tcp_input.c

来自「eCos操作系统源码」· C语言 代码 · 共 2,238 行 · 第 1/5 页

//==========================================================================
//
//      src/sys/netinet/tcp_input.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, 1990, 1993, 1994, 1995
 *	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.
 *
 *	@(#)tcp_input.c	8.12 (Berkeley) 5/24/95
 * $FreeBSD: src/sys/netinet/tcp_input.c,v 1.107.2.16 2001/08/22 00:59:12 silby Exp $
 */

#include <sys/param.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/sysctl.h>
#include <sys/socketvar.h>

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

#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/ip_icmp.h>	/* for ICMP_BANDLIM		*/
#include <netinet/in_var.h>
#include <netinet/icmp_var.h>	/* for ICMP_BANDLIM		*/
#include <netinet/in_pcb.h>
#include <netinet/ip_var.h>
#ifdef INET6
#include <netinet/ip6.h>
#include <netinet/icmp6.h>
#include <netinet6/nd6.h>
#include <netinet6/ip6_var.h>
#include <netinet6/in6_pcb.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>

#ifdef INET6
#include <netinet6/tcp6_var.h>
#endif
#include <netinet/tcpip.h>
#ifdef TCPDEBUG
#include <netinet/tcp_debug.h>

u_char tcp_saveipgen[40]; /* the size must be of max ip header, now IPv6 */
struct tcphdr tcp_savetcp;
#endif /* TCPDEBUG */

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

static int	tcprexmtthresh = 3;
tcp_cc	tcp_ccgen;

struct	tcpstat tcpstat;
SYSCTL_STRUCT(_net_inet_tcp, TCPCTL_STATS, stats, CTLFLAG_RD, 
    &tcpstat , tcpstat, "TCP statistics (struct tcpstat, netinet/tcp_var.h)");

static int log_in_vain = 0;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, log_in_vain, CTLFLAG_RW, 
    &log_in_vain, 0, "Log all incoming TCP connections");

static int blackhole = 0;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, blackhole, CTLFLAG_RW,
	&blackhole, 0, "Do not send RST when dropping refused connections");

int tcp_delack_enabled = 1;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, delayed_ack, CTLFLAG_RW, 
    &tcp_delack_enabled, 0, 
    "Delay ACK to try and piggyback it onto a data packet");

int tcp_lq_overflow = 1;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, tcp_lq_overflow, CTLFLAG_RW,
    &tcp_lq_overflow, 0, 
    "Listen Queue Overflow");

#ifdef TCP_DROP_SYNFIN
static int drop_synfin = 0;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, drop_synfin, CTLFLAG_RW,
    &drop_synfin, 0, "Drop TCP packets with SYN+FIN set");
#endif

struct inpcbhead tcb;
#define	tcb6	tcb  /* for KAME src sync over BSD*'s */
struct inpcbinfo tcbinfo;

static void	 tcp_dooptions __P((struct tcpcb *,
	    u_char *, int, struct tcphdr *, struct tcpopt *));
static void	 tcp_pulloutofband __P((struct socket *,
	    struct tcphdr *, struct mbuf *, int));
static int	 tcp_reass __P((struct tcpcb *, struct tcphdr *, int *,
				struct mbuf *));
static void	 tcp_xmit_timer __P((struct tcpcb *, int));
static int	 tcp_newreno __P((struct tcpcb *, struct tcphdr *));

/* Neighbor Discovery, Neighbor Unreachability Detection Upper layer hint. */
#ifdef INET6
#define ND6_HINT(tp) \
do { \
	if ((tp) && (tp)->t_inpcb && \
	    ((tp)->t_inpcb->inp_vflag & INP_IPV6) != 0 && \
	    (tp)->t_inpcb->in6p_route.ro_rt) \
		nd6_nud_hint((tp)->t_inpcb->in6p_route.ro_rt, NULL, 0); \
} while (0)
#else
#define ND6_HINT(tp)
#endif

/*
 * Indicate whether this ack should be delayed.
 */
#define DELAY_ACK(tp) \
	(tcp_delack_enabled && !callout_pending(tp->tt_delack))

static int
tcp_reass(tp, th, tlenp, m)
	register struct tcpcb *tp;
	register struct tcphdr *th;
	int *tlenp;
	struct mbuf *m;
{
	struct tseg_qent *q;
	struct tseg_qent *p = NULL;
	struct tseg_qent *nq;
	struct tseg_qent *te;
	struct socket *so = tp->t_inpcb->inp_socket;
	int flags;

	/*
	 * Call with th==0 after become established to
	 * force pre-ESTABLISHED data up to user socket.
	 */
	if (th == 0)
		goto present;

	/* Allocate a new queue entry. If we can't, just drop the pkt. XXX */
	MALLOC(te, struct tseg_qent *, sizeof (struct tseg_qent), M_TSEGQ,
	       M_NOWAIT);
	if (te == NULL) {
		tcpstat.tcps_rcvmemdrop++;
		m_freem(m);
		return (0);
	}

	/*
	 * Find a segment which begins after this one does.
	 */
	LIST_FOREACH(q, &tp->t_segq, tqe_q) {
		if (SEQ_GT(q->tqe_th->th_seq, th->th_seq))
			break;
		p = q;
	}

	/*
	 * If there is a preceding segment, it may provide some of
	 * our data already.  If so, drop the data from the incoming
	 * segment.  If it provides all of our data, drop us.
	 */
	if (p != NULL) {
		register int i;
		/* conversion to int (in i) handles seq wraparound */
		i = p->tqe_th->th_seq + p->tqe_len - th->th_seq;
		if (i > 0) {
			if (i >= *tlenp) {
				tcpstat.tcps_rcvduppack++;
				tcpstat.tcps_rcvdupbyte += *tlenp;
				m_freem(m);
				FREE(te, M_TSEGQ);
				/*
				 * Try to present any queued data
				 * at the left window edge to the user.
				 * This is needed after the 3-WHS
				 * completes.
				 */
				goto present;	/* ??? */
			}
			m_adj(m, i);
			*tlenp -= i;
			th->th_seq += i;
		}
	}
	tcpstat.tcps_rcvoopack++;
	tcpstat.tcps_rcvoobyte += *tlenp;

	/*
	 * While we overlap succeeding segments trim them or,
	 * if they are completely covered, dequeue them.
	 */
	while (q) {
		register int i = (th->th_seq + *tlenp) - q->tqe_th->th_seq;
		if (i <= 0)
			break;
		if (i < q->tqe_len) {
			q->tqe_th->th_seq += i;
			q->tqe_len -= i;
			m_adj(q->tqe_m, i);
			break;
		}

		nq = LIST_NEXT(q, tqe_q);
		LIST_REMOVE(q, tqe_q);
		m_freem(q->tqe_m);
		FREE(q, M_TSEGQ);
		q = nq;
	}

	/* Insert the new segment queue entry into place. */
	te->tqe_m = m;
	te->tqe_th = th;
	te->tqe_len = *tlenp;

	if (p == NULL) {
		LIST_INSERT_HEAD(&tp->t_segq, te, tqe_q);
	} else {
		LIST_INSERT_AFTER(p, te, tqe_q);
	}

present:
	/*
	 * Present data to user, advancing rcv_nxt through
	 * completed sequence space.
	 */
	if (!TCPS_HAVEESTABLISHED(tp->t_state))
		return (0);
	q = LIST_FIRST(&tp->t_segq);
	if (!q || q->tqe_th->th_seq != tp->rcv_nxt)
		return (0);
	do {
		tp->rcv_nxt += q->tqe_len;
		flags = q->tqe_th->th_flags & TH_FIN;
		nq = LIST_NEXT(q, tqe_q);
		LIST_REMOVE(q, tqe_q);
		if (so->so_state & SS_CANTRCVMORE)
			m_freem(q->tqe_m);
		else
			sbappend(&so->so_rcv, q->tqe_m);
		FREE(q, M_TSEGQ);
		q = nq;
	} while (q && q->tqe_th->th_seq == tp->rcv_nxt);
	ND6_HINT(tp);
	sorwakeup(so);
	return (flags);
}

/*
 * TCP input routine, follows pages 65-76 of the
 * protocol specification dated September, 1981 very closely.
 */
#ifdef INET6
int
tcp6_input(mp, offp, proto)
	struct mbuf **mp;
	int *offp, proto;
{
	register struct mbuf *m = *mp;
	struct in6_ifaddr *ia6;

	IP6_EXTHDR_CHECK(m, *offp, sizeof(struct tcphdr), IPPROTO_DONE);

	/*
	 * draft-itojun-ipv6-tcp-to-anycast
	 * better place to put this in?
	 */
	ia6 = ip6_getdstifaddr(m);
	if (ia6 && (ia6->ia6_flags & IN6_IFF_ANYCAST)) {		
		struct ip6_hdr *ip6;

		ip6 = mtod(m, struct ip6_hdr *);
		icmp6_error(m, ICMP6_DST_UNREACH, ICMP6_DST_UNREACH_ADDR,
			    (caddr_t)&ip6->ip6_dst - (caddr_t)ip6);
		return IPPROTO_DONE;
	}

	tcp_input(m, *offp);
	return IPPROTO_DONE;
}
#endif

void
tcp_input(m, off0)
	register struct mbuf *m;
	int off0;
{
	register struct tcphdr *th;
	register struct ip *ip = NULL;
	register struct ipovly *ipov;
	register struct inpcb *inp;
	u_char *optp = NULL;
	int optlen = 0;
	int len, tlen, off;
	int drop_hdrlen;
	register struct tcpcb *tp = 0;
	register int thflags;
	struct socket *so = 0;
	int todrop, acked, ourfinisacked, needoutput = 0;
	struct in_addr laddr;
#ifdef INET6
	struct in6_addr laddr6;
#endif
	int dropsocket = 0;
	int iss = 0;
	u_long tiwin;
	struct tcpopt to;		/* options in this segment */
	struct rmxp_tao *taop;		/* pointer to our TAO cache entry */
	struct rmxp_tao	tao_noncached;	/* in case there's no cached entry */
#ifdef TCPDEBUG
	short ostate = 0;
#endif
#ifdef INET6
	struct ip6_hdr *ip6 = NULL;
	int isipv6;
#endif /* INET6 */
	int rstreason; /* For badport_bandlim accounting purposes */

#ifdef INET6
	isipv6 = (mtod(m, struct ip *)->ip_v == 6) ? 1 : 0;
#endif
	bzero((char *)&to, sizeof(to));

	tcpstat.tcps_rcvtotal++;

#ifdef INET6
	if (isipv6) {
		/* IP6_EXTHDR_CHECK() is already done at tcp6_input() */
		ip6 = mtod(m, struct ip6_hdr *);
		tlen = sizeof(*ip6) + ntohs(ip6->ip6_plen) - off0;
		if (in6_cksum(m, IPPROTO_TCP, off0, tlen)) {
			tcpstat.tcps_rcvbadsum++;
			goto drop;
		}
		th = (struct tcphdr *)((caddr_t)ip6 + off0);

		/*
		 * Be proactive about unspecified IPv6 address in source.
		 * As we use all-zero to indicate unbounded/unconnected pcb,
		 * unspecified IPv6 address can be used to confuse us.
		 *
		 * Note that packets with unspecified IPv6 destination is
		 * already dropped in ip6_input.
		 */
		if (IN6_IS_ADDR_UNSPECIFIED(&ip6->ip6_src)) {
			/* XXX stat */
			goto drop;
		}
	} else
#endif /* INET6 */
      {
	/*
	 * Get IP and TCP header together in first mbuf.
	 * Note: IP leaves IP header in first mbuf.
	 */
	if (off0 > sizeof (struct ip)) {
		ip_stripoptions(m, (struct mbuf *)0);
		off0 = sizeof(struct ip);
	}
	if (m->m_len < sizeof (struct tcpiphdr)) {
		if ((m = m_pullup(m, sizeof (struct tcpiphdr))) == 0) {
			tcpstat.tcps_rcvshort++;
			return;
		}
	}
	ip = mtod(m, struct ip *);
	ipov = (struct ipovly *)ip;
	th = (struct tcphdr *)((caddr_t)ip + off0);
	tlen = ip->ip_len;

	if (m->m_pkthdr.csum_flags & CSUM_DATA_VALID) {
		if (m->m_pkthdr.csum_flags & CSUM_PSEUDO_HDR)
                	th->th_sum = m->m_pkthdr.csum_data;
		else
	                th->th_sum = in_pseudo(ip->ip_src.s_addr,
			    ip->ip_dst.s_addr, htonl(m->m_pkthdr.csum_data +
			    ip->ip_len + IPPROTO_TCP));
		th->th_sum ^= 0xffff;
	} else {
		/*
		 * Checksum extended TCP header and data.
		 */
		len = sizeof (struct ip) + tlen;
		bzero(ipov->ih_x1, sizeof(ipov->ih_x1));
		ipov->ih_len = (u_short)tlen;
		HTONS(ipov->ih_len);
		th->th_sum = in_cksum(m, len);
	}
	if (th->th_sum) {
		tcpstat.tcps_rcvbadsum++;
		goto drop;
	}
#ifdef INET6
	/* Re-initialization for later version check */
	ip->ip_v = IPVERSION;
#endif
      }