Merge branch 'generic-ipi' into generic-ipi-for-linus

Conflicts:

	arch/powerpc/Kconfig
	arch/s390/kernel/time.c
	arch/x86/kernel/apic_32.c
	arch/x86/kernel/cpu/perfctr-watchdog.c
	arch/x86/kernel/i8259_64.c
	arch/x86/kernel/ldt.c
	arch/x86/kernel/nmi_64.c
	arch/x86/kernel/smpboot.c
	arch/x86/xen/smp.c
	include/asm-x86/hw_irq_32.h
	include/asm-x86/hw_irq_64.h
	include/asm-x86/mach-default/irq_vectors.h
	include/asm-x86/mach-voyager/irq_vectors.h
	include/asm-x86/smp.h
	kernel/Makefile

Signed-off-by: Ingo Molnar <mingo@elte.hu>

30 files changed, 140 insertions, 406 deletions

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 2642b4bf41b..96e0c2ebc38 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -170,6 +170,7 @@ config GENERIC_PENDING_IRQ
 config X86_SMP
 	bool
 	depends on SMP && ((X86_32 && !X86_VOYAGER) || X86_64)
+	select USE_GENERIC_SMP_HELPERS
 	default y
 
 config X86_32_SMP
diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index 3e58b676d23..a437d027f20 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1340,6 +1340,10 @@ void __init smp_intr_init(void)
 
 	/* IPI for generic function call */
 	alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
+
+	/* IPI for single call function */
+	set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR,
+				call_function_single_interrupt);
 }
 #endif
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 98741074518..c4a7ec31394 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -364,7 +364,7 @@ static void mcheck_check_cpu(void *info)
 
 static void mcheck_timer(struct work_struct *work)
 {
-	on_each_cpu(mcheck_check_cpu, NULL, 1, 1);
+	on_each_cpu(mcheck_check_cpu, NULL, 1);
 
 	/*
 	 * Alert userspace if needed.  If we logged an MCE, reduce the
@@ -621,7 +621,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
 	 * Collect entries that were still getting written before the
 	 * synchronize.
 	 */
-	on_each_cpu(collect_tscs, cpu_tsc, 1, 1);
+	on_each_cpu(collect_tscs, cpu_tsc, 1);
 	for (i = next; i < MCE_LOG_LEN; i++) {
 		if (mcelog.entry[i].finished &&
 		    mcelog.entry[i].tsc < cpu_tsc[mcelog.entry[i].cpu]) {
@@ -746,7 +746,7 @@ static void mce_restart(void)
 	if (next_interval)
 		cancel_delayed_work(&mcheck_work);
 	/* Timer race is harmless here */
-	on_each_cpu(mce_init, NULL, 1, 1);
+	on_each_cpu(mce_init, NULL, 1);
 	next_interval = check_interval * HZ;
 	if (next_interval)
 		schedule_delayed_work(&mcheck_work,
diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c
index 00ccb6c14ec..cc1fccdd31e 100644
--- a/arch/x86/kernel/cpu/mcheck/non-fatal.c
+++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c
@@ -59,7 +59,7 @@ static DECLARE_DELAYED_WORK(mce_work, mce_work_fn);
 
 static void mce_work_fn(struct work_struct *work)
 {
-	on_each_cpu(mce_checkregs, NULL, 1, 1);
+	on_each_cpu(mce_checkregs, NULL, 1);
 	schedule_delayed_work(&mce_work, round_jiffies_relative(MCE_RATE));
 }
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 105afe12beb..6f23969c8fa 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -223,7 +223,7 @@ static void set_mtrr(unsigned int reg, unsigned long base,
 	atomic_set(&data.gate,0);
 
 	/*  Start the ball rolling on other CPUs  */
-	if (smp_call_function(ipi_handler, &data, 1, 0) != 0)
+	if (smp_call_function(ipi_handler, &data, 0) != 0)
 		panic("mtrr: timed out waiting for other CPUs\n");
 
 	local_irq_save(flags);
@@ -1682,7 +1682,7 @@ void mtrr_ap_init(void)
  */
 void mtrr_save_state(void)
 {
-	smp_call_function_single(0, mtrr_save_fixed_ranges, NULL, 1, 1);
+	smp_call_function_single(0, mtrr_save_fixed_ranges, NULL, 1);
 }
 
 static int __init mtrr_init_finialize(void)
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index 2e9bef6e3aa..6d4bdc02388 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -189,7 +189,7 @@ void disable_lapic_nmi_watchdog(void)
 	if (atomic_read(&nmi_active) <= 0)
 		return;
 
-	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 1);
 
 	if (wd_ops)
 		wd_ops->unreserve();
@@ -213,7 +213,7 @@ void enable_lapic_nmi_watchdog(void)
 		return;
 	}
 
-	on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+	on_each_cpu(setup_apic_nmi_watchdog, NULL, 1);
 	touch_nmi_watchdog();
 }
 
diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c
index 71f1c2654be..2de5fa2bbf7 100644
--- a/arch/x86/kernel/cpuid.c
+++ b/arch/x86/kernel/cpuid.c
@@ -96,7 +96,7 @@ static ssize_t cpuid_read(struct file *file, char __user *buf,
 	for (; count; count -= 16) {
 		cmd.eax = pos;
 		cmd.ecx = pos >> 32;
-		smp_call_function_single(cpu, cpuid_smp_cpuid, &cmd, 1, 1);
+		smp_call_function_single(cpu, cpuid_smp_cpuid, &cmd, 1);
 		if (copy_to_user(tmp, &cmd, 16))
 			return -EFAULT;
 		tmp += 16;
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index ba41bf42748..ae63e584c34 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -816,6 +816,9 @@ END(invalidate_interrupt\num)
 ENTRY(call_function_interrupt)
 	apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt
 END(call_function_interrupt)
+ENTRY(call_function_single_interrupt)
+	apicinterrupt CALL_FUNCTION_SINGLE_VECTOR,smp_call_function_single_interrupt
+END(call_function_single_interrupt)
 ENTRY(irq_move_cleanup_interrupt)
 	apicinterrupt IRQ_MOVE_CLEANUP_VECTOR,smp_irq_move_cleanup_interrupt
 END(irq_move_cleanup_interrupt)
diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 603261a5885..558abf4c796 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -1569,7 +1569,7 @@ void /*__init*/ print_local_APIC(void *dummy)
 
 void print_all_local_APICs(void)
 {
-	on_each_cpu(print_local_APIC, NULL, 1, 1);
+	on_each_cpu(print_local_APIC, NULL, 1);
 }
 
 void /*__init*/ print_PIC(void)
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index b16ef029cf8..6510cde36b3 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1160,7 +1160,7 @@ void __apicdebuginit print_local_APIC(void * dummy)
 
 void print_all_local_APICs (void)
 {
-	on_each_cpu(print_local_APIC, NULL, 1, 1);
+	on_each_cpu(print_local_APIC, NULL, 1);
 }
 
 void __apicdebuginit print_PIC(void)
diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 31f49e8f46a..0373e88de95 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -199,6 +199,10 @@ void __init native_init_IRQ(void)
 	/* IPI for generic function call */
 	alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
 
+	/* IPI for generic single function call */
+	alloc_intr_gate(CALL_FUNCTION_SINGLE_VECTOR,
+			call_function_single_interrupt);
+
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
 #endif
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index 21f2bae98c1..a8449571858 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -68,7 +68,7 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
 		load_LDT(pc);
 		mask = cpumask_of_cpu(smp_processor_id());
 		if (!cpus_equal(current->mm->cpu_vm_mask, mask))
-			smp_call_function(flush_ldt, current->mm, 1, 1);
+			smp_call_function(flush_ldt, current->mm, 1);
 		preempt_enable();
 #else
 		load_LDT(pc);
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 716b89284be..ec024b3baad 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -130,7 +130,7 @@ int __init check_nmi_watchdog(void)
 
 #ifdef CONFIG_SMP
 	if (nmi_watchdog == NMI_LOCAL_APIC)
-		smp_call_function(nmi_cpu_busy, (void *)&endflag, 0, 0);
+		smp_call_function(nmi_cpu_busy, (void *)&endflag, 0);
 #endif
 
 	for_each_possible_cpu(cpu)
@@ -272,7 +272,7 @@ static void __acpi_nmi_enable(void *__unused)
 void acpi_nmi_enable(void)
 {
 	if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC)
-		on_each_cpu(__acpi_nmi_enable, NULL, 0, 1);
+		on_each_cpu(__acpi_nmi_enable, NULL, 1);
 }
 
 static void __acpi_nmi_disable(void *__unused)
@@ -286,7 +286,7 @@ static void __acpi_nmi_disable(void *__unused)
 void acpi_nmi_disable(void)
 {
 	if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC)
-		on_each_cpu(__acpi_nmi_disable, NULL, 0, 1);
+		on_each_cpu(__acpi_nmi_disable, NULL, 1);
 }
 
 void setup_apic_nmi_watchdog(void *unused)
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 4061d63aabe..7dceea94723 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -132,7 +132,7 @@ void cpu_idle_wait(void)
 {
 	smp_mb();
 	/* kick all the CPUs so that they exit out of pm_idle */
-	smp_call_function(do_nothing, NULL, 0, 1);
+	smp_call_function(do_nothing, NULL, 1);
 }
 EXPORT_SYMBOL_GPL(cpu_idle_wait);
 
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 0cb7aadc87c..361b7a4c640 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -121,132 +121,23 @@ static void native_smp_send_reschedule(int cpu)
 	send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
 }
 
-/*
- * Structure and data for smp_call_function(). This is designed to minimise
- * static memory requirements. It also looks cleaner.
- */
-static DEFINE_SPINLOCK(call_lock);
-
-struct call_data_struct {
-	void (*func) (void *info);
-	void *info;
-	atomic_t started;
-	atomic_t finished;
-	int wait;
-};
-
-void lock_ipi_call_lock(void)
+void native_send_call_func_single_ipi(int cpu)
 {
-	spin_lock_irq(&call_lock);
-}
-
-void unlock_ipi_call_lock(void)
-{
-	spin_unlock_irq(&call_lock);
-}
-
-static struct call_data_struct *call_data;
-
-static void __smp_call_function(void (*func) (void *info), void *info,
-				int nonatomic, int wait)
-{
-	struct call_data_struct data;
-	int cpus = num_online_cpus() - 1;
-
-	if (!cpus)
-		return;
-
-	data.func = func;
-	data.info = info;
-	atomic_set(&data.started, 0);
-	data.wait = wait;
-	if (wait)
-		atomic_set(&data.finished, 0);
-
-	call_data = &data;
-	mb();
-
-	/* Send a message to all other CPUs and wait for them to respond */
-	send_IPI_allbutself(CALL_FUNCTION_VECTOR);
-
-	/* Wait for response */
-	while (atomic_read(&data.started) != cpus)
-		cpu_relax();
-
-	if (wait)
-		while (atomic_read(&data.finished) != cpus)
-			cpu_relax();
+	send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNCTION_SINGLE_VECTOR);
 }
 
-
-/**
- * smp_call_function_mask(): Run a function on a set of other CPUs.
- * @mask: The set of cpus to run on.  Must not include the current cpu.
- * @func: The function to run. This must be fast and non-blocking.
- * @info: An arbitrary pointer to pass to the function.
- * @wait: If true, wait (atomically) until function has completed on other CPUs.
- *
-  * Returns 0 on success, else a negative status code.
- *
- * If @wait is true, then returns once @func has returned; otherwise
- * it returns just before the target cpu calls @func.
- *
- * You must not call this function with disabled interrupts or from a
- * hardware interrupt handler or from a bottom half handler.
- */
-static int
-native_smp_call_function_mask(cpumask_t mask,
-			      void (*func)(void *), void *info,
-			      int wait)
+void native_send_call_func_ipi(cpumask_t mask)
 {
-	struct call_data_struct data;
 	cpumask_t allbutself;
-	int cpus;
-
-	/* Can deadlock when called with interrupts disabled */
-	WARN_ON(irqs_disabled());
-
-	/* Holding any lock stops cpus from going down. */
-	spin_lock(&call_lock);
 
 	allbutself = cpu_online_map;
 	cpu_clear(smp_processor_id(), allbutself);
 
-	cpus_and(mask, mask, allbutself);
-	cpus = cpus_weight(mask);
-
-	if (!cpus) {
-		spin_unlock(&call_lock);
-		return 0;
-	}
-
-	data.func = func;
-	data.info = info;
-	atomic_set(&data.started, 0);
-	data.wait = wait;
-	if (wait)
-		atomic_set(&data.finished, 0);
-
-	call_data = &data;
-	wmb();
-
-	/* Send a message to other CPUs */
 	if (cpus_equal(mask, allbutself) &&
 	    cpus_equal(cpu_online_map, cpu_callout_map))
 		send_IPI_allbutself(CALL_FUNCTION_VECTOR);
 	else
 		send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
-
-	/* Wait for response */
-	while (atomic_read(&data.started) != cpus)
-		cpu_relax();
-
-	if (wait)
-		while (atomic_read(&data.finished) != cpus)
-			cpu_relax();
-	spin_unlock(&call_lock);
-
-	return 0;
 }
 
 static void stop_this_cpu(void *dummy)
@@ -268,18 +159,13 @@ static void stop_this_cpu(void *dummy)
 
 static void native_smp_send_stop(void)
 {
-	int nolock;
 	unsigned long flags;
 
 	if (reboot_force)
 		return;
 
-	/* Don't deadlock on the call lock in panic */
-	nolock = !spin_trylock(&call_lock);
+	smp_call_function(stop_this_cpu, NULL, 0);
 	local_irq_save(flags);
-	__smp_call_function(stop_this_cpu, NULL, 0, 0);
-	if (!nolock)
-		spin_unlock(&call_lock);
 	disable_local_APIC();
 	local_irq_restore(flags);
 }
@@ -301,33 +187,28 @@ void smp_reschedule_interrupt(struct pt_regs *regs)
 
 void smp_call_function_interrupt(struct pt_regs *regs)
 {
-	void (*func) (void *info) = call_data->func;
-	void *info = call_data->info;
-	int wait = call_data->wait;
-
 	ack_APIC_irq();
-	/*
-	 * Notify initiating CPU that I've grabbed the data and am
-	 * about to execute the function
-	 */
-	mb();
-	atomic_inc(&call_data->started);
-	/*
-	 * At this point the info structure may be out of scope unless wait==1
-	 */
 	irq_enter();
-	(*func)(info);
+	generic_smp_call_function_interrupt();
 #ifdef CONFIG_X86_32
 	__get_cpu_var(irq_stat).irq_call_count++;
 #else
 	add_pda(irq_call_count, 1);
 #endif
 	irq_exit();
+}
 
-	if (wait) {
-		mb();
-		atomic_inc(&call_data->finished);
-	}
+void smp_call_function_single_interrupt(struct pt_regs *regs)
+{
+	ack_APIC_irq();
+	irq_enter();
+	generic_smp_call_function_single_interrupt();
+#ifdef CONFIG_X86_32
+	__get_cpu_var(irq_stat).irq_call_count++;
+#else
+	add_pda(irq_call_count, 1);
+#endif
+	irq_exit();
 }
 
 struct smp_ops smp_ops = {
@@ -338,7 +219,8 @@ struct smp_ops smp_ops = {
 
 	.smp_send_stop = native_smp_send_stop,
 	.smp_send_reschedule = native_smp_send_reschedule,
-	.smp_call_function_mask = native_smp_call_function_mask,
+
+	.send_call_func_ipi = native_send_call_func_ipi,
+	.send_call_func_single_ipi = native_send_call_func_single_ipi,
 };
 EXPORT_SYMBOL_GPL(smp_ops);
-
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index f35c2d8016a..687376ab07e 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -327,12 +327,12 @@ static void __cpuinit start_secondary(void *unused)
 	 * lock helps us to not include this cpu in a currently in progress
 	 * smp_call_function().
 	 */
-	lock_ipi_call_lock();
+	ipi_call_lock_irq();
 #ifdef CONFIG_X86_IO_APIC
 	setup_vector_irq(smp_processor_id());
 #endif
 	cpu_set(smp_processor_id(), cpu_online_map);
-	unlock_ipi_call_lock();
+	ipi_call_unlock_irq();
 	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
 
 	setup_secondary_clock();
diff --git a/arch/x86/kernel/smpcommon.c b/arch/x86/kernel/smpcommon.c
index 3449064d141..99941b37eca 100644
--- a/arch/x86/kernel/smpcommon.c
+++ b/arch/x86/kernel/smpcommon.c
@@ -25,59 +25,3 @@ __cpuinit void init_gdt(int cpu)
 	per_cpu(cpu_number, cpu) = cpu;
 }
 #endif
-
-/**
- * smp_call_function(): Run a function on all other CPUs.
- * @func: The function to run. This must be fast and non-blocking.
- * @info: An arbitrary pointer to pass to the function.
- * @nonatomic: Unused.
- * @wait: If true, wait (atomically) until function has completed on other CPUs.
- *
- * Returns 0 on success, else a negative status code.
- *
- * If @wait is true, then returns once @func has returned; otherwise
- * it returns just before the target cpu calls @func.
- *
- * You must not call this function with disabled interrupts or from a
- * hardware interrupt handler or from a bottom half handler.
- */
-int smp_call_function(void (*func) (void *info), void *info, int nonatomic,
-		      int wait)
-{
-	return smp_call_function_mask(cpu_online_map, func, info, wait);
-}
-EXPORT_SYMBOL(smp_call_function);
-
-/**
- * smp_call_function_single - Run a function on a specific CPU
- * @cpu: The target CPU.  Cannot be the calling CPU.
- * @func: The function to run. This must be fast and non-blocking.
- * @info: An arbitrary pointer to pass to the function.
- * @nonatomic: Unused.
- * @wait: If true, wait until function has completed on other CPUs.
- *
- * Returns 0 on success, else a negative status code.
- *
- * If @wait is true, then returns once @func has returned; otherwise
- * it returns just before the target cpu calls @func.
- */
-int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
-			     int nonatomic, int wait)
-{
-	/* prevent preemption and reschedule on another processor */
-	int ret;
-	int me = get_cpu();
-	if (cpu == me) {
-		local_irq_disable();
-		func(info);
-		local_irq_enable();
-		put_cpu();
-		return 0;
-	}
-
-	ret = smp_call_function_mask(cpumask_of_cpu(cpu), func, info, wait);
-
-	put_cpu();
-	return ret;
-}
-EXPORT_SYMBOL(smp_call_function_single);
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index 9bb2363851a..fec1ecedc9b 100644
--- a/arch/x86/kernel/tlb_32.c
+++ b/arch/x86/kernel/tlb_32.c
@@ -238,6 +238,6 @@ static void do_flush_tlb_all(void *info)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
+	on_each_cpu(do_flush_tlb_all, NULL, 1);
 }
 
diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c
index 5039d0f097a..dcbf7a1159e 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -275,5 +275,5 @@ static void do_flush_tlb_all(void *info)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
+	on_each_cpu(do_flush_tlb_all, NULL, 1);
 }
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index e50740d3231..0b8b6690a86 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -279,7 +279,7 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg)
 {
 	long cpu = (long)arg;
 	if (action == CPU_ONLINE || action == CPU_ONLINE_FROZEN)
-		smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1);
+		smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 1);
 	return NOTIFY_DONE;
 }
 
@@ -302,7 +302,7 @@ static int __init vsyscall_init(void)
 #ifdef CONFIG_SYSCTL
 	register_sysctl_table(kernel_root_table2);
 #endif
-	on_each_cpu(cpu_vsyscall_init, NULL, 0, 1);
+	on_each_cpu(cpu_vsyscall_init, NULL, 1);
 	hotcpu_notifier(cpu_vsyscall_notifier, 0);
 	return 0;
 }
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 540e9517907..10ce6ee4c49 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -335,7 +335,7 @@ static void vcpu_clear(struct vcpu_vmx *vmx)
 {
 	if (vmx->vcpu.cpu == -1)
 		return;
-	smp_call_function_single(vmx->vcpu.cpu, __vcpu_clear, vmx, 0, 1);
+	smp_call_function_single(vmx->vcpu.cpu, __vcpu_clear, vmx, 1);
 	vmx->launched = 0;
 }
 
@@ -2968,7 +2968,7 @@ static void vmx_free_vmcs(struct kvm_vcpu *vcpu)
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 
 	if (vmx->vmcs) {
-		on_each_cpu(__vcpu_clear, vmx, 0, 1);
+		on_each_cpu(__vcpu_clear, vmx, 1);
 		free_vmcs(vmx->vmcs);
 		vmx->vmcs = NULL;
 	}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 63a77caa59f..0faa2546b1c 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -4044,6 +4044,6 @@ void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
 	 * So need not to call smp_call_function_single() in that case.
 	 */
 	if (vcpu->guest_mode && vcpu->cpu != cpu)
-		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
+		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0);
 	put_cpu();
 }
diff --git a/arch/x86/lib/msr-on-cpu.c b/arch/x86/lib/msr-on-cpu.c
index 57d043fa893..d5a2b39f882 100644
--- a/arch/x86/lib/msr-on-cpu.c
+++ b/arch/x86/lib/msr-on-cpu.c
@@ -30,10 +30,10 @@ static int _rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h, int safe)
 
 	rv.msr_no = msr_no;
 	if (safe) {
-		smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 1);
 		err = rv.err;
 	} else {
-		smp_call_function_single(cpu, __rdmsr_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __rdmsr_on_cpu, &rv, 1);
 	}
 	*l = rv.l;
 	*h = rv.h;
@@ -64,10 +64,10 @@ static int _wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h, int safe)
 	rv.l = l;
 	rv.h = h;
 	if (safe) {
-		smp_call_function_single(cpu, __wrmsr_safe_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __wrmsr_safe_on_cpu, &rv, 1);
 		err = rv.err;
 	} else {
-		smp_call_function_single(cpu, __wrmsr_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __wrmsr_on_cpu, &rv, 1);
 	}
 
 	return err;
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 8dedd01e909..ee0fba09215 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -950,94 +950,24 @@ static void smp_stop_cpu_function(void *dummy)
 		halt();
 }
 
-static DEFINE_SPINLOCK(call_lock);
-
-struct call_data_struct {
-	void (*func) (void *info);
-	void *info;
-	volatile unsigned long started;
-	volatile unsigned long finished;
-	int wait;
-};
-
-static struct call_data_struct *call_data;
-
 /* execute a thread on a new CPU.  The function to be called must be
  * previously set up.  This is used to schedule a function for
  * execution on all CPUs - set up the function then broadcast a
  * function_interrupt CPI to come here on each CPU */
 static void smp_call_function_interrupt(void)
 {
-	void (*func) (void *info) = call_data->func;
-	void *info = call_data->info;
-	/* must take copy of wait because call_data may be replaced
-	 * unless the function is waiting for us to finish */
-	int wait = call_data->wait;
-	__u8 cpu = smp_processor_id();
-
-	/*
-	 * Notify initiating CPU that I've grabbed the data and am
-	 * about to execute the function
-	 */
-	mb();
-	if (!test_and_clear_bit(cpu, &call_data->started)) {
-		/* If the bit wasn't set, this could be a replay */
-		printk(KERN_WARNING "VOYAGER SMP: CPU %d received call funtion"
-		       " with no call pending\n", cpu);
-		return;
-	}
-	/*
-	 * At this point the info structure may be out of scope unless wait==1
-	 */
 	irq_enter();
-	(*func) (info);
+	generic_smp_call_function_interrupt();
 	__get_cpu_var(irq_stat).irq_call_count++;
 	irq_exit();
-	if (wait) {
-		mb();
-		clear_bit(cpu, &call_data->finished);
-	}
 }
 
-static int
-voyager_smp_call_function_mask(cpumask_t cpumask,
-			       void (*func) (void *info), void *info, int wait)
+static void smp_call_function_single_interrupt(void)
 {
-	struct call_data_struct data;
-	u32 mask = cpus_addr(cpumask)[0];
-
-	mask &= ~(1 << smp_processor_id());
-
-	if (!mask)
-		return 0;
-
-	/* Can deadlock when called with interrupts disabled */
-	WARN_ON(irqs_disabled());
-
-	data.func = func;
-	data.info = info;
-	data.started = mask;
-	data.wait = wait;
-	if (wait)
-		data.finished = mask;
-
-	spin_lock(&call_lock);
-	call_data = &data;
-	wmb();
-	/* Send a message to all other CPUs and wait for them to respond */
-	send_CPI(mask, VIC_CALL_FUNCTION_CPI);
-
-	/* Wait for response */
-	while (data.started)
-		barrier();
-
-	if (wait)
-		while (data.finished)
-			barrier();
-
-	spin_unlock(&call_lock);
-
-	return 0;
+	irq_enter();
+	generic_smp_call_function_single_interrupt();
+	__get_cpu_var(irq_stat).irq_call_count++;
+	irq_exit();
 }
 
 /* Sorry about the name.  In an APIC based system, the APICs
@@ -1094,6 +1024,12 @@ void smp_qic_call_function_interrupt(struct pt_regs *regs)
 	smp_call_function_interrupt();
 }
 
+void smp_qic_call_function_single_interrupt(struct pt_regs *regs)
+{
+	ack_QIC_CPI(QIC_CALL_FUNCTION_SINGLE_CPI);
+	smp_call_function_single_interrupt();
+}
+
 void smp_vic_cpi_interrupt(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs = set_irq_regs(regs);
@@ -1114,6 +1050,8 @@ void smp_vic_cpi_interrupt(struct pt_regs *regs)
 		smp_enable_irq_interrupt();
 	if (test_and_clear_bit(VIC_CALL_FUNCTION_CPI, &vic_cpi_mailbox[cpu]))
 		smp_call_function_interrupt();
+	if (test_and_clear_bit(VIC_CALL_FUNCTION_SINGLE_CPI, &vic_cpi_mailbox[cpu]))
+		smp_call_function_single_interrupt();
 	set_irq_regs(old_regs);
 }
 
@@ -1129,7 +1067,7 @@ static void do_flush_tlb_all(void *info)
 /* flush the TLB of every active CPU in the system */
 void flush_tlb_all(void)
 {
-	on_each_cpu(do_flush_tlb_all, 0, 1, 1);
+	on_each_cpu(do_flush_tlb_all, 0, 1);
 }
 
 /* send a reschedule CPI to one CPU by physical CPU number*/
@@ -1161,7 +1099,7 @@ int safe_smp_processor_id(void)
 /* broadcast a halt to all other CPUs */
 static void voyager_smp_send_stop(void)
 {
-	smp_call_function(smp_stop_cpu_function, NULL, 1, 1);
+	smp_call_function(smp_stop_cpu_function, NULL, 1);
 }
 
 /* this function is triggered in time.c when a clock tick fires
@@ -1848,5 +1786,7 @@ struct smp_ops smp_ops = {
 
 	.smp_send_stop = voyager_smp_send_stop,
 	.smp_send_reschedule = voyager_smp_send_reschedule,
-	.smp_call_function_mask = voyager_smp_call_function_mask,
+
+	.send_call_func_ipi = native_send_call_func_ipi,
+	.send_call_func_single_ipi = native_send_call_func_single_ipi,
 };
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 47f4e2e4a09..65c6e46bf05 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -141,7 +141,7 @@ static void cpa_flush_all(unsigned long cache)
 {
 	BUG_ON(irqs_disabled());
 
-	on_each_cpu(__cpa_flush_all, (void *) cache, 1, 1);
+	on_each_cpu(__cpa_flush_all, (void *) cache, 1);
 }
 
 static void __cpa_flush_range(void *arg)
@@ -162,7 +162,7 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache)
 	BUG_ON(irqs_disabled());
 	WARN_ON(PAGE_ALIGN(start) != start);
 
-	on_each_cpu(__cpa_flush_range, NULL, 1, 1);
+	on_each_cpu(__cpa_flush_range, NULL, 1);
 
 	if (!cache)
 		return;
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 2b6ad5b9f9d..7f3329b55d2 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -218,8 +218,8 @@ static int nmi_setup(void)
 		}
 
 	}
-	on_each_cpu(nmi_save_registers, NULL, 0, 1);
-	on_each_cpu(nmi_cpu_setup, NULL, 0, 1);
+	on_each_cpu(nmi_save_registers, NULL, 1);
+	on_each_cpu(nmi_cpu_setup, NULL, 1);
 	nmi_enabled = 1;
 	return 0;
 }
@@ -271,7 +271,7 @@ static void nmi_shutdown(void)
 {
 	struct op_msrs *msrs = &get_cpu_var(cpu_msrs);
 	nmi_enabled = 0;
-	on_each_cpu(nmi_cpu_shutdown, NULL, 0, 1);
+	on_each_cpu(nmi_cpu_shutdown, NULL, 1);
 	unregister_die_notifier(&profile_exceptions_nb);
 	model->shutdown(msrs);
 	free_msrs();
@@ -286,7 +286,7 @@ static void nmi_cpu_start(void *dummy)
 
 static int nmi_start(void)
 {
-	on_each_cpu(nmi_cpu_start, NULL, 0, 1);
+	on_each_cpu(nmi_cpu_start, NULL, 1);
 	return 0;
 }
 
@@ -298,7 +298,7 @@ static void nmi_cpu_stop(void *dummy)
 
 static void nmi_stop(void)
 {
-	on_each_cpu(nmi_cpu_stop, NULL, 0, 1);
+	on_each_cpu(nmi_cpu_stop, NULL, 1);
 }
 
 struct op_counter_config counter_config[OP_MAX_COUNTER];
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index dcd4e51f2f1..bb508456ef5 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1214,7 +1214,9 @@ static const struct smp_ops xen_smp_ops __initdata = {
 
 	.smp_send_stop = xen_smp_send_stop,
 	.smp_send_reschedule = xen_smp_send_reschedule,
-	.smp_call_function_mask = xen_smp_call_function_mask,
+
+	.send_call_func_ipi = xen_smp_send_call_function_ipi,
+	.send_call_func_single_ipi = xen_smp_send_call_function_single_ipi,
 };
 #endif	/* CONFIG_SMP */
 
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 42b3b9ed641..ff0aa74afaa 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -796,7 +796,7 @@ static void drop_mm_ref(struct mm_struct *mm)
 	}
 
 	if (!cpus_empty(mask))
-		xen_smp_call_function_mask(mask, drop_other_mm_ref, mm, 1);
+		smp_call_function_mask(mask, drop_other_mm_ref, mm, 1);
 }
 #else
 static void drop_mm_ref(struct mm_struct *mm)
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index d2e3c20127d..233156f39b7 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -36,27 +36,14 @@
 #include "mmu.h"
 
 cpumask_t xen_cpu_initialized_map;
-static DEFINE_PER_CPU(int, resched_irq) = -1;
-static DEFINE_PER_CPU(int, callfunc_irq) = -1;
-static DEFINE_PER_CPU(int, debug_irq) = -1;
-
-/*
- * Structure and data for smp_call_function(). This is designed to minimise
- * static memory requirements. It also looks cleaner.
- */
-static DEFINE_SPINLOCK(call_lock);
 
-struct call_data_struct {
-	void (*func) (void *info);
-	void *info;
-	atomic_t started;
-	atomic_t finished;
-	int wait;
-};
+static DEFINE_PER_CPU(int, resched_irq);
+static DEFINE_PER_CPU(int, callfunc_irq);
+static DEFINE_PER_CPU(int, callfuncsingle_irq);
+static DEFINE_PER_CPU(int, debug_irq) = -1;
 
 static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id);
-
-static struct call_data_struct *call_data;
+static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id);
 
 /*
  * Reschedule call back. Nothing to do,
@@ -128,6 +115,17 @@ static int xen_smp_intr_init(unsigned int cpu)
 		goto fail;
 	per_cpu(debug_irq, cpu) = rc;
 
+	callfunc_name = kasprintf(GFP_KERNEL, "callfuncsingle%d", cpu);
+	rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_SINGLE_VECTOR,
+				    cpu,
+				    xen_call_function_single_interrupt,
+				    IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				    callfunc_name,
+				    NULL);
+	if (rc < 0)
+		goto fail;
+	per_cpu(callfuncsingle_irq, cpu) = rc;
+
 	return 0;
 
  fail:
@@ -137,6 +135,9 @@ static int xen_smp_intr_init(unsigned int cpu)
 		unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
 	if (per_cpu(debug_irq, cpu) >= 0)
 		unbind_from_irqhandler(per_cpu(debug_irq, cpu), NULL);
+	if (per_cpu(callfuncsingle_irq, cpu) >= 0)
+		unbind_from_irqhandler(per_cpu(callfuncsingle_irq, cpu), NULL);
+
 	return rc;
 }
 
@@ -336,7 +337,7 @@ static void stop_self(void *v)
 
 void xen_smp_send_stop(void)
 {
-	smp_call_function(stop_self, NULL, 0, 0);
+	smp_call_function(stop_self, NULL, 0);
 }
 
 void xen_smp_send_reschedule(int cpu)
@@ -344,7 +345,6 @@ void xen_smp_send_reschedule(int cpu)
 	xen_send_IPI_one(cpu, XEN_RESCHEDULE_VECTOR);
 }
 
-
 static void xen_send_IPI_mask(cpumask_t mask, enum ipi_vector vector)
 {
 	unsigned cpu;
@@ -355,83 +355,42 @@ static void xen_send_IPI_mask(cpumask_t mask, enum ipi_vector vector)
 		xen_send_IPI_one(cpu, vector);
 }
 
+void xen_smp_send_call_function_ipi(cpumask_t mask)
+{
+	int cpu;
+
+	xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
+
+	/* Make sure other vcpus get a chance to run if they need to. */
+	for_each_cpu_mask(cpu, mask) {
+		if (xen_vcpu_stolen(cpu)) {
+			HYPERVISOR_sched_op(SCHEDOP_yield, 0);
+			break;
+		}
+	}
+}
+
+void xen_smp_send_call_function_single_ipi(int cpu)
+{
+	xen_send_IPI_mask(cpumask_of_cpu(cpu), XEN_CALL_FUNCTION_SINGLE_VECTOR);
+}
+
 static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id)
 {
-	void (*func) (void *info) = call_data->func;
-	void *info = call_data->info;
-	int wait = call_data->wait;
-
-	/*
-	 * Notify initiating CPU that I've grabbed the data and am
-	 * about to execute the function
-	 */
-	mb();
-	atomic_inc(&call_data->started);
-	/*
-	 * At this point the info structure may be out of scope unless wait==1
-	 */
 	irq_enter();
-	(*func)(info);
+	generic_smp_call_function_interrupt();
 	__get_cpu_var(irq_stat).irq_call_count++;
 	irq_exit();
 
-	if (wait) {
-		mb();		/* commit everything before setting finished */
-		atomic_inc(&call_data->finished);
-	}
-
 	return IRQ_HANDLED;
 }
 
-int xen_smp_call_function_mask(cpumask_t mask, void (*func)(void *),
-			       void *info, int wait)
+static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id)
 {
-	struct call_data_struct data;
-	int cpus, cpu;
-	bool yield;
-
-	/* Holding any lock stops cpus from going down. */
-	spin_lock(&call_lock);
-
-	cpu_clear(smp_processor_id(), mask);
-
-	cpus = cpus_weight(mask);
-	if (!cpus) {
-		spin_unlock(&call_lock);
-		return 0;
-	}
-
-	/* Can deadlock when called with interrupts disabled */
-	WARN_ON(irqs_disabled());
-
-	data.func = func;
-	data.info = info;
-	atomic_set(&data.started, 0);
-	data.wait = wait;
-	if (wait)
-		atomic_set(&data.finished, 0);
-
-	call_data = &data;
-	mb();			/* write everything before IPI */
-
-	/* Send a message to other CPUs and wait for them to respond */
-	xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
-
-	/* Make sure other vcpus get a chance to run if they need to. */
-	yield = false;
-	for_each_cpu_mask(cpu, mask)
-		if (xen_vcpu_stolen(cpu))
-			yield = true;
-
-	if (yield)
-		HYPERVISOR_sched_op(SCHEDOP_yield, 0);
-
-	/* Wait for response */
-	while (atomic_read(&data.started) != cpus ||
-	       (wait && atomic_read(&data.finished) != cpus))
-		cpu_relax();
-
-	spin_unlock(&call_lock);
+	irq_enter();
+	generic_smp_call_function_single_interrupt();
+	__get_cpu_var(irq_stat).irq_call_count++;
+	irq_exit();
 
-	return 0;
+	return IRQ_HANDLED;
 }
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index d852ddbb344..6f4b1045c1c 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -55,13 +55,8 @@ void xen_smp_cpus_done(unsigned int max_cpus);
 
 void xen_smp_send_stop(void);
 void xen_smp_send_reschedule(int cpu);
-int xen_smp_call_function (void (*func) (void *info), void *info, int nonatomic,
-			   int wait);
-int xen_smp_call_function_single(int cpu, void (*func) (void *info), void *info,
-				 int nonatomic, int wait);
-
-int xen_smp_call_function_mask(cpumask_t mask, void (*func)(void *),
-			       void *info, int wait);
+void xen_smp_send_call_function_ipi(cpumask_t mask);
+void xen_smp_send_call_function_single_ipi(int cpu);
 
 extern cpumask_t xen_cpu_initialized_map;