xref: /openbmc/linux/net/tipc/socket.c (revision abe9af53)
1 /*
2  * net/tipc/socket.c: TIPC socket API
3  *
4  * Copyright (c) 2001-2007, 2012-2017, Ericsson AB
5  * Copyright (c) 2004-2008, 2010-2013, Wind River Systems
6  * All rights reserved.
7  *
8  * Redistribution and use in source and binary forms, with or without
9  * modification, are permitted provided that the following conditions are met:
10  *
11  * 1. Redistributions of source code must retain the above copyright
12  *    notice, this list of conditions and the following disclaimer.
13  * 2. Redistributions in binary form must reproduce the above copyright
14  *    notice, this list of conditions and the following disclaimer in the
15  *    documentation and/or other materials provided with the distribution.
16  * 3. Neither the names of the copyright holders nor the names of its
17  *    contributors may be used to endorse or promote products derived from
18  *    this software without specific prior written permission.
19  *
20  * Alternatively, this software may be distributed under the terms of the
21  * GNU General Public License ("GPL") version 2 as published by the Free
22  * Software Foundation.
23  *
24  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
25  * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27  * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
28  * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
29  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
30  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
31  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
32  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
33  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
34  * POSSIBILITY OF SUCH DAMAGE.
35  */
36 
37 #include <linux/rhashtable.h>
38 #include <linux/sched/signal.h>
39 
40 #include "core.h"
41 #include "name_table.h"
42 #include "node.h"
43 #include "link.h"
44 #include "name_distr.h"
45 #include "socket.h"
46 #include "bcast.h"
47 #include "netlink.h"
48 #include "group.h"
49 #include "trace.h"
50 
51 #define NAGLE_START_INIT	4
52 #define NAGLE_START_MAX		1024
53 #define CONN_TIMEOUT_DEFAULT    8000    /* default connect timeout = 8s */
54 #define CONN_PROBING_INTV	msecs_to_jiffies(3600000)  /* [ms] => 1 h */
55 #define TIPC_MAX_PORT		0xffffffff
56 #define TIPC_MIN_PORT		1
57 #define TIPC_ACK_RATE		4       /* ACK at 1/4 of rcv window size */
58 
59 enum {
60 	TIPC_LISTEN = TCP_LISTEN,
61 	TIPC_ESTABLISHED = TCP_ESTABLISHED,
62 	TIPC_OPEN = TCP_CLOSE,
63 	TIPC_DISCONNECTING = TCP_CLOSE_WAIT,
64 	TIPC_CONNECTING = TCP_SYN_SENT,
65 };
66 
67 struct sockaddr_pair {
68 	struct sockaddr_tipc sock;
69 	struct sockaddr_tipc member;
70 };
71 
72 /**
73  * struct tipc_sock - TIPC socket structure
74  * @sk: socket - interacts with 'port' and with user via the socket API
75  * @conn_type: TIPC type used when connection was established
76  * @conn_instance: TIPC instance used when connection was established
77  * @published: non-zero if port has one or more associated names
78  * @max_pkt: maximum packet size "hint" used when building messages sent by port
79  * @maxnagle: maximum size of msg which can be subject to nagle
80  * @portid: unique port identity in TIPC socket hash table
81  * @phdr: preformatted message header used when sending messages
82  * #cong_links: list of congested links
83  * @publications: list of publications for port
84  * @blocking_link: address of the congested link we are currently sleeping on
85  * @pub_count: total # of publications port has made during its lifetime
86  * @conn_timeout: the time we can wait for an unresponded setup request
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  * @rcv_unacked: # messages read by user, but not yet acked back to peer
91  * @peer: 'connected' peer for dgram/rdm
92  * @node: hash table node
93  * @mc_method: cookie for use between socket and broadcast layer
94  * @rcu: rcu struct for tipc_sock
95  */
96 struct tipc_sock {
97 	struct sock sk;
98 	u32 conn_type;
99 	u32 conn_instance;
100 	int published;
101 	u32 max_pkt;
102 	u32 maxnagle;
103 	u32 portid;
104 	struct tipc_msg phdr;
105 	struct list_head cong_links;
106 	struct list_head publications;
107 	u32 pub_count;
108 	atomic_t dupl_rcvcnt;
109 	u16 conn_timeout;
110 	bool probe_unacked;
111 	u16 cong_link_cnt;
112 	u16 snt_unacked;
113 	u16 snd_win;
114 	u16 peer_caps;
115 	u16 rcv_unacked;
116 	u16 rcv_win;
117 	struct sockaddr_tipc peer;
118 	struct rhash_head node;
119 	struct tipc_mc_method mc_method;
120 	struct rcu_head rcu;
121 	struct tipc_group *group;
122 	u32 oneway;
123 	u32 nagle_start;
124 	u16 snd_backlog;
125 	u16 msg_acc;
126 	u16 pkt_cnt;
127 	bool expect_ack;
128 	bool nodelay;
129 	bool group_is_open;
130 };
131 
132 static int tipc_sk_backlog_rcv(struct sock *sk, struct sk_buff *skb);
133 static void tipc_data_ready(struct sock *sk);
134 static void tipc_write_space(struct sock *sk);
135 static void tipc_sock_destruct(struct sock *sk);
136 static int tipc_release(struct socket *sock);
137 static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags,
138 		       bool kern);
139 static void tipc_sk_timeout(struct timer_list *t);
140 static int tipc_sk_publish(struct tipc_sock *tsk, uint scope,
141 			   struct tipc_name_seq const *seq);
142 static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope,
143 			    struct tipc_name_seq const *seq);
144 static int tipc_sk_leave(struct tipc_sock *tsk);
145 static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid);
146 static int tipc_sk_insert(struct tipc_sock *tsk);
147 static void tipc_sk_remove(struct tipc_sock *tsk);
148 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz);
149 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz);
150 static void tipc_sk_push_backlog(struct tipc_sock *tsk, bool nagle_ack);
151 
152 static const struct proto_ops packet_ops;
153 static const struct proto_ops stream_ops;
154 static const struct proto_ops msg_ops;
155 static struct proto tipc_proto;
156 static const struct rhashtable_params tsk_rht_params;
157 
158 static u32 tsk_own_node(struct tipc_sock *tsk)
159 {
160 	return msg_prevnode(&tsk->phdr);
161 }
162 
163 static u32 tsk_peer_node(struct tipc_sock *tsk)
164 {
165 	return msg_destnode(&tsk->phdr);
166 }
167 
168 static u32 tsk_peer_port(struct tipc_sock *tsk)
169 {
170 	return msg_destport(&tsk->phdr);
171 }
172 
173 static  bool tsk_unreliable(struct tipc_sock *tsk)
174 {
175 	return msg_src_droppable(&tsk->phdr) != 0;
176 }
177 
178 static void tsk_set_unreliable(struct tipc_sock *tsk, bool unreliable)
179 {
180 	msg_set_src_droppable(&tsk->phdr, unreliable ? 1 : 0);
181 }
182 
183 static bool tsk_unreturnable(struct tipc_sock *tsk)
184 {
185 	return msg_dest_droppable(&tsk->phdr) != 0;
186 }
187 
188 static void tsk_set_unreturnable(struct tipc_sock *tsk, bool unreturnable)
189 {
190 	msg_set_dest_droppable(&tsk->phdr, unreturnable ? 1 : 0);
191 }
192 
193 static int tsk_importance(struct tipc_sock *tsk)
194 {
195 	return msg_importance(&tsk->phdr);
196 }
197 
198 static struct tipc_sock *tipc_sk(const struct sock *sk)
199 {
200 	return container_of(sk, struct tipc_sock, sk);
201 }
202 
203 int tsk_set_importance(struct sock *sk, int imp)
204 {
205 	if (imp > TIPC_CRITICAL_IMPORTANCE)
206 		return -EINVAL;
207 	msg_set_importance(&tipc_sk(sk)->phdr, (u32)imp);
208 	return 0;
209 }
210 
211 static bool tsk_conn_cong(struct tipc_sock *tsk)
212 {
213 	return tsk->snt_unacked > tsk->snd_win;
214 }
215 
216 static u16 tsk_blocks(int len)
217 {
218 	return ((len / FLOWCTL_BLK_SZ) + 1);
219 }
220 
221 /* tsk_blocks(): translate a buffer size in bytes to number of
222  * advertisable blocks, taking into account the ratio truesize(len)/len
223  * We can trust that this ratio is always < 4 for len >= FLOWCTL_BLK_SZ
224  */
225 static u16 tsk_adv_blocks(int len)
226 {
227 	return len / FLOWCTL_BLK_SZ / 4;
228 }
229 
230 /* tsk_inc(): increment counter for sent or received data
231  * - If block based flow control is not supported by peer we
232  *   fall back to message based ditto, incrementing the counter
233  */
234 static u16 tsk_inc(struct tipc_sock *tsk, int msglen)
235 {
236 	if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL))
237 		return ((msglen / FLOWCTL_BLK_SZ) + 1);
238 	return 1;
239 }
240 
241 /* tsk_set_nagle - enable/disable nagle property by manipulating maxnagle
242  */
243 static void tsk_set_nagle(struct tipc_sock *tsk)
244 {
245 	struct sock *sk = &tsk->sk;
246 
247 	tsk->maxnagle = 0;
248 	if (sk->sk_type != SOCK_STREAM)
249 		return;
250 	if (tsk->nodelay)
251 		return;
252 	if (!(tsk->peer_caps & TIPC_NAGLE))
253 		return;
254 	/* Limit node local buffer size to avoid receive queue overflow */
255 	if (tsk->max_pkt == MAX_MSG_SIZE)
256 		tsk->maxnagle = 1500;
257 	else
258 		tsk->maxnagle = tsk->max_pkt;
259 }
260 
261 /**
262  * tsk_advance_rx_queue - discard first buffer in socket receive queue
263  *
264  * Caller must hold socket lock
265  */
266 static void tsk_advance_rx_queue(struct sock *sk)
267 {
268 	trace_tipc_sk_advance_rx(sk, NULL, TIPC_DUMP_SK_RCVQ, " ");
269 	kfree_skb(__skb_dequeue(&sk->sk_receive_queue));
270 }
271 
272 /* tipc_sk_respond() : send response message back to sender
273  */
274 static void tipc_sk_respond(struct sock *sk, struct sk_buff *skb, int err)
275 {
276 	u32 selector;
277 	u32 dnode;
278 	u32 onode = tipc_own_addr(sock_net(sk));
279 
280 	if (!tipc_msg_reverse(onode, &skb, err))
281 		return;
282 
283 	trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_NONE, "@sk_respond!");
284 	dnode = msg_destnode(buf_msg(skb));
285 	selector = msg_origport(buf_msg(skb));
286 	tipc_node_xmit_skb(sock_net(sk), skb, dnode, selector);
287 }
288 
289 /**
290  * tsk_rej_rx_queue - reject all buffers in socket receive queue
291  *
292  * Caller must hold socket lock
293  */
294 static void tsk_rej_rx_queue(struct sock *sk, int error)
295 {
296 	struct sk_buff *skb;
297 
298 	while ((skb = __skb_dequeue(&sk->sk_receive_queue)))
299 		tipc_sk_respond(sk, skb, error);
300 }
301 
302 static bool tipc_sk_connected(struct sock *sk)
303 {
304 	return sk->sk_state == TIPC_ESTABLISHED;
305 }
306 
307 /* tipc_sk_type_connectionless - check if the socket is datagram socket
308  * @sk: socket
309  *
310  * Returns true if connection less, false otherwise
311  */
312 static bool tipc_sk_type_connectionless(struct sock *sk)
313 {
314 	return sk->sk_type == SOCK_RDM || sk->sk_type == SOCK_DGRAM;
315 }
316 
317 /* tsk_peer_msg - verify if message was sent by connected port's peer
318  *
319  * Handles cases where the node's network address has changed from
320  * the default of <0.0.0> to its configured setting.
321  */
322 static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg)
323 {
324 	struct sock *sk = &tsk->sk;
325 	u32 self = tipc_own_addr(sock_net(sk));
326 	u32 peer_port = tsk_peer_port(tsk);
327 	u32 orig_node, peer_node;
328 
329 	if (unlikely(!tipc_sk_connected(sk)))
330 		return false;
331 
332 	if (unlikely(msg_origport(msg) != peer_port))
333 		return false;
334 
335 	orig_node = msg_orignode(msg);
336 	peer_node = tsk_peer_node(tsk);
337 
338 	if (likely(orig_node == peer_node))
339 		return true;
340 
341 	if (!orig_node && peer_node == self)
342 		return true;
343 
344 	if (!peer_node && orig_node == self)
345 		return true;
346 
347 	return false;
348 }
349 
350 /* tipc_set_sk_state - set the sk_state of the socket
351  * @sk: socket
352  *
353  * Caller must hold socket lock
354  *
355  * Returns 0 on success, errno otherwise
356  */
357 static int tipc_set_sk_state(struct sock *sk, int state)
358 {
359 	int oldsk_state = sk->sk_state;
360 	int res = -EINVAL;
361 
362 	switch (state) {
363 	case TIPC_OPEN:
364 		res = 0;
365 		break;
366 	case TIPC_LISTEN:
367 	case TIPC_CONNECTING:
368 		if (oldsk_state == TIPC_OPEN)
369 			res = 0;
370 		break;
371 	case TIPC_ESTABLISHED:
372 		if (oldsk_state == TIPC_CONNECTING ||
373 		    oldsk_state == TIPC_OPEN)
374 			res = 0;
375 		break;
376 	case TIPC_DISCONNECTING:
377 		if (oldsk_state == TIPC_CONNECTING ||
378 		    oldsk_state == TIPC_ESTABLISHED)
379 			res = 0;
380 		break;
381 	}
382 
383 	if (!res)
384 		sk->sk_state = state;
385 
386 	return res;
387 }
388 
389 static int tipc_sk_sock_err(struct socket *sock, long *timeout)
390 {
391 	struct sock *sk = sock->sk;
392 	int err = sock_error(sk);
393 	int typ = sock->type;
394 
395 	if (err)
396 		return err;
397 	if (typ == SOCK_STREAM || typ == SOCK_SEQPACKET) {
398 		if (sk->sk_state == TIPC_DISCONNECTING)
399 			return -EPIPE;
400 		else if (!tipc_sk_connected(sk))
401 			return -ENOTCONN;
402 	}
403 	if (!*timeout)
404 		return -EAGAIN;
405 	if (signal_pending(current))
406 		return sock_intr_errno(*timeout);
407 
408 	return 0;
409 }
410 
411 #define tipc_wait_for_cond(sock_, timeo_, condition_)			       \
412 ({                                                                             \
413 	DEFINE_WAIT_FUNC(wait_, woken_wake_function);                          \
414 	struct sock *sk_;						       \
415 	int rc_;							       \
416 									       \
417 	while ((rc_ = !(condition_))) {					       \
418 		/* coupled with smp_wmb() in tipc_sk_proto_rcv() */            \
419 		smp_rmb();                                                     \
420 		sk_ = (sock_)->sk;					       \
421 		rc_ = tipc_sk_sock_err((sock_), timeo_);		       \
422 		if (rc_)						       \
423 			break;						       \
424 		add_wait_queue(sk_sleep(sk_), &wait_);                         \
425 		release_sock(sk_);					       \
426 		*(timeo_) = wait_woken(&wait_, TASK_INTERRUPTIBLE, *(timeo_)); \
427 		sched_annotate_sleep();				               \
428 		lock_sock(sk_);						       \
429 		remove_wait_queue(sk_sleep(sk_), &wait_);		       \
430 	}								       \
431 	rc_;								       \
432 })
433 
434 /**
435  * tipc_sk_create - create a TIPC socket
436  * @net: network namespace (must be default network)
437  * @sock: pre-allocated socket structure
438  * @protocol: protocol indicator (must be 0)
439  * @kern: caused by kernel or by userspace?
440  *
441  * This routine creates additional data structures used by the TIPC socket,
442  * initializes them, and links them together.
443  *
444  * Returns 0 on success, errno otherwise
445  */
446 static int tipc_sk_create(struct net *net, struct socket *sock,
447 			  int protocol, int kern)
448 {
449 	const struct proto_ops *ops;
450 	struct sock *sk;
451 	struct tipc_sock *tsk;
452 	struct tipc_msg *msg;
453 
454 	/* Validate arguments */
455 	if (unlikely(protocol != 0))
456 		return -EPROTONOSUPPORT;
457 
458 	switch (sock->type) {
459 	case SOCK_STREAM:
460 		ops = &stream_ops;
461 		break;
462 	case SOCK_SEQPACKET:
463 		ops = &packet_ops;
464 		break;
465 	case SOCK_DGRAM:
466 	case SOCK_RDM:
467 		ops = &msg_ops;
468 		break;
469 	default:
470 		return -EPROTOTYPE;
471 	}
472 
473 	/* Allocate socket's protocol area */
474 	sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto, kern);
475 	if (sk == NULL)
476 		return -ENOMEM;
477 
478 	tsk = tipc_sk(sk);
479 	tsk->max_pkt = MAX_PKT_DEFAULT;
480 	tsk->maxnagle = 0;
481 	tsk->nagle_start = NAGLE_START_INIT;
482 	INIT_LIST_HEAD(&tsk->publications);
483 	INIT_LIST_HEAD(&tsk->cong_links);
484 	msg = &tsk->phdr;
485 
486 	/* Finish initializing socket data structures */
487 	sock->ops = ops;
488 	sock_init_data(sock, sk);
489 	tipc_set_sk_state(sk, TIPC_OPEN);
490 	if (tipc_sk_insert(tsk)) {
491 		pr_warn("Socket create failed; port number exhausted\n");
492 		return -EINVAL;
493 	}
494 
495 	/* Ensure tsk is visible before we read own_addr. */
496 	smp_mb();
497 
498 	tipc_msg_init(tipc_own_addr(net), msg, TIPC_LOW_IMPORTANCE,
499 		      TIPC_NAMED_MSG, NAMED_H_SIZE, 0);
500 
501 	msg_set_origport(msg, tsk->portid);
502 	timer_setup(&sk->sk_timer, tipc_sk_timeout, 0);
503 	sk->sk_shutdown = 0;
504 	sk->sk_backlog_rcv = tipc_sk_backlog_rcv;
505 	sk->sk_rcvbuf = sysctl_tipc_rmem[1];
506 	sk->sk_data_ready = tipc_data_ready;
507 	sk->sk_write_space = tipc_write_space;
508 	sk->sk_destruct = tipc_sock_destruct;
509 	tsk->conn_timeout = CONN_TIMEOUT_DEFAULT;
510 	tsk->group_is_open = true;
511 	atomic_set(&tsk->dupl_rcvcnt, 0);
512 
513 	/* Start out with safe limits until we receive an advertised window */
514 	tsk->snd_win = tsk_adv_blocks(RCVBUF_MIN);
515 	tsk->rcv_win = tsk->snd_win;
516 
517 	if (tipc_sk_type_connectionless(sk)) {
518 		tsk_set_unreturnable(tsk, true);
519 		if (sock->type == SOCK_DGRAM)
520 			tsk_set_unreliable(tsk, true);
521 	}
522 	__skb_queue_head_init(&tsk->mc_method.deferredq);
523 	trace_tipc_sk_create(sk, NULL, TIPC_DUMP_NONE, " ");
524 	return 0;
525 }
526 
527 static void tipc_sk_callback(struct rcu_head *head)
528 {
529 	struct tipc_sock *tsk = container_of(head, struct tipc_sock, rcu);
530 
531 	sock_put(&tsk->sk);
532 }
533 
534 /* Caller should hold socket lock for the socket. */
535 static void __tipc_shutdown(struct socket *sock, int error)
536 {
537 	struct sock *sk = sock->sk;
538 	struct tipc_sock *tsk = tipc_sk(sk);
539 	struct net *net = sock_net(sk);
540 	long timeout = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT);
541 	u32 dnode = tsk_peer_node(tsk);
542 	struct sk_buff *skb;
543 
544 	/* Avoid that hi-prio shutdown msgs bypass msgs in link wakeup queue */
545 	tipc_wait_for_cond(sock, &timeout, (!tsk->cong_link_cnt &&
546 					    !tsk_conn_cong(tsk)));
547 
548 	/* Push out delayed messages if in Nagle mode */
549 	tipc_sk_push_backlog(tsk, false);
550 	/* Remove pending SYN */
551 	__skb_queue_purge(&sk->sk_write_queue);
552 
553 	/* Remove partially received buffer if any */
554 	skb = skb_peek(&sk->sk_receive_queue);
555 	if (skb && TIPC_SKB_CB(skb)->bytes_read) {
556 		__skb_unlink(skb, &sk->sk_receive_queue);
557 		kfree_skb(skb);
558 	}
559 
560 	/* Reject all unreceived messages if connectionless */
561 	if (tipc_sk_type_connectionless(sk)) {
562 		tsk_rej_rx_queue(sk, error);
563 		return;
564 	}
565 
566 	switch (sk->sk_state) {
567 	case TIPC_CONNECTING:
568 	case TIPC_ESTABLISHED:
569 		tipc_set_sk_state(sk, TIPC_DISCONNECTING);
570 		tipc_node_remove_conn(net, dnode, tsk->portid);
571 		/* Send a FIN+/- to its peer */
572 		skb = __skb_dequeue(&sk->sk_receive_queue);
573 		if (skb) {
574 			__skb_queue_purge(&sk->sk_receive_queue);
575 			tipc_sk_respond(sk, skb, error);
576 			break;
577 		}
578 		skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE,
579 				      TIPC_CONN_MSG, SHORT_H_SIZE, 0, dnode,
580 				      tsk_own_node(tsk), tsk_peer_port(tsk),
581 				      tsk->portid, error);
582 		if (skb)
583 			tipc_node_xmit_skb(net, skb, dnode, tsk->portid);
584 		break;
585 	case TIPC_LISTEN:
586 		/* Reject all SYN messages */
587 		tsk_rej_rx_queue(sk, error);
588 		break;
589 	default:
590 		__skb_queue_purge(&sk->sk_receive_queue);
591 		break;
592 	}
593 }
594 
595 /**
596  * tipc_release - destroy a TIPC socket
597  * @sock: socket to destroy
598  *
599  * This routine cleans up any messages that are still queued on the socket.
600  * For DGRAM and RDM socket types, all queued messages are rejected.
601  * For SEQPACKET and STREAM socket types, the first message is rejected
602  * and any others are discarded.  (If the first message on a STREAM socket
603  * is partially-read, it is discarded and the next one is rejected instead.)
604  *
605  * NOTE: Rejected messages are not necessarily returned to the sender!  They
606  * are returned or discarded according to the "destination droppable" setting
607  * specified for the message by the sender.
608  *
609  * Returns 0 on success, errno otherwise
610  */
611 static int tipc_release(struct socket *sock)
612 {
613 	struct sock *sk = sock->sk;
614 	struct tipc_sock *tsk;
615 
616 	/*
617 	 * Exit if socket isn't fully initialized (occurs when a failed accept()
618 	 * releases a pre-allocated child socket that was never used)
619 	 */
620 	if (sk == NULL)
621 		return 0;
622 
623 	tsk = tipc_sk(sk);
624 	lock_sock(sk);
625 
626 	trace_tipc_sk_release(sk, NULL, TIPC_DUMP_ALL, " ");
627 	__tipc_shutdown(sock, TIPC_ERR_NO_PORT);
628 	sk->sk_shutdown = SHUTDOWN_MASK;
629 	tipc_sk_leave(tsk);
630 	tipc_sk_withdraw(tsk, 0, NULL);
631 	__skb_queue_purge(&tsk->mc_method.deferredq);
632 	sk_stop_timer(sk, &sk->sk_timer);
633 	tipc_sk_remove(tsk);
634 
635 	sock_orphan(sk);
636 	/* Reject any messages that accumulated in backlog queue */
637 	release_sock(sk);
638 	tipc_dest_list_purge(&tsk->cong_links);
639 	tsk->cong_link_cnt = 0;
640 	call_rcu(&tsk->rcu, tipc_sk_callback);
641 	sock->sk = NULL;
642 
643 	return 0;
644 }
645 
646 /**
647  * tipc_bind - associate or disassocate TIPC name(s) with a socket
648  * @sock: socket structure
649  * @uaddr: socket address describing name(s) and desired operation
650  * @uaddr_len: size of socket address data structure
651  *
652  * Name and name sequence binding is indicated using a positive scope value;
653  * a negative scope value unbinds the specified name.  Specifying no name
654  * (i.e. a socket address length of 0) unbinds all names from the socket.
655  *
656  * Returns 0 on success, errno otherwise
657  *
658  * NOTE: This routine doesn't need to take the socket lock since it doesn't
659  *       access any non-constant socket information.
660  */
661 static int tipc_bind(struct socket *sock, struct sockaddr *uaddr,
662 		     int uaddr_len)
663 {
664 	struct sock *sk = sock->sk;
665 	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
666 	struct tipc_sock *tsk = tipc_sk(sk);
667 	int res = -EINVAL;
668 
669 	lock_sock(sk);
670 	if (unlikely(!uaddr_len)) {
671 		res = tipc_sk_withdraw(tsk, 0, NULL);
672 		goto exit;
673 	}
674 	if (tsk->group) {
675 		res = -EACCES;
676 		goto exit;
677 	}
678 	if (uaddr_len < sizeof(struct sockaddr_tipc)) {
679 		res = -EINVAL;
680 		goto exit;
681 	}
682 	if (addr->family != AF_TIPC) {
683 		res = -EAFNOSUPPORT;
684 		goto exit;
685 	}
686 
687 	if (addr->addrtype == TIPC_ADDR_NAME)
688 		addr->addr.nameseq.upper = addr->addr.nameseq.lower;
689 	else if (addr->addrtype != TIPC_ADDR_NAMESEQ) {
690 		res = -EAFNOSUPPORT;
691 		goto exit;
692 	}
693 
694 	if ((addr->addr.nameseq.type < TIPC_RESERVED_TYPES) &&
695 	    (addr->addr.nameseq.type != TIPC_TOP_SRV) &&
696 	    (addr->addr.nameseq.type != TIPC_CFG_SRV)) {
697 		res = -EACCES;
698 		goto exit;
699 	}
700 
701 	res = (addr->scope >= 0) ?
702 		tipc_sk_publish(tsk, addr->scope, &addr->addr.nameseq) :
703 		tipc_sk_withdraw(tsk, -addr->scope, &addr->addr.nameseq);
704 exit:
705 	release_sock(sk);
706 	return res;
707 }
708 
709 /**
710  * tipc_getname - get port ID of socket or peer socket
711  * @sock: socket structure
712  * @uaddr: area for returned socket address
713  * @peer: 0 = own ID, 1 = current peer ID, 2 = current/former peer ID
714  *
715  * Returns 0 on success, errno otherwise
716  *
717  * NOTE: This routine doesn't need to take the socket lock since it only
718  *       accesses socket information that is unchanging (or which changes in
719  *       a completely predictable manner).
720  */
721 static int tipc_getname(struct socket *sock, struct sockaddr *uaddr,
722 			int peer)
723 {
724 	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
725 	struct sock *sk = sock->sk;
726 	struct tipc_sock *tsk = tipc_sk(sk);
727 
728 	memset(addr, 0, sizeof(*addr));
729 	if (peer) {
730 		if ((!tipc_sk_connected(sk)) &&
731 		    ((peer != 2) || (sk->sk_state != TIPC_DISCONNECTING)))
732 			return -ENOTCONN;
733 		addr->addr.id.ref = tsk_peer_port(tsk);
734 		addr->addr.id.node = tsk_peer_node(tsk);
735 	} else {
736 		addr->addr.id.ref = tsk->portid;
737 		addr->addr.id.node = tipc_own_addr(sock_net(sk));
738 	}
739 
740 	addr->addrtype = TIPC_ADDR_ID;
741 	addr->family = AF_TIPC;
742 	addr->scope = 0;
743 	addr->addr.name.domain = 0;
744 
745 	return sizeof(*addr);
746 }
747 
748 /**
749  * tipc_poll - read and possibly block on pollmask
750  * @file: file structure associated with the socket
751  * @sock: socket for which to calculate the poll bits
752  * @wait: ???
753  *
754  * Returns pollmask value
755  *
756  * COMMENTARY:
757  * It appears that the usual socket locking mechanisms are not useful here
758  * since the pollmask info is potentially out-of-date the moment this routine
759  * exits.  TCP and other protocols seem to rely on higher level poll routines
760  * to handle any preventable race conditions, so TIPC will do the same ...
761  *
762  * IMPORTANT: The fact that a read or write operation is indicated does NOT
763  * imply that the operation will succeed, merely that it should be performed
764  * and will not block.
765  */
766 static __poll_t tipc_poll(struct file *file, struct socket *sock,
767 			      poll_table *wait)
768 {
769 	struct sock *sk = sock->sk;
770 	struct tipc_sock *tsk = tipc_sk(sk);
771 	__poll_t revents = 0;
772 
773 	sock_poll_wait(file, sock, wait);
774 	trace_tipc_sk_poll(sk, NULL, TIPC_DUMP_ALL, " ");
775 
776 	if (sk->sk_shutdown & RCV_SHUTDOWN)
777 		revents |= EPOLLRDHUP | EPOLLIN | EPOLLRDNORM;
778 	if (sk->sk_shutdown == SHUTDOWN_MASK)
779 		revents |= EPOLLHUP;
780 
781 	switch (sk->sk_state) {
782 	case TIPC_ESTABLISHED:
783 		if (!tsk->cong_link_cnt && !tsk_conn_cong(tsk))
784 			revents |= EPOLLOUT;
785 		fallthrough;
786 	case TIPC_LISTEN:
787 	case TIPC_CONNECTING:
788 		if (!skb_queue_empty_lockless(&sk->sk_receive_queue))
789 			revents |= EPOLLIN | EPOLLRDNORM;
790 		break;
791 	case TIPC_OPEN:
792 		if (tsk->group_is_open && !tsk->cong_link_cnt)
793 			revents |= EPOLLOUT;
794 		if (!tipc_sk_type_connectionless(sk))
795 			break;
796 		if (skb_queue_empty_lockless(&sk->sk_receive_queue))
797 			break;
798 		revents |= EPOLLIN | EPOLLRDNORM;
799 		break;
800 	case TIPC_DISCONNECTING:
801 		revents = EPOLLIN | EPOLLRDNORM | EPOLLHUP;
802 		break;
803 	}
804 	return revents;
805 }
806 
807 /**
808  * tipc_sendmcast - send multicast message
809  * @sock: socket structure
810  * @seq: destination address
811  * @msg: message to send
812  * @dlen: length of data to send
813  * @timeout: timeout to wait for wakeup
814  *
815  * Called from function tipc_sendmsg(), which has done all sanity checks
816  * Returns the number of bytes sent on success, or errno
817  */
818 static int tipc_sendmcast(struct  socket *sock, struct tipc_name_seq *seq,
819 			  struct msghdr *msg, size_t dlen, long timeout)
820 {
821 	struct sock *sk = sock->sk;
822 	struct tipc_sock *tsk = tipc_sk(sk);
823 	struct tipc_msg *hdr = &tsk->phdr;
824 	struct net *net = sock_net(sk);
825 	int mtu = tipc_bcast_get_mtu(net);
826 	struct tipc_mc_method *method = &tsk->mc_method;
827 	struct sk_buff_head pkts;
828 	struct tipc_nlist dsts;
829 	int rc;
830 
831 	if (tsk->group)
832 		return -EACCES;
833 
834 	/* Block or return if any destination link is congested */
835 	rc = tipc_wait_for_cond(sock, &timeout, !tsk->cong_link_cnt);
836 	if (unlikely(rc))
837 		return rc;
838 
839 	/* Lookup destination nodes */
840 	tipc_nlist_init(&dsts, tipc_own_addr(net));
841 	tipc_nametbl_lookup_dst_nodes(net, seq->type, seq->lower,
842 				      seq->upper, &dsts);
843 	if (!dsts.local && !dsts.remote)
844 		return -EHOSTUNREACH;
845 
846 	/* Build message header */
847 	msg_set_type(hdr, TIPC_MCAST_MSG);
848 	msg_set_hdr_sz(hdr, MCAST_H_SIZE);
849 	msg_set_lookup_scope(hdr, TIPC_CLUSTER_SCOPE);
850 	msg_set_destport(hdr, 0);
851 	msg_set_destnode(hdr, 0);
852 	msg_set_nametype(hdr, seq->type);
853 	msg_set_namelower(hdr, seq->lower);
854 	msg_set_nameupper(hdr, seq->upper);
855 
856 	/* Build message as chain of buffers */
857 	__skb_queue_head_init(&pkts);
858 	rc = tipc_msg_build(hdr, msg, 0, dlen, mtu, &pkts);
859 
860 	/* Send message if build was successful */
861 	if (unlikely(rc == dlen)) {
862 		trace_tipc_sk_sendmcast(sk, skb_peek(&pkts),
863 					TIPC_DUMP_SK_SNDQ, " ");
864 		rc = tipc_mcast_xmit(net, &pkts, method, &dsts,
865 				     &tsk->cong_link_cnt);
866 	}
867 
868 	tipc_nlist_purge(&dsts);
869 
870 	return rc ? rc : dlen;
871 }
872 
873 /**
874  * tipc_send_group_msg - send a message to a member in the group
875  * @net: network namespace
876  * @m: message to send
877  * @mb: group member
878  * @dnode: destination node
879  * @dport: destination port
880  * @dlen: total length of message data
881  */
882 static int tipc_send_group_msg(struct net *net, struct tipc_sock *tsk,
883 			       struct msghdr *m, struct tipc_member *mb,
884 			       u32 dnode, u32 dport, int dlen)
885 {
886 	u16 bc_snd_nxt = tipc_group_bc_snd_nxt(tsk->group);
887 	struct tipc_mc_method *method = &tsk->mc_method;
888 	int blks = tsk_blocks(GROUP_H_SIZE + dlen);
889 	struct tipc_msg *hdr = &tsk->phdr;
890 	struct sk_buff_head pkts;
891 	int mtu, rc;
892 
893 	/* Complete message header */
894 	msg_set_type(hdr, TIPC_GRP_UCAST_MSG);
895 	msg_set_hdr_sz(hdr, GROUP_H_SIZE);
896 	msg_set_destport(hdr, dport);
897 	msg_set_destnode(hdr, dnode);
898 	msg_set_grp_bc_seqno(hdr, bc_snd_nxt);
899 
900 	/* Build message as chain of buffers */
901 	__skb_queue_head_init(&pkts);
902 	mtu = tipc_node_get_mtu(net, dnode, tsk->portid, false);
903 	rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
904 	if (unlikely(rc != dlen))
905 		return rc;
906 
907 	/* Send message */
908 	rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid);
909 	if (unlikely(rc == -ELINKCONG)) {
910 		tipc_dest_push(&tsk->cong_links, dnode, 0);
911 		tsk->cong_link_cnt++;
912 	}
913 
914 	/* Update send window */
915 	tipc_group_update_member(mb, blks);
916 
917 	/* A broadcast sent within next EXPIRE period must follow same path */
918 	method->rcast = true;
919 	method->mandatory = true;
920 	return dlen;
921 }
922 
923 /**
924  * tipc_send_group_unicast - send message to a member in the group
925  * @sock: socket structure
926  * @m: message to send
927  * @dlen: total length of message data
928  * @timeout: timeout to wait for wakeup
929  *
930  * Called from function tipc_sendmsg(), which has done all sanity checks
931  * Returns the number of bytes sent on success, or errno
932  */
933 static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m,
934 				   int dlen, long timeout)
935 {
936 	struct sock *sk = sock->sk;
937 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
938 	int blks = tsk_blocks(GROUP_H_SIZE + dlen);
939 	struct tipc_sock *tsk = tipc_sk(sk);
940 	struct net *net = sock_net(sk);
941 	struct tipc_member *mb = NULL;
942 	u32 node, port;
943 	int rc;
944 
945 	node = dest->addr.id.node;
946 	port = dest->addr.id.ref;
947 	if (!port && !node)
948 		return -EHOSTUNREACH;
949 
950 	/* Block or return if destination link or member is congested */
951 	rc = tipc_wait_for_cond(sock, &timeout,
952 				!tipc_dest_find(&tsk->cong_links, node, 0) &&
953 				tsk->group &&
954 				!tipc_group_cong(tsk->group, node, port, blks,
955 						 &mb));
956 	if (unlikely(rc))
957 		return rc;
958 
959 	if (unlikely(!mb))
960 		return -EHOSTUNREACH;
961 
962 	rc = tipc_send_group_msg(net, tsk, m, mb, node, port, dlen);
963 
964 	return rc ? rc : dlen;
965 }
966 
967 /**
968  * tipc_send_group_anycast - send message to any member with given identity
969  * @sock: socket structure
970  * @m: message to send
971  * @dlen: total length of message data
972  * @timeout: timeout to wait for wakeup
973  *
974  * Called from function tipc_sendmsg(), which has done all sanity checks
975  * Returns the number of bytes sent on success, or errno
976  */
977 static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
978 				   int dlen, long timeout)
979 {
980 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
981 	struct sock *sk = sock->sk;
982 	struct tipc_sock *tsk = tipc_sk(sk);
983 	struct list_head *cong_links = &tsk->cong_links;
984 	int blks = tsk_blocks(GROUP_H_SIZE + dlen);
985 	struct tipc_msg *hdr = &tsk->phdr;
986 	struct tipc_member *first = NULL;
987 	struct tipc_member *mbr = NULL;
988 	struct net *net = sock_net(sk);
989 	u32 node, port, exclude;
990 	struct list_head dsts;
991 	u32 type, inst, scope;
992 	int lookups = 0;
993 	int dstcnt, rc;
994 	bool cong;
995 
996 	INIT_LIST_HEAD(&dsts);
997 
998 	type = msg_nametype(hdr);
999 	inst = dest->addr.name.name.instance;
1000 	scope = msg_lookup_scope(hdr);
1001 
1002 	while (++lookups < 4) {
1003 		exclude = tipc_group_exclude(tsk->group);
1004 
1005 		first = NULL;
1006 
1007 		/* Look for a non-congested destination member, if any */
1008 		while (1) {
1009 			if (!tipc_nametbl_lookup(net, type, inst, scope, &dsts,
1010 						 &dstcnt, exclude, false))
1011 				return -EHOSTUNREACH;
1012 			tipc_dest_pop(&dsts, &node, &port);
1013 			cong = tipc_group_cong(tsk->group, node, port, blks,
1014 					       &mbr);
1015 			if (!cong)
1016 				break;
1017 			if (mbr == first)
1018 				break;
1019 			if (!first)
1020 				first = mbr;
1021 		}
1022 
1023 		/* Start over if destination was not in member list */
1024 		if (unlikely(!mbr))
1025 			continue;
1026 
1027 		if (likely(!cong && !tipc_dest_find(cong_links, node, 0)))
1028 			break;
1029 
1030 		/* Block or return if destination link or member is congested */
1031 		rc = tipc_wait_for_cond(sock, &timeout,
1032 					!tipc_dest_find(cong_links, node, 0) &&
1033 					tsk->group &&
1034 					!tipc_group_cong(tsk->group, node, port,
1035 							 blks, &mbr));
1036 		if (unlikely(rc))
1037 			return rc;
1038 
1039 		/* Send, unless destination disappeared while waiting */
1040 		if (likely(mbr))
1041 			break;
1042 	}
1043 
1044 	if (unlikely(lookups >= 4))
1045 		return -EHOSTUNREACH;
1046 
1047 	rc = tipc_send_group_msg(net, tsk, m, mbr, node, port, dlen);
1048 
1049 	return rc ? rc : dlen;
1050 }
1051 
1052 /**
1053  * tipc_send_group_bcast - send message to all members in communication group
1054  * @sock: socket structure
1055  * @m: message to send
1056  * @dlen: total length of message data
1057  * @timeout: timeout to wait for wakeup
1058  *
1059  * Called from function tipc_sendmsg(), which has done all sanity checks
1060  * Returns the number of bytes sent on success, or errno
1061  */
1062 static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
1063 				 int dlen, long timeout)
1064 {
1065 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
1066 	struct sock *sk = sock->sk;
1067 	struct net *net = sock_net(sk);
1068 	struct tipc_sock *tsk = tipc_sk(sk);
1069 	struct tipc_nlist *dsts;
1070 	struct tipc_mc_method *method = &tsk->mc_method;
1071 	bool ack = method->mandatory && method->rcast;
1072 	int blks = tsk_blocks(MCAST_H_SIZE + dlen);
1073 	struct tipc_msg *hdr = &tsk->phdr;
1074 	int mtu = tipc_bcast_get_mtu(net);
1075 	struct sk_buff_head pkts;
1076 	int rc = -EHOSTUNREACH;
1077 
1078 	/* Block or return if any destination link or member is congested */
1079 	rc = tipc_wait_for_cond(sock, &timeout,
1080 				!tsk->cong_link_cnt && tsk->group &&
1081 				!tipc_group_bc_cong(tsk->group, blks));
1082 	if (unlikely(rc))
1083 		return rc;
1084 
1085 	dsts = tipc_group_dests(tsk->group);
1086 	if (!dsts->local && !dsts->remote)
1087 		return -EHOSTUNREACH;
1088 
1089 	/* Complete message header */
1090 	if (dest) {
1091 		msg_set_type(hdr, TIPC_GRP_MCAST_MSG);
1092 		msg_set_nameinst(hdr, dest->addr.name.name.instance);
1093 	} else {
1094 		msg_set_type(hdr, TIPC_GRP_BCAST_MSG);
1095 		msg_set_nameinst(hdr, 0);
1096 	}
1097 	msg_set_hdr_sz(hdr, GROUP_H_SIZE);
1098 	msg_set_destport(hdr, 0);
1099 	msg_set_destnode(hdr, 0);
1100 	msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(tsk->group));
1101 
1102 	/* Avoid getting stuck with repeated forced replicasts */
1103 	msg_set_grp_bc_ack_req(hdr, ack);
1104 
1105 	/* Build message as chain of buffers */
1106 	__skb_queue_head_init(&pkts);
1107 	rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
1108 	if (unlikely(rc != dlen))
1109 		return rc;
1110 
1111 	/* Send message */
1112 	rc = tipc_mcast_xmit(net, &pkts, method, dsts, &tsk->cong_link_cnt);
1113 	if (unlikely(rc))
1114 		return rc;
1115 
1116 	/* Update broadcast sequence number and send windows */
1117 	tipc_group_update_bc_members(tsk->group, blks, ack);
1118 
1119 	/* Broadcast link is now free to choose method for next broadcast */
1120 	method->mandatory = false;
1121 	method->expires = jiffies;
1122 
1123 	return dlen;
1124 }
1125 
1126 /**
1127  * tipc_send_group_mcast - send message to all members with given identity
1128  * @sock: socket structure
1129  * @m: message to send
1130  * @dlen: total length of message data
1131  * @timeout: timeout to wait for wakeup
1132  *
1133  * Called from function tipc_sendmsg(), which has done all sanity checks
1134  * Returns the number of bytes sent on success, or errno
1135  */
1136 static int tipc_send_group_mcast(struct socket *sock, struct msghdr *m,
1137 				 int dlen, long timeout)
1138 {
1139 	struct sock *sk = sock->sk;
1140 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
1141 	struct tipc_sock *tsk = tipc_sk(sk);
1142 	struct tipc_group *grp = tsk->group;
1143 	struct tipc_msg *hdr = &tsk->phdr;
1144 	struct net *net = sock_net(sk);
1145 	u32 type, inst, scope, exclude;
1146 	struct list_head dsts;
1147 	u32 dstcnt;
1148 
1149 	INIT_LIST_HEAD(&dsts);
1150 
1151 	type = msg_nametype(hdr);
1152 	inst = dest->addr.name.name.instance;
1153 	scope = msg_lookup_scope(hdr);
1154 	exclude = tipc_group_exclude(grp);
1155 
1156 	if (!tipc_nametbl_lookup(net, type, inst, scope, &dsts,
1157 				 &dstcnt, exclude, true))
1158 		return -EHOSTUNREACH;
1159 
1160 	if (dstcnt == 1) {
1161 		tipc_dest_pop(&dsts, &dest->addr.id.node, &dest->addr.id.ref);
1162 		return tipc_send_group_unicast(sock, m, dlen, timeout);
1163 	}
1164 
1165 	tipc_dest_list_purge(&dsts);
1166 	return tipc_send_group_bcast(sock, m, dlen, timeout);
1167 }
1168 
1169 /**
1170  * tipc_sk_mcast_rcv - Deliver multicast messages to all destination sockets
1171  * @arrvq: queue with arriving messages, to be cloned after destination lookup
1172  * @inputq: queue with cloned messages, delivered to socket after dest lookup
1173  *
1174  * Multi-threaded: parallel calls with reference to same queues may occur
1175  */
1176 void tipc_sk_mcast_rcv(struct net *net, struct sk_buff_head *arrvq,
1177 		       struct sk_buff_head *inputq)
1178 {
1179 	u32 self = tipc_own_addr(net);
1180 	u32 type, lower, upper, scope;
1181 	struct sk_buff *skb, *_skb;
1182 	u32 portid, onode;
1183 	struct sk_buff_head tmpq;
1184 	struct list_head dports;
1185 	struct tipc_msg *hdr;
1186 	int user, mtyp, hlen;
1187 	bool exact;
1188 
1189 	__skb_queue_head_init(&tmpq);
1190 	INIT_LIST_HEAD(&dports);
1191 
1192 	skb = tipc_skb_peek(arrvq, &inputq->lock);
1193 	for (; skb; skb = tipc_skb_peek(arrvq, &inputq->lock)) {
1194 		hdr = buf_msg(skb);
1195 		user = msg_user(hdr);
1196 		mtyp = msg_type(hdr);
1197 		hlen = skb_headroom(skb) + msg_hdr_sz(hdr);
1198 		onode = msg_orignode(hdr);
1199 		type = msg_nametype(hdr);
1200 
1201 		if (mtyp == TIPC_GRP_UCAST_MSG || user == GROUP_PROTOCOL) {
1202 			spin_lock_bh(&inputq->lock);
1203 			if (skb_peek(arrvq) == skb) {
1204 				__skb_dequeue(arrvq);
1205 				__skb_queue_tail(inputq, skb);
1206 			}
1207 			kfree_skb(skb);
1208 			spin_unlock_bh(&inputq->lock);
1209 			continue;
1210 		}
1211 
1212 		/* Group messages require exact scope match */
1213 		if (msg_in_group(hdr)) {
1214 			lower = 0;
1215 			upper = ~0;
1216 			scope = msg_lookup_scope(hdr);
1217 			exact = true;
1218 		} else {
1219 			/* TIPC_NODE_SCOPE means "any scope" in this context */
1220 			if (onode == self)
1221 				scope = TIPC_NODE_SCOPE;
1222 			else
1223 				scope = TIPC_CLUSTER_SCOPE;
1224 			exact = false;
1225 			lower = msg_namelower(hdr);
1226 			upper = msg_nameupper(hdr);
1227 		}
1228 
1229 		/* Create destination port list: */
1230 		tipc_nametbl_mc_lookup(net, type, lower, upper,
1231 				       scope, exact, &dports);
1232 
1233 		/* Clone message per destination */
1234 		while (tipc_dest_pop(&dports, NULL, &portid)) {
1235 			_skb = __pskb_copy(skb, hlen, GFP_ATOMIC);
1236 			if (_skb) {
1237 				msg_set_destport(buf_msg(_skb), portid);
1238 				__skb_queue_tail(&tmpq, _skb);
1239 				continue;
1240 			}
1241 			pr_warn("Failed to clone mcast rcv buffer\n");
1242 		}
1243 		/* Append to inputq if not already done by other thread */
1244 		spin_lock_bh(&inputq->lock);
1245 		if (skb_peek(arrvq) == skb) {
1246 			skb_queue_splice_tail_init(&tmpq, inputq);
1247 			kfree_skb(__skb_dequeue(arrvq));
1248 		}
1249 		spin_unlock_bh(&inputq->lock);
1250 		__skb_queue_purge(&tmpq);
1251 		kfree_skb(skb);
1252 	}
1253 	tipc_sk_rcv(net, inputq);
1254 }
1255 
1256 /* tipc_sk_push_backlog(): send accumulated buffers in socket write queue
1257  *                         when socket is in Nagle mode
1258  */
1259 static void tipc_sk_push_backlog(struct tipc_sock *tsk, bool nagle_ack)
1260 {
1261 	struct sk_buff_head *txq = &tsk->sk.sk_write_queue;
1262 	struct sk_buff *skb = skb_peek_tail(txq);
1263 	struct net *net = sock_net(&tsk->sk);
1264 	u32 dnode = tsk_peer_node(tsk);
1265 	int rc;
1266 
1267 	if (nagle_ack) {
1268 		tsk->pkt_cnt += skb_queue_len(txq);
1269 		if (!tsk->pkt_cnt || tsk->msg_acc / tsk->pkt_cnt < 2) {
1270 			tsk->oneway = 0;
1271 			if (tsk->nagle_start < NAGLE_START_MAX)
1272 				tsk->nagle_start *= 2;
1273 			tsk->expect_ack = false;
1274 			pr_debug("tsk %10u: bad nagle %u -> %u, next start %u!\n",
1275 				 tsk->portid, tsk->msg_acc, tsk->pkt_cnt,
1276 				 tsk->nagle_start);
1277 		} else {
1278 			tsk->nagle_start = NAGLE_START_INIT;
1279 			if (skb) {
1280 				msg_set_ack_required(buf_msg(skb));
1281 				tsk->expect_ack = true;
1282 			} else {
1283 				tsk->expect_ack = false;
1284 			}
1285 		}
1286 		tsk->msg_acc = 0;
1287 		tsk->pkt_cnt = 0;
1288 	}
1289 
1290 	if (!skb || tsk->cong_link_cnt)
1291 		return;
1292 
1293 	/* Do not send SYN again after congestion */
1294 	if (msg_is_syn(buf_msg(skb)))
1295 		return;
1296 
1297 	if (tsk->msg_acc)
1298 		tsk->pkt_cnt += skb_queue_len(txq);
1299 	tsk->snt_unacked += tsk->snd_backlog;
1300 	tsk->snd_backlog = 0;
1301 	rc = tipc_node_xmit(net, txq, dnode, tsk->portid);
1302 	if (rc == -ELINKCONG)
1303 		tsk->cong_link_cnt = 1;
1304 }
1305 
1306 /**
1307  * tipc_sk_conn_proto_rcv - receive a connection mng protocol message
1308  * @tsk: receiving socket
1309  * @skb: pointer to message buffer.
1310  */
1311 static void tipc_sk_conn_proto_rcv(struct tipc_sock *tsk, struct sk_buff *skb,
1312 				   struct sk_buff_head *inputq,
1313 				   struct sk_buff_head *xmitq)
1314 {
1315 	struct tipc_msg *hdr = buf_msg(skb);
1316 	u32 onode = tsk_own_node(tsk);
1317 	struct sock *sk = &tsk->sk;
1318 	int mtyp = msg_type(hdr);
1319 	bool was_cong;
1320 
1321 	/* Ignore if connection cannot be validated: */
1322 	if (!tsk_peer_msg(tsk, hdr)) {
1323 		trace_tipc_sk_drop_msg(sk, skb, TIPC_DUMP_NONE, "@proto_rcv!");
1324 		goto exit;
1325 	}
1326 
1327 	if (unlikely(msg_errcode(hdr))) {
1328 		tipc_set_sk_state(sk, TIPC_DISCONNECTING);
1329 		tipc_node_remove_conn(sock_net(sk), tsk_peer_node(tsk),
1330 				      tsk_peer_port(tsk));
1331 		sk->sk_state_change(sk);
1332 
1333 		/* State change is ignored if socket already awake,
1334 		 * - convert msg to abort msg and add to inqueue
1335 		 */
1336 		msg_set_user(hdr, TIPC_CRITICAL_IMPORTANCE);
1337 		msg_set_type(hdr, TIPC_CONN_MSG);
1338 		msg_set_size(hdr, BASIC_H_SIZE);
1339 		msg_set_hdr_sz(hdr, BASIC_H_SIZE);
1340 		__skb_queue_tail(inputq, skb);
1341 		return;
1342 	}
1343 
1344 	tsk->probe_unacked = false;
1345 
1346 	if (mtyp == CONN_PROBE) {
1347 		msg_set_type(hdr, CONN_PROBE_REPLY);
1348 		if (tipc_msg_reverse(onode, &skb, TIPC_OK))
1349 			__skb_queue_tail(xmitq, skb);
1350 		return;
1351 	} else if (mtyp == CONN_ACK) {
1352 		was_cong = tsk_conn_cong(tsk);
1353 		tipc_sk_push_backlog(tsk, msg_nagle_ack(hdr));
1354 		tsk->snt_unacked -= msg_conn_ack(hdr);
1355 		if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL)
1356 			tsk->snd_win = msg_adv_win(hdr);
1357 		if (was_cong && !tsk_conn_cong(tsk))
1358 			sk->sk_write_space(sk);
1359 	} else if (mtyp != CONN_PROBE_REPLY) {
1360 		pr_warn("Received unknown CONN_PROTO msg\n");
1361 	}
1362 exit:
1363 	kfree_skb(skb);
1364 }
1365 
1366 /**
1367  * tipc_sendmsg - send message in connectionless manner
1368  * @sock: socket structure
1369  * @m: message to send
1370  * @dsz: amount of user data to be sent
1371  *
1372  * Message must have an destination specified explicitly.
1373  * Used for SOCK_RDM and SOCK_DGRAM messages,
1374  * and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections.
1375  * (Note: 'SYN+' is prohibited on SOCK_STREAM.)
1376  *
1377  * Returns the number of bytes sent on success, or errno otherwise
1378  */
1379 static int tipc_sendmsg(struct socket *sock,
1380 			struct msghdr *m, size_t dsz)
1381 {
1382 	struct sock *sk = sock->sk;
1383 	int ret;
1384 
1385 	lock_sock(sk);
1386 	ret = __tipc_sendmsg(sock, m, dsz);
1387 	release_sock(sk);
1388 
1389 	return ret;
1390 }
1391 
1392 static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dlen)
1393 {
1394 	struct sock *sk = sock->sk;
1395 	struct net *net = sock_net(sk);
1396 	struct tipc_sock *tsk = tipc_sk(sk);
1397 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
1398 	long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
1399 	struct list_head *clinks = &tsk->cong_links;
1400 	bool syn = !tipc_sk_type_connectionless(sk);
1401 	struct tipc_group *grp = tsk->group;
1402 	struct tipc_msg *hdr = &tsk->phdr;
1403 	struct tipc_name_seq *seq;
1404 	struct sk_buff_head pkts;
1405 	u32 dport = 0, dnode = 0;
1406 	u32 type = 0, inst = 0;
1407 	int mtu, rc;
1408 
1409 	if (unlikely(dlen > TIPC_MAX_USER_MSG_SIZE))
1410 		return -EMSGSIZE;
1411 
1412 	if (likely(dest)) {
1413 		if (unlikely(m->msg_namelen < sizeof(*dest)))
1414 			return -EINVAL;
1415 		if (unlikely(dest->family != AF_TIPC))
1416 			return -EINVAL;
1417 	}
1418 
1419 	if (grp) {
1420 		if (!dest)
1421 			return tipc_send_group_bcast(sock, m, dlen, timeout);
1422 		if (dest->addrtype == TIPC_ADDR_NAME)
1423 			return tipc_send_group_anycast(sock, m, dlen, timeout);
1424 		if (dest->addrtype == TIPC_ADDR_ID)
1425 			return tipc_send_group_unicast(sock, m, dlen, timeout);
1426 		if (dest->addrtype == TIPC_ADDR_MCAST)
1427 			return tipc_send_group_mcast(sock, m, dlen, timeout);
1428 		return -EINVAL;
1429 	}
1430 
1431 	if (unlikely(!dest)) {
1432 		dest = &tsk->peer;
1433 		if (!syn && dest->family != AF_TIPC)
1434 			return -EDESTADDRREQ;
1435 	}
1436 
1437 	if (unlikely(syn)) {
1438 		if (sk->sk_state == TIPC_LISTEN)
1439 			return -EPIPE;
1440 		if (sk->sk_state != TIPC_OPEN)
1441 			return -EISCONN;
1442 		if (tsk->published)
1443 			return -EOPNOTSUPP;
1444 		if (dest->addrtype == TIPC_ADDR_NAME) {
1445 			tsk->conn_type = dest->addr.name.name.type;
1446 			tsk->conn_instance = dest->addr.name.name.instance;
1447 		}
1448 		msg_set_syn(hdr, 1);
1449 	}
1450 
1451 	seq = &dest->addr.nameseq;
1452 	if (dest->addrtype == TIPC_ADDR_MCAST)
1453 		return tipc_sendmcast(sock, seq, m, dlen, timeout);
1454 
1455 	if (dest->addrtype == TIPC_ADDR_NAME) {
1456 		type = dest->addr.name.name.type;
1457 		inst = dest->addr.name.name.instance;
1458 		dnode = dest->addr.name.domain;
1459 		dport = tipc_nametbl_translate(net, type, inst, &dnode);
1460 		if (unlikely(!dport && !dnode))
1461 			return -EHOSTUNREACH;
1462 	} else if (dest->addrtype == TIPC_ADDR_ID) {
1463 		dnode = dest->addr.id.node;
1464 	} else {
1465 		return -EINVAL;
1466 	}
1467 
1468 	/* Block or return if destination link is congested */
1469 	rc = tipc_wait_for_cond(sock, &timeout,
1470 				!tipc_dest_find(clinks, dnode, 0));
1471 	if (unlikely(rc))
1472 		return rc;
1473 
1474 	if (dest->addrtype == TIPC_ADDR_NAME) {
1475 		msg_set_type(hdr, TIPC_NAMED_MSG);
1476 		msg_set_hdr_sz(hdr, NAMED_H_SIZE);
1477 		msg_set_nametype(hdr, type);
1478 		msg_set_nameinst(hdr, inst);
1479 		msg_set_lookup_scope(hdr, tipc_node2scope(dnode));
1480 		msg_set_destnode(hdr, dnode);
1481 		msg_set_destport(hdr, dport);
1482 	} else { /* TIPC_ADDR_ID */
1483 		msg_set_type(hdr, TIPC_DIRECT_MSG);
1484 		msg_set_lookup_scope(hdr, 0);
1485 		msg_set_destnode(hdr, dnode);
1486 		msg_set_destport(hdr, dest->addr.id.ref);
1487 		msg_set_hdr_sz(hdr, BASIC_H_SIZE);
1488 	}
1489 
1490 	__skb_queue_head_init(&pkts);
1491 	mtu = tipc_node_get_mtu(net, dnode, tsk->portid, true);
1492 	rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
1493 	if (unlikely(rc != dlen))
1494 		return rc;
1495 	if (unlikely(syn && !tipc_msg_skb_clone(&pkts, &sk->sk_write_queue))) {
1496 		__skb_queue_purge(&pkts);
1497 		return -ENOMEM;
1498 	}
1499 
1500 	trace_tipc_sk_sendmsg(sk, skb_peek(&pkts), TIPC_DUMP_SK_SNDQ, " ");
1501 	rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid);
1502 	if (unlikely(rc == -ELINKCONG)) {
1503 		tipc_dest_push(clinks, dnode, 0);
1504 		tsk->cong_link_cnt++;
1505 		rc = 0;
1506 	}
1507 
1508 	if (unlikely(syn && !rc))
1509 		tipc_set_sk_state(sk, TIPC_CONNECTING);
1510 
1511 	return rc ? rc : dlen;
1512 }
1513 
1514 /**
1515  * tipc_sendstream - send stream-oriented data
1516  * @sock: socket structure
1517  * @m: data to send
1518  * @dsz: total length of data to be transmitted
1519  *
1520  * Used for SOCK_STREAM data.
1521  *
1522  * Returns the number of bytes sent on success (or partial success),
1523  * or errno if no data sent
1524  */
1525 static int tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dsz)
1526 {
1527 	struct sock *sk = sock->sk;
1528 	int ret;
1529 
1530 	lock_sock(sk);
1531 	ret = __tipc_sendstream(sock, m, dsz);
1532 	release_sock(sk);
1533 
1534 	return ret;
1535 }
1536 
1537 static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dlen)
1538 {
1539 	struct sock *sk = sock->sk;
1540 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
1541 	long timeout = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
1542 	struct sk_buff_head *txq = &sk->sk_write_queue;
1543 	struct tipc_sock *tsk = tipc_sk(sk);
1544 	struct tipc_msg *hdr = &tsk->phdr;
1545 	struct net *net = sock_net(sk);
1546 	struct sk_buff *skb;
1547 	u32 dnode = tsk_peer_node(tsk);
1548 	int maxnagle = tsk->maxnagle;
1549 	int maxpkt = tsk->max_pkt;
1550 	int send, sent = 0;
1551 	int blocks, rc = 0;
1552 
1553 	if (unlikely(dlen > INT_MAX))
1554 		return -EMSGSIZE;
1555 
1556 	/* Handle implicit connection setup */
1557 	if (unlikely(dest)) {
1558 		rc = __tipc_sendmsg(sock, m, dlen);
1559 		if (dlen && dlen == rc) {
1560 			tsk->peer_caps = tipc_node_get_capabilities(net, dnode);
1561 			tsk->snt_unacked = tsk_inc(tsk, dlen + msg_hdr_sz(hdr));
1562 		}
1563 		return rc;
1564 	}
1565 
1566 	do {
1567 		rc = tipc_wait_for_cond(sock, &timeout,
1568 					(!tsk->cong_link_cnt &&
1569 					 !tsk_conn_cong(tsk) &&
1570 					 tipc_sk_connected(sk)));
1571 		if (unlikely(rc))
1572 			break;
1573 		send = min_t(size_t, dlen - sent, TIPC_MAX_USER_MSG_SIZE);
1574 		blocks = tsk->snd_backlog;
1575 		if (tsk->oneway++ >= tsk->nagle_start && maxnagle &&
1576 		    send <= maxnagle) {
1577 			rc = tipc_msg_append(hdr, m, send, maxnagle, txq);
1578 			if (unlikely(rc < 0))
1579 				break;
1580 			blocks += rc;
1581 			tsk->msg_acc++;
1582 			if (blocks <= 64 && tsk->expect_ack) {
1583 				tsk->snd_backlog = blocks;
1584 				sent += send;
1585 				break;
1586 			} else if (blocks > 64) {
1587 				tsk->pkt_cnt += skb_queue_len(txq);
1588 			} else {
1589 				skb = skb_peek_tail(txq);
1590 				if (skb) {
1591 					msg_set_ack_required(buf_msg(skb));
1592 					tsk->expect_ack = true;
1593 				} else {
1594 					tsk->expect_ack = false;
1595 				}
1596 				tsk->msg_acc = 0;
1597 				tsk->pkt_cnt = 0;
1598 			}
1599 		} else {
1600 			rc = tipc_msg_build(hdr, m, sent, send, maxpkt, txq);
1601 			if (unlikely(rc != send))
1602 				break;
1603 			blocks += tsk_inc(tsk, send + MIN_H_SIZE);
1604 		}
1605 		trace_tipc_sk_sendstream(sk, skb_peek(txq),
1606 					 TIPC_DUMP_SK_SNDQ, " ");
1607 		rc = tipc_node_xmit(net, txq, dnode, tsk->portid);
1608 		if (unlikely(rc == -ELINKCONG)) {
1609 			tsk->cong_link_cnt = 1;
1610 			rc = 0;
1611 		}
1612 		if (likely(!rc)) {
1613 			tsk->snt_unacked += blocks;
1614 			tsk->snd_backlog = 0;
1615 			sent += send;
1616 		}
1617 	} while (sent < dlen && !rc);
1618 
1619 	return sent ? sent : rc;
1620 }
1621 
1622 /**
1623  * tipc_send_packet - send a connection-oriented message
1624  * @sock: socket structure
1625  * @m: message to send
1626  * @dsz: length of data to be transmitted
1627  *
1628  * Used for SOCK_SEQPACKET messages.
1629  *
1630  * Returns the number of bytes sent on success, or errno otherwise
1631  */
1632 static int tipc_send_packet(struct socket *sock, struct msghdr *m, size_t dsz)
1633 {
1634 	if (dsz > TIPC_MAX_USER_MSG_SIZE)
1635 		return -EMSGSIZE;
1636 
1637 	return tipc_sendstream(sock, m, dsz);
1638 }
1639 
1640 /* tipc_sk_finish_conn - complete the setup of a connection
1641  */
1642 static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port,
1643 				u32 peer_node)
1644 {
1645 	struct sock *sk = &tsk->sk;
1646 	struct net *net = sock_net(sk);
1647 	struct tipc_msg *msg = &tsk->phdr;
1648 
1649 	msg_set_syn(msg, 0);
1650 	msg_set_destnode(msg, peer_node);
1651 	msg_set_destport(msg, peer_port);
1652 	msg_set_type(msg, TIPC_CONN_MSG);
1653 	msg_set_lookup_scope(msg, 0);
1654 	msg_set_hdr_sz(msg, SHORT_H_SIZE);
1655 
1656 	sk_reset_timer(sk, &sk->sk_timer, jiffies + CONN_PROBING_INTV);
1657 	tipc_set_sk_state(sk, TIPC_ESTABLISHED);
1658 	tipc_node_add_conn(net, peer_node, tsk->portid, peer_port);
1659 	tsk->max_pkt = tipc_node_get_mtu(net, peer_node, tsk->portid, true);
1660 	tsk->peer_caps = tipc_node_get_capabilities(net, peer_node);
1661 	tsk_set_nagle(tsk);
1662 	__skb_queue_purge(&sk->sk_write_queue);
1663 	if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL)
1664 		return;
1665 
1666 	/* Fall back to message based flow control */
1667 	tsk->rcv_win = FLOWCTL_MSG_WIN;
1668 	tsk->snd_win = FLOWCTL_MSG_WIN;
1669 }
1670 
1671 /**
1672  * tipc_sk_set_orig_addr - capture sender's address for received message
1673  * @m: descriptor for message info
1674  * @skb: received message
1675  *
1676  * Note: Address is not captured if not requested by receiver.
1677  */
1678 static void tipc_sk_set_orig_addr(struct msghdr *m, struct sk_buff *skb)
1679 {
1680 	DECLARE_SOCKADDR(struct sockaddr_pair *, srcaddr, m->msg_name);
1681 	struct tipc_msg *hdr = buf_msg(skb);
1682 
1683 	if (!srcaddr)
1684 		return;
1685 
1686 	srcaddr->sock.family = AF_TIPC;
1687 	srcaddr->sock.addrtype = TIPC_ADDR_ID;
1688 	srcaddr->sock.scope = 0;
1689 	srcaddr->sock.addr.id.ref = msg_origport(hdr);
1690 	srcaddr->sock.addr.id.node = msg_orignode(hdr);
1691 	srcaddr->sock.addr.name.domain = 0;
1692 	m->msg_namelen = sizeof(struct sockaddr_tipc);
1693 
1694 	if (!msg_in_group(hdr))
1695 		return;
1696 
1697 	/* Group message users may also want to know sending member's id */
1698 	srcaddr->member.family = AF_TIPC;
1699 	srcaddr->member.addrtype = TIPC_ADDR_NAME;
1700 	srcaddr->member.scope = 0;
1701 	srcaddr->member.addr.name.name.type = msg_nametype(hdr);
1702 	srcaddr->member.addr.name.name.instance = TIPC_SKB_CB(skb)->orig_member;
1703 	srcaddr->member.addr.name.domain = 0;
1704 	m->msg_namelen = sizeof(*srcaddr);
1705 }
1706 
1707 /**
1708  * tipc_sk_anc_data_recv - optionally capture ancillary data for received message
1709  * @m: descriptor for message info
1710  * @skb: received message buffer
1711  * @tsk: TIPC port associated with message
1712  *
1713  * Note: Ancillary data is not captured if not requested by receiver.
1714  *
1715  * Returns 0 if successful, otherwise errno
1716  */
1717 static int tipc_sk_anc_data_recv(struct msghdr *m, struct sk_buff *skb,
1718 				 struct tipc_sock *tsk)
1719 {
1720 	struct tipc_msg *msg;
1721 	u32 anc_data[3];
1722 	u32 err;
1723 	u32 dest_type;
1724 	int has_name;
1725 	int res;
1726 
1727 	if (likely(m->msg_controllen == 0))
1728 		return 0;
1729 	msg = buf_msg(skb);
1730 
1731 	/* Optionally capture errored message object(s) */
1732 	err = msg ? msg_errcode(msg) : 0;
1733 	if (unlikely(err)) {
1734 		anc_data[0] = err;
1735 		anc_data[1] = msg_data_sz(msg);
1736 		res = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, anc_data);
1737 		if (res)
1738 			return res;
1739 		if (anc_data[1]) {
1740 			if (skb_linearize(skb))
1741 				return -ENOMEM;
1742 			msg = buf_msg(skb);
1743 			res = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, anc_data[1],
1744 				       msg_data(msg));
1745 			if (res)
1746 				return res;
1747 		}
1748 	}
1749 
1750 	/* Optionally capture message destination object */
1751 	dest_type = msg ? msg_type(msg) : TIPC_DIRECT_MSG;
1752 	switch (dest_type) {
1753 	case TIPC_NAMED_MSG:
1754 		has_name = 1;
1755 		anc_data[0] = msg_nametype(msg);
1756 		anc_data[1] = msg_namelower(msg);
1757 		anc_data[2] = msg_namelower(msg);
1758 		break;
1759 	case TIPC_MCAST_MSG:
1760 		has_name = 1;
1761 		anc_data[0] = msg_nametype(msg);
1762 		anc_data[1] = msg_namelower(msg);
1763 		anc_data[2] = msg_nameupper(msg);
1764 		break;
1765 	case TIPC_CONN_MSG:
1766 		has_name = (tsk->conn_type != 0);
1767 		anc_data[0] = tsk->conn_type;
1768 		anc_data[1] = tsk->conn_instance;
1769 		anc_data[2] = tsk->conn_instance;
1770 		break;
1771 	default:
1772 		has_name = 0;
1773 	}
1774 	if (has_name) {
1775 		res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data);
1776 		if (res)
1777 			return res;
1778 	}
1779 
1780 	return 0;
1781 }
1782 
1783 static struct sk_buff *tipc_sk_build_ack(struct tipc_sock *tsk)
1784 {
1785 	struct sock *sk = &tsk->sk;
1786 	struct sk_buff *skb = NULL;
1787 	struct tipc_msg *msg;
1788 	u32 peer_port = tsk_peer_port(tsk);
1789 	u32 dnode = tsk_peer_node(tsk);
1790 
1791 	if (!tipc_sk_connected(sk))
1792 		return NULL;
1793 	skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0,
1794 			      dnode, tsk_own_node(tsk), peer_port,
1795 			      tsk->portid, TIPC_OK);
1796 	if (!skb)
1797 		return NULL;
1798 	msg = buf_msg(skb);
1799 	msg_set_conn_ack(msg, tsk->rcv_unacked);
1800 	tsk->rcv_unacked = 0;
1801 
1802 	/* Adjust to and advertize the correct window limit */
1803 	if (tsk->peer_caps & TIPC_BLOCK_FLOWCTL) {
1804 		tsk->rcv_win = tsk_adv_blocks(tsk->sk.sk_rcvbuf);
1805 		msg_set_adv_win(msg, tsk->rcv_win);
1806 	}
1807 	return skb;
1808 }
1809 
1810 static void tipc_sk_send_ack(struct tipc_sock *tsk)
1811 {
1812 	struct sk_buff *skb;
1813 
1814 	skb = tipc_sk_build_ack(tsk);
1815 	if (!skb)
1816 		return;
1817 
1818 	tipc_node_xmit_skb(sock_net(&tsk->sk), skb, tsk_peer_node(tsk),
1819 			   msg_link_selector(buf_msg(skb)));
1820 }
1821 
1822 static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop)
1823 {
1824 	struct sock *sk = sock->sk;
1825 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
1826 	long timeo = *timeop;
1827 	int err = sock_error(sk);
1828 
1829 	if (err)
1830 		return err;
1831 
1832 	for (;;) {
1833 		if (timeo && skb_queue_empty(&sk->sk_receive_queue)) {
1834 			if (sk->sk_shutdown & RCV_SHUTDOWN) {
1835 				err = -ENOTCONN;
1836 				break;
1837 			}
1838 			add_wait_queue(sk_sleep(sk), &wait);
1839 			release_sock(sk);
1840 			timeo = wait_woken(&wait, TASK_INTERRUPTIBLE, timeo);
1841 			sched_annotate_sleep();
1842 			lock_sock(sk);
1843 			remove_wait_queue(sk_sleep(sk), &wait);
1844 		}
1845 		err = 0;
1846 		if (!skb_queue_empty(&sk->sk_receive_queue))
1847 			break;
1848 		err = -EAGAIN;
1849 		if (!timeo)
1850 			break;
1851 		err = sock_intr_errno(timeo);
1852 		if (signal_pending(current))
1853 			break;
1854 
1855 		err = sock_error(sk);
1856 		if (err)
1857 			break;
1858 	}
1859 	*timeop = timeo;
1860 	return err;
1861 }
1862 
1863 /**
1864  * tipc_recvmsg - receive packet-oriented message
1865  * @m: descriptor for message info
1866  * @buflen: length of user buffer area
1867  * @flags: receive flags
1868  *
1869  * Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages.
1870  * If the complete message doesn't fit in user area, truncate it.
1871  *
1872  * Returns size of returned message data, errno otherwise
1873  */
1874 static int tipc_recvmsg(struct socket *sock, struct msghdr *m,
1875 			size_t buflen,	int flags)
1876 {
1877 	struct sock *sk = sock->sk;
1878 	bool connected = !tipc_sk_type_connectionless(sk);
1879 	struct tipc_sock *tsk = tipc_sk(sk);
1880 	int rc, err, hlen, dlen, copy;
1881 	struct sk_buff_head xmitq;
1882 	struct tipc_msg *hdr;
1883 	struct sk_buff *skb;
1884 	bool grp_evt;
1885 	long timeout;
1886 
1887 	/* Catch invalid receive requests */
1888 	if (unlikely(!buflen))
1889 		return -EINVAL;
1890 
1891 	lock_sock(sk);
1892 	if (unlikely(connected && sk->sk_state == TIPC_OPEN)) {
1893 		rc = -ENOTCONN;
1894 		goto exit;
1895 	}
1896 	timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
1897 
1898 	/* Step rcv queue to first msg with data or error; wait if necessary */
1899 	do {
1900 		rc = tipc_wait_for_rcvmsg(sock, &timeout);
1901 		if (unlikely(rc))
1902 			goto exit;
1903 		skb = skb_peek(&sk->sk_receive_queue);
1904 		hdr = buf_msg(skb);
1905 		dlen = msg_data_sz(hdr);
1906 		hlen = msg_hdr_sz(hdr);
1907 		err = msg_errcode(hdr);
1908 		grp_evt = msg_is_grp_evt(hdr);
1909 		if (likely(dlen || err))
1910 			break;
1911 		tsk_advance_rx_queue(sk);
1912 	} while (1);
1913 
1914 	/* Collect msg meta data, including error code and rejected data */
1915 	tipc_sk_set_orig_addr(m, skb);
1916 	rc = tipc_sk_anc_data_recv(m, skb, tsk);
1917 	if (unlikely(rc))
1918 		goto exit;
1919 	hdr = buf_msg(skb);
1920 
1921 	/* Capture data if non-error msg, otherwise just set return value */
1922 	if (likely(!err)) {
1923 		copy = min_t(int, dlen, buflen);
1924 		if (unlikely(copy != dlen))
1925 			m->msg_flags |= MSG_TRUNC;
1926 		rc = skb_copy_datagram_msg(skb, hlen, m, copy);
1927 	} else {
1928 		copy = 0;
1929 		rc = 0;
1930 		if (err != TIPC_CONN_SHUTDOWN && connected && !m->msg_control)
1931 			rc = -ECONNRESET;
1932 	}
1933 	if (unlikely(rc))
1934 		goto exit;
1935 
1936 	/* Mark message as group event if applicable */
1937 	if (unlikely(grp_evt)) {
1938 		if (msg_grp_evt(hdr) == TIPC_WITHDRAWN)
1939 			m->msg_flags |= MSG_EOR;
1940 		m->msg_flags |= MSG_OOB;
1941 		copy = 0;
1942 	}
1943 
1944 	/* Caption of data or error code/rejected data was successful */
1945 	if (unlikely(flags & MSG_PEEK))
1946 		goto exit;
1947 
1948 	/* Send group flow control advertisement when applicable */
1949 	if (tsk->group && msg_in_group(hdr) && !grp_evt) {
1950 		__skb_queue_head_init(&xmitq);
1951 		tipc_group_update_rcv_win(tsk->group, tsk_blocks(hlen + dlen),
1952 					  msg_orignode(hdr), msg_origport(hdr),
1953 					  &xmitq);
1954 		tipc_node_distr_xmit(sock_net(sk), &xmitq);
1955 	}
1956 
1957 	tsk_advance_rx_queue(sk);
1958 
1959 	if (likely(!connected))
1960 		goto exit;
1961 
1962 	/* Send connection flow control advertisement when applicable */
1963 	tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen);
1964 	if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE)
1965 		tipc_sk_send_ack(tsk);
1966 exit:
1967 	release_sock(sk);
1968 	return rc ? rc : copy;
1969 }
1970 
1971 /**
1972  * tipc_recvstream - receive stream-oriented data
1973  * @m: descriptor for message info
1974  * @buflen: total size of user buffer area
1975  * @flags: receive flags
1976  *
1977  * Used for SOCK_STREAM messages only.  If not enough data is available
1978  * will optionally wait for more; never truncates data.
1979  *
1980  * Returns size of returned message data, errno otherwise
1981  */
1982 static int tipc_recvstream(struct socket *sock, struct msghdr *m,
1983 			   size_t buflen, int flags)
1984 {
1985 	struct sock *sk = sock->sk;
1986 	struct tipc_sock *tsk = tipc_sk(sk);
1987 	struct sk_buff *skb;
1988 	struct tipc_msg *hdr;
1989 	struct tipc_skb_cb *skb_cb;
1990 	bool peek = flags & MSG_PEEK;
1991 	int offset, required, copy, copied = 0;
1992 	int hlen, dlen, err, rc;
1993 	long timeout;
1994 
1995 	/* Catch invalid receive attempts */
1996 	if (unlikely(!buflen))
1997 		return -EINVAL;
1998 
1999 	lock_sock(sk);
2000 
2001 	if (unlikely(sk->sk_state == TIPC_OPEN)) {
2002 		rc = -ENOTCONN;
2003 		goto exit;
2004 	}
2005 	required = sock_rcvlowat(sk, flags & MSG_WAITALL, buflen);
2006 	timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
2007 
2008 	do {
2009 		/* Look at first msg in receive queue; wait if necessary */
2010 		rc = tipc_wait_for_rcvmsg(sock, &timeout);
2011 		if (unlikely(rc))
2012 			break;
2013 		skb = skb_peek(&sk->sk_receive_queue);
2014 		skb_cb = TIPC_SKB_CB(skb);
2015 		hdr = buf_msg(skb);
2016 		dlen = msg_data_sz(hdr);
2017 		hlen = msg_hdr_sz(hdr);
2018 		err = msg_errcode(hdr);
2019 
2020 		/* Discard any empty non-errored (SYN-) message */
2021 		if (unlikely(!dlen && !err)) {
2022 			tsk_advance_rx_queue(sk);
2023 			continue;
2024 		}
2025 
2026 		/* Collect msg meta data, incl. error code and rejected data */
2027 		if (!copied) {
2028 			tipc_sk_set_orig_addr(m, skb);
2029 			rc = tipc_sk_anc_data_recv(m, skb, tsk);
2030 			if (rc)
2031 				break;
2032 			hdr = buf_msg(skb);
2033 		}
2034 
2035 		/* Copy data if msg ok, otherwise return error/partial data */
2036 		if (likely(!err)) {
2037 			offset = skb_cb->bytes_read;
2038 			copy = min_t(int, dlen - offset, buflen - copied);
2039 			rc = skb_copy_datagram_msg(skb, hlen + offset, m, copy);
2040 			if (unlikely(rc))
2041 				break;
2042 			copied += copy;
2043 			offset += copy;
2044 			if (unlikely(offset < dlen)) {
2045 				if (!peek)
2046 					skb_cb->bytes_read = offset;
2047 				break;
2048 			}
2049 		} else {
2050 			rc = 0;
2051 			if ((err != TIPC_CONN_SHUTDOWN) && !m->msg_control)
2052 				rc = -ECONNRESET;
2053 			if (copied || rc)
2054 				break;
2055 		}
2056 
2057 		if (unlikely(peek))
2058 			break;
2059 
2060 		tsk_advance_rx_queue(sk);
2061 
2062 		/* Send connection flow control advertisement when applicable */
2063 		tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen);
2064 		if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE)
2065 			tipc_sk_send_ack(tsk);
2066 
2067 		/* Exit if all requested data or FIN/error received */
2068 		if (copied == buflen || err)
2069 			break;
2070 
2071 	} while (!skb_queue_empty(&sk->sk_receive_queue) || copied < required);
2072 exit:
2073 	release_sock(sk);
2074 	return copied ? copied : rc;
2075 }
2076 
2077 /**
2078  * tipc_write_space - wake up thread if port congestion is released
2079  * @sk: socket
2080  */
2081 static void tipc_write_space(struct sock *sk)
2082 {
2083 	struct socket_wq *wq;
2084 
2085 	rcu_read_lock();
2086 	wq = rcu_dereference(sk->sk_wq);
2087 	if (skwq_has_sleeper(wq))
2088 		wake_up_interruptible_sync_poll(&wq->wait, EPOLLOUT |
2089 						EPOLLWRNORM | EPOLLWRBAND);
2090 	rcu_read_unlock();
2091 }
2092 
2093 /**
2094  * tipc_data_ready - wake up threads to indicate messages have been received
2095  * @sk: socket
2096  */
2097 static void tipc_data_ready(struct sock *sk)
2098 {
2099 	struct socket_wq *wq;
2100 
2101 	rcu_read_lock();
2102 	wq = rcu_dereference(sk->sk_wq);
2103 	if (skwq_has_sleeper(wq))
2104 		wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN |
2105 						EPOLLRDNORM | EPOLLRDBAND);
2106 	rcu_read_unlock();
2107 }
2108 
2109 static void tipc_sock_destruct(struct sock *sk)
2110 {
2111 	__skb_queue_purge(&sk->sk_receive_queue);
2112 }
2113 
2114 static void tipc_sk_proto_rcv(struct sock *sk,
2115 			      struct sk_buff_head *inputq,
2116 			      struct sk_buff_head *xmitq)
2117 {
2118 	struct sk_buff *skb = __skb_dequeue(inputq);
2119 	struct tipc_sock *tsk = tipc_sk(sk);
2120 	struct tipc_msg *hdr = buf_msg(skb);
2121 	struct tipc_group *grp = tsk->group;
2122 	bool wakeup = false;
2123 
2124 	switch (msg_user(hdr)) {
2125 	case CONN_MANAGER:
2126 		tipc_sk_conn_proto_rcv(tsk, skb, inputq, xmitq);
2127 		return;
2128 	case SOCK_WAKEUP:
2129 		tipc_dest_del(&tsk->cong_links, msg_orignode(hdr), 0);
2130 		/* coupled with smp_rmb() in tipc_wait_for_cond() */
2131 		smp_wmb();
2132 		tsk->cong_link_cnt--;
2133 		wakeup = true;
2134 		tipc_sk_push_backlog(tsk, false);
2135 		break;
2136 	case GROUP_PROTOCOL:
2137 		tipc_group_proto_rcv(grp, &wakeup, hdr, inputq, xmitq);
2138 		break;
2139 	case TOP_SRV:
2140 		tipc_group_member_evt(tsk->group, &wakeup, &sk->sk_rcvbuf,
2141 				      hdr, inputq, xmitq);
2142 		break;
2143 	default:
2144 		break;
2145 	}
2146 
2147 	if (wakeup)
2148 		sk->sk_write_space(sk);
2149 
2150 	kfree_skb(skb);
2151 }
2152 
2153 /**
2154  * tipc_sk_filter_connect - check incoming message for a connection-based socket
2155  * @tsk: TIPC socket
2156  * @skb: pointer to message buffer.
2157  * @xmitq: for Nagle ACK if any
2158  * Returns true if message should be added to receive queue, false otherwise
2159  */
2160 static bool tipc_sk_filter_connect(struct tipc_sock *tsk, struct sk_buff *skb,
2161 				   struct sk_buff_head *xmitq)
2162 {
2163 	struct sock *sk = &tsk->sk;
2164 	struct net *net = sock_net(sk);
2165 	struct tipc_msg *hdr = buf_msg(skb);
2166 	bool con_msg = msg_connected(hdr);
2167 	u32 pport = tsk_peer_port(tsk);
2168 	u32 pnode = tsk_peer_node(tsk);
2169 	u32 oport = msg_origport(hdr);
2170 	u32 onode = msg_orignode(hdr);
2171 	int err = msg_errcode(hdr);
2172 	unsigned long delay;
2173 
2174 	if (unlikely(msg_mcast(hdr)))
2175 		return false;
2176 	tsk->oneway = 0;
2177 
2178 	switch (sk->sk_state) {
2179 	case TIPC_CONNECTING:
2180 		/* Setup ACK */
2181 		if (likely(con_msg)) {
2182 			if (err)
2183 				break;
2184 			tipc_sk_finish_conn(tsk, oport, onode);
2185 			msg_set_importance(&tsk->phdr, msg_importance(hdr));
2186 			/* ACK+ message with data is added to receive queue */
2187 			if (msg_data_sz(hdr))
2188 				return true;
2189 			/* Empty ACK-, - wake up sleeping connect() and drop */
2190 			sk->sk_state_change(sk);
2191 			msg_set_dest_droppable(hdr, 1);
2192 			return false;
2193 		}
2194 		/* Ignore connectionless message if not from listening socket */
2195 		if (oport != pport || onode != pnode)
2196 			return false;
2197 
2198 		/* Rejected SYN */
2199 		if (err != TIPC_ERR_OVERLOAD)
2200 			break;
2201 
2202 		/* Prepare for new setup attempt if we have a SYN clone */
2203 		if (skb_queue_empty(&sk->sk_write_queue))
2204 			break;
2205 		get_random_bytes(&delay, 2);
2206 		delay %= (tsk->conn_timeout / 4);
2207 		delay = msecs_to_jiffies(delay + 100);
2208 		sk_reset_timer(sk, &sk->sk_timer, jiffies + delay);
2209 		return false;
2210 	case TIPC_OPEN:
2211 	case TIPC_DISCONNECTING:
2212 		return false;
2213 	case TIPC_LISTEN:
2214 		/* Accept only SYN message */
2215 		if (!msg_is_syn(hdr) &&
2216 		    tipc_node_get_capabilities(net, onode) & TIPC_SYN_BIT)
2217 			return false;
2218 		if (!con_msg && !err)
2219 			return true;
2220 		return false;
2221 	case TIPC_ESTABLISHED:
2222 		if (!skb_queue_empty(&sk->sk_write_queue))
2223 			tipc_sk_push_backlog(tsk, false);
2224 		/* Accept only connection-based messages sent by peer */
2225 		if (likely(con_msg && !err && pport == oport &&
2226 			   pnode == onode)) {
2227 			if (msg_ack_required(hdr)) {
2228 				struct sk_buff *skb;
2229 
2230 				skb = tipc_sk_build_ack(tsk);
2231 				if (skb) {
2232 					msg_set_nagle_ack(buf_msg(skb));
2233 					__skb_queue_tail(xmitq, skb);
2234 				}
2235 			}
2236 			return true;
2237 		}
2238 		if (!tsk_peer_msg(tsk, hdr))
2239 			return false;
2240 		if (!err)
2241 			return true;
2242 		tipc_set_sk_state(sk, TIPC_DISCONNECTING);
2243 		tipc_node_remove_conn(net, pnode, tsk->portid);
2244 		sk->sk_state_change(sk);
2245 		return true;
2246 	default:
2247 		pr_err("Unknown sk_state %u\n", sk->sk_state);
2248 	}
2249 	/* Abort connection setup attempt */
2250 	tipc_set_sk_state(sk, TIPC_DISCONNECTING);
2251 	sk->sk_err = ECONNREFUSED;
2252 	sk->sk_state_change(sk);
2253 	return true;
2254 }
2255 
2256 /**
2257  * rcvbuf_limit - get proper overload limit of socket receive queue
2258  * @sk: socket
2259  * @skb: message
2260  *
2261  * For connection oriented messages, irrespective of importance,
2262  * default queue limit is 2 MB.
2263  *
2264  * For connectionless messages, queue limits are based on message
2265  * importance as follows:
2266  *
2267  * TIPC_LOW_IMPORTANCE       (2 MB)
2268  * TIPC_MEDIUM_IMPORTANCE    (4 MB)
2269  * TIPC_HIGH_IMPORTANCE      (8 MB)
2270  * TIPC_CRITICAL_IMPORTANCE  (16 MB)
2271  *
2272  * Returns overload limit according to corresponding message importance
2273  */
2274 static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *skb)
2275 {
2276 	struct tipc_sock *tsk = tipc_sk(sk);
2277 	struct tipc_msg *hdr = buf_msg(skb);
2278 
2279 	if (unlikely(msg_in_group(hdr)))
2280 		return READ_ONCE(sk->sk_rcvbuf);
2281 
2282 	if (unlikely(!msg_connected(hdr)))
2283 		return READ_ONCE(sk->sk_rcvbuf) << msg_importance(hdr);
2284 
2285 	if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL))
2286 		return READ_ONCE(sk->sk_rcvbuf);
2287 
2288 	return FLOWCTL_MSG_LIM;
2289 }
2290 
2291 /**
2292  * tipc_sk_filter_rcv - validate incoming message
2293  * @sk: socket
2294  * @skb: pointer to message.
2295  *
2296  * Enqueues message on receive queue if acceptable; optionally handles
2297  * disconnect indication for a connected socket.
2298  *
2299  * Called with socket lock already taken
2300  *
2301  */
2302 static void tipc_sk_filter_rcv(struct sock *sk, struct sk_buff *skb,
2303 			       struct sk_buff_head *xmitq)
2304 {
2305 	bool sk_conn = !tipc_sk_type_connectionless(sk);
2306 	struct tipc_sock *tsk = tipc_sk(sk);
2307 	struct tipc_group *grp = tsk->group;
2308 	struct tipc_msg *hdr = buf_msg(skb);
2309 	struct net *net = sock_net(sk);
2310 	struct sk_buff_head inputq;
2311 	int mtyp = msg_type(hdr);
2312 	int limit, err = TIPC_OK;
2313 
2314 	trace_tipc_sk_filter_rcv(sk, skb, TIPC_DUMP_ALL, " ");
2315 	TIPC_SKB_CB(skb)->bytes_read = 0;
2316 	__skb_queue_head_init(&inputq);
2317 	__skb_queue_tail(&inputq, skb);
2318 
2319 	if (unlikely(!msg_isdata(hdr)))
2320 		tipc_sk_proto_rcv(sk, &inputq, xmitq);
2321 
2322 	if (unlikely(grp))
2323 		tipc_group_filter_msg(grp, &inputq, xmitq);
2324 
2325 	if (unlikely(!grp) && mtyp == TIPC_MCAST_MSG)
2326 		tipc_mcast_filter_msg(net, &tsk->mc_method.deferredq, &inputq);
2327 
2328 	/* Validate and add to receive buffer if there is space */
2329 	while ((skb = __skb_dequeue(&inputq))) {
2330 		hdr = buf_msg(skb);
2331 		limit = rcvbuf_limit(sk, skb);
2332 		if ((sk_conn && !tipc_sk_filter_connect(tsk, skb, xmitq)) ||
2333 		    (!sk_conn && msg_connected(hdr)) ||
2334 		    (!grp && msg_in_group(hdr)))
2335 			err = TIPC_ERR_NO_PORT;
2336 		else if (sk_rmem_alloc_get(sk) + skb->truesize >= limit) {
2337 			trace_tipc_sk_dump(sk, skb, TIPC_DUMP_ALL,
2338 					   "err_overload2!");
2339 			atomic_inc(&sk->sk_drops);
2340 			err = TIPC_ERR_OVERLOAD;
2341 		}
2342 
2343 		if (unlikely(err)) {
2344 			if (tipc_msg_reverse(tipc_own_addr(net), &skb, err)) {
2345 				trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_NONE,
2346 						      "@filter_rcv!");
2347 				__skb_queue_tail(xmitq, skb);
2348 			}
2349 			err = TIPC_OK;
2350 			continue;
2351 		}
2352 		__skb_queue_tail(&sk->sk_receive_queue, skb);
2353 		skb_set_owner_r(skb, sk);
2354 		trace_tipc_sk_overlimit2(sk, skb, TIPC_DUMP_ALL,
2355 					 "rcvq >90% allocated!");
2356 		sk->sk_data_ready(sk);
2357 	}
2358 }
2359 
2360 /**
2361  * tipc_sk_backlog_rcv - handle incoming message from backlog queue
2362  * @sk: socket
2363  * @skb: message
2364  *
2365  * Caller must hold socket lock
2366  */
2367 static int tipc_sk_backlog_rcv(struct sock *sk, struct sk_buff *skb)
2368 {
2369 	unsigned int before = sk_rmem_alloc_get(sk);
2370 	struct sk_buff_head xmitq;
2371 	unsigned int added;
2372 
2373 	__skb_queue_head_init(&xmitq);
2374 
2375 	tipc_sk_filter_rcv(sk, skb, &xmitq);
2376 	added = sk_rmem_alloc_get(sk) - before;
2377 	atomic_add(added, &tipc_sk(sk)->dupl_rcvcnt);
2378 
2379 	/* Send pending response/rejected messages, if any */
2380 	tipc_node_distr_xmit(sock_net(sk), &xmitq);
2381 	return 0;
2382 }
2383 
2384 /**
2385  * tipc_sk_enqueue - extract all buffers with destination 'dport' from
2386  *                   inputq and try adding them to socket or backlog queue
2387  * @inputq: list of incoming buffers with potentially different destinations
2388  * @sk: socket where the buffers should be enqueued
2389  * @dport: port number for the socket
2390  *
2391  * Caller must hold socket lock
2392  */
2393 static void tipc_sk_enqueue(struct sk_buff_head *inputq, struct sock *sk,
2394 			    u32 dport, struct sk_buff_head *xmitq)
2395 {
2396 	unsigned long time_limit = jiffies + 2;
2397 	struct sk_buff *skb;
2398 	unsigned int lim;
2399 	atomic_t *dcnt;
2400 	u32 onode;
2401 
2402 	while (skb_queue_len(inputq)) {
2403 		if (unlikely(time_after_eq(jiffies, time_limit)))
2404 			return;
2405 
2406 		skb = tipc_skb_dequeue(inputq, dport);
2407 		if (unlikely(!skb))
2408 			return;
2409 
2410 		/* Add message directly to receive queue if possible */
2411 		if (!sock_owned_by_user(sk)) {
2412 			tipc_sk_filter_rcv(sk, skb, xmitq);
2413 			continue;
2414 		}
2415 
2416 		/* Try backlog, compensating for double-counted bytes */
2417 		dcnt = &tipc_sk(sk)->dupl_rcvcnt;
2418 		if (!sk->sk_backlog.len)
2419 			atomic_set(dcnt, 0);
2420 		lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt);
2421 		if (likely(!sk_add_backlog(sk, skb, lim))) {
2422 			trace_tipc_sk_overlimit1(sk, skb, TIPC_DUMP_ALL,
2423 						 "bklg & rcvq >90% allocated!");
2424 			continue;
2425 		}
2426 
2427 		trace_tipc_sk_dump(sk, skb, TIPC_DUMP_ALL, "err_overload!");
2428 		/* Overload => reject message back to sender */
2429 		onode = tipc_own_addr(sock_net(sk));
2430 		atomic_inc(&sk->sk_drops);
2431 		if (tipc_msg_reverse(onode, &skb, TIPC_ERR_OVERLOAD)) {
2432 			trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_ALL,
2433 					      "@sk_enqueue!");
2434 			__skb_queue_tail(xmitq, skb);
2435 		}
2436 		break;
2437 	}
2438 }
2439 
2440 /**
2441  * tipc_sk_rcv - handle a chain of incoming buffers
2442  * @inputq: buffer list containing the buffers
2443  * Consumes all buffers in list until inputq is empty
2444  * Note: may be called in multiple threads referring to the same queue
2445  */
2446 void tipc_sk_rcv(struct net *net, struct sk_buff_head *inputq)
2447 {
2448 	struct sk_buff_head xmitq;
2449 	u32 dnode, dport = 0;
2450 	int err;
2451 	struct tipc_sock *tsk;
2452 	struct sock *sk;
2453 	struct sk_buff *skb;
2454 
2455 	__skb_queue_head_init(&xmitq);
2456 	while (skb_queue_len(inputq)) {
2457 		dport = tipc_skb_peek_port(inputq, dport);
2458 		tsk = tipc_sk_lookup(net, dport);
2459 
2460 		if (likely(tsk)) {
2461 			sk = &tsk->sk;
2462 			if (likely(spin_trylock_bh(&sk->sk_lock.slock))) {
2463 				tipc_sk_enqueue(inputq, sk, dport, &xmitq);
2464 				spin_unlock_bh(&sk->sk_lock.slock);
2465 			}
2466 			/* Send pending response/rejected messages, if any */
2467 			tipc_node_distr_xmit(sock_net(sk), &xmitq);
2468 			sock_put(sk);
2469 			continue;
2470 		}
2471 		/* No destination socket => dequeue skb if still there */
2472 		skb = tipc_skb_dequeue(inputq, dport);
2473 		if (!skb)
2474 			return;
2475 
2476 		/* Try secondary lookup if unresolved named message */
2477 		err = TIPC_ERR_NO_PORT;
2478 		if (tipc_msg_lookup_dest(net, skb, &err))
2479 			goto xmit;
2480 
2481 		/* Prepare for message rejection */
2482 		if (!tipc_msg_reverse(tipc_own_addr(net), &skb, err))
2483 			continue;
2484 
2485 		trace_tipc_sk_rej_msg(NULL, skb, TIPC_DUMP_NONE, "@sk_rcv!");
2486 xmit:
2487 		dnode = msg_destnode(buf_msg(skb));
2488 		tipc_node_xmit_skb(net, skb, dnode, dport);
2489 	}
2490 }
2491 
2492 static int tipc_wait_for_connect(struct socket *sock, long *timeo_p)
2493 {
2494 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
2495 	struct sock *sk = sock->sk;
2496 	int done;
2497 
2498 	do {
2499 		int err = sock_error(sk);
2500 		if (err)
2501 			return err;
2502 		if (!*timeo_p)
2503 			return -ETIMEDOUT;
2504 		if (signal_pending(current))
2505 			return sock_intr_errno(*timeo_p);
2506 		if (sk->sk_state == TIPC_DISCONNECTING)
2507 			break;
2508 
2509 		add_wait_queue(sk_sleep(sk), &wait);
2510 		done = sk_wait_event(sk, timeo_p, tipc_sk_connected(sk),
2511 				     &wait);
2512 		remove_wait_queue(sk_sleep(sk), &wait);
2513 	} while (!done);
2514 	return 0;
2515 }
2516 
2517 static bool tipc_sockaddr_is_sane(struct sockaddr_tipc *addr)
2518 {
2519 	if (addr->family != AF_TIPC)
2520 		return false;
2521 	if (addr->addrtype == TIPC_SERVICE_RANGE)
2522 		return (addr->addr.nameseq.lower <= addr->addr.nameseq.upper);
2523 	return (addr->addrtype == TIPC_SERVICE_ADDR ||
2524 		addr->addrtype == TIPC_SOCKET_ADDR);
2525 }
2526 
2527 /**
2528  * tipc_connect - establish a connection to another TIPC port
2529  * @sock: socket structure
2530  * @dest: socket address for destination port
2531  * @destlen: size of socket address data structure
2532  * @flags: file-related flags associated with socket
2533  *
2534  * Returns 0 on success, errno otherwise
2535  */
2536 static int tipc_connect(struct socket *sock, struct sockaddr *dest,
2537 			int destlen, int flags)
2538 {
2539 	struct sock *sk = sock->sk;
2540 	struct tipc_sock *tsk = tipc_sk(sk);
2541 	struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest;
2542 	struct msghdr m = {NULL,};
2543 	long timeout = (flags & O_NONBLOCK) ? 0 : tsk->conn_timeout;
2544 	int previous;
2545 	int res = 0;
2546 
2547 	if (destlen != sizeof(struct sockaddr_tipc))
2548 		return -EINVAL;
2549 
2550 	lock_sock(sk);
2551 
2552 	if (tsk->group) {
2553 		res = -EINVAL;
2554 		goto exit;
2555 	}
2556 
2557 	if (dst->family == AF_UNSPEC) {
2558 		memset(&tsk->peer, 0, sizeof(struct sockaddr_tipc));
2559 		if (!tipc_sk_type_connectionless(sk))
2560 			res = -EINVAL;
2561 		goto exit;
2562 	}
2563 	if (!tipc_sockaddr_is_sane(dst)) {
2564 		res = -EINVAL;
2565 		goto exit;
2566 	}
2567 	/* DGRAM/RDM connect(), just save the destaddr */
2568 	if (tipc_sk_type_connectionless(sk)) {
2569 		memcpy(&tsk->peer, dest, destlen);
2570 		goto exit;
2571 	} else if (dst->addrtype == TIPC_SERVICE_RANGE) {
2572 		res = -EINVAL;
2573 		goto exit;
2574 	}
2575 
2576 	previous = sk->sk_state;
2577 
2578 	switch (sk->sk_state) {
2579 	case TIPC_OPEN:
2580 		/* Send a 'SYN-' to destination */
2581 		m.msg_name = dest;
2582 		m.msg_namelen = destlen;
2583 
2584 		/* If connect is in non-blocking case, set MSG_DONTWAIT to
2585 		 * indicate send_msg() is never blocked.
2586 		 */
2587 		if (!timeout)
2588 			m.msg_flags = MSG_DONTWAIT;
2589 
2590 		res = __tipc_sendmsg(sock, &m, 0);
2591 		if ((res < 0) && (res != -EWOULDBLOCK))
2592 			goto exit;
2593 
2594 		/* Just entered TIPC_CONNECTING state; the only
2595 		 * difference is that return value in non-blocking
2596 		 * case is EINPROGRESS, rather than EALREADY.
2597 		 */
2598 		res = -EINPROGRESS;
2599 		fallthrough;
2600 	case TIPC_CONNECTING:
2601 		if (!timeout) {
2602 			if (previous == TIPC_CONNECTING)
2603 				res = -EALREADY;
2604 			goto exit;
2605 		}
2606 		timeout = msecs_to_jiffies(timeout);
2607 		/* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */
2608 		res = tipc_wait_for_connect(sock, &timeout);
2609 		break;
2610 	case TIPC_ESTABLISHED:
2611 		res = -EISCONN;
2612 		break;
2613 	default:
2614 		res = -EINVAL;
2615 	}
2616 
2617 exit:
2618 	release_sock(sk);
2619 	return res;
2620 }
2621 
2622 /**
2623  * tipc_listen - allow socket to listen for incoming connections
2624  * @sock: socket structure
2625  * @len: (unused)
2626  *
2627  * Returns 0 on success, errno otherwise
2628  */
2629 static int tipc_listen(struct socket *sock, int len)
2630 {
2631 	struct sock *sk = sock->sk;
2632 	int res;
2633 
2634 	lock_sock(sk);
2635 	res = tipc_set_sk_state(sk, TIPC_LISTEN);
2636 	release_sock(sk);
2637 
2638 	return res;
2639 }
2640 
2641 static int tipc_wait_for_accept(struct socket *sock, long timeo)
2642 {
2643 	struct sock *sk = sock->sk;
2644 	DEFINE_WAIT(wait);
2645 	int err;
2646 
2647 	/* True wake-one mechanism for incoming connections: only
2648 	 * one process gets woken up, not the 'whole herd'.
2649 	 * Since we do not 'race & poll' for established sockets
2650 	 * anymore, the common case will execute the loop only once.
2651 	*/
2652 	for (;;) {
2653 		prepare_to_wait_exclusive(sk_sleep(sk), &wait,
2654 					  TASK_INTERRUPTIBLE);
2655 		if (timeo && skb_queue_empty(&sk->sk_receive_queue)) {
2656 			release_sock(sk);
2657 			timeo = schedule_timeout(timeo);
2658 			lock_sock(sk);
2659 		}
2660 		err = 0;
2661 		if (!skb_queue_empty(&sk->sk_receive_queue))
2662 			break;
2663 		err = -EAGAIN;
2664 		if (!timeo)
2665 			break;
2666 		err = sock_intr_errno(timeo);
2667 		if (signal_pending(current))
2668 			break;
2669 	}
2670 	finish_wait(sk_sleep(sk), &wait);
2671 	return err;
2672 }
2673 
2674 /**
2675  * tipc_accept - wait for connection request
2676  * @sock: listening socket
2677  * @new_sock: new socket that is to be connected
2678  * @flags: file-related flags associated with socket
2679  *
2680  * Returns 0 on success, errno otherwise
2681  */
2682 static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags,
2683 		       bool kern)
2684 {
2685 	struct sock *new_sk, *sk = sock->sk;
2686 	struct sk_buff *buf;
2687 	struct tipc_sock *new_tsock;
2688 	struct tipc_msg *msg;
2689 	long timeo;
2690 	int res;
2691 
2692 	lock_sock(sk);
2693 
2694 	if (sk->sk_state != TIPC_LISTEN) {
2695 		res = -EINVAL;
2696 		goto exit;
2697 	}
2698 	timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK);
2699 	res = tipc_wait_for_accept(sock, timeo);
2700 	if (res)
2701 		goto exit;
2702 
2703 	buf = skb_peek(&sk->sk_receive_queue);
2704 
2705 	res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, kern);
2706 	if (res)
2707 		goto exit;
2708 	security_sk_clone(sock->sk, new_sock->sk);
2709 
2710 	new_sk = new_sock->sk;
2711 	new_tsock = tipc_sk(new_sk);
2712 	msg = buf_msg(buf);
2713 
2714 	/* we lock on new_sk; but lockdep sees the lock on sk */
2715 	lock_sock_nested(new_sk, SINGLE_DEPTH_NESTING);
2716 
2717 	/*
2718 	 * Reject any stray messages received by new socket
2719 	 * before the socket lock was taken (very, very unlikely)
2720 	 */
2721 	tsk_rej_rx_queue(new_sk, TIPC_ERR_NO_PORT);
2722 
2723 	/* Connect new socket to it's peer */
2724 	tipc_sk_finish_conn(new_tsock, msg_origport(msg), msg_orignode(msg));
2725 
2726 	tsk_set_importance(new_sk, msg_importance(msg));
2727 	if (msg_named(msg)) {
2728 		new_tsock->conn_type = msg_nametype(msg);
2729 		new_tsock->conn_instance = msg_nameinst(msg);
2730 	}
2731 
2732 	/*
2733 	 * Respond to 'SYN-' by discarding it & returning 'ACK'-.
2734 	 * Respond to 'SYN+' by queuing it on new socket.
2735 	 */
2736 	if (!msg_data_sz(msg)) {
2737 		struct msghdr m = {NULL,};
2738 
2739 		tsk_advance_rx_queue(sk);
2740 		__tipc_sendstream(new_sock, &m, 0);
2741 	} else {
2742 		__skb_dequeue(&sk->sk_receive_queue);
2743 		__skb_queue_head(&new_sk->sk_receive_queue, buf);
2744 		skb_set_owner_r(buf, new_sk);
2745 	}
2746 	release_sock(new_sk);
2747 exit:
2748 	release_sock(sk);
2749 	return res;
2750 }
2751 
2752 /**
2753  * tipc_shutdown - shutdown socket connection
2754  * @sock: socket structure
2755  * @how: direction to close (must be SHUT_RDWR)
2756  *
2757  * Terminates connection (if necessary), then purges socket's receive queue.
2758  *
2759  * Returns 0 on success, errno otherwise
2760  */
2761 static int tipc_shutdown(struct socket *sock, int how)
2762 {
2763 	struct sock *sk = sock->sk;
2764 	int res;
2765 
2766 	if (how != SHUT_RDWR)
2767 		return -EINVAL;
2768 
2769 	lock_sock(sk);
2770 
2771 	trace_tipc_sk_shutdown(sk, NULL, TIPC_DUMP_ALL, " ");
2772 	__tipc_shutdown(sock, TIPC_CONN_SHUTDOWN);
2773 	sk->sk_shutdown = SHUTDOWN_MASK;
2774 
2775 	if (sk->sk_state == TIPC_DISCONNECTING) {
2776 		/* Discard any unreceived messages */
2777 		__skb_queue_purge(&sk->sk_receive_queue);
2778 
2779 		res = 0;
2780 	} else {
2781 		res = -ENOTCONN;
2782 	}
2783 	/* Wake up anyone sleeping in poll. */
2784 	sk->sk_state_change(sk);
2785 
2786 	release_sock(sk);
2787 	return res;
2788 }
2789 
2790 static void tipc_sk_check_probing_state(struct sock *sk,
2791 					struct sk_buff_head *list)
2792 {
2793 	struct tipc_sock *tsk = tipc_sk(sk);
2794 	u32 pnode = tsk_peer_node(tsk);
2795 	u32 pport = tsk_peer_port(tsk);
2796 	u32 self = tsk_own_node(tsk);
2797 	u32 oport = tsk->portid;
2798 	struct sk_buff *skb;
2799 
2800 	if (tsk->probe_unacked) {
2801 		tipc_set_sk_state(sk, TIPC_DISCONNECTING);
2802 		sk->sk_err = ECONNABORTED;
2803 		tipc_node_remove_conn(sock_net(sk), pnode, pport);
2804 		sk->sk_state_change(sk);
2805 		return;
2806 	}
2807 	/* Prepare new probe */
2808 	skb = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE, 0,
2809 			      pnode, self, pport, oport, TIPC_OK);
2810 	if (skb)
2811 		__skb_queue_tail(list, skb);
2812 	tsk->probe_unacked = true;
2813 	sk_reset_timer(sk, &sk->sk_timer, jiffies + CONN_PROBING_INTV);
2814 }
2815 
2816 static void tipc_sk_retry_connect(struct sock *sk, struct sk_buff_head *list)
2817 {
2818 	struct tipc_sock *tsk = tipc_sk(sk);
2819 
2820 	/* Try again later if dest link is congested */
2821 	if (tsk->cong_link_cnt) {
2822 		sk_reset_timer(sk, &sk->sk_timer, msecs_to_jiffies(100));
2823 		return;
2824 	}
2825 	/* Prepare SYN for retransmit */
2826 	tipc_msg_skb_clone(&sk->sk_write_queue, list);
2827 }
2828 
2829 static void tipc_sk_timeout(struct timer_list *t)
2830 {
2831 	struct sock *sk = from_timer(sk, t, sk_timer);
2832 	struct tipc_sock *tsk = tipc_sk(sk);
2833 	u32 pnode = tsk_peer_node(tsk);
2834 	struct sk_buff_head list;
2835 	int rc = 0;
2836 
2837 	__skb_queue_head_init(&list);
2838 	bh_lock_sock(sk);
2839 
2840 	/* Try again later if socket is busy */
2841 	if (sock_owned_by_user(sk)) {
2842 		sk_reset_timer(sk, &sk->sk_timer, jiffies + HZ / 20);
2843 		bh_unlock_sock(sk);
2844 		sock_put(sk);
2845 		return;
2846 	}
2847 
2848 	if (sk->sk_state == TIPC_ESTABLISHED)
2849 		tipc_sk_check_probing_state(sk, &list);
2850 	else if (sk->sk_state == TIPC_CONNECTING)
2851 		tipc_sk_retry_connect(sk, &list);
2852 
2853 	bh_unlock_sock(sk);
2854 
2855 	if (!skb_queue_empty(&list))
2856 		rc = tipc_node_xmit(sock_net(sk), &list, pnode, tsk->portid);
2857 
2858 	/* SYN messages may cause link congestion */
2859 	if (rc == -ELINKCONG) {
2860 		tipc_dest_push(&tsk->cong_links, pnode, 0);
2861 		tsk->cong_link_cnt = 1;
2862 	}
2863 	sock_put(sk);
2864 }
2865 
2866 static int tipc_sk_publish(struct tipc_sock *tsk, uint scope,
2867 			   struct tipc_name_seq const *seq)
2868 {
2869 	struct sock *sk = &tsk->sk;
2870 	struct net *net = sock_net(sk);
2871 	struct publication *publ;
2872 	u32 key;
2873 
2874 	if (scope != TIPC_NODE_SCOPE)
2875 		scope = TIPC_CLUSTER_SCOPE;
2876 
2877 	if (tipc_sk_connected(sk))
2878 		return -EINVAL;
2879 	key = tsk->portid + tsk->pub_count + 1;
2880 	if (key == tsk->portid)
2881 		return -EADDRINUSE;
2882 
2883 	publ = tipc_nametbl_publish(net, seq->type, seq->lower, seq->upper,
2884 				    scope, tsk->portid, key);
2885 	if (unlikely(!publ))
2886 		return -EINVAL;
2887 
2888 	list_add(&publ->binding_sock, &tsk->publications);
2889 	tsk->pub_count++;
2890 	tsk->published = 1;
2891 	return 0;
2892 }
2893 
2894 static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope,
2895 			    struct tipc_name_seq const *seq)
2896 {
2897 	struct net *net = sock_net(&tsk->sk);
2898 	struct publication *publ;
2899 	struct publication *safe;
2900 	int rc = -EINVAL;
2901 
2902 	if (scope != TIPC_NODE_SCOPE)
2903 		scope = TIPC_CLUSTER_SCOPE;
2904 
2905 	list_for_each_entry_safe(publ, safe, &tsk->publications, binding_sock) {
2906 		if (seq) {
2907 			if (publ->scope != scope)
2908 				continue;
2909 			if (publ->type != seq->type)
2910 				continue;
2911 			if (publ->lower != seq->lower)
2912 				continue;
2913 			if (publ->upper != seq->upper)
2914 				break;
2915 			tipc_nametbl_withdraw(net, publ->type, publ->lower,
2916 					      publ->upper, publ->key);
2917 			rc = 0;
2918 			break;
2919 		}
2920 		tipc_nametbl_withdraw(net, publ->type, publ->lower,
2921 				      publ->upper, publ->key);
2922 		rc = 0;
2923 	}
2924 	if (list_empty(&tsk->publications))
2925 		tsk->published = 0;
2926 	return rc;
2927 }
2928 
2929 /* tipc_sk_reinit: set non-zero address in all existing sockets
2930  *                 when we go from standalone to network mode.
2931  */
2932 void tipc_sk_reinit(struct net *net)
2933 {
2934 	struct tipc_net *tn = net_generic(net, tipc_net_id);
2935 	struct rhashtable_iter iter;
2936 	struct tipc_sock *tsk;
2937 	struct tipc_msg *msg;
2938 
2939 	rhashtable_walk_enter(&tn->sk_rht, &iter);
2940 
2941 	do {
2942 		rhashtable_walk_start(&iter);
2943 
2944 		while ((tsk = rhashtable_walk_next(&iter)) && !IS_ERR(tsk)) {
2945 			sock_hold(&tsk->sk);
2946 			rhashtable_walk_stop(&iter);
2947 			lock_sock(&tsk->sk);
2948 			msg = &tsk->phdr;
2949 			msg_set_prevnode(msg, tipc_own_addr(net));
2950 			msg_set_orignode(msg, tipc_own_addr(net));
2951 			release_sock(&tsk->sk);
2952 			rhashtable_walk_start(&iter);
2953 			sock_put(&tsk->sk);
2954 		}
2955 
2956 		rhashtable_walk_stop(&iter);
2957 	} while (tsk == ERR_PTR(-EAGAIN));
2958 
2959 	rhashtable_walk_exit(&iter);
2960 }
2961 
2962 static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid)
2963 {
2964 	struct tipc_net *tn = net_generic(net, tipc_net_id);
2965 	struct tipc_sock *tsk;
2966 
2967 	rcu_read_lock();
2968 	tsk = rhashtable_lookup(&tn->sk_rht, &portid, tsk_rht_params);
2969 	if (tsk)
2970 		sock_hold(&tsk->sk);
2971 	rcu_read_unlock();
2972 
2973 	return tsk;
2974 }
2975 
2976 static int tipc_sk_insert(struct tipc_sock *tsk)
2977 {
2978 	struct sock *sk = &tsk->sk;
2979 	struct net *net = sock_net(sk);
2980 	struct tipc_net *tn = net_generic(net, tipc_net_id);
2981 	u32 remaining = (TIPC_MAX_PORT - TIPC_MIN_PORT) + 1;
2982 	u32 portid = prandom_u32() % remaining + TIPC_MIN_PORT;
2983 
2984 	while (remaining--) {
2985 		portid++;
2986 		if ((portid < TIPC_MIN_PORT) || (portid > TIPC_MAX_PORT))
2987 			portid = TIPC_MIN_PORT;
2988 		tsk->portid = portid;
2989 		sock_hold(&tsk->sk);
2990 		if (!rhashtable_lookup_insert_fast(&tn->sk_rht, &tsk->node,
2991 						   tsk_rht_params))
2992 			return 0;
2993 		sock_put(&tsk->sk);
2994 	}
2995 
2996 	return -1;
2997 }
2998 
2999 static void tipc_sk_remove(struct tipc_sock *tsk)
3000 {
3001 	struct sock *sk = &tsk->sk;
3002 	struct tipc_net *tn = net_generic(sock_net(sk), tipc_net_id);
3003 
3004 	if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params)) {
3005 		WARN_ON(refcount_read(&sk->sk_refcnt) == 1);
3006 		__sock_put(sk);
3007 	}
3008 }
3009 
3010 static const struct rhashtable_params tsk_rht_params = {
3011 	.nelem_hint = 192,
3012 	.head_offset = offsetof(struct tipc_sock, node),
3013 	.key_offset = offsetof(struct tipc_sock, portid),
3014 	.key_len = sizeof(u32), /* portid */
3015 	.max_size = 1048576,
3016 	.min_size = 256,
3017 	.automatic_shrinking = true,
3018 };
3019 
3020 int tipc_sk_rht_init(struct net *net)
3021 {
3022 	struct tipc_net *tn = net_generic(net, tipc_net_id);
3023 
3024 	return rhashtable_init(&tn->sk_rht, &tsk_rht_params);
3025 }
3026 
3027 void tipc_sk_rht_destroy(struct net *net)
3028 {
3029 	struct tipc_net *tn = net_generic(net, tipc_net_id);
3030 
3031 	/* Wait for socket readers to complete */
3032 	synchronize_net();
3033 
3034 	rhashtable_destroy(&tn->sk_rht);
3035 }
3036 
3037 static int tipc_sk_join(struct tipc_sock *tsk, struct tipc_group_req *mreq)
3038 {
3039 	struct net *net = sock_net(&tsk->sk);
3040 	struct tipc_group *grp = tsk->group;
3041 	struct tipc_msg *hdr = &tsk->phdr;
3042 	struct tipc_name_seq seq;
3043 	int rc;
3044 
3045 	if (mreq->type < TIPC_RESERVED_TYPES)
3046 		return -EACCES;
3047 	if (mreq->scope > TIPC_NODE_SCOPE)
3048 		return -EINVAL;
3049 	if (grp)
3050 		return -EACCES;
3051 	grp = tipc_group_create(net, tsk->portid, mreq, &tsk->group_is_open);
3052 	if (!grp)
3053 		return -ENOMEM;
3054 	tsk->group = grp;
3055 	msg_set_lookup_scope(hdr, mreq->scope);
3056 	msg_set_nametype(hdr, mreq->type);
3057 	msg_set_dest_droppable(hdr, true);
3058 	seq.type = mreq->type;
3059 	seq.lower = mreq->instance;
3060 	seq.upper = seq.lower;
3061 	tipc_nametbl_build_group(net, grp, mreq->type, mreq->scope);
3062 	rc = tipc_sk_publish(tsk, mreq->scope, &seq);
3063 	if (rc) {
3064 		tipc_group_delete(net, grp);
3065 		tsk->group = NULL;
3066 		return rc;
3067 	}
3068 	/* Eliminate any risk that a broadcast overtakes sent JOINs */
3069 	tsk->mc_method.rcast = true;
3070 	tsk->mc_method.mandatory = true;
3071 	tipc_group_join(net, grp, &tsk->sk.sk_rcvbuf);
3072 	return rc;
3073 }
3074 
3075 static int tipc_sk_leave(struct tipc_sock *tsk)
3076 {
3077 	struct net *net = sock_net(&tsk->sk);
3078 	struct tipc_group *grp = tsk->group;
3079 	struct tipc_name_seq seq;
3080 	int scope;
3081 
3082 	if (!grp)
3083 		return -EINVAL;
3084 	tipc_group_self(grp, &seq, &scope);
3085 	tipc_group_delete(net, grp);
3086 	tsk->group = NULL;
3087 	tipc_sk_withdraw(tsk, scope, &seq);
3088 	return 0;
3089 }
3090 
3091 /**
3092  * tipc_setsockopt - set socket option
3093  * @sock: socket structure
3094  * @lvl: option level
3095  * @opt: option identifier
3096  * @ov: pointer to new option value
3097  * @ol: length of option value
3098  *
3099  * For stream sockets only, accepts and ignores all IPPROTO_TCP options
3100  * (to ease compatibility).
3101  *
3102  * Returns 0 on success, errno otherwise
3103  */
3104 static int tipc_setsockopt(struct socket *sock, int lvl, int opt,
3105 			   sockptr_t ov, unsigned int ol)
3106 {
3107 	struct sock *sk = sock->sk;
3108 	struct tipc_sock *tsk = tipc_sk(sk);
3109 	struct tipc_group_req mreq;
3110 	u32 value = 0;
3111 	int res = 0;
3112 
3113 	if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
3114 		return 0;
3115 	if (lvl != SOL_TIPC)
3116 		return -ENOPROTOOPT;
3117 
3118 	switch (opt) {
3119 	case TIPC_IMPORTANCE:
3120 	case TIPC_SRC_DROPPABLE:
3121 	case TIPC_DEST_DROPPABLE:
3122 	case TIPC_CONN_TIMEOUT:
3123 	case TIPC_NODELAY:
3124 		if (ol < sizeof(value))
3125 			return -EINVAL;
3126 		if (copy_from_sockptr(&value, ov, sizeof(u32)))
3127 			return -EFAULT;
3128 		break;
3129 	case TIPC_GROUP_JOIN:
3130 		if (ol < sizeof(mreq))
3131 			return -EINVAL;
3132 		if (copy_from_sockptr(&mreq, ov, sizeof(mreq)))
3133 			return -EFAULT;
3134 		break;
3135 	default:
3136 		if (!sockptr_is_null(ov) || ol)
3137 			return -EINVAL;
3138 	}
3139 
3140 	lock_sock(sk);
3141 
3142 	switch (opt) {
3143 	case TIPC_IMPORTANCE:
3144 		res = tsk_set_importance(sk, value);
3145 		break;
3146 	case TIPC_SRC_DROPPABLE:
3147 		if (sock->type != SOCK_STREAM)
3148 			tsk_set_unreliable(tsk, value);
3149 		else
3150 			res = -ENOPROTOOPT;
3151 		break;
3152 	case TIPC_DEST_DROPPABLE:
3153 		tsk_set_unreturnable(tsk, value);
3154 		break;
3155 	case TIPC_CONN_TIMEOUT:
3156 		tipc_sk(sk)->conn_timeout = value;
3157 		break;
3158 	case TIPC_MCAST_BROADCAST:
3159 		tsk->mc_method.rcast = false;
3160 		tsk->mc_method.mandatory = true;
3161 		break;
3162 	case TIPC_MCAST_REPLICAST:
3163 		tsk->mc_method.rcast = true;
3164 		tsk->mc_method.mandatory = true;
3165 		break;
3166 	case TIPC_GROUP_JOIN:
3167 		res = tipc_sk_join(tsk, &mreq);
3168 		break;
3169 	case TIPC_GROUP_LEAVE:
3170 		res = tipc_sk_leave(tsk);
3171 		break;
3172 	case TIPC_NODELAY:
3173 		tsk->nodelay = !!value;
3174 		tsk_set_nagle(tsk);
3175 		break;
3176 	default:
3177 		res = -EINVAL;
3178 	}
3179 
3180 	release_sock(sk);
3181 
3182 	return res;
3183 }
3184 
3185 /**
3186  * tipc_getsockopt - get socket option
3187  * @sock: socket structure
3188  * @lvl: option level
3189  * @opt: option identifier
3190  * @ov: receptacle for option value
3191  * @ol: receptacle for length of option value
3192  *
3193  * For stream sockets only, returns 0 length result for all IPPROTO_TCP options
3194  * (to ease compatibility).
3195  *
3196  * Returns 0 on success, errno otherwise
3197  */
3198 static int tipc_getsockopt(struct socket *sock, int lvl, int opt,
3199 			   char __user *ov, int __user *ol)
3200 {
3201 	struct sock *sk = sock->sk;
3202 	struct tipc_sock *tsk = tipc_sk(sk);
3203 	struct tipc_name_seq seq;
3204 	int len, scope;
3205 	u32 value;
3206 	int res;
3207 
3208 	if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
3209 		return put_user(0, ol);
3210 	if (lvl != SOL_TIPC)
3211 		return -ENOPROTOOPT;
3212 	res = get_user(len, ol);
3213 	if (res)
3214 		return res;
3215 
3216 	lock_sock(sk);
3217 
3218 	switch (opt) {
3219 	case TIPC_IMPORTANCE:
3220 		value = tsk_importance(tsk);
3221 		break;
3222 	case TIPC_SRC_DROPPABLE:
3223 		value = tsk_unreliable(tsk);
3224 		break;
3225 	case TIPC_DEST_DROPPABLE:
3226 		value = tsk_unreturnable(tsk);
3227 		break;
3228 	case TIPC_CONN_TIMEOUT:
3229 		value = tsk->conn_timeout;
3230 		/* no need to set "res", since already 0 at this point */
3231 		break;
3232 	case TIPC_NODE_RECVQ_DEPTH:
3233 		value = 0; /* was tipc_queue_size, now obsolete */
3234 		break;
3235 	case TIPC_SOCK_RECVQ_DEPTH:
3236 		value = skb_queue_len(&sk->sk_receive_queue);
3237 		break;
3238 	case TIPC_SOCK_RECVQ_USED:
3239 		value = sk_rmem_alloc_get(sk);
3240 		break;
3241 	case TIPC_GROUP_JOIN:
3242 		seq.type = 0;
3243 		if (tsk->group)
3244 			tipc_group_self(tsk->group, &seq, &scope);
3245 		value = seq.type;
3246 		break;
3247 	default:
3248 		res = -EINVAL;
3249 	}
3250 
3251 	release_sock(sk);
3252 
3253 	if (res)
3254 		return res;	/* "get" failed */
3255 
3256 	if (len < sizeof(value))
3257 		return -EINVAL;
3258 
3259 	if (copy_to_user(ov, &value, sizeof(value)))
3260 		return -EFAULT;
3261 
3262 	return put_user(sizeof(value), ol);
3263 }
3264 
3265 static int tipc_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
3266 {
3267 	struct net *net = sock_net(sock->sk);
3268 	struct tipc_sioc_nodeid_req nr = {0};
3269 	struct tipc_sioc_ln_req lnr;
3270 	void __user *argp = (void __user *)arg;
3271 
3272 	switch (cmd) {
3273 	case SIOCGETLINKNAME:
3274 		if (copy_from_user(&lnr, argp, sizeof(lnr)))
3275 			return -EFAULT;
3276 		if (!tipc_node_get_linkname(net,
3277 					    lnr.bearer_id & 0xffff, lnr.peer,
3278 					    lnr.linkname, TIPC_MAX_LINK_NAME)) {
3279 			if (copy_to_user(argp, &lnr, sizeof(lnr)))
3280 				return -EFAULT;
3281 			return 0;
3282 		}
3283 		return -EADDRNOTAVAIL;
3284 	case SIOCGETNODEID:
3285 		if (copy_from_user(&nr, argp, sizeof(nr)))
3286 			return -EFAULT;
3287 		if (!tipc_node_get_id(net, nr.peer, nr.node_id))
3288 			return -EADDRNOTAVAIL;
3289 		if (copy_to_user(argp, &nr, sizeof(nr)))
3290 			return -EFAULT;
3291 		return 0;
3292 	default:
3293 		return -ENOIOCTLCMD;
3294 	}
3295 }
3296 
3297 static int tipc_socketpair(struct socket *sock1, struct socket *sock2)
3298 {
3299 	struct tipc_sock *tsk2 = tipc_sk(sock2->sk);
3300 	struct tipc_sock *tsk1 = tipc_sk(sock1->sk);
3301 	u32 onode = tipc_own_addr(sock_net(sock1->sk));
3302 
3303 	tsk1->peer.family = AF_TIPC;
3304 	tsk1->peer.addrtype = TIPC_ADDR_ID;
3305 	tsk1->peer.scope = TIPC_NODE_SCOPE;
3306 	tsk1->peer.addr.id.ref = tsk2->portid;
3307 	tsk1->peer.addr.id.node = onode;
3308 	tsk2->peer.family = AF_TIPC;
3309 	tsk2->peer.addrtype = TIPC_ADDR_ID;
3310 	tsk2->peer.scope = TIPC_NODE_SCOPE;
3311 	tsk2->peer.addr.id.ref = tsk1->portid;
3312 	tsk2->peer.addr.id.node = onode;
3313 
3314 	tipc_sk_finish_conn(tsk1, tsk2->portid, onode);
3315 	tipc_sk_finish_conn(tsk2, tsk1->portid, onode);
3316 	return 0;
3317 }
3318 
3319 /* Protocol switches for the various types of TIPC sockets */
3320 
3321 static const struct proto_ops msg_ops = {
3322 	.owner		= THIS_MODULE,
3323 	.family		= AF_TIPC,
3324 	.release	= tipc_release,
3325 	.bind		= tipc_bind,
3326 	.connect	= tipc_connect,
3327 	.socketpair	= tipc_socketpair,
3328 	.accept		= sock_no_accept,
3329 	.getname	= tipc_getname,
3330 	.poll		= tipc_poll,
3331 	.ioctl		= tipc_ioctl,
3332 	.listen		= sock_no_listen,
3333 	.shutdown	= tipc_shutdown,
3334 	.setsockopt	= tipc_setsockopt,
3335 	.getsockopt	= tipc_getsockopt,
3336 	.sendmsg	= tipc_sendmsg,
3337 	.recvmsg	= tipc_recvmsg,
3338 	.mmap		= sock_no_mmap,
3339 	.sendpage	= sock_no_sendpage
3340 };
3341 
3342 static const struct proto_ops packet_ops = {
3343 	.owner		= THIS_MODULE,
3344 	.family		= AF_TIPC,
3345 	.release	= tipc_release,
3346 	.bind		= tipc_bind,
3347 	.connect	= tipc_connect,
3348 	.socketpair	= tipc_socketpair,
3349 	.accept		= tipc_accept,
3350 	.getname	= tipc_getname,
3351 	.poll		= tipc_poll,
3352 	.ioctl		= tipc_ioctl,
3353 	.listen		= tipc_listen,
3354 	.shutdown	= tipc_shutdown,
3355 	.setsockopt	= tipc_setsockopt,
3356 	.getsockopt	= tipc_getsockopt,
3357 	.sendmsg	= tipc_send_packet,
3358 	.recvmsg	= tipc_recvmsg,
3359 	.mmap		= sock_no_mmap,
3360 	.sendpage	= sock_no_sendpage
3361 };
3362 
3363 static const struct proto_ops stream_ops = {
3364 	.owner		= THIS_MODULE,
3365 	.family		= AF_TIPC,
3366 	.release	= tipc_release,
3367 	.bind		= tipc_bind,
3368 	.connect	= tipc_connect,
3369 	.socketpair	= tipc_socketpair,
3370 	.accept		= tipc_accept,
3371 	.getname	= tipc_getname,
3372 	.poll		= tipc_poll,
3373 	.ioctl		= tipc_ioctl,
3374 	.listen		= tipc_listen,
3375 	.shutdown	= tipc_shutdown,
3376 	.setsockopt	= tipc_setsockopt,
3377 	.getsockopt	= tipc_getsockopt,
3378 	.sendmsg	= tipc_sendstream,
3379 	.recvmsg	= tipc_recvstream,
3380 	.mmap		= sock_no_mmap,
3381 	.sendpage	= sock_no_sendpage
3382 };
3383 
3384 static const struct net_proto_family tipc_family_ops = {
3385 	.owner		= THIS_MODULE,
3386 	.family		= AF_TIPC,
3387 	.create		= tipc_sk_create
3388 };
3389 
3390 static struct proto tipc_proto = {
3391 	.name		= "TIPC",
3392 	.owner		= THIS_MODULE,
3393 	.obj_size	= sizeof(struct tipc_sock),
3394 	.sysctl_rmem	= sysctl_tipc_rmem
3395 };
3396 
3397 /**
3398  * tipc_socket_init - initialize TIPC socket interface
3399  *
3400  * Returns 0 on success, errno otherwise
3401  */
3402 int tipc_socket_init(void)
3403 {
3404 	int res;
3405 
3406 	res = proto_register(&tipc_proto, 1);
3407 	if (res) {
3408 		pr_err("Failed to register TIPC protocol type\n");
3409 		goto out;
3410 	}
3411 
3412 	res = sock_register(&tipc_family_ops);
3413 	if (res) {
3414 		pr_err("Failed to register TIPC socket type\n");
3415 		proto_unregister(&tipc_proto);
3416 		goto out;
3417 	}
3418  out:
3419 	return res;
3420 }
3421 
3422 /**
3423  * tipc_socket_stop - stop TIPC socket interface
3424  */
3425 void tipc_socket_stop(void)
3426 {
3427 	sock_unregister(tipc_family_ops.family);
3428 	proto_unregister(&tipc_proto);
3429 }
3430 
3431 /* Caller should hold socket lock for the passed tipc socket. */
3432 static int __tipc_nl_add_sk_con(struct sk_buff *skb, struct tipc_sock *tsk)
3433 {
3434 	u32 peer_node;
3435 	u32 peer_port;
3436 	struct nlattr *nest;
3437 
3438 	peer_node = tsk_peer_node(tsk);
3439 	peer_port = tsk_peer_port(tsk);
3440 
3441 	nest = nla_nest_start_noflag(skb, TIPC_NLA_SOCK_CON);
3442 	if (!nest)
3443 		return -EMSGSIZE;
3444 
3445 	if (nla_put_u32(skb, TIPC_NLA_CON_NODE, peer_node))
3446 		goto msg_full;
3447 	if (nla_put_u32(skb, TIPC_NLA_CON_SOCK, peer_port))
3448 		goto msg_full;
3449 
3450 	if (tsk->conn_type != 0) {
3451 		if (nla_put_flag(skb, TIPC_NLA_CON_FLAG))
3452 			goto msg_full;
3453 		if (nla_put_u32(skb, TIPC_NLA_CON_TYPE, tsk->conn_type))
3454 			goto msg_full;
3455 		if (nla_put_u32(skb, TIPC_NLA_CON_INST, tsk->conn_instance))
3456 			goto msg_full;
3457 	}
3458 	nla_nest_end(skb, nest);
3459 
3460 	return 0;
3461 
3462 msg_full:
3463 	nla_nest_cancel(skb, nest);
3464 
3465 	return -EMSGSIZE;
3466 }
3467 
3468 static int __tipc_nl_add_sk_info(struct sk_buff *skb, struct tipc_sock
3469 			  *tsk)
3470 {
3471 	struct net *net = sock_net(skb->sk);
3472 	struct sock *sk = &tsk->sk;
3473 
3474 	if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->portid) ||
3475 	    nla_put_u32(skb, TIPC_NLA_SOCK_ADDR, tipc_own_addr(net)))
3476 		return -EMSGSIZE;
3477 
3478 	if (tipc_sk_connected(sk)) {
3479 		if (__tipc_nl_add_sk_con(skb, tsk))
3480 			return -EMSGSIZE;
3481 	} else if (!list_empty(&tsk->publications)) {
3482 		if (nla_put_flag(skb, TIPC_NLA_SOCK_HAS_PUBL))
3483 			return -EMSGSIZE;
3484 	}
3485 	return 0;
3486 }
3487 
3488 /* Caller should hold socket lock for the passed tipc socket. */
3489 static int __tipc_nl_add_sk(struct sk_buff *skb, struct netlink_callback *cb,
3490 			    struct tipc_sock *tsk)
3491 {
3492 	struct nlattr *attrs;
3493 	void *hdr;
3494 
3495 	hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
3496 			  &tipc_genl_family, NLM_F_MULTI, TIPC_NL_SOCK_GET);
3497 	if (!hdr)
3498 		goto msg_cancel;
3499 
3500 	attrs = nla_nest_start_noflag(skb, TIPC_NLA_SOCK);
3501 	if (!attrs)
3502 		goto genlmsg_cancel;
3503 
3504 	if (__tipc_nl_add_sk_info(skb, tsk))
3505 		goto attr_msg_cancel;
3506 
3507 	nla_nest_end(skb, attrs);
3508 	genlmsg_end(skb, hdr);
3509 
3510 	return 0;
3511 
3512 attr_msg_cancel:
3513 	nla_nest_cancel(skb, attrs);
3514 genlmsg_cancel:
3515 	genlmsg_cancel(skb, hdr);
3516 msg_cancel:
3517 	return -EMSGSIZE;
3518 }
3519 
3520 int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
3521 		    int (*skb_handler)(struct sk_buff *skb,
3522 				       struct netlink_callback *cb,
3523 				       struct tipc_sock *tsk))
3524 {
3525 	struct rhashtable_iter *iter = (void *)cb->args[4];
3526 	struct tipc_sock *tsk;
3527 	int err;
3528 
3529 	rhashtable_walk_start(iter);
3530 	while ((tsk = rhashtable_walk_next(iter)) != NULL) {
3531 		if (IS_ERR(tsk)) {
3532 			err = PTR_ERR(tsk);
3533 			if (err == -EAGAIN) {
3534 				err = 0;
3535 				continue;
3536 			}
3537 			break;
3538 		}
3539 
3540 		sock_hold(&tsk->sk);
3541 		rhashtable_walk_stop(iter);
3542 		lock_sock(&tsk->sk);
3543 		err = skb_handler(skb, cb, tsk);
3544 		if (err) {
3545 			release_sock(&tsk->sk);
3546 			sock_put(&tsk->sk);
3547 			goto out;
3548 		}
3549 		release_sock(&tsk->sk);
3550 		rhashtable_walk_start(iter);
3551 		sock_put(&tsk->sk);
3552 	}
3553 	rhashtable_walk_stop(iter);
3554 out:
3555 	return skb->len;
3556 }
3557 EXPORT_SYMBOL(tipc_nl_sk_walk);
3558 
3559 int tipc_dump_start(struct netlink_callback *cb)
3560 {
3561 	return __tipc_dump_start(cb, sock_net(cb->skb->sk));
3562 }
3563 EXPORT_SYMBOL(tipc_dump_start);
3564 
3565 int __tipc_dump_start(struct netlink_callback *cb, struct net *net)
3566 {
3567 	/* tipc_nl_name_table_dump() uses cb->args[0...3]. */
3568 	struct rhashtable_iter *iter = (void *)cb->args[4];
3569 	struct tipc_net *tn = tipc_net(net);
3570 
3571 	if (!iter) {
3572 		iter = kmalloc(sizeof(*iter), GFP_KERNEL);
3573 		if (!iter)
3574 			return -ENOMEM;
3575 
3576 		cb->args[4] = (long)iter;
3577 	}
3578 
3579 	rhashtable_walk_enter(&tn->sk_rht, iter);
3580 	return 0;
3581 }
3582 
3583 int tipc_dump_done(struct netlink_callback *cb)
3584 {
3585 	struct rhashtable_iter *hti = (void *)cb->args[4];
3586 
3587 	rhashtable_walk_exit(hti);
3588 	kfree(hti);
3589 	return 0;
3590 }
3591 EXPORT_SYMBOL(tipc_dump_done);
3592 
3593 int tipc_sk_fill_sock_diag(struct sk_buff *skb, struct netlink_callback *cb,
3594 			   struct tipc_sock *tsk, u32 sk_filter_state,
3595 			   u64 (*tipc_diag_gen_cookie)(struct sock *sk))
3596 {
3597 	struct sock *sk = &tsk->sk;
3598 	struct nlattr *attrs;
3599 	struct nlattr *stat;
3600 
3601 	/*filter response w.r.t sk_state*/
3602 	if (!(sk_filter_state & (1 << sk->sk_state)))
3603 		return 0;
3604 
3605 	attrs = nla_nest_start_noflag(skb, TIPC_NLA_SOCK);
3606 	if (!attrs)
3607 		goto msg_cancel;
3608 
3609 	if (__tipc_nl_add_sk_info(skb, tsk))
3610 		goto attr_msg_cancel;
3611 
3612 	if (nla_put_u32(skb, TIPC_NLA_SOCK_TYPE, (u32)sk->sk_type) ||
3613 	    nla_put_u32(skb, TIPC_NLA_SOCK_TIPC_STATE, (u32)sk->sk_state) ||
3614 	    nla_put_u32(skb, TIPC_NLA_SOCK_INO, sock_i_ino(sk)) ||
3615 	    nla_put_u32(skb, TIPC_NLA_SOCK_UID,
3616 			from_kuid_munged(sk_user_ns(NETLINK_CB(cb->skb).sk),
3617 					 sock_i_uid(sk))) ||
3618 	    nla_put_u64_64bit(skb, TIPC_NLA_SOCK_COOKIE,
3619 			      tipc_diag_gen_cookie(sk),
3620 			      TIPC_NLA_SOCK_PAD))
3621 		goto attr_msg_cancel;
3622 
3623 	stat = nla_nest_start_noflag(skb, TIPC_NLA_SOCK_STAT);
3624 	if (!stat)
3625 		goto attr_msg_cancel;
3626 
3627 	if (nla_put_u32(skb, TIPC_NLA_SOCK_STAT_RCVQ,
3628 			skb_queue_len(&sk->sk_receive_queue)) ||
3629 	    nla_put_u32(skb, TIPC_NLA_SOCK_STAT_SENDQ,
3630 			skb_queue_len(&sk->sk_write_queue)) ||
3631 	    nla_put_u32(skb, TIPC_NLA_SOCK_STAT_DROP,
3632 			atomic_read(&sk->sk_drops)))
3633 		goto stat_msg_cancel;
3634 
3635 	if (tsk->cong_link_cnt &&
3636 	    nla_put_flag(skb, TIPC_NLA_SOCK_STAT_LINK_CONG))
3637 		goto stat_msg_cancel;
3638 
3639 	if (tsk_conn_cong(tsk) &&
3640 	    nla_put_flag(skb, TIPC_NLA_SOCK_STAT_CONN_CONG))
3641 		goto stat_msg_cancel;
3642 
3643 	nla_nest_end(skb, stat);
3644 
3645 	if (tsk->group)
3646 		if (tipc_group_fill_sock_diag(tsk->group, skb))
3647 			goto stat_msg_cancel;
3648 
3649 	nla_nest_end(skb, attrs);
3650 
3651 	return 0;
3652 
3653 stat_msg_cancel:
3654 	nla_nest_cancel(skb, stat);
3655 attr_msg_cancel:
3656 	nla_nest_cancel(skb, attrs);
3657 msg_cancel:
3658 	return -EMSGSIZE;
3659 }
3660 EXPORT_SYMBOL(tipc_sk_fill_sock_diag);
3661 
3662 int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb)
3663 {
3664 	return tipc_nl_sk_walk(skb, cb, __tipc_nl_add_sk);
3665 }
3666 
3667 /* Caller should hold socket lock for the passed tipc socket. */
3668 static int __tipc_nl_add_sk_publ(struct sk_buff *skb,
3669 				 struct netlink_callback *cb,
3670 				 struct publication *publ)
3671 {
3672 	void *hdr;
3673 	struct nlattr *attrs;
3674 
3675 	hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
3676 			  &tipc_genl_family, NLM_F_MULTI, TIPC_NL_PUBL_GET);
3677 	if (!hdr)
3678 		goto msg_cancel;
3679 
3680 	attrs = nla_nest_start_noflag(skb, TIPC_NLA_PUBL);
3681 	if (!attrs)
3682 		goto genlmsg_cancel;
3683 
3684 	if (nla_put_u32(skb, TIPC_NLA_PUBL_KEY, publ->key))
3685 		goto attr_msg_cancel;
3686 	if (nla_put_u32(skb, TIPC_NLA_PUBL_TYPE, publ->type))
3687 		goto attr_msg_cancel;
3688 	if (nla_put_u32(skb, TIPC_NLA_PUBL_LOWER, publ->lower))
3689 		goto attr_msg_cancel;
3690 	if (nla_put_u32(skb, TIPC_NLA_PUBL_UPPER, publ->upper))
3691 		goto attr_msg_cancel;
3692 
3693 	nla_nest_end(skb, attrs);
3694 	genlmsg_end(skb, hdr);
3695 
3696 	return 0;
3697 
3698 attr_msg_cancel:
3699 	nla_nest_cancel(skb, attrs);
3700 genlmsg_cancel:
3701 	genlmsg_cancel(skb, hdr);
3702 msg_cancel:
3703 	return -EMSGSIZE;
3704 }
3705 
3706 /* Caller should hold socket lock for the passed tipc socket. */
3707 static int __tipc_nl_list_sk_publ(struct sk_buff *skb,
3708 				  struct netlink_callback *cb,
3709 				  struct tipc_sock *tsk, u32 *last_publ)
3710 {
3711 	int err;
3712 	struct publication *p;
3713 
3714 	if (*last_publ) {
3715 		list_for_each_entry(p, &tsk->publications, binding_sock) {
3716 			if (p->key == *last_publ)
3717 				break;
3718 		}
3719 		if (p->key != *last_publ) {
3720 			/* We never set seq or call nl_dump_check_consistent()
3721 			 * this means that setting prev_seq here will cause the
3722 			 * consistence check to fail in the netlink callback
3723 			 * handler. Resulting in the last NLMSG_DONE message
3724 			 * having the NLM_F_DUMP_INTR flag set.
3725 			 */
3726 			cb->prev_seq = 1;
3727 			*last_publ = 0;
3728 			return -EPIPE;
3729 		}
3730 	} else {
3731 		p = list_first_entry(&tsk->publications, struct publication,
3732 				     binding_sock);
3733 	}
3734 
3735 	list_for_each_entry_from(p, &tsk->publications, binding_sock) {
3736 		err = __tipc_nl_add_sk_publ(skb, cb, p);
3737 		if (err) {
3738 			*last_publ = p->key;
3739 			return err;
3740 		}
3741 	}
3742 	*last_publ = 0;
3743 
3744 	return 0;
3745 }
3746 
3747 int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb)
3748 {
3749 	int err;
3750 	u32 tsk_portid = cb->args[0];
3751 	u32 last_publ = cb->args[1];
3752 	u32 done = cb->args[2];
3753 	struct net *net = sock_net(skb->sk);
3754 	struct tipc_sock *tsk;
3755 
3756 	if (!tsk_portid) {
3757 		struct nlattr **attrs = genl_dumpit_info(cb)->attrs;
3758 		struct nlattr *sock[TIPC_NLA_SOCK_MAX + 1];
3759 
3760 		if (!attrs[TIPC_NLA_SOCK])
3761 			return -EINVAL;
3762 
3763 		err = nla_parse_nested_deprecated(sock, TIPC_NLA_SOCK_MAX,
3764 						  attrs[TIPC_NLA_SOCK],
3765 						  tipc_nl_sock_policy, NULL);
3766 		if (err)
3767 			return err;
3768 
3769 		if (!sock[TIPC_NLA_SOCK_REF])
3770 			return -EINVAL;
3771 
3772 		tsk_portid = nla_get_u32(sock[TIPC_NLA_SOCK_REF]);
3773 	}
3774 
3775 	if (done)
3776 		return 0;
3777 
3778 	tsk = tipc_sk_lookup(net, tsk_portid);
3779 	if (!tsk)
3780 		return -EINVAL;
3781 
3782 	lock_sock(&tsk->sk);
3783 	err = __tipc_nl_list_sk_publ(skb, cb, tsk, &last_publ);
3784 	if (!err)
3785 		done = 1;
3786 	release_sock(&tsk->sk);
3787 	sock_put(&tsk->sk);
3788 
3789 	cb->args[0] = tsk_portid;
3790 	cb->args[1] = last_publ;
3791 	cb->args[2] = done;
3792 
3793 	return skb->len;
3794 }
3795 
3796 /**
3797  * tipc_sk_filtering - check if a socket should be traced
3798  * @sk: the socket to be examined
3799  * @sysctl_tipc_sk_filter[]: the socket tuple for filtering,
3800  *  (portid, sock type, name type, name lower, name upper)
3801  *
3802  * Returns true if the socket meets the socket tuple data
3803  * (value 0 = 'any') or when there is no tuple set (all = 0),
3804  * otherwise false
3805  */
3806 bool tipc_sk_filtering(struct sock *sk)
3807 {
3808 	struct tipc_sock *tsk;
3809 	struct publication *p;
3810 	u32 _port, _sktype, _type, _lower, _upper;
3811 	u32 type = 0, lower = 0, upper = 0;
3812 
3813 	if (!sk)
3814 		return true;
3815 
3816 	tsk = tipc_sk(sk);
3817 
3818 	_port = sysctl_tipc_sk_filter[0];
3819 	_sktype = sysctl_tipc_sk_filter[1];
3820 	_type = sysctl_tipc_sk_filter[2];
3821 	_lower = sysctl_tipc_sk_filter[3];
3822 	_upper = sysctl_tipc_sk_filter[4];
3823 
3824 	if (!_port && !_sktype && !_type && !_lower && !_upper)
3825 		return true;
3826 
3827 	if (_port)
3828 		return (_port == tsk->portid);
3829 
3830 	if (_sktype && _sktype != sk->sk_type)
3831 		return false;
3832 
3833 	if (tsk->published) {
3834 		p = list_first_entry_or_null(&tsk->publications,
3835 					     struct publication, binding_sock);
3836 		if (p) {
3837 			type = p->type;
3838 			lower = p->lower;
3839 			upper = p->upper;
3840 		}
3841 	}
3842 
3843 	if (!tipc_sk_type_connectionless(sk)) {
3844 		type = tsk->conn_type;
3845 		lower = tsk->conn_instance;
3846 		upper = tsk->conn_instance;
3847 	}
3848 
3849 	if ((_type && _type != type) || (_lower && _lower != lower) ||
3850 	    (_upper && _upper != upper))
3851 		return false;
3852 
3853 	return true;
3854 }
3855 
3856 u32 tipc_sock_get_portid(struct sock *sk)
3857 {
3858 	return (sk) ? (tipc_sk(sk))->portid : 0;
3859 }
3860 
3861 /**
3862  * tipc_sk_overlimit1 - check if socket rx queue is about to be overloaded,
3863  *			both the rcv and backlog queues are considered
3864  * @sk: tipc sk to be checked
3865  * @skb: tipc msg to be checked
3866  *
3867  * Returns true if the socket rx queue allocation is > 90%, otherwise false
3868  */
3869 
3870 bool tipc_sk_overlimit1(struct sock *sk, struct sk_buff *skb)
3871 {
3872 	atomic_t *dcnt = &tipc_sk(sk)->dupl_rcvcnt;
3873 	unsigned int lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt);
3874 	unsigned int qsize = sk->sk_backlog.len + sk_rmem_alloc_get(sk);
3875 
3876 	return (qsize > lim * 90 / 100);
3877 }
3878 
3879 /**
3880  * tipc_sk_overlimit2 - check if socket rx queue is about to be overloaded,
3881  *			only the rcv queue is considered
3882  * @sk: tipc sk to be checked
3883  * @skb: tipc msg to be checked
3884  *
3885  * Returns true if the socket rx queue allocation is > 90%, otherwise false
3886  */
3887 
3888 bool tipc_sk_overlimit2(struct sock *sk, struct sk_buff *skb)
3889 {
3890 	unsigned int lim = rcvbuf_limit(sk, skb);
3891 	unsigned int qsize = sk_rmem_alloc_get(sk);
3892 
3893 	return (qsize > lim * 90 / 100);
3894 }
3895 
3896 /**
3897  * tipc_sk_dump - dump TIPC socket
3898  * @sk: tipc sk to be dumped
3899  * @dqueues: bitmask to decide if any socket queue to be dumped?
3900  *           - TIPC_DUMP_NONE: don't dump socket queues
3901  *           - TIPC_DUMP_SK_SNDQ: dump socket send queue
3902  *           - TIPC_DUMP_SK_RCVQ: dump socket rcv queue
3903  *           - TIPC_DUMP_SK_BKLGQ: dump socket backlog queue
3904  *           - TIPC_DUMP_ALL: dump all the socket queues above
3905  * @buf: returned buffer of dump data in format
3906  */
3907 int tipc_sk_dump(struct sock *sk, u16 dqueues, char *buf)
3908 {
3909 	int i = 0;
3910 	size_t sz = (dqueues) ? SK_LMAX : SK_LMIN;
3911 	struct tipc_sock *tsk;
3912 	struct publication *p;
3913 	bool tsk_connected;
3914 
3915 	if (!sk) {
3916 		i += scnprintf(buf, sz, "sk data: (null)\n");
3917 		return i;
3918 	}
3919 
3920 	tsk = tipc_sk(sk);
3921 	tsk_connected = !tipc_sk_type_connectionless(sk);
3922 
3923 	i += scnprintf(buf, sz, "sk data: %u", sk->sk_type);
3924 	i += scnprintf(buf + i, sz - i, " %d", sk->sk_state);
3925 	i += scnprintf(buf + i, sz - i, " %x", tsk_own_node(tsk));
3926 	i += scnprintf(buf + i, sz - i, " %u", tsk->portid);
3927 	i += scnprintf(buf + i, sz - i, " | %u", tsk_connected);
3928 	if (tsk_connected) {
3929 		i += scnprintf(buf + i, sz - i, " %x", tsk_peer_node(tsk));
3930 		i += scnprintf(buf + i, sz - i, " %u", tsk_peer_port(tsk));
3931 		i += scnprintf(buf + i, sz - i, " %u", tsk->conn_type);
3932 		i += scnprintf(buf + i, sz - i, " %u", tsk->conn_instance);
3933 	}
3934 	i += scnprintf(buf + i, sz - i, " | %u", tsk->published);
3935 	if (tsk->published) {
3936 		p = list_first_entry_or_null(&tsk->publications,
3937 					     struct publication, binding_sock);
3938 		i += scnprintf(buf + i, sz - i, " %u", (p) ? p->type : 0);
3939 		i += scnprintf(buf + i, sz - i, " %u", (p) ? p->lower : 0);
3940 		i += scnprintf(buf + i, sz - i, " %u", (p) ? p->upper : 0);
3941 	}
3942 	i += scnprintf(buf + i, sz - i, " | %u", tsk->snd_win);
3943 	i += scnprintf(buf + i, sz - i, " %u", tsk->rcv_win);
3944 	i += scnprintf(buf + i, sz - i, " %u", tsk->max_pkt);
3945 	i += scnprintf(buf + i, sz - i, " %x", tsk->peer_caps);
3946 	i += scnprintf(buf + i, sz - i, " %u", tsk->cong_link_cnt);
3947 	i += scnprintf(buf + i, sz - i, " %u", tsk->snt_unacked);
3948 	i += scnprintf(buf + i, sz - i, " %u", tsk->rcv_unacked);
3949 	i += scnprintf(buf + i, sz - i, " %u", atomic_read(&tsk->dupl_rcvcnt));
3950 	i += scnprintf(buf + i, sz - i, " %u", sk->sk_shutdown);
3951 	i += scnprintf(buf + i, sz - i, " | %d", sk_wmem_alloc_get(sk));
3952 	i += scnprintf(buf + i, sz - i, " %d", sk->sk_sndbuf);
3953 	i += scnprintf(buf + i, sz - i, " | %d", sk_rmem_alloc_get(sk));
3954 	i += scnprintf(buf + i, sz - i, " %d", sk->sk_rcvbuf);
3955 	i += scnprintf(buf + i, sz - i, " | %d\n", READ_ONCE(sk->sk_backlog.len));
3956 
3957 	if (dqueues & TIPC_DUMP_SK_SNDQ) {
3958 		i += scnprintf(buf + i, sz - i, "sk_write_queue: ");
3959 		i += tipc_list_dump(&sk->sk_write_queue, false, buf + i);
3960 	}
3961 
3962 	if (dqueues & TIPC_DUMP_SK_RCVQ) {
3963 		i += scnprintf(buf + i, sz - i, "sk_receive_queue: ");
3964 		i += tipc_list_dump(&sk->sk_receive_queue, false, buf + i);
3965 	}
3966 
3967 	if (dqueues & TIPC_DUMP_SK_BKLGQ) {
3968 		i += scnprintf(buf + i, sz - i, "sk_backlog:\n  head ");
3969 		i += tipc_skb_dump(sk->sk_backlog.head, false, buf + i);
3970 		if (sk->sk_backlog.tail != sk->sk_backlog.head) {
3971 			i += scnprintf(buf + i, sz - i, "  tail ");
3972 			i += tipc_skb_dump(sk->sk_backlog.tail, false,
3973 					   buf + i);
3974 		}
3975 	}
3976 
3977 	return i;
3978 }
3979