linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: linux-arm-kernel@lists.infradead.org
Cc: linux-kernel@vger.kernel.org, Will Deacon <will@kernel.org>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Marc Zyngier <maz@kernel.org>,
	Lorenzo Colitti <lorenzo@google.com>,
	John Stultz <john.stultz@linaro.org>,
	Stephen Boyd <sboyd@kernel.org>,
	kernel-team@android.com
Subject: [PATCH 4/5] tick/broadcast: Program wakeup timer when entering idle if required
Date: Thu, 20 May 2021 19:47:04 +0100	[thread overview]
Message-ID: <20210520184705.10845-5-will@kernel.org> (raw)
In-Reply-To: <20210520184705.10845-1-will@kernel.org>

When configuring the broadcast timer on entry to and exit from deep idle
states, prefer a per-CPU wakeup timer if one exists.

On entry to idle, stop the tick device and transfer the next event into
the oneshot wakeup device, which will serve as the wakeup from idle. To
avoid the overhead of additional hardware accesses on exit from idle,
leave the timer armed and treat the inevitable interrupt as a (possibly
spurious) tick event.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Lorenzo Colitti <lorenzo@google.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Stephen Boyd <sboyd@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
---
 kernel/time/tick-broadcast.c | 38 +++++++++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 8bd8cd69c8c9..ba5264e210d9 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -96,6 +96,15 @@ static struct clock_event_device *tick_get_oneshot_wakeup_device(int cpu)
 	return per_cpu(tick_oneshot_wakeup_device, cpu);
 }
 
+static void tick_oneshot_wakeup_handler(struct clock_event_device *wd)
+{
+	/*
+	 * If we woke up early and the tick was reprogrammed in the
+	 * meantime then this may be spurious but harmless.
+	 */
+	tick_receive_broadcast();
+}
+
 static bool tick_set_oneshot_wakeup_device(struct clock_event_device *newdev,
 					   int cpu)
 {
@@ -116,6 +125,7 @@ static bool tick_set_oneshot_wakeup_device(struct clock_event_device *newdev,
 	if (curdev && newdev->rating <= curdev->rating)
 		return false;
 
+	newdev->event_handler = tick_oneshot_wakeup_handler;
 set_device:
 	per_cpu(tick_oneshot_wakeup_device, cpu) = newdev;
 	return true;
@@ -903,16 +913,42 @@ static int ___tick_broadcast_oneshot_control(enum tick_broadcast_state state,
 	return ret;
 }
 
+static int tick_oneshot_wakeup_control(enum tick_broadcast_state state,
+				       struct tick_device *td,
+				       int cpu)
+{
+	struct clock_event_device *dev, *wd;
+
+	dev = td->evtdev;
+	if (td->mode != TICKDEV_MODE_ONESHOT)
+		return -EINVAL;
+
+	wd = tick_get_oneshot_wakeup_device(cpu);
+	if (!wd)
+		return -ENODEV;
+
+	if (state == TICK_BROADCAST_ENTER) {
+		clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT_STOPPED);
+		clockevents_switch_state(wd, CLOCK_EVT_STATE_ONESHOT);
+		clockevents_program_event(wd, dev->next_event, 1);
+	}
+
+	return 0;
+}
+
 int __tick_broadcast_oneshot_control(enum tick_broadcast_state state)
 {
 	struct tick_device *td = this_cpu_ptr(&tick_cpu_device);
 	int cpu = smp_processor_id();
 
+	if (!tick_oneshot_wakeup_control(state, td, cpu))
+		return 0;
+
 	if (tick_broadcast_device.evtdev)
 		return ___tick_broadcast_oneshot_control(state, td, cpu);
 
 	/*
-	 * If there is no broadcast device, tell the caller not
+	 * If there is no broadcast or wakeup device, tell the caller not
 	 * to go into deep idle.
 	 */
 	return -EBUSY;
-- 
2.31.1.818.g46aad6cb9e-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  parent reply	other threads:[~2021-05-20 18:50 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-20 18:47 [PATCH 0/5] tick/broadcast: Allow per-cpu timers to be used instead of broadcast Will Deacon
2021-05-20 18:47 ` [PATCH 1/5] tick/broadcast: Drop unneeded CONFIG_GENERIC_CLOCKEVENTS_BROADCAST guard Will Deacon
2021-05-20 18:47 ` [PATCH 2/5] tick/broadcast: Split __tick_broadcast_oneshot_control() into a helper Will Deacon
2021-05-20 18:47 ` [PATCH 3/5] tick/broadcast: Prefer per-cpu oneshot wakeup timers to broadcast Will Deacon
2021-05-21  2:25   ` Mika Penttilä
2021-05-21 11:25     ` Will Deacon
2021-05-21 12:18       ` Will Deacon
2021-05-21 13:49       ` Thomas Gleixner
2021-05-21 14:43         ` Will Deacon
2021-05-20 18:47 ` Will Deacon [this message]
2021-05-20 18:47 ` [PATCH 5/5] timer_list: Print name of per-cpu wakeup device Will Deacon

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=20210520184705.10845-5-will@kernel.org \
    --to=will@kernel.org \
    --cc=fweisbec@gmail.com \
    --cc=john.stultz@linaro.org \
    --cc=kernel-team@android.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lorenzo@google.com \
    --cc=maz@kernel.org \
    --cc=sboyd@kernel.org \
    --cc=tglx@linutronix.de \
    --subject='Re: [PATCH 4/5] tick/broadcast: Program wakeup timer when entering idle if required' \
    /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

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).