From: "Rafael J. Wysocki" <rjw@rjwysocki.net>
To: Linux PM <linux-pm@vger.kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Thomas Gleixner <tglx@linutronix.de>,
Paul McKenney <paulmck@linux.vnet.ibm.com>,
Thomas Ilsche <thomas.ilsche@tu-dresden.de>,
Doug Smythies <dsmythies@telus.net>,
Rik van Riel <riel@surriel.com>,
Aubrey Li <aubrey.li@linux.intel.com>,
Mike Galbraith <mgalbraith@suse.de>,
LKML <linux-kernel@vger.kernel.org>,
Len Brown <len.brown@intel.com>
Subject: [PATCH v9 10/10] cpuidle: menu: Avoid selecting shallow states with stopped tick
Date: Wed, 04 Apr 2018 10:50:36 +0200 [thread overview]
Message-ID: <6542020.eHGLEK9V0J@aspire.rjw.lan> (raw)
In-Reply-To: <1736751.LdhZHb50jq@aspire.rjw.lan>
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
If the scheduler tick has been stopped already and the governor
selects a shallow idle state, the CPU can spend a long time in that
state if the selection is based on an inaccurate prediction of idle
time. That effect turns out to be relevant, so it needs to be
mitigated.
To that end, modify the menu governor to discard the result of the
idle time prediction if the tick is stopped and the predicted idle
time is less than the tick period length, unless the tick timer is
going to expire soon.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
v8 -> v9: No changes.
---
drivers/cpuidle/governors/menu.c | 29 ++++++++++++++++++++++-------
1 file changed, 22 insertions(+), 7 deletions(-)
Index: linux-pm/drivers/cpuidle/governors/menu.c
===================================================================
--- linux-pm.orig/drivers/cpuidle/governors/menu.c
+++ linux-pm/drivers/cpuidle/governors/menu.c
@@ -352,13 +352,28 @@ static int menu_select(struct cpuidle_dr
*/
data->predicted_us = min(data->predicted_us, expected_interval);
- /*
- * Use the performance multiplier and the user-configurable
- * latency_req to determine the maximum exit latency.
- */
- interactivity_req = data->predicted_us / performance_multiplier(nr_iowaiters, cpu_load);
- if (latency_req > interactivity_req)
- latency_req = interactivity_req;
+ if (tick_nohz_tick_stopped()) {
+ /*
+ * If the tick is already stopped, the cost of possible short
+ * idle duration misprediction is much higher, because the CPU
+ * may be stuck in a shallow idle state for a long time as a
+ * result of it. In that case say we might mispredict and try
+ * to force the CPU into a state for which we would have stopped
+ * the tick, unless the tick timer is going to expire really
+ * soon anyway.
+ */
+ if (data->predicted_us < TICK_USEC)
+ data->predicted_us = min_t(unsigned int, TICK_USEC,
+ ktime_to_us(delta_next));
+ } else {
+ /*
+ * Use the performance multiplier and the user-configurable
+ * latency_req to determine the maximum exit latency.
+ */
+ interactivity_req = data->predicted_us / performance_multiplier(nr_iowaiters, cpu_load);
+ if (latency_req > interactivity_req)
+ latency_req = interactivity_req;
+ }
expected_interval = data->predicted_us;
/*
next prev parent reply other threads:[~2018-04-04 8:53 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-04 8:32 [PATCH v9 00/10] sched/cpuidle: Idle loop rework Rafael J. Wysocki
2018-04-04 8:33 ` [PATCH v9 01/10] time: tick-sched: Reorganize idle tick management code Rafael J. Wysocki
2018-04-04 8:34 ` [PATCH v9 02/10] sched: idle: Do not stop the tick upfront in the idle loop Rafael J. Wysocki
2018-04-04 8:36 ` [PATCH v9 03/10] sched: idle: Do not stop the tick before cpuidle_idle_call() Rafael J. Wysocki
2018-04-04 8:38 ` [PATCH v9 04/10] jiffies: Introduce USER_TICK_USEC and redefine TICK_USEC Rafael J. Wysocki
2018-04-06 1:09 ` Frederic Weisbecker
2018-04-06 7:20 ` Rafael J. Wysocki
2018-04-04 8:39 ` [PATCH v9 05/10] cpuidle: Return nohz hint from cpuidle_select() Rafael J. Wysocki
2018-04-06 2:44 ` Frederic Weisbecker
2018-04-06 7:24 ` Rafael J. Wysocki
2018-04-06 14:19 ` Frederic Weisbecker
2018-04-06 7:58 ` Peter Zijlstra
2018-04-06 14:23 ` Frederic Weisbecker
2018-04-06 8:11 ` Rafael J. Wysocki
2018-04-06 12:56 ` Rafael J. Wysocki
2018-04-06 15:28 ` Frederic Weisbecker
2018-04-06 14:28 ` Frederic Weisbecker
2018-04-04 8:41 ` [PATCH v9 06/10] time: tick-sched: Split tick_nohz_stop_sched_tick() Rafael J. Wysocki
2018-04-07 2:36 ` Frederic Weisbecker
2018-04-07 16:36 ` Rafael J. Wysocki
2018-04-04 8:45 ` [PATCH v9 07/10] time: hrtimer: Introduce hrtimer_next_event_without() Rafael J. Wysocki
2018-04-07 14:46 ` Frederic Weisbecker
2018-04-08 8:20 ` Rafael J. Wysocki
2018-04-08 17:58 ` Frederic Weisbecker
2018-04-04 8:47 ` [PATCH v9 08/10] sched: idle: Select idle state before stopping the tick Rafael J. Wysocki
2018-04-09 2:41 ` Frederic Weisbecker
2018-04-04 8:49 ` [PATCH v9 09/10] cpuidle: menu: Refine idle state selection for running tick Rafael J. Wysocki
2018-04-05 12:27 ` Peter Zijlstra
2018-04-05 13:51 ` Rafael J. Wysocki
2018-04-05 12:32 ` Peter Zijlstra
2018-04-05 13:52 ` Rafael J. Wysocki
2018-04-04 8:50 ` Rafael J. Wysocki [this message]
2018-04-05 12:47 ` [PATCH v9 10/10] cpuidle: menu: Avoid selecting shallow states with stopped tick Peter Zijlstra
2018-04-05 13:49 ` Rafael J. Wysocki
2018-04-05 14:11 ` Peter Zijlstra
2018-04-05 14:13 ` Peter Zijlstra
2018-04-05 14:29 ` Rafael J. Wysocki
2018-04-08 16:32 ` [PATCH v9 00/10] sched/cpuidle: Idle loop rework Rafael J. Wysocki
2018-04-09 15:58 ` Thomas Ilsche
2018-04-10 7:03 ` Rafael J. Wysocki
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=6542020.eHGLEK9V0J@aspire.rjw.lan \
--to=rjw@rjwysocki.net \
--cc=aubrey.li@linux.intel.com \
--cc=dsmythies@telus.net \
--cc=fweisbec@gmail.com \
--cc=len.brown@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mgalbraith@suse.de \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=riel@surriel.com \
--cc=tglx@linutronix.de \
--cc=thomas.ilsche@tu-dresden.de \
/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
all inboxes | Powered by JetHome®