1 /* $NetBSD: if_bridge.c,v 1.31 2005/06/01 19:45:34 jdc Exp $ */ 2 3 /*- 4 * SPDX-License-Identifier: BSD-4-Clause 5 * 6 * Copyright 2001 Wasabi Systems, Inc. 7 * All rights reserved. 8 * 9 * Written by Jason R. Thorpe for Wasabi Systems, Inc. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 3. All advertising materials mentioning features or use of this software 20 * must display the following acknowledgement: 21 * This product includes software developed for the NetBSD Project by 22 * Wasabi Systems, Inc. 23 * 4. The name of Wasabi Systems, Inc. may not be used to endorse 24 * or promote products derived from this software without specific prior 25 * written permission. 26 * 27 * THIS SOFTWARE IS PROVIDED BY WASABI SYSTEMS, INC. ``AS IS'' AND 28 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 29 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 30 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL WASABI SYSTEMS, INC 31 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 32 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 33 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 34 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 35 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 37 * POSSIBILITY OF SUCH DAMAGE. 38 */ 39 40 /* 41 * Copyright (c) 1999, 2000 Jason L. Wright (jason@thought.net) 42 * All rights reserved. 43 * 44 * Redistribution and use in source and binary forms, with or without 45 * modification, are permitted provided that the following conditions 46 * are met: 47 * 1. Redistributions of source code must retain the above copyright 48 * notice, this list of conditions and the following disclaimer. 49 * 2. Redistributions in binary form must reproduce the above copyright 50 * notice, this list of conditions and the following disclaimer in the 51 * documentation and/or other materials provided with the distribution. 52 * 53 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR 54 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 55 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 56 * DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, 57 * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 58 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR 59 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 60 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, 61 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN 62 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 63 * POSSIBILITY OF SUCH DAMAGE. 64 * 65 * OpenBSD: if_bridge.c,v 1.60 2001/06/15 03:38:33 itojun Exp 66 */ 67 68 /* 69 * Network interface bridge support. 70 * 71 * TODO: 72 * 73 * - Currently only supports Ethernet-like interfaces (Ethernet, 74 * 802.11, VLANs on Ethernet, etc.) Figure out a nice way 75 * to bridge other types of interfaces (maybe consider 76 * heterogeneous bridges). 77 */ 78 79 #include <sys/cdefs.h> 80 __FBSDID("$FreeBSD$"); 81 82 #include "opt_inet.h" 83 #include "opt_inet6.h" 84 85 #include <sys/param.h> 86 #include <sys/eventhandler.h> 87 #include <sys/mbuf.h> 88 #include <sys/malloc.h> 89 #include <sys/protosw.h> 90 #include <sys/systm.h> 91 #include <sys/jail.h> 92 #include <sys/time.h> 93 #include <sys/socket.h> /* for net/if.h */ 94 #include <sys/sockio.h> 95 #include <sys/ctype.h> /* string functions */ 96 #include <sys/kernel.h> 97 #include <sys/random.h> 98 #include <sys/syslog.h> 99 #include <sys/sysctl.h> 100 #include <vm/uma.h> 101 #include <sys/module.h> 102 #include <sys/priv.h> 103 #include <sys/proc.h> 104 #include <sys/lock.h> 105 #include <sys/mutex.h> 106 107 #include <net/bpf.h> 108 #include <net/if.h> 109 #include <net/if_clone.h> 110 #include <net/if_dl.h> 111 #include <net/if_types.h> 112 #include <net/if_var.h> 113 #include <net/pfil.h> 114 #include <net/vnet.h> 115 116 #include <netinet/in.h> 117 #include <netinet/in_systm.h> 118 #include <netinet/in_var.h> 119 #include <netinet/ip.h> 120 #include <netinet/ip_var.h> 121 #ifdef INET6 122 #include <netinet/ip6.h> 123 #include <netinet6/ip6_var.h> 124 #include <netinet6/in6_ifattach.h> 125 #endif 126 #if defined(INET) || defined(INET6) 127 #include <netinet/ip_carp.h> 128 #endif 129 #include <machine/in_cksum.h> 130 #include <netinet/if_ether.h> 131 #include <net/bridgestp.h> 132 #include <net/if_bridgevar.h> 133 #include <net/if_llc.h> 134 #include <net/if_vlan_var.h> 135 136 #include <net/route.h> 137 138 /* 139 * Size of the route hash table. Must be a power of two. 140 */ 141 #ifndef BRIDGE_RTHASH_SIZE 142 #define BRIDGE_RTHASH_SIZE 1024 143 #endif 144 145 #define BRIDGE_RTHASH_MASK (BRIDGE_RTHASH_SIZE - 1) 146 147 /* 148 * Default maximum number of addresses to cache. 149 */ 150 #ifndef BRIDGE_RTABLE_MAX 151 #define BRIDGE_RTABLE_MAX 2000 152 #endif 153 154 /* 155 * Timeout (in seconds) for entries learned dynamically. 156 */ 157 #ifndef BRIDGE_RTABLE_TIMEOUT 158 #define BRIDGE_RTABLE_TIMEOUT (20 * 60) /* same as ARP */ 159 #endif 160 161 /* 162 * Number of seconds between walks of the route list. 163 */ 164 #ifndef BRIDGE_RTABLE_PRUNE_PERIOD 165 #define BRIDGE_RTABLE_PRUNE_PERIOD (5 * 60) 166 #endif 167 168 /* 169 * List of capabilities to possibly mask on the member interface. 170 */ 171 #define BRIDGE_IFCAPS_MASK (IFCAP_TOE|IFCAP_TSO|IFCAP_TXCSUM|\ 172 IFCAP_TXCSUM_IPV6) 173 174 /* 175 * List of capabilities to strip 176 */ 177 #define BRIDGE_IFCAPS_STRIP IFCAP_LRO 178 179 /* 180 * Bridge interface list entry. 181 */ 182 struct bridge_iflist { 183 LIST_ENTRY(bridge_iflist) bif_next; 184 struct ifnet *bif_ifp; /* member if */ 185 struct bstp_port bif_stp; /* STP state */ 186 uint32_t bif_flags; /* member if flags */ 187 int bif_savedcaps; /* saved capabilities */ 188 uint32_t bif_addrmax; /* max # of addresses */ 189 uint32_t bif_addrcnt; /* cur. # of addresses */ 190 uint32_t bif_addrexceeded;/* # of address violations */ 191 }; 192 193 /* 194 * Bridge route node. 195 */ 196 struct bridge_rtnode { 197 LIST_ENTRY(bridge_rtnode) brt_hash; /* hash table linkage */ 198 LIST_ENTRY(bridge_rtnode) brt_list; /* list linkage */ 199 struct bridge_iflist *brt_dst; /* destination if */ 200 unsigned long brt_expire; /* expiration time */ 201 uint8_t brt_flags; /* address flags */ 202 uint8_t brt_addr[ETHER_ADDR_LEN]; 203 uint16_t brt_vlan; /* vlan id */ 204 }; 205 #define brt_ifp brt_dst->bif_ifp 206 207 /* 208 * Software state for each bridge. 209 */ 210 struct bridge_softc { 211 struct ifnet *sc_ifp; /* make this an interface */ 212 LIST_ENTRY(bridge_softc) sc_list; 213 struct mtx sc_mtx; 214 struct cv sc_cv; 215 uint32_t sc_brtmax; /* max # of addresses */ 216 uint32_t sc_brtcnt; /* cur. # of addresses */ 217 uint32_t sc_brttimeout; /* rt timeout in seconds */ 218 struct callout sc_brcallout; /* bridge callout */ 219 uint32_t sc_iflist_ref; /* refcount for sc_iflist */ 220 uint32_t sc_iflist_xcnt; /* refcount for sc_iflist */ 221 LIST_HEAD(, bridge_iflist) sc_iflist; /* member interface list */ 222 LIST_HEAD(, bridge_rtnode) *sc_rthash; /* our forwarding table */ 223 LIST_HEAD(, bridge_rtnode) sc_rtlist; /* list version of above */ 224 uint32_t sc_rthash_key; /* key for hash */ 225 LIST_HEAD(, bridge_iflist) sc_spanlist; /* span ports list */ 226 struct bstp_state sc_stp; /* STP state */ 227 uint32_t sc_brtexceeded; /* # of cache drops */ 228 struct ifnet *sc_ifaddr; /* member mac copied from */ 229 u_char sc_defaddr[6]; /* Default MAC address */ 230 }; 231 232 VNET_DEFINE_STATIC(struct mtx, bridge_list_mtx); 233 #define V_bridge_list_mtx VNET(bridge_list_mtx) 234 static eventhandler_tag bridge_detach_cookie; 235 236 int bridge_rtable_prune_period = BRIDGE_RTABLE_PRUNE_PERIOD; 237 238 uma_zone_t bridge_rtnode_zone; 239 240 static int bridge_clone_create(struct if_clone *, int, caddr_t); 241 static void bridge_clone_destroy(struct ifnet *); 242 243 static int bridge_ioctl(struct ifnet *, u_long, caddr_t); 244 static void bridge_mutecaps(struct bridge_softc *); 245 static void bridge_set_ifcap(struct bridge_softc *, struct bridge_iflist *, 246 int); 247 static void bridge_ifdetach(void *arg __unused, struct ifnet *); 248 static void bridge_init(void *); 249 static void bridge_dummynet(struct mbuf *, struct ifnet *); 250 static void bridge_stop(struct ifnet *, int); 251 static int bridge_transmit(struct ifnet *, struct mbuf *); 252 static void bridge_qflush(struct ifnet *); 253 static struct mbuf *bridge_input(struct ifnet *, struct mbuf *); 254 static int bridge_output(struct ifnet *, struct mbuf *, struct sockaddr *, 255 struct rtentry *); 256 static int bridge_enqueue(struct bridge_softc *, struct ifnet *, 257 struct mbuf *); 258 static void bridge_rtdelete(struct bridge_softc *, struct ifnet *ifp, int); 259 260 static void bridge_forward(struct bridge_softc *, struct bridge_iflist *, 261 struct mbuf *m); 262 263 static void bridge_timer(void *); 264 265 static void bridge_broadcast(struct bridge_softc *, struct ifnet *, 266 struct mbuf *, int); 267 static void bridge_span(struct bridge_softc *, struct mbuf *); 268 269 static int bridge_rtupdate(struct bridge_softc *, const uint8_t *, 270 uint16_t, struct bridge_iflist *, int, uint8_t); 271 static struct ifnet *bridge_rtlookup(struct bridge_softc *, const uint8_t *, 272 uint16_t); 273 static void bridge_rttrim(struct bridge_softc *); 274 static void bridge_rtage(struct bridge_softc *); 275 static void bridge_rtflush(struct bridge_softc *, int); 276 static int bridge_rtdaddr(struct bridge_softc *, const uint8_t *, 277 uint16_t); 278 279 static void bridge_rtable_init(struct bridge_softc *); 280 static void bridge_rtable_fini(struct bridge_softc *); 281 282 static int bridge_rtnode_addr_cmp(const uint8_t *, const uint8_t *); 283 static struct bridge_rtnode *bridge_rtnode_lookup(struct bridge_softc *, 284 const uint8_t *, uint16_t); 285 static int bridge_rtnode_insert(struct bridge_softc *, 286 struct bridge_rtnode *); 287 static void bridge_rtnode_destroy(struct bridge_softc *, 288 struct bridge_rtnode *); 289 static void bridge_rtable_expire(struct ifnet *, int); 290 static void bridge_state_change(struct ifnet *, int); 291 292 static struct bridge_iflist *bridge_lookup_member(struct bridge_softc *, 293 const char *name); 294 static struct bridge_iflist *bridge_lookup_member_if(struct bridge_softc *, 295 struct ifnet *ifp); 296 static void bridge_delete_member(struct bridge_softc *, 297 struct bridge_iflist *, int); 298 static void bridge_delete_span(struct bridge_softc *, 299 struct bridge_iflist *); 300 301 static int bridge_ioctl_add(struct bridge_softc *, void *); 302 static int bridge_ioctl_del(struct bridge_softc *, void *); 303 static int bridge_ioctl_gifflags(struct bridge_softc *, void *); 304 static int bridge_ioctl_sifflags(struct bridge_softc *, void *); 305 static int bridge_ioctl_scache(struct bridge_softc *, void *); 306 static int bridge_ioctl_gcache(struct bridge_softc *, void *); 307 static int bridge_ioctl_gifs(struct bridge_softc *, void *); 308 static int bridge_ioctl_rts(struct bridge_softc *, void *); 309 static int bridge_ioctl_saddr(struct bridge_softc *, void *); 310 static int bridge_ioctl_sto(struct bridge_softc *, void *); 311 static int bridge_ioctl_gto(struct bridge_softc *, void *); 312 static int bridge_ioctl_daddr(struct bridge_softc *, void *); 313 static int bridge_ioctl_flush(struct bridge_softc *, void *); 314 static int bridge_ioctl_gpri(struct bridge_softc *, void *); 315 static int bridge_ioctl_spri(struct bridge_softc *, void *); 316 static int bridge_ioctl_ght(struct bridge_softc *, void *); 317 static int bridge_ioctl_sht(struct bridge_softc *, void *); 318 static int bridge_ioctl_gfd(struct bridge_softc *, void *); 319 static int bridge_ioctl_sfd(struct bridge_softc *, void *); 320 static int bridge_ioctl_gma(struct bridge_softc *, void *); 321 static int bridge_ioctl_sma(struct bridge_softc *, void *); 322 static int bridge_ioctl_sifprio(struct bridge_softc *, void *); 323 static int bridge_ioctl_sifcost(struct bridge_softc *, void *); 324 static int bridge_ioctl_sifmaxaddr(struct bridge_softc *, void *); 325 static int bridge_ioctl_addspan(struct bridge_softc *, void *); 326 static int bridge_ioctl_delspan(struct bridge_softc *, void *); 327 static int bridge_ioctl_gbparam(struct bridge_softc *, void *); 328 static int bridge_ioctl_grte(struct bridge_softc *, void *); 329 static int bridge_ioctl_gifsstp(struct bridge_softc *, void *); 330 static int bridge_ioctl_sproto(struct bridge_softc *, void *); 331 static int bridge_ioctl_stxhc(struct bridge_softc *, void *); 332 static int bridge_pfil(struct mbuf **, struct ifnet *, struct ifnet *, 333 int); 334 static int bridge_ip_checkbasic(struct mbuf **mp); 335 #ifdef INET6 336 static int bridge_ip6_checkbasic(struct mbuf **mp); 337 #endif /* INET6 */ 338 static int bridge_fragment(struct ifnet *, struct mbuf **mp, 339 struct ether_header *, int, struct llc *); 340 static void bridge_linkstate(struct ifnet *ifp); 341 static void bridge_linkcheck(struct bridge_softc *sc); 342 343 344 /* The default bridge vlan is 1 (IEEE 802.1Q-2003 Table 9-2) */ 345 #define VLANTAGOF(_m) \ 346 (_m->m_flags & M_VLANTAG) ? EVL_VLANOFTAG(_m->m_pkthdr.ether_vtag) : 1 347 348 static struct bstp_cb_ops bridge_ops = { 349 .bcb_state = bridge_state_change, 350 .bcb_rtage = bridge_rtable_expire 351 }; 352 353 SYSCTL_DECL(_net_link); 354 static SYSCTL_NODE(_net_link, IFT_BRIDGE, bridge, CTLFLAG_RW, 0, "Bridge"); 355 356 /* only pass IP[46] packets when pfil is enabled */ 357 VNET_DEFINE_STATIC(int, pfil_onlyip) = 1; 358 #define V_pfil_onlyip VNET(pfil_onlyip) 359 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_onlyip, 360 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_onlyip), 0, 361 "Only pass IP packets when pfil is enabled"); 362 363 /* run pfil hooks on the bridge interface */ 364 VNET_DEFINE_STATIC(int, pfil_bridge) = 1; 365 #define V_pfil_bridge VNET(pfil_bridge) 366 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_bridge, 367 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_bridge), 0, 368 "Packet filter on the bridge interface"); 369 370 /* layer2 filter with ipfw */ 371 VNET_DEFINE_STATIC(int, pfil_ipfw); 372 #define V_pfil_ipfw VNET(pfil_ipfw) 373 374 /* layer2 ARP filter with ipfw */ 375 VNET_DEFINE_STATIC(int, pfil_ipfw_arp); 376 #define V_pfil_ipfw_arp VNET(pfil_ipfw_arp) 377 SYSCTL_INT(_net_link_bridge, OID_AUTO, ipfw_arp, 378 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_ipfw_arp), 0, 379 "Filter ARP packets through IPFW layer2"); 380 381 /* run pfil hooks on the member interface */ 382 VNET_DEFINE_STATIC(int, pfil_member) = 1; 383 #define V_pfil_member VNET(pfil_member) 384 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_member, 385 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_member), 0, 386 "Packet filter on the member interface"); 387 388 /* run pfil hooks on the physical interface for locally destined packets */ 389 VNET_DEFINE_STATIC(int, pfil_local_phys); 390 #define V_pfil_local_phys VNET(pfil_local_phys) 391 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_local_phys, 392 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_local_phys), 0, 393 "Packet filter on the physical interface for locally destined packets"); 394 395 /* log STP state changes */ 396 VNET_DEFINE_STATIC(int, log_stp); 397 #define V_log_stp VNET(log_stp) 398 SYSCTL_INT(_net_link_bridge, OID_AUTO, log_stp, 399 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(log_stp), 0, 400 "Log STP state changes"); 401 402 /* share MAC with first bridge member */ 403 VNET_DEFINE_STATIC(int, bridge_inherit_mac); 404 #define V_bridge_inherit_mac VNET(bridge_inherit_mac) 405 SYSCTL_INT(_net_link_bridge, OID_AUTO, inherit_mac, 406 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(bridge_inherit_mac), 0, 407 "Inherit MAC address from the first bridge member"); 408 409 VNET_DEFINE_STATIC(int, allow_llz_overlap) = 0; 410 #define V_allow_llz_overlap VNET(allow_llz_overlap) 411 SYSCTL_INT(_net_link_bridge, OID_AUTO, allow_llz_overlap, 412 CTLFLAG_RW | CTLFLAG_VNET, &VNET_NAME(allow_llz_overlap), 0, 413 "Allow overlap of link-local scope " 414 "zones of a bridge interface and the member interfaces"); 415 416 struct bridge_control { 417 int (*bc_func)(struct bridge_softc *, void *); 418 int bc_argsize; 419 int bc_flags; 420 }; 421 422 #define BC_F_COPYIN 0x01 /* copy arguments in */ 423 #define BC_F_COPYOUT 0x02 /* copy arguments out */ 424 #define BC_F_SUSER 0x04 /* do super-user check */ 425 426 const struct bridge_control bridge_control_table[] = { 427 { bridge_ioctl_add, sizeof(struct ifbreq), 428 BC_F_COPYIN|BC_F_SUSER }, 429 { bridge_ioctl_del, sizeof(struct ifbreq), 430 BC_F_COPYIN|BC_F_SUSER }, 431 432 { bridge_ioctl_gifflags, sizeof(struct ifbreq), 433 BC_F_COPYIN|BC_F_COPYOUT }, 434 { bridge_ioctl_sifflags, sizeof(struct ifbreq), 435 BC_F_COPYIN|BC_F_SUSER }, 436 437 { bridge_ioctl_scache, sizeof(struct ifbrparam), 438 BC_F_COPYIN|BC_F_SUSER }, 439 { bridge_ioctl_gcache, sizeof(struct ifbrparam), 440 BC_F_COPYOUT }, 441 442 { bridge_ioctl_gifs, sizeof(struct ifbifconf), 443 BC_F_COPYIN|BC_F_COPYOUT }, 444 { bridge_ioctl_rts, sizeof(struct ifbaconf), 445 BC_F_COPYIN|BC_F_COPYOUT }, 446 447 { bridge_ioctl_saddr, sizeof(struct ifbareq), 448 BC_F_COPYIN|BC_F_SUSER }, 449 450 { bridge_ioctl_sto, sizeof(struct ifbrparam), 451 BC_F_COPYIN|BC_F_SUSER }, 452 { bridge_ioctl_gto, sizeof(struct ifbrparam), 453 BC_F_COPYOUT }, 454 455 { bridge_ioctl_daddr, sizeof(struct ifbareq), 456 BC_F_COPYIN|BC_F_SUSER }, 457 458 { bridge_ioctl_flush, sizeof(struct ifbreq), 459 BC_F_COPYIN|BC_F_SUSER }, 460 461 { bridge_ioctl_gpri, sizeof(struct ifbrparam), 462 BC_F_COPYOUT }, 463 { bridge_ioctl_spri, sizeof(struct ifbrparam), 464 BC_F_COPYIN|BC_F_SUSER }, 465 466 { bridge_ioctl_ght, sizeof(struct ifbrparam), 467 BC_F_COPYOUT }, 468 { bridge_ioctl_sht, sizeof(struct ifbrparam), 469 BC_F_COPYIN|BC_F_SUSER }, 470 471 { bridge_ioctl_gfd, sizeof(struct ifbrparam), 472 BC_F_COPYOUT }, 473 { bridge_ioctl_sfd, sizeof(struct ifbrparam), 474 BC_F_COPYIN|BC_F_SUSER }, 475 476 { bridge_ioctl_gma, sizeof(struct ifbrparam), 477 BC_F_COPYOUT }, 478 { bridge_ioctl_sma, sizeof(struct ifbrparam), 479 BC_F_COPYIN|BC_F_SUSER }, 480 481 { bridge_ioctl_sifprio, sizeof(struct ifbreq), 482 BC_F_COPYIN|BC_F_SUSER }, 483 484 { bridge_ioctl_sifcost, sizeof(struct ifbreq), 485 BC_F_COPYIN|BC_F_SUSER }, 486 487 { bridge_ioctl_addspan, sizeof(struct ifbreq), 488 BC_F_COPYIN|BC_F_SUSER }, 489 { bridge_ioctl_delspan, sizeof(struct ifbreq), 490 BC_F_COPYIN|BC_F_SUSER }, 491 492 { bridge_ioctl_gbparam, sizeof(struct ifbropreq), 493 BC_F_COPYOUT }, 494 495 { bridge_ioctl_grte, sizeof(struct ifbrparam), 496 BC_F_COPYOUT }, 497 498 { bridge_ioctl_gifsstp, sizeof(struct ifbpstpconf), 499 BC_F_COPYIN|BC_F_COPYOUT }, 500 501 { bridge_ioctl_sproto, sizeof(struct ifbrparam), 502 BC_F_COPYIN|BC_F_SUSER }, 503 504 { bridge_ioctl_stxhc, sizeof(struct ifbrparam), 505 BC_F_COPYIN|BC_F_SUSER }, 506 507 { bridge_ioctl_sifmaxaddr, sizeof(struct ifbreq), 508 BC_F_COPYIN|BC_F_SUSER }, 509 510 }; 511 const int bridge_control_table_size = nitems(bridge_control_table); 512 513 VNET_DEFINE_STATIC(LIST_HEAD(, bridge_softc), bridge_list); 514 #define V_bridge_list VNET(bridge_list) 515 #define BRIDGE_LIST_LOCK_INIT(x) mtx_init(&V_bridge_list_mtx, \ 516 "if_bridge list", NULL, MTX_DEF) 517 #define BRIDGE_LIST_LOCK_DESTROY(x) mtx_destroy(&V_bridge_list_mtx) 518 #define BRIDGE_LIST_LOCK(x) mtx_lock(&V_bridge_list_mtx) 519 #define BRIDGE_LIST_UNLOCK(x) mtx_unlock(&V_bridge_list_mtx) 520 521 VNET_DEFINE_STATIC(struct if_clone *, bridge_cloner); 522 #define V_bridge_cloner VNET(bridge_cloner) 523 524 static const char bridge_name[] = "bridge"; 525 526 static void 527 vnet_bridge_init(const void *unused __unused) 528 { 529 530 BRIDGE_LIST_LOCK_INIT(); 531 LIST_INIT(&V_bridge_list); 532 V_bridge_cloner = if_clone_simple(bridge_name, 533 bridge_clone_create, bridge_clone_destroy, 0); 534 } 535 VNET_SYSINIT(vnet_bridge_init, SI_SUB_PROTO_IFATTACHDOMAIN, SI_ORDER_ANY, 536 vnet_bridge_init, NULL); 537 538 static void 539 vnet_bridge_uninit(const void *unused __unused) 540 { 541 542 if_clone_detach(V_bridge_cloner); 543 V_bridge_cloner = NULL; 544 BRIDGE_LIST_LOCK_DESTROY(); 545 } 546 VNET_SYSUNINIT(vnet_bridge_uninit, SI_SUB_PSEUDO, SI_ORDER_ANY, 547 vnet_bridge_uninit, NULL); 548 549 static int 550 bridge_modevent(module_t mod, int type, void *data) 551 { 552 553 switch (type) { 554 case MOD_LOAD: 555 bridge_rtnode_zone = uma_zcreate("bridge_rtnode", 556 sizeof(struct bridge_rtnode), NULL, NULL, NULL, NULL, 557 UMA_ALIGN_PTR, 0); 558 bridge_dn_p = bridge_dummynet; 559 bridge_detach_cookie = EVENTHANDLER_REGISTER( 560 ifnet_departure_event, bridge_ifdetach, NULL, 561 EVENTHANDLER_PRI_ANY); 562 break; 563 case MOD_UNLOAD: 564 EVENTHANDLER_DEREGISTER(ifnet_departure_event, 565 bridge_detach_cookie); 566 uma_zdestroy(bridge_rtnode_zone); 567 bridge_dn_p = NULL; 568 break; 569 default: 570 return (EOPNOTSUPP); 571 } 572 return (0); 573 } 574 575 static moduledata_t bridge_mod = { 576 "if_bridge", 577 bridge_modevent, 578 0 579 }; 580 581 DECLARE_MODULE(if_bridge, bridge_mod, SI_SUB_PSEUDO, SI_ORDER_ANY); 582 MODULE_VERSION(if_bridge, 1); 583 MODULE_DEPEND(if_bridge, bridgestp, 1, 1, 1); 584 585 /* 586 * handler for net.link.bridge.ipfw 587 */ 588 static int 589 sysctl_pfil_ipfw(SYSCTL_HANDLER_ARGS) 590 { 591 int enable = V_pfil_ipfw; 592 int error; 593 594 error = sysctl_handle_int(oidp, &enable, 0, req); 595 enable &= 1; 596 597 if (enable != V_pfil_ipfw) { 598 V_pfil_ipfw = enable; 599 600 /* 601 * Disable pfil so that ipfw doesnt run twice, if the user 602 * really wants both then they can re-enable pfil_bridge and/or 603 * pfil_member. Also allow non-ip packets as ipfw can filter by 604 * layer2 type. 605 */ 606 if (V_pfil_ipfw) { 607 V_pfil_onlyip = 0; 608 V_pfil_bridge = 0; 609 V_pfil_member = 0; 610 } 611 } 612 613 return (error); 614 } 615 SYSCTL_PROC(_net_link_bridge, OID_AUTO, ipfw, 616 CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_VNET, 617 &VNET_NAME(pfil_ipfw), 0, &sysctl_pfil_ipfw, "I", 618 "Layer2 filter with IPFW"); 619 620 /* 621 * bridge_clone_create: 622 * 623 * Create a new bridge instance. 624 */ 625 static int 626 bridge_clone_create(struct if_clone *ifc, int unit, caddr_t params) 627 { 628 struct bridge_softc *sc, *sc2; 629 struct ifnet *bifp, *ifp; 630 int fb, retry; 631 unsigned long hostid; 632 633 sc = malloc(sizeof(*sc), M_DEVBUF, M_WAITOK|M_ZERO); 634 ifp = sc->sc_ifp = if_alloc(IFT_ETHER); 635 if (ifp == NULL) { 636 free(sc, M_DEVBUF); 637 return (ENOSPC); 638 } 639 640 BRIDGE_LOCK_INIT(sc); 641 sc->sc_brtmax = BRIDGE_RTABLE_MAX; 642 sc->sc_brttimeout = BRIDGE_RTABLE_TIMEOUT; 643 644 /* Initialize our routing table. */ 645 bridge_rtable_init(sc); 646 647 callout_init_mtx(&sc->sc_brcallout, &sc->sc_mtx, 0); 648 649 LIST_INIT(&sc->sc_iflist); 650 LIST_INIT(&sc->sc_spanlist); 651 652 ifp->if_softc = sc; 653 if_initname(ifp, bridge_name, unit); 654 ifp->if_flags = IFF_BROADCAST | IFF_SIMPLEX | IFF_MULTICAST; 655 ifp->if_ioctl = bridge_ioctl; 656 ifp->if_transmit = bridge_transmit; 657 ifp->if_qflush = bridge_qflush; 658 ifp->if_init = bridge_init; 659 ifp->if_type = IFT_BRIDGE; 660 661 /* 662 * Generate an ethernet address with a locally administered address. 663 * 664 * Since we are using random ethernet addresses for the bridge, it is 665 * possible that we might have address collisions, so make sure that 666 * this hardware address isn't already in use on another bridge. 667 * The first try uses the hostid and falls back to arc4rand(). 668 */ 669 fb = 0; 670 getcredhostid(curthread->td_ucred, &hostid); 671 do { 672 if (fb || hostid == 0) { 673 arc4rand(sc->sc_defaddr, ETHER_ADDR_LEN, 1); 674 sc->sc_defaddr[0] &= ~1;/* clear multicast bit */ 675 sc->sc_defaddr[0] |= 2; /* set the LAA bit */ 676 } else { 677 sc->sc_defaddr[0] = 0x2; 678 sc->sc_defaddr[1] = (hostid >> 24) & 0xff; 679 sc->sc_defaddr[2] = (hostid >> 16) & 0xff; 680 sc->sc_defaddr[3] = (hostid >> 8 ) & 0xff; 681 sc->sc_defaddr[4] = hostid & 0xff; 682 sc->sc_defaddr[5] = ifp->if_dunit & 0xff; 683 } 684 685 fb = 1; 686 retry = 0; 687 BRIDGE_LIST_LOCK(); 688 LIST_FOREACH(sc2, &V_bridge_list, sc_list) { 689 bifp = sc2->sc_ifp; 690 if (memcmp(sc->sc_defaddr, 691 IF_LLADDR(bifp), ETHER_ADDR_LEN) == 0) { 692 retry = 1; 693 break; 694 } 695 } 696 BRIDGE_LIST_UNLOCK(); 697 } while (retry == 1); 698 699 bstp_attach(&sc->sc_stp, &bridge_ops); 700 ether_ifattach(ifp, sc->sc_defaddr); 701 /* Now undo some of the damage... */ 702 ifp->if_baudrate = 0; 703 ifp->if_type = IFT_BRIDGE; 704 705 BRIDGE_LIST_LOCK(); 706 LIST_INSERT_HEAD(&V_bridge_list, sc, sc_list); 707 BRIDGE_LIST_UNLOCK(); 708 709 return (0); 710 } 711 712 /* 713 * bridge_clone_destroy: 714 * 715 * Destroy a bridge instance. 716 */ 717 static void 718 bridge_clone_destroy(struct ifnet *ifp) 719 { 720 struct bridge_softc *sc = ifp->if_softc; 721 struct bridge_iflist *bif; 722 723 BRIDGE_LOCK(sc); 724 725 bridge_stop(ifp, 1); 726 ifp->if_flags &= ~IFF_UP; 727 728 while ((bif = LIST_FIRST(&sc->sc_iflist)) != NULL) 729 bridge_delete_member(sc, bif, 0); 730 731 while ((bif = LIST_FIRST(&sc->sc_spanlist)) != NULL) { 732 bridge_delete_span(sc, bif); 733 } 734 735 BRIDGE_UNLOCK(sc); 736 737 callout_drain(&sc->sc_brcallout); 738 739 BRIDGE_LIST_LOCK(); 740 LIST_REMOVE(sc, sc_list); 741 BRIDGE_LIST_UNLOCK(); 742 743 bstp_detach(&sc->sc_stp); 744 ether_ifdetach(ifp); 745 if_free(ifp); 746 747 /* Tear down the routing table. */ 748 bridge_rtable_fini(sc); 749 750 BRIDGE_LOCK_DESTROY(sc); 751 free(sc, M_DEVBUF); 752 } 753 754 /* 755 * bridge_ioctl: 756 * 757 * Handle a control request from the operator. 758 */ 759 static int 760 bridge_ioctl(struct ifnet *ifp, u_long cmd, caddr_t data) 761 { 762 struct bridge_softc *sc = ifp->if_softc; 763 struct ifreq *ifr = (struct ifreq *)data; 764 struct bridge_iflist *bif; 765 struct thread *td = curthread; 766 union { 767 struct ifbreq ifbreq; 768 struct ifbifconf ifbifconf; 769 struct ifbareq ifbareq; 770 struct ifbaconf ifbaconf; 771 struct ifbrparam ifbrparam; 772 struct ifbropreq ifbropreq; 773 } args; 774 struct ifdrv *ifd = (struct ifdrv *) data; 775 const struct bridge_control *bc; 776 int error = 0; 777 778 switch (cmd) { 779 780 case SIOCADDMULTI: 781 case SIOCDELMULTI: 782 break; 783 784 case SIOCGDRVSPEC: 785 case SIOCSDRVSPEC: 786 if (ifd->ifd_cmd >= bridge_control_table_size) { 787 error = EINVAL; 788 break; 789 } 790 bc = &bridge_control_table[ifd->ifd_cmd]; 791 792 if (cmd == SIOCGDRVSPEC && 793 (bc->bc_flags & BC_F_COPYOUT) == 0) { 794 error = EINVAL; 795 break; 796 } 797 else if (cmd == SIOCSDRVSPEC && 798 (bc->bc_flags & BC_F_COPYOUT) != 0) { 799 error = EINVAL; 800 break; 801 } 802 803 if (bc->bc_flags & BC_F_SUSER) { 804 error = priv_check(td, PRIV_NET_BRIDGE); 805 if (error) 806 break; 807 } 808 809 if (ifd->ifd_len != bc->bc_argsize || 810 ifd->ifd_len > sizeof(args)) { 811 error = EINVAL; 812 break; 813 } 814 815 bzero(&args, sizeof(args)); 816 if (bc->bc_flags & BC_F_COPYIN) { 817 error = copyin(ifd->ifd_data, &args, ifd->ifd_len); 818 if (error) 819 break; 820 } 821 822 BRIDGE_LOCK(sc); 823 error = (*bc->bc_func)(sc, &args); 824 BRIDGE_UNLOCK(sc); 825 if (error) 826 break; 827 828 if (bc->bc_flags & BC_F_COPYOUT) 829 error = copyout(&args, ifd->ifd_data, ifd->ifd_len); 830 831 break; 832 833 case SIOCSIFFLAGS: 834 if (!(ifp->if_flags & IFF_UP) && 835 (ifp->if_drv_flags & IFF_DRV_RUNNING)) { 836 /* 837 * If interface is marked down and it is running, 838 * then stop and disable it. 839 */ 840 BRIDGE_LOCK(sc); 841 bridge_stop(ifp, 1); 842 BRIDGE_UNLOCK(sc); 843 } else if ((ifp->if_flags & IFF_UP) && 844 !(ifp->if_drv_flags & IFF_DRV_RUNNING)) { 845 /* 846 * If interface is marked up and it is stopped, then 847 * start it. 848 */ 849 (*ifp->if_init)(sc); 850 } 851 break; 852 853 case SIOCSIFMTU: 854 if (ifr->ifr_mtu < 576) { 855 error = EINVAL; 856 break; 857 } 858 if (LIST_EMPTY(&sc->sc_iflist)) { 859 sc->sc_ifp->if_mtu = ifr->ifr_mtu; 860 break; 861 } 862 BRIDGE_LOCK(sc); 863 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 864 if (bif->bif_ifp->if_mtu != ifr->ifr_mtu) { 865 log(LOG_NOTICE, "%s: invalid MTU: %u(%s)" 866 " != %d\n", sc->sc_ifp->if_xname, 867 bif->bif_ifp->if_mtu, 868 bif->bif_ifp->if_xname, ifr->ifr_mtu); 869 error = EINVAL; 870 break; 871 } 872 } 873 if (!error) 874 sc->sc_ifp->if_mtu = ifr->ifr_mtu; 875 BRIDGE_UNLOCK(sc); 876 break; 877 default: 878 /* 879 * drop the lock as ether_ioctl() will call bridge_start() and 880 * cause the lock to be recursed. 881 */ 882 error = ether_ioctl(ifp, cmd, data); 883 break; 884 } 885 886 return (error); 887 } 888 889 /* 890 * bridge_mutecaps: 891 * 892 * Clear or restore unwanted capabilities on the member interface 893 */ 894 static void 895 bridge_mutecaps(struct bridge_softc *sc) 896 { 897 struct bridge_iflist *bif; 898 int enabled, mask; 899 900 /* Initial bitmask of capabilities to test */ 901 mask = BRIDGE_IFCAPS_MASK; 902 903 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 904 /* Every member must support it or its disabled */ 905 mask &= bif->bif_savedcaps; 906 } 907 908 BRIDGE_XLOCK(sc); 909 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 910 enabled = bif->bif_ifp->if_capenable; 911 enabled &= ~BRIDGE_IFCAPS_STRIP; 912 /* strip off mask bits and enable them again if allowed */ 913 enabled &= ~BRIDGE_IFCAPS_MASK; 914 enabled |= mask; 915 BRIDGE_UNLOCK(sc); 916 bridge_set_ifcap(sc, bif, enabled); 917 BRIDGE_LOCK(sc); 918 } 919 BRIDGE_XDROP(sc); 920 921 } 922 923 static void 924 bridge_set_ifcap(struct bridge_softc *sc, struct bridge_iflist *bif, int set) 925 { 926 struct ifnet *ifp = bif->bif_ifp; 927 struct ifreq ifr; 928 int error; 929 930 BRIDGE_UNLOCK_ASSERT(sc); 931 932 bzero(&ifr, sizeof(ifr)); 933 ifr.ifr_reqcap = set; 934 935 if (ifp->if_capenable != set) { 936 error = (*ifp->if_ioctl)(ifp, SIOCSIFCAP, (caddr_t)&ifr); 937 if (error) 938 if_printf(sc->sc_ifp, 939 "error setting capabilities on %s: %d\n", 940 ifp->if_xname, error); 941 if ((ifp->if_capenable & ~set) != 0) 942 if_printf(sc->sc_ifp, 943 "can't disable some capabilities on %s: 0x%x\n", 944 ifp->if_xname, ifp->if_capenable & ~set); 945 } 946 } 947 948 /* 949 * bridge_lookup_member: 950 * 951 * Lookup a bridge member interface. 952 */ 953 static struct bridge_iflist * 954 bridge_lookup_member(struct bridge_softc *sc, const char *name) 955 { 956 struct bridge_iflist *bif; 957 struct ifnet *ifp; 958 959 BRIDGE_LOCK_ASSERT(sc); 960 961 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 962 ifp = bif->bif_ifp; 963 if (strcmp(ifp->if_xname, name) == 0) 964 return (bif); 965 } 966 967 return (NULL); 968 } 969 970 /* 971 * bridge_lookup_member_if: 972 * 973 * Lookup a bridge member interface by ifnet*. 974 */ 975 static struct bridge_iflist * 976 bridge_lookup_member_if(struct bridge_softc *sc, struct ifnet *member_ifp) 977 { 978 struct bridge_iflist *bif; 979 980 BRIDGE_LOCK_ASSERT(sc); 981 982 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 983 if (bif->bif_ifp == member_ifp) 984 return (bif); 985 } 986 987 return (NULL); 988 } 989 990 /* 991 * bridge_delete_member: 992 * 993 * Delete the specified member interface. 994 */ 995 static void 996 bridge_delete_member(struct bridge_softc *sc, struct bridge_iflist *bif, 997 int gone) 998 { 999 struct ifnet *ifs = bif->bif_ifp; 1000 struct ifnet *fif = NULL; 1001 1002 BRIDGE_LOCK_ASSERT(sc); 1003 1004 if (bif->bif_flags & IFBIF_STP) 1005 bstp_disable(&bif->bif_stp); 1006 1007 ifs->if_bridge = NULL; 1008 BRIDGE_XLOCK(sc); 1009 LIST_REMOVE(bif, bif_next); 1010 BRIDGE_XDROP(sc); 1011 1012 /* 1013 * If removing the interface that gave the bridge its mac address, set 1014 * the mac address of the bridge to the address of the next member, or 1015 * to its default address if no members are left. 1016 */ 1017 if (V_bridge_inherit_mac && sc->sc_ifaddr == ifs) { 1018 if (LIST_EMPTY(&sc->sc_iflist)) { 1019 bcopy(sc->sc_defaddr, 1020 IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN); 1021 sc->sc_ifaddr = NULL; 1022 } else { 1023 fif = LIST_FIRST(&sc->sc_iflist)->bif_ifp; 1024 bcopy(IF_LLADDR(fif), 1025 IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN); 1026 sc->sc_ifaddr = fif; 1027 } 1028 EVENTHANDLER_INVOKE(iflladdr_event, sc->sc_ifp); 1029 } 1030 1031 bridge_linkcheck(sc); 1032 bridge_mutecaps(sc); /* recalcuate now this interface is removed */ 1033 bridge_rtdelete(sc, ifs, IFBF_FLUSHALL); 1034 KASSERT(bif->bif_addrcnt == 0, 1035 ("%s: %d bridge routes referenced", __func__, bif->bif_addrcnt)); 1036 1037 ifs->if_bridge_output = NULL; 1038 ifs->if_bridge_input = NULL; 1039 ifs->if_bridge_linkstate = NULL; 1040 BRIDGE_UNLOCK(sc); 1041 if (!gone) { 1042 switch (ifs->if_type) { 1043 case IFT_ETHER: 1044 case IFT_L2VLAN: 1045 /* 1046 * Take the interface out of promiscuous mode, but only 1047 * if it was promiscuous in the first place. It might 1048 * not be if we're in the bridge_ioctl_add() error path. 1049 */ 1050 if (ifs->if_flags & IFF_PROMISC) 1051 (void) ifpromisc(ifs, 0); 1052 break; 1053 1054 case IFT_GIF: 1055 break; 1056 1057 default: 1058 #ifdef DIAGNOSTIC 1059 panic("bridge_delete_member: impossible"); 1060 #endif 1061 break; 1062 } 1063 /* reneable any interface capabilities */ 1064 bridge_set_ifcap(sc, bif, bif->bif_savedcaps); 1065 } 1066 bstp_destroy(&bif->bif_stp); /* prepare to free */ 1067 BRIDGE_LOCK(sc); 1068 free(bif, M_DEVBUF); 1069 } 1070 1071 /* 1072 * bridge_delete_span: 1073 * 1074 * Delete the specified span interface. 1075 */ 1076 static void 1077 bridge_delete_span(struct bridge_softc *sc, struct bridge_iflist *bif) 1078 { 1079 BRIDGE_LOCK_ASSERT(sc); 1080 1081 KASSERT(bif->bif_ifp->if_bridge == NULL, 1082 ("%s: not a span interface", __func__)); 1083 1084 LIST_REMOVE(bif, bif_next); 1085 free(bif, M_DEVBUF); 1086 } 1087 1088 static int 1089 bridge_ioctl_add(struct bridge_softc *sc, void *arg) 1090 { 1091 struct ifbreq *req = arg; 1092 struct bridge_iflist *bif = NULL; 1093 struct ifnet *ifs; 1094 int error = 0; 1095 1096 ifs = ifunit(req->ifbr_ifsname); 1097 if (ifs == NULL) 1098 return (ENOENT); 1099 if (ifs->if_ioctl == NULL) /* must be supported */ 1100 return (EINVAL); 1101 1102 /* If it's in the span list, it can't be a member. */ 1103 LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1104 if (ifs == bif->bif_ifp) 1105 return (EBUSY); 1106 1107 if (ifs->if_bridge == sc) 1108 return (EEXIST); 1109 1110 if (ifs->if_bridge != NULL) 1111 return (EBUSY); 1112 1113 switch (ifs->if_type) { 1114 case IFT_ETHER: 1115 case IFT_L2VLAN: 1116 case IFT_GIF: 1117 /* permitted interface types */ 1118 break; 1119 default: 1120 return (EINVAL); 1121 } 1122 1123 #ifdef INET6 1124 /* 1125 * Two valid inet6 addresses with link-local scope must not be 1126 * on the parent interface and the member interfaces at the 1127 * same time. This restriction is needed to prevent violation 1128 * of link-local scope zone. Attempts to add a member 1129 * interface which has inet6 addresses when the parent has 1130 * inet6 triggers removal of all inet6 addresses on the member 1131 * interface. 1132 */ 1133 1134 /* Check if the parent interface has a link-local scope addr. */ 1135 if (V_allow_llz_overlap == 0 && 1136 in6ifa_llaonifp(sc->sc_ifp) != NULL) { 1137 /* 1138 * If any, remove all inet6 addresses from the member 1139 * interfaces. 1140 */ 1141 BRIDGE_XLOCK(sc); 1142 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1143 if (in6ifa_llaonifp(bif->bif_ifp)) { 1144 BRIDGE_UNLOCK(sc); 1145 in6_ifdetach(bif->bif_ifp); 1146 BRIDGE_LOCK(sc); 1147 if_printf(sc->sc_ifp, 1148 "IPv6 addresses on %s have been removed " 1149 "before adding it as a member to prevent " 1150 "IPv6 address scope violation.\n", 1151 bif->bif_ifp->if_xname); 1152 } 1153 } 1154 BRIDGE_XDROP(sc); 1155 if (in6ifa_llaonifp(ifs)) { 1156 BRIDGE_UNLOCK(sc); 1157 in6_ifdetach(ifs); 1158 BRIDGE_LOCK(sc); 1159 if_printf(sc->sc_ifp, 1160 "IPv6 addresses on %s have been removed " 1161 "before adding it as a member to prevent " 1162 "IPv6 address scope violation.\n", 1163 ifs->if_xname); 1164 } 1165 } 1166 #endif 1167 /* Allow the first Ethernet member to define the MTU */ 1168 if (LIST_EMPTY(&sc->sc_iflist)) 1169 sc->sc_ifp->if_mtu = ifs->if_mtu; 1170 else if (sc->sc_ifp->if_mtu != ifs->if_mtu) { 1171 if_printf(sc->sc_ifp, "invalid MTU: %u(%s) != %u\n", 1172 ifs->if_mtu, ifs->if_xname, sc->sc_ifp->if_mtu); 1173 return (EINVAL); 1174 } 1175 1176 bif = malloc(sizeof(*bif), M_DEVBUF, M_NOWAIT|M_ZERO); 1177 if (bif == NULL) 1178 return (ENOMEM); 1179 1180 bif->bif_ifp = ifs; 1181 bif->bif_flags = IFBIF_LEARNING | IFBIF_DISCOVER; 1182 bif->bif_savedcaps = ifs->if_capenable; 1183 1184 /* 1185 * Assign the interface's MAC address to the bridge if it's the first 1186 * member and the MAC address of the bridge has not been changed from 1187 * the default randomly generated one. 1188 */ 1189 if (V_bridge_inherit_mac && LIST_EMPTY(&sc->sc_iflist) && 1190 !memcmp(IF_LLADDR(sc->sc_ifp), sc->sc_defaddr, ETHER_ADDR_LEN)) { 1191 bcopy(IF_LLADDR(ifs), IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN); 1192 sc->sc_ifaddr = ifs; 1193 EVENTHANDLER_INVOKE(iflladdr_event, sc->sc_ifp); 1194 } 1195 1196 ifs->if_bridge = sc; 1197 ifs->if_bridge_output = bridge_output; 1198 ifs->if_bridge_input = bridge_input; 1199 ifs->if_bridge_linkstate = bridge_linkstate; 1200 bstp_create(&sc->sc_stp, &bif->bif_stp, bif->bif_ifp); 1201 /* 1202 * XXX: XLOCK HERE!?! 1203 * 1204 * NOTE: insert_***HEAD*** should be safe for the traversals. 1205 */ 1206 LIST_INSERT_HEAD(&sc->sc_iflist, bif, bif_next); 1207 1208 /* Set interface capabilities to the intersection set of all members */ 1209 bridge_mutecaps(sc); 1210 bridge_linkcheck(sc); 1211 1212 /* Place the interface into promiscuous mode */ 1213 switch (ifs->if_type) { 1214 case IFT_ETHER: 1215 case IFT_L2VLAN: 1216 BRIDGE_UNLOCK(sc); 1217 error = ifpromisc(ifs, 1); 1218 BRIDGE_LOCK(sc); 1219 break; 1220 } 1221 1222 if (error) 1223 bridge_delete_member(sc, bif, 0); 1224 return (error); 1225 } 1226 1227 static int 1228 bridge_ioctl_del(struct bridge_softc *sc, void *arg) 1229 { 1230 struct ifbreq *req = arg; 1231 struct bridge_iflist *bif; 1232 1233 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1234 if (bif == NULL) 1235 return (ENOENT); 1236 1237 bridge_delete_member(sc, bif, 0); 1238 1239 return (0); 1240 } 1241 1242 static int 1243 bridge_ioctl_gifflags(struct bridge_softc *sc, void *arg) 1244 { 1245 struct ifbreq *req = arg; 1246 struct bridge_iflist *bif; 1247 struct bstp_port *bp; 1248 1249 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1250 if (bif == NULL) 1251 return (ENOENT); 1252 1253 bp = &bif->bif_stp; 1254 req->ifbr_ifsflags = bif->bif_flags; 1255 req->ifbr_state = bp->bp_state; 1256 req->ifbr_priority = bp->bp_priority; 1257 req->ifbr_path_cost = bp->bp_path_cost; 1258 req->ifbr_portno = bif->bif_ifp->if_index & 0xfff; 1259 req->ifbr_proto = bp->bp_protover; 1260 req->ifbr_role = bp->bp_role; 1261 req->ifbr_stpflags = bp->bp_flags; 1262 req->ifbr_addrcnt = bif->bif_addrcnt; 1263 req->ifbr_addrmax = bif->bif_addrmax; 1264 req->ifbr_addrexceeded = bif->bif_addrexceeded; 1265 1266 /* Copy STP state options as flags */ 1267 if (bp->bp_operedge) 1268 req->ifbr_ifsflags |= IFBIF_BSTP_EDGE; 1269 if (bp->bp_flags & BSTP_PORT_AUTOEDGE) 1270 req->ifbr_ifsflags |= IFBIF_BSTP_AUTOEDGE; 1271 if (bp->bp_ptp_link) 1272 req->ifbr_ifsflags |= IFBIF_BSTP_PTP; 1273 if (bp->bp_flags & BSTP_PORT_AUTOPTP) 1274 req->ifbr_ifsflags |= IFBIF_BSTP_AUTOPTP; 1275 if (bp->bp_flags & BSTP_PORT_ADMEDGE) 1276 req->ifbr_ifsflags |= IFBIF_BSTP_ADMEDGE; 1277 if (bp->bp_flags & BSTP_PORT_ADMCOST) 1278 req->ifbr_ifsflags |= IFBIF_BSTP_ADMCOST; 1279 return (0); 1280 } 1281 1282 static int 1283 bridge_ioctl_sifflags(struct bridge_softc *sc, void *arg) 1284 { 1285 struct ifbreq *req = arg; 1286 struct bridge_iflist *bif; 1287 struct bstp_port *bp; 1288 int error; 1289 1290 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1291 if (bif == NULL) 1292 return (ENOENT); 1293 bp = &bif->bif_stp; 1294 1295 if (req->ifbr_ifsflags & IFBIF_SPAN) 1296 /* SPAN is readonly */ 1297 return (EINVAL); 1298 1299 if (req->ifbr_ifsflags & IFBIF_STP) { 1300 if ((bif->bif_flags & IFBIF_STP) == 0) { 1301 error = bstp_enable(&bif->bif_stp); 1302 if (error) 1303 return (error); 1304 } 1305 } else { 1306 if ((bif->bif_flags & IFBIF_STP) != 0) 1307 bstp_disable(&bif->bif_stp); 1308 } 1309 1310 /* Pass on STP flags */ 1311 bstp_set_edge(bp, req->ifbr_ifsflags & IFBIF_BSTP_EDGE ? 1 : 0); 1312 bstp_set_autoedge(bp, req->ifbr_ifsflags & IFBIF_BSTP_AUTOEDGE ? 1 : 0); 1313 bstp_set_ptp(bp, req->ifbr_ifsflags & IFBIF_BSTP_PTP ? 1 : 0); 1314 bstp_set_autoptp(bp, req->ifbr_ifsflags & IFBIF_BSTP_AUTOPTP ? 1 : 0); 1315 1316 /* Save the bits relating to the bridge */ 1317 bif->bif_flags = req->ifbr_ifsflags & IFBIFMASK; 1318 1319 return (0); 1320 } 1321 1322 static int 1323 bridge_ioctl_scache(struct bridge_softc *sc, void *arg) 1324 { 1325 struct ifbrparam *param = arg; 1326 1327 sc->sc_brtmax = param->ifbrp_csize; 1328 bridge_rttrim(sc); 1329 1330 return (0); 1331 } 1332 1333 static int 1334 bridge_ioctl_gcache(struct bridge_softc *sc, void *arg) 1335 { 1336 struct ifbrparam *param = arg; 1337 1338 param->ifbrp_csize = sc->sc_brtmax; 1339 1340 return (0); 1341 } 1342 1343 static int 1344 bridge_ioctl_gifs(struct bridge_softc *sc, void *arg) 1345 { 1346 struct ifbifconf *bifc = arg; 1347 struct bridge_iflist *bif; 1348 struct ifbreq breq; 1349 char *buf, *outbuf; 1350 int count, buflen, len, error = 0; 1351 1352 count = 0; 1353 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) 1354 count++; 1355 LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1356 count++; 1357 1358 buflen = sizeof(breq) * count; 1359 if (bifc->ifbic_len == 0) { 1360 bifc->ifbic_len = buflen; 1361 return (0); 1362 } 1363 BRIDGE_UNLOCK(sc); 1364 outbuf = malloc(buflen, M_TEMP, M_WAITOK | M_ZERO); 1365 BRIDGE_LOCK(sc); 1366 1367 count = 0; 1368 buf = outbuf; 1369 len = min(bifc->ifbic_len, buflen); 1370 bzero(&breq, sizeof(breq)); 1371 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1372 if (len < sizeof(breq)) 1373 break; 1374 1375 strlcpy(breq.ifbr_ifsname, bif->bif_ifp->if_xname, 1376 sizeof(breq.ifbr_ifsname)); 1377 /* Fill in the ifbreq structure */ 1378 error = bridge_ioctl_gifflags(sc, &breq); 1379 if (error) 1380 break; 1381 memcpy(buf, &breq, sizeof(breq)); 1382 count++; 1383 buf += sizeof(breq); 1384 len -= sizeof(breq); 1385 } 1386 LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) { 1387 if (len < sizeof(breq)) 1388 break; 1389 1390 strlcpy(breq.ifbr_ifsname, bif->bif_ifp->if_xname, 1391 sizeof(breq.ifbr_ifsname)); 1392 breq.ifbr_ifsflags = bif->bif_flags; 1393 breq.ifbr_portno = bif->bif_ifp->if_index & 0xfff; 1394 memcpy(buf, &breq, sizeof(breq)); 1395 count++; 1396 buf += sizeof(breq); 1397 len -= sizeof(breq); 1398 } 1399 1400 BRIDGE_UNLOCK(sc); 1401 bifc->ifbic_len = sizeof(breq) * count; 1402 error = copyout(outbuf, bifc->ifbic_req, bifc->ifbic_len); 1403 BRIDGE_LOCK(sc); 1404 free(outbuf, M_TEMP); 1405 return (error); 1406 } 1407 1408 static int 1409 bridge_ioctl_rts(struct bridge_softc *sc, void *arg) 1410 { 1411 struct ifbaconf *bac = arg; 1412 struct bridge_rtnode *brt; 1413 struct ifbareq bareq; 1414 char *buf, *outbuf; 1415 int count, buflen, len, error = 0; 1416 1417 if (bac->ifbac_len == 0) 1418 return (0); 1419 1420 count = 0; 1421 LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) 1422 count++; 1423 buflen = sizeof(bareq) * count; 1424 1425 BRIDGE_UNLOCK(sc); 1426 outbuf = malloc(buflen, M_TEMP, M_WAITOK | M_ZERO); 1427 BRIDGE_LOCK(sc); 1428 1429 count = 0; 1430 buf = outbuf; 1431 len = min(bac->ifbac_len, buflen); 1432 bzero(&bareq, sizeof(bareq)); 1433 LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) { 1434 if (len < sizeof(bareq)) 1435 goto out; 1436 strlcpy(bareq.ifba_ifsname, brt->brt_ifp->if_xname, 1437 sizeof(bareq.ifba_ifsname)); 1438 memcpy(bareq.ifba_dst, brt->brt_addr, sizeof(brt->brt_addr)); 1439 bareq.ifba_vlan = brt->brt_vlan; 1440 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 1441 time_uptime < brt->brt_expire) 1442 bareq.ifba_expire = brt->brt_expire - time_uptime; 1443 else 1444 bareq.ifba_expire = 0; 1445 bareq.ifba_flags = brt->brt_flags; 1446 1447 memcpy(buf, &bareq, sizeof(bareq)); 1448 count++; 1449 buf += sizeof(bareq); 1450 len -= sizeof(bareq); 1451 } 1452 out: 1453 BRIDGE_UNLOCK(sc); 1454 bac->ifbac_len = sizeof(bareq) * count; 1455 error = copyout(outbuf, bac->ifbac_req, bac->ifbac_len); 1456 BRIDGE_LOCK(sc); 1457 free(outbuf, M_TEMP); 1458 return (error); 1459 } 1460 1461 static int 1462 bridge_ioctl_saddr(struct bridge_softc *sc, void *arg) 1463 { 1464 struct ifbareq *req = arg; 1465 struct bridge_iflist *bif; 1466 int error; 1467 1468 bif = bridge_lookup_member(sc, req->ifba_ifsname); 1469 if (bif == NULL) 1470 return (ENOENT); 1471 1472 error = bridge_rtupdate(sc, req->ifba_dst, req->ifba_vlan, bif, 1, 1473 req->ifba_flags); 1474 1475 return (error); 1476 } 1477 1478 static int 1479 bridge_ioctl_sto(struct bridge_softc *sc, void *arg) 1480 { 1481 struct ifbrparam *param = arg; 1482 1483 sc->sc_brttimeout = param->ifbrp_ctime; 1484 return (0); 1485 } 1486 1487 static int 1488 bridge_ioctl_gto(struct bridge_softc *sc, void *arg) 1489 { 1490 struct ifbrparam *param = arg; 1491 1492 param->ifbrp_ctime = sc->sc_brttimeout; 1493 return (0); 1494 } 1495 1496 static int 1497 bridge_ioctl_daddr(struct bridge_softc *sc, void *arg) 1498 { 1499 struct ifbareq *req = arg; 1500 1501 return (bridge_rtdaddr(sc, req->ifba_dst, req->ifba_vlan)); 1502 } 1503 1504 static int 1505 bridge_ioctl_flush(struct bridge_softc *sc, void *arg) 1506 { 1507 struct ifbreq *req = arg; 1508 1509 bridge_rtflush(sc, req->ifbr_ifsflags); 1510 return (0); 1511 } 1512 1513 static int 1514 bridge_ioctl_gpri(struct bridge_softc *sc, void *arg) 1515 { 1516 struct ifbrparam *param = arg; 1517 struct bstp_state *bs = &sc->sc_stp; 1518 1519 param->ifbrp_prio = bs->bs_bridge_priority; 1520 return (0); 1521 } 1522 1523 static int 1524 bridge_ioctl_spri(struct bridge_softc *sc, void *arg) 1525 { 1526 struct ifbrparam *param = arg; 1527 1528 return (bstp_set_priority(&sc->sc_stp, param->ifbrp_prio)); 1529 } 1530 1531 static int 1532 bridge_ioctl_ght(struct bridge_softc *sc, void *arg) 1533 { 1534 struct ifbrparam *param = arg; 1535 struct bstp_state *bs = &sc->sc_stp; 1536 1537 param->ifbrp_hellotime = bs->bs_bridge_htime >> 8; 1538 return (0); 1539 } 1540 1541 static int 1542 bridge_ioctl_sht(struct bridge_softc *sc, void *arg) 1543 { 1544 struct ifbrparam *param = arg; 1545 1546 return (bstp_set_htime(&sc->sc_stp, param->ifbrp_hellotime)); 1547 } 1548 1549 static int 1550 bridge_ioctl_gfd(struct bridge_softc *sc, void *arg) 1551 { 1552 struct ifbrparam *param = arg; 1553 struct bstp_state *bs = &sc->sc_stp; 1554 1555 param->ifbrp_fwddelay = bs->bs_bridge_fdelay >> 8; 1556 return (0); 1557 } 1558 1559 static int 1560 bridge_ioctl_sfd(struct bridge_softc *sc, void *arg) 1561 { 1562 struct ifbrparam *param = arg; 1563 1564 return (bstp_set_fdelay(&sc->sc_stp, param->ifbrp_fwddelay)); 1565 } 1566 1567 static int 1568 bridge_ioctl_gma(struct bridge_softc *sc, void *arg) 1569 { 1570 struct ifbrparam *param = arg; 1571 struct bstp_state *bs = &sc->sc_stp; 1572 1573 param->ifbrp_maxage = bs->bs_bridge_max_age >> 8; 1574 return (0); 1575 } 1576 1577 static int 1578 bridge_ioctl_sma(struct bridge_softc *sc, void *arg) 1579 { 1580 struct ifbrparam *param = arg; 1581 1582 return (bstp_set_maxage(&sc->sc_stp, param->ifbrp_maxage)); 1583 } 1584 1585 static int 1586 bridge_ioctl_sifprio(struct bridge_softc *sc, void *arg) 1587 { 1588 struct ifbreq *req = arg; 1589 struct bridge_iflist *bif; 1590 1591 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1592 if (bif == NULL) 1593 return (ENOENT); 1594 1595 return (bstp_set_port_priority(&bif->bif_stp, req->ifbr_priority)); 1596 } 1597 1598 static int 1599 bridge_ioctl_sifcost(struct bridge_softc *sc, void *arg) 1600 { 1601 struct ifbreq *req = arg; 1602 struct bridge_iflist *bif; 1603 1604 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1605 if (bif == NULL) 1606 return (ENOENT); 1607 1608 return (bstp_set_path_cost(&bif->bif_stp, req->ifbr_path_cost)); 1609 } 1610 1611 static int 1612 bridge_ioctl_sifmaxaddr(struct bridge_softc *sc, void *arg) 1613 { 1614 struct ifbreq *req = arg; 1615 struct bridge_iflist *bif; 1616 1617 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1618 if (bif == NULL) 1619 return (ENOENT); 1620 1621 bif->bif_addrmax = req->ifbr_addrmax; 1622 return (0); 1623 } 1624 1625 static int 1626 bridge_ioctl_addspan(struct bridge_softc *sc, void *arg) 1627 { 1628 struct ifbreq *req = arg; 1629 struct bridge_iflist *bif = NULL; 1630 struct ifnet *ifs; 1631 1632 ifs = ifunit(req->ifbr_ifsname); 1633 if (ifs == NULL) 1634 return (ENOENT); 1635 1636 LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1637 if (ifs == bif->bif_ifp) 1638 return (EBUSY); 1639 1640 if (ifs->if_bridge != NULL) 1641 return (EBUSY); 1642 1643 switch (ifs->if_type) { 1644 case IFT_ETHER: 1645 case IFT_GIF: 1646 case IFT_L2VLAN: 1647 break; 1648 default: 1649 return (EINVAL); 1650 } 1651 1652 bif = malloc(sizeof(*bif), M_DEVBUF, M_NOWAIT|M_ZERO); 1653 if (bif == NULL) 1654 return (ENOMEM); 1655 1656 bif->bif_ifp = ifs; 1657 bif->bif_flags = IFBIF_SPAN; 1658 1659 LIST_INSERT_HEAD(&sc->sc_spanlist, bif, bif_next); 1660 1661 return (0); 1662 } 1663 1664 static int 1665 bridge_ioctl_delspan(struct bridge_softc *sc, void *arg) 1666 { 1667 struct ifbreq *req = arg; 1668 struct bridge_iflist *bif; 1669 struct ifnet *ifs; 1670 1671 ifs = ifunit(req->ifbr_ifsname); 1672 if (ifs == NULL) 1673 return (ENOENT); 1674 1675 LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1676 if (ifs == bif->bif_ifp) 1677 break; 1678 1679 if (bif == NULL) 1680 return (ENOENT); 1681 1682 bridge_delete_span(sc, bif); 1683 1684 return (0); 1685 } 1686 1687 static int 1688 bridge_ioctl_gbparam(struct bridge_softc *sc, void *arg) 1689 { 1690 struct ifbropreq *req = arg; 1691 struct bstp_state *bs = &sc->sc_stp; 1692 struct bstp_port *root_port; 1693 1694 req->ifbop_maxage = bs->bs_bridge_max_age >> 8; 1695 req->ifbop_hellotime = bs->bs_bridge_htime >> 8; 1696 req->ifbop_fwddelay = bs->bs_bridge_fdelay >> 8; 1697 1698 root_port = bs->bs_root_port; 1699 if (root_port == NULL) 1700 req->ifbop_root_port = 0; 1701 else 1702 req->ifbop_root_port = root_port->bp_ifp->if_index; 1703 1704 req->ifbop_holdcount = bs->bs_txholdcount; 1705 req->ifbop_priority = bs->bs_bridge_priority; 1706 req->ifbop_protocol = bs->bs_protover; 1707 req->ifbop_root_path_cost = bs->bs_root_pv.pv_cost; 1708 req->ifbop_bridgeid = bs->bs_bridge_pv.pv_dbridge_id; 1709 req->ifbop_designated_root = bs->bs_root_pv.pv_root_id; 1710 req->ifbop_designated_bridge = bs->bs_root_pv.pv_dbridge_id; 1711 req->ifbop_last_tc_time.tv_sec = bs->bs_last_tc_time.tv_sec; 1712 req->ifbop_last_tc_time.tv_usec = bs->bs_last_tc_time.tv_usec; 1713 1714 return (0); 1715 } 1716 1717 static int 1718 bridge_ioctl_grte(struct bridge_softc *sc, void *arg) 1719 { 1720 struct ifbrparam *param = arg; 1721 1722 param->ifbrp_cexceeded = sc->sc_brtexceeded; 1723 return (0); 1724 } 1725 1726 static int 1727 bridge_ioctl_gifsstp(struct bridge_softc *sc, void *arg) 1728 { 1729 struct ifbpstpconf *bifstp = arg; 1730 struct bridge_iflist *bif; 1731 struct bstp_port *bp; 1732 struct ifbpstpreq bpreq; 1733 char *buf, *outbuf; 1734 int count, buflen, len, error = 0; 1735 1736 count = 0; 1737 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1738 if ((bif->bif_flags & IFBIF_STP) != 0) 1739 count++; 1740 } 1741 1742 buflen = sizeof(bpreq) * count; 1743 if (bifstp->ifbpstp_len == 0) { 1744 bifstp->ifbpstp_len = buflen; 1745 return (0); 1746 } 1747 1748 BRIDGE_UNLOCK(sc); 1749 outbuf = malloc(buflen, M_TEMP, M_WAITOK | M_ZERO); 1750 BRIDGE_LOCK(sc); 1751 1752 count = 0; 1753 buf = outbuf; 1754 len = min(bifstp->ifbpstp_len, buflen); 1755 bzero(&bpreq, sizeof(bpreq)); 1756 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1757 if (len < sizeof(bpreq)) 1758 break; 1759 1760 if ((bif->bif_flags & IFBIF_STP) == 0) 1761 continue; 1762 1763 bp = &bif->bif_stp; 1764 bpreq.ifbp_portno = bif->bif_ifp->if_index & 0xfff; 1765 bpreq.ifbp_fwd_trans = bp->bp_forward_transitions; 1766 bpreq.ifbp_design_cost = bp->bp_desg_pv.pv_cost; 1767 bpreq.ifbp_design_port = bp->bp_desg_pv.pv_port_id; 1768 bpreq.ifbp_design_bridge = bp->bp_desg_pv.pv_dbridge_id; 1769 bpreq.ifbp_design_root = bp->bp_desg_pv.pv_root_id; 1770 1771 memcpy(buf, &bpreq, sizeof(bpreq)); 1772 count++; 1773 buf += sizeof(bpreq); 1774 len -= sizeof(bpreq); 1775 } 1776 1777 BRIDGE_UNLOCK(sc); 1778 bifstp->ifbpstp_len = sizeof(bpreq) * count; 1779 error = copyout(outbuf, bifstp->ifbpstp_req, bifstp->ifbpstp_len); 1780 BRIDGE_LOCK(sc); 1781 free(outbuf, M_TEMP); 1782 return (error); 1783 } 1784 1785 static int 1786 bridge_ioctl_sproto(struct bridge_softc *sc, void *arg) 1787 { 1788 struct ifbrparam *param = arg; 1789 1790 return (bstp_set_protocol(&sc->sc_stp, param->ifbrp_proto)); 1791 } 1792 1793 static int 1794 bridge_ioctl_stxhc(struct bridge_softc *sc, void *arg) 1795 { 1796 struct ifbrparam *param = arg; 1797 1798 return (bstp_set_holdcount(&sc->sc_stp, param->ifbrp_txhc)); 1799 } 1800 1801 /* 1802 * bridge_ifdetach: 1803 * 1804 * Detach an interface from a bridge. Called when a member 1805 * interface is detaching. 1806 */ 1807 static void 1808 bridge_ifdetach(void *arg __unused, struct ifnet *ifp) 1809 { 1810 struct bridge_softc *sc = ifp->if_bridge; 1811 struct bridge_iflist *bif; 1812 1813 if (ifp->if_flags & IFF_RENAMING) 1814 return; 1815 if (V_bridge_cloner == NULL) { 1816 /* 1817 * This detach handler can be called after 1818 * vnet_bridge_uninit(). Just return in that case. 1819 */ 1820 return; 1821 } 1822 /* Check if the interface is a bridge member */ 1823 if (sc != NULL) { 1824 BRIDGE_LOCK(sc); 1825 1826 bif = bridge_lookup_member_if(sc, ifp); 1827 if (bif != NULL) 1828 bridge_delete_member(sc, bif, 1); 1829 1830 BRIDGE_UNLOCK(sc); 1831 return; 1832 } 1833 1834 /* Check if the interface is a span port */ 1835 BRIDGE_LIST_LOCK(); 1836 LIST_FOREACH(sc, &V_bridge_list, sc_list) { 1837 BRIDGE_LOCK(sc); 1838 LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1839 if (ifp == bif->bif_ifp) { 1840 bridge_delete_span(sc, bif); 1841 break; 1842 } 1843 1844 BRIDGE_UNLOCK(sc); 1845 } 1846 BRIDGE_LIST_UNLOCK(); 1847 } 1848 1849 /* 1850 * bridge_init: 1851 * 1852 * Initialize a bridge interface. 1853 */ 1854 static void 1855 bridge_init(void *xsc) 1856 { 1857 struct bridge_softc *sc = (struct bridge_softc *)xsc; 1858 struct ifnet *ifp = sc->sc_ifp; 1859 1860 if (ifp->if_drv_flags & IFF_DRV_RUNNING) 1861 return; 1862 1863 BRIDGE_LOCK(sc); 1864 callout_reset(&sc->sc_brcallout, bridge_rtable_prune_period * hz, 1865 bridge_timer, sc); 1866 1867 ifp->if_drv_flags |= IFF_DRV_RUNNING; 1868 bstp_init(&sc->sc_stp); /* Initialize Spanning Tree */ 1869 1870 BRIDGE_UNLOCK(sc); 1871 } 1872 1873 /* 1874 * bridge_stop: 1875 * 1876 * Stop the bridge interface. 1877 */ 1878 static void 1879 bridge_stop(struct ifnet *ifp, int disable) 1880 { 1881 struct bridge_softc *sc = ifp->if_softc; 1882 1883 BRIDGE_LOCK_ASSERT(sc); 1884 1885 if ((ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) 1886 return; 1887 1888 callout_stop(&sc->sc_brcallout); 1889 bstp_stop(&sc->sc_stp); 1890 1891 bridge_rtflush(sc, IFBF_FLUSHDYN); 1892 1893 ifp->if_drv_flags &= ~IFF_DRV_RUNNING; 1894 } 1895 1896 /* 1897 * bridge_enqueue: 1898 * 1899 * Enqueue a packet on a bridge member interface. 1900 * 1901 */ 1902 static int 1903 bridge_enqueue(struct bridge_softc *sc, struct ifnet *dst_ifp, struct mbuf *m) 1904 { 1905 int len, err = 0; 1906 short mflags; 1907 struct mbuf *m0; 1908 1909 /* We may be sending a fragment so traverse the mbuf */ 1910 for (; m; m = m0) { 1911 m0 = m->m_nextpkt; 1912 m->m_nextpkt = NULL; 1913 len = m->m_pkthdr.len; 1914 mflags = m->m_flags; 1915 1916 /* 1917 * If underlying interface can not do VLAN tag insertion itself 1918 * then attach a packet tag that holds it. 1919 */ 1920 if ((m->m_flags & M_VLANTAG) && 1921 (dst_ifp->if_capenable & IFCAP_VLAN_HWTAGGING) == 0) { 1922 m = ether_vlanencap(m, m->m_pkthdr.ether_vtag); 1923 if (m == NULL) { 1924 if_printf(dst_ifp, 1925 "unable to prepend VLAN header\n"); 1926 if_inc_counter(dst_ifp, IFCOUNTER_OERRORS, 1); 1927 continue; 1928 } 1929 m->m_flags &= ~M_VLANTAG; 1930 } 1931 1932 M_ASSERTPKTHDR(m); /* We shouldn't transmit mbuf without pkthdr */ 1933 if ((err = dst_ifp->if_transmit(dst_ifp, m))) { 1934 m_freem(m0); 1935 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 1936 break; 1937 } 1938 1939 if_inc_counter(sc->sc_ifp, IFCOUNTER_OPACKETS, 1); 1940 if_inc_counter(sc->sc_ifp, IFCOUNTER_OBYTES, len); 1941 if (mflags & M_MCAST) 1942 if_inc_counter(sc->sc_ifp, IFCOUNTER_OMCASTS, 1); 1943 } 1944 1945 return (err); 1946 } 1947 1948 /* 1949 * bridge_dummynet: 1950 * 1951 * Receive a queued packet from dummynet and pass it on to the output 1952 * interface. 1953 * 1954 * The mbuf has the Ethernet header already attached. 1955 */ 1956 static void 1957 bridge_dummynet(struct mbuf *m, struct ifnet *ifp) 1958 { 1959 struct bridge_softc *sc; 1960 1961 sc = ifp->if_bridge; 1962 1963 /* 1964 * The packet didnt originate from a member interface. This should only 1965 * ever happen if a member interface is removed while packets are 1966 * queued for it. 1967 */ 1968 if (sc == NULL) { 1969 m_freem(m); 1970 return; 1971 } 1972 1973 if (PFIL_HOOKED_OUT(V_inet_pfil_head) 1974 #ifdef INET6 1975 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 1976 #endif 1977 ) { 1978 if (bridge_pfil(&m, sc->sc_ifp, ifp, PFIL_OUT) != 0) 1979 return; 1980 if (m == NULL) 1981 return; 1982 } 1983 1984 bridge_enqueue(sc, ifp, m); 1985 } 1986 1987 /* 1988 * bridge_output: 1989 * 1990 * Send output from a bridge member interface. This 1991 * performs the bridging function for locally originated 1992 * packets. 1993 * 1994 * The mbuf has the Ethernet header already attached. We must 1995 * enqueue or free the mbuf before returning. 1996 */ 1997 static int 1998 bridge_output(struct ifnet *ifp, struct mbuf *m, struct sockaddr *sa, 1999 struct rtentry *rt) 2000 { 2001 struct ether_header *eh; 2002 struct ifnet *dst_if; 2003 struct bridge_softc *sc; 2004 uint16_t vlan; 2005 2006 if (m->m_len < ETHER_HDR_LEN) { 2007 m = m_pullup(m, ETHER_HDR_LEN); 2008 if (m == NULL) 2009 return (0); 2010 } 2011 2012 eh = mtod(m, struct ether_header *); 2013 sc = ifp->if_bridge; 2014 vlan = VLANTAGOF(m); 2015 2016 BRIDGE_LOCK(sc); 2017 2018 /* 2019 * If bridge is down, but the original output interface is up, 2020 * go ahead and send out that interface. Otherwise, the packet 2021 * is dropped below. 2022 */ 2023 if ((sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) { 2024 dst_if = ifp; 2025 goto sendunicast; 2026 } 2027 2028 /* 2029 * If the packet is a multicast, or we don't know a better way to 2030 * get there, send to all interfaces. 2031 */ 2032 if (ETHER_IS_MULTICAST(eh->ether_dhost)) 2033 dst_if = NULL; 2034 else 2035 dst_if = bridge_rtlookup(sc, eh->ether_dhost, vlan); 2036 if (dst_if == NULL) { 2037 struct bridge_iflist *bif; 2038 struct mbuf *mc; 2039 int error = 0, used = 0; 2040 2041 bridge_span(sc, m); 2042 2043 BRIDGE_LOCK2REF(sc, error); 2044 if (error) { 2045 m_freem(m); 2046 return (0); 2047 } 2048 2049 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 2050 dst_if = bif->bif_ifp; 2051 2052 if (dst_if->if_type == IFT_GIF) 2053 continue; 2054 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2055 continue; 2056 2057 /* 2058 * If this is not the original output interface, 2059 * and the interface is participating in spanning 2060 * tree, make sure the port is in a state that 2061 * allows forwarding. 2062 */ 2063 if (dst_if != ifp && (bif->bif_flags & IFBIF_STP) && 2064 bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2065 continue; 2066 2067 if (LIST_NEXT(bif, bif_next) == NULL) { 2068 used = 1; 2069 mc = m; 2070 } else { 2071 mc = m_copypacket(m, M_NOWAIT); 2072 if (mc == NULL) { 2073 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 2074 continue; 2075 } 2076 } 2077 2078 bridge_enqueue(sc, dst_if, mc); 2079 } 2080 if (used == 0) 2081 m_freem(m); 2082 BRIDGE_UNREF(sc); 2083 return (0); 2084 } 2085 2086 sendunicast: 2087 /* 2088 * XXX Spanning tree consideration here? 2089 */ 2090 2091 bridge_span(sc, m); 2092 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) { 2093 m_freem(m); 2094 BRIDGE_UNLOCK(sc); 2095 return (0); 2096 } 2097 2098 BRIDGE_UNLOCK(sc); 2099 bridge_enqueue(sc, dst_if, m); 2100 return (0); 2101 } 2102 2103 /* 2104 * bridge_transmit: 2105 * 2106 * Do output on a bridge. 2107 * 2108 */ 2109 static int 2110 bridge_transmit(struct ifnet *ifp, struct mbuf *m) 2111 { 2112 struct bridge_softc *sc; 2113 struct ether_header *eh; 2114 struct ifnet *dst_if; 2115 int error = 0; 2116 2117 sc = ifp->if_softc; 2118 2119 ETHER_BPF_MTAP(ifp, m); 2120 2121 eh = mtod(m, struct ether_header *); 2122 2123 BRIDGE_LOCK(sc); 2124 if (((m->m_flags & (M_BCAST|M_MCAST)) == 0) && 2125 (dst_if = bridge_rtlookup(sc, eh->ether_dhost, 1)) != NULL) { 2126 BRIDGE_UNLOCK(sc); 2127 error = bridge_enqueue(sc, dst_if, m); 2128 } else 2129 bridge_broadcast(sc, ifp, m, 0); 2130 2131 return (error); 2132 } 2133 2134 /* 2135 * The ifp->if_qflush entry point for if_bridge(4) is no-op. 2136 */ 2137 static void 2138 bridge_qflush(struct ifnet *ifp __unused) 2139 { 2140 } 2141 2142 /* 2143 * bridge_forward: 2144 * 2145 * The forwarding function of the bridge. 2146 * 2147 * NOTE: Releases the lock on return. 2148 */ 2149 static void 2150 bridge_forward(struct bridge_softc *sc, struct bridge_iflist *sbif, 2151 struct mbuf *m) 2152 { 2153 struct bridge_iflist *dbif; 2154 struct ifnet *src_if, *dst_if, *ifp; 2155 struct ether_header *eh; 2156 uint16_t vlan; 2157 uint8_t *dst; 2158 int error; 2159 2160 src_if = m->m_pkthdr.rcvif; 2161 ifp = sc->sc_ifp; 2162 2163 if_inc_counter(ifp, IFCOUNTER_IPACKETS, 1); 2164 if_inc_counter(ifp, IFCOUNTER_IBYTES, m->m_pkthdr.len); 2165 vlan = VLANTAGOF(m); 2166 2167 if ((sbif->bif_flags & IFBIF_STP) && 2168 sbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2169 goto drop; 2170 2171 eh = mtod(m, struct ether_header *); 2172 dst = eh->ether_dhost; 2173 2174 /* If the interface is learning, record the address. */ 2175 if (sbif->bif_flags & IFBIF_LEARNING) { 2176 error = bridge_rtupdate(sc, eh->ether_shost, vlan, 2177 sbif, 0, IFBAF_DYNAMIC); 2178 /* 2179 * If the interface has addresses limits then deny any source 2180 * that is not in the cache. 2181 */ 2182 if (error && sbif->bif_addrmax) 2183 goto drop; 2184 } 2185 2186 if ((sbif->bif_flags & IFBIF_STP) != 0 && 2187 sbif->bif_stp.bp_state == BSTP_IFSTATE_LEARNING) 2188 goto drop; 2189 2190 /* 2191 * At this point, the port either doesn't participate 2192 * in spanning tree or it is in the forwarding state. 2193 */ 2194 2195 /* 2196 * If the packet is unicast, destined for someone on 2197 * "this" side of the bridge, drop it. 2198 */ 2199 if ((m->m_flags & (M_BCAST|M_MCAST)) == 0) { 2200 dst_if = bridge_rtlookup(sc, dst, vlan); 2201 if (src_if == dst_if) 2202 goto drop; 2203 } else { 2204 /* 2205 * Check if its a reserved multicast address, any address 2206 * listed in 802.1D section 7.12.6 may not be forwarded by the 2207 * bridge. 2208 * This is currently 01-80-C2-00-00-00 to 01-80-C2-00-00-0F 2209 */ 2210 if (dst[0] == 0x01 && dst[1] == 0x80 && 2211 dst[2] == 0xc2 && dst[3] == 0x00 && 2212 dst[4] == 0x00 && dst[5] <= 0x0f) 2213 goto drop; 2214 2215 /* ...forward it to all interfaces. */ 2216 if_inc_counter(ifp, IFCOUNTER_IMCASTS, 1); 2217 dst_if = NULL; 2218 } 2219 2220 /* 2221 * If we have a destination interface which is a member of our bridge, 2222 * OR this is a unicast packet, push it through the bpf(4) machinery. 2223 * For broadcast or multicast packets, don't bother because it will 2224 * be reinjected into ether_input. We do this before we pass the packets 2225 * through the pfil(9) framework, as it is possible that pfil(9) will 2226 * drop the packet, or possibly modify it, making it difficult to debug 2227 * firewall issues on the bridge. 2228 */ 2229 if (dst_if != NULL || (m->m_flags & (M_BCAST | M_MCAST)) == 0) 2230 ETHER_BPF_MTAP(ifp, m); 2231 2232 /* run the packet filter */ 2233 if (PFIL_HOOKED_IN(V_inet_pfil_head) 2234 #ifdef INET6 2235 || PFIL_HOOKED_IN(V_inet6_pfil_head) 2236 #endif 2237 ) { 2238 BRIDGE_UNLOCK(sc); 2239 if (bridge_pfil(&m, ifp, src_if, PFIL_IN) != 0) 2240 return; 2241 if (m == NULL) 2242 return; 2243 BRIDGE_LOCK(sc); 2244 } 2245 2246 if (dst_if == NULL) { 2247 bridge_broadcast(sc, src_if, m, 1); 2248 return; 2249 } 2250 2251 /* 2252 * At this point, we're dealing with a unicast frame 2253 * going to a different interface. 2254 */ 2255 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2256 goto drop; 2257 2258 dbif = bridge_lookup_member_if(sc, dst_if); 2259 if (dbif == NULL) 2260 /* Not a member of the bridge (anymore?) */ 2261 goto drop; 2262 2263 /* Private segments can not talk to each other */ 2264 if (sbif->bif_flags & dbif->bif_flags & IFBIF_PRIVATE) 2265 goto drop; 2266 2267 if ((dbif->bif_flags & IFBIF_STP) && 2268 dbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2269 goto drop; 2270 2271 BRIDGE_UNLOCK(sc); 2272 2273 if (PFIL_HOOKED_OUT(V_inet_pfil_head) 2274 #ifdef INET6 2275 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 2276 #endif 2277 ) { 2278 if (bridge_pfil(&m, ifp, dst_if, PFIL_OUT) != 0) 2279 return; 2280 if (m == NULL) 2281 return; 2282 } 2283 2284 bridge_enqueue(sc, dst_if, m); 2285 return; 2286 2287 drop: 2288 BRIDGE_UNLOCK(sc); 2289 m_freem(m); 2290 } 2291 2292 /* 2293 * bridge_input: 2294 * 2295 * Receive input from a member interface. Queue the packet for 2296 * bridging if it is not for us. 2297 */ 2298 static struct mbuf * 2299 bridge_input(struct ifnet *ifp, struct mbuf *m) 2300 { 2301 struct bridge_softc *sc = ifp->if_bridge; 2302 struct bridge_iflist *bif, *bif2; 2303 struct ifnet *bifp; 2304 struct ether_header *eh; 2305 struct mbuf *mc, *mc2; 2306 uint16_t vlan; 2307 int error; 2308 2309 if ((sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) 2310 return (m); 2311 2312 bifp = sc->sc_ifp; 2313 vlan = VLANTAGOF(m); 2314 2315 /* 2316 * Implement support for bridge monitoring. If this flag has been 2317 * set on this interface, discard the packet once we push it through 2318 * the bpf(4) machinery, but before we do, increment the byte and 2319 * packet counters associated with this interface. 2320 */ 2321 if ((bifp->if_flags & IFF_MONITOR) != 0) { 2322 m->m_pkthdr.rcvif = bifp; 2323 ETHER_BPF_MTAP(bifp, m); 2324 if_inc_counter(bifp, IFCOUNTER_IPACKETS, 1); 2325 if_inc_counter(bifp, IFCOUNTER_IBYTES, m->m_pkthdr.len); 2326 m_freem(m); 2327 return (NULL); 2328 } 2329 BRIDGE_LOCK(sc); 2330 bif = bridge_lookup_member_if(sc, ifp); 2331 if (bif == NULL) { 2332 BRIDGE_UNLOCK(sc); 2333 return (m); 2334 } 2335 2336 eh = mtod(m, struct ether_header *); 2337 2338 bridge_span(sc, m); 2339 2340 if (m->m_flags & (M_BCAST|M_MCAST)) { 2341 /* Tap off 802.1D packets; they do not get forwarded. */ 2342 if (memcmp(eh->ether_dhost, bstp_etheraddr, 2343 ETHER_ADDR_LEN) == 0) { 2344 bstp_input(&bif->bif_stp, ifp, m); /* consumes mbuf */ 2345 BRIDGE_UNLOCK(sc); 2346 return (NULL); 2347 } 2348 2349 if ((bif->bif_flags & IFBIF_STP) && 2350 bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) { 2351 BRIDGE_UNLOCK(sc); 2352 return (m); 2353 } 2354 2355 /* 2356 * Make a deep copy of the packet and enqueue the copy 2357 * for bridge processing; return the original packet for 2358 * local processing. 2359 */ 2360 mc = m_dup(m, M_NOWAIT); 2361 if (mc == NULL) { 2362 BRIDGE_UNLOCK(sc); 2363 return (m); 2364 } 2365 2366 /* Perform the bridge forwarding function with the copy. */ 2367 bridge_forward(sc, bif, mc); 2368 2369 /* 2370 * Reinject the mbuf as arriving on the bridge so we have a 2371 * chance at claiming multicast packets. We can not loop back 2372 * here from ether_input as a bridge is never a member of a 2373 * bridge. 2374 */ 2375 KASSERT(bifp->if_bridge == NULL, 2376 ("loop created in bridge_input")); 2377 mc2 = m_dup(m, M_NOWAIT); 2378 if (mc2 != NULL) { 2379 /* Keep the layer3 header aligned */ 2380 int i = min(mc2->m_pkthdr.len, max_protohdr); 2381 mc2 = m_copyup(mc2, i, ETHER_ALIGN); 2382 } 2383 if (mc2 != NULL) { 2384 mc2->m_pkthdr.rcvif = bifp; 2385 (*bifp->if_input)(bifp, mc2); 2386 } 2387 2388 /* Return the original packet for local processing. */ 2389 return (m); 2390 } 2391 2392 if ((bif->bif_flags & IFBIF_STP) && 2393 bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) { 2394 BRIDGE_UNLOCK(sc); 2395 return (m); 2396 } 2397 2398 #if (defined(INET) || defined(INET6)) 2399 # define OR_CARP_CHECK_WE_ARE_DST(iface) \ 2400 || ((iface)->if_carp \ 2401 && (*carp_forus_p)((iface), eh->ether_dhost)) 2402 # define OR_CARP_CHECK_WE_ARE_SRC(iface) \ 2403 || ((iface)->if_carp \ 2404 && (*carp_forus_p)((iface), eh->ether_shost)) 2405 #else 2406 # define OR_CARP_CHECK_WE_ARE_DST(iface) 2407 # define OR_CARP_CHECK_WE_ARE_SRC(iface) 2408 #endif 2409 2410 #ifdef INET6 2411 # define OR_PFIL_HOOKED_INET6 \ 2412 || PFIL_HOOKED_IN(V_inet6_pfil_head) 2413 #else 2414 # define OR_PFIL_HOOKED_INET6 2415 #endif 2416 2417 #define GRAB_OUR_PACKETS(iface) \ 2418 if ((iface)->if_type == IFT_GIF) \ 2419 continue; \ 2420 /* It is destined for us. */ \ 2421 if (memcmp(IF_LLADDR((iface)), eh->ether_dhost, ETHER_ADDR_LEN) == 0 \ 2422 OR_CARP_CHECK_WE_ARE_DST((iface)) \ 2423 ) { \ 2424 if ((iface)->if_type == IFT_BRIDGE) { \ 2425 ETHER_BPF_MTAP(iface, m); \ 2426 if_inc_counter(iface, IFCOUNTER_IPACKETS, 1); \ 2427 if_inc_counter(iface, IFCOUNTER_IBYTES, m->m_pkthdr.len); \ 2428 /* Filter on the physical interface. */ \ 2429 if (V_pfil_local_phys && \ 2430 (PFIL_HOOKED_IN(V_inet_pfil_head) \ 2431 OR_PFIL_HOOKED_INET6)) { \ 2432 if (bridge_pfil(&m, NULL, ifp, \ 2433 PFIL_IN) != 0 || m == NULL) { \ 2434 BRIDGE_UNLOCK(sc); \ 2435 return (NULL); \ 2436 } \ 2437 eh = mtod(m, struct ether_header *); \ 2438 } \ 2439 } \ 2440 if (bif->bif_flags & IFBIF_LEARNING) { \ 2441 error = bridge_rtupdate(sc, eh->ether_shost, \ 2442 vlan, bif, 0, IFBAF_DYNAMIC); \ 2443 if (error && bif->bif_addrmax) { \ 2444 BRIDGE_UNLOCK(sc); \ 2445 m_freem(m); \ 2446 return (NULL); \ 2447 } \ 2448 } \ 2449 m->m_pkthdr.rcvif = iface; \ 2450 BRIDGE_UNLOCK(sc); \ 2451 return (m); \ 2452 } \ 2453 \ 2454 /* We just received a packet that we sent out. */ \ 2455 if (memcmp(IF_LLADDR((iface)), eh->ether_shost, ETHER_ADDR_LEN) == 0 \ 2456 OR_CARP_CHECK_WE_ARE_SRC((iface)) \ 2457 ) { \ 2458 BRIDGE_UNLOCK(sc); \ 2459 m_freem(m); \ 2460 return (NULL); \ 2461 } 2462 2463 /* 2464 * Unicast. Make sure it's not for the bridge. 2465 */ 2466 do { GRAB_OUR_PACKETS(bifp) } while (0); 2467 2468 /* 2469 * Give a chance for ifp at first priority. This will help when the 2470 * packet comes through the interface like VLAN's with the same MACs 2471 * on several interfaces from the same bridge. This also will save 2472 * some CPU cycles in case the destination interface and the input 2473 * interface (eq ifp) are the same. 2474 */ 2475 do { GRAB_OUR_PACKETS(ifp) } while (0); 2476 2477 /* Now check the all bridge members. */ 2478 LIST_FOREACH(bif2, &sc->sc_iflist, bif_next) { 2479 GRAB_OUR_PACKETS(bif2->bif_ifp) 2480 } 2481 2482 #undef OR_CARP_CHECK_WE_ARE_DST 2483 #undef OR_CARP_CHECK_WE_ARE_SRC 2484 #undef OR_PFIL_HOOKED_INET6 2485 #undef GRAB_OUR_PACKETS 2486 2487 /* Perform the bridge forwarding function. */ 2488 bridge_forward(sc, bif, m); 2489 2490 return (NULL); 2491 } 2492 2493 /* 2494 * bridge_broadcast: 2495 * 2496 * Send a frame to all interfaces that are members of 2497 * the bridge, except for the one on which the packet 2498 * arrived. 2499 * 2500 * NOTE: Releases the lock on return. 2501 */ 2502 static void 2503 bridge_broadcast(struct bridge_softc *sc, struct ifnet *src_if, 2504 struct mbuf *m, int runfilt) 2505 { 2506 struct bridge_iflist *dbif, *sbif; 2507 struct mbuf *mc; 2508 struct ifnet *dst_if; 2509 int error = 0, used = 0, i; 2510 2511 sbif = bridge_lookup_member_if(sc, src_if); 2512 2513 BRIDGE_LOCK2REF(sc, error); 2514 if (error) { 2515 m_freem(m); 2516 return; 2517 } 2518 2519 /* Filter on the bridge interface before broadcasting */ 2520 if (runfilt && (PFIL_HOOKED_OUT(V_inet_pfil_head) 2521 #ifdef INET6 2522 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 2523 #endif 2524 )) { 2525 if (bridge_pfil(&m, sc->sc_ifp, NULL, PFIL_OUT) != 0) 2526 goto out; 2527 if (m == NULL) 2528 goto out; 2529 } 2530 2531 LIST_FOREACH(dbif, &sc->sc_iflist, bif_next) { 2532 dst_if = dbif->bif_ifp; 2533 if (dst_if == src_if) 2534 continue; 2535 2536 /* Private segments can not talk to each other */ 2537 if (sbif && (sbif->bif_flags & dbif->bif_flags & IFBIF_PRIVATE)) 2538 continue; 2539 2540 if ((dbif->bif_flags & IFBIF_STP) && 2541 dbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2542 continue; 2543 2544 if ((dbif->bif_flags & IFBIF_DISCOVER) == 0 && 2545 (m->m_flags & (M_BCAST|M_MCAST)) == 0) 2546 continue; 2547 2548 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2549 continue; 2550 2551 if (LIST_NEXT(dbif, bif_next) == NULL) { 2552 mc = m; 2553 used = 1; 2554 } else { 2555 mc = m_dup(m, M_NOWAIT); 2556 if (mc == NULL) { 2557 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 2558 continue; 2559 } 2560 } 2561 2562 /* 2563 * Filter on the output interface. Pass a NULL bridge interface 2564 * pointer so we do not redundantly filter on the bridge for 2565 * each interface we broadcast on. 2566 */ 2567 if (runfilt && (PFIL_HOOKED_OUT(V_inet_pfil_head) 2568 #ifdef INET6 2569 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 2570 #endif 2571 )) { 2572 if (used == 0) { 2573 /* Keep the layer3 header aligned */ 2574 i = min(mc->m_pkthdr.len, max_protohdr); 2575 mc = m_copyup(mc, i, ETHER_ALIGN); 2576 if (mc == NULL) { 2577 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 2578 continue; 2579 } 2580 } 2581 if (bridge_pfil(&mc, NULL, dst_if, PFIL_OUT) != 0) 2582 continue; 2583 if (mc == NULL) 2584 continue; 2585 } 2586 2587 bridge_enqueue(sc, dst_if, mc); 2588 } 2589 if (used == 0) 2590 m_freem(m); 2591 2592 out: 2593 BRIDGE_UNREF(sc); 2594 } 2595 2596 /* 2597 * bridge_span: 2598 * 2599 * Duplicate a packet out one or more interfaces that are in span mode, 2600 * the original mbuf is unmodified. 2601 */ 2602 static void 2603 bridge_span(struct bridge_softc *sc, struct mbuf *m) 2604 { 2605 struct bridge_iflist *bif; 2606 struct ifnet *dst_if; 2607 struct mbuf *mc; 2608 2609 if (LIST_EMPTY(&sc->sc_spanlist)) 2610 return; 2611 2612 LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) { 2613 dst_if = bif->bif_ifp; 2614 2615 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2616 continue; 2617 2618 mc = m_copypacket(m, M_NOWAIT); 2619 if (mc == NULL) { 2620 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 2621 continue; 2622 } 2623 2624 bridge_enqueue(sc, dst_if, mc); 2625 } 2626 } 2627 2628 /* 2629 * bridge_rtupdate: 2630 * 2631 * Add a bridge routing entry. 2632 */ 2633 static int 2634 bridge_rtupdate(struct bridge_softc *sc, const uint8_t *dst, uint16_t vlan, 2635 struct bridge_iflist *bif, int setflags, uint8_t flags) 2636 { 2637 struct bridge_rtnode *brt; 2638 int error; 2639 2640 BRIDGE_LOCK_ASSERT(sc); 2641 2642 /* Check the source address is valid and not multicast. */ 2643 if (ETHER_IS_MULTICAST(dst) || 2644 (dst[0] == 0 && dst[1] == 0 && dst[2] == 0 && 2645 dst[3] == 0 && dst[4] == 0 && dst[5] == 0) != 0) 2646 return (EINVAL); 2647 2648 /* 802.1p frames map to vlan 1 */ 2649 if (vlan == 0) 2650 vlan = 1; 2651 2652 /* 2653 * A route for this destination might already exist. If so, 2654 * update it, otherwise create a new one. 2655 */ 2656 if ((brt = bridge_rtnode_lookup(sc, dst, vlan)) == NULL) { 2657 if (sc->sc_brtcnt >= sc->sc_brtmax) { 2658 sc->sc_brtexceeded++; 2659 return (ENOSPC); 2660 } 2661 /* Check per interface address limits (if enabled) */ 2662 if (bif->bif_addrmax && bif->bif_addrcnt >= bif->bif_addrmax) { 2663 bif->bif_addrexceeded++; 2664 return (ENOSPC); 2665 } 2666 2667 /* 2668 * Allocate a new bridge forwarding node, and 2669 * initialize the expiration time and Ethernet 2670 * address. 2671 */ 2672 brt = uma_zalloc(bridge_rtnode_zone, M_NOWAIT | M_ZERO); 2673 if (brt == NULL) 2674 return (ENOMEM); 2675 2676 if (bif->bif_flags & IFBIF_STICKY) 2677 brt->brt_flags = IFBAF_STICKY; 2678 else 2679 brt->brt_flags = IFBAF_DYNAMIC; 2680 2681 memcpy(brt->brt_addr, dst, ETHER_ADDR_LEN); 2682 brt->brt_vlan = vlan; 2683 2684 if ((error = bridge_rtnode_insert(sc, brt)) != 0) { 2685 uma_zfree(bridge_rtnode_zone, brt); 2686 return (error); 2687 } 2688 brt->brt_dst = bif; 2689 bif->bif_addrcnt++; 2690 } 2691 2692 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 2693 brt->brt_dst != bif) { 2694 brt->brt_dst->bif_addrcnt--; 2695 brt->brt_dst = bif; 2696 brt->brt_dst->bif_addrcnt++; 2697 } 2698 2699 if ((flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) 2700 brt->brt_expire = time_uptime + sc->sc_brttimeout; 2701 if (setflags) 2702 brt->brt_flags = flags; 2703 2704 return (0); 2705 } 2706 2707 /* 2708 * bridge_rtlookup: 2709 * 2710 * Lookup the destination interface for an address. 2711 */ 2712 static struct ifnet * 2713 bridge_rtlookup(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan) 2714 { 2715 struct bridge_rtnode *brt; 2716 2717 BRIDGE_LOCK_ASSERT(sc); 2718 2719 if ((brt = bridge_rtnode_lookup(sc, addr, vlan)) == NULL) 2720 return (NULL); 2721 2722 return (brt->brt_ifp); 2723 } 2724 2725 /* 2726 * bridge_rttrim: 2727 * 2728 * Trim the routine table so that we have a number 2729 * of routing entries less than or equal to the 2730 * maximum number. 2731 */ 2732 static void 2733 bridge_rttrim(struct bridge_softc *sc) 2734 { 2735 struct bridge_rtnode *brt, *nbrt; 2736 2737 BRIDGE_LOCK_ASSERT(sc); 2738 2739 /* Make sure we actually need to do this. */ 2740 if (sc->sc_brtcnt <= sc->sc_brtmax) 2741 return; 2742 2743 /* Force an aging cycle; this might trim enough addresses. */ 2744 bridge_rtage(sc); 2745 if (sc->sc_brtcnt <= sc->sc_brtmax) 2746 return; 2747 2748 LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2749 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) { 2750 bridge_rtnode_destroy(sc, brt); 2751 if (sc->sc_brtcnt <= sc->sc_brtmax) 2752 return; 2753 } 2754 } 2755 } 2756 2757 /* 2758 * bridge_timer: 2759 * 2760 * Aging timer for the bridge. 2761 */ 2762 static void 2763 bridge_timer(void *arg) 2764 { 2765 struct bridge_softc *sc = arg; 2766 2767 BRIDGE_LOCK_ASSERT(sc); 2768 2769 bridge_rtage(sc); 2770 2771 if (sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) 2772 callout_reset(&sc->sc_brcallout, 2773 bridge_rtable_prune_period * hz, bridge_timer, sc); 2774 } 2775 2776 /* 2777 * bridge_rtage: 2778 * 2779 * Perform an aging cycle. 2780 */ 2781 static void 2782 bridge_rtage(struct bridge_softc *sc) 2783 { 2784 struct bridge_rtnode *brt, *nbrt; 2785 2786 BRIDGE_LOCK_ASSERT(sc); 2787 2788 LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2789 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) { 2790 if (time_uptime >= brt->brt_expire) 2791 bridge_rtnode_destroy(sc, brt); 2792 } 2793 } 2794 } 2795 2796 /* 2797 * bridge_rtflush: 2798 * 2799 * Remove all dynamic addresses from the bridge. 2800 */ 2801 static void 2802 bridge_rtflush(struct bridge_softc *sc, int full) 2803 { 2804 struct bridge_rtnode *brt, *nbrt; 2805 2806 BRIDGE_LOCK_ASSERT(sc); 2807 2808 LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2809 if (full || (brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) 2810 bridge_rtnode_destroy(sc, brt); 2811 } 2812 } 2813 2814 /* 2815 * bridge_rtdaddr: 2816 * 2817 * Remove an address from the table. 2818 */ 2819 static int 2820 bridge_rtdaddr(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan) 2821 { 2822 struct bridge_rtnode *brt; 2823 int found = 0; 2824 2825 BRIDGE_LOCK_ASSERT(sc); 2826 2827 /* 2828 * If vlan is zero then we want to delete for all vlans so the lookup 2829 * may return more than one. 2830 */ 2831 while ((brt = bridge_rtnode_lookup(sc, addr, vlan)) != NULL) { 2832 bridge_rtnode_destroy(sc, brt); 2833 found = 1; 2834 } 2835 2836 return (found ? 0 : ENOENT); 2837 } 2838 2839 /* 2840 * bridge_rtdelete: 2841 * 2842 * Delete routes to a speicifc member interface. 2843 */ 2844 static void 2845 bridge_rtdelete(struct bridge_softc *sc, struct ifnet *ifp, int full) 2846 { 2847 struct bridge_rtnode *brt, *nbrt; 2848 2849 BRIDGE_LOCK_ASSERT(sc); 2850 2851 LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2852 if (brt->brt_ifp == ifp && (full || 2853 (brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC)) 2854 bridge_rtnode_destroy(sc, brt); 2855 } 2856 } 2857 2858 /* 2859 * bridge_rtable_init: 2860 * 2861 * Initialize the route table for this bridge. 2862 */ 2863 static void 2864 bridge_rtable_init(struct bridge_softc *sc) 2865 { 2866 int i; 2867 2868 sc->sc_rthash = malloc(sizeof(*sc->sc_rthash) * BRIDGE_RTHASH_SIZE, 2869 M_DEVBUF, M_WAITOK); 2870 2871 for (i = 0; i < BRIDGE_RTHASH_SIZE; i++) 2872 LIST_INIT(&sc->sc_rthash[i]); 2873 2874 sc->sc_rthash_key = arc4random(); 2875 LIST_INIT(&sc->sc_rtlist); 2876 } 2877 2878 /* 2879 * bridge_rtable_fini: 2880 * 2881 * Deconstruct the route table for this bridge. 2882 */ 2883 static void 2884 bridge_rtable_fini(struct bridge_softc *sc) 2885 { 2886 2887 KASSERT(sc->sc_brtcnt == 0, 2888 ("%s: %d bridge routes referenced", __func__, sc->sc_brtcnt)); 2889 free(sc->sc_rthash, M_DEVBUF); 2890 } 2891 2892 /* 2893 * The following hash function is adapted from "Hash Functions" by Bob Jenkins 2894 * ("Algorithm Alley", Dr. Dobbs Journal, September 1997). 2895 */ 2896 #define mix(a, b, c) \ 2897 do { \ 2898 a -= b; a -= c; a ^= (c >> 13); \ 2899 b -= c; b -= a; b ^= (a << 8); \ 2900 c -= a; c -= b; c ^= (b >> 13); \ 2901 a -= b; a -= c; a ^= (c >> 12); \ 2902 b -= c; b -= a; b ^= (a << 16); \ 2903 c -= a; c -= b; c ^= (b >> 5); \ 2904 a -= b; a -= c; a ^= (c >> 3); \ 2905 b -= c; b -= a; b ^= (a << 10); \ 2906 c -= a; c -= b; c ^= (b >> 15); \ 2907 } while (/*CONSTCOND*/0) 2908 2909 static __inline uint32_t 2910 bridge_rthash(struct bridge_softc *sc, const uint8_t *addr) 2911 { 2912 uint32_t a = 0x9e3779b9, b = 0x9e3779b9, c = sc->sc_rthash_key; 2913 2914 b += addr[5] << 8; 2915 b += addr[4]; 2916 a += addr[3] << 24; 2917 a += addr[2] << 16; 2918 a += addr[1] << 8; 2919 a += addr[0]; 2920 2921 mix(a, b, c); 2922 2923 return (c & BRIDGE_RTHASH_MASK); 2924 } 2925 2926 #undef mix 2927 2928 static int 2929 bridge_rtnode_addr_cmp(const uint8_t *a, const uint8_t *b) 2930 { 2931 int i, d; 2932 2933 for (i = 0, d = 0; i < ETHER_ADDR_LEN && d == 0; i++) { 2934 d = ((int)a[i]) - ((int)b[i]); 2935 } 2936 2937 return (d); 2938 } 2939 2940 /* 2941 * bridge_rtnode_lookup: 2942 * 2943 * Look up a bridge route node for the specified destination. Compare the 2944 * vlan id or if zero then just return the first match. 2945 */ 2946 static struct bridge_rtnode * 2947 bridge_rtnode_lookup(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan) 2948 { 2949 struct bridge_rtnode *brt; 2950 uint32_t hash; 2951 int dir; 2952 2953 BRIDGE_LOCK_ASSERT(sc); 2954 2955 hash = bridge_rthash(sc, addr); 2956 LIST_FOREACH(brt, &sc->sc_rthash[hash], brt_hash) { 2957 dir = bridge_rtnode_addr_cmp(addr, brt->brt_addr); 2958 if (dir == 0 && (brt->brt_vlan == vlan || vlan == 0)) 2959 return (brt); 2960 if (dir > 0) 2961 return (NULL); 2962 } 2963 2964 return (NULL); 2965 } 2966 2967 /* 2968 * bridge_rtnode_insert: 2969 * 2970 * Insert the specified bridge node into the route table. We 2971 * assume the entry is not already in the table. 2972 */ 2973 static int 2974 bridge_rtnode_insert(struct bridge_softc *sc, struct bridge_rtnode *brt) 2975 { 2976 struct bridge_rtnode *lbrt; 2977 uint32_t hash; 2978 int dir; 2979 2980 BRIDGE_LOCK_ASSERT(sc); 2981 2982 hash = bridge_rthash(sc, brt->brt_addr); 2983 2984 lbrt = LIST_FIRST(&sc->sc_rthash[hash]); 2985 if (lbrt == NULL) { 2986 LIST_INSERT_HEAD(&sc->sc_rthash[hash], brt, brt_hash); 2987 goto out; 2988 } 2989 2990 do { 2991 dir = bridge_rtnode_addr_cmp(brt->brt_addr, lbrt->brt_addr); 2992 if (dir == 0 && brt->brt_vlan == lbrt->brt_vlan) 2993 return (EEXIST); 2994 if (dir > 0) { 2995 LIST_INSERT_BEFORE(lbrt, brt, brt_hash); 2996 goto out; 2997 } 2998 if (LIST_NEXT(lbrt, brt_hash) == NULL) { 2999 LIST_INSERT_AFTER(lbrt, brt, brt_hash); 3000 goto out; 3001 } 3002 lbrt = LIST_NEXT(lbrt, brt_hash); 3003 } while (lbrt != NULL); 3004 3005 #ifdef DIAGNOSTIC 3006 panic("bridge_rtnode_insert: impossible"); 3007 #endif 3008 3009 out: 3010 LIST_INSERT_HEAD(&sc->sc_rtlist, brt, brt_list); 3011 sc->sc_brtcnt++; 3012 3013 return (0); 3014 } 3015 3016 /* 3017 * bridge_rtnode_destroy: 3018 * 3019 * Destroy a bridge rtnode. 3020 */ 3021 static void 3022 bridge_rtnode_destroy(struct bridge_softc *sc, struct bridge_rtnode *brt) 3023 { 3024 BRIDGE_LOCK_ASSERT(sc); 3025 3026 LIST_REMOVE(brt, brt_hash); 3027 3028 LIST_REMOVE(brt, brt_list); 3029 sc->sc_brtcnt--; 3030 brt->brt_dst->bif_addrcnt--; 3031 uma_zfree(bridge_rtnode_zone, brt); 3032 } 3033 3034 /* 3035 * bridge_rtable_expire: 3036 * 3037 * Set the expiry time for all routes on an interface. 3038 */ 3039 static void 3040 bridge_rtable_expire(struct ifnet *ifp, int age) 3041 { 3042 struct bridge_softc *sc = ifp->if_bridge; 3043 struct bridge_rtnode *brt; 3044 3045 BRIDGE_LOCK(sc); 3046 3047 /* 3048 * If the age is zero then flush, otherwise set all the expiry times to 3049 * age for the interface 3050 */ 3051 if (age == 0) 3052 bridge_rtdelete(sc, ifp, IFBF_FLUSHDYN); 3053 else { 3054 LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) { 3055 /* Cap the expiry time to 'age' */ 3056 if (brt->brt_ifp == ifp && 3057 brt->brt_expire > time_uptime + age && 3058 (brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) 3059 brt->brt_expire = time_uptime + age; 3060 } 3061 } 3062 BRIDGE_UNLOCK(sc); 3063 } 3064 3065 /* 3066 * bridge_state_change: 3067 * 3068 * Callback from the bridgestp code when a port changes states. 3069 */ 3070 static void 3071 bridge_state_change(struct ifnet *ifp, int state) 3072 { 3073 struct bridge_softc *sc = ifp->if_bridge; 3074 static const char *stpstates[] = { 3075 "disabled", 3076 "listening", 3077 "learning", 3078 "forwarding", 3079 "blocking", 3080 "discarding" 3081 }; 3082 3083 CURVNET_SET(ifp->if_vnet); 3084 if (V_log_stp) 3085 log(LOG_NOTICE, "%s: state changed to %s on %s\n", 3086 sc->sc_ifp->if_xname, stpstates[state], ifp->if_xname); 3087 CURVNET_RESTORE(); 3088 } 3089 3090 /* 3091 * Send bridge packets through pfil if they are one of the types pfil can deal 3092 * with, or if they are ARP or REVARP. (pfil will pass ARP and REVARP without 3093 * question.) If *bifp or *ifp are NULL then packet filtering is skipped for 3094 * that interface. 3095 */ 3096 static int 3097 bridge_pfil(struct mbuf **mp, struct ifnet *bifp, struct ifnet *ifp, int dir) 3098 { 3099 int snap, error, i, hlen; 3100 struct ether_header *eh1, eh2; 3101 struct ip *ip; 3102 struct llc llc1; 3103 u_int16_t ether_type; 3104 pfil_return_t rv; 3105 3106 snap = 0; 3107 error = -1; /* Default error if not error == 0 */ 3108 3109 #if 0 3110 /* we may return with the IP fields swapped, ensure its not shared */ 3111 KASSERT(M_WRITABLE(*mp), ("%s: modifying a shared mbuf", __func__)); 3112 #endif 3113 3114 if (V_pfil_bridge == 0 && V_pfil_member == 0 && V_pfil_ipfw == 0) 3115 return (0); /* filtering is disabled */ 3116 3117 i = min((*mp)->m_pkthdr.len, max_protohdr); 3118 if ((*mp)->m_len < i) { 3119 *mp = m_pullup(*mp, i); 3120 if (*mp == NULL) { 3121 printf("%s: m_pullup failed\n", __func__); 3122 return (-1); 3123 } 3124 } 3125 3126 eh1 = mtod(*mp, struct ether_header *); 3127 ether_type = ntohs(eh1->ether_type); 3128 3129 /* 3130 * Check for SNAP/LLC. 3131 */ 3132 if (ether_type < ETHERMTU) { 3133 struct llc *llc2 = (struct llc *)(eh1 + 1); 3134 3135 if ((*mp)->m_len >= ETHER_HDR_LEN + 8 && 3136 llc2->llc_dsap == LLC_SNAP_LSAP && 3137 llc2->llc_ssap == LLC_SNAP_LSAP && 3138 llc2->llc_control == LLC_UI) { 3139 ether_type = htons(llc2->llc_un.type_snap.ether_type); 3140 snap = 1; 3141 } 3142 } 3143 3144 /* 3145 * If we're trying to filter bridge traffic, don't look at anything 3146 * other than IP and ARP traffic. If the filter doesn't understand 3147 * IPv6, don't allow IPv6 through the bridge either. This is lame 3148 * since if we really wanted, say, an AppleTalk filter, we are hosed, 3149 * but of course we don't have an AppleTalk filter to begin with. 3150 * (Note that since pfil doesn't understand ARP it will pass *ALL* 3151 * ARP traffic.) 3152 */ 3153 switch (ether_type) { 3154 case ETHERTYPE_ARP: 3155 case ETHERTYPE_REVARP: 3156 if (V_pfil_ipfw_arp == 0) 3157 return (0); /* Automatically pass */ 3158 break; 3159 3160 case ETHERTYPE_IP: 3161 #ifdef INET6 3162 case ETHERTYPE_IPV6: 3163 #endif /* INET6 */ 3164 break; 3165 default: 3166 /* 3167 * Check to see if the user wants to pass non-ip 3168 * packets, these will not be checked by pfil(9) and 3169 * passed unconditionally so the default is to drop. 3170 */ 3171 if (V_pfil_onlyip) 3172 goto bad; 3173 } 3174 3175 /* Run the packet through pfil before stripping link headers */ 3176 if (PFIL_HOOKED_OUT(V_link_pfil_head) && V_pfil_ipfw != 0 && 3177 dir == PFIL_OUT && ifp != NULL) { 3178 switch (pfil_run_hooks(V_link_pfil_head, mp, ifp, dir, NULL)) { 3179 case PFIL_DROPPED: 3180 return (EPERM); 3181 case PFIL_CONSUMED: 3182 return (0); 3183 } 3184 } 3185 3186 /* Strip off the Ethernet header and keep a copy. */ 3187 m_copydata(*mp, 0, ETHER_HDR_LEN, (caddr_t) &eh2); 3188 m_adj(*mp, ETHER_HDR_LEN); 3189 3190 /* Strip off snap header, if present */ 3191 if (snap) { 3192 m_copydata(*mp, 0, sizeof(struct llc), (caddr_t) &llc1); 3193 m_adj(*mp, sizeof(struct llc)); 3194 } 3195 3196 /* 3197 * Check the IP header for alignment and errors 3198 */ 3199 if (dir == PFIL_IN) { 3200 switch (ether_type) { 3201 case ETHERTYPE_IP: 3202 error = bridge_ip_checkbasic(mp); 3203 break; 3204 #ifdef INET6 3205 case ETHERTYPE_IPV6: 3206 error = bridge_ip6_checkbasic(mp); 3207 break; 3208 #endif /* INET6 */ 3209 default: 3210 error = 0; 3211 } 3212 if (error) 3213 goto bad; 3214 } 3215 3216 error = 0; 3217 3218 /* 3219 * Run the packet through pfil 3220 */ 3221 rv = PFIL_PASS; 3222 switch (ether_type) { 3223 case ETHERTYPE_IP: 3224 /* 3225 * Run pfil on the member interface and the bridge, both can 3226 * be skipped by clearing pfil_member or pfil_bridge. 3227 * 3228 * Keep the order: 3229 * in_if -> bridge_if -> out_if 3230 */ 3231 if (V_pfil_bridge && dir == PFIL_OUT && bifp != NULL && (rv = 3232 pfil_run_hooks(V_inet_pfil_head, mp, bifp, dir, NULL)) != 3233 PFIL_PASS) 3234 break; 3235 3236 if (V_pfil_member && ifp != NULL && (rv = 3237 pfil_run_hooks(V_inet_pfil_head, mp, ifp, dir, NULL)) != 3238 PFIL_PASS) 3239 break; 3240 3241 if (V_pfil_bridge && dir == PFIL_IN && bifp != NULL && (rv = 3242 pfil_run_hooks(V_inet_pfil_head, mp, bifp, dir, NULL)) != 3243 PFIL_PASS) 3244 break; 3245 3246 /* check if we need to fragment the packet */ 3247 /* bridge_fragment generates a mbuf chain of packets */ 3248 /* that already include eth headers */ 3249 if (V_pfil_member && ifp != NULL && dir == PFIL_OUT) { 3250 i = (*mp)->m_pkthdr.len; 3251 if (i > ifp->if_mtu) { 3252 error = bridge_fragment(ifp, mp, &eh2, snap, 3253 &llc1); 3254 return (error); 3255 } 3256 } 3257 3258 /* Recalculate the ip checksum. */ 3259 ip = mtod(*mp, struct ip *); 3260 hlen = ip->ip_hl << 2; 3261 if (hlen < sizeof(struct ip)) 3262 goto bad; 3263 if (hlen > (*mp)->m_len) { 3264 if ((*mp = m_pullup(*mp, hlen)) == NULL) 3265 goto bad; 3266 ip = mtod(*mp, struct ip *); 3267 if (ip == NULL) 3268 goto bad; 3269 } 3270 ip->ip_sum = 0; 3271 if (hlen == sizeof(struct ip)) 3272 ip->ip_sum = in_cksum_hdr(ip); 3273 else 3274 ip->ip_sum = in_cksum(*mp, hlen); 3275 3276 break; 3277 #ifdef INET6 3278 case ETHERTYPE_IPV6: 3279 if (V_pfil_bridge && dir == PFIL_OUT && bifp != NULL && (rv = 3280 pfil_run_hooks(V_inet6_pfil_head, mp, bifp, dir, NULL)) != 3281 PFIL_PASS) 3282 break; 3283 3284 if (V_pfil_member && ifp != NULL && (rv = 3285 pfil_run_hooks(V_inet6_pfil_head, mp, ifp, dir, NULL)) != 3286 PFIL_PASS) 3287 break; 3288 3289 if (V_pfil_bridge && dir == PFIL_IN && bifp != NULL && (rv = 3290 pfil_run_hooks(V_inet6_pfil_head, mp, bifp, dir, NULL)) != 3291 PFIL_PASS) 3292 break; 3293 break; 3294 #endif 3295 } 3296 3297 switch (rv) { 3298 case PFIL_CONSUMED: 3299 return (0); 3300 case PFIL_DROPPED: 3301 return (EPERM); 3302 default: 3303 break; 3304 } 3305 3306 error = -1; 3307 3308 /* 3309 * Finally, put everything back the way it was and return 3310 */ 3311 if (snap) { 3312 M_PREPEND(*mp, sizeof(struct llc), M_NOWAIT); 3313 if (*mp == NULL) 3314 return (error); 3315 bcopy(&llc1, mtod(*mp, caddr_t), sizeof(struct llc)); 3316 } 3317 3318 M_PREPEND(*mp, ETHER_HDR_LEN, M_NOWAIT); 3319 if (*mp == NULL) 3320 return (error); 3321 bcopy(&eh2, mtod(*mp, caddr_t), ETHER_HDR_LEN); 3322 3323 return (0); 3324 3325 bad: 3326 m_freem(*mp); 3327 *mp = NULL; 3328 return (error); 3329 } 3330 3331 /* 3332 * Perform basic checks on header size since 3333 * pfil assumes ip_input has already processed 3334 * it for it. Cut-and-pasted from ip_input.c. 3335 * Given how simple the IPv6 version is, 3336 * does the IPv4 version really need to be 3337 * this complicated? 3338 * 3339 * XXX Should we update ipstat here, or not? 3340 * XXX Right now we update ipstat but not 3341 * XXX csum_counter. 3342 */ 3343 static int 3344 bridge_ip_checkbasic(struct mbuf **mp) 3345 { 3346 struct mbuf *m = *mp; 3347 struct ip *ip; 3348 int len, hlen; 3349 u_short sum; 3350 3351 if (*mp == NULL) 3352 return (-1); 3353 3354 if (IP_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { 3355 if ((m = m_copyup(m, sizeof(struct ip), 3356 (max_linkhdr + 3) & ~3)) == NULL) { 3357 /* XXXJRT new stat, please */ 3358 KMOD_IPSTAT_INC(ips_toosmall); 3359 goto bad; 3360 } 3361 } else if (__predict_false(m->m_len < sizeof (struct ip))) { 3362 if ((m = m_pullup(m, sizeof (struct ip))) == NULL) { 3363 KMOD_IPSTAT_INC(ips_toosmall); 3364 goto bad; 3365 } 3366 } 3367 ip = mtod(m, struct ip *); 3368 if (ip == NULL) goto bad; 3369 3370 if (ip->ip_v != IPVERSION) { 3371 KMOD_IPSTAT_INC(ips_badvers); 3372 goto bad; 3373 } 3374 hlen = ip->ip_hl << 2; 3375 if (hlen < sizeof(struct ip)) { /* minimum header length */ 3376 KMOD_IPSTAT_INC(ips_badhlen); 3377 goto bad; 3378 } 3379 if (hlen > m->m_len) { 3380 if ((m = m_pullup(m, hlen)) == NULL) { 3381 KMOD_IPSTAT_INC(ips_badhlen); 3382 goto bad; 3383 } 3384 ip = mtod(m, struct ip *); 3385 if (ip == NULL) goto bad; 3386 } 3387 3388 if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) { 3389 sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID); 3390 } else { 3391 if (hlen == sizeof(struct ip)) { 3392 sum = in_cksum_hdr(ip); 3393 } else { 3394 sum = in_cksum(m, hlen); 3395 } 3396 } 3397 if (sum) { 3398 KMOD_IPSTAT_INC(ips_badsum); 3399 goto bad; 3400 } 3401 3402 /* Retrieve the packet length. */ 3403 len = ntohs(ip->ip_len); 3404 3405 /* 3406 * Check for additional length bogosity 3407 */ 3408 if (len < hlen) { 3409 KMOD_IPSTAT_INC(ips_badlen); 3410 goto bad; 3411 } 3412 3413 /* 3414 * Check that the amount of data in the buffers 3415 * is as at least much as the IP header would have us expect. 3416 * Drop packet if shorter than we expect. 3417 */ 3418 if (m->m_pkthdr.len < len) { 3419 KMOD_IPSTAT_INC(ips_tooshort); 3420 goto bad; 3421 } 3422 3423 /* Checks out, proceed */ 3424 *mp = m; 3425 return (0); 3426 3427 bad: 3428 *mp = m; 3429 return (-1); 3430 } 3431 3432 #ifdef INET6 3433 /* 3434 * Same as above, but for IPv6. 3435 * Cut-and-pasted from ip6_input.c. 3436 * XXX Should we update ip6stat, or not? 3437 */ 3438 static int 3439 bridge_ip6_checkbasic(struct mbuf **mp) 3440 { 3441 struct mbuf *m = *mp; 3442 struct ip6_hdr *ip6; 3443 3444 /* 3445 * If the IPv6 header is not aligned, slurp it up into a new 3446 * mbuf with space for link headers, in the event we forward 3447 * it. Otherwise, if it is aligned, make sure the entire base 3448 * IPv6 header is in the first mbuf of the chain. 3449 */ 3450 if (IP6_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { 3451 struct ifnet *inifp = m->m_pkthdr.rcvif; 3452 if ((m = m_copyup(m, sizeof(struct ip6_hdr), 3453 (max_linkhdr + 3) & ~3)) == NULL) { 3454 /* XXXJRT new stat, please */ 3455 IP6STAT_INC(ip6s_toosmall); 3456 in6_ifstat_inc(inifp, ifs6_in_hdrerr); 3457 goto bad; 3458 } 3459 } else if (__predict_false(m->m_len < sizeof(struct ip6_hdr))) { 3460 struct ifnet *inifp = m->m_pkthdr.rcvif; 3461 if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) { 3462 IP6STAT_INC(ip6s_toosmall); 3463 in6_ifstat_inc(inifp, ifs6_in_hdrerr); 3464 goto bad; 3465 } 3466 } 3467 3468 ip6 = mtod(m, struct ip6_hdr *); 3469 3470 if ((ip6->ip6_vfc & IPV6_VERSION_MASK) != IPV6_VERSION) { 3471 IP6STAT_INC(ip6s_badvers); 3472 in6_ifstat_inc(m->m_pkthdr.rcvif, ifs6_in_hdrerr); 3473 goto bad; 3474 } 3475 3476 /* Checks out, proceed */ 3477 *mp = m; 3478 return (0); 3479 3480 bad: 3481 *mp = m; 3482 return (-1); 3483 } 3484 #endif /* INET6 */ 3485 3486 /* 3487 * bridge_fragment: 3488 * 3489 * Fragment mbuf chain in multiple packets and prepend ethernet header. 3490 */ 3491 static int 3492 bridge_fragment(struct ifnet *ifp, struct mbuf **mp, struct ether_header *eh, 3493 int snap, struct llc *llc) 3494 { 3495 struct mbuf *m = *mp, *nextpkt = NULL, *mprev = NULL, *mcur = NULL; 3496 struct ip *ip; 3497 int error = -1; 3498 3499 if (m->m_len < sizeof(struct ip) && 3500 (m = m_pullup(m, sizeof(struct ip))) == NULL) 3501 goto dropit; 3502 ip = mtod(m, struct ip *); 3503 3504 m->m_pkthdr.csum_flags |= CSUM_IP; 3505 error = ip_fragment(ip, &m, ifp->if_mtu, ifp->if_hwassist); 3506 if (error) 3507 goto dropit; 3508 3509 /* 3510 * Walk the chain and re-add the Ethernet header for 3511 * each mbuf packet. 3512 */ 3513 for (mcur = m; mcur; mcur = mcur->m_nextpkt) { 3514 nextpkt = mcur->m_nextpkt; 3515 mcur->m_nextpkt = NULL; 3516 if (snap) { 3517 M_PREPEND(mcur, sizeof(struct llc), M_NOWAIT); 3518 if (mcur == NULL) { 3519 error = ENOBUFS; 3520 if (mprev != NULL) 3521 mprev->m_nextpkt = nextpkt; 3522 goto dropit; 3523 } 3524 bcopy(llc, mtod(mcur, caddr_t),sizeof(struct llc)); 3525 } 3526 3527 M_PREPEND(mcur, ETHER_HDR_LEN, M_NOWAIT); 3528 if (mcur == NULL) { 3529 error = ENOBUFS; 3530 if (mprev != NULL) 3531 mprev->m_nextpkt = nextpkt; 3532 goto dropit; 3533 } 3534 bcopy(eh, mtod(mcur, caddr_t), ETHER_HDR_LEN); 3535 3536 /* 3537 * The previous two M_PREPEND could have inserted one or two 3538 * mbufs in front so we have to update the previous packet's 3539 * m_nextpkt. 3540 */ 3541 mcur->m_nextpkt = nextpkt; 3542 if (mprev != NULL) 3543 mprev->m_nextpkt = mcur; 3544 else { 3545 /* The first mbuf in the original chain needs to be 3546 * updated. */ 3547 *mp = mcur; 3548 } 3549 mprev = mcur; 3550 } 3551 3552 KMOD_IPSTAT_INC(ips_fragmented); 3553 return (error); 3554 3555 dropit: 3556 for (mcur = *mp; mcur; mcur = m) { /* droping the full packet chain */ 3557 m = mcur->m_nextpkt; 3558 m_freem(mcur); 3559 } 3560 return (error); 3561 } 3562 3563 static void 3564 bridge_linkstate(struct ifnet *ifp) 3565 { 3566 struct bridge_softc *sc = ifp->if_bridge; 3567 struct bridge_iflist *bif; 3568 3569 BRIDGE_LOCK(sc); 3570 bif = bridge_lookup_member_if(sc, ifp); 3571 if (bif == NULL) { 3572 BRIDGE_UNLOCK(sc); 3573 return; 3574 } 3575 bridge_linkcheck(sc); 3576 BRIDGE_UNLOCK(sc); 3577 3578 bstp_linkstate(&bif->bif_stp); 3579 } 3580 3581 static void 3582 bridge_linkcheck(struct bridge_softc *sc) 3583 { 3584 struct bridge_iflist *bif; 3585 int new_link, hasls; 3586 3587 BRIDGE_LOCK_ASSERT(sc); 3588 new_link = LINK_STATE_DOWN; 3589 hasls = 0; 3590 /* Our link is considered up if at least one of our ports is active */ 3591 LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 3592 if (bif->bif_ifp->if_capabilities & IFCAP_LINKSTATE) 3593 hasls++; 3594 if (bif->bif_ifp->if_link_state == LINK_STATE_UP) { 3595 new_link = LINK_STATE_UP; 3596 break; 3597 } 3598 } 3599 if (!LIST_EMPTY(&sc->sc_iflist) && !hasls) { 3600 /* If no interfaces support link-state then we default to up */ 3601 new_link = LINK_STATE_UP; 3602 } 3603 if_link_state_change(sc->sc_ifp, new_link); 3604 } 3605