4.4BSD snapshot (revision 8.1); add 1993 to copyright
[unix-history] / usr / src / sys / netinet / ip_icmp.c
/*
* Copyright (c) 1982, 1986, 1988, 1993
* The Regents of the University of California. All rights reserved.
*
* %sccs.include.redist.c%
*
* @(#)ip_icmp.c 8.1 (Berkeley) %G%
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/time.h>
#include <sys/kernel.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/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;
#ifdef ICMPPRINTFS
int icmpprintfs = 0;
#endif
extern struct protosw inetsw[];
/*
* Generate an error packet of type error
* in response to bad packet ip.
*/
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;
register struct mbuf *m;
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 &&
dtom(oip)->m_len >= oiplen + ICMP_MINLEN &&
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, 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_len = htons((u_short)(nip->ip_len + oiplen));
/*
* 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_len = m->m_len;
nip->ip_hl = sizeof(struct ip) >> 2;
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 sockproto icmproto = { AF_INET, IPPROTO_ICMP };
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
icmp_input(m, hlen)
register struct mbuf *m;
int hlen;
{
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 *, struct ip *));
int code;
extern u_char ip_protox[];
/*
* Locate icmp structure in mbuf, and check
* that not corrupted and of at least minimum length.
*/
#ifdef ICMPPRINTFS
if (icmpprintfs)
printf("icmp_input from %x to %x, len %d\n",
ntohl(ip->ip_src.s_addr), ntohl(ip->ip_dst.s_addr),
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)) == 0) {
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:
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:
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:
/*
* 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;
}
NTOHS(icp->icmp_ip.ip_len);
#ifdef ICMPPRINTFS
if (icmpprintfs)
printf("deliver to protocol %d\n", icp->icmp_ip.ip_p);
#endif
icmpsrc.sin_addr = icp->icmp_ip.ip_dst;
if (ctlfunc = inetsw[ip_protox[icp->icmp_ip.ip_p]].pr_ctlinput)
(*ctlfunc)(code, (struct sockaddr *)&icmpsrc,
&icp->icmp_ip);
break;
badcode:
icmpstat.icps_badcode++;
break;
case ICMP_ECHO:
icp->icmp_type = ICMP_ECHOREPLY;
goto reflect;
case ICMP_TSTAMP:
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:
#define satosin(sa) ((struct sockaddr_in *)(sa))
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)
break;
switch (ip->ip_dst.s_addr) {
case INADDR_BROADCAST:
case INADDR_ANY:
icmpdst.sin_addr = ip->ip_src;
break;
default:
icmpdst.sin_addr = ip->ip_dst;
}
ia = (struct in_ifaddr *)ifaof_ifpforaddr(
(struct sockaddr *)&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 = satosin(&ia->ia_broadaddr)->sin_addr;
else if (ia->ia_ifp->if_flags & IFF_POINTOPOINT)
ip->ip_src = satosin(&ia->ia_dstaddr)->sin_addr;
}
reflect:
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:
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)
printf("redirect dst %x to %x\n", icp->icmp_ip.ip_dst,
icp->icmp_gwaddr);
#endif
icmpsrc.sin_addr = icp->icmp_ip.ip_dst;
rtredirect((struct sockaddr *)&icmpsrc,
(struct sockaddr *)&icmpdst,
(struct sockaddr *)0, RTF_GATEWAY | RTF_HOST,
(struct sockaddr *)&icmpgw, (struct rtentry **)0);
pfctlinput(PRC_REDIRECT_HOST, (struct sockaddr *)&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, *ip_srcroute();
int optlen = (ip->ip_hl << 2) - sizeof(struct ip);
if (!in_canforward(ip->ip_src) &&
((ntohl(ip->ip_src.s_addr) & IN_CLASSA_NET) !=
(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; ia; ia = ia->ia_next) {
if (t.s_addr == IA_SIN(ia)->sin_addr.s_addr)
break;
if ((ia->ia_ifp->if_flags & IFF_BROADCAST) &&
t.s_addr == satosin(&ia->ia_broadaddr)->sin_addr.s_addr)
break;
}
icmpdst.sin_addr = t;
if (ia == (struct in_ifaddr *)0)
ia = (struct in_ifaddr *)ifaof_ifpforaddr(
(struct sockaddr *)&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)
ia = in_ifaddr;
t = IA_SIN(ia)->sin_addr;
ip->ip_src = t;
ip->ip_ttl = MAXTTL;
if (optlen > 0) {
register u_char *cp;
struct mbuf *m = dtom(ip);
int opt, cnt, off;
u_int len;
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 {
len = cp[IPOPT_OLEN];
if (len <= 0 || 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) {
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)
printf("icmp_send dst %x src %x\n", ip->ip_dst, ip->ip_src);
#endif
(void) ip_output(m, opts, NULL, 0, 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;
{
extern int ip_ttl;
/* 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));
default:
return (ENOPROTOOPT);
}
/* NOTREACHED */
}