tcp_usrreq.c

来自「eCos操作系统源码」· C语言 代码 · 共 1,132 行 · 第 1/2 页

			 * initialize maxseg/maxopd using peer's cached
			 * MSS.
			 */
#ifdef INET6
			if (isipv6)
				error = tcp6_connect(tp, nam, p);
			else
#endif /* INET6 */
			error = tcp_connect(tp, nam, p);
			if (error)
				goto out;
			tp->snd_wnd = TTCP_CLIENT_SND_WND;
			tcp_mss(tp, -1);
		}

		if (flags & PRUS_EOF) {
			/*
			 * Close the send side of the connection after
			 * the data is sent.
			 */
			socantsendmore(so);
			tp = tcp_usrclosed(tp);
		}
		if (tp != NULL) {
			if (flags & PRUS_MORETOCOME)
				tp->t_flags |= TF_MORETOCOME;
			error = tcp_output(tp);
			if (flags & PRUS_MORETOCOME)
				tp->t_flags &= ~TF_MORETOCOME;
		}
	} else {
		if (sbspace(&so->so_snd) < -512) {
			m_freem(m);
			error = ENOBUFS;
			goto out;
		}
		/*
		 * According to RFC961 (Assigned Protocols),
		 * the urgent pointer points to the last octet
		 * of urgent data.  We continue, however,
		 * to consider it to indicate the first octet
		 * of data past the urgent section.
		 * Otherwise, snd_up should be one lower.
		 */
		sbappend(&so->so_snd, m);
		if (nam && tp->t_state < TCPS_SYN_SENT) {
			/*
			 * Do implied connect if not yet connected,
			 * initialize window to default value, and
			 * initialize maxseg/maxopd using peer's cached
			 * MSS.
			 */
#ifdef INET6
			if (isipv6)
				error = tcp6_connect(tp, nam, p);
			else
#endif /* INET6 */
			error = tcp_connect(tp, nam, p);
			if (error)
				goto out;
			tp->snd_wnd = TTCP_CLIENT_SND_WND;
			tcp_mss(tp, -1);
		}
		tp->snd_up = tp->snd_una + so->so_snd.sb_cc;
		tp->t_force = 1;
		error = tcp_output(tp);
		tp->t_force = 0;
	}
	COMMON_END((flags & PRUS_OOB) ? PRU_SENDOOB : 
		   ((flags & PRUS_EOF) ? PRU_SEND_EOF : PRU_SEND));
}

/*
 * Abort the TCP.
 */
static int
tcp_usr_abort(struct socket *so)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;

	COMMON_START();
	tp = tcp_drop(tp, ECONNABORTED);
	COMMON_END(PRU_ABORT);
}

/*
 * Receive out-of-band data.
 */
static int
tcp_usr_rcvoob(struct socket *so, struct mbuf *m, int flags)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;

	COMMON_START();
	if ((so->so_oobmark == 0 &&
	     (so->so_state & SS_RCVATMARK) == 0) ||
	    so->so_options & SO_OOBINLINE ||
	    tp->t_oobflags & TCPOOB_HADDATA) {
		error = EINVAL;
		goto out;
	}
	if ((tp->t_oobflags & TCPOOB_HAVEDATA) == 0) {
		error = EWOULDBLOCK;
		goto out;
	}
	m->m_len = 1;
	*mtod(m, caddr_t) = tp->t_iobc;
	if ((flags & MSG_PEEK) == 0)
		tp->t_oobflags ^= (TCPOOB_HAVEDATA | TCPOOB_HADDATA);
	COMMON_END(PRU_RCVOOB);
}

/* xxx - should be const */
struct pr_usrreqs tcp_usrreqs = {
	tcp_usr_abort, tcp_usr_accept, tcp_usr_attach, tcp_usr_bind,
	tcp_usr_connect, pru_connect2_notsupp, in_control, tcp_usr_detach,
	tcp_usr_disconnect, tcp_usr_listen, in_setpeeraddr, tcp_usr_rcvd,
	tcp_usr_rcvoob, tcp_usr_send, pru_sense_null, tcp_usr_shutdown,
	in_setsockaddr, sosend, soreceive, sopoll
};

#ifdef INET6
struct pr_usrreqs tcp6_usrreqs = {
	tcp_usr_abort, tcp6_usr_accept, tcp_usr_attach, tcp6_usr_bind,
	tcp6_usr_connect, pru_connect2_notsupp, in6_control, tcp_usr_detach,
	tcp_usr_disconnect, tcp6_usr_listen, in6_mapped_peeraddr, tcp_usr_rcvd,
	tcp_usr_rcvoob, tcp_usr_send, pru_sense_null, tcp_usr_shutdown,
	in6_mapped_sockaddr, sosend, soreceive, sopoll
};
#endif /* INET6 */

/*
 * Common subroutine to open a TCP connection to remote host specified
 * by struct sockaddr_in in mbuf *nam.  Call in_pcbbind to assign a local
 * port number if needed.  Call in_pcbladdr to do the routing and to choose
 * a local host address (interface).  If there is an existing incarnation
 * of the same connection in TIME-WAIT state and if the remote host was
 * sending CC options and if the connection duration was < MSL, then
 * truncate the previous TIME-WAIT state and proceed.
 * Initialize connection parameters and enter SYN-SENT state.
 */
static int
tcp_connect(tp, nam, p)
	register struct tcpcb *tp;
	struct sockaddr *nam;
	struct proc *p;
{
	struct inpcb *inp = tp->t_inpcb, *oinp;
	struct socket *so = inp->inp_socket;
	struct tcpcb *otp;
	struct sockaddr_in *sin = (struct sockaddr_in *)nam;
	struct sockaddr_in *ifaddr;
	struct rmxp_tao *taop;
	struct rmxp_tao tao_noncached;
	int error;

	if (inp->inp_lport == 0) {
		error = in_pcbbind(inp, (struct sockaddr *)0, p);
		if (error)
			return error;
	}

	/*
	 * Cannot simply call in_pcbconnect, because there might be an
	 * earlier incarnation of this same connection still in
	 * TIME_WAIT state, creating an ADDRINUSE error.
	 */
	error = in_pcbladdr(inp, nam, &ifaddr);
	if (error)
		return error;
	oinp = in_pcblookup_hash(inp->inp_pcbinfo,
	    sin->sin_addr, sin->sin_port,
	    inp->inp_laddr.s_addr != INADDR_ANY ? inp->inp_laddr
						: ifaddr->sin_addr,
	    inp->inp_lport,  0, NULL);
	if (oinp) {
		if (oinp != inp && (otp = intotcpcb(oinp)) != NULL &&
		otp->t_state == TCPS_TIME_WAIT &&
		    (ticks - otp->t_starttime) < tcp_msl &&
		    (otp->t_flags & TF_RCVD_CC))
			otp = tcp_close(otp);
		else
			return EADDRINUSE;
	}
	if (inp->inp_laddr.s_addr == INADDR_ANY)
		inp->inp_laddr = ifaddr->sin_addr;
	inp->inp_faddr = sin->sin_addr;
	inp->inp_fport = sin->sin_port;
	in_pcbrehash(inp);

	/* Compute window scaling to request.  */
	while (tp->request_r_scale < TCP_MAX_WINSHIFT &&
	    (TCP_MAXWIN << tp->request_r_scale) < so->so_rcv.sb_hiwat)
		tp->request_r_scale++;

	soisconnecting(so);
	tcpstat.tcps_connattempt++;
	tp->t_state = TCPS_SYN_SENT;
	callout_reset(tp->tt_keep, tcp_keepinit, tcp_timer_keep, tp);
	tp->iss = tcp_new_isn(tp);
	tcp_sendseqinit(tp);

	/*
	 * Generate a CC value for this connection and
	 * check whether CC or CCnew should be used.
	 */
	if ((taop = tcp_gettaocache(tp->t_inpcb)) == NULL) {
		taop = &tao_noncached;
		bzero(taop, sizeof(*taop));
	}

	tp->cc_send = CC_INC(tcp_ccgen);
	if (taop->tao_ccsent != 0 &&
	    CC_GEQ(tp->cc_send, taop->tao_ccsent)) {
		taop->tao_ccsent = tp->cc_send;
	} else {
		taop->tao_ccsent = 0;
		tp->t_flags |= TF_SENDCCNEW;
	}

	return 0;
}

#ifdef INET6
static int
tcp6_connect(tp, nam, p)
	register struct tcpcb *tp;
	struct sockaddr *nam;
	struct proc *p;
{
	struct inpcb *inp = tp->t_inpcb, *oinp;
	struct socket *so = inp->inp_socket;
	struct tcpcb *otp;
	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)nam;
	struct in6_addr *addr6;
	struct rmxp_tao *taop;
	struct rmxp_tao tao_noncached;
	int error;

	if (inp->inp_lport == 0) {
		error = in6_pcbbind(inp, (struct sockaddr *)0, p);
		if (error)
			return error;
	}

	/*
	 * Cannot simply call in_pcbconnect, because there might be an
	 * earlier incarnation of this same connection still in
	 * TIME_WAIT state, creating an ADDRINUSE error.
	 */
	error = in6_pcbladdr(inp, nam, &addr6);
	if (error)
		return error;
	oinp = in6_pcblookup_hash(inp->inp_pcbinfo,
				  &sin6->sin6_addr, sin6->sin6_port,
				  IN6_IS_ADDR_UNSPECIFIED(&inp->in6p_laddr)
				  ? addr6
				  : &inp->in6p_laddr,
				  inp->inp_lport,  0, NULL);
	if (oinp) {
		if (oinp != inp && (otp = intotcpcb(oinp)) != NULL &&
		    otp->t_state == TCPS_TIME_WAIT &&
		    (ticks - otp->t_starttime) < tcp_msl &&
		    (otp->t_flags & TF_RCVD_CC))
			otp = tcp_close(otp);
		else
			return EADDRINUSE;
	}
	if (IN6_IS_ADDR_UNSPECIFIED(&inp->in6p_laddr))
		inp->in6p_laddr = *addr6;
	inp->in6p_faddr = sin6->sin6_addr;
	inp->inp_fport = sin6->sin6_port;
	if ((sin6->sin6_flowinfo & IPV6_FLOWINFO_MASK) != (u_int32_t)NULL)
		inp->in6p_flowinfo = sin6->sin6_flowinfo;
	in_pcbrehash(inp);

	/* Compute window scaling to request.  */
	while (tp->request_r_scale < TCP_MAX_WINSHIFT &&
	    (TCP_MAXWIN << tp->request_r_scale) < so->so_rcv.sb_hiwat)
		tp->request_r_scale++;

	soisconnecting(so);
	tcpstat.tcps_connattempt++;
	tp->t_state = TCPS_SYN_SENT;
	callout_reset(tp->tt_keep, tcp_keepinit, tcp_timer_keep, tp);
	tp->iss = tcp_new_isn(tp);
	tcp_sendseqinit(tp);

	/*
	 * Generate a CC value for this connection and
	 * check whether CC or CCnew should be used.
	 */
	if ((taop = tcp_gettaocache(tp->t_inpcb)) == NULL) {
		taop = &tao_noncached;
		bzero(taop, sizeof(*taop));
	}

	tp->cc_send = CC_INC(tcp_ccgen);
	if (taop->tao_ccsent != 0 &&
	    CC_GEQ(tp->cc_send, taop->tao_ccsent)) {
		taop->tao_ccsent = tp->cc_send;
	} else {
		taop->tao_ccsent = 0;
		tp->t_flags |= TF_SENDCCNEW;
	}

	return 0;
}
#endif /* INET6 */

/*
 * The new sockopt interface makes it possible for us to block in the
 * copyin/out step (if we take a page fault).  Taking a page fault at
 * splnet() is probably a Bad Thing.  (Since sockets and pcbs both now
 * use TSM, there probably isn't any need for this function to run at
 * splnet() any more.  This needs more examination.)
 */
int
tcp_ctloutput(so, sopt)
	struct socket *so;
	struct sockopt *sopt;
{
	int	error, opt, optval, s;
	struct	inpcb *inp;
	struct	tcpcb *tp;

	error = 0;
	s = splnet();		/* XXX */
	inp = sotoinpcb(so);
	if (inp == NULL) {
		splx(s);
		return (ECONNRESET);
	}
	if (sopt->sopt_level != IPPROTO_TCP) {
#ifdef INET6
		if (INP_CHECK_SOCKAF(so, AF_INET6))
			error = ip6_ctloutput(so, sopt);
		else
#endif /* INET6 */
		error = ip_ctloutput(so, sopt);
		splx(s);
		return (error);
	}
	tp = intotcpcb(inp);

	switch (sopt->sopt_dir) {
	case SOPT_SET:
		switch (sopt->sopt_name) {
		case TCP_NODELAY:
		case TCP_NOOPT:
		case TCP_NOPUSH:
			error = sooptcopyin(sopt, &optval, sizeof optval,
					    sizeof optval);
			if (error)
				break;

			switch (sopt->sopt_name) {
			case TCP_NODELAY:
				opt = TF_NODELAY;
				break;
			case TCP_NOOPT:
				opt = TF_NOOPT;
				break;
			case TCP_NOPUSH:
				opt = TF_NOPUSH;
				break;
			default:
				opt = 0; /* dead code to fool gcc */
				break;
			}

			if (optval)
				tp->t_flags |= opt;
			else
				tp->t_flags &= ~opt;
			break;

		case TCP_MAXSEG:
			error = sooptcopyin(sopt, &optval, sizeof optval,
					    sizeof optval);
			if (error)
				break;

			if (optval > 0 && optval <= tp->t_maxseg)
				tp->t_maxseg = optval;
			else
				error = EINVAL;
			break;

		default:
			error = ENOPROTOOPT;
			break;
		}
		break;

	case SOPT_GET:
		switch (sopt->sopt_name) {
		case TCP_NODELAY:
			optval = tp->t_flags & TF_NODELAY;
			break;
		case TCP_MAXSEG:
			optval = tp->t_maxseg;
			break;
		case TCP_NOOPT:
			optval = tp->t_flags & TF_NOOPT;
			break;
		case TCP_NOPUSH:
			optval = tp->t_flags & TF_NOPUSH;
			break;
		default:
			error = ENOPROTOOPT;
			break;
		}
		if (error == 0)
			error = sooptcopyout(sopt, &optval, sizeof optval);
		break;
	}
	splx(s);
	return (error);
}

/*
 * tcp_sendspace and tcp_recvspace are the default send and receive window
 * sizes, respectively.  These are obsolescent (this information should
 * be set by the route).
 */
u_long	tcp_sendspace = 1024*16;
SYSCTL_INT(_net_inet_tcp, TCPCTL_SENDSPACE, sendspace, CTLFLAG_RW, 
    &tcp_sendspace , 0, "Maximum outgoing TCP datagram size");
u_long	tcp_recvspace = 1024*16;
SYSCTL_INT(_net_inet_tcp, TCPCTL_RECVSPACE, recvspace, CTLFLAG_RW, 
    &tcp_recvspace , 0, "Maximum incoming TCP datagram size");

/*
 * Attach TCP protocol to socket, allocating
 * internet protocol control block, tcp control block,
 * bufer space, and entering LISTEN state if to accept connections.
 */
static int
tcp_attach(so, p)
	struct socket *so;
	struct proc *p;
{
	register struct tcpcb *tp;
	struct inpcb *inp;
	int error;
#ifdef INET6
	int isipv6 = INP_CHECK_SOCKAF(so, AF_INET6) != (int)NULL;
#endif

	if (so->so_snd.sb_hiwat == 0 || so->so_rcv.sb_hiwat == 0) {
		error = soreserve(so, tcp_sendspace, tcp_recvspace);
		if (error)
			return (error);
	}
	error = in_pcballoc(so, &tcbinfo, p);
	if (error)
		return (error);
	inp = sotoinpcb(so);
#ifdef INET6
	if (isipv6) {
		inp->inp_vflag |= INP_IPV6;
		inp->in6p_hops = -1;	/* use kernel default */
	}
	else
#endif
	inp->inp_vflag |= INP_IPV4;
	tp = tcp_newtcpcb(inp);
	if (tp == 0) {
		int nofd = so->so_state & SS_NOFDREF;	/* XXX */

		so->so_state &= ~SS_NOFDREF;	/* don't free the socket yet */
#ifdef INET6
		if (isipv6)
			in6_pcbdetach(inp);
		else
#endif
		in_pcbdetach(inp);
		so->so_state |= nofd;
		return (ENOBUFS);
	}
	tp->t_state = TCPS_CLOSED;
	return (0);
}

/*
 * Initiate (or continue) disconnect.
 * If embryonic state, just send reset (once).
 * If in ``let data drain'' option and linger null, just drop.
 * Otherwise (hard), mark socket disconnecting and drop
 * current input data; switch states based on user close, and
 * send segment to peer (with FIN).
 */
static struct tcpcb *
tcp_disconnect(tp)
	register struct tcpcb *tp;
{
	struct socket *so = tp->t_inpcb->inp_socket;

	if (tp->t_state < TCPS_ESTABLISHED)
		tp = tcp_close(tp);
	else if ((so->so_options & SO_LINGER) && so->so_linger == 0)
		tp = tcp_drop(tp, 0);
	else {
		soisdisconnecting(so);
		sbflush(&so->so_rcv);
		tp = tcp_usrclosed(tp);
		if (tp)
			(void) tcp_output(tp);
	}
	return (tp);
}

/*
 * User issued close, and wish to trail through shutdown states:
 * if never received SYN, just forget it.  If got a SYN from peer,
 * but haven't sent FIN, then go to FIN_WAIT_1 state to send peer a FIN.
 * If already got a FIN from peer, then almost done; go to LAST_ACK
 * state.  In all other cases, have already sent FIN to peer (e.g.
 * after PRU_SHUTDOWN), and just have to play tedious game waiting
 * for peer to send FIN or not respond to keep-alives, etc.
 * We can let the user exit from the close as soon as the FIN is acked.
 */
static struct tcpcb *
tcp_usrclosed(tp)
	register struct tcpcb *tp;
{

	switch (tp->t_state) {

	case TCPS_CLOSED:
	case TCPS_LISTEN:
		tp->t_state = TCPS_CLOSED;
		tp = tcp_close(tp);
		break;

	case TCPS_SYN_SENT:
	case TCPS_SYN_RECEIVED:
		tp->t_flags |= TF_NEEDFIN;
		break;

	case TCPS_ESTABLISHED:
		tp->t_state = TCPS_FIN_WAIT_1;
		break;

	case TCPS_CLOSE_WAIT:
		tp->t_state = TCPS_LAST_ACK;
		break;
	}
	if (tp && tp->t_state >= TCPS_FIN_WAIT_2) {
		soisdisconnected(tp->t_inpcb->inp_socket);
		/* To prevent the connection hanging in FIN_WAIT_2 forever. */
		if (tp->t_state == TCPS_FIN_WAIT_2)
			callout_reset(tp->tt_2msl, tcp_maxidle,
				      tcp_timer_2msl, tp);
	}
	return (tp);
}