xref: /openbmc/linux/net/tipc/socket.c (revision 18da174d)
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 <trace/events/sock.h>
41 
42 #include "core.h"
43 #include "name_table.h"
44 #include "node.h"
45 #include "link.h"
46 #include "name_distr.h"
47 #include "socket.h"
48 #include "bcast.h"
49 #include "netlink.h"
50 #include "group.h"
51 #include "trace.h"
52 
53 #define NAGLE_START_INIT	4
54 #define NAGLE_START_MAX		1024
55 #define CONN_TIMEOUT_DEFAULT    8000    /* default connect timeout = 8s */
56 #define CONN_PROBING_INTV	msecs_to_jiffies(3600000)  /* [ms] => 1 h */
57 #define TIPC_MAX_PORT		0xffffffff
58 #define TIPC_MIN_PORT		1
59 #define TIPC_ACK_RATE		4       /* ACK at 1/4 of rcv window size */
60 
61 enum {
62 	TIPC_LISTEN = TCP_LISTEN,
63 	TIPC_ESTABLISHED = TCP_ESTABLISHED,
64 	TIPC_OPEN = TCP_CLOSE,
65 	TIPC_DISCONNECTING = TCP_CLOSE_WAIT,
66 	TIPC_CONNECTING = TCP_SYN_SENT,
67 };
68 
69 struct sockaddr_pair {
70 	struct sockaddr_tipc sock;
71 	struct sockaddr_tipc member;
72 };
73 
74 /**
75  * struct tipc_sock - TIPC socket structure
76  * @sk: socket - interacts with 'port' and with user via the socket API
77  * @max_pkt: maximum packet size "hint" used when building messages sent by port
78  * @maxnagle: maximum size of msg which can be subject to nagle
79  * @portid: unique port identity in TIPC socket hash table
80  * @phdr: preformatted message header used when sending messages
81  * @cong_links: list of congested links
82  * @publications: list of publications for port
83  * @blocking_link: address of the congested link we are currently sleeping on
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 int tipc_accept(struct socket *sock, struct socket *new_sock, int flags,
151 		       bool kern);
152 static void tipc_sk_timeout(struct timer_list *t);
153 static int tipc_sk_publish(struct tipc_sock *tsk, struct tipc_uaddr *ua);
154 static int tipc_sk_withdraw(struct tipc_sock *tsk, struct tipc_uaddr *ua);
155 static int tipc_sk_leave(struct tipc_sock *tsk);
156 static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid);
157 static int tipc_sk_insert(struct tipc_sock *tsk);
158 static void tipc_sk_remove(struct tipc_sock *tsk);
159 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz);
160 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz);
161 static void tipc_sk_push_backlog(struct tipc_sock *tsk, bool nagle_ack);
162 static int tipc_wait_for_connect(struct socket *sock, long *timeo_p);
163 
164 static const struct proto_ops packet_ops;
165 static const struct proto_ops stream_ops;
166 static const struct proto_ops msg_ops;
167 static struct proto tipc_proto;
168 static const struct rhashtable_params tsk_rht_params;
169 
170 static u32 tsk_own_node(struct tipc_sock *tsk)
171 {
172 	return msg_prevnode(&tsk->phdr);
173 }
174 
175 static u32 tsk_peer_node(struct tipc_sock *tsk)
176 {
177 	return msg_destnode(&tsk->phdr);
178 }
179 
180 static u32 tsk_peer_port(struct tipc_sock *tsk)
181 {
182 	return msg_destport(&tsk->phdr);
183 }
184 
185 static  bool tsk_unreliable(struct tipc_sock *tsk)
186 {
187 	return msg_src_droppable(&tsk->phdr) != 0;
188 }
189 
190 static void tsk_set_unreliable(struct tipc_sock *tsk, bool unreliable)
191 {
192 	msg_set_src_droppable(&tsk->phdr, unreliable ? 1 : 0);
193 }
194 
195 static bool tsk_unreturnable(struct tipc_sock *tsk)
196 {
197 	return msg_dest_droppable(&tsk->phdr) != 0;
198 }
199 
200 static void tsk_set_unreturnable(struct tipc_sock *tsk, bool unreturnable)
201 {
202 	msg_set_dest_droppable(&tsk->phdr, unreturnable ? 1 : 0);
203 }
204 
205 static int tsk_importance(struct tipc_sock *tsk)
206 {
207 	return msg_importance(&tsk->phdr);
208 }
209 
210 static struct tipc_sock *tipc_sk(const struct sock *sk)
211 {
212 	return container_of(sk, struct tipc_sock, sk);
213 }
214 
215 int tsk_set_importance(struct sock *sk, int imp)
216 {
217 	if (imp > TIPC_CRITICAL_IMPORTANCE)
218 		return -EINVAL;
219 	msg_set_importance(&tipc_sk(sk)->phdr, (u32)imp);
220 	return 0;
221 }
222 
223 static bool tsk_conn_cong(struct tipc_sock *tsk)
224 {
225 	return tsk->snt_unacked > tsk->snd_win;
226 }
227 
228 static u16 tsk_blocks(int len)
229 {
230 	return ((len / FLOWCTL_BLK_SZ) + 1);
231 }
232 
233 /* tsk_blocks(): translate a buffer size in bytes to number of
234  * advertisable blocks, taking into account the ratio truesize(len)/len
235  * We can trust that this ratio is always < 4 for len >= FLOWCTL_BLK_SZ
236  */
237 static u16 tsk_adv_blocks(int len)
238 {
239 	return len / FLOWCTL_BLK_SZ / 4;
240 }
241 
242 /* tsk_inc(): increment counter for sent or received data
243  * - If block based flow control is not supported by peer we
244  *   fall back to message based ditto, incrementing the counter
245  */
246 static u16 tsk_inc(struct tipc_sock *tsk, int msglen)
247 {
248 	if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL))
249 		return ((msglen / FLOWCTL_BLK_SZ) + 1);
250 	return 1;
251 }
252 
253 /* tsk_set_nagle - enable/disable nagle property by manipulating maxnagle
254  */
255 static void tsk_set_nagle(struct tipc_sock *tsk)
256 {
257 	struct sock *sk = &tsk->sk;
258 
259 	tsk->maxnagle = 0;
260 	if (sk->sk_type != SOCK_STREAM)
261 		return;
262 	if (tsk->nodelay)
263 		return;
264 	if (!(tsk->peer_caps & TIPC_NAGLE))
265 		return;
266 	/* Limit node local buffer size to avoid receive queue overflow */
267 	if (tsk->max_pkt == MAX_MSG_SIZE)
268 		tsk->maxnagle = 1500;
269 	else
270 		tsk->maxnagle = tsk->max_pkt;
271 }
272 
273 /**
274  * tsk_advance_rx_queue - discard first buffer in socket receive queue
275  * @sk: network socket
276  *
277  * Caller must hold socket lock
278  */
279 static void tsk_advance_rx_queue(struct sock *sk)
280 {
281 	trace_tipc_sk_advance_rx(sk, NULL, TIPC_DUMP_SK_RCVQ, " ");
282 	kfree_skb(__skb_dequeue(&sk->sk_receive_queue));
283 }
284 
285 /* tipc_sk_respond() : send response message back to sender
286  */
287 static void tipc_sk_respond(struct sock *sk, struct sk_buff *skb, int err)
288 {
289 	u32 selector;
290 	u32 dnode;
291 	u32 onode = tipc_own_addr(sock_net(sk));
292 
293 	if (!tipc_msg_reverse(onode, &skb, err))
294 		return;
295 
296 	trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_NONE, "@sk_respond!");
297 	dnode = msg_destnode(buf_msg(skb));
298 	selector = msg_origport(buf_msg(skb));
299 	tipc_node_xmit_skb(sock_net(sk), skb, dnode, selector);
300 }
301 
302 /**
303  * tsk_rej_rx_queue - reject all buffers in socket receive queue
304  * @sk: network socket
305  * @error: response error code
306  *
307  * Caller must hold socket lock
308  */
309 static void tsk_rej_rx_queue(struct sock *sk, int error)
310 {
311 	struct sk_buff *skb;
312 
313 	while ((skb = __skb_dequeue(&sk->sk_receive_queue)))
314 		tipc_sk_respond(sk, skb, error);
315 }
316 
317 static bool tipc_sk_connected(const struct sock *sk)
318 {
319 	return READ_ONCE(sk->sk_state) == TIPC_ESTABLISHED;
320 }
321 
322 /* tipc_sk_type_connectionless - check if the socket is datagram socket
323  * @sk: socket
324  *
325  * Returns true if connection less, false otherwise
326  */
327 static bool tipc_sk_type_connectionless(struct sock *sk)
328 {
329 	return sk->sk_type == SOCK_RDM || sk->sk_type == SOCK_DGRAM;
330 }
331 
332 /* tsk_peer_msg - verify if message was sent by connected port's peer
333  *
334  * Handles cases where the node's network address has changed from
335  * the default of <0.0.0> to its configured setting.
336  */
337 static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg)
338 {
339 	struct sock *sk = &tsk->sk;
340 	u32 self = tipc_own_addr(sock_net(sk));
341 	u32 peer_port = tsk_peer_port(tsk);
342 	u32 orig_node, peer_node;
343 
344 	if (unlikely(!tipc_sk_connected(sk)))
345 		return false;
346 
347 	if (unlikely(msg_origport(msg) != peer_port))
348 		return false;
349 
350 	orig_node = msg_orignode(msg);
351 	peer_node = tsk_peer_node(tsk);
352 
353 	if (likely(orig_node == peer_node))
354 		return true;
355 
356 	if (!orig_node && peer_node == self)
357 		return true;
358 
359 	if (!peer_node && orig_node == self)
360 		return true;
361 
362 	return false;
363 }
364 
365 /* tipc_set_sk_state - set the sk_state of the socket
366  * @sk: socket
367  *
368  * Caller must hold socket lock
369  *
370  * Returns 0 on success, errno otherwise
371  */
372 static int tipc_set_sk_state(struct sock *sk, int state)
373 {
374 	int oldsk_state = sk->sk_state;
375 	int res = -EINVAL;
376 
377 	switch (state) {
378 	case TIPC_OPEN:
379 		res = 0;
380 		break;
381 	case TIPC_LISTEN:
382 	case TIPC_CONNECTING:
383 		if (oldsk_state == TIPC_OPEN)
384 			res = 0;
385 		break;
386 	case TIPC_ESTABLISHED:
387 		if (oldsk_state == TIPC_CONNECTING ||
388 		    oldsk_state == TIPC_OPEN)
389 			res = 0;
390 		break;
391 	case TIPC_DISCONNECTING:
392 		if (oldsk_state == TIPC_CONNECTING ||
393 		    oldsk_state == TIPC_ESTABLISHED)
394 			res = 0;
395 		break;
396 	}
397 
398 	if (!res)
399 		sk->sk_state = state;
400 
401 	return res;
402 }
403 
404 static int tipc_sk_sock_err(struct socket *sock, long *timeout)
405 {
406 	struct sock *sk = sock->sk;
407 	int err = sock_error(sk);
408 	int typ = sock->type;
409 
410 	if (err)
411 		return err;
412 	if (typ == SOCK_STREAM || typ == SOCK_SEQPACKET) {
413 		if (sk->sk_state == TIPC_DISCONNECTING)
414 			return -EPIPE;
415 		else if (!tipc_sk_connected(sk))
416 			return -ENOTCONN;
417 	}
418 	if (!*timeout)
419 		return -EAGAIN;
420 	if (signal_pending(current))
421 		return sock_intr_errno(*timeout);
422 
423 	return 0;
424 }
425 
426 #define tipc_wait_for_cond(sock_, timeo_, condition_)			       \
427 ({                                                                             \
428 	DEFINE_WAIT_FUNC(wait_, woken_wake_function);                          \
429 	struct sock *sk_;						       \
430 	int rc_;							       \
431 									       \
432 	while ((rc_ = !(condition_))) {					       \
433 		/* coupled with smp_wmb() in tipc_sk_proto_rcv() */            \
434 		smp_rmb();                                                     \
435 		sk_ = (sock_)->sk;					       \
436 		rc_ = tipc_sk_sock_err((sock_), timeo_);		       \
437 		if (rc_)						       \
438 			break;						       \
439 		add_wait_queue(sk_sleep(sk_), &wait_);                         \
440 		release_sock(sk_);					       \
441 		*(timeo_) = wait_woken(&wait_, TASK_INTERRUPTIBLE, *(timeo_)); \
442 		sched_annotate_sleep();				               \
443 		lock_sock(sk_);						       \
444 		remove_wait_queue(sk_sleep(sk_), &wait_);		       \
445 	}								       \
446 	rc_;								       \
447 })
448 
449 /**
450  * tipc_sk_create - create a TIPC socket
451  * @net: network namespace (must be default network)
452  * @sock: pre-allocated socket structure
453  * @protocol: protocol indicator (must be 0)
454  * @kern: caused by kernel or by userspace?
455  *
456  * This routine creates additional data structures used by the TIPC socket,
457  * initializes them, and links them together.
458  *
459  * Return: 0 on success, errno otherwise
460  */
461 static int tipc_sk_create(struct net *net, struct socket *sock,
462 			  int protocol, int kern)
463 {
464 	const struct proto_ops *ops;
465 	struct sock *sk;
466 	struct tipc_sock *tsk;
467 	struct tipc_msg *msg;
468 
469 	/* Validate arguments */
470 	if (unlikely(protocol != 0))
471 		return -EPROTONOSUPPORT;
472 
473 	switch (sock->type) {
474 	case SOCK_STREAM:
475 		ops = &stream_ops;
476 		break;
477 	case SOCK_SEQPACKET:
478 		ops = &packet_ops;
479 		break;
480 	case SOCK_DGRAM:
481 	case SOCK_RDM:
482 		ops = &msg_ops;
483 		break;
484 	default:
485 		return -EPROTOTYPE;
486 	}
487 
488 	/* Allocate socket's protocol area */
489 	sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto, kern);
490 	if (sk == NULL)
491 		return -ENOMEM;
492 
493 	tsk = tipc_sk(sk);
494 	tsk->max_pkt = MAX_PKT_DEFAULT;
495 	tsk->maxnagle = 0;
496 	tsk->nagle_start = NAGLE_START_INIT;
497 	INIT_LIST_HEAD(&tsk->publications);
498 	INIT_LIST_HEAD(&tsk->cong_links);
499 	msg = &tsk->phdr;
500 
501 	/* Finish initializing socket data structures */
502 	sock->ops = ops;
503 	sock_init_data(sock, sk);
504 	tipc_set_sk_state(sk, TIPC_OPEN);
505 	if (tipc_sk_insert(tsk)) {
506 		sk_free(sk);
507 		pr_warn("Socket create failed; port number exhausted\n");
508 		return -EINVAL;
509 	}
510 
511 	/* Ensure tsk is visible before we read own_addr. */
512 	smp_mb();
513 
514 	tipc_msg_init(tipc_own_addr(net), msg, TIPC_LOW_IMPORTANCE,
515 		      TIPC_NAMED_MSG, NAMED_H_SIZE, 0);
516 
517 	msg_set_origport(msg, tsk->portid);
518 	timer_setup(&sk->sk_timer, tipc_sk_timeout, 0);
519 	sk->sk_shutdown = 0;
520 	sk->sk_backlog_rcv = tipc_sk_backlog_rcv;
521 	sk->sk_rcvbuf = READ_ONCE(sysctl_tipc_rmem[1]);
522 	sk->sk_data_ready = tipc_data_ready;
523 	sk->sk_write_space = tipc_write_space;
524 	sk->sk_destruct = tipc_sock_destruct;
525 	tsk->conn_timeout = CONN_TIMEOUT_DEFAULT;
526 	tsk->group_is_open = true;
527 	atomic_set(&tsk->dupl_rcvcnt, 0);
528 
529 	/* Start out with safe limits until we receive an advertised window */
530 	tsk->snd_win = tsk_adv_blocks(RCVBUF_MIN);
531 	tsk->rcv_win = tsk->snd_win;
532 
533 	if (tipc_sk_type_connectionless(sk)) {
534 		tsk_set_unreturnable(tsk, true);
535 		if (sock->type == SOCK_DGRAM)
536 			tsk_set_unreliable(tsk, true);
537 	}
538 	__skb_queue_head_init(&tsk->mc_method.deferredq);
539 	trace_tipc_sk_create(sk, NULL, TIPC_DUMP_NONE, " ");
540 	return 0;
541 }
542 
543 static void tipc_sk_callback(struct rcu_head *head)
544 {
545 	struct tipc_sock *tsk = container_of(head, struct tipc_sock, rcu);
546 
547 	sock_put(&tsk->sk);
548 }
549 
550 /* Caller should hold socket lock for the socket. */
551 static void __tipc_shutdown(struct socket *sock, int error)
552 {
553 	struct sock *sk = sock->sk;
554 	struct tipc_sock *tsk = tipc_sk(sk);
555 	struct net *net = sock_net(sk);
556 	long timeout = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT);
557 	u32 dnode = tsk_peer_node(tsk);
558 	struct sk_buff *skb;
559 
560 	/* Avoid that hi-prio shutdown msgs bypass msgs in link wakeup queue */
561 	tipc_wait_for_cond(sock, &timeout, (!tsk->cong_link_cnt &&
562 					    !tsk_conn_cong(tsk)));
563 
564 	/* Push out delayed messages if in Nagle mode */
565 	tipc_sk_push_backlog(tsk, false);
566 	/* Remove pending SYN */
567 	__skb_queue_purge(&sk->sk_write_queue);
568 
569 	/* Remove partially received buffer if any */
570 	skb = skb_peek(&sk->sk_receive_queue);
571 	if (skb && TIPC_SKB_CB(skb)->bytes_read) {
572 		__skb_unlink(skb, &sk->sk_receive_queue);
573 		kfree_skb(skb);
574 	}
575 
576 	/* Reject all unreceived messages if connectionless */
577 	if (tipc_sk_type_connectionless(sk)) {
578 		tsk_rej_rx_queue(sk, error);
579 		return;
580 	}
581 
582 	switch (sk->sk_state) {
583 	case TIPC_CONNECTING:
584 	case TIPC_ESTABLISHED:
585 		tipc_set_sk_state(sk, TIPC_DISCONNECTING);
586 		tipc_node_remove_conn(net, dnode, tsk->portid);
587 		/* Send a FIN+/- to its peer */
588 		skb = __skb_dequeue(&sk->sk_receive_queue);
589 		if (skb) {
590 			__skb_queue_purge(&sk->sk_receive_queue);
591 			tipc_sk_respond(sk, skb, error);
592 			break;
593 		}
594 		skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE,
595 				      TIPC_CONN_MSG, SHORT_H_SIZE, 0, dnode,
596 				      tsk_own_node(tsk), tsk_peer_port(tsk),
597 				      tsk->portid, error);
598 		if (skb)
599 			tipc_node_xmit_skb(net, skb, dnode, tsk->portid);
600 		break;
601 	case TIPC_LISTEN:
602 		/* Reject all SYN messages */
603 		tsk_rej_rx_queue(sk, error);
604 		break;
605 	default:
606 		__skb_queue_purge(&sk->sk_receive_queue);
607 		break;
608 	}
609 }
610 
611 /**
612  * tipc_release - destroy a TIPC socket
613  * @sock: socket to destroy
614  *
615  * This routine cleans up any messages that are still queued on the socket.
616  * For DGRAM and RDM socket types, all queued messages are rejected.
617  * For SEQPACKET and STREAM socket types, the first message is rejected
618  * and any others are discarded.  (If the first message on a STREAM socket
619  * is partially-read, it is discarded and the next one is rejected instead.)
620  *
621  * NOTE: Rejected messages are not necessarily returned to the sender!  They
622  * are returned or discarded according to the "destination droppable" setting
623  * specified for the message by the sender.
624  *
625  * Return: 0 on success, errno otherwise
626  */
627 static int tipc_release(struct socket *sock)
628 {
629 	struct sock *sk = sock->sk;
630 	struct tipc_sock *tsk;
631 
632 	/*
633 	 * Exit if socket isn't fully initialized (occurs when a failed accept()
634 	 * releases a pre-allocated child socket that was never used)
635 	 */
636 	if (sk == NULL)
637 		return 0;
638 
639 	tsk = tipc_sk(sk);
640 	lock_sock(sk);
641 
642 	trace_tipc_sk_release(sk, NULL, TIPC_DUMP_ALL, " ");
643 	__tipc_shutdown(sock, TIPC_ERR_NO_PORT);
644 	sk->sk_shutdown = SHUTDOWN_MASK;
645 	tipc_sk_leave(tsk);
646 	tipc_sk_withdraw(tsk, NULL);
647 	__skb_queue_purge(&tsk->mc_method.deferredq);
648 	sk_stop_timer(sk, &sk->sk_timer);
649 	tipc_sk_remove(tsk);
650 
651 	sock_orphan(sk);
652 	/* Reject any messages that accumulated in backlog queue */
653 	release_sock(sk);
654 	tipc_dest_list_purge(&tsk->cong_links);
655 	tsk->cong_link_cnt = 0;
656 	call_rcu(&tsk->rcu, tipc_sk_callback);
657 	sock->sk = NULL;
658 
659 	return 0;
660 }
661 
662 /**
663  * __tipc_bind - associate or disassocate TIPC name(s) with a socket
664  * @sock: socket structure
665  * @skaddr: socket address describing name(s) and desired operation
666  * @alen: size of socket address data structure
667  *
668  * Name and name sequence binding are indicated using a positive scope value;
669  * a negative scope value unbinds the specified name.  Specifying no name
670  * (i.e. a socket address length of 0) unbinds all names from the socket.
671  *
672  * Return: 0 on success, errno otherwise
673  *
674  * NOTE: This routine doesn't need to take the socket lock since it doesn't
675  *       access any non-constant socket information.
676  */
677 static int __tipc_bind(struct socket *sock, struct sockaddr *skaddr, int alen)
678 {
679 	struct tipc_uaddr *ua = (struct tipc_uaddr *)skaddr;
680 	struct tipc_sock *tsk = tipc_sk(sock->sk);
681 	bool unbind = false;
682 
683 	if (unlikely(!alen))
684 		return tipc_sk_withdraw(tsk, NULL);
685 
686 	if (ua->addrtype == TIPC_SERVICE_ADDR) {
687 		ua->addrtype = TIPC_SERVICE_RANGE;
688 		ua->sr.upper = ua->sr.lower;
689 	}
690 	if (ua->scope < 0) {
691 		unbind = true;
692 		ua->scope = -ua->scope;
693 	}
694 	/* Users may still use deprecated TIPC_ZONE_SCOPE */
695 	if (ua->scope != TIPC_NODE_SCOPE)
696 		ua->scope = TIPC_CLUSTER_SCOPE;
697 
698 	if (tsk->group)
699 		return -EACCES;
700 
701 	if (unbind)
702 		return tipc_sk_withdraw(tsk, ua);
703 	return tipc_sk_publish(tsk, ua);
704 }
705 
706 int tipc_sk_bind(struct socket *sock, struct sockaddr *skaddr, int alen)
707 {
708 	int res;
709 
710 	lock_sock(sock->sk);
711 	res = __tipc_bind(sock, skaddr, alen);
712 	release_sock(sock->sk);
713 	return res;
714 }
715 
716 static int tipc_bind(struct socket *sock, struct sockaddr *skaddr, int alen)
717 {
718 	struct tipc_uaddr *ua = (struct tipc_uaddr *)skaddr;
719 	u32 atype = ua->addrtype;
720 
721 	if (alen) {
722 		if (!tipc_uaddr_valid(ua, alen))
723 			return -EINVAL;
724 		if (atype == TIPC_SOCKET_ADDR)
725 			return -EAFNOSUPPORT;
726 		if (ua->sr.type < TIPC_RESERVED_TYPES) {
727 			pr_warn_once("Can't bind to reserved service type %u\n",
728 				     ua->sr.type);
729 			return -EACCES;
730 		}
731 	}
732 	return tipc_sk_bind(sock, skaddr, alen);
733 }
734 
735 /**
736  * tipc_getname - get port ID of socket or peer socket
737  * @sock: socket structure
738  * @uaddr: area for returned socket address
739  * @peer: 0 = own ID, 1 = current peer ID, 2 = current/former peer ID
740  *
741  * Return: 0 on success, errno otherwise
742  *
743  * NOTE: This routine doesn't need to take the socket lock since it only
744  *       accesses socket information that is unchanging (or which changes in
745  *       a completely predictable manner).
746  */
747 static int tipc_getname(struct socket *sock, struct sockaddr *uaddr,
748 			int peer)
749 {
750 	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
751 	struct sock *sk = sock->sk;
752 	struct tipc_sock *tsk = tipc_sk(sk);
753 
754 	memset(addr, 0, sizeof(*addr));
755 	if (peer) {
756 		if ((!tipc_sk_connected(sk)) &&
757 		    ((peer != 2) || (sk->sk_state != TIPC_DISCONNECTING)))
758 			return -ENOTCONN;
759 		addr->addr.id.ref = tsk_peer_port(tsk);
760 		addr->addr.id.node = tsk_peer_node(tsk);
761 	} else {
762 		addr->addr.id.ref = tsk->portid;
763 		addr->addr.id.node = tipc_own_addr(sock_net(sk));
764 	}
765 
766 	addr->addrtype = TIPC_SOCKET_ADDR;
767 	addr->family = AF_TIPC;
768 	addr->scope = 0;
769 	addr->addr.name.domain = 0;
770 
771 	return sizeof(*addr);
772 }
773 
774 /**
775  * tipc_poll - read and possibly block on pollmask
776  * @file: file structure associated with the socket
777  * @sock: socket for which to calculate the poll bits
778  * @wait: ???
779  *
780  * Return: pollmask value
781  *
782  * COMMENTARY:
783  * It appears that the usual socket locking mechanisms are not useful here
784  * since the pollmask info is potentially out-of-date the moment this routine
785  * exits.  TCP and other protocols seem to rely on higher level poll routines
786  * to handle any preventable race conditions, so TIPC will do the same ...
787  *
788  * IMPORTANT: The fact that a read or write operation is indicated does NOT
789  * imply that the operation will succeed, merely that it should be performed
790  * and will not block.
791  */
792 static __poll_t tipc_poll(struct file *file, struct socket *sock,
793 			      poll_table *wait)
794 {
795 	struct sock *sk = sock->sk;
796 	struct tipc_sock *tsk = tipc_sk(sk);
797 	__poll_t revents = 0;
798 
799 	sock_poll_wait(file, sock, wait);
800 	trace_tipc_sk_poll(sk, NULL, TIPC_DUMP_ALL, " ");
801 
802 	if (sk->sk_shutdown & RCV_SHUTDOWN)
803 		revents |= EPOLLRDHUP | EPOLLIN | EPOLLRDNORM;
804 	if (sk->sk_shutdown == SHUTDOWN_MASK)
805 		revents |= EPOLLHUP;
806 
807 	switch (sk->sk_state) {
808 	case TIPC_ESTABLISHED:
809 		if (!tsk->cong_link_cnt && !tsk_conn_cong(tsk))
810 			revents |= EPOLLOUT;
811 		fallthrough;
812 	case TIPC_LISTEN:
813 	case TIPC_CONNECTING:
814 		if (!skb_queue_empty_lockless(&sk->sk_receive_queue))
815 			revents |= EPOLLIN | EPOLLRDNORM;
816 		break;
817 	case TIPC_OPEN:
818 		if (tsk->group_is_open && !tsk->cong_link_cnt)
819 			revents |= EPOLLOUT;
820 		if (!tipc_sk_type_connectionless(sk))
821 			break;
822 		if (skb_queue_empty_lockless(&sk->sk_receive_queue))
823 			break;
824 		revents |= EPOLLIN | EPOLLRDNORM;
825 		break;
826 	case TIPC_DISCONNECTING:
827 		revents = EPOLLIN | EPOLLRDNORM | EPOLLHUP;
828 		break;
829 	}
830 	return revents;
831 }
832 
833 /**
834  * tipc_sendmcast - send multicast message
835  * @sock: socket structure
836  * @ua: destination address struct
837  * @msg: message to send
838  * @dlen: length of data to send
839  * @timeout: timeout to wait for wakeup
840  *
841  * Called from function tipc_sendmsg(), which has done all sanity checks
842  * Return: the number of bytes sent on success, or errno
843  */
844 static int tipc_sendmcast(struct  socket *sock, struct tipc_uaddr *ua,
845 			  struct msghdr *msg, size_t dlen, long timeout)
846 {
847 	struct sock *sk = sock->sk;
848 	struct tipc_sock *tsk = tipc_sk(sk);
849 	struct tipc_msg *hdr = &tsk->phdr;
850 	struct net *net = sock_net(sk);
851 	int mtu = tipc_bcast_get_mtu(net);
852 	struct sk_buff_head pkts;
853 	struct tipc_nlist dsts;
854 	int rc;
855 
856 	if (tsk->group)
857 		return -EACCES;
858 
859 	/* Block or return if any destination link is congested */
860 	rc = tipc_wait_for_cond(sock, &timeout, !tsk->cong_link_cnt);
861 	if (unlikely(rc))
862 		return rc;
863 
864 	/* Lookup destination nodes */
865 	tipc_nlist_init(&dsts, tipc_own_addr(net));
866 	tipc_nametbl_lookup_mcast_nodes(net, ua, &dsts);
867 	if (!dsts.local && !dsts.remote)
868 		return -EHOSTUNREACH;
869 
870 	/* Build message header */
871 	msg_set_type(hdr, TIPC_MCAST_MSG);
872 	msg_set_hdr_sz(hdr, MCAST_H_SIZE);
873 	msg_set_lookup_scope(hdr, TIPC_CLUSTER_SCOPE);
874 	msg_set_destport(hdr, 0);
875 	msg_set_destnode(hdr, 0);
876 	msg_set_nametype(hdr, ua->sr.type);
877 	msg_set_namelower(hdr, ua->sr.lower);
878 	msg_set_nameupper(hdr, ua->sr.upper);
879 
880 	/* Build message as chain of buffers */
881 	__skb_queue_head_init(&pkts);
882 	rc = tipc_msg_build(hdr, msg, 0, dlen, mtu, &pkts);
883 
884 	/* Send message if build was successful */
885 	if (unlikely(rc == dlen)) {
886 		trace_tipc_sk_sendmcast(sk, skb_peek(&pkts),
887 					TIPC_DUMP_SK_SNDQ, " ");
888 		rc = tipc_mcast_xmit(net, &pkts, &tsk->mc_method, &dsts,
889 				     &tsk->cong_link_cnt);
890 	}
891 
892 	tipc_nlist_purge(&dsts);
893 
894 	return rc ? rc : dlen;
895 }
896 
897 /**
898  * tipc_send_group_msg - send a message to a member in the group
899  * @net: network namespace
900  * @tsk: tipc socket
901  * @m: message to send
902  * @mb: group member
903  * @dnode: destination node
904  * @dport: destination port
905  * @dlen: total length of message data
906  */
907 static int tipc_send_group_msg(struct net *net, struct tipc_sock *tsk,
908 			       struct msghdr *m, struct tipc_member *mb,
909 			       u32 dnode, u32 dport, int dlen)
910 {
911 	u16 bc_snd_nxt = tipc_group_bc_snd_nxt(tsk->group);
912 	struct tipc_mc_method *method = &tsk->mc_method;
913 	int blks = tsk_blocks(GROUP_H_SIZE + dlen);
914 	struct tipc_msg *hdr = &tsk->phdr;
915 	struct sk_buff_head pkts;
916 	int mtu, rc;
917 
918 	/* Complete message header */
919 	msg_set_type(hdr, TIPC_GRP_UCAST_MSG);
920 	msg_set_hdr_sz(hdr, GROUP_H_SIZE);
921 	msg_set_destport(hdr, dport);
922 	msg_set_destnode(hdr, dnode);
923 	msg_set_grp_bc_seqno(hdr, bc_snd_nxt);
924 
925 	/* Build message as chain of buffers */
926 	__skb_queue_head_init(&pkts);
927 	mtu = tipc_node_get_mtu(net, dnode, tsk->portid, false);
928 	rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
929 	if (unlikely(rc != dlen))
930 		return rc;
931 
932 	/* Send message */
933 	rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid);
934 	if (unlikely(rc == -ELINKCONG)) {
935 		tipc_dest_push(&tsk->cong_links, dnode, 0);
936 		tsk->cong_link_cnt++;
937 	}
938 
939 	/* Update send window */
940 	tipc_group_update_member(mb, blks);
941 
942 	/* A broadcast sent within next EXPIRE period must follow same path */
943 	method->rcast = true;
944 	method->mandatory = true;
945 	return dlen;
946 }
947 
948 /**
949  * tipc_send_group_unicast - send message to a member in the group
950  * @sock: socket structure
951  * @m: message to send
952  * @dlen: total length of message data
953  * @timeout: timeout to wait for wakeup
954  *
955  * Called from function tipc_sendmsg(), which has done all sanity checks
956  * Return: the number of bytes sent on success, or errno
957  */
958 static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m,
959 				   int dlen, long timeout)
960 {
961 	struct sock *sk = sock->sk;
962 	struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name;
963 	int blks = tsk_blocks(GROUP_H_SIZE + dlen);
964 	struct tipc_sock *tsk = tipc_sk(sk);
965 	struct net *net = sock_net(sk);
966 	struct tipc_member *mb = NULL;
967 	u32 node, port;
968 	int rc;
969 
970 	node = ua->sk.node;
971 	port = ua->sk.ref;
972 	if (!port && !node)
973 		return -EHOSTUNREACH;
974 
975 	/* Block or return if destination link or member is congested */
976 	rc = tipc_wait_for_cond(sock, &timeout,
977 				!tipc_dest_find(&tsk->cong_links, node, 0) &&
978 				tsk->group &&
979 				!tipc_group_cong(tsk->group, node, port, blks,
980 						 &mb));
981 	if (unlikely(rc))
982 		return rc;
983 
984 	if (unlikely(!mb))
985 		return -EHOSTUNREACH;
986 
987 	rc = tipc_send_group_msg(net, tsk, m, mb, node, port, dlen);
988 
989 	return rc ? rc : dlen;
990 }
991 
992 /**
993  * tipc_send_group_anycast - send message to any member with given identity
994  * @sock: socket structure
995  * @m: message to send
996  * @dlen: total length of message data
997  * @timeout: timeout to wait for wakeup
998  *
999  * Called from function tipc_sendmsg(), which has done all sanity checks
1000  * Return: the number of bytes sent on success, or errno
1001  */
1002 static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
1003 				   int dlen, long timeout)
1004 {
1005 	struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name;
1006 	struct sock *sk = sock->sk;
1007 	struct tipc_sock *tsk = tipc_sk(sk);
1008 	struct list_head *cong_links = &tsk->cong_links;
1009 	int blks = tsk_blocks(GROUP_H_SIZE + dlen);
1010 	struct tipc_msg *hdr = &tsk->phdr;
1011 	struct tipc_member *first = NULL;
1012 	struct tipc_member *mbr = NULL;
1013 	struct net *net = sock_net(sk);
1014 	u32 node, port, exclude;
1015 	struct list_head dsts;
1016 	int lookups = 0;
1017 	int dstcnt, rc;
1018 	bool cong;
1019 
1020 	INIT_LIST_HEAD(&dsts);
1021 	ua->sa.type = msg_nametype(hdr);
1022 	ua->scope = msg_lookup_scope(hdr);
1023 
1024 	while (++lookups < 4) {
1025 		exclude = tipc_group_exclude(tsk->group);
1026 
1027 		first = NULL;
1028 
1029 		/* Look for a non-congested destination member, if any */
1030 		while (1) {
1031 			if (!tipc_nametbl_lookup_group(net, ua, &dsts, &dstcnt,
1032 						       exclude, false))
1033 				return -EHOSTUNREACH;
1034 			tipc_dest_pop(&dsts, &node, &port);
1035 			cong = tipc_group_cong(tsk->group, node, port, blks,
1036 					       &mbr);
1037 			if (!cong)
1038 				break;
1039 			if (mbr == first)
1040 				break;
1041 			if (!first)
1042 				first = mbr;
1043 		}
1044 
1045 		/* Start over if destination was not in member list */
1046 		if (unlikely(!mbr))
1047 			continue;
1048 
1049 		if (likely(!cong && !tipc_dest_find(cong_links, node, 0)))
1050 			break;
1051 
1052 		/* Block or return if destination link or member is congested */
1053 		rc = tipc_wait_for_cond(sock, &timeout,
1054 					!tipc_dest_find(cong_links, node, 0) &&
1055 					tsk->group &&
1056 					!tipc_group_cong(tsk->group, node, port,
1057 							 blks, &mbr));
1058 		if (unlikely(rc))
1059 			return rc;
1060 
1061 		/* Send, unless destination disappeared while waiting */
1062 		if (likely(mbr))
1063 			break;
1064 	}
1065 
1066 	if (unlikely(lookups >= 4))
1067 		return -EHOSTUNREACH;
1068 
1069 	rc = tipc_send_group_msg(net, tsk, m, mbr, node, port, dlen);
1070 
1071 	return rc ? rc : dlen;
1072 }
1073 
1074 /**
1075  * tipc_send_group_bcast - send message to all members in communication group
1076  * @sock: socket structure
1077  * @m: message to send
1078  * @dlen: total length of message data
1079  * @timeout: timeout to wait for wakeup
1080  *
1081  * Called from function tipc_sendmsg(), which has done all sanity checks
1082  * Return: the number of bytes sent on success, or errno
1083  */
1084 static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
1085 				 int dlen, long timeout)
1086 {
1087 	struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name;
1088 	struct sock *sk = sock->sk;
1089 	struct net *net = sock_net(sk);
1090 	struct tipc_sock *tsk = tipc_sk(sk);
1091 	struct tipc_nlist *dsts;
1092 	struct tipc_mc_method *method = &tsk->mc_method;
1093 	bool ack = method->mandatory && method->rcast;
1094 	int blks = tsk_blocks(MCAST_H_SIZE + dlen);
1095 	struct tipc_msg *hdr = &tsk->phdr;
1096 	int mtu = tipc_bcast_get_mtu(net);
1097 	struct sk_buff_head pkts;
1098 	int rc = -EHOSTUNREACH;
1099 
1100 	/* Block or return if any destination link or member is congested */
1101 	rc = tipc_wait_for_cond(sock, &timeout,
1102 				!tsk->cong_link_cnt && tsk->group &&
1103 				!tipc_group_bc_cong(tsk->group, blks));
1104 	if (unlikely(rc))
1105 		return rc;
1106 
1107 	dsts = tipc_group_dests(tsk->group);
1108 	if (!dsts->local && !dsts->remote)
1109 		return -EHOSTUNREACH;
1110 
1111 	/* Complete message header */
1112 	if (ua) {
1113 		msg_set_type(hdr, TIPC_GRP_MCAST_MSG);
1114 		msg_set_nameinst(hdr, ua->sa.instance);
1115 	} else {
1116 		msg_set_type(hdr, TIPC_GRP_BCAST_MSG);
1117 		msg_set_nameinst(hdr, 0);
1118 	}
1119 	msg_set_hdr_sz(hdr, GROUP_H_SIZE);
1120 	msg_set_destport(hdr, 0);
1121 	msg_set_destnode(hdr, 0);
1122 	msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(tsk->group));
1123 
1124 	/* Avoid getting stuck with repeated forced replicasts */
1125 	msg_set_grp_bc_ack_req(hdr, ack);
1126 
1127 	/* Build message as chain of buffers */
1128 	__skb_queue_head_init(&pkts);
1129 	rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
1130 	if (unlikely(rc != dlen))
1131 		return rc;
1132 
1133 	/* Send message */
1134 	rc = tipc_mcast_xmit(net, &pkts, method, dsts, &tsk->cong_link_cnt);
1135 	if (unlikely(rc))
1136 		return rc;
1137 
1138 	/* Update broadcast sequence number and send windows */
1139 	tipc_group_update_bc_members(tsk->group, blks, ack);
1140 
1141 	/* Broadcast link is now free to choose method for next broadcast */
1142 	method->mandatory = false;
1143 	method->expires = jiffies;
1144 
1145 	return dlen;
1146 }
1147 
1148 /**
1149  * tipc_send_group_mcast - send message to all members with given identity
1150  * @sock: socket structure
1151  * @m: message to send
1152  * @dlen: total length of message data
1153  * @timeout: timeout to wait for wakeup
1154  *
1155  * Called from function tipc_sendmsg(), which has done all sanity checks
1156  * Return: the number of bytes sent on success, or errno
1157  */
1158 static int tipc_send_group_mcast(struct socket *sock, struct msghdr *m,
1159 				 int dlen, long timeout)
1160 {
1161 	struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name;
1162 	struct sock *sk = sock->sk;
1163 	struct tipc_sock *tsk = tipc_sk(sk);
1164 	struct tipc_group *grp = tsk->group;
1165 	struct tipc_msg *hdr = &tsk->phdr;
1166 	struct net *net = sock_net(sk);
1167 	struct list_head dsts;
1168 	u32 dstcnt, exclude;
1169 
1170 	INIT_LIST_HEAD(&dsts);
1171 	ua->sa.type = msg_nametype(hdr);
1172 	ua->scope = msg_lookup_scope(hdr);
1173 	exclude = tipc_group_exclude(grp);
1174 
1175 	if (!tipc_nametbl_lookup_group(net, ua, &dsts, &dstcnt, exclude, true))
1176 		return -EHOSTUNREACH;
1177 
1178 	if (dstcnt == 1) {
1179 		tipc_dest_pop(&dsts, &ua->sk.node, &ua->sk.ref);
1180 		return tipc_send_group_unicast(sock, m, dlen, timeout);
1181 	}
1182 
1183 	tipc_dest_list_purge(&dsts);
1184 	return tipc_send_group_bcast(sock, m, dlen, timeout);
1185 }
1186 
1187 /**
1188  * tipc_sk_mcast_rcv - Deliver multicast messages to all destination sockets
1189  * @net: the associated network namespace
1190  * @arrvq: queue with arriving messages, to be cloned after destination lookup
1191  * @inputq: queue with cloned messages, delivered to socket after dest lookup
1192  *
1193  * Multi-threaded: parallel calls with reference to same queues may occur
1194  */
1195 void tipc_sk_mcast_rcv(struct net *net, struct sk_buff_head *arrvq,
1196 		       struct sk_buff_head *inputq)
1197 {
1198 	u32 self = tipc_own_addr(net);
1199 	struct sk_buff *skb, *_skb;
1200 	u32 portid, onode;
1201 	struct sk_buff_head tmpq;
1202 	struct list_head dports;
1203 	struct tipc_msg *hdr;
1204 	struct tipc_uaddr ua;
1205 	int user, mtyp, hlen;
1206 
1207 	__skb_queue_head_init(&tmpq);
1208 	INIT_LIST_HEAD(&dports);
1209 	ua.addrtype = TIPC_SERVICE_RANGE;
1210 
1211 	/* tipc_skb_peek() increments the head skb's reference counter */
1212 	skb = tipc_skb_peek(arrvq, &inputq->lock);
1213 	for (; skb; skb = tipc_skb_peek(arrvq, &inputq->lock)) {
1214 		hdr = buf_msg(skb);
1215 		user = msg_user(hdr);
1216 		mtyp = msg_type(hdr);
1217 		hlen = skb_headroom(skb) + msg_hdr_sz(hdr);
1218 		onode = msg_orignode(hdr);
1219 		ua.sr.type = msg_nametype(hdr);
1220 		ua.sr.lower = msg_namelower(hdr);
1221 		ua.sr.upper = msg_nameupper(hdr);
1222 		if (onode == self)
1223 			ua.scope = TIPC_ANY_SCOPE;
1224 		else
1225 			ua.scope = TIPC_CLUSTER_SCOPE;
1226 
1227 		if (mtyp == TIPC_GRP_UCAST_MSG || user == GROUP_PROTOCOL) {
1228 			spin_lock_bh(&inputq->lock);
1229 			if (skb_peek(arrvq) == skb) {
1230 				__skb_dequeue(arrvq);
1231 				__skb_queue_tail(inputq, skb);
1232 			}
1233 			kfree_skb(skb);
1234 			spin_unlock_bh(&inputq->lock);
1235 			continue;
1236 		}
1237 
1238 		/* Group messages require exact scope match */
1239 		if (msg_in_group(hdr)) {
1240 			ua.sr.lower = 0;
1241 			ua.sr.upper = ~0;
1242 			ua.scope = msg_lookup_scope(hdr);
1243 		}
1244 
1245 		/* Create destination port list: */
1246 		tipc_nametbl_lookup_mcast_sockets(net, &ua, &dports);
1247 
1248 		/* Clone message per destination */
1249 		while (tipc_dest_pop(&dports, NULL, &portid)) {
1250 			_skb = __pskb_copy(skb, hlen, GFP_ATOMIC);
1251 			if (_skb) {
1252 				msg_set_destport(buf_msg(_skb), portid);
1253 				__skb_queue_tail(&tmpq, _skb);
1254 				continue;
1255 			}
1256 			pr_warn("Failed to clone mcast rcv buffer\n");
1257 		}
1258 		/* Append clones to inputq only if skb is still head of arrvq */
1259 		spin_lock_bh(&inputq->lock);
1260 		if (skb_peek(arrvq) == skb) {
1261 			skb_queue_splice_tail_init(&tmpq, inputq);
1262 			/* Decrement the skb's refcnt */
1263 			kfree_skb(__skb_dequeue(arrvq));
1264 		}
1265 		spin_unlock_bh(&inputq->lock);
1266 		__skb_queue_purge(&tmpq);
1267 		kfree_skb(skb);
1268 	}
1269 	tipc_sk_rcv(net, inputq);
1270 }
1271 
1272 /* tipc_sk_push_backlog(): send accumulated buffers in socket write queue
1273  *                         when socket is in Nagle mode
1274  */
1275 static void tipc_sk_push_backlog(struct tipc_sock *tsk, bool nagle_ack)
1276 {
1277 	struct sk_buff_head *txq = &tsk->sk.sk_write_queue;
1278 	struct sk_buff *skb = skb_peek_tail(txq);
1279 	struct net *net = sock_net(&tsk->sk);
1280 	u32 dnode = tsk_peer_node(tsk);
1281 	int rc;
1282 
1283 	if (nagle_ack) {
1284 		tsk->pkt_cnt += skb_queue_len(txq);
1285 		if (!tsk->pkt_cnt || tsk->msg_acc / tsk->pkt_cnt < 2) {
1286 			tsk->oneway = 0;
1287 			if (tsk->nagle_start < NAGLE_START_MAX)
1288 				tsk->nagle_start *= 2;
1289 			tsk->expect_ack = false;
1290 			pr_debug("tsk %10u: bad nagle %u -> %u, next start %u!\n",
1291 				 tsk->portid, tsk->msg_acc, tsk->pkt_cnt,
1292 				 tsk->nagle_start);
1293 		} else {
1294 			tsk->nagle_start = NAGLE_START_INIT;
1295 			if (skb) {
1296 				msg_set_ack_required(buf_msg(skb));
1297 				tsk->expect_ack = true;
1298 			} else {
1299 				tsk->expect_ack = false;
1300 			}
1301 		}
1302 		tsk->msg_acc = 0;
1303 		tsk->pkt_cnt = 0;
1304 	}
1305 
1306 	if (!skb || tsk->cong_link_cnt)
1307 		return;
1308 
1309 	/* Do not send SYN again after congestion */
1310 	if (msg_is_syn(buf_msg(skb)))
1311 		return;
1312 
1313 	if (tsk->msg_acc)
1314 		tsk->pkt_cnt += skb_queue_len(txq);
1315 	tsk->snt_unacked += tsk->snd_backlog;
1316 	tsk->snd_backlog = 0;
1317 	rc = tipc_node_xmit(net, txq, dnode, tsk->portid);
1318 	if (rc == -ELINKCONG)
1319 		tsk->cong_link_cnt = 1;
1320 }
1321 
1322 /**
1323  * tipc_sk_conn_proto_rcv - receive a connection mng protocol message
1324  * @tsk: receiving socket
1325  * @skb: pointer to message buffer.
1326  * @inputq: buffer list containing the buffers
1327  * @xmitq: output message area
1328  */
1329 static void tipc_sk_conn_proto_rcv(struct tipc_sock *tsk, struct sk_buff *skb,
1330 				   struct sk_buff_head *inputq,
1331 				   struct sk_buff_head *xmitq)
1332 {
1333 	struct tipc_msg *hdr = buf_msg(skb);
1334 	u32 onode = tsk_own_node(tsk);
1335 	struct sock *sk = &tsk->sk;
1336 	int mtyp = msg_type(hdr);
1337 	bool was_cong;
1338 
1339 	/* Ignore if connection cannot be validated: */
1340 	if (!tsk_peer_msg(tsk, hdr)) {
1341 		trace_tipc_sk_drop_msg(sk, skb, TIPC_DUMP_NONE, "@proto_rcv!");
1342 		goto exit;
1343 	}
1344 
1345 	if (unlikely(msg_errcode(hdr))) {
1346 		tipc_set_sk_state(sk, TIPC_DISCONNECTING);
1347 		tipc_node_remove_conn(sock_net(sk), tsk_peer_node(tsk),
1348 				      tsk_peer_port(tsk));
1349 		sk->sk_state_change(sk);
1350 
1351 		/* State change is ignored if socket already awake,
1352 		 * - convert msg to abort msg and add to inqueue
1353 		 */
1354 		msg_set_user(hdr, TIPC_CRITICAL_IMPORTANCE);
1355 		msg_set_type(hdr, TIPC_CONN_MSG);
1356 		msg_set_size(hdr, BASIC_H_SIZE);
1357 		msg_set_hdr_sz(hdr, BASIC_H_SIZE);
1358 		__skb_queue_tail(inputq, skb);
1359 		return;
1360 	}
1361 
1362 	tsk->probe_unacked = false;
1363 
1364 	if (mtyp == CONN_PROBE) {
1365 		msg_set_type(hdr, CONN_PROBE_REPLY);
1366 		if (tipc_msg_reverse(onode, &skb, TIPC_OK))
1367 			__skb_queue_tail(xmitq, skb);
1368 		return;
1369 	} else if (mtyp == CONN_ACK) {
1370 		was_cong = tsk_conn_cong(tsk);
1371 		tipc_sk_push_backlog(tsk, msg_nagle_ack(hdr));
1372 		tsk->snt_unacked -= msg_conn_ack(hdr);
1373 		if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL)
1374 			tsk->snd_win = msg_adv_win(hdr);
1375 		if (was_cong && !tsk_conn_cong(tsk))
1376 			sk->sk_write_space(sk);
1377 	} else if (mtyp != CONN_PROBE_REPLY) {
1378 		pr_warn("Received unknown CONN_PROTO msg\n");
1379 	}
1380 exit:
1381 	kfree_skb(skb);
1382 }
1383 
1384 /**
1385  * tipc_sendmsg - send message in connectionless manner
1386  * @sock: socket structure
1387  * @m: message to send
1388  * @dsz: amount of user data to be sent
1389  *
1390  * Message must have an destination specified explicitly.
1391  * Used for SOCK_RDM and SOCK_DGRAM messages,
1392  * and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections.
1393  * (Note: 'SYN+' is prohibited on SOCK_STREAM.)
1394  *
1395  * Return: the number of bytes sent on success, or errno otherwise
1396  */
1397 static int tipc_sendmsg(struct socket *sock,
1398 			struct msghdr *m, size_t dsz)
1399 {
1400 	struct sock *sk = sock->sk;
1401 	int ret;
1402 
1403 	lock_sock(sk);
1404 	ret = __tipc_sendmsg(sock, m, dsz);
1405 	release_sock(sk);
1406 
1407 	return ret;
1408 }
1409 
1410 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dlen)
1411 {
1412 	struct sock *sk = sock->sk;
1413 	struct net *net = sock_net(sk);
1414 	struct tipc_sock *tsk = tipc_sk(sk);
1415 	struct tipc_uaddr *ua = (struct tipc_uaddr *)m->msg_name;
1416 	long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
1417 	struct list_head *clinks = &tsk->cong_links;
1418 	bool syn = !tipc_sk_type_connectionless(sk);
1419 	struct tipc_group *grp = tsk->group;
1420 	struct tipc_msg *hdr = &tsk->phdr;
1421 	struct tipc_socket_addr skaddr;
1422 	struct sk_buff_head pkts;
1423 	int atype, mtu, rc;
1424 
1425 	if (unlikely(dlen > TIPC_MAX_USER_MSG_SIZE))
1426 		return -EMSGSIZE;
1427 
1428 	if (ua) {
1429 		if (!tipc_uaddr_valid(ua, m->msg_namelen))
1430 			return -EINVAL;
1431 		atype = ua->addrtype;
1432 	}
1433 
1434 	/* If socket belongs to a communication group follow other paths */
1435 	if (grp) {
1436 		if (!ua)
1437 			return tipc_send_group_bcast(sock, m, dlen, timeout);
1438 		if (atype == TIPC_SERVICE_ADDR)
1439 			return tipc_send_group_anycast(sock, m, dlen, timeout);
1440 		if (atype == TIPC_SOCKET_ADDR)
1441 			return tipc_send_group_unicast(sock, m, dlen, timeout);
1442 		if (atype == TIPC_SERVICE_RANGE)
1443 			return tipc_send_group_mcast(sock, m, dlen, timeout);
1444 		return -EINVAL;
1445 	}
1446 
1447 	if (!ua) {
1448 		ua = (struct tipc_uaddr *)&tsk->peer;
1449 		if (!syn && ua->family != AF_TIPC)
1450 			return -EDESTADDRREQ;
1451 		atype = ua->addrtype;
1452 	}
1453 
1454 	if (unlikely(syn)) {
1455 		if (sk->sk_state == TIPC_LISTEN)
1456 			return -EPIPE;
1457 		if (sk->sk_state != TIPC_OPEN)
1458 			return -EISCONN;
1459 		if (tsk->published)
1460 			return -EOPNOTSUPP;
1461 		if (atype == TIPC_SERVICE_ADDR)
1462 			tsk->conn_addrtype = atype;
1463 		msg_set_syn(hdr, 1);
1464 	}
1465 
1466 	memset(&skaddr, 0, sizeof(skaddr));
1467 
1468 	/* Determine destination */
1469 	if (atype == TIPC_SERVICE_RANGE) {
1470 		return tipc_sendmcast(sock, ua, m, dlen, timeout);
1471 	} else if (atype == TIPC_SERVICE_ADDR) {
1472 		skaddr.node = ua->lookup_node;
1473 		ua->scope = tipc_node2scope(skaddr.node);
1474 		if (!tipc_nametbl_lookup_anycast(net, ua, &skaddr))
1475 			return -EHOSTUNREACH;
1476 	} else if (atype == TIPC_SOCKET_ADDR) {
1477 		skaddr = ua->sk;
1478 	} else {
1479 		return -EINVAL;
1480 	}
1481 
1482 	/* Block or return if destination link is congested */
1483 	rc = tipc_wait_for_cond(sock, &timeout,
1484 				!tipc_dest_find(clinks, skaddr.node, 0));
1485 	if (unlikely(rc))
1486 		return rc;
1487 
1488 	/* Finally build message header */
1489 	msg_set_destnode(hdr, skaddr.node);
1490 	msg_set_destport(hdr, skaddr.ref);
1491 	if (atype == TIPC_SERVICE_ADDR) {
1492 		msg_set_type(hdr, TIPC_NAMED_MSG);
1493 		msg_set_hdr_sz(hdr, NAMED_H_SIZE);
1494 		msg_set_nametype(hdr, ua->sa.type);
1495 		msg_set_nameinst(hdr, ua->sa.instance);
1496 		msg_set_lookup_scope(hdr, ua->scope);
1497 	} else { /* TIPC_SOCKET_ADDR */
1498 		msg_set_type(hdr, TIPC_DIRECT_MSG);
1499 		msg_set_lookup_scope(hdr, 0);
1500 		msg_set_hdr_sz(hdr, BASIC_H_SIZE);
1501 	}
1502 
1503 	/* Add message body */
1504 	__skb_queue_head_init(&pkts);
1505 	mtu = tipc_node_get_mtu(net, skaddr.node, tsk->portid, true);
1506 	rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
1507 	if (unlikely(rc != dlen))
1508 		return rc;
1509 	if (unlikely(syn && !tipc_msg_skb_clone(&pkts, &sk->sk_write_queue))) {
1510 		__skb_queue_purge(&pkts);
1511 		return -ENOMEM;
1512 	}
1513 
1514 	/* Send message */
1515 	trace_tipc_sk_sendmsg(sk, skb_peek(&pkts), TIPC_DUMP_SK_SNDQ, " ");
1516 	rc = tipc_node_xmit(net, &pkts, skaddr.node, tsk->portid);
1517 	if (unlikely(rc == -ELINKCONG)) {
1518 		tipc_dest_push(clinks, skaddr.node, 0);
1519 		tsk->cong_link_cnt++;
1520 		rc = 0;
1521 	}
1522 
1523 	if (unlikely(syn && !rc)) {
1524 		tipc_set_sk_state(sk, TIPC_CONNECTING);
1525 		if (dlen && timeout) {
1526 			timeout = msecs_to_jiffies(timeout);
1527 			tipc_wait_for_connect(sock, &timeout);
1528 		}
1529 	}
1530 
1531 	return rc ? rc : dlen;
1532 }
1533 
1534 /**
1535  * tipc_sendstream - send stream-oriented data
1536  * @sock: socket structure
1537  * @m: data to send
1538  * @dsz: total length of data to be transmitted
1539  *
1540  * Used for SOCK_STREAM data.
1541  *
1542  * Return: the number of bytes sent on success (or partial success),
1543  * or errno if no data sent
1544  */
1545 static int tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz)
1546 {
1547 	struct sock *sk = sock->sk;
1548 	int ret;
1549 
1550 	lock_sock(sk);
1551 	ret = __tipc_sendstream(sock, m, dsz);
1552 	release_sock(sk);
1553 
1554 	return ret;
1555 }
1556 
1557 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dlen)
1558 {
1559 	struct sock *sk = sock->sk;
1560 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
1561 	long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
1562 	struct sk_buff_head *txq = &sk->sk_write_queue;
1563 	struct tipc_sock *tsk = tipc_sk(sk);
1564 	struct tipc_msg *hdr = &tsk->phdr;
1565 	struct net *net = sock_net(sk);
1566 	struct sk_buff *skb;
1567 	u32 dnode = tsk_peer_node(tsk);
1568 	int maxnagle = tsk->maxnagle;
1569 	int maxpkt = tsk->max_pkt;
1570 	int send, sent = 0;
1571 	int blocks, rc = 0;
1572 
1573 	if (unlikely(dlen > INT_MAX))
1574 		return -EMSGSIZE;
1575 
1576 	/* Handle implicit connection setup */
1577 	if (unlikely(dest && sk->sk_state == TIPC_OPEN)) {
1578 		rc = __tipc_sendmsg(sock, m, dlen);
1579 		if (dlen && dlen == rc) {
1580 			tsk->peer_caps = tipc_node_get_capabilities(net, dnode);
1581 			tsk->snt_unacked = tsk_inc(tsk, dlen + msg_hdr_sz(hdr));
1582 		}
1583 		return rc;
1584 	}
1585 
1586 	do {
1587 		rc = tipc_wait_for_cond(sock, &timeout,
1588 					(!tsk->cong_link_cnt &&
1589 					 !tsk_conn_cong(tsk) &&
1590 					 tipc_sk_connected(sk)));
1591 		if (unlikely(rc))
1592 			break;
1593 		send = min_t(size_t, dlen - sent, TIPC_MAX_USER_MSG_SIZE);
1594 		blocks = tsk->snd_backlog;
1595 		if (tsk->oneway++ >= tsk->nagle_start && maxnagle &&
1596 		    send <= maxnagle) {
1597 			rc = tipc_msg_append(hdr, m, send, maxnagle, txq);
1598 			if (unlikely(rc < 0))
1599 				break;
1600 			blocks += rc;
1601 			tsk->msg_acc++;
1602 			if (blocks <= 64 && tsk->expect_ack) {
1603 				tsk->snd_backlog = blocks;
1604 				sent += send;
1605 				break;
1606 			} else if (blocks > 64) {
1607 				tsk->pkt_cnt += skb_queue_len(txq);
1608 			} else {
1609 				skb = skb_peek_tail(txq);
1610 				if (skb) {
1611 					msg_set_ack_required(buf_msg(skb));
1612 					tsk->expect_ack = true;
1613 				} else {
1614 					tsk->expect_ack = false;
1615 				}
1616 				tsk->msg_acc = 0;
1617 				tsk->pkt_cnt = 0;
1618 			}
1619 		} else {
1620 			rc = tipc_msg_build(hdr, m, sent, send, maxpkt, txq);
1621 			if (unlikely(rc != send))
1622 				break;
1623 			blocks += tsk_inc(tsk, send + MIN_H_SIZE);
1624 		}
1625 		trace_tipc_sk_sendstream(sk, skb_peek(txq),
1626 					 TIPC_DUMP_SK_SNDQ, " ");
1627 		rc = tipc_node_xmit(net, txq, dnode, tsk->portid);
1628 		if (unlikely(rc == -ELINKCONG)) {
1629 			tsk->cong_link_cnt = 1;
1630 			rc = 0;
1631 		}
1632 		if (likely(!rc)) {
1633 			tsk->snt_unacked += blocks;
1634 			tsk->snd_backlog = 0;
1635 			sent += send;
1636 		}
1637 	} while (sent < dlen && !rc);
1638 
1639 	return sent ? sent : rc;
1640 }
1641 
1642 /**
1643  * tipc_send_packet - send a connection-oriented message
1644  * @sock: socket structure
1645  * @m: message to send
1646  * @dsz: length of data to be transmitted
1647  *
1648  * Used for SOCK_SEQPACKET messages.
1649  *
1650  * Return: the number of bytes sent on success, or errno otherwise
1651  */
1652 static int tipc_send_packet(struct socket *sock, struct msghdr *m, size_t dsz)
1653 {
1654 	if (dsz > TIPC_MAX_USER_MSG_SIZE)
1655 		return -EMSGSIZE;
1656 
1657 	return tipc_sendstream(sock, m, dsz);
1658 }
1659 
1660 /* tipc_sk_finish_conn - complete the setup of a connection
1661  */
1662 static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port,
1663 				u32 peer_node)
1664 {
1665 	struct sock *sk = &tsk->sk;
1666 	struct net *net = sock_net(sk);
1667 	struct tipc_msg *msg = &tsk->phdr;
1668 
1669 	msg_set_syn(msg, 0);
1670 	msg_set_destnode(msg, peer_node);
1671 	msg_set_destport(msg, peer_port);
1672 	msg_set_type(msg, TIPC_CONN_MSG);
1673 	msg_set_lookup_scope(msg, 0);
1674 	msg_set_hdr_sz(msg, SHORT_H_SIZE);
1675 
1676 	sk_reset_timer(sk, &sk->sk_timer, jiffies + CONN_PROBING_INTV);
1677 	tipc_set_sk_state(sk, TIPC_ESTABLISHED);
1678 	tipc_node_add_conn(net, peer_node, tsk->portid, peer_port);
1679 	tsk->max_pkt = tipc_node_get_mtu(net, peer_node, tsk->portid, true);
1680 	tsk->peer_caps = tipc_node_get_capabilities(net, peer_node);
1681 	tsk_set_nagle(tsk);
1682 	__skb_queue_purge(&sk->sk_write_queue);
1683 	if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL)
1684 		return;
1685 
1686 	/* Fall back to message based flow control */
1687 	tsk->rcv_win = FLOWCTL_MSG_WIN;
1688 	tsk->snd_win = FLOWCTL_MSG_WIN;
1689 }
1690 
1691 /**
1692  * tipc_sk_set_orig_addr - capture sender's address for received message
1693  * @m: descriptor for message info
1694  * @skb: received message
1695  *
1696  * Note: Address is not captured if not requested by receiver.
1697  */
1698 static void tipc_sk_set_orig_addr(struct msghdr *m, struct sk_buff *skb)
1699 {
1700 	DECLARE_SOCKADDR(struct sockaddr_pair *, srcaddr, m->msg_name);
1701 	struct tipc_msg *hdr = buf_msg(skb);
1702 
1703 	if (!srcaddr)
1704 		return;
1705 
1706 	srcaddr->sock.family = AF_TIPC;
1707 	srcaddr->sock.addrtype = TIPC_SOCKET_ADDR;
1708 	srcaddr->sock.scope = 0;
1709 	srcaddr->sock.addr.id.ref = msg_origport(hdr);
1710 	srcaddr->sock.addr.id.node = msg_orignode(hdr);
1711 	srcaddr->sock.addr.name.domain = 0;
1712 	m->msg_namelen = sizeof(struct sockaddr_tipc);
1713 
1714 	if (!msg_in_group(hdr))
1715 		return;
1716 
1717 	/* Group message users may also want to know sending member's id */
1718 	srcaddr->member.family = AF_TIPC;
1719 	srcaddr->member.addrtype = TIPC_SERVICE_ADDR;
1720 	srcaddr->member.scope = 0;
1721 	srcaddr->member.addr.name.name.type = msg_nametype(hdr);
1722 	srcaddr->member.addr.name.name.instance = TIPC_SKB_CB(skb)->orig_member;
1723 	srcaddr->member.addr.name.domain = 0;
1724 	m->msg_namelen = sizeof(*srcaddr);
1725 }
1726 
1727 /**
1728  * tipc_sk_anc_data_recv - optionally capture ancillary data for received message
1729  * @m: descriptor for message info
1730  * @skb: received message buffer
1731  * @tsk: TIPC port associated with message
1732  *
1733  * Note: Ancillary data is not captured if not requested by receiver.
1734  *
1735  * Return: 0 if successful, otherwise errno
1736  */
1737 static int tipc_sk_anc_data_recv(struct msghdr *m, struct sk_buff *skb,
1738 				 struct tipc_sock *tsk)
1739 {
1740 	struct tipc_msg *hdr;
1741 	u32 data[3] = {0,};
1742 	bool has_addr;
1743 	int dlen, rc;
1744 
1745 	if (likely(m->msg_controllen == 0))
1746 		return 0;
1747 
1748 	hdr = buf_msg(skb);
1749 	dlen = msg_data_sz(hdr);
1750 
1751 	/* Capture errored message object, if any */
1752 	if (msg_errcode(hdr)) {
1753 		if (skb_linearize(skb))
1754 			return -ENOMEM;
1755 		hdr = buf_msg(skb);
1756 		data[0] = msg_errcode(hdr);
1757 		data[1] = dlen;
1758 		rc = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, data);
1759 		if (rc || !dlen)
1760 			return rc;
1761 		rc = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, dlen, msg_data(hdr));
1762 		if (rc)
1763 			return rc;
1764 	}
1765 
1766 	/* Capture TIPC_SERVICE_ADDR/RANGE destination address, if any */
1767 	switch (msg_type(hdr)) {
1768 	case TIPC_NAMED_MSG:
1769 		has_addr = true;
1770 		data[0] = msg_nametype(hdr);
1771 		data[1] = msg_namelower(hdr);
1772 		data[2] = data[1];
1773 		break;
1774 	case TIPC_MCAST_MSG:
1775 		has_addr = true;
1776 		data[0] = msg_nametype(hdr);
1777 		data[1] = msg_namelower(hdr);
1778 		data[2] = msg_nameupper(hdr);
1779 		break;
1780 	case TIPC_CONN_MSG:
1781 		has_addr = !!tsk->conn_addrtype;
1782 		data[0] = msg_nametype(&tsk->phdr);
1783 		data[1] = msg_nameinst(&tsk->phdr);
1784 		data[2] = data[1];
1785 		break;
1786 	default:
1787 		has_addr = false;
1788 	}
1789 	if (!has_addr)
1790 		return 0;
1791 	return put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, data);
1792 }
1793 
1794 static struct sk_buff *tipc_sk_build_ack(struct tipc_sock *tsk)
1795 {
1796 	struct sock *sk = &tsk->sk;
1797 	struct sk_buff *skb = NULL;
1798 	struct tipc_msg *msg;
1799 	u32 peer_port = tsk_peer_port(tsk);
1800 	u32 dnode = tsk_peer_node(tsk);
1801 
1802 	if (!tipc_sk_connected(sk))
1803 		return NULL;
1804 	skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0,
1805 			      dnode, tsk_own_node(tsk), peer_port,
1806 			      tsk->portid, TIPC_OK);
1807 	if (!skb)
1808 		return NULL;
1809 	msg = buf_msg(skb);
1810 	msg_set_conn_ack(msg, tsk->rcv_unacked);
1811 	tsk->rcv_unacked = 0;
1812 
1813 	/* Adjust to and advertize the correct window limit */
1814 	if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) {
1815 		tsk->rcv_win = tsk_adv_blocks(tsk->sk.sk_rcvbuf);
1816 		msg_set_adv_win(msg, tsk->rcv_win);
1817 	}
1818 	return skb;
1819 }
1820 
1821 static void tipc_sk_send_ack(struct tipc_sock *tsk)
1822 {
1823 	struct sk_buff *skb;
1824 
1825 	skb = tipc_sk_build_ack(tsk);
1826 	if (!skb)
1827 		return;
1828 
1829 	tipc_node_xmit_skb(sock_net(&tsk->sk), skb, tsk_peer_node(tsk),
1830 			   msg_link_selector(buf_msg(skb)));
1831 }
1832 
1833 static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop)
1834 {
1835 	struct sock *sk = sock->sk;
1836 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
1837 	long timeo = *timeop;
1838 	int err = sock_error(sk);
1839 
1840 	if (err)
1841 		return err;
1842 
1843 	for (;;) {
1844 		if (timeo && skb_queue_empty(&sk->sk_receive_queue)) {
1845 			if (sk->sk_shutdown & RCV_SHUTDOWN) {
1846 				err = -ENOTCONN;
1847 				break;
1848 			}
1849 			add_wait_queue(sk_sleep(sk), &wait);
1850 			release_sock(sk);
1851 			timeo = wait_woken(&wait, TASK_INTERRUPTIBLE, timeo);
1852 			sched_annotate_sleep();
1853 			lock_sock(sk);
1854 			remove_wait_queue(sk_sleep(sk), &wait);
1855 		}
1856 		err = 0;
1857 		if (!skb_queue_empty(&sk->sk_receive_queue))
1858 			break;
1859 		err = -EAGAIN;
1860 		if (!timeo)
1861 			break;
1862 		err = sock_intr_errno(timeo);
1863 		if (signal_pending(current))
1864 			break;
1865 
1866 		err = sock_error(sk);
1867 		if (err)
1868 			break;
1869 	}
1870 	*timeop = timeo;
1871 	return err;
1872 }
1873 
1874 /**
1875  * tipc_recvmsg - receive packet-oriented message
1876  * @sock: network socket
1877  * @m: descriptor for message info
1878  * @buflen: length of user buffer area
1879  * @flags: receive flags
1880  *
1881  * Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages.
1882  * If the complete message doesn't fit in user area, truncate it.
1883  *
1884  * Return: size of returned message data, errno otherwise
1885  */
1886 static int tipc_recvmsg(struct socket *sock, struct msghdr *m,
1887 			size_t buflen,	int flags)
1888 {
1889 	struct sock *sk = sock->sk;
1890 	bool connected = !tipc_sk_type_connectionless(sk);
1891 	struct tipc_sock *tsk = tipc_sk(sk);
1892 	int rc, err, hlen, dlen, copy;
1893 	struct tipc_skb_cb *skb_cb;
1894 	struct sk_buff_head xmitq;
1895 	struct tipc_msg *hdr;
1896 	struct sk_buff *skb;
1897 	bool grp_evt;
1898 	long timeout;
1899 
1900 	/* Catch invalid receive requests */
1901 	if (unlikely(!buflen))
1902 		return -EINVAL;
1903 
1904 	lock_sock(sk);
1905 	if (unlikely(connected && sk->sk_state == TIPC_OPEN)) {
1906 		rc = -ENOTCONN;
1907 		goto exit;
1908 	}
1909 	timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
1910 
1911 	/* Step rcv queue to first msg with data or error; wait if necessary */
1912 	do {
1913 		rc = tipc_wait_for_rcvmsg(sock, &timeout);
1914 		if (unlikely(rc))
1915 			goto exit;
1916 		skb = skb_peek(&sk->sk_receive_queue);
1917 		skb_cb = TIPC_SKB_CB(skb);
1918 		hdr = buf_msg(skb);
1919 		dlen = msg_data_sz(hdr);
1920 		hlen = msg_hdr_sz(hdr);
1921 		err = msg_errcode(hdr);
1922 		grp_evt = msg_is_grp_evt(hdr);
1923 		if (likely(dlen || err))
1924 			break;
1925 		tsk_advance_rx_queue(sk);
1926 	} while (1);
1927 
1928 	/* Collect msg meta data, including error code and rejected data */
1929 	tipc_sk_set_orig_addr(m, skb);
1930 	rc = tipc_sk_anc_data_recv(m, skb, tsk);
1931 	if (unlikely(rc))
1932 		goto exit;
1933 	hdr = buf_msg(skb);
1934 
1935 	/* Capture data if non-error msg, otherwise just set return value */
1936 	if (likely(!err)) {
1937 		int offset = skb_cb->bytes_read;
1938 
1939 		copy = min_t(int, dlen - offset, buflen);
1940 		rc = skb_copy_datagram_msg(skb, hlen + offset, m, copy);
1941 		if (unlikely(rc))
1942 			goto exit;
1943 		if (unlikely(offset + copy < dlen)) {
1944 			if (flags & MSG_EOR) {
1945 				if (!(flags & MSG_PEEK))
1946 					skb_cb->bytes_read = offset + copy;
1947 			} else {
1948 				m->msg_flags |= MSG_TRUNC;
1949 				skb_cb->bytes_read = 0;
1950 			}
1951 		} else {
1952 			if (flags & MSG_EOR)
1953 				m->msg_flags |= MSG_EOR;
1954 			skb_cb->bytes_read = 0;
1955 		}
1956 	} else {
1957 		copy = 0;
1958 		rc = 0;
1959 		if (err != TIPC_CONN_SHUTDOWN && connected && !m->msg_control) {
1960 			rc = -ECONNRESET;
1961 			goto exit;
1962 		}
1963 	}
1964 
1965 	/* Mark message as group event if applicable */
1966 	if (unlikely(grp_evt)) {
1967 		if (msg_grp_evt(hdr) == TIPC_WITHDRAWN)
1968 			m->msg_flags |= MSG_EOR;
1969 		m->msg_flags |= MSG_OOB;
1970 		copy = 0;
1971 	}
1972 
1973 	/* Caption of data or error code/rejected data was successful */
1974 	if (unlikely(flags & MSG_PEEK))
1975 		goto exit;
1976 
1977 	/* Send group flow control advertisement when applicable */
1978 	if (tsk->group && msg_in_group(hdr) && !grp_evt) {
1979 		__skb_queue_head_init(&xmitq);
1980 		tipc_group_update_rcv_win(tsk->group, tsk_blocks(hlen + dlen),
1981 					  msg_orignode(hdr), msg_origport(hdr),
1982 					  &xmitq);
1983 		tipc_node_distr_xmit(sock_net(sk), &xmitq);
1984 	}
1985 
1986 	if (skb_cb->bytes_read)
1987 		goto exit;
1988 
1989 	tsk_advance_rx_queue(sk);
1990 
1991 	if (likely(!connected))
1992 		goto exit;
1993 
1994 	/* Send connection flow control advertisement when applicable */
1995 	tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen);
1996 	if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE)
1997 		tipc_sk_send_ack(tsk);
1998 exit:
1999 	release_sock(sk);
2000 	return rc ? rc : copy;
2001 }
2002 
2003 /**
2004  * tipc_recvstream - receive stream-oriented data
2005  * @sock: network socket
2006  * @m: descriptor for message info
2007  * @buflen: total size of user buffer area
2008  * @flags: receive flags
2009  *
2010  * Used for SOCK_STREAM messages only.  If not enough data is available
2011  * will optionally wait for more; never truncates data.
2012  *
2013  * Return: size of returned message data, errno otherwise
2014  */
2015 static int tipc_recvstream(struct socket *sock, struct msghdr *m,
2016 			   size_t buflen, int flags)
2017 {
2018 	struct sock *sk = sock->sk;
2019 	struct tipc_sock *tsk = tipc_sk(sk);
2020 	struct sk_buff *skb;
2021 	struct tipc_msg *hdr;
2022 	struct tipc_skb_cb *skb_cb;
2023 	bool peek = flags & MSG_PEEK;
2024 	int offset, required, copy, copied = 0;
2025 	int hlen, dlen, err, rc;
2026 	long timeout;
2027 
2028 	/* Catch invalid receive attempts */
2029 	if (unlikely(!buflen))
2030 		return -EINVAL;
2031 
2032 	lock_sock(sk);
2033 
2034 	if (unlikely(sk->sk_state == TIPC_OPEN)) {
2035 		rc = -ENOTCONN;
2036 		goto exit;
2037 	}
2038 	required = sock_rcvlowat(sk, flags & MSG_WAITALL, buflen);
2039 	timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
2040 
2041 	do {
2042 		/* Look at first msg in receive queue; wait if necessary */
2043 		rc = tipc_wait_for_rcvmsg(sock, &timeout);
2044 		if (unlikely(rc))
2045 			break;
2046 		skb = skb_peek(&sk->sk_receive_queue);
2047 		skb_cb = TIPC_SKB_CB(skb);
2048 		hdr = buf_msg(skb);
2049 		dlen = msg_data_sz(hdr);
2050 		hlen = msg_hdr_sz(hdr);
2051 		err = msg_errcode(hdr);
2052 
2053 		/* Discard any empty non-errored (SYN-) message */
2054 		if (unlikely(!dlen && !err)) {
2055 			tsk_advance_rx_queue(sk);
2056 			continue;
2057 		}
2058 
2059 		/* Collect msg meta data, incl. error code and rejected data */
2060 		if (!copied) {
2061 			tipc_sk_set_orig_addr(m, skb);
2062 			rc = tipc_sk_anc_data_recv(m, skb, tsk);
2063 			if (rc)
2064 				break;
2065 			hdr = buf_msg(skb);
2066 		}
2067 
2068 		/* Copy data if msg ok, otherwise return error/partial data */
2069 		if (likely(!err)) {
2070 			offset = skb_cb->bytes_read;
2071 			copy = min_t(int, dlen - offset, buflen - copied);
2072 			rc = skb_copy_datagram_msg(skb, hlen + offset, m, copy);
2073 			if (unlikely(rc))
2074 				break;
2075 			copied += copy;
2076 			offset += copy;
2077 			if (unlikely(offset < dlen)) {
2078 				if (!peek)
2079 					skb_cb->bytes_read = offset;
2080 				break;
2081 			}
2082 		} else {
2083 			rc = 0;
2084 			if ((err != TIPC_CONN_SHUTDOWN) && !m->msg_control)
2085 				rc = -ECONNRESET;
2086 			if (copied || rc)
2087 				break;
2088 		}
2089 
2090 		if (unlikely(peek))
2091 			break;
2092 
2093 		tsk_advance_rx_queue(sk);
2094 
2095 		/* Send connection flow control advertisement when applicable */
2096 		tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen);
2097 		if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE)
2098 			tipc_sk_send_ack(tsk);
2099 
2100 		/* Exit if all requested data or FIN/error received */
2101 		if (copied == buflen || err)
2102 			break;
2103 
2104 	} while (!skb_queue_empty(&sk->sk_receive_queue) || copied < required);
2105 exit:
2106 	release_sock(sk);
2107 	return copied ? copied : rc;
2108 }
2109 
2110 /**
2111  * tipc_write_space - wake up thread if port congestion is released
2112  * @sk: socket
2113  */
2114 static void tipc_write_space(struct sock *sk)
2115 {
2116 	struct socket_wq *wq;
2117 
2118 	rcu_read_lock();
2119 	wq = rcu_dereference(sk->sk_wq);
2120 	if (skwq_has_sleeper(wq))
2121 		wake_up_interruptible_sync_poll(&wq->wait, EPOLLOUT |
2122 						EPOLLWRNORM | EPOLLWRBAND);
2123 	rcu_read_unlock();
2124 }
2125 
2126 /**
2127  * tipc_data_ready - wake up threads to indicate messages have been received
2128  * @sk: socket
2129  */
2130 static void tipc_data_ready(struct sock *sk)
2131 {
2132 	struct socket_wq *wq;
2133 
2134 	trace_sk_data_ready(sk);
2135 
2136 	rcu_read_lock();
2137 	wq = rcu_dereference(sk->sk_wq);
2138 	if (skwq_has_sleeper(wq))
2139 		wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN |
2140 						EPOLLRDNORM | EPOLLRDBAND);
2141 	rcu_read_unlock();
2142 }
2143 
2144 static void tipc_sock_destruct(struct sock *sk)
2145 {
2146 	__skb_queue_purge(&sk->sk_receive_queue);
2147 }
2148 
2149 static void tipc_sk_proto_rcv(struct sock *sk,
2150 			      struct sk_buff_head *inputq,
2151 			      struct sk_buff_head *xmitq)
2152 {
2153 	struct sk_buff *skb = __skb_dequeue(inputq);
2154 	struct tipc_sock *tsk = tipc_sk(sk);
2155 	struct tipc_msg *hdr = buf_msg(skb);
2156 	struct tipc_group *grp = tsk->group;
2157 	bool wakeup = false;
2158 
2159 	switch (msg_user(hdr)) {
2160 	case CONN_MANAGER:
2161 		tipc_sk_conn_proto_rcv(tsk, skb, inputq, xmitq);
2162 		return;
2163 	case SOCK_WAKEUP:
2164 		tipc_dest_del(&tsk->cong_links, msg_orignode(hdr), 0);
2165 		/* coupled with smp_rmb() in tipc_wait_for_cond() */
2166 		smp_wmb();
2167 		tsk->cong_link_cnt--;
2168 		wakeup = true;
2169 		tipc_sk_push_backlog(tsk, false);
2170 		break;
2171 	case GROUP_PROTOCOL:
2172 		tipc_group_proto_rcv(grp, &wakeup, hdr, inputq, xmitq);
2173 		break;
2174 	case TOP_SRV:
2175 		tipc_group_member_evt(tsk->group, &wakeup, &sk->sk_rcvbuf,
2176 				      hdr, inputq, xmitq);
2177 		break;
2178 	default:
2179 		break;
2180 	}
2181 
2182 	if (wakeup)
2183 		sk->sk_write_space(sk);
2184 
2185 	kfree_skb(skb);
2186 }
2187 
2188 /**
2189  * tipc_sk_filter_connect - check incoming message for a connection-based socket
2190  * @tsk: TIPC socket
2191  * @skb: pointer to message buffer.
2192  * @xmitq: for Nagle ACK if any
2193  * Return: true if message should be added to receive queue, false otherwise
2194  */
2195 static bool tipc_sk_filter_connect(struct tipc_sock *tsk, struct sk_buff *skb,
2196 				   struct sk_buff_head *xmitq)
2197 {
2198 	struct sock *sk = &tsk->sk;
2199 	struct net *net = sock_net(sk);
2200 	struct tipc_msg *hdr = buf_msg(skb);
2201 	bool con_msg = msg_connected(hdr);
2202 	u32 pport = tsk_peer_port(tsk);
2203 	u32 pnode = tsk_peer_node(tsk);
2204 	u32 oport = msg_origport(hdr);
2205 	u32 onode = msg_orignode(hdr);
2206 	int err = msg_errcode(hdr);
2207 	unsigned long delay;
2208 
2209 	if (unlikely(msg_mcast(hdr)))
2210 		return false;
2211 	tsk->oneway = 0;
2212 
2213 	switch (sk->sk_state) {
2214 	case TIPC_CONNECTING:
2215 		/* Setup ACK */
2216 		if (likely(con_msg)) {
2217 			if (err)
2218 				break;
2219 			tipc_sk_finish_conn(tsk, oport, onode);
2220 			msg_set_importance(&tsk->phdr, msg_importance(hdr));
2221 			/* ACK+ message with data is added to receive queue */
2222 			if (msg_data_sz(hdr))
2223 				return true;
2224 			/* Empty ACK-, - wake up sleeping connect() and drop */
2225 			sk->sk_state_change(sk);
2226 			msg_set_dest_droppable(hdr, 1);
2227 			return false;
2228 		}
2229 		/* Ignore connectionless message if not from listening socket */
2230 		if (oport != pport || onode != pnode)
2231 			return false;
2232 
2233 		/* Rejected SYN */
2234 		if (err != TIPC_ERR_OVERLOAD)
2235 			break;
2236 
2237 		/* Prepare for new setup attempt if we have a SYN clone */
2238 		if (skb_queue_empty(&sk->sk_write_queue))
2239 			break;
2240 		get_random_bytes(&delay, 2);
2241 		delay %= (tsk->conn_timeout / 4);
2242 		delay = msecs_to_jiffies(delay + 100);
2243 		sk_reset_timer(sk, &sk->sk_timer, jiffies + delay);
2244 		return false;
2245 	case TIPC_OPEN:
2246 	case TIPC_DISCONNECTING:
2247 		return false;
2248 	case TIPC_LISTEN:
2249 		/* Accept only SYN message */
2250 		if (!msg_is_syn(hdr) &&
2251 		    tipc_node_get_capabilities(net, onode) & TIPC_SYN_BIT)
2252 			return false;
2253 		if (!con_msg && !err)
2254 			return true;
2255 		return false;
2256 	case TIPC_ESTABLISHED:
2257 		if (!skb_queue_empty(&sk->sk_write_queue))
2258 			tipc_sk_push_backlog(tsk, false);
2259 		/* Accept only connection-based messages sent by peer */
2260 		if (likely(con_msg && !err && pport == oport &&
2261 			   pnode == onode)) {
2262 			if (msg_ack_required(hdr)) {
2263 				struct sk_buff *skb;
2264 
2265 				skb = tipc_sk_build_ack(tsk);
2266 				if (skb) {
2267 					msg_set_nagle_ack(buf_msg(skb));
2268 					__skb_queue_tail(xmitq, skb);
2269 				}
2270 			}
2271 			return true;
2272 		}
2273 		if (!tsk_peer_msg(tsk, hdr))
2274 			return false;
2275 		if (!err)
2276 			return true;
2277 		tipc_set_sk_state(sk, TIPC_DISCONNECTING);
2278 		tipc_node_remove_conn(net, pnode, tsk->portid);
2279 		sk->sk_state_change(sk);
2280 		return true;
2281 	default:
2282 		pr_err("Unknown sk_state %u\n", sk->sk_state);
2283 	}
2284 	/* Abort connection setup attempt */
2285 	tipc_set_sk_state(sk, TIPC_DISCONNECTING);
2286 	sk->sk_err = ECONNREFUSED;
2287 	sk->sk_state_change(sk);
2288 	return true;
2289 }
2290 
2291 /**
2292  * rcvbuf_limit - get proper overload limit of socket receive queue
2293  * @sk: socket
2294  * @skb: message
2295  *
2296  * For connection oriented messages, irrespective of importance,
2297  * default queue limit is 2 MB.
2298  *
2299  * For connectionless messages, queue limits are based on message
2300  * importance as follows:
2301  *
2302  * TIPC_LOW_IMPORTANCE       (2 MB)
2303  * TIPC_MEDIUM_IMPORTANCE    (4 MB)
2304  * TIPC_HIGH_IMPORTANCE      (8 MB)
2305  * TIPC_CRITICAL_IMPORTANCE  (16 MB)
2306  *
2307  * Return: overload limit according to corresponding message importance
2308  */
2309 static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *skb)
2310 {
2311 	struct tipc_sock *tsk = tipc_sk(sk);
2312 	struct tipc_msg *hdr = buf_msg(skb);
2313 
2314 	if (unlikely(msg_in_group(hdr)))
2315 		return READ_ONCE(sk->sk_rcvbuf);
2316 
2317 	if (unlikely(!msg_connected(hdr)))
2318 		return READ_ONCE(sk->sk_rcvbuf) << msg_importance(hdr);
2319 
2320 	if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL))
2321 		return READ_ONCE(sk->sk_rcvbuf);
2322 
2323 	return FLOWCTL_MSG_LIM;
2324 }
2325 
2326 /**
2327  * tipc_sk_filter_rcv - validate incoming message
2328  * @sk: socket
2329  * @skb: pointer to message.
2330  * @xmitq: output message area (FIXME)
2331  *
2332  * Enqueues message on receive queue if acceptable; optionally handles
2333  * disconnect indication for a connected socket.
2334  *
2335  * Called with socket lock already taken
2336  */
2337 static void tipc_sk_filter_rcv(struct sock *sk, struct sk_buff *skb,
2338 			       struct sk_buff_head *xmitq)
2339 {
2340 	bool sk_conn = !tipc_sk_type_connectionless(sk);
2341 	struct tipc_sock *tsk = tipc_sk(sk);
2342 	struct tipc_group *grp = tsk->group;
2343 	struct tipc_msg *hdr = buf_msg(skb);
2344 	struct net *net = sock_net(sk);
2345 	struct sk_buff_head inputq;
2346 	int mtyp = msg_type(hdr);
2347 	int limit, err = TIPC_OK;
2348 
2349 	trace_tipc_sk_filter_rcv(sk, skb, TIPC_DUMP_ALL, " ");
2350 	TIPC_SKB_CB(skb)->bytes_read = 0;
2351 	__skb_queue_head_init(&inputq);
2352 	__skb_queue_tail(&inputq, skb);
2353 
2354 	if (unlikely(!msg_isdata(hdr)))
2355 		tipc_sk_proto_rcv(sk, &inputq, xmitq);
2356 
2357 	if (unlikely(grp))
2358 		tipc_group_filter_msg(grp, &inputq, xmitq);
2359 
2360 	if (unlikely(!grp) && mtyp == TIPC_MCAST_MSG)
2361 		tipc_mcast_filter_msg(net, &tsk->mc_method.deferredq, &inputq);
2362 
2363 	/* Validate and add to receive buffer if there is space */
2364 	while ((skb = __skb_dequeue(&inputq))) {
2365 		hdr = buf_msg(skb);
2366 		limit = rcvbuf_limit(sk, skb);
2367 		if ((sk_conn && !tipc_sk_filter_connect(tsk, skb, xmitq)) ||
2368 		    (!sk_conn && msg_connected(hdr)) ||
2369 		    (!grp && msg_in_group(hdr)))
2370 			err = TIPC_ERR_NO_PORT;
2371 		else if (sk_rmem_alloc_get(sk) + skb->truesize >= limit) {
2372 			trace_tipc_sk_dump(sk, skb, TIPC_DUMP_ALL,
2373 					   "err_overload2!");
2374 			atomic_inc(&sk->sk_drops);
2375 			err = TIPC_ERR_OVERLOAD;
2376 		}
2377 
2378 		if (unlikely(err)) {
2379 			if (tipc_msg_reverse(tipc_own_addr(net), &skb, err)) {
2380 				trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_NONE,
2381 						      "@filter_rcv!");
2382 				__skb_queue_tail(xmitq, skb);
2383 			}
2384 			err = TIPC_OK;
2385 			continue;
2386 		}
2387 		__skb_queue_tail(&sk->sk_receive_queue, skb);
2388 		skb_set_owner_r(skb, sk);
2389 		trace_tipc_sk_overlimit2(sk, skb, TIPC_DUMP_ALL,
2390 					 "rcvq >90% allocated!");
2391 		sk->sk_data_ready(sk);
2392 	}
2393 }
2394 
2395 /**
2396  * tipc_sk_backlog_rcv - handle incoming message from backlog queue
2397  * @sk: socket
2398  * @skb: message
2399  *
2400  * Caller must hold socket lock
2401  */
2402 static int tipc_sk_backlog_rcv(struct sock *sk, struct sk_buff *skb)
2403 {
2404 	unsigned int before = sk_rmem_alloc_get(sk);
2405 	struct sk_buff_head xmitq;
2406 	unsigned int added;
2407 
2408 	__skb_queue_head_init(&xmitq);
2409 
2410 	tipc_sk_filter_rcv(sk, skb, &xmitq);
2411 	added = sk_rmem_alloc_get(sk) - before;
2412 	atomic_add(added, &tipc_sk(sk)->dupl_rcvcnt);
2413 
2414 	/* Send pending response/rejected messages, if any */
2415 	tipc_node_distr_xmit(sock_net(sk), &xmitq);
2416 	return 0;
2417 }
2418 
2419 /**
2420  * tipc_sk_enqueue - extract all buffers with destination 'dport' from
2421  *                   inputq and try adding them to socket or backlog queue
2422  * @inputq: list of incoming buffers with potentially different destinations
2423  * @sk: socket where the buffers should be enqueued
2424  * @dport: port number for the socket
2425  * @xmitq: output queue
2426  *
2427  * Caller must hold socket lock
2428  */
2429 static void tipc_sk_enqueue(struct sk_buff_head *inputq, struct sock *sk,
2430 			    u32 dport, struct sk_buff_head *xmitq)
2431 {
2432 	unsigned long time_limit = jiffies + usecs_to_jiffies(20000);
2433 	struct sk_buff *skb;
2434 	unsigned int lim;
2435 	atomic_t *dcnt;
2436 	u32 onode;
2437 
2438 	while (skb_queue_len(inputq)) {
2439 		if (unlikely(time_after_eq(jiffies, time_limit)))
2440 			return;
2441 
2442 		skb = tipc_skb_dequeue(inputq, dport);
2443 		if (unlikely(!skb))
2444 			return;
2445 
2446 		/* Add message directly to receive queue if possible */
2447 		if (!sock_owned_by_user(sk)) {
2448 			tipc_sk_filter_rcv(sk, skb, xmitq);
2449 			continue;
2450 		}
2451 
2452 		/* Try backlog, compensating for double-counted bytes */
2453 		dcnt = &tipc_sk(sk)->dupl_rcvcnt;
2454 		if (!sk->sk_backlog.len)
2455 			atomic_set(dcnt, 0);
2456 		lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt);
2457 		if (likely(!sk_add_backlog(sk, skb, lim))) {
2458 			trace_tipc_sk_overlimit1(sk, skb, TIPC_DUMP_ALL,
2459 						 "bklg & rcvq >90% allocated!");
2460 			continue;
2461 		}
2462 
2463 		trace_tipc_sk_dump(sk, skb, TIPC_DUMP_ALL, "err_overload!");
2464 		/* Overload => reject message back to sender */
2465 		onode = tipc_own_addr(sock_net(sk));
2466 		atomic_inc(&sk->sk_drops);
2467 		if (tipc_msg_reverse(onode, &skb, TIPC_ERR_OVERLOAD)) {
2468 			trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_ALL,
2469 					      "@sk_enqueue!");
2470 			__skb_queue_tail(xmitq, skb);
2471 		}
2472 		break;
2473 	}
2474 }
2475 
2476 /**
2477  * tipc_sk_rcv - handle a chain of incoming buffers
2478  * @net: the associated network namespace
2479  * @inputq: buffer list containing the buffers
2480  * Consumes all buffers in list until inputq is empty
2481  * Note: may be called in multiple threads referring to the same queue
2482  */
2483 void tipc_sk_rcv(struct net *net, struct sk_buff_head *inputq)
2484 {
2485 	struct sk_buff_head xmitq;
2486 	u32 dnode, dport = 0;
2487 	int err;
2488 	struct tipc_sock *tsk;
2489 	struct sock *sk;
2490 	struct sk_buff *skb;
2491 
2492 	__skb_queue_head_init(&xmitq);
2493 	while (skb_queue_len(inputq)) {
2494 		dport = tipc_skb_peek_port(inputq, dport);
2495 		tsk = tipc_sk_lookup(net, dport);
2496 
2497 		if (likely(tsk)) {
2498 			sk = &tsk->sk;
2499 			if (likely(spin_trylock_bh(&sk->sk_lock.slock))) {
2500 				tipc_sk_enqueue(inputq, sk, dport, &xmitq);
2501 				spin_unlock_bh(&sk->sk_lock.slock);
2502 			}
2503 			/* Send pending response/rejected messages, if any */
2504 			tipc_node_distr_xmit(sock_net(sk), &xmitq);
2505 			sock_put(sk);
2506 			continue;
2507 		}
2508 		/* No destination socket => dequeue skb if still there */
2509 		skb = tipc_skb_dequeue(inputq, dport);
2510 		if (!skb)
2511 			return;
2512 
2513 		/* Try secondary lookup if unresolved named message */
2514 		err = TIPC_ERR_NO_PORT;
2515 		if (tipc_msg_lookup_dest(net, skb, &err))
2516 			goto xmit;
2517 
2518 		/* Prepare for message rejection */
2519 		if (!tipc_msg_reverse(tipc_own_addr(net), &skb, err))
2520 			continue;
2521 
2522 		trace_tipc_sk_rej_msg(NULL, skb, TIPC_DUMP_NONE, "@sk_rcv!");
2523 xmit:
2524 		dnode = msg_destnode(buf_msg(skb));
2525 		tipc_node_xmit_skb(net, skb, dnode, dport);
2526 	}
2527 }
2528 
2529 static int tipc_wait_for_connect(struct socket *sock, long *timeo_p)
2530 {
2531 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
2532 	struct sock *sk = sock->sk;
2533 	int done;
2534 
2535 	do {
2536 		int err = sock_error(sk);
2537 		if (err)
2538 			return err;
2539 		if (!*timeo_p)
2540 			return -ETIMEDOUT;
2541 		if (signal_pending(current))
2542 			return sock_intr_errno(*timeo_p);
2543 		if (sk->sk_state == TIPC_DISCONNECTING)
2544 			break;
2545 
2546 		add_wait_queue(sk_sleep(sk), &wait);
2547 		done = sk_wait_event(sk, timeo_p, tipc_sk_connected(sk),
2548 				     &wait);
2549 		remove_wait_queue(sk_sleep(sk), &wait);
2550 	} while (!done);
2551 	return 0;
2552 }
2553 
2554 static bool tipc_sockaddr_is_sane(struct sockaddr_tipc *addr)
2555 {
2556 	if (addr->family != AF_TIPC)
2557 		return false;
2558 	if (addr->addrtype == TIPC_SERVICE_RANGE)
2559 		return (addr->addr.nameseq.lower <= addr->addr.nameseq.upper);
2560 	return (addr->addrtype == TIPC_SERVICE_ADDR ||
2561 		addr->addrtype == TIPC_SOCKET_ADDR);
2562 }
2563 
2564 /**
2565  * tipc_connect - establish a connection to another TIPC port
2566  * @sock: socket structure
2567  * @dest: socket address for destination port
2568  * @destlen: size of socket address data structure
2569  * @flags: file-related flags associated with socket
2570  *
2571  * Return: 0 on success, errno otherwise
2572  */
2573 static int tipc_connect(struct socket *sock, struct sockaddr *dest,
2574 			int destlen, int flags)
2575 {
2576 	struct sock *sk = sock->sk;
2577 	struct tipc_sock *tsk = tipc_sk(sk);
2578 	struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest;
2579 	struct msghdr m = {NULL,};
2580 	long timeout = (flags & O_NONBLOCK) ? 0 : tsk->conn_timeout;
2581 	int previous;
2582 	int res = 0;
2583 
2584 	if (destlen != sizeof(struct sockaddr_tipc))
2585 		return -EINVAL;
2586 
2587 	lock_sock(sk);
2588 
2589 	if (tsk->group) {
2590 		res = -EINVAL;
2591 		goto exit;
2592 	}
2593 
2594 	if (dst->family == AF_UNSPEC) {
2595 		memset(&tsk->peer, 0, sizeof(struct sockaddr_tipc));
2596 		if (!tipc_sk_type_connectionless(sk))
2597 			res = -EINVAL;
2598 		goto exit;
2599 	}
2600 	if (!tipc_sockaddr_is_sane(dst)) {
2601 		res = -EINVAL;
2602 		goto exit;
2603 	}
2604 	/* DGRAM/RDM connect(), just save the destaddr */
2605 	if (tipc_sk_type_connectionless(sk)) {
2606 		memcpy(&tsk->peer, dest, destlen);
2607 		goto exit;
2608 	} else if (dst->addrtype == TIPC_SERVICE_RANGE) {
2609 		res = -EINVAL;
2610 		goto exit;
2611 	}
2612 
2613 	previous = sk->sk_state;
2614 
2615 	switch (sk->sk_state) {
2616 	case TIPC_OPEN:
2617 		/* Send a 'SYN-' to destination */
2618 		m.msg_name = dest;
2619 		m.msg_namelen = destlen;
2620 		iov_iter_kvec(&m.msg_iter, ITER_SOURCE, NULL, 0, 0);
2621 
2622 		/* If connect is in non-blocking case, set MSG_DONTWAIT to
2623 		 * indicate send_msg() is never blocked.
2624 		 */
2625 		if (!timeout)
2626 			m.msg_flags = MSG_DONTWAIT;
2627 
2628 		res = __tipc_sendmsg(sock, &m, 0);
2629 		if ((res < 0) && (res != -EWOULDBLOCK))
2630 			goto exit;
2631 
2632 		/* Just entered TIPC_CONNECTING state; the only
2633 		 * difference is that return value in non-blocking
2634 		 * case is EINPROGRESS, rather than EALREADY.
2635 		 */
2636 		res = -EINPROGRESS;
2637 		fallthrough;
2638 	case TIPC_CONNECTING:
2639 		if (!timeout) {
2640 			if (previous == TIPC_CONNECTING)
2641 				res = -EALREADY;
2642 			goto exit;
2643 		}
2644 		timeout = msecs_to_jiffies(timeout);
2645 		/* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */
2646 		res = tipc_wait_for_connect(sock, &timeout);
2647 		break;
2648 	case TIPC_ESTABLISHED:
2649 		res = -EISCONN;
2650 		break;
2651 	default:
2652 		res = -EINVAL;
2653 	}
2654 
2655 exit:
2656 	release_sock(sk);
2657 	return res;
2658 }
2659 
2660 /**
2661  * tipc_listen - allow socket to listen for incoming connections
2662  * @sock: socket structure
2663  * @len: (unused)
2664  *
2665  * Return: 0 on success, errno otherwise
2666  */
2667 static int tipc_listen(struct socket *sock, int len)
2668 {
2669 	struct sock *sk = sock->sk;
2670 	int res;
2671 
2672 	lock_sock(sk);
2673 	res = tipc_set_sk_state(sk, TIPC_LISTEN);
2674 	release_sock(sk);
2675 
2676 	return res;
2677 }
2678 
2679 static int tipc_wait_for_accept(struct socket *sock, long timeo)
2680 {
2681 	struct sock *sk = sock->sk;
2682 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
2683 	int err;
2684 
2685 	/* True wake-one mechanism for incoming connections: only
2686 	 * one process gets woken up, not the 'whole herd'.
2687 	 * Since we do not 'race & poll' for established sockets
2688 	 * anymore, the common case will execute the loop only once.
2689 	*/
2690 	for (;;) {
2691 		if (timeo && skb_queue_empty(&sk->sk_receive_queue)) {
2692 			add_wait_queue(sk_sleep(sk), &wait);
2693 			release_sock(sk);
2694 			timeo = wait_woken(&wait, TASK_INTERRUPTIBLE, timeo);
2695 			lock_sock(sk);
2696 			remove_wait_queue(sk_sleep(sk), &wait);
2697 		}
2698 		err = 0;
2699 		if (!skb_queue_empty(&sk->sk_receive_queue))
2700 			break;
2701 		err = -EAGAIN;
2702 		if (!timeo)
2703 			break;
2704 		err = sock_intr_errno(timeo);
2705 		if (signal_pending(current))
2706 			break;
2707 	}
2708 	return err;
2709 }
2710 
2711 /**
2712  * tipc_accept - wait for connection request
2713  * @sock: listening socket
2714  * @new_sock: new socket that is to be connected
2715  * @flags: file-related flags associated with socket
2716  * @kern: caused by kernel or by userspace?
2717  *
2718  * Return: 0 on success, errno otherwise
2719  */
2720 static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags,
2721 		       bool kern)
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, 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, 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 = from_timer(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 		WARN_ON(refcount_read(&sk->sk_refcnt) == 1);
3042 		__sock_put(sk);
3043 	}
3044 }
3045 
3046 static const struct rhashtable_params tsk_rht_params = {
3047 	.nelem_hint = 192,
3048 	.head_offset = offsetof(struct tipc_sock, node),
3049 	.key_offset = offsetof(struct tipc_sock, portid),
3050 	.key_len = sizeof(u32), /* portid */
3051 	.max_size = 1048576,
3052 	.min_size = 256,
3053 	.automatic_shrinking = true,
3054 };
3055 
3056 int tipc_sk_rht_init(struct net *net)
3057 {
3058 	struct tipc_net *tn = net_generic(net, tipc_net_id);
3059 
3060 	return rhashtable_init(&tn->sk_rht, &tsk_rht_params);
3061 }
3062 
3063 void tipc_sk_rht_destroy(struct net *net)
3064 {
3065 	struct tipc_net *tn = net_generic(net, tipc_net_id);
3066 
3067 	/* Wait for socket readers to complete */
3068 	synchronize_net();
3069 
3070 	rhashtable_destroy(&tn->sk_rht);
3071 }
3072 
3073 static int tipc_sk_join(struct tipc_sock *tsk, struct tipc_group_req *mreq)
3074 {
3075 	struct net *net = sock_net(&tsk->sk);
3076 	struct tipc_group *grp = tsk->group;
3077 	struct tipc_msg *hdr = &tsk->phdr;
3078 	struct tipc_uaddr ua;
3079 	int rc;
3080 
3081 	if (mreq->type < TIPC_RESERVED_TYPES)
3082 		return -EACCES;
3083 	if (mreq->scope > TIPC_NODE_SCOPE)
3084 		return -EINVAL;
3085 	if (mreq->scope != TIPC_NODE_SCOPE)
3086 		mreq->scope = TIPC_CLUSTER_SCOPE;
3087 	if (grp)
3088 		return -EACCES;
3089 	grp = tipc_group_create(net, tsk->portid, mreq, &tsk->group_is_open);
3090 	if (!grp)
3091 		return -ENOMEM;
3092 	tsk->group = grp;
3093 	msg_set_lookup_scope(hdr, mreq->scope);
3094 	msg_set_nametype(hdr, mreq->type);
3095 	msg_set_dest_droppable(hdr, true);
3096 	tipc_uaddr(&ua, TIPC_SERVICE_RANGE, mreq->scope,
3097 		   mreq->type, mreq->instance, mreq->instance);
3098 	tipc_nametbl_build_group(net, grp, &ua);
3099 	rc = tipc_sk_publish(tsk, &ua);
3100 	if (rc) {
3101 		tipc_group_delete(net, grp);
3102 		tsk->group = NULL;
3103 		return rc;
3104 	}
3105 	/* Eliminate any risk that a broadcast overtakes sent JOINs */
3106 	tsk->mc_method.rcast = true;
3107 	tsk->mc_method.mandatory = true;
3108 	tipc_group_join(net, grp, &tsk->sk.sk_rcvbuf);
3109 	return rc;
3110 }
3111 
3112 static int tipc_sk_leave(struct tipc_sock *tsk)
3113 {
3114 	struct net *net = sock_net(&tsk->sk);
3115 	struct tipc_group *grp = tsk->group;
3116 	struct tipc_uaddr ua;
3117 	int scope;
3118 
3119 	if (!grp)
3120 		return -EINVAL;
3121 	ua.addrtype = TIPC_SERVICE_RANGE;
3122 	tipc_group_self(grp, &ua.sr, &scope);
3123 	ua.scope = scope;
3124 	tipc_group_delete(net, grp);
3125 	tsk->group = NULL;
3126 	tipc_sk_withdraw(tsk, &ua);
3127 	return 0;
3128 }
3129 
3130 /**
3131  * tipc_setsockopt - set socket option
3132  * @sock: socket structure
3133  * @lvl: option level
3134  * @opt: option identifier
3135  * @ov: pointer to new option value
3136  * @ol: length of option value
3137  *
3138  * For stream sockets only, accepts and ignores all IPPROTO_TCP options
3139  * (to ease compatibility).
3140  *
3141  * Return: 0 on success, errno otherwise
3142  */
3143 static int tipc_setsockopt(struct socket *sock, int lvl, int opt,
3144 			   sockptr_t ov, unsigned int ol)
3145 {
3146 	struct sock *sk = sock->sk;
3147 	struct tipc_sock *tsk = tipc_sk(sk);
3148 	struct tipc_group_req mreq;
3149 	u32 value = 0;
3150 	int res = 0;
3151 
3152 	if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
3153 		return 0;
3154 	if (lvl != SOL_TIPC)
3155 		return -ENOPROTOOPT;
3156 
3157 	switch (opt) {
3158 	case TIPC_IMPORTANCE:
3159 	case TIPC_SRC_DROPPABLE:
3160 	case TIPC_DEST_DROPPABLE:
3161 	case TIPC_CONN_TIMEOUT:
3162 	case TIPC_NODELAY:
3163 		if (ol < sizeof(value))
3164 			return -EINVAL;
3165 		if (copy_from_sockptr(&value, ov, sizeof(u32)))
3166 			return -EFAULT;
3167 		break;
3168 	case TIPC_GROUP_JOIN:
3169 		if (ol < sizeof(mreq))
3170 			return -EINVAL;
3171 		if (copy_from_sockptr(&mreq, ov, sizeof(mreq)))
3172 			return -EFAULT;
3173 		break;
3174 	default:
3175 		if (!sockptr_is_null(ov) || ol)
3176 			return -EINVAL;
3177 	}
3178 
3179 	lock_sock(sk);
3180 
3181 	switch (opt) {
3182 	case TIPC_IMPORTANCE:
3183 		res = tsk_set_importance(sk, value);
3184 		break;
3185 	case TIPC_SRC_DROPPABLE:
3186 		if (sock->type != SOCK_STREAM)
3187 			tsk_set_unreliable(tsk, value);
3188 		else
3189 			res = -ENOPROTOOPT;
3190 		break;
3191 	case TIPC_DEST_DROPPABLE:
3192 		tsk_set_unreturnable(tsk, value);
3193 		break;
3194 	case TIPC_CONN_TIMEOUT:
3195 		tipc_sk(sk)->conn_timeout = value;
3196 		break;
3197 	case TIPC_MCAST_BROADCAST:
3198 		tsk->mc_method.rcast = false;
3199 		tsk->mc_method.mandatory = true;
3200 		break;
3201 	case TIPC_MCAST_REPLICAST:
3202 		tsk->mc_method.rcast = true;
3203 		tsk->mc_method.mandatory = true;
3204 		break;
3205 	case TIPC_GROUP_JOIN:
3206 		res = tipc_sk_join(tsk, &mreq);
3207 		break;
3208 	case TIPC_GROUP_LEAVE:
3209 		res = tipc_sk_leave(tsk);
3210 		break;
3211 	case TIPC_NODELAY:
3212 		tsk->nodelay = !!value;
3213 		tsk_set_nagle(tsk);
3214 		break;
3215 	default:
3216 		res = -EINVAL;
3217 	}
3218 
3219 	release_sock(sk);
3220 
3221 	return res;
3222 }
3223 
3224 /**
3225  * tipc_getsockopt - get socket option
3226  * @sock: socket structure
3227  * @lvl: option level
3228  * @opt: option identifier
3229  * @ov: receptacle for option value
3230  * @ol: receptacle for length of option value
3231  *
3232  * For stream sockets only, returns 0 length result for all IPPROTO_TCP options
3233  * (to ease compatibility).
3234  *
3235  * Return: 0 on success, errno otherwise
3236  */
3237 static int tipc_getsockopt(struct socket *sock, int lvl, int opt,
3238 			   char __user *ov, int __user *ol)
3239 {
3240 	struct sock *sk = sock->sk;
3241 	struct tipc_sock *tsk = tipc_sk(sk);
3242 	struct tipc_service_range seq;
3243 	int len, scope;
3244 	u32 value;
3245 	int res;
3246 
3247 	if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
3248 		return put_user(0, ol);
3249 	if (lvl != SOL_TIPC)
3250 		return -ENOPROTOOPT;
3251 	res = get_user(len, ol);
3252 	if (res)
3253 		return res;
3254 
3255 	lock_sock(sk);
3256 
3257 	switch (opt) {
3258 	case TIPC_IMPORTANCE:
3259 		value = tsk_importance(tsk);
3260 		break;
3261 	case TIPC_SRC_DROPPABLE:
3262 		value = tsk_unreliable(tsk);
3263 		break;
3264 	case TIPC_DEST_DROPPABLE:
3265 		value = tsk_unreturnable(tsk);
3266 		break;
3267 	case TIPC_CONN_TIMEOUT:
3268 		value = tsk->conn_timeout;
3269 		/* no need to set "res", since already 0 at this point */
3270 		break;
3271 	case TIPC_NODE_RECVQ_DEPTH:
3272 		value = 0; /* was tipc_queue_size, now obsolete */
3273 		break;
3274 	case TIPC_SOCK_RECVQ_DEPTH:
3275 		value = skb_queue_len(&sk->sk_receive_queue);
3276 		break;
3277 	case TIPC_SOCK_RECVQ_USED:
3278 		value = sk_rmem_alloc_get(sk);
3279 		break;
3280 	case TIPC_GROUP_JOIN:
3281 		seq.type = 0;
3282 		if (tsk->group)
3283 			tipc_group_self(tsk->group, &seq, &scope);
3284 		value = seq.type;
3285 		break;
3286 	default:
3287 		res = -EINVAL;
3288 	}
3289 
3290 	release_sock(sk);
3291 
3292 	if (res)
3293 		return res;	/* "get" failed */
3294 
3295 	if (len < sizeof(value))
3296 		return -EINVAL;
3297 
3298 	if (copy_to_user(ov, &value, sizeof(value)))
3299 		return -EFAULT;
3300 
3301 	return put_user(sizeof(value), ol);
3302 }
3303 
3304 static int tipc_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
3305 {
3306 	struct net *net = sock_net(sock->sk);
3307 	struct tipc_sioc_nodeid_req nr = {0};
3308 	struct tipc_sioc_ln_req lnr;
3309 	void __user *argp = (void __user *)arg;
3310 
3311 	switch (cmd) {
3312 	case SIOCGETLINKNAME:
3313 		if (copy_from_user(&lnr, argp, sizeof(lnr)))
3314 			return -EFAULT;
3315 		if (!tipc_node_get_linkname(net,
3316 					    lnr.bearer_id & 0xffff, lnr.peer,
3317 					    lnr.linkname, TIPC_MAX_LINK_NAME)) {
3318 			if (copy_to_user(argp, &lnr, sizeof(lnr)))
3319 				return -EFAULT;
3320 			return 0;
3321 		}
3322 		return -EADDRNOTAVAIL;
3323 	case SIOCGETNODEID:
3324 		if (copy_from_user(&nr, argp, sizeof(nr)))
3325 			return -EFAULT;
3326 		if (!tipc_node_get_id(net, nr.peer, nr.node_id))
3327 			return -EADDRNOTAVAIL;
3328 		if (copy_to_user(argp, &nr, sizeof(nr)))
3329 			return -EFAULT;
3330 		return 0;
3331 	default:
3332 		return -ENOIOCTLCMD;
3333 	}
3334 }
3335 
3336 static int tipc_socketpair(struct socket *sock1, struct socket *sock2)
3337 {
3338 	struct tipc_sock *tsk2 = tipc_sk(sock2->sk);
3339 	struct tipc_sock *tsk1 = tipc_sk(sock1->sk);
3340 	u32 onode = tipc_own_addr(sock_net(sock1->sk));
3341 
3342 	tsk1->peer.family = AF_TIPC;
3343 	tsk1->peer.addrtype = TIPC_SOCKET_ADDR;
3344 	tsk1->peer.scope = TIPC_NODE_SCOPE;
3345 	tsk1->peer.addr.id.ref = tsk2->portid;
3346 	tsk1->peer.addr.id.node = onode;
3347 	tsk2->peer.family = AF_TIPC;
3348 	tsk2->peer.addrtype = TIPC_SOCKET_ADDR;
3349 	tsk2->peer.scope = TIPC_NODE_SCOPE;
3350 	tsk2->peer.addr.id.ref = tsk1->portid;
3351 	tsk2->peer.addr.id.node = onode;
3352 
3353 	tipc_sk_finish_conn(tsk1, tsk2->portid, onode);
3354 	tipc_sk_finish_conn(tsk2, tsk1->portid, onode);
3355 	return 0;
3356 }
3357 
3358 /* Protocol switches for the various types of TIPC sockets */
3359 
3360 static const struct proto_ops msg_ops = {
3361 	.owner		= THIS_MODULE,
3362 	.family		= AF_TIPC,
3363 	.release	= tipc_release,
3364 	.bind		= tipc_bind,
3365 	.connect	= tipc_connect,
3366 	.socketpair	= tipc_socketpair,
3367 	.accept		= sock_no_accept,
3368 	.getname	= tipc_getname,
3369 	.poll		= tipc_poll,
3370 	.ioctl		= tipc_ioctl,
3371 	.listen		= sock_no_listen,
3372 	.shutdown	= tipc_shutdown,
3373 	.setsockopt	= tipc_setsockopt,
3374 	.getsockopt	= tipc_getsockopt,
3375 	.sendmsg	= tipc_sendmsg,
3376 	.recvmsg	= tipc_recvmsg,
3377 	.mmap		= sock_no_mmap,
3378 	.sendpage	= sock_no_sendpage
3379 };
3380 
3381 static const struct proto_ops packet_ops = {
3382 	.owner		= THIS_MODULE,
3383 	.family		= AF_TIPC,
3384 	.release	= tipc_release,
3385 	.bind		= tipc_bind,
3386 	.connect	= tipc_connect,
3387 	.socketpair	= tipc_socketpair,
3388 	.accept		= tipc_accept,
3389 	.getname	= tipc_getname,
3390 	.poll		= tipc_poll,
3391 	.ioctl		= tipc_ioctl,
3392 	.listen		= tipc_listen,
3393 	.shutdown	= tipc_shutdown,
3394 	.setsockopt	= tipc_setsockopt,
3395 	.getsockopt	= tipc_getsockopt,
3396 	.sendmsg	= tipc_send_packet,
3397 	.recvmsg	= tipc_recvmsg,
3398 	.mmap		= sock_no_mmap,
3399 	.sendpage	= sock_no_sendpage
3400 };
3401 
3402 static const struct proto_ops stream_ops = {
3403 	.owner		= THIS_MODULE,
3404 	.family		= AF_TIPC,
3405 	.release	= tipc_release,
3406 	.bind		= tipc_bind,
3407 	.connect	= tipc_connect,
3408 	.socketpair	= tipc_socketpair,
3409 	.accept		= tipc_accept,
3410 	.getname	= tipc_getname,
3411 	.poll		= tipc_poll,
3412 	.ioctl		= tipc_ioctl,
3413 	.listen		= tipc_listen,
3414 	.shutdown	= tipc_shutdown,
3415 	.setsockopt	= tipc_setsockopt,
3416 	.getsockopt	= tipc_getsockopt,
3417 	.sendmsg	= tipc_sendstream,
3418 	.recvmsg	= tipc_recvstream,
3419 	.mmap		= sock_no_mmap,
3420 	.sendpage	= sock_no_sendpage
3421 };
3422 
3423 static const struct net_proto_family tipc_family_ops = {
3424 	.owner		= THIS_MODULE,
3425 	.family		= AF_TIPC,
3426 	.create		= tipc_sk_create
3427 };
3428 
3429 static struct proto tipc_proto = {
3430 	.name		= "TIPC",
3431 	.owner		= THIS_MODULE,
3432 	.obj_size	= sizeof(struct tipc_sock),
3433 	.sysctl_rmem	= sysctl_tipc_rmem
3434 };
3435 
3436 /**
3437  * tipc_socket_init - initialize TIPC socket interface
3438  *
3439  * Return: 0 on success, errno otherwise
3440  */
3441 int tipc_socket_init(void)
3442 {
3443 	int res;
3444 
3445 	res = proto_register(&tipc_proto, 1);
3446 	if (res) {
3447 		pr_err("Failed to register TIPC protocol type\n");
3448 		goto out;
3449 	}
3450 
3451 	res = sock_register(&tipc_family_ops);
3452 	if (res) {
3453 		pr_err("Failed to register TIPC socket type\n");
3454 		proto_unregister(&tipc_proto);
3455 		goto out;
3456 	}
3457  out:
3458 	return res;
3459 }
3460 
3461 /**
3462  * tipc_socket_stop - stop TIPC socket interface
3463  */
3464 void tipc_socket_stop(void)
3465 {
3466 	sock_unregister(tipc_family_ops.family);
3467 	proto_unregister(&tipc_proto);
3468 }
3469 
3470 /* Caller should hold socket lock for the passed tipc socket. */
3471 static int __tipc_nl_add_sk_con(struct sk_buff *skb, struct tipc_sock *tsk)
3472 {
3473 	u32 peer_node, peer_port;
3474 	u32 conn_type, conn_instance;
3475 	struct nlattr *nest;
3476 
3477 	peer_node = tsk_peer_node(tsk);
3478 	peer_port = tsk_peer_port(tsk);
3479 	conn_type = msg_nametype(&tsk->phdr);
3480 	conn_instance = msg_nameinst(&tsk->phdr);
3481 	nest = nla_nest_start_noflag(skb, TIPC_NLA_SOCK_CON);
3482 	if (!nest)
3483 		return -EMSGSIZE;
3484 
3485 	if (nla_put_u32(skb, TIPC_NLA_CON_NODE, peer_node))
3486 		goto msg_full;
3487 	if (nla_put_u32(skb, TIPC_NLA_CON_SOCK, peer_port))
3488 		goto msg_full;
3489 
3490 	if (tsk->conn_addrtype != 0) {
3491 		if (nla_put_flag(skb, TIPC_NLA_CON_FLAG))
3492 			goto msg_full;
3493 		if (nla_put_u32(skb, TIPC_NLA_CON_TYPE, conn_type))
3494 			goto msg_full;
3495 		if (nla_put_u32(skb, TIPC_NLA_CON_INST, conn_instance))
3496 			goto msg_full;
3497 	}
3498 	nla_nest_end(skb, nest);
3499 
3500 	return 0;
3501 
3502 msg_full:
3503 	nla_nest_cancel(skb, nest);
3504 
3505 	return -EMSGSIZE;
3506 }
3507 
3508 static int __tipc_nl_add_sk_info(struct sk_buff *skb, struct tipc_sock
3509 			  *tsk)
3510 {
3511 	struct net *net = sock_net(skb->sk);
3512 	struct sock *sk = &tsk->sk;
3513 
3514 	if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->portid) ||
3515 	    nla_put_u32(skb, TIPC_NLA_SOCK_ADDR, tipc_own_addr(net)))
3516 		return -EMSGSIZE;
3517 
3518 	if (tipc_sk_connected(sk)) {
3519 		if (__tipc_nl_add_sk_con(skb, tsk))
3520 			return -EMSGSIZE;
3521 	} else if (!list_empty(&tsk->publications)) {
3522 		if (nla_put_flag(skb, TIPC_NLA_SOCK_HAS_PUBL))
3523 			return -EMSGSIZE;
3524 	}
3525 	return 0;
3526 }
3527 
3528 /* Caller should hold socket lock for the passed tipc socket. */
3529 static int __tipc_nl_add_sk(struct sk_buff *skb, struct netlink_callback *cb,
3530 			    struct tipc_sock *tsk)
3531 {
3532 	struct nlattr *attrs;
3533 	void *hdr;
3534 
3535 	hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
3536 			  &tipc_genl_family, NLM_F_MULTI, TIPC_NL_SOCK_GET);
3537 	if (!hdr)
3538 		goto msg_cancel;
3539 
3540 	attrs = nla_nest_start_noflag(skb, TIPC_NLA_SOCK);
3541 	if (!attrs)
3542 		goto genlmsg_cancel;
3543 
3544 	if (__tipc_nl_add_sk_info(skb, tsk))
3545 		goto attr_msg_cancel;
3546 
3547 	nla_nest_end(skb, attrs);
3548 	genlmsg_end(skb, hdr);
3549 
3550 	return 0;
3551 
3552 attr_msg_cancel:
3553 	nla_nest_cancel(skb, attrs);
3554 genlmsg_cancel:
3555 	genlmsg_cancel(skb, hdr);
3556 msg_cancel:
3557 	return -EMSGSIZE;
3558 }
3559 
3560 int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
3561 		    int (*skb_handler)(struct sk_buff *skb,
3562 				       struct netlink_callback *cb,
3563 				       struct tipc_sock *tsk))
3564 {
3565 	struct rhashtable_iter *iter = (void *)cb->args[4];
3566 	struct tipc_sock *tsk;
3567 	int err;
3568 
3569 	rhashtable_walk_start(iter);
3570 	while ((tsk = rhashtable_walk_next(iter)) != NULL) {
3571 		if (IS_ERR(tsk)) {
3572 			err = PTR_ERR(tsk);
3573 			if (err == -EAGAIN) {
3574 				err = 0;
3575 				continue;
3576 			}
3577 			break;
3578 		}
3579 
3580 		sock_hold(&tsk->sk);
3581 		rhashtable_walk_stop(iter);
3582 		lock_sock(&tsk->sk);
3583 		err = skb_handler(skb, cb, tsk);
3584 		if (err) {
3585 			release_sock(&tsk->sk);
3586 			sock_put(&tsk->sk);
3587 			goto out;
3588 		}
3589 		release_sock(&tsk->sk);
3590 		rhashtable_walk_start(iter);
3591 		sock_put(&tsk->sk);
3592 	}
3593 	rhashtable_walk_stop(iter);
3594 out:
3595 	return skb->len;
3596 }
3597 EXPORT_SYMBOL(tipc_nl_sk_walk);
3598 
3599 int tipc_dump_start(struct netlink_callback *cb)
3600 {
3601 	return __tipc_dump_start(cb, sock_net(cb->skb->sk));
3602 }
3603 EXPORT_SYMBOL(tipc_dump_start);
3604 
3605 int __tipc_dump_start(struct netlink_callback *cb, struct net *net)
3606 {
3607 	/* tipc_nl_name_table_dump() uses cb->args[0...3]. */
3608 	struct rhashtable_iter *iter = (void *)cb->args[4];
3609 	struct tipc_net *tn = tipc_net(net);
3610 
3611 	if (!iter) {
3612 		iter = kmalloc(sizeof(*iter), GFP_KERNEL);
3613 		if (!iter)
3614 			return -ENOMEM;
3615 
3616 		cb->args[4] = (long)iter;
3617 	}
3618 
3619 	rhashtable_walk_enter(&tn->sk_rht, iter);
3620 	return 0;
3621 }
3622 
3623 int tipc_dump_done(struct netlink_callback *cb)
3624 {
3625 	struct rhashtable_iter *hti = (void *)cb->args[4];
3626 
3627 	rhashtable_walk_exit(hti);
3628 	kfree(hti);
3629 	return 0;
3630 }
3631 EXPORT_SYMBOL(tipc_dump_done);
3632 
3633 int tipc_sk_fill_sock_diag(struct sk_buff *skb, struct netlink_callback *cb,
3634 			   struct tipc_sock *tsk, u32 sk_filter_state,
3635 			   u64 (*tipc_diag_gen_cookie)(struct sock *sk))
3636 {
3637 	struct sock *sk = &tsk->sk;
3638 	struct nlattr *attrs;
3639 	struct nlattr *stat;
3640 
3641 	/*filter response w.r.t sk_state*/
3642 	if (!(sk_filter_state & (1 << sk->sk_state)))
3643 		return 0;
3644 
3645 	attrs = nla_nest_start_noflag(skb, TIPC_NLA_SOCK);
3646 	if (!attrs)
3647 		goto msg_cancel;
3648 
3649 	if (__tipc_nl_add_sk_info(skb, tsk))
3650 		goto attr_msg_cancel;
3651 
3652 	if (nla_put_u32(skb, TIPC_NLA_SOCK_TYPE, (u32)sk->sk_type) ||
3653 	    nla_put_u32(skb, TIPC_NLA_SOCK_TIPC_STATE, (u32)sk->sk_state) ||
3654 	    nla_put_u32(skb, TIPC_NLA_SOCK_INO, sock_i_ino(sk)) ||
3655 	    nla_put_u32(skb, TIPC_NLA_SOCK_UID,
3656 			from_kuid_munged(sk_user_ns(NETLINK_CB(cb->skb).sk),
3657 					 sock_i_uid(sk))) ||
3658 	    nla_put_u64_64bit(skb, TIPC_NLA_SOCK_COOKIE,
3659 			      tipc_diag_gen_cookie(sk),
3660 			      TIPC_NLA_SOCK_PAD))
3661 		goto attr_msg_cancel;
3662 
3663 	stat = nla_nest_start_noflag(skb, TIPC_NLA_SOCK_STAT);
3664 	if (!stat)
3665 		goto attr_msg_cancel;
3666 
3667 	if (nla_put_u32(skb, TIPC_NLA_SOCK_STAT_RCVQ,
3668 			skb_queue_len(&sk->sk_receive_queue)) ||
3669 	    nla_put_u32(skb, TIPC_NLA_SOCK_STAT_SENDQ,
3670 			skb_queue_len(&sk->sk_write_queue)) ||
3671 	    nla_put_u32(skb, TIPC_NLA_SOCK_STAT_DROP,
3672 			atomic_read(&sk->sk_drops)))
3673 		goto stat_msg_cancel;
3674 
3675 	if (tsk->cong_link_cnt &&
3676 	    nla_put_flag(skb, TIPC_NLA_SOCK_STAT_LINK_CONG))
3677 		goto stat_msg_cancel;
3678 
3679 	if (tsk_conn_cong(tsk) &&
3680 	    nla_put_flag(skb, TIPC_NLA_SOCK_STAT_CONN_CONG))
3681 		goto stat_msg_cancel;
3682 
3683 	nla_nest_end(skb, stat);
3684 
3685 	if (tsk->group)
3686 		if (tipc_group_fill_sock_diag(tsk->group, skb))
3687 			goto stat_msg_cancel;
3688 
3689 	nla_nest_end(skb, attrs);
3690 
3691 	return 0;
3692 
3693 stat_msg_cancel:
3694 	nla_nest_cancel(skb, stat);
3695 attr_msg_cancel:
3696 	nla_nest_cancel(skb, attrs);
3697 msg_cancel:
3698 	return -EMSGSIZE;
3699 }
3700 EXPORT_SYMBOL(tipc_sk_fill_sock_diag);
3701 
3702 int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb)
3703 {
3704 	return tipc_nl_sk_walk(skb, cb, __tipc_nl_add_sk);
3705 }
3706 
3707 /* Caller should hold socket lock for the passed tipc socket. */
3708 static int __tipc_nl_add_sk_publ(struct sk_buff *skb,
3709 				 struct netlink_callback *cb,
3710 				 struct publication *publ)
3711 {
3712 	void *hdr;
3713 	struct nlattr *attrs;
3714 
3715 	hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
3716 			  &tipc_genl_family, NLM_F_MULTI, TIPC_NL_PUBL_GET);
3717 	if (!hdr)
3718 		goto msg_cancel;
3719 
3720 	attrs = nla_nest_start_noflag(skb, TIPC_NLA_PUBL);
3721 	if (!attrs)
3722 		goto genlmsg_cancel;
3723 
3724 	if (nla_put_u32(skb, TIPC_NLA_PUBL_KEY, publ->key))
3725 		goto attr_msg_cancel;
3726 	if (nla_put_u32(skb, TIPC_NLA_PUBL_TYPE, publ->sr.type))
3727 		goto attr_msg_cancel;
3728 	if (nla_put_u32(skb, TIPC_NLA_PUBL_LOWER, publ->sr.lower))
3729 		goto attr_msg_cancel;
3730 	if (nla_put_u32(skb, TIPC_NLA_PUBL_UPPER, publ->sr.upper))
3731 		goto attr_msg_cancel;
3732 
3733 	nla_nest_end(skb, attrs);
3734 	genlmsg_end(skb, hdr);
3735 
3736 	return 0;
3737 
3738 attr_msg_cancel:
3739 	nla_nest_cancel(skb, attrs);
3740 genlmsg_cancel:
3741 	genlmsg_cancel(skb, hdr);
3742 msg_cancel:
3743 	return -EMSGSIZE;
3744 }
3745 
3746 /* Caller should hold socket lock for the passed tipc socket. */
3747 static int __tipc_nl_list_sk_publ(struct sk_buff *skb,
3748 				  struct netlink_callback *cb,
3749 				  struct tipc_sock *tsk, u32 *last_publ)
3750 {
3751 	int err;
3752 	struct publication *p;
3753 
3754 	if (*last_publ) {
3755 		list_for_each_entry(p, &tsk->publications, binding_sock) {
3756 			if (p->key == *last_publ)
3757 				break;
3758 		}
3759 		if (list_entry_is_head(p, &tsk->publications, binding_sock)) {
3760 			/* We never set seq or call nl_dump_check_consistent()
3761 			 * this means that setting prev_seq here will cause the
3762 			 * consistence check to fail in the netlink callback
3763 			 * handler. Resulting in the last NLMSG_DONE message
3764 			 * having the NLM_F_DUMP_INTR flag set.
3765 			 */
3766 			cb->prev_seq = 1;
3767 			*last_publ = 0;
3768 			return -EPIPE;
3769 		}
3770 	} else {
3771 		p = list_first_entry(&tsk->publications, struct publication,
3772 				     binding_sock);
3773 	}
3774 
3775 	list_for_each_entry_from(p, &tsk->publications, binding_sock) {
3776 		err = __tipc_nl_add_sk_publ(skb, cb, p);
3777 		if (err) {
3778 			*last_publ = p->key;
3779 			return err;
3780 		}
3781 	}
3782 	*last_publ = 0;
3783 
3784 	return 0;
3785 }
3786 
3787 int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb)
3788 {
3789 	int err;
3790 	u32 tsk_portid = cb->args[0];
3791 	u32 last_publ = cb->args[1];
3792 	u32 done = cb->args[2];
3793 	struct net *net = sock_net(skb->sk);
3794 	struct tipc_sock *tsk;
3795 
3796 	if (!tsk_portid) {
3797 		struct nlattr **attrs = genl_dumpit_info(cb)->attrs;
3798 		struct nlattr *sock[TIPC_NLA_SOCK_MAX + 1];
3799 
3800 		if (!attrs[TIPC_NLA_SOCK])
3801 			return -EINVAL;
3802 
3803 		err = nla_parse_nested_deprecated(sock, TIPC_NLA_SOCK_MAX,
3804 						  attrs[TIPC_NLA_SOCK],
3805 						  tipc_nl_sock_policy, NULL);
3806 		if (err)
3807 			return err;
3808 
3809 		if (!sock[TIPC_NLA_SOCK_REF])
3810 			return -EINVAL;
3811 
3812 		tsk_portid = nla_get_u32(sock[TIPC_NLA_SOCK_REF]);
3813 	}
3814 
3815 	if (done)
3816 		return 0;
3817 
3818 	tsk = tipc_sk_lookup(net, tsk_portid);
3819 	if (!tsk)
3820 		return -EINVAL;
3821 
3822 	lock_sock(&tsk->sk);
3823 	err = __tipc_nl_list_sk_publ(skb, cb, tsk, &last_publ);
3824 	if (!err)
3825 		done = 1;
3826 	release_sock(&tsk->sk);
3827 	sock_put(&tsk->sk);
3828 
3829 	cb->args[0] = tsk_portid;
3830 	cb->args[1] = last_publ;
3831 	cb->args[2] = done;
3832 
3833 	return skb->len;
3834 }
3835 
3836 /**
3837  * tipc_sk_filtering - check if a socket should be traced
3838  * @sk: the socket to be examined
3839  *
3840  * @sysctl_tipc_sk_filter is used as the socket tuple for filtering:
3841  * (portid, sock type, name type, name lower, name upper)
3842  *
3843  * Return: true if the socket meets the socket tuple data
3844  * (value 0 = 'any') or when there is no tuple set (all = 0),
3845  * otherwise false
3846  */
3847 bool tipc_sk_filtering(struct sock *sk)
3848 {
3849 	struct tipc_sock *tsk;
3850 	struct publication *p;
3851 	u32 _port, _sktype, _type, _lower, _upper;
3852 	u32 type = 0, lower = 0, upper = 0;
3853 
3854 	if (!sk)
3855 		return true;
3856 
3857 	tsk = tipc_sk(sk);
3858 
3859 	_port = sysctl_tipc_sk_filter[0];
3860 	_sktype = sysctl_tipc_sk_filter[1];
3861 	_type = sysctl_tipc_sk_filter[2];
3862 	_lower = sysctl_tipc_sk_filter[3];
3863 	_upper = sysctl_tipc_sk_filter[4];
3864 
3865 	if (!_port && !_sktype && !_type && !_lower && !_upper)
3866 		return true;
3867 
3868 	if (_port)
3869 		return (_port == tsk->portid);
3870 
3871 	if (_sktype && _sktype != sk->sk_type)
3872 		return false;
3873 
3874 	if (tsk->published) {
3875 		p = list_first_entry_or_null(&tsk->publications,
3876 					     struct publication, binding_sock);
3877 		if (p) {
3878 			type = p->sr.type;
3879 			lower = p->sr.lower;
3880 			upper = p->sr.upper;
3881 		}
3882 	}
3883 
3884 	if (!tipc_sk_type_connectionless(sk)) {
3885 		type = msg_nametype(&tsk->phdr);
3886 		lower = msg_nameinst(&tsk->phdr);
3887 		upper = lower;
3888 	}
3889 
3890 	if ((_type && _type != type) || (_lower && _lower != lower) ||
3891 	    (_upper && _upper != upper))
3892 		return false;
3893 
3894 	return true;
3895 }
3896 
3897 u32 tipc_sock_get_portid(struct sock *sk)
3898 {
3899 	return (sk) ? (tipc_sk(sk))->portid : 0;
3900 }
3901 
3902 /**
3903  * tipc_sk_overlimit1 - check if socket rx queue is about to be overloaded,
3904  *			both the rcv and backlog queues are considered
3905  * @sk: tipc sk to be checked
3906  * @skb: tipc msg to be checked
3907  *
3908  * Return: true if the socket rx queue allocation is > 90%, otherwise false
3909  */
3910 
3911 bool tipc_sk_overlimit1(struct sock *sk, struct sk_buff *skb)
3912 {
3913 	atomic_t *dcnt = &tipc_sk(sk)->dupl_rcvcnt;
3914 	unsigned int lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt);
3915 	unsigned int qsize = sk->sk_backlog.len + sk_rmem_alloc_get(sk);
3916 
3917 	return (qsize > lim * 90 / 100);
3918 }
3919 
3920 /**
3921  * tipc_sk_overlimit2 - check if socket rx queue is about to be overloaded,
3922  *			only the rcv queue is considered
3923  * @sk: tipc sk to be checked
3924  * @skb: tipc msg to be checked
3925  *
3926  * Return: true if the socket rx queue allocation is > 90%, otherwise false
3927  */
3928 
3929 bool tipc_sk_overlimit2(struct sock *sk, struct sk_buff *skb)
3930 {
3931 	unsigned int lim = rcvbuf_limit(sk, skb);
3932 	unsigned int qsize = sk_rmem_alloc_get(sk);
3933 
3934 	return (qsize > lim * 90 / 100);
3935 }
3936 
3937 /**
3938  * tipc_sk_dump - dump TIPC socket
3939  * @sk: tipc sk to be dumped
3940  * @dqueues: bitmask to decide if any socket queue to be dumped?
3941  *           - TIPC_DUMP_NONE: don't dump socket queues
3942  *           - TIPC_DUMP_SK_SNDQ: dump socket send queue
3943  *           - TIPC_DUMP_SK_RCVQ: dump socket rcv queue
3944  *           - TIPC_DUMP_SK_BKLGQ: dump socket backlog queue
3945  *           - TIPC_DUMP_ALL: dump all the socket queues above
3946  * @buf: returned buffer of dump data in format
3947  */
3948 int tipc_sk_dump(struct sock *sk, u16 dqueues, char *buf)
3949 {
3950 	int i = 0;
3951 	size_t sz = (dqueues) ? SK_LMAX : SK_LMIN;
3952 	u32 conn_type, conn_instance;
3953 	struct tipc_sock *tsk;
3954 	struct publication *p;
3955 	bool tsk_connected;
3956 
3957 	if (!sk) {
3958 		i += scnprintf(buf, sz, "sk data: (null)\n");
3959 		return i;
3960 	}
3961 
3962 	tsk = tipc_sk(sk);
3963 	tsk_connected = !tipc_sk_type_connectionless(sk);
3964 
3965 	i += scnprintf(buf, sz, "sk data: %u", sk->sk_type);
3966 	i += scnprintf(buf + i, sz - i, " %d", sk->sk_state);
3967 	i += scnprintf(buf + i, sz - i, " %x", tsk_own_node(tsk));
3968 	i += scnprintf(buf + i, sz - i, " %u", tsk->portid);
3969 	i += scnprintf(buf + i, sz - i, " | %u", tsk_connected);
3970 	if (tsk_connected) {
3971 		i += scnprintf(buf + i, sz - i, " %x", tsk_peer_node(tsk));
3972 		i += scnprintf(buf + i, sz - i, " %u", tsk_peer_port(tsk));
3973 		conn_type = msg_nametype(&tsk->phdr);
3974 		conn_instance = msg_nameinst(&tsk->phdr);
3975 		i += scnprintf(buf + i, sz - i, " %u", conn_type);
3976 		i += scnprintf(buf + i, sz - i, " %u", conn_instance);
3977 	}
3978 	i += scnprintf(buf + i, sz - i, " | %u", tsk->published);
3979 	if (tsk->published) {
3980 		p = list_first_entry_or_null(&tsk->publications,
3981 					     struct publication, binding_sock);
3982 		i += scnprintf(buf + i, sz - i, " %u", (p) ? p->sr.type : 0);
3983 		i += scnprintf(buf + i, sz - i, " %u", (p) ? p->sr.lower : 0);
3984 		i += scnprintf(buf + i, sz - i, " %u", (p) ? p->sr.upper : 0);
3985 	}
3986 	i += scnprintf(buf + i, sz - i, " | %u", tsk->snd_win);
3987 	i += scnprintf(buf + i, sz - i, " %u", tsk->rcv_win);
3988 	i += scnprintf(buf + i, sz - i, " %u", tsk->max_pkt);
3989 	i += scnprintf(buf + i, sz - i, " %x", tsk->peer_caps);
3990 	i += scnprintf(buf + i, sz - i, " %u", tsk->cong_link_cnt);
3991 	i += scnprintf(buf + i, sz - i, " %u", tsk->snt_unacked);
3992 	i += scnprintf(buf + i, sz - i, " %u", tsk->rcv_unacked);
3993 	i += scnprintf(buf + i, sz - i, " %u", atomic_read(&tsk->dupl_rcvcnt));
3994 	i += scnprintf(buf + i, sz - i, " %u", sk->sk_shutdown);
3995 	i += scnprintf(buf + i, sz - i, " | %d", sk_wmem_alloc_get(sk));
3996 	i += scnprintf(buf + i, sz - i, " %d", sk->sk_sndbuf);
3997 	i += scnprintf(buf + i, sz - i, " | %d", sk_rmem_alloc_get(sk));
3998 	i += scnprintf(buf + i, sz - i, " %d", sk->sk_rcvbuf);
3999 	i += scnprintf(buf + i, sz - i, " | %d\n", READ_ONCE(sk->sk_backlog.len));
4000 
4001 	if (dqueues & TIPC_DUMP_SK_SNDQ) {
4002 		i += scnprintf(buf + i, sz - i, "sk_write_queue: ");
4003 		i += tipc_list_dump(&sk->sk_write_queue, false, buf + i);
4004 	}
4005 
4006 	if (dqueues & TIPC_DUMP_SK_RCVQ) {
4007 		i += scnprintf(buf + i, sz - i, "sk_receive_queue: ");
4008 		i += tipc_list_dump(&sk->sk_receive_queue, false, buf + i);
4009 	}
4010 
4011 	if (dqueues & TIPC_DUMP_SK_BKLGQ) {
4012 		i += scnprintf(buf + i, sz - i, "sk_backlog:\n  head ");
4013 		i += tipc_skb_dump(sk->sk_backlog.head, false, buf + i);
4014 		if (sk->sk_backlog.tail != sk->sk_backlog.head) {
4015 			i += scnprintf(buf + i, sz - i, "  tail ");
4016 			i += tipc_skb_dump(sk->sk_backlog.tail, false,
4017 					   buf + i);
4018 		}
4019 	}
4020 
4021 	return i;
4022 }
4023