From 08bfc9cb76e26d2489c401fb1d2de58d06fd1ded Mon Sep 17 00:00:00 2001
From: Erik Hugne <erik.hugne@ericsson.com>
Date: Thu, 22 Jan 2015 17:10:32 +0100
Subject: flow_dissector: add tipc support
The flows are hashed on the sending node address, which allows us
to spread out the TIPC link processing to RPS enabled cores. There
is no point to include the destination address in the hash as that
will always be the same for all inbound links. We have experimented
with a 3-tuple hash over [srcnode, sport, dport], but this showed to
give slightly lower performance because of increased lock contention
when the same link was handled by multiple cores.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
net/core/flow_dissector.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
(limited to 'net/core/flow_dissector.c')
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 45084938c403..beb83d1ac1c6 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -178,6 +178,20 @@ ipv6:
return false;
}
}
+ case htons(ETH_P_TIPC): {
+ struct {
+ __be32 pre[3];
+ __be32 srcnode;
+ } *hdr, _hdr;
+ hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr);
+ if (!hdr)
+ return false;
+ flow->src = hdr->srcnode;
+ flow->dst = 0;
+ flow->n_proto = proto;
+ flow->thoff = (u16)nhoff;
+ return true;
+ }
case htons(ETH_P_FCOE):
flow->thoff = (u16)(nhoff + FCOE_HEADER_LEN);
/* fall through */
--
cgit v1.2.3
From 2bd82484bb4c5db1d5dc983ac7c409b2782e0154 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 3 Feb 2015 23:48:24 -0800
Subject: xps: fix xps for stacked devices
A typical qdisc setup is the following :
bond0 : bonding device, using HTB hierarchy
eth1/eth2 : slaves, multiqueue NIC, using MQ + FQ qdisc
XPS allows to spread packets on specific tx queues, based on the cpu
doing the send.
Problem is that dequeues from bond0 qdisc can happen on random cpus,
due to the fact that qdisc_run() can dequeue a batch of packets.
CPUA -> queue packet P1 on bond0 qdisc, P1->ooo_okay=1
CPUA -> queue packet P2 on bond0 qdisc, P2->ooo_okay=0
CPUB -> dequeue packet P1 from bond0
enqueue packet on eth1/eth2
CPUC -> dequeue packet P2 from bond0
enqueue packet on eth1/eth2 using sk cache (ooo_okay is 0)
get_xps_queue() then might select wrong queue for P1, since current cpu
might be different than CPUA.
P2 might be sent on the old queue (stored in sk->sk_tx_queue_mapping),
if CPUC runs a bit faster (or CPUB spins a bit on qdisc lock)
Effect of this bug is TCP reorders, and more generally not optimal
TX queue placement. (A victim bulk flow can be migrated to the wrong TX
queue for a while)
To fix this, we have to record sender cpu number the first time
dev_queue_xmit() is called for one tx skb.
We can union napi_id (used on receive path) and sender_cpu,
granted we clear sender_cpu in skb_scrub_packet() (credit to Willem for
this union idea)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Nandita Dukkipati <nanditad@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
include/linux/skbuff.h | 7 +++++--
net/core/flow_dissector.c | 7 ++++++-
net/core/skbuff.c | 4 ++++
3 files changed, 15 insertions(+), 3 deletions(-)
(limited to 'net/core/flow_dissector.c')
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 85ab7d72b54c..2748ff639144 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -626,8 +626,11 @@ struct sk_buff {
__u32 hash;
__be16 vlan_proto;
__u16 vlan_tci;
-#ifdef CONFIG_NET_RX_BUSY_POLL
- unsigned int napi_id;
+#if defined(CONFIG_NET_RX_BUSY_POLL) || defined(CONFIG_XPS)
+ union {
+ unsigned int napi_id;
+ unsigned int sender_cpu;
+ };
#endif
#ifdef CONFIG_NETWORK_SECMARK
__u32 secmark;
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index beb83d1ac1c6..2c35c02a931e 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -422,7 +422,7 @@ static inline int get_xps_queue(struct net_device *dev, struct sk_buff *skb)
dev_maps = rcu_dereference(dev->xps_maps);
if (dev_maps) {
map = rcu_dereference(
- dev_maps->cpu_map[raw_smp_processor_id()]);
+ dev_maps->cpu_map[skb->sender_cpu - 1]);
if (map) {
if (map->len == 1)
queue_index = map->queues[0];
@@ -468,6 +468,11 @@ struct netdev_queue *netdev_pick_tx(struct net_device *dev,
{
int queue_index = 0;
+#ifdef CONFIG_XPS
+ if (skb->sender_cpu == 0)
+ skb->sender_cpu = raw_smp_processor_id() + 1;
+#endif
+
if (dev->real_num_tx_queues != 1) {
const struct net_device_ops *ops = dev->netdev_ops;
if (ops->ndo_select_queue)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index a5bff2767f15..88c613eab142 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -825,6 +825,9 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
#ifdef CONFIG_NET_RX_BUSY_POLL
CHECK_SKB_FIELD(napi_id);
#endif
+#ifdef CONFIG_XPS
+ CHECK_SKB_FIELD(sender_cpu);
+#endif
#ifdef CONFIG_NET_SCHED
CHECK_SKB_FIELD(tc_index);
#ifdef CONFIG_NET_CLS_ACT
@@ -4169,6 +4172,7 @@ void skb_scrub_packet(struct sk_buff *skb, bool xnet)
skb->ignore_df = 0;
skb_dst_drop(skb);
skb->mark = 0;
+ skb->sender_cpu = 0;
skb_init_secmark(skb);
secpath_reset(skb);
nf_reset(skb);
--
cgit v1.2.3