irq_work: Add generic hardirq context callbacks

Provide a mechanism that allows running code in IRQ context. It is
most useful for NMI code that needs to interact with the rest of the
system -- like wakeup a task to drain buffers.

Perf currently has such a mechanism, so extract that and provide it as
a generic feature, independent of perf so that others may also
benefit.

The IRQ context callback is generated through self-IPIs where
possible, or on architectures like powerpc the decrementer (the
built-in timer facility) is set to generate an interrupt immediately.

Architectures that don't have anything like this get to do with a
callback from the timer tick. These architectures can call
irq_work_run() at the tail of any IRQ handlers that might enqueue such
work (like the perf IRQ handler) to avoid undue latencies in
processing the work.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Kyle McMartin <kyle@mcmartin.ca>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
[ various fixes ]
Signed-off-by: Huang Ying <ying.huang@intel.com>
LKML-Reference: <1287036094.7768.291.camel@yhuang-dev>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

32 files changed, 105 insertions, 132 deletions

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index b9647bb66d13..d04ccd73af45 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -9,6 +9,7 @@ config ALPHA
         select HAVE_IDE
         select HAVE_OPROFILE
         select HAVE_SYSCALL_WRAPPERS
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select HAVE_DMA_ATTRS
         help
diff --git a/arch/alpha/include/asm/perf_event.h b/arch/alpha/include/asm/perf_event.h
index 4157cd3c44a9..fe792ca818f6 100644
--- a/arch/alpha/include/asm/perf_event.h
+++ b/arch/alpha/include/asm/perf_event.h
@@ -1,11 +1,6 @@
 #ifndef __ASM_ALPHA_PERF_EVENT_H
 #define __ASM_ALPHA_PERF_EVENT_H
 
-/* Alpha only supports software events through this interface. */
-extern void set_perf_event_pending(void);
-
-#define PERF_EVENT_INDEX_OFFSET 0
-
 #ifdef CONFIG_PERF_EVENTS
 extern void init_hw_perf_events(void);
 #else
diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c
index 396af1799ea4..0f1d8493cfca 100644
--- a/arch/alpha/kernel/time.c
+++ b/arch/alpha/kernel/time.c
@@ -41,7 +41,7 @@
 #include <linux/init.h>
 #include <linux/bcd.h>
 #include <linux/profile.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -83,25 +83,25 @@ static struct {
 
 unsigned long est_cycle_freq;
 
-#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_IRQ_WORK
 
-DEFINE_PER_CPU(u8, perf_event_pending);
+DEFINE_PER_CPU(u8, irq_work_pending);
 
-#define set_perf_event_pending_flag()  __get_cpu_var(perf_event_pending) = 1
-#define test_perf_event_pending()      __get_cpu_var(perf_event_pending)
-#define clear_perf_event_pending()     __get_cpu_var(perf_event_pending) = 0
+#define set_irq_work_pending_flag()  __get_cpu_var(irq_work_pending) = 1
+#define test_irq_work_pending()      __get_cpu_var(irq_work_pending)
+#define clear_irq_work_pending()     __get_cpu_var(irq_work_pending) = 0
 
-void set_perf_event_pending(void)
+void set_irq_work_pending(void)
 {
-        set_perf_event_pending_flag();
+        set_irq_work_pending_flag();
 }
 
-#else  /* CONFIG_PERF_EVENTS */
+#else  /* CONFIG_IRQ_WORK */
 
-#define test_perf_event_pending()      0
-#define clear_perf_event_pending()
+#define test_irq_work_pending()      0
+#define clear_irq_work_pending()
 
-#endif /* CONFIG_PERF_EVENTS */
+#endif /* CONFIG_IRQ_WORK */
 
 
 static inline __u32 rpcc(void)
@@ -191,9 +191,9 @@ irqreturn_t timer_interrupt(int irq, void *dev)
 
         write_sequnlock(&xtime_lock);
 
-        if (test_perf_event_pending()) {
-                clear_perf_event_pending();
-                perf_event_do_pending();
+        if (test_irq_work_pending()) {
+                clear_irq_work_pending();
+                irq_work_run();
         }
 
 #ifndef CONFIG_SMP
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 88c97bc7a6f5..7c0dfccd05bd 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -23,6 +23,7 @@ config ARM
         select HAVE_KERNEL_GZIP
         select HAVE_KERNEL_LZO
         select HAVE_KERNEL_LZMA
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select PERF_USE_VMALLOC
         select HAVE_REGS_AND_STACK_ACCESS_API
diff --git a/arch/arm/include/asm/perf_event.h b/arch/arm/include/asm/perf_event.h
index b5799a3b7117..c4aa4e8c6af9 100644
--- a/arch/arm/include/asm/perf_event.h
+++ b/arch/arm/include/asm/perf_event.h
@@ -12,18 +12,6 @@
 #ifndef __ARM_PERF_EVENT_H__
 #define __ARM_PERF_EVENT_H__
 
-/*
- * NOP: on *most* (read: all supported) ARM platforms, the performance
- * counter interrupts are regular interrupts and not an NMI. This
- * means that when we receive the interrupt we can call
- * perf_event_do_pending() that handles all of the work with
- * interrupts disabled.
- */
-static inline void
-set_perf_event_pending(void)
-{
-}
-
 /* ARM performance counters start from 1 (in the cp15 accesses) so use the
  * same indexes here for consistency. */
 #define PERF_EVENT_INDEX_OFFSET 1
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 6cc6521881aa..49643b1467e6 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -1092,7 +1092,7 @@ armv6pmu_handle_irq(int irq_num,
          * platforms that can have the PMU interrupts raised as an NMI, this
          * will not work.
          */
-        perf_event_do_pending();
+        irq_work_run();
 
         return IRQ_HANDLED;
 }
@@ -2068,7 +2068,7 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev)
          * platforms that can have the PMU interrupts raised as an NMI, this
          * will not work.
          */
-        perf_event_do_pending();
+        irq_work_run();
 
         return IRQ_HANDLED;
 }
@@ -2436,7 +2436,7 @@ xscale1pmu_handle_irq(int irq_num, void *dev)
                         armpmu->disable(hwc, idx);
         }
 
-        perf_event_do_pending();
+        irq_work_run();
 
         /*
          * Re-enable the PMU.
@@ -2763,7 +2763,7 @@ xscale2pmu_handle_irq(int irq_num, void *dev)
                         armpmu->disable(hwc, idx);
         }
 
-        perf_event_do_pending();
+        irq_work_run();
 
         /*
          * Re-enable the PMU.
diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 16399bd24993..0f2417df6323 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -7,6 +7,7 @@ config FRV
         default y
         select HAVE_IDE
         select HAVE_ARCH_TRACEHOOK
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
 
 config ZONE_DMA
diff --git a/arch/frv/lib/Makefile b/arch/frv/lib/Makefile
index f4709756d0d9..4ff2fb1e6b16 100644
--- a/arch/frv/lib/Makefile
+++ b/arch/frv/lib/Makefile
@@ -5,4 +5,4 @@
 lib-y := \
         __ashldi3.o __lshrdi3.o __muldi3.o __ashrdi3.o __negdi2.o __ucmpdi2.o \
         checksum.o memcpy.o memset.o atomic-ops.o atomic64-ops.o \
-        outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o perf_event.o
+        outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o
diff --git a/arch/frv/lib/perf_event.c b/arch/frv/lib/perf_event.c
deleted file mode 100644
index 9ac5acfd2e91..000000000000
--- a/arch/frv/lib/perf_event.c
+++ /dev/null
@@ -1,19 +0,0 @@
-/* Performance event handling
- *
- * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-
-#include <linux/perf_event.h>
-
-/*
- * mark the performance event as pending
- */
-void set_perf_event_pending(void)
-{
-}
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 907417d187e1..79a04a9394d5 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -16,6 +16,7 @@ config PARISC
         select RTC_DRV_GENERIC
         select INIT_ALL_POSSIBLE
         select BUG
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select GENERIC_ATOMIC64 if !64BIT
         help
diff --git a/arch/parisc/include/asm/perf_event.h b/arch/parisc/include/asm/perf_event.h
index cc146427d8f9..1e0fd8ba6c03 100644
--- a/arch/parisc/include/asm/perf_event.h
+++ b/arch/parisc/include/asm/perf_event.h
@@ -1,7 +1,6 @@
 #ifndef __ASM_PARISC_PERF_EVENT_H
 #define __ASM_PARISC_PERF_EVENT_H
 
-/* parisc only supports software events through this interface. */
-static inline void set_perf_event_pending(void) { }
+/* Empty, just to avoid compiling error */
 
 #endif /* __ASM_PARISC_PERF_EVENT_H */
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 631e5a0fb6ab..4b1e521d966f 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -138,6 +138,7 @@ config PPC
         select HAVE_OPROFILE
         select HAVE_SYSCALL_WRAPPERS if PPC64
         select GENERIC_ATOMIC64 if PPC32
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select HAVE_REGS_AND_STACK_ACCESS_API
         select HAVE_HW_BREAKPOINT if PERF_EVENTS && PPC_BOOK3S_64
diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h
index 1ff6662f7faf..9b287fdd8ea3 100644
--- a/arch/powerpc/include/asm/paca.h
+++ b/arch/powerpc/include/asm/paca.h
@@ -129,7 +129,7 @@ struct paca_struct {
         u8 soft_enabled;                /* irq soft-enable flag */
         u8 hard_enabled;                /* set if irqs are enabled in MSR */
         u8 io_sync;                     /* writel() needs spin_unlock sync */
-        u8 perf_event_pending;          /* PM interrupt while soft-disabled */
+        u8 irq_work_pending;            /* IRQ_WORK interrupt while soft-disable */
 
         /* Stuff for accurate time accounting */
         u64 user_time;                  /* accumulated usermode TB ticks */
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 8533b3b83f5d..54888eb10c3b 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -53,7 +53,7 @@
 #include <linux/posix-timers.h>
 #include <linux/irq.h>
 #include <linux/delay.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <asm/trace.h>
 
 #include <asm/io.h>
@@ -493,60 +493,60 @@ void __init iSeries_time_init_early(void)
 }
 #endif /* CONFIG_PPC_ISERIES */
 
-#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_IRQ_WORK
 
 /*
  * 64-bit uses a byte in the PACA, 32-bit uses a per-cpu variable...
  */
 #ifdef CONFIG_PPC64
-static inline unsigned long test_perf_event_pending(void)
+static inline unsigned long test_irq_work_pending(void)
 {
         unsigned long x;
 
         asm volatile("lbz %0,%1(13)"
                 : "=r" (x)
-                : "i" (offsetof(struct paca_struct, perf_event_pending)));
+                : "i" (offsetof(struct paca_struct, irq_work_pending)));
         return x;
 }
 
-static inline void set_perf_event_pending_flag(void)
+static inline void set_irq_work_pending_flag(void)
 {
         asm volatile("stb %0,%1(13)" : :
                 "r" (1),
-                "i" (offsetof(struct paca_struct, perf_event_pending)));
+                "i" (offsetof(struct paca_struct, irq_work_pending)));
 }
 
-static inline void clear_perf_event_pending(void)
+static inline void clear_irq_work_pending(void)
 {
         asm volatile("stb %0,%1(13)" : :
                 "r" (0),
-                "i" (offsetof(struct paca_struct, perf_event_pending)));
+                "i" (offsetof(struct paca_struct, irq_work_pending)));
 }
 
 #else /* 32-bit */
 
-DEFINE_PER_CPU(u8, perf_event_pending);
+DEFINE_PER_CPU(u8, irq_work_pending);
 
-#define set_perf_event_pending_flag()   __get_cpu_var(perf_event_pending) = 1
-#define test_perf_event_pending()       __get_cpu_var(perf_event_pending)
-#define clear_perf_event_pending()      __get_cpu_var(perf_event_pending) = 0
+#define set_irq_work_pending_flag()     __get_cpu_var(irq_work_pending) = 1
+#define test_irq_work_pending()         __get_cpu_var(irq_work_pending)
+#define clear_irq_work_pending()        __get_cpu_var(irq_work_pending) = 0
 
 #endif /* 32 vs 64 bit */
 
-void set_perf_event_pending(void)
+void set_irq_work_pending(void)
 {
         preempt_disable();
-        set_perf_event_pending_flag();
+        set_irq_work_pending_flag();
         set_dec(1);
         preempt_enable();
 }
 
-#else  /* CONFIG_PERF_EVENTS */
+#else  /* CONFIG_IRQ_WORK */
 
-#define test_perf_event_pending()       0
-#define clear_perf_event_pending()
+#define test_irq_work_pending() 0
+#define clear_irq_work_pending()
 
-#endif /* CONFIG_PERF_EVENTS */
+#endif /* CONFIG_IRQ_WORK */
 
 /*
  * For iSeries shared processors, we have to let the hypervisor
@@ -587,9 +587,9 @@ void timer_interrupt(struct pt_regs * regs)
 
         calculate_steal_time();
 
-        if (test_perf_event_pending()) {
-                clear_perf_event_pending();
-                perf_event_do_pending();
+        if (test_irq_work_pending()) {
+                clear_irq_work_pending();
+                irq_work_run();
         }
 
 #ifdef CONFIG_PPC_ISERIES
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index f0777a47e3a5..958f0dadeadf 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -95,6 +95,7 @@ config S390
         select HAVE_KVM if 64BIT
         select HAVE_ARCH_TRACEHOOK
         select INIT_ALL_POSSIBLE
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select HAVE_KERNEL_GZIP
         select HAVE_KERNEL_BZIP2
diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h
index 3840cbe77637..a75f168d2718 100644
--- a/arch/s390/include/asm/perf_event.h
+++ b/arch/s390/include/asm/perf_event.h
@@ -4,7 +4,6 @@
  * Copyright 2009 Martin Schwidefsky, IBM Corporation.
  */
 
-static inline void set_perf_event_pending(void) {}
-static inline void clear_perf_event_pending(void) {}
+/* Empty, just to avoid compiling error */
 
 #define PERF_EVENT_INDEX_OFFSET 0
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 35b6c3f85173..35b6879628a0 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -16,6 +16,7 @@ config SUPERH
         select HAVE_ARCH_TRACEHOOK
         select HAVE_DMA_API_DEBUG
         select HAVE_DMA_ATTRS
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select PERF_USE_VMALLOC
         select HAVE_KERNEL_GZIP
diff --git a/arch/sh/include/asm/perf_event.h b/arch/sh/include/asm/perf_event.h
index 3d0c9f36d150..14308bed7ea5 100644
--- a/arch/sh/include/asm/perf_event.h
+++ b/arch/sh/include/asm/perf_event.h
@@ -26,11 +26,4 @@ extern int register_sh_pmu(struct sh_pmu *);
 extern int reserve_pmc_hardware(void);
 extern void release_pmc_hardware(void);
 
-static inline void set_perf_event_pending(void)
-{
-        /* Nothing to see here, move along. */
-}
-
-#define PERF_EVENT_INDEX_OFFSET 0
-
 #endif /* __ASM_SH_PERF_EVENT_H */
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 9212cd42a832..3e9d31401fb2 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -26,6 +26,7 @@ config SPARC
         select ARCH_WANT_OPTIONAL_GPIOLIB
         select RTC_CLASS
         select RTC_DRV_M48T59
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select PERF_USE_VMALLOC
         select HAVE_DMA_ATTRS
@@ -54,6 +55,7 @@ config SPARC64
         select RTC_DRV_BQ4802
         select RTC_DRV_SUN4V
         select RTC_DRV_STARFIRE
+        select HAVE_IRQ_WORK
         select HAVE_PERF_EVENTS
         select PERF_USE_VMALLOC
 
diff --git a/arch/sparc/include/asm/perf_event.h b/arch/sparc/include/asm/perf_event.h
index 727af70646cb..6e8bfa1786da 100644
--- a/arch/sparc/include/asm/perf_event.h
+++ b/arch/sparc/include/asm/perf_event.h
@@ -1,10 +1,6 @@
 #ifndef __ASM_SPARC_PERF_EVENT_H
 #define __ASM_SPARC_PERF_EVENT_H
 
-extern void set_perf_event_pending(void);
-
-#define PERF_EVENT_INDEX_OFFSET 0
-
 #ifdef CONFIG_PERF_EVENTS
 #include <asm/ptrace.h>
 
diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c
index c4a6a50b4849..b87873c0e8ea 100644
--- a/arch/sparc/kernel/pcr.c
+++ b/arch/sparc/kernel/pcr.c
@@ -7,7 +7,7 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <linux/ftrace.h>
 
 #include <asm/pil.h>
@@ -43,14 +43,14 @@ void __irq_entry deferred_pcr_work_irq(int irq, struct pt_regs *regs)
 
         old_regs = set_irq_regs(regs);
         irq_enter();
-#ifdef CONFIG_PERF_EVENTS
-        perf_event_do_pending();
+#ifdef CONFIG_IRQ_WORK
+        irq_work_run();
 #endif
         irq_exit();
         set_irq_regs(old_regs);
 }
 
-void set_perf_event_pending(void)
+void arch_irq_work_raise(void)
 {
         set_softint(1 << PIL_DEFERRED_PCR_WORK);
 }
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 9815221976a7..fd227d6b8d9c 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -25,6 +25,7 @@ config X86
         select HAVE_IDE
         select HAVE_OPROFILE
         select HAVE_PERF_EVENTS if (!M386 && !M486)
+        select HAVE_IRQ_WORK
         select HAVE_IOREMAP_PROT
         select HAVE_KPROBES
         select ARCH_WANT_OPTIONAL_GPIOLIB
diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h
index 8e8ec663a98f..b8e96a18676b 100644
--- a/arch/x86/include/asm/entry_arch.h
+++ b/arch/x86/include/asm/entry_arch.h
@@ -49,8 +49,8 @@ BUILD_INTERRUPT(apic_timer_interrupt,LOCAL_TIMER_VECTOR)
 BUILD_INTERRUPT(error_interrupt,ERROR_APIC_VECTOR)
 BUILD_INTERRUPT(spurious_interrupt,SPURIOUS_APIC_VECTOR)
 
-#ifdef CONFIG_PERF_EVENTS
-BUILD_INTERRUPT(perf_pending_interrupt, LOCAL_PENDING_VECTOR)
+#ifdef CONFIG_IRQ_WORK
+BUILD_INTERRUPT(irq_work_interrupt, IRQ_WORK_VECTOR)
 #endif
 
 #ifdef CONFIG_X86_THERMAL_VECTOR
diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h
index aeab29aee617..55e4de613f0e 100644
--- a/arch/x86/include/asm/hardirq.h
+++ b/arch/x86/include/asm/hardirq.h
@@ -14,7 +14,7 @@ typedef struct {
 #endif
         unsigned int x86_platform_ipis; /* arch dependent */
         unsigned int apic_perf_irqs;
-        unsigned int apic_pending_irqs;
+        unsigned int apic_irq_work_irqs;
 #ifdef CONFIG_SMP
         unsigned int irq_resched_count;
         unsigned int irq_call_count;
diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
index 46c0fe05f230..3a54a1ca1a02 100644
--- a/arch/x86/include/asm/hw_irq.h
+++ b/arch/x86/include/asm/hw_irq.h
@@ -29,7 +29,7 @@
 extern void apic_timer_interrupt(void);
 extern void x86_platform_ipi(void);
 extern void error_interrupt(void);
-extern void perf_pending_interrupt(void);
+extern void irq_work_interrupt(void);
 
 extern void spurious_interrupt(void);
 extern void thermal_interrupt(void);
diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h
index e2ca30092557..6af0894dafb4 100644
--- a/arch/x86/include/asm/irq_vectors.h
+++ b/arch/x86/include/asm/irq_vectors.h
@@ -114,9 +114,9 @@
 #define X86_PLATFORM_IPI_VECTOR         0xed
 
 /*
- * Performance monitoring pending work vector:
+ * IRQ work vector:
  */
-#define LOCAL_PENDING_VECTOR            0xec
+#define IRQ_WORK_VECTOR                 0xec
 
 #define UV_BAU_MESSAGE                  0xea
 
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 9d3f485e5dd0..7490bf8d1459 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -35,6 +35,7 @@ obj-y			:= process_$(BITS).o signal.o entry_$(BITS).o
 obj-y                   += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o
 obj-y                   += time.o ioport.o ldt.o dumpstack.o
 obj-y                   += setup.o x86_init.o i8259.o irqinit.o jump_label.o
+obj-$(CONFIG_IRQ_WORK)  += irq_work.o
 obj-$(CONFIG_X86_VISWS) += visws_quirks.o
 obj-$(CONFIG_X86_32)    += probe_roms_32.o
 obj-$(CONFIG_X86_32)    += sys_i386_32.o i386_ksyms_32.o
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index e2513f26ba8b..fe73c1844a9a 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1196,25 +1196,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs)
         return handled;
 }
 
-void smp_perf_pending_interrupt(struct pt_regs *regs)
-{
-        irq_enter();
-        ack_APIC_irq();
-        inc_irq_stat(apic_pending_irqs);
-        perf_event_do_pending();
-        irq_exit();
-}
-
-void set_perf_event_pending(void)
-{
-#ifdef CONFIG_X86_LOCAL_APIC
-        if (!x86_pmu.apic || !x86_pmu_initialized())
-                return;
-
-        apic->send_IPI_self(LOCAL_PENDING_VECTOR);
-#endif
-}
-
 void perf_events_lapic_init(void)
 {
         if (!x86_pmu.apic || !x86_pmu_initialized())
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 17be5ec7cbba..c375c79065f8 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -1023,9 +1023,9 @@ apicinterrupt ERROR_APIC_VECTOR \
 apicinterrupt SPURIOUS_APIC_VECTOR \
         spurious_interrupt smp_spurious_interrupt
 
-#ifdef CONFIG_PERF_EVENTS
-apicinterrupt LOCAL_PENDING_VECTOR \
-        perf_pending_interrupt smp_perf_pending_interrupt
+#ifdef CONFIG_IRQ_WORK
+apicinterrupt IRQ_WORK_VECTOR \
+        irq_work_interrupt smp_irq_work_interrupt
 #endif
 
 /*
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 91fd0c70a18a..44edb03fc9ec 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -67,10 +67,10 @@ static int show_other_interrupts(struct seq_file *p, int prec)
         for_each_online_cpu(j)
                 seq_printf(p, "%10u ", irq_stats(j)->apic_perf_irqs);
         seq_printf(p, "  Performance monitoring interrupts\n");
-        seq_printf(p, "%*s: ", prec, "PND");
+        seq_printf(p, "%*s: ", prec, "IWI");
         for_each_online_cpu(j)
-                seq_printf(p, "%10u ", irq_stats(j)->apic_pending_irqs);
-        seq_printf(p, "  Performance pending work\n");
+                seq_printf(p, "%10u ", irq_stats(j)->apic_irq_work_irqs);
+        seq_printf(p, "  IRQ work interrupts\n");
 #endif
         if (x86_platform_ipi_callback) {
                 seq_printf(p, "%*s: ", prec, "PLT");
@@ -185,7 +185,7 @@ u64 arch_irq_stat_cpu(unsigned int cpu)
         sum += irq_stats(cpu)->apic_timer_irqs;
         sum += irq_stats(cpu)->irq_spurious_count;
         sum += irq_stats(cpu)->apic_perf_irqs;
-        sum += irq_stats(cpu)->apic_pending_irqs;
+        sum += irq_stats(cpu)->apic_irq_work_irqs;
 #endif
         if (x86_platform_ipi_callback)
                 sum += irq_stats(cpu)->x86_platform_ipis;
diff --git a/arch/x86/kernel/irq_work.c b/arch/x86/kernel/irq_work.c
new file mode 100644
index 000000000000..ca8f703a1e70
--- /dev/null
+++ b/arch/x86/kernel/irq_work.c
@@ -0,0 +1,30 @@
+/*
+ * x86 specific code for irq_work
+ *
+ * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/irq_work.h>
+#include <linux/hardirq.h>
+#include <asm/apic.h>
+
+void smp_irq_work_interrupt(struct pt_regs *regs)
+{
+        irq_enter();
+        ack_APIC_irq();
+        inc_irq_stat(apic_irq_work_irqs);
+        irq_work_run();
+        irq_exit();
+}
+
+void arch_irq_work_raise(void)
+{
+#ifdef CONFIG_X86_LOCAL_APIC
+        if (!cpu_has_apic)
+                return;
+
+        apic->send_IPI_self(IRQ_WORK_VECTOR);
+        apic_wait_icr_idle();
+#endif
+}
diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c
index 990ae7cfc578..713969b9266b 100644
--- a/arch/x86/kernel/irqinit.c
+++ b/arch/x86/kernel/irqinit.c
@@ -224,9 +224,9 @@ static void __init apic_intr_init(void)
         alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
         alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
 
-        /* Performance monitoring interrupts: */
-# ifdef CONFIG_PERF_EVENTS
-        alloc_intr_gate(LOCAL_PENDING_VECTOR, perf_pending_interrupt);
+        /* IRQ work interrupts: */
+# ifdef CONFIG_IRQ_WORK
+        alloc_intr_gate(IRQ_WORK_VECTOR, irq_work_interrupt);
 # endif
 
 #endif