Merge git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched

* git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched:
  sched: fix new task startup crash
  sched: fix !SYSFS build breakage
  sched: fix improper load balance across sched domain
  sched: more robust sd-sysctl entry freeing

4 files changed, 49 insertions, 20 deletions

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 3de5aa210feb..c204ab0d4df1 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -535,10 +535,12 @@ struct user_struct {
 
 #ifdef CONFIG_FAIR_USER_SCHED
 	struct task_group *tg;
+#ifdef CONFIG_SYSFS
 	struct kset kset;
 	struct subsys_attribute user_attr;
 	struct work_struct work;
 #endif
+#endif
 };
 
 #ifdef CONFIG_FAIR_USER_SCHED
diff --git a/kernel/sched.c b/kernel/sched.c
index c4889abc00b6..92721d1534b8 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1712,7 +1712,7 @@ void fastcall wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 
 	p->prio = effective_prio(p);
 
-	if (!p->sched_class->task_new || !current->se.on_rq || !rq->cfs.curr) {
+	if (!p->sched_class->task_new || !current->se.on_rq) {
 		activate_task(rq, p, 0);
 	} else {
 		/*
@@ -2336,7 +2336,7 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 	unsigned long max_pull;
 	unsigned long busiest_load_per_task, busiest_nr_running;
 	unsigned long this_load_per_task, this_nr_running;
-	int load_idx;
+	int load_idx, group_imb = 0;
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
 	int power_savings_balance = 1;
 	unsigned long leader_nr_running = 0, min_load_per_task = 0;
@@ -2355,9 +2355,10 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		load_idx = sd->idle_idx;
 
 	do {
-		unsigned long load, group_capacity;
+		unsigned long load, group_capacity, max_cpu_load, min_cpu_load;
 		int local_group;
 		int i;
+		int __group_imb = 0;
 		unsigned int balance_cpu = -1, first_idle_cpu = 0;
 		unsigned long sum_nr_running, sum_weighted_load;
 
@@ -2368,6 +2369,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 
 		/* Tally up the load of all CPUs in the group */
 		sum_weighted_load = sum_nr_running = avg_load = 0;
+		max_cpu_load = 0;
+		min_cpu_load = ~0UL;
 
 		for_each_cpu_mask(i, group->cpumask) {
 			struct rq *rq;
@@ -2388,8 +2391,13 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 				}
 
 				load = target_load(i, load_idx);
-			} else
+			} else {
 				load = source_load(i, load_idx);
+				if (load > max_cpu_load)
+					max_cpu_load = load;
+				if (min_cpu_load > load)
+					min_cpu_load = load;
+			}
 
 			avg_load += load;
 			sum_nr_running += rq->nr_running;
@@ -2415,6 +2423,9 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		avg_load = sg_div_cpu_power(group,
 				avg_load * SCHED_LOAD_SCALE);
 
+		if ((max_cpu_load - min_cpu_load) > SCHED_LOAD_SCALE)
+			__group_imb = 1;
+
 		group_capacity = group->__cpu_power / SCHED_LOAD_SCALE;
 
 		if (local_group) {
@@ -2423,11 +2434,12 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 			this_nr_running = sum_nr_running;
 			this_load_per_task = sum_weighted_load;
 		} else if (avg_load > max_load &&
-			   sum_nr_running > group_capacity) {
+			   (sum_nr_running > group_capacity || __group_imb)) {
 			max_load = avg_load;
 			busiest = group;
 			busiest_nr_running = sum_nr_running;
 			busiest_load_per_task = sum_weighted_load;
+			group_imb = __group_imb;
 		}
 
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
@@ -2499,6 +2511,9 @@ group_next:
 		goto out_balanced;
 
 	busiest_load_per_task /= busiest_nr_running;
+	if (group_imb)
+		busiest_load_per_task = min(busiest_load_per_task, avg_load);
+
 	/*
 	 * We're trying to get all the cpus to the average_load, so we don't
 	 * want to push ourselves above the average load, nor do we wish to
@@ -5282,11 +5297,20 @@ static struct ctl_table *sd_alloc_ctl_entry(int n)
 
 static void sd_free_ctl_entry(struct ctl_table **tablep)
 {
-	struct ctl_table *entry = *tablep;
+	struct ctl_table *entry;
 
-	for (entry = *tablep; entry->procname; entry++)
+	/*
+	 * In the intermediate directories, both the child directory and
+	 * procname are dynamically allocated and could fail but the mode
+	 * will always be set.  In the lowest directory the names are
+	 * static strings and all have proc handlers.
+	 */
+	for (entry = *tablep; entry->mode; entry++) {
 		if (entry->child)
 			sd_free_ctl_entry(&entry->child);
+		if (entry->proc_handler == NULL)
+			kfree(entry->procname);
+	}
 
 	kfree(*tablep);
 	*tablep = NULL;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index a17b785d7000..166ed6db600b 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1031,12 +1031,8 @@ static void task_new_fair(struct rq *rq, struct task_struct *p)
 		swap(curr->vruntime, se->vruntime);
 	}
 
-	update_stats_enqueue(cfs_rq, se);
-	check_spread(cfs_rq, se);
-	check_spread(cfs_rq, curr);
-	__enqueue_entity(cfs_rq, se);
-	account_entity_enqueue(cfs_rq, se);
 	se->peer_preempt = 0;
+	enqueue_task_fair(rq, p, 0);
 	resched_task(rq->curr);
 }
 
diff --git a/kernel/user.c b/kernel/user.c
index 9cb6f6403561..e91331c457e2 100644
--- a/kernel/user.c
+++ b/kernel/user.c
@@ -84,9 +84,6 @@ static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)
 
 #ifdef CONFIG_FAIR_USER_SCHED
 
-static struct kobject uids_kobject; /* represents /sys/kernel/uids directory */
-static DEFINE_MUTEX(uids_mutex);
-
 static void sched_destroy_user(struct user_struct *up)
 {
 	sched_destroy_group(up->tg);
@@ -108,6 +105,19 @@ static void sched_switch_user(struct task_struct *p)
 	sched_move_task(p);
 }
 
+#else	/* CONFIG_FAIR_USER_SCHED */
+
+static void sched_destroy_user(struct user_struct *up) { }
+static int sched_create_user(struct user_struct *up) { return 0; }
+static void sched_switch_user(struct task_struct *p) { }
+
+#endif	/* CONFIG_FAIR_USER_SCHED */
+
+#if defined(CONFIG_FAIR_USER_SCHED) && defined(CONFIG_SYSFS)
+
+static struct kobject uids_kobject; /* represents /sys/kernel/uids directory */
+static DEFINE_MUTEX(uids_mutex);
+
 static inline void uids_mutex_lock(void)
 {
 	mutex_lock(&uids_mutex);
@@ -254,11 +264,8 @@ static inline void free_user(struct user_struct *up, unsigned long flags)
 	schedule_work(&up->work);
 }
 
-#else	/* CONFIG_FAIR_USER_SCHED */
+#else	/* CONFIG_FAIR_USER_SCHED && CONFIG_SYSFS */
 
-static void sched_destroy_user(struct user_struct *up) { }
-static int sched_create_user(struct user_struct *up) { return 0; }
-static void sched_switch_user(struct task_struct *p) { }
 static inline int user_kobject_create(struct user_struct *up) { return 0; }
 static inline void uids_mutex_lock(void) { }
 static inline void uids_mutex_unlock(void) { }
@@ -277,7 +284,7 @@ static inline void free_user(struct user_struct *up, unsigned long flags)
 	kmem_cache_free(uid_cachep, up);
 }
 
-#endif	/* CONFIG_FAIR_USER_SCHED */
+#endif
 
 /*
  * Locate the user_struct for the passed UID.  If found, take a ref on it.  The