[PATCH] NUMA scheduler fixup

From: Ingo Molnar <mingo@elte.hu>

This cleans up the NUMA scheduler to have separate triggers for idle
and busy rebalance ratio, preventing some complex interactions where
idle cpus can cause busy ones to busy rebalance more often. The core
functionality of this patch is rebalace_tick().

Tested in -mjb tree for over 1 month ... no problems at all found -
retested on top of 64-bk3 with new scheduler updates.

kernel/sched.c

@@ -33,6 +33,12 @@
 #include <linux/timer.h>
 #include <linux/rcupdate.h>
 
+#ifdef CONFIG_NUMA
+#define cpu_to_node_mask(cpu) node_to_cpumask(cpu_to_node(cpu))
+#else
+#define cpu_to_node_mask(cpu) (cpu_online_map)
+#endif
+
 /*
  * Convert user-nice values [ -20 ... 0 ... 19 ]
  * to static priority [ MAX_RT_PRIO..MAX_PRIO-1 ],
@@ -154,10 +160,9 @@ struct runqueue {
 	task_t *curr, *idle;
 	struct mm_struct *prev_mm;
 	prio_array_t *active, *expired, arrays[2];
-	int prev_nr_running[NR_CPUS];
+	int prev_cpu_load[NR_CPUS];
 #ifdef CONFIG_NUMA
 	atomic_t *node_nr_running;
-	unsigned int nr_balanced;
 	int prev_node_load[MAX_NUMNODES];
 #endif
 	task_t *migration_thread;
@@ -867,29 +872,6 @@ static int find_busiest_node(int this_node)
 	return node;
 }
 
-static inline unsigned long cpus_to_balance(int this_cpu, runqueue_t *this_rq)
-{
-	int this_node = cpu_to_node(this_cpu);
-	/*
-	 * Avoid rebalancing between nodes too often.
-	 * We rebalance globally once every NODE_BALANCE_RATE load balances.
-	 */
-	if (++(this_rq->nr_balanced) == NODE_BALANCE_RATE) {
-		int node = find_busiest_node(this_node);
-		this_rq->nr_balanced = 0;
-		if (node >= 0)
-			return (node_to_cpumask(node) | (1UL << this_cpu));
-	}
-	return node_to_cpumask(this_node);
-}
-
-#else /* !CONFIG_NUMA */
-
-static inline unsigned long cpus_to_balance(int this_cpu, runqueue_t *this_rq)
-{
-	return cpu_online_map;
-}
-
 #endif /* CONFIG_NUMA */
 
 #if CONFIG_SMP
@@ -909,10 +891,10 @@ static inline unsigned int double_lock_balance(runqueue_t *this_rq,
 			spin_lock(&busiest->lock);
 			spin_lock(&this_rq->lock);
 			/* Need to recalculate nr_running */
-			if (idle || (this_rq->nr_running > this_rq->prev_nr_running[this_cpu]))
+			if (idle || (this_rq->nr_running > this_rq->prev_cpu_load[this_cpu]))
 				nr_running = this_rq->nr_running;
 			else
-				nr_running = this_rq->prev_nr_running[this_cpu];
+				nr_running = this_rq->prev_cpu_load[this_cpu];
 		} else
 			spin_lock(&busiest->lock);
 	}
@@ -949,10 +931,10 @@ static inline runqueue_t *find_busiest_queue(runqueue_t *this_rq, int this_cpu,
 	 * that case we are less picky about moving a task across CPUs and
 	 * take what can be taken.
 	 */
-	if (idle || (this_rq->nr_running > this_rq->prev_nr_running[this_cpu]))
+	if (idle || (this_rq->nr_running > this_rq->prev_cpu_load[this_cpu]))
 		nr_running = this_rq->nr_running;
 	else
-		nr_running = this_rq->prev_nr_running[this_cpu];
+		nr_running = this_rq->prev_cpu_load[this_cpu];
 
 	busiest = NULL;
 	max_load = 1;
@@ -961,11 +943,11 @@ static inline runqueue_t *find_busiest_queue(runqueue_t *this_rq, int this_cpu,
 			continue;
 
 		rq_src = cpu_rq(i);
-		if (idle || (rq_src->nr_running < this_rq->prev_nr_running[i]))
+		if (idle || (rq_src->nr_running < this_rq->prev_cpu_load[i]))
 			load = rq_src->nr_running;
 		else
-			load = this_rq->prev_nr_running[i];
-		this_rq->prev_nr_running[i] = rq_src->nr_running;
+			load = this_rq->prev_cpu_load[i];
+		this_rq->prev_cpu_load[i] = rq_src->nr_running;
 
 		if ((load > max_load) && (rq_src != this_rq)) {
 			busiest = rq_src;
@@ -1029,7 +1011,7 @@ static inline void pull_task(runqueue_t *src_rq, prio_array_t *src_array, task_t
  * We call this with the current runqueue locked,
  * irqs disabled.
  */
-static void load_balance(runqueue_t *this_rq, int idle)
+static void load_balance(runqueue_t *this_rq, int idle, unsigned long cpumask)
 {
 	int imbalance, idx, this_cpu = smp_processor_id();
 	runqueue_t *busiest;
@@ -1037,8 +1019,7 @@ static void load_balance(runqueue_t *this_rq, int idle)
 	struct list_head *head, *curr;
 	task_t *tmp;
 
-	busiest = find_busiest_queue(this_rq, this_cpu, idle, &imbalance,
-					cpus_to_balance(this_cpu, this_rq));
+	busiest = find_busiest_queue(this_rq, this_cpu, idle, &imbalance, cpumask);
 	if (!busiest)
 		goto out;
 
@@ -1113,21 +1094,75 @@ static void load_balance(runqueue_t *this_rq, int idle)
  * frequency and balancing agressivity depends on whether the CPU is
  * idle or not.
  *
- * busy-rebalance every 250 msecs. idle-rebalance every 1 msec. (or on
+ * busy-rebalance every 200 msecs. idle-rebalance every 1 msec. (or on
  * systems with HZ=100, every 10 msecs.)
+ *
+ * On NUMA, do a node-rebalance every 400 msecs.
  */
-#define BUSY_REBALANCE_TICK (HZ/4 ?: 1)
 #define IDLE_REBALANCE_TICK (HZ/1000 ?: 1)
+#define BUSY_REBALANCE_TICK (HZ/5 ?: 1)
+#define IDLE_NODE_REBALANCE_TICK (IDLE_REBALANCE_TICK * 5)
+#define BUSY_NODE_REBALANCE_TICK (BUSY_REBALANCE_TICK * 100)
 
-static inline void idle_tick(runqueue_t *rq)
+#if CONFIG_NUMA
+static void balance_node(runqueue_t *this_rq, int idle, int this_cpu)
 {
-	if (jiffies % IDLE_REBALANCE_TICK)
-		return;
-	spin_lock(&rq->lock);
-	load_balance(rq, 1);
-	spin_unlock(&rq->lock);
+	int node = find_busiest_node(cpu_to_node(this_cpu));
+	unsigned long cpumask, this_cpumask = 1UL << this_cpu;
+
+	if (node >= 0) {
+		cpumask = node_to_cpumask(node) | this_cpumask;
+		spin_lock(&this_rq->lock);
+		load_balance(this_rq, idle, cpumask);
+		spin_unlock(&this_rq->lock);
+	}
 }
+#endif
 
+static void rebalance_tick(runqueue_t *this_rq, int idle)
+{
+#if CONFIG_NUMA
+	int this_cpu = smp_processor_id();
+#endif
+	unsigned long j = jiffies;
+
+	/*
+	 * First do inter-node rebalancing, then intra-node rebalancing,
+	 * if both events happen in the same tick. The inter-node
+	 * rebalancing does not necessarily have to create a perfect
+	 * balance within the node, since we load-balance the most loaded
+	 * node with the current CPU. (ie. other CPUs in the local node
+	 * are not balanced.)
+	 */
+	if (idle) {
+#if CONFIG_NUMA
+		if (!(j % IDLE_NODE_REBALANCE_TICK))
+			balance_node(this_rq, idle, this_cpu);
+#endif
+		if (!(j % IDLE_REBALANCE_TICK)) {
+			spin_lock(&this_rq->lock);
+			load_balance(this_rq, 0, cpu_to_node_mask(this_cpu));
+			spin_unlock(&this_rq->lock);
+		}
+		return;
+	}
+#if CONFIG_NUMA
+	if (!(j % BUSY_NODE_REBALANCE_TICK))
+		balance_node(this_rq, idle, this_cpu);
+#endif
+	if (!(j % BUSY_REBALANCE_TICK)) {
+		spin_lock(&this_rq->lock);
+		load_balance(this_rq, idle, cpu_to_node_mask(this_cpu));
+		spin_unlock(&this_rq->lock);
+	}
+}
+#else
+/*
+ * on UP we do not need to balance between CPUs:
+ */
+static inline void rebalance_tick(runqueue_t *this_rq, int idle)
+{
+}
 #endif
 
 DEFINE_PER_CPU(struct kernel_stat, kstat) = { { 0 } };
@@ -1170,9 +1205,7 @@ void scheduler_tick(int user_ticks, int sys_ticks)
 			kstat_cpu(cpu).cpustat.iowait += sys_ticks;
 		else
 			kstat_cpu(cpu).cpustat.idle += sys_ticks;
-#if CONFIG_SMP
-		idle_tick(rq);
-#endif
+		rebalance_tick(rq, 1);
 		return;
 	}
 	if (TASK_NICE(p) > 0)
@@ -1228,11 +1261,8 @@ void scheduler_tick(int user_ticks, int sys_ticks)
 			enqueue_task(p, rq->active);
 	}
 out:
-#if CONFIG_SMP
-	if (!(jiffies % BUSY_REBALANCE_TICK))
-		load_balance(rq, 0);
-#endif
 	spin_unlock(&rq->lock);
+	rebalance_tick(rq, 0);
 }
 
 void scheduling_functions_start_here(void) { }
@@ -1291,7 +1321,7 @@ asmlinkage void schedule(void)
 pick_next_task:
 	if (unlikely(!rq->nr_running)) {
 #if CONFIG_SMP
-		load_balance(rq, 1);
+		load_balance(rq, 1, cpu_to_node_mask(smp_processor_id()));
 		if (rq->nr_running)
 			goto pick_next_task;
 #endif