【Linux4.1.12源码分析】协议栈报文接收之netif_receive_skb函数分析

netif_receive_skb函数是协议栈报文接收的入口，一般由驱动调用，把报文送入协议栈，4.1.12内核做了对sk的封装，目的是什么没还未搞清楚。

1、netif_receive_skb函数

static inline int netif_receive_skb(struct sk_buff *skb)
{
	return netif_receive_skb_sk(skb->sk, skb);
}

2、netif_receive_skb_sk函数

int netif_receive_skb_sk(struct sock *sk, struct sk_buff *skb)
{
	trace_netif_receive_skb_entry(skb);

	return netif_receive_skb_internal(skb);
}

3、netif_receive_skb_internal函数

static int netif_receive_skb_internal(struct sk_buff *skb)
{
	int ret;

	net_timestamp_check(netdev_tstamp_prequeue, skb);

	if (skb_defer_rx_timestamp(skb))
		return NET_RX_SUCCESS;

	rcu_read_lock();

#ifdef CONFIG_RPS
	if (static_key_false(&rps_needed)) {
		struct rps_dev_flow voidflow, *rflow = &voidflow;
		int cpu = get_rps_cpu(skb->dev, skb, &rflow);		//支持rps，选择报文处理的CPU

		if (cpu >= 0) {
			ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail);	//交给相应的CPU处理
			rcu_read_unlock();
			return ret;
		}
	}
#endif
	ret = __netif_receive_skb(skb);
	rcu_read_unlock();
	return ret;
}

4、__netif_receive_skb函数

static int __netif_receive_skb(struct sk_buff *skb)
{
	int ret;

	if (sk_memalloc_socks() && skb_pfmemalloc(skb)) {
		unsigned long pflags = current->flags;

		/*
		 * PFMEMALLOC skbs are special, they should
		 * - be delivered to SOCK_MEMALLOC sockets only
		 * - stay away from userspace
		 * - have bounded memory usage
		 *
		 * Use PF_MEMALLOC as this saves us from propagating the allocation
		 * context down to all allocation sites.
		 */
		current->flags |= PF_MEMALLOC;
		ret = __netif_receive_skb_core(skb, true);
		tsk_restore_flags(current, pflags, PF_MEMALLOC);
	} else
		ret = __netif_receive_skb_core(skb, false);

	return ret;
}

5、__netif_receive_skb_core函数

static int __netif_receive_skb_core(struct sk_buff *skb, bool pfmemalloc)
{
	struct packet_type *ptype, *pt_prev;
	rx_handler_func_t *rx_handler;
	struct net_device *orig_dev;
	bool deliver_exact = false;
	int ret = NET_RX_DROP;
	__be16 type;

	net_timestamp_check(!netdev_tstamp_prequeue, skb);

	trace_netif_receive_skb(skb);

	orig_dev = skb->dev;

	skb_reset_network_header(skb);		//重置network header，此时skb已经指向IP头（没有vlan的情况下）
	if (!skb_transport_header_was_set(skb))
		skb_reset_transport_header(skb);
	skb_reset_mac_len(skb);		//重置mac len

	pt_prev = NULL;

another_round:
	skb->skb_iif = skb->dev->ifindex;	//设置报文的iif值

	__this_cpu_inc(softnet_data.processed);

	if (skb->protocol == cpu_to_be16(ETH_P_8021Q) ||
	    skb->protocol == cpu_to_be16(ETH_P_8021AD)) {	//vxlan报文处理，剥除vxlan头
		skb = skb_vlan_untag(skb);	//剥除vlan头
		if (unlikely(!skb))
			goto out;
	}

#ifdef CONFIG_NET_CLS_ACT
	if (skb->tc_verd & TC_NCLS) {
		skb->tc_verd = CLR_TC_NCLS(skb->tc_verd);
		goto ncls;
	}
#endif

	if (pfmemalloc)			//此类报文不允许ptype_all处理，即tcpdump也抓不到
		goto skip_taps;

	list_for_each_entry_rcu(ptype, &ptype_all, list) {  //遍历ptype_all，如果有则做相应处理，例如raw socket和tcpdump实现
		if (pt_prev)
			ret = deliver_skb(skb, pt_prev, orig_dev);
		pt_prev = ptype;
	}

	list_for_each_entry_rcu(ptype, &skb->dev->ptype_all, list) {	//设备上注册ptype_all，做相应的处理，更加精细的控制
		if (pt_prev)
			ret = deliver_skb(skb, pt_prev, orig_dev);
		pt_prev = ptype;
	}

skip_taps:
#ifdef CONFIG_NET_CLS_ACT
	if (static_key_false(&ingress_needed)) {
		skb = handle_ing(skb, &pt_prev, &ret, orig_dev);
		if (!skb)
			goto out;
	}

	skb->tc_verd = 0;
ncls:
#endif
	if (pfmemalloc && !skb_pfmemalloc_protocol(skb))
		goto drop;

	if (skb_vlan_tag_present(skb)) {
		if (pt_prev) {
			ret = deliver_skb(skb, pt_prev, orig_dev);
			pt_prev = NULL;
		}
		if (vlan_do_receive(&skb))
			goto another_round;
		else if (unlikely(!skb))
			goto out;
	}

	rx_handler = rcu_dereference(skb->dev->rx_handler);	//设备rx_handler，加入OVS时会注册为OVS的入口函数
	if (rx_handler) {
		if (pt_prev) {
			ret = deliver_skb(skb, pt_prev, orig_dev);
			pt_prev = NULL;
		}
		switch (rx_handler(&skb)) {		//执行rx_handler处理，例如进入OVS，OVS不支持报头中携带vlan的报文
		case RX_HANDLER_CONSUMED:
			ret = NET_RX_SUCCESS;
			goto out;
		case RX_HANDLER_ANOTHER:
			goto another_round;
		case RX_HANDLER_EXACT:
			deliver_exact = true;
		case RX_HANDLER_PASS:
			break;
		default:
			BUG();
		}
	}
	
	if (unlikely(skb_vlan_tag_present(skb))) {<span style="white-space:pre">	</span>
		if (skb_vlan_tag_get_id(skb))
			skb->pkt_type = PACKET_OTHERHOST;
		/* Note: we might in the future use prio bits
		 * and set skb->priority like in vlan_do_receive()
		 * For the time being, just ignore Priority Code Point
		 */
		skb->vlan_tci = 0;
	}

	type = skb->protocol;

	/* deliver only exact match when indicated */
	if (likely(!deliver_exact)) {
		deliver_ptype_list_skb(skb, &pt_prev, orig_dev, type,	//根据全局定义的协议处理报文
				       &ptype_base[ntohs(type) &
						   PTYPE_HASH_MASK]);
	}

	deliver_ptype_list_skb(skb, &pt_prev, orig_dev, type,	//根据设备上注册的协议进行处理
			       &orig_dev->ptype_specific);

	if (unlikely(skb->dev != orig_dev)) {
		deliver_ptype_list_skb(skb, &pt_prev, orig_dev, type,
				       &skb->dev->ptype_specific); //如果设备发生变化，那么还需要针对新设备的注册协议进行处理
	}

	if (pt_prev) {
		if (unlikely(skb_orphan_frags(skb, GFP_ATOMIC)))
			goto drop;
		else
			ret = pt_prev->func(skb, skb->dev, pt_prev, orig_dev);	//调用协议处理
	} else {
drop:
		atomic_long_inc(&skb->dev->rx_dropped);
		kfree_skb(skb);
		/* Jamal, now you will not able to escape explaining
		 * me how you were going to use this. :-)
		 */
		ret = NET_RX_DROP;
	}

out:
	return ret;
}

__netif_receive_skb_core函数主要有几个处理：

1、vlan报文的处理，主要是循环把vlan头剥掉，如果qinq场景，两个vlan都会被剥掉；

2、交给rx_handler处理，例如OVS、linux bridge等；

3、ptype_all处理，例如抓包程序、raw socket等；

4、ptype_base处理，交给协议栈处理，例如ip、arp、rarp等；

后续将逐个针对这些场景进行分析。