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

//==========================================================================
//
//      src/sys/netinet/tcp_usrreq.c
//
//==========================================================================
//####BSDCOPYRIGHTBEGIN####
//
// -------------------------------------------
//
// Portions of this software may have been derived from OpenBSD, 
// FreeBSD or other sources, and are covered by the appropriate
// copyright disclaimers included herein.
//
// Portions created by Red Hat are
// Copyright (C) 2002 Red Hat, Inc. All Rights Reserved.
//
// -------------------------------------------
//
//####BSDCOPYRIGHTEND####
//==========================================================================

/*
 * Copyright (c) 1982, 1986, 1988, 1993
 *	The Regents of the University of California.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	From: @(#)tcp_usrreq.c	8.2 (Berkeley) 1/3/94
 * $FreeBSD: src/sys/netinet/tcp_usrreq.c,v 1.51.2.9 2001/08/22 00:59:12 silby Exp $
 */

#include <sys/param.h>
#include <sys/sysctl.h>
#include <sys/mbuf.h>
#ifdef INET6
#include <sys/domain.h>
#endif /* INET6 */
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/protosw.h>

#include <net/if.h>
#include <net/route.h>

#include <netinet/in.h>
#include <netinet/in_systm.h>
#ifdef INET6
#include <netinet/ip6.h>
#endif
#include <netinet/in_pcb.h>
#ifdef INET6
#include <netinet6/in6_pcb.h>
#endif
#include <netinet/in_var.h>
#include <netinet/ip_var.h>
#ifdef INET6
#include <netinet6/ip6_var.h>
#endif
#include <netinet/tcp.h>
#include <netinet/tcp_fsm.h>
#include <netinet/tcp_seq.h>
#include <netinet/tcp_timer.h>
#include <netinet/tcp_var.h>
#include <netinet/tcpip.h>
#ifdef TCPDEBUG
#include <netinet/tcp_debug.h>
#endif

#ifdef IPSEC
#include <netinet6/ipsec.h>
#endif /*IPSEC*/

/*
 * TCP protocol interface to socket abstraction.
 */
extern	char *tcpstates[];	/* XXX ??? */

static int	tcp_attach __P((struct socket *, struct proc *));
static int	tcp_connect __P((struct tcpcb *, struct sockaddr *, 
				 struct proc *));
#ifdef INET6
static int	tcp6_connect __P((struct tcpcb *, struct sockaddr *,
				 struct proc *));
#endif /* INET6 */
static struct tcpcb *
		tcp_disconnect __P((struct tcpcb *));
static struct tcpcb *
		tcp_usrclosed __P((struct tcpcb *));

#ifdef TCPDEBUG
#define	TCPDEBUG0	int ostate = 0
#define	TCPDEBUG1()	ostate = tp ? tp->t_state : 0
#define	TCPDEBUG2(req)	if (tp && (so->so_options & SO_DEBUG)) \
				tcp_trace(TA_USER, ostate, tp, 0, 0, req)
#else
#define	TCPDEBUG0
#define	TCPDEBUG1()
#define	TCPDEBUG2(req)
#endif

/*
 * TCP attaches to socket via pru_attach(), reserving space,
 * and an internet control block.
 */
static int
tcp_usr_attach(struct socket *so, int proto, struct proc *p)
{
	int s = splnet();
	int error;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp = 0;
	TCPDEBUG0;

	TCPDEBUG1();
	if (inp) {
		error = EISCONN;
		goto out;
	}

	error = tcp_attach(so, p);
	if (error)
		goto out;

	if ((so->so_options & SO_LINGER) && so->so_linger == 0)
		so->so_linger = TCP_LINGERTIME;
	tp = sototcpcb(so);
out:
	TCPDEBUG2(PRU_ATTACH);
	splx(s);
	return error;
}

/*
 * pru_detach() detaches the TCP protocol from the socket.
 * If the protocol state is non-embryonic, then can't
 * do this directly: have to initiate a pru_disconnect(),
 * which may finish later; embryonic TCB's can just
 * be discarded here.
 */
static int
tcp_usr_detach(struct socket *so)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;
	TCPDEBUG0;

	if (inp == 0) {
		splx(s);
		return EINVAL;	/* XXX */
	}
	tp = intotcpcb(inp);
	TCPDEBUG1();
	tp = tcp_disconnect(tp);

	TCPDEBUG2(PRU_DETACH);
	splx(s);
	return error;
}

#define	COMMON_START()	TCPDEBUG0; \
			do { \
				     if (inp == 0) { \
					     splx(s); \
					     return EINVAL; \
				     } \
				     tp = intotcpcb(inp); \
				     TCPDEBUG1(); \
		     } while(0)
			     
#define COMMON_END(req)	out: TCPDEBUG2(req); splx(s); return error; goto out


/*
 * Give the socket an address.
 */
static int
tcp_usr_bind(struct socket *so, struct sockaddr *nam, struct proc *p)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;
	struct sockaddr_in *sinp;

	COMMON_START();

	/*
	 * Must check for multicast addresses and disallow binding
	 * to them.
	 */
	sinp = (struct sockaddr_in *)nam;
	if (sinp->sin_family == AF_INET &&
	    IN_MULTICAST(ntohl(sinp->sin_addr.s_addr))) {
		error = EAFNOSUPPORT;
		goto out;
	}
	error = in_pcbbind(inp, nam, p);
	if (error)
		goto out;
	COMMON_END(PRU_BIND);

}

#ifdef INET6
static int
tcp6_usr_bind(struct socket *so, struct sockaddr *nam, struct proc *p)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;
	struct sockaddr_in6 *sin6p;

	COMMON_START();

	/*
	 * Must check for multicast addresses and disallow binding
	 * to them.
	 */
	sin6p = (struct sockaddr_in6 *)nam;
	if (sin6p->sin6_family == AF_INET6 &&
	    IN6_IS_ADDR_MULTICAST(&sin6p->sin6_addr)) {
		error = EAFNOSUPPORT;
		goto out;
	}
	inp->inp_vflag &= ~INP_IPV4;
	inp->inp_vflag |= INP_IPV6;
	if (ip6_mapped_addr_on && (inp->inp_flags & IN6P_IPV6_V6ONLY) == 0) {
		if (IN6_IS_ADDR_UNSPECIFIED(&sin6p->sin6_addr))
			inp->inp_vflag |= INP_IPV4;
		else if (IN6_IS_ADDR_V4MAPPED(&sin6p->sin6_addr)) {
			struct sockaddr_in sin;

			in6_sin6_2_sin(&sin, sin6p);
			inp->inp_vflag |= INP_IPV4;
			inp->inp_vflag &= ~INP_IPV6;
			error = in_pcbbind(inp, (struct sockaddr *)&sin, p);
			goto out;
		}
	}
	error = in6_pcbbind(inp, nam, p);
	if (error)
		goto out;
	COMMON_END(PRU_BIND);
}
#endif /* INET6 */

/*
 * Prepare to accept connections.
 */
static int
tcp_usr_listen(struct socket *so, struct proc *p)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;

	COMMON_START();
	if (inp->inp_lport == 0)
		error = in_pcbbind(inp, (struct sockaddr *)0, p);
	if (error == 0)
		tp->t_state = TCPS_LISTEN;
	COMMON_END(PRU_LISTEN);
}

#ifdef INET6
static int
tcp6_usr_listen(struct socket *so, struct proc *p)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;

	COMMON_START();
	if (inp->inp_lport == 0) {
		inp->inp_vflag &= ~INP_IPV4;
		if (ip6_mapped_addr_on &&
		    (inp->inp_flags & IN6P_IPV6_V6ONLY) == 0)
			inp->inp_vflag |= INP_IPV4;
		error = in6_pcbbind(inp, (struct sockaddr *)0, p);
	}
	if (error == 0)
		tp->t_state = TCPS_LISTEN;
	COMMON_END(PRU_LISTEN);
}
#endif /* INET6 */

/*
 * Initiate connection to peer.
 * Create a template for use in transmissions on this connection.
 * Enter SYN_SENT state, and mark socket as connecting.
 * Start keep-alive timer, and seed output sequence space.
 * Send initial segment on connection.
 */
static int
tcp_usr_connect(struct socket *so, struct sockaddr *nam, struct proc *p)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;
	struct sockaddr_in *sinp;

	COMMON_START();

	/*
	 * Must disallow TCP ``connections'' to multicast addresses.
	 */
	sinp = (struct sockaddr_in *)nam;
	if (sinp->sin_family == AF_INET
	    && IN_MULTICAST(ntohl(sinp->sin_addr.s_addr))) {
		error = EAFNOSUPPORT;
		goto out;
	}


	if ((error = tcp_connect(tp, nam, p)) != 0)
		goto out;
	error = tcp_output(tp);
	COMMON_END(PRU_CONNECT);
}

#ifdef INET6
static int
tcp6_usr_connect(struct socket *so, struct sockaddr *nam, struct proc *p)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;
	struct sockaddr_in6 *sin6p;

	COMMON_START();

	/*
	 * Must disallow TCP ``connections'' to multicast addresses.
	 */
	sin6p = (struct sockaddr_in6 *)nam;
	if (sin6p->sin6_family == AF_INET6
	    && IN6_IS_ADDR_MULTICAST(&sin6p->sin6_addr)) {
		error = EAFNOSUPPORT;
		goto out;
	}

	if (IN6_IS_ADDR_V4MAPPED(&sin6p->sin6_addr)) {
		struct sockaddr_in sin;

		if (!ip6_mapped_addr_on ||
		    (inp->inp_flags & IN6P_IPV6_V6ONLY))
			return(EINVAL);

		in6_sin6_2_sin(&sin, sin6p);
		inp->inp_vflag |= INP_IPV4;
		inp->inp_vflag &= ~INP_IPV6;
		if ((error = tcp_connect(tp, (struct sockaddr *)&sin, p)) != 0)
			goto out;
		error = tcp_output(tp);
		goto out;
	}
	inp->inp_vflag &= ~INP_IPV4;
	inp->inp_vflag |= INP_IPV6;
	if ((error = tcp6_connect(tp, nam, p)) != 0)
		goto out;
	error = tcp_output(tp);
	COMMON_END(PRU_CONNECT);
}
#endif /* INET6 */

/*
 * Initiate disconnect from peer.
 * If connection never passed embryonic stage, just drop;
 * else if don't need to let data drain, then can just drop anyways,
 * else have to begin TCP shutdown process: mark socket disconnecting,
 * drain unread data, state switch to reflect user close, and
 * send segment (e.g. FIN) to peer.  Socket will be really disconnected
 * when peer sends FIN and acks ours.
 *
 * SHOULD IMPLEMENT LATER PRU_CONNECT VIA REALLOC TCPCB.
 */
static int
tcp_usr_disconnect(struct socket *so)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;

	COMMON_START();
	tp = tcp_disconnect(tp);
	COMMON_END(PRU_DISCONNECT);
}

/*
 * Accept a connection.  Essentially all the work is
 * done at higher levels; just return the address
 * of the peer, storing through addr.
 */
static int
tcp_usr_accept(struct socket *so, struct sockaddr **nam)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp = NULL;
	TCPDEBUG0;

	if (so->so_state & SS_ISDISCONNECTED) {
		error = ECONNABORTED;
		goto out;
	}
	if (inp == 0) {
		splx(s);
		return (EINVAL);
	}
	tp = intotcpcb(inp);
	TCPDEBUG1();
	in_setpeeraddr(so, nam);
	COMMON_END(PRU_ACCEPT);
}

#ifdef INET6
static int
tcp6_usr_accept(struct socket *so, struct sockaddr **nam)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp = NULL;
	TCPDEBUG0;

	if (so->so_state & SS_ISDISCONNECTED) {
		error = ECONNABORTED;
		goto out;
	}
	if (inp == 0) {
		splx(s);
		return (EINVAL);
	}
	tp = intotcpcb(inp);
	TCPDEBUG1();
	in6_mapped_peeraddr(so, nam);
	COMMON_END(PRU_ACCEPT);
}
#endif /* INET6 */
/*
 * Mark the connection as being incapable of further output.
 */
static int
tcp_usr_shutdown(struct socket *so)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;

	COMMON_START();
	socantsendmore(so);
	tp = tcp_usrclosed(tp);
	if (tp)
		error = tcp_output(tp);
	COMMON_END(PRU_SHUTDOWN);
}

/*
 * After a receive, possibly send window update to peer.
 */
static int
tcp_usr_rcvd(struct socket *so, int flags)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;

	COMMON_START();
	tcp_output(tp);
	COMMON_END(PRU_RCVD);
}

/*
 * Do a send by putting data in output queue and updating urgent
 * marker if URG set.  Possibly send more data.  Unlike the other
 * pru_*() routines, the mbuf chains are our responsibility.  We
 * must either enqueue them or free them.  The other pru_* routines
 * generally are caller-frees.
 */
static int
tcp_usr_send(struct socket *so, int flags, struct mbuf *m, 
	     struct sockaddr *nam, struct mbuf *control, struct proc *p)
{
	int s = splnet();
	int error = 0;
	struct inpcb *inp = sotoinpcb(so);
	struct tcpcb *tp;
#ifdef INET6
	int isipv6;
#endif
	TCPDEBUG0;

	if (inp == NULL) {
		/*
		 * OOPS! we lost a race, the TCP session got reset after
		 * we checked SS_CANTSENDMORE, eg: while doing uiomove or a
		 * network interrupt in the non-splnet() section of sosend().
		 */
		if (m)
			m_freem(m);
		if (control)
			m_freem(control);
		error = ECONNRESET;	/* XXX EPIPE? */
		tp = NULL;
		TCPDEBUG1();
		goto out;
	}
#ifdef INET6
	isipv6 = nam && nam->sa_family == AF_INET6;
#endif /* INET6 */
	tp = intotcpcb(inp);
	TCPDEBUG1();
	if (control) {
		/* TCP doesn't do control messages (rights, creds, etc) */
		if (control->m_len) {
			m_freem(control);
			if (m)
				m_freem(m);
			error = EINVAL;
			goto out;
		}
		m_freem(control);	/* empty control, just free it */
	}
	if(!(flags & PRUS_OOB)) {
		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