9 files changed, 159 insertions, 100 deletions

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index c24875bd9c5b..d614d2472798 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -359,13 +359,14 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
 
 /*
  * The following particular system cpumasks and operations manage
- * possible, present and online cpus.  Each of them is a fixed size
+ * possible, present, active and online cpus.  Each of them is a fixed size
  * bitmap of size NR_CPUS.
  *
  *  #ifdef CONFIG_HOTPLUG_CPU
  *     cpu_possible_map - has bit 'cpu' set iff cpu is populatable
  *     cpu_present_map  - has bit 'cpu' set iff cpu is populated
  *     cpu_online_map   - has bit 'cpu' set iff cpu available to scheduler
+ *     cpu_active_map   - has bit 'cpu' set iff cpu available to migration
  *  #else
  *     cpu_possible_map - has bit 'cpu' set iff cpu is populated
  *     cpu_present_map  - copy of cpu_possible_map
@@ -416,6 +417,7 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
 extern cpumask_t cpu_possible_map;
 extern cpumask_t cpu_online_map;
 extern cpumask_t cpu_present_map;
+extern cpumask_t cpu_active_map;
 
 #if NR_CPUS > 1
 #define num_online_cpus()       cpus_weight(cpu_online_map)
@@ -424,6 +426,7 @@ extern cpumask_t 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)
+#define cpu_active(cpu)         cpu_isset((cpu), cpu_active_map)
 #else
 #define num_online_cpus()       1
 #define num_possible_cpus()     1
@@ -431,6 +434,7 @@ extern cpumask_t cpu_present_map;
 #define cpu_online(cpu)         ((cpu) == 0)
 #define cpu_possible(cpu)       ((cpu) == 0)
 #define cpu_present(cpu)        ((cpu) == 0)
+#define cpu_active(cpu)         ((cpu) == 0)
 #endif
 
 #define cpu_is_offline(cpu)     unlikely(!cpu_online(cpu))
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 038578362b47..e8f450c499b0 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -78,6 +78,8 @@ extern void cpuset_track_online_nodes(void);
 
 extern int current_cpuset_is_being_rebound(void);
 
+extern void rebuild_sched_domains(void);
+
 #else /* !CONFIG_CPUSETS */
 
 static inline int cpuset_init_early(void) { return 0; }
@@ -156,6 +158,11 @@ static inline int current_cpuset_is_being_rebound(void)
         return 0;
 }
 
+static inline void rebuild_sched_domains(void)
+{
+        partition_sched_domains(0, NULL, NULL);
+}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1941d8b5cf11..26da921530fe 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -824,7 +824,16 @@ extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
                                     struct sched_domain_attr *dattr_new);
 extern int arch_reinit_sched_domains(void);
 
-#endif  /* CONFIG_SMP */
+#else /* CONFIG_SMP */
+
+struct sched_domain_attr;
+
+static inline void
+partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+                        struct sched_domain_attr *dattr_new)
+{
+}
+#endif  /* !CONFIG_SMP */
 
 struct io_context;                      /* See blkdev.h */
 #define NGROUPS_SMALL           32
diff --git a/init/main.c b/init/main.c
index edeace036fd9..dd25259530ea 100644
--- a/init/main.c
+++ b/init/main.c
@@ -415,6 +415,13 @@ static void __init smp_init(void)
 {
         unsigned int cpu;
 
+        /*
+         * Set up the current CPU as possible to migrate to.
+         * The other ones will be done by cpu_up/cpu_down()
+         */
+        cpu = smp_processor_id();
+        cpu_set(cpu, cpu_active_map);
+
         /* FIXME: This should be done in userspace --RR */
         for_each_present_cpu(cpu) {
                 if (num_online_cpus() >= setup_max_cpus)
diff --git a/kernel/cpu.c b/kernel/cpu.c
index cfb1d43ab801..033603c1d7c3 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -64,6 +64,8 @@ void __init cpu_hotplug_init(void)
         cpu_hotplug.refcount = 0;
 }
 
+cpumask_t cpu_active_map;
+
 #ifdef CONFIG_HOTPLUG_CPU
 
 void get_online_cpus(void)
@@ -291,11 +293,30 @@ int __ref cpu_down(unsigned int cpu)
         int err = 0;
 
         cpu_maps_update_begin();
-        if (cpu_hotplug_disabled)
+
+        if (cpu_hotplug_disabled) {
                 err = -EBUSY;
-        else
-                err = _cpu_down(cpu, 0);
+                goto out;
+        }
+
+        cpu_clear(cpu, cpu_active_map);
+
+        /*
+         * Make sure the all cpus did the reschedule and are not
+         * using stale version of the cpu_active_map.
+         * This is not strictly necessary becuase stop_machine()
+         * that we run down the line already provides the required
+         * synchronization. But it's really a side effect and we do not
+         * want to depend on the innards of the stop_machine here.
+         */
+        synchronize_sched();
+
+        err = _cpu_down(cpu, 0);
 
+        if (cpu_online(cpu))
+                cpu_set(cpu, cpu_active_map);
+
+out:
         cpu_maps_update_done();
         return err;
 }
@@ -355,11 +376,18 @@ int __cpuinit cpu_up(unsigned int cpu)
         }
 
         cpu_maps_update_begin();
-        if (cpu_hotplug_disabled)
+
+        if (cpu_hotplug_disabled) {
                 err = -EBUSY;
-        else
-                err = _cpu_up(cpu, 0);
+                goto out;
+        }
+
+        err = _cpu_up(cpu, 0);
 
+        if (cpu_online(cpu))
+                cpu_set(cpu, cpu_active_map);
+
+out:
         cpu_maps_update_done();
         return err;
 }
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 459d601947a8..3c3ef02f65f1 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -564,7 +564,7 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c)
  *      partition_sched_domains().
  */
 
-static void rebuild_sched_domains(void)
+void rebuild_sched_domains(void)
 {
         struct kfifo *q;        /* queue of cpusets to be scanned */
         struct cpuset *cp;      /* scans q */
diff --git a/kernel/sched.c b/kernel/sched.c
index c13c75e9f9f7..85cf246cfdf5 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2802,7 +2802,7 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu)
 
         rq = task_rq_lock(p, &flags);
         if (!cpu_isset(dest_cpu, p->cpus_allowed)
-            || unlikely(cpu_is_offline(dest_cpu)))
+            || unlikely(!cpu_active(dest_cpu)))
                 goto out;
 
         /* force the process onto the specified CPU */
@@ -3770,7 +3770,7 @@ int select_nohz_load_balancer(int stop_tick)
                 /*
                  * If we are going offline and still the leader, give up!
                  */
-                if (cpu_is_offline(cpu) &&
+                if (!cpu_active(cpu) &&
                     atomic_read(&nohz.load_balancer) == cpu) {
                         if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
                                 BUG();
@@ -5794,7 +5794,7 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
         struct rq *rq_dest, *rq_src;
         int ret = 0, on_rq;
 
-        if (unlikely(cpu_is_offline(dest_cpu)))
+        if (unlikely(!cpu_active(dest_cpu)))
                 return ret;
 
         rq_src = cpu_rq(src_cpu);
@@ -7472,18 +7472,6 @@ void __attribute__((weak)) arch_update_cpu_topology(void)
 }
 
 /*
- * Free current domain masks.
- * Called after all cpus are attached to NULL domain.
- */
-static void free_sched_domains(void)
-{
-        ndoms_cur = 0;
-        if (doms_cur != &fallback_doms)
-                kfree(doms_cur);
-        doms_cur = &fallback_doms;
-}
-
-/*
  * Set up scheduler domains and groups. Callers must hold the hotplug lock.
  * For now this just excludes isolated cpus, but could be used to
  * exclude other special cases in the future.
@@ -7561,7 +7549,7 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur,
  * ownership of it and will kfree it when done with it. If the caller
  * failed the kmalloc call, then it can pass in doms_new == NULL,
  * and partition_sched_domains() will fallback to the single partition
- * 'fallback_doms'.
+ * 'fallback_doms', it also forces the domains to be rebuilt.
  *
  * Call with hotplug lock held
  */
@@ -7575,12 +7563,8 @@ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
         /* always unregister in case we don't destroy any domains */
         unregister_sched_domain_sysctl();
 
-        if (doms_new == NULL) {
-                ndoms_new = 1;
-                doms_new = &fallback_doms;
-                cpus_andnot(doms_new[0], cpu_online_map, cpu_isolated_map);
-                dattr_new = NULL;
-        }
+        if (doms_new == NULL)
+                ndoms_new = 0;
 
         /* Destroy deleted domains */
         for (i = 0; i < ndoms_cur; i++) {
@@ -7595,6 +7579,14 @@ match1:
                 ;
         }
 
+        if (doms_new == NULL) {
+                ndoms_cur = 0;
+                ndoms_new = 1;
+                doms_new = &fallback_doms;
+                cpus_andnot(doms_new[0], cpu_online_map, cpu_isolated_map);
+                dattr_new = NULL;
+        }
+
         /* Build new domains */
         for (i = 0; i < ndoms_new; i++) {
                 for (j = 0; j < ndoms_cur; j++) {
@@ -7625,17 +7617,10 @@ match2:
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
 int arch_reinit_sched_domains(void)
 {
-        int err;
-
         get_online_cpus();
-        mutex_lock(&sched_domains_mutex);
-        detach_destroy_domains(&cpu_online_map);
-        free_sched_domains();
-        err = arch_init_sched_domains(&cpu_online_map);
-        mutex_unlock(&sched_domains_mutex);
+        rebuild_sched_domains();
         put_online_cpus();
-
-        return err;
+        return 0;
 }
 
 static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
@@ -7701,59 +7686,49 @@ int sched_create_sysfs_power_savings_entries(struct sysdev_class *cls)
 }
 #endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */
 
+#ifndef CONFIG_CPUSETS
 /*
- * Force a reinitialization of the sched domains hierarchy. The domains
- * and groups cannot be updated in place without racing with the balancing
- * code, so we temporarily attach all running cpus to the NULL domain
- * which will prevent rebalancing while the sched domains are recalculated.
+ * Add online and remove offline CPUs from the scheduler domains.
+ * When cpusets are enabled they take over this function.
  */
 static int update_sched_domains(struct notifier_block *nfb,
                                 unsigned long action, void *hcpu)
 {
+        switch (action) {
+        case CPU_ONLINE:
+        case CPU_ONLINE_FROZEN:
+        case CPU_DEAD:
+        case CPU_DEAD_FROZEN:
+                partition_sched_domains(0, NULL, NULL);
+                return NOTIFY_OK;
+
+        default:
+                return NOTIFY_DONE;
+        }
+}
+#endif
+
+static int update_runtime(struct notifier_block *nfb,
+                                unsigned long action, void *hcpu)
+{
         int cpu = (int)(long)hcpu;
 
         switch (action) {
         case CPU_DOWN_PREPARE:
         case CPU_DOWN_PREPARE_FROZEN:
                 disable_runtime(cpu_rq(cpu));
-                /* fall-through */
-        case CPU_UP_PREPARE:
-        case CPU_UP_PREPARE_FROZEN:
-                detach_destroy_domains(&cpu_online_map);
-                free_sched_domains();
                 return NOTIFY_OK;
 
-
         case CPU_DOWN_FAILED:
         case CPU_DOWN_FAILED_FROZEN:
         case CPU_ONLINE:
         case CPU_ONLINE_FROZEN:
                 enable_runtime(cpu_rq(cpu));
-                /* fall-through */
-        case CPU_UP_CANCELED:
-        case CPU_UP_CANCELED_FROZEN:
-        case CPU_DEAD:
-        case CPU_DEAD_FROZEN:
-                /*
-                 * Fall through and re-initialise the domains.
-                 */
-                break;
+                return NOTIFY_OK;
+
         default:
                 return NOTIFY_DONE;
         }
-
-#ifndef CONFIG_CPUSETS
-        /*
-         * Create default domain partitioning if cpusets are disabled.
-         * Otherwise we let cpusets rebuild the domains based on the
-         * current setup.
-         */
-
-        /* The hotplug lock is already held by cpu_up/cpu_down */
-        arch_init_sched_domains(&cpu_online_map);
-#endif
-
-        return NOTIFY_OK;
 }
 
 void __init sched_init_smp(void)
@@ -7773,8 +7748,15 @@ void __init sched_init_smp(void)
                 cpu_set(smp_processor_id(), non_isolated_cpus);
         mutex_unlock(&sched_domains_mutex);
         put_online_cpus();
+
+#ifndef CONFIG_CPUSETS
         /* XXX: Theoretical race here - CPU may be hotplugged now */
         hotcpu_notifier(update_sched_domains, 0);
+#endif
+
+        /* RT runtime code needs to handle some hotplug events */
+        hotcpu_notifier(update_runtime, 0);
+
         init_hrtick();
 
         /* Move init over to a non-isolated CPU */
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 6893b3ed65fe..7f700263f04c 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1003,6 +1003,8 @@ static void yield_task_fair(struct rq *rq)
  * not idle and an idle cpu is available.  The span of cpus to
  * search starts with cpus closest then further out as needed,
  * so we always favor a closer, idle cpu.
+ * Domains may include CPUs that are not usable for migration,
+ * hence we need to mask them out (cpu_active_map)
  *
  * Returns the CPU we should wake onto.
  */
@@ -1030,6 +1032,7 @@ 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);
+                        cpus_and(tmp, tmp, cpu_active_map);
                         for_each_cpu_mask(i, tmp) {
                                 if (idle_cpu(i)) {
                                         if (i != task_cpu(p)) {
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 147004c651c0..24621cea8bb0 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -601,11 +601,7 @@ static void __enqueue_rt_entity(struct sched_rt_entity *rt_se)
         if (group_rq && (rt_rq_throttled(group_rq) || !group_rq->rt_nr_running))
                 return;
 
-        if (rt_se->nr_cpus_allowed == 1)
-                list_add(&rt_se->run_list, queue);
-        else
-                list_add_tail(&rt_se->run_list, queue);
-
+        list_add_tail(&rt_se->run_list, queue);
         __set_bit(rt_se_prio(rt_se), array->bitmap);
 
         inc_rt_tasks(rt_se, rt_rq);
@@ -690,32 +686,34 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
  * Put task to the end of the run list without the overhead of dequeue
  * followed by enqueue.
  */
-static
-void requeue_rt_entity(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se)
+static void
+requeue_rt_entity(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se, int head)
 {
-        struct rt_prio_array *array = &rt_rq->active;
-
         if (on_rt_rq(rt_se)) {
-                list_del_init(&rt_se->run_list);
-                list_add_tail(&rt_se->run_list,
-                              array->queue + rt_se_prio(rt_se));
+                struct rt_prio_array *array = &rt_rq->active;
+                struct list_head *queue = array->queue + rt_se_prio(rt_se);
+
+                if (head)
+                        list_move(&rt_se->run_list, queue);
+                else
+                        list_move_tail(&rt_se->run_list, queue);
         }
 }
 
-static void requeue_task_rt(struct rq *rq, struct task_struct *p)
+static void requeue_task_rt(struct rq *rq, struct task_struct *p, int head)
 {
         struct sched_rt_entity *rt_se = &p->rt;
         struct rt_rq *rt_rq;
 
         for_each_sched_rt_entity(rt_se) {
                 rt_rq = rt_rq_of_se(rt_se);
-                requeue_rt_entity(rt_rq, rt_se);
+                requeue_rt_entity(rt_rq, rt_se, head);
         }
 }
 
 static void yield_task_rt(struct rq *rq)
 {
-        requeue_task_rt(rq, rq->curr);
+        requeue_task_rt(rq, rq->curr, 0);
 }
 
 #ifdef CONFIG_SMP
@@ -755,6 +753,30 @@ static int select_task_rq_rt(struct task_struct *p, int sync)
          */
         return task_cpu(p);
 }
+
+static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p)
+{
+        cpumask_t mask;
+
+        if (rq->curr->rt.nr_cpus_allowed == 1)
+                return;
+
+        if (p->rt.nr_cpus_allowed != 1
+            && cpupri_find(&rq->rd->cpupri, p, &mask))
+                return;
+
+        if (!cpupri_find(&rq->rd->cpupri, rq->curr, &mask))
+                return;
+
+        /*
+         * There appears to be other cpus that can accept
+         * current and none to run 'p', so lets reschedule
+         * to try and push current away:
+         */
+        requeue_task_rt(rq, p, 1);
+        resched_task(rq->curr);
+}
+
 #endif /* CONFIG_SMP */
 
 /*
@@ -780,18 +802,8 @@ static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p)
          * to move current somewhere else, making room for our non-migratable
          * task.
          */
-        if((p->prio == rq->curr->prio)
-           && p->rt.nr_cpus_allowed == 1
-           && rq->curr->rt.nr_cpus_allowed != 1) {
-                cpumask_t mask;
-
-                if (cpupri_find(&rq->rd->cpupri, rq->curr, &mask))
-                        /*
-                         * There appears to be other cpus that can accept
-                         * current, so lets reschedule to try and push it away
-                         */
-                        resched_task(rq->curr);
-        }
+        if (p->prio == rq->curr->prio && !need_resched())
+                check_preempt_equal_prio(rq, p);
 #endif
 }
 
@@ -924,6 +936,13 @@ static int find_lowest_rq(struct task_struct *task)
                 return -1; /* No targets found */
 
         /*
+         * Only consider CPUs that are usable for migration.
+         * I guess we might want to change cpupri_find() to ignore those
+         * in the first place.
+         */
+        cpus_and(*lowest_mask, *lowest_mask, cpu_active_map);
+
+        /*
          * At this point we have built a mask of cpus representing the
          * lowest priority tasks in the system.  Now we want to elect
          * the best one based on our affinity and topology.
@@ -1417,7 +1436,7 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued)
          * on the queue:
          */
         if (p->rt.run_list.prev != p->rt.run_list.next) {
-                requeue_task_rt(rq, p);
+                requeue_task_rt(rq, p, 0);
                 set_tsk_need_resched(p);
         }
 }