etf: Drop all expired packets

Currently on dequeue() ETF only drops the first expired packet, which
causes a problem if the next packet is already expired. When this
happens, the watchdog will be configured with a time in the past, fire
straight way and the packet will finally be dropped once the dequeue()
function of the qdisc is called again.

We can save quite a few cycles and improve the overall behavior of the
qdisc if we drop all expired packets if the next packet is expired.
This should allow ETF to recover faster from bad situations. But
packet drops are still a very serious warning that the requirements
imposed on the system aren't reasonable.

This was inspired by how the implementation of hrtimers use the
rb_tree inside the kernel.

Signed-off-by: Jesus Sanchez-Palencia <jesus.s.palencia@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/sch_etf.c b/net/sched/sch_etf.c
index bfe04748d5f04112db016a22258b8af02863f439..1150f22983dfe6e5e3b8aaf6619b22702341b6cd 100644 (file)
--- a/net/sched/sch_etf.c
+++ b/net/sched/sch_etf.c
@@ -190,29 +190,35 @@ static int etf_enqueue_timesortedlist(struct sk_buff *nskb, struct Qdisc *sch,
        return NET_XMIT_SUCCESS;
 }
 
-static void timesortedlist_drop(struct Qdisc *sch, struct sk_buff *skb)
+static void timesortedlist_drop(struct Qdisc *sch, struct sk_buff *skb,
+                               ktime_t now)
 {
        struct etf_sched_data *q = qdisc_priv(sch);
        struct sk_buff *to_free = NULL;
+       struct sk_buff *tmp = NULL;
 
-       rb_erase_cached(&skb->rbnode, &q->head);
+       skb_rbtree_walk_from_safe(skb, tmp) {
+               if (ktime_after(skb->tstamp, now))
+                       break;
 
-       /* The rbnode field in the skb re-uses these fields, now that
-        * we are done with the rbnode, reset them.
-        */
-       skb->next = NULL;
-       skb->prev = NULL;
-       skb->dev = qdisc_dev(sch);
+               rb_erase_cached(&skb->rbnode, &q->head);
 
-       qdisc_qstats_backlog_dec(sch, skb);
+               /* The rbnode field in the skb re-uses these fields, now that
+                * we are done with the rbnode, reset them.
+                */
+               skb->next = NULL;
+               skb->prev = NULL;
+               skb->dev = qdisc_dev(sch);
 
-       report_sock_error(skb, ECANCELED, SO_EE_CODE_TXTIME_MISSED);
+               report_sock_error(skb, ECANCELED, SO_EE_CODE_TXTIME_MISSED);
 
-       qdisc_drop(skb, sch, &to_free);
-       kfree_skb_list(to_free);
-       qdisc_qstats_overlimit(sch);
+               qdisc_qstats_backlog_dec(sch, skb);
+               qdisc_drop(skb, sch, &to_free);
+               qdisc_qstats_overlimit(sch);
+               sch->q.qlen--;
+       }
 
-       sch->q.qlen--;
+       kfree_skb_list(to_free);
 }
 
 static void timesortedlist_remove(struct Qdisc *sch, struct sk_buff *skb)
@@ -251,7 +257,7 @@ static struct sk_buff *etf_dequeue_timesortedlist(struct Qdisc *sch)
 
        /* Drop if packet has expired while in queue. */
        if (ktime_before(skb->tstamp, now)) {
-               timesortedlist_drop(sch, skb);
+               timesortedlist_drop(sch, skb, now);
                skb = NULL;
                goto out;
        }