Message ID | 20200303171159.246992-1-bryan.odonoghue@linaro.org (mailing list archive) |
---|---|
Headers | show |
Series | Enable Qualcomm QCS 404 HS/SS USB | expand |
On Tue 03 Mar 09:11 PST 2020, Bryan O'Donoghue wrote: > V1: > This series enables the Primary and Secondary USB controllers on the > QCS404, associated PHYs, role-switching and DTS descriptions. > Finally took the time to give this a spin on my QCS404 dev board. Tested-by: Bjorn Andersson <bjorn.andersson@linaro.org> As this touches three different subsystems, and doesn't have have compile time dependencies between the parts, I would suggest that as you fix up the build error I reported yesterday you send v8 as three different series - one per maintainer/subsystem. That way we avoid any questions about whom should merge what parts and in what order. Thanks, Bjorn > The series takes in a number of patches worked on by a number of people > over the past few years from downstream, through to previous upstream > submissions for both of these interfaces. Additional work has been done to > enable USB role-switching. > > 1. USB SS > - extcon has been dropped in favour of gpio-usb-conn as discussed and > agreed previously by Jorge, Bjorn, Stephen Boyd and Jack Pham [1]. > > - Regulator API has been updated following suggestions from Bjorn. > > - Sanitzation of the DT compatible name - dropped "snps" entirely > from the name - it made almost no sense to me and doesn't appear > consistent with similar naming conventions for Snopsys based IP. > > 2. USB HS > - Regulator API changes here. > - Dropped "snps" from the namespace for similar reasons as above. > - Dropped "28nm" from the namespace, looked superfluous. > - Changed "High-Speed" to "Hi-Speed". > - [2] > > 3. DWC3 Role switching > - At the time usb-gpio-conn was discussed it was mentioned that > role-switching was absent from the DWC3 driver. > - John Stultz has some patches in-flight for that, that I've included in > this series for completeness. > - Adds my SoB to relevant patches. > - Drops gerrit ChangeId. > > 4. DWC3 usb-gpio-conn > Needs to be a child node of the DWC3 driver so some code and DT binding > is required for that. > > 5. QCOM-DWC3 > Since we are role-switching with an external PMIC supplying VBUS we want > to skip past toggling VBUS from QCOM-DWC3 controlled registers, so a > patch is added to the qcom-dwc3 driver to do that. > > References: > > 1: USB SS PHY for Qualcomm's QCS404 > https://lwn.net/ml/devicetree/20190207111734.24171-1-jorge.ramirez-ortiz@linaro.org/ > > 2: Add Synopsys High-Speed USB PHY driver for Qualcomm SoCs > https://lore.kernel.org/linux-arm-msm/20181127100722.9993-3-shawn.guo@linaro.org/ > > https://www.spinics.net/lists/linux-usb/msg190003.html > > V2: > - Fixes yaml error - Rob's YAML robot > - Exclusive control of reset in PHY drivers - Philipp Zabel > > V3: > - Fixes typo generating YAML warning - Rob's YAML robot > > V4: > > https://lore.kernel.org/linux-arm-msm/20200122185610.131930-1-bryan.odonoghue@linaro.org/ > > - Churn names of PHYs - Rob Herring > Rob questioned the name of the SuperSpeed PHY in V3. > Looking at available data > > usb-hs-28nm - There are two possible PHYs on 28nm litho node > so it makes sense to name the PHY relating to its relevant > litho node. > > usb-ss - This is not litho node dependent and is used on multiple SoCs > and litho nodes. > > - Drop default mode for role switching - Felipe Balbi > Felipe asked if the default mode for role switching was > required and the answer is no. It makes no difference > becuase the USB ID pin ultimately dictates the mode of operation. > > - Change "gpio_usb_connector" to "connector" - Rob > This was a minor change in terms of DTS but, means I need to look for the > DTS compatible string as opposed to a label given in the DTS. > No matter what the name of the label, this is he right thing to do. > > - Used IS_ENABLED() - Felipe > The logic is the same but IS_ENABLED() is used now. > > - Retained example of USB connector in dwc.txt - Rob, Felipe > Rob pointed out adding the connector was redundant as the documentation > already implies it. > Felipe seemed in favour of I think adding the example. > I've dropped the documentation of the connector and kept the example. > https://lore.kernel.org/linux-arm-msm/20200122185610.131930-7-bryan.odonoghue@linaro.org/ > > - Added example of usb-role-switch in dwc3.txt - BOD > > - Incorporated various inputs from Rob on DTS/YAML > - Added required: > - Added additionalProperties: > - Renamed "phy" clock to "ahb" > - maxItems dropped as indicated > > V5: > - https://lkml.org/lkml/2020/2/6/913 > > - Adds a notifier to DWC3 - BOD > This is done in order to allow propagation of role-switch events from the > DWC3 core to an associated binding layer. > > - Re-use the existent EXTCON VBUS power lane toggle logic - Jack Pham > Jack flagged this for inclusion and as a result we need to make a > small change to the qcom binding layer. > > - Squash DTS changes - BOD > I've squashed down some of the DTS changes to stop the patch count in > this series going up any further. > > V6: > - https://lkml.org/lkml/2020/2/7/632 > > - Add RB Jack Pham patch # 11 > > - Fix a stale description in git log patch # 10 > > V7: > - https://lkml.org/lkml/2020/2/10/258 > > - Adds RB Rob Herring as indicated > > - Updates the naming of the USB SS PHY for the particular SoC. > The name of the driver is maintained, as previously mentioned, this > IP appears to be in use on the 20nm and 28nm nodes. However taking > Rob's feedback on the naming of the SoC specific bit the code being added > is specified @ 28nm. > If/when we come to add in 20nm for this IP we can add a new compatible > for 20nm. > Rob Herring > > Bjorn Andersson (1): > arm64: dts: qcom: qcs404: Add USB devices and PHYs > > Bryan O'Donoghue (11): > dt-bindings: usb: dwc3: Add a gpio-usb-connector example > dt-bindings: usb: dwc3: Add a usb-role-switch to the example > usb: dwc3: qcom: Add support for usb-conn-gpio connectors > usb: dwc3: Add support for usb-conn-gpio connectors > usb: dwc3: Add support for a role-switch notifier > usb: dwc3: qcom: Enable gpio-usb-conn based role-switching > arm64: dts: qcom: qcs404-evb: Define VBUS pins > arm64: dts: qcom: qcs404-evb: Define USB ID pin > arm64: dts: qcom: qcs404-evb: Describe external VBUS regulator > arm64: dts: qcom: qcs404-evb: Raise vreg_l12_3p3 minimum voltage > arm64: dts: qcom: qcs404-evb: Enable USB controllers > > Jorge Ramirez-Ortiz (3): > dt-bindings: phy: remove qcom-dwc3-usb-phy > dt-bindings: Add Qualcomm USB SuperSpeed PHY bindings > phy: qualcomm: usb: Add SuperSpeed PHY driver > > Shawn Guo (1): > phy: qualcomm: Add Synopsys 28nm Hi-Speed USB PHY driver > > Sriharsha Allenki (1): > dt-bindings: phy: Add Qualcomm Synopsys Hi-Speed USB PHY binding > > Yu Chen (1): > usb: dwc3: Registering a role switch in the DRD code. > > .../bindings/phy/qcom,usb-hs-28nm.yaml | 90 ++++ > .../devicetree/bindings/phy/qcom,usb-ss.yaml | 83 ++++ > .../bindings/phy/qcom-dwc3-usb-phy.txt | 37 -- > .../devicetree/bindings/usb/dwc3.txt | 9 + > arch/arm64/boot/dts/qcom/qcs404-evb.dtsi | 90 +++- > arch/arm64/boot/dts/qcom/qcs404.dtsi | 100 +++++ > drivers/phy/qualcomm/Kconfig | 20 + > drivers/phy/qualcomm/Makefile | 2 + > drivers/phy/qualcomm/phy-qcom-usb-hs-28nm.c | 415 ++++++++++++++++++ > drivers/phy/qualcomm/phy-qcom-usb-ss.c | 246 +++++++++++ > drivers/usb/dwc3/core.h | 22 + > drivers/usb/dwc3/drd.c | 119 ++++- > drivers/usb/dwc3/dwc3-qcom.c | 31 +- > 13 files changed, 1223 insertions(+), 41 deletions(-) > create mode 100644 Documentation/devicetree/bindings/phy/qcom,usb-hs-28nm.yaml > create mode 100644 Documentation/devicetree/bindings/phy/qcom,usb-ss.yaml > delete mode 100644 Documentation/devicetree/bindings/phy/qcom-dwc3-usb-phy.txt > create mode 100644 drivers/phy/qualcomm/phy-qcom-usb-hs-28nm.c > create mode 100644 drivers/phy/qualcomm/phy-qcom-usb-ss.c > > -- > 2.25.1 >
On 08/03/2020 05:23, Bjorn Andersson wrote: > On Tue 03 Mar 09:11 PST 2020, Bryan O'Donoghue wrote: > >> V1: >> This series enables the Primary and Secondary USB controllers on the >> QCS404, associated PHYs, role-switching and DTS descriptions. >> > > Finally took the time to give this a spin on my QCS404 dev board. > > Tested-by: Bjorn Andersson <bjorn.andersson@linaro.org> thanks ! > > As this touches three different subsystems, and doesn't have have > compile time dependencies between the parts, I would suggest that as you > fix up the build error I reported yesterday you send v8 as three > different series - one per maintainer/subsystem. That way we avoid any > questions about whom should merge what parts and in what order. ack, will do. --- bod