ip_input.c

RTEMS (Real-Time Executive for Multiprocessor Systems) is a free open source real-time operating sys

		case IPOPT_RR:
			if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) {
				code = &cp[IPOPT_OFFSET] - (u_char *)ip;
				goto bad;
			}
			/*
			 * If no space remains, ignore.
			 */
			off--;			/* 0 origin */
			if (off > optlen - sizeof(struct in_addr))
				break;
			(void)memcpy(&ipaddr.sin_addr, &ip->ip_dst,
			    sizeof(ipaddr.sin_addr));
			/*
			 * locate outgoing interface; if we're the destination,
			 * use the incoming interface (should be same).
			 */
			if ((ia = (INA)ifa_ifwithaddr((SA)&ipaddr)) == 0 &&
			    (ia = ip_rtaddr(ipaddr.sin_addr)) == 0) {
				type = ICMP_UNREACH;
				code = ICMP_UNREACH_HOST;
				goto bad;
			}
			(void)memcpy(cp + off, &(IA_SIN(ia)->sin_addr),
			    sizeof(struct in_addr));
			cp[IPOPT_OFFSET] += sizeof(struct in_addr);
			break;

		case IPOPT_TS:
			code = cp - (u_char *)ip;
			ipt = (struct ip_timestamp *)cp;
			if (ipt->ipt_len < 5)
				goto bad;
			if (ipt->ipt_ptr > ipt->ipt_len - sizeof (long)) {
				if (++ipt->ipt_oflw == 0)
					goto bad;
				break;
			}
			sin = (struct in_addr *)(cp + ipt->ipt_ptr - 1);
			switch (ipt->ipt_flg) {

			case IPOPT_TS_TSONLY:
				break;

			case IPOPT_TS_TSANDADDR:
				if (ipt->ipt_ptr + sizeof(n_time) +
				    sizeof(struct in_addr) > ipt->ipt_len)
					goto bad;
				ipaddr.sin_addr = dst;
				ia = (INA)ifaof_ifpforaddr((SA)&ipaddr,
							    m->m_pkthdr.rcvif);
				if (ia == 0)
					continue;
				(void)memcpy(sin, &IA_SIN(ia)->sin_addr,
				    sizeof(struct in_addr));
				ipt->ipt_ptr += sizeof(struct in_addr);
				break;

			case IPOPT_TS_PRESPEC:
				if (ipt->ipt_ptr + sizeof(n_time) +
				    sizeof(struct in_addr) > ipt->ipt_len)
					goto bad;
				(void)memcpy(&ipaddr.sin_addr, sin,
				    sizeof(struct in_addr));
				if (ifa_ifwithaddr((SA)&ipaddr) == 0)
					continue;
				ipt->ipt_ptr += sizeof(struct in_addr);
				break;

			default:
				goto bad;
			}
			ntime = iptime();
			(void)memcpy(cp + ipt->ipt_ptr - 1, &ntime,
			    sizeof(n_time));
			ipt->ipt_ptr += sizeof(n_time);
		}
	}
	if (forward && ipforwarding) {
		ip_forward(m, 1);
		return (1);
	}
	return (0);
bad:
	ip->ip_len -= IP_VHL_HL(ip->ip_vhl) << 2;   /* XXX icmp_error adds in hdr length */
	icmp_error(m, type, code, 0, 0);
	ipstat.ips_badoptions++;
	return (1);
}

/*
 * Given address of next destination (final or next hop),
 * return internet address info of interface to be used to get there.
 */
static struct in_ifaddr *
ip_rtaddr(dst)
	 struct in_addr dst;
{
	register struct sockaddr_in *sin;

	sin = (struct sockaddr_in *) &ipforward_rt.ro_dst;

	if (ipforward_rt.ro_rt == 0 || dst.s_addr != sin->sin_addr.s_addr) {
		if (ipforward_rt.ro_rt) {
			RTFREE(ipforward_rt.ro_rt);
			ipforward_rt.ro_rt = 0;
		}
		sin->sin_family = AF_INET;
		sin->sin_len = sizeof(*sin);
		sin->sin_addr = dst;

		rtalloc_ign(&ipforward_rt, RTF_PRCLONING);
	}
	if (ipforward_rt.ro_rt == 0)
		return ((struct in_ifaddr *)0);
	return ((struct in_ifaddr *) ipforward_rt.ro_rt->rt_ifa);
}

/*
 * Save incoming source route for use in replies,
 * to be picked up later by ip_srcroute if the receiver is interested.
 */
void
save_rte(option, dst)
	u_char *option;
	struct in_addr dst;
{
	unsigned olen;

	olen = option[IPOPT_OLEN];
#ifdef DIAGNOSTIC
	if (ipprintfs)
		printf("save_rte: olen %d\n", olen);
#endif
	if (olen > sizeof(ip_srcrt) - (1 + sizeof(dst)))
		return;
	bcopy(option, ip_srcrt.srcopt, olen);
	ip_nhops = (olen - IPOPT_OFFSET - 1) / sizeof(struct in_addr);
	ip_srcrt.dst = dst;
}

/*
 * Retrieve incoming source route for use in replies,
 * in the same form used by setsockopt.
 * The first hop is placed before the options, will be removed later.
 */
struct mbuf *
ip_srcroute()
{
	register struct in_addr *p, *q;
	register struct mbuf *m;

	if (ip_nhops == 0)
		return ((struct mbuf *)0);
	m = m_get(M_DONTWAIT, MT_SOOPTS);
	if (m == 0)
		return ((struct mbuf *)0);

#define OPTSIZ	(sizeof(ip_srcrt.nop) + sizeof(ip_srcrt.srcopt))

	/* length is (nhops+1)*sizeof(addr) + sizeof(nop + srcrt header) */
	m->m_len = ip_nhops * sizeof(struct in_addr) + sizeof(struct in_addr) +
	    OPTSIZ;
#ifdef DIAGNOSTIC
	if (ipprintfs)
		printf("ip_srcroute: nhops %d mlen %d", ip_nhops, m->m_len);
#endif

	/*
	 * First save first hop for return route
	 */
	p = &ip_srcrt.route[ip_nhops - 1];
	*(mtod(m, struct in_addr *)) = *p--;
#ifdef DIAGNOSTIC
	if (ipprintfs)
		printf(" hops %lx", ntohl(mtod(m, struct in_addr *)->s_addr));
#endif

	/*
	 * Copy option fields and padding (nop) to mbuf.
	 */
	ip_srcrt.nop = IPOPT_NOP;
	ip_srcrt.srcopt[IPOPT_OFFSET] = IPOPT_MINOFF;
	(void)memcpy(mtod(m, caddr_t) + sizeof(struct in_addr),
	    &ip_srcrt.nop, OPTSIZ);
	q = (struct in_addr *)(mtod(m, caddr_t) +
	    sizeof(struct in_addr) + OPTSIZ);
#undef OPTSIZ
	/*
	 * Record return path as an IP source route,
	 * reversing the path (pointers are now aligned).
	 */
	while (p >= ip_srcrt.route) {
#ifdef DIAGNOSTIC
		if (ipprintfs)
			printf(" %lx", ntohl(q->s_addr));
#endif
		*q++ = *p--;
	}
	/*
	 * Last hop goes to final destination.
	 */
	*q = ip_srcrt.dst;
#ifdef DIAGNOSTIC
	if (ipprintfs)
		printf(" %lx\n", ntohl(q->s_addr));
#endif
	return (m);
}

/*
 * Strip out IP options, at higher
 * level protocol in the kernel.
 * Second argument is buffer to which options
 * will be moved, and return value is their length.
 * XXX should be deleted; last arg currently ignored.
 */
void
ip_stripoptions(m, mopt)
	register struct mbuf *m;
	struct mbuf *mopt;
{
	register int i;
	struct ip *ip = mtod(m, struct ip *);
	register caddr_t opts;
	int olen;

	olen = (IP_VHL_HL(ip->ip_vhl) << 2) - sizeof (struct ip);
	opts = (caddr_t)(ip + 1);
	i = m->m_len - (sizeof (struct ip) + olen);
	bcopy(opts + olen, opts, (unsigned)i);
	m->m_len -= olen;
	if (m->m_flags & M_PKTHDR)
		m->m_pkthdr.len -= olen;
	ip->ip_vhl = IP_MAKE_VHL(IPVERSION, sizeof(struct ip) >> 2);
}

u_char inetctlerrmap[PRC_NCMDS] = {
	0,		0,		0,		0,
	0,		EMSGSIZE,	EHOSTDOWN,	EHOSTUNREACH,
	EHOSTUNREACH,	EHOSTUNREACH,	ECONNREFUSED,	ECONNREFUSED,
	EMSGSIZE,	EHOSTUNREACH,	0,		0,
	0,		0,		0,		0,
	ENOPROTOOPT
};

/*
 * Forward a packet.  If some error occurs return the sender
 * an icmp packet.  Note we can't always generate a meaningful
 * icmp message because icmp doesn't have a large enough repertoire
 * of codes and types.
 *
 * If not forwarding, just drop the packet.  This could be confusing
 * if ipforwarding was zero but some routing protocol was advancing
 * us as a gateway to somewhere.  However, we must let the routing
 * protocol deal with that.
 *
 * The srcrt parameter indicates whether the packet is being forwarded
 * via a source route.
 */
static void
ip_forward(m, srcrt)
	struct mbuf *m;
	int srcrt;
{
	register struct ip *ip = mtod(m, struct ip *);
	register struct sockaddr_in *sin;
	register struct rtentry *rt;
	int error, type = 0, code = 0;
	struct mbuf *mcopy;
	n_long dest;
	struct ifnet *destifp;

	dest = 0;
#ifdef DIAGNOSTIC
	if (ipprintfs)
		printf("forward: src %lx dst %lx ttl %x\n",
			ip->ip_src.s_addr, ip->ip_dst.s_addr, ip->ip_ttl);
#endif


	if (m->m_flags & M_BCAST || in_canforward(ip->ip_dst) == 0) {
		ipstat.ips_cantforward++;
		m_freem(m);
		return;
	}
	HTONS(ip->ip_id);
	if (ip->ip_ttl <= IPTTLDEC) {
		icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS, dest, 0);
		return;
	}
	ip->ip_ttl -= IPTTLDEC;

	sin = (struct sockaddr_in *)&ipforward_rt.ro_dst;
	if ((rt = ipforward_rt.ro_rt) == 0 ||
	    ip->ip_dst.s_addr != sin->sin_addr.s_addr) {
		if (ipforward_rt.ro_rt) {
			RTFREE(ipforward_rt.ro_rt);
			ipforward_rt.ro_rt = 0;
		}
		sin->sin_family = AF_INET;
		sin->sin_len = sizeof(*sin);
		sin->sin_addr = ip->ip_dst;

		rtalloc_ign(&ipforward_rt, RTF_PRCLONING);
		if (ipforward_rt.ro_rt == 0) {
			icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, dest, 0);
			return;
		}
		rt = ipforward_rt.ro_rt;
	}

	/*
	 * Save at most 64 bytes of the packet in case
	 * we need to generate an ICMP message to the src.
	 */
	mcopy = m_copy(m, 0, imin((int)ip->ip_len, 64));

	/*
	 * If forwarding packet using same interface that it came in on,
	 * perhaps should send a redirect to sender to shortcut a hop.
	 * Only send redirect if source is sending directly to us,
	 * and if packet was not source routed (or has any options).
	 * Also, don't send redirect if forwarding using a default route
	 * or a route modified by a redirect.
	 */
#define	satosin(sa)	((struct sockaddr_in *)(sa))
	if (rt->rt_ifp == m->m_pkthdr.rcvif &&
	    (rt->rt_flags & (RTF_DYNAMIC|RTF_MODIFIED)) == 0 &&
	    satosin(rt_key(rt))->sin_addr.s_addr != 0 &&
	    ipsendredirects && !srcrt) {
#define	RTA(rt)	((struct in_ifaddr *)(rt->rt_ifa))
		u_long src = ntohl(ip->ip_src.s_addr);

		if (RTA(rt) &&
		    (src & RTA(rt)->ia_subnetmask) == RTA(rt)->ia_subnet) {
		    if (rt->rt_flags & RTF_GATEWAY)
			dest = satosin(rt->rt_gateway)->sin_addr.s_addr;
		    else
			dest = ip->ip_dst.s_addr;
		    /* Router requirements says to only send host redirects */
		    type = ICMP_REDIRECT;
		    code = ICMP_REDIRECT_HOST;
#ifdef DIAGNOSTIC
		    if (ipprintfs)
		        printf("redirect (%d) to %lx\n", code, (u_long)dest);
#endif
		}
	}

	error = ip_output(m, (struct mbuf *)0, &ipforward_rt, 
			  IP_FORWARDING, 0);
	if (error)
		ipstat.ips_cantforward++;
	else {
		ipstat.ips_forward++;
		if (type)
			ipstat.ips_redirectsent++;
		else {
			if (mcopy)
				m_freem(mcopy);
			return;
		}
	}
	if (mcopy == NULL)
		return;
	destifp = NULL;

	switch (error) {

	case 0:				/* forwarded, but need redirect */
		/* type, code set above */
		break;

	case ENETUNREACH:		/* shouldn't happen, checked above */
	case EHOSTUNREACH:
	case ENETDOWN:
	case EHOSTDOWN:
	default:
		type = ICMP_UNREACH;
		code = ICMP_UNREACH_HOST;
		break;

	case EMSGSIZE:
		type = ICMP_UNREACH;
		code = ICMP_UNREACH_NEEDFRAG;
		if (ipforward_rt.ro_rt)
			destifp = ipforward_rt.ro_rt->rt_ifp;
		ipstat.ips_cantfrag++;
		break;

	case ENOBUFS:
		type = ICMP_SOURCEQUENCH;
		code = 0;
		break;
	}
	icmp_error(mcopy, type, code, dest, destifp);
}

void
ip_savecontrol(inp, mp, ip, m)
	register struct inpcb *inp;
	register struct mbuf **mp;
	register struct ip *ip;
	register struct mbuf *m;
{
	if (inp->inp_socket->so_options & SO_TIMESTAMP) {
		struct timeval tv;

		microtime(&tv);
		*mp = sbcreatecontrol((caddr_t) &tv, sizeof(tv),
			SCM_TIMESTAMP, SOL_SOCKET);
		if (*mp)
			mp = &(*mp)->m_next;
	}
	if (inp->inp_flags & INP_RECVDSTADDR) {
		*mp = sbcreatecontrol((caddr_t) &ip->ip_dst,
		    sizeof(struct in_addr), IP_RECVDSTADDR, IPPROTO_IP);
		if (*mp)
			mp = &(*mp)->m_next;
	}
#ifdef notyet
	/* XXX
	 * Moving these out of udp_input() made them even more broken
	 * than they already were.
	 */
	/* options were tossed already */
	if (inp->inp_flags & INP_RECVOPTS) {
		*mp = sbcreatecontrol((caddr_t) opts_deleted_above,
		    sizeof(struct in_addr), IP_RECVOPTS, IPPROTO_IP);
		if (*mp)
			mp = &(*mp)->m_next;
	}
	/* ip_srcroute doesn't do what we want here, need to fix */
	if (inp->inp_flags & INP_RECVRETOPTS) {
		*mp = sbcreatecontrol((caddr_t) ip_srcroute(),
		    sizeof(struct in_addr), IP_RECVRETOPTS, IPPROTO_IP);
		if (*mp)
			mp = &(*mp)->m_next;
	}
#endif
	if (inp->inp_flags & INP_RECVIF) {
		struct sockaddr_dl sdl;

		sdl.sdl_len = offsetof(struct sockaddr_dl, sdl_data[0]);
		sdl.sdl_family = AF_LINK;
		sdl.sdl_index = m->m_pkthdr.rcvif ?
			m->m_pkthdr.rcvif->if_index : 0;
		sdl.sdl_nlen = sdl.sdl_alen = sdl.sdl_slen = 0;
		*mp = sbcreatecontrol((caddr_t) &sdl, sdl.sdl_len,
			IP_RECVIF, IPPROTO_IP);
		if (*mp)
			mp = &(*mp)->m_next;
	}
}

int
ip_rsvp_init(struct socket *so)
{
	if (so->so_type != SOCK_RAW ||
	    so->so_proto->pr_protocol != IPPROTO_RSVP)
	  return EOPNOTSUPP;

	if (ip_rsvpd != NULL)
	  return EADDRINUSE;

	ip_rsvpd = so;
	/*
	 * This may seem silly, but we need to be sure we don't over-increment
	 * the RSVP counter, in case something slips up.
	 */
	if (!ip_rsvp_on) {
		ip_rsvp_on = 1;
		rsvp_on++;
	}

	return 0;
}

int
ip_rsvp_done(void)
{
	ip_rsvpd = NULL;
	/*
	 * This may seem silly, but we need to be sure we don't over-decrement
	 * the RSVP counter, in case something slips up.
	 */
	if (ip_rsvp_on) {
		ip_rsvp_on = 0;
		rsvp_on--;
	}
	return 0;
}