udp的connect && send

上一篇文章疑惑地方是：udp bind 同一个port 是否可行？以及如何查找sk 管理skbbuff

此时疑惑的是：udp没有bind 一个ip：port 去connect 会发生什么，connect 后发出send 调用呢?此过程中涉及到的ip port 如何填充呢？ 如果socket没有connect也没有bind，那sendto/send会发生什么？

 

UDP的connect系统调用

　　udp虽然不是基于链接的协议，但是内核也提供了connect 系统调用，通过connect传输控制块记录的目的地址和目的端口，并且根据这个目的ip-port 选择目的路由，怎样发送的时候就可以通过write send发送，而不需要使用sendto，同时接收数据的时候也不必调用recvfrom获取数据报文发送者，而可以用read recv recvmsg。

udp的connect核心函数为：

/*
 *    Automatically bind an unbound socket.
 */

static int inet_autobind(struct sock *sk)
{
    struct inet_sock *inet;
    /* We may need to bind the socket. */
    lock_sock(sk);
    inet = inet_sk(sk);
    if (!inet->inet_num) {
        if (sk->sk_prot->get_port(sk, 0)) {
            release_sock(sk);
            return -EAGAIN;
        }
        inet->inet_sport = htons(inet->inet_num);
    }
    release_sock(sk);
    return 0;
}

int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr,
               int addr_len, int flags)
{
    struct sock *sk = sock->sk;

    if (addr_len < sizeof(uaddr->sa_family))
        return -EINVAL;
    if (uaddr->sa_family == AF_UNSPEC)
        return sk->sk_prot->disconnect(sk, flags);

    if (!inet_sk(sk)->inet_num && inet_autobind(sk))
        return -EAGAIN;
    return sk->sk_prot->connect(sk, uaddr, addr_len);
}

可以看到：udp connect的时候如果没有bind 一个port ， 就回自动bind 一个port

 最后会调用一个协议相关的connect ：ip4_datagram_connect

int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
    int res;

    lock_sock(sk);
    res = __ip4_datagram_connect(sk, uaddr, addr_len);
    release_sock(sk);
    return res;
}

 

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;

    sk_dst_reset(sk);//复位路由高速缓冲区的入口地址

    oif = sk->sk_bound_dev_if;
    saddr = inet->inet_saddr;
    //如果建立连接的目的地址是组播地址，如果源ip 以及出口接口没有设置则重新设置
    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;
    //查找出口路由
    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);
    if (IS_ERR(rt)) {
        err = PTR_ERR(rt);
        if (err == -ENETUNREACH)
            IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
        goto out;
    }

    if ((rt->rt_flags & RTCF_BROADCAST) && !sock_flag(sk, SOCK_BROADCAST)) {
        ip_rt_put(rt);
        err = -EACCES;
        goto out;
    }
    if (!inet->inet_saddr)// 如果 socket 没有bind ip 则设置路由出口的src ip
        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;//可能是组播原因
    inet->inet_dport = usin->sin_port;
    sk->sk_state = TCP_ESTABLISHED;
    sk_set_txhash(sk);
    inet->inet_id = jiffies;

    sk_dst_set(sk, &rt->dst);
    err = 0;
out:
    return err;
}

 

从上述代码可一看到，udp socket 即使没有bind ip port，在connect的时候根据路由结果，会去主动填充src ip 以及src port 同时同时要注意组播的处理方式；

UDP  套接字的send调用

由于对同一个socket 可以连续发起send调用发送到不同目的ip port的报文，所以除非是connect后的socket，每次send 应该是需要从新路由获取出口信息以及源ip port 信息

 

int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
{
    /*套接字的网络层表示          ------ 转换成INET套接字的表示*/
    struct inet_sock *inet = inet_sk(sk);
    struct udp_sock *up = udp_sk(sk);
    struct flowi4 fl4_stack;
    struct flowi4 *fl4;
    int ulen = len;
    struct ipcm_cookie ipc;
    struct rtable *rt = NULL;
    int free = 0;
    int connected = 0;
    __be32 daddr, faddr, saddr;
    __be16 dport;
    u8  tos;
    int err, is_udplite = IS_UDPLITE(sk);
    int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
    int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
    struct sk_buff *skb;
    struct ip_options_data opt_copy;

    if (len > 0xFFFF)
        return -EMSGSIZE;

    /*
     *    Check the flags.
     */

    if (msg->msg_flags & MSG_OOB) /* Mirror BSD error message compatibility */
        return -EOPNOTSUPP;

    ipc.opt = NULL;
    ipc.tx_flags = 0;
    ipc.ttl = 0;
    ipc.tos = -1;

    getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag;

    fl4 = &inet->cork.fl.u.ip4;
    if (up->pending) {
        /*
         * There are pending frames.
         * The socket lock must be held while it's corked.
         */ /*当前的sock有等待发送的数据，直接将数据追加？？    
         注意 ：应用层发送数据时 自家来加锁控制乱序 ， 同时接收端也应该自己控制乱序*/
        lock_sock(sk);
        if (likely(up->pending)) {
            if (unlikely(up->pending != AF_INET)) {
                release_sock(sk);
                return -EINVAL;
            }
            goto do_append_data;
        }
        release_sock(sk);
    }
    ulen += sizeof(struct udphdr);

    /*
     *    Get and verify the address.
     */
    if (msg->msg_name) {
        DECLARE_SOCKADDR(struct sockaddr_in *, usin, msg->msg_name);
        if (msg->msg_namelen < sizeof(*usin))
            return -EINVAL;
        if (usin->sin_family != AF_INET) {
            if (usin->sin_family != AF_UNSPEC)
                return -EAFNOSUPPORT;
        }
        // 获取目的ip地址和端口
        daddr = usin->sin_addr.s_addr;
        dport = usin->sin_port;
        if (dport == 0)
            return -EINVAL;
    } else {// 如果参数中没有目的ip port 则 检测是否为 connect过的socket
        if (sk->sk_state != TCP_ESTABLISHED)
            return -EDESTADDRREQ;
        daddr = inet->inet_daddr;
        dport = inet->inet_dport;
        /* Open fast path for connected socket.
           Route will not be used, if at least one option is set.
         */
        connected = 1;
    }
    /*获取存储在 socket 上的源地址、发送网络设备索引（device index）和时间戳选项*/
    ipc.addr = inet->inet_saddr;

    ipc.oif = sk->sk_bound_dev_if;

    sock_tx_timestamp(sk, &ipc.tx_flags);

    if (msg->msg_controllen) {/*msg中控制信息处理*/
        err = ip_cmsg_send(sock_net(sk), msg, &ipc,
                   sk->sk_family == AF_INET6);
        if (unlikely(err)) {
            kfree(ipc.opt);
            return err;
        }
        if (ipc.opt)
            free = 1;
        connected = 0;
    }
    if (!ipc.opt) {/*如果发送的数据中没有IP选项控制信息，则从正在使用的socket中获取IP选项信息*/
        struct ip_options_rcu *inet_opt;

        rcu_read_lock();
        inet_opt = rcu_dereference(inet->inet_opt);
        if (inet_opt) {
            memcpy(&opt_copy, inet_opt,
                   sizeof(*inet_opt) + inet_opt->opt.optlen);
            ipc.opt = &opt_copy.opt;
        }
        rcu_read_unlock();
    }

    saddr = ipc.addr;
    ipc.addr = faddr = daddr;

    if (ipc.opt && ipc.opt->opt.srr) {
        if (!daddr)
            return -EINVAL;
        // 如果存在严格ip 选项， 则不能根据目的ip 来选路， 需要根据opt 中的下一站ip 地址来选路，
        faddr = ipc.opt->opt.faddr;
        connected = 0;/*重新进行本地路由*/
    }
    tos = get_rttos(&ipc, inet);
    if (sock_flag(sk, SOCK_LOCALROUTE) ||
        (msg->msg_flags & MSG_DONTROUTE) ||
        (ipc.opt && ipc.opt->opt.is_strictroute)) {
        tos |= RTO_ONLINK;
        /*如果设置了 dontroute 或者ip 中存在严格源选路 
        则说明目的地址或者下一条必定位于本地子网*/ 
        connected = 0;/*只进行本地路由*/
    }

    if (ipv4_is_multicast(daddr)) {
        if (!ipc.oif)
            ipc.oif = inet->mc_index;
        if (!saddr)
            saddr = inet->mc_addr;
        connected = 0;/*只进行本地路由*/
    } else if (!ipc.oif)
        ipc.oif = inet->uc_index;

    if (connected)/*目标路由检查*/
        rt = (struct rtable *)sk_dst_check(sk, 0);

    if (!rt) {//如果没有路由，建立一次路由
        struct net *net = sock_net(sk);
        __u8 flow_flags = inet_sk_flowi_flags(sk);

        fl4 = &fl4_stack;

        flowi4_init_output(fl4, ipc.oif, sk->sk_mark, tos,
                   RT_SCOPE_UNIVERSE, sk->sk_protocol,
                   flow_flags,
                   faddr, saddr, dport, inet->inet_sport);

        if (!saddr && ipc.oif) {
            err = l3mdev_get_saddr(net, ipc.oif, fl4);
            if (err < 0)
                goto out;
        }

        security_sk_classify_flow(sk, flowi4_to_flowi(fl4));
        rt = ip_route_output_flow(net, fl4, sk);
        if (IS_ERR(rt)) {
            err = PTR_ERR(rt);
            rt = NULL;
            if (err == -ENETUNREACH)
                IP_INC_STATS(net, IPSTATS_MIB_OUTNOROUTES);
            goto out;
        }

        err = -EACCES;
        if ((rt->rt_flags & RTCF_BROADCAST) &&
            !sock_flag(sk, SOCK_BROADCAST))
            goto out;
        if (connected)
            sk_dst_set(sk, dst_clone(&rt->dst));
    }

    if (msg->msg_flags&MSG_CONFIRM)
        goto do_confirm;// 应用层知道网关有效可达， 跳转到do_confirm 确认路由缓存
back_from_confirm:

    saddr = fl4->saddr;
    if (!ipc.addr)
        daddr = ipc.addr = fl4->daddr;

    /* Lockless fast path for the non-corking case. */
    if (!corkreq) {
        skb = ip_make_skb(sk, fl4, getfrag, msg, ulen,
                  sizeof(struct udphdr), &ipc, &rt,
                  msg->msg_flags);
        err = PTR_ERR(skb);
        if (!IS_ERR_OR_NULL(skb))
            err = udp_send_skb(skb, fl4);
        goto out;
    }

    lock_sock(sk);
    if (unlikely(up->pending)) {
        /* The socket is already corked while preparing it. */
        /* ... which is an evident application bug. --ANK */
        release_sock(sk);

        net_dbg_ratelimited("cork app bug 2\n");
        err = -EINVAL;
        goto out;
    }
    /*
     *    Now cork the socket to pend data.
     更新缓存 目的地址 目的端口 源ip port 信息 设置pending标志，表示正在处理数据
     */
    fl4 = &inet->cork.fl.u.ip4;
    fl4->daddr = daddr;
    fl4->saddr = saddr;
    fl4->fl4_dport = dport;
    fl4->fl4_sport = inet->inet_sport;
    up->pending = AF_INET;
/*
1、首先获取到目标地址和端口，原地址和端口，目的端口不能为0,
目标地址为广播地址或者开头为224操作不同。
2、.虽然是面向无连接的，但是需要起始点到目的地之间的链路，发送数据必须有一条路由缓存。
如果目的地址改变，则重新建立一条路由并获取缓存
3、发包中没有包含校验和的操作，只含有数据拼接和发送数据前的准备工作
    */
do_append_data:
    up->len += ulen;/*udp组ip包*/
    err = ip_append_data(sk, fl4, getfrag, msg, ulen,
                 sizeof(struct udphdr), &ipc, &rt,
                 corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
    if (err)
        udp_flush_pending_frames(sk);
    else if (!corkreq)
        err = udp_push_pending_frames(sk);
    else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
        up->pending = 0;
    release_sock(sk);

out:
    ip_rt_put(rt);
    if (free)
        kfree(ipc.opt);
    if (!err)
        return len;
    /*
     * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space.  Reporting
     * ENOBUFS might not be good (it's not tunable per se), but otherwise
     * we don't have a good statistic (IpOutDiscards but it can be too many
     * things).  We could add another new stat but at least for now that
     * seems like overkill.
     */
    if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
        UDP_INC_STATS_USER(sock_net(sk),
                UDP_MIB_SNDBUFERRORS, is_udplite);
    }
    return err;

do_confirm:
    if (msg->msg_flags & MSG_PROBE)
        dst_confirm_neigh(&rt->dst, &fl4->daddr);
    if (!(msg->msg_flags&MSG_PROBE) || len)
        goto back_from_confirm;
    err = 0;
    goto out;
}

 

 问题：

1. 多线程同时向一个socket send pkt， 如果线程1  lock_sock --->sockfd  线程2 被阻塞！！ 
2. 如果线程1 send 后release sockfd， 但是sock 出于pending状态， 线程2 发送则直接add data 到socket ，那就是直接使用socket的inet 信息发送数据，没有更新路由数据了！
3. udp 发送socket 的时候会根据目的ip port 信息来路由，对于源IP port  必须bind 一个吗？ 目前看没必要！因为ip_route_output_flow 会根据目的ip地址查找路由，如果saddr为空则会将路由信息填充进去

 

#include <sys/types.h>
#include <sys/socket.h>
#include<pthread.h>
#include <netinet/in.h>
#include <stdio.h>
#include <string.h>
#include <unistd.h>
#include <stdlib.h>
 
 
int main(int argc, char **argv)
{
    char buff[512];
 
    printf("This is a UDP client\n");
    struct sockaddr_in addr;
    int sock;
 
    if ( (sock=socket(AF_INET, SOCK_DGRAM, 0)) <0)
    {
        perror("socket");
        exit(1);
    }
    addr.sin_family = AF_INET;
    addr.sin_port = htons(3702);
    addr.sin_addr.s_addr = inet_addr("10.10.10.10");
    if (addr.sin_addr.s_addr == INADDR_NONE)
    {
        printf("Incorrect ip address!");
        close(sock);
        exit(1);
    }
 
    while (1)
    {
        /* 接受用户输入 */
        bzero(buff, 512);
        if (fgets(buff, 511, stdin) == (char *) EOF)//从键盘输入内容
            exit(0);
        if( sendto(sock, buff, strlen(buff), 0, (struct sockaddr *)&addr, sizeof(addr)) < 0){
            perror("sendto");
            close(sock);
            break;
        }else{
            printf("clinet send success!\n");
        }
 
    }
    
    return 0;
}

 

#include <sys/types.h>
#include <sys/socket.h>
#include<pthread.h>
#include <netinet/in.h>
#include <stdio.h>
#include <string.h>
#include <unistd.h>
#include <stdlib.h>
 
int main(int argc, char **argv)
{
    printf("Welcome! This is a UDP server, I can only received message from client and reply with same message\n");
    
    struct sockaddr_in addr;
    addr.sin_family = AF_INET;
    addr.sin_port = htons(3702);
    addr.sin_addr.s_addr = htonl(INADDR_ANY);
 
    int sock;
    if ( (sock = socket(AF_INET, SOCK_DGRAM, 0)) < 0)
    {
        perror("socket");
        exit(1);
    }
    if (bind(sock, (struct sockaddr *)&addr, sizeof(addr)) < 0)
    {
        perror("bind");
        exit(1);
    }
    char buff[512];
    struct sockaddr_in clientAddr;
    int n;
    int len = sizeof(clientAddr);
    while (1)
    {
        n = recvfrom(sock, buff, 511, 0, (struct sockaddr*)&clientAddr, &len);
        if (n>0)
        {
            buff[n] = 0;
            printf("%s %u says: %s\n", inet_ntoa(clientAddr.sin_addr), ntohs(clientAddr.sin_port), buff);
            n = sendto(sock, buff, n, 0, (struct sockaddr *)&clientAddr, sizeof(clientAddr));
            if (n < 0)
            {
                perror("sendto");
                break;
            }
        }
        else
        {
            perror("recv");
            break;
        }
    }
    return 0;
}

上述有client 和server 例子 直接测试使用

关于详细ip_append_data /udp_push_pending_frames  发送数据，后续再看