]>
Commit | Line | Data |
---|---|---|
6714d8e8 KH |
1 | /* -*- mode: c; c-basic-offset: 8; -*- |
2 | * vim: noexpandtab sw=8 ts=8 sts=0: | |
3 | * | |
4 | * dlmdomain.c | |
5 | * | |
6 | * defines domain join / leave apis | |
7 | * | |
8 | * Copyright (C) 2004 Oracle. All rights reserved. | |
9 | * | |
10 | * This program is free software; you can redistribute it and/or | |
11 | * modify it under the terms of the GNU General Public | |
12 | * License as published by the Free Software Foundation; either | |
13 | * version 2 of the License, or (at your option) any later version. | |
14 | * | |
15 | * This program is distributed in the hope that it will be useful, | |
16 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
17 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
18 | * General Public License for more details. | |
19 | * | |
20 | * You should have received a copy of the GNU General Public | |
21 | * License along with this program; if not, write to the | |
22 | * Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
23 | * Boston, MA 021110-1307, USA. | |
24 | * | |
25 | */ | |
26 | ||
27 | #include <linux/module.h> | |
28 | #include <linux/types.h> | |
29 | #include <linux/slab.h> | |
30 | #include <linux/highmem.h> | |
31 | #include <linux/utsname.h> | |
32 | #include <linux/init.h> | |
33 | #include <linux/spinlock.h> | |
34 | #include <linux/delay.h> | |
35 | #include <linux/err.h> | |
6325b4a2 | 36 | #include <linux/debugfs.h> |
6714d8e8 KH |
37 | |
38 | #include "cluster/heartbeat.h" | |
39 | #include "cluster/nodemanager.h" | |
40 | #include "cluster/tcp.h" | |
41 | ||
42 | #include "dlmapi.h" | |
43 | #include "dlmcommon.h" | |
6714d8e8 | 44 | #include "dlmdomain.h" |
6325b4a2 | 45 | #include "dlmdebug.h" |
6714d8e8 KH |
46 | |
47 | #include "dlmver.h" | |
48 | ||
49 | #define MLOG_MASK_PREFIX (ML_DLM|ML_DLM_DOMAIN) | |
50 | #include "cluster/masklog.h" | |
51 | ||
1faf2894 SE |
52 | /* |
53 | * ocfs2 node maps are array of long int, which limits to send them freely | |
54 | * across the wire due to endianness issues. To workaround this, we convert | |
55 | * long ints to byte arrays. Following 3 routines are helper functions to | |
56 | * set/test/copy bits within those array of bytes | |
57 | */ | |
58 | static inline void byte_set_bit(u8 nr, u8 map[]) | |
59 | { | |
60 | map[nr >> 3] |= (1UL << (nr & 7)); | |
61 | } | |
62 | ||
63 | static inline int byte_test_bit(u8 nr, u8 map[]) | |
64 | { | |
65 | return ((1UL << (nr & 7)) & (map[nr >> 3])) != 0; | |
66 | } | |
67 | ||
68 | static inline void byte_copymap(u8 dmap[], unsigned long smap[], | |
69 | unsigned int sz) | |
70 | { | |
71 | unsigned int nn; | |
72 | ||
73 | if (!sz) | |
74 | return; | |
75 | ||
76 | memset(dmap, 0, ((sz + 7) >> 3)); | |
77 | for (nn = 0 ; nn < sz; nn++) | |
78 | if (test_bit(nn, smap)) | |
79 | byte_set_bit(nn, dmap); | |
80 | } | |
81 | ||
03d864c0 DP |
82 | static void dlm_free_pagevec(void **vec, int pages) |
83 | { | |
84 | while (pages--) | |
85 | free_page((unsigned long)vec[pages]); | |
86 | kfree(vec); | |
87 | } | |
88 | ||
89 | static void **dlm_alloc_pagevec(int pages) | |
90 | { | |
91 | void **vec = kmalloc(pages * sizeof(void *), GFP_KERNEL); | |
92 | int i; | |
93 | ||
94 | if (!vec) | |
95 | return NULL; | |
96 | ||
97 | for (i = 0; i < pages; i++) | |
98 | if (!(vec[i] = (void *)__get_free_page(GFP_KERNEL))) | |
99 | goto out_free; | |
c8f33b6e | 100 | |
685f1adb | 101 | mlog(0, "Allocated DLM hash pagevec; %d pages (%lu expected), %lu buckets per page\n", |
f5a923d1 MF |
102 | pages, (unsigned long)DLM_HASH_PAGES, |
103 | (unsigned long)DLM_BUCKETS_PER_PAGE); | |
03d864c0 DP |
104 | return vec; |
105 | out_free: | |
106 | dlm_free_pagevec(vec, i); | |
107 | return NULL; | |
108 | } | |
109 | ||
6714d8e8 KH |
110 | /* |
111 | * | |
112 | * spinlock lock ordering: if multiple locks are needed, obey this ordering: | |
113 | * dlm_domain_lock | |
114 | * struct dlm_ctxt->spinlock | |
115 | * struct dlm_lock_resource->spinlock | |
116 | * struct dlm_ctxt->master_lock | |
117 | * struct dlm_ctxt->ast_lock | |
118 | * dlm_master_list_entry->spinlock | |
119 | * dlm_lock->spinlock | |
120 | * | |
121 | */ | |
122 | ||
34af946a | 123 | DEFINE_SPINLOCK(dlm_domain_lock); |
6714d8e8 KH |
124 | LIST_HEAD(dlm_domains); |
125 | static DECLARE_WAIT_QUEUE_HEAD(dlm_domain_events); | |
126 | ||
d24fbcda JB |
127 | /* |
128 | * The supported protocol version for DLM communication. Running domains | |
129 | * will have a negotiated version with the same major number and a minor | |
130 | * number equal or smaller. The dlm_ctxt->dlm_locking_proto field should | |
131 | * be used to determine what a running domain is actually using. | |
132 | */ | |
133 | static const struct dlm_protocol_version dlm_protocol = { | |
134 | .pv_major = 1, | |
135 | .pv_minor = 0, | |
136 | }; | |
137 | ||
6714d8e8 KH |
138 | #define DLM_DOMAIN_BACKOFF_MS 200 |
139 | ||
d74c9803 KH |
140 | static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data, |
141 | void **ret_data); | |
142 | static int dlm_assert_joined_handler(struct o2net_msg *msg, u32 len, void *data, | |
143 | void **ret_data); | |
144 | static int dlm_cancel_join_handler(struct o2net_msg *msg, u32 len, void *data, | |
145 | void **ret_data); | |
146 | static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, | |
147 | void **ret_data); | |
d24fbcda JB |
148 | static int dlm_protocol_compare(struct dlm_protocol_version *existing, |
149 | struct dlm_protocol_version *request); | |
6714d8e8 KH |
150 | |
151 | static void dlm_unregister_domain_handlers(struct dlm_ctxt *dlm); | |
152 | ||
153 | void __dlm_unhash_lockres(struct dlm_lock_resource *lockres) | |
154 | { | |
78062cb2 SM |
155 | if (!hlist_unhashed(&lockres->hash_node)) { |
156 | hlist_del_init(&lockres->hash_node); | |
157 | dlm_lockres_put(lockres); | |
158 | } | |
6714d8e8 KH |
159 | } |
160 | ||
161 | void __dlm_insert_lockres(struct dlm_ctxt *dlm, | |
162 | struct dlm_lock_resource *res) | |
163 | { | |
81f2094a | 164 | struct hlist_head *bucket; |
6714d8e8 KH |
165 | struct qstr *q; |
166 | ||
167 | assert_spin_locked(&dlm->spinlock); | |
168 | ||
169 | q = &res->lockname; | |
03d864c0 | 170 | bucket = dlm_lockres_hash(dlm, q->hash); |
6714d8e8 KH |
171 | |
172 | /* get a reference for our hashtable */ | |
173 | dlm_lockres_get(res); | |
174 | ||
81f2094a | 175 | hlist_add_head(&res->hash_node, bucket); |
6714d8e8 KH |
176 | } |
177 | ||
ba2bf218 KH |
178 | struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm, |
179 | const char *name, | |
180 | unsigned int len, | |
181 | unsigned int hash) | |
6714d8e8 | 182 | { |
81f2094a | 183 | struct hlist_head *bucket; |
4198985f | 184 | struct hlist_node *list; |
6714d8e8 KH |
185 | |
186 | mlog_entry("%.*s\n", len, name); | |
187 | ||
188 | assert_spin_locked(&dlm->spinlock); | |
189 | ||
03d864c0 DP |
190 | bucket = dlm_lockres_hash(dlm, hash); |
191 | ||
4198985f DP |
192 | hlist_for_each(list, bucket) { |
193 | struct dlm_lock_resource *res = hlist_entry(list, | |
194 | struct dlm_lock_resource, hash_node); | |
195 | if (res->lockname.name[0] != name[0]) | |
196 | continue; | |
197 | if (unlikely(res->lockname.len != len)) | |
198 | continue; | |
199 | if (memcmp(res->lockname.name + 1, name + 1, len - 1)) | |
200 | continue; | |
201 | dlm_lockres_get(res); | |
202 | return res; | |
6714d8e8 | 203 | } |
4198985f | 204 | return NULL; |
6714d8e8 KH |
205 | } |
206 | ||
ba2bf218 KH |
207 | /* intended to be called by functions which do not care about lock |
208 | * resources which are being purged (most net _handler functions). | |
209 | * this will return NULL for any lock resource which is found but | |
210 | * currently in the process of dropping its mastery reference. | |
211 | * use __dlm_lookup_lockres_full when you need the lock resource | |
212 | * regardless (e.g. dlm_get_lock_resource) */ | |
213 | struct dlm_lock_resource * __dlm_lookup_lockres(struct dlm_ctxt *dlm, | |
214 | const char *name, | |
215 | unsigned int len, | |
216 | unsigned int hash) | |
217 | { | |
218 | struct dlm_lock_resource *res = NULL; | |
219 | ||
220 | mlog_entry("%.*s\n", len, name); | |
221 | ||
222 | assert_spin_locked(&dlm->spinlock); | |
223 | ||
224 | res = __dlm_lookup_lockres_full(dlm, name, len, hash); | |
225 | if (res) { | |
226 | spin_lock(&res->spinlock); | |
227 | if (res->state & DLM_LOCK_RES_DROPPING_REF) { | |
228 | spin_unlock(&res->spinlock); | |
229 | dlm_lockres_put(res); | |
230 | return NULL; | |
231 | } | |
232 | spin_unlock(&res->spinlock); | |
233 | } | |
234 | ||
235 | return res; | |
236 | } | |
237 | ||
6714d8e8 KH |
238 | struct dlm_lock_resource * dlm_lookup_lockres(struct dlm_ctxt *dlm, |
239 | const char *name, | |
240 | unsigned int len) | |
241 | { | |
242 | struct dlm_lock_resource *res; | |
a3d33291 | 243 | unsigned int hash = dlm_lockid_hash(name, len); |
6714d8e8 KH |
244 | |
245 | spin_lock(&dlm->spinlock); | |
a3d33291 | 246 | res = __dlm_lookup_lockres(dlm, name, len, hash); |
6714d8e8 KH |
247 | spin_unlock(&dlm->spinlock); |
248 | return res; | |
249 | } | |
250 | ||
251 | static struct dlm_ctxt * __dlm_lookup_domain_full(const char *domain, int len) | |
252 | { | |
253 | struct dlm_ctxt *tmp = NULL; | |
254 | struct list_head *iter; | |
255 | ||
256 | assert_spin_locked(&dlm_domain_lock); | |
257 | ||
258 | /* tmp->name here is always NULL terminated, | |
259 | * but domain may not be! */ | |
260 | list_for_each(iter, &dlm_domains) { | |
261 | tmp = list_entry (iter, struct dlm_ctxt, list); | |
262 | if (strlen(tmp->name) == len && | |
263 | memcmp(tmp->name, domain, len)==0) | |
264 | break; | |
265 | tmp = NULL; | |
266 | } | |
267 | ||
268 | return tmp; | |
269 | } | |
270 | ||
271 | /* For null terminated domain strings ONLY */ | |
272 | static struct dlm_ctxt * __dlm_lookup_domain(const char *domain) | |
273 | { | |
274 | assert_spin_locked(&dlm_domain_lock); | |
275 | ||
276 | return __dlm_lookup_domain_full(domain, strlen(domain)); | |
277 | } | |
278 | ||
279 | ||
280 | /* returns true on one of two conditions: | |
281 | * 1) the domain does not exist | |
282 | * 2) the domain exists and it's state is "joined" */ | |
283 | static int dlm_wait_on_domain_helper(const char *domain) | |
284 | { | |
285 | int ret = 0; | |
286 | struct dlm_ctxt *tmp = NULL; | |
287 | ||
288 | spin_lock(&dlm_domain_lock); | |
289 | ||
290 | tmp = __dlm_lookup_domain(domain); | |
291 | if (!tmp) | |
292 | ret = 1; | |
293 | else if (tmp->dlm_state == DLM_CTXT_JOINED) | |
294 | ret = 1; | |
295 | ||
296 | spin_unlock(&dlm_domain_lock); | |
297 | return ret; | |
298 | } | |
299 | ||
300 | static void dlm_free_ctxt_mem(struct dlm_ctxt *dlm) | |
301 | { | |
6325b4a2 SM |
302 | dlm_destroy_debugfs_subroot(dlm); |
303 | ||
81f2094a | 304 | if (dlm->lockres_hash) |
03d864c0 | 305 | dlm_free_pagevec((void **)dlm->lockres_hash, DLM_HASH_PAGES); |
6714d8e8 KH |
306 | |
307 | if (dlm->name) | |
308 | kfree(dlm->name); | |
309 | ||
310 | kfree(dlm); | |
311 | } | |
312 | ||
313 | /* A little strange - this function will be called while holding | |
314 | * dlm_domain_lock and is expected to be holding it on the way out. We | |
315 | * will however drop and reacquire it multiple times */ | |
316 | static void dlm_ctxt_release(struct kref *kref) | |
317 | { | |
318 | struct dlm_ctxt *dlm; | |
319 | ||
320 | dlm = container_of(kref, struct dlm_ctxt, dlm_refs); | |
321 | ||
322 | BUG_ON(dlm->num_joins); | |
323 | BUG_ON(dlm->dlm_state == DLM_CTXT_JOINED); | |
324 | ||
325 | /* we may still be in the list if we hit an error during join. */ | |
326 | list_del_init(&dlm->list); | |
327 | ||
328 | spin_unlock(&dlm_domain_lock); | |
329 | ||
330 | mlog(0, "freeing memory from domain %s\n", dlm->name); | |
331 | ||
332 | wake_up(&dlm_domain_events); | |
333 | ||
334 | dlm_free_ctxt_mem(dlm); | |
335 | ||
336 | spin_lock(&dlm_domain_lock); | |
337 | } | |
338 | ||
339 | void dlm_put(struct dlm_ctxt *dlm) | |
340 | { | |
341 | spin_lock(&dlm_domain_lock); | |
342 | kref_put(&dlm->dlm_refs, dlm_ctxt_release); | |
343 | spin_unlock(&dlm_domain_lock); | |
344 | } | |
345 | ||
346 | static void __dlm_get(struct dlm_ctxt *dlm) | |
347 | { | |
348 | kref_get(&dlm->dlm_refs); | |
349 | } | |
350 | ||
351 | /* given a questionable reference to a dlm object, gets a reference if | |
352 | * it can find it in the list, otherwise returns NULL in which case | |
353 | * you shouldn't trust your pointer. */ | |
354 | struct dlm_ctxt *dlm_grab(struct dlm_ctxt *dlm) | |
355 | { | |
356 | struct list_head *iter; | |
357 | struct dlm_ctxt *target = NULL; | |
358 | ||
359 | spin_lock(&dlm_domain_lock); | |
360 | ||
361 | list_for_each(iter, &dlm_domains) { | |
362 | target = list_entry (iter, struct dlm_ctxt, list); | |
363 | ||
364 | if (target == dlm) { | |
365 | __dlm_get(target); | |
366 | break; | |
367 | } | |
368 | ||
369 | target = NULL; | |
370 | } | |
371 | ||
372 | spin_unlock(&dlm_domain_lock); | |
373 | ||
374 | return target; | |
375 | } | |
376 | ||
377 | int dlm_domain_fully_joined(struct dlm_ctxt *dlm) | |
378 | { | |
379 | int ret; | |
380 | ||
381 | spin_lock(&dlm_domain_lock); | |
382 | ret = (dlm->dlm_state == DLM_CTXT_JOINED) || | |
383 | (dlm->dlm_state == DLM_CTXT_IN_SHUTDOWN); | |
384 | spin_unlock(&dlm_domain_lock); | |
385 | ||
386 | return ret; | |
387 | } | |
388 | ||
3156d267 KH |
389 | static void dlm_destroy_dlm_worker(struct dlm_ctxt *dlm) |
390 | { | |
391 | if (dlm->dlm_worker) { | |
392 | flush_workqueue(dlm->dlm_worker); | |
393 | destroy_workqueue(dlm->dlm_worker); | |
394 | dlm->dlm_worker = NULL; | |
395 | } | |
396 | } | |
397 | ||
6714d8e8 KH |
398 | static void dlm_complete_dlm_shutdown(struct dlm_ctxt *dlm) |
399 | { | |
400 | dlm_unregister_domain_handlers(dlm); | |
007dce53 | 401 | dlm_debug_shutdown(dlm); |
6714d8e8 KH |
402 | dlm_complete_thread(dlm); |
403 | dlm_complete_recovery_thread(dlm); | |
3156d267 | 404 | dlm_destroy_dlm_worker(dlm); |
6714d8e8 KH |
405 | |
406 | /* We've left the domain. Now we can take ourselves out of the | |
407 | * list and allow the kref stuff to help us free the | |
408 | * memory. */ | |
409 | spin_lock(&dlm_domain_lock); | |
410 | list_del_init(&dlm->list); | |
411 | spin_unlock(&dlm_domain_lock); | |
412 | ||
413 | /* Wake up anyone waiting for us to remove this domain */ | |
414 | wake_up(&dlm_domain_events); | |
415 | } | |
416 | ||
ba2bf218 | 417 | static int dlm_migrate_all_locks(struct dlm_ctxt *dlm) |
6714d8e8 | 418 | { |
ba2bf218 | 419 | int i, num, n, ret = 0; |
6714d8e8 | 420 | struct dlm_lock_resource *res; |
ba2bf218 KH |
421 | struct hlist_node *iter; |
422 | struct hlist_head *bucket; | |
423 | int dropped; | |
6714d8e8 KH |
424 | |
425 | mlog(0, "Migrating locks from domain %s\n", dlm->name); | |
ba2bf218 KH |
426 | |
427 | num = 0; | |
6714d8e8 | 428 | spin_lock(&dlm->spinlock); |
81f2094a | 429 | for (i = 0; i < DLM_HASH_BUCKETS; i++) { |
ba2bf218 KH |
430 | redo_bucket: |
431 | n = 0; | |
432 | bucket = dlm_lockres_hash(dlm, i); | |
433 | iter = bucket->first; | |
434 | while (iter) { | |
435 | n++; | |
436 | res = hlist_entry(iter, struct dlm_lock_resource, | |
437 | hash_node); | |
6714d8e8 | 438 | dlm_lockres_get(res); |
ba2bf218 KH |
439 | /* migrate, if necessary. this will drop the dlm |
440 | * spinlock and retake it if it does migration. */ | |
441 | dropped = dlm_empty_lockres(dlm, res); | |
442 | ||
443 | spin_lock(&res->spinlock); | |
444 | __dlm_lockres_calc_usage(dlm, res); | |
445 | iter = res->hash_node.next; | |
446 | spin_unlock(&res->spinlock); | |
447 | ||
6714d8e8 | 448 | dlm_lockres_put(res); |
ba2bf218 | 449 | |
ba2bf218 KH |
450 | if (dropped) |
451 | goto redo_bucket; | |
6714d8e8 | 452 | } |
0d01af6e | 453 | cond_resched_lock(&dlm->spinlock); |
ba2bf218 KH |
454 | num += n; |
455 | mlog(0, "%s: touched %d lockreses in bucket %d " | |
456 | "(tot=%d)\n", dlm->name, n, i, num); | |
6714d8e8 KH |
457 | } |
458 | spin_unlock(&dlm->spinlock); | |
ba2bf218 KH |
459 | wake_up(&dlm->dlm_thread_wq); |
460 | ||
461 | /* let the dlm thread take care of purging, keep scanning until | |
462 | * nothing remains in the hash */ | |
463 | if (num) { | |
464 | mlog(0, "%s: %d lock resources in hash last pass\n", | |
465 | dlm->name, num); | |
466 | ret = -EAGAIN; | |
467 | } | |
6714d8e8 | 468 | mlog(0, "DONE Migrating locks from domain %s\n", dlm->name); |
ba2bf218 | 469 | return ret; |
6714d8e8 KH |
470 | } |
471 | ||
472 | static int dlm_no_joining_node(struct dlm_ctxt *dlm) | |
473 | { | |
474 | int ret; | |
475 | ||
476 | spin_lock(&dlm->spinlock); | |
477 | ret = dlm->joining_node == DLM_LOCK_RES_OWNER_UNKNOWN; | |
478 | spin_unlock(&dlm->spinlock); | |
479 | ||
480 | return ret; | |
481 | } | |
482 | ||
483 | static void dlm_mark_domain_leaving(struct dlm_ctxt *dlm) | |
484 | { | |
485 | /* Yikes, a double spinlock! I need domain_lock for the dlm | |
486 | * state and the dlm spinlock for join state... Sorry! */ | |
487 | again: | |
488 | spin_lock(&dlm_domain_lock); | |
489 | spin_lock(&dlm->spinlock); | |
490 | ||
491 | if (dlm->joining_node != DLM_LOCK_RES_OWNER_UNKNOWN) { | |
492 | mlog(0, "Node %d is joining, we wait on it.\n", | |
493 | dlm->joining_node); | |
494 | spin_unlock(&dlm->spinlock); | |
495 | spin_unlock(&dlm_domain_lock); | |
496 | ||
497 | wait_event(dlm->dlm_join_events, dlm_no_joining_node(dlm)); | |
498 | goto again; | |
499 | } | |
500 | ||
501 | dlm->dlm_state = DLM_CTXT_LEAVING; | |
502 | spin_unlock(&dlm->spinlock); | |
503 | spin_unlock(&dlm_domain_lock); | |
504 | } | |
505 | ||
506 | static void __dlm_print_nodes(struct dlm_ctxt *dlm) | |
507 | { | |
508 | int node = -1; | |
509 | ||
510 | assert_spin_locked(&dlm->spinlock); | |
511 | ||
781ee3e2 | 512 | printk(KERN_INFO "ocfs2_dlm: Nodes in domain (\"%s\"): ", dlm->name); |
6714d8e8 KH |
513 | |
514 | while ((node = find_next_bit(dlm->domain_map, O2NM_MAX_NODES, | |
515 | node + 1)) < O2NM_MAX_NODES) { | |
781ee3e2 | 516 | printk("%d ", node); |
6714d8e8 | 517 | } |
781ee3e2 | 518 | printk("\n"); |
6714d8e8 KH |
519 | } |
520 | ||
d74c9803 KH |
521 | static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, |
522 | void **ret_data) | |
6714d8e8 KH |
523 | { |
524 | struct dlm_ctxt *dlm = data; | |
525 | unsigned int node; | |
526 | struct dlm_exit_domain *exit_msg = (struct dlm_exit_domain *) msg->buf; | |
527 | ||
528 | mlog_entry("%p %u %p", msg, len, data); | |
529 | ||
530 | if (!dlm_grab(dlm)) | |
531 | return 0; | |
532 | ||
533 | node = exit_msg->node_idx; | |
534 | ||
781ee3e2 | 535 | printk(KERN_INFO "ocfs2_dlm: Node %u leaves domain %s\n", node, dlm->name); |
6714d8e8 KH |
536 | |
537 | spin_lock(&dlm->spinlock); | |
538 | clear_bit(node, dlm->domain_map); | |
539 | __dlm_print_nodes(dlm); | |
540 | ||
541 | /* notify anything attached to the heartbeat events */ | |
542 | dlm_hb_event_notify_attached(dlm, node, 0); | |
543 | ||
544 | spin_unlock(&dlm->spinlock); | |
545 | ||
546 | dlm_put(dlm); | |
547 | ||
548 | return 0; | |
549 | } | |
550 | ||
551 | static int dlm_send_one_domain_exit(struct dlm_ctxt *dlm, | |
552 | unsigned int node) | |
553 | { | |
554 | int status; | |
555 | struct dlm_exit_domain leave_msg; | |
556 | ||
557 | mlog(0, "Asking node %u if we can leave the domain %s me = %u\n", | |
558 | node, dlm->name, dlm->node_num); | |
559 | ||
560 | memset(&leave_msg, 0, sizeof(leave_msg)); | |
561 | leave_msg.node_idx = dlm->node_num; | |
562 | ||
563 | status = o2net_send_message(DLM_EXIT_DOMAIN_MSG, dlm->key, | |
564 | &leave_msg, sizeof(leave_msg), node, | |
565 | NULL); | |
566 | ||
567 | mlog(0, "status return %d from o2net_send_message\n", status); | |
568 | ||
569 | return status; | |
570 | } | |
571 | ||
572 | ||
573 | static void dlm_leave_domain(struct dlm_ctxt *dlm) | |
574 | { | |
575 | int node, clear_node, status; | |
576 | ||
577 | /* At this point we've migrated away all our locks and won't | |
578 | * accept mastership of new ones. The dlm is responsible for | |
579 | * almost nothing now. We make sure not to confuse any joining | |
580 | * nodes and then commence shutdown procedure. */ | |
581 | ||
582 | spin_lock(&dlm->spinlock); | |
583 | /* Clear ourselves from the domain map */ | |
584 | clear_bit(dlm->node_num, dlm->domain_map); | |
585 | while ((node = find_next_bit(dlm->domain_map, O2NM_MAX_NODES, | |
586 | 0)) < O2NM_MAX_NODES) { | |
587 | /* Drop the dlm spinlock. This is safe wrt the domain_map. | |
588 | * -nodes cannot be added now as the | |
589 | * query_join_handlers knows to respond with OK_NO_MAP | |
590 | * -we catch the right network errors if a node is | |
591 | * removed from the map while we're sending him the | |
592 | * exit message. */ | |
593 | spin_unlock(&dlm->spinlock); | |
594 | ||
595 | clear_node = 1; | |
596 | ||
597 | status = dlm_send_one_domain_exit(dlm, node); | |
598 | if (status < 0 && | |
599 | status != -ENOPROTOOPT && | |
600 | status != -ENOTCONN) { | |
601 | mlog(ML_NOTICE, "Error %d sending domain exit message " | |
602 | "to node %d\n", status, node); | |
603 | ||
604 | /* Not sure what to do here but lets sleep for | |
605 | * a bit in case this was a transient | |
606 | * error... */ | |
607 | msleep(DLM_DOMAIN_BACKOFF_MS); | |
608 | clear_node = 0; | |
609 | } | |
610 | ||
611 | spin_lock(&dlm->spinlock); | |
612 | /* If we're not clearing the node bit then we intend | |
613 | * to loop back around to try again. */ | |
614 | if (clear_node) | |
615 | clear_bit(node, dlm->domain_map); | |
616 | } | |
617 | spin_unlock(&dlm->spinlock); | |
618 | } | |
619 | ||
620 | int dlm_joined(struct dlm_ctxt *dlm) | |
621 | { | |
622 | int ret = 0; | |
623 | ||
624 | spin_lock(&dlm_domain_lock); | |
625 | ||
626 | if (dlm->dlm_state == DLM_CTXT_JOINED) | |
627 | ret = 1; | |
628 | ||
629 | spin_unlock(&dlm_domain_lock); | |
630 | ||
631 | return ret; | |
632 | } | |
633 | ||
634 | int dlm_shutting_down(struct dlm_ctxt *dlm) | |
635 | { | |
636 | int ret = 0; | |
637 | ||
638 | spin_lock(&dlm_domain_lock); | |
639 | ||
640 | if (dlm->dlm_state == DLM_CTXT_IN_SHUTDOWN) | |
641 | ret = 1; | |
642 | ||
643 | spin_unlock(&dlm_domain_lock); | |
644 | ||
645 | return ret; | |
646 | } | |
647 | ||
648 | void dlm_unregister_domain(struct dlm_ctxt *dlm) | |
649 | { | |
650 | int leave = 0; | |
29576f8b | 651 | struct dlm_lock_resource *res; |
6714d8e8 KH |
652 | |
653 | spin_lock(&dlm_domain_lock); | |
654 | BUG_ON(dlm->dlm_state != DLM_CTXT_JOINED); | |
655 | BUG_ON(!dlm->num_joins); | |
656 | ||
657 | dlm->num_joins--; | |
658 | if (!dlm->num_joins) { | |
659 | /* We mark it "in shutdown" now so new register | |
660 | * requests wait until we've completely left the | |
661 | * domain. Don't use DLM_CTXT_LEAVING yet as we still | |
662 | * want new domain joins to communicate with us at | |
663 | * least until we've completed migration of our | |
664 | * resources. */ | |
665 | dlm->dlm_state = DLM_CTXT_IN_SHUTDOWN; | |
666 | leave = 1; | |
667 | } | |
668 | spin_unlock(&dlm_domain_lock); | |
669 | ||
670 | if (leave) { | |
671 | mlog(0, "shutting down domain %s\n", dlm->name); | |
672 | ||
673 | /* We changed dlm state, notify the thread */ | |
674 | dlm_kick_thread(dlm, NULL); | |
675 | ||
ba2bf218 | 676 | while (dlm_migrate_all_locks(dlm)) { |
2f5bf1f2 SM |
677 | /* Give dlm_thread time to purge the lockres' */ |
678 | msleep(500); | |
ba2bf218 KH |
679 | mlog(0, "%s: more migration to do\n", dlm->name); |
680 | } | |
29576f8b SM |
681 | |
682 | /* This list should be empty. If not, print remaining lockres */ | |
683 | if (!list_empty(&dlm->tracking_list)) { | |
684 | mlog(ML_ERROR, "Following lockres' are still on the " | |
685 | "tracking list:\n"); | |
686 | list_for_each_entry(res, &dlm->tracking_list, tracking) | |
687 | dlm_print_one_lock_resource(res); | |
688 | } | |
689 | ||
6714d8e8 KH |
690 | dlm_mark_domain_leaving(dlm); |
691 | dlm_leave_domain(dlm); | |
692 | dlm_complete_dlm_shutdown(dlm); | |
693 | } | |
694 | dlm_put(dlm); | |
695 | } | |
696 | EXPORT_SYMBOL_GPL(dlm_unregister_domain); | |
697 | ||
d24fbcda JB |
698 | static int dlm_query_join_proto_check(char *proto_type, int node, |
699 | struct dlm_protocol_version *ours, | |
700 | struct dlm_protocol_version *request) | |
701 | { | |
702 | int rc; | |
703 | struct dlm_protocol_version proto = *request; | |
704 | ||
705 | if (!dlm_protocol_compare(ours, &proto)) { | |
706 | mlog(0, | |
707 | "node %u wanted to join with %s locking protocol " | |
708 | "%u.%u, we respond with %u.%u\n", | |
709 | node, proto_type, | |
710 | request->pv_major, | |
711 | request->pv_minor, | |
712 | proto.pv_major, proto.pv_minor); | |
713 | request->pv_minor = proto.pv_minor; | |
714 | rc = 0; | |
715 | } else { | |
716 | mlog(ML_NOTICE, | |
717 | "Node %u wanted to join with %s locking " | |
718 | "protocol %u.%u, but we have %u.%u, disallowing\n", | |
719 | node, proto_type, | |
720 | request->pv_major, | |
721 | request->pv_minor, | |
722 | ours->pv_major, | |
723 | ours->pv_minor); | |
724 | rc = 1; | |
725 | } | |
726 | ||
727 | return rc; | |
728 | } | |
729 | ||
0f71b7b4 JB |
730 | /* |
731 | * struct dlm_query_join_packet is made up of four one-byte fields. They | |
732 | * are effectively in big-endian order already. However, little-endian | |
733 | * machines swap them before putting the packet on the wire (because | |
734 | * query_join's response is a status, and that status is treated as a u32 | |
735 | * on the wire). Thus, a big-endian and little-endian machines will treat | |
736 | * this structure differently. | |
737 | * | |
738 | * The solution is to have little-endian machines swap the structure when | |
739 | * converting from the structure to the u32 representation. This will | |
740 | * result in the structure having the correct format on the wire no matter | |
741 | * the host endian format. | |
742 | */ | |
743 | static void dlm_query_join_packet_to_wire(struct dlm_query_join_packet *packet, | |
744 | u32 *wire) | |
745 | { | |
746 | union dlm_query_join_response response; | |
747 | ||
748 | response.packet = *packet; | |
749 | *wire = cpu_to_be32(response.intval); | |
750 | } | |
751 | ||
752 | static void dlm_query_join_wire_to_packet(u32 wire, | |
753 | struct dlm_query_join_packet *packet) | |
754 | { | |
755 | union dlm_query_join_response response; | |
756 | ||
757 | response.intval = cpu_to_be32(wire); | |
758 | *packet = response.packet; | |
759 | } | |
760 | ||
d74c9803 KH |
761 | static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data, |
762 | void **ret_data) | |
6714d8e8 KH |
763 | { |
764 | struct dlm_query_join_request *query; | |
0f71b7b4 JB |
765 | struct dlm_query_join_packet packet = { |
766 | .code = JOIN_DISALLOW, | |
d24fbcda | 767 | }; |
6714d8e8 | 768 | struct dlm_ctxt *dlm = NULL; |
0f71b7b4 | 769 | u32 response; |
1faf2894 | 770 | u8 nodenum; |
6714d8e8 KH |
771 | |
772 | query = (struct dlm_query_join_request *) msg->buf; | |
773 | ||
774 | mlog(0, "node %u wants to join domain %s\n", query->node_idx, | |
775 | query->domain); | |
776 | ||
777 | /* | |
778 | * If heartbeat doesn't consider the node live, tell it | |
779 | * to back off and try again. This gives heartbeat a chance | |
780 | * to catch up. | |
781 | */ | |
782 | if (!o2hb_check_node_heartbeating(query->node_idx)) { | |
783 | mlog(0, "node %u is not in our live map yet\n", | |
784 | query->node_idx); | |
785 | ||
0f71b7b4 | 786 | packet.code = JOIN_DISALLOW; |
6714d8e8 KH |
787 | goto respond; |
788 | } | |
789 | ||
0f71b7b4 | 790 | packet.code = JOIN_OK_NO_MAP; |
6714d8e8 KH |
791 | |
792 | spin_lock(&dlm_domain_lock); | |
793 | dlm = __dlm_lookup_domain_full(query->domain, query->name_len); | |
1faf2894 SE |
794 | if (!dlm) |
795 | goto unlock_respond; | |
796 | ||
797 | /* | |
798 | * There is a small window where the joining node may not see the | |
799 | * node(s) that just left but still part of the cluster. DISALLOW | |
800 | * join request if joining node has different node map. | |
801 | */ | |
802 | nodenum=0; | |
803 | while (nodenum < O2NM_MAX_NODES) { | |
804 | if (test_bit(nodenum, dlm->domain_map)) { | |
805 | if (!byte_test_bit(nodenum, query->node_map)) { | |
e4968476 SM |
806 | mlog(0, "disallow join as node %u does not " |
807 | "have node %u in its nodemap\n", | |
808 | query->node_idx, nodenum); | |
0f71b7b4 | 809 | packet.code = JOIN_DISALLOW; |
1faf2894 SE |
810 | goto unlock_respond; |
811 | } | |
812 | } | |
813 | nodenum++; | |
814 | } | |
815 | ||
6714d8e8 | 816 | /* Once the dlm ctxt is marked as leaving then we don't want |
e2faea4c KH |
817 | * to be put in someone's domain map. |
818 | * Also, explicitly disallow joining at certain troublesome | |
819 | * times (ie. during recovery). */ | |
6714d8e8 | 820 | if (dlm && dlm->dlm_state != DLM_CTXT_LEAVING) { |
e2faea4c | 821 | int bit = query->node_idx; |
6714d8e8 KH |
822 | spin_lock(&dlm->spinlock); |
823 | ||
824 | if (dlm->dlm_state == DLM_CTXT_NEW && | |
825 | dlm->joining_node == DLM_LOCK_RES_OWNER_UNKNOWN) { | |
826 | /*If this is a brand new context and we | |
827 | * haven't started our join process yet, then | |
828 | * the other node won the race. */ | |
0f71b7b4 | 829 | packet.code = JOIN_OK_NO_MAP; |
6714d8e8 KH |
830 | } else if (dlm->joining_node != DLM_LOCK_RES_OWNER_UNKNOWN) { |
831 | /* Disallow parallel joins. */ | |
0f71b7b4 | 832 | packet.code = JOIN_DISALLOW; |
e2faea4c | 833 | } else if (dlm->reco.state & DLM_RECO_STATE_ACTIVE) { |
e4968476 | 834 | mlog(0, "node %u trying to join, but recovery " |
e2faea4c | 835 | "is ongoing.\n", bit); |
0f71b7b4 | 836 | packet.code = JOIN_DISALLOW; |
e2faea4c | 837 | } else if (test_bit(bit, dlm->recovery_map)) { |
e4968476 | 838 | mlog(0, "node %u trying to join, but it " |
e2faea4c | 839 | "still needs recovery.\n", bit); |
0f71b7b4 | 840 | packet.code = JOIN_DISALLOW; |
e2faea4c | 841 | } else if (test_bit(bit, dlm->domain_map)) { |
e4968476 | 842 | mlog(0, "node %u trying to join, but it " |
e2faea4c KH |
843 | "is still in the domain! needs recovery?\n", |
844 | bit); | |
0f71b7b4 | 845 | packet.code = JOIN_DISALLOW; |
6714d8e8 KH |
846 | } else { |
847 | /* Alright we're fully a part of this domain | |
848 | * so we keep some state as to who's joining | |
849 | * and indicate to him that needs to be fixed | |
850 | * up. */ | |
d24fbcda JB |
851 | |
852 | /* Make sure we speak compatible locking protocols. */ | |
853 | if (dlm_query_join_proto_check("DLM", bit, | |
854 | &dlm->dlm_locking_proto, | |
855 | &query->dlm_proto)) { | |
0f71b7b4 | 856 | packet.code = JOIN_PROTOCOL_MISMATCH; |
d24fbcda JB |
857 | } else if (dlm_query_join_proto_check("fs", bit, |
858 | &dlm->fs_locking_proto, | |
859 | &query->fs_proto)) { | |
0f71b7b4 | 860 | packet.code = JOIN_PROTOCOL_MISMATCH; |
d24fbcda | 861 | } else { |
0f71b7b4 JB |
862 | packet.dlm_minor = query->dlm_proto.pv_minor; |
863 | packet.fs_minor = query->fs_proto.pv_minor; | |
864 | packet.code = JOIN_OK; | |
d24fbcda JB |
865 | __dlm_set_joining_node(dlm, query->node_idx); |
866 | } | |
6714d8e8 KH |
867 | } |
868 | ||
869 | spin_unlock(&dlm->spinlock); | |
870 | } | |
1faf2894 | 871 | unlock_respond: |
6714d8e8 KH |
872 | spin_unlock(&dlm_domain_lock); |
873 | ||
874 | respond: | |
0f71b7b4 | 875 | mlog(0, "We respond with %u\n", packet.code); |
6714d8e8 | 876 | |
0f71b7b4 JB |
877 | dlm_query_join_packet_to_wire(&packet, &response); |
878 | return response; | |
6714d8e8 KH |
879 | } |
880 | ||
d74c9803 KH |
881 | static int dlm_assert_joined_handler(struct o2net_msg *msg, u32 len, void *data, |
882 | void **ret_data) | |
6714d8e8 KH |
883 | { |
884 | struct dlm_assert_joined *assert; | |
885 | struct dlm_ctxt *dlm = NULL; | |
886 | ||
887 | assert = (struct dlm_assert_joined *) msg->buf; | |
888 | ||
889 | mlog(0, "node %u asserts join on domain %s\n", assert->node_idx, | |
890 | assert->domain); | |
891 | ||
892 | spin_lock(&dlm_domain_lock); | |
893 | dlm = __dlm_lookup_domain_full(assert->domain, assert->name_len); | |
894 | /* XXX should we consider no dlm ctxt an error? */ | |
895 | if (dlm) { | |
896 | spin_lock(&dlm->spinlock); | |
897 | ||
898 | /* Alright, this node has officially joined our | |
899 | * domain. Set him in the map and clean up our | |
900 | * leftover join state. */ | |
901 | BUG_ON(dlm->joining_node != assert->node_idx); | |
902 | set_bit(assert->node_idx, dlm->domain_map); | |
903 | __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); | |
904 | ||
781ee3e2 SM |
905 | printk(KERN_INFO "ocfs2_dlm: Node %u joins domain %s\n", |
906 | assert->node_idx, dlm->name); | |
6714d8e8 KH |
907 | __dlm_print_nodes(dlm); |
908 | ||
909 | /* notify anything attached to the heartbeat events */ | |
910 | dlm_hb_event_notify_attached(dlm, assert->node_idx, 1); | |
911 | ||
912 | spin_unlock(&dlm->spinlock); | |
913 | } | |
914 | spin_unlock(&dlm_domain_lock); | |
915 | ||
916 | return 0; | |
917 | } | |
918 | ||
d74c9803 KH |
919 | static int dlm_cancel_join_handler(struct o2net_msg *msg, u32 len, void *data, |
920 | void **ret_data) | |
6714d8e8 KH |
921 | { |
922 | struct dlm_cancel_join *cancel; | |
923 | struct dlm_ctxt *dlm = NULL; | |
924 | ||
925 | cancel = (struct dlm_cancel_join *) msg->buf; | |
926 | ||
927 | mlog(0, "node %u cancels join on domain %s\n", cancel->node_idx, | |
928 | cancel->domain); | |
929 | ||
930 | spin_lock(&dlm_domain_lock); | |
931 | dlm = __dlm_lookup_domain_full(cancel->domain, cancel->name_len); | |
932 | ||
933 | if (dlm) { | |
934 | spin_lock(&dlm->spinlock); | |
935 | ||
936 | /* Yikes, this guy wants to cancel his join. No | |
937 | * problem, we simply cleanup our join state. */ | |
938 | BUG_ON(dlm->joining_node != cancel->node_idx); | |
939 | __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); | |
940 | ||
941 | spin_unlock(&dlm->spinlock); | |
942 | } | |
943 | spin_unlock(&dlm_domain_lock); | |
944 | ||
945 | return 0; | |
946 | } | |
947 | ||
948 | static int dlm_send_one_join_cancel(struct dlm_ctxt *dlm, | |
949 | unsigned int node) | |
950 | { | |
951 | int status; | |
952 | struct dlm_cancel_join cancel_msg; | |
953 | ||
954 | memset(&cancel_msg, 0, sizeof(cancel_msg)); | |
955 | cancel_msg.node_idx = dlm->node_num; | |
956 | cancel_msg.name_len = strlen(dlm->name); | |
957 | memcpy(cancel_msg.domain, dlm->name, cancel_msg.name_len); | |
958 | ||
959 | status = o2net_send_message(DLM_CANCEL_JOIN_MSG, DLM_MOD_KEY, | |
960 | &cancel_msg, sizeof(cancel_msg), node, | |
961 | NULL); | |
962 | if (status < 0) { | |
963 | mlog_errno(status); | |
964 | goto bail; | |
965 | } | |
966 | ||
967 | bail: | |
968 | return status; | |
969 | } | |
970 | ||
971 | /* map_size should be in bytes. */ | |
972 | static int dlm_send_join_cancels(struct dlm_ctxt *dlm, | |
973 | unsigned long *node_map, | |
974 | unsigned int map_size) | |
975 | { | |
976 | int status, tmpstat; | |
977 | unsigned int node; | |
978 | ||
979 | if (map_size != (BITS_TO_LONGS(O2NM_MAX_NODES) * | |
980 | sizeof(unsigned long))) { | |
981 | mlog(ML_ERROR, | |
982 | "map_size %u != BITS_TO_LONGS(O2NM_MAX_NODES) %u\n", | |
3a4780a8 | 983 | map_size, (unsigned)BITS_TO_LONGS(O2NM_MAX_NODES)); |
6714d8e8 KH |
984 | return -EINVAL; |
985 | } | |
986 | ||
987 | status = 0; | |
988 | node = -1; | |
989 | while ((node = find_next_bit(node_map, O2NM_MAX_NODES, | |
990 | node + 1)) < O2NM_MAX_NODES) { | |
991 | if (node == dlm->node_num) | |
992 | continue; | |
993 | ||
994 | tmpstat = dlm_send_one_join_cancel(dlm, node); | |
995 | if (tmpstat) { | |
996 | mlog(ML_ERROR, "Error return %d cancelling join on " | |
997 | "node %d\n", tmpstat, node); | |
998 | if (!status) | |
999 | status = tmpstat; | |
1000 | } | |
1001 | } | |
1002 | ||
1003 | if (status) | |
1004 | mlog_errno(status); | |
1005 | return status; | |
1006 | } | |
1007 | ||
1008 | static int dlm_request_join(struct dlm_ctxt *dlm, | |
1009 | int node, | |
d24fbcda | 1010 | enum dlm_query_join_response_code *response) |
6714d8e8 | 1011 | { |
d24fbcda | 1012 | int status; |
6714d8e8 | 1013 | struct dlm_query_join_request join_msg; |
0f71b7b4 JB |
1014 | struct dlm_query_join_packet packet; |
1015 | u32 join_resp; | |
6714d8e8 KH |
1016 | |
1017 | mlog(0, "querying node %d\n", node); | |
1018 | ||
1019 | memset(&join_msg, 0, sizeof(join_msg)); | |
1020 | join_msg.node_idx = dlm->node_num; | |
1021 | join_msg.name_len = strlen(dlm->name); | |
1022 | memcpy(join_msg.domain, dlm->name, join_msg.name_len); | |
d24fbcda JB |
1023 | join_msg.dlm_proto = dlm->dlm_locking_proto; |
1024 | join_msg.fs_proto = dlm->fs_locking_proto; | |
6714d8e8 | 1025 | |
1faf2894 SE |
1026 | /* copy live node map to join message */ |
1027 | byte_copymap(join_msg.node_map, dlm->live_nodes_map, O2NM_MAX_NODES); | |
1028 | ||
6714d8e8 | 1029 | status = o2net_send_message(DLM_QUERY_JOIN_MSG, DLM_MOD_KEY, &join_msg, |
d24fbcda | 1030 | sizeof(join_msg), node, |
0f71b7b4 | 1031 | &join_resp); |
6714d8e8 KH |
1032 | if (status < 0 && status != -ENOPROTOOPT) { |
1033 | mlog_errno(status); | |
1034 | goto bail; | |
1035 | } | |
0f71b7b4 | 1036 | dlm_query_join_wire_to_packet(join_resp, &packet); |
6714d8e8 KH |
1037 | |
1038 | /* -ENOPROTOOPT from the net code means the other side isn't | |
1039 | listening for our message type -- that's fine, it means | |
1040 | his dlm isn't up, so we can consider him a 'yes' but not | |
1041 | joined into the domain. */ | |
1042 | if (status == -ENOPROTOOPT) { | |
1043 | status = 0; | |
1044 | *response = JOIN_OK_NO_MAP; | |
0f71b7b4 JB |
1045 | } else if (packet.code == JOIN_DISALLOW || |
1046 | packet.code == JOIN_OK_NO_MAP) { | |
1047 | *response = packet.code; | |
1048 | } else if (packet.code == JOIN_PROTOCOL_MISMATCH) { | |
d24fbcda JB |
1049 | mlog(ML_NOTICE, |
1050 | "This node requested DLM locking protocol %u.%u and " | |
1051 | "filesystem locking protocol %u.%u. At least one of " | |
1052 | "the protocol versions on node %d is not compatible, " | |
1053 | "disconnecting\n", | |
1054 | dlm->dlm_locking_proto.pv_major, | |
1055 | dlm->dlm_locking_proto.pv_minor, | |
1056 | dlm->fs_locking_proto.pv_major, | |
1057 | dlm->fs_locking_proto.pv_minor, | |
1058 | node); | |
1059 | status = -EPROTO; | |
0f71b7b4 JB |
1060 | *response = packet.code; |
1061 | } else if (packet.code == JOIN_OK) { | |
1062 | *response = packet.code; | |
d24fbcda | 1063 | /* Use the same locking protocol as the remote node */ |
0f71b7b4 JB |
1064 | dlm->dlm_locking_proto.pv_minor = packet.dlm_minor; |
1065 | dlm->fs_locking_proto.pv_minor = packet.fs_minor; | |
d24fbcda JB |
1066 | mlog(0, |
1067 | "Node %d responds JOIN_OK with DLM locking protocol " | |
1068 | "%u.%u and fs locking protocol %u.%u\n", | |
1069 | node, | |
1070 | dlm->dlm_locking_proto.pv_major, | |
1071 | dlm->dlm_locking_proto.pv_minor, | |
1072 | dlm->fs_locking_proto.pv_major, | |
1073 | dlm->fs_locking_proto.pv_minor); | |
6714d8e8 KH |
1074 | } else { |
1075 | status = -EINVAL; | |
d24fbcda | 1076 | mlog(ML_ERROR, "invalid response %d from node %u\n", |
0f71b7b4 | 1077 | packet.code, node); |
6714d8e8 KH |
1078 | } |
1079 | ||
1080 | mlog(0, "status %d, node %d response is %d\n", status, node, | |
0f71b7b4 | 1081 | *response); |
6714d8e8 KH |
1082 | |
1083 | bail: | |
1084 | return status; | |
1085 | } | |
1086 | ||
1087 | static int dlm_send_one_join_assert(struct dlm_ctxt *dlm, | |
1088 | unsigned int node) | |
1089 | { | |
1090 | int status; | |
1091 | struct dlm_assert_joined assert_msg; | |
1092 | ||
1093 | mlog(0, "Sending join assert to node %u\n", node); | |
1094 | ||
1095 | memset(&assert_msg, 0, sizeof(assert_msg)); | |
1096 | assert_msg.node_idx = dlm->node_num; | |
1097 | assert_msg.name_len = strlen(dlm->name); | |
1098 | memcpy(assert_msg.domain, dlm->name, assert_msg.name_len); | |
1099 | ||
1100 | status = o2net_send_message(DLM_ASSERT_JOINED_MSG, DLM_MOD_KEY, | |
1101 | &assert_msg, sizeof(assert_msg), node, | |
1102 | NULL); | |
1103 | if (status < 0) | |
1104 | mlog_errno(status); | |
1105 | ||
1106 | return status; | |
1107 | } | |
1108 | ||
1109 | static void dlm_send_join_asserts(struct dlm_ctxt *dlm, | |
1110 | unsigned long *node_map) | |
1111 | { | |
1112 | int status, node, live; | |
1113 | ||
1114 | status = 0; | |
1115 | node = -1; | |
1116 | while ((node = find_next_bit(node_map, O2NM_MAX_NODES, | |
1117 | node + 1)) < O2NM_MAX_NODES) { | |
1118 | if (node == dlm->node_num) | |
1119 | continue; | |
1120 | ||
1121 | do { | |
1122 | /* It is very important that this message be | |
1123 | * received so we spin until either the node | |
1124 | * has died or it gets the message. */ | |
1125 | status = dlm_send_one_join_assert(dlm, node); | |
1126 | ||
1127 | spin_lock(&dlm->spinlock); | |
1128 | live = test_bit(node, dlm->live_nodes_map); | |
1129 | spin_unlock(&dlm->spinlock); | |
1130 | ||
1131 | if (status) { | |
1132 | mlog(ML_ERROR, "Error return %d asserting " | |
1133 | "join on node %d\n", status, node); | |
1134 | ||
1135 | /* give us some time between errors... */ | |
1136 | if (live) | |
1137 | msleep(DLM_DOMAIN_BACKOFF_MS); | |
1138 | } | |
1139 | } while (status && live); | |
1140 | } | |
1141 | } | |
1142 | ||
1143 | struct domain_join_ctxt { | |
1144 | unsigned long live_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; | |
1145 | unsigned long yes_resp_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; | |
1146 | }; | |
1147 | ||
1148 | static int dlm_should_restart_join(struct dlm_ctxt *dlm, | |
1149 | struct domain_join_ctxt *ctxt, | |
d24fbcda | 1150 | enum dlm_query_join_response_code response) |
6714d8e8 KH |
1151 | { |
1152 | int ret; | |
1153 | ||
1154 | if (response == JOIN_DISALLOW) { | |
1155 | mlog(0, "Latest response of disallow -- should restart\n"); | |
1156 | return 1; | |
1157 | } | |
1158 | ||
1159 | spin_lock(&dlm->spinlock); | |
1160 | /* For now, we restart the process if the node maps have | |
1161 | * changed at all */ | |
1162 | ret = memcmp(ctxt->live_map, dlm->live_nodes_map, | |
1163 | sizeof(dlm->live_nodes_map)); | |
1164 | spin_unlock(&dlm->spinlock); | |
1165 | ||
1166 | if (ret) | |
1167 | mlog(0, "Node maps changed -- should restart\n"); | |
1168 | ||
1169 | return ret; | |
1170 | } | |
1171 | ||
1172 | static int dlm_try_to_join_domain(struct dlm_ctxt *dlm) | |
1173 | { | |
1174 | int status = 0, tmpstat, node; | |
1175 | struct domain_join_ctxt *ctxt; | |
d24fbcda | 1176 | enum dlm_query_join_response_code response = JOIN_DISALLOW; |
6714d8e8 KH |
1177 | |
1178 | mlog_entry("%p", dlm); | |
1179 | ||
cd861280 | 1180 | ctxt = kzalloc(sizeof(*ctxt), GFP_KERNEL); |
6714d8e8 KH |
1181 | if (!ctxt) { |
1182 | status = -ENOMEM; | |
1183 | mlog_errno(status); | |
1184 | goto bail; | |
1185 | } | |
1186 | ||
1187 | /* group sem locking should work for us here -- we're already | |
1188 | * registered for heartbeat events so filling this should be | |
1189 | * atomic wrt getting those handlers called. */ | |
1190 | o2hb_fill_node_map(dlm->live_nodes_map, sizeof(dlm->live_nodes_map)); | |
1191 | ||
1192 | spin_lock(&dlm->spinlock); | |
1193 | memcpy(ctxt->live_map, dlm->live_nodes_map, sizeof(ctxt->live_map)); | |
1194 | ||
1195 | __dlm_set_joining_node(dlm, dlm->node_num); | |
1196 | ||
1197 | spin_unlock(&dlm->spinlock); | |
1198 | ||
1199 | node = -1; | |
1200 | while ((node = find_next_bit(ctxt->live_map, O2NM_MAX_NODES, | |
1201 | node + 1)) < O2NM_MAX_NODES) { | |
1202 | if (node == dlm->node_num) | |
1203 | continue; | |
1204 | ||
1205 | status = dlm_request_join(dlm, node, &response); | |
1206 | if (status < 0) { | |
1207 | mlog_errno(status); | |
1208 | goto bail; | |
1209 | } | |
1210 | ||
1211 | /* Ok, either we got a response or the node doesn't have a | |
1212 | * dlm up. */ | |
1213 | if (response == JOIN_OK) | |
1214 | set_bit(node, ctxt->yes_resp_map); | |
1215 | ||
1216 | if (dlm_should_restart_join(dlm, ctxt, response)) { | |
1217 | status = -EAGAIN; | |
1218 | goto bail; | |
1219 | } | |
1220 | } | |
1221 | ||
1222 | mlog(0, "Yay, done querying nodes!\n"); | |
1223 | ||
1224 | /* Yay, everyone agree's we can join the domain. My domain is | |
1225 | * comprised of all nodes who were put in the | |
1226 | * yes_resp_map. Copy that into our domain map and send a join | |
1227 | * assert message to clean up everyone elses state. */ | |
1228 | spin_lock(&dlm->spinlock); | |
1229 | memcpy(dlm->domain_map, ctxt->yes_resp_map, | |
1230 | sizeof(ctxt->yes_resp_map)); | |
1231 | set_bit(dlm->node_num, dlm->domain_map); | |
1232 | spin_unlock(&dlm->spinlock); | |
1233 | ||
1234 | dlm_send_join_asserts(dlm, ctxt->yes_resp_map); | |
1235 | ||
1236 | /* Joined state *must* be set before the joining node | |
1237 | * information, otherwise the query_join handler may read no | |
1238 | * current joiner but a state of NEW and tell joining nodes | |
1239 | * we're not in the domain. */ | |
1240 | spin_lock(&dlm_domain_lock); | |
1241 | dlm->dlm_state = DLM_CTXT_JOINED; | |
1242 | dlm->num_joins++; | |
1243 | spin_unlock(&dlm_domain_lock); | |
1244 | ||
1245 | bail: | |
1246 | spin_lock(&dlm->spinlock); | |
1247 | __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); | |
1248 | if (!status) | |
1249 | __dlm_print_nodes(dlm); | |
1250 | spin_unlock(&dlm->spinlock); | |
1251 | ||
1252 | if (ctxt) { | |
1253 | /* Do we need to send a cancel message to any nodes? */ | |
1254 | if (status < 0) { | |
1255 | tmpstat = dlm_send_join_cancels(dlm, | |
1256 | ctxt->yes_resp_map, | |
1257 | sizeof(ctxt->yes_resp_map)); | |
1258 | if (tmpstat < 0) | |
1259 | mlog_errno(tmpstat); | |
1260 | } | |
1261 | kfree(ctxt); | |
1262 | } | |
1263 | ||
1264 | mlog(0, "returning %d\n", status); | |
1265 | return status; | |
1266 | } | |
1267 | ||
1268 | static void dlm_unregister_domain_handlers(struct dlm_ctxt *dlm) | |
1269 | { | |
14829422 JB |
1270 | o2hb_unregister_callback(NULL, &dlm->dlm_hb_up); |
1271 | o2hb_unregister_callback(NULL, &dlm->dlm_hb_down); | |
6714d8e8 KH |
1272 | o2net_unregister_handler_list(&dlm->dlm_domain_handlers); |
1273 | } | |
1274 | ||
1275 | static int dlm_register_domain_handlers(struct dlm_ctxt *dlm) | |
1276 | { | |
1277 | int status; | |
1278 | ||
1279 | mlog(0, "registering handlers.\n"); | |
1280 | ||
1281 | o2hb_setup_callback(&dlm->dlm_hb_down, O2HB_NODE_DOWN_CB, | |
1282 | dlm_hb_node_down_cb, dlm, DLM_HB_NODE_DOWN_PRI); | |
14829422 | 1283 | status = o2hb_register_callback(NULL, &dlm->dlm_hb_down); |
6714d8e8 KH |
1284 | if (status) |
1285 | goto bail; | |
1286 | ||
1287 | o2hb_setup_callback(&dlm->dlm_hb_up, O2HB_NODE_UP_CB, | |
1288 | dlm_hb_node_up_cb, dlm, DLM_HB_NODE_UP_PRI); | |
14829422 | 1289 | status = o2hb_register_callback(NULL, &dlm->dlm_hb_up); |
6714d8e8 KH |
1290 | if (status) |
1291 | goto bail; | |
1292 | ||
1293 | status = o2net_register_handler(DLM_MASTER_REQUEST_MSG, dlm->key, | |
1294 | sizeof(struct dlm_master_request), | |
1295 | dlm_master_request_handler, | |
d74c9803 | 1296 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1297 | if (status) |
1298 | goto bail; | |
1299 | ||
1300 | status = o2net_register_handler(DLM_ASSERT_MASTER_MSG, dlm->key, | |
1301 | sizeof(struct dlm_assert_master), | |
1302 | dlm_assert_master_handler, | |
3b8118cf KH |
1303 | dlm, dlm_assert_master_post_handler, |
1304 | &dlm->dlm_domain_handlers); | |
6714d8e8 KH |
1305 | if (status) |
1306 | goto bail; | |
1307 | ||
1308 | status = o2net_register_handler(DLM_CREATE_LOCK_MSG, dlm->key, | |
1309 | sizeof(struct dlm_create_lock), | |
1310 | dlm_create_lock_handler, | |
d74c9803 | 1311 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1312 | if (status) |
1313 | goto bail; | |
1314 | ||
1315 | status = o2net_register_handler(DLM_CONVERT_LOCK_MSG, dlm->key, | |
1316 | DLM_CONVERT_LOCK_MAX_LEN, | |
1317 | dlm_convert_lock_handler, | |
d74c9803 | 1318 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1319 | if (status) |
1320 | goto bail; | |
1321 | ||
1322 | status = o2net_register_handler(DLM_UNLOCK_LOCK_MSG, dlm->key, | |
1323 | DLM_UNLOCK_LOCK_MAX_LEN, | |
1324 | dlm_unlock_lock_handler, | |
d74c9803 | 1325 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1326 | if (status) |
1327 | goto bail; | |
1328 | ||
1329 | status = o2net_register_handler(DLM_PROXY_AST_MSG, dlm->key, | |
1330 | DLM_PROXY_AST_MAX_LEN, | |
1331 | dlm_proxy_ast_handler, | |
d74c9803 | 1332 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1333 | if (status) |
1334 | goto bail; | |
1335 | ||
1336 | status = o2net_register_handler(DLM_EXIT_DOMAIN_MSG, dlm->key, | |
1337 | sizeof(struct dlm_exit_domain), | |
1338 | dlm_exit_domain_handler, | |
d74c9803 | 1339 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1340 | if (status) |
1341 | goto bail; | |
1342 | ||
ba2bf218 KH |
1343 | status = o2net_register_handler(DLM_DEREF_LOCKRES_MSG, dlm->key, |
1344 | sizeof(struct dlm_deref_lockres), | |
1345 | dlm_deref_lockres_handler, | |
d74c9803 | 1346 | dlm, NULL, &dlm->dlm_domain_handlers); |
ba2bf218 KH |
1347 | if (status) |
1348 | goto bail; | |
1349 | ||
6714d8e8 KH |
1350 | status = o2net_register_handler(DLM_MIGRATE_REQUEST_MSG, dlm->key, |
1351 | sizeof(struct dlm_migrate_request), | |
1352 | dlm_migrate_request_handler, | |
d74c9803 | 1353 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1354 | if (status) |
1355 | goto bail; | |
1356 | ||
1357 | status = o2net_register_handler(DLM_MIG_LOCKRES_MSG, dlm->key, | |
1358 | DLM_MIG_LOCKRES_MAX_LEN, | |
1359 | dlm_mig_lockres_handler, | |
d74c9803 | 1360 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1361 | if (status) |
1362 | goto bail; | |
1363 | ||
1364 | status = o2net_register_handler(DLM_MASTER_REQUERY_MSG, dlm->key, | |
1365 | sizeof(struct dlm_master_requery), | |
1366 | dlm_master_requery_handler, | |
d74c9803 | 1367 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1368 | if (status) |
1369 | goto bail; | |
1370 | ||
1371 | status = o2net_register_handler(DLM_LOCK_REQUEST_MSG, dlm->key, | |
1372 | sizeof(struct dlm_lock_request), | |
1373 | dlm_request_all_locks_handler, | |
d74c9803 | 1374 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1375 | if (status) |
1376 | goto bail; | |
1377 | ||
1378 | status = o2net_register_handler(DLM_RECO_DATA_DONE_MSG, dlm->key, | |
1379 | sizeof(struct dlm_reco_data_done), | |
1380 | dlm_reco_data_done_handler, | |
d74c9803 | 1381 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1382 | if (status) |
1383 | goto bail; | |
1384 | ||
1385 | status = o2net_register_handler(DLM_BEGIN_RECO_MSG, dlm->key, | |
1386 | sizeof(struct dlm_begin_reco), | |
1387 | dlm_begin_reco_handler, | |
d74c9803 | 1388 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1389 | if (status) |
1390 | goto bail; | |
1391 | ||
1392 | status = o2net_register_handler(DLM_FINALIZE_RECO_MSG, dlm->key, | |
1393 | sizeof(struct dlm_finalize_reco), | |
1394 | dlm_finalize_reco_handler, | |
d74c9803 | 1395 | dlm, NULL, &dlm->dlm_domain_handlers); |
6714d8e8 KH |
1396 | if (status) |
1397 | goto bail; | |
1398 | ||
1399 | bail: | |
1400 | if (status) | |
1401 | dlm_unregister_domain_handlers(dlm); | |
1402 | ||
1403 | return status; | |
1404 | } | |
1405 | ||
1406 | static int dlm_join_domain(struct dlm_ctxt *dlm) | |
1407 | { | |
1408 | int status; | |
0dd82141 SM |
1409 | unsigned int backoff; |
1410 | unsigned int total_backoff = 0; | |
6714d8e8 KH |
1411 | |
1412 | BUG_ON(!dlm); | |
1413 | ||
1414 | mlog(0, "Join domain %s\n", dlm->name); | |
1415 | ||
1416 | status = dlm_register_domain_handlers(dlm); | |
1417 | if (status) { | |
1418 | mlog_errno(status); | |
1419 | goto bail; | |
1420 | } | |
1421 | ||
007dce53 SM |
1422 | status = dlm_debug_init(dlm); |
1423 | if (status < 0) { | |
1424 | mlog_errno(status); | |
1425 | goto bail; | |
1426 | } | |
1427 | ||
6714d8e8 KH |
1428 | status = dlm_launch_thread(dlm); |
1429 | if (status < 0) { | |
1430 | mlog_errno(status); | |
1431 | goto bail; | |
1432 | } | |
1433 | ||
1434 | status = dlm_launch_recovery_thread(dlm); | |
1435 | if (status < 0) { | |
1436 | mlog_errno(status); | |
1437 | goto bail; | |
1438 | } | |
1439 | ||
3156d267 KH |
1440 | dlm->dlm_worker = create_singlethread_workqueue("dlm_wq"); |
1441 | if (!dlm->dlm_worker) { | |
1442 | status = -ENOMEM; | |
1443 | mlog_errno(status); | |
1444 | goto bail; | |
1445 | } | |
1446 | ||
6714d8e8 | 1447 | do { |
6714d8e8 KH |
1448 | status = dlm_try_to_join_domain(dlm); |
1449 | ||
1450 | /* If we're racing another node to the join, then we | |
1451 | * need to back off temporarily and let them | |
1452 | * complete. */ | |
0dd82141 | 1453 | #define DLM_JOIN_TIMEOUT_MSECS 90000 |
6714d8e8 KH |
1454 | if (status == -EAGAIN) { |
1455 | if (signal_pending(current)) { | |
1456 | status = -ERESTARTSYS; | |
1457 | goto bail; | |
1458 | } | |
1459 | ||
0dd82141 SM |
1460 | if (total_backoff > |
1461 | msecs_to_jiffies(DLM_JOIN_TIMEOUT_MSECS)) { | |
1462 | status = -ERESTARTSYS; | |
1463 | mlog(ML_NOTICE, "Timed out joining dlm domain " | |
1464 | "%s after %u msecs\n", dlm->name, | |
1465 | jiffies_to_msecs(total_backoff)); | |
1466 | goto bail; | |
1467 | } | |
1468 | ||
6714d8e8 KH |
1469 | /* |
1470 | * <chip> After you! | |
1471 | * <dale> No, after you! | |
1472 | * <chip> I insist! | |
1473 | * <dale> But you first! | |
1474 | * ... | |
1475 | */ | |
1476 | backoff = (unsigned int)(jiffies & 0x3); | |
1477 | backoff *= DLM_DOMAIN_BACKOFF_MS; | |
0dd82141 | 1478 | total_backoff += backoff; |
6714d8e8 KH |
1479 | mlog(0, "backoff %d\n", backoff); |
1480 | msleep(backoff); | |
1481 | } | |
1482 | } while (status == -EAGAIN); | |
1483 | ||
1484 | if (status < 0) { | |
1485 | mlog_errno(status); | |
1486 | goto bail; | |
1487 | } | |
1488 | ||
1489 | status = 0; | |
1490 | bail: | |
1491 | wake_up(&dlm_domain_events); | |
1492 | ||
1493 | if (status) { | |
1494 | dlm_unregister_domain_handlers(dlm); | |
007dce53 | 1495 | dlm_debug_shutdown(dlm); |
6714d8e8 KH |
1496 | dlm_complete_thread(dlm); |
1497 | dlm_complete_recovery_thread(dlm); | |
3156d267 | 1498 | dlm_destroy_dlm_worker(dlm); |
6714d8e8 KH |
1499 | } |
1500 | ||
1501 | return status; | |
1502 | } | |
1503 | ||
1504 | static struct dlm_ctxt *dlm_alloc_ctxt(const char *domain, | |
1505 | u32 key) | |
1506 | { | |
1507 | int i; | |
6325b4a2 | 1508 | int ret; |
6714d8e8 KH |
1509 | struct dlm_ctxt *dlm = NULL; |
1510 | ||
cd861280 | 1511 | dlm = kzalloc(sizeof(*dlm), GFP_KERNEL); |
6714d8e8 KH |
1512 | if (!dlm) { |
1513 | mlog_errno(-ENOMEM); | |
1514 | goto leave; | |
1515 | } | |
1516 | ||
1517 | dlm->name = kmalloc(strlen(domain) + 1, GFP_KERNEL); | |
1518 | if (dlm->name == NULL) { | |
1519 | mlog_errno(-ENOMEM); | |
1520 | kfree(dlm); | |
1521 | dlm = NULL; | |
1522 | goto leave; | |
1523 | } | |
1524 | ||
03d864c0 | 1525 | dlm->lockres_hash = (struct hlist_head **)dlm_alloc_pagevec(DLM_HASH_PAGES); |
81f2094a | 1526 | if (!dlm->lockres_hash) { |
6714d8e8 KH |
1527 | mlog_errno(-ENOMEM); |
1528 | kfree(dlm->name); | |
1529 | kfree(dlm); | |
1530 | dlm = NULL; | |
1531 | goto leave; | |
1532 | } | |
6714d8e8 | 1533 | |
03d864c0 DP |
1534 | for (i = 0; i < DLM_HASH_BUCKETS; i++) |
1535 | INIT_HLIST_HEAD(dlm_lockres_hash(dlm, i)); | |
6714d8e8 KH |
1536 | |
1537 | strcpy(dlm->name, domain); | |
1538 | dlm->key = key; | |
1539 | dlm->node_num = o2nm_this_node(); | |
1540 | ||
6325b4a2 SM |
1541 | ret = dlm_create_debugfs_subroot(dlm); |
1542 | if (ret < 0) { | |
1543 | dlm_free_pagevec((void **)dlm->lockres_hash, DLM_HASH_PAGES); | |
1544 | kfree(dlm->name); | |
1545 | kfree(dlm); | |
1546 | dlm = NULL; | |
1547 | goto leave; | |
1548 | } | |
1549 | ||
6714d8e8 KH |
1550 | spin_lock_init(&dlm->spinlock); |
1551 | spin_lock_init(&dlm->master_lock); | |
1552 | spin_lock_init(&dlm->ast_lock); | |
1553 | INIT_LIST_HEAD(&dlm->list); | |
1554 | INIT_LIST_HEAD(&dlm->dirty_list); | |
1555 | INIT_LIST_HEAD(&dlm->reco.resources); | |
1556 | INIT_LIST_HEAD(&dlm->reco.received); | |
1557 | INIT_LIST_HEAD(&dlm->reco.node_data); | |
1558 | INIT_LIST_HEAD(&dlm->purge_list); | |
1559 | INIT_LIST_HEAD(&dlm->dlm_domain_handlers); | |
29576f8b | 1560 | INIT_LIST_HEAD(&dlm->tracking_list); |
6714d8e8 KH |
1561 | dlm->reco.state = 0; |
1562 | ||
1563 | INIT_LIST_HEAD(&dlm->pending_asts); | |
1564 | INIT_LIST_HEAD(&dlm->pending_basts); | |
1565 | ||
1566 | mlog(0, "dlm->recovery_map=%p, &(dlm->recovery_map[0])=%p\n", | |
1567 | dlm->recovery_map, &(dlm->recovery_map[0])); | |
1568 | ||
1569 | memset(dlm->recovery_map, 0, sizeof(dlm->recovery_map)); | |
1570 | memset(dlm->live_nodes_map, 0, sizeof(dlm->live_nodes_map)); | |
1571 | memset(dlm->domain_map, 0, sizeof(dlm->domain_map)); | |
1572 | ||
1573 | dlm->dlm_thread_task = NULL; | |
1574 | dlm->dlm_reco_thread_task = NULL; | |
3156d267 | 1575 | dlm->dlm_worker = NULL; |
6714d8e8 KH |
1576 | init_waitqueue_head(&dlm->dlm_thread_wq); |
1577 | init_waitqueue_head(&dlm->dlm_reco_thread_wq); | |
1578 | init_waitqueue_head(&dlm->reco.event); | |
1579 | init_waitqueue_head(&dlm->ast_wq); | |
1580 | init_waitqueue_head(&dlm->migration_wq); | |
1581 | INIT_LIST_HEAD(&dlm->master_list); | |
1582 | INIT_LIST_HEAD(&dlm->mle_hb_events); | |
1583 | ||
1584 | dlm->joining_node = DLM_LOCK_RES_OWNER_UNKNOWN; | |
1585 | init_waitqueue_head(&dlm->dlm_join_events); | |
1586 | ||
1587 | dlm->reco.new_master = O2NM_INVALID_NODE_NUM; | |
1588 | dlm->reco.dead_node = O2NM_INVALID_NODE_NUM; | |
1589 | atomic_set(&dlm->local_resources, 0); | |
1590 | atomic_set(&dlm->remote_resources, 0); | |
1591 | atomic_set(&dlm->unknown_resources, 0); | |
1592 | ||
1593 | spin_lock_init(&dlm->work_lock); | |
1594 | INIT_LIST_HEAD(&dlm->work_list); | |
c4028958 | 1595 | INIT_WORK(&dlm->dispatched_work, dlm_dispatch_work); |
6714d8e8 KH |
1596 | |
1597 | kref_init(&dlm->dlm_refs); | |
1598 | dlm->dlm_state = DLM_CTXT_NEW; | |
1599 | ||
1600 | INIT_LIST_HEAD(&dlm->dlm_eviction_callbacks); | |
1601 | ||
1602 | mlog(0, "context init: refcount %u\n", | |
1603 | atomic_read(&dlm->dlm_refs.refcount)); | |
1604 | ||
1605 | leave: | |
1606 | return dlm; | |
1607 | } | |
1608 | ||
1609 | /* | |
d24fbcda JB |
1610 | * Compare a requested locking protocol version against the current one. |
1611 | * | |
1612 | * If the major numbers are different, they are incompatible. | |
1613 | * If the current minor is greater than the request, they are incompatible. | |
1614 | * If the current minor is less than or equal to the request, they are | |
1615 | * compatible, and the requester should run at the current minor version. | |
1616 | */ | |
1617 | static int dlm_protocol_compare(struct dlm_protocol_version *existing, | |
1618 | struct dlm_protocol_version *request) | |
1619 | { | |
1620 | if (existing->pv_major != request->pv_major) | |
1621 | return 1; | |
1622 | ||
1623 | if (existing->pv_minor > request->pv_minor) | |
1624 | return 1; | |
1625 | ||
1626 | if (existing->pv_minor < request->pv_minor) | |
1627 | request->pv_minor = existing->pv_minor; | |
1628 | ||
1629 | return 0; | |
1630 | } | |
1631 | ||
1632 | /* | |
1633 | * dlm_register_domain: one-time setup per "domain". | |
1634 | * | |
1635 | * The filesystem passes in the requested locking version via proto. | |
1636 | * If registration was successful, proto will contain the negotiated | |
1637 | * locking protocol. | |
6714d8e8 KH |
1638 | */ |
1639 | struct dlm_ctxt * dlm_register_domain(const char *domain, | |
d24fbcda JB |
1640 | u32 key, |
1641 | struct dlm_protocol_version *fs_proto) | |
6714d8e8 KH |
1642 | { |
1643 | int ret; | |
1644 | struct dlm_ctxt *dlm = NULL; | |
1645 | struct dlm_ctxt *new_ctxt = NULL; | |
1646 | ||
1647 | if (strlen(domain) > O2NM_MAX_NAME_LEN) { | |
1648 | ret = -ENAMETOOLONG; | |
1649 | mlog(ML_ERROR, "domain name length too long\n"); | |
1650 | goto leave; | |
1651 | } | |
1652 | ||
1653 | if (!o2hb_check_local_node_heartbeating()) { | |
1654 | mlog(ML_ERROR, "the local node has not been configured, or is " | |
1655 | "not heartbeating\n"); | |
1656 | ret = -EPROTO; | |
1657 | goto leave; | |
1658 | } | |
1659 | ||
1660 | mlog(0, "register called for domain \"%s\"\n", domain); | |
1661 | ||
1662 | retry: | |
1663 | dlm = NULL; | |
1664 | if (signal_pending(current)) { | |
1665 | ret = -ERESTARTSYS; | |
1666 | mlog_errno(ret); | |
1667 | goto leave; | |
1668 | } | |
1669 | ||
1670 | spin_lock(&dlm_domain_lock); | |
1671 | ||
1672 | dlm = __dlm_lookup_domain(domain); | |
1673 | if (dlm) { | |
1674 | if (dlm->dlm_state != DLM_CTXT_JOINED) { | |
1675 | spin_unlock(&dlm_domain_lock); | |
1676 | ||
1677 | mlog(0, "This ctxt is not joined yet!\n"); | |
1678 | wait_event_interruptible(dlm_domain_events, | |
1679 | dlm_wait_on_domain_helper( | |
1680 | domain)); | |
1681 | goto retry; | |
1682 | } | |
1683 | ||
d24fbcda JB |
1684 | if (dlm_protocol_compare(&dlm->fs_locking_proto, fs_proto)) { |
1685 | mlog(ML_ERROR, | |
1686 | "Requested locking protocol version is not " | |
1687 | "compatible with already registered domain " | |
1688 | "\"%s\"\n", domain); | |
1689 | ret = -EPROTO; | |
1690 | goto leave; | |
1691 | } | |
1692 | ||
6714d8e8 KH |
1693 | __dlm_get(dlm); |
1694 | dlm->num_joins++; | |
1695 | ||
1696 | spin_unlock(&dlm_domain_lock); | |
1697 | ||
1698 | ret = 0; | |
1699 | goto leave; | |
1700 | } | |
1701 | ||
1702 | /* doesn't exist */ | |
1703 | if (!new_ctxt) { | |
1704 | spin_unlock(&dlm_domain_lock); | |
1705 | ||
1706 | new_ctxt = dlm_alloc_ctxt(domain, key); | |
1707 | if (new_ctxt) | |
1708 | goto retry; | |
1709 | ||
1710 | ret = -ENOMEM; | |
1711 | mlog_errno(ret); | |
1712 | goto leave; | |
1713 | } | |
1714 | ||
1715 | /* a little variable switch-a-roo here... */ | |
1716 | dlm = new_ctxt; | |
1717 | new_ctxt = NULL; | |
1718 | ||
1719 | /* add the new domain */ | |
1720 | list_add_tail(&dlm->list, &dlm_domains); | |
1721 | spin_unlock(&dlm_domain_lock); | |
1722 | ||
d24fbcda JB |
1723 | /* |
1724 | * Pass the locking protocol version into the join. If the join | |
1725 | * succeeds, it will have the negotiated protocol set. | |
1726 | */ | |
1727 | dlm->dlm_locking_proto = dlm_protocol; | |
1728 | dlm->fs_locking_proto = *fs_proto; | |
1729 | ||
6714d8e8 KH |
1730 | ret = dlm_join_domain(dlm); |
1731 | if (ret) { | |
1732 | mlog_errno(ret); | |
1733 | dlm_put(dlm); | |
1734 | goto leave; | |
1735 | } | |
1736 | ||
d24fbcda JB |
1737 | /* Tell the caller what locking protocol we negotiated */ |
1738 | *fs_proto = dlm->fs_locking_proto; | |
1739 | ||
6714d8e8 KH |
1740 | ret = 0; |
1741 | leave: | |
1742 | if (new_ctxt) | |
1743 | dlm_free_ctxt_mem(new_ctxt); | |
1744 | ||
1745 | if (ret < 0) | |
1746 | dlm = ERR_PTR(ret); | |
1747 | ||
1748 | return dlm; | |
1749 | } | |
1750 | EXPORT_SYMBOL_GPL(dlm_register_domain); | |
1751 | ||
1752 | static LIST_HEAD(dlm_join_handlers); | |
1753 | ||
1754 | static void dlm_unregister_net_handlers(void) | |
1755 | { | |
1756 | o2net_unregister_handler_list(&dlm_join_handlers); | |
1757 | } | |
1758 | ||
1759 | static int dlm_register_net_handlers(void) | |
1760 | { | |
1761 | int status = 0; | |
1762 | ||
1763 | status = o2net_register_handler(DLM_QUERY_JOIN_MSG, DLM_MOD_KEY, | |
1764 | sizeof(struct dlm_query_join_request), | |
1765 | dlm_query_join_handler, | |
d74c9803 | 1766 | NULL, NULL, &dlm_join_handlers); |
6714d8e8 KH |
1767 | if (status) |
1768 | goto bail; | |
1769 | ||
1770 | status = o2net_register_handler(DLM_ASSERT_JOINED_MSG, DLM_MOD_KEY, | |
1771 | sizeof(struct dlm_assert_joined), | |
1772 | dlm_assert_joined_handler, | |
d74c9803 | 1773 | NULL, NULL, &dlm_join_handlers); |
6714d8e8 KH |
1774 | if (status) |
1775 | goto bail; | |
1776 | ||
1777 | status = o2net_register_handler(DLM_CANCEL_JOIN_MSG, DLM_MOD_KEY, | |
1778 | sizeof(struct dlm_cancel_join), | |
1779 | dlm_cancel_join_handler, | |
d74c9803 | 1780 | NULL, NULL, &dlm_join_handlers); |
6714d8e8 KH |
1781 | |
1782 | bail: | |
1783 | if (status < 0) | |
1784 | dlm_unregister_net_handlers(); | |
1785 | ||
1786 | return status; | |
1787 | } | |
1788 | ||
1789 | /* Domain eviction callback handling. | |
1790 | * | |
1791 | * The file system requires notification of node death *before* the | |
1792 | * dlm completes it's recovery work, otherwise it may be able to | |
1793 | * acquire locks on resources requiring recovery. Since the dlm can | |
1794 | * evict a node from it's domain *before* heartbeat fires, a similar | |
1795 | * mechanism is required. */ | |
1796 | ||
1797 | /* Eviction is not expected to happen often, so a per-domain lock is | |
1798 | * not necessary. Eviction callbacks are allowed to sleep for short | |
1799 | * periods of time. */ | |
1800 | static DECLARE_RWSEM(dlm_callback_sem); | |
1801 | ||
1802 | void dlm_fire_domain_eviction_callbacks(struct dlm_ctxt *dlm, | |
1803 | int node_num) | |
1804 | { | |
1805 | struct list_head *iter; | |
1806 | struct dlm_eviction_cb *cb; | |
1807 | ||
1808 | down_read(&dlm_callback_sem); | |
1809 | list_for_each(iter, &dlm->dlm_eviction_callbacks) { | |
1810 | cb = list_entry(iter, struct dlm_eviction_cb, ec_item); | |
1811 | ||
1812 | cb->ec_func(node_num, cb->ec_data); | |
1813 | } | |
1814 | up_read(&dlm_callback_sem); | |
1815 | } | |
1816 | ||
1817 | void dlm_setup_eviction_cb(struct dlm_eviction_cb *cb, | |
1818 | dlm_eviction_func *f, | |
1819 | void *data) | |
1820 | { | |
1821 | INIT_LIST_HEAD(&cb->ec_item); | |
1822 | cb->ec_func = f; | |
1823 | cb->ec_data = data; | |
1824 | } | |
1825 | EXPORT_SYMBOL_GPL(dlm_setup_eviction_cb); | |
1826 | ||
1827 | void dlm_register_eviction_cb(struct dlm_ctxt *dlm, | |
1828 | struct dlm_eviction_cb *cb) | |
1829 | { | |
1830 | down_write(&dlm_callback_sem); | |
1831 | list_add_tail(&cb->ec_item, &dlm->dlm_eviction_callbacks); | |
1832 | up_write(&dlm_callback_sem); | |
1833 | } | |
1834 | EXPORT_SYMBOL_GPL(dlm_register_eviction_cb); | |
1835 | ||
1836 | void dlm_unregister_eviction_cb(struct dlm_eviction_cb *cb) | |
1837 | { | |
1838 | down_write(&dlm_callback_sem); | |
1839 | list_del_init(&cb->ec_item); | |
1840 | up_write(&dlm_callback_sem); | |
1841 | } | |
1842 | EXPORT_SYMBOL_GPL(dlm_unregister_eviction_cb); | |
1843 | ||
1844 | static int __init dlm_init(void) | |
1845 | { | |
1846 | int status; | |
1847 | ||
1848 | dlm_print_version(); | |
1849 | ||
1850 | status = dlm_init_mle_cache(); | |
12eb0035 SM |
1851 | if (status) { |
1852 | mlog(ML_ERROR, "Could not create o2dlm_mle slabcache\n"); | |
724bdca9 SM |
1853 | goto error; |
1854 | } | |
1855 | ||
1856 | status = dlm_init_master_caches(); | |
1857 | if (status) { | |
1858 | mlog(ML_ERROR, "Could not create o2dlm_lockres and " | |
1859 | "o2dlm_lockname slabcaches\n"); | |
1860 | goto error; | |
1861 | } | |
1862 | ||
1863 | status = dlm_init_lock_cache(); | |
1864 | if (status) { | |
1865 | mlog(ML_ERROR, "Count not create o2dlm_lock slabcache\n"); | |
1866 | goto error; | |
12eb0035 | 1867 | } |
6714d8e8 KH |
1868 | |
1869 | status = dlm_register_net_handlers(); | |
1870 | if (status) { | |
724bdca9 SM |
1871 | mlog(ML_ERROR, "Unable to register network handlers\n"); |
1872 | goto error; | |
6714d8e8 KH |
1873 | } |
1874 | ||
6325b4a2 SM |
1875 | status = dlm_create_debugfs_root(); |
1876 | if (status) | |
1877 | goto error; | |
1878 | ||
6714d8e8 | 1879 | return 0; |
724bdca9 | 1880 | error: |
6325b4a2 | 1881 | dlm_unregister_net_handlers(); |
724bdca9 SM |
1882 | dlm_destroy_lock_cache(); |
1883 | dlm_destroy_master_caches(); | |
1884 | dlm_destroy_mle_cache(); | |
1885 | return -1; | |
6714d8e8 KH |
1886 | } |
1887 | ||
1888 | static void __exit dlm_exit (void) | |
1889 | { | |
6325b4a2 | 1890 | dlm_destroy_debugfs_root(); |
6714d8e8 | 1891 | dlm_unregister_net_handlers(); |
724bdca9 SM |
1892 | dlm_destroy_lock_cache(); |
1893 | dlm_destroy_master_caches(); | |
6714d8e8 KH |
1894 | dlm_destroy_mle_cache(); |
1895 | } | |
1896 | ||
1897 | MODULE_AUTHOR("Oracle"); | |
1898 | MODULE_LICENSE("GPL"); | |
1899 | ||
1900 | module_init(dlm_init); | |
1901 | module_exit(dlm_exit); |