1 /* $NetBSD: ip_output.c,v 1.304 2018/04/29 11:51:08 maxv 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 * 49 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 50 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 51 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 52 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 53 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 54 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 55 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 56 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 57 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 58 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 59 * POSSIBILITY OF SUCH DAMAGE. 60 */ 61 62 /* 63 * Copyright (c) 1982, 1986, 1988, 1990, 1993 64 * The Regents of the University of California. All rights reserved. 65 * 66 * Redistribution and use in source and binary forms, with or without 67 * modification, are permitted provided that the following conditions 68 * are met: 69 * 1. Redistributions of source code must retain the above copyright 70 * notice, this list of conditions and the following disclaimer. 71 * 2. Redistributions in binary form must reproduce the above copyright 72 * notice, this list of conditions and the following disclaimer in the 73 * documentation and/or other materials provided with the distribution. 74 * 3. Neither the name of the University nor the names of its contributors 75 * may be used to endorse or promote products derived from this software 76 * without specific prior written permission. 77 * 78 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 79 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 80 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 81 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 82 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 83 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 84 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 85 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 86 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 87 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 88 * SUCH DAMAGE. 89 * 90 * @(#)ip_output.c 8.3 (Berkeley) 1/21/94 91 */ 92 93 #include <sys/cdefs.h> 94 __KERNEL_RCSID(0, "$NetBSD: ip_output.c,v 1.304 2018/04/29 11:51:08 maxv Exp $"); 95 96 #ifdef _KERNEL_OPT 97 #include "opt_inet.h" 98 #include "opt_ipsec.h" 99 #include "opt_mrouting.h" 100 #include "opt_net_mpsafe.h" 101 #include "opt_mpls.h" 102 #endif 103 104 #include "arp.h" 105 106 #include <sys/param.h> 107 #include <sys/kmem.h> 108 #include <sys/mbuf.h> 109 #include <sys/socket.h> 110 #include <sys/socketvar.h> 111 #include <sys/kauth.h> 112 #include <sys/systm.h> 113 #include <sys/syslog.h> 114 115 #include <net/if.h> 116 #include <net/if_types.h> 117 #include <net/route.h> 118 #include <net/pfil.h> 119 120 #include <netinet/in.h> 121 #include <netinet/in_systm.h> 122 #include <netinet/ip.h> 123 #include <netinet/in_pcb.h> 124 #include <netinet/in_var.h> 125 #include <netinet/ip_var.h> 126 #include <netinet/ip_private.h> 127 #include <netinet/in_offload.h> 128 #include <netinet/portalgo.h> 129 #include <netinet/udp.h> 130 #include <netinet/udp_var.h> 131 132 #ifdef INET6 133 #include <netinet6/ip6_var.h> 134 #endif 135 136 #ifdef MROUTING 137 #include <netinet/ip_mroute.h> 138 #endif 139 140 #ifdef IPSEC 141 #include <netipsec/ipsec.h> 142 #include <netipsec/key.h> 143 #endif 144 145 #ifdef MPLS 146 #include <netmpls/mpls.h> 147 #include <netmpls/mpls_var.h> 148 #endif 149 150 static int ip_pcbopts(struct inpcb *, const struct sockopt *); 151 static struct mbuf *ip_insertoptions(struct mbuf *, struct mbuf *, int *); 152 static struct ifnet *ip_multicast_if(struct in_addr *, int *); 153 static void ip_mloopback(struct ifnet *, struct mbuf *, 154 const struct sockaddr_in *); 155 static int ip_ifaddrvalid(const struct in_ifaddr *); 156 157 extern pfil_head_t *inet_pfil_hook; /* XXX */ 158 159 int ip_do_loopback_cksum = 0; 160 161 static int 162 ip_mark_mpls(struct ifnet * const ifp, struct mbuf * const m, 163 const struct rtentry *rt) 164 { 165 int error = 0; 166 #ifdef MPLS 167 union mpls_shim msh; 168 169 if (rt == NULL || rt_gettag(rt) == NULL || 170 rt_gettag(rt)->sa_family != AF_MPLS || 171 (m->m_flags & (M_MCAST | M_BCAST)) != 0 || 172 ifp->if_type != IFT_ETHER) 173 return 0; 174 175 msh.s_addr = MPLS_GETSADDR(rt); 176 if (msh.shim.label != MPLS_LABEL_IMPLNULL) { 177 struct m_tag *mtag; 178 /* 179 * XXX tentative solution to tell ether_output 180 * it's MPLS. Need some more efficient solution. 181 */ 182 mtag = m_tag_get(PACKET_TAG_MPLS, 183 sizeof(int) /* dummy */, 184 M_NOWAIT); 185 if (mtag == NULL) 186 return ENOMEM; 187 m_tag_prepend(m, mtag); 188 } 189 #endif 190 return error; 191 } 192 193 /* 194 * Send an IP packet to a host. 195 */ 196 int 197 ip_if_output(struct ifnet * const ifp, struct mbuf * const m, 198 const struct sockaddr * const dst, const struct rtentry *rt) 199 { 200 int error = 0; 201 202 if (rt != NULL) { 203 error = rt_check_reject_route(rt, ifp); 204 if (error != 0) { 205 m_freem(m); 206 return error; 207 } 208 } 209 210 error = ip_mark_mpls(ifp, m, rt); 211 if (error != 0) { 212 m_freem(m); 213 return error; 214 } 215 216 error = if_output_lock(ifp, ifp, m, dst, rt); 217 218 return error; 219 } 220 221 /* 222 * IP output. The packet in mbuf chain m contains a skeletal IP 223 * header (with len, off, ttl, proto, tos, src, dst). 224 * The mbuf chain containing the packet will be freed. 225 * The mbuf opt, if present, will not be freed. 226 */ 227 int 228 ip_output(struct mbuf *m0, struct mbuf *opt, struct route *ro, int flags, 229 struct ip_moptions *imo, struct inpcb *inp) 230 { 231 struct rtentry *rt; 232 struct ip *ip; 233 struct ifnet *ifp, *mifp = NULL; 234 struct mbuf *m = m0; 235 int len, hlen, error = 0; 236 struct route iproute; 237 const struct sockaddr_in *dst; 238 struct in_ifaddr *ia = NULL; 239 struct ifaddr *ifa; 240 int isbroadcast; 241 int sw_csum; 242 u_long mtu; 243 bool natt_frag = false; 244 bool rtmtu_nolock; 245 union { 246 struct sockaddr sa; 247 struct sockaddr_in sin; 248 } udst, usrc; 249 struct sockaddr *rdst = &udst.sa; /* real IP destination, as 250 * opposed to the nexthop 251 */ 252 struct psref psref, psref_ia; 253 int bound; 254 bool bind_need_restore = false; 255 256 len = 0; 257 258 MCLAIM(m, &ip_tx_mowner); 259 260 KASSERT((m->m_flags & M_PKTHDR) != 0); 261 KASSERT((m->m_pkthdr.csum_flags & (M_CSUM_TCPv6|M_CSUM_UDPv6)) == 0); 262 KASSERT((m->m_pkthdr.csum_flags & (M_CSUM_TCPv4|M_CSUM_UDPv4)) != 263 (M_CSUM_TCPv4|M_CSUM_UDPv4)); 264 KASSERT(m->m_len >= sizeof(struct ip)); 265 266 hlen = sizeof(struct ip); 267 if (opt) { 268 m = ip_insertoptions(m, opt, &len); 269 hlen = len; 270 } 271 ip = mtod(m, struct ip *); 272 273 /* 274 * Fill in IP header. 275 */ 276 if ((flags & (IP_FORWARDING|IP_RAWOUTPUT)) == 0) { 277 ip->ip_v = IPVERSION; 278 ip->ip_off = htons(0); 279 /* ip->ip_id filled in after we find out source ia */ 280 ip->ip_hl = hlen >> 2; 281 IP_STATINC(IP_STAT_LOCALOUT); 282 } else { 283 hlen = ip->ip_hl << 2; 284 } 285 286 /* 287 * Route packet. 288 */ 289 if (ro == NULL) { 290 memset(&iproute, 0, sizeof(iproute)); 291 ro = &iproute; 292 } 293 sockaddr_in_init(&udst.sin, &ip->ip_dst, 0); 294 dst = satocsin(rtcache_getdst(ro)); 295 296 /* 297 * If there is a cached route, check that it is to the same 298 * destination and is still up. If not, free it and try again. 299 * The address family should also be checked in case of sharing 300 * the cache with IPv6. 301 */ 302 if (dst && (dst->sin_family != AF_INET || 303 !in_hosteq(dst->sin_addr, ip->ip_dst))) 304 rtcache_free(ro); 305 306 /* XXX must be before rtcache operations */ 307 bound = curlwp_bind(); 308 bind_need_restore = true; 309 310 if ((rt = rtcache_validate(ro)) == NULL && 311 (rt = rtcache_update(ro, 1)) == NULL) { 312 dst = &udst.sin; 313 error = rtcache_setdst(ro, &udst.sa); 314 if (error != 0) 315 goto bad; 316 } 317 318 /* 319 * If routing to interface only, short circuit routing lookup. 320 */ 321 if (flags & IP_ROUTETOIF) { 322 ifa = ifa_ifwithladdr_psref(sintocsa(dst), &psref_ia); 323 if (ifa == NULL) { 324 IP_STATINC(IP_STAT_NOROUTE); 325 error = ENETUNREACH; 326 goto bad; 327 } 328 /* ia is already referenced by psref_ia */ 329 ia = ifatoia(ifa); 330 331 ifp = ia->ia_ifp; 332 mtu = ifp->if_mtu; 333 ip->ip_ttl = 1; 334 isbroadcast = in_broadcast(dst->sin_addr, ifp); 335 } else if (((IN_MULTICAST(ip->ip_dst.s_addr) || 336 ip->ip_dst.s_addr == INADDR_BROADCAST) || 337 (flags & IP_ROUTETOIFINDEX)) && 338 imo != NULL && imo->imo_multicast_if_index != 0) { 339 ifp = mifp = if_get_byindex(imo->imo_multicast_if_index, &psref); 340 if (ifp == NULL) { 341 IP_STATINC(IP_STAT_NOROUTE); 342 error = ENETUNREACH; 343 goto bad; 344 } 345 mtu = ifp->if_mtu; 346 ia = in_get_ia_from_ifp_psref(ifp, &psref_ia); 347 if (ia == NULL) { 348 error = EADDRNOTAVAIL; 349 goto bad; 350 } 351 if (IN_MULTICAST(ip->ip_dst.s_addr) || 352 ip->ip_dst.s_addr == INADDR_BROADCAST) { 353 isbroadcast = 0; 354 } else { 355 /* IP_ROUTETOIFINDEX */ 356 isbroadcast = in_broadcast(dst->sin_addr, ifp); 357 if ((isbroadcast == 0) && ((ifp->if_flags & 358 (IFF_LOOPBACK | IFF_POINTOPOINT)) == 0) && 359 (in_direct(dst->sin_addr, ifp) == 0)) { 360 /* gateway address required */ 361 if (rt == NULL) 362 rt = rtcache_init(ro); 363 if (rt == NULL || rt->rt_ifp != ifp) { 364 IP_STATINC(IP_STAT_NOROUTE); 365 error = EHOSTUNREACH; 366 goto bad; 367 } 368 rt->rt_use++; 369 if (rt->rt_flags & RTF_GATEWAY) 370 dst = satosin(rt->rt_gateway); 371 if (rt->rt_flags & RTF_HOST) 372 isbroadcast = 373 rt->rt_flags & RTF_BROADCAST; 374 } 375 } 376 } else { 377 if (rt == NULL) 378 rt = rtcache_init(ro); 379 if (rt == NULL) { 380 IP_STATINC(IP_STAT_NOROUTE); 381 error = EHOSTUNREACH; 382 goto bad; 383 } 384 if (ifa_is_destroying(rt->rt_ifa)) { 385 rtcache_unref(rt, ro); 386 rt = NULL; 387 IP_STATINC(IP_STAT_NOROUTE); 388 error = EHOSTUNREACH; 389 goto bad; 390 } 391 ifa_acquire(rt->rt_ifa, &psref_ia); 392 ia = ifatoia(rt->rt_ifa); 393 ifp = rt->rt_ifp; 394 if ((mtu = rt->rt_rmx.rmx_mtu) == 0) 395 mtu = ifp->if_mtu; 396 rt->rt_use++; 397 if (rt->rt_flags & RTF_GATEWAY) 398 dst = satosin(rt->rt_gateway); 399 if (rt->rt_flags & RTF_HOST) 400 isbroadcast = rt->rt_flags & RTF_BROADCAST; 401 else 402 isbroadcast = in_broadcast(dst->sin_addr, ifp); 403 } 404 rtmtu_nolock = rt && (rt->rt_rmx.rmx_locks & RTV_MTU) == 0; 405 406 if (IN_MULTICAST(ip->ip_dst.s_addr) || 407 (ip->ip_dst.s_addr == INADDR_BROADCAST)) { 408 bool inmgroup; 409 410 m->m_flags |= (ip->ip_dst.s_addr == INADDR_BROADCAST) ? 411 M_BCAST : M_MCAST; 412 /* 413 * See if the caller provided any multicast options 414 */ 415 if (imo != NULL) 416 ip->ip_ttl = imo->imo_multicast_ttl; 417 else 418 ip->ip_ttl = IP_DEFAULT_MULTICAST_TTL; 419 420 /* 421 * if we don't know the outgoing ifp yet, we can't generate 422 * output 423 */ 424 if (!ifp) { 425 IP_STATINC(IP_STAT_NOROUTE); 426 error = ENETUNREACH; 427 goto bad; 428 } 429 430 /* 431 * If the packet is multicast or broadcast, confirm that 432 * the outgoing interface can transmit it. 433 */ 434 if (((m->m_flags & M_MCAST) && 435 (ifp->if_flags & IFF_MULTICAST) == 0) || 436 ((m->m_flags & M_BCAST) && 437 (ifp->if_flags & (IFF_BROADCAST|IFF_POINTOPOINT)) == 0)) { 438 IP_STATINC(IP_STAT_NOROUTE); 439 error = ENETUNREACH; 440 goto bad; 441 } 442 /* 443 * If source address not specified yet, use an address 444 * of outgoing interface. 445 */ 446 if (in_nullhost(ip->ip_src)) { 447 struct in_ifaddr *xia; 448 struct ifaddr *xifa; 449 struct psref _psref; 450 451 xia = in_get_ia_from_ifp_psref(ifp, &_psref); 452 if (!xia) { 453 error = EADDRNOTAVAIL; 454 goto bad; 455 } 456 xifa = &xia->ia_ifa; 457 if (xifa->ifa_getifa != NULL) { 458 ia4_release(xia, &_psref); 459 /* FIXME ifa_getifa is NOMPSAFE */ 460 xia = ifatoia((*xifa->ifa_getifa)(xifa, rdst)); 461 if (xia == NULL) { 462 error = EADDRNOTAVAIL; 463 goto bad; 464 } 465 ia4_acquire(xia, &_psref); 466 } 467 ip->ip_src = xia->ia_addr.sin_addr; 468 ia4_release(xia, &_psref); 469 } 470 471 inmgroup = in_multi_group(ip->ip_dst, ifp, flags); 472 if (inmgroup && (imo == NULL || imo->imo_multicast_loop)) { 473 /* 474 * If we belong to the destination multicast group 475 * on the outgoing interface, and the caller did not 476 * forbid loopback, loop back a copy. 477 */ 478 ip_mloopback(ifp, m, &udst.sin); 479 } 480 #ifdef MROUTING 481 else { 482 /* 483 * If we are acting as a multicast router, perform 484 * multicast forwarding as if the packet had just 485 * arrived on the interface to which we are about 486 * to send. The multicast forwarding function 487 * recursively calls this function, using the 488 * IP_FORWARDING flag to prevent infinite recursion. 489 * 490 * Multicasts that are looped back by ip_mloopback(), 491 * above, will be forwarded by the ip_input() routine, 492 * if necessary. 493 */ 494 extern struct socket *ip_mrouter; 495 496 if (ip_mrouter && (flags & IP_FORWARDING) == 0) { 497 if (ip_mforward(m, ifp) != 0) { 498 m_freem(m); 499 goto done; 500 } 501 } 502 } 503 #endif 504 /* 505 * Multicasts with a time-to-live of zero may be looped- 506 * back, above, but must not be transmitted on a network. 507 * Also, multicasts addressed to the loopback interface 508 * are not sent -- the above call to ip_mloopback() will 509 * loop back a copy if this host actually belongs to the 510 * destination group on the loopback interface. 511 */ 512 if (ip->ip_ttl == 0 || (ifp->if_flags & IFF_LOOPBACK) != 0) { 513 m_freem(m); 514 goto done; 515 } 516 goto sendit; 517 } 518 519 /* 520 * If source address not specified yet, use address 521 * of outgoing interface. 522 */ 523 if (in_nullhost(ip->ip_src)) { 524 struct ifaddr *xifa; 525 526 xifa = &ia->ia_ifa; 527 if (xifa->ifa_getifa != NULL) { 528 ia4_release(ia, &psref_ia); 529 /* FIXME ifa_getifa is NOMPSAFE */ 530 ia = ifatoia((*xifa->ifa_getifa)(xifa, rdst)); 531 if (ia == NULL) { 532 error = EADDRNOTAVAIL; 533 goto bad; 534 } 535 ia4_acquire(ia, &psref_ia); 536 } 537 ip->ip_src = ia->ia_addr.sin_addr; 538 } 539 540 /* 541 * Packets with Class-D address as source are not valid per 542 * RFC1112. 543 */ 544 if (IN_MULTICAST(ip->ip_src.s_addr)) { 545 IP_STATINC(IP_STAT_ODROPPED); 546 error = EADDRNOTAVAIL; 547 goto bad; 548 } 549 550 /* 551 * Look for broadcast address and verify user is allowed to 552 * send such a packet. 553 */ 554 if (isbroadcast) { 555 if ((ifp->if_flags & IFF_BROADCAST) == 0) { 556 error = EADDRNOTAVAIL; 557 goto bad; 558 } 559 if ((flags & IP_ALLOWBROADCAST) == 0) { 560 error = EACCES; 561 goto bad; 562 } 563 /* don't allow broadcast messages to be fragmented */ 564 if (ntohs(ip->ip_len) > ifp->if_mtu) { 565 error = EMSGSIZE; 566 goto bad; 567 } 568 m->m_flags |= M_BCAST; 569 } else 570 m->m_flags &= ~M_BCAST; 571 572 sendit: 573 if ((flags & (IP_FORWARDING|IP_NOIPNEWID)) == 0) { 574 if (m->m_pkthdr.len < IP_MINFRAGSIZE) { 575 ip->ip_id = 0; 576 } else if ((m->m_pkthdr.csum_flags & M_CSUM_TSOv4) == 0) { 577 ip->ip_id = ip_newid(ia); 578 } else { 579 /* 580 * TSO capable interfaces (typically?) increment 581 * ip_id for each segment. 582 * "allocate" enough ids here to increase the chance 583 * for them to be unique. 584 * 585 * note that the following calculation is not 586 * needed to be precise. wasting some ip_id is fine. 587 */ 588 589 unsigned int segsz = m->m_pkthdr.segsz; 590 unsigned int datasz = ntohs(ip->ip_len) - hlen; 591 unsigned int num = howmany(datasz, segsz); 592 593 ip->ip_id = ip_newid_range(ia, num); 594 } 595 } 596 if (ia != NULL) { 597 ia4_release(ia, &psref_ia); 598 ia = NULL; 599 } 600 601 /* 602 * If we're doing Path MTU Discovery, we need to set DF unless 603 * the route's MTU is locked. 604 */ 605 if ((flags & IP_MTUDISC) != 0 && rtmtu_nolock) { 606 ip->ip_off |= htons(IP_DF); 607 } 608 609 #ifdef IPSEC 610 if (ipsec_used) { 611 bool ipsec_done = false; 612 613 /* Perform IPsec processing, if any. */ 614 error = ipsec4_output(m, inp, flags, &mtu, &natt_frag, 615 &ipsec_done); 616 if (error || ipsec_done) 617 goto done; 618 } 619 #endif 620 621 /* 622 * Run through list of hooks for output packets. 623 */ 624 error = pfil_run_hooks(inet_pfil_hook, &m, ifp, PFIL_OUT); 625 if (error) 626 goto done; 627 if (m == NULL) 628 goto done; 629 630 ip = mtod(m, struct ip *); 631 hlen = ip->ip_hl << 2; 632 633 m->m_pkthdr.csum_data |= hlen << 16; 634 635 /* 636 * search for the source address structure to 637 * maintain output statistics, and verify address 638 * validity 639 */ 640 KASSERT(ia == NULL); 641 sockaddr_in_init(&usrc.sin, &ip->ip_src, 0); 642 ifa = ifaof_ifpforaddr_psref(&usrc.sa, ifp, &psref_ia); 643 if (ifa != NULL) 644 ia = ifatoia(ifa); 645 646 /* 647 * Ensure we only send from a valid address. 648 * A NULL address is valid because the packet could be 649 * generated from a packet filter. 650 */ 651 if (ia != NULL && (flags & IP_FORWARDING) == 0 && 652 (error = ip_ifaddrvalid(ia)) != 0) 653 { 654 ARPLOG(LOG_ERR, 655 "refusing to send from invalid address %s (pid %d)\n", 656 ARPLOGADDR(&ip->ip_src), curproc->p_pid); 657 IP_STATINC(IP_STAT_ODROPPED); 658 if (error == 1) 659 /* 660 * Address exists, but is tentative or detached. 661 * We can't send from it because it's invalid, 662 * so we drop the packet. 663 */ 664 error = 0; 665 else 666 error = EADDRNOTAVAIL; 667 goto bad; 668 } 669 670 /* Maybe skip checksums on loopback interfaces. */ 671 if (IN_NEED_CHECKSUM(ifp, M_CSUM_IPv4)) { 672 m->m_pkthdr.csum_flags |= M_CSUM_IPv4; 673 } 674 sw_csum = m->m_pkthdr.csum_flags & ~ifp->if_csum_flags_tx; 675 676 /* 677 * If small enough for mtu of path, or if using TCP segmentation 678 * offload, can just send directly. 679 */ 680 if (ntohs(ip->ip_len) <= mtu || 681 (m->m_pkthdr.csum_flags & M_CSUM_TSOv4) != 0) { 682 const struct sockaddr *sa; 683 684 #if IFA_STATS 685 if (ia) 686 ia->ia_ifa.ifa_data.ifad_outbytes += ntohs(ip->ip_len); 687 #endif 688 /* 689 * Always initialize the sum to 0! Some HW assisted 690 * checksumming requires this. 691 */ 692 ip->ip_sum = 0; 693 694 if ((m->m_pkthdr.csum_flags & M_CSUM_TSOv4) == 0) { 695 /* 696 * Perform any checksums that the hardware can't do 697 * for us. 698 * 699 * XXX Does any hardware require the {th,uh}_sum 700 * XXX fields to be 0? 701 */ 702 if (sw_csum & M_CSUM_IPv4) { 703 KASSERT(IN_NEED_CHECKSUM(ifp, M_CSUM_IPv4)); 704 ip->ip_sum = in_cksum(m, hlen); 705 m->m_pkthdr.csum_flags &= ~M_CSUM_IPv4; 706 } 707 if (sw_csum & (M_CSUM_TCPv4|M_CSUM_UDPv4)) { 708 if (IN_NEED_CHECKSUM(ifp, 709 sw_csum & (M_CSUM_TCPv4|M_CSUM_UDPv4))) { 710 in_delayed_cksum(m); 711 } 712 m->m_pkthdr.csum_flags &= 713 ~(M_CSUM_TCPv4|M_CSUM_UDPv4); 714 } 715 } 716 717 sa = (m->m_flags & M_MCAST) ? sintocsa(rdst) : sintocsa(dst); 718 if (__predict_true( 719 (m->m_pkthdr.csum_flags & M_CSUM_TSOv4) == 0 || 720 (ifp->if_capenable & IFCAP_TSOv4) != 0)) { 721 error = ip_if_output(ifp, m, sa, rt); 722 } else { 723 error = ip_tso_output(ifp, m, sa, rt); 724 } 725 goto done; 726 } 727 728 /* 729 * We can't use HW checksumming if we're about to fragment the packet. 730 * 731 * XXX Some hardware can do this. 732 */ 733 if (m->m_pkthdr.csum_flags & (M_CSUM_TCPv4|M_CSUM_UDPv4)) { 734 if (IN_NEED_CHECKSUM(ifp, 735 m->m_pkthdr.csum_flags & (M_CSUM_TCPv4|M_CSUM_UDPv4))) { 736 in_delayed_cksum(m); 737 } 738 m->m_pkthdr.csum_flags &= ~(M_CSUM_TCPv4|M_CSUM_UDPv4); 739 } 740 741 /* 742 * Too large for interface; fragment if possible. 743 * Must be able to put at least 8 bytes per fragment. 744 */ 745 if (ntohs(ip->ip_off) & IP_DF) { 746 if (flags & IP_RETURNMTU) { 747 KASSERT(inp != NULL); 748 inp->inp_errormtu = mtu; 749 } 750 error = EMSGSIZE; 751 IP_STATINC(IP_STAT_CANTFRAG); 752 goto bad; 753 } 754 755 error = ip_fragment(m, ifp, mtu); 756 if (error) { 757 m = NULL; 758 goto bad; 759 } 760 761 for (; m; m = m0) { 762 m0 = m->m_nextpkt; 763 m->m_nextpkt = NULL; 764 if (error) { 765 m_freem(m); 766 continue; 767 } 768 #if IFA_STATS 769 if (ia) 770 ia->ia_ifa.ifa_data.ifad_outbytes += ntohs(ip->ip_len); 771 #endif 772 /* 773 * If we get there, the packet has not been handled by 774 * IPsec whereas it should have. Now that it has been 775 * fragmented, re-inject it in ip_output so that IPsec 776 * processing can occur. 777 */ 778 if (natt_frag) { 779 error = ip_output(m, opt, ro, 780 flags | IP_RAWOUTPUT | IP_NOIPNEWID, 781 imo, inp); 782 } else { 783 KASSERT((m->m_pkthdr.csum_flags & 784 (M_CSUM_UDPv4 | M_CSUM_TCPv4)) == 0); 785 error = ip_if_output(ifp, m, 786 (m->m_flags & M_MCAST) ? 787 sintocsa(rdst) : sintocsa(dst), rt); 788 } 789 } 790 if (error == 0) { 791 IP_STATINC(IP_STAT_FRAGMENTED); 792 } 793 794 done: 795 ia4_release(ia, &psref_ia); 796 rtcache_unref(rt, ro); 797 if (ro == &iproute) { 798 rtcache_free(&iproute); 799 } 800 if (mifp != NULL) { 801 if_put(mifp, &psref); 802 } 803 if (bind_need_restore) 804 curlwp_bindx(bound); 805 return error; 806 807 bad: 808 m_freem(m); 809 goto done; 810 } 811 812 int 813 ip_fragment(struct mbuf *m, struct ifnet *ifp, u_long mtu) 814 { 815 struct ip *ip, *mhip; 816 struct mbuf *m0; 817 int len, hlen, off; 818 int mhlen, firstlen; 819 struct mbuf **mnext; 820 int sw_csum = m->m_pkthdr.csum_flags; 821 int fragments = 0; 822 int error = 0; 823 int ipoff; 824 bool mff; 825 826 ip = mtod(m, struct ip *); 827 hlen = ip->ip_hl << 2; 828 829 /* XXX: Why don't we remove IP_RF? */ 830 ipoff = ntohs(ip->ip_off) & ~IP_MF; 831 832 mff = (ip->ip_off & htons(IP_MF)) != 0; 833 834 if (ifp != NULL) 835 sw_csum &= ~ifp->if_csum_flags_tx; 836 837 len = (mtu - hlen) &~ 7; 838 if (len < 8) { 839 m_freem(m); 840 return EMSGSIZE; 841 } 842 843 firstlen = len; 844 mnext = &m->m_nextpkt; 845 846 /* 847 * Loop through length of segment after first fragment, 848 * make new header and copy data of each part and link onto chain. 849 */ 850 m0 = m; 851 mhlen = sizeof(struct ip); 852 for (off = hlen + len; off < ntohs(ip->ip_len); off += len) { 853 MGETHDR(m, M_DONTWAIT, MT_HEADER); 854 if (m == NULL) { 855 error = ENOBUFS; 856 IP_STATINC(IP_STAT_ODROPPED); 857 goto sendorfree; 858 } 859 MCLAIM(m, m0->m_owner); 860 861 *mnext = m; 862 mnext = &m->m_nextpkt; 863 864 m->m_data += max_linkhdr; 865 mhip = mtod(m, struct ip *); 866 *mhip = *ip; 867 868 /* we must inherit MCAST and BCAST flags */ 869 m->m_flags |= m0->m_flags & (M_MCAST|M_BCAST); 870 871 if (hlen > sizeof(struct ip)) { 872 mhlen = ip_optcopy(ip, mhip) + sizeof(struct ip); 873 mhip->ip_hl = mhlen >> 2; 874 } 875 m->m_len = mhlen; 876 877 mhip->ip_off = ((off - hlen) >> 3) + ipoff; 878 if (mff) 879 mhip->ip_off |= IP_MF; 880 if (off + len >= ntohs(ip->ip_len)) 881 len = ntohs(ip->ip_len) - off; 882 else 883 mhip->ip_off |= IP_MF; 884 HTONS(mhip->ip_off); 885 886 mhip->ip_len = htons((u_int16_t)(len + mhlen)); 887 m->m_next = m_copym(m0, off, len, M_DONTWAIT); 888 if (m->m_next == NULL) { 889 error = ENOBUFS; 890 IP_STATINC(IP_STAT_ODROPPED); 891 goto sendorfree; 892 } 893 894 m->m_pkthdr.len = mhlen + len; 895 m_reset_rcvif(m); 896 897 mhip->ip_sum = 0; 898 KASSERT((m->m_pkthdr.csum_flags & M_CSUM_IPv4) == 0); 899 if (sw_csum & M_CSUM_IPv4) { 900 mhip->ip_sum = in_cksum(m, mhlen); 901 } else { 902 /* 903 * checksum is hw-offloaded or not necessary. 904 */ 905 m->m_pkthdr.csum_flags |= 906 m0->m_pkthdr.csum_flags & M_CSUM_IPv4; 907 m->m_pkthdr.csum_data |= mhlen << 16; 908 KASSERT(!(ifp != NULL && 909 IN_NEED_CHECKSUM(ifp, M_CSUM_IPv4)) || 910 (m->m_pkthdr.csum_flags & M_CSUM_IPv4) != 0); 911 } 912 IP_STATINC(IP_STAT_OFRAGMENTS); 913 fragments++; 914 } 915 916 /* 917 * Update first fragment by trimming what's been copied out 918 * and updating header, then send each fragment (in order). 919 */ 920 m = m0; 921 m_adj(m, hlen + firstlen - ntohs(ip->ip_len)); 922 m->m_pkthdr.len = hlen + firstlen; 923 ip->ip_len = htons((u_int16_t)m->m_pkthdr.len); 924 ip->ip_off |= htons(IP_MF); 925 ip->ip_sum = 0; 926 if (sw_csum & M_CSUM_IPv4) { 927 ip->ip_sum = in_cksum(m, hlen); 928 m->m_pkthdr.csum_flags &= ~M_CSUM_IPv4; 929 } else { 930 /* 931 * checksum is hw-offloaded or not necessary. 932 */ 933 KASSERT(!(ifp != NULL && IN_NEED_CHECKSUM(ifp, M_CSUM_IPv4)) || 934 (m->m_pkthdr.csum_flags & M_CSUM_IPv4) != 0); 935 KASSERT(M_CSUM_DATA_IPv4_IPHL(m->m_pkthdr.csum_data) >= 936 sizeof(struct ip)); 937 } 938 939 sendorfree: 940 /* 941 * If there is no room for all the fragments, don't queue 942 * any of them. 943 */ 944 if (ifp != NULL) { 945 IFQ_LOCK(&ifp->if_snd); 946 if (ifp->if_snd.ifq_maxlen - ifp->if_snd.ifq_len < fragments && 947 error == 0) { 948 error = ENOBUFS; 949 IP_STATINC(IP_STAT_ODROPPED); 950 IFQ_INC_DROPS(&ifp->if_snd); 951 } 952 IFQ_UNLOCK(&ifp->if_snd); 953 } 954 if (error) { 955 for (m = m0; m; m = m0) { 956 m0 = m->m_nextpkt; 957 m->m_nextpkt = NULL; 958 m_freem(m); 959 } 960 } 961 962 return error; 963 } 964 965 /* 966 * Process a delayed payload checksum calculation. 967 */ 968 void 969 in_delayed_cksum(struct mbuf *m) 970 { 971 struct ip *ip; 972 u_int16_t csum, offset; 973 974 ip = mtod(m, struct ip *); 975 offset = ip->ip_hl << 2; 976 csum = in4_cksum(m, 0, offset, ntohs(ip->ip_len) - offset); 977 if (csum == 0 && (m->m_pkthdr.csum_flags & M_CSUM_UDPv4) != 0) 978 csum = 0xffff; 979 980 offset += M_CSUM_DATA_IPv4_OFFSET(m->m_pkthdr.csum_data); 981 982 if ((offset + sizeof(u_int16_t)) > m->m_len) { 983 /* This happens when ip options were inserted */ 984 m_copyback(m, offset, sizeof(csum), (void *)&csum); 985 } else { 986 *(u_int16_t *)(mtod(m, char *) + offset) = csum; 987 } 988 } 989 990 /* 991 * Determine the maximum length of the options to be inserted; 992 * we would far rather allocate too much space rather than too little. 993 */ 994 u_int 995 ip_optlen(struct inpcb *inp) 996 { 997 struct mbuf *m = inp->inp_options; 998 999 if (m && m->m_len > offsetof(struct ipoption, ipopt_dst)) { 1000 return (m->m_len - offsetof(struct ipoption, ipopt_dst)); 1001 } 1002 return 0; 1003 } 1004 1005 /* 1006 * Insert IP options into preformed packet. 1007 * Adjust IP destination as required for IP source routing, 1008 * as indicated by a non-zero in_addr at the start of the options. 1009 */ 1010 static struct mbuf * 1011 ip_insertoptions(struct mbuf *m, struct mbuf *opt, int *phlen) 1012 { 1013 struct ipoption *p = mtod(opt, struct ipoption *); 1014 struct mbuf *n; 1015 struct ip *ip = mtod(m, struct ip *); 1016 unsigned optlen; 1017 1018 optlen = opt->m_len - sizeof(p->ipopt_dst); 1019 KASSERT(optlen % 4 == 0); 1020 if (optlen + ntohs(ip->ip_len) > IP_MAXPACKET) 1021 return m; /* XXX should fail */ 1022 if (!in_nullhost(p->ipopt_dst)) 1023 ip->ip_dst = p->ipopt_dst; 1024 if (M_READONLY(m) || M_LEADINGSPACE(m) < optlen) { 1025 MGETHDR(n, M_DONTWAIT, MT_HEADER); 1026 if (n == NULL) 1027 return m; 1028 MCLAIM(n, m->m_owner); 1029 M_MOVE_PKTHDR(n, m); 1030 m->m_len -= sizeof(struct ip); 1031 m->m_data += sizeof(struct ip); 1032 n->m_next = m; 1033 n->m_len = optlen + sizeof(struct ip); 1034 n->m_data += max_linkhdr; 1035 memcpy(mtod(n, void *), ip, sizeof(struct ip)); 1036 m = n; 1037 } else { 1038 m->m_data -= optlen; 1039 m->m_len += optlen; 1040 memmove(mtod(m, void *), ip, sizeof(struct ip)); 1041 } 1042 m->m_pkthdr.len += optlen; 1043 ip = mtod(m, struct ip *); 1044 memcpy(ip + 1, p->ipopt_list, optlen); 1045 *phlen = sizeof(struct ip) + optlen; 1046 ip->ip_len = htons(ntohs(ip->ip_len) + optlen); 1047 return m; 1048 } 1049 1050 /* 1051 * Copy options from ipsrc to ipdst, omitting those not copied during 1052 * fragmentation. 1053 */ 1054 int 1055 ip_optcopy(struct ip *ipsrc, struct ip *ipdst) 1056 { 1057 u_char *cp, *dp; 1058 int opt, optlen, cnt; 1059 1060 cp = (u_char *)(ipsrc + 1); 1061 dp = (u_char *)(ipdst + 1); 1062 cnt = (ipsrc->ip_hl << 2) - sizeof(struct ip); 1063 for (; cnt > 0; cnt -= optlen, cp += optlen) { 1064 opt = cp[0]; 1065 if (opt == IPOPT_EOL) 1066 break; 1067 if (opt == IPOPT_NOP) { 1068 /* Preserve for IP mcast tunnel's LSRR alignment. */ 1069 *dp++ = IPOPT_NOP; 1070 optlen = 1; 1071 continue; 1072 } 1073 1074 KASSERT(cnt >= IPOPT_OLEN + sizeof(*cp)); 1075 optlen = cp[IPOPT_OLEN]; 1076 KASSERT(optlen >= IPOPT_OLEN + sizeof(*cp) && optlen < cnt); 1077 1078 /* Invalid lengths should have been caught by ip_dooptions. */ 1079 if (optlen > cnt) 1080 optlen = cnt; 1081 if (IPOPT_COPIED(opt)) { 1082 bcopy((void *)cp, (void *)dp, (unsigned)optlen); 1083 dp += optlen; 1084 } 1085 } 1086 1087 for (optlen = dp - (u_char *)(ipdst+1); optlen & 0x3; optlen++) { 1088 *dp++ = IPOPT_EOL; 1089 } 1090 1091 return optlen; 1092 } 1093 1094 /* 1095 * IP socket option processing. 1096 */ 1097 int 1098 ip_ctloutput(int op, struct socket *so, struct sockopt *sopt) 1099 { 1100 struct inpcb *inp = sotoinpcb(so); 1101 struct ip *ip = &inp->inp_ip; 1102 int inpflags = inp->inp_flags; 1103 int optval = 0, error = 0; 1104 struct in_pktinfo pktinfo; 1105 1106 KASSERT(solocked(so)); 1107 1108 if (sopt->sopt_level != IPPROTO_IP) { 1109 if (sopt->sopt_level == SOL_SOCKET && sopt->sopt_name == SO_NOHEADER) 1110 return 0; 1111 return ENOPROTOOPT; 1112 } 1113 1114 switch (op) { 1115 case PRCO_SETOPT: 1116 switch (sopt->sopt_name) { 1117 case IP_OPTIONS: 1118 #ifdef notyet 1119 case IP_RETOPTS: 1120 #endif 1121 error = ip_pcbopts(inp, sopt); 1122 break; 1123 1124 case IP_TOS: 1125 case IP_TTL: 1126 case IP_MINTTL: 1127 case IP_RECVOPTS: 1128 case IP_RECVRETOPTS: 1129 case IP_RECVDSTADDR: 1130 case IP_RECVIF: 1131 case IP_RECVPKTINFO: 1132 case IP_RECVTTL: 1133 error = sockopt_getint(sopt, &optval); 1134 if (error) 1135 break; 1136 1137 switch (sopt->sopt_name) { 1138 case IP_TOS: 1139 ip->ip_tos = optval; 1140 break; 1141 1142 case IP_TTL: 1143 ip->ip_ttl = optval; 1144 break; 1145 1146 case IP_MINTTL: 1147 if (optval > 0 && optval <= MAXTTL) 1148 inp->inp_ip_minttl = optval; 1149 else 1150 error = EINVAL; 1151 break; 1152 #define OPTSET(bit) \ 1153 if (optval) \ 1154 inpflags |= bit; \ 1155 else \ 1156 inpflags &= ~bit; 1157 1158 case IP_RECVOPTS: 1159 OPTSET(INP_RECVOPTS); 1160 break; 1161 1162 case IP_RECVPKTINFO: 1163 OPTSET(INP_RECVPKTINFO); 1164 break; 1165 1166 case IP_RECVRETOPTS: 1167 OPTSET(INP_RECVRETOPTS); 1168 break; 1169 1170 case IP_RECVDSTADDR: 1171 OPTSET(INP_RECVDSTADDR); 1172 break; 1173 1174 case IP_RECVIF: 1175 OPTSET(INP_RECVIF); 1176 break; 1177 1178 case IP_RECVTTL: 1179 OPTSET(INP_RECVTTL); 1180 break; 1181 } 1182 break; 1183 case IP_PKTINFO: 1184 error = sockopt_getint(sopt, &optval); 1185 if (!error) { 1186 /* Linux compatibility */ 1187 OPTSET(INP_RECVPKTINFO); 1188 break; 1189 } 1190 error = sockopt_get(sopt, &pktinfo, sizeof(pktinfo)); 1191 if (error) 1192 break; 1193 1194 if (pktinfo.ipi_ifindex == 0) { 1195 inp->inp_prefsrcip = pktinfo.ipi_addr; 1196 break; 1197 } 1198 1199 /* Solaris compatibility */ 1200 struct ifnet *ifp; 1201 struct in_ifaddr *ia; 1202 int s; 1203 1204 /* pick up primary address */ 1205 s = pserialize_read_enter(); 1206 ifp = if_byindex(pktinfo.ipi_ifindex); 1207 if (ifp == NULL) { 1208 pserialize_read_exit(s); 1209 error = EADDRNOTAVAIL; 1210 break; 1211 } 1212 ia = in_get_ia_from_ifp(ifp); 1213 if (ia == NULL) { 1214 pserialize_read_exit(s); 1215 error = EADDRNOTAVAIL; 1216 break; 1217 } 1218 inp->inp_prefsrcip = IA_SIN(ia)->sin_addr; 1219 pserialize_read_exit(s); 1220 break; 1221 break; 1222 #undef OPTSET 1223 1224 case IP_MULTICAST_IF: 1225 case IP_MULTICAST_TTL: 1226 case IP_MULTICAST_LOOP: 1227 case IP_ADD_MEMBERSHIP: 1228 case IP_DROP_MEMBERSHIP: 1229 error = ip_setmoptions(&inp->inp_moptions, sopt); 1230 break; 1231 1232 case IP_PORTRANGE: 1233 error = sockopt_getint(sopt, &optval); 1234 if (error) 1235 break; 1236 1237 switch (optval) { 1238 case IP_PORTRANGE_DEFAULT: 1239 case IP_PORTRANGE_HIGH: 1240 inpflags &= ~(INP_LOWPORT); 1241 break; 1242 1243 case IP_PORTRANGE_LOW: 1244 inpflags |= INP_LOWPORT; 1245 break; 1246 1247 default: 1248 error = EINVAL; 1249 break; 1250 } 1251 break; 1252 1253 case IP_PORTALGO: 1254 error = sockopt_getint(sopt, &optval); 1255 if (error) 1256 break; 1257 1258 error = portalgo_algo_index_select( 1259 (struct inpcb_hdr *)inp, optval); 1260 break; 1261 1262 #if defined(IPSEC) 1263 case IP_IPSEC_POLICY: 1264 if (ipsec_enabled) { 1265 error = ipsec_set_policy(inp, 1266 sopt->sopt_data, sopt->sopt_size, 1267 curlwp->l_cred); 1268 break; 1269 } 1270 /*FALLTHROUGH*/ 1271 #endif /* IPSEC */ 1272 1273 default: 1274 error = ENOPROTOOPT; 1275 break; 1276 } 1277 break; 1278 1279 case PRCO_GETOPT: 1280 switch (sopt->sopt_name) { 1281 case IP_OPTIONS: 1282 case IP_RETOPTS: { 1283 struct mbuf *mopts = inp->inp_options; 1284 1285 if (mopts) { 1286 struct mbuf *m; 1287 1288 m = m_copym(mopts, 0, M_COPYALL, M_DONTWAIT); 1289 if (m == NULL) { 1290 error = ENOBUFS; 1291 break; 1292 } 1293 error = sockopt_setmbuf(sopt, m); 1294 } 1295 break; 1296 } 1297 case IP_TOS: 1298 case IP_TTL: 1299 case IP_MINTTL: 1300 case IP_RECVOPTS: 1301 case IP_RECVRETOPTS: 1302 case IP_RECVDSTADDR: 1303 case IP_RECVIF: 1304 case IP_RECVPKTINFO: 1305 case IP_RECVTTL: 1306 case IP_ERRORMTU: 1307 switch (sopt->sopt_name) { 1308 case IP_TOS: 1309 optval = ip->ip_tos; 1310 break; 1311 1312 case IP_TTL: 1313 optval = ip->ip_ttl; 1314 break; 1315 1316 case IP_MINTTL: 1317 optval = inp->inp_ip_minttl; 1318 break; 1319 1320 case IP_ERRORMTU: 1321 optval = inp->inp_errormtu; 1322 break; 1323 1324 #define OPTBIT(bit) (inpflags & bit ? 1 : 0) 1325 1326 case IP_RECVOPTS: 1327 optval = OPTBIT(INP_RECVOPTS); 1328 break; 1329 1330 case IP_RECVPKTINFO: 1331 optval = OPTBIT(INP_RECVPKTINFO); 1332 break; 1333 1334 case IP_RECVRETOPTS: 1335 optval = OPTBIT(INP_RECVRETOPTS); 1336 break; 1337 1338 case IP_RECVDSTADDR: 1339 optval = OPTBIT(INP_RECVDSTADDR); 1340 break; 1341 1342 case IP_RECVIF: 1343 optval = OPTBIT(INP_RECVIF); 1344 break; 1345 1346 case IP_RECVTTL: 1347 optval = OPTBIT(INP_RECVTTL); 1348 break; 1349 } 1350 error = sockopt_setint(sopt, optval); 1351 break; 1352 1353 case IP_PKTINFO: 1354 switch (sopt->sopt_size) { 1355 case sizeof(int): 1356 /* Linux compatibility */ 1357 optval = OPTBIT(INP_RECVPKTINFO); 1358 error = sockopt_setint(sopt, optval); 1359 break; 1360 case sizeof(struct in_pktinfo): 1361 /* Solaris compatibility */ 1362 pktinfo.ipi_ifindex = 0; 1363 pktinfo.ipi_addr = inp->inp_prefsrcip; 1364 error = sockopt_set(sopt, &pktinfo, 1365 sizeof(pktinfo)); 1366 break; 1367 default: 1368 /* 1369 * While size is stuck at 0, and, later, if 1370 * the caller doesn't use an exactly sized 1371 * recipient for the data, default to Linux 1372 * compatibility 1373 */ 1374 optval = OPTBIT(INP_RECVPKTINFO); 1375 error = sockopt_setint(sopt, optval); 1376 break; 1377 } 1378 break; 1379 1380 #if 0 /* defined(IPSEC) */ 1381 case IP_IPSEC_POLICY: 1382 { 1383 struct mbuf *m = NULL; 1384 1385 /* XXX this will return EINVAL as sopt is empty */ 1386 error = ipsec_get_policy(inp, sopt->sopt_data, 1387 sopt->sopt_size, &m); 1388 if (error == 0) 1389 error = sockopt_setmbuf(sopt, m); 1390 break; 1391 } 1392 #endif /*IPSEC*/ 1393 1394 case IP_MULTICAST_IF: 1395 case IP_MULTICAST_TTL: 1396 case IP_MULTICAST_LOOP: 1397 case IP_ADD_MEMBERSHIP: 1398 case IP_DROP_MEMBERSHIP: 1399 error = ip_getmoptions(inp->inp_moptions, sopt); 1400 break; 1401 1402 case IP_PORTRANGE: 1403 if (inpflags & INP_LOWPORT) 1404 optval = IP_PORTRANGE_LOW; 1405 else 1406 optval = IP_PORTRANGE_DEFAULT; 1407 error = sockopt_setint(sopt, optval); 1408 break; 1409 1410 case IP_PORTALGO: 1411 optval = inp->inp_portalgo; 1412 error = sockopt_setint(sopt, optval); 1413 break; 1414 1415 default: 1416 error = ENOPROTOOPT; 1417 break; 1418 } 1419 break; 1420 } 1421 1422 if (!error) { 1423 inp->inp_flags = inpflags; 1424 } 1425 return error; 1426 } 1427 1428 static int 1429 ip_pktinfo_prepare(const struct in_pktinfo *pktinfo, struct ip_pktopts *pktopts, 1430 int *flags, kauth_cred_t cred) 1431 { 1432 struct ip_moptions *imo; 1433 int error = 0; 1434 bool addrset = false; 1435 1436 if (!in_nullhost(pktinfo->ipi_addr)) { 1437 pktopts->ippo_laddr.sin_addr = pktinfo->ipi_addr; 1438 /* EADDRNOTAVAIL? */ 1439 error = in_pcbbindableaddr(&pktopts->ippo_laddr, cred); 1440 if (error != 0) 1441 return error; 1442 addrset = true; 1443 } 1444 1445 if (pktinfo->ipi_ifindex != 0) { 1446 if (!addrset) { 1447 struct ifnet *ifp; 1448 struct in_ifaddr *ia; 1449 int s; 1450 1451 /* pick up primary address */ 1452 s = pserialize_read_enter(); 1453 ifp = if_byindex(pktinfo->ipi_ifindex); 1454 if (ifp == NULL) { 1455 pserialize_read_exit(s); 1456 return EADDRNOTAVAIL; 1457 } 1458 ia = in_get_ia_from_ifp(ifp); 1459 if (ia == NULL) { 1460 pserialize_read_exit(s); 1461 return EADDRNOTAVAIL; 1462 } 1463 pktopts->ippo_laddr.sin_addr = IA_SIN(ia)->sin_addr; 1464 pserialize_read_exit(s); 1465 } 1466 1467 /* 1468 * If specified ipi_ifindex, 1469 * use copied or locally initialized ip_moptions. 1470 * Original ip_moptions must not be modified. 1471 */ 1472 imo = &pktopts->ippo_imobuf; /* local buf in pktopts */ 1473 if (pktopts->ippo_imo != NULL) { 1474 memcpy(imo, pktopts->ippo_imo, sizeof(*imo)); 1475 } else { 1476 memset(imo, 0, sizeof(*imo)); 1477 imo->imo_multicast_ttl = IP_DEFAULT_MULTICAST_TTL; 1478 imo->imo_multicast_loop = IP_DEFAULT_MULTICAST_LOOP; 1479 } 1480 imo->imo_multicast_if_index = pktinfo->ipi_ifindex; 1481 pktopts->ippo_imo = imo; 1482 *flags |= IP_ROUTETOIFINDEX; 1483 } 1484 return error; 1485 } 1486 1487 /* 1488 * Set up IP outgoing packet options. Even if control is NULL, 1489 * pktopts->ippo_laddr and pktopts->ippo_imo are set and used. 1490 */ 1491 int 1492 ip_setpktopts(struct mbuf *control, struct ip_pktopts *pktopts, int *flags, 1493 struct inpcb *inp, kauth_cred_t cred) 1494 { 1495 struct cmsghdr *cm; 1496 struct in_pktinfo pktinfo; 1497 int error; 1498 1499 pktopts->ippo_imo = inp->inp_moptions; 1500 1501 struct in_addr *ia = in_nullhost(inp->inp_prefsrcip) ? &inp->inp_laddr : 1502 &inp->inp_prefsrcip; 1503 sockaddr_in_init(&pktopts->ippo_laddr, ia, 0); 1504 1505 if (control == NULL) 1506 return 0; 1507 1508 /* 1509 * XXX: Currently, we assume all the optional information is 1510 * stored in a single mbuf. 1511 */ 1512 if (control->m_next) 1513 return EINVAL; 1514 1515 for (; control->m_len > 0; 1516 control->m_data += CMSG_ALIGN(cm->cmsg_len), 1517 control->m_len -= CMSG_ALIGN(cm->cmsg_len)) { 1518 cm = mtod(control, struct cmsghdr *); 1519 if ((control->m_len < sizeof(*cm)) || 1520 (cm->cmsg_len == 0) || 1521 (cm->cmsg_len > control->m_len)) { 1522 return EINVAL; 1523 } 1524 if (cm->cmsg_level != IPPROTO_IP) 1525 continue; 1526 1527 switch (cm->cmsg_type) { 1528 case IP_PKTINFO: 1529 if (cm->cmsg_len != CMSG_LEN(sizeof(pktinfo))) 1530 return EINVAL; 1531 memcpy(&pktinfo, CMSG_DATA(cm), sizeof(pktinfo)); 1532 error = ip_pktinfo_prepare(&pktinfo, pktopts, flags, 1533 cred); 1534 if (error) 1535 return error; 1536 break; 1537 case IP_SENDSRCADDR: /* FreeBSD compatibility */ 1538 if (cm->cmsg_len != CMSG_LEN(sizeof(struct in_addr))) 1539 return EINVAL; 1540 pktinfo.ipi_ifindex = 0; 1541 pktinfo.ipi_addr = 1542 ((struct in_pktinfo *)CMSG_DATA(cm))->ipi_addr; 1543 error = ip_pktinfo_prepare(&pktinfo, pktopts, flags, 1544 cred); 1545 if (error) 1546 return error; 1547 break; 1548 default: 1549 return ENOPROTOOPT; 1550 } 1551 } 1552 return 0; 1553 } 1554 1555 /* 1556 * Set up IP options in pcb for insertion in output packets. 1557 * Store in mbuf with pointer in pcbopt, adding pseudo-option 1558 * with destination address if source routed. 1559 */ 1560 static int 1561 ip_pcbopts(struct inpcb *inp, const struct sockopt *sopt) 1562 { 1563 struct mbuf *m; 1564 const u_char *cp; 1565 u_char *dp; 1566 int cnt; 1567 1568 KASSERT(inp_locked(inp)); 1569 1570 /* Turn off any old options. */ 1571 if (inp->inp_options) { 1572 m_free(inp->inp_options); 1573 } 1574 inp->inp_options = NULL; 1575 if ((cnt = sopt->sopt_size) == 0) { 1576 /* Only turning off any previous options. */ 1577 return 0; 1578 } 1579 cp = sopt->sopt_data; 1580 1581 if (cnt % 4) { 1582 /* Must be 4-byte aligned, because there's no padding. */ 1583 return EINVAL; 1584 } 1585 1586 m = m_get(M_DONTWAIT, MT_SOOPTS); 1587 if (m == NULL) 1588 return ENOBUFS; 1589 1590 dp = mtod(m, u_char *); 1591 memset(dp, 0, sizeof(struct in_addr)); 1592 dp += sizeof(struct in_addr); 1593 m->m_len = sizeof(struct in_addr); 1594 1595 /* 1596 * IP option list according to RFC791. Each option is of the form 1597 * 1598 * [optval] [olen] [(olen - 2) data bytes] 1599 * 1600 * We validate the list and copy options to an mbuf for prepending 1601 * to data packets. The IP first-hop destination address will be 1602 * stored before actual options and is zero if unset. 1603 */ 1604 while (cnt > 0) { 1605 uint8_t optval, olen, offset; 1606 1607 optval = cp[IPOPT_OPTVAL]; 1608 1609 if (optval == IPOPT_EOL || optval == IPOPT_NOP) { 1610 olen = 1; 1611 } else { 1612 if (cnt < IPOPT_OLEN + 1) 1613 goto bad; 1614 1615 olen = cp[IPOPT_OLEN]; 1616 if (olen < IPOPT_OLEN + 1 || olen > cnt) 1617 goto bad; 1618 } 1619 1620 if (optval == IPOPT_LSRR || optval == IPOPT_SSRR) { 1621 /* 1622 * user process specifies route as: 1623 * ->A->B->C->D 1624 * D must be our final destination (but we can't 1625 * check that since we may not have connected yet). 1626 * A is first hop destination, which doesn't appear in 1627 * actual IP option, but is stored before the options. 1628 */ 1629 if (olen < IPOPT_OFFSET + 1 + sizeof(struct in_addr)) 1630 goto bad; 1631 1632 offset = cp[IPOPT_OFFSET]; 1633 memcpy(mtod(m, u_char *), cp + IPOPT_OFFSET + 1, 1634 sizeof(struct in_addr)); 1635 1636 cp += sizeof(struct in_addr); 1637 cnt -= sizeof(struct in_addr); 1638 olen -= sizeof(struct in_addr); 1639 1640 if (m->m_len + olen > MAX_IPOPTLEN + sizeof(struct in_addr)) 1641 goto bad; 1642 1643 memcpy(dp, cp, olen); 1644 dp[IPOPT_OPTVAL] = optval; 1645 dp[IPOPT_OLEN] = olen; 1646 dp[IPOPT_OFFSET] = offset; 1647 break; 1648 } else { 1649 if (m->m_len + olen > MAX_IPOPTLEN + sizeof(struct in_addr)) 1650 goto bad; 1651 1652 memcpy(dp, cp, olen); 1653 break; 1654 } 1655 1656 dp += olen; 1657 m->m_len += olen; 1658 1659 if (optval == IPOPT_EOL) 1660 break; 1661 1662 cp += olen; 1663 cnt -= olen; 1664 } 1665 1666 inp->inp_options = m; 1667 return 0; 1668 1669 bad: 1670 (void)m_free(m); 1671 return EINVAL; 1672 } 1673 1674 /* 1675 * following RFC1724 section 3.3, 0.0.0.0/8 is interpreted as interface index. 1676 * Must be called in a pserialize critical section. 1677 */ 1678 static struct ifnet * 1679 ip_multicast_if(struct in_addr *a, int *ifindexp) 1680 { 1681 int ifindex; 1682 struct ifnet *ifp = NULL; 1683 struct in_ifaddr *ia; 1684 1685 if (ifindexp) 1686 *ifindexp = 0; 1687 if (ntohl(a->s_addr) >> 24 == 0) { 1688 ifindex = ntohl(a->s_addr) & 0xffffff; 1689 ifp = if_byindex(ifindex); 1690 if (!ifp) 1691 return NULL; 1692 if (ifindexp) 1693 *ifindexp = ifindex; 1694 } else { 1695 IN_ADDRHASH_READER_FOREACH(ia, a->s_addr) { 1696 if (in_hosteq(ia->ia_addr.sin_addr, *a) && 1697 (ia->ia_ifp->if_flags & IFF_MULTICAST) != 0) { 1698 ifp = ia->ia_ifp; 1699 if (if_is_deactivated(ifp)) 1700 ifp = NULL; 1701 break; 1702 } 1703 } 1704 } 1705 return ifp; 1706 } 1707 1708 static int 1709 ip_getoptval(const struct sockopt *sopt, u_int8_t *val, u_int maxval) 1710 { 1711 u_int tval; 1712 u_char cval; 1713 int error; 1714 1715 if (sopt == NULL) 1716 return EINVAL; 1717 1718 switch (sopt->sopt_size) { 1719 case sizeof(u_char): 1720 error = sockopt_get(sopt, &cval, sizeof(u_char)); 1721 tval = cval; 1722 break; 1723 1724 case sizeof(u_int): 1725 error = sockopt_get(sopt, &tval, sizeof(u_int)); 1726 break; 1727 1728 default: 1729 error = EINVAL; 1730 } 1731 1732 if (error) 1733 return error; 1734 1735 if (tval > maxval) 1736 return EINVAL; 1737 1738 *val = tval; 1739 return 0; 1740 } 1741 1742 static int 1743 ip_get_membership(const struct sockopt *sopt, struct ifnet **ifp, 1744 struct psref *psref, struct in_addr *ia, bool add) 1745 { 1746 int error; 1747 struct ip_mreq mreq; 1748 1749 error = sockopt_get(sopt, &mreq, sizeof(mreq)); 1750 if (error) 1751 return error; 1752 1753 if (!IN_MULTICAST(mreq.imr_multiaddr.s_addr)) 1754 return EINVAL; 1755 1756 memcpy(ia, &mreq.imr_multiaddr, sizeof(*ia)); 1757 1758 if (in_nullhost(mreq.imr_interface)) { 1759 union { 1760 struct sockaddr dst; 1761 struct sockaddr_in dst4; 1762 } u; 1763 struct route ro; 1764 1765 if (!add) { 1766 *ifp = NULL; 1767 return 0; 1768 } 1769 /* 1770 * If no interface address was provided, use the interface of 1771 * the route to the given multicast address. 1772 */ 1773 struct rtentry *rt; 1774 memset(&ro, 0, sizeof(ro)); 1775 1776 sockaddr_in_init(&u.dst4, ia, 0); 1777 error = rtcache_setdst(&ro, &u.dst); 1778 if (error != 0) 1779 return error; 1780 *ifp = (rt = rtcache_init(&ro)) != NULL ? rt->rt_ifp : NULL; 1781 if (*ifp != NULL) { 1782 if (if_is_deactivated(*ifp)) 1783 *ifp = NULL; 1784 else 1785 if_acquire(*ifp, psref); 1786 } 1787 rtcache_unref(rt, &ro); 1788 rtcache_free(&ro); 1789 } else { 1790 int s = pserialize_read_enter(); 1791 *ifp = ip_multicast_if(&mreq.imr_interface, NULL); 1792 if (!add && *ifp == NULL) { 1793 pserialize_read_exit(s); 1794 return EADDRNOTAVAIL; 1795 } 1796 if (*ifp != NULL) { 1797 if (if_is_deactivated(*ifp)) 1798 *ifp = NULL; 1799 else 1800 if_acquire(*ifp, psref); 1801 } 1802 pserialize_read_exit(s); 1803 } 1804 return 0; 1805 } 1806 1807 /* 1808 * Add a multicast group membership. 1809 * Group must be a valid IP multicast address. 1810 */ 1811 static int 1812 ip_add_membership(struct ip_moptions *imo, const struct sockopt *sopt) 1813 { 1814 struct ifnet *ifp = NULL; // XXX: gcc [ppc] 1815 struct in_addr ia; 1816 int i, error, bound; 1817 struct psref psref; 1818 1819 /* imo is protected by solock or referenced only by the caller */ 1820 1821 bound = curlwp_bind(); 1822 if (sopt->sopt_size == sizeof(struct ip_mreq)) 1823 error = ip_get_membership(sopt, &ifp, &psref, &ia, true); 1824 else { 1825 #ifdef INET6 1826 error = ip6_get_membership(sopt, &ifp, &psref, &ia, sizeof(ia)); 1827 #else 1828 error = EINVAL; 1829 #endif 1830 } 1831 1832 if (error) 1833 goto out; 1834 1835 /* 1836 * See if we found an interface, and confirm that it 1837 * supports multicast. 1838 */ 1839 if (ifp == NULL || (ifp->if_flags & IFF_MULTICAST) == 0) { 1840 error = EADDRNOTAVAIL; 1841 goto out; 1842 } 1843 1844 /* 1845 * See if the membership already exists or if all the 1846 * membership slots are full. 1847 */ 1848 for (i = 0; i < imo->imo_num_memberships; ++i) { 1849 if (imo->imo_membership[i]->inm_ifp == ifp && 1850 in_hosteq(imo->imo_membership[i]->inm_addr, ia)) 1851 break; 1852 } 1853 if (i < imo->imo_num_memberships) { 1854 error = EADDRINUSE; 1855 goto out; 1856 } 1857 1858 if (i == IP_MAX_MEMBERSHIPS) { 1859 error = ETOOMANYREFS; 1860 goto out; 1861 } 1862 1863 /* 1864 * Everything looks good; add a new record to the multicast 1865 * address list for the given interface. 1866 */ 1867 IFNET_LOCK(ifp); 1868 imo->imo_membership[i] = in_addmulti(&ia, ifp); 1869 IFNET_UNLOCK(ifp); 1870 if (imo->imo_membership[i] == NULL) { 1871 error = ENOBUFS; 1872 goto out; 1873 } 1874 1875 ++imo->imo_num_memberships; 1876 error = 0; 1877 out: 1878 if_put(ifp, &psref); 1879 curlwp_bindx(bound); 1880 return error; 1881 } 1882 1883 /* 1884 * Drop a multicast group membership. 1885 * Group must be a valid IP multicast address. 1886 */ 1887 static int 1888 ip_drop_membership(struct ip_moptions *imo, const struct sockopt *sopt) 1889 { 1890 struct in_addr ia = { .s_addr = 0 }; // XXX: gcc [ppc] 1891 struct ifnet *ifp = NULL; // XXX: gcc [ppc] 1892 int i, error, bound; 1893 struct psref psref; 1894 1895 /* imo is protected by solock or referenced only by the caller */ 1896 1897 bound = curlwp_bind(); 1898 if (sopt->sopt_size == sizeof(struct ip_mreq)) 1899 error = ip_get_membership(sopt, &ifp, &psref, &ia, false); 1900 else { 1901 #ifdef INET6 1902 error = ip6_get_membership(sopt, &ifp, &psref, &ia, sizeof(ia)); 1903 #else 1904 error = EINVAL; 1905 #endif 1906 } 1907 1908 if (error) 1909 goto out; 1910 1911 /* 1912 * Find the membership in the membership array. 1913 */ 1914 for (i = 0; i < imo->imo_num_memberships; ++i) { 1915 if ((ifp == NULL || 1916 imo->imo_membership[i]->inm_ifp == ifp) && 1917 in_hosteq(imo->imo_membership[i]->inm_addr, ia)) 1918 break; 1919 } 1920 if (i == imo->imo_num_memberships) { 1921 error = EADDRNOTAVAIL; 1922 goto out; 1923 } 1924 1925 /* 1926 * Give up the multicast address record to which the 1927 * membership points. 1928 */ 1929 struct ifnet *inm_ifp = imo->imo_membership[i]->inm_ifp; 1930 IFNET_LOCK(inm_ifp); 1931 in_delmulti(imo->imo_membership[i]); 1932 IFNET_UNLOCK(inm_ifp); 1933 1934 /* 1935 * Remove the gap in the membership array. 1936 */ 1937 for (++i; i < imo->imo_num_memberships; ++i) 1938 imo->imo_membership[i-1] = imo->imo_membership[i]; 1939 --imo->imo_num_memberships; 1940 error = 0; 1941 out: 1942 if_put(ifp, &psref); 1943 curlwp_bindx(bound); 1944 return error; 1945 } 1946 1947 /* 1948 * Set the IP multicast options in response to user setsockopt(). 1949 */ 1950 int 1951 ip_setmoptions(struct ip_moptions **pimo, const struct sockopt *sopt) 1952 { 1953 struct ip_moptions *imo = *pimo; 1954 struct in_addr addr; 1955 struct ifnet *ifp; 1956 int ifindex, error = 0; 1957 1958 /* The passed imo isn't NULL, it should be protected by solock */ 1959 1960 if (!imo) { 1961 /* 1962 * No multicast option buffer attached to the pcb; 1963 * allocate one and initialize to default values. 1964 */ 1965 imo = kmem_intr_alloc(sizeof(*imo), KM_NOSLEEP); 1966 if (imo == NULL) 1967 return ENOBUFS; 1968 1969 imo->imo_multicast_if_index = 0; 1970 imo->imo_multicast_addr.s_addr = INADDR_ANY; 1971 imo->imo_multicast_ttl = IP_DEFAULT_MULTICAST_TTL; 1972 imo->imo_multicast_loop = IP_DEFAULT_MULTICAST_LOOP; 1973 imo->imo_num_memberships = 0; 1974 *pimo = imo; 1975 } 1976 1977 switch (sopt->sopt_name) { 1978 case IP_MULTICAST_IF: { 1979 int s; 1980 /* 1981 * Select the interface for outgoing multicast packets. 1982 */ 1983 error = sockopt_get(sopt, &addr, sizeof(addr)); 1984 if (error) 1985 break; 1986 1987 /* 1988 * INADDR_ANY is used to remove a previous selection. 1989 * When no interface is selected, a default one is 1990 * chosen every time a multicast packet is sent. 1991 */ 1992 if (in_nullhost(addr)) { 1993 imo->imo_multicast_if_index = 0; 1994 break; 1995 } 1996 /* 1997 * The selected interface is identified by its local 1998 * IP address. Find the interface and confirm that 1999 * it supports multicasting. 2000 */ 2001 s = pserialize_read_enter(); 2002 ifp = ip_multicast_if(&addr, &ifindex); 2003 if (ifp == NULL || (ifp->if_flags & IFF_MULTICAST) == 0) { 2004 pserialize_read_exit(s); 2005 error = EADDRNOTAVAIL; 2006 break; 2007 } 2008 imo->imo_multicast_if_index = ifp->if_index; 2009 pserialize_read_exit(s); 2010 if (ifindex) 2011 imo->imo_multicast_addr = addr; 2012 else 2013 imo->imo_multicast_addr.s_addr = INADDR_ANY; 2014 break; 2015 } 2016 2017 case IP_MULTICAST_TTL: 2018 /* 2019 * Set the IP time-to-live for outgoing multicast packets. 2020 */ 2021 error = ip_getoptval(sopt, &imo->imo_multicast_ttl, MAXTTL); 2022 break; 2023 2024 case IP_MULTICAST_LOOP: 2025 /* 2026 * Set the loopback flag for outgoing multicast packets. 2027 * Must be zero or one. 2028 */ 2029 error = ip_getoptval(sopt, &imo->imo_multicast_loop, 1); 2030 break; 2031 2032 case IP_ADD_MEMBERSHIP: /* IPV6_JOIN_GROUP */ 2033 error = ip_add_membership(imo, sopt); 2034 break; 2035 2036 case IP_DROP_MEMBERSHIP: /* IPV6_LEAVE_GROUP */ 2037 error = ip_drop_membership(imo, sopt); 2038 break; 2039 2040 default: 2041 error = EOPNOTSUPP; 2042 break; 2043 } 2044 2045 /* 2046 * If all options have default values, no need to keep the mbuf. 2047 */ 2048 if (imo->imo_multicast_if_index == 0 && 2049 imo->imo_multicast_ttl == IP_DEFAULT_MULTICAST_TTL && 2050 imo->imo_multicast_loop == IP_DEFAULT_MULTICAST_LOOP && 2051 imo->imo_num_memberships == 0) { 2052 kmem_intr_free(imo, sizeof(*imo)); 2053 *pimo = NULL; 2054 } 2055 2056 return error; 2057 } 2058 2059 /* 2060 * Return the IP multicast options in response to user getsockopt(). 2061 */ 2062 int 2063 ip_getmoptions(struct ip_moptions *imo, struct sockopt *sopt) 2064 { 2065 struct in_addr addr; 2066 uint8_t optval; 2067 int error = 0; 2068 2069 /* imo is protected by solock or refereced only by the caller */ 2070 2071 switch (sopt->sopt_name) { 2072 case IP_MULTICAST_IF: 2073 if (imo == NULL || imo->imo_multicast_if_index == 0) 2074 addr = zeroin_addr; 2075 else if (imo->imo_multicast_addr.s_addr) { 2076 /* return the value user has set */ 2077 addr = imo->imo_multicast_addr; 2078 } else { 2079 struct ifnet *ifp; 2080 struct in_ifaddr *ia = NULL; 2081 int s = pserialize_read_enter(); 2082 2083 ifp = if_byindex(imo->imo_multicast_if_index); 2084 if (ifp != NULL) { 2085 ia = in_get_ia_from_ifp(ifp); 2086 } 2087 addr = ia ? ia->ia_addr.sin_addr : zeroin_addr; 2088 pserialize_read_exit(s); 2089 } 2090 error = sockopt_set(sopt, &addr, sizeof(addr)); 2091 break; 2092 2093 case IP_MULTICAST_TTL: 2094 optval = imo ? imo->imo_multicast_ttl 2095 : IP_DEFAULT_MULTICAST_TTL; 2096 2097 error = sockopt_set(sopt, &optval, sizeof(optval)); 2098 break; 2099 2100 case IP_MULTICAST_LOOP: 2101 optval = imo ? imo->imo_multicast_loop 2102 : IP_DEFAULT_MULTICAST_LOOP; 2103 2104 error = sockopt_set(sopt, &optval, sizeof(optval)); 2105 break; 2106 2107 default: 2108 error = EOPNOTSUPP; 2109 } 2110 2111 return error; 2112 } 2113 2114 /* 2115 * Discard the IP multicast options. 2116 */ 2117 void 2118 ip_freemoptions(struct ip_moptions *imo) 2119 { 2120 int i; 2121 2122 /* The owner of imo (inp) should be protected by solock */ 2123 2124 if (imo != NULL) { 2125 for (i = 0; i < imo->imo_num_memberships; ++i) { 2126 struct in_multi *inm = imo->imo_membership[i]; 2127 struct ifnet *ifp = inm->inm_ifp; 2128 IFNET_LOCK(ifp); 2129 in_delmulti(inm); 2130 /* ifp should not leave thanks to solock */ 2131 IFNET_UNLOCK(ifp); 2132 } 2133 2134 kmem_intr_free(imo, sizeof(*imo)); 2135 } 2136 } 2137 2138 /* 2139 * Routine called from ip_output() to loop back a copy of an IP multicast 2140 * packet to the input queue of a specified interface. Note that this 2141 * calls the output routine of the loopback "driver", but with an interface 2142 * pointer that might NOT be lo0ifp -- easier than replicating that code here. 2143 */ 2144 static void 2145 ip_mloopback(struct ifnet *ifp, struct mbuf *m, const struct sockaddr_in *dst) 2146 { 2147 struct ip *ip; 2148 struct mbuf *copym; 2149 2150 copym = m_copypacket(m, M_DONTWAIT); 2151 if (copym != NULL && 2152 (copym->m_flags & M_EXT || copym->m_len < sizeof(struct ip))) 2153 copym = m_pullup(copym, sizeof(struct ip)); 2154 if (copym == NULL) 2155 return; 2156 /* 2157 * We don't bother to fragment if the IP length is greater 2158 * than the interface's MTU. Can this possibly matter? 2159 */ 2160 ip = mtod(copym, struct ip *); 2161 2162 if (copym->m_pkthdr.csum_flags & (M_CSUM_TCPv4|M_CSUM_UDPv4)) { 2163 in_delayed_cksum(copym); 2164 copym->m_pkthdr.csum_flags &= 2165 ~(M_CSUM_TCPv4|M_CSUM_UDPv4); 2166 } 2167 2168 ip->ip_sum = 0; 2169 ip->ip_sum = in_cksum(copym, ip->ip_hl << 2); 2170 KERNEL_LOCK_UNLESS_NET_MPSAFE(); 2171 (void)looutput(ifp, copym, sintocsa(dst), NULL); 2172 KERNEL_UNLOCK_UNLESS_NET_MPSAFE(); 2173 } 2174 2175 /* 2176 * Ensure sending address is valid. 2177 * Returns 0 on success, -1 if an error should be sent back or 1 2178 * if the packet could be dropped without error (protocol dependent). 2179 */ 2180 static int 2181 ip_ifaddrvalid(const struct in_ifaddr *ia) 2182 { 2183 2184 if (ia->ia_addr.sin_addr.s_addr == INADDR_ANY) 2185 return 0; 2186 2187 if (ia->ia4_flags & IN_IFF_DUPLICATED) 2188 return -1; 2189 else if (ia->ia4_flags & (IN_IFF_TENTATIVE | IN_IFF_DETACHED)) 2190 return 1; 2191 2192 return 0; 2193 } 2194