RE: [PATCH v11] pwm: opencores: Add PWM driver support

[William Qiu]

Hello,

Sorry for late reply.

> -----Original Message-----
> From: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
> Sent: 2024年4月14日 1:01
> To: William Qiu <william.qiu@xxxxxxxxxxxxxxxx>
> Cc: linux-kernel@xxxxxxxxxxxxxxx; linux-pwm@xxxxxxxxxxxxxxx; Hal Feng
> <hal.feng@xxxxxxxxxxxxxxxx>; Philipp Zabel <p.zabel@xxxxxxxxxxxxxx>
> Subject: Re: [PATCH v11] pwm: opencores: Add PWM driver support
> 
> Hello,
> 
> thanks for your patience to wait for my review.
> 
> On Fri, Feb 23, 2024 at 04:43:32PM +0800, William Qiu wrote:
> > Add driver for OpenCores PWM Controller. And add compatibility code
> > which based on StarFive SoC.
> >
> > Co-developed-by: Hal Feng <hal.feng@xxxxxxxxxxxxxxxx>
> > Signed-off-by: Hal Feng <hal.feng@xxxxxxxxxxxxxxxx>
> > Signed-off-by: William Qiu <william.qiu@xxxxxxxxxxxxxxxx>
> > ---
> >  MAINTAINERS              |   7 ++
> >  drivers/pwm/Kconfig      |  12 ++
> >  drivers/pwm/Makefile     |   1 +
> >  drivers/pwm/pwm-ocores.c | 232
> > +++++++++++++++++++++++++++++++++++++++
> >  4 files changed, 252 insertions(+)
> >  create mode 100644 drivers/pwm/pwm-ocores.c
> >
> > diff --git a/MAINTAINERS b/MAINTAINERS index
> > 9ed4d3868539..12ea5e86fc23 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -16414,6 +16414,13 @@ F:	Documentation/i2c/busses/i2c-ocores.rst
> >  F:	drivers/i2c/busses/i2c-ocores.c
> >  F:	include/linux/platform_data/i2c-ocores.h
> >
> > +OPENCORES PWM DRIVER
> > +M:	William Qiu <william.qiu@xxxxxxxxxxxxxxxx>
> > +M:	Hal Feng <hal.feng@xxxxxxxxxxxxxxxx>
> > +S:	Supported
> > +F:	Documentation/devicetree/bindings/pwm/opencores,pwm.yaml
> > +F:	drivers/pwm/pwm-ocores.c
> > +
> >  OPENRISC ARCHITECTURE
> >  M:	Jonas Bonn <jonas@xxxxxxxxxxxx>
> >  M:	Stefan Kristiansson <stefan.kristiansson@xxxxxxxxxxxxx>
> > diff --git a/drivers/pwm/Kconfig b/drivers/pwm/Kconfig index
> > 4b956d661755..d87e1bb350ba 100644
> > --- a/drivers/pwm/Kconfig
> > +++ b/drivers/pwm/Kconfig
> > @@ -444,6 +444,18 @@ config PWM_NTXEC
> >  	  controller found in certain e-book readers designed by the original
> >  	  design manufacturer Netronix.
> >
> > +config PWM_OCORES
> > +	tristate "OpenCores PWM support"
> 
> OpenCores PTC PWM support?
> 

Will update.

> > +	depends on HAS_IOMEM && OF
> > +	depends on COMMON_CLK && RESET_CONTROLLER
> > +	depends on ARCH_STARFIVE || COMPILE_TEST
> > +	help
> > +	  If you say yes to this option, support will be included for the
> > +	  OpenCores PWM. For details see https://opencores.org/projects/ptc.
> > +
> > +	  To compile this driver as a module, choose M here: the module
> > +	  will be called pwm-ocores.
> > +
> >  config PWM_OMAP_DMTIMER
> >  	tristate "OMAP Dual-Mode Timer PWM support"
> >  	depends on OF
> > diff --git a/drivers/pwm/Makefile b/drivers/pwm/Makefile index
> > c5ec9e168ee7..517c4f643058 100644
> > --- a/drivers/pwm/Makefile
> > +++ b/drivers/pwm/Makefile
> > @@ -40,6 +40,7 @@ obj-$(CONFIG_PWM_MICROCHIP_CORE)	+=
> pwm-microchip-core.o
> >  obj-$(CONFIG_PWM_MTK_DISP)	+= pwm-mtk-disp.o
> >  obj-$(CONFIG_PWM_MXS)		+= pwm-mxs.o
> >  obj-$(CONFIG_PWM_NTXEC)		+= pwm-ntxec.o
> > +obj-$(CONFIG_PWM_OCORES)	+= pwm-ocores.o
> >  obj-$(CONFIG_PWM_OMAP_DMTIMER)	+= pwm-omap-dmtimer.o
> >  obj-$(CONFIG_PWM_PCA9685)	+= pwm-pca9685.o
> >  obj-$(CONFIG_PWM_PXA)		+= pwm-pxa.o
> > diff --git a/drivers/pwm/pwm-ocores.c b/drivers/pwm/pwm-ocores.c new
> > file mode 100644 index 000000000000..874bc630bf2d
> > --- /dev/null
> > +++ b/drivers/pwm/pwm-ocores.c
> > @@ -0,0 +1,232 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * OpenCores PWM Driver
> > + *
> > + * https://opencores.org/projects/ptc
> > + *
> > + * Copyright (C) 2018-2023 StarFive Technology Co., Ltd.
> > + *
> > + * Limitations:
> > + * - The hardware only do inverted polarity.
> > + * - The hardware minimum period / duty_cycle is (1 / pwm_apb clock
> frequency) ns.
> > + * - The hardware maximum period / duty_cycle is (U32_MAX / pwm_apb
> clock frequency) ns.
> > + */
> > +
> > +#include <linux/clk.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/reset.h>
> > +#include <linux/slab.h>
> > +
> > +/* OCPWM_CTRL register bits*/
> > +#define REG_OCPWM_EN      BIT(0)
> > +#define REG_OCPWM_ECLK    BIT(1)
> > +#define REG_OCPWM_NEC     BIT(2)
> > +#define REG_OCPWM_OE      BIT(3)
> > +#define REG_OCPWM_SIGNLE  BIT(4)
> > +#define REG_OCPWM_INTE    BIT(5)
> > +#define REG_OCPWM_INT     BIT(6)
> > +#define REG_OCPWM_CNTRRST BIT(7)
> > +#define REG_OCPWM_CAPTE   BIT(8)
> > +
> > +struct ocores_pwm_device {
> > +	struct pwm_chip chip;
> > +	struct clk *clk;
> > +	struct reset_control *rst;
> > +	const struct ocores_pwm_data *data;
> > +	void __iomem *regs;
> > +	u32 clk_rate; /* PWM APB clock frequency */ };
> > +
> > +struct ocores_pwm_data {
> > +	void __iomem *(*get_ch_base)(void __iomem *base, unsigned int
> > +channel); };
> > +
> > +static inline u32 ocores_readl(struct ocores_pwm_device *ddata,
> 
> ocores_pwm_readl is a tad longer (which is annoying), but IMHO the advantage
> of the longer name (no clash with other ocores IP drivers, being able to easily
> setup ftrace filtering for all functions in this
> driver) outweighs the shorter name. Can you please update accordingly.
> (There are a few more symbols that the same treatment.)
> 

Will update.

> > +			       unsigned int channel,
> > +			       unsigned int offset)
> > +{
> > +	void __iomem *base = ddata->data->get_ch_base ?
> > +			     ddata->data->get_ch_base(ddata->regs, channel) :
> ddata->regs;
> > +
> > +	return readl(base + offset);
> > +}
> > [...]
> > +static void __iomem *starfive_jh71x0_get_ch_base(void __iomem *base,
> > +						 unsigned int channel)
> > +{
> > +	unsigned int offset = (channel > 3 ? 1 << 15 : 0) + (channel & 3) *
> > +0x10;
> 
> offset = (channel & 4) << 13 | (channel & 3) << 4
> 
> results in the same offsets and can be compiled to more efficient code.
> (Well, at least on ARM, I suspect the same applies to riscv.)
> 

I'll try it first, and update it later.

> > +	return base + offset;
> > +}
> > +
> > +static int ocores_pwm_get_state(struct pwm_chip *chip,
> > +				struct pwm_device *pwm,
> > +				struct pwm_state *state)
> > +{
> > +	struct ocores_pwm_device *ddata = chip_to_ocores(chip);
> > +	u32 period_data, duty_data, ctrl_data;
> > +
> > +	period_data = ocores_readl(ddata, pwm->hwpwm, 0x8);
> > +	duty_data = ocores_readl(ddata, pwm->hwpwm, 0x4);
> > +	ctrl_data = ocores_readl(ddata, pwm->hwpwm, 0xC);
> 
> Can you please give symbolic names to these offsets?
> 

Will name it.

> > [...]
> > +
> > +static int ocores_pwm_apply(struct pwm_chip *chip,
> > +			    struct pwm_device *pwm,
> > +			    const struct pwm_state *state) {
> > +	struct ocores_pwm_device *ddata = chip_to_ocores(chip);
> > +	u32 ctrl_data = 0;
> > +	u64 period_data, duty_data;
> > +
> > +	if (state->polarity != PWM_POLARITY_INVERSED)
> > +		return -EINVAL;
> > +
> > +	ctrl_data = ocores_readl(ddata, pwm->hwpwm, 0xC);
> > +	ocores_writel(ddata, pwm->hwpwm, 0xC, 0);
> > +
> > +	period_data = DIV_ROUND_DOWN_ULL(state->period * ddata->clk_rate,
> > +NSEC_PER_SEC);
> 
> The multiplication might overflow. Please use mul_u64_u32_div and in .probe
> assert that ddata->clk_rate <= NSEC_PER_SEC.
> 

Will update.

> > +	if (period_data <= U32_MAX)
> > +		ocores_writel(ddata, pwm->hwpwm, 0x8, (u32)period_data);
> > +	else
> > +		return -EINVAL;
> 
> Please make this:
> 
> 	if (period_data <= U32_MAX)
> 		period_data = U32_MAX;
> 
> What happens if period_data == 0? I guess this is a problem and you should
> return -EINVAL in that case.
> 

Will fix.

> > +	duty_data = DIV_ROUND_DOWN_ULL(state->duty_cycle *
> ddata->clk_rate, NSEC_PER_SEC);
> > +	if (duty_data <= U32_MAX)
> > +		ocores_writel(ddata, pwm->hwpwm, 0x4, (u32)duty_data);
> > +	else
> > +		return -EINVAL;
> > +
> > +	ocores_writel(ddata, pwm->hwpwm, 0xC, 0);
> 
> What is the effect on this one? I guess it disables the output? Is this necessary?
> Does updating the configuration complete the currently running period?
> 
> Please document in the Limitations paragraph if there are possible glitches (e.g.
> when the period register is written but the duty_cycle register not yet) and if
> the current period is completed.
> 

Will check it.

> > [...]
> > +static int ocores_pwm_probe(struct platform_device *pdev) {
> > +	const struct of_device_id *id;
> > +	struct device *dev = &pdev->dev;
> > +	struct ocores_pwm_device *ddata;
> > +	struct pwm_chip *chip;
> > +	int ret;
> > +
> > +	id = of_match_device(ocores_pwm_of_match, dev);
> > +	if (!id)
> > +		return -EINVAL;
> > +
> > +	ddata = devm_kzalloc(dev, sizeof(*ddata), GFP_KERNEL);
> > +	if (!ddata)
> > +		return -ENOMEM;
> > +
> > +	ddata->data = id->data;
> > +	chip = &ddata->chip;
> 
> This needs updating with the changes that got into 6.9-rc1. See
> ae8635e99c5cc752e204ab9ee8869ec54a9223f0 for a commit you might want
> to take as a template for the change needed here.
> 

Will fix.

> > +	chip->dev = dev;
> > +	chip->ops = &ocores_pwm_ops;
> > +	chip->npwm = 8;
> > +
> > +	ddata->regs = devm_platform_ioremap_resource(pdev, 0);
> > +	if (IS_ERR(ddata->regs))
> > +		return dev_err_probe(dev, PTR_ERR(ddata->regs),
> > +				     "Unable to map IO resources\n");
> > +
> > +	ddata->clk = devm_clk_get_enabled(dev, NULL);
> 
> This member is only used here in .probe(). So it doesn't need to be stored in
> struct ocores_pwm_device.
> 

Will drop it from struct ocores_pwm_device.

> > +	if (IS_ERR(ddata->clk))
> > +		return dev_err_probe(dev, PTR_ERR(ddata->clk),
> > +				     "Unable to get pwm's clock\n");
> > +
> > +	ddata->rst = devm_reset_control_get_optional_exclusive(dev, NULL);
> 
> Same as for clk, rst is only used here.
> 

...

> > +	if (IS_ERR(ddata->rst))
> > +		return dev_err_probe(dev, PTR_ERR(ddata->rst),
> > +				     "Unable to get pwm's reset\n");
> > +
> > +	reset_control_deassert(ddata->rst);
> > +
> > +	ret = devm_add_action_or_reset(dev, ocores_reset_control_assert,
> ddata->rst);
> > +	if (ret)
> > +		return ret;
> 
> Please add a call to devm_clk_rate_exclusive_get() before storing the rate and
> relying on it not changing.
> 

Will add.

> > +	ddata->clk_rate = clk_get_rate(ddata->clk);
> > +	if (ddata->clk_rate <= 0)
> > +		return dev_err_probe(dev, ddata->clk_rate,
> > +				     "Unable to get clock's rate\n");
> > +
> > +	ret = devm_pwmchip_add(dev, chip);
> > +	if (ret < 0)
> > +		return dev_err_probe(dev, ret, "Could not register PWM chip\n");
> > +
> > +	platform_set_drvdata(pdev, ddata);
> 
> This is unused.
> 

Will drop.

> > +	return ret;
> 
> Here ret is always 0, please use return 0 here.
> 

Will update.

> > +}
> > +
> > +static struct platform_driver ocores_pwm_driver = {
> > +	.probe = ocores_pwm_probe,
> > +	.driver = {
> > +		.name = "ocores-pwm",
> > +		.of_match_table = ocores_pwm_of_match,
> > +	},
> > +};
> > +module_platform_driver(ocores_pwm_driver);
> > +
> > +MODULE_AUTHOR("Jieqin Chen");
> > +MODULE_AUTHOR("Hal Feng <hal.feng@xxxxxxxxxxxxxxxx>");
> > +MODULE_DESCRIPTION("OpenCores PWM PTC driver");
> 
> The hardware unit is called PTC (PWM/Timer/Counter), so "OpenCores PTC
> PWM driver" would be more appropriate?!
> 

Will update.

> > +MODULE_LICENSE("GPL");
> 
> Best regards
> Uwe
> 
> --
> Pengutronix e.K.                           | Uwe Kleine-König
> |
> Industrial Linux Solutions                 | https://www.pengutronix.de/ |

Thank you for taking time to review this patch and give helpful suggestions.

Best regards,
William