1 /* 2 * net/tipc/socket.c: TIPC socket API 3 * 4 * Copyright (c) 2001-2007, 2012-2017, Ericsson AB 5 * Copyright (c) 2004-2008, 2010-2013, Wind River Systems 6 * All rights reserved. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions are met: 10 * 11 * 1. Redistributions of source code must retain the above copyright 12 * notice, this list of conditions and the following disclaimer. 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in the 15 * documentation and/or other materials provided with the distribution. 16 * 3. Neither the names of the copyright holders nor the names of its 17 * contributors may be used to endorse or promote products derived from 18 * this software without specific prior written permission. 19 * 20 * Alternatively, this software may be distributed under the terms of the 21 * GNU General Public License ("GPL") version 2 as published by the Free 22 * Software Foundation. 23 * 24 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" 25 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 27 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE 28 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 29 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 30 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 31 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 32 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 33 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 34 * POSSIBILITY OF SUCH DAMAGE. 35 */ 36 37 #include <linux/rhashtable.h> 38 #include <linux/sched/signal.h> 39 40 #include "core.h" 41 #include "name_table.h" 42 #include "node.h" 43 #include "link.h" 44 #include "name_distr.h" 45 #include "socket.h" 46 #include "bcast.h" 47 #include "netlink.h" 48 #include "group.h" 49 50 #define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */ 51 #define CONN_PROBING_INTV msecs_to_jiffies(3600000) /* [ms] => 1 h */ 52 #define TIPC_FWD_MSG 1 53 #define TIPC_MAX_PORT 0xffffffff 54 #define TIPC_MIN_PORT 1 55 #define TIPC_ACK_RATE 4 /* ACK at 1/4 of of rcv window size */ 56 57 enum { 58 TIPC_LISTEN = TCP_LISTEN, 59 TIPC_ESTABLISHED = TCP_ESTABLISHED, 60 TIPC_OPEN = TCP_CLOSE, 61 TIPC_DISCONNECTING = TCP_CLOSE_WAIT, 62 TIPC_CONNECTING = TCP_SYN_SENT, 63 }; 64 65 struct sockaddr_pair { 66 struct sockaddr_tipc sock; 67 struct sockaddr_tipc member; 68 }; 69 70 /** 71 * struct tipc_sock - TIPC socket structure 72 * @sk: socket - interacts with 'port' and with user via the socket API 73 * @conn_type: TIPC type used when connection was established 74 * @conn_instance: TIPC instance used when connection was established 75 * @published: non-zero if port has one or more associated names 76 * @max_pkt: maximum packet size "hint" used when building messages sent by port 77 * @portid: unique port identity in TIPC socket hash table 78 * @phdr: preformatted message header used when sending messages 79 * #cong_links: list of congested links 80 * @publications: list of publications for port 81 * @blocking_link: address of the congested link we are currently sleeping on 82 * @pub_count: total # of publications port has made during its lifetime 83 * @probing_state: 84 * @conn_timeout: the time we can wait for an unresponded setup request 85 * @dupl_rcvcnt: number of bytes counted twice, in both backlog and rcv queue 86 * @cong_link_cnt: number of congested links 87 * @snt_unacked: # messages sent by socket, and not yet acked by peer 88 * @rcv_unacked: # messages read by user, but not yet acked back to peer 89 * @peer: 'connected' peer for dgram/rdm 90 * @node: hash table node 91 * @mc_method: cookie for use between socket and broadcast layer 92 * @rcu: rcu struct for tipc_sock 93 */ 94 struct tipc_sock { 95 struct sock sk; 96 u32 conn_type; 97 u32 conn_instance; 98 int published; 99 u32 max_pkt; 100 u32 portid; 101 struct tipc_msg phdr; 102 struct list_head cong_links; 103 struct list_head publications; 104 u32 pub_count; 105 uint conn_timeout; 106 atomic_t dupl_rcvcnt; 107 bool probe_unacked; 108 u16 cong_link_cnt; 109 u16 snt_unacked; 110 u16 snd_win; 111 u16 peer_caps; 112 u16 rcv_unacked; 113 u16 rcv_win; 114 struct sockaddr_tipc peer; 115 struct rhash_head node; 116 struct tipc_mc_method mc_method; 117 struct rcu_head rcu; 118 struct tipc_group *group; 119 }; 120 121 static int tipc_sk_backlog_rcv(struct sock *sk, struct sk_buff *skb); 122 static void tipc_data_ready(struct sock *sk); 123 static void tipc_write_space(struct sock *sk); 124 static void tipc_sock_destruct(struct sock *sk); 125 static int tipc_release(struct socket *sock); 126 static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, 127 bool kern); 128 static void tipc_sk_timeout(struct timer_list *t); 129 static int tipc_sk_publish(struct tipc_sock *tsk, uint scope, 130 struct tipc_name_seq const *seq); 131 static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope, 132 struct tipc_name_seq const *seq); 133 static int tipc_sk_leave(struct tipc_sock *tsk); 134 static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid); 135 static int tipc_sk_insert(struct tipc_sock *tsk); 136 static void tipc_sk_remove(struct tipc_sock *tsk); 137 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz); 138 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz); 139 140 static const struct proto_ops packet_ops; 141 static const struct proto_ops stream_ops; 142 static const struct proto_ops msg_ops; 143 static struct proto tipc_proto; 144 static const struct rhashtable_params tsk_rht_params; 145 146 static u32 tsk_own_node(struct tipc_sock *tsk) 147 { 148 return msg_prevnode(&tsk->phdr); 149 } 150 151 static u32 tsk_peer_node(struct tipc_sock *tsk) 152 { 153 return msg_destnode(&tsk->phdr); 154 } 155 156 static u32 tsk_peer_port(struct tipc_sock *tsk) 157 { 158 return msg_destport(&tsk->phdr); 159 } 160 161 static bool tsk_unreliable(struct tipc_sock *tsk) 162 { 163 return msg_src_droppable(&tsk->phdr) != 0; 164 } 165 166 static void tsk_set_unreliable(struct tipc_sock *tsk, bool unreliable) 167 { 168 msg_set_src_droppable(&tsk->phdr, unreliable ? 1 : 0); 169 } 170 171 static bool tsk_unreturnable(struct tipc_sock *tsk) 172 { 173 return msg_dest_droppable(&tsk->phdr) != 0; 174 } 175 176 static void tsk_set_unreturnable(struct tipc_sock *tsk, bool unreturnable) 177 { 178 msg_set_dest_droppable(&tsk->phdr, unreturnable ? 1 : 0); 179 } 180 181 static int tsk_importance(struct tipc_sock *tsk) 182 { 183 return msg_importance(&tsk->phdr); 184 } 185 186 static int tsk_set_importance(struct tipc_sock *tsk, int imp) 187 { 188 if (imp > TIPC_CRITICAL_IMPORTANCE) 189 return -EINVAL; 190 msg_set_importance(&tsk->phdr, (u32)imp); 191 return 0; 192 } 193 194 static struct tipc_sock *tipc_sk(const struct sock *sk) 195 { 196 return container_of(sk, struct tipc_sock, sk); 197 } 198 199 static bool tsk_conn_cong(struct tipc_sock *tsk) 200 { 201 return tsk->snt_unacked > tsk->snd_win; 202 } 203 204 static u16 tsk_blocks(int len) 205 { 206 return ((len / FLOWCTL_BLK_SZ) + 1); 207 } 208 209 /* tsk_blocks(): translate a buffer size in bytes to number of 210 * advertisable blocks, taking into account the ratio truesize(len)/len 211 * We can trust that this ratio is always < 4 for len >= FLOWCTL_BLK_SZ 212 */ 213 static u16 tsk_adv_blocks(int len) 214 { 215 return len / FLOWCTL_BLK_SZ / 4; 216 } 217 218 /* tsk_inc(): increment counter for sent or received data 219 * - If block based flow control is not supported by peer we 220 * fall back to message based ditto, incrementing the counter 221 */ 222 static u16 tsk_inc(struct tipc_sock *tsk, int msglen) 223 { 224 if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL)) 225 return ((msglen / FLOWCTL_BLK_SZ) + 1); 226 return 1; 227 } 228 229 /** 230 * tsk_advance_rx_queue - discard first buffer in socket receive queue 231 * 232 * Caller must hold socket lock 233 */ 234 static void tsk_advance_rx_queue(struct sock *sk) 235 { 236 kfree_skb(__skb_dequeue(&sk->sk_receive_queue)); 237 } 238 239 /* tipc_sk_respond() : send response message back to sender 240 */ 241 static void tipc_sk_respond(struct sock *sk, struct sk_buff *skb, int err) 242 { 243 u32 selector; 244 u32 dnode; 245 u32 onode = tipc_own_addr(sock_net(sk)); 246 247 if (!tipc_msg_reverse(onode, &skb, err)) 248 return; 249 250 dnode = msg_destnode(buf_msg(skb)); 251 selector = msg_origport(buf_msg(skb)); 252 tipc_node_xmit_skb(sock_net(sk), skb, dnode, selector); 253 } 254 255 /** 256 * tsk_rej_rx_queue - reject all buffers in socket receive queue 257 * 258 * Caller must hold socket lock 259 */ 260 static void tsk_rej_rx_queue(struct sock *sk) 261 { 262 struct sk_buff *skb; 263 264 while ((skb = __skb_dequeue(&sk->sk_receive_queue))) 265 tipc_sk_respond(sk, skb, TIPC_ERR_NO_PORT); 266 } 267 268 static bool tipc_sk_connected(struct sock *sk) 269 { 270 return sk->sk_state == TIPC_ESTABLISHED; 271 } 272 273 /* tipc_sk_type_connectionless - check if the socket is datagram socket 274 * @sk: socket 275 * 276 * Returns true if connection less, false otherwise 277 */ 278 static bool tipc_sk_type_connectionless(struct sock *sk) 279 { 280 return sk->sk_type == SOCK_RDM || sk->sk_type == SOCK_DGRAM; 281 } 282 283 /* tsk_peer_msg - verify if message was sent by connected port's peer 284 * 285 * Handles cases where the node's network address has changed from 286 * the default of <0.0.0> to its configured setting. 287 */ 288 static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg) 289 { 290 struct sock *sk = &tsk->sk; 291 struct tipc_net *tn = net_generic(sock_net(sk), tipc_net_id); 292 u32 peer_port = tsk_peer_port(tsk); 293 u32 orig_node; 294 u32 peer_node; 295 296 if (unlikely(!tipc_sk_connected(sk))) 297 return false; 298 299 if (unlikely(msg_origport(msg) != peer_port)) 300 return false; 301 302 orig_node = msg_orignode(msg); 303 peer_node = tsk_peer_node(tsk); 304 305 if (likely(orig_node == peer_node)) 306 return true; 307 308 if (!orig_node && (peer_node == tn->own_addr)) 309 return true; 310 311 if (!peer_node && (orig_node == tn->own_addr)) 312 return true; 313 314 return false; 315 } 316 317 /* tipc_set_sk_state - set the sk_state of the socket 318 * @sk: socket 319 * 320 * Caller must hold socket lock 321 * 322 * Returns 0 on success, errno otherwise 323 */ 324 static int tipc_set_sk_state(struct sock *sk, int state) 325 { 326 int oldsk_state = sk->sk_state; 327 int res = -EINVAL; 328 329 switch (state) { 330 case TIPC_OPEN: 331 res = 0; 332 break; 333 case TIPC_LISTEN: 334 case TIPC_CONNECTING: 335 if (oldsk_state == TIPC_OPEN) 336 res = 0; 337 break; 338 case TIPC_ESTABLISHED: 339 if (oldsk_state == TIPC_CONNECTING || 340 oldsk_state == TIPC_OPEN) 341 res = 0; 342 break; 343 case TIPC_DISCONNECTING: 344 if (oldsk_state == TIPC_CONNECTING || 345 oldsk_state == TIPC_ESTABLISHED) 346 res = 0; 347 break; 348 } 349 350 if (!res) 351 sk->sk_state = state; 352 353 return res; 354 } 355 356 static int tipc_sk_sock_err(struct socket *sock, long *timeout) 357 { 358 struct sock *sk = sock->sk; 359 int err = sock_error(sk); 360 int typ = sock->type; 361 362 if (err) 363 return err; 364 if (typ == SOCK_STREAM || typ == SOCK_SEQPACKET) { 365 if (sk->sk_state == TIPC_DISCONNECTING) 366 return -EPIPE; 367 else if (!tipc_sk_connected(sk)) 368 return -ENOTCONN; 369 } 370 if (!*timeout) 371 return -EAGAIN; 372 if (signal_pending(current)) 373 return sock_intr_errno(*timeout); 374 375 return 0; 376 } 377 378 #define tipc_wait_for_cond(sock_, timeo_, condition_) \ 379 ({ \ 380 struct sock *sk_; \ 381 int rc_; \ 382 \ 383 while ((rc_ = !(condition_))) { \ 384 DEFINE_WAIT_FUNC(wait_, woken_wake_function); \ 385 sk_ = (sock_)->sk; \ 386 rc_ = tipc_sk_sock_err((sock_), timeo_); \ 387 if (rc_) \ 388 break; \ 389 prepare_to_wait(sk_sleep(sk_), &wait_, TASK_INTERRUPTIBLE); \ 390 release_sock(sk_); \ 391 *(timeo_) = wait_woken(&wait_, TASK_INTERRUPTIBLE, *(timeo_)); \ 392 sched_annotate_sleep(); \ 393 lock_sock(sk_); \ 394 remove_wait_queue(sk_sleep(sk_), &wait_); \ 395 } \ 396 rc_; \ 397 }) 398 399 /** 400 * tipc_sk_create - create a TIPC socket 401 * @net: network namespace (must be default network) 402 * @sock: pre-allocated socket structure 403 * @protocol: protocol indicator (must be 0) 404 * @kern: caused by kernel or by userspace? 405 * 406 * This routine creates additional data structures used by the TIPC socket, 407 * initializes them, and links them together. 408 * 409 * Returns 0 on success, errno otherwise 410 */ 411 static int tipc_sk_create(struct net *net, struct socket *sock, 412 int protocol, int kern) 413 { 414 struct tipc_net *tn; 415 const struct proto_ops *ops; 416 struct sock *sk; 417 struct tipc_sock *tsk; 418 struct tipc_msg *msg; 419 420 /* Validate arguments */ 421 if (unlikely(protocol != 0)) 422 return -EPROTONOSUPPORT; 423 424 switch (sock->type) { 425 case SOCK_STREAM: 426 ops = &stream_ops; 427 break; 428 case SOCK_SEQPACKET: 429 ops = &packet_ops; 430 break; 431 case SOCK_DGRAM: 432 case SOCK_RDM: 433 ops = &msg_ops; 434 break; 435 default: 436 return -EPROTOTYPE; 437 } 438 439 /* Allocate socket's protocol area */ 440 sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto, kern); 441 if (sk == NULL) 442 return -ENOMEM; 443 444 tsk = tipc_sk(sk); 445 tsk->max_pkt = MAX_PKT_DEFAULT; 446 INIT_LIST_HEAD(&tsk->publications); 447 INIT_LIST_HEAD(&tsk->cong_links); 448 msg = &tsk->phdr; 449 tn = net_generic(sock_net(sk), tipc_net_id); 450 451 /* Finish initializing socket data structures */ 452 sock->ops = ops; 453 sock_init_data(sock, sk); 454 tipc_set_sk_state(sk, TIPC_OPEN); 455 if (tipc_sk_insert(tsk)) { 456 pr_warn("Socket create failed; port number exhausted\n"); 457 return -EINVAL; 458 } 459 460 /* Ensure tsk is visible before we read own_addr. */ 461 smp_mb(); 462 463 tipc_msg_init(tn->own_addr, msg, TIPC_LOW_IMPORTANCE, TIPC_NAMED_MSG, 464 NAMED_H_SIZE, 0); 465 466 msg_set_origport(msg, tsk->portid); 467 timer_setup(&sk->sk_timer, tipc_sk_timeout, 0); 468 sk->sk_shutdown = 0; 469 sk->sk_backlog_rcv = tipc_sk_backlog_rcv; 470 sk->sk_rcvbuf = sysctl_tipc_rmem[1]; 471 sk->sk_data_ready = tipc_data_ready; 472 sk->sk_write_space = tipc_write_space; 473 sk->sk_destruct = tipc_sock_destruct; 474 tsk->conn_timeout = CONN_TIMEOUT_DEFAULT; 475 atomic_set(&tsk->dupl_rcvcnt, 0); 476 477 /* Start out with safe limits until we receive an advertised window */ 478 tsk->snd_win = tsk_adv_blocks(RCVBUF_MIN); 479 tsk->rcv_win = tsk->snd_win; 480 481 if (tipc_sk_type_connectionless(sk)) { 482 tsk_set_unreturnable(tsk, true); 483 if (sock->type == SOCK_DGRAM) 484 tsk_set_unreliable(tsk, true); 485 } 486 487 return 0; 488 } 489 490 static void tipc_sk_callback(struct rcu_head *head) 491 { 492 struct tipc_sock *tsk = container_of(head, struct tipc_sock, rcu); 493 494 sock_put(&tsk->sk); 495 } 496 497 /* Caller should hold socket lock for the socket. */ 498 static void __tipc_shutdown(struct socket *sock, int error) 499 { 500 struct sock *sk = sock->sk; 501 struct tipc_sock *tsk = tipc_sk(sk); 502 struct net *net = sock_net(sk); 503 long timeout = CONN_TIMEOUT_DEFAULT; 504 u32 dnode = tsk_peer_node(tsk); 505 struct sk_buff *skb; 506 507 /* Avoid that hi-prio shutdown msgs bypass msgs in link wakeup queue */ 508 tipc_wait_for_cond(sock, &timeout, (!tsk->cong_link_cnt && 509 !tsk_conn_cong(tsk))); 510 511 /* Reject all unreceived messages, except on an active connection 512 * (which disconnects locally & sends a 'FIN+' to peer). 513 */ 514 while ((skb = __skb_dequeue(&sk->sk_receive_queue)) != NULL) { 515 if (TIPC_SKB_CB(skb)->bytes_read) { 516 kfree_skb(skb); 517 continue; 518 } 519 if (!tipc_sk_type_connectionless(sk) && 520 sk->sk_state != TIPC_DISCONNECTING) { 521 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 522 tipc_node_remove_conn(net, dnode, tsk->portid); 523 } 524 tipc_sk_respond(sk, skb, error); 525 } 526 527 if (tipc_sk_type_connectionless(sk)) 528 return; 529 530 if (sk->sk_state != TIPC_DISCONNECTING) { 531 skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, 532 TIPC_CONN_MSG, SHORT_H_SIZE, 0, dnode, 533 tsk_own_node(tsk), tsk_peer_port(tsk), 534 tsk->portid, error); 535 if (skb) 536 tipc_node_xmit_skb(net, skb, dnode, tsk->portid); 537 tipc_node_remove_conn(net, dnode, tsk->portid); 538 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 539 } 540 } 541 542 /** 543 * tipc_release - destroy a TIPC socket 544 * @sock: socket to destroy 545 * 546 * This routine cleans up any messages that are still queued on the socket. 547 * For DGRAM and RDM socket types, all queued messages are rejected. 548 * For SEQPACKET and STREAM socket types, the first message is rejected 549 * and any others are discarded. (If the first message on a STREAM socket 550 * is partially-read, it is discarded and the next one is rejected instead.) 551 * 552 * NOTE: Rejected messages are not necessarily returned to the sender! They 553 * are returned or discarded according to the "destination droppable" setting 554 * specified for the message by the sender. 555 * 556 * Returns 0 on success, errno otherwise 557 */ 558 static int tipc_release(struct socket *sock) 559 { 560 struct sock *sk = sock->sk; 561 struct tipc_sock *tsk; 562 563 /* 564 * Exit if socket isn't fully initialized (occurs when a failed accept() 565 * releases a pre-allocated child socket that was never used) 566 */ 567 if (sk == NULL) 568 return 0; 569 570 tsk = tipc_sk(sk); 571 lock_sock(sk); 572 573 __tipc_shutdown(sock, TIPC_ERR_NO_PORT); 574 sk->sk_shutdown = SHUTDOWN_MASK; 575 tipc_sk_leave(tsk); 576 tipc_sk_withdraw(tsk, 0, NULL); 577 sk_stop_timer(sk, &sk->sk_timer); 578 tipc_sk_remove(tsk); 579 580 /* Reject any messages that accumulated in backlog queue */ 581 release_sock(sk); 582 tipc_dest_list_purge(&tsk->cong_links); 583 tsk->cong_link_cnt = 0; 584 call_rcu(&tsk->rcu, tipc_sk_callback); 585 sock->sk = NULL; 586 587 return 0; 588 } 589 590 /** 591 * tipc_bind - associate or disassocate TIPC name(s) with a socket 592 * @sock: socket structure 593 * @uaddr: socket address describing name(s) and desired operation 594 * @uaddr_len: size of socket address data structure 595 * 596 * Name and name sequence binding is indicated using a positive scope value; 597 * a negative scope value unbinds the specified name. Specifying no name 598 * (i.e. a socket address length of 0) unbinds all names from the socket. 599 * 600 * Returns 0 on success, errno otherwise 601 * 602 * NOTE: This routine doesn't need to take the socket lock since it doesn't 603 * access any non-constant socket information. 604 */ 605 static int tipc_bind(struct socket *sock, struct sockaddr *uaddr, 606 int uaddr_len) 607 { 608 struct sock *sk = sock->sk; 609 struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; 610 struct tipc_sock *tsk = tipc_sk(sk); 611 int res = -EINVAL; 612 613 lock_sock(sk); 614 if (unlikely(!uaddr_len)) { 615 res = tipc_sk_withdraw(tsk, 0, NULL); 616 goto exit; 617 } 618 if (tsk->group) { 619 res = -EACCES; 620 goto exit; 621 } 622 if (uaddr_len < sizeof(struct sockaddr_tipc)) { 623 res = -EINVAL; 624 goto exit; 625 } 626 if (addr->family != AF_TIPC) { 627 res = -EAFNOSUPPORT; 628 goto exit; 629 } 630 631 if (addr->addrtype == TIPC_ADDR_NAME) 632 addr->addr.nameseq.upper = addr->addr.nameseq.lower; 633 else if (addr->addrtype != TIPC_ADDR_NAMESEQ) { 634 res = -EAFNOSUPPORT; 635 goto exit; 636 } 637 638 if ((addr->addr.nameseq.type < TIPC_RESERVED_TYPES) && 639 (addr->addr.nameseq.type != TIPC_TOP_SRV) && 640 (addr->addr.nameseq.type != TIPC_CFG_SRV)) { 641 res = -EACCES; 642 goto exit; 643 } 644 645 res = (addr->scope > 0) ? 646 tipc_sk_publish(tsk, addr->scope, &addr->addr.nameseq) : 647 tipc_sk_withdraw(tsk, -addr->scope, &addr->addr.nameseq); 648 exit: 649 release_sock(sk); 650 return res; 651 } 652 653 /** 654 * tipc_getname - get port ID of socket or peer socket 655 * @sock: socket structure 656 * @uaddr: area for returned socket address 657 * @uaddr_len: area for returned length of socket address 658 * @peer: 0 = own ID, 1 = current peer ID, 2 = current/former peer ID 659 * 660 * Returns 0 on success, errno otherwise 661 * 662 * NOTE: This routine doesn't need to take the socket lock since it only 663 * accesses socket information that is unchanging (or which changes in 664 * a completely predictable manner). 665 */ 666 static int tipc_getname(struct socket *sock, struct sockaddr *uaddr, 667 int *uaddr_len, int peer) 668 { 669 struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; 670 struct sock *sk = sock->sk; 671 struct tipc_sock *tsk = tipc_sk(sk); 672 struct tipc_net *tn = net_generic(sock_net(sock->sk), tipc_net_id); 673 674 memset(addr, 0, sizeof(*addr)); 675 if (peer) { 676 if ((!tipc_sk_connected(sk)) && 677 ((peer != 2) || (sk->sk_state != TIPC_DISCONNECTING))) 678 return -ENOTCONN; 679 addr->addr.id.ref = tsk_peer_port(tsk); 680 addr->addr.id.node = tsk_peer_node(tsk); 681 } else { 682 addr->addr.id.ref = tsk->portid; 683 addr->addr.id.node = tn->own_addr; 684 } 685 686 *uaddr_len = sizeof(*addr); 687 addr->addrtype = TIPC_ADDR_ID; 688 addr->family = AF_TIPC; 689 addr->scope = 0; 690 addr->addr.name.domain = 0; 691 692 return 0; 693 } 694 695 /** 696 * tipc_poll - read and possibly block on pollmask 697 * @file: file structure associated with the socket 698 * @sock: socket for which to calculate the poll bits 699 * @wait: ??? 700 * 701 * Returns pollmask value 702 * 703 * COMMENTARY: 704 * It appears that the usual socket locking mechanisms are not useful here 705 * since the pollmask info is potentially out-of-date the moment this routine 706 * exits. TCP and other protocols seem to rely on higher level poll routines 707 * to handle any preventable race conditions, so TIPC will do the same ... 708 * 709 * IMPORTANT: The fact that a read or write operation is indicated does NOT 710 * imply that the operation will succeed, merely that it should be performed 711 * and will not block. 712 */ 713 static unsigned int tipc_poll(struct file *file, struct socket *sock, 714 poll_table *wait) 715 { 716 struct sock *sk = sock->sk; 717 struct tipc_sock *tsk = tipc_sk(sk); 718 struct tipc_group *grp = tsk->group; 719 u32 revents = 0; 720 721 sock_poll_wait(file, sk_sleep(sk), wait); 722 723 if (sk->sk_shutdown & RCV_SHUTDOWN) 724 revents |= POLLRDHUP | POLLIN | POLLRDNORM; 725 if (sk->sk_shutdown == SHUTDOWN_MASK) 726 revents |= POLLHUP; 727 728 switch (sk->sk_state) { 729 case TIPC_ESTABLISHED: 730 if (!tsk->cong_link_cnt && !tsk_conn_cong(tsk)) 731 revents |= POLLOUT; 732 /* fall thru' */ 733 case TIPC_LISTEN: 734 case TIPC_CONNECTING: 735 if (!skb_queue_empty(&sk->sk_receive_queue)) 736 revents |= POLLIN | POLLRDNORM; 737 break; 738 case TIPC_OPEN: 739 if (!grp || tipc_group_size(grp)) 740 if (!tsk->cong_link_cnt) 741 revents |= POLLOUT; 742 if (!tipc_sk_type_connectionless(sk)) 743 break; 744 if (skb_queue_empty(&sk->sk_receive_queue)) 745 break; 746 revents |= POLLIN | POLLRDNORM; 747 break; 748 case TIPC_DISCONNECTING: 749 revents = POLLIN | POLLRDNORM | POLLHUP; 750 break; 751 } 752 return revents; 753 } 754 755 /** 756 * tipc_sendmcast - send multicast message 757 * @sock: socket structure 758 * @seq: destination address 759 * @msg: message to send 760 * @dlen: length of data to send 761 * @timeout: timeout to wait for wakeup 762 * 763 * Called from function tipc_sendmsg(), which has done all sanity checks 764 * Returns the number of bytes sent on success, or errno 765 */ 766 static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq, 767 struct msghdr *msg, size_t dlen, long timeout) 768 { 769 struct sock *sk = sock->sk; 770 struct tipc_sock *tsk = tipc_sk(sk); 771 struct tipc_msg *hdr = &tsk->phdr; 772 struct net *net = sock_net(sk); 773 int mtu = tipc_bcast_get_mtu(net); 774 struct tipc_mc_method *method = &tsk->mc_method; 775 u32 domain = addr_domain(net, TIPC_CLUSTER_SCOPE); 776 struct sk_buff_head pkts; 777 struct tipc_nlist dsts; 778 int rc; 779 780 if (tsk->group) 781 return -EACCES; 782 783 /* Block or return if any destination link is congested */ 784 rc = tipc_wait_for_cond(sock, &timeout, !tsk->cong_link_cnt); 785 if (unlikely(rc)) 786 return rc; 787 788 /* Lookup destination nodes */ 789 tipc_nlist_init(&dsts, tipc_own_addr(net)); 790 tipc_nametbl_lookup_dst_nodes(net, seq->type, seq->lower, 791 seq->upper, domain, &dsts); 792 if (!dsts.local && !dsts.remote) 793 return -EHOSTUNREACH; 794 795 /* Build message header */ 796 msg_set_type(hdr, TIPC_MCAST_MSG); 797 msg_set_hdr_sz(hdr, MCAST_H_SIZE); 798 msg_set_lookup_scope(hdr, TIPC_CLUSTER_SCOPE); 799 msg_set_destport(hdr, 0); 800 msg_set_destnode(hdr, 0); 801 msg_set_nametype(hdr, seq->type); 802 msg_set_namelower(hdr, seq->lower); 803 msg_set_nameupper(hdr, seq->upper); 804 805 /* Build message as chain of buffers */ 806 skb_queue_head_init(&pkts); 807 rc = tipc_msg_build(hdr, msg, 0, dlen, mtu, &pkts); 808 809 /* Send message if build was successful */ 810 if (unlikely(rc == dlen)) 811 rc = tipc_mcast_xmit(net, &pkts, method, &dsts, 812 &tsk->cong_link_cnt); 813 814 tipc_nlist_purge(&dsts); 815 816 return rc ? rc : dlen; 817 } 818 819 /** 820 * tipc_send_group_msg - send a message to a member in the group 821 * @net: network namespace 822 * @m: message to send 823 * @mb: group member 824 * @dnode: destination node 825 * @dport: destination port 826 * @dlen: total length of message data 827 */ 828 static int tipc_send_group_msg(struct net *net, struct tipc_sock *tsk, 829 struct msghdr *m, struct tipc_member *mb, 830 u32 dnode, u32 dport, int dlen) 831 { 832 u16 bc_snd_nxt = tipc_group_bc_snd_nxt(tsk->group); 833 struct tipc_mc_method *method = &tsk->mc_method; 834 int blks = tsk_blocks(GROUP_H_SIZE + dlen); 835 struct tipc_msg *hdr = &tsk->phdr; 836 struct sk_buff_head pkts; 837 int mtu, rc; 838 839 /* Complete message header */ 840 msg_set_type(hdr, TIPC_GRP_UCAST_MSG); 841 msg_set_hdr_sz(hdr, GROUP_H_SIZE); 842 msg_set_destport(hdr, dport); 843 msg_set_destnode(hdr, dnode); 844 msg_set_grp_bc_seqno(hdr, bc_snd_nxt); 845 846 /* Build message as chain of buffers */ 847 skb_queue_head_init(&pkts); 848 mtu = tipc_node_get_mtu(net, dnode, tsk->portid); 849 rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); 850 if (unlikely(rc != dlen)) 851 return rc; 852 853 /* Send message */ 854 rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid); 855 if (unlikely(rc == -ELINKCONG)) { 856 tipc_dest_push(&tsk->cong_links, dnode, 0); 857 tsk->cong_link_cnt++; 858 } 859 860 /* Update send window */ 861 tipc_group_update_member(mb, blks); 862 863 /* A broadcast sent within next EXPIRE period must follow same path */ 864 method->rcast = true; 865 method->mandatory = true; 866 return dlen; 867 } 868 869 /** 870 * tipc_send_group_unicast - send message to a member in the group 871 * @sock: socket structure 872 * @m: message to send 873 * @dlen: total length of message data 874 * @timeout: timeout to wait for wakeup 875 * 876 * Called from function tipc_sendmsg(), which has done all sanity checks 877 * Returns the number of bytes sent on success, or errno 878 */ 879 static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m, 880 int dlen, long timeout) 881 { 882 struct sock *sk = sock->sk; 883 DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); 884 int blks = tsk_blocks(GROUP_H_SIZE + dlen); 885 struct tipc_sock *tsk = tipc_sk(sk); 886 struct tipc_group *grp = tsk->group; 887 struct net *net = sock_net(sk); 888 struct tipc_member *mb = NULL; 889 u32 node, port; 890 int rc; 891 892 node = dest->addr.id.node; 893 port = dest->addr.id.ref; 894 if (!port && !node) 895 return -EHOSTUNREACH; 896 897 /* Block or return if destination link or member is congested */ 898 rc = tipc_wait_for_cond(sock, &timeout, 899 !tipc_dest_find(&tsk->cong_links, node, 0) && 900 !tipc_group_cong(grp, node, port, blks, &mb)); 901 if (unlikely(rc)) 902 return rc; 903 904 if (unlikely(!mb)) 905 return -EHOSTUNREACH; 906 907 rc = tipc_send_group_msg(net, tsk, m, mb, node, port, dlen); 908 909 return rc ? rc : dlen; 910 } 911 912 /** 913 * tipc_send_group_anycast - send message to any member with given identity 914 * @sock: socket structure 915 * @m: message to send 916 * @dlen: total length of message data 917 * @timeout: timeout to wait for wakeup 918 * 919 * Called from function tipc_sendmsg(), which has done all sanity checks 920 * Returns the number of bytes sent on success, or errno 921 */ 922 static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m, 923 int dlen, long timeout) 924 { 925 DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); 926 struct sock *sk = sock->sk; 927 struct tipc_sock *tsk = tipc_sk(sk); 928 struct list_head *cong_links = &tsk->cong_links; 929 int blks = tsk_blocks(GROUP_H_SIZE + dlen); 930 struct tipc_group *grp = tsk->group; 931 struct tipc_member *first = NULL; 932 struct tipc_member *mbr = NULL; 933 struct net *net = sock_net(sk); 934 u32 node, port, exclude; 935 u32 type, inst, domain; 936 struct list_head dsts; 937 int lookups = 0; 938 int dstcnt, rc; 939 bool cong; 940 941 INIT_LIST_HEAD(&dsts); 942 943 type = dest->addr.name.name.type; 944 inst = dest->addr.name.name.instance; 945 domain = addr_domain(net, dest->scope); 946 exclude = tipc_group_exclude(grp); 947 948 while (++lookups < 4) { 949 first = NULL; 950 951 /* Look for a non-congested destination member, if any */ 952 while (1) { 953 if (!tipc_nametbl_lookup(net, type, inst, domain, &dsts, 954 &dstcnt, exclude, false)) 955 return -EHOSTUNREACH; 956 tipc_dest_pop(&dsts, &node, &port); 957 cong = tipc_group_cong(grp, node, port, blks, &mbr); 958 if (!cong) 959 break; 960 if (mbr == first) 961 break; 962 if (!first) 963 first = mbr; 964 } 965 966 /* Start over if destination was not in member list */ 967 if (unlikely(!mbr)) 968 continue; 969 970 if (likely(!cong && !tipc_dest_find(cong_links, node, 0))) 971 break; 972 973 /* Block or return if destination link or member is congested */ 974 rc = tipc_wait_for_cond(sock, &timeout, 975 !tipc_dest_find(cong_links, node, 0) && 976 !tipc_group_cong(grp, node, port, 977 blks, &mbr)); 978 if (unlikely(rc)) 979 return rc; 980 981 /* Send, unless destination disappeared while waiting */ 982 if (likely(mbr)) 983 break; 984 } 985 986 if (unlikely(lookups >= 4)) 987 return -EHOSTUNREACH; 988 989 rc = tipc_send_group_msg(net, tsk, m, mbr, node, port, dlen); 990 991 return rc ? rc : dlen; 992 } 993 994 /** 995 * tipc_send_group_bcast - send message to all members in communication group 996 * @sk: socket structure 997 * @m: message to send 998 * @dlen: total length of message data 999 * @timeout: timeout to wait for wakeup 1000 * 1001 * Called from function tipc_sendmsg(), which has done all sanity checks 1002 * Returns the number of bytes sent on success, or errno 1003 */ 1004 static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m, 1005 int dlen, long timeout) 1006 { 1007 DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); 1008 struct sock *sk = sock->sk; 1009 struct net *net = sock_net(sk); 1010 struct tipc_sock *tsk = tipc_sk(sk); 1011 struct tipc_group *grp = tsk->group; 1012 struct tipc_nlist *dsts = tipc_group_dests(grp); 1013 struct tipc_mc_method *method = &tsk->mc_method; 1014 bool ack = method->mandatory && method->rcast; 1015 int blks = tsk_blocks(MCAST_H_SIZE + dlen); 1016 struct tipc_msg *hdr = &tsk->phdr; 1017 int mtu = tipc_bcast_get_mtu(net); 1018 struct sk_buff_head pkts; 1019 int rc = -EHOSTUNREACH; 1020 1021 if (!dsts->local && !dsts->remote) 1022 return -EHOSTUNREACH; 1023 1024 /* Block or return if any destination link or member is congested */ 1025 rc = tipc_wait_for_cond(sock, &timeout, !tsk->cong_link_cnt && 1026 !tipc_group_bc_cong(grp, blks)); 1027 if (unlikely(rc)) 1028 return rc; 1029 1030 /* Complete message header */ 1031 if (dest) { 1032 msg_set_type(hdr, TIPC_GRP_MCAST_MSG); 1033 msg_set_nameinst(hdr, dest->addr.name.name.instance); 1034 } else { 1035 msg_set_type(hdr, TIPC_GRP_BCAST_MSG); 1036 msg_set_nameinst(hdr, 0); 1037 } 1038 msg_set_hdr_sz(hdr, GROUP_H_SIZE); 1039 msg_set_destport(hdr, 0); 1040 msg_set_destnode(hdr, 0); 1041 msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(grp)); 1042 1043 /* Avoid getting stuck with repeated forced replicasts */ 1044 msg_set_grp_bc_ack_req(hdr, ack); 1045 1046 /* Build message as chain of buffers */ 1047 skb_queue_head_init(&pkts); 1048 rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); 1049 if (unlikely(rc != dlen)) 1050 return rc; 1051 1052 /* Send message */ 1053 rc = tipc_mcast_xmit(net, &pkts, method, dsts, &tsk->cong_link_cnt); 1054 if (unlikely(rc)) 1055 return rc; 1056 1057 /* Update broadcast sequence number and send windows */ 1058 tipc_group_update_bc_members(tsk->group, blks, ack); 1059 1060 /* Broadcast link is now free to choose method for next broadcast */ 1061 method->mandatory = false; 1062 method->expires = jiffies; 1063 1064 return dlen; 1065 } 1066 1067 /** 1068 * tipc_send_group_mcast - send message to all members with given identity 1069 * @sock: socket structure 1070 * @m: message to send 1071 * @dlen: total length of message data 1072 * @timeout: timeout to wait for wakeup 1073 * 1074 * Called from function tipc_sendmsg(), which has done all sanity checks 1075 * Returns the number of bytes sent on success, or errno 1076 */ 1077 static int tipc_send_group_mcast(struct socket *sock, struct msghdr *m, 1078 int dlen, long timeout) 1079 { 1080 struct sock *sk = sock->sk; 1081 DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); 1082 struct tipc_name_seq *seq = &dest->addr.nameseq; 1083 struct tipc_sock *tsk = tipc_sk(sk); 1084 struct tipc_group *grp = tsk->group; 1085 struct net *net = sock_net(sk); 1086 u32 domain, exclude, dstcnt; 1087 struct list_head dsts; 1088 1089 INIT_LIST_HEAD(&dsts); 1090 1091 if (seq->lower != seq->upper) 1092 return -ENOTSUPP; 1093 1094 domain = addr_domain(net, dest->scope); 1095 exclude = tipc_group_exclude(grp); 1096 if (!tipc_nametbl_lookup(net, seq->type, seq->lower, domain, 1097 &dsts, &dstcnt, exclude, true)) 1098 return -EHOSTUNREACH; 1099 1100 if (dstcnt == 1) { 1101 tipc_dest_pop(&dsts, &dest->addr.id.node, &dest->addr.id.ref); 1102 return tipc_send_group_unicast(sock, m, dlen, timeout); 1103 } 1104 1105 tipc_dest_list_purge(&dsts); 1106 return tipc_send_group_bcast(sock, m, dlen, timeout); 1107 } 1108 1109 /** 1110 * tipc_sk_mcast_rcv - Deliver multicast messages to all destination sockets 1111 * @arrvq: queue with arriving messages, to be cloned after destination lookup 1112 * @inputq: queue with cloned messages, delivered to socket after dest lookup 1113 * 1114 * Multi-threaded: parallel calls with reference to same queues may occur 1115 */ 1116 void tipc_sk_mcast_rcv(struct net *net, struct sk_buff_head *arrvq, 1117 struct sk_buff_head *inputq) 1118 { 1119 u32 scope = TIPC_CLUSTER_SCOPE; 1120 u32 self = tipc_own_addr(net); 1121 struct sk_buff *skb, *_skb; 1122 u32 lower = 0, upper = ~0; 1123 struct sk_buff_head tmpq; 1124 u32 portid, oport, onode; 1125 struct list_head dports; 1126 struct tipc_msg *msg; 1127 int user, mtyp, hsz; 1128 1129 __skb_queue_head_init(&tmpq); 1130 INIT_LIST_HEAD(&dports); 1131 1132 skb = tipc_skb_peek(arrvq, &inputq->lock); 1133 for (; skb; skb = tipc_skb_peek(arrvq, &inputq->lock)) { 1134 msg = buf_msg(skb); 1135 user = msg_user(msg); 1136 mtyp = msg_type(msg); 1137 if (mtyp == TIPC_GRP_UCAST_MSG || user == GROUP_PROTOCOL) { 1138 spin_lock_bh(&inputq->lock); 1139 if (skb_peek(arrvq) == skb) { 1140 __skb_dequeue(arrvq); 1141 __skb_queue_tail(inputq, skb); 1142 } 1143 refcount_dec(&skb->users); 1144 spin_unlock_bh(&inputq->lock); 1145 continue; 1146 } 1147 hsz = skb_headroom(skb) + msg_hdr_sz(msg); 1148 oport = msg_origport(msg); 1149 onode = msg_orignode(msg); 1150 if (onode == self) 1151 scope = TIPC_NODE_SCOPE; 1152 1153 /* Create destination port list and message clones: */ 1154 if (!msg_in_group(msg)) { 1155 lower = msg_namelower(msg); 1156 upper = msg_nameupper(msg); 1157 } 1158 tipc_nametbl_mc_translate(net, msg_nametype(msg), lower, upper, 1159 scope, &dports); 1160 while (tipc_dest_pop(&dports, NULL, &portid)) { 1161 _skb = __pskb_copy(skb, hsz, GFP_ATOMIC); 1162 if (_skb) { 1163 msg_set_destport(buf_msg(_skb), portid); 1164 __skb_queue_tail(&tmpq, _skb); 1165 continue; 1166 } 1167 pr_warn("Failed to clone mcast rcv buffer\n"); 1168 } 1169 /* Append to inputq if not already done by other thread */ 1170 spin_lock_bh(&inputq->lock); 1171 if (skb_peek(arrvq) == skb) { 1172 skb_queue_splice_tail_init(&tmpq, inputq); 1173 kfree_skb(__skb_dequeue(arrvq)); 1174 } 1175 spin_unlock_bh(&inputq->lock); 1176 __skb_queue_purge(&tmpq); 1177 kfree_skb(skb); 1178 } 1179 tipc_sk_rcv(net, inputq); 1180 } 1181 1182 /** 1183 * tipc_sk_conn_proto_rcv - receive a connection mng protocol message 1184 * @tsk: receiving socket 1185 * @skb: pointer to message buffer. 1186 */ 1187 static void tipc_sk_conn_proto_rcv(struct tipc_sock *tsk, struct sk_buff *skb, 1188 struct sk_buff_head *xmitq) 1189 { 1190 struct tipc_msg *hdr = buf_msg(skb); 1191 u32 onode = tsk_own_node(tsk); 1192 struct sock *sk = &tsk->sk; 1193 int mtyp = msg_type(hdr); 1194 bool conn_cong; 1195 1196 /* Ignore if connection cannot be validated: */ 1197 if (!tsk_peer_msg(tsk, hdr)) 1198 goto exit; 1199 1200 if (unlikely(msg_errcode(hdr))) { 1201 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 1202 tipc_node_remove_conn(sock_net(sk), tsk_peer_node(tsk), 1203 tsk_peer_port(tsk)); 1204 sk->sk_state_change(sk); 1205 goto exit; 1206 } 1207 1208 tsk->probe_unacked = false; 1209 1210 if (mtyp == CONN_PROBE) { 1211 msg_set_type(hdr, CONN_PROBE_REPLY); 1212 if (tipc_msg_reverse(onode, &skb, TIPC_OK)) 1213 __skb_queue_tail(xmitq, skb); 1214 return; 1215 } else if (mtyp == CONN_ACK) { 1216 conn_cong = tsk_conn_cong(tsk); 1217 tsk->snt_unacked -= msg_conn_ack(hdr); 1218 if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) 1219 tsk->snd_win = msg_adv_win(hdr); 1220 if (conn_cong) 1221 sk->sk_write_space(sk); 1222 } else if (mtyp != CONN_PROBE_REPLY) { 1223 pr_warn("Received unknown CONN_PROTO msg\n"); 1224 } 1225 exit: 1226 kfree_skb(skb); 1227 } 1228 1229 /** 1230 * tipc_sendmsg - send message in connectionless manner 1231 * @sock: socket structure 1232 * @m: message to send 1233 * @dsz: amount of user data to be sent 1234 * 1235 * Message must have an destination specified explicitly. 1236 * Used for SOCK_RDM and SOCK_DGRAM messages, 1237 * and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections. 1238 * (Note: 'SYN+' is prohibited on SOCK_STREAM.) 1239 * 1240 * Returns the number of bytes sent on success, or errno otherwise 1241 */ 1242 static int tipc_sendmsg(struct socket *sock, 1243 struct msghdr *m, size_t dsz) 1244 { 1245 struct sock *sk = sock->sk; 1246 int ret; 1247 1248 lock_sock(sk); 1249 ret = __tipc_sendmsg(sock, m, dsz); 1250 release_sock(sk); 1251 1252 return ret; 1253 } 1254 1255 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dlen) 1256 { 1257 struct sock *sk = sock->sk; 1258 struct net *net = sock_net(sk); 1259 struct tipc_sock *tsk = tipc_sk(sk); 1260 DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); 1261 long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); 1262 struct list_head *clinks = &tsk->cong_links; 1263 bool syn = !tipc_sk_type_connectionless(sk); 1264 struct tipc_group *grp = tsk->group; 1265 struct tipc_msg *hdr = &tsk->phdr; 1266 struct tipc_name_seq *seq; 1267 struct sk_buff_head pkts; 1268 u32 type, inst, domain; 1269 u32 dnode, dport; 1270 int mtu, rc; 1271 1272 if (unlikely(dlen > TIPC_MAX_USER_MSG_SIZE)) 1273 return -EMSGSIZE; 1274 1275 if (likely(dest)) { 1276 if (unlikely(m->msg_namelen < sizeof(*dest))) 1277 return -EINVAL; 1278 if (unlikely(dest->family != AF_TIPC)) 1279 return -EINVAL; 1280 } 1281 1282 if (grp) { 1283 if (!dest) 1284 return tipc_send_group_bcast(sock, m, dlen, timeout); 1285 if (dest->addrtype == TIPC_ADDR_NAME) 1286 return tipc_send_group_anycast(sock, m, dlen, timeout); 1287 if (dest->addrtype == TIPC_ADDR_ID) 1288 return tipc_send_group_unicast(sock, m, dlen, timeout); 1289 if (dest->addrtype == TIPC_ADDR_MCAST) 1290 return tipc_send_group_mcast(sock, m, dlen, timeout); 1291 return -EINVAL; 1292 } 1293 1294 if (unlikely(!dest)) { 1295 dest = &tsk->peer; 1296 if (!syn || dest->family != AF_TIPC) 1297 return -EDESTADDRREQ; 1298 } 1299 1300 if (unlikely(syn)) { 1301 if (sk->sk_state == TIPC_LISTEN) 1302 return -EPIPE; 1303 if (sk->sk_state != TIPC_OPEN) 1304 return -EISCONN; 1305 if (tsk->published) 1306 return -EOPNOTSUPP; 1307 if (dest->addrtype == TIPC_ADDR_NAME) { 1308 tsk->conn_type = dest->addr.name.name.type; 1309 tsk->conn_instance = dest->addr.name.name.instance; 1310 } 1311 } 1312 1313 seq = &dest->addr.nameseq; 1314 if (dest->addrtype == TIPC_ADDR_MCAST) 1315 return tipc_sendmcast(sock, seq, m, dlen, timeout); 1316 1317 if (dest->addrtype == TIPC_ADDR_NAME) { 1318 type = dest->addr.name.name.type; 1319 inst = dest->addr.name.name.instance; 1320 domain = dest->addr.name.domain; 1321 dnode = domain; 1322 msg_set_type(hdr, TIPC_NAMED_MSG); 1323 msg_set_hdr_sz(hdr, NAMED_H_SIZE); 1324 msg_set_nametype(hdr, type); 1325 msg_set_nameinst(hdr, inst); 1326 msg_set_lookup_scope(hdr, tipc_addr_scope(domain)); 1327 dport = tipc_nametbl_translate(net, type, inst, &dnode); 1328 msg_set_destnode(hdr, dnode); 1329 msg_set_destport(hdr, dport); 1330 if (unlikely(!dport && !dnode)) 1331 return -EHOSTUNREACH; 1332 } else if (dest->addrtype == TIPC_ADDR_ID) { 1333 dnode = dest->addr.id.node; 1334 msg_set_type(hdr, TIPC_DIRECT_MSG); 1335 msg_set_lookup_scope(hdr, 0); 1336 msg_set_destnode(hdr, dnode); 1337 msg_set_destport(hdr, dest->addr.id.ref); 1338 msg_set_hdr_sz(hdr, BASIC_H_SIZE); 1339 } 1340 1341 /* Block or return if destination link is congested */ 1342 rc = tipc_wait_for_cond(sock, &timeout, 1343 !tipc_dest_find(clinks, dnode, 0)); 1344 if (unlikely(rc)) 1345 return rc; 1346 1347 skb_queue_head_init(&pkts); 1348 mtu = tipc_node_get_mtu(net, dnode, tsk->portid); 1349 rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); 1350 if (unlikely(rc != dlen)) 1351 return rc; 1352 1353 rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid); 1354 if (unlikely(rc == -ELINKCONG)) { 1355 tipc_dest_push(clinks, dnode, 0); 1356 tsk->cong_link_cnt++; 1357 rc = 0; 1358 } 1359 1360 if (unlikely(syn && !rc)) 1361 tipc_set_sk_state(sk, TIPC_CONNECTING); 1362 1363 return rc ? rc : dlen; 1364 } 1365 1366 /** 1367 * tipc_sendstream - send stream-oriented data 1368 * @sock: socket structure 1369 * @m: data to send 1370 * @dsz: total length of data to be transmitted 1371 * 1372 * Used for SOCK_STREAM data. 1373 * 1374 * Returns the number of bytes sent on success (or partial success), 1375 * or errno if no data sent 1376 */ 1377 static int tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz) 1378 { 1379 struct sock *sk = sock->sk; 1380 int ret; 1381 1382 lock_sock(sk); 1383 ret = __tipc_sendstream(sock, m, dsz); 1384 release_sock(sk); 1385 1386 return ret; 1387 } 1388 1389 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dlen) 1390 { 1391 struct sock *sk = sock->sk; 1392 DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); 1393 long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); 1394 struct tipc_sock *tsk = tipc_sk(sk); 1395 struct tipc_msg *hdr = &tsk->phdr; 1396 struct net *net = sock_net(sk); 1397 struct sk_buff_head pkts; 1398 u32 dnode = tsk_peer_node(tsk); 1399 int send, sent = 0; 1400 int rc = 0; 1401 1402 skb_queue_head_init(&pkts); 1403 1404 if (unlikely(dlen > INT_MAX)) 1405 return -EMSGSIZE; 1406 1407 /* Handle implicit connection setup */ 1408 if (unlikely(dest)) { 1409 rc = __tipc_sendmsg(sock, m, dlen); 1410 if (dlen && (dlen == rc)) 1411 tsk->snt_unacked = tsk_inc(tsk, dlen + msg_hdr_sz(hdr)); 1412 return rc; 1413 } 1414 1415 do { 1416 rc = tipc_wait_for_cond(sock, &timeout, 1417 (!tsk->cong_link_cnt && 1418 !tsk_conn_cong(tsk) && 1419 tipc_sk_connected(sk))); 1420 if (unlikely(rc)) 1421 break; 1422 1423 send = min_t(size_t, dlen - sent, TIPC_MAX_USER_MSG_SIZE); 1424 rc = tipc_msg_build(hdr, m, sent, send, tsk->max_pkt, &pkts); 1425 if (unlikely(rc != send)) 1426 break; 1427 1428 rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid); 1429 if (unlikely(rc == -ELINKCONG)) { 1430 tsk->cong_link_cnt = 1; 1431 rc = 0; 1432 } 1433 if (likely(!rc)) { 1434 tsk->snt_unacked += tsk_inc(tsk, send + MIN_H_SIZE); 1435 sent += send; 1436 } 1437 } while (sent < dlen && !rc); 1438 1439 return sent ? sent : rc; 1440 } 1441 1442 /** 1443 * tipc_send_packet - send a connection-oriented message 1444 * @sock: socket structure 1445 * @m: message to send 1446 * @dsz: length of data to be transmitted 1447 * 1448 * Used for SOCK_SEQPACKET messages. 1449 * 1450 * Returns the number of bytes sent on success, or errno otherwise 1451 */ 1452 static int tipc_send_packet(struct socket *sock, struct msghdr *m, size_t dsz) 1453 { 1454 if (dsz > TIPC_MAX_USER_MSG_SIZE) 1455 return -EMSGSIZE; 1456 1457 return tipc_sendstream(sock, m, dsz); 1458 } 1459 1460 /* tipc_sk_finish_conn - complete the setup of a connection 1461 */ 1462 static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port, 1463 u32 peer_node) 1464 { 1465 struct sock *sk = &tsk->sk; 1466 struct net *net = sock_net(sk); 1467 struct tipc_msg *msg = &tsk->phdr; 1468 1469 msg_set_destnode(msg, peer_node); 1470 msg_set_destport(msg, peer_port); 1471 msg_set_type(msg, TIPC_CONN_MSG); 1472 msg_set_lookup_scope(msg, 0); 1473 msg_set_hdr_sz(msg, SHORT_H_SIZE); 1474 1475 sk_reset_timer(sk, &sk->sk_timer, jiffies + CONN_PROBING_INTV); 1476 tipc_set_sk_state(sk, TIPC_ESTABLISHED); 1477 tipc_node_add_conn(net, peer_node, tsk->portid, peer_port); 1478 tsk->max_pkt = tipc_node_get_mtu(net, peer_node, tsk->portid); 1479 tsk->peer_caps = tipc_node_get_capabilities(net, peer_node); 1480 if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) 1481 return; 1482 1483 /* Fall back to message based flow control */ 1484 tsk->rcv_win = FLOWCTL_MSG_WIN; 1485 tsk->snd_win = FLOWCTL_MSG_WIN; 1486 } 1487 1488 /** 1489 * tipc_sk_set_orig_addr - capture sender's address for received message 1490 * @m: descriptor for message info 1491 * @hdr: received message header 1492 * 1493 * Note: Address is not captured if not requested by receiver. 1494 */ 1495 static void tipc_sk_set_orig_addr(struct msghdr *m, struct sk_buff *skb) 1496 { 1497 DECLARE_SOCKADDR(struct sockaddr_pair *, srcaddr, m->msg_name); 1498 struct tipc_msg *hdr = buf_msg(skb); 1499 1500 if (!srcaddr) 1501 return; 1502 1503 srcaddr->sock.family = AF_TIPC; 1504 srcaddr->sock.addrtype = TIPC_ADDR_ID; 1505 srcaddr->sock.addr.id.ref = msg_origport(hdr); 1506 srcaddr->sock.addr.id.node = msg_orignode(hdr); 1507 srcaddr->sock.addr.name.domain = 0; 1508 srcaddr->sock.scope = 0; 1509 m->msg_namelen = sizeof(struct sockaddr_tipc); 1510 1511 if (!msg_in_group(hdr)) 1512 return; 1513 1514 /* Group message users may also want to know sending member's id */ 1515 srcaddr->member.family = AF_TIPC; 1516 srcaddr->member.addrtype = TIPC_ADDR_NAME; 1517 srcaddr->member.addr.name.name.type = msg_nametype(hdr); 1518 srcaddr->member.addr.name.name.instance = TIPC_SKB_CB(skb)->orig_member; 1519 srcaddr->member.addr.name.domain = 0; 1520 m->msg_namelen = sizeof(*srcaddr); 1521 } 1522 1523 /** 1524 * tipc_sk_anc_data_recv - optionally capture ancillary data for received message 1525 * @m: descriptor for message info 1526 * @msg: received message header 1527 * @tsk: TIPC port associated with message 1528 * 1529 * Note: Ancillary data is not captured if not requested by receiver. 1530 * 1531 * Returns 0 if successful, otherwise errno 1532 */ 1533 static int tipc_sk_anc_data_recv(struct msghdr *m, struct tipc_msg *msg, 1534 struct tipc_sock *tsk) 1535 { 1536 u32 anc_data[3]; 1537 u32 err; 1538 u32 dest_type; 1539 int has_name; 1540 int res; 1541 1542 if (likely(m->msg_controllen == 0)) 1543 return 0; 1544 1545 /* Optionally capture errored message object(s) */ 1546 err = msg ? msg_errcode(msg) : 0; 1547 if (unlikely(err)) { 1548 anc_data[0] = err; 1549 anc_data[1] = msg_data_sz(msg); 1550 res = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, anc_data); 1551 if (res) 1552 return res; 1553 if (anc_data[1]) { 1554 res = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, anc_data[1], 1555 msg_data(msg)); 1556 if (res) 1557 return res; 1558 } 1559 } 1560 1561 /* Optionally capture message destination object */ 1562 dest_type = msg ? msg_type(msg) : TIPC_DIRECT_MSG; 1563 switch (dest_type) { 1564 case TIPC_NAMED_MSG: 1565 has_name = 1; 1566 anc_data[0] = msg_nametype(msg); 1567 anc_data[1] = msg_namelower(msg); 1568 anc_data[2] = msg_namelower(msg); 1569 break; 1570 case TIPC_MCAST_MSG: 1571 has_name = 1; 1572 anc_data[0] = msg_nametype(msg); 1573 anc_data[1] = msg_namelower(msg); 1574 anc_data[2] = msg_nameupper(msg); 1575 break; 1576 case TIPC_CONN_MSG: 1577 has_name = (tsk->conn_type != 0); 1578 anc_data[0] = tsk->conn_type; 1579 anc_data[1] = tsk->conn_instance; 1580 anc_data[2] = tsk->conn_instance; 1581 break; 1582 default: 1583 has_name = 0; 1584 } 1585 if (has_name) { 1586 res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data); 1587 if (res) 1588 return res; 1589 } 1590 1591 return 0; 1592 } 1593 1594 static void tipc_sk_send_ack(struct tipc_sock *tsk) 1595 { 1596 struct sock *sk = &tsk->sk; 1597 struct net *net = sock_net(sk); 1598 struct sk_buff *skb = NULL; 1599 struct tipc_msg *msg; 1600 u32 peer_port = tsk_peer_port(tsk); 1601 u32 dnode = tsk_peer_node(tsk); 1602 1603 if (!tipc_sk_connected(sk)) 1604 return; 1605 skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, 1606 dnode, tsk_own_node(tsk), peer_port, 1607 tsk->portid, TIPC_OK); 1608 if (!skb) 1609 return; 1610 msg = buf_msg(skb); 1611 msg_set_conn_ack(msg, tsk->rcv_unacked); 1612 tsk->rcv_unacked = 0; 1613 1614 /* Adjust to and advertize the correct window limit */ 1615 if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) { 1616 tsk->rcv_win = tsk_adv_blocks(tsk->sk.sk_rcvbuf); 1617 msg_set_adv_win(msg, tsk->rcv_win); 1618 } 1619 tipc_node_xmit_skb(net, skb, dnode, msg_link_selector(msg)); 1620 } 1621 1622 static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) 1623 { 1624 struct sock *sk = sock->sk; 1625 DEFINE_WAIT(wait); 1626 long timeo = *timeop; 1627 int err = sock_error(sk); 1628 1629 if (err) 1630 return err; 1631 1632 for (;;) { 1633 prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); 1634 if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { 1635 if (sk->sk_shutdown & RCV_SHUTDOWN) { 1636 err = -ENOTCONN; 1637 break; 1638 } 1639 release_sock(sk); 1640 timeo = schedule_timeout(timeo); 1641 lock_sock(sk); 1642 } 1643 err = 0; 1644 if (!skb_queue_empty(&sk->sk_receive_queue)) 1645 break; 1646 err = -EAGAIN; 1647 if (!timeo) 1648 break; 1649 err = sock_intr_errno(timeo); 1650 if (signal_pending(current)) 1651 break; 1652 1653 err = sock_error(sk); 1654 if (err) 1655 break; 1656 } 1657 finish_wait(sk_sleep(sk), &wait); 1658 *timeop = timeo; 1659 return err; 1660 } 1661 1662 /** 1663 * tipc_recvmsg - receive packet-oriented message 1664 * @m: descriptor for message info 1665 * @buflen: length of user buffer area 1666 * @flags: receive flags 1667 * 1668 * Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages. 1669 * If the complete message doesn't fit in user area, truncate it. 1670 * 1671 * Returns size of returned message data, errno otherwise 1672 */ 1673 static int tipc_recvmsg(struct socket *sock, struct msghdr *m, 1674 size_t buflen, int flags) 1675 { 1676 struct sock *sk = sock->sk; 1677 bool connected = !tipc_sk_type_connectionless(sk); 1678 struct tipc_sock *tsk = tipc_sk(sk); 1679 int rc, err, hlen, dlen, copy; 1680 struct sk_buff_head xmitq; 1681 struct tipc_msg *hdr; 1682 struct sk_buff *skb; 1683 bool grp_evt; 1684 long timeout; 1685 1686 /* Catch invalid receive requests */ 1687 if (unlikely(!buflen)) 1688 return -EINVAL; 1689 1690 lock_sock(sk); 1691 if (unlikely(connected && sk->sk_state == TIPC_OPEN)) { 1692 rc = -ENOTCONN; 1693 goto exit; 1694 } 1695 timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); 1696 1697 /* Step rcv queue to first msg with data or error; wait if necessary */ 1698 do { 1699 rc = tipc_wait_for_rcvmsg(sock, &timeout); 1700 if (unlikely(rc)) 1701 goto exit; 1702 skb = skb_peek(&sk->sk_receive_queue); 1703 hdr = buf_msg(skb); 1704 dlen = msg_data_sz(hdr); 1705 hlen = msg_hdr_sz(hdr); 1706 err = msg_errcode(hdr); 1707 grp_evt = msg_is_grp_evt(hdr); 1708 if (likely(dlen || err)) 1709 break; 1710 tsk_advance_rx_queue(sk); 1711 } while (1); 1712 1713 /* Collect msg meta data, including error code and rejected data */ 1714 tipc_sk_set_orig_addr(m, skb); 1715 rc = tipc_sk_anc_data_recv(m, hdr, tsk); 1716 if (unlikely(rc)) 1717 goto exit; 1718 1719 /* Capture data if non-error msg, otherwise just set return value */ 1720 if (likely(!err)) { 1721 copy = min_t(int, dlen, buflen); 1722 if (unlikely(copy != dlen)) 1723 m->msg_flags |= MSG_TRUNC; 1724 rc = skb_copy_datagram_msg(skb, hlen, m, copy); 1725 } else { 1726 copy = 0; 1727 rc = 0; 1728 if (err != TIPC_CONN_SHUTDOWN && connected && !m->msg_control) 1729 rc = -ECONNRESET; 1730 } 1731 if (unlikely(rc)) 1732 goto exit; 1733 1734 /* Mark message as group event if applicable */ 1735 if (unlikely(grp_evt)) { 1736 if (msg_grp_evt(hdr) == TIPC_WITHDRAWN) 1737 m->msg_flags |= MSG_EOR; 1738 m->msg_flags |= MSG_OOB; 1739 copy = 0; 1740 } 1741 1742 /* Caption of data or error code/rejected data was successful */ 1743 if (unlikely(flags & MSG_PEEK)) 1744 goto exit; 1745 1746 /* Send group flow control advertisement when applicable */ 1747 if (tsk->group && msg_in_group(hdr) && !grp_evt) { 1748 skb_queue_head_init(&xmitq); 1749 tipc_group_update_rcv_win(tsk->group, tsk_blocks(hlen + dlen), 1750 msg_orignode(hdr), msg_origport(hdr), 1751 &xmitq); 1752 tipc_node_distr_xmit(sock_net(sk), &xmitq); 1753 } 1754 1755 tsk_advance_rx_queue(sk); 1756 1757 if (likely(!connected)) 1758 goto exit; 1759 1760 /* Send connection flow control advertisement when applicable */ 1761 tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen); 1762 if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE) 1763 tipc_sk_send_ack(tsk); 1764 exit: 1765 release_sock(sk); 1766 return rc ? rc : copy; 1767 } 1768 1769 /** 1770 * tipc_recvstream - receive stream-oriented data 1771 * @m: descriptor for message info 1772 * @buflen: total size of user buffer area 1773 * @flags: receive flags 1774 * 1775 * Used for SOCK_STREAM messages only. If not enough data is available 1776 * will optionally wait for more; never truncates data. 1777 * 1778 * Returns size of returned message data, errno otherwise 1779 */ 1780 static int tipc_recvstream(struct socket *sock, struct msghdr *m, 1781 size_t buflen, int flags) 1782 { 1783 struct sock *sk = sock->sk; 1784 struct tipc_sock *tsk = tipc_sk(sk); 1785 struct sk_buff *skb; 1786 struct tipc_msg *hdr; 1787 struct tipc_skb_cb *skb_cb; 1788 bool peek = flags & MSG_PEEK; 1789 int offset, required, copy, copied = 0; 1790 int hlen, dlen, err, rc; 1791 long timeout; 1792 1793 /* Catch invalid receive attempts */ 1794 if (unlikely(!buflen)) 1795 return -EINVAL; 1796 1797 lock_sock(sk); 1798 1799 if (unlikely(sk->sk_state == TIPC_OPEN)) { 1800 rc = -ENOTCONN; 1801 goto exit; 1802 } 1803 required = sock_rcvlowat(sk, flags & MSG_WAITALL, buflen); 1804 timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); 1805 1806 do { 1807 /* Look at first msg in receive queue; wait if necessary */ 1808 rc = tipc_wait_for_rcvmsg(sock, &timeout); 1809 if (unlikely(rc)) 1810 break; 1811 skb = skb_peek(&sk->sk_receive_queue); 1812 skb_cb = TIPC_SKB_CB(skb); 1813 hdr = buf_msg(skb); 1814 dlen = msg_data_sz(hdr); 1815 hlen = msg_hdr_sz(hdr); 1816 err = msg_errcode(hdr); 1817 1818 /* Discard any empty non-errored (SYN-) message */ 1819 if (unlikely(!dlen && !err)) { 1820 tsk_advance_rx_queue(sk); 1821 continue; 1822 } 1823 1824 /* Collect msg meta data, incl. error code and rejected data */ 1825 if (!copied) { 1826 tipc_sk_set_orig_addr(m, skb); 1827 rc = tipc_sk_anc_data_recv(m, hdr, tsk); 1828 if (rc) 1829 break; 1830 } 1831 1832 /* Copy data if msg ok, otherwise return error/partial data */ 1833 if (likely(!err)) { 1834 offset = skb_cb->bytes_read; 1835 copy = min_t(int, dlen - offset, buflen - copied); 1836 rc = skb_copy_datagram_msg(skb, hlen + offset, m, copy); 1837 if (unlikely(rc)) 1838 break; 1839 copied += copy; 1840 offset += copy; 1841 if (unlikely(offset < dlen)) { 1842 if (!peek) 1843 skb_cb->bytes_read = offset; 1844 break; 1845 } 1846 } else { 1847 rc = 0; 1848 if ((err != TIPC_CONN_SHUTDOWN) && !m->msg_control) 1849 rc = -ECONNRESET; 1850 if (copied || rc) 1851 break; 1852 } 1853 1854 if (unlikely(peek)) 1855 break; 1856 1857 tsk_advance_rx_queue(sk); 1858 1859 /* Send connection flow control advertisement when applicable */ 1860 tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen); 1861 if (unlikely(tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE)) 1862 tipc_sk_send_ack(tsk); 1863 1864 /* Exit if all requested data or FIN/error received */ 1865 if (copied == buflen || err) 1866 break; 1867 1868 } while (!skb_queue_empty(&sk->sk_receive_queue) || copied < required); 1869 exit: 1870 release_sock(sk); 1871 return copied ? copied : rc; 1872 } 1873 1874 /** 1875 * tipc_write_space - wake up thread if port congestion is released 1876 * @sk: socket 1877 */ 1878 static void tipc_write_space(struct sock *sk) 1879 { 1880 struct socket_wq *wq; 1881 1882 rcu_read_lock(); 1883 wq = rcu_dereference(sk->sk_wq); 1884 if (skwq_has_sleeper(wq)) 1885 wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | 1886 POLLWRNORM | POLLWRBAND); 1887 rcu_read_unlock(); 1888 } 1889 1890 /** 1891 * tipc_data_ready - wake up threads to indicate messages have been received 1892 * @sk: socket 1893 * @len: the length of messages 1894 */ 1895 static void tipc_data_ready(struct sock *sk) 1896 { 1897 struct socket_wq *wq; 1898 1899 rcu_read_lock(); 1900 wq = rcu_dereference(sk->sk_wq); 1901 if (skwq_has_sleeper(wq)) 1902 wake_up_interruptible_sync_poll(&wq->wait, POLLIN | 1903 POLLRDNORM | POLLRDBAND); 1904 rcu_read_unlock(); 1905 } 1906 1907 static void tipc_sock_destruct(struct sock *sk) 1908 { 1909 __skb_queue_purge(&sk->sk_receive_queue); 1910 } 1911 1912 static void tipc_sk_proto_rcv(struct sock *sk, 1913 struct sk_buff_head *inputq, 1914 struct sk_buff_head *xmitq) 1915 { 1916 struct sk_buff *skb = __skb_dequeue(inputq); 1917 struct tipc_sock *tsk = tipc_sk(sk); 1918 struct tipc_msg *hdr = buf_msg(skb); 1919 struct tipc_group *grp = tsk->group; 1920 bool wakeup = false; 1921 1922 switch (msg_user(hdr)) { 1923 case CONN_MANAGER: 1924 tipc_sk_conn_proto_rcv(tsk, skb, xmitq); 1925 return; 1926 case SOCK_WAKEUP: 1927 tipc_dest_del(&tsk->cong_links, msg_orignode(hdr), 0); 1928 tsk->cong_link_cnt--; 1929 wakeup = true; 1930 break; 1931 case GROUP_PROTOCOL: 1932 tipc_group_proto_rcv(grp, &wakeup, hdr, inputq, xmitq); 1933 break; 1934 case TOP_SRV: 1935 tipc_group_member_evt(tsk->group, &wakeup, &sk->sk_rcvbuf, 1936 skb, inputq, xmitq); 1937 skb = NULL; 1938 break; 1939 default: 1940 break; 1941 } 1942 1943 if (wakeup) 1944 sk->sk_write_space(sk); 1945 1946 kfree_skb(skb); 1947 } 1948 1949 /** 1950 * tipc_filter_connect - Handle incoming message for a connection-based socket 1951 * @tsk: TIPC socket 1952 * @skb: pointer to message buffer. Set to NULL if buffer is consumed 1953 * 1954 * Returns true if everything ok, false otherwise 1955 */ 1956 static bool tipc_sk_filter_connect(struct tipc_sock *tsk, struct sk_buff *skb) 1957 { 1958 struct sock *sk = &tsk->sk; 1959 struct net *net = sock_net(sk); 1960 struct tipc_msg *hdr = buf_msg(skb); 1961 u32 pport = msg_origport(hdr); 1962 u32 pnode = msg_orignode(hdr); 1963 1964 if (unlikely(msg_mcast(hdr))) 1965 return false; 1966 1967 switch (sk->sk_state) { 1968 case TIPC_CONNECTING: 1969 /* Accept only ACK or NACK message */ 1970 if (unlikely(!msg_connected(hdr))) { 1971 if (pport != tsk_peer_port(tsk) || 1972 pnode != tsk_peer_node(tsk)) 1973 return false; 1974 1975 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 1976 sk->sk_err = ECONNREFUSED; 1977 sk->sk_state_change(sk); 1978 return true; 1979 } 1980 1981 if (unlikely(msg_errcode(hdr))) { 1982 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 1983 sk->sk_err = ECONNREFUSED; 1984 sk->sk_state_change(sk); 1985 return true; 1986 } 1987 1988 if (unlikely(!msg_isdata(hdr))) { 1989 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 1990 sk->sk_err = EINVAL; 1991 sk->sk_state_change(sk); 1992 return true; 1993 } 1994 1995 tipc_sk_finish_conn(tsk, msg_origport(hdr), msg_orignode(hdr)); 1996 msg_set_importance(&tsk->phdr, msg_importance(hdr)); 1997 1998 /* If 'ACK+' message, add to socket receive queue */ 1999 if (msg_data_sz(hdr)) 2000 return true; 2001 2002 /* If empty 'ACK-' message, wake up sleeping connect() */ 2003 sk->sk_data_ready(sk); 2004 2005 /* 'ACK-' message is neither accepted nor rejected: */ 2006 msg_set_dest_droppable(hdr, 1); 2007 return false; 2008 2009 case TIPC_OPEN: 2010 case TIPC_DISCONNECTING: 2011 break; 2012 case TIPC_LISTEN: 2013 /* Accept only SYN message */ 2014 if (!msg_connected(hdr) && !(msg_errcode(hdr))) 2015 return true; 2016 break; 2017 case TIPC_ESTABLISHED: 2018 /* Accept only connection-based messages sent by peer */ 2019 if (unlikely(!tsk_peer_msg(tsk, hdr))) 2020 return false; 2021 2022 if (unlikely(msg_errcode(hdr))) { 2023 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 2024 /* Let timer expire on it's own */ 2025 tipc_node_remove_conn(net, tsk_peer_node(tsk), 2026 tsk->portid); 2027 sk->sk_state_change(sk); 2028 } 2029 return true; 2030 default: 2031 pr_err("Unknown sk_state %u\n", sk->sk_state); 2032 } 2033 2034 return false; 2035 } 2036 2037 /** 2038 * rcvbuf_limit - get proper overload limit of socket receive queue 2039 * @sk: socket 2040 * @skb: message 2041 * 2042 * For connection oriented messages, irrespective of importance, 2043 * default queue limit is 2 MB. 2044 * 2045 * For connectionless messages, queue limits are based on message 2046 * importance as follows: 2047 * 2048 * TIPC_LOW_IMPORTANCE (2 MB) 2049 * TIPC_MEDIUM_IMPORTANCE (4 MB) 2050 * TIPC_HIGH_IMPORTANCE (8 MB) 2051 * TIPC_CRITICAL_IMPORTANCE (16 MB) 2052 * 2053 * Returns overload limit according to corresponding message importance 2054 */ 2055 static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *skb) 2056 { 2057 struct tipc_sock *tsk = tipc_sk(sk); 2058 struct tipc_msg *hdr = buf_msg(skb); 2059 2060 if (unlikely(msg_in_group(hdr))) 2061 return sk->sk_rcvbuf; 2062 2063 if (unlikely(!msg_connected(hdr))) 2064 return sk->sk_rcvbuf << msg_importance(hdr); 2065 2066 if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL)) 2067 return sk->sk_rcvbuf; 2068 2069 return FLOWCTL_MSG_LIM; 2070 } 2071 2072 /** 2073 * tipc_sk_filter_rcv - validate incoming message 2074 * @sk: socket 2075 * @skb: pointer to message. 2076 * 2077 * Enqueues message on receive queue if acceptable; optionally handles 2078 * disconnect indication for a connected socket. 2079 * 2080 * Called with socket lock already taken 2081 * 2082 */ 2083 static void tipc_sk_filter_rcv(struct sock *sk, struct sk_buff *skb, 2084 struct sk_buff_head *xmitq) 2085 { 2086 bool sk_conn = !tipc_sk_type_connectionless(sk); 2087 struct tipc_sock *tsk = tipc_sk(sk); 2088 struct tipc_group *grp = tsk->group; 2089 struct tipc_msg *hdr = buf_msg(skb); 2090 struct net *net = sock_net(sk); 2091 struct sk_buff_head inputq; 2092 int limit, err = TIPC_OK; 2093 2094 TIPC_SKB_CB(skb)->bytes_read = 0; 2095 __skb_queue_head_init(&inputq); 2096 __skb_queue_tail(&inputq, skb); 2097 2098 if (unlikely(!msg_isdata(hdr))) 2099 tipc_sk_proto_rcv(sk, &inputq, xmitq); 2100 2101 if (unlikely(grp)) 2102 tipc_group_filter_msg(grp, &inputq, xmitq); 2103 2104 /* Validate and add to receive buffer if there is space */ 2105 while ((skb = __skb_dequeue(&inputq))) { 2106 hdr = buf_msg(skb); 2107 limit = rcvbuf_limit(sk, skb); 2108 if ((sk_conn && !tipc_sk_filter_connect(tsk, skb)) || 2109 (!sk_conn && msg_connected(hdr)) || 2110 (!grp && msg_in_group(hdr))) 2111 err = TIPC_ERR_NO_PORT; 2112 else if (sk_rmem_alloc_get(sk) + skb->truesize >= limit) 2113 err = TIPC_ERR_OVERLOAD; 2114 2115 if (unlikely(err)) { 2116 tipc_skb_reject(net, err, skb, xmitq); 2117 err = TIPC_OK; 2118 continue; 2119 } 2120 __skb_queue_tail(&sk->sk_receive_queue, skb); 2121 skb_set_owner_r(skb, sk); 2122 sk->sk_data_ready(sk); 2123 } 2124 } 2125 2126 /** 2127 * tipc_sk_backlog_rcv - handle incoming message from backlog queue 2128 * @sk: socket 2129 * @skb: message 2130 * 2131 * Caller must hold socket lock 2132 */ 2133 static int tipc_sk_backlog_rcv(struct sock *sk, struct sk_buff *skb) 2134 { 2135 unsigned int before = sk_rmem_alloc_get(sk); 2136 struct sk_buff_head xmitq; 2137 unsigned int added; 2138 2139 __skb_queue_head_init(&xmitq); 2140 2141 tipc_sk_filter_rcv(sk, skb, &xmitq); 2142 added = sk_rmem_alloc_get(sk) - before; 2143 atomic_add(added, &tipc_sk(sk)->dupl_rcvcnt); 2144 2145 /* Send pending response/rejected messages, if any */ 2146 tipc_node_distr_xmit(sock_net(sk), &xmitq); 2147 return 0; 2148 } 2149 2150 /** 2151 * tipc_sk_enqueue - extract all buffers with destination 'dport' from 2152 * inputq and try adding them to socket or backlog queue 2153 * @inputq: list of incoming buffers with potentially different destinations 2154 * @sk: socket where the buffers should be enqueued 2155 * @dport: port number for the socket 2156 * 2157 * Caller must hold socket lock 2158 */ 2159 static void tipc_sk_enqueue(struct sk_buff_head *inputq, struct sock *sk, 2160 u32 dport, struct sk_buff_head *xmitq) 2161 { 2162 unsigned long time_limit = jiffies + 2; 2163 struct sk_buff *skb; 2164 unsigned int lim; 2165 atomic_t *dcnt; 2166 u32 onode; 2167 2168 while (skb_queue_len(inputq)) { 2169 if (unlikely(time_after_eq(jiffies, time_limit))) 2170 return; 2171 2172 skb = tipc_skb_dequeue(inputq, dport); 2173 if (unlikely(!skb)) 2174 return; 2175 2176 /* Add message directly to receive queue if possible */ 2177 if (!sock_owned_by_user(sk)) { 2178 tipc_sk_filter_rcv(sk, skb, xmitq); 2179 continue; 2180 } 2181 2182 /* Try backlog, compensating for double-counted bytes */ 2183 dcnt = &tipc_sk(sk)->dupl_rcvcnt; 2184 if (!sk->sk_backlog.len) 2185 atomic_set(dcnt, 0); 2186 lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt); 2187 if (likely(!sk_add_backlog(sk, skb, lim))) 2188 continue; 2189 2190 /* Overload => reject message back to sender */ 2191 onode = tipc_own_addr(sock_net(sk)); 2192 if (tipc_msg_reverse(onode, &skb, TIPC_ERR_OVERLOAD)) 2193 __skb_queue_tail(xmitq, skb); 2194 break; 2195 } 2196 } 2197 2198 /** 2199 * tipc_sk_rcv - handle a chain of incoming buffers 2200 * @inputq: buffer list containing the buffers 2201 * Consumes all buffers in list until inputq is empty 2202 * Note: may be called in multiple threads referring to the same queue 2203 */ 2204 void tipc_sk_rcv(struct net *net, struct sk_buff_head *inputq) 2205 { 2206 struct sk_buff_head xmitq; 2207 u32 dnode, dport = 0; 2208 int err; 2209 struct tipc_sock *tsk; 2210 struct sock *sk; 2211 struct sk_buff *skb; 2212 2213 __skb_queue_head_init(&xmitq); 2214 while (skb_queue_len(inputq)) { 2215 dport = tipc_skb_peek_port(inputq, dport); 2216 tsk = tipc_sk_lookup(net, dport); 2217 2218 if (likely(tsk)) { 2219 sk = &tsk->sk; 2220 if (likely(spin_trylock_bh(&sk->sk_lock.slock))) { 2221 tipc_sk_enqueue(inputq, sk, dport, &xmitq); 2222 spin_unlock_bh(&sk->sk_lock.slock); 2223 } 2224 /* Send pending response/rejected messages, if any */ 2225 tipc_node_distr_xmit(sock_net(sk), &xmitq); 2226 sock_put(sk); 2227 continue; 2228 } 2229 /* No destination socket => dequeue skb if still there */ 2230 skb = tipc_skb_dequeue(inputq, dport); 2231 if (!skb) 2232 return; 2233 2234 /* Try secondary lookup if unresolved named message */ 2235 err = TIPC_ERR_NO_PORT; 2236 if (tipc_msg_lookup_dest(net, skb, &err)) 2237 goto xmit; 2238 2239 /* Prepare for message rejection */ 2240 if (!tipc_msg_reverse(tipc_own_addr(net), &skb, err)) 2241 continue; 2242 xmit: 2243 dnode = msg_destnode(buf_msg(skb)); 2244 tipc_node_xmit_skb(net, skb, dnode, dport); 2245 } 2246 } 2247 2248 static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) 2249 { 2250 DEFINE_WAIT_FUNC(wait, woken_wake_function); 2251 struct sock *sk = sock->sk; 2252 int done; 2253 2254 do { 2255 int err = sock_error(sk); 2256 if (err) 2257 return err; 2258 if (!*timeo_p) 2259 return -ETIMEDOUT; 2260 if (signal_pending(current)) 2261 return sock_intr_errno(*timeo_p); 2262 2263 add_wait_queue(sk_sleep(sk), &wait); 2264 done = sk_wait_event(sk, timeo_p, 2265 sk->sk_state != TIPC_CONNECTING, &wait); 2266 remove_wait_queue(sk_sleep(sk), &wait); 2267 } while (!done); 2268 return 0; 2269 } 2270 2271 /** 2272 * tipc_connect - establish a connection to another TIPC port 2273 * @sock: socket structure 2274 * @dest: socket address for destination port 2275 * @destlen: size of socket address data structure 2276 * @flags: file-related flags associated with socket 2277 * 2278 * Returns 0 on success, errno otherwise 2279 */ 2280 static int tipc_connect(struct socket *sock, struct sockaddr *dest, 2281 int destlen, int flags) 2282 { 2283 struct sock *sk = sock->sk; 2284 struct tipc_sock *tsk = tipc_sk(sk); 2285 struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; 2286 struct msghdr m = {NULL,}; 2287 long timeout = (flags & O_NONBLOCK) ? 0 : tsk->conn_timeout; 2288 int previous; 2289 int res = 0; 2290 2291 if (destlen != sizeof(struct sockaddr_tipc)) 2292 return -EINVAL; 2293 2294 lock_sock(sk); 2295 2296 if (tsk->group) { 2297 res = -EINVAL; 2298 goto exit; 2299 } 2300 2301 if (dst->family == AF_UNSPEC) { 2302 memset(&tsk->peer, 0, sizeof(struct sockaddr_tipc)); 2303 if (!tipc_sk_type_connectionless(sk)) 2304 res = -EINVAL; 2305 goto exit; 2306 } else if (dst->family != AF_TIPC) { 2307 res = -EINVAL; 2308 } 2309 if (dst->addrtype != TIPC_ADDR_ID && dst->addrtype != TIPC_ADDR_NAME) 2310 res = -EINVAL; 2311 if (res) 2312 goto exit; 2313 2314 /* DGRAM/RDM connect(), just save the destaddr */ 2315 if (tipc_sk_type_connectionless(sk)) { 2316 memcpy(&tsk->peer, dest, destlen); 2317 goto exit; 2318 } 2319 2320 previous = sk->sk_state; 2321 2322 switch (sk->sk_state) { 2323 case TIPC_OPEN: 2324 /* Send a 'SYN-' to destination */ 2325 m.msg_name = dest; 2326 m.msg_namelen = destlen; 2327 2328 /* If connect is in non-blocking case, set MSG_DONTWAIT to 2329 * indicate send_msg() is never blocked. 2330 */ 2331 if (!timeout) 2332 m.msg_flags = MSG_DONTWAIT; 2333 2334 res = __tipc_sendmsg(sock, &m, 0); 2335 if ((res < 0) && (res != -EWOULDBLOCK)) 2336 goto exit; 2337 2338 /* Just entered TIPC_CONNECTING state; the only 2339 * difference is that return value in non-blocking 2340 * case is EINPROGRESS, rather than EALREADY. 2341 */ 2342 res = -EINPROGRESS; 2343 /* fall thru' */ 2344 case TIPC_CONNECTING: 2345 if (!timeout) { 2346 if (previous == TIPC_CONNECTING) 2347 res = -EALREADY; 2348 goto exit; 2349 } 2350 timeout = msecs_to_jiffies(timeout); 2351 /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ 2352 res = tipc_wait_for_connect(sock, &timeout); 2353 break; 2354 case TIPC_ESTABLISHED: 2355 res = -EISCONN; 2356 break; 2357 default: 2358 res = -EINVAL; 2359 } 2360 2361 exit: 2362 release_sock(sk); 2363 return res; 2364 } 2365 2366 /** 2367 * tipc_listen - allow socket to listen for incoming connections 2368 * @sock: socket structure 2369 * @len: (unused) 2370 * 2371 * Returns 0 on success, errno otherwise 2372 */ 2373 static int tipc_listen(struct socket *sock, int len) 2374 { 2375 struct sock *sk = sock->sk; 2376 int res; 2377 2378 lock_sock(sk); 2379 res = tipc_set_sk_state(sk, TIPC_LISTEN); 2380 release_sock(sk); 2381 2382 return res; 2383 } 2384 2385 static int tipc_wait_for_accept(struct socket *sock, long timeo) 2386 { 2387 struct sock *sk = sock->sk; 2388 DEFINE_WAIT(wait); 2389 int err; 2390 2391 /* True wake-one mechanism for incoming connections: only 2392 * one process gets woken up, not the 'whole herd'. 2393 * Since we do not 'race & poll' for established sockets 2394 * anymore, the common case will execute the loop only once. 2395 */ 2396 for (;;) { 2397 prepare_to_wait_exclusive(sk_sleep(sk), &wait, 2398 TASK_INTERRUPTIBLE); 2399 if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { 2400 release_sock(sk); 2401 timeo = schedule_timeout(timeo); 2402 lock_sock(sk); 2403 } 2404 err = 0; 2405 if (!skb_queue_empty(&sk->sk_receive_queue)) 2406 break; 2407 err = -EAGAIN; 2408 if (!timeo) 2409 break; 2410 err = sock_intr_errno(timeo); 2411 if (signal_pending(current)) 2412 break; 2413 } 2414 finish_wait(sk_sleep(sk), &wait); 2415 return err; 2416 } 2417 2418 /** 2419 * tipc_accept - wait for connection request 2420 * @sock: listening socket 2421 * @newsock: new socket that is to be connected 2422 * @flags: file-related flags associated with socket 2423 * 2424 * Returns 0 on success, errno otherwise 2425 */ 2426 static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, 2427 bool kern) 2428 { 2429 struct sock *new_sk, *sk = sock->sk; 2430 struct sk_buff *buf; 2431 struct tipc_sock *new_tsock; 2432 struct tipc_msg *msg; 2433 long timeo; 2434 int res; 2435 2436 lock_sock(sk); 2437 2438 if (sk->sk_state != TIPC_LISTEN) { 2439 res = -EINVAL; 2440 goto exit; 2441 } 2442 timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); 2443 res = tipc_wait_for_accept(sock, timeo); 2444 if (res) 2445 goto exit; 2446 2447 buf = skb_peek(&sk->sk_receive_queue); 2448 2449 res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, kern); 2450 if (res) 2451 goto exit; 2452 security_sk_clone(sock->sk, new_sock->sk); 2453 2454 new_sk = new_sock->sk; 2455 new_tsock = tipc_sk(new_sk); 2456 msg = buf_msg(buf); 2457 2458 /* we lock on new_sk; but lockdep sees the lock on sk */ 2459 lock_sock_nested(new_sk, SINGLE_DEPTH_NESTING); 2460 2461 /* 2462 * Reject any stray messages received by new socket 2463 * before the socket lock was taken (very, very unlikely) 2464 */ 2465 tsk_rej_rx_queue(new_sk); 2466 2467 /* Connect new socket to it's peer */ 2468 tipc_sk_finish_conn(new_tsock, msg_origport(msg), msg_orignode(msg)); 2469 2470 tsk_set_importance(new_tsock, msg_importance(msg)); 2471 if (msg_named(msg)) { 2472 new_tsock->conn_type = msg_nametype(msg); 2473 new_tsock->conn_instance = msg_nameinst(msg); 2474 } 2475 2476 /* 2477 * Respond to 'SYN-' by discarding it & returning 'ACK'-. 2478 * Respond to 'SYN+' by queuing it on new socket. 2479 */ 2480 if (!msg_data_sz(msg)) { 2481 struct msghdr m = {NULL,}; 2482 2483 tsk_advance_rx_queue(sk); 2484 __tipc_sendstream(new_sock, &m, 0); 2485 } else { 2486 __skb_dequeue(&sk->sk_receive_queue); 2487 __skb_queue_head(&new_sk->sk_receive_queue, buf); 2488 skb_set_owner_r(buf, new_sk); 2489 } 2490 release_sock(new_sk); 2491 exit: 2492 release_sock(sk); 2493 return res; 2494 } 2495 2496 /** 2497 * tipc_shutdown - shutdown socket connection 2498 * @sock: socket structure 2499 * @how: direction to close (must be SHUT_RDWR) 2500 * 2501 * Terminates connection (if necessary), then purges socket's receive queue. 2502 * 2503 * Returns 0 on success, errno otherwise 2504 */ 2505 static int tipc_shutdown(struct socket *sock, int how) 2506 { 2507 struct sock *sk = sock->sk; 2508 int res; 2509 2510 if (how != SHUT_RDWR) 2511 return -EINVAL; 2512 2513 lock_sock(sk); 2514 2515 __tipc_shutdown(sock, TIPC_CONN_SHUTDOWN); 2516 sk->sk_shutdown = SEND_SHUTDOWN; 2517 2518 if (sk->sk_state == TIPC_DISCONNECTING) { 2519 /* Discard any unreceived messages */ 2520 __skb_queue_purge(&sk->sk_receive_queue); 2521 2522 /* Wake up anyone sleeping in poll */ 2523 sk->sk_state_change(sk); 2524 res = 0; 2525 } else { 2526 res = -ENOTCONN; 2527 } 2528 2529 release_sock(sk); 2530 return res; 2531 } 2532 2533 static void tipc_sk_timeout(struct timer_list *t) 2534 { 2535 struct sock *sk = from_timer(sk, t, sk_timer); 2536 struct tipc_sock *tsk = tipc_sk(sk); 2537 u32 peer_port = tsk_peer_port(tsk); 2538 u32 peer_node = tsk_peer_node(tsk); 2539 u32 own_node = tsk_own_node(tsk); 2540 u32 own_port = tsk->portid; 2541 struct net *net = sock_net(sk); 2542 struct sk_buff *skb = NULL; 2543 2544 bh_lock_sock(sk); 2545 if (!tipc_sk_connected(sk)) 2546 goto exit; 2547 2548 /* Try again later if socket is busy */ 2549 if (sock_owned_by_user(sk)) { 2550 sk_reset_timer(sk, &sk->sk_timer, jiffies + HZ / 20); 2551 goto exit; 2552 } 2553 2554 if (tsk->probe_unacked) { 2555 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 2556 tipc_node_remove_conn(net, peer_node, peer_port); 2557 sk->sk_state_change(sk); 2558 goto exit; 2559 } 2560 /* Send new probe */ 2561 skb = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE, 0, 2562 peer_node, own_node, peer_port, own_port, 2563 TIPC_OK); 2564 tsk->probe_unacked = true; 2565 sk_reset_timer(sk, &sk->sk_timer, jiffies + CONN_PROBING_INTV); 2566 exit: 2567 bh_unlock_sock(sk); 2568 if (skb) 2569 tipc_node_xmit_skb(net, skb, peer_node, own_port); 2570 sock_put(sk); 2571 } 2572 2573 static int tipc_sk_publish(struct tipc_sock *tsk, uint scope, 2574 struct tipc_name_seq const *seq) 2575 { 2576 struct sock *sk = &tsk->sk; 2577 struct net *net = sock_net(sk); 2578 struct publication *publ; 2579 u32 key; 2580 2581 if (tipc_sk_connected(sk)) 2582 return -EINVAL; 2583 key = tsk->portid + tsk->pub_count + 1; 2584 if (key == tsk->portid) 2585 return -EADDRINUSE; 2586 2587 publ = tipc_nametbl_publish(net, seq->type, seq->lower, seq->upper, 2588 scope, tsk->portid, key); 2589 if (unlikely(!publ)) 2590 return -EINVAL; 2591 2592 list_add(&publ->pport_list, &tsk->publications); 2593 tsk->pub_count++; 2594 tsk->published = 1; 2595 return 0; 2596 } 2597 2598 static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope, 2599 struct tipc_name_seq const *seq) 2600 { 2601 struct net *net = sock_net(&tsk->sk); 2602 struct publication *publ; 2603 struct publication *safe; 2604 int rc = -EINVAL; 2605 2606 list_for_each_entry_safe(publ, safe, &tsk->publications, pport_list) { 2607 if (seq) { 2608 if (publ->scope != scope) 2609 continue; 2610 if (publ->type != seq->type) 2611 continue; 2612 if (publ->lower != seq->lower) 2613 continue; 2614 if (publ->upper != seq->upper) 2615 break; 2616 tipc_nametbl_withdraw(net, publ->type, publ->lower, 2617 publ->ref, publ->key); 2618 rc = 0; 2619 break; 2620 } 2621 tipc_nametbl_withdraw(net, publ->type, publ->lower, 2622 publ->ref, publ->key); 2623 rc = 0; 2624 } 2625 if (list_empty(&tsk->publications)) 2626 tsk->published = 0; 2627 return rc; 2628 } 2629 2630 /* tipc_sk_reinit: set non-zero address in all existing sockets 2631 * when we go from standalone to network mode. 2632 */ 2633 void tipc_sk_reinit(struct net *net) 2634 { 2635 struct tipc_net *tn = net_generic(net, tipc_net_id); 2636 struct rhashtable_iter iter; 2637 struct tipc_sock *tsk; 2638 struct tipc_msg *msg; 2639 2640 rhashtable_walk_enter(&tn->sk_rht, &iter); 2641 2642 do { 2643 tsk = ERR_PTR(rhashtable_walk_start(&iter)); 2644 if (IS_ERR(tsk)) 2645 goto walk_stop; 2646 2647 while ((tsk = rhashtable_walk_next(&iter)) && !IS_ERR(tsk)) { 2648 spin_lock_bh(&tsk->sk.sk_lock.slock); 2649 msg = &tsk->phdr; 2650 msg_set_prevnode(msg, tn->own_addr); 2651 msg_set_orignode(msg, tn->own_addr); 2652 spin_unlock_bh(&tsk->sk.sk_lock.slock); 2653 } 2654 walk_stop: 2655 rhashtable_walk_stop(&iter); 2656 } while (tsk == ERR_PTR(-EAGAIN)); 2657 } 2658 2659 static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid) 2660 { 2661 struct tipc_net *tn = net_generic(net, tipc_net_id); 2662 struct tipc_sock *tsk; 2663 2664 rcu_read_lock(); 2665 tsk = rhashtable_lookup_fast(&tn->sk_rht, &portid, tsk_rht_params); 2666 if (tsk) 2667 sock_hold(&tsk->sk); 2668 rcu_read_unlock(); 2669 2670 return tsk; 2671 } 2672 2673 static int tipc_sk_insert(struct tipc_sock *tsk) 2674 { 2675 struct sock *sk = &tsk->sk; 2676 struct net *net = sock_net(sk); 2677 struct tipc_net *tn = net_generic(net, tipc_net_id); 2678 u32 remaining = (TIPC_MAX_PORT - TIPC_MIN_PORT) + 1; 2679 u32 portid = prandom_u32() % remaining + TIPC_MIN_PORT; 2680 2681 while (remaining--) { 2682 portid++; 2683 if ((portid < TIPC_MIN_PORT) || (portid > TIPC_MAX_PORT)) 2684 portid = TIPC_MIN_PORT; 2685 tsk->portid = portid; 2686 sock_hold(&tsk->sk); 2687 if (!rhashtable_lookup_insert_fast(&tn->sk_rht, &tsk->node, 2688 tsk_rht_params)) 2689 return 0; 2690 sock_put(&tsk->sk); 2691 } 2692 2693 return -1; 2694 } 2695 2696 static void tipc_sk_remove(struct tipc_sock *tsk) 2697 { 2698 struct sock *sk = &tsk->sk; 2699 struct tipc_net *tn = net_generic(sock_net(sk), tipc_net_id); 2700 2701 if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params)) { 2702 WARN_ON(refcount_read(&sk->sk_refcnt) == 1); 2703 __sock_put(sk); 2704 } 2705 } 2706 2707 static const struct rhashtable_params tsk_rht_params = { 2708 .nelem_hint = 192, 2709 .head_offset = offsetof(struct tipc_sock, node), 2710 .key_offset = offsetof(struct tipc_sock, portid), 2711 .key_len = sizeof(u32), /* portid */ 2712 .max_size = 1048576, 2713 .min_size = 256, 2714 .automatic_shrinking = true, 2715 }; 2716 2717 int tipc_sk_rht_init(struct net *net) 2718 { 2719 struct tipc_net *tn = net_generic(net, tipc_net_id); 2720 2721 return rhashtable_init(&tn->sk_rht, &tsk_rht_params); 2722 } 2723 2724 void tipc_sk_rht_destroy(struct net *net) 2725 { 2726 struct tipc_net *tn = net_generic(net, tipc_net_id); 2727 2728 /* Wait for socket readers to complete */ 2729 synchronize_net(); 2730 2731 rhashtable_destroy(&tn->sk_rht); 2732 } 2733 2734 static int tipc_sk_join(struct tipc_sock *tsk, struct tipc_group_req *mreq) 2735 { 2736 struct net *net = sock_net(&tsk->sk); 2737 u32 domain = addr_domain(net, mreq->scope); 2738 struct tipc_group *grp = tsk->group; 2739 struct tipc_msg *hdr = &tsk->phdr; 2740 struct tipc_name_seq seq; 2741 int rc; 2742 2743 if (mreq->type < TIPC_RESERVED_TYPES) 2744 return -EACCES; 2745 if (grp) 2746 return -EACCES; 2747 grp = tipc_group_create(net, tsk->portid, mreq); 2748 if (!grp) 2749 return -ENOMEM; 2750 tsk->group = grp; 2751 msg_set_lookup_scope(hdr, mreq->scope); 2752 msg_set_nametype(hdr, mreq->type); 2753 msg_set_dest_droppable(hdr, true); 2754 seq.type = mreq->type; 2755 seq.lower = mreq->instance; 2756 seq.upper = seq.lower; 2757 tipc_nametbl_build_group(net, grp, mreq->type, domain); 2758 rc = tipc_sk_publish(tsk, mreq->scope, &seq); 2759 if (rc) { 2760 tipc_group_delete(net, grp); 2761 tsk->group = NULL; 2762 } 2763 2764 /* Eliminate any risk that a broadcast overtakes the sent JOIN */ 2765 tsk->mc_method.rcast = true; 2766 tsk->mc_method.mandatory = true; 2767 return rc; 2768 } 2769 2770 static int tipc_sk_leave(struct tipc_sock *tsk) 2771 { 2772 struct net *net = sock_net(&tsk->sk); 2773 struct tipc_group *grp = tsk->group; 2774 struct tipc_name_seq seq; 2775 int scope; 2776 2777 if (!grp) 2778 return -EINVAL; 2779 tipc_group_self(grp, &seq, &scope); 2780 tipc_group_delete(net, grp); 2781 tsk->group = NULL; 2782 tipc_sk_withdraw(tsk, scope, &seq); 2783 return 0; 2784 } 2785 2786 /** 2787 * tipc_setsockopt - set socket option 2788 * @sock: socket structure 2789 * @lvl: option level 2790 * @opt: option identifier 2791 * @ov: pointer to new option value 2792 * @ol: length of option value 2793 * 2794 * For stream sockets only, accepts and ignores all IPPROTO_TCP options 2795 * (to ease compatibility). 2796 * 2797 * Returns 0 on success, errno otherwise 2798 */ 2799 static int tipc_setsockopt(struct socket *sock, int lvl, int opt, 2800 char __user *ov, unsigned int ol) 2801 { 2802 struct sock *sk = sock->sk; 2803 struct tipc_sock *tsk = tipc_sk(sk); 2804 struct tipc_group_req mreq; 2805 u32 value = 0; 2806 int res = 0; 2807 2808 if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM)) 2809 return 0; 2810 if (lvl != SOL_TIPC) 2811 return -ENOPROTOOPT; 2812 2813 switch (opt) { 2814 case TIPC_IMPORTANCE: 2815 case TIPC_SRC_DROPPABLE: 2816 case TIPC_DEST_DROPPABLE: 2817 case TIPC_CONN_TIMEOUT: 2818 if (ol < sizeof(value)) 2819 return -EINVAL; 2820 if (get_user(value, (u32 __user *)ov)) 2821 return -EFAULT; 2822 break; 2823 case TIPC_GROUP_JOIN: 2824 if (ol < sizeof(mreq)) 2825 return -EINVAL; 2826 if (copy_from_user(&mreq, ov, sizeof(mreq))) 2827 return -EFAULT; 2828 break; 2829 default: 2830 if (ov || ol) 2831 return -EINVAL; 2832 } 2833 2834 lock_sock(sk); 2835 2836 switch (opt) { 2837 case TIPC_IMPORTANCE: 2838 res = tsk_set_importance(tsk, value); 2839 break; 2840 case TIPC_SRC_DROPPABLE: 2841 if (sock->type != SOCK_STREAM) 2842 tsk_set_unreliable(tsk, value); 2843 else 2844 res = -ENOPROTOOPT; 2845 break; 2846 case TIPC_DEST_DROPPABLE: 2847 tsk_set_unreturnable(tsk, value); 2848 break; 2849 case TIPC_CONN_TIMEOUT: 2850 tipc_sk(sk)->conn_timeout = value; 2851 break; 2852 case TIPC_MCAST_BROADCAST: 2853 tsk->mc_method.rcast = false; 2854 tsk->mc_method.mandatory = true; 2855 break; 2856 case TIPC_MCAST_REPLICAST: 2857 tsk->mc_method.rcast = true; 2858 tsk->mc_method.mandatory = true; 2859 break; 2860 case TIPC_GROUP_JOIN: 2861 res = tipc_sk_join(tsk, &mreq); 2862 break; 2863 case TIPC_GROUP_LEAVE: 2864 res = tipc_sk_leave(tsk); 2865 break; 2866 default: 2867 res = -EINVAL; 2868 } 2869 2870 release_sock(sk); 2871 2872 return res; 2873 } 2874 2875 /** 2876 * tipc_getsockopt - get socket option 2877 * @sock: socket structure 2878 * @lvl: option level 2879 * @opt: option identifier 2880 * @ov: receptacle for option value 2881 * @ol: receptacle for length of option value 2882 * 2883 * For stream sockets only, returns 0 length result for all IPPROTO_TCP options 2884 * (to ease compatibility). 2885 * 2886 * Returns 0 on success, errno otherwise 2887 */ 2888 static int tipc_getsockopt(struct socket *sock, int lvl, int opt, 2889 char __user *ov, int __user *ol) 2890 { 2891 struct sock *sk = sock->sk; 2892 struct tipc_sock *tsk = tipc_sk(sk); 2893 struct tipc_name_seq seq; 2894 int len, scope; 2895 u32 value; 2896 int res; 2897 2898 if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM)) 2899 return put_user(0, ol); 2900 if (lvl != SOL_TIPC) 2901 return -ENOPROTOOPT; 2902 res = get_user(len, ol); 2903 if (res) 2904 return res; 2905 2906 lock_sock(sk); 2907 2908 switch (opt) { 2909 case TIPC_IMPORTANCE: 2910 value = tsk_importance(tsk); 2911 break; 2912 case TIPC_SRC_DROPPABLE: 2913 value = tsk_unreliable(tsk); 2914 break; 2915 case TIPC_DEST_DROPPABLE: 2916 value = tsk_unreturnable(tsk); 2917 break; 2918 case TIPC_CONN_TIMEOUT: 2919 value = tsk->conn_timeout; 2920 /* no need to set "res", since already 0 at this point */ 2921 break; 2922 case TIPC_NODE_RECVQ_DEPTH: 2923 value = 0; /* was tipc_queue_size, now obsolete */ 2924 break; 2925 case TIPC_SOCK_RECVQ_DEPTH: 2926 value = skb_queue_len(&sk->sk_receive_queue); 2927 break; 2928 case TIPC_GROUP_JOIN: 2929 seq.type = 0; 2930 if (tsk->group) 2931 tipc_group_self(tsk->group, &seq, &scope); 2932 value = seq.type; 2933 break; 2934 default: 2935 res = -EINVAL; 2936 } 2937 2938 release_sock(sk); 2939 2940 if (res) 2941 return res; /* "get" failed */ 2942 2943 if (len < sizeof(value)) 2944 return -EINVAL; 2945 2946 if (copy_to_user(ov, &value, sizeof(value))) 2947 return -EFAULT; 2948 2949 return put_user(sizeof(value), ol); 2950 } 2951 2952 static int tipc_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) 2953 { 2954 struct sock *sk = sock->sk; 2955 struct tipc_sioc_ln_req lnr; 2956 void __user *argp = (void __user *)arg; 2957 2958 switch (cmd) { 2959 case SIOCGETLINKNAME: 2960 if (copy_from_user(&lnr, argp, sizeof(lnr))) 2961 return -EFAULT; 2962 if (!tipc_node_get_linkname(sock_net(sk), 2963 lnr.bearer_id & 0xffff, lnr.peer, 2964 lnr.linkname, TIPC_MAX_LINK_NAME)) { 2965 if (copy_to_user(argp, &lnr, sizeof(lnr))) 2966 return -EFAULT; 2967 return 0; 2968 } 2969 return -EADDRNOTAVAIL; 2970 default: 2971 return -ENOIOCTLCMD; 2972 } 2973 } 2974 2975 static int tipc_socketpair(struct socket *sock1, struct socket *sock2) 2976 { 2977 struct tipc_sock *tsk2 = tipc_sk(sock2->sk); 2978 struct tipc_sock *tsk1 = tipc_sk(sock1->sk); 2979 u32 onode = tipc_own_addr(sock_net(sock1->sk)); 2980 2981 tsk1->peer.family = AF_TIPC; 2982 tsk1->peer.addrtype = TIPC_ADDR_ID; 2983 tsk1->peer.scope = TIPC_NODE_SCOPE; 2984 tsk1->peer.addr.id.ref = tsk2->portid; 2985 tsk1->peer.addr.id.node = onode; 2986 tsk2->peer.family = AF_TIPC; 2987 tsk2->peer.addrtype = TIPC_ADDR_ID; 2988 tsk2->peer.scope = TIPC_NODE_SCOPE; 2989 tsk2->peer.addr.id.ref = tsk1->portid; 2990 tsk2->peer.addr.id.node = onode; 2991 2992 tipc_sk_finish_conn(tsk1, tsk2->portid, onode); 2993 tipc_sk_finish_conn(tsk2, tsk1->portid, onode); 2994 return 0; 2995 } 2996 2997 /* Protocol switches for the various types of TIPC sockets */ 2998 2999 static const struct proto_ops msg_ops = { 3000 .owner = THIS_MODULE, 3001 .family = AF_TIPC, 3002 .release = tipc_release, 3003 .bind = tipc_bind, 3004 .connect = tipc_connect, 3005 .socketpair = tipc_socketpair, 3006 .accept = sock_no_accept, 3007 .getname = tipc_getname, 3008 .poll = tipc_poll, 3009 .ioctl = tipc_ioctl, 3010 .listen = sock_no_listen, 3011 .shutdown = tipc_shutdown, 3012 .setsockopt = tipc_setsockopt, 3013 .getsockopt = tipc_getsockopt, 3014 .sendmsg = tipc_sendmsg, 3015 .recvmsg = tipc_recvmsg, 3016 .mmap = sock_no_mmap, 3017 .sendpage = sock_no_sendpage 3018 }; 3019 3020 static const struct proto_ops packet_ops = { 3021 .owner = THIS_MODULE, 3022 .family = AF_TIPC, 3023 .release = tipc_release, 3024 .bind = tipc_bind, 3025 .connect = tipc_connect, 3026 .socketpair = tipc_socketpair, 3027 .accept = tipc_accept, 3028 .getname = tipc_getname, 3029 .poll = tipc_poll, 3030 .ioctl = tipc_ioctl, 3031 .listen = tipc_listen, 3032 .shutdown = tipc_shutdown, 3033 .setsockopt = tipc_setsockopt, 3034 .getsockopt = tipc_getsockopt, 3035 .sendmsg = tipc_send_packet, 3036 .recvmsg = tipc_recvmsg, 3037 .mmap = sock_no_mmap, 3038 .sendpage = sock_no_sendpage 3039 }; 3040 3041 static const struct proto_ops stream_ops = { 3042 .owner = THIS_MODULE, 3043 .family = AF_TIPC, 3044 .release = tipc_release, 3045 .bind = tipc_bind, 3046 .connect = tipc_connect, 3047 .socketpair = tipc_socketpair, 3048 .accept = tipc_accept, 3049 .getname = tipc_getname, 3050 .poll = tipc_poll, 3051 .ioctl = tipc_ioctl, 3052 .listen = tipc_listen, 3053 .shutdown = tipc_shutdown, 3054 .setsockopt = tipc_setsockopt, 3055 .getsockopt = tipc_getsockopt, 3056 .sendmsg = tipc_sendstream, 3057 .recvmsg = tipc_recvstream, 3058 .mmap = sock_no_mmap, 3059 .sendpage = sock_no_sendpage 3060 }; 3061 3062 static const struct net_proto_family tipc_family_ops = { 3063 .owner = THIS_MODULE, 3064 .family = AF_TIPC, 3065 .create = tipc_sk_create 3066 }; 3067 3068 static struct proto tipc_proto = { 3069 .name = "TIPC", 3070 .owner = THIS_MODULE, 3071 .obj_size = sizeof(struct tipc_sock), 3072 .sysctl_rmem = sysctl_tipc_rmem 3073 }; 3074 3075 /** 3076 * tipc_socket_init - initialize TIPC socket interface 3077 * 3078 * Returns 0 on success, errno otherwise 3079 */ 3080 int tipc_socket_init(void) 3081 { 3082 int res; 3083 3084 res = proto_register(&tipc_proto, 1); 3085 if (res) { 3086 pr_err("Failed to register TIPC protocol type\n"); 3087 goto out; 3088 } 3089 3090 res = sock_register(&tipc_family_ops); 3091 if (res) { 3092 pr_err("Failed to register TIPC socket type\n"); 3093 proto_unregister(&tipc_proto); 3094 goto out; 3095 } 3096 out: 3097 return res; 3098 } 3099 3100 /** 3101 * tipc_socket_stop - stop TIPC socket interface 3102 */ 3103 void tipc_socket_stop(void) 3104 { 3105 sock_unregister(tipc_family_ops.family); 3106 proto_unregister(&tipc_proto); 3107 } 3108 3109 /* Caller should hold socket lock for the passed tipc socket. */ 3110 static int __tipc_nl_add_sk_con(struct sk_buff *skb, struct tipc_sock *tsk) 3111 { 3112 u32 peer_node; 3113 u32 peer_port; 3114 struct nlattr *nest; 3115 3116 peer_node = tsk_peer_node(tsk); 3117 peer_port = tsk_peer_port(tsk); 3118 3119 nest = nla_nest_start(skb, TIPC_NLA_SOCK_CON); 3120 3121 if (nla_put_u32(skb, TIPC_NLA_CON_NODE, peer_node)) 3122 goto msg_full; 3123 if (nla_put_u32(skb, TIPC_NLA_CON_SOCK, peer_port)) 3124 goto msg_full; 3125 3126 if (tsk->conn_type != 0) { 3127 if (nla_put_flag(skb, TIPC_NLA_CON_FLAG)) 3128 goto msg_full; 3129 if (nla_put_u32(skb, TIPC_NLA_CON_TYPE, tsk->conn_type)) 3130 goto msg_full; 3131 if (nla_put_u32(skb, TIPC_NLA_CON_INST, tsk->conn_instance)) 3132 goto msg_full; 3133 } 3134 nla_nest_end(skb, nest); 3135 3136 return 0; 3137 3138 msg_full: 3139 nla_nest_cancel(skb, nest); 3140 3141 return -EMSGSIZE; 3142 } 3143 3144 /* Caller should hold socket lock for the passed tipc socket. */ 3145 static int __tipc_nl_add_sk(struct sk_buff *skb, struct netlink_callback *cb, 3146 struct tipc_sock *tsk) 3147 { 3148 int err; 3149 void *hdr; 3150 struct nlattr *attrs; 3151 struct net *net = sock_net(skb->sk); 3152 struct tipc_net *tn = net_generic(net, tipc_net_id); 3153 struct sock *sk = &tsk->sk; 3154 3155 hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, 3156 &tipc_genl_family, NLM_F_MULTI, TIPC_NL_SOCK_GET); 3157 if (!hdr) 3158 goto msg_cancel; 3159 3160 attrs = nla_nest_start(skb, TIPC_NLA_SOCK); 3161 if (!attrs) 3162 goto genlmsg_cancel; 3163 if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->portid)) 3164 goto attr_msg_cancel; 3165 if (nla_put_u32(skb, TIPC_NLA_SOCK_ADDR, tn->own_addr)) 3166 goto attr_msg_cancel; 3167 3168 if (tipc_sk_connected(sk)) { 3169 err = __tipc_nl_add_sk_con(skb, tsk); 3170 if (err) 3171 goto attr_msg_cancel; 3172 } else if (!list_empty(&tsk->publications)) { 3173 if (nla_put_flag(skb, TIPC_NLA_SOCK_HAS_PUBL)) 3174 goto attr_msg_cancel; 3175 } 3176 nla_nest_end(skb, attrs); 3177 genlmsg_end(skb, hdr); 3178 3179 return 0; 3180 3181 attr_msg_cancel: 3182 nla_nest_cancel(skb, attrs); 3183 genlmsg_cancel: 3184 genlmsg_cancel(skb, hdr); 3185 msg_cancel: 3186 return -EMSGSIZE; 3187 } 3188 3189 int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb) 3190 { 3191 int err; 3192 struct tipc_sock *tsk; 3193 const struct bucket_table *tbl; 3194 struct rhash_head *pos; 3195 struct net *net = sock_net(skb->sk); 3196 struct tipc_net *tn = net_generic(net, tipc_net_id); 3197 u32 tbl_id = cb->args[0]; 3198 u32 prev_portid = cb->args[1]; 3199 3200 rcu_read_lock(); 3201 tbl = rht_dereference_rcu((&tn->sk_rht)->tbl, &tn->sk_rht); 3202 for (; tbl_id < tbl->size; tbl_id++) { 3203 rht_for_each_entry_rcu(tsk, pos, tbl, tbl_id, node) { 3204 spin_lock_bh(&tsk->sk.sk_lock.slock); 3205 if (prev_portid && prev_portid != tsk->portid) { 3206 spin_unlock_bh(&tsk->sk.sk_lock.slock); 3207 continue; 3208 } 3209 3210 err = __tipc_nl_add_sk(skb, cb, tsk); 3211 if (err) { 3212 prev_portid = tsk->portid; 3213 spin_unlock_bh(&tsk->sk.sk_lock.slock); 3214 goto out; 3215 } 3216 prev_portid = 0; 3217 spin_unlock_bh(&tsk->sk.sk_lock.slock); 3218 } 3219 } 3220 out: 3221 rcu_read_unlock(); 3222 cb->args[0] = tbl_id; 3223 cb->args[1] = prev_portid; 3224 3225 return skb->len; 3226 } 3227 3228 /* Caller should hold socket lock for the passed tipc socket. */ 3229 static int __tipc_nl_add_sk_publ(struct sk_buff *skb, 3230 struct netlink_callback *cb, 3231 struct publication *publ) 3232 { 3233 void *hdr; 3234 struct nlattr *attrs; 3235 3236 hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, 3237 &tipc_genl_family, NLM_F_MULTI, TIPC_NL_PUBL_GET); 3238 if (!hdr) 3239 goto msg_cancel; 3240 3241 attrs = nla_nest_start(skb, TIPC_NLA_PUBL); 3242 if (!attrs) 3243 goto genlmsg_cancel; 3244 3245 if (nla_put_u32(skb, TIPC_NLA_PUBL_KEY, publ->key)) 3246 goto attr_msg_cancel; 3247 if (nla_put_u32(skb, TIPC_NLA_PUBL_TYPE, publ->type)) 3248 goto attr_msg_cancel; 3249 if (nla_put_u32(skb, TIPC_NLA_PUBL_LOWER, publ->lower)) 3250 goto attr_msg_cancel; 3251 if (nla_put_u32(skb, TIPC_NLA_PUBL_UPPER, publ->upper)) 3252 goto attr_msg_cancel; 3253 3254 nla_nest_end(skb, attrs); 3255 genlmsg_end(skb, hdr); 3256 3257 return 0; 3258 3259 attr_msg_cancel: 3260 nla_nest_cancel(skb, attrs); 3261 genlmsg_cancel: 3262 genlmsg_cancel(skb, hdr); 3263 msg_cancel: 3264 return -EMSGSIZE; 3265 } 3266 3267 /* Caller should hold socket lock for the passed tipc socket. */ 3268 static int __tipc_nl_list_sk_publ(struct sk_buff *skb, 3269 struct netlink_callback *cb, 3270 struct tipc_sock *tsk, u32 *last_publ) 3271 { 3272 int err; 3273 struct publication *p; 3274 3275 if (*last_publ) { 3276 list_for_each_entry(p, &tsk->publications, pport_list) { 3277 if (p->key == *last_publ) 3278 break; 3279 } 3280 if (p->key != *last_publ) { 3281 /* We never set seq or call nl_dump_check_consistent() 3282 * this means that setting prev_seq here will cause the 3283 * consistence check to fail in the netlink callback 3284 * handler. Resulting in the last NLMSG_DONE message 3285 * having the NLM_F_DUMP_INTR flag set. 3286 */ 3287 cb->prev_seq = 1; 3288 *last_publ = 0; 3289 return -EPIPE; 3290 } 3291 } else { 3292 p = list_first_entry(&tsk->publications, struct publication, 3293 pport_list); 3294 } 3295 3296 list_for_each_entry_from(p, &tsk->publications, pport_list) { 3297 err = __tipc_nl_add_sk_publ(skb, cb, p); 3298 if (err) { 3299 *last_publ = p->key; 3300 return err; 3301 } 3302 } 3303 *last_publ = 0; 3304 3305 return 0; 3306 } 3307 3308 int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb) 3309 { 3310 int err; 3311 u32 tsk_portid = cb->args[0]; 3312 u32 last_publ = cb->args[1]; 3313 u32 done = cb->args[2]; 3314 struct net *net = sock_net(skb->sk); 3315 struct tipc_sock *tsk; 3316 3317 if (!tsk_portid) { 3318 struct nlattr **attrs; 3319 struct nlattr *sock[TIPC_NLA_SOCK_MAX + 1]; 3320 3321 err = tipc_nlmsg_parse(cb->nlh, &attrs); 3322 if (err) 3323 return err; 3324 3325 if (!attrs[TIPC_NLA_SOCK]) 3326 return -EINVAL; 3327 3328 err = nla_parse_nested(sock, TIPC_NLA_SOCK_MAX, 3329 attrs[TIPC_NLA_SOCK], 3330 tipc_nl_sock_policy, NULL); 3331 if (err) 3332 return err; 3333 3334 if (!sock[TIPC_NLA_SOCK_REF]) 3335 return -EINVAL; 3336 3337 tsk_portid = nla_get_u32(sock[TIPC_NLA_SOCK_REF]); 3338 } 3339 3340 if (done) 3341 return 0; 3342 3343 tsk = tipc_sk_lookup(net, tsk_portid); 3344 if (!tsk) 3345 return -EINVAL; 3346 3347 lock_sock(&tsk->sk); 3348 err = __tipc_nl_list_sk_publ(skb, cb, tsk, &last_publ); 3349 if (!err) 3350 done = 1; 3351 release_sock(&tsk->sk); 3352 sock_put(&tsk->sk); 3353 3354 cb->args[0] = tsk_portid; 3355 cb->args[1] = last_publ; 3356 cb->args[2] = done; 3357 3358 return skb->len; 3359 } 3360