diff options
| author |   Eric Dumazet <edumazet@google.com> | 2016-06-21 23:16:49 -0700 |
|---|---|---|
| committer |   David S. Miller <davem@davemloft.net> | 2016-06-25 12:19:35 -0400 |
| commit | 520ac30f45519b0a82dd92117c181d1d6144677b (patch) | |
| tree | a3189ae1ab3b5be6213716e42ddaec082578e774 /net/sched/sch_atm.c | |
| parent | Merge branch 'liquidio-next' (diff) | |
| download | linux-dev-520ac30f45519b0a82dd92117c181d1d6144677b.tar.xz linux-dev-520ac30f45519b0a82dd92117c181d1d6144677b.zip | |
net_sched: drop packets after root qdisc lock is released
Qdisc performance suffers when packets are dropped at enqueue()
time because drops (kfree_skb()) are done while qdisc lock is held,
delaying a dequeue() draining the queue.
Nominal throughput can be reduced by 50 % when this happens,
at a time we would like the dequeue() to proceed as fast as possible.
Even FQ is vulnerable to this problem, while one of FQ goals was
to provide some flow isolation.
This patch adds a 'struct sk_buff **to_free' parameter to all
qdisc->enqueue(), and in qdisc_drop() helper.
I measured a performance increase of up to 12 %, but this patch
is a prereq so that future batches in enqueue() can fly.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to '')
| -rw-r--r-- | net/sched/sch_atm.c | 9 |
1 files changed, 5 insertions, 4 deletions
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index e04ea6994d1c..481e4f12aeb4 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -357,7 +357,8 @@ static struct tcf_proto __rcu **atm_tc_find_tcf(struct Qdisc *sch, /* --------------------------- Qdisc operations ---------------------------- */ -static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct atm_qdisc_data *p = qdisc_priv(sch); struct atm_flow_data *flow; @@ -398,10 +399,10 @@ done: switch (result) { case TC_ACT_QUEUED: case TC_ACT_STOLEN: - kfree_skb(skb); + __qdisc_drop(skb, to_free); return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; case TC_ACT_SHOT: - kfree_skb(skb); + __qdisc_drop(skb, to_free); goto drop; case TC_ACT_RECLASSIFY: if (flow->excess) @@ -413,7 +414,7 @@ done: #endif } - ret = qdisc_enqueue(skb, flow->q); + ret = qdisc_enqueue(skb, flow->q, to_free); if (ret != NET_XMIT_SUCCESS) { drop: __maybe_unused if (net_xmit_drop_count(ret)) { |