1// SPDX-License-Identifier: GPL-2.0-or-later 2/* SCTP kernel implementation 3 * (C) Copyright IBM Corp. 2001, 2004 4 * Copyright (c) 1999-2000 Cisco, Inc. 5 * Copyright (c) 1999-2001 Motorola, Inc. 6 * Copyright (c) 2001 Intel Corp. 7 * Copyright (c) 2001 Nokia, Inc. 8 * Copyright (c) 2001 La Monte H.P. Yarroll 9 * 10 * This file is part of the SCTP kernel implementation 11 * 12 * Initialization/cleanup for SCTP protocol support. 13 * 14 * Please send any bug reports or fixes you make to the 15 * email address(es): 16 * lksctp developers <linux-sctp@vger.kernel.org> 17 * 18 * Written or modified by: 19 * La Monte H.P. Yarroll <piggy@acm.org> 20 * Karl Knutson <karl@athena.chicago.il.us> 21 * Jon Grimm <jgrimm@us.ibm.com> 22 * Sridhar Samudrala <sri@us.ibm.com> 23 * Daisy Chang <daisyc@us.ibm.com> 24 * Ardelle Fan <ardelle.fan@intel.com> 25 */ 26 27#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt 28 29#include <linux/module.h> 30#include <linux/init.h> 31#include <linux/netdevice.h> 32#include <linux/inetdevice.h> 33#include <linux/seq_file.h> 34#include <linux/memblock.h> 35#include <linux/highmem.h> 36#include <linux/swap.h> 37#include <linux/slab.h> 38#include <net/net_namespace.h> 39#include <net/protocol.h> 40#include <net/ip.h> 41#include <net/ipv6.h> 42#include <net/route.h> 43#include <net/sctp/sctp.h> 44#include <net/addrconf.h> 45#include <net/inet_common.h> 46#include <net/inet_ecn.h> 47 48#define MAX_SCTP_PORT_HASH_ENTRIES (64 * 1024) 49 50/* Global data structures. */ 51struct sctp_globals sctp_globals __read_mostly; 52 53struct idr sctp_assocs_id; 54DEFINE_SPINLOCK(sctp_assocs_id_lock); 55 56static struct sctp_pf *sctp_pf_inet6_specific; 57static struct sctp_pf *sctp_pf_inet_specific; 58static struct sctp_af *sctp_af_v4_specific; 59static struct sctp_af *sctp_af_v6_specific; 60 61struct kmem_cache *sctp_chunk_cachep __read_mostly; 62struct kmem_cache *sctp_bucket_cachep __read_mostly; 63 64long sysctl_sctp_mem[3]; 65int sysctl_sctp_rmem[3]; 66int sysctl_sctp_wmem[3]; 67 68/* Private helper to extract ipv4 address and stash them in 69 * the protocol structure. 70 */ 71static void sctp_v4_copy_addrlist(struct list_head *addrlist, 72 struct net_device *dev) 73{ 74 struct in_device *in_dev; 75 struct in_ifaddr *ifa; 76 struct sctp_sockaddr_entry *addr; 77 78 rcu_read_lock(); 79 if ((in_dev = __in_dev_get_rcu(dev)) == NULL) { 80 rcu_read_unlock(); 81 return; 82 } 83 84 in_dev_for_each_ifa_rcu(ifa, in_dev) { 85 /* Add the address to the local list. */ 86 addr = kzalloc(sizeof(*addr), GFP_ATOMIC); 87 if (addr) { 88 addr->a.v4.sin_family = AF_INET; 89 addr->a.v4.sin_addr.s_addr = ifa->ifa_local; 90 addr->valid = 1; 91 INIT_LIST_HEAD(&addr->list); 92 list_add_tail(&addr->list, addrlist); 93 } 94 } 95 96 rcu_read_unlock(); 97} 98 99/* Extract our IP addresses from the system and stash them in the 100 * protocol structure. 101 */ 102static void sctp_get_local_addr_list(struct net *net) 103{ 104 struct net_device *dev; 105 struct list_head *pos; 106 struct sctp_af *af; 107 108 rcu_read_lock(); 109 for_each_netdev_rcu(net, dev) { 110 list_for_each(pos, &sctp_address_families) { 111 af = list_entry(pos, struct sctp_af, list); 112 af->copy_addrlist(&net->sctp.local_addr_list, dev); 113 } 114 } 115 rcu_read_unlock(); 116} 117 118/* Free the existing local addresses. */ 119static void sctp_free_local_addr_list(struct net *net) 120{ 121 struct sctp_sockaddr_entry *addr; 122 struct list_head *pos, *temp; 123 124 list_for_each_safe(pos, temp, &net->sctp.local_addr_list) { 125 addr = list_entry(pos, struct sctp_sockaddr_entry, list); 126 list_del(pos); 127 kfree(addr); 128 } 129} 130 131/* Copy the local addresses which are valid for 'scope' into 'bp'. */ 132int sctp_copy_local_addr_list(struct net *net, struct sctp_bind_addr *bp, 133 enum sctp_scope scope, gfp_t gfp, int copy_flags) 134{ 135 struct sctp_sockaddr_entry *addr; 136 union sctp_addr laddr; 137 int error = 0; 138 139 rcu_read_lock(); 140 list_for_each_entry_rcu(addr, &net->sctp.local_addr_list, list) { 141 if (!addr->valid) 142 continue; 143 if (!sctp_in_scope(net, &addr->a, scope)) 144 continue; 145 146 /* Now that the address is in scope, check to see if 147 * the address type is really supported by the local 148 * sock as well as the remote peer. 149 */ 150 if (addr->a.sa.sa_family == AF_INET && 151 (!(copy_flags & SCTP_ADDR4_ALLOWED) || 152 !(copy_flags & SCTP_ADDR4_PEERSUPP))) 153 continue; 154 if (addr->a.sa.sa_family == AF_INET6 && 155 (!(copy_flags & SCTP_ADDR6_ALLOWED) || 156 !(copy_flags & SCTP_ADDR6_PEERSUPP))) 157 continue; 158 159 laddr = addr->a; 160 /* also works for setting ipv6 address port */ 161 laddr.v4.sin_port = htons(bp->port); 162 if (sctp_bind_addr_state(bp, &laddr) != -1) 163 continue; 164 165 error = sctp_add_bind_addr(bp, &addr->a, sizeof(addr->a), 166 SCTP_ADDR_SRC, GFP_ATOMIC); 167 if (error) 168 break; 169 } 170 171 rcu_read_unlock(); 172 return error; 173} 174 175/* Copy over any ip options */ 176static void sctp_v4_copy_ip_options(struct sock *sk, struct sock *newsk) 177{ 178 struct inet_sock *newinet, *inet = inet_sk(sk); 179 struct ip_options_rcu *inet_opt, *newopt = NULL; 180 181 newinet = inet_sk(newsk); 182 183 rcu_read_lock(); 184 inet_opt = rcu_dereference(inet->inet_opt); 185 if (inet_opt) { 186 newopt = sock_kmalloc(newsk, sizeof(*inet_opt) + 187 inet_opt->opt.optlen, GFP_ATOMIC); 188 if (newopt) 189 memcpy(newopt, inet_opt, sizeof(*inet_opt) + 190 inet_opt->opt.optlen); 191 else 192 pr_err("%s: Failed to copy ip options\n", __func__); 193 } 194 RCU_INIT_POINTER(newinet->inet_opt, newopt); 195 rcu_read_unlock(); 196} 197 198/* Account for the IP options */ 199static int sctp_v4_ip_options_len(struct sock *sk) 200{ 201 struct inet_sock *inet = inet_sk(sk); 202 struct ip_options_rcu *inet_opt; 203 int len = 0; 204 205 rcu_read_lock(); 206 inet_opt = rcu_dereference(inet->inet_opt); 207 if (inet_opt) 208 len = inet_opt->opt.optlen; 209 210 rcu_read_unlock(); 211 return len; 212} 213 214/* Initialize a sctp_addr from in incoming skb. */ 215static void sctp_v4_from_skb(union sctp_addr *addr, struct sk_buff *skb, 216 int is_saddr) 217{ 218 /* Always called on head skb, so this is safe */ 219 struct sctphdr *sh = sctp_hdr(skb); 220 struct sockaddr_in *sa = &addr->v4; 221 222 addr->v4.sin_family = AF_INET; 223 224 if (is_saddr) { 225 sa->sin_port = sh->source; 226 sa->sin_addr.s_addr = ip_hdr(skb)->saddr; 227 } else { 228 sa->sin_port = sh->dest; 229 sa->sin_addr.s_addr = ip_hdr(skb)->daddr; 230 } 231 memset(sa->sin_zero, 0, sizeof(sa->sin_zero)); 232} 233 234/* Initialize an sctp_addr from a socket. */ 235static void sctp_v4_from_sk(union sctp_addr *addr, struct sock *sk) 236{ 237 addr->v4.sin_family = AF_INET; 238 addr->v4.sin_port = 0; 239 addr->v4.sin_addr.s_addr = inet_sk(sk)->inet_rcv_saddr; 240 memset(addr->v4.sin_zero, 0, sizeof(addr->v4.sin_zero)); 241} 242 243/* Initialize sk->sk_rcv_saddr from sctp_addr. */ 244static void sctp_v4_to_sk_saddr(union sctp_addr *addr, struct sock *sk) 245{ 246 inet_sk(sk)->inet_rcv_saddr = addr->v4.sin_addr.s_addr; 247} 248 249/* Initialize sk->sk_daddr from sctp_addr. */ 250static void sctp_v4_to_sk_daddr(union sctp_addr *addr, struct sock *sk) 251{ 252 inet_sk(sk)->inet_daddr = addr->v4.sin_addr.s_addr; 253} 254 255/* Initialize a sctp_addr from an address parameter. */ 256static bool sctp_v4_from_addr_param(union sctp_addr *addr, 257 union sctp_addr_param *param, 258 __be16 port, int iif) 259{ 260 if (ntohs(param->v4.param_hdr.length) < sizeof(struct sctp_ipv4addr_param)) 261 return false; 262 263 addr->v4.sin_family = AF_INET; 264 addr->v4.sin_port = port; 265 addr->v4.sin_addr.s_addr = param->v4.addr.s_addr; 266 memset(addr->v4.sin_zero, 0, sizeof(addr->v4.sin_zero)); 267 268 return true; 269} 270 271/* Initialize an address parameter from a sctp_addr and return the length 272 * of the address parameter. 273 */ 274static int sctp_v4_to_addr_param(const union sctp_addr *addr, 275 union sctp_addr_param *param) 276{ 277 int length = sizeof(struct sctp_ipv4addr_param); 278 279 param->v4.param_hdr.type = SCTP_PARAM_IPV4_ADDRESS; 280 param->v4.param_hdr.length = htons(length); 281 param->v4.addr.s_addr = addr->v4.sin_addr.s_addr; 282 283 return length; 284} 285 286/* Initialize a sctp_addr from a dst_entry. */ 287static void sctp_v4_dst_saddr(union sctp_addr *saddr, struct flowi4 *fl4, 288 __be16 port) 289{ 290 saddr->v4.sin_family = AF_INET; 291 saddr->v4.sin_port = port; 292 saddr->v4.sin_addr.s_addr = fl4->saddr; 293 memset(saddr->v4.sin_zero, 0, sizeof(saddr->v4.sin_zero)); 294} 295 296/* Compare two addresses exactly. */ 297static int sctp_v4_cmp_addr(const union sctp_addr *addr1, 298 const union sctp_addr *addr2) 299{ 300 if (addr1->sa.sa_family != addr2->sa.sa_family) 301 return 0; 302 if (addr1->v4.sin_port != addr2->v4.sin_port) 303 return 0; 304 if (addr1->v4.sin_addr.s_addr != addr2->v4.sin_addr.s_addr) 305 return 0; 306 307 return 1; 308} 309 310/* Initialize addr struct to INADDR_ANY. */ 311static void sctp_v4_inaddr_any(union sctp_addr *addr, __be16 port) 312{ 313 addr->v4.sin_family = AF_INET; 314 addr->v4.sin_addr.s_addr = htonl(INADDR_ANY); 315 addr->v4.sin_port = port; 316 memset(addr->v4.sin_zero, 0, sizeof(addr->v4.sin_zero)); 317} 318 319/* Is this a wildcard address? */ 320static int sctp_v4_is_any(const union sctp_addr *addr) 321{ 322 return htonl(INADDR_ANY) == addr->v4.sin_addr.s_addr; 323} 324 325/* This function checks if the address is a valid address to be used for 326 * SCTP binding. 327 * 328 * Output: 329 * Return 0 - If the address is a non-unicast or an illegal address. 330 * Return 1 - If the address is a unicast. 331 */ 332static int sctp_v4_addr_valid(union sctp_addr *addr, 333 struct sctp_sock *sp, 334 const struct sk_buff *skb) 335{ 336 /* IPv4 addresses not allowed */ 337 if (sp && ipv6_only_sock(sctp_opt2sk(sp))) 338 return 0; 339 340 /* Is this a non-unicast address or a unusable SCTP address? */ 341 if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr)) 342 return 0; 343 344 /* Is this a broadcast address? */ 345 if (skb && skb_rtable(skb)->rt_flags & RTCF_BROADCAST) 346 return 0; 347 348 return 1; 349} 350 351/* Should this be available for binding? */ 352static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp) 353{ 354 struct net *net = sock_net(&sp->inet.sk); 355 int ret = inet_addr_type(net, addr->v4.sin_addr.s_addr); 356 357 358 if (addr->v4.sin_addr.s_addr != htonl(INADDR_ANY) && 359 ret != RTN_LOCAL && 360 !sp->inet.freebind && 361 !READ_ONCE(net->ipv4.sysctl_ip_nonlocal_bind)) 362 return 0; 363 364 if (ipv6_only_sock(sctp_opt2sk(sp))) 365 return 0; 366 367 return 1; 368} 369 370/* Checking the loopback, private and other address scopes as defined in 371 * RFC 1918. The IPv4 scoping is based on the draft for SCTP IPv4 372 * scoping <draft-stewart-tsvwg-sctp-ipv4-00.txt>. 373 * 374 * Level 0 - unusable SCTP addresses 375 * Level 1 - loopback address 376 * Level 2 - link-local addresses 377 * Level 3 - private addresses. 378 * Level 4 - global addresses 379 * For INIT and INIT-ACK address list, let L be the level of 380 * requested destination address, sender and receiver 381 * SHOULD include all of its addresses with level greater 382 * than or equal to L. 383 * 384 * IPv4 scoping can be controlled through sysctl option 385 * net.sctp.addr_scope_policy 386 */ 387static enum sctp_scope sctp_v4_scope(union sctp_addr *addr) 388{ 389 enum sctp_scope retval; 390 391 /* Check for unusable SCTP addresses. */ 392 if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr)) { 393 retval = SCTP_SCOPE_UNUSABLE; 394 } else if (ipv4_is_loopback(addr->v4.sin_addr.s_addr)) { 395 retval = SCTP_SCOPE_LOOPBACK; 396 } else if (ipv4_is_linklocal_169(addr->v4.sin_addr.s_addr)) { 397 retval = SCTP_SCOPE_LINK; 398 } else if (ipv4_is_private_10(addr->v4.sin_addr.s_addr) || 399 ipv4_is_private_172(addr->v4.sin_addr.s_addr) || 400 ipv4_is_private_192(addr->v4.sin_addr.s_addr) || 401 ipv4_is_test_198(addr->v4.sin_addr.s_addr)) { 402 retval = SCTP_SCOPE_PRIVATE; 403 } else { 404 retval = SCTP_SCOPE_GLOBAL; 405 } 406 407 return retval; 408} 409 410/* Returns a valid dst cache entry for the given source and destination ip 411 * addresses. If an association is passed, trys to get a dst entry with a 412 * source address that matches an address in the bind address list. 413 */ 414static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr, 415 struct flowi *fl, struct sock *sk) 416{ 417 struct sctp_association *asoc = t->asoc; 418 struct rtable *rt; 419 struct flowi _fl; 420 struct flowi4 *fl4 = &_fl.u.ip4; 421 struct sctp_bind_addr *bp; 422 struct sctp_sockaddr_entry *laddr; 423 struct dst_entry *dst = NULL; 424 union sctp_addr *daddr = &t->ipaddr; 425 union sctp_addr dst_saddr; 426 __u8 tos = inet_sk(sk)->tos; 427 428 if (t->dscp & SCTP_DSCP_SET_MASK) 429 tos = t->dscp & SCTP_DSCP_VAL_MASK; 430 memset(&_fl, 0x0, sizeof(_fl)); 431 fl4->daddr = daddr->v4.sin_addr.s_addr; 432 fl4->fl4_dport = daddr->v4.sin_port; 433 fl4->flowi4_proto = IPPROTO_SCTP; 434 if (asoc) { 435 fl4->flowi4_tos = RT_CONN_FLAGS_TOS(asoc->base.sk, tos); 436 fl4->flowi4_oif = asoc->base.sk->sk_bound_dev_if; 437 fl4->fl4_sport = htons(asoc->base.bind_addr.port); 438 } 439 if (saddr) { 440 fl4->saddr = saddr->v4.sin_addr.s_addr; 441 if (!fl4->fl4_sport) 442 fl4->fl4_sport = saddr->v4.sin_port; 443 } 444 445 pr_debug("%s: dst:%pI4, src:%pI4 - ", __func__, &fl4->daddr, 446 &fl4->saddr); 447 448 rt = ip_route_output_key(sock_net(sk), fl4); 449 if (!IS_ERR(rt)) { 450 dst = &rt->dst; 451 t->dst = dst; 452 memcpy(fl, &_fl, sizeof(_fl)); 453 } 454 455 /* If there is no association or if a source address is passed, no 456 * more validation is required. 457 */ 458 if (!asoc || saddr) 459 goto out; 460 461 bp = &asoc->base.bind_addr; 462 463 if (dst) { 464 /* Walk through the bind address list and look for a bind 465 * address that matches the source address of the returned dst. 466 */ 467 sctp_v4_dst_saddr(&dst_saddr, fl4, htons(bp->port)); 468 rcu_read_lock(); 469 list_for_each_entry_rcu(laddr, &bp->address_list, list) { 470 if (!laddr->valid || (laddr->state == SCTP_ADDR_DEL) || 471 (laddr->state != SCTP_ADDR_SRC && 472 !asoc->src_out_of_asoc_ok)) 473 continue; 474 if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a)) 475 goto out_unlock; 476 } 477 rcu_read_unlock(); 478 479 /* None of the bound addresses match the source address of the 480 * dst. So release it. 481 */ 482 dst_release(dst); 483 dst = NULL; 484 } 485 486 /* Walk through the bind address list and try to get a dst that 487 * matches a bind address as the source address. 488 */ 489 rcu_read_lock(); 490 list_for_each_entry_rcu(laddr, &bp->address_list, list) { 491 struct net_device *odev; 492 493 if (!laddr->valid) 494 continue; 495 if (laddr->state != SCTP_ADDR_SRC || 496 AF_INET != laddr->a.sa.sa_family) 497 continue; 498 499 fl4->fl4_sport = laddr->a.v4.sin_port; 500 flowi4_update_output(fl4, 501 asoc->base.sk->sk_bound_dev_if, 502 RT_CONN_FLAGS_TOS(asoc->base.sk, tos), 503 daddr->v4.sin_addr.s_addr, 504 laddr->a.v4.sin_addr.s_addr); 505 506 rt = ip_route_output_key(sock_net(sk), fl4); 507 if (IS_ERR(rt)) 508 continue; 509 510 /* Ensure the src address belongs to the output 511 * interface. 512 */ 513 odev = __ip_dev_find(sock_net(sk), laddr->a.v4.sin_addr.s_addr, 514 false); 515 if (!odev || odev->ifindex != fl4->flowi4_oif) { 516 if (!dst) { 517 dst = &rt->dst; 518 t->dst = dst; 519 memcpy(fl, &_fl, sizeof(_fl)); 520 } else { 521 dst_release(&rt->dst); 522 } 523 continue; 524 } 525 526 dst_release(dst); 527 dst = &rt->dst; 528 t->dst = dst; 529 memcpy(fl, &_fl, sizeof(_fl)); 530 break; 531 } 532 533out_unlock: 534 rcu_read_unlock(); 535out: 536 if (dst) { 537 pr_debug("rt_dst:%pI4, rt_src:%pI4\n", 538 &fl->u.ip4.daddr, &fl->u.ip4.saddr); 539 } else { 540 t->dst = NULL; 541 pr_debug("no route\n"); 542 } 543} 544 545/* For v4, the source address is cached in the route entry(dst). So no need 546 * to cache it separately and hence this is an empty routine. 547 */ 548static void sctp_v4_get_saddr(struct sctp_sock *sk, 549 struct sctp_transport *t, 550 struct flowi *fl) 551{ 552 union sctp_addr *saddr = &t->saddr; 553 struct rtable *rt = (struct rtable *)t->dst; 554 555 if (rt) { 556 saddr->v4.sin_family = AF_INET; 557 saddr->v4.sin_addr.s_addr = fl->u.ip4.saddr; 558 } 559} 560 561/* What interface did this skb arrive on? */ 562static int sctp_v4_skb_iif(const struct sk_buff *skb) 563{ 564 return inet_iif(skb); 565} 566 567/* Was this packet marked by Explicit Congestion Notification? */ 568static int sctp_v4_is_ce(const struct sk_buff *skb) 569{ 570 return INET_ECN_is_ce(ip_hdr(skb)->tos); 571} 572 573/* Create and initialize a new sk for the socket returned by accept(). */ 574static struct sock *sctp_v4_create_accept_sk(struct sock *sk, 575 struct sctp_association *asoc, 576 bool kern) 577{ 578 struct sock *newsk = sk_alloc(sock_net(sk), PF_INET, GFP_KERNEL, 579 sk->sk_prot, kern); 580 struct inet_sock *newinet; 581 582 if (!newsk) 583 goto out; 584 585 sock_init_data(NULL, newsk); 586 587 sctp_copy_sock(newsk, sk, asoc); 588 sock_reset_flag(newsk, SOCK_ZAPPED); 589 590 sctp_v4_copy_ip_options(sk, newsk); 591 592 newinet = inet_sk(newsk); 593 594 newinet->inet_daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr; 595 596 sk_refcnt_debug_inc(newsk); 597 598 if (newsk->sk_prot->init(newsk)) { 599 sk_common_release(newsk); 600 newsk = NULL; 601 } 602 603out: 604 return newsk; 605} 606 607static int sctp_v4_addr_to_user(struct sctp_sock *sp, union sctp_addr *addr) 608{ 609 /* No address mapping for V4 sockets */ 610 memset(addr->v4.sin_zero, 0, sizeof(addr->v4.sin_zero)); 611 return sizeof(struct sockaddr_in); 612} 613 614/* Dump the v4 addr to the seq file. */ 615static void sctp_v4_seq_dump_addr(struct seq_file *seq, union sctp_addr *addr) 616{ 617 seq_printf(seq, "%pI4 ", &addr->v4.sin_addr); 618} 619 620static void sctp_v4_ecn_capable(struct sock *sk) 621{ 622 INET_ECN_xmit(sk); 623} 624 625static void sctp_addr_wq_timeout_handler(struct timer_list *t) 626{ 627 struct net *net = from_timer(net, t, sctp.addr_wq_timer); 628 struct sctp_sockaddr_entry *addrw, *temp; 629 struct sctp_sock *sp; 630 631 spin_lock_bh(&net->sctp.addr_wq_lock); 632 633 list_for_each_entry_safe(addrw, temp, &net->sctp.addr_waitq, list) { 634 pr_debug("%s: the first ent in wq:%p is addr:%pISc for cmd:%d at " 635 "entry:%p\n", __func__, &net->sctp.addr_waitq, &addrw->a.sa, 636 addrw->state, addrw); 637 638#if IS_ENABLED(CONFIG_IPV6) 639 /* Now we send an ASCONF for each association */ 640 /* Note. we currently don't handle link local IPv6 addressees */ 641 if (addrw->a.sa.sa_family == AF_INET6) { 642 struct in6_addr *in6; 643 644 if (ipv6_addr_type(&addrw->a.v6.sin6_addr) & 645 IPV6_ADDR_LINKLOCAL) 646 goto free_next; 647 648 in6 = (struct in6_addr *)&addrw->a.v6.sin6_addr; 649 if (ipv6_chk_addr(net, in6, NULL, 0) == 0 && 650 addrw->state == SCTP_ADDR_NEW) { 651 unsigned long timeo_val; 652 653 pr_debug("%s: this is on DAD, trying %d sec " 654 "later\n", __func__, 655 SCTP_ADDRESS_TICK_DELAY); 656 657 timeo_val = jiffies; 658 timeo_val += msecs_to_jiffies(SCTP_ADDRESS_TICK_DELAY); 659 mod_timer(&net->sctp.addr_wq_timer, timeo_val); 660 break; 661 } 662 } 663#endif 664 list_for_each_entry(sp, &net->sctp.auto_asconf_splist, auto_asconf_list) { 665 struct sock *sk; 666 667 sk = sctp_opt2sk(sp); 668 /* ignore bound-specific endpoints */ 669 if (!sctp_is_ep_boundall(sk)) 670 continue; 671 bh_lock_sock(sk); 672 if (sctp_asconf_mgmt(sp, addrw) < 0) 673 pr_debug("%s: sctp_asconf_mgmt failed\n", __func__); 674 bh_unlock_sock(sk); 675 } 676#if IS_ENABLED(CONFIG_IPV6) 677free_next: 678#endif 679 list_del(&addrw->list); 680 kfree(addrw); 681 } 682 spin_unlock_bh(&net->sctp.addr_wq_lock); 683} 684 685static void sctp_free_addr_wq(struct net *net) 686{ 687 struct sctp_sockaddr_entry *addrw; 688 struct sctp_sockaddr_entry *temp; 689 690 spin_lock_bh(&net->sctp.addr_wq_lock); 691 del_timer(&net->sctp.addr_wq_timer); 692 list_for_each_entry_safe(addrw, temp, &net->sctp.addr_waitq, list) { 693 list_del(&addrw->list); 694 kfree(addrw); 695 } 696 spin_unlock_bh(&net->sctp.addr_wq_lock); 697} 698 699/* lookup the entry for the same address in the addr_waitq 700 * sctp_addr_wq MUST be locked 701 */ 702static struct sctp_sockaddr_entry *sctp_addr_wq_lookup(struct net *net, 703 struct sctp_sockaddr_entry *addr) 704{ 705 struct sctp_sockaddr_entry *addrw; 706 707 list_for_each_entry(addrw, &net->sctp.addr_waitq, list) { 708 if (addrw->a.sa.sa_family != addr->a.sa.sa_family) 709 continue; 710 if (addrw->a.sa.sa_family == AF_INET) { 711 if (addrw->a.v4.sin_addr.s_addr == 712 addr->a.v4.sin_addr.s_addr) 713 return addrw; 714 } else if (addrw->a.sa.sa_family == AF_INET6) { 715 if (ipv6_addr_equal(&addrw->a.v6.sin6_addr, 716 &addr->a.v6.sin6_addr)) 717 return addrw; 718 } 719 } 720 return NULL; 721} 722 723void sctp_addr_wq_mgmt(struct net *net, struct sctp_sockaddr_entry *addr, int cmd) 724{ 725 struct sctp_sockaddr_entry *addrw; 726 unsigned long timeo_val; 727 728 /* first, we check if an opposite message already exist in the queue. 729 * If we found such message, it is removed. 730 * This operation is a bit stupid, but the DHCP client attaches the 731 * new address after a couple of addition and deletion of that address 732 */ 733 734 spin_lock_bh(&net->sctp.addr_wq_lock); 735 /* Offsets existing events in addr_wq */ 736 addrw = sctp_addr_wq_lookup(net, addr); 737 if (addrw) { 738 if (addrw->state != cmd) { 739 pr_debug("%s: offsets existing entry for %d, addr:%pISc " 740 "in wq:%p\n", __func__, addrw->state, &addrw->a.sa, 741 &net->sctp.addr_waitq); 742 743 list_del(&addrw->list); 744 kfree(addrw); 745 } 746 spin_unlock_bh(&net->sctp.addr_wq_lock); 747 return; 748 } 749 750 /* OK, we have to add the new address to the wait queue */ 751 addrw = kmemdup(addr, sizeof(struct sctp_sockaddr_entry), GFP_ATOMIC); 752 if (addrw == NULL) { 753 spin_unlock_bh(&net->sctp.addr_wq_lock); 754 return; 755 } 756 addrw->state = cmd; 757 list_add_tail(&addrw->list, &net->sctp.addr_waitq); 758 759 pr_debug("%s: add new entry for cmd:%d, addr:%pISc in wq:%p\n", 760 __func__, addrw->state, &addrw->a.sa, &net->sctp.addr_waitq); 761 762 if (!timer_pending(&net->sctp.addr_wq_timer)) { 763 timeo_val = jiffies; 764 timeo_val += msecs_to_jiffies(SCTP_ADDRESS_TICK_DELAY); 765 mod_timer(&net->sctp.addr_wq_timer, timeo_val); 766 } 767 spin_unlock_bh(&net->sctp.addr_wq_lock); 768} 769 770/* Event handler for inet address addition/deletion events. 771 * The sctp_local_addr_list needs to be protocted by a spin lock since 772 * multiple notifiers (say IPv4 and IPv6) may be running at the same 773 * time and thus corrupt the list. 774 * The reader side is protected with RCU. 775 */ 776static int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev, 777 void *ptr) 778{ 779 struct in_ifaddr *ifa = (struct in_ifaddr *)ptr; 780 struct sctp_sockaddr_entry *addr = NULL; 781 struct sctp_sockaddr_entry *temp; 782 struct net *net = dev_net(ifa->ifa_dev->dev); 783 int found = 0; 784 785 switch (ev) { 786 case NETDEV_UP: 787 addr = kzalloc(sizeof(*addr), GFP_ATOMIC); 788 if (addr) { 789 addr->a.v4.sin_family = AF_INET; 790 addr->a.v4.sin_addr.s_addr = ifa->ifa_local; 791 addr->valid = 1; 792 spin_lock_bh(&net->sctp.local_addr_lock); 793 list_add_tail_rcu(&addr->list, &net->sctp.local_addr_list); 794 sctp_addr_wq_mgmt(net, addr, SCTP_ADDR_NEW); 795 spin_unlock_bh(&net->sctp.local_addr_lock); 796 } 797 break; 798 case NETDEV_DOWN: 799 spin_lock_bh(&net->sctp.local_addr_lock); 800 list_for_each_entry_safe(addr, temp, 801 &net->sctp.local_addr_list, list) { 802 if (addr->a.sa.sa_family == AF_INET && 803 addr->a.v4.sin_addr.s_addr == 804 ifa->ifa_local) { 805 sctp_addr_wq_mgmt(net, addr, SCTP_ADDR_DEL); 806 found = 1; 807 addr->valid = 0; 808 list_del_rcu(&addr->list); 809 break; 810 } 811 } 812 spin_unlock_bh(&net->sctp.local_addr_lock); 813 if (found) 814 kfree_rcu(addr, rcu); 815 break; 816 } 817 818 return NOTIFY_DONE; 819} 820 821/* 822 * Initialize the control inode/socket with a control endpoint data 823 * structure. This endpoint is reserved exclusively for the OOTB processing. 824 */ 825static int sctp_ctl_sock_init(struct net *net) 826{ 827 int err; 828 sa_family_t family = PF_INET; 829 830 if (sctp_get_pf_specific(PF_INET6)) 831 family = PF_INET6; 832 833 err = inet_ctl_sock_create(&net->sctp.ctl_sock, family, 834 SOCK_SEQPACKET, IPPROTO_SCTP, net); 835 836 /* If IPv6 socket could not be created, try the IPv4 socket */ 837 if (err < 0 && family == PF_INET6) 838 err = inet_ctl_sock_create(&net->sctp.ctl_sock, AF_INET, 839 SOCK_SEQPACKET, IPPROTO_SCTP, 840 net); 841 842 if (err < 0) { 843 pr_err("Failed to create the SCTP control socket\n"); 844 return err; 845 } 846 return 0; 847} 848 849/* Register address family specific functions. */ 850int sctp_register_af(struct sctp_af *af) 851{ 852 switch (af->sa_family) { 853 case AF_INET: 854 if (sctp_af_v4_specific) 855 return 0; 856 sctp_af_v4_specific = af; 857 break; 858 case AF_INET6: 859 if (sctp_af_v6_specific) 860 return 0; 861 sctp_af_v6_specific = af; 862 break; 863 default: 864 return 0; 865 } 866 867 INIT_LIST_HEAD(&af->list); 868 list_add_tail(&af->list, &sctp_address_families); 869 return 1; 870} 871 872/* Get the table of functions for manipulating a particular address 873 * family. 874 */ 875struct sctp_af *sctp_get_af_specific(sa_family_t family) 876{ 877 switch (family) { 878 case AF_INET: 879 return sctp_af_v4_specific; 880 case AF_INET6: 881 return sctp_af_v6_specific; 882 default: 883 return NULL; 884 } 885} 886 887/* Common code to initialize a AF_INET msg_name. */ 888static void sctp_inet_msgname(char *msgname, int *addr_len) 889{ 890 struct sockaddr_in *sin; 891 892 sin = (struct sockaddr_in *)msgname; 893 *addr_len = sizeof(struct sockaddr_in); 894 sin->sin_family = AF_INET; 895 memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); 896} 897 898/* Copy the primary address of the peer primary address as the msg_name. */ 899static void sctp_inet_event_msgname(struct sctp_ulpevent *event, char *msgname, 900 int *addr_len) 901{ 902 struct sockaddr_in *sin, *sinfrom; 903 904 if (msgname) { 905 struct sctp_association *asoc; 906 907 asoc = event->asoc; 908 sctp_inet_msgname(msgname, addr_len); 909 sin = (struct sockaddr_in *)msgname; 910 sinfrom = &asoc->peer.primary_addr.v4; 911 sin->sin_port = htons(asoc->peer.port); 912 sin->sin_addr.s_addr = sinfrom->sin_addr.s_addr; 913 } 914} 915 916/* Initialize and copy out a msgname from an inbound skb. */ 917static void sctp_inet_skb_msgname(struct sk_buff *skb, char *msgname, int *len) 918{ 919 if (msgname) { 920 struct sctphdr *sh = sctp_hdr(skb); 921 struct sockaddr_in *sin = (struct sockaddr_in *)msgname; 922 923 sctp_inet_msgname(msgname, len); 924 sin->sin_port = sh->source; 925 sin->sin_addr.s_addr = ip_hdr(skb)->saddr; 926 } 927} 928 929/* Do we support this AF? */ 930static int sctp_inet_af_supported(sa_family_t family, struct sctp_sock *sp) 931{ 932 /* PF_INET only supports AF_INET addresses. */ 933 return AF_INET == family; 934} 935 936/* Address matching with wildcards allowed. */ 937static int sctp_inet_cmp_addr(const union sctp_addr *addr1, 938 const union sctp_addr *addr2, 939 struct sctp_sock *opt) 940{ 941 /* PF_INET only supports AF_INET addresses. */ 942 if (addr1->sa.sa_family != addr2->sa.sa_family) 943 return 0; 944 if (htonl(INADDR_ANY) == addr1->v4.sin_addr.s_addr || 945 htonl(INADDR_ANY) == addr2->v4.sin_addr.s_addr) 946 return 1; 947 if (addr1->v4.sin_addr.s_addr == addr2->v4.sin_addr.s_addr) 948 return 1; 949 950 return 0; 951} 952 953/* Verify that provided sockaddr looks bindable. Common verification has 954 * already been taken care of. 955 */ 956static int sctp_inet_bind_verify(struct sctp_sock *opt, union sctp_addr *addr) 957{ 958 return sctp_v4_available(addr, opt); 959} 960 961/* Verify that sockaddr looks sendable. Common verification has already 962 * been taken care of. 963 */ 964static int sctp_inet_send_verify(struct sctp_sock *opt, union sctp_addr *addr) 965{ 966 return 1; 967} 968 969/* Fill in Supported Address Type information for INIT and INIT-ACK 970 * chunks. Returns number of addresses supported. 971 */ 972static int sctp_inet_supported_addrs(const struct sctp_sock *opt, 973 __be16 *types) 974{ 975 types[0] = SCTP_PARAM_IPV4_ADDRESS; 976 return 1; 977} 978 979/* Wrapper routine that calls the ip transmit routine. */ 980static inline int sctp_v4_xmit(struct sk_buff *skb, 981 struct sctp_transport *transport) 982{ 983 struct inet_sock *inet = inet_sk(skb->sk); 984 __u8 dscp = inet->tos; 985 986 pr_debug("%s: skb:%p, len:%d, src:%pI4, dst:%pI4\n", __func__, skb, 987 skb->len, &transport->fl.u.ip4.saddr, 988 &transport->fl.u.ip4.daddr); 989 990 if (transport->dscp & SCTP_DSCP_SET_MASK) 991 dscp = transport->dscp & SCTP_DSCP_VAL_MASK; 992 993 inet->pmtudisc = transport->param_flags & SPP_PMTUD_ENABLE ? 994 IP_PMTUDISC_DO : IP_PMTUDISC_DONT; 995 996 SCTP_INC_STATS(sock_net(&inet->sk), SCTP_MIB_OUTSCTPPACKS); 997 998 return __ip_queue_xmit(&inet->sk, skb, &transport->fl, dscp); 999} 1000 1001static struct sctp_af sctp_af_inet; 1002 1003static struct sctp_pf sctp_pf_inet = { 1004 .event_msgname = sctp_inet_event_msgname, 1005 .skb_msgname = sctp_inet_skb_msgname, 1006 .af_supported = sctp_inet_af_supported, 1007 .cmp_addr = sctp_inet_cmp_addr, 1008 .bind_verify = sctp_inet_bind_verify, 1009 .send_verify = sctp_inet_send_verify, 1010 .supported_addrs = sctp_inet_supported_addrs, 1011 .create_accept_sk = sctp_v4_create_accept_sk, 1012 .addr_to_user = sctp_v4_addr_to_user, 1013 .to_sk_saddr = sctp_v4_to_sk_saddr, 1014 .to_sk_daddr = sctp_v4_to_sk_daddr, 1015 .copy_ip_options = sctp_v4_copy_ip_options, 1016 .af = &sctp_af_inet 1017}; 1018 1019/* Notifier for inetaddr addition/deletion events. */ 1020static struct notifier_block sctp_inetaddr_notifier = { 1021 .notifier_call = sctp_inetaddr_event, 1022}; 1023 1024/* Socket operations. */ 1025static const struct proto_ops inet_seqpacket_ops = { 1026 .family = PF_INET, 1027 .owner = THIS_MODULE, 1028 .release = inet_release, /* Needs to be wrapped... */ 1029 .bind = inet_bind, 1030 .connect = sctp_inet_connect, 1031 .socketpair = sock_no_socketpair, 1032 .accept = inet_accept, 1033 .getname = inet_getname, /* Semantics are different. */ 1034 .poll = sctp_poll, 1035 .ioctl = inet_ioctl, 1036 .gettstamp = sock_gettstamp, 1037 .listen = sctp_inet_listen, 1038 .shutdown = inet_shutdown, /* Looks harmless. */ 1039 .setsockopt = sock_common_setsockopt, /* IP_SOL IP_OPTION is a problem */ 1040 .getsockopt = sock_common_getsockopt, 1041 .sendmsg = inet_sendmsg, 1042 .recvmsg = inet_recvmsg, 1043 .mmap = sock_no_mmap, 1044 .sendpage = sock_no_sendpage, 1045}; 1046 1047/* Registration with AF_INET family. */ 1048static struct inet_protosw sctp_seqpacket_protosw = { 1049 .type = SOCK_SEQPACKET, 1050 .protocol = IPPROTO_SCTP, 1051 .prot = &sctp_prot, 1052 .ops = &inet_seqpacket_ops, 1053 .flags = SCTP_PROTOSW_FLAG 1054}; 1055static struct inet_protosw sctp_stream_protosw = { 1056 .type = SOCK_STREAM, 1057 .protocol = IPPROTO_SCTP, 1058 .prot = &sctp_prot, 1059 .ops = &inet_seqpacket_ops, 1060 .flags = SCTP_PROTOSW_FLAG 1061}; 1062 1063/* Register with IP layer. */ 1064static const struct net_protocol sctp_protocol = { 1065 .handler = sctp_rcv, 1066 .err_handler = sctp_v4_err, 1067 .no_policy = 1, 1068 .netns_ok = 1, 1069 .icmp_strict_tag_validation = 1, 1070}; 1071 1072/* IPv4 address related functions. */ 1073static struct sctp_af sctp_af_inet = { 1074 .sa_family = AF_INET, 1075 .sctp_xmit = sctp_v4_xmit, 1076 .setsockopt = ip_setsockopt, 1077 .getsockopt = ip_getsockopt, 1078 .get_dst = sctp_v4_get_dst, 1079 .get_saddr = sctp_v4_get_saddr, 1080 .copy_addrlist = sctp_v4_copy_addrlist, 1081 .from_skb = sctp_v4_from_skb, 1082 .from_sk = sctp_v4_from_sk, 1083 .from_addr_param = sctp_v4_from_addr_param, 1084 .to_addr_param = sctp_v4_to_addr_param, 1085 .cmp_addr = sctp_v4_cmp_addr, 1086 .addr_valid = sctp_v4_addr_valid, 1087 .inaddr_any = sctp_v4_inaddr_any, 1088 .is_any = sctp_v4_is_any, 1089 .available = sctp_v4_available, 1090 .scope = sctp_v4_scope, 1091 .skb_iif = sctp_v4_skb_iif, 1092 .is_ce = sctp_v4_is_ce, 1093 .seq_dump_addr = sctp_v4_seq_dump_addr, 1094 .ecn_capable = sctp_v4_ecn_capable, 1095 .net_header_len = sizeof(struct iphdr), 1096 .sockaddr_len = sizeof(struct sockaddr_in), 1097 .ip_options_len = sctp_v4_ip_options_len, 1098}; 1099 1100struct sctp_pf *sctp_get_pf_specific(sa_family_t family) 1101{ 1102 switch (family) { 1103 case PF_INET: 1104 return sctp_pf_inet_specific; 1105 case PF_INET6: 1106 return sctp_pf_inet6_specific; 1107 default: 1108 return NULL; 1109 } 1110} 1111 1112/* Register the PF specific function table. */ 1113int sctp_register_pf(struct sctp_pf *pf, sa_family_t family) 1114{ 1115 switch (family) { 1116 case PF_INET: 1117 if (sctp_pf_inet_specific) 1118 return 0; 1119 sctp_pf_inet_specific = pf; 1120 break; 1121 case PF_INET6: 1122 if (sctp_pf_inet6_specific) 1123 return 0; 1124 sctp_pf_inet6_specific = pf; 1125 break; 1126 default: 1127 return 0; 1128 } 1129 return 1; 1130} 1131 1132static inline int init_sctp_mibs(struct net *net) 1133{ 1134 net->sctp.sctp_statistics = alloc_percpu(struct sctp_mib); 1135 if (!net->sctp.sctp_statistics) 1136 return -ENOMEM; 1137 return 0; 1138} 1139 1140static inline void cleanup_sctp_mibs(struct net *net) 1141{ 1142 free_percpu(net->sctp.sctp_statistics); 1143} 1144 1145static void sctp_v4_pf_init(void) 1146{ 1147 /* Initialize the SCTP specific PF functions. */ 1148 sctp_register_pf(&sctp_pf_inet, PF_INET); 1149 sctp_register_af(&sctp_af_inet); 1150} 1151 1152static void sctp_v4_pf_exit(void) 1153{ 1154 list_del(&sctp_af_inet.list); 1155} 1156 1157static int sctp_v4_protosw_init(void) 1158{ 1159 int rc; 1160 1161 rc = proto_register(&sctp_prot, 1); 1162 if (rc) 1163 return rc; 1164 1165 /* Register SCTP(UDP and TCP style) with socket layer. */ 1166 inet_register_protosw(&sctp_seqpacket_protosw); 1167 inet_register_protosw(&sctp_stream_protosw); 1168 1169 return 0; 1170} 1171 1172static void sctp_v4_protosw_exit(void) 1173{ 1174 inet_unregister_protosw(&sctp_stream_protosw); 1175 inet_unregister_protosw(&sctp_seqpacket_protosw); 1176 proto_unregister(&sctp_prot); 1177} 1178 1179static int sctp_v4_add_protocol(void) 1180{ 1181 /* Register notifier for inet address additions/deletions. */ 1182 register_inetaddr_notifier(&sctp_inetaddr_notifier); 1183 1184 /* Register SCTP with inet layer. */ 1185 if (inet_add_protocol(&sctp_protocol, IPPROTO_SCTP) < 0) 1186 return -EAGAIN; 1187 1188 return 0; 1189} 1190 1191static void sctp_v4_del_protocol(void) 1192{ 1193 inet_del_protocol(&sctp_protocol, IPPROTO_SCTP); 1194 unregister_inetaddr_notifier(&sctp_inetaddr_notifier); 1195} 1196 1197static int __net_init sctp_defaults_init(struct net *net) 1198{ 1199 int status; 1200 1201 /* 1202 * 14. Suggested SCTP Protocol Parameter Values 1203 */ 1204 /* The following protocol parameters are RECOMMENDED: */ 1205 /* RTO.Initial - 3 seconds */ 1206 net->sctp.rto_initial = SCTP_RTO_INITIAL; 1207 /* RTO.Min - 1 second */ 1208 net->sctp.rto_min = SCTP_RTO_MIN; 1209 /* RTO.Max - 60 seconds */ 1210 net->sctp.rto_max = SCTP_RTO_MAX; 1211 /* RTO.Alpha - 1/8 */ 1212 net->sctp.rto_alpha = SCTP_RTO_ALPHA; 1213 /* RTO.Beta - 1/4 */ 1214 net->sctp.rto_beta = SCTP_RTO_BETA; 1215 1216 /* Valid.Cookie.Life - 60 seconds */ 1217 net->sctp.valid_cookie_life = SCTP_DEFAULT_COOKIE_LIFE; 1218 1219 /* Whether Cookie Preservative is enabled(1) or not(0) */ 1220 net->sctp.cookie_preserve_enable = 1; 1221 1222 /* Default sctp sockets to use md5 as their hmac alg */ 1223#if defined (CONFIG_SCTP_DEFAULT_COOKIE_HMAC_MD5) 1224 net->sctp.sctp_hmac_alg = "md5"; 1225#elif defined (CONFIG_SCTP_DEFAULT_COOKIE_HMAC_SHA1) 1226 net->sctp.sctp_hmac_alg = "sha1"; 1227#else 1228 net->sctp.sctp_hmac_alg = NULL; 1229#endif 1230 1231 /* Max.Burst - 4 */ 1232 net->sctp.max_burst = SCTP_DEFAULT_MAX_BURST; 1233 1234 /* Disable of Primary Path Switchover by default */ 1235 net->sctp.ps_retrans = SCTP_PS_RETRANS_MAX; 1236 1237 /* Enable pf state by default */ 1238 net->sctp.pf_enable = 1; 1239 1240 /* Ignore pf exposure feature by default */ 1241 net->sctp.pf_expose = SCTP_PF_EXPOSE_UNSET; 1242 1243 /* Association.Max.Retrans - 10 attempts 1244 * Path.Max.Retrans - 5 attempts (per destination address) 1245 * Max.Init.Retransmits - 8 attempts 1246 */ 1247 net->sctp.max_retrans_association = 10; 1248 net->sctp.max_retrans_path = 5; 1249 net->sctp.max_retrans_init = 8; 1250 1251 /* Sendbuffer growth - do per-socket accounting */ 1252 net->sctp.sndbuf_policy = 0; 1253 1254 /* Rcvbuffer growth - do per-socket accounting */ 1255 net->sctp.rcvbuf_policy = 0; 1256 1257 /* HB.interval - 30 seconds */ 1258 net->sctp.hb_interval = SCTP_DEFAULT_TIMEOUT_HEARTBEAT; 1259 1260 /* delayed SACK timeout */ 1261 net->sctp.sack_timeout = SCTP_DEFAULT_TIMEOUT_SACK; 1262 1263 /* Disable ADDIP by default. */ 1264 net->sctp.addip_enable = 0; 1265 net->sctp.addip_noauth = 0; 1266 net->sctp.default_auto_asconf = 0; 1267 1268 /* Enable PR-SCTP by default. */ 1269 net->sctp.prsctp_enable = 1; 1270 1271 /* Disable RECONF by default. */ 1272 net->sctp.reconf_enable = 0; 1273 1274 /* Disable AUTH by default. */ 1275 net->sctp.auth_enable = 0; 1276 1277 /* Enable ECN by default. */ 1278 net->sctp.ecn_enable = 1; 1279 1280 /* Set SCOPE policy to enabled */ 1281 net->sctp.scope_policy = SCTP_SCOPE_POLICY_ENABLE; 1282 1283 /* Set the default rwnd update threshold */ 1284 net->sctp.rwnd_upd_shift = SCTP_DEFAULT_RWND_SHIFT; 1285 1286 /* Initialize maximum autoclose timeout. */ 1287 net->sctp.max_autoclose = INT_MAX / HZ; 1288 1289 status = sctp_sysctl_net_register(net); 1290 if (status) 1291 goto err_sysctl_register; 1292 1293 /* Allocate and initialise sctp mibs. */ 1294 status = init_sctp_mibs(net); 1295 if (status) 1296 goto err_init_mibs; 1297 1298#ifdef CONFIG_PROC_FS 1299 /* Initialize proc fs directory. */ 1300 status = sctp_proc_init(net); 1301 if (status) 1302 goto err_init_proc; 1303#endif 1304 1305 sctp_dbg_objcnt_init(net); 1306 1307 /* Initialize the local address list. */ 1308 INIT_LIST_HEAD(&net->sctp.local_addr_list); 1309 spin_lock_init(&net->sctp.local_addr_lock); 1310 sctp_get_local_addr_list(net); 1311 1312 /* Initialize the address event list */ 1313 INIT_LIST_HEAD(&net->sctp.addr_waitq); 1314 INIT_LIST_HEAD(&net->sctp.auto_asconf_splist); 1315 spin_lock_init(&net->sctp.addr_wq_lock); 1316 net->sctp.addr_wq_timer.expires = 0; 1317 timer_setup(&net->sctp.addr_wq_timer, sctp_addr_wq_timeout_handler, 0); 1318 1319 return 0; 1320 1321#ifdef CONFIG_PROC_FS 1322err_init_proc: 1323 cleanup_sctp_mibs(net); 1324#endif 1325err_init_mibs: 1326 sctp_sysctl_net_unregister(net); 1327err_sysctl_register: 1328 return status; 1329} 1330 1331static void __net_exit sctp_defaults_exit(struct net *net) 1332{ 1333 /* Free the local address list */ 1334 sctp_free_addr_wq(net); 1335 sctp_free_local_addr_list(net); 1336 1337#ifdef CONFIG_PROC_FS 1338 remove_proc_subtree("sctp", net->proc_net); 1339 net->sctp.proc_net_sctp = NULL; 1340#endif 1341 cleanup_sctp_mibs(net); 1342 sctp_sysctl_net_unregister(net); 1343} 1344 1345static struct pernet_operations sctp_defaults_ops = { 1346 .init = sctp_defaults_init, 1347 .exit = sctp_defaults_exit, 1348}; 1349 1350static int __net_init sctp_ctrlsock_init(struct net *net) 1351{ 1352 int status; 1353 1354 /* Initialize the control inode/socket for handling OOTB packets. */ 1355 status = sctp_ctl_sock_init(net); 1356 if (status) 1357 pr_err("Failed to initialize the SCTP control sock\n"); 1358 1359 return status; 1360} 1361 1362static void __net_exit sctp_ctrlsock_exit(struct net *net) 1363{ 1364 /* Free the control endpoint. */ 1365 inet_ctl_sock_destroy(net->sctp.ctl_sock); 1366} 1367 1368static struct pernet_operations sctp_ctrlsock_ops = { 1369 .init = sctp_ctrlsock_init, 1370 .exit = sctp_ctrlsock_exit, 1371}; 1372 1373/* Initialize the universe into something sensible. */ 1374static __init int sctp_init(void) 1375{ 1376 unsigned long nr_pages = totalram_pages(); 1377 unsigned long limit; 1378 unsigned long goal; 1379 int max_entry_order; 1380 int num_entries; 1381 int max_share; 1382 int status; 1383 int order; 1384 int i; 1385 1386 sock_skb_cb_check_size(sizeof(struct sctp_ulpevent)); 1387 1388 /* Allocate bind_bucket and chunk caches. */ 1389 status = -ENOBUFS; 1390 sctp_bucket_cachep = kmem_cache_create("sctp_bind_bucket", 1391 sizeof(struct sctp_bind_bucket), 1392 0, SLAB_HWCACHE_ALIGN, 1393 NULL); 1394 if (!sctp_bucket_cachep) 1395 goto out; 1396 1397 sctp_chunk_cachep = kmem_cache_create("sctp_chunk", 1398 sizeof(struct sctp_chunk), 1399 0, SLAB_HWCACHE_ALIGN, 1400 NULL); 1401 if (!sctp_chunk_cachep) 1402 goto err_chunk_cachep; 1403 1404 status = percpu_counter_init(&sctp_sockets_allocated, 0, GFP_KERNEL); 1405 if (status) 1406 goto err_percpu_counter_init; 1407 1408 /* Implementation specific variables. */ 1409 1410 /* Initialize default stream count setup information. */ 1411 sctp_max_instreams = SCTP_DEFAULT_INSTREAMS; 1412 sctp_max_outstreams = SCTP_DEFAULT_OUTSTREAMS; 1413 1414 /* Initialize handle used for association ids. */ 1415 idr_init(&sctp_assocs_id); 1416 1417 limit = nr_free_buffer_pages() / 8; 1418 limit = max(limit, 128UL); 1419 sysctl_sctp_mem[0] = limit / 4 * 3; 1420 sysctl_sctp_mem[1] = limit; 1421 sysctl_sctp_mem[2] = sysctl_sctp_mem[0] * 2; 1422 1423 /* Set per-socket limits to no more than 1/128 the pressure threshold*/ 1424 limit = (sysctl_sctp_mem[1]) << (PAGE_SHIFT - 7); 1425 max_share = min(4UL*1024*1024, limit); 1426 1427 sysctl_sctp_rmem[0] = SK_MEM_QUANTUM; /* give each asoc 1 page min */ 1428 sysctl_sctp_rmem[1] = 1500 * SKB_TRUESIZE(1); 1429 sysctl_sctp_rmem[2] = max(sysctl_sctp_rmem[1], max_share); 1430 1431 sysctl_sctp_wmem[0] = SK_MEM_QUANTUM; 1432 sysctl_sctp_wmem[1] = 16*1024; 1433 sysctl_sctp_wmem[2] = max(64*1024, max_share); 1434 1435 /* Size and allocate the association hash table. 1436 * The methodology is similar to that of the tcp hash tables. 1437 * Though not identical. Start by getting a goal size 1438 */ 1439 if (nr_pages >= (128 * 1024)) 1440 goal = nr_pages >> (22 - PAGE_SHIFT); 1441 else 1442 goal = nr_pages >> (24 - PAGE_SHIFT); 1443 1444 /* Then compute the page order for said goal */ 1445 order = get_order(goal); 1446 1447 /* Now compute the required page order for the maximum sized table we 1448 * want to create 1449 */ 1450 max_entry_order = get_order(MAX_SCTP_PORT_HASH_ENTRIES * 1451 sizeof(struct sctp_bind_hashbucket)); 1452 1453 /* Limit the page order by that maximum hash table size */ 1454 order = min(order, max_entry_order); 1455 1456 /* Allocate and initialize the endpoint hash table. */ 1457 sctp_ep_hashsize = 64; 1458 sctp_ep_hashtable = 1459 kmalloc_array(64, sizeof(struct sctp_hashbucket), GFP_KERNEL); 1460 if (!sctp_ep_hashtable) { 1461 pr_err("Failed endpoint_hash alloc\n"); 1462 status = -ENOMEM; 1463 goto err_ehash_alloc; 1464 } 1465 for (i = 0; i < sctp_ep_hashsize; i++) { 1466 rwlock_init(&sctp_ep_hashtable[i].lock); 1467 INIT_HLIST_HEAD(&sctp_ep_hashtable[i].chain); 1468 } 1469 1470 /* Allocate and initialize the SCTP port hash table. 1471 * Note that order is initalized to start at the max sized 1472 * table we want to support. If we can't get that many pages 1473 * reduce the order and try again 1474 */ 1475 do { 1476 sctp_port_hashtable = (struct sctp_bind_hashbucket *) 1477 __get_free_pages(GFP_KERNEL | __GFP_NOWARN, order); 1478 } while (!sctp_port_hashtable && --order > 0); 1479 1480 if (!sctp_port_hashtable) { 1481 pr_err("Failed bind hash alloc\n"); 1482 status = -ENOMEM; 1483 goto err_bhash_alloc; 1484 } 1485 1486 /* Now compute the number of entries that will fit in the 1487 * port hash space we allocated 1488 */ 1489 num_entries = (1UL << order) * PAGE_SIZE / 1490 sizeof(struct sctp_bind_hashbucket); 1491 1492 /* And finish by rounding it down to the nearest power of two. 1493 * This wastes some memory of course, but it's needed because 1494 * the hash function operates based on the assumption that 1495 * the number of entries is a power of two. 1496 */ 1497 sctp_port_hashsize = rounddown_pow_of_two(num_entries); 1498 1499 for (i = 0; i < sctp_port_hashsize; i++) { 1500 spin_lock_init(&sctp_port_hashtable[i].lock); 1501 INIT_HLIST_HEAD(&sctp_port_hashtable[i].chain); 1502 } 1503 1504 status = sctp_transport_hashtable_init(); 1505 if (status) 1506 goto err_thash_alloc; 1507 1508 pr_info("Hash tables configured (bind %d/%d)\n", sctp_port_hashsize, 1509 num_entries); 1510 1511 sctp_sysctl_register(); 1512 1513 INIT_LIST_HEAD(&sctp_address_families); 1514 sctp_v4_pf_init(); 1515 sctp_v6_pf_init(); 1516 sctp_sched_ops_init(); 1517 1518 status = register_pernet_subsys(&sctp_defaults_ops); 1519 if (status) 1520 goto err_register_defaults; 1521 1522 status = sctp_v4_protosw_init(); 1523 if (status) 1524 goto err_protosw_init; 1525 1526 status = sctp_v6_protosw_init(); 1527 if (status) 1528 goto err_v6_protosw_init; 1529 1530 status = register_pernet_subsys(&sctp_ctrlsock_ops); 1531 if (status) 1532 goto err_register_ctrlsock; 1533 1534 status = sctp_v4_add_protocol(); 1535 if (status) 1536 goto err_add_protocol; 1537 1538 /* Register SCTP with inet6 layer. */ 1539 status = sctp_v6_add_protocol(); 1540 if (status) 1541 goto err_v6_add_protocol; 1542 1543 if (sctp_offload_init() < 0) 1544 pr_crit("%s: Cannot add SCTP protocol offload\n", __func__); 1545 1546out: 1547 return status; 1548err_v6_add_protocol: 1549 sctp_v4_del_protocol(); 1550err_add_protocol: 1551 unregister_pernet_subsys(&sctp_ctrlsock_ops); 1552err_register_ctrlsock: 1553 sctp_v6_protosw_exit(); 1554err_v6_protosw_init: 1555 sctp_v4_protosw_exit(); 1556err_protosw_init: 1557 unregister_pernet_subsys(&sctp_defaults_ops); 1558err_register_defaults: 1559 sctp_v4_pf_exit(); 1560 sctp_v6_pf_exit(); 1561 sctp_sysctl_unregister(); 1562 free_pages((unsigned long)sctp_port_hashtable, 1563 get_order(sctp_port_hashsize * 1564 sizeof(struct sctp_bind_hashbucket))); 1565err_bhash_alloc: 1566 sctp_transport_hashtable_destroy(); 1567err_thash_alloc: 1568 kfree(sctp_ep_hashtable); 1569err_ehash_alloc: 1570 percpu_counter_destroy(&sctp_sockets_allocated); 1571err_percpu_counter_init: 1572 kmem_cache_destroy(sctp_chunk_cachep); 1573err_chunk_cachep: 1574 kmem_cache_destroy(sctp_bucket_cachep); 1575 goto out; 1576} 1577 1578/* Exit handler for the SCTP protocol. */ 1579static __exit void sctp_exit(void) 1580{ 1581 /* BUG. This should probably do something useful like clean 1582 * up all the remaining associations and all that memory. 1583 */ 1584 1585 /* Unregister with inet6/inet layers. */ 1586 sctp_v6_del_protocol(); 1587 sctp_v4_del_protocol(); 1588 1589 unregister_pernet_subsys(&sctp_ctrlsock_ops); 1590 1591 /* Free protosw registrations */ 1592 sctp_v6_protosw_exit(); 1593 sctp_v4_protosw_exit(); 1594 1595 unregister_pernet_subsys(&sctp_defaults_ops); 1596 1597 /* Unregister with socket layer. */ 1598 sctp_v6_pf_exit(); 1599 sctp_v4_pf_exit(); 1600 1601 sctp_sysctl_unregister(); 1602 1603 free_pages((unsigned long)sctp_port_hashtable, 1604 get_order(sctp_port_hashsize * 1605 sizeof(struct sctp_bind_hashbucket))); 1606 kfree(sctp_ep_hashtable); 1607 sctp_transport_hashtable_destroy(); 1608 1609 percpu_counter_destroy(&sctp_sockets_allocated); 1610 1611 rcu_barrier(); /* Wait for completion of call_rcu()'s */ 1612 1613 kmem_cache_destroy(sctp_chunk_cachep); 1614 kmem_cache_destroy(sctp_bucket_cachep); 1615} 1616 1617module_init(sctp_init); 1618module_exit(sctp_exit); 1619 1620/* 1621 * __stringify doesn't likes enums, so use IPPROTO_SCTP value (132) directly. 1622 */ 1623MODULE_ALIAS("net-pf-" __stringify(PF_INET) "-proto-132"); 1624MODULE_ALIAS("net-pf-" __stringify(PF_INET6) "-proto-132"); 1625MODULE_AUTHOR("Linux Kernel SCTP developers <linux-sctp@vger.kernel.org>"); 1626MODULE_DESCRIPTION("Support for the SCTP protocol (RFC2960)"); 1627module_param_named(no_checksums, sctp_checksum_disable, bool, 0644); 1628MODULE_PARM_DESC(no_checksums, "Disable checksums computing and verification"); 1629MODULE_LICENSE("GPL"); 1630