[PATCH v3] cpufreq: schedutil: rate limits for SCHED_DEADLINE

[PATCH v3] cpufreq: schedutil: rate limits for SCHED_DEADLINE

[Claudio Scordino]

When the SCHED_DEADLINE scheduling class increases the CPU utilization,
we should not wait for the rate limit, otherwise we 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>
CC: Ingo Molnar <mingo@redhat.com>
CC: Patrick Bellasi <patrick.bellasi@arm.com>
CC: Dietmar Eggemann <dietmar.eggemann@arm.com>
CC: Morten Rasmussen <morten.rasmussen@arm.com>
CC: Juri Lelli <juri.lelli@redhat.com>
CC: Viresh Kumar <viresh.kumar@linaro.org>
CC: Vincent Guittot <vincent.guittot@linaro.org>
CC: Todd Kjos <tkjos@android.com>
CC: Joel Fernandes <joelaf@google.com>
CC: linux-pm@vger.kernel.org
CC: linux-kernel@vger.kernel.org
---
Changes from v2:
 - Rate limit ignored also in case of "fast switch"
 - Specific routine added
---
Changes from v1:
 - Logic moved from sugov_should_update_freq() to
   sugov_update_single()/_shared() to not duplicate data structures
 - Rate limit not ignored in case of "fast switch"
---
 kernel/sched/cpufreq_schedutil.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 7936f54..13f9cce 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -260,6 +260,17 @@ 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 set_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)
 {
@@ -273,6 +284,8 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
 	sugov_set_iowait_boost(sg_cpu, time);
 	sg_cpu->last_update = time;
 
+	set_dl_rate_limit(sg_cpu, sg_policy);
+
 	if (!sugov_should_update_freq(sg_policy, time))
 		return;
 
@@ -354,6 +367,8 @@ static void sugov_update_shared(struct update_util_data *hook, u64 time,
 
 	raw_spin_lock(&sg_policy->update_lock);
 
+	set_dl_rate_limit(sg_cpu, sg_policy);
+
 	sugov_get_util(sg_cpu);
 	sg_cpu->flags = flags;
 
-- 
2.7.4

Re: [PATCH v3] cpufreq: schedutil: rate limits for SCHED_DEADLINE

[Viresh Kumar]

On 08-03-18, 11:29, Claudio Scordino wrote:
> When the SCHED_DEADLINE scheduling class increases the CPU utilization,
> we should not wait for the rate limit, otherwise we 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>
> CC: Ingo Molnar <mingo@redhat.com>
> CC: Patrick Bellasi <patrick.bellasi@arm.com>
> CC: Dietmar Eggemann <dietmar.eggemann@arm.com>
> CC: Morten Rasmussen <morten.rasmussen@arm.com>
> CC: Juri Lelli <juri.lelli@redhat.com>
> CC: Viresh Kumar <viresh.kumar@linaro.org>
> CC: Vincent Guittot <vincent.guittot@linaro.org>
> CC: Todd Kjos <tkjos@android.com>
> CC: Joel Fernandes <joelaf@google.com>
> CC: linux-pm@vger.kernel.org
> CC: linux-kernel@vger.kernel.org
> ---
> Changes from v2:
>  - Rate limit ignored also in case of "fast switch"
>  - Specific routine added
> ---
> Changes from v1:
>  - Logic moved from sugov_should_update_freq() to
>    sugov_update_single()/_shared() to not duplicate data structures
>  - Rate limit not ignored in case of "fast switch"
> ---
>  kernel/sched/cpufreq_schedutil.c | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> index 7936f54..13f9cce 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -260,6 +260,17 @@ 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 set_dl_rate_limit(struct sugov_cpu *sg_cpu, struct sugov_policy *sg_policy)

Maybe it could be renamed as ignore_dl_rate_limit() ? Lets see what others have
to say. But looks fine otherwise.

Acked-by: Viresh Kumar <viresh.kumar@linaro.org>

-- 
viresh

Re: [PATCH v3] cpufreq: schedutil: rate limits for SCHED_DEADLINE

[Rafael J. Wysocki]

On Thu, Mar 8, 2018 at 11:29 AM, Claudio Scordino
<claudio@evidence.eu.com> wrote:
> When the SCHED_DEADLINE scheduling class increases the CPU utilization,
> we should not wait for the rate limit, otherwise we 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>
> CC: Ingo Molnar <mingo@redhat.com>
> CC: Patrick Bellasi <patrick.bellasi@arm.com>
> CC: Dietmar Eggemann <dietmar.eggemann@arm.com>
> CC: Morten Rasmussen <morten.rasmussen@arm.com>
> CC: Juri Lelli <juri.lelli@redhat.com>
> CC: Viresh Kumar <viresh.kumar@linaro.org>
> CC: Vincent Guittot <vincent.guittot@linaro.org>
> CC: Todd Kjos <tkjos@android.com>
> CC: Joel Fernandes <joelaf@google.com>
> CC: linux-pm@vger.kernel.org
> CC: linux-kernel@vger.kernel.org
> ---
> Changes from v2:
>  - Rate limit ignored also in case of "fast switch"
>  - Specific routine added
> ---
> Changes from v1:
>  - Logic moved from sugov_should_update_freq() to
>    sugov_update_single()/_shared() to not duplicate data structures
>  - Rate limit not ignored in case of "fast switch"
> ---
>  kernel/sched/cpufreq_schedutil.c | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
>
> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> index 7936f54..13f9cce 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -260,6 +260,17 @@ 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

I wouldn't break the line here

> +void set_dl_rate_limit(struct sugov_cpu *sg_cpu, struct sugov_policy *sg_policy)

and the name might be better as Viresh said, but overall

Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

> +{
> +       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)
>  {
> @@ -273,6 +284,8 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
>         sugov_set_iowait_boost(sg_cpu, time);
>         sg_cpu->last_update = time;
>
> +       set_dl_rate_limit(sg_cpu, sg_policy);
> +
>         if (!sugov_should_update_freq(sg_policy, time))
>                 return;
>
> @@ -354,6 +367,8 @@ static void sugov_update_shared(struct update_util_data *hook, u64 time,
>
>         raw_spin_lock(&sg_policy->update_lock);
>
> +       set_dl_rate_limit(sg_cpu, sg_policy);
> +
>         sugov_get_util(sg_cpu);
>         sg_cpu->flags = flags;
>
> --
> 2.7.4
>