Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Juri Lelli <juri.lelli@redhat.com>
To:     peterz@infradead.org, mingo@redhat.com, rostedt@goodmis.org,
        tj@kernel.org
Cc:     linux-kernel@vger.kernel.org, luca.abeni@santannapisa.it,
        claudio@evidence.eu.com, tommaso.cucinotta@santannapisa.it,
        bristot@redhat.com, mathieu.poirier@linaro.org, lizefan@huawei.com,
        longman@redhat.com, dietmar.eggemann@arm.com,
        cgroups@vger.kernel.org, Juri Lelli <juri.lelli@redhat.com>
Subject: [PATCH v9 4/8] sched/deadline: Fix bandwidth accounting at all levels after offline migration
Date:   Fri, 19 Jul 2019 15:59:56 +0200
Message-Id: <20190719140000.31694-5-juri.lelli@redhat.com>
In-Reply-To: <20190719140000.31694-1-juri.lelli@redhat.com>
References: <20190719140000.31694-1-juri.lelli@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

If a task happens to be throttled while the CPU it was running on gets
hotplugged off, the bandwidth associated with the task is not correctly
migrated with it when the replenishment timer fires (offline_migration).

Fix things up, for this_bw, running_bw and total_bw, when replenishment
timer fires and task is migrated (dl_task_offline_migration()).

Signed-off-by: Juri Lelli <juri.lelli@redhat.com>
---
 kernel/sched/deadline.c | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 4cedcf8d6b03..f0166ab8c6b4 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -529,6 +529,7 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq);
 static struct rq *dl_task_offline_migration(struct rq *rq, struct task_struct *p)
 {
 	struct rq *later_rq = NULL;
+	struct dl_bw *dl_b;
 
 	later_rq = find_lock_later_rq(p, rq);
 	if (!later_rq) {
@@ -557,6 +558,38 @@ static struct rq *dl_task_offline_migration(struct rq *rq, struct task_struct *p
 		double_lock_balance(rq, later_rq);
 	}
 
+	if (p->dl.dl_non_contending || p->dl.dl_throttled) {
+		/*
+		 * Inactive timer is armed (or callback is running, but
+		 * waiting for us to release rq locks). In any case, when it
+		 * will file (or continue), it will see running_bw of this
+		 * task migrated to later_rq (and correctly handle it).
+		 */
+		sub_running_bw(&p->dl, &rq->dl);
+		sub_rq_bw(&p->dl, &rq->dl);
+
+		add_rq_bw(&p->dl, &later_rq->dl);
+		add_running_bw(&p->dl, &later_rq->dl);
+	} else {
+		sub_rq_bw(&p->dl, &rq->dl);
+		add_rq_bw(&p->dl, &later_rq->dl);
+	}
+
+	/*
+	 * And we finally need to fixup root_domain(s) bandwidth accounting,
+	 * since p is still hanging out in the old (now moved to default) root
+	 * domain.
+	 */
+	dl_b = &rq->rd->dl_bw;
+	raw_spin_lock(&dl_b->lock);
+	__dl_sub(dl_b, p->dl.dl_bw, cpumask_weight(rq->rd->span));
+	raw_spin_unlock(&dl_b->lock);
+
+	dl_b = &later_rq->rd->dl_bw;
+	raw_spin_lock(&dl_b->lock);
+	__dl_add(dl_b, p->dl.dl_bw, cpumask_weight(later_rq->rd->span));
+	raw_spin_unlock(&dl_b->lock);
+
 	set_task_cpu(p, later_rq->cpu);
 	double_unlock_balance(later_rq, rq);
 
-- 
2.17.2