diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index c80d1c2..5f63ec5 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -390,7 +390,6 @@
 	ret = qdisc_enqueue(skb, cl->q);
 	if (ret == NET_XMIT_SUCCESS) {
 		sch->q.qlen++;
-		qdisc_bstats_update(sch, skb);
 		cbq_mark_toplevel(q, cl);
 		if (!cl->next_alive)
 			cbq_activate_class(cl);
@@ -649,7 +648,6 @@
 		ret = qdisc_enqueue(skb, cl->q);
 		if (ret == NET_XMIT_SUCCESS) {
 			sch->q.qlen++;
-			qdisc_bstats_update(sch, skb);
 			if (!cl->next_alive)
 				cbq_activate_class(cl);
 			return 0;
@@ -971,6 +969,7 @@
 
 		skb = cbq_dequeue_1(sch);
 		if (skb) {
+			qdisc_bstats_update(sch, skb);
 			sch->q.qlen--;
 			sch->flags &= ~TCQ_F_THROTTLED;
 			return skb;
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index de55e64..6b7fe4a 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -376,7 +376,6 @@
 	}
 
 	bstats_update(&cl->bstats, skb);
-	qdisc_bstats_update(sch, skb);
 
 	sch->q.qlen++;
 	return err;
@@ -403,6 +402,7 @@
 			skb = qdisc_dequeue_peeked(cl->qdisc);
 			if (cl->qdisc->q.qlen == 0)
 				list_del(&cl->alist);
+			qdisc_bstats_update(sch, skb);
 			sch->q.qlen--;
 			return skb;
 		}
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 60f4bdd..0f7bf3f 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -260,7 +260,6 @@
 		return err;
 	}
 
-	qdisc_bstats_update(sch, skb);
 	sch->q.qlen++;
 
 	return NET_XMIT_SUCCESS;
@@ -283,6 +282,7 @@
 	if (skb == NULL)
 		return NULL;
 
+	qdisc_bstats_update(sch, skb);
 	sch->q.qlen--;
 
 	index = skb->tc_index & (p->indices - 1);
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index aa4d633..d468b47 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -46,17 +46,14 @@
 
 static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 {
-	struct sk_buff *skb_head;
 	struct fifo_sched_data *q = qdisc_priv(sch);
 
 	if (likely(skb_queue_len(&sch->q) < q->limit))
 		return qdisc_enqueue_tail(skb, sch);
 
 	/* queue full, remove one skb to fulfill the limit */
-	skb_head = qdisc_dequeue_head(sch);
+	__qdisc_queue_drop_head(sch, &sch->q);
 	sch->qstats.drops++;
-	kfree_skb(skb_head);
-
 	qdisc_enqueue_tail(skb, sch);
 
 	return NET_XMIT_CN;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 2e45791..14a799d 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1600,7 +1600,6 @@
 		set_active(cl, qdisc_pkt_len(skb));
 
 	bstats_update(&cl->bstats, skb);
-	qdisc_bstats_update(sch, skb);
 	sch->q.qlen++;
 
 	return NET_XMIT_SUCCESS;
@@ -1666,6 +1665,7 @@
 	}
 
 	sch->flags &= ~TCQ_F_THROTTLED;
+	qdisc_bstats_update(sch, skb);
 	sch->q.qlen--;
 
 	return skb;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 984c1b0..fc12fe6 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -574,7 +574,6 @@
 	}
 
 	sch->q.qlen++;
-	qdisc_bstats_update(sch, skb);
 	return NET_XMIT_SUCCESS;
 }
 
@@ -842,7 +841,7 @@
 
 static struct sk_buff *htb_dequeue(struct Qdisc *sch)
 {
-	struct sk_buff *skb = NULL;
+	struct sk_buff *skb;
 	struct htb_sched *q = qdisc_priv(sch);
 	int level;
 	psched_time_t next_event;
@@ -851,6 +850,8 @@
 	/* try to dequeue direct packets as high prio (!) to minimize cpu work */
 	skb = __skb_dequeue(&q->direct_queue);
 	if (skb != NULL) {
+ok:
+		qdisc_bstats_update(sch, skb);
 		sch->flags &= ~TCQ_F_THROTTLED;
 		sch->q.qlen--;
 		return skb;
@@ -884,11 +885,8 @@
 			int prio = ffz(m);
 			m |= 1 << prio;
 			skb = htb_dequeue_tree(q, prio, level);
-			if (likely(skb != NULL)) {
-				sch->q.qlen--;
-				sch->flags &= ~TCQ_F_THROTTLED;
-				goto fin;
-			}
+			if (likely(skb != NULL))
+				goto ok;
 		}
 	}
 	sch->qstats.overlimits++;
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 21f13da..436a2e7 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -83,7 +83,6 @@
 
 	ret = qdisc_enqueue(skb, qdisc);
 	if (ret == NET_XMIT_SUCCESS) {
-		qdisc_bstats_update(sch, skb);
 		sch->q.qlen++;
 		return NET_XMIT_SUCCESS;
 	}
@@ -112,6 +111,7 @@
 			qdisc = q->queues[q->curband];
 			skb = qdisc->dequeue(qdisc);
 			if (skb) {
+				qdisc_bstats_update(sch, skb);
 				sch->q.qlen--;
 				return skb;
 			}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 1c4bce8..6a3006b 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -240,7 +240,6 @@
 
 	if (likely(ret == NET_XMIT_SUCCESS)) {
 		sch->q.qlen++;
-		qdisc_bstats_update(sch, skb);
 	} else if (net_xmit_drop_count(ret)) {
 		sch->qstats.drops++;
 	}
@@ -289,6 +288,7 @@
 				skb->tstamp.tv64 = 0;
 #endif
 			pr_debug("netem_dequeue: return skb=%p\n", skb);
+			qdisc_bstats_update(sch, skb);
 			sch->q.qlen--;
 			return skb;
 		}
@@ -476,7 +476,6 @@
 		__skb_queue_after(list, skb, nskb);
 
 		sch->qstats.backlog += qdisc_pkt_len(nskb);
-		qdisc_bstats_update(sch, nskb);
 
 		return NET_XMIT_SUCCESS;
 	}
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 966158d..fbd710d 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -84,7 +84,6 @@
 
 	ret = qdisc_enqueue(skb, qdisc);
 	if (ret == NET_XMIT_SUCCESS) {
-		qdisc_bstats_update(sch, skb);
 		sch->q.qlen++;
 		return NET_XMIT_SUCCESS;
 	}
@@ -116,6 +115,7 @@
 		struct Qdisc *qdisc = q->queues[prio];
 		struct sk_buff *skb = qdisc->dequeue(qdisc);
 		if (skb) {
+			qdisc_bstats_update(sch, skb);
 			sch->q.qlen--;
 			return skb;
 		}
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index a6009c5..9f98dbd 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -94,7 +94,6 @@
 
 	ret = qdisc_enqueue(skb, child);
 	if (likely(ret == NET_XMIT_SUCCESS)) {
-		qdisc_bstats_update(sch, skb);
 		sch->q.qlen++;
 	} else if (net_xmit_drop_count(ret)) {
 		q->stats.pdrop++;
@@ -114,11 +113,13 @@
 	struct Qdisc *child = q->qdisc;
 
 	skb = child->dequeue(child);
-	if (skb)
+	if (skb) {
+		qdisc_bstats_update(sch, skb);
 		sch->q.qlen--;
-	else if (!red_is_idling(&q->parms))
-		red_start_of_idle_period(&q->parms);
-
+	} else {
+		if (!red_is_idling(&q->parms))
+			red_start_of_idle_period(&q->parms);
+	}
 	return skb;
 }
 
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 239ec53..edea8ce 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -402,10 +402,8 @@
 		q->tail = slot;
 		slot->allot = q->scaled_quantum;
 	}
-	if (++sch->q.qlen <= q->limit) {
-		qdisc_bstats_update(sch, skb);
+	if (++sch->q.qlen <= q->limit)
 		return NET_XMIT_SUCCESS;
-	}
 
 	sfq_drop(sch);
 	return NET_XMIT_CN;
@@ -445,6 +443,7 @@
 	}
 	skb = slot_dequeue_head(slot);
 	sfq_dec(q, a);
+	qdisc_bstats_update(sch, skb);
 	sch->q.qlen--;
 	sch->qstats.backlog -= qdisc_pkt_len(skb);
 
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 77565e7..e931658 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -134,7 +134,6 @@
 	}
 
 	sch->q.qlen++;
-	qdisc_bstats_update(sch, skb);
 	return NET_XMIT_SUCCESS;
 }
 
@@ -187,6 +186,7 @@
 			q->ptokens = ptoks;
 			sch->q.qlen--;
 			sch->flags &= ~TCQ_F_THROTTLED;
+			qdisc_bstats_update(sch, skb);
 			return skb;
 		}
 
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 84ce48e..d84e732 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -87,7 +87,6 @@
 
 	if (q->q.qlen < dev->tx_queue_len) {
 		__skb_queue_tail(&q->q, skb);
-		qdisc_bstats_update(sch, skb);
 		return NET_XMIT_SUCCESS;
 	}
 
@@ -111,6 +110,8 @@
 			dat->m->slaves = sch;
 			netif_wake_queue(m);
 		}
+	} else {
+		qdisc_bstats_update(sch, skb);
 	}
 	sch->q.qlen = dat->q.qlen + dat_queue->qdisc->q.qlen;
 	return skb;