Re: [linux-sunxi] [PATCH v2 4/4] ARM: PWM: add allwinner sun8i pwm support.

[André Przywara <andre.przywara@xxxxxxx>]

Hi,

On 25/02/18 13:53, hao_zhang wrote:
> This patch add allwinner sun8i pwm support.

Again, the subject line is too generic. Mention the R40?

Can you elaborate here a bit? Mention that is used on the R40, but not
other sun8i SoCs, for instance. And mention that this is very different
from the sun4i-pwm device, so justifies a new driver. Possibly mention
some features? And that we for now just implement a subset of them.

> 
> Signed-off-by: hao_zhang <hao5781286@xxxxxxxxx>
> ---
>  drivers/pwm/Kconfig     |  10 ++
>  drivers/pwm/Makefile    |   1 +
>  drivers/pwm/pwm-sun8i.c | 401 ++++++++++++++++++++++++++++++++++++++++++++++++

I am not too happy with this name, but I guess there are no better
alternatives, so it's probably OK to keep it.

>  3 files changed, 412 insertions(+)
>  create mode 100644 drivers/pwm/pwm-sun8i.c
> 
> diff --git a/drivers/pwm/Kconfig b/drivers/pwm/Kconfig
> index 763ee50..7e68d0f 100644
> --- a/drivers/pwm/Kconfig
> +++ b/drivers/pwm/Kconfig
> @@ -444,6 +444,16 @@ config PWM_SUN4I
>  	  To compile this driver as a module, choose M here: the module
>  	  will be called pwm-sun4i.
>  
> +config PWM_SUN8I
> +	tristate "Allwinner PWM SUN8I support"
> +	depends on ARCH_SUNXI || COMPILE_TEST
> +	depends on HAS_IOMEM && COMMON_CLK
> +	help
> +	  Generic PWM framework driver for Allwinner SoCs.

Mmh, not really. So far there is only one SoC using this. Maybe:
	  Driver for the enhanced PWM IP used in some newer Allwinner
	  SoCs.

> +
> +	  To compile this driver as a module, choose M here: the module
> +	  will be called pwm-sun8i.
> +
>  config PWM_TEGRA
>  	tristate "NVIDIA Tegra PWM support"
>  	depends on ARCH_TEGRA
> diff --git a/drivers/pwm/Makefile b/drivers/pwm/Makefile
> index 0258a74..cd6bf40 100644
> --- a/drivers/pwm/Makefile
> +++ b/drivers/pwm/Makefile
> @@ -44,6 +44,7 @@ obj-$(CONFIG_PWM_STM32)		+= pwm-stm32.o
>  obj-$(CONFIG_PWM_STM32_LP)	+= pwm-stm32-lp.o
>  obj-$(CONFIG_PWM_STMPE)		+= pwm-stmpe.o
>  obj-$(CONFIG_PWM_SUN4I)		+= pwm-sun4i.o
> +obj-$(CONFIG_PWM_SUN8I)		+= pwm-sun8i.o
>  obj-$(CONFIG_PWM_TEGRA)		+= pwm-tegra.o
>  obj-$(CONFIG_PWM_TIECAP)	+= pwm-tiecap.o
>  obj-$(CONFIG_PWM_TIEHRPWM)	+= pwm-tiehrpwm.o
> diff --git a/drivers/pwm/pwm-sun8i.c b/drivers/pwm/pwm-sun8i.c
> new file mode 100644
> index 0000000..cf23b0a
> --- /dev/null
> +++ b/drivers/pwm/pwm-sun8i.c
> @@ -0,0 +1,401 @@
> +#include <linux/bitops.h>
> +#include <linux/clk.h>
> +#include <linux/delay.h>
> +#include <linux/err.h>
> +#include <linux/io.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/of_device.h>
> +#include <linux/platform_device.h>
> +#include <linux/pwm.h>
> +#include <linux/slab.h>
> +#include <linux/spinlock.h>
> +#include <linux/time.h>
> +#include <linux/regmap.h>
> +
> +#define PWM_IRQ_ENABLE_REG	0x0000
> +#define PCIE(ch)	BIT(ch)

Can you please align those:
#define PWM_IRQ_ENABLE_REG	0x0000
#define PCIE(ch)		BIT(ch)

And all those below as well? Which means you might want to insert
another tab to cater for those longer symbols.

> +
> +#define PWM_IRQ_STATUS_REG	0x0004
> +#define PIS(ch)	BIT(ch)
> +
> +#define CAPTURE_IRQ_ENABLE_REG	0x0010
> +#define CFIE(ch)	BIT(ch << 1 + 1)
> +#define CRIE(ch)	BIT(ch << 1)
> +
> +#define CAPTURE_IRQ_STATUS_REG	0x0014
> +#define CFIS(ch)	BIT(ch << 1 + 1)
> +#define CRIS(ch)	BIT(ch << 1)
> +
> +#define CLK_CFG_REG(ch)	(0x0020 + (ch >> 1) * 4)
> +#define CLK_SRC	BIT(7)
> +#define CLK_SRC_BYPASS_SEC	BIT(6)
> +#define CLK_SRC_BYPASS_FIR	BIT(5)
> +#define CLK_GATING	BIT(4)
> +#define CLK_DIV_M	GENMASK(3, 0)
> +
> +#define PWM_DZ_CTR_REG(ch)	(0x0030 + (ch >> 1) * 4)
> +#define PWM_DZ_INTV	GENMASK(15, 8)
> +#define PWM_DZ_EN	BIT(0)
> +
> +#define PWM_ENABLE_REG	0x0040
> +#define PWM_EN(ch)	BIT(ch)
> +
> +#define CAPTURE_ENABLE_REG	0x0044
> +#define CAP_EN(ch)	BIT(ch)
> +
> +#define PWM_CTR_REG(ch)	(0x0060 + ch * 0x20)
> +#define PWM_PERIOD_RDY	BIT(11)
> +#define PWM_PUL_START	BIT(10)
> +#define PWM_MODE	BIT(9)
> +#define PWM_ACT_STA	BIT(8)
> +#define PWM_PRESCAL_K	GENMASK(7, 0)
> +
> +#define PWM_PERIOD_REG(ch)	(0x0064 + ch * 0x20)
> +#define PWM_ENTIRE_CYCLE	GENMASK(31, 16)
> +#define PWM_ACT_CYCLE	GENMASK(15, 0)
> +
> +#define PWM_CNT_REG(ch)	(0x0068 + ch * 0x20)
> +#define PWM_CNT_VAL	GENMASK(15, 0)
> +
> +#define CAPTURE_CTR_REG(ch)	(0x006c + ch * 0x20)
> +#define CAPTURE_CRLF	BIT(2)
> +#define CAPTURE_CFLF	BIT(1)
> +#define CAPINV	BIT(0)
> +
> +#define CAPTURE_RISE_REG(ch)	(0x0070 + ch * 0x20)
> +#define CAPTURE_CRLR	GENMASK(15, 0)
> +
> +#define CAPTURE_FALL_REG(ch)	(0x0074 + ch * 0x20)
> +#define CAPTURE_CFLR	GENMASK(15, 0)
> +
> +struct sun8i_pwm_data {
> +	bool has_prescaler_bypass;
> +	bool has_rdy;
> +	unsigned int npwm;
> +};

I believe you don't need this structure. See below.

> +
> +struct sun8i_pwm_chip {
> +	struct pwm_chip chip;
> +	struct clk *clk;
> +	void __iomem *base;
> +	spinlock_t ctrl_lock;
> +	const struct sun8i_pwm_data *data;
> +	struct regmap *regmap;
> +};
> +
> +static const u16 div_m_table[] = {
> +	1,
> +	2,
> +	4,
> +	8,
> +	16,
> +	32,
> +	64,
> +	128,
> +	256
> +};

That looks very much like: "1U << x" to me.

> +
> +static inline struct sun8i_pwm_chip *to_sun8i_pwm_chip(struct pwm_chip *chip)

No need for "inline", the compiler knows better. static is enough.

> +{
> +	return container_of(chip, struct sun8i_pwm_chip, chip);
> +}
> +
> +static u32 sun8i_pwm_read(struct sun8i_pwm_chip *sun8i_pwm,
> +		unsigned long offset)

Can you please align those continuation lines properly? The first
character in the new line should be aligned to the first character of
the first argument. Use tabs first, then fill up with spaces:

static u32 sun8i_pwm_read(struct sun8i_pwm_chip *sun8i_pwm,
			  unsigned long offset)

This applies to the rest of the file as well.

> +{
> +	u32 val;
> +
> +	regmap_read(sun8i_pwm->regmap, offset, &val);
> +
> +	return val;
> +}
> +
> +static inline void sun8i_pwm_set_bit(struct sun8i_pwm_chip *sun8i_pwm,

no inline (for those below as well)

> +		unsigned long reg, u32 bit)
> +{
> +	regmap_update_bits(sun8i_pwm->regmap, reg, bit, bit);
> +}
> +
> +static inline void sun8i_pwm_clear_bit(struct sun8i_pwm_chip *sun8i_pwm,
> +		unsigned long reg, u32 bit)
> +{
> +	regmap_update_bits(sun8i_pwm->regmap, reg, bit, 0);
> +}
> +
> +static inline void sun8i_pwm_set_value(struct sun8i_pwm_chip *sun8i_pwm,
> +		unsigned long reg, u32 mask, u32 val)
> +{
> +	regmap_update_bits(sun8i_pwm->regmap, reg, mask, val);
> +}
> +
> +static void sun8i_pwm_set_polarity(struct sun8i_pwm_chip *chip, u32 ch,
> +		enum pwm_polarity polarity)
> +{
> +	if (polarity == PWM_POLARITY_NORMAL)
> +		sun8i_pwm_set_bit(chip, PWM_CTR_REG(ch), PWM_ACT_STA);
> +	else
> +		sun8i_pwm_clear_bit(chip, PWM_CTR_REG(ch), PWM_ACT_STA);
> +}
> +
> +static int sun8i_pwm_config(struct sun8i_pwm_chip *sun8i_pwm, u8 ch,
> +		struct pwm_state *state)
> +{
> +	u64 clk_rate, clk_div, val;
> +	u16 prescaler = 0;
> +	u8 id = 0;
> +
> +	clk_rate = clk_get_rate(sun8i_pwm->clk);
> +
> +	if (clk_rate == 24000000)
> +		sun8i_pwm_clear_bit(sun8i_pwm, CLK_CFG_REG(ch), CLK_SRC);
> +	else
> +		sun8i_pwm_set_bit(sun8i_pwm, CLK_CFG_REG(ch), CLK_SRC);

This hardcoded 24MHz looks slightly dodgy and should be replaced with
some proper code to select the best matching clock, out of a number of
them given in the DT (see the DT binding mail).
Without thinking too deeply about it, I guess we try which clocks gives
the least error for the given configuration. The frequency alone might
be a good first guide.
If you can't be bothered with coding this, we might just go ahead with
the first specified clock and always use this, for now.

> +
> +	if (sun8i_pwm->data->has_prescaler_bypass) {

What is this about? I think this is a misunderstanding:
The bypass bits allows to directly pass on the input clock to the output
pin, without any actual PWM properties. So if one channel is (by
chance?) configured for a 50% duty cycle and the same frequency as one
of the input clocks, you might want to use the bypass bit instead. But I
don't see many advantages in doing so, so I guess we can ignore it in a
generic PWM driver.
Anyway using some hardcoded value from the "data" structure looks just
wrong to me. I guess you can just remove this, along with the
has_prescaler_bypass variable from the sun8i_pwm_data structure.

> +		/* pwm output bypass */
> +		if (ch % 2)
> +			sun8i_pwm_set_bit(sun8i_pwm, CLK_CFG_REG(ch),
> +					CLK_SRC_BYPASS_FIR);
> +		else
> +			sun8i_pwm_set_bit(sun8i_pwm, CLK_CFG_REG(ch),
> +					CLK_SRC_BYPASS_SEC);
> +		return 0;
> +	}
> +
> +	val = state->period * clk_rate;
> +	do_div(val, NSEC_PER_SEC);
> +	if (val < 1) {
> +		dev_err(sun8i_pwm->chip.dev,
> +				"Period expects a larger value\n");

Alignment.
And you might want to hook in here to select a higher frequency input clock.

> +		return -EINVAL;
> +	}
> +
> +	/* calculate and set prescalar, div table, pwn entrie cycle */

                             prescaler             PWM entire

though I believe this "entire cycle" term is an Allwinner invention.
Wouldn't period be a better term here, also matching the framework?

> +	clk_div = val;
> +
> +	while (clk_div > 65535) {
> +		prescaler++;
> +		clk_div = val;
> +		do_div(clk_div, prescaler + 1);
> +		do_div(clk_div, div_m_table[id]);

			        1U << id

> +
> +		if (prescaler == 255) {
> +			prescaler = 0;
> +			id++;
> +			if (id == 9)
> +				return -EINVAL;
> +		}
> +	}
> +
> +	sun8i_pwm_set_value(sun8i_pwm, PWM_PERIOD_REG(ch),
> +			PWM_ENTIRE_CYCLE, clk_div << 16);
> +	sun8i_pwm_set_value(sun8i_pwm, PWM_CTR_REG(ch),
> +			PWM_PRESCAL_K, prescaler << 0);
> +	sun8i_pwm_set_value(sun8i_pwm, CLK_CFG_REG(ch),
> +			CLK_DIV_M, id << 0);
> +
> +	/* set duty cycle */
> +	val = (prescaler + 1) * div_m_table[id] * clk_div;

				(1U << id)

You might want to check for the range, though.

> +	val = state->period;
> +	do_div(val, clk_div);
> +	clk_div = state->duty_cycle;
> +	do_div(clk_div, val);
> +
> +	sun8i_pwm_set_value(sun8i_pwm, PWM_PERIOD_REG(ch),
> +			PWM_ACT_CYCLE, clk_div << 0);
> +
> +	return 0;
> +}
> +
> +static int sun8i_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> +		struct pwm_state *state)
> +{
> +	int ret;
> +	struct sun8i_pwm_chip *sun8i_pwm = to_sun8i_pwm_chip(chip);
> +	struct pwm_state cstate;
> +
> +	pwm_get_state(pwm, &cstate);
> +	if (!cstate.enabled) {
> +		ret = clk_prepare_enable(sun8i_pwm->clk);
> +		if (ret) {
> +			dev_err(chip->dev, "Failed to enable PWM clock\n");
> +			return ret;
> +		}
> +	}
> +
> +	spin_lock(&sun8i_pwm->ctrl_lock);
> +
> +	if ((cstate.period != state->period) ||
> +			(cstate.duty_cycle != state->duty_cycle)) {
> +		ret = sun8i_pwm_config(sun8i_pwm, pwm->hwpwm, state);
> +		if (ret) {
> +			spin_unlock(&sun8i_pwm->ctrl_lock);
> +			dev_err(chip->dev, "Failed to config PWM\n");
> +			return ret;
> +		}
> +	}
> +
> +	if (state->polarity != cstate.polarity)
> +		sun8i_pwm_set_polarity(sun8i_pwm, pwm->hwpwm, state->polarity);
> +
> +	if (state->enabled) {
> +		sun8i_pwm_set_bit(sun8i_pwm,
> +				CLK_CFG_REG(pwm->hwpwm), CLK_GATING);
> +
> +		sun8i_pwm_set_bit(sun8i_pwm,
> +				PWM_ENABLE_REG, PWM_EN(pwm->hwpwm));
> +	} else {
> +		sun8i_pwm_clear_bit(sun8i_pwm,
> +				CLK_CFG_REG(pwm->hwpwm), CLK_GATING);
> +
> +		sun8i_pwm_clear_bit(sun8i_pwm,
> +				PWM_ENABLE_REG, PWM_EN(pwm->hwpwm));
> +	}
> +
> +	spin_unlock(&sun8i_pwm->ctrl_lock);
> +
> +	return 0;
> +}
> +
> +static void sun8i_pwm_get_state(struct pwm_chip *chip, struct pwm_device *pwm,
> +		struct pwm_state *state)
> +{
> +	struct sun8i_pwm_chip *sun8i_pwm = to_sun8i_pwm_chip(chip);
> +	u64 clk_rate, tmp;
> +	u32 val;
> +	u16 clk_div, act_cycle;
> +	u8 prescal, id;

You might want to add a channel variable to increase readability:
	int channel = pwm->hwpwm;

> +
> +	clk_rate = clk_get_rate(sun8i_pwm->clk);
> +
> +	val = sun8i_pwm_read(sun8i_pwm, PWM_CTR_REG(pwm->hwpwm));
> +	if (PWM_ACT_STA & val)
> +		state->polarity = PWM_POLARITY_NORMAL;
> +	else
> +		state->polarity = PWM_POLARITY_INVERSED;
> +
> +	prescal = PWM_PRESCAL_K & val;
> +
> +	val = sun8i_pwm_read(sun8i_pwm, PWM_ENABLE_REG);
> +	if (PWM_EN(pwm->hwpwm) & val)
> +		state->enabled = true;
> +	else
> +		state->enabled = false;
> +
> +	val = sun8i_pwm_read(sun8i_pwm, PWM_PERIOD_REG(pwm->hwpwm));
> +	act_cycle = PWM_ACT_CYCLE & val;
> +	clk_div = val >> 16;
> +
> +	val = sun8i_pwm_read(sun8i_pwm, CLK_CFG_REG(pwm->hwpwm));
> +	id = CLK_DIV_M & val;
> +
> +	tmp = act_cycle * prescal * div_m_table[id] * NSEC_PER_SEC;
> +	state->duty_cycle = DIV_ROUND_CLOSEST_ULL(tmp, clk_rate);
> +	tmp = clk_div * prescal * div_m_table[id] * NSEC_PER_SEC;
> +	state->period = DIV_ROUND_CLOSEST_ULL(tmp, clk_rate);
> +}
> +
> +static const struct regmap_config sun8i_pwm_regmap_config = {
> +	.reg_bits = 32,
> +	.reg_stride = 4,
> +	.val_bits = 32,
> +	.max_register = CAPTURE_FALL_REG(7),
> +};
> +
> +static const struct pwm_ops sun8i_pwm_ops = {
> +	.apply = sun8i_pwm_apply,
> +	.get_state = sun8i_pwm_get_state,
> +	.owner = THIS_MODULE,
> +};
> +
> +static const struct sun8i_pwm_data sun8i_pwm_data_r40 = {
> +	.has_prescaler_bypass = false,

This is not needed (see my comment above).

> +	.has_rdy = true,

And this is not used. Copied from sun4i? Where it interestingly isn't
used either ;-)

> +	.npwm = 8,

I would really love to see this being moved to the DT (see my other mail
to Thierry about the generic property).

This would mean you don't need a SoC specific structure at all.

> +};
> +
> +static const struct of_device_id sun8i_pwm_dt_ids[] = {
> +	{
> +		.compatible = "allwinner,sun8i-r40-pwm",
> +		.data = &sun8i_pwm_data_r40,
> +	},
> +	{},
> +};
> +MODULE_DEVICE_TABLE(of, sun8i_pwm_dt_ids);
> +
> +static int sun8i_pwm_probe(struct platform_device *pdev)
> +{
> +	struct sun8i_pwm_chip *pwm;
> +	struct resource *res;
> +	int ret;
> +	const struct of_device_id *match;
> +
> +	match = of_match_device(sun8i_pwm_dt_ids, &pdev->dev);
> +	if (!match) {
> +		dev_err(&pdev->dev, "Error: No device match found\n");
> +		return -ENODEV;
> +	}
> +
> +	pwm = devm_kzalloc(&pdev->dev, sizeof(*pwm), GFP_KERNEL);
> +	if (!pwm)
> +		return -ENOMEM;
> +
> +	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> +	pwm->base = devm_ioremap_resource(&pdev->dev, res);
> +	if (IS_ERR(pwm->base))
> +		return PTR_ERR(pwm->base);
> +
> +	pwm->regmap = devm_regmap_init_mmio(&pdev->dev, pwm->base,
> +			&sun8i_pwm_regmap_config);
> +	if (IS_ERR(pwm->regmap)) {
> +		dev_err(&pdev->dev, "Failed to create regmap\n");
> +		return PTR_ERR(pwm->regmap);
> +	}
> +
> +	pwm->clk = devm_clk_get(&pdev->dev, NULL);
> +	if (IS_ERR(pwm->clk))
> +		return PTR_ERR(pwm->clk);

This would need to be extended to get multiple clocks.

> +
> +	pwm->data = match->data;
> +	pwm->chip.dev = &pdev->dev;
> +	pwm->chip.ops = &sun8i_pwm_ops;
> +	pwm->chip.base = -1;
> +	pwm->chip.npwm = pwm->data->npwm;

It should be fairly easy to initialise this from some DT property.

That's it for the my first review round. Haven't checked the actual
algorithm and bit assignments yet.
Did you manage to test this?

Cheers,
Andre.

> +	pwm->chip.of_xlate = of_pwm_xlate_with_flags;
> +	pwm->chip.of_pwm_n_cells = 3;
> +
> +	spin_lock_init(&pwm->ctrl_lock);
> +
> +	ret = pwmchip_add(&pwm->chip);
> +	if (ret < 0) {
> +		dev_err(&pdev->dev, "Failed to add PWM chip: %d\n", ret);
> +		return ret;
> +	}
> +
> +	platform_set_drvdata(pdev, pwm);
> +
> +	return 0;
> +}
> +
> +static int sun8i_pwm_remove(struct platform_device *pdev)
> +{
> +	struct sun8i_pwm_chip *pwm = platform_get_drvdata(pdev);
> +
> +	return pwmchip_remove(&pwm->chip);
> +}
> +
> +static struct platform_driver sun8i_pwm_driver = {
> +	.driver = {
> +		.name = "sun8i-pwm",
> +		.of_match_table = sun8i_pwm_dt_ids,
> +	},
> +	.probe = sun8i_pwm_probe,
> +	.remove = sun8i_pwm_remove,
> +};
> +module_platform_driver(sun8i_pwm_driver);
> +
> +MODULE_ALIAS("platform: sun8i-pwm");
> +MODULE_AUTHOR("Hao Zhang <hao5781286@xxxxxxxxx>");
> +MODULE_DESCRIPTION("Allwinner sun8i PWM driver");
> +MODULE_LICENSE("GPL v2");
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html