From 3b16cf874861436725c43ba0b68bdd799297be7c Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 26 Jun 2008 11:21:54 +0200
Subject: x86: convert to generic helpers for IPI function calls

This converts x86, x86-64, and xen to use the new helpers for
smp_call_function() and friends, and adds support for
smp_call_function_single().

Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 arch/x86/Kconfig                    |   1 +
 arch/x86/kernel/apic_32.c           |   4 +
 arch/x86/kernel/entry_64.S          |   3 +
 arch/x86/kernel/i8259_64.c          |   4 +
 arch/x86/kernel/smp.c               | 158 +++++-------------------------------
 arch/x86/kernel/smpboot.c           |   4 +-
 arch/x86/kernel/smpcommon.c         |  56 -------------
 arch/x86/mach-voyager/voyager_smp.c |  94 ++++-----------------
 arch/x86/xen/enlighten.c            |   4 +-
 arch/x86/xen/mmu.c                  |   2 +-
 arch/x86/xen/smp.c                  | 133 +++++++++++-------------------
 arch/x86/xen/xen-ops.h              |   9 +-
 12 files changed, 103 insertions(+), 369 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e0edaaa..2f3fbeb 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -168,6 +168,7 @@ 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 4b99b1b..71017f7 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1358,6 +1358,10 @@ void __init smp_intr_init(void)
 
 	/* IPI for generic function call */
 	set_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/entry_64.S b/arch/x86/kernel/entry_64.S
index 556a8df5..6d1fe27 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -711,6 +711,9 @@ 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/i8259_64.c b/arch/x86/kernel/i8259_64.c
index fa57a15..00d2ccd 100644
--- a/arch/x86/kernel/i8259_64.c
+++ b/arch/x86/kernel/i8259_64.c
@@ -494,6 +494,10 @@ void __init native_init_IRQ(void)
 	/* IPI for generic function call */
 	set_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
 
+	/* IPI for generic single function call */
+	set_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/smp.c b/arch/x86/kernel/smp.c
index 0cb7aad..575aa3d 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, 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(void)
+{
+	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 56078d6..8964789 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -345,7 +345,7 @@ 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_64
 	spin_lock(&vector_lock);
 
@@ -357,7 +357,7 @@ static void __cpuinit start_secondary(void *unused)
 	spin_unlock(&vector_lock);
 #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 3449064..99941b3 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/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 8acbf0c..cb34407 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -955,94 +955,24 @@ 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
@@ -1099,6 +1029,12 @@ 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);
@@ -1119,6 +1055,8 @@ 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);
 }
 
@@ -1862,5 +1800,7 @@ 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/xen/enlighten.c b/arch/x86/xen/enlighten.c
index f09c1c6..8e31778 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1108,7 +1108,9 @@ 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 df40bf7..5c01590 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -558,7 +558,7 @@ 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 94e6900..b3786e7 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -36,27 +36,14 @@
 #include "mmu.h"
 
 static 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,
@@ -122,6 +109,17 @@ 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:
@@ -131,6 +129,9 @@ 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;
 }
 
@@ -338,7 +339,6 @@ 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;
@@ -349,83 +349,42 @@ 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 f1063ae..a636ab5 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -46,13 +46,8 @@ 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);
 
 
 /* Declare an asm function, along with symbols needed to make it
-- 
cgit v1.1


From 8691e5a8f691cc2a4fda0651e8d307aaba0e7d68 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 6 Jun 2008 11:18:06 +0200
Subject: smp_call_function: get rid of the unused nonatomic/retry argument

It's never used and the comments refer to nonatomic and retry
interchangably. So get rid of it.

Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 arch/x86/kernel/cpu/mtrr/main.c     | 4 ++--
 arch/x86/kernel/cpuid.c             | 2 +-
 arch/x86/kernel/ldt.c               | 2 +-
 arch/x86/kernel/nmi_32.c            | 2 +-
 arch/x86/kernel/nmi_64.c            | 2 +-
 arch/x86/kernel/smp.c               | 2 +-
 arch/x86/kernel/vsyscall_64.c       | 2 +-
 arch/x86/kvm/vmx.c                  | 2 +-
 arch/x86/kvm/x86.c                  | 2 +-
 arch/x86/lib/msr-on-cpu.c           | 8 ++++----
 arch/x86/mach-voyager/voyager_smp.c | 2 +-
 arch/x86/xen/smp.c                  | 2 +-
 12 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 6a1e278..290652c 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -222,7 +222,7 @@ 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);
@@ -822,7 +822,7 @@ 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/cpuid.c b/arch/x86/kernel/cpuid.c
index daff52a..336dd43 100644
--- a/arch/x86/kernel/cpuid.c
+++ b/arch/x86/kernel/cpuid.c
@@ -95,7 +95,7 @@ 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/ldt.c b/arch/x86/kernel/ldt.c
index 0224c36..cb0a6398 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, NULL, 1, 1);
+			smp_call_function(flush_ldt, NULL, 1);
 		preempt_enable();
 #else
 		load_LDT(pc);
diff --git a/arch/x86/kernel/nmi_32.c b/arch/x86/kernel/nmi_32.c
index 84160f7..5562dab 100644
--- a/arch/x86/kernel/nmi_32.c
+++ b/arch/x86/kernel/nmi_32.c
@@ -87,7 +87,7 @@ 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)
diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c
index 5a29ded..2f1e4f5 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -96,7 +96,7 @@ 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 (cpu = 0; cpu < NR_CPUS; cpu++)
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 575aa3d..56546e8 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -164,7 +164,7 @@ static void native_smp_send_stop(void)
 	if (reboot_force)
 		return;
 
-	smp_call_function(stop_this_cpu, NULL, 0, 0);
+	smp_call_function(stop_this_cpu, NULL, 0);
 	local_irq_save(flags);
 	disable_local_APIC();
 	local_irq_restore(flags);
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index 61efa2f..0a03d57 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -278,7 +278,7 @@ 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;
 }
 
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 540e951..5534fe5 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;
 }
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 63a77ca..0faa254 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 57d043f..d5a2b39 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 cb34407..04f596e 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -1113,7 +1113,7 @@ 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
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index b3786e7..a1651d0 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -331,7 +331,7 @@ 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)
-- 
cgit v1.1


From 15c8b6c1aaaf1c4edd67e2f02e4d8e1bd1a51c0d Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 9 May 2008 09:39:44 +0200
Subject: on_each_cpu(): kill unused 'retry' parameter

It's not even passed on to smp_call_function() anymore, since that
was removed. So kill it.

Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 arch/x86/kernel/cpu/mcheck/mce_64.c    |  6 +++---
 arch/x86/kernel/cpu/mcheck/non-fatal.c |  2 +-
 arch/x86/kernel/cpu/perfctr-watchdog.c |  4 ++--
 arch/x86/kernel/io_apic_32.c           |  2 +-
 arch/x86/kernel/io_apic_64.c           |  2 +-
 arch/x86/kernel/nmi_32.c               |  4 ++--
 arch/x86/kernel/nmi_64.c               |  4 ++--
 arch/x86/kernel/tlb_32.c               |  2 +-
 arch/x86/kernel/tlb_64.c               |  2 +-
 arch/x86/kernel/vsyscall_64.c          |  2 +-
 arch/x86/kvm/vmx.c                     |  2 +-
 arch/x86/mach-voyager/voyager_smp.c    |  2 +-
 arch/x86/mm/pageattr.c                 |  4 ++--
 arch/x86/oprofile/nmi_int.c            | 10 +++++-----
 14 files changed, 24 insertions(+), 24 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index e07e8c0..43b7cb5 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -363,7 +363,7 @@ 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
@@ -612,7 +612,7 @@ 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]) {
@@ -737,7 +737,7 @@ 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 00ccb6c..cc1fccd 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/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index f9ae93a..58043f0 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -180,7 +180,7 @@ 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);
 	wd_ops->unreserve();
 
 	BUG_ON(atomic_read(&nmi_active) != 0);
@@ -202,7 +202,7 @@ 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/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 4dc8600d9..720640f 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -1565,7 +1565,7 @@ 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 ef1a8df..4504c7f 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1146,7 +1146,7 @@ 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/nmi_32.c b/arch/x86/kernel/nmi_32.c
index 5562dab..11008e0 100644
--- a/arch/x86/kernel/nmi_32.c
+++ b/arch/x86/kernel/nmi_32.c
@@ -218,7 +218,7 @@ 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)
@@ -232,7 +232,7 @@ 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/nmi_64.c b/arch/x86/kernel/nmi_64.c
index 2f1e4f5..bbdcb17 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -225,7 +225,7 @@ 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)
@@ -239,7 +239,7 @@ 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/tlb_32.c b/arch/x86/kernel/tlb_32.c
index 9bb2363..fec1ece 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 a1f07d7..184a367 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -270,5 +270,5 @@ 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 0a03d57..0dcae19 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -301,7 +301,7 @@ 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 5534fe5..10ce6ee 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -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/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 04f596e..abea084 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -1072,7 +1072,7 @@ 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);
 }
 
 /* used to set up the trampoline for other CPUs when the memory manager
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 60bcb5b..9b836ba 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -106,7 +106,7 @@ 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)
@@ -127,7 +127,7 @@ 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 cc48d3f..3238ad3 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();
@@ -285,7 +285,7 @@ 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;
 }
 
@@ -297,7 +297,7 @@ 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];
-- 
cgit v1.1


From 127a237a1ff49fa5b8e00af91e841598aeea3513 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 27 Jun 2008 11:48:22 +0200
Subject: fix "smp_call_function: get rid of the unused nonatomic/retry
 argument"

fix:

arch/x86/kernel/process.c: In function 'cpu_idle_wait':
arch/x86/kernel/process.c:64: error: too many arguments to function 'smp_call_function'

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/process.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index ba370dc..2dad8fef 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -61,7 +61,7 @@ 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);
 
-- 
cgit v1.1


From 8594698ebddeef5443b7da8258ae33b3eaca61d5 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 27 Jun 2008 21:20:17 +0200
Subject: stacktrace: fix modular build, export print_stack_trace and
 save_stack_trace

fix:

ERROR: "print_stack_trace" [kernel/backtracetest.ko] undefined!
ERROR: "save_stack_trace" [kernel/backtracetest.ko] undefined!

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

and fix:

  Building modules, stage 2.
  MODPOST 376 modules
ERROR: "print_stack_trace" [kernel/backtracetest.ko] undefined!
make[1]: *** [__modpost] Error 1

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/stacktrace.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
index c28c342..a03e7f6 100644
--- a/arch/x86/kernel/stacktrace.c
+++ b/arch/x86/kernel/stacktrace.c
@@ -74,6 +74,7 @@ void save_stack_trace(struct stack_trace *trace)
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
+EXPORT_SYMBOL_GPL(save_stack_trace);
 
 void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 {
@@ -81,3 +82,4 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
+EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
-- 
cgit v1.1


From 5e374fb62621aca9522f76c2317c9acda75a8e88 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 1 Jul 2008 13:12:04 +0200
Subject: generic-ipi: fixlet

create proper stackframe.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 56546e8..361b7a4 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -198,7 +198,7 @@ void smp_call_function_interrupt(struct pt_regs *regs)
 	irq_exit();
 }
 
-void smp_call_function_single_interrupt(void)
+void smp_call_function_single_interrupt(struct pt_regs *regs)
 {
 	ack_APIC_irq();
 	irq_enter();
-- 
cgit v1.1


From 1c776bf87c855a6e823e13c3667f0cf7c14635bd Mon Sep 17 00:00:00 2001
From: Krzysztof Oledzki <ole@ans.pl>
Date: Wed, 4 Jun 2008 03:40:17 +0200
Subject: x86: add another PCI ID for ICH6 force-hpet

Tested on Asus P5GDC-V

$ lspci -n -n |grep ISA
00:1f.0 ISA bridge [0601]: Intel Corporation 82801FB/FR (ICH6/ICH6R) LPC Interface Bridge [8086:2640] (rev 03)

Force enabled HPET at base address 0xfed00000
hpet clockevent registered
hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
hpet0: 3 64-bit timers, 14318180 Hz

Signed-off-by: Krzysztof Piotr Oledzki <ole@ans.pl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/quirks.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c
index d89a648..06e1fd6 100644
--- a/arch/x86/kernel/quirks.c
+++ b/arch/x86/kernel/quirks.c
@@ -158,6 +158,8 @@ static void ich_force_enable_hpet(struct pci_dev *dev)
 
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ESB2_0,
 			 ich_force_enable_hpet);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_0,
+			 ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1,
 			 ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_0,
-- 
cgit v1.1


From 4c2a997c34c0aa952ba9c247b0c2043526054919 Mon Sep 17 00:00:00 2001
From: Joe Buehler <aspam@cox.net>
Date: Mon, 9 Jun 2008 08:55:20 -0400
Subject: x86: add PCI ID for 6300ESB force hpet

00:1f.0 ISA bridge: Intel Corporation 6300ESB LPC Interface Controller (rev 02)
00:1f.0 Class 0601: 8086:25a1 (rev 02)

kernel: pci 0000:00:1f.0: Force enabled HPET at 0xfed00000
kernel: hpet clockevent registered
kernel: hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
kernel: hpet0: 3 64-bit timers, 14318180 Hz

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/quirks.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c
index 06e1fd6..f327aba 100644
--- a/arch/x86/kernel/quirks.c
+++ b/arch/x86/kernel/quirks.c
@@ -257,6 +257,8 @@ static void old_ich_force_enable_hpet_user(struct pci_dev *dev)
 		old_ich_force_enable_hpet(dev);
 }
 
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ESB_1,
+			 old_ich_force_enable_hpet_user);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_0,
 			 old_ich_force_enable_hpet_user);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_12,
-- 
cgit v1.1


From 3bf2e77453a87c22eb57ed4926760ac131c84459 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Sun, 13 Jul 2008 21:18:02 -0700
Subject: x86, suspend, acpi: enter Big Real Mode

The explanation for recent video BIOS suspend quirk failures is that
the VESA BIOS expects to be entered in Big Real Mode (*.limit = 0xffffffff)
instead of ordinary Real Mode (*.limit = 0xffff).

This patch changes the segment descriptors to Big Real Mode instead.

The segment descriptor registers (what Intel calls "segment cache") is
always active.  The only thing that changes based on CR0.PE is how it is
*loaded* and the interpretation of the CS flags.

The segment descriptor registers contain of the following sub-registers:
selector (the "visible" part), base, limit and flags.  In protected mode
or long mode, they are loaded from descriptors (or fs.base or gs.base can
be manipulated directly in long mode.)  In real mode, the only thing
changed by a segment register load is the selector and the base, where the
base <- selector << 4.  In particular, *the limit and the flags are not
changed*.

As far as the handling of the CS flags: a code segment cannot be writable
in protected mode, whereas it is "just another segment" in real mode, so
there is some kind of quirk that kicks in for this when CR0.PE <- 0.  I'm
not sure if this is accomplished by actually changing the cs.flags register
or just changing the interpretation; it might be something that is
CPU-specific.  In particular, the Transmeta CPUs had an explicit "CS is
writable if you're in real mode" override, so even if you had loaded CS
with an execute-only segment it'd be writable (but not readable!) on return
to real mode.  I'm not at all sure if that is how other CPUs behave.

Signed-off-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/sleep.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index 36af01f..130711f1 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -23,6 +23,15 @@ static unsigned long acpi_realmode;
 static char temp_stack[10240];
 #endif
 
+/* XXX: this macro should move to asm-x86/segment.h and be shared with the
+   boot code... */
+#define GDT_ENTRY(flags, base, limit)		\
+	(((u64)(base & 0xff000000) << 32) |	\
+	 ((u64)flags << 40) |			\
+	 ((u64)(limit & 0x00ff0000) << 32) |	\
+	 ((u64)(base & 0x00ffffff) << 16) |	\
+	 ((u64)(limit & 0x0000ffff)))
+
 /**
  * acpi_save_state_mem - save kernel state
  *
@@ -58,11 +67,11 @@ int acpi_save_state_mem(void)
 			((char *)&header->wakeup_gdt - (char *)acpi_realmode))
 				<< 16);
 	/* GDT[1]: real-mode-like code segment */
-	header->wakeup_gdt[1] = (0x009bULL << 40) +
-		((u64)acpi_wakeup_address << 16) + 0xffff;
+	header->wakeup_gdt[1] =
+		GDT_ENTRY(0x809b, acpi_wakeup_address, 0xfffff);
 	/* GDT[2]: real-mode-like data segment */
-	header->wakeup_gdt[2] = (0x0093ULL << 40) +
-		((u64)acpi_wakeup_address << 16) + 0xffff;
+	header->wakeup_gdt[2] =
+		GDT_ENTRY(0x8093, acpi_wakeup_address, 0xfffff);
 
 #ifndef CONFIG_64BIT
 	store_gdt((struct desc_ptr *)&header->pmode_gdt);
-- 
cgit v1.1


From 065cb3dfe24978651caedfa54da585388ad15dde Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Mon, 14 Jul 2008 11:44:26 -0700
Subject: x86, suspend, acpi: correct and add comments about Big Real Mode

Explain that we set up the descriptors for Big Real Mode, and why we
do so.  In particular, one system that is known to fail without it is
the Lenovo X61.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/acpi/sleep.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index 130711f1..41bb130 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -60,16 +60,25 @@ int acpi_save_state_mem(void)
 	header->video_mode = saved_video_mode;
 
 	header->wakeup_jmp_seg = acpi_wakeup_address >> 4;
+
+	/*
+	 * Set up the wakeup GDT.  We set these up as Big Real Mode,
+	 * that is, with limits set to 4 GB.  At least the Lenovo
+	 * Thinkpad X61 is known to need this for the video BIOS
+	 * initialization quirk to work; this is likely to also
+	 * be the case for other laptops or integrated video devices.
+	 */
+
 	/* GDT[0]: GDT self-pointer */
 	header->wakeup_gdt[0] =
 		(u64)(sizeof(header->wakeup_gdt) - 1) +
 		((u64)(acpi_wakeup_address +
 			((char *)&header->wakeup_gdt - (char *)acpi_realmode))
 				<< 16);
-	/* GDT[1]: real-mode-like code segment */
+	/* GDT[1]: big real mode-like code segment */
 	header->wakeup_gdt[1] =
 		GDT_ENTRY(0x809b, acpi_wakeup_address, 0xfffff);
-	/* GDT[2]: real-mode-like data segment */
+	/* GDT[2]: big real mode-like data segment */
 	header->wakeup_gdt[2] =
 		GDT_ENTRY(0x8093, acpi_wakeup_address, 0xfffff);
 
-- 
cgit v1.1


From aba3728ce2e8ce85e1e5f6b275131e9332256789 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 15 Jul 2008 14:48:48 +0200
Subject: x86: sanitize Kconfig

Set default n for MEMTEST and MTRR_SANITIZER and fix the help texts.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig | 22 +++++++++-------------
 1 file changed, 9 insertions(+), 13 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6958d6b..2642b4b 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -447,7 +447,6 @@ config PARAVIRT_DEBUG
 config MEMTEST
 	bool "Memtest"
 	depends on X86_64
-	default y
 	help
 	  This option adds a kernel parameter 'memtest', which allows memtest
 	  to be set.
@@ -455,7 +454,7 @@ config MEMTEST
 		memtest=1, mean do 1 test pattern;
 		...
 		memtest=4, mean do 4 test patterns.
-	  If you are unsure how to answer this question, answer Y.
+	  If you are unsure how to answer this question, answer N.
 
 config X86_SUMMIT_NUMA
 	def_bool y
@@ -1135,21 +1134,18 @@ config MTRR
 	  See <file:Documentation/mtrr.txt> for more information.
 
 config MTRR_SANITIZER
-	def_bool y
+	bool
 	prompt "MTRR cleanup support"
 	depends on MTRR
 	help
-	  Convert MTRR layout from continuous to discrete, so some X driver
-	  could add WB entries.
+	  Convert MTRR layout from continuous to discrete, so X drivers can
+	  add writeback entries.
 
-	  Say N here if you see bootup problems (boot crash, boot hang,
-	  spontaneous reboots).
+	  Can be disabled with disable_mtrr_cleanup on the kernel command line.
+	  The largest mtrr entry size for a continous block can be set with
+	  mtrr_chunk_size.
 
-	  Could be disabled with disable_mtrr_cleanup. Also mtrr_chunk_size
-	  could be used to send largest mtrr entry size for continuous block
-	  to hold holes (aka. UC entries)
-
-	  If unsure, say Y.
+	  If unsure, say N.
 
 config MTRR_SANITIZER_ENABLE_DEFAULT
 	int "MTRR cleanup enable value (0-1)"
@@ -1166,7 +1162,7 @@ config MTRR_SANITIZER_SPARE_REG_NR_DEFAULT
 	depends on MTRR_SANITIZER
 	help
 	  mtrr cleanup spare entries default, it can be changed via
-	  mtrr_spare_reg_nr=
+	  mtrr_spare_reg_nr=N on the kernel command line.
 
 config X86_PAT
 	bool
-- 
cgit v1.1


From b3c9816b9fa9a7b75ab36111eb76eca03e5bab78 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Tue, 15 Jul 2008 22:03:56 +0200
Subject: generic-ipi: merge fix
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

fix merge fallout:

arch/x86/pci/amd_bus.c: In function ‘enable_pci_io_ecs':
arch/x86/pci/amd_bus.c:581: error: too many arguments to function ‘on_each_cpu'

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/pci/amd_bus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/pci/amd_bus.c b/arch/x86/pci/amd_bus.c
index a18141a..dbf5323 100644
--- a/arch/x86/pci/amd_bus.c
+++ b/arch/x86/pci/amd_bus.c
@@ -578,7 +578,7 @@ static int __init enable_pci_io_ecs(void)
 	/* assume all cpus from fam10h have IO ECS */
         if (boot_cpu_data.x86 < 0x10)
 		return 0;
-	on_each_cpu(enable_pci_io_ecs_per_cpu, NULL, 1, 1);
+	on_each_cpu(enable_pci_io_ecs_per_cpu, NULL, 1);
 	pci_probe |= PCI_HAS_IO_ECS;
 	return 0;
 }
-- 
cgit v1.1


From 431ceb83f703a343bdd14350480a2224fa4bfedf Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 15 Jul 2008 22:08:04 +0200
Subject: x86: fix TSC build error on 32bit

Dave Hansen reported a build error on 32bit which went unnoticed
as newer gcc versions seem to optimize unused static functions
away before compiling them.

Make vread_tsc() depend on CONFIG_X86_64

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/tsc.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 3c36f92..7603c05 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -358,6 +358,7 @@ static cycle_t read_tsc(void)
 		ret : clocksource_tsc.cycle_last;
 }
 
+#ifdef CONFIG_X86_64
 static cycle_t __vsyscall_fn vread_tsc(void)
 {
 	cycle_t ret = (cycle_t)vget_cycles();
@@ -365,6 +366,7 @@ static cycle_t __vsyscall_fn vread_tsc(void)
 	return ret >= __vsyscall_gtod_data.clock.cycle_last ?
 		ret : __vsyscall_gtod_data.clock.cycle_last;
 }
+#endif
 
 static struct clocksource clocksource_tsc = {
 	.name                   = "tsc",
-- 
cgit v1.1