Please note that diffs are not public domain; they are subject to the copyright notices on the relevant files. =================================================================== RCS file: /ftp/cvs/cvsroot/src/sys/netinet/ip_input.c,v rcsdiff: /ftp/cvs/cvsroot/src/sys/netinet/ip_input.c,v: warning: Unknown phrases like `commitid ...;' are present. retrieving revision 1.47 retrieving revision 1.114.4.9 diff -u -p -r1.47 -r1.114.4.9 --- src/sys/netinet/ip_input.c 1997/02/25 08:35:42 1.47 +++ src/sys/netinet/ip_input.c 2002/11/10 19:58:05 1.114.4.9 @@ -1,4 +1,70 @@ -/* $NetBSD: ip_input.c,v 1.47 1997/02/25 08:35:42 cjs Exp $ */ +/* $NetBSD: ip_input.c,v 1.114.4.9 2002/11/10 19:58:05 itojun Exp $ */ + +/* + * Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. + * 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. Neither the name of the project 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 PROJECT 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 PROJECT OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +/*- + * Copyright (c) 1998 The NetBSD Foundation, Inc. + * All rights reserved. + * + * This code is derived from software contributed to The NetBSD Foundation + * by Public Access Networks Corporation ("Panix"). It was developed under + * contract to Panix by Eric Haszlakiewicz and Thor Lancelot Simon. + * + * 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 NetBSD + * Foundation, Inc. and its contributors. + * 4. Neither the name of The NetBSD Foundation 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 NETBSD FOUNDATION, INC. 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 FOUNDATION OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ /* * Copyright (c) 1982, 1986, 1988, 1993 @@ -35,6 +101,11 @@ * @(#)ip_input.c 8.2 (Berkeley) 1/4/94 */ +#include "opt_gateway.h" +#include "opt_pfil_hooks.h" +#include "opt_ipsec.h" +#include "opt_mrouting.h" + #include #include #include @@ -47,6 +118,7 @@ #include #include #include +#include #include #include @@ -63,9 +135,21 @@ #include #include #include +/* just for gif_ttl */ +#include +#include "gif.h" +#include +#include "gre.h" -/* XXX should really put this in libkern.h */ -#define offsetof(type, member) ((size_t)(&((type *)0)->member)) +#ifdef MROUTING +#include +#endif + +#ifdef IPSEC +#include +#include +#include +#endif #ifndef IPFORWARDING #ifdef GATEWAY @@ -81,8 +165,15 @@ #define IPFORWSRCRT 1 /* forward source-routed packets */ #endif #ifndef IPALLOWSRCRT -#define IPALLOWSRCRT 0 /* reject all source-routed packets */ +#define IPALLOWSRCRT 1 /* allow source-routed packets */ #endif +#ifndef IPMTUDISC +#define IPMTUDISC 0 +#endif +#ifndef IPMTUDISCTIMEOUT +#define IPMTUDISCTIMEOUT (10 * 60) /* as per RFC 1191 */ +#endif + /* * Note: DIRECTED_BROADCAST is handled this way so that previous * configuration using this option will Just Work. @@ -100,16 +191,78 @@ int ip_defttl = IPDEFTTL; int ip_forwsrcrt = IPFORWSRCRT; int ip_directedbcast = IPDIRECTEDBCAST; int ip_allowsrcrt = IPALLOWSRCRT; +int ip_mtudisc = IPMTUDISC; +u_int ip_mtudisc_timeout = IPMTUDISCTIMEOUT; #ifdef DIAGNOSTIC int ipprintfs = 0; #endif +struct rttimer_queue *ip_mtudisc_timeout_q = NULL; + extern struct domain inetdomain; -extern struct protosw inetsw[]; -u_char ip_protox[IPPROTO_MAX]; int ipqmaxlen = IFQ_MAXLEN; struct in_ifaddrhead in_ifaddr; +struct in_ifaddrhashhead *in_ifaddrhashtbl; struct ifqueue ipintrq; +struct ipstat ipstat; +u_int16_t ip_id; + +struct ipqhead ipq; +int ipq_locked; +int ip_nfragpackets = 0; +int ip_maxfragpackets = 200; + +static __inline int ipq_lock_try __P((void)); +static __inline void ipq_unlock __P((void)); + +static __inline int +ipq_lock_try() +{ + int s; + + s = splimp(); + if (ipq_locked) { + splx(s); + return (0); + } + ipq_locked = 1; + splx(s); + return (1); +} + +static __inline void +ipq_unlock() +{ + int s; + + s = splimp(); + ipq_locked = 0; + splx(s); +} + +#ifdef DIAGNOSTIC +#define IPQ_LOCK() \ +do { \ + if (ipq_lock_try() == 0) { \ + printf("%s:%d: ipq already locked\n", __FILE__, __LINE__); \ + panic("ipq_lock"); \ + } \ +} while (0) +#define IPQ_LOCK_CHECK() \ +do { \ + if (ipq_locked == 0) { \ + printf("%s:%d: ipq lock not held\n", __FILE__, __LINE__); \ + panic("ipq lock check"); \ + } \ +} while (0) +#else +#define IPQ_LOCK() (void) ipq_lock_try() +#define IPQ_LOCK_CHECK() /* nothing */ +#endif + +#define IPQ_UNLOCK() ipq_unlock() + +struct pool ipqent_pool; /* * We need to save the IP options in case a protocol wants to respond @@ -135,8 +288,11 @@ static void save_rte __P((u_char *, stru void ip_init() { - register struct protosw *pr; - register int i; + struct protosw *pr; + int i; + + pool_init(&ipqent_pool, sizeof(struct ipqent), 0, 0, 0, "ipqepl", + 0, NULL, NULL, M_IPQ); pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); if (pr == 0) @@ -152,44 +308,70 @@ ip_init() ip_id = time.tv_sec & 0xffff; ipintrq.ifq_maxlen = ipqmaxlen; TAILQ_INIT(&in_ifaddr); + in_ifaddrhashtbl = + hashinit(IN_IFADDR_HASH_SIZE, M_IFADDR, M_WAITOK, &in_ifaddrhash); + ip_mtudisc_timeout_q = rt_timer_queue_create(ip_mtudisc_timeout); +#ifdef GATEWAY + ipflow_init(); +#endif } struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; struct route ipforward_rt; /* + * IP software interrupt routine + */ +void +ipintr() +{ + int s; + struct mbuf *m; + + while (1) { + s = splimp(); + IF_DEQUEUE(&ipintrq, m); + splx(s); + if (m == 0) + return; + ip_input(m); + } +} + +/* * Ip input routine. Checksum and byte swap header. If fragmented * try to reassemble. Process options. Pass to next level. */ void -ipintr() +ip_input(struct mbuf *m) { - register struct ip *ip = NULL; - register struct mbuf *m; - register struct ipq *fp; - register struct in_ifaddr *ia; + struct ip *ip = NULL; + struct ipq *fp; + struct in_ifaddr *ia; + struct ifaddr *ifa; struct ipqent *ipqe; - int hlen = 0, mff, len, s; + int hlen = 0, mff, len; + int downmatch; #ifdef PFIL_HOOKS struct packet_filter_hook *pfh; struct mbuf *m0; int rv; #endif /* PFIL_HOOKS */ -next: - /* - * Get next datagram off input queue and get IP header - * in first mbuf. - */ - s = splimp(); - IF_DEQUEUE(&ipintrq, m); - splx(s); - if (m == 0) - return; #ifdef DIAGNOSTIC if ((m->m_flags & M_PKTHDR) == 0) panic("ipintr no HDR"); #endif +#ifdef IPSEC + /* + * should the inner packet be considered authentic? + * see comment in ah4_input(). + */ + if (m) { + m->m_flags &= ~M_AUTHIPHDR; + m->m_flags &= ~M_AUTHIPDGM; + } +#endif /* * If no IP addresses have been set yet but the interfaces * are receiving, can't do anything with incoming packets yet. @@ -200,7 +382,7 @@ next: if (m->m_len < sizeof (struct ip) && (m = m_pullup(m, sizeof (struct ip))) == 0) { ipstat.ips_toosmall++; - goto next; + return; } ip = mtod(m, struct ip *); if (ip->ip_v != IPVERSION) { @@ -215,11 +397,21 @@ next: if (hlen > m->m_len) { if ((m = m_pullup(m, hlen)) == 0) { ipstat.ips_badhlen++; - goto next; + return; } ip = mtod(m, struct ip *); } - if ((ip->ip_sum = in_cksum(m, hlen)) != 0) { + + /* + * RFC1122: packets with a multicast source address are + * not allowed. + */ + if (IN_MULTICAST(ip->ip_src.s_addr)) { + /* XXX stat */ + goto bad; + } + + if (in_cksum(m, hlen) != 0) { ipstat.ips_badsum++; goto bad; } @@ -228,11 +420,18 @@ next: * Convert fields to host representation. */ NTOHS(ip->ip_len); - NTOHS(ip->ip_id); NTOHS(ip->ip_off); len = ip->ip_len; /* + * Check for additional length bogosity + */ + if (len < hlen) { + ipstat.ips_badlen++; + goto bad; + } + + /* * 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. @@ -250,18 +449,49 @@ next: m_adj(m, len - m->m_pkthdr.len); } +#ifdef IPSEC + /* ipflow (IP fast fowarding) is not compatible with IPsec. */ + m->m_flags &= ~M_CANFASTFWD; +#else + /* + * Assume that we can create a fast-forward IP flow entry + * based on this packet. + */ + m->m_flags |= M_CANFASTFWD; +#endif + #ifdef PFIL_HOOKS +#ifdef IPSEC + /* + * let ipfilter look at packet on the wire, + * not the decapsulated packet. + */ + if (ipsec_gethist(m, NULL)) + goto nofilt; +#endif /* - * Run through list of hooks for input packets. + * 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; - for (pfh = pfil_hook_get(PFIL_IN); pfh; pfh = pfh->pfil_link.le_next) + pfh = pfil_hook_get(PFIL_IN, &inetsw[ip_protox[IPPROTO_IP]].pr_pfh); + for (; pfh; pfh = pfh->pfil_link.tqe_next) if (pfh->pfil_func) { - rv = pfh->pfil_func(ip, hlen, m->m_pkthdr.rcvif, 0, &m0); - ip = mtod(m = m0, struct ip *); + rv = pfh->pfil_func(ip, hlen, + m->m_pkthdr.rcvif, 0, &m0); if (rv) - goto next; + return; + m = m0; + if (m == NULL) + return; + ip = mtod(m, struct ip *); } +#ifdef IPSEC +nofilt:; +#endif #endif /* PFIL_HOOKS */ /* @@ -272,17 +502,33 @@ next: */ ip_nhops = 0; /* for source routed packets */ if (hlen > sizeof (struct ip) && ip_dooptions(m)) - goto next; + return; /* * Check our list of addresses, to see if the packet is for us. - */ - for (ia = in_ifaddr.tqh_first; ia; ia = ia->ia_list.tqe_next) { - if (in_hosteq(ip->ip_dst, ia->ia_addr.sin_addr)) - goto ours; - if (((ip_directedbcast == 0) || (ip_directedbcast && - ia->ia_ifp == m->m_pkthdr.rcvif)) && - (ia->ia_ifp->if_flags & IFF_BROADCAST)) { + * + * Traditional 4.4BSD did not consult IFF_UP at all. + * The behavior here is to treat addresses on !IFF_UP interface + * as not mine. + */ + downmatch = 0; + for (ia = IN_IFADDR_HASH(ip->ip_dst.s_addr).lh_first; + ia != NULL; + ia = ia->ia_hash.le_next) { + if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { + if ((ia->ia_ifp->if_flags & IFF_UP) != 0) + break; + else + downmatch++; + } + } + if (ia != NULL) + goto ours; + if (m->m_pkthdr.rcvif->if_flags & IFF_BROADCAST) { + for (ifa = m->m_pkthdr.rcvif->if_addrlist.tqh_first; + ifa != NULL; ifa = ifa->ifa_list.tqe_next) { + if (ifa->ifa_addr->sa_family != AF_INET) continue; + ia = ifatoia(ifa); if (in_hosteq(ip->ip_dst, ia->ia_broadaddr.sin_addr) || in_hosteq(ip->ip_dst, ia->ia_netbroadcast) || /* @@ -292,6 +538,12 @@ next: ip->ip_dst.s_addr == ia->ia_subnet || ip->ip_dst.s_addr == ia->ia_net) goto ours; + /* + * An interface with IP address zero accepts + * all packets that arrive on that interface. + */ + if (in_nullhost(ia->ia_addr.sin_addr)) + goto ours; } } if (IN_MULTICAST(ip->ip_dst.s_addr)) { @@ -302,7 +554,7 @@ next: if (m->m_flags & M_EXT) { if ((m = m_pullup(m, hlen)) == 0) { ipstat.ips_toosmall++; - goto next; + return; } ip = mtod(m, struct ip *); } @@ -320,13 +572,11 @@ next: * as expected when ip_mforward() is called from * ip_output().) */ - ip->ip_id = htons(ip->ip_id); if (ip_mforward(m, m->m_pkthdr.rcvif) != 0) { ipstat.ips_cantforward++; m_freem(m); - goto next; + return; } - ip->ip_id = ntohs(ip->ip_id); /* * The process-level routing demon needs to receive @@ -346,7 +596,7 @@ next: if (inm == NULL) { ipstat.ips_cantforward++; m_freem(m); - goto next; + return; } goto ours; } @@ -360,9 +610,28 @@ next: if (ipforwarding == 0) { ipstat.ips_cantforward++; m_freem(m); - } else + } else { + /* + * If ip_dst matched any of my address on !IFF_UP interface, + * and there's no IFF_UP interface that matches ip_dst, + * send icmp unreach. Forwarding it will result in in-kernel + * forwarding loop till TTL goes to 0. + */ + if (downmatch) { + icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, 0, 0); + ipstat.ips_cantforward++; + return; + } +#ifdef IPSEC + if (ipsec4_in_reject(m, NULL)) { + ipsecstat.in_polvio++; + goto bad; + } +#endif + ip_forward(m, 0); - goto next; + } + return; ours: /* @@ -373,17 +642,11 @@ ours: * but it's not worth the time; just let them time out.) */ if (ip->ip_off & ~(IP_DF|IP_RF)) { - if (m->m_flags & M_EXT) { /* XXX */ - if ((m = m_pullup(m, sizeof (struct ip))) == 0) { - ipstat.ips_toosmall++; - goto next; - } - ip = mtod(m, struct ip *); - } /* * Look for queue of fragments * of this datagram. */ + IPQ_LOCK(); for (fp = ipq.lh_first; fp != NULL; fp = fp->ipq_q.le_next) if (ip->ip_id == fp->ipq_id && in_hosteq(ip->ip_src, fp->ipq_src) && @@ -407,6 +670,7 @@ found: */ if (ip->ip_len == 0 || (ip->ip_len & 0x7) != 0) { ipstat.ips_badfrags++; + IPQ_UNLOCK(); goto bad; } } @@ -419,34 +683,58 @@ found: */ if (mff || ip->ip_off) { ipstat.ips_fragments++; - MALLOC(ipqe, struct ipqent *, sizeof (struct ipqent), - M_IPQ, M_NOWAIT); + ipqe = pool_get(&ipqent_pool, PR_NOWAIT); if (ipqe == NULL) { ipstat.ips_rcvmemdrop++; + IPQ_UNLOCK(); goto bad; } ipqe->ipqe_mff = mff; + ipqe->ipqe_m = m; ipqe->ipqe_ip = ip; - ip = ip_reass(ipqe, fp); - if (ip == 0) - goto next; + m = ip_reass(ipqe, fp); + if (m == 0) { + IPQ_UNLOCK(); + return; + } ipstat.ips_reassembled++; - m = dtom(ip); + ip = mtod(m, struct ip *); + hlen = ip->ip_hl << 2; + ip->ip_len += hlen; } else if (fp) ip_freef(fp); - } else - ip->ip_len -= hlen; + IPQ_UNLOCK(); + } + +#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. */ +#if IFA_STATS + ia->ia_ifa.ifa_data.ifad_inbytes += ip->ip_len; +#endif ipstat.ips_delivered++; - (*inetsw[ip_protox[ip->ip_p]].pr_input)(m, hlen); - goto next; + { + int off = hlen, nh = ip->ip_p; + + (*inetsw[ip_protox[nh]].pr_input)(m, off, nh); + return; + } bad: m_freem(m); - goto next; } /* @@ -455,18 +743,20 @@ bad: * reassembly of this datagram already exists, then it * is given as fp; otherwise have to make a chain. */ -struct ip * +struct mbuf * ip_reass(ipqe, fp) - register struct ipqent *ipqe; - register struct ipq *fp; + struct ipqent *ipqe; + struct ipq *fp; { - register struct mbuf *m = dtom(ipqe->ipqe_ip); - register struct ipqent *nq, *p, *q; + struct mbuf *m = ipqe->ipqe_m; + struct ipqent *nq, *p, *q; struct ip *ip; struct mbuf *t; int hlen = ipqe->ipqe_ip->ip_hl << 2; int i, next; + IPQ_LOCK_CHECK(); + /* * Presence of header sizes in mbufs * would confuse code below. @@ -478,9 +768,21 @@ ip_reass(ipqe, fp) * If first fragment to arrive, create a reassembly queue. */ if (fp == 0) { - if ((t = m_get(M_DONTWAIT, MT_FTABLE)) == NULL) + /* + * 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) + ; + else if (ip_nfragpackets >= ip_maxfragpackets) + goto dropfrag; + ip_nfragpackets++; + MALLOC(fp, struct ipq *, sizeof (struct ipq), + M_FTABLE, M_NOWAIT); + if (fp == NULL) goto dropfrag; - fp = mtod(t, struct ipq *); LIST_INSERT_HEAD(&ipq, fp, ipq_q); fp->ipq_ttl = IPFRAGTTL; fp->ipq_p = ipqe->ipqe_ip->ip_p; @@ -511,7 +813,7 @@ ip_reass(ipqe, fp) if (i > 0) { if (i >= ipqe->ipqe_ip->ip_len) goto dropfrag; - m_adj(dtom(ipqe->ipqe_ip), i); + m_adj(ipqe->ipqe_m, i); ipqe->ipqe_ip->ip_off += i; ipqe->ipqe_ip->ip_len -= i; } @@ -528,13 +830,13 @@ ip_reass(ipqe, fp) if (i < q->ipqe_ip->ip_len) { q->ipqe_ip->ip_len -= i; q->ipqe_ip->ip_off += i; - m_adj(dtom(q->ipqe_ip), i); + m_adj(q->ipqe_m, i); break; } nq = q->ipqe_q.le_next; - m_freem(dtom(q->ipqe_ip)); + m_freem(q->ipqe_m); LIST_REMOVE(q, ipqe_q); - FREE(q, M_IPQ); + pool_put(&ipqent_pool, q); } insert: @@ -568,16 +870,16 @@ insert: ip_freef(fp); return (0); } - m = dtom(q->ipqe_ip); + m = q->ipqe_m; t = m->m_next; m->m_next = 0; m_cat(m, t); nq = q->ipqe_q.le_next; - FREE(q, M_IPQ); + pool_put(&ipqent_pool, q); for (q = nq; q != NULL; q = nq) { - t = dtom(q->ipqe_ip); + t = q->ipqe_m; nq = q->ipqe_q.le_next; - FREE(q, M_IPQ); + pool_put(&ipqent_pool, q); m_cat(m, t); } @@ -591,22 +893,23 @@ insert: ip->ip_src = fp->ipq_src; ip->ip_dst = fp->ipq_dst; LIST_REMOVE(fp, ipq_q); - (void) m_free(dtom(fp)); + FREE(fp, M_FTABLE); + ip_nfragpackets--; m->m_len += (ip->ip_hl << 2); m->m_data -= (ip->ip_hl << 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; m; m = m->m_next) - plen += m->m_len; - t->m_pkthdr.len = plen; + int plen = 0; + for (t = m; t; t = t->m_next) + plen += t->m_len; + m->m_pkthdr.len = plen; } - return (ip); + return (m); dropfrag: ipstat.ips_fragdropped++; m_freem(m); - FREE(ipqe, M_IPQ); + pool_put(&ipqent_pool, ipqe); return (0); } @@ -618,16 +921,19 @@ void ip_freef(fp) struct ipq *fp; { - register struct ipqent *q, *p; + struct ipqent *q, *p; + + IPQ_LOCK_CHECK(); for (q = fp->ipq_fragq.lh_first; q != NULL; q = p) { p = q->ipqe_q.le_next; - m_freem(dtom(q->ipqe_ip)); + m_freem(q->ipqe_m); LIST_REMOVE(q, ipqe_q); - FREE(q, M_IPQ); + pool_put(&ipqent_pool, q); } LIST_REMOVE(fp, ipq_q); - (void) m_free(dtom(fp)); + FREE(fp, M_FTABLE); + ip_nfragpackets--; } /* @@ -638,9 +944,10 @@ ip_freef(fp) void ip_slowtimo() { - register struct ipq *fp, *nfp; + struct ipq *fp, *nfp; int s = splsoftnet(); + IPQ_LOCK(); for (fp = ipq.lh_first; fp != NULL; fp = nfp) { nfp = fp->ipq_q.le_next; if (--fp->ipq_ttl == 0) { @@ -648,6 +955,21 @@ ip_slowtimo() ip_freef(fp); } } + /* + * 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. + */ + if (ip_maxfragpackets < 0) + ; + else { + while (ip_nfragpackets > ip_maxfragpackets && ipq.lh_first) + ip_freef(ipq.lh_first); + } + IPQ_UNLOCK(); +#ifdef GATEWAY + ipflow_slowtimo(); +#endif splx(s); } @@ -658,10 +980,19 @@ void ip_drain() { + /* + * We may be called from a device's interrupt context. If + * the ipq is already busy, just bail out now. + */ + if (ipq_lock_try() == 0) + return; + while (ipq.lh_first != NULL) { ipstat.ips_fragdropped++; ip_freef(ipq.lh_first); } + + IPQ_UNLOCK(); } /* @@ -675,12 +1006,12 @@ int ip_dooptions(m) struct mbuf *m; { - register struct ip *ip = mtod(m, struct ip *); - register u_char *cp; - register struct ip_timestamp *ipt; - register struct in_ifaddr *ia; + struct ip *ip = mtod(m, struct ip *); + u_char *cp, *cp0; + struct ip_timestamp *ipt; + struct in_ifaddr *ia; int opt, optlen, cnt, off, code, type = ICMP_PARAMPROB, forward = 0; - struct in_addr *sin, dst; + struct in_addr dst; n_time ntime; dst = ip->ip_dst; @@ -693,8 +1024,12 @@ ip_dooptions(m) 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 <= 0 || optlen > cnt) { + if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } @@ -720,6 +1055,10 @@ ip_dooptions(m) code = ICMP_UNREACH_NET_PROHIB; goto bad; } + 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; @@ -739,7 +1078,7 @@ ip_dooptions(m) break; } off--; /* 0 origin */ - if (off > optlen - sizeof(struct in_addr)) { + if ((off + sizeof(struct in_addr)) > optlen) { /* * End of source route. Should be for us. */ @@ -751,11 +1090,9 @@ ip_dooptions(m) */ bcopy((caddr_t)(cp + off), (caddr_t)&ipaddr.sin_addr, sizeof(ipaddr.sin_addr)); - if (opt == IPOPT_SSRR) { -#define INA struct in_ifaddr * -#define SA struct sockaddr * - ia = (INA)ifa_ifwithladdr((SA)&ipaddr); - } else + if (opt == IPOPT_SSRR) + ia = ifatoia(ifa_ifwithaddr(sintosa(&ipaddr))); + else ia = ip_rtaddr(ipaddr.sin_addr); if (ia == 0) { type = ICMP_UNREACH; @@ -773,6 +1110,10 @@ ip_dooptions(m) break; case IPOPT_RR: + 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; @@ -781,7 +1122,7 @@ ip_dooptions(m) * If no space remains, ignore. */ off--; /* 0 origin */ - if (off > optlen - sizeof(struct in_addr)) + if ((off + sizeof(struct in_addr)) > optlen) break; bcopy((caddr_t)(&ip->ip_dst), (caddr_t)&ipaddr.sin_addr, sizeof(ipaddr.sin_addr)); @@ -789,8 +1130,9 @@ ip_dooptions(m) * 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) { + if ((ia = ifatoia(ifa_ifwithaddr(sintosa(&ipaddr)))) + == NULL && + (ia = ip_rtaddr(ipaddr.sin_addr)) == NULL) { type = ICMP_UNREACH; code = ICMP_UNREACH_HOST; goto bad; @@ -803,49 +1145,69 @@ ip_dooptions(m) case IPOPT_TS: code = cp - (u_char *)ip; ipt = (struct ip_timestamp *)cp; - if (ipt->ipt_len < 5) + if (ipt->ipt_len < 4 || ipt->ipt_len > 40) { + code = (u_char *)&ipt->ipt_len - (u_char *)ip; goto bad; + } + if (ipt->ipt_ptr < 5) { + code = (u_char *)&ipt->ipt_ptr - (u_char *)ip; + goto bad; + } if (ipt->ipt_ptr > ipt->ipt_len - sizeof (int32_t)) { - if (++ipt->ipt_oflw == 0) + if (++ipt->ipt_oflw == 0) { + code = (u_char *)&ipt->ipt_ptr - + (u_char *)ip; goto bad; + } break; } - sin = (struct in_addr *)(cp + ipt->ipt_ptr - 1); + cp0 = (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) + if (ipt->ipt_ptr - 1 + sizeof(n_time) + + sizeof(struct in_addr) > ipt->ipt_len) { + code = (u_char *)&ipt->ipt_ptr - + (u_char *)ip; goto bad; + } ipaddr.sin_addr = dst; - ia = (INA)ifaof_ifpforaddr((SA)&ipaddr, - m->m_pkthdr.rcvif); + ia = ifatoia(ifaof_ifpforaddr(sintosa(&ipaddr), + m->m_pkthdr.rcvif)); if (ia == 0) continue; - bcopy((caddr_t)&ia->ia_addr.sin_addr, - (caddr_t)sin, sizeof(struct in_addr)); + bcopy(&ia->ia_addr.sin_addr, + cp0, 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) + if (ipt->ipt_ptr - 1 + sizeof(n_time) + + sizeof(struct in_addr) > ipt->ipt_len) { + code = (u_char *)&ipt->ipt_ptr - + (u_char *)ip; goto bad; - bcopy((caddr_t)sin, (caddr_t)&ipaddr.sin_addr, + } + bcopy(cp0, &ipaddr.sin_addr, sizeof(struct in_addr)); - if (ifa_ifwithaddr((SA)&ipaddr) == 0) + if (ifatoia(ifa_ifwithaddr(sintosa(&ipaddr))) + == NULL) continue; ipt->ipt_ptr += sizeof(struct in_addr); break; default: + /* XXX can't take &ipt->ipt_flg */ + code = (u_char *)&ipt->ipt_ptr - + (u_char *)ip + 1; goto bad; } ntime = iptime(); - bcopy((caddr_t)&ntime, (caddr_t)cp + ipt->ipt_ptr - 1, + cp0 = (u_char *) &ntime; /* XXX grumble, GCC... */ + bcopy(cp0, (caddr_t)cp + ipt->ipt_ptr - 1, sizeof(n_time)); ipt->ipt_ptr += sizeof(n_time); } @@ -861,7 +1223,6 @@ ip_dooptions(m) } return (0); bad: - ip->ip_len -= ip->ip_hl << 2; /* XXX icmp_error adds in hdr length */ icmp_error(m, type, code, 0, 0); ipstat.ips_badoptions++; return (1); @@ -875,7 +1236,7 @@ struct in_ifaddr * ip_rtaddr(dst) struct in_addr dst; { - register struct sockaddr_in *sin; + struct sockaddr_in *sin; sin = satosin(&ipforward_rt.ro_dst); @@ -910,7 +1271,7 @@ save_rte(option, dst) #ifdef DIAGNOSTIC if (ipprintfs) printf("save_rte: olen %d\n", olen); -#endif +#endif /* 0 */ if (olen > sizeof(ip_srcrt) - (1 + sizeof(dst))) return; bcopy((caddr_t)option, (caddr_t)ip_srcrt.srcopt, olen); @@ -926,8 +1287,8 @@ save_rte(option, dst) struct mbuf * ip_srcroute() { - register struct in_addr *p, *q; - register struct mbuf *m; + struct in_addr *p, *q; + struct mbuf *m; if (ip_nhops == 0) return ((struct mbuf *)0); @@ -996,22 +1357,23 @@ ip_srcroute() */ void ip_stripoptions(m, mopt) - register struct mbuf *m; + struct mbuf *m; struct mbuf *mopt; { - register int i; + int i; struct ip *ip = mtod(m, struct ip *); - register caddr_t opts; + caddr_t opts; int olen; - olen = (ip->ip_hl<<2) - sizeof (struct ip); + olen = (ip->ip_hl << 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_hl = sizeof(struct ip) >> 2; + ip->ip_len -= olen; + ip->ip_hl = sizeof (struct ip) >> 2; } int inetctlerrmap[PRC_NCMDS] = { @@ -1042,26 +1404,29 @@ 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; + struct ip *ip = mtod(m, struct ip *); + struct sockaddr_in *sin; + struct rtentry *rt; int error, type = 0, code = 0; struct mbuf *mcopy; n_long dest; struct ifnet *destifp; +#ifdef IPSEC + struct ifnet dummyifp; +#endif dest = 0; #ifdef DIAGNOSTIC if (ipprintfs) - printf("forward: src %x dst %x ttl %x\n", - ip->ip_src.s_addr, ip->ip_dst.s_addr, ip->ip_ttl); + printf("forward: src %2.2x dst %2.2x ttl %x\n", + ntohl(ip->ip_src.s_addr), + ntohl(ip->ip_dst.s_addr), ip->ip_ttl); #endif - if (m->m_flags & M_BCAST || in_canforward(ip->ip_dst) == 0) { + if (m->m_flags & (M_BCAST|M_MCAST) || 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; @@ -1090,8 +1455,11 @@ ip_forward(m, srcrt) /* * Save at most 68 bytes of the packet in case * we need to generate an ICMP message to the src. + * Pullup to avoid sharing mbuf cluster between m and mcopy. */ - mcopy = m_copy(m, 0, imin((int)ip->ip_len, 68)); + mcopy = m_copym(m, 0, imin((int)ip->ip_len, 68), M_DONTWAIT); + if (mcopy) + mcopy = m_pullup(mcopy, ip->ip_hl << 2); /* * If forwarding packet using same interface that it came in on, @@ -1108,20 +1476,28 @@ ip_forward(m, srcrt) if (rt->rt_ifa && (ip->ip_src.s_addr & ifatoia(rt->rt_ifa)->ia_subnetmask) == ifatoia(rt->rt_ifa)->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; + 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 %x\n", code, (u_int32_t)dest); + if (ipprintfs) + printf("redirect (%d) to %x\n", code, + (u_int32_t)dest); #endif } } +#ifdef IPSEC + /* Don't lookup socket in forwading case */ + (void)ipsec_setsocket(m, NULL); +#endif error = ip_output(m, (struct mbuf *)0, &ipforward_rt, (IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), 0); if (error) @@ -1131,8 +1507,13 @@ ip_forward(m, srcrt) if (type) ipstat.ips_redirectsent++; else { - if (mcopy) + if (mcopy) { +#ifdef GATEWAY + if (mcopy->m_flags & M_CANFASTFWD) + ipflow_create(&ipforward_rt, mcopy); +#endif m_freem(mcopy); + } return; } } @@ -1158,8 +1539,61 @@ ip_forward(m, srcrt) 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; + int ipsecerror; + size_t 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; @@ -1173,10 +1607,10 @@ ip_forward(m, srcrt) void ip_savecontrol(inp, mp, ip, m) - register struct inpcb *inp; - register struct mbuf **mp; - register struct ip *ip; - register struct mbuf *m; + struct inpcb *inp; + struct mbuf **mp; + struct ip *ip; + struct mbuf *m; { if (inp->inp_socket->so_options & SO_TIMESTAMP) { @@ -1240,6 +1674,10 @@ ip_sysctl(name, namelen, oldp, oldlenp, void *newp; size_t newlen; { + extern int subnetsarelocal, hostzeroisbroadcast; + + int error, old; + /* All sysctl names at this level are terminal. */ if (namelen != 1) return (ENOTDIR); @@ -1270,6 +1708,110 @@ ip_sysctl(name, namelen, oldp, oldlenp, case IPCTL_ALLOWSRCRT: return (sysctl_int(oldp, oldlenp, newp, newlen, &ip_allowsrcrt)); + case IPCTL_SUBNETSARELOCAL: + return (sysctl_int(oldp, oldlenp, newp, newlen, + &subnetsarelocal)); + case IPCTL_MTUDISC: + error = sysctl_int(oldp, oldlenp, newp, newlen, + &ip_mtudisc); + if (ip_mtudisc != 0 && ip_mtudisc_timeout_q == NULL) { + ip_mtudisc_timeout_q = + rt_timer_queue_create(ip_mtudisc_timeout); + } else if (ip_mtudisc == 0 && ip_mtudisc_timeout_q != NULL) { + rt_timer_queue_destroy(ip_mtudisc_timeout_q, TRUE); + ip_mtudisc_timeout_q = NULL; + } + return error; + case IPCTL_ANONPORTMIN: + old = anonportmin; + error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmin); + if (anonportmin >= anonportmax || anonportmin < 0 + || anonportmin > 65535 +#ifndef IPNOPRIVPORTS + || anonportmin < IPPORT_RESERVED +#endif + ) { + anonportmin = old; + return (EINVAL); + } + return (error); + case IPCTL_ANONPORTMAX: + old = anonportmax; + error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmax); + if (anonportmin >= anonportmax || anonportmax < 0 + || anonportmax > 65535 +#ifndef IPNOPRIVPORTS + || anonportmax < IPPORT_RESERVED +#endif + ) { + anonportmax = old; + return (EINVAL); + } + return (error); + case IPCTL_MTUDISCTIMEOUT: + error = sysctl_int(oldp, oldlenp, newp, newlen, + &ip_mtudisc_timeout); + if (ip_mtudisc_timeout_q != NULL) + rt_timer_queue_change(ip_mtudisc_timeout_q, + ip_mtudisc_timeout); + return (error); +#ifdef GATEWAY + case IPCTL_MAXFLOWS: + { + int s; + + error = sysctl_int(oldp, oldlenp, newp, newlen, + &ip_maxflows); + s = splsoftnet(); + ipflow_reap(0); + splx(s); + return (error); + } +#endif + case IPCTL_HOSTZEROBROADCAST: + return (sysctl_int(oldp, oldlenp, newp, newlen, + &hostzeroisbroadcast)); +#if NGIF > 0 + case IPCTL_GIF_TTL: + return(sysctl_int(oldp, oldlenp, newp, newlen, + &ip_gif_ttl)); +#endif + +#if NGRE > 0 + case IPCTL_GRE_TTL: + return(sysctl_int(oldp, oldlenp, newp, newlen, + &ip_gre_ttl)); +#endif + +#ifndef IPNOPRIVPORTS + case IPCTL_LOWPORTMIN: + old = lowportmin; + error = sysctl_int(oldp, oldlenp, newp, newlen, &lowportmin); + if (lowportmin >= lowportmax + || lowportmin > IPPORT_RESERVEDMAX + || lowportmin < IPPORT_RESERVEDMIN + ) { + lowportmin = old; + return (EINVAL); + } + return (error); + case IPCTL_LOWPORTMAX: + old = lowportmax; + error = sysctl_int(oldp, oldlenp, newp, newlen, &lowportmax); + if (lowportmin >= lowportmax + || lowportmax > IPPORT_RESERVEDMAX + || lowportmax < IPPORT_RESERVEDMIN + ) { + lowportmax = old; + return (EINVAL); + } + return (error); +#endif + + case IPCTL_MAXFRAGPACKETS: + return (sysctl_int(oldp, oldlenp, newp, newlen, + &ip_maxfragpackets)); + default: return (EOPNOTSUPP); }