[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ef55499c-4edc-60fe-d936-83a2df78ee88@canonical.com>
Date: Thu, 24 Jun 2021 10:45:13 +0200
From: Krzysztof Kozlowski <krzysztof.kozlowski@...onical.com>
To: Jagan Teki <jagan@...rulasolutions.com>,
Peng Fan <peng.fan@....com>, Shawn Guo <shawnguo@...nel.org>,
Sascha Hauer <s.hauer@...gutronix.de>,
Tomasz Figa <t.figa@...sung.com>,
Fancy Fang <chen.fang@....com>
Cc: devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
dri-devel@...ts.freedesktop.org, linux-phy@...ts.infradead.org,
linux-kernel@...r.kernel.org, NXP Linux Team <linux-imx@....com>,
linux-amarula@...rulasolutions.com,
Anthony Brandon <anthony@...rulasolutions.com>,
Francis Laniel <francis.laniel@...rulasolutions.com>,
Matteo Lisi <matteo.lisi@...icam.com>,
Milco Pratesi <milco.pratesi@...icam.com>,
Kishon Vijay Abraham I <kishon@...com>,
Vinod Koul <vkoul@...nel.org>
Subject: Re: [RFC PATCH 4/9] phy: samsung: Add SEC DSIM DPHY driver
On 21/06/2021 09:24, Jagan Teki wrote:
> Samsung SEC MIPI DSIM DPHY controller is part of registers
> available in SEC MIPI DSIM bridge for NXP's i.MX8M Mini and
> Nano Processors.
>
> Add phy driver for it.
>
> Cc: Kishon Vijay Abraham I <kishon@...com>
> Cc: Vinod Koul <vkoul@...nel.org>
> Signed-off-by: Jagan Teki <jagan@...rulasolutions.com>
> ---
> drivers/phy/samsung/Kconfig | 9 +
> drivers/phy/samsung/Makefile | 1 +
> drivers/phy/samsung/phy-sec-dsim-dphy.c | 236 ++++++++++++++++++++++++
> 3 files changed, 246 insertions(+)
> create mode 100644 drivers/phy/samsung/phy-sec-dsim-dphy.c
You add a driver for a Samsung's component, so please Cc respective
folks. They might help you with it or provide comments to avoid
duplication of drivers/code:
Around phy:
Marek Szyprowski <m.szyprowski@...sung.com>
Jaehoon Chung <jh80.chung@...sung.com>
Around MIPI/DRM:
Inki Dae <inki.dae@...sung.com>
Seung-Woo Kim <sw0312.kim@...sung.com>
Andrzej Hajda <a.hajda@...sung.com>
and:
linux-samsung-soc@...r.kernel.org
>
> diff --git a/drivers/phy/samsung/Kconfig b/drivers/phy/samsung/Kconfig
> index e20d2fcc9fe7..e80d40d1278c 100644
> --- a/drivers/phy/samsung/Kconfig
> +++ b/drivers/phy/samsung/Kconfig
> @@ -103,3 +103,12 @@ config PHY_EXYNOS5250_SATA
> Exynos5250 based SoCs.This SerDes/Phy supports SATA 1.5 Gb/s,
> SATA 3.0 Gb/s, SATA 6.0 Gb/s speeds. It supports one SATA host
> port to accept one SATA device.
> +
> +config PHY_SEC_DSIM_DPHY
SEC is not a codename of a project/product/silicon. It is a company
name, so basically you called this "PHY_SAMSUNG_DSIM_DPHY" which is too
generic.
> + tristate "Samsung SEC MIPI DSIM DPHY driver"
What is Samsung SEC? Either Samsung or SEC, not both.
> + depends on OF && HAS_IOMEM
> + select GENERIC_PHY
> + select REGMAP_MMIO
> + help
> + Enable this to add support for the SEC MIPI DSIM DPHY as found
> + on NXP's i.MX8M Mini and Nano family of SOCs.
> diff --git a/drivers/phy/samsung/Makefile b/drivers/phy/samsung/Makefile
> index 3959100fe8a2..4d46c7ec0072 100644
> --- a/drivers/phy/samsung/Makefile
> +++ b/drivers/phy/samsung/Makefile
> @@ -11,3 +11,4 @@ phy-exynos-usb2-$(CONFIG_PHY_EXYNOS5250_USB2) += phy-exynos5250-usb2.o
> phy-exynos-usb2-$(CONFIG_PHY_S5PV210_USB2) += phy-s5pv210-usb2.o
> obj-$(CONFIG_PHY_EXYNOS5_USBDRD) += phy-exynos5-usbdrd.o
> obj-$(CONFIG_PHY_EXYNOS5250_SATA) += phy-exynos5250-sata.o
> +obj-$(CONFIG_PHY_SEC_DSIM_DPHY) += phy-sec-dsim-dphy.o
> diff --git a/drivers/phy/samsung/phy-sec-dsim-dphy.c b/drivers/phy/samsung/phy-sec-dsim-dphy.c
> new file mode 100644
> index 000000000000..31de4a774b5f
> --- /dev/null
> +++ b/drivers/phy/samsung/phy-sec-dsim-dphy.c
> @@ -0,0 +1,236 @@
> +// SPDX-License-Identifier: GPL-2.0+
> +/*
> + * Copyright (C) 2018 NXP
> + * Copyright (C) 2021 Amarula Solutions(India)
> + */
> +
> +#include <linux/clk.h>
> +#include <linux/clk-provider.h>
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/of_platform.h>
> +#include <linux/phy/phy.h>
> +#include <linux/regmap.h>
> +
> +#define DSI_PHYCTRL_B1 0x00
> +#define DSI_PHYCTRL_B2 0x04
> +#define DSI_PHYCTRL_M1 0x08
> +#define DSI_PHYCTRL_M2 0x0c
> +#define DSI_PHYTIMING 0x10
> +#define DSI_PHYTIMING1 0x14
> +#define DSI_PHYTIMING2 0x18
> +
> +/* phytiming */
> +#define M_TLPXCTL_MASK GENMASK(15, 8)
What is the "M_" prefix for?
> +#define M_TLPXCTL(x) FIELD_PREP(M_TLPXCTL_MASK, (x))
> +#define M_THSEXITCTL_MASK GENMASK(7, 0)
> +#define M_THSEXITCTL(x) FIELD_PREP(M_THSEXITCTL_MASK, (x))
> +
> +/* phytiming1 */
> +#define M_TCLKPRPRCTL_MASK GENMASK(31, 24)
> +#define M_TCLKPRPRCTL(x) FIELD_PREP(M_TCLKPRPRCTL_MASK, (x))
> +#define M_TCLKZEROCTL_MASK GENMASK(23, 16)
> +#define M_TCLKZEROCTL(x) FIELD_PREP(M_TCLKZEROCTL_MASK, (x))
> +#define M_TCLKPOSTCTL_MASK GENMASK(15, 8)
> +#define M_TCLKPOSTCTL(x) FIELD_PREP(M_TCLKPOSTCTL_MASK, (x))
> +#define M_TCLKTRAILCTL_MASK GENMASK(7, 0)
> +#define M_TCLKTRAILCTL(x) FIELD_PREP(M_TCLKTRAILCTL_MASK, (x))
> +
> +/* phytiming2 */
> +#define M_THSPRPRCTL_MASK GENMASK(23, 16)
> +#define M_THSPRPRCTL(x) FIELD_PREP(M_THSPRPRCTL_MASK, (x))
> +#define M_THSZEROCTL_MASK GENMASK(15, 8)
> +#define M_THSZEROCTL(x) FIELD_PREP(M_THSZEROCTL_MASK, (x))
> +#define M_THSTRAILCTL_MASK GENMASK(7, 0)
> +#define M_THSTRAILCTL(x) FIELD_PREP(M_THSTRAILCTL_MASK, (x))
> +
> +struct dsim_dphy_plat_data {
Remove the m_ prefix. Please document all fields.
> + unsigned int m_tlpxctl;
> + unsigned int m_thsexitctl;
> + unsigned int m_tclkprprctl;
> + unsigned int m_tclkzeroctl;
> + unsigned int m_tclkpostctl;
> + unsigned int m_tclktrailctl;
> + unsigned int m_thsprprctl;
> + unsigned int m_thszeroctl;
> + unsigned int m_thstrailctl;
> +};
> +
> +struct dsim_dphy {
> + struct regmap *regmap;
> + struct clk *phy_ref_clk;
> + const struct dsim_dphy_plat_data *pdata;
> +};
> +
> +static const struct regmap_config dsim_dphy_regmap_config = {
> + .reg_bits = 8,
> + .val_bits = 32,
> + .reg_stride = 4,
> + .max_register = DSI_PHYTIMING2,
> + .name = "mipi-dphy",
> +};
> +
> +static int dsim_dphy_init(struct phy *phy)
> +{
> + struct dsim_dphy *dphy = phy_get_drvdata(phy);
> + const struct dsim_dphy_plat_data *pdata = dphy->pdata;
> + u32 reg;
> +
> + /* phytiming */
> + regmap_read(dphy->regmap, DSI_PHYTIMING, ®);
> +
> + reg &= ~M_TLPXCTL_MASK;
> + reg |= M_TLPXCTL(pdata->m_tlpxctl);
> + reg &= ~M_THSEXITCTL_MASK;
> + reg |= M_THSEXITCTL(pdata->m_thsexitctl);
> + regmap_write(dphy->regmap, DSI_PHYTIMING, reg);
> +
> + /* phytiming1 */
> + regmap_read(dphy->regmap, DSI_PHYTIMING1, ®);
> +
> + reg &= ~M_TCLKPRPRCTL_MASK;
> + reg |= M_TCLKPRPRCTL(pdata->m_tclkprprctl);
> + reg &= ~M_TCLKZEROCTL_MASK;
> + reg |= M_TCLKZEROCTL(pdata->m_tclkzeroctl);
> + reg &= ~M_TCLKPOSTCTL_MASK;
> + reg |= M_TCLKPOSTCTL(pdata->m_tclkpostctl);
> + reg &= ~M_TCLKTRAILCTL_MASK;
> + reg |= M_TCLKTRAILCTL(pdata->m_tclktrailctl);
> + regmap_write(dphy->regmap, DSI_PHYTIMING1, reg);
> +
> + /* phytiming2 */
> + regmap_read(dphy->regmap, DSI_PHYTIMING2, ®);
> +
> + reg &= ~M_THSPRPRCTL_MASK;
> + reg |= M_THSPRPRCTL(pdata->m_thsprprctl);
> + reg &= ~M_THSZEROCTL_MASK;
> + reg |= M_THSZEROCTL(pdata->m_thszeroctl);
> + reg &= ~M_THSTRAILCTL_MASK;
> + reg |= M_THSTRAILCTL(pdata->m_thstrailctl);
> + regmap_write(dphy->regmap, DSI_PHYTIMING2, reg);
> +
> + return 0;
> +}
> +
> +static int dsim_dphy_exit(struct phy *phy)
> +{
> + return 0;
> +}
> +
> +static int dsim_dphy_power_on(struct phy *phy)
> +{
> + struct dsim_dphy *dphy = phy_get_drvdata(phy);
> + int ret;
> +
> + ret = clk_prepare_enable(dphy->phy_ref_clk);
> + if (ret < 0)
> + return ret;
> +
> + return ret;
> +}
> +
> +static int dsim_dphy_power_off(struct phy *phy)
> +{
> + struct dsim_dphy *dphy = phy_get_drvdata(phy);
> +
> + clk_disable_unprepare(dphy->phy_ref_clk);
> +
> + return 0;
> +}
> +
> +static const struct phy_ops dsim_dphy_phy_ops = {
> + .init = dsim_dphy_init,
> + .exit = dsim_dphy_exit,
> + .power_on = dsim_dphy_power_on,
> + .power_off = dsim_dphy_power_off,
> + .owner = THIS_MODULE,
> +};
> +
> +static const struct dsim_dphy_plat_data imx8mm_dphy_plat_data = {
> + /* phytiming */
> + .m_tlpxctl = 0x06,
> + .m_thsexitctl = 0x0b,
> + /* phytiming1 */
> + .m_tclkprprctl = 0x07,
> + .m_tclkzeroctl = 0x26,
> + .m_tclkpostctl = 0x0d,
> + .m_tclktrailctl = 0x08,
> + /* phytimings2 */
> + .m_thsprprctl = 0x08,
> + .m_thszeroctl = 0x0d,
> + .m_thstrailctl = 0x0b,
> +};
> +
> +static const struct of_device_id dsim_dphy_of_match[] = {
> + {
> + .compatible = "fsl,imx8mm-sec-dsim-dphy",
> + .data = &imx8mm_dphy_plat_data,
> + },
> + { /* sentinel */ },
> +};
> +MODULE_DEVICE_TABLE(of, dsim_dphy_of_match);
> +
> +static int dsim_dphy_probe(struct platform_device *pdev)
> +{
> + struct device *dev = &pdev->dev;
> + struct device_node *np = dev->of_node;
> + struct phy_provider *phy_provider;
> + struct dsim_dphy *dphy;
> + struct phy *phy;
> + void __iomem *base;
> +
> + if (!np)
> + return -ENODEV;
How can this driver bind without 'np'? How is it possible?
> +
> + dphy = devm_kzalloc(dev, sizeof(*dphy), GFP_KERNEL);
> + if (!dphy)
> + return -ENOMEM;
> +
> + dphy->pdata = of_device_get_match_data(&pdev->dev);
> + if (!dphy->pdata)
> + return -EINVAL;
> +
> + base = devm_platform_ioremap_resource(pdev, 0);
> + if (IS_ERR(base))
> + return PTR_ERR(base);
> +
> + dphy->regmap = devm_regmap_init_mmio(&pdev->dev, base,
> + &dsim_dphy_regmap_config);
> + if (IS_ERR(dphy->regmap)) {
> + dev_err(dev, "failed create the DPHY regmap\n");
> + return PTR_ERR(dphy->regmap);
> + }
> +
> + dphy->phy_ref_clk = devm_clk_get(&pdev->dev, "phy_ref");
> + if (IS_ERR(dphy->phy_ref_clk)) {
> + dev_err(dev, "failed to get phy_ref clock\n");
> + return PTR_ERR(dphy->phy_ref_clk);
> + }
> +
> + dev_set_drvdata(dev, dphy);
> +
> + phy = devm_phy_create(dev, np, &dsim_dphy_phy_ops);
> + if (IS_ERR(phy)) {
> + dev_err(dev, "failed to create phy %ld\n", PTR_ERR(phy));
> + return PTR_ERR(phy);
> + }
> + phy_set_drvdata(phy, dphy);
> +
> + phy_provider = devm_of_phy_provider_register(dev, of_phy_simple_xlate);
> +
> + return PTR_ERR_OR_ZERO(phy_provider);
> +}
> +
> +static struct platform_driver dsim_dphy_driver = {
> + .probe = dsim_dphy_probe,
> + .driver = {
> + .name = "sec-dsim-dphy",
"sec" it's too generic, what if next time we have another one from
Samsung? sec2? Please choose a name matching the actual component.
> + .of_match_table = dsim_dphy_of_match,
> + }
> +};
> +module_platform_driver(dsim_dphy_driver);
> +
> +MODULE_AUTHOR("Jagan Teki <jagan@...rulasolutions.com>");
> +MODULE_DESCRIPTION("Samsung SEC MIPI DSIM DPHY driver");> +MODULE_LICENSE("GPL");
>
Best regards,
Krzysztof
Powered by blists - more mailing lists