On 4/10/2020 7:18 PM, Stephen Boyd wrote: > Quoting Wesley Cheng (2020-04-07 15:23:27) >> diff --git a/drivers/phy/qualcomm/Kconfig b/drivers/phy/qualcomm/Kconfig >> index e46824d..3158c14 100644 >> --- a/drivers/phy/qualcomm/Kconfig >> +++ b/drivers/phy/qualcomm/Kconfig >> @@ -85,6 +85,16 @@ config PHY_QCOM_USB_HS >> Support for the USB high-speed ULPI compliant phy on Qualcomm >> chipsets. >> >> +config PHY_QCOM_USB_SNPS_HS_7NM >> + tristate "Qualcomm 7nm USB HS PHY module" >> + depends on OF && (ARCH_QCOM || COMPILE_TEST) >> + select GENERIC_PHY >> + help >> + Enable support for the USB high-speed SNPS phy on Qualcomm 7nm >> + chipsets. This PHY has differences in the register map compared >> + to the 28nm variant. The PHY is paired with a Synopsys DWC3 USB >> + controller on MSM SOCs. > > s/MSM/Qualcomm/? > Hi Stephen, Thanks for the catch. Will replace with Qualcomm. >> + >> config PHY_QCOM_USB_HSIC >> tristate "Qualcomm USB HSIC ULPI PHY module" >> depends on USB_ULPI_BUS >> diff --git a/drivers/phy/qualcomm/phy-qcom-snps-7nm.c b/drivers/phy/qualcomm/phy-qcom-snps-7nm.c >> new file mode 100644 >> index 0000000..00a46cd >> --- /dev/null >> +++ b/drivers/phy/qualcomm/phy-qcom-snps-7nm.c >> @@ -0,0 +1,294 @@ > [...] >> + >> +/** >> + * struct qcom_snps_hsphy - structure holding snps hs phy attributes > > snps hs phy attributes > > structure and holding are redundant. > Understood. >> + * >> + * @phy: generic phy >> + * @base: iomapped memory space for qubs2 phy >> + * >> + * @cfg_ahb_clk: AHB2PHY interface clock >> + * @ref_clk: phy reference clock >> + * @iface_clk: phy interface clock >> + * @phy_reset: phy reset control >> + * @vregs: regulator supplies bulk data >> + >> + * @phy_initialized: if PHY has been initialized correctly > > Please drop space between members in the kernel-doc. > Vinod actually pointed this out already, and I fixed this in v6. >> + * >> + */ >> +struct qcom_snps_hsphy { >> + struct phy *phy; >> + void __iomem *base; >> + >> + struct clk *cfg_ahb_clk; >> + struct clk *ref_clk; >> + struct reset_control *phy_reset; >> + struct regulator_bulk_data vregs[SNPS_HS_NUM_VREGS]; >> + >> + bool phy_initialized; >> +}; >> + >> +static inline void qcom_snps_hsphy_write_mask(void __iomem *base, u32 offset, >> + u32 mask, u32 val) >> +{ >> + u32 reg; >> + >> + reg = readl(base + offset); >> + reg &= ~mask; >> + reg |= val & mask; >> + writel(reg, base + offset); > > Make these relaxed? > >> + >> + /* Ensure above write is completed */ >> + readl(base + offset); > > And this one too? > Agreed, using the _relaxed() variant for these will be fine. I'll replace them in the next revision. >> +} >> + >> +static int qcom_snps_hsphy_init(struct phy *phy) >> +{ >> + struct qcom_snps_hsphy *hsphy = phy_get_drvdata(phy); >> + int ret; >> + >> + dev_vdbg(&phy->dev, "%s(): Initializing SNPS HS phy\n", __func__); >> + >> + /* turn on regulator supplies */ > > These comments don't tell us anything besides what the code is doing so > they don't seem useful. > Will remove the comments. >> + ret = regulator_bulk_enable(ARRAY_SIZE(hsphy->vregs), hsphy->vregs); >> + if (ret) >> + return ret; >> + >> + /* enable ahb interface clock to program phy */ >> + ret = clk_prepare_enable(hsphy->cfg_ahb_clk); >> + if (ret) { >> + dev_err(&phy->dev, "failed to enable cfg ahb clock, %d\n", ret); >> + goto poweroff_phy; >> + } >> + >> + /* Perform phy reset */ >> + ret = reset_control_assert(hsphy->phy_reset); >> + if (ret) { >> + dev_err(&phy->dev, "failed to assert phy_reset, %d\n", ret); >> + goto disable_ahb_clk; >> + } >> + >> + /* 100 us delay to keep PHY in reset mode */ >> + usleep_range(100, 150); >> + >> + ret = reset_control_deassert(hsphy->phy_reset); >> + if (ret) { >> + dev_err(&phy->dev, "failed to de-assert phy_reset, %d\n", ret); >> + goto disable_ahb_clk; >> + } >> + >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_CFG0, >> + UTMI_PHY_CMN_CTRL_OVERRIDE_EN, >> + UTMI_PHY_CMN_CTRL_OVERRIDE_EN); >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_UTMI_CTRL5, >> + POR, POR); >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_HS_PHY_CTRL_COMMON0, >> + FSEL_MASK, 0); >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_HS_PHY_CTRL_COMMON1, >> + PLLBTUNE, PLLBTUNE); >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_REFCLK_CTRL, >> + REFCLK_SEL_DEFAULT, REFCLK_SEL_MASK); >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_HS_PHY_CTRL_COMMON1, >> + VBUSVLDEXTSEL0, VBUSVLDEXTSEL0); >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_HS_PHY_CTRL1, >> + VBUSVLDEXT0, VBUSVLDEXT0); >> + >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_HS_PHY_CTRL_COMMON2, >> + VREGBYPASS, VREGBYPASS); >> + >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_HS_PHY_CTRL2, >> + USB2_SUSPEND_N_SEL | USB2_SUSPEND_N, >> + USB2_SUSPEND_N_SEL | USB2_SUSPEND_N); >> + >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_UTMI_CTRL0, >> + SLEEPM, SLEEPM); >> + >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_UTMI_CTRL5, >> + POR, 0); >> + >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_HS_PHY_CTRL2, >> + USB2_SUSPEND_N_SEL, 0); >> + >> + qcom_snps_hsphy_write_mask(hsphy->base, USB2_PHY_USB_PHY_CFG0, >> + UTMI_PHY_CMN_CTRL_OVERRIDE_EN, 0); >> + >> + hsphy->phy_initialized = true; >> + >> + return 0; >> + >> +disable_ahb_clk: >> + clk_disable_unprepare(hsphy->cfg_ahb_clk); >> +poweroff_phy: >> + regulator_bulk_disable(ARRAY_SIZE(hsphy->vregs), hsphy->vregs); >> + >> + return ret; >> +} >> + >> +static int qcom_snps_hsphy_exit(struct phy *phy) >> +{ >> + struct qcom_snps_hsphy *hsphy = phy_get_drvdata(phy); >> + >> + reset_control_assert(hsphy->phy_reset); >> + clk_disable_unprepare(hsphy->cfg_ahb_clk); >> + regulator_bulk_disable(ARRAY_SIZE(hsphy->vregs), hsphy->vregs); >> + hsphy->phy_initialized = false; >> + >> + return 0; >> +} >> + >> +static const struct phy_ops qcom_snps_hsphy_gen_ops = { >> + .init = qcom_snps_hsphy_init, >> + .exit = qcom_snps_hsphy_exit, >> + .owner = THIS_MODULE, >> +}; >> + >> +static const struct of_device_id qcom_snps_hsphy_of_match_table[] = { >> + { >> + .compatible = "qcom,sm8150-usb-hs-phy", >> + }, >> + { >> + .compatible = "qcom,usb-snps-hs-7nm-phy", >> + }, >> + { }, > > Can this be on one line? > > { .compatible = "qcom,sm8150-usb-hs-phy", }, > { .compatible = "qcom,usb-snps-hs-7nm-phy", }, > { } > > Also drop comma on the last element so nothing can follow without > causing compile error. > Got it. >> +}; >> +MODULE_DEVICE_TABLE(of, qcom_snps_hsphy_of_match_table); >> + >> +static int qcom_snps_hsphy_probe(struct platform_device *pdev) >> +{ >> + struct device *dev = &pdev->dev; >> + struct qcom_snps_hsphy *hsphy; >> + struct phy_provider *phy_provider; >> + struct phy *generic_phy; >> + struct resource *res; >> + int ret, i; >> + int num; >> + >> + hsphy = devm_kzalloc(dev, sizeof(*hsphy), GFP_KERNEL); >> + if (!hsphy) >> + return -ENOMEM; >> + >> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); >> + hsphy->base = devm_ioremap_resource(dev, res); > > devm_platform_ioremap_resource()? > Sure, will replace + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + hsphy->base = devm_ioremap_resource(dev, res); to: + hsphy->base = devm_platform_ioremap_resource(pdev, 0); >> + if (IS_ERR(hsphy->base)) >> + return PTR_ERR(hsphy->base); >> + >> + hsphy->ref_clk = devm_clk_get(dev, "ref"); >> + if (IS_ERR(hsphy->ref_clk)) { >> + ret = PTR_ERR(hsphy->ref_clk); >> + if (ret != -EPROBE_DEFER) >> + dev_err(dev, "failed to get ref clk, %d\n", ret); >> + return ret; >> + } >> + >> + hsphy->phy_reset = devm_reset_control_get_exclusive(&pdev->dev, NULL); >> + if (IS_ERR(hsphy->phy_reset)) { >> + dev_err(dev, "failed to get phy core reset\n"); >> + return PTR_ERR(hsphy->phy_reset); >> + } >> + >> + num = ARRAY_SIZE(hsphy->vregs); >> + for (i = 0; i < num; i++) >> + hsphy->vregs[i].supply = qcom_snps_hsphy_vreg_names[i]; >> + >> + ret = devm_regulator_bulk_get(dev, num, hsphy->vregs); >> + if (ret) { >> + if (ret != -EPROBE_DEFER) >> + dev_err(dev, "failed to get regulator supplies: %d\n", >> + ret); >> + return ret; >> + } >> + >> + generic_phy = devm_phy_create(dev, NULL, &qcom_snps_hsphy_gen_ops); >> + if (IS_ERR(generic_phy)) { >> + ret = PTR_ERR(generic_phy); >> + dev_err(dev, "failed to create phy, %d\n", ret); >> + return ret; >> + } >> + hsphy->phy = generic_phy; >> + >> + dev_set_drvdata(dev, hsphy); >> + phy_set_drvdata(generic_phy, hsphy); >> + >> + phy_provider = devm_of_phy_provider_register(dev, of_phy_simple_xlate); >> + if (!IS_ERR(phy_provider)) >> + dev_info(dev, "Registered Qcom-SNPS HS phy\n"); > > dev_dbg()? Or just remove it? > I'd like to keep some kind of print here. Will replace with dev_dbg so it can be enabled when required. Thanks again! >> + >> + return PTR_ERR_OR_ZERO(phy_provider); >> +} >> + >> +static struct platform_driver qcom_snps_hsphy_driver = { >> + .probe = qcom_snps_hsphy_probe, >> + .driver = { >> + .name = "qcom-snps-hs-7nm-phy", >> + .of_match_table = qcom_snps_hsphy_of_match_table, >> + }, >> +}; >> + >> +module_platform_driver(qcom_snps_hsphy_driver); >> + >> +MODULE_DESCRIPTION("Qualcomm 7nm USB HS PHY driver"); -- The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project