Index: head/sys/netinet/ip_input.c =================================================================== --- head/sys/netinet/ip_input.c (revision 88592) +++ head/sys/netinet/ip_input.c (revision 88593) @@ -1,1897 +1,1905 @@ /* * 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. * * @(#)ip_input.c 8.2 (Berkeley) 1/4/94 * $FreeBSD$ */ #define _IP_VHL #include "opt_bootp.h" #include "opt_ipfw.h" #include "opt_ipdn.h" #include "opt_ipdivert.h" #include "opt_ipfilter.h" #include "opt_ipstealth.h" #include "opt_ipsec.h" #include "opt_pfil_hooks.h" #include "opt_random_ip_id.h" #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #ifdef IPSEC #include #include #endif int rsvp_on = 0; static int ip_rsvp_on; struct socket *ip_rsvpd; int ipforwarding = 0; SYSCTL_INT(_net_inet_ip, IPCTL_FORWARDING, forwarding, CTLFLAG_RW, &ipforwarding, 0, "Enable IP forwarding between interfaces"); static int ipsendredirects = 1; /* XXX */ SYSCTL_INT(_net_inet_ip, IPCTL_SENDREDIRECTS, redirect, CTLFLAG_RW, &ipsendredirects, 0, "Enable sending IP redirects"); int ip_defttl = IPDEFTTL; SYSCTL_INT(_net_inet_ip, IPCTL_DEFTTL, ttl, CTLFLAG_RW, &ip_defttl, 0, "Maximum TTL on IP packets"); static int ip_dosourceroute = 0; SYSCTL_INT(_net_inet_ip, IPCTL_SOURCEROUTE, sourceroute, CTLFLAG_RW, &ip_dosourceroute, 0, "Enable forwarding source routed IP packets"); static int ip_acceptsourceroute = 0; SYSCTL_INT(_net_inet_ip, IPCTL_ACCEPTSOURCEROUTE, accept_sourceroute, CTLFLAG_RW, &ip_acceptsourceroute, 0, "Enable accepting source routed IP packets"); static int ip_keepfaith = 0; SYSCTL_INT(_net_inet_ip, IPCTL_KEEPFAITH, keepfaith, CTLFLAG_RW, &ip_keepfaith, 0, "Enable packet capture for FAITH IPv4->IPv6 translater daemon"); static int ip_nfragpackets = 0; static int ip_maxfragpackets; /* initialized in ip_init() */ SYSCTL_INT(_net_inet_ip, OID_AUTO, maxfragpackets, CTLFLAG_RW, &ip_maxfragpackets, 0, "Maximum number of IPv4 fragment reassembly queue entries"); /* * XXX - Setting ip_checkinterface mostly implements the receive side of * the Strong ES model described in RFC 1122, but since the routing table * and transmit implementation do not implement the Strong ES model, * setting this to 1 results in an odd hybrid. * * XXX - ip_checkinterface currently must be disabled if you use ipnat * to translate the destination address to another local interface. * * XXX - ip_checkinterface must be disabled if you add IP aliases * to the loopback interface instead of the interface where the * packets for those addresses are received. */ static int ip_checkinterface = 1; SYSCTL_INT(_net_inet_ip, OID_AUTO, check_interface, CTLFLAG_RW, &ip_checkinterface, 0, "Verify packet arrives on correct interface"); #ifdef DIAGNOSTIC static int ipprintfs = 0; #endif static int ipqmaxlen = IFQ_MAXLEN; extern struct domain inetdomain; extern struct protosw inetsw[]; u_char ip_protox[IPPROTO_MAX]; struct in_ifaddrhead in_ifaddrhead; /* first inet address */ struct in_ifaddrhashhead *in_ifaddrhashtbl; /* inet addr hash table */ u_long in_ifaddrhmask; /* mask for hash table */ SYSCTL_INT(_net_inet_ip, IPCTL_INTRQMAXLEN, intr_queue_maxlen, CTLFLAG_RW, &ipintrq.ifq_maxlen, 0, "Maximum size of the IP input queue"); SYSCTL_INT(_net_inet_ip, IPCTL_INTRQDROPS, intr_queue_drops, CTLFLAG_RD, &ipintrq.ifq_drops, 0, "Number of packets dropped from the IP input queue"); struct ipstat ipstat; SYSCTL_STRUCT(_net_inet_ip, IPCTL_STATS, stats, CTLFLAG_RW, &ipstat, ipstat, "IP statistics (struct ipstat, netinet/ip_var.h)"); /* Packet reassembly stuff */ #define IPREASS_NHASH_LOG2 6 #define IPREASS_NHASH (1 << IPREASS_NHASH_LOG2) #define IPREASS_HMASK (IPREASS_NHASH - 1) #define IPREASS_HASH(x,y) \ (((((x) & 0xF) | ((((x) >> 8) & 0xF) << 4)) ^ (y)) & IPREASS_HMASK) static TAILQ_HEAD(ipqhead, ipq) ipq[IPREASS_NHASH]; static int nipq = 0; /* total # of reass queues */ static int maxnipq; const int ipintrq_present = 1; #ifdef IPCTL_DEFMTU SYSCTL_INT(_net_inet_ip, IPCTL_DEFMTU, mtu, CTLFLAG_RW, &ip_mtu, 0, "Default MTU"); #endif #ifdef IPSTEALTH static int ipstealth = 0; SYSCTL_INT(_net_inet_ip, OID_AUTO, stealth, CTLFLAG_RW, &ipstealth, 0, ""); #endif /* Firewall hooks */ ip_fw_chk_t *ip_fw_chk_ptr; int fw_enable = 1 ; /* Dummynet hooks */ ip_dn_io_t *ip_dn_io_ptr; /* * We need to save the IP options in case a protocol wants to respond * to an incoming packet over the same route if the packet got here * using IP source routing. This allows connection establishment and * maintenance when the remote end is on a network that is not known * to us. */ static int ip_nhops = 0; static struct ip_srcrt { struct in_addr dst; /* final destination */ char nop; /* one NOP to align */ char srcopt[IPOPT_OFFSET + 1]; /* OPTVAL, OLEN and OFFSET */ struct in_addr route[MAX_IPOPTLEN/sizeof(struct in_addr)]; } ip_srcrt; struct sockaddr_in *ip_fw_fwd_addr; static void save_rte __P((u_char *, struct in_addr)); static int ip_dooptions __P((struct mbuf *)); static void ip_forward __P((struct mbuf *, int)); static void ip_freef __P((struct ipqhead *, struct ipq *)); #ifdef IPDIVERT static struct mbuf *ip_reass __P((struct mbuf *, struct ipqhead *, struct ipq *, u_int32_t *, u_int16_t *)); #else static struct mbuf *ip_reass __P((struct mbuf *, struct ipqhead *, struct ipq *)); #endif static void ipintr __P((void)); /* * IP initialization: fill in IP protocol switch table. * All protocols not implemented in kernel go to raw IP protocol handler. */ void ip_init() { register struct protosw *pr; register int i; TAILQ_INIT(&in_ifaddrhead); in_ifaddrhashtbl = hashinit(INADDR_NHASH, M_IFADDR, &in_ifaddrhmask); pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); if (pr == 0) panic("ip_init"); for (i = 0; i < IPPROTO_MAX; i++) ip_protox[i] = pr - inetsw; for (pr = inetdomain.dom_protosw; pr < inetdomain.dom_protoswNPROTOSW; pr++) if (pr->pr_domain->dom_family == PF_INET && pr->pr_protocol && pr->pr_protocol != IPPROTO_RAW) ip_protox[pr->pr_protocol] = pr - inetsw; for (i = 0; i < IPREASS_NHASH; i++) TAILQ_INIT(&ipq[i]); maxnipq = nmbclusters / 4; ip_maxfragpackets = nmbclusters / 4; #ifndef RANDOM_IP_ID ip_id = time_second & 0xffff; #endif ipintrq.ifq_maxlen = ipqmaxlen; mtx_init(&ipintrq.ifq_mtx, "ip_inq", MTX_DEF); register_netisr(NETISR_IP, ipintr); } static struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; struct route ipforward_rt; /* * Ip input routine. Checksum and byte swap header. If fragmented * try to reassemble. Process options. Pass to next level. */ void ip_input(struct mbuf *m) { struct ip *ip; struct ipq *fp; struct in_ifaddr *ia = NULL; struct ifaddr *ifa; int i, hlen, checkif; u_short sum; u_int16_t divert_cookie; /* firewall cookie */ struct in_addr pkt_dst; #ifdef IPDIVERT u_int32_t divert_info = 0; /* packet divert/tee info */ #endif struct ip_fw *rule = NULL; #ifdef PFIL_HOOKS struct packet_filter_hook *pfh; struct mbuf *m0; int rv; #endif /* PFIL_HOOKS */ #ifdef IPDIVERT /* Get and reset firewall cookie */ divert_cookie = ip_divert_cookie; ip_divert_cookie = 0; #else divert_cookie = 0; #endif /* * dummynet packet are prepended a vestigial mbuf with * m_type = MT_DUMMYNET and m_data pointing to the matching * rule. */ if (m->m_type == MT_DUMMYNET) { rule = (struct ip_fw *)(m->m_data) ; m = m->m_next ; ip = mtod(m, struct ip *); hlen = IP_VHL_HL(ip->ip_vhl) << 2; goto iphack ; } else rule = NULL ; #ifdef DIAGNOSTIC if (m == NULL || (m->m_flags & M_PKTHDR) == 0) panic("ip_input no HDR"); #endif ipstat.ips_total++; if (m->m_pkthdr.len < sizeof(struct ip)) goto tooshort; if (m->m_len < sizeof (struct ip) && (m = m_pullup(m, sizeof (struct ip))) == 0) { ipstat.ips_toosmall++; return; } ip = mtod(m, struct ip *); if (IP_VHL_V(ip->ip_vhl) != IPVERSION) { ipstat.ips_badvers++; goto bad; } hlen = IP_VHL_HL(ip->ip_vhl) << 2; if (hlen < sizeof(struct ip)) { /* minimum header length */ ipstat.ips_badhlen++; goto bad; } if (hlen > m->m_len) { if ((m = m_pullup(m, hlen)) == 0) { ipstat.ips_badhlen++; return; } ip = mtod(m, struct ip *); } /* 127/8 must not appear on wire - RFC1122 */ if ((ntohl(ip->ip_dst.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET || (ntohl(ip->ip_src.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET) { if ((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0) { ipstat.ips_badaddr++; goto bad; } } if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) { sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID); } else { if (hlen == sizeof(struct ip)) { sum = in_cksum_hdr(ip); } else { sum = in_cksum(m, hlen); } } if (sum) { ipstat.ips_badsum++; goto bad; } /* * Convert fields to host representation. */ NTOHS(ip->ip_len); if (ip->ip_len < hlen) { ipstat.ips_badlen++; goto bad; } NTOHS(ip->ip_off); /* * Check that the amount of data in the buffers * is as at least much as the IP header would have us expect. * Trim mbufs if longer than we expect. * Drop packet if shorter than we expect. */ if (m->m_pkthdr.len < ip->ip_len) { tooshort: ipstat.ips_tooshort++; goto bad; } if (m->m_pkthdr.len > ip->ip_len) { if (m->m_len == m->m_pkthdr.len) { m->m_len = ip->ip_len; m->m_pkthdr.len = ip->ip_len; } else m_adj(m, ip->ip_len - m->m_pkthdr.len); } #ifdef IPSEC if (ipsec_gethist(m, NULL)) goto pass; #endif /* * IpHack's section. * Right now when no processing on packet has done * and it is still fresh out of network we do our black * deals with it. * - Firewall: deny/allow/divert * - Xlate: translate packet's addr/port (NAT). * - Pipe: pass pkt through dummynet. * - Wrap: fake packet's addr/port * - Encapsulate: put it in another IP and send out. */ iphack: #ifdef PFIL_HOOKS /* * Run through list of hooks for input packets. If there are any * filters which require that additional packets in the flow are * not fast-forwarded, they must clear the M_CANFASTFWD flag. * Note that filters must _never_ set this flag, as another filter * in the list may have previously cleared it. */ m0 = m; pfh = pfil_hook_get(PFIL_IN, &inetsw[ip_protox[IPPROTO_IP]].pr_pfh); for (; pfh; pfh = TAILQ_NEXT(pfh, pfil_link)) if (pfh->pfil_func) { rv = pfh->pfil_func(ip, hlen, m->m_pkthdr.rcvif, 0, &m0); if (rv) return; m = m0; if (m == NULL) return; ip = mtod(m, struct ip *); } #endif /* PFIL_HOOKS */ if (fw_enable && IPFW_LOADED) { #ifdef IPFIREWALL_FORWARD /* * If we've been forwarded from the output side, then * skip the firewall a second time */ if (ip_fw_fwd_addr) goto ours; #endif /* IPFIREWALL_FORWARD */ /* * See the comment in ip_output for the return values * produced by the firewall. */ i = ip_fw_chk_ptr(&ip, hlen, NULL, &divert_cookie, &m, &rule, &ip_fw_fwd_addr); if (i & IP_FW_PORT_DENY_FLAG) { /* XXX new interface-denied */ if (m) m_freem(m); return; } if (m == NULL) { /* Packet discarded by firewall */ static int __debug=10; if (__debug > 0) { printf( "firewall returns NULL, please update!\n"); __debug--; } return; } if (i == 0 && ip_fw_fwd_addr == NULL) /* common case */ goto pass; if (DUMMYNET_LOADED && (i & IP_FW_PORT_DYNT_FLAG) != 0) { /* Send packet to the appropriate pipe */ ip_dn_io_ptr(i&0xffff,DN_TO_IP_IN,m,NULL,NULL,0, rule, 0); return; } #ifdef IPDIVERT if (i != 0 && (i & IP_FW_PORT_DYNT_FLAG) == 0) { /* Divert or tee packet */ divert_info = i; goto ours; } #endif #ifdef IPFIREWALL_FORWARD if (i == 0 && ip_fw_fwd_addr != NULL) goto pass; #endif /* * if we get here, the packet must be dropped */ m_freem(m); return; } pass: /* * Process options and, if not destined for us, * ship it on. ip_dooptions returns 1 when an * error was detected (causing an icmp message * to be sent and the original packet to be freed). */ ip_nhops = 0; /* for source routed packets */ if (hlen > sizeof (struct ip) && ip_dooptions(m)) { #ifdef IPFIREWALL_FORWARD ip_fw_fwd_addr = NULL; #endif return; } /* greedy RSVP, snatches any PATH packet of the RSVP protocol and no * matter if it is destined to another node, or whether it is * a multicast one, RSVP wants it! and prevents it from being forwarded * anywhere else. Also checks if the rsvp daemon is running before * grabbing the packet. */ if (rsvp_on && ip->ip_p==IPPROTO_RSVP) goto ours; /* * Check our list of addresses, to see if the packet is for us. * If we don't have any addresses, assume any unicast packet * we receive might be for us (and let the upper layers deal * with it). */ if (TAILQ_EMPTY(&in_ifaddrhead) && (m->m_flags & (M_MCAST|M_BCAST)) == 0) goto ours; /* * Cache the destination address of the packet; this may be * changed by use of 'ipfw fwd'. */ pkt_dst = ip_fw_fwd_addr == NULL ? ip->ip_dst : ip_fw_fwd_addr->sin_addr; /* * Enable a consistency check between the destination address * and the arrival interface for a unicast packet (the RFC 1122 * strong ES model) if IP forwarding is disabled and the packet * is not locally generated and the packet is not subject to * 'ipfw fwd'. * * XXX - Checking also should be disabled if the destination * address is ipnat'ed to a different interface. * * XXX - Checking is incompatible with IP aliases added * to the loopback interface instead of the interface where * the packets are received. */ checkif = ip_checkinterface && (ipforwarding == 0) && m->m_pkthdr.rcvif != NULL && ((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0) && (ip_fw_fwd_addr == NULL); /* * Check for exact addresses in the hash bucket. */ LIST_FOREACH(ia, INADDR_HASH(pkt_dst.s_addr), ia_hash) { /* * If the address matches, verify that the packet * arrived via the correct interface if checking is * enabled. */ if (IA_SIN(ia)->sin_addr.s_addr == pkt_dst.s_addr && (!checkif || ia->ia_ifp == m->m_pkthdr.rcvif)) goto ours; } /* * Check for broadcast addresses. * * Only accept broadcast packets that arrive via the matching * interface. Reception of forwarded directed broadcasts would * be handled via ip_forward() and ether_output() with the loopback * into the stack for SIMPLEX interfaces handled by ether_output(). */ if (m->m_pkthdr.rcvif->if_flags & IFF_BROADCAST) { TAILQ_FOREACH(ifa, &m->m_pkthdr.rcvif->if_addrhead, ifa_link) { if (ifa->ifa_addr->sa_family != AF_INET) continue; ia = ifatoia(ifa); if (satosin(&ia->ia_broadaddr)->sin_addr.s_addr == pkt_dst.s_addr) goto ours; if (ia->ia_netbroadcast.s_addr == pkt_dst.s_addr) goto ours; #ifdef BOOTP_COMPAT if (IA_SIN(ia)->sin_addr.s_addr == INADDR_ANY) goto ours; #endif } } if (IN_MULTICAST(ntohl(ip->ip_dst.s_addr))) { struct in_multi *inm; if (ip_mrouter) { /* * If we are acting as a multicast router, all * incoming multicast packets are passed to the * kernel-level multicast forwarding function. * The packet is returned (relatively) intact; if * ip_mforward() returns a non-zero value, the packet * must be discarded, else it may be accepted below. */ if (ip_mforward(ip, m->m_pkthdr.rcvif, m, 0) != 0) { ipstat.ips_cantforward++; m_freem(m); return; } /* * The process-level routing demon needs to receive * all multicast IGMP packets, whether or not this * host belongs to their destination groups. */ if (ip->ip_p == IPPROTO_IGMP) goto ours; ipstat.ips_forward++; } /* * See if we belong to the destination multicast group on the * arrival interface. */ IN_LOOKUP_MULTI(ip->ip_dst, m->m_pkthdr.rcvif, inm); if (inm == NULL) { ipstat.ips_notmember++; m_freem(m); return; } goto ours; } if (ip->ip_dst.s_addr == (u_long)INADDR_BROADCAST) goto ours; if (ip->ip_dst.s_addr == INADDR_ANY) goto ours; /* * FAITH(Firewall Aided Internet Translator) */ if (m->m_pkthdr.rcvif && m->m_pkthdr.rcvif->if_type == IFT_FAITH) { if (ip_keepfaith) { if (ip->ip_p == IPPROTO_TCP || ip->ip_p == IPPROTO_ICMP) goto ours; } m_freem(m); return; } /* * Not for us; forward if possible and desirable. */ if (ipforwarding == 0) { ipstat.ips_cantforward++; m_freem(m); } else ip_forward(m, 0); #ifdef IPFIREWALL_FORWARD ip_fw_fwd_addr = NULL; #endif return; ours: /* Count the packet in the ip address stats */ if (ia != NULL) { ia->ia_ifa.if_ipackets++; ia->ia_ifa.if_ibytes += m->m_pkthdr.len; } /* * If offset or IP_MF are set, must reassemble. * Otherwise, nothing need be done. * (We could look in the reassembly queue to see * if the packet was previously fragmented, * but it's not worth the time; just let them time out.) */ if (ip->ip_off & (IP_MF | IP_OFFMASK)) { sum = IPREASS_HASH(ip->ip_src.s_addr, ip->ip_id); /* * Look for queue of fragments * of this datagram. */ TAILQ_FOREACH(fp, &ipq[sum], ipq_list) if (ip->ip_id == fp->ipq_id && ip->ip_src.s_addr == fp->ipq_src.s_addr && ip->ip_dst.s_addr == fp->ipq_dst.s_addr && ip->ip_p == fp->ipq_p) goto found; fp = 0; /* check if there's a place for the new queue */ if (nipq > maxnipq) { /* * drop something from the tail of the current queue * before proceeding further */ struct ipq *q = TAILQ_LAST(&ipq[sum], ipqhead); if (q == NULL) { /* gak */ for (i = 0; i < IPREASS_NHASH; i++) { struct ipq *r = TAILQ_LAST(&ipq[i], ipqhead); if (r) { ip_freef(&ipq[i], r); break; } } } else ip_freef(&ipq[sum], q); } found: /* * Adjust ip_len to not reflect header, * convert offset of this to bytes. */ ip->ip_len -= hlen; if (ip->ip_off & IP_MF) { /* * Make sure that fragments have a data length * that's a non-zero multiple of 8 bytes. */ if (ip->ip_len == 0 || (ip->ip_len & 0x7) != 0) { ipstat.ips_toosmall++; /* XXX */ goto bad; } m->m_flags |= M_FRAG; } ip->ip_off <<= 3; /* * Attempt reassembly; if it succeeds, proceed. */ ipstat.ips_fragments++; m->m_pkthdr.header = ip; #ifdef IPDIVERT m = ip_reass(m, &ipq[sum], fp, &divert_info, &divert_cookie); #else m = ip_reass(m, &ipq[sum], fp); #endif if (m == 0) { #ifdef IPFIREWALL_FORWARD ip_fw_fwd_addr = NULL; #endif return; } ipstat.ips_reassembled++; ip = mtod(m, struct ip *); /* Get the header length of the reassembled packet */ hlen = IP_VHL_HL(ip->ip_vhl) << 2; #ifdef IPDIVERT /* Restore original checksum before diverting packet */ if (divert_info != 0) { ip->ip_len += hlen; HTONS(ip->ip_len); HTONS(ip->ip_off); ip->ip_sum = 0; if (hlen == sizeof(struct ip)) ip->ip_sum = in_cksum_hdr(ip); else ip->ip_sum = in_cksum(m, hlen); NTOHS(ip->ip_off); NTOHS(ip->ip_len); ip->ip_len -= hlen; } #endif } else ip->ip_len -= hlen; #ifdef IPDIVERT /* * Divert or tee packet to the divert protocol if required. * * If divert_info is zero then cookie should be too, so we shouldn't * need to clear them here. Assume divert_packet() does so also. */ if (divert_info != 0) { struct mbuf *clone = NULL; /* Clone packet if we're doing a 'tee' */ if ((divert_info & IP_FW_PORT_TEE_FLAG) != 0) clone = m_dup(m, M_DONTWAIT); /* Restore packet header fields to original values */ ip->ip_len += hlen; HTONS(ip->ip_len); HTONS(ip->ip_off); /* Deliver packet to divert input routine */ ip_divert_cookie = divert_cookie; divert_packet(m, 1, divert_info & 0xffff); ipstat.ips_delivered++; /* If 'tee', continue with original packet */ if (clone == NULL) return; m = clone; ip = mtod(m, struct ip *); } #endif #ifdef IPSEC /* * enforce IPsec policy checking if we are seeing last header. * note that we do not visit this with protocols with pcb layer * code - like udp/tcp/raw ip. */ if ((inetsw[ip_protox[ip->ip_p]].pr_flags & PR_LASTHDR) != 0 && ipsec4_in_reject(m, NULL)) { ipsecstat.in_polvio++; goto bad; } #endif /* * Switch out to protocol's input routine. */ ipstat.ips_delivered++; { int off = hlen; (*inetsw[ip_protox[ip->ip_p]].pr_input)(m, off); #ifdef IPFIREWALL_FORWARD ip_fw_fwd_addr = NULL; /* tcp needed it */ #endif return; } bad: #ifdef IPFIREWALL_FORWARD ip_fw_fwd_addr = NULL; #endif m_freem(m); } /* * IP software interrupt routine - to go away sometime soon */ static void ipintr(void) { struct mbuf *m; while (1) { IF_DEQUEUE(&ipintrq, m); if (m == 0) return; ip_input(m); } } /* * Take incoming datagram fragment and try to reassemble it into * whole datagram. If a chain for reassembly of this datagram already * exists, then it is given as fp; otherwise have to make a chain. * * When IPDIVERT enabled, keep additional state with each packet that * tells us if we need to divert or tee the packet we're building. */ static struct mbuf * #ifdef IPDIVERT ip_reass(m, head, fp, divinfo, divcookie) #else ip_reass(m, head, fp) #endif struct mbuf *m; struct ipqhead *head; struct ipq *fp; #ifdef IPDIVERT u_int32_t *divinfo; u_int16_t *divcookie; #endif { struct ip *ip = mtod(m, struct ip *); register struct mbuf *p, *q, *nq; struct mbuf *t; int hlen = IP_VHL_HL(ip->ip_vhl) << 2; int i, next; /* * Presence of header sizes in mbufs * would confuse code below. */ m->m_data += hlen; m->m_len -= hlen; /* * If first fragment to arrive, create a reassembly queue. */ if (fp == 0) { /* * Enforce upper bound on number of fragmented packets * for which we attempt reassembly; * If maxfrag is 0, never accept fragments. * If maxfrag is -1, accept all fragments without limitation. */ if ((ip_maxfragpackets >= 0) && (ip_nfragpackets >= ip_maxfragpackets)) goto dropfrag; ip_nfragpackets++; if ((t = m_get(M_DONTWAIT, MT_FTABLE)) == NULL) goto dropfrag; fp = mtod(t, struct ipq *); TAILQ_INSERT_HEAD(head, fp, ipq_list); nipq++; fp->ipq_ttl = IPFRAGTTL; fp->ipq_p = ip->ip_p; fp->ipq_id = ip->ip_id; fp->ipq_src = ip->ip_src; fp->ipq_dst = ip->ip_dst; fp->ipq_frags = m; m->m_nextpkt = NULL; #ifdef IPDIVERT fp->ipq_div_info = 0; fp->ipq_div_cookie = 0; #endif goto inserted; } #define GETIP(m) ((struct ip*)((m)->m_pkthdr.header)) /* * Find a segment which begins after this one does. */ for (p = NULL, q = fp->ipq_frags; q; p = q, q = q->m_nextpkt) if (GETIP(q)->ip_off > ip->ip_off) 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, otherwise * stick new segment in the proper place. * * If some of the data is dropped from the the preceding * segment, then it's checksum is invalidated. */ if (p) { i = GETIP(p)->ip_off + GETIP(p)->ip_len - ip->ip_off; if (i > 0) { if (i >= ip->ip_len) goto dropfrag; m_adj(m, i); m->m_pkthdr.csum_flags = 0; ip->ip_off += i; ip->ip_len -= i; } m->m_nextpkt = p->m_nextpkt; p->m_nextpkt = m; } else { m->m_nextpkt = fp->ipq_frags; fp->ipq_frags = m; } /* * While we overlap succeeding segments trim them or, * if they are completely covered, dequeue them. */ for (; q != NULL && ip->ip_off + ip->ip_len > GETIP(q)->ip_off; q = nq) { i = (ip->ip_off + ip->ip_len) - GETIP(q)->ip_off; if (i < GETIP(q)->ip_len) { GETIP(q)->ip_len -= i; GETIP(q)->ip_off += i; m_adj(q, i); q->m_pkthdr.csum_flags = 0; break; } nq = q->m_nextpkt; m->m_nextpkt = nq; m_freem(q); } inserted: #ifdef IPDIVERT /* * Transfer firewall instructions to the fragment structure. * Any fragment diverting causes the whole packet to divert. */ fp->ipq_div_info = *divinfo; fp->ipq_div_cookie = *divcookie; *divinfo = 0; *divcookie = 0; #endif /* * Check for complete reassembly. */ next = 0; for (p = NULL, q = fp->ipq_frags; q; p = q, q = q->m_nextpkt) { if (GETIP(q)->ip_off != next) return (0); next += GETIP(q)->ip_len; } /* Make sure the last packet didn't have the IP_MF flag */ if (p->m_flags & M_FRAG) return (0); /* * Reassembly is complete. Make sure the packet is a sane size. */ q = fp->ipq_frags; ip = GETIP(q); if (next + (IP_VHL_HL(ip->ip_vhl) << 2) > IP_MAXPACKET) { ipstat.ips_toolong++; ip_freef(head, fp); return (0); } /* * Concatenate fragments. */ m = q; t = m->m_next; m->m_next = 0; m_cat(m, t); nq = q->m_nextpkt; q->m_nextpkt = 0; for (q = nq; q != NULL; q = nq) { nq = q->m_nextpkt; q->m_nextpkt = NULL; m->m_pkthdr.csum_flags &= q->m_pkthdr.csum_flags; m->m_pkthdr.csum_data += q->m_pkthdr.csum_data; m_cat(m, q); } #ifdef IPDIVERT /* * Extract firewall instructions from the fragment structure. */ *divinfo = fp->ipq_div_info; *divcookie = fp->ipq_div_cookie; #endif /* * Create header for new ip packet by * modifying header of first packet; * dequeue and discard fragment reassembly header. * Make header visible. */ ip->ip_len = next; ip->ip_src = fp->ipq_src; ip->ip_dst = fp->ipq_dst; TAILQ_REMOVE(head, fp, ipq_list); nipq--; (void) m_free(dtom(fp)); ip_nfragpackets--; m->m_len += (IP_VHL_HL(ip->ip_vhl) << 2); m->m_data -= (IP_VHL_HL(ip->ip_vhl) << 2); /* some debugging cruft by sklower, below, will go away soon */ if (m->m_flags & M_PKTHDR) { /* XXX this should be done elsewhere */ register int plen = 0; for (t = m; t; t = t->m_next) plen += t->m_len; m->m_pkthdr.len = plen; } return (m); dropfrag: #ifdef IPDIVERT *divinfo = 0; *divcookie = 0; #endif ipstat.ips_fragdropped++; m_freem(m); return (0); #undef GETIP } /* * Free a fragment reassembly header and all * associated datagrams. */ static void ip_freef(fhp, fp) struct ipqhead *fhp; struct ipq *fp; { register struct mbuf *q; while (fp->ipq_frags) { q = fp->ipq_frags; fp->ipq_frags = q->m_nextpkt; m_freem(q); } TAILQ_REMOVE(fhp, fp, ipq_list); (void) m_free(dtom(fp)); ip_nfragpackets--; nipq--; } /* * IP timer processing; * if a timer expires on a reassembly * queue, discard it. */ void ip_slowtimo() { register struct ipq *fp; int s = splnet(); int i; for (i = 0; i < IPREASS_NHASH; i++) { for(fp = TAILQ_FIRST(&ipq[i]); fp;) { struct ipq *fpp; fpp = fp; fp = TAILQ_NEXT(fp, ipq_list); if(--fpp->ipq_ttl == 0) { ipstat.ips_fragtimeout++; ip_freef(&ipq[i], fpp); } } } /* * If we are over the maximum number of fragments * (due to the limit being lowered), drain off * enough to get down to the new limit. */ for (i = 0; i < IPREASS_NHASH; i++) { if (ip_maxfragpackets >= 0) { while (ip_nfragpackets > ip_maxfragpackets && !TAILQ_EMPTY(&ipq[i])) { ipstat.ips_fragdropped++; ip_freef(&ipq[i], TAILQ_FIRST(&ipq[i])); } } } ipflow_slowtimo(); splx(s); } /* * Drain off all datagram fragments. */ void ip_drain() { int i; for (i = 0; i < IPREASS_NHASH; i++) { while(!TAILQ_EMPTY(&ipq[i])) { ipstat.ips_fragdropped++; ip_freef(&ipq[i], TAILQ_FIRST(&ipq[i])); } } in_rtqdrain(); } /* * Do option processing on a datagram, * possibly discarding it if bad options are encountered, * or forwarding it if source-routed. * Returns 1 if packet has been forwarded/freed, * 0 if the packet should be processed further. */ static int ip_dooptions(m) struct mbuf *m; { register struct ip *ip = mtod(m, struct ip *); register u_char *cp; register struct in_ifaddr *ia; int opt, optlen, cnt, off, code, type = ICMP_PARAMPROB, forward = 0; struct in_addr *sin, dst; n_time ntime; dst = ip->ip_dst; cp = (u_char *)(ip + 1); cnt = (IP_VHL_HL(ip->ip_vhl) << 2) - sizeof (struct ip); for (; cnt > 0; cnt -= optlen, cp += optlen) { opt = cp[IPOPT_OPTVAL]; if (opt == IPOPT_EOL) break; if (opt == IPOPT_NOP) optlen = 1; else { if (cnt < IPOPT_OLEN + sizeof(*cp)) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } optlen = cp[IPOPT_OLEN]; if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } } switch (opt) { default: break; /* * Source routing with record. * Find interface with current destination address. * If none on this machine then drop if strictly routed, * or do nothing if loosely routed. * Record interface address and bring up next address * component. If strictly routed make sure next * address is on directly accessible net. */ case IPOPT_LSRR: case IPOPT_SSRR: if (optlen < IPOPT_OFFSET + sizeof(*cp)) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } ipaddr.sin_addr = ip->ip_dst; ia = (struct in_ifaddr *) ifa_ifwithaddr((struct sockaddr *)&ipaddr); if (ia == 0) { if (opt == IPOPT_SSRR) { type = ICMP_UNREACH; code = ICMP_UNREACH_SRCFAIL; goto bad; } if (!ip_dosourceroute) goto nosourcerouting; /* * Loose routing, and not at next destination * yet; nothing to do except forward. */ break; } off--; /* 0 origin */ if (off > optlen - (int)sizeof(struct in_addr)) { /* * End of source route. Should be for us. */ if (!ip_acceptsourceroute) goto nosourcerouting; save_rte(cp, ip->ip_src); break; } if (!ip_dosourceroute) { if (ipforwarding) { char buf[16]; /* aaa.bbb.ccc.ddd\0 */ /* * Acting as a router, so generate ICMP */ nosourcerouting: strcpy(buf, inet_ntoa(ip->ip_dst)); log(LOG_WARNING, "attempted source route from %s to %s\n", inet_ntoa(ip->ip_src), buf); type = ICMP_UNREACH; code = ICMP_UNREACH_SRCFAIL; goto bad; } else { /* * Not acting as a router, so silently drop. */ ipstat.ips_cantforward++; m_freem(m); return (1); } } /* * locate outgoing interface */ (void)memcpy(&ipaddr.sin_addr, cp + off, sizeof(ipaddr.sin_addr)); if (opt == IPOPT_SSRR) { #define INA struct in_ifaddr * #define SA struct sockaddr * if ((ia = (INA)ifa_ifwithdstaddr((SA)&ipaddr)) == 0) ia = (INA)ifa_ifwithnet((SA)&ipaddr); } else ia = ip_rtaddr(ipaddr.sin_addr, &ipforward_rt); if (ia == 0) { type = ICMP_UNREACH; code = ICMP_UNREACH_SRCFAIL; goto bad; } ip->ip_dst = ipaddr.sin_addr; (void)memcpy(cp + off, &(IA_SIN(ia)->sin_addr), sizeof(struct in_addr)); cp[IPOPT_OFFSET] += sizeof(struct in_addr); /* * Let ip_intr's mcast routing check handle mcast pkts */ forward = !IN_MULTICAST(ntohl(ip->ip_dst.s_addr)); break; case IPOPT_RR: if (optlen < IPOPT_OFFSET + sizeof(*cp)) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } 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 - (int)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, &ipforward_rt)) == 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; if (optlen < 4 || optlen > 40) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } if ((off = cp[IPOPT_OFFSET]) < 5) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } if (off > optlen - (int)sizeof(int32_t)) { cp[IPOPT_OFFSET + 1] += (1 << 4); if ((cp[IPOPT_OFFSET + 1] & 0xf0) == 0) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } break; } off--; /* 0 origin */ sin = (struct in_addr *)(cp + off); switch (cp[IPOPT_OFFSET + 1] & 0x0f) { case IPOPT_TS_TSONLY: break; case IPOPT_TS_TSANDADDR: if (off + sizeof(n_time) + sizeof(struct in_addr) > optlen) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; 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)); cp[IPOPT_OFFSET] += sizeof(struct in_addr); break; case IPOPT_TS_PRESPEC: if (off + sizeof(n_time) + sizeof(struct in_addr) > optlen) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } (void)memcpy(&ipaddr.sin_addr, sin, sizeof(struct in_addr)); if (ifa_ifwithaddr((SA)&ipaddr) == 0) continue; cp[IPOPT_OFFSET] += sizeof(struct in_addr); break; default: code = &cp[IPOPT_OFFSET + 1] - (u_char *)ip; goto bad; } ntime = iptime(); (void)memcpy(cp + off, &ntime, sizeof(n_time)); cp[IPOPT_OFFSET] += sizeof(n_time); } } if (forward && ipforwarding) { ip_forward(m, 1); return (1); } return (0); bad: 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. */ struct in_ifaddr * ip_rtaddr(dst, rt) struct in_addr dst; struct route *rt; { register struct sockaddr_in *sin; sin = (struct sockaddr_in *)&rt->ro_dst; if (rt->ro_rt == 0 || !(rt->ro_rt->rt_flags & RTF_UP) || dst.s_addr != sin->sin_addr.s_addr) { if (rt->ro_rt) { RTFREE(rt->ro_rt); rt->ro_rt = 0; } sin->sin_family = AF_INET; sin->sin_len = sizeof(*sin); sin->sin_addr = dst; rtalloc_ign(rt, RTF_PRCLONING); } if (rt->ro_rt == 0) return ((struct in_ifaddr *)0); return (ifatoia(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_HEADER); 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", (u_long)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", (u_long)ntohl(q->s_addr)); #endif *q++ = *p--; } /* * Last hop goes to final destination. */ *q = ip_srcrt.dst; #ifdef DIAGNOSTIC if (ipprintfs) printf(" %lx\n", (u_long)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, ECONNREFUSED }; /* * 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 rtentry *rt; int error, type = 0, code = 0; struct mbuf *mcopy; n_long dest; + struct in_addr pkt_dst; struct ifnet *destifp; #ifdef IPSEC struct ifnet dummyifp; #endif dest = 0; + /* + * Cache the destination address of the packet; this may be + * changed by use of 'ipfw fwd'. + */ + pkt_dst = ip_fw_fwd_addr == NULL ? + ip->ip_dst : ip_fw_fwd_addr->sin_addr; + #ifdef DIAGNOSTIC if (ipprintfs) printf("forward: src %lx dst %lx ttl %x\n", - (u_long)ip->ip_src.s_addr, (u_long)ip->ip_dst.s_addr, + (u_long)ip->ip_src.s_addr, (u_long)pkt_dst.s_addr, ip->ip_ttl); #endif - if (m->m_flags & (M_BCAST|M_MCAST) || in_canforward(ip->ip_dst) == 0) { + if (m->m_flags & (M_BCAST|M_MCAST) || in_canforward(pkt_dst) == 0) { ipstat.ips_cantforward++; m_freem(m); return; } #ifdef IPSTEALTH if (!ipstealth) { #endif if (ip->ip_ttl <= IPTTLDEC) { icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS, dest, 0); return; } #ifdef IPSTEALTH } #endif - if (ip_rtaddr(ip->ip_dst, &ipforward_rt) == 0) { + if (ip_rtaddr(pkt_dst, &ipforward_rt) == 0) { icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, dest, 0); return; } else rt = ipforward_rt.ro_rt; /* * Save the IP header and at most 8 bytes of the payload, * in case we need to generate an ICMP message to the src. * * We don't use m_copy() because it might return a reference * to a shared cluster. Both this function and ip_output() * assume exclusive access to the IP header in `m', so any * data in a cluster may change before we reach icmp_error(). */ MGET(mcopy, M_DONTWAIT, m->m_type); if (mcopy != NULL) { M_COPY_PKTHDR(mcopy, m); mcopy->m_len = imin((IP_VHL_HL(ip->ip_vhl) << 2) + 8, (int)ip->ip_len); m_copydata(m, 0, mcopy->m_len, mtod(mcopy, caddr_t)); } #ifdef IPSTEALTH if (!ipstealth) { #endif ip->ip_ttl -= IPTTLDEC; #ifdef IPSTEALTH } #endif /* * 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. */ 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) { + ipsendredirects && !srcrt && !ip_fw_fwd_addr) { #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; + dest = pkt_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) { ipflow_create(&ipforward_rt, 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; #ifndef IPSEC if (ipforward_rt.ro_rt) destifp = ipforward_rt.ro_rt->rt_ifp; #else /* * If the packet is routed over IPsec tunnel, tell the * originator the tunnel MTU. * tunnel MTU = if MTU - sizeof(IP) - ESP/AH hdrsiz * XXX quickhack!!! */ if (ipforward_rt.ro_rt) { struct secpolicy *sp = NULL; int ipsecerror; int ipsechdr; struct route *ro; sp = ipsec4_getpolicybyaddr(mcopy, IPSEC_DIR_OUTBOUND, IP_FORWARDING, &ipsecerror); if (sp == NULL) destifp = ipforward_rt.ro_rt->rt_ifp; else { /* count IPsec header size */ ipsechdr = ipsec4_hdrsiz(mcopy, IPSEC_DIR_OUTBOUND, NULL); /* * find the correct route for outer IPv4 * header, compute tunnel MTU. * * XXX BUG ALERT * The "dummyifp" code relies upon the fact * that icmp_error() touches only ifp->if_mtu. */ /*XXX*/ destifp = NULL; if (sp->req != NULL && sp->req->sav != NULL && sp->req->sav->sah != NULL) { ro = &sp->req->sav->sah->sa_route; if (ro->ro_rt && ro->ro_rt->rt_ifp) { dummyifp.if_mtu = ro->ro_rt->rt_ifp->if_mtu; dummyifp.if_mtu -= ipsechdr; destifp = &dummyifp; } } key_freesp(sp); } } #endif /*IPSEC*/ ipstat.ips_cantfrag++; break; case ENOBUFS: type = ICMP_SOURCEQUENCH; code = 0; break; case EACCES: /* ipfw denied packet */ m_freem(mcopy); return; } 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 ifnet *ifp; struct sdlbuf { struct sockaddr_dl sdl; u_char pad[32]; } sdlbuf; struct sockaddr_dl *sdp; struct sockaddr_dl *sdl2 = &sdlbuf.sdl; if (((ifp = m->m_pkthdr.rcvif)) && ( ifp->if_index && (ifp->if_index <= if_index))) { sdp = (struct sockaddr_dl *) (ifaddr_byindex(ifp->if_index)->ifa_addr); /* * Change our mind and don't try copy. */ if ((sdp->sdl_family != AF_LINK) || (sdp->sdl_len > sizeof(sdlbuf))) { goto makedummy; } bcopy(sdp, sdl2, sdp->sdl_len); } else { makedummy: sdl2->sdl_len = offsetof(struct sockaddr_dl, sdl_data[0]); sdl2->sdl_family = AF_LINK; sdl2->sdl_index = 0; sdl2->sdl_nlen = sdl2->sdl_alen = sdl2->sdl_slen = 0; } *mp = sbcreatecontrol((caddr_t) sdl2, sdl2->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; } Index: head/sys/netinet/ip_output.c =================================================================== --- head/sys/netinet/ip_output.c (revision 88592) +++ head/sys/netinet/ip_output.c (revision 88593) @@ -1,1961 +1,1967 @@ /* * Copyright (c) 1982, 1986, 1988, 1990, 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. * * @(#)ip_output.c 8.3 (Berkeley) 1/21/94 * $FreeBSD$ */ #define _IP_VHL #include "opt_ipfw.h" #include "opt_ipdn.h" #include "opt_ipdivert.h" #include "opt_ipfilter.h" #include "opt_ipsec.h" #include "opt_pfil_hooks.h" #include "opt_random_ip_id.h" #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include static MALLOC_DEFINE(M_IPMOPTS, "ip_moptions", "internet multicast options"); #ifdef IPSEC #include #include #ifdef IPSEC_DEBUG #include #else #define KEYDEBUG(lev,arg) #endif #endif /*IPSEC*/ #include #include #ifdef IPFIREWALL_FORWARD_DEBUG #define print_ip(a) printf("%ld.%ld.%ld.%ld",(ntohl(a.s_addr)>>24)&0xFF,\ (ntohl(a.s_addr)>>16)&0xFF,\ (ntohl(a.s_addr)>>8)&0xFF,\ (ntohl(a.s_addr))&0xFF); #endif u_short ip_id; static struct mbuf *ip_insertoptions __P((struct mbuf *, struct mbuf *, int *)); static struct ifnet *ip_multicast_if __P((struct in_addr *, int *)); static void ip_mloopback __P((struct ifnet *, struct mbuf *, struct sockaddr_in *, int)); static int ip_getmoptions __P((struct sockopt *, struct ip_moptions *)); static int ip_pcbopts __P((int, struct mbuf **, struct mbuf *)); static int ip_setmoptions __P((struct sockopt *, struct ip_moptions **)); int ip_optcopy __P((struct ip *, struct ip *)); extern struct protosw inetsw[]; /* * IP output. The packet in mbuf chain m contains a skeletal IP * header (with len, off, ttl, proto, tos, src, dst). * The mbuf chain containing the packet will be freed. * The mbuf opt, if present, will not be freed. */ int ip_output(m0, opt, ro, flags, imo) struct mbuf *m0; struct mbuf *opt; struct route *ro; int flags; struct ip_moptions *imo; { struct ip *ip, *mhip; struct ifnet *ifp; struct mbuf *m = m0; int hlen = sizeof (struct ip); int len, off, error = 0; struct route iproute; struct sockaddr_in *dst; struct in_ifaddr *ia; int isbroadcast, sw_csum; + struct in_addr pkt_dst; #ifdef IPSEC struct socket *so = NULL; struct secpolicy *sp = NULL; #endif u_int16_t divert_cookie; /* firewall cookie */ #ifdef PFIL_HOOKS struct packet_filter_hook *pfh; struct mbuf *m1; int rv; #endif /* PFIL_HOOKS */ #ifdef IPFIREWALL_FORWARD int fwd_rewrite_src = 0; #endif struct ip_fw *rule = NULL; #ifdef IPDIVERT /* Get and reset firewall cookie */ divert_cookie = ip_divert_cookie; ip_divert_cookie = 0; #else divert_cookie = 0; #endif /* * dummynet packet are prepended a vestigial mbuf with * m_type = MT_DUMMYNET and m_data pointing to the matching * rule. */ if (m->m_type == MT_DUMMYNET) { /* * the packet was already tagged, so part of the * processing was already done, and we need to go down. * Get parameters from the header. */ rule = (struct ip_fw *)(m->m_data) ; opt = NULL ; ro = & ( ((struct dn_pkt *)m)->ro ) ; imo = NULL ; dst = ((struct dn_pkt *)m)->dn_dst ; ifp = ((struct dn_pkt *)m)->ifp ; flags = ((struct dn_pkt *)m)->flags ; m0 = m = m->m_next ; #ifdef IPSEC so = ipsec_getsocket(m); (void)ipsec_setsocket(m, NULL); #endif ip = mtod(m, struct ip *); hlen = IP_VHL_HL(ip->ip_vhl) << 2 ; ia = ifatoia(ro->ro_rt->rt_ifa); goto sendit; } else rule = NULL ; #ifdef IPSEC so = ipsec_getsocket(m); (void)ipsec_setsocket(m, NULL); #endif #ifdef DIAGNOSTIC if ((m->m_flags & M_PKTHDR) == 0) panic("ip_output no HDR"); #endif if (opt) { m = ip_insertoptions(m, opt, &len); hlen = len; } ip = mtod(m, struct ip *); + pkt_dst = ip_fw_fwd_addr == NULL + ? ip->ip_dst : ip_fw_fwd_addr->sin_addr; + /* * Fill in IP header. */ if ((flags & (IP_FORWARDING|IP_RAWOUTPUT)) == 0) { ip->ip_vhl = IP_MAKE_VHL(IPVERSION, hlen >> 2); ip->ip_off &= IP_DF; #ifdef RANDOM_IP_ID ip->ip_id = ip_randomid(); #else ip->ip_id = htons(ip_id++); #endif ipstat.ips_localout++; } else { hlen = IP_VHL_HL(ip->ip_vhl) << 2; } /* Route packet. */ if (ro == NULL) { ro = &iproute; bzero(ro, sizeof(*ro)); } dst = (struct sockaddr_in *)&ro->ro_dst; /* * If there is a cached route, * check that it is to the same destination * and is still up. If not, free it and try again. */ if (ro->ro_rt && ((ro->ro_rt->rt_flags & RTF_UP) == 0 || - dst->sin_addr.s_addr != ip->ip_dst.s_addr)) { + dst->sin_addr.s_addr != pkt_dst.s_addr)) { RTFREE(ro->ro_rt); ro->ro_rt = (struct rtentry *)0; } if (ro->ro_rt == 0) { dst->sin_family = AF_INET; dst->sin_len = sizeof(*dst); - dst->sin_addr = ip->ip_dst; + dst->sin_addr = pkt_dst; } /* * If routing to interface only, * short circuit routing lookup. */ if (flags & IP_ROUTETOIF) { if ((ia = ifatoia(ifa_ifwithdstaddr(sintosa(dst)))) == 0 && (ia = ifatoia(ifa_ifwithnet(sintosa(dst)))) == 0) { ipstat.ips_noroute++; error = ENETUNREACH; goto bad; } ifp = ia->ia_ifp; ip->ip_ttl = 1; isbroadcast = in_broadcast(dst->sin_addr, ifp); } else if (IN_MULTICAST(ntohl(ip->ip_dst.s_addr)) && imo != NULL && imo->imo_multicast_ifp != NULL) { /* * Bypass the normal routing lookup for multicast * packets if the interface is specified. */ ifp = imo->imo_multicast_ifp; IFP_TO_IA(ifp, ia); isbroadcast = 0; /* fool gcc */ } else { /* * If this is the case, we probably don't want to allocate * a protocol-cloned route since we didn't get one from the * ULP. This lets TCP do its thing, while not burdening * forwarding or ICMP with the overhead of cloning a route. * Of course, we still want to do any cloning requested by * the link layer, as this is probably required in all cases * for correct operation (as it is for ARP). */ if (ro->ro_rt == 0) rtalloc_ign(ro, RTF_PRCLONING); if (ro->ro_rt == 0) { ipstat.ips_noroute++; error = EHOSTUNREACH; goto bad; } ia = ifatoia(ro->ro_rt->rt_ifa); ifp = ro->ro_rt->rt_ifp; ro->ro_rt->rt_use++; if (ro->ro_rt->rt_flags & RTF_GATEWAY) dst = (struct sockaddr_in *)ro->ro_rt->rt_gateway; if (ro->ro_rt->rt_flags & RTF_HOST) isbroadcast = (ro->ro_rt->rt_flags & RTF_BROADCAST); else isbroadcast = in_broadcast(dst->sin_addr, ifp); } - if (IN_MULTICAST(ntohl(ip->ip_dst.s_addr))) { + if (IN_MULTICAST(ntohl(pkt_dst.s_addr))) { struct in_multi *inm; m->m_flags |= M_MCAST; /* * IP destination address is multicast. Make sure "dst" * still points to the address in "ro". (It may have been * changed to point to a gateway address, above.) */ dst = (struct sockaddr_in *)&ro->ro_dst; /* * See if the caller provided any multicast options */ if (imo != NULL) { ip->ip_ttl = imo->imo_multicast_ttl; if (imo->imo_multicast_vif != -1) ip->ip_src.s_addr = ip_mcast_src(imo->imo_multicast_vif); } else ip->ip_ttl = IP_DEFAULT_MULTICAST_TTL; /* * Confirm that the outgoing interface supports multicast. */ if ((imo == NULL) || (imo->imo_multicast_vif == -1)) { if ((ifp->if_flags & IFF_MULTICAST) == 0) { ipstat.ips_noroute++; error = ENETUNREACH; goto bad; } } /* * If source address not specified yet, use address * of outgoing interface. */ if (ip->ip_src.s_addr == INADDR_ANY) { /* Interface may have no addresses. */ if (ia != NULL) ip->ip_src = IA_SIN(ia)->sin_addr; } - IN_LOOKUP_MULTI(ip->ip_dst, ifp, inm); + IN_LOOKUP_MULTI(pkt_dst, ifp, inm); if (inm != NULL && (imo == NULL || imo->imo_multicast_loop)) { /* * If we belong to the destination multicast group * on the outgoing interface, and the caller did not * forbid loopback, loop back a copy. */ ip_mloopback(ifp, m, dst, hlen); } else { /* * If we are acting as a multicast router, perform * multicast forwarding as if the packet had just * arrived on the interface to which we are about * to send. The multicast forwarding function * recursively calls this function, using the * IP_FORWARDING flag to prevent infinite recursion. * * Multicasts that are looped back by ip_mloopback(), * above, will be forwarded by the ip_input() routine, * if necessary. */ if (ip_mrouter && (flags & IP_FORWARDING) == 0) { /* * Check if rsvp daemon is running. If not, don't * set ip_moptions. This ensures that the packet * is multicast and not just sent down one link * as prescribed by rsvpd. */ if (!rsvp_on) imo = NULL; if (ip_mforward(ip, ifp, m, imo) != 0) { m_freem(m); goto done; } } } /* * Multicasts with a time-to-live of zero may be looped- * back, above, but must not be transmitted on a network. * Also, multicasts addressed to the loopback interface * are not sent -- the above call to ip_mloopback() will * loop back a copy if this host actually belongs to the * destination group on the loopback interface. */ if (ip->ip_ttl == 0 || ifp->if_flags & IFF_LOOPBACK) { m_freem(m); goto done; } goto sendit; } #ifndef notdef /* * If source address not specified yet, use address * of outgoing interface. */ if (ip->ip_src.s_addr == INADDR_ANY) { /* Interface may have no addresses. */ if (ia != NULL) { ip->ip_src = IA_SIN(ia)->sin_addr; #ifdef IPFIREWALL_FORWARD /* Keep note that we did this - if the firewall changes * the next-hop, our interface may change, changing the * default source IP. It's a shame so much effort happens * twice. Oh well. */ fwd_rewrite_src++; #endif /* IPFIREWALL_FORWARD */ } } #endif /* notdef */ /* * Verify that we have any chance at all of being able to queue * the packet or packet fragments */ if ((ifp->if_snd.ifq_len + ip->ip_len / ifp->if_mtu + 1) >= ifp->if_snd.ifq_maxlen) { error = ENOBUFS; ipstat.ips_odropped++; goto bad; } /* * Look for broadcast address and * and verify user is allowed to send * such a packet. */ if (isbroadcast) { if ((ifp->if_flags & IFF_BROADCAST) == 0) { error = EADDRNOTAVAIL; goto bad; } if ((flags & IP_ALLOWBROADCAST) == 0) { error = EACCES; goto bad; } /* don't allow broadcast messages to be fragmented */ if ((u_short)ip->ip_len > ifp->if_mtu) { error = EMSGSIZE; goto bad; } m->m_flags |= M_BCAST; } else { m->m_flags &= ~M_BCAST; } sendit: #ifdef IPSEC /* get SP for this packet */ if (so == NULL) sp = ipsec4_getpolicybyaddr(m, IPSEC_DIR_OUTBOUND, flags, &error); else sp = ipsec4_getpolicybysock(m, IPSEC_DIR_OUTBOUND, so, &error); if (sp == NULL) { ipsecstat.out_inval++; goto bad; } error = 0; /* check policy */ switch (sp->policy) { case IPSEC_POLICY_DISCARD: /* * This packet is just discarded. */ ipsecstat.out_polvio++; goto bad; case IPSEC_POLICY_BYPASS: case IPSEC_POLICY_NONE: /* no need to do IPsec. */ goto skip_ipsec; case IPSEC_POLICY_IPSEC: if (sp->req == NULL) { /* acquire a policy */ error = key_spdacquire(sp); goto bad; } break; case IPSEC_POLICY_ENTRUST: default: printf("ip_output: Invalid policy found. %d\n", sp->policy); } { struct ipsec_output_state state; bzero(&state, sizeof(state)); state.m = m; if (flags & IP_ROUTETOIF) { state.ro = &iproute; bzero(&iproute, sizeof(iproute)); } else state.ro = ro; state.dst = (struct sockaddr *)dst; ip->ip_sum = 0; /* * XXX * delayed checksums are not currently compatible with IPsec */ if (m->m_pkthdr.csum_flags & CSUM_DELAY_DATA) { in_delayed_cksum(m); m->m_pkthdr.csum_flags &= ~CSUM_DELAY_DATA; } HTONS(ip->ip_len); HTONS(ip->ip_off); error = ipsec4_output(&state, sp, flags); m = state.m; if (flags & IP_ROUTETOIF) { /* * if we have tunnel mode SA, we may need to ignore * IP_ROUTETOIF. */ if (state.ro != &iproute || state.ro->ro_rt != NULL) { flags &= ~IP_ROUTETOIF; ro = state.ro; } } else ro = state.ro; dst = (struct sockaddr_in *)state.dst; if (error) { /* mbuf is already reclaimed in ipsec4_output. */ m0 = NULL; switch (error) { case EHOSTUNREACH: case ENETUNREACH: case EMSGSIZE: case ENOBUFS: case ENOMEM: break; default: printf("ip4_output (ipsec): error code %d\n", error); /*fall through*/ case ENOENT: /* don't show these error codes to the user */ error = 0; break; } goto bad; } } /* be sure to update variables that are affected by ipsec4_output() */ ip = mtod(m, struct ip *); #ifdef _IP_VHL hlen = IP_VHL_HL(ip->ip_vhl) << 2; #else hlen = ip->ip_hl << 2; #endif if (ro->ro_rt == NULL) { if ((flags & IP_ROUTETOIF) == 0) { printf("ip_output: " "can't update route after IPsec processing\n"); error = EHOSTUNREACH; /*XXX*/ goto bad; } } else { ia = ifatoia(ro->ro_rt->rt_ifa); ifp = ro->ro_rt->rt_ifp; } /* make it flipped, again. */ NTOHS(ip->ip_len); NTOHS(ip->ip_off); skip_ipsec: #endif /*IPSEC*/ /* * IpHack's section. * - Xlate: translate packet's addr/port (NAT). * - Firewall: deny/allow/etc. * - Wrap: fake packet's addr/port * - Encapsulate: put it in another IP and send out. */ #ifdef PFIL_HOOKS /* * Run through list of hooks for output packets. */ m1 = m; pfh = pfil_hook_get(PFIL_OUT, &inetsw[ip_protox[IPPROTO_IP]].pr_pfh); for (; pfh; pfh = TAILQ_NEXT(pfh, pfil_link)) if (pfh->pfil_func) { rv = pfh->pfil_func(ip, hlen, ifp, 1, &m1); if (rv) { error = EHOSTUNREACH; goto done; } m = m1; if (m == NULL) goto done; ip = mtod(m, struct ip *); } #endif /* PFIL_HOOKS */ /* * Check with the firewall... + * but not if we are already being fwd'd from a firewall. */ - if (fw_enable && IPFW_LOADED) { + if (fw_enable && IPFW_LOADED && !ip_fw_fwd_addr) { struct sockaddr_in *old = dst; off = ip_fw_chk_ptr(&ip, hlen, ifp, &divert_cookie, &m, &rule, &dst); /* * On return we must do the following: * m == NULL -> drop the pkt (old interface, deprecated) * (off & IP_FW_PORT_DENY_FLAG) -> drop the pkt (new interface) * 1<=off<= 0xffff -> DIVERT * (off & IP_FW_PORT_DYNT_FLAG) -> send to a DUMMYNET pipe * (off & IP_FW_PORT_TEE_FLAG) -> TEE the packet * dst != old -> IPFIREWALL_FORWARD * off==0, dst==old -> accept * If some of the above modules are not compiled in, then * we should't have to check the corresponding condition * (because the ipfw control socket should not accept * unsupported rules), but better play safe and drop * packets in case of doubt. */ if (off & IP_FW_PORT_DENY_FLAG) { /* XXX new interface-denied */ if (m) m_freem(m); error = EACCES; goto done; } if (!m) { /* firewall said to reject */ static int __debug=10; if (__debug > 0) { printf( "firewall returns NULL, please update!\n"); __debug--; } error = EACCES; goto done; } if (off == 0 && dst == old) /* common case */ goto pass; if (DUMMYNET_LOADED && (off & IP_FW_PORT_DYNT_FLAG) != 0) { /* * pass the pkt to dummynet. Need to include * pipe number, m, ifp, ro, dst because these are * not recomputed in the next pass. * All other parameters have been already used and * so they are not needed anymore. * XXX note: if the ifp or ro entry are deleted * while a pkt is in dummynet, we are in trouble! */ error = ip_dn_io_ptr(off & 0xffff, DN_TO_IP_OUT, m, ifp, ro, dst, rule, flags); goto done; } #ifdef IPDIVERT if (off != 0 && (off & IP_FW_PORT_DYNT_FLAG) == 0) { struct mbuf *clone = NULL; /* Clone packet if we're doing a 'tee' */ if ((off & IP_FW_PORT_TEE_FLAG) != 0) clone = m_dup(m, M_DONTWAIT); /* * XXX * delayed checksums are not currently compatible * with divert sockets. */ if (m->m_pkthdr.csum_flags & CSUM_DELAY_DATA) { in_delayed_cksum(m); m->m_pkthdr.csum_flags &= ~CSUM_DELAY_DATA; } /* Restore packet header fields to original values */ HTONS(ip->ip_len); HTONS(ip->ip_off); /* Deliver packet to divert input routine */ ip_divert_cookie = divert_cookie; divert_packet(m, 0, off & 0xffff); /* If 'tee', continue with original packet */ if (clone != NULL) { m = clone; ip = mtod(m, struct ip *); goto pass; } goto done; } #endif #ifdef IPFIREWALL_FORWARD /* Here we check dst to make sure it's directly reachable on the * interface we previously thought it was. * If it isn't (which may be likely in some situations) we have * to re-route it (ie, find a route for the next-hop and the * associated interface) and set them here. This is nested * forwarding which in most cases is undesirable, except where * such control is nigh impossible. So we do it here. * And I'm babbling. */ if (off == 0 && old != dst) { struct in_ifaddr *ia; /* It's changed... */ /* There must be a better way to do this next line... */ static struct route sro_fwd, *ro_fwd = &sro_fwd; #ifdef IPFIREWALL_FORWARD_DEBUG printf("IPFIREWALL_FORWARD: New dst ip: "); print_ip(dst->sin_addr); printf("\n"); #endif /* * We need to figure out if we have been forwarded * to a local socket. If so then we should somehow * "loop back" to ip_input, and get directed to the * PCB as if we had received this packet. This is * because it may be dificult to identify the packets * you want to forward until they are being output * and have selected an interface. (e.g. locally * initiated packets) If we used the loopback inteface, * we would not be able to control what happens * as the packet runs through ip_input() as * it is done through a ISR. */ LIST_FOREACH(ia, INADDR_HASH(dst->sin_addr.s_addr), ia_hash) { /* * If the addr to forward to is one * of ours, we pretend to * be the destination for this packet. */ if (IA_SIN(ia)->sin_addr.s_addr == dst->sin_addr.s_addr) break; } if (ia) { /* tell ip_input "dont filter" */ ip_fw_fwd_addr = dst; if (m->m_pkthdr.rcvif == NULL) m->m_pkthdr.rcvif = ifunit("lo0"); if (m->m_pkthdr.csum_flags & CSUM_DELAY_DATA) { m->m_pkthdr.csum_flags |= CSUM_DATA_VALID | CSUM_PSEUDO_HDR; m0->m_pkthdr.csum_data = 0xffff; } m->m_pkthdr.csum_flags |= CSUM_IP_CHECKED | CSUM_IP_VALID; HTONS(ip->ip_len); HTONS(ip->ip_off); ip_input(m); goto done; } /* Some of the logic for this was * nicked from above. * * This rewrites the cached route in a local PCB. * Is this what we want to do? */ bcopy(dst, &ro_fwd->ro_dst, sizeof(*dst)); ro_fwd->ro_rt = 0; rtalloc_ign(ro_fwd, RTF_PRCLONING); if (ro_fwd->ro_rt == 0) { ipstat.ips_noroute++; error = EHOSTUNREACH; goto bad; } ia = ifatoia(ro_fwd->ro_rt->rt_ifa); ifp = ro_fwd->ro_rt->rt_ifp; ro_fwd->ro_rt->rt_use++; if (ro_fwd->ro_rt->rt_flags & RTF_GATEWAY) dst = (struct sockaddr_in *)ro_fwd->ro_rt->rt_gateway; if (ro_fwd->ro_rt->rt_flags & RTF_HOST) isbroadcast = (ro_fwd->ro_rt->rt_flags & RTF_BROADCAST); else isbroadcast = in_broadcast(dst->sin_addr, ifp); if (ro->ro_rt) RTFREE(ro->ro_rt); ro->ro_rt = ro_fwd->ro_rt; dst = (struct sockaddr_in *)&ro_fwd->ro_dst; /* * If we added a default src ip earlier, * which would have been gotten from the-then * interface, do it again, from the new one. */ if (fwd_rewrite_src) ip->ip_src = IA_SIN(ia)->sin_addr; goto pass ; } #endif /* IPFIREWALL_FORWARD */ /* * if we get here, none of the above matches, and * we have to drop the pkt */ m_freem(m); error = EACCES; /* not sure this is the right error msg */ goto done; } + ip_fw_fwd_addr = NULL; pass: m->m_pkthdr.csum_flags |= CSUM_IP; sw_csum = m->m_pkthdr.csum_flags & ~ifp->if_hwassist; if (sw_csum & CSUM_DELAY_DATA) { in_delayed_cksum(m); sw_csum &= ~CSUM_DELAY_DATA; } m->m_pkthdr.csum_flags &= ifp->if_hwassist; /* * If small enough for interface, or the interface will take * care of the fragmentation for us, can just send directly. */ if ((u_short)ip->ip_len <= ifp->if_mtu || ifp->if_hwassist & CSUM_FRAGMENT) { HTONS(ip->ip_len); HTONS(ip->ip_off); ip->ip_sum = 0; if (sw_csum & CSUM_DELAY_IP) { if (ip->ip_vhl == IP_VHL_BORING) { ip->ip_sum = in_cksum_hdr(ip); } else { ip->ip_sum = in_cksum(m, hlen); } } /* Record statistics for this interface address. */ if (!(flags & IP_FORWARDING) && ia) { ia->ia_ifa.if_opackets++; ia->ia_ifa.if_obytes += m->m_pkthdr.len; } #ifdef IPSEC /* clean ipsec history once it goes out of the node */ ipsec_delaux(m); #endif error = (*ifp->if_output)(ifp, m, (struct sockaddr *)dst, ro->ro_rt); goto done; } /* * Too large for interface; fragment if possible. * Must be able to put at least 8 bytes per fragment. */ if (ip->ip_off & IP_DF) { error = EMSGSIZE; /* * This case can happen if the user changed the MTU * of an interface after enabling IP on it. Because * most netifs don't keep track of routes pointing to * them, there is no way for one to update all its * routes when the MTU is changed. */ if ((ro->ro_rt->rt_flags & (RTF_UP | RTF_HOST)) && !(ro->ro_rt->rt_rmx.rmx_locks & RTV_MTU) && (ro->ro_rt->rt_rmx.rmx_mtu > ifp->if_mtu)) { ro->ro_rt->rt_rmx.rmx_mtu = ifp->if_mtu; } ipstat.ips_cantfrag++; goto bad; } len = (ifp->if_mtu - hlen) &~ 7; if (len < 8) { error = EMSGSIZE; goto bad; } /* * if the interface will not calculate checksums on * fragmented packets, then do it here. */ if (m->m_pkthdr.csum_flags & CSUM_DELAY_DATA && (ifp->if_hwassist & CSUM_IP_FRAGS) == 0) { in_delayed_cksum(m); m->m_pkthdr.csum_flags &= ~CSUM_DELAY_DATA; } { int mhlen, firstlen = len; struct mbuf **mnext = &m->m_nextpkt; int nfrags = 1; /* * Loop through length of segment after first fragment, * make new header and copy data of each part and link onto chain. */ m0 = m; mhlen = sizeof (struct ip); for (off = hlen + len; off < (u_short)ip->ip_len; off += len) { MGETHDR(m, M_DONTWAIT, MT_HEADER); if (m == 0) { error = ENOBUFS; ipstat.ips_odropped++; goto sendorfree; } m->m_flags |= (m0->m_flags & M_MCAST) | M_FRAG; m->m_data += max_linkhdr; mhip = mtod(m, struct ip *); *mhip = *ip; if (hlen > sizeof (struct ip)) { mhlen = ip_optcopy(ip, mhip) + sizeof (struct ip); mhip->ip_vhl = IP_MAKE_VHL(IPVERSION, mhlen >> 2); } m->m_len = mhlen; mhip->ip_off = ((off - hlen) >> 3) + ip->ip_off; if (off + len >= (u_short)ip->ip_len) len = (u_short)ip->ip_len - off; else mhip->ip_off |= IP_MF; mhip->ip_len = htons((u_short)(len + mhlen)); m->m_next = m_copy(m0, off, len); if (m->m_next == 0) { (void) m_free(m); error = ENOBUFS; /* ??? */ ipstat.ips_odropped++; goto sendorfree; } m->m_pkthdr.len = mhlen + len; m->m_pkthdr.rcvif = (struct ifnet *)0; m->m_pkthdr.csum_flags = m0->m_pkthdr.csum_flags; HTONS(mhip->ip_off); mhip->ip_sum = 0; if (sw_csum & CSUM_DELAY_IP) { if (mhip->ip_vhl == IP_VHL_BORING) { mhip->ip_sum = in_cksum_hdr(mhip); } else { mhip->ip_sum = in_cksum(m, mhlen); } } *mnext = m; mnext = &m->m_nextpkt; nfrags++; } ipstat.ips_ofragments += nfrags; /* set first/last markers for fragment chain */ m->m_flags |= M_LASTFRAG; m0->m_flags |= M_FIRSTFRAG | M_FRAG; m0->m_pkthdr.csum_data = nfrags; /* * Update first fragment by trimming what's been copied out * and updating header, then send each fragment (in order). */ m = m0; m_adj(m, hlen + firstlen - (u_short)ip->ip_len); m->m_pkthdr.len = hlen + firstlen; ip->ip_len = htons((u_short)m->m_pkthdr.len); ip->ip_off |= IP_MF; HTONS(ip->ip_off); ip->ip_sum = 0; if (sw_csum & CSUM_DELAY_IP) { if (ip->ip_vhl == IP_VHL_BORING) { ip->ip_sum = in_cksum_hdr(ip); } else { ip->ip_sum = in_cksum(m, hlen); } } sendorfree: for (m = m0; m; m = m0) { m0 = m->m_nextpkt; m->m_nextpkt = 0; #ifdef IPSEC /* clean ipsec history once it goes out of the node */ ipsec_delaux(m); #endif if (error == 0) { /* Record statistics for this interface address. */ if (ia != NULL) { ia->ia_ifa.if_opackets++; ia->ia_ifa.if_obytes += m->m_pkthdr.len; } error = (*ifp->if_output)(ifp, m, (struct sockaddr *)dst, ro->ro_rt); } else m_freem(m); } if (error == 0) ipstat.ips_fragmented++; } done: #ifdef IPSEC if (ro == &iproute && ro->ro_rt) { RTFREE(ro->ro_rt); ro->ro_rt = NULL; } if (sp != NULL) { KEYDEBUG(KEYDEBUG_IPSEC_STAMP, printf("DP ip_output call free SP:%p\n", sp)); key_freesp(sp); } #endif /* IPSEC */ return (error); bad: m_freem(m); goto done; } void in_delayed_cksum(struct mbuf *m) { struct ip *ip; u_short csum, offset; ip = mtod(m, struct ip *); offset = IP_VHL_HL(ip->ip_vhl) << 2 ; csum = in_cksum_skip(m, ip->ip_len, offset); if (m->m_pkthdr.csum_flags & CSUM_UDP && csum == 0) csum = 0xffff; offset += m->m_pkthdr.csum_data; /* checksum offset */ if (offset + sizeof(u_short) > m->m_len) { printf("delayed m_pullup, m->len: %d off: %d p: %d\n", m->m_len, offset, ip->ip_p); /* * XXX * this shouldn't happen, but if it does, the * correct behavior may be to insert the checksum * in the existing chain instead of rearranging it. */ m = m_pullup(m, offset + sizeof(u_short)); } *(u_short *)(m->m_data + offset) = csum; } /* * Insert IP options into preformed packet. * Adjust IP destination as required for IP source routing, * as indicated by a non-zero in_addr at the start of the options. * * XXX This routine assumes that the packet has no options in place. */ static struct mbuf * ip_insertoptions(m, opt, phlen) register struct mbuf *m; struct mbuf *opt; int *phlen; { register struct ipoption *p = mtod(opt, struct ipoption *); struct mbuf *n; register struct ip *ip = mtod(m, struct ip *); unsigned optlen; optlen = opt->m_len - sizeof(p->ipopt_dst); if (optlen + (u_short)ip->ip_len > IP_MAXPACKET) return (m); /* XXX should fail */ if (p->ipopt_dst.s_addr) ip->ip_dst = p->ipopt_dst; if (m->m_flags & M_EXT || m->m_data - optlen < m->m_pktdat) { MGETHDR(n, M_DONTWAIT, MT_HEADER); if (n == 0) return (m); n->m_pkthdr.rcvif = (struct ifnet *)0; n->m_pkthdr.len = m->m_pkthdr.len + optlen; m->m_len -= sizeof(struct ip); m->m_data += sizeof(struct ip); n->m_next = m; m = n; m->m_len = optlen + sizeof(struct ip); m->m_data += max_linkhdr; (void)memcpy(mtod(m, void *), ip, sizeof(struct ip)); } else { m->m_data -= optlen; m->m_len += optlen; m->m_pkthdr.len += optlen; ovbcopy((caddr_t)ip, mtod(m, caddr_t), sizeof(struct ip)); } ip = mtod(m, struct ip *); bcopy(p->ipopt_list, ip + 1, optlen); *phlen = sizeof(struct ip) + optlen; ip->ip_vhl = IP_MAKE_VHL(IPVERSION, *phlen >> 2); ip->ip_len += optlen; return (m); } /* * Copy options from ip to jp, * omitting those not copied during fragmentation. */ int ip_optcopy(ip, jp) struct ip *ip, *jp; { register u_char *cp, *dp; int opt, optlen, cnt; cp = (u_char *)(ip + 1); dp = (u_char *)(jp + 1); cnt = (IP_VHL_HL(ip->ip_vhl) << 2) - sizeof (struct ip); for (; cnt > 0; cnt -= optlen, cp += optlen) { opt = cp[0]; if (opt == IPOPT_EOL) break; if (opt == IPOPT_NOP) { /* Preserve for IP mcast tunnel's LSRR alignment. */ *dp++ = IPOPT_NOP; optlen = 1; continue; } #ifdef DIAGNOSTIC if (cnt < IPOPT_OLEN + sizeof(*cp)) panic("malformed IPv4 option passed to ip_optcopy"); #endif optlen = cp[IPOPT_OLEN]; #ifdef DIAGNOSTIC if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) panic("malformed IPv4 option passed to ip_optcopy"); #endif /* bogus lengths should have been caught by ip_dooptions */ if (optlen > cnt) optlen = cnt; if (IPOPT_COPIED(opt)) { bcopy(cp, dp, optlen); dp += optlen; } } for (optlen = dp - (u_char *)(jp+1); optlen & 0x3; optlen++) *dp++ = IPOPT_EOL; return (optlen); } /* * IP socket option processing. */ int ip_ctloutput(so, sopt) struct socket *so; struct sockopt *sopt; { struct inpcb *inp = sotoinpcb(so); int error, optval; error = optval = 0; if (sopt->sopt_level != IPPROTO_IP) { return (EINVAL); } switch (sopt->sopt_dir) { case SOPT_SET: switch (sopt->sopt_name) { case IP_OPTIONS: #ifdef notyet case IP_RETOPTS: #endif { struct mbuf *m; if (sopt->sopt_valsize > MLEN) { error = EMSGSIZE; break; } MGET(m, sopt->sopt_td ? M_TRYWAIT : M_DONTWAIT, MT_HEADER); if (m == 0) { error = ENOBUFS; break; } m->m_len = sopt->sopt_valsize; error = sooptcopyin(sopt, mtod(m, char *), m->m_len, m->m_len); return (ip_pcbopts(sopt->sopt_name, &inp->inp_options, m)); } case IP_TOS: case IP_TTL: case IP_RECVOPTS: case IP_RECVRETOPTS: case IP_RECVDSTADDR: case IP_RECVIF: case IP_FAITH: error = sooptcopyin(sopt, &optval, sizeof optval, sizeof optval); if (error) break; switch (sopt->sopt_name) { case IP_TOS: inp->inp_ip_tos = optval; break; case IP_TTL: inp->inp_ip_ttl = optval; break; #define OPTSET(bit) \ if (optval) \ inp->inp_flags |= bit; \ else \ inp->inp_flags &= ~bit; case IP_RECVOPTS: OPTSET(INP_RECVOPTS); break; case IP_RECVRETOPTS: OPTSET(INP_RECVRETOPTS); break; case IP_RECVDSTADDR: OPTSET(INP_RECVDSTADDR); break; case IP_RECVIF: OPTSET(INP_RECVIF); break; case IP_FAITH: OPTSET(INP_FAITH); break; } break; #undef OPTSET case IP_MULTICAST_IF: case IP_MULTICAST_VIF: case IP_MULTICAST_TTL: case IP_MULTICAST_LOOP: case IP_ADD_MEMBERSHIP: case IP_DROP_MEMBERSHIP: error = ip_setmoptions(sopt, &inp->inp_moptions); break; case IP_PORTRANGE: error = sooptcopyin(sopt, &optval, sizeof optval, sizeof optval); if (error) break; switch (optval) { case IP_PORTRANGE_DEFAULT: inp->inp_flags &= ~(INP_LOWPORT); inp->inp_flags &= ~(INP_HIGHPORT); break; case IP_PORTRANGE_HIGH: inp->inp_flags &= ~(INP_LOWPORT); inp->inp_flags |= INP_HIGHPORT; break; case IP_PORTRANGE_LOW: inp->inp_flags &= ~(INP_HIGHPORT); inp->inp_flags |= INP_LOWPORT; break; default: error = EINVAL; break; } break; #ifdef IPSEC case IP_IPSEC_POLICY: { caddr_t req; size_t len = 0; int priv; struct mbuf *m; int optname; if ((error = soopt_getm(sopt, &m)) != 0) /* XXX */ break; if ((error = soopt_mcopyin(sopt, m)) != 0) /* XXX */ break; priv = (sopt->sopt_td != NULL && suser_td(sopt->sopt_td) != 0) ? 0 : 1; req = mtod(m, caddr_t); len = m->m_len; optname = sopt->sopt_name; error = ipsec4_set_policy(inp, optname, req, len, priv); m_freem(m); break; } #endif /*IPSEC*/ default: error = ENOPROTOOPT; break; } break; case SOPT_GET: switch (sopt->sopt_name) { case IP_OPTIONS: case IP_RETOPTS: if (inp->inp_options) error = sooptcopyout(sopt, mtod(inp->inp_options, char *), inp->inp_options->m_len); else sopt->sopt_valsize = 0; break; case IP_TOS: case IP_TTL: case IP_RECVOPTS: case IP_RECVRETOPTS: case IP_RECVDSTADDR: case IP_RECVIF: case IP_PORTRANGE: case IP_FAITH: switch (sopt->sopt_name) { case IP_TOS: optval = inp->inp_ip_tos; break; case IP_TTL: optval = inp->inp_ip_ttl; break; #define OPTBIT(bit) (inp->inp_flags & bit ? 1 : 0) case IP_RECVOPTS: optval = OPTBIT(INP_RECVOPTS); break; case IP_RECVRETOPTS: optval = OPTBIT(INP_RECVRETOPTS); break; case IP_RECVDSTADDR: optval = OPTBIT(INP_RECVDSTADDR); break; case IP_RECVIF: optval = OPTBIT(INP_RECVIF); break; case IP_PORTRANGE: if (inp->inp_flags & INP_HIGHPORT) optval = IP_PORTRANGE_HIGH; else if (inp->inp_flags & INP_LOWPORT) optval = IP_PORTRANGE_LOW; else optval = 0; break; case IP_FAITH: optval = OPTBIT(INP_FAITH); break; } error = sooptcopyout(sopt, &optval, sizeof optval); break; case IP_MULTICAST_IF: case IP_MULTICAST_VIF: case IP_MULTICAST_TTL: case IP_MULTICAST_LOOP: case IP_ADD_MEMBERSHIP: case IP_DROP_MEMBERSHIP: error = ip_getmoptions(sopt, inp->inp_moptions); break; #ifdef IPSEC case IP_IPSEC_POLICY: { struct mbuf *m = NULL; caddr_t req = NULL; size_t len = 0; if (m != 0) { req = mtod(m, caddr_t); len = m->m_len; } error = ipsec4_get_policy(sotoinpcb(so), req, len, &m); if (error == 0) error = soopt_mcopyout(sopt, m); /* XXX */ if (error == 0) m_freem(m); break; } #endif /*IPSEC*/ default: error = ENOPROTOOPT; break; } break; } return (error); } /* * Set up IP options in pcb for insertion in output packets. * Store in mbuf with pointer in pcbopt, adding pseudo-option * with destination address if source routed. */ static int ip_pcbopts(optname, pcbopt, m) int optname; struct mbuf **pcbopt; register struct mbuf *m; { register int cnt, optlen; register u_char *cp; u_char opt; /* turn off any old options */ if (*pcbopt) (void)m_free(*pcbopt); *pcbopt = 0; if (m == (struct mbuf *)0 || m->m_len == 0) { /* * Only turning off any previous options. */ if (m) (void)m_free(m); return (0); } if (m->m_len % sizeof(int32_t)) goto bad; /* * IP first-hop destination address will be stored before * actual options; move other options back * and clear it when none present. */ if (m->m_data + m->m_len + sizeof(struct in_addr) >= &m->m_dat[MLEN]) goto bad; cnt = m->m_len; m->m_len += sizeof(struct in_addr); cp = mtod(m, u_char *) + sizeof(struct in_addr); ovbcopy(mtod(m, caddr_t), (caddr_t)cp, (unsigned)cnt); bzero(mtod(m, caddr_t), sizeof(struct in_addr)); for (; cnt > 0; cnt -= optlen, cp += optlen) { opt = cp[IPOPT_OPTVAL]; if (opt == IPOPT_EOL) break; if (opt == IPOPT_NOP) optlen = 1; else { if (cnt < IPOPT_OLEN + sizeof(*cp)) goto bad; optlen = cp[IPOPT_OLEN]; if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) goto bad; } switch (opt) { default: break; case IPOPT_LSRR: case IPOPT_SSRR: /* * user process specifies route as: * ->A->B->C->D * D must be our final destination (but we can't * check that since we may not have connected yet). * A is first hop destination, which doesn't appear in * actual IP option, but is stored before the options. */ if (optlen < IPOPT_MINOFF - 1 + sizeof(struct in_addr)) goto bad; m->m_len -= sizeof(struct in_addr); cnt -= sizeof(struct in_addr); optlen -= sizeof(struct in_addr); cp[IPOPT_OLEN] = optlen; /* * Move first hop before start of options. */ bcopy((caddr_t)&cp[IPOPT_OFFSET+1], mtod(m, caddr_t), sizeof(struct in_addr)); /* * Then copy rest of options back * to close up the deleted entry. */ ovbcopy((caddr_t)(&cp[IPOPT_OFFSET+1] + sizeof(struct in_addr)), (caddr_t)&cp[IPOPT_OFFSET+1], (unsigned)cnt + sizeof(struct in_addr)); break; } } if (m->m_len > MAX_IPOPTLEN + sizeof(struct in_addr)) goto bad; *pcbopt = m; return (0); bad: (void)m_free(m); return (EINVAL); } /* * XXX * The whole multicast option thing needs to be re-thought. * Several of these options are equally applicable to non-multicast * transmission, and one (IP_MULTICAST_TTL) totally duplicates a * standard option (IP_TTL). */ /* * following RFC1724 section 3.3, 0.0.0.0/8 is interpreted as interface index. */ static struct ifnet * ip_multicast_if(a, ifindexp) struct in_addr *a; int *ifindexp; { int ifindex; struct ifnet *ifp; if (ifindexp) *ifindexp = 0; if (ntohl(a->s_addr) >> 24 == 0) { ifindex = ntohl(a->s_addr) & 0xffffff; if (ifindex < 0 || if_index < ifindex) return NULL; ifp = ifnet_byindex(ifindex); if (ifindexp) *ifindexp = ifindex; } else { INADDR_TO_IFP(*a, ifp); } return ifp; } /* * Set the IP multicast options in response to user setsockopt(). */ static int ip_setmoptions(sopt, imop) struct sockopt *sopt; struct ip_moptions **imop; { int error = 0; int i; struct in_addr addr; struct ip_mreq mreq; struct ifnet *ifp; struct ip_moptions *imo = *imop; struct route ro; struct sockaddr_in *dst; int ifindex; int s; if (imo == NULL) { /* * No multicast option buffer attached to the pcb; * allocate one and initialize to default values. */ imo = (struct ip_moptions*)malloc(sizeof(*imo), M_IPMOPTS, M_WAITOK); if (imo == NULL) return (ENOBUFS); *imop = imo; imo->imo_multicast_ifp = NULL; imo->imo_multicast_addr.s_addr = INADDR_ANY; imo->imo_multicast_vif = -1; imo->imo_multicast_ttl = IP_DEFAULT_MULTICAST_TTL; imo->imo_multicast_loop = IP_DEFAULT_MULTICAST_LOOP; imo->imo_num_memberships = 0; } switch (sopt->sopt_name) { /* store an index number for the vif you wanna use in the send */ case IP_MULTICAST_VIF: if (legal_vif_num == 0) { error = EOPNOTSUPP; break; } error = sooptcopyin(sopt, &i, sizeof i, sizeof i); if (error) break; if (!legal_vif_num(i) && (i != -1)) { error = EINVAL; break; } imo->imo_multicast_vif = i; break; case IP_MULTICAST_IF: /* * Select the interface for outgoing multicast packets. */ error = sooptcopyin(sopt, &addr, sizeof addr, sizeof addr); if (error) break; /* * INADDR_ANY is used to remove a previous selection. * When no interface is selected, a default one is * chosen every time a multicast packet is sent. */ if (addr.s_addr == INADDR_ANY) { imo->imo_multicast_ifp = NULL; break; } /* * The selected interface is identified by its local * IP address. Find the interface and confirm that * it supports multicasting. */ s = splimp(); ifp = ip_multicast_if(&addr, &ifindex); if (ifp == NULL || (ifp->if_flags & IFF_MULTICAST) == 0) { splx(s); error = EADDRNOTAVAIL; break; } imo->imo_multicast_ifp = ifp; if (ifindex) imo->imo_multicast_addr = addr; else imo->imo_multicast_addr.s_addr = INADDR_ANY; splx(s); break; case IP_MULTICAST_TTL: /* * Set the IP time-to-live for outgoing multicast packets. * The original multicast API required a char argument, * which is inconsistent with the rest of the socket API. * We allow either a char or an int. */ if (sopt->sopt_valsize == 1) { u_char ttl; error = sooptcopyin(sopt, &ttl, 1, 1); if (error) break; imo->imo_multicast_ttl = ttl; } else { u_int ttl; error = sooptcopyin(sopt, &ttl, sizeof ttl, sizeof ttl); if (error) break; if (ttl > 255) error = EINVAL; else imo->imo_multicast_ttl = ttl; } break; case IP_MULTICAST_LOOP: /* * Set the loopback flag for outgoing multicast packets. * Must be zero or one. The original multicast API required a * char argument, which is inconsistent with the rest * of the socket API. We allow either a char or an int. */ if (sopt->sopt_valsize == 1) { u_char loop; error = sooptcopyin(sopt, &loop, 1, 1); if (error) break; imo->imo_multicast_loop = !!loop; } else { u_int loop; error = sooptcopyin(sopt, &loop, sizeof loop, sizeof loop); if (error) break; imo->imo_multicast_loop = !!loop; } break; case IP_ADD_MEMBERSHIP: /* * Add a multicast group membership. * Group must be a valid IP multicast address. */ error = sooptcopyin(sopt, &mreq, sizeof mreq, sizeof mreq); if (error) break; if (!IN_MULTICAST(ntohl(mreq.imr_multiaddr.s_addr))) { error = EINVAL; break; } s = splimp(); /* * If no interface address was provided, use the interface of * the route to the given multicast address. */ if (mreq.imr_interface.s_addr == INADDR_ANY) { bzero((caddr_t)&ro, sizeof(ro)); dst = (struct sockaddr_in *)&ro.ro_dst; dst->sin_len = sizeof(*dst); dst->sin_family = AF_INET; dst->sin_addr = mreq.imr_multiaddr; rtalloc(&ro); if (ro.ro_rt == NULL) { error = EADDRNOTAVAIL; splx(s); break; } ifp = ro.ro_rt->rt_ifp; rtfree(ro.ro_rt); } else { ifp = ip_multicast_if(&mreq.imr_interface, NULL); } /* * See if we found an interface, and confirm that it * supports multicast. */ if (ifp == NULL || (ifp->if_flags & IFF_MULTICAST) == 0) { error = EADDRNOTAVAIL; splx(s); break; } /* * See if the membership already exists or if all the * membership slots are full. */ for (i = 0; i < imo->imo_num_memberships; ++i) { if (imo->imo_membership[i]->inm_ifp == ifp && imo->imo_membership[i]->inm_addr.s_addr == mreq.imr_multiaddr.s_addr) break; } if (i < imo->imo_num_memberships) { error = EADDRINUSE; splx(s); break; } if (i == IP_MAX_MEMBERSHIPS) { error = ETOOMANYREFS; splx(s); break; } /* * Everything looks good; add a new record to the multicast * address list for the given interface. */ if ((imo->imo_membership[i] = in_addmulti(&mreq.imr_multiaddr, ifp)) == NULL) { error = ENOBUFS; splx(s); break; } ++imo->imo_num_memberships; splx(s); break; case IP_DROP_MEMBERSHIP: /* * Drop a multicast group membership. * Group must be a valid IP multicast address. */ error = sooptcopyin(sopt, &mreq, sizeof mreq, sizeof mreq); if (error) break; if (!IN_MULTICAST(ntohl(mreq.imr_multiaddr.s_addr))) { error = EINVAL; break; } s = splimp(); /* * If an interface address was specified, get a pointer * to its ifnet structure. */ if (mreq.imr_interface.s_addr == INADDR_ANY) ifp = NULL; else { ifp = ip_multicast_if(&mreq.imr_interface, NULL); if (ifp == NULL) { error = EADDRNOTAVAIL; splx(s); break; } } /* * Find the membership in the membership array. */ for (i = 0; i < imo->imo_num_memberships; ++i) { if ((ifp == NULL || imo->imo_membership[i]->inm_ifp == ifp) && imo->imo_membership[i]->inm_addr.s_addr == mreq.imr_multiaddr.s_addr) break; } if (i == imo->imo_num_memberships) { error = EADDRNOTAVAIL; splx(s); break; } /* * Give up the multicast address record to which the * membership points. */ in_delmulti(imo->imo_membership[i]); /* * Remove the gap in the membership array. */ for (++i; i < imo->imo_num_memberships; ++i) imo->imo_membership[i-1] = imo->imo_membership[i]; --imo->imo_num_memberships; splx(s); break; default: error = EOPNOTSUPP; break; } /* * If all options have default values, no need to keep the mbuf. */ if (imo->imo_multicast_ifp == NULL && imo->imo_multicast_vif == -1 && imo->imo_multicast_ttl == IP_DEFAULT_MULTICAST_TTL && imo->imo_multicast_loop == IP_DEFAULT_MULTICAST_LOOP && imo->imo_num_memberships == 0) { free(*imop, M_IPMOPTS); *imop = NULL; } return (error); } /* * Return the IP multicast options in response to user getsockopt(). */ static int ip_getmoptions(sopt, imo) struct sockopt *sopt; register struct ip_moptions *imo; { struct in_addr addr; struct in_ifaddr *ia; int error, optval; u_char coptval; error = 0; switch (sopt->sopt_name) { case IP_MULTICAST_VIF: if (imo != NULL) optval = imo->imo_multicast_vif; else optval = -1; error = sooptcopyout(sopt, &optval, sizeof optval); break; case IP_MULTICAST_IF: if (imo == NULL || imo->imo_multicast_ifp == NULL) addr.s_addr = INADDR_ANY; else if (imo->imo_multicast_addr.s_addr) { /* return the value user has set */ addr = imo->imo_multicast_addr; } else { IFP_TO_IA(imo->imo_multicast_ifp, ia); addr.s_addr = (ia == NULL) ? INADDR_ANY : IA_SIN(ia)->sin_addr.s_addr; } error = sooptcopyout(sopt, &addr, sizeof addr); break; case IP_MULTICAST_TTL: if (imo == 0) optval = coptval = IP_DEFAULT_MULTICAST_TTL; else optval = coptval = imo->imo_multicast_ttl; if (sopt->sopt_valsize == 1) error = sooptcopyout(sopt, &coptval, 1); else error = sooptcopyout(sopt, &optval, sizeof optval); break; case IP_MULTICAST_LOOP: if (imo == 0) optval = coptval = IP_DEFAULT_MULTICAST_LOOP; else optval = coptval = imo->imo_multicast_loop; if (sopt->sopt_valsize == 1) error = sooptcopyout(sopt, &coptval, 1); else error = sooptcopyout(sopt, &optval, sizeof optval); break; default: error = ENOPROTOOPT; break; } return (error); } /* * Discard the IP multicast options. */ void ip_freemoptions(imo) register struct ip_moptions *imo; { register int i; if (imo != NULL) { for (i = 0; i < imo->imo_num_memberships; ++i) in_delmulti(imo->imo_membership[i]); free(imo, M_IPMOPTS); } } /* * Routine called from ip_output() to loop back a copy of an IP multicast * packet to the input queue of a specified interface. Note that this * calls the output routine of the loopback "driver", but with an interface * pointer that might NOT be a loopback interface -- evil, but easier than * replicating that code here. */ static void ip_mloopback(ifp, m, dst, hlen) struct ifnet *ifp; register struct mbuf *m; register struct sockaddr_in *dst; int hlen; { register struct ip *ip; struct mbuf *copym; copym = m_copy(m, 0, M_COPYALL); if (copym != NULL && (copym->m_flags & M_EXT || copym->m_len < hlen)) copym = m_pullup(copym, hlen); if (copym != NULL) { /* * We don't bother to fragment if the IP length is greater * than the interface's MTU. Can this possibly matter? */ ip = mtod(copym, struct ip *); HTONS(ip->ip_len); HTONS(ip->ip_off); ip->ip_sum = 0; if (ip->ip_vhl == IP_VHL_BORING) { ip->ip_sum = in_cksum_hdr(ip); } else { ip->ip_sum = in_cksum(copym, hlen); } /* * NB: * It's not clear whether there are any lingering * reentrancy problems in other areas which might * be exposed by using ip_input directly (in * particular, everything which modifies the packet * in-place). Yet another option is using the * protosw directly to deliver the looped back * packet. For the moment, we'll err on the side * of safety by using if_simloop(). */ #if 1 /* XXX */ if (dst->sin_family != AF_INET) { printf("ip_mloopback: bad address family %d\n", dst->sin_family); dst->sin_family = AF_INET; } #endif #ifdef notdef copym->m_pkthdr.rcvif = ifp; ip_input(copym); #else /* if the checksum hasn't been computed, mark it as valid */ if (copym->m_pkthdr.csum_flags & CSUM_DELAY_DATA) { copym->m_pkthdr.csum_flags |= CSUM_DATA_VALID | CSUM_PSEUDO_HDR; copym->m_pkthdr.csum_data = 0xffff; } if_simloop(ifp, copym, dst->sin_family, 0); #endif } }