| Message ID | 20230501121111.1058190-1-bryan.odonoghue@linaro.org |
|---|---|
| Headers | show |
| Series | Add Qualcomm PMIC TPCM support | expand |
On 02/05/2023 12:47, Konrad Dybcio wrote: >> connector { >> compatible = "usb-c-connector"; >> ports { >> #address-cells = <1>; >> #size-cells = <0>; >> >> port@0 { >> reg = <0>; >> pm8150b_role_switch_out: endpoint { >> remote-endpoint = <&dwc3_role_switch_in>; >> }; >> }; > ==== Hmm, I'm actually recoiling from doing this now that I look at it. - Every other implementation defines the connector in the platfrom dts not the core dtsi - The connector would have to be named something like pm8150b_type_connector which is not consistent with the rest of the Type-C connector namespace Yes it can be done that the connector with the ports can be defined in the pm8150b dtsi but TBH I think it is more confusing that way and not consistent with other implementations. I'm citing prior art here I just think this is nicer/easier to follow - arch/arm64/boot/dts/qcom/sc8280xp.dtsi - arch/arm64/boot/dts/qcom/sc8280xp-lenovo-thinkpad-x13s.dts So for now https://imgflip.com/i/7kw6ck --- bod
I'm dropping the previous V1-V5 log because the text is getting too long here's a link to the previous cover letter instead Link: https://lore.kernel.org/linux-usb/20230413113438.1577658-1-bryan.odonoghue@linaro.org/ Bootable tree Link: https://git.codelinaro.org/bryan.odonoghue/kernel/-/tree/linux-next-23-04-28-pm8150b-tcpm-qcom-wrapper-typec-mux-bjorn Please note patch#5 and patch#6 should be merged in tandem not separately. V6: - I found an off-by-one bug when connecting a Type-C DP cable from Amazon which sends the max PD PDU - bod patch#7 - Drops the partial PHY orientation patch in favour of Bjorn's sc8280xp series - Updates the commit log for the driver per Heikki's comment patch#7 - Adds Heikki's Acked-by to the driver patch#7 - Ammends commit log text - Marijn patch#4 - Typos highlithed - Marijn patch#4 - Adds Reviewed-by - Krzysztof patch#1 - Makes port@0 input required - Krzysztof patch#4 - Incorporate's Krzysztof's comments for quotation marks, minItems patch#5 - Adds Reviewed-by - Krzysztof patch#6 - Defines port/endpoint in sm8250.dtsi - patch#13 Konrad - Luca. Would appreciate your testing. I don't see anything wrong with your reference dts Link: https://github.com/z3ntu/linux/blob/5db89528bd6f0f2e6027f2ad64a9178699be9bc0/arch/arm64/boot/dts/qcom/sm7225-fairphone-fp4.dts If we still can't get it working for you with Bjorn's patches maybe I can find a few cycles to debug with a FP4 after landing the 8250 bit. Caleb has something working for the rb3 based on these patches so, the gap can't be too big. Here's how I'm testing using an RB5/SM8250 A) Confirm device mode works in both orientations - Boot to a Linux shell - Run usb-ecm-up.sh [1] - This provides usb0 on the device side @ 192.168.8.2 - Attach Type-C cable to host PC - On host PC "sudo ifconfig usb0 192.168.8.1" - Ping 192.168.8.2 confirming ping works - Switch cable orientation - On the PC side re-assign the ip address "sudo ifconfig usb0 192.168.8.1" - Ping again confirming connectivity B) Test automatic host-mode transition, orientation and SuperSpeed - Remove cable from PC - Attach a Type-C USB key - Validate the orientation detection cat /sys/class/typec/port0/orientation normal - Confirm the RB5 sees the USB key as a SuperSpeed device - Remove USB key and physically invert it along the horizontal axis - Re-insert the key - Confirm the RB5 sees the USB key as a SuperSpeed device - Validate the orientation detection cat /sys/class/typec/port0/orientation reverse C) Test Type-C accessory automatic host-mode transition, orientation and SuperSpeed - Remove USB key from RB5 - Attach a Type-C USB accessory - Validate the orientation detection cat /sys/class/typec/port0/orientation normal - Attach a SuperSpeed capable USB key to the Type-C accessory - Confirm the RB5 sees the USB key as a SuperSpeed device - Remove the USB key from the Type-C accessory - Remove the Type-C accessory from the RB5 - Physically invert the Type-C accessory along the horizontal axis - Re-insert the Type-C accessory - Validate the orientation detection cat /sys/class/typec/port0/orientation reverse - Re-insert the key - Confirm the RB5 sees the USB key as a SuperSpeed device D) Goto Test-A - Confirming that automatic switch to device mode happens - Repeat the above loop in ~ any order [1] usb-ecm-up.sh #!/usr/bin/env bash # load libcomposite module modprobe libcomposite # ensure function is loaded modprobe usb_f_ecm modprobe usb_f_ncm mount -t configfs none /sys/kernel/config/ # create a gadget mkdir /sys/kernel/config/usb_gadget/g0 # cd to its configfs node cd /sys/kernel/config/usb_gadget/g0 # configure it (vid/pid can be anything if USB Class is used for driver compat) echo 0x0525 > idVendor echo 0xa4a4 > idProduct # configure its serial/mfg/product mkdir strings/0x409 echo 0xCAFEBABE > strings/0x409/serialnumber echo Linaro > strings/0x409/manufacturer echo qrb5165-rb5 > strings/0x409/product # create configs mkdir configs/c.1 mkdir configs/c.1/strings/0x409 # create the function (name must match a usb_f_<name> module such as 'acm') mkdir functions/ncm.0 echo "CDC ECM" > configs/c.1/strings/0x409/configuration # associate function with config ln -s functions/ncm.0 configs/c.1 # Set USB version 3.1 echo 0x0310 > bcdUSB echo "super-speed-plus" > max_speed # enable gadget by binding it to a UDC from /sys/class/udc echo a600000.usb > UDC # to unbind it: echo "" > UDC; sleep 1; rm -rf /sys/kernel/config/usb_gadget/g0 sleep 1 ifconfig usb0 192.168.8.2 Bryan O'Donoghue (13): dt-bindings: regulator: qcom,usb-vbus-regulator: Mark reg as required dt-bindings: regulator: qcom,usb-vbus-regulator: Mark regulator-*-microamp required dt-bindings: phy: qcom,sc7180-qmp-usb3-dp-phy: Add orientation-switch as optional dt-bindings: phy: qcom,sc7180-qmp-usb3-dp-phy: Add input and output ports dt-bindings: usb: Add Qualcomm PMIC Type-C dt-bindings: mfd: qcom,spmi-pmic: Add typec to SPMI device types usb: typec: qcom: Add Qualcomm PMIC Type-C driver arm64: dts: qcom: sm8250: Define ports for qmpphy orientation-switching arm64: dts: qcom: pm8150b: Add a TCPM description arm64: dts: qcom: qrb5165-rb5: Switch on Type-C VBUS boost arm64: dts: qcom: qrb5165-rb5: Switch on basic TCPM arm64: dts: qcom: qrb5165-rb5: Switch on TCPM usb-role-switching for usb_1 arm64: dts: qcom: qrb5165-rb5: Switch on TCPM orientation-switch for usb_1_qmpphy .../bindings/mfd/qcom,spmi-pmic.yaml | 4 + .../phy/qcom,sc7180-qmp-usb3-dp-phy.yaml | 44 ++ .../regulator/qcom,usb-vbus-regulator.yaml | 10 +- .../bindings/usb/qcom,pmic-typec.yaml | 190 ++++++ MAINTAINERS | 10 + arch/arm64/boot/dts/qcom/pm8150b.dtsi | 40 ++ arch/arm64/boot/dts/qcom/qrb5165-rb5.dts | 56 +- arch/arm64/boot/dts/qcom/sm8250.dtsi | 17 + drivers/usb/typec/Kconfig | 13 - drivers/usb/typec/Makefile | 1 - drivers/usb/typec/qcom-pmic-typec.c | 261 -------- drivers/usb/typec/tcpm/Kconfig | 11 + drivers/usb/typec/tcpm/Makefile | 1 + drivers/usb/typec/tcpm/qcom/Makefile | 6 + drivers/usb/typec/tcpm/qcom/qcom_pmic_typec.c | 362 +++++++++++ .../typec/tcpm/qcom/qcom_pmic_typec_pdphy.c | 528 +++++++++++++++++ .../typec/tcpm/qcom/qcom_pmic_typec_pdphy.h | 115 ++++ .../typec/tcpm/qcom/qcom_pmic_typec_port.c | 560 ++++++++++++++++++ .../typec/tcpm/qcom/qcom_pmic_typec_port.h | 194 ++++++ 19 files changed, 2146 insertions(+), 277 deletions(-) create mode 100644 Documentation/devicetree/bindings/usb/qcom,pmic-typec.yaml delete mode 100644 drivers/usb/typec/qcom-pmic-typec.c create mode 100644 drivers/usb/typec/tcpm/qcom/Makefile create mode 100644 drivers/usb/typec/tcpm/qcom/qcom_pmic_typec.c create mode 100644 drivers/usb/typec/tcpm/qcom/qcom_pmic_typec_pdphy.c create mode 100644 drivers/usb/typec/tcpm/qcom/qcom_pmic_typec_pdphy.h create mode 100644 drivers/usb/typec/tcpm/qcom/qcom_pmic_typec_port.c create mode 100644 drivers/usb/typec/tcpm/qcom/qcom_pmic_typec_port.h