sched: revert load_balance_monitor() changes

The following commits cause a number of regressions:

  commit 58e2d4ca581167c2a079f4ee02be2f0bc52e8729
  Author: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
  Date:   Fri Jan 25 21:08:00 2008 +0100
  sched: group scheduling, change how cpu load is calculated

  commit 6b2d7700266b9402e12824e11e0099ae6a4a6a79
  Author: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
  Date:   Fri Jan 25 21:08:00 2008 +0100
  sched: group scheduler, fix fairness of cpu bandwidth allocation for task groups

Namely:
 - very frequent wakeups on SMP, reported by PowerTop users.
 - cacheline trashing on (large) SMP
 - some latencies larger than 500ms

While there is a mergeable patch to fix the latter, the former issues
are not fixable in a manner suitable for .25 (we're at -rc3 now).

Hence we revert them and try again in v2.6.26.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Tested-by: Alexey Zaytsev <alexey.zaytsev@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

1 files changed, 35 insertions, 80 deletions

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index c8e6492c592..3df4d46994c 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -727,8 +727,6 @@ static inline struct sched_entity *parent_entity(struct sched_entity *se)
 	return se->parent;
 }
 
-#define GROUP_IMBALANCE_PCT	20
-
 #else	/* CONFIG_FAIR_GROUP_SCHED */
 
 #define for_each_sched_entity(se) \
@@ -819,26 +817,15 @@ hrtick_start_fair(struct rq *rq, struct task_struct *p)
 static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup)
 {
 	struct cfs_rq *cfs_rq;
-	struct sched_entity *se = &p->se,
-			    *topse = NULL;	/* Highest schedulable entity */
-	int incload = 1;
+	struct sched_entity *se = &p->se;
 
 	for_each_sched_entity(se) {
-		topse = se;
-		if (se->on_rq) {
-			incload = 0;
+		if (se->on_rq)
 			break;
-		}
 		cfs_rq = cfs_rq_of(se);
 		enqueue_entity(cfs_rq, se, wakeup);
 		wakeup = 1;
 	}
-	/* Increment cpu load if we just enqueued the first task of a group on
-	 * 'rq->cpu'. 'topse' represents the group to which task 'p' belongs
-	 * at the highest grouping level.
-	 */
-	if (incload)
-		inc_cpu_load(rq, topse->load.weight);
 
 	hrtick_start_fair(rq, rq->curr);
 }
@@ -851,28 +838,16 @@ static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup)
 static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int sleep)
 {
 	struct cfs_rq *cfs_rq;
-	struct sched_entity *se = &p->se,
-			    *topse = NULL; 	/* Highest schedulable entity */
-	int decload = 1;
+	struct sched_entity *se = &p->se;
 
 	for_each_sched_entity(se) {
-		topse = se;
 		cfs_rq = cfs_rq_of(se);
 		dequeue_entity(cfs_rq, se, sleep);
 		/* Don't dequeue parent if it has other entities besides us */
-		if (cfs_rq->load.weight) {
-			if (parent_entity(se))
-				decload = 0;
+		if (cfs_rq->load.weight)
 			break;
-		}
 		sleep = 1;
 	}
-	/* Decrement cpu load if we just dequeued the last task of a group on
-	 * 'rq->cpu'. 'topse' represents the group to which task 'p' belongs
-	 * at the highest grouping level.
-	 */
-	if (decload)
-		dec_cpu_load(rq, topse->load.weight);
 
 	hrtick_start_fair(rq, rq->curr);
 }
@@ -1186,6 +1161,25 @@ static struct task_struct *load_balance_next_fair(void *arg)
 	return __load_balance_iterator(cfs_rq, cfs_rq->rb_load_balance_curr);
 }
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
+static int cfs_rq_best_prio(struct cfs_rq *cfs_rq)
+{
+	struct sched_entity *curr;
+	struct task_struct *p;
+
+	if (!cfs_rq->nr_running || !first_fair(cfs_rq))
+		return MAX_PRIO;
+
+	curr = cfs_rq->curr;
+	if (!curr)
+		curr = __pick_next_entity(cfs_rq);
+
+	p = task_of(curr);
+
+	return p->prio;
+}
+#endif
+
 static unsigned long
 load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		  unsigned long max_load_move,
@@ -1195,45 +1189,28 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 	struct cfs_rq *busy_cfs_rq;
 	long rem_load_move = max_load_move;
 	struct rq_iterator cfs_rq_iterator;
-	unsigned long load_moved;
 
 	cfs_rq_iterator.start = load_balance_start_fair;
 	cfs_rq_iterator.next = load_balance_next_fair;
 
 	for_each_leaf_cfs_rq(busiest, busy_cfs_rq) {
 #ifdef CONFIG_FAIR_GROUP_SCHED
-		struct cfs_rq *this_cfs_rq = busy_cfs_rq->tg->cfs_rq[this_cpu];
-		unsigned long maxload, task_load, group_weight;
-		unsigned long thisload, per_task_load;
-		struct sched_entity *se = busy_cfs_rq->tg->se[busiest->cpu];
-
-		task_load = busy_cfs_rq->load.weight;
-		group_weight = se->load.weight;
+		struct cfs_rq *this_cfs_rq;
+		long imbalance;
+		unsigned long maxload;
 
-		/*
-		 * 'group_weight' is contributed by tasks of total weight
-		 * 'task_load'. To move 'rem_load_move' worth of weight only,
-		 * we need to move a maximum task load of:
-		 *
-		 * 	maxload = (remload / group_weight) * task_load;
-		 */
-		maxload = (rem_load_move * task_load) / group_weight;
+		this_cfs_rq = cpu_cfs_rq(busy_cfs_rq, this_cpu);
 
-		if (!maxload || !task_load)
+		imbalance = busy_cfs_rq->load.weight - this_cfs_rq->load.weight;
+		/* Don't pull if this_cfs_rq has more load than busy_cfs_rq */
+		if (imbalance <= 0)
 			continue;
 
-		per_task_load = task_load / busy_cfs_rq->nr_running;
-		/*
-		 * balance_tasks will try to forcibly move atleast one task if
-		 * possible (because of SCHED_LOAD_SCALE_FUZZ). Avoid that if
-		 * maxload is less than GROUP_IMBALANCE_FUZZ% the per_task_load.
-		 */
-		 if (100 * maxload < GROUP_IMBALANCE_PCT * per_task_load)
-			continue;
+		/* Don't pull more than imbalance/2 */
+		imbalance /= 2;
+		maxload = min(rem_load_move, imbalance);
 
-		/* Disable priority-based load balance */
-		*this_best_prio = 0;
-		thisload = this_cfs_rq->load.weight;
+		*this_best_prio = cfs_rq_best_prio(this_cfs_rq);
 #else
 # define maxload rem_load_move
 #endif
@@ -1242,33 +1219,11 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		 * load_balance_[start|next]_fair iterators
 		 */
 		cfs_rq_iterator.arg = busy_cfs_rq;
-		load_moved = balance_tasks(this_rq, this_cpu, busiest,
+		rem_load_move -= balance_tasks(this_rq, this_cpu, busiest,
 					       maxload, sd, idle, all_pinned,
 					       this_best_prio,
 					       &cfs_rq_iterator);
 
-#ifdef CONFIG_FAIR_GROUP_SCHED
-		/*
-		 * load_moved holds the task load that was moved. The
-		 * effective (group) weight moved would be:
-		 * 	load_moved_eff = load_moved/task_load * group_weight;
-		 */
-		load_moved = (group_weight * load_moved) / task_load;
-
-		/* Adjust shares on both cpus to reflect load_moved */
-		group_weight -= load_moved;
-		set_se_shares(se, group_weight);
-
-		se = busy_cfs_rq->tg->se[this_cpu];
-		if (!thisload)
-			group_weight = load_moved;
-		else
-			group_weight = se->load.weight + load_moved;
-		set_se_shares(se, group_weight);
-#endif
-
-		rem_load_move -= load_moved;
-
 		if (rem_load_move <= 0)
 			break;
 	}