From: Dietmar Eggemann <dietmar.eggemann@arm.com>
To: linux-kernel@vger.kernel.org,
Peter Zijlstra <peterz@infradead.org>,
Quentin Perret <quentin.perret@arm.com>,
Thara Gopinath <thara.gopinath@linaro.org>
Cc: linux-pm@vger.kernel.org,
Morten Rasmussen <morten.rasmussen@arm.com>,
Chris Redpath <chris.redpath@arm.com>,
Patrick Bellasi <patrick.bellasi@arm.com>,
Valentin Schneider <valentin.schneider@arm.com>,
"Rafael J . Wysocki" <rjw@rjwysocki.net>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Vincent Guittot <vincent.guittot@linaro.org>,
Viresh Kumar <viresh.kumar@linaro.org>,
Todd Kjos <tkjos@google.com>, Joel Fernandes <joelaf@google.com>
Subject: [RFC PATCH 4/6] sched/fair: Introduce an energy estimation helper function
Date: Tue, 20 Mar 2018 09:43:10 +0000 [thread overview]
Message-ID: <20180320094312.24081-5-dietmar.eggemann@arm.com> (raw)
In-Reply-To: <20180320094312.24081-1-dietmar.eggemann@arm.com>
From: Quentin Perret <quentin.perret@arm.com>
In preparation for the definition of an energy-aware wakeup path, a
helper function is provided to estimate the consequence on system energy
when a specific task wakes-up on a specific CPU. compute_energy()
estimates the OPPs to be reached by all frequency domains and estimates
the consumption of each online CPU according to its energy model and its
percentage of busy time.
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Quentin Perret <quentin.perret@arm.com>
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
---
kernel/sched/fair.c | 81 +++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 81 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6c72a5e7b1b0..76bd46502486 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6409,6 +6409,30 @@ static inline int cpu_overutilized(int cpu)
}
/*
+ * Returns the util of "cpu" if "p" wakes up on "dst_cpu".
+ */
+static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu)
+{
+ unsigned long util = cpu_rq(cpu)->cfs.avg.util_avg;
+ unsigned long capacity = capacity_orig_of(cpu);
+
+ /*
+ * If p is where it should be, or if it has no impact on cpu, there is
+ * not much to do.
+ */
+ if ((task_cpu(p) == dst_cpu) || (cpu != task_cpu(p) && cpu != dst_cpu))
+ goto clamp_util;
+
+ if (dst_cpu == cpu)
+ util += task_util(p);
+ else
+ util = max_t(long, util - task_util(p), 0);
+
+clamp_util:
+ return (util >= capacity) ? capacity : util;
+}
+
+/*
* Disable WAKE_AFFINE in the case where task @p doesn't fit in the
* capacity of either the waking CPU @cpu or the previous CPU @prev_cpu.
*
@@ -6432,6 +6456,63 @@ static int wake_cap(struct task_struct *p, int cpu, int prev_cpu)
return !util_fits_capacity(task_util(p), min_cap);
}
+static struct capacity_state *find_cap_state(int cpu, unsigned long util)
+{
+ struct sched_energy_model *em = *per_cpu_ptr(energy_model, cpu);
+ struct capacity_state *cs = NULL;
+ int i;
+
+ /*
+ * As the goal is to estimate the OPP reached for a specific util
+ * value, mimic the behaviour of schedutil with a 1.25 coefficient
+ */
+ util += util >> 2;
+
+ for (i = 0; i < em->nr_cap_states; i++) {
+ cs = &em->cap_states[i];
+ if (cs->cap >= util)
+ break;
+ }
+
+ return cs;
+}
+
+static unsigned long compute_energy(struct task_struct *p, int dst_cpu)
+{
+ unsigned long util, fdom_max_util;
+ struct capacity_state *cs;
+ unsigned long energy = 0;
+ struct freq_domain *fdom;
+ int cpu;
+
+ for_each_freq_domain(fdom) {
+ fdom_max_util = 0;
+ for_each_cpu_and(cpu, &(fdom->span), cpu_online_mask) {
+ util = cpu_util_next(cpu, p, dst_cpu);
+ fdom_max_util = max(util, fdom_max_util);
+ }
+
+ /*
+ * Here we assume that the capacity states of CPUs belonging to
+ * the same frequency domains are shared. Hence, we look at the
+ * capacity state of the first CPU and re-use it for all.
+ */
+ cpu = cpumask_first(&(fdom->span));
+ cs = find_cap_state(cpu, fdom_max_util);
+
+ /*
+ * The energy consumed by each CPU is derived from the power
+ * it dissipates at the expected OPP and its percentage of
+ * busy time.
+ */
+ for_each_cpu_and(cpu, &(fdom->span), cpu_online_mask) {
+ util = cpu_util_next(cpu, p, dst_cpu);
+ energy += cs->power * util / cs->cap;
+ }
+ }
+ return energy;
+}
+
/*
* select_task_rq_fair: Select target runqueue for the waking task in domains
* that have the 'sd_flag' flag set. In practice, this is SD_BALANCE_WAKE,
--
2.11.0
next prev parent reply other threads:[~2018-03-20 9:43 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-20 9:43 [RFC PATCH 0/6] Energy Aware Scheduling Dietmar Eggemann
2018-03-20 9:43 ` [RFC PATCH 1/6] sched/fair: Create util_fits_capacity() Dietmar Eggemann
2018-03-20 9:43 ` [RFC PATCH 2/6] sched: Introduce energy models of CPUs Dietmar Eggemann
2018-03-20 9:52 ` Greg Kroah-Hartman
2018-03-21 0:45 ` Quentin Perret
2018-03-25 13:48 ` Quentin Perret
2018-03-26 22:26 ` Dietmar Eggemann
2018-04-09 12:01 ` Peter Zijlstra
2018-04-09 13:45 ` Quentin Perret
2018-04-09 15:32 ` Peter Zijlstra
2018-04-09 16:42 ` Quentin Perret
2018-04-10 6:55 ` Rafael J. Wysocki
2018-04-10 9:31 ` Quentin Perret
2018-04-10 10:20 ` Rafael J. Wysocki
2018-03-20 9:43 ` [RFC PATCH 3/6] sched: Add over-utilization/tipping point indicator Dietmar Eggemann
2018-04-09 9:40 ` Peter Zijlstra
2018-04-09 9:47 ` Peter Zijlstra
2018-04-09 9:53 ` Dietmar Eggemann
2018-04-09 11:49 ` Peter Zijlstra
2018-03-20 9:43 ` Dietmar Eggemann [this message]
2018-03-21 9:04 ` [RFC PATCH 4/6] sched/fair: Introduce an energy estimation helper function Juri Lelli
2018-03-21 12:26 ` Patrick Bellasi
2018-03-21 12:59 ` Juri Lelli
2018-03-21 13:55 ` Quentin Perret
2018-03-21 15:15 ` Juri Lelli
2018-03-21 16:26 ` Morten Rasmussen
2018-03-21 17:02 ` Juri Lelli
2018-03-21 14:02 ` Quentin Perret
2018-03-21 21:15 ` Dietmar Eggemann
2018-03-21 12:39 ` Patrick Bellasi
2018-03-21 14:26 ` Quentin Perret
2018-03-21 14:50 ` Juri Lelli
2018-03-21 15:54 ` Patrick Bellasi
2018-03-22 5:05 ` Quentin Perret
2018-03-20 9:43 ` [RFC PATCH 5/6] sched/fair: Select an energy-efficient CPU on task wake-up Dietmar Eggemann
2018-03-21 15:35 ` Patrick Bellasi
2018-03-22 20:10 ` Joel Fernandes
2018-03-23 15:47 ` Morten Rasmussen
2018-03-24 1:13 ` Joel Fernandes
2018-03-24 1:34 ` Quentin Perret
2018-03-24 6:06 ` Joel Fernandes
2018-03-24 1:22 ` Quentin Perret
2018-03-25 1:52 ` Quentin Perret
2018-03-22 16:27 ` Joel Fernandes
2018-03-22 18:06 ` Patrick Bellasi
2018-03-22 20:19 ` Joel Fernandes
2018-03-24 1:47 ` Quentin Perret
2018-03-25 0:12 ` Joel Fernandes
2018-03-23 16:00 ` Morten Rasmussen
2018-03-24 0:36 ` Joel Fernandes
2018-03-25 1:38 ` Quentin Perret
2018-03-20 9:43 ` [RFC PATCH 6/6] drivers: base: arch_topology.c: Enable EAS for arm/arm64 platforms Dietmar Eggemann
2018-03-20 9:49 ` Greg Kroah-Hartman
2018-03-20 15:20 ` Dietmar Eggemann
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180320094312.24081-5-dietmar.eggemann@arm.com \
--to=dietmar.eggemann@arm.com \
--cc=chris.redpath@arm.com \
--cc=gregkh@linuxfoundation.org \
--cc=joelaf@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=morten.rasmussen@arm.com \
--cc=patrick.bellasi@arm.com \
--cc=peterz@infradead.org \
--cc=quentin.perret@arm.com \
--cc=rjw@rjwysocki.net \
--cc=thara.gopinath@linaro.org \
--cc=tkjos@google.com \
--cc=valentin.schneider@arm.com \
--cc=vincent.guittot@linaro.org \
--cc=viresh.kumar@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).