From 2d26ff6c2b39eccfe45e08834b4a5fa8028803ea Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Fri, 8 Nov 2019 14:16:00 +0100
Subject: [PATCH] sched/core: Further clarify sched_class::set_next_task()

References: bnc#1155798 (CPU scheduler functional and performance backports)
Patch-mainline: v5.5-rc1
Git-commit: a0e813f26ebcb25c0b5e504498fbd796cca1a4ba

It turns out there really is something special to the first
set_next_task() invocation. In specific the 'change' pattern really
should not cause balance callbacks.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: juri.lelli@redhat.com
Cc: ktkhai@virtuozzo.com
Cc: mgorman@suse.de
Cc: qais.yousef@arm.com
Cc: qperret@google.com
Cc: rostedt@goodmis.org
Cc: valentin.schneider@arm.com
Cc: vincent.guittot@linaro.org
Fixes: f95d4eaee6d0 ("sched/{rt,deadline}: Fix set_next_task vs pick_next_task")
Link: https://lkml.kernel.org/r/20191108131909.775434698@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
 kernel/sched/deadline.c  | 7 +++++--
 kernel/sched/fair.c      | 2 +-
 kernel/sched/idle.c      | 4 ++--
 kernel/sched/rt.c        | 7 +++++--
 kernel/sched/sched.h     | 4 ++--
 kernel/sched/stop_task.c | 4 ++--
 6 files changed, 17 insertions(+), 11 deletions(-)

diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 9bf088e591ff..04315c56c3b5 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1743,13 +1743,16 @@ static void start_hrtick_dl(struct rq *rq, struct task_struct *p)
 }
 #endif
 
-static void set_next_task_dl(struct rq *rq, struct task_struct *p)
+static void set_next_task_dl(struct rq *rq, struct task_struct *p, bool first)
 {
 	p->se.exec_start = rq_clock_task(rq);
 
 	/* You can't push away the running task */
 	dequeue_pushable_dl_task(rq, p);
 
+	if (!first)
+		return;
+
 	if (hrtick_enabled(rq))
 		start_hrtick_dl(rq, p);
 
@@ -1782,7 +1785,7 @@ static struct task_struct *pick_next_task_dl(struct rq *rq)
 	dl_se = pick_next_dl_entity(rq, dl_rq);
 	BUG_ON(!dl_se);
 	p = dl_task_of(dl_se);
-	set_next_task_dl(rq, p);
+	set_next_task_dl(rq, p, true);
 	return p;
 }
 
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 3dfd81699f62..666e4c55cc5f 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -10360,7 +10360,7 @@ static void switched_to_fair(struct rq *rq, struct task_struct *p)
  * This routine is mostly called to set cfs_rq->curr field when a task
  * migrates between groups/classes.
  */
-static void set_next_task_fair(struct rq *rq, struct task_struct *p)
+static void set_next_task_fair(struct rq *rq, struct task_struct *p, bool first)
 {
 	struct sched_entity *se = &p->se;
 
diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index af42c08141e9..53154d01a639 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -385,7 +385,7 @@ static void put_prev_task_idle(struct rq *rq, struct task_struct *prev)
 {
 }
 
-static void set_next_task_idle(struct rq *rq, struct task_struct *next)
+static void set_next_task_idle(struct rq *rq, struct task_struct *next, bool first)
 {
 	update_idle_core(rq);
 	schedstat_inc(rq->sched_goidle);
@@ -395,7 +395,7 @@ struct task_struct *pick_next_task_idle(struct rq *rq)
 {
 	struct task_struct *next = rq->idle;
 
-	set_next_task_idle(rq, next);
+	set_next_task_idle(rq, next, true);
 
 	return next;
 }
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 9a1d9d5184ec..8466eadffa11 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1514,13 +1514,16 @@ static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int flag
 #endif
 }
 
-static inline void set_next_task_rt(struct rq *rq, struct task_struct *p)
+static inline void set_next_task_rt(struct rq *rq, struct task_struct *p, bool first)
 {
 	p->se.exec_start = rq_clock_task(rq);
 
 	/* The running task is never eligible for pushing */
 	dequeue_pushable_task(rq, p);
 
+	if (!first)
+		return;
+
 	/*
 	 * If prev task was rt, put_prev_task() has already updated the
 	 * utilization. We only care of the case where we start to schedule a
@@ -1571,7 +1574,7 @@ static struct task_struct *pick_next_task_rt(struct rq *rq)
 		return NULL;
 
 	p = _pick_next_task_rt(rq);
-	set_next_task_rt(rq, p);
+	set_next_task_rt(rq, p, true);
 	return p;
 }
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 18d54d7c1b75..8038c80bc77f 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1706,7 +1706,7 @@ struct sched_class {
 	struct task_struct *(*pick_next_task)(struct rq *rq);
 
 	void (*put_prev_task)(struct rq *rq, struct task_struct *p);
-	void (*set_next_task)(struct rq *rq, struct task_struct *p);
+	void (*set_next_task)(struct rq *rq, struct task_struct *p, bool first);
 
 #ifdef CONFIG_SMP
 	int (*balance)(struct rq *rq, struct task_struct *prev, struct rq_flags *rf);
@@ -1758,7 +1758,7 @@ static inline void put_prev_task(struct rq *rq, struct task_struct *prev)
 static inline void set_next_task(struct rq *rq, struct task_struct *next)
 {
 	WARN_ON_ONCE(rq->curr != next);
-	next->sched_class->set_next_task(rq, next);
+	next->sched_class->set_next_task(rq, next, false);
 }
 
 #ifdef CONFIG_SMP
diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
index 0aefdfb79b36..4c9e9975684f 100644
--- a/kernel/sched/stop_task.c
+++ b/kernel/sched/stop_task.c
@@ -29,7 +29,7 @@ check_preempt_curr_stop(struct rq *rq, struct task_struct *p, int flags)
 	/* we're never preempted */
 }
 
-static void set_next_task_stop(struct rq *rq, struct task_struct *stop)
+static void set_next_task_stop(struct rq *rq, struct task_struct *stop, bool first)
 {
 	stop->se.exec_start = rq_clock_task(rq);
 }
@@ -39,7 +39,7 @@ static struct task_struct *pick_next_task_stop(struct rq *rq)
 	if (!sched_stop_runnable(rq))
 		return NULL;
 
-	set_next_task_stop(rq, rq->stop);
+	set_next_task_stop(rq, rq->stop, true);
 	return rq->stop;
 }