1994-05-24 10:09:53 +00:00
|
|
|
/*
|
1995-10-03 16:54:17 +00:00
|
|
|
* Copyright (c) 1982, 1986, 1988, 1990, 1993, 1994, 1995
|
1994-05-24 10:09:53 +00:00
|
|
|
* The Regents of the University of California. All rights reserved.
|
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
* are met:
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
* 3. All advertising materials mentioning features or use of this software
|
|
|
|
* must display the following acknowledgement:
|
|
|
|
* This product includes software developed by the University of
|
|
|
|
* California, Berkeley and its contributors.
|
|
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*
|
1995-10-03 16:54:17 +00:00
|
|
|
* @(#)tcp_input.c 8.12 (Berkeley) 5/24/95
|
1996-03-22 18:09:21 +00:00
|
|
|
* $Id: tcp_input.c,v 1.38 1996/03/11 15:13:29 davidg Exp $
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
#ifndef TUBA_INCLUDE
|
|
|
|
#include <sys/param.h>
|
1996-03-11 15:13:58 +00:00
|
|
|
#include <sys/queue.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/systm.h>
|
1995-11-09 20:23:09 +00:00
|
|
|
#include <sys/kernel.h>
|
|
|
|
#include <sys/sysctl.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/malloc.h>
|
|
|
|
#include <sys/mbuf.h>
|
|
|
|
#include <sys/protosw.h>
|
|
|
|
#include <sys/socket.h>
|
|
|
|
#include <sys/socketvar.h>
|
|
|
|
#include <sys/errno.h>
|
|
|
|
|
1995-10-03 16:54:17 +00:00
|
|
|
#include <machine/cpu.h> /* before tcp_seq.h, for tcp_random18() */
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <net/if.h>
|
|
|
|
#include <net/route.h>
|
|
|
|
|
|
|
|
#include <netinet/in.h>
|
|
|
|
#include <netinet/in_systm.h>
|
|
|
|
#include <netinet/ip.h>
|
|
|
|
#include <netinet/in_pcb.h>
|
|
|
|
#include <netinet/ip_var.h>
|
|
|
|
#include <netinet/tcp.h>
|
|
|
|
#include <netinet/tcp_fsm.h>
|
|
|
|
#include <netinet/tcp_seq.h>
|
|
|
|
#include <netinet/tcp_timer.h>
|
|
|
|
#include <netinet/tcp_var.h>
|
|
|
|
#include <netinet/tcpip.h>
|
1994-09-15 10:36:56 +00:00
|
|
|
#ifdef TCPDEBUG
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <netinet/tcp_debug.h>
|
1995-12-14 09:55:16 +00:00
|
|
|
static struct tcpiphdr tcp_saveti;
|
1994-09-15 10:36:56 +00:00
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1995-11-14 20:34:56 +00:00
|
|
|
static int tcprexmtthresh = 3;
|
1995-02-14 02:35:19 +00:00
|
|
|
tcp_seq tcp_iss;
|
|
|
|
tcp_cc tcp_ccgen;
|
1995-11-14 20:34:56 +00:00
|
|
|
|
1995-02-14 02:35:19 +00:00
|
|
|
struct tcpstat tcpstat;
|
1995-11-09 20:23:09 +00:00
|
|
|
SYSCTL_STRUCT(_net_inet_tcp, TCPCTL_STATS, stats,
|
|
|
|
CTLFLAG_RD, &tcpstat , tcpstat, "");
|
1995-11-14 20:34:56 +00:00
|
|
|
|
1995-02-14 02:35:19 +00:00
|
|
|
u_long tcp_now;
|
1995-04-09 01:29:31 +00:00
|
|
|
struct inpcbhead tcb;
|
|
|
|
struct inpcbinfo tcbinfo;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1995-11-14 20:34:56 +00:00
|
|
|
static void tcp_dooptions __P((struct tcpcb *,
|
|
|
|
u_char *, int, struct tcpiphdr *, struct tcpopt *));
|
|
|
|
static void tcp_pulloutofband __P((struct socket *,
|
|
|
|
struct tcpiphdr *, struct mbuf *));
|
|
|
|
static int tcp_reass __P((struct tcpcb *, struct tcpiphdr *, struct mbuf *));
|
|
|
|
static void tcp_xmit_timer __P((struct tcpcb *, int));
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
#endif /* TUBA_INCLUDE */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Insert segment ti into reassembly queue of tcp with
|
|
|
|
* control block tp. Return TH_FIN if reassembly now includes
|
|
|
|
* a segment with FIN. The macro form does the common case inline
|
|
|
|
* (segment is the next to be received on an established connection,
|
|
|
|
* and the queue is empty), avoiding linkage into and removal
|
|
|
|
* from the queue and repetition of various conversions.
|
|
|
|
* Set DELACK for segments received in order, but ack immediately
|
|
|
|
* when segments are out of order (so fast retransmit can work).
|
|
|
|
*/
|
1995-05-11 01:41:06 +00:00
|
|
|
#ifdef TCP_ACK_HACK
|
1994-05-24 10:09:53 +00:00
|
|
|
#define TCP_REASS(tp, ti, m, so, flags) { \
|
|
|
|
if ((ti)->ti_seq == (tp)->rcv_nxt && \
|
|
|
|
(tp)->seg_next == (struct tcpiphdr *)(tp) && \
|
|
|
|
(tp)->t_state == TCPS_ESTABLISHED) { \
|
1995-04-10 17:37:46 +00:00
|
|
|
if (ti->ti_flags & TH_PUSH) \
|
|
|
|
tp->t_flags |= TF_ACKNOW; \
|
|
|
|
else \
|
|
|
|
tp->t_flags |= TF_DELACK; \
|
1994-05-24 10:09:53 +00:00
|
|
|
(tp)->rcv_nxt += (ti)->ti_len; \
|
|
|
|
flags = (ti)->ti_flags & TH_FIN; \
|
|
|
|
tcpstat.tcps_rcvpack++;\
|
|
|
|
tcpstat.tcps_rcvbyte += (ti)->ti_len;\
|
|
|
|
sbappend(&(so)->so_rcv, (m)); \
|
|
|
|
sorwakeup(so); \
|
|
|
|
} else { \
|
|
|
|
(flags) = tcp_reass((tp), (ti), (m)); \
|
|
|
|
tp->t_flags |= TF_ACKNOW; \
|
|
|
|
} \
|
|
|
|
}
|
1995-05-11 01:41:06 +00:00
|
|
|
#else
|
|
|
|
#define TCP_REASS(tp, ti, m, so, flags) { \
|
|
|
|
if ((ti)->ti_seq == (tp)->rcv_nxt && \
|
|
|
|
(tp)->seg_next == (struct tcpiphdr *)(tp) && \
|
|
|
|
(tp)->t_state == TCPS_ESTABLISHED) { \
|
|
|
|
tp->t_flags |= TF_DELACK; \
|
|
|
|
(tp)->rcv_nxt += (ti)->ti_len; \
|
|
|
|
flags = (ti)->ti_flags & TH_FIN; \
|
|
|
|
tcpstat.tcps_rcvpack++;\
|
|
|
|
tcpstat.tcps_rcvbyte += (ti)->ti_len;\
|
|
|
|
sbappend(&(so)->so_rcv, (m)); \
|
|
|
|
sorwakeup(so); \
|
|
|
|
} else { \
|
|
|
|
(flags) = tcp_reass((tp), (ti), (m)); \
|
|
|
|
tp->t_flags |= TF_ACKNOW; \
|
|
|
|
} \
|
|
|
|
}
|
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
#ifndef TUBA_INCLUDE
|
|
|
|
|
1995-11-14 20:34:56 +00:00
|
|
|
static int
|
1994-05-24 10:09:53 +00:00
|
|
|
tcp_reass(tp, ti, m)
|
|
|
|
register struct tcpcb *tp;
|
|
|
|
register struct tcpiphdr *ti;
|
|
|
|
struct mbuf *m;
|
|
|
|
{
|
|
|
|
register struct tcpiphdr *q;
|
|
|
|
struct socket *so = tp->t_inpcb->inp_socket;
|
|
|
|
int flags;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Call with ti==0 after become established to
|
|
|
|
* force pre-ESTABLISHED data up to user socket.
|
|
|
|
*/
|
|
|
|
if (ti == 0)
|
|
|
|
goto present;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Find a segment which begins after this one does.
|
|
|
|
*/
|
|
|
|
for (q = tp->seg_next; q != (struct tcpiphdr *)tp;
|
|
|
|
q = (struct tcpiphdr *)q->ti_next)
|
|
|
|
if (SEQ_GT(q->ti_seq, ti->ti_seq))
|
|
|
|
break;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If there is a preceding segment, it may provide some of
|
|
|
|
* our data already. If so, drop the data from the incoming
|
|
|
|
* segment. If it provides all of our data, drop us.
|
|
|
|
*/
|
|
|
|
if ((struct tcpiphdr *)q->ti_prev != (struct tcpiphdr *)tp) {
|
|
|
|
register int i;
|
|
|
|
q = (struct tcpiphdr *)q->ti_prev;
|
|
|
|
/* conversion to int (in i) handles seq wraparound */
|
|
|
|
i = q->ti_seq + q->ti_len - ti->ti_seq;
|
|
|
|
if (i > 0) {
|
|
|
|
if (i >= ti->ti_len) {
|
|
|
|
tcpstat.tcps_rcvduppack++;
|
|
|
|
tcpstat.tcps_rcvdupbyte += ti->ti_len;
|
|
|
|
m_freem(m);
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* Try to present any queued data
|
|
|
|
* at the left window edge to the user.
|
|
|
|
* This is needed after the 3-WHS
|
|
|
|
* completes.
|
|
|
|
*/
|
|
|
|
goto present; /* ??? */
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
m_adj(m, i);
|
|
|
|
ti->ti_len -= i;
|
|
|
|
ti->ti_seq += i;
|
|
|
|
}
|
|
|
|
q = (struct tcpiphdr *)(q->ti_next);
|
|
|
|
}
|
|
|
|
tcpstat.tcps_rcvoopack++;
|
|
|
|
tcpstat.tcps_rcvoobyte += ti->ti_len;
|
|
|
|
REASS_MBUF(ti) = m; /* XXX */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* While we overlap succeeding segments trim them or,
|
|
|
|
* if they are completely covered, dequeue them.
|
|
|
|
*/
|
|
|
|
while (q != (struct tcpiphdr *)tp) {
|
|
|
|
register int i = (ti->ti_seq + ti->ti_len) - q->ti_seq;
|
|
|
|
if (i <= 0)
|
|
|
|
break;
|
|
|
|
if (i < q->ti_len) {
|
|
|
|
q->ti_seq += i;
|
|
|
|
q->ti_len -= i;
|
|
|
|
m_adj(REASS_MBUF(q), i);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
q = (struct tcpiphdr *)q->ti_next;
|
|
|
|
m = REASS_MBUF((struct tcpiphdr *)q->ti_prev);
|
|
|
|
remque(q->ti_prev);
|
|
|
|
m_freem(m);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Stick new segment in its place.
|
|
|
|
*/
|
|
|
|
insque(ti, q->ti_prev);
|
|
|
|
|
|
|
|
present:
|
|
|
|
/*
|
|
|
|
* Present data to user, advancing rcv_nxt through
|
|
|
|
* completed sequence space.
|
|
|
|
*/
|
1995-02-09 23:13:27 +00:00
|
|
|
if (!TCPS_HAVEESTABLISHED(tp->t_state))
|
1994-05-24 10:09:53 +00:00
|
|
|
return (0);
|
|
|
|
ti = tp->seg_next;
|
|
|
|
if (ti == (struct tcpiphdr *)tp || ti->ti_seq != tp->rcv_nxt)
|
|
|
|
return (0);
|
|
|
|
do {
|
|
|
|
tp->rcv_nxt += ti->ti_len;
|
|
|
|
flags = ti->ti_flags & TH_FIN;
|
|
|
|
remque(ti);
|
|
|
|
m = REASS_MBUF(ti);
|
|
|
|
ti = (struct tcpiphdr *)ti->ti_next;
|
|
|
|
if (so->so_state & SS_CANTRCVMORE)
|
|
|
|
m_freem(m);
|
|
|
|
else
|
|
|
|
sbappend(&so->so_rcv, m);
|
|
|
|
} while (ti != (struct tcpiphdr *)tp && ti->ti_seq == tp->rcv_nxt);
|
|
|
|
sorwakeup(so);
|
|
|
|
return (flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* TCP input routine, follows pages 65-76 of the
|
|
|
|
* protocol specification dated September, 1981 very closely.
|
|
|
|
*/
|
|
|
|
void
|
|
|
|
tcp_input(m, iphlen)
|
|
|
|
register struct mbuf *m;
|
|
|
|
int iphlen;
|
|
|
|
{
|
|
|
|
register struct tcpiphdr *ti;
|
|
|
|
register struct inpcb *inp;
|
1995-10-03 16:54:17 +00:00
|
|
|
u_char *optp = NULL;
|
1994-05-25 09:21:21 +00:00
|
|
|
int optlen = 0;
|
1994-05-24 10:09:53 +00:00
|
|
|
int len, tlen, off;
|
|
|
|
register struct tcpcb *tp = 0;
|
|
|
|
register int tiflags;
|
1994-05-25 09:21:21 +00:00
|
|
|
struct socket *so = 0;
|
1994-05-24 10:09:53 +00:00
|
|
|
int todrop, acked, ourfinisacked, needoutput = 0;
|
|
|
|
struct in_addr laddr;
|
|
|
|
int dropsocket = 0;
|
|
|
|
int iss = 0;
|
1995-02-09 23:13:27 +00:00
|
|
|
u_long tiwin;
|
|
|
|
struct tcpopt to; /* options in this segment */
|
|
|
|
struct rmxp_tao *taop; /* pointer to our TAO cache entry */
|
|
|
|
struct rmxp_tao tao_noncached; /* in case there's no cached entry */
|
1994-09-15 10:36:56 +00:00
|
|
|
#ifdef TCPDEBUG
|
|
|
|
short ostate = 0;
|
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1995-02-09 23:13:27 +00:00
|
|
|
bzero((char *)&to, sizeof(to));
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
tcpstat.tcps_rcvtotal++;
|
|
|
|
/*
|
|
|
|
* Get IP and TCP header together in first mbuf.
|
|
|
|
* Note: IP leaves IP header in first mbuf.
|
|
|
|
*/
|
|
|
|
ti = mtod(m, struct tcpiphdr *);
|
|
|
|
if (iphlen > sizeof (struct ip))
|
|
|
|
ip_stripoptions(m, (struct mbuf *)0);
|
|
|
|
if (m->m_len < sizeof (struct tcpiphdr)) {
|
|
|
|
if ((m = m_pullup(m, sizeof (struct tcpiphdr))) == 0) {
|
|
|
|
tcpstat.tcps_rcvshort++;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
ti = mtod(m, struct tcpiphdr *);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Checksum extended TCP header and data.
|
|
|
|
*/
|
|
|
|
tlen = ((struct ip *)ti)->ip_len;
|
|
|
|
len = sizeof (struct ip) + tlen;
|
|
|
|
ti->ti_next = ti->ti_prev = 0;
|
|
|
|
ti->ti_x1 = 0;
|
|
|
|
ti->ti_len = (u_short)tlen;
|
|
|
|
HTONS(ti->ti_len);
|
1994-10-02 17:48:58 +00:00
|
|
|
ti->ti_sum = in_cksum(m, len);
|
|
|
|
if (ti->ti_sum) {
|
1994-05-24 10:09:53 +00:00
|
|
|
tcpstat.tcps_rcvbadsum++;
|
|
|
|
goto drop;
|
|
|
|
}
|
|
|
|
#endif /* TUBA_INCLUDE */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check that TCP offset makes sense,
|
|
|
|
* pull out TCP options and adjust length. XXX
|
|
|
|
*/
|
|
|
|
off = ti->ti_off << 2;
|
|
|
|
if (off < sizeof (struct tcphdr) || off > tlen) {
|
|
|
|
tcpstat.tcps_rcvbadoff++;
|
|
|
|
goto drop;
|
|
|
|
}
|
|
|
|
tlen -= off;
|
|
|
|
ti->ti_len = tlen;
|
|
|
|
if (off > sizeof (struct tcphdr)) {
|
|
|
|
if (m->m_len < sizeof(struct ip) + off) {
|
|
|
|
if ((m = m_pullup(m, sizeof (struct ip) + off)) == 0) {
|
|
|
|
tcpstat.tcps_rcvshort++;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
ti = mtod(m, struct tcpiphdr *);
|
|
|
|
}
|
|
|
|
optlen = off - sizeof (struct tcphdr);
|
1995-10-03 16:54:17 +00:00
|
|
|
optp = mtod(m, u_char *) + sizeof (struct tcpiphdr);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
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);
|
|
|
|
|
1995-04-05 10:32:14 +00:00
|
|
|
/*
|
|
|
|
* 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);
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* Locate pcb for segment.
|
|
|
|
*/
|
|
|
|
findpcb:
|
1995-05-03 07:16:53 +00:00
|
|
|
/*
|
|
|
|
* First look for an exact match.
|
|
|
|
*/
|
|
|
|
inp = in_pcblookuphash(&tcbinfo, ti->ti_src, ti->ti_sport,
|
|
|
|
ti->ti_dst, ti->ti_dport);
|
|
|
|
/*
|
|
|
|
* ...and if that fails, do a wildcard search.
|
|
|
|
*/
|
|
|
|
if (inp == NULL) {
|
|
|
|
inp = in_pcblookup(&tcb, ti->ti_src, ti->ti_sport,
|
|
|
|
ti->ti_dst, ti->ti_dport, INPLOOKUP_WILDCARD);
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
1995-05-03 07:16:53 +00:00
|
|
|
if (inp == NULL)
|
1994-05-24 10:09:53 +00:00
|
|
|
goto dropwithreset;
|
|
|
|
tp = intotcpcb(inp);
|
|
|
|
if (tp == 0)
|
|
|
|
goto dropwithreset;
|
|
|
|
if (tp->t_state == TCPS_CLOSED)
|
|
|
|
goto drop;
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/* 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)) {
|
1994-09-15 10:36:56 +00:00
|
|
|
#ifdef TCPDEBUG
|
1994-05-24 10:09:53 +00:00
|
|
|
if (so->so_options & SO_DEBUG) {
|
|
|
|
ostate = tp->t_state;
|
|
|
|
tcp_saveti = *ti;
|
|
|
|
}
|
1994-09-15 10:36:56 +00:00
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
if (so->so_options & SO_ACCEPTCONN) {
|
1995-02-09 23:13:27 +00:00
|
|
|
register struct tcpcb *tp0 = tp;
|
1994-05-24 10:09:53 +00:00
|
|
|
so = sonewconn(so, 0);
|
1996-02-26 21:47:13 +00:00
|
|
|
if (so == 0) {
|
|
|
|
tcpstat.tcps_listendrop++;
|
1994-05-24 10:09:53 +00:00
|
|
|
goto drop;
|
1996-02-26 21:47:13 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* 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;
|
1995-04-09 01:29:31 +00:00
|
|
|
in_pcbrehash(inp);
|
1994-05-24 10:09:53 +00:00
|
|
|
#if BSD>=43
|
|
|
|
inp->inp_options = ip_srcroute();
|
|
|
|
#endif
|
|
|
|
tp = intotcpcb(inp);
|
|
|
|
tp->t_state = TCPS_LISTEN;
|
1995-02-09 23:13:27 +00:00
|
|
|
tp->t_flags |= tp0->t_flags & (TF_NOPUSH|TF_NOOPT);
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1995-02-09 23:13:27 +00:00
|
|
|
/* Compute proper scaling value from buffer space */
|
1994-05-24 10:09:53 +00:00
|
|
|
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).
|
|
|
|
*/
|
1996-02-22 11:46:39 +00:00
|
|
|
if (tp->t_state != TCPS_LISTEN)
|
|
|
|
tcp_dooptions(tp, optp, optlen, ti, &to);
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1994-05-24 10:09:53 +00:00
|
|
|
* 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.
|
1995-02-09 23:13:27 +00:00
|
|
|
* Make sure that the hidden state-flags are also off.
|
|
|
|
* Since we check for TCPS_ESTABLISHED above, it can only
|
|
|
|
* be TH_NEEDSYN.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
if (tp->t_state == TCPS_ESTABLISHED &&
|
|
|
|
(tiflags & (TH_SYN|TH_FIN|TH_RST|TH_URG|TH_ACK)) == TH_ACK &&
|
1995-02-09 23:13:27 +00:00
|
|
|
((tp->t_flags & (TF_NEEDSYN|TF_NEEDFIN)) == 0) &&
|
|
|
|
((to.to_flag & TOF_TS) == 0 ||
|
|
|
|
TSTMP_GEQ(to.to_tsval, tp->ts_recent)) &&
|
|
|
|
/*
|
|
|
|
* Using the CC option is compulsory if once started:
|
|
|
|
* the segment is OK if no T/TCP was negotiated or
|
|
|
|
* if the segment has a CC option equal to CCrecv
|
|
|
|
*/
|
|
|
|
((tp->t_flags & (TF_REQ_CC|TF_RCVD_CC)) != (TF_REQ_CC|TF_RCVD_CC) ||
|
|
|
|
(to.to_flag & TOF_CC) != 0 && to.to_cc == tp->cc_recv) &&
|
1994-05-24 10:09:53 +00:00
|
|
|
ti->ti_seq == tp->rcv_nxt &&
|
|
|
|
tiwin && tiwin == tp->snd_wnd &&
|
|
|
|
tp->snd_nxt == tp->snd_max) {
|
|
|
|
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1994-05-24 10:09:53 +00:00
|
|
|
* If last ACK falls within this segment's sequence numbers,
|
1995-02-09 23:13:27 +00:00
|
|
|
* record the timestamp.
|
|
|
|
* NOTE that the test is modified according to the latest
|
|
|
|
* proposal of the tcplw@cray.com list (Braden 1993/04/26).
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((to.to_flag & TOF_TS) != 0 &&
|
|
|
|
SEQ_LEQ(ti->ti_seq, tp->last_ack_sent)) {
|
|
|
|
tp->ts_recent_age = tcp_now;
|
|
|
|
tp->ts_recent = to.to_tsval;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (ti->ti_len == 0) {
|
|
|
|
if (SEQ_GT(ti->ti_ack, tp->snd_una) &&
|
|
|
|
SEQ_LEQ(ti->ti_ack, tp->snd_max) &&
|
1996-03-22 18:09:21 +00:00
|
|
|
tp->snd_cwnd >= tp->snd_wnd &&
|
|
|
|
tp->t_dupacks < tcprexmtthresh) {
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* this is a pure ack for outstanding data.
|
|
|
|
*/
|
|
|
|
++tcpstat.tcps_predack;
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((to.to_flag & TOF_TS) != 0)
|
|
|
|
tcp_xmit_timer(tp,
|
|
|
|
tcp_now - to.to_tsecr + 1);
|
1994-05-24 10:09:53 +00:00
|
|
|
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;
|
|
|
|
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_flags & SB_NOTIFY)
|
|
|
|
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.
|
|
|
|
*/
|
|
|
|
++tcpstat.tcps_preddat;
|
|
|
|
tp->rcv_nxt += ti->ti_len;
|
|
|
|
tcpstat.tcps_rcvpack++;
|
|
|
|
tcpstat.tcps_rcvbyte += ti->ti_len;
|
|
|
|
/*
|
1995-04-05 10:32:14 +00:00
|
|
|
* Add data to socket buffer.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
sbappend(&so->so_rcv, m);
|
|
|
|
sorwakeup(so);
|
1995-05-11 01:41:06 +00:00
|
|
|
#ifdef TCP_ACK_HACK
|
1995-03-27 07:12:24 +00:00
|
|
|
/*
|
|
|
|
* If this is a short packet, then ACK now - with Nagel
|
|
|
|
* congestion avoidance sender won't send more until
|
|
|
|
* he gets an ACK.
|
|
|
|
*/
|
1995-04-10 17:16:10 +00:00
|
|
|
if (tiflags & TH_PUSH) {
|
1995-03-27 07:12:24 +00:00
|
|
|
tp->t_flags |= TF_ACKNOW;
|
|
|
|
tcp_output(tp);
|
|
|
|
} else {
|
|
|
|
tp->t_flags |= TF_DELACK;
|
|
|
|
}
|
1995-05-11 01:41:06 +00:00
|
|
|
#else
|
|
|
|
tp->t_flags |= TF_DELACK;
|
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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;
|
|
|
|
|
|
|
|
if (tiflags & TH_RST)
|
|
|
|
goto drop;
|
|
|
|
if (tiflags & TH_ACK)
|
|
|
|
goto dropwithreset;
|
|
|
|
if ((tiflags & TH_SYN) == 0)
|
|
|
|
goto drop;
|
|
|
|
/*
|
|
|
|
* 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) ||
|
1994-05-26 09:51:33 +00:00
|
|
|
IN_MULTICAST(ntohl(ti->ti_dst.s_addr)))
|
1994-05-24 10:09:53 +00:00
|
|
|
goto drop;
|
|
|
|
am = m_get(M_DONTWAIT, MT_SONAME); /* XXX */
|
|
|
|
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;
|
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((taop = tcp_gettaocache(inp)) == NULL) {
|
|
|
|
taop = &tao_noncached;
|
|
|
|
bzero(taop, sizeof(*taop));
|
|
|
|
}
|
1996-02-22 11:46:39 +00:00
|
|
|
tcp_dooptions(tp, optp, optlen, ti, &to);
|
1994-05-24 10:09:53 +00:00
|
|
|
if (iss)
|
|
|
|
tp->iss = iss;
|
|
|
|
else
|
|
|
|
tp->iss = tcp_iss;
|
1995-10-03 16:54:17 +00:00
|
|
|
tcp_iss += TCP_ISSINCR/4;
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->irs = ti->ti_seq;
|
|
|
|
tcp_sendseqinit(tp);
|
|
|
|
tcp_rcvseqinit(tp);
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* Initialization of the tcpcb for transaction;
|
|
|
|
* set SND.WND = SEG.WND,
|
|
|
|
* initialize CCsend and CCrecv.
|
|
|
|
*/
|
|
|
|
tp->snd_wnd = tiwin; /* initial send-window */
|
|
|
|
tp->cc_send = CC_INC(tcp_ccgen);
|
|
|
|
tp->cc_recv = to.to_cc;
|
|
|
|
/*
|
|
|
|
* Perform TAO test on incoming CC (SEG.CC) option, if any.
|
1995-05-30 08:16:23 +00:00
|
|
|
* - compare SEG.CC against cached CC from the same host,
|
1995-02-09 23:13:27 +00:00
|
|
|
* if any.
|
|
|
|
* - if SEG.CC > chached value, SYN must be new and is accepted
|
|
|
|
* immediately: save new CC in the cache, mark the socket
|
|
|
|
* connected, enter ESTABLISHED state, turn on flag to
|
|
|
|
* send a SYN in the next segment.
|
|
|
|
* A virtual advertised window is set in rcv_adv to
|
|
|
|
* initialize SWS prevention. Then enter normal segment
|
|
|
|
* processing: drop SYN, process data and FIN.
|
|
|
|
* - otherwise do a normal 3-way handshake.
|
|
|
|
*/
|
|
|
|
if ((to.to_flag & TOF_CC) != 0) {
|
|
|
|
if (taop->tao_cc != 0 && CC_GT(to.to_cc, taop->tao_cc)) {
|
|
|
|
taop->tao_cc = to.to_cc;
|
|
|
|
tp->t_state = TCPS_ESTABLISHED;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If there is a FIN, or if there is data and the
|
|
|
|
* connection is local, then delay SYN,ACK(SYN) in
|
|
|
|
* the hope of piggy-backing it on a response
|
|
|
|
* segment. Otherwise must send ACK now in case
|
|
|
|
* the other side is slow starting.
|
|
|
|
*/
|
|
|
|
if ((tiflags & TH_FIN) || (ti->ti_len != 0 &&
|
|
|
|
in_localaddr(inp->inp_faddr)))
|
|
|
|
tp->t_flags |= (TF_DELACK | TF_NEEDSYN);
|
|
|
|
else
|
|
|
|
tp->t_flags |= (TF_ACKNOW | TF_NEEDSYN);
|
1996-01-31 08:22:24 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Limit the `virtual advertised window' to TCP_MAXWIN
|
|
|
|
* here. Even if we requested window scaling, it will
|
|
|
|
* become effective only later when our SYN is acked.
|
|
|
|
*/
|
|
|
|
tp->rcv_adv += min(tp->rcv_wnd, TCP_MAXWIN);
|
1995-02-09 23:13:27 +00:00
|
|
|
tcpstat.tcps_connects++;
|
|
|
|
soisconnected(so);
|
|
|
|
tp->t_timer[TCPT_KEEP] = TCPTV_KEEP_INIT;
|
|
|
|
dropsocket = 0; /* committed to socket */
|
|
|
|
tcpstat.tcps_accepts++;
|
|
|
|
goto trimthenstep6;
|
|
|
|
}
|
|
|
|
/* else do standard 3-way handshake */
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* No CC option, but maybe CC.NEW:
|
|
|
|
* invalidate cached value.
|
|
|
|
*/
|
|
|
|
taop->tao_cc = 0;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* TAO test failed or there was no CC option,
|
|
|
|
* do a standard 3-way handshake.
|
|
|
|
*/
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->t_flags |= TF_ACKNOW;
|
|
|
|
tp->t_state = TCPS_SYN_RECEIVED;
|
|
|
|
tp->t_timer[TCPT_KEEP] = TCPTV_KEEP_INIT;
|
|
|
|
dropsocket = 0; /* committed to socket */
|
|
|
|
tcpstat.tcps_accepts++;
|
|
|
|
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:
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((taop = tcp_gettaocache(inp)) == NULL) {
|
|
|
|
taop = &tao_noncached;
|
|
|
|
bzero(taop, sizeof(*taop));
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((tiflags & TH_ACK) &&
|
|
|
|
(SEQ_LEQ(ti->ti_ack, tp->iss) ||
|
|
|
|
SEQ_GT(ti->ti_ack, tp->snd_max))) {
|
|
|
|
/*
|
|
|
|
* If we have a cached CCsent for the remote host,
|
|
|
|
* hence we haven't just crashed and restarted,
|
|
|
|
* do not send a RST. This may be a retransmission
|
|
|
|
* from the other side after our earlier ACK was lost.
|
|
|
|
* Our new SYN, when it arrives, will serve as the
|
|
|
|
* needed ACK.
|
|
|
|
*/
|
|
|
|
if (taop->tao_ccsent != 0)
|
|
|
|
goto drop;
|
|
|
|
else
|
|
|
|
goto dropwithreset;
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
if (tiflags & TH_RST) {
|
|
|
|
if (tiflags & TH_ACK)
|
|
|
|
tp = tcp_drop(tp, ECONNREFUSED);
|
|
|
|
goto drop;
|
|
|
|
}
|
|
|
|
if ((tiflags & TH_SYN) == 0)
|
|
|
|
goto drop;
|
1995-02-09 23:13:27 +00:00
|
|
|
tp->snd_wnd = ti->ti_win; /* initial send window */
|
|
|
|
tp->cc_recv = to.to_cc; /* foreign CC */
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->irs = ti->ti_seq;
|
|
|
|
tcp_rcvseqinit(tp);
|
1995-11-03 22:31:54 +00:00
|
|
|
if (tiflags & TH_ACK) {
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* Our SYN was acked. If segment contains CC.ECHO
|
|
|
|
* option, check it to make sure this segment really
|
|
|
|
* matches our SYN. If not, just drop it as old
|
|
|
|
* duplicate, but send an RST if we're still playing
|
|
|
|
* by the old rules.
|
|
|
|
*/
|
|
|
|
if ((to.to_flag & TOF_CCECHO) &&
|
|
|
|
tp->cc_send != to.to_ccecho) {
|
|
|
|
if (taop->tao_ccsent != 0)
|
|
|
|
goto drop;
|
|
|
|
else
|
|
|
|
goto dropwithreset;
|
|
|
|
}
|
1995-11-03 22:31:54 +00:00
|
|
|
tcpstat.tcps_connects++;
|
|
|
|
soisconnected(so);
|
|
|
|
/* Do window scaling on this connection? */
|
|
|
|
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;
|
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
/* Segment is acceptable, update cache if undefined. */
|
|
|
|
if (taop->tao_ccsent == 0)
|
|
|
|
taop->tao_ccsent = to.to_ccecho;
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1995-02-09 23:13:27 +00:00
|
|
|
tp->rcv_adv += tp->rcv_wnd;
|
|
|
|
tp->snd_una++; /* SYN is acked */
|
|
|
|
/*
|
|
|
|
* If there's data, delay ACK; if there's also a FIN
|
|
|
|
* ACKNOW will be turned on later.
|
|
|
|
*/
|
|
|
|
if (ti->ti_len != 0)
|
|
|
|
tp->t_flags |= TF_DELACK;
|
|
|
|
else
|
|
|
|
tp->t_flags |= TF_ACKNOW;
|
|
|
|
/*
|
|
|
|
* Received <SYN,ACK> in SYN_SENT[*] state.
|
|
|
|
* Transitions:
|
|
|
|
* SYN_SENT --> ESTABLISHED
|
|
|
|
* SYN_SENT* --> FIN_WAIT_1
|
|
|
|
*/
|
|
|
|
if (tp->t_flags & TF_NEEDFIN) {
|
|
|
|
tp->t_state = TCPS_FIN_WAIT_1;
|
|
|
|
tp->t_flags &= ~TF_NEEDFIN;
|
|
|
|
tiflags &= ~TH_SYN;
|
|
|
|
} else
|
|
|
|
tp->t_state = TCPS_ESTABLISHED;
|
|
|
|
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* Received initial SYN in SYN-SENT[*] state => simul-
|
|
|
|
* taneous open. If segment contains CC option and there is
|
1995-05-30 08:16:23 +00:00
|
|
|
* a cached CC, apply TAO test; if it succeeds, connection is
|
1995-02-09 23:13:27 +00:00
|
|
|
* half-synchronized. Otherwise, do 3-way handshake:
|
|
|
|
* SYN-SENT -> SYN-RECEIVED
|
|
|
|
* SYN-SENT* -> SYN-RECEIVED*
|
|
|
|
* If there was no CC option, clear cached CC value.
|
|
|
|
*/
|
|
|
|
tp->t_flags |= TF_ACKNOW;
|
1995-05-30 08:16:23 +00:00
|
|
|
tp->t_timer[TCPT_REXMT] = 0;
|
1995-02-09 23:13:27 +00:00
|
|
|
if (to.to_flag & TOF_CC) {
|
|
|
|
if (taop->tao_cc != 0 &&
|
|
|
|
CC_GT(to.to_cc, taop->tao_cc)) {
|
|
|
|
/*
|
1995-05-30 08:16:23 +00:00
|
|
|
* update cache and make transition:
|
1995-02-09 23:13:27 +00:00
|
|
|
* SYN-SENT -> ESTABLISHED*
|
|
|
|
* SYN-SENT* -> FIN-WAIT-1*
|
|
|
|
*/
|
|
|
|
taop->tao_cc = to.to_cc;
|
|
|
|
if (tp->t_flags & TF_NEEDFIN) {
|
|
|
|
tp->t_state = TCPS_FIN_WAIT_1;
|
|
|
|
tp->t_flags &= ~TF_NEEDFIN;
|
|
|
|
} else
|
|
|
|
tp->t_state = TCPS_ESTABLISHED;
|
|
|
|
tp->t_flags |= TF_NEEDSYN;
|
|
|
|
} else
|
|
|
|
tp->t_state = TCPS_SYN_RECEIVED;
|
|
|
|
} else {
|
|
|
|
/* CC.NEW or no option => invalidate cache */
|
|
|
|
taop->tao_cc = 0;
|
|
|
|
tp->t_state = TCPS_SYN_RECEIVED;
|
|
|
|
}
|
1995-05-30 08:16:23 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
trimthenstep6:
|
|
|
|
/*
|
|
|
|
* Advance ti->ti_seq to correspond to first data byte.
|
|
|
|
* If data, trim to stay within window,
|
|
|
|
* dropping FIN if necessary.
|
|
|
|
*/
|
|
|
|
ti->ti_seq++;
|
|
|
|
if (ti->ti_len > tp->rcv_wnd) {
|
|
|
|
todrop = ti->ti_len - tp->rcv_wnd;
|
|
|
|
m_adj(m, -todrop);
|
|
|
|
ti->ti_len = tp->rcv_wnd;
|
|
|
|
tiflags &= ~TH_FIN;
|
|
|
|
tcpstat.tcps_rcvpackafterwin++;
|
|
|
|
tcpstat.tcps_rcvbyteafterwin += todrop;
|
|
|
|
}
|
|
|
|
tp->snd_wl1 = ti->ti_seq - 1;
|
|
|
|
tp->rcv_up = ti->ti_seq;
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* Client side of transaction: already sent SYN and data.
|
|
|
|
* If the remote host used T/TCP to validate the SYN,
|
|
|
|
* our data will be ACK'd; if so, enter normal data segment
|
|
|
|
* processing in the middle of step 5, ack processing.
|
|
|
|
* Otherwise, goto step 6.
|
1995-05-30 08:16:23 +00:00
|
|
|
*/
|
1995-02-09 23:13:27 +00:00
|
|
|
if (tiflags & TH_ACK)
|
|
|
|
goto process_ACK;
|
1994-05-24 10:09:53 +00:00
|
|
|
goto step6;
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* If the state is LAST_ACK or CLOSING or TIME_WAIT:
|
|
|
|
* if segment contains a SYN and CC [not CC.NEW] option:
|
|
|
|
* if state == TIME_WAIT and connection duration > MSL,
|
|
|
|
* drop packet and send RST;
|
|
|
|
*
|
|
|
|
* if SEG.CC > CCrecv then is new SYN, and can implicitly
|
|
|
|
* ack the FIN (and data) in retransmission queue.
|
|
|
|
* Complete close and delete TCPCB. Then reprocess
|
|
|
|
* segment, hoping to find new TCPCB in LISTEN state;
|
|
|
|
*
|
|
|
|
* else must be old SYN; drop it.
|
1995-05-30 08:16:23 +00:00
|
|
|
* else do normal processing.
|
|
|
|
*/
|
1995-02-09 23:13:27 +00:00
|
|
|
case TCPS_LAST_ACK:
|
|
|
|
case TCPS_CLOSING:
|
|
|
|
case TCPS_TIME_WAIT:
|
|
|
|
if ((tiflags & TH_SYN) &&
|
|
|
|
(to.to_flag & TOF_CC) && tp->cc_recv != 0) {
|
|
|
|
if (tp->t_state == TCPS_TIME_WAIT &&
|
|
|
|
tp->t_duration > TCPTV_MSL)
|
|
|
|
goto dropwithreset;
|
|
|
|
if (CC_GT(to.to_cc, tp->cc_recv)) {
|
|
|
|
tp = tcp_close(tp);
|
|
|
|
goto findpcb;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
goto drop;
|
|
|
|
}
|
|
|
|
break; /* continue normal processing */
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* States other than LISTEN or SYN_SENT.
|
|
|
|
* First check timestamp, if present.
|
1995-02-09 23:13:27 +00:00
|
|
|
* Then check the connection count, if present.
|
1995-05-30 08:16:23 +00:00
|
|
|
* Then check that at least some bytes of segment are within
|
1994-05-24 10:09:53 +00:00
|
|
|
* receive window. If segment begins before rcv_nxt,
|
|
|
|
* drop leading data (and SYN); if nothing left, just ack.
|
1995-02-09 23:13:27 +00:00
|
|
|
*
|
1994-05-24 10:09:53 +00:00
|
|
|
* RFC 1323 PAWS: If we have a timestamp reply on this segment
|
|
|
|
* and it's less than ts_recent, drop it.
|
|
|
|
*/
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((to.to_flag & TOF_TS) != 0 && (tiflags & TH_RST) == 0 &&
|
|
|
|
tp->ts_recent && TSTMP_LT(to.to_tsval, tp->ts_recent)) {
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
/* Check to see if ts_recent is over 24 days old. */
|
|
|
|
if ((int)(tcp_now - tp->ts_recent_age) > TCP_PAWS_IDLE) {
|
|
|
|
/*
|
|
|
|
* Invalidate ts_recent. If this segment updates
|
|
|
|
* ts_recent, the age will be reset later and ts_recent
|
|
|
|
* will get a valid value. If it does not, setting
|
|
|
|
* ts_recent to zero will at least satisfy the
|
|
|
|
* requirement that zero be placed in the timestamp
|
|
|
|
* echo reply when ts_recent isn't valid. The
|
|
|
|
* age isn't reset until we get a valid ts_recent
|
|
|
|
* because we don't want out-of-order segments to be
|
|
|
|
* dropped when ts_recent is old.
|
|
|
|
*/
|
|
|
|
tp->ts_recent = 0;
|
|
|
|
} else {
|
|
|
|
tcpstat.tcps_rcvduppack++;
|
|
|
|
tcpstat.tcps_rcvdupbyte += ti->ti_len;
|
|
|
|
tcpstat.tcps_pawsdrop++;
|
|
|
|
goto dropafterack;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* T/TCP mechanism
|
|
|
|
* If T/TCP was negotiated and the segment doesn't have CC,
|
|
|
|
* or if it's CC is wrong then drop the segment.
|
|
|
|
* RST segments do not have to comply with this.
|
|
|
|
*/
|
|
|
|
if ((tp->t_flags & (TF_REQ_CC|TF_RCVD_CC)) == (TF_REQ_CC|TF_RCVD_CC) &&
|
|
|
|
((to.to_flag & TOF_CC) == 0 || tp->cc_recv != to.to_cc) &&
|
|
|
|
(tiflags & TH_RST) == 0)
|
|
|
|
goto dropafterack;
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
todrop = tp->rcv_nxt - ti->ti_seq;
|
|
|
|
if (todrop > 0) {
|
|
|
|
if (tiflags & TH_SYN) {
|
|
|
|
tiflags &= ~TH_SYN;
|
|
|
|
ti->ti_seq++;
|
1995-05-30 08:16:23 +00:00
|
|
|
if (ti->ti_urp > 1)
|
1994-05-24 10:09:53 +00:00
|
|
|
ti->ti_urp--;
|
|
|
|
else
|
|
|
|
tiflags &= ~TH_URG;
|
|
|
|
todrop--;
|
|
|
|
}
|
1995-02-16 01:39:19 +00:00
|
|
|
/*
|
|
|
|
* Following if statement from Stevens, vol. 2, p. 960.
|
|
|
|
*/
|
|
|
|
if (todrop > ti->ti_len
|
|
|
|
|| (todrop == ti->ti_len && (tiflags & TH_FIN) == 0)) {
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
1995-02-16 01:39:19 +00:00
|
|
|
* Any valid FIN must be to the left of the window.
|
|
|
|
* At this point the FIN must be a duplicate or out
|
|
|
|
* of sequence; drop it.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1995-02-16 01:39:19 +00:00
|
|
|
tiflags &= ~TH_FIN;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Send an ACK to resynchronize and drop any data.
|
|
|
|
* But keep on processing for RST or ACK.
|
|
|
|
*/
|
|
|
|
tp->t_flags |= TF_ACKNOW;
|
|
|
|
todrop = ti->ti_len;
|
|
|
|
tcpstat.tcps_rcvduppack++;
|
|
|
|
tcpstat.tcps_rcvdupbyte += todrop;
|
1994-05-24 10:09:53 +00:00
|
|
|
} else {
|
|
|
|
tcpstat.tcps_rcvpartduppack++;
|
|
|
|
tcpstat.tcps_rcvpartdupbyte += todrop;
|
|
|
|
}
|
|
|
|
m_adj(m, todrop);
|
|
|
|
ti->ti_seq += todrop;
|
|
|
|
ti->ti_len -= todrop;
|
|
|
|
if (ti->ti_urp > todrop)
|
|
|
|
ti->ti_urp -= todrop;
|
|
|
|
else {
|
|
|
|
tiflags &= ~TH_URG;
|
|
|
|
ti->ti_urp = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If new data are received on a connection after the
|
|
|
|
* user processes are gone, then RST the other end.
|
|
|
|
*/
|
|
|
|
if ((so->so_state & SS_NOFDREF) &&
|
|
|
|
tp->t_state > TCPS_CLOSE_WAIT && ti->ti_len) {
|
|
|
|
tp = tcp_close(tp);
|
|
|
|
tcpstat.tcps_rcvafterclose++;
|
|
|
|
goto dropwithreset;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If segment ends after window, drop trailing data
|
|
|
|
* (and PUSH and FIN); if nothing left, just ACK.
|
|
|
|
*/
|
|
|
|
todrop = (ti->ti_seq+ti->ti_len) - (tp->rcv_nxt+tp->rcv_wnd);
|
|
|
|
if (todrop > 0) {
|
|
|
|
tcpstat.tcps_rcvpackafterwin++;
|
|
|
|
if (todrop >= ti->ti_len) {
|
|
|
|
tcpstat.tcps_rcvbyteafterwin += ti->ti_len;
|
|
|
|
/*
|
|
|
|
* If a new connection request is received
|
|
|
|
* while in TIME_WAIT, drop the old connection
|
|
|
|
* and start over if the sequence numbers
|
|
|
|
* are above the previous ones.
|
|
|
|
*/
|
|
|
|
if (tiflags & TH_SYN &&
|
|
|
|
tp->t_state == TCPS_TIME_WAIT &&
|
|
|
|
SEQ_GT(ti->ti_seq, tp->rcv_nxt)) {
|
|
|
|
iss = tp->rcv_nxt + TCP_ISSINCR;
|
|
|
|
tp = tcp_close(tp);
|
|
|
|
goto findpcb;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* If window is closed can only take segments at
|
|
|
|
* window edge, and have to drop data and PUSH from
|
|
|
|
* incoming segments. Continue processing, but
|
|
|
|
* remember to ack. Otherwise, drop segment
|
|
|
|
* and ack.
|
|
|
|
*/
|
|
|
|
if (tp->rcv_wnd == 0 && ti->ti_seq == tp->rcv_nxt) {
|
|
|
|
tp->t_flags |= TF_ACKNOW;
|
|
|
|
tcpstat.tcps_rcvwinprobe++;
|
|
|
|
} else
|
|
|
|
goto dropafterack;
|
|
|
|
} else
|
|
|
|
tcpstat.tcps_rcvbyteafterwin += todrop;
|
|
|
|
m_adj(m, -todrop);
|
|
|
|
ti->ti_len -= todrop;
|
|
|
|
tiflags &= ~(TH_PUSH|TH_FIN);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If last ACK falls within this segment's sequence numbers,
|
|
|
|
* record its timestamp.
|
1995-02-09 23:13:27 +00:00
|
|
|
* NOTE that the test is modified according to the latest
|
|
|
|
* proposal of the tcplw@cray.com list (Braden 1993/04/26).
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1995-05-30 08:16:23 +00:00
|
|
|
if ((to.to_flag & TOF_TS) != 0 &&
|
1995-02-09 23:13:27 +00:00
|
|
|
SEQ_LEQ(ti->ti_seq, tp->last_ack_sent)) {
|
|
|
|
tp->ts_recent_age = tcp_now;
|
|
|
|
tp->ts_recent = to.to_tsval;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the RST bit is set examine the state:
|
|
|
|
* SYN_RECEIVED STATE:
|
|
|
|
* If passive open, return to LISTEN state.
|
|
|
|
* If active open, inform user that connection was refused.
|
|
|
|
* ESTABLISHED, FIN_WAIT_1, FIN_WAIT2, CLOSE_WAIT STATES:
|
|
|
|
* Inform user that connection was reset, and close tcb.
|
|
|
|
* CLOSING, LAST_ACK, TIME_WAIT STATES
|
|
|
|
* Close the tcb.
|
|
|
|
*/
|
|
|
|
if (tiflags&TH_RST) switch (tp->t_state) {
|
|
|
|
|
|
|
|
case TCPS_SYN_RECEIVED:
|
|
|
|
so->so_error = ECONNREFUSED;
|
|
|
|
goto close;
|
|
|
|
|
|
|
|
case TCPS_ESTABLISHED:
|
|
|
|
case TCPS_FIN_WAIT_1:
|
|
|
|
case TCPS_FIN_WAIT_2:
|
|
|
|
case TCPS_CLOSE_WAIT:
|
|
|
|
so->so_error = ECONNRESET;
|
|
|
|
close:
|
|
|
|
tp->t_state = TCPS_CLOSED;
|
|
|
|
tcpstat.tcps_drops++;
|
|
|
|
tp = tcp_close(tp);
|
|
|
|
goto drop;
|
|
|
|
|
|
|
|
case TCPS_CLOSING:
|
|
|
|
case TCPS_LAST_ACK:
|
|
|
|
case TCPS_TIME_WAIT:
|
|
|
|
tp = tcp_close(tp);
|
|
|
|
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;
|
|
|
|
}
|
|
|
|
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* If the ACK bit is off: if in SYN-RECEIVED state or SENDSYN
|
|
|
|
* flag is on (half-synchronized state), then queue data for
|
|
|
|
* later processing; else drop segment and return.
|
|
|
|
*/
|
|
|
|
if ((tiflags & TH_ACK) == 0) {
|
|
|
|
if (tp->t_state == TCPS_SYN_RECEIVED ||
|
|
|
|
(tp->t_flags & TF_NEEDSYN))
|
|
|
|
goto step6;
|
|
|
|
else
|
|
|
|
goto drop;
|
|
|
|
}
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* Ack processing.
|
|
|
|
*/
|
|
|
|
switch (tp->t_state) {
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In SYN_RECEIVED state if the ack ACKs our SYN then enter
|
|
|
|
* ESTABLISHED state and continue processing, otherwise
|
|
|
|
* send an RST.
|
|
|
|
*/
|
|
|
|
case TCPS_SYN_RECEIVED:
|
|
|
|
if (SEQ_GT(tp->snd_una, ti->ti_ack) ||
|
|
|
|
SEQ_GT(ti->ti_ack, tp->snd_max))
|
|
|
|
goto dropwithreset;
|
1995-02-09 23:13:27 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
tcpstat.tcps_connects++;
|
|
|
|
soisconnected(so);
|
|
|
|
/* 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;
|
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* Upon successful completion of 3-way handshake,
|
|
|
|
* update cache.CC if it was undefined, pass any queued
|
|
|
|
* data to the user, and advance state appropriately.
|
|
|
|
*/
|
|
|
|
if ((taop = tcp_gettaocache(inp)) != NULL &&
|
|
|
|
taop->tao_cc == 0)
|
|
|
|
taop->tao_cc = tp->cc_recv;
|
|
|
|
|
|
|
|
/*
|
1995-05-30 08:16:23 +00:00
|
|
|
* Make transitions:
|
1995-02-09 23:13:27 +00:00
|
|
|
* SYN-RECEIVED -> ESTABLISHED
|
|
|
|
* SYN-RECEIVED* -> FIN-WAIT-1
|
|
|
|
*/
|
|
|
|
if (tp->t_flags & TF_NEEDFIN) {
|
|
|
|
tp->t_state = TCPS_FIN_WAIT_1;
|
|
|
|
tp->t_flags &= ~TF_NEEDFIN;
|
|
|
|
} else
|
|
|
|
tp->t_state = TCPS_ESTABLISHED;
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1995-02-09 23:13:27 +00:00
|
|
|
* If segment contains data or ACK, will call tcp_reass()
|
|
|
|
* later; if not, do so now to pass queued data to user.
|
|
|
|
*/
|
|
|
|
if (ti->ti_len == 0 && (tiflags & TH_FIN) == 0)
|
|
|
|
(void) tcp_reass(tp, (struct tcpiphdr *)0,
|
|
|
|
(struct mbuf *)0);
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->snd_wl1 = ti->ti_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(ti->ti_ack, tp->snd_una)) {
|
|
|
|
if (ti->ti_len == 0 && 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
|
1995-05-30 08:16:23 +00:00
|
|
|
* network (they're now cached at the receiver)
|
1994-05-24 10:09:53 +00:00
|
|
|
* 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 ||
|
|
|
|
ti->ti_ack != tp->snd_una)
|
|
|
|
tp->t_dupacks = 0;
|
|
|
|
else if (++tp->t_dupacks == tcprexmtthresh) {
|
|
|
|
tcp_seq onxt = tp->snd_nxt;
|
|
|
|
u_int win =
|
|
|
|
min(tp->snd_wnd, tp->snd_cwnd) / 2 /
|
|
|
|
tp->t_maxseg;
|
|
|
|
|
|
|
|
if (win < 2)
|
|
|
|
win = 2;
|
|
|
|
tp->snd_ssthresh = win * tp->t_maxseg;
|
|
|
|
tp->t_timer[TCPT_REXMT] = 0;
|
|
|
|
tp->t_rtt = 0;
|
|
|
|
tp->snd_nxt = ti->ti_ack;
|
|
|
|
tp->snd_cwnd = tp->t_maxseg;
|
|
|
|
(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) {
|
|
|
|
tp->snd_cwnd += tp->t_maxseg;
|
|
|
|
(void) tcp_output(tp);
|
|
|
|
goto drop;
|
|
|
|
}
|
|
|
|
} else
|
|
|
|
tp->t_dupacks = 0;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* If the congestion window was inflated to account
|
|
|
|
* for the other side's cached packets, retract it.
|
|
|
|
*/
|
1996-03-22 18:09:21 +00:00
|
|
|
if (tp->t_dupacks >= tcprexmtthresh &&
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->snd_cwnd > tp->snd_ssthresh)
|
|
|
|
tp->snd_cwnd = tp->snd_ssthresh;
|
|
|
|
tp->t_dupacks = 0;
|
|
|
|
if (SEQ_GT(ti->ti_ack, tp->snd_max)) {
|
|
|
|
tcpstat.tcps_rcvacktoomuch++;
|
|
|
|
goto dropafterack;
|
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
1995-05-30 08:16:23 +00:00
|
|
|
* If we reach this point, ACK is not a duplicate,
|
1995-02-09 23:13:27 +00:00
|
|
|
* i.e., it ACKs something we sent.
|
|
|
|
*/
|
|
|
|
if (tp->t_flags & TF_NEEDSYN) {
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1996-01-31 08:22:24 +00:00
|
|
|
* T/TCP: Connection was half-synchronized, and our
|
|
|
|
* SYN has been ACK'd (so connection is now fully
|
|
|
|
* synchronized). Go to non-starred state,
|
|
|
|
* increment snd_una for ACK of SYN, and check if
|
|
|
|
* we can do window scaling.
|
1995-02-09 23:13:27 +00:00
|
|
|
*/
|
|
|
|
tp->t_flags &= ~TF_NEEDSYN;
|
|
|
|
tp->snd_una++;
|
1996-01-31 08:22:24 +00:00
|
|
|
/* 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;
|
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
process_ACK:
|
1994-05-24 10:09:53 +00:00
|
|
|
acked = ti->ti_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.
|
|
|
|
*/
|
1995-02-09 23:13:27 +00:00
|
|
|
if (to.to_flag & TOF_TS)
|
|
|
|
tcp_xmit_timer(tp, tcp_now - to.to_tsecr + 1);
|
1994-05-24 10:09:53 +00:00
|
|
|
else if (tp->t_rtt && SEQ_GT(ti->ti_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 (ti->ti_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;
|
1995-02-09 23:13:27 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If no data (only SYN) was ACK'd,
|
|
|
|
* skip rest of ACK processing.
|
|
|
|
*/
|
|
|
|
if (acked == 0)
|
|
|
|
goto step6;
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* 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
|
1994-10-13 18:36:32 +00:00
|
|
|
* (maxseg^2 / cwnd per packet).
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
{
|
|
|
|
register u_int cw = tp->snd_cwnd;
|
|
|
|
register u_int incr = tp->t_maxseg;
|
|
|
|
|
|
|
|
if (cw > tp->snd_ssthresh)
|
1994-10-13 18:36:32 +00:00
|
|
|
incr = incr * incr / cw;
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->snd_cwnd = min(cw + incr, TCP_MAXWIN<<tp->snd_scale);
|
|
|
|
}
|
|
|
|
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 (so->so_snd.sb_flags & SB_NOTIFY)
|
|
|
|
sowwakeup(so);
|
|
|
|
tp->snd_una = ti->ti_ack;
|
|
|
|
if (SEQ_LT(tp->snd_nxt, tp->snd_una))
|
|
|
|
tp->snd_nxt = tp->snd_una;
|
|
|
|
|
|
|
|
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);
|
1995-02-09 23:13:27 +00:00
|
|
|
/* Shorten TIME_WAIT [RFC-1644, p.28] */
|
|
|
|
if (tp->cc_recv != 0 &&
|
|
|
|
tp->t_duration < TCPTV_MSL)
|
|
|
|
tp->t_timer[TCPT_2MSL] =
|
|
|
|
tp->t_rxtcur * TCPTV_TWTRUNC;
|
|
|
|
else
|
|
|
|
tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
|
1994-05-24 10:09:53 +00:00
|
|
|
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) &&
|
1995-05-30 08:16:23 +00:00
|
|
|
(SEQ_LT(tp->snd_wl1, ti->ti_seq) ||
|
1994-10-02 17:48:58 +00:00
|
|
|
(tp->snd_wl1 == ti->ti_seq && (SEQ_LT(tp->snd_wl2, ti->ti_ack) ||
|
|
|
|
(tp->snd_wl2 == ti->ti_ack && tiwin > tp->snd_wnd))))) {
|
1994-05-24 10:09:53 +00:00
|
|
|
/* keep track of pure window updates */
|
|
|
|
if (ti->ti_len == 0 &&
|
|
|
|
tp->snd_wl2 == ti->ti_ack && tiwin > tp->snd_wnd)
|
|
|
|
tcpstat.tcps_rcvwinupd++;
|
|
|
|
tp->snd_wnd = tiwin;
|
|
|
|
tp->snd_wl1 = ti->ti_seq;
|
|
|
|
tp->snd_wl2 = ti->ti_ack;
|
|
|
|
if (tp->snd_wnd > tp->max_sndwnd)
|
|
|
|
tp->max_sndwnd = tp->snd_wnd;
|
|
|
|
needoutput = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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 > sb_max) {
|
|
|
|
ti->ti_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
|
1995-05-30 08:16:23 +00:00
|
|
|
* a FIN has been received from the remote side.
|
1994-05-24 10:09:53 +00:00
|
|
|
* 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
|
1995-05-30 08:16:23 +00:00
|
|
|
* of data past the urgent section as the original
|
1994-05-24 10:09:53 +00:00
|
|
|
* 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_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.
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
if (ti->ti_urp <= (u_long)ti->ti_len
|
1994-05-24 10:09:53 +00:00
|
|
|
#ifdef SO_OOBINLINE
|
|
|
|
&& (so->so_options & SO_OOBINLINE) == 0
|
|
|
|
#endif
|
|
|
|
)
|
|
|
|
tcp_pulloutofband(so, ti, m);
|
|
|
|
} 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: /* XXX */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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_hiwat - (tp->rcv_adv - tp->rcv_nxt);
|
|
|
|
} else {
|
|
|
|
m_freem(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) {
|
|
|
|
socantrcvmore(so);
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* If connection is half-synchronized
|
1995-07-31 10:24:22 +00:00
|
|
|
* (ie NEEDSYN flag on) then delay ACK,
|
1995-02-09 23:13:27 +00:00
|
|
|
* so it may be piggybacked when SYN is sent.
|
|
|
|
* Otherwise, since we received a FIN then no
|
1995-05-30 08:16:23 +00:00
|
|
|
* more input can be expected, send ACK now.
|
1995-02-09 23:13:27 +00:00
|
|
|
*/
|
|
|
|
if (tp->t_flags & TF_NEEDSYN)
|
|
|
|
tp->t_flags |= TF_DELACK;
|
1995-05-30 08:16:23 +00:00
|
|
|
else
|
1995-02-09 23:13:27 +00:00
|
|
|
tp->t_flags |= TF_ACKNOW;
|
1994-05-24 10:09:53 +00:00
|
|
|
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:
|
|
|
|
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,
|
1995-05-30 08:16:23 +00:00
|
|
|
* starting the time-wait timer, turning off the other
|
1994-05-24 10:09:53 +00:00
|
|
|
* standard timers.
|
|
|
|
*/
|
|
|
|
case TCPS_FIN_WAIT_2:
|
|
|
|
tp->t_state = TCPS_TIME_WAIT;
|
|
|
|
tcp_canceltimers(tp);
|
1995-02-09 23:13:27 +00:00
|
|
|
/* Shorten TIME_WAIT [RFC-1644, p.28] */
|
|
|
|
if (tp->cc_recv != 0 &&
|
|
|
|
tp->t_duration < TCPTV_MSL) {
|
|
|
|
tp->t_timer[TCPT_2MSL] =
|
|
|
|
tp->t_rxtcur * TCPTV_TWTRUNC;
|
|
|
|
/* For transaction client, force ACK now. */
|
|
|
|
tp->t_flags |= TF_ACKNOW;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
tp->t_timer[TCPT_2MSL] = 2 * TCPTV_MSL;
|
1994-05-24 10:09:53 +00:00
|
|
|
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;
|
|
|
|
}
|
|
|
|
}
|
1994-09-15 10:36:56 +00:00
|
|
|
#ifdef TCPDEBUG
|
1994-05-24 10:09:53 +00:00
|
|
|
if (so->so_options & SO_DEBUG)
|
|
|
|
tcp_trace(TA_INPUT, ostate, tp, &tcp_saveti, 0);
|
1994-09-15 10:36:56 +00:00
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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;
|
1995-02-09 23:13:27 +00:00
|
|
|
#ifdef TCPDEBUG
|
|
|
|
if (so->so_options & SO_DEBUG)
|
|
|
|
tcp_trace(TA_DROP, ostate, tp, &tcp_saveti, 0);
|
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
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) ||
|
1994-05-26 09:51:33 +00:00
|
|
|
IN_MULTICAST(ntohl(ti->ti_dst.s_addr)))
|
1994-05-24 10:09:53 +00:00
|
|
|
goto drop;
|
1995-02-09 23:13:27 +00:00
|
|
|
#ifdef TCPDEBUG
|
|
|
|
if (tp == 0 || (tp->t_inpcb->inp_socket->so_options & SO_DEBUG))
|
|
|
|
tcp_trace(TA_DROP, ostate, tp, &tcp_saveti, 0);
|
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
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);
|
|
|
|
}
|
|
|
|
/* destroy temporarily created socket */
|
|
|
|
if (dropsocket)
|
|
|
|
(void) soabort(so);
|
|
|
|
return;
|
|
|
|
|
|
|
|
drop:
|
|
|
|
/*
|
|
|
|
* Drop space held by incoming segment and return.
|
|
|
|
*/
|
1994-09-15 10:36:56 +00:00
|
|
|
#ifdef TCPDEBUG
|
1995-02-09 23:13:27 +00:00
|
|
|
if (tp == 0 || (tp->t_inpcb->inp_socket->so_options & SO_DEBUG))
|
|
|
|
tcp_trace(TA_DROP, ostate, tp, &tcp_saveti, 0);
|
1994-09-15 10:36:56 +00:00
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
m_freem(m);
|
|
|
|
/* destroy temporarily created socket */
|
|
|
|
if (dropsocket)
|
|
|
|
(void) soabort(so);
|
|
|
|
return;
|
|
|
|
#ifndef TUBA_INCLUDE
|
|
|
|
}
|
|
|
|
|
1995-11-14 20:34:56 +00:00
|
|
|
static void
|
1995-02-09 23:13:27 +00:00
|
|
|
tcp_dooptions(tp, cp, cnt, ti, to)
|
1994-05-24 10:09:53 +00:00
|
|
|
struct tcpcb *tp;
|
|
|
|
u_char *cp;
|
|
|
|
int cnt;
|
|
|
|
struct tcpiphdr *ti;
|
1995-02-09 23:13:27 +00:00
|
|
|
struct tcpopt *to;
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
1995-02-09 23:13:27 +00:00
|
|
|
u_short mss = 0;
|
1994-05-24 10:09:53 +00:00
|
|
|
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);
|
|
|
|
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;
|
1995-02-09 23:13:27 +00:00
|
|
|
to->to_flag |= TOF_TS;
|
|
|
|
bcopy((char *)cp + 2,
|
|
|
|
(char *)&to->to_tsval, sizeof(to->to_tsval));
|
|
|
|
NTOHL(to->to_tsval);
|
|
|
|
bcopy((char *)cp + 6,
|
|
|
|
(char *)&to->to_tsecr, sizeof(to->to_tsecr));
|
|
|
|
NTOHL(to->to_tsecr);
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1994-05-24 10:09:53 +00:00
|
|
|
* 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;
|
1995-02-09 23:13:27 +00:00
|
|
|
tp->ts_recent = to->to_tsval;
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->ts_recent_age = tcp_now;
|
|
|
|
}
|
|
|
|
break;
|
1995-02-09 23:13:27 +00:00
|
|
|
case TCPOPT_CC:
|
|
|
|
if (optlen != TCPOLEN_CC)
|
|
|
|
continue;
|
1995-05-09 12:32:06 +00:00
|
|
|
to->to_flag |= TOF_CC;
|
1995-02-09 23:13:27 +00:00
|
|
|
bcopy((char *)cp + 2,
|
|
|
|
(char *)&to->to_cc, sizeof(to->to_cc));
|
|
|
|
NTOHL(to->to_cc);
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1995-02-09 23:13:27 +00:00
|
|
|
* A CC or CC.new option received in a SYN makes
|
|
|
|
* it ok to send CC in subsequent segments.
|
|
|
|
*/
|
|
|
|
if (ti->ti_flags & TH_SYN)
|
|
|
|
tp->t_flags |= TF_RCVD_CC;
|
|
|
|
break;
|
|
|
|
case TCPOPT_CCNEW:
|
|
|
|
if (optlen != TCPOLEN_CC)
|
|
|
|
continue;
|
|
|
|
if (!(ti->ti_flags & TH_SYN))
|
|
|
|
continue;
|
|
|
|
to->to_flag |= TOF_CCNEW;
|
|
|
|
bcopy((char *)cp + 2,
|
|
|
|
(char *)&to->to_cc, sizeof(to->to_cc));
|
|
|
|
NTOHL(to->to_cc);
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1995-02-09 23:13:27 +00:00
|
|
|
* A CC or CC.new option received in a SYN makes
|
|
|
|
* it ok to send CC in subsequent segments.
|
|
|
|
*/
|
|
|
|
tp->t_flags |= TF_RCVD_CC;
|
|
|
|
break;
|
|
|
|
case TCPOPT_CCECHO:
|
|
|
|
if (optlen != TCPOLEN_CC)
|
|
|
|
continue;
|
|
|
|
if (!(ti->ti_flags & TH_SYN))
|
|
|
|
continue;
|
|
|
|
to->to_flag |= TOF_CCECHO;
|
|
|
|
bcopy((char *)cp + 2,
|
|
|
|
(char *)&to->to_ccecho, sizeof(to->to_ccecho));
|
|
|
|
NTOHL(to->to_ccecho);
|
|
|
|
break;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
if (ti->ti_flags & TH_SYN)
|
|
|
|
tcp_mss(tp, mss); /* sets t_maxseg */
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
1995-11-14 20:34:56 +00:00
|
|
|
static void
|
1994-05-24 10:09:53 +00:00
|
|
|
tcp_pulloutofband(so, ti, m)
|
|
|
|
struct socket *so;
|
|
|
|
struct tcpiphdr *ti;
|
|
|
|
register struct mbuf *m;
|
|
|
|
{
|
|
|
|
int cnt = ti->ti_urp - 1;
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
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.
|
|
|
|
*/
|
1995-11-14 20:34:56 +00:00
|
|
|
static void
|
1994-05-24 10:09:53 +00:00
|
|
|
tcp_xmit_timer(tp, rtt)
|
|
|
|
register struct tcpcb *tp;
|
|
|
|
short rtt;
|
|
|
|
{
|
1996-03-22 18:09:21 +00:00
|
|
|
#ifdef notdef
|
1994-05-24 10:09:53 +00:00
|
|
|
register short delta;
|
|
|
|
|
|
|
|
tcpstat.tcps_rttupdated++;
|
1995-06-29 18:11:24 +00:00
|
|
|
tp->t_rttupdated++;
|
1994-05-24 10:09:53 +00:00
|
|
|
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 {
|
1995-05-30 08:16:23 +00:00
|
|
|
/*
|
1994-05-24 10:09:53 +00:00
|
|
|
* 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);
|
|
|
|
}
|
1996-03-22 18:09:21 +00:00
|
|
|
#else /* Peterson paper */
|
|
|
|
register int delta;
|
|
|
|
|
|
|
|
tcpstat.tcps_rttupdated++;
|
|
|
|
tp->t_rttupdated++;
|
|
|
|
if (tp->t_srtt != 0) {
|
|
|
|
/*
|
|
|
|
* srtt is stored as fixed point with 5 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) << TCP_DELTA_SHIFT)
|
|
|
|
- (tp->t_srtt >> (TCP_RTT_SHIFT - TCP_DELTA_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 4 bits after the
|
|
|
|
* binary point (scaled by 16). 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 - TCP_DELTA_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);
|
|
|
|
}
|
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
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).
|
|
|
|
*/
|
1996-03-22 18:09:21 +00:00
|
|
|
#ifdef notdef
|
1994-05-24 10:09:53 +00:00
|
|
|
TCPT_RANGESET(tp->t_rxtcur, TCP_REXMTVAL(tp),
|
|
|
|
tp->t_rttmin, TCPTV_REXMTMAX);
|
1996-03-22 18:09:21 +00:00
|
|
|
#else /* Peterson */
|
|
|
|
TCPT_RANGESET(tp->t_rxtcur, TCP_REXMTVAL(tp),
|
|
|
|
max(tp->t_rttmin, TCPTV_MIN + rtt - 1), TCPTV_REXMTMAX);
|
|
|
|
#endif
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* 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.
|
1995-02-09 23:13:27 +00:00
|
|
|
*
|
|
|
|
* Also take into account the space needed for options that we
|
|
|
|
* send regularly. Make maxseg shorter by that amount to assure
|
|
|
|
* that we can send maxseg amount of data even when the options
|
|
|
|
* are present. Store the upper limit of the length of options plus
|
|
|
|
* data in maxopd.
|
|
|
|
*
|
|
|
|
* NOTE that this routine is only called when we process an incoming
|
|
|
|
* segment, for outgoing segments only tcp_mssopt is called.
|
|
|
|
*
|
|
|
|
* In case of T/TCP, we call this routine during implicit connection
|
|
|
|
* setup as well (offer = -1), to initialize maxseg from the cached
|
|
|
|
* MSS of our peer.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1995-02-09 23:13:27 +00:00
|
|
|
void
|
1994-05-24 10:09:53 +00:00
|
|
|
tcp_mss(tp, offer)
|
1995-02-09 23:13:27 +00:00
|
|
|
struct tcpcb *tp;
|
|
|
|
int offer;
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
|
|
|
register struct rtentry *rt;
|
|
|
|
struct ifnet *ifp;
|
|
|
|
register int rtt, mss;
|
|
|
|
u_long bufsize;
|
|
|
|
struct inpcb *inp;
|
|
|
|
struct socket *so;
|
1995-02-09 23:13:27 +00:00
|
|
|
struct rmxp_tao *taop;
|
|
|
|
int origoffer = offer;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
inp = tp->t_inpcb;
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((rt = tcp_rtlookup(inp)) == NULL) {
|
|
|
|
tp->t_maxopd = tp->t_maxseg = tcp_mssdflt;
|
|
|
|
return;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
ifp = rt->rt_ifp;
|
|
|
|
so = inp->inp_socket;
|
|
|
|
|
1995-02-09 23:13:27 +00:00
|
|
|
taop = rmx_taop(rt->rt_rmx);
|
|
|
|
/*
|
|
|
|
* Offer == -1 means that we didn't receive SYN yet,
|
|
|
|
* use cached value in that case;
|
|
|
|
*/
|
|
|
|
if (offer == -1)
|
|
|
|
offer = taop->tao_mssopt;
|
|
|
|
/*
|
|
|
|
* Offer == 0 means that there was no MSS on the SYN segment,
|
|
|
|
* in this case we use tcp_mssdflt.
|
|
|
|
*/
|
|
|
|
if (offer == 0)
|
|
|
|
offer = tcp_mssdflt;
|
|
|
|
else
|
|
|
|
/*
|
|
|
|
* Sanity check: make sure that maxopd will be large
|
|
|
|
* enough to allow some data on segments even is the
|
|
|
|
* all the option space is used (40bytes). Otherwise
|
|
|
|
* funny things may happen in tcp_output.
|
|
|
|
*/
|
|
|
|
offer = max(offer, 64);
|
|
|
|
taop->tao_mssopt = offer;
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* 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)) {
|
|
|
|
/*
|
1995-02-09 23:13:27 +00:00
|
|
|
* XXX the lock bit for RTT indicates that the value
|
1994-05-24 10:09:53 +00:00
|
|
|
* 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));
|
1995-07-10 15:39:16 +00:00
|
|
|
tcpstat.tcps_usedrtt++;
|
|
|
|
if (rt->rt_rmx.rmx_rttvar) {
|
1994-05-24 10:09:53 +00:00
|
|
|
tp->t_rttvar = rt->rt_rmx.rmx_rttvar /
|
|
|
|
(RTM_RTTUNIT / (PR_SLOWHZ * TCP_RTTVAR_SCALE));
|
1995-07-10 15:39:16 +00:00
|
|
|
tcpstat.tcps_usedrttvar++;
|
|
|
|
} else {
|
1994-05-24 10:09:53 +00:00
|
|
|
/* default variation is +- 1 rtt */
|
|
|
|
tp->t_rttvar =
|
|
|
|
tp->t_srtt * TCP_RTTVAR_SCALE / TCP_RTT_SCALE;
|
1995-07-10 15:39:16 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
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
|
|
|
|
{
|
|
|
|
mss = ifp->if_mtu - sizeof(struct tcpiphdr);
|
1995-02-09 23:13:27 +00:00
|
|
|
if (!in_localaddr(inp->inp_faddr))
|
|
|
|
mss = min(mss, tcp_mssdflt);
|
|
|
|
}
|
|
|
|
mss = min(mss, offer);
|
|
|
|
/*
|
|
|
|
* maxopd stores the maximum length of data AND options
|
|
|
|
* in a segment; maxseg is the amount of data in a normal
|
|
|
|
* segment. We need to store this value (maxopd) apart
|
|
|
|
* from maxseg, because now every segment carries options
|
|
|
|
* and thus we normally have somewhat less data in segments.
|
|
|
|
*/
|
|
|
|
tp->t_maxopd = mss;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In case of T/TCP, origoffer==-1 indicates, that no segments
|
|
|
|
* were received yet. In this case we just guess, otherwise
|
|
|
|
* we do the same as before T/TCP.
|
|
|
|
*/
|
|
|
|
if ((tp->t_flags & (TF_REQ_TSTMP|TF_NOOPT)) == TF_REQ_TSTMP &&
|
|
|
|
(origoffer == -1 ||
|
|
|
|
(tp->t_flags & TF_RCVD_TSTMP) == TF_RCVD_TSTMP))
|
|
|
|
mss -= TCPOLEN_TSTAMP_APPA;
|
|
|
|
if ((tp->t_flags & (TF_REQ_CC|TF_NOOPT)) == TF_REQ_CC &&
|
|
|
|
(origoffer == -1 ||
|
|
|
|
(tp->t_flags & TF_RCVD_CC) == TF_RCVD_CC))
|
|
|
|
mss -= TCPOLEN_CC_APPA;
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
#if (MCLBYTES & (MCLBYTES - 1)) == 0
|
|
|
|
if (mss > MCLBYTES)
|
|
|
|
mss &= ~(MCLBYTES-1);
|
|
|
|
#else
|
|
|
|
if (mss > MCLBYTES)
|
|
|
|
mss = mss / MCLBYTES * MCLBYTES;
|
|
|
|
#endif
|
|
|
|
/*
|
1995-02-09 23:13:27 +00:00
|
|
|
* 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.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
#ifdef RTV_SPIPE
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((bufsize = rt->rt_rmx.rmx_sendpipe) == 0)
|
1994-05-24 10:09:53 +00:00
|
|
|
#endif
|
1995-02-09 23:13:27 +00:00
|
|
|
bufsize = so->so_snd.sb_hiwat;
|
|
|
|
if (bufsize < mss)
|
|
|
|
mss = bufsize;
|
|
|
|
else {
|
|
|
|
bufsize = roundup(bufsize, mss);
|
|
|
|
if (bufsize > sb_max)
|
|
|
|
bufsize = sb_max;
|
|
|
|
(void)sbreserve(&so->so_snd, bufsize);
|
|
|
|
}
|
|
|
|
tp->t_maxseg = mss;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
#ifdef RTV_RPIPE
|
1995-02-09 23:13:27 +00:00
|
|
|
if ((bufsize = rt->rt_rmx.rmx_recvpipe) == 0)
|
1994-05-24 10:09:53 +00:00
|
|
|
#endif
|
1995-02-09 23:13:27 +00:00
|
|
|
bufsize = so->so_rcv.sb_hiwat;
|
|
|
|
if (bufsize > mss) {
|
|
|
|
bufsize = roundup(bufsize, mss);
|
|
|
|
if (bufsize > sb_max)
|
|
|
|
bufsize = sb_max;
|
|
|
|
(void)sbreserve(&so->so_rcv, bufsize);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
/*
|
|
|
|
* Don't force slow-start on local network.
|
|
|
|
*/
|
|
|
|
if (!in_localaddr(inp->inp_faddr))
|
|
|
|
tp->snd_cwnd = mss;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
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);
|
1995-07-10 15:39:16 +00:00
|
|
|
tcpstat.tcps_usedssthresh++;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
1995-02-09 23:13:27 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Determine the MSS option to send on an outgoing SYN.
|
|
|
|
*/
|
|
|
|
int
|
|
|
|
tcp_mssopt(tp)
|
|
|
|
struct tcpcb *tp;
|
|
|
|
{
|
|
|
|
struct rtentry *rt;
|
|
|
|
|
|
|
|
rt = tcp_rtlookup(tp->t_inpcb);
|
|
|
|
if (rt == NULL)
|
|
|
|
return tcp_mssdflt;
|
|
|
|
|
|
|
|
return rt->rt_ifp->if_mtu - sizeof(struct tcpiphdr);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
#endif /* TUBA_INCLUDE */
|