net: release dst entry in dev_hard_start_xmit()

One point of contention in high network loads is the dst_release() performed
when a transmited skb is freed. This is because NIC tx completion calls
dev_kree_skb() long after original call to dev_queue_xmit(skb).

CPU cache is cold and the atomic op in dst_release() stalls. On SMP, this is
quite visible if one CPU is 100% handling softirqs for a network device,
since dst_clone() is done by other cpus, involving cache line ping pongs.

It seems right place to release dst is in dev_hard_start_xmit(), for most
devices but ones that are virtual, and some exceptions.

David Miller suggested to define a new device flag, set in alloc_netdev_mq()
(so that most devices set it at init time), and carefuly unset in devices
which dont want a NULL skb->dst in their ndo_start_xmit().

List of devices that must clear this flag is :

- loopback device, because it calls netif_rx() and quoting Patrick :
    "ip_route_input() doesn't accept loopback addresses, so loopback packets
     already need to have a dst_entry attached."
- appletalk/ipddp.c : needs skb->dst in its xmit function

- And all devices that call again dev_queue_xmit() from their xmit function
(as some classifiers need skb->dst) : bonding, vlan, macvlan, eql, ifb, hdlc_fr

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

2 files changed, 10 insertions, 0 deletions

diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 8faacee68633..ff7572ac5481 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -739,6 +739,7 @@ void vlan_setup(struct net_device *dev)
         ether_setup(dev);
 
         dev->priv_flags         |= IFF_802_1Q_VLAN;
+        dev->priv_flags         &= ~IFF_XMIT_DST_RELEASE;
         dev->tx_queue_len       = 0;
 
         dev->netdev_ops         = &vlan_netdev_ops;
diff --git a/net/core/dev.c b/net/core/dev.c
index 6d3630d16271..92ebeca29901 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1688,6 +1688,14 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
                                 goto gso;
                 }
 
+                /*
+                 * If device doesnt need skb->dst, release it right now while
+                 * its hot in this cpu cache
+                 */
+                if ((dev->priv_flags & IFF_XMIT_DST_RELEASE) && skb->dst) {
+                        dst_release(skb->dst);
+                        skb->dst = NULL;
+                }
                 rc = ops->ndo_start_xmit(skb, dev);
                 /*
                  * TODO: if skb_orphan() was called by
@@ -5045,6 +5053,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
         netdev_init_queues(dev);
 
         INIT_LIST_HEAD(&dev->napi_list);
+        dev->priv_flags = IFF_XMIT_DST_RELEASE;
         setup(dev);
         strcpy(dev->name, name);
         return dev;