Skip to contentSkip to conversation

Polarfire SoC GPIO support

PATCH Superseded Conor Dooley 2 participants 11 messages Oct 31, 2024 – Nov 1, 2024

Subsystems (2)

v9 of 10 Superseded Latest is v10 →

lore
Share & tools
lore
#0 lore raw

v9 cover Polarfire SoC GPIO support

From: Conor Dooley <[email protected]>

Hey all,

Lewis is no longer at Microchip, so I've taken over the GPIO controller patchset that he had been working on prior to that:
https://lore.kernel.org/linux-gpio/[email protected]/

One thing that was wrong with Lewis' series was that it could only, depending on the iteration of the series, support GPIOs that had their interrupts muxed or GPIOs that had dedicated interrupts at the parent interrupt controller. I found that to be problematic, because the hardware itself always has a mix of muxed and dedicated interrupts and so there was always a controller rendered unusable for interrupts. I attempted to fix this by remodelling how the interrupt hierarchy in the devicetree is described, with a mux added between the GPIO controllers and the platform's interrupt controller. v7 introduced an irqchip driver for the mux between the GPIO controllers and PLIC to handle that problem.

After some discussion with Linus on v7, I've opted to strip out the interrupt handling entirely, in order to upstream this piecemeal. Interrupt controller support will be added at a later date, when I've sorted out the bits that Thomas did not approve of.

The binding for this GPIO controller is already upstream, so there's just one patch here now.

Cheers,
Conor.

v7: https://lore.kernel.org/linux-gpio/20240723-supervise-drown-d5d3b303e7fd@wendy/

CC Daire McNamara <[email protected]>
CC Linus Walleij <[email protected]>
CC Bartosz Golaszewski <[email protected]>

Lewis Hanly (1):
gpio: mpfs: add polarfire soc gpio support

 drivers/gpio/Kconfig     |   6 ++
 drivers/gpio/Makefile    |   1 +
 drivers/gpio/gpio-mpfs.c | 170 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 177 insertions(+)
 create mode 100644 drivers/gpio/gpio-mpfs.c
Signature · 1 line
2.45.2
#1 lore raw

PATCH v9 1/1 gpio: mpfs: add polarfire soc gpio support

From: Lewis Hanly <[email protected]>

Add a driver to support the Polarfire SoC gpio controller. Interrupt controller support is unavailable for now and will be added at a later date.

---
v9:
- remove remove()
- drop OF_GPIO from kconfig
- add a prefix to MAX_NUM_GPIO

v8:
- drop interrupt support
- replace regular mmio acesses with regmap (nice complexity reduction)
- change mpfs_gpio_get() so that it can report non-zero when the line direction is output.

---
 drivers/gpio/Kconfig     |   6 ++
 drivers/gpio/Makefile    |   1 +
 drivers/gpio/gpio-mpfs.c | 170 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 177 insertions(+)
 create mode 100644 drivers/gpio/gpio-mpfs.c
drivers/gpio/Kconfig +6 -0
index d93cd4f722b40..811263b033c89 100644
@@ -549,6 +549,12 @@ config GPIO_PL061
 	help
 	  Say yes here to support the PrimeCell PL061 GPIO device.
 
+config GPIO_POLARFIRE_SOC
+	bool "Microchip FPGA GPIO support"
+	select REGMAP_MMIO
+	help
+	  Say yes here to support the GPIO controllers on Microchip FPGAs.
+
 config GPIO_PXA
 	bool "PXA GPIO support"
 	depends on ARCH_PXA || ARCH_MMP || COMPILE_TEST
drivers/gpio/Makefile +1 -0
index 1429e8c0229b9..fc66e6388c76c 100644
@@ -133,6 +133,7 @@ obj-$(CONFIG_GPIO_PCI_IDIO_16)		+= gpio-pci-idio-16.o
 obj-$(CONFIG_GPIO_PISOSR)		+= gpio-pisosr.o
 obj-$(CONFIG_GPIO_PL061)		+= gpio-pl061.o
 obj-$(CONFIG_GPIO_PMIC_EIC_SPRD)	+= gpio-pmic-eic-sprd.o
+obj-$(CONFIG_GPIO_POLARFIRE_SOC)	+= gpio-mpfs.o
 obj-$(CONFIG_GPIO_PXA)			+= gpio-pxa.o
 obj-$(CONFIG_GPIO_RASPBERRYPI_EXP)	+= gpio-raspberrypi-exp.o
 obj-$(CONFIG_GPIO_RC5T583)		+= gpio-rc5t583.o
drivers/gpio/gpio-mpfs.c new file +170 -0

170 lines not on this page · raw message

Signature · 1 line
2.45.2
#2 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 66 lines · Conor Dooley

From: Lewis Hanly <[email protected]>

Add a driver to support the Polarfire SoC gpio controller. Interrupt controller support is unavailable for now and will be added at a later date.

---
v9:
- remove remove()
- drop OF_GPIO from kconfig
- add a prefix to MAX_NUM_GPIO

v8:
- drop interrupt support
- replace regular mmio acesses with regmap (nice complexity reduction)
- change mpfs_gpio_get() so that it can report non-zero when the line direction is output.

---
 drivers/gpio/Kconfig     |   6 ++
 drivers/gpio/Makefile    |   1 +
 drivers/gpio/gpio-mpfs.c | 170 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 177 insertions(+)
 create mode 100644 drivers/gpio/gpio-mpfs.c
drivers/gpio/Kconfig +6 -0
index d93cd4f722b40..811263b033c89 100644
@@ -549,6 +549,12 @@ config GPIO_PL061
        help
          Say yes here to support the PrimeCell PL061 GPIO device.
 
+config GPIO_POLARFIRE_SOC
+       bool "Microchip FPGA GPIO support"
+       select REGMAP_MMIO
+       help
+         Say yes here to support the GPIO controllers on Microchip FPGAs.
+
 config GPIO_PXA
        bool "PXA GPIO support"
        depends on ARCH_PXA || ARCH_MMP || COMPILE_TEST
drivers/gpio/Makefile +1 -0
index 1429e8c0229b9..fc66e6388c76c 100644
@@ -133,6 +133,7 @@ obj-$(CONFIG_GPIO_PCI_IDIO_16)              += gpio-pci-idio-16.o
 obj-$(CONFIG_GPIO_PISOSR)              += gpio-pisosr.o
 obj-$(CONFIG_GPIO_PL061)               += gpio-pl061.o
 obj-$(CONFIG_GPIO_PMIC_EIC_SPRD)       += gpio-pmic-eic-sprd.o
+obj-$(CONFIG_GPIO_POLARFIRE_SOC)       += gpio-mpfs.o
 obj-$(CONFIG_GPIO_PXA)                 += gpio-pxa.o
 obj-$(CONFIG_GPIO_RASPBERRYPI_EXP)     += gpio-raspberrypi-exp.o
 obj-$(CONFIG_GPIO_RC5T583)             += gpio-rc5t583.o
drivers/gpio/gpio-mpfs.c new file +3 -0
new file mode 100644
index 0000000000000..e5017307b0fe6
@@ -0,0 +1,170 @@
+// SPDX-License-Identifier: (GPL-2.0)
+/*
+ * Microchip PolarFire SoC (MPFS) GPIO controller driver

drivers/gpio/gpio-mpfs.c line 13

Quoted 10 lines
drivers/gpio/gpio-mpfs.c new file +10 -0
+ *
+ * Copyright (c) 2018-2024 Microchip Technology Inc. and its subsidiaries
+ */
+
+#include <linux/clk.h>
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/gpio/driver.h>
+#include <linux/init.h>
+#include <linux/of.h>

Sorry for noticing it only now but you don't use any symbol from this header so it can be dropped.

Quoted 15 lines
drivers/gpio/gpio-mpfs.c new file +15 -0
+#include <linux/mod_devicetable.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/spinlock.h>
+
+#define MPFS_GPIO_CTRL(i)              (0x4 * (i))
+#define MPFS_MAX_NUM_GPIO              32
+#define MPFS_GPIO_EN_INT               3
+#define MPFS_GPIO_EN_OUT_BUF           BIT(2)
+#define MPFS_GPIO_EN_IN                        BIT(1)
+#define MPFS_GPIO_EN_OUT               BIT(0)
+#define MPFS_GPIO_DIR_MASK             GENMASK(2, 0)
+
+#define MPFS_GPIO_TYPE_INT_EDGE_BOTH   0x80
+#define MPFS_GPIO_TYPE_INT_EDGE_NEG    0x60

drivers/gpio/gpio-mpfs.c line 38

Quoted 10 lines
drivers/gpio/gpio-mpfs.c new file +10 -0
+#define MPFS_GPIO_TYPE_INT_EDGE_POS    0x40
+#define MPFS_GPIO_TYPE_INT_LEVEL_LOW   0x20
+#define MPFS_GPIO_TYPE_INT_LEVEL_HIGH  0x00
+#define MPFS_GPIO_TYPE_INT_MASK                GENMASK(7, 5)
+#define MPFS_IRQ_REG                   0x80
+#define MPFS_INP_REG                   0x84
+#define MPFS_OUTP_REG                  0x88
+
+struct mpfs_gpio_chip {
+       struct clk *clk;

After the initial clk_get() you never reference it again so there's no need to store the address.

Quoted 96 lines
drivers/gpio/gpio-mpfs.c new file +96 -0
+       struct regmap *regs;
+       struct gpio_chip gc;
+};
+
+static const struct regmap_config mpfs_gpio_regmap_config = {
+       .reg_bits = 32,
+       .reg_stride = 4,
+       .val_bits = 32,
+};
+
+static int mpfs_gpio_direction_input(struct gpio_chip *gc, unsigned int gpio_index)
+{
+       struct mpfs_gpio_chip *mpfs_gpio = gpiochip_get_data(gc);
+
+       regmap_update_bits(mpfs_gpio->regs, MPFS_GPIO_CTRL(gpio_index),
+                          MPFS_GPIO_DIR_MASK, MPFS_GPIO_EN_IN);
+
+       return 0;
+}
+
+static int mpfs_gpio_direction_output(struct gpio_chip *gc, unsigned int gpio_index, int value)
+{
+       struct mpfs_gpio_chip *mpfs_gpio = gpiochip_get_data(gc);
+
+       regmap_update_bits(mpfs_gpio->regs, MPFS_GPIO_CTRL(gpio_index),
+                          MPFS_GPIO_DIR_MASK, MPFS_GPIO_EN_IN);
+       regmap_update_bits(mpfs_gpio->regs, MPFS_OUTP_REG, BIT(gpio_index),
+                          value << gpio_index);
+
+       return 0;
+}
+
+static int mpfs_gpio_get_direction(struct gpio_chip *gc,
+                                  unsigned int gpio_index)
+{
+       struct mpfs_gpio_chip *mpfs_gpio = gpiochip_get_data(gc);
+       unsigned int gpio_cfg;
+
+       regmap_read(mpfs_gpio->regs, MPFS_GPIO_CTRL(gpio_index), &gpio_cfg);
+       if (gpio_cfg & MPFS_GPIO_EN_IN)
+               return GPIO_LINE_DIRECTION_IN;
+
+       return GPIO_LINE_DIRECTION_OUT;
+}
+
+static int mpfs_gpio_get(struct gpio_chip *gc, unsigned int gpio_index)
+{
+       struct mpfs_gpio_chip *mpfs_gpio = gpiochip_get_data(gc);
+
+       if (mpfs_gpio_get_direction(gc, gpio_index) == GPIO_LINE_DIRECTION_OUT)
+               return regmap_test_bits(mpfs_gpio->regs, MPFS_OUTP_REG, BIT(gpio_index));
+       else
+               return regmap_test_bits(mpfs_gpio->regs, MPFS_INP_REG, BIT(gpio_index));
+}
+
+static void mpfs_gpio_set(struct gpio_chip *gc, unsigned int gpio_index, int value)
+{
+       struct mpfs_gpio_chip *mpfs_gpio = gpiochip_get_data(gc);
+
+       mpfs_gpio_get(gc, gpio_index);
+
+       regmap_update_bits(mpfs_gpio->regs, MPFS_OUTP_REG, BIT(gpio_index),
+                          value << gpio_index);
+
+       mpfs_gpio_get(gc, gpio_index);
+}
+
+static int mpfs_gpio_probe(struct platform_device *pdev)
+{
+       struct device *dev = &pdev->dev;
+       struct mpfs_gpio_chip *mpfs_gpio;
+       struct clk *clk;
+       void __iomem *base;
+       int ret, ngpios;
+
+       mpfs_gpio = devm_kzalloc(dev, sizeof(*mpfs_gpio), GFP_KERNEL);
+       if (!mpfs_gpio)
+               return -ENOMEM;
+
+       base = devm_platform_ioremap_resource(pdev, 0);
+       if (IS_ERR(base))
+               return dev_err_probe(dev, PTR_ERR(base), "failed to ioremap memory resource\n");
+
+       mpfs_gpio->regs = devm_regmap_init_mmio(dev, base, &mpfs_gpio_regmap_config);
+       if (IS_ERR(mpfs_gpio->regs))
+               return dev_err_probe(dev, PTR_ERR(mpfs_gpio->regs),
+                                    "failed to initialise regmap\n");
+
+       clk = devm_clk_get_enabled(dev, NULL);
+       if (IS_ERR(clk))
+               return dev_err_probe(dev, PTR_ERR(clk), "failed to get and enable clock\n");
+
+       mpfs_gpio->clk = clk;
+
+       ngpios = MPFS_MAX_NUM_GPIO;
+       device_property_read_u32(dev, "ngpios", &ngpios);

drivers/gpio/gpio-mpfs.c line 144

Quoted 10 lines
drivers/gpio/gpio-mpfs.c new file +10 -0
+       if (ngpios > MPFS_MAX_NUM_GPIO)
+               ngpios = MPFS_MAX_NUM_GPIO;
+
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

drivers/gpio/gpio-mpfs.c line 153

Quoted 9 lines
drivers/gpio/gpio-mpfs.c new file +9 -0
+       mpfs_gpio->gc.label = dev_name(dev);
+       mpfs_gpio->gc.parent = dev;
+       mpfs_gpio->gc.owner = THIS_MODULE;
+
+       ret = devm_gpiochip_add_data(dev, &mpfs_gpio->gc, mpfs_gpio);
+       if (ret)
+               return ret;
+
+       platform_set_drvdata(pdev, mpfs_gpio);

There's no platform_get_drvdata() in the driver so you can drop this too.

Bart

Quoted 20 lines
drivers/gpio/gpio-mpfs.c new file +17 -1
+
+       return 0;
+}
+
+static const struct of_device_id mpfs_gpio_of_ids[] = {
+       { .compatible = "microchip,mpfs-gpio", },
+       { /* end of list */ }
+};
+
+static struct platform_driver mpfs_gpio_driver = {
+       .probe = mpfs_gpio_probe,
+       .driver = {
+               .name = "microchip,mpfs-gpio",
+               .of_match_table = mpfs_gpio_of_ids,
+       },
+};
+builtin_platform_driver(mpfs_gpio_driver);
--

2.45.2

#3 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 11 lines · Bartosz Golaszewski

drivers/gpio/gpio-mpfs.c line 144

Quoted 7 lines · Conor Dooley
drivers/gpio/gpio-mpfs.c new file +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

#4 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 14 lines · Conor Dooley
Quoted 11 lines · Bartosz Golaszewski

drivers/gpio/gpio-mpfs.c line 144

Quoted 7 lines · Conor Dooley
drivers/gpio/gpio-mpfs.c new file +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Quoted 8 lines

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

Does dev have an fwnode correctly assigned? What does dev_fwnode(dev) return?

Bart

#5 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 19 lines · Bartosz Golaszewski
Quoted 18 lines · Conor Dooley
Quoted 11 lines · Bartosz Golaszewski

drivers/gpio/gpio-mpfs.c line 144

Quoted 7 lines · Conor Dooley
drivers/gpio/gpio-mpfs.c new file +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

Quoted 6 lines
Quoted 4 lines

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

Does dev have an fwnode correctly assigned? What does dev_fwnode(dev) return?

It's not a null pointer or something obviously wrong by virtue of being null-adjacent, it is a virtual address but not one that %ps can identify.

#6 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 26 lines · Conor Dooley
Quoted 24 lines · Bartosz Golaszewski
Quoted 22 lines · Conor Dooley
Quoted 11 lines · Bartosz Golaszewski

drivers/gpio/gpio-mpfs.c line 144

Quoted 7 lines · Conor Dooley
drivers/gpio/gpio-mpfs.c new file +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

Quoted 4 lines
Quoted 1 line

Does dev have an fwnode correctly assigned? What does dev_fwnode(dev) return?

It's not a null pointer or something obviously wrong by virtue of being null-adjacent, it is a virtual address but not one that %ps can identify.

This is an OF system right? If you do dev_of_node(dev), does the returned node->name show the OF node you expect? Does it have the "ngpios" property? Can you read it with of_property_read_u32()?

Bart

#7 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 27 lines · Bartosz Golaszewski
Quoted 26 lines · Conor Dooley
Quoted 24 lines · Bartosz Golaszewski
Quoted 22 lines · Conor Dooley
Quoted 11 lines · Bartosz Golaszewski

drivers/gpio/gpio-mpfs.c line 144

Quoted 7 lines · Conor Dooley
drivers/gpio/gpio-mpfs.c new file +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

Does dev have an fwnode correctly assigned? What does dev_fwnode(dev) return?

It's not a null pointer or something obviously wrong by virtue of being null-adjacent, it is a virtual address but not one that %ps can identify.

This is an OF system right? If you do dev_of_node(dev), does the returned node->name show the OF node you expect?

Yes.

Quoted 2 lines

Does it have the
"ngpios" property? Can you read it with of_property_read_u32()?

No, EINVAL there too.

#8 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 31 lines · Conor Dooley
Quoted 29 lines · Bartosz Golaszewski
Quoted 28 lines · Conor Dooley
Quoted 25 lines · Bartosz Golaszewski
Quoted 14 lines · Conor Dooley
Quoted 11 lines · Bartosz Golaszewski

Review

Quoted 7 lines · Conor Dooley
(quoted diff) +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Quoted 8 lines

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

Does dev have an fwnode correctly assigned? What does dev_fwnode(dev) return?

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

It's not a null pointer or something obviously wrong by virtue of being null-adjacent, it is a virtual address but not one that %ps can identify.

This is an OF system right? If you do dev_of_node(dev), does the returned node->name show the OF node you expect?

Yes.

I mean inside gpiochip_get_ngpios(), sorry for confusion.

Quoted 5 lines
Quoted 3 lines

Does it have the
"ngpios" property? Can you read it with of_property_read_u32()?

No, EINVAL there too.

I assume the node is not assigned correctly. What if in your probe you
do: gc->fwnode = dev_fwnode(dev)?

Bart

#9 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 35 lines · Bartosz Golaszewski
Quoted 34 lines · Conor Dooley
Quoted 32 lines · Bartosz Golaszewski
Quoted 30 lines · Conor Dooley
Quoted 19 lines · Bartosz Golaszewski
Quoted 18 lines · Conor Dooley
Quoted 11 lines · Bartosz Golaszewski

Review

Quoted 7 lines · Conor Dooley
(quoted diff) +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

Quoted 6 lines
Quoted 4 lines

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

Does dev have an fwnode correctly assigned? What does dev_fwnode(dev) return?

It's not a null pointer or something obviously wrong by virtue of being null-adjacent, it is a virtual address but not one that %ps can identify.

Quoted 6 lines
Quoted 4 lines
Quoted 2 lines

This is an OF system right? If you do dev_of_node(dev), does the returned node->name show the OF node you expect?

Yes.

I mean inside gpiochip_get_ngpios(), sorry for confusion.

That is what I checked actually, didn't think you'd ask me to check the one in probe that works.

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

Does it have the
"ngpios" property? Can you read it with of_property_read_u32()?

No, EINVAL there too.

I assume the node is not assigned correctly. What if in your probe you
do: gc->fwnode = dev_fwnode(dev)?

Makes no difference, same probe failure as before...

...but I just realised something: ngpios isn't a required property for this device as it has a default of 32 and 32 is how many gpios this controller has. Simple oversight, hours wasted. That's always the way of it I suppose. The core code can't be used here I suppose, since ngpios is optional.

#10 lore raw

Re: [PATCH v9 1/1] gpio: mpfs: add polarfire soc gpio support

Quoted 55 lines · Conor Dooley
Quoted 35 lines · Bartosz Golaszewski
Quoted 34 lines · Conor Dooley
Quoted 32 lines · Bartosz Golaszewski
Quoted 30 lines · Conor Dooley
Quoted 19 lines · Bartosz Golaszewski
Quoted 18 lines · Conor Dooley
Quoted 11 lines · Bartosz Golaszewski

Review

Quoted 7 lines · Conor Dooley
(quoted diff) +7 -0
+       mpfs_gpio->gc.direction_input = mpfs_gpio_direction_input;
+       mpfs_gpio->gc.direction_output = mpfs_gpio_direction_output;
+       mpfs_gpio->gc.get_direction = mpfs_gpio_get_direction;
+       mpfs_gpio->gc.get = mpfs_gpio_get;
+       mpfs_gpio->gc.set = mpfs_gpio_set;
+       mpfs_gpio->gc.base = -1;
+       mpfs_gpio->gc.ngpio = ngpios;

The "ngpios" property will be parsed by GPIO core so no need to set it.

Are you sure it'll work here? I tried dropping the ngpios parsing, but I get:
gpiochip_add_data_with_key: GPIOs 0..-1 (20122000.gpio) failed to register, -22

Quoted 6 lines
Quoted 4 lines

That's coming from the device_property_read_u32(dev, "ngpios", &ngpios); in gpiochip_get_ngpios(). Checking against the bluefield driver and the code in gpiochip_add_data_with_key(), it's not immediately obvious what I am missing.

Does dev have an fwnode correctly assigned? What does dev_fwnode(dev) return?

It's not a null pointer or something obviously wrong by virtue of being null-adjacent, it is a virtual address but not one that %ps can identify.

Quoted 6 lines
Quoted 4 lines
Quoted 2 lines

This is an OF system right? If you do dev_of_node(dev), does the returned node->name show the OF node you expect?

Yes.

I mean inside gpiochip_get_ngpios(), sorry for confusion.

That is what I checked actually, didn't think you'd ask me to check the one in probe that works.

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

Does it have the
"ngpios" property? Can you read it with of_property_read_u32()?

No, EINVAL there too.

I assume the node is not assigned correctly. What if in your probe you
do: gc->fwnode = dev_fwnode(dev)?

Quoted 8 lines

Makes no difference, same probe failure as before...

...but I just realised something: ngpios isn't a required property for this device as it has a default of 32 and 32 is how many gpios this controller has. Simple oversight, hours wasted. That's always the way of it I suppose. The core code can't be used here I suppose, since ngpios is optional.

Ah right. If we get to gpiochip_get_ngpios() without a GPIO number set, the property becomes mandatory.

Nevermind my comment from the review then.

Bart