TCP发送源码学习(3)--tcp_transmit_skb
http://blog.chinaunix.net/uid-9543173-id-3560665.html
一、tcp_transmit_skb
static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
gfp_t gfp_mask)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
struct inet_sock *inet;
struct tcp_sock *tp;
struct tcp_skb_cb *tcb;
struct tcp_out_options opts;
unsigned tcp_options_size, tcp_header_size;
struct tcp_md5sig_key *md5;
struct tcphdr *th;
int err;
BUG_ON(!skb || !tcp_skb_pcount(skb));
/* If congestion control is doing timestamping, we must
* take such a timestamp before we potentially clone/copy.
*/
/*如果拥塞控制需要做时间才有,则必须在克隆或者拷贝报文之前设置一个时间戳。
linux支持了多达十种拥塞控制算法,但并不是每种算中都需要做时间采样的,
因此在设置时间戳前先判断当前的拥塞算法是否需要做时间采样。*/
if (icsk->icsk_ca_ops->flags & TCP_CONG_RTT_STAMP)
__net_timestamp(skb);
/*根据传递进来的clone_it参数来确定是否需要克隆待发送的报文。*/
if (likely(clone_it)) {
/*如果skb已经被clone,则只能复制该skb的数据到新分配的skb中*/
if (unlikely(skb_cloned(skb)))
skb = pskb_copy(skb, gfp_mask);
else
/*clone新的skb*/
skb = skb_clone(skb, gfp_mask);
if (unlikely(!skb))
return -ENOBUFS;
}
/*获取INET层和TCP层的传输控制块、skb中的TCP私有数据块。*/
inet = inet_sk(sk);
tp = tcp_sk(sk);
tcb = TCP_SKB_CB(skb);
memset(&opts, 0, sizeof(opts));
/*根据TCP选项重新调整TCP首部的长度。*/
/*判断当前TCP报文是否是SYN段,因为有些选项只能出现在SYN报文中,需做特别处理。*/
if (unlikely(tcb->flags & TCPHDR_SYN))
tcp_options_size = tcp_syn_options(sk, skb, &opts, &md5);
else
tcp_options_size = tcp_established_options(sk, skb, &opts, &md5);
/*tcp首部的总长度等于可选长度加上struct tcphdr。*/
tcp_header_size = tcp_options_size + sizeof(struct tcphdr);
/*如果已发出但未确认的数据包数目为零,则只初始化拥塞控制,并开始跟踪该连接的RTT。*/
if (tcp_packets_in_flight(tp) == 0)
tcp_ca_event(sk, CA_EVENT_TX_START);
/*调用skb_push()在数据部分的头部添加TCP首部,长度即为之前计算得到的那个tcp_header_size,实际上是把data指针往上移。*/
skb_push(skb, tcp_header_size);
skb_reset_transport_header(skb);
/*SKB已添加到发送队列中,但是从SKB的角度去看还不知道他是属于哪个传输控制块,因此调用skb_set_owner_w设置该SKB的宿主。*/
skb_set_owner_w(skb, sk);
/* Build TCP header and checksum it. */
/*填充TCP首部中的源端口source、目的端口dest、TCP报文的序号seq、确认序号ack_seq以及各个标志位*/
th = tcp_hdr(skb);
th->source = inet->inet_sport;
th->dest = inet->inet_dport;
th->seq = htonl(tcb->seq);
th->ack_seq = htonl(tp->rcv_nxt);
*(((__be16 *)th) + 6) = htons(((tcp_header_size >> 2) << 12) |
tcb->flags);
/*分两种情况设置TCP首部的接收窗口的大小*/
if (unlikely(tcb->flags & TCPHDR_SYN)) {
/* RFC1323: The window in SYN & SYN/ACK segments
* is never scaled.
*/
/*如果是SYN段,则设置接收窗口初始值为rcv_wnd*/
th->window = htons(min(tp->rcv_wnd, 65535U));
} else {
/*如果是其他的报文,则调用tcp_select_window()计算当前接收窗口的大小。*/
th->window = htons(tcp_select_window(sk));
}
/*初始化TCP首部的校验码和紧急指针,具体请参考TCP协议中的首部定义。*/
th->check = 0;
th->urg_ptr = 0;
/* The urg_mode check is necessary during a below snd_una win probe */
if (unlikely(tcp_urg_mode(tp) && before(tcb->seq, tp->snd_up))) {
if (before(tp->snd_up, tcb->seq + 0x10000)) {
th->urg_ptr = htons(tp->snd_up - tcb->seq);
th->urg = 1;
} else if (after(tcb->seq + 0xFFFF, tp->snd_nxt)) {
th->urg_ptr = htons(0xFFFF);
th->urg = 1;
}
}
tcp_options_write((__be32 *)(th + 1), tp, &opts);
if (likely((tcb->flags & TCPHDR_SYN) == 0))
TCP_ECN_send(sk, skb, tcp_header_size);
#ifdef CONFIG_TCP_MD5SIG
/* Calculate the MD5 hash, as we have all we need now */
if (md5) {
sk_nocaps_add(sk, NETIF_F_GSO_MASK);
tp->af_specific->calc_md5_hash(opts.hash_location,
md5, sk, NULL, skb);
}
#endif
icsk->icsk_af_ops->send_check(sk, skb);
if (likely(tcb->flags & TCPHDR_ACK))
tcp_event_ack_sent(sk, tcp_skb_pcount(skb));
if (skb->len != tcp_header_size)
tcp_event_data_sent(tp, skb, sk);
if (after(tcb->end_seq, tp->snd_nxt) || tcb->seq == tcb->end_seq)
TCP_ADD_STATS(sock_net(sk), TCP_MIB_OUTSEGS,
tcp_skb_pcount(skb));
/*调用发送接口queue_xmit发送报文,进入到ip层,如果失败返回错误码。在TCP中该接口实现函数为ip_queue_xmit()*/
err = icsk->icsk_af_ops->queue_xmit(skb);
if (likely(err <= 0))
return err;
tcp_enter_cwr(sk, 1);
return net_xmit_eval(err);
}