1 /* 2 * net/tipc/socket.c: TIPC socket API 3 * 4 * Copyright (c) 2001-2007, 2012-2019, Ericsson AB 5 * Copyright (c) 2004-2008, 2010-2013, Wind River Systems 6 * Copyright (c) 2020-2021, Red Hat Inc 7 * All rights reserved. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions are met: 11 * 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 3. Neither the names of the copyright holders nor the names of its 18 * contributors may be used to endorse or promote products derived from 19 * this software without specific prior written permission. 20 * 21 * Alternatively, this software may be distributed under the terms of the 22 * GNU General Public License ("GPL") version 2 as published by the Free 23 * Software Foundation. 24 * 25 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" 26 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 27 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 28 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE 29 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 30 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 31 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 32 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 33 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 34 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 35 * POSSIBILITY OF SUCH DAMAGE. 36 */ 37 38 #include <linux/rhashtable.h> 39 #include <linux/sched/signal.h> 40 #include <linux/uio.h> 41 #include <trace/events/sock.h> 42 43 #include "core.h" 44 #include "name_table.h" 45 #include "node.h" 46 #include "link.h" 47 #include "name_distr.h" 48 #include "socket.h" 49 #include "bcast.h" 50 #include "netlink.h" 51 #include "group.h" 52 #include "trace.h" 53 54 #define NAGLE_START_INIT 4 55 #define NAGLE_START_MAX 1024 56 #define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */ 57 #define CONN_PROBING_INTV msecs_to_jiffies(3600000) /* [ms] => 1 h */ 58 #define TIPC_MAX_PORT 0xffffffff 59 #define TIPC_MIN_PORT 1 60 #define TIPC_ACK_RATE 4 /* ACK at 1/4 of rcv window size */ 61 62 enum { 63 TIPC_LISTEN = TCP_LISTEN, 64 TIPC_ESTABLISHED = TCP_ESTABLISHED, 65 TIPC_OPEN = TCP_CLOSE, 66 TIPC_DISCONNECTING = TCP_CLOSE_WAIT, 67 TIPC_CONNECTING = TCP_SYN_SENT, 68 }; 69 70 struct sockaddr_pair { 71 struct sockaddr_tipc sock; 72 struct sockaddr_tipc member; 73 }; 74 75 /** 76 * struct tipc_sock - TIPC socket structure 77 * @sk: socket - interacts with 'port' and with user via the socket API 78 * @max_pkt: maximum packet size "hint" used when building messages sent by port 79 * @maxnagle: maximum size of msg which can be subject to nagle 80 * @portid: unique port identity in TIPC socket hash table 81 * @phdr: preformatted message header used when sending messages 82 * @cong_links: list of congested links 83 * @publications: list of publications for port 84 * @pub_count: total # of publications port has made during its lifetime 85 * @conn_timeout: the time we can wait for an unresponded setup request 86 * @probe_unacked: probe has not received ack yet 87 * @dupl_rcvcnt: number of bytes counted twice, in both backlog and rcv queue 88 * @cong_link_cnt: number of congested links 89 * @snt_unacked: # messages sent by socket, and not yet acked by peer 90 * @snd_win: send window size 91 * @peer_caps: peer capabilities mask 92 * @rcv_unacked: # messages read by user, but not yet acked back to peer 93 * @rcv_win: receive window size 94 * @peer: 'connected' peer for dgram/rdm 95 * @node: hash table node 96 * @mc_method: cookie for use between socket and broadcast layer 97 * @rcu: rcu struct for tipc_sock 98 * @group: TIPC communications group 99 * @oneway: message count in one direction (FIXME) 100 * @nagle_start: current nagle value 101 * @snd_backlog: send backlog count 102 * @msg_acc: messages accepted; used in managing backlog and nagle 103 * @pkt_cnt: TIPC socket packet count 104 * @expect_ack: whether this TIPC socket is expecting an ack 105 * @nodelay: setsockopt() TIPC_NODELAY setting 106 * @group_is_open: TIPC socket group is fully open (FIXME) 107 * @published: true if port has one or more associated names 108 * @conn_addrtype: address type used when establishing connection 109 */ 110 struct tipc_sock { 111 struct sock sk; 112 u32 max_pkt; 113 u32 maxnagle; 114 u32 portid; 115 struct tipc_msg phdr; 116 struct list_head cong_links; 117 struct list_head publications; 118 u32 pub_count; 119 atomic_t dupl_rcvcnt; 120 u16 conn_timeout; 121 bool probe_unacked; 122 u16 cong_link_cnt; 123 u16 snt_unacked; 124 u16 snd_win; 125 u16 peer_caps; 126 u16 rcv_unacked; 127 u16 rcv_win; 128 struct sockaddr_tipc peer; 129 struct rhash_head node; 130 struct tipc_mc_method mc_method; 131 struct rcu_head rcu; 132 struct tipc_group *group; 133 u32 oneway; 134 u32 nagle_start; 135 u16 snd_backlog; 136 u16 msg_acc; 137 u16 pkt_cnt; 138 bool expect_ack; 139 bool nodelay; 140 bool group_is_open; 141 bool published; 142 u8 conn_addrtype; 143 }; 144 145 static int tipc_sk_backlog_rcv(struct sock *sk, struct sk_buff *skb); 146 static void tipc_data_ready(struct sock *sk); 147 static void tipc_write_space(struct sock *sk); 148 static void tipc_sock_destruct(struct sock *sk); 149 static int tipc_release(struct socket *sock); 150 static void tipc_sk_timeout(struct timer_list *t); 151 static int tipc_sk_publish(struct tipc_sock *tsk, struct tipc_uaddr *ua); 152 static int tipc_sk_withdraw(struct tipc_sock *tsk, struct tipc_uaddr *ua); 153 static int tipc_sk_leave(struct tipc_sock *tsk); 154 static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid); 155 static int tipc_sk_insert(struct tipc_sock *tsk); 156 static void tipc_sk_remove(struct tipc_sock *tsk); 157 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz); 158 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz); 159 static void tipc_sk_push_backlog(struct tipc_sock *tsk, bool nagle_ack); 160 static int tipc_wait_for_connect(struct socket *sock, long *timeo_p); 161 162 static const struct proto_ops packet_ops; 163 static const struct proto_ops stream_ops; 164 static const struct proto_ops msg_ops; 165 static struct proto tipc_proto; 166 static const struct rhashtable_params tsk_rht_params; 167 168 static u32 tsk_own_node(struct tipc_sock *tsk) 169 { 170 return msg_prevnode(&tsk->phdr); 171 } 172 173 static u32 tsk_peer_node(struct tipc_sock *tsk) 174 { 175 return msg_destnode(&tsk->phdr); 176 } 177 178 static u32 tsk_peer_port(struct tipc_sock *tsk) 179 { 180 return msg_destport(&tsk->phdr); 181 } 182 183 static bool tsk_unreliable(struct tipc_sock *tsk) 184 { 185 return msg_src_droppable(&tsk->phdr) != 0; 186 } 187 188 static void tsk_set_unreliable(struct tipc_sock *tsk, bool unreliable) 189 { 190 msg_set_src_droppable(&tsk->phdr, unreliable ? 1 : 0); 191 } 192 193 static bool tsk_unreturnable(struct tipc_sock *tsk) 194 { 195 return msg_dest_droppable(&tsk->phdr) != 0; 196 } 197 198 static void tsk_set_unreturnable(struct tipc_sock *tsk, bool unreturnable) 199 { 200 msg_set_dest_droppable(&tsk->phdr, unreturnable ? 1 : 0); 201 } 202 203 static int tsk_importance(struct tipc_sock *tsk) 204 { 205 return msg_importance(&tsk->phdr); 206 } 207 208 static struct tipc_sock *tipc_sk(const struct sock *sk) 209 { 210 return container_of(sk, struct tipc_sock, sk); 211 } 212 213 int tsk_set_importance(struct sock *sk, int imp) 214 { 215 if (imp > TIPC_CRITICAL_IMPORTANCE) 216 return -EINVAL; 217 msg_set_importance(&tipc_sk(sk)->phdr, (u32)imp); 218 return 0; 219 } 220 221 static bool tsk_conn_cong(struct tipc_sock *tsk) 222 { 223 return tsk->snt_unacked > tsk->snd_win; 224 } 225 226 static u16 tsk_blocks(int len) 227 { 228 return ((len / FLOWCTL_BLK_SZ) + 1); 229 } 230 231 /* tsk_blocks(): translate a buffer size in bytes to number of 232 * advertisable blocks, taking into account the ratio truesize(len)/len 233 * We can trust that this ratio is always < 4 for len >= FLOWCTL_BLK_SZ 234 */ 235 static u16 tsk_adv_blocks(int len) 236 { 237 return len / FLOWCTL_BLK_SZ / 4; 238 } 239 240 /* tsk_inc(): increment counter for sent or received data 241 * - If block based flow control is not supported by peer we 242 * fall back to message based ditto, incrementing the counter 243 */ 244 static u16 tsk_inc(struct tipc_sock *tsk, int msglen) 245 { 246 if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL)) 247 return ((msglen / FLOWCTL_BLK_SZ) + 1); 248 return 1; 249 } 250 251 /* tsk_set_nagle - enable/disable nagle property by manipulating maxnagle 252 */ 253 static void tsk_set_nagle(struct tipc_sock *tsk) 254 { 255 struct sock *sk = &tsk->sk; 256 257 tsk->maxnagle = 0; 258 if (sk->sk_type != SOCK_STREAM) 259 return; 260 if (tsk->nodelay) 261 return; 262 if (!(tsk->peer_caps & TIPC_NAGLE)) 263 return; 264 /* Limit node local buffer size to avoid receive queue overflow */ 265 if (tsk->max_pkt == MAX_MSG_SIZE) 266 tsk->maxnagle = 1500; 267 else 268 tsk->maxnagle = tsk->max_pkt; 269 } 270 271 /** 272 * tsk_advance_rx_queue - discard first buffer in socket receive queue 273 * @sk: network socket 274 * 275 * Caller must hold socket lock 276 */ 277 static void tsk_advance_rx_queue(struct sock *sk) 278 { 279 trace_tipc_sk_advance_rx(sk, NULL, TIPC_DUMP_SK_RCVQ, " "); 280 kfree_skb(__skb_dequeue(&sk->sk_receive_queue)); 281 } 282 283 /* tipc_sk_respond() : send response message back to sender 284 */ 285 static void tipc_sk_respond(struct sock *sk, struct sk_buff *skb, int err) 286 { 287 u32 selector; 288 u32 dnode; 289 u32 onode = tipc_own_addr(sock_net(sk)); 290 291 if (!tipc_msg_reverse(onode, &skb, err)) 292 return; 293 294 trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_NONE, "@sk_respond!"); 295 dnode = msg_destnode(buf_msg(skb)); 296 selector = msg_origport(buf_msg(skb)); 297 tipc_node_xmit_skb(sock_net(sk), skb, dnode, selector); 298 } 299 300 /** 301 * tsk_rej_rx_queue - reject all buffers in socket receive queue 302 * @sk: network socket 303 * @error: response error code 304 * 305 * Caller must hold socket lock 306 */ 307 static void tsk_rej_rx_queue(struct sock *sk, int error) 308 { 309 struct sk_buff *skb; 310 311 while ((skb = __skb_dequeue(&sk->sk_receive_queue))) 312 tipc_sk_respond(sk, skb, error); 313 } 314 315 static bool tipc_sk_connected(const struct sock *sk) 316 { 317 return READ_ONCE(sk->sk_state) == TIPC_ESTABLISHED; 318 } 319 320 /* tipc_sk_type_connectionless - check if the socket is datagram socket 321 * @sk: socket 322 * 323 * Returns true if connection less, false otherwise 324 */ 325 static bool tipc_sk_type_connectionless(struct sock *sk) 326 { 327 return sk->sk_type == SOCK_RDM || sk->sk_type == SOCK_DGRAM; 328 } 329 330 /* tsk_peer_msg - verify if message was sent by connected port's peer 331 * 332 * Handles cases where the node's network address has changed from 333 * the default of <0.0.0> to its configured setting. 334 */ 335 static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg) 336 { 337 struct sock *sk = &tsk->sk; 338 u32 self = tipc_own_addr(sock_net(sk)); 339 u32 peer_port = tsk_peer_port(tsk); 340 u32 orig_node, peer_node; 341 342 if (unlikely(!tipc_sk_connected(sk))) 343 return false; 344 345 if (unlikely(msg_origport(msg) != peer_port)) 346 return false; 347 348 orig_node = msg_orignode(msg); 349 peer_node = tsk_peer_node(tsk); 350 351 if (likely(orig_node == peer_node)) 352 return true; 353 354 if (!orig_node && peer_node == self) 355 return true; 356 357 if (!peer_node && orig_node == self) 358 return true; 359 360 return false; 361 } 362 363 /* tipc_set_sk_state - set the sk_state of the socket 364 * @sk: socket 365 * 366 * Caller must hold socket lock 367 * 368 * Returns 0 on success, errno otherwise 369 */ 370 static int tipc_set_sk_state(struct sock *sk, int state) 371 { 372 int oldsk_state = sk->sk_state; 373 int res = -EINVAL; 374 375 switch (state) { 376 case TIPC_OPEN: 377 res = 0; 378 break; 379 case TIPC_LISTEN: 380 case TIPC_CONNECTING: 381 if (oldsk_state == TIPC_OPEN) 382 res = 0; 383 break; 384 case TIPC_ESTABLISHED: 385 if (oldsk_state == TIPC_CONNECTING || 386 oldsk_state == TIPC_OPEN) 387 res = 0; 388 break; 389 case TIPC_DISCONNECTING: 390 if (oldsk_state == TIPC_CONNECTING || 391 oldsk_state == TIPC_ESTABLISHED) 392 res = 0; 393 break; 394 } 395 396 if (!res) 397 sk->sk_state = state; 398 399 return res; 400 } 401 402 static int tipc_sk_sock_err(struct socket *sock, long *timeout) 403 { 404 struct sock *sk = sock->sk; 405 int err = sock_error(sk); 406 int typ = sock->type; 407 408 if (err) 409 return err; 410 if (typ == SOCK_STREAM || typ == SOCK_SEQPACKET) { 411 if (sk->sk_state == TIPC_DISCONNECTING) 412 return -EPIPE; 413 else if (!tipc_sk_connected(sk)) 414 return -ENOTCONN; 415 } 416 if (!*timeout) 417 return -EAGAIN; 418 if (signal_pending(current)) 419 return sock_intr_errno(*timeout); 420 421 return 0; 422 } 423 424 #define tipc_wait_for_cond(sock_, timeo_, condition_) \ 425 ({ \ 426 DEFINE_WAIT_FUNC(wait_, woken_wake_function); \ 427 struct sock *sk_; \ 428 int rc_; \ 429 \ 430 while ((rc_ = !(condition_))) { \ 431 /* coupled with smp_wmb() in tipc_sk_proto_rcv() */ \ 432 smp_rmb(); \ 433 sk_ = (sock_)->sk; \ 434 rc_ = tipc_sk_sock_err((sock_), timeo_); \ 435 if (rc_) \ 436 break; \ 437 add_wait_queue(sk_sleep(sk_), &wait_); \ 438 release_sock(sk_); \ 439 *(timeo_) = wait_woken(&wait_, TASK_INTERRUPTIBLE, *(timeo_)); \ 440 sched_annotate_sleep(); \ 441 lock_sock(sk_); \ 442 remove_wait_queue(sk_sleep(sk_), &wait_); \ 443 } \ 444 rc_; \ 445 }) 446 447 /** 448 * tipc_sk_create - create a TIPC socket 449 * @net: network namespace (must be default network) 450 * @sock: pre-allocated socket structure 451 * @protocol: protocol indicator (must be 0) 452 * @kern: caused by kernel or by userspace? 453 * 454 * This routine creates additional data structures used by the TIPC socket, 455 * initializes them, and links them together. 456 * 457 * Return: 0 on success, errno otherwise 458 */ 459 static int tipc_sk_create(struct net *net, struct socket *sock, 460 int protocol, int kern) 461 { 462 const struct proto_ops *ops; 463 struct sock *sk; 464 struct tipc_sock *tsk; 465 struct tipc_msg *msg; 466 467 /* Validate arguments */ 468 if (unlikely(protocol != 0)) 469 return -EPROTONOSUPPORT; 470 471 switch (sock->type) { 472 case SOCK_STREAM: 473 ops = &stream_ops; 474 break; 475 case SOCK_SEQPACKET: 476 ops = &packet_ops; 477 break; 478 case SOCK_DGRAM: 479 case SOCK_RDM: 480 ops = &msg_ops; 481 break; 482 default: 483 return -EPROTOTYPE; 484 } 485 486 /* Allocate socket's protocol area */ 487 sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto, kern); 488 if (sk == NULL) 489 return -ENOMEM; 490 491 tsk = tipc_sk(sk); 492 tsk->max_pkt = MAX_PKT_DEFAULT; 493 tsk->maxnagle = 0; 494 tsk->nagle_start = NAGLE_START_INIT; 495 INIT_LIST_HEAD(&tsk->publications); 496 INIT_LIST_HEAD(&tsk->cong_links); 497 msg = &tsk->phdr; 498 499 /* Finish initializing socket data structures */ 500 sock->ops = ops; 501 sock_init_data(sock, sk); 502 tipc_set_sk_state(sk, TIPC_OPEN); 503 if (tipc_sk_insert(tsk)) { 504 sk_free(sk); 505 pr_warn("Socket create failed; port number exhausted\n"); 506 return -EINVAL; 507 } 508 509 /* Ensure tsk is visible before we read own_addr. */ 510 smp_mb(); 511 512 tipc_msg_init(tipc_own_addr(net), msg, TIPC_LOW_IMPORTANCE, 513 TIPC_NAMED_MSG, NAMED_H_SIZE, 0); 514 515 msg_set_origport(msg, tsk->portid); 516 timer_setup(&sk->sk_timer, tipc_sk_timeout, 0); 517 sk->sk_shutdown = 0; 518 sk->sk_backlog_rcv = tipc_sk_backlog_rcv; 519 sk->sk_rcvbuf = READ_ONCE(sysctl_tipc_rmem[1]); 520 sk->sk_data_ready = tipc_data_ready; 521 sk->sk_write_space = tipc_write_space; 522 sk->sk_destruct = tipc_sock_destruct; 523 tsk->conn_timeout = CONN_TIMEOUT_DEFAULT; 524 tsk->group_is_open = true; 525 atomic_set(&tsk->dupl_rcvcnt, 0); 526 527 /* Start out with safe limits until we receive an advertised window */ 528 tsk->snd_win = tsk_adv_blocks(RCVBUF_MIN); 529 tsk->rcv_win = tsk->snd_win; 530 531 if (tipc_sk_type_connectionless(sk)) { 532 tsk_set_unreturnable(tsk, true); 533 if (sock->type == SOCK_DGRAM) 534 tsk_set_unreliable(tsk, true); 535 } 536 __skb_queue_head_init(&tsk->mc_method.deferredq); 537 trace_tipc_sk_create(sk, NULL, TIPC_DUMP_NONE, " "); 538 return 0; 539 } 540 541 static void tipc_sk_callback(struct rcu_head *head) 542 { 543 struct tipc_sock *tsk = container_of(head, struct tipc_sock, rcu); 544 545 sock_put(&tsk->sk); 546 } 547 548 /* Caller should hold socket lock for the socket. */ 549 static void __tipc_shutdown(struct socket *sock, int error) 550 { 551 struct sock *sk = sock->sk; 552 struct tipc_sock *tsk = tipc_sk(sk); 553 struct net *net = sock_net(sk); 554 long timeout = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT); 555 u32 dnode = tsk_peer_node(tsk); 556 struct sk_buff *skb; 557 558 /* Avoid that hi-prio shutdown msgs bypass msgs in link wakeup queue */ 559 tipc_wait_for_cond(sock, &timeout, (!tsk->cong_link_cnt && 560 !tsk_conn_cong(tsk))); 561 562 /* Push out delayed messages if in Nagle mode */ 563 tipc_sk_push_backlog(tsk, false); 564 /* Remove pending SYN */ 565 __skb_queue_purge(&sk->sk_write_queue); 566 567 /* Remove partially received buffer if any */ 568 skb = skb_peek(&sk->sk_receive_queue); 569 if (skb && TIPC_SKB_CB(skb)->bytes_read) { 570 __skb_unlink(skb, &sk->sk_receive_queue); 571 kfree_skb(skb); 572 } 573 574 /* Reject all unreceived messages if connectionless */ 575 if (tipc_sk_type_connectionless(sk)) { 576 tsk_rej_rx_queue(sk, error); 577 return; 578 } 579 580 switch (sk->sk_state) { 581 case TIPC_CONNECTING: 582 case TIPC_ESTABLISHED: 583 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 584 tipc_node_remove_conn(net, dnode, tsk->portid); 585 /* Send a FIN+/- to its peer */ 586 skb = __skb_dequeue(&sk->sk_receive_queue); 587 if (skb) { 588 __skb_queue_purge(&sk->sk_receive_queue); 589 tipc_sk_respond(sk, skb, error); 590 break; 591 } 592 skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, 593 TIPC_CONN_MSG, SHORT_H_SIZE, 0, dnode, 594 tsk_own_node(tsk), tsk_peer_port(tsk), 595 tsk->portid, error); 596 if (skb) 597 tipc_node_xmit_skb(net, skb, dnode, tsk->portid); 598 break; 599 case TIPC_LISTEN: 600 /* Reject all SYN messages */ 601 tsk_rej_rx_queue(sk, error); 602 break; 603 default: 604 __skb_queue_purge(&sk->sk_receive_queue); 605 break; 606 } 607 } 608 609 /** 610 * tipc_release - destroy a TIPC socket 611 * @sock: socket to destroy 612 * 613 * This routine cleans up any messages that are still queued on the socket. 614 * For DGRAM and RDM socket types, all queued messages are rejected. 615 * For SEQPACKET and STREAM socket types, the first message is rejected 616 * and any others are discarded. (If the first message on a STREAM socket 617 * is partially-read, it is discarded and the next one is rejected instead.) 618 * 619 * NOTE: Rejected messages are not necessarily returned to the sender! They 620 * are returned or discarded according to the "destination droppable" setting 621 * specified for the message by the sender. 622 * 623 * Return: 0 on success, errno otherwise 624 */ 625 static int tipc_release(struct socket *sock) 626 { 627 struct sock *sk = sock->sk; 628 struct tipc_sock *tsk; 629 630 /* 631 * Exit if socket isn't fully initialized (occurs when a failed accept() 632 * releases a pre-allocated child socket that was never used) 633 */ 634 if (sk == NULL) 635 return 0; 636 637 tsk = tipc_sk(sk); 638 lock_sock(sk); 639 640 trace_tipc_sk_release(sk, NULL, TIPC_DUMP_ALL, " "); 641 __tipc_shutdown(sock, TIPC_ERR_NO_PORT); 642 sk->sk_shutdown = SHUTDOWN_MASK; 643 tipc_sk_leave(tsk); 644 tipc_sk_withdraw(tsk, NULL); 645 __skb_queue_purge(&tsk->mc_method.deferredq); 646 sk_stop_timer(sk, &sk->sk_timer); 647 tipc_sk_remove(tsk); 648 649 sock_orphan(sk); 650 /* Reject any messages that accumulated in backlog queue */ 651 release_sock(sk); 652 tipc_dest_list_purge(&tsk->cong_links); 653 tsk->cong_link_cnt = 0; 654 call_rcu(&tsk->rcu, tipc_sk_callback); 655 sock->sk = NULL; 656 657 return 0; 658 } 659 660 /** 661 * __tipc_bind - associate or disassociate TIPC name(s) with a socket 662 * @sock: socket structure 663 * @skaddr: socket address describing name(s) and desired operation 664 * @alen: size of socket address data structure 665 * 666 * Name and name sequence binding are indicated using a positive scope value; 667 * a negative scope value unbinds the specified name. Specifying no name 668 * (i.e. a socket address length of 0) unbinds all names from the socket. 669 * 670 * Return: 0 on success, errno otherwise 671 * 672 * NOTE: This routine doesn't need to take the socket lock since it doesn't 673 * access any non-constant socket information. 674 */ 675 static int __tipc_bind(struct socket *sock, struct sockaddr *skaddr, int alen) 676 { 677 struct tipc_uaddr *ua = (struct tipc_uaddr *)skaddr; 678 struct tipc_sock *tsk = tipc_sk(sock->sk); 679 bool unbind = false; 680 681 if (unlikely(!alen)) 682 return tipc_sk_withdraw(tsk, NULL); 683 684 if (ua->addrtype == TIPC_SERVICE_ADDR) { 685 ua->addrtype = TIPC_SERVICE_RANGE; 686 ua->sr.upper = ua->sr.lower; 687 } 688 if (ua->scope < 0) { 689 unbind = true; 690 ua->scope = -ua->scope; 691 } 692 /* Users may still use deprecated TIPC_ZONE_SCOPE */ 693 if (ua->scope != TIPC_NODE_SCOPE) 694 ua->scope = TIPC_CLUSTER_SCOPE; 695 696 if (tsk->group) 697 return -EACCES; 698 699 if (unbind) 700 return tipc_sk_withdraw(tsk, ua); 701 return tipc_sk_publish(tsk, ua); 702 } 703 704 int tipc_sk_bind(struct socket *sock, struct sockaddr *skaddr, int alen) 705 { 706 int res; 707 708 lock_sock(sock->sk); 709 res = __tipc_bind(sock, skaddr, alen); 710 release_sock(sock->sk); 711 return res; 712 } 713 714 static int tipc_bind(struct socket *sock, struct sockaddr_unsized *skaddr, int alen) 715 { 716 struct tipc_uaddr *ua = (struct tipc_uaddr *)skaddr; 717 u32 atype = ua->addrtype; 718 719 if (alen) { 720 if (!tipc_uaddr_valid(ua, alen)) 721 return -EINVAL; 722 if (atype == TIPC_SOCKET_ADDR) 723 return -EAFNOSUPPORT; 724 if (ua->sr.type < TIPC_RESERVED_TYPES) { 725 pr_warn_once("Can't bind to reserved service type %u\n", 726 ua->sr.type); 727 return -EACCES; 728 } 729 } 730 return tipc_sk_bind(sock, (struct sockaddr *)skaddr, alen); 731 } 732 733 /** 734 * tipc_getname - get port ID of socket or peer socket 735 * @sock: socket structure 736 * @uaddr: area for returned socket address 737 * @peer: 0 = own ID, 1 = current peer ID, 2 = current/former peer ID 738 * 739 * Return: 0 on success, errno otherwise 740 * 741 * NOTE: This routine doesn't need to take the socket lock since it only 742 * accesses socket information that is unchanging (or which changes in 743 * a completely predictable manner). 744 */ 745 static int tipc_getname(struct socket *sock, struct sockaddr *uaddr, 746 int peer) 747 { 748 struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; 749 struct sock *sk = sock->sk; 750 struct tipc_sock *tsk = tipc_sk(sk); 751 752 memset(addr, 0, sizeof(*addr)); 753 if (peer) { 754 if ((!tipc_sk_connected(sk)) && 755 ((peer != 2) || (sk->sk_state != TIPC_DISCONNECTING))) 756 return -ENOTCONN; 757 addr->addr.id.ref = tsk_peer_port(tsk); 758 addr->addr.id.node = tsk_peer_node(tsk); 759 } else { 760 addr->addr.id.ref = tsk->portid; 761 addr->addr.id.node = tipc_own_addr(sock_net(sk)); 762 } 763 764 addr->addrtype = TIPC_SOCKET_ADDR; 765 addr->family = AF_TIPC; 766 addr->scope = 0; 767 addr->addr.name.domain = 0; 768 769 return sizeof(*addr); 770 } 771 772 /** 773 * tipc_poll - read and possibly block on pollmask 774 * @file: file structure associated with the socket 775 * @sock: socket for which to calculate the poll bits 776 * @wait: ??? 777 * 778 * Return: pollmask value 779 * 780 * COMMENTARY: 781 * It appears that the usual socket locking mechanisms are not useful here 782 * since the pollmask info is potentially out-of-date the moment this routine 783 * exits. TCP and other protocols seem to rely on higher level poll routines 784 * to handle any preventable race conditions, so TIPC will do the same ... 785 * 786 * IMPORTANT: The fact that a read or write operation is indicated does NOT 787 * imply that the operation will succeed, merely that it should be performed 788 * and will not block. 789 */ 790 static __poll_t tipc_poll(struct file *file, struct socket *sock, 791 poll_table *wait) 792 { 793 struct sock *sk = sock->sk; 794 struct tipc_sock *tsk = tipc_sk(sk); 795 __poll_t revents = 0; 796 797 sock_poll_wait(file, sock, wait); 798 trace_tipc_sk_poll(sk, NULL, TIPC_DUMP_ALL, " "); 799 800 if (sk->sk_shutdown & RCV_SHUTDOWN) 801 revents |= EPOLLRDHUP | EPOLLIN | EPOLLRDNORM; 802 if (sk->sk_shutdown == SHUTDOWN_MASK) 803 revents |= EPOLLHUP; 804 805 switch (sk->sk_state) { 806 case TIPC_ESTABLISHED: 807 if (!tsk->cong_link_cnt && !tsk_conn_cong(tsk)) 808 revents |= EPOLLOUT; 809 fallthrough; 810 case TIPC_LISTEN: 811 case TIPC_CONNECTING: 812 if (!skb_queue_empty_lockless(&sk->sk_receive_queue)) 813 revents |= EPOLLIN | EPOLLRDNORM; 814 break; 815 case TIPC_OPEN: 816 if (tsk->group_is_open && !tsk->cong_link_cnt) 817 revents |= EPOLLOUT; 818 if (!tipc_sk_type_connectionless(sk)) 819 break; 820 if (skb_queue_empty_lockless(&sk->sk_receive_queue)) 821 break; 822 revents |= EPOLLIN | EPOLLRDNORM; 823 break; 824 case TIPC_DISCONNECTING: 825 revents = EPOLLIN | EPOLLRDNORM | EPOLLHUP; 826 break; 827 } 828 return revents; 829 } 830 831 /** 832 * tipc_sendmcast - send multicast message 833 * @sock: socket structure 834 * @ua: destination address struct 835 * @msg: message to send 836 * @dlen: length of data to send 837 * @timeout: timeout to wait for wakeup 838 * 839 * Called from function tipc_sendmsg(), which has done all sanity checks 840 * Return: the number of bytes sent on success, or errno 841 */ 842 static int tipc_sendmcast(struct socket *sock, struct tipc_uaddr *ua, 843 struct msghdr *msg, size_t dlen, long timeout) 844 { 845 struct sock *sk = sock->sk; 846 struct tipc_sock *tsk = tipc_sk(sk); 847 struct tipc_msg *hdr = &tsk->phdr; 848 struct net *net = sock_net(sk); 849 int mtu = tipc_bcast_get_mtu(net); 850 struct sk_buff_head pkts; 851 struct tipc_nlist dsts; 852 int rc; 853 854 if (tsk->group) 855 return -EACCES; 856 857 /* Block or return if any destination link is congested */ 858 rc = tipc_wait_for_cond(sock, &timeout, !tsk->cong_link_cnt); 859 if (unlikely(rc)) 860 return rc; 861 862 /* Lookup destination nodes */ 863 tipc_nlist_init(&dsts, tipc_own_addr(net)); 864 tipc_nametbl_lookup_mcast_nodes(net, ua, &dsts); 865 if (!dsts.local && !dsts.remote) 866 return -EHOSTUNREACH; 867 868 /* Build message header */ 869 msg_set_type(hdr, TIPC_MCAST_MSG); 870 msg_set_hdr_sz(hdr, MCAST_H_SIZE); 871 msg_set_lookup_scope(hdr, TIPC_CLUSTER_SCOPE); 872 msg_set_destport(hdr, 0); 873 msg_set_destnode(hdr, 0); 874 msg_set_nametype(hdr, ua->sr.type); 875 msg_set_namelower(hdr, ua->sr.lower); 876 msg_set_nameupper(hdr, ua->sr.upper); 877 878 /* Build message as chain of buffers */ 879 __skb_queue_head_init(&pkts); 880 rc = tipc_msg_build(hdr, msg, 0, dlen, mtu, &pkts); 881 882 /* Send message if build was successful */ 883 if (unlikely(rc == dlen)) { 884 trace_tipc_sk_sendmcast(sk, skb_peek(&pkts), 885 TIPC_DUMP_SK_SNDQ, " "); 886 rc = tipc_mcast_xmit(net, &pkts, &tsk->mc_method, &dsts, 887 &tsk->cong_link_cnt); 888 } 889 890 tipc_nlist_purge(&dsts); 891 892 return rc ? rc : dlen; 893 } 894 895 /** 896 * tipc_send_group_msg - send a message to a member in the group 897 * @net: network namespace 898 * @tsk: tipc socket 899 * @m: message to send 900 * @mb: group member 901 * @dnode: destination node 902 * @dport: destination port 903 * @dlen: total length of message data 904 */ 905 static int tipc_send_group_msg(struct net *net, struct tipc_sock *tsk, 906 struct msghdr *m, struct tipc_member *mb, 907 u32 dnode, u32 dport, int dlen) 908 { 909 u16 bc_snd_nxt = tipc_group_bc_snd_nxt(tsk->group); 910 struct tipc_mc_method *method = &tsk->mc_method; 911 int blks = tsk_blocks(GROUP_H_SIZE + dlen); 912 struct tipc_msg *hdr = &tsk->phdr; 913 struct sk_buff_head pkts; 914 int mtu, rc; 915 916 /* Complete message header */ 917 msg_set_type(hdr, TIPC_GRP_UCAST_MSG); 918 msg_set_hdr_sz(hdr, GROUP_H_SIZE); 919 msg_set_destport(hdr, dport); 920 msg_set_destnode(hdr, dnode); 921 msg_set_grp_bc_seqno(hdr, bc_snd_nxt); 922 923 /* Build message as chain of buffers */ 924 __skb_queue_head_init(&pkts); 925 mtu = tipc_node_get_mtu(net, dnode, tsk->portid, false); 926 rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); 927 if (unlikely(rc != dlen)) 928 return rc; 929 930 /* Send message */ 931 rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid); 932 if (unlikely(rc == -ELINKCONG)) { 933 tipc_dest_push(&tsk->cong_links, dnode, 0); 934 tsk->cong_link_cnt++; 935 } 936 937 /* Update send window */ 938 tipc_group_update_member(mb, blks); 939 940 /* A broadcast sent within next EXPIRE period must follow same path */ 941 method->rcast = true; 942 method->mandatory = true; 943 return dlen; 944 } 945 946 /** 947 * tipc_send_group_unicast - send message to a member in the group 948 * @sock: socket structure 949 * @m: message to send 950 * @dlen: total length of message data 951 * @timeout: timeout to wait for wakeup 952 * 953 * Called from function tipc_sendmsg(), which has done all sanity checks 954 * Return: the number of bytes sent on success, or errno 955 */ 956 static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m, 957 int dlen, long timeout) 958 { 959 struct sock *sk = sock->sk; 960 struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name; 961 int blks = tsk_blocks(GROUP_H_SIZE + dlen); 962 struct tipc_sock *tsk = tipc_sk(sk); 963 struct net *net = sock_net(sk); 964 struct tipc_member *mb = NULL; 965 u32 node, port; 966 int rc; 967 968 node = ua->sk.node; 969 port = ua->sk.ref; 970 if (!port && !node) 971 return -EHOSTUNREACH; 972 973 /* Block or return if destination link or member is congested */ 974 rc = tipc_wait_for_cond(sock, &timeout, 975 !tipc_dest_find(&tsk->cong_links, node, 0) && 976 tsk->group && 977 !tipc_group_cong(tsk->group, node, port, blks, 978 &mb)); 979 if (unlikely(rc)) 980 return rc; 981 982 if (unlikely(!mb)) 983 return -EHOSTUNREACH; 984 985 rc = tipc_send_group_msg(net, tsk, m, mb, node, port, dlen); 986 987 return rc ? rc : dlen; 988 } 989 990 /** 991 * tipc_send_group_anycast - send message to any member with given identity 992 * @sock: socket structure 993 * @m: message to send 994 * @dlen: total length of message data 995 * @timeout: timeout to wait for wakeup 996 * 997 * Called from function tipc_sendmsg(), which has done all sanity checks 998 * Return: the number of bytes sent on success, or errno 999 */ 1000 static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m, 1001 int dlen, long timeout) 1002 { 1003 struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name; 1004 struct sock *sk = sock->sk; 1005 struct tipc_sock *tsk = tipc_sk(sk); 1006 struct list_head *cong_links = &tsk->cong_links; 1007 int blks = tsk_blocks(GROUP_H_SIZE + dlen); 1008 struct tipc_msg *hdr = &tsk->phdr; 1009 struct tipc_member *first = NULL; 1010 struct tipc_member *mbr = NULL; 1011 struct net *net = sock_net(sk); 1012 u32 node, port, exclude; 1013 LIST_HEAD(dsts); 1014 int lookups = 0; 1015 int dstcnt, rc; 1016 bool cong; 1017 1018 ua->sa.type = msg_nametype(hdr); 1019 ua->scope = msg_lookup_scope(hdr); 1020 1021 while (++lookups < 4) { 1022 exclude = tipc_group_exclude(tsk->group); 1023 1024 first = NULL; 1025 1026 /* Look for a non-congested destination member, if any */ 1027 while (1) { 1028 if (!tipc_nametbl_lookup_group(net, ua, &dsts, &dstcnt, 1029 exclude, false)) 1030 return -EHOSTUNREACH; 1031 tipc_dest_pop(&dsts, &node, &port); 1032 cong = tipc_group_cong(tsk->group, node, port, blks, 1033 &mbr); 1034 if (!cong) 1035 break; 1036 if (mbr == first) 1037 break; 1038 if (!first) 1039 first = mbr; 1040 } 1041 1042 /* Start over if destination was not in member list */ 1043 if (unlikely(!mbr)) 1044 continue; 1045 1046 if (likely(!cong && !tipc_dest_find(cong_links, node, 0))) 1047 break; 1048 1049 /* Block or return if destination link or member is congested */ 1050 rc = tipc_wait_for_cond(sock, &timeout, 1051 !tipc_dest_find(cong_links, node, 0) && 1052 tsk->group && 1053 !tipc_group_cong(tsk->group, node, port, 1054 blks, &mbr)); 1055 if (unlikely(rc)) 1056 return rc; 1057 1058 /* Send, unless destination disappeared while waiting */ 1059 if (likely(mbr)) 1060 break; 1061 } 1062 1063 if (unlikely(lookups >= 4)) 1064 return -EHOSTUNREACH; 1065 1066 rc = tipc_send_group_msg(net, tsk, m, mbr, node, port, dlen); 1067 1068 return rc ? rc : dlen; 1069 } 1070 1071 /** 1072 * tipc_send_group_bcast - send message to all members in communication group 1073 * @sock: socket structure 1074 * @m: message to send 1075 * @dlen: total length of message data 1076 * @timeout: timeout to wait for wakeup 1077 * 1078 * Called from function tipc_sendmsg(), which has done all sanity checks 1079 * Return: the number of bytes sent on success, or errno 1080 */ 1081 static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m, 1082 int dlen, long timeout) 1083 { 1084 struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name; 1085 struct sock *sk = sock->sk; 1086 struct net *net = sock_net(sk); 1087 struct tipc_sock *tsk = tipc_sk(sk); 1088 struct tipc_nlist *dsts; 1089 struct tipc_mc_method *method = &tsk->mc_method; 1090 bool ack = method->mandatory && method->rcast; 1091 int blks = tsk_blocks(MCAST_H_SIZE + dlen); 1092 struct tipc_msg *hdr = &tsk->phdr; 1093 int mtu = tipc_bcast_get_mtu(net); 1094 struct sk_buff_head pkts; 1095 int rc = -EHOSTUNREACH; 1096 1097 /* Block or return if any destination link or member is congested */ 1098 rc = tipc_wait_for_cond(sock, &timeout, 1099 !tsk->cong_link_cnt && tsk->group && 1100 !tipc_group_bc_cong(tsk->group, blks)); 1101 if (unlikely(rc)) 1102 return rc; 1103 1104 dsts = tipc_group_dests(tsk->group); 1105 if (!dsts->local && !dsts->remote) 1106 return -EHOSTUNREACH; 1107 1108 /* Complete message header */ 1109 if (ua) { 1110 msg_set_type(hdr, TIPC_GRP_MCAST_MSG); 1111 msg_set_nameinst(hdr, ua->sa.instance); 1112 } else { 1113 msg_set_type(hdr, TIPC_GRP_BCAST_MSG); 1114 msg_set_nameinst(hdr, 0); 1115 } 1116 msg_set_hdr_sz(hdr, GROUP_H_SIZE); 1117 msg_set_destport(hdr, 0); 1118 msg_set_destnode(hdr, 0); 1119 msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(tsk->group)); 1120 1121 /* Avoid getting stuck with repeated forced replicasts */ 1122 msg_set_grp_bc_ack_req(hdr, ack); 1123 1124 /* Build message as chain of buffers */ 1125 __skb_queue_head_init(&pkts); 1126 rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); 1127 if (unlikely(rc != dlen)) 1128 return rc; 1129 1130 /* Send message */ 1131 rc = tipc_mcast_xmit(net, &pkts, method, dsts, &tsk->cong_link_cnt); 1132 if (unlikely(rc)) 1133 return rc; 1134 1135 /* Update broadcast sequence number and send windows */ 1136 tipc_group_update_bc_members(tsk->group, blks, ack); 1137 1138 /* Broadcast link is now free to choose method for next broadcast */ 1139 method->mandatory = false; 1140 method->expires = jiffies; 1141 1142 return dlen; 1143 } 1144 1145 /** 1146 * tipc_send_group_mcast - send message to all members with given identity 1147 * @sock: socket structure 1148 * @m: message to send 1149 * @dlen: total length of message data 1150 * @timeout: timeout to wait for wakeup 1151 * 1152 * Called from function tipc_sendmsg(), which has done all sanity checks 1153 * Return: the number of bytes sent on success, or errno 1154 */ 1155 static int tipc_send_group_mcast(struct socket *sock, struct msghdr *m, 1156 int dlen, long timeout) 1157 { 1158 struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name; 1159 struct sock *sk = sock->sk; 1160 struct tipc_sock *tsk = tipc_sk(sk); 1161 struct tipc_group *grp = tsk->group; 1162 struct tipc_msg *hdr = &tsk->phdr; 1163 struct net *net = sock_net(sk); 1164 u32 dstcnt, exclude; 1165 LIST_HEAD(dsts); 1166 1167 ua->sa.type = msg_nametype(hdr); 1168 ua->scope = msg_lookup_scope(hdr); 1169 exclude = tipc_group_exclude(grp); 1170 1171 if (!tipc_nametbl_lookup_group(net, ua, &dsts, &dstcnt, exclude, true)) 1172 return -EHOSTUNREACH; 1173 1174 if (dstcnt == 1) { 1175 tipc_dest_pop(&dsts, &ua->sk.node, &ua->sk.ref); 1176 return tipc_send_group_unicast(sock, m, dlen, timeout); 1177 } 1178 1179 tipc_dest_list_purge(&dsts); 1180 return tipc_send_group_bcast(sock, m, dlen, timeout); 1181 } 1182 1183 /** 1184 * tipc_sk_mcast_rcv - Deliver multicast messages to all destination sockets 1185 * @net: the associated network namespace 1186 * @arrvq: queue with arriving messages, to be cloned after destination lookup 1187 * @inputq: queue with cloned messages, delivered to socket after dest lookup 1188 * 1189 * Multi-threaded: parallel calls with reference to same queues may occur 1190 */ 1191 void tipc_sk_mcast_rcv(struct net *net, struct sk_buff_head *arrvq, 1192 struct sk_buff_head *inputq) 1193 { 1194 u32 self = tipc_own_addr(net); 1195 struct sk_buff *skb, *_skb; 1196 u32 portid, onode; 1197 struct sk_buff_head tmpq; 1198 struct list_head dports; 1199 struct tipc_msg *hdr; 1200 struct tipc_uaddr ua; 1201 int user, mtyp, hlen; 1202 1203 __skb_queue_head_init(&tmpq); 1204 INIT_LIST_HEAD(&dports); 1205 ua.addrtype = TIPC_SERVICE_RANGE; 1206 1207 /* tipc_skb_peek() increments the head skb's reference counter */ 1208 skb = tipc_skb_peek(arrvq, &inputq->lock); 1209 for (; skb; skb = tipc_skb_peek(arrvq, &inputq->lock)) { 1210 hdr = buf_msg(skb); 1211 user = msg_user(hdr); 1212 mtyp = msg_type(hdr); 1213 hlen = skb_headroom(skb) + msg_hdr_sz(hdr); 1214 onode = msg_orignode(hdr); 1215 ua.sr.type = msg_nametype(hdr); 1216 ua.sr.lower = msg_namelower(hdr); 1217 ua.sr.upper = msg_nameupper(hdr); 1218 if (onode == self) 1219 ua.scope = TIPC_ANY_SCOPE; 1220 else 1221 ua.scope = TIPC_CLUSTER_SCOPE; 1222 1223 if (mtyp == TIPC_GRP_UCAST_MSG || user == GROUP_PROTOCOL) { 1224 spin_lock_bh(&inputq->lock); 1225 if (skb_peek(arrvq) == skb) { 1226 __skb_dequeue(arrvq); 1227 __skb_queue_tail(inputq, skb); 1228 } 1229 kfree_skb(skb); 1230 spin_unlock_bh(&inputq->lock); 1231 continue; 1232 } 1233 1234 /* Group messages require exact scope match */ 1235 if (msg_in_group(hdr)) { 1236 ua.sr.lower = 0; 1237 ua.sr.upper = ~0; 1238 ua.scope = msg_lookup_scope(hdr); 1239 } 1240 1241 /* Create destination port list: */ 1242 tipc_nametbl_lookup_mcast_sockets(net, &ua, &dports); 1243 1244 /* Clone message per destination */ 1245 while (tipc_dest_pop(&dports, NULL, &portid)) { 1246 _skb = __pskb_copy(skb, hlen, GFP_ATOMIC); 1247 if (_skb) { 1248 msg_set_destport(buf_msg(_skb), portid); 1249 __skb_queue_tail(&tmpq, _skb); 1250 continue; 1251 } 1252 pr_warn("Failed to clone mcast rcv buffer\n"); 1253 } 1254 /* Append clones to inputq only if skb is still head of arrvq */ 1255 spin_lock_bh(&inputq->lock); 1256 if (skb_peek(arrvq) == skb) { 1257 skb_queue_splice_tail_init(&tmpq, inputq); 1258 /* Decrement the skb's refcnt */ 1259 kfree_skb(__skb_dequeue(arrvq)); 1260 } 1261 spin_unlock_bh(&inputq->lock); 1262 __skb_queue_purge(&tmpq); 1263 kfree_skb(skb); 1264 } 1265 tipc_sk_rcv(net, inputq); 1266 } 1267 1268 /* tipc_sk_push_backlog(): send accumulated buffers in socket write queue 1269 * when socket is in Nagle mode 1270 */ 1271 static void tipc_sk_push_backlog(struct tipc_sock *tsk, bool nagle_ack) 1272 { 1273 struct sk_buff_head *txq = &tsk->sk.sk_write_queue; 1274 struct sk_buff *skb = skb_peek_tail(txq); 1275 struct net *net = sock_net(&tsk->sk); 1276 u32 dnode = tsk_peer_node(tsk); 1277 int rc; 1278 1279 if (nagle_ack) { 1280 tsk->pkt_cnt += skb_queue_len(txq); 1281 if (!tsk->pkt_cnt || tsk->msg_acc / tsk->pkt_cnt < 2) { 1282 tsk->oneway = 0; 1283 if (tsk->nagle_start < NAGLE_START_MAX) 1284 tsk->nagle_start *= 2; 1285 tsk->expect_ack = false; 1286 pr_debug("tsk %10u: bad nagle %u -> %u, next start %u!\n", 1287 tsk->portid, tsk->msg_acc, tsk->pkt_cnt, 1288 tsk->nagle_start); 1289 } else { 1290 tsk->nagle_start = NAGLE_START_INIT; 1291 if (skb) { 1292 msg_set_ack_required(buf_msg(skb)); 1293 tsk->expect_ack = true; 1294 } else { 1295 tsk->expect_ack = false; 1296 } 1297 } 1298 tsk->msg_acc = 0; 1299 tsk->pkt_cnt = 0; 1300 } 1301 1302 if (!skb || tsk->cong_link_cnt) 1303 return; 1304 1305 /* Do not send SYN again after congestion */ 1306 if (msg_is_syn(buf_msg(skb))) 1307 return; 1308 1309 if (tsk->msg_acc) 1310 tsk->pkt_cnt += skb_queue_len(txq); 1311 tsk->snt_unacked += tsk->snd_backlog; 1312 tsk->snd_backlog = 0; 1313 rc = tipc_node_xmit(net, txq, dnode, tsk->portid); 1314 if (rc == -ELINKCONG) 1315 tsk->cong_link_cnt = 1; 1316 } 1317 1318 /** 1319 * tipc_sk_conn_proto_rcv - receive a connection mng protocol message 1320 * @tsk: receiving socket 1321 * @skb: pointer to message buffer. 1322 * @inputq: buffer list containing the buffers 1323 * @xmitq: output message area 1324 */ 1325 static void tipc_sk_conn_proto_rcv(struct tipc_sock *tsk, struct sk_buff *skb, 1326 struct sk_buff_head *inputq, 1327 struct sk_buff_head *xmitq) 1328 { 1329 struct tipc_msg *hdr = buf_msg(skb); 1330 u32 onode = tsk_own_node(tsk); 1331 struct sock *sk = &tsk->sk; 1332 int mtyp = msg_type(hdr); 1333 bool was_cong; 1334 1335 /* Ignore if connection cannot be validated: */ 1336 if (!tsk_peer_msg(tsk, hdr)) { 1337 trace_tipc_sk_drop_msg(sk, skb, TIPC_DUMP_NONE, "@proto_rcv!"); 1338 goto exit; 1339 } 1340 1341 if (unlikely(msg_errcode(hdr))) { 1342 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 1343 tipc_node_remove_conn(sock_net(sk), tsk_peer_node(tsk), 1344 tsk_peer_port(tsk)); 1345 sk->sk_state_change(sk); 1346 1347 /* State change is ignored if socket already awake, 1348 * - convert msg to abort msg and add to inqueue 1349 */ 1350 msg_set_user(hdr, TIPC_CRITICAL_IMPORTANCE); 1351 msg_set_type(hdr, TIPC_CONN_MSG); 1352 msg_set_size(hdr, BASIC_H_SIZE); 1353 msg_set_hdr_sz(hdr, BASIC_H_SIZE); 1354 __skb_queue_tail(inputq, skb); 1355 return; 1356 } 1357 1358 tsk->probe_unacked = false; 1359 1360 if (mtyp == CONN_PROBE) { 1361 msg_set_type(hdr, CONN_PROBE_REPLY); 1362 if (tipc_msg_reverse(onode, &skb, TIPC_OK)) 1363 __skb_queue_tail(xmitq, skb); 1364 return; 1365 } else if (mtyp == CONN_ACK) { 1366 if (tsk->snt_unacked < msg_conn_ack(hdr)) 1367 goto exit; 1368 1369 was_cong = tsk_conn_cong(tsk); 1370 tipc_sk_push_backlog(tsk, msg_nagle_ack(hdr)); 1371 tsk->snt_unacked -= msg_conn_ack(hdr); 1372 if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) 1373 tsk->snd_win = msg_adv_win(hdr); 1374 if (was_cong && !tsk_conn_cong(tsk)) 1375 sk->sk_write_space(sk); 1376 } else if (mtyp != CONN_PROBE_REPLY) { 1377 pr_warn("Received unknown CONN_PROTO msg\n"); 1378 } 1379 exit: 1380 kfree_skb(skb); 1381 } 1382 1383 /** 1384 * tipc_sendmsg - send message in connectionless manner 1385 * @sock: socket structure 1386 * @m: message to send 1387 * @dsz: amount of user data to be sent 1388 * 1389 * Message must have an destination specified explicitly. 1390 * Used for SOCK_RDM and SOCK_DGRAM messages, 1391 * and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections. 1392 * (Note: 'SYN+' is prohibited on SOCK_STREAM.) 1393 * 1394 * Return: the number of bytes sent on success, or errno otherwise 1395 */ 1396 static int tipc_sendmsg(struct socket *sock, 1397 struct msghdr *m, size_t dsz) 1398 { 1399 struct sock *sk = sock->sk; 1400 int ret; 1401 1402 lock_sock(sk); 1403 ret = __tipc_sendmsg(sock, m, dsz); 1404 release_sock(sk); 1405 1406 return ret; 1407 } 1408 1409 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dlen) 1410 { 1411 struct sock *sk = sock->sk; 1412 struct net *net = sock_net(sk); 1413 struct tipc_sock *tsk = tipc_sk(sk); 1414 struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name; 1415 long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); 1416 struct list_head *clinks = &tsk->cong_links; 1417 bool syn = !tipc_sk_type_connectionless(sk); 1418 struct tipc_group *grp = tsk->group; 1419 struct tipc_msg *hdr = &tsk->phdr; 1420 struct tipc_socket_addr skaddr; 1421 struct sk_buff_head pkts; 1422 int atype, mtu, rc; 1423 1424 if (unlikely(dlen > TIPC_MAX_USER_MSG_SIZE)) 1425 return -EMSGSIZE; 1426 1427 if (ua) { 1428 if (!tipc_uaddr_valid(ua, m->msg_namelen)) 1429 return -EINVAL; 1430 atype = ua->addrtype; 1431 } 1432 1433 /* If socket belongs to a communication group follow other paths */ 1434 if (grp) { 1435 if (!ua) 1436 return tipc_send_group_bcast(sock, m, dlen, timeout); 1437 if (atype == TIPC_SERVICE_ADDR) 1438 return tipc_send_group_anycast(sock, m, dlen, timeout); 1439 if (atype == TIPC_SOCKET_ADDR) 1440 return tipc_send_group_unicast(sock, m, dlen, timeout); 1441 if (atype == TIPC_SERVICE_RANGE) 1442 return tipc_send_group_mcast(sock, m, dlen, timeout); 1443 return -EINVAL; 1444 } 1445 1446 if (!ua) { 1447 ua = (struct tipc_uaddr *)&tsk->peer; 1448 if (!syn && ua->family != AF_TIPC) 1449 return -EDESTADDRREQ; 1450 atype = ua->addrtype; 1451 } 1452 1453 if (unlikely(syn)) { 1454 if (sk->sk_state == TIPC_LISTEN) 1455 return -EPIPE; 1456 if (sk->sk_state != TIPC_OPEN) 1457 return -EISCONN; 1458 if (tsk->published) 1459 return -EOPNOTSUPP; 1460 if (atype == TIPC_SERVICE_ADDR) 1461 tsk->conn_addrtype = atype; 1462 msg_set_syn(hdr, 1); 1463 } 1464 1465 memset(&skaddr, 0, sizeof(skaddr)); 1466 1467 /* Determine destination */ 1468 if (atype == TIPC_SERVICE_RANGE) { 1469 return tipc_sendmcast(sock, ua, m, dlen, timeout); 1470 } else if (atype == TIPC_SERVICE_ADDR) { 1471 skaddr.node = ua->lookup_node; 1472 ua->scope = tipc_node2scope(skaddr.node); 1473 if (!tipc_nametbl_lookup_anycast(net, ua, &skaddr)) 1474 return -EHOSTUNREACH; 1475 } else if (atype == TIPC_SOCKET_ADDR) { 1476 skaddr = ua->sk; 1477 } else { 1478 return -EINVAL; 1479 } 1480 1481 /* Block or return if destination link is congested */ 1482 rc = tipc_wait_for_cond(sock, &timeout, 1483 !tipc_dest_find(clinks, skaddr.node, 0)); 1484 if (unlikely(rc)) 1485 return rc; 1486 1487 /* Finally build message header */ 1488 msg_set_destnode(hdr, skaddr.node); 1489 msg_set_destport(hdr, skaddr.ref); 1490 if (atype == TIPC_SERVICE_ADDR) { 1491 msg_set_type(hdr, TIPC_NAMED_MSG); 1492 msg_set_hdr_sz(hdr, NAMED_H_SIZE); 1493 msg_set_nametype(hdr, ua->sa.type); 1494 msg_set_nameinst(hdr, ua->sa.instance); 1495 msg_set_lookup_scope(hdr, ua->scope); 1496 } else { /* TIPC_SOCKET_ADDR */ 1497 msg_set_type(hdr, TIPC_DIRECT_MSG); 1498 msg_set_lookup_scope(hdr, 0); 1499 msg_set_hdr_sz(hdr, BASIC_H_SIZE); 1500 } 1501 1502 /* Add message body */ 1503 __skb_queue_head_init(&pkts); 1504 mtu = tipc_node_get_mtu(net, skaddr.node, tsk->portid, true); 1505 rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); 1506 if (unlikely(rc != dlen)) 1507 return rc; 1508 if (unlikely(syn && !tipc_msg_skb_clone(&pkts, &sk->sk_write_queue))) { 1509 __skb_queue_purge(&pkts); 1510 return -ENOMEM; 1511 } 1512 1513 /* Send message */ 1514 trace_tipc_sk_sendmsg(sk, skb_peek(&pkts), TIPC_DUMP_SK_SNDQ, " "); 1515 rc = tipc_node_xmit(net, &pkts, skaddr.node, tsk->portid); 1516 if (unlikely(rc == -ELINKCONG)) { 1517 tipc_dest_push(clinks, skaddr.node, 0); 1518 tsk->cong_link_cnt++; 1519 rc = 0; 1520 } 1521 1522 if (unlikely(syn && !rc)) { 1523 tipc_set_sk_state(sk, TIPC_CONNECTING); 1524 if (dlen && timeout) { 1525 timeout = msecs_to_jiffies(timeout); 1526 tipc_wait_for_connect(sock, &timeout); 1527 } 1528 } 1529 1530 return rc ? rc : dlen; 1531 } 1532 1533 /** 1534 * tipc_sendstream - send stream-oriented data 1535 * @sock: socket structure 1536 * @m: data to send 1537 * @dsz: total length of data to be transmitted 1538 * 1539 * Used for SOCK_STREAM data. 1540 * 1541 * Return: the number of bytes sent on success (or partial success), 1542 * or errno if no data sent 1543 */ 1544 static int tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz) 1545 { 1546 struct sock *sk = sock->sk; 1547 int ret; 1548 1549 lock_sock(sk); 1550 ret = __tipc_sendstream(sock, m, dsz); 1551 release_sock(sk); 1552 1553 return ret; 1554 } 1555 1556 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dlen) 1557 { 1558 struct sock *sk = sock->sk; 1559 DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); 1560 long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); 1561 struct sk_buff_head *txq = &sk->sk_write_queue; 1562 struct tipc_sock *tsk = tipc_sk(sk); 1563 struct tipc_msg *hdr = &tsk->phdr; 1564 struct net *net = sock_net(sk); 1565 struct sk_buff *skb; 1566 u32 dnode = tsk_peer_node(tsk); 1567 int maxnagle = tsk->maxnagle; 1568 int maxpkt = tsk->max_pkt; 1569 int send, sent = 0; 1570 int blocks, rc = 0; 1571 1572 if (unlikely(dlen > INT_MAX)) 1573 return -EMSGSIZE; 1574 1575 /* Handle implicit connection setup */ 1576 if (unlikely(dest && sk->sk_state == TIPC_OPEN)) { 1577 rc = __tipc_sendmsg(sock, m, dlen); 1578 if (dlen && dlen == rc) { 1579 tsk->peer_caps = tipc_node_get_capabilities(net, dnode); 1580 tsk->snt_unacked = tsk_inc(tsk, dlen + msg_hdr_sz(hdr)); 1581 } 1582 return rc; 1583 } 1584 1585 do { 1586 rc = tipc_wait_for_cond(sock, &timeout, 1587 (!tsk->cong_link_cnt && 1588 !tsk_conn_cong(tsk) && 1589 tipc_sk_connected(sk))); 1590 if (unlikely(rc)) 1591 break; 1592 send = min_t(size_t, dlen - sent, TIPC_MAX_USER_MSG_SIZE); 1593 blocks = tsk->snd_backlog; 1594 if (tsk->oneway++ >= tsk->nagle_start && maxnagle && 1595 send <= maxnagle) { 1596 rc = tipc_msg_append(hdr, m, send, maxnagle, txq); 1597 if (unlikely(rc < 0)) 1598 break; 1599 blocks += rc; 1600 tsk->msg_acc++; 1601 if (blocks <= 64 && tsk->expect_ack) { 1602 tsk->snd_backlog = blocks; 1603 sent += send; 1604 break; 1605 } else if (blocks > 64) { 1606 tsk->pkt_cnt += skb_queue_len(txq); 1607 } else { 1608 skb = skb_peek_tail(txq); 1609 if (skb) { 1610 msg_set_ack_required(buf_msg(skb)); 1611 tsk->expect_ack = true; 1612 } else { 1613 tsk->expect_ack = false; 1614 } 1615 tsk->msg_acc = 0; 1616 tsk->pkt_cnt = 0; 1617 } 1618 } else { 1619 rc = tipc_msg_build(hdr, m, sent, send, maxpkt, txq); 1620 if (unlikely(rc != send)) 1621 break; 1622 blocks += tsk_inc(tsk, send + MIN_H_SIZE); 1623 } 1624 trace_tipc_sk_sendstream(sk, skb_peek(txq), 1625 TIPC_DUMP_SK_SNDQ, " "); 1626 rc = tipc_node_xmit(net, txq, dnode, tsk->portid); 1627 if (unlikely(rc == -ELINKCONG)) { 1628 tsk->cong_link_cnt = 1; 1629 rc = 0; 1630 } 1631 if (likely(!rc)) { 1632 tsk->snt_unacked += blocks; 1633 tsk->snd_backlog = 0; 1634 sent += send; 1635 } 1636 } while (sent < dlen && !rc); 1637 1638 return sent ? sent : rc; 1639 } 1640 1641 /** 1642 * tipc_send_packet - send a connection-oriented message 1643 * @sock: socket structure 1644 * @m: message to send 1645 * @dsz: length of data to be transmitted 1646 * 1647 * Used for SOCK_SEQPACKET messages. 1648 * 1649 * Return: the number of bytes sent on success, or errno otherwise 1650 */ 1651 static int tipc_send_packet(struct socket *sock, struct msghdr *m, size_t dsz) 1652 { 1653 if (dsz > TIPC_MAX_USER_MSG_SIZE) 1654 return -EMSGSIZE; 1655 1656 return tipc_sendstream(sock, m, dsz); 1657 } 1658 1659 /* tipc_sk_finish_conn - complete the setup of a connection 1660 */ 1661 static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port, 1662 u32 peer_node) 1663 { 1664 struct sock *sk = &tsk->sk; 1665 struct net *net = sock_net(sk); 1666 struct tipc_msg *msg = &tsk->phdr; 1667 1668 msg_set_syn(msg, 0); 1669 msg_set_destnode(msg, peer_node); 1670 msg_set_destport(msg, peer_port); 1671 msg_set_type(msg, TIPC_CONN_MSG); 1672 msg_set_lookup_scope(msg, 0); 1673 msg_set_hdr_sz(msg, SHORT_H_SIZE); 1674 1675 sk_reset_timer(sk, &sk->sk_timer, jiffies + CONN_PROBING_INTV); 1676 tipc_set_sk_state(sk, TIPC_ESTABLISHED); 1677 tipc_node_add_conn(net, peer_node, tsk->portid, peer_port); 1678 tsk->max_pkt = tipc_node_get_mtu(net, peer_node, tsk->portid, true); 1679 tsk->peer_caps = tipc_node_get_capabilities(net, peer_node); 1680 tsk_set_nagle(tsk); 1681 __skb_queue_purge(&sk->sk_write_queue); 1682 if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) 1683 return; 1684 1685 /* Fall back to message based flow control */ 1686 tsk->rcv_win = FLOWCTL_MSG_WIN; 1687 tsk->snd_win = FLOWCTL_MSG_WIN; 1688 } 1689 1690 /** 1691 * tipc_sk_set_orig_addr - capture sender's address for received message 1692 * @m: descriptor for message info 1693 * @skb: received message 1694 * 1695 * Note: Address is not captured if not requested by receiver. 1696 */ 1697 static void tipc_sk_set_orig_addr(struct msghdr *m, struct sk_buff *skb) 1698 { 1699 DECLARE_SOCKADDR(struct sockaddr_pair *, srcaddr, m->msg_name); 1700 struct tipc_msg *hdr = buf_msg(skb); 1701 1702 if (!srcaddr) 1703 return; 1704 1705 srcaddr->sock.family = AF_TIPC; 1706 srcaddr->sock.addrtype = TIPC_SOCKET_ADDR; 1707 srcaddr->sock.scope = 0; 1708 srcaddr->sock.addr.id.ref = msg_origport(hdr); 1709 srcaddr->sock.addr.id.node = msg_orignode(hdr); 1710 srcaddr->sock.addr.name.domain = 0; 1711 m->msg_namelen = sizeof(struct sockaddr_tipc); 1712 1713 if (!msg_in_group(hdr)) 1714 return; 1715 1716 /* Group message users may also want to know sending member's id */ 1717 srcaddr->member.family = AF_TIPC; 1718 srcaddr->member.addrtype = TIPC_SERVICE_ADDR; 1719 srcaddr->member.scope = 0; 1720 srcaddr->member.addr.name.name.type = msg_nametype(hdr); 1721 srcaddr->member.addr.name.name.instance = TIPC_SKB_CB(skb)->orig_member; 1722 srcaddr->member.addr.name.domain = 0; 1723 m->msg_namelen = sizeof(*srcaddr); 1724 } 1725 1726 /** 1727 * tipc_sk_anc_data_recv - optionally capture ancillary data for received message 1728 * @m: descriptor for message info 1729 * @skb: received message buffer 1730 * @tsk: TIPC port associated with message 1731 * 1732 * Note: Ancillary data is not captured if not requested by receiver. 1733 * 1734 * Return: 0 if successful, otherwise errno 1735 */ 1736 static int tipc_sk_anc_data_recv(struct msghdr *m, struct sk_buff *skb, 1737 struct tipc_sock *tsk) 1738 { 1739 struct tipc_msg *hdr; 1740 u32 data[3] = {0,}; 1741 bool has_addr; 1742 int dlen, rc; 1743 1744 if (likely(m->msg_controllen == 0)) 1745 return 0; 1746 1747 hdr = buf_msg(skb); 1748 dlen = msg_data_sz(hdr); 1749 1750 /* Capture errored message object, if any */ 1751 if (msg_errcode(hdr)) { 1752 if (skb_linearize(skb)) 1753 return -ENOMEM; 1754 hdr = buf_msg(skb); 1755 data[0] = msg_errcode(hdr); 1756 data[1] = dlen; 1757 rc = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, data); 1758 if (rc || !dlen) 1759 return rc; 1760 rc = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, dlen, msg_data(hdr)); 1761 if (rc) 1762 return rc; 1763 } 1764 1765 /* Capture TIPC_SERVICE_ADDR/RANGE destination address, if any */ 1766 switch (msg_type(hdr)) { 1767 case TIPC_NAMED_MSG: 1768 has_addr = true; 1769 data[0] = msg_nametype(hdr); 1770 data[1] = msg_namelower(hdr); 1771 data[2] = data[1]; 1772 break; 1773 case TIPC_MCAST_MSG: 1774 has_addr = true; 1775 data[0] = msg_nametype(hdr); 1776 data[1] = msg_namelower(hdr); 1777 data[2] = msg_nameupper(hdr); 1778 break; 1779 case TIPC_CONN_MSG: 1780 has_addr = !!tsk->conn_addrtype; 1781 data[0] = msg_nametype(&tsk->phdr); 1782 data[1] = msg_nameinst(&tsk->phdr); 1783 data[2] = data[1]; 1784 break; 1785 default: 1786 has_addr = false; 1787 } 1788 if (!has_addr) 1789 return 0; 1790 return put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, data); 1791 } 1792 1793 static struct sk_buff *tipc_sk_build_ack(struct tipc_sock *tsk) 1794 { 1795 struct sock *sk = &tsk->sk; 1796 struct sk_buff *skb = NULL; 1797 struct tipc_msg *msg; 1798 u32 peer_port = tsk_peer_port(tsk); 1799 u32 dnode = tsk_peer_node(tsk); 1800 1801 if (!tipc_sk_connected(sk)) 1802 return NULL; 1803 skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, 1804 dnode, tsk_own_node(tsk), peer_port, 1805 tsk->portid, TIPC_OK); 1806 if (!skb) 1807 return NULL; 1808 msg = buf_msg(skb); 1809 msg_set_conn_ack(msg, tsk->rcv_unacked); 1810 tsk->rcv_unacked = 0; 1811 1812 /* Adjust to and advertize the correct window limit */ 1813 if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) { 1814 tsk->rcv_win = tsk_adv_blocks(tsk->sk.sk_rcvbuf); 1815 msg_set_adv_win(msg, tsk->rcv_win); 1816 } 1817 return skb; 1818 } 1819 1820 static void tipc_sk_send_ack(struct tipc_sock *tsk) 1821 { 1822 struct sk_buff *skb; 1823 1824 skb = tipc_sk_build_ack(tsk); 1825 if (!skb) 1826 return; 1827 1828 tipc_node_xmit_skb(sock_net(&tsk->sk), skb, tsk_peer_node(tsk), 1829 msg_link_selector(buf_msg(skb))); 1830 } 1831 1832 static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) 1833 { 1834 struct sock *sk = sock->sk; 1835 DEFINE_WAIT_FUNC(wait, woken_wake_function); 1836 long timeo = *timeop; 1837 int err = sock_error(sk); 1838 1839 if (err) 1840 return err; 1841 1842 for (;;) { 1843 if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { 1844 if (sk->sk_shutdown & RCV_SHUTDOWN) { 1845 err = -ENOTCONN; 1846 break; 1847 } 1848 add_wait_queue(sk_sleep(sk), &wait); 1849 release_sock(sk); 1850 timeo = wait_woken(&wait, TASK_INTERRUPTIBLE, timeo); 1851 sched_annotate_sleep(); 1852 lock_sock(sk); 1853 remove_wait_queue(sk_sleep(sk), &wait); 1854 } 1855 err = 0; 1856 if (!skb_queue_empty(&sk->sk_receive_queue)) 1857 break; 1858 err = -EAGAIN; 1859 if (!timeo) 1860 break; 1861 err = sock_intr_errno(timeo); 1862 if (signal_pending(current)) 1863 break; 1864 1865 err = sock_error(sk); 1866 if (err) 1867 break; 1868 } 1869 *timeop = timeo; 1870 return err; 1871 } 1872 1873 /** 1874 * tipc_recvmsg - receive packet-oriented message 1875 * @sock: network socket 1876 * @m: descriptor for message info 1877 * @buflen: length of user buffer area 1878 * @flags: receive flags 1879 * 1880 * Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages. 1881 * If the complete message doesn't fit in user area, truncate it. 1882 * 1883 * Return: size of returned message data, errno otherwise 1884 */ 1885 static int tipc_recvmsg(struct socket *sock, struct msghdr *m, 1886 size_t buflen, int flags) 1887 { 1888 struct sock *sk = sock->sk; 1889 bool connected = !tipc_sk_type_connectionless(sk); 1890 struct tipc_sock *tsk = tipc_sk(sk); 1891 int rc, err, hlen, dlen, copy; 1892 struct tipc_skb_cb *skb_cb; 1893 struct sk_buff_head xmitq; 1894 struct tipc_msg *hdr; 1895 struct sk_buff *skb; 1896 bool grp_evt; 1897 long timeout; 1898 1899 /* Catch invalid receive requests */ 1900 if (unlikely(!buflen)) 1901 return -EINVAL; 1902 1903 lock_sock(sk); 1904 if (unlikely(connected && sk->sk_state == TIPC_OPEN)) { 1905 rc = -ENOTCONN; 1906 goto exit; 1907 } 1908 timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); 1909 1910 /* Step rcv queue to first msg with data or error; wait if necessary */ 1911 do { 1912 rc = tipc_wait_for_rcvmsg(sock, &timeout); 1913 if (unlikely(rc)) 1914 goto exit; 1915 skb = skb_peek(&sk->sk_receive_queue); 1916 skb_cb = TIPC_SKB_CB(skb); 1917 hdr = buf_msg(skb); 1918 dlen = msg_data_sz(hdr); 1919 hlen = msg_hdr_sz(hdr); 1920 err = msg_errcode(hdr); 1921 grp_evt = msg_is_grp_evt(hdr); 1922 if (likely(dlen || err)) 1923 break; 1924 tsk_advance_rx_queue(sk); 1925 } while (1); 1926 1927 /* Collect msg meta data, including error code and rejected data */ 1928 tipc_sk_set_orig_addr(m, skb); 1929 rc = tipc_sk_anc_data_recv(m, skb, tsk); 1930 if (unlikely(rc)) 1931 goto exit; 1932 hdr = buf_msg(skb); 1933 1934 /* Capture data if non-error msg, otherwise just set return value */ 1935 if (likely(!err)) { 1936 int offset = skb_cb->bytes_read; 1937 1938 copy = min_t(int, dlen - offset, buflen); 1939 rc = skb_copy_datagram_msg(skb, hlen + offset, m, copy); 1940 if (unlikely(rc)) 1941 goto exit; 1942 if (unlikely(offset + copy < dlen)) { 1943 if (flags & MSG_EOR) { 1944 if (!(flags & MSG_PEEK)) 1945 skb_cb->bytes_read = offset + copy; 1946 } else { 1947 m->msg_flags |= MSG_TRUNC; 1948 skb_cb->bytes_read = 0; 1949 } 1950 } else { 1951 if (flags & MSG_EOR) 1952 m->msg_flags |= MSG_EOR; 1953 skb_cb->bytes_read = 0; 1954 } 1955 } else { 1956 copy = 0; 1957 rc = 0; 1958 if (err != TIPC_CONN_SHUTDOWN && connected && !m->msg_control) { 1959 rc = -ECONNRESET; 1960 goto exit; 1961 } 1962 } 1963 1964 /* Mark message as group event if applicable */ 1965 if (unlikely(grp_evt)) { 1966 if (msg_grp_evt(hdr) == TIPC_WITHDRAWN) 1967 m->msg_flags |= MSG_EOR; 1968 m->msg_flags |= MSG_OOB; 1969 copy = 0; 1970 } 1971 1972 /* Caption of data or error code/rejected data was successful */ 1973 if (unlikely(flags & MSG_PEEK)) 1974 goto exit; 1975 1976 /* Send group flow control advertisement when applicable */ 1977 if (tsk->group && msg_in_group(hdr) && !grp_evt) { 1978 __skb_queue_head_init(&xmitq); 1979 tipc_group_update_rcv_win(tsk->group, tsk_blocks(hlen + dlen), 1980 msg_orignode(hdr), msg_origport(hdr), 1981 &xmitq); 1982 tipc_node_distr_xmit(sock_net(sk), &xmitq); 1983 } 1984 1985 if (skb_cb->bytes_read) 1986 goto exit; 1987 1988 tsk_advance_rx_queue(sk); 1989 1990 if (likely(!connected)) 1991 goto exit; 1992 1993 /* Send connection flow control advertisement when applicable */ 1994 tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen); 1995 if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE) 1996 tipc_sk_send_ack(tsk); 1997 exit: 1998 release_sock(sk); 1999 return rc ? rc : copy; 2000 } 2001 2002 /** 2003 * tipc_recvstream - receive stream-oriented data 2004 * @sock: network socket 2005 * @m: descriptor for message info 2006 * @buflen: total size of user buffer area 2007 * @flags: receive flags 2008 * 2009 * Used for SOCK_STREAM messages only. If not enough data is available 2010 * will optionally wait for more; never truncates data. 2011 * 2012 * Return: size of returned message data, errno otherwise 2013 */ 2014 static int tipc_recvstream(struct socket *sock, struct msghdr *m, 2015 size_t buflen, int flags) 2016 { 2017 struct sock *sk = sock->sk; 2018 struct tipc_sock *tsk = tipc_sk(sk); 2019 struct sk_buff *skb; 2020 struct tipc_msg *hdr; 2021 struct tipc_skb_cb *skb_cb; 2022 bool peek = flags & MSG_PEEK; 2023 int offset, required, copy, copied = 0; 2024 int hlen, dlen, err, rc; 2025 long timeout; 2026 2027 /* Catch invalid receive attempts */ 2028 if (unlikely(!buflen)) 2029 return -EINVAL; 2030 2031 lock_sock(sk); 2032 2033 if (unlikely(sk->sk_state == TIPC_OPEN)) { 2034 rc = -ENOTCONN; 2035 goto exit; 2036 } 2037 required = sock_rcvlowat(sk, flags & MSG_WAITALL, buflen); 2038 timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); 2039 2040 do { 2041 /* Look at first msg in receive queue; wait if necessary */ 2042 rc = tipc_wait_for_rcvmsg(sock, &timeout); 2043 if (unlikely(rc)) 2044 break; 2045 skb = skb_peek(&sk->sk_receive_queue); 2046 skb_cb = TIPC_SKB_CB(skb); 2047 hdr = buf_msg(skb); 2048 dlen = msg_data_sz(hdr); 2049 hlen = msg_hdr_sz(hdr); 2050 err = msg_errcode(hdr); 2051 2052 /* Discard any empty non-errored (SYN-) message */ 2053 if (unlikely(!dlen && !err)) { 2054 tsk_advance_rx_queue(sk); 2055 continue; 2056 } 2057 2058 /* Collect msg meta data, incl. error code and rejected data */ 2059 if (!copied) { 2060 tipc_sk_set_orig_addr(m, skb); 2061 rc = tipc_sk_anc_data_recv(m, skb, tsk); 2062 if (rc) 2063 break; 2064 hdr = buf_msg(skb); 2065 } 2066 2067 /* Copy data if msg ok, otherwise return error/partial data */ 2068 if (likely(!err)) { 2069 offset = skb_cb->bytes_read; 2070 copy = min_t(int, dlen - offset, buflen - copied); 2071 rc = skb_copy_datagram_msg(skb, hlen + offset, m, copy); 2072 if (unlikely(rc)) 2073 break; 2074 copied += copy; 2075 offset += copy; 2076 if (unlikely(offset < dlen)) { 2077 if (!peek) 2078 skb_cb->bytes_read = offset; 2079 break; 2080 } 2081 } else { 2082 rc = 0; 2083 if ((err != TIPC_CONN_SHUTDOWN) && !m->msg_control) 2084 rc = -ECONNRESET; 2085 if (copied || rc) 2086 break; 2087 } 2088 2089 if (unlikely(peek)) 2090 break; 2091 2092 tsk_advance_rx_queue(sk); 2093 2094 /* Send connection flow control advertisement when applicable */ 2095 tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen); 2096 if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE) 2097 tipc_sk_send_ack(tsk); 2098 2099 /* Exit if all requested data or FIN/error received */ 2100 if (copied == buflen || err) 2101 break; 2102 2103 } while (!skb_queue_empty(&sk->sk_receive_queue) || copied < required); 2104 exit: 2105 release_sock(sk); 2106 return copied ? copied : rc; 2107 } 2108 2109 /** 2110 * tipc_write_space - wake up thread if port congestion is released 2111 * @sk: socket 2112 */ 2113 static void tipc_write_space(struct sock *sk) 2114 { 2115 struct socket_wq *wq; 2116 2117 rcu_read_lock(); 2118 wq = rcu_dereference(sk->sk_wq); 2119 if (skwq_has_sleeper(wq)) 2120 wake_up_interruptible_sync_poll(&wq->wait, EPOLLOUT | 2121 EPOLLWRNORM | EPOLLWRBAND); 2122 rcu_read_unlock(); 2123 } 2124 2125 /** 2126 * tipc_data_ready - wake up threads to indicate messages have been received 2127 * @sk: socket 2128 */ 2129 static void tipc_data_ready(struct sock *sk) 2130 { 2131 struct socket_wq *wq; 2132 2133 trace_sk_data_ready(sk); 2134 2135 rcu_read_lock(); 2136 wq = rcu_dereference(sk->sk_wq); 2137 if (skwq_has_sleeper(wq)) 2138 wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN | 2139 EPOLLRDNORM | EPOLLRDBAND); 2140 rcu_read_unlock(); 2141 } 2142 2143 static void tipc_sock_destruct(struct sock *sk) 2144 { 2145 __skb_queue_purge(&sk->sk_receive_queue); 2146 } 2147 2148 static void tipc_sk_proto_rcv(struct sock *sk, 2149 struct sk_buff_head *inputq, 2150 struct sk_buff_head *xmitq) 2151 { 2152 struct sk_buff *skb = __skb_dequeue(inputq); 2153 struct tipc_sock *tsk = tipc_sk(sk); 2154 struct tipc_msg *hdr = buf_msg(skb); 2155 struct tipc_group *grp = tsk->group; 2156 bool wakeup = false; 2157 2158 switch (msg_user(hdr)) { 2159 case CONN_MANAGER: 2160 tipc_sk_conn_proto_rcv(tsk, skb, inputq, xmitq); 2161 return; 2162 case SOCK_WAKEUP: 2163 tipc_dest_del(&tsk->cong_links, msg_orignode(hdr), 0); 2164 /* coupled with smp_rmb() in tipc_wait_for_cond() */ 2165 smp_wmb(); 2166 tsk->cong_link_cnt--; 2167 wakeup = true; 2168 tipc_sk_push_backlog(tsk, false); 2169 break; 2170 case GROUP_PROTOCOL: 2171 tipc_group_proto_rcv(grp, &wakeup, hdr, inputq, xmitq); 2172 break; 2173 case TOP_SRV: 2174 tipc_group_member_evt(tsk->group, &wakeup, &sk->sk_rcvbuf, 2175 hdr, inputq, xmitq); 2176 break; 2177 default: 2178 break; 2179 } 2180 2181 if (wakeup) 2182 sk->sk_write_space(sk); 2183 2184 kfree_skb(skb); 2185 } 2186 2187 /** 2188 * tipc_sk_filter_connect - check incoming message for a connection-based socket 2189 * @tsk: TIPC socket 2190 * @skb: pointer to message buffer. 2191 * @xmitq: for Nagle ACK if any 2192 * Return: true if message should be added to receive queue, false otherwise 2193 */ 2194 static bool tipc_sk_filter_connect(struct tipc_sock *tsk, struct sk_buff *skb, 2195 struct sk_buff_head *xmitq) 2196 { 2197 struct sock *sk = &tsk->sk; 2198 struct net *net = sock_net(sk); 2199 struct tipc_msg *hdr = buf_msg(skb); 2200 bool con_msg = msg_connected(hdr); 2201 u32 pport = tsk_peer_port(tsk); 2202 u32 pnode = tsk_peer_node(tsk); 2203 u32 oport = msg_origport(hdr); 2204 u32 onode = msg_orignode(hdr); 2205 int err = msg_errcode(hdr); 2206 unsigned long delay; 2207 2208 if (unlikely(msg_mcast(hdr))) 2209 return false; 2210 tsk->oneway = 0; 2211 2212 switch (sk->sk_state) { 2213 case TIPC_CONNECTING: 2214 /* Setup ACK */ 2215 if (likely(con_msg)) { 2216 if (err) 2217 break; 2218 tipc_sk_finish_conn(tsk, oport, onode); 2219 msg_set_importance(&tsk->phdr, msg_importance(hdr)); 2220 /* ACK+ message with data is added to receive queue */ 2221 if (msg_data_sz(hdr)) 2222 return true; 2223 /* Empty ACK-, - wake up sleeping connect() and drop */ 2224 sk->sk_state_change(sk); 2225 msg_set_dest_droppable(hdr, 1); 2226 return false; 2227 } 2228 /* Ignore connectionless message if not from listening socket */ 2229 if (oport != pport || onode != pnode) 2230 return false; 2231 2232 /* Rejected SYN */ 2233 if (err != TIPC_ERR_OVERLOAD) 2234 break; 2235 2236 /* Prepare for new setup attempt if we have a SYN clone */ 2237 if (skb_queue_empty(&sk->sk_write_queue)) 2238 break; 2239 get_random_bytes(&delay, 2); 2240 if (tsk->conn_timeout < 4) 2241 tsk->conn_timeout = 4; 2242 delay %= (tsk->conn_timeout / 4); 2243 delay = msecs_to_jiffies(delay + 100); 2244 sk_reset_timer(sk, &sk->sk_timer, jiffies + delay); 2245 return false; 2246 case TIPC_OPEN: 2247 case TIPC_DISCONNECTING: 2248 return false; 2249 case TIPC_LISTEN: 2250 /* Accept only SYN message */ 2251 if (!msg_is_syn(hdr) && 2252 tipc_node_get_capabilities(net, onode) & TIPC_SYN_BIT) 2253 return false; 2254 if (!con_msg && !err) 2255 return true; 2256 return false; 2257 case TIPC_ESTABLISHED: 2258 if (!skb_queue_empty(&sk->sk_write_queue)) 2259 tipc_sk_push_backlog(tsk, false); 2260 /* Accept only connection-based messages sent by peer */ 2261 if (likely(con_msg && !err && pport == oport && 2262 pnode == onode)) { 2263 if (msg_ack_required(hdr)) { 2264 struct sk_buff *skb; 2265 2266 skb = tipc_sk_build_ack(tsk); 2267 if (skb) { 2268 msg_set_nagle_ack(buf_msg(skb)); 2269 __skb_queue_tail(xmitq, skb); 2270 } 2271 } 2272 return true; 2273 } 2274 if (!tsk_peer_msg(tsk, hdr)) 2275 return false; 2276 if (!err) 2277 return true; 2278 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 2279 tipc_node_remove_conn(net, pnode, tsk->portid); 2280 sk->sk_state_change(sk); 2281 return true; 2282 default: 2283 pr_err("Unknown sk_state %u\n", sk->sk_state); 2284 } 2285 /* Abort connection setup attempt */ 2286 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 2287 sk->sk_err = ECONNREFUSED; 2288 sk->sk_state_change(sk); 2289 return true; 2290 } 2291 2292 /** 2293 * rcvbuf_limit - get proper overload limit of socket receive queue 2294 * @sk: socket 2295 * @skb: message 2296 * 2297 * For connection oriented messages, irrespective of importance, 2298 * default queue limit is 2 MB. 2299 * 2300 * For connectionless messages, queue limits are based on message 2301 * importance as follows: 2302 * 2303 * TIPC_LOW_IMPORTANCE (2 MB) 2304 * TIPC_MEDIUM_IMPORTANCE (4 MB) 2305 * TIPC_HIGH_IMPORTANCE (8 MB) 2306 * TIPC_CRITICAL_IMPORTANCE (16 MB) 2307 * 2308 * Return: overload limit according to corresponding message importance 2309 */ 2310 static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *skb) 2311 { 2312 struct tipc_sock *tsk = tipc_sk(sk); 2313 struct tipc_msg *hdr = buf_msg(skb); 2314 2315 if (unlikely(msg_in_group(hdr))) 2316 return READ_ONCE(sk->sk_rcvbuf); 2317 2318 if (unlikely(!msg_connected(hdr))) 2319 return READ_ONCE(sk->sk_rcvbuf) << msg_importance(hdr); 2320 2321 if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL)) 2322 return READ_ONCE(sk->sk_rcvbuf); 2323 2324 return FLOWCTL_MSG_LIM; 2325 } 2326 2327 /** 2328 * tipc_sk_filter_rcv - validate incoming message 2329 * @sk: socket 2330 * @skb: pointer to message. 2331 * @xmitq: output message area (FIXME) 2332 * 2333 * Enqueues message on receive queue if acceptable; optionally handles 2334 * disconnect indication for a connected socket. 2335 * 2336 * Called with socket lock already taken 2337 */ 2338 static void tipc_sk_filter_rcv(struct sock *sk, struct sk_buff *skb, 2339 struct sk_buff_head *xmitq) 2340 { 2341 bool sk_conn = !tipc_sk_type_connectionless(sk); 2342 struct tipc_sock *tsk = tipc_sk(sk); 2343 struct tipc_group *grp = tsk->group; 2344 struct tipc_msg *hdr = buf_msg(skb); 2345 struct net *net = sock_net(sk); 2346 struct sk_buff_head inputq; 2347 int mtyp = msg_type(hdr); 2348 int limit, err = TIPC_OK; 2349 2350 trace_tipc_sk_filter_rcv(sk, skb, TIPC_DUMP_ALL, " "); 2351 TIPC_SKB_CB(skb)->bytes_read = 0; 2352 __skb_queue_head_init(&inputq); 2353 __skb_queue_tail(&inputq, skb); 2354 2355 if (unlikely(!msg_isdata(hdr))) 2356 tipc_sk_proto_rcv(sk, &inputq, xmitq); 2357 2358 if (unlikely(grp)) 2359 tipc_group_filter_msg(grp, &inputq, xmitq); 2360 2361 if (unlikely(!grp) && mtyp == TIPC_MCAST_MSG) 2362 tipc_mcast_filter_msg(net, &tsk->mc_method.deferredq, &inputq); 2363 2364 /* Validate and add to receive buffer if there is space */ 2365 while ((skb = __skb_dequeue(&inputq))) { 2366 hdr = buf_msg(skb); 2367 limit = rcvbuf_limit(sk, skb); 2368 if ((sk_conn && !tipc_sk_filter_connect(tsk, skb, xmitq)) || 2369 (!sk_conn && msg_connected(hdr)) || 2370 (!grp && msg_in_group(hdr))) 2371 err = TIPC_ERR_NO_PORT; 2372 else if (sk_rmem_alloc_get(sk) + skb->truesize >= limit) { 2373 trace_tipc_sk_dump(sk, skb, TIPC_DUMP_ALL, 2374 "err_overload2!"); 2375 sk_drops_inc(sk); 2376 err = TIPC_ERR_OVERLOAD; 2377 } 2378 2379 if (unlikely(err)) { 2380 if (tipc_msg_reverse(tipc_own_addr(net), &skb, err)) { 2381 trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_NONE, 2382 "@filter_rcv!"); 2383 __skb_queue_tail(xmitq, skb); 2384 } 2385 err = TIPC_OK; 2386 continue; 2387 } 2388 __skb_queue_tail(&sk->sk_receive_queue, skb); 2389 skb_set_owner_r(skb, sk); 2390 trace_tipc_sk_overlimit2(sk, skb, TIPC_DUMP_ALL, 2391 "rcvq >90% allocated!"); 2392 sk->sk_data_ready(sk); 2393 } 2394 } 2395 2396 /** 2397 * tipc_sk_backlog_rcv - handle incoming message from backlog queue 2398 * @sk: socket 2399 * @skb: message 2400 * 2401 * Caller must hold socket lock 2402 */ 2403 static int tipc_sk_backlog_rcv(struct sock *sk, struct sk_buff *skb) 2404 { 2405 unsigned int before = sk_rmem_alloc_get(sk); 2406 struct sk_buff_head xmitq; 2407 unsigned int added; 2408 2409 __skb_queue_head_init(&xmitq); 2410 2411 tipc_sk_filter_rcv(sk, skb, &xmitq); 2412 added = sk_rmem_alloc_get(sk) - before; 2413 atomic_add(added, &tipc_sk(sk)->dupl_rcvcnt); 2414 2415 /* Send pending response/rejected messages, if any */ 2416 tipc_node_distr_xmit(sock_net(sk), &xmitq); 2417 return 0; 2418 } 2419 2420 /** 2421 * tipc_sk_enqueue - extract all buffers with destination 'dport' from 2422 * inputq and try adding them to socket or backlog queue 2423 * @inputq: list of incoming buffers with potentially different destinations 2424 * @sk: socket where the buffers should be enqueued 2425 * @dport: port number for the socket 2426 * @xmitq: output queue 2427 * 2428 * Caller must hold socket lock 2429 */ 2430 static void tipc_sk_enqueue(struct sk_buff_head *inputq, struct sock *sk, 2431 u32 dport, struct sk_buff_head *xmitq) 2432 { 2433 unsigned long time_limit = jiffies + usecs_to_jiffies(20000); 2434 struct sk_buff *skb; 2435 unsigned int lim; 2436 atomic_t *dcnt; 2437 u32 onode; 2438 2439 while (skb_queue_len(inputq)) { 2440 if (unlikely(time_after_eq(jiffies, time_limit))) 2441 return; 2442 2443 skb = tipc_skb_dequeue(inputq, dport); 2444 if (unlikely(!skb)) 2445 return; 2446 2447 /* Add message directly to receive queue if possible */ 2448 if (!sock_owned_by_user(sk)) { 2449 tipc_sk_filter_rcv(sk, skb, xmitq); 2450 continue; 2451 } 2452 2453 /* Try backlog, compensating for double-counted bytes */ 2454 dcnt = &tipc_sk(sk)->dupl_rcvcnt; 2455 if (!sk->sk_backlog.len) 2456 atomic_set(dcnt, 0); 2457 lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt); 2458 if (likely(!sk_add_backlog(sk, skb, lim))) { 2459 trace_tipc_sk_overlimit1(sk, skb, TIPC_DUMP_SK_BKLGQ, 2460 "bklg & rcvq >90% allocated!"); 2461 continue; 2462 } 2463 2464 trace_tipc_sk_dump(sk, skb, TIPC_DUMP_SK_BKLGQ, "err_overload!"); 2465 /* Overload => reject message back to sender */ 2466 onode = tipc_own_addr(sock_net(sk)); 2467 sk_drops_inc(sk); 2468 if (tipc_msg_reverse(onode, &skb, TIPC_ERR_OVERLOAD)) { 2469 trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_SK_BKLGQ, 2470 "@sk_enqueue!"); 2471 __skb_queue_tail(xmitq, skb); 2472 } 2473 break; 2474 } 2475 } 2476 2477 /** 2478 * tipc_sk_rcv - handle a chain of incoming buffers 2479 * @net: the associated network namespace 2480 * @inputq: buffer list containing the buffers 2481 * Consumes all buffers in list until inputq is empty 2482 * Note: may be called in multiple threads referring to the same queue 2483 */ 2484 void tipc_sk_rcv(struct net *net, struct sk_buff_head *inputq) 2485 { 2486 struct sk_buff_head xmitq; 2487 u32 dnode, dport = 0; 2488 int err; 2489 struct tipc_sock *tsk; 2490 struct sock *sk; 2491 struct sk_buff *skb; 2492 2493 __skb_queue_head_init(&xmitq); 2494 while (skb_queue_len(inputq)) { 2495 dport = tipc_skb_peek_port(inputq, dport); 2496 tsk = tipc_sk_lookup(net, dport); 2497 2498 if (likely(tsk)) { 2499 sk = &tsk->sk; 2500 if (likely(spin_trylock_bh(&sk->sk_lock.slock))) { 2501 tipc_sk_enqueue(inputq, sk, dport, &xmitq); 2502 spin_unlock_bh(&sk->sk_lock.slock); 2503 } 2504 /* Send pending response/rejected messages, if any */ 2505 tipc_node_distr_xmit(sock_net(sk), &xmitq); 2506 sock_put(sk); 2507 continue; 2508 } 2509 /* No destination socket => dequeue skb if still there */ 2510 skb = tipc_skb_dequeue(inputq, dport); 2511 if (!skb) 2512 return; 2513 2514 /* Try secondary lookup if unresolved named message */ 2515 err = TIPC_ERR_NO_PORT; 2516 if (tipc_msg_lookup_dest(net, skb, &err)) 2517 goto xmit; 2518 2519 /* Prepare for message rejection */ 2520 if (!tipc_msg_reverse(tipc_own_addr(net), &skb, err)) 2521 continue; 2522 2523 trace_tipc_sk_rej_msg(NULL, skb, TIPC_DUMP_NONE, "@sk_rcv!"); 2524 xmit: 2525 dnode = msg_destnode(buf_msg(skb)); 2526 tipc_node_xmit_skb(net, skb, dnode, dport); 2527 } 2528 } 2529 2530 static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) 2531 { 2532 DEFINE_WAIT_FUNC(wait, woken_wake_function); 2533 struct sock *sk = sock->sk; 2534 int done; 2535 2536 do { 2537 int err = sock_error(sk); 2538 if (err) 2539 return err; 2540 if (!*timeo_p) 2541 return -ETIMEDOUT; 2542 if (signal_pending(current)) 2543 return sock_intr_errno(*timeo_p); 2544 if (sk->sk_state == TIPC_DISCONNECTING) 2545 break; 2546 2547 add_wait_queue(sk_sleep(sk), &wait); 2548 done = sk_wait_event(sk, timeo_p, tipc_sk_connected(sk), 2549 &wait); 2550 remove_wait_queue(sk_sleep(sk), &wait); 2551 } while (!done); 2552 return 0; 2553 } 2554 2555 static bool tipc_sockaddr_is_sane(struct sockaddr_tipc *addr) 2556 { 2557 if (addr->family != AF_TIPC) 2558 return false; 2559 if (addr->addrtype == TIPC_SERVICE_RANGE) 2560 return (addr->addr.nameseq.lower <= addr->addr.nameseq.upper); 2561 return (addr->addrtype == TIPC_SERVICE_ADDR || 2562 addr->addrtype == TIPC_SOCKET_ADDR); 2563 } 2564 2565 /** 2566 * tipc_connect - establish a connection to another TIPC port 2567 * @sock: socket structure 2568 * @dest: socket address for destination port 2569 * @destlen: size of socket address data structure 2570 * @flags: file-related flags associated with socket 2571 * 2572 * Return: 0 on success, errno otherwise 2573 */ 2574 static int tipc_connect(struct socket *sock, struct sockaddr_unsized *dest, 2575 int destlen, int flags) 2576 { 2577 struct sock *sk = sock->sk; 2578 struct tipc_sock *tsk = tipc_sk(sk); 2579 struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; 2580 struct msghdr m = {NULL,}; 2581 long timeout = (flags & O_NONBLOCK) ? 0 : tsk->conn_timeout; 2582 int previous; 2583 int res = 0; 2584 2585 if (destlen != sizeof(struct sockaddr_tipc)) 2586 return -EINVAL; 2587 2588 lock_sock(sk); 2589 2590 if (tsk->group) { 2591 res = -EINVAL; 2592 goto exit; 2593 } 2594 2595 if (dst->family == AF_UNSPEC) { 2596 memset(&tsk->peer, 0, sizeof(struct sockaddr_tipc)); 2597 if (!tipc_sk_type_connectionless(sk)) 2598 res = -EINVAL; 2599 goto exit; 2600 } 2601 if (!tipc_sockaddr_is_sane(dst)) { 2602 res = -EINVAL; 2603 goto exit; 2604 } 2605 /* DGRAM/RDM connect(), just save the destaddr */ 2606 if (tipc_sk_type_connectionless(sk)) { 2607 memcpy(&tsk->peer, dest, destlen); 2608 goto exit; 2609 } else if (dst->addrtype == TIPC_SERVICE_RANGE) { 2610 res = -EINVAL; 2611 goto exit; 2612 } 2613 2614 previous = sk->sk_state; 2615 2616 switch (sk->sk_state) { 2617 case TIPC_OPEN: 2618 /* Send a 'SYN-' to destination */ 2619 m.msg_name = dest; 2620 m.msg_namelen = destlen; 2621 iov_iter_kvec(&m.msg_iter, ITER_SOURCE, NULL, 0, 0); 2622 2623 /* If connect is in non-blocking case, set MSG_DONTWAIT to 2624 * indicate send_msg() is never blocked. 2625 */ 2626 if (!timeout) 2627 m.msg_flags = MSG_DONTWAIT; 2628 2629 res = __tipc_sendmsg(sock, &m, 0); 2630 if ((res < 0) && (res != -EWOULDBLOCK)) 2631 goto exit; 2632 2633 /* Just entered TIPC_CONNECTING state; the only 2634 * difference is that return value in non-blocking 2635 * case is EINPROGRESS, rather than EALREADY. 2636 */ 2637 res = -EINPROGRESS; 2638 fallthrough; 2639 case TIPC_CONNECTING: 2640 if (!timeout) { 2641 if (previous == TIPC_CONNECTING) 2642 res = -EALREADY; 2643 goto exit; 2644 } 2645 timeout = msecs_to_jiffies(timeout); 2646 /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ 2647 res = tipc_wait_for_connect(sock, &timeout); 2648 break; 2649 case TIPC_ESTABLISHED: 2650 res = -EISCONN; 2651 break; 2652 default: 2653 res = -EINVAL; 2654 } 2655 2656 exit: 2657 release_sock(sk); 2658 return res; 2659 } 2660 2661 /** 2662 * tipc_listen - allow socket to listen for incoming connections 2663 * @sock: socket structure 2664 * @len: (unused) 2665 * 2666 * Return: 0 on success, errno otherwise 2667 */ 2668 static int tipc_listen(struct socket *sock, int len) 2669 { 2670 struct sock *sk = sock->sk; 2671 int res; 2672 2673 lock_sock(sk); 2674 res = tipc_set_sk_state(sk, TIPC_LISTEN); 2675 release_sock(sk); 2676 2677 return res; 2678 } 2679 2680 static int tipc_wait_for_accept(struct socket *sock, long timeo) 2681 { 2682 struct sock *sk = sock->sk; 2683 DEFINE_WAIT_FUNC(wait, woken_wake_function); 2684 int err; 2685 2686 /* True wake-one mechanism for incoming connections: only 2687 * one process gets woken up, not the 'whole herd'. 2688 * Since we do not 'race & poll' for established sockets 2689 * anymore, the common case will execute the loop only once. 2690 */ 2691 for (;;) { 2692 if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { 2693 add_wait_queue(sk_sleep(sk), &wait); 2694 release_sock(sk); 2695 timeo = wait_woken(&wait, TASK_INTERRUPTIBLE, timeo); 2696 lock_sock(sk); 2697 remove_wait_queue(sk_sleep(sk), &wait); 2698 } 2699 err = 0; 2700 if (!skb_queue_empty(&sk->sk_receive_queue)) 2701 break; 2702 err = -EAGAIN; 2703 if (!timeo) 2704 break; 2705 err = sock_intr_errno(timeo); 2706 if (signal_pending(current)) 2707 break; 2708 } 2709 return err; 2710 } 2711 2712 /** 2713 * tipc_accept - wait for connection request 2714 * @sock: listening socket 2715 * @new_sock: new socket that is to be connected 2716 * @arg: arguments for accept 2717 * 2718 * Return: 0 on success, errno otherwise 2719 */ 2720 static int tipc_accept(struct socket *sock, struct socket *new_sock, 2721 struct proto_accept_arg *arg) 2722 { 2723 struct sock *new_sk, *sk = sock->sk; 2724 struct tipc_sock *new_tsock; 2725 struct msghdr m = {NULL,}; 2726 struct tipc_msg *msg; 2727 struct sk_buff *buf; 2728 long timeo; 2729 int res; 2730 2731 lock_sock(sk); 2732 2733 if (sk->sk_state != TIPC_LISTEN) { 2734 res = -EINVAL; 2735 goto exit; 2736 } 2737 timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); 2738 res = tipc_wait_for_accept(sock, timeo); 2739 if (res) 2740 goto exit; 2741 2742 buf = skb_peek(&sk->sk_receive_queue); 2743 2744 res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, arg->kern); 2745 if (res) 2746 goto exit; 2747 security_sk_clone(sock->sk, new_sock->sk); 2748 2749 new_sk = new_sock->sk; 2750 new_tsock = tipc_sk(new_sk); 2751 msg = buf_msg(buf); 2752 2753 /* we lock on new_sk; but lockdep sees the lock on sk */ 2754 lock_sock_nested(new_sk, SINGLE_DEPTH_NESTING); 2755 2756 /* 2757 * Reject any stray messages received by new socket 2758 * before the socket lock was taken (very, very unlikely) 2759 */ 2760 tsk_rej_rx_queue(new_sk, TIPC_ERR_NO_PORT); 2761 2762 /* Connect new socket to it's peer */ 2763 tipc_sk_finish_conn(new_tsock, msg_origport(msg), msg_orignode(msg)); 2764 2765 tsk_set_importance(new_sk, msg_importance(msg)); 2766 if (msg_named(msg)) { 2767 new_tsock->conn_addrtype = TIPC_SERVICE_ADDR; 2768 msg_set_nametype(&new_tsock->phdr, msg_nametype(msg)); 2769 msg_set_nameinst(&new_tsock->phdr, msg_nameinst(msg)); 2770 } 2771 2772 /* 2773 * Respond to 'SYN-' by discarding it & returning 'ACK'. 2774 * Respond to 'SYN+' by queuing it on new socket & returning 'ACK'. 2775 */ 2776 if (!msg_data_sz(msg)) { 2777 tsk_advance_rx_queue(sk); 2778 } else { 2779 __skb_dequeue(&sk->sk_receive_queue); 2780 __skb_queue_head(&new_sk->sk_receive_queue, buf); 2781 skb_set_owner_r(buf, new_sk); 2782 } 2783 iov_iter_kvec(&m.msg_iter, ITER_SOURCE, NULL, 0, 0); 2784 __tipc_sendstream(new_sock, &m, 0); 2785 release_sock(new_sk); 2786 exit: 2787 release_sock(sk); 2788 return res; 2789 } 2790 2791 /** 2792 * tipc_shutdown - shutdown socket connection 2793 * @sock: socket structure 2794 * @how: direction to close (must be SHUT_RDWR) 2795 * 2796 * Terminates connection (if necessary), then purges socket's receive queue. 2797 * 2798 * Return: 0 on success, errno otherwise 2799 */ 2800 static int tipc_shutdown(struct socket *sock, int how) 2801 { 2802 struct sock *sk = sock->sk; 2803 int res; 2804 2805 if (how != SHUT_RDWR) 2806 return -EINVAL; 2807 2808 lock_sock(sk); 2809 2810 trace_tipc_sk_shutdown(sk, NULL, TIPC_DUMP_ALL, " "); 2811 __tipc_shutdown(sock, TIPC_CONN_SHUTDOWN); 2812 sk->sk_shutdown = SHUTDOWN_MASK; 2813 2814 if (sk->sk_state == TIPC_DISCONNECTING) { 2815 /* Discard any unreceived messages */ 2816 __skb_queue_purge(&sk->sk_receive_queue); 2817 2818 res = 0; 2819 } else { 2820 res = -ENOTCONN; 2821 } 2822 /* Wake up anyone sleeping in poll. */ 2823 sk->sk_state_change(sk); 2824 2825 release_sock(sk); 2826 return res; 2827 } 2828 2829 static void tipc_sk_check_probing_state(struct sock *sk, 2830 struct sk_buff_head *list) 2831 { 2832 struct tipc_sock *tsk = tipc_sk(sk); 2833 u32 pnode = tsk_peer_node(tsk); 2834 u32 pport = tsk_peer_port(tsk); 2835 u32 self = tsk_own_node(tsk); 2836 u32 oport = tsk->portid; 2837 struct sk_buff *skb; 2838 2839 if (tsk->probe_unacked) { 2840 tipc_set_sk_state(sk, TIPC_DISCONNECTING); 2841 sk->sk_err = ECONNABORTED; 2842 tipc_node_remove_conn(sock_net(sk), pnode, pport); 2843 sk->sk_state_change(sk); 2844 return; 2845 } 2846 /* Prepare new probe */ 2847 skb = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE, 0, 2848 pnode, self, pport, oport, TIPC_OK); 2849 if (skb) 2850 __skb_queue_tail(list, skb); 2851 tsk->probe_unacked = true; 2852 sk_reset_timer(sk, &sk->sk_timer, jiffies + CONN_PROBING_INTV); 2853 } 2854 2855 static void tipc_sk_retry_connect(struct sock *sk, struct sk_buff_head *list) 2856 { 2857 struct tipc_sock *tsk = tipc_sk(sk); 2858 2859 /* Try again later if dest link is congested */ 2860 if (tsk->cong_link_cnt) { 2861 sk_reset_timer(sk, &sk->sk_timer, 2862 jiffies + msecs_to_jiffies(100)); 2863 return; 2864 } 2865 /* Prepare SYN for retransmit */ 2866 tipc_msg_skb_clone(&sk->sk_write_queue, list); 2867 } 2868 2869 static void tipc_sk_timeout(struct timer_list *t) 2870 { 2871 struct sock *sk = timer_container_of(sk, t, sk_timer); 2872 struct tipc_sock *tsk = tipc_sk(sk); 2873 u32 pnode = tsk_peer_node(tsk); 2874 struct sk_buff_head list; 2875 int rc = 0; 2876 2877 __skb_queue_head_init(&list); 2878 bh_lock_sock(sk); 2879 2880 /* Try again later if socket is busy */ 2881 if (sock_owned_by_user(sk)) { 2882 sk_reset_timer(sk, &sk->sk_timer, jiffies + HZ / 20); 2883 bh_unlock_sock(sk); 2884 sock_put(sk); 2885 return; 2886 } 2887 2888 if (sk->sk_state == TIPC_ESTABLISHED) 2889 tipc_sk_check_probing_state(sk, &list); 2890 else if (sk->sk_state == TIPC_CONNECTING) 2891 tipc_sk_retry_connect(sk, &list); 2892 2893 bh_unlock_sock(sk); 2894 2895 if (!skb_queue_empty(&list)) 2896 rc = tipc_node_xmit(sock_net(sk), &list, pnode, tsk->portid); 2897 2898 /* SYN messages may cause link congestion */ 2899 if (rc == -ELINKCONG) { 2900 tipc_dest_push(&tsk->cong_links, pnode, 0); 2901 tsk->cong_link_cnt = 1; 2902 } 2903 sock_put(sk); 2904 } 2905 2906 static int tipc_sk_publish(struct tipc_sock *tsk, struct tipc_uaddr *ua) 2907 { 2908 struct sock *sk = &tsk->sk; 2909 struct net *net = sock_net(sk); 2910 struct tipc_socket_addr skaddr; 2911 struct publication *p; 2912 u32 key; 2913 2914 if (tipc_sk_connected(sk)) 2915 return -EINVAL; 2916 key = tsk->portid + tsk->pub_count + 1; 2917 if (key == tsk->portid) 2918 return -EADDRINUSE; 2919 skaddr.ref = tsk->portid; 2920 skaddr.node = tipc_own_addr(net); 2921 p = tipc_nametbl_publish(net, ua, &skaddr, key); 2922 if (unlikely(!p)) 2923 return -EINVAL; 2924 2925 list_add(&p->binding_sock, &tsk->publications); 2926 tsk->pub_count++; 2927 tsk->published = true; 2928 return 0; 2929 } 2930 2931 static int tipc_sk_withdraw(struct tipc_sock *tsk, struct tipc_uaddr *ua) 2932 { 2933 struct net *net = sock_net(&tsk->sk); 2934 struct publication *safe, *p; 2935 struct tipc_uaddr _ua; 2936 int rc = -EINVAL; 2937 2938 list_for_each_entry_safe(p, safe, &tsk->publications, binding_sock) { 2939 if (!ua) { 2940 tipc_uaddr(&_ua, TIPC_SERVICE_RANGE, p->scope, 2941 p->sr.type, p->sr.lower, p->sr.upper); 2942 tipc_nametbl_withdraw(net, &_ua, &p->sk, p->key); 2943 continue; 2944 } 2945 /* Unbind specific publication */ 2946 if (p->scope != ua->scope) 2947 continue; 2948 if (p->sr.type != ua->sr.type) 2949 continue; 2950 if (p->sr.lower != ua->sr.lower) 2951 continue; 2952 if (p->sr.upper != ua->sr.upper) 2953 break; 2954 tipc_nametbl_withdraw(net, ua, &p->sk, p->key); 2955 rc = 0; 2956 break; 2957 } 2958 if (list_empty(&tsk->publications)) { 2959 tsk->published = 0; 2960 rc = 0; 2961 } 2962 return rc; 2963 } 2964 2965 /* tipc_sk_reinit: set non-zero address in all existing sockets 2966 * when we go from standalone to network mode. 2967 */ 2968 void tipc_sk_reinit(struct net *net) 2969 { 2970 struct tipc_net *tn = net_generic(net, tipc_net_id); 2971 struct rhashtable_iter iter; 2972 struct tipc_sock *tsk; 2973 struct tipc_msg *msg; 2974 2975 rhashtable_walk_enter(&tn->sk_rht, &iter); 2976 2977 do { 2978 rhashtable_walk_start(&iter); 2979 2980 while ((tsk = rhashtable_walk_next(&iter)) && !IS_ERR(tsk)) { 2981 sock_hold(&tsk->sk); 2982 rhashtable_walk_stop(&iter); 2983 lock_sock(&tsk->sk); 2984 msg = &tsk->phdr; 2985 msg_set_prevnode(msg, tipc_own_addr(net)); 2986 msg_set_orignode(msg, tipc_own_addr(net)); 2987 release_sock(&tsk->sk); 2988 rhashtable_walk_start(&iter); 2989 sock_put(&tsk->sk); 2990 } 2991 2992 rhashtable_walk_stop(&iter); 2993 } while (tsk == ERR_PTR(-EAGAIN)); 2994 2995 rhashtable_walk_exit(&iter); 2996 } 2997 2998 static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid) 2999 { 3000 struct tipc_net *tn = net_generic(net, tipc_net_id); 3001 struct tipc_sock *tsk; 3002 3003 rcu_read_lock(); 3004 tsk = rhashtable_lookup(&tn->sk_rht, &portid, tsk_rht_params); 3005 if (tsk) 3006 sock_hold(&tsk->sk); 3007 rcu_read_unlock(); 3008 3009 return tsk; 3010 } 3011 3012 static int tipc_sk_insert(struct tipc_sock *tsk) 3013 { 3014 struct sock *sk = &tsk->sk; 3015 struct net *net = sock_net(sk); 3016 struct tipc_net *tn = net_generic(net, tipc_net_id); 3017 u32 remaining = (TIPC_MAX_PORT - TIPC_MIN_PORT) + 1; 3018 u32 portid = get_random_u32_below(remaining) + TIPC_MIN_PORT; 3019 3020 while (remaining--) { 3021 portid++; 3022 if ((portid < TIPC_MIN_PORT) || (portid > TIPC_MAX_PORT)) 3023 portid = TIPC_MIN_PORT; 3024 tsk->portid = portid; 3025 sock_hold(&tsk->sk); 3026 if (!rhashtable_lookup_insert_fast(&tn->sk_rht, &tsk->node, 3027 tsk_rht_params)) 3028 return 0; 3029 sock_put(&tsk->sk); 3030 } 3031 3032 return -1; 3033 } 3034 3035 static void tipc_sk_remove(struct tipc_sock *tsk) 3036 { 3037 struct sock *sk = &tsk->sk; 3038 struct tipc_net *tn = net_generic(sock_net(sk), tipc_net_id); 3039 3040 if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params)) 3041 __sock_put(sk); 3042 } 3043 3044 static const struct rhashtable_params tsk_rht_params = { 3045 .nelem_hint = 192, 3046 .head_offset = offsetof(struct tipc_sock, node), 3047 .key_offset = offsetof(struct tipc_sock, portid), 3048 .key_len = sizeof(u32), /* portid */ 3049 .max_size = 1048576, 3050 .min_size = 256, 3051 .automatic_shrinking = true, 3052 }; 3053 3054 int tipc_sk_rht_init(struct net *net) 3055 { 3056 struct tipc_net *tn = net_generic(net, tipc_net_id); 3057 3058 return rhashtable_init(&tn->sk_rht, &tsk_rht_params); 3059 } 3060 3061 void tipc_sk_rht_destroy(struct net *net) 3062 { 3063 struct tipc_net *tn = net_generic(net, tipc_net_id); 3064 3065 /* Wait for socket readers to complete */ 3066 synchronize_net(); 3067 3068 rhashtable_destroy(&tn->sk_rht); 3069 } 3070 3071 static int tipc_sk_join(struct tipc_sock *tsk, struct tipc_group_req *mreq) 3072 { 3073 struct net *net = sock_net(&tsk->sk); 3074 struct tipc_group *grp = tsk->group; 3075 struct tipc_msg *hdr = &tsk->phdr; 3076 struct tipc_uaddr ua; 3077 int rc; 3078 3079 if (mreq->type < TIPC_RESERVED_TYPES) 3080 return -EACCES; 3081 if (mreq->scope > TIPC_NODE_SCOPE) 3082 return -EINVAL; 3083 if (mreq->scope != TIPC_NODE_SCOPE) 3084 mreq->scope = TIPC_CLUSTER_SCOPE; 3085 if (grp) 3086 return -EACCES; 3087 grp = tipc_group_create(net, tsk->portid, mreq, &tsk->group_is_open); 3088 if (!grp) 3089 return -ENOMEM; 3090 tsk->group = grp; 3091 msg_set_lookup_scope(hdr, mreq->scope); 3092 msg_set_nametype(hdr, mreq->type); 3093 msg_set_dest_droppable(hdr, true); 3094 tipc_uaddr(&ua, TIPC_SERVICE_RANGE, mreq->scope, 3095 mreq->type, mreq->instance, mreq->instance); 3096 tipc_nametbl_build_group(net, grp, &ua); 3097 rc = tipc_sk_publish(tsk, &ua); 3098 if (rc) { 3099 tipc_group_delete(net, grp); 3100 tsk->group = NULL; 3101 return rc; 3102 } 3103 /* Eliminate any risk that a broadcast overtakes sent JOINs */ 3104 tsk->mc_method.rcast = true; 3105 tsk->mc_method.mandatory = true; 3106 tipc_group_join(net, grp, &tsk->sk.sk_rcvbuf); 3107 return rc; 3108 } 3109 3110 static int tipc_sk_leave(struct tipc_sock *tsk) 3111 { 3112 struct net *net = sock_net(&tsk->sk); 3113 struct tipc_group *grp = tsk->group; 3114 struct tipc_uaddr ua; 3115 int scope; 3116 3117 if (!grp) 3118 return -EINVAL; 3119 ua.addrtype = TIPC_SERVICE_RANGE; 3120 tipc_group_self(grp, &ua.sr, &scope); 3121 ua.scope = scope; 3122 tipc_group_delete(net, grp); 3123 tsk->group = NULL; 3124 tipc_sk_withdraw(tsk, &ua); 3125 return 0; 3126 } 3127 3128 /** 3129 * tipc_setsockopt - set socket option 3130 * @sock: socket structure 3131 * @lvl: option level 3132 * @opt: option identifier 3133 * @ov: pointer to new option value 3134 * @ol: length of option value 3135 * 3136 * For stream sockets only, accepts and ignores all IPPROTO_TCP options 3137 * (to ease compatibility). 3138 * 3139 * Return: 0 on success, errno otherwise 3140 */ 3141 static int tipc_setsockopt(struct socket *sock, int lvl, int opt, 3142 sockptr_t ov, unsigned int ol) 3143 { 3144 struct sock *sk = sock->sk; 3145 struct tipc_sock *tsk = tipc_sk(sk); 3146 struct tipc_group_req mreq; 3147 u32 value = 0; 3148 int res = 0; 3149 3150 if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM)) 3151 return 0; 3152 if (lvl != SOL_TIPC) 3153 return -ENOPROTOOPT; 3154 3155 switch (opt) { 3156 case TIPC_IMPORTANCE: 3157 case TIPC_SRC_DROPPABLE: 3158 case TIPC_DEST_DROPPABLE: 3159 case TIPC_CONN_TIMEOUT: 3160 case TIPC_NODELAY: 3161 if (ol < sizeof(value)) 3162 return -EINVAL; 3163 if (copy_from_sockptr(&value, ov, sizeof(u32))) 3164 return -EFAULT; 3165 break; 3166 case TIPC_GROUP_JOIN: 3167 if (ol < sizeof(mreq)) 3168 return -EINVAL; 3169 if (copy_from_sockptr(&mreq, ov, sizeof(mreq))) 3170 return -EFAULT; 3171 break; 3172 default: 3173 if (!sockptr_is_null(ov) || ol) 3174 return -EINVAL; 3175 } 3176 3177 lock_sock(sk); 3178 3179 switch (opt) { 3180 case TIPC_IMPORTANCE: 3181 res = tsk_set_importance(sk, value); 3182 break; 3183 case TIPC_SRC_DROPPABLE: 3184 if (sock->type != SOCK_STREAM) 3185 tsk_set_unreliable(tsk, value); 3186 else 3187 res = -ENOPROTOOPT; 3188 break; 3189 case TIPC_DEST_DROPPABLE: 3190 tsk_set_unreturnable(tsk, value); 3191 break; 3192 case TIPC_CONN_TIMEOUT: 3193 tipc_sk(sk)->conn_timeout = value; 3194 break; 3195 case TIPC_MCAST_BROADCAST: 3196 tsk->mc_method.rcast = false; 3197 tsk->mc_method.mandatory = true; 3198 break; 3199 case TIPC_MCAST_REPLICAST: 3200 tsk->mc_method.rcast = true; 3201 tsk->mc_method.mandatory = true; 3202 break; 3203 case TIPC_GROUP_JOIN: 3204 res = tipc_sk_join(tsk, &mreq); 3205 break; 3206 case TIPC_GROUP_LEAVE: 3207 res = tipc_sk_leave(tsk); 3208 break; 3209 case TIPC_NODELAY: 3210 tsk->nodelay = !!value; 3211 tsk_set_nagle(tsk); 3212 break; 3213 default: 3214 res = -EINVAL; 3215 } 3216 3217 release_sock(sk); 3218 3219 return res; 3220 } 3221 3222 /** 3223 * tipc_getsockopt - get socket option 3224 * @sock: socket structure 3225 * @lvl: option level 3226 * @opt: option identifier 3227 * @sopt: socket option container (input buffer length, output value/length) 3228 * 3229 * For stream sockets only, returns 0 length result for all IPPROTO_TCP options 3230 * (to ease compatibility). 3231 * 3232 * Return: 0 on success, errno otherwise 3233 */ 3234 static int tipc_getsockopt(struct socket *sock, int lvl, int opt, 3235 sockopt_t *sopt) 3236 { 3237 struct sock *sk = sock->sk; 3238 struct tipc_sock *tsk = tipc_sk(sk); 3239 struct tipc_service_range seq; 3240 int len, scope; 3241 int res = 0; 3242 u32 value; 3243 3244 if (lvl == IPPROTO_TCP && sock->type == SOCK_STREAM) { 3245 sopt->optlen = 0; 3246 return 0; 3247 } 3248 if (lvl != SOL_TIPC) 3249 return -ENOPROTOOPT; 3250 len = sopt->optlen; 3251 3252 lock_sock(sk); 3253 3254 switch (opt) { 3255 case TIPC_IMPORTANCE: 3256 value = tsk_importance(tsk); 3257 break; 3258 case TIPC_SRC_DROPPABLE: 3259 value = tsk_unreliable(tsk); 3260 break; 3261 case TIPC_DEST_DROPPABLE: 3262 value = tsk_unreturnable(tsk); 3263 break; 3264 case TIPC_CONN_TIMEOUT: 3265 value = tsk->conn_timeout; 3266 break; 3267 case TIPC_NODE_RECVQ_DEPTH: 3268 value = 0; /* was tipc_queue_size, now obsolete */ 3269 break; 3270 case TIPC_SOCK_RECVQ_DEPTH: 3271 value = skb_queue_len(&sk->sk_receive_queue); 3272 break; 3273 case TIPC_SOCK_RECVQ_USED: 3274 value = sk_rmem_alloc_get(sk); 3275 break; 3276 case TIPC_GROUP_JOIN: 3277 seq.type = 0; 3278 if (tsk->group) 3279 tipc_group_self(tsk->group, &seq, &scope); 3280 value = seq.type; 3281 break; 3282 default: 3283 res = -EINVAL; 3284 } 3285 3286 release_sock(sk); 3287 3288 if (res) 3289 return res; 3290 3291 if (len < sizeof(value)) 3292 return -EINVAL; 3293 3294 if (copy_to_iter(&value, sizeof(value), &sopt->iter_out) != 3295 sizeof(value)) 3296 return -EFAULT; 3297 sopt->optlen = sizeof(value); 3298 3299 return 0; 3300 } 3301 3302 static int tipc_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) 3303 { 3304 struct net *net = sock_net(sock->sk); 3305 struct tipc_sioc_nodeid_req nr = {0}; 3306 struct tipc_sioc_ln_req lnr; 3307 void __user *argp = (void __user *)arg; 3308 3309 switch (cmd) { 3310 case SIOCGETLINKNAME: 3311 if (copy_from_user(&lnr, argp, sizeof(lnr))) 3312 return -EFAULT; 3313 if (!tipc_node_get_linkname(net, 3314 lnr.bearer_id & 0xffff, lnr.peer, 3315 lnr.linkname, TIPC_MAX_LINK_NAME)) { 3316 if (copy_to_user(argp, &lnr, sizeof(lnr))) 3317 return -EFAULT; 3318 return 0; 3319 } 3320 return -EADDRNOTAVAIL; 3321 case SIOCGETNODEID: 3322 if (copy_from_user(&nr, argp, sizeof(nr))) 3323 return -EFAULT; 3324 if (!tipc_node_get_id(net, nr.peer, nr.node_id)) 3325 return -EADDRNOTAVAIL; 3326 if (copy_to_user(argp, &nr, sizeof(nr))) 3327 return -EFAULT; 3328 return 0; 3329 default: 3330 return -ENOIOCTLCMD; 3331 } 3332 } 3333 3334 static int tipc_socketpair(struct socket *sock1, struct socket *sock2) 3335 { 3336 struct tipc_sock *tsk2 = tipc_sk(sock2->sk); 3337 struct tipc_sock *tsk1 = tipc_sk(sock1->sk); 3338 u32 onode = tipc_own_addr(sock_net(sock1->sk)); 3339 3340 tsk1->peer.family = AF_TIPC; 3341 tsk1->peer.addrtype = TIPC_SOCKET_ADDR; 3342 tsk1->peer.scope = TIPC_NODE_SCOPE; 3343 tsk1->peer.addr.id.ref = tsk2->portid; 3344 tsk1->peer.addr.id.node = onode; 3345 tsk2->peer.family = AF_TIPC; 3346 tsk2->peer.addrtype = TIPC_SOCKET_ADDR; 3347 tsk2->peer.scope = TIPC_NODE_SCOPE; 3348 tsk2->peer.addr.id.ref = tsk1->portid; 3349 tsk2->peer.addr.id.node = onode; 3350 3351 tipc_sk_finish_conn(tsk1, tsk2->portid, onode); 3352 tipc_sk_finish_conn(tsk2, tsk1->portid, onode); 3353 return 0; 3354 } 3355 3356 /* Protocol switches for the various types of TIPC sockets */ 3357 3358 static const struct proto_ops msg_ops = { 3359 .owner = THIS_MODULE, 3360 .family = AF_TIPC, 3361 .release = tipc_release, 3362 .bind = tipc_bind, 3363 .connect = tipc_connect, 3364 .socketpair = tipc_socketpair, 3365 .accept = sock_no_accept, 3366 .getname = tipc_getname, 3367 .poll = tipc_poll, 3368 .ioctl = tipc_ioctl, 3369 .listen = sock_no_listen, 3370 .shutdown = tipc_shutdown, 3371 .setsockopt = tipc_setsockopt, 3372 .getsockopt_iter = tipc_getsockopt, 3373 .sendmsg = tipc_sendmsg, 3374 .recvmsg = tipc_recvmsg, 3375 .mmap = sock_no_mmap, 3376 }; 3377 3378 static const struct proto_ops packet_ops = { 3379 .owner = THIS_MODULE, 3380 .family = AF_TIPC, 3381 .release = tipc_release, 3382 .bind = tipc_bind, 3383 .connect = tipc_connect, 3384 .socketpair = tipc_socketpair, 3385 .accept = tipc_accept, 3386 .getname = tipc_getname, 3387 .poll = tipc_poll, 3388 .ioctl = tipc_ioctl, 3389 .listen = tipc_listen, 3390 .shutdown = tipc_shutdown, 3391 .setsockopt = tipc_setsockopt, 3392 .getsockopt_iter = tipc_getsockopt, 3393 .sendmsg = tipc_send_packet, 3394 .recvmsg = tipc_recvmsg, 3395 .mmap = sock_no_mmap, 3396 }; 3397 3398 static const struct proto_ops stream_ops = { 3399 .owner = THIS_MODULE, 3400 .family = AF_TIPC, 3401 .release = tipc_release, 3402 .bind = tipc_bind, 3403 .connect = tipc_connect, 3404 .socketpair = tipc_socketpair, 3405 .accept = tipc_accept, 3406 .getname = tipc_getname, 3407 .poll = tipc_poll, 3408 .ioctl = tipc_ioctl, 3409 .listen = tipc_listen, 3410 .shutdown = tipc_shutdown, 3411 .setsockopt = tipc_setsockopt, 3412 .getsockopt_iter = tipc_getsockopt, 3413 .sendmsg = tipc_sendstream, 3414 .recvmsg = tipc_recvstream, 3415 .mmap = sock_no_mmap, 3416 }; 3417 3418 static const struct net_proto_family tipc_family_ops = { 3419 .owner = THIS_MODULE, 3420 .family = AF_TIPC, 3421 .create = tipc_sk_create 3422 }; 3423 3424 static struct proto tipc_proto = { 3425 .name = "TIPC", 3426 .owner = THIS_MODULE, 3427 .obj_size = sizeof(struct tipc_sock), 3428 .sysctl_rmem = sysctl_tipc_rmem 3429 }; 3430 3431 /** 3432 * tipc_socket_init - initialize TIPC socket interface 3433 * 3434 * Return: 0 on success, errno otherwise 3435 */ 3436 int tipc_socket_init(void) 3437 { 3438 int res; 3439 3440 res = proto_register(&tipc_proto, 1); 3441 if (res) { 3442 pr_err("Failed to register TIPC protocol type\n"); 3443 goto out; 3444 } 3445 3446 res = sock_register(&tipc_family_ops); 3447 if (res) { 3448 pr_err("Failed to register TIPC socket type\n"); 3449 proto_unregister(&tipc_proto); 3450 goto out; 3451 } 3452 out: 3453 return res; 3454 } 3455 3456 /** 3457 * tipc_socket_stop - stop TIPC socket interface 3458 */ 3459 void tipc_socket_stop(void) 3460 { 3461 sock_unregister(tipc_family_ops.family); 3462 proto_unregister(&tipc_proto); 3463 } 3464 3465 /* Caller should hold socket lock for the passed tipc socket. */ 3466 static int __tipc_nl_add_sk_con(struct sk_buff *skb, struct tipc_sock *tsk) 3467 { 3468 u32 peer_node, peer_port; 3469 u32 conn_type, conn_instance; 3470 struct nlattr *nest; 3471 3472 peer_node = tsk_peer_node(tsk); 3473 peer_port = tsk_peer_port(tsk); 3474 conn_type = msg_nametype(&tsk->phdr); 3475 conn_instance = msg_nameinst(&tsk->phdr); 3476 nest = nla_nest_start_noflag(skb, TIPC_NLA_SOCK_CON); 3477 if (!nest) 3478 return -EMSGSIZE; 3479 3480 if (nla_put_u32(skb, TIPC_NLA_CON_NODE, peer_node)) 3481 goto msg_full; 3482 if (nla_put_u32(skb, TIPC_NLA_CON_SOCK, peer_port)) 3483 goto msg_full; 3484 3485 if (tsk->conn_addrtype != 0) { 3486 if (nla_put_flag(skb, TIPC_NLA_CON_FLAG)) 3487 goto msg_full; 3488 if (nla_put_u32(skb, TIPC_NLA_CON_TYPE, conn_type)) 3489 goto msg_full; 3490 if (nla_put_u32(skb, TIPC_NLA_CON_INST, conn_instance)) 3491 goto msg_full; 3492 } 3493 nla_nest_end(skb, nest); 3494 3495 return 0; 3496 3497 msg_full: 3498 nla_nest_cancel(skb, nest); 3499 3500 return -EMSGSIZE; 3501 } 3502 3503 static int __tipc_nl_add_sk_info(struct sk_buff *skb, struct tipc_sock 3504 *tsk) 3505 { 3506 struct net *net = sock_net(skb->sk); 3507 struct sock *sk = &tsk->sk; 3508 3509 if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->portid) || 3510 nla_put_u32(skb, TIPC_NLA_SOCK_ADDR, tipc_own_addr(net))) 3511 return -EMSGSIZE; 3512 3513 if (tipc_sk_connected(sk)) { 3514 if (__tipc_nl_add_sk_con(skb, tsk)) 3515 return -EMSGSIZE; 3516 } else if (!list_empty(&tsk->publications)) { 3517 if (nla_put_flag(skb, TIPC_NLA_SOCK_HAS_PUBL)) 3518 return -EMSGSIZE; 3519 } 3520 return 0; 3521 } 3522 3523 /* Caller should hold socket lock for the passed tipc socket. */ 3524 static int __tipc_nl_add_sk(struct sk_buff *skb, struct netlink_callback *cb, 3525 struct tipc_sock *tsk) 3526 { 3527 struct nlattr *attrs; 3528 void *hdr; 3529 3530 hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, 3531 &tipc_genl_family, NLM_F_MULTI, TIPC_NL_SOCK_GET); 3532 if (!hdr) 3533 goto msg_cancel; 3534 3535 attrs = nla_nest_start_noflag(skb, TIPC_NLA_SOCK); 3536 if (!attrs) 3537 goto genlmsg_cancel; 3538 3539 if (__tipc_nl_add_sk_info(skb, tsk)) 3540 goto attr_msg_cancel; 3541 3542 nla_nest_end(skb, attrs); 3543 genlmsg_end(skb, hdr); 3544 3545 return 0; 3546 3547 attr_msg_cancel: 3548 nla_nest_cancel(skb, attrs); 3549 genlmsg_cancel: 3550 genlmsg_cancel(skb, hdr); 3551 msg_cancel: 3552 return -EMSGSIZE; 3553 } 3554 3555 int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb, 3556 int (*skb_handler)(struct sk_buff *skb, 3557 struct netlink_callback *cb, 3558 struct tipc_sock *tsk)) 3559 { 3560 struct rhashtable_iter *iter = (void *)cb->args[4]; 3561 struct tipc_sock *tsk; 3562 int err; 3563 3564 rhashtable_walk_start(iter); 3565 while ((tsk = rhashtable_walk_next(iter)) != NULL) { 3566 if (IS_ERR(tsk)) { 3567 if (PTR_ERR(tsk) == -EAGAIN) 3568 continue; 3569 break; 3570 } 3571 3572 sock_hold(&tsk->sk); 3573 rhashtable_walk_stop(iter); 3574 lock_sock(&tsk->sk); 3575 err = skb_handler(skb, cb, tsk); 3576 if (err) { 3577 release_sock(&tsk->sk); 3578 sock_put(&tsk->sk); 3579 goto out; 3580 } 3581 release_sock(&tsk->sk); 3582 rhashtable_walk_start(iter); 3583 sock_put(&tsk->sk); 3584 } 3585 rhashtable_walk_stop(iter); 3586 out: 3587 return skb->len; 3588 } 3589 EXPORT_SYMBOL(tipc_nl_sk_walk); 3590 3591 int tipc_dump_start(struct netlink_callback *cb) 3592 { 3593 return __tipc_dump_start(cb, sock_net(cb->skb->sk)); 3594 } 3595 EXPORT_SYMBOL(tipc_dump_start); 3596 3597 int __tipc_dump_start(struct netlink_callback *cb, struct net *net) 3598 { 3599 /* tipc_nl_name_table_dump() uses cb->args[0...3]. */ 3600 struct rhashtable_iter *iter = (void *)cb->args[4]; 3601 struct tipc_net *tn = tipc_net(net); 3602 3603 if (!iter) { 3604 iter = kmalloc_obj(*iter); 3605 if (!iter) 3606 return -ENOMEM; 3607 3608 cb->args[4] = (long)iter; 3609 } 3610 3611 rhashtable_walk_enter(&tn->sk_rht, iter); 3612 return 0; 3613 } 3614 3615 int tipc_dump_done(struct netlink_callback *cb) 3616 { 3617 struct rhashtable_iter *hti = (void *)cb->args[4]; 3618 3619 rhashtable_walk_exit(hti); 3620 kfree(hti); 3621 return 0; 3622 } 3623 EXPORT_SYMBOL(tipc_dump_done); 3624 3625 int tipc_sk_fill_sock_diag(struct sk_buff *skb, struct netlink_callback *cb, 3626 struct tipc_sock *tsk, u32 sk_filter_state, 3627 u64 (*tipc_diag_gen_cookie)(struct sock *sk)) 3628 { 3629 struct sock *sk = &tsk->sk; 3630 struct nlattr *attrs; 3631 struct nlattr *stat; 3632 3633 /*filter response w.r.t sk_state*/ 3634 if (!(sk_filter_state & (1 << sk->sk_state))) 3635 return 0; 3636 3637 attrs = nla_nest_start_noflag(skb, TIPC_NLA_SOCK); 3638 if (!attrs) 3639 goto msg_cancel; 3640 3641 if (__tipc_nl_add_sk_info(skb, tsk)) 3642 goto attr_msg_cancel; 3643 3644 if (nla_put_u32(skb, TIPC_NLA_SOCK_TYPE, (u32)sk->sk_type) || 3645 nla_put_u32(skb, TIPC_NLA_SOCK_TIPC_STATE, (u32)sk->sk_state) || 3646 nla_put_u32(skb, TIPC_NLA_SOCK_INO, sock_i_ino(sk)) || 3647 nla_put_u32(skb, TIPC_NLA_SOCK_UID, 3648 from_kuid_munged(sk_user_ns(NETLINK_CB(cb->skb).sk), 3649 sk_uid(sk))) || 3650 nla_put_u64_64bit(skb, TIPC_NLA_SOCK_COOKIE, 3651 tipc_diag_gen_cookie(sk), 3652 TIPC_NLA_SOCK_PAD)) 3653 goto attr_msg_cancel; 3654 3655 stat = nla_nest_start_noflag(skb, TIPC_NLA_SOCK_STAT); 3656 if (!stat) 3657 goto attr_msg_cancel; 3658 3659 if (nla_put_u32(skb, TIPC_NLA_SOCK_STAT_RCVQ, 3660 skb_queue_len(&sk->sk_receive_queue)) || 3661 nla_put_u32(skb, TIPC_NLA_SOCK_STAT_SENDQ, 3662 skb_queue_len(&sk->sk_write_queue)) || 3663 nla_put_u32(skb, TIPC_NLA_SOCK_STAT_DROP, 3664 sk_drops_read(sk))) 3665 goto stat_msg_cancel; 3666 3667 if (tsk->cong_link_cnt && 3668 nla_put_flag(skb, TIPC_NLA_SOCK_STAT_LINK_CONG)) 3669 goto stat_msg_cancel; 3670 3671 if (tsk_conn_cong(tsk) && 3672 nla_put_flag(skb, TIPC_NLA_SOCK_STAT_CONN_CONG)) 3673 goto stat_msg_cancel; 3674 3675 nla_nest_end(skb, stat); 3676 3677 if (tsk->group) 3678 if (tipc_group_fill_sock_diag(tsk->group, skb)) 3679 goto stat_msg_cancel; 3680 3681 nla_nest_end(skb, attrs); 3682 3683 return 0; 3684 3685 stat_msg_cancel: 3686 nla_nest_cancel(skb, stat); 3687 attr_msg_cancel: 3688 nla_nest_cancel(skb, attrs); 3689 msg_cancel: 3690 return -EMSGSIZE; 3691 } 3692 EXPORT_SYMBOL(tipc_sk_fill_sock_diag); 3693 3694 int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb) 3695 { 3696 return tipc_nl_sk_walk(skb, cb, __tipc_nl_add_sk); 3697 } 3698 3699 /* Caller should hold socket lock for the passed tipc socket. */ 3700 static int __tipc_nl_add_sk_publ(struct sk_buff *skb, 3701 struct netlink_callback *cb, 3702 struct publication *publ) 3703 { 3704 void *hdr; 3705 struct nlattr *attrs; 3706 3707 hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, 3708 &tipc_genl_family, NLM_F_MULTI, TIPC_NL_PUBL_GET); 3709 if (!hdr) 3710 goto msg_cancel; 3711 3712 attrs = nla_nest_start_noflag(skb, TIPC_NLA_PUBL); 3713 if (!attrs) 3714 goto genlmsg_cancel; 3715 3716 if (nla_put_u32(skb, TIPC_NLA_PUBL_KEY, publ->key)) 3717 goto attr_msg_cancel; 3718 if (nla_put_u32(skb, TIPC_NLA_PUBL_TYPE, publ->sr.type)) 3719 goto attr_msg_cancel; 3720 if (nla_put_u32(skb, TIPC_NLA_PUBL_LOWER, publ->sr.lower)) 3721 goto attr_msg_cancel; 3722 if (nla_put_u32(skb, TIPC_NLA_PUBL_UPPER, publ->sr.upper)) 3723 goto attr_msg_cancel; 3724 3725 nla_nest_end(skb, attrs); 3726 genlmsg_end(skb, hdr); 3727 3728 return 0; 3729 3730 attr_msg_cancel: 3731 nla_nest_cancel(skb, attrs); 3732 genlmsg_cancel: 3733 genlmsg_cancel(skb, hdr); 3734 msg_cancel: 3735 return -EMSGSIZE; 3736 } 3737 3738 /* Caller should hold socket lock for the passed tipc socket. */ 3739 static int __tipc_nl_list_sk_publ(struct sk_buff *skb, 3740 struct netlink_callback *cb, 3741 struct tipc_sock *tsk, u32 *last_publ) 3742 { 3743 int err; 3744 struct publication *p; 3745 3746 if (*last_publ) { 3747 list_for_each_entry(p, &tsk->publications, binding_sock) { 3748 if (p->key == *last_publ) 3749 break; 3750 } 3751 if (list_entry_is_head(p, &tsk->publications, binding_sock)) { 3752 /* We never set seq or call nl_dump_check_consistent() 3753 * this means that setting prev_seq here will cause the 3754 * consistence check to fail in the netlink callback 3755 * handler. Resulting in the last NLMSG_DONE message 3756 * having the NLM_F_DUMP_INTR flag set. 3757 */ 3758 cb->prev_seq = 1; 3759 *last_publ = 0; 3760 return -EPIPE; 3761 } 3762 } else { 3763 p = list_first_entry(&tsk->publications, struct publication, 3764 binding_sock); 3765 } 3766 3767 list_for_each_entry_from(p, &tsk->publications, binding_sock) { 3768 err = __tipc_nl_add_sk_publ(skb, cb, p); 3769 if (err) { 3770 *last_publ = p->key; 3771 return err; 3772 } 3773 } 3774 *last_publ = 0; 3775 3776 return 0; 3777 } 3778 3779 int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb) 3780 { 3781 int err; 3782 u32 tsk_portid = cb->args[0]; 3783 u32 last_publ = cb->args[1]; 3784 u32 done = cb->args[2]; 3785 struct net *net = sock_net(skb->sk); 3786 struct tipc_sock *tsk; 3787 3788 if (!tsk_portid) { 3789 struct nlattr **attrs = genl_dumpit_info(cb)->info.attrs; 3790 struct nlattr *sock[TIPC_NLA_SOCK_MAX + 1]; 3791 3792 if (!attrs[TIPC_NLA_SOCK]) 3793 return -EINVAL; 3794 3795 err = nla_parse_nested_deprecated(sock, TIPC_NLA_SOCK_MAX, 3796 attrs[TIPC_NLA_SOCK], 3797 tipc_nl_sock_policy, NULL); 3798 if (err) 3799 return err; 3800 3801 if (!sock[TIPC_NLA_SOCK_REF]) 3802 return -EINVAL; 3803 3804 tsk_portid = nla_get_u32(sock[TIPC_NLA_SOCK_REF]); 3805 } 3806 3807 if (done) 3808 return 0; 3809 3810 tsk = tipc_sk_lookup(net, tsk_portid); 3811 if (!tsk) 3812 return -EINVAL; 3813 3814 lock_sock(&tsk->sk); 3815 err = __tipc_nl_list_sk_publ(skb, cb, tsk, &last_publ); 3816 if (!err) 3817 done = 1; 3818 release_sock(&tsk->sk); 3819 sock_put(&tsk->sk); 3820 3821 cb->args[0] = tsk_portid; 3822 cb->args[1] = last_publ; 3823 cb->args[2] = done; 3824 3825 return skb->len; 3826 } 3827 3828 /** 3829 * tipc_sk_filtering - check if a socket should be traced 3830 * @sk: the socket to be examined 3831 * 3832 * @sysctl_tipc_sk_filter is used as the socket tuple for filtering: 3833 * (portid, sock type, name type, name lower, name upper) 3834 * 3835 * Return: true if the socket meets the socket tuple data 3836 * (value 0 = 'any') or when there is no tuple set (all = 0), 3837 * otherwise false 3838 */ 3839 bool tipc_sk_filtering(struct sock *sk) 3840 { 3841 struct tipc_sock *tsk; 3842 struct publication *p; 3843 u32 _port, _sktype, _type, _lower, _upper; 3844 u32 type = 0, lower = 0, upper = 0; 3845 3846 if (!sk) 3847 return true; 3848 3849 tsk = tipc_sk(sk); 3850 3851 _port = sysctl_tipc_sk_filter[0]; 3852 _sktype = sysctl_tipc_sk_filter[1]; 3853 _type = sysctl_tipc_sk_filter[2]; 3854 _lower = sysctl_tipc_sk_filter[3]; 3855 _upper = sysctl_tipc_sk_filter[4]; 3856 3857 if (!_port && !_sktype && !_type && !_lower && !_upper) 3858 return true; 3859 3860 if (_port) 3861 return (_port == tsk->portid); 3862 3863 if (_sktype && _sktype != sk->sk_type) 3864 return false; 3865 3866 if (tsk->published) { 3867 p = list_first_entry_or_null(&tsk->publications, 3868 struct publication, binding_sock); 3869 if (p) { 3870 type = p->sr.type; 3871 lower = p->sr.lower; 3872 upper = p->sr.upper; 3873 } 3874 } 3875 3876 if (!tipc_sk_type_connectionless(sk)) { 3877 type = msg_nametype(&tsk->phdr); 3878 lower = msg_nameinst(&tsk->phdr); 3879 upper = lower; 3880 } 3881 3882 if ((_type && _type != type) || (_lower && _lower != lower) || 3883 (_upper && _upper != upper)) 3884 return false; 3885 3886 return true; 3887 } 3888 3889 u32 tipc_sock_get_portid(struct sock *sk) 3890 { 3891 return (sk) ? (tipc_sk(sk))->portid : 0; 3892 } 3893 3894 /** 3895 * tipc_sk_overlimit1 - check if socket rx queue is about to be overloaded, 3896 * both the rcv and backlog queues are considered 3897 * @sk: tipc sk to be checked 3898 * @skb: tipc msg to be checked 3899 * 3900 * Return: true if the socket rx queue allocation is > 90%, otherwise false 3901 */ 3902 3903 bool tipc_sk_overlimit1(struct sock *sk, struct sk_buff *skb) 3904 { 3905 atomic_t *dcnt = &tipc_sk(sk)->dupl_rcvcnt; 3906 unsigned int lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt); 3907 unsigned int qsize = sk->sk_backlog.len + sk_rmem_alloc_get(sk); 3908 3909 return (qsize > lim * 90 / 100); 3910 } 3911 3912 /** 3913 * tipc_sk_overlimit2 - check if socket rx queue is about to be overloaded, 3914 * only the rcv queue is considered 3915 * @sk: tipc sk to be checked 3916 * @skb: tipc msg to be checked 3917 * 3918 * Return: true if the socket rx queue allocation is > 90%, otherwise false 3919 */ 3920 3921 bool tipc_sk_overlimit2(struct sock *sk, struct sk_buff *skb) 3922 { 3923 unsigned int lim = rcvbuf_limit(sk, skb); 3924 unsigned int qsize = sk_rmem_alloc_get(sk); 3925 3926 return (qsize > lim * 90 / 100); 3927 } 3928 3929 /** 3930 * tipc_sk_dump - dump TIPC socket 3931 * @sk: tipc sk to be dumped 3932 * @dqueues: bitmask to decide if any socket queue to be dumped? 3933 * - TIPC_DUMP_NONE: don't dump socket queues 3934 * - TIPC_DUMP_SK_SNDQ: dump socket send queue 3935 * - TIPC_DUMP_SK_RCVQ: dump socket rcv queue 3936 * - TIPC_DUMP_SK_BKLGQ: dump socket backlog queue 3937 * - TIPC_DUMP_ALL: dump all the socket queues above 3938 * @buf: returned buffer of dump data in format 3939 */ 3940 int tipc_sk_dump(struct sock *sk, u16 dqueues, char *buf) 3941 { 3942 int i = 0; 3943 size_t sz = (dqueues) ? SK_LMAX : SK_LMIN; 3944 u32 conn_type, conn_instance; 3945 struct tipc_sock *tsk; 3946 struct publication *p; 3947 bool tsk_connected; 3948 3949 if (!sk) { 3950 i += scnprintf(buf, sz, "sk data: (null)\n"); 3951 return i; 3952 } 3953 3954 tsk = tipc_sk(sk); 3955 tsk_connected = !tipc_sk_type_connectionless(sk); 3956 3957 i += scnprintf(buf, sz, "sk data: %u", sk->sk_type); 3958 i += scnprintf(buf + i, sz - i, " %d", sk->sk_state); 3959 i += scnprintf(buf + i, sz - i, " %x", tsk_own_node(tsk)); 3960 i += scnprintf(buf + i, sz - i, " %u", tsk->portid); 3961 i += scnprintf(buf + i, sz - i, " | %u", tsk_connected); 3962 if (tsk_connected) { 3963 i += scnprintf(buf + i, sz - i, " %x", tsk_peer_node(tsk)); 3964 i += scnprintf(buf + i, sz - i, " %u", tsk_peer_port(tsk)); 3965 conn_type = msg_nametype(&tsk->phdr); 3966 conn_instance = msg_nameinst(&tsk->phdr); 3967 i += scnprintf(buf + i, sz - i, " %u", conn_type); 3968 i += scnprintf(buf + i, sz - i, " %u", conn_instance); 3969 } 3970 i += scnprintf(buf + i, sz - i, " | %u", tsk->published); 3971 if (tsk->published) { 3972 p = list_first_entry_or_null(&tsk->publications, 3973 struct publication, binding_sock); 3974 i += scnprintf(buf + i, sz - i, " %u", (p) ? p->sr.type : 0); 3975 i += scnprintf(buf + i, sz - i, " %u", (p) ? p->sr.lower : 0); 3976 i += scnprintf(buf + i, sz - i, " %u", (p) ? p->sr.upper : 0); 3977 } 3978 i += scnprintf(buf + i, sz - i, " | %u", tsk->snd_win); 3979 i += scnprintf(buf + i, sz - i, " %u", tsk->rcv_win); 3980 i += scnprintf(buf + i, sz - i, " %u", tsk->max_pkt); 3981 i += scnprintf(buf + i, sz - i, " %x", tsk->peer_caps); 3982 i += scnprintf(buf + i, sz - i, " %u", tsk->cong_link_cnt); 3983 i += scnprintf(buf + i, sz - i, " %u", tsk->snt_unacked); 3984 i += scnprintf(buf + i, sz - i, " %u", tsk->rcv_unacked); 3985 i += scnprintf(buf + i, sz - i, " %u", atomic_read(&tsk->dupl_rcvcnt)); 3986 i += scnprintf(buf + i, sz - i, " %u", sk->sk_shutdown); 3987 i += scnprintf(buf + i, sz - i, " | %d", sk_wmem_alloc_get(sk)); 3988 i += scnprintf(buf + i, sz - i, " %d", sk->sk_sndbuf); 3989 i += scnprintf(buf + i, sz - i, " | %d", sk_rmem_alloc_get(sk)); 3990 i += scnprintf(buf + i, sz - i, " %d", sk->sk_rcvbuf); 3991 i += scnprintf(buf + i, sz - i, " | %d\n", READ_ONCE(sk->sk_backlog.len)); 3992 3993 if (dqueues & TIPC_DUMP_SK_SNDQ) { 3994 i += scnprintf(buf + i, sz - i, "sk_write_queue: "); 3995 i += tipc_list_dump(&sk->sk_write_queue, false, buf + i); 3996 } 3997 3998 if (dqueues & TIPC_DUMP_SK_RCVQ) { 3999 i += scnprintf(buf + i, sz - i, "sk_receive_queue: "); 4000 i += tipc_list_dump(&sk->sk_receive_queue, false, buf + i); 4001 } 4002 4003 if (dqueues & TIPC_DUMP_SK_BKLGQ) { 4004 i += scnprintf(buf + i, sz - i, "sk_backlog:\n head "); 4005 i += tipc_skb_dump(sk->sk_backlog.head, false, buf + i); 4006 if (sk->sk_backlog.tail != sk->sk_backlog.head) { 4007 i += scnprintf(buf + i, sz - i, " tail "); 4008 i += tipc_skb_dump(sk->sk_backlog.tail, false, 4009 buf + i); 4010 } 4011 } 4012 4013 return i; 4014 } 4015