ip_input.c revision 1.230
1/* $NetBSD: ip_input.c,v 1.230 2006/09/08 20:58:58 elad Exp $ */ 2 3/* 4 * Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 3. Neither the name of the project nor the names of its contributors 16 * may be used to endorse or promote products derived from this software 17 * without specific prior written permission. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND 20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 22 * ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE 23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 29 * SUCH DAMAGE. 30 */ 31 32/*- 33 * Copyright (c) 1998 The NetBSD Foundation, Inc. 34 * All rights reserved. 35 * 36 * This code is derived from software contributed to The NetBSD Foundation 37 * by Public Access Networks Corporation ("Panix"). It was developed under 38 * contract to Panix by Eric Haszlakiewicz and Thor Lancelot Simon. 39 * 40 * Redistribution and use in source and binary forms, with or without 41 * modification, are permitted provided that the following conditions 42 * are met: 43 * 1. Redistributions of source code must retain the above copyright 44 * notice, this list of conditions and the following disclaimer. 45 * 2. Redistributions in binary form must reproduce the above copyright 46 * notice, this list of conditions and the following disclaimer in the 47 * documentation and/or other materials provided with the distribution. 48 * 3. All advertising materials mentioning features or use of this software 49 * must display the following acknowledgement: 50 * This product includes software developed by the NetBSD 51 * Foundation, Inc. and its contributors. 52 * 4. Neither the name of The NetBSD Foundation nor the names of its 53 * contributors may be used to endorse or promote products derived 54 * from this software without specific prior written permission. 55 * 56 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 57 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 58 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 59 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 60 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 61 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 62 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 63 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 64 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 65 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 66 * POSSIBILITY OF SUCH DAMAGE. 67 */ 68 69/* 70 * Copyright (c) 1982, 1986, 1988, 1993 71 * The Regents of the University of California. All rights reserved. 72 * 73 * Redistribution and use in source and binary forms, with or without 74 * modification, are permitted provided that the following conditions 75 * are met: 76 * 1. Redistributions of source code must retain the above copyright 77 * notice, this list of conditions and the following disclaimer. 78 * 2. Redistributions in binary form must reproduce the above copyright 79 * notice, this list of conditions and the following disclaimer in the 80 * documentation and/or other materials provided with the distribution. 81 * 3. Neither the name of the University nor the names of its contributors 82 * may be used to endorse or promote products derived from this software 83 * without specific prior written permission. 84 * 85 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 86 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 87 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 88 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 89 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 90 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 91 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 92 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 93 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 94 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 95 * SUCH DAMAGE. 96 * 97 * @(#)ip_input.c 8.2 (Berkeley) 1/4/94 98 */ 99 100#include <sys/cdefs.h> 101__KERNEL_RCSID(0, "$NetBSD: ip_input.c,v 1.230 2006/09/08 20:58:58 elad Exp $"); 102 103#include "opt_inet.h" 104#include "opt_gateway.h" 105#include "opt_pfil_hooks.h" 106#include "opt_ipsec.h" 107#include "opt_mrouting.h" 108#include "opt_mbuftrace.h" 109#include "opt_inet_csum.h" 110 111#include <sys/param.h> 112#include <sys/systm.h> 113#include <sys/malloc.h> 114#include <sys/mbuf.h> 115#include <sys/domain.h> 116#include <sys/protosw.h> 117#include <sys/socket.h> 118#include <sys/socketvar.h> 119#include <sys/errno.h> 120#include <sys/time.h> 121#include <sys/kernel.h> 122#include <sys/pool.h> 123#include <sys/sysctl.h> 124#include <sys/kauth.h> 125 126#include <net/if.h> 127#include <net/if_dl.h> 128#include <net/route.h> 129#include <net/pfil.h> 130 131#include <netinet/in.h> 132#include <netinet/in_systm.h> 133#include <netinet/ip.h> 134#include <netinet/in_pcb.h> 135#include <netinet/in_proto.h> 136#include <netinet/in_var.h> 137#include <netinet/ip_var.h> 138#include <netinet/ip_icmp.h> 139/* just for gif_ttl */ 140#include <netinet/in_gif.h> 141#include "gif.h" 142#include <net/if_gre.h> 143#include "gre.h" 144 145#ifdef MROUTING 146#include <netinet/ip_mroute.h> 147#endif 148 149#ifdef IPSEC 150#include <netinet6/ipsec.h> 151#include <netkey/key.h> 152#endif 153#ifdef FAST_IPSEC 154#include <netipsec/ipsec.h> 155#include <netipsec/key.h> 156#endif /* FAST_IPSEC*/ 157 158#ifndef IPFORWARDING 159#ifdef GATEWAY 160#define IPFORWARDING 1 /* forward IP packets not for us */ 161#else /* GATEWAY */ 162#define IPFORWARDING 0 /* don't forward IP packets not for us */ 163#endif /* GATEWAY */ 164#endif /* IPFORWARDING */ 165#ifndef IPSENDREDIRECTS 166#define IPSENDREDIRECTS 1 167#endif 168#ifndef IPFORWSRCRT 169#define IPFORWSRCRT 1 /* forward source-routed packets */ 170#endif 171#ifndef IPALLOWSRCRT 172#define IPALLOWSRCRT 1 /* allow source-routed packets */ 173#endif 174#ifndef IPMTUDISC 175#define IPMTUDISC 1 176#endif 177#ifndef IPMTUDISCTIMEOUT 178#define IPMTUDISCTIMEOUT (10 * 60) /* as per RFC 1191 */ 179#endif 180 181/* 182 * Note: DIRECTED_BROADCAST is handled this way so that previous 183 * configuration using this option will Just Work. 184 */ 185#ifndef IPDIRECTEDBCAST 186#ifdef DIRECTED_BROADCAST 187#define IPDIRECTEDBCAST 1 188#else 189#define IPDIRECTEDBCAST 0 190#endif /* DIRECTED_BROADCAST */ 191#endif /* IPDIRECTEDBCAST */ 192int ipforwarding = IPFORWARDING; 193int ipsendredirects = IPSENDREDIRECTS; 194int ip_defttl = IPDEFTTL; 195int ip_forwsrcrt = IPFORWSRCRT; 196int ip_directedbcast = IPDIRECTEDBCAST; 197int ip_allowsrcrt = IPALLOWSRCRT; 198int ip_mtudisc = IPMTUDISC; 199int ip_mtudisc_timeout = IPMTUDISCTIMEOUT; 200#ifdef DIAGNOSTIC 201int ipprintfs = 0; 202#endif 203 204int ip_do_randomid = 0; 205 206/* 207 * XXX - Setting ip_checkinterface mostly implements the receive side of 208 * the Strong ES model described in RFC 1122, but since the routing table 209 * and transmit implementation do not implement the Strong ES model, 210 * setting this to 1 results in an odd hybrid. 211 * 212 * XXX - ip_checkinterface currently must be disabled if you use ipnat 213 * to translate the destination address to another local interface. 214 * 215 * XXX - ip_checkinterface must be disabled if you add IP aliases 216 * to the loopback interface instead of the interface where the 217 * packets for those addresses are received. 218 */ 219int ip_checkinterface = 0; 220 221 222struct rttimer_queue *ip_mtudisc_timeout_q = NULL; 223 224int ipqmaxlen = IFQ_MAXLEN; 225u_long in_ifaddrhash; /* size of hash table - 1 */ 226int in_ifaddrentries; /* total number of addrs */ 227struct in_ifaddrhead in_ifaddrhead; 228struct in_ifaddrhashhead *in_ifaddrhashtbl; 229u_long in_multihash; /* size of hash table - 1 */ 230int in_multientries; /* total number of addrs */ 231struct in_multihashhead *in_multihashtbl; 232struct ifqueue ipintrq; 233struct ipstat ipstat; 234uint16_t ip_id; 235 236#ifdef PFIL_HOOKS 237struct pfil_head inet_pfil_hook; 238#endif 239 240/* 241 * Cached copy of nmbclusters. If nbclusters is different, 242 * recalculate IP parameters derived from nmbclusters. 243 */ 244static int ip_nmbclusters; /* copy of nmbclusters */ 245static void ip_nmbclusters_changed(void); /* recalc limits */ 246 247#define CHECK_NMBCLUSTER_PARAMS() \ 248do { \ 249 if (__predict_false(ip_nmbclusters != nmbclusters)) \ 250 ip_nmbclusters_changed(); \ 251} while (/*CONSTCOND*/0) 252 253/* IP datagram reassembly queues (hashed) */ 254#define IPREASS_NHASH_LOG2 6 255#define IPREASS_NHASH (1 << IPREASS_NHASH_LOG2) 256#define IPREASS_HMASK (IPREASS_NHASH - 1) 257#define IPREASS_HASH(x,y) \ 258 (((((x) & 0xF) | ((((x) >> 8) & 0xF) << 4)) ^ (y)) & IPREASS_HMASK) 259struct ipqhead ipq[IPREASS_NHASH]; 260int ipq_locked; 261static int ip_nfragpackets; /* packets in reass queue */ 262static int ip_nfrags; /* total fragments in reass queues */ 263 264int ip_maxfragpackets = 200; /* limit on packets. XXX sysctl */ 265int ip_maxfrags; /* limit on fragments. XXX sysctl */ 266 267 268/* 269 * Additive-Increase/Multiplicative-Decrease (AIMD) strategy for 270 * IP reassembly queue buffer managment. 271 * 272 * We keep a count of total IP fragments (NB: not fragmented packets!) 273 * awaiting reassembly (ip_nfrags) and a limit (ip_maxfrags) on fragments. 274 * If ip_nfrags exceeds ip_maxfrags the limit, we drop half the 275 * total fragments in reassembly queues.This AIMD policy avoids 276 * repeatedly deleting single packets under heavy fragmentation load 277 * (e.g., from lossy NFS peers). 278 */ 279static u_int ip_reass_ttl_decr(u_int ticks); 280static void ip_reass_drophalf(void); 281 282 283static inline int ipq_lock_try(void); 284static inline void ipq_unlock(void); 285 286static inline int 287ipq_lock_try(void) 288{ 289 int s; 290 291 /* 292 * Use splvm() -- we're blocking things that would cause 293 * mbuf allocation. 294 */ 295 s = splvm(); 296 if (ipq_locked) { 297 splx(s); 298 return (0); 299 } 300 ipq_locked = 1; 301 splx(s); 302 return (1); 303} 304 305static inline void 306ipq_unlock(void) 307{ 308 int s; 309 310 s = splvm(); 311 ipq_locked = 0; 312 splx(s); 313} 314 315#ifdef DIAGNOSTIC 316#define IPQ_LOCK() \ 317do { \ 318 if (ipq_lock_try() == 0) { \ 319 printf("%s:%d: ipq already locked\n", __FILE__, __LINE__); \ 320 panic("ipq_lock"); \ 321 } \ 322} while (/*CONSTCOND*/ 0) 323#define IPQ_LOCK_CHECK() \ 324do { \ 325 if (ipq_locked == 0) { \ 326 printf("%s:%d: ipq lock not held\n", __FILE__, __LINE__); \ 327 panic("ipq lock check"); \ 328 } \ 329} while (/*CONSTCOND*/ 0) 330#else 331#define IPQ_LOCK() (void) ipq_lock_try() 332#define IPQ_LOCK_CHECK() /* nothing */ 333#endif 334 335#define IPQ_UNLOCK() ipq_unlock() 336 337POOL_INIT(inmulti_pool, sizeof(struct in_multi), 0, 0, 0, "inmltpl", NULL); 338POOL_INIT(ipqent_pool, sizeof(struct ipqent), 0, 0, 0, "ipqepl", NULL); 339 340#ifdef INET_CSUM_COUNTERS 341#include <sys/device.h> 342 343struct evcnt ip_hwcsum_bad = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 344 NULL, "inet", "hwcsum bad"); 345struct evcnt ip_hwcsum_ok = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 346 NULL, "inet", "hwcsum ok"); 347struct evcnt ip_swcsum = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 348 NULL, "inet", "swcsum"); 349 350#define INET_CSUM_COUNTER_INCR(ev) (ev)->ev_count++ 351 352EVCNT_ATTACH_STATIC(ip_hwcsum_bad); 353EVCNT_ATTACH_STATIC(ip_hwcsum_ok); 354EVCNT_ATTACH_STATIC(ip_swcsum); 355 356#else 357 358#define INET_CSUM_COUNTER_INCR(ev) /* nothing */ 359 360#endif /* INET_CSUM_COUNTERS */ 361 362/* 363 * We need to save the IP options in case a protocol wants to respond 364 * to an incoming packet over the same route if the packet got here 365 * using IP source routing. This allows connection establishment and 366 * maintenance when the remote end is on a network that is not known 367 * to us. 368 */ 369int ip_nhops = 0; 370static struct ip_srcrt { 371 struct in_addr dst; /* final destination */ 372 char nop; /* one NOP to align */ 373 char srcopt[IPOPT_OFFSET + 1]; /* OPTVAL, OLEN and OFFSET */ 374 struct in_addr route[MAX_IPOPTLEN/sizeof(struct in_addr)]; 375} ip_srcrt; 376 377static void save_rte(u_char *, struct in_addr); 378 379#ifdef MBUFTRACE 380struct mowner ip_rx_mowner = { "internet", "rx" }; 381struct mowner ip_tx_mowner = { "internet", "tx" }; 382#endif 383 384/* 385 * Compute IP limits derived from the value of nmbclusters. 386 */ 387static void 388ip_nmbclusters_changed(void) 389{ 390 ip_maxfrags = nmbclusters / 4; 391 ip_nmbclusters = nmbclusters; 392} 393 394/* 395 * IP initialization: fill in IP protocol switch table. 396 * All protocols not implemented in kernel go to raw IP protocol handler. 397 */ 398void 399ip_init(void) 400{ 401 const struct protosw *pr; 402 int i; 403 404 pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); 405 if (pr == 0) 406 panic("ip_init"); 407 for (i = 0; i < IPPROTO_MAX; i++) 408 ip_protox[i] = pr - inetsw; 409 for (pr = inetdomain.dom_protosw; 410 pr < inetdomain.dom_protoswNPROTOSW; pr++) 411 if (pr->pr_domain->dom_family == PF_INET && 412 pr->pr_protocol && pr->pr_protocol != IPPROTO_RAW) 413 ip_protox[pr->pr_protocol] = pr - inetsw; 414 415 for (i = 0; i < IPREASS_NHASH; i++) 416 LIST_INIT(&ipq[i]); 417 418 ip_id = time_second & 0xfffff; 419 420 ipintrq.ifq_maxlen = ipqmaxlen; 421 ip_nmbclusters_changed(); 422 423 TAILQ_INIT(&in_ifaddrhead); 424 in_ifaddrhashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IFADDR, 425 M_WAITOK, &in_ifaddrhash); 426 in_multihashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IPMADDR, 427 M_WAITOK, &in_multihash); 428 ip_mtudisc_timeout_q = rt_timer_queue_create(ip_mtudisc_timeout); 429#ifdef GATEWAY 430 ipflow_init(); 431#endif 432 433#ifdef PFIL_HOOKS 434 /* Register our Packet Filter hook. */ 435 inet_pfil_hook.ph_type = PFIL_TYPE_AF; 436 inet_pfil_hook.ph_af = AF_INET; 437 i = pfil_head_register(&inet_pfil_hook); 438 if (i != 0) 439 printf("ip_init: WARNING: unable to register pfil hook, " 440 "error %d\n", i); 441#endif /* PFIL_HOOKS */ 442 443#ifdef MBUFTRACE 444 MOWNER_ATTACH(&ip_tx_mowner); 445 MOWNER_ATTACH(&ip_rx_mowner); 446#endif /* MBUFTRACE */ 447} 448 449struct sockaddr_in ipaddr = { 450 .sin_len = sizeof(ipaddr), 451 .sin_family = AF_INET, 452}; 453struct route ipforward_rt; 454 455/* 456 * IP software interrupt routine 457 */ 458void 459ipintr(void) 460{ 461 int s; 462 struct mbuf *m; 463 464 while (1) { 465 s = splnet(); 466 IF_DEQUEUE(&ipintrq, m); 467 splx(s); 468 if (m == 0) 469 return; 470 MCLAIM(m, &ip_rx_mowner); 471 ip_input(m); 472 } 473} 474 475/* 476 * Ip input routine. Checksum and byte swap header. If fragmented 477 * try to reassemble. Process options. Pass to next level. 478 */ 479void 480ip_input(struct mbuf *m) 481{ 482 struct ip *ip = NULL; 483 struct ipq *fp; 484 struct in_ifaddr *ia; 485 struct ifaddr *ifa; 486 struct ipqent *ipqe; 487 int hlen = 0, mff, len; 488 int downmatch; 489 int checkif; 490 int srcrt = 0; 491 u_int hash; 492#ifdef FAST_IPSEC 493 struct m_tag *mtag; 494 struct tdb_ident *tdbi; 495 struct secpolicy *sp; 496 int s, error; 497#endif /* FAST_IPSEC */ 498 499 MCLAIM(m, &ip_rx_mowner); 500#ifdef DIAGNOSTIC 501 if ((m->m_flags & M_PKTHDR) == 0) 502 panic("ipintr no HDR"); 503#endif 504 505 /* 506 * If no IP addresses have been set yet but the interfaces 507 * are receiving, can't do anything with incoming packets yet. 508 */ 509 if (TAILQ_FIRST(&in_ifaddrhead) == 0) 510 goto bad; 511 ipstat.ips_total++; 512 /* 513 * If the IP header is not aligned, slurp it up into a new 514 * mbuf with space for link headers, in the event we forward 515 * it. Otherwise, if it is aligned, make sure the entire 516 * base IP header is in the first mbuf of the chain. 517 */ 518 if (IP_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { 519 if ((m = m_copyup(m, sizeof(struct ip), 520 (max_linkhdr + 3) & ~3)) == NULL) { 521 /* XXXJRT new stat, please */ 522 ipstat.ips_toosmall++; 523 return; 524 } 525 } else if (__predict_false(m->m_len < sizeof (struct ip))) { 526 if ((m = m_pullup(m, sizeof (struct ip))) == NULL) { 527 ipstat.ips_toosmall++; 528 return; 529 } 530 } 531 ip = mtod(m, struct ip *); 532 if (ip->ip_v != IPVERSION) { 533 ipstat.ips_badvers++; 534 goto bad; 535 } 536 hlen = ip->ip_hl << 2; 537 if (hlen < sizeof(struct ip)) { /* minimum header length */ 538 ipstat.ips_badhlen++; 539 goto bad; 540 } 541 if (hlen > m->m_len) { 542 if ((m = m_pullup(m, hlen)) == 0) { 543 ipstat.ips_badhlen++; 544 return; 545 } 546 ip = mtod(m, struct ip *); 547 } 548 549 /* 550 * RFC1122: packets with a multicast source address are 551 * not allowed. 552 */ 553 if (IN_MULTICAST(ip->ip_src.s_addr)) { 554 ipstat.ips_badaddr++; 555 goto bad; 556 } 557 558 /* 127/8 must not appear on wire - RFC1122 */ 559 if ((ntohl(ip->ip_dst.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET || 560 (ntohl(ip->ip_src.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET) { 561 if ((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0) { 562 ipstat.ips_badaddr++; 563 goto bad; 564 } 565 } 566 567 switch (m->m_pkthdr.csum_flags & 568 ((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_IPv4) | 569 M_CSUM_IPv4_BAD)) { 570 case M_CSUM_IPv4|M_CSUM_IPv4_BAD: 571 INET_CSUM_COUNTER_INCR(&ip_hwcsum_bad); 572 goto badcsum; 573 574 case M_CSUM_IPv4: 575 /* Checksum was okay. */ 576 INET_CSUM_COUNTER_INCR(&ip_hwcsum_ok); 577 break; 578 579 default: 580 /* 581 * Must compute it ourselves. Maybe skip checksum on 582 * loopback interfaces. 583 */ 584 if (__predict_true(!(m->m_pkthdr.rcvif->if_flags & 585 IFF_LOOPBACK) || ip_do_loopback_cksum)) { 586 INET_CSUM_COUNTER_INCR(&ip_swcsum); 587 if (in_cksum(m, hlen) != 0) 588 goto badcsum; 589 } 590 break; 591 } 592 593 /* Retrieve the packet length. */ 594 len = ntohs(ip->ip_len); 595 596 /* 597 * Check for additional length bogosity 598 */ 599 if (len < hlen) { 600 ipstat.ips_badlen++; 601 goto bad; 602 } 603 604 /* 605 * Check that the amount of data in the buffers 606 * is as at least much as the IP header would have us expect. 607 * Trim mbufs if longer than we expect. 608 * Drop packet if shorter than we expect. 609 */ 610 if (m->m_pkthdr.len < len) { 611 ipstat.ips_tooshort++; 612 goto bad; 613 } 614 if (m->m_pkthdr.len > len) { 615 if (m->m_len == m->m_pkthdr.len) { 616 m->m_len = len; 617 m->m_pkthdr.len = len; 618 } else 619 m_adj(m, len - m->m_pkthdr.len); 620 } 621 622#if defined(IPSEC) 623 /* ipflow (IP fast forwarding) is not compatible with IPsec. */ 624 m->m_flags &= ~M_CANFASTFWD; 625#else 626 /* 627 * Assume that we can create a fast-forward IP flow entry 628 * based on this packet. 629 */ 630 m->m_flags |= M_CANFASTFWD; 631#endif 632 633#ifdef PFIL_HOOKS 634 /* 635 * Run through list of hooks for input packets. If there are any 636 * filters which require that additional packets in the flow are 637 * not fast-forwarded, they must clear the M_CANFASTFWD flag. 638 * Note that filters must _never_ set this flag, as another filter 639 * in the list may have previously cleared it. 640 */ 641 /* 642 * let ipfilter look at packet on the wire, 643 * not the decapsulated packet. 644 */ 645#ifdef IPSEC 646 if (!ipsec_getnhist(m)) 647#elif defined(FAST_IPSEC) 648 if (!ipsec_indone(m)) 649#else 650 if (1) 651#endif 652 { 653 struct in_addr odst; 654 655 odst = ip->ip_dst; 656 if (pfil_run_hooks(&inet_pfil_hook, &m, m->m_pkthdr.rcvif, 657 PFIL_IN) != 0) 658 return; 659 if (m == NULL) 660 return; 661 ip = mtod(m, struct ip *); 662 hlen = ip->ip_hl << 2; 663 /* 664 * XXX The setting of "srcrt" here is to prevent ip_forward() 665 * from generating ICMP redirects for packets that have 666 * been redirected by a hook back out on to the same LAN that 667 * they came from and is not an indication that the packet 668 * is being inffluenced by source routing options. This 669 * allows things like 670 * "rdr tlp0 0/0 port 80 -> 1.1.1.200 3128 tcp" 671 * where tlp0 is both on the 1.1.1.0/24 network and is the 672 * default route for hosts on 1.1.1.0/24. Of course this 673 * also requires a "map tlp0 ..." to complete the story. 674 * One might argue whether or not this kind of network config. 675 * should be supported in this manner... 676 */ 677 srcrt = (odst.s_addr != ip->ip_dst.s_addr); 678 } 679#endif /* PFIL_HOOKS */ 680 681#ifdef ALTQ 682 /* XXX Temporary until ALTQ is changed to use a pfil hook */ 683 if (altq_input != NULL && (*altq_input)(m, AF_INET) == 0) { 684 /* packet dropped by traffic conditioner */ 685 return; 686 } 687#endif 688 689 /* 690 * Process options and, if not destined for us, 691 * ship it on. ip_dooptions returns 1 when an 692 * error was detected (causing an icmp message 693 * to be sent and the original packet to be freed). 694 */ 695 ip_nhops = 0; /* for source routed packets */ 696 if (hlen > sizeof (struct ip) && ip_dooptions(m)) 697 return; 698 699 /* 700 * Enable a consistency check between the destination address 701 * and the arrival interface for a unicast packet (the RFC 1122 702 * strong ES model) if IP forwarding is disabled and the packet 703 * is not locally generated. 704 * 705 * XXX - Checking also should be disabled if the destination 706 * address is ipnat'ed to a different interface. 707 * 708 * XXX - Checking is incompatible with IP aliases added 709 * to the loopback interface instead of the interface where 710 * the packets are received. 711 * 712 * XXX - We need to add a per ifaddr flag for this so that 713 * we get finer grain control. 714 */ 715 checkif = ip_checkinterface && (ipforwarding == 0) && 716 (m->m_pkthdr.rcvif != NULL) && 717 ((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0); 718 719 /* 720 * Check our list of addresses, to see if the packet is for us. 721 * 722 * Traditional 4.4BSD did not consult IFF_UP at all. 723 * The behavior here is to treat addresses on !IFF_UP interface 724 * as not mine. 725 */ 726 downmatch = 0; 727 LIST_FOREACH(ia, &IN_IFADDR_HASH(ip->ip_dst.s_addr), ia_hash) { 728 if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { 729 if (checkif && ia->ia_ifp != m->m_pkthdr.rcvif) 730 continue; 731 if ((ia->ia_ifp->if_flags & IFF_UP) != 0) 732 break; 733 else 734 downmatch++; 735 } 736 } 737 if (ia != NULL) 738 goto ours; 739 if (m->m_pkthdr.rcvif && m->m_pkthdr.rcvif->if_flags & IFF_BROADCAST) { 740 IFADDR_FOREACH(ifa, m->m_pkthdr.rcvif) { 741 if (ifa->ifa_addr->sa_family != AF_INET) 742 continue; 743 ia = ifatoia(ifa); 744 if (in_hosteq(ip->ip_dst, ia->ia_broadaddr.sin_addr) || 745 in_hosteq(ip->ip_dst, ia->ia_netbroadcast) || 746 /* 747 * Look for all-0's host part (old broadcast addr), 748 * either for subnet or net. 749 */ 750 ip->ip_dst.s_addr == ia->ia_subnet || 751 ip->ip_dst.s_addr == ia->ia_net) 752 goto ours; 753 /* 754 * An interface with IP address zero accepts 755 * all packets that arrive on that interface. 756 */ 757 if (in_nullhost(ia->ia_addr.sin_addr)) 758 goto ours; 759 } 760 } 761 if (IN_MULTICAST(ip->ip_dst.s_addr)) { 762 struct in_multi *inm; 763#ifdef MROUTING 764 extern struct socket *ip_mrouter; 765 766 if (ip_mrouter) { 767 /* 768 * If we are acting as a multicast router, all 769 * incoming multicast packets are passed to the 770 * kernel-level multicast forwarding function. 771 * The packet is returned (relatively) intact; if 772 * ip_mforward() returns a non-zero value, the packet 773 * must be discarded, else it may be accepted below. 774 * 775 * (The IP ident field is put in the same byte order 776 * as expected when ip_mforward() is called from 777 * ip_output().) 778 */ 779 if (ip_mforward(m, m->m_pkthdr.rcvif) != 0) { 780 ipstat.ips_cantforward++; 781 m_freem(m); 782 return; 783 } 784 785 /* 786 * The process-level routing demon needs to receive 787 * all multicast IGMP packets, whether or not this 788 * host belongs to their destination groups. 789 */ 790 if (ip->ip_p == IPPROTO_IGMP) 791 goto ours; 792 ipstat.ips_forward++; 793 } 794#endif 795 /* 796 * See if we belong to the destination multicast group on the 797 * arrival interface. 798 */ 799 IN_LOOKUP_MULTI(ip->ip_dst, m->m_pkthdr.rcvif, inm); 800 if (inm == NULL) { 801 ipstat.ips_cantforward++; 802 m_freem(m); 803 return; 804 } 805 goto ours; 806 } 807 if (ip->ip_dst.s_addr == INADDR_BROADCAST || 808 in_nullhost(ip->ip_dst)) 809 goto ours; 810 811 /* 812 * Not for us; forward if possible and desirable. 813 */ 814 if (ipforwarding == 0) { 815 ipstat.ips_cantforward++; 816 m_freem(m); 817 } else { 818 /* 819 * If ip_dst matched any of my address on !IFF_UP interface, 820 * and there's no IFF_UP interface that matches ip_dst, 821 * send icmp unreach. Forwarding it will result in in-kernel 822 * forwarding loop till TTL goes to 0. 823 */ 824 if (downmatch) { 825 icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, 0, 0); 826 ipstat.ips_cantforward++; 827 return; 828 } 829#ifdef IPSEC 830 if (ipsec4_in_reject(m, NULL)) { 831 ipsecstat.in_polvio++; 832 goto bad; 833 } 834#endif 835#ifdef FAST_IPSEC 836 mtag = m_tag_find(m, PACKET_TAG_IPSEC_IN_DONE, NULL); 837 s = splsoftnet(); 838 if (mtag != NULL) { 839 tdbi = (struct tdb_ident *)(mtag + 1); 840 sp = ipsec_getpolicy(tdbi, IPSEC_DIR_INBOUND); 841 } else { 842 sp = ipsec_getpolicybyaddr(m, IPSEC_DIR_INBOUND, 843 IP_FORWARDING, &error); 844 } 845 if (sp == NULL) { /* NB: can happen if error */ 846 splx(s); 847 /*XXX error stat???*/ 848 DPRINTF(("ip_input: no SP for forwarding\n")); /*XXX*/ 849 goto bad; 850 } 851 852 /* 853 * Check security policy against packet attributes. 854 */ 855 error = ipsec_in_reject(sp, m); 856 KEY_FREESP(&sp); 857 splx(s); 858 if (error) { 859 ipstat.ips_cantforward++; 860 goto bad; 861 } 862 863 /* 864 * Peek at the outbound SP for this packet to determine if 865 * it's a Fast Forward candidate. 866 */ 867 mtag = m_tag_find(m, PACKET_TAG_IPSEC_PENDING_TDB, NULL); 868 if (mtag != NULL) 869 m->m_flags &= ~M_CANFASTFWD; 870 else { 871 s = splsoftnet(); 872 sp = ipsec4_checkpolicy(m, IPSEC_DIR_OUTBOUND, 873 (IP_FORWARDING | 874 (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), 875 &error, NULL); 876 if (sp != NULL) { 877 m->m_flags &= ~M_CANFASTFWD; 878 KEY_FREESP(&sp); 879 } 880 splx(s); 881 } 882#endif /* FAST_IPSEC */ 883 884 ip_forward(m, srcrt); 885 } 886 return; 887 888ours: 889 /* 890 * If offset or IP_MF are set, must reassemble. 891 * Otherwise, nothing need be done. 892 * (We could look in the reassembly queue to see 893 * if the packet was previously fragmented, 894 * but it's not worth the time; just let them time out.) 895 */ 896 if (ip->ip_off & ~htons(IP_DF|IP_RF)) { 897 898 /* 899 * Look for queue of fragments 900 * of this datagram. 901 */ 902 IPQ_LOCK(); 903 hash = IPREASS_HASH(ip->ip_src.s_addr, ip->ip_id); 904 /* XXX LIST_FOREACH(fp, &ipq[hash], ipq_q) */ 905 for (fp = LIST_FIRST(&ipq[hash]); fp != NULL; 906 fp = LIST_NEXT(fp, ipq_q)) { 907 if (ip->ip_id == fp->ipq_id && 908 in_hosteq(ip->ip_src, fp->ipq_src) && 909 in_hosteq(ip->ip_dst, fp->ipq_dst) && 910 ip->ip_p == fp->ipq_p) 911 goto found; 912 913 } 914 fp = 0; 915found: 916 917 /* 918 * Adjust ip_len to not reflect header, 919 * set ipqe_mff if more fragments are expected, 920 * convert offset of this to bytes. 921 */ 922 ip->ip_len = htons(ntohs(ip->ip_len) - hlen); 923 mff = (ip->ip_off & htons(IP_MF)) != 0; 924 if (mff) { 925 /* 926 * Make sure that fragments have a data length 927 * that's a non-zero multiple of 8 bytes. 928 */ 929 if (ntohs(ip->ip_len) == 0 || 930 (ntohs(ip->ip_len) & 0x7) != 0) { 931 ipstat.ips_badfrags++; 932 IPQ_UNLOCK(); 933 goto bad; 934 } 935 } 936 ip->ip_off = htons((ntohs(ip->ip_off) & IP_OFFMASK) << 3); 937 938 /* 939 * If datagram marked as having more fragments 940 * or if this is not the first fragment, 941 * attempt reassembly; if it succeeds, proceed. 942 */ 943 if (mff || ip->ip_off != htons(0)) { 944 ipstat.ips_fragments++; 945 ipqe = pool_get(&ipqent_pool, PR_NOWAIT); 946 if (ipqe == NULL) { 947 ipstat.ips_rcvmemdrop++; 948 IPQ_UNLOCK(); 949 goto bad; 950 } 951 ipqe->ipqe_mff = mff; 952 ipqe->ipqe_m = m; 953 ipqe->ipqe_ip = ip; 954 m = ip_reass(ipqe, fp, &ipq[hash]); 955 if (m == 0) { 956 IPQ_UNLOCK(); 957 return; 958 } 959 ipstat.ips_reassembled++; 960 ip = mtod(m, struct ip *); 961 hlen = ip->ip_hl << 2; 962 ip->ip_len = htons(ntohs(ip->ip_len) + hlen); 963 } else 964 if (fp) 965 ip_freef(fp); 966 IPQ_UNLOCK(); 967 } 968 969#if defined(IPSEC) 970 /* 971 * enforce IPsec policy checking if we are seeing last header. 972 * note that we do not visit this with protocols with pcb layer 973 * code - like udp/tcp/raw ip. 974 */ 975 if ((inetsw[ip_protox[ip->ip_p]].pr_flags & PR_LASTHDR) != 0 && 976 ipsec4_in_reject(m, NULL)) { 977 ipsecstat.in_polvio++; 978 goto bad; 979 } 980#endif 981#ifdef FAST_IPSEC 982 /* 983 * enforce IPsec policy checking if we are seeing last header. 984 * note that we do not visit this with protocols with pcb layer 985 * code - like udp/tcp/raw ip. 986 */ 987 if ((inetsw[ip_protox[ip->ip_p]].pr_flags & PR_LASTHDR) != 0) { 988 /* 989 * Check if the packet has already had IPsec processing 990 * done. If so, then just pass it along. This tag gets 991 * set during AH, ESP, etc. input handling, before the 992 * packet is returned to the ip input queue for delivery. 993 */ 994 mtag = m_tag_find(m, PACKET_TAG_IPSEC_IN_DONE, NULL); 995 s = splsoftnet(); 996 if (mtag != NULL) { 997 tdbi = (struct tdb_ident *)(mtag + 1); 998 sp = ipsec_getpolicy(tdbi, IPSEC_DIR_INBOUND); 999 } else { 1000 sp = ipsec_getpolicybyaddr(m, IPSEC_DIR_INBOUND, 1001 IP_FORWARDING, &error); 1002 } 1003 if (sp != NULL) { 1004 /* 1005 * Check security policy against packet attributes. 1006 */ 1007 error = ipsec_in_reject(sp, m); 1008 KEY_FREESP(&sp); 1009 } else { 1010 /* XXX error stat??? */ 1011 error = EINVAL; 1012DPRINTF(("ip_input: no SP, packet discarded\n"));/*XXX*/ 1013 goto bad; 1014 } 1015 splx(s); 1016 if (error) 1017 goto bad; 1018 } 1019#endif /* FAST_IPSEC */ 1020 1021 /* 1022 * Switch out to protocol's input routine. 1023 */ 1024#if IFA_STATS 1025 if (ia && ip) 1026 ia->ia_ifa.ifa_data.ifad_inbytes += ntohs(ip->ip_len); 1027#endif 1028 ipstat.ips_delivered++; 1029 { 1030 int off = hlen, nh = ip->ip_p; 1031 1032 (*inetsw[ip_protox[nh]].pr_input)(m, off, nh); 1033 return; 1034 } 1035bad: 1036 m_freem(m); 1037 return; 1038 1039badcsum: 1040 ipstat.ips_badsum++; 1041 m_freem(m); 1042} 1043 1044/* 1045 * Take incoming datagram fragment and try to 1046 * reassemble it into whole datagram. If a chain for 1047 * reassembly of this datagram already exists, then it 1048 * is given as fp; otherwise have to make a chain. 1049 */ 1050struct mbuf * 1051ip_reass(struct ipqent *ipqe, struct ipq *fp, struct ipqhead *ipqhead) 1052{ 1053 struct mbuf *m = ipqe->ipqe_m; 1054 struct ipqent *nq, *p, *q; 1055 struct ip *ip; 1056 struct mbuf *t; 1057 int hlen = ipqe->ipqe_ip->ip_hl << 2; 1058 int i, next; 1059 1060 IPQ_LOCK_CHECK(); 1061 1062 /* 1063 * Presence of header sizes in mbufs 1064 * would confuse code below. 1065 */ 1066 m->m_data += hlen; 1067 m->m_len -= hlen; 1068 1069#ifdef notyet 1070 /* make sure fragment limit is up-to-date */ 1071 CHECK_NMBCLUSTER_PARAMS(); 1072 1073 /* If we have too many fragments, drop the older half. */ 1074 if (ip_nfrags >= ip_maxfrags) 1075 ip_reass_drophalf(void); 1076#endif 1077 1078 /* 1079 * We are about to add a fragment; increment frag count. 1080 */ 1081 ip_nfrags++; 1082 1083 /* 1084 * If first fragment to arrive, create a reassembly queue. 1085 */ 1086 if (fp == 0) { 1087 /* 1088 * Enforce upper bound on number of fragmented packets 1089 * for which we attempt reassembly; 1090 * If maxfrag is 0, never accept fragments. 1091 * If maxfrag is -1, accept all fragments without limitation. 1092 */ 1093 if (ip_maxfragpackets < 0) 1094 ; 1095 else if (ip_nfragpackets >= ip_maxfragpackets) 1096 goto dropfrag; 1097 ip_nfragpackets++; 1098 MALLOC(fp, struct ipq *, sizeof (struct ipq), 1099 M_FTABLE, M_NOWAIT); 1100 if (fp == NULL) 1101 goto dropfrag; 1102 LIST_INSERT_HEAD(ipqhead, fp, ipq_q); 1103 fp->ipq_nfrags = 1; 1104 fp->ipq_ttl = IPFRAGTTL; 1105 fp->ipq_p = ipqe->ipqe_ip->ip_p; 1106 fp->ipq_id = ipqe->ipqe_ip->ip_id; 1107 TAILQ_INIT(&fp->ipq_fragq); 1108 fp->ipq_src = ipqe->ipqe_ip->ip_src; 1109 fp->ipq_dst = ipqe->ipqe_ip->ip_dst; 1110 p = NULL; 1111 goto insert; 1112 } else { 1113 fp->ipq_nfrags++; 1114 } 1115 1116 /* 1117 * Find a segment which begins after this one does. 1118 */ 1119 for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; 1120 p = q, q = TAILQ_NEXT(q, ipqe_q)) 1121 if (ntohs(q->ipqe_ip->ip_off) > ntohs(ipqe->ipqe_ip->ip_off)) 1122 break; 1123 1124 /* 1125 * If there is a preceding segment, it may provide some of 1126 * our data already. If so, drop the data from the incoming 1127 * segment. If it provides all of our data, drop us. 1128 */ 1129 if (p != NULL) { 1130 i = ntohs(p->ipqe_ip->ip_off) + ntohs(p->ipqe_ip->ip_len) - 1131 ntohs(ipqe->ipqe_ip->ip_off); 1132 if (i > 0) { 1133 if (i >= ntohs(ipqe->ipqe_ip->ip_len)) 1134 goto dropfrag; 1135 m_adj(ipqe->ipqe_m, i); 1136 ipqe->ipqe_ip->ip_off = 1137 htons(ntohs(ipqe->ipqe_ip->ip_off) + i); 1138 ipqe->ipqe_ip->ip_len = 1139 htons(ntohs(ipqe->ipqe_ip->ip_len) - i); 1140 } 1141 } 1142 1143 /* 1144 * While we overlap succeeding segments trim them or, 1145 * if they are completely covered, dequeue them. 1146 */ 1147 for (; q != NULL && 1148 ntohs(ipqe->ipqe_ip->ip_off) + ntohs(ipqe->ipqe_ip->ip_len) > 1149 ntohs(q->ipqe_ip->ip_off); q = nq) { 1150 i = (ntohs(ipqe->ipqe_ip->ip_off) + 1151 ntohs(ipqe->ipqe_ip->ip_len)) - ntohs(q->ipqe_ip->ip_off); 1152 if (i < ntohs(q->ipqe_ip->ip_len)) { 1153 q->ipqe_ip->ip_len = 1154 htons(ntohs(q->ipqe_ip->ip_len) - i); 1155 q->ipqe_ip->ip_off = 1156 htons(ntohs(q->ipqe_ip->ip_off) + i); 1157 m_adj(q->ipqe_m, i); 1158 break; 1159 } 1160 nq = TAILQ_NEXT(q, ipqe_q); 1161 m_freem(q->ipqe_m); 1162 TAILQ_REMOVE(&fp->ipq_fragq, q, ipqe_q); 1163 pool_put(&ipqent_pool, q); 1164 fp->ipq_nfrags--; 1165 ip_nfrags--; 1166 } 1167 1168insert: 1169 /* 1170 * Stick new segment in its place; 1171 * check for complete reassembly. 1172 */ 1173 if (p == NULL) { 1174 TAILQ_INSERT_HEAD(&fp->ipq_fragq, ipqe, ipqe_q); 1175 } else { 1176 TAILQ_INSERT_AFTER(&fp->ipq_fragq, p, ipqe, ipqe_q); 1177 } 1178 next = 0; 1179 for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; 1180 p = q, q = TAILQ_NEXT(q, ipqe_q)) { 1181 if (ntohs(q->ipqe_ip->ip_off) != next) 1182 return (0); 1183 next += ntohs(q->ipqe_ip->ip_len); 1184 } 1185 if (p->ipqe_mff) 1186 return (0); 1187 1188 /* 1189 * Reassembly is complete. Check for a bogus message size and 1190 * concatenate fragments. 1191 */ 1192 q = TAILQ_FIRST(&fp->ipq_fragq); 1193 ip = q->ipqe_ip; 1194 if ((next + (ip->ip_hl << 2)) > IP_MAXPACKET) { 1195 ipstat.ips_toolong++; 1196 ip_freef(fp); 1197 return (0); 1198 } 1199 m = q->ipqe_m; 1200 t = m->m_next; 1201 m->m_next = 0; 1202 m_cat(m, t); 1203 nq = TAILQ_NEXT(q, ipqe_q); 1204 pool_put(&ipqent_pool, q); 1205 for (q = nq; q != NULL; q = nq) { 1206 t = q->ipqe_m; 1207 nq = TAILQ_NEXT(q, ipqe_q); 1208 pool_put(&ipqent_pool, q); 1209 m_cat(m, t); 1210 } 1211 ip_nfrags -= fp->ipq_nfrags; 1212 1213 /* 1214 * Create header for new ip packet by 1215 * modifying header of first packet; 1216 * dequeue and discard fragment reassembly header. 1217 * Make header visible. 1218 */ 1219 ip->ip_len = htons(next); 1220 ip->ip_src = fp->ipq_src; 1221 ip->ip_dst = fp->ipq_dst; 1222 LIST_REMOVE(fp, ipq_q); 1223 FREE(fp, M_FTABLE); 1224 ip_nfragpackets--; 1225 m->m_len += (ip->ip_hl << 2); 1226 m->m_data -= (ip->ip_hl << 2); 1227 /* some debugging cruft by sklower, below, will go away soon */ 1228 if (m->m_flags & M_PKTHDR) { /* XXX this should be done elsewhere */ 1229 int plen = 0; 1230 for (t = m; t; t = t->m_next) 1231 plen += t->m_len; 1232 m->m_pkthdr.len = plen; 1233 m->m_pkthdr.csum_flags = 0; 1234 } 1235 return (m); 1236 1237dropfrag: 1238 if (fp != 0) 1239 fp->ipq_nfrags--; 1240 ip_nfrags--; 1241 ipstat.ips_fragdropped++; 1242 m_freem(m); 1243 pool_put(&ipqent_pool, ipqe); 1244 return (0); 1245} 1246 1247/* 1248 * Free a fragment reassembly header and all 1249 * associated datagrams. 1250 */ 1251void 1252ip_freef(struct ipq *fp) 1253{ 1254 struct ipqent *q, *p; 1255 u_int nfrags = 0; 1256 1257 IPQ_LOCK_CHECK(); 1258 1259 for (q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; q = p) { 1260 p = TAILQ_NEXT(q, ipqe_q); 1261 m_freem(q->ipqe_m); 1262 nfrags++; 1263 TAILQ_REMOVE(&fp->ipq_fragq, q, ipqe_q); 1264 pool_put(&ipqent_pool, q); 1265 } 1266 1267 if (nfrags != fp->ipq_nfrags) 1268 printf("ip_freef: nfrags %d != %d\n", fp->ipq_nfrags, nfrags); 1269 ip_nfrags -= nfrags; 1270 LIST_REMOVE(fp, ipq_q); 1271 FREE(fp, M_FTABLE); 1272 ip_nfragpackets--; 1273} 1274 1275/* 1276 * IP reassembly TTL machinery for multiplicative drop. 1277 */ 1278static u_int fragttl_histo[(IPFRAGTTL+1)]; 1279 1280 1281/* 1282 * Decrement TTL of all reasembly queue entries by `ticks'. 1283 * Count number of distinct fragments (as opposed to partial, fragmented 1284 * datagrams) in the reassembly queue. While we traverse the entire 1285 * reassembly queue, compute and return the median TTL over all fragments. 1286 */ 1287static u_int 1288ip_reass_ttl_decr(u_int ticks) 1289{ 1290 u_int nfrags, median, dropfraction, keepfraction; 1291 struct ipq *fp, *nfp; 1292 int i; 1293 1294 nfrags = 0; 1295 memset(fragttl_histo, 0, sizeof fragttl_histo); 1296 1297 for (i = 0; i < IPREASS_NHASH; i++) { 1298 for (fp = LIST_FIRST(&ipq[i]); fp != NULL; fp = nfp) { 1299 fp->ipq_ttl = ((fp->ipq_ttl <= ticks) ? 1300 0 : fp->ipq_ttl - ticks); 1301 nfp = LIST_NEXT(fp, ipq_q); 1302 if (fp->ipq_ttl == 0) { 1303 ipstat.ips_fragtimeout++; 1304 ip_freef(fp); 1305 } else { 1306 nfrags += fp->ipq_nfrags; 1307 fragttl_histo[fp->ipq_ttl] += fp->ipq_nfrags; 1308 } 1309 } 1310 } 1311 1312 KASSERT(ip_nfrags == nfrags); 1313 1314 /* Find median (or other drop fraction) in histogram. */ 1315 dropfraction = (ip_nfrags / 2); 1316 keepfraction = ip_nfrags - dropfraction; 1317 for (i = IPFRAGTTL, median = 0; i >= 0; i--) { 1318 median += fragttl_histo[i]; 1319 if (median >= keepfraction) 1320 break; 1321 } 1322 1323 /* Return TTL of median (or other fraction). */ 1324 return (u_int)i; 1325} 1326 1327void 1328ip_reass_drophalf(void) 1329{ 1330 1331 u_int median_ticks; 1332 /* 1333 * Compute median TTL of all fragments, and count frags 1334 * with that TTL or lower (roughly half of all fragments). 1335 */ 1336 median_ticks = ip_reass_ttl_decr(0); 1337 1338 /* Drop half. */ 1339 median_ticks = ip_reass_ttl_decr(median_ticks); 1340 1341} 1342 1343/* 1344 * IP timer processing; 1345 * if a timer expires on a reassembly 1346 * queue, discard it. 1347 */ 1348void 1349ip_slowtimo(void) 1350{ 1351 static u_int dropscanidx = 0; 1352 u_int i; 1353 u_int median_ttl; 1354 int s = splsoftnet(); 1355 1356 IPQ_LOCK(); 1357 1358 /* Age TTL of all fragments by 1 tick .*/ 1359 median_ttl = ip_reass_ttl_decr(1); 1360 1361 /* make sure fragment limit is up-to-date */ 1362 CHECK_NMBCLUSTER_PARAMS(); 1363 1364 /* If we have too many fragments, drop the older half. */ 1365 if (ip_nfrags > ip_maxfrags) 1366 ip_reass_ttl_decr(median_ttl); 1367 1368 /* 1369 * If we are over the maximum number of fragmented packets 1370 * (due to the limit being lowered), drain off 1371 * enough to get down to the new limit. Start draining 1372 * from the reassembly hashqueue most recently drained. 1373 */ 1374 if (ip_maxfragpackets < 0) 1375 ; 1376 else { 1377 int wrapped = 0; 1378 1379 i = dropscanidx; 1380 while (ip_nfragpackets > ip_maxfragpackets && wrapped == 0) { 1381 while (LIST_FIRST(&ipq[i]) != NULL) 1382 ip_freef(LIST_FIRST(&ipq[i])); 1383 if (++i >= IPREASS_NHASH) { 1384 i = 0; 1385 } 1386 /* 1387 * Dont scan forever even if fragment counters are 1388 * wrong: stop after scanning entire reassembly queue. 1389 */ 1390 if (i == dropscanidx) 1391 wrapped = 1; 1392 } 1393 dropscanidx = i; 1394 } 1395 IPQ_UNLOCK(); 1396#ifdef GATEWAY 1397 ipflow_slowtimo(); 1398#endif 1399 splx(s); 1400} 1401 1402/* 1403 * Drain off all datagram fragments. 1404 */ 1405void 1406ip_drain(void) 1407{ 1408 1409 /* 1410 * We may be called from a device's interrupt context. If 1411 * the ipq is already busy, just bail out now. 1412 */ 1413 if (ipq_lock_try() == 0) 1414 return; 1415 1416 /* 1417 * Drop half the total fragments now. If more mbufs are needed, 1418 * we will be called again soon. 1419 */ 1420 ip_reass_drophalf(); 1421 1422 IPQ_UNLOCK(); 1423} 1424 1425/* 1426 * Do option processing on a datagram, 1427 * possibly discarding it if bad options are encountered, 1428 * or forwarding it if source-routed. 1429 * Returns 1 if packet has been forwarded/freed, 1430 * 0 if the packet should be processed further. 1431 */ 1432int 1433ip_dooptions(struct mbuf *m) 1434{ 1435 struct ip *ip = mtod(m, struct ip *); 1436 u_char *cp, *cp0; 1437 struct ip_timestamp *ipt; 1438 struct in_ifaddr *ia; 1439 int opt, optlen, cnt, off, code, type = ICMP_PARAMPROB, forward = 0; 1440 struct in_addr dst; 1441 n_time ntime; 1442 1443 dst = ip->ip_dst; 1444 cp = (u_char *)(ip + 1); 1445 cnt = (ip->ip_hl << 2) - sizeof (struct ip); 1446 for (; cnt > 0; cnt -= optlen, cp += optlen) { 1447 opt = cp[IPOPT_OPTVAL]; 1448 if (opt == IPOPT_EOL) 1449 break; 1450 if (opt == IPOPT_NOP) 1451 optlen = 1; 1452 else { 1453 if (cnt < IPOPT_OLEN + sizeof(*cp)) { 1454 code = &cp[IPOPT_OLEN] - (u_char *)ip; 1455 goto bad; 1456 } 1457 optlen = cp[IPOPT_OLEN]; 1458 if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) { 1459 code = &cp[IPOPT_OLEN] - (u_char *)ip; 1460 goto bad; 1461 } 1462 } 1463 switch (opt) { 1464 1465 default: 1466 break; 1467 1468 /* 1469 * Source routing with record. 1470 * Find interface with current destination address. 1471 * If none on this machine then drop if strictly routed, 1472 * or do nothing if loosely routed. 1473 * Record interface address and bring up next address 1474 * component. If strictly routed make sure next 1475 * address is on directly accessible net. 1476 */ 1477 case IPOPT_LSRR: 1478 case IPOPT_SSRR: 1479 if (ip_allowsrcrt == 0) { 1480 type = ICMP_UNREACH; 1481 code = ICMP_UNREACH_NET_PROHIB; 1482 goto bad; 1483 } 1484 if (optlen < IPOPT_OFFSET + sizeof(*cp)) { 1485 code = &cp[IPOPT_OLEN] - (u_char *)ip; 1486 goto bad; 1487 } 1488 if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { 1489 code = &cp[IPOPT_OFFSET] - (u_char *)ip; 1490 goto bad; 1491 } 1492 ipaddr.sin_addr = ip->ip_dst; 1493 ia = ifatoia(ifa_ifwithaddr(sintosa(&ipaddr))); 1494 if (ia == 0) { 1495 if (opt == IPOPT_SSRR) { 1496 type = ICMP_UNREACH; 1497 code = ICMP_UNREACH_SRCFAIL; 1498 goto bad; 1499 } 1500 /* 1501 * Loose routing, and not at next destination 1502 * yet; nothing to do except forward. 1503 */ 1504 break; 1505 } 1506 off--; /* 0 origin */ 1507 if ((off + sizeof(struct in_addr)) > optlen) { 1508 /* 1509 * End of source route. Should be for us. 1510 */ 1511 save_rte(cp, ip->ip_src); 1512 break; 1513 } 1514 /* 1515 * locate outgoing interface 1516 */ 1517 bcopy((caddr_t)(cp + off), (caddr_t)&ipaddr.sin_addr, 1518 sizeof(ipaddr.sin_addr)); 1519 if (opt == IPOPT_SSRR) 1520 ia = ifatoia(ifa_ifwithladdr(sintosa(&ipaddr))); 1521 else 1522 ia = ip_rtaddr(ipaddr.sin_addr); 1523 if (ia == 0) { 1524 type = ICMP_UNREACH; 1525 code = ICMP_UNREACH_SRCFAIL; 1526 goto bad; 1527 } 1528 ip->ip_dst = ipaddr.sin_addr; 1529 bcopy((caddr_t)&ia->ia_addr.sin_addr, 1530 (caddr_t)(cp + off), sizeof(struct in_addr)); 1531 cp[IPOPT_OFFSET] += sizeof(struct in_addr); 1532 /* 1533 * Let ip_intr's mcast routing check handle mcast pkts 1534 */ 1535 forward = !IN_MULTICAST(ip->ip_dst.s_addr); 1536 break; 1537 1538 case IPOPT_RR: 1539 if (optlen < IPOPT_OFFSET + sizeof(*cp)) { 1540 code = &cp[IPOPT_OLEN] - (u_char *)ip; 1541 goto bad; 1542 } 1543 if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { 1544 code = &cp[IPOPT_OFFSET] - (u_char *)ip; 1545 goto bad; 1546 } 1547 /* 1548 * If no space remains, ignore. 1549 */ 1550 off--; /* 0 origin */ 1551 if ((off + sizeof(struct in_addr)) > optlen) 1552 break; 1553 bcopy((caddr_t)(&ip->ip_dst), (caddr_t)&ipaddr.sin_addr, 1554 sizeof(ipaddr.sin_addr)); 1555 /* 1556 * locate outgoing interface; if we're the destination, 1557 * use the incoming interface (should be same). 1558 */ 1559 if ((ia = ifatoia(ifa_ifwithaddr(sintosa(&ipaddr)))) 1560 == NULL && 1561 (ia = ip_rtaddr(ipaddr.sin_addr)) == NULL) { 1562 type = ICMP_UNREACH; 1563 code = ICMP_UNREACH_HOST; 1564 goto bad; 1565 } 1566 bcopy((caddr_t)&ia->ia_addr.sin_addr, 1567 (caddr_t)(cp + off), sizeof(struct in_addr)); 1568 cp[IPOPT_OFFSET] += sizeof(struct in_addr); 1569 break; 1570 1571 case IPOPT_TS: 1572 code = cp - (u_char *)ip; 1573 ipt = (struct ip_timestamp *)cp; 1574 if (ipt->ipt_len < 4 || ipt->ipt_len > 40) { 1575 code = (u_char *)&ipt->ipt_len - (u_char *)ip; 1576 goto bad; 1577 } 1578 if (ipt->ipt_ptr < 5) { 1579 code = (u_char *)&ipt->ipt_ptr - (u_char *)ip; 1580 goto bad; 1581 } 1582 if (ipt->ipt_ptr > ipt->ipt_len - sizeof (int32_t)) { 1583 if (++ipt->ipt_oflw == 0) { 1584 code = (u_char *)&ipt->ipt_ptr - 1585 (u_char *)ip; 1586 goto bad; 1587 } 1588 break; 1589 } 1590 cp0 = (cp + ipt->ipt_ptr - 1); 1591 switch (ipt->ipt_flg) { 1592 1593 case IPOPT_TS_TSONLY: 1594 break; 1595 1596 case IPOPT_TS_TSANDADDR: 1597 if (ipt->ipt_ptr - 1 + sizeof(n_time) + 1598 sizeof(struct in_addr) > ipt->ipt_len) { 1599 code = (u_char *)&ipt->ipt_ptr - 1600 (u_char *)ip; 1601 goto bad; 1602 } 1603 ipaddr.sin_addr = dst; 1604 ia = ifatoia(ifaof_ifpforaddr(sintosa(&ipaddr), 1605 m->m_pkthdr.rcvif)); 1606 if (ia == 0) 1607 continue; 1608 bcopy(&ia->ia_addr.sin_addr, 1609 cp0, sizeof(struct in_addr)); 1610 ipt->ipt_ptr += sizeof(struct in_addr); 1611 break; 1612 1613 case IPOPT_TS_PRESPEC: 1614 if (ipt->ipt_ptr - 1 + sizeof(n_time) + 1615 sizeof(struct in_addr) > ipt->ipt_len) { 1616 code = (u_char *)&ipt->ipt_ptr - 1617 (u_char *)ip; 1618 goto bad; 1619 } 1620 bcopy(cp0, &ipaddr.sin_addr, 1621 sizeof(struct in_addr)); 1622 if (ifatoia(ifa_ifwithaddr(sintosa(&ipaddr))) 1623 == NULL) 1624 continue; 1625 ipt->ipt_ptr += sizeof(struct in_addr); 1626 break; 1627 1628 default: 1629 /* XXX can't take &ipt->ipt_flg */ 1630 code = (u_char *)&ipt->ipt_ptr - 1631 (u_char *)ip + 1; 1632 goto bad; 1633 } 1634 ntime = iptime(); 1635 cp0 = (u_char *) &ntime; /* XXX grumble, GCC... */ 1636 bcopy(cp0, (caddr_t)cp + ipt->ipt_ptr - 1, 1637 sizeof(n_time)); 1638 ipt->ipt_ptr += sizeof(n_time); 1639 } 1640 } 1641 if (forward) { 1642 if (ip_forwsrcrt == 0) { 1643 type = ICMP_UNREACH; 1644 code = ICMP_UNREACH_SRCFAIL; 1645 goto bad; 1646 } 1647 ip_forward(m, 1); 1648 return (1); 1649 } 1650 return (0); 1651bad: 1652 icmp_error(m, type, code, 0, 0); 1653 ipstat.ips_badoptions++; 1654 return (1); 1655} 1656 1657/* 1658 * Given address of next destination (final or next hop), 1659 * return internet address info of interface to be used to get there. 1660 */ 1661struct in_ifaddr * 1662ip_rtaddr(struct in_addr dst) 1663{ 1664 struct sockaddr_in *sin; 1665 1666 sin = satosin(&ipforward_rt.ro_dst); 1667 1668 if (ipforward_rt.ro_rt == 0 || !in_hosteq(dst, sin->sin_addr)) { 1669 if (ipforward_rt.ro_rt) { 1670 RTFREE(ipforward_rt.ro_rt); 1671 ipforward_rt.ro_rt = 0; 1672 } 1673 sin->sin_family = AF_INET; 1674 sin->sin_len = sizeof(*sin); 1675 sin->sin_addr = dst; 1676 1677 rtalloc(&ipforward_rt); 1678 } 1679 if (ipforward_rt.ro_rt == 0) 1680 return ((struct in_ifaddr *)0); 1681 return (ifatoia(ipforward_rt.ro_rt->rt_ifa)); 1682} 1683 1684/* 1685 * Save incoming source route for use in replies, 1686 * to be picked up later by ip_srcroute if the receiver is interested. 1687 */ 1688void 1689save_rte(u_char *option, struct in_addr dst) 1690{ 1691 unsigned olen; 1692 1693 olen = option[IPOPT_OLEN]; 1694#ifdef DIAGNOSTIC 1695 if (ipprintfs) 1696 printf("save_rte: olen %d\n", olen); 1697#endif /* 0 */ 1698 if (olen > sizeof(ip_srcrt) - (1 + sizeof(dst))) 1699 return; 1700 bcopy((caddr_t)option, (caddr_t)ip_srcrt.srcopt, olen); 1701 ip_nhops = (olen - IPOPT_OFFSET - 1) / sizeof(struct in_addr); 1702 ip_srcrt.dst = dst; 1703} 1704 1705/* 1706 * Retrieve incoming source route for use in replies, 1707 * in the same form used by setsockopt. 1708 * The first hop is placed before the options, will be removed later. 1709 */ 1710struct mbuf * 1711ip_srcroute(void) 1712{ 1713 struct in_addr *p, *q; 1714 struct mbuf *m; 1715 1716 if (ip_nhops == 0) 1717 return ((struct mbuf *)0); 1718 m = m_get(M_DONTWAIT, MT_SOOPTS); 1719 if (m == 0) 1720 return ((struct mbuf *)0); 1721 1722 MCLAIM(m, &inetdomain.dom_mowner); 1723#define OPTSIZ (sizeof(ip_srcrt.nop) + sizeof(ip_srcrt.srcopt)) 1724 1725 /* length is (nhops+1)*sizeof(addr) + sizeof(nop + srcrt header) */ 1726 m->m_len = ip_nhops * sizeof(struct in_addr) + sizeof(struct in_addr) + 1727 OPTSIZ; 1728#ifdef DIAGNOSTIC 1729 if (ipprintfs) 1730 printf("ip_srcroute: nhops %d mlen %d", ip_nhops, m->m_len); 1731#endif 1732 1733 /* 1734 * First save first hop for return route 1735 */ 1736 p = &ip_srcrt.route[ip_nhops - 1]; 1737 *(mtod(m, struct in_addr *)) = *p--; 1738#ifdef DIAGNOSTIC 1739 if (ipprintfs) 1740 printf(" hops %x", ntohl(mtod(m, struct in_addr *)->s_addr)); 1741#endif 1742 1743 /* 1744 * Copy option fields and padding (nop) to mbuf. 1745 */ 1746 ip_srcrt.nop = IPOPT_NOP; 1747 ip_srcrt.srcopt[IPOPT_OFFSET] = IPOPT_MINOFF; 1748 bcopy((caddr_t)&ip_srcrt.nop, 1749 mtod(m, caddr_t) + sizeof(struct in_addr), OPTSIZ); 1750 q = (struct in_addr *)(mtod(m, caddr_t) + 1751 sizeof(struct in_addr) + OPTSIZ); 1752#undef OPTSIZ 1753 /* 1754 * Record return path as an IP source route, 1755 * reversing the path (pointers are now aligned). 1756 */ 1757 while (p >= ip_srcrt.route) { 1758#ifdef DIAGNOSTIC 1759 if (ipprintfs) 1760 printf(" %x", ntohl(q->s_addr)); 1761#endif 1762 *q++ = *p--; 1763 } 1764 /* 1765 * Last hop goes to final destination. 1766 */ 1767 *q = ip_srcrt.dst; 1768#ifdef DIAGNOSTIC 1769 if (ipprintfs) 1770 printf(" %x\n", ntohl(q->s_addr)); 1771#endif 1772 return (m); 1773} 1774 1775/* 1776 * Strip out IP options, at higher 1777 * level protocol in the kernel. 1778 * Second argument is buffer to which options 1779 * will be moved, and return value is their length. 1780 * XXX should be deleted; last arg currently ignored. 1781 */ 1782void 1783ip_stripoptions(struct mbuf *m, struct mbuf *mopt) 1784{ 1785 int i; 1786 struct ip *ip = mtod(m, struct ip *); 1787 caddr_t opts; 1788 int olen; 1789 1790 olen = (ip->ip_hl << 2) - sizeof (struct ip); 1791 opts = (caddr_t)(ip + 1); 1792 i = m->m_len - (sizeof (struct ip) + olen); 1793 bcopy(opts + olen, opts, (unsigned)i); 1794 m->m_len -= olen; 1795 if (m->m_flags & M_PKTHDR) 1796 m->m_pkthdr.len -= olen; 1797 ip->ip_len = htons(ntohs(ip->ip_len) - olen); 1798 ip->ip_hl = sizeof (struct ip) >> 2; 1799} 1800 1801const int inetctlerrmap[PRC_NCMDS] = { 1802 0, 0, 0, 0, 1803 0, EMSGSIZE, EHOSTDOWN, EHOSTUNREACH, 1804 EHOSTUNREACH, EHOSTUNREACH, ECONNREFUSED, ECONNREFUSED, 1805 EMSGSIZE, EHOSTUNREACH, 0, 0, 1806 0, 0, 0, 0, 1807 ENOPROTOOPT 1808}; 1809 1810/* 1811 * Forward a packet. If some error occurs return the sender 1812 * an icmp packet. Note we can't always generate a meaningful 1813 * icmp message because icmp doesn't have a large enough repertoire 1814 * of codes and types. 1815 * 1816 * If not forwarding, just drop the packet. This could be confusing 1817 * if ipforwarding was zero but some routing protocol was advancing 1818 * us as a gateway to somewhere. However, we must let the routing 1819 * protocol deal with that. 1820 * 1821 * The srcrt parameter indicates whether the packet is being forwarded 1822 * via a source route. 1823 */ 1824void 1825ip_forward(struct mbuf *m, int srcrt) 1826{ 1827 struct ip *ip = mtod(m, struct ip *); 1828 struct sockaddr_in *sin; 1829 struct rtentry *rt; 1830 int error, type = 0, code = 0, destmtu = 0; 1831 struct mbuf *mcopy; 1832 n_long dest; 1833 1834 /* 1835 * We are now in the output path. 1836 */ 1837 MCLAIM(m, &ip_tx_mowner); 1838 1839 /* 1840 * Clear any in-bound checksum flags for this packet. 1841 */ 1842 m->m_pkthdr.csum_flags = 0; 1843 1844 dest = 0; 1845#ifdef DIAGNOSTIC 1846 if (ipprintfs) { 1847 printf("forward: src %s ", inet_ntoa(ip->ip_src)); 1848 printf("dst %s ttl %x\n", inet_ntoa(ip->ip_dst), ip->ip_ttl); 1849 } 1850#endif 1851 if (m->m_flags & (M_BCAST|M_MCAST) || in_canforward(ip->ip_dst) == 0) { 1852 ipstat.ips_cantforward++; 1853 m_freem(m); 1854 return; 1855 } 1856 if (ip->ip_ttl <= IPTTLDEC) { 1857 icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS, dest, 0); 1858 return; 1859 } 1860 1861 sin = satosin(&ipforward_rt.ro_dst); 1862 if ((rt = ipforward_rt.ro_rt) == 0 || 1863 !in_hosteq(ip->ip_dst, sin->sin_addr)) { 1864 if (ipforward_rt.ro_rt) { 1865 RTFREE(ipforward_rt.ro_rt); 1866 ipforward_rt.ro_rt = 0; 1867 } 1868 sin->sin_family = AF_INET; 1869 sin->sin_len = sizeof(struct sockaddr_in); 1870 sin->sin_addr = ip->ip_dst; 1871 1872 rtalloc(&ipforward_rt); 1873 if (ipforward_rt.ro_rt == 0) { 1874 icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_NET, dest, 0); 1875 return; 1876 } 1877 rt = ipforward_rt.ro_rt; 1878 } 1879 1880 /* 1881 * Save at most 68 bytes of the packet in case 1882 * we need to generate an ICMP message to the src. 1883 * Pullup to avoid sharing mbuf cluster between m and mcopy. 1884 */ 1885 mcopy = m_copym(m, 0, imin(ntohs(ip->ip_len), 68), M_DONTWAIT); 1886 if (mcopy) 1887 mcopy = m_pullup(mcopy, ip->ip_hl << 2); 1888 1889 ip->ip_ttl -= IPTTLDEC; 1890 1891 /* 1892 * If forwarding packet using same interface that it came in on, 1893 * perhaps should send a redirect to sender to shortcut a hop. 1894 * Only send redirect if source is sending directly to us, 1895 * and if packet was not source routed (or has any options). 1896 * Also, don't send redirect if forwarding using a default route 1897 * or a route modified by a redirect. 1898 */ 1899 if (rt->rt_ifp == m->m_pkthdr.rcvif && 1900 (rt->rt_flags & (RTF_DYNAMIC|RTF_MODIFIED)) == 0 && 1901 !in_nullhost(satosin(rt_key(rt))->sin_addr) && 1902 ipsendredirects && !srcrt) { 1903 if (rt->rt_ifa && 1904 (ip->ip_src.s_addr & ifatoia(rt->rt_ifa)->ia_subnetmask) == 1905 ifatoia(rt->rt_ifa)->ia_subnet) { 1906 if (rt->rt_flags & RTF_GATEWAY) 1907 dest = satosin(rt->rt_gateway)->sin_addr.s_addr; 1908 else 1909 dest = ip->ip_dst.s_addr; 1910 /* 1911 * Router requirements says to only send host 1912 * redirects. 1913 */ 1914 type = ICMP_REDIRECT; 1915 code = ICMP_REDIRECT_HOST; 1916#ifdef DIAGNOSTIC 1917 if (ipprintfs) 1918 printf("redirect (%d) to %x\n", code, 1919 (u_int32_t)dest); 1920#endif 1921 } 1922 } 1923 1924 error = ip_output(m, (struct mbuf *)0, &ipforward_rt, 1925 (IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), 1926 (struct ip_moptions *)NULL, (struct socket *)NULL); 1927 1928 if (error) 1929 ipstat.ips_cantforward++; 1930 else { 1931 ipstat.ips_forward++; 1932 if (type) 1933 ipstat.ips_redirectsent++; 1934 else { 1935 if (mcopy) { 1936#ifdef GATEWAY 1937 if (mcopy->m_flags & M_CANFASTFWD) 1938 ipflow_create(&ipforward_rt, mcopy); 1939#endif 1940 m_freem(mcopy); 1941 } 1942 return; 1943 } 1944 } 1945 if (mcopy == NULL) 1946 return; 1947 1948 switch (error) { 1949 1950 case 0: /* forwarded, but need redirect */ 1951 /* type, code set above */ 1952 break; 1953 1954 case ENETUNREACH: /* shouldn't happen, checked above */ 1955 case EHOSTUNREACH: 1956 case ENETDOWN: 1957 case EHOSTDOWN: 1958 default: 1959 type = ICMP_UNREACH; 1960 code = ICMP_UNREACH_HOST; 1961 break; 1962 1963 case EMSGSIZE: 1964 type = ICMP_UNREACH; 1965 code = ICMP_UNREACH_NEEDFRAG; 1966#if !defined(IPSEC) && !defined(FAST_IPSEC) 1967 if (ipforward_rt.ro_rt) 1968 destmtu = ipforward_rt.ro_rt->rt_ifp->if_mtu; 1969#else 1970 /* 1971 * If the packet is routed over IPsec tunnel, tell the 1972 * originator the tunnel MTU. 1973 * tunnel MTU = if MTU - sizeof(IP) - ESP/AH hdrsiz 1974 * XXX quickhack!!! 1975 */ 1976 if (ipforward_rt.ro_rt) { 1977 struct secpolicy *sp; 1978 int ipsecerror; 1979 size_t ipsechdr; 1980 struct route *ro; 1981 1982 sp = ipsec4_getpolicybyaddr(mcopy, 1983 IPSEC_DIR_OUTBOUND, IP_FORWARDING, 1984 &ipsecerror); 1985 1986 if (sp == NULL) 1987 destmtu = ipforward_rt.ro_rt->rt_ifp->if_mtu; 1988 else { 1989 /* count IPsec header size */ 1990 ipsechdr = ipsec4_hdrsiz(mcopy, 1991 IPSEC_DIR_OUTBOUND, NULL); 1992 1993 /* 1994 * find the correct route for outer IPv4 1995 * header, compute tunnel MTU. 1996 */ 1997 1998 if (sp->req != NULL 1999 && sp->req->sav != NULL 2000 && sp->req->sav->sah != NULL) { 2001 ro = &sp->req->sav->sah->sa_route; 2002 if (ro->ro_rt && ro->ro_rt->rt_ifp) { 2003 destmtu = 2004 ro->ro_rt->rt_rmx.rmx_mtu ? 2005 ro->ro_rt->rt_rmx.rmx_mtu : 2006 ro->ro_rt->rt_ifp->if_mtu; 2007 destmtu -= ipsechdr; 2008 } 2009 } 2010 2011#ifdef IPSEC 2012 key_freesp(sp); 2013#else 2014 KEY_FREESP(&sp); 2015#endif 2016 } 2017 } 2018#endif /*IPSEC*/ 2019 ipstat.ips_cantfrag++; 2020 break; 2021 2022 case ENOBUFS: 2023#if 1 2024 /* 2025 * a router should not generate ICMP_SOURCEQUENCH as 2026 * required in RFC1812 Requirements for IP Version 4 Routers. 2027 * source quench could be a big problem under DoS attacks, 2028 * or if the underlying interface is rate-limited. 2029 */ 2030 if (mcopy) 2031 m_freem(mcopy); 2032 return; 2033#else 2034 type = ICMP_SOURCEQUENCH; 2035 code = 0; 2036 break; 2037#endif 2038 } 2039 icmp_error(mcopy, type, code, dest, destmtu); 2040} 2041 2042void 2043ip_savecontrol(struct inpcb *inp, struct mbuf **mp, struct ip *ip, 2044 struct mbuf *m) 2045{ 2046 2047 if (inp->inp_socket->so_options & SO_TIMESTAMP) { 2048 struct timeval tv; 2049 2050 microtime(&tv); 2051 *mp = sbcreatecontrol((caddr_t) &tv, sizeof(tv), 2052 SCM_TIMESTAMP, SOL_SOCKET); 2053 if (*mp) 2054 mp = &(*mp)->m_next; 2055 } 2056 if (inp->inp_flags & INP_RECVDSTADDR) { 2057 *mp = sbcreatecontrol((caddr_t) &ip->ip_dst, 2058 sizeof(struct in_addr), IP_RECVDSTADDR, IPPROTO_IP); 2059 if (*mp) 2060 mp = &(*mp)->m_next; 2061 } 2062#ifdef notyet 2063 /* 2064 * XXX 2065 * Moving these out of udp_input() made them even more broken 2066 * than they already were. 2067 * - fenner@parc.xerox.com 2068 */ 2069 /* options were tossed already */ 2070 if (inp->inp_flags & INP_RECVOPTS) { 2071 *mp = sbcreatecontrol((caddr_t) opts_deleted_above, 2072 sizeof(struct in_addr), IP_RECVOPTS, IPPROTO_IP); 2073 if (*mp) 2074 mp = &(*mp)->m_next; 2075 } 2076 /* ip_srcroute doesn't do what we want here, need to fix */ 2077 if (inp->inp_flags & INP_RECVRETOPTS) { 2078 *mp = sbcreatecontrol((caddr_t) ip_srcroute(), 2079 sizeof(struct in_addr), IP_RECVRETOPTS, IPPROTO_IP); 2080 if (*mp) 2081 mp = &(*mp)->m_next; 2082 } 2083#endif 2084 if (inp->inp_flags & INP_RECVIF) { 2085 struct sockaddr_dl sdl; 2086 2087 sdl.sdl_len = offsetof(struct sockaddr_dl, sdl_data[0]); 2088 sdl.sdl_family = AF_LINK; 2089 sdl.sdl_index = m->m_pkthdr.rcvif ? 2090 m->m_pkthdr.rcvif->if_index : 0; 2091 sdl.sdl_nlen = sdl.sdl_alen = sdl.sdl_slen = 0; 2092 *mp = sbcreatecontrol((caddr_t) &sdl, sdl.sdl_len, 2093 IP_RECVIF, IPPROTO_IP); 2094 if (*mp) 2095 mp = &(*mp)->m_next; 2096 } 2097} 2098 2099/* 2100 * sysctl helper routine for net.inet.ip.forwsrcrt. 2101 */ 2102static int 2103sysctl_net_inet_ip_forwsrcrt(SYSCTLFN_ARGS) 2104{ 2105 int error, tmp; 2106 struct sysctlnode node; 2107 2108 node = *rnode; 2109 tmp = ip_forwsrcrt; 2110 node.sysctl_data = &tmp; 2111 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2112 if (error || newp == NULL) 2113 return (error); 2114 2115 if (kauth_authorize_network(l->l_cred, KAUTH_NETWORK_FORWSRCRT, 2116 NULL, NULL, NULL, NULL) != KAUTH_RESULT_ALLOW) 2117 return (EPERM); 2118 2119 ip_forwsrcrt = tmp; 2120 2121 return (0); 2122} 2123 2124/* 2125 * sysctl helper routine for net.inet.ip.mtudisctimeout. checks the 2126 * range of the new value and tweaks timers if it changes. 2127 */ 2128static int 2129sysctl_net_inet_ip_pmtudto(SYSCTLFN_ARGS) 2130{ 2131 int error, tmp; 2132 struct sysctlnode node; 2133 2134 node = *rnode; 2135 tmp = ip_mtudisc_timeout; 2136 node.sysctl_data = &tmp; 2137 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2138 if (error || newp == NULL) 2139 return (error); 2140 if (tmp < 0) 2141 return (EINVAL); 2142 2143 ip_mtudisc_timeout = tmp; 2144 rt_timer_queue_change(ip_mtudisc_timeout_q, ip_mtudisc_timeout); 2145 2146 return (0); 2147} 2148 2149#ifdef GATEWAY 2150/* 2151 * sysctl helper routine for net.inet.ip.maxflows. apparently if 2152 * maxflows is even looked up, we "reap flows". 2153 */ 2154static int 2155sysctl_net_inet_ip_maxflows(SYSCTLFN_ARGS) 2156{ 2157 int s; 2158 2159 s = sysctl_lookup(SYSCTLFN_CALL(rnode)); 2160 if (s) 2161 return (s); 2162 2163 s = splsoftnet(); 2164 ipflow_reap(0); 2165 splx(s); 2166 2167 return (0); 2168} 2169#endif /* GATEWAY */ 2170 2171 2172SYSCTL_SETUP(sysctl_net_inet_ip_setup, "sysctl net.inet.ip subtree setup") 2173{ 2174 extern int subnetsarelocal, hostzeroisbroadcast; 2175 2176 sysctl_createv(clog, 0, NULL, NULL, 2177 CTLFLAG_PERMANENT, 2178 CTLTYPE_NODE, "net", NULL, 2179 NULL, 0, NULL, 0, 2180 CTL_NET, CTL_EOL); 2181 sysctl_createv(clog, 0, NULL, NULL, 2182 CTLFLAG_PERMANENT, 2183 CTLTYPE_NODE, "inet", 2184 SYSCTL_DESCR("PF_INET related settings"), 2185 NULL, 0, NULL, 0, 2186 CTL_NET, PF_INET, CTL_EOL); 2187 sysctl_createv(clog, 0, NULL, NULL, 2188 CTLFLAG_PERMANENT, 2189 CTLTYPE_NODE, "ip", 2190 SYSCTL_DESCR("IPv4 related settings"), 2191 NULL, 0, NULL, 0, 2192 CTL_NET, PF_INET, IPPROTO_IP, CTL_EOL); 2193 2194 sysctl_createv(clog, 0, NULL, NULL, 2195 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2196 CTLTYPE_INT, "forwarding", 2197 SYSCTL_DESCR("Enable forwarding of INET datagrams"), 2198 NULL, 0, &ipforwarding, 0, 2199 CTL_NET, PF_INET, IPPROTO_IP, 2200 IPCTL_FORWARDING, CTL_EOL); 2201 sysctl_createv(clog, 0, NULL, NULL, 2202 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2203 CTLTYPE_INT, "redirect", 2204 SYSCTL_DESCR("Enable sending of ICMP redirect messages"), 2205 NULL, 0, &ipsendredirects, 0, 2206 CTL_NET, PF_INET, IPPROTO_IP, 2207 IPCTL_SENDREDIRECTS, CTL_EOL); 2208 sysctl_createv(clog, 0, NULL, NULL, 2209 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2210 CTLTYPE_INT, "ttl", 2211 SYSCTL_DESCR("Default TTL for an INET datagram"), 2212 NULL, 0, &ip_defttl, 0, 2213 CTL_NET, PF_INET, IPPROTO_IP, 2214 IPCTL_DEFTTL, CTL_EOL); 2215#ifdef IPCTL_DEFMTU 2216 sysctl_createv(clog, 0, NULL, NULL, 2217 CTLFLAG_PERMANENT /* |CTLFLAG_READWRITE? */, 2218 CTLTYPE_INT, "mtu", 2219 SYSCTL_DESCR("Default MTA for an INET route"), 2220 NULL, 0, &ip_mtu, 0, 2221 CTL_NET, PF_INET, IPPROTO_IP, 2222 IPCTL_DEFMTU, CTL_EOL); 2223#endif /* IPCTL_DEFMTU */ 2224 sysctl_createv(clog, 0, NULL, NULL, 2225 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2226 CTLTYPE_INT, "forwsrcrt", 2227 SYSCTL_DESCR("Enable forwarding of source-routed " 2228 "datagrams"), 2229 sysctl_net_inet_ip_forwsrcrt, 0, &ip_forwsrcrt, 0, 2230 CTL_NET, PF_INET, IPPROTO_IP, 2231 IPCTL_FORWSRCRT, CTL_EOL); 2232 sysctl_createv(clog, 0, NULL, NULL, 2233 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2234 CTLTYPE_INT, "directed-broadcast", 2235 SYSCTL_DESCR("Enable forwarding of broadcast datagrams"), 2236 NULL, 0, &ip_directedbcast, 0, 2237 CTL_NET, PF_INET, IPPROTO_IP, 2238 IPCTL_DIRECTEDBCAST, CTL_EOL); 2239 sysctl_createv(clog, 0, NULL, NULL, 2240 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2241 CTLTYPE_INT, "allowsrcrt", 2242 SYSCTL_DESCR("Accept source-routed datagrams"), 2243 NULL, 0, &ip_allowsrcrt, 0, 2244 CTL_NET, PF_INET, IPPROTO_IP, 2245 IPCTL_ALLOWSRCRT, CTL_EOL); 2246 sysctl_createv(clog, 0, NULL, NULL, 2247 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2248 CTLTYPE_INT, "subnetsarelocal", 2249 SYSCTL_DESCR("Whether logical subnets are considered " 2250 "local"), 2251 NULL, 0, &subnetsarelocal, 0, 2252 CTL_NET, PF_INET, IPPROTO_IP, 2253 IPCTL_SUBNETSARELOCAL, CTL_EOL); 2254 sysctl_createv(clog, 0, NULL, NULL, 2255 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2256 CTLTYPE_INT, "mtudisc", 2257 SYSCTL_DESCR("Use RFC1191 Path MTU Discovery"), 2258 NULL, 0, &ip_mtudisc, 0, 2259 CTL_NET, PF_INET, IPPROTO_IP, 2260 IPCTL_MTUDISC, CTL_EOL); 2261 sysctl_createv(clog, 0, NULL, NULL, 2262 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2263 CTLTYPE_INT, "anonportmin", 2264 SYSCTL_DESCR("Lowest ephemeral port number to assign"), 2265 sysctl_net_inet_ip_ports, 0, &anonportmin, 0, 2266 CTL_NET, PF_INET, IPPROTO_IP, 2267 IPCTL_ANONPORTMIN, CTL_EOL); 2268 sysctl_createv(clog, 0, NULL, NULL, 2269 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2270 CTLTYPE_INT, "anonportmax", 2271 SYSCTL_DESCR("Highest ephemeral port number to assign"), 2272 sysctl_net_inet_ip_ports, 0, &anonportmax, 0, 2273 CTL_NET, PF_INET, IPPROTO_IP, 2274 IPCTL_ANONPORTMAX, CTL_EOL); 2275 sysctl_createv(clog, 0, NULL, NULL, 2276 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2277 CTLTYPE_INT, "mtudisctimeout", 2278 SYSCTL_DESCR("Lifetime of a Path MTU Discovered route"), 2279 sysctl_net_inet_ip_pmtudto, 0, &ip_mtudisc_timeout, 0, 2280 CTL_NET, PF_INET, IPPROTO_IP, 2281 IPCTL_MTUDISCTIMEOUT, CTL_EOL); 2282#ifdef GATEWAY 2283 sysctl_createv(clog, 0, NULL, NULL, 2284 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2285 CTLTYPE_INT, "maxflows", 2286 SYSCTL_DESCR("Number of flows for fast forwarding"), 2287 sysctl_net_inet_ip_maxflows, 0, &ip_maxflows, 0, 2288 CTL_NET, PF_INET, IPPROTO_IP, 2289 IPCTL_MAXFLOWS, CTL_EOL); 2290#endif /* GATEWAY */ 2291 sysctl_createv(clog, 0, NULL, NULL, 2292 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2293 CTLTYPE_INT, "hostzerobroadcast", 2294 SYSCTL_DESCR("All zeroes address is broadcast address"), 2295 NULL, 0, &hostzeroisbroadcast, 0, 2296 CTL_NET, PF_INET, IPPROTO_IP, 2297 IPCTL_HOSTZEROBROADCAST, CTL_EOL); 2298#if NGIF > 0 2299 sysctl_createv(clog, 0, NULL, NULL, 2300 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2301 CTLTYPE_INT, "gifttl", 2302 SYSCTL_DESCR("Default TTL for a gif tunnel datagram"), 2303 NULL, 0, &ip_gif_ttl, 0, 2304 CTL_NET, PF_INET, IPPROTO_IP, 2305 IPCTL_GIF_TTL, CTL_EOL); 2306#endif /* NGIF */ 2307#ifndef IPNOPRIVPORTS 2308 sysctl_createv(clog, 0, NULL, NULL, 2309 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2310 CTLTYPE_INT, "lowportmin", 2311 SYSCTL_DESCR("Lowest privileged ephemeral port number " 2312 "to assign"), 2313 sysctl_net_inet_ip_ports, 0, &lowportmin, 0, 2314 CTL_NET, PF_INET, IPPROTO_IP, 2315 IPCTL_LOWPORTMIN, CTL_EOL); 2316 sysctl_createv(clog, 0, NULL, NULL, 2317 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2318 CTLTYPE_INT, "lowportmax", 2319 SYSCTL_DESCR("Highest privileged ephemeral port number " 2320 "to assign"), 2321 sysctl_net_inet_ip_ports, 0, &lowportmax, 0, 2322 CTL_NET, PF_INET, IPPROTO_IP, 2323 IPCTL_LOWPORTMAX, CTL_EOL); 2324#endif /* IPNOPRIVPORTS */ 2325 sysctl_createv(clog, 0, NULL, NULL, 2326 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2327 CTLTYPE_INT, "maxfragpackets", 2328 SYSCTL_DESCR("Maximum number of fragments to retain for " 2329 "possible reassembly"), 2330 NULL, 0, &ip_maxfragpackets, 0, 2331 CTL_NET, PF_INET, IPPROTO_IP, 2332 IPCTL_MAXFRAGPACKETS, CTL_EOL); 2333#if NGRE > 0 2334 sysctl_createv(clog, 0, NULL, NULL, 2335 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2336 CTLTYPE_INT, "grettl", 2337 SYSCTL_DESCR("Default TTL for a gre tunnel datagram"), 2338 NULL, 0, &ip_gre_ttl, 0, 2339 CTL_NET, PF_INET, IPPROTO_IP, 2340 IPCTL_GRE_TTL, CTL_EOL); 2341#endif /* NGRE */ 2342 sysctl_createv(clog, 0, NULL, NULL, 2343 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2344 CTLTYPE_INT, "checkinterface", 2345 SYSCTL_DESCR("Enable receive side of Strong ES model " 2346 "from RFC1122"), 2347 NULL, 0, &ip_checkinterface, 0, 2348 CTL_NET, PF_INET, IPPROTO_IP, 2349 IPCTL_CHECKINTERFACE, CTL_EOL); 2350 sysctl_createv(clog, 0, NULL, NULL, 2351 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2352 CTLTYPE_INT, "random_id", 2353 SYSCTL_DESCR("Assign random ip_id values"), 2354 NULL, 0, &ip_do_randomid, 0, 2355 CTL_NET, PF_INET, IPPROTO_IP, 2356 IPCTL_RANDOMID, CTL_EOL); 2357 sysctl_createv(clog, 0, NULL, NULL, 2358 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2359 CTLTYPE_INT, "do_loopback_cksum", 2360 SYSCTL_DESCR("Perform IP checksum on loopback"), 2361 NULL, 0, &ip_do_loopback_cksum, 0, 2362 CTL_NET, PF_INET, IPPROTO_IP, 2363 IPCTL_LOOPBACKCKSUM, CTL_EOL); 2364 sysctl_createv(clog, 0, NULL, NULL, 2365 CTLFLAG_PERMANENT, 2366 CTLTYPE_STRUCT, "stats", 2367 SYSCTL_DESCR("IP statistics"), 2368 NULL, 0, &ipstat, sizeof(ipstat), 2369 CTL_NET, PF_INET, IPPROTO_IP, IPCTL_STATS, 2370 CTL_EOL); 2371} 2372