Skip to contentSkip to conversation

pwm: opencores: Add PWM driver support

PATCH Superseded William Qiu 3 participants 8 messages Jul 2, 2024 – Aug 7, 2024 PWM

v13 of 17 Superseded Latest is v17 →

lore
Share & tools
lore
#0 lore raw

PATCH v13 1/1 pwm: opencores: Add PWM driver support

Add driver for OpenCores PWM Controller. And add compatibility code which based on StarFive SoC.

---
 MAINTAINERS              |   7 ++
 drivers/pwm/Kconfig      |  12 ++
 drivers/pwm/Makefile     |   1 +
 drivers/pwm/pwm-ocores.c | 239 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 259 insertions(+)
 create mode 100644 drivers/pwm/pwm-ocores.c
MAINTAINERS +7 -0
index 3c4fdf74a3f9..3b547ede2ce5 100644
@@ -16824,6 +16824,13 @@ 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 <[email protected]>
+M:	Hal Feng <[email protected]>
+S:	Supported
+F:	Documentation/devicetree/bindings/pwm/opencores,pwm.yaml
+F:	drivers/pwm/pwm-ocores.c
+
 OPENRISC ARCHITECTURE
 M:	Jonas Bonn <[email protected]>
 M:	Stefan Kristiansson <[email protected]>
drivers/pwm/Kconfig +12 -0
index 1dd7921194f5..42158bc1c8bc 100644
@@ -440,6 +440,18 @@ config PWM_NTXEC
 	  controller found in certain e-book readers designed by the original
 	  design manufacturer Netronix.
 
+config PWM_OCORES
+	tristate "OpenCores PTC PWM support"
+	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
drivers/pwm/Makefile +1 -0
index 90913519f11a..7a44d8afe044 100644
@@ -39,6 +39,7 @@ 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
drivers/pwm/pwm-ocores.c new file +239 -0

239 lines not on this page · raw message

Signature · 1 line
2.34.1
#1 lore raw
Quoted 37 lines · William Qiu

Add driver for OpenCores PWM Controller. And add compatibility code which based on StarFive SoC.

---
 MAINTAINERS              |   7 ++
 drivers/pwm/Kconfig      |  12 ++
 drivers/pwm/Makefile     |   1 +
 drivers/pwm/pwm-ocores.c | 239 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 259 insertions(+)
 create mode 100644 drivers/pwm/pwm-ocores.c
MAINTAINERS +7 -0
index 3c4fdf74a3f9..3b547ede2ce5 100644
@@ -16824,6 +16824,13 @@ 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 <[email protected]>
+M:	Hal Feng <[email protected]>
+S:	Supported
+F:	Documentation/devicetree/bindings/pwm/opencores,pwm.yaml
+F:	drivers/pwm/pwm-ocores.c
+
 OPENRISC ARCHITECTURE
 M:	Jonas Bonn <[email protected]>
 M:	Stefan Kristiansson <[email protected]>
drivers/pwm/Kconfig +0 -0
index 1dd7921194f5..42158bc1c8bc 100644

drivers/pwm/Kconfig config PWM_NTXEC · line 446

Quoted 7 lines
drivers/pwm/Kconfig +4 -0
@@ -440,6 +440,18 @@ config PWM_NTXEC
 	  controller found in certain e-book readers designed by the original
 	  design manufacturer Netronix.
 
+config PWM_OCORES
+	tristate "OpenCores PTC PWM support"
+	depends on HAS_IOMEM && OF
+	depends on COMMON_CLK && RESET_CONTROLLER

The reset controller API has stubs in case RESET_CONTROLLER is disabled, so in general reset consumers don't need to depend on it.

regards
Philipp

#2 lore raw
Quoted 61 lines

-----Original Message-----
From: Philipp Zabel <[email protected]>
Sent: 2024年7月2日 17:22
To: William Qiu <[email protected]>; [email protected]; [email protected]
Cc: Uwe Kleine-König <[email protected]>; Hal Feng <[email protected]>
Subject: Re: [PATCH v13] pwm: opencores: Add PWM driver support

Quoted 38 lines · William Qiu

Add driver for OpenCores PWM Controller. And add compatibility code which based on StarFive SoC.

---
 MAINTAINERS              |   7 ++
 drivers/pwm/Kconfig      |  12 ++
 drivers/pwm/Makefile     |   1 +
 drivers/pwm/pwm-ocores.c | 239
+++++++++++++++++++++++++++++++++++++++
 4 files changed, 259 insertions(+)
 create mode 100644 drivers/pwm/pwm-ocores.c
MAINTAINERS index renamed from MAINTAINERS +0 -0

3c4fdf74a3f9..3b547ede2ce5 100644

MAINTAINERS +7 -0
@@ -16824,6 +16824,13 @@ 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 <[email protected]>
+M:	Hal Feng <[email protected]>
+S:	Supported
+F:	Documentation/devicetree/bindings/pwm/opencores,pwm.yaml
+F:	drivers/pwm/pwm-ocores.c
+
 OPENRISC ARCHITECTURE
 M:	Jonas Bonn <[email protected]>
 M:	Stefan Kristiansson <[email protected]>
drivers/pwm/Kconfig index renamed from drivers/pwm/Kconfig +0 -0

1dd7921194f5..42158bc1c8bc 100644

drivers/pwm/Kconfig config PWM_NTXEC · line 446

Quoted 7 lines
drivers/pwm/Kconfig +4 -0
@@ -440,6 +440,18 @@ config PWM_NTXEC
 	  controller found in certain e-book readers designed by the original
 	  design manufacturer Netronix.
 
+config PWM_OCORES
+	tristate "OpenCores PTC PWM support"
+	depends on HAS_IOMEM && OF
+	depends on COMMON_CLK && RESET_CONTROLLER

The reset controller API has stubs in case RESET_CONTROLLER is disabled, so in general reset consumers don't need to depend on it.

Quoted 2 lines

regards
Philipp

I see, I'll drop it.

Best Regards,
William

#3 lore raw
In reply to William Qiu
Quoted 335 lines

-----Original Message-----
From: William Qiu
Sent: 2024年7月2日 16:39
To: [email protected]; [email protected]
Cc: Uwe Kleine-König <[email protected]>; Hal Feng <[email protected]>; Philipp Zabel <[email protected]>; William Qiu <[email protected]>
Subject: [PATCH v13] pwm: opencores: Add PWM driver support

Add driver for OpenCores PWM Controller. And add compatibility code which based on StarFive SoC.

---
 MAINTAINERS              |   7 ++
 drivers/pwm/Kconfig      |  12 ++
 drivers/pwm/Makefile     |   1 +
 drivers/pwm/pwm-ocores.c | 239
+++++++++++++++++++++++++++++++++++++++
 4 files changed, 259 insertions(+)
 create mode 100644 drivers/pwm/pwm-ocores.c
MAINTAINERS +7 -0
index 3c4fdf74a3f9..3b547ede2ce5 100644
@@ -16824,6 +16824,13 @@ 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 <[email protected]>
+M:	Hal Feng <[email protected]>
+S:	Supported
+F:	Documentation/devicetree/bindings/pwm/opencores,pwm.yaml
+F:	drivers/pwm/pwm-ocores.c
+
 OPENRISC ARCHITECTURE
 M:	Jonas Bonn <[email protected]>
 M:	Stefan Kristiansson <[email protected]>
drivers/pwm/Kconfig index renamed from drivers/pwm/Kconfig +0 -0

1dd7921194f5..42158bc1c8bc 100644

drivers/pwm/Kconfig +12 -0
@@ -440,6 +440,18 @@ config PWM_NTXEC
 	  controller found in certain e-book readers designed by the original
 	  design manufacturer Netronix.
 
+config PWM_OCORES
+	tristate "OpenCores PTC PWM support"
+	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
drivers/pwm/Makefile index renamed from drivers/pwm/Makefile +0 -0

90913519f11a..7a44d8afe044 100644

drivers/pwm/Makefile +0 -0
@@ -39,6 +39,7 @@ 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

drivers/pwm/Makefile +1 -0
+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
drivers/pwm/pwm-ocores.c new file renamed from drivers/pwm/pwm-ocores.c +0 -0

mode 100644 index 000000000000..c8f08aa14e44

drivers/pwm/pwm-ocores.c new file +11 -0
@@ -0,0 +1,239 @@
+// 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 supports 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.
+ * - The hardware is set to a low level immediately when disabledThe hardware is set to
+ * a low level immediately when disabled.
+ * - The hardware will have a conversion cycle when reconfiguring.
+ */

drivers/pwm/pwm-ocores.c new file +43 -0
+
+#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>
+
+/* OpenCores Register offsets */
+#define REG_OCPWM_CNTR    0x0
+#define REG_OCPWM_HRC     0x4
+#define REG_OCPWM_LRC     0x8
+#define REG_OCPWM_CTRL    0xC
+
+/* OCPWM_CTRL register bits*/
+#define REG_OCPWM_CNTR_EN      BIT(0)
+#define REG_OCPWM_CNTR_ECLK    BIT(1)
+#define REG_OCPWM_CNTR_NEC     BIT(2)
+#define REG_OCPWM_CNTR_OE      BIT(3)
+#define REG_OCPWM_CNTR_SIGNLE  BIT(4)
+#define REG_OCPWM_CNTR_INTE    BIT(5)
+#define REG_OCPWM_CNTR_INT     BIT(6)
+#define REG_OCPWM_CNTR_RST     BIT(7)
+#define REG_OCPWM_CNTR_CAPTE   BIT(8)
+
+struct ocores_pwm_device {
+	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_pwm_readl(struct ocores_pwm_device *ddata,
+				   unsigned int channel,
+				   unsigned int offset)
+{
+	void __iomem *base = ddata->data->get_ch_base ?
+			     ddata->data->get_ch_base(ddata->regs, channel) :

ddata->regs;

drivers/pwm/pwm-ocores.c new file +10 -0
+
+	return readl(base + offset);
+}
+
+static inline void ocores_pwm_writel(struct ocores_pwm_device *ddata,
+				     unsigned int channel,
+				     unsigned int offset, u32 val)
+{
+	void __iomem *base = ddata->data->get_ch_base ?
+			     ddata->data->get_ch_base(ddata->regs, channel) :

ddata->regs;

drivers/pwm/pwm-ocores.c new file +25 -0
+
+	writel(val, base + offset);
+}
+
+static inline struct ocores_pwm_device *chip_to_ocores(struct pwm_chip
+*chip) {
+	return pwmchip_get_drvdata(chip);
+}
+
+static void __iomem *starfive_get_ch_base(void __iomem *base,
+					  unsigned int channel)
+{
+	unsigned int offset = (channel & 4) << 13 | (channel & 3) << 4;
+
+	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_pwm_readl(ddata, pwm->hwpwm,

REG_OCPWM_LRC);

drivers/pwm/pwm-ocores.c new file +1 -0
+	duty_data = ocores_pwm_readl(ddata, pwm->hwpwm,

REG_OCPWM_HRC);

drivers/pwm/pwm-ocores.c new file +1 -0
+	ctrl_data = ocores_pwm_readl(ddata, pwm->hwpwm,

REG_OCPWM_CTRL);

drivers/pwm/pwm-ocores.c new file +2 -0
+
+	state->period = DIV_ROUND_UP_ULL((u64)period_data * NSEC_PER_SEC,

ddata->clk_rate);

drivers/pwm/pwm-ocores.c new file +1 -0
+	state->duty_cycle = DIV_ROUND_UP_ULL((u64)duty_data *

NSEC_PER_SEC, ddata->clk_rate);

drivers/pwm/pwm-ocores.c new file +18 -0
+	state->polarity = PWM_POLARITY_INVERSED;
+	state->enabled = (ctrl_data & REG_OCPWM_CNTR_EN) ? true : false;
+
+	return 0;
+}
+
+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;
+
+	period_data = mul_u64_u32_div(state->period, ddata->clk_rate,

NSEC_PER_SEC);

drivers/pwm/pwm-ocores.c new file +9 -0
+	if (!period_data)
+		return -EINVAL;
+
+	if (period_data > U32_MAX)
+		period_data = U32_MAX;
+
+	ocores_pwm_writel(ddata, pwm->hwpwm, 0x8, (u32)period_data);
+
+	duty_data = mul_u64_u32_div(state->duty_cycle, ddata->clk_rate,

NSEC_PER_SEC);

drivers/pwm/pwm-ocores.c new file +2 -0
+	if (duty_data <= U32_MAX)
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_HRC,

(u32)duty_data);

drivers/pwm/pwm-ocores.c new file +6 -0
+	else if (duty_data > U32_MAX)
+		duty_data = U32_MAX;
+	else
+		return -EINVAL;
+
+	ctrl_data = ocores_pwm_readl(ddata, pwm->hwpwm,

REG_OCPWM_CTRL);

drivers/pwm/pwm-ocores.c new file +3 -0
+	if (state->enabled)
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_CTRL,
+				  ctrl_data | REG_OCPWM_CNTR_EN |

REG_OCPWM_CNTR_OE);

drivers/pwm/pwm-ocores.c new file +3 -0
+	else
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_CTRL,
+				  ctrl_data & ~(REG_OCPWM_CNTR_EN |

REG_OCPWM_CNTR_OE));

drivers/pwm/pwm-ocores.c new file +68 -0
+
+	return 0;
+}
+
+static const struct pwm_ops ocores_pwm_ops = {
+	.get_state = ocores_pwm_get_state,
+	.apply = ocores_pwm_apply,
+};
+
+static const struct ocores_pwm_data starfive_pwm_data = {
+	.get_ch_base = starfive_get_ch_base,
+};
+
+static const struct of_device_id ocores_pwm_of_match[] = {
+	{ .compatible = "opencores,pwm-v1" },
+	{ .compatible = "starfive,jh7100-pwm", .data = &starfive_pwm_data},
+	{ .compatible = "starfive,jh7110-pwm", .data = &starfive_pwm_data},
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, ocores_pwm_of_match);
+
+static void ocores_pwm_reset_control_assert(void *data) {
+	reset_control_assert(data);
+}
+
+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;
+	struct clk *clk;
+	struct reset_control *rst;
+	int ret;
+
+	id = of_match_device(ocores_pwm_of_match, dev);
+	if (!id)
+		return -EINVAL;
+
+	chip = devm_pwmchip_alloc(&pdev->dev, 8, sizeof(*ddata));
+	if (IS_ERR(chip))
+		return -ENOMEM;
+
+	ddata = chip_to_ocores(chip);
+	ddata->data = id->data;
+	chip->ops = &ocores_pwm_ops;
+
+	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");
+
+	clk = devm_clk_get_enabled(dev, NULL);
+	if (IS_ERR(clk))
+		return dev_err_probe(dev, PTR_ERR(clk),
+				     "Unable to get pwm's clock\n");
+
+	ret = devm_clk_rate_exclusive_get(dev, clk);
+	if (ret)
+		return ret;
+
+	rst = devm_reset_control_get_optional_exclusive(dev, NULL);
+	if (IS_ERR(rst))
+		return dev_err_probe(dev, PTR_ERR(rst),
+				     "Unable to get pwm's reset\n");
+
+	reset_control_deassert(rst);
+
+	ret = devm_add_action_or_reset(dev, ocores_pwm_reset_control_assert,

rst);

drivers/pwm/pwm-ocores.c new file +27 -0
+	if (ret)
+		return ret;
+
+	ddata->clk_rate = clk_get_rate(clk);
+	if (ddata->clk_rate > NSEC_PER_SEC)
+		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");
+
+	return 0;
+}
+
+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 <[email protected]>");
+MODULE_DESCRIPTION("OpenCores PTC PWM driver");
Quoted 3 lines

MODULE_LICENSE("GPL");
--
2.34.1

Hi Uwe,

Could you please help me review this patch series to see if there is anything that needs to be modified?

Thanks for taking time to review this patch series.

Best Regards,
William

#4 lore raw
In reply to William Qiu

Hello William,

Quoted 22 lines · William Qiu

Add driver for OpenCores PWM Controller. And add compatibility code which based on StarFive SoC.

---
 MAINTAINERS              |   7 ++
 drivers/pwm/Kconfig      |  12 ++
 drivers/pwm/Makefile     |   1 +
 drivers/pwm/pwm-ocores.c | 239 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 259 insertions(+)
 create mode 100644 drivers/pwm/pwm-ocores.c

[...]

drivers/pwm/pwm-ocores.c new file +2 -0
new file mode 100644
index 000000000000..c8f08aa14e44
@@ -0,0 +1,239 @@
+// SPDX-License-Identifier: GPL-2.0
+/*

drivers/pwm/pwm-ocores.c line 12

Quoted 10 lines
drivers/pwm/pwm-ocores.c new file +10 -0
+ * OpenCores PWM Driver
+ *
+ * https://opencores.org/projects/ptc
+ *
+ * Copyright (C) 2018-2023 StarFive Technology Co., Ltd.
+ *
+ * Limitations:
+ * - The hardware only supports 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.

Nitpick: s/ ns//

drivers/pwm/pwm-ocores.c line 14

Quoted 2 lines
drivers/pwm/pwm-ocores.c new file +2 -0
+ * - The hardware is set to a low level immediately when disabledThe hardware is set to
+ *   a low level immediately when disabled.

This sentence is duplicated somehow. Maybe also: s/hardware/output/ ?

drivers/pwm/pwm-ocores.c line 15

Quoted 1 line
drivers/pwm/pwm-ocores.c new file +1 -0
+ * - The hardware will have a conversion cycle when reconfiguring.

I don't understand that.

Quoted 21 lines
drivers/pwm/pwm-ocores.c new file +21 -0
+ */
+
+#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>
+
+/* OpenCores Register offsets */
+#define REG_OCPWM_CNTR    0x0
+#define REG_OCPWM_HRC     0x4
+#define REG_OCPWM_LRC     0x8
+#define REG_OCPWM_CTRL    0xC
+
+/* OCPWM_CTRL register bits*/
+#define REG_OCPWM_CNTR_EN      BIT(0)
+#define REG_OCPWM_CNTR_ECLK    BIT(1)

drivers/pwm/pwm-ocores.c line 46

Quoted 10 lines
drivers/pwm/pwm-ocores.c new file +10 -0
+#define REG_OCPWM_CNTR_NEC     BIT(2)
+#define REG_OCPWM_CNTR_OE      BIT(3)
+#define REG_OCPWM_CNTR_SIGNLE  BIT(4)
+#define REG_OCPWM_CNTR_INTE    BIT(5)
+#define REG_OCPWM_CNTR_INT     BIT(6)
+#define REG_OCPWM_CNTR_RST     BIT(7)
+#define REG_OCPWM_CNTR_CAPTE   BIT(8)
+
+struct ocores_pwm_device {
+	const struct ocores_pwm_data *data;

I admit I didn't try to compile this, but I wonder if this doesn't result in a compiler warning given that struct ocores_pwm_data is only defined below.

Quoted 17 lines
drivers/pwm/pwm-ocores.c new file +16 -0
+	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 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)

drivers/pwm/pwm-ocores.c line 125

Quoted 10 lines
drivers/pwm/pwm-ocores.c new file +10 -0
+		return -EINVAL;
+
+	period_data = mul_u64_u32_div(state->period, ddata->clk_rate, NSEC_PER_SEC);
+	if (!period_data)
+		return -EINVAL;
+
+	if (period_data > U32_MAX)
+		period_data = U32_MAX;
+
+	ocores_pwm_writel(ddata, pwm->hwpwm, 0x8, (u32)period_data);

s/0x8/REG_OCPWM_LRC/ ?

That cast can be dropped.

drivers/pwm/pwm-ocores.c line 133

Quoted 7 lines
drivers/pwm/pwm-ocores.c new file +7 -0
+	duty_data = mul_u64_u32_div(state->duty_cycle, ddata->clk_rate, NSEC_PER_SEC);
+	if (duty_data <= U32_MAX)
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_HRC, (u32)duty_data);
+	else if (duty_data > U32_MAX)
+		duty_data = U32_MAX;
+	else
+		return -EINVAL;

That looks wrong. If duty_data > U32_MAX you assign duty_data but don't reuse this variable later.

Quoted 44 lines
drivers/pwm/pwm-ocores.c new file +43 -0
+
+	ctrl_data = ocores_pwm_readl(ddata, pwm->hwpwm, REG_OCPWM_CTRL);
+	if (state->enabled)
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_CTRL,
+				  ctrl_data | REG_OCPWM_CNTR_EN | REG_OCPWM_CNTR_OE);
+	else
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_CTRL,
+				  ctrl_data & ~(REG_OCPWM_CNTR_EN | REG_OCPWM_CNTR_OE));
+
+	return 0;
+}
[...]
+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;
+	struct clk *clk;
+	struct reset_control *rst;
+	int ret;
+
+	id = of_match_device(ocores_pwm_of_match, dev);
+	if (!id)
+		return -EINVAL;
+
+	chip = devm_pwmchip_alloc(&pdev->dev, 8, sizeof(*ddata));
+	if (IS_ERR(chip))
+		return -ENOMEM;
+
+	ddata = chip_to_ocores(chip);
+	ddata->data = id->data;
+	chip->ops = &ocores_pwm_ops;
+
+	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");
+
+	clk = devm_clk_get_enabled(dev, NULL);
+	if (IS_ERR(clk))
+		return dev_err_probe(dev, PTR_ERR(clk),
+				     "Unable to get pwm's clock\n");
+

drivers/pwm/pwm-ocores.c line 209

Quoted 10 lines
drivers/pwm/pwm-ocores.c new file +10 -0
+	ret = devm_clk_rate_exclusive_get(dev, clk);
+	if (ret)
+		return ret;
+
+	rst = devm_reset_control_get_optional_exclusive(dev, NULL);
+	if (IS_ERR(rst))
+		return dev_err_probe(dev, PTR_ERR(rst),
+				     "Unable to get pwm's reset\n");
+
+	reset_control_deassert(rst);

I think you want to check the return value of reset_control_deassert().

drivers/pwm/pwm-ocores.c line 218

Quoted 8 lines
drivers/pwm/pwm-ocores.c new file +8 -0
+	ret = devm_add_action_or_reset(dev, ocores_pwm_reset_control_assert, rst);
+	if (ret)
+		return ret;
+
+	ddata->clk_rate = clk_get_rate(clk);
+	if (ddata->clk_rate > NSEC_PER_SEC)
+		return dev_err_probe(dev, ddata->clk_rate,
+				     "Unable to get clock's rate\n");

Best regards
Uwe

#5 lore raw
Quoted 42 lines

-----Original Message-----
From: Uwe Kleine-König <[email protected]>
Sent: 2024年8月7日 1:19
To: William Qiu <[email protected]>
Cc: [email protected]; [email protected]; Hal Feng <[email protected]>; Philipp Zabel <[email protected]>
Subject: Re: [PATCH v13] pwm: opencores: Add PWM driver support

Hello William,

Quoted 20 lines · William Qiu

Add driver for OpenCores PWM Controller. And add compatibility code which based on StarFive SoC.

---
 MAINTAINERS              |   7 ++
 drivers/pwm/Kconfig      |  12 ++
 drivers/pwm/Makefile     |   1 +
 drivers/pwm/pwm-ocores.c | 239
+++++++++++++++++++++++++++++++++++++++
 4 files changed, 259 insertions(+)
 create mode 100644 drivers/pwm/pwm-ocores.c [...] diff --git

a/drivers/pwm/pwm-ocores.c b/drivers/pwm/pwm-ocores.c new file mode 100644 index 000000000000..c8f08aa14e44

drivers/pwm/pwm-ocores.c new file +1 -0
@@ -0,0 +1,239 @@
+// SPDX-License-Identifier: GPL-2.0

drivers/pwm/pwm-ocores.c line 11

Quoted 10 lines
drivers/pwm/pwm-ocores.c new file +10 -0
+/*
+ * OpenCores PWM Driver
+ *
+ * https://opencores.org/projects/ptc
+ *
+ * Copyright (C) 2018-2023 StarFive Technology Co., Ltd.
+ *
+ * Limitations:
+ * - The hardware only supports inverted polarity.
+ * - The hardware minimum period / duty_cycle is (1 / pwm_apb clock

frequency) ns.

Quoted 5 lines

drivers/pwm/pwm-ocores.c line 12

Quoted 1 line
drivers/pwm/pwm-ocores.c new file +1 -0
+ * - The hardware maximum period / duty_cycle is (U32_MAX / pwm_apb

clock frequency) ns.

Nitpick: s/ ns//

Will update.

Quoted 6 lines

drivers/pwm/pwm-ocores.c line 13

Quoted 1 line
drivers/pwm/pwm-ocores.c new file +1 -0
+ * - The hardware is set to a low level immediately when disabledThe

hardware is set to

drivers/pwm/pwm-ocores.c line 14

Quoted 1 line
drivers/pwm/pwm-ocores.c new file +1 -0
+ *   a low level immediately when disabled.

This sentence is duplicated somehow. Maybe also: s/hardware/output/ ?

Will update.

Quoted 4 lines

drivers/pwm/pwm-ocores.c line 15

Quoted 1 line
drivers/pwm/pwm-ocores.c new file +1 -0
+ * - The hardware will have a conversion cycle when reconfiguring.

I don't understand that.
.

For example, after the PWM duty cycle is changed from 50% to 80%, it is not a direct change, but there is a conversion period. The waveform during the conversion period will vary depending on whether it is high or low when reconfigured.

Quoted 26 lines
Quoted 26 lines
drivers/pwm/pwm-ocores.c new file +26 -0
+ */
+
+#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>
+
+/* OpenCores Register offsets */
+#define REG_OCPWM_CNTR    0x0
+#define REG_OCPWM_HRC     0x4
+#define REG_OCPWM_LRC     0x8
+#define REG_OCPWM_CTRL    0xC
+
+/* OCPWM_CTRL register bits*/
+#define REG_OCPWM_CNTR_EN      BIT(0)
+#define REG_OCPWM_CNTR_ECLK    BIT(1)
+#define REG_OCPWM_CNTR_NEC     BIT(2)
+#define REG_OCPWM_CNTR_OE      BIT(3)
+#define REG_OCPWM_CNTR_SIGNLE  BIT(4)
+#define REG_OCPWM_CNTR_INTE    BIT(5)
+#define REG_OCPWM_CNTR_INT     BIT(6)
Quoted 9 lines

drivers/pwm/pwm-ocores.c line 46

Quoted 5 lines
drivers/pwm/pwm-ocores.c new file +5 -0
+#define REG_OCPWM_CNTR_RST     BIT(7)
+#define REG_OCPWM_CNTR_CAPTE   BIT(8)
+
+struct ocores_pwm_device {
+	const struct ocores_pwm_data *data;

I admit I didn't try to compile this, but I wonder if this doesn't result in a compiler warning given that struct ocores_pwm_data is only defined below.

Will update.

Quoted 19 lines
Quoted 8 lines
drivers/pwm/pwm-ocores.c new file +7 -0
+	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 int ocores_pwm_apply(struct pwm_chip *chip,

drivers/pwm/pwm-ocores.c

Quoted 10 lines
drivers/pwm/pwm-ocores.c new file +10 -0
+			    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;
+
+	period_data = mul_u64_u32_div(state->period, ddata->clk_rate,

NSEC_PER_SEC);

Quoted 10 lines

drivers/pwm/pwm-ocores.c line 125

Quoted 7 lines
drivers/pwm/pwm-ocores.c new file +7 -0
+	if (!period_data)
+		return -EINVAL;
+
+	if (period_data > U32_MAX)
+		period_data = U32_MAX;
+
+	ocores_pwm_writel(ddata, pwm->hwpwm, 0x8, (u32)period_data);

s/0x8/REG_OCPWM_LRC/ ?

Will update.

Quoted 7 lines

That cast can be dropped.

drivers/pwm/pwm-ocores.c

Quoted 1 line
drivers/pwm/pwm-ocores.c new file +1 -0
+	duty_data = mul_u64_u32_div(state->duty_cycle, ddata->clk_rate,

NSEC_PER_SEC);

drivers/pwm/pwm-ocores.c

Quoted 2 lines
drivers/pwm/pwm-ocores.c new file +2 -0
+	if (duty_data <= U32_MAX)
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_HRC,

(u32)duty_data);

Quoted 8 lines

drivers/pwm/pwm-ocores.c line 133

Quoted 4 lines
drivers/pwm/pwm-ocores.c new file +4 -0
+	else if (duty_data > U32_MAX)
+		duty_data = U32_MAX;
+	else
+		return -EINVAL;

That looks wrong. If duty_data > U32_MAX you assign duty_data but don't reuse this variable later.

Will update.

Quoted 50 lines

drivers/pwm/pwm-ocores.c

Quoted 2 lines
drivers/pwm/pwm-ocores.c new file +2 -0
+
+	ctrl_data = ocores_pwm_readl(ddata, pwm->hwpwm,

REG_OCPWM_CTRL);

drivers/pwm/pwm-ocores.c

Quoted 3 lines
drivers/pwm/pwm-ocores.c new file +3 -0
+	if (state->enabled)
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_CTRL,
+				  ctrl_data | REG_OCPWM_CNTR_EN |

REG_OCPWM_CNTR_OE);

drivers/pwm/pwm-ocores.c

Quoted 3 lines
drivers/pwm/pwm-ocores.c new file +3 -0
+	else
+		ocores_pwm_writel(ddata, pwm->hwpwm, REG_OCPWM_CTRL,
+				  ctrl_data & ~(REG_OCPWM_CNTR_EN |

REG_OCPWM_CNTR_OE));

Quoted 39 lines
drivers/pwm/pwm-ocores.c new file +38 -0
+
+	return 0;
+}
[...]
+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;
+	struct clk *clk;
+	struct reset_control *rst;
+	int ret;
+
+	id = of_match_device(ocores_pwm_of_match, dev);
+	if (!id)
+		return -EINVAL;
+
+	chip = devm_pwmchip_alloc(&pdev->dev, 8, sizeof(*ddata));
+	if (IS_ERR(chip))
+		return -ENOMEM;
+
+	ddata = chip_to_ocores(chip);
+	ddata->data = id->data;
+	chip->ops = &ocores_pwm_ops;
+
+	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");
+
+	clk = devm_clk_get_enabled(dev, NULL);
+	if (IS_ERR(clk))
+		return dev_err_probe(dev, PTR_ERR(clk),
+				     "Unable to get pwm's clock\n");
+
+	ret = devm_clk_rate_exclusive_get(dev, clk);
+	if (ret)
+		return ret;
+
Quoted 9 lines

drivers/pwm/pwm-ocores.c line 209

Quoted 6 lines
drivers/pwm/pwm-ocores.c new file +6 -0
+	rst = devm_reset_control_get_optional_exclusive(dev, NULL);
+	if (IS_ERR(rst))
+		return dev_err_probe(dev, PTR_ERR(rst),
+				     "Unable to get pwm's reset\n");
+
+	reset_control_deassert(rst);

I think you want to check the return value of reset_control_deassert().

Will add.

Quoted 12 lines

drivers/pwm/pwm-ocores.c

Quoted 1 line
drivers/pwm/pwm-ocores.c new file +1 -0
+	ret = devm_add_action_or_reset(dev, ocores_pwm_reset_control_assert,

rst);

drivers/pwm/pwm-ocores.c line 218

Quoted 7 lines
drivers/pwm/pwm-ocores.c new file +7 -0
+	if (ret)
+		return ret;
+
+	ddata->clk_rate = clk_get_rate(clk);
+	if (ddata->clk_rate > NSEC_PER_SEC)
+		return dev_err_probe(dev, ddata->clk_rate,
+				     "Unable to get clock's rate\n");

Best regards
Uwe

#6 lore raw

Hello William,

Quoted 4 lines · William Qiu
Quoted 4 lines

-----Original Message-----
From: Uwe Kleine-König <[email protected]> [...]
On Tue, Jul 02, 2024 at 04:38:48PM +0800, William Qiu wrote:

Quoted 8 lines
Quoted 4 lines
Quoted 1 line

+ * - The hardware will have a conversion cycle when reconfiguring.

I don't understand that.
.

For example, after the PWM duty cycle is changed from 50% to 80%, it is not a direct change, but there is a conversion period. The waveform during the conversion period will vary depending on whether it is high or low when reconfigured.

Is it "just" that the new settings become active immediately and so if you change the relative duty_cycle from 50% to 80% and the current period is "done" between 50% and 80% you get three instead of one level change during that period?

If it's that, I'd describe that as: When configuration changes are done, they get active immediately without resetting the counter. This might result in one period affected by both old and new settings.

Best regards
Uwe

#7 lore raw
Quoted 30 lines

-----Original Message-----
From: Uwe Kleine-König <[email protected]>
Sent: 2024年8月7日 13:57
To: William Qiu <[email protected]>
Cc: [email protected]; [email protected]; Hal Feng <[email protected]>; Philipp Zabel <[email protected]>
Subject: Re: [PATCH v13] pwm: opencores: Add PWM driver support

Hello William,

Quoted 11 lines · William Qiu
Quoted 7 lines

-----Original Message-----
From: Uwe Kleine-König <[email protected]> [...] On Tue,

Quoted 1 line · William Qiu

+ * - The hardware will have a conversion cycle when reconfiguring.

I don't understand that.
.

For example, after the PWM duty cycle is changed from 50% to 80%, it is not a direct change, but there is a conversion period. The waveform during the conversion period will vary depending on whether it is high or low when reconfigured.

Is it "just" that the new settings become active immediately and so if you change the relative duty_cycle from 50% to 80% and the current period is "done" between 50% and 80% you get three instead of one level change during that period?

Quoted 6 lines

If it's that, I'd describe that as: When configuration changes are done, they get active immediately without resetting the counter. This might result in one period affected by both old and new settings.

Best regards
Uwe

Fine, I'll describe just like that.

Best Regards,
William