[PATCH v2 07/13] sched/fair: util_est: Take into account periodic tasks

[Qais Yousef <qyousef@layalina.io>]

The new faster rampup is great for performance. But terrible for power.
We want the faster rampup to be only applied for tasks that are
transitioning from one periodic/steady state to another periodic/steady
state. But if they are stably periodic, then the faster rampup doesn't
make sense as util_avg describes their computational demand accurately
and we can rely on that to make accurate decision. And preserve the
power savings from being exact with the resources we give to this task
(ie: smaller DVFS headroom).

We detect periodic tasks based on util_avg across util_est_update()
calls. If it is rising, then the task is going through a transition.

We rely on util_avg being stable for periodic tasks with very little
variations around one stable point.

Signed-off-by: Qais Yousef <qyousef@layalina.io>
---
 include/linux/sched.h |  2 ++
 kernel/sched/core.c   |  2 ++
 kernel/sched/fair.c   | 35 ++++++++++++++++++++++++-----------
 3 files changed, 28 insertions(+), 11 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index b61da16861e7..70517497e80b 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -907,6 +907,8 @@ struct task_struct {
 	struct uclamp_se		uclamp[UCLAMP_CNT];
 #endif
 
+	unsigned long			util_avg_dequeued;
+
 	struct sched_statistics         stats;
 
 #ifdef CONFIG_PREEMPT_NOTIFIERS
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index fe14fd4a2d53..82189bdc85b7 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -4409,6 +4409,8 @@ static void __sched_fork(u64 clone_flags, struct task_struct *p)
 #endif
 #endif
 
+	p->util_avg_dequeued		= 0;
+
 #ifdef CONFIG_SCHEDSTATS
 	/* Even if schedstat is disabled, there should not be garbage */
 	memset(&p->stats, 0, sizeof(p->stats));
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index c6363ec5de9d..d9729da3901a 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5041,6 +5041,11 @@ static inline unsigned long task_util(struct task_struct *p)
 	return READ_ONCE(p->se.avg.util_avg);
 }
 
+static inline unsigned long task_util_dequeued(struct task_struct *p)
+{
+	return READ_ONCE(p->util_avg_dequeued);
+}
+
 static inline unsigned long task_runnable(struct task_struct *p)
 {
 	return READ_ONCE(p->se.avg.runnable_avg);
@@ -5108,18 +5113,22 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
 	 * quickly to settle down to our new util_avg.
 	 */
 	if (!task_sleep) {
-		u64 delta = p->se.delta_exec;
-		unsigned int prev_ewma = ewma & ~UTIL_AVG_UNCHANGED;
+		if (task_util(p) > task_util_dequeued(p) &&
+		    task_util(p) - task_util_dequeued(p) > UTIL_EST_MARGIN) {
+			u64 delta = p->se.delta_exec;
+			unsigned int prev_ewma = ewma & ~UTIL_AVG_UNCHANGED;
 
-		do_div(delta, 1000);
-		ewma = approximate_util_avg(prev_ewma, delta);
-		/*
-		 * Keep accumulating delta_exec if it is too small to cause
-		 * a change.
-		 */
-		if (ewma != prev_ewma)
-			p->se.delta_exec = 0;
-		goto done;
+			do_div(delta, 1000);
+			ewma = approximate_util_avg(prev_ewma, delta);
+			/*
+			 * Keep accumulating delta_exec if it is too small to cause
+			 * a change.
+			 */
+			if (ewma != prev_ewma)
+				p->se.delta_exec = 0;
+			goto done_running;
+		}
+		return;
 	} else {
 		p->se.delta_exec = 0;
 	}
@@ -5134,6 +5143,9 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
 	/* Get utilization at dequeue */
 	dequeued = task_util(p);
 
+	if (!task_on_rq_migrating(p))
+		p->util_avg_dequeued = dequeued;
+
 	/*
 	 * Reset EWMA on utilization increases, the moving average is used only
 	 * to smooth utilization decreases.
@@ -5180,6 +5192,7 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
 	ewma >>= UTIL_EST_WEIGHT_SHIFT;
 done:
 	ewma |= UTIL_AVG_UNCHANGED;
+done_running:
 	WRITE_ONCE(p->se.avg.util_est, ewma);
 
 	trace_sched_util_est_se_tp(&p->se);
-- 
2.34.1