Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed,  8 Jun 2022 19:55:15 -0700
Message-Id: <20220609025515.2086253-1-joshdon@google.com>
Mime-Version: 1.0
Subject: [PATCH] sched: allow newidle balancing to bail out of load_balance
From:   Josh Don <joshdon@google.com>
To:     Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>
Cc:     Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Valentin Schneider <vschneid@redhat.com>,
        linux-kernel@vger.kernel.org, Josh Don <joshdon@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

While doing newidle load balancing, it is possible for new tasks to
arrive, such as with pending wakeups. newidle_balance() already accounts
for this by exiting the sched_domain load_balance() iteration if it
detects these cases. This is very important for minimizing wakeup
latency.

However, if we are already in load_balance(), we may stay there for a
while before returning back to newidle_balance(). This is most
exacerbated if we enter a 'goto redo' loop in the LBF_ALL_PINNED case. A
very straightforward workaround to this is to adjust should_we_balance()
to bail out if we're doing a CPU_NEWLY_IDLE balance and new tasks are
detected.

This was tested with the following reproduction:
- two threads that take turns sleeping and waking each other up are
  affined to two cores
- a large number of threads with 100% utilization are pinned to all
  other cores

Without this patch, wakeup latency was ~120us for the pair of threads,
almost entirely spent in load_balance(). With this patch, wakeup latency
is ~6us.

Signed-off-by: Josh Don <joshdon@google.com>
---
 kernel/sched/fair.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 8c5b74f66bd3..5abf30117824 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9834,9 +9834,15 @@ static int should_we_balance(struct lb_env *env)
 	/*
 	 * In the newly idle case, we will allow all the CPUs
 	 * to do the newly idle load balance.
+	 *
+	 * However, we bail out if we already have tasks or a wakeup pending,
+	 * to optimize wakeup latency.
 	 */
-	if (env->idle == CPU_NEWLY_IDLE)
+	if (env->idle == CPU_NEWLY_IDLE) {
+		if (env->dst_rq->nr_running > 0 || env->dst_rq->ttwu_pending)
+			return 0;
 		return 1;
+	}
 
 	/* Try to find first idle CPU */
 	for_each_cpu_and(cpu, group_balance_mask(sg), env->cpus) {
-- 
2.36.1.476.g0c4daa206d-goog