Re: [PATCH 3/4] pwm: Add Aspeed ast2600 PWM support

[Uwe Kleine-König]

Hello Billy,

On Mon, Apr 12, 2021 at 05:54:56PM +0800, Billy Tsai wrote:
> This patch add the support of PWM controller which can find at aspeed
> ast2600 soc chip. This controller supoorts up to 16 channels.
> 
> Signed-off-by: Billy Tsai <billy_tsai@xxxxxxxxxxxxxx>
> ---
>  drivers/pwm/pwm-aspeed-g6.c | 291 ++++++++++++++++++++++++++++++++++++
>  1 file changed, 291 insertions(+)
>  create mode 100644 drivers/pwm/pwm-aspeed-g6.c
> 
> diff --git a/drivers/pwm/pwm-aspeed-g6.c b/drivers/pwm/pwm-aspeed-g6.c
> new file mode 100644
> index 000000000000..4bb4f97453c6
> --- /dev/null
> +++ b/drivers/pwm/pwm-aspeed-g6.c
> @@ -0,0 +1,291 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * Copyright (C) ASPEED Technology Inc.

Don't you need to add a year here?

> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 or later as
> + * published by the Free Software Foundation.

Hmm, the comment and the SPDX-License-Identifier contradict each other.
The idea of the latter is that the former isn't needed.

> + */

Is there documentation available in the internet for this hardware? If
yes, please mention a link here.

Also describe the hardware here similar to how e.g.
drivers/pwm/pwm-sifive.c does it. Please stick to the same format for
easy grepping.

> +
> +#include <linux/clk.h>
> +#include <linux/errno.h>
> +#include <linux/delay.h>
> +#include <linux/io.h>
> +#include <linux/kernel.h>
> +#include <linux/mfd/syscon.h>
> +#include <linux/module.h>
> +#include <linux/of_platform.h>
> +#include <linux/of_device.h>
> +#include <linux/platform_device.h>
> +#include <linux/sysfs.h>
> +#include <linux/reset.h>
> +#include <linux/regmap.h>
> +#include <linux/pwm.h>
> +/* The channel number of Aspeed pwm controller */
> +#define ASPEED_NR_PWMS 16
> +/* PWM Control Register */
> +#define ASPEED_PWM_CTRL_CH(ch) ((ch * 0x10) + 0x00)

#define ASPEED_PWM_CTRL_CH(ch) (((ch) * 0x10) + 0x00)

> +#define PWM_LOAD_SEL_AS_WDT BIT(19)
> +#define LOAD_SEL_FALLING 0
> +#define LOAD_SEL_RIGING 1
> +#define PWM_DUTY_LOAD_AS_WDT_EN BIT(18)
> +#define PWM_DUTY_SYNC_DIS BIT(17)
> +#define PWM_CLK_ENABLE BIT(16)
> +#define PWM_LEVEL_OUTPUT BIT(15)
> +#define PWM_INVERSE BIT(14)
> +#define PWM_OPEN_DRAIN_EN BIT(13)
> +#define PWM_PIN_EN BIT(12)
> +#define PWM_CLK_DIV_H_SHIFT 8
> +#define PWM_CLK_DIV_H_MASK (0xf << PWM_CLK_DIV_H_SHIFT)
> +#define PWM_CLK_DIV_L_SHIFT 0
> +#define PWM_CLK_DIV_L_MASK (0xff << PWM_CLK_DIV_L_SHIFT)
> +/* PWM Duty Cycle Register */
> +#define ASPEED_PWM_DUTY_CYCLE_CH(ch) ((ch * 0x10) + 0x04)
> +#define PWM_PERIOD_SHIFT (24)
> +#define PWM_PERIOD_MASK (0xff << PWM_PERIOD_SHIFT)
> +#define PWM_POINT_AS_WDT_SHIFT (16)
> +#define PWM_POINT_AS_WDT_MASK (0xff << PWM_POINT_AS_WDT_SHIFT)
> +#define PWM_FALLING_POINT_SHIFT (8)
> +#define PWM_FALLING_POINT_MASK (0xffff << PWM_FALLING_POINT_SHIFT)
> +#define PWM_RISING_POINT_SHIFT (0)
> +#define PWM_RISING_POINT_MASK (0xffff << PWM_RISING_POINT_SHIFT)
> +/* PWM default value */
> +#define DEFAULT_PWM_PERIOD 0xff
> +#define DEFAULT_TARGET_PWM_FREQ 25000
> +#define DEFAULT_DUTY_PT 10
> +#define DEFAULT_WDT_RELOAD_DUTY_PT 16

You could spend a few empty lines to make this better readable. Also
please use a consistent driver-specific prefix for your defines and
consider using the macros from <linux/bitfield.h>. Also defines
for bitfields should contain the register name.

> +struct aspeed_pwm_data {
> +	struct pwm_chip chip;
> +	struct regmap *regmap;
> +	unsigned long clk_freq;
> +	struct reset_control *reset;
> +};
> +/**
> + * struct aspeed_pwm - per-PWM driver data
> + * @freq: cached pwm freq
> + */
> +struct aspeed_pwm {
> +	u32 freq;
> +};

This is actually unused, please drop it. (You save a value in it, but
make never use of it.)

> +static void aspeed_set_pwm_channel_enable(struct regmap *regmap, u8 pwm_channel,
> +					  bool enable)
> +{
> +	regmap_update_bits(regmap, ASPEED_PWM_CTRL_CH(pwm_channel),
> +			   (PWM_CLK_ENABLE | PWM_PIN_EN),
> +			   enable ? (PWM_CLK_ENABLE | PWM_PIN_EN) : 0);

What is the semantic of PIN_EN?

> +}
> +/*
> + * The PWM frequency = HCLK(200Mhz) / (clock division L bit *
> + * clock division H bit * (period bit + 1))
> + */
> +static void aspeed_set_pwm_freq(struct aspeed_pwm_data *priv,
> +				struct pwm_device *pwm, u32 freq)
> +{
> +	u32 target_div, cal_freq;
> +	u32 tmp_div_h, tmp_div_l, diff, min_diff = INT_MAX;
> +	u32 div_h = BIT(5) - 1, div_l = BIT(8) - 1;
> +	u8 div_found;
> +	u32 index = pwm->hwpwm;
> +	struct aspeed_pwm *channel = pwm_get_chip_data(pwm);
> +
> +	cal_freq = priv->clk_freq / (DEFAULT_PWM_PERIOD + 1);
> +	target_div = DIV_ROUND_UP(cal_freq, freq);
> +	div_found = 0;
> +	/* calculate for target frequence */

s/frequence/frequency/

> +	for (tmp_div_h = 0; tmp_div_h < 0x10; tmp_div_h++) {
> +		tmp_div_l = target_div / BIT(tmp_div_h) - 1;
> +
> +		if (tmp_div_l < 0 || tmp_div_l > 255)
> +			continue;
> +
> +		diff = freq - cal_freq / (BIT(tmp_div_h) * (tmp_div_l + 1));
> +		if (abs(diff) < abs(min_diff)) {
> +			min_diff = diff;
> +			div_l = tmp_div_l;
> +			div_h = tmp_div_h;
> +			div_found = 1;
> +			if (diff == 0)
> +				break;
> +		}
> +	}

If my understanding is right (i.e. H divides by a power of two and L by
an integer) this can be simplified.

> +	if (div_found == 0) {
> +		pr_debug("target freq: %d too slow set minimal frequency\n",
> +			 freq);
> +	}
> +	channel->freq = cal_freq / (BIT(div_h) * (div_l + 1));
> +	pr_debug("div h %x, l : %x pwm out clk %d\n", div_h, div_l,
> +		 channel->freq);
> +	pr_debug("hclk %ld, target pwm freq %d, real pwm freq %d\n",
> +		 priv->clk_freq, freq, channel->freq);
> +
> +	regmap_update_bits(priv->regmap, ASPEED_PWM_CTRL_CH(index),
> +			   (PWM_CLK_DIV_H_MASK | PWM_CLK_DIV_L_MASK),
> +			   (div_h << PWM_CLK_DIV_H_SHIFT) |
> +				   (div_l << PWM_CLK_DIV_L_SHIFT));
> +}
> +
> +static void aspeed_set_pwm_duty(struct aspeed_pwm_data *priv,
> +				struct pwm_device *pwm, u32 duty_pt)
> +{
> +	u32 index = pwm->hwpwm;
> +
> +	if (duty_pt == 0) {
> +		aspeed_set_pwm_channel_enable(priv->regmap, index, false);
> +	} else {
> +		regmap_update_bits(priv->regmap,
> +				   ASPEED_PWM_DUTY_CYCLE_CH(index),
> +				   PWM_FALLING_POINT_MASK,
> +				   duty_pt << PWM_FALLING_POINT_SHIFT);
> +		aspeed_set_pwm_channel_enable(priv->regmap, index, true);
> +	}
> +}
> +
> +static void aspeed_set_pwm_polarity(struct aspeed_pwm_data *priv,
> +				    struct pwm_device *pwm, u8 polarity)
> +{
> +	u32 index = pwm->hwpwm;
> +
> +	regmap_update_bits(priv->regmap, ASPEED_PWM_CTRL_CH(index), PWM_INVERSE,
> +			   (polarity) ? PWM_INVERSE : 0);
> +}
> +
> +static int aspeed_pwm_request(struct pwm_chip *chip, struct pwm_device *pwm)
> +{
> +	struct device *dev = chip->dev;
> +	struct aspeed_pwm_data *priv = dev_get_drvdata(dev);
> +	struct aspeed_pwm *channel;
> +	u32 index = pwm->hwpwm;
> +	/*
> +	 * Fixed the period to the max value and rising point to 0
> +	 * for high resolution and simplified frequency calculation.

s/^H//

> +	 */
> +	regmap_update_bits(priv->regmap, ASPEED_PWM_DUTY_CYCLE_CH(index),
> +			   PWM_PERIOD_MASK,
> +			   DEFAULT_PWM_PERIOD << PWM_PERIOD_SHIFT);
> +
> +	regmap_update_bits(priv->regmap, ASPEED_PWM_DUTY_CYCLE_CH(index),
> +			   PWM_RISING_POINT_MASK, 0);

Only .apply is supposed to modify the PWM's configuration.

> +	channel = devm_kzalloc(dev, sizeof(*channel), GFP_KERNEL);
> +	if (!channel)
> +		return -ENOMEM;

Don't use devm_kzalloc if freeing isn't done at device cleanup time.

> +	pwm_set_chip_data(pwm, channel);
> +
> +	return 0;

> +}
> +
> +static void aspeed_pwm_free(struct pwm_chip *chip, struct pwm_device *pwm)
> +{
> +	struct device *dev = chip->dev;
> +	struct aspeed_pwm *channel = pwm_get_chip_data(pwm);
> +
> +	devm_kfree(dev, channel);
> +}
> +
> +static int aspeed_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> +			    const struct pwm_state *state)
> +{
> +	struct device *dev = chip->dev;
> +	struct aspeed_pwm_data *priv = dev_get_drvdata(dev);

Please consider using

	priv = container_of(chip, struct aspeed_pwm_data, chip);

(preferably wrapped in a macro) which is more type safe and more
effective to calculate.

> +	struct pwm_state *cur_state = &pwm->state;
> +	u32 freq = DIV_ROUND_UP_ULL(1000000000, state->period);
> +	u32 duty_pt = DIV_ROUND_UP_ULL(
> +		state->duty_cycle * (DEFAULT_PWM_PERIOD + 1), state->period);

You're loosing precision here.

> +	dev_dbg(dev, "freq: %d, duty_pt: %d", freq, duty_pt);
> +	if (state->enabled) {
> +		aspeed_set_pwm_freq(priv, pwm, freq);
> +		aspeed_set_pwm_duty(priv, pwm, duty_pt);
> +		aspeed_set_pwm_polarity(priv, pwm, state->polarity);

How does the hardware behave between these calls? E.g. can it happen
that it already emits a normal period when inversed polarity is
requested just before aspeed_set_pwm_polarity is called? Or there is a
period with the previous duty cycle and the new period?

> +	} else {
> +		aspeed_set_pwm_duty(priv, pwm, 0);
> +	}
> +	cur_state->period = state->period;
> +	cur_state->duty_cycle = state->duty_cycle;
> +	cur_state->polarity = state->polarity;
> +	cur_state->enabled = state->enabled;

The driver is not supposed to modify pwm->state.

> +	return 0;
> +}

From your code I understood: The period of the signal is

	(PWM_PERIOD + 1) * (2 ** DIV_H) * (DIV_L + 1) / 200 MHz

. The duty cycle is

	PWM_FALLING_POINT * (2 ** DIV_H) * (DIV_L + 1) / 200 MHz

. So the PWM cannot provide a 100% relative duty cycle.

Is this right?

> +static const struct pwm_ops aspeed_pwm_ops = {
> +	.request = aspeed_pwm_request,
> +	.free = aspeed_pwm_free,
> +	.apply = aspeed_pwm_apply,

Please test your driver with PWM_DEBUG enabled.

> +	.owner = THIS_MODULE,
> +};
> +
> +static int aspeed_pwm_probe(struct platform_device *pdev)
> +{
> +	struct device *dev = &pdev->dev;
> +	struct clk *clk;
> +	int ret;
> +	struct aspeed_pwm_data *priv;
> +	struct device_node *np;
> +
> +	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
> +	if (!priv)
> +		return -ENOMEM;
> +
> +	np = pdev->dev.parent->of_node;
> +	if (!of_device_is_compatible(np, "aspeed,ast2600-pwm-tach")) {
> +		dev_err(dev, "unsupported pwm device binding\n");
> +		return -ENODEV;
> +	}

Is this pwm-tach an mfd?

> +	priv->regmap = syscon_node_to_regmap(np);
> +	if (IS_ERR(priv->regmap)) {
> +		dev_err(dev, "Couldn't get regmap\n");
> +		return -ENODEV;
> +	}
> +
> +	clk = devm_clk_get(dev, NULL);
> +	if (IS_ERR(clk))
> +		return -ENODEV;
> +	priv->clk_freq = clk_get_rate(clk);

If you intend to use this clock, you have to enable it.

> +	priv->reset = reset_control_get_shared(&pdev->dev, NULL);
> +	if (IS_ERR(priv->reset)) {
> +		dev_err(&pdev->dev, "can't get aspeed_pwm_tacho reset\n");
> +		return PTR_ERR(priv->reset);
> +	}
> +	reset_control_deassert(priv->reset);

missing error checking

> +	priv->chip.dev = dev;
> +	priv->chip.ops = &aspeed_pwm_ops;
> +	priv->chip.base = -1;

This isn't necessary since f9a8ee8c8bcd118e800d88772c6457381db45224,
please drop the assignment to base.

> +	priv->chip.npwm = ASPEED_NR_PWMS;
> +	priv->chip.of_xlate = of_pwm_xlate_with_flags;
> +	priv->chip.of_pwm_n_cells = 3;
> +
> +	ret = pwmchip_add(&priv->chip);
> +	if (ret < 0) {
> +		dev_err(&pdev->dev, "failed to add PWM chip: %d\n", ret);

Please use %pe to make the error messages better readable.

> +		return ret;
> +	}
> +	dev_set_drvdata(dev, priv);
> +	return ret;
> +}
> +
> +static const struct of_device_id of_pwm_match_table[] = {
> +	{
> +		.compatible = "aspeed,ast2600-pwm",
> +	},
> +	{},
> +};
> +MODULE_DEVICE_TABLE(of, of_pwm_match_table);
> +
> +static struct platform_driver aspeed_pwm_driver = {
> +	.probe		= aspeed_pwm_probe,

Please implement a .remove callback.

> +	.driver		= {
> +		.name	= "aspeed_pwm",
> +		.of_match_table = of_pwm_match_table,
> +	},
> +};

Best regards
Uwe

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

Attachment: signature.asc
Description: PGP signature