From 5a308f40bfe27fcfd1db3970afe18b635f23c182 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 14 Jul 2012 03:16:27 +0000 Subject: [PATCH] netem: refine early skb orphaning netem does an early orphaning of skbs. Doing so breaks TCP Small Queue or any mechanism relying on socket sk_wmem_alloc feedback. Ideally, we should perform this orphaning after the rate module and before the delay module, to mimic what happens on a real link : skb orphaning is indeed normally done at TX completion, before the transit on the link. +-------+ +--------+ +---------------+ +-----------------+ + Qdisc +---> Device +--> TX completion +--> links / hops +-> + + + xmit + + skb orphaning + + propagation + +-------+ +--------+ +---------------+ +-----------------+ < rate limiting > < delay, drops, reorders > If netem is used without delay feature (drops, reorders, rate limiting), then we should avoid early skb orphaning, to keep pressure on sockets as long as packets are still in qdisc queue. Ideally, netem should be refactored to implement delay module as the last stage. Current algorithm merges the two phases (rate limiting + delay) so its not correct. Signed-off-by: Eric Dumazet Cc: Hagen Paul Pfeifer Cc: Mark Gordon Cc: Andreas Terzis Cc: Yuchung Cheng Acked-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/sched/sch_netem.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index c412ad0d0308..298c0ddfb57e 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -380,7 +380,14 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; } - skb_orphan(skb); + /* If a delay is expected, orphan the skb. (orphaning usually takes + * place at TX completion time, so _before_ the link transit delay) + * Ideally, this orphaning should be done after the rate limiting + * module, because this breaks TCP Small Queue, and other mechanisms + * based on socket sk_wmem_alloc. + */ + if (q->latency || q->jitter) + skb_orphan(skb); /* * If we need to duplicate packet, then re-insert at top of the -- 2.30.2