On Mon, Oct 28, 2019 at 11:32:49AM +0200, Roger Quadros wrote: > The J721e platform comes with 2 Cadence USB3 controller > instances. This driver supports the TI specific wrapper > on this platform. > > Signed-off-by: Roger Quadros <rogerq@xxxxxx> > Signed-off-by: Sekhar Nori <nsekhar@xxxxxx> > Reviewed-by: Pawel Laszczak <pawell@xxxxxxxxxxx> > --- > drivers/usb/cdns3/Kconfig | 10 ++ > drivers/usb/cdns3/Makefile | 1 + > drivers/usb/cdns3/cdns3-ti.c | 236 +++++++++++++++++++++++++++++++++++ > 3 files changed, 247 insertions(+) > create mode 100644 drivers/usb/cdns3/cdns3-ti.c > > diff --git a/drivers/usb/cdns3/Kconfig b/drivers/usb/cdns3/Kconfig > index d0331613a355..2a1e89d12ed9 100644 > --- a/drivers/usb/cdns3/Kconfig > +++ b/drivers/usb/cdns3/Kconfig > @@ -43,4 +43,14 @@ config USB_CDNS3_PCI_WRAP > If you choose to build this driver as module it will > be dynamically linked and module will be called cdns3-pci.ko > > +config USB_CDNS3_TI > + tristate "Cadence USB3 support on TI platforms" > + depends on ARCH_K3 || COMPILE_TEST > + default USB_CDNS3 > + help > + Say 'Y' or 'M' here if you are building for Texas Instruments > + platforms that contain Cadence USB3 controller core. > + > + e.g. J721e. > + > endif > diff --git a/drivers/usb/cdns3/Makefile b/drivers/usb/cdns3/Makefile > index a703547350bb..948e6b88d1a9 100644 > --- a/drivers/usb/cdns3/Makefile > +++ b/drivers/usb/cdns3/Makefile > @@ -14,3 +14,4 @@ endif > cdns3-$(CONFIG_USB_CDNS3_HOST) += host.o > > obj-$(CONFIG_USB_CDNS3_PCI_WRAP) += cdns3-pci-wrap.o > +obj-$(CONFIG_USB_CDNS3_TI) += cdns3-ti.o > diff --git a/drivers/usb/cdns3/cdns3-ti.c b/drivers/usb/cdns3/cdns3-ti.c > new file mode 100644 > index 000000000000..c6a79ca15858 > --- /dev/null > +++ b/drivers/usb/cdns3/cdns3-ti.c > @@ -0,0 +1,236 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/** > + * cdns3-ti.c - TI specific Glue layer for Cadence USB Controller > + * > + * Copyright (C) 2019 Texas Instruments Incorporated - http://www.ti.com > + */ > + > +#include <linux/bits.h> > +#include <linux/clk.h> > +#include <linux/module.h> > +#include <linux/kernel.h> > +#include <linux/interrupt.h> > +#include <linux/platform_device.h> > +#include <linux/dma-mapping.h> > +#include <linux/io.h> > +#include <linux/of_platform.h> > +#include <linux/pm_runtime.h> > + > +/* USB Wrapper register offsets */ > +#define USBSS_PID 0x0 > +#define USBSS_W1 0x4 > +#define USBSS_STATIC_CONFIG 0x8 > +#define USBSS_PHY_TEST 0xc > +#define USBSS_DEBUG_CTRL 0x10 > +#define USBSS_DEBUG_INFO 0x14 > +#define USBSS_DEBUG_LINK_STATE 0x18 > +#define USBSS_DEVICE_CTRL 0x1c > + > +/* Wrapper 1 register bits */ > +#define USBSS_W1_PWRUP_RST BIT(0) > +#define USBSS_W1_OVERCURRENT_SEL BIT(8) > +#define USBSS_W1_MODESTRAP_SEL BIT(9) > +#define USBSS_W1_OVERCURRENT BIT(16) > +#define USBSS_W1_MODESTRAP_MASK GENMASK(18, 17) > +#define USBSS_W1_MODESTRAP_SHIFT 17 > +#define USBSS_W1_USB2_ONLY BIT(19) > + > +/* Static config register bits */ > +#define USBSS1_STATIC_PLL_REF_SEL_MASK GENMASK(8, 5) > +#define USBSS1_STATIC_PLL_REF_SEL_SHIFT 5 > +#define USBSS1_STATIC_LOOPBACK_MODE_MASK GENMASK(4, 3) > +#define USBSS1_STATIC_LOOPBACK_MODE_SHIFT 3 > +#define USBSS1_STATIC_VBUS_SEL_MASK GENMASK(2, 1) > +#define USBSS1_STATIC_VBUS_SEL_SHIFT 1 > +#define USBSS1_STATIC_LANE_REVERSE BIT(0) > + > +/* Modestrap modes */ > +enum modestrap_mode { USBSS_MODESTRAP_MODE_NONE, > + USBSS_MODESTRAP_MODE_HOST, > + USBSS_MODESTRAP_MODE_PERIPHERAL}; > + > +struct cdns_ti { > + struct device *dev; > + void __iomem *usbss; > + int usb2_only:1; > + int vbus_divider:1; 'bool' instead of bitfields? Makes it more obvious, right? > + struct clk *usb2_refclk; > + struct clk *lpm_clk; > +}; > + > +static const int cdns_ti_rate_table[] = { /* in KHZ */ > + 9600, > + 10000, > + 12000, > + 19200, > + 20000, > + 24000, > + 25000, > + 26000, > + 38400, > + 40000, > + 58000, > + 50000, > + 52000, > +}; > + > +static inline u32 cdns_ti_readl(struct cdns_ti *data, u32 offset) > +{ > + return readl(data->usbss + offset); > +} Does sparse like this function? > + > +static inline void cdns_ti_writel(struct cdns_ti *data, u32 offset, u32 value) > +{ > + writel(value, data->usbss + offset); Same here, have you run sparse on this code? > +} > + > +static int cdns_ti_probe(struct platform_device *pdev) > +{ > + struct device *dev = &pdev->dev; > + struct device_node *node = pdev->dev.of_node; > + struct cdns_ti *data; > + int error; > + u32 reg; > + int rate_code, i; > + unsigned long rate; > + > + data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); > + if (!data) > + return -ENOMEM; > + > + platform_set_drvdata(pdev, data); > + > + data->dev = dev; > + > + data->usbss = devm_platform_ioremap_resource(pdev, 0); > + if (IS_ERR(data->usbss)) { > + dev_err(dev, "can't map IOMEM resource\n"); Doesn't the function print an error? > + return PTR_ERR(data->usbss); > + } > + > + data->usb2_refclk = devm_clk_get(dev, "ref"); > + if (IS_ERR(data->usb2_refclk)) { > + dev_err(dev, "can't get usb2_refclk\n"); Again, doesn't the function print an error? > + return PTR_ERR(data->usb2_refclk); > + } > + > + data->lpm_clk = devm_clk_get(dev, "lpm"); > + if (IS_ERR(data->lpm_clk)) { > + dev_err(dev, "can't get lpm_clk\n"); Same? > + return PTR_ERR(data->lpm_clk); > + } > + > + rate = clk_get_rate(data->usb2_refclk); > + rate /= 1000; /* To KHz */ > + for (i = 0; i < ARRAY_SIZE(cdns_ti_rate_table); i++) { > + if (cdns_ti_rate_table[i] == rate) > + break; > + } > + > + if (i == ARRAY_SIZE(cdns_ti_rate_table)) { > + dev_err(dev, "unsupported usb2_refclk rate: %lu KHz\n", rate); What can userspace do about this? > + return -EINVAL; > + } > + > + rate_code = i; > + > + pm_runtime_enable(dev); > + error = pm_runtime_get_sync(dev); > + if (error < 0) { > + dev_err(dev, "pm_runtime_get_sync failed: %d\n", error); Again, the call should print the error, right? > + goto err_get; > + } > + > + /* assert RESET */ > + reg = cdns_ti_readl(data, USBSS_W1); > + reg &= ~USBSS_W1_PWRUP_RST; > + cdns_ti_writel(data, USBSS_W1, reg); > + > + /* set static config */ > + reg = cdns_ti_readl(data, USBSS_STATIC_CONFIG); > + reg &= ~USBSS1_STATIC_PLL_REF_SEL_MASK; > + reg |= rate_code << USBSS1_STATIC_PLL_REF_SEL_SHIFT; > + > + reg &= ~USBSS1_STATIC_VBUS_SEL_MASK; > + data->vbus_divider = device_property_read_bool(dev, "ti,vbus-divider"); > + if (data->vbus_divider) > + reg |= 1 << USBSS1_STATIC_VBUS_SEL_SHIFT; > + > + cdns_ti_writel(data, USBSS_STATIC_CONFIG, reg); > + reg = cdns_ti_readl(data, USBSS_STATIC_CONFIG); > + > + /* set USB2_ONLY mode if requested */ > + reg = cdns_ti_readl(data, USBSS_W1); > + data->usb2_only = device_property_read_bool(dev, "ti,usb2-only"); > + if (data->usb2_only) > + reg |= USBSS_W1_USB2_ONLY; > + > + /* set default modestrap */ > + reg |= USBSS_W1_MODESTRAP_SEL; > + reg &= ~USBSS_W1_MODESTRAP_MASK; > + reg |= USBSS_MODESTRAP_MODE_NONE << USBSS_W1_MODESTRAP_SHIFT; > + cdns_ti_writel(data, USBSS_W1, reg); > + > + /* de-assert RESET */ > + reg |= USBSS_W1_PWRUP_RST; > + cdns_ti_writel(data, USBSS_W1, reg); > + > + error = of_platform_populate(node, NULL, NULL, dev); > + if (error) { > + dev_err(dev, "failed to create children: %d\n", error); Again, error in the caller? thanks, greg k-h
