path: root/sys/netinet/ip_icmp.c

/*	$OpenBSD: ip_icmp.c,v 1.37 2001/06/08 03:53:46 angelos Exp $	*/
/*	$NetBSD: ip_icmp.c,v 1.19 1996/02/13 23:42:22 christos Exp $	*/

/*
 * Copyright (c) 1982, 1986, 1988, 1993
 *	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.
 *
 *	@(#)COPYRIGHT	1.1 (NRL) 17 January 1995
 * 
 * NRL grants permission for redistribution and use in source and binary
 * forms, with or without modification, of the software and documentation
 * created at NRL 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 acknowledgements:
 * 	This product includes software developed by the University of
 * 	California, Berkeley and its contributors.
 * 	This product includes software developed at the Information
 * 	Technology Division, US Naval Research Laboratory.
 * 4. Neither the name of the NRL nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 * 
 * THE SOFTWARE PROVIDED BY NRL IS PROVIDED BY NRL 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 NRL 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.
 * 
 * The views and conclusions contained in the software and documentation
 * are those of the authors and should not be interpreted as representing
 * official policies, either expressed or implied, of the US Naval
 * Research Laboratory (NRL).
 */

#include <sys/param.h>
#include <sys/mbuf.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/sysctl.h>

#include <net/if.h>
#include <net/route.h>

#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/in_var.h>
#include <netinet/ip.h>
#include <netinet/ip_icmp.h>
#include <netinet/ip_var.h>
#include <netinet/icmp_var.h>

/*
 * ICMP routines: error generation, receive packet processing, and
 * routines to turnaround packets back to the originator, and
 * host table maintenance routines.
 */

int	icmpmaskrepl = 0;
int	icmpbmcastecho = 0;
#ifdef ICMPPRINTFS
int	icmpprintfs = 0;
#endif
int	icmperrppslim = 100;
int	icmperrpps_count = 0;
struct timeval icmperrppslim_last;

void icmp_mtudisc_timeout __P((struct rtentry *, struct rttimer *));
int icmp_ratelimit __P((const struct in_addr *, const int, const int));

extern	struct protosw inetsw[];

/*
 * Generate an error packet of type error
 * in response to bad packet ip.
 *
 * The ip packet inside has ip_off and ip_len in host byte order.
 */
void
icmp_error(n, type, code, dest, destifp)
	struct mbuf *n;
	int type, code;
	n_long dest;
	struct ifnet *destifp;
{
	register struct ip *oip = mtod(n, struct ip *), *nip;
	register unsigned oiplen = oip->ip_hl << 2;
	register struct icmp *icp;
	struct mbuf *m, m0;
	unsigned icmplen;

#ifdef ICMPPRINTFS
	if (icmpprintfs)
		printf("icmp_error(%x, %d, %d)\n", oip, type, code);
#endif
	if (type != ICMP_REDIRECT)
		icmpstat.icps_error++;
	/*
	 * Don't send error if not the first fragment of message.
	 * Don't error if the old packet protocol was ICMP
	 * error message, only known informational types.
	 */
	if (oip->ip_off &~ (IP_MF|IP_DF))
		goto freeit;
	if (oip->ip_p == IPPROTO_ICMP && type != ICMP_REDIRECT &&
	  n->m_len >= oiplen + ICMP_MINLEN &&
	  !ICMP_INFOTYPE(((struct icmp *)((caddr_t)oip + oiplen))->icmp_type)) {
		icmpstat.icps_oldicmp++;
		goto freeit;
	}
	/* Don't send error in response to a multicast or broadcast packet */
	if (n->m_flags & (M_BCAST|M_MCAST))
		goto freeit;

	/*
	 * First, do a rate limitation check.
 	 */
	if (icmp_ratelimit(&oip->ip_src, type, code)) {
		/* XXX stat */
		goto freeit;
	}

	/*
	 * Now, formulate icmp message
	 */
	m = m_gethdr(M_DONTWAIT, MT_HEADER);
	if (m == NULL)
		goto freeit;
	icmplen = oiplen + min(8, oip->ip_len);
	m->m_len = icmplen + ICMP_MINLEN;
	MH_ALIGN(m, m->m_len);
	icp = mtod(m, struct icmp *);
	if ((u_int)type > ICMP_MAXTYPE)
		panic("icmp_error");
	icmpstat.icps_outhist[type]++;
	icp->icmp_type = type;
	if (type == ICMP_REDIRECT)
		icp->icmp_gwaddr.s_addr = dest;
	else {
		icp->icmp_void = 0;
		/* 
		 * The following assignments assume an overlay with the
		 * zeroed icmp_void field.
		 */
		if (type == ICMP_PARAMPROB) {
			icp->icmp_pptr = code;
			code = 0;
		} else if (type == ICMP_UNREACH &&
		    code == ICMP_UNREACH_NEEDFRAG && destifp)
			icp->icmp_nextmtu = htons(destifp->if_mtu);
	}

	icp->icmp_code = code;
	bcopy((caddr_t)oip, (caddr_t)&icp->icmp_ip, icmplen);
	nip = &icp->icmp_ip;
	nip->ip_off = htons(nip->ip_off);
	nip->ip_len = htons(nip->ip_len);

	m0.m_next = NULL;			/* correct nip->ip_sum */
	m0.m_data = (char *)nip;
	m0.m_len = nip->ip_hl << 2;
	nip->ip_sum = 0;
	nip->ip_sum = in_cksum(&m0, nip->ip_hl << 2);

	/*
	 * Now, copy old ip header (without options)
	 * in front of icmp message.
	 */
	if (m->m_data - sizeof(struct ip) < m->m_pktdat)
		panic("icmp len");
	m->m_data -= sizeof(struct ip);
	m->m_len += sizeof(struct ip);
	m->m_pkthdr.len = m->m_len;
	m->m_pkthdr.rcvif = n->m_pkthdr.rcvif;
	nip = mtod(m, struct ip *);
	bcopy((caddr_t)oip, (caddr_t)nip, sizeof(struct ip));
	nip->ip_off = htons(nip->ip_off);
	nip->ip_len = m->m_len;
	nip->ip_hl = sizeof(struct ip) >> 2;
	nip->ip_p = IPPROTO_ICMP;
	nip->ip_tos = 0;
	icmp_reflect(m);

freeit:
	m_freem(n);
}

static struct sockaddr_in icmpsrc = { sizeof (struct sockaddr_in), AF_INET };
static struct sockaddr_in icmpdst = { sizeof (struct sockaddr_in), AF_INET };
static struct sockaddr_in icmpgw = { sizeof (struct sockaddr_in), AF_INET };
struct sockaddr_in icmpmask = { 8, 0 };

/*
 * Process a received ICMP message.
 */
void
#if __STDC__
icmp_input(struct mbuf *m, ...)
#else
icmp_input(m, va_alist)
	struct mbuf *m;
	va_dcl
#endif
{
	register struct icmp *icp;
	register struct ip *ip = mtod(m, struct ip *);
	int icmplen = ip->ip_len;
	register int i;
	struct in_ifaddr *ia;
	void *(*ctlfunc) __P((int, struct sockaddr *, void *));
	int code;
	extern u_char ip_protox[];
	int hlen;
	va_list ap;

	va_start(ap, m);
	hlen = va_arg(ap, int);
	va_end(ap);

	/*
	 * Locate icmp structure in mbuf, and check
	 * that not corrupted and of at least minimum length.
	 */
#ifdef ICMPPRINTFS
	if (icmpprintfs) {
		char buf[4*sizeof "123"];

		strcpy(buf, inet_ntoa(ip->ip_dst));
		printf("icmp_input from %s to %s, len %d\n",
			inet_ntoa(ip->ip_src), buf, icmplen);
	}
#endif
	if (icmplen < ICMP_MINLEN) {
		icmpstat.icps_tooshort++;
		goto freeit;
	}
	i = hlen + min(icmplen, ICMP_ADVLENMIN);
	if (m->m_len < i && (m = m_pullup(m, i)) == NULL)  {
		icmpstat.icps_tooshort++;
		return;
	}
	ip = mtod(m, struct ip *);
	m->m_len -= hlen;
	m->m_data += hlen;
	icp = mtod(m, struct icmp *);
	if (in_cksum(m, icmplen)) {
		icmpstat.icps_checksum++;
		goto freeit;
	}
	m->m_len += hlen;
	m->m_data -= hlen;

#ifdef ICMPPRINTFS
	/*
	 * Message type specific processing.
	 */
	if (icmpprintfs)
		printf("icmp_input, type %d code %d\n", icp->icmp_type,
		    icp->icmp_code);
#endif
	if (icp->icmp_type > ICMP_MAXTYPE)
		goto raw;
	icmpstat.icps_inhist[icp->icmp_type]++;
	code = icp->icmp_code;
	switch (icp->icmp_type) {

	case ICMP_UNREACH:
		switch (code) {
		case ICMP_UNREACH_NET:
		case ICMP_UNREACH_HOST:
		case ICMP_UNREACH_PROTOCOL:
		case ICMP_UNREACH_PORT:
		case ICMP_UNREACH_SRCFAIL:
			code += PRC_UNREACH_NET;
			break;

		case ICMP_UNREACH_NEEDFRAG:
#if 0 /*NRL INET6*/
			if (icp->icmp_nextmtu) {
				extern int ipv6_trans_mtu
				    __P((struct mbuf **, int, int));
				struct mbuf *m0 = m;

				/*
				 * Do cool v4-related path MTU, for now,
				 * only v6-in-v4 can handle it.
				 */
				if (icmplen >= ICMP_V6ADVLENMIN &&
				    icmplen >= ICMP_V6ADVLEN(icp) &&
				    icp->icmp_ip.ip_p == IPPROTO_IPV6) {
					/*
					 * ipv6_trans_mtu returns 1 if
					 * the mbuf is still intact.
					 */
					if (ipv6_trans_mtu(&m0,icp->icmp_nextmtu,
					    hlen + ICMP_V6ADVLEN(icp))) {
						m = m0;
						goto raw;
					} else
						return;
				}
			}
#endif /* INET6 */
			code = PRC_MSGSIZE;
			break;
				
		case ICMP_UNREACH_NET_UNKNOWN:
		case ICMP_UNREACH_NET_PROHIB:
		case ICMP_UNREACH_TOSNET:
			code = PRC_UNREACH_NET;
			break;

		case ICMP_UNREACH_HOST_UNKNOWN:
		case ICMP_UNREACH_ISOLATED:
		case ICMP_UNREACH_HOST_PROHIB:
		case ICMP_UNREACH_TOSHOST:
		case ICMP_UNREACH_FILTER_PROHIB:
		case ICMP_UNREACH_HOST_PRECEDENCE:
		case ICMP_UNREACH_PRECEDENCE_CUTOFF:
			code = PRC_UNREACH_HOST;
			break;

		default:
			goto badcode;
		}
		goto deliver;

	case ICMP_TIMXCEED:
		if (code > 1)
			goto badcode;
		code += PRC_TIMXCEED_INTRANS;
		goto deliver;

	case ICMP_PARAMPROB:
		if (code > 1)
			goto badcode;
		code = PRC_PARAMPROB;
		goto deliver;

	case ICMP_SOURCEQUENCH:
		if (code)
			goto badcode;
		code = PRC_QUENCH;
	deliver:
		/* Free packet atttributes */
		if (m->m_flags & M_PKTHDR)
			m_tag_delete_chain(m, NULL);

		/*
		 * Problem with datagram; advise higher level routines.
		 */
		if (icmplen < ICMP_ADVLENMIN || icmplen < ICMP_ADVLEN(icp) ||
		    icp->icmp_ip.ip_hl < (sizeof(struct ip) >> 2)) {
			icmpstat.icps_badlen++;
			goto freeit;
		}
		if (IN_MULTICAST(icp->icmp_ip.ip_dst.s_addr))
			goto badcode;
		NTOHS(icp->icmp_ip.ip_len);
#ifdef INET6
		/* Get more contiguous data for a v6 in v4 ICMP message. */
		if (icp->icmp_ip.ip_p == IPPROTO_IPV6) {
			if (icmplen < ICMP_V6ADVLENMIN || 
			    icmplen < ICMP_V6ADVLEN(icp)) {
				icmpstat.icps_badlen++;
				goto freeit;
			} else {
				if ((m = m_pullup(m, (ip->ip_hl << 2) +
				    ICMP_V6ADVLEN(icp))) == NULL) {
					icmpstat.icps_tooshort++;
					return;
				}
				ip = mtod(m, struct ip *);
				icp = (struct icmp *)(m->m_data + (ip->ip_hl << 2));
			}
		}
#endif /* INET6 */
#ifdef ICMPPRINTFS
		if (icmpprintfs)
			printf("deliver to protocol %d\n", icp->icmp_ip.ip_p);
#endif
		icmpsrc.sin_addr = icp->icmp_ip.ip_dst;
		/*
		 * XXX if the packet contains [IPv4 AH TCP], we can't make a
		 * notification to TCP layer.
		 */
		ctlfunc = inetsw[ip_protox[icp->icmp_ip.ip_p]].pr_ctlinput;
		if (ctlfunc)
			(*ctlfunc)(code, sintosa(&icmpsrc), &icp->icmp_ip);
		break;

	badcode:
		icmpstat.icps_badcode++;
		break;

	case ICMP_ECHO:
		if (!icmpbmcastecho &&
		    (m->m_flags & (M_MCAST | M_BCAST)) != 0) {
			icmpstat.icps_bmcastecho++;
			break;
		}
		icp->icmp_type = ICMP_ECHOREPLY;
		goto reflect;

	case ICMP_TSTAMP:
		if (!icmpbmcastecho &&
		    (m->m_flags & (M_MCAST | M_BCAST)) != 0) {
			icmpstat.icps_bmcastecho++;
			break;
		}
		if (icmplen < ICMP_TSLEN) {
			icmpstat.icps_badlen++;
			break;
		}
		icp->icmp_type = ICMP_TSTAMPREPLY;
		icp->icmp_rtime = iptime();
		icp->icmp_ttime = icp->icmp_rtime;	/* bogus, do later! */
		goto reflect;
		
	case ICMP_MASKREQ:
		if (icmpmaskrepl == 0)
			break;
		/*
		 * We are not able to respond with all ones broadcast
		 * unless we receive it over a point-to-point interface.
		 */
		if (icmplen < ICMP_MASKLEN) {
			icmpstat.icps_badlen++;
			break;
		}
		if (ip->ip_dst.s_addr == INADDR_BROADCAST ||
		    ip->ip_dst.s_addr == INADDR_ANY)
			icmpdst.sin_addr = ip->ip_src;
		else
			icmpdst.sin_addr = ip->ip_dst;
		ia = ifatoia(ifaof_ifpforaddr(sintosa(&icmpdst),
		    m->m_pkthdr.rcvif));
		if (ia == 0)
			break;
		icp->icmp_type = ICMP_MASKREPLY;
		icp->icmp_mask = ia->ia_sockmask.sin_addr.s_addr;
		if (ip->ip_src.s_addr == 0) {
			if (ia->ia_ifp->if_flags & IFF_BROADCAST)
				ip->ip_src = ia->ia_broadaddr.sin_addr;
			else if (ia->ia_ifp->if_flags & IFF_POINTOPOINT)
				ip->ip_src = ia->ia_dstaddr.sin_addr;
		}
reflect:
		/* Free packet atttributes */
		if (m->m_flags & M_PKTHDR)
			m_tag_delete_chain(m, NULL);

		ip->ip_len += hlen;	/* since ip_input deducts this */
		icmpstat.icps_reflect++;
		icmpstat.icps_outhist[icp->icmp_type]++;
		icmp_reflect(m);
		return;

	case ICMP_REDIRECT:
		/* Free packet atttributes */
		if (m->m_flags & M_PKTHDR)
			m_tag_delete_chain(m, NULL);
		if (code > 3)
			goto badcode;
		if (icmplen < ICMP_ADVLENMIN || icmplen < ICMP_ADVLEN(icp) ||
		    icp->icmp_ip.ip_hl < (sizeof(struct ip) >> 2)) {
			icmpstat.icps_badlen++;
			break;
		}
		/*
		 * Short circuit routing redirects to force
		 * immediate change in the kernel's routing
		 * tables.  The message is also handed to anyone
		 * listening on a raw socket (e.g. the routing
		 * daemon for use in updating its tables).
		 */
		icmpgw.sin_addr = ip->ip_src;
		icmpdst.sin_addr = icp->icmp_gwaddr;
#ifdef	ICMPPRINTFS
		if (icmpprintfs) {
			char buf[4 * sizeof "123"];
			strcpy(buf, inet_ntoa(icp->icmp_ip.ip_dst));

			printf("redirect dst %s to %s\n",
			    buf, inet_ntoa(icp->icmp_gwaddr));
		}
#endif
		icmpsrc.sin_addr = icp->icmp_ip.ip_dst;
		rtredirect(sintosa(&icmpsrc), sintosa(&icmpdst),
		    (struct sockaddr *)0, RTF_GATEWAY | RTF_HOST,
		    sintosa(&icmpgw), (struct rtentry **)0);
		pfctlinput(PRC_REDIRECT_HOST, sintosa(&icmpsrc));
		break;

	/*
	 * No kernel processing for the following;
	 * just fall through to send to raw listener.
	 */
	case ICMP_ECHOREPLY:
	case ICMP_ROUTERADVERT:
	case ICMP_ROUTERSOLICIT:
	case ICMP_TSTAMPREPLY:
	case ICMP_IREQREPLY:
	case ICMP_MASKREPLY:
	default:
		break;
	}

raw:
	rip_input(m);
	return;

freeit:
	m_freem(m);
}

/*
 * Reflect the ip packet back to the source
 */
void
icmp_reflect(m)
	struct mbuf *m;
{
	register struct ip *ip = mtod(m, struct ip *);
	register struct in_ifaddr *ia;
	struct in_addr t;
	struct mbuf *opts = 0;
	int optlen = (ip->ip_hl << 2) - sizeof(struct ip);

	if (!in_canforward(ip->ip_src) &&
	    ((ip->ip_src.s_addr & IN_CLASSA_NET) !=
	     htonl(IN_LOOPBACKNET << IN_CLASSA_NSHIFT))) {
		m_freem(m);	/* Bad return address */
		goto done;	/* ip_output() will check for broadcast */
	}
	t = ip->ip_dst;
	ip->ip_dst = ip->ip_src;
	/*
	 * If the incoming packet was addressed directly to us,
	 * use dst as the src for the reply.  Otherwise (broadcast
	 * or anonymous), use the address which corresponds
	 * to the incoming interface.
	 */
	for (ia = in_ifaddr.tqh_first; ia; ia = ia->ia_list.tqe_next) {
		if (t.s_addr == ia->ia_addr.sin_addr.s_addr)
			break;
		if ((ia->ia_ifp->if_flags & IFF_BROADCAST) &&
		    t.s_addr == ia->ia_broadaddr.sin_addr.s_addr)
			break;
	}
	icmpdst.sin_addr = t;
	if (ia == (struct in_ifaddr *)0)
		ia = ifatoia(ifaof_ifpforaddr(sintosa(&icmpdst),
					      m->m_pkthdr.rcvif));
	/*
	 * The following happens if the packet was not addressed to us,
	 * and was received on an interface with no IP address.
	 */
	if (ia == (struct in_ifaddr *)0) {
	        struct sockaddr_in *dst;
		struct route ro;

		bzero((caddr_t) &ro, sizeof(ro));
		dst = satosin(&ro.ro_dst);
		dst->sin_family = AF_INET;
		dst->sin_len = sizeof(*dst);
		dst->sin_addr = t;

		rtalloc(&ro);
		if (ro.ro_rt == 0) 
		{
		    ipstat.ips_noroute++;
		    goto done;
		}

		ia = ifatoia(ro.ro_rt->rt_ifa);
		ro.ro_rt->rt_use++;
                RTFREE(ro.ro_rt);
        }

	t = ia->ia_addr.sin_addr;
	ip->ip_src = t;
	ip->ip_ttl = MAXTTL;

	if (optlen > 0) {
		register u_char *cp;
		int opt, cnt;
		u_int len;

		/*
		 * Retrieve any source routing from the incoming packet;
		 * add on any record-route or timestamp options.
		 */
		cp = (u_char *) (ip + 1);
		if ((opts = ip_srcroute()) == 0 &&
		    (opts = m_gethdr(M_DONTWAIT, MT_HEADER))) {
			opts->m_len = sizeof(struct in_addr);
			mtod(opts, struct in_addr *)->s_addr = 0;
		}
		if (opts) {
#ifdef ICMPPRINTFS
		    if (icmpprintfs)
			    printf("icmp_reflect optlen %d rt %d => ",
				optlen, opts->m_len);
#endif
		    for (cnt = optlen; cnt > 0; cnt -= len, cp += len) {
			    opt = cp[IPOPT_OPTVAL];
			    if (opt == IPOPT_EOL)
				    break;
			    if (opt == IPOPT_NOP)
				    len = 1;
			    else {
				    if (cnt < IPOPT_OLEN + sizeof(*cp))
					    break;
				    len = cp[IPOPT_OLEN];
				    if (len < IPOPT_OLEN + sizeof(*cp) ||
				        len > cnt)
					    break;
			    }
			    /*
			     * Should check for overflow, but it "can't happen"
			     */
			    if (opt == IPOPT_RR || opt == IPOPT_TS || 
				opt == IPOPT_SECURITY) {
				    bcopy((caddr_t)cp,
					mtod(opts, caddr_t) + opts->m_len, len);
				    opts->m_len += len;
			    }
		    }
		    /* Terminate & pad, if necessary */
		    if ((cnt = opts->m_len % 4) != 0) {
			    for (; cnt < 4; cnt++) {
				    *(mtod(opts, caddr_t) + opts->m_len) =
					IPOPT_EOL;
				    opts->m_len++;
			    }
		    }
#ifdef ICMPPRINTFS
		    if (icmpprintfs)
			    printf("%d\n", opts->m_len);
#endif
		}
		/*
		 * Now strip out original options by copying rest of first
		 * mbuf's data back, and adjust the IP length.
		 */
		ip->ip_len -= optlen;
		ip->ip_hl = sizeof(struct ip) >> 2;
		m->m_len -= optlen;
		if (m->m_flags & M_PKTHDR)
			m->m_pkthdr.len -= optlen;
		optlen += sizeof(struct ip);
		bcopy((caddr_t)ip + optlen, (caddr_t)(ip + 1),
			 (unsigned)(m->m_len - sizeof(struct ip)));
	}
	m->m_flags &= ~(M_BCAST|M_MCAST);
	icmp_send(m, opts);
done:
	if (opts)
		(void)m_free(opts);
}

/*
 * Send an icmp packet back to the ip level,
 * after supplying a checksum.
 */
void
icmp_send(m, opts)
	register struct mbuf *m;
	struct mbuf *opts;
{
	register struct ip *ip = mtod(m, struct ip *);
	register int hlen;
	register struct icmp *icp;

	hlen = ip->ip_hl << 2;
	m->m_data += hlen;
	m->m_len -= hlen;
	icp = mtod(m, struct icmp *);
	icp->icmp_cksum = 0;
	icp->icmp_cksum = in_cksum(m, ip->ip_len - hlen);
	m->m_data -= hlen;
	m->m_len += hlen;
#ifdef ICMPPRINTFS
	if (icmpprintfs) {
		char buf[4 * sizeof "123"];

		strcpy(buf, inet_ntoa(ip->ip_dst));
		printf("icmp_send dst %s src %s\n",
		    buf, inet_ntoa(ip->ip_src));
	}
#endif
	(void) ip_output(m, opts, NULL, 0, NULL, NULL);
}

n_time
iptime()
{
	struct timeval atv;
	u_long t;

	microtime(&atv);
	t = (atv.tv_sec % (24*60*60)) * 1000 + atv.tv_usec / 1000;
	return (htonl(t));
}

int
icmp_sysctl(name, namelen, oldp, oldlenp, newp, newlen)
	int *name;
	u_int namelen;
	void *oldp;
	size_t *oldlenp;
	void *newp;
	size_t newlen;
{

	/* All sysctl names at this level are terminal. */
	if (namelen != 1)
		return (ENOTDIR);

	switch (name[0]) {
	case ICMPCTL_MASKREPL:
		return (sysctl_int(oldp, oldlenp, newp, newlen, &icmpmaskrepl));
	case ICMPCTL_BMCASTECHO:
		return (sysctl_int(oldp, oldlenp, newp, newlen, &icmpbmcastecho));
	case ICMPCTL_ERRPPSLIMIT:
		return (sysctl_int(oldp, oldlenp, newp, newlen,
		    &icmperrppslim));
		break;
	default:
		return (ENOPROTOOPT);
	}
	/* NOTREACHED */
}

struct rtentry *
icmp_mtudisc_clone(struct sockaddr *dst)
{
	struct rtentry *rt;
	int error;

	rt = rtalloc1(dst, 1);
	if (rt == 0)
		return (NULL);
    
	/* If we didn't get a host route, allocate one */
    
	if ((rt->rt_flags & RTF_HOST) == 0) {
		struct rtentry *nrt;

		error = rtrequest((int) RTM_ADD, dst, 
		    (struct sockaddr *) rt->rt_gateway,
		    (struct sockaddr *) 0, 
		    RTF_GATEWAY | RTF_HOST | RTF_DYNAMIC, &nrt);
		if (error) {
			rtfree(rt);
			return (NULL);
		}
		nrt->rt_rmx = rt->rt_rmx;
		rtfree(rt);
		rt = nrt;
	}
	error = rt_timer_add(rt, icmp_mtudisc_timeout, ip_mtudisc_timeout_q);
	if (error) {
		rtfree(rt);
		return (NULL);
	}

	return (rt);
}

void
icmp_mtudisc(icp)
	struct icmp *icp;
{
	struct rtentry *rt;
	struct sockaddr *dst = sintosa(&icmpsrc);
	u_long mtu = ntohs(icp->icmp_nextmtu);  /* Why a long?  IPv6 */
	
	/* Table of common MTUs: */

	static u_short mtu_table[] = {65535, 65280, 32000, 17914, 9180, 8166, 
				      4352, 2002, 1492, 1006, 508, 296, 68, 0};

	rt = icmp_mtudisc_clone(dst);
	if (rt == 0)
		return;

	if (mtu == 0) {
		int i = 0;

		mtu = icp->icmp_ip.ip_len; /* NTOHS happened in deliver: */
		/* Some 4.2BSD-based routers incorrectly adjust the ip_len */
		if (mtu > rt->rt_rmx.rmx_mtu && rt->rt_rmx.rmx_mtu != 0)
			mtu -= (icp->icmp_ip.ip_hl << 2);

		/* If we still can't guess a value, try the route */

		if (mtu == 0) {
			mtu = rt->rt_rmx.rmx_mtu;

			/* If no route mtu, default to the interface mtu */

			if (mtu == 0)
				mtu = rt->rt_ifp->if_mtu;
		}

		for (i = 0; i < sizeof(mtu_table) / sizeof(mtu_table[0]); i++)
			if (mtu > mtu_table[i]) {
				mtu = mtu_table[i];
				break;
			}
	}

	/*
	 * XXX:   RTV_MTU is overloaded, since the admin can set it
	 *	  to turn off PMTU for a route, and the kernel can
	 *	  set it to indicate a serious problem with PMTU
	 *	  on a route.  We should be using a separate flag
	 *	  for the kernel to indicate this.
	 */

	if ((rt->rt_rmx.rmx_locks & RTV_MTU) == 0) {
		if (mtu < 296 || mtu > rt->rt_ifp->if_mtu)
			rt->rt_rmx.rmx_locks |= RTV_MTU;
		else if (rt->rt_rmx.rmx_mtu > mtu || 
			 rt->rt_rmx.rmx_mtu == 0)
			rt->rt_rmx.rmx_mtu = mtu;
	}

	rtfree(rt);
}

void
icmp_mtudisc_timeout(rt, r)
	struct rtentry *rt;
	struct rttimer *r;
{
	if (rt == NULL)
		panic("icmp_mtudisc_timeout:  bad route to timeout");
	if ((rt->rt_flags & (RTF_DYNAMIC | RTF_HOST)) == 
	    (RTF_DYNAMIC | RTF_HOST)) {
		void *(*ctlfunc) __P((int, struct sockaddr *, void *));
		extern u_char ip_protox[];
		struct sockaddr_in sa;

		sa = *(struct sockaddr_in *)rt_key(rt);
		rtrequest((int) RTM_DELETE, (struct sockaddr *)rt_key(rt),
		    rt->rt_gateway, rt_mask(rt), rt->rt_flags, 0);

		/* Notify TCP layer of increased Path MTU estimate */
		ctlfunc = inetsw[ip_protox[IPPROTO_TCP]].pr_ctlinput;
		if (ctlfunc)
			(*ctlfunc)(PRC_MTUINC,(struct sockaddr *)&sa, NULL);
	} else {
		if ((rt->rt_rmx.rmx_locks & RTV_MTU) == 0) {
			rt->rt_rmx.rmx_mtu = 0;
		}
	}
}

/*
 * Perform rate limit check.
 * Returns 0 if it is okay to send the icmp packet.
 * Returns 1 if the router SHOULD NOT send this icmp packet due to rate
 * limitation.
 *
 * XXX per-destination/type check necessary?
 */
int
icmp_ratelimit(dst, type, code)
	const struct in_addr *dst;
	const int type;			/* not used at this moment */
	const int code;			/* not used at this moment */
{

	/* PPS limit */
	if (!ppsratecheck(&icmperrppslim_last, &icmperrpps_count,
	    icmperrppslim)) {
		/* The packet is subject to rate limit */
		return 1;
	}

	/*okay to send*/
	return 0;
}