summaryrefslogtreecommitdiffstats
path: root/fs/dlm/member.c
blob: 3fda3832cf6a62761e5aa8294c4c1e81919267eb (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
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
/******************************************************************************
*******************************************************************************
**
**  Copyright (C) 2005-2011 Red Hat, Inc.  All rights reserved.
**
**  This copyrighted material is made available to anyone wishing to use,
**  modify, copy, or redistribute it subject to the terms and conditions
**  of the GNU General Public License v.2.
**
*******************************************************************************
******************************************************************************/

#include "dlm_internal.h"
#include "lockspace.h"
#include "member.h"
#include "recoverd.h"
#include "recover.h"
#include "rcom.h"
#include "config.h"
#include "lowcomms.h"

int dlm_slots_version(struct dlm_header *h)
{
	if ((h->h_version & 0x0000FFFF) < DLM_HEADER_SLOTS)
		return 0;
	return 1;
}

void dlm_slot_save(struct dlm_ls *ls, struct dlm_rcom *rc,
		   struct dlm_member *memb)
{
	struct rcom_config *rf = (struct rcom_config *)rc->rc_buf;

	if (!dlm_slots_version(&rc->rc_header))
		return;

	memb->slot = le16_to_cpu(rf->rf_our_slot);
	memb->generation = le32_to_cpu(rf->rf_generation);
}

void dlm_slots_copy_out(struct dlm_ls *ls, struct dlm_rcom *rc)
{
	struct dlm_slot *slot;
	struct rcom_slot *ro;
	int i;

	ro = (struct rcom_slot *)(rc->rc_buf + sizeof(struct rcom_config));

	/* ls_slots array is sparse, but not rcom_slots */

	for (i = 0; i < ls->ls_slots_size; i++) {
		slot = &ls->ls_slots[i];
		if (!slot->nodeid)
			continue;
		ro->ro_nodeid = cpu_to_le32(slot->nodeid);
		ro->ro_slot = cpu_to_le16(slot->slot);
		ro++;
	}
}

#define SLOT_DEBUG_LINE 128

static void log_slots(struct dlm_ls *ls, uint32_t gen, int num_slots,
		      struct rcom_slot *ro0, struct dlm_slot *array,
		      int array_size)
{
	char line[SLOT_DEBUG_LINE];
	int len = SLOT_DEBUG_LINE - 1;
	int pos = 0;
	int ret, i;

	memset(line, 0, sizeof(line));

	if (array) {
		for (i = 0; i < array_size; i++) {
			if (!array[i].nodeid)
				continue;

			ret = snprintf(line + pos, len - pos, " %d:%d",
				       array[i].slot, array[i].nodeid);
			if (ret >= len - pos)
				break;
			pos += ret;
		}
	} else if (ro0) {
		for (i = 0; i < num_slots; i++) {
			ret = snprintf(line + pos, len - pos, " %d:%d",
				       ro0[i].ro_slot, ro0[i].ro_nodeid);
			if (ret >= len - pos)
				break;
			pos += ret;
		}
	}

	log_rinfo(ls, "generation %u slots %d%s", gen, num_slots, line);
}

int dlm_slots_copy_in(struct dlm_ls *ls)
{
	struct dlm_member *memb;
	struct dlm_rcom *rc = ls->ls_recover_buf;
	struct rcom_config *rf = (struct rcom_config *)rc->rc_buf;
	struct rcom_slot *ro0, *ro;
	int our_nodeid = dlm_our_nodeid();
	int i, num_slots;
	uint32_t gen;

	if (!dlm_slots_version(&rc->rc_header))
		return -1;

	gen = le32_to_cpu(rf->rf_generation);
	if (gen <= ls->ls_generation) {
		log_error(ls, "dlm_slots_copy_in gen %u old %u",
			  gen, ls->ls_generation);
	}
	ls->ls_generation = gen;

	num_slots = le16_to_cpu(rf->rf_num_slots);
	if (!num_slots)
		return -1;

	ro0 = (struct rcom_slot *)(rc->rc_buf + sizeof(struct rcom_config));

	for (i = 0, ro = ro0; i < num_slots; i++, ro++) {
		ro->ro_nodeid = le32_to_cpu(ro->ro_nodeid);
		ro->ro_slot = le16_to_cpu(ro->ro_slot);
	}

	log_slots(ls, gen, num_slots, ro0, NULL, 0);

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		for (i = 0, ro = ro0; i < num_slots; i++, ro++) {
			if (ro->ro_nodeid != memb->nodeid)
				continue;
			memb->slot = ro->ro_slot;
			memb->slot_prev = memb->slot;
			break;
		}

		if (memb->nodeid == our_nodeid) {
			if (ls->ls_slot && ls->ls_slot != memb->slot) {
				log_error(ls, "dlm_slots_copy_in our slot "
					  "changed %d %d", ls->ls_slot,
					  memb->slot);
				return -1;
			}

			if (!ls->ls_slot)
				ls->ls_slot = memb->slot;
		}

		if (!memb->slot) {
			log_error(ls, "dlm_slots_copy_in nodeid %d no slot",
				   memb->nodeid);
			return -1;
		}
	}

	return 0;
}

/* for any nodes that do not support slots, we will not have set memb->slot
   in wait_status_all(), so memb->slot will remain -1, and we will not
   assign slots or set ls_num_slots here */

int dlm_slots_assign(struct dlm_ls *ls, int *num_slots, int *slots_size,
		     struct dlm_slot **slots_out, uint32_t *gen_out)
{
	struct dlm_member *memb;
	struct dlm_slot *array;
	int our_nodeid = dlm_our_nodeid();
	int array_size, max_slots, i;
	int need = 0;
	int max = 0;
	int num = 0;
	uint32_t gen = 0;

	/* our own memb struct will have slot -1 gen 0 */

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (memb->nodeid == our_nodeid) {
			memb->slot = ls->ls_slot;
			memb->generation = ls->ls_generation;
			break;
		}
	}

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (memb->generation > gen)
			gen = memb->generation;

		/* node doesn't support slots */

		if (memb->slot == -1)
			return -1;

		/* node needs a slot assigned */

		if (!memb->slot)
			need++;

		/* node has a slot assigned */

		num++;

		if (!max || max < memb->slot)
			max = memb->slot;

		/* sanity check, once slot is assigned it shouldn't change */

		if (memb->slot_prev && memb->slot && memb->slot_prev != memb->slot) {
			log_error(ls, "nodeid %d slot changed %d %d",
				  memb->nodeid, memb->slot_prev, memb->slot);
			return -1;
		}
		memb->slot_prev = memb->slot;
	}

	array_size = max + need;
	array = kcalloc(array_size, sizeof(*array), GFP_NOFS);
	if (!array)
		return -ENOMEM;

	num = 0;

	/* fill in slots (offsets) that are used */

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (!memb->slot)
			continue;

		if (memb->slot > array_size) {
			log_error(ls, "invalid slot number %d", memb->slot);
			kfree(array);
			return -1;
		}

		array[memb->slot - 1].nodeid = memb->nodeid;
		array[memb->slot - 1].slot = memb->slot;
		num++;
	}

	/* assign new slots from unused offsets */

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (memb->slot)
			continue;

		for (i = 0; i < array_size; i++) {
			if (array[i].nodeid)
				continue;

			memb->slot = i + 1;
			memb->slot_prev = memb->slot;
			array[i].nodeid = memb->nodeid;
			array[i].slot = memb->slot;
			num++;

			if (!ls->ls_slot && memb->nodeid == our_nodeid)
				ls->ls_slot = memb->slot;
			break;
		}

		if (!memb->slot) {
			log_error(ls, "no free slot found");
			kfree(array);
			return -1;
		}
	}

	gen++;

	log_slots(ls, gen, num, NULL, array, array_size);

	max_slots = (dlm_config.ci_buffer_size - sizeof(struct dlm_rcom) -
		     sizeof(struct rcom_config)) / sizeof(struct rcom_slot);

	if (num > max_slots) {
		log_error(ls, "num_slots %d exceeds max_slots %d",
			  num, max_slots);
		kfree(array);
		return -1;
	}

	*gen_out = gen;
	*slots_out = array;
	*slots_size = array_size;
	*num_slots = num;
	return 0;
}

static void add_ordered_member(struct dlm_ls *ls, struct dlm_member *new)
{
	struct dlm_member *memb = NULL;
	struct list_head *tmp;
	struct list_head *newlist = &new->list;
	struct list_head *head = &ls->ls_nodes;

	list_for_each(tmp, head) {
		memb = list_entry(tmp, struct dlm_member, list);
		if (new->nodeid < memb->nodeid)
			break;
	}

	if (!memb)
		list_add_tail(newlist, head);
	else {
		/* FIXME: can use list macro here */
		newlist->prev = tmp->prev;
		newlist->next = tmp;
		tmp->prev->next = newlist;
		tmp->prev = newlist;
	}
}

static int dlm_add_member(struct dlm_ls *ls, struct dlm_config_node *node)
{
	struct dlm_member *memb;
	int error;

	memb = kzalloc(sizeof(*memb), GFP_NOFS);
	if (!memb)
		return -ENOMEM;

	error = dlm_lowcomms_connect_node(node->nodeid);
	if (error < 0) {
		kfree(memb);
		return error;
	}

	memb->nodeid = node->nodeid;
	memb->weight = node->weight;
	memb->comm_seq = node->comm_seq;
	add_ordered_member(ls, memb);
	ls->ls_num_nodes++;
	return 0;
}

static struct dlm_member *find_memb(struct list_head *head, int nodeid)
{
	struct dlm_member *memb;

	list_for_each_entry(memb, head, list) {
		if (memb->nodeid == nodeid)
			return memb;
	}
	return NULL;
}

int dlm_is_member(struct dlm_ls *ls, int nodeid)
{
	if (find_memb(&ls->ls_nodes, nodeid))
		return 1;
	return 0;
}

int dlm_is_removed(struct dlm_ls *ls, int nodeid)
{
	if (find_memb(&ls->ls_nodes_gone, nodeid))
		return 1;
	return 0;
}

static void clear_memb_list(struct list_head *head)
{
	struct dlm_member *memb;

	while (!list_empty(head)) {
		memb = list_entry(head->next, struct dlm_member, list);
		list_del(&memb->list);
		kfree(memb);
	}
}

void dlm_clear_members(struct dlm_ls *ls)
{
	clear_memb_list(&ls->ls_nodes);
	ls->ls_num_nodes = 0;
}

void dlm_clear_members_gone(struct dlm_ls *ls)
{
	clear_memb_list(&ls->ls_nodes_gone);
}

static void make_member_array(struct dlm_ls *ls)
{
	struct dlm_member *memb;
	int i, w, x = 0, total = 0, all_zero = 0, *array;

	kfree(ls->ls_node_array);
	ls->ls_node_array = NULL;

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (memb->weight)
			total += memb->weight;
	}

	/* all nodes revert to weight of 1 if all have weight 0 */

	if (!total) {
		total = ls->ls_num_nodes;
		all_zero = 1;
	}

	ls->ls_total_weight = total;
	array = kmalloc_array(total, sizeof(*array), GFP_NOFS);
	if (!array)
		return;

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (!all_zero && !memb->weight)
			continue;

		if (all_zero)
			w = 1;
		else
			w = memb->weight;

		DLM_ASSERT(x < total, printk("total %d x %d\n", total, x););

		for (i = 0; i < w; i++)
			array[x++] = memb->nodeid;
	}

	ls->ls_node_array = array;
}

/* send a status request to all members just to establish comms connections */

static int ping_members(struct dlm_ls *ls)
{
	struct dlm_member *memb;
	int error = 0;

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		error = dlm_recovery_stopped(ls);
		if (error)
			break;
		error = dlm_rcom_status(ls, memb->nodeid, 0);
		if (error)
			break;
	}
	if (error)
		log_rinfo(ls, "ping_members aborted %d last nodeid %d",
			  error, ls->ls_recover_nodeid);
	return error;
}

static void dlm_lsop_recover_prep(struct dlm_ls *ls)
{
	if (!ls->ls_ops || !ls->ls_ops->recover_prep)
		return;
	ls->ls_ops->recover_prep(ls->ls_ops_arg);
}

static void dlm_lsop_recover_slot(struct dlm_ls *ls, struct dlm_member *memb)
{
	struct dlm_slot slot;
	uint32_t seq;
	int error;

	if (!ls->ls_ops || !ls->ls_ops->recover_slot)
		return;

	/* if there is no comms connection with this node
	   or the present comms connection is newer
	   than the one when this member was added, then
	   we consider the node to have failed (versus
	   being removed due to dlm_release_lockspace) */

	error = dlm_comm_seq(memb->nodeid, &seq);

	if (!error && seq == memb->comm_seq)
		return;

	slot.nodeid = memb->nodeid;
	slot.slot = memb->slot;

	ls->ls_ops->recover_slot(ls->ls_ops_arg, &slot);
}

void dlm_lsop_recover_done(struct dlm_ls *ls)
{
	struct dlm_member *memb;
	struct dlm_slot *slots;
	int i, num;

	if (!ls->ls_ops || !ls->ls_ops->recover_done)
		return;

	num = ls->ls_num_nodes;
	slots = kcalloc(num, sizeof(*slots), GFP_KERNEL);
	if (!slots)
		return;

	i = 0;
	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (i == num) {
			log_error(ls, "dlm_lsop_recover_done bad num %d", num);
			goto out;
		}
		slots[i].nodeid = memb->nodeid;
		slots[i].slot = memb->slot;
		i++;
	}

	ls->ls_ops->recover_done(ls->ls_ops_arg, slots, num,
				 ls->ls_slot, ls->ls_generation);
 out:
	kfree(slots);
}

static struct dlm_config_node *find_config_node(struct dlm_recover *rv,
						int nodeid)
{
	int i;

	for (i = 0; i < rv->nodes_count; i++) {
		if (rv->nodes[i].nodeid == nodeid)
			return &rv->nodes[i];
	}
	return NULL;
}

int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out)
{
	struct dlm_member *memb, *safe;
	struct dlm_config_node *node;
	int i, error, neg = 0, low = -1;

	/* previously removed members that we've not finished removing need to
	   count as a negative change so the "neg" recovery steps will happen */

	list_for_each_entry(memb, &ls->ls_nodes_gone, list) {
		log_rinfo(ls, "prev removed member %d", memb->nodeid);
		neg++;
	}

	/* move departed members from ls_nodes to ls_nodes_gone */

	list_for_each_entry_safe(memb, safe, &ls->ls_nodes, list) {
		node = find_config_node(rv, memb->nodeid);
		if (node && !node->new)
			continue;

		if (!node) {
			log_rinfo(ls, "remove member %d", memb->nodeid);
		} else {
			/* removed and re-added */
			log_rinfo(ls, "remove member %d comm_seq %u %u",
				  memb->nodeid, memb->comm_seq, node->comm_seq);
		}

		neg++;
		list_move(&memb->list, &ls->ls_nodes_gone);
		ls->ls_num_nodes--;
		dlm_lsop_recover_slot(ls, memb);
	}

	/* add new members to ls_nodes */

	for (i = 0; i < rv->nodes_count; i++) {
		node = &rv->nodes[i];
		if (dlm_is_member(ls, node->nodeid))
			continue;
		dlm_add_member(ls, node);
		log_rinfo(ls, "add member %d", node->nodeid);
	}

	list_for_each_entry(memb, &ls->ls_nodes, list) {
		if (low == -1 || memb->nodeid < low)
			low = memb->nodeid;
	}
	ls->ls_low_nodeid = low;

	make_member_array(ls);
	*neg_out = neg;

	error = ping_members(ls);
	if (!error || error == -EPROTO) {
		/* new_lockspace() may be waiting to know if the config
		   is good or bad */
		ls->ls_members_result = error;
		complete(&ls->ls_members_done);
	}

	log_rinfo(ls, "dlm_recover_members %d nodes", ls->ls_num_nodes);
	return error;
}

/* Userspace guarantees that dlm_ls_stop() has completed on all nodes before
   dlm_ls_start() is called on any of them to start the new recovery. */

int dlm_ls_stop(struct dlm_ls *ls)
{
	int new;

	/*
	 * Prevent dlm_recv from being in the middle of something when we do
	 * the stop.  This includes ensuring dlm_recv isn't processing a
	 * recovery message (rcom), while dlm_recoverd is aborting and
	 * resetting things from an in-progress recovery.  i.e. we want
	 * dlm_recoverd to abort its recovery without worrying about dlm_recv
	 * processing an rcom at the same time.  Stopping dlm_recv also makes
	 * it easy for dlm_receive_message() to check locking stopped and add a
	 * message to the requestqueue without races.
	 */

	down_write(&ls->ls_recv_active);

	/*
	 * Abort any recovery that's in progress (see RECOVER_STOP,
	 * dlm_recovery_stopped()) and tell any other threads running in the
	 * dlm to quit any processing (see RUNNING, dlm_locking_stopped()).
	 */

	spin_lock(&ls->ls_recover_lock);
	set_bit(LSFL_RECOVER_STOP, &ls->ls_flags);
	new = test_and_clear_bit(LSFL_RUNNING, &ls->ls_flags);
	ls->ls_recover_seq++;
	spin_unlock(&ls->ls_recover_lock);

	/*
	 * Let dlm_recv run again, now any normal messages will be saved on the
	 * requestqueue for later.
	 */

	up_write(&ls->ls_recv_active);

	/*
	 * This in_recovery lock does two things:
	 * 1) Keeps this function from returning until all threads are out
	 *    of locking routines and locking is truly stopped.
	 * 2) Keeps any new requests from being processed until it's unlocked
	 *    when recovery is complete.
	 */

	if (new) {
		set_bit(LSFL_RECOVER_DOWN, &ls->ls_flags);
		wake_up_process(ls->ls_recoverd_task);
		wait_event(ls->ls_recover_lock_wait,
			   test_bit(LSFL_RECOVER_LOCK, &ls->ls_flags));
	}

	/*
	 * The recoverd suspend/resume makes sure that dlm_recoverd (if
	 * running) has noticed RECOVER_STOP above and quit processing the
	 * previous recovery.
	 */

	dlm_recoverd_suspend(ls);

	spin_lock(&ls->ls_recover_lock);
	kfree(ls->ls_slots);
	ls->ls_slots = NULL;
	ls->ls_num_slots = 0;
	ls->ls_slots_size = 0;
	ls->ls_recover_status = 0;
	spin_unlock(&ls->ls_recover_lock);

	dlm_recoverd_resume(ls);

	if (!ls->ls_recover_begin)
		ls->ls_recover_begin = jiffies;

	dlm_lsop_recover_prep(ls);
	return 0;
}

int dlm_ls_start(struct dlm_ls *ls)
{
	struct dlm_recover *rv, *rv_old;
	struct dlm_config_node *nodes;
	int error, count;

	rv = kzalloc(sizeof(*rv), GFP_NOFS);
	if (!rv)
		return -ENOMEM;

	error = dlm_config_nodes(ls->ls_name, &nodes, &count);
	if (error < 0)
		goto fail;

	spin_lock(&ls->ls_recover_lock);

	/* the lockspace needs to be stopped before it can be started */

	if (!dlm_locking_stopped(ls)) {
		spin_unlock(&ls->ls_recover_lock);
		log_error(ls, "start ignored: lockspace running");
		error = -EINVAL;
		goto fail;
	}

	rv->nodes = nodes;
	rv->nodes_count = count;
	rv->seq = ++ls->ls_recover_seq;
	rv_old = ls->ls_recover_args;
	ls->ls_recover_args = rv;
	spin_unlock(&ls->ls_recover_lock);

	if (rv_old) {
		log_error(ls, "unused recovery %llx %d",
			  (unsigned long long)rv_old->seq, rv_old->nodes_count);
		kfree(rv_old->nodes);
		kfree(rv_old);
	}

	set_bit(LSFL_RECOVER_WORK, &ls->ls_flags);
	wake_up_process(ls->ls_recoverd_task);
	return 0;

 fail:
	kfree(rv);
	kfree(nodes);
	return error;
}

OpenPOWER on IntegriCloud