Merge branch 'linus' into x86/cleanups

Conflicts:
	arch/x86/kernel/reboot.c

36 files changed, 1235 insertions, 673 deletions

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 88dd768eab6..d364df03c1d 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -109,6 +109,8 @@ obj-$(CONFIG_MICROCODE)			+= microcode.o
 
 obj-$(CONFIG_X86_CHECK_BIOS_CORRUPTION) += check.o
 
+obj-$(CONFIG_SWIOTLB)			+= pci-swiotlb_64.o # NB rename without _64
+
 ###
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
@@ -122,7 +124,6 @@ ifeq ($(CONFIG_X86_64),y)
         obj-$(CONFIG_GART_IOMMU)	+= pci-gart_64.o aperture_64.o
         obj-$(CONFIG_CALGARY_IOMMU)	+= pci-calgary_64.o tce_64.o
         obj-$(CONFIG_AMD_IOMMU)		+= amd_iommu_init.o amd_iommu.o
-        obj-$(CONFIG_SWIOTLB)		+= pci-swiotlb_64.o
 
         obj-$(CONFIG_PCI_MMCONFIG)	+= mmconf-fam10h_64.o
 endif
diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c
index 66198cbe464..d652515e285 100644
--- a/arch/x86/kernel/apic.c
+++ b/arch/x86/kernel/apic.c
@@ -119,8 +119,6 @@ EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok);
 
 int first_system_vector = 0xfe;
 
-char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE};
-
 /*
  * Debug level, exported for io_apic.c
  */
@@ -142,7 +140,7 @@ static int lapic_next_event(unsigned long delta,
 			    struct clock_event_device *evt);
 static void lapic_timer_setup(enum clock_event_mode mode,
 			      struct clock_event_device *evt);
-static void lapic_timer_broadcast(cpumask_t mask);
+static void lapic_timer_broadcast(const cpumask_t *mask);
 static void apic_pm_activate(void);
 
 /*
@@ -455,7 +453,7 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 /*
  * Local APIC timer broadcast function
  */
-static void lapic_timer_broadcast(cpumask_t mask)
+static void lapic_timer_broadcast(const cpumask_t *mask)
 {
 #ifdef CONFIG_SMP
 	send_IPI_mask(mask, LOCAL_TIMER_VECTOR);
@@ -471,7 +469,7 @@ static void __cpuinit setup_APIC_timer(void)
 	struct clock_event_device *levt = &__get_cpu_var(lapic_events);
 
 	memcpy(levt, &lapic_clockevent, sizeof(*levt));
-	levt->cpumask = cpumask_of_cpu(smp_processor_id());
+	levt->cpumask = cpumask_of(smp_processor_id());
 
 	clockevents_register_device(levt);
 }
@@ -1807,28 +1805,32 @@ void disconnect_bsp_APIC(int virt_wire_setup)
 void __cpuinit generic_processor_info(int apicid, int version)
 {
 	int cpu;
-	cpumask_t tmp_map;
 
 	/*
 	 * Validate version
 	 */
 	if (version == 0x0) {
 		pr_warning("BIOS bug, APIC version is 0 for CPU#%d! "
-			"fixing up to 0x10. (tell your hw vendor)\n",
-			version);
+			   "fixing up to 0x10. (tell your hw vendor)\n",
+				version);
 		version = 0x10;
 	}
 	apic_version[apicid] = version;
 
-	if (num_processors >= NR_CPUS) {
-		pr_warning("WARNING: NR_CPUS limit of %i reached."
-			"  Processor ignored.\n", NR_CPUS);
+	if (num_processors >= nr_cpu_ids) {
+		int max = nr_cpu_ids;
+		int thiscpu = max + disabled_cpus;
+
+		pr_warning(
+			"ACPI: NR_CPUS/possible_cpus limit of %i reached."
+			"  Processor %d/0x%x ignored.\n", max, thiscpu, apicid);
+
+		disabled_cpus++;
 		return;
 	}
 
 	num_processors++;
-	cpus_complement(tmp_map, cpu_present_map);
-	cpu = first_cpu(tmp_map);
+	cpu = cpumask_next_zero(-1, cpu_present_mask);
 
 	physid_set(apicid, phys_cpu_present_map);
 	if (apicid == boot_cpu_physical_apicid) {
@@ -1878,8 +1880,8 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	}
 #endif
 
-	cpu_set(cpu, cpu_possible_map);
-	cpu_set(cpu, cpu_present_map);
+	set_cpu_possible(cpu, true);
+	set_cpu_present(cpu, true);
 }
 
 #ifdef CONFIG_X86_64
@@ -2081,7 +2083,7 @@ __cpuinit int apic_is_clustered_box(void)
 	bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid);
 	bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		/* are we being called early in kernel startup? */
 		if (bios_cpu_apicid) {
 			id = bios_cpu_apicid[i];
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 68b5d8681cb..c6ecda64f5f 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -534,31 +534,16 @@ static void __cpuinit free_cache_attributes(unsigned int cpu)
 	per_cpu(cpuid4_info, cpu) = NULL;
 }
 
-static int __cpuinit detect_cache_attributes(unsigned int cpu)
+static void get_cpu_leaves(void *_retval)
 {
-	struct _cpuid4_info	*this_leaf;
-	unsigned long		j;
-	int			retval;
-	cpumask_t		oldmask;
-
-	if (num_cache_leaves == 0)
-		return -ENOENT;
-
-	per_cpu(cpuid4_info, cpu) = kzalloc(
-	    sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL);
-	if (per_cpu(cpuid4_info, cpu) == NULL)
-		return -ENOMEM;
-
-	oldmask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-	if (retval)
-		goto out;
+	int j, *retval = _retval, cpu = smp_processor_id();
 
 	/* Do cpuid and store the results */
 	for (j = 0; j < num_cache_leaves; j++) {
+		struct _cpuid4_info *this_leaf;
 		this_leaf = CPUID4_INFO_IDX(cpu, j);
-		retval = cpuid4_cache_lookup(j, this_leaf);
-		if (unlikely(retval < 0)) {
+		*retval = cpuid4_cache_lookup(j, this_leaf);
+		if (unlikely(*retval < 0)) {
 			int i;
 
 			for (i = 0; i < j; i++)
@@ -567,9 +552,21 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 		}
 		cache_shared_cpu_map_setup(cpu, j);
 	}
-	set_cpus_allowed_ptr(current, &oldmask);
+}
+
+static int __cpuinit detect_cache_attributes(unsigned int cpu)
+{
+	int			retval;
+
+	if (num_cache_leaves == 0)
+		return -ENOENT;
+
+	per_cpu(cpuid4_info, cpu) = kzalloc(
+	    sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL);
+	if (per_cpu(cpuid4_info, cpu) == NULL)
+		return -ENOMEM;
 
-out:
+	smp_call_function_single(cpu, get_cpu_leaves, &retval, true);
 	if (retval) {
 		kfree(per_cpu(cpuid4_info, cpu));
 		per_cpu(cpuid4_info, cpu) = NULL;
@@ -626,8 +623,8 @@ static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf,
 		cpumask_t *mask = &this_leaf->shared_cpu_map;
 
 		n = type?
-			cpulist_scnprintf(buf, len-2, *mask):
-			cpumask_scnprintf(buf, len-2, *mask);
+			cpulist_scnprintf(buf, len-2, mask) :
+			cpumask_scnprintf(buf, len-2, mask);
 		buf[n++] = '\n';
 		buf[n] = '\0';
 	}
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index 748c8f9e7a0..a5a5e053037 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -83,34 +83,41 @@ static DEFINE_PER_CPU(unsigned char, bank_map);	/* see which banks are on */
  * CPU Initialization
  */
 
+struct thresh_restart {
+	struct threshold_block *b;
+	int reset;
+	u16 old_limit;
+};
+
 /* must be called with correct cpu affinity */
-static void threshold_restart_bank(struct threshold_block *b,
-				   int reset, u16 old_limit)
+static long threshold_restart_bank(void *_tr)
 {
+	struct thresh_restart *tr = _tr;
 	u32 mci_misc_hi, mci_misc_lo;
 
-	rdmsr(b->address, mci_misc_lo, mci_misc_hi);
+	rdmsr(tr->b->address, mci_misc_lo, mci_misc_hi);
 
-	if (b->threshold_limit < (mci_misc_hi & THRESHOLD_MAX))
-		reset = 1;	/* limit cannot be lower than err count */
+	if (tr->b->threshold_limit < (mci_misc_hi & THRESHOLD_MAX))
+		tr->reset = 1;	/* limit cannot be lower than err count */
 
-	if (reset) {		/* reset err count and overflow bit */
+	if (tr->reset) {		/* reset err count and overflow bit */
 		mci_misc_hi =
 		    (mci_misc_hi & ~(MASK_ERR_COUNT_HI | MASK_OVERFLOW_HI)) |
-		    (THRESHOLD_MAX - b->threshold_limit);
-	} else if (old_limit) {	/* change limit w/o reset */
+		    (THRESHOLD_MAX - tr->b->threshold_limit);
+	} else if (tr->old_limit) {	/* change limit w/o reset */
 		int new_count = (mci_misc_hi & THRESHOLD_MAX) +
-		    (old_limit - b->threshold_limit);
+		    (tr->old_limit - tr->b->threshold_limit);
 		mci_misc_hi = (mci_misc_hi & ~MASK_ERR_COUNT_HI) |
 		    (new_count & THRESHOLD_MAX);
 	}
 
-	b->interrupt_enable ?
+	tr->b->interrupt_enable ?
 	    (mci_misc_hi = (mci_misc_hi & ~MASK_INT_TYPE_HI) | INT_TYPE_APIC) :
 	    (mci_misc_hi &= ~MASK_INT_TYPE_HI);
 
 	mci_misc_hi |= MASK_COUNT_EN_HI;
-	wrmsr(b->address, mci_misc_lo, mci_misc_hi);
+	wrmsr(tr->b->address, mci_misc_lo, mci_misc_hi);
+	return 0;
 }
 
 /* cpu init entry point, called from mce.c with preempt off */
@@ -120,6 +127,7 @@ void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c)
 	unsigned int cpu = smp_processor_id();
 	u8 lvt_off;
 	u32 low = 0, high = 0, address = 0;
+	struct thresh_restart tr;
 
 	for (bank = 0; bank < NR_BANKS; ++bank) {
 		for (block = 0; block < NR_BLOCKS; ++block) {
@@ -162,7 +170,10 @@ void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c)
 			wrmsr(address, low, high);
 
 			threshold_defaults.address = address;
-			threshold_restart_bank(&threshold_defaults, 0, 0);
+			tr.b = &threshold_defaults;
+			tr.reset = 0;
+			tr.old_limit = 0;
+			threshold_restart_bank(&tr);
 		}
 	}
 }
@@ -251,20 +262,6 @@ struct threshold_attr {
 	ssize_t(*store) (struct threshold_block *, const char *, size_t count);
 };
 
-static void affinity_set(unsigned int cpu, cpumask_t *oldmask,
-					   cpumask_t *newmask)
-{
-	*oldmask = current->cpus_allowed;
-	cpus_clear(*newmask);
-	cpu_set(cpu, *newmask);
-	set_cpus_allowed_ptr(current, newmask);
-}
-
-static void affinity_restore(const cpumask_t *oldmask)
-{
-	set_cpus_allowed_ptr(current, oldmask);
-}
-
 #define SHOW_FIELDS(name)                                           \
 static ssize_t show_ ## name(struct threshold_block * b, char *buf) \
 {                                                                   \
@@ -277,15 +274,16 @@ static ssize_t store_interrupt_enable(struct threshold_block *b,
 				      const char *buf, size_t count)
 {
 	char *end;
-	cpumask_t oldmask, newmask;
+	struct thresh_restart tr;
 	unsigned long new = simple_strtoul(buf, &end, 0);
 	if (end == buf)
 		return -EINVAL;
 	b->interrupt_enable = !!new;
 
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 0, 0);
-	affinity_restore(&oldmask);
+	tr.b = b;
+	tr.reset = 0;
+	tr.old_limit = 0;
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 
 	return end - buf;
 }
@@ -294,8 +292,7 @@ static ssize_t store_threshold_limit(struct threshold_block *b,
 				     const char *buf, size_t count)
 {
 	char *end;
-	cpumask_t oldmask, newmask;
-	u16 old;
+	struct thresh_restart tr;
 	unsigned long new = simple_strtoul(buf, &end, 0);
 	if (end == buf)
 		return -EINVAL;
@@ -303,34 +300,36 @@ static ssize_t store_threshold_limit(struct threshold_block *b,
 		new = THRESHOLD_MAX;
 	if (new < 1)
 		new = 1;
-	old = b->threshold_limit;
+	tr.old_limit = b->threshold_limit;
 	b->threshold_limit = new;
+	tr.b = b;
+	tr.reset = 0;
 
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 0, old);
-	affinity_restore(&oldmask);
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 
 	return end - buf;
 }
 
-static ssize_t show_error_count(struct threshold_block *b, char *buf)
+static long local_error_count(void *_b)
 {
-	u32 high, low;
-	cpumask_t oldmask, newmask;
-	affinity_set(b->cpu, &oldmask, &newmask);
+	struct threshold_block *b = _b;
+	u32 low, high;
+
 	rdmsr(b->address, low, high);
-	affinity_restore(&oldmask);
-	return sprintf(buf, "%x\n",
-		       (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit));
+	return (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit);
+}
+
+static ssize_t show_error_count(struct threshold_block *b, char *buf)
+{
+	return sprintf(buf, "%lx\n", work_on_cpu(b->cpu, local_error_count, b));
 }
 
 static ssize_t store_error_count(struct threshold_block *b,
 				 const char *buf, size_t count)
 {
-	cpumask_t oldmask, newmask;
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 1, 0);
-	affinity_restore(&oldmask);
+	struct thresh_restart tr = { .b = b, .reset = 1, .old_limit = 0 };
+
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 	return 1;
 }
 
@@ -463,12 +462,19 @@ out_free:
 	return err;
 }
 
+static long local_allocate_threshold_blocks(void *_bank)
+{
+	unsigned int *bank = _bank;
+
+	return allocate_threshold_blocks(smp_processor_id(), *bank, 0,
+					 MSR_IA32_MC0_MISC + *bank * 4);
+}
+
 /* symlinks sibling shared banks to first core.  first core owns dir/files. */
 static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 {
 	int i, err = 0;
 	struct threshold_bank *b = NULL;
-	cpumask_t oldmask, newmask;
 	char name[32];
 
 	sprintf(name, "threshold_bank%i", bank);
@@ -519,11 +525,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 
 	per_cpu(threshold_banks, cpu)[bank] = b;
 
-	affinity_set(cpu, &oldmask, &newmask);
-	err = allocate_threshold_blocks(cpu, bank, 0,
-					MSR_IA32_MC0_MISC + bank * 4);
-	affinity_restore(&oldmask);
-
+	err = work_on_cpu(cpu, local_allocate_threshold_blocks, &bank);
 	if (err)
 		goto out_free;
 
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 4e8d77f01ee..b59ddcc88cd 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -14,14 +14,6 @@
 #include <asm/pat.h>
 #include "mtrr.h"
 
-struct mtrr_state {
-	struct mtrr_var_range var_ranges[MAX_VAR_RANGES];
-	mtrr_type fixed_ranges[NUM_FIXED_RANGES];
-	unsigned char enabled;
-	unsigned char have_fixed;
-	mtrr_type def_type;
-};
-
 struct fixed_range_block {
 	int base_msr; /* start address of an MTRR block */
 	int ranges;   /* number of MTRRs in this block  */
@@ -35,10 +27,12 @@ static struct fixed_range_block fixed_range_blocks[] = {
 };
 
 static unsigned long smp_changes_mask;
-static struct mtrr_state mtrr_state = {};
 static int mtrr_state_set;
 u64 mtrr_tom2;
 
+struct mtrr_state_type mtrr_state = {};
+EXPORT_SYMBOL_GPL(mtrr_state);
+
 #undef MODULE_PARAM_PREFIX
 #define MODULE_PARAM_PREFIX "mtrr."
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 44fcb237bd5..d259e5d2e05 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -49,7 +49,7 @@
 
 u32 num_var_ranges = 0;
 
-unsigned int mtrr_usage_table[MAX_VAR_RANGES];
+unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 static DEFINE_MUTEX(mtrr_mutex);
 
 u64 size_or_mask, size_and_mask;
@@ -574,7 +574,7 @@ struct mtrr_value {
 	unsigned long	lsize;
 };
 
-static struct mtrr_value mtrr_state[MAX_VAR_RANGES];
+static struct mtrr_value mtrr_state[MTRR_MAX_VAR_RANGES];
 
 static int mtrr_save(struct sys_device * sysdev, pm_message_t state)
 {
diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index 2dc4ec656b2..ffd60409cc6 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -8,11 +8,6 @@
 #define MTRRcap_MSR     0x0fe
 #define MTRRdefType_MSR 0x2ff
 
-#define MTRRphysBase_MSR(reg) (0x200 + 2 * (reg))
-#define MTRRphysMask_MSR(reg) (0x200 + 2 * (reg) + 1)
-
-#define NUM_FIXED_RANGES 88
-#define MAX_VAR_RANGES 256
 #define MTRRfix64K_00000_MSR 0x250
 #define MTRRfix16K_80000_MSR 0x258
 #define MTRRfix16K_A0000_MSR 0x259
@@ -29,11 +24,7 @@
 #define MTRR_CHANGE_MASK_VARIABLE  0x02
 #define MTRR_CHANGE_MASK_DEFTYPE   0x04
 
-/* In the Intel processor's MTRR interface, the MTRR type is always held in
-   an 8 bit field: */
-typedef u8 mtrr_type;
-
-extern unsigned int mtrr_usage_table[MAX_VAR_RANGES];
+extern unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 
 struct mtrr_ops {
 	u32	vendor;
@@ -70,13 +61,6 @@ struct set_mtrr_context {
 	u32 ccr3;
 };
 
-struct mtrr_var_range {
-	u32 base_lo;
-	u32 base_hi;
-	u32 mask_lo;
-	u32 mask_hi;
-};
-
 void set_mtrr_done(struct set_mtrr_context *ctxt);
 void set_mtrr_cache_disable(struct set_mtrr_context *ctxt);
 void set_mtrr_prepare_save(struct set_mtrr_context *ctxt);
diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index d84a852e4cd..c689d19e35a 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -26,6 +26,7 @@
 #include <linux/kdebug.h>
 #include <asm/smp.h>
 #include <asm/reboot.h>
+#include <asm/virtext.h>
 
 #include <mach_ipi.h>
 
@@ -49,6 +50,15 @@ static void kdump_nmi_callback(int cpu, struct die_args *args)
 #endif
 	crash_save_cpu(regs, cpu);
 
+	/* Disable VMX or SVM if needed.
+	 *
+	 * We need to disable virtualization on all CPUs.
+	 * Having VMX or SVM enabled on any CPU may break rebooting
+	 * after the kdump kernel has finished its task.
+	 */
+	cpu_emergency_vmxoff();
+	cpu_emergency_svm_disable();
+
 	disable_local_APIC();
 }
 
@@ -80,6 +90,14 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
 	local_irq_disable();
 
 	kdump_nmi_shootdown_cpus();
+
+	/* Booting kdump kernel with VMX or SVM enabled won't work,
+	 * because (among other limitations) we can't disable paging
+	 * with the virt flags.
+	 */
+	cpu_emergency_vmxoff();
+	cpu_emergency_svm_disable();
+
 	lapic_shutdown();
 #if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c
index c0262791bda..34185488e4f 100644
--- a/arch/x86/kernel/genapic_flat_64.c
+++ b/arch/x86/kernel/genapic_flat_64.c
@@ -30,12 +30,12 @@ static int flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 1;
 }
 
-static cpumask_t flat_target_cpus(void)
+static const struct cpumask *flat_target_cpus(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 
-static cpumask_t flat_vector_allocation_domain(int cpu)
+static void flat_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
@@ -45,8 +45,8 @@ static cpumask_t flat_vector_allocation_domain(int cpu)
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	cpumask_t domain = { { [0] = APIC_ALL_CPUS, } };
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
 }
 
 /*
@@ -69,9 +69,8 @@ static void flat_init_apic_ldr(void)
 	apic_write(APIC_LDR, val);
 }
 
-static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
+static inline void _flat_send_IPI_mask(unsigned long mask, int vector)
 {
-	unsigned long mask = cpus_addr(cpumask)[0];
 	unsigned long flags;
 
 	local_irq_save(flags);
@@ -79,20 +78,41 @@ static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
 	local_irq_restore(flags);
 }
 
+static void flat_send_IPI_mask(const struct cpumask *cpumask, int vector)
+{
+	unsigned long mask = cpumask_bits(cpumask)[0];
+
+	_flat_send_IPI_mask(mask, vector);
+}
+
+static void flat_send_IPI_mask_allbutself(const struct cpumask *cpumask,
+					  int vector)
+{
+	unsigned long mask = cpumask_bits(cpumask)[0];
+	int cpu = smp_processor_id();
+
+	if (cpu < BITS_PER_LONG)
+		clear_bit(cpu, &mask);
+	_flat_send_IPI_mask(mask, vector);
+}
+
 static void flat_send_IPI_allbutself(int vector)
 {
+	int cpu = smp_processor_id();
 #ifdef	CONFIG_HOTPLUG_CPU
 	int hotplug = 1;
 #else
 	int hotplug = 0;
 #endif
 	if (hotplug || vector == NMI_VECTOR) {
-		cpumask_t allbutme = cpu_online_map;
+		if (!cpumask_equal(cpu_online_mask, cpumask_of(cpu))) {
+			unsigned long mask = cpumask_bits(cpu_online_mask)[0];
 
-		cpu_clear(smp_processor_id(), allbutme);
+			if (cpu < BITS_PER_LONG)
+				clear_bit(cpu, &mask);
 
-		if (!cpus_empty(allbutme))
-			flat_send_IPI_mask(allbutme, vector);
+			_flat_send_IPI_mask(mask, vector);
+		}
 	} else if (num_online_cpus() > 1) {
 		__send_IPI_shortcut(APIC_DEST_ALLBUT, vector,APIC_DEST_LOGICAL);
 	}
@@ -101,7 +121,7 @@ static void flat_send_IPI_allbutself(int vector)
 static void flat_send_IPI_all(int vector)
 {
 	if (vector == NMI_VECTOR)
-		flat_send_IPI_mask(cpu_online_map, vector);
+		flat_send_IPI_mask(cpu_online_mask, vector);
 	else
 		__send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
 }
@@ -135,9 +155,18 @@ 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(cpumask_t cpumask)
+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)
 {
-	return cpus_addr(cpumask)[0] & APIC_ALL_CPUS;
+	unsigned long mask1 = cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
+	unsigned long mask2 = cpumask_bits(andmask)[0] & APIC_ALL_CPUS;
+
+	return mask1 & mask2;
 }
 
 static unsigned int phys_pkg_id(int index_msb)
@@ -157,8 +186,10 @@ struct genapic apic_flat =  {
 	.send_IPI_all = flat_send_IPI_all,
 	.send_IPI_allbutself = flat_send_IPI_allbutself,
 	.send_IPI_mask = flat_send_IPI_mask,
+	.send_IPI_mask_allbutself = flat_send_IPI_mask_allbutself,
 	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = flat_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = flat_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
@@ -188,35 +219,39 @@ static int physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 0;
 }
 
-static cpumask_t physflat_target_cpus(void)
+static const struct cpumask *physflat_target_cpus(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 
-static cpumask_t physflat_vector_allocation_domain(int cpu)
+static void physflat_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	return cpumask_of_cpu(cpu);
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
-static void physflat_send_IPI_mask(cpumask_t cpumask, int vector)
+static void physflat_send_IPI_mask(const struct cpumask *cpumask, int vector)
 {
 	send_IPI_mask_sequence(cpumask, vector);
 }
 
-static void physflat_send_IPI_allbutself(int vector)
+static void physflat_send_IPI_mask_allbutself(const struct cpumask *cpumask,
+					      int vector)
 {
-	cpumask_t allbutme = cpu_online_map;
+	send_IPI_mask_allbutself(cpumask, vector);
+}
 
-	cpu_clear(smp_processor_id(), allbutme);
-	physflat_send_IPI_mask(allbutme, vector);
+static void physflat_send_IPI_allbutself(int vector)
+{
+	send_IPI_mask_allbutself(cpu_online_mask, vector);
 }
 
 static void physflat_send_IPI_all(int vector)
 {
-	physflat_send_IPI_mask(cpu_online_map, vector);
+	physflat_send_IPI_mask(cpu_online_mask, vector);
 }
 
-static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int physflat_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -224,13 +259,31 @@ static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = cpumask_first(cpumask);
 	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int
+physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+				const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 struct genapic apic_physflat =  {
 	.name = "physical flat",
 	.acpi_madt_oem_check = physflat_acpi_madt_oem_check,
@@ -243,8 +296,10 @@ struct genapic apic_physflat =  {
 	.send_IPI_all = physflat_send_IPI_all,
 	.send_IPI_allbutself = physflat_send_IPI_allbutself,
 	.send_IPI_mask = physflat_send_IPI_mask,
+	.send_IPI_mask_allbutself = physflat_send_IPI_mask_allbutself,
 	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = physflat_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = physflat_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c
index f6a2c8eb48a..6ce497cc372 100644
--- a/arch/x86/kernel/genx2apic_cluster.c
+++ b/arch/x86/kernel/genx2apic_cluster.c
@@ -22,19 +22,18 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t x2apic_target_cpus(void)
+static const struct cpumask *x2apic_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
 /*
  * for now each logical cpu is in its own vector allocation domain.
  */
-static cpumask_t x2apic_vector_allocation_domain(int cpu)
+static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
@@ -56,32 +55,53 @@ static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
  * at once. We have 16 cpu's in a cluster. This will minimize IPI register
  * writes.
  */
-static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned long query_cpu;
 
 	local_irq_save(flags);
-	for_each_cpu_mask(query_cpu, mask) {
-		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_logical_apicid, query_cpu),
-				       vector, APIC_DEST_LOGICAL);
-	}
+	for_each_cpu(query_cpu, mask)
+		__x2apic_send_IPI_dest(
+			per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+			vector, APIC_DEST_LOGICAL);
 	local_irq_restore(flags);
 }
 
-static void x2apic_send_IPI_allbutself(int vector)
+static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask,
+					    int vector)
 {
-	cpumask_t mask = cpu_online_map;
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	cpu_clear(smp_processor_id(), mask);
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+				vector, APIC_DEST_LOGICAL);
+	local_irq_restore(flags);
+}
+
+static void x2apic_send_IPI_allbutself(int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	if (!cpus_empty(mask))
-		x2apic_send_IPI_mask(mask, vector);
+	local_irq_save(flags);
+	for_each_online_cpu(query_cpu)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+				vector, APIC_DEST_LOGICAL);
+	local_irq_restore(flags);
 }
 
 static void x2apic_send_IPI_all(int vector)
 {
-	x2apic_send_IPI_mask(cpu_online_map, vector);
+	x2apic_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int x2apic_apic_id_registered(void)
@@ -89,21 +109,38 @@ static int x2apic_apic_id_registered(void)
 	return 1;
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
 	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	cpu = cpumask_first(cpumask);
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_logical_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_logical_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -150,8 +187,10 @@ struct genapic apic_x2apic_cluster = {
 	.send_IPI_all = x2apic_send_IPI_all,
 	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
 	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself,
 	.send_IPI_self = x2apic_send_IPI_self,
 	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c
index a177c7880ab..21bcc0e098b 100644
--- a/arch/x86/kernel/genx2apic_phys.c
+++ b/arch/x86/kernel/genx2apic_phys.c
@@ -29,16 +29,15 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t x2apic_target_cpus(void)
+static const struct cpumask *x2apic_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
-static cpumask_t x2apic_vector_allocation_domain(int cpu)
+static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
@@ -54,32 +53,54 @@ static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
 	x2apic_icr_write(cfg, apicid);
 }
 
-static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned long query_cpu;
 
 	local_irq_save(flags);
-	for_each_cpu_mask(query_cpu, mask) {
+	for_each_cpu(query_cpu, mask) {
 		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu),
 				       vector, APIC_DEST_PHYSICAL);
 	}
 	local_irq_restore(flags);
 }
 
-static void x2apic_send_IPI_allbutself(int vector)
+static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask,
+					    int vector)
 {
-	cpumask_t mask = cpu_online_map;
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask) {
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_apicid, query_cpu),
+				vector, APIC_DEST_PHYSICAL);
+	}
+	local_irq_restore(flags);
+}
 
-	cpu_clear(smp_processor_id(), mask);
+static void x2apic_send_IPI_allbutself(int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	if (!cpus_empty(mask))
-		x2apic_send_IPI_mask(mask, vector);
+	local_irq_save(flags);
+	for_each_online_cpu(query_cpu)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_apicid, query_cpu),
+				vector, APIC_DEST_PHYSICAL);
+	local_irq_restore(flags);
 }
 
 static void x2apic_send_IPI_all(int vector)
 {
-	x2apic_send_IPI_mask(cpu_online_map, vector);
+	x2apic_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int x2apic_apic_id_registered(void)
@@ -87,7 +108,7 @@ static int x2apic_apic_id_registered(void)
 	return 1;
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -95,13 +116,30 @@ static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	cpu = cpumask_first(cpumask);
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -145,8 +183,10 @@ struct genapic apic_x2apic_phys = {
 	.send_IPI_all = x2apic_send_IPI_all,
 	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
 	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself,
 	.send_IPI_self = x2apic_send_IPI_self,
 	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
index dece1728973..b193e082f6c 100644
--- a/arch/x86/kernel/genx2apic_uv_x.c
+++ b/arch/x86/kernel/genx2apic_uv_x.c
@@ -79,16 +79,15 @@ EXPORT_SYMBOL(sn_rtc_cycles_per_second);
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t uv_target_cpus(void)
+static const struct cpumask *uv_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
-static cpumask_t uv_vector_allocation_domain(int cpu)
+static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
@@ -127,28 +126,37 @@ static void uv_send_IPI_one(int cpu, int vector)
 	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
 }
 
-static void uv_send_IPI_mask(cpumask_t mask, int vector)
+static void uv_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned int cpu;
 
-	for_each_possible_cpu(cpu)
-		if (cpu_isset(cpu, mask))
+	for_each_cpu(cpu, mask)
+		uv_send_IPI_one(cpu, vector);
+}
+
+static void uv_send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+	unsigned int cpu;
+	unsigned int this_cpu = smp_processor_id();
+
+	for_each_cpu(cpu, mask)
+		if (cpu != this_cpu)
 			uv_send_IPI_one(cpu, vector);
 }
 
 static void uv_send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
-
-	cpu_clear(smp_processor_id(), mask);
+	unsigned int cpu;
+	unsigned int this_cpu = smp_processor_id();
 
-	if (!cpus_empty(mask))
-		uv_send_IPI_mask(mask, vector);
+	for_each_online_cpu(cpu)
+		if (cpu != this_cpu)
+			uv_send_IPI_one(cpu, vector);
 }
 
 static void uv_send_IPI_all(int vector)
 {
-	uv_send_IPI_mask(cpu_online_map, vector);
+	uv_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int uv_apic_id_registered(void)
@@ -160,7 +168,7 @@ static void uv_init_apic_ldr(void)
 {
 }
 
-static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int uv_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -168,13 +176,30 @@ static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = cpumask_first(cpumask);
 	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+					      const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -222,8 +247,10 @@ struct genapic apic_x2apic_uv_x = {
 	.send_IPI_all = uv_send_IPI_all,
 	.send_IPI_allbutself = uv_send_IPI_allbutself,
 	.send_IPI_mask = uv_send_IPI_mask,
+	.send_IPI_mask_allbutself = uv_send_IPI_mask_allbutself,
 	.send_IPI_self = uv_send_IPI_self,
 	.cpu_mask_to_apicid = uv_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = uv_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 3f0a3edf0a5..cd759ad9069 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -248,7 +248,7 @@ static void hpet_legacy_clockevent_register(void)
 	 * Start hpet with the boot cpu mask and make it
 	 * global after the IO_APIC has been initialized.
 	 */
-	hpet_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	hpet_clockevent.cpumask = cpumask_of(smp_processor_id());
 	clockevents_register_device(&hpet_clockevent);
 	global_clock_event = &hpet_clockevent;
 	printk(KERN_DEBUG "hpet clockevent registered\n");
@@ -303,7 +303,7 @@ static void hpet_set_mode(enum clock_event_mode mode,
 			struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
 			hpet_setup_msi_irq(hdev->irq);
 			disable_irq(hdev->irq);
-			irq_set_affinity(hdev->irq, cpumask_of_cpu(hdev->cpu));
+			irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu));
 			enable_irq(hdev->irq);
 		}
 		break;
@@ -451,7 +451,7 @@ static int hpet_setup_irq(struct hpet_dev *dev)
 		return -1;
 
 	disable_irq(dev->irq);
-	irq_set_affinity(dev->irq, cpumask_of_cpu(dev->cpu));
+	irq_set_affinity(dev->irq, cpumask_of(dev->cpu));
 	enable_irq(dev->irq);
 
 	printk(KERN_DEBUG "hpet: %s irq %d for MSI\n",
@@ -502,7 +502,7 @@ static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu)
 	/* 5 usec minimum reprogramming delta. */
 	evt->min_delta_ns = 5000;
 
-	evt->cpumask = cpumask_of_cpu(hdev->cpu);
+	evt->cpumask = cpumask_of(hdev->cpu);
 	clockevents_register_device(evt);
 }
 
@@ -813,7 +813,7 @@ int __init hpet_enable(void)
 
 out_nohpet:
 	hpet_clear_mapping();
-	boot_hpet_disable = 1;
+	hpet_address = 0;
 	return 0;
 }
 
@@ -836,10 +836,11 @@ static __init int hpet_late_init(void)
 
 		hpet_address = force_hpet_address;
 		hpet_enable();
-		if (!hpet_virt_address)
-			return -ENODEV;
 	}
 
+	if (!hpet_virt_address)
+		return -ENODEV;
+
 	hpet_reserve_platform_timers(hpet_readl(HPET_ID));
 
 	for_each_online_cpu(cpu) {
diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c
index c1b5e3ece1f..10f92fb532f 100644
--- a/arch/x86/kernel/i8253.c
+++ b/arch/x86/kernel/i8253.c
@@ -114,7 +114,7 @@ void __init setup_pit_timer(void)
 	 * Start pit with the boot cpu mask and make it global after the
 	 * IO_APIC has been initialized.
 	 */
-	pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	pit_clockevent.cpumask = cpumask_of(smp_processor_id());
 	pit_clockevent.mult = div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC,
 				     pit_clockevent.shift);
 	pit_clockevent.max_delta_ns =
diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index d39918076bb..df3bf269bea 100644
--- a/arch/x86/kernel/init_task.c
+++ b/arch/x86/kernel/init_task.c
@@ -10,7 +10,6 @@
 #include <asm/pgtable.h>
 #include <asm/desc.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index b8c8a8e9934..69911722b9d 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -108,94 +108,277 @@ static int __init parse_noapic(char *str)
 early_param("noapic", parse_noapic);
 
 struct irq_pin_list;
+
+/*
+ * This is performance-critical, we want to do it O(1)
+ *
+ * the indexing order of this array favors 1:1 mappings
+ * between pins and IRQs.
+ */
+
+struct irq_pin_list {
+	int apic, pin;
+	struct irq_pin_list *next;
+};
+
+static struct irq_pin_list *get_one_free_irq_2_pin(int cpu)
+{
+	struct irq_pin_list *pin;
+	int node;
+
+	node = cpu_to_node(cpu);
+
+	pin = kzalloc_node(sizeof(*pin), GFP_ATOMIC, node);
+	printk(KERN_DEBUG "  alloc irq_2_pin on cpu %d node %d\n", cpu, node);
+
+	return pin;
+}
+
 struct irq_cfg {
-	unsigned int irq;
 	struct irq_pin_list *irq_2_pin;
-	cpumask_t domain;
-	cpumask_t old_domain;
+	cpumask_var_t domain;
+	cpumask_var_t old_domain;
 	unsigned move_cleanup_count;
 	u8 vector;
 	u8 move_in_progress : 1;
+#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
+	u8 move_desc_pending : 1;
+#endif
 };
 
 /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */
+#ifdef CONFIG_SPARSE_IRQ
+static struct irq_cfg irq_cfgx[] = {
+#else
 static struct irq_cfg irq_cfgx[NR_IRQS] = {
-	[0]  = { .irq =  0, .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR,  },
-	[1]  = { .irq =  1, .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR,  },
-	[2]  = { .irq =  2, .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR,  },
-	[3]  = { .irq =  3, .domain = CPU_MASK_ALL, .vector = IRQ3_VECTOR,  },
-	[4]  = { .irq =  4, .domain = CPU_MASK_ALL, .vector = IRQ4_VECTOR,  },
-	[5]  = { .irq =  5, .domain = CPU_MASK_ALL, .vector = IRQ5_VECTOR,  },
-	[6]  = { .irq =  6, .domain = CPU_MASK_ALL, .vector = IRQ6_VECTOR,  },
-	[7]  = { .irq =  7, .domain = CPU_MASK_ALL, .vector = IRQ7_VECTOR,  },
-	[8]  = { .irq =  8, .domain = CPU_MASK_ALL, .vector = IRQ8_VECTOR,  },
-	[9]  = { .irq =  9, .domain = CPU_MASK_ALL, .vector = IRQ9_VECTOR,  },
-	[10] = { .irq = 10, .domain = CPU_MASK_ALL, .vector = IRQ10_VECTOR, },
-	[11] = { .irq = 11, .domain = CPU_MASK_ALL, .vector = IRQ11_VECTOR, },
-	[12] = { .irq = 12, .domain = CPU_MASK_ALL, .vector = IRQ12_VECTOR, },
-	[13] = { .irq = 13, .domain = CPU_MASK_ALL, .vector = IRQ13_VECTOR, },
-	[14] = { .irq = 14, .domain = CPU_MASK_ALL, .vector = IRQ14_VECTOR, },
-	[15] = { .irq = 15, .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, },
+#endif
+	[0]  = { .vector = IRQ0_VECTOR,  },
+	[1]  = { .vector = IRQ1_VECTOR,  },
+	[2]  = { .vector = IRQ2_VECTOR,  },
+	[3]  = { .vector = IRQ3_VECTOR,  },
+	[4]  = { .vector = IRQ4_VECTOR,  },
+	[5]  = { .vector = IRQ5_VECTOR,  },
+	[6]  = { .vector = IRQ6_VECTOR,  },
+	[7]  = { .vector = IRQ7_VECTOR,  },
+	[8]  = { .vector = IRQ8_VECTOR,  },
+	[9]  = { .vector = IRQ9_VECTOR,  },
+	[10] = { .vector = IRQ10_VECTOR, },
+	[11] = { .vector = IRQ11_VECTOR, },
+	[12] = { .vector = IRQ12_VECTOR, },
+	[13] = { .vector = IRQ13_VECTOR, },
+	[14] = { .vector = IRQ14_VECTOR, },
+	[15] = { .vector = IRQ15_VECTOR, },
 };
 
-#define for_each_irq_cfg(irq, cfg)		\
-	for (irq = 0, cfg = irq_cfgx; irq < nr_irqs; irq++, cfg++)
+int __init arch_early_irq_init(void)
+{
+	struct irq_cfg *cfg;
+	struct irq_desc *desc;
+	int count;
+	int i;
+
+	cfg = irq_cfgx;
+	count = ARRAY_SIZE(irq_cfgx);
+
+	for (i = 0; i < count; i++) {
+		desc = irq_to_desc(i);
+		desc->chip_data = &cfg[i];
+		alloc_bootmem_cpumask_var(&cfg[i].domain);
+		alloc_bootmem_cpumask_var(&cfg[i].old_domain);
+		if (i < NR_IRQS_LEGACY)
+			cpumask_setall(cfg[i].domain);
+	}
+
+	return 0;
+}
 
+#ifdef CONFIG_SPARSE_IRQ
 static struct irq_cfg *irq_cfg(unsigned int irq)
 {
-	return irq < nr_irqs ? irq_cfgx + irq : NULL;
+	struct irq_cfg *cfg = NULL;
+	struct irq_desc *desc;
+
+	desc = irq_to_desc(irq);
+	if (desc)
+		cfg = desc->chip_data;
+
+	return cfg;
 }
 
-static struct irq_cfg *irq_cfg_alloc(unsigned int irq)
+static struct irq_cfg *get_one_free_irq_cfg(int cpu)
 {
-	return irq_cfg(irq);
+	struct irq_cfg *cfg;
+	int node;
+
+	node = cpu_to_node(cpu);
+
+	cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node);
+	if (cfg) {
+		/* FIXME: needs alloc_cpumask_var_node() */
+		if (!alloc_cpumask_var(&cfg->domain, GFP_ATOMIC)) {
+			kfree(cfg);
+			cfg = NULL;
+		} else if (!alloc_cpumask_var(&cfg->old_domain, GFP_ATOMIC)) {
+			free_cpumask_var(cfg->domain);
+			kfree(cfg);
+			cfg = NULL;
+		} else {
+			cpumask_clear(cfg->domain);
+			cpumask_clear(cfg->old_domain);
+		}
+	}
+	printk(KERN_DEBUG "  alloc irq_cfg on cpu %d node %d\n", cpu, node);
+
+	return cfg;
 }
 
-/*
- * Rough estimation of how many shared IRQs there are, can be changed
- * anytime.
- */
-#define MAX_PLUS_SHARED_IRQS NR_IRQS
-#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS)
+int arch_init_chip_data(struct irq_desc *desc, int cpu)
+{
+	struct irq_cfg *cfg;
 
-/*
- * This is performance-critical, we want to do it O(1)
- *
- * the indexing order of this array favors 1:1 mappings
- * between pins and IRQs.
- */
+	cfg = desc->chip_data;
+	if (!cfg) {
+		desc->chip_data = get_one_free_irq_cfg(cpu);
+		if (!desc->chip_data) {
+			printk(KERN_ERR "can not alloc irq_cfg\n");
+			BUG_ON(1);
+		}
+	}
 
-struct irq_pin_list {
-	int apic, pin;
-	struct irq_pin_list *next;
-};
+	return 0;
+}
 
-static struct irq_pin_list irq_2_pin_head[PIN_MAP_SIZE];
-static struct irq_pin_list *irq_2_pin_ptr;
+#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
 
-static void __init irq_2_pin_init(void)
+static void
+init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int cpu)
 {
-	struct irq_pin_list *pin = irq_2_pin_head;
-	int i;
+	struct irq_pin_list *old_entry, *head, *tail, *entry;
+
+	cfg->irq_2_pin = NULL;
+	old_entry = old_cfg->irq_2_pin;
+	if (!old_entry)
+		return;
+
+	entry = get_one_free_irq_2_pin(cpu);
+	if (!entry)
+		return;
 
-	for (i = 1; i < PIN_MAP_SIZE; i++)
-		pin[i-1].next = &pin[i];
+	entry->apic	= old_entry->apic;
+	entry->pin	= old_entry->pin;
+	head		= entry;
+	tail		= entry;
+	old_entry	= old_entry->next;
+	while (old_entry) {
+		entry = get_one_free_irq_2_pin(cpu);
+		if (!entry) {
+			entry = head;
+			while (entry) {
+				head = entry->next;
+				kfree(entry);
+				entry = head;
+			}
+			/* still use the old one */
+			return;
+		}
+		entry->apic	= old_entry->apic;
+		entry->pin	= old_entry->pin;
+		tail->next	= entry;
+		tail		= entry;
+		old_entry	= old_entry->next;
+	}
 
-	irq_2_pin_ptr = &pin[0];
+	tail->next = NULL;
+	cfg->irq_2_pin = head;
 }
 
-static struct irq_pin_list *get_one_free_irq_2_pin(void)
+static void free_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg)
 {
-	struct irq_pin_list *pin = irq_2_pin_ptr;
+	struct irq_pin_list *entry, *next;
 
-	if (!pin)
-		panic("can not get more irq_2_pin\n");
+	if (old_cfg->irq_2_pin == cfg->irq_2_pin)
+		return;
 
-	irq_2_pin_ptr = pin->next;
-	pin->next = NULL;
-	return pin;
+	entry = old_cfg->irq_2_pin;
+
+	while (entry) {
+		next = entry->next;
+		kfree(entry);
+		entry = next;
+	}
+	old_cfg->irq_2_pin = NULL;
+}
+
+void arch_init_copy_chip_data(struct irq_desc *old_desc,
+				 struct irq_desc *desc, int cpu)
+{
+	struct irq_cfg *cfg;
+	struct irq_cfg *old_cfg;
+
+	cfg = get_one_free_irq_cfg(cpu);
+
+	if (!cfg)
+		return;
+
+	desc->chip_data = cfg;
+
+	old_cfg = old_desc->chip_data;
+
+	memcpy(cfg, old_cfg, sizeof(struct irq_cfg));
+
+	init_copy_irq_2_pin(old_cfg, cfg, cpu);
 }
 
+static void free_irq_cfg(struct irq_cfg *old_cfg)
+{
+	kfree(old_cfg);
+}
+
+void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc)
+{
+	struct irq_cfg *old_cfg, *cfg;
+
+	old_cfg = old_desc->chip_data;
+	cfg = desc->chip_data;
+
+	if (old_cfg == cfg)
+		return;
+
+	if (old_cfg) {
+		free_irq_2_pin(old_cfg, cfg);
+		free_irq_cfg(old_cfg);
+		old_desc->chip_data = NULL;
+	}
+}
+
+static void
+set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask)
+{
+	struct irq_cfg *cfg = desc->chip_data;
+
+	if (!cfg->move_in_progress) {
+		/* it means that domain is not changed */
+		if (!cpumask_intersects(&desc->affinity, mask))
+			cfg->move_desc_pending = 1;
+	}
+}
+#endif
+
+#else
+static struct irq_cfg *irq_cfg(unsigned int irq)
+{
+	return irq < nr_irqs ? irq_cfgx + irq : NULL;
+}
+
+#endif
+
+#ifndef CONFIG_NUMA_MIGRATE_IRQ_DESC
+static inline void
+set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask)
+{
+}
+#endif
+
 struct io_apic {
 	unsigned int index;
 	unsigned int unused[3];
@@ -237,11 +420,10 @@ static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned
 	writel(value, &io_apic->data);
 }
 
-static bool io_apic_level_ack_pending(unsigned int irq)
+static bool io_apic_level_ack_pending(struct irq_cfg *cfg)
 {
 	struct irq_pin_list *entry;
 	unsigned long flags;
-	struct irq_cfg *cfg = irq_cfg(irq);
 
 	spin_lock_irqsave(&ioapic_lock, flags);
 	entry = cfg->irq_2_pin;
@@ -323,13 +505,32 @@ static void ioapic_mask_entry(int apic, int pin)
 }
 
 #ifdef CONFIG_SMP
-static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector)
+static void send_cleanup_vector(struct irq_cfg *cfg)
+{
+	cpumask_var_t cleanup_mask;
+
+	if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) {
+		unsigned int i;
+		cfg->move_cleanup_count = 0;
+		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
+			cfg->move_cleanup_count++;
+		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
+			send_IPI_mask(cpumask_of(i), IRQ_MOVE_CLEANUP_VECTOR);
+	} else {
+		cpumask_and(cleanup_mask, cfg->old_domain, cpu_online_mask);
+		cfg->move_cleanup_count = cpumask_weight(cleanup_mask);
+		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
+		free_cpumask_var(cleanup_mask);
+	}
+	cfg->move_in_progress = 0;
+}
+
+static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg)
 {
 	int apic, pin;
-	struct irq_cfg *cfg;
 	struct irq_pin_list *entry;
+	u8 vector = cfg->vector;
 
-	cfg = irq_cfg(irq);
 	entry = cfg->irq_2_pin;
 	for (;;) {
 		unsigned int reg;
@@ -359,36 +560,61 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector)
 	}
 }
 
-static int assign_irq_vector(int irq, cpumask_t mask);
+static int
+assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask);
 
-static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+/*
+ * Either sets desc->affinity to a valid value, and returns cpu_mask_to_apicid
+ * of that, or returns BAD_APICID and leaves desc->affinity untouched.
+ */
+static unsigned int
+set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask)
+{
+	struct irq_cfg *cfg;
+	unsigned int irq;
+
+	if (!cpumask_intersects(mask, cpu_online_mask))
+		return BAD_APICID;
+
+	irq = desc->irq;
+	cfg = desc->chip_data;
+	if (assign_irq_vector(irq, cfg, mask))
+		return BAD_APICID;
+
+	cpumask_and(&desc->affinity, cfg->domain, mask);
+	set_extra_move_desc(desc, mask);
+	return cpu_mask_to_apicid_and(&desc->affinity, cpu_online_mask);
+}
+
+static void
+set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	unsigned long flags;
 	unsigned int dest;
-	cpumask_t tmp;
-	struct irq_desc *desc;
+	unsigned int irq;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
-		return;
+	irq = desc->irq;
+	cfg = desc->chip_data;
 
-	cfg = irq_cfg(irq);
-	if (assign_irq_vector(irq, mask))
-		return;
+	spin_lock_irqsave(&ioapic_lock, flags);
+	dest = set_desc_affinity(desc, mask);
+	if (dest != BAD_APICID) {
+		/* Only the high 8 bits are valid. */
+		dest = SET_APIC_LOGICAL_ID(dest);
+		__target_IO_APIC_irq(irq, dest, cfg);
+	}
+	spin_unlock_irqrestore(&ioapic_lock, flags);
+}
 
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
-	/*
-	 * Only the high 8 bits are valid.
-	 */
-	dest = SET_APIC_LOGICAL_ID(dest);
+static void
+set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask)
+{
+	struct irq_desc *desc;
 
 	desc = irq_to_desc(irq);
-	spin_lock_irqsave(&ioapic_lock, flags);
-	__target_IO_APIC_irq(irq, dest, cfg->vector);
-	desc->affinity = mask;
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+
+	set_ioapic_affinity_irq_desc(desc, mask);
 }
 #endif /* CONFIG_SMP */
 
@@ -397,16 +623,18 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
  * shared ISA-space IRQs, so we have to support them. We are super
  * fast in the common case, and fast for shared ISA-space IRQs.
  */
-static void add_pin_to_irq(unsigned int irq, int apic, int pin)
+static void add_pin_to_irq_cpu(struct irq_cfg *cfg, int cpu, int apic, int pin)
 {
-	struct irq_cfg *cfg;
 	struct irq_pin_list *entry;
 
-	/* first time to refer irq_cfg, so with new */
-	cfg = irq_cfg_alloc(irq);
 	entry = cfg->irq_2_pin;
 	if (!entry) {
-		entry = get_one_free_irq_2_pin();
+		entry = get_one_free_irq_2_pin(cpu);
+		if (!entry) {
+			printk(KERN_ERR "can not alloc irq_2_pin to add %d - %d\n",
+					apic, pin);
+			return;
+		}
 		cfg->irq_2_pin = entry;
 		entry->apic = apic;
 		entry->pin = pin;
@@ -421,7 +649,7 @@ static void add_pin_to_irq(unsigned int irq, int apic, int pin)
 		entry = entry->next;
 	}
 
-	entry->next = get_one_free_irq_2_pin();
+	entry->next = get_one_free_irq_2_pin(cpu);
 	entry = entry->next;
 	entry->apic = apic;
 	entry->pin = pin;
@@ -430,11 +658,10 @@ static void add_pin_to_irq(unsigned int irq, int apic, int pin)
 /*
  * Reroute an IRQ to a different pin.
  */
-static void __init replace_pin_at_irq(unsigned int irq,
+static void __init replace_pin_at_irq_cpu(struct irq_cfg *cfg, int cpu,
 				      int oldapic, int oldpin,
 				      int newapic, int newpin)
 {
-	struct irq_cfg *cfg = irq_cfg(irq);
 	struct irq_pin_list *entry = cfg->irq_2_pin;
 	int replaced = 0;
 
@@ -451,18 +678,16 @@ static void __init replace_pin_at_irq(unsigned int irq,
 
 	/* why? call replace before add? */
 	if (!replaced)
-		add_pin_to_irq(irq, newapic, newpin);
+		add_pin_to_irq_cpu(cfg, cpu, newapic, newpin);
 }
 
-static inline void io_apic_modify_irq(unsigned int irq,
+static inline void io_apic_modify_irq(struct irq_cfg *cfg,
 				int mask_and, int mask_or,
 				void (*final)(struct irq_pin_list *entry))
 {
 	int pin;
-	struct irq_cfg *cfg;
 	struct irq_pin_list *entry;
 
-	cfg = irq_cfg(irq);
 	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
 		unsigned int reg;
 		pin = entry->pin;
@@ -475,9 +700,9 @@ static inline void io_apic_modify_irq(unsigned int irq,
 	}
 }
 
-static void __unmask_IO_APIC_irq(unsigned int irq)
+static void __unmask_IO_APIC_irq(struct irq_cfg *cfg)
 {
-	io_apic_modify_irq(irq, ~IO_APIC_REDIR_MASKED, 0, NULL);
+	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL);
 }
 
 #ifdef CONFIG_X86_64
@@ -492,47 +717,64 @@ static void io_apic_sync(struct irq_pin_list *entry)
 	readl(&io_apic->data);
 }
 
-static void __mask_IO_APIC_irq(unsigned int irq)
+static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
 {
-	io_apic_modify_irq(irq, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
+	io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
 }
 #else /* CONFIG_X86_32 */
-static void __mask_IO_APIC_irq(unsigned int irq)
+static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
 {
-	io_apic_modify_irq(irq, ~0, IO_APIC_REDIR_MASKED, NULL);
+	io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, NULL);
 }
 
-static void __mask_and_edge_IO_APIC_irq(unsigned int irq)
+static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg)
 {
-	io_apic_modify_irq(irq, ~IO_APIC_REDIR_LEVEL_TRIGGER,
+	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_LEVEL_TRIGGER,
 			IO_APIC_REDIR_MASKED, NULL);
 }
 
-static void __unmask_and_level_IO_APIC_irq(unsigned int irq)
+static void __unmask_and_level_IO_APIC_irq(struct irq_cfg *cfg)
 {
-	io_apic_modify_irq(irq, ~IO_APIC_REDIR_MASKED,
+	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED,
 			IO_APIC_REDIR_LEVEL_TRIGGER, NULL);
 }
 #endif /* CONFIG_X86_32 */
 
-static void mask_IO_APIC_irq (unsigned int irq)
+static void mask_IO_APIC_irq_desc(struct irq_desc *desc)
 {
+	struct irq_cfg *cfg = desc->chip_data;
 	unsigned long flags;
 
+	BUG_ON(!cfg);
+
 	spin_lock_irqsave(&ioapic_lock, flags);
-	__mask_IO_APIC_irq(irq);
+	__mask_IO_APIC_irq(cfg);
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void unmask_IO_APIC_irq (unsigned int irq)
+static void unmask_IO_APIC_irq_desc(struct irq_desc *desc)
 {
+	struct irq_cfg *cfg = desc->chip_data;
 	unsigned long flags;
 
 	spin_lock_irqsave(&ioapic_lock, flags);
-	__unmask_IO_APIC_irq(irq);
+	__unmask_IO_APIC_irq(cfg);
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
+static void mask_IO_APIC_irq(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	mask_IO_APIC_irq_desc(desc);
+}
+static void unmask_IO_APIC_irq(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	unmask_IO_APIC_irq_desc(desc);
+}
+
 static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 {
 	struct IO_APIC_route_entry entry;
@@ -809,7 +1051,7 @@ EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
  */
 static int EISA_ELCR(unsigned int irq)
 {
-	if (irq < 16) {
+	if (irq < NR_IRQS_LEGACY) {
 		unsigned int port = 0x4d0 + (irq >> 3);
 		return (inb(port) >> (irq & 7)) & 1;
 	}
@@ -1034,7 +1276,8 @@ void unlock_vector_lock(void)
 	spin_unlock(&vector_lock);
 }
 
-static int __assign_irq_vector(int irq, cpumask_t mask)
+static int
+__assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 {
 	/*
 	 * NOTE! The local APIC isn't very good at handling
@@ -1049,52 +1292,49 @@ static int __assign_irq_vector(int irq, cpumask_t mask)
 	 */
 	static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0;
 	unsigned int old_vector;
-	int cpu;
-	struct irq_cfg *cfg;
-
-	cfg = irq_cfg(irq);
-
-	/* Only try and allocate irqs on cpus that are present */
-	cpus_and(mask, mask, cpu_online_map);
+	int cpu, err;
+	cpumask_var_t tmp_mask;
 
 	if ((cfg->move_in_progress) || cfg->move_cleanup_count)
 		return -EBUSY;
 
+	if (!alloc_cpumask_var(&tmp_mask, GFP_ATOMIC))
+		return -ENOMEM;
+
 	old_vector = cfg->vector;
 	if (old_vector) {
-		cpumask_t tmp;
-		cpus_and(tmp, cfg->domain, mask);
-		if (!cpus_empty(tmp))
+		cpumask_and(tmp_mask, mask, cpu_online_mask);
+		cpumask_and(tmp_mask, cfg->domain, tmp_mask);
+		if (!cpumask_empty(tmp_mask)) {
+			free_cpumask_var(tmp_mask);
 			return 0;
+		}
 	}
 
-	for_each_cpu_mask_nr(cpu, mask) {
-		cpumask_t domain, new_mask;
+	/* Only try and allocate irqs on cpus that are present */
+	err = -ENOSPC;
+	for_each_cpu_and(cpu, mask, cpu_online_mask) {
 		int new_cpu;
 		int vector, offset;
 
-		domain = vector_allocation_domain(cpu);
-		cpus_and(new_mask, domain, cpu_online_map);
+		vector_allocation_domain(cpu, tmp_mask);
 
 		vector = current_vector;
 		offset = current_offset;
 next:
 		vector += 8;
 		if (vector >= first_system_vector) {
-			/* If we run out of vectors on large boxen, must share them. */
+			/* If out of vectors on large boxen, must share them. */
 			offset = (offset + 1) % 8;
 			vector = FIRST_DEVICE_VECTOR + offset;
 		}
 		if (unlikely(current_vector == vector))
 			continue;
-#ifdef CONFIG_X86_64
-		if (vector == IA32_SYSCALL_VECTOR)
-			goto next;
-#else
-		if (vector == SYSCALL_VECTOR)
+
+		if (test_bit(vector, used_vectors))
 			goto next;
-#endif
-		for_each_cpu_mask_nr(new_cpu, new_mask)
+
+		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask)
 			if (per_cpu(vector_irq, new_cpu)[vector] != -1)
 				goto next;
 		/* Found one! */
@@ -1102,49 +1342,47 @@ next:
 		current_offset = offset;
 		if (old_vector) {
 			cfg->move_in_progress = 1;
-			cfg->old_domain = cfg->domain;
+			cpumask_copy(cfg->old_domain, cfg->domain);
 		}
-		for_each_cpu_mask_nr(new_cpu, new_mask)
+		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask)
 			per_cpu(vector_irq, new_cpu)[vector] = irq;
 		cfg->vector = vector;
-		cfg->domain = domain;
-		return 0;
+		cpumask_copy(cfg->domain, tmp_mask);
+		err = 0;
+		break;
 	}
-	return -ENOSPC;
+	free_cpumask_var(tmp_mask);
+	return err;
 }
 
-static int assign_irq_vector(int irq, cpumask_t mask)
+static int
+assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 {
 	int err;
 	unsigned long flags;
 
 	spin_lock_irqsave(&vector_lock, flags);
-	err = __assign_irq_vector(irq, mask);
+	err = __assign_irq_vector(irq, cfg, mask);
 	spin_unlock_irqrestore(&vector_lock, flags);
 	return err;
 }
 
-static void __clear_irq_vector(int irq)
+static void __clear_irq_vector(int irq, struct irq_cfg *cfg)
 {
-	struct irq_cfg *cfg;
-	cpumask_t mask;
 	int cpu, vector;
 
-	cfg = irq_cfg(irq);
 	BUG_ON(!cfg->vector);
 
 	vector = cfg->vector;
-	cpus_and(mask, cfg->domain, cpu_online_map);
-	for_each_cpu_mask_nr(cpu, mask)
+	for_each_cpu_and(cpu, cfg->domain, cpu_online_mask)
 		per_cpu(vector_irq, cpu)[vector] = -1;
 
 	cfg->vector = 0;
-	cpus_clear(cfg->domain);
+	cpumask_clear(cfg->domain);
 
 	if (likely(!cfg->move_in_progress))
 		return;
-	cpus_and(mask, cfg->old_domain, cpu_online_map);
-	for_each_cpu_mask_nr(cpu, mask) {
+	for_each_cpu_and(cpu, cfg->old_domain, cpu_online_mask) {
 		for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS;
 								vector++) {
 			if (per_cpu(vector_irq, cpu)[vector] != irq)
@@ -1162,10 +1400,12 @@ void __setup_vector_irq(int cpu)
 	/* This function must be called with vector_lock held */
 	int irq, vector;
 	struct irq_cfg *cfg;
+	struct irq_desc *desc;
 
 	/* Mark the inuse vectors */
-	for_each_irq_cfg(irq, cfg) {
-		if (!cpu_isset(cpu, cfg->domain))
+	for_each_irq_desc(irq, desc) {
+		cfg = desc->chip_data;
+		if (!cpumask_test_cpu(cpu, cfg->domain))
 			continue;
 		vector = cfg->vector;
 		per_cpu(vector_irq, cpu)[vector] = irq;
@@ -1177,7 +1417,7 @@ void __setup_vector_irq(int cpu)
 			continue;
 
 		cfg = irq_cfg(irq);
-		if (!cpu_isset(cpu, cfg->domain))
+		if (!cpumask_test_cpu(cpu, cfg->domain))
 			per_cpu(vector_irq, cpu)[vector] = -1;
 	}
 }
@@ -1215,11 +1455,8 @@ static inline int IO_APIC_irq_trigger(int irq)
 }
 #endif
 
-static void ioapic_register_intr(int irq, unsigned long trigger)
+static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long trigger)
 {
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
 
 	if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
 	    trigger == IOAPIC_LEVEL)
@@ -1311,23 +1548,22 @@ static int setup_ioapic_entry(int apic, int irq,
 	return 0;
 }
 
-static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq,
+static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_desc *desc,
 			      int trigger, int polarity)
 {
 	struct irq_cfg *cfg;
 	struct IO_APIC_route_entry entry;
-	cpumask_t mask;
+	unsigned int dest;
 
 	if (!IO_APIC_IRQ(irq))
 		return;
 
-	cfg = irq_cfg(irq);
+	cfg = desc->chip_data;
 
-	mask = TARGET_CPUS;
-	if (assign_irq_vector(irq, mask))
+	if (assign_irq_vector(irq, cfg, TARGET_CPUS))
 		return;
 
-	cpus_and(mask, cfg->domain, mask);
+	dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 	apic_printk(APIC_VERBOSE,KERN_DEBUG
 		    "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> "
@@ -1337,16 +1573,15 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq,
 
 
 	if (setup_ioapic_entry(mp_ioapics[apic].mp_apicid, irq, &entry,
-			       cpu_mask_to_apicid(mask), trigger, polarity,
-			       cfg->vector)) {
+			       dest, trigger, polarity, cfg->vector)) {
 		printk("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
 		       mp_ioapics[apic].mp_apicid, pin);
-		__clear_irq_vector(irq);
+		__clear_irq_vector(irq, cfg);
 		return;
 	}
 
-	ioapic_register_intr(irq, trigger);
-	if (irq < 16)
+	ioapic_register_intr(irq, desc, trigger);
+	if (irq < NR_IRQS_LEGACY)
 		disable_8259A_irq(irq);
 
 	ioapic_write_entry(apic, pin, entry);
@@ -1356,6 +1591,9 @@ static void __init setup_IO_APIC_irqs(void)
 {
 	int apic, pin, idx, irq;
 	int notcon = 0;
+	struct irq_desc *desc;
+	struct irq_cfg *cfg;
+	int cpu = boot_cpu_id;
 
 	apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
 
@@ -1387,9 +1625,15 @@ static void __init setup_IO_APIC_irqs(void)
 			if (multi_timer_check(apic, irq))
 				continue;
 #endif
-			add_pin_to_irq(irq, apic, pin);
+			desc = irq_to_desc_alloc_cpu(irq, cpu);
+			if (!desc) {
+				printk(KERN_INFO "can not get irq_desc for %d\n", irq);
+				continue;
+			}
+			cfg = desc->chip_data;
+			add_pin_to_irq_cpu(cfg, cpu, apic, pin);
 
-			setup_IO_APIC_irq(apic, pin, irq,
+			setup_IO_APIC_irq(apic, pin, irq, desc,
 					irq_trigger(idx), irq_polarity(idx));
 		}
 	}
@@ -1448,6 +1692,7 @@ __apicdebuginit(void) print_IO_APIC(void)
 	union IO_APIC_reg_03 reg_03;
 	unsigned long flags;
 	struct irq_cfg *cfg;
+	struct irq_desc *desc;
 	unsigned int irq;
 
 	if (apic_verbosity == APIC_QUIET)
@@ -1537,8 +1782,11 @@ __apicdebuginit(void) print_IO_APIC(void)
 	}
 	}
 	printk(KERN_DEBUG "IRQ to pin mappings:\n");
-	for_each_irq_cfg(irq, cfg) {
-		struct irq_pin_list *entry = cfg->irq_2_pin;
+	for_each_irq_desc(irq, desc) {
+		struct irq_pin_list *entry;
+
+		cfg = desc->chip_data;
+		entry = cfg->irq_2_pin;
 		if (!entry)
 			continue;
 		printk(KERN_DEBUG "IRQ%d ", irq);
@@ -2022,14 +2270,16 @@ static unsigned int startup_ioapic_irq(unsigned int irq)
 {
 	int was_pending = 0;
 	unsigned long flags;
+	struct irq_cfg *cfg;
 
 	spin_lock_irqsave(&ioapic_lock, flags);
-	if (irq < 16) {
+	if (irq < NR_IRQS_LEGACY) {
 		disable_8259A_irq(irq);
 		if (i8259A_irq_pending(irq))
 			was_pending = 1;
 	}
-	__unmask_IO_APIC_irq(irq);
+	cfg = irq_cfg(irq);
+	__unmask_IO_APIC_irq(cfg);
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 
 	return was_pending;
@@ -2043,7 +2293,7 @@ static int ioapic_retrigger_irq(unsigned int irq)
 	unsigned long flags;
 
 	spin_lock_irqsave(&vector_lock, flags);
-	send_IPI_mask(cpumask_of_cpu(first_cpu(cfg->domain)), cfg->vector);
+	send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector);
 	spin_unlock_irqrestore(&vector_lock, flags);
 
 	return 1;
@@ -2092,35 +2342,35 @@ static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration);
  * as simple as edge triggered migration and we can do the irq migration
  * with a simple atomic update to IO-APIC RTE.
  */
-static void migrate_ioapic_irq(int irq, cpumask_t mask)
+static void
+migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
-	struct irq_desc *desc;
-	cpumask_t tmp, cleanup_mask;
 	struct irte irte;
 	int modify_ioapic_rte;
 	unsigned int dest;
 	unsigned long flags;
+	unsigned int irq;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
+	irq = desc->irq;
 	if (get_irte(irq, &irte))
 		return;
 
-	if (assign_irq_vector(irq, mask))
+	cfg = desc->chip_data;
+	if (assign_irq_vector(irq, cfg, mask))
 		return;
 
-	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
+	set_extra_move_desc(desc, mask);
+
+	dest = cpu_mask_to_apicid_and(cfg->domain, mask);
 
-	desc = irq_to_desc(irq);
 	modify_ioapic_rte = desc->status & IRQ_LEVEL;
 	if (modify_ioapic_rte) {
 		spin_lock_irqsave(&ioapic_lock, flags);
-		__target_IO_APIC_irq(irq, dest, cfg->vector);
+		__target_IO_APIC_irq(irq, dest, cfg);
 		spin_unlock_irqrestore(&ioapic_lock, flags);
 	}
 
@@ -2132,24 +2382,20 @@ static void migrate_ioapic_irq(int irq, cpumask_t mask)
 	 */
 	modify_irte(irq, &irte);
 
-	if (cfg->move_in_progress) {
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
+	if (cfg->move_in_progress)
+		send_cleanup_vector(cfg);
 
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 
-static int migrate_irq_remapped_level(int irq)
+static int migrate_irq_remapped_level_desc(struct irq_desc *desc)
 {
 	int ret = -1;
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct irq_cfg *cfg = desc->chip_data;
 
-	mask_IO_APIC_irq(irq);
+	mask_IO_APIC_irq_desc(desc);
 
-	if (io_apic_level_ack_pending(irq)) {
+	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
@@ -2161,14 +2407,15 @@ static int migrate_irq_remapped_level(int irq)
 	}
 
 	/* everthing is clear. we have right of way */
-	migrate_ioapic_irq(irq, desc->pending_mask);
+	migrate_ioapic_irq_desc(desc, &desc->pending_mask);
 
 	ret = 0;
 	desc->status &= ~IRQ_MOVE_PENDING;
-	cpus_clear(desc->pending_mask);
+	cpumask_clear(&desc->pending_mask);
 
 unmask:
-	unmask_IO_APIC_irq(irq);
+	unmask_IO_APIC_irq_desc(desc);
+
 	return ret;
 }
 
@@ -2189,7 +2436,7 @@ static void ir_irq_migration(struct work_struct *work)
 				continue;
 			}
 
-			desc->chip->set_affinity(irq, desc->pending_mask);
+			desc->chip->set_affinity(irq, &desc->pending_mask);
 			spin_unlock_irqrestore(&desc->lock, flags);
 		}
 	}
@@ -2198,18 +2445,24 @@ static void ir_irq_migration(struct work_struct *work)
 /*
  * Migrates the IRQ destination in the process context.
  */
-static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
+					    const struct cpumask *mask)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-
 	if (desc->status & IRQ_LEVEL) {
 		desc->status |= IRQ_MOVE_PENDING;
-		desc->pending_mask = mask;
-		migrate_irq_remapped_level(irq);
+		cpumask_copy(&desc->pending_mask, mask);
+		migrate_irq_remapped_level_desc(desc);
 		return;
 	}
 
-	migrate_ioapic_irq(irq, mask);
+	migrate_ioapic_irq_desc(desc, mask);
+}
+static void set_ir_ioapic_affinity_irq(unsigned int irq,
+				       const struct cpumask *mask)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	set_ir_ioapic_affinity_irq_desc(desc, mask);
 }
 #endif
 
@@ -2228,6 +2481,9 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
 		struct irq_cfg *cfg;
 		irq = __get_cpu_var(vector_irq)[vector];
 
+		if (irq == -1)
+			continue;
+
 		desc = irq_to_desc(irq);
 		if (!desc)
 			continue;
@@ -2237,7 +2493,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
 		if (!cfg->move_cleanup_count)
 			goto unlock;
 
-		if ((vector == cfg->vector) && cpu_isset(me, cfg->domain))
+		if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
 			goto unlock;
 
 		__get_cpu_var(vector_irq)[vector] = -1;
@@ -2249,28 +2505,44 @@ unlock:
 	irq_exit();
 }
 
-static void irq_complete_move(unsigned int irq)
+static void irq_complete_move(struct irq_desc **descp)
 {
-	struct irq_cfg *cfg = irq_cfg(irq);
+	struct irq_desc *desc = *descp;
+	struct irq_cfg *cfg = desc->chip_data;
 	unsigned vector, me;
 
-	if (likely(!cfg->move_in_progress))
+	if (likely(!cfg->move_in_progress)) {
+#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
+		if (likely(!cfg->move_desc_pending))
+			return;
+
+		/* domain has not changed, but affinity did */
+		me = smp_processor_id();
+		if (cpu_isset(me, desc->affinity)) {
+			*descp = desc = move_irq_desc(desc, me);
+			/* get the new one */
+			cfg = desc->chip_data;
+			cfg->move_desc_pending = 0;
+		}
+#endif
 		return;
+	}
 
 	vector = ~get_irq_regs()->orig_ax;
 	me = smp_processor_id();
-	if ((vector == cfg->vector) && cpu_isset(me, cfg->domain)) {
-		cpumask_t cleanup_mask;
+#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
+		*descp = desc = move_irq_desc(desc, me);
+		/* get the new one */
+		cfg = desc->chip_data;
+#endif
 
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
+	if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
+		send_cleanup_vector(cfg);
 }
 #else
-static inline void irq_complete_move(unsigned int irq) {}
+static inline void irq_complete_move(struct irq_desc **descp) {}
 #endif
+
 #ifdef CONFIG_INTR_REMAP
 static void ack_x2apic_level(unsigned int irq)
 {
@@ -2281,11 +2553,14 @@ static void ack_x2apic_edge(unsigned int irq)
 {
 	ack_x2APIC_irq();
 }
+
 #endif
 
 static void ack_apic_edge(unsigned int irq)
 {
-	irq_complete_move(irq);
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	irq_complete_move(&desc);
 	move_native_irq(irq);
 	ack_APIC_irq();
 }
@@ -2294,18 +2569,21 @@ atomic_t irq_mis_count;
 
 static void ack_apic_level(unsigned int irq)
 {
+	struct irq_desc *desc = irq_to_desc(irq);
+
 #ifdef CONFIG_X86_32
 	unsigned long v;
 	int i;
 #endif
+	struct irq_cfg *cfg;
 	int do_unmask_irq = 0;
 
-	irq_complete_move(irq);
+	irq_complete_move(&desc);
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	/* If we are moving the irq we need to mask it */
-	if (unlikely(irq_to_desc(irq)->status & IRQ_MOVE_PENDING)) {
+	if (unlikely(desc->status & IRQ_MOVE_PENDING)) {
 		do_unmask_irq = 1;
-		mask_IO_APIC_irq(irq);
+		mask_IO_APIC_irq_desc(desc);
 	}
 #endif
 
@@ -2329,7 +2607,8 @@ static void ack_apic_level(unsigned int irq)
 	* operation to prevent an edge-triggered interrupt escaping meanwhile.
 	* The idea is from Manfred Spraul.  --macro
 	*/
-	i = irq_cfg(irq)->vector;
+	cfg = desc->chip_data;
+	i = cfg->vector;
 
 	v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
 #endif
@@ -2368,17 +2647,18 @@ static void ack_apic_level(unsigned int irq)
 		 * accurate and is causing problems then it is a hardware bug
 		 * and you can go talk to the chipset vendor about it.
 		 */
-		if (!io_apic_level_ack_pending(irq))
+		cfg = desc->chip_data;
+		if (!io_apic_level_ack_pending(cfg))
 			move_masked_irq(irq);
-		unmask_IO_APIC_irq(irq);
+		unmask_IO_APIC_irq_desc(desc);
 	}
 
 #ifdef CONFIG_X86_32
 	if (!(v & (1 << (i & 0x1f)))) {
 		atomic_inc(&irq_mis_count);
 		spin_lock(&ioapic_lock);
-		__mask_and_edge_IO_APIC_irq(irq);
-		__unmask_and_level_IO_APIC_irq(irq);
+		__mask_and_edge_IO_APIC_irq(cfg);
+		__unmask_and_level_IO_APIC_irq(cfg);
 		spin_unlock(&ioapic_lock);
 	}
 #endif
@@ -2429,20 +2709,19 @@ static inline void init_IO_APIC_traps(void)
 	 * Also, we've got to be careful not to trash gate
 	 * 0x80, because int 0x80 is hm, kind of importantish. ;)
 	 */
-	for_each_irq_cfg(irq, cfg) {
-		if (IO_APIC_IRQ(irq) && !cfg->vector) {
+	for_each_irq_desc(irq, desc) {
+		cfg = desc->chip_data;
+		if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) {
 			/*
 			 * Hmm.. We don't have an entry for this,
 			 * so default to an old-fashioned 8259
 			 * interrupt if we can..
 			 */
-			if (irq < 16)
+			if (irq < NR_IRQS_LEGACY)
 				make_8259A_irq(irq);
-			else {
-				desc = irq_to_desc(irq);
+			else
 				/* Strange. Oh, well.. */
 				desc->chip = &no_irq_chip;
-			}
 		}
 	}
 }
@@ -2467,7 +2746,7 @@ static void unmask_lapic_irq(unsigned int irq)
 	apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
 }
 
-static void ack_lapic_irq (unsigned int irq)
+static void ack_lapic_irq(unsigned int irq)
 {
 	ack_APIC_irq();
 }
@@ -2479,11 +2758,8 @@ static struct irq_chip lapic_chip __read_mostly = {
 	.ack		= ack_lapic_irq,
 };
 
-static void lapic_register_intr(int irq)
+static void lapic_register_intr(int irq, struct irq_desc *desc)
 {
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
 	desc->status &= ~IRQ_LEVEL;
 	set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq,
 				      "edge");
@@ -2587,7 +2863,9 @@ int timer_through_8259 __initdata;
  */
 static inline void __init check_timer(void)
 {
-	struct irq_cfg *cfg = irq_cfg(0);
+	struct irq_desc *desc = irq_to_desc(0);
+	struct irq_cfg *cfg = desc->chip_data;
+	int cpu = boot_cpu_id;
 	int apic1, pin1, apic2, pin2;
 	unsigned long flags;
 	unsigned int ver;
@@ -2602,7 +2880,7 @@ static inline void __init check_timer(void)
 	 * get/set the timer IRQ vector:
 	 */
 	disable_8259A_irq(0);
-	assign_irq_vector(0, TARGET_CPUS);
+	assign_irq_vector(0, cfg, TARGET_CPUS);
 
 	/*
 	 * As IRQ0 is to be enabled in the 8259A, the virtual
@@ -2653,10 +2931,10 @@ static inline void __init check_timer(void)
 		 * Ok, does IRQ0 through the IOAPIC work?
 		 */
 		if (no_pin1) {
-			add_pin_to_irq(0, apic1, pin1);
+			add_pin_to_irq_cpu(cfg, cpu, apic1, pin1);
 			setup_timer_IRQ0_pin(apic1, pin1, cfg->vector);
 		}
-		unmask_IO_APIC_irq(0);
+		unmask_IO_APIC_irq_desc(desc);
 		if (timer_irq_works()) {
 			if (nmi_watchdog == NMI_IO_APIC) {
 				setup_nmi();
@@ -2682,9 +2960,9 @@ static inline void __init check_timer(void)
 		/*
 		 * legacy devices should be connected to IO APIC #0
 		 */
-		replace_pin_at_irq(0, apic1, pin1, apic2, pin2);
+		replace_pin_at_irq_cpu(cfg, cpu, apic1, pin1, apic2, pin2);
 		setup_timer_IRQ0_pin(apic2, pin2, cfg->vector);
-		unmask_IO_APIC_irq(0);
+		unmask_IO_APIC_irq_desc(desc);
 		enable_8259A_irq(0);
 		if (timer_irq_works()) {
 			apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
@@ -2716,7 +2994,7 @@ static inline void __init check_timer(void)
 	apic_printk(APIC_QUIET, KERN_INFO
 		    "...trying to set up timer as Virtual Wire IRQ...\n");
 
-	lapic_register_intr(0);
+	lapic_register_intr(0, desc);
 	apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector);	/* Fixed mode */
 	enable_8259A_irq(0);
 
@@ -2901,22 +3179,26 @@ unsigned int create_irq_nr(unsigned int irq_want)
 	unsigned int irq;
 	unsigned int new;
 	unsigned long flags;
-	struct irq_cfg *cfg_new;
-
-	irq_want = nr_irqs - 1;
+	struct irq_cfg *cfg_new = NULL;
+	int cpu = boot_cpu_id;
+	struct irq_desc *desc_new = NULL;
 
 	irq = 0;
 	spin_lock_irqsave(&vector_lock, flags);
-	for (new = irq_want; new > 0; new--) {
+	for (new = irq_want; new < NR_IRQS; new++) {
 		if (platform_legacy_irq(new))
 			continue;
-		cfg_new = irq_cfg(new);
-		if (cfg_new && cfg_new->vector != 0)
+
+		desc_new = irq_to_desc_alloc_cpu(new, cpu);
+		if (!desc_new) {
+			printk(KERN_INFO "can not get irq_desc for %d\n", new);
 			continue;
-		/* check if need to create one */
-		if (!cfg_new)
-			cfg_new = irq_cfg_alloc(new);
-		if (__assign_irq_vector(new, TARGET_CPUS) == 0)
+		}
+		cfg_new = desc_new->chip_data;
+
+		if (cfg_new->vector != 0)
+			continue;
+		if (__assign_irq_vector(new, cfg_new, TARGET_CPUS) == 0)
 			irq = new;
 		break;
 	}
@@ -2924,15 +3206,21 @@ unsigned int create_irq_nr(unsigned int irq_want)
 
 	if (irq > 0) {
 		dynamic_irq_init(irq);
+		/* restore it, in case dynamic_irq_init clear it */
+		if (desc_new)
+			desc_new->chip_data = cfg_new;
 	}
 	return irq;
 }
 
+static int nr_irqs_gsi = NR_IRQS_LEGACY;
 int create_irq(void)
 {
+	unsigned int irq_want;
 	int irq;
 
-	irq = create_irq_nr(nr_irqs - 1);
+	irq_want = nr_irqs_gsi;
+	irq = create_irq_nr(irq_want);
 
 	if (irq == 0)
 		irq = -1;
@@ -2943,14 +3231,22 @@ int create_irq(void)
 void destroy_irq(unsigned int irq)
 {
 	unsigned long flags;
+	struct irq_cfg *cfg;
+	struct irq_desc *desc;
 
+	/* store it, in case dynamic_irq_cleanup clear it */
+	desc = irq_to_desc(irq);
+	cfg = desc->chip_data;
 	dynamic_irq_cleanup(irq);
+	/* connect back irq_cfg */
+	if (desc)
+		desc->chip_data = cfg;
 
 #ifdef CONFIG_INTR_REMAP
 	free_irte(irq);
 #endif
 	spin_lock_irqsave(&vector_lock, flags);
-	__clear_irq_vector(irq);
+	__clear_irq_vector(irq, cfg);
 	spin_unlock_irqrestore(&vector_lock, flags);
 }
 
@@ -2963,16 +3259,13 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 	struct irq_cfg *cfg;
 	int err;
 	unsigned dest;
-	cpumask_t tmp;
 
-	tmp = TARGET_CPUS;
-	err = assign_irq_vector(irq, tmp);
+	cfg = irq_cfg(irq);
+	err = assign_irq_vector(irq, cfg, TARGET_CPUS);
 	if (err)
 		return err;
 
-	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, tmp);
-	dest = cpu_mask_to_apicid(tmp);
+	dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 #ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
@@ -3026,64 +3319,48 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 }
 
 #ifdef CONFIG_SMP
-static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
+	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
-	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
-	if (assign_irq_vector(irq, mask))
-		return;
+	cfg = desc->chip_data;
 
-	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
-
-	read_msi_msg(irq, &msg);
+	read_msi_msg_desc(desc, &msg);
 
 	msg.data &= ~MSI_DATA_VECTOR_MASK;
 	msg.data |= MSI_DATA_VECTOR(cfg->vector);
 	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
-	write_msi_msg(irq, &msg);
-	desc = irq_to_desc(irq);
-	desc->affinity = mask;
+	write_msi_msg_desc(desc, &msg);
 }
-
 #ifdef CONFIG_INTR_REMAP
 /*
  * Migrate the MSI irq to another cpumask. This migration is
  * done in the process context using interrupt-remapping hardware.
  */
-static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void
+ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
-	struct irq_cfg *cfg;
+	struct irq_desc *desc = irq_to_desc(irq);
+	struct irq_cfg *cfg = desc->chip_data;
 	unsigned int dest;
-	cpumask_t tmp, cleanup_mask;
 	struct irte irte;
-	struct irq_desc *desc;
-
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
-		return;
 
 	if (get_irte(irq, &irte))
 		return;
 
-	if (assign_irq_vector(irq, mask))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
-	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
-
 	irte.vector = cfg->vector;
 	irte.dest_id = IRTE_DEST(dest);
 
@@ -3097,16 +3374,10 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 	 * at the new destination. So, time to cleanup the previous
 	 * vector allocation.
 	 */
-	if (cfg->move_in_progress) {
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
-
-	desc = irq_to_desc(irq);
-	desc->affinity = mask;
+	if (cfg->move_in_progress)
+		send_cleanup_vector(cfg);
 }
+
 #endif
 #endif /* CONFIG_SMP */
 
@@ -3165,7 +3436,7 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec)
 }
 #endif
 
-static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq)
+static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 {
 	int ret;
 	struct msi_msg msg;
@@ -3174,7 +3445,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq)
 	if (ret < 0)
 		return ret;
 
-	set_irq_msi(irq, desc);
+	set_irq_msi(irq, msidesc);
 	write_msi_msg(irq, &msg);
 
 #ifdef CONFIG_INTR_REMAP
@@ -3194,26 +3465,13 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq)
 	return 0;
 }
 
-static unsigned int build_irq_for_pci_dev(struct pci_dev *dev)
-{
-	unsigned int irq;
-
-	irq = dev->bus->number;
-	irq <<= 8;
-	irq |= dev->devfn;
-	irq <<= 12;
-
-	return irq;
-}
-
-int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
+int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc)
 {
 	unsigned int irq;
 	int ret;
 	unsigned int irq_want;
 
-	irq_want = build_irq_for_pci_dev(dev) + 0x100;
-
+	irq_want = nr_irqs_gsi;
 	irq = create_irq_nr(irq_want);
 	if (irq == 0)
 		return -1;
@@ -3227,7 +3485,7 @@ int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
 		goto error;
 no_ir:
 #endif
-	ret = setup_msi_irq(dev, desc, irq);
+	ret = setup_msi_irq(dev, msidesc, irq);
 	if (ret < 0) {
 		destroy_irq(irq);
 		return ret;
@@ -3245,7 +3503,7 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 {
 	unsigned int irq;
 	int ret, sub_handle;
-	struct msi_desc *desc;
+	struct msi_desc *msidesc;
 	unsigned int irq_want;
 
 #ifdef CONFIG_INTR_REMAP
@@ -3253,10 +3511,11 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 	int index = 0;
 #endif
 
-	irq_want = build_irq_for_pci_dev(dev) + 0x100;
+	irq_want = nr_irqs_gsi;
 	sub_handle = 0;
-	list_for_each_entry(desc, &dev->msi_list, list) {
-		irq = create_irq_nr(irq_want--);
+	list_for_each_entry(msidesc, &dev->msi_list, list) {
+		irq = create_irq_nr(irq_want);
+		irq_want++;
 		if (irq == 0)
 			return -1;
 #ifdef CONFIG_INTR_REMAP
@@ -3288,7 +3547,7 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 		}
 no_ir:
 #endif
-		ret = setup_msi_irq(dev, desc, irq);
+		ret = setup_msi_irq(dev, msidesc, irq);
 		if (ret < 0)
 			goto error;
 		sub_handle++;
@@ -3307,24 +3566,18 @@ void arch_teardown_msi_irq(unsigned int irq)
 
 #ifdef CONFIG_DMAR
 #ifdef CONFIG_SMP
-static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
+	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
-	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
-	if (assign_irq_vector(irq, mask))
-		return;
-
-	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
+	cfg = desc->chip_data;
 
 	dmar_msi_read(irq, &msg);
 
@@ -3334,9 +3587,8 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	dmar_msi_write(irq, &msg);
-	desc = irq_to_desc(irq);
-	desc->affinity = mask;
 }
+
 #endif /* CONFIG_SMP */
 
 struct irq_chip dmar_msi_type = {
@@ -3368,24 +3620,18 @@ int arch_setup_dmar_msi(unsigned int irq)
 #ifdef CONFIG_HPET_TIMER
 
 #ifdef CONFIG_SMP
-static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
+	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
-	struct irq_desc *desc;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
-	if (assign_irq_vector(irq, mask))
-		return;
-
-	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
+	cfg = desc->chip_data;
 
 	hpet_msi_read(irq, &msg);
 
@@ -3395,9 +3641,8 @@ static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	hpet_msi_write(irq, &msg);
-	desc = irq_to_desc(irq);
-	desc->affinity = mask;
 }
+
 #endif /* CONFIG_SMP */
 
 struct irq_chip hpet_msi_type = {
@@ -3450,28 +3695,21 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector)
 	write_ht_irq_msg(irq, &msg);
 }
 
-static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
+	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	unsigned int dest;
-	cpumask_t tmp;
-	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
-	if (assign_irq_vector(irq, mask))
-		return;
-
-	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
+	cfg = desc->chip_data;
 
 	target_ht_irq(irq, dest, cfg->vector);
-	desc = irq_to_desc(irq);
-	desc->affinity = mask;
 }
+
 #endif
 
 static struct irq_chip ht_irq_chip = {
@@ -3489,17 +3727,14 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
 {
 	struct irq_cfg *cfg;
 	int err;
-	cpumask_t tmp;
 
-	tmp = TARGET_CPUS;
-	err = assign_irq_vector(irq, tmp);
+	cfg = irq_cfg(irq);
+	err = assign_irq_vector(irq, cfg, TARGET_CPUS);
 	if (!err) {
 		struct ht_irq_msg msg;
 		unsigned dest;
 
-		cfg = irq_cfg(irq);
-		cpus_and(tmp, cfg->domain, tmp);
-		dest = cpu_mask_to_apicid(tmp);
+		dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 		msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest);
 
@@ -3535,7 +3770,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
 int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 		       unsigned long mmr_offset)
 {
-	const cpumask_t *eligible_cpu = get_cpu_mask(cpu);
+	const struct cpumask *eligible_cpu = cpumask_of(cpu);
 	struct irq_cfg *cfg;
 	int mmr_pnode;
 	unsigned long mmr_value;
@@ -3543,7 +3778,9 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 	unsigned long flags;
 	int err;
 
-	err = assign_irq_vector(irq, *eligible_cpu);
+	cfg = irq_cfg(irq);
+
+	err = assign_irq_vector(irq, cfg, eligible_cpu);
 	if (err != 0)
 		return err;
 
@@ -3552,8 +3789,6 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 				      irq_name);
 	spin_unlock_irqrestore(&vector_lock, flags);
 
-	cfg = irq_cfg(irq);
-
 	mmr_value = 0;
 	entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
 	BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long));
@@ -3564,7 +3799,7 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 	entry->polarity = 0;
 	entry->trigger = 0;
 	entry->mask = 0;
-	entry->dest = cpu_mask_to_apicid(*eligible_cpu);
+	entry->dest = cpu_mask_to_apicid(eligible_cpu);
 
 	mmr_pnode = uv_blade_to_pnode(mmr_blade);
 	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
@@ -3605,9 +3840,16 @@ int __init io_apic_get_redir_entries (int ioapic)
 	return reg_01.bits.entries;
 }
 
-int __init probe_nr_irqs(void)
+void __init probe_nr_irqs_gsi(void)
 {
-	return NR_IRQS;
+	int idx;
+	int nr = 0;
+
+	for (idx = 0; idx < nr_ioapics; idx++)
+		nr += io_apic_get_redir_entries(idx) + 1;
+
+	if (nr > nr_irqs_gsi)
+		nr_irqs_gsi = nr;
 }
 
 /* --------------------------------------------------------------------------
@@ -3706,19 +3948,31 @@ int __init io_apic_get_version(int ioapic)
 
 int io_apic_set_pci_routing (int ioapic, int pin, int irq, int triggering, int polarity)
 {
+	struct irq_desc *desc;
+	struct irq_cfg *cfg;
+	int cpu = boot_cpu_id;
+
 	if (!IO_APIC_IRQ(irq)) {
 		apic_printk(APIC_QUIET,KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n",
 			ioapic);
 		return -EINVAL;
 	}
 
+	desc = irq_to_desc_alloc_cpu(irq, cpu);
+	if (!desc) {
+		printk(KERN_INFO "can not get irq_desc %d\n", irq);
+		return 0;
+	}
+
 	/*
 	 * IRQs < 16 are already in the irq_2_pin[] map
 	 */
-	if (irq >= 16)
-		add_pin_to_irq(irq, ioapic, pin);
+	if (irq >= NR_IRQS_LEGACY) {
+		cfg = desc->chip_data;
+		add_pin_to_irq_cpu(cfg, cpu, ioapic, pin);
+	}
 
-	setup_IO_APIC_irq(ioapic, pin, irq, triggering, polarity);
+	setup_IO_APIC_irq(ioapic, pin, irq, desc, triggering, polarity);
 
 	return 0;
 }
@@ -3756,7 +4010,7 @@ void __init setup_ioapic_dest(void)
 	int pin, ioapic, irq, irq_entry;
 	struct irq_desc *desc;
 	struct irq_cfg *cfg;
-	cpumask_t mask;
+	const struct cpumask *mask;
 
 	if (skip_ioapic_setup == 1)
 		return;
@@ -3772,9 +4026,10 @@ void __init setup_ioapic_dest(void)
 			 * when you have too many devices, because at that time only boot
 			 * cpu is online.
 			 */
-			cfg = irq_cfg(irq);
+			desc = irq_to_desc(irq);
+			cfg = desc->chip_data;
 			if (!cfg->vector) {
-				setup_IO_APIC_irq(ioapic, pin, irq,
+				setup_IO_APIC_irq(ioapic, pin, irq, desc,
 						  irq_trigger(irq_entry),
 						  irq_polarity(irq_entry));
 				continue;
@@ -3784,19 +4039,18 @@ void __init setup_ioapic_dest(void)
 			/*
 			 * Honour affinities which have been set in early boot
 			 */
-			desc = irq_to_desc(irq);
 			if (desc->status &
 			    (IRQ_NO_BALANCING | IRQ_AFFINITY_SET))
-				mask = desc->affinity;
+				mask = &desc->affinity;
 			else
 				mask = TARGET_CPUS;
 
 #ifdef CONFIG_INTR_REMAP
 			if (intr_remapping_enabled)
-				set_ir_ioapic_affinity_irq(irq, mask);
+				set_ir_ioapic_affinity_irq_desc(desc, mask);
 			else
 #endif
-				set_ioapic_affinity_irq(irq, mask);
+				set_ioapic_affinity_irq_desc(desc, mask);
 		}
 
 	}
@@ -3845,7 +4099,6 @@ void __init ioapic_init_mappings(void)
 	struct resource *ioapic_res;
 	int i;
 
-	irq_2_pin_init();
 	ioapic_res = ioapic_setup_resources();
 	for (i = 0; i < nr_ioapics; i++) {
 		if (smp_found_config) {
diff --git a/arch/x86/kernel/ipi.c b/arch/x86/kernel/ipi.c
index f1c688e46f3..285bbf8831f 100644
--- a/arch/x86/kernel/ipi.c
+++ b/arch/x86/kernel/ipi.c
@@ -116,18 +116,18 @@ static inline void __send_IPI_dest_field(unsigned long mask, int vector)
 /*
  * This is only used on smaller machines.
  */
-void send_IPI_mask_bitmask(cpumask_t cpumask, int vector)
+void send_IPI_mask_bitmask(const struct cpumask *cpumask, int vector)
 {
-	unsigned long mask = cpus_addr(cpumask)[0];
+	unsigned long mask = cpumask_bits(cpumask)[0];
 	unsigned long flags;
 
 	local_irq_save(flags);
-	WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]);
+	WARN_ON(mask & ~cpumask_bits(cpu_online_mask)[0]);
 	__send_IPI_dest_field(mask, vector);
 	local_irq_restore(flags);
 }
 
-void send_IPI_mask_sequence(cpumask_t mask, int vector)
+void send_IPI_mask_sequence(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned int query_cpu;
@@ -139,12 +139,24 @@ void send_IPI_mask_sequence(cpumask_t mask, int vector)
 	 */
 
 	local_irq_save(flags);
-	for_each_possible_cpu(query_cpu) {
-		if (cpu_isset(query_cpu, mask)) {
+	for_each_cpu(query_cpu, mask)
+		__send_IPI_dest_field(cpu_to_logical_apicid(query_cpu), vector);
+	local_irq_restore(flags);
+}
+
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+	unsigned long flags;
+	unsigned int query_cpu;
+	unsigned int this_cpu = smp_processor_id();
+
+	/* See Hack comment above */
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask)
+		if (query_cpu != this_cpu)
 			__send_IPI_dest_field(cpu_to_logical_apicid(query_cpu),
 					      vector);
-		}
-	}
 	local_irq_restore(flags);
 }
 
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index d1d4dc52f64..bce53e1352a 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -9,6 +9,7 @@
 #include <asm/apic.h>
 #include <asm/io_apic.h>
 #include <asm/smp.h>
+#include <asm/irq.h>
 
 atomic_t irq_err_count;
 
@@ -118,6 +119,9 @@ int show_interrupts(struct seq_file *p, void *v)
 	}
 
 	desc = irq_to_desc(i);
+	if (!desc)
+		return 0;
+
 	spin_lock_irqsave(&desc->lock, flags);
 #ifndef CONFIG_SMP
 	any_count = kstat_irqs(i);
@@ -187,3 +191,5 @@ u64 arch_irq_stat(void)
 #endif
 	return sum;
 }
+
+EXPORT_SYMBOL_GPL(vector_used_by_percpu_irq);
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index a51382672de..9dc5588f336 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -233,25 +233,28 @@ unsigned int do_IRQ(struct pt_regs *regs)
 #ifdef CONFIG_HOTPLUG_CPU
 #include <mach_apic.h>
 
-void fixup_irqs(cpumask_t map)
+/* A cpu has been removed from cpu_online_mask.  Reset irq affinities. */
+void fixup_irqs(void)
 {
 	unsigned int irq;
 	static int warned;
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		cpumask_t mask;
+		const struct cpumask *affinity;
 
+		if (!desc)
+			continue;
 		if (irq == 2)
 			continue;
 
-		cpus_and(mask, desc->affinity, map);
-		if (any_online_cpu(mask) == NR_CPUS) {
+		affinity = &desc->affinity;
+		if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
 			printk("Breaking affinity for irq %i\n", irq);
-			mask = map;
+			affinity = cpu_all_mask;
 		}
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, affinity);
 		else if (desc->action && !(warned++))
 			printk("Cannot set affinity for irq %i\n", irq);
 	}
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index 1df869e5bd0..6383d50f82e 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -80,40 +80,43 @@ asmlinkage unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-void fixup_irqs(cpumask_t map)
+/* A cpu has been removed from cpu_online_mask.  Reset irq affinities. */
+void fixup_irqs(void)
 {
 	unsigned int irq;
 	static int warned;
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		cpumask_t mask;
 		int break_affinity = 0;
 		int set_affinity = 1;
+		const struct cpumask *affinity;
 
+		if (!desc)
+			continue;
 		if (irq == 2)
 			continue;
 
 		/* interrupt's are disabled at this point */
 		spin_lock(&desc->lock);
 
+		affinity = &desc->affinity;
 		if (!irq_has_action(irq) ||
-		    cpus_equal(desc->affinity, map)) {
+		    cpumask_equal(affinity, cpu_online_mask)) {
 			spin_unlock(&desc->lock);
 			continue;
 		}
 
-		cpus_and(mask, desc->affinity, map);
-		if (cpus_empty(mask)) {
+		if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
 			break_affinity = 1;
-			mask = map;
+			affinity = cpu_all_mask;
 		}
 
 		if (desc->chip->mask)
 			desc->chip->mask(irq);
 
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, affinity);
 		else if (!(warned++))
 			set_affinity = 0;
 
diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c
index 607db63044a..84723295f88 100644
--- a/arch/x86/kernel/irqinit_32.c
+++ b/arch/x86/kernel/irqinit_32.c
@@ -68,8 +68,7 @@ void __init init_ISA_irqs (void)
 	/*
 	 * 16 old-style INTA-cycle interrupts:
 	 */
-	for (i = 0; i < 16; i++) {
-		/* first time call this irq_desc */
+	for (i = 0; i < NR_IRQS_LEGACY; i++) {
 		struct irq_desc *desc = irq_to_desc(i);
 
 		desc->status = IRQ_DISABLED;
@@ -111,6 +110,18 @@ DEFINE_PER_CPU(vector_irq_t, vector_irq) = {
 	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
 };
 
+int vector_used_by_percpu_irq(unsigned int vector)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		if (per_cpu(vector_irq, cpu)[vector] != -1)
+			return 1;
+	}
+
+	return 0;
+}
+
 /* Overridden in paravirt.c */
 void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ")));
 
@@ -147,10 +158,12 @@ void __init native_init_IRQ(void)
 	alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
 
 	/* IPI for single call function */
-	set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, call_function_single_interrupt);
+	alloc_intr_gate(CALL_FUNCTION_SINGLE_VECTOR,
+				 call_function_single_interrupt);
 
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
+	set_bit(IRQ_MOVE_CLEANUP_VECTOR, used_vectors);
 #endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 8670b3ce626..31ebfe38e96 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -69,6 +69,18 @@ DEFINE_PER_CPU(vector_irq_t, vector_irq) = {
 	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
 };
 
+int vector_used_by_percpu_irq(unsigned int vector)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		if (per_cpu(vector_irq, cpu)[vector] != -1)
+			return 1;
+	}
+
+	return 0;
+}
+
 void __init init_ISA_irqs(void)
 {
 	int i;
@@ -76,8 +88,7 @@ void __init init_ISA_irqs(void)
 	init_bsp_APIC();
 	init_8259A(0);
 
-	for (i = 0; i < 16; i++) {
-		/* first time call this irq_desc */
+	for (i = 0; i < NR_IRQS_LEGACY; i++) {
 		struct irq_desc *desc = irq_to_desc(i);
 
 		desc->status = IRQ_DISABLED;
@@ -122,6 +133,7 @@ static void __init smp_intr_init(void)
 
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
+	set_bit(IRQ_MOVE_CLEANUP_VECTOR, used_vectors);
 #endif
 }
 
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index e169ae9b6a6..652fce6d2cc 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -89,17 +89,17 @@ static cycle_t kvm_clock_read(void)
  */
 static unsigned long kvm_get_tsc_khz(void)
 {
-	return preset_lpj;
+	struct pvclock_vcpu_time_info *src;
+	src = &per_cpu(hv_clock, 0);
+	return pvclock_tsc_khz(src);
 }
 
 static void kvm_get_preset_lpj(void)
 {
-	struct pvclock_vcpu_time_info *src;
 	unsigned long khz;
 	u64 lpj;
 
-	src = &per_cpu(hv_clock, 0);
-	khz = pvclock_tsc_khz(src);
+	khz = kvm_get_tsc_khz();
 
 	lpj = ((u64)khz * 1000);
 	do_div(lpj, HZ);
@@ -194,5 +194,7 @@ void __init kvmclock_init(void)
 #endif
 		kvm_get_preset_lpj();
 		clocksource_register(&kvm_clock);
+		pv_info.paravirt_enabled = 1;
+		pv_info.name = "KVM";
 	}
 }
diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c
index 3b599518c32..c12314c9e86 100644
--- a/arch/x86/kernel/mfgpt_32.c
+++ b/arch/x86/kernel/mfgpt_32.c
@@ -287,7 +287,7 @@ static struct clock_event_device mfgpt_clockevent = {
 	.set_mode = mfgpt_set_mode,
 	.set_next_event = mfgpt_next_event,
 	.rating = 250,
-	.cpumask = CPU_MASK_ALL,
+	.cpumask = cpu_all_mask,
 	.shift = 32
 };
 
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 7a3dfceb90e..19a1044a0cd 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -101,11 +101,15 @@ static void __init dma32_free_bootmem(void)
 	dma32_bootmem_ptr = NULL;
 	dma32_bootmem_size = 0;
 }
+#endif
 
 void __init pci_iommu_alloc(void)
 {
+#ifdef CONFIG_X86_64
 	/* free the range so iommu could get some range less than 4G */
 	dma32_free_bootmem();
+#endif
+
 	/*
 	 * The order of these functions is important for
 	 * fall-back/fail-over reasons
@@ -121,15 +125,6 @@ void __init pci_iommu_alloc(void)
 	pci_swiotlb_init();
 }
 
-unsigned long iommu_nr_pages(unsigned long addr, unsigned long len)
-{
-	unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE);
-
-	return size >> PAGE_SHIFT;
-}
-EXPORT_SYMBOL(iommu_nr_pages);
-#endif
-
 void *dma_generic_alloc_coherent(struct device *dev, size_t size,
 				 dma_addr_t *dma_addr, gfp_t flag)
 {
diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index 3c539d111ab..242c3440687 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -3,6 +3,8 @@
 #include <linux/pci.h>
 #include <linux/cache.h>
 #include <linux/module.h>
+#include <linux/swiotlb.h>
+#include <linux/bootmem.h>
 #include <linux/dma-mapping.h>
 
 #include <asm/iommu.h>
@@ -11,6 +13,31 @@
 
 int swiotlb __read_mostly;
 
+void *swiotlb_alloc_boot(size_t size, unsigned long nslabs)
+{
+	return alloc_bootmem_low_pages(size);
+}
+
+void *swiotlb_alloc(unsigned order, unsigned long nslabs)
+{
+	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
+}
+
+dma_addr_t swiotlb_phys_to_bus(phys_addr_t paddr)
+{
+	return paddr;
+}
+
+phys_addr_t swiotlb_bus_to_phys(dma_addr_t baddr)
+{
+	return baddr;
+}
+
+int __weak swiotlb_arch_range_needs_mapping(void *ptr, size_t size)
+{
+	return 0;
+}
+
 static dma_addr_t
 swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size,
 			int direction)
@@ -50,8 +77,10 @@ struct dma_mapping_ops swiotlb_dma_ops = {
 void __init pci_swiotlb_init(void)
 {
 	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
+#ifdef CONFIG_X86_64
 	if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN)
 	       swiotlb = 1;
+#endif
 	if (swiotlb_force)
 		swiotlb = 1;
 	if (swiotlb) {
diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c
index 67465ed8931..309949e9e1c 100644
--- a/arch/x86/kernel/quirks.c
+++ b/arch/x86/kernel/quirks.c
@@ -168,6 +168,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_31,
 			 ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_1,
 			 ich_force_enable_hpet);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_4,
+			 ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH9_7,
 			 ich_force_enable_hpet);
 
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index a90913cccfb..bf088c61fa4 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -13,6 +13,7 @@
 #include <asm/reboot_fixups.h>
 #include <asm/reboot.h>
 #include <asm/pci_x86.h>
+#include <asm/virtext.h>
 
 #ifdef CONFIG_X86_32
 # include <linux/dmi.h>
@@ -39,6 +40,12 @@ int reboot_force;
 static int reboot_cpu = -1;
 #endif
 
+/* This is set if we need to go through the 'emergency' path.
+ * When machine_emergency_restart() is called, we may be on
+ * an inconsistent state and won't be able to do a clean cleanup
+ */
+static int reboot_emergency;
+
 /* This is set by the PCI code if either type 1 or type 2 PCI is detected */
 bool port_cf9_safe = false;
 
@@ -368,6 +375,48 @@ static inline void kb_wait(void)
 	}
 }
 
+static void vmxoff_nmi(int cpu, struct die_args *args)
+{
+	cpu_emergency_vmxoff();
+}
+
+/* Use NMIs as IPIs to tell all CPUs to disable virtualization
+ */
+static void emergency_vmx_disable_all(void)
+{
+	/* Just make sure we won't change CPUs while doing this */
+	local_irq_disable();
+
+	/* We need to disable VMX on all CPUs before rebooting, otherwise
+	 * we risk hanging up the machine, because the CPU ignore INIT
+	 * signals when VMX is enabled.
+	 *
+	 * We can't take any locks and we may be on an inconsistent
+	 * state, so we use NMIs as IPIs to tell the other CPUs to disable
+	 * VMX and halt.
+	 *
+	 * For safety, we will avoid running the nmi_shootdown_cpus()
+	 * stuff unnecessarily, but we don't have a way to check
+	 * if other CPUs have VMX enabled. So we will call it only if the
+	 * CPU we are running on has VMX enabled.
+	 *
+	 * We will miss cases where VMX is not enabled on all CPUs. This
+	 * shouldn't do much harm because KVM always enable VMX on all
+	 * CPUs anyway. But we can miss it on the small window where KVM
+	 * is still enabling VMX.
+	 */
+	if (cpu_has_vmx() && cpu_vmx_enabled()) {
+		/* Disable VMX on this CPU.
+		 */
+		cpu_vmxoff();
+
+		/* Halt and disable VMX on the other CPUs */
+		nmi_shootdown_cpus(vmxoff_nmi);
+
+	}
+}
+
+
 void __attribute__((weak)) mach_reboot_fixups(void)
 {
 }
@@ -376,6 +425,9 @@ static void native_machine_emergency_restart(void)
 {
 	int i;
 
+	if (reboot_emergency)
+		emergency_vmx_disable_all();
+
 	/* Tell the BIOS if we want cold or warm reboot */
 	*((unsigned short *)__va(0x472)) = reboot_mode;
 
@@ -482,13 +534,19 @@ void native_machine_shutdown(void)
 #endif
 }
 
+static void __machine_emergency_restart(int emergency)
+{
+	reboot_emergency = emergency;
+	machine_ops.emergency_restart();
+}
+
 static void native_machine_restart(char *__unused)
 {
 	printk("machine restart\n");
 
 	if (!reboot_force)
 		machine_shutdown();
-	machine_emergency_restart();
+	__machine_emergency_restart(0);
 }
 
 static void native_machine_halt(void)
@@ -532,7 +590,7 @@ void machine_shutdown(void)
 
 void machine_emergency_restart(void)
 {
-	machine_ops.emergency_restart();
+	__machine_emergency_restart(1);
 }
 
 void machine_restart(char *cmd)
@@ -592,10 +650,7 @@ static int crash_nmi_callback(struct notifier_block *self,
 
 static void smp_send_nmi_allbutself(void)
 {
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(safe_smp_processor_id(), mask);
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, NMI_VECTOR);
+	send_IPI_allbutself(NMI_VECTOR);
 }
 
 static struct notifier_block crash_nmi_nb = {
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 08e02e8453c..ae0d8042cf6 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -953,7 +953,7 @@ void __init setup_arch(char **cmdline_p)
 	ioapic_init_mappings();
 
 	/* need to wait for io_apic is mapped */
-	nr_irqs = probe_nr_irqs();
+	probe_nr_irqs_gsi();
 
 	kvm_guest_init();
 
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index ae0c0d3bb77..0b63b08e753 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -152,6 +152,11 @@ void __init setup_per_cpu_areas(void)
 	old_size = PERCPU_ENOUGH_ROOM;
 	align = max_t(unsigned long, PAGE_SIZE, align);
 	size = roundup(old_size, align);
+
+	printk(KERN_INFO
+		"NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n",
+		NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids);
+
 	printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
 			  size);
 
@@ -168,24 +173,24 @@ void __init setup_per_cpu_areas(void)
 			       "cpu %d has no node %d or node-local memory\n",
 				cpu, node);
 			if (ptr)
-				printk(KERN_DEBUG "per cpu data for cpu%d at %016lx\n",
+				printk(KERN_DEBUG
+					"per cpu data for cpu%d at %016lx\n",
 					 cpu, __pa(ptr));
 		}
 		else {
 			ptr = __alloc_bootmem_node(NODE_DATA(node), size, align,
 							__pa(MAX_DMA_ADDRESS));
 			if (ptr)
-				printk(KERN_DEBUG "per cpu data for cpu%d on node%d at %016lx\n",
-					 cpu, node, __pa(ptr));
+				printk(KERN_DEBUG
+					"per cpu data for cpu%d on node%d "
+					"at %016lx\n",
+					cpu, node, __pa(ptr));
 		}
 #endif
 		per_cpu_offset(cpu) = ptr - __per_cpu_start;
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 	}
 
-	printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d, nr_node_ids %d\n",
-		NR_CPUS, nr_cpu_ids, nr_node_ids);
-
 	/* Setup percpu data maps */
 	setup_per_cpu_maps();
 
@@ -282,7 +287,7 @@ static void __cpuinit numa_set_cpumask(int cpu, int enable)
 	else
 		cpu_clear(cpu, *mask);
 
-	cpulist_scnprintf(buf, sizeof(buf), *mask);
+	cpulist_scnprintf(buf, sizeof(buf), mask);
 	printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n",
 		enable? "numa_add_cpu":"numa_remove_cpu", cpu, node, buf);
  }
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 7e558db362c..beea2649a24 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -118,22 +118,22 @@ static void native_smp_send_reschedule(int cpu)
 		WARN_ON(1);
 		return;
 	}
-	send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
+	send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR);
 }
 
 void native_send_call_func_single_ipi(int cpu)
 {
-	send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNCTION_SINGLE_VECTOR);
+	send_IPI_mask(cpumask_of(cpu), CALL_FUNCTION_SINGLE_VECTOR);
 }
 
-void native_send_call_func_ipi(cpumask_t mask)
+void native_send_call_func_ipi(const struct cpumask *mask)
 {
 	cpumask_t allbutself;
 
 	allbutself = cpu_online_map;
 	cpu_clear(smp_processor_id(), allbutself);
 
-	if (cpus_equal(mask, allbutself) &&
+	if (cpus_equal(*mask, allbutself) &&
 	    cpus_equal(cpu_online_map, cpu_callout_map))
 		send_IPI_allbutself(CALL_FUNCTION_VECTOR);
 	else
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index f8500c96944..31869bf5fab 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -102,14 +102,8 @@ EXPORT_SYMBOL(smp_num_siblings);
 /* Last level cache ID of each logical CPU */
 DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
 
-/* bitmap of online cpus */
-cpumask_t cpu_online_map __read_mostly;
-EXPORT_SYMBOL(cpu_online_map);
-
 cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
 
 /* representing HT siblings of each logical CPU */
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map);
@@ -1260,6 +1254,15 @@ void __init native_smp_cpus_done(unsigned int max_cpus)
 	check_nmi_watchdog();
 }
 
+static int __initdata setup_possible_cpus = -1;
+static int __init _setup_possible_cpus(char *str)
+{
+	get_option(&str, &setup_possible_cpus);
+	return 0;
+}
+early_param("possible_cpus", _setup_possible_cpus);
+
+
 /*
  * cpu_possible_map should be static, it cannot change as cpu's
  * are onlined, or offlined. The reason is per-cpu data-structures
@@ -1272,7 +1275,7 @@ void __init native_smp_cpus_done(unsigned int max_cpus)
  *
  * Three ways to find out the number of additional hotplug CPUs:
  * - If the BIOS specified disabled CPUs in ACPI/mptables use that.
- * - The user can overwrite it with additional_cpus=NUM
+ * - The user can overwrite it with possible_cpus=NUM
  * - Otherwise don't reserve additional CPUs.
  * We do this because additional CPUs waste a lot of memory.
  * -AK
@@ -1285,9 +1288,17 @@ __init void prefill_possible_map(void)
 	if (!num_processors)
 		num_processors = 1;
 
-	possible = num_processors + disabled_cpus;
-	if (possible > NR_CPUS)
-		possible = NR_CPUS;
+	if (setup_possible_cpus == -1)
+		possible = num_processors + disabled_cpus;
+	else
+		possible = setup_possible_cpus;
+
+	if (possible > CONFIG_NR_CPUS) {
+		printk(KERN_WARNING
+			"%d Processors exceeds NR_CPUS limit of %d\n",
+			possible, CONFIG_NR_CPUS);
+		possible = CONFIG_NR_CPUS;
+	}
 
 	printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n",
 		possible, max_t(int, possible - num_processors, 0));
@@ -1352,7 +1363,7 @@ void cpu_disable_common(void)
 	lock_vector_lock();
 	remove_cpu_from_maps(cpu);
 	unlock_vector_lock();
-	fixup_irqs(cpu_online_map);
+	fixup_irqs();
 }
 
 int native_cpu_disable(void)
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index 8da059f949b..ce505464224 100644
--- a/arch/x86/kernel/tlb_32.c
+++ b/arch/x86/kernel/tlb_32.c
@@ -163,7 +163,7 @@ void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
 	 * We have to send the IPI only to
 	 * CPUs affected.
 	 */
-	send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR);
+	send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR);
 
 	while (!cpus_empty(flush_cpumask))
 		/* nothing. lockup detection does not belong here */
diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c
index 29887d7081a..f8be6f1d2e4 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -191,7 +191,7 @@ void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
 	 * We have to send the IPI only to
 	 * CPUs affected.
 	 */
-	send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR_START + sender);
+	send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR_START + sender);
 
 	while (!cpus_empty(f->flush_cpumask))
 		cpu_relax();
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 961e26a69d5..ce6650eb64e 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -72,9 +72,6 @@
 
 #include "cpu/mcheck/mce.h"
 
-DECLARE_BITMAP(used_vectors, NR_VECTORS);
-EXPORT_SYMBOL_GPL(used_vectors);
-
 asmlinkage int system_call(void);
 
 /* Do we ignore FPU interrupts ? */
@@ -89,6 +86,9 @@ gate_desc idt_table[256]
 	__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
 #endif
 
+DECLARE_BITMAP(used_vectors, NR_VECTORS);
+EXPORT_SYMBOL_GPL(used_vectors);
+
 static int ignore_nmis;
 
 static inline void conditional_sti(struct pt_regs *regs)
@@ -946,9 +946,7 @@ dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code)
 
 void __init trap_init(void)
 {
-#ifdef CONFIG_X86_32
 	int i;
-#endif
 
 #ifdef CONFIG_EISA
 	void __iomem *p = early_ioremap(0x0FFFD9, 4);
@@ -1005,11 +1003,15 @@ void __init trap_init(void)
 	}
 
 	set_system_trap_gate(SYSCALL_VECTOR, &system_call);
+#endif
 
 	/* Reserve all the builtin and the syscall vector: */
 	for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++)
 		set_bit(i, used_vectors);
 
+#ifdef CONFIG_X86_64
+	set_bit(IA32_SYSCALL_VECTOR, used_vectors);
+#else
 	set_bit(SYSCALL_VECTOR, used_vectors);
 #endif
 	/*
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
index 254ee07f863..c4c1f9e0940 100644
--- a/arch/x86/kernel/vmiclock_32.c
+++ b/arch/x86/kernel/vmiclock_32.c
@@ -226,7 +226,7 @@ static void __devinit vmi_time_init_clockevent(void)
 	/* Upper bound is clockevent's use of ulong for cycle deltas. */
 	evt->max_delta_ns = clockevent_delta2ns(ULONG_MAX, evt);
 	evt->min_delta_ns = clockevent_delta2ns(1, evt);
-	evt->cpumask = cpumask_of_cpu(cpu);
+	evt->cpumask = cpumask_of(cpu);
 
 	printk(KERN_WARNING "vmi: registering clock event %s. mult=%lu shift=%u\n",
 	       evt->name, evt->mult, evt->shift);