内核listen的backlog和简单的三次握手分析
1. 背景：
前面通过抓包分析了listen backlog对全连接和半连接的影响，本文将从内核源码(kernel 2.6.32)上简单了解下服务端三次握手的过程以及backlog在中间所起的作用。
2. 三次握手：
2.1 服务端监听：
在system_call后，通过fd号获取相应的socket，及对backlog最值进行限制后，然后进入inet_listen函数进行处理。
int inet_listen(struct socket *sock, int backlog)
{
struct sock *sk = sock->sk;
unsigned char old_state;
int err;
lock_sock(sk);
err = -EINVAL;
if (sock->state != SS_UNCONNECTED || sock->type != SOCK_STREAM)
goto out;
old_state = sk->sk_state;
if (!((1 << old_state) & (TCPF_CLOSE | TCPF_LISTEN)))
goto out;
/* Really, if the socket is already in listen state
* we can only allow the backlog to be adjusted.
*/
if (old_state != TCP_LISTEN) {
err = inet_csk_listen_start(sk, backlog);
if (err)
goto out;
}
sk->sk_max_ack_backlog = backlog;
err = 0;
out:
release_sock(sk);
return err;
}static inline int sk_acceptq_is_full(struct sock *sk)
{
return sk->sk_ack_backlog > sk->sk_max_ack_backlog;
}int inet_csk_listen_start(struct sock *sk, const int nr_table_entries)
{
struct inet_sock *inet = inet_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
int rc = reqsk_queue_alloc(&icsk->icsk_accept_queue, nr_table_entries);
if (rc != 0)
return rc;
/* 初始化backlog为0，回到上层后赋值，见前文。 */
sk->sk_max_ack_backlog = 0;
sk->sk_ack_backlog = 0;
inet_csk_delack_init(sk);
/* There is race window here: we announce ourselves listening,
* but this transition is still not validated by get_port().
* It is OK, because this socket enters to hash table only
* after validation is complete.
*/
/* 检查端口是否被占用。 */
sk->sk_state = TCP_LISTEN;
if (!sk->sk_prot->get_port(sk, inet->num)) {
inet->sport = htons(inet->num);
sk_dst_reset(sk);
sk->sk_prot->hash(sk);
return 0;
}
sk->sk_state = TCP_CLOSE;
__reqsk_queue_destroy(&icsk->icsk_accept_queue);
return -EADDRINUSE;
}int reqsk_queue_alloc(struct request_sock_queue *queue, unsigned int nr_table_entries)
{
size_t lopt_size = sizeof(struct listen_sock);
/** struct listen_sock - listen state
*
* @max_qlen_log - log_2 of maximal queued SYNs/REQUESTs
*/
/*
struct listen_sock {
/* 哈希大小。 */
u8 max_qlen_log;
/* 3 bytes hole, try to use */
/* 用来记录半连接数量和未超时的半连接数量。 */
int qlen;
int qlen_young;
/* 将用于半连接超时重传。 */
int clock_hand;
/* 用于计算哈希值。 */
u32 hash_rnd;
u32 nr_table_entries;
/* 柔性数组。 */
struct request_sock *syn_table[0];
};
listen_sock是一个柔性数组，是一个用来存放半连接的哈希表，
哈希表大小为backlog向上取2的幂次。
*/
struct listen_sock *lopt;
/* 柔性数组大小计算，在[8, sysctl_max_syn_backlog]之间的backlog向上取2的幂。 */
nr_table_entries = min_t(u32, nr_table_entries, sysctl_max_syn_backlog);
nr_table_entries = max_t(u32, nr_table_entries, 8);
nr_table_entries = roundup_pow_of_two(nr_table_entries + 1);
lopt_size += nr_table_entries * sizeof(struct request_sock *);
if (lopt_size > PAGE_SIZE)
lopt = __vmalloc(lopt_size, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, PAGE_KERNEL);
else
lopt = kzalloc(lopt_size, GFP_KERNEL);
if (lopt == NULL)
return -ENOMEM;
/* 哈希至少8。 */
for (lopt->max_qlen_log = 3; (1 << lopt->max_qlen_log) < nr_table_entries; lopt->max_qlen_log++);
get_random_bytes(&lopt->hash_rnd, sizeof(lopt->hash_rnd));
rwlock_init(&queue->syn_wait_lock);
queue->rskq_accept_head = NULL;
lopt->nr_table_entries = nr_table_entries;
write_lock_bh(&queue->syn_wait_lock);
queue->listen_opt = lopt;
write_unlock_bh(&queue->syn_wait_lock);
return 0;
}2.2 第一次握手的syn到达：
当有报文到达，ipv4的tcp处理入口为tcp_v4_do_rcv，下面代码中去掉了一些校验的处理和已建立连接的处理，只留下监听的主要流程。第一次握手时候收到的是syn包，因此在tcp_v4_hnd_req之后走到tcp_rcv_state_process处理当前收到的包。
int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
{
struct sock *rsk;
if (sk->sk_state == TCP_LISTEN) {
struct sock *nsk = tcp_v4_hnd_req(sk, skb);
/* 错误，丢弃包。 */
if (!nsk)
goto discard;
/* 收到ack. */
if (nsk != sk) {
if (tcp_child_process(sk, nsk, skb)) {
rsk = nsk;
goto reset;
}
return 0;
}
}
/* 处理报文。 */
TCP_CHECK_TIMER(sk);
if (tcp_rcv_state_process(sk, skb, tcp_hdr(skb), skb->len)) {
rsk = sk;
goto reset;
}
TCP_CHECK_TIMER(sk);
return 0;
reset:
tcp_v4_send_reset(rsk, skb);
discard:
kfree_skb(skb);
/* Be careful here. If this function gets more complicated and
* gcc suffers from register pressure on the x86, sk (in %ebx)
* might be destroyed here. This current version compiles correctly,
* but you have been warned.
*/
return 0;
}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;
/* 半连接中查找，找到的会进行检查，如果半连接中没有，查找已完成的连接，都不是则进入syn接收。 */
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);
nsk = inet_lookup_established(sock_net(sk), &tcp_hashinfo, iph->saddr,
th->source, iph->daddr, th->dest, inet_iif(skb));
if (nsk) {
if (nsk->sk_state != TCP_TIME_WAIT) {
bh_lock_sock(nsk);
return nsk;
}
inet_twsk_put(inet_twsk(nsk));
return NULL;
}
#ifdef CONFIG_SYN_COOKIES
if (!th->rst && !th->syn && th->ack)
sk = cookie_v4_check(sk, skb, &(IPCB(skb)->opt));
#endif
return sk;
}接下来在处理包的函数tcp_rcv_state_process中的第一个状态机，进入监听状态且收到syn包的处理函数tcp_v4_conn_request。
简化一部分代码，保留主要流程如下：
1. 检查半连接队列是否已满，满的大小为backlog修正值向上取2的幂次，满的话会丢弃这个syn包。
2. 检查完成队列是否空，且新的请求数是不是大于1，注意，这里新的请求指的是刚刚发来syn，还未重传或者确认的请求，新的请求在tcp_check_req接收了ack进入连接状态以后会重置qlen_young，或者在第一次重传的时候重置，个人理解是把服务端的压力分摊到客户端上，让重传发生在客户端，减少服务端的压力。
3. 可以创建半连接，则会初始化一个request_sock，并发送syn + ack，同时将这个req添加到监听sock的syn_table中，下一次tcp_v4_hnd_req的时候会在半连接表中找到该链接。
4. 添加进哈希表以后会激活对应socket的keepalive事件，后面会分析。
int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
{
...
/* TW buckets are converted to open requests without
* limitations, they conserve resources and peer is
* evidently real one.
*/
if (inet_csk_reqsk_queue_is_full(sk) && !isn) {
#ifdef CONFIG_SYN_COOKIES
if (sysctl_tcp_syncookies) {
want_cookie = 1;
} else
#endif
goto drop;
}
/* Accept backlog is full. If we have already queued enough
* of warm entries in syn queue, drop request. It is better than
* clogging syn queue with openreqs with exponentially increasing
* timeout.
*/
/* 当接收队列已满，且新的未重试请求数量大于1，服务端会暂时丢弃这个报文，
* 个人理解是为了降低服务端的压力，将重传放到客户端（重传syn），
* 而不是服务端（syn + ack）。
*/
if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1)
goto drop;
req = inet_reqsk_alloc(&tcp_request_sock_ops);
if (!req)
goto drop;
#ifdef CONFIG_TCP_MD5SIG
tcp_rsk(req)->af_specific = &tcp_request_sock_ipv4_ops;
#endif
tcp_clear_options(&tmp_opt);
tmp_opt.mss_clamp = 536;
tmp_opt.user_mss = tcp_sk(sk)->rx_opt.user_mss;
tcp_parse_options(skb, &tmp_opt, 0);
if (want_cookie && !tmp_opt.saw_tstamp)
tcp_clear_options(&tmp_opt);
tmp_opt.tstamp_ok = tmp_opt.saw_tstamp;
tcp_openreq_init(req, &tmp_opt, skb);
ireq = inet_rsk(req);
ireq->loc_addr = daddr;
ireq->rmt_addr = saddr;
ireq->no_srccheck = inet_sk(sk)->transparent;
ireq->opt = tcp_v4_save_options(sk, skb);
if (security_inet_conn_request(sk, skb, req))
goto drop_and_free;
if (!want_cookie)
TCP_ECN_create_request(req, tcp_hdr(skb));
...
/* 发送syn + ack。 */
if (__tcp_v4_send_synack(sk, req, dst) || want_cookie)
goto drop_and_free;
/* 添加req到半连接的syn_table中，同时启动一个超时检测的定时器。 */
inet_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT);
return 0;
drop_and_release:
dst_release(dst);
drop_and_free:
reqsk_free(req);
drop:
return 0;
}第一次握手主要有一个细节，就是上面提到的，队列已满时候，会适当丢弃syn。
2.3 第三次握手：
第三次握手时候服务端收到客户端反馈的ack，这时候依旧是在tcp_v4_do_rcv中处理这个包，但是不同的是在tcp_v4_hnd_req的时候，会在syn_table中查找到req，这时候会进入tcp_check_req检查这个半连接。详细检查可以见参考文档[2]，当检查完成后，会创建一个子sock用来处理连接，主要处理在tcp_v4_syn_recv_sock函数中完成。当子sock创建失败，如接受队列已满，则会根据系统的sysctl_tcp_abort_on_overflow标志判断是否需要向对端发送RST，或者是简单丢弃该包，等待后续的重传。
如果子sock建立成功了，则会从哈希桶中移除，且减少相应的半连接计数，移除相应的定时器。
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);
inet_csk_reqsk_queue_removed(sk, req);
inet_csk_reqsk_queue_add(sk, req, child);
return child;
listen_overflow:
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))
req->rsk_ops->send_reset(sk, skb);
inet_csk_reqsk_queue_drop(sk, req, prev);
return NULL;
} 定时器总入口为tcp_keepalive_timer，对应的监听状态的入口为inet_csk_reqsk_queue_prune。
void inet_csk_reqsk_queue_prune(struct sock *parent,
const unsigned long interval,
const unsigned long timeout,
const unsigned long max_rto)
{
struct inet_connection_sock *icsk = inet_csk(parent);
struct request_sock_queue *queue = &icsk->icsk_accept_queue;
struct listen_sock *lopt = queue->listen_opt;
int max_retries = icsk->icsk_syn_retries ? : sysctl_tcp_synack_retries;
int thresh = max_retries;
unsigned long now = jiffies;
struct request_sock **reqp, *req;
int i, budget;
if (lopt == NULL || lopt->qlen == 0)
return;
/* Normally all the openreqs are young and become mature
* (i.e. converted to established socket) for first timeout.
* If synack was not acknowledged for 1 second, it means
* one of the following things: synack was lost, ack was lost,
* rtt is high or nobody planned to ack (i.e. synflood).
* When server is a bit loaded, queue is populated with old
* open requests, reducing effective size of queue.
* When server is well loaded, queue size reduces to zero
* after several minutes of work. It is not synflood,
* it is normal operation. The solution is pruning
* too old entries overriding normal timeout, when
* situation becomes dangerous.
*
* Essentially, we reserve half of room for young
* embrions; and abort old ones without pity, if old
* ones are about to clog our table.
*/
/* 当半连接数量增大，但是young增大的速度比较平缓，这时候thresh会逐渐变小，
* 半连接就更容易过期。
*/
if (lopt->qlen>>(lopt->max_qlen_log-1)) {
int young = (lopt->qlen_young<<1);
while (thresh > 2) {
if (lopt->qlen < young)
break;
thresh--;
young <<= 1;
}
}
if (queue->rskq_defer_accept)
max_retries = queue->rskq_defer_accept;
budget = 2 * (lopt->nr_table_entries / (timeout / interval));
i = lopt->clock_hand;
/* 遍历桶。 */
do {
reqp=&lopt->syn_table[i];
while ((req = *reqp) != NULL) {
/* 已经达到超时时间。 */
if (time_after_eq(now, req->expires)) {
int expire = 0, resend = 0;
/* 超时和过期的计算。 */
syn_ack_recalc(req, thresh, max_retries,
queue->rskq_defer_accept,
&expire, &resend);
req->rsk_ops->syn_ack_timeout(parent, req);
if (!expire &&
(!resend ||
!inet_rtx_syn_ack(parent, req) ||
inet_rsk(req)->acked)) {
/* 超时重传。 */
unsigned long timeo;
/* 首次超时则这个请求不再是young。 */
if (req->num_timeout++ == 0)
lopt->qlen_young--;
/* 超时时间指数增加。 */
timeo = min(timeout << req->num_timeout,
max_rto);
req->expires = now + timeo;
reqp = &req->dl_next;
continue;
}
/* Drop this request */
/* 过期。 */
inet_csk_reqsk_queue_unlink(parent, req, reqp);
reqsk_queue_removed(queue, req);
reqsk_free(req);
continue;
}
reqp = &req->dl_next;
}
i = (i + 1) & (lopt->nr_table_entries - 1);
} while (--budget > 0);
lopt->clock_hand = i;
/* 刷新定时事件。 */
if (lopt->qlen)
inet_csk_reset_keepalive_timer(parent, interval);
}3. 参考文献：
[1]. 函数调用关系： http://dedecms.com/knowledge/servers/linux-bsd/2012/1217/17745_3.html
[2]. http://blog.csdn.net/zhangskd/article/details/17923917
[3]. 定时器：http://blog.csdn.net/zhangskd/article/details/35281345