ip_output-设置输出设备和协议,然后经过POST_ROUTING钩子点,最后调用ip_finish_output; ip_finish_output-对skb进行分片判断,需要分片,则分片后输出,不需要分片则知直接输出; ip_finish_output2-对skb的头部空间进行检查,看是否能够容纳下二层头部,若空间不足,则需要重新申请skb;然后,获取邻居子系统,并通过邻居子系...
IP层发包最后会调用ip_finish_output2函数,我们从该函数入手分析邻居子系统。 ip_finish_output2函数 AI检测代码解析 1. static inline int ip_finish_output2(struct sock *sk, struct sk_buff *skb) 2. { 3. struct dst_entry *dst = skb_dst(skb); 4. struct rtable *rt = (struct rtable *)ds...
也就是说啊,内核中的收发包的路径,很可能是经理driver_recv --> tcp -->driver_send这样一个过程,是个很长的路径呢... 从dev_queue_xmit中来的数据包全部都是从ip_finish_output中来的吗? 所以,应该经常能抓到ip_local_out才对,因为tcp是需要发确认帧的, 不过就下面这些内容来看,内核里的tcp/ip的协议...
Re: ip_finish_output2: No header cache and no neighbo... Patrick McHardy Re: ip_finish_output2: No header cache and no nei... Andi Kleen Re: ip_finish_output2: No header cache and no... David S. Miller Re: ip_finish_output2: No header cache an... Patrick McHardy Reply via ...
Re: ip_finish_output2: No header cache and no neighbo...Andi Kleen Re: ip_finish_output2: No header cache and no nei...Herbert Xu Re: ip_finish_output2: No header cache and no neighbo...Patrick McHardy Re: ip_finish_output2: No header cache and no nei...Herbert Xu ...