Merge branch 'linus' into perfcounters/core-v2

Merge reason: we have gathered quite a few conflicts, need to merge upstream

Conflicts:
	arch/powerpc/kernel/Makefile
	arch/x86/ia32/ia32entry.S
	arch/x86/include/asm/hardirq.h
	arch/x86/include/asm/unistd_32.h
	arch/x86/include/asm/unistd_64.h
	arch/x86/kernel/cpu/common.c
	arch/x86/kernel/irq.c
	arch/x86/kernel/syscall_table_32.S
	arch/x86/mm/iomap_32.c
	include/linux/sched.h
	kernel/Makefile

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

13 files changed, 414 insertions, 425 deletions

diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 4732768c534..b0e5e712a7a 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -47,6 +47,7 @@
 #include <asm/idle.h>
 #include <asm/mtrr.h>
 #include <asm/smp.h>
+#include <asm/mce.h>
 
 unsigned int num_processors;
 
@@ -811,7 +812,7 @@ void clear_local_APIC(void)
 	u32 v;
 
 	/* APIC hasn't been mapped yet */
-	if (!apic_phys)
+	if (!x2apic && !apic_phys)
 		return;
 
 	maxlvt = lapic_get_maxlvt();
@@ -845,6 +846,14 @@ void clear_local_APIC(void)
 		apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED);
 	}
 #endif
+#ifdef CONFIG_X86_MCE_INTEL
+	if (maxlvt >= 6) {
+		v = apic_read(APIC_LVTCMCI);
+		if (!(v & APIC_LVT_MASKED))
+			apic_write(APIC_LVTCMCI, v | APIC_LVT_MASKED);
+	}
+#endif
+
 	/*
 	 * Clean APIC state for other OSs:
 	 */
@@ -1245,6 +1254,12 @@ void __cpuinit setup_local_APIC(void)
 	apic_write(APIC_LVT1, value);
 
 	preempt_enable();
+
+#ifdef CONFIG_X86_MCE_INTEL
+	/* Recheck CMCI information after local APIC is up on CPU #0 */
+	if (smp_processor_id() == 0)
+		cmci_recheck();
+#endif
 }
 
 void __cpuinit end_local_APIC_setup(void)
@@ -1323,15 +1338,16 @@ void __init enable_IR_x2apic(void)
 		return;
 	}
 
-	local_irq_save(flags);
-	mask_8259A();
-
-	ret = save_mask_IO_APIC_setup();
+	ret = save_IO_APIC_setup();
 	if (ret) {
 		pr_info("Saving IO-APIC state failed: %d\n", ret);
 		goto end;
 	}
 
+	local_irq_save(flags);
+	mask_IO_APIC_setup();
+	mask_8259A();
+
 	ret = enable_intr_remapping(1);
 
 	if (ret && x2apic_preenabled) {
@@ -1356,10 +1372,10 @@ end_restore:
 	else
 		reinit_intr_remapped_IO_APIC(x2apic_preenabled);
 
-end:
 	unmask_8259A();
 	local_irq_restore(flags);
 
+end:
 	if (!ret) {
 		if (!x2apic_preenabled)
 			pr_info("Enabled x2apic and interrupt-remapping\n");
@@ -1512,12 +1528,10 @@ void __init early_init_lapic_mapping(void)
  */
 void __init init_apic_mappings(void)
 {
-#ifdef CONFIG_X86_X2APIC
 	if (x2apic) {
 		boot_cpu_physical_apicid = read_apic_id();
 		return;
 	}
-#endif
 
 	/*
 	 * If no local APIC can be found then set up a fake all
@@ -1961,12 +1975,9 @@ static int lapic_resume(struct sys_device *dev)
 
 	local_irq_save(flags);
 
-#ifdef CONFIG_X86_X2APIC
 	if (x2apic)
 		enable_x2apic();
-	else
-#endif
-	{
+	else {
 		/*
 		 * Make sure the APICBASE points to the right address
 		 *
diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c
index 3b002995e14..0014714ea97 100644
--- a/arch/x86/kernel/apic/apic_flat_64.c
+++ b/arch/x86/kernel/apic/apic_flat_64.c
@@ -159,20 +159,6 @@ static int flat_apic_id_registered(void)
 	return physid_isset(read_xapic_id(), phys_cpu_present_map);
 }
 
-static unsigned int flat_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	return cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
-}
-
-static unsigned int flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-						const struct cpumask *andmask)
-{
-	unsigned long mask1 = cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
-	unsigned long mask2 = cpumask_bits(andmask)[0] & APIC_ALL_CPUS;
-
-	return mask1 & mask2;
-}
-
 static int flat_phys_pkg_id(int initial_apic_id, int index_msb)
 {
 	return hard_smp_processor_id() >> index_msb;
@@ -213,8 +199,8 @@ struct apic apic_flat =  {
 	.set_apic_id			= set_apic_id,
 	.apic_id_mask			= 0xFFu << 24,
 
-	.cpu_mask_to_apicid		= flat_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= flat_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid		= default_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= flat_send_IPI_mask,
 	.send_IPI_mask_allbutself	= flat_send_IPI_mask_allbutself,
@@ -222,7 +208,6 @@ struct apic apic_flat =  {
 	.send_IPI_all			= flat_send_IPI_all,
 	.send_IPI_self			= apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
@@ -373,7 +358,6 @@ struct apic apic_physflat =  {
 	.send_IPI_all			= physflat_send_IPI_all,
 	.send_IPI_self			= apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c
index 0b1093394fd..676cdac385c 100644
--- a/arch/x86/kernel/apic/bigsmp_32.c
+++ b/arch/x86/kernel/apic/bigsmp_32.c
@@ -16,32 +16,31 @@
 #include <asm/apic.h>
 #include <asm/ipi.h>
 
-static inline unsigned bigsmp_get_apic_id(unsigned long x)
+static unsigned bigsmp_get_apic_id(unsigned long x)
 {
 	return (x >> 24) & 0xFF;
 }
 
-static inline int bigsmp_apic_id_registered(void)
+static int bigsmp_apic_id_registered(void)
 {
 	return 1;
 }
 
-static inline const cpumask_t *bigsmp_target_cpus(void)
+static const struct cpumask *bigsmp_target_cpus(void)
 {
 #ifdef CONFIG_SMP
-	return &cpu_online_map;
+	return cpu_online_mask;
 #else
-	return &cpumask_of_cpu(0);
+	return cpumask_of(0);
 #endif
 }
 
-static inline unsigned long
-bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid)
+static unsigned long bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid)
 {
 	return 0;
 }
 
-static inline unsigned long bigsmp_check_apicid_present(int bit)
+static unsigned long bigsmp_check_apicid_present(int bit)
 {
 	return 1;
 }
@@ -64,7 +63,7 @@ static inline unsigned long calculate_ldr(int cpu)
  * an APIC.  See e.g. "AP-388 82489DX User's Manual" (Intel
  * document number 292116).  So here it goes...
  */
-static inline void bigsmp_init_apic_ldr(void)
+static void bigsmp_init_apic_ldr(void)
 {
 	unsigned long val;
 	int cpu = smp_processor_id();
@@ -74,19 +73,19 @@ static inline void bigsmp_init_apic_ldr(void)
 	apic_write(APIC_LDR, val);
 }
 
-static inline void bigsmp_setup_apic_routing(void)
+static void bigsmp_setup_apic_routing(void)
 {
 	printk(KERN_INFO
 		"Enabling APIC mode:  Physflat.  Using %d I/O APICs\n",
 		nr_ioapics);
 }
 
-static inline int bigsmp_apicid_to_node(int logical_apicid)
+static int bigsmp_apicid_to_node(int logical_apicid)
 {
 	return apicid_2_node[hard_smp_processor_id()];
 }
 
-static inline int bigsmp_cpu_present_to_apicid(int mps_cpu)
+static int bigsmp_cpu_present_to_apicid(int mps_cpu)
 {
 	if (mps_cpu < nr_cpu_ids)
 		return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu);
@@ -94,7 +93,7 @@ static inline int bigsmp_cpu_present_to_apicid(int mps_cpu)
 	return BAD_APICID;
 }
 
-static inline physid_mask_t bigsmp_apicid_to_cpu_present(int phys_apicid)
+static physid_mask_t bigsmp_apicid_to_cpu_present(int phys_apicid)
 {
 	return physid_mask_of_physid(phys_apicid);
 }
@@ -107,29 +106,24 @@ static inline int bigsmp_cpu_to_logical_apicid(int cpu)
 	return cpu_physical_id(cpu);
 }
 
-static inline physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map)
+static physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map)
 {
 	/* For clustered we don't have a good way to do this yet - hack */
 	return physids_promote(0xFFL);
 }
 
-static inline void bigsmp_setup_portio_remap(void)
-{
-}
-
-static inline int bigsmp_check_phys_apicid_present(int boot_cpu_physical_apicid)
+static int bigsmp_check_phys_apicid_present(int boot_cpu_physical_apicid)
 {
 	return 1;
 }
 
 /* As we are using single CPU as destination, pick only one CPU here */
-static inline unsigned int bigsmp_cpu_mask_to_apicid(const cpumask_t *cpumask)
+static unsigned int bigsmp_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
-	return bigsmp_cpu_to_logical_apicid(first_cpu(*cpumask));
+	return bigsmp_cpu_to_logical_apicid(cpumask_first(cpumask));
 }
 
-static inline unsigned int
-bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+static unsigned int bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
 			      const struct cpumask *andmask)
 {
 	int cpu;
@@ -148,7 +142,7 @@ bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
 	return BAD_APICID;
 }
 
-static inline int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
+static int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
 {
 	return cpuid_apic >> index_msb;
 }
@@ -158,12 +152,12 @@ static inline void bigsmp_send_IPI_mask(const struct cpumask *mask, int vector)
 	default_send_IPI_mask_sequence_phys(mask, vector);
 }
 
-static inline void bigsmp_send_IPI_allbutself(int vector)
+static void bigsmp_send_IPI_allbutself(int vector)
 {
 	default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector);
 }
 
-static inline void bigsmp_send_IPI_all(int vector)
+static void bigsmp_send_IPI_all(int vector)
 {
 	bigsmp_send_IPI_mask(cpu_online_mask, vector);
 }
@@ -194,10 +188,10 @@ static const struct dmi_system_id bigsmp_dmi_table[] = {
 	{ } /* NULL entry stops DMI scanning */
 };
 
-static void bigsmp_vector_allocation_domain(int cpu, cpumask_t *retmask)
+static void bigsmp_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpus_clear(*retmask);
-	cpu_set(cpu, *retmask);
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 static int probe_bigsmp(void)
@@ -256,7 +250,6 @@ struct apic apic_bigsmp = {
 	.send_IPI_all			= bigsmp_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c
index 320f2d2e4e5..1c11b819f24 100644
--- a/arch/x86/kernel/apic/es7000_32.c
+++ b/arch/x86/kernel/apic/es7000_32.c
@@ -163,22 +163,17 @@ static int wakeup_secondary_cpu_via_mip(int cpu, unsigned long eip)
 	return 0;
 }
 
-static int __init es7000_update_apic(void)
+static int es7000_apic_is_cluster(void)
 {
-	apic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
-
 	/* MPENTIUMIII */
 	if (boot_cpu_data.x86 == 6 &&
-	    (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11)) {
-		es7000_update_apic_to_cluster();
-		apic->wait_for_init_deassert = NULL;
-		apic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
-	}
+	    (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11))
+		return 1;
 
 	return 0;
 }
 
-static void __init setup_unisys(void)
+static void setup_unisys(void)
 {
 	/*
 	 * Determine the generation of the ES7000 currently running.
@@ -192,14 +187,12 @@ static void __init setup_unisys(void)
 	else
 		es7000_plat = ES7000_CLASSIC;
 	ioapic_renumber_irq = es7000_rename_gsi;
-
-	x86_quirks->update_apic = es7000_update_apic;
 }
 
 /*
  * Parse the OEM Table:
  */
-static int __init parse_unisys_oem(char *oemptr)
+static int parse_unisys_oem(char *oemptr)
 {
 	int			i;
 	int 			success = 0;
@@ -261,7 +254,7 @@ static int __init parse_unisys_oem(char *oemptr)
 }
 
 #ifdef CONFIG_ACPI
-static int __init find_unisys_acpi_oem_table(unsigned long *oem_addr)
+static int find_unisys_acpi_oem_table(unsigned long *oem_addr)
 {
 	struct acpi_table_header *header = NULL;
 	struct es7000_oem_table *table;
@@ -292,7 +285,7 @@ static int __init find_unisys_acpi_oem_table(unsigned long *oem_addr)
 	return 0;
 }
 
-static void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr)
+static void unmap_unisys_acpi_oem_table(unsigned long oem_addr)
 {
 	if (!oem_addr)
 		return;
@@ -310,8 +303,10 @@ static int es7000_check_dsdt(void)
 	return 0;
 }
 
+static int es7000_acpi_ret;
+
 /* Hook from generic ACPI tables.c */
-static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 {
 	unsigned long oem_addr = 0;
 	int check_dsdt;
@@ -332,10 +327,26 @@ static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 		 */
 		unmap_unisys_acpi_oem_table(oem_addr);
 	}
-	return ret;
+
+	es7000_acpi_ret = ret;
+
+	return ret && !es7000_apic_is_cluster();
 }
+
+static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id)
+{
+	int ret = es7000_acpi_ret;
+
+	return ret && es7000_apic_is_cluster();
+}
+
 #else /* !CONFIG_ACPI: */
-static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	return 0;
+}
+
+static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id)
 {
 	return 0;
 }
@@ -349,8 +360,7 @@ static void es7000_spin(int n)
 		rep_nop();
 }
 
-static int __init
-es7000_mip_write(struct mip_reg *mip_reg)
+static int es7000_mip_write(struct mip_reg *mip_reg)
 {
 	int status = 0;
 	int spin;
@@ -383,7 +393,7 @@ es7000_mip_write(struct mip_reg *mip_reg)
 	return status;
 }
 
-static void __init es7000_enable_apic_mode(void)
+static void es7000_enable_apic_mode(void)
 {
 	struct mip_reg es7000_mip_reg;
 	int mip_status;
@@ -400,7 +410,7 @@ static void __init es7000_enable_apic_mode(void)
 		WARN(1, "Command failed, status = %x\n", mip_status);
 }
 
-static void es7000_vector_allocation_domain(int cpu, cpumask_t *retmask)
+static void es7000_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -410,17 +420,15 @@ static void es7000_vector_allocation_domain(int cpu, cpumask_t *retmask)
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	*retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } };
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
 }
 
 
 static void es7000_wait_for_init_deassert(atomic_t *deassert)
 {
-#ifndef CONFIG_ES7000_CLUSTERED_APIC
 	while (!atomic_read(deassert))
 		cpu_relax();
-#endif
-	return;
 }
 
 static unsigned int es7000_get_apic_id(unsigned long x)
@@ -448,14 +456,14 @@ static int es7000_apic_id_registered(void)
 	return 1;
 }
 
-static const cpumask_t *target_cpus_cluster(void)
+static const struct cpumask *target_cpus_cluster(void)
 {
-	return &CPU_MASK_ALL;
+	return cpu_all_mask;
 }
 
-static const cpumask_t *es7000_target_cpus(void)
+static const struct cpumask *es7000_target_cpus(void)
 {
-	return &cpumask_of_cpu(smp_processor_id());
+	return cpumask_of(smp_processor_id());
 }
 
 static unsigned long
@@ -510,7 +518,7 @@ static void es7000_setup_apic_routing(void)
 	  "Enabling APIC mode:  %s. Using %d I/O APICs, target cpus %lx\n",
 		(apic_version[apic] == 0x14) ?
 			"Physical Cluster" : "Logical Cluster",
-		nr_ioapics, cpus_addr(*es7000_target_cpus())[0]);
+		nr_ioapics, cpumask_bits(es7000_target_cpus())[0]);
 }
 
 static int es7000_apicid_to_node(int logical_apicid)
@@ -565,72 +573,24 @@ static int es7000_check_phys_apicid_present(int cpu_physical_apicid)
 	return 1;
 }
 
-static unsigned int
-es7000_cpu_mask_to_apicid_cluster(const struct cpumask *cpumask)
-{
-	int cpus_found = 0;
-	int num_bits_set;
-	int apicid;
-	int cpu;
-
-	num_bits_set = cpumask_weight(cpumask);
-	/* Return id to all */
-	if (num_bits_set == nr_cpu_ids)
-		return 0xFF;
-	/*
-	 * The cpus in the mask must all be on the apic cluster.  If are not
-	 * on the same apicid cluster return default value of target_cpus():
-	 */
-	cpu = cpumask_first(cpumask);
-	apicid = es7000_cpu_to_logical_apicid(cpu);
-
-	while (cpus_found < num_bits_set) {
-		if (cpumask_test_cpu(cpu, cpumask)) {
-			int new_apicid = es7000_cpu_to_logical_apicid(cpu);
-
-			if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-				WARN(1, "Not a valid mask!");
-
-				return 0xFF;
-			}
-			apicid = new_apicid;
-			cpus_found++;
-		}
-		cpu++;
-	}
-	return apicid;
-}
-
-static unsigned int es7000_cpu_mask_to_apicid(const cpumask_t *cpumask)
+static unsigned int es7000_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
-	int cpus_found = 0;
-	int num_bits_set;
-	int apicid;
-	int cpu;
+	unsigned int round = 0;
+	int cpu, uninitialized_var(apicid);
 
-	num_bits_set = cpus_weight(*cpumask);
-	/* Return id to all */
-	if (num_bits_set == nr_cpu_ids)
-		return es7000_cpu_to_logical_apicid(0);
 	/*
-	 * The cpus in the mask must all be on the apic cluster.  If are not
-	 * on the same apicid cluster return default value of target_cpus():
+	 * The cpus in the mask must all be on the apic cluster.
 	 */
-	cpu = first_cpu(*cpumask);
-	apicid = es7000_cpu_to_logical_apicid(cpu);
-	while (cpus_found < num_bits_set) {
-		if (cpu_isset(cpu, *cpumask)) {
-			int new_apicid = es7000_cpu_to_logical_apicid(cpu);
+	for_each_cpu(cpu, cpumask) {
+		int new_apicid = es7000_cpu_to_logical_apicid(cpu);
 
-			if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-				printk("%s: Not a valid mask!\n", __func__);
+		if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
+			WARN(1, "Not a valid mask!");
 
-				return es7000_cpu_to_logical_apicid(0);
-			}
-			apicid = new_apicid;
-			cpus_found++;
+			return BAD_APICID;
 		}
-		cpu++;
+		apicid = new_apicid;
+		round++;
 	}
 	return apicid;
 }
@@ -659,37 +619,103 @@ static int es7000_phys_pkg_id(int cpuid_apic, int index_msb)
 	return cpuid_apic >> index_msb;
 }
 
-void __init es7000_update_apic_to_cluster(void)
-{
-	apic->target_cpus = target_cpus_cluster;
-	apic->irq_delivery_mode = dest_LowestPrio;
-	/* logical delivery broadcast to all procs: */
-	apic->irq_dest_mode = 1;
-
-	apic->init_apic_ldr = es7000_init_apic_ldr_cluster;
-
-	apic->cpu_mask_to_apicid = es7000_cpu_mask_to_apicid_cluster;
-}
-
 static int probe_es7000(void)
 {
 	/* probed later in mptable/ACPI hooks */
 	return 0;
 }
 
-static __init int
-es7000_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid)
+static int es7000_mps_ret;
+static int es7000_mps_oem_check(struct mpc_table *mpc, char *oem,
+		char *productid)
 {
+	int ret = 0;
+
 	if (mpc->oemptr) {
 		struct mpc_oemtable *oem_table =
 			(struct mpc_oemtable *)mpc->oemptr;
 
 		if (!strncmp(oem, "UNISYS", 6))
-			return parse_unisys_oem((char *)oem_table);
+			ret = parse_unisys_oem((char *)oem_table);
 	}
-	return 0;
+
+	es7000_mps_ret = ret;
+
+	return ret && !es7000_apic_is_cluster();
 }
 
+static int es7000_mps_oem_check_cluster(struct mpc_table *mpc, char *oem,
+		char *productid)
+{
+	int ret = es7000_mps_ret;
+
+	return ret && es7000_apic_is_cluster();
+}
+
+struct apic apic_es7000_cluster = {
+
+	.name				= "es7000",
+	.probe				= probe_es7000,
+	.acpi_madt_oem_check		= es7000_acpi_madt_oem_check_cluster,
+	.apic_id_registered		= es7000_apic_id_registered,
+
+	.irq_delivery_mode		= dest_LowestPrio,
+	/* logical delivery broadcast to all procs: */
+	.irq_dest_mode			= 1,
+
+	.target_cpus			= target_cpus_cluster,
+	.disable_esr			= 1,
+	.dest_logical			= 0,
+	.check_apicid_used		= es7000_check_apicid_used,
+	.check_apicid_present		= es7000_check_apicid_present,
+
+	.vector_allocation_domain	= es7000_vector_allocation_domain,
+	.init_apic_ldr			= es7000_init_apic_ldr_cluster,
+
+	.ioapic_phys_id_map		= es7000_ioapic_phys_id_map,
+	.setup_apic_routing		= es7000_setup_apic_routing,
+	.multi_timer_check		= NULL,
+	.apicid_to_node			= es7000_apicid_to_node,
+	.cpu_to_logical_apicid		= es7000_cpu_to_logical_apicid,
+	.cpu_present_to_apicid		= es7000_cpu_present_to_apicid,
+	.apicid_to_cpu_present		= es7000_apicid_to_cpu_present,
+	.setup_portio_remap		= NULL,
+	.check_phys_apicid_present	= es7000_check_phys_apicid_present,
+	.enable_apic_mode		= es7000_enable_apic_mode,
+	.phys_pkg_id			= es7000_phys_pkg_id,
+	.mps_oem_check			= es7000_mps_oem_check_cluster,
+
+	.get_apic_id			= es7000_get_apic_id,
+	.set_apic_id			= NULL,
+	.apic_id_mask			= 0xFF << 24,
+
+	.cpu_mask_to_apicid		= es7000_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and		= es7000_cpu_mask_to_apicid_and,
+
+	.send_IPI_mask			= es7000_send_IPI_mask,
+	.send_IPI_mask_allbutself	= NULL,
+	.send_IPI_allbutself		= es7000_send_IPI_allbutself,
+	.send_IPI_all			= es7000_send_IPI_all,
+	.send_IPI_self			= default_send_IPI_self,
+
+	.wakeup_secondary_cpu		= wakeup_secondary_cpu_via_mip,
+
+	.trampoline_phys_low		= 0x467,
+	.trampoline_phys_high		= 0x469,
+
+	.wait_for_init_deassert		= NULL,
+
+	/* Nothing to do for most platforms, since cleared by the INIT cycle: */
+	.smp_callin_clear_local_apic	= NULL,
+	.inquire_remote_apic		= default_inquire_remote_apic,
+
+	.read				= native_apic_mem_read,
+	.write				= native_apic_mem_write,
+	.icr_read			= native_apic_icr_read,
+	.icr_write			= native_apic_icr_write,
+	.wait_icr_idle			= native_apic_wait_icr_idle,
+	.safe_wait_icr_idle		= native_safe_apic_wait_icr_idle,
+};
 
 struct apic apic_es7000 = {
 
@@ -737,8 +763,6 @@ struct apic apic_es7000 = {
 	.send_IPI_all			= es7000_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
-
 	.trampoline_phys_low		= 0x467,
 	.trampoline_phys_high		= 0x469,
 
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 00e6071cefc..1bb5c6cee3e 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -389,6 +389,8 @@ struct io_apic {
 	unsigned int index;
 	unsigned int unused[3];
 	unsigned int data;
+	unsigned int unused2[11];
+	unsigned int eoi;
 };
 
 static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
@@ -397,6 +399,12 @@ static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
 		+ (mp_ioapics[idx].apicaddr & ~PAGE_MASK);
 }
 
+static inline void io_apic_eoi(unsigned int apic, unsigned int vector)
+{
+	struct io_apic __iomem *io_apic = io_apic_base(apic);
+	writel(vector, &io_apic->eoi);
+}
+
 static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg)
 {
 	struct io_apic __iomem *io_apic = io_apic_base(apic);
@@ -546,16 +554,12 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq
 
 		apic = entry->apic;
 		pin = entry->pin;
-#ifdef CONFIG_INTR_REMAP
 		/*
 		 * With interrupt-remapping, destination information comes
 		 * from interrupt-remapping table entry.
 		 */
 		if (!irq_remapped(irq))
 			io_apic_write(apic, 0x11 + pin*2, dest);
-#else
-		io_apic_write(apic, 0x11 + pin*2, dest);
-#endif
 		reg = io_apic_read(apic, 0x10 + pin*2);
 		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
 		reg |= vector;
@@ -588,10 +592,12 @@ set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask)
 	if (assign_irq_vector(irq, cfg, mask))
 		return BAD_APICID;
 
-	cpumask_and(desc->affinity, cfg->domain, mask);
+	/* check that before desc->addinity get updated */
 	set_extra_move_desc(desc, mask);
 
-	return apic->cpu_mask_to_apicid_and(desc->affinity, cpu_online_mask);
+	cpumask_copy(desc->affinity, mask);
+
+	return apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain);
 }
 
 static void
@@ -849,9 +855,9 @@ __setup("pirq=", ioapic_pirq_setup);
 static struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS];
 
 /*
- * Saves and masks all the unmasked IO-APIC RTE's
+ * Saves all the IO-APIC RTE's
  */
-int save_mask_IO_APIC_setup(void)
+int save_IO_APIC_setup(void)
 {
 	union IO_APIC_reg_01 reg_01;
 	unsigned long flags;
@@ -876,16 +882,9 @@ int save_mask_IO_APIC_setup(void)
 	}
 
 	for (apic = 0; apic < nr_ioapics; apic++)
-		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
-			struct IO_APIC_route_entry entry;
-
-			entry = early_ioapic_entries[apic][pin] =
+		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++)
+			early_ioapic_entries[apic][pin] =
 				ioapic_read_entry(apic, pin);
-			if (!entry.mask) {
-				entry.mask = 1;
-				ioapic_write_entry(apic, pin, entry);
-			}
-		}
 
 	return 0;
 
@@ -898,6 +897,25 @@ nomem:
 	return -ENOMEM;
 }
 
+void mask_IO_APIC_setup(void)
+{
+	int apic, pin;
+
+	for (apic = 0; apic < nr_ioapics; apic++) {
+		if (!early_ioapic_entries[apic])
+			break;
+		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+			struct IO_APIC_route_entry entry;
+
+			entry = early_ioapic_entries[apic][pin];
+			if (!entry.mask) {
+				entry.mask = 1;
+				ioapic_write_entry(apic, pin, entry);
+			}
+		}
+	}
+}
+
 void restore_IO_APIC_setup(void)
 {
 	int apic, pin;
@@ -1411,9 +1429,7 @@ void __setup_vector_irq(int cpu)
 }
 
 static struct irq_chip ioapic_chip;
-#ifdef CONFIG_INTR_REMAP
 static struct irq_chip ir_ioapic_chip;
-#endif
 
 #define IOAPIC_AUTO     -1
 #define IOAPIC_EDGE     0
@@ -1452,7 +1468,6 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t
 	else
 		desc->status &= ~IRQ_LEVEL;
 
-#ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
 		desc->status |= IRQ_MOVE_PCNTXT;
 		if (trigger)
@@ -1464,7 +1479,7 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t
 						      handle_edge_irq, "edge");
 		return;
 	}
-#endif
+
 	if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
 	    trigger == IOAPIC_LEVEL)
 		set_irq_chip_and_handler_name(irq, &ioapic_chip,
@@ -1478,14 +1493,13 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t
 int setup_ioapic_entry(int apic_id, int irq,
 		       struct IO_APIC_route_entry *entry,
 		       unsigned int destination, int trigger,
-		       int polarity, int vector)
+		       int polarity, int vector, int pin)
 {
 	/*
 	 * add it to the IO-APIC irq-routing table:
 	 */
 	memset(entry,0,sizeof(*entry));
 
-#ifdef CONFIG_INTR_REMAP
 	if (intr_remapping_enabled) {
 		struct intel_iommu *iommu = map_ioapic_to_ir(apic_id);
 		struct irte irte;
@@ -1504,7 +1518,14 @@ int setup_ioapic_entry(int apic_id, int irq,
 
 		irte.present = 1;
 		irte.dst_mode = apic->irq_dest_mode;
-		irte.trigger_mode = trigger;
+		/*
+		 * Trigger mode in the IRTE will always be edge, and the
+		 * actual level or edge trigger will be setup in the IO-APIC
+		 * RTE. This will help simplify level triggered irq migration.
+		 * For more details, see the comments above explainig IO-APIC
+		 * irq migration in the presence of interrupt-remapping.
+		 */
+		irte.trigger_mode = 0;
 		irte.dlvry_mode = apic->irq_delivery_mode;
 		irte.vector = vector;
 		irte.dest_id = IRTE_DEST(destination);
@@ -1515,18 +1536,21 @@ int setup_ioapic_entry(int apic_id, int irq,
 		ir_entry->zero = 0;
 		ir_entry->format = 1;
 		ir_entry->index = (index & 0x7fff);
-	} else
-#endif
-	{
+		/*
+		 * IO-APIC RTE will be configured with virtual vector.
+		 * irq handler will do the explicit EOI to the io-apic.
+		 */
+		ir_entry->vector = pin;
+	} else {
 		entry->delivery_mode = apic->irq_delivery_mode;
 		entry->dest_mode = apic->irq_dest_mode;
 		entry->dest = destination;
+		entry->vector = vector;
 	}
 
 	entry->mask = 0;				/* enable IRQ */
 	entry->trigger = trigger;
 	entry->polarity = polarity;
-	entry->vector = vector;
 
 	/* Mask level triggered irqs.
 	 * Use IRQ_DELAYED_DISABLE for edge triggered irqs.
@@ -1561,7 +1585,7 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq
 
 
 	if (setup_ioapic_entry(mp_ioapics[apic_id].apicid, irq, &entry,
-			       dest, trigger, polarity, cfg->vector)) {
+			       dest, trigger, polarity, cfg->vector, pin)) {
 		printk("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
 		       mp_ioapics[apic_id].apicid, pin);
 		__clear_irq_vector(irq, cfg);
@@ -1642,10 +1666,8 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin,
 {
 	struct IO_APIC_route_entry entry;
 
-#ifdef CONFIG_INTR_REMAP
 	if (intr_remapping_enabled)
 		return;
-#endif
 
 	memset(&entry, 0, sizeof(entry));
 
@@ -2040,8 +2062,13 @@ void disable_IO_APIC(void)
 	 * If the i8259 is routed through an IOAPIC
 	 * Put that IOAPIC in virtual wire mode
 	 * so legacy interrupts can be delivered.
+	 *
+	 * With interrupt-remapping, for now we will use virtual wire A mode,
+	 * as virtual wire B is little complex (need to configure both
+	 * IOAPIC RTE aswell as interrupt-remapping table entry).
+	 * As this gets called during crash dump, keep this simple for now.
 	 */
-	if (ioapic_i8259.pin != -1) {
+	if (ioapic_i8259.pin != -1 && !intr_remapping_enabled) {
 		struct IO_APIC_route_entry entry;
 
 		memset(&entry, 0, sizeof(entry));
@@ -2061,7 +2088,10 @@ void disable_IO_APIC(void)
 		ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry);
 	}
 
-	disconnect_bsp_APIC(ioapic_i8259.pin != -1);
+	/*
+	 * Use virtual wire A mode when interrupt remapping is enabled.
+	 */
+	disconnect_bsp_APIC(!intr_remapping_enabled && ioapic_i8259.pin != -1);
 }
 
 #ifdef CONFIG_X86_32
@@ -2303,37 +2333,24 @@ static int ioapic_retrigger_irq(unsigned int irq)
 #ifdef CONFIG_SMP
 
 #ifdef CONFIG_INTR_REMAP
-static void ir_irq_migration(struct work_struct *work);
-
-static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration);
 
 /*
  * Migrate the IO-APIC irq in the presence of intr-remapping.
  *
- * For edge triggered, irq migration is a simple atomic update(of vector
- * and cpu destination) of IRTE and flush the hardware cache.
- *
- * For level triggered, we need to modify the io-apic RTE aswell with the update
- * vector information, along with modifying IRTE with vector and destination.
- * So irq migration for level triggered is little  bit more complex compared to
- * edge triggered migration. But the good news is, we use the same algorithm
- * for level triggered migration as we have today, only difference being,
- * we now initiate the irq migration from process context instead of the
- * interrupt context.
+ * For both level and edge triggered, irq migration is a simple atomic
+ * update(of vector and cpu destination) of IRTE and flush the hardware cache.
  *
- * In future, when we do a directed EOI (combined with cpu EOI broadcast
- * suppression) to the IO-APIC, level triggered irq migration will also be
- * as simple as edge triggered migration and we can do the irq migration
- * with a simple atomic update to IO-APIC RTE.
+ * For level triggered, we eliminate the io-apic RTE modification (with the
+ * updated vector information), by using a virtual vector (io-apic pin number).
+ * Real vector that is used for interrupting cpu will be coming from
+ * the interrupt-remapping table entry.
  */
 static void
 migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	struct irte irte;
-	int modify_ioapic_rte;
 	unsigned int dest;
-	unsigned long flags;
 	unsigned int irq;
 
 	if (!cpumask_intersects(mask, cpu_online_mask))
@@ -2351,13 +2368,6 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 
 	dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask);
 
-	modify_ioapic_rte = desc->status & IRQ_LEVEL;
-	if (modify_ioapic_rte) {
-		spin_lock_irqsave(&ioapic_lock, flags);
-		__target_IO_APIC_irq(irq, dest, cfg);
-		spin_unlock_irqrestore(&ioapic_lock, flags);
-	}
-
 	irte.vector = cfg->vector;
 	irte.dest_id = IRTE_DEST(dest);
 
@@ -2372,73 +2382,12 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 	cpumask_copy(desc->affinity, mask);
 }
 
-static int migrate_irq_remapped_level_desc(struct irq_desc *desc)
-{
-	int ret = -1;
-	struct irq_cfg *cfg = desc->chip_data;
-
-	mask_IO_APIC_irq_desc(desc);
-
-	if (io_apic_level_ack_pending(cfg)) {
-		/*
-		 * Interrupt in progress. Migrating irq now will change the
-		 * vector information in the IO-APIC RTE and that will confuse
-		 * the EOI broadcast performed by cpu.
-		 * So, delay the irq migration to the next instance.
-		 */
-		schedule_delayed_work(&ir_migration_work, 1);
-		goto unmask;
-	}
-
-	/* everthing is clear. we have right of way */
-	migrate_ioapic_irq_desc(desc, desc->pending_mask);
-
-	ret = 0;
-	desc->status &= ~IRQ_MOVE_PENDING;
-	cpumask_clear(desc->pending_mask);
-
-unmask:
-	unmask_IO_APIC_irq_desc(desc);
-
-	return ret;
-}
-
-static void ir_irq_migration(struct work_struct *work)
-{
-	unsigned int irq;
-	struct irq_desc *desc;
-
-	for_each_irq_desc(irq, desc) {
-		if (desc->status & IRQ_MOVE_PENDING) {
-			unsigned long flags;
-
-			spin_lock_irqsave(&desc->lock, flags);
-			if (!desc->chip->set_affinity ||
-			    !(desc->status & IRQ_MOVE_PENDING)) {
-				desc->status &= ~IRQ_MOVE_PENDING;
-				spin_unlock_irqrestore(&desc->lock, flags);
-				continue;
-			}
-
-			desc->chip->set_affinity(irq, desc->pending_mask);
-			spin_unlock_irqrestore(&desc->lock, flags);
-		}
-	}
-}
-
 /*
  * Migrates the IRQ destination in the process context.
  */
 static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
 					    const struct cpumask *mask)
 {
-	if (desc->status & IRQ_LEVEL) {
-		desc->status |= IRQ_MOVE_PENDING;
-		cpumask_copy(desc->pending_mask, mask);
-		migrate_irq_remapped_level_desc(desc);
-		return;
-	}
-
 	migrate_ioapic_irq_desc(desc, mask);
 }
 static void set_ir_ioapic_affinity_irq(unsigned int irq,
@@ -2448,6 +2397,11 @@ static void set_ir_ioapic_affinity_irq(unsigned int irq,
 
 	set_ir_ioapic_affinity_irq_desc(desc, mask);
 }
+#else
+static inline void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
+						   const struct cpumask *mask)
+{
+}
 #endif
 
 asmlinkage void smp_irq_move_cleanup_interrupt(void)
@@ -2461,6 +2415,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
 	me = smp_processor_id();
 	for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) {
 		unsigned int irq;
+		unsigned int irr;
 		struct irq_desc *desc;
 		struct irq_cfg *cfg;
 		irq = __get_cpu_var(vector_irq)[vector];
@@ -2480,6 +2435,18 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
 		if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
 			goto unlock;
 
+		irr = apic_read(APIC_IRR + (vector / 32 * 0x10));
+		/*
+		 * Check if the vector that needs to be cleanedup is
+		 * registered at the cpu's IRR. If so, then this is not
+		 * the best time to clean it up. Lets clean it up in the
+		 * next attempt by sending another IRQ_MOVE_CLEANUP_VECTOR
+		 * to myself.
+		 */
+		if (irr  & (1 << (vector % 32))) {
+			apic->send_IPI_self(IRQ_MOVE_CLEANUP_VECTOR);
+			goto unlock;
+		}
 		__get_cpu_var(vector_irq)[vector] = -1;
 		cfg->move_cleanup_count--;
 unlock:
@@ -2529,9 +2496,44 @@ static inline void irq_complete_move(struct irq_desc **descp) {}
 #endif
 
 #ifdef CONFIG_INTR_REMAP
+static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
+{
+	int apic, pin;
+	struct irq_pin_list *entry;
+
+	entry = cfg->irq_2_pin;
+	for (;;) {
+
+		if (!entry)
+			break;
+
+		apic = entry->apic;
+		pin = entry->pin;
+		io_apic_eoi(apic, pin);
+		entry = entry->next;
+	}
+}
+
+static void
+eoi_ioapic_irq(struct irq_desc *desc)
+{
+	struct irq_cfg *cfg;
+	unsigned long flags;
+	unsigned int irq;
+
+	irq = desc->irq;
+	cfg = desc->chip_data;
+
+	spin_lock_irqsave(&ioapic_lock, flags);
+	__eoi_ioapic_irq(irq, cfg);
+	spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
 static void ack_x2apic_level(unsigned int irq)
 {
+	struct irq_desc *desc = irq_to_desc(irq);
 	ack_x2APIC_irq();
+	eoi_ioapic_irq(desc);
 }
 
 static void ack_x2apic_edge(unsigned int irq)
@@ -2662,20 +2664,20 @@ static struct irq_chip ioapic_chip __read_mostly = {
 	.retrigger	= ioapic_retrigger_irq,
 };
 
-#ifdef CONFIG_INTR_REMAP
 static struct irq_chip ir_ioapic_chip __read_mostly = {
 	.name		= "IR-IO-APIC",
 	.startup	= startup_ioapic_irq,
 	.mask		= mask_IO_APIC_irq,
 	.unmask		= unmask_IO_APIC_irq,
+#ifdef CONFIG_INTR_REMAP
 	.ack		= ack_x2apic_edge,
 	.eoi		= ack_x2apic_level,
 #ifdef CONFIG_SMP
 	.set_affinity	= set_ir_ioapic_affinity_irq,
 #endif
+#endif
 	.retrigger	= ioapic_retrigger_irq,
 };
-#endif
 
 static inline void init_IO_APIC_traps(void)
 {
@@ -2901,10 +2903,8 @@ static inline void __init check_timer(void)
 	 * 8259A.
 	 */
 	if (pin1 == -1) {
-#ifdef CONFIG_INTR_REMAP
 		if (intr_remapping_enabled)
 			panic("BIOS bug: timer not connected to IO-APIC");
-#endif
 		pin1 = pin2;
 		apic1 = apic2;
 		no_pin1 = 1;
@@ -2940,10 +2940,8 @@ static inline void __init check_timer(void)
 				clear_IO_APIC_pin(0, pin1);
 			goto out;
 		}
-#ifdef CONFIG_INTR_REMAP
 		if (intr_remapping_enabled)
 			panic("timer doesn't work through Interrupt-remapped IO-APIC");
-#endif
 		local_irq_disable();
 		clear_IO_APIC_pin(apic1, pin1);
 		if (!no_pin1)
@@ -3237,9 +3235,7 @@ void destroy_irq(unsigned int irq)
 	if (desc)
 		desc->chip_data = cfg;
 
-#ifdef CONFIG_INTR_REMAP
 	free_irte(irq);
-#endif
 	spin_lock_irqsave(&vector_lock, flags);
 	__clear_irq_vector(irq, cfg);
 	spin_unlock_irqrestore(&vector_lock, flags);
@@ -3265,7 +3261,6 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 
 	dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus());
 
-#ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
 		struct irte irte;
 		int ir_index;
@@ -3291,10 +3286,13 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 				  MSI_ADDR_IR_SHV |
 				  MSI_ADDR_IR_INDEX1(ir_index) |
 				  MSI_ADDR_IR_INDEX2(ir_index);
-	} else
-#endif
-	{
-		msg->address_hi = MSI_ADDR_BASE_HI;
+	} else {
+		if (x2apic_enabled())
+			msg->address_hi = MSI_ADDR_BASE_HI |
+					  MSI_ADDR_EXT_DEST_ID(dest);
+		else
+			msg->address_hi = MSI_ADDR_BASE_HI;
+
 		msg->address_lo =
 			MSI_ADDR_BASE_LO |
 			((apic->irq_dest_mode == 0) ?
@@ -3394,15 +3392,16 @@ static struct irq_chip msi_chip = {
 	.retrigger	= ioapic_retrigger_irq,
 };
 
-#ifdef CONFIG_INTR_REMAP
 static struct irq_chip msi_ir_chip = {
 	.name		= "IR-PCI-MSI",
 	.unmask		= unmask_msi_irq,
 	.mask		= mask_msi_irq,
+#ifdef CONFIG_INTR_REMAP
 	.ack		= ack_x2apic_edge,
 #ifdef CONFIG_SMP
 	.set_affinity	= ir_set_msi_irq_affinity,
 #endif
+#endif
 	.retrigger	= ioapic_retrigger_irq,
 };
 
@@ -3432,7 +3431,6 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec)
 	}
 	return index;
 }
-#endif
 
 static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 {
@@ -3446,7 +3444,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 	set_irq_msi(irq, msidesc);
 	write_msi_msg(irq, &msg);
 
-#ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
 		struct irq_desc *desc = irq_to_desc(irq);
 		/*
@@ -3455,7 +3452,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 		desc->status |= IRQ_MOVE_PCNTXT;
 		set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge");
 	} else
-#endif
 		set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge");
 
 	dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq);
@@ -3469,11 +3465,12 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 	int ret, sub_handle;
 	struct msi_desc *msidesc;
 	unsigned int irq_want;
-
-#ifdef CONFIG_INTR_REMAP
-	struct intel_iommu *iommu = 0;
+	struct intel_iommu *iommu = NULL;
 	int index = 0;
-#endif
+
+	/* x86 doesn't support multiple MSI yet */
+	if (type == PCI_CAP_ID_MSI && nvec > 1)
+		return 1;
 
 	irq_want = nr_irqs_gsi;
 	sub_handle = 0;
@@ -3482,7 +3479,6 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 		if (irq == 0)
 			return -1;
 		irq_want = irq + 1;
-#ifdef CONFIG_INTR_REMAP
 		if (!intr_remapping_enabled)
 			goto no_ir;
 
@@ -3510,7 +3506,6 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 			set_irte_irq(irq, iommu, index, sub_handle);
 		}
 no_ir:
-#endif
 		ret = setup_msi_irq(dev, msidesc, irq);
 		if (ret < 0)
 			goto error;
@@ -3528,7 +3523,7 @@ void arch_teardown_msi_irq(unsigned int irq)
 	destroy_irq(irq);
 }
 
-#ifdef CONFIG_DMAR
+#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP)
 #ifdef CONFIG_SMP
 static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
@@ -3609,7 +3604,7 @@ static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 
 #endif /* CONFIG_SMP */
 
-struct irq_chip hpet_msi_type = {
+static struct irq_chip hpet_msi_type = {
 	.name = "HPET_MSI",
 	.unmask = hpet_msi_unmask,
 	.mask = hpet_msi_mask,
@@ -4045,11 +4040,9 @@ void __init setup_ioapic_dest(void)
 			else
 				mask = apic->target_cpus();
 
-#ifdef CONFIG_INTR_REMAP
 			if (intr_remapping_enabled)
 				set_ir_ioapic_affinity_irq_desc(desc, mask);
 			else
-#endif
 				set_ioapic_affinity_irq_desc(desc, mask);
 		}
 
@@ -4142,9 +4135,12 @@ static int __init ioapic_insert_resources(void)
 	struct resource *r = ioapic_resources;
 
 	if (!r) {
-		printk(KERN_ERR
-		       "IO APIC resources could be not be allocated.\n");
-		return -1;
+		if (nr_ioapics > 0) {
+			printk(KERN_ERR
+				"IO APIC resources couldn't be allocated.\n");
+			return -1;
+		}
+		return 0;
 	}
 
 	for (i = 0; i < nr_ioapics; i++) {
diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index bdfad80c3cf..d6bd6240715 100644
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -39,7 +39,7 @@
 int unknown_nmi_panic;
 int nmi_watchdog_enabled;
 
-static cpumask_t backtrace_mask = CPU_MASK_NONE;
+static cpumask_var_t backtrace_mask;
 
 /* nmi_active:
  * >0: the lapic NMI watchdog is active, but can be disabled
@@ -138,6 +138,7 @@ int __init check_nmi_watchdog(void)
 	if (!prev_nmi_count)
 		goto error;
 
+	alloc_cpumask_var(&backtrace_mask, GFP_KERNEL);
 	printk(KERN_INFO "Testing NMI watchdog ... ");
 
 #ifdef CONFIG_SMP
@@ -413,14 +414,14 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 		touched = 1;
 	}
 
-	if (cpu_isset(cpu, backtrace_mask)) {
+	if (cpumask_test_cpu(cpu, backtrace_mask)) {
 		static DEFINE_SPINLOCK(lock);	/* Serialise the printks */
 
 		spin_lock(&lock);
 		printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu);
 		dump_stack();
 		spin_unlock(&lock);
-		cpu_clear(cpu, backtrace_mask);
+		cpumask_clear_cpu(cpu, backtrace_mask);
 	}
 
 	/* Could check oops_in_progress here too, but it's safer not to */
@@ -554,10 +555,10 @@ void __trigger_all_cpu_backtrace(void)
 {
 	int i;
 
-	backtrace_mask = cpu_online_map;
+	cpumask_copy(backtrace_mask, cpu_online_mask);
 	/* Wait for up to 10 seconds for all CPUs to do the backtrace */
 	for (i = 0; i < 10 * 1000; i++) {
-		if (cpus_empty(backtrace_mask))
+		if (cpumask_empty(backtrace_mask))
 			break;
 		mdelay(1);
 	}
diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c
index d9d6d61eed8..533e59c6fc8 100644
--- a/arch/x86/kernel/apic/numaq_32.c
+++ b/arch/x86/kernel/apic/numaq_32.c
@@ -69,7 +69,7 @@ struct mpc_trans {
 /* x86_quirks member */
 static int				mpc_record;
 
-static __cpuinitdata struct mpc_trans	*translation_table[MAX_MPC_ENTRY];
+static struct mpc_trans			*translation_table[MAX_MPC_ENTRY];
 
 int					mp_bus_id_to_node[MAX_MP_BUSSES];
 int					mp_bus_id_to_local[MAX_MP_BUSSES];
@@ -256,13 +256,6 @@ static int __init numaq_setup_ioapic_ids(void)
 	return 1;
 }
 
-static int __init numaq_update_apic(void)
-{
-	apic->wakeup_cpu = wakeup_secondary_cpu_via_nmi;
-
-	return 0;
-}
-
 static struct x86_quirks numaq_x86_quirks __initdata = {
 	.arch_pre_time_init		= numaq_pre_time_init,
 	.arch_time_init			= NULL,
@@ -278,7 +271,6 @@ static struct x86_quirks numaq_x86_quirks __initdata = {
 	.mpc_oem_pci_bus		= mpc_oem_pci_bus,
 	.smp_read_mpc_oem		= smp_read_mpc_oem,
 	.setup_ioapic_ids		= numaq_setup_ioapic_ids,
-	.update_apic			= numaq_update_apic,
 };
 
 static __init void early_check_numaq(void)
@@ -342,9 +334,9 @@ static inline void numaq_smp_callin_clear_local_apic(void)
 	clear_local_APIC();
 }
 
-static inline const cpumask_t *numaq_target_cpus(void)
+static inline const struct cpumask *numaq_target_cpus(void)
 {
-	return &CPU_MASK_ALL;
+	return cpu_all_mask;
 }
 
 static inline unsigned long
@@ -435,7 +427,7 @@ static inline int numaq_check_phys_apicid_present(int boot_cpu_physical_apicid)
  * We use physical apicids here, not logical, so just return the default
  * physical broadcast to stop people from breaking us
  */
-static inline unsigned int numaq_cpu_mask_to_apicid(const cpumask_t *cpumask)
+static unsigned int numaq_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	return 0x0F;
 }
@@ -470,7 +462,7 @@ static int probe_numaq(void)
 	return found_numaq;
 }
 
-static void numaq_vector_allocation_domain(int cpu, cpumask_t *retmask)
+static void numaq_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -480,7 +472,8 @@ static void numaq_vector_allocation_domain(int cpu, cpumask_t *retmask)
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	*retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } };
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
 }
 
 static void numaq_setup_portio_remap(void)
@@ -546,7 +539,7 @@ struct apic apic_numaq = {
 	.send_IPI_all			= numaq_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
+	.wakeup_secondary_cpu		= wakeup_secondary_cpu_via_nmi,
 	.trampoline_phys_low		= NUMAQ_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= NUMAQ_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c
index 3a730fa574b..01eda2ac65e 100644
--- a/arch/x86/kernel/apic/probe_32.c
+++ b/arch/x86/kernel/apic/probe_32.c
@@ -83,7 +83,8 @@ static void default_vector_allocation_domain(int cpu, struct cpumask *retmask)
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	*retmask = (cpumask_t) { { [0] = APIC_ALL_CPUS } };
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
 }
 
 /* should be called last. */
@@ -138,7 +139,6 @@ struct apic apic_default = {
 	.send_IPI_all			= default_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
@@ -159,6 +159,7 @@ extern struct apic apic_numaq;
 extern struct apic apic_summit;
 extern struct apic apic_bigsmp;
 extern struct apic apic_es7000;
+extern struct apic apic_es7000_cluster;
 extern struct apic apic_default;
 
 struct apic *apic = &apic_default;
@@ -176,6 +177,7 @@ static struct apic *apic_probe[] __initdata = {
 #endif
 #ifdef CONFIG_X86_ES7000
 	&apic_es7000,
+	&apic_es7000_cluster,
 #endif
 	&apic_default,	/* must be last */
 	NULL,
@@ -197,9 +199,6 @@ static int __init parse_apic(char *arg)
 		}
 	}
 
-	if (x86_quirks->update_apic)
-		x86_quirks->update_apic();
-
 	/* Parsed again by __setup for debug/verbose */
 	return 0;
 }
@@ -218,8 +217,6 @@ void __init generic_bigsmp_probe(void)
 	if (!cmdline_apic && apic == &apic_default) {
 		if (apic_bigsmp.probe()) {
 			apic = &apic_bigsmp;
-			if (x86_quirks->update_apic)
-				x86_quirks->update_apic();
 			printk(KERN_INFO "Overriding APIC driver with %s\n",
 			       apic->name);
 		}
@@ -240,9 +237,6 @@ void __init generic_apic_probe(void)
 		/* Not visible without early console */
 		if (!apic_probe[i])
 			panic("Didn't find an APIC driver");
-
-		if (x86_quirks->update_apic)
-			x86_quirks->update_apic();
 	}
 	printk(KERN_INFO "Using APIC driver %s\n", apic->name);
 }
@@ -262,8 +256,6 @@ generic_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid)
 
 		if (!cmdline_apic) {
 			apic = apic_probe[i];
-			if (x86_quirks->update_apic)
-				x86_quirks->update_apic();
 			printk(KERN_INFO "Switched to APIC driver `%s'.\n",
 			       apic->name);
 		}
@@ -284,8 +276,6 @@ int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 
 		if (!cmdline_apic) {
 			apic = apic_probe[i];
-			if (x86_quirks->update_apic)
-				x86_quirks->update_apic();
 			printk(KERN_INFO "Switched to APIC driver `%s'.\n",
 			       apic->name);
 		}
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index e7c163661c7..1783652bb0e 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -69,8 +69,12 @@ void __init default_setup_apic_routing(void)
 		printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
 	}
 
-	if (x86_quirks->update_apic)
-		x86_quirks->update_apic();
+	/*
+	 * Now that apic routing model is selected, configure the
+	 * fault handling for intr remapping.
+	 */
+	if (intr_remapping_enabled)
+		enable_drhd_fault_handling();
 }
 
 /* Same for both flat and physical. */
diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c
index 32838b57a94..9cfe1f415d8 100644
--- a/arch/x86/kernel/apic/summit_32.c
+++ b/arch/x86/kernel/apic/summit_32.c
@@ -53,23 +53,19 @@ static unsigned summit_get_apic_id(unsigned long x)
 	return (x >> 24) & 0xFF;
 }
 
-static inline void summit_send_IPI_mask(const cpumask_t *mask, int vector)
+static inline void summit_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	default_send_IPI_mask_sequence_logical(mask, vector);
 }
 
 static void summit_send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(smp_processor_id(), mask);
-
-	if (!cpus_empty(mask))
-		summit_send_IPI_mask(&mask, vector);
+	default_send_IPI_mask_allbutself_logical(cpu_online_mask, vector);
 }
 
 static void summit_send_IPI_all(int vector)
 {
-	summit_send_IPI_mask(&cpu_online_map, vector);
+	summit_send_IPI_mask(cpu_online_mask, vector);
 }
 
 #include <asm/tsc.h>
@@ -77,9 +73,9 @@ static void summit_send_IPI_all(int vector)
 extern int use_cyclone;
 
 #ifdef CONFIG_X86_SUMMIT_NUMA
-extern void setup_summit(void);
+static void setup_summit(void);
 #else
-#define setup_summit()	{}
+static inline void setup_summit(void) {}
 #endif
 
 static int summit_mps_oem_check(struct mpc_table *mpc, char *oem,
@@ -186,13 +182,13 @@ static inline int is_WPEG(struct rio_detail *rio){
 
 #define SUMMIT_APIC_DFR_VALUE	(APIC_DFR_CLUSTER)
 
-static const cpumask_t *summit_target_cpus(void)
+static const struct cpumask *summit_target_cpus(void)
 {
 	/* CPU_MASK_ALL (0xff) has undefined behaviour with
 	 * dest_LowestPrio mode logical clustered apic interrupt routing
 	 * Just start on cpu 0.  IRQ balancing will spread load
 	 */
-	return &cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
 static unsigned long summit_check_apicid_used(physid_mask_t bitmap, int apicid)
@@ -289,35 +285,23 @@ static int summit_check_phys_apicid_present(int boot_cpu_physical_apicid)
 	return 1;
 }
 
-static unsigned int summit_cpu_mask_to_apicid(const cpumask_t *cpumask)
+static unsigned int summit_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
-	int cpus_found = 0;
-	int num_bits_set;
-	int apicid;
-	int cpu;
+	unsigned int round = 0;
+	int cpu, apicid = 0;
 
-	num_bits_set = cpus_weight(*cpumask);
-	if (num_bits_set >= nr_cpu_ids)
-		return BAD_APICID;
 	/*
 	 * The cpus in the mask must all be on the apic cluster.
 	 */
-	cpu = first_cpu(*cpumask);
-	apicid = summit_cpu_to_logical_apicid(cpu);
-
-	while (cpus_found < num_bits_set) {
-		if (cpu_isset(cpu, *cpumask)) {
-			int new_apicid = summit_cpu_to_logical_apicid(cpu);
+	for_each_cpu(cpu, cpumask) {
+		int new_apicid = summit_cpu_to_logical_apicid(cpu);
 
-			if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-				printk("%s: Not a valid mask!\n", __func__);
-
-				return BAD_APICID;
-			}
-			apicid = apicid | new_apicid;
-			cpus_found++;
+		if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
+			printk("%s: Not a valid mask!\n", __func__);
+			return BAD_APICID;
 		}
-		cpu++;
+		apicid |= new_apicid;
+		round++;
 	}
 	return apicid;
 }
@@ -358,7 +342,7 @@ static int probe_summit(void)
 	return 0;
 }
 
-static void summit_vector_allocation_domain(int cpu, cpumask_t *retmask)
+static void summit_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -368,19 +352,20 @@ static void summit_vector_allocation_domain(int cpu, cpumask_t *retmask)
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	*retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } };
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
 }
 
 #ifdef CONFIG_X86_SUMMIT_NUMA
-static struct rio_table_hdr *rio_table_hdr __initdata;
-static struct scal_detail   *scal_devs[MAX_NUMNODES] __initdata;
-static struct rio_detail    *rio_devs[MAX_NUMNODES*4] __initdata;
+static struct rio_table_hdr *rio_table_hdr;
+static struct scal_detail   *scal_devs[MAX_NUMNODES];
+static struct rio_detail    *rio_devs[MAX_NUMNODES*4];
 
 #ifndef CONFIG_X86_NUMAQ
-static int mp_bus_id_to_node[MAX_MP_BUSSES] __initdata;
+static int mp_bus_id_to_node[MAX_MP_BUSSES];
 #endif
 
-static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
+static int setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
 {
 	int twister = 0, node = 0;
 	int i, bus, num_buses;
@@ -442,7 +427,7 @@ static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
 	return bus;
 }
 
-static int __init build_detail_arrays(void)
+static int build_detail_arrays(void)
 {
 	unsigned long ptr;
 	int i, scal_detail_size, rio_detail_size;
@@ -476,7 +461,7 @@ static int __init build_detail_arrays(void)
 	return 1;
 }
 
-void __init setup_summit(void)
+void setup_summit(void)
 {
 	unsigned long		ptr;
 	unsigned short		offset;
@@ -574,7 +559,6 @@ struct apic apic_summit = {
 	.send_IPI_all			= summit_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c
index 354b9c45601..4a903e2f0d1 100644
--- a/arch/x86/kernel/apic/x2apic_cluster.c
+++ b/arch/x86/kernel/apic/x2apic_cluster.c
@@ -57,6 +57,8 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		__x2apic_send_IPI_dest(
@@ -73,6 +75,8 @@ static void
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		if (query_cpu == this_cpu)
@@ -90,6 +94,8 @@ static void x2apic_send_IPI_allbutself(int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_online_cpu(query_cpu) {
 		if (query_cpu == this_cpu)
@@ -224,7 +230,6 @@ struct apic apic_x2apic_cluster = {
 	.send_IPI_all			= x2apic_send_IPI_all,
 	.send_IPI_self			= x2apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c
index 5bcb174409b..a284359627e 100644
--- a/arch/x86/kernel/apic/x2apic_phys.c
+++ b/arch/x86/kernel/apic/x2apic_phys.c
@@ -58,6 +58,8 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu),
@@ -73,6 +75,8 @@ static void
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		if (query_cpu != this_cpu)
@@ -89,6 +93,8 @@ static void x2apic_send_IPI_allbutself(int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_online_cpu(query_cpu) {
 		if (query_cpu == this_cpu)
@@ -213,7 +219,6 @@ struct apic apic_x2apic_phys = {
 	.send_IPI_all			= x2apic_send_IPI_all,
 	.send_IPI_self			= x2apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index 20b4ad07c3a..1248318436e 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -7,28 +7,28 @@
  *
  * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved.
  */
-
-#include <linux/kernel.h>
-#include <linux/threads.h>
-#include <linux/cpu.h>
 #include <linux/cpumask.h>
+#include <linux/hardirq.h>
+#include <linux/proc_fs.h>
+#include <linux/threads.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/string.h>
 #include <linux/ctype.h>
-#include <linux/init.h>
 #include <linux/sched.h>
-#include <linux/module.h>
-#include <linux/hardirq.h>
 #include <linux/timer.h>
-#include <linux/proc_fs.h>
-#include <asm/current.h>
-#include <asm/smp.h>
-#include <asm/apic.h>
-#include <asm/ipi.h>
-#include <asm/pgtable.h>
-#include <asm/uv/uv.h>
+#include <linux/cpu.h>
+#include <linux/init.h>
+
 #include <asm/uv/uv_mmrs.h>
 #include <asm/uv/uv_hub.h>
+#include <asm/current.h>
+#include <asm/pgtable.h>
 #include <asm/uv/bios.h>
+#include <asm/uv/uv.h>
+#include <asm/apic.h>
+#include <asm/ipi.h>
+#include <asm/smp.h>
 
 DEFINE_PER_CPU(int, x2apic_extra_bits);
 
@@ -91,40 +91,39 @@ static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask)
 	cpumask_set_cpu(cpu, retmask);
 }
 
-int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
+static int uv_wakeup_secondary(int phys_apicid, unsigned long start_rip)
 {
+#ifdef CONFIG_SMP
 	unsigned long val;
 	int pnode;
 
 	pnode = uv_apicid_to_pnode(phys_apicid);
 	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
 	    (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
-	    (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+	    ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
 	    APIC_DM_INIT;
 	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
 	mdelay(10);
 
 	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
 	    (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
-	    (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+	    ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
 	    APIC_DM_STARTUP;
 	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
+
+	atomic_set(&init_deasserted, 1);
+#endif
 	return 0;
 }
 
 static void uv_send_IPI_one(int cpu, int vector)
 {
-	unsigned long val, apicid;
+	unsigned long apicid;
 	int pnode;
 
 	apicid = per_cpu(x86_cpu_to_apicid, cpu);
 	pnode = uv_apicid_to_pnode(apicid);
-
-	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
-	      (apicid << UVH_IPI_INT_APIC_ID_SHFT) |
-	      (vector << UVH_IPI_INT_VECTOR_SHFT);
-
-	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
+	uv_hub_send_ipi(pnode, apicid, vector);
 }
 
 static void uv_send_IPI_mask(const struct cpumask *mask, int vector)
@@ -285,7 +284,7 @@ struct apic apic_x2apic_uv_x = {
 	.send_IPI_all			= uv_send_IPI_all,
 	.send_IPI_self			= uv_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
+	.wakeup_secondary_cpu		= uv_wakeup_secondary,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
@@ -365,7 +364,7 @@ static __init void map_high(char *id, unsigned long base, int shift,
 	paddr = base << shift;
 	bytes = (1UL << shift) * (max_pnode + 1);
 	printk(KERN_INFO "UV: Map %s_HI 0x%lx - 0x%lx\n", id, paddr,
-	       					paddr + bytes);
+						paddr + bytes);
 	if (map_type == map_uc)
 		init_extra_mapping_uc(paddr, bytes);
 	else
@@ -528,7 +527,7 @@ late_initcall(uv_init_heartbeat);
 
 /*
  * Called on each cpu to initialize the per_cpu UV data area.
- * 	ZZZ hotplug not supported yet
+ * FIXME: hotplug not supported yet
  */
 void __cpuinit uv_cpu_init(void)
 {