host0_xhci { - dr_mode = "host"; - status = "okay"; usb-role-switch; + status = "okay"; port { usb_host0_xhci_drd_sw: endpoint {
br, Hugh
Quoted 50 lines · Hugh Cole-Baker
Hi John,
Quoted 33 lines · John Clark
The USB-C port on the NanoPC-T6 was not providing VBUS (
vbus5v0_typecregulator disabled, gpio-58 out lo) due to misconfiguration. The original setup with regulator-always-on and regulator-boot-on forced the port on, masking the issue, but removing these properties revealed that the fusb302 driver was not enabling the regulator dynamically.Signed-off-by John Clark <[email protected]>--- .../boot/dts/rockchip/rk3588-nanopc-t6.dtsi | 21 ++++++++++--------- 1 file changed, 11 insertions(+), 10 deletions(-)arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +1
-4@@ -174,8 +174,6 @@ vbus5v0_typec: regulator-vbus5v0-typec { gpio = <&gpio1 RK_PD2 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&typec5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_typec";@@ -465,24 +461,30 @@ regulator-state-mem { &i2c6 { - clock-frequency = <200000>; status = "okay"; - fusb302: typec-portc@22 { + usbc0: usb-typec@22 { compatible = "fcs,fusb302"; connector { compatible = "usb-c-connector"; data-role = "dual";arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi regulator-state-mem { · line 477
Quoted 10 lines
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +7
-1label = "USB-C"; - power-role = "source"; + /* fusb302 supports PD Rev 2.0 Ver 1.2 */ + pd-revision = /bits/ 8 <0x2 0x0 0x1 0x2>; + power-role = "dual"; + op-sink-microwatt = <1000000>; + sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)>; source-pdos = <PDO_FIXED(5000, 2000, PDO_FIXED_USB_COMM)>; + try-power-role = "source"; + typec-power-opmode = "3.0A";According to the manufacturer wiki [1] "Power Output Capacity" table, the USB-C port maximum output is 5V/2A. So I think "1.5A" would be a better value here.
Quoted 1 line
Hi Hugh,
Thank you for your feedback and for referencing the manufacturer’s wiki. To clarify, the line in question:
source-pdos = <PDO_FIXED(5000, 2000, PDO_FIXED_USB_COMM)>;
was not modified in my patch. This value, which sets the USB-C port’s source capability to 5V/2A, is unchanged from the existing device tree.
Furthermore, the schematic for the NanoPC-T6 and the vendor’s upstream Linux device tree (e.g., FriendlyARM’s kernel at https://github.com/friendlyarm/kernel-rockchip/blob/nanopi6-v6.1.y/arch/arm64/boot/dts/rockchip/rk3588-nanopi6-rev01.dts#L346) also specify the source capability as 5V/2A. Given this consistency across the schematic and vendor sources, I believe 2A is an appropriate value for this configuration.
Please let me know if you have further concerns or suggestions!
Best regards,
John Clark
Quoted 53 lines · John Clark
Quoted 50 lines · Hugh Cole-Baker
Hi John,
Quoted 33 lines · John Clark
The USB-C port on the NanoPC-T6 was not providing VBUS (
vbus5v0_typecregulator disabled, gpio-58 out lo) due to misconfiguration. The original setup with regulator-always-on and regulator-boot-on forced the port on, masking the issue, but removing these properties revealed that the fusb302 driver was not enabling the regulator dynamically.Signed-off-by John Clark <[email protected]>--- .../boot/dts/rockchip/rk3588-nanopc-t6.dtsi | 21 ++++++++++--------- 1 file changed, 11 insertions(+), 10 deletions(-)arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +1
-4@@ -174,8 +174,6 @@ vbus5v0_typec: regulator-vbus5v0-typec { gpio = <&gpio1 RK_PD2 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&typec5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_typec";@@ -465,24 +461,30 @@ regulator-state-mem { &i2c6 { - clock-frequency = <200000>; status = "okay"; - fusb302: typec-portc@22 { + usbc0: usb-typec@22 { compatible = "fcs,fusb302"; connector { compatible = "usb-c-connector"; data-role = "dual";arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi regulator-state-mem { · line 477
Quoted 10 lines
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +7
-1label = "USB-C"; - power-role = "source"; + /* fusb302 supports PD Rev 2.0 Ver 1.2 */ + pd-revision = /bits/ 8 <0x2 0x0 0x1 0x2>; + power-role = "dual"; + op-sink-microwatt = <1000000>; + sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)>; source-pdos = <PDO_FIXED(5000, 2000, PDO_FIXED_USB_COMM)>; + try-power-role = "source"; + typec-power-opmode = "3.0A";According to the manufacturer wiki [1] "Power Output Capacity" table, the USB-C port maximum output is 5V/2A. So I think "1.5A" would be a better value here.
Quoted 1 line
Quoted 7 lines
Hi Hugh,
Thank you for your feedback and for referencing the manufacturer’s wiki. To clarify, the line in question:
source-pdos = <
PDO_FIXED(5000, 2000, PDO_FIXED_USB_COMM)>;was not modified in my patch. This value, which sets the USB-C port’s source capability to 5V/2A, is unchanged from the existing device tree.
source-pdos is OK, I was referring to the typec-power-opmode = "3.0A" line just above where I commented. As I read it, that'd mean the board would advertise via the CC pins that it could source 5V/3A when USB PD is not in use. The wiki says it can only source up to 5V/2A.
br, Hugh
Quoted 7 lines
Furthermore, the schematic for the NanoPC-T6 and the vendor’s upstream Linux device tree (e.g., FriendlyARM’s kernel at https://github.com/friendlyarm/kernel-rockchip/blob/nanopi6-v6.1.y/arch/arm64/boot/dts/rockchip/rk3588-nanopi6-rev01.dts#L346) also specify the source capability as 5V/2A. Given this consistency across the schematic and vendor sources, I believe 2A is an appropriate value for this configuration.
Please let me know if you have further concerns or suggestions!
Best regards,
John Clark
PATCH v2 1/1 arm64: dts: rockchip: rk3588-nanopc-t6: fix usb-c port functionality
In reply to John ClarkThe USB-C port on the NanoPC-T6 was not providing VBUS (vbus5v0_typec regulator disabled, gpio-58 out lo) due to misconfiguration. The original setup with regulator-always-on and regulator-boot-on forced the port on, masking the issue, but removing these properties revealed that the fusb302 driver was not enabling the regulator dynamically.
Changes:
- Removed regulator-always-on and regulator-boot-on from vbus5v0_typec and vbus5v0_usb to allow driver control.
- Changed power-role from "source" to "dual" in the usb-c-connector to support OTG functionality.
- Added pd-revision = /bits/ 8 <0x2 0x0 0x1 0x2> to the FUSB302MPX node to specify USB Power Delivery (PD) Revision 2.0, Version 1.2, ensuring the driver correctly advertises PD capabilities and negotiates power roles (source/sink).
- Added op-sink-microwatt and sink-pdos for proper sink mode configuration (1W min, 15W max).
- Added typec-power-opmode = "1.5A" to enable 1.5A fallback for non-PD USB-C devices, aligning with the 5V/2A hardware limit.
- Set try-power-role to "source" to prioritize VBUS enablement.
- Adjusted usb_host0_xhci dr_mode from "host" to "otg" and added usb-role-switch for dual-role support.
Testing:
- Verified VBUS (5V) delivery to a sink device (USB thumb drive).
- Confirmed USB host mode with lsusb detecting connected devices.
- Validated USB device mode with adb devices when connected to a PC.
- Tested dual-role (OTG) functionality with try-power-role set to "source" and "sink"; "source" prioritizes faster VBUS activation.
- Validated functionality with a mobile device, including USB Power Delivery, file transfer, USB tethering, MIDI, and image transfer.
- Tested USB-C Ethernet adapter compatibility in host mode.
- Tested USB-C hub compatibility in host mode.
---
Changes in v2:
- Changed typec-power-opmode from "3.0A" to "1.5A" to conservatively align with the 5V/2A hardware limit, addressing feedback from Hugh Cole-Baker.
--- .../boot/dts/rockchip/rk3588-nanopc-t6.dtsi | 21 ++++++++++--------- 1 file changed, 11 insertions(+), 10 deletions(-)
arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi +11 -10
@@ -174,8 +174,6 @@ vbus5v0_typec: regulator-vbus5v0-typec { gpio = <&gpio1 RK_PD2 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&typec5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_typec"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>;
@@ -188,8 +186,6 @@ vbus5v0_usb: regulator-vbus5v0-usb { gpio = <&gpio4 RK_PB0 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&usb5v_pwren>; - regulator-always-on; - regulator-boot-on; regulator-name = "vbus5v0_usb"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>;
@@ -465,24 +461,30 @@ regulator-state-mem { }; &i2c6 { - clock-frequency = <200000>; status = "okay"; - fusb302: typec-portc@22 { + usbc0: usb-typec@22 { compatible = "fcs,fusb302"; reg = <0x22>; interrupt-parent = <&gpio0>; interrupts = <RK_PD3 IRQ_TYPE_LEVEL_LOW>; - pinctrl-0 = <&usbc0_int>; pinctrl-names = "default"; + pinctrl-0 = <&usbc0_int>; vbus-supply = <&vbus5v0_typec>; + status = "okay"; connector { compatible = "usb-c-connector"; data-role = "dual"; label = "USB-C"; - power-role = "source"; + op-sink-microwatt = <1000000>; + /* fusb302 supports PD Rev 2.0 Ver 1.2 */ + pd-revision = /bits/ 8 <0x2 0x0 0x1 0x2>; + power-role = "dual"; + sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)>; source-pdos = <PDO_FIXED(5000, 2000, PDO_FIXED_USB_COMM)>; + try-power-role = "source"; + typec-power-opmode = "1.5A"; ports { #address-cells = <1>;
@@ -1135,9 +1137,8 @@ &usb_host0_ohci { }; &usb_host0_xhci { - dr_mode = "host"; - status = "okay"; usb-role-switch; + status = "okay"; port { usb_host0_xhci_drd_sw: endpoint {
Signature · 1 line
2.39.5
Quoted 23 lines · John Clark
The USB-C port on the NanoPC-T6 was not providing VBUS (
vbus5v0_typecregulator disabled, gpio-58 out lo) due to misconfiguration. The original setup with regulator-always-on and regulator-boot-on forced the port on, masking the issue, but removing these properties revealed that the fusb302 driver was not enabling the regulator dynamically.Changes:
- Removed regulator-always-on and regulator-boot-on fromvbus5v0_typecandvbus5v0_usbto allow driver control.
- Changed power-role from "source" to "dual" in the usb-c-connector to support OTG functionality.
- Added pd-revision = /bits/ 8 <0x20x00x10x2> to the FUSB302MPX node to specify USB Power Delivery (PD) Revision 2.0, Version 1.2, ensuring the driver correctly advertises PD capabilities and negotiates power roles (source/sink).
- Added op-sink-microwatt and sink-pdos for proper sink mode configuration (1W min, 15W max).
- Added typec-power-opmode = "1.5A" to enable 1.5A fallback for non-PD USB-C devices, aligning with the 5V/2A hardware limit.
- Set try-power-role to "source" to prioritize VBUS enablement.
- Adjustedusb_host0_xhcidr_modefrom "host" to "otg" and added usb-role-switch for dual-role support.
Quoted 1 line
[...]
Applied, thanks!
[1/1] arm64: dts: rockchip: rk3588-nanopc-t6: fix usb-c port functionality
commit: 60087bcbd1206a546c570c453dee5f5d961ef5b3
Best regards,
Signature · 1 line
Heiko Stuebner <[email protected]>