本文主要分析：三次握手中最后一个ACK段到达时，服务器端的处理路径。
内核版本：3.6
Author：zhangskd @ csdn blog
函数路径
以下是第三次握手时，服务端接收到ACK后的处理路径。
接收入口
1. 状态为ESTABLISHED时，用tcp_rcv_established()接收处理。
2. 状态为LISTEN时，说明这个sock处于监听状态，用于被动打开的接收处理，包括SYN和ACK。
3. 当状态不为ESTABLISHED或TIME_WAIT时，用tcp_rcv_state_process()处理。
int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
{
struct sock *rsk;
#ifdef CONFIG_TCP_MD5SIG
/* We really want to reject the packet as early as possible if :
* We're expecting an MD5'd packet and this is no MD5 tcp option.
* There is an MD5 option and we're not expecting one.
*/
if (tcp_v4_inbound_md5_hash(sk, skb))
goto discard;
#endif
/* 当状态为ESTABLISHED时，用tcp_rcv_established()接收处理 */
if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */
struct dst_entry *dst = sk->sk_rx_dst;
sock_rps_save_rxhash(sk, skb);
if (dst) {
if (inet_sk(sk)->rx_dst_ifindex != skb->skb_iif || dst->ops->check(dst, 0) == NULL) {
dst_release(dst);
sk->sk_rx_dst = NULL;
}
}
/* 连接已建立时的处理路径 */
if (tcp_rcv_established(sk, skb, tcp_hdr(skb), skb->len)) {
rsk = sk;
goto reset;
}
return 0;
}
/* 检查报文长度、报文校验和 */
if (skb->len < tcp_hdrlen(skb) || tcp_checksum_complete(skb))
goto csum_err;
/* 如果这个sock处于监听状态，被动打开时的处理，包括收到SYN或ACK */
if (sk->sk_state == TCP_LISTEN) {
/* 返回值：
* NULL，错误
* nsk == sk，接收到SYN
* nsk != sk，接收到ACK
*/
struct sock *nsk = tcp_v4_hnd_req(sk, skb); /* 接收ACK的处理 */
if (! nsk)
goto discard;
if (nsk != sk) { /* 接收到ACK时 */
sock_rps_save_rxhash(nsk, skb);
if (tcp_child_process(sk, nsk, skb)) { /* 处理新的sock */
rsk = nsk;
goto reset;
}
return 0;
}
} else
sock_rps_save_rx(sk, skb);
/* 处理除了ESTABLISHED和TIME_WAIT之外的所有状态 */
if (tcp_rcv_state_process(sk, skb, tcp_hdr(skb), skb->len)) {
rsk = sk;
goto reset;
}
return 0;
reset:
tcp_v4_send_reset(rsk, skb); /* 发送RST包 */
discard:
kfree_skb(skb);
return 0;
csum_err:
TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
goto discard;
}
收到SYN段后，服务器端会分配一个连接请求块，并初始化这个连接请求块。
构造和发送SYNACK段。
然后把这个连接请求块链入半连接队列中，启动超时定时器。
之后如果再收到ACK，就能完成三次握手了。
static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
{
struct tcphdr *th = tcp_hdr(skb);
const struct iphdr *iph = ip_hdr(skb);
struct sock *nsk;
struct request_sock **prev;
/* 在半连接队列中查找是否已有符合的连接请求块，如果有，则说明这是三次握手的最后一个ACK。*/
struct request_sock *req = inet_csk_search_req(sk, &prev, th->source, iph->saddr, iph->daddr);
if (req)
return tcp_check_req(sk, skb, req, prev); /* 服务器端处理三次握手的最后一个ACK */
/* 如果在半连接队列中没找到，则在ESTABLISHED状态的哈希表中查找。*/
nsk = inet_lookup_established(sock_net(sk), &tcp_hashinfo, iph->saddr, th->source,
iph->daddr, th->dest, inet_iif(skb));
if (nsk) { /* 如果在ehash表中找到对应的sock，且不处于TIME_WAIT状态 */
if (nsk->sk_state != TCP_TIME_WAIT) {
bh_lock_sock(nsk);
return nsk;
}
inet_twsk_put(inet_twsk(nsk)); /* 释放tw结构体 */
return NULL;
}
#ifdef CONFIG_SYN_COOKIES
/* 如果使用SYN Cookie，则检查cookie是否合法，合法则直接完成三次握手 */
if (! th->syn)
sk = cookie_v4_check(sk, skb, &(IPCB(skb)->opt));
#endif
return sk;
}
在表示半连接队列的哈希表中，寻找符合条件的连接请求块。
struct request_sock *inet_csk_search_req(const struct sock *sk, struct request_sock ***prevp,
const __be16 rport, const __be32 raddr, const __be32 laddr)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
struct listen_sock *lopt = icsk->icsk_accept_queue.listen_opt; /* 半连接队列 */
struct request_sock *req, **prev;
/* 通过哈希值，找到哈希桶，然后遍历哈希桶寻找符合条件的连接请求块 */
for(prev = &lopt->syn_table[inet_synq_hash(raddr, rport, lopt->hash_rnd, lopt->nr_table_entries)];
(req = *prev) != NULL; prev = &req->dl_next) {
const struct inet_request_sock *ireq = inet_rsk(req);
if (ireq->rmt_port == rport && ireq->rmt_addr == raddr && ireq->loc_addr = laddr
&& AF_INET_FAMILY(req->rsk_ops->family)) {
WARN_ON(req->sk); /* 连接尚未建立，sk应该为NULL */
*prevp = prev; /* 保存此req指针的指针 */
break;
}
}
return req;
}
第三次握手
inet_csk_search_req()在半连接队列中查找是否已有符合的连接请求块，如果有，则说明这可能是三次握手的最后一个ACK。
接着调用tcp_check_req()来进行验证，如果合法，则完成三次握手。
/* Process an incoming packet for SYN_RECV sockets represented as a request_sock. */
struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, struct request_sock *req,
struct request_sock **prev)
{
struct tcp_options_received tmp_opt;
const u8 *hash_location;
struct sock *child;
const struct tcphdr *th = tcp_hdr(skb);
__be32 flg = tcp_flag_word(th) & (TCP_FLAG_RST | TCP_FLAG_SYN | TCP_FLAG_ACK);
bool paws_reject = false;
tmp_opt.saw_tstamp = 0;
/* 如果此ACK带有选项 */
if (th->doff > (sizeof(struct tcphdr) >> 2)) {
tcp_parse_options(skb, &tmp_opt, &hash_location, 0, NULL); /* 解析TCP选项，保存到实例中 */
if (tmp_opt.saw_tstamp) {
tmp_opt.ts_recent = req->ts_recent; /* 客户端发送SYN段的时间 */
/* We do not store true stamp, but it is not required,
* it can be estimated (approximately) from another data.
*/
tmp_opt.ts_recent_stamp = get_seconds() - ((TCP_TIMEOUT_INIT/HZ) << req->retrans);
paws_reject = tcp_paws_reject(&tmp_opt, th->rst); /* 检查客户端时间戳是否回绕 */
}
}
/* Check for pure retransmitted SYN. 处理重传的SYN */
if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn && flg == TCP_FLAG_SYN
&& ! paws_reject) {
/* 重新发送SYNACK。
* 实例为tcp_request_sock_ops，调用tcp_v4_rtx_synack()
*/
req->rsk_ops->rtx_syn_ack(sk, req, NULL);
return NULL;
}
/* 如果接收段包含ACK标志，但确认序号不对，则返回监听sock。
* 然后在tcp_v4_do_rcv()中发送RST段。
*/
if ((flg & TCP_FLAG_ACK) && (TCP_SKB_CB(skb)->ack_seq !=
tcp_rsk(req)->snt_isn + 1 + tcp_s_data_size(tcp_sk(sk))))
return sk;
/* 如果发生了回绕，或者接收序号不在接收窗口内 */
if (paws_reject || ! tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq,
tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) {
/* Out of window: send ACK and drop. */
if (! (flg & TCP_FLAG_RST))
/* 发送ACK段。
* 实例为tcp_request_sock_ops，调用tcp_v4_reqsk_send_ack()
*/
req->rsk_ops->send_ack(sk, skb, req);
if (paws_reject)
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
return NULL;
}
/* In sequence, PAWS is ok. */
if (tmp_opt.saw_tstamp && ! after(TCP_SKB_CB(skb)->seq, tcp_rsk(req)->rcv_isn + 1))
req->ts_recent = tmp_opt.rcv_tsval; /* 保存ACK段的时间戳 */
if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn) {
/* Truncate SYN, it is out of window starting at tcp_rsk(req)->rcv_isn + 1 */
flg &= ~TCP_FLAG_SYN;
}
if (flg & (TCP_FLAG_RST | TCP_FLAG_SYN)) {
TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_ATTEMPTFAILS);
goto embryonic_reset;
}
/* ACK sequence verified above, just make sure ACK is set.
* If ACK not set, just silently drop the packet.
*/
if (! (flg & TCP_FLAG_ACK))
return NULL;
/* 如果设置了TCP_DEFER_ACCEPT选项，则不接收纯ACK，等待有负荷的数据包到达后，
* 再建立连接。直接丢弃纯ACK。
*/
if (req->retrans < inet_csk(sk)->icsk_accept_queue.rskq_defer_accept &&
TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) {
inet_rsk(req)->acked = 1;
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPDEFERACCEPTDROP);
return NULL;
}
if (tmp_opt.saw_tstamp && tmp_opt.rcv_tsecr)
tcp_rsk(req)->snt_synack = tmp_opt.rcv_tsecr;
else if (req->retrans) /* don't take RTT sample if retrans && ~TS */
tcp_rsk(req)->snt_synack = 0;
/* OK, ACK is valid, create big socket and feed this segment to it.
* This segment must move socket to established state. If it will be dropped
* after socket is created, wait for troubles.
*/
/* 三次握手完成以后，调用tcp_v4_syn_recv_sock()创建和初始化一个新的传输控制块 */
child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL);
if (child == NULL)
goto listen_overflow;
inet_csk_reqsk_queue_unlink(sk, req, prev); /* 把连接请求块从半连接队列中删除 */
inet_csk_reqsk_queue_removed(sk, req); /* 更新半连接队列的长度，如果为0，则删除定时器 */
/* 把完成三次握手的连接请求块，和新的sock关联起来，并把它移入全连接队列中 */
inet_csk_reqsk_queue_add(sk, req, child);
return child;
listen_overflow:
/* tcp_abort_on_overflow表示全连接队列满了，是给客户端发RST段，还是默默丢弃 */
if (! sysctl_tcp_abort_on_overflow) {
inet_rsk(req)->acked = 1;
return NULL;
}
embryonic_reset:
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_EMBRYONICRSTS);
if (! (flg & TCP_FLAG_RST))
/* 实例为tcp_request_sock_ops，调用tcp_v4_send_reset()。*/
req->rsk_ops->send_reset(sk, skb);
/* 把连接请求块从半连接队列中删除，更新半连接队列 */
inet_csk_reqsk_queue_drop(sk, req, prev);
return NULL;
}
是否发生了回绕。
static inline bool tcp_paws_reject(const struct tcp_options_received *rx_opt, int rst)
{
if (tcp_paws_check(rx_opt, 0))
return false;
/* ACK段包含RST标志 */
if (rst && get_seconds() >= rx_opt->ts_recent_stamp + TCP_PAWS_MSL)
return false;
return true;
}
检查客户端的时间戳是否合法。
要求客户端发送SYN的时间戳 <= 客户端重传SYN的时间戳 、客户端发送ACK的时间戳。
static inline bool tcp_paws_check(const struct tcp_options_received *rx_opt, int paws_win)
{
if ((s32) (rx_opt->ts_recent - rx_opt->rcv_tsval) <= paws_win)
return true;
/* 重传时间超过24天？*/
if (unlikely(get_seconds() >= rx_opt->ts_recent_stamp + TCP_PAWS_24DAYS))
return true;
/* Some OSes send SYN and SYNACK messages with tsval = 0 tsecr = 0,
* then following tcp messages have valid values. Ignore 0 value, or else 'negative'
* tsval might forbid us to accept their packets.
*/
if (! rx_opt->ts_recent)
return true;
return false;
}
检查序号是否合法。
/* @seq：接收段的序号。
* @end_seq：接收段的结束序号。
* @s_win：接收窗口的起始序号。
* @e_win：接收窗口的结束序号。
*/
static bool tcp_in_window(u32 seq, u32 end_seq, u32 s_win, u32 e_win)
{
if (seq == s_win)
return true;
if (after(end_seq, s_win) && before(seq, e_win))
return true;
return seq == e_win && seq == end_seq;
}
连接请求块操作
request_sock_ops为处理连接请求块的函数指针表，对于TCP，它的实例为tcp_request_sock_ops。
struct request_sock_ops tcp_request_sock_ops __read_mostly = {
.family = PF_INET,
.obj_size = sizeof(struct tcp_request_sock),
.rtx_syn_ack = tcp_v4_rtx_synack, /* 重传SYNACK段 */
.send_ack = tcp_v4_reqsk_send_ack, /* 发送ACK段 */
.destructor = tcp_v4_reqsk_destructor,
.send_reset = tcp_v4_send_reset, /* 发送RST段 */
.syn_ack_timeout = tcp_syn_ack_timeout, /* SYNACK段超时处理 */
};
(1) 重传SYNACK段
static int tcp_v4_rtx_synack(struct sock *sk, struct request_sock *req, struct request_values *rvp)
{
TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS);
return tcp_v4_send_synack(sk, NULL, req, rvp, 0, false);
}
我们在上一篇中已分析过tcp_v4_send_synack()，它主要用于构造和发送SYNACK段。
(2) 发送ACK段
在tcp_check_req()中，如果接收到的ACK段时间戳不合法、或者序号不在接收窗口内，且不含RST标志，
则需要给客户端发送一个ACK。
static void tcp_v4_reqsk_send_ack(struct sock *sk, struct sk_buff *skb, struct request_sock *req)
{
tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1,
req->rcv_wnd, req->ts_recent, 0,
tcp_md5_do_lookup(sk, (union tcp_md5_addr *) &ip_hdr(skb)->daddr, AF_INET),
inet_rsk(req)->no_srccheck ? IP_REPLY_ARG_NOSRCCHECK : 0, ip_hdr(skb)->tos);
}
static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts, int oif,
struct tcp_md5sig_key *key, int reply_flags, u8 tos)
{
const struct tcphdr *th = tcp_hdr(skb);
struct {
struct tcphdr th;
__be32 opt[(TCPOLEN_TSTAMP_ALIGNED >> 2)
#ifdef CONFIG_TCP_MD5SIG
+ (TCPOLEN_MD5SIG_ALIGNED >> 2)
#endif
];
} rep;
struct ip_reply_arg arg;
struct net *net = dev_net(skb_dst(skb)->dev);
memset(&rep.th, 0, sizeof(struct tcphdr));
memset(&arg, 0, sizeof(arg));
arg.iov[0].iov_base = (unsigned char *) &rep;
arg.iov[0].iov_len = sizeof(rep.th);
if (ts) { /* 时间戳 */
rep.opt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | (TCPOPT_TIMESTAMP << 8) |
TCPOLEN_TIMESTAMP);
rep.opt[1] = htonl(tcp_time_stamp);
rep.opt[2] = htonl(ts);
arg.iov[0].iov_len += TCPOLEN_TSTAMP_ALIGNED;
}
/* Swap the send and the receive. */
rep.th.dest = th->source;
rep.th.source = th->dest;
rep.th.doff = arg.iov[0].iov_len / 4;
rep.th.seq = htonl(seq);
rep.th.ack_seq = htonl(ack);
rep.th.ack = 1;
rep.th.window = htons(win);
#ifdef CONFIG_TCP_MD5SIG
if (key) {
int offset = (ts) ? 3 : 0;
rep.opt[offset++] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
(TCPOPT_MD5SIG << 8) | TCPOLEN_MD5SIG);
arg.iov[0].iov_len += TCPOLEN_MD5SIG_ALIGNED;
rep.th.doff = arg.iov[0].iov_len / 4;
tcp_v4_md5_hash_addr((__u8 *) &rep.opt[offset], key, ip_hdr(skb)->saddr,
ip_hdr(skb)->daddr, &rep.th);
}
#endif
arg.flags = reply_flags;
arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr, ip_hdr(skb)->saddr,
arg.iov[0].iov_len, IPPROTO_TCP, 0); /* 累加伪首部 */
arg.csumoffset = offsetof(struct tcphdr, check) / 2;
if (oif)
arg.bound_dev_if = oif;
arg.tos = tos;
/* 调用IP层函数，发送此ACK段 */
ip_send_unicast_reply(net, skb, ip_hdr(skb)->saddr, ip_hdr(skb)->daddr, &arg, arg.iov[0].iov_len);
TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
}struct ip_reply_arg {
struct kvec iov[1];
int flags;
__wsum csum;
int csumoffset; /* u16 offset of csum in iov[0].iov_base */
int bound_dev_if;
u8 tos;
};
struct kvec {
void *iov_base;
size_t iov_len;
};
(3) 发送RST段
检测到对端异常时，发送RST段。
static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb)
{
const struct tcphdr *th = tcp_hdr(skb);
struct {
struct tcphdr th;
#ifdef CONFIG_TCP_MD5SIG
__be32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2) ];
#endif
} rep;
struct ip_reply_arg arg; /* 数据报的控制信息 */
#ifdef CONFIG_TCP_MD5SIG
struct tcp_md5sig_key *key;
const __u8 *hash_location = NULL;
unsigned char newhash[16];
int genhash;
struct sock *sk1 = NULL;
#endif
struct net *net;
/* Never send a reset in response to a reset. */
if (th->rst)
return;
if (skb_rtable(skb)->rt_type != RTN_LOCAL)
return;
/* Swap the send and the receive. */
memset(&rep, 0, sizeof(rep));
rep.th.dest = th->source;
rep.th.source = th->dest;
rep.th.doff = sizeof(struct tcphdr) / 4;
rep.th.rst = 1;
if (th->ack) {
rep.th.seq = th->ack_seq;
} else {
rep.th.ack = 1;
rep.th.ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin + skb->len - (th->doff << 2));
}
memset(&arg, 0, sizeof(arg));
arg.iov[0].iov_base = (unsigned char *) &rep;
arg.iov[0].iov_len = sizeof(rep.th);
#ifdef CONFIG_TCP_MD5SIG
/* 此处省略MD5选项的处理 */
...
#endif
arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr, ip_hdr(skb)->saddr,
arg.iov[0].iov_len, IPPROTO_TCP, 0);
arg.csumoffset = offsetof(struct tcphdr, check) / 2;
arg.flags = (sk && inet_sk(sk)->transparent) ? IP_REPLY_ARG_NOSRCCHECK : 0;
/* When socket is gone, all binding information is lost.
* routing might fail in this case. No choice here, if we choose to force input interface,
* we will misroute in case of asymmetric route.
*/
if (sk)
arg.bound_dev_if = sk->sk_bound_dev_if;
net = dev_net(skb_dst(skb)->dev);
arg.tos = ip_hdr(skb)->tos;
/* 调用IP层函数发送 */
ip_send_unicast_reply(net, skb, ip_hdr(skb)->saddr, ip_hdr(skb)->daddr, &arg, arg.iov[0].iov_len);
TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
TCP_INC_STATS_BH(net, TCP_MIB_OUTRSTS);
#ifdef CONFIG_TCP_MD5SIG
/* 省略MD5处理 */
...
#endif
}
(4) 析构函数
释放request_sock实例前调用。
/* IPv4 request_sock destructor. */
static void tcp_v4_reqsk_destructor(struct request_sock *req)
{
kfree(inet_rsk(req)->opt); /* 释放IP选项实例 */
}
(5) 超时处理函数
不是真正的SYNACK超时处理函数，简单更新下统计变量。
void tcp_syn_ack_timeout(struct sock *sk, struct request_sock *req)
{
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPTIMEOUTS);
}