[PATCH] sch_sfq: avoid giving spurious NET_XMIT_CN signals

[PATCH] sch_sfq: avoid giving spurious NET_XMIT_CN signals

[Eric Dumazet]

While chasing a possible net_sched bug, I found that IP fragments have
litle chance to pass a congestioned SFQ qdisc :

- Say SFQ qdisc is full because one flow is non responsive.
- ip_fragment() wants to send two fragments belonging to an idle flow.
- sfq_enqueue() queues first packet, but see queue limit reached :
- sfq_enqueue() drops one packet from 'big consumer', and returns
NET_XMIT_CN.
- ip_fragment() cancel remaining fragments.

This patch restores fairness, making sure we return NET_XMIT_CN only if
we dropped a packet from the same flow.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Patrick McHardy <kaber@trash.net>
CC: Jarek Poplawski <jarkao2@gmail.com>
CC: Jamal Hadi Salim <hadi@cyberus.ca>
CC: Stephen Hemminger <shemminger@vyatta.com>
---
 net/sched/sch_sfq.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 7ef87f9..b1d00f8 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -361,7 +361,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
 	struct sfq_sched_data *q = qdisc_priv(sch);
 	unsigned int hash;
-	sfq_index x;
+	sfq_index x, qlen;
 	struct sfq_slot *slot;
 	int uninitialized_var(ret);
 
@@ -405,8 +405,12 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	if (++sch->q.qlen <= q->limit)
 		return NET_XMIT_SUCCESS;
 
+	qlen = slot->qlen;
 	sfq_drop(sch);
-	return NET_XMIT_CN;
+	/* Return Congestion Notification only if we dropped a packet
+	 * from this flow.
+	 */
+	return (qlen != slot->qlen) ? NET_XMIT_CN : NET_XMIT_SUCCESS;
 }
 
 static struct sk_buff *

Re: [PATCH] sch_sfq: avoid giving spurious NET_XMIT_CN signals

[David Miller]

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 23 May 2011 23:02:42 +0200

> While chasing a possible net_sched bug, I found that IP fragments have
> litle chance to pass a congestioned SFQ qdisc :
> 
> - Say SFQ qdisc is full because one flow is non responsive.
> - ip_fragment() wants to send two fragments belonging to an idle flow.
> - sfq_enqueue() queues first packet, but see queue limit reached :
> - sfq_enqueue() drops one packet from 'big consumer', and returns
> NET_XMIT_CN.
> - ip_fragment() cancel remaining fragments.
> 
> This patch restores fairness, making sure we return NET_XMIT_CN only if
> we dropped a packet from the same flow.
> 
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>

Looks good, applied, thanks Eric.