1 /* 2 * Copyright 2001 Wasabi Systems, Inc. 3 * All rights reserved. 4 * 5 * Written by Jason R. Thorpe for Wasabi Systems, Inc. 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. All advertising materials mentioning features or use of this software 16 * must display the following acknowledgement: 17 * This product includes software developed for the NetBSD Project by 18 * Wasabi Systems, Inc. 19 * 4. The name of Wasabi Systems, Inc. may not be used to endorse 20 * or promote products derived from this software without specific prior 21 * written permission. 22 * 23 * THIS SOFTWARE IS PROVIDED BY WASABI SYSTEMS, INC. ``AS IS'' AND 24 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 25 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 26 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL WASABI SYSTEMS, INC 27 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 28 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 29 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 30 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 31 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 32 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 33 * POSSIBILITY OF SUCH DAMAGE. 34 */ 35 36 /* 37 * Copyright (c) 1999, 2000 Jason L. Wright (jason@thought.net) 38 * All rights reserved. 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 Jason L. Wright 51 * 4. The name of the author may not be used to endorse or promote products 52 * derived from this software without specific prior written permission. 53 * 54 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR 55 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 56 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 57 * DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, 58 * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 59 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR 60 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 61 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, 62 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN 63 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 64 * POSSIBILITY OF SUCH DAMAGE. 65 * 66 * $OpenBSD: if_bridge.c,v 1.60 2001/06/15 03:38:33 itojun Exp $ 67 * $NetBSD: if_bridge.c,v 1.31 2005/06/01 19:45:34 jdc Exp $ 68 * $FreeBSD: src/sys/net/if_bridge.c,v 1.26 2005/10/13 23:05:55 thompsa Exp $ 69 */ 70 71 /* 72 * Network interface bridge support. 73 * 74 * TODO: 75 * 76 * - Currently only supports Ethernet-like interfaces (Ethernet, 77 * 802.11, VLANs on Ethernet, etc.) Figure out a nice way 78 * to bridge other types of interfaces (FDDI-FDDI, and maybe 79 * consider heterogenous bridges). 80 * 81 * 82 * Bridge's route information is duplicated to each CPUs: 83 * 84 * CPU0 CPU1 CPU2 CPU3 85 * +-----------+ +-----------+ +-----------+ +-----------+ 86 * | rtnode | | rtnode | | rtnode | | rtnode | 87 * | | | | | | | | 88 * | dst eaddr | | dst eaddr | | dst eaddr | | dst eaddr | 89 * +-----------+ +-----------+ +-----------+ +-----------+ 90 * | | | | 91 * | | | | 92 * | | +----------+ | | 93 * | | | rtinfo | | | 94 * | +---->| |<---+ | 95 * | | flags | | 96 * +-------------->| timeout |<-------------+ 97 * | dst_ifp | 98 * +----------+ 99 * 100 * We choose to put timeout and dst_ifp into shared part, so updating 101 * them will be cheaper than using message forwarding. Also there is 102 * not need to use spinlock to protect the updating: timeout and dst_ifp 103 * is not related and specific field's updating order has no importance. 104 * The cache pollution by the share part should not be heavy: in a stable 105 * setup, dst_ifp probably will be not changed in rtnode's life time, 106 * while timeout is refreshed once per second; most of the time, timeout 107 * and dst_ifp are read-only accessed. 108 * 109 * 110 * Bridge route information installation on bridge_input path: 111 * 112 * CPU0 CPU1 CPU2 CPU3 113 * 114 * tcp_thread2 115 * | 116 * alloc nmsg 117 * snd nmsg | 118 * w/o rtinfo | 119 * ifnet0<-----------------------+ 120 * | : 121 * lookup dst : 122 * rtnode exists?(Y)free nmsg : 123 * |(N) : 124 * | 125 * alloc rtinfo 126 * alloc rtnode 127 * install rtnode 128 * | 129 * +---------->ifnet1 130 * : fwd nmsg | 131 * : w/ rtinfo | 132 * : | 133 * : | 134 * alloc rtnode 135 * (w/ nmsg's rtinfo) 136 * install rtnode 137 * | 138 * +---------->ifnet2 139 * : fwd nmsg | 140 * : w/ rtinfo | 141 * : | 142 * : same as ifnet1 143 * | 144 * +---------->ifnet3 145 * : fwd nmsg | 146 * : w/ rtinfo | 147 * : | 148 * : same as ifnet1 149 * free nmsg 150 * : 151 * : 152 * 153 * The netmsgs forwarded between protocol threads and ifnet threads are 154 * allocated with (M_WAITOK|M_NULLOK), so it will not fail under most 155 * cases (route information is too precious to be not installed :). 156 * Since multiple threads may try to install route information for the 157 * same dst eaddr, we look up route information in ifnet0. However, this 158 * looking up only need to be performed on ifnet0, which is the start 159 * point of the route information installation process. 160 * 161 * 162 * Bridge route information deleting/flushing: 163 * 164 * CPU0 CPU1 CPU2 CPU3 165 * 166 * netisr0 167 * | 168 * find suitable rtnodes, 169 * mark their rtinfo dead 170 * | 171 * | domsg <------------------------------------------+ 172 * | | replymsg 173 * | | 174 * V fwdmsg fwdmsg fwdmsg | 175 * ifnet0 --------> ifnet1 --------> ifnet2 --------> ifnet3 176 * delete rtnodes delete rtnodes delete rtnodes delete rtnodes 177 * w/ dead rtinfo w/ dead rtinfo w/ dead rtinfo w/ dead rtinfo 178 * free dead rtinfos 179 * 180 * All deleting/flushing operations are serialized by netisr0, so each 181 * operation only reaps the route information marked dead by itself. 182 * 183 * 184 * Bridge route information adding/deleting/flushing: 185 * Since all operation is serialized by the fixed message flow between 186 * ifnet threads, it is not possible to create corrupted per-cpu route 187 * information. 188 * 189 * 190 * 191 * Percpu member interface list iteration with blocking operation: 192 * Since one bridge could only delete one member interface at a time and 193 * the deleted member interface is not freed after netmsg_service_sync(), 194 * following way is used to make sure that even if the certain member 195 * interface is ripped from the percpu list during the blocking operation, 196 * the iteration still could keep going: 197 * 198 * TAILQ_FOREACH_MUTABLE(bif, sc->sc_iflists[mycpuid], bif_next, nbif) { 199 * blocking operation; 200 * blocking operation; 201 * ... 202 * ... 203 * if (nbif != NULL && !nbif->bif_onlist) { 204 * KKASSERT(bif->bif_onlist); 205 * nbif = TAILQ_NEXT(bif, bif_next); 206 * } 207 * } 208 * 209 * As mentioned above only one member interface could be unlinked from the 210 * percpu member interface list, so either bif or nbif may be not on the list, 211 * but _not_ both. To keep the list iteration, we don't care about bif, but 212 * only nbif. Since removed member interface will only be freed after we 213 * finish our work, it is safe to access any field in an unlinked bif (here 214 * bif_onlist). If nbif is no longer on the list, then bif must be on the 215 * list, so we change nbif to the next element of bif and keep going. 216 */ 217 218 #include "opt_inet.h" 219 #include "opt_inet6.h" 220 221 #include <sys/param.h> 222 #include <sys/mbuf.h> 223 #include <sys/malloc.h> 224 #include <sys/protosw.h> 225 #include <sys/systm.h> 226 #include <sys/time.h> 227 #include <sys/socket.h> /* for net/if.h */ 228 #include <sys/sockio.h> 229 #include <sys/ctype.h> /* string functions */ 230 #include <sys/kernel.h> 231 #include <sys/random.h> 232 #include <sys/sysctl.h> 233 #include <sys/module.h> 234 #include <sys/proc.h> 235 #include <sys/priv.h> 236 #include <sys/lock.h> 237 #include <sys/thread.h> 238 #include <sys/thread2.h> 239 #include <sys/mpipe.h> 240 241 #include <net/bpf.h> 242 #include <net/if.h> 243 #include <net/if_dl.h> 244 #include <net/if_types.h> 245 #include <net/if_var.h> 246 #include <net/pfil.h> 247 #include <net/ifq_var.h> 248 #include <net/if_clone.h> 249 250 #include <netinet/in.h> /* for struct arpcom */ 251 #include <netinet/in_systm.h> 252 #include <netinet/in_var.h> 253 #include <netinet/ip.h> 254 #include <netinet/ip_var.h> 255 #ifdef INET6 256 #include <netinet/ip6.h> 257 #include <netinet6/ip6_var.h> 258 #endif 259 #include <netinet/if_ether.h> /* for struct arpcom */ 260 #include <net/bridge/if_bridgevar.h> 261 #include <net/if_llc.h> 262 #include <net/netmsg2.h> 263 264 #include <net/route.h> 265 #include <sys/in_cksum.h> 266 267 /* 268 * Size of the route hash table. Must be a power of two. 269 */ 270 #ifndef BRIDGE_RTHASH_SIZE 271 #define BRIDGE_RTHASH_SIZE 1024 272 #endif 273 274 #define BRIDGE_RTHASH_MASK (BRIDGE_RTHASH_SIZE - 1) 275 276 /* 277 * Maximum number of addresses to cache. 278 */ 279 #ifndef BRIDGE_RTABLE_MAX 280 #define BRIDGE_RTABLE_MAX 100 281 #endif 282 283 /* 284 * Spanning tree defaults. 285 */ 286 #define BSTP_DEFAULT_MAX_AGE (20 * 256) 287 #define BSTP_DEFAULT_HELLO_TIME (2 * 256) 288 #define BSTP_DEFAULT_FORWARD_DELAY (15 * 256) 289 #define BSTP_DEFAULT_HOLD_TIME (1 * 256) 290 #define BSTP_DEFAULT_BRIDGE_PRIORITY 0x8000 291 #define BSTP_DEFAULT_PORT_PRIORITY 0x80 292 #define BSTP_DEFAULT_PATH_COST 55 293 294 /* 295 * Timeout (in seconds) for entries learned dynamically. 296 */ 297 #ifndef BRIDGE_RTABLE_TIMEOUT 298 #define BRIDGE_RTABLE_TIMEOUT (20 * 60) /* same as ARP */ 299 #endif 300 301 /* 302 * Number of seconds between walks of the route list. 303 */ 304 #ifndef BRIDGE_RTABLE_PRUNE_PERIOD 305 #define BRIDGE_RTABLE_PRUNE_PERIOD (5 * 60) 306 #endif 307 308 /* 309 * List of capabilities to mask on the member interface. 310 */ 311 #define BRIDGE_IFCAPS_MASK IFCAP_TXCSUM 312 313 typedef int (*bridge_ctl_t)(struct bridge_softc *, void *); 314 315 struct netmsg_brctl { 316 struct netmsg_base base; 317 bridge_ctl_t bc_func; 318 struct bridge_softc *bc_sc; 319 void *bc_arg; 320 }; 321 322 struct netmsg_brsaddr { 323 struct netmsg_base base; 324 struct bridge_softc *br_softc; 325 struct ifnet *br_dst_if; 326 struct bridge_rtinfo *br_rtinfo; 327 int br_setflags; 328 uint8_t br_dst[ETHER_ADDR_LEN]; 329 uint8_t br_flags; 330 }; 331 332 struct netmsg_braddbif { 333 struct netmsg_base base; 334 struct bridge_softc *br_softc; 335 struct bridge_ifinfo *br_bif_info; 336 struct ifnet *br_bif_ifp; 337 }; 338 339 struct netmsg_brdelbif { 340 struct netmsg_base base; 341 struct bridge_softc *br_softc; 342 struct bridge_ifinfo *br_bif_info; 343 struct bridge_iflist_head *br_bif_list; 344 }; 345 346 struct netmsg_brsflags { 347 struct netmsg_base base; 348 struct bridge_softc *br_softc; 349 struct bridge_ifinfo *br_bif_info; 350 uint32_t br_bif_flags; 351 }; 352 353 eventhandler_tag bridge_detach_cookie = NULL; 354 355 extern struct mbuf *(*bridge_input_p)(struct ifnet *, struct mbuf *); 356 extern int (*bridge_output_p)(struct ifnet *, struct mbuf *); 357 extern void (*bridge_dn_p)(struct mbuf *, struct ifnet *); 358 extern struct ifnet *(*bridge_interface_p)(void *if_bridge); 359 360 static int bridge_rtable_prune_period = BRIDGE_RTABLE_PRUNE_PERIOD; 361 362 static int bridge_clone_create(struct if_clone *, int, caddr_t); 363 static int bridge_clone_destroy(struct ifnet *); 364 365 static int bridge_ioctl(struct ifnet *, u_long, caddr_t, struct ucred *); 366 static void bridge_mutecaps(struct bridge_ifinfo *, struct ifnet *, int); 367 static void bridge_ifdetach(void *, struct ifnet *); 368 static void bridge_init(void *); 369 static int bridge_from_us(struct bridge_softc *, struct ether_header *); 370 static void bridge_stop(struct ifnet *); 371 static void bridge_start(struct ifnet *, struct ifaltq_subque *); 372 static struct mbuf *bridge_input(struct ifnet *, struct mbuf *); 373 static int bridge_output(struct ifnet *, struct mbuf *); 374 static struct ifnet *bridge_interface(void *if_bridge); 375 376 static void bridge_forward(struct bridge_softc *, struct mbuf *m); 377 378 static void bridge_timer_handler(netmsg_t); 379 static void bridge_timer(void *); 380 381 static void bridge_start_bcast(struct bridge_softc *, struct mbuf *); 382 static void bridge_broadcast(struct bridge_softc *, struct ifnet *, 383 struct mbuf *); 384 static void bridge_span(struct bridge_softc *, struct mbuf *); 385 386 static int bridge_rtupdate(struct bridge_softc *, const uint8_t *, 387 struct ifnet *, uint8_t); 388 static struct ifnet *bridge_rtlookup(struct bridge_softc *, const uint8_t *); 389 static void bridge_rtreap(struct bridge_softc *); 390 static void bridge_rtreap_async(struct bridge_softc *); 391 static void bridge_rttrim(struct bridge_softc *); 392 static int bridge_rtage_finddead(struct bridge_softc *); 393 static void bridge_rtage(struct bridge_softc *); 394 static void bridge_rtflush(struct bridge_softc *, int); 395 static int bridge_rtdaddr(struct bridge_softc *, const uint8_t *); 396 static int bridge_rtsaddr(struct bridge_softc *, const uint8_t *, 397 struct ifnet *, uint8_t); 398 static void bridge_rtmsg_sync(struct bridge_softc *sc); 399 static void bridge_rtreap_handler(netmsg_t); 400 static void bridge_rtinstall_handler(netmsg_t); 401 static int bridge_rtinstall_oncpu(struct bridge_softc *, const uint8_t *, 402 struct ifnet *, int, uint8_t, struct bridge_rtinfo **); 403 404 static void bridge_rtable_init(struct bridge_softc *); 405 static void bridge_rtable_fini(struct bridge_softc *); 406 407 static int bridge_rtnode_addr_cmp(const uint8_t *, const uint8_t *); 408 static struct bridge_rtnode *bridge_rtnode_lookup(struct bridge_softc *, 409 const uint8_t *); 410 static void bridge_rtnode_insert(struct bridge_softc *, 411 struct bridge_rtnode *); 412 static void bridge_rtnode_destroy(struct bridge_softc *, 413 struct bridge_rtnode *); 414 415 static struct bridge_iflist *bridge_lookup_member(struct bridge_softc *, 416 const char *name); 417 static struct bridge_iflist *bridge_lookup_member_if(struct bridge_softc *, 418 struct ifnet *ifp); 419 static struct bridge_iflist *bridge_lookup_member_ifinfo(struct bridge_softc *, 420 struct bridge_ifinfo *); 421 static void bridge_delete_member(struct bridge_softc *, 422 struct bridge_iflist *, int); 423 static void bridge_delete_span(struct bridge_softc *, 424 struct bridge_iflist *); 425 426 static int bridge_control(struct bridge_softc *, u_long, 427 bridge_ctl_t, void *); 428 static int bridge_ioctl_init(struct bridge_softc *, void *); 429 static int bridge_ioctl_stop(struct bridge_softc *, void *); 430 static int bridge_ioctl_add(struct bridge_softc *, void *); 431 static int bridge_ioctl_del(struct bridge_softc *, void *); 432 static void bridge_ioctl_fillflags(struct bridge_softc *sc, 433 struct bridge_iflist *bif, struct ifbreq *req); 434 static int bridge_ioctl_gifflags(struct bridge_softc *, void *); 435 static int bridge_ioctl_sifflags(struct bridge_softc *, void *); 436 static int bridge_ioctl_scache(struct bridge_softc *, void *); 437 static int bridge_ioctl_gcache(struct bridge_softc *, void *); 438 static int bridge_ioctl_gifs(struct bridge_softc *, void *); 439 static int bridge_ioctl_rts(struct bridge_softc *, void *); 440 static int bridge_ioctl_saddr(struct bridge_softc *, void *); 441 static int bridge_ioctl_sto(struct bridge_softc *, void *); 442 static int bridge_ioctl_gto(struct bridge_softc *, void *); 443 static int bridge_ioctl_daddr(struct bridge_softc *, void *); 444 static int bridge_ioctl_flush(struct bridge_softc *, void *); 445 static int bridge_ioctl_gpri(struct bridge_softc *, void *); 446 static int bridge_ioctl_spri(struct bridge_softc *, void *); 447 static int bridge_ioctl_reinit(struct bridge_softc *, void *); 448 static int bridge_ioctl_ght(struct bridge_softc *, void *); 449 static int bridge_ioctl_sht(struct bridge_softc *, void *); 450 static int bridge_ioctl_gfd(struct bridge_softc *, void *); 451 static int bridge_ioctl_sfd(struct bridge_softc *, void *); 452 static int bridge_ioctl_gma(struct bridge_softc *, void *); 453 static int bridge_ioctl_sma(struct bridge_softc *, void *); 454 static int bridge_ioctl_sifprio(struct bridge_softc *, void *); 455 static int bridge_ioctl_sifcost(struct bridge_softc *, void *); 456 static int bridge_ioctl_addspan(struct bridge_softc *, void *); 457 static int bridge_ioctl_delspan(struct bridge_softc *, void *); 458 static int bridge_ioctl_sifbondwght(struct bridge_softc *, void *); 459 static int bridge_pfil(struct mbuf **, struct ifnet *, struct ifnet *, 460 int); 461 static int bridge_ip_checkbasic(struct mbuf **mp); 462 #ifdef INET6 463 static int bridge_ip6_checkbasic(struct mbuf **mp); 464 #endif /* INET6 */ 465 static int bridge_fragment(struct ifnet *, struct mbuf *, 466 struct ether_header *, int, struct llc *); 467 static void bridge_enqueue_handler(netmsg_t); 468 static void bridge_handoff(struct bridge_softc *, struct ifnet *, 469 struct mbuf *, int); 470 471 static void bridge_del_bif_handler(netmsg_t); 472 static void bridge_add_bif_handler(netmsg_t); 473 static void bridge_del_bif(struct bridge_softc *, struct bridge_ifinfo *, 474 struct bridge_iflist_head *); 475 static void bridge_add_bif(struct bridge_softc *, struct bridge_ifinfo *, 476 struct ifnet *); 477 478 SYSCTL_DECL(_net_link); 479 SYSCTL_NODE(_net_link, IFT_BRIDGE, bridge, CTLFLAG_RW, 0, "Bridge"); 480 481 static int pfil_onlyip = 1; /* only pass IP[46] packets when pfil is enabled */ 482 static int pfil_bridge = 1; /* run pfil hooks on the bridge interface */ 483 static int pfil_member = 1; /* run pfil hooks on the member interface */ 484 static int bridge_debug; 485 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_onlyip, CTLFLAG_RW, 486 &pfil_onlyip, 0, "Only pass IP packets when pfil is enabled"); 487 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_bridge, CTLFLAG_RW, 488 &pfil_bridge, 0, "Packet filter on the bridge interface"); 489 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_member, CTLFLAG_RW, 490 &pfil_member, 0, "Packet filter on the member interface"); 491 SYSCTL_INT(_net_link_bridge, OID_AUTO, debug, CTLFLAG_RW, 492 &bridge_debug, 0, "Bridge debug mode"); 493 494 struct bridge_control_arg { 495 union { 496 struct ifbreq ifbreq; 497 struct ifbifconf ifbifconf; 498 struct ifbareq ifbareq; 499 struct ifbaconf ifbaconf; 500 struct ifbrparam ifbrparam; 501 } bca_u; 502 int bca_len; 503 void *bca_uptr; 504 void *bca_kptr; 505 }; 506 507 struct bridge_control { 508 bridge_ctl_t bc_func; 509 int bc_argsize; 510 int bc_flags; 511 }; 512 513 #define BC_F_COPYIN 0x01 /* copy arguments in */ 514 #define BC_F_COPYOUT 0x02 /* copy arguments out */ 515 #define BC_F_SUSER 0x04 /* do super-user check */ 516 517 const struct bridge_control bridge_control_table[] = { 518 { bridge_ioctl_add, sizeof(struct ifbreq), 519 BC_F_COPYIN|BC_F_SUSER }, 520 { bridge_ioctl_del, sizeof(struct ifbreq), 521 BC_F_COPYIN|BC_F_SUSER }, 522 523 { bridge_ioctl_gifflags, sizeof(struct ifbreq), 524 BC_F_COPYIN|BC_F_COPYOUT }, 525 { bridge_ioctl_sifflags, sizeof(struct ifbreq), 526 BC_F_COPYIN|BC_F_SUSER }, 527 528 { bridge_ioctl_scache, sizeof(struct ifbrparam), 529 BC_F_COPYIN|BC_F_SUSER }, 530 { bridge_ioctl_gcache, sizeof(struct ifbrparam), 531 BC_F_COPYOUT }, 532 533 { bridge_ioctl_gifs, sizeof(struct ifbifconf), 534 BC_F_COPYIN|BC_F_COPYOUT }, 535 { bridge_ioctl_rts, sizeof(struct ifbaconf), 536 BC_F_COPYIN|BC_F_COPYOUT }, 537 538 { bridge_ioctl_saddr, sizeof(struct ifbareq), 539 BC_F_COPYIN|BC_F_SUSER }, 540 541 { bridge_ioctl_sto, sizeof(struct ifbrparam), 542 BC_F_COPYIN|BC_F_SUSER }, 543 { bridge_ioctl_gto, sizeof(struct ifbrparam), 544 BC_F_COPYOUT }, 545 546 { bridge_ioctl_daddr, sizeof(struct ifbareq), 547 BC_F_COPYIN|BC_F_SUSER }, 548 549 { bridge_ioctl_flush, sizeof(struct ifbreq), 550 BC_F_COPYIN|BC_F_SUSER }, 551 552 { bridge_ioctl_gpri, sizeof(struct ifbrparam), 553 BC_F_COPYOUT }, 554 { bridge_ioctl_spri, sizeof(struct ifbrparam), 555 BC_F_COPYIN|BC_F_SUSER }, 556 557 { bridge_ioctl_ght, sizeof(struct ifbrparam), 558 BC_F_COPYOUT }, 559 { bridge_ioctl_sht, sizeof(struct ifbrparam), 560 BC_F_COPYIN|BC_F_SUSER }, 561 562 { bridge_ioctl_gfd, sizeof(struct ifbrparam), 563 BC_F_COPYOUT }, 564 { bridge_ioctl_sfd, sizeof(struct ifbrparam), 565 BC_F_COPYIN|BC_F_SUSER }, 566 567 { bridge_ioctl_gma, sizeof(struct ifbrparam), 568 BC_F_COPYOUT }, 569 { bridge_ioctl_sma, sizeof(struct ifbrparam), 570 BC_F_COPYIN|BC_F_SUSER }, 571 572 { bridge_ioctl_sifprio, sizeof(struct ifbreq), 573 BC_F_COPYIN|BC_F_SUSER }, 574 575 { bridge_ioctl_sifcost, sizeof(struct ifbreq), 576 BC_F_COPYIN|BC_F_SUSER }, 577 578 { bridge_ioctl_addspan, sizeof(struct ifbreq), 579 BC_F_COPYIN|BC_F_SUSER }, 580 { bridge_ioctl_delspan, sizeof(struct ifbreq), 581 BC_F_COPYIN|BC_F_SUSER }, 582 583 { bridge_ioctl_sifbondwght, sizeof(struct ifbreq), 584 BC_F_COPYIN|BC_F_SUSER }, 585 586 }; 587 static const int bridge_control_table_size = NELEM(bridge_control_table); 588 589 LIST_HEAD(, bridge_softc) bridge_list; 590 591 struct if_clone bridge_cloner = IF_CLONE_INITIALIZER("bridge", 592 bridge_clone_create, 593 bridge_clone_destroy, 0, IF_MAXUNIT); 594 595 static int 596 bridge_modevent(module_t mod, int type, void *data) 597 { 598 switch (type) { 599 case MOD_LOAD: 600 LIST_INIT(&bridge_list); 601 if_clone_attach(&bridge_cloner); 602 bridge_input_p = bridge_input; 603 bridge_output_p = bridge_output; 604 bridge_interface_p = bridge_interface; 605 bridge_detach_cookie = EVENTHANDLER_REGISTER( 606 ifnet_detach_event, bridge_ifdetach, NULL, 607 EVENTHANDLER_PRI_ANY); 608 #if 0 /* notyet */ 609 bstp_linkstate_p = bstp_linkstate; 610 #endif 611 break; 612 case MOD_UNLOAD: 613 if (!LIST_EMPTY(&bridge_list)) 614 return (EBUSY); 615 EVENTHANDLER_DEREGISTER(ifnet_detach_event, 616 bridge_detach_cookie); 617 if_clone_detach(&bridge_cloner); 618 bridge_input_p = NULL; 619 bridge_output_p = NULL; 620 bridge_interface_p = NULL; 621 #if 0 /* notyet */ 622 bstp_linkstate_p = NULL; 623 #endif 624 break; 625 default: 626 return (EOPNOTSUPP); 627 } 628 return (0); 629 } 630 631 static moduledata_t bridge_mod = { 632 "if_bridge", 633 bridge_modevent, 634 0 635 }; 636 637 DECLARE_MODULE(if_bridge, bridge_mod, SI_SUB_PSEUDO, SI_ORDER_ANY); 638 639 640 /* 641 * bridge_clone_create: 642 * 643 * Create a new bridge instance. 644 */ 645 static int 646 bridge_clone_create(struct if_clone *ifc, int unit, caddr_t param __unused) 647 { 648 struct bridge_softc *sc; 649 struct ifnet *ifp; 650 u_char eaddr[6]; 651 int cpu, rnd; 652 653 sc = kmalloc(sizeof(*sc), M_DEVBUF, M_WAITOK | M_ZERO); 654 ifp = sc->sc_ifp = &sc->sc_if; 655 656 sc->sc_brtmax = BRIDGE_RTABLE_MAX; 657 sc->sc_brttimeout = BRIDGE_RTABLE_TIMEOUT; 658 sc->sc_bridge_max_age = BSTP_DEFAULT_MAX_AGE; 659 sc->sc_bridge_hello_time = BSTP_DEFAULT_HELLO_TIME; 660 sc->sc_bridge_forward_delay = BSTP_DEFAULT_FORWARD_DELAY; 661 sc->sc_bridge_priority = BSTP_DEFAULT_BRIDGE_PRIORITY; 662 sc->sc_hold_time = BSTP_DEFAULT_HOLD_TIME; 663 664 /* Initialize our routing table. */ 665 bridge_rtable_init(sc); 666 667 callout_init(&sc->sc_brcallout); 668 netmsg_init(&sc->sc_brtimemsg, NULL, &netisr_adone_rport, 669 MSGF_DROPABLE, bridge_timer_handler); 670 sc->sc_brtimemsg.lmsg.u.ms_resultp = sc; 671 672 callout_init(&sc->sc_bstpcallout); 673 netmsg_init(&sc->sc_bstptimemsg, NULL, &netisr_adone_rport, 674 MSGF_DROPABLE, bstp_tick_handler); 675 sc->sc_bstptimemsg.lmsg.u.ms_resultp = sc; 676 677 /* Initialize per-cpu member iface lists */ 678 sc->sc_iflists = kmalloc(sizeof(*sc->sc_iflists) * ncpus, 679 M_DEVBUF, M_WAITOK); 680 for (cpu = 0; cpu < ncpus; ++cpu) 681 TAILQ_INIT(&sc->sc_iflists[cpu]); 682 683 TAILQ_INIT(&sc->sc_spanlist); 684 685 ifp->if_softc = sc; 686 if_initname(ifp, ifc->ifc_name, unit); 687 ifp->if_mtu = ETHERMTU; 688 ifp->if_flags = IFF_BROADCAST | IFF_MULTICAST; 689 ifp->if_ioctl = bridge_ioctl; 690 ifp->if_start = bridge_start; 691 ifp->if_init = bridge_init; 692 ifp->if_type = IFT_ETHER; 693 ifq_set_maxlen(&ifp->if_snd, ifqmaxlen); 694 ifq_set_ready(&ifp->if_snd); 695 ifp->if_hdrlen = ETHER_HDR_LEN; 696 697 /* 698 * Generate a random ethernet address and use the private AC:DE:48 699 * OUI code. 700 */ 701 rnd = karc4random(); 702 bcopy(&rnd, &eaddr[0], 4); /* ETHER_ADDR_LEN == 6 */ 703 rnd = karc4random(); 704 bcopy(&rnd, &eaddr[2], 4); /* ETHER_ADDR_LEN == 6 */ 705 706 eaddr[0] &= ~1; /* clear multicast bit */ 707 eaddr[0] |= 2; /* set the LAA bit */ 708 709 ether_ifattach(ifp, eaddr, NULL); 710 /* Now undo some of the damage... */ 711 ifp->if_baudrate = 0; 712 /*ifp->if_type = IFT_BRIDGE;*/ 713 714 crit_enter(); /* XXX MP */ 715 LIST_INSERT_HEAD(&bridge_list, sc, sc_list); 716 crit_exit(); 717 718 return (0); 719 } 720 721 static void 722 bridge_delete_dispatch(netmsg_t msg) 723 { 724 struct bridge_softc *sc = msg->lmsg.u.ms_resultp; 725 struct ifnet *bifp = sc->sc_ifp; 726 struct bridge_iflist *bif; 727 728 ifnet_serialize_all(bifp); 729 730 while ((bif = TAILQ_FIRST(&sc->sc_iflists[mycpuid])) != NULL) 731 bridge_delete_member(sc, bif, 0); 732 733 while ((bif = TAILQ_FIRST(&sc->sc_spanlist)) != NULL) 734 bridge_delete_span(sc, bif); 735 736 ifnet_deserialize_all(bifp); 737 738 lwkt_replymsg(&msg->lmsg, 0); 739 } 740 741 /* 742 * bridge_clone_destroy: 743 * 744 * Destroy a bridge instance. 745 */ 746 static int 747 bridge_clone_destroy(struct ifnet *ifp) 748 { 749 struct bridge_softc *sc = ifp->if_softc; 750 struct netmsg_base msg; 751 752 ifnet_serialize_all(ifp); 753 754 bridge_stop(ifp); 755 ifp->if_flags &= ~IFF_UP; 756 757 ifnet_deserialize_all(ifp); 758 759 netmsg_init(&msg, NULL, &curthread->td_msgport, 760 0, bridge_delete_dispatch); 761 msg.lmsg.u.ms_resultp = sc; 762 lwkt_domsg(BRIDGE_CFGPORT, &msg.lmsg, 0); 763 764 crit_enter(); /* XXX MP */ 765 LIST_REMOVE(sc, sc_list); 766 crit_exit(); 767 768 ether_ifdetach(ifp); 769 770 /* Tear down the routing table. */ 771 bridge_rtable_fini(sc); 772 773 /* Free per-cpu member iface lists */ 774 kfree(sc->sc_iflists, M_DEVBUF); 775 776 kfree(sc, M_DEVBUF); 777 778 return 0; 779 } 780 781 /* 782 * bridge_ioctl: 783 * 784 * Handle a control request from the operator. 785 */ 786 static int 787 bridge_ioctl(struct ifnet *ifp, u_long cmd, caddr_t data, struct ucred *cr) 788 { 789 struct bridge_softc *sc = ifp->if_softc; 790 struct bridge_control_arg args; 791 struct ifdrv *ifd = (struct ifdrv *) data; 792 const struct bridge_control *bc; 793 int error = 0; 794 795 ASSERT_IFNET_SERIALIZED_ALL(ifp); 796 797 switch (cmd) { 798 case SIOCADDMULTI: 799 case SIOCDELMULTI: 800 break; 801 802 case SIOCGDRVSPEC: 803 case SIOCSDRVSPEC: 804 if (ifd->ifd_cmd >= bridge_control_table_size) { 805 error = EINVAL; 806 break; 807 } 808 bc = &bridge_control_table[ifd->ifd_cmd]; 809 810 if (cmd == SIOCGDRVSPEC && 811 (bc->bc_flags & BC_F_COPYOUT) == 0) { 812 error = EINVAL; 813 break; 814 } else if (cmd == SIOCSDRVSPEC && 815 (bc->bc_flags & BC_F_COPYOUT)) { 816 error = EINVAL; 817 break; 818 } 819 820 if (bc->bc_flags & BC_F_SUSER) { 821 error = priv_check_cred(cr, PRIV_ROOT, NULL_CRED_OKAY); 822 if (error) 823 break; 824 } 825 826 if (ifd->ifd_len != bc->bc_argsize || 827 ifd->ifd_len > sizeof(args.bca_u)) { 828 error = EINVAL; 829 break; 830 } 831 832 memset(&args, 0, sizeof(args)); 833 if (bc->bc_flags & BC_F_COPYIN) { 834 error = copyin(ifd->ifd_data, &args.bca_u, 835 ifd->ifd_len); 836 if (error) 837 break; 838 } 839 840 error = bridge_control(sc, cmd, bc->bc_func, &args); 841 if (error) { 842 KKASSERT(args.bca_len == 0 && args.bca_kptr == NULL); 843 break; 844 } 845 846 if (bc->bc_flags & BC_F_COPYOUT) { 847 error = copyout(&args, ifd->ifd_data, ifd->ifd_len); 848 if (args.bca_len != 0) { 849 KKASSERT(args.bca_kptr != NULL); 850 if (!error) { 851 error = copyout(args.bca_kptr, 852 args.bca_uptr, args.bca_len); 853 } 854 kfree(args.bca_kptr, M_TEMP); 855 } else { 856 KKASSERT(args.bca_kptr == NULL); 857 } 858 } else { 859 KKASSERT(args.bca_len == 0 && args.bca_kptr == NULL); 860 } 861 break; 862 863 case SIOCSIFFLAGS: 864 if (!(ifp->if_flags & IFF_UP) && 865 (ifp->if_flags & IFF_RUNNING)) { 866 /* 867 * If interface is marked down and it is running, 868 * then stop it. 869 */ 870 bridge_stop(ifp); 871 } else if ((ifp->if_flags & IFF_UP) && 872 !(ifp->if_flags & IFF_RUNNING)) { 873 /* 874 * If interface is marked up and it is stopped, then 875 * start it. 876 */ 877 ifp->if_init(sc); 878 } 879 880 /* 881 * If running and link flag state change we have to 882 * reinitialize as well. 883 */ 884 if ((ifp->if_flags & IFF_RUNNING) && 885 (ifp->if_flags & (IFF_LINK0|IFF_LINK1|IFF_LINK2)) != 886 sc->sc_copy_flags) { 887 sc->sc_copy_flags = ifp->if_flags & 888 (IFF_LINK0|IFF_LINK1|IFF_LINK2); 889 bridge_control(sc, 0, bridge_ioctl_reinit, NULL); 890 } 891 892 break; 893 894 case SIOCSIFMTU: 895 /* Do not allow the MTU to be changed on the bridge */ 896 error = EINVAL; 897 break; 898 899 default: 900 error = ether_ioctl(ifp, cmd, data); 901 break; 902 } 903 return (error); 904 } 905 906 /* 907 * bridge_mutecaps: 908 * 909 * Clear or restore unwanted capabilities on the member interface 910 */ 911 static void 912 bridge_mutecaps(struct bridge_ifinfo *bif_info, struct ifnet *ifp, int mute) 913 { 914 struct ifreq ifr; 915 int error; 916 917 if (ifp->if_ioctl == NULL) 918 return; 919 920 bzero(&ifr, sizeof(ifr)); 921 ifr.ifr_reqcap = ifp->if_capenable; 922 923 if (mute) { 924 /* mask off and save capabilities */ 925 bif_info->bifi_mutecap = ifr.ifr_reqcap & BRIDGE_IFCAPS_MASK; 926 if (bif_info->bifi_mutecap != 0) 927 ifr.ifr_reqcap &= ~BRIDGE_IFCAPS_MASK; 928 } else { 929 /* restore muted capabilities */ 930 ifr.ifr_reqcap |= bif_info->bifi_mutecap; 931 } 932 933 if (bif_info->bifi_mutecap != 0) { 934 ifnet_serialize_all(ifp); 935 error = ifp->if_ioctl(ifp, SIOCSIFCAP, (caddr_t)&ifr, NULL); 936 ifnet_deserialize_all(ifp); 937 } 938 } 939 940 /* 941 * bridge_lookup_member: 942 * 943 * Lookup a bridge member interface. 944 */ 945 static struct bridge_iflist * 946 bridge_lookup_member(struct bridge_softc *sc, const char *name) 947 { 948 struct bridge_iflist *bif; 949 950 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) { 951 if (strcmp(bif->bif_ifp->if_xname, name) == 0) 952 return (bif); 953 } 954 return (NULL); 955 } 956 957 /* 958 * bridge_lookup_member_if: 959 * 960 * Lookup a bridge member interface by ifnet*. 961 */ 962 static struct bridge_iflist * 963 bridge_lookup_member_if(struct bridge_softc *sc, struct ifnet *member_ifp) 964 { 965 struct bridge_iflist *bif; 966 967 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) { 968 if (bif->bif_ifp == member_ifp) 969 return (bif); 970 } 971 return (NULL); 972 } 973 974 /* 975 * bridge_lookup_member_ifinfo: 976 * 977 * Lookup a bridge member interface by bridge_ifinfo. 978 */ 979 static struct bridge_iflist * 980 bridge_lookup_member_ifinfo(struct bridge_softc *sc, 981 struct bridge_ifinfo *bif_info) 982 { 983 struct bridge_iflist *bif; 984 985 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) { 986 if (bif->bif_info == bif_info) 987 return (bif); 988 } 989 return (NULL); 990 } 991 992 /* 993 * bridge_delete_member: 994 * 995 * Delete the specified member interface. 996 */ 997 static void 998 bridge_delete_member(struct bridge_softc *sc, struct bridge_iflist *bif, 999 int gone) 1000 { 1001 struct ifnet *ifs = bif->bif_ifp; 1002 struct ifnet *bifp = sc->sc_ifp; 1003 struct bridge_ifinfo *bif_info = bif->bif_info; 1004 struct bridge_iflist_head saved_bifs; 1005 1006 ASSERT_IFNET_SERIALIZED_ALL(bifp); 1007 KKASSERT(bif_info != NULL); 1008 1009 ifs->if_bridge = NULL; 1010 1011 /* 1012 * Release bridge interface's serializer: 1013 * - To avoid possible dead lock. 1014 * - Various sync operation will block the current thread. 1015 */ 1016 ifnet_deserialize_all(bifp); 1017 1018 if (!gone) { 1019 switch (ifs->if_type) { 1020 case IFT_ETHER: 1021 case IFT_L2VLAN: 1022 /* 1023 * Take the interface out of promiscuous mode. 1024 */ 1025 ifpromisc(ifs, 0); 1026 bridge_mutecaps(bif_info, ifs, 0); 1027 break; 1028 1029 case IFT_GIF: 1030 break; 1031 1032 default: 1033 panic("bridge_delete_member: impossible"); 1034 break; 1035 } 1036 } 1037 1038 /* 1039 * Remove bifs from percpu linked list. 1040 * 1041 * Removed bifs are not freed immediately, instead, 1042 * they are saved in saved_bifs. They will be freed 1043 * after we make sure that no one is accessing them, 1044 * i.e. after following netmsg_service_sync() 1045 */ 1046 TAILQ_INIT(&saved_bifs); 1047 bridge_del_bif(sc, bif_info, &saved_bifs); 1048 1049 /* 1050 * Make sure that all protocol threads: 1051 * o see 'ifs' if_bridge is changed 1052 * o know that bif is removed from the percpu linked list 1053 */ 1054 netmsg_service_sync(); 1055 1056 /* 1057 * Free the removed bifs 1058 */ 1059 KKASSERT(!TAILQ_EMPTY(&saved_bifs)); 1060 while ((bif = TAILQ_FIRST(&saved_bifs)) != NULL) { 1061 TAILQ_REMOVE(&saved_bifs, bif, bif_next); 1062 kfree(bif, M_DEVBUF); 1063 } 1064 1065 /* See the comment in bridge_ioctl_stop() */ 1066 bridge_rtmsg_sync(sc); 1067 bridge_rtdelete(sc, ifs, IFBF_FLUSHALL | IFBF_FLUSHSYNC); 1068 1069 ifnet_serialize_all(bifp); 1070 1071 if (bifp->if_flags & IFF_RUNNING) 1072 bstp_initialization(sc); 1073 1074 /* 1075 * Free the bif_info after bstp_initialization(), so that 1076 * bridge_softc.sc_root_port will not reference a dangling 1077 * pointer. 1078 */ 1079 kfree(bif_info, M_DEVBUF); 1080 } 1081 1082 /* 1083 * bridge_delete_span: 1084 * 1085 * Delete the specified span interface. 1086 */ 1087 static void 1088 bridge_delete_span(struct bridge_softc *sc, struct bridge_iflist *bif) 1089 { 1090 KASSERT(bif->bif_ifp->if_bridge == NULL, 1091 ("%s: not a span interface", __func__)); 1092 1093 TAILQ_REMOVE(&sc->sc_iflists[mycpuid], bif, bif_next); 1094 kfree(bif, M_DEVBUF); 1095 } 1096 1097 static int 1098 bridge_ioctl_init(struct bridge_softc *sc, void *arg __unused) 1099 { 1100 struct ifnet *ifp = sc->sc_ifp; 1101 1102 if (ifp->if_flags & IFF_RUNNING) 1103 return 0; 1104 1105 callout_reset(&sc->sc_brcallout, bridge_rtable_prune_period * hz, 1106 bridge_timer, sc); 1107 1108 ifp->if_flags |= IFF_RUNNING; 1109 bstp_initialization(sc); 1110 return 0; 1111 } 1112 1113 static int 1114 bridge_ioctl_stop(struct bridge_softc *sc, void *arg __unused) 1115 { 1116 struct ifnet *ifp = sc->sc_ifp; 1117 struct lwkt_msg *lmsg; 1118 1119 if ((ifp->if_flags & IFF_RUNNING) == 0) 1120 return 0; 1121 1122 callout_stop(&sc->sc_brcallout); 1123 1124 crit_enter(); 1125 lmsg = &sc->sc_brtimemsg.lmsg; 1126 if ((lmsg->ms_flags & MSGF_DONE) == 0) { 1127 /* Pending to be processed; drop it */ 1128 lwkt_dropmsg(lmsg); 1129 } 1130 crit_exit(); 1131 1132 bstp_stop(sc); 1133 1134 ifp->if_flags &= ~IFF_RUNNING; 1135 1136 ifnet_deserialize_all(ifp); 1137 1138 /* Let everyone know that we are stopped */ 1139 netmsg_service_sync(); 1140 1141 /* 1142 * Sync ifnetX msgports in the order we forward rtnode 1143 * installation message. This is used to make sure that 1144 * all rtnode installation messages sent by bridge_rtupdate() 1145 * during above netmsg_service_sync() are flushed. 1146 */ 1147 bridge_rtmsg_sync(sc); 1148 bridge_rtflush(sc, IFBF_FLUSHDYN | IFBF_FLUSHSYNC); 1149 1150 ifnet_serialize_all(ifp); 1151 return 0; 1152 } 1153 1154 static int 1155 bridge_ioctl_add(struct bridge_softc *sc, void *arg) 1156 { 1157 struct ifbreq *req = arg; 1158 struct bridge_iflist *bif; 1159 struct bridge_ifinfo *bif_info; 1160 struct ifnet *ifs, *bifp; 1161 int error = 0; 1162 1163 bifp = sc->sc_ifp; 1164 ASSERT_IFNET_SERIALIZED_ALL(bifp); 1165 1166 ifs = ifunit(req->ifbr_ifsname); 1167 if (ifs == NULL) 1168 return (ENOENT); 1169 1170 /* If it's in the span list, it can't be a member. */ 1171 TAILQ_FOREACH(bif, &sc->sc_spanlist, bif_next) 1172 if (ifs == bif->bif_ifp) 1173 return (EBUSY); 1174 1175 /* Allow the first Ethernet member to define the MTU */ 1176 if (ifs->if_type != IFT_GIF) { 1177 if (TAILQ_EMPTY(&sc->sc_iflists[mycpuid])) { 1178 bifp->if_mtu = ifs->if_mtu; 1179 } else if (bifp->if_mtu != ifs->if_mtu) { 1180 if_printf(bifp, "invalid MTU for %s\n", ifs->if_xname); 1181 return (EINVAL); 1182 } 1183 } 1184 1185 if (ifs->if_bridge == sc) 1186 return (EEXIST); 1187 1188 if (ifs->if_bridge != NULL) 1189 return (EBUSY); 1190 1191 bif_info = kmalloc(sizeof(*bif_info), M_DEVBUF, M_WAITOK | M_ZERO); 1192 bif_info->bifi_priority = BSTP_DEFAULT_PORT_PRIORITY; 1193 bif_info->bifi_path_cost = BSTP_DEFAULT_PATH_COST; 1194 bif_info->bifi_ifp = ifs; 1195 bif_info->bifi_bond_weight = 1; 1196 1197 /* 1198 * Release bridge interface's serializer: 1199 * - To avoid possible dead lock. 1200 * - Various sync operation will block the current thread. 1201 */ 1202 ifnet_deserialize_all(bifp); 1203 1204 switch (ifs->if_type) { 1205 case IFT_ETHER: 1206 case IFT_L2VLAN: 1207 /* 1208 * Place the interface into promiscuous mode. 1209 */ 1210 error = ifpromisc(ifs, 1); 1211 if (error) { 1212 ifnet_serialize_all(bifp); 1213 goto out; 1214 } 1215 bridge_mutecaps(bif_info, ifs, 1); 1216 break; 1217 1218 case IFT_GIF: /* :^) */ 1219 break; 1220 1221 default: 1222 error = EINVAL; 1223 ifnet_serialize_all(bifp); 1224 goto out; 1225 } 1226 1227 /* 1228 * Add bifs to percpu linked lists 1229 */ 1230 bridge_add_bif(sc, bif_info, ifs); 1231 1232 ifnet_serialize_all(bifp); 1233 1234 if (bifp->if_flags & IFF_RUNNING) 1235 bstp_initialization(sc); 1236 else 1237 bstp_stop(sc); 1238 1239 /* 1240 * Everything has been setup, so let the member interface 1241 * deliver packets to this bridge on its input/output path. 1242 */ 1243 ifs->if_bridge = sc; 1244 out: 1245 if (error) { 1246 if (bif_info != NULL) 1247 kfree(bif_info, M_DEVBUF); 1248 } 1249 return (error); 1250 } 1251 1252 static int 1253 bridge_ioctl_del(struct bridge_softc *sc, void *arg) 1254 { 1255 struct ifbreq *req = arg; 1256 struct bridge_iflist *bif; 1257 1258 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1259 if (bif == NULL) 1260 return (ENOENT); 1261 1262 bridge_delete_member(sc, bif, 0); 1263 1264 return (0); 1265 } 1266 1267 static int 1268 bridge_ioctl_gifflags(struct bridge_softc *sc, void *arg) 1269 { 1270 struct ifbreq *req = arg; 1271 struct bridge_iflist *bif; 1272 1273 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1274 if (bif == NULL) 1275 return (ENOENT); 1276 bridge_ioctl_fillflags(sc, bif, req); 1277 return (0); 1278 } 1279 1280 static void 1281 bridge_ioctl_fillflags(struct bridge_softc *sc, struct bridge_iflist *bif, 1282 struct ifbreq *req) 1283 { 1284 req->ifbr_ifsflags = bif->bif_flags; 1285 req->ifbr_state = bif->bif_state; 1286 req->ifbr_priority = bif->bif_priority; 1287 req->ifbr_path_cost = bif->bif_path_cost; 1288 req->ifbr_bond_weight = bif->bif_bond_weight; 1289 req->ifbr_portno = bif->bif_ifp->if_index & 0xff; 1290 if (bif->bif_flags & IFBIF_STP) { 1291 req->ifbr_peer_root = bif->bif_peer_root; 1292 req->ifbr_peer_bridge = bif->bif_peer_bridge; 1293 req->ifbr_peer_cost = bif->bif_peer_cost; 1294 req->ifbr_peer_port = bif->bif_peer_port; 1295 if (bstp_supersedes_port_info(sc, bif)) { 1296 req->ifbr_designated_root = bif->bif_peer_root; 1297 req->ifbr_designated_bridge = bif->bif_peer_bridge; 1298 req->ifbr_designated_cost = bif->bif_peer_cost; 1299 req->ifbr_designated_port = bif->bif_peer_port; 1300 } else { 1301 req->ifbr_designated_root = sc->sc_bridge_id; 1302 req->ifbr_designated_bridge = sc->sc_bridge_id; 1303 req->ifbr_designated_cost = bif->bif_path_cost + 1304 bif->bif_peer_cost; 1305 req->ifbr_designated_port = bif->bif_port_id; 1306 } 1307 } else { 1308 req->ifbr_peer_root = 0; 1309 req->ifbr_peer_bridge = 0; 1310 req->ifbr_peer_cost = 0; 1311 req->ifbr_peer_port = 0; 1312 req->ifbr_designated_root = 0; 1313 req->ifbr_designated_bridge = 0; 1314 req->ifbr_designated_cost = 0; 1315 req->ifbr_designated_port = 0; 1316 } 1317 } 1318 1319 static int 1320 bridge_ioctl_sifflags(struct bridge_softc *sc, void *arg) 1321 { 1322 struct ifbreq *req = arg; 1323 struct bridge_iflist *bif; 1324 struct ifnet *bifp = sc->sc_ifp; 1325 1326 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1327 if (bif == NULL) 1328 return (ENOENT); 1329 1330 if (req->ifbr_ifsflags & IFBIF_SPAN) { 1331 /* SPAN is readonly */ 1332 return (EINVAL); 1333 } 1334 1335 if (req->ifbr_ifsflags & IFBIF_STP) { 1336 switch (bif->bif_ifp->if_type) { 1337 case IFT_ETHER: 1338 /* These can do spanning tree. */ 1339 break; 1340 1341 default: 1342 /* Nothing else can. */ 1343 return (EINVAL); 1344 } 1345 } 1346 1347 bif->bif_flags = (bif->bif_flags & IFBIF_KEEPMASK) | 1348 (req->ifbr_ifsflags & ~IFBIF_KEEPMASK); 1349 if (bifp->if_flags & IFF_RUNNING) 1350 bstp_initialization(sc); 1351 1352 return (0); 1353 } 1354 1355 static int 1356 bridge_ioctl_scache(struct bridge_softc *sc, void *arg) 1357 { 1358 struct ifbrparam *param = arg; 1359 struct ifnet *ifp = sc->sc_ifp; 1360 1361 sc->sc_brtmax = param->ifbrp_csize; 1362 1363 ifnet_deserialize_all(ifp); 1364 bridge_rttrim(sc); 1365 ifnet_serialize_all(ifp); 1366 1367 return (0); 1368 } 1369 1370 static int 1371 bridge_ioctl_gcache(struct bridge_softc *sc, void *arg) 1372 { 1373 struct ifbrparam *param = arg; 1374 1375 param->ifbrp_csize = sc->sc_brtmax; 1376 1377 return (0); 1378 } 1379 1380 static int 1381 bridge_ioctl_gifs(struct bridge_softc *sc, void *arg) 1382 { 1383 struct bridge_control_arg *bc_arg = arg; 1384 struct ifbifconf *bifc = arg; 1385 struct bridge_iflist *bif; 1386 struct ifbreq *breq; 1387 int count, len; 1388 1389 count = 0; 1390 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) 1391 count++; 1392 TAILQ_FOREACH(bif, &sc->sc_spanlist, bif_next) 1393 count++; 1394 1395 if (bifc->ifbic_len == 0) { 1396 bifc->ifbic_len = sizeof(*breq) * count; 1397 return 0; 1398 } else if (count == 0 || bifc->ifbic_len < sizeof(*breq)) { 1399 bifc->ifbic_len = 0; 1400 return 0; 1401 } 1402 1403 len = min(bifc->ifbic_len, sizeof(*breq) * count); 1404 KKASSERT(len >= sizeof(*breq)); 1405 1406 breq = kmalloc(len, M_TEMP, M_WAITOK | M_NULLOK | M_ZERO); 1407 if (breq == NULL) { 1408 bifc->ifbic_len = 0; 1409 return ENOMEM; 1410 } 1411 bc_arg->bca_kptr = breq; 1412 1413 count = 0; 1414 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) { 1415 if (len < sizeof(*breq)) 1416 break; 1417 1418 strlcpy(breq->ifbr_ifsname, bif->bif_ifp->if_xname, 1419 sizeof(breq->ifbr_ifsname)); 1420 bridge_ioctl_fillflags(sc, bif, breq); 1421 breq++; 1422 count++; 1423 len -= sizeof(*breq); 1424 } 1425 TAILQ_FOREACH(bif, &sc->sc_spanlist, bif_next) { 1426 if (len < sizeof(*breq)) 1427 break; 1428 1429 strlcpy(breq->ifbr_ifsname, bif->bif_ifp->if_xname, 1430 sizeof(breq->ifbr_ifsname)); 1431 breq->ifbr_ifsflags = bif->bif_flags; 1432 breq->ifbr_portno = bif->bif_ifp->if_index & 0xff; 1433 breq++; 1434 count++; 1435 len -= sizeof(*breq); 1436 } 1437 1438 bifc->ifbic_len = sizeof(*breq) * count; 1439 KKASSERT(bifc->ifbic_len > 0); 1440 1441 bc_arg->bca_len = bifc->ifbic_len; 1442 bc_arg->bca_uptr = bifc->ifbic_req; 1443 return 0; 1444 } 1445 1446 static int 1447 bridge_ioctl_rts(struct bridge_softc *sc, void *arg) 1448 { 1449 struct bridge_control_arg *bc_arg = arg; 1450 struct ifbaconf *bac = arg; 1451 struct bridge_rtnode *brt; 1452 struct ifbareq *bareq; 1453 int count, len; 1454 1455 count = 0; 1456 LIST_FOREACH(brt, &sc->sc_rtlists[mycpuid], brt_list) 1457 count++; 1458 1459 if (bac->ifbac_len == 0) { 1460 bac->ifbac_len = sizeof(*bareq) * count; 1461 return 0; 1462 } else if (count == 0 || bac->ifbac_len < sizeof(*bareq)) { 1463 bac->ifbac_len = 0; 1464 return 0; 1465 } 1466 1467 len = min(bac->ifbac_len, sizeof(*bareq) * count); 1468 KKASSERT(len >= sizeof(*bareq)); 1469 1470 bareq = kmalloc(len, M_TEMP, M_WAITOK | M_NULLOK | M_ZERO); 1471 if (bareq == NULL) { 1472 bac->ifbac_len = 0; 1473 return ENOMEM; 1474 } 1475 bc_arg->bca_kptr = bareq; 1476 1477 count = 0; 1478 LIST_FOREACH(brt, &sc->sc_rtlists[mycpuid], brt_list) { 1479 struct bridge_rtinfo *bri = brt->brt_info; 1480 unsigned long expire; 1481 1482 if (len < sizeof(*bareq)) 1483 break; 1484 1485 strlcpy(bareq->ifba_ifsname, bri->bri_ifp->if_xname, 1486 sizeof(bareq->ifba_ifsname)); 1487 memcpy(bareq->ifba_dst, brt->brt_addr, sizeof(brt->brt_addr)); 1488 expire = bri->bri_expire; 1489 if ((bri->bri_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 1490 time_second < expire) 1491 bareq->ifba_expire = expire - time_second; 1492 else 1493 bareq->ifba_expire = 0; 1494 bareq->ifba_flags = bri->bri_flags; 1495 bareq++; 1496 count++; 1497 len -= sizeof(*bareq); 1498 } 1499 1500 bac->ifbac_len = sizeof(*bareq) * count; 1501 KKASSERT(bac->ifbac_len > 0); 1502 1503 bc_arg->bca_len = bac->ifbac_len; 1504 bc_arg->bca_uptr = bac->ifbac_req; 1505 return 0; 1506 } 1507 1508 static int 1509 bridge_ioctl_saddr(struct bridge_softc *sc, void *arg) 1510 { 1511 struct ifbareq *req = arg; 1512 struct bridge_iflist *bif; 1513 struct ifnet *ifp = sc->sc_ifp; 1514 int error; 1515 1516 ASSERT_IFNET_SERIALIZED_ALL(ifp); 1517 1518 bif = bridge_lookup_member(sc, req->ifba_ifsname); 1519 if (bif == NULL) 1520 return (ENOENT); 1521 1522 ifnet_deserialize_all(ifp); 1523 error = bridge_rtsaddr(sc, req->ifba_dst, bif->bif_ifp, 1524 req->ifba_flags); 1525 ifnet_serialize_all(ifp); 1526 return (error); 1527 } 1528 1529 static int 1530 bridge_ioctl_sto(struct bridge_softc *sc, void *arg) 1531 { 1532 struct ifbrparam *param = arg; 1533 1534 sc->sc_brttimeout = param->ifbrp_ctime; 1535 1536 return (0); 1537 } 1538 1539 static int 1540 bridge_ioctl_gto(struct bridge_softc *sc, void *arg) 1541 { 1542 struct ifbrparam *param = arg; 1543 1544 param->ifbrp_ctime = sc->sc_brttimeout; 1545 1546 return (0); 1547 } 1548 1549 static int 1550 bridge_ioctl_daddr(struct bridge_softc *sc, void *arg) 1551 { 1552 struct ifbareq *req = arg; 1553 struct ifnet *ifp = sc->sc_ifp; 1554 int error; 1555 1556 ifnet_deserialize_all(ifp); 1557 error = bridge_rtdaddr(sc, req->ifba_dst); 1558 ifnet_serialize_all(ifp); 1559 return error; 1560 } 1561 1562 static int 1563 bridge_ioctl_flush(struct bridge_softc *sc, void *arg) 1564 { 1565 struct ifbreq *req = arg; 1566 struct ifnet *ifp = sc->sc_ifp; 1567 1568 ifnet_deserialize_all(ifp); 1569 bridge_rtflush(sc, req->ifbr_ifsflags | IFBF_FLUSHSYNC); 1570 ifnet_serialize_all(ifp); 1571 1572 return (0); 1573 } 1574 1575 static int 1576 bridge_ioctl_gpri(struct bridge_softc *sc, void *arg) 1577 { 1578 struct ifbrparam *param = arg; 1579 1580 param->ifbrp_prio = sc->sc_bridge_priority; 1581 1582 return (0); 1583 } 1584 1585 static int 1586 bridge_ioctl_spri(struct bridge_softc *sc, void *arg) 1587 { 1588 struct ifbrparam *param = arg; 1589 1590 sc->sc_bridge_priority = param->ifbrp_prio; 1591 1592 if (sc->sc_ifp->if_flags & IFF_RUNNING) 1593 bstp_initialization(sc); 1594 1595 return (0); 1596 } 1597 1598 static int 1599 bridge_ioctl_reinit(struct bridge_softc *sc, void *arg __unused) 1600 { 1601 if (sc->sc_ifp->if_flags & IFF_RUNNING) 1602 bstp_initialization(sc); 1603 return (0); 1604 } 1605 1606 static int 1607 bridge_ioctl_ght(struct bridge_softc *sc, void *arg) 1608 { 1609 struct ifbrparam *param = arg; 1610 1611 param->ifbrp_hellotime = sc->sc_bridge_hello_time >> 8; 1612 1613 return (0); 1614 } 1615 1616 static int 1617 bridge_ioctl_sht(struct bridge_softc *sc, void *arg) 1618 { 1619 struct ifbrparam *param = arg; 1620 1621 if (param->ifbrp_hellotime == 0) 1622 return (EINVAL); 1623 sc->sc_bridge_hello_time = param->ifbrp_hellotime << 8; 1624 1625 if (sc->sc_ifp->if_flags & IFF_RUNNING) 1626 bstp_initialization(sc); 1627 1628 return (0); 1629 } 1630 1631 static int 1632 bridge_ioctl_gfd(struct bridge_softc *sc, void *arg) 1633 { 1634 struct ifbrparam *param = arg; 1635 1636 param->ifbrp_fwddelay = sc->sc_bridge_forward_delay >> 8; 1637 1638 return (0); 1639 } 1640 1641 static int 1642 bridge_ioctl_sfd(struct bridge_softc *sc, void *arg) 1643 { 1644 struct ifbrparam *param = arg; 1645 1646 if (param->ifbrp_fwddelay == 0) 1647 return (EINVAL); 1648 sc->sc_bridge_forward_delay = param->ifbrp_fwddelay << 8; 1649 1650 if (sc->sc_ifp->if_flags & IFF_RUNNING) 1651 bstp_initialization(sc); 1652 1653 return (0); 1654 } 1655 1656 static int 1657 bridge_ioctl_gma(struct bridge_softc *sc, void *arg) 1658 { 1659 struct ifbrparam *param = arg; 1660 1661 param->ifbrp_maxage = sc->sc_bridge_max_age >> 8; 1662 1663 return (0); 1664 } 1665 1666 static int 1667 bridge_ioctl_sma(struct bridge_softc *sc, void *arg) 1668 { 1669 struct ifbrparam *param = arg; 1670 1671 if (param->ifbrp_maxage == 0) 1672 return (EINVAL); 1673 sc->sc_bridge_max_age = param->ifbrp_maxage << 8; 1674 1675 if (sc->sc_ifp->if_flags & IFF_RUNNING) 1676 bstp_initialization(sc); 1677 1678 return (0); 1679 } 1680 1681 static int 1682 bridge_ioctl_sifprio(struct bridge_softc *sc, void *arg) 1683 { 1684 struct ifbreq *req = arg; 1685 struct bridge_iflist *bif; 1686 1687 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1688 if (bif == NULL) 1689 return (ENOENT); 1690 1691 bif->bif_priority = req->ifbr_priority; 1692 1693 if (sc->sc_ifp->if_flags & IFF_RUNNING) 1694 bstp_initialization(sc); 1695 1696 return (0); 1697 } 1698 1699 static int 1700 bridge_ioctl_sifcost(struct bridge_softc *sc, void *arg) 1701 { 1702 struct ifbreq *req = arg; 1703 struct bridge_iflist *bif; 1704 1705 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1706 if (bif == NULL) 1707 return (ENOENT); 1708 1709 bif->bif_path_cost = req->ifbr_path_cost; 1710 1711 if (sc->sc_ifp->if_flags & IFF_RUNNING) 1712 bstp_initialization(sc); 1713 1714 return (0); 1715 } 1716 1717 static int 1718 bridge_ioctl_sifbondwght(struct bridge_softc *sc, void *arg) 1719 { 1720 struct ifbreq *req = arg; 1721 struct bridge_iflist *bif; 1722 1723 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1724 if (bif == NULL) 1725 return (ENOENT); 1726 1727 bif->bif_bond_weight = req->ifbr_bond_weight; 1728 1729 /* no reinit needed */ 1730 1731 return (0); 1732 } 1733 1734 static int 1735 bridge_ioctl_addspan(struct bridge_softc *sc, void *arg) 1736 { 1737 struct ifbreq *req = arg; 1738 struct bridge_iflist *bif; 1739 struct ifnet *ifs; 1740 struct bridge_ifinfo *bif_info; 1741 1742 ifs = ifunit(req->ifbr_ifsname); 1743 if (ifs == NULL) 1744 return (ENOENT); 1745 1746 TAILQ_FOREACH(bif, &sc->sc_spanlist, bif_next) 1747 if (ifs == bif->bif_ifp) 1748 return (EBUSY); 1749 1750 if (ifs->if_bridge != NULL) 1751 return (EBUSY); 1752 1753 switch (ifs->if_type) { 1754 case IFT_ETHER: 1755 case IFT_GIF: 1756 case IFT_L2VLAN: 1757 break; 1758 1759 default: 1760 return (EINVAL); 1761 } 1762 1763 /* 1764 * bif_info is needed for bif_flags 1765 */ 1766 bif_info = kmalloc(sizeof(*bif_info), M_DEVBUF, M_WAITOK | M_ZERO); 1767 bif_info->bifi_ifp = ifs; 1768 1769 bif = kmalloc(sizeof(*bif), M_DEVBUF, M_WAITOK | M_ZERO); 1770 bif->bif_ifp = ifs; 1771 bif->bif_info = bif_info; 1772 bif->bif_flags = IFBIF_SPAN; 1773 /* NOTE: span bif does not need bridge_ifinfo */ 1774 1775 TAILQ_INSERT_HEAD(&sc->sc_spanlist, bif, bif_next); 1776 1777 sc->sc_span = 1; 1778 1779 return (0); 1780 } 1781 1782 static int 1783 bridge_ioctl_delspan(struct bridge_softc *sc, void *arg) 1784 { 1785 struct ifbreq *req = arg; 1786 struct bridge_iflist *bif; 1787 struct ifnet *ifs; 1788 1789 ifs = ifunit(req->ifbr_ifsname); 1790 if (ifs == NULL) 1791 return (ENOENT); 1792 1793 TAILQ_FOREACH(bif, &sc->sc_spanlist, bif_next) 1794 if (ifs == bif->bif_ifp) 1795 break; 1796 1797 if (bif == NULL) 1798 return (ENOENT); 1799 1800 bridge_delete_span(sc, bif); 1801 1802 if (TAILQ_EMPTY(&sc->sc_spanlist)) 1803 sc->sc_span = 0; 1804 1805 return (0); 1806 } 1807 1808 static void 1809 bridge_ifdetach_dispatch(netmsg_t msg) 1810 { 1811 struct ifnet *ifp, *bifp; 1812 struct bridge_softc *sc; 1813 struct bridge_iflist *bif; 1814 1815 ifp = msg->lmsg.u.ms_resultp; 1816 sc = ifp->if_bridge; 1817 1818 /* Check if the interface is a bridge member */ 1819 if (sc != NULL) { 1820 bifp = sc->sc_ifp; 1821 1822 ifnet_serialize_all(bifp); 1823 1824 bif = bridge_lookup_member_if(sc, ifp); 1825 if (bif != NULL) { 1826 bridge_delete_member(sc, bif, 1); 1827 } else { 1828 /* XXX Why bif will be NULL? */ 1829 } 1830 1831 ifnet_deserialize_all(bifp); 1832 goto reply; 1833 } 1834 1835 crit_enter(); /* XXX MP */ 1836 1837 /* Check if the interface is a span port */ 1838 LIST_FOREACH(sc, &bridge_list, sc_list) { 1839 bifp = sc->sc_ifp; 1840 1841 ifnet_serialize_all(bifp); 1842 1843 TAILQ_FOREACH(bif, &sc->sc_spanlist, bif_next) 1844 if (ifp == bif->bif_ifp) { 1845 bridge_delete_span(sc, bif); 1846 break; 1847 } 1848 1849 ifnet_deserialize_all(bifp); 1850 } 1851 1852 crit_exit(); 1853 1854 reply: 1855 lwkt_replymsg(&msg->lmsg, 0); 1856 } 1857 1858 /* 1859 * bridge_ifdetach: 1860 * 1861 * Detach an interface from a bridge. Called when a member 1862 * interface is detaching. 1863 */ 1864 static void 1865 bridge_ifdetach(void *arg __unused, struct ifnet *ifp) 1866 { 1867 struct netmsg_base msg; 1868 1869 netmsg_init(&msg, NULL, &curthread->td_msgport, 1870 0, bridge_ifdetach_dispatch); 1871 msg.lmsg.u.ms_resultp = ifp; 1872 1873 lwkt_domsg(BRIDGE_CFGPORT, &msg.lmsg, 0); 1874 } 1875 1876 /* 1877 * bridge_init: 1878 * 1879 * Initialize a bridge interface. 1880 */ 1881 static void 1882 bridge_init(void *xsc) 1883 { 1884 bridge_control(xsc, SIOCSIFFLAGS, bridge_ioctl_init, NULL); 1885 } 1886 1887 /* 1888 * bridge_stop: 1889 * 1890 * Stop the bridge interface. 1891 */ 1892 static void 1893 bridge_stop(struct ifnet *ifp) 1894 { 1895 bridge_control(ifp->if_softc, SIOCSIFFLAGS, bridge_ioctl_stop, NULL); 1896 } 1897 1898 /* 1899 * Returns TRUE if the packet is being sent 'from us'... from our bridge 1900 * interface or from any member of our bridge interface. This is used 1901 * later on to force the MAC to be the MAC of our bridge interface. 1902 */ 1903 static int 1904 bridge_from_us(struct bridge_softc *sc, struct ether_header *eh) 1905 { 1906 struct bridge_iflist *bif; 1907 1908 if (memcmp(eh->ether_shost, IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN) == 0) 1909 return (1); 1910 1911 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) { 1912 if (memcmp(eh->ether_shost, IF_LLADDR(bif->bif_ifp), 1913 ETHER_ADDR_LEN) == 0) { 1914 return (1); 1915 } 1916 } 1917 return (0); 1918 } 1919 1920 /* 1921 * bridge_enqueue: 1922 * 1923 * Enqueue a packet on a bridge member interface. 1924 * 1925 */ 1926 void 1927 bridge_enqueue(struct ifnet *dst_ifp, struct mbuf *m) 1928 { 1929 struct netmsg_packet *nmp; 1930 1931 mbuftrackid(m, 64); 1932 1933 nmp = &m->m_hdr.mh_netmsg; 1934 netmsg_init(&nmp->base, NULL, &netisr_apanic_rport, 1935 0, bridge_enqueue_handler); 1936 nmp->nm_packet = m; 1937 nmp->base.lmsg.u.ms_resultp = dst_ifp; 1938 1939 lwkt_sendmsg(netisr_portfn(mycpu->gd_cpuid), &nmp->base.lmsg); 1940 } 1941 1942 /* 1943 * After looking up dst_if in our forwarding table we still have to 1944 * deal with channel bonding. Find the best interface in the bonding set. 1945 */ 1946 static struct ifnet * 1947 bridge_select_unicast(struct bridge_softc *sc, struct ifnet *dst_if, 1948 int from_blocking, struct mbuf *m) 1949 { 1950 struct bridge_iflist *bif, *nbif; 1951 struct ifnet *alt_if; 1952 int alt_priority; 1953 int priority; 1954 1955 /* 1956 * Unicast, kinda replicates the output side of bridge_output(). 1957 * 1958 * Even though this is a uni-cast packet we may have to select 1959 * an interface from a bonding set. 1960 */ 1961 bif = bridge_lookup_member_if(sc, dst_if); 1962 if (bif == NULL) { 1963 /* Not a member of the bridge (anymore?) */ 1964 return NULL; 1965 } 1966 1967 /* 1968 * If STP is enabled on the target we are an equal opportunity 1969 * employer and do not necessarily output to dst_if. Instead 1970 * scan available links with the same MAC as the current dst_if 1971 * and choose the best one. 1972 * 1973 * We also need to do this because arp entries tag onto a particular 1974 * interface and if it happens to be dead then the packets will 1975 * go into a bit bucket. 1976 * 1977 * If LINK2 is set the matching links are bonded and we-round robin. 1978 * (the MAC address must be the same for the participating links). 1979 * In this case links in a STP FORWARDING or BONDED state are 1980 * allowed for unicast packets. 1981 */ 1982 if (bif->bif_flags & IFBIF_STP) { 1983 alt_if = NULL; 1984 alt_priority = 0; 1985 priority = 0; 1986 1987 TAILQ_FOREACH_MUTABLE(bif, &sc->sc_iflists[mycpuid], 1988 bif_next, nbif) { 1989 /* 1990 * dst_if may imply a bonding set so we must compare 1991 * MAC addresses. 1992 */ 1993 if (memcmp(IF_LLADDR(bif->bif_ifp), 1994 IF_LLADDR(dst_if), 1995 ETHER_ADDR_LEN) != 0) { 1996 continue; 1997 } 1998 1999 if ((bif->bif_ifp->if_flags & IFF_RUNNING) == 0) 2000 continue; 2001 2002 /* 2003 * NOTE: We allow tranmissions through a BLOCKING 2004 * or LEARNING interface only as a last resort. 2005 * We DISALLOW both cases if the receiving 2006 * 2007 * NOTE: If we send a packet through a learning 2008 * interface the receiving end (if also in 2009 * LEARNING) will throw it away, so this is 2010 * the ultimate last resort. 2011 */ 2012 switch(bif->bif_state) { 2013 case BSTP_IFSTATE_BLOCKING: 2014 if (from_blocking == 0 && 2015 bif->bif_priority + 256 > alt_priority) { 2016 alt_priority = bif->bif_priority + 256; 2017 alt_if = bif->bif_ifp; 2018 } 2019 continue; 2020 case BSTP_IFSTATE_LEARNING: 2021 if (from_blocking == 0 && 2022 bif->bif_priority > alt_priority) { 2023 alt_priority = bif->bif_priority; 2024 alt_if = bif->bif_ifp; 2025 } 2026 continue; 2027 case BSTP_IFSTATE_L1BLOCKING: 2028 case BSTP_IFSTATE_LISTENING: 2029 case BSTP_IFSTATE_DISABLED: 2030 continue; 2031 default: 2032 /* FORWARDING, BONDED */ 2033 break; 2034 } 2035 2036 /* 2037 * XXX we need to use the toepliz hash or 2038 * something like that instead of 2039 * round-robining. 2040 */ 2041 if (sc->sc_ifp->if_flags & IFF_LINK2) { 2042 dst_if = bif->bif_ifp; 2043 if (++bif->bif_bond_count >= 2044 bif->bif_bond_weight) { 2045 bif->bif_bond_count = 0; 2046 TAILQ_REMOVE(&sc->sc_iflists[mycpuid], 2047 bif, bif_next); 2048 TAILQ_INSERT_TAIL( 2049 &sc->sc_iflists[mycpuid], 2050 bif, bif_next); 2051 } 2052 priority = 1; 2053 break; 2054 } 2055 2056 /* 2057 * Select best interface in the FORWARDING or 2058 * BONDED set. Well, there shouldn't be any 2059 * in a BONDED state if LINK2 is not set (they 2060 * will all be in a BLOCKING) state, but there 2061 * could be a transitory condition here. 2062 */ 2063 if (bif->bif_priority > priority) { 2064 priority = bif->bif_priority; 2065 dst_if = bif->bif_ifp; 2066 } 2067 } 2068 2069 /* 2070 * If no suitable interfaces were found but a suitable 2071 * alternative interface was found, use the alternative 2072 * interface. 2073 */ 2074 if (priority == 0 && alt_if) 2075 dst_if = alt_if; 2076 } 2077 2078 /* 2079 * At this point, we're dealing with a unicast frame 2080 * going to a different interface. 2081 */ 2082 if ((dst_if->if_flags & IFF_RUNNING) == 0) 2083 dst_if = NULL; 2084 return (dst_if); 2085 } 2086 2087 2088 /* 2089 * bridge_output: 2090 * 2091 * Send output from a bridge member interface. This 2092 * performs the bridging function for locally originated 2093 * packets. 2094 * 2095 * The mbuf has the Ethernet header already attached. We must 2096 * enqueue or free the mbuf before returning. 2097 */ 2098 static int 2099 bridge_output(struct ifnet *ifp, struct mbuf *m) 2100 { 2101 struct bridge_softc *sc = ifp->if_bridge; 2102 struct bridge_iflist *bif, *nbif; 2103 struct ether_header *eh; 2104 struct ifnet *dst_if, *alt_if, *bifp; 2105 int from_us; 2106 int alt_priority; 2107 2108 ASSERT_IFNET_NOT_SERIALIZED_ALL(ifp); 2109 mbuftrackid(m, 65); 2110 2111 /* 2112 * Make sure that we are still a member of a bridge interface. 2113 */ 2114 if (sc == NULL) { 2115 m_freem(m); 2116 return (0); 2117 } 2118 bifp = sc->sc_ifp; 2119 2120 /* 2121 * Acquire header 2122 */ 2123 if (m->m_len < ETHER_HDR_LEN) { 2124 m = m_pullup(m, ETHER_HDR_LEN); 2125 if (m == NULL) { 2126 bifp->if_oerrors++; 2127 return (0); 2128 } 2129 } 2130 eh = mtod(m, struct ether_header *); 2131 from_us = bridge_from_us(sc, eh); 2132 2133 /* 2134 * If bridge is down, but the original output interface is up, 2135 * go ahead and send out that interface. Otherwise, the packet 2136 * is dropped below. 2137 */ 2138 if ((bifp->if_flags & IFF_RUNNING) == 0) { 2139 dst_if = ifp; 2140 goto sendunicast; 2141 } 2142 2143 /* 2144 * If the packet is a multicast, or we don't know a better way to 2145 * get there, send to all interfaces. 2146 */ 2147 if (ETHER_IS_MULTICAST(eh->ether_dhost)) 2148 dst_if = NULL; 2149 else 2150 dst_if = bridge_rtlookup(sc, eh->ether_dhost); 2151 2152 if (dst_if == NULL) { 2153 struct mbuf *mc; 2154 int used = 0; 2155 int found = 0; 2156 2157 if (sc->sc_span) 2158 bridge_span(sc, m); 2159 2160 alt_if = NULL; 2161 alt_priority = 0; 2162 TAILQ_FOREACH_MUTABLE(bif, &sc->sc_iflists[mycpuid], 2163 bif_next, nbif) { 2164 dst_if = bif->bif_ifp; 2165 2166 if ((dst_if->if_flags & IFF_RUNNING) == 0) 2167 continue; 2168 2169 /* 2170 * If this is not the original output interface, 2171 * and the interface is participating in spanning 2172 * tree, make sure the port is in a state that 2173 * allows forwarding. 2174 * 2175 * We keep track of a possible backup IF if we are 2176 * unable to find any interfaces to forward through. 2177 * 2178 * NOTE: Currently round-robining is not implemented 2179 * across bonded interface groups (needs an 2180 * algorithm to track each group somehow). 2181 * 2182 * Similarly we track only one alternative 2183 * interface if no suitable interfaces are 2184 * found. 2185 */ 2186 if (dst_if != ifp && 2187 (bif->bif_flags & IFBIF_STP) != 0) { 2188 switch (bif->bif_state) { 2189 case BSTP_IFSTATE_BONDED: 2190 if (bif->bif_priority + 512 > 2191 alt_priority) { 2192 alt_priority = 2193 bif->bif_priority + 512; 2194 alt_if = bif->bif_ifp; 2195 } 2196 continue; 2197 case BSTP_IFSTATE_BLOCKING: 2198 if (bif->bif_priority + 256 > 2199 alt_priority) { 2200 alt_priority = 2201 bif->bif_priority + 256; 2202 alt_if = bif->bif_ifp; 2203 } 2204 continue; 2205 case BSTP_IFSTATE_LEARNING: 2206 if (bif->bif_priority > alt_priority) { 2207 alt_priority = 2208 bif->bif_priority; 2209 alt_if = bif->bif_ifp; 2210 } 2211 continue; 2212 case BSTP_IFSTATE_L1BLOCKING: 2213 case BSTP_IFSTATE_LISTENING: 2214 case BSTP_IFSTATE_DISABLED: 2215 continue; 2216 default: 2217 /* FORWARDING */ 2218 break; 2219 } 2220 } 2221 2222 KKASSERT(used == 0); 2223 if (TAILQ_NEXT(bif, bif_next) == NULL) { 2224 used = 1; 2225 mc = m; 2226 } else { 2227 mc = m_copypacket(m, MB_DONTWAIT); 2228 if (mc == NULL) { 2229 bifp->if_oerrors++; 2230 continue; 2231 } 2232 } 2233 2234 /* 2235 * If the packet is 'from' us override ether_shost. 2236 */ 2237 bridge_handoff(sc, dst_if, mc, from_us); 2238 found = 1; 2239 2240 if (nbif != NULL && !nbif->bif_onlist) { 2241 KKASSERT(bif->bif_onlist); 2242 nbif = TAILQ_NEXT(bif, bif_next); 2243 } 2244 } 2245 2246 /* 2247 * If we couldn't find anything use the backup interface 2248 * if we have one. 2249 */ 2250 if (found == 0 && alt_if) { 2251 KKASSERT(used == 0); 2252 mc = m; 2253 used = 1; 2254 bridge_handoff(sc, alt_if, mc, from_us); 2255 } 2256 2257 if (used == 0) 2258 m_freem(m); 2259 return (0); 2260 } 2261 2262 /* 2263 * Unicast 2264 */ 2265 sendunicast: 2266 dst_if = bridge_select_unicast(sc, dst_if, 0, m); 2267 2268 if (sc->sc_span) 2269 bridge_span(sc, m); 2270 if (dst_if == NULL) 2271 m_freem(m); 2272 else 2273 bridge_handoff(sc, dst_if, m, from_us); 2274 return (0); 2275 } 2276 2277 /* 2278 * Returns the bridge interface associated with an ifc. 2279 * Pass ifp->if_bridge (must not be NULL). Used by the ARP 2280 * code to supply the bridge for the is-at info, making 2281 * the bridge responsible for matching local addresses. 2282 * 2283 * Without this the ARP code will supply bridge member interfaces 2284 * for the is-at which makes it difficult the bridge to fail-over 2285 * interfaces (amoung other things). 2286 */ 2287 static struct ifnet * 2288 bridge_interface(void *if_bridge) 2289 { 2290 struct bridge_softc *sc = if_bridge; 2291 return (sc->sc_ifp); 2292 } 2293 2294 /* 2295 * bridge_start: 2296 * 2297 * Start output on a bridge. 2298 */ 2299 static void 2300 bridge_start(struct ifnet *ifp, struct ifaltq_subque *ifsq) 2301 { 2302 struct bridge_softc *sc = ifp->if_softc; 2303 2304 ASSERT_ALTQ_SQ_DEFAULT(ifp, ifsq); 2305 ASSERT_IFNET_SERIALIZED_TX(ifp, ifsq); 2306 2307 ifsq_set_oactive(ifsq); 2308 for (;;) { 2309 struct ifnet *dst_if = NULL; 2310 struct ether_header *eh; 2311 struct mbuf *m; 2312 2313 m = ifsq_dequeue(ifsq, NULL); 2314 if (m == NULL) 2315 break; 2316 mbuftrackid(m, 75); 2317 2318 if (m->m_len < sizeof(*eh)) { 2319 m = m_pullup(m, sizeof(*eh)); 2320 if (m == NULL) { 2321 ifp->if_oerrors++; 2322 continue; 2323 } 2324 } 2325 eh = mtod(m, struct ether_header *); 2326 2327 BPF_MTAP(ifp, m); 2328 ifp->if_opackets++; 2329 2330 if ((m->m_flags & (M_BCAST|M_MCAST)) == 0) 2331 dst_if = bridge_rtlookup(sc, eh->ether_dhost); 2332 2333 /* 2334 * Multicast or broadcast 2335 */ 2336 if (dst_if == NULL) { 2337 bridge_start_bcast(sc, m); 2338 continue; 2339 } 2340 2341 /* 2342 * Unicast 2343 */ 2344 dst_if = bridge_select_unicast(sc, dst_if, 0, m); 2345 2346 if (dst_if == NULL) 2347 m_freem(m); 2348 else 2349 bridge_enqueue(dst_if, m); 2350 } 2351 ifsq_clr_oactive(ifsq); 2352 } 2353 2354 /* 2355 * bridge_forward: 2356 * 2357 * Forward packets received on a bridge interface via the input 2358 * path. 2359 * 2360 * This implements the forwarding function of the bridge. 2361 */ 2362 static void 2363 bridge_forward(struct bridge_softc *sc, struct mbuf *m) 2364 { 2365 struct bridge_iflist *bif; 2366 struct ifnet *src_if, *dst_if, *ifp; 2367 struct ether_header *eh; 2368 int from_blocking; 2369 2370 mbuftrackid(m, 66); 2371 src_if = m->m_pkthdr.rcvif; 2372 ifp = sc->sc_ifp; 2373 2374 ASSERT_IFNET_NOT_SERIALIZED_ALL(ifp); 2375 2376 ifp->if_ipackets++; 2377 ifp->if_ibytes += m->m_pkthdr.len; 2378 2379 /* 2380 * Look up the bridge_iflist. 2381 */ 2382 bif = bridge_lookup_member_if(sc, src_if); 2383 if (bif == NULL) { 2384 /* Interface is not a bridge member (anymore?) */ 2385 m_freem(m); 2386 return; 2387 } 2388 2389 /* 2390 * In spanning tree mode receiving a packet from an interface 2391 * in a BLOCKING state is allowed, it could be a member of last 2392 * resort from the sender's point of view, but forwarding it is 2393 * not allowed. 2394 * 2395 * The sender's spanning tree will eventually sync up and the 2396 * sender will go into a BLOCKING state too (but this still may be 2397 * an interface of last resort during state changes). 2398 */ 2399 if (bif->bif_flags & IFBIF_STP) { 2400 switch (bif->bif_state) { 2401 case BSTP_IFSTATE_L1BLOCKING: 2402 case BSTP_IFSTATE_LISTENING: 2403 case BSTP_IFSTATE_DISABLED: 2404 m_freem(m); 2405 return; 2406 default: 2407 /* learning, blocking, bonded, forwarding */ 2408 break; 2409 } 2410 } 2411 from_blocking = (bif->bif_state == BSTP_IFSTATE_BLOCKING); 2412 2413 eh = mtod(m, struct ether_header *); 2414 2415 /* 2416 * If the interface is learning, and the source 2417 * address is valid and not multicast, record 2418 * the address. 2419 */ 2420 if ((bif->bif_flags & IFBIF_LEARNING) != 0 && 2421 from_blocking == 0 && 2422 ETHER_IS_MULTICAST(eh->ether_shost) == 0 && 2423 (eh->ether_shost[0] == 0 && 2424 eh->ether_shost[1] == 0 && 2425 eh->ether_shost[2] == 0 && 2426 eh->ether_shost[3] == 0 && 2427 eh->ether_shost[4] == 0 && 2428 eh->ether_shost[5] == 0) == 0) { 2429 bridge_rtupdate(sc, eh->ether_shost, src_if, IFBAF_DYNAMIC); 2430 } 2431 2432 /* 2433 * Don't forward from an interface in the listening or learning 2434 * state. That is, in the learning state we learn information 2435 * but we throw away the packets. 2436 * 2437 * We let through packets on interfaces in the blocking state. 2438 * The blocking state is applicable to the send side, not the 2439 * receive side. 2440 */ 2441 if ((bif->bif_flags & IFBIF_STP) != 0 && 2442 (bif->bif_state == BSTP_IFSTATE_LISTENING || 2443 bif->bif_state == BSTP_IFSTATE_LEARNING)) { 2444 m_freem(m); 2445 return; 2446 } 2447 2448 /* 2449 * At this point, the port either doesn't participate 2450 * in spanning tree or it is in the forwarding state. 2451 */ 2452 2453 /* 2454 * If the packet is unicast, destined for someone on 2455 * "this" side of the bridge, drop it. 2456 * 2457 * src_if implies the entire bonding set so we have to compare MAC 2458 * addresses and not just if pointers. 2459 */ 2460 if ((m->m_flags & (M_BCAST|M_MCAST)) == 0) { 2461 dst_if = bridge_rtlookup(sc, eh->ether_dhost); 2462 if (dst_if && memcmp(IF_LLADDR(src_if), IF_LLADDR(dst_if), 2463 ETHER_ADDR_LEN) == 0) { 2464 m_freem(m); 2465 return; 2466 } 2467 } else { 2468 /* ...forward it to all interfaces. */ 2469 ifp->if_imcasts++; 2470 dst_if = NULL; 2471 } 2472 2473 /* 2474 * Brodcast if we do not have forwarding information. However, if 2475 * we received the packet on a blocking interface we do not do this 2476 * (unless you really want to blow up your network). 2477 */ 2478 if (dst_if == NULL) { 2479 if (from_blocking) 2480 m_freem(m); 2481 else 2482 bridge_broadcast(sc, src_if, m); 2483 return; 2484 } 2485 2486 dst_if = bridge_select_unicast(sc, dst_if, from_blocking, m); 2487 2488 if (dst_if == NULL) { 2489 m_freem(m); 2490 return; 2491 } 2492 2493 if (inet_pfil_hook.ph_hashooks > 0 2494 #ifdef INET6 2495 || inet6_pfil_hook.ph_hashooks > 0 2496 #endif 2497 ) { 2498 if (bridge_pfil(&m, ifp, src_if, PFIL_IN) != 0) 2499 return; 2500 if (m == NULL) 2501 return; 2502 2503 if (bridge_pfil(&m, ifp, dst_if, PFIL_OUT) != 0) 2504 return; 2505 if (m == NULL) 2506 return; 2507 } 2508 bridge_handoff(sc, dst_if, m, 0); 2509 } 2510 2511 /* 2512 * bridge_input: 2513 * 2514 * Receive input from a member interface. Queue the packet for 2515 * bridging if it is not for us. 2516 */ 2517 static struct mbuf * 2518 bridge_input(struct ifnet *ifp, struct mbuf *m) 2519 { 2520 struct bridge_softc *sc = ifp->if_bridge; 2521 struct bridge_iflist *bif; 2522 struct ifnet *bifp, *new_ifp; 2523 struct ether_header *eh; 2524 struct mbuf *mc, *mc2; 2525 2526 ASSERT_IFNET_NOT_SERIALIZED_ALL(ifp); 2527 mbuftrackid(m, 67); 2528 2529 /* 2530 * Make sure that we are still a member of a bridge interface. 2531 */ 2532 if (sc == NULL) 2533 return m; 2534 2535 new_ifp = NULL; 2536 bifp = sc->sc_ifp; 2537 2538 if ((bifp->if_flags & IFF_RUNNING) == 0) 2539 goto out; 2540 2541 /* 2542 * Implement support for bridge monitoring. If this flag has been 2543 * set on this interface, discard the packet once we push it through 2544 * the bpf(4) machinery, but before we do, increment various counters 2545 * associated with this bridge. 2546 */ 2547 if (bifp->if_flags & IFF_MONITOR) { 2548 /* Change input interface to this bridge */ 2549 m->m_pkthdr.rcvif = bifp; 2550 2551 BPF_MTAP(bifp, m); 2552 2553 /* Update bridge's ifnet statistics */ 2554 bifp->if_ipackets++; 2555 bifp->if_ibytes += m->m_pkthdr.len; 2556 if (m->m_flags & (M_MCAST | M_BCAST)) 2557 bifp->if_imcasts++; 2558 2559 m_freem(m); 2560 m = NULL; 2561 goto out; 2562 } 2563 2564 /* 2565 * Handle the ether_header 2566 * 2567 * In all cases if the packet is destined for us via our MAC 2568 * we must clear BRIDGE_MBUF_TAGGED to ensure that we don't 2569 * repeat the source MAC out the same interface. 2570 * 2571 * This first test against our bridge MAC is the fast-path. 2572 * 2573 * NOTE! The bridge interface can serve as an endpoint for 2574 * communication but normally there are no IPs associated 2575 * with it so you cannot route through it. Instead what 2576 * you do is point your default route *THROUGH* the bridge 2577 * to the actual default router for one of the bridged spaces. 2578 * 2579 * Another possibility is to put all your IP specifications 2580 * on the bridge instead of on the individual interfaces. If 2581 * you do this it should be possible to use the bridge as an 2582 * end point and route (rather than switch) through it using 2583 * the default route or ipfw forwarding rules. 2584 */ 2585 2586 /* 2587 * Acquire header 2588 */ 2589 if (m->m_len < ETHER_HDR_LEN) { 2590 m = m_pullup(m, ETHER_HDR_LEN); 2591 if (m == NULL) 2592 goto out; 2593 } 2594 eh = mtod(m, struct ether_header *); 2595 m->m_pkthdr.fw_flags |= BRIDGE_MBUF_TAGGED; 2596 bcopy(eh, &m->m_pkthdr.br.ether, sizeof(*eh)); 2597 2598 if ((bridge_debug & 1) && 2599 (ntohs(eh->ether_type) == ETHERTYPE_ARP || 2600 ntohs(eh->ether_type) == ETHERTYPE_REVARP)) { 2601 kprintf("%02x:%02x:%02x:%02x:%02x:%02x " 2602 "%02x:%02x:%02x:%02x:%02x:%02x type %04x " 2603 "lla %02x:%02x:%02x:%02x:%02x:%02x\n", 2604 eh->ether_dhost[0], 2605 eh->ether_dhost[1], 2606 eh->ether_dhost[2], 2607 eh->ether_dhost[3], 2608 eh->ether_dhost[4], 2609 eh->ether_dhost[5], 2610 eh->ether_shost[0], 2611 eh->ether_shost[1], 2612 eh->ether_shost[2], 2613 eh->ether_shost[3], 2614 eh->ether_shost[4], 2615 eh->ether_shost[5], 2616 eh->ether_type, 2617 ((u_char *)IF_LLADDR(bifp))[0], 2618 ((u_char *)IF_LLADDR(bifp))[1], 2619 ((u_char *)IF_LLADDR(bifp))[2], 2620 ((u_char *)IF_LLADDR(bifp))[3], 2621 ((u_char *)IF_LLADDR(bifp))[4], 2622 ((u_char *)IF_LLADDR(bifp))[5] 2623 ); 2624 } 2625 2626 if (memcmp(eh->ether_dhost, IF_LLADDR(bifp), ETHER_ADDR_LEN) == 0) { 2627 /* 2628 * If the packet is for us, set the packets source as the 2629 * bridge, and return the packet back to ifnet.if_input for 2630 * local processing. 2631 */ 2632 m->m_pkthdr.fw_flags &= ~BRIDGE_MBUF_TAGGED; 2633 KASSERT(bifp->if_bridge == NULL, 2634 ("loop created in bridge_input")); 2635 if (pfil_member != 0) { 2636 if (inet_pfil_hook.ph_hashooks > 0 2637 #ifdef INET6 2638 || inet6_pfil_hook.ph_hashooks > 0 2639 #endif 2640 ) { 2641 if (bridge_pfil(&m, NULL, ifp, PFIL_IN) != 0) 2642 goto out; 2643 if (m == NULL) 2644 goto out; 2645 } 2646 } 2647 new_ifp = bifp; 2648 goto out; 2649 } 2650 2651 /* 2652 * Tap all packets arriving on the bridge, no matter if 2653 * they are local destinations or not. In is in. 2654 */ 2655 BPF_MTAP(bifp, m); 2656 2657 bif = bridge_lookup_member_if(sc, ifp); 2658 if (bif == NULL) 2659 goto out; 2660 2661 if (sc->sc_span) 2662 bridge_span(sc, m); 2663 2664 if (m->m_flags & (M_BCAST | M_MCAST)) { 2665 /* 2666 * Tap off 802.1D packets; they do not get forwarded. 2667 */ 2668 if (memcmp(eh->ether_dhost, bstp_etheraddr, 2669 ETHER_ADDR_LEN) == 0) { 2670 ifnet_serialize_all(bifp); 2671 bstp_input(sc, bif, m); 2672 ifnet_deserialize_all(bifp); 2673 2674 /* m is freed by bstp_input */ 2675 m = NULL; 2676 goto out; 2677 } 2678 2679 /* 2680 * Other than 802.11d packets, ignore packets if the 2681 * interface is not in a good state. 2682 * 2683 * NOTE: Broadcast/mcast packets received on a blocking or 2684 * learning interface are allowed for local processing. 2685 * 2686 * The sending side of a blocked port will stop 2687 * transmitting when a better alternative is found. 2688 * However, later on we will disallow the forwarding 2689 * of bcast/mcsat packets over a blocking interface. 2690 */ 2691 if (bif->bif_flags & IFBIF_STP) { 2692 switch (bif->bif_state) { 2693 case BSTP_IFSTATE_L1BLOCKING: 2694 case BSTP_IFSTATE_LISTENING: 2695 case BSTP_IFSTATE_DISABLED: 2696 goto out; 2697 default: 2698 /* blocking, learning, bonded, forwarding */ 2699 break; 2700 } 2701 } 2702 2703 /* 2704 * Make a deep copy of the packet and enqueue the copy 2705 * for bridge processing; return the original packet for 2706 * local processing. 2707 */ 2708 mc = m_dup(m, MB_DONTWAIT); 2709 if (mc == NULL) 2710 goto out; 2711 2712 /* 2713 * It's just too dangerous to allow bcast/mcast over a 2714 * blocked interface, eventually the network will sort 2715 * itself out and a better path will be found. 2716 */ 2717 if ((bif->bif_flags & IFBIF_STP) == 0 || 2718 bif->bif_state != BSTP_IFSTATE_BLOCKING) { 2719 bridge_forward(sc, mc); 2720 } 2721 2722 /* 2723 * Reinject the mbuf as arriving on the bridge so we have a 2724 * chance at claiming multicast packets. We can not loop back 2725 * here from ether_input as a bridge is never a member of a 2726 * bridge. 2727 */ 2728 KASSERT(bifp->if_bridge == NULL, 2729 ("loop created in bridge_input")); 2730 mc2 = m_dup(m, MB_DONTWAIT); 2731 #ifdef notyet 2732 if (mc2 != NULL) { 2733 /* Keep the layer3 header aligned */ 2734 int i = min(mc2->m_pkthdr.len, max_protohdr); 2735 mc2 = m_copyup(mc2, i, ETHER_ALIGN); 2736 } 2737 #endif 2738 if (mc2 != NULL) { 2739 /* 2740 * Don't tap to bpf(4) again; we have already done 2741 * the tapping. 2742 * 2743 * Leave m_pkthdr.rcvif alone, so ARP replies are 2744 * processed as coming in on the correct interface. 2745 * 2746 * Clear the bridge flag for local processing in 2747 * case the packet gets routed. 2748 */ 2749 mc2->m_pkthdr.fw_flags &= ~BRIDGE_MBUF_TAGGED; 2750 ether_reinput_oncpu(bifp, mc2, 0); 2751 } 2752 2753 /* Return the original packet for local processing. */ 2754 goto out; 2755 } 2756 2757 /* 2758 * Input of a unicast packet. We have to allow unicast packets 2759 * input from links in the BLOCKING state as this might be an 2760 * interface of last resort. 2761 * 2762 * NOTE: We explicitly ignore normal packets received on a link 2763 * in the BLOCKING state. The point of being in that state 2764 * is to avoid getting duplicate packets. 2765 * 2766 * HOWEVER, if LINK2 is set the normal spanning tree code 2767 * will mark an interface BLOCKING to avoid multi-cast/broadcast 2768 * loops. Unicast packets CAN still loop if we allow the 2769 * case (hence we only do it in LINK2), but it isn't quite as 2770 * bad as a broadcast packet looping. 2771 */ 2772 if (bif->bif_flags & IFBIF_STP) { 2773 switch (bif->bif_state) { 2774 case BSTP_IFSTATE_L1BLOCKING: 2775 case BSTP_IFSTATE_LISTENING: 2776 case BSTP_IFSTATE_DISABLED: 2777 goto out; 2778 default: 2779 /* blocking, bonded, forwarding, learning */ 2780 break; 2781 } 2782 } 2783 2784 /* 2785 * Unicast. Make sure it's not for us. 2786 * 2787 * This loop is MPSAFE; the only blocking operation (bridge_rtupdate) 2788 * is followed by breaking out of the loop. 2789 */ 2790 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) { 2791 if (bif->bif_ifp->if_type != IFT_ETHER) 2792 continue; 2793 2794 /* 2795 * It is destined for an interface linked to the bridge. 2796 * We want the bridge itself to take care of link level 2797 * forwarding to member interfaces so reinput on the bridge. 2798 * i.e. if you ping an IP on a target interface associated 2799 * with the bridge, the arp is-at response should indicate 2800 * the bridge MAC. 2801 * 2802 * Only update our addr list when learning if the port 2803 * is not in a blocking state. If it is we still allow 2804 * the packet but we do not try to learn from it. 2805 */ 2806 if (memcmp(IF_LLADDR(bif->bif_ifp), eh->ether_dhost, 2807 ETHER_ADDR_LEN) == 0) { 2808 if (bif->bif_ifp != ifp) { 2809 /* XXX loop prevention */ 2810 m->m_flags |= M_ETHER_BRIDGED; 2811 } 2812 if ((bif->bif_flags & IFBIF_LEARNING) && 2813 bif->bif_state != BSTP_IFSTATE_BLOCKING) { 2814 bridge_rtupdate(sc, eh->ether_shost, 2815 ifp, IFBAF_DYNAMIC); 2816 } 2817 new_ifp = bifp; /* not bif->bif_ifp */ 2818 m->m_pkthdr.fw_flags &= ~BRIDGE_MBUF_TAGGED; 2819 goto out; 2820 } 2821 2822 /* 2823 * Ignore received packets that were sent by us. 2824 */ 2825 if (memcmp(IF_LLADDR(bif->bif_ifp), eh->ether_shost, 2826 ETHER_ADDR_LEN) == 0) { 2827 m_freem(m); 2828 m = NULL; 2829 goto out; 2830 } 2831 } 2832 2833 /* 2834 * It isn't for us. 2835 * 2836 * Perform the bridge forwarding function, but disallow bridging 2837 * to interfaces in the blocking state if the packet came in on 2838 * an interface in the blocking state. 2839 */ 2840 bridge_forward(sc, m); 2841 m = NULL; 2842 2843 /* 2844 * ether_reinput_oncpu() will reprocess rcvif as 2845 * coming from new_ifp (since we do not specify 2846 * REINPUT_KEEPRCVIF). 2847 */ 2848 out: 2849 if (new_ifp != NULL) { 2850 /* 2851 * Clear the bridge flag for local processing in 2852 * case the packet gets routed. 2853 */ 2854 ether_reinput_oncpu(new_ifp, m, REINPUT_RUNBPF); 2855 m = NULL; 2856 } 2857 return (m); 2858 } 2859 2860 /* 2861 * bridge_start_bcast: 2862 * 2863 * Broadcast the packet sent from bridge to all member 2864 * interfaces. 2865 * This is a simplified version of bridge_broadcast(), however, 2866 * this function expects caller to hold bridge's serializer. 2867 */ 2868 static void 2869 bridge_start_bcast(struct bridge_softc *sc, struct mbuf *m) 2870 { 2871 struct bridge_iflist *bif; 2872 struct mbuf *mc; 2873 struct ifnet *dst_if, *alt_if, *bifp; 2874 int used = 0; 2875 int found = 0; 2876 int alt_priority; 2877 2878 mbuftrackid(m, 68); 2879 bifp = sc->sc_ifp; 2880 ASSERT_IFNET_SERIALIZED_ALL(bifp); 2881 2882 /* 2883 * Following loop is MPSAFE; nothing is blocking 2884 * in the loop body. 2885 * 2886 * NOTE: We transmit through an member in the BLOCKING state only 2887 * as a last resort. 2888 */ 2889 alt_if = NULL; 2890 alt_priority = 0; 2891 2892 TAILQ_FOREACH(bif, &sc->sc_iflists[mycpuid], bif_next) { 2893 dst_if = bif->bif_ifp; 2894 2895 if (bif->bif_flags & IFBIF_STP) { 2896 switch (bif->bif_state) { 2897 case BSTP_IFSTATE_BLOCKING: 2898 if (bif->bif_priority > alt_priority) { 2899 alt_priority = bif->bif_priority; 2900 alt_if = bif->bif_ifp; 2901 } 2902 /* fall through */ 2903 case BSTP_IFSTATE_L1BLOCKING: 2904 case BSTP_IFSTATE_DISABLED: 2905 continue; 2906 default: 2907 /* listening, learning, bonded, forwarding */ 2908 break; 2909 } 2910 } 2911 2912 if ((bif->bif_flags & IFBIF_DISCOVER) == 0 && 2913 (m->m_flags & (M_BCAST|M_MCAST)) == 0) 2914 continue; 2915 2916 if ((dst_if->if_flags & IFF_RUNNING) == 0) 2917 continue; 2918 2919 if (TAILQ_NEXT(bif, bif_next) == NULL) { 2920 mc = m; 2921 used = 1; 2922 } else { 2923 mc = m_copypacket(m, MB_DONTWAIT); 2924 if (mc == NULL) { 2925 bifp->if_oerrors++; 2926 continue; 2927 } 2928 } 2929 found = 1; 2930 bridge_enqueue(dst_if, mc); 2931 } 2932 2933 if (found == 0 && alt_if) { 2934 KKASSERT(used == 0); 2935 mc = m; 2936 used = 1; 2937 bridge_enqueue(alt_if, mc); 2938 } 2939 2940 if (used == 0) 2941 m_freem(m); 2942 } 2943 2944 /* 2945 * bridge_broadcast: 2946 * 2947 * Send a frame to all interfaces that are members of 2948 * the bridge, except for the one on which the packet 2949 * arrived. 2950 */ 2951 static void 2952 bridge_broadcast(struct bridge_softc *sc, struct ifnet *src_if, 2953 struct mbuf *m) 2954 { 2955 struct bridge_iflist *bif, *nbif; 2956 struct ether_header *eh; 2957 struct mbuf *mc; 2958 struct ifnet *dst_if, *alt_if, *bifp; 2959 int used; 2960 int found; 2961 int alt_priority; 2962 int from_us; 2963 2964 mbuftrackid(m, 69); 2965 bifp = sc->sc_ifp; 2966 ASSERT_IFNET_NOT_SERIALIZED_ALL(bifp); 2967 2968 eh = mtod(m, struct ether_header *); 2969 from_us = bridge_from_us(sc, eh); 2970 2971 if (inet_pfil_hook.ph_hashooks > 0 2972 #ifdef INET6 2973 || inet6_pfil_hook.ph_hashooks > 0 2974 #endif 2975 ) { 2976 if (bridge_pfil(&m, bifp, src_if, PFIL_IN) != 0) 2977 return; 2978 if (m == NULL) 2979 return; 2980 2981 /* Filter on the bridge interface before broadcasting */ 2982 if (bridge_pfil(&m, bifp, NULL, PFIL_OUT) != 0) 2983 return; 2984 if (m == NULL) 2985 return; 2986 } 2987 2988 alt_if = NULL; 2989 alt_priority = 0; 2990 found = 0; 2991 used = 0; 2992 2993 TAILQ_FOREACH_MUTABLE(bif, &sc->sc_iflists[mycpuid], bif_next, nbif) { 2994 dst_if = bif->bif_ifp; 2995 2996 if ((dst_if->if_flags & IFF_RUNNING) == 0) 2997 continue; 2998 2999 /* 3000 * Don't bounce the packet out the same interface it came 3001 * in on. We have to test MAC addresses because a packet 3002 * can come in a bonded interface and we don't want it to 3003 * be echod out the forwarding interface for the same bonding 3004 * set. 3005 */ 3006 if (src_if && memcmp(IF_LLADDR(src_if), IF_LLADDR(dst_if), 3007 ETHER_ADDR_LEN) == 0) { 3008 continue; 3009 } 3010 3011 /* 3012 * Generally speaking we only broadcast through forwarding 3013 * interfaces. If no interfaces are available we select 3014 * a BONDED, BLOCKING, or LEARNING interface to forward 3015 * through. 3016 */ 3017 if (bif->bif_flags & IFBIF_STP) { 3018 switch (bif->bif_state) { 3019 case BSTP_IFSTATE_BONDED: 3020 if (bif->bif_priority + 512 > alt_priority) { 3021 alt_priority = bif->bif_priority + 512; 3022 alt_if = bif->bif_ifp; 3023 } 3024 continue; 3025 case BSTP_IFSTATE_BLOCKING: 3026 if (bif->bif_priority + 256 > alt_priority) { 3027 alt_priority = bif->bif_priority + 256; 3028 alt_if = bif->bif_ifp; 3029 } 3030 continue; 3031 case BSTP_IFSTATE_LEARNING: 3032 if (bif->bif_priority > alt_priority) { 3033 alt_priority = bif->bif_priority; 3034 alt_if = bif->bif_ifp; 3035 } 3036 continue; 3037 case BSTP_IFSTATE_L1BLOCKING: 3038 case BSTP_IFSTATE_DISABLED: 3039 case BSTP_IFSTATE_LISTENING: 3040 continue; 3041 default: 3042 /* forwarding */ 3043 break; 3044 } 3045 } 3046 3047 if ((bif->bif_flags & IFBIF_DISCOVER) == 0 && 3048 (m->m_flags & (M_BCAST|M_MCAST)) == 0) { 3049 continue; 3050 } 3051 3052 if (TAILQ_NEXT(bif, bif_next) == NULL) { 3053 mc = m; 3054 used = 1; 3055 } else { 3056 mc = m_copypacket(m, MB_DONTWAIT); 3057 if (mc == NULL) { 3058 sc->sc_ifp->if_oerrors++; 3059 continue; 3060 } 3061 } 3062 found = 1; 3063 3064 /* 3065 * Filter on the output interface. Pass a NULL bridge 3066 * interface pointer so we do not redundantly filter on 3067 * the bridge for each interface we broadcast on. 3068 */ 3069 if (inet_pfil_hook.ph_hashooks > 0 3070 #ifdef INET6 3071 || inet6_pfil_hook.ph_hashooks > 0 3072 #endif 3073 ) { 3074 if (bridge_pfil(&mc, NULL, dst_if, PFIL_OUT) != 0) 3075 continue; 3076 if (mc == NULL) 3077 continue; 3078 } 3079 bridge_handoff(sc, dst_if, mc, from_us); 3080 3081 if (nbif != NULL && !nbif->bif_onlist) { 3082 KKASSERT(bif->bif_onlist); 3083 nbif = TAILQ_NEXT(bif, bif_next); 3084 } 3085 } 3086 3087 if (found == 0 && alt_if) { 3088 KKASSERT(used == 0); 3089 mc = m; 3090 used = 1; 3091 bridge_enqueue(alt_if, mc); 3092 } 3093 3094 if (used == 0) 3095 m_freem(m); 3096 } 3097 3098 /* 3099 * bridge_span: 3100 * 3101 * Duplicate a packet out one or more interfaces that are in span mode, 3102 * the original mbuf is unmodified. 3103 */ 3104 static void 3105 bridge_span(struct bridge_softc *sc, struct mbuf *m) 3106 { 3107 struct bridge_iflist *bif; 3108 struct ifnet *dst_if, *bifp; 3109 struct mbuf *mc; 3110 3111 mbuftrackid(m, 70); 3112 bifp = sc->sc_ifp; 3113 ifnet_serialize_all(bifp); 3114 3115 TAILQ_FOREACH(bif, &sc->sc_spanlist, bif_next) { 3116 dst_if = bif->bif_ifp; 3117 3118 if ((dst_if->if_flags & IFF_RUNNING) == 0) 3119 continue; 3120 3121 mc = m_copypacket(m, MB_DONTWAIT); 3122 if (mc == NULL) { 3123 sc->sc_ifp->if_oerrors++; 3124 continue; 3125 } 3126 bridge_enqueue(dst_if, mc); 3127 } 3128 3129 ifnet_deserialize_all(bifp); 3130 } 3131 3132 static void 3133 bridge_rtmsg_sync_handler(netmsg_t msg) 3134 { 3135 ifnet_forwardmsg(&msg->lmsg, mycpuid + 1); 3136 } 3137 3138 static void 3139 bridge_rtmsg_sync(struct bridge_softc *sc) 3140 { 3141 struct netmsg_base msg; 3142 3143 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 3144 3145 netmsg_init(&msg, NULL, &curthread->td_msgport, 3146 0, bridge_rtmsg_sync_handler); 3147 ifnet_domsg(&msg.lmsg, 0); 3148 } 3149 3150 static __inline void 3151 bridge_rtinfo_update(struct bridge_rtinfo *bri, struct ifnet *dst_if, 3152 int setflags, uint8_t flags, uint32_t timeo) 3153 { 3154 if ((bri->bri_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 3155 bri->bri_ifp != dst_if) 3156 bri->bri_ifp = dst_if; 3157 if ((flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 3158 bri->bri_expire != time_second + timeo) 3159 bri->bri_expire = time_second + timeo; 3160 if (setflags) 3161 bri->bri_flags = flags; 3162 } 3163 3164 static int 3165 bridge_rtinstall_oncpu(struct bridge_softc *sc, const uint8_t *dst, 3166 struct ifnet *dst_if, int setflags, uint8_t flags, 3167 struct bridge_rtinfo **bri0) 3168 { 3169 struct bridge_rtnode *brt; 3170 struct bridge_rtinfo *bri; 3171 3172 if (mycpuid == 0) { 3173 brt = bridge_rtnode_lookup(sc, dst); 3174 if (brt != NULL) { 3175 /* 3176 * rtnode for 'dst' already exists. We inform the 3177 * caller about this by leaving bri0 as NULL. The 3178 * caller will terminate the intallation upon getting 3179 * NULL bri0. However, we still need to update the 3180 * rtinfo. 3181 */ 3182 KKASSERT(*bri0 == NULL); 3183 3184 /* Update rtinfo */ 3185 bridge_rtinfo_update(brt->brt_info, dst_if, setflags, 3186 flags, sc->sc_brttimeout); 3187 return 0; 3188 } 3189 3190 /* 3191 * We only need to check brtcnt on CPU0, since if limit 3192 * is to be exceeded, ENOSPC is returned. Caller knows 3193 * this and will terminate the installation. 3194 */ 3195 if (sc->sc_brtcnt >= sc->sc_brtmax) 3196 return ENOSPC; 3197 3198 KKASSERT(*bri0 == NULL); 3199 bri = kmalloc(sizeof(struct bridge_rtinfo), M_DEVBUF, 3200 M_WAITOK | M_ZERO); 3201 *bri0 = bri; 3202 3203 /* Setup rtinfo */ 3204 bri->bri_flags = IFBAF_DYNAMIC; 3205 bridge_rtinfo_update(bri, dst_if, setflags, flags, 3206 sc->sc_brttimeout); 3207 } else { 3208 bri = *bri0; 3209 KKASSERT(bri != NULL); 3210 } 3211 3212 brt = kmalloc(sizeof(struct bridge_rtnode), M_DEVBUF, 3213 M_WAITOK | M_ZERO); 3214 memcpy(brt->brt_addr, dst, ETHER_ADDR_LEN); 3215 brt->brt_info = bri; 3216 3217 bridge_rtnode_insert(sc, brt); 3218 return 0; 3219 } 3220 3221 static void 3222 bridge_rtinstall_handler(netmsg_t msg) 3223 { 3224 struct netmsg_brsaddr *brmsg = (struct netmsg_brsaddr *)msg; 3225 int error; 3226 3227 error = bridge_rtinstall_oncpu(brmsg->br_softc, 3228 brmsg->br_dst, brmsg->br_dst_if, 3229 brmsg->br_setflags, brmsg->br_flags, 3230 &brmsg->br_rtinfo); 3231 if (error) { 3232 KKASSERT(mycpuid == 0 && brmsg->br_rtinfo == NULL); 3233 lwkt_replymsg(&brmsg->base.lmsg, error); 3234 return; 3235 } else if (brmsg->br_rtinfo == NULL) { 3236 /* rtnode already exists for 'dst' */ 3237 KKASSERT(mycpuid == 0); 3238 lwkt_replymsg(&brmsg->base.lmsg, 0); 3239 return; 3240 } 3241 ifnet_forwardmsg(&brmsg->base.lmsg, mycpuid + 1); 3242 } 3243 3244 /* 3245 * bridge_rtupdate: 3246 * 3247 * Add/Update a bridge routing entry. 3248 */ 3249 static int 3250 bridge_rtupdate(struct bridge_softc *sc, const uint8_t *dst, 3251 struct ifnet *dst_if, uint8_t flags) 3252 { 3253 struct bridge_rtnode *brt; 3254 3255 /* 3256 * A route for this destination might already exist. If so, 3257 * update it, otherwise create a new one. 3258 */ 3259 if ((brt = bridge_rtnode_lookup(sc, dst)) == NULL) { 3260 struct netmsg_brsaddr *brmsg; 3261 3262 if (sc->sc_brtcnt >= sc->sc_brtmax) 3263 return ENOSPC; 3264 3265 brmsg = kmalloc(sizeof(*brmsg), M_LWKTMSG, M_WAITOK | M_NULLOK); 3266 if (brmsg == NULL) 3267 return ENOMEM; 3268 3269 netmsg_init(&brmsg->base, NULL, &netisr_afree_rport, 3270 0, bridge_rtinstall_handler); 3271 memcpy(brmsg->br_dst, dst, ETHER_ADDR_LEN); 3272 brmsg->br_dst_if = dst_if; 3273 brmsg->br_flags = flags; 3274 brmsg->br_setflags = 0; 3275 brmsg->br_softc = sc; 3276 brmsg->br_rtinfo = NULL; 3277 3278 ifnet_sendmsg(&brmsg->base.lmsg, 0); 3279 return 0; 3280 } 3281 bridge_rtinfo_update(brt->brt_info, dst_if, 0, flags, 3282 sc->sc_brttimeout); 3283 return 0; 3284 } 3285 3286 static int 3287 bridge_rtsaddr(struct bridge_softc *sc, const uint8_t *dst, 3288 struct ifnet *dst_if, uint8_t flags) 3289 { 3290 struct netmsg_brsaddr brmsg; 3291 3292 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 3293 3294 netmsg_init(&brmsg.base, NULL, &curthread->td_msgport, 3295 0, bridge_rtinstall_handler); 3296 memcpy(brmsg.br_dst, dst, ETHER_ADDR_LEN); 3297 brmsg.br_dst_if = dst_if; 3298 brmsg.br_flags = flags; 3299 brmsg.br_setflags = 1; 3300 brmsg.br_softc = sc; 3301 brmsg.br_rtinfo = NULL; 3302 3303 return ifnet_domsg(&brmsg.base.lmsg, 0); 3304 } 3305 3306 /* 3307 * bridge_rtlookup: 3308 * 3309 * Lookup the destination interface for an address. 3310 */ 3311 static struct ifnet * 3312 bridge_rtlookup(struct bridge_softc *sc, const uint8_t *addr) 3313 { 3314 struct bridge_rtnode *brt; 3315 3316 if ((brt = bridge_rtnode_lookup(sc, addr)) == NULL) 3317 return NULL; 3318 return brt->brt_info->bri_ifp; 3319 } 3320 3321 static void 3322 bridge_rtreap_handler(netmsg_t msg) 3323 { 3324 struct bridge_softc *sc = msg->lmsg.u.ms_resultp; 3325 struct bridge_rtnode *brt, *nbrt; 3326 3327 LIST_FOREACH_MUTABLE(brt, &sc->sc_rtlists[mycpuid], brt_list, nbrt) { 3328 if (brt->brt_info->bri_dead) 3329 bridge_rtnode_destroy(sc, brt); 3330 } 3331 ifnet_forwardmsg(&msg->lmsg, mycpuid + 1); 3332 } 3333 3334 static void 3335 bridge_rtreap(struct bridge_softc *sc) 3336 { 3337 struct netmsg_base msg; 3338 3339 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 3340 3341 netmsg_init(&msg, NULL, &curthread->td_msgport, 3342 0, bridge_rtreap_handler); 3343 msg.lmsg.u.ms_resultp = sc; 3344 3345 ifnet_domsg(&msg.lmsg, 0); 3346 } 3347 3348 static void 3349 bridge_rtreap_async(struct bridge_softc *sc) 3350 { 3351 struct netmsg_base *msg; 3352 3353 msg = kmalloc(sizeof(*msg), M_LWKTMSG, M_WAITOK); 3354 3355 netmsg_init(msg, NULL, &netisr_afree_rport, 3356 0, bridge_rtreap_handler); 3357 msg->lmsg.u.ms_resultp = sc; 3358 3359 ifnet_sendmsg(&msg->lmsg, 0); 3360 } 3361 3362 /* 3363 * bridge_rttrim: 3364 * 3365 * Trim the routine table so that we have a number 3366 * of routing entries less than or equal to the 3367 * maximum number. 3368 */ 3369 static void 3370 bridge_rttrim(struct bridge_softc *sc) 3371 { 3372 struct bridge_rtnode *brt; 3373 int dead; 3374 3375 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 3376 3377 /* Make sure we actually need to do this. */ 3378 if (sc->sc_brtcnt <= sc->sc_brtmax) 3379 return; 3380 3381 /* 3382 * Find out how many rtnodes are dead 3383 */ 3384 dead = bridge_rtage_finddead(sc); 3385 KKASSERT(dead <= sc->sc_brtcnt); 3386 3387 if (sc->sc_brtcnt - dead <= sc->sc_brtmax) { 3388 /* Enough dead rtnodes are found */ 3389 bridge_rtreap(sc); 3390 return; 3391 } 3392 3393 /* 3394 * Kill some dynamic rtnodes to meet the brtmax 3395 */ 3396 LIST_FOREACH(brt, &sc->sc_rtlists[mycpuid], brt_list) { 3397 struct bridge_rtinfo *bri = brt->brt_info; 3398 3399 if (bri->bri_dead) { 3400 /* 3401 * We have counted this rtnode in 3402 * bridge_rtage_finddead() 3403 */ 3404 continue; 3405 } 3406 3407 if ((bri->bri_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) { 3408 bri->bri_dead = 1; 3409 ++dead; 3410 KKASSERT(dead <= sc->sc_brtcnt); 3411 3412 if (sc->sc_brtcnt - dead <= sc->sc_brtmax) { 3413 /* Enough rtnodes are collected */ 3414 break; 3415 } 3416 } 3417 } 3418 if (dead) 3419 bridge_rtreap(sc); 3420 } 3421 3422 /* 3423 * bridge_timer: 3424 * 3425 * Aging timer for the bridge. 3426 */ 3427 static void 3428 bridge_timer(void *arg) 3429 { 3430 struct bridge_softc *sc = arg; 3431 struct netmsg_base *msg; 3432 3433 KKASSERT(mycpuid == BRIDGE_CFGCPU); 3434 3435 crit_enter(); 3436 3437 if (callout_pending(&sc->sc_brcallout) || 3438 !callout_active(&sc->sc_brcallout)) { 3439 crit_exit(); 3440 return; 3441 } 3442 callout_deactivate(&sc->sc_brcallout); 3443 3444 msg = &sc->sc_brtimemsg; 3445 KKASSERT(msg->lmsg.ms_flags & MSGF_DONE); 3446 lwkt_sendmsg(BRIDGE_CFGPORT, &msg->lmsg); 3447 3448 crit_exit(); 3449 } 3450 3451 static void 3452 bridge_timer_handler(netmsg_t msg) 3453 { 3454 struct bridge_softc *sc = msg->lmsg.u.ms_resultp; 3455 3456 KKASSERT(&curthread->td_msgport == BRIDGE_CFGPORT); 3457 3458 crit_enter(); 3459 /* Reply ASAP */ 3460 lwkt_replymsg(&msg->lmsg, 0); 3461 crit_exit(); 3462 3463 bridge_rtage(sc); 3464 if (sc->sc_ifp->if_flags & IFF_RUNNING) { 3465 callout_reset(&sc->sc_brcallout, 3466 bridge_rtable_prune_period * hz, bridge_timer, sc); 3467 } 3468 } 3469 3470 static int 3471 bridge_rtage_finddead(struct bridge_softc *sc) 3472 { 3473 struct bridge_rtnode *brt; 3474 int dead = 0; 3475 3476 LIST_FOREACH(brt, &sc->sc_rtlists[mycpuid], brt_list) { 3477 struct bridge_rtinfo *bri = brt->brt_info; 3478 3479 if ((bri->bri_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 3480 time_second >= bri->bri_expire) { 3481 bri->bri_dead = 1; 3482 ++dead; 3483 KKASSERT(dead <= sc->sc_brtcnt); 3484 } 3485 } 3486 return dead; 3487 } 3488 3489 /* 3490 * bridge_rtage: 3491 * 3492 * Perform an aging cycle. 3493 */ 3494 static void 3495 bridge_rtage(struct bridge_softc *sc) 3496 { 3497 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 3498 3499 if (bridge_rtage_finddead(sc)) 3500 bridge_rtreap(sc); 3501 } 3502 3503 /* 3504 * bridge_rtflush: 3505 * 3506 * Remove all dynamic addresses from the bridge. 3507 */ 3508 static void 3509 bridge_rtflush(struct bridge_softc *sc, int bf) 3510 { 3511 struct bridge_rtnode *brt; 3512 int reap; 3513 3514 reap = 0; 3515 LIST_FOREACH(brt, &sc->sc_rtlists[mycpuid], brt_list) { 3516 struct bridge_rtinfo *bri = brt->brt_info; 3517 3518 if ((bf & IFBF_FLUSHALL) || 3519 (bri->bri_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) { 3520 bri->bri_dead = 1; 3521 reap = 1; 3522 } 3523 } 3524 if (reap) { 3525 if (bf & IFBF_FLUSHSYNC) 3526 bridge_rtreap(sc); 3527 else 3528 bridge_rtreap_async(sc); 3529 } 3530 } 3531 3532 /* 3533 * bridge_rtdaddr: 3534 * 3535 * Remove an address from the table. 3536 */ 3537 static int 3538 bridge_rtdaddr(struct bridge_softc *sc, const uint8_t *addr) 3539 { 3540 struct bridge_rtnode *brt; 3541 3542 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 3543 3544 if ((brt = bridge_rtnode_lookup(sc, addr)) == NULL) 3545 return (ENOENT); 3546 3547 /* TODO: add a cheaper delete operation */ 3548 brt->brt_info->bri_dead = 1; 3549 bridge_rtreap(sc); 3550 return (0); 3551 } 3552 3553 /* 3554 * bridge_rtdelete: 3555 * 3556 * Delete routes to a speicifc member interface. 3557 */ 3558 void 3559 bridge_rtdelete(struct bridge_softc *sc, struct ifnet *ifp, int bf) 3560 { 3561 struct bridge_rtnode *brt; 3562 int reap; 3563 3564 reap = 0; 3565 LIST_FOREACH(brt, &sc->sc_rtlists[mycpuid], brt_list) { 3566 struct bridge_rtinfo *bri = brt->brt_info; 3567 3568 if (bri->bri_ifp == ifp && 3569 ((bf & IFBF_FLUSHALL) || 3570 (bri->bri_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC)) { 3571 bri->bri_dead = 1; 3572 reap = 1; 3573 } 3574 } 3575 if (reap) { 3576 if (bf & IFBF_FLUSHSYNC) 3577 bridge_rtreap(sc); 3578 else 3579 bridge_rtreap_async(sc); 3580 } 3581 } 3582 3583 /* 3584 * bridge_rtable_init: 3585 * 3586 * Initialize the route table for this bridge. 3587 */ 3588 static void 3589 bridge_rtable_init(struct bridge_softc *sc) 3590 { 3591 int cpu; 3592 3593 /* 3594 * Initialize per-cpu hash tables 3595 */ 3596 sc->sc_rthashs = kmalloc(sizeof(*sc->sc_rthashs) * ncpus, 3597 M_DEVBUF, M_WAITOK); 3598 for (cpu = 0; cpu < ncpus; ++cpu) { 3599 int i; 3600 3601 sc->sc_rthashs[cpu] = 3602 kmalloc(sizeof(struct bridge_rtnode_head) * BRIDGE_RTHASH_SIZE, 3603 M_DEVBUF, M_WAITOK); 3604 3605 for (i = 0; i < BRIDGE_RTHASH_SIZE; i++) 3606 LIST_INIT(&sc->sc_rthashs[cpu][i]); 3607 } 3608 sc->sc_rthash_key = karc4random(); 3609 3610 /* 3611 * Initialize per-cpu lists 3612 */ 3613 sc->sc_rtlists = kmalloc(sizeof(struct bridge_rtnode_head) * ncpus, 3614 M_DEVBUF, M_WAITOK); 3615 for (cpu = 0; cpu < ncpus; ++cpu) 3616 LIST_INIT(&sc->sc_rtlists[cpu]); 3617 } 3618 3619 /* 3620 * bridge_rtable_fini: 3621 * 3622 * Deconstruct the route table for this bridge. 3623 */ 3624 static void 3625 bridge_rtable_fini(struct bridge_softc *sc) 3626 { 3627 int cpu; 3628 3629 /* 3630 * Free per-cpu hash tables 3631 */ 3632 for (cpu = 0; cpu < ncpus; ++cpu) 3633 kfree(sc->sc_rthashs[cpu], M_DEVBUF); 3634 kfree(sc->sc_rthashs, M_DEVBUF); 3635 3636 /* 3637 * Free per-cpu lists 3638 */ 3639 kfree(sc->sc_rtlists, M_DEVBUF); 3640 } 3641 3642 /* 3643 * The following hash function is adapted from "Hash Functions" by Bob Jenkins 3644 * ("Algorithm Alley", Dr. Dobbs Journal, September 1997). 3645 */ 3646 #define mix(a, b, c) \ 3647 do { \ 3648 a -= b; a -= c; a ^= (c >> 13); \ 3649 b -= c; b -= a; b ^= (a << 8); \ 3650 c -= a; c -= b; c ^= (b >> 13); \ 3651 a -= b; a -= c; a ^= (c >> 12); \ 3652 b -= c; b -= a; b ^= (a << 16); \ 3653 c -= a; c -= b; c ^= (b >> 5); \ 3654 a -= b; a -= c; a ^= (c >> 3); \ 3655 b -= c; b -= a; b ^= (a << 10); \ 3656 c -= a; c -= b; c ^= (b >> 15); \ 3657 } while (/*CONSTCOND*/0) 3658 3659 static __inline uint32_t 3660 bridge_rthash(struct bridge_softc *sc, const uint8_t *addr) 3661 { 3662 uint32_t a = 0x9e3779b9, b = 0x9e3779b9, c = sc->sc_rthash_key; 3663 3664 b += addr[5] << 8; 3665 b += addr[4]; 3666 a += addr[3] << 24; 3667 a += addr[2] << 16; 3668 a += addr[1] << 8; 3669 a += addr[0]; 3670 3671 mix(a, b, c); 3672 3673 return (c & BRIDGE_RTHASH_MASK); 3674 } 3675 3676 #undef mix 3677 3678 static int 3679 bridge_rtnode_addr_cmp(const uint8_t *a, const uint8_t *b) 3680 { 3681 int i, d; 3682 3683 for (i = 0, d = 0; i < ETHER_ADDR_LEN && d == 0; i++) { 3684 d = ((int)a[i]) - ((int)b[i]); 3685 } 3686 3687 return (d); 3688 } 3689 3690 /* 3691 * bridge_rtnode_lookup: 3692 * 3693 * Look up a bridge route node for the specified destination. 3694 */ 3695 static struct bridge_rtnode * 3696 bridge_rtnode_lookup(struct bridge_softc *sc, const uint8_t *addr) 3697 { 3698 struct bridge_rtnode *brt; 3699 uint32_t hash; 3700 int dir; 3701 3702 hash = bridge_rthash(sc, addr); 3703 LIST_FOREACH(brt, &sc->sc_rthashs[mycpuid][hash], brt_hash) { 3704 dir = bridge_rtnode_addr_cmp(addr, brt->brt_addr); 3705 if (dir == 0) 3706 return (brt); 3707 if (dir > 0) 3708 return (NULL); 3709 } 3710 3711 return (NULL); 3712 } 3713 3714 /* 3715 * bridge_rtnode_insert: 3716 * 3717 * Insert the specified bridge node into the route table. 3718 * Caller has to make sure that rtnode does not exist. 3719 */ 3720 static void 3721 bridge_rtnode_insert(struct bridge_softc *sc, struct bridge_rtnode *brt) 3722 { 3723 struct bridge_rtnode *lbrt; 3724 uint32_t hash; 3725 int dir; 3726 3727 hash = bridge_rthash(sc, brt->brt_addr); 3728 3729 lbrt = LIST_FIRST(&sc->sc_rthashs[mycpuid][hash]); 3730 if (lbrt == NULL) { 3731 LIST_INSERT_HEAD(&sc->sc_rthashs[mycpuid][hash], 3732 brt, brt_hash); 3733 goto out; 3734 } 3735 3736 do { 3737 dir = bridge_rtnode_addr_cmp(brt->brt_addr, lbrt->brt_addr); 3738 KASSERT(dir != 0, ("rtnode already exist")); 3739 3740 if (dir > 0) { 3741 LIST_INSERT_BEFORE(lbrt, brt, brt_hash); 3742 goto out; 3743 } 3744 if (LIST_NEXT(lbrt, brt_hash) == NULL) { 3745 LIST_INSERT_AFTER(lbrt, brt, brt_hash); 3746 goto out; 3747 } 3748 lbrt = LIST_NEXT(lbrt, brt_hash); 3749 } while (lbrt != NULL); 3750 3751 panic("no suitable position found for rtnode"); 3752 out: 3753 LIST_INSERT_HEAD(&sc->sc_rtlists[mycpuid], brt, brt_list); 3754 if (mycpuid == 0) { 3755 /* 3756 * Update the brtcnt. 3757 * We only need to do it once and we do it on CPU0. 3758 */ 3759 sc->sc_brtcnt++; 3760 } 3761 } 3762 3763 /* 3764 * bridge_rtnode_destroy: 3765 * 3766 * Destroy a bridge rtnode. 3767 */ 3768 static void 3769 bridge_rtnode_destroy(struct bridge_softc *sc, struct bridge_rtnode *brt) 3770 { 3771 LIST_REMOVE(brt, brt_hash); 3772 LIST_REMOVE(brt, brt_list); 3773 3774 if (mycpuid + 1 == ncpus) { 3775 /* Free rtinfo associated with rtnode on the last cpu */ 3776 kfree(brt->brt_info, M_DEVBUF); 3777 } 3778 kfree(brt, M_DEVBUF); 3779 3780 if (mycpuid == 0) { 3781 /* Update brtcnt only on CPU0 */ 3782 sc->sc_brtcnt--; 3783 } 3784 } 3785 3786 static __inline int 3787 bridge_post_pfil(struct mbuf *m) 3788 { 3789 if (m->m_pkthdr.fw_flags & IPFORWARD_MBUF_TAGGED) 3790 return EOPNOTSUPP; 3791 3792 /* Not yet */ 3793 if (m->m_pkthdr.fw_flags & DUMMYNET_MBUF_TAGGED) 3794 return EOPNOTSUPP; 3795 3796 return 0; 3797 } 3798 3799 /* 3800 * Send bridge packets through pfil if they are one of the types pfil can deal 3801 * with, or if they are ARP or REVARP. (pfil will pass ARP and REVARP without 3802 * question.) If *bifp or *ifp are NULL then packet filtering is skipped for 3803 * that interface. 3804 */ 3805 static int 3806 bridge_pfil(struct mbuf **mp, struct ifnet *bifp, struct ifnet *ifp, int dir) 3807 { 3808 int snap, error, i, hlen; 3809 struct ether_header *eh1, eh2; 3810 struct ip *ip; 3811 struct llc llc1; 3812 u_int16_t ether_type; 3813 3814 snap = 0; 3815 error = -1; /* Default error if not error == 0 */ 3816 3817 if (pfil_bridge == 0 && pfil_member == 0) 3818 return (0); /* filtering is disabled */ 3819 3820 i = min((*mp)->m_pkthdr.len, max_protohdr); 3821 if ((*mp)->m_len < i) { 3822 *mp = m_pullup(*mp, i); 3823 if (*mp == NULL) { 3824 kprintf("%s: m_pullup failed\n", __func__); 3825 return (-1); 3826 } 3827 } 3828 3829 eh1 = mtod(*mp, struct ether_header *); 3830 ether_type = ntohs(eh1->ether_type); 3831 3832 /* 3833 * Check for SNAP/LLC. 3834 */ 3835 if (ether_type < ETHERMTU) { 3836 struct llc *llc2 = (struct llc *)(eh1 + 1); 3837 3838 if ((*mp)->m_len >= ETHER_HDR_LEN + 8 && 3839 llc2->llc_dsap == LLC_SNAP_LSAP && 3840 llc2->llc_ssap == LLC_SNAP_LSAP && 3841 llc2->llc_control == LLC_UI) { 3842 ether_type = htons(llc2->llc_un.type_snap.ether_type); 3843 snap = 1; 3844 } 3845 } 3846 3847 /* 3848 * If we're trying to filter bridge traffic, don't look at anything 3849 * other than IP and ARP traffic. If the filter doesn't understand 3850 * IPv6, don't allow IPv6 through the bridge either. This is lame 3851 * since if we really wanted, say, an AppleTalk filter, we are hosed, 3852 * but of course we don't have an AppleTalk filter to begin with. 3853 * (Note that since pfil doesn't understand ARP it will pass *ALL* 3854 * ARP traffic.) 3855 */ 3856 switch (ether_type) { 3857 case ETHERTYPE_ARP: 3858 case ETHERTYPE_REVARP: 3859 return (0); /* Automatically pass */ 3860 3861 case ETHERTYPE_IP: 3862 #ifdef INET6 3863 case ETHERTYPE_IPV6: 3864 #endif /* INET6 */ 3865 break; 3866 3867 default: 3868 /* 3869 * Check to see if the user wants to pass non-ip 3870 * packets, these will not be checked by pfil(9) 3871 * and passed unconditionally so the default is to drop. 3872 */ 3873 if (pfil_onlyip) 3874 goto bad; 3875 } 3876 3877 /* Strip off the Ethernet header and keep a copy. */ 3878 m_copydata(*mp, 0, ETHER_HDR_LEN, (caddr_t) &eh2); 3879 m_adj(*mp, ETHER_HDR_LEN); 3880 3881 /* Strip off snap header, if present */ 3882 if (snap) { 3883 m_copydata(*mp, 0, sizeof(struct llc), (caddr_t) &llc1); 3884 m_adj(*mp, sizeof(struct llc)); 3885 } 3886 3887 /* 3888 * Check the IP header for alignment and errors 3889 */ 3890 if (dir == PFIL_IN) { 3891 switch (ether_type) { 3892 case ETHERTYPE_IP: 3893 error = bridge_ip_checkbasic(mp); 3894 break; 3895 #ifdef INET6 3896 case ETHERTYPE_IPV6: 3897 error = bridge_ip6_checkbasic(mp); 3898 break; 3899 #endif /* INET6 */ 3900 default: 3901 error = 0; 3902 } 3903 if (error) 3904 goto bad; 3905 } 3906 3907 error = 0; 3908 3909 /* 3910 * Run the packet through pfil 3911 */ 3912 switch (ether_type) { 3913 case ETHERTYPE_IP: 3914 /* 3915 * before calling the firewall, swap fields the same as 3916 * IP does. here we assume the header is contiguous 3917 */ 3918 ip = mtod(*mp, struct ip *); 3919 3920 ip->ip_len = ntohs(ip->ip_len); 3921 ip->ip_off = ntohs(ip->ip_off); 3922 3923 /* 3924 * Run pfil on the member interface and the bridge, both can 3925 * be skipped by clearing pfil_member or pfil_bridge. 3926 * 3927 * Keep the order: 3928 * in_if -> bridge_if -> out_if 3929 */ 3930 if (pfil_bridge && dir == PFIL_OUT && bifp != NULL) { 3931 error = pfil_run_hooks(&inet_pfil_hook, mp, bifp, dir); 3932 if (*mp == NULL || error != 0) /* filter may consume */ 3933 break; 3934 error = bridge_post_pfil(*mp); 3935 if (error) 3936 break; 3937 } 3938 3939 if (pfil_member && ifp != NULL) { 3940 error = pfil_run_hooks(&inet_pfil_hook, mp, ifp, dir); 3941 if (*mp == NULL || error != 0) /* filter may consume */ 3942 break; 3943 error = bridge_post_pfil(*mp); 3944 if (error) 3945 break; 3946 } 3947 3948 if (pfil_bridge && dir == PFIL_IN && bifp != NULL) { 3949 error = pfil_run_hooks(&inet_pfil_hook, mp, bifp, dir); 3950 if (*mp == NULL || error != 0) /* filter may consume */ 3951 break; 3952 error = bridge_post_pfil(*mp); 3953 if (error) 3954 break; 3955 } 3956 3957 /* check if we need to fragment the packet */ 3958 if (pfil_member && ifp != NULL && dir == PFIL_OUT) { 3959 i = (*mp)->m_pkthdr.len; 3960 if (i > ifp->if_mtu) { 3961 error = bridge_fragment(ifp, *mp, &eh2, snap, 3962 &llc1); 3963 return (error); 3964 } 3965 } 3966 3967 /* Recalculate the ip checksum and restore byte ordering */ 3968 ip = mtod(*mp, struct ip *); 3969 hlen = ip->ip_hl << 2; 3970 if (hlen < sizeof(struct ip)) 3971 goto bad; 3972 if (hlen > (*mp)->m_len) { 3973 if ((*mp = m_pullup(*mp, hlen)) == NULL) 3974 goto bad; 3975 ip = mtod(*mp, struct ip *); 3976 if (ip == NULL) 3977 goto bad; 3978 } 3979 ip->ip_len = htons(ip->ip_len); 3980 ip->ip_off = htons(ip->ip_off); 3981 ip->ip_sum = 0; 3982 if (hlen == sizeof(struct ip)) 3983 ip->ip_sum = in_cksum_hdr(ip); 3984 else 3985 ip->ip_sum = in_cksum(*mp, hlen); 3986 3987 break; 3988 #ifdef INET6 3989 case ETHERTYPE_IPV6: 3990 if (pfil_bridge && dir == PFIL_OUT && bifp != NULL) 3991 error = pfil_run_hooks(&inet6_pfil_hook, mp, bifp, 3992 dir); 3993 3994 if (*mp == NULL || error != 0) /* filter may consume */ 3995 break; 3996 3997 if (pfil_member && ifp != NULL) 3998 error = pfil_run_hooks(&inet6_pfil_hook, mp, ifp, 3999 dir); 4000 4001 if (*mp == NULL || error != 0) /* filter may consume */ 4002 break; 4003 4004 if (pfil_bridge && dir == PFIL_IN && bifp != NULL) 4005 error = pfil_run_hooks(&inet6_pfil_hook, mp, bifp, 4006 dir); 4007 break; 4008 #endif 4009 default: 4010 error = 0; 4011 break; 4012 } 4013 4014 if (*mp == NULL) 4015 return (error); 4016 if (error != 0) 4017 goto bad; 4018 4019 error = -1; 4020 4021 /* 4022 * Finally, put everything back the way it was and return 4023 */ 4024 if (snap) { 4025 M_PREPEND(*mp, sizeof(struct llc), MB_DONTWAIT); 4026 if (*mp == NULL) 4027 return (error); 4028 bcopy(&llc1, mtod(*mp, caddr_t), sizeof(struct llc)); 4029 } 4030 4031 M_PREPEND(*mp, ETHER_HDR_LEN, MB_DONTWAIT); 4032 if (*mp == NULL) 4033 return (error); 4034 bcopy(&eh2, mtod(*mp, caddr_t), ETHER_HDR_LEN); 4035 4036 return (0); 4037 4038 bad: 4039 m_freem(*mp); 4040 *mp = NULL; 4041 return (error); 4042 } 4043 4044 /* 4045 * Perform basic checks on header size since 4046 * pfil assumes ip_input has already processed 4047 * it for it. Cut-and-pasted from ip_input.c. 4048 * Given how simple the IPv6 version is, 4049 * does the IPv4 version really need to be 4050 * this complicated? 4051 * 4052 * XXX Should we update ipstat here, or not? 4053 * XXX Right now we update ipstat but not 4054 * XXX csum_counter. 4055 */ 4056 static int 4057 bridge_ip_checkbasic(struct mbuf **mp) 4058 { 4059 struct mbuf *m = *mp; 4060 struct ip *ip; 4061 int len, hlen; 4062 u_short sum; 4063 4064 if (*mp == NULL) 4065 return (-1); 4066 #if 0 /* notyet */ 4067 if (IP_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { 4068 if ((m = m_copyup(m, sizeof(struct ip), 4069 (max_linkhdr + 3) & ~3)) == NULL) { 4070 /* XXXJRT new stat, please */ 4071 ipstat.ips_toosmall++; 4072 goto bad; 4073 } 4074 } else 4075 #endif 4076 #ifndef __predict_false 4077 #define __predict_false(x) x 4078 #endif 4079 if (__predict_false(m->m_len < sizeof (struct ip))) { 4080 if ((m = m_pullup(m, sizeof (struct ip))) == NULL) { 4081 ipstat.ips_toosmall++; 4082 goto bad; 4083 } 4084 } 4085 ip = mtod(m, struct ip *); 4086 if (ip == NULL) goto bad; 4087 4088 if (ip->ip_v != IPVERSION) { 4089 ipstat.ips_badvers++; 4090 goto bad; 4091 } 4092 hlen = ip->ip_hl << 2; 4093 if (hlen < sizeof(struct ip)) { /* minimum header length */ 4094 ipstat.ips_badhlen++; 4095 goto bad; 4096 } 4097 if (hlen > m->m_len) { 4098 if ((m = m_pullup(m, hlen)) == NULL) { 4099 ipstat.ips_badhlen++; 4100 goto bad; 4101 } 4102 ip = mtod(m, struct ip *); 4103 if (ip == NULL) goto bad; 4104 } 4105 4106 if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) { 4107 sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID); 4108 } else { 4109 if (hlen == sizeof(struct ip)) { 4110 sum = in_cksum_hdr(ip); 4111 } else { 4112 sum = in_cksum(m, hlen); 4113 } 4114 } 4115 if (sum) { 4116 ipstat.ips_badsum++; 4117 goto bad; 4118 } 4119 4120 /* Retrieve the packet length. */ 4121 len = ntohs(ip->ip_len); 4122 4123 /* 4124 * Check for additional length bogosity 4125 */ 4126 if (len < hlen) { 4127 ipstat.ips_badlen++; 4128 goto bad; 4129 } 4130 4131 /* 4132 * Check that the amount of data in the buffers 4133 * is as at least much as the IP header would have us expect. 4134 * Drop packet if shorter than we expect. 4135 */ 4136 if (m->m_pkthdr.len < len) { 4137 ipstat.ips_tooshort++; 4138 goto bad; 4139 } 4140 4141 /* Checks out, proceed */ 4142 *mp = m; 4143 return (0); 4144 4145 bad: 4146 *mp = m; 4147 return (-1); 4148 } 4149 4150 #ifdef INET6 4151 /* 4152 * Same as above, but for IPv6. 4153 * Cut-and-pasted from ip6_input.c. 4154 * XXX Should we update ip6stat, or not? 4155 */ 4156 static int 4157 bridge_ip6_checkbasic(struct mbuf **mp) 4158 { 4159 struct mbuf *m = *mp; 4160 struct ip6_hdr *ip6; 4161 4162 /* 4163 * If the IPv6 header is not aligned, slurp it up into a new 4164 * mbuf with space for link headers, in the event we forward 4165 * it. Otherwise, if it is aligned, make sure the entire base 4166 * IPv6 header is in the first mbuf of the chain. 4167 */ 4168 #if 0 /* notyet */ 4169 if (IP6_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { 4170 struct ifnet *inifp = m->m_pkthdr.rcvif; 4171 if ((m = m_copyup(m, sizeof(struct ip6_hdr), 4172 (max_linkhdr + 3) & ~3)) == NULL) { 4173 /* XXXJRT new stat, please */ 4174 ip6stat.ip6s_toosmall++; 4175 in6_ifstat_inc(inifp, ifs6_in_hdrerr); 4176 goto bad; 4177 } 4178 } else 4179 #endif 4180 if (__predict_false(m->m_len < sizeof(struct ip6_hdr))) { 4181 struct ifnet *inifp = m->m_pkthdr.rcvif; 4182 if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) { 4183 ip6stat.ip6s_toosmall++; 4184 in6_ifstat_inc(inifp, ifs6_in_hdrerr); 4185 goto bad; 4186 } 4187 } 4188 4189 ip6 = mtod(m, struct ip6_hdr *); 4190 4191 if ((ip6->ip6_vfc & IPV6_VERSION_MASK) != IPV6_VERSION) { 4192 ip6stat.ip6s_badvers++; 4193 in6_ifstat_inc(m->m_pkthdr.rcvif, ifs6_in_hdrerr); 4194 goto bad; 4195 } 4196 4197 /* Checks out, proceed */ 4198 *mp = m; 4199 return (0); 4200 4201 bad: 4202 *mp = m; 4203 return (-1); 4204 } 4205 #endif /* INET6 */ 4206 4207 /* 4208 * bridge_fragment: 4209 * 4210 * Return a fragmented mbuf chain. 4211 */ 4212 static int 4213 bridge_fragment(struct ifnet *ifp, struct mbuf *m, struct ether_header *eh, 4214 int snap, struct llc *llc) 4215 { 4216 struct mbuf *m0; 4217 struct ip *ip; 4218 int error = -1; 4219 4220 if (m->m_len < sizeof(struct ip) && 4221 (m = m_pullup(m, sizeof(struct ip))) == NULL) 4222 goto out; 4223 ip = mtod(m, struct ip *); 4224 4225 error = ip_fragment(ip, &m, ifp->if_mtu, ifp->if_hwassist, 4226 CSUM_DELAY_IP); 4227 if (error) 4228 goto out; 4229 4230 /* walk the chain and re-add the Ethernet header */ 4231 for (m0 = m; m0; m0 = m0->m_nextpkt) { 4232 if (error == 0) { 4233 if (snap) { 4234 M_PREPEND(m0, sizeof(struct llc), MB_DONTWAIT); 4235 if (m0 == NULL) { 4236 error = ENOBUFS; 4237 continue; 4238 } 4239 bcopy(llc, mtod(m0, caddr_t), 4240 sizeof(struct llc)); 4241 } 4242 M_PREPEND(m0, ETHER_HDR_LEN, MB_DONTWAIT); 4243 if (m0 == NULL) { 4244 error = ENOBUFS; 4245 continue; 4246 } 4247 bcopy(eh, mtod(m0, caddr_t), ETHER_HDR_LEN); 4248 } else 4249 m_freem(m); 4250 } 4251 4252 if (error == 0) 4253 ipstat.ips_fragmented++; 4254 4255 return (error); 4256 4257 out: 4258 if (m != NULL) 4259 m_freem(m); 4260 return (error); 4261 } 4262 4263 static void 4264 bridge_enqueue_handler(netmsg_t msg) 4265 { 4266 struct netmsg_packet *nmp; 4267 struct ifnet *dst_ifp; 4268 struct mbuf *m; 4269 4270 nmp = &msg->packet; 4271 m = nmp->nm_packet; 4272 dst_ifp = nmp->base.lmsg.u.ms_resultp; 4273 mbuftrackid(m, 71); 4274 4275 bridge_handoff(dst_ifp->if_bridge, dst_ifp, m, 1); 4276 } 4277 4278 static void 4279 bridge_handoff(struct bridge_softc *sc, struct ifnet *dst_ifp, 4280 struct mbuf *m, int from_us) 4281 { 4282 struct mbuf *m0; 4283 struct ifnet *bifp; 4284 4285 bifp = sc->sc_ifp; 4286 mbuftrackid(m, 72); 4287 4288 /* We may be sending a fragment so traverse the mbuf */ 4289 for (; m; m = m0) { 4290 struct altq_pktattr pktattr; 4291 4292 m0 = m->m_nextpkt; 4293 m->m_nextpkt = NULL; 4294 4295 /* 4296 * If being sent from our host override ether_shost 4297 * with the bridge MAC. This is mandatory for ARP 4298 * so things don't get confused. In particular we 4299 * don't want ARPs to get associated with link interfaces 4300 * under the bridge which might or might not stay valid. 4301 * 4302 * Also override ether_shost when relaying a packet out 4303 * the same interface it came in on, due to multi-homed 4304 * addresses & default routes, otherwise switches will 4305 * get very confused. 4306 * 4307 * Otherwise if we are in transparent mode. 4308 */ 4309 if (from_us || m->m_pkthdr.rcvif == dst_ifp) { 4310 m_copyback(m, 4311 offsetof(struct ether_header, ether_shost), 4312 ETHER_ADDR_LEN, IF_LLADDR(sc->sc_ifp)); 4313 } else if ((bifp->if_flags & IFF_LINK0) && 4314 (m->m_pkthdr.fw_flags & BRIDGE_MBUF_TAGGED)) { 4315 m_copyback(m, 4316 offsetof(struct ether_header, ether_shost), 4317 ETHER_ADDR_LEN, 4318 m->m_pkthdr.br.ether.ether_shost); 4319 } /* else retain shost */ 4320 4321 if (ifq_is_enabled(&dst_ifp->if_snd)) 4322 altq_etherclassify(&dst_ifp->if_snd, m, &pktattr); 4323 4324 ifq_dispatch(dst_ifp, m, &pktattr); 4325 } 4326 } 4327 4328 static void 4329 bridge_control_dispatch(netmsg_t msg) 4330 { 4331 struct netmsg_brctl *bc_msg = (struct netmsg_brctl *)msg; 4332 struct ifnet *bifp = bc_msg->bc_sc->sc_ifp; 4333 int error; 4334 4335 ifnet_serialize_all(bifp); 4336 error = bc_msg->bc_func(bc_msg->bc_sc, bc_msg->bc_arg); 4337 ifnet_deserialize_all(bifp); 4338 4339 lwkt_replymsg(&bc_msg->base.lmsg, error); 4340 } 4341 4342 static int 4343 bridge_control(struct bridge_softc *sc, u_long cmd, 4344 bridge_ctl_t bc_func, void *bc_arg) 4345 { 4346 struct ifnet *bifp = sc->sc_ifp; 4347 struct netmsg_brctl bc_msg; 4348 int error; 4349 4350 ASSERT_IFNET_SERIALIZED_ALL(bifp); 4351 4352 bzero(&bc_msg, sizeof(bc_msg)); 4353 4354 netmsg_init(&bc_msg.base, NULL, &curthread->td_msgport, 4355 0, bridge_control_dispatch); 4356 bc_msg.bc_func = bc_func; 4357 bc_msg.bc_sc = sc; 4358 bc_msg.bc_arg = bc_arg; 4359 4360 ifnet_deserialize_all(bifp); 4361 error = lwkt_domsg(BRIDGE_CFGPORT, &bc_msg.base.lmsg, 0); 4362 ifnet_serialize_all(bifp); 4363 return error; 4364 } 4365 4366 static void 4367 bridge_add_bif_handler(netmsg_t msg) 4368 { 4369 struct netmsg_braddbif *amsg = (struct netmsg_braddbif *)msg; 4370 struct bridge_softc *sc; 4371 struct bridge_iflist *bif; 4372 4373 sc = amsg->br_softc; 4374 4375 bif = kmalloc(sizeof(*bif), M_DEVBUF, M_WAITOK | M_ZERO); 4376 bif->bif_ifp = amsg->br_bif_ifp; 4377 bif->bif_onlist = 1; 4378 bif->bif_info = amsg->br_bif_info; 4379 4380 /* 4381 * runs through bif_info 4382 */ 4383 bif->bif_flags = IFBIF_LEARNING | IFBIF_DISCOVER; 4384 4385 TAILQ_INSERT_HEAD(&sc->sc_iflists[mycpuid], bif, bif_next); 4386 4387 ifnet_forwardmsg(&amsg->base.lmsg, mycpuid + 1); 4388 } 4389 4390 static void 4391 bridge_add_bif(struct bridge_softc *sc, struct bridge_ifinfo *bif_info, 4392 struct ifnet *ifp) 4393 { 4394 struct netmsg_braddbif amsg; 4395 4396 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 4397 4398 netmsg_init(&amsg.base, NULL, &curthread->td_msgport, 4399 0, bridge_add_bif_handler); 4400 amsg.br_softc = sc; 4401 amsg.br_bif_info = bif_info; 4402 amsg.br_bif_ifp = ifp; 4403 4404 ifnet_domsg(&amsg.base.lmsg, 0); 4405 } 4406 4407 static void 4408 bridge_del_bif_handler(netmsg_t msg) 4409 { 4410 struct netmsg_brdelbif *dmsg = (struct netmsg_brdelbif *)msg; 4411 struct bridge_softc *sc; 4412 struct bridge_iflist *bif; 4413 4414 sc = dmsg->br_softc; 4415 4416 /* 4417 * Locate the bif associated with the br_bif_info 4418 * on the current CPU 4419 */ 4420 bif = bridge_lookup_member_ifinfo(sc, dmsg->br_bif_info); 4421 KKASSERT(bif != NULL && bif->bif_onlist); 4422 4423 /* Remove the bif from the current CPU's iflist */ 4424 bif->bif_onlist = 0; 4425 TAILQ_REMOVE(dmsg->br_bif_list, bif, bif_next); 4426 4427 /* Save the removed bif for later freeing */ 4428 TAILQ_INSERT_HEAD(dmsg->br_bif_list, bif, bif_next); 4429 4430 ifnet_forwardmsg(&dmsg->base.lmsg, mycpuid + 1); 4431 } 4432 4433 static void 4434 bridge_del_bif(struct bridge_softc *sc, struct bridge_ifinfo *bif_info, 4435 struct bridge_iflist_head *saved_bifs) 4436 { 4437 struct netmsg_brdelbif dmsg; 4438 4439 ASSERT_IFNET_NOT_SERIALIZED_ALL(sc->sc_ifp); 4440 4441 netmsg_init(&dmsg.base, NULL, &curthread->td_msgport, 4442 0, bridge_del_bif_handler); 4443 dmsg.br_softc = sc; 4444 dmsg.br_bif_info = bif_info; 4445 dmsg.br_bif_list = saved_bifs; 4446 4447 ifnet_domsg(&dmsg.base.lmsg, 0); 4448 } 4449