bcast.c 26.4 KB
Newer Older
P
Per Liden 已提交
1 2
/*
 * net/tipc/bcast.c: TIPC broadcast code
3
 *
4
 * Copyright (c) 2004-2006, 2014-2015, Ericsson AB
P
Per Liden 已提交
5
 * Copyright (c) 2004, Intel Corporation.
6
 * Copyright (c) 2005, 2010-2011, Wind River Systems
P
Per Liden 已提交
7 8
 * All rights reserved.
 *
P
Per Liden 已提交
9
 * Redistribution and use in source and binary forms, with or without
P
Per Liden 已提交
10 11
 * modification, are permitted provided that the following conditions are met:
 *
P
Per Liden 已提交
12 13 14 15 16 17 18 19
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the names of the copyright holders nor the names of its
 *    contributors may be used to endorse or promote products derived from
 *    this software without specific prior written permission.
P
Per Liden 已提交
20
 *
P
Per Liden 已提交
21 22 23 24 25 26 27 28 29 30 31 32 33 34
 * Alternatively, this software may be distributed under the terms of the
 * GNU General Public License ("GPL") version 2 as published by the Free
 * Software Foundation.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
P
Per Liden 已提交
35 36 37
 * POSSIBILITY OF SUCH DAMAGE.
 */

38 39
#include "socket.h"
#include "msg.h"
P
Per Liden 已提交
40
#include "bcast.h"
41
#include "name_distr.h"
42
#include "core.h"
P
Per Liden 已提交
43

44 45
#define	MAX_PKT_DEFAULT_MCAST	1500	/* bcast link max packet size (fixed) */
#define	BCLINK_WIN_DEFAULT	20	/* bcast link window size (default) */
P
Per Liden 已提交
46

47
const char tipc_bclink_name[] = "broadcast-link";
P
Per Liden 已提交
48

49 50 51
static void tipc_nmap_diff(struct tipc_node_map *nm_a,
			   struct tipc_node_map *nm_b,
			   struct tipc_node_map *nm_diff);
52 53
static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node);
static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node);
P
Per Liden 已提交
54

55
static void tipc_bclink_lock(struct net *net)
56
{
57 58 59
	struct tipc_net *tn = net_generic(net, tipc_net_id);

	spin_lock_bh(&tn->bclink->lock);
60 61
}

62
static void tipc_bclink_unlock(struct net *net)
63
{
64
	struct tipc_net *tn = net_generic(net, tipc_net_id);
65

66
	spin_unlock_bh(&tn->bclink->lock);
67 68
}

69 70 71 72 73 74 75
void tipc_bclink_input(struct net *net)
{
	struct tipc_net *tn = net_generic(net, tipc_net_id);

	tipc_sk_mcast_rcv(net, &tn->bclink->arrvq, &tn->bclink->inputq);
}

76 77 78 79 80
uint  tipc_bclink_get_mtu(void)
{
	return MAX_PKT_DEFAULT_MCAST;
}

S
Sam Ravnborg 已提交
81
static u32 bcbuf_acks(struct sk_buff *buf)
P
Per Liden 已提交
82
{
83
	return (u32)(unsigned long)TIPC_SKB_CB(buf)->handle;
P
Per Liden 已提交
84 85
}

S
Sam Ravnborg 已提交
86
static void bcbuf_set_acks(struct sk_buff *buf, u32 acks)
P
Per Liden 已提交
87
{
88
	TIPC_SKB_CB(buf)->handle = (void *)(unsigned long)acks;
P
Per Liden 已提交
89 90
}

S
Sam Ravnborg 已提交
91
static void bcbuf_decr_acks(struct sk_buff *buf)
P
Per Liden 已提交
92 93 94 95
{
	bcbuf_set_acks(buf, bcbuf_acks(buf) - 1);
}

96
void tipc_bclink_add_node(struct net *net, u32 addr)
97
{
98 99 100 101 102
	struct tipc_net *tn = net_generic(net, tipc_net_id);

	tipc_bclink_lock(net);
	tipc_nmap_add(&tn->bclink->bcast_nodes, addr);
	tipc_bclink_unlock(net);
103 104
}

105
void tipc_bclink_remove_node(struct net *net, u32 addr)
106
{
107 108 109 110
	struct tipc_net *tn = net_generic(net, tipc_net_id);

	tipc_bclink_lock(net);
	tipc_nmap_remove(&tn->bclink->bcast_nodes, addr);
111 112 113 114 115

	/* Last node? => reset backlog queue */
	if (!tn->bclink->bcast_nodes.count)
		tipc_link_purge_backlog(&tn->bclink->link);

116
	tipc_bclink_unlock(net);
117
}
P
Per Liden 已提交
118

119
static void bclink_set_last_sent(struct net *net)
120
{
121 122 123
	struct tipc_net *tn = net_generic(net, tipc_net_id);
	struct tipc_link *bcl = tn->bcl;

124
	bcl->silent_intv_cnt = mod(bcl->snd_nxt - 1);
125 126
}

127
u32 tipc_bclink_get_last_sent(struct net *net)
128
{
129 130
	struct tipc_net *tn = net_generic(net, tipc_net_id);

131
	return tn->bcl->silent_intv_cnt;
132 133
}

134
static void bclink_update_last_sent(struct tipc_node *node, u32 seqno)
P
Per Liden 已提交
135
{
136 137
	node->bclink.last_sent = less_eq(node->bclink.last_sent, seqno) ?
						seqno : node->bclink.last_sent;
P
Per Liden 已提交
138 139
}

140
/**
141 142
 * tipc_bclink_retransmit_to - get most recent node to request retransmission
 *
143
 * Called with bclink_lock locked
144
 */
145
struct tipc_node *tipc_bclink_retransmit_to(struct net *net)
146
{
147 148 149
	struct tipc_net *tn = net_generic(net, tipc_net_id);

	return tn->bclink->retransmit_to;
150 151
}

152
/**
P
Per Liden 已提交
153 154 155
 * bclink_retransmit_pkt - retransmit broadcast packets
 * @after: sequence number of last packet to *not* retransmit
 * @to: sequence number of last packet to retransmit
156
 *
157
 * Called with bclink_lock locked
P
Per Liden 已提交
158
 */
159
static void bclink_retransmit_pkt(struct tipc_net *tn, u32 after, u32 to)
P
Per Liden 已提交
160
{
161
	struct sk_buff *skb;
162
	struct tipc_link *bcl = tn->bcl;
P
Per Liden 已提交
163

J
Jon Paul Maloy 已提交
164
	skb_queue_walk(&bcl->transmq, skb) {
165 166
		if (more(buf_seqno(skb), after)) {
			tipc_link_retransmit(bcl, skb, mod(to - after));
167
			break;
168
		}
169
	}
P
Per Liden 已提交
170 171
}

172 173 174 175 176
/**
 * tipc_bclink_wakeup_users - wake up pending users
 *
 * Called with no locks taken
 */
177
void tipc_bclink_wakeup_users(struct net *net)
178
{
179
	struct tipc_net *tn = net_generic(net, tipc_net_id);
180

181
	tipc_sk_rcv(net, &tn->bclink->link.wakeupq);
182 183
}

184
/**
185
 * tipc_bclink_acknowledge - handle acknowledgement of broadcast packets
P
Per Liden 已提交
186 187
 * @n_ptr: node that sent acknowledgement info
 * @acked: broadcast sequence # that has been acknowledged
188
 *
189
 * Node is locked, bclink_lock unlocked.
P
Per Liden 已提交
190
 */
191
void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked)
P
Per Liden 已提交
192
{
193
	struct sk_buff *skb, *tmp;
P
Per Liden 已提交
194
	unsigned int released = 0;
195 196
	struct net *net = n_ptr->net;
	struct tipc_net *tn = net_generic(net, tipc_net_id);
P
Per Liden 已提交
197

198 199 200
	if (unlikely(!n_ptr->bclink.recv_permitted))
		return;

201
	tipc_bclink_lock(net);
202

203
	/* Bail out if tx queue is empty (no clean up is required) */
J
Jon Paul Maloy 已提交
204
	skb = skb_peek(&tn->bcl->transmq);
205
	if (!skb)
206 207 208 209 210 211 212 213 214
		goto exit;

	/* Determine which messages need to be acknowledged */
	if (acked == INVALID_LINK_SEQ) {
		/*
		 * Contact with specified node has been lost, so need to
		 * acknowledge sent messages only (if other nodes still exist)
		 * or both sent and unsent messages (otherwise)
		 */
215
		if (tn->bclink->bcast_nodes.count)
216
			acked = tn->bcl->silent_intv_cnt;
217
		else
218
			acked = tn->bcl->snd_nxt;
219 220 221 222 223
	} else {
		/*
		 * Bail out if specified sequence number does not correspond
		 * to a message that has been sent and not yet acknowledged
		 */
224
		if (less(acked, buf_seqno(skb)) ||
225
		    less(tn->bcl->silent_intv_cnt, acked) ||
226 227 228 229 230
		    less_eq(acked, n_ptr->bclink.acked))
			goto exit;
	}

	/* Skip over packets that node has previously acknowledged */
J
Jon Paul Maloy 已提交
231
	skb_queue_walk(&tn->bcl->transmq, skb) {
232 233 234
		if (more(buf_seqno(skb), n_ptr->bclink.acked))
			break;
	}
P
Per Liden 已提交
235 236

	/* Update packets that node is now acknowledging */
J
Jon Paul Maloy 已提交
237
	skb_queue_walk_from_safe(&tn->bcl->transmq, skb, tmp) {
238 239
		if (more(buf_seqno(skb), acked))
			break;
J
Jon Paul Maloy 已提交
240 241
		bcbuf_decr_acks(skb);
		bclink_set_last_sent(net);
242
		if (bcbuf_acks(skb) == 0) {
J
Jon Paul Maloy 已提交
243
			__skb_unlink(skb, &tn->bcl->transmq);
244
			kfree_skb(skb);
P
Per Liden 已提交
245 246 247 248 249 250
			released = 1;
		}
	}
	n_ptr->bclink.acked = acked;

	/* Try resolving broadcast link congestion, if necessary */
J
Jon Paul Maloy 已提交
251
	if (unlikely(skb_peek(&tn->bcl->backlogq))) {
252 253
		tipc_link_push_packets(tn->bcl);
		bclink_set_last_sent(net);
254
	}
255
	if (unlikely(released && !skb_queue_empty(&tn->bcl->wakeupq)))
256
		n_ptr->action_flags |= TIPC_WAKEUP_BCAST_USERS;
257
exit:
258
	tipc_bclink_unlock(net);
P
Per Liden 已提交
259 260
}

261
/**
262
 * tipc_bclink_update_link_state - update broadcast link state
263
 *
Y
Ying Xue 已提交
264
 * RCU and node lock set
P
Per Liden 已提交
265
 */
266
void tipc_bclink_update_link_state(struct tipc_node *n_ptr,
267
				   u32 last_sent)
P
Per Liden 已提交
268
{
269
	struct sk_buff *buf;
270
	struct net *net = n_ptr->net;
271
	struct tipc_net *tn = net_generic(net, tipc_net_id);
P
Per Liden 已提交
272

273 274 275
	/* Ignore "stale" link state info */
	if (less_eq(last_sent, n_ptr->bclink.last_in))
		return;
P
Per Liden 已提交
276

277 278 279 280 281 282 283 284 285 286 287 288 289 290 291
	/* Update link synchronization state; quit if in sync */
	bclink_update_last_sent(n_ptr, last_sent);

	if (n_ptr->bclink.last_sent == n_ptr->bclink.last_in)
		return;

	/* Update out-of-sync state; quit if loss is still unconfirmed */
	if ((++n_ptr->bclink.oos_state) == 1) {
		if (n_ptr->bclink.deferred_size < (TIPC_MIN_LINK_WIN / 2))
			return;
		n_ptr->bclink.oos_state++;
	}

	/* Don't NACK if one has been recently sent (or seen) */
	if (n_ptr->bclink.oos_state & 0x1)
P
Per Liden 已提交
292 293
		return;

294
	/* Send NACK */
295
	buf = tipc_buf_acquire(INT_H_SIZE);
P
Per Liden 已提交
296
	if (buf) {
297
		struct tipc_msg *msg = buf_msg(buf);
J
Jon Paul Maloy 已提交
298
		struct sk_buff *skb = skb_peek(&n_ptr->bclink.deferdq);
299
		u32 to = skb ? buf_seqno(skb) - 1 : n_ptr->bclink.last_sent;
300

301
		tipc_msg_init(tn->own_addr, msg, BCAST_PROTOCOL, STATE_MSG,
302
			      INT_H_SIZE, n_ptr->addr);
303
		msg_set_non_seq(msg, 1);
304
		msg_set_mc_netid(msg, tn->net_id);
305 306
		msg_set_bcast_ack(msg, n_ptr->bclink.last_in);
		msg_set_bcgap_after(msg, n_ptr->bclink.last_in);
307
		msg_set_bcgap_to(msg, to);
P
Per Liden 已提交
308

309
		tipc_bclink_lock(net);
310
		tipc_bearer_send(net, MAX_BEARERS, buf, NULL);
311 312
		tn->bcl->stats.sent_nacks++;
		tipc_bclink_unlock(net);
313
		kfree_skb(buf);
P
Per Liden 已提交
314

315
		n_ptr->bclink.oos_state++;
P
Per Liden 已提交
316 317 318
	}
}

319
/**
320
 * bclink_peek_nack - monitor retransmission requests sent by other nodes
P
Per Liden 已提交
321
 *
322 323
 * Delay any upcoming NACK by this node if another node has already
 * requested the first message this node is going to ask for.
P
Per Liden 已提交
324
 */
325
static void bclink_peek_nack(struct net *net, struct tipc_msg *msg)
P
Per Liden 已提交
326
{
327
	struct tipc_node *n_ptr = tipc_node_find(net, msg_destnode(msg));
P
Per Liden 已提交
328

329
	if (unlikely(!n_ptr))
P
Per Liden 已提交
330
		return;
331

332
	tipc_node_lock(n_ptr);
333
	if (n_ptr->bclink.recv_permitted &&
334 335 336
	    (n_ptr->bclink.last_in != n_ptr->bclink.last_sent) &&
	    (n_ptr->bclink.last_in == msg_bcgap_after(msg)))
		n_ptr->bclink.oos_state = 2;
337
	tipc_node_unlock(n_ptr);
338
	tipc_node_put(n_ptr);
P
Per Liden 已提交
339 340
}

341
/* tipc_bclink_xmit - deliver buffer chain to all nodes in cluster
342
 *                    and to identified node local sockets
343
 * @net: the applicable net namespace
344
 * @list: chain of buffers containing message
345 346 347
 * Consumes the buffer chain, except when returning -ELINKCONG
 * Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE
 */
348
int tipc_bclink_xmit(struct net *net, struct sk_buff_head *list)
349
{
350 351 352
	struct tipc_net *tn = net_generic(net, tipc_net_id);
	struct tipc_link *bcl = tn->bcl;
	struct tipc_bclink *bclink = tn->bclink;
353 354
	int rc = 0;
	int bc = 0;
355
	struct sk_buff *skb;
356 357
	struct sk_buff_head arrvq;
	struct sk_buff_head inputq;
358 359

	/* Prepare clone of message for local node */
360 361 362
	skb = tipc_msg_reassemble(list);
	if (unlikely(!skb)) {
		__skb_queue_purge(list);
363 364
		return -EHOSTUNREACH;
	}
365
	/* Broadcast to all nodes */
366
	if (likely(bclink)) {
367
		tipc_bclink_lock(net);
368
		if (likely(bclink->bcast_nodes.count)) {
369
			rc = __tipc_link_xmit(net, bcl, list);
370
			if (likely(!rc)) {
J
Jon Paul Maloy 已提交
371
				u32 len = skb_queue_len(&bcl->transmq);
372

373
				bclink_set_last_sent(net);
374
				bcl->stats.queue_sz_counts++;
375
				bcl->stats.accu_queue_sz += len;
376 377 378
			}
			bc = 1;
		}
379
		tipc_bclink_unlock(net);
380 381 382
	}

	if (unlikely(!bc))
383
		__skb_queue_purge(list);
384

385
	if (unlikely(rc)) {
386
		kfree_skb(skb);
387 388 389 390 391 392 393
		return rc;
	}
	/* Deliver message clone */
	__skb_queue_head_init(&arrvq);
	skb_queue_head_init(&inputq);
	__skb_queue_tail(&arrvq, skb);
	tipc_sk_mcast_rcv(net, &arrvq, &inputq);
394 395 396
	return rc;
}

397
/**
398 399
 * bclink_accept_pkt - accept an incoming, in-sequence broadcast packet
 *
400
 * Called with both sending node's lock and bclink_lock taken.
401 402 403
 */
static void bclink_accept_pkt(struct tipc_node *node, u32 seqno)
{
404 405
	struct tipc_net *tn = net_generic(node->net, tipc_net_id);

406 407 408
	bclink_update_last_sent(node, seqno);
	node->bclink.last_in = seqno;
	node->bclink.oos_state = 0;
409
	tn->bcl->stats.recv_info++;
410 411 412 413 414

	/*
	 * Unicast an ACK periodically, ensuring that
	 * all nodes in the cluster don't ACK at the same time
	 */
415
	if (((seqno - tn->own_addr) % TIPC_MIN_LINK_WIN) == 0) {
416
		tipc_link_proto_xmit(node_active_link(node, node->addr),
417
				     STATE_MSG, 0, 0, 0, 0);
418
		tn->bcl->stats.sent_acks++;
419 420 421
	}
}

422
/**
423
 * tipc_bclink_rcv - receive a broadcast packet, and deliver upwards
424
 *
Y
Ying Xue 已提交
425
 * RCU is locked, no other locks set
P
Per Liden 已提交
426
 */
427
void tipc_bclink_rcv(struct net *net, struct sk_buff *buf)
428
{
429
	struct tipc_net *tn = net_generic(net, tipc_net_id);
430
	struct tipc_link *bcl = tn->bcl;
P
Per Liden 已提交
431
	struct tipc_msg *msg = buf_msg(buf);
432
	struct tipc_node *node;
P
Per Liden 已提交
433 434
	u32 next_in;
	u32 seqno;
435
	int deferred = 0;
436 437
	int pos = 0;
	struct sk_buff *iskb;
438
	struct sk_buff_head *arrvq, *inputq;
P
Per Liden 已提交
439

440
	/* Screen out unwanted broadcast messages */
441
	if (msg_mc_netid(msg) != tn->net_id)
442 443
		goto exit;

444
	node = tipc_node_find(net, msg_prevnode(msg));
445 446 447 448
	if (unlikely(!node))
		goto exit;

	tipc_node_lock(node);
449
	if (unlikely(!node->bclink.recv_permitted))
450
		goto unlock;
P
Per Liden 已提交
451

452
	/* Handle broadcast protocol message */
P
Per Liden 已提交
453
	if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) {
454 455
		if (msg_type(msg) != STATE_MSG)
			goto unlock;
456
		if (msg_destnode(msg) == tn->own_addr) {
457
			tipc_bclink_acknowledge(node, msg_bcast_ack(msg));
458
			tipc_bclink_lock(net);
P
Per Liden 已提交
459
			bcl->stats.recv_nacks++;
460 461
			tn->bclink->retransmit_to = node;
			bclink_retransmit_pkt(tn, msg_bcgap_after(msg),
P
Per Liden 已提交
462
					      msg_bcgap_to(msg));
463
			tipc_bclink_unlock(net);
464
			tipc_node_unlock(node);
P
Per Liden 已提交
465
		} else {
466
			tipc_node_unlock(node);
467
			bclink_peek_nack(net, msg);
P
Per Liden 已提交
468
		}
469
		tipc_node_put(node);
470
		goto exit;
P
Per Liden 已提交
471 472
	}

473
	/* Handle in-sequence broadcast message */
P
Per Liden 已提交
474
	seqno = msg_seqno(msg);
475
	next_in = mod(node->bclink.last_in + 1);
476 477
	arrvq = &tn->bclink->arrvq;
	inputq = &tn->bclink->inputq;
P
Per Liden 已提交
478 479

	if (likely(seqno == next_in)) {
480
receive:
481
		/* Deliver message to destination */
P
Per Liden 已提交
482
		if (likely(msg_isdata(msg))) {
483
			tipc_bclink_lock(net);
484
			bclink_accept_pkt(node, seqno);
485 486 487 488
			spin_lock_bh(&inputq->lock);
			__skb_queue_tail(arrvq, buf);
			spin_unlock_bh(&inputq->lock);
			node->action_flags |= TIPC_BCAST_MSG_EVT;
489
			tipc_bclink_unlock(net);
490
			tipc_node_unlock(node);
P
Per Liden 已提交
491
		} else if (msg_user(msg) == MSG_BUNDLER) {
492
			tipc_bclink_lock(net);
493
			bclink_accept_pkt(node, seqno);
P
Per Liden 已提交
494 495
			bcl->stats.recv_bundles++;
			bcl->stats.recv_bundled += msg_msgcnt(msg);
496 497 498 499 500 501 502
			pos = 0;
			while (tipc_msg_extract(buf, &iskb, &pos)) {
				spin_lock_bh(&inputq->lock);
				__skb_queue_tail(arrvq, iskb);
				spin_unlock_bh(&inputq->lock);
			}
			node->action_flags |= TIPC_BCAST_MSG_EVT;
503
			tipc_bclink_unlock(net);
504
			tipc_node_unlock(node);
P
Per Liden 已提交
505
		} else if (msg_user(msg) == MSG_FRAGMENTER) {
506
			tipc_bclink_lock(net);
507
			bclink_accept_pkt(node, seqno);
508 509 510 511 512
			tipc_buf_append(&node->bclink.reasm_buf, &buf);
			if (unlikely(!buf && !node->bclink.reasm_buf)) {
				tipc_bclink_unlock(net);
				goto unlock;
			}
P
Per Liden 已提交
513
			bcl->stats.recv_fragments++;
514
			if (buf) {
P
Per Liden 已提交
515
				bcl->stats.recv_fragmented++;
516
				msg = buf_msg(buf);
517
				tipc_bclink_unlock(net);
E
Erik Hugne 已提交
518 519
				goto receive;
			}
520
			tipc_bclink_unlock(net);
521
			tipc_node_unlock(node);
P
Per Liden 已提交
522
		} else {
523
			tipc_bclink_lock(net);
524
			bclink_accept_pkt(node, seqno);
525
			tipc_bclink_unlock(net);
526
			tipc_node_unlock(node);
527
			kfree_skb(buf);
P
Per Liden 已提交
528
		}
529
		buf = NULL;
530 531

		/* Determine new synchronization state */
532
		tipc_node_lock(node);
533 534 535
		if (unlikely(!tipc_node_is_up(node)))
			goto unlock;

536
		if (node->bclink.last_in == node->bclink.last_sent)
537 538
			goto unlock;

J
Jon Paul Maloy 已提交
539
		if (skb_queue_empty(&node->bclink.deferdq)) {
540 541 542 543
			node->bclink.oos_state = 1;
			goto unlock;
		}

J
Jon Paul Maloy 已提交
544
		msg = buf_msg(skb_peek(&node->bclink.deferdq));
545 546 547 548 549 550
		seqno = msg_seqno(msg);
		next_in = mod(next_in + 1);
		if (seqno != next_in)
			goto unlock;

		/* Take in-sequence message from deferred queue & deliver it */
J
Jon Paul Maloy 已提交
551
		buf = __skb_dequeue(&node->bclink.deferdq);
552 553 554 555 556
		goto receive;
	}

	/* Handle out-of-sequence broadcast message */
	if (less(next_in, seqno)) {
J
Jon Paul Maloy 已提交
557
		deferred = tipc_link_defer_pkt(&node->bclink.deferdq,
558
					       buf);
559
		bclink_update_last_sent(node, seqno);
560
		buf = NULL;
561
	}
562

563
	tipc_bclink_lock(net);
564

565 566
	if (deferred)
		bcl->stats.deferred_recv++;
567 568
	else
		bcl->stats.duplicates++;
569

570
	tipc_bclink_unlock(net);
571

572
unlock:
573
	tipc_node_unlock(node);
574
	tipc_node_put(node);
575
exit:
576
	kfree_skb(buf);
P
Per Liden 已提交
577 578
}

579
u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr)
P
Per Liden 已提交
580
{
581
	return (n_ptr->bclink.recv_permitted &&
582
		(tipc_bclink_get_last_sent(n_ptr->net) != n_ptr->bclink.acked));
P
Per Liden 已提交
583 584 585 586
}


/**
587
 * tipc_bcbearer_send - send a packet through the broadcast pseudo-bearer
588
 *
589 590
 * Send packet over as many bearers as necessary to reach all nodes
 * that have joined the broadcast link.
591
 *
592 593
 * Returns 0 (packet sent successfully) under all circumstances,
 * since the broadcast link's pseudo-bearer never blocks
P
Per Liden 已提交
594
 */
595 596
static int tipc_bcbearer_send(struct net *net, struct sk_buff *buf,
			      struct tipc_bearer *unused1,
A
Adrian Bunk 已提交
597
			      struct tipc_media_addr *unused2)
P
Per Liden 已提交
598 599
{
	int bp_index;
600
	struct tipc_msg *msg = buf_msg(buf);
601
	struct tipc_net *tn = net_generic(net, tipc_net_id);
602 603
	struct tipc_bcbearer *bcbearer = tn->bcbearer;
	struct tipc_bclink *bclink = tn->bclink;
P
Per Liden 已提交
604

605
	/* Prepare broadcast link message for reliable transmission,
606 607 608 609
	 * if first time trying to send it;
	 * preparation is skipped for broadcast link protocol messages
	 * since they are sent in an unreliable manner and don't need it
	 */
P
Per Liden 已提交
610
	if (likely(!msg_non_seq(buf_msg(buf)))) {
611
		bcbuf_set_acks(buf, bclink->bcast_nodes.count);
612
		msg_set_non_seq(msg, 1);
613
		msg_set_mc_netid(msg, tn->net_id);
614
		tn->bcl->stats.sent_info++;
615
		if (WARN_ON(!bclink->bcast_nodes.count)) {
616 617 618
			dump_stack();
			return 0;
		}
P
Per Liden 已提交
619 620 621
	}

	/* Send buffer over bearers until all targets reached */
622
	bcbearer->remains = bclink->bcast_nodes;
P
Per Liden 已提交
623 624

	for (bp_index = 0; bp_index < MAX_BEARERS; bp_index++) {
625 626
		struct tipc_bearer *p = bcbearer->bpairs[bp_index].primary;
		struct tipc_bearer *s = bcbearer->bpairs[bp_index].secondary;
627 628
		struct tipc_bearer *bp[2] = {p, s};
		struct tipc_bearer *b = bp[msg_link_selector(msg)];
629
		struct sk_buff *tbuf;
P
Per Liden 已提交
630 631

		if (!p)
632
			break; /* No more bearers to try */
633 634
		if (!b)
			b = p;
635
		tipc_nmap_diff(&bcbearer->remains, &b->nodes,
636
			       &bcbearer->remains_new);
637
		if (bcbearer->remains_new.count == bcbearer->remains.count)
638
			continue; /* Nothing added by bearer pair */
P
Per Liden 已提交
639

640 641
		if (bp_index == 0) {
			/* Use original buffer for first bearer */
642
			tipc_bearer_send(net, b->identity, buf, &b->bcast_addr);
643 644
		} else {
			/* Avoid concurrent buffer access */
645
			tbuf = pskb_copy_for_clone(buf, GFP_ATOMIC);
646 647
			if (!tbuf)
				break;
648 649
			tipc_bearer_send(net, b->identity, tbuf,
					 &b->bcast_addr);
650 651
			kfree_skb(tbuf); /* Bearer keeps a clone */
		}
652
		if (bcbearer->remains_new.count == 0)
653
			break; /* All targets reached */
P
Per Liden 已提交
654

655
		bcbearer->remains = bcbearer->remains_new;
P
Per Liden 已提交
656
	}
657

658
	return 0;
P
Per Liden 已提交
659 660 661
}

/**
662
 * tipc_bcbearer_sort - create sets of bearer pairs used by broadcast bearer
P
Per Liden 已提交
663
 */
664 665
void tipc_bcbearer_sort(struct net *net, struct tipc_node_map *nm_ptr,
			u32 node, bool action)
P
Per Liden 已提交
666
{
667
	struct tipc_net *tn = net_generic(net, tipc_net_id);
668
	struct tipc_bcbearer *bcbearer = tn->bcbearer;
669 670
	struct tipc_bcbearer_pair *bp_temp = bcbearer->bpairs_temp;
	struct tipc_bcbearer_pair *bp_curr;
Y
Ying Xue 已提交
671
	struct tipc_bearer *b;
P
Per Liden 已提交
672 673 674
	int b_index;
	int pri;

675
	tipc_bclink_lock(net);
P
Per Liden 已提交
676

677 678 679 680 681
	if (action)
		tipc_nmap_add(nm_ptr, node);
	else
		tipc_nmap_remove(nm_ptr, node);

P
Per Liden 已提交
682 683 684
	/* Group bearers by priority (can assume max of two per priority) */
	memset(bp_temp, 0, sizeof(bcbearer->bpairs_temp));

Y
Ying Xue 已提交
685
	rcu_read_lock();
P
Per Liden 已提交
686
	for (b_index = 0; b_index < MAX_BEARERS; b_index++) {
687
		b = rcu_dereference_rtnl(tn->bearer_list[b_index]);
688
		if (!b || !b->nodes.count)
P
Per Liden 已提交
689 690 691 692 693 694 695
			continue;

		if (!bp_temp[b->priority].primary)
			bp_temp[b->priority].primary = b;
		else
			bp_temp[b->priority].secondary = b;
	}
Y
Ying Xue 已提交
696
	rcu_read_unlock();
P
Per Liden 已提交
697 698 699 700 701

	/* Create array of bearer pairs for broadcasting */
	bp_curr = bcbearer->bpairs;
	memset(bcbearer->bpairs, 0, sizeof(bcbearer->bpairs));

P
Per Liden 已提交
702
	for (pri = TIPC_MAX_LINK_PRI; pri >= 0; pri--) {
P
Per Liden 已提交
703 704 705 706 707 708 709

		if (!bp_temp[pri].primary)
			continue;

		bp_curr->primary = bp_temp[pri].primary;

		if (bp_temp[pri].secondary) {
710 711
			if (tipc_nmap_equal(&bp_temp[pri].primary->nodes,
					    &bp_temp[pri].secondary->nodes)) {
P
Per Liden 已提交
712 713 714 715 716 717 718 719 720 721
				bp_curr->secondary = bp_temp[pri].secondary;
			} else {
				bp_curr++;
				bp_curr->primary = bp_temp[pri].secondary;
			}
		}

		bp_curr++;
	}

722
	tipc_bclink_unlock(net);
P
Per Liden 已提交
723 724
}

725 726
static int __tipc_nl_add_bc_link_stat(struct sk_buff *skb,
				      struct tipc_stats *stats)
727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775
{
	int i;
	struct nlattr *nest;

	struct nla_map {
		__u32 key;
		__u32 val;
	};

	struct nla_map map[] = {
		{TIPC_NLA_STATS_RX_INFO, stats->recv_info},
		{TIPC_NLA_STATS_RX_FRAGMENTS, stats->recv_fragments},
		{TIPC_NLA_STATS_RX_FRAGMENTED, stats->recv_fragmented},
		{TIPC_NLA_STATS_RX_BUNDLES, stats->recv_bundles},
		{TIPC_NLA_STATS_RX_BUNDLED, stats->recv_bundled},
		{TIPC_NLA_STATS_TX_INFO, stats->sent_info},
		{TIPC_NLA_STATS_TX_FRAGMENTS, stats->sent_fragments},
		{TIPC_NLA_STATS_TX_FRAGMENTED, stats->sent_fragmented},
		{TIPC_NLA_STATS_TX_BUNDLES, stats->sent_bundles},
		{TIPC_NLA_STATS_TX_BUNDLED, stats->sent_bundled},
		{TIPC_NLA_STATS_RX_NACKS, stats->recv_nacks},
		{TIPC_NLA_STATS_RX_DEFERRED, stats->deferred_recv},
		{TIPC_NLA_STATS_TX_NACKS, stats->sent_nacks},
		{TIPC_NLA_STATS_TX_ACKS, stats->sent_acks},
		{TIPC_NLA_STATS_RETRANSMITTED, stats->retransmitted},
		{TIPC_NLA_STATS_DUPLICATES, stats->duplicates},
		{TIPC_NLA_STATS_LINK_CONGS, stats->link_congs},
		{TIPC_NLA_STATS_MAX_QUEUE, stats->max_queue_sz},
		{TIPC_NLA_STATS_AVG_QUEUE, stats->queue_sz_counts ?
			(stats->accu_queue_sz / stats->queue_sz_counts) : 0}
	};

	nest = nla_nest_start(skb, TIPC_NLA_LINK_STATS);
	if (!nest)
		return -EMSGSIZE;

	for (i = 0; i <  ARRAY_SIZE(map); i++)
		if (nla_put_u32(skb, map[i].key, map[i].val))
			goto msg_full;

	nla_nest_end(skb, nest);

	return 0;
msg_full:
	nla_nest_cancel(skb, nest);

	return -EMSGSIZE;
}

776
int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg)
777 778 779 780 781
{
	int err;
	void *hdr;
	struct nlattr *attrs;
	struct nlattr *prop;
782 783
	struct tipc_net *tn = net_generic(net, tipc_net_id);
	struct tipc_link *bcl = tn->bcl;
784 785 786 787

	if (!bcl)
		return 0;

788
	tipc_bclink_lock(net);
789

790
	hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family,
791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
			  NLM_F_MULTI, TIPC_NL_LINK_GET);
	if (!hdr)
		return -EMSGSIZE;

	attrs = nla_nest_start(msg->skb, TIPC_NLA_LINK);
	if (!attrs)
		goto msg_full;

	/* The broadcast link is always up */
	if (nla_put_flag(msg->skb, TIPC_NLA_LINK_UP))
		goto attr_msg_full;

	if (nla_put_flag(msg->skb, TIPC_NLA_LINK_BROADCAST))
		goto attr_msg_full;
	if (nla_put_string(msg->skb, TIPC_NLA_LINK_NAME, bcl->name))
		goto attr_msg_full;
807
	if (nla_put_u32(msg->skb, TIPC_NLA_LINK_RX, bcl->rcv_nxt))
808
		goto attr_msg_full;
809
	if (nla_put_u32(msg->skb, TIPC_NLA_LINK_TX, bcl->snd_nxt))
810 811 812 813 814
		goto attr_msg_full;

	prop = nla_nest_start(msg->skb, TIPC_NLA_LINK_PROP);
	if (!prop)
		goto attr_msg_full;
815
	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->window))
816 817 818 819 820 821 822
		goto prop_msg_full;
	nla_nest_end(msg->skb, prop);

	err = __tipc_nl_add_bc_link_stat(msg->skb, &bcl->stats);
	if (err)
		goto attr_msg_full;

823
	tipc_bclink_unlock(net);
824 825 826 827 828 829 830 831 832 833
	nla_nest_end(msg->skb, attrs);
	genlmsg_end(msg->skb, hdr);

	return 0;

prop_msg_full:
	nla_nest_cancel(msg->skb, prop);
attr_msg_full:
	nla_nest_cancel(msg->skb, attrs);
msg_full:
834
	tipc_bclink_unlock(net);
835 836 837 838
	genlmsg_cancel(msg->skb, hdr);

	return -EMSGSIZE;
}
P
Per Liden 已提交
839

840
int tipc_bclink_reset_stats(struct net *net)
P
Per Liden 已提交
841
{
842 843 844
	struct tipc_net *tn = net_generic(net, tipc_net_id);
	struct tipc_link *bcl = tn->bcl;

P
Per Liden 已提交
845 846 847
	if (!bcl)
		return -ENOPROTOOPT;

848
	tipc_bclink_lock(net);
P
Per Liden 已提交
849
	memset(&bcl->stats, 0, sizeof(bcl->stats));
850
	tipc_bclink_unlock(net);
851
	return 0;
P
Per Liden 已提交
852 853
}

854
int tipc_bclink_set_queue_limits(struct net *net, u32 limit)
P
Per Liden 已提交
855
{
856 857 858
	struct tipc_net *tn = net_generic(net, tipc_net_id);
	struct tipc_link *bcl = tn->bcl;

P
Per Liden 已提交
859 860 861 862 863
	if (!bcl)
		return -ENOPROTOOPT;
	if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN))
		return -EINVAL;

864
	tipc_bclink_lock(net);
865
	tipc_link_set_queue_limits(bcl, limit);
866
	tipc_bclink_unlock(net);
867
	return 0;
P
Per Liden 已提交
868 869
}

870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890
int tipc_nl_bc_link_set(struct net *net, struct nlattr *attrs[])
{
	int err;
	u32 win;
	struct nlattr *props[TIPC_NLA_PROP_MAX + 1];

	if (!attrs[TIPC_NLA_LINK_PROP])
		return -EINVAL;

	err = tipc_nl_parse_link_prop(attrs[TIPC_NLA_LINK_PROP], props);
	if (err)
		return err;

	if (!props[TIPC_NLA_PROP_WIN])
		return -EOPNOTSUPP;

	win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);

	return tipc_bclink_set_queue_limits(net, win);
}

891
int tipc_bclink_init(struct net *net)
P
Per Liden 已提交
892
{
893
	struct tipc_net *tn = net_generic(net, tipc_net_id);
894 895 896
	struct tipc_bcbearer *bcbearer;
	struct tipc_bclink *bclink;
	struct tipc_link *bcl;
897

898 899 900 901 902 903 904 905 906 907 908
	bcbearer = kzalloc(sizeof(*bcbearer), GFP_ATOMIC);
	if (!bcbearer)
		return -ENOMEM;

	bclink = kzalloc(sizeof(*bclink), GFP_ATOMIC);
	if (!bclink) {
		kfree(bcbearer);
		return -ENOMEM;
	}

	bcl = &bclink->link;
P
Per Liden 已提交
909
	bcbearer->bearer.media = &bcbearer->media;
910
	bcbearer->media.send_msg = tipc_bcbearer_send;
911
	sprintf(bcbearer->media.name, "tipc-broadcast");
P
Per Liden 已提交
912

913
	spin_lock_init(&bclink->lock);
J
Jon Paul Maloy 已提交
914 915 916
	__skb_queue_head_init(&bcl->transmq);
	__skb_queue_head_init(&bcl->backlogq);
	__skb_queue_head_init(&bcl->deferdq);
917
	skb_queue_head_init(&bcl->wakeupq);
918
	bcl->snd_nxt = 1;
I
Ingo Molnar 已提交
919
	spin_lock_init(&bclink->node.lock);
920 921
	__skb_queue_head_init(&bclink->arrvq);
	skb_queue_head_init(&bclink->inputq);
P
Per Liden 已提交
922
	bcl->owner = &bclink->node;
923
	bcl->owner->net = net;
924
	bcl->mtu = MAX_PKT_DEFAULT_MCAST;
925
	tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT);
926
	bcl->bearer_id = MAX_BEARERS;
927
	rcu_assign_pointer(tn->bearer_list[MAX_BEARERS], &bcbearer->bearer);
P
Per Liden 已提交
928
	bcl->state = WORKING_WORKING;
929 930
	bcl->pmsg = (struct tipc_msg *)&bcl->proto_msg;
	msg_set_prevnode(bcl->pmsg, tn->own_addr);
931
	strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME);
932 933 934
	tn->bcbearer = bcbearer;
	tn->bclink = bclink;
	tn->bcl = bcl;
935
	return 0;
P
Per Liden 已提交
936 937
}

938
void tipc_bclink_stop(struct net *net)
P
Per Liden 已提交
939
{
940 941
	struct tipc_net *tn = net_generic(net, tipc_net_id);

942 943 944
	tipc_bclink_lock(net);
	tipc_link_purge_queues(tn->bcl);
	tipc_bclink_unlock(net);
945

946
	RCU_INIT_POINTER(tn->bearer_list[BCBEARER], NULL);
947
	synchronize_net();
948 949
	kfree(tn->bcbearer);
	kfree(tn->bclink);
P
Per Liden 已提交
950 951
}

952 953 954
/**
 * tipc_nmap_add - add a node to a node map
 */
955
static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node)
956 957 958 959 960 961 962 963 964 965 966 967 968 969
{
	int n = tipc_node(node);
	int w = n / WSIZE;
	u32 mask = (1 << (n % WSIZE));

	if ((nm_ptr->map[w] & mask) == 0) {
		nm_ptr->count++;
		nm_ptr->map[w] |= mask;
	}
}

/**
 * tipc_nmap_remove - remove a node from a node map
 */
970
static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node)
971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987
{
	int n = tipc_node(node);
	int w = n / WSIZE;
	u32 mask = (1 << (n % WSIZE));

	if ((nm_ptr->map[w] & mask) != 0) {
		nm_ptr->map[w] &= ~mask;
		nm_ptr->count--;
	}
}

/**
 * tipc_nmap_diff - find differences between node maps
 * @nm_a: input node map A
 * @nm_b: input node map B
 * @nm_diff: output node map A-B (i.e. nodes of A that are not in B)
 */
988 989 990
static void tipc_nmap_diff(struct tipc_node_map *nm_a,
			   struct tipc_node_map *nm_b,
			   struct tipc_node_map *nm_diff)
991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008
{
	int stop = ARRAY_SIZE(nm_a->map);
	int w;
	int b;
	u32 map;

	memset(nm_diff, 0, sizeof(*nm_diff));
	for (w = 0; w < stop; w++) {
		map = nm_a->map[w] ^ (nm_a->map[w] & nm_b->map[w]);
		nm_diff->map[w] = map;
		if (map != 0) {
			for (b = 0 ; b < WSIZE; b++) {
				if (map & (1 << b))
					nm_diff->count++;
			}
		}
	}
}