| Message ID | 1600771612-30727-1-git-send-email-jun.li@nxp.com |
|---|---|
| Headers | show |
| Series | add NXP imx8mp usb support | expand |
On Tue, 22 Sep 2020 at 12:56, Li Jun <jun.li@nxp.com> wrote: > > imx8mp integrates 2 identical dwc3 based USB3 controllers and > Synopsys phys, each instance has additional wakeup logic to > support low power mode, so the glue layer need a node with dwc3 > core sub node. > > Signed-off-by: Li Jun <jun.li@nxp.com> > --- > arch/arm64/boot/dts/freescale/imx8mp.dtsi | 88 +++++++++++++++++++++++++++++++ > 1 file changed, 88 insertions(+) > > diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi b/arch/arm64/boot/dts/freescale/imx8mp.dtsi > index 9de2aa1..1b7ed4c 100644 > --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi > +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi > @@ -750,5 +750,93 @@ > reg = <0x3d800000 0x400000>; > interrupts = <GIC_SPI 98 IRQ_TYPE_LEVEL_HIGH>; > }; > + > + usb3_phy0: usb-phy@381f0040 { > + compatible = "fsl,imx8mp-usb-phy"; > + reg = <0x381f0040 0x40>; > + clocks = <&clk IMX8MP_CLK_USB_PHY_ROOT>; > + clock-names = "phy"; > + assigned-clocks = <&clk IMX8MP_CLK_USB_PHY_REF>; > + assigned-clock-parents = <&clk IMX8MP_CLK_24M>; > + #phy-cells = <0>; > + status = "disabled"; > + }; > + > + usb3_0: usb@32f10100 { > + compatible = "fsl,imx8mp-dwc3"; > + reg = <0x32f10100 0x8>; > + clocks = <&clk IMX8MP_CLK_HSIO_ROOT>, > + <&clk IMX8MP_CLK_USB_ROOT>; > + clock-names = "hsio", "suspend"; > + interrupts = <GIC_SPI 148 IRQ_TYPE_LEVEL_HIGH>; > + #address-cells = <1>; > + #size-cells = <1>; > + ranges; > + status = "disabled"; > + > + usb_dwc3_0: dwc3@38100000 { > + compatible = "snps,dwc3"; > + reg = <0x38100000 0x10000>; > + clocks = <&clk IMX8MP_CLK_HSIO_AXI>, > + <&clk IMX8MP_CLK_USB_CORE_REF>, > + <&clk IMX8MP_CLK_USB_ROOT>; > + clock-names = "bus_early", "ref", "suspend"; > + assigned-clocks = <&clk IMX8MP_CLK_HSIO_AXI>; > + assigned-clock-parents = <&clk IMX8MP_SYS_PLL2_500M>; > + assigned-clock-rates = <500000000>; > + interrupts = <GIC_SPI 40 IRQ_TYPE_LEVEL_HIGH>; > + phys = <&usb3_phy0>, <&usb3_phy0>; > + phy-names = "usb2-phy", "usb3-phy"; > + snps,dis-u2-freeclk-exists-quirk; > + xhci-64bit-support-disable; > + status = "disabled"; > + }; > + > + }; > + > + usb3_phy1: usb-phy@382f0040 { > + compatible = "fsl,imx8mp-usb-phy"; > + reg = <0x382f0040 0x40>; > + clocks = <&clk IMX8MP_CLK_USB_PHY_ROOT>; > + clock-names = "phy"; > + assigned-clocks = <&clk IMX8MP_CLK_USB_PHY_REF>; > + assigned-clock-parents = <&clk IMX8MP_CLK_24M>; > + #phy-cells = <0>; > + status = "disabled"; > + }; > + > + usb3_1: usb@32f10108 { > + compatible = "fsl,imx8mp-dwc3"; > + reg = <0x32f10108 0x8>; > + clocks = <&clk IMX8MP_CLK_HSIO_ROOT>, > + <&clk IMX8MP_CLK_USB_ROOT>; > + clock-names = "hsio", "suspend"; > + assigned-clocks = <&clk IMX8MP_CLK_HSIO_AXI_SRC>; This does not validate against your own schema: arch/arm64/boot/dts/freescale/imx8mp-evk.dt.yaml: usb@32f10108: 'assigned-clock-parents', 'assigned-clock-rates', 'assigned-clocks' do not match any of the regexes: '^dwc3@[0-9a-f]+$', 'pinctrl-[0-9]+' Please, stop adding new schema and DTS which from day one have warnings/violations. It's really a lot of effort (and commits) to clean this up later. Best regards, Krzysztof
On Tue, Sep 22, 2020 at 06:46:51PM +0800, Li Jun wrote: > imx8mp integrates 2 identical dwc3 based USB3 controllers and > Synopsys phys, each instance has additional wakeup logic to > support low power mode, so the glue layer need a node with dwc3 > core sub node. > > Signed-off-by: Li Jun <jun.li@nxp.com> > --- > arch/arm64/boot/dts/freescale/imx8mp.dtsi | 88 +++++++++++++++++++++++++++++++ > 1 file changed, 88 insertions(+) > > diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi b/arch/arm64/boot/dts/freescale/imx8mp.dtsi > index 9de2aa1..1b7ed4c 100644 > --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi > +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi > + usb3_1: usb@32f10108 { > + compatible = "fsl,imx8mp-dwc3"; > + reg = <0x32f10108 0x8>; > + clocks = <&clk IMX8MP_CLK_HSIO_ROOT>, > + <&clk IMX8MP_CLK_USB_ROOT>; > + clock-names = "hsio", "suspend"; > + assigned-clocks = <&clk IMX8MP_CLK_HSIO_AXI_SRC>; In Linux-5.9-rc6 this clock doesn't exist anymore. Should be IMX8MP_CLK_HSIO_AXI > + assigned-clock-parents = <&clk IMX8MP_SYS_PLL2_500M>; > + assigned-clock-rates = <500000000>; > + interrupts = <GIC_SPI 149 IRQ_TYPE_LEVEL_HIGH>; > + #address-cells = <1>; > + #size-cells = <1>; > + ranges; > + status = "disabled"; > + > + usb_dwc3_1: dwc3@38200000 { > + compatible = "snps,dwc3"; > + reg = <0x38200000 0x10000>; > + clocks = <&clk IMX8MP_CLK_HSIO_AXI>, > + <&clk IMX8MP_CLK_USB_CORE_REF>, > + <&clk IMX8MP_CLK_USB_ROOT>; > + clock-names = "bus_early", "ref", "suspend"; > + assigned-clocks = <&clk IMX8MP_CLK_HSIO_AXI>; > + assigned-clock-parents = <&clk IMX8MP_SYS_PLL2_500M>; > + assigned-clock-rates = <500000000>; > + interrupts = <GIC_SPI 41 IRQ_TYPE_LEVEL_HIGH>; > + phys = <&usb3_phy1>, <&usb3_phy1>; > + phy-names = "usb2-phy", "usb3-phy"; > + snps,dis-u2-freeclk-exists-quirk; > + xhci-64bit-support-disable; > + status = "disabled"; Does it make sense for a board to enable the parent node and leave this one disabled? If not you can drop this status = "disabled" here. Sascha
> -----Original Message----- > From: Sascha Hauer <s.hauer@pengutronix.de> > Sent: Wednesday, September 23, 2020 3:51 PM > To: Jun Li <jun.li@nxp.com> > Cc: robh+dt@kernel.org; shawnguo@kernel.org; balbi@kernel.org; > mathias.nyman@intel.com; gregkh@linuxfoundation.org; > kernel@pengutronix.de; festevam@gmail.com; dl-linux-imx > <linux-imx@nxp.com>; Anson Huang <anson.huang@nxp.com>; Aisheng Dong > <aisheng.dong@nxp.com>; Peng Fan <peng.fan@nxp.com>; Andy Duan > <fugang.duan@nxp.com>; Joakim Zhang <qiangqing.zhang@nxp.com>; Horia > Geanta <horia.geanta@nxp.com>; linux-usb@vger.kernel.org; > devicetree@vger.kernel.org; linux-arm-kernel@lists.infradead.org > Subject: Re: [PATCH v3 5/6] arm64: dtsi: imx8mp: add usb nodes > > On Tue, Sep 22, 2020 at 06:46:51PM +0800, Li Jun wrote: > > imx8mp integrates 2 identical dwc3 based USB3 controllers and Synopsys > > phys, each instance has additional wakeup logic to support low power > > mode, so the glue layer need a node with dwc3 core sub node. > > > > Signed-off-by: Li Jun <jun.li@nxp.com> > > --- > > arch/arm64/boot/dts/freescale/imx8mp.dtsi | 88 > > +++++++++++++++++++++++++++++++ > > 1 file changed, 88 insertions(+) > > > > diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi > > b/arch/arm64/boot/dts/freescale/imx8mp.dtsi > > index 9de2aa1..1b7ed4c 100644 > > --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi > > +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi > > + usb3_1: usb@32f10108 { > > + compatible = "fsl,imx8mp-dwc3"; > > + reg = <0x32f10108 0x8>; > > + clocks = <&clk IMX8MP_CLK_HSIO_ROOT>, > > + <&clk IMX8MP_CLK_USB_ROOT>; > > + clock-names = "hsio", "suspend"; > > + assigned-clocks = <&clk IMX8MP_CLK_HSIO_AXI_SRC>; > > In Linux-5.9-rc6 this clock doesn't exist anymore. Should be > IMX8MP_CLK_HSIO_AXI Will change. > > > + assigned-clock-parents = <&clk IMX8MP_SYS_PLL2_500M>; > > + assigned-clock-rates = <500000000>; > > + interrupts = <GIC_SPI 149 IRQ_TYPE_LEVEL_HIGH>; > > + #address-cells = <1>; > > + #size-cells = <1>; > > + ranges; > > + status = "disabled"; > > + > > + usb_dwc3_1: dwc3@38200000 { > > + compatible = "snps,dwc3"; > > + reg = <0x38200000 0x10000>; > > + clocks = <&clk IMX8MP_CLK_HSIO_AXI>, > > + <&clk IMX8MP_CLK_USB_CORE_REF>, > > + <&clk IMX8MP_CLK_USB_ROOT>; > > + clock-names = "bus_early", "ref", "suspend"; > > + assigned-clocks = <&clk IMX8MP_CLK_HSIO_AXI>; > > + assigned-clock-parents = <&clk IMX8MP_SYS_PLL2_500M>; > > + assigned-clock-rates = <500000000>; > > + interrupts = <GIC_SPI 41 IRQ_TYPE_LEVEL_HIGH>; > > + phys = <&usb3_phy1>, <&usb3_phy1>; > > + phy-names = "usb2-phy", "usb3-phy"; > > + snps,dis-u2-freeclk-exists-quirk; > > + xhci-64bit-support-disable; > > + status = "disabled"; > > Does it make sense for a board to enable the parent node and leave this one > disabled? If not you can drop this status = "disabled" here. OK, will drop it. Thanks Li Jun > > Sascha > > -- > Pengutronix e.K. | | > Steuerwalder Str. 21 | > https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.pe > ngutronix.de%2F&data=02%7C01%7Cjun.li%40nxp.com%7Cadbbbfe2edec4cca1 > 38908d85f956382%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C6373644425 > 20571896&sdata=07ku8kTQCv8lv8qdRiOcU3CX4lly3503LO4bUDIbjow%3D&r > eserved=0 | > 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | > Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
NXP imx8MPlus integrates 2 indentical dwc3 3.30b IP with additional wakeup logic to support low power, this wakeup logic has a separated interrupt which can generate events with suspend clock(32K); due to SoC integration limitation, a few quriks required, instead of create new properties flags, introduce platform data and pass it from glue layer to dwc3 core, those xhci private data can further pass to xhci-plat. changes for v3: - Add dwc3 core related clocks into dwc3 core node, and glue layer driver only handle the clocks(hsio and suspend) for glue block, this is to match real HW. - Change to use property "xhci-64bit-support-disable" to disable 64bit DMA as imx8mp USB integration actully can't support it, so remove platform data in v2. - Some changes of imx8mp usb driver binding doc to address comments from Rob Changes for v2: - Drop the 2 patches for new property("snps,xhci-dis-64bit-support-quirk") introduction, as suggested, imply by SoC compatible string, this is done by introduce dwc3 core platform data and pass the xhci_plat_priv to xhci-plat for those xhci quirks, so a new patch added: [1/5] usb: dwc3: add platform data to dwc3 core device to pass data. this patch is based on Peter's one patch which is also in review: https://patchwork.kernel.org/patch/11640945/ - dts change, use the USB power function of TRL logic instead of a always-on regulator to control vbus on/off. - Some changes to address Peter's command on patch [2/5]. Li Jun (6): dt-bindings: usb: xhci: add property to disable xhci 64bit support usb: host: xhci-plat: add support of XHCI_NO_64BIT_SUPPORT via property dt-bindings: usb: dwc3-imx8mp: add imx8mp dwc3 glue bindings usb: dwc3: add imx8mp dwc3 glue layer driver arm64: dtsi: imx8mp: add usb nodes arm64: dts: imx8mp-evk: enable usb1 as host mode .../devicetree/bindings/usb/fsl,imx8mp-dwc3.yaml | 99 ++++++ Documentation/devicetree/bindings/usb/usb-xhci.txt | 3 + arch/arm64/boot/dts/freescale/imx8mp-evk.dts | 21 ++ arch/arm64/boot/dts/freescale/imx8mp.dtsi | 88 +++++ drivers/usb/dwc3/Kconfig | 10 + drivers/usb/dwc3/Makefile | 1 + drivers/usb/dwc3/dwc3-imx8mp.c | 363 +++++++++++++++++++++ drivers/usb/host/xhci-plat.c | 4 + 8 files changed, 589 insertions(+) create mode 100644 Documentation/devicetree/bindings/usb/fsl,imx8mp-dwc3.yaml create mode 100644 drivers/usb/dwc3/dwc3-imx8mp.c