sched: debug, improve migration statistics

add new migration statistics when SCHED_DEBUG and SCHEDSTATS
is enabled. Available in /proc/<PID>/sched.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

include/linux/sched.h

@@ -931,6 +931,24 @@ struct sched_entity {
 	u64			block_max;
 	u64			exec_max;
 	u64			slice_max;
+
+	u64			nr_migrations;
+	u64			nr_migrations_cold;
+	u64			nr_failed_migrations_affine;
+	u64			nr_failed_migrations_running;
+	u64			nr_failed_migrations_hot;
+	u64			nr_forced_migrations;
+	u64			nr_forced2_migrations;
+
+	u64			nr_wakeups;
+	u64			nr_wakeups_sync;
+	u64			nr_wakeups_migrate;
+	u64			nr_wakeups_local;
+	u64			nr_wakeups_remote;
+	u64			nr_wakeups_affine;
+	u64			nr_wakeups_affine_attempts;
+	u64			nr_wakeups_passive;
+	u64			nr_wakeups_idle;
 #endif
 
 #ifdef CONFIG_FAIR_GROUP_SCHED

kernel/sched.c

@@ -1005,6 +1005,23 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 
 #ifdef CONFIG_SMP
 
+/*
+ * Is this task likely cache-hot:
+ */
+static inline int
+task_hot(struct task_struct *p, u64 now, struct sched_domain *sd)
+{
+	s64 delta;
+
+	if (p->sched_class != &fair_sched_class)
+		return 0;
+
+	delta = now - p->se.exec_start;
+
+	return delta < (s64)sysctl_sched_migration_cost;
+}
+
+
 void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 {
 	int old_cpu = task_cpu(p);
@@ -1022,6 +1039,11 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 		p->se.sleep_start -= clock_offset;
 	if (p->se.block_start)
 		p->se.block_start -= clock_offset;
+	if (old_cpu != new_cpu) {
+		schedstat_inc(p, se.nr_migrations);
+		if (task_hot(p, old_rq->clock, NULL))
+			schedstat_inc(p, se.nr_forced2_migrations);
+	}
 #endif
 	p->se.vruntime -= old_cfsrq->min_vruntime -
 					 new_cfsrq->min_vruntime;
@@ -1394,9 +1416,14 @@ static int wake_idle(int cpu, struct task_struct *p)
 		if (sd->flags & SD_WAKE_IDLE) {
 			cpus_and(tmp, sd->span, p->cpus_allowed);
 			for_each_cpu_mask(i, tmp) {
-				if (idle_cpu(i))
+				if (idle_cpu(i)) {
+					if (i != task_cpu(p)) {
+						schedstat_inc(p,
+							se.nr_wakeups_idle);
+					}
 					return i;
 				}
+			}
 		} else {
 			break;
 		}
@@ -1426,7 +1453,7 @@ static inline int wake_idle(int cpu, struct task_struct *p)
  */
 static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 {
-	int cpu, this_cpu, success = 0;
+	int cpu, orig_cpu, this_cpu, success = 0;
 	unsigned long flags;
 	long old_state;
 	struct rq *rq;
@@ -1445,6 +1472,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 		goto out_running;
 
 	cpu = task_cpu(p);
+	orig_cpu = cpu;
 	this_cpu = smp_processor_id();
 
 #ifdef CONFIG_SMP
@@ -1488,6 +1516,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 			unsigned long tl = this_load;
 			unsigned long tl_per_task;
 
+			schedstat_inc(p, se.nr_wakeups_affine_attempts);
 			tl_per_task = cpu_avg_load_per_task(this_cpu);
 
 			/*
@@ -1507,6 +1536,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 				 * there is no bad imbalance.
 				 */
 				schedstat_inc(this_sd, ttwu_move_affine);
+				schedstat_inc(p, se.nr_wakeups_affine);
 				goto out_set_cpu;
 			}
 		}
@@ -1518,6 +1548,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 		if (this_sd->flags & SD_WAKE_BALANCE) {
 			if (imbalance*this_load <= 100*load) {
 				schedstat_inc(this_sd, ttwu_move_balance);
+				schedstat_inc(p, se.nr_wakeups_passive);
 				goto out_set_cpu;
 			}
 		}
@@ -1543,6 +1574,15 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 
 out_activate:
 #endif /* CONFIG_SMP */
+	schedstat_inc(p, se.nr_wakeups);
+	if (sync)
+		schedstat_inc(p, se.nr_wakeups_sync);
+	if (orig_cpu != cpu)
+		schedstat_inc(p, se.nr_wakeups_migrate);
+	if (cpu == this_cpu)
+		schedstat_inc(p, se.nr_wakeups_local);
+	else
+		schedstat_inc(p, se.nr_wakeups_remote);
 	update_rq_clock(rq);
 	activate_task(rq, p, 1);
 	/*
@@ -2118,22 +2158,6 @@ static void pull_task(struct rq *src_rq, struct task_struct *p,
 	check_preempt_curr(this_rq, p);
 }
 
-/*
- * Is this task likely cache-hot:
- */
-static inline int
-task_hot(struct task_struct *p, u64 now, struct sched_domain *sd)
-{
-	s64 delta;
-
-	if (p->sched_class != &fair_sched_class)
-		return 0;
-
-	delta = now - p->se.exec_start;
-
-	return delta < (s64)sysctl_sched_migration_cost;
-}
-
 /*
  * can_migrate_task - may task p from runqueue rq be migrated to this_cpu?
  */
@@ -2148,12 +2172,16 @@ 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 (!cpu_isset(this_cpu, p->cpus_allowed)) {
+		schedstat_inc(p, se.nr_failed_migrations_affine);
 		return 0;
+	}
 	*all_pinned = 0;
 
-	if (task_running(rq, p))
+	if (task_running(rq, p)) {
+		schedstat_inc(p, se.nr_failed_migrations_running);
 		return 0;
+	}
 
 	/*
 	 * Aggressive migration if:
@@ -2163,14 +2191,18 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu,
 
 	if (sd->nr_balance_failed > sd->cache_nice_tries) {
 #ifdef CONFIG_SCHEDSTATS
-		if (task_hot(p, rq->clock, sd))
+		if (task_hot(p, rq->clock, sd)) {
 			schedstat_inc(sd, lb_hot_gained[idle]);
+			schedstat_inc(p, se.nr_forced_migrations);
+		}
 #endif
 		return 1;
 	}
 
-	if (task_hot(p, rq->clock, sd))
+	if (task_hot(p, rq->clock, sd)) {
+		schedstat_inc(p, se.nr_failed_migrations_hot);
 		return 0;
+	}
 	return 1;
 }
 

kernel/sched_debug.c

@@ -260,6 +260,7 @@ __initcall(init_sched_debug_procfs);
 
 void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 {
+	unsigned long nr_switches;
 	unsigned long flags;
 	int num_threads = 1;
 
@@ -273,8 +274,12 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 	SEQ_printf(m, "%s (%d, #threads: %d)\n", p->comm, p->pid, num_threads);
 	SEQ_printf(m,
 		"---------------------------------------------------------\n");
+#define __P(F) \
+	SEQ_printf(m, "%-35s:%21Ld\n", #F, (long long)F)
 #define P(F) \
 	SEQ_printf(m, "%-35s:%21Ld\n", #F, (long long)p->F)
+#define __PN(F) \
+	SEQ_printf(m, "%-35s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)F))
 #define PN(F) \
 	SEQ_printf(m, "%-35s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)p->F))
 
@@ -282,6 +287,8 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 	PN(se.vruntime);
 	PN(se.sum_exec_runtime);
 
+	nr_switches = p->nvcsw + p->nivcsw;
+
 #ifdef CONFIG_SCHEDSTATS
 	PN(se.wait_start);
 	PN(se.sleep_start);
@@ -292,14 +299,55 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 	PN(se.slice_max);
 	PN(se.wait_max);
 	P(sched_info.bkl_count);
+	P(se.nr_migrations);
+	P(se.nr_migrations_cold);
+	P(se.nr_failed_migrations_affine);
+	P(se.nr_failed_migrations_running);
+	P(se.nr_failed_migrations_hot);
+	P(se.nr_forced_migrations);
+	P(se.nr_forced2_migrations);
+	P(se.nr_wakeups);
+	P(se.nr_wakeups_sync);
+	P(se.nr_wakeups_migrate);
+	P(se.nr_wakeups_local);
+	P(se.nr_wakeups_remote);
+	P(se.nr_wakeups_affine);
+	P(se.nr_wakeups_affine_attempts);
+	P(se.nr_wakeups_passive);
+	P(se.nr_wakeups_idle);
+
+	{
+		u64 avg_atom, avg_per_cpu;
+
+		avg_atom = p->se.sum_exec_runtime;
+		if (nr_switches)
+			do_div(avg_atom, nr_switches);
+		else
+			avg_atom = -1LL;
+
+		avg_per_cpu = p->se.sum_exec_runtime;
+		if (p->se.nr_migrations)
+			avg_per_cpu = div64_64(avg_per_cpu, p->se.nr_migrations);
+		else
+			avg_per_cpu = -1LL;
+
+		__PN(avg_atom);
+		__PN(avg_per_cpu);
+	}
 #endif
+	__P(nr_switches);
+	SEQ_printf(m, "%-35s:%21Ld\n",
+		   "nr_voluntary_switches", (long long)p->nvcsw);
 	SEQ_printf(m, "%-35s:%21Ld\n",
-		   "nr_switches", (long long)(p->nvcsw + p->nivcsw));
+		   "nr_involuntary_switches", (long long)p->nivcsw);
+
 	P(se.load.weight);
 	P(policy);
 	P(prio);
-#undef P
 #undef PN
+#undef __PN
+#undef P
+#undef __P
 
 	{
 		u64 t0, t1;
@@ -314,13 +362,32 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 void proc_sched_set_task(struct task_struct *p)
 {
 #ifdef CONFIG_SCHEDSTATS
+	p->se.wait_max				= 0;
 	p->se.sleep_max				= 0;
+	p->se.sum_sleep_runtime			= 0;
 	p->se.block_max				= 0;
 	p->se.exec_max				= 0;
 	p->se.slice_max				= 0;
-	p->se.wait_max			= 0;
+	p->se.nr_migrations			= 0;
+	p->se.nr_migrations_cold		= 0;
+	p->se.nr_failed_migrations_affine	= 0;
+	p->se.nr_failed_migrations_running	= 0;
+	p->se.nr_failed_migrations_hot		= 0;
+	p->se.nr_forced_migrations		= 0;
+	p->se.nr_forced2_migrations		= 0;
+	p->se.nr_wakeups			= 0;
+	p->se.nr_wakeups_sync			= 0;
+	p->se.nr_wakeups_migrate		= 0;
+	p->se.nr_wakeups_local			= 0;
+	p->se.nr_wakeups_remote			= 0;
+	p->se.nr_wakeups_affine			= 0;
+	p->se.nr_wakeups_affine_attempts	= 0;
+	p->se.nr_wakeups_passive		= 0;
+	p->se.nr_wakeups_idle			= 0;
 	p->sched_info.bkl_count			= 0;
 #endif
 	p->se.sum_exec_runtime			= 0;
 	p->se.prev_sum_exec_runtime		= 0;
+	p->nvcsw				= 0;
+	p->nivcsw				= 0;
 }