sh: intc: Split up the INTC code.

This splits up the sh intc core in to something more vaguely resembling
a subsystem. Most of the functionality was alread fairly well
compartmentalized, and there were only a handful of interdependencies
that needed to be resolved in the process.

This also serves as future-proofing for the genirq and sparseirq rework,
which will make some of the split out functionality wholly generic,
allowing things to be killed off in place with minimal migration pain.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

drivers/sh/Kconfig

-config INTC_USERIMASK
-	bool "Userspace interrupt masking support"
-	depends on ARCH_SHMOBILE || (SUPERH && CPU_SH4A)
-	help
-	  This enables support for hardware-assisted userspace hardirq
-	  masking.
+menu "SuperH / SH-Mobile Driver Options"
 
-	  SH-4A and newer interrupt blocks all support a special shadowed
-	  page with all non-masking registers obscured when mapped in to
-	  userspace. This is primarily for use by userspace device
-	  drivers that are using special priority levels.
+source "drivers/sh/intc/Kconfig"
 
-	  If in doubt, say N.
-
-config INTC_BALANCING
-	bool "Hardware IRQ balancing support"
-	depends on SMP && SUPERH && CPU_SHX3
-	help
-	  This enables support for IRQ auto-distribution mode on SH-X3
-	  SMP parts. All of the balancing and CPU wakeup decisions are
-	  taken care of automatically by hardware for distributed
-	  vectors.
-
-	  If in doubt, say N.
-
-config INTC_MAPPING_DEBUG
-	bool "Expose IRQ to per-controller id mapping via debugfs"
-	depends on DEBUG_FS
-	help
-	  This will create a debugfs entry for showing the relationship
-	  between system IRQs and the per-controller id tables.
-
-	  If in doubt, say N.
+endmenu

drivers/sh/Makefile

 #
 # Makefile for the SuperH specific drivers.
 #
-obj-y	:= clk.o intc.o
+obj-y	:= clk.o intc/
 
 obj-$(CONFIG_SUPERHYWAY)	+= superhyway/
 obj-$(CONFIG_MAPLE)		+= maple/

drivers/sh/intc/access.c

+/*
+ * Common INTC2 register accessors
+ *
+ * Copyright (C) 2007, 2008 Magnus Damm
+ * Copyright (C) 2009, 2010 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/io.h>
+#include "internals.h"
+
+unsigned long intc_phys_to_virt(struct intc_desc_int *d, unsigned long address)
+{
+	struct intc_window *window;
+	int k;
+
+	/* scan through physical windows and convert address */
+	for (k = 0; k < d->nr_windows; k++) {
+		window = d->window + k;
+
+		if (address < window->phys)
+			continue;
+
+		if (address >= (window->phys + window->size))
+			continue;
+
+		address -= window->phys;
+		address += (unsigned long)window->virt;
+
+		return address;
+	}
+
+	/* no windows defined, register must be 1:1 mapped virt:phys */
+	return address;
+}
+
+unsigned int intc_get_reg(struct intc_desc_int *d, unsigned long address)
+{
+	unsigned int k;
+
+	address = intc_phys_to_virt(d, address);
+
+	for (k = 0; k < d->nr_reg; k++) {
+		if (d->reg[k] == address)
+			return k;
+	}
+
+	BUG();
+	return 0;
+}
+
+unsigned int intc_set_field_from_handle(unsigned int value,
+					unsigned int field_value,
+					unsigned int handle)
+{
+	unsigned int width = _INTC_WIDTH(handle);
+	unsigned int shift = _INTC_SHIFT(handle);
+
+	value &= ~(((1 << width) - 1) << shift);
+	value |= field_value << shift;
+	return value;
+}
+
+unsigned long intc_get_field_from_handle(unsigned int value, unsigned int handle)
+{
+	unsigned int width = _INTC_WIDTH(handle);
+	unsigned int shift = _INTC_SHIFT(handle);
+	unsigned int mask = ((1 << width) - 1) << shift;
+
+	return (value & mask) >> shift;
+}
+
+static unsigned long test_8(unsigned long addr, unsigned long h,
+			    unsigned long ignore)
+{
+	return intc_get_field_from_handle(__raw_readb(addr), h);
+}
+
+static unsigned long test_16(unsigned long addr, unsigned long h,
+			     unsigned long ignore)
+{
+	return intc_get_field_from_handle(__raw_readw(addr), h);
+}
+
+static unsigned long test_32(unsigned long addr, unsigned long h,
+			     unsigned long ignore)
+{
+	return intc_get_field_from_handle(__raw_readl(addr), h);
+}
+
+static unsigned long write_8(unsigned long addr, unsigned long h,
+			     unsigned long data)
+{
+	__raw_writeb(intc_set_field_from_handle(0, data, h), addr);
+	(void)__raw_readb(addr);	/* Defeat write posting */
+	return 0;
+}
+
+static unsigned long write_16(unsigned long addr, unsigned long h,
+			      unsigned long data)
+{
+	__raw_writew(intc_set_field_from_handle(0, data, h), addr);
+	(void)__raw_readw(addr);	/* Defeat write posting */
+	return 0;
+}
+
+static unsigned long write_32(unsigned long addr, unsigned long h,
+			      unsigned long data)
+{
+	__raw_writel(intc_set_field_from_handle(0, data, h), addr);
+	(void)__raw_readl(addr);	/* Defeat write posting */
+	return 0;
+}
+
+static unsigned long modify_8(unsigned long addr, unsigned long h,
+			      unsigned long data)
+{
+	unsigned long flags;
+	unsigned int value;
+	local_irq_save(flags);
+	value = intc_set_field_from_handle(__raw_readb(addr), data, h);
+	__raw_writeb(value, addr);
+	(void)__raw_readb(addr);	/* Defeat write posting */
+	local_irq_restore(flags);
+	return 0;
+}
+
+static unsigned long modify_16(unsigned long addr, unsigned long h,
+			       unsigned long data)
+{
+	unsigned long flags;
+	unsigned int value;
+	local_irq_save(flags);
+	value = intc_set_field_from_handle(__raw_readw(addr), data, h);
+	__raw_writew(value, addr);
+	(void)__raw_readw(addr);	/* Defeat write posting */
+	local_irq_restore(flags);
+	return 0;
+}
+
+static unsigned long modify_32(unsigned long addr, unsigned long h,
+			       unsigned long data)
+{
+	unsigned long flags;
+	unsigned int value;
+	local_irq_save(flags);
+	value = intc_set_field_from_handle(__raw_readl(addr), data, h);
+	__raw_writel(value, addr);
+	(void)__raw_readl(addr);	/* Defeat write posting */
+	local_irq_restore(flags);
+	return 0;
+}
+
+static unsigned long intc_mode_field(unsigned long addr,
+				     unsigned long handle,
+				     unsigned long (*fn)(unsigned long,
+						unsigned long,
+						unsigned long),
+				     unsigned int irq)
+{
+	return fn(addr, handle, ((1 << _INTC_WIDTH(handle)) - 1));
+}
+
+static unsigned long intc_mode_zero(unsigned long addr,
+				    unsigned long handle,
+				    unsigned long (*fn)(unsigned long,
+					       unsigned long,
+					       unsigned long),
+				    unsigned int irq)
+{
+	return fn(addr, handle, 0);
+}
+
+static unsigned long intc_mode_prio(unsigned long addr,
+				    unsigned long handle,
+				    unsigned long (*fn)(unsigned long,
+					       unsigned long,
+					       unsigned long),
+				    unsigned int irq)
+{
+	return fn(addr, handle, intc_get_prio_level(irq));
+}
+
+unsigned long (*intc_reg_fns[])(unsigned long addr,
+				unsigned long h,
+				unsigned long data) = {
+	[REG_FN_TEST_BASE + 0] = test_8,
+	[REG_FN_TEST_BASE + 1] = test_16,
+	[REG_FN_TEST_BASE + 3] = test_32,
+	[REG_FN_WRITE_BASE + 0] = write_8,
+	[REG_FN_WRITE_BASE + 1] = write_16,
+	[REG_FN_WRITE_BASE + 3] = write_32,
+	[REG_FN_MODIFY_BASE + 0] = modify_8,
+	[REG_FN_MODIFY_BASE + 1] = modify_16,
+	[REG_FN_MODIFY_BASE + 3] = modify_32,
+};
+
+unsigned long (*intc_enable_fns[])(unsigned long addr,
+				   unsigned long handle,
+				   unsigned long (*fn)(unsigned long,
+					    unsigned long,
+					    unsigned long),
+				   unsigned int irq) = {
+	[MODE_ENABLE_REG] = intc_mode_field,
+	[MODE_MASK_REG] = intc_mode_zero,
+	[MODE_DUAL_REG] = intc_mode_field,
+	[MODE_PRIO_REG] = intc_mode_prio,
+	[MODE_PCLR_REG] = intc_mode_prio,
+};
+
+unsigned long (*intc_disable_fns[])(unsigned long addr,
+				    unsigned long handle,
+				    unsigned long (*fn)(unsigned long,
+					     unsigned long,
+					     unsigned long),
+				    unsigned int irq) = {
+	[MODE_ENABLE_REG] = intc_mode_zero,
+	[MODE_MASK_REG] = intc_mode_field,
+	[MODE_DUAL_REG] = intc_mode_field,
+	[MODE_PRIO_REG] = intc_mode_zero,
+	[MODE_PCLR_REG] = intc_mode_field,
+};
+
+unsigned long (*intc_enable_noprio_fns[])(unsigned long addr,
+					  unsigned long handle,
+					  unsigned long (*fn)(unsigned long,
+						unsigned long,
+						unsigned long),
+					  unsigned int irq) = {
+	[MODE_ENABLE_REG] = intc_mode_field,
+	[MODE_MASK_REG] = intc_mode_zero,
+	[MODE_DUAL_REG] = intc_mode_field,
+	[MODE_PRIO_REG] = intc_mode_field,
+	[MODE_PCLR_REG] = intc_mode_field,
+};

drivers/sh/intc/balancing.c

+/*
+ * Support for hardware-managed IRQ auto-distribution.
+ *
+ * Copyright (C) 2010  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include "internals.h"
+
+static unsigned long dist_handle[NR_IRQS];
+
+void intc_balancing_enable(unsigned int irq)
+{
+	struct intc_desc_int *d = get_intc_desc(irq);
+	unsigned long handle = dist_handle[irq];
+	unsigned long addr;
+
+	if (irq_balancing_disabled(irq) || !handle)
+		return;
+
+	addr = INTC_REG(d, _INTC_ADDR_D(handle), 0);
+	intc_reg_fns[_INTC_FN(handle)](addr, handle, 1);
+}
+
+void intc_balancing_disable(unsigned int irq)
+{
+	struct intc_desc_int *d = get_intc_desc(irq);
+	unsigned long handle = dist_handle[irq];
+	unsigned long addr;
+
+	if (irq_balancing_disabled(irq) || !handle)
+		return;
+
+	addr = INTC_REG(d, _INTC_ADDR_D(handle), 0);
+	intc_reg_fns[_INTC_FN(handle)](addr, handle, 0);
+}
+
+static unsigned int intc_dist_data(struct intc_desc *desc,
+				   struct intc_desc_int *d,
+				   intc_enum enum_id)
+{
+	struct intc_mask_reg *mr = desc->hw.mask_regs;
+	unsigned int i, j, fn, mode;
+	unsigned long reg_e, reg_d;
+
+	for (i = 0; mr && enum_id && i < desc->hw.nr_mask_regs; i++) {
+		mr = desc->hw.mask_regs + i;
+
+		/*
+		 * Skip this entry if there's no auto-distribution
+		 * register associated with it.
+		 */
+		if (!mr->dist_reg)
+			continue;
+
+		for (j = 0; j < ARRAY_SIZE(mr->enum_ids); j++) {
+			if (mr->enum_ids[j] != enum_id)
+				continue;
+
+			fn = REG_FN_MODIFY_BASE;
+			mode = MODE_ENABLE_REG;
+			reg_e = mr->dist_reg;
+			reg_d = mr->dist_reg;
+
+			fn += (mr->reg_width >> 3) - 1;
+			return _INTC_MK(fn, mode,
+					intc_get_reg(d, reg_e),
+					intc_get_reg(d, reg_d),
+					1,
+					(mr->reg_width - 1) - j);
+		}
+	}
+
+	/*
+	 * It's possible we've gotten here with no distribution options
+	 * available for the IRQ in question, so we just skip over those.
+	 */
+	return 0;
+}
+
+void intc_set_dist_handle(unsigned int irq, struct intc_desc *desc,
+			  struct intc_desc_int *d, intc_enum id)
+{
+	unsigned long flags;
+
+	/*
+	 * Nothing to do for this IRQ.
+	 */
+	if (!desc->hw.mask_regs)
+		return;
+
+	raw_spin_lock_irqsave(&intc_big_lock, flags);
+	dist_handle[irq] = intc_dist_data(desc, d, id);
+	raw_spin_unlock_irqrestore(&intc_big_lock, flags);
+}

drivers/sh/intc/chip.c

+/*
+ * IRQ chip definitions for INTC IRQs.
+ *
+ * Copyright (C) 2007, 2008 Magnus Damm
+ * Copyright (C) 2009, 2010 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/cpumask.h>
+#include <linux/io.h>
+#include "internals.h"
+
+void _intc_enable(unsigned int irq, unsigned long handle)
+{
+	struct intc_desc_int *d = get_intc_desc(irq);
+	unsigned long addr;
+	unsigned int cpu;
+
+	for (cpu = 0; cpu < SMP_NR(d, _INTC_ADDR_E(handle)); cpu++) {
+#ifdef CONFIG_SMP
+		if (!cpumask_test_cpu(cpu, irq_to_desc(irq)->affinity))
+			continue;
+#endif
+		addr = INTC_REG(d, _INTC_ADDR_E(handle), cpu);
+		intc_enable_fns[_INTC_MODE(handle)](addr, handle, intc_reg_fns\
+						    [_INTC_FN(handle)], irq);
+	}
+
+	intc_balancing_enable(irq);
+}
+
+static void intc_enable(unsigned int irq)
+{
+	_intc_enable(irq, (unsigned long)get_irq_chip_data(irq));
+}
+
+static void intc_disable(unsigned int irq)
+{
+	struct intc_desc_int *d = get_intc_desc(irq);
+	unsigned long handle = (unsigned long)get_irq_chip_data(irq);
+	unsigned long addr;
+	unsigned int cpu;
+
+	intc_balancing_disable(irq);
+
+	for (cpu = 0; cpu < SMP_NR(d, _INTC_ADDR_D(handle)); cpu++) {
+#ifdef CONFIG_SMP
+		if (!cpumask_test_cpu(cpu, irq_to_desc(irq)->affinity))
+			continue;
+#endif
+		addr = INTC_REG(d, _INTC_ADDR_D(handle), cpu);
+		intc_disable_fns[_INTC_MODE(handle)](addr, handle,intc_reg_fns\
+						     [_INTC_FN(handle)], irq);
+	}
+}
+
+static int intc_set_wake(unsigned int irq, unsigned int on)
+{
+	return 0; /* allow wakeup, but setup hardware in intc_suspend() */
+}
+
+#ifdef CONFIG_SMP
+/*
+ * This is held with the irq desc lock held, so we don't require any
+ * additional locking here at the intc desc level. The affinity mask is
+ * later tested in the enable/disable paths.
+ */
+static int intc_set_affinity(unsigned int irq, const struct cpumask *cpumask)
+{
+	if (!cpumask_intersects(cpumask, cpu_online_mask))
+		return -1;
+
+	cpumask_copy(irq_to_desc(irq)->affinity, cpumask);
+
+	return 0;
+}
+#endif
+
+static void intc_mask_ack(unsigned int irq)
+{
+	struct intc_desc_int *d = get_intc_desc(irq);
+	unsigned long handle = intc_get_ack_handle(irq);
+	unsigned long addr;
+
+	intc_disable(irq);
+
+	/* read register and write zero only to the associated bit */
+	if (handle) {
+		unsigned int value;
+
+		addr = INTC_REG(d, _INTC_ADDR_D(handle), 0);
+		value = intc_set_field_from_handle(0, 1, handle);
+
+		switch (_INTC_FN(handle)) {
+		case REG_FN_MODIFY_BASE + 0:	/* 8bit */
+			__raw_readb(addr);
+			__raw_writeb(0xff ^ value, addr);
+			break;
+		case REG_FN_MODIFY_BASE + 1:	/* 16bit */
+			__raw_readw(addr);
+			__raw_writew(0xffff ^ value, addr);
+			break;
+		case REG_FN_MODIFY_BASE + 3:	/* 32bit */
+			__raw_readl(addr);
+			__raw_writel(0xffffffff ^ value, addr);
+			break;
+		default:
+			BUG();
+			break;
+		}
+	}
+}
+
+static struct intc_handle_int *intc_find_irq(struct intc_handle_int *hp,
+					     unsigned int nr_hp,
+					     unsigned int irq)
+{
+	int i;
+
+	/*
+	 * this doesn't scale well, but...
+	 *
+	 * this function should only be used for cerain uncommon
+	 * operations such as intc_set_priority() and intc_set_type()
+	 * and in those rare cases performance doesn't matter that much.
+	 * keeping the memory footprint low is more important.
+	 *
+	 * one rather simple way to speed this up and still keep the
+	 * memory footprint down is to make sure the array is sorted
+	 * and then perform a bisect to lookup the irq.
+	 */
+	for (i = 0; i < nr_hp; i++) {
+		if ((hp + i)->irq != irq)
+			continue;
+
+		return hp + i;
+	}
+
+	return NULL;
+}
+
+int intc_set_priority(unsigned int irq, unsigned int prio)
+{
+	struct intc_desc_int *d = get_intc_desc(irq);
+	struct intc_handle_int *ihp;
+
+	if (!intc_get_prio_level(irq) || prio <= 1)
+		return -EINVAL;
+
+	ihp = intc_find_irq(d->prio, d->nr_prio, irq);
+	if (ihp) {
+		if (prio >= (1 << _INTC_WIDTH(ihp->handle)))
+			return -EINVAL;
+
+		intc_set_prio_level(irq, prio);
+
+		/*
+		 * only set secondary masking method directly
+		 * primary masking method is using intc_prio_level[irq]
+		 * priority level will be set during next enable()
+		 */
+		if (_INTC_FN(ihp->handle) != REG_FN_ERR)
+			_intc_enable(irq, ihp->handle);
+	}
+	return 0;
+}
+
+#define VALID(x) (x | 0x80)
+
+static unsigned char intc_irq_sense_table[IRQ_TYPE_SENSE_MASK + 1] = {
+	[IRQ_TYPE_EDGE_FALLING] = VALID(0),
+	[IRQ_TYPE_EDGE_RISING] = VALID(1),
+	[IRQ_TYPE_LEVEL_LOW] = VALID(2),
+	/* SH7706, SH7707 and SH7709 do not support high level triggered */
+#if !defined(CONFIG_CPU_SUBTYPE_SH7706) && \
+    !defined(CONFIG_CPU_SUBTYPE_SH7707) && \
+    !defined(CONFIG_CPU_SUBTYPE_SH7709)
+	[IRQ_TYPE_LEVEL_HIGH] = VALID(3),
+#endif
+};
+
+static int intc_set_type(unsigned int irq, unsigned int type)
+{
+	struct intc_desc_int *d = get_intc_desc(irq);
+	unsigned char value = intc_irq_sense_table[type & IRQ_TYPE_SENSE_MASK];
+	struct intc_handle_int *ihp;
+	unsigned long addr;
+
+	if (!value)
+		return -EINVAL;
+
+	ihp = intc_find_irq(d->sense, d->nr_sense, irq);
+	if (ihp) {
+		addr = INTC_REG(d, _INTC_ADDR_E(ihp->handle), 0);
+		intc_reg_fns[_INTC_FN(ihp->handle)](addr, ihp->handle, value);
+	}
+
+	return 0;
+}
+
+struct irq_chip intc_irq_chip	= {
+	.mask		= intc_disable,
+	.unmask		= intc_enable,
+	.mask_ack	= intc_mask_ack,
+	.enable		= intc_enable,
+	.disable	= intc_disable,
+	.shutdown	= intc_disable,
+	.set_type	= intc_set_type,
+	.set_wake	= intc_set_wake,
+#ifdef CONFIG_SMP
+	.set_affinity	= intc_set_affinity,
+#endif
+};

drivers/sh/intc/dynamic.c

+/*
+ * Dynamic IRQ management
+ *
+ * Copyright (C) 2010  Paul Mundt
+ *
+ * Modelled after arch/x86/kernel/apic/io_apic.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#define pr_fmt(fmt) "intc: " fmt
+
+#include <linux/irq.h>
+#include <linux/bitmap.h>
+#include <linux/spinlock.h>
+#include "internals.h" /* only for activate_irq() damage.. */
+
+/*
+ * The intc_irq_map provides a global map of bound IRQ vectors for a
+ * given platform. Allocation of IRQs are either static through the CPU
+ * vector map, or dynamic in the case of board mux vectors or MSI.
+ *
+ * As this is a central point for all IRQ controllers on the system,
+ * each of the available sources are mapped out here. This combined with
+ * sparseirq makes it quite trivial to keep the vector map tightly packed
+ * when dynamically creating IRQs, as well as tying in to otherwise
+ * unused irq_desc positions in the sparse array.
+ */
+static DECLARE_BITMAP(intc_irq_map, NR_IRQS);
+static DEFINE_RAW_SPINLOCK(vector_lock);
+
+/*
+ * Dynamic IRQ allocation and deallocation
+ */
+unsigned int create_irq_nr(unsigned int irq_want, int node)
+{
+	unsigned int irq = 0, new;
+	unsigned long flags;
+	struct irq_desc *desc;
+
+	raw_spin_lock_irqsave(&vector_lock, flags);
+
+	/*
+	 * First try the wanted IRQ
+	 */
+	if (test_and_set_bit(irq_want, intc_irq_map) == 0) {
+		new = irq_want;
+	} else {
+		/* .. then fall back to scanning. */
+		new = find_first_zero_bit(intc_irq_map, nr_irqs);
+		if (unlikely(new == nr_irqs))
+			goto out_unlock;
+
+		__set_bit(new, intc_irq_map);
+	}
+
+	desc = irq_to_desc_alloc_node(new, node);
+	if (unlikely(!desc)) {
+		pr_err("can't get irq_desc for %d\n", new);
+		goto out_unlock;
+	}
+
+	desc = move_irq_desc(desc, node);
+	irq = new;
+
+out_unlock:
+	raw_spin_unlock_irqrestore(&vector_lock, flags);
+
+	if (irq > 0) {
+		dynamic_irq_init(irq);
+		activate_irq(irq);
+	}
+
+	return irq;
+}
+
+int create_irq(void)
+{
+	int nid = cpu_to_node(smp_processor_id());
+	int irq;
+
+	irq = create_irq_nr(NR_IRQS_LEGACY, nid);
+	if (irq == 0)
+		irq = -1;
+
+	return irq;
+}
+
+void destroy_irq(unsigned int irq)
+{
+	unsigned long flags;
+
+	dynamic_irq_cleanup(irq);
+
+	raw_spin_lock_irqsave(&vector_lock, flags);
+	__clear_bit(irq, intc_irq_map);
+	raw_spin_unlock_irqrestore(&vector_lock, flags);
+}
+
+int reserve_irq_vector(unsigned int irq)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	raw_spin_lock_irqsave(&vector_lock, flags);
+	if (test_and_set_bit(irq, intc_irq_map))
+		ret = -EBUSY;
+	raw_spin_unlock_irqrestore(&vector_lock, flags);
+
+	return ret;
+}
+
+void reserve_intc_vectors(struct intc_vect *vectors, unsigned int nr_vecs)
+{
+	unsigned long flags;
+	int i;
+
+	raw_spin_lock_irqsave(&vector_lock, flags);
+	for (i = 0; i < nr_vecs; i++)
+		__set_bit(evt2irq(vectors[i].vect), intc_irq_map);
+	raw_spin_unlock_irqrestore(&vector_lock, flags);
+}
+
+void reserve_irq_legacy(void)
+{
+	unsigned long flags;
+	int i, j;
+
+	raw_spin_lock_irqsave(&vector_lock, flags);
+	j = find_first_bit(intc_irq_map, nr_irqs);
+	for (i = 0; i < j; i++)
+		__set_bit(i, intc_irq_map);
+	raw_spin_unlock_irqrestore(&vector_lock, flags);
+}

drivers/sh/intc/handle.c

+/*
+ * Shared interrupt handling code for IPR and INTC2 types of IRQs.
+ *
+ * Copyright (C) 2007, 2008 Magnus Damm
+ * Copyright (C) 2009, 2010 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/spinlock.h>
+#include "internals.h"
+
+static unsigned long ack_handle[NR_IRQS];
+
+static intc_enum __init intc_grp_id(struct intc_desc *desc,
+				    intc_enum enum_id)
+{
+	struct intc_group *g = desc->hw.groups;
+	unsigned int i, j;
+
+	for (i = 0; g && enum_id && i < desc->hw.nr_groups; i++) {
+		g = desc->hw.groups + i;
+
+		for (j = 0; g->enum_ids[j]; j++) {
+			if (g->enum_ids[j] != enum_id)
+				continue;
+
+			return g->enum_id;
+		}
+	}
+
+	return 0;
+}
+
+static unsigned int __init _intc_mask_data(struct intc_desc *desc,
+					   struct intc_desc_int *d,
+					   intc_enum enum_id,
+					   unsigned int *reg_idx,
+					   unsigned int *fld_idx)
+{
+	struct intc_mask_reg *mr = desc->hw.mask_regs;
+	unsigned int fn, mode;
+	unsigned long reg_e, reg_d;
+
+	while (mr && enum_id && *reg_idx < desc->hw.nr_mask_regs) {
+		mr = desc->hw.mask_regs + *reg_idx;
+
+		for (; *fld_idx < ARRAY_SIZE(mr->enum_ids); (*fld_idx)++) {
+			if (mr->enum_ids[*fld_idx] != enum_id)
+				continue;
+
+			if (mr->set_reg && mr->clr_reg) {
+				fn = REG_FN_WRITE_BASE;
+				mode = MODE_DUAL_REG;
+				reg_e = mr->clr_reg;
+				reg_d = mr->set_reg;
+			} else {
+				fn = REG_FN_MODIFY_BASE;
+				if (mr->set_reg) {
+					mode = MODE_ENABLE_REG;
+					reg_e = mr->set_reg;
+					reg_d = mr->set_reg;
+				} else {
+					mode = MODE_MASK_REG;
+					reg_e = mr->clr_reg;
+					reg_d = mr->clr_reg;
+				}
+			}
+
+			fn += (mr->reg_width >> 3) - 1;
+			return _INTC_MK(fn, mode,
+					intc_get_reg(d, reg_e),
+					intc_get_reg(d, reg_d),
+					1,
+					(mr->reg_width - 1) - *fld_idx);
+		}
+
+		*fld_idx = 0;
+		(*reg_idx)++;
+	}
+
+	return 0;
+}
+
+unsigned int __init
+intc_get_mask_handle(struct intc_desc *desc, struct intc_desc_int *d,
+		     intc_enum enum_id, int do_grps)
+{
+	unsigned int i = 0;
+	unsigned int j = 0;
+	unsigned int ret;
+
+	ret = _intc_mask_data(desc, d, enum_id, &i, &j);
+	if (ret)
+		return ret;
+
+	if (do_grps)
+		return intc_get_mask_handle(desc, d, intc_grp_id(desc, enum_id), 0);
+
+	return 0;
+}
+
+static unsigned int __init _intc_prio_data(struct intc_desc *desc,
+					   struct intc_desc_int *d,
+					   intc_enum enum_id,
+					   unsigned int *reg_idx,
+					   unsigned int *fld_idx)
+{
+	struct intc_prio_reg *pr = desc->hw.prio_regs;
+	unsigned int fn, n, mode, bit;
+	unsigned long reg_e, reg_d;
+
+	while (pr && enum_id && *reg_idx < desc->hw.nr_prio_regs) {
+		pr = desc->hw.prio_regs + *reg_idx;
+
+		for (; *fld_idx < ARRAY_SIZE(pr->enum_ids); (*fld_idx)++) {
+			if (pr->enum_ids[*fld_idx] != enum_id)
+				continue;
+
+			if (pr->set_reg && pr->clr_reg) {
+				fn = REG_FN_WRITE_BASE;
+				mode = MODE_PCLR_REG;
+				reg_e = pr->set_reg;
+				reg_d = pr->clr_reg;
+			} else {
+				fn = REG_FN_MODIFY_BASE;
+				mode = MODE_PRIO_REG;
+				if (!pr->set_reg)
+					BUG();
+				reg_e = pr->set_reg;
+				reg_d = pr->set_reg;
+			}
+
+			fn += (pr->reg_width >> 3) - 1;
+			n = *fld_idx + 1;
+
+			BUG_ON(n * pr->field_width > pr->reg_width);
+
+			bit = pr->reg_width - (n * pr->field_width);
+
+			return _INTC_MK(fn, mode,
+					intc_get_reg(d, reg_e),
+					intc_get_reg(d, reg_d),
+					pr->field_width, bit);
+		}
+
+		*fld_idx = 0;
+		(*reg_idx)++;
+	}
+
+	return 0;
+}
+
+unsigned int __init
+intc_get_prio_handle(struct intc_desc *desc, struct intc_desc_int *d,
+		     intc_enum enum_id, int do_grps)
+{
+	unsigned int i = 0;
+	unsigned int j = 0;
+	unsigned int ret;
+
+	ret = _intc_prio_data(desc, d, enum_id, &i, &j);
+	if (ret)
+		return ret;
+
+	if (do_grps)
+		return intc_get_prio_handle(desc, d, intc_grp_id(desc, enum_id), 0);
+
+	return 0;
+}
+
+static unsigned int __init intc_ack_data(struct intc_desc *desc,
+					  struct intc_desc_int *d,
+					  intc_enum enum_id)
+{
+	struct intc_mask_reg *mr = desc->hw.ack_regs;
+	unsigned int i, j, fn, mode;
+	unsigned long reg_e, reg_d;
+
+	for (i = 0; mr && enum_id && i < desc->hw.nr_ack_regs; i++) {
+		mr = desc->hw.ack_regs + i;
+
+		for (j = 0; j < ARRAY_SIZE(mr->enum_ids); j++) {
+			if (mr->enum_ids[j] != enum_id)
+				continue;
+
+			fn = REG_FN_MODIFY_BASE;
+			mode = MODE_ENABLE_REG;
+			reg_e = mr->set_reg;
+			reg_d = mr->set_reg;
+
+			fn += (mr->reg_width >> 3) - 1;
+			return _INTC_MK(fn, mode,
+					intc_get_reg(d, reg_e),
+					intc_get_reg(d, reg_d),
+					1,
+					(mr->reg_width - 1) - j);
+		}
+	}
+
+	return 0;
+}
+
+static void intc_enable_disable(struct intc_desc_int *d,
+				unsigned long handle, int do_enable)
+{
+	unsigned long addr;
+	unsigned int cpu;
+	unsigned long (*fn)(unsigned long, unsigned long,
+		   unsigned long (*)(unsigned long, unsigned long,
+				     unsigned long),
+		   unsigned int);
+
+	if (do_enable) {
+		for (cpu = 0; cpu < SMP_NR(d, _INTC_ADDR_E(handle)); cpu++) {
+			addr = INTC_REG(d, _INTC_ADDR_E(handle), cpu);
+			fn = intc_enable_noprio_fns[_INTC_MODE(handle)];
+			fn(addr, handle, intc_reg_fns[_INTC_FN(handle)], 0);
+		}
+	} else {
+		for (cpu = 0; cpu < SMP_NR(d, _INTC_ADDR_D(handle)); cpu++) {
+			addr = INTC_REG(d, _INTC_ADDR_D(handle), cpu);
+			fn = intc_disable_fns[_INTC_MODE(handle)];
+			fn(addr, handle, intc_reg_fns[_INTC_FN(handle)], 0);
+		}
+	}
+}
+
+void __init intc_enable_disable_enum(struct intc_desc *desc,
+				     struct intc_desc_int *d,
+				     intc_enum enum_id, int enable)
+{
+	unsigned int i, j, data;
+
+	/* go through and enable/disable all mask bits */
+	i = j = 0;
+	do {
+		data = _intc_mask_data(desc, d, enum_id, &i, &j);
+		if (data)
+			intc_enable_disable(d, data, enable);
+		j++;
+	} while (data);
+
+	/* go through and enable/disable all priority fields */
+	i = j = 0;
+	do {
+		data = _intc_prio_data(desc, d, enum_id, &i, &j);
+		if (data)
+			intc_enable_disable(d, data, enable);
+
+		j++;
+	} while (data);
+}
+
+unsigned int __init
+intc_get_sense_handle(struct intc_desc *desc, struct intc_desc_int *d,
+		      intc_enum enum_id)
+{
+	struct intc_sense_reg *sr = desc->hw.sense_regs;
+	unsigned int i, j, fn, bit;
+
+	for (i = 0; sr && enum_id && i < desc->hw.nr_sense_regs; i++) {
+		sr = desc->hw.sense_regs + i;
+
+		for (j = 0; j < ARRAY_SIZE(sr->enum_ids); j++) {
+			if (sr->enum_ids[j] != enum_id)
+				continue;
+
+			fn = REG_FN_MODIFY_BASE;
+			fn += (sr->reg_width >> 3) - 1;
+
+			BUG_ON((j + 1) * sr->field_width > sr->reg_width);
+
+			bit = sr->reg_width - ((j + 1) * sr->field_width);
+
+			return _INTC_MK(fn, 0, intc_get_reg(d, sr->reg),
+					0, sr->field_width, bit);
+		}
+	}
+
+	return 0;
+}
+
+
+void intc_set_ack_handle(unsigned int irq, struct intc_desc *desc,
+			 struct intc_desc_int *d, intc_enum id)
+{
+	unsigned long flags;
+
+	/*
+	 * Nothing to do for this IRQ.
+	 */
+	if (!desc->hw.ack_regs)
+		return;
+
+	raw_spin_lock_irqsave(&intc_big_lock, flags);
+	ack_handle[irq] = intc_ack_data(desc, d, id);
+	raw_spin_unlock_irqrestore(&intc_big_lock, flags);
+}
+
+unsigned long intc_get_ack_handle(unsigned int irq)
+{
+	return ack_handle[irq];
+}

drivers/sh/intc/internals.h

+#include <linux/sh_intc.h>
+#include <linux/irq.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/radix-tree.h>
+#include <linux/sysdev.h>
+
+#define _INTC_MK(fn, mode, addr_e, addr_d, width, shift) \
+	((shift) | ((width) << 5) | ((fn) << 9) | ((mode) << 13) | \
+	 ((addr_e) << 16) | ((addr_d << 24)))
+
+#define _INTC_SHIFT(h)		(h & 0x1f)
+#define _INTC_WIDTH(h)		((h >> 5) & 0xf)
+#define _INTC_FN(h)		((h >> 9) & 0xf)
+#define _INTC_MODE(h)		((h >> 13) & 0x7)
+#define _INTC_ADDR_E(h)		((h >> 16) & 0xff)
+#define _INTC_ADDR_D(h)		((h >> 24) & 0xff)
+
+#ifdef CONFIG_SMP
+#define IS_SMP(x)		(x.smp)
+#define INTC_REG(d, x, c)	(d->reg[(x)] + ((d->smp[(x)] & 0xff) * c))
+#define SMP_NR(d, x)		((d->smp[(x)] >> 8) ? (d->smp[(x)] >> 8) : 1)
+#else
+#define IS_SMP(x)		0
+#define INTC_REG(d, x, c)	(d->reg[(x)])
+#define SMP_NR(d, x)		1
+#endif
+
+struct intc_handle_int {
+	unsigned int irq;
+	unsigned long handle;
+};
+
+struct intc_window {
+	phys_addr_t phys;
+	void __iomem *virt;
+	unsigned long size;
+};
+
+struct intc_map_entry {
+	intc_enum enum_id;
+	struct intc_desc_int *desc;
+};
+
+struct intc_subgroup_entry {
+	unsigned int pirq;
+	intc_enum enum_id;
+	unsigned long handle;
+};
+
+struct intc_desc_int {
+	struct list_head list;
+	struct sys_device sysdev;
+	struct radix_tree_root tree;
+	pm_message_t state;
+	raw_spinlock_t lock;
+	unsigned int index;
+	unsigned long *reg;
+#ifdef CONFIG_SMP
+	unsigned long *smp;
+#endif
+	unsigned int nr_reg;
+	struct intc_handle_int *prio;
+	unsigned int nr_prio;
+	struct intc_handle_int *sense;
+	unsigned int nr_sense;
+	struct intc_window *window;
+	unsigned int nr_windows;
+	struct irq_chip chip;
+};
+
+
+enum {
+	REG_FN_ERR = 0,
+	REG_FN_TEST_BASE = 1,
+	REG_FN_WRITE_BASE = 5,
+	REG_FN_MODIFY_BASE = 9
+};
+
+enum {	MODE_ENABLE_REG = 0, /* Bit(s) set -> interrupt enabled */
+	MODE_MASK_REG,       /* Bit(s) set -> interrupt disabled */
+	MODE_DUAL_REG,       /* Two registers, set bit to enable / disable */
+	MODE_PRIO_REG,       /* Priority value written to enable interrupt */
+	MODE_PCLR_REG,       /* Above plus all bits set to disable interrupt */
+};
+
+static inline struct intc_desc_int *get_intc_desc(unsigned int irq)
+{
+	struct irq_chip *chip = get_irq_chip(irq);
+
+	return container_of(chip, struct intc_desc_int, chip);
+}
+
+/*
+ * Grumble.
+ */
+static inline void activate_irq(int irq)
+{
+#ifdef CONFIG_ARM
+	/* ARM requires an extra step to clear IRQ_NOREQUEST, which it
+	 * sets on behalf of every irq_chip.  Also sets IRQ_NOPROBE.
+	 */
+	set_irq_flags(irq, IRQF_VALID);
+#else
+	/* same effect on other architectures */
+	set_irq_noprobe(irq);
+#endif
+}
+
+/* access.c */
+extern unsigned long
+(*intc_reg_fns[])(unsigned long addr, unsigned long h, unsigned long data);
+
+extern unsigned long
+(*intc_enable_fns[])(unsigned long addr, unsigned long handle,
+		     unsigned long (*fn)(unsigned long,
+				unsigned long, unsigned long),
+		     unsigned int irq);
+extern unsigned long
+(*intc_disable_fns[])(unsigned long addr, unsigned long handle,
+		      unsigned long (*fn)(unsigned long,
+				unsigned long, unsigned long),
+		      unsigned int irq);
+extern unsigned long
+(*intc_enable_noprio_fns[])(unsigned long addr, unsigned long handle,
+		            unsigned long (*fn)(unsigned long,
+				unsigned long, unsigned long),
+			    unsigned int irq);
+
+unsigned long intc_phys_to_virt(struct intc_desc_int *d, unsigned long address);
+unsigned int intc_get_reg(struct intc_desc_int *d, unsigned long address);
+unsigned int intc_set_field_from_handle(unsigned int value,
+			    unsigned int field_value,
+			    unsigned int handle);
+unsigned long intc_get_field_from_handle(unsigned int value,
+					 unsigned int handle);
+
+/* balancing.c */
+#ifdef CONFIG_INTC_BALANCING
+void intc_balancing_enable(unsigned int irq);
+void intc_balancing_disable(unsigned int irq);
+void intc_set_dist_handle(unsigned int irq, struct intc_desc *desc,
+			  struct intc_desc_int *d, intc_enum id);
+#else
+void intc_balancing_enable(unsigned int irq) { }
+void intc_balancing_disable(unsigned int irq) { }
+void intc_set_dist_handle(unsigned int irq, struct intc_desc *desc,
+			  struct intc_desc_int *d, intc_enum id) { }
+#endif
+
+/* chip.c */
+extern struct irq_chip intc_irq_chip;
+void _intc_enable(unsigned int irq, unsigned long handle);
+
+/* core.c */
+extern struct list_head intc_list;
+extern raw_spinlock_t intc_big_lock;
+extern unsigned int nr_intc_controllers;
+extern struct sysdev_class intc_sysdev_class;
+
+unsigned int intc_get_dfl_prio_level(void);
+unsigned int intc_get_prio_level(unsigned int irq);
+void intc_set_prio_level(unsigned int irq, unsigned int level);
+
+/* handle.c */
+unsigned int intc_get_mask_handle(struct intc_desc *desc,
+				  struct intc_desc_int *d,
+				  intc_enum enum_id, int do_grps);
+unsigned int intc_get_prio_handle(struct intc_desc *desc,
+				  struct intc_desc_int *d,
+				  intc_enum enum_id, int do_grps);
+unsigned int intc_get_sense_handle(struct intc_desc *desc,
+				   struct intc_desc_int *d,
+				   intc_enum enum_id);
+void intc_set_ack_handle(unsigned int irq, struct intc_desc *desc,
+			 struct intc_desc_int *d, intc_enum id);
+unsigned long intc_get_ack_handle(unsigned int irq);
+void intc_enable_disable_enum(struct intc_desc *desc, struct intc_desc_int *d,
+			      intc_enum enum_id, int enable);
+
+/* virq.c */
+void intc_subgroup_init(struct intc_desc *desc, struct intc_desc_int *d);
+void intc_irq_xlate_set(unsigned int irq, intc_enum id, struct intc_desc_int *d);
+struct intc_map_entry *intc_irq_xlate_get(unsigned int irq);

drivers/sh/intc/userimask.c

+/*
+ * Support for hardware-assisted userspace interrupt masking.
+ *
+ * Copyright (C) 2010  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#define pr_fmt(fmt) "intc: " fmt
+
+#include <linux/errno.h>
+#include <linux/sysdev.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <asm/sizes.h>
+#include "internals.h"
+
+static void __iomem *uimask;
+
+static ssize_t
+show_intc_userimask(struct sysdev_class *cls,
+		    struct sysdev_class_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%d\n", (__raw_readl(uimask) >> 4) & 0xf);
+}
+
+static ssize_t
+store_intc_userimask(struct sysdev_class *cls,
+		     struct sysdev_class_attribute *attr,
+		     const char *buf, size_t count)
+{
+	unsigned long level;
+
+	level = simple_strtoul(buf, NULL, 10);
+
+	/*
+	 * Minimal acceptable IRQ levels are in the 2 - 16 range, but
+	 * these are chomped so as to not interfere with normal IRQs.
+	 *
+	 * Level 1 is a special case on some CPUs in that it's not
+	 * directly settable, but given that USERIMASK cuts off below a
+	 * certain level, we don't care about this limitation here.
+	 * Level 0 on the other hand equates to user masking disabled.
+	 *
+	 * We use the default priority level as a cut off so that only
+	 * special case opt-in IRQs can be mangled.
+	 */
+	if (level >= intc_get_dfl_prio_level())
+		return -EINVAL;
+
+	__raw_writel(0xa5 << 24 | level << 4, uimask);
+
+	return count;
+}
+
+static SYSDEV_CLASS_ATTR(userimask, S_IRUSR | S_IWUSR,
+			 show_intc_userimask, store_intc_userimask);
+
+
+static int __init userimask_sysdev_init(void)
+{
+	if (unlikely(!uimask))
+		return -ENXIO;
+
+	return sysdev_class_create_file(&intc_sysdev_class, &attr_userimask);
+}
+late_initcall(userimask_sysdev_init);
+
+int register_intc_userimask(unsigned long addr)
+{
+	if (unlikely(uimask))
+		return -EBUSY;
+
+	uimask = ioremap_nocache(addr, SZ_4K);
+	if (unlikely(!uimask))
+		return -ENOMEM;
+
+	pr_info("userimask support registered for levels 0 -> %d\n",
+		intc_get_dfl_prio_level() - 1);
+
+	return 0;
+}

drivers/sh/intc/virq-debugfs.c

+/*
+ * Support for virtual IRQ subgroups debugfs mapping.
+ *
+ * Copyright (C) 2010  Paul Mundt
+ *
+ * Modelled after arch/powerpc/kernel/irq.c.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/seq_file.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/debugfs.h>
+#include "internals.h"
+
+static int intc_irq_xlate_debug(struct seq_file *m, void *priv)
+{
+	int i;
+
+	seq_printf(m, "%-5s  %-7s  %-15s\n", "irq", "enum", "chip name");
+
+	for (i = 1; i < nr_irqs; i++) {
+		struct intc_map_entry *entry = intc_irq_xlate_get(i);
+		struct intc_desc_int *desc = entry->desc;
+
+		if (!desc)
+			continue;
+
+		seq_printf(m, "%5d  ", i);
+		seq_printf(m, "0x%05x  ", entry->enum_id);
+		seq_printf(m, "%-15s\n", desc->chip.name);
+	}
+
+	return 0;
+}
+
+static int intc_irq_xlate_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, intc_irq_xlate_debug, inode->i_private);
+}
+
+static const struct file_operations intc_irq_xlate_fops = {
+	.open = intc_irq_xlate_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+static int __init intc_irq_xlate_init(void)
+{
+	/*
+	 * XXX.. use arch_debugfs_dir here when all of the intc users are
+	 * converted.
+	 */
+	if (debugfs_create_file("intc_irq_xlate", S_IRUGO, NULL, NULL,
+				&intc_irq_xlate_fops) == NULL)
+		return -ENOMEM;
+
+	return 0;
+}
+fs_initcall(intc_irq_xlate_init);

drivers/sh/intc/virq.c

+/*
+ * Support for virtual IRQ subgroups.
+ *
+ * Copyright (C) 2010  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#define pr_fmt(fmt) "intc: " fmt
+
+#include <linux/slab.h>
+#include <linux/irq.h>
+#include <linux/list.h>
+#include <linux/radix-tree.h>
+#include <linux/spinlock.h>
+#include "internals.h"
+
+static struct intc_map_entry intc_irq_xlate[NR_IRQS];
+
+struct intc_virq_list {
+	unsigned int irq;
+	struct intc_virq_list *next;
+};
+
+#define for_each_virq(entry, head) \
+	for (entry = head; entry; entry = entry->next)
+
+/*
+ * Tags for the radix tree
+ */
+#define INTC_TAG_VIRQ_NEEDS_ALLOC	0
+
+void intc_irq_xlate_set(unsigned int irq, intc_enum id, struct intc_desc_int *d)
+{
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&intc_big_lock, flags);
+	intc_irq_xlate[irq].enum_id = id;
+	intc_irq_xlate[irq].desc = d;
+	raw_spin_unlock_irqrestore(&intc_big_lock, flags);
+}
+
+struct intc_map_entry *intc_irq_xlate_get(unsigned int irq)
+{
+	return intc_irq_xlate + irq;
+}
+
+int intc_irq_lookup(const char *chipname, intc_enum enum_id)
+{
+	struct intc_map_entry *ptr;
+	struct intc_desc_int *d;
+	int irq = -1;
+
+	list_for_each_entry(d, &intc_list, list) {
+		int tagged;
+
+		if (strcmp(d->chip.name, chipname) != 0)
+			continue;
+
+		/*
+		 * Catch early lookups for subgroup VIRQs that have not
+		 * yet been allocated an IRQ. This already includes a
+		 * fast-path out if the tree is untagged, so there is no
+		 * need to explicitly test the root tree.
+		 */
+		tagged = radix_tree_tag_get(&d->tree, enum_id,
+					    INTC_TAG_VIRQ_NEEDS_ALLOC);
+		if (unlikely(tagged))
+			break;
+
+		ptr = radix_tree_lookup(&d->tree, enum_id);
+		if (ptr) {
+			irq = ptr - intc_irq_xlate;
+			break;
+		}
+	}
+
+	return irq;
+}
+EXPORT_SYMBOL_GPL(intc_irq_lookup);
+
+static int add_virq_to_pirq(unsigned int irq, unsigned int virq)
+{
+	struct intc_virq_list **last, *entry;
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	/* scan for duplicates */
+	last = (struct intc_virq_list **)&desc->handler_data;
+	for_each_virq(entry, desc->handler_data) {
+		if (entry->irq == virq)
+			return 0;
+		last = &entry->next;
+	}
+
+	entry = kzalloc(sizeof(struct intc_virq_list), GFP_ATOMIC);
+	if (!entry) {
+		pr_err("can't allocate VIRQ mapping for %d\n", virq);
+		return -ENOMEM;
+	}
+
+	entry->irq = virq;
+
+	*last = entry;
+
+	return 0;
+}
+
+static void intc_virq_handler(unsigned int irq, struct irq_desc *desc)
+{
+	struct intc_virq_list *entry, *vlist = get_irq_data(irq);
+	struct intc_desc_int *d = get_intc_desc(irq);
+
+	desc->chip->mask_ack(irq);
+
+	for_each_virq(entry, vlist) {
+		unsigned long addr, handle;
+
+		handle = (unsigned long)get_irq_data(entry->irq);
+		addr = INTC_REG(d, _INTC_ADDR_E(handle), 0);
+
+		if (intc_reg_fns[_INTC_FN(handle)](addr, handle, 0))
+			generic_handle_irq(entry->irq);
+	}
+
+	desc->chip->unmask(irq);
+}
+
+static unsigned long __init intc_subgroup_data(struct intc_subgroup *subgroup,
+					       struct intc_desc_int *d,
+					       unsigned int index)
+{
+	unsigned int fn = REG_FN_TEST_BASE + (subgroup->reg_width >> 3) - 1;
+
+	return _INTC_MK(fn, MODE_ENABLE_REG, intc_get_reg(d, subgroup->reg),
+			0, 1, (subgroup->reg_width - 1) - index);
+}
+
+static void __init intc_subgroup_init_one(struct intc_desc *desc,
+					  struct intc_desc_int *d,
+					  struct intc_subgroup *subgroup)
+{
+	struct intc_map_entry *mapped;
+	unsigned int pirq;
+	unsigned long flags;
+	int i;
+
+	mapped = radix_tree_lookup(&d->tree, subgroup->parent_id);
+	if (!mapped) {
+		WARN_ON(1);
+		return;
+	}
+
+	pirq = mapped - intc_irq_xlate;
+
+	raw_spin_lock_irqsave(&d->lock, flags);
+
+	for (i = 0; i < ARRAY_SIZE(subgroup->enum_ids); i++) {
+		struct intc_subgroup_entry *entry;
+		int err;
+
+		if (!subgroup->enum_ids[i])
+			continue;
+
+		entry = kmalloc(sizeof(*entry), GFP_NOWAIT);
+		if (!entry)
+			break;
+
+		entry->pirq = pirq;
+		entry->enum_id = subgroup->enum_ids[i];
+		entry->handle = intc_subgroup_data(subgroup, d, i);
+
+		err = radix_tree_insert(&d->tree, entry->enum_id, entry);
+		if (unlikely(err < 0))
+			break;
+
+		radix_tree_tag_set(&d->tree, entry->enum_id,
+				   INTC_TAG_VIRQ_NEEDS_ALLOC);
+	}
+
+	raw_spin_unlock_irqrestore(&d->lock, flags);
+}
+
+void __init intc_subgroup_init(struct intc_desc *desc, struct intc_desc_int *d)
+{
+	int i;
+
+	if (!desc->hw.subgroups)
+		return;
+
+	for (i = 0; i < desc->hw.nr_subgroups; i++)
+		intc_subgroup_init_one(desc, d, desc->hw.subgroups + i);
+}
+
+static void __init intc_subgroup_map(struct intc_desc_int *d)
+{
+	struct intc_subgroup_entry *entries[32];
+	unsigned long flags;
+	unsigned int nr_found;
+	int i;
+
+	raw_spin_lock_irqsave(&d->lock, flags);
+
+restart:
+	nr_found = radix_tree_gang_lookup_tag_slot(&d->tree,
+			(void ***)entries, 0, ARRAY_SIZE(entries),
+			INTC_TAG_VIRQ_NEEDS_ALLOC);
+
+	for (i = 0; i < nr_found; i++) {
+		struct intc_subgroup_entry *entry;
+		int irq;
+
+		entry = radix_tree_deref_slot((void **)entries[i]);
+		if (unlikely(!entry))
+			continue;
+		if (unlikely(entry == RADIX_TREE_RETRY))
+			goto restart;
+
+		irq = create_irq();
+		if (unlikely(irq < 0)) {
+			pr_err("no more free IRQs, bailing..\n");
+			break;
+		}
+
+		pr_info("Setting up a chained VIRQ from %d -> %d\n",
+			irq, entry->pirq);
+
+		intc_irq_xlate_set(irq, entry->enum_id, d);
+
+		set_irq_chip_and_handler_name(irq, get_irq_chip(entry->pirq),
+					      handle_simple_irq, "virq");
+		set_irq_chip_data(irq, get_irq_chip_data(entry->pirq));
+
+		set_irq_data(irq, (void *)entry->handle);
+
+		set_irq_chained_handler(entry->pirq, intc_virq_handler);
+		add_virq_to_pirq(entry->pirq, irq);
+
+		radix_tree_tag_clear(&d->tree, entry->enum_id,
+				     INTC_TAG_VIRQ_NEEDS_ALLOC);
+		radix_tree_replace_slot((void **)entries[i],
+					&intc_irq_xlate[irq]);
+	}
+
+	raw_spin_unlock_irqrestore(&d->lock, flags);
+}
+
+void __init intc_finalize(void)
+{
+	struct intc_desc_int *d;
+
+	list_for_each_entry(d, &intc_list, list)
+		if (radix_tree_tagged(&d->tree, INTC_TAG_VIRQ_NEEDS_ALLOC))
+			intc_subgroup_map(d);
+}

include/linux/sh_intc.h

@@ -114,7 +114,7 @@ struct intc_desc symbol __initdata = {					\
 			   prio_regs, sense_regs, ack_regs),		\
 }
 
-int __init register_intc_controller(struct intc_desc *desc);
+int register_intc_controller(struct intc_desc *desc);
 void reserve_intc_vectors(struct intc_vect *vectors, unsigned int nr_vecs);
 int intc_set_priority(unsigned int irq, unsigned int prio);
 int intc_irq_lookup(const char *chipname, intc_enum enum_id);