From: tip-bot for Vincent Guittot <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: viresh.kumar@linaro.org, peterz@infradead.org,
tglx@linutronix.de, torvalds@linux-foundation.org, hpa@zytor.com,
linux-kernel@vger.kernel.org, vincent.guittot@linaro.org,
mingo@kernel.org
Subject: [tip:sched/core] cpufreq/schedutil: Take time spent in interrupts into account
Date: Sun, 15 Jul 2018 16:29:37 -0700 [thread overview]
Message-ID: <tip-9033ea11889f88f243445495f72441e22256d5e9@git.kernel.org> (raw)
In-Reply-To: <1530200714-4504-8-git-send-email-vincent.guittot@linaro.org>
Commit-ID: 9033ea11889f88f243445495f72441e22256d5e9
Gitweb: https://git.kernel.org/tip/9033ea11889f88f243445495f72441e22256d5e9
Author: Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate: Thu, 28 Jun 2018 17:45:10 +0200
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Sun, 15 Jul 2018 23:51:21 +0200
cpufreq/schedutil: Take time spent in interrupts into account
The time spent executing IRQ handlers can be significant but it is not reflected
in the utilization of CPU when deciding to choose an OPP. Now that we have
access to this metric, schedutil can take it into account when selecting
the OPP for a CPU.
RQS utilization don't see the time spend under interrupt context and report
their value in the normal context time window. We need to compensate this when
adding interrupt utilization
The CPU utilization is:
IRQ util_avg + (1 - IRQ util_avg / max capacity ) * /Sum rq util_avg
A test with iperf on hikey (octo arm64) gives the following speedup:
iperf -c server_address -r -t 5
w/o patch w/ patch
Tx 276 Mbits/sec 304 Mbits/sec +10%
Rx 299 Mbits/sec 328 Mbits/sec +9%
8 iterations
stdev is lower than 1%
Only WFI idle state is enabled (shallowest idle state).
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: claudio@evidence.eu.com
Cc: daniel.lezcano@linaro.org
Cc: dietmar.eggemann@arm.com
Cc: joel@joelfernandes.org
Cc: juri.lelli@redhat.com
Cc: luca.abeni@santannapisa.it
Cc: patrick.bellasi@arm.com
Cc: quentin.perret@arm.com
Cc: rjw@rjwysocki.net
Cc: valentin.schneider@arm.com
Link: http://lkml.kernel.org/r/1530200714-4504-8-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/sched/cpufreq_schedutil.c | 25 +++++++++++++++++++++----
kernel/sched/sched.h | 13 +++++++++++++
2 files changed, 34 insertions(+), 4 deletions(-)
diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 07760bc7f69a..7016bde9d194 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -58,6 +58,7 @@ struct sugov_cpu {
unsigned long util_dl;
unsigned long bw_dl;
unsigned long util_rt;
+ unsigned long util_irq;
unsigned long max;
/* The field below is for single-CPU policies only: */
@@ -190,21 +191,30 @@ static void sugov_get_util(struct sugov_cpu *sg_cpu)
sg_cpu->util_dl = cpu_util_dl(rq);
sg_cpu->bw_dl = cpu_bw_dl(rq);
sg_cpu->util_rt = cpu_util_rt(rq);
+ sg_cpu->util_irq = cpu_util_irq(rq);
}
static unsigned long sugov_aggregate_util(struct sugov_cpu *sg_cpu)
{
struct rq *rq = cpu_rq(sg_cpu->cpu);
- unsigned long util;
+ unsigned long util, max = sg_cpu->max;
if (rt_rq_is_runnable(&rq->rt))
return sg_cpu->max;
+ if (unlikely(sg_cpu->util_irq >= max))
+ return max;
+
+ /* Sum rq utilization */
util = sg_cpu->util_cfs;
util += sg_cpu->util_rt;
- if ((util + sg_cpu->util_dl) >= sg_cpu->max)
- return sg_cpu->max;
+ /*
+ * Interrupt time is not seen by RQS utilization so we can compare
+ * them with the CPU capacity
+ */
+ if ((util + sg_cpu->util_dl) >= max)
+ return max;
/*
* As there is still idle time on the CPU, we need to compute the
@@ -220,10 +230,17 @@ static unsigned long sugov_aggregate_util(struct sugov_cpu *sg_cpu)
* ready for such an interface. So, we only do the latter for now.
*/
+ /* Weight RQS utilization to normal context window */
+ util *= (max - sg_cpu->util_irq);
+ util /= max;
+
+ /* Add interrupt utilization */
+ util += sg_cpu->util_irq;
+
/* Add DL bandwidth requirement */
util += sg_cpu->bw_dl;
- return min(sg_cpu->max, util);
+ return min(max, util);
}
/**
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index b26d0c9948dd..b2833e2b4b6a 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -2228,4 +2228,17 @@ static inline unsigned long cpu_util_rt(struct rq *rq)
{
return rq->avg_rt.util_avg;
}
+
+#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING)
+static inline unsigned long cpu_util_irq(struct rq *rq)
+{
+ return rq->avg_irq.util_avg;
+}
+#else
+static inline unsigned long cpu_util_irq(struct rq *rq)
+{
+ return 0;
+}
+
+#endif
#endif
next prev parent reply other threads:[~2018-07-15 23:29 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-28 15:45 [PATCH v7 00/11] track CPU utilization Vincent Guittot
2018-06-28 15:45 ` [PATCH 01/11] sched/pelt: Move pelt related code in a dedicated file Vincent Guittot
2018-07-15 23:26 ` [tip:sched/core] sched/pelt: Move PELT " tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 02/11] sched/rt: add rt_rq utilization tracking Vincent Guittot
2018-07-15 23:27 ` [tip:sched/core] sched/rt: Add " tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 03/11] cpufreq/schedutil: use rt " Vincent Guittot
2018-07-06 5:56 ` Viresh Kumar
2018-07-15 23:27 ` [tip:sched/core] cpufreq/schedutil: Use RT " tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 04/11] sched/dl: add dl_rq " Vincent Guittot
2018-07-15 23:28 ` [tip:sched/core] sched/dl: Add " tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 05/11] cpufreq/schedutil: use dl " Vincent Guittot
2018-07-06 5:59 ` Viresh Kumar
2018-07-15 23:28 ` [tip:sched/core] cpufreq/schedutil: Use DL " tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 06/11] sched/irq: add irq " Vincent Guittot
2018-07-15 23:29 ` [tip:sched/core] sched/irq: Add IRQ " tip-bot for Vincent Guittot
2018-07-26 3:09 ` [PATCH 06/11] sched/irq: add irq " Wanpeng Li
2018-07-30 16:43 ` Vincent Guittot
2018-07-31 3:32 ` Wanpeng Li
2018-07-31 8:21 ` Vincent Guittot
2018-06-28 15:45 ` [PATCH 07/11] cpufreq/schedutil: take into account interrupt Vincent Guittot
2018-07-06 6:00 ` Viresh Kumar
2018-07-06 9:14 ` Peter Zijlstra
2018-07-06 9:21 ` Vincent Guittot
2018-07-15 23:29 ` tip-bot for Vincent Guittot [this message]
2018-06-28 15:45 ` [PATCH 08/11] sched: schedutil: remove sugov_aggregate_util() Vincent Guittot
2018-07-06 6:02 ` Viresh Kumar
2018-07-15 23:30 ` [tip:sched/core] sched/cpufreq: Remove sugov_aggregate_util() tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 09/11] sched: use pelt for scale_rt_capacity() Vincent Guittot
2018-07-15 22:15 ` Ingo Molnar
2018-07-15 22:46 ` Joe Perches
2018-07-16 11:24 ` Vincent Guittot
2018-07-16 11:39 ` Ingo Molnar
2018-07-15 23:32 ` [tip:sched/core] sched/core: Use PELT " tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 10/11] sched: remove rt_avg code Vincent Guittot
2018-07-15 23:33 ` [tip:sched/core] sched/core: Remove the " tip-bot for Vincent Guittot
2018-06-28 15:45 ` [PATCH 11/11] proc/sched: remove unused sched_time_avg_ms Vincent Guittot
2018-06-28 15:51 ` Luis R. Rodriguez
2018-06-29 5:49 ` Vincent Guittot
2018-07-15 23:33 ` [tip:sched/core] sched/sysctl: Remove unused sched_time_avg_ms sysctl tip-bot for Vincent Guittot
2018-07-05 12:36 ` [PATCH v7 00/11] track CPU utilization Peter Zijlstra
2018-07-05 13:32 ` Vincent Guittot
2018-07-06 6:05 ` Viresh Kumar
2018-07-06 9:18 ` Peter Zijlstra
2018-07-15 23:34 ` [tip:sched/core] sched/cpufreq: Clarify sugov_get_util() tip-bot for Peter Zijlstra
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=tip-9033ea11889f88f243445495f72441e22256d5e9@git.kernel.org \
--to=tipbot@zytor.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--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).