Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Scott Wood <swood@redhat.com>
To:     Steven Rostedt <rostedt@goodmis.org>,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Vincent Guittot <vincent.guittot@linaro.org>
Cc:     Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Rik van Riel <riel@surriel.com>,
        Mel Gorman <mgorman@suse.de>, linux-kernel@vger.kernel.org,
        linux-rt-users <linux-rt-users@vger.kernel.org>
Subject: [RFC PATCH 3/3] sched,rt: break out of load balancing if an RT task appears
Date:   Tue, 28 Apr 2020 00:02:42 -0500
Message-Id: <20200428050242.17717-4-swood@redhat.com>
In-Reply-To: <20200428050242.17717-1-swood@redhat.com>
References: <20200428050242.17717-1-swood@redhat.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

From: Rik van Riel <riel@redhat.com>

Bugzilla: 1331562

The CFS load balancer can take a little while, to the point of
it having a special LBF_NEED_BREAK flag, when the task moving
code takes a breather.

However, at that point it will jump right back in to load balancing,
without checking whether the CPU has gained any runnable real time
(or deadline) tasks.

Only idle_balance used to check for runnable real time tasks on a
CPU. This patch moves that check into a separate inline function,
and calls that function in load_balance, at approximately the same
granularity that LBF_NEED_BREAK happens.

Besides breaking out of load_balance, this patch also clears
continue_balancing, in order for rebalance_domains to break out
of its loop when a realtime task becomes runnable.

Signed-off-by: Rik van Riel <riel@redhat.com>
Reported-by: Clark Williams <williams@redhat.com>
Signed-off-by: Clark Williams <williams@redhat.com>
---
 kernel/sched/fair.c  | 19 +++++++++++++++++--
 kernel/sched/sched.h |  6 ++++++
 2 files changed, 23 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index dfde7f0ce3db..e7437e4e40b4 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9377,10 +9377,16 @@ voluntary_active_balance(struct lb_env *env)
 	return 0;
 }
=20
-static int need_active_balance(struct lb_env *env)
+static int need_active_balance(struct lb_env *env, int *continue_balanci=
ng)
 {
 	struct sched_domain *sd =3D env->sd;
=20
+	/* Run the realtime task now; load balance later. */
+	if (rq_has_runnable_rt_task(env->dst_rq)) {
+		*continue_balancing =3D 0;
+		return 0;
+	}
+
 	if (voluntary_active_balance(env))
 		return 1;
=20
@@ -9394,6 +9400,10 @@ static int should_we_balance(struct lb_env *env)
 	struct sched_group *sg =3D env->sd->groups;
 	int cpu, balance_cpu =3D -1;
=20
+	/* Run the realtime task now; load balance later. */
+	if (rq_has_runnable_rt_task(env->dst_rq))
+		return 0;
+
 	/*
 	 * Ensure the balancing environment is consistent; can happen
 	 * when the softirq triggers 'during' hotplug.
@@ -9521,6 +9531,11 @@ static int load_balance(int this_cpu, struct rq *t=
his_rq,
=20
 		local_irq_restore(rf.flags);
=20
+		if (rq_has_runnable_rt_task(this_rq)) {
+			*continue_balancing =3D 0;
+			goto out;
+		}
+
 		if (env.flags & LBF_NEED_BREAK) {
 			env.flags &=3D ~LBF_NEED_BREAK;
 			goto more_balance;
@@ -9604,7 +9619,7 @@ static int load_balance(int this_cpu, struct rq *th=
is_rq,
 		if (idle !=3D CPU_NEWLY_IDLE)
 			sd->nr_balance_failed++;
=20
-		if (need_active_balance(&env)) {
+		if (need_active_balance(&env, continue_balancing)) {
 			unsigned long flags;
=20
 			raw_spin_lock_irqsave(&busiest->lock, flags);
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 3d97c51544d7..a2a01dfd2bea 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1878,6 +1878,12 @@ static inline struct cpuidle_state *idle_get_state=
(struct rq *rq)
=20
 	return rq->idle_state;
 }
+
+/* Is there a task of a high priority class? */
+static inline bool rq_has_runnable_rt_task(struct rq *rq)
+{
+	return unlikely(rq->nr_running !=3D rq->cfs.h_nr_running);
+}
 #else
 static inline void idle_set_state(struct rq *rq,
 				  struct cpuidle_state *idle_state)
--=20
2.18.2