When nohz or nohz_full is configured, the concurrency calls of
tick_do_update_jiffies64 increases, and the conflict between
jiffies_lock and jiffies_seq increases, especially in multi-core
scenarios.
However, it is unnecessary to update the jiffies_seq lock multiple
times in a tick period, so the critical region of the jiffies_seq
can be reduced to reduce latency overheads. By the way,
last_jiffies_update is protected by jiffies_lock, so reducing the
jiffies_seq critical area is safe.
Signed-off-by: Yunfeng Ye <yeyunf...@huawei.com>
---
kernel/time/tick-sched.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index f0199a4ba1ad..41fb1400439b 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -66,11 +66,11 @@ static void tick_do_update_jiffies64(ktime_t now)
/* Reevaluate with jiffies_lock held */
raw_spin_lock(&jiffies_lock);
- write_seqcount_begin(&jiffies_seq);
delta = ktime_sub(now, last_jiffies_update);
if (delta >= tick_period) {
+ write_seqcount_begin(&jiffies_seq);
delta = ktime_sub(delta, tick_period);
/* Pairs with the lockless read in this function. */
WRITE_ONCE(last_jiffies_update,
@@ -91,12 +91,11 @@ static void tick_do_update_jiffies64(ktime_t now)
/* Keep the tick_next_period variable up to date */
tick_next_period = ktime_add(last_jiffies_update, tick_period);
- } else {
write_seqcount_end(&jiffies_seq);
+ } else {
raw_spin_unlock(&jiffies_lock);
return;
}
- write_seqcount_end(&jiffies_seq);
raw_spin_unlock(&jiffies_lock);
update_wall_time();
}
--
2.18.4
