tcp_output.c

来自「eCos操作系统源码」· C语言 代码 · 共 949 行 · 第 1/2 页

		 */
		case TH_SYN:
			opt[optlen++] = TCPOPT_NOP;
			opt[optlen++] = TCPOPT_NOP;
			opt[optlen++] = tp->t_flags & TF_SENDCCNEW ?
						TCPOPT_CCNEW : TCPOPT_CC;
			opt[optlen++] = TCPOLEN_CC;
			*(u_int32_t *)&opt[optlen] = htonl(tp->cc_send);
 			optlen += 4;
			break;

		/*
		 * This is a SYN,ACK; send CC and CC.echo if we received
		 * CC from our peer.
		 */
		case (TH_SYN|TH_ACK):
			if (tp->t_flags & TF_RCVD_CC) {
				opt[optlen++] = TCPOPT_NOP;
				opt[optlen++] = TCPOPT_NOP;
				opt[optlen++] = TCPOPT_CC;
				opt[optlen++] = TCPOLEN_CC;
				*(u_int32_t *)&opt[optlen] =
					htonl(tp->cc_send);
				optlen += 4;
				opt[optlen++] = TCPOPT_NOP;
				opt[optlen++] = TCPOPT_NOP;
				opt[optlen++] = TCPOPT_CCECHO;
				opt[optlen++] = TCPOLEN_CC;
				*(u_int32_t *)&opt[optlen] =
					htonl(tp->cc_recv);
				optlen += 4;
			}
			break;
		}
 	}

 	hdrlen += optlen;

#ifdef INET6
	if (isipv6)
		ipoptlen = ip6_optlen(tp->t_inpcb);
	else
#endif
      {
	if (tp->t_inpcb->inp_options) {
		ipoptlen = tp->t_inpcb->inp_options->m_len -
				offsetof(struct ipoption, ipopt_list);
	} else {
		ipoptlen = 0;
	}
      }
#ifdef IPSEC
	ipoptlen += ipsec_hdrsiz_tcp(tp);
#endif

	/*
	 * Adjust data length if insertion of options will
	 * bump the packet length beyond the t_maxopd length.
	 * Clear the FIN bit because we cut off the tail of
	 * the segment.
	 */
	if (len + optlen + ipoptlen > tp->t_maxopd) {
		/*
		 * If there is still more to send, don't close the connection.
		 */
		flags &= ~TH_FIN;
		len = tp->t_maxopd - optlen - ipoptlen;
		sendalot = 1;
	}

/*#ifdef DIAGNOSTIC*/
#ifdef INET6
 	if (max_linkhdr + hdrlen > MCLBYTES)
		panic("tcphdr too big");
#else
 	if (max_linkhdr + hdrlen > MHLEN)
		panic("tcphdr too big");
#endif
/*#endif*/

	/*
	 * Grab a header mbuf, attaching a copy of data to
	 * be transmitted, and initialize the header from
	 * the template for sends on this connection.
	 */
	if (len) {
		if (tp->t_force && len == 1)
			tcpstat.tcps_sndprobe++;
		else if (SEQ_LT(tp->snd_nxt, tp->snd_max)) {
			tcpstat.tcps_sndrexmitpack++;
			tcpstat.tcps_sndrexmitbyte += len;
		} else {
			tcpstat.tcps_sndpack++;
			tcpstat.tcps_sndbyte += len;
		}
#ifdef notyet
		if ((m = m_copypack(so->so_snd.sb_mb, off,
		    (int)len, max_linkhdr + hdrlen)) == 0) {
			error = ENOBUFS;
			goto out;
		}
		/*
		 * m_copypack left space for our hdr; use it.
		 */
		m->m_len += hdrlen;
		m->m_data -= hdrlen;
#else
		MGETHDR(m, M_DONTWAIT, MT_HEADER);
		if (m == NULL) {
			error = ENOBUFS;
			goto out;
		}
#ifdef INET6
		if (MHLEN < hdrlen + max_linkhdr) {
			MCLGET(m, M_DONTWAIT);
			if ((m->m_flags & M_EXT) == 0) {
				m_freem(m);
				error = ENOBUFS;
				goto out;
			}
		}
#endif
		m->m_data += max_linkhdr;
		m->m_len = hdrlen;
		if (len <= MHLEN - hdrlen - max_linkhdr) {
			m_copydata(so->so_snd.sb_mb, off, (int) len,
			    mtod(m, caddr_t) + hdrlen);
			m->m_len += len;
		} else {
			m->m_next = m_copy(so->so_snd.sb_mb, off, (int) len);
			if (m->m_next == 0) {
				(void) m_free(m);
				error = ENOBUFS;
				goto out;
			}
		}
#endif
		/*
		 * If we're sending everything we've got, set PUSH.
		 * (This will keep happy those implementations which only
		 * give data to the user when a buffer fills or
		 * a PUSH comes in.)
		 */
		if (off + len == so->so_snd.sb_cc)
			flags |= TH_PUSH;
	} else {
		if (tp->t_flags & TF_ACKNOW)
			tcpstat.tcps_sndacks++;
		else if (flags & (TH_SYN|TH_FIN|TH_RST))
			tcpstat.tcps_sndctrl++;
		else if (SEQ_GT(tp->snd_up, tp->snd_una))
			tcpstat.tcps_sndurg++;
		else
			tcpstat.tcps_sndwinup++;

		MGETHDR(m, M_DONTWAIT, MT_HEADER);
		if (m == NULL) {
			error = ENOBUFS;
			goto out;
		}
#ifdef INET6
		if (isipv6 && (MHLEN < hdrlen + max_linkhdr) &&
		    MHLEN >= hdrlen) {
			MH_ALIGN(m, hdrlen);
		} else
#endif
		m->m_data += max_linkhdr;
		m->m_len = hdrlen;
	}
	m->m_pkthdr.rcvif = (struct ifnet *)0;
#ifdef INET6
	if (isipv6) {
		ip6 = mtod(m, struct ip6_hdr *);
		th = (struct tcphdr *)(ip6 + 1);
		tcp_fillheaders(tp, ip6, th);
	} else
#endif /* INET6 */
      {
	ip = mtod(m, struct ip *);
	ipov = (struct ipovly *)ip;
	th = (struct tcphdr *)(ip + 1);
	/* this picks up the pseudo header (w/o the length) */
	tcp_fillheaders(tp, ip, th);
      }

	/*
	 * Fill in fields, remembering maximum advertised
	 * window for use in delaying messages about window sizes.
	 * If resending a FIN, be sure not to use a new sequence number.
	 */
	if (flags & TH_FIN && tp->t_flags & TF_SENTFIN &&
	    tp->snd_nxt == tp->snd_max)
		tp->snd_nxt--;
	/*
	 * If we are doing retransmissions, then snd_nxt will
	 * not reflect the first unsent octet.  For ACK only
	 * packets, we do not want the sequence number of the
	 * retransmitted packet, we want the sequence number
	 * of the next unsent octet.  So, if there is no data
	 * (and no SYN or FIN), use snd_max instead of snd_nxt
	 * when filling in ti_seq.  But if we are in persist
	 * state, snd_max might reflect one byte beyond the
	 * right edge of the window, so use snd_nxt in that
	 * case, since we know we aren't doing a retransmission.
	 * (retransmit and persist are mutually exclusive...)
	 */
	if (len || (flags & (TH_SYN|TH_FIN)) 
	    || callout_active(tp->tt_persist))
		th->th_seq = htonl(tp->snd_nxt);
	else
		th->th_seq = htonl(tp->snd_max);
	th->th_ack = htonl(tp->rcv_nxt);
	if (optlen) {
		bcopy(opt, th + 1, optlen);
		th->th_off = (sizeof (struct tcphdr) + optlen) >> 2;
	}
	th->th_flags = flags;
	/*
	 * Calculate receive window.  Don't shrink window,
	 * but avoid silly window syndrome.
	 */
	if (win < (long)(so->so_rcv.sb_hiwat / 4) && win < (long)tp->t_maxseg)
		win = 0;
	if (win < (long)(tp->rcv_adv - tp->rcv_nxt))
		win = (long)(tp->rcv_adv - tp->rcv_nxt);
	if (win > (long)TCP_MAXWIN << tp->rcv_scale)
		win = (long)TCP_MAXWIN << tp->rcv_scale;
	th->th_win = htons((u_short) (win>>tp->rcv_scale));
	if (SEQ_GT(tp->snd_up, tp->snd_nxt)) {
		th->th_urp = htons((u_short)(tp->snd_up - tp->snd_nxt));
		th->th_flags |= TH_URG;
	} else
		/*
		 * If no urgent pointer to send, then we pull
		 * the urgent pointer to the left edge of the send window
		 * so that it doesn't drift into the send window on sequence
		 * number wraparound.
		 */
		tp->snd_up = tp->snd_una;		/* drag it along */

	/*
	 * Put TCP length in extended header, and then
	 * checksum extended header and data.
	 */
	m->m_pkthdr.len = hdrlen + len; /* in6_cksum() need this */
#ifdef INET6
	if (isipv6)
		/*
		 * ip6_plen is not need to be filled now, and will be filled
		 * in ip6_output.
		 */
		th->th_sum = in6_cksum(m, IPPROTO_TCP, sizeof(struct ip6_hdr),
				       sizeof(struct tcphdr) + optlen + len);
	else
#endif /* INET6 */
      {
	m->m_pkthdr.csum_flags = CSUM_TCP;
	m->m_pkthdr.csum_data = offsetof(struct tcphdr, th_sum);
	if (len + optlen)
		th->th_sum = in_addword(th->th_sum, 
		    htons((u_short)(optlen + len)));

	/* IP version must be set here for ipv4/ipv6 checking later */
      }

	/*
	 * In transmit state, time the transmission and arrange for
	 * the retransmit.  In persist state, just set snd_max.
	 */
	if (tp->t_force == 0 || !callout_active(tp->tt_persist)) {
		tcp_seq startseq = tp->snd_nxt;

		/*
		 * Advance snd_nxt over sequence space of this segment.
		 */
		if (flags & (TH_SYN|TH_FIN)) {
			if (flags & TH_SYN)
				tp->snd_nxt++;
			if (flags & TH_FIN) {
				tp->snd_nxt++;
				tp->t_flags |= TF_SENTFIN;
			}
		}
		tp->snd_nxt += len;
		if (SEQ_GT(tp->snd_nxt, tp->snd_max)) {
			tp->snd_max = tp->snd_nxt;
			/*
			 * Time this transmission if not a retransmission and
			 * not currently timing anything.
			 */
			if (tp->t_rtttime == 0) {
				tp->t_rtttime = ticks;
				tp->t_rtseq = startseq;
				tcpstat.tcps_segstimed++;
			}
		}

		/*
		 * Set retransmit timer if not currently set,
		 * and not doing an ack or a keep-alive probe.
		 * Initial value for retransmit timer is smoothed
		 * round-trip time + 2 * round-trip time variance.
		 * Initialize shift counter which is used for backoff
		 * of retransmit time.
		 */
		if (!callout_active(tp->tt_rexmt) &&
		    tp->snd_nxt != tp->snd_una) {
			if (callout_active(tp->tt_persist)) {
				callout_stop(tp->tt_persist);
				tp->t_rxtshift = 0;
			}
			callout_reset(tp->tt_rexmt, tp->t_rxtcur,
				      tcp_timer_rexmt, tp);
		}
	} else
		if (SEQ_GT(tp->snd_nxt + len, tp->snd_max))
			tp->snd_max = tp->snd_nxt + len;

#ifdef TCPDEBUG
	/*
	 * Trace.
	 */
	if (so->so_options & SO_DEBUG)
		tcp_trace(TA_OUTPUT, tp->t_state, tp, mtod(m, void *), th, 0);
#endif

	/*
	 * Fill in IP length and desired time to live and
	 * send to IP level.  There should be a better way
	 * to handle ttl and tos; we could keep them in
	 * the template, but need a way to checksum without them.
	 */
	/*
	 * m->m_pkthdr.len should have been set before cksum calcuration,
	 * because in6_cksum() need it.
	 */
#ifdef INET6
	if (isipv6) {
		/*
		 * we separately set hoplimit for every segment, since the
		 * user might want to change the value via setsockopt.
		 * Also, desired default hop limit might be changed via
		 * Neighbor Discovery.
		 */
		ip6->ip6_hlim = in6_selecthlim(tp->t_inpcb,
					       tp->t_inpcb->in6p_route.ro_rt ?
					       tp->t_inpcb->in6p_route.ro_rt->rt_ifp
					       : NULL);

		/* TODO: IPv6 IP6TOS_ECT bit on */
#ifdef IPSEC
		if (ipsec_setsocket(m, so) != 0) {
			m_freem(m);
			error = ENOBUFS;
			goto out;
		}
#endif /*IPSEC*/
		error = ip6_output(m,
			    tp->t_inpcb->in6p_outputopts,
			    &tp->t_inpcb->in6p_route,
			    (so->so_options & SO_DONTROUTE), NULL, NULL);
	} else
#endif /* INET6 */
    {
	struct rtentry *rt;
	ip->ip_len = m->m_pkthdr.len;
#ifdef INET6
 	if (INP_CHECK_SOCKAF(so, AF_INET6))
 		ip->ip_ttl = in6_selecthlim(tp->t_inpcb,
 					    tp->t_inpcb->in6p_route.ro_rt ?
 					    tp->t_inpcb->in6p_route.ro_rt->rt_ifp
 					    : NULL);
 	else
#endif /* INET6 */
	ip->ip_ttl = tp->t_inpcb->inp_ip_ttl;	/* XXX */
	ip->ip_tos = tp->t_inpcb->inp_ip_tos;	/* XXX */
	/*
	 * See if we should do MTU discovery.  We do it only if the following
	 * are true:
	 *	1) we have a valid route to the destination
	 *	2) the MTU is not locked (if it is, then discovery has been
	 *	   disabled)
	 */
	if (path_mtu_discovery
	    && (rt = tp->t_inpcb->inp_route.ro_rt)
	    && rt->rt_flags & RTF_UP
	    && !(rt->rt_rmx.rmx_locks & RTV_MTU)) {
		ip->ip_off |= IP_DF;
	}
#ifdef IPSEC
 	ipsec_setsocket(m, so);
#endif /*IPSEC*/
	error = ip_output(m, tp->t_inpcb->inp_options, &tp->t_inpcb->inp_route,
	    (so->so_options & SO_DONTROUTE), 0);
    }
	if (error) {

		/*
		 * We know that the packet was lost, so back out the
		 * sequence number advance, if any.
		 */
		if (tp->t_force == 0 || !callout_active(tp->tt_persist)) {
			/*
			 * No need to check for TH_FIN here because
			 * the TF_SENTFIN flag handles that case.
			 */
			if ((flags & TH_SYN) == 0)
				tp->snd_nxt -= len;
		}

out:
		if (error == ENOBUFS) {
	                if (!callout_active(tp->tt_rexmt) &&
                            !callout_active(tp->tt_persist))
	                        callout_reset(tp->tt_rexmt, tp->t_rxtcur,
                                      tcp_timer_rexmt, tp);
			tcp_quench(tp->t_inpcb, 0);
			return (0);
		}
		if (error == EMSGSIZE) {
			/*
			 * ip_output() will have already fixed the route
			 * for us.  tcp_mtudisc() will, as its last action,
			 * initiate retransmission, so it is important to
			 * not do so here.
			 */
			tcp_mtudisc(tp->t_inpcb, 0);
			return 0;
		}
		if ((error == EHOSTUNREACH || error == ENETDOWN)
		    && TCPS_HAVERCVDSYN(tp->t_state)) {
			tp->t_softerror = error;
			return (0);
		}
		return (error);
	}
	tcpstat.tcps_sndtotal++;

	/*
	 * Data sent (as far as we can tell).
	 * If this advertises a larger window than any other segment,
	 * then remember the size of the advertised window.
	 * Any pending ACK has now been sent.
	 */
	if (win > 0 && SEQ_GT(tp->rcv_nxt+win, tp->rcv_adv))
		tp->rcv_adv = tp->rcv_nxt + win;
	tp->last_ack_sent = tp->rcv_nxt;
	tp->t_flags &= ~TF_ACKNOW;
	if (tcp_delack_enabled)
		callout_stop(tp->tt_delack);
	if (sendalot && (!tcp_do_newreno || --maxburst))
		goto again;
	return (0);
}

void
tcp_setpersist(tp)
	register struct tcpcb *tp;
{
	int t = ((tp->t_srtt >> 2) + tp->t_rttvar) >> 1;
	int tt;

	if (callout_active(tp->tt_rexmt))
		panic("tcp_setpersist: retransmit pending");
	/*
	 * Start/restart persistance timer.
	 */
	TCPT_RANGESET(tt, t * tcp_backoff[tp->t_rxtshift],
		      TCPTV_PERSMIN, TCPTV_PERSMAX);
	callout_reset(tp->tt_persist, tt, tcp_timer_persist, tp);
	if (tp->t_rxtshift < tcp_rexmit_shift_max)
		tp->t_rxtshift++;
}