-
connect 系统调用 分析
#include <sys/types.h> /* See NOTES */
#include <sys/socket.h>
int connect(int sockfd, const struct sockaddr *addr, socklen_t addrlen);
其中的参数解释如下:
·int sockfd :套接字描述符。
·const struct sockaddr*addr :要连接的地址。
·socklen_t addrlen :要连接的地址长度。
返回值 0 表示成功, -1 表示失败。
connect 的用途是使用指定的套接字去连接指定的地址。对于面向连接的协议(套接字类型为
SOCK_STREAM ), connect 只能成功一次(当然要如此,因为真正的连接已经建立了)。如果重复调
用 connect ,会返回 -1 表示失败,同时错误码为 EISCONN 。而对于非面向连接的协议(套接字类型为
SOCK_DGRAM ),则可以执行多次 connect (因为这时的 connect 仅仅是设置了默认的目的地址)。
对于 TCP 套接字来说, connect 实际上是要真正地进行三次握手,所以其默认是一个阻塞操作。那么
是否可以写一个非阻塞的 TCP connect 代码呢?
/* * Attempt to connect to a socket with the server address. The address * is in user space so we verify it is OK and move it to kernel space. * * For 1003.1g we need to add clean support for a bind to AF_UNSPEC to * break bindings * * NOTE: 1003.1g draft 6.3 is broken with respect to AX.25/NetROM and * other SEQPACKET protocols that take time to connect() as it doesn't * include the -EINPROGRESS status for such sockets. */ SYSCALL_DEFINE3(connect, int, fd, struct sockaddr __user *, uservaddr, int, addrlen) { struct socket *sock; struct sockaddr_storage address; int err, fput_needed; /* 通过文件描述符fd,找到对应的socket实例。 * 以fd为索引从当前进程的文件描述符表files_struct实例中找到对应的file实例, * 然后从file实例的private_data成员中获取socket实例。 */ sock = sockfd_lookup_light(fd, &err, &fput_needed); if (!sock) goto out; /* 把套接字地址从用户空间拷贝到内核空间 */ err = move_addr_to_kernel(uservaddr, addrlen, &address); if (err < 0) goto out_put; err = security_socket_connect(sock, (struct sockaddr *)&address, addrlen); if (err) goto out_put; /* 调用Socket层的操作函数,如果是SOCK_STREAM,则proto_ops为inet_stream_ops, * 函数指针指向inet_stream_connect()。 */ err = sock->ops->connect(sock, (struct sockaddr *)&address, addrlen, sock->file->f_flags); out_put: fput_light(sock->file, fput_needed); out: return err; }
int inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags) { int err; lock_sock(sock->sk);//进入互斥区 err = __inet_stream_connect(sock, uaddr, addr_len, flags); release_sock(sock->sk); return err; }
/* * Connect to a remote host. There is regrettably still a little * TCP 'magic' in here. */ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; int err; long timeo; /* 长度合法性检查*/ if (addr_len < sizeof(uaddr->sa_family)) return -EINVAL; if (uaddr->sa_family == AF_UNSPEC) {/* 如果协议族为 AF_UNSPEC ,则先执行*/ err = sk->sk_prot->disconnect(sk, flags); /* 根据是否成功断开连接,来设置socket状态 */ sock->state = err ? SS_DISCONNECTING : SS_UNCONNECTED; goto out; } switch (sock->state) { default: err = -EINVAL; goto out; /* 此套接口已经和对端的套接口相连接了,即连接已经建立 */ case SS_CONNECTED: err = -EISCONN;/* Transport endpoint is already connected */ goto out; case SS_CONNECTING:/*连接正在建立中 */ err = -EALREADY;/* Operation already in progress */ /* Fall out of switch with err, set for this state */ break; case SS_UNCONNECTED: err = -EISCONN; if (sk->sk_state != TCP_CLOSE) goto out; /* 如果使用的是TCP,则sk_prot为tcp_prot,connect为tcp_v4_connect() */ err = sk->sk_prot->connect(sk, uaddr, addr_len);/* 发送SYN包 */ if (err < 0) goto out; /* 发出SYN包后socket状态设为正在连接 */ sock->state = SS_CONNECTING; /* Just entered SS_CONNECTING state; the only * difference is that return value in non-blocking * case is EINPROGRESS, rather than EALREADY. */ err = -EINPROGRESS; break; } /* sock的发送超时时间,非阻塞则为0 */ timeo = sock_sndtimeo(sk, flags & O_NONBLOCK); /* 发出SYN包后,等待后续握手的完成 */ if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) { int writebias = (sk->sk_protocol == IPPROTO_TCP) && tcp_sk(sk)->fastopen_req && tcp_sk(sk)->fastopen_req->data ? 1 : 0; /* 如果是非阻塞的,那么就直接返回错误码-EINPROGRESS。 * socket为阻塞时,使用inet_wait_for_connect()来等待协议栈的处理: * 1. 使用SO_SNDTIMEO,睡眠时间超过timeo就返回0,之后返回错误码-EINPROGRESS。 * 2. 收到信号,就返回剩余的等待时间。之后会返回错误码-ERESTARTSYS或-EINTR。 * 3. 三次握手成功,被sock I/O事件处理函数唤醒,之后会返回0。 */ /* Error code is set above */ if (!timeo || !inet_wait_for_connect(sk, timeo, writebias)) goto out; err = sock_intr_errno(timeo); /* 进程收到信号,如果err为-ERESTARTSYS,接下来库函数会重新调用connect() */ if (signal_pending(current)) goto out; } /* Connection was closed by RST, timeout, ICMP error * or another process disconnected us. */ if (sk->sk_state == TCP_CLOSE) goto sock_error; /* sk->sk_err may be not zero now, if RECVERR was ordered by user * and error was received after socket entered established state. * Hence, it is handled normally after connect() return successfully. */ /* 更新socket状态为连接已建立 */ sock->state = SS_CONNECTED; err = 0; out: return err; sock_error: err = sock_error(sk) ? : -ECONNABORTED; sock->state = SS_UNCONNECTED; if (sk->sk_prot->disconnect(sk, flags)) sock->state = SS_DISCONNECTING; goto out; } EXPORT_SYMB
static long inet_wait_for_connect(struct sock *sk, long timeo, int writebias) { DEFINE_WAIT(wait); /* 把等待任务加入到socket的等待队列头部,把进程的状态设为TASK_INTERRUPTIBLE */ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); sk->sk_write_pending += writebias; /* Basic assumption: if someone sets sk->sk_err, he _must_ * change state of the socket from TCP_SYN_*. * Connect() does not allow to get error notifications * without closing the socket. */ /* 完成三次握手后,状态就会变为TCP_ESTABLISHED,从而退出循环 */ while ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) { release_sock(sk); /* 进入睡眠,直到超时或收到信号,或者被I/O事件处理函数唤醒。 * 1. 如果是收到信号退出的,timeo为剩余的jiffies。 * 2. 如果使用了SO_SNDTIMEO选项,超时退出后,timeo为0。 * 3. 如果没有使用SO_SNDTIMEO选项,timeo为无穷大,即MAX_SCHEDULE_TIMEOUT, * 那么返回值也是这个,而超时时间不定。为了无限阻塞,需要上面的while循环。 */ timeo = schedule_timeout(timeo); lock_sock(sk); /* 如果进程有待处理的信号,或者睡眠超时了,退出循环,之后会返回错误码 */ if (signal_pending(current) || !timeo) break; prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); } /* 等待结束时,把等待进程从等待队列中删除,把当前进程的状态设为TASK_RUNNING */ finish_wait(sk_sleep(sk), &wait); sk->sk_write_pending -= writebias; return timeo; } /**/ 进程的唤醒 三次握手中,当客户端收到SYNACK、发出ACK后,连接就成功建立了。 此时连接的状态从TCP_SYN_SENT或TCP_SYN_RECV变为TCP_ESTABLISHED,sock的状态发生变化, 会调用sock_def_wakeup()来处理连接状态变化事件,唤醒进程,connect()就能成功返回了。 sock_def_wakeup()的函数调用路径如下: tcp_v4_rcv tcp_v4_do_rcv tcp_rcv_state_process tcp_rcv_synsent_state_process tcp_finish_connect sock_def_wakeup wake_up_interruptible_all __wake_up
void tcp_finish_connect(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); tcp_set_state(sk, TCP_ESTABLISHED); ------------------------ ---------------------------- if (!sock_flag(sk, SOCK_DEAD)) { sk->sk_state_change(sk);---->// 指向sock_def_wakeup /* 如果使用了异步通知,则发送SIGIO通知进程可写 */ sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT); } } static inline void sk_wake_async(struct sock *sk, int how, int band) { if (sock_flag(sk, SOCK_FASYNC)) sock_wake_async(sk->sk_socket, how, band); } static void sock_def_wakeup(struct sock *sk) { struct socket_wq *wq; rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); if (wq_has_sleeper(wq)) wake_up_interruptible_all(&wq->wait); rcu_read_unlock(); } //最终调用__wake_up_common(),由于nr_exclusive为0,会把此socket上所有的等待进程都唤醒
udp_prot 是 UDP 协议中所有自定义操作函数的集合。其 connect 的实现函数为 ip4_datagram_connect 。
其主要是设置了目的 IP 、端口和路由信息
int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_in *usin = (struct sockaddr_in *) uaddr; struct flowi4 *fl4; struct rtable *rt; __be32 saddr; int oif; int err; if (addr_len < sizeof(*usin)) return -EINVAL; if (usin->sin_family != AF_INET) return -EAFNOSUPPORT; //复位路由高速缓冲区的入口地址 sk_dst_reset(sk); lock_sock(sk); //和套接字绑定的网络设备索引号 oif = sk->sk_bound_dev_if; saddr = inet->inet_saddr; //如果建立连接的地址是组传送地址,meiyou jiu 重新初始化oif和原地址 if (ipv4_is_multicast(usin->sin_addr.s_addr)) { if (!oif) oif = inet->mc_index; if (!saddr) saddr = inet->mc_addr; } fl4 = &inet->cork.fl.u.ip4; /* 调用ip_route_connet寻找路由, 源路由主要根据源地址、源端口、目的地址、目的端口、输出网络设备额索引号, 如果寻找路由失败就返回错误,如果寻找的路由是广播地址路由就要是否路由在高速 缓冲区的入口并返回错误。寻找路由成功就把套接字的状态变量sk_state设置为TCP_ESTABLISHED, 并把路由保存到套接字的sk->sk_dst_cache数据域 */ rt = ip_route_connect(fl4, usin->sin_addr.s_addr, saddr, RT_CONN_FLAGS(sk), oif, sk->sk_protocol, inet->inet_sport, usin->sin_port, sk, true); if (IS_ERR(rt)) { err = PTR_ERR(rt); if (err == -ENETUNREACH) IP_INC_STATS_BH(sock_net(sk), IPSTATS_MIB_OUTNOROUTES); goto out; } //寻找的路由是广播地址路由,则释放该路由在路由缓冲区的入口 if ((rt->rt_flags & RTCF_BROADCAST) && !sock_flag(sk, SOCK_BROADCAST)) { ip_rt_put(rt); err = -EACCES; goto out; } if (!inet->inet_saddr)//从路由表中获取的信息更新udp的原地址 inet->inet_saddr = fl4->saddr; /* Update source address */ if (!inet->inet_rcv_saddr) { inet->inet_rcv_saddr = fl4->saddr; if (sk->sk_prot->rehash) sk->sk_prot->rehash(sk); }//更新目的地址和目的端口,源端口已经给定了 inet->inet_daddr = fl4->daddr; inet->inet_dport = usin->sin_port; sk->sk_state = TCP_ESTABLISHED; inet->inet_id = jiffies; sk_dst_set(sk, &rt->dst); err = 0; out: release_sock(sk); return err; }