tcp_input.c

完整的TCP/IP源代码

	 	/*
		 * In SYN_RECEIVED and ESTABLISHED STATES
		 * enter the CLOSE_WAIT state.
		 */
		case TCPS_SYN_RECEIVED:
		case TCPS_ESTABLISHED:
			tp->t_state = TCPS_CLOSE_WAIT;
			break;

	 	/*
		 * If still in FIN_WAIT_1 STATE FIN has not been acked so
		 * enter the CLOSING state.
		 */
		case TCPS_FIN_WAIT_1:
			tp->t_state = TCPS_CLOSING;
			break;

	 	/*
		 * In FIN_WAIT_2 state enter the TIME_WAIT state,
		 * starting the time-wait timer, turning off the other 
		 * standard timers.
		 */
		case TCPS_FIN_WAIT_2:
			tp->t_state = TCPS_TIME_WAIT;
			tcp_canceltimers(tp);
			tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
			soisdisconnected(so);
			break;

		/*
		 * In TIME_WAIT state restart the 2 MSL time_wait timer.
		 */
		case TCPS_TIME_WAIT:
			tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
			break;
		}
	}
	if (so->so_options & SO_DEBUG)
		(*tcpTraceRtn)(TA_INPUT, ostate, tp, &tcp_saveti, 0);

	/*
	 * Return any desired output.
	 */
	if (needoutput || (tp->t_flags & TF_ACKNOW))
		(void) tcp_output(tp);
	return;

dropafterack:
	/*
	 * Generate an ACK dropping incoming segment if it occupies
	 * sequence space, where the ACK reflects our state.
	 */
	if (tiflags & TH_RST)
		goto drop;
	m_freem(m);
	tp->t_flags |= TF_ACKNOW;
	(void) tcp_output(tp);
	return;

dropwithreset:
	/*
	 * Generate a RST, dropping incoming segment.
	 * Make ACK acceptable to originator of segment.
	 * Don't bother to respond if destination was broadcast/multicast.
	 */
	if ((tiflags & TH_RST) || m->m_flags & (M_BCAST|M_MCAST) ||
	    IN_MULTICAST(ntohl(ti->ti_dst.s_addr)))
		goto drop;
	if (tiflags & TH_ACK)
		tcp_respond(tp, ti, m, (tcp_seq)0, ti->ti_ack, TH_RST);
	else {
		if (tiflags & TH_SYN)
			ti->ti_len++;
		tcp_respond(tp, ti, m, ti->ti_seq+ti->ti_len, (tcp_seq)0,
		    TH_RST|TH_ACK);
	}

        /* MIB-II Count # of Resets Sent */

         tcpOutRsts++;

	/* destroy temporarily created socket */
	if (dropsocket)
		(void) soabort(so);
	return;

drop:
	/*
	 * Drop space held by incoming segment and return.
	 */
	if (tp && (tp->t_inpcb->inp_socket->so_options & SO_DEBUG))
		(*tcpTraceRtn)(TA_DROP, ostate, tp, &tcp_saveti, 0);
	m_freem(m);
	/* destroy temporarily created socket */
	if (dropsocket)
		(void) soabort(so);
	return;
#ifndef TUBA_INCLUDE
}

void
tcp_dooptions(tp, cp, cnt, ti, ts_present, ts_val, ts_ecr)
	struct tcpcb *tp;
	u_char *cp;
	int cnt;
	struct tcpiphdr *ti;
	int *ts_present;
	u_long *ts_val, *ts_ecr;
{
	u_short mss;
	int opt, optlen;

	for (; cnt > 0; cnt -= optlen, cp += optlen) {
		opt = cp[0];
		if (opt == TCPOPT_EOL)
			break;
		if (opt == TCPOPT_NOP)
			optlen = 1;
		else {
			optlen = cp[1];
			if (optlen <= 0)
				break;
		}
		switch (opt) {

		default:
			continue;

		case TCPOPT_MAXSEG:
			if (optlen != TCPOLEN_MAXSEG)
				continue;
			if (!(ti->ti_flags & TH_SYN))
				continue;
			bcopy((char *) cp + 2, (char *) &mss, sizeof(mss));
			NTOHS(mss);
			(void) tcp_mss(tp, mss);	/* sets t_maxseg */
			break;

		case TCPOPT_WINDOW:
			if (optlen != TCPOLEN_WINDOW)
				continue;
			if (!(ti->ti_flags & TH_SYN))
				continue;
			tp->t_flags |= TF_RCVD_SCALE;
			tp->requested_s_scale = min(cp[2], TCP_MAX_WINSHIFT);
			break;

		case TCPOPT_TIMESTAMP:
			if (optlen != TCPOLEN_TIMESTAMP)
				continue;
			*ts_present = 1;
			bcopy((char *)cp + 2, (char *) ts_val, sizeof(*ts_val));
			NTOHL(*ts_val);
			bcopy((char *)cp + 6, (char *) ts_ecr, sizeof(*ts_ecr));
			NTOHL(*ts_ecr);

			/* 
			 * A timestamp received in a SYN makes
			 * it ok to send timestamp requests and replies.
			 */
			if (ti->ti_flags & TH_SYN) {
				tp->t_flags |= TF_RCVD_TSTMP;
				tp->ts_recent = *ts_val;
				tp->ts_recent_age = tcp_now;
			}
			break;
		}
	}
}

/*
 * Pull out of band byte out of a segment so
 * it doesn't appear in the user's data queue.
 * It is still reflected in the segment length for
 * sequencing purposes.
 */
void
tcp_pulloutofband(so, ti, m)
	struct socket *so;
	struct tcpiphdr *ti;
	register struct mbuf *m;
{
	int cnt = ti->ti_urp - 1;
	
	while (cnt >= 0) {
		if (m->m_len > cnt) {
			char *cp = mtod(m, caddr_t) + cnt;
			struct tcpcb *tp = sototcpcb(so);

			tp->t_iobc = *cp;
			tp->t_oobflags |= TCPOOB_HAVEDATA;
			bcopy(cp+1, cp, (unsigned)(m->m_len - cnt - 1));
			m->m_len--;
			return;
		}
		cnt -= m->m_len;
		m = m->m_next;
		if (m == 0)
			break;
	}
	panic("tcp_pulloutofband");
}

/*
 * Collect new round-trip time estimate
 * and update averages and current timeout.
 */
void
tcp_xmit_timer(tp, rtt)
	register struct tcpcb *tp;
	short rtt;
{
	register short delta;

	tcpstat.tcps_rttupdated++;
	if (tp->t_srtt != 0) {
		/*
		 * srtt is stored as fixed point with 3 bits after the
		 * binary point (i.e., scaled by 8).  The following magic
		 * is equivalent to the smoothing algorithm in rfc793 with
		 * an alpha of .875 (srtt = rtt/8 + srtt*7/8 in fixed
		 * point).  Adjust rtt to origin 0.
		 */
		delta = rtt - 1 - (tp->t_srtt >> TCP_RTT_SHIFT);
		if ((tp->t_srtt += delta) <= 0)
			tp->t_srtt = 1;
		/*
		 * We accumulate a smoothed rtt variance (actually, a
		 * smoothed mean difference), then set the retransmit
		 * timer to smoothed rtt + 4 times the smoothed variance.
		 * rttvar is stored as fixed point with 2 bits after the
		 * binary point (scaled by 4).  The following is
		 * equivalent to rfc793 smoothing with an alpha of .75
		 * (rttvar = rttvar*3/4 + |delta| / 4).  This replaces
		 * rfc793's wired-in beta.
		 */
		if (delta < 0)
			delta = -delta;
		delta -= (tp->t_rttvar >> TCP_RTTVAR_SHIFT);
		if ((tp->t_rttvar += delta) <= 0)
			tp->t_rttvar = 1;
	} else {
		/* 
		 * No rtt measurement yet - use the unsmoothed rtt.
		 * Set the variance to half the rtt (so our first
		 * retransmit happens at 3*rtt).
		 */
		tp->t_srtt = rtt << TCP_RTT_SHIFT;
		tp->t_rttvar = rtt << (TCP_RTTVAR_SHIFT - 1);
	}
	tp->t_rtt = 0;
	tp->t_rxtshift = 0;

	/*
	 * the retransmit should happen at rtt + 4 * rttvar.
	 * Because of the way we do the smoothing, srtt and rttvar
	 * will each average +1/2 tick of bias.  When we compute
	 * the retransmit timer, we want 1/2 tick of rounding and
	 * 1 extra tick because of +-1/2 tick uncertainty in the
	 * firing of the timer.  The bias will give us exactly the
	 * 1.5 tick we need.  But, because the bias is
	 * statistical, we have to test that we don't drop below
	 * the minimum feasible timer (which is 2 ticks).
	 */
	TCPT_RANGESET(tp->t_rxtcur, TCP_REXMTVAL(tp),
	    tp->t_rttmin, TCPTV_REXMTMAX);
	
	/*
	 * We received an ack for a packet that wasn't retransmitted;
	 * it is probably safe to discard any error indications we've
	 * received recently.  This isn't quite right, but close enough
	 * for now (a route might have failed after we sent a segment,
	 * and the return path might not be symmetrical).
	 */
	tp->t_softerror = 0;
}

/*
 * Determine a reasonable value for maxseg size.
 * If the route is known, check route for mtu.
 * If none, use an mss that can be handled on the outgoing
 * interface without forcing IP to fragment;
 * If no route is found, route has no mtu,
 * or the destination isn't local, use a default, hopefully conservative
 * size (usually 512 or the default IP max size, but no more than the mtu
 * of the interface), as we can't discover anything about intervening
 * gateways or networks.  We also initialize the congestion/slow start
 * window to be a single segment if the destination isn't local.
 * While looking at the routing entry, we also initialize other path-dependent
 * parameters from pre-set or cached values in the routing entry.
 */
int
tcp_mss(tp, offer)
	register struct tcpcb *tp;
	u_int offer;
{
	struct route *ro;
	register struct rtentry *rt;
	struct ifnet *ifp;
	register int rtt, mss;
	u_long bufsize;
	struct inpcb *inp;
	struct socket *so;
	extern int tcp_mssdflt;

	inp = tp->t_inpcb;
	ro = &inp->inp_route;

	if ((rt = ro->ro_rt) == (struct rtentry *)0) {
		/* No route yet, so try to acquire one */
		if (inp->inp_faddr.s_addr != INADDR_ANY) {
			ro->ro_dst.sa_family = AF_INET;
			ro->ro_dst.sa_len = sizeof(ro->ro_dst);
			((struct sockaddr_in *) &ro->ro_dst)->sin_addr =
				inp->inp_faddr;
			TOS_SET (&ro->ro_dst, inp->inp_ip.ip_tos);
			rtalloc(ro);
		}
		if ((rt = ro->ro_rt) == (struct rtentry *)0)
			return (tcp_mssdflt);
	}
	ifp = rt->rt_ifp;
	so = inp->inp_socket;

#ifdef RTV_MTU	/* if route characteristics exist ... */
	/*
	 * While we're here, check if there's an initial rtt
	 * or rttvar.  Convert from the route-table units
	 * to scaled multiples of the slow timeout timer.
	 */
	if (tp->t_srtt == 0 && (rtt = rt->rt_rmx.rmx_rtt)) {
		/*
		 * XXX the lock bit for MTU indicates that the value
		 * is also a minimum value; this is subject to time.
		 */
		if (rt->rt_rmx.rmx_locks & RTV_RTT)
			tp->t_rttmin = rtt / (RTM_RTTUNIT / PR_SLOWHZ);
		tp->t_srtt = rtt / (RTM_RTTUNIT / (PR_SLOWHZ * TCP_RTT_SCALE));
		if (rt->rt_rmx.rmx_rttvar)
			tp->t_rttvar = rt->rt_rmx.rmx_rttvar /
			    (RTM_RTTUNIT / (PR_SLOWHZ * TCP_RTTVAR_SCALE));
		else
			/* default variation is +- 1 rtt */
			tp->t_rttvar =
			    tp->t_srtt * TCP_RTTVAR_SCALE / TCP_RTT_SCALE;
		TCPT_RANGESET(tp->t_rxtcur,
		    ((tp->t_srtt >> 2) + tp->t_rttvar) >> 1,
		    tp->t_rttmin, TCPTV_REXMTMAX);
	}
	/*
	 * if there's an mtu associated with the route, use it
	 */
	if (rt->rt_rmx.rmx_mtu)
		mss = rt->rt_rmx.rmx_mtu - sizeof(struct tcpiphdr);
	else
#endif /* RTV_MTU */
	{
		mss = ifp->if_mtu - sizeof(struct tcpiphdr);
		if (!in_localaddr(inp->inp_faddr))
			mss = min(mss, tcp_mssdflt);
	}
	/*
	 * The current mss, t_maxseg, is initialized to the default value.
	 * If we compute a smaller value, reduce the current mss.
	 * If we compute a larger value, return it for use in sending
	 * a max seg size option, but don't store it for use
	 * unless we received an offer at least that large from peer.
	 * However, do not accept offers under 32 bytes.
	 */
	if (offer)
		mss = min(mss, offer);
	mss = max(mss, 32);		/* sanity */
	if (mss < tp->t_maxseg || offer != 0) {
		/*
		 * If there's a pipesize, change the socket buffer
		 * to that size.  Make the socket buffers an integral
		 * number of mss units; if the mss is larger than
		 * the socket buffer, decrease the mss.
		 */
#ifdef RTV_SPIPE
		if ((bufsize = rt->rt_rmx.rmx_sendpipe) == 0)
#endif
			bufsize = so->so_snd.sb_hiwat;
		if (bufsize < mss)
			mss = bufsize;
		tp->t_maxseg = mss;

#ifdef RTV_RPIPE
		if ((bufsize = rt->rt_rmx.rmx_recvpipe) == 0)
#endif
			bufsize = so->so_rcv.sb_hiwat;
	}
	tp->snd_cwnd = mss;

#ifdef RTV_SSTHRESH
	if (rt->rt_rmx.rmx_ssthresh) {
		/*
		 * There's some sort of gateway or interface
		 * buffer limit on the path.  Use this to set
		 * the slow start threshhold, but set the
		 * threshold to no less than 2*mss.
		 */
		tp->snd_ssthresh = max(2 * mss, rt->rt_rmx.rmx_ssthresh);
	}
#endif /* RTV_MTU */
	return (mss);
}
#endif /* TUBA_INCLUDE */

#ifdef TCP_DEBUG /* XXX just for debugging */
void tcbShow
    (
    FAST struct inpcb *pinPcb		/* pointer to the pcb control block */
    )
    {
    struct tcpcb *	pTcpCb;	/* pointer to tcp Control block */

    pTcpCb = (struct tcpcb *) pinPcb->inp_ppcb;

    printf ("send unacknowledged: %8u\n", pTcpCb->snd_una); 
    printf ("send next: %8u\n", pTcpCb->snd_nxt); 
    printf ("send window update seg seqnumber: %8u\n", pTcpCb->snd_wl1);
    printf ("send window updata seg ack number: %8u\n", pTcpCb->snd_wl2); 
    printf ("send window: %8u\n", pTcpCb->snd_wnd); 
    printf ("recv window: %8u\n", pTcpCb->rcv_wnd);
    printf ("recv next: %8u\n", pTcpCb->rcv_nxt);
    printf ("recv advertised window by other end: %8u\n", pTcpCb->rcv_adv);
    printf ("send highest sequence number sent: %8u\n", pTcpCb->snd_max);
    printf ("congestion controlled window: %8u\n", pTcpCb->snd_cwnd);
    printf ("congestion threshold: %8u\n", pTcpCb->snd_ssthresh);
    printf ("largest window peer has offered: %8u\n", pTcpCb->max_sndwnd);
    printf ("send scale %8u\n", pTcpCb->snd_scale) ;
    printf ("recv scale %8u\n", pTcpCb->rcv_scale) ;
    printf ("send pending window scale %8u\n", pTcpCb->request_r_scale) ;
    printf ("recv pending window scale %8u\n", pTcpCb->requested_s_scale) ;
    
    }
#endif /* XXX just for debugging */