tcp_input.c

MIPS处理器的bootloader,龙芯就是用的修改过的PMON2

			goto drop;
		}
	}

	/*
	 * If a SYN is in the window, then this is an
	 * error and we send an RST and drop the connection.
	 */
	if (tiflags & TH_SYN) {
		tp = tcp_drop(tp, ECONNRESET);
		goto dropwithreset;
	}

	/*
	 * If the ACK bit is off we drop the segment and return.
	 */
	if ((tiflags & TH_ACK) == 0) {
		if (tp->t_flags & TF_ACKNOW)
			goto dropafterack;
		else
			goto drop;
	}
	
	/*
	 * Ack processing.
	 */
	switch (tp->t_state) {

	/*
	 * In SYN_RECEIVED state, the ack ACKs our SYN, so enter
	 * ESTABLISHED state and continue processing.
	 * The ACK was checked above.
	 */
	case TCPS_SYN_RECEIVED:
		tcpstat.tcps_connects++;
		soisconnected(so);
		tp->t_state = TCPS_ESTABLISHED;
		/* Do window scaling? */
		if ((tp->t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) ==
			(TF_RCVD_SCALE|TF_REQ_SCALE)) {
			tp->snd_scale = tp->requested_s_scale;
			tp->rcv_scale = tp->request_r_scale;
		}
		(void) tcp_reass(tp, (struct tcphdr *)0, (struct mbuf *)0,
				 &tlen);
		tp->snd_wl1 = th->th_seq - 1;
		/* fall into ... */

	/*
	 * In ESTABLISHED state: drop duplicate ACKs; ACK out of range
	 * ACKs.  If the ack is in the range
	 *	tp->snd_una < ti->ti_ack <= tp->snd_max
	 * then advance tp->snd_una to ti->ti_ack and drop
	 * data from the retransmission queue.  If this ACK reflects
	 * more up to date window information we update our window information.
	 */
	case TCPS_ESTABLISHED:
	case TCPS_FIN_WAIT_1:
	case TCPS_FIN_WAIT_2:
	case TCPS_CLOSE_WAIT:
	case TCPS_CLOSING:
	case TCPS_LAST_ACK:
	case TCPS_TIME_WAIT:
		if (SEQ_LEQ(th->th_ack, tp->snd_una)) {
			/*
			 * Duplicate/old ACK processing.
			 * Increments t_dupacks:
			 *	Pure duplicate (same seq/ack/window, no data)
			 * Doesn't affect t_dupacks:
			 *	Data packets.
			 *	Normal window updates (window opens)
			 * Resets t_dupacks:
			 *	New data ACKed.
			 *	Window shrinks
			 *	Old ACK
			 */
			if (tlen)
				break;
			/*
			 * If we get an old ACK, there is probably packet
			 * reordering going on.  Be conservative and reset
			 * t_dupacks so that we are less agressive in
			 * doing a fast retransmit.
			 */
			if (th->th_ack != tp->snd_una) {
				tp->t_dupacks = 0;
				break;
			}
			if (tiwin == tp->snd_wnd) {
				tcpstat.tcps_rcvdupack++;
				/*
				 * If we have outstanding data (other than
				 * a window probe), this is a completely
				 * duplicate ack (ie, window info didn't
				 * change), the ack is the biggest we've
				 * seen and we've seen exactly our rexmt
				 * threshhold of them, assume a packet
				 * has been dropped and retransmit it.
				 * Kludge snd_nxt & the congestion
				 * window so we send only this one
				 * packet.
				 *
				 * We know we're losing at the current
				 * window size so do congestion avoidance
				 * (set ssthresh to half the current window
				 * and pull our congestion window back to
				 * the new ssthresh).
				 *
				 * Dup acks mean that packets have left the
				 * network (they're now cached at the receiver) 
				 * so bump cwnd by the amount in the receiver
				 * to keep a constant cwnd packets in the
				 * network.
				 */
				if (tp->t_timer[TCPT_REXMT] == 0)
					tp->t_dupacks = 0;
#if defined(TCP_SACK) && defined(TCP_FACK)
				/* 
				 * In FACK, can enter fast rec. if the receiver
				 * reports a reass. queue longer than 3 segs.
				 */
				else if (++tp->t_dupacks == tcprexmtthresh ||
				    ((SEQ_GT(tp->snd_fack, tcprexmtthresh * 
				    tp->t_maxseg + tp->snd_una)) &&
				    SEQ_GT(tp->snd_una, tp->snd_last))) {
#else
				else if (++tp->t_dupacks == tcprexmtthresh) {
#endif /* TCP_FACK */
					tcp_seq onxt = tp->snd_nxt;
					u_long win =
					    ulmin(tp->snd_wnd, tp->snd_cwnd) /
						2 / tp->t_maxseg;

#if defined(TCP_SACK) 
					if (SEQ_LT(th->th_ack, tp->snd_last)){
					    	/* 
						 * False fast retx after 
						 * timeout.  Do not cut window.
						 */
						tp->snd_cwnd += tp->t_maxseg;
						tp->t_dupacks = 0;
						(void) tcp_output(tp); 
						goto drop;
					}
#endif
					if (win < 2)
						win = 2;
					tp->snd_ssthresh = win * tp->t_maxseg;
#if defined(TCP_SACK)
					tp->snd_last = tp->snd_max;
#endif
#ifdef TCP_SACK
                    			if (!tp->sack_disable) {
						tp->t_timer[TCPT_REXMT] = 0;
						tp->t_rtt = 0;
						tcpstat.tcps_sndrexmitfast++;
#if defined(TCP_SACK) && defined(TCP_FACK) 
						(void) tcp_output(tp);
						/*
						 * During FR, snd_cwnd is held
						 * constant for FACK.
						 */
						tp->snd_cwnd = tp->snd_ssthresh;
						tp->t_dupacks = tcprexmtthresh;
#else
						/* 
						 * tcp_output() will send
						 * oldest SACK-eligible rtx.
						 */
						(void) tcp_output(tp);
						tp->snd_cwnd = tp->snd_ssthresh+
					           tp->t_maxseg * tp->t_dupacks;
#endif /* TCP_FACK */
						goto drop;
					}
#endif /* TCP_SACK */
					tp->t_timer[TCPT_REXMT] = 0;
					tp->t_rtt = 0;
					tp->snd_nxt = th->th_ack;
					tp->snd_cwnd = tp->t_maxseg;
					tcpstat.tcps_sndrexmitfast++;
					(void) tcp_output(tp);

					tp->snd_cwnd = tp->snd_ssthresh +
					    tp->t_maxseg * tp->t_dupacks;
					if (SEQ_GT(onxt, tp->snd_nxt))
						tp->snd_nxt = onxt;
					goto drop;
				} else if (tp->t_dupacks > tcprexmtthresh) {
#if defined(TCP_SACK) && defined(TCP_FACK)
					/* 
					 * while (awnd < cwnd) 
					 *         sendsomething(); 
					 */
					if (!tp->sack_disable) {
						if (tp->snd_awnd < tp->snd_cwnd)
							tcp_output(tp);
						goto drop;
					}
#endif /* TCP_FACK */
					tp->snd_cwnd += tp->t_maxseg;
					(void) tcp_output(tp);
					goto drop;
				}
			} else if (tiwin < tp->snd_wnd) {
				/*
				 * The window was retracted!  Previous dup
				 * ACKs may have been due to packets arriving
				 * after the shrunken window, not a missing
				 * packet, so play it safe and reset t_dupacks
				 */
				tp->t_dupacks = 0;
			}
			break;
		}
		/*
		 * If the congestion window was inflated to account
		 * for the other side's cached packets, retract it.
		 */
#if defined(TCP_SACK)
		if (!tp->sack_disable) {
			if (tp->t_dupacks >= tcprexmtthresh) {
				/* Check for a partial ACK */
				if (tcp_sack_partialack(tp, th)) {
#if defined(TCP_SACK) && defined(TCP_FACK)
					/* Force call to tcp_output */
					if (tp->snd_awnd < tp->snd_cwnd) 
						needoutput = 1;
#else
					tp->snd_cwnd += tp->t_maxseg;
					needoutput = 1;
#endif /* TCP_FACK */
				} else {
					/* Out of fast recovery */
					tp->snd_cwnd = tp->snd_ssthresh;
					if (tcp_seq_subtract(tp->snd_max, 
					    th->th_ack) < tp->snd_ssthresh)
						tp->snd_cwnd = 
						   tcp_seq_subtract(tp->snd_max,
					           th->th_ack) + tp->t_maxseg;
					tp->t_dupacks = 0;
#if defined(TCP_SACK) && defined(TCP_FACK)
					if (SEQ_GT(th->th_ack, tp->snd_fack))
						tp->snd_fack = th->th_ack;
#endif /* TCP_FACK */
				}
			} 
		} else {
			if (tp->t_dupacks >= tcprexmtthresh && 
			    !tcp_newreno(tp, th)) {
				/* Out of fast recovery */
				tp->snd_cwnd = tp->snd_ssthresh;
				if (tcp_seq_subtract(tp->snd_max, th->th_ack) <
			  	    tp->snd_ssthresh)
					tp->snd_cwnd = 
					    tcp_seq_subtract(tp->snd_max,
					    th->th_ack) + tp->t_maxseg;
				tp->t_dupacks = 0;
			}
		}
		if (tp->t_dupacks < tcprexmtthresh)
			tp->t_dupacks = 0;
#else /* else no TCP_SACK */
		if (tp->t_dupacks >= tcprexmtthresh &&
		    tp->snd_cwnd > tp->snd_ssthresh)
			tp->snd_cwnd = tp->snd_ssthresh;
		tp->t_dupacks = 0;
#endif
		if (SEQ_GT(th->th_ack, tp->snd_max)) {
			tcpstat.tcps_rcvacktoomuch++;
			goto dropafterack;
		}
		acked = th->th_ack - tp->snd_una;
		tcpstat.tcps_rcvackpack++;
		tcpstat.tcps_rcvackbyte += acked;

		/*
		 * If we have a timestamp reply, update smoothed
		 * round trip time.  If no timestamp is present but
		 * transmit timer is running and timed sequence
		 * number was acked, update smoothed round trip time.
		 * Since we now have an rtt measurement, cancel the
		 * timer backoff (cf., Phil Karn's retransmit alg.).
		 * Recompute the initial retransmit timer.
		 */
		if (ts_present)
			tcp_xmit_timer(tp, tcp_now-ts_ecr+1);
		else if (tp->t_rtt && SEQ_GT(th->th_ack, tp->t_rtseq))
			tcp_xmit_timer(tp,tp->t_rtt);

		/*
		 * If all outstanding data is acked, stop retransmit
		 * timer and remember to restart (more output or persist).
		 * If there is more data to be acked, restart retransmit
		 * timer, using current (possibly backed-off) value.
		 */
		if (th->th_ack == tp->snd_max) {
			tp->t_timer[TCPT_REXMT] = 0;
			needoutput = 1;
		} else if (tp->t_timer[TCPT_PERSIST] == 0)
			tp->t_timer[TCPT_REXMT] = tp->t_rxtcur;
		/*
		 * When new data is acked, open the congestion window.
		 * If the window gives us less than ssthresh packets
		 * in flight, open exponentially (maxseg per packet).
		 * Otherwise open linearly: maxseg per window
		 * (maxseg^2 / cwnd per packet).
		 */
		{
		register u_int cw = tp->snd_cwnd;
		register u_int incr = tp->t_maxseg;

		if (cw > tp->snd_ssthresh)
			incr = incr * incr / cw;
#if defined (TCP_SACK)
		if (SEQ_GEQ(th->th_ack, tp->snd_last)) 
#endif
		tp->snd_cwnd = min(cw + incr, TCP_MAXWIN<<tp->snd_scale);
		}
		ND6_HINT(tp);
		if (acked > so->so_snd.sb_cc) {
			tp->snd_wnd -= so->so_snd.sb_cc;
			sbdrop(&so->so_snd, (int)so->so_snd.sb_cc);
			ourfinisacked = 1;
		} else {
			sbdrop(&so->so_snd, acked);
			tp->snd_wnd -= acked;
			ourfinisacked = 0;
		}
		if (sb_notify(&so->so_snd))
			sowwakeup(so);
		tp->snd_una = th->th_ack;
		if (SEQ_LT(tp->snd_nxt, tp->snd_una))
			tp->snd_nxt = tp->snd_una;
#if defined (TCP_SACK) && defined (TCP_FACK)
		if (SEQ_GT(tp->snd_una, tp->snd_fack))
			tp->snd_fack = tp->snd_una;
#endif

		switch (tp->t_state) {

		/*
		 * In FIN_WAIT_1 STATE in addition to the processing
		 * for the ESTABLISHED state if our FIN is now acknowledged
		 * then enter FIN_WAIT_2.
		 */
		case TCPS_FIN_WAIT_1:
			if (ourfinisacked) {
				/*
				 * If we can't receive any more
				 * data, then closing user can proceed.
				 * Starting the timer is contrary to the
				 * specification, but if we don't get a FIN
				 * we'll hang forever.
				 */
				if (so->so_state & SS_CANTRCVMORE) {
					soisdisconnected(so);
					tp->t_timer[TCPT_2MSL] = tcp_maxidle;
				}
				tp->t_state = TCPS_FIN_WAIT_2;
			}
			break;

		/*
		 * In CLOSING STATE in addition to the processing for
		 * the ESTABLISHED state if the ACK acknowledges our FIN
		 * then enter the TIME-WAIT state, otherwise ignore
		 * the segment.
		 */
		case TCPS_CLOSING:
			if (ourfinisacked) {
				tp->t_state = TCPS_TIME_WAIT;
				tcp_canceltimers(tp);
				tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
				soisdisconnected(so);
			}
			break;

		/*
		 * In LAST_ACK, we may still be waiting for data to drain
		 * and/or to be acked, as well as for the ack of our FIN.
		 * If our FIN is now acknowledged, delete the TCB,
		 * enter the closed state and return.
		 */
		case TCPS_LAST_ACK:
			if (ourfinisacked) {
				tp = tcp_close(tp);
				goto drop;
			}
			break;

		/*
		 * In TIME_WAIT state the only thing that should arrive
		 * is a retransmission of the remote FIN.  Acknowledge
		 * it and restart the finack timer.
		 */
		case TCPS_TIME_WAIT:
			tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
			goto dropafterack;
		}
	}

step6:
	/*
	 * Update window information.
	 * Don't look at window if no ACK: TAC's send garbage on first SYN.
	 */
	if ((tiflags & TH_ACK) && (SEQ_LT(tp->snd_wl1, th->th_seq) ||
	    (tp->snd_wl1 == th->th_seq && SEQ_LT(tp->snd_wl2, th->th_ack)) ||
	    (tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd))) {
		/* keep track of pure window updates */
		if (tlen == 0 &&
		    tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd)
			tcpstat.tcps_rcvwinupd++;
		tp->snd_wnd = tiwin;
		tp->snd_wl1 = th->th_seq;
		tp->snd_wl2 = th->th_ack;
		if (tp->snd_wnd > tp->max_sndwnd)
			tp->max_sndwnd = tp->snd_wnd;
		needoutput = 1;
	}

	/*
	 * Process segments with URG.
	 */
	if ((tiflags & TH_URG) && th->th_urp &&
	    TCPS_HAVERCVDFIN(tp->t_state) == 0) {
		/*
		 * This is a kludge, but if we receive and accept
		 * random urgent pointers, we'll crash in
		 * soreceive.  It's hard to imagine someone
		 * actually wanting to send this much urgent data.
		 */
		if (th->th_urp + so->so_rcv.sb_cc > sb_max) {
			th->th_urp = 0;			/* XXX */
			tiflags &= ~TH_URG;		/* XXX */
			goto dodata;			/* XXX */
		}
		/*
		 * If this segment advances the known urgent pointer,
		 * then mark the data stream.  This should not happen
		 * in CLOSE_WAIT, CLOSING, LAST_ACK or TIME_WAIT STATES since
		 * a FIN has been received from the remote side. 
		 * In these states we ignore the URG.
		 *
		 * 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 as the original 
		 * spec states (in one of two places).
		 */
		if (SEQ_GT(th->th_seq+th->th_urp, tp->rcv_up)) {
			tp->rcv_up = th->th_seq + th->th_urp;
			so->so_oobmark = so->so_rcv.sb_cc +
			    (tp->rcv_up - tp->rcv_nxt) - 1;
			if (so->so_oobmark == 0)
				so->so_state |= SS_RCVATMARK;
			sohasoutofband(so);
			tp->t_oobflags &= ~(TCPOOB_HAVEDATA | TCPOOB_HADDATA);
		}
		/*
		 * Remove out of band data so doesn't get presented to user.
		 * This can happen independent of advancing the URG pointer,
		 * but if two URG's are pending at once, some out-of-band
		 * data may creep in... ick.
		 */
		if (th->th_urp <= (u_int16_t) tlen
#ifdef SO_OOBINLINE
		     && (so->so_options & SO_OOBINLINE) == 0
#endif
		     )
		        tcp_pulloutofband(so, th->th_urp, m, hdroptlen);
	} else