qfq: don't leak skb if kzalloc fails
authorFlorian Westphal <fw@strlen.de>
Wed, 8 Jun 2016 21:23:01 +0000 (23:23 +0200)
committerDavid S. Miller <davem@davemloft.net>
Thu, 9 Jun 2016 05:41:38 +0000 (22:41 -0700)
When we need to create a new aggregate to enqueue the skb we call kzalloc.
If that fails we returned ENOBUFS without freeing the skb.

Spotted during code review.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/sched/sch_qfq.c

index 8d2d8d953432e83d8523cc42923793eb852f599c..f18857febdad0adcae567beaebd2520ec9932ce6 100644 (file)
@@ -1235,8 +1235,10 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
                         cl->agg->lmax, qdisc_pkt_len(skb), cl->common.classid);
                err = qfq_change_agg(sch, cl, cl->agg->class_weight,
                                     qdisc_pkt_len(skb));
-               if (err)
-                       return err;
+               if (err) {
+                       cl->qstats.drops++;
+                       return qdisc_drop(skb, sch);
+               }
        }
 
        err = qdisc_enqueue(skb, cl->qdisc);