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 1.135.2.4 retrieving revision 1.190 diff -u -p -r1.135.2.4 -r1.190 --- src/sys/netinet/ip_input.c 2002/06/23 17:50:51 1.135.2.4 +++ src/sys/netinet/ip_input.c 2003/12/06 23:56:10 1.190 @@ -1,4 +1,4 @@ -/* $NetBSD: ip_input.c,v 1.135.2.4 2002/06/23 17:50:51 jdolecek Exp $ */ +/* $NetBSD: ip_input.c,v 1.190 2003/12/06 23:56:10 jonathan Exp $ */ /* * Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. @@ -78,11 +78,7 @@ * 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 + * 3. 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. * @@ -102,12 +98,14 @@ */ #include -__KERNEL_RCSID(0, "$NetBSD: ip_input.c,v 1.135.2.4 2002/06/23 17:50:51 jdolecek Exp $"); +__KERNEL_RCSID(0, "$NetBSD: ip_input.c,v 1.190 2003/12/06 23:56:10 jonathan Exp $"); +#include "opt_inet.h" #include "opt_gateway.h" #include "opt_pfil_hooks.h" #include "opt_ipsec.h" #include "opt_mrouting.h" +#include "opt_mbuftrace.h" #include "opt_inet_csum.h" #include @@ -150,6 +148,10 @@ __KERNEL_RCSID(0, "$NetBSD: ip_input.c,v #include #include #endif +#ifdef FAST_IPSEC +#include +#include +#endif /* FAST_IPSEC*/ #ifndef IPFORWARDING #ifdef GATEWAY @@ -192,28 +194,58 @@ int ip_forwsrcrt = IPFORWSRCRT; int ip_directedbcast = IPDIRECTEDBCAST; int ip_allowsrcrt = IPALLOWSRCRT; int ip_mtudisc = IPMTUDISC; -u_int ip_mtudisc_timeout = IPMTUDISCTIMEOUT; +int ip_mtudisc_timeout = IPMTUDISCTIMEOUT; #ifdef DIAGNOSTIC int ipprintfs = 0; #endif +int ip_do_randomid = 0; + +/* + * 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. + */ +int ip_checkinterface = 0; + + struct rttimer_queue *ip_mtudisc_timeout_q = NULL; extern struct domain inetdomain; int ipqmaxlen = IFQ_MAXLEN; u_long in_ifaddrhash; /* size of hash table - 1 */ int in_ifaddrentries; /* total number of addrs */ -struct in_ifaddrhead in_ifaddr; +struct in_ifaddrhead in_ifaddrhead; struct in_ifaddrhashhead *in_ifaddrhashtbl; +u_long in_multihash; /* size of hash table - 1 */ +int in_multientries; /* total number of addrs */ +struct in_multihashhead *in_multihashtbl; struct ifqueue ipintrq; struct ipstat ipstat; -u_int16_t ip_id; +uint16_t ip_id; #ifdef PFIL_HOOKS struct pfil_head inet_pfil_hook; #endif -struct ipqhead ipq; +/* IP datagram reassembly queues (hashed) */ +#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) +struct ipqhead ipq[IPREASS_NHASH]; +#ifdef notyet +static int nipq = 0; /* total # of reass queues */ +#endif int ipq_locked; int ip_nfragpackets = 0; int ip_maxfragpackets = 200; @@ -257,14 +289,14 @@ do { \ printf("%s:%d: ipq already locked\n", __FILE__, __LINE__); \ panic("ipq_lock"); \ } \ -} while (0) +} while (/*CONSTCOND*/ 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) +} while (/*CONSTCOND*/ 0) #else #define IPQ_LOCK() (void) ipq_lock_try() #define IPQ_LOCK_CHECK() /* nothing */ @@ -272,6 +304,7 @@ do { \ #define IPQ_UNLOCK() ipq_unlock() +struct pool inmulti_pool; struct pool ipqent_pool; #ifdef INET_CSUM_COUNTERS @@ -309,6 +342,11 @@ static struct ip_srcrt { static void save_rte __P((u_char *, struct in_addr)); +#ifdef MBUFTRACE +struct mowner ip_rx_mowner = { "internet", "rx" }; +struct mowner ip_tx_mowner = { "internet", "tx" }; +#endif + /* * IP initialization: fill in IP protocol switch table. * All protocols not implemented in kernel go to raw IP protocol handler. @@ -319,6 +357,8 @@ ip_init() struct protosw *pr; int i; + pool_init(&inmulti_pool, sizeof(struct in_multi), 0, 0, 0, "inmltpl", + NULL); pool_init(&ipqent_pool, sizeof(struct ipqent), 0, 0, 0, "ipqepl", NULL); @@ -332,15 +372,17 @@ ip_init() if (pr->pr_domain->dom_family == PF_INET && pr->pr_protocol && pr->pr_protocol != IPPROTO_RAW) ip_protox[pr->pr_protocol] = pr - inetsw; - LIST_INIT(&ipq); - ip_id = time.tv_sec & 0xffff; + for (i = 0; i < IPREASS_NHASH; i++) + LIST_INIT(&ipq[i]); + + ip_id = time.tv_sec & 0xfffff; ipintrq.ifq_maxlen = ipqmaxlen; - TAILQ_INIT(&in_ifaddr); + TAILQ_INIT(&in_ifaddrhead); in_ifaddrhashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IFADDR, M_WAITOK, &in_ifaddrhash); - if (ip_mtudisc != 0) - ip_mtudisc_timeout_q = - rt_timer_queue_create(ip_mtudisc_timeout); + in_multihashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IPMADDR, + M_WAITOK, &in_multihash); + ip_mtudisc_timeout_q = rt_timer_queue_create(ip_mtudisc_timeout); #ifdef GATEWAY ipflow_init(); #endif @@ -360,6 +402,11 @@ ip_init() evcnt_attach_static(&ip_hwcsum_ok); evcnt_attach_static(&ip_swcsum); #endif /* INET_CSUM_COUNTERS */ + +#ifdef MBUFTRACE + MOWNER_ATTACH(&ip_tx_mowner); + MOWNER_ATTACH(&ip_rx_mowner); +#endif /* MBUFTRACE */ } struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; @@ -380,6 +427,7 @@ ipintr() splx(s); if (m == 0) return; + MCLAIM(m, &ip_rx_mowner); ip_input(m); } } @@ -398,32 +446,47 @@ ip_input(struct mbuf *m) struct ipqent *ipqe; int hlen = 0, mff, len; int downmatch; + int checkif; + int srcrt = 0; + u_int hash; +#ifdef FAST_IPSEC + struct m_tag *mtag; + struct tdb_ident *tdbi; + struct secpolicy *sp; + int s, error; +#endif /* FAST_IPSEC */ + MCLAIM(m, &ip_rx_mowner); #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. */ - if (TAILQ_FIRST(&in_ifaddr) == 0) + if (TAILQ_FIRST(&in_ifaddrhead) == 0) goto bad; ipstat.ips_total++; - if (m->m_len < sizeof (struct ip) && - (m = m_pullup(m, sizeof (struct ip))) == 0) { - ipstat.ips_toosmall++; - return; + /* + * If the IP header is not aligned, slurp it up into a new + * mbuf with space for link headers, in the event we forward + * it. Otherwise, if it is aligned, make sure the entire + * base IP header is in the first mbuf of the chain. + */ + if (IP_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { + if ((m = m_copyup(m, sizeof(struct ip), + (max_linkhdr + 3) & ~3)) == NULL) { + /* XXXJRT new stat, please */ + ipstat.ips_toosmall++; + return; + } + } else if (__predict_false(m->m_len < sizeof (struct ip))) { + if ((m = m_pullup(m, sizeof (struct ip))) == NULL) { + ipstat.ips_toosmall++; + return; + } } ip = mtod(m, struct ip *); if (ip->ip_v != IPVERSION) { @@ -510,7 +573,7 @@ ip_input(struct mbuf *m) m_adj(m, len - m->m_pkthdr.len); } -#ifdef IPSEC +#if defined(IPSEC) || defined(FAST_IPSEC) /* ipflow (IP fast forwarding) is not compatible with IPsec. */ m->m_flags &= ~M_CANFASTFWD; #else @@ -535,17 +598,23 @@ ip_input(struct mbuf *m) */ #ifdef IPSEC if (!ipsec_getnhist(m)) +#elif defined(FAST_IPSEC) + if (!ipsec_indone(m)) #else if (1) #endif { + struct in_addr odst; + + odst = ip->ip_dst; if (pfil_run_hooks(&inet_pfil_hook, &m, m->m_pkthdr.rcvif, - PFIL_IN) != 0) - return; + PFIL_IN) != 0) + return; if (m == NULL) return; ip = mtod(m, struct ip *); hlen = ip->ip_hl << 2; + srcrt = (odst.s_addr != ip->ip_dst.s_addr); } #endif /* PFIL_HOOKS */ @@ -558,12 +627,6 @@ ip_input(struct mbuf *m) #endif /* - * Convert fields to host representation. - */ - NTOHS(ip->ip_len); - NTOHS(ip->ip_off); - - /* * Process options and, if not destined for us, * ship it on. ip_dooptions returns 1 when an * error was detected (causing an icmp message @@ -574,6 +637,26 @@ ip_input(struct mbuf *m) return; /* + * 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. + * + * 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. + * + * XXX - We need to add a per ifaddr flag for this so that + * we get finer grain control. + */ + checkif = ip_checkinterface && (ipforwarding == 0) && + (m->m_pkthdr.rcvif != NULL) && + ((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0); + + /* * Check our list of addresses, to see if the packet is for us. * * Traditional 4.4BSD did not consult IFF_UP at all. @@ -583,6 +666,8 @@ ip_input(struct mbuf *m) downmatch = 0; LIST_FOREACH(ia, &IN_IFADDR_HASH(ip->ip_dst.s_addr), ia_hash) { if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { + if (checkif && ia->ia_ifp != m->m_pkthdr.rcvif) + continue; if ((ia->ia_ifp->if_flags & IFF_UP) != 0) break; else @@ -695,8 +780,36 @@ ip_input(struct mbuf *m) goto bad; } #endif +#ifdef FAST_IPSEC + mtag = m_tag_find(m, PACKET_TAG_IPSEC_IN_DONE, NULL); + s = splsoftnet(); + if (mtag != NULL) { + tdbi = (struct tdb_ident *)(mtag + 1); + sp = ipsec_getpolicy(tdbi, IPSEC_DIR_INBOUND); + } else { + sp = ipsec_getpolicybyaddr(m, IPSEC_DIR_INBOUND, + IP_FORWARDING, &error); + } + if (sp == NULL) { /* NB: can happen if error */ + splx(s); + /*XXX error stat???*/ + DPRINTF(("ip_input: no SP for forwarding\n")); /*XXX*/ + goto bad; + } - ip_forward(m, 0); + /* + * Check security policy against packet attributes. + */ + error = ipsec_in_reject(sp, m); + KEY_FREESP(&sp); + splx(s); + if (error) { + ipstat.ips_cantforward++; + goto bad; + } +#endif /* FAST_IPSEC */ + + ip_forward(m, srcrt); } return; @@ -708,18 +821,31 @@ ours: * if the packet was previously fragmented, * but it's not worth the time; just let them time out.) */ - if (ip->ip_off & ~(IP_DF|IP_RF)) { + if (ip->ip_off & ~htons(IP_DF|IP_RF)) { + if (M_READONLY(m)) { + if ((m = m_pullup(m, hlen)) == NULL) { + ipstat.ips_toosmall++; + goto bad; + } + ip = mtod(m, struct ip *); + } + /* * Look for queue of fragments * of this datagram. */ IPQ_LOCK(); - LIST_FOREACH(fp, &ipq, ipq_q) + hash = IPREASS_HASH(ip->ip_src.s_addr, ip->ip_id); + /* XXX LIST_FOREACH(fp, &ipq[hash], ipq_q) */ + for (fp = LIST_FIRST(&ipq[hash]); fp != NULL; + fp = LIST_NEXT(fp, ipq_q)) { if (ip->ip_id == fp->ipq_id && in_hosteq(ip->ip_src, fp->ipq_src) && in_hosteq(ip->ip_dst, fp->ipq_dst) && ip->ip_p == fp->ipq_p) goto found; + + } fp = 0; found: @@ -728,27 +854,28 @@ found: * set ipqe_mff if more fragments are expected, * convert offset of this to bytes. */ - ip->ip_len -= hlen; - mff = (ip->ip_off & IP_MF) != 0; + ip->ip_len = htons(ntohs(ip->ip_len) - hlen); + mff = (ip->ip_off & htons(IP_MF)) != 0; if (mff) { /* * 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) { + if (ntohs(ip->ip_len) == 0 || + (ntohs(ip->ip_len) & 0x7) != 0) { ipstat.ips_badfrags++; IPQ_UNLOCK(); goto bad; } } - ip->ip_off <<= 3; + ip->ip_off = htons((ntohs(ip->ip_off) & IP_OFFMASK) << 3); /* * If datagram marked as having more fragments * or if this is not the first fragment, * attempt reassembly; if it succeeds, proceed. */ - if (mff || ip->ip_off) { + if (mff || ip->ip_off != htons(0)) { ipstat.ips_fragments++; ipqe = pool_get(&ipqent_pool, PR_NOWAIT); if (ipqe == NULL) { @@ -759,7 +886,7 @@ found: ipqe->ipqe_mff = mff; ipqe->ipqe_m = m; ipqe->ipqe_ip = ip; - m = ip_reass(ipqe, fp); + m = ip_reass(ipqe, fp, &ipq[hash]); if (m == 0) { IPQ_UNLOCK(); return; @@ -767,14 +894,14 @@ found: ipstat.ips_reassembled++; ip = mtod(m, struct ip *); hlen = ip->ip_hl << 2; - ip->ip_len += hlen; + ip->ip_len = htons(ntohs(ip->ip_len) + hlen); } else if (fp) ip_freef(fp); IPQ_UNLOCK(); } -#ifdef IPSEC +#if defined(IPSEC) /* * enforce IPsec policy checking if we are seeing last header. * note that we do not visit this with protocols with pcb layer @@ -786,13 +913,52 @@ found: goto bad; } #endif +#if FAST_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) { + /* + * Check if the packet has already had IPsec processing + * done. If so, then just pass it along. This tag gets + * set during AH, ESP, etc. input handling, before the + * packet is returned to the ip input queue for delivery. + */ + mtag = m_tag_find(m, PACKET_TAG_IPSEC_IN_DONE, NULL); + s = splsoftnet(); + if (mtag != NULL) { + tdbi = (struct tdb_ident *)(mtag + 1); + sp = ipsec_getpolicy(tdbi, IPSEC_DIR_INBOUND); + } else { + sp = ipsec_getpolicybyaddr(m, IPSEC_DIR_INBOUND, + IP_FORWARDING, &error); + } + if (sp != NULL) { + /* + * Check security policy against packet attributes. + */ + error = ipsec_in_reject(sp, m); + KEY_FREESP(&sp); + } else { + /* XXX error stat??? */ + error = EINVAL; +DPRINTF(("ip_input: no SP, packet discarded\n"));/*XXX*/ + goto bad; + } + splx(s); + if (error) + goto bad; + } +#endif /* FAST_IPSEC */ /* * Switch out to protocol's input routine. */ #if IFA_STATS if (ia && ip) - ia->ia_ifa.ifa_data.ifad_inbytes += ip->ip_len; + ia->ia_ifa.ifa_data.ifad_inbytes += ntohs(ip->ip_len); #endif ipstat.ips_delivered++; { @@ -817,9 +983,10 @@ badcsum: * is given as fp; otherwise have to make a chain. */ struct mbuf * -ip_reass(ipqe, fp) +ip_reass(ipqe, fp, ipqhead) struct ipqent *ipqe; struct ipq *fp; + struct ipqhead *ipqhead; { struct mbuf *m = ipqe->ipqe_m; struct ipqent *nq, *p, *q; @@ -856,7 +1023,7 @@ ip_reass(ipqe, fp) M_FTABLE, M_NOWAIT); if (fp == NULL) goto dropfrag; - LIST_INSERT_HEAD(&ipq, fp, ipq_q); + LIST_INSERT_HEAD(ipqhead, fp, ipq_q); fp->ipq_ttl = IPFRAGTTL; fp->ipq_p = ipqe->ipqe_ip->ip_p; fp->ipq_id = ipqe->ipqe_ip->ip_id; @@ -872,7 +1039,7 @@ ip_reass(ipqe, fp) */ for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; p = q, q = TAILQ_NEXT(q, ipqe_q)) - if (q->ipqe_ip->ip_off > ipqe->ipqe_ip->ip_off) + if (ntohs(q->ipqe_ip->ip_off) > ntohs(ipqe->ipqe_ip->ip_off)) break; /* @@ -881,14 +1048,16 @@ ip_reass(ipqe, fp) * segment. If it provides all of our data, drop us. */ if (p != NULL) { - i = p->ipqe_ip->ip_off + p->ipqe_ip->ip_len - - ipqe->ipqe_ip->ip_off; + i = ntohs(p->ipqe_ip->ip_off) + ntohs(p->ipqe_ip->ip_len) - + ntohs(ipqe->ipqe_ip->ip_off); if (i > 0) { - if (i >= ipqe->ipqe_ip->ip_len) + if (i >= ntohs(ipqe->ipqe_ip->ip_len)) goto dropfrag; m_adj(ipqe->ipqe_m, i); - ipqe->ipqe_ip->ip_off += i; - ipqe->ipqe_ip->ip_len -= i; + ipqe->ipqe_ip->ip_off = + htons(ntohs(ipqe->ipqe_ip->ip_off) + i); + ipqe->ipqe_ip->ip_len = + htons(ntohs(ipqe->ipqe_ip->ip_len) - i); } } @@ -896,13 +1065,16 @@ ip_reass(ipqe, fp) * While we overlap succeeding segments trim them or, * if they are completely covered, dequeue them. */ - for (; q != NULL && ipqe->ipqe_ip->ip_off + ipqe->ipqe_ip->ip_len > - q->ipqe_ip->ip_off; q = nq) { - i = (ipqe->ipqe_ip->ip_off + ipqe->ipqe_ip->ip_len) - - q->ipqe_ip->ip_off; - if (i < q->ipqe_ip->ip_len) { - q->ipqe_ip->ip_len -= i; - q->ipqe_ip->ip_off += i; + for (; q != NULL && + ntohs(ipqe->ipqe_ip->ip_off) + ntohs(ipqe->ipqe_ip->ip_len) > + ntohs(q->ipqe_ip->ip_off); q = nq) { + i = (ntohs(ipqe->ipqe_ip->ip_off) + + ntohs(ipqe->ipqe_ip->ip_len)) - ntohs(q->ipqe_ip->ip_off); + if (i < ntohs(q->ipqe_ip->ip_len)) { + q->ipqe_ip->ip_len = + htons(ntohs(q->ipqe_ip->ip_len) - i); + q->ipqe_ip->ip_off = + htons(ntohs(q->ipqe_ip->ip_off) + i); m_adj(q->ipqe_m, i); break; } @@ -925,9 +1097,9 @@ insert: next = 0; for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; p = q, q = TAILQ_NEXT(q, ipqe_q)) { - if (q->ipqe_ip->ip_off != next) + if (ntohs(q->ipqe_ip->ip_off) != next) return (0); - next += q->ipqe_ip->ip_len; + next += ntohs(q->ipqe_ip->ip_len); } if (p->ipqe_mff) return (0); @@ -962,7 +1134,7 @@ insert: * dequeue and discard fragment reassembly header. * Make header visible. */ - ip->ip_len = next; + ip->ip_len = htons(next); ip->ip_src = fp->ipq_src; ip->ip_dst = fp->ipq_dst; LIST_REMOVE(fp, ipq_q); @@ -1017,27 +1189,47 @@ ip_freef(fp) void ip_slowtimo() { + static unsigned dropscanidx = 0; + unsigned i; struct ipq *fp, *nfp; int s = splsoftnet(); IPQ_LOCK(); - for (fp = LIST_FIRST(&ipq); fp != NULL; fp = nfp) { - nfp = LIST_NEXT(fp, ipq_q); - if (--fp->ipq_ttl == 0) { - ipstat.ips_fragtimeout++; - ip_freef(fp); + for (i = 0; i < IPREASS_NHASH; i++) { + for (fp = LIST_FIRST(&ipq[i]); fp != NULL; fp = nfp) { + nfp = LIST_NEXT(fp, ipq_q); + if (--fp->ipq_ttl == 0) { + ipstat.ips_fragtimeout++; + 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. + * enough to get down to the new limit. Start draining + * from the reassembly hashqueue most recently drained. */ if (ip_maxfragpackets < 0) ; else { - while (ip_nfragpackets > ip_maxfragpackets && LIST_FIRST(&ipq)) - ip_freef(LIST_FIRST(&ipq)); + int wrapped = 0; + + i = dropscanidx; + while (ip_nfragpackets > ip_maxfragpackets && wrapped == 0) { + while (LIST_FIRST(&ipq[i]) != NULL) + ip_freef(LIST_FIRST(&ipq[i])); + if (++i >= IPREASS_NHASH) { + i = 0; + } + /* + * Dont scan forever even if fragment counters are + * wrong: stop after scanning entire reassembly queue. + */ + if (i == dropscanidx) + wrapped = 1; + } + dropscanidx = i; } IPQ_UNLOCK(); #ifdef GATEWAY @@ -1052,6 +1244,7 @@ ip_slowtimo() void ip_drain() { + int i; /* * We may be called from a device's interrupt context. If @@ -1060,9 +1253,14 @@ ip_drain() if (ipq_lock_try() == 0) return; - while (LIST_FIRST(&ipq) != NULL) { - ipstat.ips_fragdropped++; - ip_freef(LIST_FIRST(&ipq)); + for (i = 0; i < IPREASS_NHASH; i++) { + struct ipqhead *ipqh = &ipq[i]; + struct ipq *fp, *nfp; + for (fp = LIST_FIRST(ipqh); fp != NULL; fp = nfp) { + nfp = LIST_NEXT(fp, ipq_q); + ip_freef(fp); + ipstat.ips_fragdropped++; + } } IPQ_UNLOCK(); @@ -1369,6 +1567,7 @@ ip_srcroute() if (m == 0) return ((struct mbuf *)0); + MCLAIM(m, &inetdomain.dom_mowner); #define OPTSIZ (sizeof(ip_srcrt.nop) + sizeof(ip_srcrt.srcopt)) /* length is (nhops+1)*sizeof(addr) + sizeof(nop + srcrt header) */ @@ -1445,7 +1644,7 @@ ip_stripoptions(m, mopt) m->m_len -= olen; if (m->m_flags & M_PKTHDR) m->m_pkthdr.len -= olen; - ip->ip_len -= olen; + ip->ip_len = htons(ntohs(ip->ip_len) - olen); ip->ip_hl = sizeof (struct ip) >> 2; } @@ -1484,11 +1683,16 @@ ip_forward(m, srcrt) struct mbuf *mcopy; n_long dest; struct ifnet *destifp; -#ifdef IPSEC +#if defined(IPSEC) || defined(FAST_IPSEC) struct ifnet dummyifp; #endif /* + * We are now in the output path. + */ + MCLAIM(m, &ip_tx_mowner); + + /* * Clear any in-bound checksum flags for this packet. */ m->m_pkthdr.csum_flags = 0; @@ -1535,7 +1739,7 @@ ip_forward(m, srcrt) * we need to generate an ICMP message to the src. * Pullup to avoid sharing mbuf cluster between m and mcopy. */ - mcopy = m_copym(m, 0, imin((int)ip->ip_len, 68), M_DONTWAIT); + mcopy = m_copym(m, 0, imin(ntohs(ip->ip_len), 68), M_DONTWAIT); if (mcopy) mcopy = m_pullup(mcopy, ip->ip_hl << 2); @@ -1572,12 +1776,10 @@ ip_forward(m, srcrt) } } -#ifdef IPSEC - /* Don't lookup socket in forwarding case */ - (void)ipsec_setsocket(m, NULL); -#endif error = ip_output(m, (struct mbuf *)0, &ipforward_rt, - (IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), 0); + (IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), + (struct ip_moptions *)NULL, (struct socket *)NULL); + if (error) ipstat.ips_cantforward++; else { @@ -1617,7 +1819,7 @@ ip_forward(m, srcrt) case EMSGSIZE: type = ICMP_UNREACH; code = ICMP_UNREACH_NEEDFRAG; -#ifndef IPSEC +#if !defined(IPSEC) && !defined(FAST_IPSEC) if (ipforward_rt.ro_rt) destifp = ipforward_rt.ro_rt->rt_ifp; #else @@ -1634,17 +1836,15 @@ ip_forward(m, srcrt) struct route *ro; sp = ipsec4_getpolicybyaddr(mcopy, - IPSEC_DIR_OUTBOUND, - IP_FORWARDING, - &ipsecerror); + 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); + IPSEC_DIR_OUTBOUND, NULL); /* * find the correct route for outer IPv4 @@ -1670,7 +1870,11 @@ ip_forward(m, srcrt) } } +#ifdef IPSEC key_freesp(sp); +#else + KEY_FREESP(&sp); +#endif } } #endif /*IPSEC*/ @@ -1757,155 +1961,184 @@ ip_savecontrol(inp, mp, ip, m) } } -int -ip_sysctl(name, namelen, oldp, oldlenp, newp, newlen) - int *name; - u_int namelen; - void *oldp; - size_t *oldlenp; - void *newp; - size_t newlen; +/* + * sysctl helper routine for net.inet.ip.mtudisctimeout. checks the + * range of the new value and tweaks timers if it changes. + */ +static int +sysctl_net_inet_ip_pmtudto(SYSCTLFN_ARGS) { - extern int subnetsarelocal, hostzeroisbroadcast; - - int error, old; + int error, tmp; + struct sysctlnode node; - /* All sysctl names at this level are terminal. */ - if (namelen != 1) - return (ENOTDIR); - - switch (name[0]) { - case IPCTL_FORWARDING: - return (sysctl_int(oldp, oldlenp, newp, newlen, &ipforwarding)); - case IPCTL_SENDREDIRECTS: - return (sysctl_int(oldp, oldlenp, newp, newlen, - &ipsendredirects)); - case IPCTL_DEFTTL: - return (sysctl_int(oldp, oldlenp, newp, newlen, &ip_defttl)); -#ifdef notyet - case IPCTL_DEFMTU: - return (sysctl_int(oldp, oldlenp, newp, newlen, &ip_mtu)); -#endif - case IPCTL_FORWSRCRT: - /* Don't allow this to change in a secure environment. */ - if (securelevel > 0) - return (sysctl_rdint(oldp, oldlenp, newp, - ip_forwsrcrt)); - else - return (sysctl_int(oldp, oldlenp, newp, newlen, - &ip_forwsrcrt)); - case IPCTL_DIRECTEDBCAST: - return (sysctl_int(oldp, oldlenp, newp, newlen, - &ip_directedbcast)); - 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); + node = *rnode; + tmp = ip_mtudisc_timeout; + node.sysctl_data = &tmp; + error = sysctl_lookup(SYSCTLFN_CALL(&node)); + if (error || newp == NULL) return (error); + if (tmp < 0) + return (EINVAL); + + ip_mtudisc_timeout = tmp; + rt_timer_queue_change(ip_mtudisc_timeout_q, ip_mtudisc_timeout); + + return (0); +} + #ifdef GATEWAY - case IPCTL_MAXFLOWS: - { - int s; +/* + * sysctl helper routine for net.inet.ip.maxflows. apparently if + * maxflows is even looked up, we "reap flows". + */ +static int +sysctl_net_inet_ip_maxflows(SYSCTLFN_ARGS) +{ + 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 + s = sysctl_lookup(SYSCTLFN_CALL(rnode)); + if (s) + return (s); + + s = splsoftnet(); + ipflow_reap(0); + splx(s); -#if NGRE > 0 - case IPCTL_GRE_TTL: - return(sysctl_int(oldp, oldlenp, newp, newlen, - &ip_gre_ttl)); -#endif + return (0); +} +#endif /* GATEWAY */ -#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)); +SYSCTL_SETUP(sysctl_net_inet_ip_setup, "sysctl net.inet.ip subtree setup") +{ + extern int subnetsarelocal, hostzeroisbroadcast; - default: - return (EOPNOTSUPP); - } - /* NOTREACHED */ + sysctl_createv(SYSCTL_PERMANENT, + CTLTYPE_NODE, "net", NULL, + NULL, 0, NULL, 0, + CTL_NET, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT, + CTLTYPE_NODE, "inet", NULL, + NULL, 0, NULL, 0, + CTL_NET, PF_INET, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT, + CTLTYPE_NODE, "ip", NULL, + NULL, 0, NULL, 0, + CTL_NET, PF_INET, IPPROTO_IP, CTL_EOL); + + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "forwarding", NULL, + NULL, 0, &ipforwarding, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_FORWARDING, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "redirect", NULL, + NULL, 0, &ipsendredirects, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_SENDREDIRECTS, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "ttl", NULL, + NULL, 0, &ip_defttl, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_DEFTTL, CTL_EOL); +#ifdef IPCTL_DEFMTU + sysctl_createv(SYSCTL_PERMANENT /* |SYSCTL_READWRITE? */, + CTLTYPE_INT, "mtu", NULL, + NULL, 0, &ip_mtu, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_DEFMTU, CTL_EOL); +#endif /* IPCTL_DEFMTU */ + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READONLY1, + CTLTYPE_INT, "forwsrcrt", NULL, + NULL, 0, &ip_forwsrcrt, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_FORWSRCRT, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "directed-broadcast", NULL, + NULL, 0, &ip_directedbcast, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_DIRECTEDBCAST, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "allowsrcrt", NULL, + NULL, 0, &ip_allowsrcrt, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_ALLOWSRCRT, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "subnetsarelocal", NULL, + NULL, 0, &subnetsarelocal, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_SUBNETSARELOCAL, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "mtudisc", NULL, + NULL, 0, &ip_mtudisc, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_MTUDISC, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "anonportmin", NULL, + sysctl_net_inet_ip_ports, 0, &anonportmin, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_ANONPORTMIN, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "anonportmax", NULL, + sysctl_net_inet_ip_ports, 0, &anonportmax, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_ANONPORTMAX, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "mtudisctimeout", NULL, + sysctl_net_inet_ip_pmtudto, 0, &ip_mtudisc_timeout, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_MTUDISCTIMEOUT, CTL_EOL); +#ifdef GATEWAY + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "maxflows", NULL, + sysctl_net_inet_ip_maxflows, 0, &ip_maxflows, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_MAXFLOWS, CTL_EOL); +#endif /* GATEWAY */ + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "hostzerobroadcast", NULL, + NULL, 0, &hostzeroisbroadcast, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_HOSTZEROBROADCAST, CTL_EOL); +#if NGIF > 0 + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "gifttl", NULL, + NULL, 0, &ip_gif_ttl, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_GIF_TTL, CTL_EOL); +#endif /* NGIF */ +#ifndef IPNOPRIVPORTS + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "lowportmin", NULL, + sysctl_net_inet_ip_ports, 0, &lowportmin, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_LOWPORTMIN, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "lowportmax", NULL, + sysctl_net_inet_ip_ports, 0, &lowportmax, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_LOWPORTMAX, CTL_EOL); +#endif /* IPNOPRIVPORTS */ + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "maxfragpackets", NULL, + NULL, 0, &ip_maxfragpackets, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_MAXFRAGPACKETS, CTL_EOL); +#if NGRE > 0 + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "grettl", NULL, + NULL, 0, &ip_gre_ttl, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_GRE_TTL, CTL_EOL); +#endif /* NGRE */ + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "checkinterface", NULL, + NULL, 0, &ip_checkinterface, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_CHECKINTERFACE, CTL_EOL); + sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, + CTLTYPE_INT, "random_id", NULL, + NULL, 0, &ip_do_randomid, 0, + CTL_NET, PF_INET, IPPROTO_IP, + IPCTL_RANDOMID, CTL_EOL); }