From: Eric Dumazet <edumazet@google.com>
Date: Mon, 16 Mar 2020 19:12:50 -0700
Subject: net_sched: do not reprogram a timer about to expire
Patch-mainline: v5.7-rc1
Git-commit: b88948fbc7cec03bd97fe5be65264681e98041e8
References: bsc#1176447

qdisc_watchdog_schedule_range_ns() can use the newly added slack
and avoid rearming the hrtimer a bit earlier than the current
value. This patch has no effect if delta_ns parameter
is zero.

Note that this means the max slack is potentially doubled.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
---
 net/sched/sch_api.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -625,8 +625,13 @@ void qdisc_watchdog_schedule_range_ns(st
 		     &qdisc_root_sleeping(wd->qdisc)->state))
 		return;
 
-	if (wd->last_expires == expires)
-		return;
+	if (hrtimer_is_queued(&wd->timer)) {
+		/* If timer is already set in [expires, expires + delta_ns],
+		 * do not reprogram it.
+		 */
+		if (wd->last_expires - expires <= delta_ns)
+			return;
+	}
 
 	wd->last_expires = expires;
 	hrtimer_start_range_ns(&wd->timer,