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 #ifdef INET6 139 /* 140 * XXX: declare here to avoid to include many inet6 related files.. 141 * should be more generalized? 142 */ 143 extern void nd6_setmtu(struct ifnet *); 144 #endif 145 146 /* 147 * Size of the route hash table. Must be a power of two. 148 */ 149 #ifndef BRIDGE_RTHASH_SIZE 150 #define BRIDGE_RTHASH_SIZE 1024 151 #endif 152 153 #define BRIDGE_RTHASH_MASK (BRIDGE_RTHASH_SIZE - 1) 154 155 /* 156 * Default maximum number of addresses to cache. 157 */ 158 #ifndef BRIDGE_RTABLE_MAX 159 #define BRIDGE_RTABLE_MAX 2000 160 #endif 161 162 /* 163 * Timeout (in seconds) for entries learned dynamically. 164 */ 165 #ifndef BRIDGE_RTABLE_TIMEOUT 166 #define BRIDGE_RTABLE_TIMEOUT (20 * 60) /* same as ARP */ 167 #endif 168 169 /* 170 * Number of seconds between walks of the route list. 171 */ 172 #ifndef BRIDGE_RTABLE_PRUNE_PERIOD 173 #define BRIDGE_RTABLE_PRUNE_PERIOD (5 * 60) 174 #endif 175 176 /* 177 * List of capabilities to possibly mask on the member interface. 178 */ 179 #define BRIDGE_IFCAPS_MASK (IFCAP_TOE|IFCAP_TSO|IFCAP_TXCSUM|\ 180 IFCAP_TXCSUM_IPV6) 181 182 /* 183 * List of capabilities to strip 184 */ 185 #define BRIDGE_IFCAPS_STRIP IFCAP_LRO 186 187 /* 188 * Bridge locking 189 * 190 * The bridge relies heavily on the epoch(9) system to protect its data 191 * structures. This means we can safely use CK_LISTs while in NET_EPOCH, but we 192 * must ensure there is only one writer at a time. 193 * 194 * That is: for read accesses we only need to be in NET_EPOCH, but for write 195 * accesses we must hold: 196 * 197 * - BRIDGE_RT_LOCK, for any change to bridge_rtnodes 198 * - BRIDGE_LOCK, for any other change 199 * 200 * The BRIDGE_LOCK is a sleepable lock, because it is held accross ioctl() 201 * calls to bridge member interfaces and these ioctl()s can sleep. 202 * The BRIDGE_RT_LOCK is a non-sleepable mutex, because it is sometimes 203 * required while we're in NET_EPOCH and then we're not allowed to sleep. 204 */ 205 #define BRIDGE_LOCK_INIT(_sc) do { \ 206 sx_init(&(_sc)->sc_sx, "if_bridge"); \ 207 mtx_init(&(_sc)->sc_rt_mtx, "if_bridge rt", NULL, MTX_DEF); \ 208 } while (0) 209 #define BRIDGE_LOCK_DESTROY(_sc) do { \ 210 sx_destroy(&(_sc)->sc_sx); \ 211 mtx_destroy(&(_sc)->sc_rt_mtx); \ 212 } while (0) 213 #define BRIDGE_LOCK(_sc) sx_xlock(&(_sc)->sc_sx) 214 #define BRIDGE_UNLOCK(_sc) sx_xunlock(&(_sc)->sc_sx) 215 #define BRIDGE_LOCK_ASSERT(_sc) sx_assert(&(_sc)->sc_sx, SX_XLOCKED) 216 #define BRIDGE_LOCK_OR_NET_EPOCH_ASSERT(_sc) \ 217 MPASS(in_epoch(net_epoch_preempt) || sx_xlocked(&(_sc)->sc_sx)) 218 #define BRIDGE_UNLOCK_ASSERT(_sc) sx_assert(&(_sc)->sc_sx, SX_UNLOCKED) 219 #define BRIDGE_RT_LOCK(_sc) mtx_lock(&(_sc)->sc_rt_mtx) 220 #define BRIDGE_RT_UNLOCK(_sc) mtx_unlock(&(_sc)->sc_rt_mtx) 221 #define BRIDGE_RT_LOCK_ASSERT(_sc) mtx_assert(&(_sc)->sc_rt_mtx, MA_OWNED) 222 #define BRIDGE_RT_LOCK_OR_NET_EPOCH_ASSERT(_sc) \ 223 MPASS(in_epoch(net_epoch_preempt) || mtx_owned(&(_sc)->sc_rt_mtx)) 224 225 /* 226 * Bridge interface list entry. 227 */ 228 struct bridge_iflist { 229 CK_LIST_ENTRY(bridge_iflist) bif_next; 230 struct ifnet *bif_ifp; /* member if */ 231 struct bstp_port bif_stp; /* STP state */ 232 uint32_t bif_flags; /* member if flags */ 233 int bif_savedcaps; /* saved capabilities */ 234 uint32_t bif_addrmax; /* max # of addresses */ 235 uint32_t bif_addrcnt; /* cur. # of addresses */ 236 uint32_t bif_addrexceeded;/* # of address violations */ 237 struct epoch_context bif_epoch_ctx; 238 }; 239 240 /* 241 * Bridge route node. 242 */ 243 struct bridge_rtnode { 244 CK_LIST_ENTRY(bridge_rtnode) brt_hash; /* hash table linkage */ 245 CK_LIST_ENTRY(bridge_rtnode) brt_list; /* list linkage */ 246 struct bridge_iflist *brt_dst; /* destination if */ 247 unsigned long brt_expire; /* expiration time */ 248 uint8_t brt_flags; /* address flags */ 249 uint8_t brt_addr[ETHER_ADDR_LEN]; 250 uint16_t brt_vlan; /* vlan id */ 251 struct vnet *brt_vnet; 252 struct epoch_context brt_epoch_ctx; 253 }; 254 #define brt_ifp brt_dst->bif_ifp 255 256 /* 257 * Software state for each bridge. 258 */ 259 struct bridge_softc { 260 struct ifnet *sc_ifp; /* make this an interface */ 261 LIST_ENTRY(bridge_softc) sc_list; 262 struct sx sc_sx; 263 struct mtx sc_rt_mtx; 264 uint32_t sc_brtmax; /* max # of addresses */ 265 uint32_t sc_brtcnt; /* cur. # of addresses */ 266 uint32_t sc_brttimeout; /* rt timeout in seconds */ 267 struct callout sc_brcallout; /* bridge callout */ 268 CK_LIST_HEAD(, bridge_iflist) sc_iflist; /* member interface list */ 269 CK_LIST_HEAD(, bridge_rtnode) *sc_rthash; /* our forwarding table */ 270 CK_LIST_HEAD(, bridge_rtnode) sc_rtlist; /* list version of above */ 271 uint32_t sc_rthash_key; /* key for hash */ 272 CK_LIST_HEAD(, bridge_iflist) sc_spanlist; /* span ports list */ 273 struct bstp_state sc_stp; /* STP state */ 274 uint32_t sc_brtexceeded; /* # of cache drops */ 275 struct ifnet *sc_ifaddr; /* member mac copied from */ 276 struct ether_addr sc_defaddr; /* Default MAC address */ 277 struct epoch_context sc_epoch_ctx; 278 }; 279 280 VNET_DEFINE_STATIC(struct sx, bridge_list_sx); 281 #define V_bridge_list_sx VNET(bridge_list_sx) 282 static eventhandler_tag bridge_detach_cookie; 283 284 int bridge_rtable_prune_period = BRIDGE_RTABLE_PRUNE_PERIOD; 285 286 VNET_DEFINE_STATIC(uma_zone_t, bridge_rtnode_zone); 287 #define V_bridge_rtnode_zone VNET(bridge_rtnode_zone) 288 289 static int bridge_clone_create(struct if_clone *, int, caddr_t); 290 static void bridge_clone_destroy(struct ifnet *); 291 292 static int bridge_ioctl(struct ifnet *, u_long, caddr_t); 293 static void bridge_mutecaps(struct bridge_softc *); 294 static void bridge_set_ifcap(struct bridge_softc *, struct bridge_iflist *, 295 int); 296 static void bridge_ifdetach(void *arg __unused, struct ifnet *); 297 static void bridge_init(void *); 298 static void bridge_dummynet(struct mbuf *, struct ifnet *); 299 static void bridge_stop(struct ifnet *, int); 300 static int bridge_transmit(struct ifnet *, struct mbuf *); 301 static void bridge_qflush(struct ifnet *); 302 static struct mbuf *bridge_input(struct ifnet *, struct mbuf *); 303 static int bridge_output(struct ifnet *, struct mbuf *, struct sockaddr *, 304 struct rtentry *); 305 static int bridge_enqueue(struct bridge_softc *, struct ifnet *, 306 struct mbuf *); 307 static void bridge_rtdelete(struct bridge_softc *, struct ifnet *ifp, int); 308 309 static void bridge_forward(struct bridge_softc *, struct bridge_iflist *, 310 struct mbuf *m); 311 312 static void bridge_timer(void *); 313 314 static void bridge_broadcast(struct bridge_softc *, struct ifnet *, 315 struct mbuf *, int); 316 static void bridge_span(struct bridge_softc *, struct mbuf *); 317 318 static int bridge_rtupdate(struct bridge_softc *, const uint8_t *, 319 uint16_t, struct bridge_iflist *, int, uint8_t); 320 static struct ifnet *bridge_rtlookup(struct bridge_softc *, const uint8_t *, 321 uint16_t); 322 static void bridge_rttrim(struct bridge_softc *); 323 static void bridge_rtage(struct bridge_softc *); 324 static void bridge_rtflush(struct bridge_softc *, int); 325 static int bridge_rtdaddr(struct bridge_softc *, const uint8_t *, 326 uint16_t); 327 328 static void bridge_rtable_init(struct bridge_softc *); 329 static void bridge_rtable_fini(struct bridge_softc *); 330 331 static int bridge_rtnode_addr_cmp(const uint8_t *, const uint8_t *); 332 static struct bridge_rtnode *bridge_rtnode_lookup(struct bridge_softc *, 333 const uint8_t *, uint16_t); 334 static int bridge_rtnode_insert(struct bridge_softc *, 335 struct bridge_rtnode *); 336 static void bridge_rtnode_destroy(struct bridge_softc *, 337 struct bridge_rtnode *); 338 static void bridge_rtable_expire(struct ifnet *, int); 339 static void bridge_state_change(struct ifnet *, int); 340 341 static struct bridge_iflist *bridge_lookup_member(struct bridge_softc *, 342 const char *name); 343 static struct bridge_iflist *bridge_lookup_member_if(struct bridge_softc *, 344 struct ifnet *ifp); 345 static void bridge_delete_member(struct bridge_softc *, 346 struct bridge_iflist *, int); 347 static void bridge_delete_span(struct bridge_softc *, 348 struct bridge_iflist *); 349 350 static int bridge_ioctl_add(struct bridge_softc *, void *); 351 static int bridge_ioctl_del(struct bridge_softc *, void *); 352 static int bridge_ioctl_gifflags(struct bridge_softc *, void *); 353 static int bridge_ioctl_sifflags(struct bridge_softc *, void *); 354 static int bridge_ioctl_scache(struct bridge_softc *, void *); 355 static int bridge_ioctl_gcache(struct bridge_softc *, void *); 356 static int bridge_ioctl_gifs(struct bridge_softc *, void *); 357 static int bridge_ioctl_rts(struct bridge_softc *, void *); 358 static int bridge_ioctl_saddr(struct bridge_softc *, void *); 359 static int bridge_ioctl_sto(struct bridge_softc *, void *); 360 static int bridge_ioctl_gto(struct bridge_softc *, void *); 361 static int bridge_ioctl_daddr(struct bridge_softc *, void *); 362 static int bridge_ioctl_flush(struct bridge_softc *, void *); 363 static int bridge_ioctl_gpri(struct bridge_softc *, void *); 364 static int bridge_ioctl_spri(struct bridge_softc *, void *); 365 static int bridge_ioctl_ght(struct bridge_softc *, void *); 366 static int bridge_ioctl_sht(struct bridge_softc *, void *); 367 static int bridge_ioctl_gfd(struct bridge_softc *, void *); 368 static int bridge_ioctl_sfd(struct bridge_softc *, void *); 369 static int bridge_ioctl_gma(struct bridge_softc *, void *); 370 static int bridge_ioctl_sma(struct bridge_softc *, void *); 371 static int bridge_ioctl_sifprio(struct bridge_softc *, void *); 372 static int bridge_ioctl_sifcost(struct bridge_softc *, void *); 373 static int bridge_ioctl_sifmaxaddr(struct bridge_softc *, void *); 374 static int bridge_ioctl_addspan(struct bridge_softc *, void *); 375 static int bridge_ioctl_delspan(struct bridge_softc *, void *); 376 static int bridge_ioctl_gbparam(struct bridge_softc *, void *); 377 static int bridge_ioctl_grte(struct bridge_softc *, void *); 378 static int bridge_ioctl_gifsstp(struct bridge_softc *, void *); 379 static int bridge_ioctl_sproto(struct bridge_softc *, void *); 380 static int bridge_ioctl_stxhc(struct bridge_softc *, void *); 381 static int bridge_pfil(struct mbuf **, struct ifnet *, struct ifnet *, 382 int); 383 static int bridge_ip_checkbasic(struct mbuf **mp); 384 #ifdef INET6 385 static int bridge_ip6_checkbasic(struct mbuf **mp); 386 #endif /* INET6 */ 387 static int bridge_fragment(struct ifnet *, struct mbuf **mp, 388 struct ether_header *, int, struct llc *); 389 static void bridge_linkstate(struct ifnet *ifp); 390 static void bridge_linkcheck(struct bridge_softc *sc); 391 392 /* The default bridge vlan is 1 (IEEE 802.1Q-2003 Table 9-2) */ 393 #define VLANTAGOF(_m) \ 394 (_m->m_flags & M_VLANTAG) ? EVL_VLANOFTAG(_m->m_pkthdr.ether_vtag) : 1 395 396 static struct bstp_cb_ops bridge_ops = { 397 .bcb_state = bridge_state_change, 398 .bcb_rtage = bridge_rtable_expire 399 }; 400 401 SYSCTL_DECL(_net_link); 402 static SYSCTL_NODE(_net_link, IFT_BRIDGE, bridge, CTLFLAG_RW | CTLFLAG_MPSAFE, 0, 403 "Bridge"); 404 405 /* only pass IP[46] packets when pfil is enabled */ 406 VNET_DEFINE_STATIC(int, pfil_onlyip) = 1; 407 #define V_pfil_onlyip VNET(pfil_onlyip) 408 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_onlyip, 409 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_onlyip), 0, 410 "Only pass IP packets when pfil is enabled"); 411 412 /* run pfil hooks on the bridge interface */ 413 VNET_DEFINE_STATIC(int, pfil_bridge) = 1; 414 #define V_pfil_bridge VNET(pfil_bridge) 415 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_bridge, 416 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_bridge), 0, 417 "Packet filter on the bridge interface"); 418 419 /* layer2 filter with ipfw */ 420 VNET_DEFINE_STATIC(int, pfil_ipfw); 421 #define V_pfil_ipfw VNET(pfil_ipfw) 422 423 /* layer2 ARP filter with ipfw */ 424 VNET_DEFINE_STATIC(int, pfil_ipfw_arp); 425 #define V_pfil_ipfw_arp VNET(pfil_ipfw_arp) 426 SYSCTL_INT(_net_link_bridge, OID_AUTO, ipfw_arp, 427 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_ipfw_arp), 0, 428 "Filter ARP packets through IPFW layer2"); 429 430 /* run pfil hooks on the member interface */ 431 VNET_DEFINE_STATIC(int, pfil_member) = 1; 432 #define V_pfil_member VNET(pfil_member) 433 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_member, 434 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_member), 0, 435 "Packet filter on the member interface"); 436 437 /* run pfil hooks on the physical interface for locally destined packets */ 438 VNET_DEFINE_STATIC(int, pfil_local_phys); 439 #define V_pfil_local_phys VNET(pfil_local_phys) 440 SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_local_phys, 441 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(pfil_local_phys), 0, 442 "Packet filter on the physical interface for locally destined packets"); 443 444 /* log STP state changes */ 445 VNET_DEFINE_STATIC(int, log_stp); 446 #define V_log_stp VNET(log_stp) 447 SYSCTL_INT(_net_link_bridge, OID_AUTO, log_stp, 448 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(log_stp), 0, 449 "Log STP state changes"); 450 451 /* share MAC with first bridge member */ 452 VNET_DEFINE_STATIC(int, bridge_inherit_mac); 453 #define V_bridge_inherit_mac VNET(bridge_inherit_mac) 454 SYSCTL_INT(_net_link_bridge, OID_AUTO, inherit_mac, 455 CTLFLAG_RWTUN | CTLFLAG_VNET, &VNET_NAME(bridge_inherit_mac), 0, 456 "Inherit MAC address from the first bridge member"); 457 458 VNET_DEFINE_STATIC(int, allow_llz_overlap) = 0; 459 #define V_allow_llz_overlap VNET(allow_llz_overlap) 460 SYSCTL_INT(_net_link_bridge, OID_AUTO, allow_llz_overlap, 461 CTLFLAG_RW | CTLFLAG_VNET, &VNET_NAME(allow_llz_overlap), 0, 462 "Allow overlap of link-local scope " 463 "zones of a bridge interface and the member interfaces"); 464 465 struct bridge_control { 466 int (*bc_func)(struct bridge_softc *, void *); 467 int bc_argsize; 468 int bc_flags; 469 }; 470 471 #define BC_F_COPYIN 0x01 /* copy arguments in */ 472 #define BC_F_COPYOUT 0x02 /* copy arguments out */ 473 #define BC_F_SUSER 0x04 /* do super-user check */ 474 475 const struct bridge_control bridge_control_table[] = { 476 { bridge_ioctl_add, sizeof(struct ifbreq), 477 BC_F_COPYIN|BC_F_SUSER }, 478 { bridge_ioctl_del, sizeof(struct ifbreq), 479 BC_F_COPYIN|BC_F_SUSER }, 480 481 { bridge_ioctl_gifflags, sizeof(struct ifbreq), 482 BC_F_COPYIN|BC_F_COPYOUT }, 483 { bridge_ioctl_sifflags, sizeof(struct ifbreq), 484 BC_F_COPYIN|BC_F_SUSER }, 485 486 { bridge_ioctl_scache, sizeof(struct ifbrparam), 487 BC_F_COPYIN|BC_F_SUSER }, 488 { bridge_ioctl_gcache, sizeof(struct ifbrparam), 489 BC_F_COPYOUT }, 490 491 { bridge_ioctl_gifs, sizeof(struct ifbifconf), 492 BC_F_COPYIN|BC_F_COPYOUT }, 493 { bridge_ioctl_rts, sizeof(struct ifbaconf), 494 BC_F_COPYIN|BC_F_COPYOUT }, 495 496 { bridge_ioctl_saddr, sizeof(struct ifbareq), 497 BC_F_COPYIN|BC_F_SUSER }, 498 499 { bridge_ioctl_sto, sizeof(struct ifbrparam), 500 BC_F_COPYIN|BC_F_SUSER }, 501 { bridge_ioctl_gto, sizeof(struct ifbrparam), 502 BC_F_COPYOUT }, 503 504 { bridge_ioctl_daddr, sizeof(struct ifbareq), 505 BC_F_COPYIN|BC_F_SUSER }, 506 507 { bridge_ioctl_flush, sizeof(struct ifbreq), 508 BC_F_COPYIN|BC_F_SUSER }, 509 510 { bridge_ioctl_gpri, sizeof(struct ifbrparam), 511 BC_F_COPYOUT }, 512 { bridge_ioctl_spri, sizeof(struct ifbrparam), 513 BC_F_COPYIN|BC_F_SUSER }, 514 515 { bridge_ioctl_ght, sizeof(struct ifbrparam), 516 BC_F_COPYOUT }, 517 { bridge_ioctl_sht, sizeof(struct ifbrparam), 518 BC_F_COPYIN|BC_F_SUSER }, 519 520 { bridge_ioctl_gfd, sizeof(struct ifbrparam), 521 BC_F_COPYOUT }, 522 { bridge_ioctl_sfd, sizeof(struct ifbrparam), 523 BC_F_COPYIN|BC_F_SUSER }, 524 525 { bridge_ioctl_gma, sizeof(struct ifbrparam), 526 BC_F_COPYOUT }, 527 { bridge_ioctl_sma, sizeof(struct ifbrparam), 528 BC_F_COPYIN|BC_F_SUSER }, 529 530 { bridge_ioctl_sifprio, sizeof(struct ifbreq), 531 BC_F_COPYIN|BC_F_SUSER }, 532 533 { bridge_ioctl_sifcost, sizeof(struct ifbreq), 534 BC_F_COPYIN|BC_F_SUSER }, 535 536 { bridge_ioctl_addspan, sizeof(struct ifbreq), 537 BC_F_COPYIN|BC_F_SUSER }, 538 { bridge_ioctl_delspan, sizeof(struct ifbreq), 539 BC_F_COPYIN|BC_F_SUSER }, 540 541 { bridge_ioctl_gbparam, sizeof(struct ifbropreq), 542 BC_F_COPYOUT }, 543 544 { bridge_ioctl_grte, sizeof(struct ifbrparam), 545 BC_F_COPYOUT }, 546 547 { bridge_ioctl_gifsstp, sizeof(struct ifbpstpconf), 548 BC_F_COPYIN|BC_F_COPYOUT }, 549 550 { bridge_ioctl_sproto, sizeof(struct ifbrparam), 551 BC_F_COPYIN|BC_F_SUSER }, 552 553 { bridge_ioctl_stxhc, sizeof(struct ifbrparam), 554 BC_F_COPYIN|BC_F_SUSER }, 555 556 { bridge_ioctl_sifmaxaddr, sizeof(struct ifbreq), 557 BC_F_COPYIN|BC_F_SUSER }, 558 559 }; 560 const int bridge_control_table_size = nitems(bridge_control_table); 561 562 VNET_DEFINE_STATIC(LIST_HEAD(, bridge_softc), bridge_list); 563 #define V_bridge_list VNET(bridge_list) 564 #define BRIDGE_LIST_LOCK_INIT(x) sx_init(&V_bridge_list_sx, \ 565 "if_bridge list") 566 #define BRIDGE_LIST_LOCK_DESTROY(x) sx_destroy(&V_bridge_list_sx) 567 #define BRIDGE_LIST_LOCK(x) sx_xlock(&V_bridge_list_sx) 568 #define BRIDGE_LIST_UNLOCK(x) sx_xunlock(&V_bridge_list_sx) 569 570 VNET_DEFINE_STATIC(struct if_clone *, bridge_cloner); 571 #define V_bridge_cloner VNET(bridge_cloner) 572 573 static const char bridge_name[] = "bridge"; 574 575 static void 576 vnet_bridge_init(const void *unused __unused) 577 { 578 579 V_bridge_rtnode_zone = uma_zcreate("bridge_rtnode", 580 sizeof(struct bridge_rtnode), NULL, NULL, NULL, NULL, 581 UMA_ALIGN_PTR, 0); 582 BRIDGE_LIST_LOCK_INIT(); 583 LIST_INIT(&V_bridge_list); 584 V_bridge_cloner = if_clone_simple(bridge_name, 585 bridge_clone_create, bridge_clone_destroy, 0); 586 } 587 VNET_SYSINIT(vnet_bridge_init, SI_SUB_PROTO_IFATTACHDOMAIN, SI_ORDER_ANY, 588 vnet_bridge_init, NULL); 589 590 static void 591 vnet_bridge_uninit(const void *unused __unused) 592 { 593 594 if_clone_detach(V_bridge_cloner); 595 V_bridge_cloner = NULL; 596 BRIDGE_LIST_LOCK_DESTROY(); 597 598 /* Callbacks may use the UMA zone. */ 599 epoch_drain_callbacks(net_epoch_preempt); 600 601 uma_zdestroy(V_bridge_rtnode_zone); 602 } 603 VNET_SYSUNINIT(vnet_bridge_uninit, SI_SUB_PSEUDO, SI_ORDER_ANY, 604 vnet_bridge_uninit, NULL); 605 606 static int 607 bridge_modevent(module_t mod, int type, void *data) 608 { 609 610 switch (type) { 611 case MOD_LOAD: 612 bridge_dn_p = bridge_dummynet; 613 bridge_detach_cookie = EVENTHANDLER_REGISTER( 614 ifnet_departure_event, bridge_ifdetach, NULL, 615 EVENTHANDLER_PRI_ANY); 616 break; 617 case MOD_UNLOAD: 618 EVENTHANDLER_DEREGISTER(ifnet_departure_event, 619 bridge_detach_cookie); 620 bridge_dn_p = NULL; 621 break; 622 default: 623 return (EOPNOTSUPP); 624 } 625 return (0); 626 } 627 628 static moduledata_t bridge_mod = { 629 "if_bridge", 630 bridge_modevent, 631 0 632 }; 633 634 DECLARE_MODULE(if_bridge, bridge_mod, SI_SUB_PSEUDO, SI_ORDER_ANY); 635 MODULE_VERSION(if_bridge, 1); 636 MODULE_DEPEND(if_bridge, bridgestp, 1, 1, 1); 637 638 /* 639 * handler for net.link.bridge.ipfw 640 */ 641 static int 642 sysctl_pfil_ipfw(SYSCTL_HANDLER_ARGS) 643 { 644 int enable = V_pfil_ipfw; 645 int error; 646 647 error = sysctl_handle_int(oidp, &enable, 0, req); 648 enable &= 1; 649 650 if (enable != V_pfil_ipfw) { 651 V_pfil_ipfw = enable; 652 653 /* 654 * Disable pfil so that ipfw doesnt run twice, if the user 655 * really wants both then they can re-enable pfil_bridge and/or 656 * pfil_member. Also allow non-ip packets as ipfw can filter by 657 * layer2 type. 658 */ 659 if (V_pfil_ipfw) { 660 V_pfil_onlyip = 0; 661 V_pfil_bridge = 0; 662 V_pfil_member = 0; 663 } 664 } 665 666 return (error); 667 } 668 SYSCTL_PROC(_net_link_bridge, OID_AUTO, ipfw, 669 CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_VNET | CTLFLAG_NEEDGIANT, 670 &VNET_NAME(pfil_ipfw), 0, &sysctl_pfil_ipfw, "I", 671 "Layer2 filter with IPFW"); 672 673 /* 674 * bridge_clone_create: 675 * 676 * Create a new bridge instance. 677 */ 678 static int 679 bridge_clone_create(struct if_clone *ifc, int unit, caddr_t params) 680 { 681 struct bridge_softc *sc; 682 struct ifnet *ifp; 683 684 sc = malloc(sizeof(*sc), M_DEVBUF, M_WAITOK|M_ZERO); 685 ifp = sc->sc_ifp = if_alloc(IFT_ETHER); 686 if (ifp == NULL) { 687 free(sc, M_DEVBUF); 688 return (ENOSPC); 689 } 690 691 BRIDGE_LOCK_INIT(sc); 692 sc->sc_brtmax = BRIDGE_RTABLE_MAX; 693 sc->sc_brttimeout = BRIDGE_RTABLE_TIMEOUT; 694 695 /* Initialize our routing table. */ 696 bridge_rtable_init(sc); 697 698 callout_init_mtx(&sc->sc_brcallout, &sc->sc_rt_mtx, 0); 699 700 CK_LIST_INIT(&sc->sc_iflist); 701 CK_LIST_INIT(&sc->sc_spanlist); 702 703 ifp->if_softc = sc; 704 if_initname(ifp, bridge_name, unit); 705 ifp->if_flags = IFF_BROADCAST | IFF_SIMPLEX | IFF_MULTICAST; 706 ifp->if_ioctl = bridge_ioctl; 707 ifp->if_transmit = bridge_transmit; 708 ifp->if_qflush = bridge_qflush; 709 ifp->if_init = bridge_init; 710 ifp->if_type = IFT_BRIDGE; 711 712 ether_gen_addr(ifp, &sc->sc_defaddr); 713 714 bstp_attach(&sc->sc_stp, &bridge_ops); 715 ether_ifattach(ifp, sc->sc_defaddr.octet); 716 /* Now undo some of the damage... */ 717 ifp->if_baudrate = 0; 718 ifp->if_type = IFT_BRIDGE; 719 720 BRIDGE_LIST_LOCK(); 721 LIST_INSERT_HEAD(&V_bridge_list, sc, sc_list); 722 BRIDGE_LIST_UNLOCK(); 723 724 return (0); 725 } 726 727 static void 728 bridge_clone_destroy_cb(struct epoch_context *ctx) 729 { 730 struct bridge_softc *sc; 731 732 sc = __containerof(ctx, struct bridge_softc, sc_epoch_ctx); 733 734 BRIDGE_LOCK_DESTROY(sc); 735 free(sc, M_DEVBUF); 736 } 737 738 /* 739 * bridge_clone_destroy: 740 * 741 * Destroy a bridge instance. 742 */ 743 static void 744 bridge_clone_destroy(struct ifnet *ifp) 745 { 746 struct bridge_softc *sc = ifp->if_softc; 747 struct bridge_iflist *bif; 748 struct epoch_tracker et; 749 750 BRIDGE_LOCK(sc); 751 752 bridge_stop(ifp, 1); 753 ifp->if_flags &= ~IFF_UP; 754 755 while ((bif = CK_LIST_FIRST(&sc->sc_iflist)) != NULL) 756 bridge_delete_member(sc, bif, 0); 757 758 while ((bif = CK_LIST_FIRST(&sc->sc_spanlist)) != NULL) { 759 bridge_delete_span(sc, bif); 760 } 761 762 /* Tear down the routing table. */ 763 bridge_rtable_fini(sc); 764 765 BRIDGE_UNLOCK(sc); 766 767 NET_EPOCH_ENTER(et); 768 769 callout_drain(&sc->sc_brcallout); 770 771 BRIDGE_LIST_LOCK(); 772 LIST_REMOVE(sc, sc_list); 773 BRIDGE_LIST_UNLOCK(); 774 775 bstp_detach(&sc->sc_stp); 776 NET_EPOCH_EXIT(et); 777 778 ether_ifdetach(ifp); 779 if_free(ifp); 780 781 NET_EPOCH_CALL(bridge_clone_destroy_cb, &sc->sc_epoch_ctx); 782 } 783 784 /* 785 * bridge_ioctl: 786 * 787 * Handle a control request from the operator. 788 */ 789 static int 790 bridge_ioctl(struct ifnet *ifp, u_long cmd, caddr_t data) 791 { 792 struct bridge_softc *sc = ifp->if_softc; 793 struct ifreq *ifr = (struct ifreq *)data; 794 struct bridge_iflist *bif; 795 struct thread *td = curthread; 796 union { 797 struct ifbreq ifbreq; 798 struct ifbifconf ifbifconf; 799 struct ifbareq ifbareq; 800 struct ifbaconf ifbaconf; 801 struct ifbrparam ifbrparam; 802 struct ifbropreq ifbropreq; 803 } args; 804 struct ifdrv *ifd = (struct ifdrv *) data; 805 const struct bridge_control *bc; 806 int error = 0, oldmtu; 807 808 BRIDGE_LOCK(sc); 809 810 switch (cmd) { 811 case SIOCADDMULTI: 812 case SIOCDELMULTI: 813 break; 814 815 case SIOCGDRVSPEC: 816 case SIOCSDRVSPEC: 817 if (ifd->ifd_cmd >= bridge_control_table_size) { 818 error = EINVAL; 819 break; 820 } 821 bc = &bridge_control_table[ifd->ifd_cmd]; 822 823 if (cmd == SIOCGDRVSPEC && 824 (bc->bc_flags & BC_F_COPYOUT) == 0) { 825 error = EINVAL; 826 break; 827 } 828 else if (cmd == SIOCSDRVSPEC && 829 (bc->bc_flags & BC_F_COPYOUT) != 0) { 830 error = EINVAL; 831 break; 832 } 833 834 if (bc->bc_flags & BC_F_SUSER) { 835 error = priv_check(td, PRIV_NET_BRIDGE); 836 if (error) 837 break; 838 } 839 840 if (ifd->ifd_len != bc->bc_argsize || 841 ifd->ifd_len > sizeof(args)) { 842 error = EINVAL; 843 break; 844 } 845 846 bzero(&args, sizeof(args)); 847 if (bc->bc_flags & BC_F_COPYIN) { 848 error = copyin(ifd->ifd_data, &args, ifd->ifd_len); 849 if (error) 850 break; 851 } 852 853 oldmtu = ifp->if_mtu; 854 error = (*bc->bc_func)(sc, &args); 855 if (error) 856 break; 857 858 /* 859 * Bridge MTU may change during addition of the first port. 860 * If it did, do network layer specific procedure. 861 */ 862 if (ifp->if_mtu != oldmtu) { 863 #ifdef INET6 864 nd6_setmtu(ifp); 865 #endif 866 rt_updatemtu(ifp); 867 } 868 869 if (bc->bc_flags & BC_F_COPYOUT) 870 error = copyout(&args, ifd->ifd_data, ifd->ifd_len); 871 872 break; 873 874 case SIOCSIFFLAGS: 875 if (!(ifp->if_flags & IFF_UP) && 876 (ifp->if_drv_flags & IFF_DRV_RUNNING)) { 877 /* 878 * If interface is marked down and it is running, 879 * then stop and disable it. 880 */ 881 bridge_stop(ifp, 1); 882 } else if ((ifp->if_flags & IFF_UP) && 883 !(ifp->if_drv_flags & IFF_DRV_RUNNING)) { 884 /* 885 * If interface is marked up and it is stopped, then 886 * start it. 887 */ 888 BRIDGE_UNLOCK(sc); 889 (*ifp->if_init)(sc); 890 BRIDGE_LOCK(sc); 891 } 892 break; 893 894 case SIOCSIFMTU: 895 if (ifr->ifr_mtu < 576) { 896 error = EINVAL; 897 break; 898 } 899 if (CK_LIST_EMPTY(&sc->sc_iflist)) { 900 sc->sc_ifp->if_mtu = ifr->ifr_mtu; 901 break; 902 } 903 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 904 if (bif->bif_ifp->if_mtu != ifr->ifr_mtu) { 905 log(LOG_NOTICE, "%s: invalid MTU: %u(%s)" 906 " != %d\n", sc->sc_ifp->if_xname, 907 bif->bif_ifp->if_mtu, 908 bif->bif_ifp->if_xname, ifr->ifr_mtu); 909 error = EINVAL; 910 break; 911 } 912 } 913 if (!error) 914 sc->sc_ifp->if_mtu = ifr->ifr_mtu; 915 break; 916 default: 917 /* 918 * drop the lock as ether_ioctl() will call bridge_start() and 919 * cause the lock to be recursed. 920 */ 921 BRIDGE_UNLOCK(sc); 922 error = ether_ioctl(ifp, cmd, data); 923 BRIDGE_LOCK(sc); 924 break; 925 } 926 927 BRIDGE_UNLOCK(sc); 928 929 return (error); 930 } 931 932 /* 933 * bridge_mutecaps: 934 * 935 * Clear or restore unwanted capabilities on the member interface 936 */ 937 static void 938 bridge_mutecaps(struct bridge_softc *sc) 939 { 940 struct bridge_iflist *bif; 941 int enabled, mask; 942 943 BRIDGE_LOCK_ASSERT(sc); 944 945 /* Initial bitmask of capabilities to test */ 946 mask = BRIDGE_IFCAPS_MASK; 947 948 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 949 /* Every member must support it or its disabled */ 950 mask &= bif->bif_savedcaps; 951 } 952 953 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 954 enabled = bif->bif_ifp->if_capenable; 955 enabled &= ~BRIDGE_IFCAPS_STRIP; 956 /* strip off mask bits and enable them again if allowed */ 957 enabled &= ~BRIDGE_IFCAPS_MASK; 958 enabled |= mask; 959 bridge_set_ifcap(sc, bif, enabled); 960 } 961 } 962 963 static void 964 bridge_set_ifcap(struct bridge_softc *sc, struct bridge_iflist *bif, int set) 965 { 966 struct ifnet *ifp = bif->bif_ifp; 967 struct ifreq ifr; 968 int error, mask, stuck; 969 970 bzero(&ifr, sizeof(ifr)); 971 ifr.ifr_reqcap = set; 972 973 if (ifp->if_capenable != set) { 974 error = (*ifp->if_ioctl)(ifp, SIOCSIFCAP, (caddr_t)&ifr); 975 if (error) 976 if_printf(sc->sc_ifp, 977 "error setting capabilities on %s: %d\n", 978 ifp->if_xname, error); 979 mask = BRIDGE_IFCAPS_MASK | BRIDGE_IFCAPS_STRIP; 980 stuck = ifp->if_capenable & mask & ~set; 981 if (stuck != 0) 982 if_printf(sc->sc_ifp, 983 "can't disable some capabilities on %s: 0x%x\n", 984 ifp->if_xname, stuck); 985 } 986 } 987 988 /* 989 * bridge_lookup_member: 990 * 991 * Lookup a bridge member interface. 992 */ 993 static struct bridge_iflist * 994 bridge_lookup_member(struct bridge_softc *sc, const char *name) 995 { 996 struct bridge_iflist *bif; 997 struct ifnet *ifp; 998 999 BRIDGE_LOCK_OR_NET_EPOCH_ASSERT(sc); 1000 1001 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1002 ifp = bif->bif_ifp; 1003 if (strcmp(ifp->if_xname, name) == 0) 1004 return (bif); 1005 } 1006 1007 return (NULL); 1008 } 1009 1010 /* 1011 * bridge_lookup_member_if: 1012 * 1013 * Lookup a bridge member interface by ifnet*. 1014 */ 1015 static struct bridge_iflist * 1016 bridge_lookup_member_if(struct bridge_softc *sc, struct ifnet *member_ifp) 1017 { 1018 struct bridge_iflist *bif; 1019 1020 BRIDGE_LOCK_OR_NET_EPOCH_ASSERT(sc); 1021 1022 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1023 if (bif->bif_ifp == member_ifp) 1024 return (bif); 1025 } 1026 1027 return (NULL); 1028 } 1029 1030 static void 1031 bridge_delete_member_cb(struct epoch_context *ctx) 1032 { 1033 struct bridge_iflist *bif; 1034 1035 bif = __containerof(ctx, struct bridge_iflist, bif_epoch_ctx); 1036 1037 free(bif, M_DEVBUF); 1038 } 1039 1040 /* 1041 * bridge_delete_member: 1042 * 1043 * Delete the specified member interface. 1044 */ 1045 static void 1046 bridge_delete_member(struct bridge_softc *sc, struct bridge_iflist *bif, 1047 int gone) 1048 { 1049 struct ifnet *ifs = bif->bif_ifp; 1050 struct ifnet *fif = NULL; 1051 struct bridge_iflist *bifl; 1052 1053 BRIDGE_LOCK_ASSERT(sc); 1054 1055 if (bif->bif_flags & IFBIF_STP) 1056 bstp_disable(&bif->bif_stp); 1057 1058 ifs->if_bridge = NULL; 1059 CK_LIST_REMOVE(bif, bif_next); 1060 1061 /* 1062 * If removing the interface that gave the bridge its mac address, set 1063 * the mac address of the bridge to the address of the next member, or 1064 * to its default address if no members are left. 1065 */ 1066 if (V_bridge_inherit_mac && sc->sc_ifaddr == ifs) { 1067 if (CK_LIST_EMPTY(&sc->sc_iflist)) { 1068 bcopy(&sc->sc_defaddr, 1069 IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN); 1070 sc->sc_ifaddr = NULL; 1071 } else { 1072 bifl = CK_LIST_FIRST(&sc->sc_iflist); 1073 fif = bifl->bif_ifp; 1074 bcopy(IF_LLADDR(fif), 1075 IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN); 1076 sc->sc_ifaddr = fif; 1077 } 1078 EVENTHANDLER_INVOKE(iflladdr_event, sc->sc_ifp); 1079 } 1080 1081 bridge_linkcheck(sc); 1082 bridge_mutecaps(sc); /* recalcuate now this interface is removed */ 1083 BRIDGE_RT_LOCK(sc); 1084 bridge_rtdelete(sc, ifs, IFBF_FLUSHALL); 1085 BRIDGE_RT_UNLOCK(sc); 1086 KASSERT(bif->bif_addrcnt == 0, 1087 ("%s: %d bridge routes referenced", __func__, bif->bif_addrcnt)); 1088 1089 ifs->if_bridge_output = NULL; 1090 ifs->if_bridge_input = NULL; 1091 ifs->if_bridge_linkstate = NULL; 1092 if (!gone) { 1093 switch (ifs->if_type) { 1094 case IFT_ETHER: 1095 case IFT_L2VLAN: 1096 /* 1097 * Take the interface out of promiscuous mode, but only 1098 * if it was promiscuous in the first place. It might 1099 * not be if we're in the bridge_ioctl_add() error path. 1100 */ 1101 if (ifs->if_flags & IFF_PROMISC) 1102 (void) ifpromisc(ifs, 0); 1103 break; 1104 1105 case IFT_GIF: 1106 break; 1107 1108 default: 1109 #ifdef DIAGNOSTIC 1110 panic("bridge_delete_member: impossible"); 1111 #endif 1112 break; 1113 } 1114 /* reneable any interface capabilities */ 1115 bridge_set_ifcap(sc, bif, bif->bif_savedcaps); 1116 } 1117 bstp_destroy(&bif->bif_stp); /* prepare to free */ 1118 1119 NET_EPOCH_CALL(bridge_delete_member_cb, &bif->bif_epoch_ctx); 1120 } 1121 1122 /* 1123 * bridge_delete_span: 1124 * 1125 * Delete the specified span interface. 1126 */ 1127 static void 1128 bridge_delete_span(struct bridge_softc *sc, struct bridge_iflist *bif) 1129 { 1130 BRIDGE_LOCK_ASSERT(sc); 1131 1132 KASSERT(bif->bif_ifp->if_bridge == NULL, 1133 ("%s: not a span interface", __func__)); 1134 1135 CK_LIST_REMOVE(bif, bif_next); 1136 1137 NET_EPOCH_CALL(bridge_delete_member_cb, &bif->bif_epoch_ctx); 1138 } 1139 1140 static int 1141 bridge_ioctl_add(struct bridge_softc *sc, void *arg) 1142 { 1143 struct ifbreq *req = arg; 1144 struct bridge_iflist *bif = NULL; 1145 struct ifnet *ifs; 1146 int error = 0; 1147 1148 ifs = ifunit(req->ifbr_ifsname); 1149 if (ifs == NULL) 1150 return (ENOENT); 1151 if (ifs->if_ioctl == NULL) /* must be supported */ 1152 return (EINVAL); 1153 1154 /* If it's in the span list, it can't be a member. */ 1155 CK_LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1156 if (ifs == bif->bif_ifp) 1157 return (EBUSY); 1158 1159 if (ifs->if_bridge == sc) 1160 return (EEXIST); 1161 1162 if (ifs->if_bridge != NULL) 1163 return (EBUSY); 1164 1165 switch (ifs->if_type) { 1166 case IFT_ETHER: 1167 case IFT_L2VLAN: 1168 case IFT_GIF: 1169 /* permitted interface types */ 1170 break; 1171 default: 1172 return (EINVAL); 1173 } 1174 1175 #ifdef INET6 1176 /* 1177 * Two valid inet6 addresses with link-local scope must not be 1178 * on the parent interface and the member interfaces at the 1179 * same time. This restriction is needed to prevent violation 1180 * of link-local scope zone. Attempts to add a member 1181 * interface which has inet6 addresses when the parent has 1182 * inet6 triggers removal of all inet6 addresses on the member 1183 * interface. 1184 */ 1185 1186 /* Check if the parent interface has a link-local scope addr. */ 1187 if (V_allow_llz_overlap == 0 && 1188 in6ifa_llaonifp(sc->sc_ifp) != NULL) { 1189 /* 1190 * If any, remove all inet6 addresses from the member 1191 * interfaces. 1192 */ 1193 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1194 if (in6ifa_llaonifp(bif->bif_ifp)) { 1195 in6_ifdetach(bif->bif_ifp); 1196 if_printf(sc->sc_ifp, 1197 "IPv6 addresses on %s have been removed " 1198 "before adding it as a member to prevent " 1199 "IPv6 address scope violation.\n", 1200 bif->bif_ifp->if_xname); 1201 } 1202 } 1203 if (in6ifa_llaonifp(ifs)) { 1204 in6_ifdetach(ifs); 1205 if_printf(sc->sc_ifp, 1206 "IPv6 addresses on %s have been removed " 1207 "before adding it as a member to prevent " 1208 "IPv6 address scope violation.\n", 1209 ifs->if_xname); 1210 } 1211 } 1212 #endif 1213 /* Allow the first Ethernet member to define the MTU */ 1214 if (CK_LIST_EMPTY(&sc->sc_iflist)) 1215 sc->sc_ifp->if_mtu = ifs->if_mtu; 1216 else if (sc->sc_ifp->if_mtu != ifs->if_mtu) { 1217 if_printf(sc->sc_ifp, "invalid MTU: %u(%s) != %u\n", 1218 ifs->if_mtu, ifs->if_xname, sc->sc_ifp->if_mtu); 1219 return (EINVAL); 1220 } 1221 1222 bif = malloc(sizeof(*bif), M_DEVBUF, M_NOWAIT|M_ZERO); 1223 if (bif == NULL) 1224 return (ENOMEM); 1225 1226 bif->bif_ifp = ifs; 1227 bif->bif_flags = IFBIF_LEARNING | IFBIF_DISCOVER; 1228 bif->bif_savedcaps = ifs->if_capenable; 1229 1230 /* 1231 * Assign the interface's MAC address to the bridge if it's the first 1232 * member and the MAC address of the bridge has not been changed from 1233 * the default randomly generated one. 1234 */ 1235 if (V_bridge_inherit_mac && CK_LIST_EMPTY(&sc->sc_iflist) && 1236 !memcmp(IF_LLADDR(sc->sc_ifp), sc->sc_defaddr.octet, ETHER_ADDR_LEN)) { 1237 bcopy(IF_LLADDR(ifs), IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN); 1238 sc->sc_ifaddr = ifs; 1239 EVENTHANDLER_INVOKE(iflladdr_event, sc->sc_ifp); 1240 } 1241 1242 ifs->if_bridge = sc; 1243 ifs->if_bridge_output = bridge_output; 1244 ifs->if_bridge_input = bridge_input; 1245 ifs->if_bridge_linkstate = bridge_linkstate; 1246 bstp_create(&sc->sc_stp, &bif->bif_stp, bif->bif_ifp); 1247 /* 1248 * XXX: XLOCK HERE!?! 1249 * 1250 * NOTE: insert_***HEAD*** should be safe for the traversals. 1251 */ 1252 CK_LIST_INSERT_HEAD(&sc->sc_iflist, bif, bif_next); 1253 1254 /* Set interface capabilities to the intersection set of all members */ 1255 bridge_mutecaps(sc); 1256 bridge_linkcheck(sc); 1257 1258 /* Place the interface into promiscuous mode */ 1259 switch (ifs->if_type) { 1260 case IFT_ETHER: 1261 case IFT_L2VLAN: 1262 error = ifpromisc(ifs, 1); 1263 break; 1264 } 1265 1266 if (error) 1267 bridge_delete_member(sc, bif, 0); 1268 return (error); 1269 } 1270 1271 static int 1272 bridge_ioctl_del(struct bridge_softc *sc, void *arg) 1273 { 1274 struct ifbreq *req = arg; 1275 struct bridge_iflist *bif; 1276 1277 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1278 if (bif == NULL) 1279 return (ENOENT); 1280 1281 bridge_delete_member(sc, bif, 0); 1282 1283 return (0); 1284 } 1285 1286 static int 1287 bridge_ioctl_gifflags(struct bridge_softc *sc, void *arg) 1288 { 1289 struct ifbreq *req = arg; 1290 struct bridge_iflist *bif; 1291 struct bstp_port *bp; 1292 1293 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1294 if (bif == NULL) 1295 return (ENOENT); 1296 1297 bp = &bif->bif_stp; 1298 req->ifbr_ifsflags = bif->bif_flags; 1299 req->ifbr_state = bp->bp_state; 1300 req->ifbr_priority = bp->bp_priority; 1301 req->ifbr_path_cost = bp->bp_path_cost; 1302 req->ifbr_portno = bif->bif_ifp->if_index & 0xfff; 1303 req->ifbr_proto = bp->bp_protover; 1304 req->ifbr_role = bp->bp_role; 1305 req->ifbr_stpflags = bp->bp_flags; 1306 req->ifbr_addrcnt = bif->bif_addrcnt; 1307 req->ifbr_addrmax = bif->bif_addrmax; 1308 req->ifbr_addrexceeded = bif->bif_addrexceeded; 1309 1310 /* Copy STP state options as flags */ 1311 if (bp->bp_operedge) 1312 req->ifbr_ifsflags |= IFBIF_BSTP_EDGE; 1313 if (bp->bp_flags & BSTP_PORT_AUTOEDGE) 1314 req->ifbr_ifsflags |= IFBIF_BSTP_AUTOEDGE; 1315 if (bp->bp_ptp_link) 1316 req->ifbr_ifsflags |= IFBIF_BSTP_PTP; 1317 if (bp->bp_flags & BSTP_PORT_AUTOPTP) 1318 req->ifbr_ifsflags |= IFBIF_BSTP_AUTOPTP; 1319 if (bp->bp_flags & BSTP_PORT_ADMEDGE) 1320 req->ifbr_ifsflags |= IFBIF_BSTP_ADMEDGE; 1321 if (bp->bp_flags & BSTP_PORT_ADMCOST) 1322 req->ifbr_ifsflags |= IFBIF_BSTP_ADMCOST; 1323 return (0); 1324 } 1325 1326 static int 1327 bridge_ioctl_sifflags(struct bridge_softc *sc, void *arg) 1328 { 1329 struct ifbreq *req = arg; 1330 struct bridge_iflist *bif; 1331 struct bstp_port *bp; 1332 int error; 1333 1334 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1335 if (bif == NULL) 1336 return (ENOENT); 1337 bp = &bif->bif_stp; 1338 1339 if (req->ifbr_ifsflags & IFBIF_SPAN) 1340 /* SPAN is readonly */ 1341 return (EINVAL); 1342 1343 if (req->ifbr_ifsflags & IFBIF_STP) { 1344 if ((bif->bif_flags & IFBIF_STP) == 0) { 1345 error = bstp_enable(&bif->bif_stp); 1346 if (error) 1347 return (error); 1348 } 1349 } else { 1350 if ((bif->bif_flags & IFBIF_STP) != 0) 1351 bstp_disable(&bif->bif_stp); 1352 } 1353 1354 /* Pass on STP flags */ 1355 bstp_set_edge(bp, req->ifbr_ifsflags & IFBIF_BSTP_EDGE ? 1 : 0); 1356 bstp_set_autoedge(bp, req->ifbr_ifsflags & IFBIF_BSTP_AUTOEDGE ? 1 : 0); 1357 bstp_set_ptp(bp, req->ifbr_ifsflags & IFBIF_BSTP_PTP ? 1 : 0); 1358 bstp_set_autoptp(bp, req->ifbr_ifsflags & IFBIF_BSTP_AUTOPTP ? 1 : 0); 1359 1360 /* Save the bits relating to the bridge */ 1361 bif->bif_flags = req->ifbr_ifsflags & IFBIFMASK; 1362 1363 return (0); 1364 } 1365 1366 static int 1367 bridge_ioctl_scache(struct bridge_softc *sc, void *arg) 1368 { 1369 struct ifbrparam *param = arg; 1370 1371 sc->sc_brtmax = param->ifbrp_csize; 1372 bridge_rttrim(sc); 1373 1374 return (0); 1375 } 1376 1377 static int 1378 bridge_ioctl_gcache(struct bridge_softc *sc, void *arg) 1379 { 1380 struct ifbrparam *param = arg; 1381 1382 param->ifbrp_csize = sc->sc_brtmax; 1383 1384 return (0); 1385 } 1386 1387 static int 1388 bridge_ioctl_gifs(struct bridge_softc *sc, void *arg) 1389 { 1390 struct ifbifconf *bifc = arg; 1391 struct bridge_iflist *bif; 1392 struct ifbreq breq; 1393 char *buf, *outbuf; 1394 int count, buflen, len, error = 0; 1395 1396 count = 0; 1397 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) 1398 count++; 1399 CK_LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1400 count++; 1401 1402 buflen = sizeof(breq) * count; 1403 if (bifc->ifbic_len == 0) { 1404 bifc->ifbic_len = buflen; 1405 return (0); 1406 } 1407 outbuf = malloc(buflen, M_TEMP, M_NOWAIT | M_ZERO); 1408 if (outbuf == NULL) 1409 return (ENOMEM); 1410 1411 count = 0; 1412 buf = outbuf; 1413 len = min(bifc->ifbic_len, buflen); 1414 bzero(&breq, sizeof(breq)); 1415 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1416 if (len < sizeof(breq)) 1417 break; 1418 1419 strlcpy(breq.ifbr_ifsname, bif->bif_ifp->if_xname, 1420 sizeof(breq.ifbr_ifsname)); 1421 /* Fill in the ifbreq structure */ 1422 error = bridge_ioctl_gifflags(sc, &breq); 1423 if (error) 1424 break; 1425 memcpy(buf, &breq, sizeof(breq)); 1426 count++; 1427 buf += sizeof(breq); 1428 len -= sizeof(breq); 1429 } 1430 CK_LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) { 1431 if (len < sizeof(breq)) 1432 break; 1433 1434 strlcpy(breq.ifbr_ifsname, bif->bif_ifp->if_xname, 1435 sizeof(breq.ifbr_ifsname)); 1436 breq.ifbr_ifsflags = bif->bif_flags; 1437 breq.ifbr_portno = bif->bif_ifp->if_index & 0xfff; 1438 memcpy(buf, &breq, sizeof(breq)); 1439 count++; 1440 buf += sizeof(breq); 1441 len -= sizeof(breq); 1442 } 1443 1444 bifc->ifbic_len = sizeof(breq) * count; 1445 error = copyout(outbuf, bifc->ifbic_req, bifc->ifbic_len); 1446 free(outbuf, M_TEMP); 1447 return (error); 1448 } 1449 1450 static int 1451 bridge_ioctl_rts(struct bridge_softc *sc, void *arg) 1452 { 1453 struct ifbaconf *bac = arg; 1454 struct bridge_rtnode *brt; 1455 struct ifbareq bareq; 1456 char *buf, *outbuf; 1457 int count, buflen, len, error = 0; 1458 1459 if (bac->ifbac_len == 0) 1460 return (0); 1461 1462 count = 0; 1463 CK_LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) 1464 count++; 1465 buflen = sizeof(bareq) * count; 1466 1467 outbuf = malloc(buflen, M_TEMP, M_NOWAIT | M_ZERO); 1468 if (outbuf == NULL) 1469 return (ENOMEM); 1470 1471 count = 0; 1472 buf = outbuf; 1473 len = min(bac->ifbac_len, buflen); 1474 bzero(&bareq, sizeof(bareq)); 1475 CK_LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) { 1476 if (len < sizeof(bareq)) 1477 goto out; 1478 strlcpy(bareq.ifba_ifsname, brt->brt_ifp->if_xname, 1479 sizeof(bareq.ifba_ifsname)); 1480 memcpy(bareq.ifba_dst, brt->brt_addr, sizeof(brt->brt_addr)); 1481 bareq.ifba_vlan = brt->brt_vlan; 1482 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 1483 time_uptime < brt->brt_expire) 1484 bareq.ifba_expire = brt->brt_expire - time_uptime; 1485 else 1486 bareq.ifba_expire = 0; 1487 bareq.ifba_flags = brt->brt_flags; 1488 1489 memcpy(buf, &bareq, sizeof(bareq)); 1490 count++; 1491 buf += sizeof(bareq); 1492 len -= sizeof(bareq); 1493 } 1494 out: 1495 bac->ifbac_len = sizeof(bareq) * count; 1496 error = copyout(outbuf, bac->ifbac_req, bac->ifbac_len); 1497 free(outbuf, M_TEMP); 1498 return (error); 1499 } 1500 1501 static int 1502 bridge_ioctl_saddr(struct bridge_softc *sc, void *arg) 1503 { 1504 struct ifbareq *req = arg; 1505 struct bridge_iflist *bif; 1506 struct epoch_tracker et; 1507 int error; 1508 1509 NET_EPOCH_ENTER(et); 1510 bif = bridge_lookup_member(sc, req->ifba_ifsname); 1511 if (bif == NULL) { 1512 NET_EPOCH_EXIT(et); 1513 return (ENOENT); 1514 } 1515 1516 /* bridge_rtupdate() may acquire the lock. */ 1517 error = bridge_rtupdate(sc, req->ifba_dst, req->ifba_vlan, bif, 1, 1518 req->ifba_flags); 1519 NET_EPOCH_EXIT(et); 1520 1521 return (error); 1522 } 1523 1524 static int 1525 bridge_ioctl_sto(struct bridge_softc *sc, void *arg) 1526 { 1527 struct ifbrparam *param = arg; 1528 1529 sc->sc_brttimeout = param->ifbrp_ctime; 1530 return (0); 1531 } 1532 1533 static int 1534 bridge_ioctl_gto(struct bridge_softc *sc, void *arg) 1535 { 1536 struct ifbrparam *param = arg; 1537 1538 param->ifbrp_ctime = sc->sc_brttimeout; 1539 return (0); 1540 } 1541 1542 static int 1543 bridge_ioctl_daddr(struct bridge_softc *sc, void *arg) 1544 { 1545 struct ifbareq *req = arg; 1546 1547 return (bridge_rtdaddr(sc, req->ifba_dst, req->ifba_vlan)); 1548 } 1549 1550 static int 1551 bridge_ioctl_flush(struct bridge_softc *sc, void *arg) 1552 { 1553 struct ifbreq *req = arg; 1554 1555 BRIDGE_RT_LOCK(sc); 1556 bridge_rtflush(sc, req->ifbr_ifsflags); 1557 BRIDGE_RT_UNLOCK(sc); 1558 1559 return (0); 1560 } 1561 1562 static int 1563 bridge_ioctl_gpri(struct bridge_softc *sc, void *arg) 1564 { 1565 struct ifbrparam *param = arg; 1566 struct bstp_state *bs = &sc->sc_stp; 1567 1568 param->ifbrp_prio = bs->bs_bridge_priority; 1569 return (0); 1570 } 1571 1572 static int 1573 bridge_ioctl_spri(struct bridge_softc *sc, void *arg) 1574 { 1575 struct ifbrparam *param = arg; 1576 1577 return (bstp_set_priority(&sc->sc_stp, param->ifbrp_prio)); 1578 } 1579 1580 static int 1581 bridge_ioctl_ght(struct bridge_softc *sc, void *arg) 1582 { 1583 struct ifbrparam *param = arg; 1584 struct bstp_state *bs = &sc->sc_stp; 1585 1586 param->ifbrp_hellotime = bs->bs_bridge_htime >> 8; 1587 return (0); 1588 } 1589 1590 static int 1591 bridge_ioctl_sht(struct bridge_softc *sc, void *arg) 1592 { 1593 struct ifbrparam *param = arg; 1594 1595 return (bstp_set_htime(&sc->sc_stp, param->ifbrp_hellotime)); 1596 } 1597 1598 static int 1599 bridge_ioctl_gfd(struct bridge_softc *sc, void *arg) 1600 { 1601 struct ifbrparam *param = arg; 1602 struct bstp_state *bs = &sc->sc_stp; 1603 1604 param->ifbrp_fwddelay = bs->bs_bridge_fdelay >> 8; 1605 return (0); 1606 } 1607 1608 static int 1609 bridge_ioctl_sfd(struct bridge_softc *sc, void *arg) 1610 { 1611 struct ifbrparam *param = arg; 1612 1613 return (bstp_set_fdelay(&sc->sc_stp, param->ifbrp_fwddelay)); 1614 } 1615 1616 static int 1617 bridge_ioctl_gma(struct bridge_softc *sc, void *arg) 1618 { 1619 struct ifbrparam *param = arg; 1620 struct bstp_state *bs = &sc->sc_stp; 1621 1622 param->ifbrp_maxage = bs->bs_bridge_max_age >> 8; 1623 return (0); 1624 } 1625 1626 static int 1627 bridge_ioctl_sma(struct bridge_softc *sc, void *arg) 1628 { 1629 struct ifbrparam *param = arg; 1630 1631 return (bstp_set_maxage(&sc->sc_stp, param->ifbrp_maxage)); 1632 } 1633 1634 static int 1635 bridge_ioctl_sifprio(struct bridge_softc *sc, void *arg) 1636 { 1637 struct ifbreq *req = arg; 1638 struct bridge_iflist *bif; 1639 1640 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1641 if (bif == NULL) 1642 return (ENOENT); 1643 1644 return (bstp_set_port_priority(&bif->bif_stp, req->ifbr_priority)); 1645 } 1646 1647 static int 1648 bridge_ioctl_sifcost(struct bridge_softc *sc, void *arg) 1649 { 1650 struct ifbreq *req = arg; 1651 struct bridge_iflist *bif; 1652 1653 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1654 if (bif == NULL) 1655 return (ENOENT); 1656 1657 return (bstp_set_path_cost(&bif->bif_stp, req->ifbr_path_cost)); 1658 } 1659 1660 static int 1661 bridge_ioctl_sifmaxaddr(struct bridge_softc *sc, void *arg) 1662 { 1663 struct ifbreq *req = arg; 1664 struct bridge_iflist *bif; 1665 1666 bif = bridge_lookup_member(sc, req->ifbr_ifsname); 1667 if (bif == NULL) 1668 return (ENOENT); 1669 1670 bif->bif_addrmax = req->ifbr_addrmax; 1671 return (0); 1672 } 1673 1674 static int 1675 bridge_ioctl_addspan(struct bridge_softc *sc, void *arg) 1676 { 1677 struct ifbreq *req = arg; 1678 struct bridge_iflist *bif = NULL; 1679 struct ifnet *ifs; 1680 1681 ifs = ifunit(req->ifbr_ifsname); 1682 if (ifs == NULL) 1683 return (ENOENT); 1684 1685 CK_LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1686 if (ifs == bif->bif_ifp) 1687 return (EBUSY); 1688 1689 if (ifs->if_bridge != NULL) 1690 return (EBUSY); 1691 1692 switch (ifs->if_type) { 1693 case IFT_ETHER: 1694 case IFT_GIF: 1695 case IFT_L2VLAN: 1696 break; 1697 default: 1698 return (EINVAL); 1699 } 1700 1701 bif = malloc(sizeof(*bif), M_DEVBUF, M_NOWAIT|M_ZERO); 1702 if (bif == NULL) 1703 return (ENOMEM); 1704 1705 bif->bif_ifp = ifs; 1706 bif->bif_flags = IFBIF_SPAN; 1707 1708 CK_LIST_INSERT_HEAD(&sc->sc_spanlist, bif, bif_next); 1709 1710 return (0); 1711 } 1712 1713 static int 1714 bridge_ioctl_delspan(struct bridge_softc *sc, void *arg) 1715 { 1716 struct ifbreq *req = arg; 1717 struct bridge_iflist *bif; 1718 struct ifnet *ifs; 1719 1720 ifs = ifunit(req->ifbr_ifsname); 1721 if (ifs == NULL) 1722 return (ENOENT); 1723 1724 CK_LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1725 if (ifs == bif->bif_ifp) 1726 break; 1727 1728 if (bif == NULL) 1729 return (ENOENT); 1730 1731 bridge_delete_span(sc, bif); 1732 1733 return (0); 1734 } 1735 1736 static int 1737 bridge_ioctl_gbparam(struct bridge_softc *sc, void *arg) 1738 { 1739 struct ifbropreq *req = arg; 1740 struct bstp_state *bs = &sc->sc_stp; 1741 struct bstp_port *root_port; 1742 1743 req->ifbop_maxage = bs->bs_bridge_max_age >> 8; 1744 req->ifbop_hellotime = bs->bs_bridge_htime >> 8; 1745 req->ifbop_fwddelay = bs->bs_bridge_fdelay >> 8; 1746 1747 root_port = bs->bs_root_port; 1748 if (root_port == NULL) 1749 req->ifbop_root_port = 0; 1750 else 1751 req->ifbop_root_port = root_port->bp_ifp->if_index; 1752 1753 req->ifbop_holdcount = bs->bs_txholdcount; 1754 req->ifbop_priority = bs->bs_bridge_priority; 1755 req->ifbop_protocol = bs->bs_protover; 1756 req->ifbop_root_path_cost = bs->bs_root_pv.pv_cost; 1757 req->ifbop_bridgeid = bs->bs_bridge_pv.pv_dbridge_id; 1758 req->ifbop_designated_root = bs->bs_root_pv.pv_root_id; 1759 req->ifbop_designated_bridge = bs->bs_root_pv.pv_dbridge_id; 1760 req->ifbop_last_tc_time.tv_sec = bs->bs_last_tc_time.tv_sec; 1761 req->ifbop_last_tc_time.tv_usec = bs->bs_last_tc_time.tv_usec; 1762 1763 return (0); 1764 } 1765 1766 static int 1767 bridge_ioctl_grte(struct bridge_softc *sc, void *arg) 1768 { 1769 struct ifbrparam *param = arg; 1770 1771 param->ifbrp_cexceeded = sc->sc_brtexceeded; 1772 return (0); 1773 } 1774 1775 static int 1776 bridge_ioctl_gifsstp(struct bridge_softc *sc, void *arg) 1777 { 1778 struct ifbpstpconf *bifstp = arg; 1779 struct bridge_iflist *bif; 1780 struct bstp_port *bp; 1781 struct ifbpstpreq bpreq; 1782 char *buf, *outbuf; 1783 int count, buflen, len, error = 0; 1784 1785 count = 0; 1786 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1787 if ((bif->bif_flags & IFBIF_STP) != 0) 1788 count++; 1789 } 1790 1791 buflen = sizeof(bpreq) * count; 1792 if (bifstp->ifbpstp_len == 0) { 1793 bifstp->ifbpstp_len = buflen; 1794 return (0); 1795 } 1796 1797 outbuf = malloc(buflen, M_TEMP, M_NOWAIT | M_ZERO); 1798 if (outbuf == NULL) 1799 return (ENOMEM); 1800 1801 count = 0; 1802 buf = outbuf; 1803 len = min(bifstp->ifbpstp_len, buflen); 1804 bzero(&bpreq, sizeof(bpreq)); 1805 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 1806 if (len < sizeof(bpreq)) 1807 break; 1808 1809 if ((bif->bif_flags & IFBIF_STP) == 0) 1810 continue; 1811 1812 bp = &bif->bif_stp; 1813 bpreq.ifbp_portno = bif->bif_ifp->if_index & 0xfff; 1814 bpreq.ifbp_fwd_trans = bp->bp_forward_transitions; 1815 bpreq.ifbp_design_cost = bp->bp_desg_pv.pv_cost; 1816 bpreq.ifbp_design_port = bp->bp_desg_pv.pv_port_id; 1817 bpreq.ifbp_design_bridge = bp->bp_desg_pv.pv_dbridge_id; 1818 bpreq.ifbp_design_root = bp->bp_desg_pv.pv_root_id; 1819 1820 memcpy(buf, &bpreq, sizeof(bpreq)); 1821 count++; 1822 buf += sizeof(bpreq); 1823 len -= sizeof(bpreq); 1824 } 1825 1826 bifstp->ifbpstp_len = sizeof(bpreq) * count; 1827 error = copyout(outbuf, bifstp->ifbpstp_req, bifstp->ifbpstp_len); 1828 free(outbuf, M_TEMP); 1829 return (error); 1830 } 1831 1832 static int 1833 bridge_ioctl_sproto(struct bridge_softc *sc, void *arg) 1834 { 1835 struct ifbrparam *param = arg; 1836 1837 return (bstp_set_protocol(&sc->sc_stp, param->ifbrp_proto)); 1838 } 1839 1840 static int 1841 bridge_ioctl_stxhc(struct bridge_softc *sc, void *arg) 1842 { 1843 struct ifbrparam *param = arg; 1844 1845 return (bstp_set_holdcount(&sc->sc_stp, param->ifbrp_txhc)); 1846 } 1847 1848 /* 1849 * bridge_ifdetach: 1850 * 1851 * Detach an interface from a bridge. Called when a member 1852 * interface is detaching. 1853 */ 1854 static void 1855 bridge_ifdetach(void *arg __unused, struct ifnet *ifp) 1856 { 1857 struct bridge_softc *sc = ifp->if_bridge; 1858 struct bridge_iflist *bif; 1859 1860 if (ifp->if_flags & IFF_RENAMING) 1861 return; 1862 if (V_bridge_cloner == NULL) { 1863 /* 1864 * This detach handler can be called after 1865 * vnet_bridge_uninit(). Just return in that case. 1866 */ 1867 return; 1868 } 1869 /* Check if the interface is a bridge member */ 1870 if (sc != NULL) { 1871 BRIDGE_LOCK(sc); 1872 1873 bif = bridge_lookup_member_if(sc, ifp); 1874 if (bif != NULL) 1875 bridge_delete_member(sc, bif, 1); 1876 1877 BRIDGE_UNLOCK(sc); 1878 return; 1879 } 1880 1881 /* Check if the interface is a span port */ 1882 BRIDGE_LIST_LOCK(); 1883 LIST_FOREACH(sc, &V_bridge_list, sc_list) { 1884 BRIDGE_LOCK(sc); 1885 CK_LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) 1886 if (ifp == bif->bif_ifp) { 1887 bridge_delete_span(sc, bif); 1888 break; 1889 } 1890 1891 BRIDGE_UNLOCK(sc); 1892 } 1893 BRIDGE_LIST_UNLOCK(); 1894 } 1895 1896 /* 1897 * bridge_init: 1898 * 1899 * Initialize a bridge interface. 1900 */ 1901 static void 1902 bridge_init(void *xsc) 1903 { 1904 struct bridge_softc *sc = (struct bridge_softc *)xsc; 1905 struct ifnet *ifp = sc->sc_ifp; 1906 1907 if (ifp->if_drv_flags & IFF_DRV_RUNNING) 1908 return; 1909 1910 BRIDGE_LOCK(sc); 1911 callout_reset(&sc->sc_brcallout, bridge_rtable_prune_period * hz, 1912 bridge_timer, sc); 1913 1914 ifp->if_drv_flags |= IFF_DRV_RUNNING; 1915 bstp_init(&sc->sc_stp); /* Initialize Spanning Tree */ 1916 1917 BRIDGE_UNLOCK(sc); 1918 } 1919 1920 /* 1921 * bridge_stop: 1922 * 1923 * Stop the bridge interface. 1924 */ 1925 static void 1926 bridge_stop(struct ifnet *ifp, int disable) 1927 { 1928 struct bridge_softc *sc = ifp->if_softc; 1929 1930 BRIDGE_LOCK_ASSERT(sc); 1931 1932 if ((ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) 1933 return; 1934 1935 BRIDGE_RT_LOCK(sc); 1936 callout_stop(&sc->sc_brcallout); 1937 1938 bstp_stop(&sc->sc_stp); 1939 1940 bridge_rtflush(sc, IFBF_FLUSHDYN); 1941 BRIDGE_RT_UNLOCK(sc); 1942 1943 ifp->if_drv_flags &= ~IFF_DRV_RUNNING; 1944 } 1945 1946 /* 1947 * bridge_enqueue: 1948 * 1949 * Enqueue a packet on a bridge member interface. 1950 * 1951 */ 1952 static int 1953 bridge_enqueue(struct bridge_softc *sc, struct ifnet *dst_ifp, struct mbuf *m) 1954 { 1955 int len, err = 0; 1956 short mflags; 1957 struct mbuf *m0; 1958 1959 /* We may be sending a fragment so traverse the mbuf */ 1960 for (; m; m = m0) { 1961 m0 = m->m_nextpkt; 1962 m->m_nextpkt = NULL; 1963 len = m->m_pkthdr.len; 1964 mflags = m->m_flags; 1965 1966 /* 1967 * If underlying interface can not do VLAN tag insertion itself 1968 * then attach a packet tag that holds it. 1969 */ 1970 if ((m->m_flags & M_VLANTAG) && 1971 (dst_ifp->if_capenable & IFCAP_VLAN_HWTAGGING) == 0) { 1972 m = ether_vlanencap(m, m->m_pkthdr.ether_vtag); 1973 if (m == NULL) { 1974 if_printf(dst_ifp, 1975 "unable to prepend VLAN header\n"); 1976 if_inc_counter(dst_ifp, IFCOUNTER_OERRORS, 1); 1977 continue; 1978 } 1979 m->m_flags &= ~M_VLANTAG; 1980 } 1981 1982 M_ASSERTPKTHDR(m); /* We shouldn't transmit mbuf without pkthdr */ 1983 if ((err = dst_ifp->if_transmit(dst_ifp, m))) { 1984 m_freem(m0); 1985 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 1986 break; 1987 } 1988 1989 if_inc_counter(sc->sc_ifp, IFCOUNTER_OPACKETS, 1); 1990 if_inc_counter(sc->sc_ifp, IFCOUNTER_OBYTES, len); 1991 if (mflags & M_MCAST) 1992 if_inc_counter(sc->sc_ifp, IFCOUNTER_OMCASTS, 1); 1993 } 1994 1995 return (err); 1996 } 1997 1998 /* 1999 * bridge_dummynet: 2000 * 2001 * Receive a queued packet from dummynet and pass it on to the output 2002 * interface. 2003 * 2004 * The mbuf has the Ethernet header already attached. 2005 */ 2006 static void 2007 bridge_dummynet(struct mbuf *m, struct ifnet *ifp) 2008 { 2009 struct bridge_softc *sc; 2010 2011 sc = ifp->if_bridge; 2012 2013 /* 2014 * The packet didnt originate from a member interface. This should only 2015 * ever happen if a member interface is removed while packets are 2016 * queued for it. 2017 */ 2018 if (sc == NULL) { 2019 m_freem(m); 2020 return; 2021 } 2022 2023 if (PFIL_HOOKED_OUT(V_inet_pfil_head) 2024 #ifdef INET6 2025 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 2026 #endif 2027 ) { 2028 if (bridge_pfil(&m, sc->sc_ifp, ifp, PFIL_OUT) != 0) 2029 return; 2030 if (m == NULL) 2031 return; 2032 } 2033 2034 bridge_enqueue(sc, ifp, m); 2035 } 2036 2037 /* 2038 * bridge_output: 2039 * 2040 * Send output from a bridge member interface. This 2041 * performs the bridging function for locally originated 2042 * packets. 2043 * 2044 * The mbuf has the Ethernet header already attached. We must 2045 * enqueue or free the mbuf before returning. 2046 */ 2047 static int 2048 bridge_output(struct ifnet *ifp, struct mbuf *m, struct sockaddr *sa, 2049 struct rtentry *rt) 2050 { 2051 struct ether_header *eh; 2052 struct ifnet *bifp, *dst_if; 2053 struct bridge_softc *sc; 2054 uint16_t vlan; 2055 2056 NET_EPOCH_ASSERT(); 2057 2058 if (m->m_len < ETHER_HDR_LEN) { 2059 m = m_pullup(m, ETHER_HDR_LEN); 2060 if (m == NULL) 2061 return (0); 2062 } 2063 2064 eh = mtod(m, struct ether_header *); 2065 sc = ifp->if_bridge; 2066 vlan = VLANTAGOF(m); 2067 2068 bifp = sc->sc_ifp; 2069 2070 /* 2071 * If bridge is down, but the original output interface is up, 2072 * go ahead and send out that interface. Otherwise, the packet 2073 * is dropped below. 2074 */ 2075 if ((bifp->if_drv_flags & IFF_DRV_RUNNING) == 0) { 2076 dst_if = ifp; 2077 goto sendunicast; 2078 } 2079 2080 /* 2081 * If the packet is a multicast, or we don't know a better way to 2082 * get there, send to all interfaces. 2083 */ 2084 if (ETHER_IS_MULTICAST(eh->ether_dhost)) 2085 dst_if = NULL; 2086 else 2087 dst_if = bridge_rtlookup(sc, eh->ether_dhost, vlan); 2088 /* Tap any traffic not passing back out the originating interface */ 2089 if (dst_if != ifp) 2090 ETHER_BPF_MTAP(bifp, m); 2091 if (dst_if == NULL) { 2092 struct bridge_iflist *bif; 2093 struct mbuf *mc; 2094 int used = 0; 2095 2096 bridge_span(sc, m); 2097 2098 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 2099 dst_if = bif->bif_ifp; 2100 2101 if (dst_if->if_type == IFT_GIF) 2102 continue; 2103 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2104 continue; 2105 2106 /* 2107 * If this is not the original output interface, 2108 * and the interface is participating in spanning 2109 * tree, make sure the port is in a state that 2110 * allows forwarding. 2111 */ 2112 if (dst_if != ifp && (bif->bif_flags & IFBIF_STP) && 2113 bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2114 continue; 2115 2116 if (CK_LIST_NEXT(bif, bif_next) == NULL) { 2117 used = 1; 2118 mc = m; 2119 } else { 2120 mc = m_copypacket(m, M_NOWAIT); 2121 if (mc == NULL) { 2122 if_inc_counter(bifp, IFCOUNTER_OERRORS, 1); 2123 continue; 2124 } 2125 } 2126 2127 bridge_enqueue(sc, dst_if, mc); 2128 } 2129 if (used == 0) 2130 m_freem(m); 2131 return (0); 2132 } 2133 2134 sendunicast: 2135 /* 2136 * XXX Spanning tree consideration here? 2137 */ 2138 2139 bridge_span(sc, m); 2140 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) { 2141 m_freem(m); 2142 return (0); 2143 } 2144 2145 bridge_enqueue(sc, dst_if, m); 2146 return (0); 2147 } 2148 2149 /* 2150 * bridge_transmit: 2151 * 2152 * Do output on a bridge. 2153 * 2154 */ 2155 static int 2156 bridge_transmit(struct ifnet *ifp, struct mbuf *m) 2157 { 2158 struct bridge_softc *sc; 2159 struct ether_header *eh; 2160 struct ifnet *dst_if; 2161 int error = 0; 2162 2163 sc = ifp->if_softc; 2164 2165 ETHER_BPF_MTAP(ifp, m); 2166 2167 eh = mtod(m, struct ether_header *); 2168 2169 if (((m->m_flags & (M_BCAST|M_MCAST)) == 0) && 2170 (dst_if = bridge_rtlookup(sc, eh->ether_dhost, 1)) != NULL) { 2171 error = bridge_enqueue(sc, dst_if, m); 2172 } else 2173 bridge_broadcast(sc, ifp, m, 0); 2174 2175 return (error); 2176 } 2177 2178 /* 2179 * The ifp->if_qflush entry point for if_bridge(4) is no-op. 2180 */ 2181 static void 2182 bridge_qflush(struct ifnet *ifp __unused) 2183 { 2184 } 2185 2186 /* 2187 * bridge_forward: 2188 * 2189 * The forwarding function of the bridge. 2190 * 2191 * NOTE: Releases the lock on return. 2192 */ 2193 static void 2194 bridge_forward(struct bridge_softc *sc, struct bridge_iflist *sbif, 2195 struct mbuf *m) 2196 { 2197 struct bridge_iflist *dbif; 2198 struct ifnet *src_if, *dst_if, *ifp; 2199 struct ether_header *eh; 2200 uint16_t vlan; 2201 uint8_t *dst; 2202 int error; 2203 2204 NET_EPOCH_ASSERT(); 2205 2206 src_if = m->m_pkthdr.rcvif; 2207 ifp = sc->sc_ifp; 2208 2209 if_inc_counter(ifp, IFCOUNTER_IPACKETS, 1); 2210 if_inc_counter(ifp, IFCOUNTER_IBYTES, m->m_pkthdr.len); 2211 vlan = VLANTAGOF(m); 2212 2213 if ((sbif->bif_flags & IFBIF_STP) && 2214 sbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2215 goto drop; 2216 2217 eh = mtod(m, struct ether_header *); 2218 dst = eh->ether_dhost; 2219 2220 /* If the interface is learning, record the address. */ 2221 if (sbif->bif_flags & IFBIF_LEARNING) { 2222 error = bridge_rtupdate(sc, eh->ether_shost, vlan, 2223 sbif, 0, IFBAF_DYNAMIC); 2224 /* 2225 * If the interface has addresses limits then deny any source 2226 * that is not in the cache. 2227 */ 2228 if (error && sbif->bif_addrmax) 2229 goto drop; 2230 } 2231 2232 if ((sbif->bif_flags & IFBIF_STP) != 0 && 2233 sbif->bif_stp.bp_state == BSTP_IFSTATE_LEARNING) 2234 goto drop; 2235 2236 /* 2237 * At this point, the port either doesn't participate 2238 * in spanning tree or it is in the forwarding state. 2239 */ 2240 2241 /* 2242 * If the packet is unicast, destined for someone on 2243 * "this" side of the bridge, drop it. 2244 */ 2245 if ((m->m_flags & (M_BCAST|M_MCAST)) == 0) { 2246 dst_if = bridge_rtlookup(sc, dst, vlan); 2247 if (src_if == dst_if) 2248 goto drop; 2249 } else { 2250 /* 2251 * Check if its a reserved multicast address, any address 2252 * listed in 802.1D section 7.12.6 may not be forwarded by the 2253 * bridge. 2254 * This is currently 01-80-C2-00-00-00 to 01-80-C2-00-00-0F 2255 */ 2256 if (dst[0] == 0x01 && dst[1] == 0x80 && 2257 dst[2] == 0xc2 && dst[3] == 0x00 && 2258 dst[4] == 0x00 && dst[5] <= 0x0f) 2259 goto drop; 2260 2261 /* ...forward it to all interfaces. */ 2262 if_inc_counter(ifp, IFCOUNTER_IMCASTS, 1); 2263 dst_if = NULL; 2264 } 2265 2266 /* 2267 * If we have a destination interface which is a member of our bridge, 2268 * OR this is a unicast packet, push it through the bpf(4) machinery. 2269 * For broadcast or multicast packets, don't bother because it will 2270 * be reinjected into ether_input. We do this before we pass the packets 2271 * through the pfil(9) framework, as it is possible that pfil(9) will 2272 * drop the packet, or possibly modify it, making it difficult to debug 2273 * firewall issues on the bridge. 2274 */ 2275 if (dst_if != NULL || (m->m_flags & (M_BCAST | M_MCAST)) == 0) 2276 ETHER_BPF_MTAP(ifp, m); 2277 2278 /* run the packet filter */ 2279 if (PFIL_HOOKED_IN(V_inet_pfil_head) 2280 #ifdef INET6 2281 || PFIL_HOOKED_IN(V_inet6_pfil_head) 2282 #endif 2283 ) { 2284 if (bridge_pfil(&m, ifp, src_if, PFIL_IN) != 0) 2285 return; 2286 if (m == NULL) 2287 return; 2288 } 2289 2290 if (dst_if == NULL) { 2291 bridge_broadcast(sc, src_if, m, 1); 2292 return; 2293 } 2294 2295 /* 2296 * At this point, we're dealing with a unicast frame 2297 * going to a different interface. 2298 */ 2299 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2300 goto drop; 2301 2302 dbif = bridge_lookup_member_if(sc, dst_if); 2303 if (dbif == NULL) 2304 /* Not a member of the bridge (anymore?) */ 2305 goto drop; 2306 2307 /* Private segments can not talk to each other */ 2308 if (sbif->bif_flags & dbif->bif_flags & IFBIF_PRIVATE) 2309 goto drop; 2310 2311 if ((dbif->bif_flags & IFBIF_STP) && 2312 dbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2313 goto drop; 2314 2315 if (PFIL_HOOKED_OUT(V_inet_pfil_head) 2316 #ifdef INET6 2317 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 2318 #endif 2319 ) { 2320 if (bridge_pfil(&m, ifp, dst_if, PFIL_OUT) != 0) 2321 return; 2322 if (m == NULL) 2323 return; 2324 } 2325 2326 bridge_enqueue(sc, dst_if, m); 2327 return; 2328 2329 drop: 2330 m_freem(m); 2331 } 2332 2333 /* 2334 * bridge_input: 2335 * 2336 * Receive input from a member interface. Queue the packet for 2337 * bridging if it is not for us. 2338 */ 2339 static struct mbuf * 2340 bridge_input(struct ifnet *ifp, struct mbuf *m) 2341 { 2342 struct bridge_softc *sc = ifp->if_bridge; 2343 struct bridge_iflist *bif, *bif2; 2344 struct ifnet *bifp; 2345 struct ether_header *eh; 2346 struct mbuf *mc, *mc2; 2347 uint16_t vlan; 2348 int error; 2349 2350 NET_EPOCH_ASSERT(); 2351 2352 if ((sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) 2353 return (m); 2354 2355 bifp = sc->sc_ifp; 2356 vlan = VLANTAGOF(m); 2357 2358 /* 2359 * Implement support for bridge monitoring. If this flag has been 2360 * set on this interface, discard the packet once we push it through 2361 * the bpf(4) machinery, but before we do, increment the byte and 2362 * packet counters associated with this interface. 2363 */ 2364 if ((bifp->if_flags & IFF_MONITOR) != 0) { 2365 m->m_pkthdr.rcvif = bifp; 2366 ETHER_BPF_MTAP(bifp, m); 2367 if_inc_counter(bifp, IFCOUNTER_IPACKETS, 1); 2368 if_inc_counter(bifp, IFCOUNTER_IBYTES, m->m_pkthdr.len); 2369 m_freem(m); 2370 return (NULL); 2371 } 2372 bif = bridge_lookup_member_if(sc, ifp); 2373 if (bif == NULL) { 2374 return (m); 2375 } 2376 2377 eh = mtod(m, struct ether_header *); 2378 2379 bridge_span(sc, m); 2380 2381 if (m->m_flags & (M_BCAST|M_MCAST)) { 2382 /* Tap off 802.1D packets; they do not get forwarded. */ 2383 if (memcmp(eh->ether_dhost, bstp_etheraddr, 2384 ETHER_ADDR_LEN) == 0) { 2385 bstp_input(&bif->bif_stp, ifp, m); /* consumes mbuf */ 2386 return (NULL); 2387 } 2388 2389 if ((bif->bif_flags & IFBIF_STP) && 2390 bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) { 2391 return (m); 2392 } 2393 2394 /* 2395 * Make a deep copy of the packet and enqueue the copy 2396 * for bridge processing; return the original packet for 2397 * local processing. 2398 */ 2399 mc = m_dup(m, M_NOWAIT); 2400 if (mc == NULL) { 2401 return (m); 2402 } 2403 2404 /* Perform the bridge forwarding function with the copy. */ 2405 bridge_forward(sc, bif, mc); 2406 2407 /* 2408 * Reinject the mbuf as arriving on the bridge so we have a 2409 * chance at claiming multicast packets. We can not loop back 2410 * here from ether_input as a bridge is never a member of a 2411 * bridge. 2412 */ 2413 KASSERT(bifp->if_bridge == NULL, 2414 ("loop created in bridge_input")); 2415 mc2 = m_dup(m, M_NOWAIT); 2416 if (mc2 != NULL) { 2417 /* Keep the layer3 header aligned */ 2418 int i = min(mc2->m_pkthdr.len, max_protohdr); 2419 mc2 = m_copyup(mc2, i, ETHER_ALIGN); 2420 } 2421 if (mc2 != NULL) { 2422 mc2->m_pkthdr.rcvif = bifp; 2423 (*bifp->if_input)(bifp, mc2); 2424 } 2425 2426 /* Return the original packet for local processing. */ 2427 return (m); 2428 } 2429 2430 if ((bif->bif_flags & IFBIF_STP) && 2431 bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) { 2432 return (m); 2433 } 2434 2435 #if (defined(INET) || defined(INET6)) 2436 # define OR_CARP_CHECK_WE_ARE_DST(iface) \ 2437 || ((iface)->if_carp \ 2438 && (*carp_forus_p)((iface), eh->ether_dhost)) 2439 # define OR_CARP_CHECK_WE_ARE_SRC(iface) \ 2440 || ((iface)->if_carp \ 2441 && (*carp_forus_p)((iface), eh->ether_shost)) 2442 #else 2443 # define OR_CARP_CHECK_WE_ARE_DST(iface) 2444 # define OR_CARP_CHECK_WE_ARE_SRC(iface) 2445 #endif 2446 2447 #ifdef INET6 2448 # define OR_PFIL_HOOKED_INET6 \ 2449 || PFIL_HOOKED_IN(V_inet6_pfil_head) 2450 #else 2451 # define OR_PFIL_HOOKED_INET6 2452 #endif 2453 2454 #define GRAB_OUR_PACKETS(iface) \ 2455 if ((iface)->if_type == IFT_GIF) \ 2456 continue; \ 2457 /* It is destined for us. */ \ 2458 if (memcmp(IF_LLADDR((iface)), eh->ether_dhost, ETHER_ADDR_LEN) == 0 \ 2459 OR_CARP_CHECK_WE_ARE_DST((iface)) \ 2460 ) { \ 2461 if (bif->bif_flags & IFBIF_LEARNING) { \ 2462 error = bridge_rtupdate(sc, eh->ether_shost, \ 2463 vlan, bif, 0, IFBAF_DYNAMIC); \ 2464 if (error && bif->bif_addrmax) { \ 2465 m_freem(m); \ 2466 return (NULL); \ 2467 } \ 2468 } \ 2469 m->m_pkthdr.rcvif = iface; \ 2470 if ((iface) == ifp) { \ 2471 /* Skip bridge processing... src == dest */ \ 2472 return (m); \ 2473 } \ 2474 /* It's passing over or to the bridge, locally. */ \ 2475 ETHER_BPF_MTAP(bifp, m); \ 2476 if_inc_counter(bifp, IFCOUNTER_IPACKETS, 1); \ 2477 if_inc_counter(bifp, IFCOUNTER_IBYTES, m->m_pkthdr.len); \ 2478 /* Filter on the physical interface. */ \ 2479 if (V_pfil_local_phys && (PFIL_HOOKED_IN(V_inet_pfil_head) \ 2480 OR_PFIL_HOOKED_INET6)) { \ 2481 if (bridge_pfil(&m, NULL, ifp, \ 2482 PFIL_IN) != 0 || m == NULL) { \ 2483 return (NULL); \ 2484 } \ 2485 } \ 2486 if ((iface) != bifp) \ 2487 ETHER_BPF_MTAP(iface, m); \ 2488 return (m); \ 2489 } \ 2490 \ 2491 /* We just received a packet that we sent out. */ \ 2492 if (memcmp(IF_LLADDR((iface)), eh->ether_shost, ETHER_ADDR_LEN) == 0 \ 2493 OR_CARP_CHECK_WE_ARE_SRC((iface)) \ 2494 ) { \ 2495 m_freem(m); \ 2496 return (NULL); \ 2497 } 2498 2499 /* 2500 * Unicast. Make sure it's not for the bridge. 2501 */ 2502 do { GRAB_OUR_PACKETS(bifp) } while (0); 2503 2504 /* 2505 * Give a chance for ifp at first priority. This will help when the 2506 * packet comes through the interface like VLAN's with the same MACs 2507 * on several interfaces from the same bridge. This also will save 2508 * some CPU cycles in case the destination interface and the input 2509 * interface (eq ifp) are the same. 2510 */ 2511 do { GRAB_OUR_PACKETS(ifp) } while (0); 2512 2513 /* Now check the all bridge members. */ 2514 CK_LIST_FOREACH(bif2, &sc->sc_iflist, bif_next) { 2515 GRAB_OUR_PACKETS(bif2->bif_ifp) 2516 } 2517 2518 #undef OR_CARP_CHECK_WE_ARE_DST 2519 #undef OR_CARP_CHECK_WE_ARE_SRC 2520 #undef OR_PFIL_HOOKED_INET6 2521 #undef GRAB_OUR_PACKETS 2522 2523 /* Perform the bridge forwarding function. */ 2524 bridge_forward(sc, bif, m); 2525 2526 return (NULL); 2527 } 2528 2529 /* 2530 * bridge_broadcast: 2531 * 2532 * Send a frame to all interfaces that are members of 2533 * the bridge, except for the one on which the packet 2534 * arrived. 2535 * 2536 * NOTE: Releases the lock on return. 2537 */ 2538 static void 2539 bridge_broadcast(struct bridge_softc *sc, struct ifnet *src_if, 2540 struct mbuf *m, int runfilt) 2541 { 2542 struct bridge_iflist *dbif, *sbif; 2543 struct mbuf *mc; 2544 struct ifnet *dst_if; 2545 int used = 0, i; 2546 2547 NET_EPOCH_ASSERT(); 2548 2549 sbif = bridge_lookup_member_if(sc, src_if); 2550 2551 /* Filter on the bridge interface before broadcasting */ 2552 if (runfilt && (PFIL_HOOKED_OUT(V_inet_pfil_head) 2553 #ifdef INET6 2554 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 2555 #endif 2556 )) { 2557 if (bridge_pfil(&m, sc->sc_ifp, NULL, PFIL_OUT) != 0) 2558 return; 2559 if (m == NULL) 2560 return; 2561 } 2562 2563 CK_LIST_FOREACH(dbif, &sc->sc_iflist, bif_next) { 2564 dst_if = dbif->bif_ifp; 2565 if (dst_if == src_if) 2566 continue; 2567 2568 /* Private segments can not talk to each other */ 2569 if (sbif && (sbif->bif_flags & dbif->bif_flags & IFBIF_PRIVATE)) 2570 continue; 2571 2572 if ((dbif->bif_flags & IFBIF_STP) && 2573 dbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) 2574 continue; 2575 2576 if ((dbif->bif_flags & IFBIF_DISCOVER) == 0 && 2577 (m->m_flags & (M_BCAST|M_MCAST)) == 0) 2578 continue; 2579 2580 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2581 continue; 2582 2583 if (CK_LIST_NEXT(dbif, bif_next) == NULL) { 2584 mc = m; 2585 used = 1; 2586 } else { 2587 mc = m_dup(m, M_NOWAIT); 2588 if (mc == NULL) { 2589 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 2590 continue; 2591 } 2592 } 2593 2594 /* 2595 * Filter on the output interface. Pass a NULL bridge interface 2596 * pointer so we do not redundantly filter on the bridge for 2597 * each interface we broadcast on. 2598 */ 2599 if (runfilt && (PFIL_HOOKED_OUT(V_inet_pfil_head) 2600 #ifdef INET6 2601 || PFIL_HOOKED_OUT(V_inet6_pfil_head) 2602 #endif 2603 )) { 2604 if (used == 0) { 2605 /* Keep the layer3 header aligned */ 2606 i = min(mc->m_pkthdr.len, max_protohdr); 2607 mc = m_copyup(mc, i, ETHER_ALIGN); 2608 if (mc == NULL) { 2609 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 2610 continue; 2611 } 2612 } 2613 if (bridge_pfil(&mc, NULL, dst_if, PFIL_OUT) != 0) 2614 continue; 2615 if (mc == NULL) 2616 continue; 2617 } 2618 2619 bridge_enqueue(sc, dst_if, mc); 2620 } 2621 if (used == 0) 2622 m_freem(m); 2623 } 2624 2625 /* 2626 * bridge_span: 2627 * 2628 * Duplicate a packet out one or more interfaces that are in span mode, 2629 * the original mbuf is unmodified. 2630 */ 2631 static void 2632 bridge_span(struct bridge_softc *sc, struct mbuf *m) 2633 { 2634 struct bridge_iflist *bif; 2635 struct ifnet *dst_if; 2636 struct mbuf *mc; 2637 2638 NET_EPOCH_ASSERT(); 2639 2640 if (CK_LIST_EMPTY(&sc->sc_spanlist)) 2641 return; 2642 2643 CK_LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) { 2644 dst_if = bif->bif_ifp; 2645 2646 if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) 2647 continue; 2648 2649 mc = m_copypacket(m, M_NOWAIT); 2650 if (mc == NULL) { 2651 if_inc_counter(sc->sc_ifp, IFCOUNTER_OERRORS, 1); 2652 continue; 2653 } 2654 2655 bridge_enqueue(sc, dst_if, mc); 2656 } 2657 } 2658 2659 /* 2660 * bridge_rtupdate: 2661 * 2662 * Add a bridge routing entry. 2663 */ 2664 static int 2665 bridge_rtupdate(struct bridge_softc *sc, const uint8_t *dst, uint16_t vlan, 2666 struct bridge_iflist *bif, int setflags, uint8_t flags) 2667 { 2668 struct bridge_rtnode *brt; 2669 int error; 2670 2671 BRIDGE_LOCK_OR_NET_EPOCH_ASSERT(sc); 2672 2673 /* Check the source address is valid and not multicast. */ 2674 if (ETHER_IS_MULTICAST(dst) || 2675 (dst[0] == 0 && dst[1] == 0 && dst[2] == 0 && 2676 dst[3] == 0 && dst[4] == 0 && dst[5] == 0) != 0) 2677 return (EINVAL); 2678 2679 /* 802.1p frames map to vlan 1 */ 2680 if (vlan == 0) 2681 vlan = 1; 2682 2683 /* 2684 * A route for this destination might already exist. If so, 2685 * update it, otherwise create a new one. 2686 */ 2687 if ((brt = bridge_rtnode_lookup(sc, dst, vlan)) == NULL) { 2688 BRIDGE_RT_LOCK(sc); 2689 2690 /* Check again, now that we have the lock. There could have 2691 * been a race and we only want to insert this once. */ 2692 if ((brt = bridge_rtnode_lookup(sc, dst, vlan)) != NULL) { 2693 BRIDGE_RT_UNLOCK(sc); 2694 return (0); 2695 } 2696 2697 if (sc->sc_brtcnt >= sc->sc_brtmax) { 2698 sc->sc_brtexceeded++; 2699 BRIDGE_RT_UNLOCK(sc); 2700 return (ENOSPC); 2701 } 2702 /* Check per interface address limits (if enabled) */ 2703 if (bif->bif_addrmax && bif->bif_addrcnt >= bif->bif_addrmax) { 2704 bif->bif_addrexceeded++; 2705 BRIDGE_RT_UNLOCK(sc); 2706 return (ENOSPC); 2707 } 2708 2709 /* 2710 * Allocate a new bridge forwarding node, and 2711 * initialize the expiration time and Ethernet 2712 * address. 2713 */ 2714 brt = uma_zalloc(V_bridge_rtnode_zone, M_NOWAIT | M_ZERO); 2715 if (brt == NULL) { 2716 BRIDGE_RT_UNLOCK(sc); 2717 return (ENOMEM); 2718 } 2719 brt->brt_vnet = curvnet; 2720 2721 if (bif->bif_flags & IFBIF_STICKY) 2722 brt->brt_flags = IFBAF_STICKY; 2723 else 2724 brt->brt_flags = IFBAF_DYNAMIC; 2725 2726 memcpy(brt->brt_addr, dst, ETHER_ADDR_LEN); 2727 brt->brt_vlan = vlan; 2728 2729 if ((error = bridge_rtnode_insert(sc, brt)) != 0) { 2730 uma_zfree(V_bridge_rtnode_zone, brt); 2731 BRIDGE_RT_UNLOCK(sc); 2732 return (error); 2733 } 2734 brt->brt_dst = bif; 2735 bif->bif_addrcnt++; 2736 2737 BRIDGE_RT_UNLOCK(sc); 2738 } 2739 2740 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC && 2741 brt->brt_dst != bif) { 2742 BRIDGE_RT_LOCK(sc); 2743 brt->brt_dst->bif_addrcnt--; 2744 brt->brt_dst = bif; 2745 brt->brt_dst->bif_addrcnt++; 2746 BRIDGE_RT_UNLOCK(sc); 2747 } 2748 2749 if ((flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) 2750 brt->brt_expire = time_uptime + sc->sc_brttimeout; 2751 if (setflags) 2752 brt->brt_flags = flags; 2753 2754 return (0); 2755 } 2756 2757 /* 2758 * bridge_rtlookup: 2759 * 2760 * Lookup the destination interface for an address. 2761 */ 2762 static struct ifnet * 2763 bridge_rtlookup(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan) 2764 { 2765 struct bridge_rtnode *brt; 2766 2767 NET_EPOCH_ASSERT(); 2768 2769 if ((brt = bridge_rtnode_lookup(sc, addr, vlan)) == NULL) 2770 return (NULL); 2771 2772 return (brt->brt_ifp); 2773 } 2774 2775 /* 2776 * bridge_rttrim: 2777 * 2778 * Trim the routine table so that we have a number 2779 * of routing entries less than or equal to the 2780 * maximum number. 2781 */ 2782 static void 2783 bridge_rttrim(struct bridge_softc *sc) 2784 { 2785 struct bridge_rtnode *brt, *nbrt; 2786 2787 NET_EPOCH_ASSERT(); 2788 BRIDGE_RT_LOCK_ASSERT(sc); 2789 2790 /* Make sure we actually need to do this. */ 2791 if (sc->sc_brtcnt <= sc->sc_brtmax) 2792 return; 2793 2794 /* Force an aging cycle; this might trim enough addresses. */ 2795 bridge_rtage(sc); 2796 if (sc->sc_brtcnt <= sc->sc_brtmax) 2797 return; 2798 2799 CK_LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2800 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) { 2801 bridge_rtnode_destroy(sc, brt); 2802 if (sc->sc_brtcnt <= sc->sc_brtmax) 2803 return; 2804 } 2805 } 2806 } 2807 2808 /* 2809 * bridge_timer: 2810 * 2811 * Aging timer for the bridge. 2812 */ 2813 static void 2814 bridge_timer(void *arg) 2815 { 2816 struct bridge_softc *sc = arg; 2817 2818 BRIDGE_RT_LOCK_ASSERT(sc); 2819 2820 /* Destruction of rtnodes requires a proper vnet context */ 2821 CURVNET_SET(sc->sc_ifp->if_vnet); 2822 bridge_rtage(sc); 2823 2824 if (sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) 2825 callout_reset(&sc->sc_brcallout, 2826 bridge_rtable_prune_period * hz, bridge_timer, sc); 2827 CURVNET_RESTORE(); 2828 } 2829 2830 /* 2831 * bridge_rtage: 2832 * 2833 * Perform an aging cycle. 2834 */ 2835 static void 2836 bridge_rtage(struct bridge_softc *sc) 2837 { 2838 struct bridge_rtnode *brt, *nbrt; 2839 2840 BRIDGE_RT_LOCK_ASSERT(sc); 2841 2842 CK_LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2843 if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) { 2844 if (time_uptime >= brt->brt_expire) 2845 bridge_rtnode_destroy(sc, brt); 2846 } 2847 } 2848 } 2849 2850 /* 2851 * bridge_rtflush: 2852 * 2853 * Remove all dynamic addresses from the bridge. 2854 */ 2855 static void 2856 bridge_rtflush(struct bridge_softc *sc, int full) 2857 { 2858 struct bridge_rtnode *brt, *nbrt; 2859 2860 BRIDGE_RT_LOCK_ASSERT(sc); 2861 2862 CK_LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2863 if (full || (brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) 2864 bridge_rtnode_destroy(sc, brt); 2865 } 2866 } 2867 2868 /* 2869 * bridge_rtdaddr: 2870 * 2871 * Remove an address from the table. 2872 */ 2873 static int 2874 bridge_rtdaddr(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan) 2875 { 2876 struct bridge_rtnode *brt; 2877 int found = 0; 2878 2879 BRIDGE_RT_LOCK(sc); 2880 2881 /* 2882 * If vlan is zero then we want to delete for all vlans so the lookup 2883 * may return more than one. 2884 */ 2885 while ((brt = bridge_rtnode_lookup(sc, addr, vlan)) != NULL) { 2886 bridge_rtnode_destroy(sc, brt); 2887 found = 1; 2888 } 2889 2890 BRIDGE_RT_UNLOCK(sc); 2891 2892 return (found ? 0 : ENOENT); 2893 } 2894 2895 /* 2896 * bridge_rtdelete: 2897 * 2898 * Delete routes to a speicifc member interface. 2899 */ 2900 static void 2901 bridge_rtdelete(struct bridge_softc *sc, struct ifnet *ifp, int full) 2902 { 2903 struct bridge_rtnode *brt, *nbrt; 2904 2905 BRIDGE_RT_LOCK_ASSERT(sc); 2906 2907 CK_LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) { 2908 if (brt->brt_ifp == ifp && (full || 2909 (brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC)) 2910 bridge_rtnode_destroy(sc, brt); 2911 } 2912 } 2913 2914 /* 2915 * bridge_rtable_init: 2916 * 2917 * Initialize the route table for this bridge. 2918 */ 2919 static void 2920 bridge_rtable_init(struct bridge_softc *sc) 2921 { 2922 int i; 2923 2924 sc->sc_rthash = malloc(sizeof(*sc->sc_rthash) * BRIDGE_RTHASH_SIZE, 2925 M_DEVBUF, M_WAITOK); 2926 2927 for (i = 0; i < BRIDGE_RTHASH_SIZE; i++) 2928 CK_LIST_INIT(&sc->sc_rthash[i]); 2929 2930 sc->sc_rthash_key = arc4random(); 2931 CK_LIST_INIT(&sc->sc_rtlist); 2932 } 2933 2934 /* 2935 * bridge_rtable_fini: 2936 * 2937 * Deconstruct the route table for this bridge. 2938 */ 2939 static void 2940 bridge_rtable_fini(struct bridge_softc *sc) 2941 { 2942 2943 KASSERT(sc->sc_brtcnt == 0, 2944 ("%s: %d bridge routes referenced", __func__, sc->sc_brtcnt)); 2945 free(sc->sc_rthash, M_DEVBUF); 2946 } 2947 2948 /* 2949 * The following hash function is adapted from "Hash Functions" by Bob Jenkins 2950 * ("Algorithm Alley", Dr. Dobbs Journal, September 1997). 2951 */ 2952 #define mix(a, b, c) \ 2953 do { \ 2954 a -= b; a -= c; a ^= (c >> 13); \ 2955 b -= c; b -= a; b ^= (a << 8); \ 2956 c -= a; c -= b; c ^= (b >> 13); \ 2957 a -= b; a -= c; a ^= (c >> 12); \ 2958 b -= c; b -= a; b ^= (a << 16); \ 2959 c -= a; c -= b; c ^= (b >> 5); \ 2960 a -= b; a -= c; a ^= (c >> 3); \ 2961 b -= c; b -= a; b ^= (a << 10); \ 2962 c -= a; c -= b; c ^= (b >> 15); \ 2963 } while (/*CONSTCOND*/0) 2964 2965 static __inline uint32_t 2966 bridge_rthash(struct bridge_softc *sc, const uint8_t *addr) 2967 { 2968 uint32_t a = 0x9e3779b9, b = 0x9e3779b9, c = sc->sc_rthash_key; 2969 2970 b += addr[5] << 8; 2971 b += addr[4]; 2972 a += addr[3] << 24; 2973 a += addr[2] << 16; 2974 a += addr[1] << 8; 2975 a += addr[0]; 2976 2977 mix(a, b, c); 2978 2979 return (c & BRIDGE_RTHASH_MASK); 2980 } 2981 2982 #undef mix 2983 2984 static int 2985 bridge_rtnode_addr_cmp(const uint8_t *a, const uint8_t *b) 2986 { 2987 int i, d; 2988 2989 for (i = 0, d = 0; i < ETHER_ADDR_LEN && d == 0; i++) { 2990 d = ((int)a[i]) - ((int)b[i]); 2991 } 2992 2993 return (d); 2994 } 2995 2996 /* 2997 * bridge_rtnode_lookup: 2998 * 2999 * Look up a bridge route node for the specified destination. Compare the 3000 * vlan id or if zero then just return the first match. 3001 */ 3002 static struct bridge_rtnode * 3003 bridge_rtnode_lookup(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan) 3004 { 3005 struct bridge_rtnode *brt; 3006 uint32_t hash; 3007 int dir; 3008 3009 BRIDGE_RT_LOCK_OR_NET_EPOCH_ASSERT(sc); 3010 3011 hash = bridge_rthash(sc, addr); 3012 CK_LIST_FOREACH(brt, &sc->sc_rthash[hash], brt_hash) { 3013 dir = bridge_rtnode_addr_cmp(addr, brt->brt_addr); 3014 if (dir == 0 && (brt->brt_vlan == vlan || vlan == 0)) 3015 return (brt); 3016 if (dir > 0) 3017 return (NULL); 3018 } 3019 3020 return (NULL); 3021 } 3022 3023 /* 3024 * bridge_rtnode_insert: 3025 * 3026 * Insert the specified bridge node into the route table. We 3027 * assume the entry is not already in the table. 3028 */ 3029 static int 3030 bridge_rtnode_insert(struct bridge_softc *sc, struct bridge_rtnode *brt) 3031 { 3032 struct bridge_rtnode *lbrt; 3033 uint32_t hash; 3034 int dir; 3035 3036 BRIDGE_RT_LOCK_ASSERT(sc); 3037 3038 hash = bridge_rthash(sc, brt->brt_addr); 3039 3040 lbrt = CK_LIST_FIRST(&sc->sc_rthash[hash]); 3041 if (lbrt == NULL) { 3042 CK_LIST_INSERT_HEAD(&sc->sc_rthash[hash], brt, brt_hash); 3043 goto out; 3044 } 3045 3046 do { 3047 dir = bridge_rtnode_addr_cmp(brt->brt_addr, lbrt->brt_addr); 3048 if (dir == 0 && brt->brt_vlan == lbrt->brt_vlan) 3049 return (EEXIST); 3050 if (dir > 0) { 3051 CK_LIST_INSERT_BEFORE(lbrt, brt, brt_hash); 3052 goto out; 3053 } 3054 if (CK_LIST_NEXT(lbrt, brt_hash) == NULL) { 3055 CK_LIST_INSERT_AFTER(lbrt, brt, brt_hash); 3056 goto out; 3057 } 3058 lbrt = CK_LIST_NEXT(lbrt, brt_hash); 3059 } while (lbrt != NULL); 3060 3061 #ifdef DIAGNOSTIC 3062 panic("bridge_rtnode_insert: impossible"); 3063 #endif 3064 3065 out: 3066 CK_LIST_INSERT_HEAD(&sc->sc_rtlist, brt, brt_list); 3067 sc->sc_brtcnt++; 3068 3069 return (0); 3070 } 3071 3072 static void 3073 bridge_rtnode_destroy_cb(struct epoch_context *ctx) 3074 { 3075 struct bridge_rtnode *brt; 3076 3077 brt = __containerof(ctx, struct bridge_rtnode, brt_epoch_ctx); 3078 3079 CURVNET_SET(brt->brt_vnet); 3080 uma_zfree(V_bridge_rtnode_zone, brt); 3081 CURVNET_RESTORE(); 3082 } 3083 3084 /* 3085 * bridge_rtnode_destroy: 3086 * 3087 * Destroy a bridge rtnode. 3088 */ 3089 static void 3090 bridge_rtnode_destroy(struct bridge_softc *sc, struct bridge_rtnode *brt) 3091 { 3092 BRIDGE_RT_LOCK_ASSERT(sc); 3093 3094 CK_LIST_REMOVE(brt, brt_hash); 3095 3096 CK_LIST_REMOVE(brt, brt_list); 3097 sc->sc_brtcnt--; 3098 brt->brt_dst->bif_addrcnt--; 3099 3100 NET_EPOCH_CALL(bridge_rtnode_destroy_cb, &brt->brt_epoch_ctx); 3101 } 3102 3103 /* 3104 * bridge_rtable_expire: 3105 * 3106 * Set the expiry time for all routes on an interface. 3107 */ 3108 static void 3109 bridge_rtable_expire(struct ifnet *ifp, int age) 3110 { 3111 struct bridge_softc *sc = ifp->if_bridge; 3112 struct bridge_rtnode *brt; 3113 3114 CURVNET_SET(ifp->if_vnet); 3115 BRIDGE_RT_LOCK(sc); 3116 3117 /* 3118 * If the age is zero then flush, otherwise set all the expiry times to 3119 * age for the interface 3120 */ 3121 if (age == 0) 3122 bridge_rtdelete(sc, ifp, IFBF_FLUSHDYN); 3123 else { 3124 CK_LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) { 3125 /* Cap the expiry time to 'age' */ 3126 if (brt->brt_ifp == ifp && 3127 brt->brt_expire > time_uptime + age && 3128 (brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) 3129 brt->brt_expire = time_uptime + age; 3130 } 3131 } 3132 BRIDGE_RT_UNLOCK(sc); 3133 CURVNET_RESTORE(); 3134 } 3135 3136 /* 3137 * bridge_state_change: 3138 * 3139 * Callback from the bridgestp code when a port changes states. 3140 */ 3141 static void 3142 bridge_state_change(struct ifnet *ifp, int state) 3143 { 3144 struct bridge_softc *sc = ifp->if_bridge; 3145 static const char *stpstates[] = { 3146 "disabled", 3147 "listening", 3148 "learning", 3149 "forwarding", 3150 "blocking", 3151 "discarding" 3152 }; 3153 3154 CURVNET_SET(ifp->if_vnet); 3155 if (V_log_stp) 3156 log(LOG_NOTICE, "%s: state changed to %s on %s\n", 3157 sc->sc_ifp->if_xname, stpstates[state], ifp->if_xname); 3158 CURVNET_RESTORE(); 3159 } 3160 3161 /* 3162 * Send bridge packets through pfil if they are one of the types pfil can deal 3163 * with, or if they are ARP or REVARP. (pfil will pass ARP and REVARP without 3164 * question.) If *bifp or *ifp are NULL then packet filtering is skipped for 3165 * that interface. 3166 */ 3167 static int 3168 bridge_pfil(struct mbuf **mp, struct ifnet *bifp, struct ifnet *ifp, int dir) 3169 { 3170 int snap, error, i, hlen; 3171 struct ether_header *eh1, eh2; 3172 struct ip *ip; 3173 struct llc llc1; 3174 u_int16_t ether_type; 3175 pfil_return_t rv; 3176 3177 snap = 0; 3178 error = -1; /* Default error if not error == 0 */ 3179 3180 #if 0 3181 /* we may return with the IP fields swapped, ensure its not shared */ 3182 KASSERT(M_WRITABLE(*mp), ("%s: modifying a shared mbuf", __func__)); 3183 #endif 3184 3185 if (V_pfil_bridge == 0 && V_pfil_member == 0 && V_pfil_ipfw == 0) 3186 return (0); /* filtering is disabled */ 3187 3188 i = min((*mp)->m_pkthdr.len, max_protohdr); 3189 if ((*mp)->m_len < i) { 3190 *mp = m_pullup(*mp, i); 3191 if (*mp == NULL) { 3192 printf("%s: m_pullup failed\n", __func__); 3193 return (-1); 3194 } 3195 } 3196 3197 eh1 = mtod(*mp, struct ether_header *); 3198 ether_type = ntohs(eh1->ether_type); 3199 3200 /* 3201 * Check for SNAP/LLC. 3202 */ 3203 if (ether_type < ETHERMTU) { 3204 struct llc *llc2 = (struct llc *)(eh1 + 1); 3205 3206 if ((*mp)->m_len >= ETHER_HDR_LEN + 8 && 3207 llc2->llc_dsap == LLC_SNAP_LSAP && 3208 llc2->llc_ssap == LLC_SNAP_LSAP && 3209 llc2->llc_control == LLC_UI) { 3210 ether_type = htons(llc2->llc_un.type_snap.ether_type); 3211 snap = 1; 3212 } 3213 } 3214 3215 /* 3216 * If we're trying to filter bridge traffic, don't look at anything 3217 * other than IP and ARP traffic. If the filter doesn't understand 3218 * IPv6, don't allow IPv6 through the bridge either. This is lame 3219 * since if we really wanted, say, an AppleTalk filter, we are hosed, 3220 * but of course we don't have an AppleTalk filter to begin with. 3221 * (Note that since pfil doesn't understand ARP it will pass *ALL* 3222 * ARP traffic.) 3223 */ 3224 switch (ether_type) { 3225 case ETHERTYPE_ARP: 3226 case ETHERTYPE_REVARP: 3227 if (V_pfil_ipfw_arp == 0) 3228 return (0); /* Automatically pass */ 3229 break; 3230 3231 case ETHERTYPE_IP: 3232 #ifdef INET6 3233 case ETHERTYPE_IPV6: 3234 #endif /* INET6 */ 3235 break; 3236 default: 3237 /* 3238 * Check to see if the user wants to pass non-ip 3239 * packets, these will not be checked by pfil(9) and 3240 * passed unconditionally so the default is to drop. 3241 */ 3242 if (V_pfil_onlyip) 3243 goto bad; 3244 } 3245 3246 /* Run the packet through pfil before stripping link headers */ 3247 if (PFIL_HOOKED_OUT(V_link_pfil_head) && V_pfil_ipfw != 0 && 3248 dir == PFIL_OUT && ifp != NULL) { 3249 switch (pfil_run_hooks(V_link_pfil_head, mp, ifp, dir, NULL)) { 3250 case PFIL_DROPPED: 3251 return (EACCES); 3252 case PFIL_CONSUMED: 3253 return (0); 3254 } 3255 } 3256 3257 /* Strip off the Ethernet header and keep a copy. */ 3258 m_copydata(*mp, 0, ETHER_HDR_LEN, (caddr_t) &eh2); 3259 m_adj(*mp, ETHER_HDR_LEN); 3260 3261 /* Strip off snap header, if present */ 3262 if (snap) { 3263 m_copydata(*mp, 0, sizeof(struct llc), (caddr_t) &llc1); 3264 m_adj(*mp, sizeof(struct llc)); 3265 } 3266 3267 /* 3268 * Check the IP header for alignment and errors 3269 */ 3270 if (dir == PFIL_IN) { 3271 switch (ether_type) { 3272 case ETHERTYPE_IP: 3273 error = bridge_ip_checkbasic(mp); 3274 break; 3275 #ifdef INET6 3276 case ETHERTYPE_IPV6: 3277 error = bridge_ip6_checkbasic(mp); 3278 break; 3279 #endif /* INET6 */ 3280 default: 3281 error = 0; 3282 } 3283 if (error) 3284 goto bad; 3285 } 3286 3287 error = 0; 3288 3289 /* 3290 * Run the packet through pfil 3291 */ 3292 rv = PFIL_PASS; 3293 switch (ether_type) { 3294 case ETHERTYPE_IP: 3295 /* 3296 * Run pfil on the member interface and the bridge, both can 3297 * be skipped by clearing pfil_member or pfil_bridge. 3298 * 3299 * Keep the order: 3300 * in_if -> bridge_if -> out_if 3301 */ 3302 if (V_pfil_bridge && dir == PFIL_OUT && bifp != NULL && (rv = 3303 pfil_run_hooks(V_inet_pfil_head, mp, bifp, dir, NULL)) != 3304 PFIL_PASS) 3305 break; 3306 3307 if (V_pfil_member && ifp != NULL && (rv = 3308 pfil_run_hooks(V_inet_pfil_head, mp, ifp, dir, NULL)) != 3309 PFIL_PASS) 3310 break; 3311 3312 if (V_pfil_bridge && dir == PFIL_IN && bifp != NULL && (rv = 3313 pfil_run_hooks(V_inet_pfil_head, mp, bifp, dir, NULL)) != 3314 PFIL_PASS) 3315 break; 3316 3317 /* check if we need to fragment the packet */ 3318 /* bridge_fragment generates a mbuf chain of packets */ 3319 /* that already include eth headers */ 3320 if (V_pfil_member && ifp != NULL && dir == PFIL_OUT) { 3321 i = (*mp)->m_pkthdr.len; 3322 if (i > ifp->if_mtu) { 3323 error = bridge_fragment(ifp, mp, &eh2, snap, 3324 &llc1); 3325 return (error); 3326 } 3327 } 3328 3329 /* Recalculate the ip checksum. */ 3330 ip = mtod(*mp, struct ip *); 3331 hlen = ip->ip_hl << 2; 3332 if (hlen < sizeof(struct ip)) 3333 goto bad; 3334 if (hlen > (*mp)->m_len) { 3335 if ((*mp = m_pullup(*mp, hlen)) == NULL) 3336 goto bad; 3337 ip = mtod(*mp, struct ip *); 3338 if (ip == NULL) 3339 goto bad; 3340 } 3341 ip->ip_sum = 0; 3342 if (hlen == sizeof(struct ip)) 3343 ip->ip_sum = in_cksum_hdr(ip); 3344 else 3345 ip->ip_sum = in_cksum(*mp, hlen); 3346 3347 break; 3348 #ifdef INET6 3349 case ETHERTYPE_IPV6: 3350 if (V_pfil_bridge && dir == PFIL_OUT && bifp != NULL && (rv = 3351 pfil_run_hooks(V_inet6_pfil_head, mp, bifp, dir, NULL)) != 3352 PFIL_PASS) 3353 break; 3354 3355 if (V_pfil_member && ifp != NULL && (rv = 3356 pfil_run_hooks(V_inet6_pfil_head, mp, ifp, dir, NULL)) != 3357 PFIL_PASS) 3358 break; 3359 3360 if (V_pfil_bridge && dir == PFIL_IN && bifp != NULL && (rv = 3361 pfil_run_hooks(V_inet6_pfil_head, mp, bifp, dir, NULL)) != 3362 PFIL_PASS) 3363 break; 3364 break; 3365 #endif 3366 } 3367 3368 switch (rv) { 3369 case PFIL_CONSUMED: 3370 return (0); 3371 case PFIL_DROPPED: 3372 return (EACCES); 3373 default: 3374 break; 3375 } 3376 3377 error = -1; 3378 3379 /* 3380 * Finally, put everything back the way it was and return 3381 */ 3382 if (snap) { 3383 M_PREPEND(*mp, sizeof(struct llc), M_NOWAIT); 3384 if (*mp == NULL) 3385 return (error); 3386 bcopy(&llc1, mtod(*mp, caddr_t), sizeof(struct llc)); 3387 } 3388 3389 M_PREPEND(*mp, ETHER_HDR_LEN, M_NOWAIT); 3390 if (*mp == NULL) 3391 return (error); 3392 bcopy(&eh2, mtod(*mp, caddr_t), ETHER_HDR_LEN); 3393 3394 return (0); 3395 3396 bad: 3397 m_freem(*mp); 3398 *mp = NULL; 3399 return (error); 3400 } 3401 3402 /* 3403 * Perform basic checks on header size since 3404 * pfil assumes ip_input has already processed 3405 * it for it. Cut-and-pasted from ip_input.c. 3406 * Given how simple the IPv6 version is, 3407 * does the IPv4 version really need to be 3408 * this complicated? 3409 * 3410 * XXX Should we update ipstat here, or not? 3411 * XXX Right now we update ipstat but not 3412 * XXX csum_counter. 3413 */ 3414 static int 3415 bridge_ip_checkbasic(struct mbuf **mp) 3416 { 3417 struct mbuf *m = *mp; 3418 struct ip *ip; 3419 int len, hlen; 3420 u_short sum; 3421 3422 if (*mp == NULL) 3423 return (-1); 3424 3425 if (IP_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { 3426 if ((m = m_copyup(m, sizeof(struct ip), 3427 (max_linkhdr + 3) & ~3)) == NULL) { 3428 /* XXXJRT new stat, please */ 3429 KMOD_IPSTAT_INC(ips_toosmall); 3430 goto bad; 3431 } 3432 } else if (__predict_false(m->m_len < sizeof (struct ip))) { 3433 if ((m = m_pullup(m, sizeof (struct ip))) == NULL) { 3434 KMOD_IPSTAT_INC(ips_toosmall); 3435 goto bad; 3436 } 3437 } 3438 ip = mtod(m, struct ip *); 3439 if (ip == NULL) goto bad; 3440 3441 if (ip->ip_v != IPVERSION) { 3442 KMOD_IPSTAT_INC(ips_badvers); 3443 goto bad; 3444 } 3445 hlen = ip->ip_hl << 2; 3446 if (hlen < sizeof(struct ip)) { /* minimum header length */ 3447 KMOD_IPSTAT_INC(ips_badhlen); 3448 goto bad; 3449 } 3450 if (hlen > m->m_len) { 3451 if ((m = m_pullup(m, hlen)) == NULL) { 3452 KMOD_IPSTAT_INC(ips_badhlen); 3453 goto bad; 3454 } 3455 ip = mtod(m, struct ip *); 3456 if (ip == NULL) goto bad; 3457 } 3458 3459 if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) { 3460 sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID); 3461 } else { 3462 if (hlen == sizeof(struct ip)) { 3463 sum = in_cksum_hdr(ip); 3464 } else { 3465 sum = in_cksum(m, hlen); 3466 } 3467 } 3468 if (sum) { 3469 KMOD_IPSTAT_INC(ips_badsum); 3470 goto bad; 3471 } 3472 3473 /* Retrieve the packet length. */ 3474 len = ntohs(ip->ip_len); 3475 3476 /* 3477 * Check for additional length bogosity 3478 */ 3479 if (len < hlen) { 3480 KMOD_IPSTAT_INC(ips_badlen); 3481 goto bad; 3482 } 3483 3484 /* 3485 * Check that the amount of data in the buffers 3486 * is as at least much as the IP header would have us expect. 3487 * Drop packet if shorter than we expect. 3488 */ 3489 if (m->m_pkthdr.len < len) { 3490 KMOD_IPSTAT_INC(ips_tooshort); 3491 goto bad; 3492 } 3493 3494 /* Checks out, proceed */ 3495 *mp = m; 3496 return (0); 3497 3498 bad: 3499 *mp = m; 3500 return (-1); 3501 } 3502 3503 #ifdef INET6 3504 /* 3505 * Same as above, but for IPv6. 3506 * Cut-and-pasted from ip6_input.c. 3507 * XXX Should we update ip6stat, or not? 3508 */ 3509 static int 3510 bridge_ip6_checkbasic(struct mbuf **mp) 3511 { 3512 struct mbuf *m = *mp; 3513 struct ip6_hdr *ip6; 3514 3515 /* 3516 * If the IPv6 header is not aligned, slurp it up into a new 3517 * mbuf with space for link headers, in the event we forward 3518 * it. Otherwise, if it is aligned, make sure the entire base 3519 * IPv6 header is in the first mbuf of the chain. 3520 */ 3521 if (IP6_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { 3522 struct ifnet *inifp = m->m_pkthdr.rcvif; 3523 if ((m = m_copyup(m, sizeof(struct ip6_hdr), 3524 (max_linkhdr + 3) & ~3)) == NULL) { 3525 /* XXXJRT new stat, please */ 3526 IP6STAT_INC(ip6s_toosmall); 3527 in6_ifstat_inc(inifp, ifs6_in_hdrerr); 3528 goto bad; 3529 } 3530 } else if (__predict_false(m->m_len < sizeof(struct ip6_hdr))) { 3531 struct ifnet *inifp = m->m_pkthdr.rcvif; 3532 if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) { 3533 IP6STAT_INC(ip6s_toosmall); 3534 in6_ifstat_inc(inifp, ifs6_in_hdrerr); 3535 goto bad; 3536 } 3537 } 3538 3539 ip6 = mtod(m, struct ip6_hdr *); 3540 3541 if ((ip6->ip6_vfc & IPV6_VERSION_MASK) != IPV6_VERSION) { 3542 IP6STAT_INC(ip6s_badvers); 3543 in6_ifstat_inc(m->m_pkthdr.rcvif, ifs6_in_hdrerr); 3544 goto bad; 3545 } 3546 3547 /* Checks out, proceed */ 3548 *mp = m; 3549 return (0); 3550 3551 bad: 3552 *mp = m; 3553 return (-1); 3554 } 3555 #endif /* INET6 */ 3556 3557 /* 3558 * bridge_fragment: 3559 * 3560 * Fragment mbuf chain in multiple packets and prepend ethernet header. 3561 */ 3562 static int 3563 bridge_fragment(struct ifnet *ifp, struct mbuf **mp, struct ether_header *eh, 3564 int snap, struct llc *llc) 3565 { 3566 struct mbuf *m = *mp, *nextpkt = NULL, *mprev = NULL, *mcur = NULL; 3567 struct ip *ip; 3568 int error = -1; 3569 3570 if (m->m_len < sizeof(struct ip) && 3571 (m = m_pullup(m, sizeof(struct ip))) == NULL) 3572 goto dropit; 3573 ip = mtod(m, struct ip *); 3574 3575 m->m_pkthdr.csum_flags |= CSUM_IP; 3576 error = ip_fragment(ip, &m, ifp->if_mtu, ifp->if_hwassist); 3577 if (error) 3578 goto dropit; 3579 3580 /* 3581 * Walk the chain and re-add the Ethernet header for 3582 * each mbuf packet. 3583 */ 3584 for (mcur = m; mcur; mcur = mcur->m_nextpkt) { 3585 nextpkt = mcur->m_nextpkt; 3586 mcur->m_nextpkt = NULL; 3587 if (snap) { 3588 M_PREPEND(mcur, sizeof(struct llc), M_NOWAIT); 3589 if (mcur == NULL) { 3590 error = ENOBUFS; 3591 if (mprev != NULL) 3592 mprev->m_nextpkt = nextpkt; 3593 goto dropit; 3594 } 3595 bcopy(llc, mtod(mcur, caddr_t),sizeof(struct llc)); 3596 } 3597 3598 M_PREPEND(mcur, ETHER_HDR_LEN, M_NOWAIT); 3599 if (mcur == NULL) { 3600 error = ENOBUFS; 3601 if (mprev != NULL) 3602 mprev->m_nextpkt = nextpkt; 3603 goto dropit; 3604 } 3605 bcopy(eh, mtod(mcur, caddr_t), ETHER_HDR_LEN); 3606 3607 /* 3608 * The previous two M_PREPEND could have inserted one or two 3609 * mbufs in front so we have to update the previous packet's 3610 * m_nextpkt. 3611 */ 3612 mcur->m_nextpkt = nextpkt; 3613 if (mprev != NULL) 3614 mprev->m_nextpkt = mcur; 3615 else { 3616 /* The first mbuf in the original chain needs to be 3617 * updated. */ 3618 *mp = mcur; 3619 } 3620 mprev = mcur; 3621 } 3622 3623 KMOD_IPSTAT_INC(ips_fragmented); 3624 return (error); 3625 3626 dropit: 3627 for (mcur = *mp; mcur; mcur = m) { /* droping the full packet chain */ 3628 m = mcur->m_nextpkt; 3629 m_freem(mcur); 3630 } 3631 return (error); 3632 } 3633 3634 static void 3635 bridge_linkstate(struct ifnet *ifp) 3636 { 3637 struct bridge_softc *sc = ifp->if_bridge; 3638 struct bridge_iflist *bif; 3639 struct epoch_tracker et; 3640 3641 NET_EPOCH_ENTER(et); 3642 3643 bif = bridge_lookup_member_if(sc, ifp); 3644 if (bif == NULL) { 3645 NET_EPOCH_EXIT(et); 3646 return; 3647 } 3648 bridge_linkcheck(sc); 3649 3650 bstp_linkstate(&bif->bif_stp); 3651 3652 NET_EPOCH_EXIT(et); 3653 } 3654 3655 static void 3656 bridge_linkcheck(struct bridge_softc *sc) 3657 { 3658 struct bridge_iflist *bif; 3659 int new_link, hasls; 3660 3661 BRIDGE_LOCK_OR_NET_EPOCH_ASSERT(sc); 3662 3663 new_link = LINK_STATE_DOWN; 3664 hasls = 0; 3665 /* Our link is considered up if at least one of our ports is active */ 3666 CK_LIST_FOREACH(bif, &sc->sc_iflist, bif_next) { 3667 if (bif->bif_ifp->if_capabilities & IFCAP_LINKSTATE) 3668 hasls++; 3669 if (bif->bif_ifp->if_link_state == LINK_STATE_UP) { 3670 new_link = LINK_STATE_UP; 3671 break; 3672 } 3673 } 3674 if (!CK_LIST_EMPTY(&sc->sc_iflist) && !hasls) { 3675 /* If no interfaces support link-state then we default to up */ 3676 new_link = LINK_STATE_UP; 3677 } 3678 if_link_state_change(sc->sc_ifp, new_link); 3679 } 3680