Re: [PATCH 3/6] phy: qcom: Introduce PCIe UNIPHY 28LP driver

From: Dmitry Baryshkov
Date: Tue Oct 03 2023 - 11:16:10 EST

Next message: Andy Shevchenko: "Re: [PATCH 34/36] treewide: rename pinctrl_gpio_direction_input_new()"
Previous message: Geert Uytterhoeven: "Re: [PATCH v2 02/28] clk: renesas: rzg2l: wait for status bit of SD mux before continuing"
In reply to: Nitheesh Sekar: "[PATCH 3/6] phy: qcom: Introduce PCIe UNIPHY 28LP driver"
Next in thread: Krzysztof Kozlowski: "Re: [PATCH 3/6] phy: qcom: Introduce PCIe UNIPHY 28LP driver"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Tue, 3 Oct 2023 at 15:09, Nitheesh Sekar <quic_nsekar@xxxxxxxxxxx> wrote:
>
> Add Qualcomm PCIe UNIPHY 28LP driver support present
> in Qualcomm IPQ5018 SoC and the phy init sequence.
>
> Signed-off-by: Nitheesh Sekar <quic_nsekar@xxxxxxxxxxx>
> ---
> drivers/phy/qualcomm/Kconfig | 12 +
> drivers/phy/qualcomm/Makefile | 1 +
> .../phy/qualcomm/phy-qcom-uniphy-pcie-28lp.c | 336 ++++++++++++++++++
> 3 files changed, 349 insertions(+)
> create mode 100644 drivers/phy/qualcomm/phy-qcom-uniphy-pcie-28lp.c
>
> diff --git a/drivers/phy/qualcomm/Kconfig b/drivers/phy/qualcomm/Kconfig
> index d891058b7c39..b7d37cd98f02 100644
> --- a/drivers/phy/qualcomm/Kconfig
> +++ b/drivers/phy/qualcomm/Kconfig
> @@ -154,6 +154,18 @@ config PHY_QCOM_M31_USB
> management. This driver is required even for peripheral only or
> host only mode configurations.
>
> +config PHY_QCOM_UNIPHY_PCIE_28LP
> + bool "PCIE UNIPHY 28LP PHY driver"
> + depends on ARCH_QCOM
> + depends on HAS_IOMEM
> + depends on OF
> + select GENERIC_PHY
> + help
> + Enable this to support the PCIe UNIPHY 28LP phy transceiver that
> + is used with PCIe controllers on Qualcomm IPQ5018 chips. It
> + handles PHY initialization, clock management required after
> + resetting the hardware and power management.
> +
> config PHY_QCOM_USB_HS
> tristate "Qualcomm USB HS PHY module"
> depends on USB_ULPI_BUS
> diff --git a/drivers/phy/qualcomm/Makefile b/drivers/phy/qualcomm/Makefile
> index ffd609ac6233..31105cd17bc9 100644
> --- a/drivers/phy/qualcomm/Makefile
> +++ b/drivers/phy/qualcomm/Makefile
> @@ -17,6 +17,7 @@ obj-$(CONFIG_PHY_QCOM_QMP_USB_LEGACY) += phy-qcom-qmp-usb-legacy.o
> obj-$(CONFIG_PHY_QCOM_QUSB2) += phy-qcom-qusb2.o
> obj-$(CONFIG_PHY_QCOM_SNPS_EUSB2) += phy-qcom-snps-eusb2.o
> obj-$(CONFIG_PHY_QCOM_EUSB2_REPEATER) += phy-qcom-eusb2-repeater.o
> +obj-$(CONFIG_PHY_QCOM_UNIPHY_PCIE_28LP) += phy-qcom-uniphy-pcie-28lp.o
> obj-$(CONFIG_PHY_QCOM_USB_HS) += phy-qcom-usb-hs.o
> obj-$(CONFIG_PHY_QCOM_USB_HSIC) += phy-qcom-usb-hsic.o
> obj-$(CONFIG_PHY_QCOM_USB_HS_28NM) += phy-qcom-usb-hs-28nm.o
> diff --git a/drivers/phy/qualcomm/phy-qcom-uniphy-pcie-28lp.c b/drivers/phy/qualcomm/phy-qcom-uniphy-pcie-28lp.c
> new file mode 100644
> index 000000000000..5ef6ae7276cf
> --- /dev/null
> +++ b/drivers/phy/qualcomm/phy-qcom-uniphy-pcie-28lp.c
> @@ -0,0 +1,336 @@
> +// SPDX-License-Identifier: GPL-2.0+
> +/*
> + * Copyright (c) 2023, The Linux Foundation. All rights reserved.
> + */
> +
> +#include <linux/clk.h>
> +#include <linux/clk-provider.h>
> +#include <linux/err.h>
> +#include <linux/io.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/platform_device.h>
> +#include <linux/phy/phy.h>
> +#include <linux/reset.h>
> +#include <linux/of_device.h>
> +#include <linux/delay.h>
> +#include <linux/mfd/syscon.h>
> +#include <linux/regmap.h>
> +
> +#define PIPE_CLK_DELAY_MIN_US 5000
> +#define PIPE_CLK_DELAY_MAX_US 5100
> +#define CDR_CTRL_REG_1 0x80
> +#define CDR_CTRL_REG_2 0x84
> +#define CDR_CTRL_REG_3 0x88
> +#define CDR_CTRL_REG_4 0x8C
> +#define CDR_CTRL_REG_5 0x90
> +#define CDR_CTRL_REG_6 0x94
> +#define CDR_CTRL_REG_7 0x98
> +#define SSCG_CTRL_REG_1 0x9c
> +#define SSCG_CTRL_REG_2 0xa0
> +#define SSCG_CTRL_REG_3 0xa4
> +#define SSCG_CTRL_REG_4 0xa8
> +#define SSCG_CTRL_REG_5 0xac
> +#define SSCG_CTRL_REG_6 0xb0
> +#define PCS_INTERNAL_CONTROL_2 0x2d8
> +
> +#define PHY_MODE_FIXED 0x1
> +
> +enum qcom_uniphy_pcie_type {
> + PHY_TYPE_PCIE = 1,
> + PHY_TYPE_PCIE_GEN2,
> + PHY_TYPE_PCIE_GEN3,
> +};
> +
> +struct uniphy_regs {
> + unsigned int offset;
> + unsigned int val;
> +};
> +
> +struct uniphy_pcie_data {

Please stick to a single symbol/struct prefix.

> + int lanes;
> + /* 2nd lane offset */
> + int lane_offset;
> + unsigned int phy_type;
> + const struct uniphy_regs *init_seq;
> + unsigned int init_seq_num;
> +};
> +
> +struct qcom_uniphy_pcie {
> + struct phy phy;
> + struct device *dev;
> + const struct uniphy_pcie_data *data;
> + struct clk_bulk_data *clks;
> + int num_clks;
> + struct reset_control *resets;
> + void __iomem *base;
> +};
> +
> +#define phy_to_dw_phy(x) container_of((x), struct qca_uni_pcie_phy, phy)
> +
> +static const struct uniphy_regs ipq5018_regs[] = {
> + {
> + .offset = SSCG_CTRL_REG_4,
> + .val = 0x1cb9,
> + },
> + {

"}, {", on the same line.

> + .offset = SSCG_CTRL_REG_5,
> + .val = 0x023a,
> + },
> + {
> + .offset = SSCG_CTRL_REG_3,
> + .val = 0xd360,
> + },
> + {
> + .offset = SSCG_CTRL_REG_1,
> + .val = 0x1,
> + },
> + {
> + .offset = SSCG_CTRL_REG_2,
> + .val = 0xeb,
> + },
> + {
> + .offset = CDR_CTRL_REG_4,
> + .val = 0x3f9,
> + },
> + {
> + .offset = CDR_CTRL_REG_5,
> + .val = 0x1c9,
> + },
> + {
> + .offset = CDR_CTRL_REG_2,
> + .val = 0x419,
> + },
> + {
> + .offset = CDR_CTRL_REG_1,
> + .val = 0x200,
> + },
> + {
> + .offset = PCS_INTERNAL_CONTROL_2,
> + .val = 0xf101,
> + },
> +};
> +
> +static const struct uniphy_pcie_data ipq5018_2x2_data = {
> + .lanes = 2,
> + .lane_offset = 0x800,
> + .phy_type = PHY_TYPE_PCIE_GEN2,
> + .init_seq = ipq5018_regs,
> + .init_seq_num = ARRAY_SIZE(ipq5018_regs),
> +};
> +
> +static void qcom_uniphy_pcie_init(struct qcom_uniphy_pcie *phy)
> +{
> + const struct uniphy_pcie_data *data = phy->data;
> + const struct uniphy_regs *init_seq;
> + void __iomem *base = phy->base;
> + int lane = 0;
> + int i;
> +
> + while (lane != data->lanes) {

for (lane = 0; ...)

> + init_seq = data->init_seq;
> +
> + for (i = 0; i < data->init_seq_num; i++, init_seq++)
> + writel(init_seq->val, base + init_seq->offset);
> +
> + if (data->lanes == 2)
> + base = base + data->lane_offset;

Drop the if(). Use +=.

> +
> + lane++;
> + }
> +}
> +
> +static int qcom_uniphy_pcie_power_off(struct phy *x)
> +{
> + struct qcom_uniphy_pcie *phy = phy_get_drvdata(x);
> +
> + reset_control_assert(phy->resets);
> +
> + clk_bulk_disable_unprepare(phy->num_clks, phy->clks);

Judging from power_on(), the order should be opposite.

> +
> + return 0;
> +}
> +
> +static int qcom_uniphy_pcie_power_on(struct phy *x)
> +{
> + int ret;
> + struct qcom_uniphy_pcie *phy = phy_get_drvdata(x);
> +
> + ret = reset_control_assert(phy->resets);
> + if (ret) {
> + dev_err(phy->dev, "reset assert failed (%d)\n", ret);
> + return ret;
> + }
> +
> + /*
> + * Delay periods before and after reset deassert are working values
> + * from downstream Codeaurora kernel
> + */
> + usleep_range(100, 150);
> +
> + ret = reset_control_deassert(phy->resets);
> + if (ret) {
> + dev_err(phy->dev, "reset deassert failed (%d)\n", ret);
> + return ret;
> + }
> +
> + usleep_range(PIPE_CLK_DELAY_MIN_US, PIPE_CLK_DELAY_MAX_US);
> +
> + ret = clk_bulk_prepare_enable(phy->num_clks, phy->clks);
> + if (ret) {
> + dev_err(phy->dev, "clk prepare and enable failed %d\n", ret);
> + return ret;

Shouldn't the reset be asserted again?

> + }
> +
> + usleep_range(30, 50);

In the same function you have one usleep_range with the defined
constants and another one using plain numbers. Could you please stick
to the uniform scheme?

> +
> + qcom_uniphy_pcie_init(phy);
> + return 0;
> +}
> +
> +static int qcom_uniphy_pcie_get_resources(struct platform_device *pdev,
> + struct qcom_uniphy_pcie *phy)

inline it

> +{
> + struct resource *res;
> +
> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + phy->base = devm_ioremap_resource(phy->dev, res);

devm_platform_get_and_ioremap_resource()

> + if (IS_ERR(phy->base)) {
> + dev_err(phy->dev, "cannot get phy registers\n");
> + return PTR_ERR(phy->base);
> + }
> +
> + phy->num_clks = devm_clk_bulk_get_all(phy->dev, &phy->clks);
> + if (phy->num_clks < 0)
> + return phy->num_clks;
> +
> + phy->resets = devm_reset_control_array_get_exclusive(phy->dev);
> + if (IS_ERR(phy->resets))
> + return PTR_ERR(phy->resets);
> +
> + return 0;
> +}
> +
> +/*
> + * Register a fixed rate pipe clock.
> + *
> + * The <s>_pipe_clksrc generated by PHY goes to the GCC that gate
> + * controls it. The <s>_pipe_clk coming out of the GCC is requested
> + * by the PHY driver for its operations.
> + * We register the <s>_pipe_clksrc here. The gcc driver takes care
> + * of assigning this <s>_pipe_clksrc as parent to <s>_pipe_clk.
> + * Below picture shows this relationship.
> + *
> + * +---------------+
> + * | PHY block |<<---------------------------------------+
> + * | | |
> + * | +-------+ | +-----+ |
> + * I/P---^-->| PLL |---^--->pipe_clksrc--->| GCC |--->pipe_clk---+
> + * clk | +-------+ | +-----+
> + * +---------------+
> + */
> +static int phy_pipe_clk_register(struct qcom_uniphy_pcie *phy,
> + struct device_node *np)
> +{
> + struct clk_fixed_rate *fixed;
> + struct clk_init_data init = { };
> + int ret;
> +
> + ret = of_property_read_string(np, "clock-output-names", &init.name);
> + if (ret) {
> + dev_err(phy->dev, "%pOFn: No clock-output-names\n", np);
> + return ret;
> + }

Can we drop clock-output-names please.

> +
> + fixed = devm_kzalloc(phy->dev, sizeof(*fixed), GFP_KERNEL);
> + if (!fixed)
> + return -ENOMEM;
> +
> + init.ops = &clk_fixed_rate_ops;
> + fixed->fixed_rate = 125000000;
> + fixed->hw.init = &init;
> +
> + ret = devm_clk_hw_register(phy->dev, &fixed->hw);
> + if (ret)
> + return ret;

devm_clk_hw_register_fixed_rate(). Then the whole function can be inlined.

> +
> + ret = devm_of_clk_add_hw_provider(phy->dev, of_clk_hw_simple_get,
> + &fixed->hw);
> + if (ret)
> + return ret;
> +
> + return 0;
> +}
> +
> +static const struct of_device_id qcom_uniphy_pcie_id_table[] = {
> + {
> + .compatible = "qcom,ipq5018-uniphy-pcie-gen2x2",

Bindings describe both 2x1 and 2x2 PHYs, but the driver supports only
2x2. Either of that should be fixed.

> + .data = &ipq5018_2x2_data,
> + },
> + { /* Sentinel */ },
> +};
> +MODULE_DEVICE_TABLE(of, qcom_uniphy_pcie_id_table);
> +
> +static const struct phy_ops pcie_ops = {
> + .power_on = qcom_uniphy_pcie_power_on,
> + .power_off = qcom_uniphy_pcie_power_off,
> + .owner = THIS_MODULE,
> +};
> +
> +static int qcom_uniphy_pcie_probe(struct platform_device *pdev)
> +{
> + struct qcom_uniphy_pcie *phy;
> + int ret;
> + struct phy *generic_phy;
> + struct phy_provider *phy_provider;
> + struct device *dev = &pdev->dev;
> + struct device_node *np = of_node_get(dev->of_node);
> +
> + phy = devm_kzalloc(&pdev->dev, sizeof(*phy), GFP_KERNEL);
> + if (!phy)
> + return -ENOMEM;
> +
> + platform_set_drvdata(pdev, phy);
> + phy->dev = &pdev->dev;
> +
> + phy->data = of_device_get_match_data(dev);
> + if (!phy->data)
> + return -EINVAL;
> +
> + ret = qcom_uniphy_pcie_get_resources(pdev, phy);
> + if (ret < 0) {
> + dev_err(&pdev->dev, "failed to get resources: %d\n", ret);
> + return ret;
> + }
> +
> + ret = phy_pipe_clk_register(phy, np);
> + if (ret)
> + dev_err(&pdev->dev, "failed to register phy pipe clk\n");
> +
> + generic_phy = devm_phy_create(phy->dev, NULL, &pcie_ops);
> + if (IS_ERR(generic_phy))
> + return PTR_ERR(generic_phy);
> +
> + phy_set_drvdata(generic_phy, phy);
> + phy_provider = devm_of_phy_provider_register(phy->dev,
> + of_phy_simple_xlate);
> + if (IS_ERR(phy_provider))
> + return PTR_ERR(phy_provider);
> +
> + return 0;
> +}
> +
> +static struct platform_driver qcom_uniphy_pcie_driver = {
> + .probe = qcom_uniphy_pcie_probe,
> + .driver = {
> + .name = "qcom-uniphy-pcie",
> + .owner = THIS_MODULE,
> + .of_match_table = qcom_uniphy_pcie_id_table,
> + },
> +};
> +
> +module_platform_driver(qcom_uniphy_pcie_driver);
> +
> +MODULE_ALIAS("platform:qcom-uniphy-pcie");
> +MODULE_LICENSE("Dual BSD/GPL");
> +MODULE_DESCRIPTION("PCIE QCOM UNIPHY driver");
> --
> 2.17.1
>

--
With best wishes
Dmitry

Next message: Andy Shevchenko: "Re: [PATCH 34/36] treewide: rename pinctrl_gpio_direction_input_new()"
Previous message: Geert Uytterhoeven: "Re: [PATCH v2 02/28] clk: renesas: rzg2l: wait for status bit of SD mux before continuing"
In reply to: Nitheesh Sekar: "[PATCH 3/6] phy: qcom: Introduce PCIe UNIPHY 28LP driver"
Next in thread: Krzysztof Kozlowski: "Re: [PATCH 3/6] phy: qcom: Introduce PCIe UNIPHY 28LP driver"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]