Merge branch 'cpus4096-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'cpus4096-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (66 commits)
  x86: export vector_used_by_percpu_irq
  x86: use logical apicid in x2apic_cluster's x2apic_cpu_mask_to_apicid_and()
  sched: nominate preferred wakeup cpu, fix
  x86: fix lguest used_vectors breakage, -v2
  x86: fix warning in arch/x86/kernel/io_apic.c
  sched: fix warning in kernel/sched.c
  sched: move test_sd_parent() to an SMP section of sched.h
  sched: add SD_BALANCE_NEWIDLE at MC and CPU level for sched_mc>0
  sched: activate active load balancing in new idle cpus
  sched: bias task wakeups to preferred semi-idle packages
  sched: nominate preferred wakeup cpu
  sched: favour lower logical cpu number for sched_mc balance
  sched: framework for sched_mc/smt_power_savings=N
  sched: convert BALANCE_FOR_xx_POWER to inline functions
  x86: use possible_cpus=NUM to extend the possible cpus allowed
  x86: fix cpu_mask_to_apicid_and to include cpu_online_mask
  x86: update io_apic.c to the new cpumask code
  x86: Introduce topology_core_cpumask()/topology_thread_cpumask()
  x86: xen: use smp_call_function_many()
  x86: use work_on_cpu in x86/kernel/cpu/mcheck/mce_amd_64.c
  ...

Fixed up trivial conflict in kernel/time/tick-sched.c manually

1 files changed, 556 insertions, 414 deletions

diff --git a/kernel/sched.c b/kernel/sched.c
index fff1c4a20b6..27ba1d642f0 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -498,18 +498,26 @@ struct rt_rq {
  */
 struct root_domain {
 	atomic_t refcount;
-	cpumask_t span;
-	cpumask_t online;
+	cpumask_var_t span;
+	cpumask_var_t online;
 
 	/*
 	 * The "RT overload" flag: it gets set if a CPU has more than
 	 * one runnable RT task.
 	 */
-	cpumask_t rto_mask;
+	cpumask_var_t rto_mask;
 	atomic_t rto_count;
 #ifdef CONFIG_SMP
 	struct cpupri cpupri;
 #endif
+#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
+	/*
+	 * Preferred wake up cpu nominated by sched_mc balance that will be
+	 * used when most cpus are idle in the system indicating overall very
+	 * low system utilisation. Triggered at POWERSAVINGS_BALANCE_WAKEUP(2)
+	 */
+	unsigned int sched_mc_preferred_wakeup_cpu;
+#endif
 };
 
 /*
@@ -1514,7 +1522,7 @@ static int tg_shares_up(struct task_group *tg, void *data)
 	struct sched_domain *sd = data;
 	int i;
 
-	for_each_cpu_mask(i, sd->span) {
+	for_each_cpu(i, sched_domain_span(sd)) {
 		/*
 		 * If there are currently no tasks on the cpu pretend there
 		 * is one of average load so that when a new task gets to
@@ -1535,7 +1543,7 @@ static int tg_shares_up(struct task_group *tg, void *data)
 	if (!sd->parent || !(sd->parent->flags & SD_LOAD_BALANCE))
 		shares = tg->shares;
 
-	for_each_cpu_mask(i, sd->span)
+	for_each_cpu(i, sched_domain_span(sd))
 		update_group_shares_cpu(tg, i, shares, rq_weight);
 
 	return 0;
@@ -2101,15 +2109,17 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
 		int i;
 
 		/* Skip over this group if it has no CPUs allowed */
-		if (!cpus_intersects(group->cpumask, p->cpus_allowed))
+		if (!cpumask_intersects(sched_group_cpus(group),
+					&p->cpus_allowed))
 			continue;
 
-		local_group = cpu_isset(this_cpu, group->cpumask);
+		local_group = cpumask_test_cpu(this_cpu,
+					       sched_group_cpus(group));
 
 		/* Tally up the load of all CPUs in the group */
 		avg_load = 0;
 
-		for_each_cpu_mask_nr(i, group->cpumask) {
+		for_each_cpu(i, sched_group_cpus(group)) {
 			/* Bias balancing toward cpus of our domain */
 			if (local_group)
 				load = source_load(i, load_idx);
@@ -2141,17 +2151,14 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
  * find_idlest_cpu - find the idlest cpu among the cpus in group.
  */
 static int
-find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu,
-		cpumask_t *tmp)
+find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 {
 	unsigned long load, min_load = ULONG_MAX;
 	int idlest = -1;
 	int i;
 
 	/* Traverse only the allowed CPUs */
-	cpus_and(*tmp, group->cpumask, p->cpus_allowed);
-
-	for_each_cpu_mask_nr(i, *tmp) {
+	for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) {
 		load = weighted_cpuload(i);
 
 		if (load < min_load || (load == min_load && i == this_cpu)) {
@@ -2193,7 +2200,6 @@ static int sched_balance_self(int cpu, int flag)
 		update_shares(sd);
 
 	while (sd) {
-		cpumask_t span, tmpmask;
 		struct sched_group *group;
 		int new_cpu, weight;
 
@@ -2202,14 +2208,13 @@ static int sched_balance_self(int cpu, int flag)
 			continue;
 		}
 
-		span = sd->span;
 		group = find_idlest_group(sd, t, cpu);
 		if (!group) {
 			sd = sd->child;
 			continue;
 		}
 
-		new_cpu = find_idlest_cpu(group, t, cpu, &tmpmask);
+		new_cpu = find_idlest_cpu(group, t, cpu);
 		if (new_cpu == -1 || new_cpu == cpu) {
 			/* Now try balancing at a lower domain level of cpu */
 			sd = sd->child;
@@ -2218,10 +2223,10 @@ static int sched_balance_self(int cpu, int flag)
 
 		/* Now try balancing at a lower domain level of new_cpu */
 		cpu = new_cpu;
+		weight = cpumask_weight(sched_domain_span(sd));
 		sd = NULL;
-		weight = cpus_weight(span);
 		for_each_domain(cpu, tmp) {
-			if (weight <= cpus_weight(tmp->span))
+			if (weight <= cpumask_weight(sched_domain_span(tmp)))
 				break;
 			if (tmp->flags & flag)
 				sd = tmp;
@@ -2266,7 +2271,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 		cpu = task_cpu(p);
 
 		for_each_domain(this_cpu, sd) {
-			if (cpu_isset(cpu, sd->span)) {
+			if (cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 				update_shares(sd);
 				break;
 			}
@@ -2315,7 +2320,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 	else {
 		struct sched_domain *sd;
 		for_each_domain(this_cpu, sd) {
-			if (cpu_isset(cpu, sd->span)) {
+			if (cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 				schedstat_inc(sd, ttwu_wake_remote);
 				break;
 			}
@@ -2846,7 +2851,7 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu)
 	struct rq *rq;
 
 	rq = task_rq_lock(p, &flags);
-	if (!cpu_isset(dest_cpu, p->cpus_allowed)
+	if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed)
 	    || unlikely(!cpu_active(dest_cpu)))
 		goto out;
 
@@ -2911,7 +2916,7 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu,
 	 * 2) cannot be migrated to this CPU due to cpus_allowed, or
 	 * 3) are cache-hot on their current CPU.
 	 */
-	if (!cpu_isset(this_cpu, p->cpus_allowed)) {
+	if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) {
 		schedstat_inc(p, se.nr_failed_migrations_affine);
 		return 0;
 	}
@@ -3086,7 +3091,7 @@ static int move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest,
 static struct sched_group *
 find_busiest_group(struct sched_domain *sd, int this_cpu,
 		   unsigned long *imbalance, enum cpu_idle_type idle,
-		   int *sd_idle, const cpumask_t *cpus, int *balance)
+		   int *sd_idle, const struct cpumask *cpus, int *balance)
 {
 	struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups;
 	unsigned long max_load, avg_load, total_load, this_load, total_pwr;
@@ -3122,10 +3127,11 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		unsigned long sum_avg_load_per_task;
 		unsigned long avg_load_per_task;
 
-		local_group = cpu_isset(this_cpu, group->cpumask);
+		local_group = cpumask_test_cpu(this_cpu,
+					       sched_group_cpus(group));
 
 		if (local_group)
-			balance_cpu = first_cpu(group->cpumask);
+			balance_cpu = cpumask_first(sched_group_cpus(group));
 
 		/* Tally up the load of all CPUs in the group */
 		sum_weighted_load = sum_nr_running = avg_load = 0;
@@ -3134,13 +3140,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		max_cpu_load = 0;
 		min_cpu_load = ~0UL;
 
-		for_each_cpu_mask_nr(i, group->cpumask) {
-			struct rq *rq;
-
-			if (!cpu_isset(i, *cpus))
-				continue;
-
-			rq = cpu_rq(i);
+		for_each_cpu_and(i, sched_group_cpus(group), cpus) {
+			struct rq *rq = cpu_rq(i);
 
 			if (*sd_idle && rq->nr_running)
 				*sd_idle = 0;
@@ -3251,8 +3252,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		 */
 		if ((sum_nr_running < min_nr_running) ||
 		    (sum_nr_running == min_nr_running &&
-		     first_cpu(group->cpumask) <
-		     first_cpu(group_min->cpumask))) {
+		     cpumask_first(sched_group_cpus(group)) >
+		     cpumask_first(sched_group_cpus(group_min)))) {
 			group_min = group;
 			min_nr_running = sum_nr_running;
 			min_load_per_task = sum_weighted_load /
@@ -3267,8 +3268,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		if (sum_nr_running <= group_capacity - 1) {
 			if (sum_nr_running > leader_nr_running ||
 			    (sum_nr_running == leader_nr_running &&
-			     first_cpu(group->cpumask) >
-			      first_cpu(group_leader->cpumask))) {
+			     cpumask_first(sched_group_cpus(group)) <
+			     cpumask_first(sched_group_cpus(group_leader)))) {
 				group_leader = group;
 				leader_nr_running = sum_nr_running;
 			}
@@ -3394,6 +3395,10 @@ out_balanced:
 
 	if (this == group_leader && group_leader != group_min) {
 		*imbalance = min_load_per_task;
+		if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP) {
+			cpu_rq(this_cpu)->rd->sched_mc_preferred_wakeup_cpu =
+				cpumask_first(sched_group_cpus(group_leader));
+		}
 		return group_min;
 	}
 #endif
@@ -3407,16 +3412,16 @@ ret:
  */
 static struct rq *
 find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
-		   unsigned long imbalance, const cpumask_t *cpus)
+		   unsigned long imbalance, const struct cpumask *cpus)
 {
 	struct rq *busiest = NULL, *rq;
 	unsigned long max_load = 0;
 	int i;
 
-	for_each_cpu_mask_nr(i, group->cpumask) {
+	for_each_cpu(i, sched_group_cpus(group)) {
 		unsigned long wl;
 
-		if (!cpu_isset(i, *cpus))
+		if (!cpumask_test_cpu(i, cpus))
 			continue;
 
 		rq = cpu_rq(i);
@@ -3446,7 +3451,7 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
  */
 static int load_balance(int this_cpu, struct rq *this_rq,
 			struct sched_domain *sd, enum cpu_idle_type idle,
-			int *balance, cpumask_t *cpus)
+			int *balance, struct cpumask *cpus)
 {
 	int ld_moved, all_pinned = 0, active_balance = 0, sd_idle = 0;
 	struct sched_group *group;
@@ -3454,7 +3459,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 	struct rq *busiest;
 	unsigned long flags;
 
-	cpus_setall(*cpus);
+	cpumask_setall(cpus);
 
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
@@ -3514,8 +3519,8 @@ redo:
 
 		/* All tasks on this runqueue were pinned by CPU affinity */
 		if (unlikely(all_pinned)) {
-			cpu_clear(cpu_of(busiest), *cpus);
-			if (!cpus_empty(*cpus))
+			cpumask_clear_cpu(cpu_of(busiest), cpus);
+			if (!cpumask_empty(cpus))
 				goto redo;
 			goto out_balanced;
 		}
@@ -3532,7 +3537,8 @@ redo:
 			/* don't kick the migration_thread, if the curr
 			 * task on busiest cpu can't be moved to this_cpu
 			 */
-			if (!cpu_isset(this_cpu, busiest->curr->cpus_allowed)) {
+			if (!cpumask_test_cpu(this_cpu,
+					      &busiest->curr->cpus_allowed)) {
 				spin_unlock_irqrestore(&busiest->lock, flags);
 				all_pinned = 1;
 				goto out_one_pinned;
@@ -3607,7 +3613,7 @@ out:
  */
 static int
 load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd,
-			cpumask_t *cpus)
+			struct cpumask *cpus)
 {
 	struct sched_group *group;
 	struct rq *busiest = NULL;
@@ -3616,7 +3622,7 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd,
 	int sd_idle = 0;
 	int all_pinned = 0;
 
-	cpus_setall(*cpus);
+	cpumask_setall(cpus);
 
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
@@ -3660,17 +3666,71 @@ redo:
 		double_unlock_balance(this_rq, busiest);
 
 		if (unlikely(all_pinned)) {
-			cpu_clear(cpu_of(busiest), *cpus);
-			if (!cpus_empty(*cpus))
+			cpumask_clear_cpu(cpu_of(busiest), cpus);
+			if (!cpumask_empty(cpus))
 				goto redo;
 		}
 	}
 
 	if (!ld_moved) {
+		int active_balance = 0;
+
 		schedstat_inc(sd, lb_failed[CPU_NEWLY_IDLE]);
 		if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
 		    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
 			return -1;
+
+		if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
+			return -1;
+
+		if (sd->nr_balance_failed++ < 2)
+			return -1;
+
+		/*
+		 * The only task running in a non-idle cpu can be moved to this
+		 * cpu in an attempt to completely freeup the other CPU
+		 * package. The same method used to move task in load_balance()
+		 * have been extended for load_balance_newidle() to speedup
+		 * consolidation at sched_mc=POWERSAVINGS_BALANCE_WAKEUP (2)
+		 *
+		 * The package power saving logic comes from
+		 * find_busiest_group().  If there are no imbalance, then
+		 * f_b_g() will return NULL.  However when sched_mc={1,2} then
+		 * f_b_g() will select a group from which a running task may be
+		 * pulled to this cpu in order to make the other package idle.
+		 * If there is no opportunity to make a package idle and if
+		 * there are no imbalance, then f_b_g() will return NULL and no
+		 * action will be taken in load_balance_newidle().
+		 *
+		 * Under normal task pull operation due to imbalance, there
+		 * will be more than one task in the source run queue and
+		 * move_tasks() will succeed.  ld_moved will be true and this
+		 * active balance code will not be triggered.
+		 */
+
+		/* Lock busiest in correct order while this_rq is held */
+		double_lock_balance(this_rq, busiest);
+
+		/*
+		 * don't kick the migration_thread, if the curr
+		 * task on busiest cpu can't be moved to this_cpu
+		 */
+		if (!cpu_isset(this_cpu, busiest->curr->cpus_allowed)) {
+			double_unlock_balance(this_rq, busiest);
+			all_pinned = 1;
+			return ld_moved;
+		}
+
+		if (!busiest->active_balance) {
+			busiest->active_balance = 1;
+			busiest->push_cpu = this_cpu;
+			active_balance = 1;
+		}
+
+		double_unlock_balance(this_rq, busiest);
+		if (active_balance)
+			wake_up_process(busiest->migration_thread);
+
 	} else
 		sd->nr_balance_failed = 0;
 
@@ -3696,7 +3756,10 @@ static void idle_balance(int this_cpu, struct rq *this_rq)
 	struct sched_domain *sd;
 	int pulled_task = 0;
 	unsigned long next_balance = jiffies + HZ;
-	cpumask_t tmpmask;
+	cpumask_var_t tmpmask;
+
+	if (!alloc_cpumask_var(&tmpmask, GFP_ATOMIC))
+		return;
 
 	for_each_domain(this_cpu, sd) {
 		unsigned long interval;
@@ -3707,7 +3770,7 @@ static void idle_balance(int this_cpu, struct rq *this_rq)
 		if (sd->flags & SD_BALANCE_NEWIDLE)
 			/* If we've pulled tasks over stop searching: */
 			pulled_task = load_balance_newidle(this_cpu, this_rq,
-							   sd, &tmpmask);
+							   sd, tmpmask);
 
 		interval = msecs_to_jiffies(sd->balance_interval);
 		if (time_after(next_balance, sd->last_balance + interval))
@@ -3722,6 +3785,7 @@ static void idle_balance(int this_cpu, struct rq *this_rq)
 		 */
 		this_rq->next_balance = next_balance;
 	}
+	free_cpumask_var(tmpmask);
 }
 
 /*
@@ -3759,7 +3823,7 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu)
 	/* Search for an sd spanning us and the target CPU. */
 	for_each_domain(target_cpu, sd) {
 		if ((sd->flags & SD_LOAD_BALANCE) &&
-		    cpu_isset(busiest_cpu, sd->span))
+		    cpumask_test_cpu(busiest_cpu, sched_domain_span(sd)))
 				break;
 	}
 
@@ -3778,10 +3842,9 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu)
 #ifdef CONFIG_NO_HZ
 static struct {
 	atomic_t load_balancer;
-	cpumask_t cpu_mask;
+	cpumask_var_t cpu_mask;
 } nohz ____cacheline_aligned = {
 	.load_balancer = ATOMIC_INIT(-1),
-	.cpu_mask = CPU_MASK_NONE,
 };
 
 /*
@@ -3809,7 +3872,7 @@ int select_nohz_load_balancer(int stop_tick)
 	int cpu = smp_processor_id();
 
 	if (stop_tick) {
-		cpu_set(cpu, nohz.cpu_mask);
+		cpumask_set_cpu(cpu, nohz.cpu_mask);
 		cpu_rq(cpu)->in_nohz_recently = 1;
 
 		/*
@@ -3823,7 +3886,7 @@ int select_nohz_load_balancer(int stop_tick)
 		}
 
 		/* time for ilb owner also to sleep */
-		if (cpus_weight(nohz.cpu_mask) == num_online_cpus()) {
+		if (cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
 			if (atomic_read(&nohz.load_balancer) == cpu)
 				atomic_set(&nohz.load_balancer, -1);
 			return 0;
@@ -3836,10 +3899,10 @@ int select_nohz_load_balancer(int stop_tick)
 		} else if (atomic_read(&nohz.load_balancer) == cpu)
 			return 1;
 	} else {
-		if (!cpu_isset(cpu, nohz.cpu_mask))
+		if (!cpumask_test_cpu(cpu, nohz.cpu_mask))
 			return 0;
 
-		cpu_clear(cpu, nohz.cpu_mask);
+		cpumask_clear_cpu(cpu, nohz.cpu_mask);
 
 		if (atomic_read(&nohz.load_balancer) == cpu)
 			if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
@@ -3867,7 +3930,11 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle)
 	unsigned long next_balance = jiffies + 60*HZ;
 	int update_next_balance = 0;
 	int need_serialize;
-	cpumask_t tmp;
+	cpumask_var_t tmp;
+
+	/* Fails alloc?  Rebalancing probably not a priority right now. */
+	if (!alloc_cpumask_var(&tmp, GFP_ATOMIC))
+		return;
 
 	for_each_domain(cpu, sd) {
 		if (!(sd->flags & SD_LOAD_BALANCE))
@@ -3892,7 +3959,7 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle)
 		}
 
 		if (time_after_eq(jiffies, sd->last_balance + interval)) {
-			if (load_balance(cpu, rq, sd, idle, &balance, &tmp)) {
+			if (load_balance(cpu, rq, sd, idle, &balance, tmp)) {
 				/*
 				 * We've pulled tasks over so either we're no
 				 * longer idle, or one of our SMT siblings is
@@ -3926,6 +3993,8 @@ out:
 	 */
 	if (likely(update_next_balance))
 		rq->next_balance = next_balance;
+
+	free_cpumask_var(tmp);
 }
 
 /*
@@ -3950,12 +4019,13 @@ static void run_rebalance_domains(struct softirq_action *h)
 	 */
 	if (this_rq->idle_at_tick &&
 	    atomic_read(&nohz.load_balancer) == this_cpu) {
-		cpumask_t cpus = nohz.cpu_mask;
 		struct rq *rq;
 		int balance_cpu;
 
-		cpu_clear(this_cpu, cpus);
-		for_each_cpu_mask_nr(balance_cpu, cpus) {
+		for_each_cpu(balance_cpu, nohz.cpu_mask) {
+			if (balance_cpu == this_cpu)
+				continue;
+
 			/*
 			 * If this cpu gets work to do, stop the load balancing
 			 * work being done for other cpus. Next load
@@ -3993,7 +4063,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu)
 		rq->in_nohz_recently = 0;
 
 		if (atomic_read(&nohz.load_balancer) == cpu) {
-			cpu_clear(cpu, nohz.cpu_mask);
+			cpumask_clear_cpu(cpu, nohz.cpu_mask);
 			atomic_set(&nohz.load_balancer, -1);
 		}
 
@@ -4006,7 +4076,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu)
 			 * TBD: Traverse the sched domains and nominate
 			 * the nearest cpu in the nohz.cpu_mask.
 			 */
-			int ilb = first_cpu(nohz.cpu_mask);
+			int ilb = cpumask_first(nohz.cpu_mask);
 
 			if (ilb < nr_cpu_ids)
 				resched_cpu(ilb);
@@ -4018,7 +4088,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu)
 	 * cpus with ticks stopped, is it time for that to stop?
 	 */
 	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) == cpu &&
-	    cpus_weight(nohz.cpu_mask) == num_online_cpus()) {
+	    cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
 		resched_cpu(cpu);
 		return;
 	}
@@ -4028,7 +4098,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu)
 	 * someone else, then no need raise the SCHED_SOFTIRQ
 	 */
 	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) != cpu &&
-	    cpu_isset(cpu, nohz.cpu_mask))
+	    cpumask_test_cpu(cpu, nohz.cpu_mask))
 		return;
 #endif
 	if (time_after_eq(jiffies, rq->next_balance))
@@ -5401,10 +5471,9 @@ out_unlock:
 	return retval;
 }
 
-long sched_setaffinity(pid_t pid, const cpumask_t *in_mask)
+long sched_setaffinity(pid_t pid, const struct cpumask *in_mask)
 {
-	cpumask_t cpus_allowed;
-	cpumask_t new_mask = *in_mask;
+	cpumask_var_t cpus_allowed, new_mask;
 	struct task_struct *p;
 	int retval;
 
@@ -5426,6 +5495,14 @@ long sched_setaffinity(pid_t pid, const cpumask_t *in_mask)
 	get_task_struct(p);
 	read_unlock(&tasklist_lock);
 
+	if (!alloc_cpumask_var(&cpus_allowed, GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto out_put_task;
+	}
+	if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto out_free_cpus_allowed;
+	}
 	retval = -EPERM;
 	if (!check_same_owner(p) && !capable(CAP_SYS_NICE))
 		goto out_unlock;
@@ -5434,37 +5511,41 @@ long sched_setaffinity(pid_t pid, const cpumask_t *in_mask)
 	if (retval)
 		goto out_unlock;
 
-	cpuset_cpus_allowed(p, &cpus_allowed);
-	cpus_and(new_mask, new_mask, cpus_allowed);
+	cpuset_cpus_allowed(p, cpus_allowed);
+	cpumask_and(new_mask, in_mask, cpus_allowed);
  again:
-	retval = set_cpus_allowed_ptr(p, &new_mask);
+	retval = set_cpus_allowed_ptr(p, new_mask);
 
 	if (!retval) {
-		cpuset_cpus_allowed(p, &cpus_allowed);
-		if (!cpus_subset(new_mask, cpus_allowed)) {
+		cpuset_cpus_allowed(p, cpus_allowed);
+		if (!cpumask_subset(new_mask, cpus_allowed)) {
 			/*
 			 * We must have raced with a concurrent cpuset
 			 * update. Just reset the cpus_allowed to the
 			 * cpuset's cpus_allowed
 			 */
-			new_mask = cpus_allowed;
+			cpumask_copy(new_mask, cpus_allowed);
 			goto again;
 		}
 	}
 out_unlock:
+	free_cpumask_var(new_mask);
+out_free_cpus_allowed:
+	free_cpumask_var(cpus_allowed);
+out_put_task:
 	put_task_struct(p);
 	put_online_cpus();
 	return retval;
 }
 
 static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len,
-			     cpumask_t *new_mask)
+			     struct cpumask *new_mask)
 {
-	if (len < sizeof(cpumask_t)) {
-		memset(new_mask, 0, sizeof(cpumask_t));
-	} else if (len > sizeof(cpumask_t)) {
-		len = sizeof(cpumask_t);
-	}
+	if (len < cpumask_size())
+		cpumask_clear(new_mask);
+	else if (len > cpumask_size())
+		len = cpumask_size();
+
 	return copy_from_user(new_mask, user_mask_ptr, len) ? -EFAULT : 0;
 }
 
@@ -5477,17 +5558,20 @@ static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len,
 asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len,
 				      unsigned long __user *user_mask_ptr)
 {
-	cpumask_t new_mask;
+	cpumask_var_t new_mask;
 	int retval;
 
-	retval = get_user_cpu_mask(user_mask_ptr, len, &new_mask);
-	if (retval)
-		return retval;
+	if (!alloc_cpumask_var(&new_mask, GFP_KERNEL))
+		return -ENOMEM;
 
-	return sched_setaffinity(pid, &new_mask);
+	retval = get_user_cpu_mask(user_mask_ptr, len, new_mask);
+	if (retval == 0)
+		retval = sched_setaffinity(pid, new_mask);
+	free_cpumask_var(new_mask);
+	return retval;
 }
 
-long sched_getaffinity(pid_t pid, cpumask_t *mask)
+long sched_getaffinity(pid_t pid, struct cpumask *mask)
 {
 	struct task_struct *p;
 	int retval;
@@ -5504,7 +5588,7 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask)
 	if (retval)
 		goto out_unlock;
 
-	cpus_and(*mask, p->cpus_allowed, cpu_online_map);
+	cpumask_and(mask, &p->cpus_allowed, cpu_online_mask);
 
 out_unlock:
 	read_unlock(&tasklist_lock);
@@ -5523,19 +5607,24 @@ asmlinkage long sys_sched_getaffinity(pid_t pid, unsigned int len,
 				      unsigned long __user *user_mask_ptr)
 {
 	int ret;
-	cpumask_t mask;
+	cpumask_var_t mask;
 
-	if (len < sizeof(cpumask_t))
+	if (len < cpumask_size())
 		return -EINVAL;
 
-	ret = sched_getaffinity(pid, &mask);
-	if (ret < 0)
-		return ret;
+	if (!alloc_cpumask_var(&mask, GFP_KERNEL))
+		return -ENOMEM;
 
-	if (copy_to_user(user_mask_ptr, &mask, sizeof(cpumask_t)))
-		return -EFAULT;
+	ret = sched_getaffinity(pid, mask);
+	if (ret == 0) {
+		if (copy_to_user(user_mask_ptr, mask, cpumask_size()))
+			ret = -EFAULT;
+		else
+			ret = cpumask_size();
+	}
+	free_cpumask_var(mask);
 
-	return sizeof(cpumask_t);
+	return ret;
 }
 
 /**
@@ -5877,7 +5966,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
 	idle->se.exec_start = sched_clock();
 
 	idle->prio = idle->normal_prio = MAX_PRIO;
-	idle->cpus_allowed = cpumask_of_cpu(cpu);
+	cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu));
 	__set_task_cpu(idle, cpu);
 
 	rq->curr = rq->idle = idle;
@@ -5904,9 +5993,9 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
  * indicates which cpus entered this state. This is used
  * in the rcu update to wait only for active cpus. For system
  * which do not switch off the HZ timer nohz_cpu_mask should
- * always be CPU_MASK_NONE.
+ * always be CPU_BITS_NONE.
  */
-cpumask_t nohz_cpu_mask = CPU_MASK_NONE;
+cpumask_var_t nohz_cpu_mask;
 
 /*
  * Increase the granularity value when there are more CPUs,
@@ -5961,7 +6050,7 @@ static inline void sched_init_granularity(void)
  * task must not exit() & deallocate itself prematurely. The
  * call is not atomic; no spinlocks may be held.
  */
-int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
+int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask)
 {
 	struct migration_req req;
 	unsigned long flags;
@@ -5969,13 +6058,13 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
 	int ret = 0;
 
 	rq = task_rq_lock(p, &flags);
-	if (!cpus_intersects(*new_mask, cpu_online_map)) {
+	if (!cpumask_intersects(new_mask, cpu_online_mask)) {
 		ret = -EINVAL;
 		goto out;
 	}
 
 	if (unlikely((p->flags & PF_THREAD_BOUND) && p != current &&
-		     !cpus_equal(p->cpus_allowed, *new_mask))) {
+		     !cpumask_equal(&p->cpus_allowed, new_mask))) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -5983,15 +6072,15 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
 	if (p->sched_class->set_cpus_allowed)
 		p->sched_class->set_cpus_allowed(p, new_mask);
 	else {
-		p->cpus_allowed = *new_mask;
-		p->rt.nr_cpus_allowed = cpus_weight(*new_mask);
+		cpumask_copy(&p->cpus_allowed, new_mask);
+		p->rt.nr_cpus_allowed = cpumask_weight(new_mask);
 	}
 
 	/* Can the task run on the task's current CPU? If so, we're done */
-	if (cpu_isset(task_cpu(p), *new_mask))
+	if (cpumask_test_cpu(task_cpu(p), new_mask))
 		goto out;
 
-	if (migrate_task(p, any_online_cpu(*new_mask), &req)) {
+	if (migrate_task(p, cpumask_any_and(cpu_online_mask, new_mask), &req)) {
 		/* Need help from migration thread: drop lock and wait. */
 		task_rq_unlock(rq, &flags);
 		wake_up_process(rq->migration_thread);
@@ -6033,7 +6122,7 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
 	if (task_cpu(p) != src_cpu)
 		goto done;
 	/* Affinity changed (again). */
-	if (!cpu_isset(dest_cpu, p->cpus_allowed))
+	if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed))
 		goto fail;
 
 	on_rq = p->se.on_rq;
@@ -6130,50 +6219,43 @@ static int __migrate_task_irq(struct task_struct *p, int src_cpu, int dest_cpu)
  */
 static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p)
 {
-	unsigned long flags;
-	cpumask_t mask;
-	struct rq *rq;
 	int dest_cpu;
+	/* FIXME: Use cpumask_of_node here. */
+	cpumask_t _nodemask = node_to_cpumask(cpu_to_node(dead_cpu));
+	const struct cpumask *nodemask = &_nodemask;
+
+again:
+	/* Look for allowed, online CPU in same node. */
+	for_each_cpu_and(dest_cpu, nodemask, cpu_online_mask)
+		if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed))
+			goto move;
+
+	/* Any allowed, online CPU? */
+	dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_online_mask);
+	if (dest_cpu < nr_cpu_ids)
+		goto move;
+
+	/* No more Mr. Nice Guy. */
+	if (dest_cpu >= nr_cpu_ids) {
+		cpuset_cpus_allowed_locked(p, &p->cpus_allowed);
+		dest_cpu = cpumask_any_and(cpu_online_mask, &p->cpus_allowed);
 
-	do {
-		/* On same node? */
-		mask = node_to_cpumask(cpu_to_node(dead_cpu));
-		cpus_and(mask, mask, p->cpus_allowed);
-		dest_cpu = any_online_cpu(mask);
-
-		/* On any allowed CPU? */
-		if (dest_cpu >= nr_cpu_ids)
-			dest_cpu = any_online_cpu(p->cpus_allowed);
-
-		/* No more Mr. Nice Guy. */
-		if (dest_cpu >= nr_cpu_ids) {
-			cpumask_t cpus_allowed;
-
-			cpuset_cpus_allowed_locked(p, &cpus_allowed);
-			/*
-			 * Try to stay on the same cpuset, where the
-			 * current cpuset may be a subset of all cpus.
-			 * The cpuset_cpus_allowed_locked() variant of
-			 * cpuset_cpus_allowed() will not block. It must be
-			 * called within calls to cpuset_lock/cpuset_unlock.
-			 */
-			rq = task_rq_lock(p, &flags);
-			p->cpus_allowed = cpus_allowed;
-			dest_cpu = any_online_cpu(p->cpus_allowed);
-			task_rq_unlock(rq, &flags);
-
-			/*
-			 * Don't tell them about moving exiting tasks or
-			 * kernel threads (both mm NULL), since they never
-			 * leave kernel.
-			 */
-			if (p->mm && printk_ratelimit()) {
-				printk(KERN_INFO "process %d (%s) no "
-				       "longer affine to cpu%d\n",
-					task_pid_nr(p), p->comm, dead_cpu);
-			}
+		/*
+		 * Don't tell them about moving exiting tasks or
+		 * kernel threads (both mm NULL), since they never
+		 * leave kernel.
+		 */
+		if (p->mm && printk_ratelimit()) {
+			printk(KERN_INFO "process %d (%s) no "
+			       "longer affine to cpu%d\n",
+			       task_pid_nr(p), p->comm, dead_cpu);
 		}
-	} while (!__migrate_task_irq(p, dead_cpu, dest_cpu));
+	}
+
+move:
+	/* It can have affinity changed while we were choosing. */
+	if (unlikely(!__migrate_task_irq(p, dead_cpu, dest_cpu)))
+		goto again;
 }
 
 /*
@@ -6185,7 +6267,7 @@ static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p)
  */
 static void migrate_nr_uninterruptible(struct rq *rq_src)
 {
-	struct rq *rq_dest = cpu_rq(any_online_cpu(*CPU_MASK_ALL_PTR));
+	struct rq *rq_dest = cpu_rq(cpumask_any(cpu_online_mask));
 	unsigned long flags;
 
 	local_irq_save(flags);
@@ -6475,7 +6557,7 @@ static void set_rq_online(struct rq *rq)
 	if (!rq->online) {
 		const struct sched_class *class;
 
-		cpu_set(rq->cpu, rq->rd->online);
+		cpumask_set_cpu(rq->cpu, rq->rd->online);
 		rq->online = 1;
 
 		for_each_class(class) {
@@ -6495,7 +6577,7 @@ static void set_rq_offline(struct rq *rq)
 				class->rq_offline(rq);
 		}
 
-		cpu_clear(rq->cpu, rq->rd->online);
+		cpumask_clear_cpu(rq->cpu, rq->rd->online);
 		rq->online = 0;
 	}
 }
@@ -6536,7 +6618,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		rq = cpu_rq(cpu);
 		spin_lock_irqsave(&rq->lock, flags);
 		if (rq->rd) {
-			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
 
 			set_rq_online(rq);
 		}
@@ -6550,7 +6632,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 			break;
 		/* Unbind it from offline cpu so it can run. Fall thru. */
 		kthread_bind(cpu_rq(cpu)->migration_thread,
-			     any_online_cpu(cpu_online_map));
+			     cpumask_any(cpu_online_mask));
 		kthread_stop(cpu_rq(cpu)->migration_thread);
 		cpu_rq(cpu)->migration_thread = NULL;
 		break;
@@ -6600,7 +6682,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		rq = cpu_rq(cpu);
 		spin_lock_irqsave(&rq->lock, flags);
 		if (rq->rd) {
-			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
 			set_rq_offline(rq);
 		}
 		spin_unlock_irqrestore(&rq->lock, flags);
@@ -6639,13 +6721,13 @@ early_initcall(migration_init);
 #ifdef CONFIG_SCHED_DEBUG
 
 static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
-				  cpumask_t *groupmask)
+				  struct cpumask *groupmask)
 {
 	struct sched_group *group = sd->groups;
 	char str[256];
 
-	cpulist_scnprintf(str, sizeof(str), sd->span);
-	cpus_clear(*groupmask);
+	cpulist_scnprintf(str, sizeof(str), sched_domain_span(sd));
+	cpumask_clear(groupmask);
 
 	printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
 
@@ -6659,11 +6741,11 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 
 	printk(KERN_CONT "span %s level %s\n", str, sd->name);
 
-	if (!cpu_isset(cpu, sd->span)) {
+	if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 		printk(KERN_ERR "ERROR: domain->span does not contain "
 				"CPU%d\n", cpu);
 	}
-	if (!cpu_isset(cpu, group->cpumask)) {
+	if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) {
 		printk(KERN_ERR "ERROR: domain->groups does not contain"
 				" CPU%d\n", cpu);
 	}
@@ -6683,31 +6765,32 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 			break;
 		}
 
-		if (!cpus_weight(group->cpumask)) {
+		if (!cpumask_weight(sched_group_cpus(group))) {
 			printk(KERN_CONT "\n");
 			printk(KERN_ERR "ERROR: empty group\n");
 			break;
 		}
 
-		if (cpus_intersects(*groupmask, group->cpumask)) {
+		if (cpumask_intersects(groupmask, sched_group_cpus(group))) {
 			printk(KERN_CONT "\n");
 			printk(KERN_ERR "ERROR: repeated CPUs\n");
 			break;
 		}
 
-		cpus_or(*groupmask, *groupmask, group->cpumask);
+		cpumask_or(groupmask, groupmask, sched_group_cpus(group));
 
-		cpulist_scnprintf(str, sizeof(str), group->cpumask);
+		cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group));
 		printk(KERN_CONT " %s", str);
 
 		group = group->next;
 	} while (group != sd->groups);
 	printk(KERN_CONT "\n");
 
-	if (!cpus_equal(sd->span, *groupmask))
+	if (!cpumask_equal(sched_domain_span(sd), groupmask))
 		printk(KERN_ERR "ERROR: groups don't span domain->span\n");
 
-	if (sd->parent && !cpus_subset(*groupmask, sd->parent->span))
+	if (sd->parent &&
+	    !cpumask_subset(groupmask, sched_domain_span(sd->parent)))
 		printk(KERN_ERR "ERROR: parent span is not a superset "
 			"of domain->span\n");
 	return 0;
@@ -6715,7 +6798,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 
 static void sched_domain_debug(struct sched_domain *sd, int cpu)
 {
-	cpumask_t *groupmask;
+	cpumask_var_t groupmask;
 	int level = 0;
 
 	if (!sd) {
@@ -6725,8 +6808,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu)
 
 	printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu);
 
-	groupmask = kmalloc(sizeof(cpumask_t), GFP_KERNEL);
-	if (!groupmask) {
+	if (!alloc_cpumask_var(&groupmask, GFP_KERNEL)) {
 		printk(KERN_DEBUG "Cannot load-balance (out of memory)\n");
 		return;
 	}
@@ -6739,7 +6821,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu)
 		if (!sd)
 			break;
 	}
-	kfree(groupmask);
+	free_cpumask_var(groupmask);
 }
 #else /* !CONFIG_SCHED_DEBUG */
 # define sched_domain_debug(sd, cpu) do { } while (0)
@@ -6747,7 +6829,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu)
 
 static int sd_degenerate(struct sched_domain *sd)
 {
-	if (cpus_weight(sd->span) == 1)
+	if (cpumask_weight(sched_domain_span(sd)) == 1)
 		return 1;
 
 	/* Following flags need at least 2 groups */
@@ -6778,7 +6860,7 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 	if (sd_degenerate(parent))
 		return 1;
 
-	if (!cpus_equal(sd->span, parent->span))
+	if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent)))
 		return 0;
 
 	/* Does parent contain flags not in child? */
@@ -6802,6 +6884,16 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 	return 1;