dlm: improve error and debug messages
[deliverable/linux.git] / fs / dlm / lockspace.c
CommitLineData
e7fd4179
DT
1/******************************************************************************
2*******************************************************************************
3**
4** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
60f98d18 5** Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved.
e7fd4179
DT
6**
7** This copyrighted material is made available to anyone wishing to use,
8** modify, copy, or redistribute it subject to the terms and conditions
9** of the GNU General Public License v.2.
10**
11*******************************************************************************
12******************************************************************************/
13
14#include "dlm_internal.h"
15#include "lockspace.h"
16#include "member.h"
17#include "recoverd.h"
e7fd4179
DT
18#include "dir.h"
19#include "lowcomms.h"
20#include "config.h"
21#include "memory.h"
22#include "lock.h"
c56b39cd 23#include "recover.h"
2896ee37 24#include "requestqueue.h"
0f8e0d9a 25#include "user.h"
23e8e1aa 26#include "ast.h"
e7fd4179 27
e7fd4179 28static int ls_count;
90135925 29static struct mutex ls_lock;
e7fd4179
DT
30static struct list_head lslist;
31static spinlock_t lslist_lock;
32static struct task_struct * scand_task;
33
34
35static ssize_t dlm_control_store(struct dlm_ls *ls, const char *buf, size_t len)
36{
37 ssize_t ret = len;
38 int n = simple_strtol(buf, NULL, 0);
39
e2de7f56
PC
40 ls = dlm_find_lockspace_local(ls->ls_local_handle);
41 if (!ls)
42 return -EINVAL;
43
e7fd4179
DT
44 switch (n) {
45 case 0:
46 dlm_ls_stop(ls);
47 break;
48 case 1:
49 dlm_ls_start(ls);
50 break;
51 default:
52 ret = -EINVAL;
53 }
e2de7f56 54 dlm_put_lockspace(ls);
e7fd4179
DT
55 return ret;
56}
57
58static ssize_t dlm_event_store(struct dlm_ls *ls, const char *buf, size_t len)
59{
60 ls->ls_uevent_result = simple_strtol(buf, NULL, 0);
61 set_bit(LSFL_UEVENT_WAIT, &ls->ls_flags);
62 wake_up(&ls->ls_uevent_wait);
63 return len;
64}
65
66static ssize_t dlm_id_show(struct dlm_ls *ls, char *buf)
67{
a1d144c7 68 return snprintf(buf, PAGE_SIZE, "%u\n", ls->ls_global_id);
e7fd4179
DT
69}
70
71static ssize_t dlm_id_store(struct dlm_ls *ls, const char *buf, size_t len)
72{
73 ls->ls_global_id = simple_strtoul(buf, NULL, 0);
74 return len;
75}
76
c56b39cd
DT
77static ssize_t dlm_recover_status_show(struct dlm_ls *ls, char *buf)
78{
79 uint32_t status = dlm_recover_status(ls);
a1d144c7 80 return snprintf(buf, PAGE_SIZE, "%x\n", status);
c56b39cd
DT
81}
82
faa0f267
DT
83static ssize_t dlm_recover_nodeid_show(struct dlm_ls *ls, char *buf)
84{
a1d144c7 85 return snprintf(buf, PAGE_SIZE, "%d\n", ls->ls_recover_nodeid);
faa0f267
DT
86}
87
e7fd4179
DT
88struct dlm_attr {
89 struct attribute attr;
90 ssize_t (*show)(struct dlm_ls *, char *);
91 ssize_t (*store)(struct dlm_ls *, const char *, size_t);
92};
93
94static struct dlm_attr dlm_attr_control = {
95 .attr = {.name = "control", .mode = S_IWUSR},
96 .store = dlm_control_store
97};
98
99static struct dlm_attr dlm_attr_event = {
100 .attr = {.name = "event_done", .mode = S_IWUSR},
101 .store = dlm_event_store
102};
103
104static struct dlm_attr dlm_attr_id = {
105 .attr = {.name = "id", .mode = S_IRUGO | S_IWUSR},
106 .show = dlm_id_show,
107 .store = dlm_id_store
108};
109
c56b39cd
DT
110static struct dlm_attr dlm_attr_recover_status = {
111 .attr = {.name = "recover_status", .mode = S_IRUGO},
112 .show = dlm_recover_status_show
113};
114
faa0f267
DT
115static struct dlm_attr dlm_attr_recover_nodeid = {
116 .attr = {.name = "recover_nodeid", .mode = S_IRUGO},
117 .show = dlm_recover_nodeid_show
118};
119
e7fd4179
DT
120static struct attribute *dlm_attrs[] = {
121 &dlm_attr_control.attr,
122 &dlm_attr_event.attr,
123 &dlm_attr_id.attr,
c56b39cd 124 &dlm_attr_recover_status.attr,
faa0f267 125 &dlm_attr_recover_nodeid.attr,
e7fd4179
DT
126 NULL,
127};
128
129static ssize_t dlm_attr_show(struct kobject *kobj, struct attribute *attr,
130 char *buf)
131{
132 struct dlm_ls *ls = container_of(kobj, struct dlm_ls, ls_kobj);
133 struct dlm_attr *a = container_of(attr, struct dlm_attr, attr);
134 return a->show ? a->show(ls, buf) : 0;
135}
136
137static ssize_t dlm_attr_store(struct kobject *kobj, struct attribute *attr,
138 const char *buf, size_t len)
139{
140 struct dlm_ls *ls = container_of(kobj, struct dlm_ls, ls_kobj);
141 struct dlm_attr *a = container_of(attr, struct dlm_attr, attr);
142 return a->store ? a->store(ls, buf, len) : len;
143}
144
ba542e3b
PC
145static void lockspace_kobj_release(struct kobject *k)
146{
147 struct dlm_ls *ls = container_of(k, struct dlm_ls, ls_kobj);
148 kfree(ls);
149}
150
52cf25d0 151static const struct sysfs_ops dlm_attr_ops = {
e7fd4179
DT
152 .show = dlm_attr_show,
153 .store = dlm_attr_store,
154};
155
156static struct kobj_type dlm_ktype = {
157 .default_attrs = dlm_attrs,
158 .sysfs_ops = &dlm_attr_ops,
ba542e3b 159 .release = lockspace_kobj_release,
e7fd4179
DT
160};
161
d405936b 162static struct kset *dlm_kset;
e7fd4179 163
e7fd4179
DT
164static int do_uevent(struct dlm_ls *ls, int in)
165{
166 int error;
167
168 if (in)
169 kobject_uevent(&ls->ls_kobj, KOBJ_ONLINE);
170 else
171 kobject_uevent(&ls->ls_kobj, KOBJ_OFFLINE);
172
8b0e7b2c
DT
173 log_debug(ls, "%s the lockspace group...", in ? "joining" : "leaving");
174
175 /* dlm_controld will see the uevent, do the necessary group management
176 and then write to sysfs to wake us */
177
e7fd4179
DT
178 error = wait_event_interruptible(ls->ls_uevent_wait,
179 test_and_clear_bit(LSFL_UEVENT_WAIT, &ls->ls_flags));
8b0e7b2c
DT
180
181 log_debug(ls, "group event done %d %d", error, ls->ls_uevent_result);
182
e7fd4179
DT
183 if (error)
184 goto out;
185
186 error = ls->ls_uevent_result;
187 out:
8b0e7b2c
DT
188 if (error)
189 log_error(ls, "group %s failed %d %d", in ? "join" : "leave",
190 error, ls->ls_uevent_result);
e7fd4179
DT
191 return error;
192}
193
b4a5d4bc
SW
194static int dlm_uevent(struct kset *kset, struct kobject *kobj,
195 struct kobj_uevent_env *env)
196{
197 struct dlm_ls *ls = container_of(kobj, struct dlm_ls, ls_kobj);
198
199 add_uevent_var(env, "LOCKSPACE=%s", ls->ls_name);
200 return 0;
201}
202
203static struct kset_uevent_ops dlm_uevent_ops = {
204 .uevent = dlm_uevent,
205};
e7fd4179 206
30727174 207int __init dlm_lockspace_init(void)
e7fd4179 208{
e7fd4179 209 ls_count = 0;
90135925 210 mutex_init(&ls_lock);
e7fd4179
DT
211 INIT_LIST_HEAD(&lslist);
212 spin_lock_init(&lslist_lock);
213
b4a5d4bc 214 dlm_kset = kset_create_and_add("dlm", &dlm_uevent_ops, kernel_kobj);
d405936b 215 if (!dlm_kset) {
8e24eea7 216 printk(KERN_WARNING "%s: can not create kset\n", __func__);
d405936b
GKH
217 return -ENOMEM;
218 }
219 return 0;
e7fd4179
DT
220}
221
222void dlm_lockspace_exit(void)
223{
d405936b 224 kset_unregister(dlm_kset);
e7fd4179
DT
225}
226
c1dcf65f
DT
227static struct dlm_ls *find_ls_to_scan(void)
228{
229 struct dlm_ls *ls;
230
231 spin_lock(&lslist_lock);
232 list_for_each_entry(ls, &lslist, ls_list) {
233 if (time_after_eq(jiffies, ls->ls_scan_time +
234 dlm_config.ci_scan_secs * HZ)) {
235 spin_unlock(&lslist_lock);
236 return ls;
237 }
238 }
239 spin_unlock(&lslist_lock);
240 return NULL;
241}
242
e7fd4179
DT
243static int dlm_scand(void *data)
244{
245 struct dlm_ls *ls;
246
247 while (!kthread_should_stop()) {
c1dcf65f
DT
248 ls = find_ls_to_scan();
249 if (ls) {
85e86edf 250 if (dlm_lock_recovery_try(ls)) {
c1dcf65f 251 ls->ls_scan_time = jiffies;
85e86edf 252 dlm_scan_rsbs(ls);
3ae1acf9 253 dlm_scan_timeout(ls);
c6ff669b 254 dlm_scan_waiters(ls);
85e86edf 255 dlm_unlock_recovery(ls);
c1dcf65f
DT
256 } else {
257 ls->ls_scan_time += HZ;
85e86edf 258 }
c6ff669b 259 continue;
85e86edf 260 }
c6ff669b 261 schedule_timeout_interruptible(dlm_config.ci_scan_secs * HZ);
e7fd4179
DT
262 }
263 return 0;
264}
265
266static int dlm_scand_start(void)
267{
268 struct task_struct *p;
269 int error = 0;
270
271 p = kthread_run(dlm_scand, NULL, "dlm_scand");
272 if (IS_ERR(p))
273 error = PTR_ERR(p);
274 else
275 scand_task = p;
276 return error;
277}
278
279static void dlm_scand_stop(void)
280{
281 kthread_stop(scand_task);
282}
283
e7fd4179
DT
284struct dlm_ls *dlm_find_lockspace_global(uint32_t id)
285{
286 struct dlm_ls *ls;
287
288 spin_lock(&lslist_lock);
289
290 list_for_each_entry(ls, &lslist, ls_list) {
291 if (ls->ls_global_id == id) {
292 ls->ls_count++;
293 goto out;
294 }
295 }
296 ls = NULL;
297 out:
298 spin_unlock(&lslist_lock);
299 return ls;
300}
301
597d0cae 302struct dlm_ls *dlm_find_lockspace_local(dlm_lockspace_t *lockspace)
e7fd4179 303{
597d0cae 304 struct dlm_ls *ls;
e7fd4179
DT
305
306 spin_lock(&lslist_lock);
597d0cae
DT
307 list_for_each_entry(ls, &lslist, ls_list) {
308 if (ls->ls_local_handle == lockspace) {
309 ls->ls_count++;
310 goto out;
311 }
312 }
313 ls = NULL;
314 out:
315 spin_unlock(&lslist_lock);
316 return ls;
317}
318
319struct dlm_ls *dlm_find_lockspace_device(int minor)
320{
321 struct dlm_ls *ls;
322
323 spin_lock(&lslist_lock);
324 list_for_each_entry(ls, &lslist, ls_list) {
325 if (ls->ls_device.minor == minor) {
326 ls->ls_count++;
327 goto out;
328 }
329 }
330 ls = NULL;
331 out:
e7fd4179
DT
332 spin_unlock(&lslist_lock);
333 return ls;
334}
335
336void dlm_put_lockspace(struct dlm_ls *ls)
337{
338 spin_lock(&lslist_lock);
339 ls->ls_count--;
340 spin_unlock(&lslist_lock);
341}
342
343static void remove_lockspace(struct dlm_ls *ls)
344{
345 for (;;) {
346 spin_lock(&lslist_lock);
347 if (ls->ls_count == 0) {
0f8e0d9a 348 WARN_ON(ls->ls_create_count != 0);
e7fd4179
DT
349 list_del(&ls->ls_list);
350 spin_unlock(&lslist_lock);
351 return;
352 }
353 spin_unlock(&lslist_lock);
354 ssleep(1);
355 }
356}
357
358static int threads_start(void)
359{
360 int error;
361
e7fd4179
DT
362 error = dlm_scand_start();
363 if (error) {
364 log_print("cannot start dlm_scand thread %d", error);
23e8e1aa 365 goto fail;
e7fd4179
DT
366 }
367
368 /* Thread for sending/receiving messages for all lockspace's */
369 error = dlm_lowcomms_start();
370 if (error) {
371 log_print("cannot start dlm lowcomms %d", error);
372 goto scand_fail;
373 }
374
375 return 0;
376
377 scand_fail:
378 dlm_scand_stop();
e7fd4179
DT
379 fail:
380 return error;
381}
382
383static void threads_stop(void)
384{
385 dlm_scand_stop();
386 dlm_lowcomms_stop();
e7fd4179
DT
387}
388
60f98d18
DT
389static int new_lockspace(const char *name, const char *cluster,
390 uint32_t flags, int lvblen,
391 const struct dlm_lockspace_ops *ops, void *ops_arg,
392 int *ops_result, dlm_lockspace_t **lockspace)
e7fd4179
DT
393{
394 struct dlm_ls *ls;
0f8e0d9a 395 int i, size, error;
79d72b54 396 int do_unreg = 0;
60f98d18 397 int namelen = strlen(name);
e7fd4179
DT
398
399 if (namelen > DLM_LOCKSPACE_LEN)
400 return -EINVAL;
401
402 if (!lvblen || (lvblen % 8))
403 return -EINVAL;
404
405 if (!try_module_get(THIS_MODULE))
406 return -EINVAL;
407
dc68c7ed 408 if (!dlm_user_daemon_available()) {
60f98d18
DT
409 log_print("dlm user daemon not available");
410 error = -EUNATCH;
411 goto out;
412 }
413
414 if (ops && ops_result) {
415 if (!dlm_config.ci_recover_callbacks)
416 *ops_result = -EOPNOTSUPP;
417 else
418 *ops_result = 0;
419 }
420
421 if (dlm_config.ci_recover_callbacks && cluster &&
422 strncmp(cluster, dlm_config.ci_cluster_name, DLM_LOCKSPACE_LEN)) {
423 log_print("dlm cluster name %s mismatch %s",
424 dlm_config.ci_cluster_name, cluster);
425 error = -EBADR;
426 goto out;
dc68c7ed
DT
427 }
428
0f8e0d9a
DT
429 error = 0;
430
431 spin_lock(&lslist_lock);
432 list_for_each_entry(ls, &lslist, ls_list) {
433 WARN_ON(ls->ls_create_count <= 0);
434 if (ls->ls_namelen != namelen)
435 continue;
436 if (memcmp(ls->ls_name, name, namelen))
437 continue;
438 if (flags & DLM_LSFL_NEWEXCL) {
439 error = -EEXIST;
440 break;
441 }
442 ls->ls_create_count++;
8511a272
DT
443 *lockspace = ls;
444 error = 1;
0f8e0d9a 445 break;
e7fd4179 446 }
0f8e0d9a
DT
447 spin_unlock(&lslist_lock);
448
0f8e0d9a 449 if (error)
8511a272 450 goto out;
0f8e0d9a
DT
451
452 error = -ENOMEM;
e7fd4179 453
573c24c4 454 ls = kzalloc(sizeof(struct dlm_ls) + namelen, GFP_NOFS);
e7fd4179
DT
455 if (!ls)
456 goto out;
e7fd4179
DT
457 memcpy(ls->ls_name, name, namelen);
458 ls->ls_namelen = namelen;
e7fd4179
DT
459 ls->ls_lvblen = lvblen;
460 ls->ls_count = 0;
461 ls->ls_flags = 0;
c1dcf65f 462 ls->ls_scan_time = jiffies;
e7fd4179 463
60f98d18
DT
464 if (ops && dlm_config.ci_recover_callbacks) {
465 ls->ls_ops = ops;
466 ls->ls_ops_arg = ops_arg;
467 }
468
3ae1acf9
DT
469 if (flags & DLM_LSFL_TIMEWARN)
470 set_bit(LSFL_TIMEWARN, &ls->ls_flags);
3ae1acf9 471
fad59c13 472 /* ls_exflags are forced to match among nodes, and we don't
0f8e0d9a
DT
473 need to require all nodes to have some flags set */
474 ls->ls_exflags = (flags & ~(DLM_LSFL_TIMEWARN | DLM_LSFL_FS |
475 DLM_LSFL_NEWEXCL));
fad59c13 476
68c817a1 477 size = dlm_config.ci_rsbtbl_size;
e7fd4179
DT
478 ls->ls_rsbtbl_size = size;
479
c282af49 480 ls->ls_rsbtbl = vmalloc(sizeof(struct dlm_rsbtable) * size);
e7fd4179
DT
481 if (!ls->ls_rsbtbl)
482 goto out_lsfree;
483 for (i = 0; i < size; i++) {
9beb3bf5
BP
484 ls->ls_rsbtbl[i].keep.rb_node = NULL;
485 ls->ls_rsbtbl[i].toss.rb_node = NULL;
c7be761a 486 spin_lock_init(&ls->ls_rsbtbl[i].lock);
e7fd4179
DT
487 }
488
3d6aa675
DT
489 idr_init(&ls->ls_lkbidr);
490 spin_lock_init(&ls->ls_lkbidr_spin);
e7fd4179 491
68c817a1 492 size = dlm_config.ci_dirtbl_size;
e7fd4179
DT
493 ls->ls_dirtbl_size = size;
494
c282af49 495 ls->ls_dirtbl = vmalloc(sizeof(struct dlm_dirtable) * size);
e7fd4179
DT
496 if (!ls->ls_dirtbl)
497 goto out_lkbfree;
498 for (i = 0; i < size; i++) {
499 INIT_LIST_HEAD(&ls->ls_dirtbl[i].list);
305a47b1 500 spin_lock_init(&ls->ls_dirtbl[i].lock);
e7fd4179
DT
501 }
502
503 INIT_LIST_HEAD(&ls->ls_waiters);
90135925 504 mutex_init(&ls->ls_waiters_mutex);
ef0c2bb0
DT
505 INIT_LIST_HEAD(&ls->ls_orphans);
506 mutex_init(&ls->ls_orphans_mutex);
3ae1acf9
DT
507 INIT_LIST_HEAD(&ls->ls_timeout);
508 mutex_init(&ls->ls_timeout_mutex);
e7fd4179 509
3881ac04
DT
510 INIT_LIST_HEAD(&ls->ls_new_rsb);
511 spin_lock_init(&ls->ls_new_rsb_spin);
512
e7fd4179
DT
513 INIT_LIST_HEAD(&ls->ls_nodes);
514 INIT_LIST_HEAD(&ls->ls_nodes_gone);
515 ls->ls_num_nodes = 0;
516 ls->ls_low_nodeid = 0;
517 ls->ls_total_weight = 0;
518 ls->ls_node_array = NULL;
519
520 memset(&ls->ls_stub_rsb, 0, sizeof(struct dlm_rsb));
521 ls->ls_stub_rsb.res_ls = ls;
522
5de6319b
DT
523 ls->ls_debug_rsb_dentry = NULL;
524 ls->ls_debug_waiters_dentry = NULL;
e7fd4179
DT
525
526 init_waitqueue_head(&ls->ls_uevent_wait);
527 ls->ls_uevent_result = 0;
8b0e7b2c
DT
528 init_completion(&ls->ls_members_done);
529 ls->ls_members_result = -1;
e7fd4179 530
23e8e1aa
DT
531 mutex_init(&ls->ls_cb_mutex);
532 INIT_LIST_HEAD(&ls->ls_cb_delay);
533
e7fd4179 534 ls->ls_recoverd_task = NULL;
90135925 535 mutex_init(&ls->ls_recoverd_active);
e7fd4179 536 spin_lock_init(&ls->ls_recover_lock);
98f176fb
DT
537 spin_lock_init(&ls->ls_rcom_spin);
538 get_random_bytes(&ls->ls_rcom_seq, sizeof(uint64_t));
e7fd4179
DT
539 ls->ls_recover_status = 0;
540 ls->ls_recover_seq = 0;
541 ls->ls_recover_args = NULL;
542 init_rwsem(&ls->ls_in_recovery);
c36258b5 543 init_rwsem(&ls->ls_recv_active);
e7fd4179 544 INIT_LIST_HEAD(&ls->ls_requestqueue);
90135925 545 mutex_init(&ls->ls_requestqueue_mutex);
597d0cae 546 mutex_init(&ls->ls_clear_proc_locks);
e7fd4179 547
573c24c4 548 ls->ls_recover_buf = kmalloc(dlm_config.ci_buffer_size, GFP_NOFS);
e7fd4179
DT
549 if (!ls->ls_recover_buf)
550 goto out_dirfree;
551
757a4271
DT
552 ls->ls_slot = 0;
553 ls->ls_num_slots = 0;
554 ls->ls_slots_size = 0;
555 ls->ls_slots = NULL;
556
e7fd4179
DT
557 INIT_LIST_HEAD(&ls->ls_recover_list);
558 spin_lock_init(&ls->ls_recover_list_lock);
559 ls->ls_recover_list_count = 0;
597d0cae 560 ls->ls_local_handle = ls;
e7fd4179
DT
561 init_waitqueue_head(&ls->ls_wait_general);
562 INIT_LIST_HEAD(&ls->ls_root_list);
563 init_rwsem(&ls->ls_root_sem);
564
565 down_write(&ls->ls_in_recovery);
566
5f88f1ea 567 spin_lock(&lslist_lock);
0f8e0d9a 568 ls->ls_create_count = 1;
5f88f1ea
DT
569 list_add(&ls->ls_list, &lslist);
570 spin_unlock(&lslist_lock);
571
23e8e1aa
DT
572 if (flags & DLM_LSFL_FS) {
573 error = dlm_callback_start(ls);
574 if (error) {
575 log_error(ls, "can't start dlm_callback %d", error);
576 goto out_delist;
577 }
578 }
579
5f88f1ea 580 /* needs to find ls in lslist */
e7fd4179
DT
581 error = dlm_recoverd_start(ls);
582 if (error) {
583 log_error(ls, "can't start dlm_recoverd %d", error);
23e8e1aa 584 goto out_callback;
e7fd4179
DT
585 }
586
901195ed
GKH
587 ls->ls_kobj.kset = dlm_kset;
588 error = kobject_init_and_add(&ls->ls_kobj, &dlm_ktype, NULL,
589 "%s", ls->ls_name);
e7fd4179 590 if (error)
23e8e1aa 591 goto out_recoverd;
901195ed 592 kobject_uevent(&ls->ls_kobj, KOBJ_ADD);
79d72b54
DT
593
594 /* let kobject handle freeing of ls if there's an error */
595 do_unreg = 1;
e7fd4179 596
8b0e7b2c
DT
597 /* This uevent triggers dlm_controld in userspace to add us to the
598 group of nodes that are members of this lockspace (managed by the
599 cluster infrastructure.) Once it's done that, it tells us who the
600 current lockspace members are (via configfs) and then tells the
601 lockspace to start running (via sysfs) in dlm_ls_start(). */
602
e7fd4179
DT
603 error = do_uevent(ls, 1);
604 if (error)
23e8e1aa 605 goto out_recoverd;
79d72b54 606
8b0e7b2c
DT
607 wait_for_completion(&ls->ls_members_done);
608 error = ls->ls_members_result;
609 if (error)
610 goto out_members;
611
79d72b54
DT
612 dlm_create_debug_file(ls);
613
614 log_debug(ls, "join complete");
e7fd4179
DT
615 *lockspace = ls;
616 return 0;
617
8b0e7b2c
DT
618 out_members:
619 do_uevent(ls, 0);
620 dlm_clear_members(ls);
621 kfree(ls->ls_node_array);
23e8e1aa 622 out_recoverd:
5f88f1ea 623 dlm_recoverd_stop(ls);
23e8e1aa
DT
624 out_callback:
625 dlm_callback_stop(ls);
79d72b54 626 out_delist:
e7fd4179
DT
627 spin_lock(&lslist_lock);
628 list_del(&ls->ls_list);
629 spin_unlock(&lslist_lock);
e7fd4179
DT
630 kfree(ls->ls_recover_buf);
631 out_dirfree:
c282af49 632 vfree(ls->ls_dirtbl);
e7fd4179 633 out_lkbfree:
3d6aa675 634 idr_destroy(&ls->ls_lkbidr);
c282af49 635 vfree(ls->ls_rsbtbl);
e7fd4179 636 out_lsfree:
79d72b54 637 if (do_unreg)
197b12d6 638 kobject_put(&ls->ls_kobj);
79d72b54
DT
639 else
640 kfree(ls);
e7fd4179
DT
641 out:
642 module_put(THIS_MODULE);
643 return error;
644}
645
60f98d18
DT
646int dlm_new_lockspace(const char *name, const char *cluster,
647 uint32_t flags, int lvblen,
648 const struct dlm_lockspace_ops *ops, void *ops_arg,
649 int *ops_result, dlm_lockspace_t **lockspace)
e7fd4179
DT
650{
651 int error = 0;
652
90135925 653 mutex_lock(&ls_lock);
e7fd4179
DT
654 if (!ls_count)
655 error = threads_start();
656 if (error)
657 goto out;
658
60f98d18
DT
659 error = new_lockspace(name, cluster, flags, lvblen, ops, ops_arg,
660 ops_result, lockspace);
e7fd4179
DT
661 if (!error)
662 ls_count++;
8511a272
DT
663 if (error > 0)
664 error = 0;
665 if (!ls_count)
8b0e7b2c 666 threads_stop();
e7fd4179 667 out:
90135925 668 mutex_unlock(&ls_lock);
e7fd4179
DT
669 return error;
670}
671
3d6aa675 672static int lkb_idr_is_local(int id, void *p, void *data)
e7fd4179 673{
3d6aa675
DT
674 struct dlm_lkb *lkb = p;
675
676 if (!lkb->lkb_nodeid)
677 return 1;
678 return 0;
679}
680
681static int lkb_idr_is_any(int id, void *p, void *data)
682{
683 return 1;
684}
685
686static int lkb_idr_free(int id, void *p, void *data)
687{
688 struct dlm_lkb *lkb = p;
689
3d6aa675
DT
690 if (lkb->lkb_lvbptr && lkb->lkb_flags & DLM_IFL_MSTCPY)
691 dlm_free_lvb(lkb->lkb_lvbptr);
692
693 dlm_free_lkb(lkb);
694 return 0;
695}
696
697/* NOTE: We check the lkbidr here rather than the resource table.
698 This is because there may be LKBs queued as ASTs that have been unlinked
699 from their RSBs and are pending deletion once the AST has been delivered */
700
701static int lockspace_busy(struct dlm_ls *ls, int force)
702{
703 int rv;
704
705 spin_lock(&ls->ls_lkbidr_spin);
706 if (force == 0) {
707 rv = idr_for_each(&ls->ls_lkbidr, lkb_idr_is_any, ls);
708 } else if (force == 1) {
709 rv = idr_for_each(&ls->ls_lkbidr, lkb_idr_is_local, ls);
710 } else {
711 rv = 0;
e7fd4179 712 }
3d6aa675
DT
713 spin_unlock(&ls->ls_lkbidr_spin);
714 return rv;
e7fd4179
DT
715}
716
717static int release_lockspace(struct dlm_ls *ls, int force)
718{
e7fd4179 719 struct dlm_rsb *rsb;
9beb3bf5 720 struct rb_node *n;
0f8e0d9a
DT
721 int i, busy, rv;
722
3d6aa675 723 busy = lockspace_busy(ls, force);
0f8e0d9a
DT
724
725 spin_lock(&lslist_lock);
726 if (ls->ls_create_count == 1) {
3d6aa675 727 if (busy) {
0f8e0d9a 728 rv = -EBUSY;
3d6aa675 729 } else {
0f8e0d9a
DT
730 /* remove_lockspace takes ls off lslist */
731 ls->ls_create_count = 0;
732 rv = 0;
733 }
734 } else if (ls->ls_create_count > 1) {
735 rv = --ls->ls_create_count;
736 } else {
737 rv = -EINVAL;
738 }
739 spin_unlock(&lslist_lock);
740
741 if (rv) {
742 log_debug(ls, "release_lockspace no remove %d", rv);
743 return rv;
744 }
e7fd4179 745
0f8e0d9a 746 dlm_device_deregister(ls);
e7fd4179 747
dc68c7ed 748 if (force < 3 && dlm_user_daemon_available())
e7fd4179
DT
749 do_uevent(ls, 0);
750
751 dlm_recoverd_stop(ls);
752
23e8e1aa
DT
753 dlm_callback_stop(ls);
754
e7fd4179
DT
755 remove_lockspace(ls);
756
757 dlm_delete_debug_file(ls);
758
e7fd4179
DT
759 kfree(ls->ls_recover_buf);
760
761 /*
762 * Free direntry structs.
763 */
764
765 dlm_dir_clear(ls);
c282af49 766 vfree(ls->ls_dirtbl);
e7fd4179
DT
767
768 /*
3d6aa675 769 * Free all lkb's in idr
e7fd4179
DT
770 */
771
3d6aa675
DT
772 idr_for_each(&ls->ls_lkbidr, lkb_idr_free, ls);
773 idr_remove_all(&ls->ls_lkbidr);
774 idr_destroy(&ls->ls_lkbidr);
e7fd4179 775
e7fd4179
DT
776 /*
777 * Free all rsb's on rsbtbl[] lists
778 */
779
780 for (i = 0; i < ls->ls_rsbtbl_size; i++) {
9beb3bf5
BP
781 while ((n = rb_first(&ls->ls_rsbtbl[i].keep))) {
782 rsb = rb_entry(n, struct dlm_rsb, res_hashnode);
783 rb_erase(n, &ls->ls_rsbtbl[i].keep);
52bda2b5 784 dlm_free_rsb(rsb);
e7fd4179
DT
785 }
786
9beb3bf5
BP
787 while ((n = rb_first(&ls->ls_rsbtbl[i].toss))) {
788 rsb = rb_entry(n, struct dlm_rsb, res_hashnode);
789 rb_erase(n, &ls->ls_rsbtbl[i].toss);
52bda2b5 790 dlm_free_rsb(rsb);
e7fd4179
DT
791 }
792 }
793
c282af49 794 vfree(ls->ls_rsbtbl);
e7fd4179 795
3881ac04
DT
796 while (!list_empty(&ls->ls_new_rsb)) {
797 rsb = list_first_entry(&ls->ls_new_rsb, struct dlm_rsb,
798 res_hashchain);
799 list_del(&rsb->res_hashchain);
800 dlm_free_rsb(rsb);
801 }
802
e7fd4179
DT
803 /*
804 * Free structures on any other lists
805 */
806
2896ee37 807 dlm_purge_requestqueue(ls);
e7fd4179
DT
808 kfree(ls->ls_recover_args);
809 dlm_clear_free_entries(ls);
810 dlm_clear_members(ls);
811 dlm_clear_members_gone(ls);
812 kfree(ls->ls_node_array);
0f8e0d9a 813 log_debug(ls, "release_lockspace final free");
197b12d6 814 kobject_put(&ls->ls_kobj);
79d72b54 815 /* The ls structure will be freed when the kobject is done with */
e7fd4179 816
e7fd4179
DT
817 module_put(THIS_MODULE);
818 return 0;
819}
820
821/*
822 * Called when a system has released all its locks and is not going to use the
823 * lockspace any longer. We free everything we're managing for this lockspace.
824 * Remaining nodes will go through the recovery process as if we'd died. The
825 * lockspace must continue to function as usual, participating in recoveries,
826 * until this returns.
827 *
828 * Force has 4 possible values:
829 * 0 - don't destroy locksapce if it has any LKBs
830 * 1 - destroy lockspace if it has remote LKBs but not if it has local LKBs
831 * 2 - destroy lockspace regardless of LKBs
832 * 3 - destroy lockspace as part of a forced shutdown
833 */
834
835int dlm_release_lockspace(void *lockspace, int force)
836{
837 struct dlm_ls *ls;
0f8e0d9a 838 int error;
e7fd4179
DT
839
840 ls = dlm_find_lockspace_local(lockspace);
841 if (!ls)
842 return -EINVAL;
843 dlm_put_lockspace(ls);
0f8e0d9a
DT
844
845 mutex_lock(&ls_lock);
846 error = release_lockspace(ls, force);
847 if (!error)
848 ls_count--;
278afcbf 849 if (!ls_count)
0f8e0d9a
DT
850 threads_stop();
851 mutex_unlock(&ls_lock);
852
853 return error;
e7fd4179
DT
854}
855
dc68c7ed
DT
856void dlm_stop_lockspaces(void)
857{
858 struct dlm_ls *ls;
859
860 restart:
861 spin_lock(&lslist_lock);
862 list_for_each_entry(ls, &lslist, ls_list) {
863 if (!test_bit(LSFL_RUNNING, &ls->ls_flags))
864 continue;
865 spin_unlock(&lslist_lock);
866 log_error(ls, "no userland control daemon, stopping lockspace");
867 dlm_ls_stop(ls);
868 goto restart;
869 }
870 spin_unlock(&lslist_lock);
871}
872
This page took 0.460356 seconds and 5 git commands to generate.