[tip: sched/core] sched/fair: Update ->next_balance correctly during newidle balance

[tip-bot2 for Vineeth Pillai (Google)]

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

Commit-ID:     1c0482707c42960ec46b88aadd6bffca8685db11
Gitweb:        https://git.kernel.org/tip/1c0482707c42960ec46b88aadd6bffca8685db11
Author:        Vineeth Pillai (Google) <vineeth@xxxxxxxxxxxxxxx>
AuthorDate:    Fri, 20 Oct 2023 01:40:28 
Committer:     Ingo Molnar <mingo@xxxxxxxxxx>
CommitterDate: Fri, 20 Oct 2023 09:56:21 +02:00

sched/fair: Update ->next_balance correctly during newidle balance

When newidle balancing triggers, we see that it constantly clobbers
rq->next_balance even when there is no newidle balance happening due to
the cost estimates.  Due to this, we see that periodic load balance
(rebalance_domains) may trigger way more often when the CPU is going in
and out of idle at a high rate but is no really idle. Repeatedly
triggering load balance there is a bad idea as it is a heavy operation.
It also causes increases in softirq.

Another issue is ->last_balance is not updated after newidle balance
causing mistakes in the ->next_balance calculations.

Fix by updating last_balance when a newidle load balance actually
happens and then updating next_balance. This is also how it is done in
other load balance paths.

Testing shows a significant drop in softirqs when running:

  $ cyclictest -i 100 -d 100 --latency=1000 -D 5 -t -m  -q

... goes from ~6,000 to ~800.

Co-developed-by: Joel Fernandes (Google) <joel@xxxxxxxxxxxxxxxxx>
Signed-off-by: Vineeth Pillai (Google) <vineeth@xxxxxxxxxxxxxxx>
Signed-off-by: Joel Fernandes (Google) <joel@xxxxxxxxxxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Juri Lelli <juri.lelli@xxxxxxxxxx>
Cc: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
Cc: Dietmar Eggemann <dietmar.eggemann@xxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: Daniel Bristot de Oliveira <bristot@xxxxxxxxxx>
Cc: Valentin Schneider <vschneid@xxxxxxxxxx>
Link: https://lore.kernel.org/r/20231020014031.919742-3-joel@xxxxxxxxxxxxxxxxx
---
 kernel/sched/fair.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 8c486ff..393d0dc 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -12122,11 +12122,7 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 
 	if (!READ_ONCE(this_rq->rd->overload) ||
 	    (sd && this_rq->avg_idle < sd->max_newidle_lb_cost)) {
-
-		if (sd)
-			update_next_balance(sd, &next_balance);
 		rcu_read_unlock();
-
 		goto out;
 	}
 	rcu_read_unlock();
@@ -12141,8 +12137,6 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 		int continue_balancing = 1;
 		u64 domain_cost;
 
-		update_next_balance(sd, &next_balance);
-
 		if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost)
 			break;
 
@@ -12155,6 +12149,8 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 			t1 = sched_clock_cpu(this_cpu);
 			domain_cost = t1 - t0;
 			update_newidle_cost(sd, domain_cost);
+			sd->last_balance = jiffies;
+			update_next_balance(sd, &next_balance);
 
 			curr_cost += domain_cost;
 			t0 = t1;