[PATCH 1/6] ARM: Samsung: Add common s5p gpio interrupt support

Marek Szyprowski m.szyprowski at samsung.com
Tue Sep 28 10:24:41 EDT 2010


Hello,

On Tuesday, September 28, 2010 4:04 PM Kukjin Kim wrote:

> Marek Szyprowski wrote:
> >
> > This patch adds common code to enable support of gpio interrupts on s5p
> > series.
> >
> > The total number of gpio pins is quite large on s5p series. Registering
> > irq support for all of them would be a resource waste. Because of that
> > the interrupt support for standard gpio pins is registered dynamically
> > by the s5p_register_gpio_interrupt() function.
> 
> Hi,
> 
> I checked only S5PV210/S5PC110 GPIO interrupt.
> 
> >
> > Signed-off-by: Marek Szyprowski <m.szyprowski at samsung.com>
> > Signed-off-by: Joonyoung Shim <jy0922.shim at samsung.com>
> > Signed-off-by: Kyungmin Park <kyungmin.park at samsung.com>
> > ---
> >  arch/arm/plat-s5p/Kconfig                      |    5 +
> >  arch/arm/plat-s5p/Makefile                     |    1 +
> >  arch/arm/plat-s5p/include/plat/irqs.h          |    5 +
> >  arch/arm/plat-s5p/irq-gpioint.c                |  243
> > ++++++++++++++++++++++++
> >  arch/arm/plat-samsung/include/plat/gpio-cfg.h  |   18 ++
> >  arch/arm/plat-samsung/include/plat/gpio-core.h |    4 +
> >  6 files changed, 276 insertions(+), 0 deletions(-)
> >  create mode 100644 arch/arm/plat-s5p/irq-gpioint.c
> >
> > diff --git a/arch/arm/plat-s5p/Kconfig b/arch/arm/plat-s5p/Kconfig
> > index 407e323..d77d518 100644
> > --- a/arch/arm/plat-s5p/Kconfig
> > +++ b/arch/arm/plat-s5p/Kconfig
> > @@ -32,6 +32,11 @@ config S5P_EXT_INT
> >  	  Use the external interrupts (other than GPIO interrupts.)
> >  	  Note: Do not choose this for S5P6440.
> >
> > +config S5P_GPIO_INT
> > +	bool
> > +	help
> > +	  Common code for the GPIO interrupts (other than external
> interrupts.)
> > +
> >  config S5P_DEV_FIMC0
> >  	bool
> >  	help
> > diff --git a/arch/arm/plat-s5p/Makefile b/arch/arm/plat-s5p/Makefile
> > index f3e917e..e0823be 100644
> > --- a/arch/arm/plat-s5p/Makefile
> > +++ b/arch/arm/plat-s5p/Makefile
> > @@ -18,6 +18,7 @@ obj-y				+= cpu.o
> >  obj-y				+= clock.o
> >  obj-y				+= irq.o
> >  obj-$(CONFIG_S5P_EXT_INT)	+= irq-eint.o
> > +obj-$(CONFIG_S5P_GPIO_INT)	+= irq-gpioint.o
> >
> >  # devices
> >
> > diff --git a/arch/arm/plat-s5p/include/plat/irqs.h b/arch/arm/plat-
> > s5p/include/plat/irqs.h
> > index 3fb3a3a..50bcf1e 100644
> > --- a/arch/arm/plat-s5p/include/plat/irqs.h
> > +++ b/arch/arm/plat-s5p/include/plat/irqs.h
> > @@ -94,4 +94,9 @@
> >  						((irq) - S5P_EINT_BASE1) : \
> >  						((irq) + 16 -
> > S5P_EINT_BASE2))
> >
> > +/* GPIO interrupt (registered by s5p_register_gpio_interrupt) */
> > +#define S5P_GPIOINT_GROUP_COUNT 4
> > +#define S5P_GPIOINT_GROUP_SIZE	8
> 
> Is it possible to support S5P SoCs GPIO interrupt with only 4-GROUPs?

Yes, it won't be a problem. Just 4 interrupt slots will be wasted. Not a bit problem
imho.

> 
> > +#define S5P_GPIOINT_COUNT	(S5P_GPIOINT_GROUP_COUNT *
> > S5P_GPIOINT_GROUP_SIZE)
> > +
> >  #endif /* __ASM_PLAT_S5P_IRQS_H */
> > diff --git a/arch/arm/plat-s5p/irq-gpioint.c
> b/arch/arm/plat-s5p/irq-gpioint.c
> > new file mode 100644
> > index 0000000..7409ae0
> > --- /dev/null
> > +++ b/arch/arm/plat-s5p/irq-gpioint.c
> > @@ -0,0 +1,243 @@
> > +/* linux/arch/arm/plat-s5p/irq-gpioint.c
> > + *
> > + * Copyright (c) 2010 Samsung Electronics Co., Ltd.
> > + * Author: Kyungmin Park <kyungmin.park at samsung.com>
> > + * Author: Joonyoung Shim <jy0922.shim at samsung.com>
> > + * Author: Marek Szyprowski <m.szyprowski at samsung.com>
> > + *
> > + *  This program is free software; you can redistribute  it and/or modify
> it
> > + *  under  the terms of  the GNU General  Public License as published by
> the
> > + *  Free Software Foundation;  either version 2 of the  License, or (at
> your
> > + *  option) any later version.
> > + *
> > + */
> > +
> > +#include <linux/kernel.h>
> > +#include <linux/interrupt.h>
> > +#include <linux/irq.h>
> > +#include <linux/io.h>
> > +#include <linux/gpio.h>
> > +
> > +#include <mach/map.h>
> > +#include <plat/gpio-core.h>
> > +#include <plat/gpio-cfg.h>
> > +
> > +#define S5P_GPIOREG(x)			(S5P_VA_GPIO + (x))
> > +
> > +#define GPIOINT_CON_OFFSET		0x700
> > +#define GPIOINT_MASK_OFFSET		0x900
> > +#define GPIOINT_PEND_OFFSET		0xA00
> > +
> > +#define GPIOINT_LEVEL_LOW		0x0
> > +#define GPIOINT_LEVEL_HIGH		0x1
> > +#define GPIOINT_EDGE_FALLING		0x2
> > +#define GPIOINT_EDGE_RISING		0x3
> > +#define GPIOINT_EDGE_BOTH		0x4
> 
> I remember Kyungmin's patch about interrupt polarity of GPIO Interrupt and
> GPIO External Interrupt.
> How about merging them into plat-s5p/include

I'm ok with this.

> Hmm...let's think the proper name...
> 
> > +
> > +static struct s3c_gpio_chip *irq_chips[S5P_GPIOINT_GROUP_MAXNR];
> 
> Where is the definition of S5P_GPIOINT_GROUP_MAXNR?

It will be defined by the proper user of the common irq-gpioint.c code. See the
next commit ("ARM: S5PC110: add support for gpio interrupts"). To avoid any
build breaks irq-gpioint.c code is compiled conditionally if selected by the
machine that requires/uses it.

> > +
> > +static int s5p_gpioint_get_group(unsigned int irq)
> > +{
> > +	struct gpio_chip *chip = get_irq_data(irq);
> > +	struct s3c_gpio_chip *s3c_chip = container_of(chip,
> > +			struct s3c_gpio_chip, chip);
> > +	int group;
> > +
> > +	for (group = 0; group < S5P_GPIOINT_GROUP_MAXNR; group++)
> > +		if (s3c_chip == irq_chips[group])
> > +			break;
> > +
> > +	return group;
> > +}
> > +
> > +static int s5p_gpioint_get_offset(unsigned int irq)
> > +{
> > +	struct gpio_chip *chip = get_irq_data(irq);
> > +	struct s3c_gpio_chip *s3c_chip = container_of(chip,
> > +			struct s3c_gpio_chip, chip);
> > +
> > +	return irq - s3c_chip->irq_base;
> > +}
> > +
> > +static void s5p_gpioint_ack(unsigned int irq)
> > +{
> > +	int group, offset, pend_offset;
> > +	unsigned int value;
> > +
> > +	group = s5p_gpioint_get_group(irq);
> > +	offset = s5p_gpioint_get_offset(irq);
> > +	pend_offset = group << 2;
> > +
> > +	value = __raw_readl(S5P_GPIOREG(GPIOINT_PEND_OFFSET) +
> > pend_offset);
> > +	value |= 1 << offset;
> > +	__raw_writel(value, S5P_GPIOREG(GPIOINT_PEND_OFFSET) +
> > pend_offset);
> > +}
> > +
> > +static void s5p_gpioint_mask(unsigned int irq)
> > +{
> > +	int group, offset, mask_offset;
> > +	unsigned int value;
> > +
> > +	group = s5p_gpioint_get_group(irq);
> > +	offset = s5p_gpioint_get_offset(irq);
> > +	mask_offset = group << 2;
> > +
> > +	value = __raw_readl(S5P_GPIOREG(GPIOINT_MASK_OFFSET) +
> > mask_offset);
> > +	value |= 1 << offset;
> > +	__raw_writel(value, S5P_GPIOREG(GPIOINT_MASK_OFFSET) +
> > mask_offset);
> > +}
> > +
> > +static void s5p_gpioint_unmask(unsigned int irq)
> > +{
> > +	int group, offset, mask_offset;
> > +	unsigned int value;
> > +
> > +	group = s5p_gpioint_get_group(irq);
> > +	offset = s5p_gpioint_get_offset(irq);
> > +	mask_offset = group << 2;
> > +
> > +	value = __raw_readl(S5P_GPIOREG(GPIOINT_MASK_OFFSET) +
> > mask_offset);
> > +	value &= ~(1 << offset);
> > +	__raw_writel(value, S5P_GPIOREG(GPIOINT_MASK_OFFSET) +
> > mask_offset);
> > +}
> > +
> > +static void s5p_gpioint_mask_ack(unsigned int irq)
> > +{
> > +	s5p_gpioint_mask(irq);
> > +	s5p_gpioint_ack(irq);
> > +}
> > +
> > +static int s5p_gpioint_set_type(unsigned int irq, unsigned int type)
> > +{
> > +	int group, offset, con_offset;
> > +	unsigned int value;
> > +
> > +	group = s5p_gpioint_get_group(irq);
> > +	offset = s5p_gpioint_get_offset(irq);
> > +	con_offset = group << 2;
> > +
> > +	switch (type) {
> > +	case IRQ_TYPE_NONE:
> > +		printk(KERN_WARNING "No irq type\n");
> > +		return -EINVAL;
> > +	case IRQ_TYPE_EDGE_RISING:
> > +		type = GPIOINT_EDGE_RISING;
> > +		break;
> > +	case IRQ_TYPE_EDGE_FALLING:
> > +		type = GPIOINT_EDGE_FALLING;
> > +		break;
> > +	case IRQ_TYPE_EDGE_BOTH:
> > +		type = GPIOINT_EDGE_BOTH;
> > +		break;
> > +	case IRQ_TYPE_LEVEL_HIGH:
> > +		type = GPIOINT_LEVEL_HIGH;
> > +		break;
> > +	case IRQ_TYPE_LEVEL_LOW:
> > +		type = GPIOINT_LEVEL_LOW;
> > +		break;
> > +	default:
> > +		BUG();
> 
> how about merging IRQ_TYPE_NONE and 'default'?
> And basically, rerely happened exception case, so it's better move
> IRQ_TYPE_NONE to later in switch case.

Ok, I can change this.

> > +	}
> > +
> > +	value = __raw_readl(S5P_GPIOREG(GPIOINT_CON_OFFSET) +
> > con_offset);
> > +	value &= ~(0xf << (offset * 0x4));
> 
> 0x7 is enough on S5PV210...but need to check another S5P SoCs.
> 
> > +	value |= (type << (offset * 0x4));
> > +	__raw_writel(value, S5P_GPIOREG(GPIOINT_CON_OFFSET) +
> > con_offset);
> > +
> > +	return 0;
> > +}
> > +
> > +struct irq_chip s5p_gpioint = {
> > +	.name		= "GPIO",
> 
> how about s5p-gpioint?

ok, no problem.

> 
> > +	.ack		= s5p_gpioint_ack,
> > +	.mask		= s5p_gpioint_mask,
> > +	.mask_ack	= s5p_gpioint_mask_ack,
> > +	.unmask		= s5p_gpioint_unmask,
> > +	.set_type	= s5p_gpioint_set_type,
> > +};
> > +
> > +static void s5p_gpioint_handler(unsigned int irq, struct irq_desc *desc)
> > +{
> > +	int group, offset, pend_offset, mask_offset;
> > +	int real_irq;
> > +	unsigned int pend, mask;
> > +
> > +	for (group = 0; group < S5P_GPIOINT_GROUP_MAXNR; group++) {
> > +		pend_offset = group << 2;
> > +		pend = __raw_readl(S5P_GPIOREG(GPIOINT_PEND_OFFSET) +
> > +				pend_offset);
> > +		if (!pend)
> > +			continue;
> > +
> > +		mask_offset = group << 2;
> > +		mask = __raw_readl(S5P_GPIOREG(GPIOINT_MASK_OFFSET) +
> > +				mask_offset);
> > +		pend &= ~mask;
> > +
> > +		for (offset = 0; offset < 8; offset++) {
> > +			if (pend & (1 << offset)) {
> > +				struct s3c_gpio_chip *chip =
> irq_chips[group];
> > +				if (chip) {
> > +					real_irq = chip->irq_base + offset;
> > +					generic_handle_irq(real_irq);
> > +				}
> > +			}
> > +		}
> > +	}
> > +}
> > +
> > +static __init int s5p_gpioint_add(struct s3c_gpio_chip *chip)
> > +{
> > +	static int used_gpioint_groups = 0;
> > +	static bool handler_registered = 0;
> > +	int irq, group = chip->group;
> > +	int i;
> > +
> > +	if (used_gpioint_groups >= S5P_GPIOINT_GROUP_COUNT)
> > +		return -ENOMEM;
> > +
> > +	chip->irq_base = S5P_GPIOINT_BASE +
> > +			 used_gpioint_groups * S5P_GPIOINT_GROUP_SIZE;
> 
> Where is S5P_GPIOINT_BASE?

see the next commit...

> 
> > +	used_gpioint_groups++;
> > +
> > +	if (!handler_registered) {
> > +		set_irq_chained_handler(IRQ_GPIOINT, s5p_gpioint_handler);
> 
> As you know, IRQ_GPIOINT defined only in the mach-s5p6442, mach-s5pc100 and
> mach-s5pv210.
> It means can be happend build error with mach-s5p64x0 or mach-s5pv310.

Nope, this code is compiled conditionally only if the platform requires/supports
it.

> 
> > +		handler_registered = 1;
> > +	}
> > +
> > +	irq_chips[group] = chip;
> > +	for (i = 0; i < chip->chip.ngpio; i++) {
> > +		irq = chip->irq_base + i;
> > +		set_irq_chip(irq, &s5p_gpioint);
> > +		set_irq_data(irq, &chip->chip);
> 
> Do we really need set_irq_data?

Yes, to get the gpio bank group number in s5p_gpioint_get_group() function.

> 
> > +		set_irq_handler(irq, handle_level_irq);
> > +		set_irq_flags(irq, IRQF_VALID);
> > +	}
> > +	return 0;
> > +}
> > +
> > +int __init s5p_register_gpio_interrupt(int pin)
> > +{
> > +	struct s3c_gpio_chip *my_chip = s3c_gpiolib_getchip(pin);
> > +	int offset, group;
> > +	int ret;
> > +
> > +	if (!my_chip)
> > +		return -EINVAL;
> > +
> > +	offset = pin - my_chip->chip.base;
> > +	group = my_chip->group;
> > +
> > +	/* check if the group has been already registered */
> > +	if (my_chip->irq_base)
> > +		return my_chip->irq_base + offset;
> > +
> > +	/* register gpio group */
> > +	ret = s5p_gpioint_add(my_chip);
> > +	if (ret == 0) {
> > +		printk(KERN_INFO "Registered interrupt support for gpio
> > group %d.\n",
> > +		       group);
> > +		return my_chip->irq_base + offset;
> > +	}
> > +	return ret;
> > +}
> > diff --git a/arch/arm/plat-samsung/include/plat/gpio-cfg.h
> b/arch/arm/plat-
> > samsung/include/plat/gpio-cfg.h
> > index db4112c..41479a0 100644
> > --- a/arch/arm/plat-samsung/include/plat/gpio-cfg.h
> > +++ b/arch/arm/plat-samsung/include/plat/gpio-cfg.h
> > @@ -169,4 +169,22 @@ extern s5p_gpio_drvstr_t s5p_gpio_get_drvstr(unsigned
> > int pin);
> >  */
> >  extern int s5p_gpio_set_drvstr(unsigned int pin, s5p_gpio_drvstr_t
> drvstr);
> >
> > +/**
> > + * s5p_register_gpio_interrupt() - register interrupt support for a gpio
> group
> > + * @pin: The pin number from the group to be registered
> > + *
> > + * This function registers gpio interrupt support for the group that the
> > + * specified pin belongs to.
> > + *
> > + * The total number of gpio pins is quite large ob s5p series.
> Registering
> > + * irq support for all of them would be a resource waste. Because of that
> the
> > + * interrupt support for standard gpio pins is registered dynamically.
> > + *
> > + * It will return the irq number of the interrupt that has been
> registered
> > + * or -ENOMEM if no more gpio interrupts can be registered. It is allowed
> > + * to call this function more than once for the same gpio group (the
> group
> > + * will be registered only once).
> > + */
> > +extern int s5p_register_gpio_interrupt(int pin);
> > +
> >  #endif /* __PLAT_GPIO_CFG_H */
> > diff --git a/arch/arm/plat-samsung/include/plat/gpio-core.h
> b/arch/arm/plat-
> > samsung/include/plat/gpio-core.h
> > index e358c7d..c22c27c 100644
> > --- a/arch/arm/plat-samsung/include/plat/gpio-core.h
> > +++ b/arch/arm/plat-samsung/include/plat/gpio-core.h
> > @@ -43,6 +43,8 @@ struct s3c_gpio_cfg;
> >   * struct s3c_gpio_chip - wrapper for specific implementation of gpio
> >   * @chip: The chip structure to be exported via gpiolib.
> >   * @base: The base pointer to the gpio configuration registers.
> > + * @group: The group register number for gpio interrupt support.
> > + * @irq_base: The base irq number.
> >   * @config: special function and pull-resistor control information.
> >   * @lock: Lock for exclusive access to this gpio bank.
> >   * @pm_save: Save information for suspend/resume support.
> > @@ -63,6 +65,8 @@ struct s3c_gpio_chip {
> >  	struct s3c_gpio_cfg	*config;
> >  	struct s3c_gpio_pm	*pm;
> >  	void __iomem		*base;
> > +	int			irq_base;
> > +	int			group;
> >  	spinlock_t		 lock;
> >  #ifdef CONFIG_PM
> >  	u32			pm_save[4];
> > --
> 
> I need sleeping :-)
> Will continue tomorrow.

Ok.

Best regards
--
Marek Szyprowski
Samsung Poland R&D Center





More information about the linux-arm-kernel mailing list