diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c
index c2502eb9aa83..9220cf46aa10 100644
--- a/arch/x86/kernel/acpi/cstate.c
+++ b/arch/x86/kernel/acpi/cstate.c
@@ -73,6 +73,7 @@ int acpi_processor_ffh_cstate_probe(unsigned int cpu,
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr(new_mask, cpu);
 	int retval;
 	unsigned int eax, ebx, ecx, edx;
 	unsigned int edx_part;
@@ -91,7 +92,7 @@ int acpi_processor_ffh_cstate_probe(unsigned int cpu,
 
 	/* Make sure we are running on right CPU */
 	saved_mask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	retval = set_cpus_allowed_ptr(current, new_mask);
 	if (retval)
 		return -1;
 
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index b0c8208df9fa..ff2fff56f0a8 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -200,10 +200,12 @@ static void drv_read(struct drv_cmd *cmd)
 static void drv_write(struct drv_cmd *cmd)
 {
 	cpumask_t saved_mask = current->cpus_allowed;
+	cpumask_of_cpu_ptr_declare(cpu_mask);
 	unsigned int i;
 
-	for_each_cpu_mask(i, cmd->mask) {
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(i));
+	for_each_cpu_mask_nr(i, cmd->mask) {
+		cpumask_of_cpu_ptr_next(cpu_mask, i);
+		set_cpus_allowed_ptr(current, cpu_mask);
 		do_drv_write(cmd);
 	}
 
@@ -267,11 +269,12 @@ static unsigned int get_measured_perf(unsigned int cpu)
 	} aperf_cur, mperf_cur;
 
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	unsigned int perf_percent;
 	unsigned int retval;
 
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, cpu_mask);
 	if (get_cpu() != cpu) {
 		/* We were not able to run on requested processor */
 		put_cpu();
@@ -337,6 +340,7 @@ static unsigned int get_measured_perf(unsigned int cpu)
 
 static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 {
+	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	struct acpi_cpufreq_data *data = per_cpu(drv_data, cpu);
 	unsigned int freq;
 	unsigned int cached_freq;
@@ -349,7 +353,7 @@ static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 	}
 
 	cached_freq = data->freq_table[data->acpi_data->state].frequency;
-	freq = extract_freq(get_cur_val(&cpumask_of_cpu(cpu)), data);
+	freq = extract_freq(get_cur_val(cpu_mask), data);
 	if (freq != cached_freq) {
 		/*
 		 * The dreaded BIOS frequency change behind our back.
@@ -451,7 +455,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 
 	freqs.old = perf->states[perf->state].core_frequency * 1000;
 	freqs.new = data->freq_table[next_state].frequency;
-	for_each_cpu_mask(i, cmd.mask) {
+	for_each_cpu_mask_nr(i, cmd.mask) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 	}
@@ -466,7 +470,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 		}
 	}
 
-	for_each_cpu_mask(i, cmd.mask) {
+	for_each_cpu_mask_nr(i, cmd.mask) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
index 199e4e05e5dc..f1685fb91fbd 100644
--- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
+++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
@@ -122,7 +122,7 @@ static int cpufreq_p4_target(struct cpufreq_policy *policy,
 		return 0;
 
 	/* notifiers */
-	for_each_cpu_mask(i, policy->cpus) {
+	for_each_cpu_mask_nr(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 	}
@@ -130,11 +130,11 @@ static int cpufreq_p4_target(struct cpufreq_policy *policy,
 	/* run on each logical CPU, see section 13.15.3 of IA32 Intel Architecture Software
 	 * Developer's Manual, Volume 3
 	 */
-	for_each_cpu_mask(i, policy->cpus)
+	for_each_cpu_mask_nr(i, policy->cpus)
 		cpufreq_p4_setdc(i, p4clockmod_table[newstate].index);
 
 	/* notifiers */
-	for_each_cpu_mask(i, policy->cpus) {
+	for_each_cpu_mask_nr(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index 206791eb46e3..53c7b6936973 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -479,11 +479,12 @@ static int core_voltage_post_transition(struct powernow_k8_data *data, u32 reqvi
 static int check_supported_cpu(unsigned int cpu)
 {
 	cpumask_t oldmask;
+	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	u32 eax, ebx, ecx, edx;
 	unsigned int rc = 0;
 
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, cpu_mask);
 
 	if (smp_processor_id() != cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", cpu);
@@ -966,7 +967,7 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data, unsigned i
 	freqs.old = find_khz_freq_from_fid(data->currfid);
 	freqs.new = find_khz_freq_from_fid(fid);
 
-	for_each_cpu_mask(i, *(data->available_cores)) {
+	for_each_cpu_mask_nr(i, *(data->available_cores)) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 	}
@@ -974,7 +975,7 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data, unsigned i
 	res = transition_fid_vid(data, fid, vid);
 	freqs.new = find_khz_freq_from_fid(data->currfid);
 
-	for_each_cpu_mask(i, *(data->available_cores)) {
+	for_each_cpu_mask_nr(i, *(data->available_cores)) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
@@ -997,7 +998,7 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned i
 	freqs.old = find_khz_freq_from_pstate(data->powernow_table, data->currpstate);
 	freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate);
 
-	for_each_cpu_mask(i, *(data->available_cores)) {
+	for_each_cpu_mask_nr(i, *(data->available_cores)) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 	}
@@ -1005,7 +1006,7 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned i
 	res = transition_pstate(data, pstate);
 	freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate);
 
-	for_each_cpu_mask(i, *(data->available_cores)) {
+	for_each_cpu_mask_nr(i, *(data->available_cores)) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
@@ -1016,6 +1017,7 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned i
 static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsigned relation)
 {
 	cpumask_t oldmask;
+	cpumask_of_cpu_ptr(cpu_mask, pol->cpu);
 	struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu);
 	u32 checkfid;
 	u32 checkvid;
@@ -1030,7 +1032,7 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi
 
 	/* only run on specific CPU from here on */
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu));
+	set_cpus_allowed_ptr(current, cpu_mask);
 
 	if (smp_processor_id() != pol->cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu);
@@ -1105,6 +1107,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 {
 	struct powernow_k8_data *data;
 	cpumask_t oldmask;
+	cpumask_of_cpu_ptr_declare(newmask);
 	int rc;
 
 	if (!cpu_online(pol->cpu))
@@ -1156,7 +1159,8 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 
 	/* only run on specific CPU from here on */
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu));
+	cpumask_of_cpu_ptr_next(newmask, pol->cpu);
+	set_cpus_allowed_ptr(current, newmask);
 
 	if (smp_processor_id() != pol->cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu);
@@ -1178,7 +1182,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 	set_cpus_allowed_ptr(current, &oldmask);
 
 	if (cpu_family == CPU_HW_PSTATE)
-		pol->cpus = cpumask_of_cpu(pol->cpu);
+		pol->cpus = *newmask;
 	else
 		pol->cpus = per_cpu(cpu_core_map, pol->cpu);
 	data->available_cores = &(pol->cpus);
@@ -1244,6 +1248,7 @@ static unsigned int powernowk8_get (unsigned int cpu)
 {
 	struct powernow_k8_data *data;
 	cpumask_t oldmask = current->cpus_allowed;
+	cpumask_of_cpu_ptr(newmask, cpu);
 	unsigned int khz = 0;
 	unsigned int first;
 
@@ -1253,7 +1258,7 @@ static unsigned int powernowk8_get (unsigned int cpu)
 	if (!data)
 		return -EINVAL;
 
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, newmask);
 	if (smp_processor_id() != cpu) {
 		printk(KERN_ERR PFX
 			"limiting to CPU %d failed in powernowk8_get\n", cpu);
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index 908dd347c67e..ca2ac13b7af2 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -28,7 +28,8 @@
 #define PFX		"speedstep-centrino: "
 #define MAINTAINER	"cpufreq@lists.linux.org.uk"
 
-#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-centrino", msg)
+#define dprintk(msg...) \
+	cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-centrino", msg)
 
 #define INTEL_MSR_RANGE	(0xffff)
 
@@ -66,11 +67,12 @@ struct cpu_model
 
 	struct cpufreq_frequency_table *op_points; /* clock/voltage pairs */
 };
-static int centrino_verify_cpu_id(const struct cpuinfo_x86 *c, const struct cpu_id *x);
+static int centrino_verify_cpu_id(const struct cpuinfo_x86 *c,
+				  const struct cpu_id *x);
 
 /* Operating points for current CPU */
-static struct cpu_model *centrino_model[NR_CPUS];
-static const struct cpu_id *centrino_cpu[NR_CPUS];
+static DEFINE_PER_CPU(struct cpu_model *, centrino_model);
+static DEFINE_PER_CPU(const struct cpu_id *, centrino_cpu);
 
 static struct cpufreq_driver centrino_driver;
 
@@ -255,7 +257,7 @@ static int centrino_cpu_init_table(struct cpufreq_policy *policy)
 		return -ENOENT;
 	}
 
-	centrino_model[policy->cpu] = model;
+	per_cpu(centrino_model, policy->cpu) = model;
 
 	dprintk("found \"%s\": max frequency: %dkHz\n",
 	       model->model_name, model->max_freq);
@@ -264,10 +266,14 @@ static int centrino_cpu_init_table(struct cpufreq_policy *policy)
 }
 
 #else
-static inline int centrino_cpu_init_table(struct cpufreq_policy *policy) { return -ENODEV; }
+static inline int centrino_cpu_init_table(struct cpufreq_policy *policy)
+{
+	return -ENODEV;
+}
 #endif /* CONFIG_X86_SPEEDSTEP_CENTRINO_TABLE */
 
-static int centrino_verify_cpu_id(const struct cpuinfo_x86 *c, const struct cpu_id *x)
+static int centrino_verify_cpu_id(const struct cpuinfo_x86 *c,
+				  const struct cpu_id *x)
 {
 	if ((c->x86 == x->x86) &&
 	    (c->x86_model == x->x86_model) &&
@@ -286,23 +292,28 @@ static unsigned extract_clock(unsigned msr, unsigned int cpu, int failsafe)
 	 * for centrino, as some DSDTs are buggy.
 	 * Ideally, this can be done using the acpi_data structure.
 	 */
-	if ((centrino_cpu[cpu] == &cpu_ids[CPU_BANIAS]) ||
-	    (centrino_cpu[cpu] == &cpu_ids[CPU_DOTHAN_A1]) ||
-	    (centrino_cpu[cpu] == &cpu_ids[CPU_DOTHAN_B0])) {
+	if ((per_cpu(centrino_cpu, cpu) == &cpu_ids[CPU_BANIAS]) ||
+	    (per_cpu(centrino_cpu, cpu) == &cpu_ids[CPU_DOTHAN_A1]) ||
+	    (per_cpu(centrino_cpu, cpu) == &cpu_ids[CPU_DOTHAN_B0])) {
 		msr = (msr >> 8) & 0xff;
 		return msr * 100000;
 	}
 
-	if ((!centrino_model[cpu]) || (!centrino_model[cpu]->op_points))
+	if ((!per_cpu(centrino_model, cpu)) ||
+	    (!per_cpu(centrino_model, cpu)->op_points))
 		return 0;
 
 	msr &= 0xffff;
-	for (i=0;centrino_model[cpu]->op_points[i].frequency != CPUFREQ_TABLE_END; i++) {
-		if (msr == centrino_model[cpu]->op_points[i].index)
-			return centrino_model[cpu]->op_points[i].frequency;
+	for (i = 0;
+		per_cpu(centrino_model, cpu)->op_points[i].frequency
+							!= CPUFREQ_TABLE_END;
+	     i++) {
+		if (msr == per_cpu(centrino_model, cpu)->op_points[i].index)
+			return per_cpu(centrino_model, cpu)->
+							op_points[i].frequency;
 	}
 	if (failsafe)
-		return centrino_model[cpu]->op_points[i-1].frequency;
+		return per_cpu(centrino_model, cpu)->op_points[i-1].frequency;
 	else
 		return 0;
 }
@@ -313,9 +324,10 @@ static unsigned int get_cur_freq(unsigned int cpu)
 	unsigned l, h;
 	unsigned clock_freq;
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr(new_mask, cpu);
 
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, new_mask);
 	if (smp_processor_id() != cpu)
 		return 0;
 
@@ -347,7 +359,8 @@ static int centrino_cpu_init(struct cpufreq_policy *policy)
 	int i;
 
 	/* Only Intel makes Enhanced Speedstep-capable CPUs */
-	if (cpu->x86_vendor != X86_VENDOR_INTEL || !cpu_has(cpu, X86_FEATURE_EST))
+	if (cpu->x86_vendor != X86_VENDOR_INTEL ||
+	    !cpu_has(cpu, X86_FEATURE_EST))
 		return -ENODEV;
 
 	if (cpu_has(cpu, X86_FEATURE_CONSTANT_TSC))
@@ -361,9 +374,9 @@ static int centrino_cpu_init(struct cpufreq_policy *policy)
 			break;
 
 	if (i != N_IDS)
-		centrino_cpu[policy->cpu] = &cpu_ids[i];
+		per_cpu(centrino_cpu, policy->cpu) = &cpu_ids[i];
 
-	if (!centrino_cpu[policy->cpu]) {
+	if (!per_cpu(centrino_cpu, policy->cpu)) {
 		dprintk("found unsupported CPU with "
 		"Enhanced SpeedStep: send /proc/cpuinfo to "
 		MAINTAINER "\n");
@@ -386,23 +399,26 @@ static int centrino_cpu_init(struct cpufreq_policy *policy)
 		/* check to see if it stuck */
 		rdmsr(MSR_IA32_MISC_ENABLE, l, h);
 		if (!(l & (1<<16))) {
-			printk(KERN_INFO PFX "couldn't enable Enhanced SpeedStep\n");
+			printk(KERN_INFO PFX
+				"couldn't enable Enhanced SpeedStep\n");
 			return -ENODEV;
 		}
 	}
 
 	freq = get_cur_freq(policy->cpu);
-
-	policy->cpuinfo.transition_latency = 10000; /* 10uS transition latency */
+	policy->cpuinfo.transition_latency = 10000;
+						/* 10uS transition latency */
 	policy->cur = freq;
 
 	dprintk("centrino_cpu_init: cur=%dkHz\n", policy->cur);
 
-	ret = cpufreq_frequency_table_cpuinfo(policy, centrino_model[policy->cpu]->op_points);
+	ret = cpufreq_frequency_table_cpuinfo(policy,
+		per_cpu(centrino_model, policy->cpu)->op_points);
 	if (ret)
 		return (ret);
 
-	cpufreq_frequency_table_get_attr(centrino_model[policy->cpu]->op_points, policy->cpu);
+	cpufreq_frequency_table_get_attr(
+		per_cpu(centrino_model, policy->cpu)->op_points, policy->cpu);
 
 	return 0;
 }
@@ -411,12 +427,12 @@ static int centrino_cpu_exit(struct cpufreq_policy *policy)
 {
 	unsigned int cpu = policy->cpu;
 
-	if (!centrino_model[cpu])
+	if (!per_cpu(centrino_model, cpu))
 		return -ENODEV;
 
 	cpufreq_frequency_table_put_attr(cpu);
 
-	centrino_model[cpu] = NULL;
+	per_cpu(centrino_model, cpu) = NULL;
 
 	return 0;
 }
@@ -430,17 +446,26 @@ static int centrino_cpu_exit(struct cpufreq_policy *policy)
  */
 static int centrino_verify (struct cpufreq_policy *policy)
 {
-	return cpufreq_frequency_table_verify(policy, centrino_model[policy->cpu]->op_points);
+	return cpufreq_frequency_table_verify(policy,
+			per_cpu(centrino_model, policy->cpu)->op_points);
 }
 
 /**
  * centrino_setpolicy - set a new CPUFreq policy
  * @policy: new policy
  * @target_freq: the target frequency
- * @relation: how that frequency relates to achieved frequency (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H)
+ * @relation: how that frequency relates to achieved frequency
+ *	(CPUFREQ_RELATION_L or CPUFREQ_RELATION_H)
  *
  * Sets a new CPUFreq policy.
  */
+struct allmasks {
+	cpumask_t		online_policy_cpus;
+	cpumask_t		saved_mask;
+	cpumask_t		set_mask;
+	cpumask_t		covered_cpus;
+};
+
 static int centrino_target (struct cpufreq_policy *policy,
 			    unsigned int target_freq,
 			    unsigned int relation)
@@ -448,48 +473,55 @@ static int centrino_target (struct cpufreq_policy *policy,
 	unsigned int    newstate = 0;
 	unsigned int	msr, oldmsr = 0, h = 0, cpu = policy->cpu;
 	struct cpufreq_freqs	freqs;
-	cpumask_t		online_policy_cpus;
-	cpumask_t		saved_mask;
-	cpumask_t		set_mask;
-	cpumask_t		covered_cpus;
 	int			retval = 0;
 	unsigned int		j, k, first_cpu, tmp;
+	CPUMASK_ALLOC(allmasks);
+	CPUMASK_PTR(online_policy_cpus, allmasks);
+	CPUMASK_PTR(saved_mask, allmasks);
+	CPUMASK_PTR(set_mask, allmasks);
+	CPUMASK_PTR(covered_cpus, allmasks);
 
-	if (unlikely(centrino_model[cpu] == NULL))
-		return -ENODEV;
+	if (unlikely(allmasks == NULL))
+		return -ENOMEM;
+
+	if (unlikely(per_cpu(centrino_model, cpu) == NULL)) {
+		retval = -ENODEV;
+		goto out;
+	}
 
 	if (unlikely(cpufreq_frequency_table_target(policy,
-			centrino_model[cpu]->op_points,
+			per_cpu(centrino_model, cpu)->op_points,
 			target_freq,
 			relation,
 			&newstate))) {
-		return -EINVAL;
+		retval = -EINVAL;
+		goto out;
 	}
 
 #ifdef CONFIG_HOTPLUG_CPU
 	/* cpufreq holds the hotplug lock, so we are safe from here on */
-	cpus_and(online_policy_cpus, cpu_online_map, policy->cpus);
+	cpus_and(*online_policy_cpus, cpu_online_map, policy->cpus);
 #else
-	online_policy_cpus = policy->cpus;
+	*online_policy_cpus = policy->cpus;
 #endif
 
-	saved_mask = current->cpus_allowed;
+	*saved_mask = current->cpus_allowed;
 	first_cpu = 1;
-	cpus_clear(covered_cpus);
-	for_each_cpu_mask(j, online_policy_cpus) {
+	cpus_clear(*covered_cpus);
+	for_each_cpu_mask_nr(j, *online_policy_cpus) {
 		/*
 		 * Support for SMP systems.
 		 * Make sure we are running on CPU that wants to change freq
 		 */
-		cpus_clear(set_mask);
+		cpus_clear(*set_mask);
 		if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY)
-			cpus_or(set_mask, set_mask, online_policy_cpus);
+			cpus_or(*set_mask, *set_mask, *online_policy_cpus);
 		else
-			cpu_set(j, set_mask);
+			cpu_set(j, *set_mask);
 
-		set_cpus_allowed_ptr(current, &set_mask);
+		set_cpus_allowed_ptr(current, set_mask);
 		preempt_disable();
-		if (unlikely(!cpu_isset(smp_processor_id(), set_mask))) {
+		if (unlikely(!cpu_isset(smp_processor_id(), *set_mask))) {
 			dprintk("couldn't limit to CPUs in this domain\n");
 			retval = -EAGAIN;
 			if (first_cpu) {
@@ -500,7 +532,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 			break;
 		}
 
-		msr = centrino_model[cpu]->op_points[newstate].index;
+		msr = per_cpu(centrino_model, cpu)->op_points[newstate].index;
 
 		if (first_cpu) {
 			rdmsr(MSR_IA32_PERF_CTL, oldmsr, h);
@@ -517,7 +549,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 			dprintk("target=%dkHz old=%d new=%d msr=%04x\n",
 				target_freq, freqs.old, freqs.new, msr);
 
-			for_each_cpu_mask(k, online_policy_cpus) {
+			for_each_cpu_mask_nr(k, *online_policy_cpus) {
 				freqs.cpu = k;
 				cpufreq_notify_transition(&freqs,
 					CPUFREQ_PRECHANGE);
@@ -536,11 +568,11 @@ static int centrino_target (struct cpufreq_policy *policy,
 			break;
 		}
 
-		cpu_set(j, covered_cpus);
+		cpu_set(j, *covered_cpus);
 		preempt_enable();
 	}
 
-	for_each_cpu_mask(k, online_policy_cpus) {
+	for_each_cpu_mask_nr(k, *online_policy_cpus) {
 		freqs.cpu = k;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
@@ -553,10 +585,12 @@ static int centrino_target (struct cpufreq_policy *policy,
 		 * Best effort undo..
 		 */
 
-		if (!cpus_empty(covered_cpus)) {
-			for_each_cpu_mask(j, covered_cpus) {
-				set_cpus_allowed_ptr(current,
-						     &cpumask_of_cpu(j));
+		if (!cpus_empty(*covered_cpus)) {
+			cpumask_of_cpu_ptr_declare(new_mask);
+
+			for_each_cpu_mask_nr(j, *covered_cpus) {
+				cpumask_of_cpu_ptr_next(new_mask, j);
+				set_cpus_allowed_ptr(current, new_mask);
 				wrmsr(MSR_IA32_PERF_CTL, oldmsr, h);
 			}
 		}
@@ -564,19 +598,22 @@ static int centrino_target (struct cpufreq_policy *policy,
 		tmp = freqs.new;
 		freqs.new = freqs.old;
 		freqs.old = tmp;
-		for_each_cpu_mask(j, online_policy_cpus) {
+		for_each_cpu_mask_nr(j, *online_policy_cpus) {
 			freqs.cpu = j;
 			cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 			cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 		}
 	}
-	set_cpus_allowed_ptr(current, &saved_mask);
-	return 0;
+	set_cpus_allowed_ptr(current, saved_mask);
+	retval = 0;
+	goto out;
 
 migrate_end:
 	preempt_enable();
-	set_cpus_allowed_ptr(current, &saved_mask);
-	return 0;
+	set_cpus_allowed_ptr(current, saved_mask);
+out:
+	CPUMASK_FREE(allmasks);
+	return retval;
 }
 
 static struct freq_attr* centrino_attr[] = {
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
index 1b50244b1fdf..2f3728dc24f6 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
@@ -244,7 +244,8 @@ static unsigned int _speedstep_get(const cpumask_t *cpus)
 
 static unsigned int speedstep_get(unsigned int cpu)
 {
-	return _speedstep_get(&cpumask_of_cpu(cpu));
+	cpumask_of_cpu_ptr(newmask, cpu);
+	return _speedstep_get(newmask);
 }
 
 /**
@@ -279,7 +280,7 @@ static int speedstep_target (struct cpufreq_policy *policy,
 
 	cpus_allowed = current->cpus_allowed;
 
-	for_each_cpu_mask(i, policy->cpus) {
+	for_each_cpu_mask_nr(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 	}
@@ -292,7 +293,7 @@ static int speedstep_target (struct cpufreq_policy *policy,
 	/* allow to be run on all CPUs */
 	set_cpus_allowed_ptr(current, &cpus_allowed);
 
-	for_each_cpu_mask(i, policy->cpus) {
+	for_each_cpu_mask_nr(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index ff517f0b8cc4..650d40f7912b 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -489,7 +489,7 @@ static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
 	int sibling;
 
 	this_leaf = CPUID4_INFO_IDX(cpu, index);
-	for_each_cpu_mask(sibling, this_leaf->shared_cpu_map) {
+	for_each_cpu_mask_nr(sibling, this_leaf->shared_cpu_map) {
 		sibling_leaf = CPUID4_INFO_IDX(sibling, index);	
 		cpu_clear(cpu, sibling_leaf->shared_cpu_map);
 	}
@@ -516,6 +516,7 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 	unsigned long		j;
 	int			retval;
 	cpumask_t		oldmask;
+	cpumask_of_cpu_ptr(newmask, cpu);
 
 	if (num_cache_leaves == 0)
 		return -ENOENT;
@@ -526,7 +527,7 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 		return -ENOMEM;
 
 	oldmask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	retval = set_cpus_allowed_ptr(current, newmask);
 	if (retval)
 		goto out;
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 9ab65be82427..65a339678ece 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -580,7 +580,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
 	char __user *buf = ubuf;
 	int i, err;
 
-	cpu_tsc = kmalloc(NR_CPUS * sizeof(long), GFP_KERNEL);
+	cpu_tsc = kmalloc(nr_cpu_ids * sizeof(long), GFP_KERNEL);
 	if (!cpu_tsc)
 		return -ENOMEM;
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index 7c9a813e1193..88736cadbaa6 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -527,7 +527,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 	if (err)
 		goto out_free;
 
-	for_each_cpu_mask(i, b->cpus) {
+	for_each_cpu_mask_nr(i, b->cpus) {
 		if (i == cpu)
 			continue;
 
@@ -617,7 +617,7 @@ static void threshold_remove_bank(unsigned int cpu, int bank)
 #endif
 
 	/* remove all sibling symlinks before unregistering */
-	for_each_cpu_mask(i, b->cpus) {
+	for_each_cpu_mask_nr(i, b->cpus) {
 		if (i == cpu)
 			continue;
 
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index 0d0d9057e7c0..a26c480b9491 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -160,7 +160,7 @@ static void *c_start(struct seq_file *m, loff_t *pos)
 {
 	if (*pos == 0)	/* just in case, cpu 0 is not the first */
 		*pos = first_cpu(cpu_online_map);
-	if ((*pos) < NR_CPUS && cpu_online(*pos))
+	if ((*pos) < nr_cpu_ids && cpu_online(*pos))
 		return &cpu_data(*pos);
 	return NULL;
 }
diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c
index 1a9c68845ee8..786548a62d38 100644
--- a/arch/x86/kernel/genapic_flat_64.c
+++ b/arch/x86/kernel/genapic_flat_64.c
@@ -168,7 +168,7 @@ static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * May as well be the first.
 	 */
 	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
index 3c3929340692..2cfcbded888a 100644
--- a/arch/x86/kernel/genx2apic_uv_x.c
+++ b/arch/x86/kernel/genx2apic_uv_x.c
@@ -98,7 +98,7 @@ static void uv_send_IPI_mask(cpumask_t mask, int vector)
 {
 	unsigned int cpu;
 
-	for (cpu = 0; cpu < NR_CPUS; ++cpu)
+	for_each_possible_cpu(cpu)
 		if (cpu_isset(cpu, mask))
 			uv_send_IPI_one(cpu, vector);
 }
@@ -132,7 +132,7 @@ static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
 	 * May as well be the first.
 	 */
 	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 64a46affd858..8269434d1707 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -732,7 +732,7 @@ static int __assign_irq_vector(int irq, cpumask_t mask)
 			return 0;
 	}
 
-	for_each_cpu_mask(cpu, mask) {
+	for_each_cpu_mask_nr(cpu, mask) {
 		cpumask_t domain, new_mask;
 		int new_cpu;
 		int vector, offset;
@@ -753,7 +753,7 @@ next:
 			continue;
 		if (vector == IA32_SYSCALL_VECTOR)
 			goto next;
-		for_each_cpu_mask(new_cpu, new_mask)
+		for_each_cpu_mask_nr(new_cpu, new_mask)
 			if (per_cpu(vector_irq, new_cpu)[vector] != -1)
 				goto next;
 		/* Found one! */
@@ -763,7 +763,7 @@ next:
 			cfg->move_in_progress = 1;
 			cfg->old_domain = cfg->domain;
 		}
-		for_each_cpu_mask(new_cpu, new_mask)
+		for_each_cpu_mask_nr(new_cpu, new_mask)
 			per_cpu(vector_irq, new_cpu)[vector] = irq;
 		cfg->vector = vector;
 		cfg->domain = domain;
@@ -795,7 +795,7 @@ static void __clear_irq_vector(int irq)
 
 	vector = cfg->vector;
 	cpus_and(mask, cfg->domain, cpu_online_map);
-	for_each_cpu_mask(cpu, mask)
+	for_each_cpu_mask_nr(cpu, mask)
 		per_cpu(vector_irq, cpu)[vector] = -1;
 
 	cfg->vector = 0;
@@ -1373,12 +1373,10 @@ static unsigned int startup_ioapic_irq(unsigned int irq)
 static int ioapic_retrigger_irq(unsigned int irq)
 {
 	struct irq_cfg *cfg = &irq_cfg[irq];
-	cpumask_t mask;
 	unsigned long flags;
 
 	spin_lock_irqsave(&vector_lock, flags);
-	mask = cpumask_of_cpu(first_cpu(cfg->domain));
-	send_IPI_mask(mask, cfg->vector);
+	send_IPI_mask(cpumask_of_cpu(first_cpu(cfg->domain)), cfg->vector);
 	spin_unlock_irqrestore(&vector_lock, flags);
 
 	return 1;
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index a8449571858a..3fee2aa50f3f 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -62,12 +62,12 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
 
 	if (reload) {
 #ifdef CONFIG_SMP
-		cpumask_t mask;
+		cpumask_of_cpu_ptr_declare(mask);
 
 		preempt_disable();
 		load_LDT(pc);
-		mask = cpumask_of_cpu(smp_processor_id());
-		if (!cpus_equal(current->mm->cpu_vm_mask, mask))
+		cpumask_of_cpu_ptr_next(mask, smp_processor_id());
+		if (!cpus_equal(current->mm->cpu_vm_mask, *mask))
 			smp_call_function(flush_ldt, current->mm, 1);
 		preempt_enable();
 #else
diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c
index fc4790638b69..6994c751590e 100644
--- a/arch/x86/kernel/microcode.c
+++ b/arch/x86/kernel/microcode.c
@@ -388,6 +388,7 @@ static int do_microcode_update (void)
 	void *new_mc = NULL;
 	int cpu;
 	cpumask_t old;
+	cpumask_of_cpu_ptr_declare(newmask);
 
 	old = current->cpus_allowed;
 
@@ -404,7 +405,8 @@ static int do_microcode_update (void)
 
 			if (!uci->valid)
 				continue;
-			set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+			cpumask_of_cpu_ptr_next(newmask, cpu);
+			set_cpus_allowed_ptr(current, newmask);
 			error = get_maching_microcode(new_mc, cpu);
 			if (error < 0)
 				goto out;
@@ -574,6 +576,7 @@ static int apply_microcode_check_cpu(int cpu)
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 	cpumask_t old;
+	cpumask_of_cpu_ptr(newmask, cpu);
 	unsigned int val[2];
 	int err = 0;
 
@@ -582,7 +585,7 @@ static int apply_microcode_check_cpu(int cpu)
 		return 0;
 
 	old = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, newmask);
 
 	/* Check if the microcode we have in memory matches the CPU */
 	if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
@@ -620,11 +623,12 @@ static int apply_microcode_check_cpu(int cpu)
 static void microcode_init_cpu(int cpu, int resume)
 {
 	cpumask_t old;
+	cpumask_of_cpu_ptr(newmask, cpu);
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 
 	old = current->cpus_allowed;
 
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, newmask);
 	mutex_lock(&microcode_mutex);
 	collect_cpu_info(cpu);
 	if (uci->valid && system_state == SYSTEM_RUNNING && !resume)
@@ -658,11 +662,12 @@ static ssize_t reload_store(struct sys_device *dev,
 		return -EINVAL;
 	if (val == 1) {
 		cpumask_t old;
+		cpumask_of_cpu_ptr(newmask, cpu);
 
 		old = current->cpus_allowed;
 
 		get_online_cpus();
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+		set_cpus_allowed_ptr(current, newmask);
 
 		mutex_lock(&microcode_mutex);
 		if (uci->valid)
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 9dcf39c02972..06a9f643817e 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -411,24 +411,28 @@ void native_machine_shutdown(void)
 {
 	/* Stop the cpus and apics */
 #ifdef CONFIG_SMP
-	int reboot_cpu_id;
 
 	/* The boot cpu is always logical cpu 0 */
-	reboot_cpu_id = 0;
+	int reboot_cpu_id = 0;
+	cpumask_of_cpu_ptr(newmask, reboot_cpu_id);
 
 #ifdef CONFIG_X86_32
 	/* See if there has been given a command line override */
 	if ((reboot_cpu != -1) && (reboot_cpu < NR_CPUS) &&
-		cpu_online(reboot_cpu))
+		cpu_online(reboot_cpu)) {
 		reboot_cpu_id = reboot_cpu;
+		cpumask_of_cpu_ptr_next(newmask, reboot_cpu_id);
+	}
 #endif
 
 	/* Make certain the cpu I'm about to reboot on is online */
-	if (!cpu_online(reboot_cpu_id))
+	if (!cpu_online(reboot_cpu_id)) {
 		reboot_cpu_id = smp_processor_id();
+		cpumask_of_cpu_ptr_next(newmask, reboot_cpu_id);
+	}
 
 	/* Make certain I only run on the appropriate processor */
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(reboot_cpu_id));
+	set_cpus_allowed_ptr(current, newmask);
 
 	/* O.K Now that I'm on the appropriate processor,
 	 * stop all of the others.
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 4b53a647bc0a..332512767f4f 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -438,7 +438,7 @@ void __cpuinit set_cpu_sibling_map(int cpu)
 	cpu_set(cpu, cpu_sibling_setup_map);
 
 	if (smp_num_siblings > 1) {
-		for_each_cpu_mask(i, cpu_sibling_setup_map) {
+		for_each_cpu_mask_nr(i, cpu_sibling_setup_map) {
 			if (c->phys_proc_id == cpu_data(i).phys_proc_id &&
 			    c->cpu_core_id == cpu_data(i).cpu_core_id) {
 				cpu_set(i, per_cpu(cpu_sibling_map, cpu));
@@ -461,7 +461,7 @@ void __cpuinit set_cpu_sibling_map(int cpu)
 		return;
 	}
 
-	for_each_cpu_mask(i, cpu_sibling_setup_map) {
+	for_each_cpu_mask_nr(i, cpu_sibling_setup_map) {
 		if (per_cpu(cpu_llc_id, cpu) != BAD_APICID &&
 		    per_cpu(cpu_llc_id, cpu) == per_cpu(cpu_llc_id, i)) {
 			cpu_set(i, c->llc_shared_map);
@@ -1219,7 +1219,7 @@ static void remove_siblinginfo(int cpu)
 	int sibling;
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 
-	for_each_cpu_mask(sibling, per_cpu(cpu_core_map, cpu)) {
+	for_each_cpu_mask_nr(sibling, per_cpu(cpu_core_map, cpu)) {
 		cpu_clear(cpu, per_cpu(cpu_core_map, sibling));
 		/*/
 		 * last thread sibling in this cpu core going down
@@ -1228,7 +1228,7 @@ static void remove_siblinginfo(int cpu)
 			cpu_data(sibling).booted_cores--;
 	}
 
-	for_each_cpu_mask(sibling, per_cpu(cpu_sibling_map, cpu))
+	for_each_cpu_mask_nr(sibling, per_cpu(cpu_sibling_map, cpu))
 		cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling));
 	cpus_clear(per_cpu(cpu_sibling_map, cpu));
 	cpus_clear(per_cpu(cpu_core_map, cpu));
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index e693812ac59a..d8faf79a0a1d 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -367,7 +367,7 @@ static void xen_send_IPI_mask(cpumask_t mask, enum ipi_vector vector)
 
 	cpus_and(mask, mask, cpu_online_map);
 
-	for_each_cpu_mask(cpu, mask)
+	for_each_cpu_mask_nr(cpu, mask)
 		xen_send_IPI_one(cpu, vector);
 }
 
@@ -378,7 +378,7 @@ static void xen_smp_send_call_function_ipi(cpumask_t mask)
 	xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
 
 	/* Make sure other vcpus get a chance to run if they need to. */
-	for_each_cpu_mask(cpu, mask) {
+	for_each_cpu_mask_nr(cpu, mask) {
 		if (xen_vcpu_stolen(cpu)) {
 			HYPERVISOR_sched_op(SCHEDOP_yield, 0);
 			break;
diff --git a/drivers/acpi/processor_throttling.c b/drivers/acpi/processor_throttling.c
index 0622ace05220..a2c3f9cfa549 100644
--- a/drivers/acpi/processor_throttling.c
+++ b/drivers/acpi/processor_throttling.c
@@ -827,6 +827,7 @@ static int acpi_processor_get_throttling_ptc(struct acpi_processor *pr)
 static int acpi_processor_get_throttling(struct acpi_processor *pr)
 {
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr_declare(new_mask);
 	int ret;
 
 	if (!pr)
@@ -838,7 +839,8 @@ static int acpi_processor_get_throttling(struct acpi_processor *pr)
 	 * Migrate task to the cpu pointed by pr.
 	 */
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pr->id));
+	cpumask_of_cpu_ptr_next(new_mask, pr->id);
+	set_cpus_allowed_ptr(current, new_mask);
 	ret = pr->throttling.acpi_processor_get_throttling(pr);
 	/* restore the previous state */
 	set_cpus_allowed_ptr(current, &saved_mask);
@@ -987,6 +989,7 @@ static int acpi_processor_set_throttling_ptc(struct acpi_processor *pr,
 int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 {
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr_declare(new_mask);
 	int ret = 0;
 	unsigned int i;
 	struct acpi_processor *match_pr;
@@ -1013,7 +1016,7 @@ int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 	 * affected cpu in order to get one proper T-state.
 	 * The notifier event is THROTTLING_PRECHANGE.
 	 */
-	for_each_cpu_mask(i, online_throttling_cpus) {
+	for_each_cpu_mask_nr(i, online_throttling_cpus) {
 		t_state.cpu = i;
 		acpi_processor_throttling_notifier(THROTTLING_PRECHANGE,
 							&t_state);
@@ -1025,7 +1028,8 @@ int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 	 * it can be called only for the cpu pointed by pr.
 	 */
 	if (p_throttling->shared_type == DOMAIN_COORD_TYPE_SW_ANY) {
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(pr->id));
+		cpumask_of_cpu_ptr_next(new_mask, pr->id);
+		set_cpus_allowed_ptr(current, new_mask);
 		ret = p_throttling->acpi_processor_set_throttling(pr,
 						t_state.target_state);
 	} else {
@@ -1034,7 +1038,7 @@ int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 		 * it is necessary to set T-state for every affected
 		 * cpus.
 		 */
-		for_each_cpu_mask(i, online_throttling_cpus) {
+		for_each_cpu_mask_nr(i, online_throttling_cpus) {
 			match_pr = per_cpu(processors, i);
 			/*
 			 * If the pointer is invalid, we will report the
@@ -1056,7 +1060,8 @@ int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 				continue;
 			}
 			t_state.cpu = i;
-			set_cpus_allowed_ptr(current, &cpumask_of_cpu(i));
+			cpumask_of_cpu_ptr_next(new_mask, i);
+			set_cpus_allowed_ptr(current, new_mask);
 			ret = match_pr->throttling.
 				acpi_processor_set_throttling(
 				match_pr, t_state.target_state);
@@ -1068,7 +1073,7 @@ int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 	 * affected cpu to update the T-states.
 	 * The notifier event is THROTTLING_POSTCHANGE
 	 */
-	for_each_cpu_mask(i, online_throttling_cpus) {
+	for_each_cpu_mask_nr(i, online_throttling_cpus) {
 		t_state.cpu = i;
 		acpi_processor_throttling_notifier(THROTTLING_POSTCHANGE,
 							&t_state);
diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 20537d507909..64f5d54f7edc 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -121,14 +121,14 @@ static ssize_t print_cpus_##type(struct sysdev_class *class, char *buf)	\
 {									\
 	return print_cpus_map(buf, &cpu_##type##_map);			\
 }									\
-struct sysdev_class_attribute attr_##type##_map = 			\
+static struct sysdev_class_attribute attr_##type##_map = 		\
 	_SYSDEV_CLASS_ATTR(type, 0444, print_cpus_##type, NULL)
 
 print_cpus_func(online);
 print_cpus_func(possible);
 print_cpus_func(present);
 
-struct sysdev_class_attribute *cpu_state_attr[] = {
+static struct sysdev_class_attribute *cpu_state_attr[] = {
 	&attr_online_map,
 	&attr_possible_map,
 	&attr_present_map,
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index ee1df0d45e81..8d6a3ff02672 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -589,7 +589,7 @@ static ssize_t show_cpus(cpumask_t mask, char *buf)
 	ssize_t i = 0;
 	unsigned int cpu;
 
-	for_each_cpu_mask(cpu, mask) {
+	for_each_cpu_mask_nr(cpu, mask) {
 		if (i)
 			i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), " ");
 		i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u", cpu);
@@ -835,7 +835,7 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 	}
 #endif
 
-	for_each_cpu_mask(j, policy->cpus) {
+	for_each_cpu_mask_nr(j, policy->cpus) {
 		if (cpu == j)
 			continue;
 
@@ -898,14 +898,14 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 	}
 
 	spin_lock_irqsave(&cpufreq_driver_lock, flags);
-	for_each_cpu_mask(j, policy->cpus) {
+	for_each_cpu_mask_nr(j, policy->cpus) {
 		per_cpu(cpufreq_cpu_data, j) = policy;
 		per_cpu(policy_cpu, j) = policy->cpu;
 	}
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
 	/* symlink affected CPUs */
-	for_each_cpu_mask(j, policy->cpus) {
+	for_each_cpu_mask_nr(j, policy->cpus) {
 		if (j == cpu)
 			continue;
 		if (!cpu_online(j))
@@ -945,7 +945,7 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 
 err_out_unregister:
 	spin_lock_irqsave(&cpufreq_driver_lock, flags);
-	for_each_cpu_mask(j, policy->cpus)
+	for_each_cpu_mask_nr(j, policy->cpus)
 		per_cpu(cpufreq_cpu_data, j) = NULL;
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
@@ -1028,7 +1028,7 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev)
 	 * the sysfs links afterwards.
 	 */
 	if (unlikely(cpus_weight(data->cpus) > 1)) {
-		for_each_cpu_mask(j, data->cpus) {
+		for_each_cpu_mask_nr(j, data->cpus) {
 			if (j == cpu)
 				continue;
 			per_cpu(cpufreq_cpu_data, j) = NULL;
@@ -1038,7 +1038,7 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev)
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
 	if (unlikely(cpus_weight(data->cpus) > 1)) {
-		for_each_cpu_mask(j, data->cpus) {
+		for_each_cpu_mask_nr(j, data->cpus) {
 			if (j == cpu)
 				continue;
 			dprintk("removing link for cpu %u\n", j);
diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c
index 5d3a04ba6ad2..fe565ee43757 100644
--- a/drivers/cpufreq/cpufreq_conservative.c
+++ b/drivers/cpufreq/cpufreq_conservative.c
@@ -497,7 +497,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 			return rc;
 		}
 
-		for_each_cpu_mask(j, policy->cpus) {
+		for_each_cpu_mask_nr(j, policy->cpus) {
 			struct cpu_dbs_info_s *j_dbs_info;
 			j_dbs_info = &per_cpu(cpu_dbs_info, j);
 			j_dbs_info->cur_policy = policy;
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index d2af20dda382..33855cb3cf16 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -367,7 +367,7 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 
 	/* Get Idle Time */
 	idle_ticks = UINT_MAX;
-	for_each_cpu_mask(j, policy->cpus) {
+	for_each_cpu_mask_nr(j, policy->cpus) {
 		cputime64_t total_idle_ticks;
 		unsigned int tmp_idle_ticks;
 		struct cpu_dbs_info_s *j_dbs_info;
@@ -521,7 +521,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 			return rc;
 		}
 
-		for_each_cpu_mask(j, policy->cpus) {
+		for_each_cpu_mask_nr(j, policy->cpus) {
 			struct cpu_dbs_info_s *j_dbs_info;
 			j_dbs_info = &per_cpu(cpu_dbs_info, j);
 			j_dbs_info->cur_policy = policy;
diff --git a/drivers/cpufreq/cpufreq_userspace.c b/drivers/cpufreq/cpufreq_userspace.c
index cb2ac01a41a1..32244aa7cc0c 100644
--- a/drivers/cpufreq/cpufreq_userspace.c
+++ b/drivers/cpufreq/cpufreq_userspace.c
@@ -30,16 +30,18 @@
 /**
  * A few values needed by the userspace governor
  */
-static unsigned int	cpu_max_freq[NR_CPUS];
-static unsigned int	cpu_min_freq[NR_CPUS];
-static unsigned int	cpu_cur_freq[NR_CPUS]; /* current CPU freq */
-static unsigned int	cpu_set_freq[NR_CPUS]; /* CPU freq desired by userspace */
-static unsigned int	cpu_is_managed[NR_CPUS];
+static DEFINE_PER_CPU(unsigned int, cpu_max_freq);
+static DEFINE_PER_CPU(unsigned int, cpu_min_freq);
+static DEFINE_PER_CPU(unsigned int, cpu_cur_freq); /* current CPU freq */
+static DEFINE_PER_CPU(unsigned int, cpu_set_freq); /* CPU freq desired by
+							userspace */
+static DEFINE_PER_CPU(unsigned int, cpu_is_managed);
 
 static DEFINE_MUTEX	(userspace_mutex);
 static int cpus_using_userspace_governor;
 
-#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_GOVERNOR, "userspace", msg)
+#define dprintk(msg...) \
+	cpufreq_debug_printk(CPUFREQ_DEBUG_GOVERNOR, "userspace", msg)
 
 /* keep track of frequency transitions */
 static int
@@ -48,12 +50,12 @@ userspace_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
 {
         struct cpufreq_freqs *freq = data;
 
-	if (!cpu_is_managed[freq->cpu])
+	if (!per_cpu(cpu_is_managed, freq->cpu))
 		return 0;
 
 	dprintk("saving cpu_cur_freq of cpu %u to be %u kHz\n",
 			freq->cpu, freq->new);
-	cpu_cur_freq[freq->cpu] = freq->new;
+	per_cpu(cpu_cur_freq, freq->cpu) = freq->new;
 
         return 0;
 }
@@ -77,15 +79,15 @@ static int cpufreq_set(struct cpufreq_policy *policy, unsigned int freq)
 	dprintk("cpufreq_set for cpu %u, freq %u kHz\n", policy->cpu, freq);
 
 	mutex_lock(&userspace_mutex);
-	if (!cpu_is_managed[policy->cpu])
+	if (!per_cpu(cpu_is_managed, policy->cpu))
 		goto err;
 
-	cpu_set_freq[policy->cpu] = freq;
+	per_cpu(cpu_set_freq, policy->cpu) = freq;
 
-	if (freq < cpu_min_freq[policy->cpu])
-		freq = cpu_min_freq[policy->cpu];
-	if (freq > cpu_max_freq[policy->cpu])
-		freq = cpu_max_freq[policy->cpu];
+	if (freq < per_cpu(cpu_min_freq, policy->cpu))
+		freq = per_cpu(cpu_min_freq, policy->cpu);
+	if (freq > per_cpu(cpu_max_freq, policy->cpu))
+		freq = per_cpu(cpu_max_freq, policy->cpu);
 
 	/*
 	 * We're safe from concurrent calls to ->target() here
@@ -104,7 +106,7 @@ static int cpufreq_set(struct cpufreq_policy *policy, unsigned int freq)
 
 static ssize_t show_speed(struct cpufreq_policy *policy, char *buf)
 {
-	return sprintf(buf, "%u\n", cpu_cur_freq[policy->cpu]);
+	return sprintf(buf, "%u\n", per_cpu(cpu_cur_freq, policy->cpu));
 }
 
 static int cpufreq_governor_userspace(struct cpufreq_policy *policy,
@@ -127,12 +129,17 @@ static int cpufreq_governor_userspace(struct cpufreq_policy *policy,
 		}
 		cpus_using_userspace_governor++;
 
-		cpu_is_managed[cpu] = 1;
-		cpu_min_freq[cpu] = policy->min;
-		cpu_max_freq[cpu] = policy->max;
-		cpu_cur_freq[cpu] = policy->cur;
-		cpu_set_freq[cpu] = policy->cur;
-		dprintk("managing cpu %u started (%u - %u kHz, currently %u kHz)\n", cpu, cpu_min_freq[cpu], cpu_max_freq[cpu], cpu_cur_freq[cpu]);
+		per_cpu(cpu_is_managed, cpu) = 1;
+		per_cpu(cpu_min_freq, cpu) = policy->min;
+		per_cpu(cpu_max_freq, cpu) = policy->max;
+		per_cpu(cpu_cur_freq, cpu) = policy->cur;
+		per_cpu(cpu_set_freq, cpu) = policy->cur;
+		dprintk("managing cpu %u started "
+			"(%u - %u kHz, currently %u kHz)\n",
+				cpu,
+				per_cpu(cpu_min_freq, cpu),
+				per_cpu(cpu_max_freq, cpu),
+				per_cpu(cpu_cur_freq, cpu));
 
 		mutex_unlock(&userspace_mutex);
 		break;
@@ -145,34 +152,34 @@ static int cpufreq_governor_userspace(struct cpufreq_policy *policy,
 					CPUFREQ_TRANSITION_NOTIFIER);
 		}
 
-		cpu_is_managed[cpu] = 0;
-		cpu_min_freq[cpu] = 0;
-		cpu_max_freq[cpu] = 0;
-		cpu_set_freq[cpu] = 0;
+		per_cpu(cpu_is_managed, cpu) = 0;
+		per_cpu(cpu_min_freq, cpu) = 0;
+		per_cpu(cpu_max_freq, cpu) = 0;
+		per_cpu(cpu_set_freq, cpu) = 0;
 		dprintk("managing cpu %u stopped\n", cpu);
 		mutex_unlock(&userspace_mutex);
 		break;
 	case CPUFREQ_GOV_LIMITS:
 		mutex_lock(&userspace_mutex);
-		dprintk("limit event for cpu %u: %u - %u kHz,"
+		dprintk("limit event for cpu %u: %u - %u kHz, "
 			"currently %u kHz, last set to %u kHz\n",
 			cpu, policy->min, policy->max,
-			cpu_cur_freq[cpu], cpu_set_freq[cpu]);
-		if (policy->max < cpu_set_freq[cpu]) {
+			per_cpu(cpu_cur_freq, cpu),
+			per_cpu(cpu_set_freq, cpu));
+		if (policy->max < per_cpu(cpu_set_freq, cpu)) {
 			__cpufreq_driver_target(policy, policy->max,
 						CPUFREQ_RELATION_H);
-		}
-		else if (policy->min > cpu_set_freq[cpu]) {
+		} else if (policy->min > per_cpu(cpu_set_freq, cpu)) {
 			__cpufreq_driver_target(policy, policy->min,
 						CPUFREQ_RELATION_L);
-		}
-		else {
-			__cpufreq_driver_target(policy, cpu_set_freq[cpu],
+		} else {
+			__cpufreq_driver_target(policy,
+						per_cpu(cpu_set_freq, cpu),
 						CPUFREQ_RELATION_L);
 		}
-		cpu_min_freq[cpu] = policy->min;
-		cpu_max_freq[cpu] = policy->max;
-		cpu_cur_freq[cpu] = policy->cur;
+		per_cpu(cpu_min_freq, cpu) = policy->min;
+		per_cpu(cpu_max_freq, cpu) = policy->max;
+		per_cpu(cpu_cur_freq, cpu) = policy->cur;
 		mutex_unlock(&userspace_mutex);
 		break;
 	}
diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c
index 25918f7dfd0f..0b624e927a6f 100644
--- a/drivers/firmware/dcdbas.c
+++ b/drivers/firmware/dcdbas.c
@@ -254,6 +254,7 @@ static ssize_t host_control_on_shutdown_store(struct device *dev,
 static int smi_request(struct smi_cmd *smi_cmd)
 {
 	cpumask_t old_mask;
+	cpumask_of_cpu_ptr(new_mask, 0);
 	int ret = 0;
 
 	if (smi_cmd->magic != SMI_CMD_MAGIC) {
@@ -264,7 +265,7 @@ static int smi_request(struct smi_cmd *smi_cmd)
 
 	/* SMI requires CPU 0 */
 	old_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(0));
+	set_cpus_allowed_ptr(current, new_mask);
 	if (smp_processor_id() != 0) {
 		dev_dbg(&dcdbas_pdev->dev, "%s: failed to get CPU 0\n",
 			__func__);
diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c
index 0792d930c481..7a64aa9b51b6 100644
--- a/drivers/infiniband/hw/ehca/ehca_irq.c
+++ b/drivers/infiniband/hw/ehca/ehca_irq.c
@@ -646,8 +646,8 @@ static inline int find_next_online_cpu(struct ehca_comp_pool *pool)
 		ehca_dmp(&cpu_online_map, sizeof(cpumask_t), "");
 
 	spin_lock_irqsave(&pool->last_cpu_lock, flags);
-	cpu = next_cpu(pool->last_cpu, cpu_online_map);
-	if (cpu == NR_CPUS)
+	cpu = next_cpu_nr(pool->last_cpu, cpu_online_map);
+	if (cpu >= nr_cpu_ids)
 		cpu = first_cpu(cpu_online_map);
 	pool->last_cpu = cpu;
 	spin_unlock_irqrestore(&pool->last_cpu_lock, flags);
diff --git a/drivers/misc/sgi-xp/xpc_main.c b/drivers/misc/sgi-xp/xpc_main.c
index 08256ed0d9a6..579b01ff82d4 100644
--- a/drivers/misc/sgi-xp/xpc_main.c
+++ b/drivers/misc/sgi-xp/xpc_main.c
@@ -229,10 +229,11 @@ xpc_hb_checker(void *ignore)
 	int last_IRQ_count = 0;
 	int new_IRQ_count;
 	int force_IRQ = 0;
+	cpumask_of_cpu_ptr(cpumask, XPC_HB_CHECK_CPU);
 
 	/* this thread was marked active by xpc_hb_init() */
 
-	set_cpus_allowed(current, cpumask_of_cpu(XPC_HB_CHECK_CPU));
+	set_cpus_allowed_ptr(current, cpumask);
 
 	/* set our heartbeating to other partitions into motion */
 	xpc_hb_check_timeout = jiffies + (xpc_hb_check_interval * HZ);
diff --git a/include/asm-x86/ipi.h b/include/asm-x86/ipi.h
index 196d63c28aa4..bb1c09f7a76c 100644
--- a/include/asm-x86/ipi.h
+++ b/include/asm-x86/ipi.h
@@ -122,7 +122,7 @@ static inline void send_IPI_mask_sequence(cpumask_t mask, int vector)
 	 * - mbligh
 	 */
 	local_irq_save(flags);
-	for_each_cpu_mask(query_cpu, mask) {
+	for_each_cpu_mask_nr(query_cpu, mask) {
 		__send_IPI_dest_field(per_cpu(x86_cpu_to_apicid, query_cpu),
 				      vector, APIC_DEST_PHYSICAL);
 	}
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 15cb82a44e89..5f58da401b43 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -134,7 +134,7 @@ extern __u32			cleared_cpu_caps[NCAPINTS];
 #ifdef CONFIG_SMP
 DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
 #define cpu_data(cpu)		per_cpu(cpu_info, cpu)
-#define current_cpu_data	cpu_data(smp_processor_id())
+#define current_cpu_data	__get_cpu_var(cpu_info)
 #else
 #define cpu_data(cpu)		boot_cpu_data
 #define current_cpu_data	boot_cpu_data
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index c24875bd9c5b..30d59d1d0626 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -17,6 +17,20 @@
  * For details of cpus_onto(), see bitmap_onto in lib/bitmap.c.
  * For details of cpus_fold(), see bitmap_fold in lib/bitmap.c.
  *
+ * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
+ * Note: The alternate operations with the suffix "_nr" are used
+ *       to limit the range of the loop to nr_cpu_ids instead of
+ *       NR_CPUS when NR_CPUS > 64 for performance reasons.
+ *       If NR_CPUS is <= 64 then most assembler bitmask
+ *       operators execute faster with a constant range, so
+ *       the operator will continue to use NR_CPUS.
+ *
+ *       Another consideration is that nr_cpu_ids is initialized
+ *       to NR_CPUS and isn't lowered until the possible cpus are
+ *       discovered (including any disabled cpus).  So early uses
+ *       will span the entire range of NR_CPUS.
+ * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
+ *
  * The available cpumask operations are:
  *
  * void cpu_set(cpu, mask)		turn on bit 'cpu' in mask
@@ -38,18 +52,60 @@
  * int cpus_empty(mask)			Is mask empty (no bits sets)?
  * int cpus_full(mask)			Is mask full (all bits sets)?
  * int cpus_weight(mask)		Hamming weigh - number of set bits
+ * int cpus_weight_nr(mask)		Same using nr_cpu_ids instead of NR_CPUS
  *
  * void cpus_shift_right(dst, src, n)	Shift right
  * void cpus_shift_left(dst, src, n)	Shift left
  *
  * int first_cpu(mask)			Number lowest set bit, or NR_CPUS
  * int next_cpu(cpu, mask)		Next cpu past 'cpu', or NR_CPUS
+ * int next_cpu_nr(cpu, mask)		Next cpu past 'cpu', or nr_cpu_ids
  *
  * cpumask_t cpumask_of_cpu(cpu)	Return cpumask with bit 'cpu' set
+ *ifdef CONFIG_HAS_CPUMASK_OF_CPU
+ * cpumask_of_cpu_ptr_declare(v)	Declares cpumask_t *v
+ * cpumask_of_cpu_ptr_next(v, cpu)	Sets v = &cpumask_of_cpu_map[cpu]
+ * cpumask_of_cpu_ptr(v, cpu)		Combines above two operations
+ *else
+ * cpumask_of_cpu_ptr_declare(v)	Declares cpumask_t _v and *v = &_v
+ * cpumask_of_cpu_ptr_next(v, cpu)	Sets _v = cpumask_of_cpu(cpu)
+ * cpumask_of_cpu_ptr(v, cpu)		Combines above two operations
+ *endif
  * CPU_MASK_ALL				Initializer - all bits set
  * CPU_MASK_NONE			Initializer - no bits set
  * unsigned long *cpus_addr(mask)	Array of unsigned long's in mask
  *
+ * CPUMASK_ALLOC kmalloc's a structure that is a composite of many cpumask_t
+ * variables, and CPUMASK_PTR provides pointers to each field.
+ *
+ * The structure should be defined something like this:
+ * struct my_cpumasks {
+ *	cpumask_t mask1;
+ *	cpumask_t mask2;
+ * };
+ *
+ * Usage is then:
+ *	CPUMASK_ALLOC(my_cpumasks);
+ *	CPUMASK_PTR(mask1, my_cpumasks);
+ *	CPUMASK_PTR(mask2, my_cpumasks);
+ *
+ *	--- DO NOT reference cpumask_t pointers until this check ---
+ *	if (my_cpumasks == NULL)
+ *		"kmalloc failed"...
+ *
+ * References are now pointers to the cpumask_t variables (*mask1, ...)
+ *
+ *if NR_CPUS > BITS_PER_LONG
+ *   CPUMASK_ALLOC(m)			Declares and allocates struct m *m =
+ *						kmalloc(sizeof(*m), GFP_KERNEL)
+ *   CPUMASK_FREE(m)			Macro for kfree(m)
+ *else
+ *   CPUMASK_ALLOC(m)			Declares struct m _m, *m = &_m
+ *   CPUMASK_FREE(m)			Nop
+ *endif
+ *   CPUMASK_PTR(v, m)			Declares cpumask_t *v = &(m->v)
+ * ------------------------------------------------------------------------
+ *
  * int cpumask_scnprintf(buf, len, mask) Format cpumask for printing
  * int cpumask_parse_user(ubuf, ulen, mask)	Parse ascii string as cpumask
  * int cpulist_scnprintf(buf, len, mask) Format cpumask as list for printing
@@ -59,7 +115,8 @@
  * void cpus_onto(dst, orig, relmap)	*dst = orig relative to relmap
  * void cpus_fold(dst, orig, sz)	dst bits = orig bits mod sz
  *
- * for_each_cpu_mask(cpu, mask)		for-loop cpu over mask
+ * for_each_cpu_mask(cpu, mask)		for-loop cpu over mask using NR_CPUS
+ * for_each_cpu_mask_nr(cpu, mask)	for-loop cpu over mask using nr_cpu_ids
  *
  * int num_online_cpus()		Number of online CPUs
  * int num_possible_cpus()		Number of all possible CPUs
@@ -216,23 +273,19 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 	bitmap_shift_left(dstp->bits, srcp->bits, n, nbits);
 }
 
-#ifdef CONFIG_SMP
-int __first_cpu(const cpumask_t *srcp);
-#define first_cpu(src) __first_cpu(&(src))
-int __next_cpu(int n, const cpumask_t *srcp);
-#define next_cpu(n, src) __next_cpu((n), &(src))
-#else
-#define first_cpu(src)		({ (void)(src); 0; })
-#define next_cpu(n, src)	({ (void)(src); 1; })
-#endif
 
 #ifdef CONFIG_HAVE_CPUMASK_OF_CPU_MAP
 extern cpumask_t *cpumask_of_cpu_map;
-#define cpumask_of_cpu(cpu)    (cpumask_of_cpu_map[cpu])
-
+#define cpumask_of_cpu(cpu)	(cpumask_of_cpu_map[cpu])
+#define	cpumask_of_cpu_ptr(v, cpu)					\
+		const cpumask_t *v = &cpumask_of_cpu(cpu)
+#define	cpumask_of_cpu_ptr_declare(v)					\
+		const cpumask_t *v
+#define cpumask_of_cpu_ptr_next(v, cpu)					\
+					v = &cpumask_of_cpu(cpu)
 #else
 #define cpumask_of_cpu(cpu)						\
-(*({									\
+({									\
 	typeof(_unused_cpumask_arg_) m;					\
 	if (sizeof(m) == sizeof(unsigned long)) {			\
 		m.bits[0] = 1UL<<(cpu);					\
@@ -240,8 +293,16 @@ extern cpumask_t *cpumask_of_cpu_map;
 		cpus_clear(m);						\
 		cpu_set((cpu), m);					\
 	}								\
-	&m;								\
-}))
+	m;								\
+})
+#define	cpumask_of_cpu_ptr(v, cpu) 					\
+		cpumask_t _##v = cpumask_of_cpu(cpu);			\
+		const cpumask_t *v = &_##v
+#define	cpumask_of_cpu_ptr_declare(v)					\
+		cpumask_t _##v;						\
+		const cpumask_t *v = &_##v
+#define cpumask_of_cpu_ptr_next(v, cpu)					\
+					_##v = cpumask_of_cpu(cpu)
 #endif
 
 #define CPU_MASK_LAST_WORD BITMAP_LAST_WORD_MASK(NR_CPUS)
@@ -281,6 +342,15 @@ extern cpumask_t cpu_mask_all;
 
 #define cpus_addr(src) ((src).bits)
 
+#if NR_CPUS > BITS_PER_LONG
+#define	CPUMASK_ALLOC(m)	struct m *m = kmalloc(sizeof(*m), GFP_KERNEL)
+#define	CPUMASK_FREE(m)		kfree(m)
+#else
+#define	CPUMASK_ALLOC(m)	struct m _m, *m = &_m
+#define	CPUMASK_FREE(m)
+#endif
+#define	CPUMASK_PTR(v, m) 	cpumask_t *v = &(m->v)
+
 #define cpumask_scnprintf(buf, len, src) \
 			__cpumask_scnprintf((buf), (len), &(src), NR_CPUS)
 static inline int __cpumask_scnprintf(char *buf, int len,
@@ -343,20 +413,49 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
 	bitmap_fold(dstp->bits, origp->bits, sz, nbits);
 }
 
-#if NR_CPUS > 1
-#define for_each_cpu_mask(cpu, mask)		\
-	for ((cpu) = first_cpu(mask);		\
-		(cpu) < NR_CPUS;		\
-		(cpu) = next_cpu((cpu), (mask)))
-#else /* NR_CPUS == 1 */
-#define for_each_cpu_mask(cpu, mask)		\
+#if NR_CPUS == 1
+
+#define nr_cpu_ids		1
+#define first_cpu(src)		({ (void)(src); 0; })
+#define next_cpu(n, src)	({ (void)(src); 1; })
+#define any_online_cpu(mask)	0
+#define for_each_cpu_mask(cpu, mask)	\
 	for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask)
-#endif /* NR_CPUS */
+
+#else /* NR_CPUS > 1 */
+
+extern int nr_cpu_ids;
+int __first_cpu(const cpumask_t *srcp);
+int __next_cpu(int n, const cpumask_t *srcp);
+int __any_online_cpu(const cpumask_t *mask);
+
+#define first_cpu(src)		__first_cpu(&(src))
+#define next_cpu(n, src)	__next_cpu((n), &(src))
+#define any_online_cpu(mask) __any_online_cpu(&(mask))
+#define for_each_cpu_mask(cpu, mask)			\
+	for ((cpu) = -1;				\
+		(cpu) = next_cpu((cpu), (mask)),	\
+		(cpu) < NR_CPUS; )
+#endif
+
+#if NR_CPUS <= 64
 
 #define next_cpu_nr(n, src)		next_cpu(n, src)
 #define cpus_weight_nr(cpumask)		cpus_weight(cpumask)
 #define for_each_cpu_mask_nr(cpu, mask)	for_each_cpu_mask(cpu, mask)
 
+#else /* NR_CPUS > 64 */
+
+int __next_cpu_nr(int n, const cpumask_t *srcp);
+#define next_cpu_nr(n, src)	__next_cpu_nr((n), &(src))
+#define cpus_weight_nr(cpumask)	__cpus_weight(&(cpumask), nr_cpu_ids)
+#define for_each_cpu_mask_nr(cpu, mask)			\
+	for ((cpu) = -1;				\
+		(cpu) = next_cpu_nr((cpu), (mask)),	\
+		(cpu) < nr_cpu_ids; )
+
+#endif /* NR_CPUS > 64 */
+
 /*
  * The following particular system cpumasks and operations manage
  * possible, present and online cpus.  Each of them is a fixed size
@@ -418,9 +517,9 @@ extern cpumask_t cpu_online_map;
 extern cpumask_t cpu_present_map;
 
 #if NR_CPUS > 1
-#define num_online_cpus()	cpus_weight(cpu_online_map)
-#define num_possible_cpus()	cpus_weight(cpu_possible_map)
-#define num_present_cpus()	cpus_weight(cpu_present_map)
+#define num_online_cpus()	cpus_weight_nr(cpu_online_map)
+#define num_possible_cpus()	cpus_weight_nr(cpu_possible_map)
+#define num_present_cpus()	cpus_weight_nr(cpu_present_map)
 #define cpu_online(cpu)		cpu_isset((cpu), cpu_online_map)
 #define cpu_possible(cpu)	cpu_isset((cpu), cpu_possible_map)
 #define cpu_present(cpu)	cpu_isset((cpu), cpu_present_map)
@@ -435,17 +534,8 @@ extern cpumask_t cpu_present_map;
 
 #define cpu_is_offline(cpu)	unlikely(!cpu_online(cpu))
 
-#ifdef CONFIG_SMP
-extern int nr_cpu_ids;
-#define any_online_cpu(mask) __any_online_cpu(&(mask))
-int __any_online_cpu(const cpumask_t *mask);
-#else
-#define nr_cpu_ids			1
-#define any_online_cpu(mask)		0
-#endif
-
-#define for_each_possible_cpu(cpu)  for_each_cpu_mask((cpu), cpu_possible_map)
-#define for_each_online_cpu(cpu)  for_each_cpu_mask((cpu), cpu_online_map)
-#define for_each_present_cpu(cpu) for_each_cpu_mask((cpu), cpu_present_map)
+#define for_each_possible_cpu(cpu) for_each_cpu_mask_nr((cpu), cpu_possible_map)
+#define for_each_online_cpu(cpu)   for_each_cpu_mask_nr((cpu), cpu_online_map)
+#define for_each_present_cpu(cpu)  for_each_cpu_mask_nr((cpu), cpu_present_map)
 
 #endif /* __LINUX_CPUMASK_H */
diff --git a/kernel/cpu.c b/kernel/cpu.c
index cfb1d43ab801..d26d0b095b3b 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -413,7 +413,7 @@ void __ref enable_nonboot_cpus(void)
 		goto out;
 
 	printk("Enabling non-boot CPUs ...\n");
-	for_each_cpu_mask(cpu, frozen_cpus) {
+	for_each_cpu_mask_nr(cpu, frozen_cpus) {
 		error = _cpu_up(cpu, 1);
 		if (!error) {
 			printk("CPU%d is up\n", cpu);
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index 16eeeaa9d618..6f8696c502f4 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -106,7 +106,7 @@ static void force_quiescent_state(struct rcu_data *rdp,
 		 */
 		cpus_and(cpumask, rcp->cpumask, cpu_online_map);
 		cpu_clear(rdp->cpu, cpumask);
-		for_each_cpu_mask(cpu, cpumask)
+		for_each_cpu_mask_nr(cpu, cpumask)
 			smp_send_reschedule(cpu);
 	}
 }
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index 6f62b77d93c4..27827931ca0d 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -756,7 +756,7 @@ rcu_try_flip_idle(void)
 
 	/* Now ask each CPU for acknowledgement of the flip. */
 
-	for_each_cpu_mask(cpu, rcu_cpu_online_map) {
+	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map) {
 		per_cpu(rcu_flip_flag, cpu) = rcu_flipped;
 		dyntick_save_progress_counter(cpu);
 	}
@@ -774,7 +774,7 @@ rcu_try_flip_waitack(void)
 	int cpu;
 
 	RCU_TRACE_ME(rcupreempt_trace_try_flip_a1);
-	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map)
 		if (rcu_try_flip_waitack_needed(cpu) &&
 		    per_cpu(rcu_flip_flag, cpu) != rcu_flip_seen) {
 			RCU_TRACE_ME(rcupreempt_trace_try_flip_ae1);
@@ -806,7 +806,7 @@ rcu_try_flip_waitzero(void)
 	/* Check to see if the sum of the "last" counters is zero. */
 
 	RCU_TRACE_ME(rcupreempt_trace_try_flip_z1);
-	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map)
 		sum += RCU_DATA_CPU(cpu)->rcu_flipctr[lastidx];
 	if (sum != 0) {
 		RCU_TRACE_ME(rcupreempt_trace_try_flip_ze1);
@@ -821,7 +821,7 @@ rcu_try_flip_waitzero(void)
 	smp_mb();  /*  ^^^^^^^^^^^^ */
 
 	/* Call for a memory barrier from each CPU. */
-	for_each_cpu_mask(cpu, rcu_cpu_online_map) {
+	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map) {
 		per_cpu(rcu_mb_flag, cpu) = rcu_mb_needed;
 		dyntick_save_progress_counter(cpu);
 	}
@@ -841,7 +841,7 @@ rcu_try_flip_waitmb(void)
 	int cpu;
 
 	RCU_TRACE_ME(rcupreempt_trace_try_flip_m1);
-	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map)
 		if (rcu_try_flip_waitmb_needed(cpu) &&
 		    per_cpu(rcu_mb_flag, cpu) != rcu_mb_done) {
 			RCU_TRACE_ME(rcupreempt_trace_try_flip_me1);
diff --git a/kernel/sched.c b/kernel/sched.c
index b1104ea5d255..df80bae68152 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2108,7 +2108,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
 		/* Tally up the load of all CPUs in the group */
 		avg_load = 0;
 
-		for_each_cpu_mask(i, group->cpumask) {
+		for_each_cpu_mask_nr(i, group->cpumask) {
 			/* Bias balancing toward cpus of our domain */
 			if (local_group)
 				load = source_load(i, load_idx);
@@ -2150,7 +2150,7 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu,
 	/* Traverse only the allowed CPUs */
 	cpus_and(*tmp, group->cpumask, p->cpus_allowed);
 
-	for_each_cpu_mask(i, *tmp) {
+	for_each_cpu_mask_nr(i, *tmp) {
 		load = weighted_cpuload(i);
 
 		if (load < min_load || (load == min_load && i == this_cpu)) {
@@ -3168,7 +3168,7 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		max_cpu_load = 0;
 		min_cpu_load = ~0UL;
 
-		for_each_cpu_mask(i, group->cpumask) {
+		for_each_cpu_mask_nr(i, group->cpumask) {
 			struct rq *rq;
 
 			if (!cpu_isset(i, *cpus))
@@ -3447,7 +3447,7 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
 	unsigned long max_load = 0;
 	int i;
 
-	for_each_cpu_mask(i, group->cpumask) {
+	for_each_cpu_mask_nr(i, group->cpumask) {
 		unsigned long wl;
 
 		if (!cpu_isset(i, *cpus))
@@ -3989,7 +3989,7 @@ static void run_rebalance_domains(struct softirq_action *h)
 		int balance_cpu;
 
 		cpu_clear(this_cpu, cpus);
-		for_each_cpu_mask(balance_cpu, cpus) {
+		for_each_cpu_mask_nr(balance_cpu, cpus) {
 			/*
 			 * If this cpu gets work to do, stop the load balancing
 			 * work being done for other cpus. Next load
@@ -6802,7 +6802,7 @@ init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map,
 
 	cpus_clear(*covered);
 
-	for_each_cpu_mask(i, *span) {
+	for_each_cpu_mask_nr(i, *span) {
 		struct sched_group *sg;
 		int group = group_fn(i, cpu_map, &sg, tmpmask);
 		int j;
@@ -6813,7 +6813,7 @@ init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map,
 		cpus_clear(sg->cpumask);
 		sg->__cpu_power = 0;
 
-		for_each_cpu_mask(j, *span) {
+		for_each_cpu_mask_nr(j, *span) {
 			if (group_fn(j, cpu_map, NULL, tmpmask) != group)
 				continue;
 
@@ -7013,7 +7013,7 @@ static void init_numa_sched_groups_power(struct sched_group *group_head)
 	if (!sg)
 		return;
 	do {
-		for_each_cpu_mask(j, sg->cpumask) {
+		for_each_cpu_mask_nr(j, sg->cpumask) {
 			struct sched_domain *sd;
 
 			sd = &per_cpu(phys_domains, j);
@@ -7038,7 +7038,7 @@ static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask)
 {
 	int cpu, i;
 
-	for_each_cpu_mask(cpu, *cpu_map) {
+	for_each_cpu_mask_nr(cpu, *cpu_map) {
 		struct sched_group **sched_group_nodes
 			= sched_group_nodes_bycpu[cpu];
 
@@ -7277,7 +7277,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 	/*
 	 * Set up domains for cpus specified by the cpu_map.
 	 */
-	for_each_cpu_mask(i, *cpu_map) {
+	for_each_cpu_mask_nr(i, *cpu_map) {
 		struct sched_domain *sd = NULL, *p;
 		SCHED_CPUMASK_VAR(nodemask, allmasks);
 
@@ -7344,7 +7344,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 
 #ifdef CONFIG_SCHED_SMT
 	/* Set up CPU (sibling) groups */
-	for_each_cpu_mask(i, *cpu_map) {
+	for_each_cpu_mask_nr(i, *cpu_map) {
 		SCHED_CPUMASK_VAR(this_sibling_map, allmasks);
 		SCHED_CPUMASK_VAR(send_covered, allmasks);
 
@@ -7361,7 +7361,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 
 #ifdef CONFIG_SCHED_MC
 	/* Set up multi-core groups */
-	for_each_cpu_mask(i, *cpu_map) {
+	for_each_cpu_mask_nr(i, *cpu_map) {
 		SCHED_CPUMASK_VAR(this_core_map, allmasks);
 		SCHED_CPUMASK_VAR(send_covered, allmasks);
 
@@ -7428,7 +7428,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 			goto error;
 		}
 		sched_group_nodes[i] = sg;
-		for_each_cpu_mask(j, *nodemask) {
+		for_each_cpu_mask_nr(j, *nodemask) {
 			struct sched_domain *sd;
 
 			sd = &per_cpu(node_domains, j);
@@ -7474,21 +7474,21 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 
 	/* Calculate CPU power for physical packages and nodes */
 #ifdef CONFIG_SCHED_SMT
-	for_each_cpu_mask(i, *cpu_map) {
+	for_each_cpu_mask_nr(i, *cpu_map) {
 		struct sched_domain *sd = &per_cpu(cpu_domains, i);
 
 		init_sched_groups_power(i, sd);
 	}
 #endif
 #ifdef CONFIG_SCHED_MC
-	for_each_cpu_mask(i, *cpu_map) {
+	for_each_cpu_mask_nr(i, *cpu_map) {
 		struct sched_domain *sd = &per_cpu(core_domains, i);
 
 		init_sched_groups_power(i, sd);
 	}
 #endif
 
-	for_each_cpu_mask(i, *cpu_map) {
+	for_each_cpu_mask_nr(i, *cpu_map) {
 		struct sched_domain *sd = &per_cpu(phys_domains, i);
 
 		init_sched_groups_power(i, sd);
@@ -7508,7 +7508,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 #endif
 
 	/* Attach the domains */
-	for_each_cpu_mask(i, *cpu_map) {
+	for_each_cpu_mask_nr(i, *cpu_map) {
 		struct sched_domain *sd;
 #ifdef CONFIG_SCHED_SMT
 		sd = &per_cpu(cpu_domains, i);
@@ -7603,7 +7603,7 @@ static void detach_destroy_domains(const cpumask_t *cpu_map)
 
 	unregister_sched_domain_sysctl();
 
-	for_each_cpu_mask(i, *cpu_map)
+	for_each_cpu_mask_nr(i, *cpu_map)
 		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
 	arch_destroy_sched_domains(cpu_map, &tmpmask);
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index f2aa987027d6..bb61fe26b62c 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1031,7 +1031,7 @@ static int wake_idle(int cpu, struct task_struct *p)
 		    || ((sd->flags & SD_WAKE_IDLE_FAR)
 			&& !task_hot(p, task_rq(p)->clock, sd))) {
 			cpus_and(tmp, sd->span, p->cpus_allowed);
-			for_each_cpu_mask(i, tmp) {
+			for_each_cpu_mask_nr(i, tmp) {
 				if (idle_cpu(i)) {
 					if (i != task_cpu(p)) {
 						schedstat_inc(p,
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 47ceac9e8552..7c9614728c59 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -240,7 +240,7 @@ static int do_balance_runtime(struct rt_rq *rt_rq)
 
 	spin_lock(&rt_b->rt_runtime_lock);
 	rt_period = ktime_to_ns(rt_b->rt_period);
-	for_each_cpu_mask(i, rd->span) {
+	for_each_cpu_mask_nr(i, rd->span) {
 		struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i);
 		s64 diff;
 
@@ -1107,7 +1107,7 @@ static int pull_rt_task(struct rq *this_rq)
 
 	next = pick_next_task_rt(this_rq);
 
-	for_each_cpu_mask(cpu, this_rq->rd->rto_mask) {
+	for_each_cpu_mask_nr(cpu, this_rq->rd->rto_mask) {
 		if (this_cpu == cpu)
 			continue;
 
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index ba9b2054ecbd..738b411ff2d3 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -33,8 +33,9 @@ static int stopmachine(void *cpu)
 {
 	int irqs_disabled = 0;
 	int prepared = 0;
+	cpumask_of_cpu_ptr(cpumask, (int)(long)cpu);
 
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu((int)(long)cpu));
+	set_cpus_allowed_ptr(current, cpumask);
 
 	/* Ack: we are alive */
 	smp_mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index 4a23517169a6..06b17547f4e7 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -301,7 +301,7 @@ static int add_del_listener(pid_t pid, cpumask_t *maskp, int isadd)
 		return -EINVAL;
 
 	if (isadd == REGISTER) {
-		for_each_cpu_mask(cpu, mask) {
+		for_each_cpu_mask_nr(cpu, mask) {
 			s = kmalloc_node(sizeof(struct listener), GFP_KERNEL,
 					 cpu_to_node(cpu));
 			if (!s)
@@ -320,7 +320,7 @@ static int add_del_listener(pid_t pid, cpumask_t *maskp, int isadd)
 
 	/* Deregister or cleanup */
 cleanup:
-	for_each_cpu_mask(cpu, mask) {
+	for_each_cpu_mask_nr(cpu, mask) {
 		listeners = &per_cpu(listener_array, cpu);
 		down_write(&listeners->sem);
 		list_for_each_entry_safe(s, tmp, &listeners->list, list) {
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index b1c2da81b050..093d4acf993b 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -145,9 +145,9 @@ static void clocksource_watchdog(unsigned long data)
 		 * Cycle through CPUs to check if the CPUs stay
 		 * synchronized to each other.
 		 */
-		int next_cpu = next_cpu(raw_smp_processor_id(), cpu_online_map);
+		int next_cpu = next_cpu_nr(raw_smp_processor_id(), cpu_online_map);
 
-		if (next_cpu >= NR_CPUS)
+		if (next_cpu >= nr_cpu_ids)
 			next_cpu = first_cpu(cpu_online_map);
 		watchdog_timer.expires += WATCHDOG_INTERVAL;
 		add_timer_on(&watchdog_timer, next_cpu);
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index f48d0f09d32f..31463d370b94 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -399,8 +399,7 @@ again:
 	mask = CPU_MASK_NONE;
 	now = ktime_get();
 	/* Find all expired events */
-	for (cpu = first_cpu(tick_broadcast_oneshot_mask); cpu != NR_CPUS;
-	     cpu = next_cpu(cpu, tick_broadcast_oneshot_mask)) {
+	for_each_cpu_mask_nr(cpu, tick_broadcast_oneshot_mask) {
 		td = &per_cpu(tick_cpu_device, cpu);
 		if (td->evtdev->next_event.tv64 <= now.tv64)
 			cpu_set(cpu, mask);
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 4f3886562b8c..bf43284d6855 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -135,7 +135,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
  */
 static void tick_setup_device(struct tick_device *td,
 			      struct clock_event_device *newdev, int cpu,
-			      cpumask_t cpumask)
+			      const cpumask_t *cpumask)
 {
 	ktime_t next_event;
 	void (*handler)(struct clock_event_device *) = NULL;
@@ -169,8 +169,8 @@ static void tick_setup_device(struct tick_device *td,
 	 * When the device is not per cpu, pin the interrupt to the
 	 * current cpu:
 	 */
-	if (!cpus_equal(newdev->cpumask, cpumask))
-		irq_set_affinity(newdev->irq, cpumask);
+	if (!cpus_equal(newdev->cpumask, *cpumask))
+		irq_set_affinity(newdev->irq, *cpumask);
 
 	/*
 	 * When global broadcasting is active, check if the current
@@ -196,20 +196,20 @@ static int tick_check_new_device(struct clock_event_device *newdev)
 	struct tick_device *td;
 	int cpu, ret = NOTIFY_OK;
 	unsigned long flags;
-	cpumask_t cpumask;
+	cpumask_of_cpu_ptr_declare(cpumask);
 
 	spin_lock_irqsave(&tick_device_lock, flags);
 
 	cpu = smp_processor_id();
+	cpumask_of_cpu_ptr_next(cpumask, cpu);
 	if (!cpu_isset(cpu, newdev->cpumask))
 		goto out_bc;
 
 	td = &per_cpu(tick_cpu_device, cpu);
 	curdev = td->evtdev;
-	cpumask = cpumask_of_cpu(cpu);
 
 	/* cpu local device ? */
-	if (!cpus_equal(newdev->cpumask, cpumask)) {
+	if (!cpus_equal(newdev->cpumask, *cpumask)) {
 
 		/*
 		 * If the cpu affinity of the device interrupt can not
@@ -222,7 +222,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
 		 * If we have a cpu local device already, do not replace it
 		 * by a non cpu local device
 		 */
-		if (curdev && cpus_equal(curdev->cpumask, cpumask))
+		if (curdev && cpus_equal(curdev->cpumask, *cpumask))
 			goto out_bc;
 	}
 
diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c
index 2301e1e7c606..63528086337c 100644
--- a/kernel/trace/trace_sysprof.c
+++ b/kernel/trace/trace_sysprof.c
@@ -213,7 +213,9 @@ static void start_stack_timers(void)
 	int cpu;
 
 	for_each_online_cpu(cpu) {
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+		cpumask_of_cpu_ptr(new_mask, cpu);
+
+		set_cpus_allowed_ptr(current, new_mask);
 		start_stack_timer(cpu);
 	}
 	set_cpus_allowed_ptr(current, &saved_mask);
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index ce7799540c91..a6d36346d10a 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -397,7 +397,7 @@ void flush_workqueue(struct workqueue_struct *wq)
 	might_sleep();
 	lock_acquire(&wq->lockdep_map, 0, 0, 0, 2, _THIS_IP_);
 	lock_release(&wq->lockdep_map, 1, _THIS_IP_);
-	for_each_cpu_mask(cpu, *cpu_map)
+	for_each_cpu_mask_nr(cpu, *cpu_map)
 		flush_cpu_workqueue(per_cpu_ptr(wq->cpu_wq, cpu));
 }
 EXPORT_SYMBOL_GPL(flush_workqueue);
@@ -477,7 +477,7 @@ static void wait_on_work(struct work_struct *work)
 	wq = cwq->wq;
 	cpu_map = wq_cpu_map(wq);
 
-	for_each_cpu_mask(cpu, *cpu_map)
+	for_each_cpu_mask_nr(cpu, *cpu_map)
 		wait_on_cpu_work(per_cpu_ptr(wq->cpu_wq, cpu), work);
 }
 
@@ -813,7 +813,7 @@ void destroy_workqueue(struct workqueue_struct *wq)
 	list_del(&wq->list);
 	spin_unlock(&workqueue_lock);
 
-	for_each_cpu_mask(cpu, *cpu_map)
+	for_each_cpu_mask_nr(cpu, *cpu_map)
 		cleanup_workqueue_thread(per_cpu_ptr(wq->cpu_wq, cpu));
 	put_online_cpus();
 
diff --git a/lib/cpumask.c b/lib/cpumask.c
index bb4f76d3c3e7..5f97dc25ef9c 100644
--- a/lib/cpumask.c
+++ b/lib/cpumask.c
@@ -15,6 +15,15 @@ int __next_cpu(int n, const cpumask_t *srcp)
 }
 EXPORT_SYMBOL(__next_cpu);
 
+#if NR_CPUS > 64
+int __next_cpu_nr(int n, const cpumask_t *srcp)
+{
+	return min_t(int, nr_cpu_ids,
+				find_next_bit(srcp->bits, nr_cpu_ids, n+1));
+}
+EXPORT_SYMBOL(__next_cpu_nr);
+#endif
+
 int __any_online_cpu(const cpumask_t *mask)
 {
 	int cpu;
diff --git a/lib/smp_processor_id.c b/lib/smp_processor_id.c
index 3b4dc098181e..c4381d9516f6 100644
--- a/lib/smp_processor_id.c
+++ b/lib/smp_processor_id.c
@@ -11,7 +11,7 @@ notrace unsigned int debug_smp_processor_id(void)
 {
 	unsigned long preempt_count = preempt_count();
 	int this_cpu = raw_smp_processor_id();
-	cpumask_t this_mask;
+	cpumask_of_cpu_ptr_declare(this_mask);
 
 	if (likely(preempt_count))
 		goto out;
@@ -23,9 +23,9 @@ notrace unsigned int debug_smp_processor_id(void)
 	 * Kernel threads bound to a single CPU can safely use
 	 * smp_processor_id():
 	 */
-	this_mask = cpumask_of_cpu(this_cpu);
+	cpumask_of_cpu_ptr_next(this_mask, this_cpu);
 
-	if (cpus_equal(current->cpus_allowed, this_mask))
+	if (cpus_equal(current->cpus_allowed, *this_mask))
 		goto out;
 
 	/*
diff --git a/mm/allocpercpu.c b/mm/allocpercpu.c
index 05f2b4009ccc..843364594e23 100644
--- a/mm/allocpercpu.c
+++ b/mm/allocpercpu.c
@@ -35,7 +35,7 @@ EXPORT_SYMBOL_GPL(percpu_depopulate);
 void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
 {
 	int cpu;
-	for_each_cpu_mask(cpu, *mask)
+	for_each_cpu_mask_nr(cpu, *mask)
 		percpu_depopulate(__pdata, cpu);
 }
 EXPORT_SYMBOL_GPL(__percpu_depopulate_mask);
@@ -86,7 +86,7 @@ int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
 	int cpu;
 
 	cpus_clear(populated);
-	for_each_cpu_mask(cpu, *mask)
+	for_each_cpu_mask_nr(cpu, *mask)
 		if (unlikely(!percpu_populate(__pdata, size, gfp, cpu))) {
 			__percpu_depopulate_mask(__pdata, &populated);
 			return -ENOMEM;
diff --git a/mm/vmstat.c b/mm/vmstat.c
index db9eabb2c5b3..c3d4a781802f 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -26,7 +26,7 @@ static void sum_vm_events(unsigned long *ret, cpumask_t *cpumask)
 
 	memset(ret, 0, NR_VM_EVENT_ITEMS * sizeof(unsigned long));
 
-	for_each_cpu_mask(cpu, *cpumask) {
+	for_each_cpu_mask_nr(cpu, *cpumask) {
 		struct vm_event_state *this = &per_cpu(vm_event_states, cpu);
 
 		for (i = 0; i < NR_VM_EVENT_ITEMS; i++)
diff --git a/net/core/dev.c b/net/core/dev.c
index 6bf217da9d8f..7463a2150b09 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2398,7 +2398,7 @@ out:
 	 */
 	if (!cpus_empty(net_dma.channel_mask)) {
 		int chan_idx;
-		for_each_cpu_mask(chan_idx, net_dma.channel_mask) {
+		for_each_cpu_mask_nr(chan_idx, net_dma.channel_mask) {
 			struct dma_chan *chan = net_dma.channels[chan_idx];
 			if (chan)
 				dma_async_memcpy_issue_pending(chan);
@@ -4533,7 +4533,7 @@ static void net_dma_rebalance(struct net_dma *net_dma)
 	i = 0;
 	cpu = first_cpu(cpu_online_map);
 
-	for_each_cpu_mask(chan_idx, net_dma->channel_mask) {
+	for_each_cpu_mask_nr(chan_idx, net_dma->channel_mask) {
 		chan = net_dma->channels[chan_idx];
 
 		n = ((num_online_cpus() / cpus_weight(net_dma->channel_mask))
diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c
index 265b1b289a32..705959b31e24 100644
--- a/net/iucv/iucv.c
+++ b/net/iucv/iucv.c
@@ -497,7 +497,7 @@ static void iucv_setmask_up(void)
 	/* Disable all cpu but the first in cpu_irq_cpumask. */
 	cpumask = iucv_irq_cpumask;
 	cpu_clear(first_cpu(iucv_irq_cpumask), cpumask);
-	for_each_cpu_mask(cpu, cpumask)
+	for_each_cpu_mask_nr(cpu, cpumask)
 		smp_call_function_single(cpu, iucv_block_cpu, NULL, 1);
 }
 
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 5a32cb7c4bb4..835d27413083 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -310,7 +310,8 @@ svc_pool_map_set_cpumask(struct task_struct *task, unsigned int pidx)
 	switch (m->mode) {
 	case SVC_POOL_PERCPU:
 	{
-		set_cpus_allowed_ptr(task, &cpumask_of_cpu(node));
+		cpumask_of_cpu_ptr(cpumask, node);
+		set_cpus_allowed_ptr(task, cpumask);
 		break;
 	}
 	case SVC_POOL_PERNODE: