Re: [PATCH v2 2/2] pwm: add T-HEAD PWM driver

From: Uwe Kleine-König
Date: Wed Oct 04 2023 - 10:01:58 EST


On Wed, Oct 04, 2023 at 05:27:31PM +0800, Jisheng Zhang wrote:
> T-HEAD SoCs such as the TH1520 contain a PWM controller used
> to control the LCD backlight, fan and so on. Add driver for it.
>
> Signed-off-by: Jisheng Zhang <jszhang@xxxxxxxxxx>
> ---
> MAINTAINERS | 1 +
> drivers/pwm/Kconfig | 11 ++
> drivers/pwm/Makefile | 1 +
> drivers/pwm/pwm-thead.c | 274 ++++++++++++++++++++++++++++++++++++++++
> 4 files changed, 287 insertions(+)
> create mode 100644 drivers/pwm/pwm-thead.c
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index d55e40060c46..86cf0926dbfc 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -18482,6 +18482,7 @@ L: linux-riscv@xxxxxxxxxxxxxxxxxxx
> S: Maintained
> F: arch/riscv/boot/dts/thead/
> F: drivers/usb/dwc3/dwc3-thead.c
> +F: drivers/pwm/pwm-thead.c
>
> RNBD BLOCK DRIVERS
> M: Md. Haris Iqbal <haris.iqbal@xxxxxxxxx>
> diff --git a/drivers/pwm/Kconfig b/drivers/pwm/Kconfig
> index 8ebcddf91f7b..428fa365a19a 100644
> --- a/drivers/pwm/Kconfig
> +++ b/drivers/pwm/Kconfig
> @@ -637,6 +637,17 @@ config PWM_TEGRA
> To compile this driver as a module, choose M here: the module
> will be called pwm-tegra.
>
> +config PWM_THEAD
> + tristate "T-HEAD PWM support"
> + depends on ARCH_THEAD || COMPILE_TEST
> + depends on HAS_IOMEM
> + help
> + Generic PWM framework driver for the PWFM controller found on THEAD
> + SoCs.
> +
> + To compile this driver as a module, choose M here: the module
> + will be called pwm-thead.
> +
> config PWM_TIECAP
> tristate "ECAP PWM support"
> depends on ARCH_OMAP2PLUS || ARCH_DAVINCI_DA8XX || ARCH_KEYSTONE || ARCH_K3 || COMPILE_TEST
> diff --git a/drivers/pwm/Makefile b/drivers/pwm/Makefile
> index c822389c2a24..4c317e6316e8 100644
> --- a/drivers/pwm/Makefile
> +++ b/drivers/pwm/Makefile
> @@ -59,6 +59,7 @@ obj-$(CONFIG_PWM_STMPE) += pwm-stmpe.o
> obj-$(CONFIG_PWM_SUN4I) += pwm-sun4i.o
> obj-$(CONFIG_PWM_SUNPLUS) += pwm-sunplus.o
> obj-$(CONFIG_PWM_TEGRA) += pwm-tegra.o
> +obj-$(CONFIG_PWM_THEAD) += pwm-thead.o
> obj-$(CONFIG_PWM_TIECAP) += pwm-tiecap.o
> obj-$(CONFIG_PWM_TIEHRPWM) += pwm-tiehrpwm.o
> obj-$(CONFIG_PWM_TWL) += pwm-twl.o
> diff --git a/drivers/pwm/pwm-thead.c b/drivers/pwm/pwm-thead.c
> new file mode 100644
> index 000000000000..ba1e3a4f1027
> --- /dev/null
> +++ b/drivers/pwm/pwm-thead.c
> @@ -0,0 +1,274 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * T-HEAD PWM driver
> + *
> + * Copyright (C) 2021 Alibaba Group Holding Limited.
> + * Copyright (C) 2023 Jisheng Zhang <jszhang@xxxxxxxxxx>
> + *
> + * Limitations:
> + * - The THEAD_PWM_START bit is only effective when 0 -> 1, which is used to
> + * start the channel, 1 -> 0 doesn't change anything. so 0 % duty cycle is
> + * used to "disable" the channel.
> + * - The PWM_CFG_UPDATE atomically updates and only updates period and duty.
> + * - To update period and duty, PWM_CFG_UPDATE needs to go through 0 -> 1 step,
> + * I.E if PWM_CFG_UPDATE is already 1, it's necessary to clear it to 0
> + * beforehand.
> + * - Polarity can only be changed if never started before.
> + */
> +
> +#include <linux/bitfield.h>
> +#include <linux/bitops.h>
> +#include <linux/clk.h>
> +#include <linux/delay.h>
> +#include <linux/err.h>
> +#include <linux/io.h>
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/of_device.h>

Looking at 0a41b0c5d97a3758ad102cec469aaa79c7d406b7 I think you want
linux/mod_devicetable.h here instead of of_device.h.

> +#include <linux/platform_device.h>
> +#include <linux/pm_runtime.h>
> +#include <linux/pwm.h>
> +#include <linux/slab.h>
> +
> +#define THEAD_PWM_MAX_NUM 6
> +#define THEAD_PWM_MAX_PERIOD GENMASK(31, 0)
> +#define THEAD_PWM_MAX_DUTY GENMASK(31, 0)
> +
> +#define THEAD_PWM_CHN_BASE(n) ((n) * 0x20)
> +#define THEAD_PWM_CTRL(n) (THEAD_PWM_CHN_BASE(n) + 0x00)
> +#define THEAD_PWM_RPT(n) (THEAD_PWM_CHN_BASE(n) + 0x04)
> +#define THEAD_PWM_PER(n) (THEAD_PWM_CHN_BASE(n) + 0x08)
> +#define THEAD_PWM_FP(n) (THEAD_PWM_CHN_BASE(n) + 0x0c)
> +#define THEAD_PWM_STATUS(n) (THEAD_PWM_CHN_BASE(n) + 0x10)
> +
> +/* bit definition PWM_CTRL */
> +#define THEAD_PWM_START BIT(0)

This is a bit in THEAD_PWM_CTRL(n), so I'd propose THEAD_PWM_CTRL_START
as a name.

> +#define THEAD_PWM_SOFT_RST BIT(1)
> +#define THEAD_PWM_CFG_UPDATE BIT(2)
> +#define THEAD_PWM_INT_EN BIT(3)
> +#define THEAD_PWM_MODE_MASK GENMASK(5, 4)

I'd drop "_MASK" here (and add _CTRL).

> +#define THEAD_PWM_ONE_SHOT_MODE FIELD_PREP(THEAD_PWM_MODE_MASK, 1)

This is unused

> +#define THEAD_PWM_CONTINUOUS_MODE FIELD_PREP(THEAD_PWM_MODE_MASK, 2)

THEAD_PWM_CTRL_MODE_CONTINUOUS ?

> +#define THEAD_PWM_EVTRIG_MASK GENMASK(7, 6)
> +#define THEAD_PWM_FPOUT BIT(8)
> +#define THEAD_PWM_INFACTOUT BIT(9)
> +
> +struct thead_pwm_chip {
> + struct pwm_chip chip;
> + void __iomem *mmio_base;
> + struct clk *clk;
> + bool ever_started;
> +};
> +
> +static inline struct thead_pwm_chip *thead_pwm_from_chip(struct pwm_chip *chip)
> +{
> + return container_of(chip, struct thead_pwm_chip, chip);
> +}
> +
> +static int thead_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> + const struct pwm_state *state)
> +{
> + struct thead_pwm_chip *priv = thead_pwm_from_chip(chip);
> + u64 period_cycle, duty_cycle, rate;
> + u32 val;
> +
> + /* if ever started, can't change the polarity */
> + if (priv->ever_started && state->polarity != pwm->state.polarity)
> + return -EINVAL;
> +
> + if (!state->enabled) {
> + if (pwm->state.enabled) {
> + val = readl(priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> + val &= ~THEAD_PWM_CFG_UPDATE;
> + writel(val, priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> +
> + writel(0, priv->mmio_base + THEAD_PWM_FP(pwm->hwpwm));
> +
> + val |= THEAD_PWM_CFG_UPDATE;
> + writel(val, priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> + }
> + return 0;
> + }
> +
> + if (!pwm->state.enabled)
> + pm_runtime_get_sync(chip->dev);

pm_runtime_get_sync() returns an int that you shouldn't ignore.

> + val = readl(priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> + val &= ~THEAD_PWM_CFG_UPDATE;
> +
> + if (state->polarity == PWM_POLARITY_INVERSED)
> + val &= ~THEAD_PWM_FPOUT;
> + else
> + val |= THEAD_PWM_FPOUT;

What happens here if the bootloader already touched that flag? Or the
driver is reloaded/rebound?

> + writel(val, priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> +
> + rate = clk_get_rate(priv->clk);
> + /*
> + * The following calculations might overflow if clk is bigger
> + * than 1 GHz. In practise it's 24MHz, so this limitation
> + * is only theoretic.
> + */
> + if (rate > (u64)NSEC_PER_SEC)

this cast isn't needed.

> + return -EINVAL;
> +
> + period_cycle = mul_u64_u64_div_u64(rate, state->period, NSEC_PER_SEC);
> + if (period_cycle > THEAD_PWM_MAX_PERIOD)
> + period_cycle = THEAD_PWM_MAX_PERIOD;
> + /*
> + * With limitation above we have period_cycle <= THEAD_PWM_MAX_PERIOD,
> + * so this cannot overflow.
> + */
> + writel((u32)period_cycle, priv->mmio_base + THEAD_PWM_PER(pwm->hwpwm));

This cast can also be dropped.

> +
> + duty_cycle = mul_u64_u64_div_u64(rate, state->duty_cycle, NSEC_PER_SEC);
> + if (duty_cycle > THEAD_PWM_MAX_DUTY)
> + duty_cycle = THEAD_PWM_MAX_DUTY;
> + /*
> + * With limitation above we have duty_cycle <= THEAD_PWM_MAX_PERIOD,
> + * so this cannot overflow.
> + */
> + writel((u32)duty_cycle, priv->mmio_base + THEAD_PWM_FP(pwm->hwpwm));

...

> +
> + val |= THEAD_PWM_CFG_UPDATE;
> + writel(val, priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> +
> + if (!pwm->state.enabled) {
> + val |= THEAD_PWM_START;
> + writel(val, priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> + priv->ever_started = true;
> + }

Further above you conditionally call pm_runtime_get_sync(), there should
be a matching pm_runtime_put().

> +
> + return 0;
> +}
> +
> +static int thead_pwm_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
> + struct pwm_state *state)
> +{
> + struct thead_pwm_chip *priv = thead_pwm_from_chip(chip);
> + u64 rate = clk_get_rate(priv->clk);
> + u32 val;
> +
> + pm_runtime_get_sync(chip->dev);
> +
> + val = readl(priv->mmio_base + THEAD_PWM_CTRL(pwm->hwpwm));
> + state->enabled = !!(val & THEAD_PWM_START);
> + if (val & THEAD_PWM_FPOUT)
> + state->polarity = PWM_POLARITY_NORMAL;
> + else
> + state->polarity = PWM_POLARITY_INVERSED;
> +
> + val = readl(priv->mmio_base + THEAD_PWM_PER(pwm->hwpwm));
> + /*
> + * val 32 bits, multiply NSEC_PER_SEC, won't overflow.
> + */
> + state->period = DIV64_U64_ROUND_UP((u64)val * NSEC_PER_SEC, rate);
> +
> + val = readl(priv->mmio_base + THEAD_PWM_FP(pwm->hwpwm));
> + /*
> + * val 32 bits, multiply NSEC_PER_SEC, won't overflow.
> + */
> + state->duty_cycle = DIV64_U64_ROUND_UP((u64)val * NSEC_PER_SEC, rate);
> +
> + pm_runtime_put_sync(chip->dev);
> +
> + return 0;
> +}
> +
> +static const struct pwm_ops thead_pwm_ops = {
> + .apply = thead_pwm_apply,
> + .get_state = thead_pwm_get_state,
> + .owner = THIS_MODULE,
> +};
> +
> +static int __maybe_unused thead_pwm_runtime_suspend(struct device *dev)
> +{
> + struct thead_pwm_chip *priv = dev_get_drvdata(dev);
> +
> + clk_disable_unprepare(priv->clk);
> +
> + return 0;
> +}
> +
> +static int __maybe_unused thead_pwm_runtime_resume(struct device *dev)
> +{
> + struct thead_pwm_chip *priv = dev_get_drvdata(dev);
> + int ret;
> +
> + ret = clk_prepare_enable(priv->clk);
> + if (ret) {
> + dev_err(dev, "failed to enable pwm clock(%d)\n", ret);
> + return ret;
> + }
> +
> + return 0;

This can be simplified to:

ret = clk_prepare_enable(priv->clk);
if (ret)
dev_err(dev, "failed to enable pwm clock(%pe)\n", ERR_PTR(ret));

return ret;

(Note that I used %pe instead of %d which is nicer for error codes)

Having said that, I'm unsure if emitting an error message is useful.
Maybe the core emits a message anyhow?

> +}
> +
> +static int thead_pwm_probe(struct platform_device *pdev)
> +{
> + u32 val = THEAD_PWM_INFACTOUT | THEAD_PWM_FPOUT | THEAD_PWM_CONTINUOUS_MODE;
> + struct thead_pwm_chip *priv;
> + int ret, i;
> +
> + priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
> + if (!priv)
> + return -ENOMEM;
> +
> + platform_set_drvdata(pdev, priv);
> +
> + priv->mmio_base = devm_platform_ioremap_resource(pdev, 0);
> + if (IS_ERR(priv->mmio_base))
> + return PTR_ERR(priv->mmio_base);
> +
> + priv->clk = devm_clk_get_enabled(&pdev->dev, NULL);
> + if (IS_ERR(priv->clk))
> + return PTR_ERR(priv->clk);
> +
> + priv->chip.ops = &thead_pwm_ops;
> + priv->chip.dev = &pdev->dev;
> + priv->chip.npwm = THEAD_PWM_MAX_NUM;
> +
> + /* set normal polarity and other proper bits for all channels */

Please don't. You're supposed to keep the state of the hardware in
probe. Consider a bootloader that enabled the backlight of an LCD that
shows a splash screen.

> + for (i = 0; i < priv->chip.npwm; i++)
> + writel(val, priv->mmio_base + THEAD_PWM_CTRL(i));
> +
> + ret = devm_pwmchip_add(&pdev->dev, &priv->chip);
> + if (ret)
> + return ret;
> +
> + pm_runtime_enable(&pdev->dev);

devm_pm_runtime_enable() and then drop .remove()

> +
> + return 0;
> +}
> +
> +static void thead_pwm_remove(struct platform_device *pdev)
> +{
> + pm_runtime_disable(&pdev->dev);
> +}
> +
> +static const struct of_device_id thead_pwm_dt_ids[] = {
> + {.compatible = "thead,th1520-pwm",},
> + {/* sentinel */}
> +};
> +MODULE_DEVICE_TABLE(of, thead_pwm_dt_ids);
> +
> +static const struct dev_pm_ops thead_pwm_pm_ops = {
> + SET_RUNTIME_PM_OPS(thead_pwm_runtime_suspend, thead_pwm_runtime_resume, NULL)
> +};
> +
> +static struct platform_driver thead_pwm_driver = {
> + .driver = {
> + .name = "thead-pwm",
> + .of_match_table = thead_pwm_dt_ids,
> + .pm = &thead_pwm_pm_ops,
> + },
> + .probe = thead_pwm_probe,
> + .remove_new = thead_pwm_remove,
> +};
> +module_platform_driver(thead_pwm_driver);

Best regards
Uwe

--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | https://www.pengutronix.de/ |

Attachment: signature.asc
Description: PGP signature