Linux源码看Socket 从Linux源码看Socket(TCP)Client端的Connect的示例详解

int clientSocket;
if((clientSocket = socket(AF_INET, SOCK_STREAM, 0)) < 0) {
	// 创建socket失败失败
 	return -1;
}
......
if(connect(clientSocket, (struct sockaddr *)&serverAddr, sizeof(serverAddr)) < 0) {
	// connect 失败
	return -1;
}
.......

sock_map_fd
	|->get_unused_fd_flags
			|->alloc_fd
				|->expand_files （ulimit)
	|->sock_alloc_file	
		|->alloc_file
			|->get_empty_filp (/proc/sys/fs/max_files)

int expand_files(struct files_struct *files, int nr
{
	......
	if (nr >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
		return -EMFILE;
	......
}

struct file *get_empty_filp(void)
{
 ......
	/*
	 * 由此可见，特权用户可以无视文件数最大大小的限制！
	 */
	if (get_nr_files() >= files_stat.max_files && !capable(CAP_SYS_ADMIN)) {
		/*
		 * percpu_counters are inaccurate. Do an expensive check before
		 * we go and fail.
		 */
		if (percpu_counter_sum_positive(&nr_files) >= files_stat.max_files)
			goto over;
	}
	
 ......
}

int connect(int sockfd,const struct sockaddr *serv_addr,socklen_t addrlen)

SYSCALL_DEFINE3(connect, ......

socket.c
SYSCALL_DEFINE3(connect, int, fd, struct sockaddr __user *, uservaddr,
		int, addrlen)
{
 ......
	err = sock->ops->connect(sock, (struct sockaddr *)&address, addrlen,
				 sock->file->f_flags);
	......
}

SYSCALL_DEFINE3(connect
	|->inet_stream_ops
		|->inet_stream_connect
			|->tcp_v4_connect
				|->tcp_set_state(sk, TCP_SYN_SENT);设置状态为TCP_SYN_SENT
			 	|->inet_hash_connect
				|->tcp_connect

void inet_get_local_port_range(int *low, int *high)
{
	unsigned int seq;

	do {
		// 顺序锁
		seq = read_seqbegin(&sysctl_local_ports.lock);

		*low = sysctl_local_ports.range[0];
		*high = sysctl_local_ports.range[1];
	} while (read_seqretry(&sysctl_local_ports.lock, seq));
}

cat /proc/sys/net/ipv4/ip_local_port_range 
32768 61000 

int __inet_hash_connect(...)
{
		// 注意，这边是static变量
		static u32 hint;
		// 这边的port_offset是用对端ip:port hash的一个值
		// 也就是说对端ip:port固定,port_offset固定
		u32 offset = hint + port_offset;
		for (i = 1; i <= remaining; i++) {
			port = low + (i + offset) % remaining;
			/* port是否占用check */
			....
			goto ok;
		}
		.......
ok:
		hint += i;
		......
}

C语言connect返回错误码为
-EADDRNOTAVAIL，对应描述为Cannot assign requested address 
对应Java的异常为
java.net.NoRouteToHostException: Cannot assign requested address (Address not available)

#define TCP_TIMEWAIT_LEN (60*HZ) /* how long to wait to destroy TIME-WAIT
				 * state, about 60 seconds	*/

__inet_hash_connect
	|->__inet_check_established
static int __inet_check_established(......)
{
	......	
	/* Check TIME-WAIT sockets first. */
	sk_nulls_for_each(sk2, node, &head->twchain) {
		tw = inet_twsk(sk2);
		// 如果在time_wait中找到一个match的port,就判断是否可重用
		if (INET_TW_MATCH(sk2, net, hash, acookie,
					saddr, daddr, ports, dif)) {
			if (twsk_unique(sk, sk2, twp))
				goto unique;
			else
				goto not_unique;
		}
	}
	......
}

int tcp_twsk_unique(......)
{
	......
	if (tcptw->tw_ts_recent_stamp &&
	 (twp == NULL || (sysctl_tcp_tw_reuse &&
			 get_seconds() - tcptw->tw_ts_recent_stamp > 1))) {
		tp->write_seq = tcptw->tw_snd_nxt + 65535 + 2
		......
		return 1;
	}
	return 0;	
}

/* And established part... */
	sk_nulls_for_each(sk2, node, &head->chain) {
		if (INET_MATCH(sk2, net, hash, acookie,
					saddr, daddr, ports, dif))
			goto not_unique;
	}

__inet_hash_connect(......)
{
		......
		if (tw) {
			inet_twsk_deschedule(tw, death_row);
			inet_twsk_put(tw);
		}
		......
}

ip_route_newports
	|->ip_route_output_flow
			|->__ip_route_output_key
				|->ip_route_output_slow
					|->fib_lookup

tcp_connect_init
	|->tcp_select_initial_window
int tcp_select_initial_window(...)
{
	......
	(*rcv_wscale) = 0;
	if (wscale_ok) {
		/* Set window scaling on max possible window
		 * See RFC1323 for an explanation of the limit to 14
		 */
		space = max_t(u32, sysctl_tcp_rmem[2], sysctl_rmem_max);
		space = min_t(u32, space, *window_clamp);
		while (space > 65535 && (*rcv_wscale) < 14) {
			space >>= 1;
			(*rcv_wscale)++;
		}
	}
	......
}	

int __inet_stream_connect(struct socket *sock,...,)
{
	// 如果设置了O_NONBLOCK则timeo为0
	timeo = sock_sndtimeo(sk, flags & O_NONBLOCK);
	......
	// 如果timeo=0即O_NONBLOCK会立刻返回
	// 否则等待timeo时间
	if (!timeo || !inet_wait_for_connect(sk, timeo, writebias))
		goto out;
}

tcp_v4_rcv
	|->tcp_v4_do_rcv
		|->tcp_rcv_state_process
			|->tcp_rcv_synsent_state_process
				|->tcp_finish_connect
					|->tcp_init_metrics 初始化度量统计
					|->tcp_init_congestion_control 初始化拥塞控制
					|->tcp_init_buffer_space 初始化buffer空间
					|->inet_csk_reset_keepalive_timer 开启包活定时器
					|->sk_state_change(sock_def_wakeup) 唤醒用户态进程
				|->tcp_send_ack 发送三次握手的最后一次握手给Server端
			|->tcp_set_state(sk, TCP_ESTABLISHED) 设置为ESTABLISHED状态