dst_output發包

不管是收到報文轉發還是本機發送報文，最后都會調用dst_output

/* Output packet to network from transport.  */
static inline int dst_output(struct net *net, struct sock *sk, struct sk_buff *skb)
{
   /*
     * 如果是單播數據包，設置的是ip_output(),
     * 如果是組播數據包，設置的是ip_mc_output().dev_queue_xmit
     */
    return skb_dst(skb)->output(net, sk, skb);
}

 單播：

/*
 * 對於單播數據包，目的路由緩存項中的輸出接口是ip_output().
 */
int ip_output(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	struct net_device *dev = skb_dst(skb)->dev;

	IP_UPD_PO_STATS(net, IPSTATS_MIB_OUT, skb->len);
	/*
	 * 設置數據包的輸出網絡設備和數據包網絡
	 * 層協議類型。
	 */
	skb->dev = dev;
	skb->protocol = htons(ETH_P_IP);
/*
	 * 經netfilter處理后，調用ip_finish_output()繼續IP數據包的輸出
	 */
	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING,
			    net, sk, skb, NULL, dev,
			    ip_finish_output,
			    !(IPCB(skb)->flags & IPSKB_REROUTED));
}

 

static int ip_finish_output(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	unsigned int mtu;

#if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM)
	/* Policy lookup after SNAT yielded a new policy */
	if (skb_dst(skb)->xfrm) {
		IPCB(skb)->flags |= IPSKB_REROUTED;
		return dst_output(net, sk, skb);
	}
#endif
/*
//如果不支持TSO或者GSO，tcp發送的時候是按照mss來組織skb的，
所以skb->len會等於mtu  所以TCP叫分段，和IP分片不一樣，只有UDP才有IP分片
//SKB不是gso類型，並且skb->len大於mtu則需要分片  
對方接受后的分片重組在netfilter中的ipv4_conntrack_defrag
*/
	mtu = ip_skb_dst_mtu(sk, skb);
	if (skb_is_gso(skb))
		return ip_finish_output_gso(net, sk, skb, mtu);
 /* 如果數據包長度大於MTU，則調用ip_fragment()
	 * 對IP數據包進行分片。
	 */
	if (skb->len > mtu || (IPCB(skb)->flags & IPSKB_FRAG_PMTU))
		return ip_fragment(net, sk, skb, mtu, ip_finish_output2);

	return ip_finish_output2(net, sk, skb);
}

 

/* ip send the packet  by ip_finish_output2*/
static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *skb)
{
	struct dst_entry *dst = skb_dst(skb);
	struct rtable *rt = (struct rtable *)dst;
	struct net_device *dev = dst->dev;
	unsigned int hh_len = LL_RESERVED_SPACE(dev);
	struct neighbour *neigh;
	u32 nexthop;

	if (rt->rt_type == RTN_MULTICAST) {
		IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTMCAST, skb->len);
	} else if (rt->rt_type == RTN_BROADCAST)
		IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTBCAST, skb->len);

	/* Be paranoid, rather than too clever. */
	 /* skb頭部空間不能存儲鏈路頭 */
	if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) {
		struct sk_buff *skb2;
		 /* 重新分配skb */
		skb2 = skb_realloc_headroom(skb, LL_RESERVED_SPACE(dev));
		if (!skb2) {
			kfree_skb(skb);
			return -ENOMEM;
		}
		if (skb->sk)/* 關聯控制塊 */
			skb_set_owner_w(skb2, skb->sk);
		consume_skb(skb); /* 釋放skb */
		skb = skb2; /* 指向新的skb */
	}

	rcu_read_lock_bh();
	/* 獲取下一跳 */
	nexthop = (__force u32) rt_nexthop(rt, ip_hdr(skb)->daddr);// get the dst ip address (u32)
	neigh = __ipv4_neigh_lookup_noref(dev, nexthop); 
	//根據目的IP查找鄰居項是否存在  
	//如果沒有則創建鄰居項，然后通過dst_neigh_output  發包
	if (unlikely(!neigh))
		neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);
	if (!IS_ERR(neigh)) {/* 成功 */
		int res = dst_neigh_output(dst, neigh, skb); /* 通過鄰居子系統輸出 */

		rcu_read_unlock_bh();
		return res;
	}
	rcu_read_unlock_bh();

	net_dbg_ratelimited("%s: No header cache and no neighbour!\n",
			    __func__);
	kfree_skb(skb);
	return -EINVAL;
}

 

最后通過令鄰居子系統，調用dev_queue_xmit 將數據報文發送給鏈路層驅動