sched/cpufreq: Rate limits for SCHED_DEADLINE
authorClaudio Scordino <claudio@evidence.eu.com>
Tue, 13 Mar 2018 10:35:40 +0000 (11:35 +0100)
committerThomas Gleixner <tglx@linutronix.de>
Fri, 23 Mar 2018 21:48:22 +0000 (22:48 +0100)
When the SCHED_DEADLINE scheduling class increases the CPU utilization, it
should not wait for the rate limit, otherwise it may miss some deadline.

Tests using rt-app on Exynos5422 with up to 10 SCHED_DEADLINE tasks have
shown reductions of even 10% of deadline misses with a negligible
increase of energy consumption (measured through Baylibre Cape).

Signed-off-by: Claudio Scordino <claudio@evidence.eu.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: linux-pm@vger.kernel.org
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Patrick Bellasi <patrick.bellasi@arm.com>
Cc: Todd Kjos <tkjos@android.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lkml.kernel.org/r/1520937340-2755-1-git-send-email-claudio@evidence.eu.com
kernel/sched/cpufreq_schedutil.c

index 89fe78ecb88ca431064bb10b30dbabe5a7f70a1d..2b124811947d896516298a9c7f314e181c38abd9 100644 (file)
@@ -267,6 +267,16 @@ static bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu)
 static inline bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu) { return false; }
 #endif /* CONFIG_NO_HZ_COMMON */
 
+/*
+ * Make sugov_should_update_freq() ignore the rate limit when DL
+ * has increased the utilization.
+ */
+static inline void ignore_dl_rate_limit(struct sugov_cpu *sg_cpu, struct sugov_policy *sg_policy)
+{
+       if (cpu_util_dl(cpu_rq(sg_cpu->cpu)) > sg_cpu->util_dl)
+               sg_policy->need_freq_update = true;
+}
+
 static void sugov_update_single(struct update_util_data *hook, u64 time,
                                unsigned int flags)
 {
@@ -279,6 +289,8 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
        sugov_set_iowait_boost(sg_cpu, time, flags);
        sg_cpu->last_update = time;
 
+       ignore_dl_rate_limit(sg_cpu, sg_policy);
+
        if (!sugov_should_update_freq(sg_policy, time))
                return;
 
@@ -356,6 +368,8 @@ sugov_update_shared(struct update_util_data *hook, u64 time, unsigned int flags)
        sugov_set_iowait_boost(sg_cpu, time, flags);
        sg_cpu->last_update = time;
 
+       ignore_dl_rate_limit(sg_cpu, sg_policy);
+
        if (sugov_should_update_freq(sg_policy, time)) {
                next_f = sugov_next_freq_shared(sg_cpu, time);
                sugov_update_commit(sg_policy, time, next_f);