summaryrefslogtreecommitdiffstats
path: root/net/tipc/monitor.c
blob: 87d4efedd09fd9768a925ac99d374c8ac6b78584 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
/*
 * net/tipc/monitor.c
 *
 * Copyright (c) 2016, Ericsson AB
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * 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.
 *
 * 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
 * POSSIBILITY OF SUCH DAMAGE.
 */

#include "core.h"
#include "addr.h"
#include "monitor.h"

#define MAX_MON_DOMAIN       64
#define MON_TIMEOUT          120000
#define MAX_PEER_DOWN_EVENTS 4

/* struct tipc_mon_domain: domain record to be transferred between peers
 * @len: actual size of domain record
 * @gen: current generation of sender's domain
 * @ack_gen: most recent generation of self's domain acked by peer
 * @member_cnt: number of domain member nodes described in this record
 * @up_map: bit map indicating which of the members the sender considers up
 * @members: identity of the domain members
 */
struct tipc_mon_domain {
	u16 len;
	u16 gen;
	u16 ack_gen;
	u16 member_cnt;
	u64 up_map;
	u32 members[MAX_MON_DOMAIN];
};

/* struct tipc_peer: state of a peer node and its domain
 * @addr: tipc node identity of peer
 * @head_map: shows which other nodes currently consider peer 'up'
 * @domain: most recent domain record from peer
 * @hash: position in hashed lookup list
 * @list: position in linked list, in circular ascending order by 'addr'
 * @applied: number of reported domain members applied on this monitor list
 * @is_up: peer is up as seen from this node
 * @is_head: peer is assigned domain head as seen from this node
 * @is_local: peer is in local domain and should be continuously monitored
 * @down_cnt: - numbers of other peers which have reported this on lost
 */
struct tipc_peer {
	u32 addr;
	struct tipc_mon_domain *domain;
	struct hlist_node hash;
	struct list_head list;
	u8 applied;
	u8 down_cnt;
	bool is_up;
	bool is_head;
	bool is_local;
};

struct tipc_monitor {
	struct hlist_head peers[NODE_HTABLE_SIZE];
	int peer_cnt;
	struct tipc_peer *self;
	rwlock_t lock;
	struct tipc_mon_domain cache;
	u16 list_gen;
	u16 dom_gen;
	struct net *net;
	struct timer_list timer;
	unsigned long timer_intv;
};

static struct tipc_monitor *tipc_monitor(struct net *net, int bearer_id)
{
	return tipc_net(net)->monitors[bearer_id];
}

const int tipc_max_domain_size = sizeof(struct tipc_mon_domain);

/* dom_rec_len(): actual length of domain record for transport
 */
static int dom_rec_len(struct tipc_mon_domain *dom, u16 mcnt)
{
	return ((void *)&dom->members - (void *)dom) + (mcnt * sizeof(u32));
}

/* dom_size() : calculate size of own domain based on number of peers
 */
static int dom_size(int peers)
{
	int i = 0;

	while ((i * i) < peers)
		i++;
	return i < MAX_MON_DOMAIN ? i : MAX_MON_DOMAIN;
}

static void map_set(u64 *up_map, int i, unsigned int v)
{
	*up_map &= ~(1 << i);
	*up_map |= (v << i);
}

static int map_get(u64 up_map, int i)
{
	return (up_map & (1 << i)) >> i;
}

static struct tipc_peer *peer_prev(struct tipc_peer *peer)
{
	return list_last_entry(&peer->list, struct tipc_peer, list);
}

static struct tipc_peer *peer_nxt(struct tipc_peer *peer)
{
	return list_first_entry(&peer->list, struct tipc_peer, list);
}

static struct tipc_peer *peer_head(struct tipc_peer *peer)
{
	while (!peer->is_head)
		peer = peer_prev(peer);
	return peer;
}

static struct tipc_peer *get_peer(struct tipc_monitor *mon, u32 addr)
{
	struct tipc_peer *peer;
	unsigned int thash = tipc_hashfn(addr);

	hlist_for_each_entry(peer, &mon->peers[thash], hash) {
		if (peer->addr == addr)
			return peer;
	}
	return NULL;
}

static struct tipc_peer *get_self(struct net *net, int bearer_id)
{
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);

	return mon->self;
}

static inline bool tipc_mon_is_active(struct net *net, struct tipc_monitor *mon)
{
	struct tipc_net *tn = tipc_net(net);

	return mon->peer_cnt > tn->mon_threshold;
}

/* mon_identify_lost_members() : - identify amd mark potentially lost members
 */
static void mon_identify_lost_members(struct tipc_peer *peer,
				      struct tipc_mon_domain *dom_bef,
				      int applied_bef)
{
	struct tipc_peer *member = peer;
	struct tipc_mon_domain *dom_aft = peer->domain;
	int applied_aft = peer->applied;
	int i;

	for (i = 0; i < applied_bef; i++) {
		member = peer_nxt(member);

		/* Do nothing if self or peer already see member as down */
		if (!member->is_up || !map_get(dom_bef->up_map, i))
			continue;

		/* Loss of local node must be detected by active probing */
		if (member->is_local)
			continue;

		/* Start probing if member was removed from applied domain */
		if (!applied_aft || (applied_aft < i)) {
			member->down_cnt = 1;
			continue;
		}

		/* Member loss is confirmed if it is still in applied domain */
		if (!map_get(dom_aft->up_map, i))
			member->down_cnt++;
	}
}

/* mon_apply_domain() : match a peer's domain record against monitor list
 */
static void mon_apply_domain(struct tipc_monitor *mon,
			     struct tipc_peer *peer)
{
	struct tipc_mon_domain *dom = peer->domain;
	struct tipc_peer *member;
	u32 addr;
	int i;

	if (!dom || !peer->is_up)
		return;

	/* Scan across domain members and match against monitor list */
	peer->applied = 0;
	member = peer_nxt(peer);
	for (i = 0; i < dom->member_cnt; i++) {
		addr = dom->members[i];
		if (addr != member->addr)
			return;
		peer->applied++;
		member = peer_nxt(member);
	}
}

/* mon_update_local_domain() : update after peer addition/removal/up/down
 */
static void mon_update_local_domain(struct tipc_monitor *mon)
{
	struct tipc_peer *self = mon->self;
	struct tipc_mon_domain *cache = &mon->cache;
	struct tipc_mon_domain *dom = self->domain;
	struct tipc_peer *peer = self;
	u64 prev_up_map = dom->up_map;
	u16 member_cnt, i;
	bool diff;

	/* Update local domain size based on current size of cluster */
	member_cnt = dom_size(mon->peer_cnt) - 1;
	self->applied = member_cnt;

	/* Update native and cached outgoing local domain records */
	dom->len = dom_rec_len(dom, member_cnt);
	diff = dom->member_cnt != member_cnt;
	dom->member_cnt = member_cnt;
	for (i = 0; i < member_cnt; i++) {
		peer = peer_nxt(peer);
		diff |= dom->members[i] != peer->addr;
		dom->members[i] = peer->addr;
		map_set(&dom->up_map, i, peer->is_up);
		cache->members[i] = htonl(peer->addr);
	}
	diff |= dom->up_map != prev_up_map;
	if (!diff)
		return;
	dom->gen = ++mon->dom_gen;
	cache->len = htons(dom->len);
	cache->gen = htons(dom->gen);
	cache->member_cnt = htons(member_cnt);
	cache->up_map = cpu_to_be64(dom->up_map);
	mon_apply_domain(mon, self);
}

/* mon_update_neighbors() : update preceding neighbors of added/removed peer
 */
static void mon_update_neighbors(struct tipc_monitor *mon,
				 struct tipc_peer *peer)
{
	int dz, i;

	dz = dom_size(mon->peer_cnt);
	for (i = 0; i < dz; i++) {
		mon_apply_domain(mon, peer);
		peer = peer_prev(peer);
	}
}

/* mon_assign_roles() : reassign peer roles after a network change
 * The monitor list is consistent at this stage; i.e., each peer is monitoring
 * a set of domain members as matched between domain record and the monitor list
 */
static void mon_assign_roles(struct tipc_monitor *mon, struct tipc_peer *head)
{
	struct tipc_peer *peer = peer_nxt(head);
	struct tipc_peer *self = mon->self;
	int i = 0;

	for (; peer != self; peer = peer_nxt(peer)) {
		peer->is_local = false;

		/* Update domain member */
		if (i++ < head->applied) {
			peer->is_head = false;
			if (head == self)
				peer->is_local = true;
			continue;
		}
		/* Assign next domain head */
		if (!peer->is_up)
			continue;
		if (peer->is_head)
			break;
		head = peer;
		head->is_head = true;
		i = 0;
	}
	mon->list_gen++;
}

void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id)
{
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
	struct tipc_peer *self = get_self(net, bearer_id);
	struct tipc_peer *peer, *prev, *head;

	write_lock_bh(&mon->lock);
	peer = get_peer(mon, addr);
	if (!peer)
		goto exit;
	prev = peer_prev(peer);
	list_del(&peer->list);
	hlist_del(&peer->hash);
	kfree(peer->domain);
	kfree(peer);
	mon->peer_cnt--;
	head = peer_head(prev);
	if (head == self)
		mon_update_local_domain(mon);
	mon_update_neighbors(mon, prev);

	/* Revert to full-mesh monitoring if we reach threshold */
	if (!tipc_mon_is_active(net, mon)) {
		list_for_each_entry(peer, &self->list, list) {
			kfree(peer->domain);
			peer->domain = NULL;
			peer->applied = 0;
		}
	}
	mon_assign_roles(mon, head);
exit:
	write_unlock_bh(&mon->lock);
}

static bool tipc_mon_add_peer(struct tipc_monitor *mon, u32 addr,
			      struct tipc_peer **peer)
{
	struct tipc_peer *self = mon->self;
	struct tipc_peer *cur, *prev, *p;

	p = kzalloc(sizeof(*p), GFP_ATOMIC);
	*peer = p;
	if (!p)
		return false;
	p->addr = addr;

	/* Add new peer to lookup list */
	INIT_LIST_HEAD(&p->list);
	hlist_add_head(&p->hash, &mon->peers[tipc_hashfn(addr)]);

	/* Sort new peer into iterator list, in ascending circular order */
	prev = self;
	list_for_each_entry(cur, &self->list, list) {
		if ((addr > prev->addr) && (addr < cur->addr))
			break;
		if (((addr < cur->addr) || (addr > prev->addr)) &&
		    (prev->addr > cur->addr))
			break;
		prev = cur;
	}
	list_add_tail(&p->list, &cur->list);
	mon->peer_cnt++;
	mon_update_neighbors(mon, p);
	return true;
}

void tipc_mon_peer_up(struct net *net, u32 addr, int bearer_id)
{
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
	struct tipc_peer *self = get_self(net, bearer_id);
	struct tipc_peer *peer, *head;

	write_lock_bh(&mon->lock);
	peer = get_peer(mon, addr);
	if (!peer && !tipc_mon_add_peer(mon, addr, &peer))
		goto exit;
	peer->is_up = true;
	head = peer_head(peer);
	if (head == self)
		mon_update_local_domain(mon);
	mon_assign_roles(mon, head);
exit:
	write_unlock_bh(&mon->lock);
}

void tipc_mon_peer_down(struct net *net, u32 addr, int bearer_id)
{
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
	struct tipc_peer *self = get_self(net, bearer_id);
	struct tipc_peer *peer, *head;
	struct tipc_mon_domain *dom;
	int applied;

	write_lock_bh(&mon->lock);
	peer = get_peer(mon, addr);
	if (!peer) {
		pr_warn("Mon: unknown link %x/%u DOWN\n", addr, bearer_id);
		goto exit;
	}
	applied = peer->applied;
	peer->applied = 0;
	dom = peer->domain;
	peer->domain = NULL;
	if (peer->is_head)
		mon_identify_lost_members(peer, dom, applied);
	kfree(dom);
	peer->is_up = false;
	peer->is_head = false;
	peer->is_local = false;
	peer->down_cnt = 0;
	head = peer_head(peer);
	if (head == self)
		mon_update_local_domain(mon);
	mon_assign_roles(mon, head);
exit:
	write_unlock_bh(&mon->lock);
}

/* tipc_mon_rcv - process monitor domain event message
 */
void tipc_mon_rcv(struct net *net, void *data, u16 dlen, u32 addr,
		  struct tipc_mon_state *state, int bearer_id)
{
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
	struct tipc_mon_domain *arrv_dom = data;
	struct tipc_mon_domain dom_bef;
	struct tipc_mon_domain *dom;
	struct tipc_peer *peer;
	u16 new_member_cnt = ntohs(arrv_dom->member_cnt);
	int new_dlen = dom_rec_len(arrv_dom, new_member_cnt);
	u16 new_gen = ntohs(arrv_dom->gen);
	u16 acked_gen = ntohs(arrv_dom->ack_gen);
	bool probing = state->probing;
	int i, applied_bef;

	state->probing = false;
	if (!dlen)
		return;

	/* Sanity check received domain record */
	if ((dlen < new_dlen) || ntohs(arrv_dom->len) != new_dlen) {
		pr_warn_ratelimited("Received illegal domain record\n");
		return;
	}

	/* Synch generation numbers with peer if link just came up */
	if (!state->synched) {
		state->peer_gen = new_gen - 1;
		state->acked_gen = acked_gen;
		state->synched = true;
	}

	if (more(acked_gen, state->acked_gen))
		state->acked_gen = acked_gen;

	/* Drop duplicate unless we are waiting for a probe response */
	if (!more(new_gen, state->peer_gen) && !probing)
		return;

	write_lock_bh(&mon->lock);
	peer = get_peer(mon, addr);
	if (!peer || !peer->is_up)
		goto exit;

	/* Peer is confirmed, stop any ongoing probing */
	peer->down_cnt = 0;

	/* Task is done for duplicate record */
	if (!more(new_gen, state->peer_gen))
		goto exit;

	state->peer_gen = new_gen;

	/* Cache current domain record for later use */
	dom_bef.member_cnt = 0;
	dom = peer->domain;
	if (dom)
		memcpy(&dom_bef, dom, dom->len);

	/* Transform and store received domain record */
	if (!dom || (dom->len < new_dlen)) {
		kfree(dom);
		dom = kmalloc(new_dlen, GFP_ATOMIC);
		peer->domain = dom;
		if (!dom)
			goto exit;
	}
	dom->len = new_dlen;
	dom->gen = new_gen;
	dom->member_cnt = new_member_cnt;
	dom->up_map = be64_to_cpu(arrv_dom->up_map);
	for (i = 0; i < new_member_cnt; i++)
		dom->members[i] = ntohl(arrv_dom->members[i]);

	/* Update peers affected by this domain record */
	applied_bef = peer->applied;
	mon_apply_domain(mon, peer);
	mon_identify_lost_members(peer, &dom_bef, applied_bef);
	mon_assign_roles(mon, peer_head(peer));
exit:
	write_unlock_bh(&mon->lock);
}

void tipc_mon_prep(struct net *net, void *data, int *dlen,
		   struct tipc_mon_state *state, int bearer_id)
{
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
	struct tipc_mon_domain *dom = data;
	u16 gen = mon->dom_gen;
	u16 len;

	if (!tipc_mon_is_active(net, mon))
		return;

	/* Send only a dummy record with ack if peer has acked our last sent */
	if (likely(state->acked_gen == gen)) {
		len = dom_rec_len(dom, 0);
		*dlen = len;
		dom->len = htons(len);
		dom->gen = htons(gen);
		dom->ack_gen = htons(state->peer_gen);
		dom->member_cnt = 0;
		return;
	}
	/* Send the full record */
	read_lock_bh(&mon->lock);
	len = ntohs(mon->cache.len);
	*dlen = len;
	memcpy(data, &mon->cache, len);
	read_unlock_bh(&mon->lock);
	dom->ack_gen = htons(state->peer_gen);
}

void tipc_mon_get_state(struct net *net, u32 addr,
			struct tipc_mon_state *state,
			int bearer_id)
{
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
	struct tipc_peer *peer;

	/* Used cached state if table has not changed */
	if (!state->probing &&
	    (state->list_gen == mon->list_gen) &&
	    (state->acked_gen == mon->dom_gen))
		return;

	read_lock_bh(&mon->lock);
	peer = get_peer(mon, addr);
	if (peer) {
		state->probing = state->acked_gen != mon->dom_gen;
		state->probing |= peer->down_cnt;
		state->reset |= peer->down_cnt >= MAX_PEER_DOWN_EVENTS;
		state->monitoring = peer->is_local;
		state->monitoring |= peer->is_head;
		state->list_gen = mon->list_gen;
	}
	read_unlock_bh(&mon->lock);
}

static void mon_timeout(unsigned long m)
{
	struct tipc_monitor *mon = (void *)m;
	struct tipc_peer *self;
	int best_member_cnt = dom_size(mon->peer_cnt) - 1;

	write_lock_bh(&mon->lock);
	self = mon->self;
	if (self && (best_member_cnt != self->applied)) {
		mon_update_local_domain(mon);
		mon_assign_roles(mon, self);
	}
	write_unlock_bh(&mon->lock);
	mod_timer(&mon->timer, jiffies + mon->timer_intv);
}

int tipc_mon_create(struct net *net, int bearer_id)
{
	struct tipc_net *tn = tipc_net(net);
	struct tipc_monitor *mon;
	struct tipc_peer *self;
	struct tipc_mon_domain *dom;

	if (tn->monitors[bearer_id])
		return 0;

	mon = kzalloc(sizeof(*mon), GFP_ATOMIC);
	self = kzalloc(sizeof(*self), GFP_ATOMIC);
	dom = kzalloc(sizeof(*dom), GFP_ATOMIC);
	if (!mon || !self || !dom) {
		kfree(mon);
		kfree(self);
		kfree(dom);
		return -ENOMEM;
	}
	tn->monitors[bearer_id] = mon;
	rwlock_init(&mon->lock);
	mon->net = net;
	mon->peer_cnt = 1;
	mon->self = self;
	self->domain = dom;
	self->addr = tipc_own_addr(net);
	self->is_up = true;
	self->is_head = true;
	INIT_LIST_HEAD(&self->list);
	setup_timer(&mon->timer, mon_timeout, (unsigned long)mon);
	mon->timer_intv = msecs_to_jiffies(MON_TIMEOUT + (tn->random & 0xffff));
	mod_timer(&mon->timer, jiffies + mon->timer_intv);
	return 0;
}

void tipc_mon_delete(struct net *net, int bearer_id)
{
	struct tipc_net *tn = tipc_net(net);
	struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
	struct tipc_peer *self = get_self(net, bearer_id);
	struct tipc_peer *peer, *tmp;

	write_lock_bh(&mon->lock);
	tn->monitors[bearer_id] = NULL;
	list_for_each_entry_safe(peer, tmp, &self->list, list) {
		list_del(&peer->list);
		hlist_del(&peer->hash);
		kfree(peer->domain);
		kfree(peer);
	}
	mon->self = NULL;
	write_unlock_bh(&mon->lock);
	del_timer_sync(&mon->timer);
	kfree(self->domain);
	kfree(self);
	kfree(mon);
}