[PATCH v4 0/2] gpio: brcm: XGS iProc GPIO driver

[PATCH v4 0/2] gpio: brcm: XGS iProc GPIO driver

[Chris Packham]

This is ported this from Broadcom's XLDK (now heavily modified). There seem to
be 3 different IP blocks for 3 separate banks of GPIOs in the iProc chips.

I've dropped everything except support for the Chip Common A GPIO
controller because the other blocks actually seem to be supportable with
other drivers. The driver itself is halfway between pinctrl-nsp-gpio.c
and pinctrl-iproc-gpio.c.

Chris Packham (2):
  dt-bindings: gpio: brcm: Add bindings for xgs-iproc
  gpio: Add xgs-iproc driver

 .../bindings/gpio/brcm,xgs-iproc-gpio.yaml    |  70 ++++
 drivers/gpio/Kconfig                          |   9 +
 drivers/gpio/Makefile                         |   1 +
 drivers/gpio/gpio-xgs-iproc.c                 | 321 ++++++++++++++++++
 4 files changed, 401 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
 create mode 100644 drivers/gpio/gpio-xgs-iproc.c

-- 
2.23.0

[PATCH v4 1/2] dt-bindings: gpio: brcm: Add bindings for xgs-iproc

[Chris Packham]

This GPIO controller is present on a number of Broadcom switch ASICs
with integrated SoCs. It is similar to the nsp-gpio and iproc-gpio
blocks but different enough to require a separate driver.

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
---

Notes:
    Changes in v4:
    - rename brcm,xgs-iproc.yaml -> brcm,xgs-iproc-gpio.yaml as suggested
    
    Changes in v3:
    - incorporate review comments from Rob and Bart
    
    Changes in v2:
    - Document as DT schema
    - Include ngpios, #gpio-cells and gpio-controller properties

 .../bindings/gpio/brcm,xgs-iproc-gpio.yaml    | 70 +++++++++++++++++++
 1 file changed, 70 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml

diff --git a/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml b/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
new file mode 100644
index 000000000000..ec1fd3a64aa2
--- /dev/null
+++ b/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
@@ -0,0 +1,70 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/gpio/brcm,xgs-iproc.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Broadcom XGS iProc GPIO controller
+
+maintainers:
+  - Chris Packham <chris.packham@alliedtelesis.co.nz>
+
+description: |
+  This controller is the Chip Common A GPIO present on a number of Broadcom
+  switch ASICs with integrated SoCs.
+
+properties:
+  compatible:
+    const: brcm,iproc-gpio-cca
+
+  reg:
+    items:
+      - description: the I/O address containing the GPIO controller
+                     registers.
+      - description: the I/O address containing the Chip Common A interrupt
+                     registers.
+
+  gpio-controller: true
+
+  '#gpio-cells':
+      const: 2
+
+  ngpios:
+    minimum: 0
+    maximum: 32
+
+  interrupt-controller: true
+
+  '#interrupt-cells':
+    const: 2
+
+  interrupts:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+  - "#gpio-cells"
+  - gpio-controller
+
+dependencies:
+  interrupt-controller: [ interrupts ]
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    gpio@18000060 {
+        compatible = "brcm,iproc-gpio-cca";
+        #gpio-cells = <2>;
+        reg = <0x18000060 0x50>,
+              <0x18000000 0x50>;
+        ngpios = <12>;
+        gpio-controller;
+        interrupt-controller;
+        #interrupt-cells = <2>;
+        interrupts = <GIC_SPI 91 IRQ_TYPE_LEVEL_HIGH>;
+    };
+
+
+...
-- 
2.23.0

[PATCH v4 2/2] gpio: Add xgs-iproc driver

[Chris Packham]

This driver supports the Chip Common A GPIO controller present on a
number of Broadcom switch ASICs with integrated SoCs. The controller is
similar to the pinctrl-nsp-gpio and pinctrl-iproc-gpio blocks but
different enough that a separate driver is required.

This has been ported from Broadcom's XLDK 5.0.3 retaining only the CCA
support (pinctrl-iproc-gpio covers CCB).

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
---

Notes:
    Changes in v4:
    - rename the config option to GPIO_BCM_XGS_IPROC and place alphabetically
    - sort #includes alphabetically
    
    Changes in v3:
    - prefix local #defines with 'IPROC'
    - use {readl,writel}_relaxed
    - remove unnecessary headers
    - actually use spinlock to guard hardware accesses
    
    Changes in v2:
    - use more of the generic infrastructure for gpio chips
    - handling the root interrupt is still done manually due to sharing with uart0.

 drivers/gpio/Kconfig          |   9 +
 drivers/gpio/Makefile         |   1 +
 drivers/gpio/gpio-xgs-iproc.c | 321 ++++++++++++++++++++++++++++++++++
 3 files changed, 331 insertions(+)
 create mode 100644 drivers/gpio/gpio-xgs-iproc.c

diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index 38e096e6925f..04396787fbb0 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -147,6 +147,15 @@ config GPIO_BCM_KONA
 	help
 	  Turn on GPIO support for Broadcom "Kona" chips.
 
+config GPIO_BCM_XGS_IPROC
+	tristate "BRCM XGS iProc GPIO support"
+	depends on OF_GPIO && (ARCH_BCM_IPROC || COMPILE_TEST)
+	select GPIO_GENERIC
+	select GPIOLIB_IRQCHIP
+	default ARCH_BCM_IPROC
+	help
+	  Say yes here to enable GPIO support for Broadcom XGS iProc SoCs.
+
 config GPIO_BRCMSTB
 	tristate "BRCMSTB GPIO support"
 	default y if (ARCH_BRCMSTB || BMIPS_GENERIC)
diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile
index d2fd19c15bae..8725d158a964 100644
--- a/drivers/gpio/Makefile
+++ b/drivers/gpio/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_GPIO_ARIZONA)		+= gpio-arizona.o
 obj-$(CONFIG_GPIO_ASPEED)		+= gpio-aspeed.o
 obj-$(CONFIG_GPIO_ATH79)		+= gpio-ath79.o
 obj-$(CONFIG_GPIO_BCM_KONA)		+= gpio-bcm-kona.o
+obj-$(CONFIG_GPIO_BCM_XGS_IPROC)	+= gpio-xgs-iproc.o
 obj-$(CONFIG_GPIO_BD70528)		+= gpio-bd70528.o
 obj-$(CONFIG_GPIO_BD9571MWV)		+= gpio-bd9571mwv.o
 obj-$(CONFIG_GPIO_BRCMSTB)		+= gpio-brcmstb.o
diff --git a/drivers/gpio/gpio-xgs-iproc.c b/drivers/gpio/gpio-xgs-iproc.c
new file mode 100644
index 000000000000..a3fdd95cc9e6
--- /dev/null
+++ b/drivers/gpio/gpio-xgs-iproc.c
@@ -0,0 +1,321 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2017 Broadcom
+ */
+
+#include <linux/gpio/driver.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+
+#define IPROC_CCA_INT_F_GPIOINT		BIT(0)
+#define IPROC_CCA_INT_STS		0x20
+#define IPROC_CCA_INT_MASK		0x24
+
+#define IPROC_GPIO_CCA_DIN		0x0
+#define IPROC_GPIO_CCA_DOUT		0x4
+#define IPROC_GPIO_CCA_OUT_EN		0x8
+#define IPROC_GPIO_CCA_INT_LEVEL	0x10
+#define IPROC_GPIO_CCA_INT_LEVEL_MASK	0x14
+#define IPROC_GPIO_CCA_INT_EVENT	0x18
+#define IPROC_GPIO_CCA_INT_EVENT_MASK	0x1C
+#define IPROC_GPIO_CCA_INT_EDGE		0x24
+
+struct iproc_gpio_chip {
+	struct irq_chip irqchip;
+	struct gpio_chip gc;
+	spinlock_t lock;
+	struct device *dev;
+	void __iomem *base;
+	void __iomem *intr;
+};
+
+static inline struct iproc_gpio_chip *
+to_iproc_gpio(struct gpio_chip *gc)
+{
+	return container_of(gc, struct iproc_gpio_chip, gc);
+}
+
+static void iproc_gpio_irq_ack(struct irq_data *d)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct iproc_gpio_chip *chip = to_iproc_gpio(gc);
+	int pin = d->hwirq;
+	unsigned long flags;
+	u32 irq = d->irq;
+	u32 irq_type, event_status = 0;
+
+	spin_lock_irqsave(&chip->lock, flags);
+	irq_type = irq_get_trigger_type(irq);
+	if (irq_type & IRQ_TYPE_EDGE_BOTH) {
+		event_status |= BIT(pin);
+		writel_relaxed(event_status,
+			       chip->base + IPROC_GPIO_CCA_INT_EVENT);
+	}
+	spin_unlock_irqrestore(&chip->lock, flags);
+}
+
+static void iproc_gpio_irq_unmask(struct irq_data *d)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct iproc_gpio_chip *chip = to_iproc_gpio(gc);
+	int pin = d->hwirq;
+	unsigned long flags;
+	u32 irq = d->irq;
+	u32 int_mask, irq_type, event_mask;
+
+	spin_lock_irqsave(&chip->lock, flags);
+	irq_type = irq_get_trigger_type(irq);
+	event_mask = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_EVENT_MASK);
+	int_mask = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_LEVEL_MASK);
+
+	if (irq_type & IRQ_TYPE_EDGE_BOTH) {
+		event_mask |= 1 << pin;
+		writel_relaxed(event_mask,
+			       chip->base + IPROC_GPIO_CCA_INT_EVENT_MASK);
+	} else {
+		int_mask |= 1 << pin;
+		writel_relaxed(int_mask,
+			       chip->base + IPROC_GPIO_CCA_INT_LEVEL_MASK);
+	}
+	spin_unlock_irqrestore(&chip->lock, flags);
+}
+
+static void iproc_gpio_irq_mask(struct irq_data *d)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct iproc_gpio_chip *chip = to_iproc_gpio(gc);
+	int pin = d->hwirq;
+	unsigned long flags;
+	u32 irq = d->irq;
+	u32 irq_type, int_mask, event_mask;
+
+	spin_lock_irqsave(&chip->lock, flags);
+	irq_type = irq_get_trigger_type(irq);
+	event_mask = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_EVENT_MASK);
+	int_mask = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_LEVEL_MASK);
+
+	if (irq_type & IRQ_TYPE_EDGE_BOTH) {
+		event_mask &= ~BIT(pin);
+		writel_relaxed(event_mask,
+			       chip->base + IPROC_GPIO_CCA_INT_EVENT_MASK);
+	} else {
+		int_mask &= ~BIT(pin);
+		writel_relaxed(int_mask,
+			       chip->base + IPROC_GPIO_CCA_INT_LEVEL_MASK);
+	}
+	spin_unlock_irqrestore(&chip->lock, flags);
+}
+
+static int iproc_gpio_irq_set_type(struct irq_data *d, u32 type)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct iproc_gpio_chip *chip = to_iproc_gpio(gc);
+	int pin = d->hwirq;
+	unsigned long flags;
+	u32 irq = d->irq;
+	u32 event_pol, int_pol;
+	int ret = 0;
+
+	spin_lock_irqsave(&chip->lock, flags);
+	switch (type & IRQ_TYPE_SENSE_MASK) {
+	case IRQ_TYPE_EDGE_RISING:
+		event_pol = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_EDGE);
+		event_pol &= ~BIT(pin);
+		writel_relaxed(event_pol, chip->base + IPROC_GPIO_CCA_INT_EDGE);
+		break;
+	case IRQ_TYPE_EDGE_FALLING:
+		event_pol = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_EDGE);
+		event_pol |= BIT(pin);
+		writel_relaxed(event_pol, chip->base + IPROC_GPIO_CCA_INT_EDGE);
+		break;
+	case IRQ_TYPE_LEVEL_HIGH:
+		int_pol = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_LEVEL);
+		int_pol &= ~BIT(pin);
+		writel_relaxed(int_pol, chip->base + IPROC_GPIO_CCA_INT_LEVEL);
+		break;
+	case IRQ_TYPE_LEVEL_LOW:
+		int_pol = readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_LEVEL);
+		int_pol |= BIT(pin);
+		writel_relaxed(int_pol, chip->base + IPROC_GPIO_CCA_INT_LEVEL);
+		break;
+	default:
+		/* should not come here */
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
+	if (type & IRQ_TYPE_LEVEL_MASK)
+		irq_set_handler_locked(irq_get_irq_data(irq), handle_level_irq);
+	else if (type & IRQ_TYPE_EDGE_BOTH)
+		irq_set_handler_locked(irq_get_irq_data(irq), handle_edge_irq);
+
+out_unlock:
+	spin_unlock_irqrestore(&chip->lock, flags);
+
+	return ret;
+}
+
+static irqreturn_t iproc_gpio_irq_handler(int irq, void *data)
+{
+	struct gpio_chip *gc = (struct gpio_chip *)data;
+	struct iproc_gpio_chip *chip = to_iproc_gpio(gc);
+	int bit;
+	unsigned long int_bits = 0;
+	u32 int_status;
+
+	/* go through the entire GPIOs and handle all interrupts */
+	int_status = readl_relaxed(chip->intr + IPROC_CCA_INT_STS);
+	if (int_status & IPROC_CCA_INT_F_GPIOINT) {
+		u32 event, level;
+
+		/* Get level and edge interrupts */
+		event =
+		    readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_EVENT_MASK);
+		event &= readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_EVENT);
+		level = readl_relaxed(chip->base + IPROC_GPIO_CCA_DIN);
+		level ^= readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_LEVEL);
+		level &=
+		    readl_relaxed(chip->base + IPROC_GPIO_CCA_INT_LEVEL_MASK);
+		int_bits = level | event;
+
+		for_each_set_bit(bit, &int_bits, gc->ngpio)
+			generic_handle_irq(irq_linear_revmap(gc->irq.domain, bit));
+	}
+
+	return int_bits ? IRQ_HANDLED : IRQ_NONE;
+}
+
+static int iproc_gpio_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *dn = pdev->dev.of_node;
+	struct iproc_gpio_chip *chip;
+	u32 num_gpios;
+	int irq, ret;
+
+	chip = devm_kzalloc(dev, sizeof(*chip), GFP_KERNEL);
+	if (!chip)
+		return -ENOMEM;
+
+	chip->dev = dev;
+	platform_set_drvdata(pdev, chip);
+	spin_lock_init(&chip->lock);
+
+	chip->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(chip->base))
+		return PTR_ERR(chip->base);
+
+	ret = bgpio_init(&chip->gc, dev, 4,
+			 chip->base + IPROC_GPIO_CCA_DIN,
+			 chip->base + IPROC_GPIO_CCA_DOUT,
+			 NULL,
+			 chip->base + IPROC_GPIO_CCA_OUT_EN,
+			 NULL,
+			 0);
+	if (ret) {
+		dev_err(dev, "unable to init GPIO chip\n");
+		return ret;
+	}
+
+	chip->gc.label = dev_name(dev);
+	if (of_property_read_u32(dn, "ngpios", &num_gpios))
+		chip->gc.ngpio = num_gpios;
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq > 0) {
+		struct gpio_irq_chip *girq;
+		struct irq_chip *irqc;
+		u32 val;
+
+		irqc = &chip->irqchip;
+		irqc->name = dev_name(dev);
+		irqc->irq_ack = iproc_gpio_irq_ack;
+		irqc->irq_mask = iproc_gpio_irq_mask;
+		irqc->irq_unmask = iproc_gpio_irq_unmask;
+		irqc->irq_set_type = iproc_gpio_irq_set_type;
+
+		chip->intr = devm_platform_ioremap_resource(pdev, 1);
+		if (IS_ERR(chip->intr))
+			return PTR_ERR(chip->intr);
+
+		/* Enable GPIO interrupts for CCA GPIO */
+		val = readl_relaxed(chip->intr + IPROC_CCA_INT_MASK);
+		val |= IPROC_CCA_INT_F_GPIOINT;
+		writel_relaxed(val, chip->intr + IPROC_CCA_INT_MASK);
+
+		/*
+		 * Directly request the irq here instead of passing
+		 * a flow-handler to gpiochip_set_chained_irqchip,
+		 * because the irq is shared.
+		 */
+		ret = devm_request_irq(dev, irq, iproc_gpio_irq_handler,
+				       IRQF_SHARED, chip->gc.label, &chip->gc);
+		if (ret) {
+			dev_err(dev, "Fail to request IRQ%d: %d\n", irq, ret);
+			return ret;
+		}
+
+		girq = &chip->gc.irq;
+		girq->chip = irqc;
+		/* This will let us handle the parent IRQ in the driver */
+		girq->parent_handler = NULL;
+		girq->num_parents = 0;
+		girq->parents = NULL;
+		girq->default_type = IRQ_TYPE_NONE;
+		girq->handler = handle_simple_irq;
+	}
+
+	ret = devm_gpiochip_add_data(dev, &chip->gc, chip);
+	if (ret) {
+		dev_err(dev, "unable to add GPIO chip\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int __exit iproc_gpio_remove(struct platform_device *pdev)
+{
+	struct iproc_gpio_chip *chip;
+
+	chip = platform_get_drvdata(pdev);
+	if (!chip)
+		return -ENODEV;
+
+	if (chip->intr) {
+		u32 val;
+
+		val = readl_relaxed(chip->intr + IPROC_CCA_INT_MASK);
+		val &= ~IPROC_CCA_INT_F_GPIOINT;
+		writel_relaxed(val, chip->intr + IPROC_CCA_INT_MASK);
+	}
+
+	return 0;
+}
+
+static const struct of_device_id bcm_iproc_gpio_of_match[] __initconst = {
+	{ .compatible = "brcm,iproc-gpio-cca" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, bcm_iproc_gpio_of_match);
+
+static struct platform_driver bcm_iproc_gpio_driver = {
+	.driver = {
+		.name = "iproc-xgs-gpio",
+		.owner = THIS_MODULE,
+		.of_match_table = bcm_iproc_gpio_of_match,
+	},
+	.probe = iproc_gpio_probe,
+	.remove = iproc_gpio_remove,
+};
+
+module_platform_driver(bcm_iproc_gpio_driver);
+
+MODULE_DESCRIPTION("XGS IPROC GPIO driver");
+MODULE_LICENSE("GPL v2");
-- 
2.23.0

Re: [PATCH v4 0/2] gpio: brcm: XGS iProc GPIO driver

[Scott Branden]

Patch Series looks ok.

On 2019-10-24 1:27 p.m., Chris Packham wrote:
> This is ported this from Broadcom's XLDK (now heavily modified). There seem to
> be 3 different IP blocks for 3 separate banks of GPIOs in the iProc chips.
>
> I've dropped everything except support for the Chip Common A GPIO
> controller because the other blocks actually seem to be supportable with
> other drivers. The driver itself is halfway between pinctrl-nsp-gpio.c
> and pinctrl-iproc-gpio.c.
>
> Chris Packham (2):
>    dt-bindings: gpio: brcm: Add bindings for xgs-iproc
>    gpio: Add xgs-iproc driver
Acked-by: Scott Branden <scott.branden@broadcom.com>
>
>   .../bindings/gpio/brcm,xgs-iproc-gpio.yaml    |  70 ++++
>   drivers/gpio/Kconfig                          |   9 +
>   drivers/gpio/Makefile                         |   1 +
>   drivers/gpio/gpio-xgs-iproc.c                 | 321 ++++++++++++++++++
>   4 files changed, 401 insertions(+)
>   create mode 100644 Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
>   create mode 100644 drivers/gpio/gpio-xgs-iproc.c
>

Re: [PATCH v4 1/2] dt-bindings: gpio: brcm: Add bindings for xgs-iproc

[Rob Herring]

On Thu, Oct 24, 2019 at 3:27 PM Chris Packham
<chris.packham@alliedtelesis.co.nz> wrote:
>
> This GPIO controller is present on a number of Broadcom switch ASICs
> with integrated SoCs. It is similar to the nsp-gpio and iproc-gpio
> blocks but different enough to require a separate driver.
>
> Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
> ---
>
> Notes:
>     Changes in v4:
>     - rename brcm,xgs-iproc.yaml -> brcm,xgs-iproc-gpio.yaml as suggested
>
>     Changes in v3:
>     - incorporate review comments from Rob and Bart
>
>     Changes in v2:
>     - Document as DT schema
>     - Include ngpios, #gpio-cells and gpio-controller properties
>
>  .../bindings/gpio/brcm,xgs-iproc-gpio.yaml    | 70 +++++++++++++++++++
>  1 file changed, 70 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
>
> diff --git a/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml b/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
> new file mode 100644
> index 000000000000..ec1fd3a64aa2
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
> @@ -0,0 +1,70 @@
> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/gpio/brcm,xgs-iproc.yaml#

Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml: $id:
path/filename 'gpio/brcm,xgs-iproc.yaml' doesn't match actual filename

Otherwise,

Reviewed-by: Rob Herring <robh@kernel.org>

Re: [PATCH v4 1/2] dt-bindings: gpio: brcm: Add bindings for xgs-iproc

[Chris Packham]

Hi Rob,

On Fri, 2019-10-25 at 12:12 -0500, Rob Herring wrote:
> On Thu, Oct 24, 2019 at 3:27 PM Chris Packham
> <chris.packham@alliedtelesis.co.nz> wrote:
> > 
> > This GPIO controller is present on a number of Broadcom switch ASICs
> > with integrated SoCs. It is similar to the nsp-gpio and iproc-gpio
> > blocks but different enough to require a separate driver.
> > 
> > Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
> > ---
> > 
> > Notes:
> >     Changes in v4:
> >     - rename brcm,xgs-iproc.yaml -> brcm,xgs-iproc-gpio.yaml as suggested
> > 
> >     Changes in v3:
> >     - incorporate review comments from Rob and Bart
> > 
> >     Changes in v2:
> >     - Document as DT schema
> >     - Include ngpios, #gpio-cells and gpio-controller properties
> > 
> >  .../bindings/gpio/brcm,xgs-iproc-gpio.yaml    | 70 +++++++++++++++++++
> >  1 file changed, 70 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
> > 
> > diff --git a/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml b/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
> > new file mode 100644
> > index 000000000000..ec1fd3a64aa2
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml
> > @@ -0,0 +1,70 @@
> > +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> > +%YAML 1.2
> > +---
> > +$id: http://devicetree.org/schemas/gpio/brcm,xgs-iproc.yaml#
> 
> Documentation/devicetree/bindings/gpio/brcm,xgs-iproc-gpio.yaml: $id:
> path/filename 'gpio/brcm,xgs-iproc.yaml' doesn't match actual filename
> 
> Otherwise,
> 
> Reviewed-by: Rob Herring <robh@kernel.org>

I'll send a quick v5 of just this patch with the minor correction and
your reviewed-by.

Re: [PATCH v4 2/2] gpio: Add xgs-iproc driver

[Linus Walleij]

On Thu, Oct 24, 2019 at 10:27 PM Chris Packham
<chris.packham@alliedtelesis.co.nz> wrote:

> This driver supports the Chip Common A GPIO controller present on a
> number of Broadcom switch ASICs with integrated SoCs. The controller is
> similar to the pinctrl-nsp-gpio and pinctrl-iproc-gpio blocks but
> different enough that a separate driver is required.
>
> This has been ported from Broadcom's XLDK 5.0.3 retaining only the CCA
> support (pinctrl-iproc-gpio covers CCB).
>
> Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
> ---
>
> Notes:
>     Changes in v4:
>     - rename the config option to GPIO_BCM_XGS_IPROC and place alphabetically
>     - sort #includes alphabetically

Patch applied, great work on this patch, it's very lean and clean now!

Got a minor thing I wanna look at, will send a patch on top if I find it
useful.

Yours,
Linus Walleij

Re: [PATCH v4 2/2] gpio: Add xgs-iproc driver

[Jon Hunter]

On 24/10/2019 21:27, Chris Packham wrote:
> This driver supports the Chip Common A GPIO controller present on a
> number of Broadcom switch ASICs with integrated SoCs. The controller is
> similar to the pinctrl-nsp-gpio and pinctrl-iproc-gpio blocks but
> different enough that a separate driver is required.
> 
> This has been ported from Broadcom's XLDK 5.0.3 retaining only the CCA
> support (pinctrl-iproc-gpio covers CCB).
> 
> Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
> ---
> 
> Notes:
>     Changes in v4:
>     - rename the config option to GPIO_BCM_XGS_IPROC and place alphabetically
>     - sort #includes alphabetically
>     
>     Changes in v3:
>     - prefix local #defines with 'IPROC'
>     - use {readl,writel}_relaxed
>     - remove unnecessary headers
>     - actually use spinlock to guard hardware accesses
>     
>     Changes in v2:
>     - use more of the generic infrastructure for gpio chips
>     - handling the root interrupt is still done manually due to sharing with uart0.
> 
>  drivers/gpio/Kconfig          |   9 +
>  drivers/gpio/Makefile         |   1 +
>  drivers/gpio/gpio-xgs-iproc.c | 321 ++++++++++++++++++++++++++++++++++
>  3 files changed, 331 insertions(+)
>  create mode 100644 drivers/gpio/gpio-xgs-iproc.c
> 
> diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
> index 38e096e6925f..04396787fbb0 100644
> --- a/drivers/gpio/Kconfig
> +++ b/drivers/gpio/Kconfig
> @@ -147,6 +147,15 @@ config GPIO_BCM_KONA
>  	help
>  	  Turn on GPIO support for Broadcom "Kona" chips.
>  
> +config GPIO_BCM_XGS_IPROC
> +	tristate "BRCM XGS iProc GPIO support"
> +	depends on OF_GPIO && (ARCH_BCM_IPROC || COMPILE_TEST)
> +	select GPIO_GENERIC
> +	select GPIOLIB_IRQCHIP
> +	default ARCH_BCM_IPROC
> +	help
> +	  Say yes here to enable GPIO support for Broadcom XGS iProc SoCs.
> +
>  config GPIO_BRCMSTB
>  	tristate "BRCMSTB GPIO support"
>  	default y if (ARCH_BRCMSTB || BMIPS_GENERIC)
> diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile
> index d2fd19c15bae..8725d158a964 100644
> --- a/drivers/gpio/Makefile
> +++ b/drivers/gpio/Makefile
> @@ -34,6 +34,7 @@ obj-$(CONFIG_GPIO_ARIZONA)		+= gpio-arizona.o
>  obj-$(CONFIG_GPIO_ASPEED)		+= gpio-aspeed.o
>  obj-$(CONFIG_GPIO_ATH79)		+= gpio-ath79.o
>  obj-$(CONFIG_GPIO_BCM_KONA)		+= gpio-bcm-kona.o
> +obj-$(CONFIG_GPIO_BCM_XGS_IPROC)	+= gpio-xgs-iproc.o
>  obj-$(CONFIG_GPIO_BD70528)		+= gpio-bd70528.o
>  obj-$(CONFIG_GPIO_BD9571MWV)		+= gpio-bd9571mwv.o
>  obj-$(CONFIG_GPIO_BRCMSTB)		+= gpio-brcmstb.o
> diff --git a/drivers/gpio/gpio-xgs-iproc.c b/drivers/gpio/gpio-xgs-iproc.c
> new file mode 100644
> index 000000000000..a3fdd95cc9e6
> --- /dev/null
> +++ b/drivers/gpio/gpio-xgs-iproc.c

...

> +static const struct of_device_id bcm_iproc_gpio_of_match[] __initconst = {
> +	{ .compatible = "brcm,iproc-gpio-cca" },
> +	{}
> +};
> +MODULE_DEVICE_TABLE(of, bcm_iproc_gpio_of_match);

This patch is generating the following warning when built as a module ...

MODPOST vmlinux.o
WARNING: vmlinux.o(.data+0x834d0): Section mismatch in reference from the variable bcm_iproc_gpio_driver to the variable .init.rodata:bcm_iproc_gpio_of_match
The variable bcm_iproc_gpio_driver references
the variable __initconst bcm_iproc_gpio_of_match
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console


This then leads to the following crash on boot ...

[   13.586799] ------------[ cut here ]------------

[   13.591406] Ignoring spurious kernel translation fault at virtual address ffff80001139a990

[   13.599659] WARNING: CPU: 0 PID: 5 at /home/jonathanh/workdir/tegra/mlt-linux_next/kernel/arch/arm64/mm/fault.c:302 __do_kernel_fault+0xd0/0x128

[   13.612582] Modules linked in: ina3221(+) ip_tables x_tables ipv6 nf_defrag_ipv6

[   13.619966] CPU: 0 PID: 5 Comm: kworker/0:0 Tainted: G S                5.4.0-rc1-00031-g6a41b6c5fc20 #11

[   13.629509] Hardware name: NVIDIA Jetson TX2 Developer Kit (DT)

[   13.635416] Workqueue: events deferred_probe_work_func

[   13.640540] pstate: 40000085 (nZcv daIf -PAN -UAO)

[   13.645318] pc : __do_kernel_fault+0xd0/0x128

[   13.649663] lr : __do_kernel_fault+0xd0/0x128

[   13.654006] sp : ffff800010073a10

[   13.657308] x29: ffff800010073a10 x28: ffff0001f5ccb700 

[   13.662605] x27: ffff0001f7185730 x26: ffff8000117aff28 

[   13.667901] x25: 0000000000000000 x24: 0000000000000025 

[   13.673198] x23: 0000000060000085 x22: ffff80001139a990 

[   13.678495] x21: ffff800010073a80 x20: 0000000000000025 

[   13.683792] x19: 0000000096000007 x18: ffffffffffffffff 

[   13.689089] x17: 0000000000000000 x16: 0000000000000000 

[   13.694386] x15: ffff8000117998c8 x14: 3131303030386666 

[   13.699681] x13: 6666207373657264 x12: 6461206c61757472 

[   13.704978] x11: 697620746120746c x10: 756166206e6f6974 

[   13.710274] x9 : 616c736e61727420 x8 : 6c656e72656b2073 

[   13.715570] x7 : 0000000000000160 x6 : ffff0001f717c180 

[   13.720867] x5 : 0000000000000001 x4 : ffff0001f717c180 

[   13.726162] x3 : 0000000000000006 x2 : 0000000000000007 

[   13.731458] x1 : aa4a2528bb290a00 x0 : 0000000000000000 

[   13.736756] Call trace:

[   13.739194]  __do_kernel_fault+0xd0/0x128

[   13.743192]  do_translation_fault+0x40/0x70

[   13.747363]  do_mem_abort+0x3c/0x98

[   13.750839]  el1_da+0x20/0x94

[   13.753799]  __of_match_node+0x40/0x88

[   13.757535]  of_match_node+0x3c/0x60

[   13.761099]  of_match_device+0x18/0x28

[   13.764837]  platform_match+0x4c/0xd0

[   13.768488]  __device_attach_driver+0x34/0xc0

[   13.772832]  bus_for_each_drv+0x70/0xc8

[   13.776655]  __device_attach+0xdc/0x140

[   13.780479]  device_initial_probe+0x10/0x18

[   13.784649]  bus_probe_device+0x94/0xa0

[   13.788471]  deferred_probe_work_func+0x6c/0xa0

[   13.792990]  process_one_work+0x1c8/0x358

[   13.796986]  worker_thread+0x48/0x460

[   13.800637]  kthread+0xf0/0x120

[   13.803767]  ret_from_fork+0x10/0x1c

[   13.807331] ---[ end trace ce728f2656bbae67 ]---


I think we need to drop the __initconst from the match table.

I will send a patch.

Cheers
Jon

-- 
nvpublic