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;
Date:   Thu, 23 Jan 2020 16:28:10 -0800
Message-Id: <20200124002811.228334-1-wvw@google.com>
Mime-Version: 1.0
Subject: [PATCH] [RFC] sched: restrict iowait boost for boosted task only
From:   Wei Wang <wvw@google.com>
Cc:     wei.vince.wang@gmail.com, qais.yousef@arm.com,
        dietmar.eggemann@arm.com, valentin.schneider@arm.com,
        chris.redpath@arm.com, qperret@google.com,
        Wei Wang <wvw@google.com>, Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
To:     unlisted-recipients:; (no To-header on input)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Currently iowait doesn't distinguish background/foreground tasks.
This may not be a problem in servers but could impact devices that
are more sensitive on energy consumption. [1]
In Android world, we have seen many cases where device run to high and
inefficient frequency unnecessarily when running some background task
which are interacting I/O.

The ultimate goal is to only apply iowait boost on UX related tasks and
leave the rest for EAS scheduler for better efficiency.

This patch limits iowait boost for tasks which are associated with boost
signal from user land. On Android specifically, those are foreground or
top app tasks.

[1] ANDROID discussion:
https://android-review.googlesource.com/c/kernel/common/+/1215695

Signed-off-by: Wei Wang <wvw@google.com>
---
 kernel/sched/fair.c  |  4 +++-
 kernel/sched/sched.h | 12 ++++++++++++
 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index ba749f579714..221c0fbcea0e 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5220,8 +5220,10 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 	 * If in_iowait is set, the code below may not trigger any cpufreq
 	 * utilization updates, so do it here explicitly with the IOWAIT flag
 	 * passed.
+	 * If CONFIG_ENERGY_MODEL and CONFIG_UCLAMP_TASK are both configured,
+	 * only boost for tasks set with non-null min-clamp.
 	 */
-	if (p->in_iowait)
+	if (iowait_boosted(p))
 		cpufreq_update_util(rq, SCHED_CPUFREQ_IOWAIT);
 
 	for_each_sched_entity(se) {
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 280a3c735935..a13d49d835ed 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -2341,6 +2341,18 @@ static inline unsigned int uclamp_util(struct rq *rq, unsigned int util)
 }
 #endif /* CONFIG_UCLAMP_TASK */
 
+#if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_UCLAMP_TASK)
+static inline bool iowait_boosted(struct task_struct *p)
+{
+	return p->in_iowait && uclamp_eff_value(p, UCLAMP_MIN) > 0;
+}
+#else /* defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_UCLAMP_TASK) */
+static inline bool iowait_boosted(struct task_struct *p)
+{
+	return p->in_iowait;
+}
+#endif /* defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_UCLAMP_TASK) */
+
 #ifdef arch_scale_freq_capacity
 # ifndef arch_scale_freq_invariant
 #  define arch_scale_freq_invariant()	true
-- 
2.25.0.341.g760bfbb309-goog