34 files changed, 175 insertions, 135 deletions
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index b0c8208df9fa..dd097b835839 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -202,7 +202,7 @@ static void drv_write(struct drv_cmd *cmd)
        cpumask_t saved_mask = current->cpus_allowed;
        unsigned int i;
-        for_each_cpu_mask(i, cmd->mask) {
+        for_each_cpu_mask_nr(i, cmd->mask) {
                set_cpus_allowed_ptr(current, &cpumask_of_cpu(i));
                do_drv_write(cmd);
        }
@@ -451,7 +451,7 @@ 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 +466,7 @@ 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..c45ca6d4dce1 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -966,7 +966,7 @@ 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 +974,7 @@ 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 +997,7 @@ 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 +1005,7 @@ 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);
        }
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index 908dd347c67e..8b0dd6f2a1ac 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -476,7 +476,7 @@ static int centrino_target (struct cpufreq_policy *policy,
        saved_mask = current->cpus_allowed;
        first_cpu = 1;
        cpus_clear(covered_cpus);
-        for_each_cpu_mask(j, online_policy_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
@@ -517,7 +517,7 @@ 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);
@@ -540,7 +540,7 @@ static int centrino_target (struct cpufreq_policy *policy,
                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);
        }
@@ -554,7 +554,7 @@ static int centrino_target (struct cpufreq_policy *policy,
                 */
                if (!cpus_empty(covered_cpus)) {
-                        for_each_cpu_mask(j, covered_cpus) {
+                        for_each_cpu_mask_nr(j, covered_cpus) {
                                set_cpus_allowed_ptr(current,
                                                     &cpumask_of_cpu(j));
                                wrmsr(MSR_IA32_PERF_CTL, oldmsr, h);
@@ -564,7 +564,7 @@ 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);
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
index 1b50244b1fdf..191f7263c61d 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
@@ -279,7 +279,7 @@ 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 +292,7 @@ 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 26d615dcb149..bfade3301c3a 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -488,7 +488,7 @@ 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);
        }
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/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index ef1a8dfcc529..e2838cbd2ff8 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -718,7 +718,7 @@ 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;
@@ -739,7 +739,7 @@ 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! */
@@ -749,7 +749,7 @@ 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;
@@ -781,7 +781,7 @@ 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;
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 3e1cecedde42..197300bfebd2 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -487,7 +487,7 @@ 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));
@@ -510,7 +510,7 @@ 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);
@@ -1300,7 +1300,7 @@ 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
@@ -1309,7 +1309,7 @@ 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 94e69000f982..7a70638797ed 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -345,7 +345,7 @@ 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);
 }
@@ -413,7 +413,7 @@ int xen_smp_call_function_mask(cpumask_t mask, void (*func)(void *),
        /* Make sure other vcpus get a chance to run if they need to. */
        yield = false;
-        for_each_cpu_mask(cpu, mask)
+        for_each_cpu_mask_nr(cpu, mask)
                if (xen_vcpu_stolen(cpu))
                        yield = true;
diff --git a/drivers/acpi/processor_throttling.c b/drivers/acpi/processor_throttling.c
index bb06738860c4..28509fbba6f9 100644
--- a/drivers/acpi/processor_throttling.c
+++ b/drivers/acpi/processor_throttling.c
@@ -1013,7 +1013,7 @@ 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);
@@ -1034,7 +1034,7 @@ 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 = processors[i];
                        /*
                         * If the pointer is invalid, we will report the
@@ -1068,7 +1068,7 @@ 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 e38dfed41d80..5000402ae092 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -119,14 +119,14 @@ 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 1d41496ed2f8..d8f8b1e0edde 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) {
                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)
                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)
         * 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;
                        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/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c
index ce1ab0571be3..43180b952c1f 100644
--- a/drivers/infiniband/hw/ehca/ehca_irq.c
+++ b/drivers/infiniband/hw/ehca/ehca_irq.c
@@ -641,8 +641,8 @@ 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);
+        cpu = next_cpu_nr(pool->last_cpu, cpu_online_map);
-        if (cpu == NR_CPUS)
+        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/include/asm-x86/ipi.h b/include/asm-x86/ipi.h
index ecc80f341f37..5f7310aa3efd 100644
--- a/include/asm-x86/ipi.h
+++ b/include/asm-x86/ipi.h
@@ -121,7 +121,7 @@ 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/linux/cpumask.h b/include/linux/cpumask.h
index c24875bd9c5b..47418b1b4103 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
@@ -38,12 +52,14 @@
 * 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
 * CPU_MASK_ALL                         Initializer - all bits set
@@ -59,7 +75,8 @@
 * 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
@@ -216,15 +233,6 @@ 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;
@@ -343,15 +351,48 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
        bitmap_fold(dstp->bits, origp->bits, sz, nbits);
 }
-#if NR_CPUS > 1
+#if NR_CPUS == 1
-#define for_each_cpu_mask(cpu, mask)            \
-        for ((cpu) = first_cpu(mask);           \
+#define nr_cpu_ids              1
-                (cpu) < NR_CPUS;                \
+#define first_cpu(src)          ({ (void)(src); 0; })
-                (cpu) = next_cpu((cpu), (mask)))
+#define next_cpu(n, src)        ({ (void)(src); 1; })
-#else /* NR_CPUS == 1 */
+#define any_online_cpu(mask)    0
-#define for_each_cpu_mask(cpu, mask)            \
+#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 */
 #define next_cpu_nr(n, src)             next_cpu(n, src)
 #define cpus_weight_nr(cpumask)         cpus_weight(cpumask)
@@ -418,9 +459,9 @@ 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_online_cpus()       cpus_weight_nr(cpu_online_map)
-#define num_possible_cpus()     cpus_weight(cpu_possible_map)
+#define num_possible_cpus()     cpus_weight_nr(cpu_possible_map)
-#define num_present_cpus()      cpus_weight(cpu_present_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 +476,8 @@ extern cpumask_t cpu_present_map;
 #define cpu_is_offline(cpu)     unlikely(!cpu_online(cpu))
-#ifdef CONFIG_SMP
+#define for_each_possible_cpu(cpu) for_each_cpu_mask_nr((cpu), cpu_possible_map)
-extern int nr_cpu_ids;
+#define for_each_online_cpu(cpu)   for_each_cpu_mask_nr((cpu), cpu_online_map)
-#define any_online_cpu(mask) __any_online_cpu(&(mask))
+#define for_each_present_cpu(cpu)  for_each_cpu_mask_nr((cpu), cpu_present_map)
-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)
 #endif /* __LINUX_CPUMASK_H */
diff --git a/kernel/cpu.c b/kernel/cpu.c
index c77bc3a1c722..50ae922c6022 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -390,7 +390,7 @@ 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 a38895a5b8e2..adde10388d0c 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 5e02b7740702..5cbd69edf5d9 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -655,7 +655,7 @@ 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);
        }
@@ -673,7 +673,7 @@ 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);
@@ -705,7 +705,7 @@ 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);
@@ -720,7 +720,7 @@ 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);
        }
@@ -740,7 +740,7 @@ 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 94ead43eda62..e6795e39c8ab 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1939,7 +1939,7 @@ 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);
@@ -1981,7 +1981,7 @@ 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)) {
@@ -2964,7 +2964,7 @@ 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))
@@ -3228,7 +3228,7 @@ 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))
@@ -3759,7 +3759,7 @@ 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
@@ -6491,7 +6491,7 @@ 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;
@@ -6502,7 +6502,7 @@ 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;
@@ -6538,9 +6538,9 @@ static int find_next_best_node(int node, nodemask_t *used_nodes)
        min_val = INT_MAX;
-        for (i = 0; i < MAX_NUMNODES; i++) {
+        for (i = 0; i < nr_node_ids; i++) {
                /* Start at @node */
-                n = (node + i) % MAX_NUMNODES;
+                n = (node + i) % nr_node_ids;
                if (!nr_cpus_node(n))
                        continue;
@@ -6702,7 +6702,7 @@ 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);
@@ -6727,14 +6727,14 @@ 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];
                if (!sched_group_nodes)
                        continue;
-                for (i = 0; i < MAX_NUMNODES; i++) {
+                for (i = 0; i < nr_node_ids; i++) {
                        struct sched_group *oldsg, *sg = sched_group_nodes[i];
                        *nodemask = node_to_cpumask(i);
@@ -6927,7 +6927,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
        /*
         * Allocate the per-node list of sched groups
         */
-        sched_group_nodes = kcalloc(MAX_NUMNODES, sizeof(struct sched_group *),
+        sched_group_nodes = kcalloc(nr_node_ids, sizeof(struct sched_group *),
                                    GFP_KERNEL);
        if (!sched_group_nodes) {
                printk(KERN_WARNING "Can not alloc sched group node list\n");
@@ -6966,7 +6966,7 @@ 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);
@@ -7033,7 +7033,7 @@ 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);
@@ -7050,7 +7050,7 @@ 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);
@@ -7066,7 +7066,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 #endif
        /* Set up physical groups */
-        for (i = 0; i < MAX_NUMNODES; i++) {
+        for (i = 0; i < nr_node_ids; i++) {
                SCHED_CPUMASK_VAR(nodemask, allmasks);
                SCHED_CPUMASK_VAR(send_covered, allmasks);
@@ -7090,7 +7090,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
                                        send_covered, tmpmask);
        }
-        for (i = 0; i < MAX_NUMNODES; i++) {
+        for (i = 0; i < nr_node_ids; i++) {
                /* Set up node groups */
                struct sched_group *sg, *prev;
                SCHED_CPUMASK_VAR(nodemask, allmasks);
@@ -7117,7 +7117,7 @@ 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);
@@ -7129,9 +7129,9 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
                cpus_or(*covered, *covered, *nodemask);
                prev = sg;
-                for (j = 0; j < MAX_NUMNODES; j++) {
+                for (j = 0; j < nr_node_ids; j++) {
                        SCHED_CPUMASK_VAR(notcovered, allmasks);
-                        int n = (i + j) % MAX_NUMNODES;
+                        int n = (i + j) % nr_node_ids;
                        node_to_cpumask_ptr(pnodemask, n);
                        cpus_complement(*notcovered, *covered);
@@ -7163,28 +7163,28 @@ 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);
        }
 #ifdef CONFIG_NUMA
-        for (i = 0; i < MAX_NUMNODES; i++)
+        for (i = 0; i < nr_node_ids; i++)
                init_numa_sched_groups_power(sched_group_nodes[i]);
        if (sd_allnodes) {
@@ -7197,7 +7197,7 @@ 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);
@@ -7292,7 +7292,7 @@ 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 08ae848b71d4..74774bde5264 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -961,7 +961,7 @@ 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 0f3c19197fa4..e757f370eb1b 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -231,7 +231,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
                return 1;
        span = sched_rt_period_mask();
-        for_each_cpu_mask(i, span) {
+        for_each_cpu_mask_nr(i, span) {
                int enqueue = 0;
                struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i);
                struct rq *rq = rq_of_rt_rq(rt_rq);
@@ -273,7 +273,7 @@ static int 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;
@@ -1006,7 +1006,7 @@ 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/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 dadde5361f32..60ceabd53f2e 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 57a1f02e5ec0..2d0a96346259 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -397,8 +397,7 @@ again:
        mask = CPU_MASK_NONE;
        now = ktime_get();
        /* Find all expired events */
-        for (cpu = first_cpu(tick_broadcast_oneshot_mask); cpu != NR_CPUS;
+        for_each_cpu_mask_nr(cpu, tick_broadcast_oneshot_mask) {
-             cpu = next_cpu(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/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/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 fca23a3bf12c..94d9d6f77e04 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2261,7 +2261,7 @@ 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);
@@ -4322,7 +4322,7 @@ 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 918970762131..8de511070593 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, 0, 1);
 }
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 01c7e311b904..d43cf8ddff67 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -603,7 +603,7 @@ __svc_create_thread(svc_thread_fn func, struct svc_serv *serv,
        error = kernel_thread((int (*)(void *)) func, rqstp, 0);
        if (have_oldmask)
-                set_cpus_allowed(current, oldmask);
+                set_cpus_allowed_ptr(current, &oldmask);
        if (error < 0)
                goto out_thread;