[tip: sched/core] sched/fair: Simplify the continue_balancing logic in sched_balance_newidle()

[tip-bot2 for Shrikanth Hegde]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     c829d6818b60c591f70c060b2bb75d76cf0cec6d
Gitweb:        https://git.kernel.org/tip/c829d6818b60c591f70c060b2bb75d76cf0cec6d
Author:        Shrikanth Hegde <sshegde@xxxxxxxxxxxxx>
AuthorDate:    Mon, 25 Mar 2024 21:09:26 +05:30
Committer:     Ingo Molnar <mingo@xxxxxxxxxx>
CommitterDate: Tue, 26 Mar 2024 20:16:20 +01:00

sched/fair: Simplify the continue_balancing logic in sched_balance_newidle()

newidle(CPU_NEWLY_IDLE) balancing doesn't stop the load-balancing if the
continue_balancing flag is reset, but the other two balancing (IDLE, BUSY)
cases do that.

newidle balance stops the load balancing if rq has a task or there
is wakeup pending. The same checks are present in should_we_balance for
newidle. Hence use the return value and simplify continue_balancing
mechanism for newidle. Update the comment surrounding it as well.

No change in functionality intended.

Signed-off-by: Shrikanth Hegde <sshegde@xxxxxxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@xxxxxxx>
Link: https://lore.kernel.org/r/20240325153926.274284-1-sshegde@xxxxxxxxxxxxx
---
 kernel/sched/fair.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 24a7530..1856e58 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -12358,6 +12358,7 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf)
 {
 	unsigned long next_balance = jiffies + HZ;
 	int this_cpu = this_rq->cpu;
+	int continue_balancing = 1;
 	u64 t0, t1, curr_cost = 0;
 	struct sched_domain *sd;
 	int pulled_task = 0;
@@ -12372,8 +12373,9 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf)
 		return 0;
 
 	/*
-	 * We must set idle_stamp _before_ calling idle_balance(), such that we
-	 * measure the duration of idle_balance() as idle time.
+	 * We must set idle_stamp _before_ calling sched_balance_rq()
+	 * for CPU_NEWLY_IDLE, such that we measure the this duration
+	 * as idle time.
 	 */
 	this_rq->idle_stamp = rq_clock(this_rq);
 
@@ -12412,7 +12414,6 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf)
 
 	rcu_read_lock();
 	for_each_domain(this_cpu, sd) {
-		int continue_balancing = 1;
 		u64 domain_cost;
 
 		update_next_balance(sd, &next_balance);
@@ -12438,8 +12439,7 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf)
 		 * Stop searching for tasks to pull if there are
 		 * now runnable tasks on this rq.
 		 */
-		if (pulled_task || this_rq->nr_running > 0 ||
-		    this_rq->ttwu_pending)
+		if (pulled_task || !continue_balancing)
 			break;
 	}
 	rcu_read_unlock();