1
0
mirror of https://git.FreeBSD.org/src.git synced 2024-12-22 11:17:19 +00:00
freebsd/usr.sbin/IPXrouted/sap_input.c
1999-08-28 01:35:59 +00:00

216 lines
6.2 KiB
C

/*
* Copyright (c) 1995 John Hay. 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 John Hay.
* 4. Neither the name of the author nor the names of any co-contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY John Hay 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 John Hay 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.
*
* $FreeBSD$
*/
/*
* IPX Routing Table Management Daemon
*/
#include "defs.h"
int dognreply = 1;
/*
* Process a newly received packet.
*/
void
sap_input(from, size)
struct sockaddr *from;
int size;
{
int newsize;
int sapchanged = 0;
struct sap_entry *sap;
struct sap_info *n;
struct interface *ifp = 0;
struct afswitch *afp;
struct sockaddr_ipx *ipxp;
ifp = if_ifwithnet(from);
ipxp = (struct sockaddr_ipx *)from;
if (ifp == 0) {
if(ftrace) {
fprintf(ftrace, "Received bogus packet from %s\n",
ipxdp_ntoa(&ipxp->sipx_addr));
}
return;
}
if (ftrace)
dumpsappacket(ftrace, "received", from, (char *)sap_msg , size);
if (from->sa_family >= AF_MAX)
return;
afp = &afswitch[from->sa_family];
size -= sizeof (u_short) /* command */;
n = sap_msg->sap;
switch (ntohs(sap_msg->sap_cmd)) {
case SAP_REQ_NEAR:
if (ftrace)
fprintf(ftrace, "Received a sap REQ_NEAR packet.\n");
if (!dognreply)
return;
sap = sap_nearestserver(n->ServType, ifp);
if (sap == NULL)
return;
sap_msg->sap_cmd = htons(SAP_RESP_NEAR);
*n = sap->sap;
n->hops = htons(ntohs(n->hops) + 1);
if (ntohs(n->hops) >= HOPCNT_INFINITY)
return;
newsize = sizeof(struct sap_info) + sizeof(struct sap_packet);
(*afp->af_output)(sapsock, 0, from, newsize);
if (ftrace) {
fprintf(ftrace, "sap_nearestserver %X %s returned:\n",
ntohs(n->ServType),
ifp->int_name);
fprintf(ftrace, " service %04X %-20.20s "
"addr %s.%04X metric %d\n",
ntohs(sap->sap.ServType),
sap->sap.ServName,
ipxdp_ntoa(&sap->sap.ipx),
ntohs(sap->sap.ipx.x_port),
ntohs(sap->sap.hops));
}
return;
case SAP_REQ:
if (ftrace)
fprintf(ftrace, "Received a sap REQ packet.\n");
sap_supply(from, 0, ifp, n->ServType, 0);
return;
case SAP_RESP_NEAR:
/* XXX We do nothing here, for the moment.
* Maybe we should check if the service is in our table?
*
*/
if (ftrace)
fprintf(ftrace, "Received a sap RESP_NEAR packet.\n");
return;
case SAP_RESP:
if (ftrace)
fprintf(ftrace, "Received a sap RESP packet.\n");
(*afp->af_canon)(from);
for (; size > 0; size -= sizeof (struct sap_info), n++) {
if (size < sizeof (struct netinfo))
break;
/*
* The idea here is that if the hop count is more
* than INFINITY it is bogus and should be discarded.
* If it is equal to INFINITY it is a message to say
* that a service went down. If we don't allready
* have it in our tables discard it. Otherwise
* update our table and set the timer to EXPIRE_TIME
* so that it is removed next time we go through the
* tables.
*/
if (ntohs(n->hops) > HOPCNT_INFINITY)
continue;
sap = sap_lookup(n->ServType, n->ServName);
if (sap == 0) {
if (ntohs(n->hops) == HOPCNT_INFINITY)
continue;
sap_add(n, from);
sapchanged = 1;
continue;
}
/*
* A clone is a different route to the same service
* with exactly the same cost (metric).
* They must all be recorded because those interfaces
* must be handled in the same way as the first route
* to that service. ie When using the split horizon
* algorithm we must look at these interfaces also.
*
* Update if from gateway and different,
* from anywhere and less hops or
* getting stale and equivalent.
*/
if (((ifp != sap->ifp) ||
!equal(&sap->source, from)) &&
(n->hops == sap->sap.hops) &&
(ntohs(n->hops) != HOPCNT_INFINITY)) {
register struct sap_entry *tsap = sap->clone;
while (tsap) {
if ((ifp == tsap->ifp) &&
equal(&tsap->source, from)) {
tsap->timer = 0;
break;
}
tsap = tsap->clone;
}
if (tsap == NULL) {
sap_add_clone(sap, n, from);
}
continue;
}
if ((ifp == sap->ifp) &&
equal(&sap->source, from) &&
(ntohs(n->hops) == ntohs(sap->sap.hops)))
sap->timer = 0;
else if (((ifp == sap->ifp) &&
equal(&sap->source, from) &&
(n->hops != sap->sap.hops)) ||
(ntohs(n->hops) < ntohs(sap->sap.hops)) ||
(sap->timer > (EXPIRE_TIME*2/3) &&
ntohs(sap->sap.hops) == ntohs(n->hops) &&
ntohs(n->hops) != HOPCNT_INFINITY)) {
sap_change(sap, n, from);
sapchanged = 1;
}
}
if (sapchanged) {
register struct sap_entry *sap;
register struct sap_hash *sh;
sap_supply_toall(1);
for (sh = sap_head; sh < &sap_head[SAPHASHSIZ]; sh++)
for (sap = sh->forw;
sap != (struct sap_entry *)sh;
sap = sap->forw)
sap->state &= ~RTS_CHANGED;
}
return;
}
}