tcp_input.c

qemu虚拟机代码

	 * Process segments with URG.
	 */
	if ((tiflags & TH_URG) && ti->ti_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 (ti->ti_urp + so->so_rcv.sb_cc > so->so_rcv.sb_datalen) {
			ti->ti_urp = 0;
			tiflags &= ~TH_URG;
			goto dodata;
		}
		/*
		 * 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(ti->ti_seq+ti->ti_urp, tp->rcv_up)) {
			tp->rcv_up = ti->ti_seq + ti->ti_urp;
			so->so_urgc =  so->so_rcv.sb_cc +
				(tp->rcv_up - tp->rcv_nxt); /* -1; */
			tp->rcv_up = ti->ti_seq + ti->ti_urp;
	 
		}
	} else
		/*
		 * If no out of band data is expected,
		 * pull receive urgent pointer along
		 * with the receive window.
		 */
		if (SEQ_GT(tp->rcv_nxt, tp->rcv_up))
			tp->rcv_up = tp->rcv_nxt;
dodata:

	/*
	 * Process the segment text, merging it into the TCP sequencing queue,
	 * and arranging for acknowledgment of receipt if necessary.
	 * This process logically involves adjusting tp->rcv_wnd as data
	 * is presented to the user (this happens in tcp_usrreq.c,
	 * case PRU_RCVD).  If a FIN has already been received on this
	 * connection then we just ignore the text.
	 */
	if ((ti->ti_len || (tiflags&TH_FIN)) &&
	    TCPS_HAVERCVDFIN(tp->t_state) == 0) {
		TCP_REASS(tp, ti, m, so, tiflags);
		/*
		 * Note the amount of data that peer has sent into
		 * our window, in order to estimate the sender's
		 * buffer size.
		 */
		len = so->so_rcv.sb_datalen - (tp->rcv_adv - tp->rcv_nxt);
	} else {
		m_free(m);
		tiflags &= ~TH_FIN;
	}

	/*
	 * If FIN is received ACK the FIN and let the user know
	 * that the connection is closing.
	 */
	if (tiflags & TH_FIN) {
		if (TCPS_HAVERCVDFIN(tp->t_state) == 0) {
			/*
			 * If we receive a FIN we can't send more data,
			 * set it SS_FDRAIN
                         * Shutdown the socket if there is no rx data in the
			 * buffer.
			 * soread() is called on completion of shutdown() and
			 * will got to TCPS_LAST_ACK, and use tcp_output()
			 * to send the FIN.
			 */
/*			sofcantrcvmore(so); */
			sofwdrain(so);
			
			tp->t_flags |= TF_ACKNOW;
			tp->rcv_nxt++;
		}
		switch (tp->t_state) {

	 	/*
		 * In SYN_RECEIVED and ESTABLISHED STATES
		 * enter the CLOSE_WAIT state.
		 */
		case TCPS_SYN_RECEIVED:
		case TCPS_ESTABLISHED:
		  if(so->so_emu == EMU_CTL)        /* no shutdown on socket */
		    tp->t_state = TCPS_LAST_ACK;
		  else 
		    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;
			soisfdisconnected(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 this is a small packet, then ACK now - with Nagel
	 *      congestion avoidance sender won't send more until
	 *      he gets an ACK.
	 * 
	 * See above.
	 */
/*	if (ti->ti_len && (unsigned)ti->ti_len < tp->t_maxseg) {
 */
/*	if ((ti->ti_len && (unsigned)ti->ti_len < tp->t_maxseg &&
 *		(so->so_iptos & IPTOS_LOWDELAY) == 0) ||
 *	       ((so->so_iptos & IPTOS_LOWDELAY) &&
 *	       ((struct tcpiphdr_2 *)ti)->first_char == (char)27)) {
 */
	if (ti->ti_len && (unsigned)ti->ti_len <= 5 &&
	    ((struct tcpiphdr_2 *)ti)->first_char == (char)27) {
		tp->t_flags |= TF_ACKNOW;
	}

	/*
	 * 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:
	/* reuses m if m!=NULL, m_free() unnecessary */
	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);
	}

	return;

drop:
	/*
	 * Drop space held by incoming segment and return.
	 */
	m_free(m);

	return;
}

 /* , ts_present, ts_val, ts_ecr) */
/*	int *ts_present;
 *	u_int32_t *ts_val, *ts_ecr;
 */
void
tcp_dooptions(tp, cp, cnt, ti)
	struct tcpcb *tp;
	u_char *cp;
	int cnt;
	struct tcpiphdr *ti;
{
	u_int16_t mss;
	int opt, optlen;

	DEBUG_CALL("tcp_dooptions");
	DEBUG_ARGS((dfd," tp = %lx  cnt=%i \n", (long )tp, cnt));

	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;
			memcpy((char *) &mss, (char *) cp + 2, 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;
 *			memcpy((char *) ts_val, (char *)cp + 2, sizeof(*ts_val));
 *			NTOHL(*ts_val);
 *			memcpy((char *) ts_ecr, (char *)cp + 6, 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.
 */

#ifdef notdef

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;
			memcpy(sp, cp+1, (unsigned)(m->m_len - cnt - 1));
			m->m_len--;
			return;
		}
		cnt -= m->m_len;
		m = m->m_next; /* XXX WRONG! Fix it! */
		if (m == 0)
			break;
	}
	panic("tcp_pulloutofband");
}

#endif /* notdef */

/*
 * Collect new round-trip time estimate
 * and update averages and current timeout.
 */

void
tcp_xmit_timer(tp, rtt)
	register struct tcpcb *tp;
	int rtt;
{
	register short delta;

	DEBUG_CALL("tcp_xmit_timer");
	DEBUG_ARG("tp = %lx", (long)tp);
	DEBUG_ARG("rtt = %d", rtt);
	
	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),
	    (short)tp->t_rttmin, TCPTV_REXMTMAX); /* XXX */
	
	/*
	 * 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 bigger than
 * an mbuf cluster (MCLBYTES), round down to nearest multiple of MCLBYTES
 * to utilize large mbufs.  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 socket *so = tp->t_socket;
	int mss;
	
	DEBUG_CALL("tcp_mss");
	DEBUG_ARG("tp = %lx", (long)tp);
	DEBUG_ARG("offer = %d", offer);
	
	mss = min(if_mtu, if_mru) - sizeof(struct tcpiphdr);
	if (offer)
		mss = min(mss, offer);
	mss = max(mss, 32);
	if (mss < tp->t_maxseg || offer != 0)
	   tp->t_maxseg = mss;
	
	tp->snd_cwnd = mss;
	
	sbreserve(&so->so_snd, tcp_sndspace+((tcp_sndspace%mss)?(mss-(tcp_sndspace%mss)):0));
	sbreserve(&so->so_rcv, tcp_rcvspace+((tcp_rcvspace%mss)?(mss-(tcp_rcvspace%mss)):0));
	
	DEBUG_MISC((dfd, " returning mss = %d\n", mss));
	
	return mss;
}