]> bbs.cooldavid.org Git - net-next-2.6.git/commitdiff
genirq: Distangle kernel/irq/handle.c
authorThomas Gleixner <tglx@linutronix.de>
Wed, 22 Sep 2010 15:09:43 +0000 (17:09 +0200)
committerThomas Gleixner <tglx@linutronix.de>
Tue, 12 Oct 2010 14:39:05 +0000 (16:39 +0200)
kernel/irq/handle.c has become a dumpground for random code in random
order. Split out the irq descriptor management and the dummy irq_chip
implementation into separate files. Cleanup the include maze while at
it.

No code change.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
kernel/irq/Makefile
kernel/irq/dummychip.c [new file with mode: 0644]
kernel/irq/handle.c
kernel/irq/irqdesc.c [new file with mode: 0644]

index 7d047808419da88e273fba8cd1efd88d9aaa5bcd..1eaab0da56dbc692239deac0bf0f3118bfef0ded 100644 (file)
@@ -1,5 +1,5 @@
 
-obj-y := handle.o manage.o spurious.o resend.o chip.o devres.o
+obj-y := irqdesc.o handle.o manage.o spurious.o resend.o chip.o dummychip.o devres.o
 obj-$(CONFIG_GENERIC_IRQ_PROBE) += autoprobe.o
 obj-$(CONFIG_PROC_FS) += proc.o
 obj-$(CONFIG_GENERIC_PENDING_IRQ) += migration.o
diff --git a/kernel/irq/dummychip.c b/kernel/irq/dummychip.c
new file mode 100644 (file)
index 0000000..918dea9
--- /dev/null
@@ -0,0 +1,68 @@
+/*
+ * Copyright (C) 1992, 1998-2006 Linus Torvalds, Ingo Molnar
+ * Copyright (C) 2005-2006, Thomas Gleixner, Russell King
+ *
+ * This file contains the dummy interrupt chip implementation
+ */
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+
+#include "internals.h"
+
+/*
+ * What should we do if we get a hw irq event on an illegal vector?
+ * Each architecture has to answer this themself.
+ */
+static void ack_bad(struct irq_data *data)
+{
+       struct irq_desc *desc = irq_data_to_desc(data);
+
+       print_irq_desc(data->irq, desc);
+       ack_bad_irq(data->irq);
+}
+
+/*
+ * NOP functions
+ */
+static void noop(struct irq_data *data) { }
+
+static unsigned int noop_ret(struct irq_data *data)
+{
+       return 0;
+}
+
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_CRUFT
+static void compat_noop(unsigned int irq) { }
+#define END_INIT .end = compat_noop
+#else
+#define END_INIT
+#endif
+
+/*
+ * Generic no controller implementation
+ */
+struct irq_chip no_irq_chip = {
+       .name           = "none",
+       .irq_startup    = noop_ret,
+       .irq_shutdown   = noop,
+       .irq_enable     = noop,
+       .irq_disable    = noop,
+       .irq_ack        = ack_bad,
+       END_INIT
+};
+
+/*
+ * Generic dummy implementation which can be used for
+ * real dumb interrupt sources
+ */
+struct irq_chip dummy_irq_chip = {
+       .name           = "dummy",
+       .irq_startup    = noop_ret,
+       .irq_shutdown   = noop,
+       .irq_enable     = noop,
+       .irq_disable    = noop,
+       .irq_ack        = noop,
+       .irq_mask       = noop,
+       .irq_unmask     = noop,
+       END_INIT
+};
index 3fcef37154a18d98a0b2ee7e35c80cbff81741dd..e2347eb6330682501e99bb3214ccfa9015a8a521 100644 (file)
  */
 
 #include <linux/irq.h>
-#include <linux/sched.h>
-#include <linux/slab.h>
-#include <linux/module.h>
 #include <linux/random.h>
+#include <linux/sched.h>
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
-#include <linux/rculist.h>
-#include <linux/hash.h>
-#include <linux/radix-tree.h>
+
 #include <trace/events/irq.h>
 
 #include "internals.h"
 
-/*
- * lockdep: we want to handle all irq_desc locks as a single lock-class:
- */
-struct lock_class_key irq_desc_lock_class;
-
 /**
  * handle_bad_irq - handle spurious and unhandled irqs
  * @irq:       the interrupt number
@@ -43,308 +34,6 @@ void handle_bad_irq(unsigned int irq, struct irq_desc *desc)
        ack_bad_irq(irq);
 }
 
-#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
-static void __init init_irq_default_affinity(void)
-{
-       alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT);
-       cpumask_setall(irq_default_affinity);
-}
-#else
-static void __init init_irq_default_affinity(void)
-{
-}
-#endif
-
-/*
- * Linux has a controller-independent interrupt architecture.
- * Every controller has a 'controller-template', that is used
- * by the main code to do the right thing. Each driver-visible
- * interrupt source is transparently wired to the appropriate
- * controller. Thus drivers need not be aware of the
- * interrupt-controller.
- *
- * The code is designed to be easily extended with new/different
- * interrupt controllers, without having to do assembly magic or
- * having to touch the generic code.
- *
- * Controller mappings for all interrupt sources:
- */
-int nr_irqs = NR_IRQS;
-EXPORT_SYMBOL_GPL(nr_irqs);
-
-#ifdef CONFIG_SPARSE_IRQ
-
-static struct irq_desc irq_desc_init = {
-       .status         = IRQ_DISABLED,
-       .handle_irq     = handle_bad_irq,
-       .depth          = 1,
-       .lock           = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
-};
-
-void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr)
-{
-       void *ptr;
-
-       ptr = kzalloc_node(nr * sizeof(*desc->kstat_irqs),
-                          GFP_ATOMIC, node);
-
-       /*
-        * don't overwite if can not get new one
-        * init_copy_kstat_irqs() could still use old one
-        */
-       if (ptr) {
-               printk(KERN_DEBUG "  alloc kstat_irqs on node %d\n", node);
-               desc->kstat_irqs = ptr;
-       }
-}
-
-static void init_one_irq_desc(int irq, struct irq_desc *desc, int node)
-{
-       memcpy(desc, &irq_desc_init, sizeof(struct irq_desc));
-
-       raw_spin_lock_init(&desc->lock);
-       desc->irq_data.irq = irq;
-#ifdef CONFIG_SMP
-       desc->irq_data.node = node;
-#endif
-       lockdep_set_class(&desc->lock, &irq_desc_lock_class);
-       init_kstat_irqs(desc, node, nr_cpu_ids);
-       if (!desc->kstat_irqs) {
-               printk(KERN_ERR "can not alloc kstat_irqs\n");
-               BUG_ON(1);
-       }
-       if (!alloc_desc_masks(desc, node, false)) {
-               printk(KERN_ERR "can not alloc irq_desc cpumasks\n");
-               BUG_ON(1);
-       }
-       init_desc_masks(desc);
-       arch_init_chip_data(desc, node);
-}
-
-/*
- * Protect the sparse_irqs:
- */
-DEFINE_RAW_SPINLOCK(sparse_irq_lock);
-
-static RADIX_TREE(irq_desc_tree, GFP_ATOMIC);
-
-static void set_irq_desc(unsigned int irq, struct irq_desc *desc)
-{
-       radix_tree_insert(&irq_desc_tree, irq, desc);
-}
-
-struct irq_desc *irq_to_desc(unsigned int irq)
-{
-       return radix_tree_lookup(&irq_desc_tree, irq);
-}
-
-void replace_irq_desc(unsigned int irq, struct irq_desc *desc)
-{
-       void **ptr;
-
-       ptr = radix_tree_lookup_slot(&irq_desc_tree, irq);
-       if (ptr)
-               radix_tree_replace_slot(ptr, desc);
-}
-
-static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_smp = {
-       [0 ... NR_IRQS_LEGACY-1] = {
-               .status         = IRQ_DISABLED,
-               .handle_irq     = handle_bad_irq,
-               .depth          = 1,
-               .lock           = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
-       }
-};
-
-static unsigned int *kstat_irqs_legacy;
-
-int __init early_irq_init(void)
-{
-       struct irq_desc *desc;
-       int legacy_count;
-       int node;
-       int i;
-
-       init_irq_default_affinity();
-
-        /* initialize nr_irqs based on nr_cpu_ids */
-       arch_probe_nr_irqs();
-       printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d\n", NR_IRQS, nr_irqs);
-
-       desc = irq_desc_legacy;
-       legacy_count = ARRAY_SIZE(irq_desc_legacy);
-       node = first_online_node;
-
-       /* allocate based on nr_cpu_ids */
-       kstat_irqs_legacy = kzalloc_node(NR_IRQS_LEGACY * nr_cpu_ids *
-                                         sizeof(int), GFP_NOWAIT, node);
-
-       irq_desc_init.irq_data.chip = &no_irq_chip;
-
-       for (i = 0; i < legacy_count; i++) {
-               desc[i].irq_data.irq = i;
-               desc[i].irq_data.chip = &no_irq_chip;
-#ifdef CONFIG_SMP
-               desc[i].irq_data.node = node;
-#endif
-               desc[i].kstat_irqs = kstat_irqs_legacy + i * nr_cpu_ids;
-               lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
-               alloc_desc_masks(&desc[i], node, true);
-               init_desc_masks(&desc[i]);
-               set_irq_desc(i, &desc[i]);
-       }
-
-       return arch_early_irq_init();
-}
-
-struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node)
-{
-       struct irq_desc *desc;
-       unsigned long flags;
-
-       if (irq >= nr_irqs) {
-               WARN(1, "irq (%d) >= nr_irqs (%d) in irq_to_desc_alloc\n",
-                       irq, nr_irqs);
-               return NULL;
-       }
-
-       desc = irq_to_desc(irq);
-       if (desc)
-               return desc;
-
-       raw_spin_lock_irqsave(&sparse_irq_lock, flags);
-
-       /* We have to check it to avoid races with another CPU */
-       desc = irq_to_desc(irq);
-       if (desc)
-               goto out_unlock;
-
-       desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node);
-
-       printk(KERN_DEBUG "  alloc irq_desc for %d on node %d\n", irq, node);
-       if (!desc) {
-               printk(KERN_ERR "can not alloc irq_desc\n");
-               BUG_ON(1);
-       }
-       init_one_irq_desc(irq, desc, node);
-
-       set_irq_desc(irq, desc);
-
-out_unlock:
-       raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
-
-       return desc;
-}
-
-#else /* !CONFIG_SPARSE_IRQ */
-
-struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
-       [0 ... NR_IRQS-1] = {
-               .status         = IRQ_DISABLED,
-               .handle_irq     = handle_bad_irq,
-               .depth          = 1,
-               .lock           = __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock),
-       }
-};
-
-static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS];
-int __init early_irq_init(void)
-{
-       struct irq_desc *desc;
-       int count;
-       int i;
-
-       init_irq_default_affinity();
-
-       printk(KERN_INFO "NR_IRQS:%d\n", NR_IRQS);
-
-       desc = irq_desc;
-       count = ARRAY_SIZE(irq_desc);
-
-       for (i = 0; i < count; i++) {
-               desc[i].irq_data.irq = i;
-               desc[i].irq_data.chip = &no_irq_chip;
-               alloc_desc_masks(&desc[i], 0, true);
-               init_desc_masks(&desc[i]);
-               desc[i].kstat_irqs = kstat_irqs_all[i];
-       }
-       return arch_early_irq_init();
-}
-
-struct irq_desc *irq_to_desc(unsigned int irq)
-{
-       return (irq < NR_IRQS) ? irq_desc + irq : NULL;
-}
-
-struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node)
-{
-       return irq_to_desc(irq);
-}
-#endif /* !CONFIG_SPARSE_IRQ */
-
-void clear_kstat_irqs(struct irq_desc *desc)
-{
-       memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
-}
-
-/*
- * What should we do if we get a hw irq event on an illegal vector?
- * Each architecture has to answer this themself.
- */
-static void ack_bad(struct irq_data *data)
-{
-       struct irq_desc *desc = irq_data_to_desc(data);
-
-       print_irq_desc(data->irq, desc);
-       ack_bad_irq(data->irq);
-}
-
-/*
- * NOP functions
- */
-static void noop(struct irq_data *data) { }
-
-static unsigned int noop_ret(struct irq_data *data)
-{
-       return 0;
-}
-
-#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
-static void compat_noop(unsigned int irq) { }
-#define END_INIT .end = compat_noop
-#else
-#define END_INIT
-#endif
-
-/*
- * Generic no controller implementation
- */
-struct irq_chip no_irq_chip = {
-       .name           = "none",
-       .irq_startup    = noop_ret,
-       .irq_shutdown   = noop,
-       .irq_enable     = noop,
-       .irq_disable    = noop,
-       .irq_ack        = ack_bad,
-       END_INIT
-};
-
-/*
- * Generic dummy implementation which can be used for
- * real dumb interrupt sources
- */
-struct irq_chip dummy_irq_chip = {
-       .name           = "dummy",
-       .irq_startup    = noop_ret,
-       .irq_shutdown   = noop,
-       .irq_enable     = noop,
-       .irq_disable    = noop,
-       .irq_ack        = noop,
-       .irq_mask       = noop,
-       .irq_unmask     = noop,
-       END_INIT
-};
-
 /*
  * Special, empty irq handler:
  */
@@ -540,21 +229,3 @@ out:
        return 1;
 }
 #endif
-
-void early_init_irq_lock_class(void)
-{
-       struct irq_desc *desc;
-       int i;
-
-       for_each_irq_desc(i, desc) {
-               lockdep_set_class(&desc->lock, &irq_desc_lock_class);
-       }
-}
-
-unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
-{
-       struct irq_desc *desc = irq_to_desc(irq);
-       return desc ? desc->kstat_irqs[cpu] : 0;
-}
-EXPORT_SYMBOL(kstat_irqs_cpu);
-
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
new file mode 100644 (file)
index 0000000..fbf8cfa
--- /dev/null
@@ -0,0 +1,269 @@
+/*
+ * Copyright (C) 1992, 1998-2006 Linus Torvalds, Ingo Molnar
+ * Copyright (C) 2005-2006, Thomas Gleixner, Russell King
+ *
+ * This file contains the interrupt descriptor management code
+ *
+ * Detailed information is available in Documentation/DocBook/genericirq
+ *
+ */
+#include <linux/irq.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/kernel_stat.h>
+#include <linux/radix-tree.h>
+
+#include "internals.h"
+
+/*
+ * lockdep: we want to handle all irq_desc locks as a single lock-class:
+ */
+struct lock_class_key irq_desc_lock_class;
+
+#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
+static void __init init_irq_default_affinity(void)
+{
+       alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT);
+       cpumask_setall(irq_default_affinity);
+}
+#else
+static void __init init_irq_default_affinity(void)
+{
+}
+#endif
+
+int nr_irqs = NR_IRQS;
+EXPORT_SYMBOL_GPL(nr_irqs);
+
+#ifdef CONFIG_SPARSE_IRQ
+
+static struct irq_desc irq_desc_init = {
+       .status         = IRQ_DISABLED,
+       .handle_irq     = handle_bad_irq,
+       .depth          = 1,
+       .lock           = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
+};
+
+void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr)
+{
+       void *ptr;
+
+       ptr = kzalloc_node(nr * sizeof(*desc->kstat_irqs),
+                          GFP_ATOMIC, node);
+
+       /*
+        * don't overwite if can not get new one
+        * init_copy_kstat_irqs() could still use old one
+        */
+       if (ptr) {
+               printk(KERN_DEBUG "  alloc kstat_irqs on node %d\n", node);
+               desc->kstat_irqs = ptr;
+       }
+}
+
+static void init_one_irq_desc(int irq, struct irq_desc *desc, int node)
+{
+       memcpy(desc, &irq_desc_init, sizeof(struct irq_desc));
+
+       raw_spin_lock_init(&desc->lock);
+       desc->irq_data.irq = irq;
+#ifdef CONFIG_SMP
+       desc->irq_data.node = node;
+#endif
+       lockdep_set_class(&desc->lock, &irq_desc_lock_class);
+       init_kstat_irqs(desc, node, nr_cpu_ids);
+       if (!desc->kstat_irqs) {
+               printk(KERN_ERR "can not alloc kstat_irqs\n");
+               BUG_ON(1);
+       }
+       if (!alloc_desc_masks(desc, node, false)) {
+               printk(KERN_ERR "can not alloc irq_desc cpumasks\n");
+               BUG_ON(1);
+       }
+       init_desc_masks(desc);
+       arch_init_chip_data(desc, node);
+}
+
+/*
+ * Protect the sparse_irqs:
+ */
+DEFINE_RAW_SPINLOCK(sparse_irq_lock);
+
+static RADIX_TREE(irq_desc_tree, GFP_ATOMIC);
+
+static void set_irq_desc(unsigned int irq, struct irq_desc *desc)
+{
+       radix_tree_insert(&irq_desc_tree, irq, desc);
+}
+
+struct irq_desc *irq_to_desc(unsigned int irq)
+{
+       return radix_tree_lookup(&irq_desc_tree, irq);
+}
+
+void replace_irq_desc(unsigned int irq, struct irq_desc *desc)
+{
+       void **ptr;
+
+       ptr = radix_tree_lookup_slot(&irq_desc_tree, irq);
+       if (ptr)
+               radix_tree_replace_slot(ptr, desc);
+}
+
+static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_smp = {
+       [0 ... NR_IRQS_LEGACY-1] = {
+               .status         = IRQ_DISABLED,
+               .handle_irq     = handle_bad_irq,
+               .depth          = 1,
+               .lock           = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
+       }
+};
+
+static unsigned int *kstat_irqs_legacy;
+
+int __init early_irq_init(void)
+{
+       struct irq_desc *desc;
+       int legacy_count;
+       int node;
+       int i;
+
+       init_irq_default_affinity();
+
+        /* initialize nr_irqs based on nr_cpu_ids */
+       arch_probe_nr_irqs();
+       printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d\n", NR_IRQS, nr_irqs);
+
+       desc = irq_desc_legacy;
+       legacy_count = ARRAY_SIZE(irq_desc_legacy);
+       node = first_online_node;
+
+       /* allocate based on nr_cpu_ids */
+       kstat_irqs_legacy = kzalloc_node(NR_IRQS_LEGACY * nr_cpu_ids *
+                                         sizeof(int), GFP_NOWAIT, node);
+
+       irq_desc_init.irq_data.chip = &no_irq_chip;
+
+       for (i = 0; i < legacy_count; i++) {
+               desc[i].irq_data.irq = i;
+               desc[i].irq_data.chip = &no_irq_chip;
+#ifdef CONFIG_SMP
+               desc[i].irq_data.node = node;
+#endif
+               desc[i].kstat_irqs = kstat_irqs_legacy + i * nr_cpu_ids;
+               lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
+               alloc_desc_masks(&desc[i], node, true);
+               init_desc_masks(&desc[i]);
+               set_irq_desc(i, &desc[i]);
+       }
+
+       return arch_early_irq_init();
+}
+
+struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node)
+{
+       struct irq_desc *desc;
+       unsigned long flags;
+
+       if (irq >= nr_irqs) {
+               WARN(1, "irq (%d) >= nr_irqs (%d) in irq_to_desc_alloc\n",
+                       irq, nr_irqs);
+               return NULL;
+       }
+
+       desc = irq_to_desc(irq);
+       if (desc)
+               return desc;
+
+       raw_spin_lock_irqsave(&sparse_irq_lock, flags);
+
+       /* We have to check it to avoid races with another CPU */
+       desc = irq_to_desc(irq);
+       if (desc)
+               goto out_unlock;
+
+       desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node);
+
+       printk(KERN_DEBUG "  alloc irq_desc for %d on node %d\n", irq, node);
+       if (!desc) {
+               printk(KERN_ERR "can not alloc irq_desc\n");
+               BUG_ON(1);
+       }
+       init_one_irq_desc(irq, desc, node);
+
+       set_irq_desc(irq, desc);
+
+out_unlock:
+       raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+
+       return desc;
+}
+
+#else /* !CONFIG_SPARSE_IRQ */
+
+struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
+       [0 ... NR_IRQS-1] = {
+               .status         = IRQ_DISABLED,
+               .handle_irq     = handle_bad_irq,
+               .depth          = 1,
+               .lock           = __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock),
+       }
+};
+
+static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS];
+int __init early_irq_init(void)
+{
+       struct irq_desc *desc;
+       int count;
+       int i;
+
+       init_irq_default_affinity();
+
+       printk(KERN_INFO "NR_IRQS:%d\n", NR_IRQS);
+
+       desc = irq_desc;
+       count = ARRAY_SIZE(irq_desc);
+
+       for (i = 0; i < count; i++) {
+               desc[i].irq_data.irq = i;
+               desc[i].irq_data.chip = &no_irq_chip;
+               alloc_desc_masks(&desc[i], 0, true);
+               init_desc_masks(&desc[i]);
+               desc[i].kstat_irqs = kstat_irqs_all[i];
+       }
+       return arch_early_irq_init();
+}
+
+struct irq_desc *irq_to_desc(unsigned int irq)
+{
+       return (irq < NR_IRQS) ? irq_desc + irq : NULL;
+}
+
+struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node)
+{
+       return irq_to_desc(irq);
+}
+#endif /* !CONFIG_SPARSE_IRQ */
+
+void clear_kstat_irqs(struct irq_desc *desc)
+{
+       memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
+}
+
+void early_init_irq_lock_class(void)
+{
+       struct irq_desc *desc;
+       int i;
+
+       for_each_irq_desc(i, desc) {
+               lockdep_set_class(&desc->lock, &irq_desc_lock_class);
+       }
+}
+
+unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
+{
+       struct irq_desc *desc = irq_to_desc(irq);
+       return desc ? desc->kstat_irqs[cpu] : 0;
+}
+EXPORT_SYMBOL(kstat_irqs_cpu);