tcp_input.c

VxWorks BSP框架源代码包含头文件和驱动

		 * Do quick retrieval of timestamp options ("options
		 * prediction?").  If timestamp is the only option and it's
		 * formatted as recommended in RFC 1323 appendix A, we
		 * quickly get the values now and not bother calling
		 * tcp_dooptions(), etc.
		 */
		if ((optlen == TCPOLEN_TSTAMP_APPA ||
		     (optlen > TCPOLEN_TSTAMP_APPA &&
			optp[TCPOLEN_TSTAMP_APPA] == TCPOPT_EOL)) &&
		     *(u_long *)optp == htonl(TCPOPT_TSTAMP_HDR) &&
		     (ti->ti_flags & TH_SYN) == 0) {
			ts_present = 1;
			ts_val = ntohl(*(u_long *)(optp + 4));
			ts_ecr = ntohl(*(u_long *)(optp + 8));
			optp = NULL;	/* we've parsed the options */
		}
	}
	tiflags = ti->ti_flags;

	/*
	 * Convert TCP protocol specific fields to host format.
	 */
	NTOHL(ti->ti_seq);
	NTOHL(ti->ti_ack);
	NTOHS(ti->ti_win);
	NTOHS(ti->ti_urp);

	/*
	 * Locate pcb for segment.
	 */
findpcb:
	inp = tcp_last_inpcb;
        if ((inp != NULL) &&
            (inp->inp_lport == ti->ti_dport) &&
	    (inp->inp_fport == ti->ti_sport) &&
	    (inp->inp_faddr.s_addr == ti->ti_src.s_addr) &&
            (inp->inp_laddr.s_addr == ti->ti_dst.s_addr))
            {
            goto pcbMatchFound;
            }
        else
            {
            ++tcpstat.tcps_pcbcachemiss;

            /*
             * Search the hash table of pcbs.
             */
            inp = in_pcblookuphash (&tcbinfo, ti->ti_src, ti->ti_sport,
                                    ti->ti_dst, ti->ti_dport, 1);
            /*
             * ...and if that fails, drop the packet.
             */
            if (inp == NULL)
                {
#ifdef WV_INSTRUMENTATION
#ifdef INCLUDE_WVNET    /* WV_NET_WARNING event */
            WV_NET_PORTIN_EVENT_4 (NET_CORE_EVENT, WV_NET_WARNING, 11, 4,
                                   ti->ti_sport, ti->ti_dport,
                                   WV_NETEVENT_TCPIN_SEARCHFAIL, WV_NET_RECV, 
                                   ti->ti_src.s_addr, ti->ti_sport,
                                   ti->ti_dst.s_addr, ti->ti_dport)
#endif  /* INCLUDE_WVNET */
#endif

                /*
                 * If the state is CLOSED (i.e., TCB does not exist) then
                 * all data in the incoming segment is discarded.
                 * If the TCB exists but is in CLOSED state, it is embryonic,
                 * but should either do a listen or a connect soon.
                 */
                goto dropwithreset;
                }
            }
        tcp_last_inpcb = inp;		/* update the cache */

pcbMatchFound:
	tp = intotcpcb(inp);
	if (tp == 0)
		goto dropwithreset;
	if (tp->t_state == TCPS_CLOSED)
		goto drop;
	
	/* Unscale the window into a 32-bit value. */
	if ((tiflags & TH_SYN) == 0)
		tiwin = ti->ti_win << tp->snd_scale;
	else
		tiwin = ti->ti_win;

	so = inp->inp_socket;
	if (so->so_options & (SO_DEBUG|SO_ACCEPTCONN)) {
		if (so->so_options & SO_DEBUG) {
			ostate = tp->t_state;
			tcp_saveti = *ti;
		}
		if (so->so_options & SO_ACCEPTCONN) {
			if ((tiflags & (TH_RST|TH_ACK|TH_SYN)) != TH_SYN) {
				/*
				 * Note: dropwithreset makes sure we don't
				 * send a reset in response to a RST.
				 */
				if (tiflags & TH_ACK) {
					tcpstat.tcps_badsyn++;
					goto dropwithreset;
				}
				goto drop;
			}
			so = sonewconn(so, 0);
			if (so == 0)
				goto drop;
			/*
			 * This is ugly, but ....
			 *
			 * Mark socket as temporary until we're
			 * committed to keeping it.  The code at
			 * ``drop'' and ``dropwithreset'' check the
			 * flag dropsocket to see if the temporary
			 * socket created here should be discarded.
			 * We mark the socket as discardable until
			 * we're committed to it below in TCPS_LISTEN.
			 */
			dropsocket++;
			inp = (struct inpcb *)so->so_pcb;
			inp->inp_laddr = ti->ti_dst;
			inp->inp_lport = ti->ti_dport;
                        in_pcbrehash(inp);
#ifdef SRCRT
#if BSD>=43
			inp->inp_options = ip_srcroute();
#endif
#endif /* SRCRT */
			tp = intotcpcb(inp);
			tp->t_state = TCPS_LISTEN;

			/* Compute proper scaling value from buffer space
			 */
			while (tp->request_r_scale < TCP_MAX_WINSHIFT &&
			   TCP_MAXWIN << tp->request_r_scale < so->so_rcv.sb_hiwat)
				tp->request_r_scale++;
		}
	}

	/*
	 * Segment received on connection.
	 * Reset idle time and keep-alive timer.
	 */
	tp->t_idle = 0;
	tp->t_timer[TCPT_KEEP] = tcp_keepidle;

	/*
	 * Process options if not in LISTEN state,
	 * else do it below (after getting remote address).
	 */
	if (optp && tp->t_state != TCPS_LISTEN)
		tcp_dooptions(tp, optp, optlen, ti,
			&ts_present, &ts_val, &ts_ecr);

	/* 
	 * Header prediction: check for the two common cases
	 * of a uni-directional data xfer.  If the packet has
	 * no control flags, is in-sequence, the window didn't
	 * change and we're not retransmitting, it's a
	 * candidate.  If the length is zero and the ack moved
	 * forward, we're the sender side of the xfer.  Just
	 * free the data acked & wake any higher level process
	 * that was blocked waiting for space.  If the length
	 * is non-zero and the ack didn't move, we're the
	 * receiver side.  If we're getting packets in-order
	 * (the reassembly queue is empty), add the data to
	 * the socket buffer and note that we need a delayed ack.
	 */
	if (tp->t_state == TCPS_ESTABLISHED &&
	    (tiflags & (TH_SYN|TH_FIN|TH_RST|TH_URG|TH_ACK)) == TH_ACK &&
	    (!ts_present || TSTMP_GEQ(ts_val, tp->ts_recent)) &&
	    ti->ti_seq == tp->rcv_nxt &&
	    tiwin && tiwin == tp->snd_wnd &&
	    tp->snd_nxt == tp->snd_max) {

		/* 
		 * If last ACK falls within this segment's sequence numbers,
		 *  record the timestamp.
		 */
		if (ts_present && SEQ_LEQ(ti->ti_seq, tp->last_ack_sent)) {
			tp->ts_recent_age = tcp_now;
			tp->ts_recent = ts_val;
		}

		if (ti->ti_len == 0) {
			if (SEQ_GT(ti->ti_ack, tp->snd_una) &&
			    SEQ_LEQ(ti->ti_ack, tp->snd_max) &&
			    tp->snd_cwnd >= tp->snd_wnd) {
				/*
				 * this is a pure ack for outstanding data.
				 */
				++tcpstat.tcps_predack;
				if (ts_present)
					tcp_xmit_timer(tp, tcp_now-ts_ecr+1);
				else if (tp->t_rtt &&
					    SEQ_GT(ti->ti_ack, tp->t_rtseq))
					tcp_xmit_timer(tp, tp->t_rtt);
				acked = ti->ti_ack - tp->snd_una;

#ifdef WV_INSTRUMENTATION
#ifdef INCLUDE_WVNET    /* WV_NET_INFO event */
            WV_NET_PORTIN_EVENT_3 (NET_CORE_EVENT, WV_NET_INFO, 27, 9,
                                   ti->ti_sport, ti->ti_dport,
                                   WV_NETEVENT_TCPIN_PUREACK, WV_NET_RECV, 
                                   so->so_fd, tp->snd_una, ti->ti_ack)
#endif  /* INCLUDE_WVNET */
#endif

				tcpstat.tcps_rcvackpack++;
				tcpstat.tcps_rcvackbyte += acked;
				sbdrop(&so->so_snd, acked);
				tp->snd_una = ti->ti_ack;
				m_freem(m);

				/*
				 * If all outstanding data are acked, stop
				 * retransmit timer, otherwise restart timer
				 * using current (possibly backed-off) value.
				 * If process is waiting for space,
				 * wakeup/selwakeup/signal.  If data
				 * are ready to send, let tcp_output
				 * decide between more output or persist.
				 */
				if (tp->snd_una == tp->snd_max)
					tp->t_timer[TCPT_REXMT] = 0;
				else if (tp->t_timer[TCPT_PERSIST] == 0)
					tp->t_timer[TCPT_REXMT] = tp->t_rxtcur;

				if ((so->so_snd.sb_want > 0) ||
				    (so->so_snd.sb_sel)      ||
				    (sowakeupHook != NULL))
				    {
				    if (sbspace (&so->so_snd) >= 
					so->so_snd.sb_lowat) 
					sowwakeup(so);
				    }
				if (so->so_snd.sb_cc)
					(void) tcp_output(tp);
				return;
			}
		} else if (ti->ti_ack == tp->snd_una &&
		    tp->seg_next == (struct tcpiphdr *)tp &&
		    ti->ti_len <= sbspace(&so->so_rcv)) {
			/*
			 * this is a pure, in-sequence data packet
			 * with nothing on the reassembly queue and
			 * we have enough buffer space to take it.
			 */

#ifdef WV_INSTRUMENTATION
#ifdef INCLUDE_WVNET    /* WV_NET_INFO event */
        WV_NET_PORTIN_EVENT_3 (NET_CORE_EVENT, WV_NET_INFO, 28, 10,
                               ti->ti_sport, ti->ti_dport,
                               WV_NETEVENT_TCPIN_PUREDATA, WV_NET_RECV, 
                               so->so_fd, ti->ti_seq, ti->ti_len)
#endif  /* INCLUDE_WVNET */
#endif

			++tcpstat.tcps_preddat;
			tp->rcv_nxt += ti->ti_len;
			tcpstat.tcps_rcvpack++;
			tcpstat.tcps_rcvbyte += ti->ti_len;
			/*
			 * Drop TCP, IP headers and TCP options then add data
			 * to socket buffer.
			 */
			m->m_data += sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);
			m->m_len -= sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);
			sbappend(&so->so_rcv, m);
			sorwakeup(so);
			if (ti->ti_flags & TH_PUSH)
				tp->t_flags |= TF_ACKNOW;
			else
				tp->t_flags |= TF_DELACK;
			return;
		}
	}

	/*
	 * Drop TCP, IP headers and TCP options.
	 */
	m->m_data += sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);
	m->m_len  -= sizeof(struct tcpiphdr)+off-sizeof(struct tcphdr);

	/*
	 * Calculate amount of space in receive window,
	 * and then do TCP input processing.
	 * Receive window is amount of space in rcv queue,
	 * but not less than advertised window.
	 */
	{ int win;

	win = sbspace(&so->so_rcv);
	if (win < 0)
		win = 0;
	tp->rcv_wnd = max(win, (int)(tp->rcv_adv - tp->rcv_nxt));
	}

	switch (tp->t_state) {

	/*
	 * If the state is LISTEN then ignore segment if it contains an RST.
	 * If the segment contains an ACK then it is bad and send a RST.
	 * If it does not contain a SYN then it is not interesting; drop it.
	 * Don't bother responding if the destination was a broadcast.
	 * Otherwise initialize tp->rcv_nxt, and tp->irs, select an initial
	 * tp->iss, and send a segment:
	 *     <SEQ=ISS><ACK=RCV_NXT><CTL=SYN,ACK>
	 * Also initialize tp->snd_nxt to tp->iss+1 and tp->snd_una to tp->iss.
	 * Fill in remote peer address fields if not previously specified.
	 * Enter SYN_RECEIVED state, and process any other fields of this
	 * segment in this state.
	 */
	case TCPS_LISTEN: {
		struct mbuf *am;
		register struct sockaddr_in *sin;

#ifdef already_done
		if (tiflags & TH_RST)
			goto drop;
		if (tiflags & TH_ACK)
			goto dropwithreset;
		if ((tiflags & TH_SYN) == 0)
			goto drop;
#endif
		/*
		 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN
		 * in_broadcast() should never return true on a received
		 * packet with M_BCAST not set.
		 */
		if (m->m_flags & (M_BCAST|M_MCAST) ||
		    IN_MULTICAST(ntohl(ti->ti_dst.s_addr)))
			goto drop;

                /* XXX land bug attack fix */
                if (_landAttackSafe)
                    if ((ti->ti_sport == ti->ti_dport) &&
                        (ti->ti_src.s_addr == ti->ti_dst.s_addr))
                        goto drop;
                
		/* XXX */
		am = mBufClGet(M_DONTWAIT, MT_SONAME, CL_SIZE_128, TRUE);
		if (am == NULL)
			goto drop;
		am->m_len = sizeof (struct sockaddr_in);
		sin = mtod(am, struct sockaddr_in *);
		sin->sin_family = AF_INET;
		sin->sin_len = sizeof(*sin);
		sin->sin_addr = ti->ti_src;
		sin->sin_port = ti->ti_sport;
		bzero((caddr_t)sin->sin_zero, sizeof(sin->sin_zero));
		laddr = inp->inp_laddr;
		if (inp->inp_laddr.s_addr == INADDR_ANY)
			inp->inp_laddr = ti->ti_dst;
		if (in_pcbconnect(inp, am)) {
			inp->inp_laddr = laddr;
			(void) m_free(am);
			goto drop;
		}
		(void) m_free(am);
		tp->t_template = tcp_template(tp);
		if (tp->t_template == 0) {
			tp = tcp_drop(tp, ENOBUFS);
			dropsocket = 0;		/* socket is already gone */
			goto drop;
		}
		if (optp)
			tcp_dooptions(tp, optp, optlen, ti,
				&ts_present, &ts_val, &ts_ecr);
		if (iss)
			tp->iss = iss;
		else
			tp->iss = tcp_iss;
		tcp_iss += TCP_ISSINCR/4 + ((0x0000ffff) & (pTcpRandHook() >> 16));
		tp->irs = ti->ti_seq;
		tcp_sendseqinit(tp);
		tcp_rcvseqinit(tp);
		tp->t_flags |= TF_ACKNOW;
		tp->t_state = TCPS_SYN_RECEIVED;
		tp->t_timer[TCPT_KEEP] = tcp_keepinit;
		dropsocket = 0;		/* committed to socket */
		tcpstat.tcps_accepts++;

#ifdef WV_INSTRUMENTATION
#ifdef INCLUDE_WVNET    /* WV_NET_INFO event */
            WV_NET_PORTIN_EVENT_3 (NET_AUX_EVENT, WV_NET_INFO, 29, 11,
                                   ti->ti_sport, ti->ti_dport,
                                   WV_NETEVENT_TCPIN_ACCEPT, WV_NET_RECV, 
                                   so->so_fd, tp->snd_nxt, tp->rcv_nxt)
#endif  /* INCLUDE_WVNET */
#endif

		goto trimthenstep6;
		}

	/*
	 * If the state is SYN_SENT:
	 *	if seg contains an ACK, but not for our SYN, drop the input.
	 *	if seg contains a RST, then drop the connection.
	 *	if seg does not contain SYN, then drop it.
	 * Otherwise this is an acceptable SYN segment
	 *	initialize tp->rcv_nxt and tp->irs
	 *	if seg contains ack then advance tp->snd_una
	 *	if SYN has been acked change to ESTABLISHED else SYN_RCVD state
	 *	arrange for segment to be acked (eventually)
	 *	continue processing rest of data/controls, beginning with URG
	 */
	case TCPS_SYN_SENT:
		if ((tiflags & TH_ACK) &&
		    (SEQ_LEQ(ti->ti_ack, tp->iss) ||
		     SEQ_GT(ti->ti_ack, tp->snd_max)))
			goto dropwithreset;
		if (tiflags & TH_RST) {
			if (tiflags & TH_ACK)
				tp = tcp_drop(tp, ECONNREFUSED);
			goto drop;
		}