[PATCH 2/2] tick: SHUTDOWN event-dev if no events are required for KTIME_MAX

[Viresh Kumar]

When expires is set to KTIME_MAX in tick_program_event(), we are sure that there
are no events enqueued for a very long time and so there is no point keeping
event device running. We will get interrupted without any work to do many a
times, for example when timer's counter overflows.

So, its better to SHUTDOWN the event device then and restart it ones we get a
request for next event. For implementing this a new field 'last_mode' is added
to 'struct clock_event_device' to keep track of last mode used.

Signed-off-by: Viresh Kumar <viresh.kumar@xxxxxxxxxx>
---
 include/linux/clockchips.h |  1 +
 kernel/time/tick-oneshot.c | 14 +++++++++++++-
 2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index 2e4cb67..36a4ca6 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -105,6 +105,7 @@ struct clock_event_device {
 	u32			mult;
 	u32			shift;
 	enum clock_event_mode	mode;
+	enum clock_event_mode	last_mode;
 	unsigned int		features;
 	unsigned long		retries;
 
diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c
index 8241090..9543815 100644
--- a/kernel/time/tick-oneshot.c
+++ b/kernel/time/tick-oneshot.c
@@ -27,8 +27,20 @@
 int tick_program_event(ktime_t expires, int force)
 {
 	struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev);
+	int ret = 0;
 
-	return clockevents_program_event(dev, expires, force);
+	/* Shut down event device if it is not required for long */
+	if (unlikely(expires.tv64 == KTIME_MAX)) {
+		dev->last_mode = dev->mode;
+		clockevents_set_mode(dev, CLOCK_EVT_MODE_SHUTDOWN);
+	} else {
+		/* restore mode when restarting event dev */
+		if (unlikely(dev->mode == CLOCK_EVT_MODE_SHUTDOWN))
+			clockevents_set_mode(dev, dev->last_mode);
+		ret = clockevents_program_event(dev, expires, force);
+	}
+
+	return ret;
 }
 
 /**
-- 
2.0.0.rc2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/