net_sched: remove generic throttled management

__QDISC_STATE_THROTTLED bit manipulation is rather expensive
for HTB and few others.

I already removed it for sch_fq in commit f2600cf0
("net: sched: avoid costly atomic operation in fq_dequeue()")
and so far nobody complained.

When one ore more packets are stuck in one or more throttled
HTB class, a htb dequeue() performs two atomic operations
to clear/set __QDISC_STATE_THROTTLED bit, while root qdisc
lock is held.

Removing this pair of atomic operations bring me a 8 % performance
increase on 200 TCP_RR tests, in presence of throttled classes.

This patch has no side effect, since nothing actually uses
disc_is_throttled() anymore.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/pkt_sched.h

@@ -67,12 +67,12 @@ struct qdisc_watchdog {
 };
 
 void qdisc_watchdog_init(struct qdisc_watchdog *wd, struct Qdisc *qdisc);
-void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires, bool throttle);
+void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires);
 
 static inline void qdisc_watchdog_schedule(struct qdisc_watchdog *wd,
 					   psched_time_t expires)
 {
-	qdisc_watchdog_schedule_ns(wd, PSCHED_TICKS2NS(expires), true);
+	qdisc_watchdog_schedule_ns(wd, PSCHED_TICKS2NS(expires));
 }
 
 void qdisc_watchdog_cancel(struct qdisc_watchdog *wd);

include/net/sch_generic.h

@@ -26,7 +26,6 @@ struct qdisc_rate_table {
 enum qdisc_state_t {
 	__QDISC_STATE_SCHED,
 	__QDISC_STATE_DEACTIVATED,
-	__QDISC_STATE_THROTTLED,
 };
 
 struct qdisc_size_table {
@@ -125,21 +124,6 @@ static inline int qdisc_avail_bulklimit(const struct netdev_queue *txq)
 #endif
 }
 
-static inline bool qdisc_is_throttled(const struct Qdisc *qdisc)
-{
-	return test_bit(__QDISC_STATE_THROTTLED, &qdisc->state) ? true : false;
-}
-
-static inline void qdisc_throttled(struct Qdisc *qdisc)
-{
-	set_bit(__QDISC_STATE_THROTTLED, &qdisc->state);
-}
-
-static inline void qdisc_unthrottled(struct Qdisc *qdisc)
-{
-	clear_bit(__QDISC_STATE_THROTTLED, &qdisc->state);
-}
-
 struct Qdisc_class_ops {
 	/* Child qdisc manipulation */
 	struct netdev_queue *	(*select_queue)(struct Qdisc *, struct tcmsg *);

net/sched/sch_api.c

@@ -583,7 +583,6 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 						 timer);
 
 	rcu_read_lock();
-	qdisc_unthrottled(wd->qdisc);
 	__netif_schedule(qdisc_root(wd->qdisc));
 	rcu_read_unlock();
 
@@ -598,15 +597,12 @@ void qdisc_watchdog_init(struct qdisc_watchdog *wd, struct Qdisc *qdisc)
 }
 EXPORT_SYMBOL(qdisc_watchdog_init);
 
-void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires, bool throttle)
+void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires)
 {
 	if (test_bit(__QDISC_STATE_DEACTIVATED,
 		     &qdisc_root_sleeping(wd->qdisc)->state))
 		return;
 
-	if (throttle)
-		qdisc_throttled(wd->qdisc);
-
 	if (wd->last_expires == expires)
 		return;
 
@@ -620,7 +616,6 @@ EXPORT_SYMBOL(qdisc_watchdog_schedule_ns);
 void qdisc_watchdog_cancel(struct qdisc_watchdog *wd)
 {
 	hrtimer_cancel(&wd->timer);
-	qdisc_unthrottled(wd->qdisc);
 }
 EXPORT_SYMBOL(qdisc_watchdog_cancel);
 

net/sched/sch_cbq.c

@@ -513,7 +513,6 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer)
 		hrtimer_start(&q->delay_timer, time, HRTIMER_MODE_ABS_PINNED);
 	}
 
-	qdisc_unthrottled(sch);
 	__netif_schedule(qdisc_root(sch));
 	return HRTIMER_NORESTART;
 }
@@ -819,7 +818,6 @@ cbq_dequeue(struct Qdisc *sch)
 		if (skb) {
 			qdisc_bstats_update(sch, skb);
 			sch->q.qlen--;
-			qdisc_unthrottled(sch);
 			return skb;
 		}
 

net/sched/sch_fq.c

@@ -445,8 +445,7 @@ static struct sk_buff *fq_dequeue(struct Qdisc *sch)
 		if (!head->first) {
 			if (q->time_next_delayed_flow != ~0ULL)
 				qdisc_watchdog_schedule_ns(&q->watchdog,
-							   q->time_next_delayed_flow,
-							   false);
+							   q->time_next_delayed_flow);
 			return NULL;
 		}
 	}

net/sched/sch_hfsc.c

@@ -1664,7 +1664,6 @@ hfsc_dequeue(struct Qdisc *sch)
 		set_passive(cl);
 	}
 
-	qdisc_unthrottled(sch);
 	qdisc_bstats_update(sch, skb);
 	qdisc_qstats_backlog_dec(sch, skb);
 	sch->q.qlen--;

net/sched/sch_htb.c

@@ -889,7 +889,6 @@ static struct sk_buff *htb_dequeue(struct Qdisc *sch)
 	if (skb != NULL) {
 ok:
 		qdisc_bstats_update(sch, skb);
-		qdisc_unthrottled(sch);
 		qdisc_qstats_backlog_dec(sch, skb);
 		sch->q.qlen--;
 		return skb;
@@ -929,7 +928,7 @@ static struct sk_buff *htb_dequeue(struct Qdisc *sch)
 	}
 	qdisc_qstats_overlimit(sch);
 	if (likely(next_event > q->now))
-		qdisc_watchdog_schedule_ns(&q->watchdog, next_event, true);
+		qdisc_watchdog_schedule_ns(&q->watchdog, next_event);
 	else
 		schedule_work(&q->work);
 fin:

net/sched/sch_netem.c

@@ -587,7 +587,6 @@ static struct sk_buff *netem_dequeue(struct Qdisc *sch)
 	if (skb) {
 		qdisc_qstats_backlog_dec(sch, skb);
 deliver:
-		qdisc_unthrottled(sch);
 		qdisc_bstats_update(sch, skb);
 		return skb;
 	}

net/sched/sch_tbf.c

@@ -254,14 +254,12 @@ static struct sk_buff *tbf_dequeue(struct Qdisc *sch)
 			q->ptokens = ptoks;
 			qdisc_qstats_backlog_dec(sch, skb);
 			sch->q.qlen--;
-			qdisc_unthrottled(sch);
 			qdisc_bstats_update(sch, skb);
 			return skb;
 		}
 
 		qdisc_watchdog_schedule_ns(&q->watchdog,
-					   now + max_t(long, -toks, -ptoks),
-					   true);
+					   now + max_t(long, -toks, -ptoks));
 
 		/* Maybe we have a shorter packet in the queue,
 		   which can be sent now. It sounds cool,