Veth pair 是一对虚拟网卡,从一张veth网卡发出的数据包可以直接到达它的peer veth,两者之间存在着虚拟链路。
Veth 网卡和常规的以太网区别仅在于xmit接口:将数据发送到其peer,触发peer的Rx 过程。
Veth 的原理示意图如下:
Veth 的实现在linux/drivers/net/veth.c 下,总体来看比较简单:
关键的数据结构:
-
struct veth_priv {
-
struct net_device __rcu *peer;
-
atomic64_t dropped;
-
};
在初始化时veth_newlink 函数中有:
点击(此处)折叠或打开
-
/*
-
* tie the deviced together
-
*/
-
-
priv = netdev_priv(dev);
-
rcu_assign_pointer(priv->peer, peer);
-
-
priv = netdev_priv(peer);
-
rcu_assign_pointer(priv->peer, dev);
在发送函数veth_xmit中有:
-
rcv = rcu_dereference(priv->peer);
-
...
-
if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
-
struct pcpu_vstats *stats = this_cpu_ptr(dev->vstats);
-
-
u64_stats_update_begin(&stats->syncp);
-
stats->bytes += length;
-
stats->packets++;
-
u64_stats_update_end(&stats->syncp);
-
} else {
应用方式:通过net namespace 做实验如下:
-
#!/bin/sh
-
-
echo "create net namespace net0 and net1"
-
ip netns add net0
-
ip netns add net1
-
-
echo "list net namespace"
-
ip netns list
-
-
echo "add veth pair v1 and vp1"
-
ip link add veth_0 type veth peer name veth_0_peer
-
ip link
-
-
echo "set veth_0 in net0"
-
ip link set veth_0 netns net0
-
echo "set veth_0_peer in net1"
-
ip link set veth_0_peer netns net1
-
-
ip netns exec net0 ip addr add local 10.0.78.3/24 dev veth_0
-
ip netns exec net0 ifconfig veth_0 up
-
ip netns exec net1 ip addr add local 10.0.78.4/24 dev veth_0_peer
-
ip netns exec net1 ifconfig veth_0_peer up
-
-
echo "show ip netns net0"
-
ip netns exec net0 ip addr
-
echo "show ip netns net1"
-
ip netns exec net1 ip addr
-
-
ip netns exec net1 ping 10.0.78.3
阅读(1730) | 评论(0) | 转发(0) |