Loading...
1/******************************************************************************
2*******************************************************************************
3**
4** Copyright (C) 2005-2010 Red Hat, Inc. All rights reserved.
5**
6** This copyrighted material is made available to anyone wishing to use,
7** modify, copy, or redistribute it subject to the terms and conditions
8** of the GNU General Public License v.2.
9**
10*******************************************************************************
11******************************************************************************/
12
13/* Central locking logic has four stages:
14
15 dlm_lock()
16 dlm_unlock()
17
18 request_lock(ls, lkb)
19 convert_lock(ls, lkb)
20 unlock_lock(ls, lkb)
21 cancel_lock(ls, lkb)
22
23 _request_lock(r, lkb)
24 _convert_lock(r, lkb)
25 _unlock_lock(r, lkb)
26 _cancel_lock(r, lkb)
27
28 do_request(r, lkb)
29 do_convert(r, lkb)
30 do_unlock(r, lkb)
31 do_cancel(r, lkb)
32
33 Stage 1 (lock, unlock) is mainly about checking input args and
34 splitting into one of the four main operations:
35
36 dlm_lock = request_lock
37 dlm_lock+CONVERT = convert_lock
38 dlm_unlock = unlock_lock
39 dlm_unlock+CANCEL = cancel_lock
40
41 Stage 2, xxxx_lock(), just finds and locks the relevant rsb which is
42 provided to the next stage.
43
44 Stage 3, _xxxx_lock(), determines if the operation is local or remote.
45 When remote, it calls send_xxxx(), when local it calls do_xxxx().
46
47 Stage 4, do_xxxx(), is the guts of the operation. It manipulates the
48 given rsb and lkb and queues callbacks.
49
50 For remote operations, send_xxxx() results in the corresponding do_xxxx()
51 function being executed on the remote node. The connecting send/receive
52 calls on local (L) and remote (R) nodes:
53
54 L: send_xxxx() -> R: receive_xxxx()
55 R: do_xxxx()
56 L: receive_xxxx_reply() <- R: send_xxxx_reply()
57*/
58#include <linux/types.h>
59#include <linux/slab.h>
60#include "dlm_internal.h"
61#include <linux/dlm_device.h>
62#include "memory.h"
63#include "lowcomms.h"
64#include "requestqueue.h"
65#include "util.h"
66#include "dir.h"
67#include "member.h"
68#include "lockspace.h"
69#include "ast.h"
70#include "lock.h"
71#include "rcom.h"
72#include "recover.h"
73#include "lvb_table.h"
74#include "user.h"
75#include "config.h"
76
77static int send_request(struct dlm_rsb *r, struct dlm_lkb *lkb);
78static int send_convert(struct dlm_rsb *r, struct dlm_lkb *lkb);
79static int send_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb);
80static int send_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb);
81static int send_grant(struct dlm_rsb *r, struct dlm_lkb *lkb);
82static int send_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int mode);
83static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb);
84static int send_remove(struct dlm_rsb *r);
85static int _request_lock(struct dlm_rsb *r, struct dlm_lkb *lkb);
86static int _cancel_lock(struct dlm_rsb *r, struct dlm_lkb *lkb);
87static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
88 struct dlm_message *ms);
89static int receive_extralen(struct dlm_message *ms);
90static void do_purge(struct dlm_ls *ls, int nodeid, int pid);
91static void del_timeout(struct dlm_lkb *lkb);
92
93/*
94 * Lock compatibilty matrix - thanks Steve
95 * UN = Unlocked state. Not really a state, used as a flag
96 * PD = Padding. Used to make the matrix a nice power of two in size
97 * Other states are the same as the VMS DLM.
98 * Usage: matrix[grmode+1][rqmode+1] (although m[rq+1][gr+1] is the same)
99 */
100
101static const int __dlm_compat_matrix[8][8] = {
102 /* UN NL CR CW PR PW EX PD */
103 {1, 1, 1, 1, 1, 1, 1, 0}, /* UN */
104 {1, 1, 1, 1, 1, 1, 1, 0}, /* NL */
105 {1, 1, 1, 1, 1, 1, 0, 0}, /* CR */
106 {1, 1, 1, 1, 0, 0, 0, 0}, /* CW */
107 {1, 1, 1, 0, 1, 0, 0, 0}, /* PR */
108 {1, 1, 1, 0, 0, 0, 0, 0}, /* PW */
109 {1, 1, 0, 0, 0, 0, 0, 0}, /* EX */
110 {0, 0, 0, 0, 0, 0, 0, 0} /* PD */
111};
112
113/*
114 * This defines the direction of transfer of LVB data.
115 * Granted mode is the row; requested mode is the column.
116 * Usage: matrix[grmode+1][rqmode+1]
117 * 1 = LVB is returned to the caller
118 * 0 = LVB is written to the resource
119 * -1 = nothing happens to the LVB
120 */
121
122const int dlm_lvb_operations[8][8] = {
123 /* UN NL CR CW PR PW EX PD*/
124 { -1, 1, 1, 1, 1, 1, 1, -1 }, /* UN */
125 { -1, 1, 1, 1, 1, 1, 1, 0 }, /* NL */
126 { -1, -1, 1, 1, 1, 1, 1, 0 }, /* CR */
127 { -1, -1, -1, 1, 1, 1, 1, 0 }, /* CW */
128 { -1, -1, -1, -1, 1, 1, 1, 0 }, /* PR */
129 { -1, 0, 0, 0, 0, 0, 1, 0 }, /* PW */
130 { -1, 0, 0, 0, 0, 0, 0, 0 }, /* EX */
131 { -1, 0, 0, 0, 0, 0, 0, 0 } /* PD */
132};
133
134#define modes_compat(gr, rq) \
135 __dlm_compat_matrix[(gr)->lkb_grmode + 1][(rq)->lkb_rqmode + 1]
136
137int dlm_modes_compat(int mode1, int mode2)
138{
139 return __dlm_compat_matrix[mode1 + 1][mode2 + 1];
140}
141
142/*
143 * Compatibility matrix for conversions with QUECVT set.
144 * Granted mode is the row; requested mode is the column.
145 * Usage: matrix[grmode+1][rqmode+1]
146 */
147
148static const int __quecvt_compat_matrix[8][8] = {
149 /* UN NL CR CW PR PW EX PD */
150 {0, 0, 0, 0, 0, 0, 0, 0}, /* UN */
151 {0, 0, 1, 1, 1, 1, 1, 0}, /* NL */
152 {0, 0, 0, 1, 1, 1, 1, 0}, /* CR */
153 {0, 0, 0, 0, 1, 1, 1, 0}, /* CW */
154 {0, 0, 0, 1, 0, 1, 1, 0}, /* PR */
155 {0, 0, 0, 0, 0, 0, 1, 0}, /* PW */
156 {0, 0, 0, 0, 0, 0, 0, 0}, /* EX */
157 {0, 0, 0, 0, 0, 0, 0, 0} /* PD */
158};
159
160void dlm_print_lkb(struct dlm_lkb *lkb)
161{
162 printk(KERN_ERR "lkb: nodeid %d id %x remid %x exflags %x flags %x\n"
163 " status %d rqmode %d grmode %d wait_type %d\n",
164 lkb->lkb_nodeid, lkb->lkb_id, lkb->lkb_remid, lkb->lkb_exflags,
165 lkb->lkb_flags, lkb->lkb_status, lkb->lkb_rqmode,
166 lkb->lkb_grmode, lkb->lkb_wait_type);
167}
168
169static void dlm_print_rsb(struct dlm_rsb *r)
170{
171 printk(KERN_ERR "rsb: nodeid %d flags %lx first %x rlc %d name %s\n",
172 r->res_nodeid, r->res_flags, r->res_first_lkid,
173 r->res_recover_locks_count, r->res_name);
174}
175
176void dlm_dump_rsb(struct dlm_rsb *r)
177{
178 struct dlm_lkb *lkb;
179
180 dlm_print_rsb(r);
181
182 printk(KERN_ERR "rsb: root_list empty %d recover_list empty %d\n",
183 list_empty(&r->res_root_list), list_empty(&r->res_recover_list));
184 printk(KERN_ERR "rsb lookup list\n");
185 list_for_each_entry(lkb, &r->res_lookup, lkb_rsb_lookup)
186 dlm_print_lkb(lkb);
187 printk(KERN_ERR "rsb grant queue:\n");
188 list_for_each_entry(lkb, &r->res_grantqueue, lkb_statequeue)
189 dlm_print_lkb(lkb);
190 printk(KERN_ERR "rsb convert queue:\n");
191 list_for_each_entry(lkb, &r->res_convertqueue, lkb_statequeue)
192 dlm_print_lkb(lkb);
193 printk(KERN_ERR "rsb wait queue:\n");
194 list_for_each_entry(lkb, &r->res_waitqueue, lkb_statequeue)
195 dlm_print_lkb(lkb);
196}
197
198/* Threads cannot use the lockspace while it's being recovered */
199
200static inline void dlm_lock_recovery(struct dlm_ls *ls)
201{
202 down_read(&ls->ls_in_recovery);
203}
204
205void dlm_unlock_recovery(struct dlm_ls *ls)
206{
207 up_read(&ls->ls_in_recovery);
208}
209
210int dlm_lock_recovery_try(struct dlm_ls *ls)
211{
212 return down_read_trylock(&ls->ls_in_recovery);
213}
214
215static inline int can_be_queued(struct dlm_lkb *lkb)
216{
217 return !(lkb->lkb_exflags & DLM_LKF_NOQUEUE);
218}
219
220static inline int force_blocking_asts(struct dlm_lkb *lkb)
221{
222 return (lkb->lkb_exflags & DLM_LKF_NOQUEUEBAST);
223}
224
225static inline int is_demoted(struct dlm_lkb *lkb)
226{
227 return (lkb->lkb_sbflags & DLM_SBF_DEMOTED);
228}
229
230static inline int is_altmode(struct dlm_lkb *lkb)
231{
232 return (lkb->lkb_sbflags & DLM_SBF_ALTMODE);
233}
234
235static inline int is_granted(struct dlm_lkb *lkb)
236{
237 return (lkb->lkb_status == DLM_LKSTS_GRANTED);
238}
239
240static inline int is_remote(struct dlm_rsb *r)
241{
242 DLM_ASSERT(r->res_nodeid >= 0, dlm_print_rsb(r););
243 return !!r->res_nodeid;
244}
245
246static inline int is_process_copy(struct dlm_lkb *lkb)
247{
248 return (lkb->lkb_nodeid && !(lkb->lkb_flags & DLM_IFL_MSTCPY));
249}
250
251static inline int is_master_copy(struct dlm_lkb *lkb)
252{
253 if (lkb->lkb_flags & DLM_IFL_MSTCPY)
254 DLM_ASSERT(lkb->lkb_nodeid, dlm_print_lkb(lkb););
255 return (lkb->lkb_flags & DLM_IFL_MSTCPY) ? 1 : 0;
256}
257
258static inline int middle_conversion(struct dlm_lkb *lkb)
259{
260 if ((lkb->lkb_grmode==DLM_LOCK_PR && lkb->lkb_rqmode==DLM_LOCK_CW) ||
261 (lkb->lkb_rqmode==DLM_LOCK_PR && lkb->lkb_grmode==DLM_LOCK_CW))
262 return 1;
263 return 0;
264}
265
266static inline int down_conversion(struct dlm_lkb *lkb)
267{
268 return (!middle_conversion(lkb) && lkb->lkb_rqmode < lkb->lkb_grmode);
269}
270
271static inline int is_overlap_unlock(struct dlm_lkb *lkb)
272{
273 return lkb->lkb_flags & DLM_IFL_OVERLAP_UNLOCK;
274}
275
276static inline int is_overlap_cancel(struct dlm_lkb *lkb)
277{
278 return lkb->lkb_flags & DLM_IFL_OVERLAP_CANCEL;
279}
280
281static inline int is_overlap(struct dlm_lkb *lkb)
282{
283 return (lkb->lkb_flags & (DLM_IFL_OVERLAP_UNLOCK |
284 DLM_IFL_OVERLAP_CANCEL));
285}
286
287static void queue_cast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
288{
289 if (is_master_copy(lkb))
290 return;
291
292 del_timeout(lkb);
293
294 DLM_ASSERT(lkb->lkb_lksb, dlm_print_lkb(lkb););
295
296 /* if the operation was a cancel, then return -DLM_ECANCEL, if a
297 timeout caused the cancel then return -ETIMEDOUT */
298 if (rv == -DLM_ECANCEL && (lkb->lkb_flags & DLM_IFL_TIMEOUT_CANCEL)) {
299 lkb->lkb_flags &= ~DLM_IFL_TIMEOUT_CANCEL;
300 rv = -ETIMEDOUT;
301 }
302
303 if (rv == -DLM_ECANCEL && (lkb->lkb_flags & DLM_IFL_DEADLOCK_CANCEL)) {
304 lkb->lkb_flags &= ~DLM_IFL_DEADLOCK_CANCEL;
305 rv = -EDEADLK;
306 }
307
308 dlm_add_cb(lkb, DLM_CB_CAST, lkb->lkb_grmode, rv, lkb->lkb_sbflags);
309}
310
311static inline void queue_cast_overlap(struct dlm_rsb *r, struct dlm_lkb *lkb)
312{
313 queue_cast(r, lkb,
314 is_overlap_unlock(lkb) ? -DLM_EUNLOCK : -DLM_ECANCEL);
315}
316
317static void queue_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rqmode)
318{
319 if (is_master_copy(lkb)) {
320 send_bast(r, lkb, rqmode);
321 } else {
322 dlm_add_cb(lkb, DLM_CB_BAST, rqmode, 0, 0);
323 }
324}
325
326/*
327 * Basic operations on rsb's and lkb's
328 */
329
330static int pre_rsb_struct(struct dlm_ls *ls)
331{
332 struct dlm_rsb *r1, *r2;
333 int count = 0;
334
335 spin_lock(&ls->ls_new_rsb_spin);
336 if (ls->ls_new_rsb_count > dlm_config.ci_new_rsb_count / 2) {
337 spin_unlock(&ls->ls_new_rsb_spin);
338 return 0;
339 }
340 spin_unlock(&ls->ls_new_rsb_spin);
341
342 r1 = dlm_allocate_rsb(ls);
343 r2 = dlm_allocate_rsb(ls);
344
345 spin_lock(&ls->ls_new_rsb_spin);
346 if (r1) {
347 list_add(&r1->res_hashchain, &ls->ls_new_rsb);
348 ls->ls_new_rsb_count++;
349 }
350 if (r2) {
351 list_add(&r2->res_hashchain, &ls->ls_new_rsb);
352 ls->ls_new_rsb_count++;
353 }
354 count = ls->ls_new_rsb_count;
355 spin_unlock(&ls->ls_new_rsb_spin);
356
357 if (!count)
358 return -ENOMEM;
359 return 0;
360}
361
362/* If ls->ls_new_rsb is empty, return -EAGAIN, so the caller can
363 unlock any spinlocks, go back and call pre_rsb_struct again.
364 Otherwise, take an rsb off the list and return it. */
365
366static int get_rsb_struct(struct dlm_ls *ls, char *name, int len,
367 struct dlm_rsb **r_ret)
368{
369 struct dlm_rsb *r;
370 int count;
371
372 spin_lock(&ls->ls_new_rsb_spin);
373 if (list_empty(&ls->ls_new_rsb)) {
374 count = ls->ls_new_rsb_count;
375 spin_unlock(&ls->ls_new_rsb_spin);
376 log_debug(ls, "find_rsb retry %d %d %s",
377 count, dlm_config.ci_new_rsb_count, name);
378 return -EAGAIN;
379 }
380
381 r = list_first_entry(&ls->ls_new_rsb, struct dlm_rsb, res_hashchain);
382 list_del(&r->res_hashchain);
383 ls->ls_new_rsb_count--;
384 spin_unlock(&ls->ls_new_rsb_spin);
385
386 r->res_ls = ls;
387 r->res_length = len;
388 memcpy(r->res_name, name, len);
389 mutex_init(&r->res_mutex);
390
391 INIT_LIST_HEAD(&r->res_hashchain);
392 INIT_LIST_HEAD(&r->res_lookup);
393 INIT_LIST_HEAD(&r->res_grantqueue);
394 INIT_LIST_HEAD(&r->res_convertqueue);
395 INIT_LIST_HEAD(&r->res_waitqueue);
396 INIT_LIST_HEAD(&r->res_root_list);
397 INIT_LIST_HEAD(&r->res_recover_list);
398
399 *r_ret = r;
400 return 0;
401}
402
403static int search_rsb_list(struct list_head *head, char *name, int len,
404 unsigned int flags, struct dlm_rsb **r_ret)
405{
406 struct dlm_rsb *r;
407 int error = 0;
408
409 list_for_each_entry(r, head, res_hashchain) {
410 if (len == r->res_length && !memcmp(name, r->res_name, len))
411 goto found;
412 }
413 *r_ret = NULL;
414 return -EBADR;
415
416 found:
417 if (r->res_nodeid && (flags & R_MASTER))
418 error = -ENOTBLK;
419 *r_ret = r;
420 return error;
421}
422
423static int _search_rsb(struct dlm_ls *ls, char *name, int len, int b,
424 unsigned int flags, struct dlm_rsb **r_ret)
425{
426 struct dlm_rsb *r;
427 int error;
428
429 error = search_rsb_list(&ls->ls_rsbtbl[b].list, name, len, flags, &r);
430 if (!error) {
431 kref_get(&r->res_ref);
432 goto out;
433 }
434 error = search_rsb_list(&ls->ls_rsbtbl[b].toss, name, len, flags, &r);
435 if (error)
436 goto out;
437
438 list_move(&r->res_hashchain, &ls->ls_rsbtbl[b].list);
439
440 if (dlm_no_directory(ls))
441 goto out;
442
443 if (r->res_nodeid == -1) {
444 rsb_clear_flag(r, RSB_MASTER_UNCERTAIN);
445 r->res_first_lkid = 0;
446 } else if (r->res_nodeid > 0) {
447 rsb_set_flag(r, RSB_MASTER_UNCERTAIN);
448 r->res_first_lkid = 0;
449 } else {
450 DLM_ASSERT(r->res_nodeid == 0, dlm_print_rsb(r););
451 DLM_ASSERT(!rsb_flag(r, RSB_MASTER_UNCERTAIN),);
452 }
453 out:
454 *r_ret = r;
455 return error;
456}
457
458/*
459 * Find rsb in rsbtbl and potentially create/add one
460 *
461 * Delaying the release of rsb's has a similar benefit to applications keeping
462 * NL locks on an rsb, but without the guarantee that the cached master value
463 * will still be valid when the rsb is reused. Apps aren't always smart enough
464 * to keep NL locks on an rsb that they may lock again shortly; this can lead
465 * to excessive master lookups and removals if we don't delay the release.
466 *
467 * Searching for an rsb means looking through both the normal list and toss
468 * list. When found on the toss list the rsb is moved to the normal list with
469 * ref count of 1; when found on normal list the ref count is incremented.
470 */
471
472static int find_rsb(struct dlm_ls *ls, char *name, int namelen,
473 unsigned int flags, struct dlm_rsb **r_ret)
474{
475 struct dlm_rsb *r = NULL;
476 uint32_t hash, bucket;
477 int error;
478
479 if (namelen > DLM_RESNAME_MAXLEN) {
480 error = -EINVAL;
481 goto out;
482 }
483
484 if (dlm_no_directory(ls))
485 flags |= R_CREATE;
486
487 hash = jhash(name, namelen, 0);
488 bucket = hash & (ls->ls_rsbtbl_size - 1);
489
490 retry:
491 if (flags & R_CREATE) {
492 error = pre_rsb_struct(ls);
493 if (error < 0)
494 goto out;
495 }
496
497 spin_lock(&ls->ls_rsbtbl[bucket].lock);
498
499 error = _search_rsb(ls, name, namelen, bucket, flags, &r);
500 if (!error)
501 goto out_unlock;
502
503 if (error == -EBADR && !(flags & R_CREATE))
504 goto out_unlock;
505
506 /* the rsb was found but wasn't a master copy */
507 if (error == -ENOTBLK)
508 goto out_unlock;
509
510 error = get_rsb_struct(ls, name, namelen, &r);
511 if (error == -EAGAIN) {
512 spin_unlock(&ls->ls_rsbtbl[bucket].lock);
513 goto retry;
514 }
515 if (error)
516 goto out_unlock;
517
518 r->res_hash = hash;
519 r->res_bucket = bucket;
520 r->res_nodeid = -1;
521 kref_init(&r->res_ref);
522
523 /* With no directory, the master can be set immediately */
524 if (dlm_no_directory(ls)) {
525 int nodeid = dlm_dir_nodeid(r);
526 if (nodeid == dlm_our_nodeid())
527 nodeid = 0;
528 r->res_nodeid = nodeid;
529 }
530 list_add(&r->res_hashchain, &ls->ls_rsbtbl[bucket].list);
531 error = 0;
532 out_unlock:
533 spin_unlock(&ls->ls_rsbtbl[bucket].lock);
534 out:
535 *r_ret = r;
536 return error;
537}
538
539/* This is only called to add a reference when the code already holds
540 a valid reference to the rsb, so there's no need for locking. */
541
542static inline void hold_rsb(struct dlm_rsb *r)
543{
544 kref_get(&r->res_ref);
545}
546
547void dlm_hold_rsb(struct dlm_rsb *r)
548{
549 hold_rsb(r);
550}
551
552static void toss_rsb(struct kref *kref)
553{
554 struct dlm_rsb *r = container_of(kref, struct dlm_rsb, res_ref);
555 struct dlm_ls *ls = r->res_ls;
556
557 DLM_ASSERT(list_empty(&r->res_root_list), dlm_print_rsb(r););
558 kref_init(&r->res_ref);
559 list_move(&r->res_hashchain, &ls->ls_rsbtbl[r->res_bucket].toss);
560 r->res_toss_time = jiffies;
561 if (r->res_lvbptr) {
562 dlm_free_lvb(r->res_lvbptr);
563 r->res_lvbptr = NULL;
564 }
565}
566
567/* When all references to the rsb are gone it's transferred to
568 the tossed list for later disposal. */
569
570static void put_rsb(struct dlm_rsb *r)
571{
572 struct dlm_ls *ls = r->res_ls;
573 uint32_t bucket = r->res_bucket;
574
575 spin_lock(&ls->ls_rsbtbl[bucket].lock);
576 kref_put(&r->res_ref, toss_rsb);
577 spin_unlock(&ls->ls_rsbtbl[bucket].lock);
578}
579
580void dlm_put_rsb(struct dlm_rsb *r)
581{
582 put_rsb(r);
583}
584
585/* See comment for unhold_lkb */
586
587static void unhold_rsb(struct dlm_rsb *r)
588{
589 int rv;
590 rv = kref_put(&r->res_ref, toss_rsb);
591 DLM_ASSERT(!rv, dlm_dump_rsb(r););
592}
593
594static void kill_rsb(struct kref *kref)
595{
596 struct dlm_rsb *r = container_of(kref, struct dlm_rsb, res_ref);
597
598 /* All work is done after the return from kref_put() so we
599 can release the write_lock before the remove and free. */
600
601 DLM_ASSERT(list_empty(&r->res_lookup), dlm_dump_rsb(r););
602 DLM_ASSERT(list_empty(&r->res_grantqueue), dlm_dump_rsb(r););
603 DLM_ASSERT(list_empty(&r->res_convertqueue), dlm_dump_rsb(r););
604 DLM_ASSERT(list_empty(&r->res_waitqueue), dlm_dump_rsb(r););
605 DLM_ASSERT(list_empty(&r->res_root_list), dlm_dump_rsb(r););
606 DLM_ASSERT(list_empty(&r->res_recover_list), dlm_dump_rsb(r););
607}
608
609/* Attaching/detaching lkb's from rsb's is for rsb reference counting.
610 The rsb must exist as long as any lkb's for it do. */
611
612static void attach_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb)
613{
614 hold_rsb(r);
615 lkb->lkb_resource = r;
616}
617
618static void detach_lkb(struct dlm_lkb *lkb)
619{
620 if (lkb->lkb_resource) {
621 put_rsb(lkb->lkb_resource);
622 lkb->lkb_resource = NULL;
623 }
624}
625
626static int create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret)
627{
628 struct dlm_lkb *lkb;
629 int rv, id;
630
631 lkb = dlm_allocate_lkb(ls);
632 if (!lkb)
633 return -ENOMEM;
634
635 lkb->lkb_nodeid = -1;
636 lkb->lkb_grmode = DLM_LOCK_IV;
637 kref_init(&lkb->lkb_ref);
638 INIT_LIST_HEAD(&lkb->lkb_ownqueue);
639 INIT_LIST_HEAD(&lkb->lkb_rsb_lookup);
640 INIT_LIST_HEAD(&lkb->lkb_time_list);
641 INIT_LIST_HEAD(&lkb->lkb_cb_list);
642 mutex_init(&lkb->lkb_cb_mutex);
643 INIT_WORK(&lkb->lkb_cb_work, dlm_callback_work);
644
645 retry:
646 rv = idr_pre_get(&ls->ls_lkbidr, GFP_NOFS);
647 if (!rv)
648 return -ENOMEM;
649
650 spin_lock(&ls->ls_lkbidr_spin);
651 rv = idr_get_new_above(&ls->ls_lkbidr, lkb, 1, &id);
652 if (!rv)
653 lkb->lkb_id = id;
654 spin_unlock(&ls->ls_lkbidr_spin);
655
656 if (rv == -EAGAIN)
657 goto retry;
658
659 if (rv < 0) {
660 log_error(ls, "create_lkb idr error %d", rv);
661 return rv;
662 }
663
664 *lkb_ret = lkb;
665 return 0;
666}
667
668static int find_lkb(struct dlm_ls *ls, uint32_t lkid, struct dlm_lkb **lkb_ret)
669{
670 struct dlm_lkb *lkb;
671
672 spin_lock(&ls->ls_lkbidr_spin);
673 lkb = idr_find(&ls->ls_lkbidr, lkid);
674 if (lkb)
675 kref_get(&lkb->lkb_ref);
676 spin_unlock(&ls->ls_lkbidr_spin);
677
678 *lkb_ret = lkb;
679 return lkb ? 0 : -ENOENT;
680}
681
682static void kill_lkb(struct kref *kref)
683{
684 struct dlm_lkb *lkb = container_of(kref, struct dlm_lkb, lkb_ref);
685
686 /* All work is done after the return from kref_put() so we
687 can release the write_lock before the detach_lkb */
688
689 DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb););
690}
691
692/* __put_lkb() is used when an lkb may not have an rsb attached to
693 it so we need to provide the lockspace explicitly */
694
695static int __put_lkb(struct dlm_ls *ls, struct dlm_lkb *lkb)
696{
697 uint32_t lkid = lkb->lkb_id;
698
699 spin_lock(&ls->ls_lkbidr_spin);
700 if (kref_put(&lkb->lkb_ref, kill_lkb)) {
701 idr_remove(&ls->ls_lkbidr, lkid);
702 spin_unlock(&ls->ls_lkbidr_spin);
703
704 detach_lkb(lkb);
705
706 /* for local/process lkbs, lvbptr points to caller's lksb */
707 if (lkb->lkb_lvbptr && is_master_copy(lkb))
708 dlm_free_lvb(lkb->lkb_lvbptr);
709 dlm_free_lkb(lkb);
710 return 1;
711 } else {
712 spin_unlock(&ls->ls_lkbidr_spin);
713 return 0;
714 }
715}
716
717int dlm_put_lkb(struct dlm_lkb *lkb)
718{
719 struct dlm_ls *ls;
720
721 DLM_ASSERT(lkb->lkb_resource, dlm_print_lkb(lkb););
722 DLM_ASSERT(lkb->lkb_resource->res_ls, dlm_print_lkb(lkb););
723
724 ls = lkb->lkb_resource->res_ls;
725 return __put_lkb(ls, lkb);
726}
727
728/* This is only called to add a reference when the code already holds
729 a valid reference to the lkb, so there's no need for locking. */
730
731static inline void hold_lkb(struct dlm_lkb *lkb)
732{
733 kref_get(&lkb->lkb_ref);
734}
735
736/* This is called when we need to remove a reference and are certain
737 it's not the last ref. e.g. del_lkb is always called between a
738 find_lkb/put_lkb and is always the inverse of a previous add_lkb.
739 put_lkb would work fine, but would involve unnecessary locking */
740
741static inline void unhold_lkb(struct dlm_lkb *lkb)
742{
743 int rv;
744 rv = kref_put(&lkb->lkb_ref, kill_lkb);
745 DLM_ASSERT(!rv, dlm_print_lkb(lkb););
746}
747
748static void lkb_add_ordered(struct list_head *new, struct list_head *head,
749 int mode)
750{
751 struct dlm_lkb *lkb = NULL;
752
753 list_for_each_entry(lkb, head, lkb_statequeue)
754 if (lkb->lkb_rqmode < mode)
755 break;
756
757 __list_add(new, lkb->lkb_statequeue.prev, &lkb->lkb_statequeue);
758}
759
760/* add/remove lkb to rsb's grant/convert/wait queue */
761
762static void add_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb, int status)
763{
764 kref_get(&lkb->lkb_ref);
765
766 DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb););
767
768 lkb->lkb_timestamp = ktime_get();
769
770 lkb->lkb_status = status;
771
772 switch (status) {
773 case DLM_LKSTS_WAITING:
774 if (lkb->lkb_exflags & DLM_LKF_HEADQUE)
775 list_add(&lkb->lkb_statequeue, &r->res_waitqueue);
776 else
777 list_add_tail(&lkb->lkb_statequeue, &r->res_waitqueue);
778 break;
779 case DLM_LKSTS_GRANTED:
780 /* convention says granted locks kept in order of grmode */
781 lkb_add_ordered(&lkb->lkb_statequeue, &r->res_grantqueue,
782 lkb->lkb_grmode);
783 break;
784 case DLM_LKSTS_CONVERT:
785 if (lkb->lkb_exflags & DLM_LKF_HEADQUE)
786 list_add(&lkb->lkb_statequeue, &r->res_convertqueue);
787 else
788 list_add_tail(&lkb->lkb_statequeue,
789 &r->res_convertqueue);
790 break;
791 default:
792 DLM_ASSERT(0, dlm_print_lkb(lkb); printk("sts=%d\n", status););
793 }
794}
795
796static void del_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb)
797{
798 lkb->lkb_status = 0;
799 list_del(&lkb->lkb_statequeue);
800 unhold_lkb(lkb);
801}
802
803static void move_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb, int sts)
804{
805 hold_lkb(lkb);
806 del_lkb(r, lkb);
807 add_lkb(r, lkb, sts);
808 unhold_lkb(lkb);
809}
810
811static int msg_reply_type(int mstype)
812{
813 switch (mstype) {
814 case DLM_MSG_REQUEST:
815 return DLM_MSG_REQUEST_REPLY;
816 case DLM_MSG_CONVERT:
817 return DLM_MSG_CONVERT_REPLY;
818 case DLM_MSG_UNLOCK:
819 return DLM_MSG_UNLOCK_REPLY;
820 case DLM_MSG_CANCEL:
821 return DLM_MSG_CANCEL_REPLY;
822 case DLM_MSG_LOOKUP:
823 return DLM_MSG_LOOKUP_REPLY;
824 }
825 return -1;
826}
827
828static int nodeid_warned(int nodeid, int num_nodes, int *warned)
829{
830 int i;
831
832 for (i = 0; i < num_nodes; i++) {
833 if (!warned[i]) {
834 warned[i] = nodeid;
835 return 0;
836 }
837 if (warned[i] == nodeid)
838 return 1;
839 }
840 return 0;
841}
842
843void dlm_scan_waiters(struct dlm_ls *ls)
844{
845 struct dlm_lkb *lkb;
846 ktime_t zero = ktime_set(0, 0);
847 s64 us;
848 s64 debug_maxus = 0;
849 u32 debug_scanned = 0;
850 u32 debug_expired = 0;
851 int num_nodes = 0;
852 int *warned = NULL;
853
854 if (!dlm_config.ci_waitwarn_us)
855 return;
856
857 mutex_lock(&ls->ls_waiters_mutex);
858
859 list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) {
860 if (ktime_equal(lkb->lkb_wait_time, zero))
861 continue;
862
863 debug_scanned++;
864
865 us = ktime_to_us(ktime_sub(ktime_get(), lkb->lkb_wait_time));
866
867 if (us < dlm_config.ci_waitwarn_us)
868 continue;
869
870 lkb->lkb_wait_time = zero;
871
872 debug_expired++;
873 if (us > debug_maxus)
874 debug_maxus = us;
875
876 if (!num_nodes) {
877 num_nodes = ls->ls_num_nodes;
878 warned = kzalloc(num_nodes * sizeof(int), GFP_KERNEL);
879 }
880 if (!warned)
881 continue;
882 if (nodeid_warned(lkb->lkb_wait_nodeid, num_nodes, warned))
883 continue;
884
885 log_error(ls, "waitwarn %x %lld %d us check connection to "
886 "node %d", lkb->lkb_id, (long long)us,
887 dlm_config.ci_waitwarn_us, lkb->lkb_wait_nodeid);
888 }
889 mutex_unlock(&ls->ls_waiters_mutex);
890 kfree(warned);
891
892 if (debug_expired)
893 log_debug(ls, "scan_waiters %u warn %u over %d us max %lld us",
894 debug_scanned, debug_expired,
895 dlm_config.ci_waitwarn_us, (long long)debug_maxus);
896}
897
898/* add/remove lkb from global waiters list of lkb's waiting for
899 a reply from a remote node */
900
901static int add_to_waiters(struct dlm_lkb *lkb, int mstype, int to_nodeid)
902{
903 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
904 int error = 0;
905
906 mutex_lock(&ls->ls_waiters_mutex);
907
908 if (is_overlap_unlock(lkb) ||
909 (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL))) {
910 error = -EINVAL;
911 goto out;
912 }
913
914 if (lkb->lkb_wait_type || is_overlap_cancel(lkb)) {
915 switch (mstype) {
916 case DLM_MSG_UNLOCK:
917 lkb->lkb_flags |= DLM_IFL_OVERLAP_UNLOCK;
918 break;
919 case DLM_MSG_CANCEL:
920 lkb->lkb_flags |= DLM_IFL_OVERLAP_CANCEL;
921 break;
922 default:
923 error = -EBUSY;
924 goto out;
925 }
926 lkb->lkb_wait_count++;
927 hold_lkb(lkb);
928
929 log_debug(ls, "addwait %x cur %d overlap %d count %d f %x",
930 lkb->lkb_id, lkb->lkb_wait_type, mstype,
931 lkb->lkb_wait_count, lkb->lkb_flags);
932 goto out;
933 }
934
935 DLM_ASSERT(!lkb->lkb_wait_count,
936 dlm_print_lkb(lkb);
937 printk("wait_count %d\n", lkb->lkb_wait_count););
938
939 lkb->lkb_wait_count++;
940 lkb->lkb_wait_type = mstype;
941 lkb->lkb_wait_time = ktime_get();
942 lkb->lkb_wait_nodeid = to_nodeid; /* for debugging */
943 hold_lkb(lkb);
944 list_add(&lkb->lkb_wait_reply, &ls->ls_waiters);
945 out:
946 if (error)
947 log_error(ls, "addwait error %x %d flags %x %d %d %s",
948 lkb->lkb_id, error, lkb->lkb_flags, mstype,
949 lkb->lkb_wait_type, lkb->lkb_resource->res_name);
950 mutex_unlock(&ls->ls_waiters_mutex);
951 return error;
952}
953
954/* We clear the RESEND flag because we might be taking an lkb off the waiters
955 list as part of process_requestqueue (e.g. a lookup that has an optimized
956 request reply on the requestqueue) between dlm_recover_waiters_pre() which
957 set RESEND and dlm_recover_waiters_post() */
958
959static int _remove_from_waiters(struct dlm_lkb *lkb, int mstype,
960 struct dlm_message *ms)
961{
962 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
963 int overlap_done = 0;
964
965 if (is_overlap_unlock(lkb) && (mstype == DLM_MSG_UNLOCK_REPLY)) {
966 log_debug(ls, "remwait %x unlock_reply overlap", lkb->lkb_id);
967 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
968 overlap_done = 1;
969 goto out_del;
970 }
971
972 if (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL_REPLY)) {
973 log_debug(ls, "remwait %x cancel_reply overlap", lkb->lkb_id);
974 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
975 overlap_done = 1;
976 goto out_del;
977 }
978
979 /* Cancel state was preemptively cleared by a successful convert,
980 see next comment, nothing to do. */
981
982 if ((mstype == DLM_MSG_CANCEL_REPLY) &&
983 (lkb->lkb_wait_type != DLM_MSG_CANCEL)) {
984 log_debug(ls, "remwait %x cancel_reply wait_type %d",
985 lkb->lkb_id, lkb->lkb_wait_type);
986 return -1;
987 }
988
989 /* Remove for the convert reply, and premptively remove for the
990 cancel reply. A convert has been granted while there's still
991 an outstanding cancel on it (the cancel is moot and the result
992 in the cancel reply should be 0). We preempt the cancel reply
993 because the app gets the convert result and then can follow up
994 with another op, like convert. This subsequent op would see the
995 lingering state of the cancel and fail with -EBUSY. */
996
997 if ((mstype == DLM_MSG_CONVERT_REPLY) &&
998 (lkb->lkb_wait_type == DLM_MSG_CONVERT) &&
999 is_overlap_cancel(lkb) && ms && !ms->m_result) {
1000 log_debug(ls, "remwait %x convert_reply zap overlap_cancel",
1001 lkb->lkb_id);
1002 lkb->lkb_wait_type = 0;
1003 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
1004 lkb->lkb_wait_count--;
1005 goto out_del;
1006 }
1007
1008 /* N.B. type of reply may not always correspond to type of original
1009 msg due to lookup->request optimization, verify others? */
1010
1011 if (lkb->lkb_wait_type) {
1012 lkb->lkb_wait_type = 0;
1013 goto out_del;
1014 }
1015
1016 log_error(ls, "remwait error %x reply %d flags %x no wait_type",
1017 lkb->lkb_id, mstype, lkb->lkb_flags);
1018 return -1;
1019
1020 out_del:
1021 /* the force-unlock/cancel has completed and we haven't recvd a reply
1022 to the op that was in progress prior to the unlock/cancel; we
1023 give up on any reply to the earlier op. FIXME: not sure when/how
1024 this would happen */
1025
1026 if (overlap_done && lkb->lkb_wait_type) {
1027 log_error(ls, "remwait error %x reply %d wait_type %d overlap",
1028 lkb->lkb_id, mstype, lkb->lkb_wait_type);
1029 lkb->lkb_wait_count--;
1030 lkb->lkb_wait_type = 0;
1031 }
1032
1033 DLM_ASSERT(lkb->lkb_wait_count, dlm_print_lkb(lkb););
1034
1035 lkb->lkb_flags &= ~DLM_IFL_RESEND;
1036 lkb->lkb_wait_count--;
1037 if (!lkb->lkb_wait_count)
1038 list_del_init(&lkb->lkb_wait_reply);
1039 unhold_lkb(lkb);
1040 return 0;
1041}
1042
1043static int remove_from_waiters(struct dlm_lkb *lkb, int mstype)
1044{
1045 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1046 int error;
1047
1048 mutex_lock(&ls->ls_waiters_mutex);
1049 error = _remove_from_waiters(lkb, mstype, NULL);
1050 mutex_unlock(&ls->ls_waiters_mutex);
1051 return error;
1052}
1053
1054/* Handles situations where we might be processing a "fake" or "stub" reply in
1055 which we can't try to take waiters_mutex again. */
1056
1057static int remove_from_waiters_ms(struct dlm_lkb *lkb, struct dlm_message *ms)
1058{
1059 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1060 int error;
1061
1062 if (ms->m_flags != DLM_IFL_STUB_MS)
1063 mutex_lock(&ls->ls_waiters_mutex);
1064 error = _remove_from_waiters(lkb, ms->m_type, ms);
1065 if (ms->m_flags != DLM_IFL_STUB_MS)
1066 mutex_unlock(&ls->ls_waiters_mutex);
1067 return error;
1068}
1069
1070static void dir_remove(struct dlm_rsb *r)
1071{
1072 int to_nodeid;
1073
1074 if (dlm_no_directory(r->res_ls))
1075 return;
1076
1077 to_nodeid = dlm_dir_nodeid(r);
1078 if (to_nodeid != dlm_our_nodeid())
1079 send_remove(r);
1080 else
1081 dlm_dir_remove_entry(r->res_ls, to_nodeid,
1082 r->res_name, r->res_length);
1083}
1084
1085/* FIXME: shouldn't this be able to exit as soon as one non-due rsb is
1086 found since they are in order of newest to oldest? */
1087
1088static int shrink_bucket(struct dlm_ls *ls, int b)
1089{
1090 struct dlm_rsb *r;
1091 int count = 0, found;
1092
1093 for (;;) {
1094 found = 0;
1095 spin_lock(&ls->ls_rsbtbl[b].lock);
1096 list_for_each_entry_reverse(r, &ls->ls_rsbtbl[b].toss,
1097 res_hashchain) {
1098 if (!time_after_eq(jiffies, r->res_toss_time +
1099 dlm_config.ci_toss_secs * HZ))
1100 continue;
1101 found = 1;
1102 break;
1103 }
1104
1105 if (!found) {
1106 spin_unlock(&ls->ls_rsbtbl[b].lock);
1107 break;
1108 }
1109
1110 if (kref_put(&r->res_ref, kill_rsb)) {
1111 list_del(&r->res_hashchain);
1112 spin_unlock(&ls->ls_rsbtbl[b].lock);
1113
1114 if (is_master(r))
1115 dir_remove(r);
1116 dlm_free_rsb(r);
1117 count++;
1118 } else {
1119 spin_unlock(&ls->ls_rsbtbl[b].lock);
1120 log_error(ls, "tossed rsb in use %s", r->res_name);
1121 }
1122 }
1123
1124 return count;
1125}
1126
1127void dlm_scan_rsbs(struct dlm_ls *ls)
1128{
1129 int i;
1130
1131 for (i = 0; i < ls->ls_rsbtbl_size; i++) {
1132 shrink_bucket(ls, i);
1133 if (dlm_locking_stopped(ls))
1134 break;
1135 cond_resched();
1136 }
1137}
1138
1139static void add_timeout(struct dlm_lkb *lkb)
1140{
1141 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1142
1143 if (is_master_copy(lkb))
1144 return;
1145
1146 if (test_bit(LSFL_TIMEWARN, &ls->ls_flags) &&
1147 !(lkb->lkb_exflags & DLM_LKF_NODLCKWT)) {
1148 lkb->lkb_flags |= DLM_IFL_WATCH_TIMEWARN;
1149 goto add_it;
1150 }
1151 if (lkb->lkb_exflags & DLM_LKF_TIMEOUT)
1152 goto add_it;
1153 return;
1154
1155 add_it:
1156 DLM_ASSERT(list_empty(&lkb->lkb_time_list), dlm_print_lkb(lkb););
1157 mutex_lock(&ls->ls_timeout_mutex);
1158 hold_lkb(lkb);
1159 list_add_tail(&lkb->lkb_time_list, &ls->ls_timeout);
1160 mutex_unlock(&ls->ls_timeout_mutex);
1161}
1162
1163static void del_timeout(struct dlm_lkb *lkb)
1164{
1165 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1166
1167 mutex_lock(&ls->ls_timeout_mutex);
1168 if (!list_empty(&lkb->lkb_time_list)) {
1169 list_del_init(&lkb->lkb_time_list);
1170 unhold_lkb(lkb);
1171 }
1172 mutex_unlock(&ls->ls_timeout_mutex);
1173}
1174
1175/* FIXME: is it safe to look at lkb_exflags, lkb_flags, lkb_timestamp, and
1176 lkb_lksb_timeout without lock_rsb? Note: we can't lock timeout_mutex
1177 and then lock rsb because of lock ordering in add_timeout. We may need
1178 to specify some special timeout-related bits in the lkb that are just to
1179 be accessed under the timeout_mutex. */
1180
1181void dlm_scan_timeout(struct dlm_ls *ls)
1182{
1183 struct dlm_rsb *r;
1184 struct dlm_lkb *lkb;
1185 int do_cancel, do_warn;
1186 s64 wait_us;
1187
1188 for (;;) {
1189 if (dlm_locking_stopped(ls))
1190 break;
1191
1192 do_cancel = 0;
1193 do_warn = 0;
1194 mutex_lock(&ls->ls_timeout_mutex);
1195 list_for_each_entry(lkb, &ls->ls_timeout, lkb_time_list) {
1196
1197 wait_us = ktime_to_us(ktime_sub(ktime_get(),
1198 lkb->lkb_timestamp));
1199
1200 if ((lkb->lkb_exflags & DLM_LKF_TIMEOUT) &&
1201 wait_us >= (lkb->lkb_timeout_cs * 10000))
1202 do_cancel = 1;
1203
1204 if ((lkb->lkb_flags & DLM_IFL_WATCH_TIMEWARN) &&
1205 wait_us >= dlm_config.ci_timewarn_cs * 10000)
1206 do_warn = 1;
1207
1208 if (!do_cancel && !do_warn)
1209 continue;
1210 hold_lkb(lkb);
1211 break;
1212 }
1213 mutex_unlock(&ls->ls_timeout_mutex);
1214
1215 if (!do_cancel && !do_warn)
1216 break;
1217
1218 r = lkb->lkb_resource;
1219 hold_rsb(r);
1220 lock_rsb(r);
1221
1222 if (do_warn) {
1223 /* clear flag so we only warn once */
1224 lkb->lkb_flags &= ~DLM_IFL_WATCH_TIMEWARN;
1225 if (!(lkb->lkb_exflags & DLM_LKF_TIMEOUT))
1226 del_timeout(lkb);
1227 dlm_timeout_warn(lkb);
1228 }
1229
1230 if (do_cancel) {
1231 log_debug(ls, "timeout cancel %x node %d %s",
1232 lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
1233 lkb->lkb_flags &= ~DLM_IFL_WATCH_TIMEWARN;
1234 lkb->lkb_flags |= DLM_IFL_TIMEOUT_CANCEL;
1235 del_timeout(lkb);
1236 _cancel_lock(r, lkb);
1237 }
1238
1239 unlock_rsb(r);
1240 unhold_rsb(r);
1241 dlm_put_lkb(lkb);
1242 }
1243}
1244
1245/* This is only called by dlm_recoverd, and we rely on dlm_ls_stop() stopping
1246 dlm_recoverd before checking/setting ls_recover_begin. */
1247
1248void dlm_adjust_timeouts(struct dlm_ls *ls)
1249{
1250 struct dlm_lkb *lkb;
1251 u64 adj_us = jiffies_to_usecs(jiffies - ls->ls_recover_begin);
1252
1253 ls->ls_recover_begin = 0;
1254 mutex_lock(&ls->ls_timeout_mutex);
1255 list_for_each_entry(lkb, &ls->ls_timeout, lkb_time_list)
1256 lkb->lkb_timestamp = ktime_add_us(lkb->lkb_timestamp, adj_us);
1257 mutex_unlock(&ls->ls_timeout_mutex);
1258
1259 if (!dlm_config.ci_waitwarn_us)
1260 return;
1261
1262 mutex_lock(&ls->ls_waiters_mutex);
1263 list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) {
1264 if (ktime_to_us(lkb->lkb_wait_time))
1265 lkb->lkb_wait_time = ktime_get();
1266 }
1267 mutex_unlock(&ls->ls_waiters_mutex);
1268}
1269
1270/* lkb is master or local copy */
1271
1272static void set_lvb_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1273{
1274 int b, len = r->res_ls->ls_lvblen;
1275
1276 /* b=1 lvb returned to caller
1277 b=0 lvb written to rsb or invalidated
1278 b=-1 do nothing */
1279
1280 b = dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1];
1281
1282 if (b == 1) {
1283 if (!lkb->lkb_lvbptr)
1284 return;
1285
1286 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1287 return;
1288
1289 if (!r->res_lvbptr)
1290 return;
1291
1292 memcpy(lkb->lkb_lvbptr, r->res_lvbptr, len);
1293 lkb->lkb_lvbseq = r->res_lvbseq;
1294
1295 } else if (b == 0) {
1296 if (lkb->lkb_exflags & DLM_LKF_IVVALBLK) {
1297 rsb_set_flag(r, RSB_VALNOTVALID);
1298 return;
1299 }
1300
1301 if (!lkb->lkb_lvbptr)
1302 return;
1303
1304 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1305 return;
1306
1307 if (!r->res_lvbptr)
1308 r->res_lvbptr = dlm_allocate_lvb(r->res_ls);
1309
1310 if (!r->res_lvbptr)
1311 return;
1312
1313 memcpy(r->res_lvbptr, lkb->lkb_lvbptr, len);
1314 r->res_lvbseq++;
1315 lkb->lkb_lvbseq = r->res_lvbseq;
1316 rsb_clear_flag(r, RSB_VALNOTVALID);
1317 }
1318
1319 if (rsb_flag(r, RSB_VALNOTVALID))
1320 lkb->lkb_sbflags |= DLM_SBF_VALNOTVALID;
1321}
1322
1323static void set_lvb_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1324{
1325 if (lkb->lkb_grmode < DLM_LOCK_PW)
1326 return;
1327
1328 if (lkb->lkb_exflags & DLM_LKF_IVVALBLK) {
1329 rsb_set_flag(r, RSB_VALNOTVALID);
1330 return;
1331 }
1332
1333 if (!lkb->lkb_lvbptr)
1334 return;
1335
1336 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1337 return;
1338
1339 if (!r->res_lvbptr)
1340 r->res_lvbptr = dlm_allocate_lvb(r->res_ls);
1341
1342 if (!r->res_lvbptr)
1343 return;
1344
1345 memcpy(r->res_lvbptr, lkb->lkb_lvbptr, r->res_ls->ls_lvblen);
1346 r->res_lvbseq++;
1347 rsb_clear_flag(r, RSB_VALNOTVALID);
1348}
1349
1350/* lkb is process copy (pc) */
1351
1352static void set_lvb_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb,
1353 struct dlm_message *ms)
1354{
1355 int b;
1356
1357 if (!lkb->lkb_lvbptr)
1358 return;
1359
1360 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1361 return;
1362
1363 b = dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1];
1364 if (b == 1) {
1365 int len = receive_extralen(ms);
1366 if (len > DLM_RESNAME_MAXLEN)
1367 len = DLM_RESNAME_MAXLEN;
1368 memcpy(lkb->lkb_lvbptr, ms->m_extra, len);
1369 lkb->lkb_lvbseq = ms->m_lvbseq;
1370 }
1371}
1372
1373/* Manipulate lkb's on rsb's convert/granted/waiting queues
1374 remove_lock -- used for unlock, removes lkb from granted
1375 revert_lock -- used for cancel, moves lkb from convert to granted
1376 grant_lock -- used for request and convert, adds lkb to granted or
1377 moves lkb from convert or waiting to granted
1378
1379 Each of these is used for master or local copy lkb's. There is
1380 also a _pc() variation used to make the corresponding change on
1381 a process copy (pc) lkb. */
1382
1383static void _remove_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1384{
1385 del_lkb(r, lkb);
1386 lkb->lkb_grmode = DLM_LOCK_IV;
1387 /* this unhold undoes the original ref from create_lkb()
1388 so this leads to the lkb being freed */
1389 unhold_lkb(lkb);
1390}
1391
1392static void remove_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1393{
1394 set_lvb_unlock(r, lkb);
1395 _remove_lock(r, lkb);
1396}
1397
1398static void remove_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
1399{
1400 _remove_lock(r, lkb);
1401}
1402
1403/* returns: 0 did nothing
1404 1 moved lock to granted
1405 -1 removed lock */
1406
1407static int revert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1408{
1409 int rv = 0;
1410
1411 lkb->lkb_rqmode = DLM_LOCK_IV;
1412
1413 switch (lkb->lkb_status) {
1414 case DLM_LKSTS_GRANTED:
1415 break;
1416 case DLM_LKSTS_CONVERT:
1417 move_lkb(r, lkb, DLM_LKSTS_GRANTED);
1418 rv = 1;
1419 break;
1420 case DLM_LKSTS_WAITING:
1421 del_lkb(r, lkb);
1422 lkb->lkb_grmode = DLM_LOCK_IV;
1423 /* this unhold undoes the original ref from create_lkb()
1424 so this leads to the lkb being freed */
1425 unhold_lkb(lkb);
1426 rv = -1;
1427 break;
1428 default:
1429 log_print("invalid status for revert %d", lkb->lkb_status);
1430 }
1431 return rv;
1432}
1433
1434static int revert_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
1435{
1436 return revert_lock(r, lkb);
1437}
1438
1439static void _grant_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1440{
1441 if (lkb->lkb_grmode != lkb->lkb_rqmode) {
1442 lkb->lkb_grmode = lkb->lkb_rqmode;
1443 if (lkb->lkb_status)
1444 move_lkb(r, lkb, DLM_LKSTS_GRANTED);
1445 else
1446 add_lkb(r, lkb, DLM_LKSTS_GRANTED);
1447 }
1448
1449 lkb->lkb_rqmode = DLM_LOCK_IV;
1450}
1451
1452static void grant_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1453{
1454 set_lvb_lock(r, lkb);
1455 _grant_lock(r, lkb);
1456 lkb->lkb_highbast = 0;
1457}
1458
1459static void grant_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb,
1460 struct dlm_message *ms)
1461{
1462 set_lvb_lock_pc(r, lkb, ms);
1463 _grant_lock(r, lkb);
1464}
1465
1466/* called by grant_pending_locks() which means an async grant message must
1467 be sent to the requesting node in addition to granting the lock if the
1468 lkb belongs to a remote node. */
1469
1470static void grant_lock_pending(struct dlm_rsb *r, struct dlm_lkb *lkb)
1471{
1472 grant_lock(r, lkb);
1473 if (is_master_copy(lkb))
1474 send_grant(r, lkb);
1475 else
1476 queue_cast(r, lkb, 0);
1477}
1478
1479/* The special CONVDEADLK, ALTPR and ALTCW flags allow the master to
1480 change the granted/requested modes. We're munging things accordingly in
1481 the process copy.
1482 CONVDEADLK: our grmode may have been forced down to NL to resolve a
1483 conversion deadlock
1484 ALTPR/ALTCW: our rqmode may have been changed to PR or CW to become
1485 compatible with other granted locks */
1486
1487static void munge_demoted(struct dlm_lkb *lkb)
1488{
1489 if (lkb->lkb_rqmode == DLM_LOCK_IV || lkb->lkb_grmode == DLM_LOCK_IV) {
1490 log_print("munge_demoted %x invalid modes gr %d rq %d",
1491 lkb->lkb_id, lkb->lkb_grmode, lkb->lkb_rqmode);
1492 return;
1493 }
1494
1495 lkb->lkb_grmode = DLM_LOCK_NL;
1496}
1497
1498static void munge_altmode(struct dlm_lkb *lkb, struct dlm_message *ms)
1499{
1500 if (ms->m_type != DLM_MSG_REQUEST_REPLY &&
1501 ms->m_type != DLM_MSG_GRANT) {
1502 log_print("munge_altmode %x invalid reply type %d",
1503 lkb->lkb_id, ms->m_type);
1504 return;
1505 }
1506
1507 if (lkb->lkb_exflags & DLM_LKF_ALTPR)
1508 lkb->lkb_rqmode = DLM_LOCK_PR;
1509 else if (lkb->lkb_exflags & DLM_LKF_ALTCW)
1510 lkb->lkb_rqmode = DLM_LOCK_CW;
1511 else {
1512 log_print("munge_altmode invalid exflags %x", lkb->lkb_exflags);
1513 dlm_print_lkb(lkb);
1514 }
1515}
1516
1517static inline int first_in_list(struct dlm_lkb *lkb, struct list_head *head)
1518{
1519 struct dlm_lkb *first = list_entry(head->next, struct dlm_lkb,
1520 lkb_statequeue);
1521 if (lkb->lkb_id == first->lkb_id)
1522 return 1;
1523
1524 return 0;
1525}
1526
1527/* Check if the given lkb conflicts with another lkb on the queue. */
1528
1529static int queue_conflict(struct list_head *head, struct dlm_lkb *lkb)
1530{
1531 struct dlm_lkb *this;
1532
1533 list_for_each_entry(this, head, lkb_statequeue) {
1534 if (this == lkb)
1535 continue;
1536 if (!modes_compat(this, lkb))
1537 return 1;
1538 }
1539 return 0;
1540}
1541
1542/*
1543 * "A conversion deadlock arises with a pair of lock requests in the converting
1544 * queue for one resource. The granted mode of each lock blocks the requested
1545 * mode of the other lock."
1546 *
1547 * Part 2: if the granted mode of lkb is preventing an earlier lkb in the
1548 * convert queue from being granted, then deadlk/demote lkb.
1549 *
1550 * Example:
1551 * Granted Queue: empty
1552 * Convert Queue: NL->EX (first lock)
1553 * PR->EX (second lock)
1554 *
1555 * The first lock can't be granted because of the granted mode of the second
1556 * lock and the second lock can't be granted because it's not first in the
1557 * list. We either cancel lkb's conversion (PR->EX) and return EDEADLK, or we
1558 * demote the granted mode of lkb (from PR to NL) if it has the CONVDEADLK
1559 * flag set and return DEMOTED in the lksb flags.
1560 *
1561 * Originally, this function detected conv-deadlk in a more limited scope:
1562 * - if !modes_compat(lkb1, lkb2) && !modes_compat(lkb2, lkb1), or
1563 * - if lkb1 was the first entry in the queue (not just earlier), and was
1564 * blocked by the granted mode of lkb2, and there was nothing on the
1565 * granted queue preventing lkb1 from being granted immediately, i.e.
1566 * lkb2 was the only thing preventing lkb1 from being granted.
1567 *
1568 * That second condition meant we'd only say there was conv-deadlk if
1569 * resolving it (by demotion) would lead to the first lock on the convert
1570 * queue being granted right away. It allowed conversion deadlocks to exist
1571 * between locks on the convert queue while they couldn't be granted anyway.
1572 *
1573 * Now, we detect and take action on conversion deadlocks immediately when
1574 * they're created, even if they may not be immediately consequential. If
1575 * lkb1 exists anywhere in the convert queue and lkb2 comes in with a granted
1576 * mode that would prevent lkb1's conversion from being granted, we do a
1577 * deadlk/demote on lkb2 right away and don't let it onto the convert queue.
1578 * I think this means that the lkb_is_ahead condition below should always
1579 * be zero, i.e. there will never be conv-deadlk between two locks that are
1580 * both already on the convert queue.
1581 */
1582
1583static int conversion_deadlock_detect(struct dlm_rsb *r, struct dlm_lkb *lkb2)
1584{
1585 struct dlm_lkb *lkb1;
1586 int lkb_is_ahead = 0;
1587
1588 list_for_each_entry(lkb1, &r->res_convertqueue, lkb_statequeue) {
1589 if (lkb1 == lkb2) {
1590 lkb_is_ahead = 1;
1591 continue;
1592 }
1593
1594 if (!lkb_is_ahead) {
1595 if (!modes_compat(lkb2, lkb1))
1596 return 1;
1597 } else {
1598 if (!modes_compat(lkb2, lkb1) &&
1599 !modes_compat(lkb1, lkb2))
1600 return 1;
1601 }
1602 }
1603 return 0;
1604}
1605
1606/*
1607 * Return 1 if the lock can be granted, 0 otherwise.
1608 * Also detect and resolve conversion deadlocks.
1609 *
1610 * lkb is the lock to be granted
1611 *
1612 * now is 1 if the function is being called in the context of the
1613 * immediate request, it is 0 if called later, after the lock has been
1614 * queued.
1615 *
1616 * References are from chapter 6 of "VAXcluster Principles" by Roy Davis
1617 */
1618
1619static int _can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now)
1620{
1621 int8_t conv = (lkb->lkb_grmode != DLM_LOCK_IV);
1622
1623 /*
1624 * 6-10: Version 5.4 introduced an option to address the phenomenon of
1625 * a new request for a NL mode lock being blocked.
1626 *
1627 * 6-11: If the optional EXPEDITE flag is used with the new NL mode
1628 * request, then it would be granted. In essence, the use of this flag
1629 * tells the Lock Manager to expedite theis request by not considering
1630 * what may be in the CONVERTING or WAITING queues... As of this
1631 * writing, the EXPEDITE flag can be used only with new requests for NL
1632 * mode locks. This flag is not valid for conversion requests.
1633 *
1634 * A shortcut. Earlier checks return an error if EXPEDITE is used in a
1635 * conversion or used with a non-NL requested mode. We also know an
1636 * EXPEDITE request is always granted immediately, so now must always
1637 * be 1. The full condition to grant an expedite request: (now &&
1638 * !conv && lkb->rqmode == DLM_LOCK_NL && (flags & EXPEDITE)) can
1639 * therefore be shortened to just checking the flag.
1640 */
1641
1642 if (lkb->lkb_exflags & DLM_LKF_EXPEDITE)
1643 return 1;
1644
1645 /*
1646 * A shortcut. Without this, !queue_conflict(grantqueue, lkb) would be
1647 * added to the remaining conditions.
1648 */
1649
1650 if (queue_conflict(&r->res_grantqueue, lkb))
1651 goto out;
1652
1653 /*
1654 * 6-3: By default, a conversion request is immediately granted if the
1655 * requested mode is compatible with the modes of all other granted
1656 * locks
1657 */
1658
1659 if (queue_conflict(&r->res_convertqueue, lkb))
1660 goto out;
1661
1662 /*
1663 * 6-5: But the default algorithm for deciding whether to grant or
1664 * queue conversion requests does not by itself guarantee that such
1665 * requests are serviced on a "first come first serve" basis. This, in
1666 * turn, can lead to a phenomenon known as "indefinate postponement".
1667 *
1668 * 6-7: This issue is dealt with by using the optional QUECVT flag with
1669 * the system service employed to request a lock conversion. This flag
1670 * forces certain conversion requests to be queued, even if they are
1671 * compatible with the granted modes of other locks on the same
1672 * resource. Thus, the use of this flag results in conversion requests
1673 * being ordered on a "first come first servce" basis.
1674 *
1675 * DCT: This condition is all about new conversions being able to occur
1676 * "in place" while the lock remains on the granted queue (assuming
1677 * nothing else conflicts.) IOW if QUECVT isn't set, a conversion
1678 * doesn't _have_ to go onto the convert queue where it's processed in
1679 * order. The "now" variable is necessary to distinguish converts
1680 * being received and processed for the first time now, because once a
1681 * convert is moved to the conversion queue the condition below applies
1682 * requiring fifo granting.
1683 */
1684
1685 if (now && conv && !(lkb->lkb_exflags & DLM_LKF_QUECVT))
1686 return 1;
1687
1688 /*
1689 * The NOORDER flag is set to avoid the standard vms rules on grant
1690 * order.
1691 */
1692
1693 if (lkb->lkb_exflags & DLM_LKF_NOORDER)
1694 return 1;
1695
1696 /*
1697 * 6-3: Once in that queue [CONVERTING], a conversion request cannot be
1698 * granted until all other conversion requests ahead of it are granted
1699 * and/or canceled.
1700 */
1701
1702 if (!now && conv && first_in_list(lkb, &r->res_convertqueue))
1703 return 1;
1704
1705 /*
1706 * 6-4: By default, a new request is immediately granted only if all
1707 * three of the following conditions are satisfied when the request is
1708 * issued:
1709 * - The queue of ungranted conversion requests for the resource is
1710 * empty.
1711 * - The queue of ungranted new requests for the resource is empty.
1712 * - The mode of the new request is compatible with the most
1713 * restrictive mode of all granted locks on the resource.
1714 */
1715
1716 if (now && !conv && list_empty(&r->res_convertqueue) &&
1717 list_empty(&r->res_waitqueue))
1718 return 1;
1719
1720 /*
1721 * 6-4: Once a lock request is in the queue of ungranted new requests,
1722 * it cannot be granted until the queue of ungranted conversion
1723 * requests is empty, all ungranted new requests ahead of it are
1724 * granted and/or canceled, and it is compatible with the granted mode
1725 * of the most restrictive lock granted on the resource.
1726 */
1727
1728 if (!now && !conv && list_empty(&r->res_convertqueue) &&
1729 first_in_list(lkb, &r->res_waitqueue))
1730 return 1;
1731 out:
1732 return 0;
1733}
1734
1735static int can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now,
1736 int *err)
1737{
1738 int rv;
1739 int8_t alt = 0, rqmode = lkb->lkb_rqmode;
1740 int8_t is_convert = (lkb->lkb_grmode != DLM_LOCK_IV);
1741
1742 if (err)
1743 *err = 0;
1744
1745 rv = _can_be_granted(r, lkb, now);
1746 if (rv)
1747 goto out;
1748
1749 /*
1750 * The CONVDEADLK flag is non-standard and tells the dlm to resolve
1751 * conversion deadlocks by demoting grmode to NL, otherwise the dlm
1752 * cancels one of the locks.
1753 */
1754
1755 if (is_convert && can_be_queued(lkb) &&
1756 conversion_deadlock_detect(r, lkb)) {
1757 if (lkb->lkb_exflags & DLM_LKF_CONVDEADLK) {
1758 lkb->lkb_grmode = DLM_LOCK_NL;
1759 lkb->lkb_sbflags |= DLM_SBF_DEMOTED;
1760 } else if (!(lkb->lkb_exflags & DLM_LKF_NODLCKWT)) {
1761 if (err)
1762 *err = -EDEADLK;
1763 else {
1764 log_print("can_be_granted deadlock %x now %d",
1765 lkb->lkb_id, now);
1766 dlm_dump_rsb(r);
1767 }
1768 }
1769 goto out;
1770 }
1771
1772 /*
1773 * The ALTPR and ALTCW flags are non-standard and tell the dlm to try
1774 * to grant a request in a mode other than the normal rqmode. It's a
1775 * simple way to provide a big optimization to applications that can
1776 * use them.
1777 */
1778
1779 if (rqmode != DLM_LOCK_PR && (lkb->lkb_exflags & DLM_LKF_ALTPR))
1780 alt = DLM_LOCK_PR;
1781 else if (rqmode != DLM_LOCK_CW && (lkb->lkb_exflags & DLM_LKF_ALTCW))
1782 alt = DLM_LOCK_CW;
1783
1784 if (alt) {
1785 lkb->lkb_rqmode = alt;
1786 rv = _can_be_granted(r, lkb, now);
1787 if (rv)
1788 lkb->lkb_sbflags |= DLM_SBF_ALTMODE;
1789 else
1790 lkb->lkb_rqmode = rqmode;
1791 }
1792 out:
1793 return rv;
1794}
1795
1796/* FIXME: I don't think that can_be_granted() can/will demote or find deadlock
1797 for locks pending on the convert list. Once verified (watch for these
1798 log_prints), we should be able to just call _can_be_granted() and not
1799 bother with the demote/deadlk cases here (and there's no easy way to deal
1800 with a deadlk here, we'd have to generate something like grant_lock with
1801 the deadlk error.) */
1802
1803/* Returns the highest requested mode of all blocked conversions; sets
1804 cw if there's a blocked conversion to DLM_LOCK_CW. */
1805
1806static int grant_pending_convert(struct dlm_rsb *r, int high, int *cw)
1807{
1808 struct dlm_lkb *lkb, *s;
1809 int hi, demoted, quit, grant_restart, demote_restart;
1810 int deadlk;
1811
1812 quit = 0;
1813 restart:
1814 grant_restart = 0;
1815 demote_restart = 0;
1816 hi = DLM_LOCK_IV;
1817
1818 list_for_each_entry_safe(lkb, s, &r->res_convertqueue, lkb_statequeue) {
1819 demoted = is_demoted(lkb);
1820 deadlk = 0;
1821
1822 if (can_be_granted(r, lkb, 0, &deadlk)) {
1823 grant_lock_pending(r, lkb);
1824 grant_restart = 1;
1825 continue;
1826 }
1827
1828 if (!demoted && is_demoted(lkb)) {
1829 log_print("WARN: pending demoted %x node %d %s",
1830 lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
1831 demote_restart = 1;
1832 continue;
1833 }
1834
1835 if (deadlk) {
1836 log_print("WARN: pending deadlock %x node %d %s",
1837 lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
1838 dlm_dump_rsb(r);
1839 continue;
1840 }
1841
1842 hi = max_t(int, lkb->lkb_rqmode, hi);
1843
1844 if (cw && lkb->lkb_rqmode == DLM_LOCK_CW)
1845 *cw = 1;
1846 }
1847
1848 if (grant_restart)
1849 goto restart;
1850 if (demote_restart && !quit) {
1851 quit = 1;
1852 goto restart;
1853 }
1854
1855 return max_t(int, high, hi);
1856}
1857
1858static int grant_pending_wait(struct dlm_rsb *r, int high, int *cw)
1859{
1860 struct dlm_lkb *lkb, *s;
1861
1862 list_for_each_entry_safe(lkb, s, &r->res_waitqueue, lkb_statequeue) {
1863 if (can_be_granted(r, lkb, 0, NULL))
1864 grant_lock_pending(r, lkb);
1865 else {
1866 high = max_t(int, lkb->lkb_rqmode, high);
1867 if (lkb->lkb_rqmode == DLM_LOCK_CW)
1868 *cw = 1;
1869 }
1870 }
1871
1872 return high;
1873}
1874
1875/* cw of 1 means there's a lock with a rqmode of DLM_LOCK_CW that's blocked
1876 on either the convert or waiting queue.
1877 high is the largest rqmode of all locks blocked on the convert or
1878 waiting queue. */
1879
1880static int lock_requires_bast(struct dlm_lkb *gr, int high, int cw)
1881{
1882 if (gr->lkb_grmode == DLM_LOCK_PR && cw) {
1883 if (gr->lkb_highbast < DLM_LOCK_EX)
1884 return 1;
1885 return 0;
1886 }
1887
1888 if (gr->lkb_highbast < high &&
1889 !__dlm_compat_matrix[gr->lkb_grmode+1][high+1])
1890 return 1;
1891 return 0;
1892}
1893
1894static void grant_pending_locks(struct dlm_rsb *r)
1895{
1896 struct dlm_lkb *lkb, *s;
1897 int high = DLM_LOCK_IV;
1898 int cw = 0;
1899
1900 DLM_ASSERT(is_master(r), dlm_dump_rsb(r););
1901
1902 high = grant_pending_convert(r, high, &cw);
1903 high = grant_pending_wait(r, high, &cw);
1904
1905 if (high == DLM_LOCK_IV)
1906 return;
1907
1908 /*
1909 * If there are locks left on the wait/convert queue then send blocking
1910 * ASTs to granted locks based on the largest requested mode (high)
1911 * found above.
1912 */
1913
1914 list_for_each_entry_safe(lkb, s, &r->res_grantqueue, lkb_statequeue) {
1915 if (lkb->lkb_bastfn && lock_requires_bast(lkb, high, cw)) {
1916 if (cw && high == DLM_LOCK_PR &&
1917 lkb->lkb_grmode == DLM_LOCK_PR)
1918 queue_bast(r, lkb, DLM_LOCK_CW);
1919 else
1920 queue_bast(r, lkb, high);
1921 lkb->lkb_highbast = high;
1922 }
1923 }
1924}
1925
1926static int modes_require_bast(struct dlm_lkb *gr, struct dlm_lkb *rq)
1927{
1928 if ((gr->lkb_grmode == DLM_LOCK_PR && rq->lkb_rqmode == DLM_LOCK_CW) ||
1929 (gr->lkb_grmode == DLM_LOCK_CW && rq->lkb_rqmode == DLM_LOCK_PR)) {
1930 if (gr->lkb_highbast < DLM_LOCK_EX)
1931 return 1;
1932 return 0;
1933 }
1934
1935 if (gr->lkb_highbast < rq->lkb_rqmode && !modes_compat(gr, rq))
1936 return 1;
1937 return 0;
1938}
1939
1940static void send_bast_queue(struct dlm_rsb *r, struct list_head *head,
1941 struct dlm_lkb *lkb)
1942{
1943 struct dlm_lkb *gr;
1944
1945 list_for_each_entry(gr, head, lkb_statequeue) {
1946 /* skip self when sending basts to convertqueue */
1947 if (gr == lkb)
1948 continue;
1949 if (gr->lkb_bastfn && modes_require_bast(gr, lkb)) {
1950 queue_bast(r, gr, lkb->lkb_rqmode);
1951 gr->lkb_highbast = lkb->lkb_rqmode;
1952 }
1953 }
1954}
1955
1956static void send_blocking_asts(struct dlm_rsb *r, struct dlm_lkb *lkb)
1957{
1958 send_bast_queue(r, &r->res_grantqueue, lkb);
1959}
1960
1961static void send_blocking_asts_all(struct dlm_rsb *r, struct dlm_lkb *lkb)
1962{
1963 send_bast_queue(r, &r->res_grantqueue, lkb);
1964 send_bast_queue(r, &r->res_convertqueue, lkb);
1965}
1966
1967/* set_master(r, lkb) -- set the master nodeid of a resource
1968
1969 The purpose of this function is to set the nodeid field in the given
1970 lkb using the nodeid field in the given rsb. If the rsb's nodeid is
1971 known, it can just be copied to the lkb and the function will return
1972 0. If the rsb's nodeid is _not_ known, it needs to be looked up
1973 before it can be copied to the lkb.
1974
1975 When the rsb nodeid is being looked up remotely, the initial lkb
1976 causing the lookup is kept on the ls_waiters list waiting for the
1977 lookup reply. Other lkb's waiting for the same rsb lookup are kept
1978 on the rsb's res_lookup list until the master is verified.
1979
1980 Return values:
1981 0: nodeid is set in rsb/lkb and the caller should go ahead and use it
1982 1: the rsb master is not available and the lkb has been placed on
1983 a wait queue
1984*/
1985
1986static int set_master(struct dlm_rsb *r, struct dlm_lkb *lkb)
1987{
1988 struct dlm_ls *ls = r->res_ls;
1989 int i, error, dir_nodeid, ret_nodeid, our_nodeid = dlm_our_nodeid();
1990
1991 if (rsb_flag(r, RSB_MASTER_UNCERTAIN)) {
1992 rsb_clear_flag(r, RSB_MASTER_UNCERTAIN);
1993 r->res_first_lkid = lkb->lkb_id;
1994 lkb->lkb_nodeid = r->res_nodeid;
1995 return 0;
1996 }
1997
1998 if (r->res_first_lkid && r->res_first_lkid != lkb->lkb_id) {
1999 list_add_tail(&lkb->lkb_rsb_lookup, &r->res_lookup);
2000 return 1;
2001 }
2002
2003 if (r->res_nodeid == 0) {
2004 lkb->lkb_nodeid = 0;
2005 return 0;
2006 }
2007
2008 if (r->res_nodeid > 0) {
2009 lkb->lkb_nodeid = r->res_nodeid;
2010 return 0;
2011 }
2012
2013 DLM_ASSERT(r->res_nodeid == -1, dlm_dump_rsb(r););
2014
2015 dir_nodeid = dlm_dir_nodeid(r);
2016
2017 if (dir_nodeid != our_nodeid) {
2018 r->res_first_lkid = lkb->lkb_id;
2019 send_lookup(r, lkb);
2020 return 1;
2021 }
2022
2023 for (i = 0; i < 2; i++) {
2024 /* It's possible for dlm_scand to remove an old rsb for
2025 this same resource from the toss list, us to create
2026 a new one, look up the master locally, and find it
2027 already exists just before dlm_scand does the
2028 dir_remove() on the previous rsb. */
2029
2030 error = dlm_dir_lookup(ls, our_nodeid, r->res_name,
2031 r->res_length, &ret_nodeid);
2032 if (!error)
2033 break;
2034 log_debug(ls, "dir_lookup error %d %s", error, r->res_name);
2035 schedule();
2036 }
2037 if (error && error != -EEXIST)
2038 return error;
2039
2040 if (ret_nodeid == our_nodeid) {
2041 r->res_first_lkid = 0;
2042 r->res_nodeid = 0;
2043 lkb->lkb_nodeid = 0;
2044 } else {
2045 r->res_first_lkid = lkb->lkb_id;
2046 r->res_nodeid = ret_nodeid;
2047 lkb->lkb_nodeid = ret_nodeid;
2048 }
2049 return 0;
2050}
2051
2052static void process_lookup_list(struct dlm_rsb *r)
2053{
2054 struct dlm_lkb *lkb, *safe;
2055
2056 list_for_each_entry_safe(lkb, safe, &r->res_lookup, lkb_rsb_lookup) {
2057 list_del_init(&lkb->lkb_rsb_lookup);
2058 _request_lock(r, lkb);
2059 schedule();
2060 }
2061}
2062
2063/* confirm_master -- confirm (or deny) an rsb's master nodeid */
2064
2065static void confirm_master(struct dlm_rsb *r, int error)
2066{
2067 struct dlm_lkb *lkb;
2068
2069 if (!r->res_first_lkid)
2070 return;
2071
2072 switch (error) {
2073 case 0:
2074 case -EINPROGRESS:
2075 r->res_first_lkid = 0;
2076 process_lookup_list(r);
2077 break;
2078
2079 case -EAGAIN:
2080 case -EBADR:
2081 case -ENOTBLK:
2082 /* the remote request failed and won't be retried (it was
2083 a NOQUEUE, or has been canceled/unlocked); make a waiting
2084 lkb the first_lkid */
2085
2086 r->res_first_lkid = 0;
2087
2088 if (!list_empty(&r->res_lookup)) {
2089 lkb = list_entry(r->res_lookup.next, struct dlm_lkb,
2090 lkb_rsb_lookup);
2091 list_del_init(&lkb->lkb_rsb_lookup);
2092 r->res_first_lkid = lkb->lkb_id;
2093 _request_lock(r, lkb);
2094 }
2095 break;
2096
2097 default:
2098 log_error(r->res_ls, "confirm_master unknown error %d", error);
2099 }
2100}
2101
2102static int set_lock_args(int mode, struct dlm_lksb *lksb, uint32_t flags,
2103 int namelen, unsigned long timeout_cs,
2104 void (*ast) (void *astparam),
2105 void *astparam,
2106 void (*bast) (void *astparam, int mode),
2107 struct dlm_args *args)
2108{
2109 int rv = -EINVAL;
2110
2111 /* check for invalid arg usage */
2112
2113 if (mode < 0 || mode > DLM_LOCK_EX)
2114 goto out;
2115
2116 if (!(flags & DLM_LKF_CONVERT) && (namelen > DLM_RESNAME_MAXLEN))
2117 goto out;
2118
2119 if (flags & DLM_LKF_CANCEL)
2120 goto out;
2121
2122 if (flags & DLM_LKF_QUECVT && !(flags & DLM_LKF_CONVERT))
2123 goto out;
2124
2125 if (flags & DLM_LKF_CONVDEADLK && !(flags & DLM_LKF_CONVERT))
2126 goto out;
2127
2128 if (flags & DLM_LKF_CONVDEADLK && flags & DLM_LKF_NOQUEUE)
2129 goto out;
2130
2131 if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_CONVERT)
2132 goto out;
2133
2134 if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_QUECVT)
2135 goto out;
2136
2137 if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_NOQUEUE)
2138 goto out;
2139
2140 if (flags & DLM_LKF_EXPEDITE && mode != DLM_LOCK_NL)
2141 goto out;
2142
2143 if (!ast || !lksb)
2144 goto out;
2145
2146 if (flags & DLM_LKF_VALBLK && !lksb->sb_lvbptr)
2147 goto out;
2148
2149 if (flags & DLM_LKF_CONVERT && !lksb->sb_lkid)
2150 goto out;
2151
2152 /* these args will be copied to the lkb in validate_lock_args,
2153 it cannot be done now because when converting locks, fields in
2154 an active lkb cannot be modified before locking the rsb */
2155
2156 args->flags = flags;
2157 args->astfn = ast;
2158 args->astparam = astparam;
2159 args->bastfn = bast;
2160 args->timeout = timeout_cs;
2161 args->mode = mode;
2162 args->lksb = lksb;
2163 rv = 0;
2164 out:
2165 return rv;
2166}
2167
2168static int set_unlock_args(uint32_t flags, void *astarg, struct dlm_args *args)
2169{
2170 if (flags & ~(DLM_LKF_CANCEL | DLM_LKF_VALBLK | DLM_LKF_IVVALBLK |
2171 DLM_LKF_FORCEUNLOCK))
2172 return -EINVAL;
2173
2174 if (flags & DLM_LKF_CANCEL && flags & DLM_LKF_FORCEUNLOCK)
2175 return -EINVAL;
2176
2177 args->flags = flags;
2178 args->astparam = astarg;
2179 return 0;
2180}
2181
2182static int validate_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
2183 struct dlm_args *args)
2184{
2185 int rv = -EINVAL;
2186
2187 if (args->flags & DLM_LKF_CONVERT) {
2188 if (lkb->lkb_flags & DLM_IFL_MSTCPY)
2189 goto out;
2190
2191 if (args->flags & DLM_LKF_QUECVT &&
2192 !__quecvt_compat_matrix[lkb->lkb_grmode+1][args->mode+1])
2193 goto out;
2194
2195 rv = -EBUSY;
2196 if (lkb->lkb_status != DLM_LKSTS_GRANTED)
2197 goto out;
2198
2199 if (lkb->lkb_wait_type)
2200 goto out;
2201
2202 if (is_overlap(lkb))
2203 goto out;
2204 }
2205
2206 lkb->lkb_exflags = args->flags;
2207 lkb->lkb_sbflags = 0;
2208 lkb->lkb_astfn = args->astfn;
2209 lkb->lkb_astparam = args->astparam;
2210 lkb->lkb_bastfn = args->bastfn;
2211 lkb->lkb_rqmode = args->mode;
2212 lkb->lkb_lksb = args->lksb;
2213 lkb->lkb_lvbptr = args->lksb->sb_lvbptr;
2214 lkb->lkb_ownpid = (int) current->pid;
2215 lkb->lkb_timeout_cs = args->timeout;
2216 rv = 0;
2217 out:
2218 if (rv)
2219 log_debug(ls, "validate_lock_args %d %x %x %x %d %d %s",
2220 rv, lkb->lkb_id, lkb->lkb_flags, args->flags,
2221 lkb->lkb_status, lkb->lkb_wait_type,
2222 lkb->lkb_resource->res_name);
2223 return rv;
2224}
2225
2226/* when dlm_unlock() sees -EBUSY with CANCEL/FORCEUNLOCK it returns 0
2227 for success */
2228
2229/* note: it's valid for lkb_nodeid/res_nodeid to be -1 when we get here
2230 because there may be a lookup in progress and it's valid to do
2231 cancel/unlockf on it */
2232
2233static int validate_unlock_args(struct dlm_lkb *lkb, struct dlm_args *args)
2234{
2235 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
2236 int rv = -EINVAL;
2237
2238 if (lkb->lkb_flags & DLM_IFL_MSTCPY) {
2239 log_error(ls, "unlock on MSTCPY %x", lkb->lkb_id);
2240 dlm_print_lkb(lkb);
2241 goto out;
2242 }
2243
2244 /* an lkb may still exist even though the lock is EOL'ed due to a
2245 cancel, unlock or failed noqueue request; an app can't use these
2246 locks; return same error as if the lkid had not been found at all */
2247
2248 if (lkb->lkb_flags & DLM_IFL_ENDOFLIFE) {
2249 log_debug(ls, "unlock on ENDOFLIFE %x", lkb->lkb_id);
2250 rv = -ENOENT;
2251 goto out;
2252 }
2253
2254 /* an lkb may be waiting for an rsb lookup to complete where the
2255 lookup was initiated by another lock */
2256
2257 if (!list_empty(&lkb->lkb_rsb_lookup)) {
2258 if (args->flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)) {
2259 log_debug(ls, "unlock on rsb_lookup %x", lkb->lkb_id);
2260 list_del_init(&lkb->lkb_rsb_lookup);
2261 queue_cast(lkb->lkb_resource, lkb,
2262 args->flags & DLM_LKF_CANCEL ?
2263 -DLM_ECANCEL : -DLM_EUNLOCK);
2264 unhold_lkb(lkb); /* undoes create_lkb() */
2265 }
2266 /* caller changes -EBUSY to 0 for CANCEL and FORCEUNLOCK */
2267 rv = -EBUSY;
2268 goto out;
2269 }
2270
2271 /* cancel not allowed with another cancel/unlock in progress */
2272
2273 if (args->flags & DLM_LKF_CANCEL) {
2274 if (lkb->lkb_exflags & DLM_LKF_CANCEL)
2275 goto out;
2276
2277 if (is_overlap(lkb))
2278 goto out;
2279
2280 /* don't let scand try to do a cancel */
2281 del_timeout(lkb);
2282
2283 if (lkb->lkb_flags & DLM_IFL_RESEND) {
2284 lkb->lkb_flags |= DLM_IFL_OVERLAP_CANCEL;
2285 rv = -EBUSY;
2286 goto out;
2287 }
2288
2289 /* there's nothing to cancel */
2290 if (lkb->lkb_status == DLM_LKSTS_GRANTED &&
2291 !lkb->lkb_wait_type) {
2292 rv = -EBUSY;
2293 goto out;
2294 }
2295
2296 switch (lkb->lkb_wait_type) {
2297 case DLM_MSG_LOOKUP:
2298 case DLM_MSG_REQUEST:
2299 lkb->lkb_flags |= DLM_IFL_OVERLAP_CANCEL;
2300 rv = -EBUSY;
2301 goto out;
2302 case DLM_MSG_UNLOCK:
2303 case DLM_MSG_CANCEL:
2304 goto out;
2305 }
2306 /* add_to_waiters() will set OVERLAP_CANCEL */
2307 goto out_ok;
2308 }
2309
2310 /* do we need to allow a force-unlock if there's a normal unlock
2311 already in progress? in what conditions could the normal unlock
2312 fail such that we'd want to send a force-unlock to be sure? */
2313
2314 if (args->flags & DLM_LKF_FORCEUNLOCK) {
2315 if (lkb->lkb_exflags & DLM_LKF_FORCEUNLOCK)
2316 goto out;
2317
2318 if (is_overlap_unlock(lkb))
2319 goto out;
2320
2321 /* don't let scand try to do a cancel */
2322 del_timeout(lkb);
2323
2324 if (lkb->lkb_flags & DLM_IFL_RESEND) {
2325 lkb->lkb_flags |= DLM_IFL_OVERLAP_UNLOCK;
2326 rv = -EBUSY;
2327 goto out;
2328 }
2329
2330 switch (lkb->lkb_wait_type) {
2331 case DLM_MSG_LOOKUP:
2332 case DLM_MSG_REQUEST:
2333 lkb->lkb_flags |= DLM_IFL_OVERLAP_UNLOCK;
2334 rv = -EBUSY;
2335 goto out;
2336 case DLM_MSG_UNLOCK:
2337 goto out;
2338 }
2339 /* add_to_waiters() will set OVERLAP_UNLOCK */
2340 goto out_ok;
2341 }
2342
2343 /* normal unlock not allowed if there's any op in progress */
2344 rv = -EBUSY;
2345 if (lkb->lkb_wait_type || lkb->lkb_wait_count)
2346 goto out;
2347
2348 out_ok:
2349 /* an overlapping op shouldn't blow away exflags from other op */
2350 lkb->lkb_exflags |= args->flags;
2351 lkb->lkb_sbflags = 0;
2352 lkb->lkb_astparam = args->astparam;
2353 rv = 0;
2354 out:
2355 if (rv)
2356 log_debug(ls, "validate_unlock_args %d %x %x %x %x %d %s", rv,
2357 lkb->lkb_id, lkb->lkb_flags, lkb->lkb_exflags,
2358 args->flags, lkb->lkb_wait_type,
2359 lkb->lkb_resource->res_name);
2360 return rv;
2361}
2362
2363/*
2364 * Four stage 4 varieties:
2365 * do_request(), do_convert(), do_unlock(), do_cancel()
2366 * These are called on the master node for the given lock and
2367 * from the central locking logic.
2368 */
2369
2370static int do_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
2371{
2372 int error = 0;
2373
2374 if (can_be_granted(r, lkb, 1, NULL)) {
2375 grant_lock(r, lkb);
2376 queue_cast(r, lkb, 0);
2377 goto out;
2378 }
2379
2380 if (can_be_queued(lkb)) {
2381 error = -EINPROGRESS;
2382 add_lkb(r, lkb, DLM_LKSTS_WAITING);
2383 add_timeout(lkb);
2384 goto out;
2385 }
2386
2387 error = -EAGAIN;
2388 queue_cast(r, lkb, -EAGAIN);
2389 out:
2390 return error;
2391}
2392
2393static void do_request_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
2394 int error)
2395{
2396 switch (error) {
2397 case -EAGAIN:
2398 if (force_blocking_asts(lkb))
2399 send_blocking_asts_all(r, lkb);
2400 break;
2401 case -EINPROGRESS:
2402 send_blocking_asts(r, lkb);
2403 break;
2404 }
2405}
2406
2407static int do_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
2408{
2409 int error = 0;
2410 int deadlk = 0;
2411
2412 /* changing an existing lock may allow others to be granted */
2413
2414 if (can_be_granted(r, lkb, 1, &deadlk)) {
2415 grant_lock(r, lkb);
2416 queue_cast(r, lkb, 0);
2417 goto out;
2418 }
2419
2420 /* can_be_granted() detected that this lock would block in a conversion
2421 deadlock, so we leave it on the granted queue and return EDEADLK in
2422 the ast for the convert. */
2423
2424 if (deadlk) {
2425 /* it's left on the granted queue */
2426 revert_lock(r, lkb);
2427 queue_cast(r, lkb, -EDEADLK);
2428 error = -EDEADLK;
2429 goto out;
2430 }
2431
2432 /* is_demoted() means the can_be_granted() above set the grmode
2433 to NL, and left us on the granted queue. This auto-demotion
2434 (due to CONVDEADLK) might mean other locks, and/or this lock, are
2435 now grantable. We have to try to grant other converting locks
2436 before we try again to grant this one. */
2437
2438 if (is_demoted(lkb)) {
2439 grant_pending_convert(r, DLM_LOCK_IV, NULL);
2440 if (_can_be_granted(r, lkb, 1)) {
2441 grant_lock(r, lkb);
2442 queue_cast(r, lkb, 0);
2443 goto out;
2444 }
2445 /* else fall through and move to convert queue */
2446 }
2447
2448 if (can_be_queued(lkb)) {
2449 error = -EINPROGRESS;
2450 del_lkb(r, lkb);
2451 add_lkb(r, lkb, DLM_LKSTS_CONVERT);
2452 add_timeout(lkb);
2453 goto out;
2454 }
2455
2456 error = -EAGAIN;
2457 queue_cast(r, lkb, -EAGAIN);
2458 out:
2459 return error;
2460}
2461
2462static void do_convert_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
2463 int error)
2464{
2465 switch (error) {
2466 case 0:
2467 grant_pending_locks(r);
2468 /* grant_pending_locks also sends basts */
2469 break;
2470 case -EAGAIN:
2471 if (force_blocking_asts(lkb))
2472 send_blocking_asts_all(r, lkb);
2473 break;
2474 case -EINPROGRESS:
2475 send_blocking_asts(r, lkb);
2476 break;
2477 }
2478}
2479
2480static int do_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
2481{
2482 remove_lock(r, lkb);
2483 queue_cast(r, lkb, -DLM_EUNLOCK);
2484 return -DLM_EUNLOCK;
2485}
2486
2487static void do_unlock_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
2488 int error)
2489{
2490 grant_pending_locks(r);
2491}
2492
2493/* returns: 0 did nothing, -DLM_ECANCEL canceled lock */
2494
2495static int do_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb)
2496{
2497 int error;
2498
2499 error = revert_lock(r, lkb);
2500 if (error) {
2501 queue_cast(r, lkb, -DLM_ECANCEL);
2502 return -DLM_ECANCEL;
2503 }
2504 return 0;
2505}
2506
2507static void do_cancel_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
2508 int error)
2509{
2510 if (error)
2511 grant_pending_locks(r);
2512}
2513
2514/*
2515 * Four stage 3 varieties:
2516 * _request_lock(), _convert_lock(), _unlock_lock(), _cancel_lock()
2517 */
2518
2519/* add a new lkb to a possibly new rsb, called by requesting process */
2520
2521static int _request_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
2522{
2523 int error;
2524
2525 /* set_master: sets lkb nodeid from r */
2526
2527 error = set_master(r, lkb);
2528 if (error < 0)
2529 goto out;
2530 if (error) {
2531 error = 0;
2532 goto out;
2533 }
2534
2535 if (is_remote(r)) {
2536 /* receive_request() calls do_request() on remote node */
2537 error = send_request(r, lkb);
2538 } else {
2539 error = do_request(r, lkb);
2540 /* for remote locks the request_reply is sent
2541 between do_request and do_request_effects */
2542 do_request_effects(r, lkb, error);
2543 }
2544 out:
2545 return error;
2546}
2547
2548/* change some property of an existing lkb, e.g. mode */
2549
2550static int _convert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
2551{
2552 int error;
2553
2554 if (is_remote(r)) {
2555 /* receive_convert() calls do_convert() on remote node */
2556 error = send_convert(r, lkb);
2557 } else {
2558 error = do_convert(r, lkb);
2559 /* for remote locks the convert_reply is sent
2560 between do_convert and do_convert_effects */
2561 do_convert_effects(r, lkb, error);
2562 }
2563
2564 return error;
2565}
2566
2567/* remove an existing lkb from the granted queue */
2568
2569static int _unlock_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
2570{
2571 int error;
2572
2573 if (is_remote(r)) {
2574 /* receive_unlock() calls do_unlock() on remote node */
2575 error = send_unlock(r, lkb);
2576 } else {
2577 error = do_unlock(r, lkb);
2578 /* for remote locks the unlock_reply is sent
2579 between do_unlock and do_unlock_effects */
2580 do_unlock_effects(r, lkb, error);
2581 }
2582
2583 return error;
2584}
2585
2586/* remove an existing lkb from the convert or wait queue */
2587
2588static int _cancel_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
2589{
2590 int error;
2591
2592 if (is_remote(r)) {
2593 /* receive_cancel() calls do_cancel() on remote node */
2594 error = send_cancel(r, lkb);
2595 } else {
2596 error = do_cancel(r, lkb);
2597 /* for remote locks the cancel_reply is sent
2598 between do_cancel and do_cancel_effects */
2599 do_cancel_effects(r, lkb, error);
2600 }
2601
2602 return error;
2603}
2604
2605/*
2606 * Four stage 2 varieties:
2607 * request_lock(), convert_lock(), unlock_lock(), cancel_lock()
2608 */
2609
2610static int request_lock(struct dlm_ls *ls, struct dlm_lkb *lkb, char *name,
2611 int len, struct dlm_args *args)
2612{
2613 struct dlm_rsb *r;
2614 int error;
2615
2616 error = validate_lock_args(ls, lkb, args);
2617 if (error)
2618 goto out;
2619
2620 error = find_rsb(ls, name, len, R_CREATE, &r);
2621 if (error)
2622 goto out;
2623
2624 lock_rsb(r);
2625
2626 attach_lkb(r, lkb);
2627 lkb->lkb_lksb->sb_lkid = lkb->lkb_id;
2628
2629 error = _request_lock(r, lkb);
2630
2631 unlock_rsb(r);
2632 put_rsb(r);
2633
2634 out:
2635 return error;
2636}
2637
2638static int convert_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
2639 struct dlm_args *args)
2640{
2641 struct dlm_rsb *r;
2642 int error;
2643
2644 r = lkb->lkb_resource;
2645
2646 hold_rsb(r);
2647 lock_rsb(r);
2648
2649 error = validate_lock_args(ls, lkb, args);
2650 if (error)
2651 goto out;
2652
2653 error = _convert_lock(r, lkb);
2654 out:
2655 unlock_rsb(r);
2656 put_rsb(r);
2657 return error;
2658}
2659
2660static int unlock_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
2661 struct dlm_args *args)
2662{
2663 struct dlm_rsb *r;
2664 int error;
2665
2666 r = lkb->lkb_resource;
2667
2668 hold_rsb(r);
2669 lock_rsb(r);
2670
2671 error = validate_unlock_args(lkb, args);
2672 if (error)
2673 goto out;
2674
2675 error = _unlock_lock(r, lkb);
2676 out:
2677 unlock_rsb(r);
2678 put_rsb(r);
2679 return error;
2680}
2681
2682static int cancel_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
2683 struct dlm_args *args)
2684{
2685 struct dlm_rsb *r;
2686 int error;
2687
2688 r = lkb->lkb_resource;
2689
2690 hold_rsb(r);
2691 lock_rsb(r);
2692
2693 error = validate_unlock_args(lkb, args);
2694 if (error)
2695 goto out;
2696
2697 error = _cancel_lock(r, lkb);
2698 out:
2699 unlock_rsb(r);
2700 put_rsb(r);
2701 return error;
2702}
2703
2704/*
2705 * Two stage 1 varieties: dlm_lock() and dlm_unlock()
2706 */
2707
2708int dlm_lock(dlm_lockspace_t *lockspace,
2709 int mode,
2710 struct dlm_lksb *lksb,
2711 uint32_t flags,
2712 void *name,
2713 unsigned int namelen,
2714 uint32_t parent_lkid,
2715 void (*ast) (void *astarg),
2716 void *astarg,
2717 void (*bast) (void *astarg, int mode))
2718{
2719 struct dlm_ls *ls;
2720 struct dlm_lkb *lkb;
2721 struct dlm_args args;
2722 int error, convert = flags & DLM_LKF_CONVERT;
2723
2724 ls = dlm_find_lockspace_local(lockspace);
2725 if (!ls)
2726 return -EINVAL;
2727
2728 dlm_lock_recovery(ls);
2729
2730 if (convert)
2731 error = find_lkb(ls, lksb->sb_lkid, &lkb);
2732 else
2733 error = create_lkb(ls, &lkb);
2734
2735 if (error)
2736 goto out;
2737
2738 error = set_lock_args(mode, lksb, flags, namelen, 0, ast,
2739 astarg, bast, &args);
2740 if (error)
2741 goto out_put;
2742
2743 if (convert)
2744 error = convert_lock(ls, lkb, &args);
2745 else
2746 error = request_lock(ls, lkb, name, namelen, &args);
2747
2748 if (error == -EINPROGRESS)
2749 error = 0;
2750 out_put:
2751 if (convert || error)
2752 __put_lkb(ls, lkb);
2753 if (error == -EAGAIN || error == -EDEADLK)
2754 error = 0;
2755 out:
2756 dlm_unlock_recovery(ls);
2757 dlm_put_lockspace(ls);
2758 return error;
2759}
2760
2761int dlm_unlock(dlm_lockspace_t *lockspace,
2762 uint32_t lkid,
2763 uint32_t flags,
2764 struct dlm_lksb *lksb,
2765 void *astarg)
2766{
2767 struct dlm_ls *ls;
2768 struct dlm_lkb *lkb;
2769 struct dlm_args args;
2770 int error;
2771
2772 ls = dlm_find_lockspace_local(lockspace);
2773 if (!ls)
2774 return -EINVAL;
2775
2776 dlm_lock_recovery(ls);
2777
2778 error = find_lkb(ls, lkid, &lkb);
2779 if (error)
2780 goto out;
2781
2782 error = set_unlock_args(flags, astarg, &args);
2783 if (error)
2784 goto out_put;
2785
2786 if (flags & DLM_LKF_CANCEL)
2787 error = cancel_lock(ls, lkb, &args);
2788 else
2789 error = unlock_lock(ls, lkb, &args);
2790
2791 if (error == -DLM_EUNLOCK || error == -DLM_ECANCEL)
2792 error = 0;
2793 if (error == -EBUSY && (flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)))
2794 error = 0;
2795 out_put:
2796 dlm_put_lkb(lkb);
2797 out:
2798 dlm_unlock_recovery(ls);
2799 dlm_put_lockspace(ls);
2800 return error;
2801}
2802
2803/*
2804 * send/receive routines for remote operations and replies
2805 *
2806 * send_args
2807 * send_common
2808 * send_request receive_request
2809 * send_convert receive_convert
2810 * send_unlock receive_unlock
2811 * send_cancel receive_cancel
2812 * send_grant receive_grant
2813 * send_bast receive_bast
2814 * send_lookup receive_lookup
2815 * send_remove receive_remove
2816 *
2817 * send_common_reply
2818 * receive_request_reply send_request_reply
2819 * receive_convert_reply send_convert_reply
2820 * receive_unlock_reply send_unlock_reply
2821 * receive_cancel_reply send_cancel_reply
2822 * receive_lookup_reply send_lookup_reply
2823 */
2824
2825static int _create_message(struct dlm_ls *ls, int mb_len,
2826 int to_nodeid, int mstype,
2827 struct dlm_message **ms_ret,
2828 struct dlm_mhandle **mh_ret)
2829{
2830 struct dlm_message *ms;
2831 struct dlm_mhandle *mh;
2832 char *mb;
2833
2834 /* get_buffer gives us a message handle (mh) that we need to
2835 pass into lowcomms_commit and a message buffer (mb) that we
2836 write our data into */
2837
2838 mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, GFP_NOFS, &mb);
2839 if (!mh)
2840 return -ENOBUFS;
2841
2842 memset(mb, 0, mb_len);
2843
2844 ms = (struct dlm_message *) mb;
2845
2846 ms->m_header.h_version = (DLM_HEADER_MAJOR | DLM_HEADER_MINOR);
2847 ms->m_header.h_lockspace = ls->ls_global_id;
2848 ms->m_header.h_nodeid = dlm_our_nodeid();
2849 ms->m_header.h_length = mb_len;
2850 ms->m_header.h_cmd = DLM_MSG;
2851
2852 ms->m_type = mstype;
2853
2854 *mh_ret = mh;
2855 *ms_ret = ms;
2856 return 0;
2857}
2858
2859static int create_message(struct dlm_rsb *r, struct dlm_lkb *lkb,
2860 int to_nodeid, int mstype,
2861 struct dlm_message **ms_ret,
2862 struct dlm_mhandle **mh_ret)
2863{
2864 int mb_len = sizeof(struct dlm_message);
2865
2866 switch (mstype) {
2867 case DLM_MSG_REQUEST:
2868 case DLM_MSG_LOOKUP:
2869 case DLM_MSG_REMOVE:
2870 mb_len += r->res_length;
2871 break;
2872 case DLM_MSG_CONVERT:
2873 case DLM_MSG_UNLOCK:
2874 case DLM_MSG_REQUEST_REPLY:
2875 case DLM_MSG_CONVERT_REPLY:
2876 case DLM_MSG_GRANT:
2877 if (lkb && lkb->lkb_lvbptr)
2878 mb_len += r->res_ls->ls_lvblen;
2879 break;
2880 }
2881
2882 return _create_message(r->res_ls, mb_len, to_nodeid, mstype,
2883 ms_ret, mh_ret);
2884}
2885
2886/* further lowcomms enhancements or alternate implementations may make
2887 the return value from this function useful at some point */
2888
2889static int send_message(struct dlm_mhandle *mh, struct dlm_message *ms)
2890{
2891 dlm_message_out(ms);
2892 dlm_lowcomms_commit_buffer(mh);
2893 return 0;
2894}
2895
2896static void send_args(struct dlm_rsb *r, struct dlm_lkb *lkb,
2897 struct dlm_message *ms)
2898{
2899 ms->m_nodeid = lkb->lkb_nodeid;
2900 ms->m_pid = lkb->lkb_ownpid;
2901 ms->m_lkid = lkb->lkb_id;
2902 ms->m_remid = lkb->lkb_remid;
2903 ms->m_exflags = lkb->lkb_exflags;
2904 ms->m_sbflags = lkb->lkb_sbflags;
2905 ms->m_flags = lkb->lkb_flags;
2906 ms->m_lvbseq = lkb->lkb_lvbseq;
2907 ms->m_status = lkb->lkb_status;
2908 ms->m_grmode = lkb->lkb_grmode;
2909 ms->m_rqmode = lkb->lkb_rqmode;
2910 ms->m_hash = r->res_hash;
2911
2912 /* m_result and m_bastmode are set from function args,
2913 not from lkb fields */
2914
2915 if (lkb->lkb_bastfn)
2916 ms->m_asts |= DLM_CB_BAST;
2917 if (lkb->lkb_astfn)
2918 ms->m_asts |= DLM_CB_CAST;
2919
2920 /* compare with switch in create_message; send_remove() doesn't
2921 use send_args() */
2922
2923 switch (ms->m_type) {
2924 case DLM_MSG_REQUEST:
2925 case DLM_MSG_LOOKUP:
2926 memcpy(ms->m_extra, r->res_name, r->res_length);
2927 break;
2928 case DLM_MSG_CONVERT:
2929 case DLM_MSG_UNLOCK:
2930 case DLM_MSG_REQUEST_REPLY:
2931 case DLM_MSG_CONVERT_REPLY:
2932 case DLM_MSG_GRANT:
2933 if (!lkb->lkb_lvbptr)
2934 break;
2935 memcpy(ms->m_extra, lkb->lkb_lvbptr, r->res_ls->ls_lvblen);
2936 break;
2937 }
2938}
2939
2940static int send_common(struct dlm_rsb *r, struct dlm_lkb *lkb, int mstype)
2941{
2942 struct dlm_message *ms;
2943 struct dlm_mhandle *mh;
2944 int to_nodeid, error;
2945
2946 to_nodeid = r->res_nodeid;
2947
2948 error = add_to_waiters(lkb, mstype, to_nodeid);
2949 if (error)
2950 return error;
2951
2952 error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh);
2953 if (error)
2954 goto fail;
2955
2956 send_args(r, lkb, ms);
2957
2958 error = send_message(mh, ms);
2959 if (error)
2960 goto fail;
2961 return 0;
2962
2963 fail:
2964 remove_from_waiters(lkb, msg_reply_type(mstype));
2965 return error;
2966}
2967
2968static int send_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
2969{
2970 return send_common(r, lkb, DLM_MSG_REQUEST);
2971}
2972
2973static int send_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
2974{
2975 int error;
2976
2977 error = send_common(r, lkb, DLM_MSG_CONVERT);
2978
2979 /* down conversions go without a reply from the master */
2980 if (!error && down_conversion(lkb)) {
2981 remove_from_waiters(lkb, DLM_MSG_CONVERT_REPLY);
2982 r->res_ls->ls_stub_ms.m_flags = DLM_IFL_STUB_MS;
2983 r->res_ls->ls_stub_ms.m_type = DLM_MSG_CONVERT_REPLY;
2984 r->res_ls->ls_stub_ms.m_result = 0;
2985 __receive_convert_reply(r, lkb, &r->res_ls->ls_stub_ms);
2986 }
2987
2988 return error;
2989}
2990
2991/* FIXME: if this lkb is the only lock we hold on the rsb, then set
2992 MASTER_UNCERTAIN to force the next request on the rsb to confirm
2993 that the master is still correct. */
2994
2995static int send_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
2996{
2997 return send_common(r, lkb, DLM_MSG_UNLOCK);
2998}
2999
3000static int send_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb)
3001{
3002 return send_common(r, lkb, DLM_MSG_CANCEL);
3003}
3004
3005static int send_grant(struct dlm_rsb *r, struct dlm_lkb *lkb)
3006{
3007 struct dlm_message *ms;
3008 struct dlm_mhandle *mh;
3009 int to_nodeid, error;
3010
3011 to_nodeid = lkb->lkb_nodeid;
3012
3013 error = create_message(r, lkb, to_nodeid, DLM_MSG_GRANT, &ms, &mh);
3014 if (error)
3015 goto out;
3016
3017 send_args(r, lkb, ms);
3018
3019 ms->m_result = 0;
3020
3021 error = send_message(mh, ms);
3022 out:
3023 return error;
3024}
3025
3026static int send_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int mode)
3027{
3028 struct dlm_message *ms;
3029 struct dlm_mhandle *mh;
3030 int to_nodeid, error;
3031
3032 to_nodeid = lkb->lkb_nodeid;
3033
3034 error = create_message(r, NULL, to_nodeid, DLM_MSG_BAST, &ms, &mh);
3035 if (error)
3036 goto out;
3037
3038 send_args(r, lkb, ms);
3039
3040 ms->m_bastmode = mode;
3041
3042 error = send_message(mh, ms);
3043 out:
3044 return error;
3045}
3046
3047static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb)
3048{
3049 struct dlm_message *ms;
3050 struct dlm_mhandle *mh;
3051 int to_nodeid, error;
3052
3053 to_nodeid = dlm_dir_nodeid(r);
3054
3055 error = add_to_waiters(lkb, DLM_MSG_LOOKUP, to_nodeid);
3056 if (error)
3057 return error;
3058
3059 error = create_message(r, NULL, to_nodeid, DLM_MSG_LOOKUP, &ms, &mh);
3060 if (error)
3061 goto fail;
3062
3063 send_args(r, lkb, ms);
3064
3065 error = send_message(mh, ms);
3066 if (error)
3067 goto fail;
3068 return 0;
3069
3070 fail:
3071 remove_from_waiters(lkb, DLM_MSG_LOOKUP_REPLY);
3072 return error;
3073}
3074
3075static int send_remove(struct dlm_rsb *r)
3076{
3077 struct dlm_message *ms;
3078 struct dlm_mhandle *mh;
3079 int to_nodeid, error;
3080
3081 to_nodeid = dlm_dir_nodeid(r);
3082
3083 error = create_message(r, NULL, to_nodeid, DLM_MSG_REMOVE, &ms, &mh);
3084 if (error)
3085 goto out;
3086
3087 memcpy(ms->m_extra, r->res_name, r->res_length);
3088 ms->m_hash = r->res_hash;
3089
3090 error = send_message(mh, ms);
3091 out:
3092 return error;
3093}
3094
3095static int send_common_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
3096 int mstype, int rv)
3097{
3098 struct dlm_message *ms;
3099 struct dlm_mhandle *mh;
3100 int to_nodeid, error;
3101
3102 to_nodeid = lkb->lkb_nodeid;
3103
3104 error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh);
3105 if (error)
3106 goto out;
3107
3108 send_args(r, lkb, ms);
3109
3110 ms->m_result = rv;
3111
3112 error = send_message(mh, ms);
3113 out:
3114 return error;
3115}
3116
3117static int send_request_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3118{
3119 return send_common_reply(r, lkb, DLM_MSG_REQUEST_REPLY, rv);
3120}
3121
3122static int send_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3123{
3124 return send_common_reply(r, lkb, DLM_MSG_CONVERT_REPLY, rv);
3125}
3126
3127static int send_unlock_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3128{
3129 return send_common_reply(r, lkb, DLM_MSG_UNLOCK_REPLY, rv);
3130}
3131
3132static int send_cancel_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3133{
3134 return send_common_reply(r, lkb, DLM_MSG_CANCEL_REPLY, rv);
3135}
3136
3137static int send_lookup_reply(struct dlm_ls *ls, struct dlm_message *ms_in,
3138 int ret_nodeid, int rv)
3139{
3140 struct dlm_rsb *r = &ls->ls_stub_rsb;
3141 struct dlm_message *ms;
3142 struct dlm_mhandle *mh;
3143 int error, nodeid = ms_in->m_header.h_nodeid;
3144
3145 error = create_message(r, NULL, nodeid, DLM_MSG_LOOKUP_REPLY, &ms, &mh);
3146 if (error)
3147 goto out;
3148
3149 ms->m_lkid = ms_in->m_lkid;
3150 ms->m_result = rv;
3151 ms->m_nodeid = ret_nodeid;
3152
3153 error = send_message(mh, ms);
3154 out:
3155 return error;
3156}
3157
3158/* which args we save from a received message depends heavily on the type
3159 of message, unlike the send side where we can safely send everything about
3160 the lkb for any type of message */
3161
3162static void receive_flags(struct dlm_lkb *lkb, struct dlm_message *ms)
3163{
3164 lkb->lkb_exflags = ms->m_exflags;
3165 lkb->lkb_sbflags = ms->m_sbflags;
3166 lkb->lkb_flags = (lkb->lkb_flags & 0xFFFF0000) |
3167 (ms->m_flags & 0x0000FFFF);
3168}
3169
3170static void receive_flags_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
3171{
3172 if (ms->m_flags == DLM_IFL_STUB_MS)
3173 return;
3174
3175 lkb->lkb_sbflags = ms->m_sbflags;
3176 lkb->lkb_flags = (lkb->lkb_flags & 0xFFFF0000) |
3177 (ms->m_flags & 0x0000FFFF);
3178}
3179
3180static int receive_extralen(struct dlm_message *ms)
3181{
3182 return (ms->m_header.h_length - sizeof(struct dlm_message));
3183}
3184
3185static int receive_lvb(struct dlm_ls *ls, struct dlm_lkb *lkb,
3186 struct dlm_message *ms)
3187{
3188 int len;
3189
3190 if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
3191 if (!lkb->lkb_lvbptr)
3192 lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
3193 if (!lkb->lkb_lvbptr)
3194 return -ENOMEM;
3195 len = receive_extralen(ms);
3196 if (len > DLM_RESNAME_MAXLEN)
3197 len = DLM_RESNAME_MAXLEN;
3198 memcpy(lkb->lkb_lvbptr, ms->m_extra, len);
3199 }
3200 return 0;
3201}
3202
3203static void fake_bastfn(void *astparam, int mode)
3204{
3205 log_print("fake_bastfn should not be called");
3206}
3207
3208static void fake_astfn(void *astparam)
3209{
3210 log_print("fake_astfn should not be called");
3211}
3212
3213static int receive_request_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
3214 struct dlm_message *ms)
3215{
3216 lkb->lkb_nodeid = ms->m_header.h_nodeid;
3217 lkb->lkb_ownpid = ms->m_pid;
3218 lkb->lkb_remid = ms->m_lkid;
3219 lkb->lkb_grmode = DLM_LOCK_IV;
3220 lkb->lkb_rqmode = ms->m_rqmode;
3221
3222 lkb->lkb_bastfn = (ms->m_asts & DLM_CB_BAST) ? &fake_bastfn : NULL;
3223 lkb->lkb_astfn = (ms->m_asts & DLM_CB_CAST) ? &fake_astfn : NULL;
3224
3225 if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
3226 /* lkb was just created so there won't be an lvb yet */
3227 lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
3228 if (!lkb->lkb_lvbptr)
3229 return -ENOMEM;
3230 }
3231
3232 return 0;
3233}
3234
3235static int receive_convert_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
3236 struct dlm_message *ms)
3237{
3238 if (lkb->lkb_status != DLM_LKSTS_GRANTED)
3239 return -EBUSY;
3240
3241 if (receive_lvb(ls, lkb, ms))
3242 return -ENOMEM;
3243
3244 lkb->lkb_rqmode = ms->m_rqmode;
3245 lkb->lkb_lvbseq = ms->m_lvbseq;
3246
3247 return 0;
3248}
3249
3250static int receive_unlock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
3251 struct dlm_message *ms)
3252{
3253 if (receive_lvb(ls, lkb, ms))
3254 return -ENOMEM;
3255 return 0;
3256}
3257
3258/* We fill in the stub-lkb fields with the info that send_xxxx_reply()
3259 uses to send a reply and that the remote end uses to process the reply. */
3260
3261static void setup_stub_lkb(struct dlm_ls *ls, struct dlm_message *ms)
3262{
3263 struct dlm_lkb *lkb = &ls->ls_stub_lkb;
3264 lkb->lkb_nodeid = ms->m_header.h_nodeid;
3265 lkb->lkb_remid = ms->m_lkid;
3266}
3267
3268/* This is called after the rsb is locked so that we can safely inspect
3269 fields in the lkb. */
3270
3271static int validate_message(struct dlm_lkb *lkb, struct dlm_message *ms)
3272{
3273 int from = ms->m_header.h_nodeid;
3274 int error = 0;
3275
3276 switch (ms->m_type) {
3277 case DLM_MSG_CONVERT:
3278 case DLM_MSG_UNLOCK:
3279 case DLM_MSG_CANCEL:
3280 if (!is_master_copy(lkb) || lkb->lkb_nodeid != from)
3281 error = -EINVAL;
3282 break;
3283
3284 case DLM_MSG_CONVERT_REPLY:
3285 case DLM_MSG_UNLOCK_REPLY:
3286 case DLM_MSG_CANCEL_REPLY:
3287 case DLM_MSG_GRANT:
3288 case DLM_MSG_BAST:
3289 if (!is_process_copy(lkb) || lkb->lkb_nodeid != from)
3290 error = -EINVAL;
3291 break;
3292
3293 case DLM_MSG_REQUEST_REPLY:
3294 if (!is_process_copy(lkb))
3295 error = -EINVAL;
3296 else if (lkb->lkb_nodeid != -1 && lkb->lkb_nodeid != from)
3297 error = -EINVAL;
3298 break;
3299
3300 default:
3301 error = -EINVAL;
3302 }
3303
3304 if (error)
3305 log_error(lkb->lkb_resource->res_ls,
3306 "ignore invalid message %d from %d %x %x %x %d",
3307 ms->m_type, from, lkb->lkb_id, lkb->lkb_remid,
3308 lkb->lkb_flags, lkb->lkb_nodeid);
3309 return error;
3310}
3311
3312static void receive_request(struct dlm_ls *ls, struct dlm_message *ms)
3313{
3314 struct dlm_lkb *lkb;
3315 struct dlm_rsb *r;
3316 int error, namelen;
3317
3318 error = create_lkb(ls, &lkb);
3319 if (error)
3320 goto fail;
3321
3322 receive_flags(lkb, ms);
3323 lkb->lkb_flags |= DLM_IFL_MSTCPY;
3324 error = receive_request_args(ls, lkb, ms);
3325 if (error) {
3326 __put_lkb(ls, lkb);
3327 goto fail;
3328 }
3329
3330 namelen = receive_extralen(ms);
3331
3332 error = find_rsb(ls, ms->m_extra, namelen, R_MASTER, &r);
3333 if (error) {
3334 __put_lkb(ls, lkb);
3335 goto fail;
3336 }
3337
3338 lock_rsb(r);
3339
3340 attach_lkb(r, lkb);
3341 error = do_request(r, lkb);
3342 send_request_reply(r, lkb, error);
3343 do_request_effects(r, lkb, error);
3344
3345 unlock_rsb(r);
3346 put_rsb(r);
3347
3348 if (error == -EINPROGRESS)
3349 error = 0;
3350 if (error)
3351 dlm_put_lkb(lkb);
3352 return;
3353
3354 fail:
3355 setup_stub_lkb(ls, ms);
3356 send_request_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
3357}
3358
3359static void receive_convert(struct dlm_ls *ls, struct dlm_message *ms)
3360{
3361 struct dlm_lkb *lkb;
3362 struct dlm_rsb *r;
3363 int error, reply = 1;
3364
3365 error = find_lkb(ls, ms->m_remid, &lkb);
3366 if (error)
3367 goto fail;
3368
3369 r = lkb->lkb_resource;
3370
3371 hold_rsb(r);
3372 lock_rsb(r);
3373
3374 error = validate_message(lkb, ms);
3375 if (error)
3376 goto out;
3377
3378 receive_flags(lkb, ms);
3379
3380 error = receive_convert_args(ls, lkb, ms);
3381 if (error) {
3382 send_convert_reply(r, lkb, error);
3383 goto out;
3384 }
3385
3386 reply = !down_conversion(lkb);
3387
3388 error = do_convert(r, lkb);
3389 if (reply)
3390 send_convert_reply(r, lkb, error);
3391 do_convert_effects(r, lkb, error);
3392 out:
3393 unlock_rsb(r);
3394 put_rsb(r);
3395 dlm_put_lkb(lkb);
3396 return;
3397
3398 fail:
3399 setup_stub_lkb(ls, ms);
3400 send_convert_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
3401}
3402
3403static void receive_unlock(struct dlm_ls *ls, struct dlm_message *ms)
3404{
3405 struct dlm_lkb *lkb;
3406 struct dlm_rsb *r;
3407 int error;
3408
3409 error = find_lkb(ls, ms->m_remid, &lkb);
3410 if (error)
3411 goto fail;
3412
3413 r = lkb->lkb_resource;
3414
3415 hold_rsb(r);
3416 lock_rsb(r);
3417
3418 error = validate_message(lkb, ms);
3419 if (error)
3420 goto out;
3421
3422 receive_flags(lkb, ms);
3423
3424 error = receive_unlock_args(ls, lkb, ms);
3425 if (error) {
3426 send_unlock_reply(r, lkb, error);
3427 goto out;
3428 }
3429
3430 error = do_unlock(r, lkb);
3431 send_unlock_reply(r, lkb, error);
3432 do_unlock_effects(r, lkb, error);
3433 out:
3434 unlock_rsb(r);
3435 put_rsb(r);
3436 dlm_put_lkb(lkb);
3437 return;
3438
3439 fail:
3440 setup_stub_lkb(ls, ms);
3441 send_unlock_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
3442}
3443
3444static void receive_cancel(struct dlm_ls *ls, struct dlm_message *ms)
3445{
3446 struct dlm_lkb *lkb;
3447 struct dlm_rsb *r;
3448 int error;
3449
3450 error = find_lkb(ls, ms->m_remid, &lkb);
3451 if (error)
3452 goto fail;
3453
3454 receive_flags(lkb, ms);
3455
3456 r = lkb->lkb_resource;
3457
3458 hold_rsb(r);
3459 lock_rsb(r);
3460
3461 error = validate_message(lkb, ms);
3462 if (error)
3463 goto out;
3464
3465 error = do_cancel(r, lkb);
3466 send_cancel_reply(r, lkb, error);
3467 do_cancel_effects(r, lkb, error);
3468 out:
3469 unlock_rsb(r);
3470 put_rsb(r);
3471 dlm_put_lkb(lkb);
3472 return;
3473
3474 fail:
3475 setup_stub_lkb(ls, ms);
3476 send_cancel_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
3477}
3478
3479static void receive_grant(struct dlm_ls *ls, struct dlm_message *ms)
3480{
3481 struct dlm_lkb *lkb;
3482 struct dlm_rsb *r;
3483 int error;
3484
3485 error = find_lkb(ls, ms->m_remid, &lkb);
3486 if (error) {
3487 log_debug(ls, "receive_grant from %d no lkb %x",
3488 ms->m_header.h_nodeid, ms->m_remid);
3489 return;
3490 }
3491
3492 r = lkb->lkb_resource;
3493
3494 hold_rsb(r);
3495 lock_rsb(r);
3496
3497 error = validate_message(lkb, ms);
3498 if (error)
3499 goto out;
3500
3501 receive_flags_reply(lkb, ms);
3502 if (is_altmode(lkb))
3503 munge_altmode(lkb, ms);
3504 grant_lock_pc(r, lkb, ms);
3505 queue_cast(r, lkb, 0);
3506 out:
3507 unlock_rsb(r);
3508 put_rsb(r);
3509 dlm_put_lkb(lkb);
3510}
3511
3512static void receive_bast(struct dlm_ls *ls, struct dlm_message *ms)
3513{
3514 struct dlm_lkb *lkb;
3515 struct dlm_rsb *r;
3516 int error;
3517
3518 error = find_lkb(ls, ms->m_remid, &lkb);
3519 if (error) {
3520 log_debug(ls, "receive_bast from %d no lkb %x",
3521 ms->m_header.h_nodeid, ms->m_remid);
3522 return;
3523 }
3524
3525 r = lkb->lkb_resource;
3526
3527 hold_rsb(r);
3528 lock_rsb(r);
3529
3530 error = validate_message(lkb, ms);
3531 if (error)
3532 goto out;
3533
3534 queue_bast(r, lkb, ms->m_bastmode);
3535 out:
3536 unlock_rsb(r);
3537 put_rsb(r);
3538 dlm_put_lkb(lkb);
3539}
3540
3541static void receive_lookup(struct dlm_ls *ls, struct dlm_message *ms)
3542{
3543 int len, error, ret_nodeid, dir_nodeid, from_nodeid, our_nodeid;
3544
3545 from_nodeid = ms->m_header.h_nodeid;
3546 our_nodeid = dlm_our_nodeid();
3547
3548 len = receive_extralen(ms);
3549
3550 dir_nodeid = dlm_hash2nodeid(ls, ms->m_hash);
3551 if (dir_nodeid != our_nodeid) {
3552 log_error(ls, "lookup dir_nodeid %d from %d",
3553 dir_nodeid, from_nodeid);
3554 error = -EINVAL;
3555 ret_nodeid = -1;
3556 goto out;
3557 }
3558
3559 error = dlm_dir_lookup(ls, from_nodeid, ms->m_extra, len, &ret_nodeid);
3560
3561 /* Optimization: we're master so treat lookup as a request */
3562 if (!error && ret_nodeid == our_nodeid) {
3563 receive_request(ls, ms);
3564 return;
3565 }
3566 out:
3567 send_lookup_reply(ls, ms, ret_nodeid, error);
3568}
3569
3570static void receive_remove(struct dlm_ls *ls, struct dlm_message *ms)
3571{
3572 int len, dir_nodeid, from_nodeid;
3573
3574 from_nodeid = ms->m_header.h_nodeid;
3575
3576 len = receive_extralen(ms);
3577
3578 dir_nodeid = dlm_hash2nodeid(ls, ms->m_hash);
3579 if (dir_nodeid != dlm_our_nodeid()) {
3580 log_error(ls, "remove dir entry dir_nodeid %d from %d",
3581 dir_nodeid, from_nodeid);
3582 return;
3583 }
3584
3585 dlm_dir_remove_entry(ls, from_nodeid, ms->m_extra, len);
3586}
3587
3588static void receive_purge(struct dlm_ls *ls, struct dlm_message *ms)
3589{
3590 do_purge(ls, ms->m_nodeid, ms->m_pid);
3591}
3592
3593static void receive_request_reply(struct dlm_ls *ls, struct dlm_message *ms)
3594{
3595 struct dlm_lkb *lkb;
3596 struct dlm_rsb *r;
3597 int error, mstype, result;
3598
3599 error = find_lkb(ls, ms->m_remid, &lkb);
3600 if (error) {
3601 log_debug(ls, "receive_request_reply from %d no lkb %x",
3602 ms->m_header.h_nodeid, ms->m_remid);
3603 return;
3604 }
3605
3606 r = lkb->lkb_resource;
3607 hold_rsb(r);
3608 lock_rsb(r);
3609
3610 error = validate_message(lkb, ms);
3611 if (error)
3612 goto out;
3613
3614 mstype = lkb->lkb_wait_type;
3615 error = remove_from_waiters(lkb, DLM_MSG_REQUEST_REPLY);
3616 if (error)
3617 goto out;
3618
3619 /* Optimization: the dir node was also the master, so it took our
3620 lookup as a request and sent request reply instead of lookup reply */
3621 if (mstype == DLM_MSG_LOOKUP) {
3622 r->res_nodeid = ms->m_header.h_nodeid;
3623 lkb->lkb_nodeid = r->res_nodeid;
3624 }
3625
3626 /* this is the value returned from do_request() on the master */
3627 result = ms->m_result;
3628
3629 switch (result) {
3630 case -EAGAIN:
3631 /* request would block (be queued) on remote master */
3632 queue_cast(r, lkb, -EAGAIN);
3633 confirm_master(r, -EAGAIN);
3634 unhold_lkb(lkb); /* undoes create_lkb() */
3635 break;
3636
3637 case -EINPROGRESS:
3638 case 0:
3639 /* request was queued or granted on remote master */
3640 receive_flags_reply(lkb, ms);
3641 lkb->lkb_remid = ms->m_lkid;
3642 if (is_altmode(lkb))
3643 munge_altmode(lkb, ms);
3644 if (result) {
3645 add_lkb(r, lkb, DLM_LKSTS_WAITING);
3646 add_timeout(lkb);
3647 } else {
3648 grant_lock_pc(r, lkb, ms);
3649 queue_cast(r, lkb, 0);
3650 }
3651 confirm_master(r, result);
3652 break;
3653
3654 case -EBADR:
3655 case -ENOTBLK:
3656 /* find_rsb failed to find rsb or rsb wasn't master */
3657 log_debug(ls, "receive_request_reply %x %x master diff %d %d",
3658 lkb->lkb_id, lkb->lkb_flags, r->res_nodeid, result);
3659 r->res_nodeid = -1;
3660 lkb->lkb_nodeid = -1;
3661
3662 if (is_overlap(lkb)) {
3663 /* we'll ignore error in cancel/unlock reply */
3664 queue_cast_overlap(r, lkb);
3665 confirm_master(r, result);
3666 unhold_lkb(lkb); /* undoes create_lkb() */
3667 } else
3668 _request_lock(r, lkb);
3669 break;
3670
3671 default:
3672 log_error(ls, "receive_request_reply %x error %d",
3673 lkb->lkb_id, result);
3674 }
3675
3676 if (is_overlap_unlock(lkb) && (result == 0 || result == -EINPROGRESS)) {
3677 log_debug(ls, "receive_request_reply %x result %d unlock",
3678 lkb->lkb_id, result);
3679 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
3680 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
3681 send_unlock(r, lkb);
3682 } else if (is_overlap_cancel(lkb) && (result == -EINPROGRESS)) {
3683 log_debug(ls, "receive_request_reply %x cancel", lkb->lkb_id);
3684 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
3685 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
3686 send_cancel(r, lkb);
3687 } else {
3688 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
3689 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
3690 }
3691 out:
3692 unlock_rsb(r);
3693 put_rsb(r);
3694 dlm_put_lkb(lkb);
3695}
3696
3697static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
3698 struct dlm_message *ms)
3699{
3700 /* this is the value returned from do_convert() on the master */
3701 switch (ms->m_result) {
3702 case -EAGAIN:
3703 /* convert would block (be queued) on remote master */
3704 queue_cast(r, lkb, -EAGAIN);
3705 break;
3706
3707 case -EDEADLK:
3708 receive_flags_reply(lkb, ms);
3709 revert_lock_pc(r, lkb);
3710 queue_cast(r, lkb, -EDEADLK);
3711 break;
3712
3713 case -EINPROGRESS:
3714 /* convert was queued on remote master */
3715 receive_flags_reply(lkb, ms);
3716 if (is_demoted(lkb))
3717 munge_demoted(lkb);
3718 del_lkb(r, lkb);
3719 add_lkb(r, lkb, DLM_LKSTS_CONVERT);
3720 add_timeout(lkb);
3721 break;
3722
3723 case 0:
3724 /* convert was granted on remote master */
3725 receive_flags_reply(lkb, ms);
3726 if (is_demoted(lkb))
3727 munge_demoted(lkb);
3728 grant_lock_pc(r, lkb, ms);
3729 queue_cast(r, lkb, 0);
3730 break;
3731
3732 default:
3733 log_error(r->res_ls, "receive_convert_reply %x error %d",
3734 lkb->lkb_id, ms->m_result);
3735 }
3736}
3737
3738static void _receive_convert_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
3739{
3740 struct dlm_rsb *r = lkb->lkb_resource;
3741 int error;
3742
3743 hold_rsb(r);
3744 lock_rsb(r);
3745
3746 error = validate_message(lkb, ms);
3747 if (error)
3748 goto out;
3749
3750 /* stub reply can happen with waiters_mutex held */
3751 error = remove_from_waiters_ms(lkb, ms);
3752 if (error)
3753 goto out;
3754
3755 __receive_convert_reply(r, lkb, ms);
3756 out:
3757 unlock_rsb(r);
3758 put_rsb(r);
3759}
3760
3761static void receive_convert_reply(struct dlm_ls *ls, struct dlm_message *ms)
3762{
3763 struct dlm_lkb *lkb;
3764 int error;
3765
3766 error = find_lkb(ls, ms->m_remid, &lkb);
3767 if (error) {
3768 log_debug(ls, "receive_convert_reply from %d no lkb %x",
3769 ms->m_header.h_nodeid, ms->m_remid);
3770 return;
3771 }
3772
3773 _receive_convert_reply(lkb, ms);
3774 dlm_put_lkb(lkb);
3775}
3776
3777static void _receive_unlock_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
3778{
3779 struct dlm_rsb *r = lkb->lkb_resource;
3780 int error;
3781
3782 hold_rsb(r);
3783 lock_rsb(r);
3784
3785 error = validate_message(lkb, ms);
3786 if (error)
3787 goto out;
3788
3789 /* stub reply can happen with waiters_mutex held */
3790 error = remove_from_waiters_ms(lkb, ms);
3791 if (error)
3792 goto out;
3793
3794 /* this is the value returned from do_unlock() on the master */
3795
3796 switch (ms->m_result) {
3797 case -DLM_EUNLOCK:
3798 receive_flags_reply(lkb, ms);
3799 remove_lock_pc(r, lkb);
3800 queue_cast(r, lkb, -DLM_EUNLOCK);
3801 break;
3802 case -ENOENT:
3803 break;
3804 default:
3805 log_error(r->res_ls, "receive_unlock_reply %x error %d",
3806 lkb->lkb_id, ms->m_result);
3807 }
3808 out:
3809 unlock_rsb(r);
3810 put_rsb(r);
3811}
3812
3813static void receive_unlock_reply(struct dlm_ls *ls, struct dlm_message *ms)
3814{
3815 struct dlm_lkb *lkb;
3816 int error;
3817
3818 error = find_lkb(ls, ms->m_remid, &lkb);
3819 if (error) {
3820 log_debug(ls, "receive_unlock_reply from %d no lkb %x",
3821 ms->m_header.h_nodeid, ms->m_remid);
3822 return;
3823 }
3824
3825 _receive_unlock_reply(lkb, ms);
3826 dlm_put_lkb(lkb);
3827}
3828
3829static void _receive_cancel_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
3830{
3831 struct dlm_rsb *r = lkb->lkb_resource;
3832 int error;
3833
3834 hold_rsb(r);
3835 lock_rsb(r);
3836
3837 error = validate_message(lkb, ms);
3838 if (error)
3839 goto out;
3840
3841 /* stub reply can happen with waiters_mutex held */
3842 error = remove_from_waiters_ms(lkb, ms);
3843 if (error)
3844 goto out;
3845
3846 /* this is the value returned from do_cancel() on the master */
3847
3848 switch (ms->m_result) {
3849 case -DLM_ECANCEL:
3850 receive_flags_reply(lkb, ms);
3851 revert_lock_pc(r, lkb);
3852 queue_cast(r, lkb, -DLM_ECANCEL);
3853 break;
3854 case 0:
3855 break;
3856 default:
3857 log_error(r->res_ls, "receive_cancel_reply %x error %d",
3858 lkb->lkb_id, ms->m_result);
3859 }
3860 out:
3861 unlock_rsb(r);
3862 put_rsb(r);
3863}
3864
3865static void receive_cancel_reply(struct dlm_ls *ls, struct dlm_message *ms)
3866{
3867 struct dlm_lkb *lkb;
3868 int error;
3869
3870 error = find_lkb(ls, ms->m_remid, &lkb);
3871 if (error) {
3872 log_debug(ls, "receive_cancel_reply from %d no lkb %x",
3873 ms->m_header.h_nodeid, ms->m_remid);
3874 return;
3875 }
3876
3877 _receive_cancel_reply(lkb, ms);
3878 dlm_put_lkb(lkb);
3879}
3880
3881static void receive_lookup_reply(struct dlm_ls *ls, struct dlm_message *ms)
3882{
3883 struct dlm_lkb *lkb;
3884 struct dlm_rsb *r;
3885 int error, ret_nodeid;
3886
3887 error = find_lkb(ls, ms->m_lkid, &lkb);
3888 if (error) {
3889 log_error(ls, "receive_lookup_reply no lkb");
3890 return;
3891 }
3892
3893 /* ms->m_result is the value returned by dlm_dir_lookup on dir node
3894 FIXME: will a non-zero error ever be returned? */
3895
3896 r = lkb->lkb_resource;
3897 hold_rsb(r);
3898 lock_rsb(r);
3899
3900 error = remove_from_waiters(lkb, DLM_MSG_LOOKUP_REPLY);
3901 if (error)
3902 goto out;
3903
3904 ret_nodeid = ms->m_nodeid;
3905 if (ret_nodeid == dlm_our_nodeid()) {
3906 r->res_nodeid = 0;
3907 ret_nodeid = 0;
3908 r->res_first_lkid = 0;
3909 } else {
3910 /* set_master() will copy res_nodeid to lkb_nodeid */
3911 r->res_nodeid = ret_nodeid;
3912 }
3913
3914 if (is_overlap(lkb)) {
3915 log_debug(ls, "receive_lookup_reply %x unlock %x",
3916 lkb->lkb_id, lkb->lkb_flags);
3917 queue_cast_overlap(r, lkb);
3918 unhold_lkb(lkb); /* undoes create_lkb() */
3919 goto out_list;
3920 }
3921
3922 _request_lock(r, lkb);
3923
3924 out_list:
3925 if (!ret_nodeid)
3926 process_lookup_list(r);
3927 out:
3928 unlock_rsb(r);
3929 put_rsb(r);
3930 dlm_put_lkb(lkb);
3931}
3932
3933static void _receive_message(struct dlm_ls *ls, struct dlm_message *ms)
3934{
3935 if (!dlm_is_member(ls, ms->m_header.h_nodeid)) {
3936 log_debug(ls, "ignore non-member message %d from %d %x %x %d",
3937 ms->m_type, ms->m_header.h_nodeid, ms->m_lkid,
3938 ms->m_remid, ms->m_result);
3939 return;
3940 }
3941
3942 switch (ms->m_type) {
3943
3944 /* messages sent to a master node */
3945
3946 case DLM_MSG_REQUEST:
3947 receive_request(ls, ms);
3948 break;
3949
3950 case DLM_MSG_CONVERT:
3951 receive_convert(ls, ms);
3952 break;
3953
3954 case DLM_MSG_UNLOCK:
3955 receive_unlock(ls, ms);
3956 break;
3957
3958 case DLM_MSG_CANCEL:
3959 receive_cancel(ls, ms);
3960 break;
3961
3962 /* messages sent from a master node (replies to above) */
3963
3964 case DLM_MSG_REQUEST_REPLY:
3965 receive_request_reply(ls, ms);
3966 break;
3967
3968 case DLM_MSG_CONVERT_REPLY:
3969 receive_convert_reply(ls, ms);
3970 break;
3971
3972 case DLM_MSG_UNLOCK_REPLY:
3973 receive_unlock_reply(ls, ms);
3974 break;
3975
3976 case DLM_MSG_CANCEL_REPLY:
3977 receive_cancel_reply(ls, ms);
3978 break;
3979
3980 /* messages sent from a master node (only two types of async msg) */
3981
3982 case DLM_MSG_GRANT:
3983 receive_grant(ls, ms);
3984 break;
3985
3986 case DLM_MSG_BAST:
3987 receive_bast(ls, ms);
3988 break;
3989
3990 /* messages sent to a dir node */
3991
3992 case DLM_MSG_LOOKUP:
3993 receive_lookup(ls, ms);
3994 break;
3995
3996 case DLM_MSG_REMOVE:
3997 receive_remove(ls, ms);
3998 break;
3999
4000 /* messages sent from a dir node (remove has no reply) */
4001
4002 case DLM_MSG_LOOKUP_REPLY:
4003 receive_lookup_reply(ls, ms);
4004 break;
4005
4006 /* other messages */
4007
4008 case DLM_MSG_PURGE:
4009 receive_purge(ls, ms);
4010 break;
4011
4012 default:
4013 log_error(ls, "unknown message type %d", ms->m_type);
4014 }
4015}
4016
4017/* If the lockspace is in recovery mode (locking stopped), then normal
4018 messages are saved on the requestqueue for processing after recovery is
4019 done. When not in recovery mode, we wait for dlm_recoverd to drain saved
4020 messages off the requestqueue before we process new ones. This occurs right
4021 after recovery completes when we transition from saving all messages on
4022 requestqueue, to processing all the saved messages, to processing new
4023 messages as they arrive. */
4024
4025static void dlm_receive_message(struct dlm_ls *ls, struct dlm_message *ms,
4026 int nodeid)
4027{
4028 if (dlm_locking_stopped(ls)) {
4029 dlm_add_requestqueue(ls, nodeid, ms);
4030 } else {
4031 dlm_wait_requestqueue(ls);
4032 _receive_message(ls, ms);
4033 }
4034}
4035
4036/* This is called by dlm_recoverd to process messages that were saved on
4037 the requestqueue. */
4038
4039void dlm_receive_message_saved(struct dlm_ls *ls, struct dlm_message *ms)
4040{
4041 _receive_message(ls, ms);
4042}
4043
4044/* This is called by the midcomms layer when something is received for
4045 the lockspace. It could be either a MSG (normal message sent as part of
4046 standard locking activity) or an RCOM (recovery message sent as part of
4047 lockspace recovery). */
4048
4049void dlm_receive_buffer(union dlm_packet *p, int nodeid)
4050{
4051 struct dlm_header *hd = &p->header;
4052 struct dlm_ls *ls;
4053 int type = 0;
4054
4055 switch (hd->h_cmd) {
4056 case DLM_MSG:
4057 dlm_message_in(&p->message);
4058 type = p->message.m_type;
4059 break;
4060 case DLM_RCOM:
4061 dlm_rcom_in(&p->rcom);
4062 type = p->rcom.rc_type;
4063 break;
4064 default:
4065 log_print("invalid h_cmd %d from %u", hd->h_cmd, nodeid);
4066 return;
4067 }
4068
4069 if (hd->h_nodeid != nodeid) {
4070 log_print("invalid h_nodeid %d from %d lockspace %x",
4071 hd->h_nodeid, nodeid, hd->h_lockspace);
4072 return;
4073 }
4074
4075 ls = dlm_find_lockspace_global(hd->h_lockspace);
4076 if (!ls) {
4077 if (dlm_config.ci_log_debug)
4078 log_print("invalid lockspace %x from %d cmd %d type %d",
4079 hd->h_lockspace, nodeid, hd->h_cmd, type);
4080
4081 if (hd->h_cmd == DLM_RCOM && type == DLM_RCOM_STATUS)
4082 dlm_send_ls_not_ready(nodeid, &p->rcom);
4083 return;
4084 }
4085
4086 /* this rwsem allows dlm_ls_stop() to wait for all dlm_recv threads to
4087 be inactive (in this ls) before transitioning to recovery mode */
4088
4089 down_read(&ls->ls_recv_active);
4090 if (hd->h_cmd == DLM_MSG)
4091 dlm_receive_message(ls, &p->message, nodeid);
4092 else
4093 dlm_receive_rcom(ls, &p->rcom, nodeid);
4094 up_read(&ls->ls_recv_active);
4095
4096 dlm_put_lockspace(ls);
4097}
4098
4099static void recover_convert_waiter(struct dlm_ls *ls, struct dlm_lkb *lkb,
4100 struct dlm_message *ms_stub)
4101{
4102 if (middle_conversion(lkb)) {
4103 hold_lkb(lkb);
4104 memset(ms_stub, 0, sizeof(struct dlm_message));
4105 ms_stub->m_flags = DLM_IFL_STUB_MS;
4106 ms_stub->m_type = DLM_MSG_CONVERT_REPLY;
4107 ms_stub->m_result = -EINPROGRESS;
4108 ms_stub->m_header.h_nodeid = lkb->lkb_nodeid;
4109 _receive_convert_reply(lkb, ms_stub);
4110
4111 /* Same special case as in receive_rcom_lock_args() */
4112 lkb->lkb_grmode = DLM_LOCK_IV;
4113 rsb_set_flag(lkb->lkb_resource, RSB_RECOVER_CONVERT);
4114 unhold_lkb(lkb);
4115
4116 } else if (lkb->lkb_rqmode >= lkb->lkb_grmode) {
4117 lkb->lkb_flags |= DLM_IFL_RESEND;
4118 }
4119
4120 /* lkb->lkb_rqmode < lkb->lkb_grmode shouldn't happen since down
4121 conversions are async; there's no reply from the remote master */
4122}
4123
4124/* A waiting lkb needs recovery if the master node has failed, or
4125 the master node is changing (only when no directory is used) */
4126
4127static int waiter_needs_recovery(struct dlm_ls *ls, struct dlm_lkb *lkb)
4128{
4129 if (dlm_is_removed(ls, lkb->lkb_nodeid))
4130 return 1;
4131
4132 if (!dlm_no_directory(ls))
4133 return 0;
4134
4135 if (dlm_dir_nodeid(lkb->lkb_resource) != lkb->lkb_nodeid)
4136 return 1;
4137
4138 return 0;
4139}
4140
4141/* Recovery for locks that are waiting for replies from nodes that are now
4142 gone. We can just complete unlocks and cancels by faking a reply from the
4143 dead node. Requests and up-conversions we flag to be resent after
4144 recovery. Down-conversions can just be completed with a fake reply like
4145 unlocks. Conversions between PR and CW need special attention. */
4146
4147void dlm_recover_waiters_pre(struct dlm_ls *ls)
4148{
4149 struct dlm_lkb *lkb, *safe;
4150 struct dlm_message *ms_stub;
4151 int wait_type, stub_unlock_result, stub_cancel_result;
4152
4153 ms_stub = kmalloc(sizeof(struct dlm_message), GFP_KERNEL);
4154 if (!ms_stub) {
4155 log_error(ls, "dlm_recover_waiters_pre no mem");
4156 return;
4157 }
4158
4159 mutex_lock(&ls->ls_waiters_mutex);
4160
4161 list_for_each_entry_safe(lkb, safe, &ls->ls_waiters, lkb_wait_reply) {
4162
4163 /* exclude debug messages about unlocks because there can be so
4164 many and they aren't very interesting */
4165
4166 if (lkb->lkb_wait_type != DLM_MSG_UNLOCK) {
4167 log_debug(ls, "recover_waiter %x nodeid %d "
4168 "msg %d to %d", lkb->lkb_id, lkb->lkb_nodeid,
4169 lkb->lkb_wait_type, lkb->lkb_wait_nodeid);
4170 }
4171
4172 /* all outstanding lookups, regardless of destination will be
4173 resent after recovery is done */
4174
4175 if (lkb->lkb_wait_type == DLM_MSG_LOOKUP) {
4176 lkb->lkb_flags |= DLM_IFL_RESEND;
4177 continue;
4178 }
4179
4180 if (!waiter_needs_recovery(ls, lkb))
4181 continue;
4182
4183 wait_type = lkb->lkb_wait_type;
4184 stub_unlock_result = -DLM_EUNLOCK;
4185 stub_cancel_result = -DLM_ECANCEL;
4186
4187 /* Main reply may have been received leaving a zero wait_type,
4188 but a reply for the overlapping op may not have been
4189 received. In that case we need to fake the appropriate
4190 reply for the overlap op. */
4191
4192 if (!wait_type) {
4193 if (is_overlap_cancel(lkb)) {
4194 wait_type = DLM_MSG_CANCEL;
4195 if (lkb->lkb_grmode == DLM_LOCK_IV)
4196 stub_cancel_result = 0;
4197 }
4198 if (is_overlap_unlock(lkb)) {
4199 wait_type = DLM_MSG_UNLOCK;
4200 if (lkb->lkb_grmode == DLM_LOCK_IV)
4201 stub_unlock_result = -ENOENT;
4202 }
4203
4204 log_debug(ls, "rwpre overlap %x %x %d %d %d",
4205 lkb->lkb_id, lkb->lkb_flags, wait_type,
4206 stub_cancel_result, stub_unlock_result);
4207 }
4208
4209 switch (wait_type) {
4210
4211 case DLM_MSG_REQUEST:
4212 lkb->lkb_flags |= DLM_IFL_RESEND;
4213 break;
4214
4215 case DLM_MSG_CONVERT:
4216 recover_convert_waiter(ls, lkb, ms_stub);
4217 break;
4218
4219 case DLM_MSG_UNLOCK:
4220 hold_lkb(lkb);
4221 memset(ms_stub, 0, sizeof(struct dlm_message));
4222 ms_stub->m_flags = DLM_IFL_STUB_MS;
4223 ms_stub->m_type = DLM_MSG_UNLOCK_REPLY;
4224 ms_stub->m_result = stub_unlock_result;
4225 ms_stub->m_header.h_nodeid = lkb->lkb_nodeid;
4226 _receive_unlock_reply(lkb, ms_stub);
4227 dlm_put_lkb(lkb);
4228 break;
4229
4230 case DLM_MSG_CANCEL:
4231 hold_lkb(lkb);
4232 memset(ms_stub, 0, sizeof(struct dlm_message));
4233 ms_stub->m_flags = DLM_IFL_STUB_MS;
4234 ms_stub->m_type = DLM_MSG_CANCEL_REPLY;
4235 ms_stub->m_result = stub_cancel_result;
4236 ms_stub->m_header.h_nodeid = lkb->lkb_nodeid;
4237 _receive_cancel_reply(lkb, ms_stub);
4238 dlm_put_lkb(lkb);
4239 break;
4240
4241 default:
4242 log_error(ls, "invalid lkb wait_type %d %d",
4243 lkb->lkb_wait_type, wait_type);
4244 }
4245 schedule();
4246 }
4247 mutex_unlock(&ls->ls_waiters_mutex);
4248 kfree(ms_stub);
4249}
4250
4251static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls)
4252{
4253 struct dlm_lkb *lkb;
4254 int found = 0;
4255
4256 mutex_lock(&ls->ls_waiters_mutex);
4257 list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) {
4258 if (lkb->lkb_flags & DLM_IFL_RESEND) {
4259 hold_lkb(lkb);
4260 found = 1;
4261 break;
4262 }
4263 }
4264 mutex_unlock(&ls->ls_waiters_mutex);
4265
4266 if (!found)
4267 lkb = NULL;
4268 return lkb;
4269}
4270
4271/* Deal with lookups and lkb's marked RESEND from _pre. We may now be the
4272 master or dir-node for r. Processing the lkb may result in it being placed
4273 back on waiters. */
4274
4275/* We do this after normal locking has been enabled and any saved messages
4276 (in requestqueue) have been processed. We should be confident that at
4277 this point we won't get or process a reply to any of these waiting
4278 operations. But, new ops may be coming in on the rsbs/locks here from
4279 userspace or remotely. */
4280
4281/* there may have been an overlap unlock/cancel prior to recovery or after
4282 recovery. if before, the lkb may still have a pos wait_count; if after, the
4283 overlap flag would just have been set and nothing new sent. we can be
4284 confident here than any replies to either the initial op or overlap ops
4285 prior to recovery have been received. */
4286
4287int dlm_recover_waiters_post(struct dlm_ls *ls)
4288{
4289 struct dlm_lkb *lkb;
4290 struct dlm_rsb *r;
4291 int error = 0, mstype, err, oc, ou;
4292
4293 while (1) {
4294 if (dlm_locking_stopped(ls)) {
4295 log_debug(ls, "recover_waiters_post aborted");
4296 error = -EINTR;
4297 break;
4298 }
4299
4300 lkb = find_resend_waiter(ls);
4301 if (!lkb)
4302 break;
4303
4304 r = lkb->lkb_resource;
4305 hold_rsb(r);
4306 lock_rsb(r);
4307
4308 mstype = lkb->lkb_wait_type;
4309 oc = is_overlap_cancel(lkb);
4310 ou = is_overlap_unlock(lkb);
4311 err = 0;
4312
4313 log_debug(ls, "recover_waiter %x nodeid %d msg %d r_nodeid %d",
4314 lkb->lkb_id, lkb->lkb_nodeid, mstype, r->res_nodeid);
4315
4316 /* At this point we assume that we won't get a reply to any
4317 previous op or overlap op on this lock. First, do a big
4318 remove_from_waiters() for all previous ops. */
4319
4320 lkb->lkb_flags &= ~DLM_IFL_RESEND;
4321 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
4322 lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
4323 lkb->lkb_wait_type = 0;
4324 lkb->lkb_wait_count = 0;
4325 mutex_lock(&ls->ls_waiters_mutex);
4326 list_del_init(&lkb->lkb_wait_reply);
4327 mutex_unlock(&ls->ls_waiters_mutex);
4328 unhold_lkb(lkb); /* for waiters list */
4329
4330 if (oc || ou) {
4331 /* do an unlock or cancel instead of resending */
4332 switch (mstype) {
4333 case DLM_MSG_LOOKUP:
4334 case DLM_MSG_REQUEST:
4335 queue_cast(r, lkb, ou ? -DLM_EUNLOCK :
4336 -DLM_ECANCEL);
4337 unhold_lkb(lkb); /* undoes create_lkb() */
4338 break;
4339 case DLM_MSG_CONVERT:
4340 if (oc) {
4341 queue_cast(r, lkb, -DLM_ECANCEL);
4342 } else {
4343 lkb->lkb_exflags |= DLM_LKF_FORCEUNLOCK;
4344 _unlock_lock(r, lkb);
4345 }
4346 break;
4347 default:
4348 err = 1;
4349 }
4350 } else {
4351 switch (mstype) {
4352 case DLM_MSG_LOOKUP:
4353 case DLM_MSG_REQUEST:
4354 _request_lock(r, lkb);
4355 if (is_master(r))
4356 confirm_master(r, 0);
4357 break;
4358 case DLM_MSG_CONVERT:
4359 _convert_lock(r, lkb);
4360 break;
4361 default:
4362 err = 1;
4363 }
4364 }
4365
4366 if (err)
4367 log_error(ls, "recover_waiters_post %x %d %x %d %d",
4368 lkb->lkb_id, mstype, lkb->lkb_flags, oc, ou);
4369 unlock_rsb(r);
4370 put_rsb(r);
4371 dlm_put_lkb(lkb);
4372 }
4373
4374 return error;
4375}
4376
4377static void purge_queue(struct dlm_rsb *r, struct list_head *queue,
4378 int (*test)(struct dlm_ls *ls, struct dlm_lkb *lkb))
4379{
4380 struct dlm_ls *ls = r->res_ls;
4381 struct dlm_lkb *lkb, *safe;
4382
4383 list_for_each_entry_safe(lkb, safe, queue, lkb_statequeue) {
4384 if (test(ls, lkb)) {
4385 rsb_set_flag(r, RSB_LOCKS_PURGED);
4386 del_lkb(r, lkb);
4387 /* this put should free the lkb */
4388 if (!dlm_put_lkb(lkb))
4389 log_error(ls, "purged lkb not released");
4390 }
4391 }
4392}
4393
4394static int purge_dead_test(struct dlm_ls *ls, struct dlm_lkb *lkb)
4395{
4396 return (is_master_copy(lkb) && dlm_is_removed(ls, lkb->lkb_nodeid));
4397}
4398
4399static int purge_mstcpy_test(struct dlm_ls *ls, struct dlm_lkb *lkb)
4400{
4401 return is_master_copy(lkb);
4402}
4403
4404static void purge_dead_locks(struct dlm_rsb *r)
4405{
4406 purge_queue(r, &r->res_grantqueue, &purge_dead_test);
4407 purge_queue(r, &r->res_convertqueue, &purge_dead_test);
4408 purge_queue(r, &r->res_waitqueue, &purge_dead_test);
4409}
4410
4411void dlm_purge_mstcpy_locks(struct dlm_rsb *r)
4412{
4413 purge_queue(r, &r->res_grantqueue, &purge_mstcpy_test);
4414 purge_queue(r, &r->res_convertqueue, &purge_mstcpy_test);
4415 purge_queue(r, &r->res_waitqueue, &purge_mstcpy_test);
4416}
4417
4418/* Get rid of locks held by nodes that are gone. */
4419
4420int dlm_purge_locks(struct dlm_ls *ls)
4421{
4422 struct dlm_rsb *r;
4423
4424 log_debug(ls, "dlm_purge_locks");
4425
4426 down_write(&ls->ls_root_sem);
4427 list_for_each_entry(r, &ls->ls_root_list, res_root_list) {
4428 hold_rsb(r);
4429 lock_rsb(r);
4430 if (is_master(r))
4431 purge_dead_locks(r);
4432 unlock_rsb(r);
4433 unhold_rsb(r);
4434
4435 schedule();
4436 }
4437 up_write(&ls->ls_root_sem);
4438
4439 return 0;
4440}
4441
4442static struct dlm_rsb *find_purged_rsb(struct dlm_ls *ls, int bucket)
4443{
4444 struct dlm_rsb *r, *r_ret = NULL;
4445
4446 spin_lock(&ls->ls_rsbtbl[bucket].lock);
4447 list_for_each_entry(r, &ls->ls_rsbtbl[bucket].list, res_hashchain) {
4448 if (!rsb_flag(r, RSB_LOCKS_PURGED))
4449 continue;
4450 hold_rsb(r);
4451 rsb_clear_flag(r, RSB_LOCKS_PURGED);
4452 r_ret = r;
4453 break;
4454 }
4455 spin_unlock(&ls->ls_rsbtbl[bucket].lock);
4456 return r_ret;
4457}
4458
4459void dlm_grant_after_purge(struct dlm_ls *ls)
4460{
4461 struct dlm_rsb *r;
4462 int bucket = 0;
4463
4464 while (1) {
4465 r = find_purged_rsb(ls, bucket);
4466 if (!r) {
4467 if (bucket == ls->ls_rsbtbl_size - 1)
4468 break;
4469 bucket++;
4470 continue;
4471 }
4472 lock_rsb(r);
4473 if (is_master(r)) {
4474 grant_pending_locks(r);
4475 confirm_master(r, 0);
4476 }
4477 unlock_rsb(r);
4478 put_rsb(r);
4479 schedule();
4480 }
4481}
4482
4483static struct dlm_lkb *search_remid_list(struct list_head *head, int nodeid,
4484 uint32_t remid)
4485{
4486 struct dlm_lkb *lkb;
4487
4488 list_for_each_entry(lkb, head, lkb_statequeue) {
4489 if (lkb->lkb_nodeid == nodeid && lkb->lkb_remid == remid)
4490 return lkb;
4491 }
4492 return NULL;
4493}
4494
4495static struct dlm_lkb *search_remid(struct dlm_rsb *r, int nodeid,
4496 uint32_t remid)
4497{
4498 struct dlm_lkb *lkb;
4499
4500 lkb = search_remid_list(&r->res_grantqueue, nodeid, remid);
4501 if (lkb)
4502 return lkb;
4503 lkb = search_remid_list(&r->res_convertqueue, nodeid, remid);
4504 if (lkb)
4505 return lkb;
4506 lkb = search_remid_list(&r->res_waitqueue, nodeid, remid);
4507 if (lkb)
4508 return lkb;
4509 return NULL;
4510}
4511
4512/* needs at least dlm_rcom + rcom_lock */
4513static int receive_rcom_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
4514 struct dlm_rsb *r, struct dlm_rcom *rc)
4515{
4516 struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
4517
4518 lkb->lkb_nodeid = rc->rc_header.h_nodeid;
4519 lkb->lkb_ownpid = le32_to_cpu(rl->rl_ownpid);
4520 lkb->lkb_remid = le32_to_cpu(rl->rl_lkid);
4521 lkb->lkb_exflags = le32_to_cpu(rl->rl_exflags);
4522 lkb->lkb_flags = le32_to_cpu(rl->rl_flags) & 0x0000FFFF;
4523 lkb->lkb_flags |= DLM_IFL_MSTCPY;
4524 lkb->lkb_lvbseq = le32_to_cpu(rl->rl_lvbseq);
4525 lkb->lkb_rqmode = rl->rl_rqmode;
4526 lkb->lkb_grmode = rl->rl_grmode;
4527 /* don't set lkb_status because add_lkb wants to itself */
4528
4529 lkb->lkb_bastfn = (rl->rl_asts & DLM_CB_BAST) ? &fake_bastfn : NULL;
4530 lkb->lkb_astfn = (rl->rl_asts & DLM_CB_CAST) ? &fake_astfn : NULL;
4531
4532 if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
4533 int lvblen = rc->rc_header.h_length - sizeof(struct dlm_rcom) -
4534 sizeof(struct rcom_lock);
4535 if (lvblen > ls->ls_lvblen)
4536 return -EINVAL;
4537 lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
4538 if (!lkb->lkb_lvbptr)
4539 return -ENOMEM;
4540 memcpy(lkb->lkb_lvbptr, rl->rl_lvb, lvblen);
4541 }
4542
4543 /* Conversions between PR and CW (middle modes) need special handling.
4544 The real granted mode of these converting locks cannot be determined
4545 until all locks have been rebuilt on the rsb (recover_conversion) */
4546
4547 if (rl->rl_wait_type == cpu_to_le16(DLM_MSG_CONVERT) &&
4548 middle_conversion(lkb)) {
4549 rl->rl_status = DLM_LKSTS_CONVERT;
4550 lkb->lkb_grmode = DLM_LOCK_IV;
4551 rsb_set_flag(r, RSB_RECOVER_CONVERT);
4552 }
4553
4554 return 0;
4555}
4556
4557/* This lkb may have been recovered in a previous aborted recovery so we need
4558 to check if the rsb already has an lkb with the given remote nodeid/lkid.
4559 If so we just send back a standard reply. If not, we create a new lkb with
4560 the given values and send back our lkid. We send back our lkid by sending
4561 back the rcom_lock struct we got but with the remid field filled in. */
4562
4563/* needs at least dlm_rcom + rcom_lock */
4564int dlm_recover_master_copy(struct dlm_ls *ls, struct dlm_rcom *rc)
4565{
4566 struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
4567 struct dlm_rsb *r;
4568 struct dlm_lkb *lkb;
4569 int error;
4570
4571 if (rl->rl_parent_lkid) {
4572 error = -EOPNOTSUPP;
4573 goto out;
4574 }
4575
4576 error = find_rsb(ls, rl->rl_name, le16_to_cpu(rl->rl_namelen),
4577 R_MASTER, &r);
4578 if (error)
4579 goto out;
4580
4581 lock_rsb(r);
4582
4583 lkb = search_remid(r, rc->rc_header.h_nodeid, le32_to_cpu(rl->rl_lkid));
4584 if (lkb) {
4585 error = -EEXIST;
4586 goto out_remid;
4587 }
4588
4589 error = create_lkb(ls, &lkb);
4590 if (error)
4591 goto out_unlock;
4592
4593 error = receive_rcom_lock_args(ls, lkb, r, rc);
4594 if (error) {
4595 __put_lkb(ls, lkb);
4596 goto out_unlock;
4597 }
4598
4599 attach_lkb(r, lkb);
4600 add_lkb(r, lkb, rl->rl_status);
4601 error = 0;
4602
4603 out_remid:
4604 /* this is the new value returned to the lock holder for
4605 saving in its process-copy lkb */
4606 rl->rl_remid = cpu_to_le32(lkb->lkb_id);
4607
4608 out_unlock:
4609 unlock_rsb(r);
4610 put_rsb(r);
4611 out:
4612 if (error)
4613 log_debug(ls, "recover_master_copy %d %x", error,
4614 le32_to_cpu(rl->rl_lkid));
4615 rl->rl_result = cpu_to_le32(error);
4616 return error;
4617}
4618
4619/* needs at least dlm_rcom + rcom_lock */
4620int dlm_recover_process_copy(struct dlm_ls *ls, struct dlm_rcom *rc)
4621{
4622 struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
4623 struct dlm_rsb *r;
4624 struct dlm_lkb *lkb;
4625 int error;
4626
4627 error = find_lkb(ls, le32_to_cpu(rl->rl_lkid), &lkb);
4628 if (error) {
4629 log_error(ls, "recover_process_copy no lkid %x",
4630 le32_to_cpu(rl->rl_lkid));
4631 return error;
4632 }
4633
4634 DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb););
4635
4636 error = le32_to_cpu(rl->rl_result);
4637
4638 r = lkb->lkb_resource;
4639 hold_rsb(r);
4640 lock_rsb(r);
4641
4642 switch (error) {
4643 case -EBADR:
4644 /* There's a chance the new master received our lock before
4645 dlm_recover_master_reply(), this wouldn't happen if we did
4646 a barrier between recover_masters and recover_locks. */
4647 log_debug(ls, "master copy not ready %x r %lx %s", lkb->lkb_id,
4648 (unsigned long)r, r->res_name);
4649 dlm_send_rcom_lock(r, lkb);
4650 goto out;
4651 case -EEXIST:
4652 log_debug(ls, "master copy exists %x", lkb->lkb_id);
4653 /* fall through */
4654 case 0:
4655 lkb->lkb_remid = le32_to_cpu(rl->rl_remid);
4656 break;
4657 default:
4658 log_error(ls, "dlm_recover_process_copy unknown error %d %x",
4659 error, lkb->lkb_id);
4660 }
4661
4662 /* an ack for dlm_recover_locks() which waits for replies from
4663 all the locks it sends to new masters */
4664 dlm_recovered_lock(r);
4665 out:
4666 unlock_rsb(r);
4667 put_rsb(r);
4668 dlm_put_lkb(lkb);
4669
4670 return 0;
4671}
4672
4673int dlm_user_request(struct dlm_ls *ls, struct dlm_user_args *ua,
4674 int mode, uint32_t flags, void *name, unsigned int namelen,
4675 unsigned long timeout_cs)
4676{
4677 struct dlm_lkb *lkb;
4678 struct dlm_args args;
4679 int error;
4680
4681 dlm_lock_recovery(ls);
4682
4683 error = create_lkb(ls, &lkb);
4684 if (error) {
4685 kfree(ua);
4686 goto out;
4687 }
4688
4689 if (flags & DLM_LKF_VALBLK) {
4690 ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
4691 if (!ua->lksb.sb_lvbptr) {
4692 kfree(ua);
4693 __put_lkb(ls, lkb);
4694 error = -ENOMEM;
4695 goto out;
4696 }
4697 }
4698
4699 /* After ua is attached to lkb it will be freed by dlm_free_lkb().
4700 When DLM_IFL_USER is set, the dlm knows that this is a userspace
4701 lock and that lkb_astparam is the dlm_user_args structure. */
4702
4703 error = set_lock_args(mode, &ua->lksb, flags, namelen, timeout_cs,
4704 fake_astfn, ua, fake_bastfn, &args);
4705 lkb->lkb_flags |= DLM_IFL_USER;
4706
4707 if (error) {
4708 __put_lkb(ls, lkb);
4709 goto out;
4710 }
4711
4712 error = request_lock(ls, lkb, name, namelen, &args);
4713
4714 switch (error) {
4715 case 0:
4716 break;
4717 case -EINPROGRESS:
4718 error = 0;
4719 break;
4720 case -EAGAIN:
4721 error = 0;
4722 /* fall through */
4723 default:
4724 __put_lkb(ls, lkb);
4725 goto out;
4726 }
4727
4728 /* add this new lkb to the per-process list of locks */
4729 spin_lock(&ua->proc->locks_spin);
4730 hold_lkb(lkb);
4731 list_add_tail(&lkb->lkb_ownqueue, &ua->proc->locks);
4732 spin_unlock(&ua->proc->locks_spin);
4733 out:
4734 dlm_unlock_recovery(ls);
4735 return error;
4736}
4737
4738int dlm_user_convert(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
4739 int mode, uint32_t flags, uint32_t lkid, char *lvb_in,
4740 unsigned long timeout_cs)
4741{
4742 struct dlm_lkb *lkb;
4743 struct dlm_args args;
4744 struct dlm_user_args *ua;
4745 int error;
4746
4747 dlm_lock_recovery(ls);
4748
4749 error = find_lkb(ls, lkid, &lkb);
4750 if (error)
4751 goto out;
4752
4753 /* user can change the params on its lock when it converts it, or
4754 add an lvb that didn't exist before */
4755
4756 ua = lkb->lkb_ua;
4757
4758 if (flags & DLM_LKF_VALBLK && !ua->lksb.sb_lvbptr) {
4759 ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
4760 if (!ua->lksb.sb_lvbptr) {
4761 error = -ENOMEM;
4762 goto out_put;
4763 }
4764 }
4765 if (lvb_in && ua->lksb.sb_lvbptr)
4766 memcpy(ua->lksb.sb_lvbptr, lvb_in, DLM_USER_LVB_LEN);
4767
4768 ua->xid = ua_tmp->xid;
4769 ua->castparam = ua_tmp->castparam;
4770 ua->castaddr = ua_tmp->castaddr;
4771 ua->bastparam = ua_tmp->bastparam;
4772 ua->bastaddr = ua_tmp->bastaddr;
4773 ua->user_lksb = ua_tmp->user_lksb;
4774
4775 error = set_lock_args(mode, &ua->lksb, flags, 0, timeout_cs,
4776 fake_astfn, ua, fake_bastfn, &args);
4777 if (error)
4778 goto out_put;
4779
4780 error = convert_lock(ls, lkb, &args);
4781
4782 if (error == -EINPROGRESS || error == -EAGAIN || error == -EDEADLK)
4783 error = 0;
4784 out_put:
4785 dlm_put_lkb(lkb);
4786 out:
4787 dlm_unlock_recovery(ls);
4788 kfree(ua_tmp);
4789 return error;
4790}
4791
4792int dlm_user_unlock(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
4793 uint32_t flags, uint32_t lkid, char *lvb_in)
4794{
4795 struct dlm_lkb *lkb;
4796 struct dlm_args args;
4797 struct dlm_user_args *ua;
4798 int error;
4799
4800 dlm_lock_recovery(ls);
4801
4802 error = find_lkb(ls, lkid, &lkb);
4803 if (error)
4804 goto out;
4805
4806 ua = lkb->lkb_ua;
4807
4808 if (lvb_in && ua->lksb.sb_lvbptr)
4809 memcpy(ua->lksb.sb_lvbptr, lvb_in, DLM_USER_LVB_LEN);
4810 if (ua_tmp->castparam)
4811 ua->castparam = ua_tmp->castparam;
4812 ua->user_lksb = ua_tmp->user_lksb;
4813
4814 error = set_unlock_args(flags, ua, &args);
4815 if (error)
4816 goto out_put;
4817
4818 error = unlock_lock(ls, lkb, &args);
4819
4820 if (error == -DLM_EUNLOCK)
4821 error = 0;
4822 /* from validate_unlock_args() */
4823 if (error == -EBUSY && (flags & DLM_LKF_FORCEUNLOCK))
4824 error = 0;
4825 if (error)
4826 goto out_put;
4827
4828 spin_lock(&ua->proc->locks_spin);
4829 /* dlm_user_add_cb() may have already taken lkb off the proc list */
4830 if (!list_empty(&lkb->lkb_ownqueue))
4831 list_move(&lkb->lkb_ownqueue, &ua->proc->unlocking);
4832 spin_unlock(&ua->proc->locks_spin);
4833 out_put:
4834 dlm_put_lkb(lkb);
4835 out:
4836 dlm_unlock_recovery(ls);
4837 kfree(ua_tmp);
4838 return error;
4839}
4840
4841int dlm_user_cancel(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
4842 uint32_t flags, uint32_t lkid)
4843{
4844 struct dlm_lkb *lkb;
4845 struct dlm_args args;
4846 struct dlm_user_args *ua;
4847 int error;
4848
4849 dlm_lock_recovery(ls);
4850
4851 error = find_lkb(ls, lkid, &lkb);
4852 if (error)
4853 goto out;
4854
4855 ua = lkb->lkb_ua;
4856 if (ua_tmp->castparam)
4857 ua->castparam = ua_tmp->castparam;
4858 ua->user_lksb = ua_tmp->user_lksb;
4859
4860 error = set_unlock_args(flags, ua, &args);
4861 if (error)
4862 goto out_put;
4863
4864 error = cancel_lock(ls, lkb, &args);
4865
4866 if (error == -DLM_ECANCEL)
4867 error = 0;
4868 /* from validate_unlock_args() */
4869 if (error == -EBUSY)
4870 error = 0;
4871 out_put:
4872 dlm_put_lkb(lkb);
4873 out:
4874 dlm_unlock_recovery(ls);
4875 kfree(ua_tmp);
4876 return error;
4877}
4878
4879int dlm_user_deadlock(struct dlm_ls *ls, uint32_t flags, uint32_t lkid)
4880{
4881 struct dlm_lkb *lkb;
4882 struct dlm_args args;
4883 struct dlm_user_args *ua;
4884 struct dlm_rsb *r;
4885 int error;
4886
4887 dlm_lock_recovery(ls);
4888
4889 error = find_lkb(ls, lkid, &lkb);
4890 if (error)
4891 goto out;
4892
4893 ua = lkb->lkb_ua;
4894
4895 error = set_unlock_args(flags, ua, &args);
4896 if (error)
4897 goto out_put;
4898
4899 /* same as cancel_lock(), but set DEADLOCK_CANCEL after lock_rsb */
4900
4901 r = lkb->lkb_resource;
4902 hold_rsb(r);
4903 lock_rsb(r);
4904
4905 error = validate_unlock_args(lkb, &args);
4906 if (error)
4907 goto out_r;
4908 lkb->lkb_flags |= DLM_IFL_DEADLOCK_CANCEL;
4909
4910 error = _cancel_lock(r, lkb);
4911 out_r:
4912 unlock_rsb(r);
4913 put_rsb(r);
4914
4915 if (error == -DLM_ECANCEL)
4916 error = 0;
4917 /* from validate_unlock_args() */
4918 if (error == -EBUSY)
4919 error = 0;
4920 out_put:
4921 dlm_put_lkb(lkb);
4922 out:
4923 dlm_unlock_recovery(ls);
4924 return error;
4925}
4926
4927/* lkb's that are removed from the waiters list by revert are just left on the
4928 orphans list with the granted orphan locks, to be freed by purge */
4929
4930static int orphan_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb)
4931{
4932 struct dlm_args args;
4933 int error;
4934
4935 hold_lkb(lkb);
4936 mutex_lock(&ls->ls_orphans_mutex);
4937 list_add_tail(&lkb->lkb_ownqueue, &ls->ls_orphans);
4938 mutex_unlock(&ls->ls_orphans_mutex);
4939
4940 set_unlock_args(0, lkb->lkb_ua, &args);
4941
4942 error = cancel_lock(ls, lkb, &args);
4943 if (error == -DLM_ECANCEL)
4944 error = 0;
4945 return error;
4946}
4947
4948/* The force flag allows the unlock to go ahead even if the lkb isn't granted.
4949 Regardless of what rsb queue the lock is on, it's removed and freed. */
4950
4951static int unlock_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb)
4952{
4953 struct dlm_args args;
4954 int error;
4955
4956 set_unlock_args(DLM_LKF_FORCEUNLOCK, lkb->lkb_ua, &args);
4957
4958 error = unlock_lock(ls, lkb, &args);
4959 if (error == -DLM_EUNLOCK)
4960 error = 0;
4961 return error;
4962}
4963
4964/* We have to release clear_proc_locks mutex before calling unlock_proc_lock()
4965 (which does lock_rsb) due to deadlock with receiving a message that does
4966 lock_rsb followed by dlm_user_add_cb() */
4967
4968static struct dlm_lkb *del_proc_lock(struct dlm_ls *ls,
4969 struct dlm_user_proc *proc)
4970{
4971 struct dlm_lkb *lkb = NULL;
4972
4973 mutex_lock(&ls->ls_clear_proc_locks);
4974 if (list_empty(&proc->locks))
4975 goto out;
4976
4977 lkb = list_entry(proc->locks.next, struct dlm_lkb, lkb_ownqueue);
4978 list_del_init(&lkb->lkb_ownqueue);
4979
4980 if (lkb->lkb_exflags & DLM_LKF_PERSISTENT)
4981 lkb->lkb_flags |= DLM_IFL_ORPHAN;
4982 else
4983 lkb->lkb_flags |= DLM_IFL_DEAD;
4984 out:
4985 mutex_unlock(&ls->ls_clear_proc_locks);
4986 return lkb;
4987}
4988
4989/* The ls_clear_proc_locks mutex protects against dlm_user_add_cb() which
4990 1) references lkb->ua which we free here and 2) adds lkbs to proc->asts,
4991 which we clear here. */
4992
4993/* proc CLOSING flag is set so no more device_reads should look at proc->asts
4994 list, and no more device_writes should add lkb's to proc->locks list; so we
4995 shouldn't need to take asts_spin or locks_spin here. this assumes that
4996 device reads/writes/closes are serialized -- FIXME: we may need to serialize
4997 them ourself. */
4998
4999void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc)
5000{
5001 struct dlm_lkb *lkb, *safe;
5002
5003 dlm_lock_recovery(ls);
5004
5005 while (1) {
5006 lkb = del_proc_lock(ls, proc);
5007 if (!lkb)
5008 break;
5009 del_timeout(lkb);
5010 if (lkb->lkb_exflags & DLM_LKF_PERSISTENT)
5011 orphan_proc_lock(ls, lkb);
5012 else
5013 unlock_proc_lock(ls, lkb);
5014
5015 /* this removes the reference for the proc->locks list
5016 added by dlm_user_request, it may result in the lkb
5017 being freed */
5018
5019 dlm_put_lkb(lkb);
5020 }
5021
5022 mutex_lock(&ls->ls_clear_proc_locks);
5023
5024 /* in-progress unlocks */
5025 list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) {
5026 list_del_init(&lkb->lkb_ownqueue);
5027 lkb->lkb_flags |= DLM_IFL_DEAD;
5028 dlm_put_lkb(lkb);
5029 }
5030
5031 list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_cb_list) {
5032 memset(&lkb->lkb_callbacks, 0,
5033 sizeof(struct dlm_callback) * DLM_CALLBACKS_SIZE);
5034 list_del_init(&lkb->lkb_cb_list);
5035 dlm_put_lkb(lkb);
5036 }
5037
5038 mutex_unlock(&ls->ls_clear_proc_locks);
5039 dlm_unlock_recovery(ls);
5040}
5041
5042static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc)
5043{
5044 struct dlm_lkb *lkb, *safe;
5045
5046 while (1) {
5047 lkb = NULL;
5048 spin_lock(&proc->locks_spin);
5049 if (!list_empty(&proc->locks)) {
5050 lkb = list_entry(proc->locks.next, struct dlm_lkb,
5051 lkb_ownqueue);
5052 list_del_init(&lkb->lkb_ownqueue);
5053 }
5054 spin_unlock(&proc->locks_spin);
5055
5056 if (!lkb)
5057 break;
5058
5059 lkb->lkb_flags |= DLM_IFL_DEAD;
5060 unlock_proc_lock(ls, lkb);
5061 dlm_put_lkb(lkb); /* ref from proc->locks list */
5062 }
5063
5064 spin_lock(&proc->locks_spin);
5065 list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) {
5066 list_del_init(&lkb->lkb_ownqueue);
5067 lkb->lkb_flags |= DLM_IFL_DEAD;
5068 dlm_put_lkb(lkb);
5069 }
5070 spin_unlock(&proc->locks_spin);
5071
5072 spin_lock(&proc->asts_spin);
5073 list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_cb_list) {
5074 memset(&lkb->lkb_callbacks, 0,
5075 sizeof(struct dlm_callback) * DLM_CALLBACKS_SIZE);
5076 list_del_init(&lkb->lkb_cb_list);
5077 dlm_put_lkb(lkb);
5078 }
5079 spin_unlock(&proc->asts_spin);
5080}
5081
5082/* pid of 0 means purge all orphans */
5083
5084static void do_purge(struct dlm_ls *ls, int nodeid, int pid)
5085{
5086 struct dlm_lkb *lkb, *safe;
5087
5088 mutex_lock(&ls->ls_orphans_mutex);
5089 list_for_each_entry_safe(lkb, safe, &ls->ls_orphans, lkb_ownqueue) {
5090 if (pid && lkb->lkb_ownpid != pid)
5091 continue;
5092 unlock_proc_lock(ls, lkb);
5093 list_del_init(&lkb->lkb_ownqueue);
5094 dlm_put_lkb(lkb);
5095 }
5096 mutex_unlock(&ls->ls_orphans_mutex);
5097}
5098
5099static int send_purge(struct dlm_ls *ls, int nodeid, int pid)
5100{
5101 struct dlm_message *ms;
5102 struct dlm_mhandle *mh;
5103 int error;
5104
5105 error = _create_message(ls, sizeof(struct dlm_message), nodeid,
5106 DLM_MSG_PURGE, &ms, &mh);
5107 if (error)
5108 return error;
5109 ms->m_nodeid = nodeid;
5110 ms->m_pid = pid;
5111
5112 return send_message(mh, ms);
5113}
5114
5115int dlm_user_purge(struct dlm_ls *ls, struct dlm_user_proc *proc,
5116 int nodeid, int pid)
5117{
5118 int error = 0;
5119
5120 if (nodeid != dlm_our_nodeid()) {
5121 error = send_purge(ls, nodeid, pid);
5122 } else {
5123 dlm_lock_recovery(ls);
5124 if (pid == current->pid)
5125 purge_proc_locks(ls, proc);
5126 else
5127 do_purge(ls, nodeid, pid);
5128 dlm_unlock_recovery(ls);
5129 }
5130 return error;
5131}
5132
1// SPDX-License-Identifier: GPL-2.0-only
2/******************************************************************************
3*******************************************************************************
4**
5** Copyright (C) 2005-2010 Red Hat, Inc. All rights reserved.
6**
7**
8*******************************************************************************
9******************************************************************************/
10
11/* Central locking logic has four stages:
12
13 dlm_lock()
14 dlm_unlock()
15
16 request_lock(ls, lkb)
17 convert_lock(ls, lkb)
18 unlock_lock(ls, lkb)
19 cancel_lock(ls, lkb)
20
21 _request_lock(r, lkb)
22 _convert_lock(r, lkb)
23 _unlock_lock(r, lkb)
24 _cancel_lock(r, lkb)
25
26 do_request(r, lkb)
27 do_convert(r, lkb)
28 do_unlock(r, lkb)
29 do_cancel(r, lkb)
30
31 Stage 1 (lock, unlock) is mainly about checking input args and
32 splitting into one of the four main operations:
33
34 dlm_lock = request_lock
35 dlm_lock+CONVERT = convert_lock
36 dlm_unlock = unlock_lock
37 dlm_unlock+CANCEL = cancel_lock
38
39 Stage 2, xxxx_lock(), just finds and locks the relevant rsb which is
40 provided to the next stage.
41
42 Stage 3, _xxxx_lock(), determines if the operation is local or remote.
43 When remote, it calls send_xxxx(), when local it calls do_xxxx().
44
45 Stage 4, do_xxxx(), is the guts of the operation. It manipulates the
46 given rsb and lkb and queues callbacks.
47
48 For remote operations, send_xxxx() results in the corresponding do_xxxx()
49 function being executed on the remote node. The connecting send/receive
50 calls on local (L) and remote (R) nodes:
51
52 L: send_xxxx() -> R: receive_xxxx()
53 R: do_xxxx()
54 L: receive_xxxx_reply() <- R: send_xxxx_reply()
55*/
56#include <trace/events/dlm.h>
57
58#include <linux/types.h>
59#include <linux/rbtree.h>
60#include <linux/slab.h>
61#include "dlm_internal.h"
62#include <linux/dlm_device.h>
63#include "memory.h"
64#include "midcomms.h"
65#include "requestqueue.h"
66#include "util.h"
67#include "dir.h"
68#include "member.h"
69#include "lockspace.h"
70#include "ast.h"
71#include "lock.h"
72#include "rcom.h"
73#include "recover.h"
74#include "lvb_table.h"
75#include "user.h"
76#include "config.h"
77
78static int send_request(struct dlm_rsb *r, struct dlm_lkb *lkb);
79static int send_convert(struct dlm_rsb *r, struct dlm_lkb *lkb);
80static int send_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb);
81static int send_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb);
82static int send_grant(struct dlm_rsb *r, struct dlm_lkb *lkb);
83static int send_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int mode);
84static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb);
85static int send_remove(struct dlm_rsb *r);
86static int _request_lock(struct dlm_rsb *r, struct dlm_lkb *lkb);
87static int _cancel_lock(struct dlm_rsb *r, struct dlm_lkb *lkb);
88static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
89 const struct dlm_message *ms, bool local);
90static int receive_extralen(const struct dlm_message *ms);
91static void do_purge(struct dlm_ls *ls, int nodeid, int pid);
92static void toss_rsb(struct kref *kref);
93
94/*
95 * Lock compatibilty matrix - thanks Steve
96 * UN = Unlocked state. Not really a state, used as a flag
97 * PD = Padding. Used to make the matrix a nice power of two in size
98 * Other states are the same as the VMS DLM.
99 * Usage: matrix[grmode+1][rqmode+1] (although m[rq+1][gr+1] is the same)
100 */
101
102static const int __dlm_compat_matrix[8][8] = {
103 /* UN NL CR CW PR PW EX PD */
104 {1, 1, 1, 1, 1, 1, 1, 0}, /* UN */
105 {1, 1, 1, 1, 1, 1, 1, 0}, /* NL */
106 {1, 1, 1, 1, 1, 1, 0, 0}, /* CR */
107 {1, 1, 1, 1, 0, 0, 0, 0}, /* CW */
108 {1, 1, 1, 0, 1, 0, 0, 0}, /* PR */
109 {1, 1, 1, 0, 0, 0, 0, 0}, /* PW */
110 {1, 1, 0, 0, 0, 0, 0, 0}, /* EX */
111 {0, 0, 0, 0, 0, 0, 0, 0} /* PD */
112};
113
114/*
115 * This defines the direction of transfer of LVB data.
116 * Granted mode is the row; requested mode is the column.
117 * Usage: matrix[grmode+1][rqmode+1]
118 * 1 = LVB is returned to the caller
119 * 0 = LVB is written to the resource
120 * -1 = nothing happens to the LVB
121 */
122
123const int dlm_lvb_operations[8][8] = {
124 /* UN NL CR CW PR PW EX PD*/
125 { -1, 1, 1, 1, 1, 1, 1, -1 }, /* UN */
126 { -1, 1, 1, 1, 1, 1, 1, 0 }, /* NL */
127 { -1, -1, 1, 1, 1, 1, 1, 0 }, /* CR */
128 { -1, -1, -1, 1, 1, 1, 1, 0 }, /* CW */
129 { -1, -1, -1, -1, 1, 1, 1, 0 }, /* PR */
130 { -1, 0, 0, 0, 0, 0, 1, 0 }, /* PW */
131 { -1, 0, 0, 0, 0, 0, 0, 0 }, /* EX */
132 { -1, 0, 0, 0, 0, 0, 0, 0 } /* PD */
133};
134
135#define modes_compat(gr, rq) \
136 __dlm_compat_matrix[(gr)->lkb_grmode + 1][(rq)->lkb_rqmode + 1]
137
138int dlm_modes_compat(int mode1, int mode2)
139{
140 return __dlm_compat_matrix[mode1 + 1][mode2 + 1];
141}
142
143/*
144 * Compatibility matrix for conversions with QUECVT set.
145 * Granted mode is the row; requested mode is the column.
146 * Usage: matrix[grmode+1][rqmode+1]
147 */
148
149static const int __quecvt_compat_matrix[8][8] = {
150 /* UN NL CR CW PR PW EX PD */
151 {0, 0, 0, 0, 0, 0, 0, 0}, /* UN */
152 {0, 0, 1, 1, 1, 1, 1, 0}, /* NL */
153 {0, 0, 0, 1, 1, 1, 1, 0}, /* CR */
154 {0, 0, 0, 0, 1, 1, 1, 0}, /* CW */
155 {0, 0, 0, 1, 0, 1, 1, 0}, /* PR */
156 {0, 0, 0, 0, 0, 0, 1, 0}, /* PW */
157 {0, 0, 0, 0, 0, 0, 0, 0}, /* EX */
158 {0, 0, 0, 0, 0, 0, 0, 0} /* PD */
159};
160
161void dlm_print_lkb(struct dlm_lkb *lkb)
162{
163 printk(KERN_ERR "lkb: nodeid %d id %x remid %x exflags %x flags %x "
164 "sts %d rq %d gr %d wait_type %d wait_nodeid %d seq %llu\n",
165 lkb->lkb_nodeid, lkb->lkb_id, lkb->lkb_remid, lkb->lkb_exflags,
166 dlm_iflags_val(lkb), lkb->lkb_status, lkb->lkb_rqmode,
167 lkb->lkb_grmode, lkb->lkb_wait_type, lkb->lkb_wait_nodeid,
168 (unsigned long long)lkb->lkb_recover_seq);
169}
170
171static void dlm_print_rsb(struct dlm_rsb *r)
172{
173 printk(KERN_ERR "rsb: nodeid %d master %d dir %d flags %lx first %x "
174 "rlc %d name %s\n",
175 r->res_nodeid, r->res_master_nodeid, r->res_dir_nodeid,
176 r->res_flags, r->res_first_lkid, r->res_recover_locks_count,
177 r->res_name);
178}
179
180void dlm_dump_rsb(struct dlm_rsb *r)
181{
182 struct dlm_lkb *lkb;
183
184 dlm_print_rsb(r);
185
186 printk(KERN_ERR "rsb: root_list empty %d recover_list empty %d\n",
187 list_empty(&r->res_root_list), list_empty(&r->res_recover_list));
188 printk(KERN_ERR "rsb lookup list\n");
189 list_for_each_entry(lkb, &r->res_lookup, lkb_rsb_lookup)
190 dlm_print_lkb(lkb);
191 printk(KERN_ERR "rsb grant queue:\n");
192 list_for_each_entry(lkb, &r->res_grantqueue, lkb_statequeue)
193 dlm_print_lkb(lkb);
194 printk(KERN_ERR "rsb convert queue:\n");
195 list_for_each_entry(lkb, &r->res_convertqueue, lkb_statequeue)
196 dlm_print_lkb(lkb);
197 printk(KERN_ERR "rsb wait queue:\n");
198 list_for_each_entry(lkb, &r->res_waitqueue, lkb_statequeue)
199 dlm_print_lkb(lkb);
200}
201
202/* Threads cannot use the lockspace while it's being recovered */
203
204static inline void dlm_lock_recovery(struct dlm_ls *ls)
205{
206 down_read(&ls->ls_in_recovery);
207}
208
209void dlm_unlock_recovery(struct dlm_ls *ls)
210{
211 up_read(&ls->ls_in_recovery);
212}
213
214int dlm_lock_recovery_try(struct dlm_ls *ls)
215{
216 return down_read_trylock(&ls->ls_in_recovery);
217}
218
219static inline int can_be_queued(struct dlm_lkb *lkb)
220{
221 return !(lkb->lkb_exflags & DLM_LKF_NOQUEUE);
222}
223
224static inline int force_blocking_asts(struct dlm_lkb *lkb)
225{
226 return (lkb->lkb_exflags & DLM_LKF_NOQUEUEBAST);
227}
228
229static inline int is_demoted(struct dlm_lkb *lkb)
230{
231 return test_bit(DLM_SBF_DEMOTED_BIT, &lkb->lkb_sbflags);
232}
233
234static inline int is_altmode(struct dlm_lkb *lkb)
235{
236 return test_bit(DLM_SBF_ALTMODE_BIT, &lkb->lkb_sbflags);
237}
238
239static inline int is_granted(struct dlm_lkb *lkb)
240{
241 return (lkb->lkb_status == DLM_LKSTS_GRANTED);
242}
243
244static inline int is_remote(struct dlm_rsb *r)
245{
246 DLM_ASSERT(r->res_nodeid >= 0, dlm_print_rsb(r););
247 return !!r->res_nodeid;
248}
249
250static inline int is_process_copy(struct dlm_lkb *lkb)
251{
252 return lkb->lkb_nodeid &&
253 !test_bit(DLM_IFL_MSTCPY_BIT, &lkb->lkb_iflags);
254}
255
256static inline int is_master_copy(struct dlm_lkb *lkb)
257{
258 return test_bit(DLM_IFL_MSTCPY_BIT, &lkb->lkb_iflags);
259}
260
261static inline int middle_conversion(struct dlm_lkb *lkb)
262{
263 if ((lkb->lkb_grmode==DLM_LOCK_PR && lkb->lkb_rqmode==DLM_LOCK_CW) ||
264 (lkb->lkb_rqmode==DLM_LOCK_PR && lkb->lkb_grmode==DLM_LOCK_CW))
265 return 1;
266 return 0;
267}
268
269static inline int down_conversion(struct dlm_lkb *lkb)
270{
271 return (!middle_conversion(lkb) && lkb->lkb_rqmode < lkb->lkb_grmode);
272}
273
274static inline int is_overlap_unlock(struct dlm_lkb *lkb)
275{
276 return test_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags);
277}
278
279static inline int is_overlap_cancel(struct dlm_lkb *lkb)
280{
281 return test_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags);
282}
283
284static inline int is_overlap(struct dlm_lkb *lkb)
285{
286 return test_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags) ||
287 test_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags);
288}
289
290static void queue_cast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
291{
292 if (is_master_copy(lkb))
293 return;
294
295 DLM_ASSERT(lkb->lkb_lksb, dlm_print_lkb(lkb););
296
297 if (rv == -DLM_ECANCEL &&
298 test_and_clear_bit(DLM_IFL_DEADLOCK_CANCEL_BIT, &lkb->lkb_iflags))
299 rv = -EDEADLK;
300
301 dlm_add_cb(lkb, DLM_CB_CAST, lkb->lkb_grmode, rv, dlm_sbflags_val(lkb));
302}
303
304static inline void queue_cast_overlap(struct dlm_rsb *r, struct dlm_lkb *lkb)
305{
306 queue_cast(r, lkb,
307 is_overlap_unlock(lkb) ? -DLM_EUNLOCK : -DLM_ECANCEL);
308}
309
310static void queue_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rqmode)
311{
312 if (is_master_copy(lkb)) {
313 send_bast(r, lkb, rqmode);
314 } else {
315 dlm_add_cb(lkb, DLM_CB_BAST, rqmode, 0, 0);
316 }
317}
318
319/*
320 * Basic operations on rsb's and lkb's
321 */
322
323/* This is only called to add a reference when the code already holds
324 a valid reference to the rsb, so there's no need for locking. */
325
326static inline void hold_rsb(struct dlm_rsb *r)
327{
328 kref_get(&r->res_ref);
329}
330
331void dlm_hold_rsb(struct dlm_rsb *r)
332{
333 hold_rsb(r);
334}
335
336/* When all references to the rsb are gone it's transferred to
337 the tossed list for later disposal. */
338
339static void put_rsb(struct dlm_rsb *r)
340{
341 struct dlm_ls *ls = r->res_ls;
342 uint32_t bucket = r->res_bucket;
343 int rv;
344
345 rv = kref_put_lock(&r->res_ref, toss_rsb,
346 &ls->ls_rsbtbl[bucket].lock);
347 if (rv)
348 spin_unlock(&ls->ls_rsbtbl[bucket].lock);
349}
350
351void dlm_put_rsb(struct dlm_rsb *r)
352{
353 put_rsb(r);
354}
355
356static int pre_rsb_struct(struct dlm_ls *ls)
357{
358 struct dlm_rsb *r1, *r2;
359 int count = 0;
360
361 spin_lock(&ls->ls_new_rsb_spin);
362 if (ls->ls_new_rsb_count > dlm_config.ci_new_rsb_count / 2) {
363 spin_unlock(&ls->ls_new_rsb_spin);
364 return 0;
365 }
366 spin_unlock(&ls->ls_new_rsb_spin);
367
368 r1 = dlm_allocate_rsb(ls);
369 r2 = dlm_allocate_rsb(ls);
370
371 spin_lock(&ls->ls_new_rsb_spin);
372 if (r1) {
373 list_add(&r1->res_hashchain, &ls->ls_new_rsb);
374 ls->ls_new_rsb_count++;
375 }
376 if (r2) {
377 list_add(&r2->res_hashchain, &ls->ls_new_rsb);
378 ls->ls_new_rsb_count++;
379 }
380 count = ls->ls_new_rsb_count;
381 spin_unlock(&ls->ls_new_rsb_spin);
382
383 if (!count)
384 return -ENOMEM;
385 return 0;
386}
387
388/* If ls->ls_new_rsb is empty, return -EAGAIN, so the caller can
389 unlock any spinlocks, go back and call pre_rsb_struct again.
390 Otherwise, take an rsb off the list and return it. */
391
392static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len,
393 struct dlm_rsb **r_ret)
394{
395 struct dlm_rsb *r;
396 int count;
397
398 spin_lock(&ls->ls_new_rsb_spin);
399 if (list_empty(&ls->ls_new_rsb)) {
400 count = ls->ls_new_rsb_count;
401 spin_unlock(&ls->ls_new_rsb_spin);
402 log_debug(ls, "find_rsb retry %d %d %s",
403 count, dlm_config.ci_new_rsb_count,
404 (const char *)name);
405 return -EAGAIN;
406 }
407
408 r = list_first_entry(&ls->ls_new_rsb, struct dlm_rsb, res_hashchain);
409 list_del(&r->res_hashchain);
410 /* Convert the empty list_head to a NULL rb_node for tree usage: */
411 memset(&r->res_hashnode, 0, sizeof(struct rb_node));
412 ls->ls_new_rsb_count--;
413 spin_unlock(&ls->ls_new_rsb_spin);
414
415 r->res_ls = ls;
416 r->res_length = len;
417 memcpy(r->res_name, name, len);
418 mutex_init(&r->res_mutex);
419
420 INIT_LIST_HEAD(&r->res_lookup);
421 INIT_LIST_HEAD(&r->res_grantqueue);
422 INIT_LIST_HEAD(&r->res_convertqueue);
423 INIT_LIST_HEAD(&r->res_waitqueue);
424 INIT_LIST_HEAD(&r->res_root_list);
425 INIT_LIST_HEAD(&r->res_recover_list);
426
427 *r_ret = r;
428 return 0;
429}
430
431static int rsb_cmp(struct dlm_rsb *r, const char *name, int nlen)
432{
433 char maxname[DLM_RESNAME_MAXLEN];
434
435 memset(maxname, 0, DLM_RESNAME_MAXLEN);
436 memcpy(maxname, name, nlen);
437 return memcmp(r->res_name, maxname, DLM_RESNAME_MAXLEN);
438}
439
440int dlm_search_rsb_tree(struct rb_root *tree, const void *name, int len,
441 struct dlm_rsb **r_ret)
442{
443 struct rb_node *node = tree->rb_node;
444 struct dlm_rsb *r;
445 int rc;
446
447 while (node) {
448 r = rb_entry(node, struct dlm_rsb, res_hashnode);
449 rc = rsb_cmp(r, name, len);
450 if (rc < 0)
451 node = node->rb_left;
452 else if (rc > 0)
453 node = node->rb_right;
454 else
455 goto found;
456 }
457 *r_ret = NULL;
458 return -EBADR;
459
460 found:
461 *r_ret = r;
462 return 0;
463}
464
465static int rsb_insert(struct dlm_rsb *rsb, struct rb_root *tree)
466{
467 struct rb_node **newn = &tree->rb_node;
468 struct rb_node *parent = NULL;
469 int rc;
470
471 while (*newn) {
472 struct dlm_rsb *cur = rb_entry(*newn, struct dlm_rsb,
473 res_hashnode);
474
475 parent = *newn;
476 rc = rsb_cmp(cur, rsb->res_name, rsb->res_length);
477 if (rc < 0)
478 newn = &parent->rb_left;
479 else if (rc > 0)
480 newn = &parent->rb_right;
481 else {
482 log_print("rsb_insert match");
483 dlm_dump_rsb(rsb);
484 dlm_dump_rsb(cur);
485 return -EEXIST;
486 }
487 }
488
489 rb_link_node(&rsb->res_hashnode, parent, newn);
490 rb_insert_color(&rsb->res_hashnode, tree);
491 return 0;
492}
493
494/*
495 * Find rsb in rsbtbl and potentially create/add one
496 *
497 * Delaying the release of rsb's has a similar benefit to applications keeping
498 * NL locks on an rsb, but without the guarantee that the cached master value
499 * will still be valid when the rsb is reused. Apps aren't always smart enough
500 * to keep NL locks on an rsb that they may lock again shortly; this can lead
501 * to excessive master lookups and removals if we don't delay the release.
502 *
503 * Searching for an rsb means looking through both the normal list and toss
504 * list. When found on the toss list the rsb is moved to the normal list with
505 * ref count of 1; when found on normal list the ref count is incremented.
506 *
507 * rsb's on the keep list are being used locally and refcounted.
508 * rsb's on the toss list are not being used locally, and are not refcounted.
509 *
510 * The toss list rsb's were either
511 * - previously used locally but not any more (were on keep list, then
512 * moved to toss list when last refcount dropped)
513 * - created and put on toss list as a directory record for a lookup
514 * (we are the dir node for the res, but are not using the res right now,
515 * but some other node is)
516 *
517 * The purpose of find_rsb() is to return a refcounted rsb for local use.
518 * So, if the given rsb is on the toss list, it is moved to the keep list
519 * before being returned.
520 *
521 * toss_rsb() happens when all local usage of the rsb is done, i.e. no
522 * more refcounts exist, so the rsb is moved from the keep list to the
523 * toss list.
524 *
525 * rsb's on both keep and toss lists are used for doing a name to master
526 * lookups. rsb's that are in use locally (and being refcounted) are on
527 * the keep list, rsb's that are not in use locally (not refcounted) and
528 * only exist for name/master lookups are on the toss list.
529 *
530 * rsb's on the toss list who's dir_nodeid is not local can have stale
531 * name/master mappings. So, remote requests on such rsb's can potentially
532 * return with an error, which means the mapping is stale and needs to
533 * be updated with a new lookup. (The idea behind MASTER UNCERTAIN and
534 * first_lkid is to keep only a single outstanding request on an rsb
535 * while that rsb has a potentially stale master.)
536 */
537
538static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len,
539 uint32_t hash, uint32_t b,
540 int dir_nodeid, int from_nodeid,
541 unsigned int flags, struct dlm_rsb **r_ret)
542{
543 struct dlm_rsb *r = NULL;
544 int our_nodeid = dlm_our_nodeid();
545 int from_local = 0;
546 int from_other = 0;
547 int from_dir = 0;
548 int create = 0;
549 int error;
550
551 if (flags & R_RECEIVE_REQUEST) {
552 if (from_nodeid == dir_nodeid)
553 from_dir = 1;
554 else
555 from_other = 1;
556 } else if (flags & R_REQUEST) {
557 from_local = 1;
558 }
559
560 /*
561 * flags & R_RECEIVE_RECOVER is from dlm_recover_master_copy, so
562 * from_nodeid has sent us a lock in dlm_recover_locks, believing
563 * we're the new master. Our local recovery may not have set
564 * res_master_nodeid to our_nodeid yet, so allow either. Don't
565 * create the rsb; dlm_recover_process_copy() will handle EBADR
566 * by resending.
567 *
568 * If someone sends us a request, we are the dir node, and we do
569 * not find the rsb anywhere, then recreate it. This happens if
570 * someone sends us a request after we have removed/freed an rsb
571 * from our toss list. (They sent a request instead of lookup
572 * because they are using an rsb from their toss list.)
573 */
574
575 if (from_local || from_dir ||
576 (from_other && (dir_nodeid == our_nodeid))) {
577 create = 1;
578 }
579
580 retry:
581 if (create) {
582 error = pre_rsb_struct(ls);
583 if (error < 0)
584 goto out;
585 }
586
587 spin_lock(&ls->ls_rsbtbl[b].lock);
588
589 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r);
590 if (error)
591 goto do_toss;
592
593 /*
594 * rsb is active, so we can't check master_nodeid without lock_rsb.
595 */
596
597 kref_get(&r->res_ref);
598 goto out_unlock;
599
600
601 do_toss:
602 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r);
603 if (error)
604 goto do_new;
605
606 /*
607 * rsb found inactive (master_nodeid may be out of date unless
608 * we are the dir_nodeid or were the master) No other thread
609 * is using this rsb because it's on the toss list, so we can
610 * look at or update res_master_nodeid without lock_rsb.
611 */
612
613 if ((r->res_master_nodeid != our_nodeid) && from_other) {
614 /* our rsb was not master, and another node (not the dir node)
615 has sent us a request */
616 log_debug(ls, "find_rsb toss from_other %d master %d dir %d %s",
617 from_nodeid, r->res_master_nodeid, dir_nodeid,
618 r->res_name);
619 error = -ENOTBLK;
620 goto out_unlock;
621 }
622
623 if ((r->res_master_nodeid != our_nodeid) && from_dir) {
624 /* don't think this should ever happen */
625 log_error(ls, "find_rsb toss from_dir %d master %d",
626 from_nodeid, r->res_master_nodeid);
627 dlm_print_rsb(r);
628 /* fix it and go on */
629 r->res_master_nodeid = our_nodeid;
630 r->res_nodeid = 0;
631 rsb_clear_flag(r, RSB_MASTER_UNCERTAIN);
632 r->res_first_lkid = 0;
633 }
634
635 if (from_local && (r->res_master_nodeid != our_nodeid)) {
636 /* Because we have held no locks on this rsb,
637 res_master_nodeid could have become stale. */
638 rsb_set_flag(r, RSB_MASTER_UNCERTAIN);
639 r->res_first_lkid = 0;
640 }
641
642 rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss);
643 error = rsb_insert(r, &ls->ls_rsbtbl[b].keep);
644 goto out_unlock;
645
646
647 do_new:
648 /*
649 * rsb not found
650 */
651
652 if (error == -EBADR && !create)
653 goto out_unlock;
654
655 error = get_rsb_struct(ls, name, len, &r);
656 if (error == -EAGAIN) {
657 spin_unlock(&ls->ls_rsbtbl[b].lock);
658 goto retry;
659 }
660 if (error)
661 goto out_unlock;
662
663 r->res_hash = hash;
664 r->res_bucket = b;
665 r->res_dir_nodeid = dir_nodeid;
666 kref_init(&r->res_ref);
667
668 if (from_dir) {
669 /* want to see how often this happens */
670 log_debug(ls, "find_rsb new from_dir %d recreate %s",
671 from_nodeid, r->res_name);
672 r->res_master_nodeid = our_nodeid;
673 r->res_nodeid = 0;
674 goto out_add;
675 }
676
677 if (from_other && (dir_nodeid != our_nodeid)) {
678 /* should never happen */
679 log_error(ls, "find_rsb new from_other %d dir %d our %d %s",
680 from_nodeid, dir_nodeid, our_nodeid, r->res_name);
681 dlm_free_rsb(r);
682 r = NULL;
683 error = -ENOTBLK;
684 goto out_unlock;
685 }
686
687 if (from_other) {
688 log_debug(ls, "find_rsb new from_other %d dir %d %s",
689 from_nodeid, dir_nodeid, r->res_name);
690 }
691
692 if (dir_nodeid == our_nodeid) {
693 /* When we are the dir nodeid, we can set the master
694 node immediately */
695 r->res_master_nodeid = our_nodeid;
696 r->res_nodeid = 0;
697 } else {
698 /* set_master will send_lookup to dir_nodeid */
699 r->res_master_nodeid = 0;
700 r->res_nodeid = -1;
701 }
702
703 out_add:
704 error = rsb_insert(r, &ls->ls_rsbtbl[b].keep);
705 out_unlock:
706 spin_unlock(&ls->ls_rsbtbl[b].lock);
707 out:
708 *r_ret = r;
709 return error;
710}
711
712/* During recovery, other nodes can send us new MSTCPY locks (from
713 dlm_recover_locks) before we've made ourself master (in
714 dlm_recover_masters). */
715
716static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len,
717 uint32_t hash, uint32_t b,
718 int dir_nodeid, int from_nodeid,
719 unsigned int flags, struct dlm_rsb **r_ret)
720{
721 struct dlm_rsb *r = NULL;
722 int our_nodeid = dlm_our_nodeid();
723 int recover = (flags & R_RECEIVE_RECOVER);
724 int error;
725
726 retry:
727 error = pre_rsb_struct(ls);
728 if (error < 0)
729 goto out;
730
731 spin_lock(&ls->ls_rsbtbl[b].lock);
732
733 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r);
734 if (error)
735 goto do_toss;
736
737 /*
738 * rsb is active, so we can't check master_nodeid without lock_rsb.
739 */
740
741 kref_get(&r->res_ref);
742 goto out_unlock;
743
744
745 do_toss:
746 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r);
747 if (error)
748 goto do_new;
749
750 /*
751 * rsb found inactive. No other thread is using this rsb because
752 * it's on the toss list, so we can look at or update
753 * res_master_nodeid without lock_rsb.
754 */
755
756 if (!recover && (r->res_master_nodeid != our_nodeid) && from_nodeid) {
757 /* our rsb is not master, and another node has sent us a
758 request; this should never happen */
759 log_error(ls, "find_rsb toss from_nodeid %d master %d dir %d",
760 from_nodeid, r->res_master_nodeid, dir_nodeid);
761 dlm_print_rsb(r);
762 error = -ENOTBLK;
763 goto out_unlock;
764 }
765
766 if (!recover && (r->res_master_nodeid != our_nodeid) &&
767 (dir_nodeid == our_nodeid)) {
768 /* our rsb is not master, and we are dir; may as well fix it;
769 this should never happen */
770 log_error(ls, "find_rsb toss our %d master %d dir %d",
771 our_nodeid, r->res_master_nodeid, dir_nodeid);
772 dlm_print_rsb(r);
773 r->res_master_nodeid = our_nodeid;
774 r->res_nodeid = 0;
775 }
776
777 rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss);
778 error = rsb_insert(r, &ls->ls_rsbtbl[b].keep);
779 goto out_unlock;
780
781
782 do_new:
783 /*
784 * rsb not found
785 */
786
787 error = get_rsb_struct(ls, name, len, &r);
788 if (error == -EAGAIN) {
789 spin_unlock(&ls->ls_rsbtbl[b].lock);
790 goto retry;
791 }
792 if (error)
793 goto out_unlock;
794
795 r->res_hash = hash;
796 r->res_bucket = b;
797 r->res_dir_nodeid = dir_nodeid;
798 r->res_master_nodeid = dir_nodeid;
799 r->res_nodeid = (dir_nodeid == our_nodeid) ? 0 : dir_nodeid;
800 kref_init(&r->res_ref);
801
802 error = rsb_insert(r, &ls->ls_rsbtbl[b].keep);
803 out_unlock:
804 spin_unlock(&ls->ls_rsbtbl[b].lock);
805 out:
806 *r_ret = r;
807 return error;
808}
809
810static int find_rsb(struct dlm_ls *ls, const void *name, int len,
811 int from_nodeid, unsigned int flags,
812 struct dlm_rsb **r_ret)
813{
814 uint32_t hash, b;
815 int dir_nodeid;
816
817 if (len > DLM_RESNAME_MAXLEN)
818 return -EINVAL;
819
820 hash = jhash(name, len, 0);
821 b = hash & (ls->ls_rsbtbl_size - 1);
822
823 dir_nodeid = dlm_hash2nodeid(ls, hash);
824
825 if (dlm_no_directory(ls))
826 return find_rsb_nodir(ls, name, len, hash, b, dir_nodeid,
827 from_nodeid, flags, r_ret);
828 else
829 return find_rsb_dir(ls, name, len, hash, b, dir_nodeid,
830 from_nodeid, flags, r_ret);
831}
832
833/* we have received a request and found that res_master_nodeid != our_nodeid,
834 so we need to return an error or make ourself the master */
835
836static int validate_master_nodeid(struct dlm_ls *ls, struct dlm_rsb *r,
837 int from_nodeid)
838{
839 if (dlm_no_directory(ls)) {
840 log_error(ls, "find_rsb keep from_nodeid %d master %d dir %d",
841 from_nodeid, r->res_master_nodeid,
842 r->res_dir_nodeid);
843 dlm_print_rsb(r);
844 return -ENOTBLK;
845 }
846
847 if (from_nodeid != r->res_dir_nodeid) {
848 /* our rsb is not master, and another node (not the dir node)
849 has sent us a request. this is much more common when our
850 master_nodeid is zero, so limit debug to non-zero. */
851
852 if (r->res_master_nodeid) {
853 log_debug(ls, "validate master from_other %d master %d "
854 "dir %d first %x %s", from_nodeid,
855 r->res_master_nodeid, r->res_dir_nodeid,
856 r->res_first_lkid, r->res_name);
857 }
858 return -ENOTBLK;
859 } else {
860 /* our rsb is not master, but the dir nodeid has sent us a
861 request; this could happen with master 0 / res_nodeid -1 */
862
863 if (r->res_master_nodeid) {
864 log_error(ls, "validate master from_dir %d master %d "
865 "first %x %s",
866 from_nodeid, r->res_master_nodeid,
867 r->res_first_lkid, r->res_name);
868 }
869
870 r->res_master_nodeid = dlm_our_nodeid();
871 r->res_nodeid = 0;
872 return 0;
873 }
874}
875
876static void __dlm_master_lookup(struct dlm_ls *ls, struct dlm_rsb *r, int our_nodeid,
877 int from_nodeid, bool toss_list, unsigned int flags,
878 int *r_nodeid, int *result)
879{
880 int fix_master = (flags & DLM_LU_RECOVER_MASTER);
881 int from_master = (flags & DLM_LU_RECOVER_DIR);
882
883 if (r->res_dir_nodeid != our_nodeid) {
884 /* should not happen, but may as well fix it and carry on */
885 log_error(ls, "%s res_dir %d our %d %s", __func__,
886 r->res_dir_nodeid, our_nodeid, r->res_name);
887 r->res_dir_nodeid = our_nodeid;
888 }
889
890 if (fix_master && dlm_is_removed(ls, r->res_master_nodeid)) {
891 /* Recovery uses this function to set a new master when
892 * the previous master failed. Setting NEW_MASTER will
893 * force dlm_recover_masters to call recover_master on this
894 * rsb even though the res_nodeid is no longer removed.
895 */
896
897 r->res_master_nodeid = from_nodeid;
898 r->res_nodeid = from_nodeid;
899 rsb_set_flag(r, RSB_NEW_MASTER);
900
901 if (toss_list) {
902 /* I don't think we should ever find it on toss list. */
903 log_error(ls, "%s fix_master on toss", __func__);
904 dlm_dump_rsb(r);
905 }
906 }
907
908 if (from_master && (r->res_master_nodeid != from_nodeid)) {
909 /* this will happen if from_nodeid became master during
910 * a previous recovery cycle, and we aborted the previous
911 * cycle before recovering this master value
912 */
913
914 log_limit(ls, "%s from_master %d master_nodeid %d res_nodeid %d first %x %s",
915 __func__, from_nodeid, r->res_master_nodeid,
916 r->res_nodeid, r->res_first_lkid, r->res_name);
917
918 if (r->res_master_nodeid == our_nodeid) {
919 log_error(ls, "from_master %d our_master", from_nodeid);
920 dlm_dump_rsb(r);
921 goto ret_assign;
922 }
923
924 r->res_master_nodeid = from_nodeid;
925 r->res_nodeid = from_nodeid;
926 rsb_set_flag(r, RSB_NEW_MASTER);
927 }
928
929 if (!r->res_master_nodeid) {
930 /* this will happen if recovery happens while we're looking
931 * up the master for this rsb
932 */
933
934 log_debug(ls, "%s master 0 to %d first %x %s", __func__,
935 from_nodeid, r->res_first_lkid, r->res_name);
936 r->res_master_nodeid = from_nodeid;
937 r->res_nodeid = from_nodeid;
938 }
939
940 if (!from_master && !fix_master &&
941 (r->res_master_nodeid == from_nodeid)) {
942 /* this can happen when the master sends remove, the dir node
943 * finds the rsb on the keep list and ignores the remove,
944 * and the former master sends a lookup
945 */
946
947 log_limit(ls, "%s from master %d flags %x first %x %s",
948 __func__, from_nodeid, flags, r->res_first_lkid,
949 r->res_name);
950 }
951
952 ret_assign:
953 *r_nodeid = r->res_master_nodeid;
954 if (result)
955 *result = DLM_LU_MATCH;
956}
957
958/*
959 * We're the dir node for this res and another node wants to know the
960 * master nodeid. During normal operation (non recovery) this is only
961 * called from receive_lookup(); master lookups when the local node is
962 * the dir node are done by find_rsb().
963 *
964 * normal operation, we are the dir node for a resource
965 * . _request_lock
966 * . set_master
967 * . send_lookup
968 * . receive_lookup
969 * . dlm_master_lookup flags 0
970 *
971 * recover directory, we are rebuilding dir for all resources
972 * . dlm_recover_directory
973 * . dlm_rcom_names
974 * remote node sends back the rsb names it is master of and we are dir of
975 * . dlm_master_lookup RECOVER_DIR (fix_master 0, from_master 1)
976 * we either create new rsb setting remote node as master, or find existing
977 * rsb and set master to be the remote node.
978 *
979 * recover masters, we are finding the new master for resources
980 * . dlm_recover_masters
981 * . recover_master
982 * . dlm_send_rcom_lookup
983 * . receive_rcom_lookup
984 * . dlm_master_lookup RECOVER_MASTER (fix_master 1, from_master 0)
985 */
986
987int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name,
988 int len, unsigned int flags, int *r_nodeid, int *result)
989{
990 struct dlm_rsb *r = NULL;
991 uint32_t hash, b;
992 int our_nodeid = dlm_our_nodeid();
993 int dir_nodeid, error;
994
995 if (len > DLM_RESNAME_MAXLEN)
996 return -EINVAL;
997
998 if (from_nodeid == our_nodeid) {
999 log_error(ls, "dlm_master_lookup from our_nodeid %d flags %x",
1000 our_nodeid, flags);
1001 return -EINVAL;
1002 }
1003
1004 hash = jhash(name, len, 0);
1005 b = hash & (ls->ls_rsbtbl_size - 1);
1006
1007 dir_nodeid = dlm_hash2nodeid(ls, hash);
1008 if (dir_nodeid != our_nodeid) {
1009 log_error(ls, "dlm_master_lookup from %d dir %d our %d h %x %d",
1010 from_nodeid, dir_nodeid, our_nodeid, hash,
1011 ls->ls_num_nodes);
1012 *r_nodeid = -1;
1013 return -EINVAL;
1014 }
1015
1016 retry:
1017 error = pre_rsb_struct(ls);
1018 if (error < 0)
1019 return error;
1020
1021 spin_lock(&ls->ls_rsbtbl[b].lock);
1022 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r);
1023 if (!error) {
1024 /* because the rsb is active, we need to lock_rsb before
1025 * checking/changing re_master_nodeid
1026 */
1027
1028 hold_rsb(r);
1029 spin_unlock(&ls->ls_rsbtbl[b].lock);
1030 lock_rsb(r);
1031
1032 __dlm_master_lookup(ls, r, our_nodeid, from_nodeid, false,
1033 flags, r_nodeid, result);
1034
1035 /* the rsb was active */
1036 unlock_rsb(r);
1037 put_rsb(r);
1038
1039 return 0;
1040 }
1041
1042 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r);
1043 if (error)
1044 goto not_found;
1045
1046 /* because the rsb is inactive (on toss list), it's not refcounted
1047 * and lock_rsb is not used, but is protected by the rsbtbl lock
1048 */
1049
1050 __dlm_master_lookup(ls, r, our_nodeid, from_nodeid, true, flags,
1051 r_nodeid, result);
1052
1053 r->res_toss_time = jiffies;
1054 /* the rsb was inactive (on toss list) */
1055 spin_unlock(&ls->ls_rsbtbl[b].lock);
1056
1057 return 0;
1058
1059 not_found:
1060 error = get_rsb_struct(ls, name, len, &r);
1061 if (error == -EAGAIN) {
1062 spin_unlock(&ls->ls_rsbtbl[b].lock);
1063 goto retry;
1064 }
1065 if (error)
1066 goto out_unlock;
1067
1068 r->res_hash = hash;
1069 r->res_bucket = b;
1070 r->res_dir_nodeid = our_nodeid;
1071 r->res_master_nodeid = from_nodeid;
1072 r->res_nodeid = from_nodeid;
1073 kref_init(&r->res_ref);
1074 r->res_toss_time = jiffies;
1075
1076 error = rsb_insert(r, &ls->ls_rsbtbl[b].toss);
1077 if (error) {
1078 /* should never happen */
1079 dlm_free_rsb(r);
1080 spin_unlock(&ls->ls_rsbtbl[b].lock);
1081 goto retry;
1082 }
1083
1084 if (result)
1085 *result = DLM_LU_ADD;
1086 *r_nodeid = from_nodeid;
1087 out_unlock:
1088 spin_unlock(&ls->ls_rsbtbl[b].lock);
1089 return error;
1090}
1091
1092static void dlm_dump_rsb_hash(struct dlm_ls *ls, uint32_t hash)
1093{
1094 struct rb_node *n;
1095 struct dlm_rsb *r;
1096 int i;
1097
1098 for (i = 0; i < ls->ls_rsbtbl_size; i++) {
1099 spin_lock(&ls->ls_rsbtbl[i].lock);
1100 for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) {
1101 r = rb_entry(n, struct dlm_rsb, res_hashnode);
1102 if (r->res_hash == hash)
1103 dlm_dump_rsb(r);
1104 }
1105 spin_unlock(&ls->ls_rsbtbl[i].lock);
1106 }
1107}
1108
1109void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len)
1110{
1111 struct dlm_rsb *r = NULL;
1112 uint32_t hash, b;
1113 int error;
1114
1115 hash = jhash(name, len, 0);
1116 b = hash & (ls->ls_rsbtbl_size - 1);
1117
1118 spin_lock(&ls->ls_rsbtbl[b].lock);
1119 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r);
1120 if (!error)
1121 goto out_dump;
1122
1123 error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r);
1124 if (error)
1125 goto out;
1126 out_dump:
1127 dlm_dump_rsb(r);
1128 out:
1129 spin_unlock(&ls->ls_rsbtbl[b].lock);
1130}
1131
1132static void toss_rsb(struct kref *kref)
1133{
1134 struct dlm_rsb *r = container_of(kref, struct dlm_rsb, res_ref);
1135 struct dlm_ls *ls = r->res_ls;
1136
1137 DLM_ASSERT(list_empty(&r->res_root_list), dlm_print_rsb(r););
1138 kref_init(&r->res_ref);
1139 rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[r->res_bucket].keep);
1140 rsb_insert(r, &ls->ls_rsbtbl[r->res_bucket].toss);
1141 r->res_toss_time = jiffies;
1142 set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[r->res_bucket].flags);
1143 if (r->res_lvbptr) {
1144 dlm_free_lvb(r->res_lvbptr);
1145 r->res_lvbptr = NULL;
1146 }
1147}
1148
1149/* See comment for unhold_lkb */
1150
1151static void unhold_rsb(struct dlm_rsb *r)
1152{
1153 int rv;
1154 rv = kref_put(&r->res_ref, toss_rsb);
1155 DLM_ASSERT(!rv, dlm_dump_rsb(r););
1156}
1157
1158static void kill_rsb(struct kref *kref)
1159{
1160 struct dlm_rsb *r = container_of(kref, struct dlm_rsb, res_ref);
1161
1162 /* All work is done after the return from kref_put() so we
1163 can release the write_lock before the remove and free. */
1164
1165 DLM_ASSERT(list_empty(&r->res_lookup), dlm_dump_rsb(r););
1166 DLM_ASSERT(list_empty(&r->res_grantqueue), dlm_dump_rsb(r););
1167 DLM_ASSERT(list_empty(&r->res_convertqueue), dlm_dump_rsb(r););
1168 DLM_ASSERT(list_empty(&r->res_waitqueue), dlm_dump_rsb(r););
1169 DLM_ASSERT(list_empty(&r->res_root_list), dlm_dump_rsb(r););
1170 DLM_ASSERT(list_empty(&r->res_recover_list), dlm_dump_rsb(r););
1171}
1172
1173/* Attaching/detaching lkb's from rsb's is for rsb reference counting.
1174 The rsb must exist as long as any lkb's for it do. */
1175
1176static void attach_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb)
1177{
1178 hold_rsb(r);
1179 lkb->lkb_resource = r;
1180}
1181
1182static void detach_lkb(struct dlm_lkb *lkb)
1183{
1184 if (lkb->lkb_resource) {
1185 put_rsb(lkb->lkb_resource);
1186 lkb->lkb_resource = NULL;
1187 }
1188}
1189
1190static int _create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret,
1191 int start, int end)
1192{
1193 struct dlm_lkb *lkb;
1194 int rv;
1195
1196 lkb = dlm_allocate_lkb(ls);
1197 if (!lkb)
1198 return -ENOMEM;
1199
1200 lkb->lkb_last_bast_mode = -1;
1201 lkb->lkb_nodeid = -1;
1202 lkb->lkb_grmode = DLM_LOCK_IV;
1203 kref_init(&lkb->lkb_ref);
1204 INIT_LIST_HEAD(&lkb->lkb_ownqueue);
1205 INIT_LIST_HEAD(&lkb->lkb_rsb_lookup);
1206 INIT_LIST_HEAD(&lkb->lkb_cb_list);
1207 INIT_LIST_HEAD(&lkb->lkb_callbacks);
1208 spin_lock_init(&lkb->lkb_cb_lock);
1209 INIT_WORK(&lkb->lkb_cb_work, dlm_callback_work);
1210
1211 idr_preload(GFP_NOFS);
1212 spin_lock(&ls->ls_lkbidr_spin);
1213 rv = idr_alloc(&ls->ls_lkbidr, lkb, start, end, GFP_NOWAIT);
1214 if (rv >= 0)
1215 lkb->lkb_id = rv;
1216 spin_unlock(&ls->ls_lkbidr_spin);
1217 idr_preload_end();
1218
1219 if (rv < 0) {
1220 log_error(ls, "create_lkb idr error %d", rv);
1221 dlm_free_lkb(lkb);
1222 return rv;
1223 }
1224
1225 *lkb_ret = lkb;
1226 return 0;
1227}
1228
1229static int create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret)
1230{
1231 return _create_lkb(ls, lkb_ret, 1, 0);
1232}
1233
1234static int find_lkb(struct dlm_ls *ls, uint32_t lkid, struct dlm_lkb **lkb_ret)
1235{
1236 struct dlm_lkb *lkb;
1237
1238 spin_lock(&ls->ls_lkbidr_spin);
1239 lkb = idr_find(&ls->ls_lkbidr, lkid);
1240 if (lkb)
1241 kref_get(&lkb->lkb_ref);
1242 spin_unlock(&ls->ls_lkbidr_spin);
1243
1244 *lkb_ret = lkb;
1245 return lkb ? 0 : -ENOENT;
1246}
1247
1248static void kill_lkb(struct kref *kref)
1249{
1250 struct dlm_lkb *lkb = container_of(kref, struct dlm_lkb, lkb_ref);
1251
1252 /* All work is done after the return from kref_put() so we
1253 can release the write_lock before the detach_lkb */
1254
1255 DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb););
1256}
1257
1258/* __put_lkb() is used when an lkb may not have an rsb attached to
1259 it so we need to provide the lockspace explicitly */
1260
1261static int __put_lkb(struct dlm_ls *ls, struct dlm_lkb *lkb)
1262{
1263 uint32_t lkid = lkb->lkb_id;
1264 int rv;
1265
1266 rv = kref_put_lock(&lkb->lkb_ref, kill_lkb,
1267 &ls->ls_lkbidr_spin);
1268 if (rv) {
1269 idr_remove(&ls->ls_lkbidr, lkid);
1270 spin_unlock(&ls->ls_lkbidr_spin);
1271
1272 detach_lkb(lkb);
1273
1274 /* for local/process lkbs, lvbptr points to caller's lksb */
1275 if (lkb->lkb_lvbptr && is_master_copy(lkb))
1276 dlm_free_lvb(lkb->lkb_lvbptr);
1277 dlm_free_lkb(lkb);
1278 }
1279
1280 return rv;
1281}
1282
1283int dlm_put_lkb(struct dlm_lkb *lkb)
1284{
1285 struct dlm_ls *ls;
1286
1287 DLM_ASSERT(lkb->lkb_resource, dlm_print_lkb(lkb););
1288 DLM_ASSERT(lkb->lkb_resource->res_ls, dlm_print_lkb(lkb););
1289
1290 ls = lkb->lkb_resource->res_ls;
1291 return __put_lkb(ls, lkb);
1292}
1293
1294/* This is only called to add a reference when the code already holds
1295 a valid reference to the lkb, so there's no need for locking. */
1296
1297static inline void hold_lkb(struct dlm_lkb *lkb)
1298{
1299 kref_get(&lkb->lkb_ref);
1300}
1301
1302static void unhold_lkb_assert(struct kref *kref)
1303{
1304 struct dlm_lkb *lkb = container_of(kref, struct dlm_lkb, lkb_ref);
1305
1306 DLM_ASSERT(false, dlm_print_lkb(lkb););
1307}
1308
1309/* This is called when we need to remove a reference and are certain
1310 it's not the last ref. e.g. del_lkb is always called between a
1311 find_lkb/put_lkb and is always the inverse of a previous add_lkb.
1312 put_lkb would work fine, but would involve unnecessary locking */
1313
1314static inline void unhold_lkb(struct dlm_lkb *lkb)
1315{
1316 kref_put(&lkb->lkb_ref, unhold_lkb_assert);
1317}
1318
1319static void lkb_add_ordered(struct list_head *new, struct list_head *head,
1320 int mode)
1321{
1322 struct dlm_lkb *lkb = NULL, *iter;
1323
1324 list_for_each_entry(iter, head, lkb_statequeue)
1325 if (iter->lkb_rqmode < mode) {
1326 lkb = iter;
1327 list_add_tail(new, &iter->lkb_statequeue);
1328 break;
1329 }
1330
1331 if (!lkb)
1332 list_add_tail(new, head);
1333}
1334
1335/* add/remove lkb to rsb's grant/convert/wait queue */
1336
1337static void add_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb, int status)
1338{
1339 kref_get(&lkb->lkb_ref);
1340
1341 DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb););
1342
1343 lkb->lkb_timestamp = ktime_get();
1344
1345 lkb->lkb_status = status;
1346
1347 switch (status) {
1348 case DLM_LKSTS_WAITING:
1349 if (lkb->lkb_exflags & DLM_LKF_HEADQUE)
1350 list_add(&lkb->lkb_statequeue, &r->res_waitqueue);
1351 else
1352 list_add_tail(&lkb->lkb_statequeue, &r->res_waitqueue);
1353 break;
1354 case DLM_LKSTS_GRANTED:
1355 /* convention says granted locks kept in order of grmode */
1356 lkb_add_ordered(&lkb->lkb_statequeue, &r->res_grantqueue,
1357 lkb->lkb_grmode);
1358 break;
1359 case DLM_LKSTS_CONVERT:
1360 if (lkb->lkb_exflags & DLM_LKF_HEADQUE)
1361 list_add(&lkb->lkb_statequeue, &r->res_convertqueue);
1362 else
1363 list_add_tail(&lkb->lkb_statequeue,
1364 &r->res_convertqueue);
1365 break;
1366 default:
1367 DLM_ASSERT(0, dlm_print_lkb(lkb); printk("sts=%d\n", status););
1368 }
1369}
1370
1371static void del_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb)
1372{
1373 lkb->lkb_status = 0;
1374 list_del(&lkb->lkb_statequeue);
1375 unhold_lkb(lkb);
1376}
1377
1378static void move_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb, int sts)
1379{
1380 hold_lkb(lkb);
1381 del_lkb(r, lkb);
1382 add_lkb(r, lkb, sts);
1383 unhold_lkb(lkb);
1384}
1385
1386static int msg_reply_type(int mstype)
1387{
1388 switch (mstype) {
1389 case DLM_MSG_REQUEST:
1390 return DLM_MSG_REQUEST_REPLY;
1391 case DLM_MSG_CONVERT:
1392 return DLM_MSG_CONVERT_REPLY;
1393 case DLM_MSG_UNLOCK:
1394 return DLM_MSG_UNLOCK_REPLY;
1395 case DLM_MSG_CANCEL:
1396 return DLM_MSG_CANCEL_REPLY;
1397 case DLM_MSG_LOOKUP:
1398 return DLM_MSG_LOOKUP_REPLY;
1399 }
1400 return -1;
1401}
1402
1403/* add/remove lkb from global waiters list of lkb's waiting for
1404 a reply from a remote node */
1405
1406static int add_to_waiters(struct dlm_lkb *lkb, int mstype, int to_nodeid)
1407{
1408 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1409 int error = 0;
1410
1411 mutex_lock(&ls->ls_waiters_mutex);
1412
1413 if (is_overlap_unlock(lkb) ||
1414 (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL))) {
1415 error = -EINVAL;
1416 goto out;
1417 }
1418
1419 if (lkb->lkb_wait_type || is_overlap_cancel(lkb)) {
1420 switch (mstype) {
1421 case DLM_MSG_UNLOCK:
1422 set_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags);
1423 break;
1424 case DLM_MSG_CANCEL:
1425 set_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags);
1426 break;
1427 default:
1428 error = -EBUSY;
1429 goto out;
1430 }
1431 lkb->lkb_wait_count++;
1432 hold_lkb(lkb);
1433
1434 log_debug(ls, "addwait %x cur %d overlap %d count %d f %x",
1435 lkb->lkb_id, lkb->lkb_wait_type, mstype,
1436 lkb->lkb_wait_count, dlm_iflags_val(lkb));
1437 goto out;
1438 }
1439
1440 DLM_ASSERT(!lkb->lkb_wait_count,
1441 dlm_print_lkb(lkb);
1442 printk("wait_count %d\n", lkb->lkb_wait_count););
1443
1444 lkb->lkb_wait_count++;
1445 lkb->lkb_wait_type = mstype;
1446 lkb->lkb_wait_nodeid = to_nodeid; /* for debugging */
1447 hold_lkb(lkb);
1448 list_add(&lkb->lkb_wait_reply, &ls->ls_waiters);
1449 out:
1450 if (error)
1451 log_error(ls, "addwait error %x %d flags %x %d %d %s",
1452 lkb->lkb_id, error, dlm_iflags_val(lkb), mstype,
1453 lkb->lkb_wait_type, lkb->lkb_resource->res_name);
1454 mutex_unlock(&ls->ls_waiters_mutex);
1455 return error;
1456}
1457
1458/* We clear the RESEND flag because we might be taking an lkb off the waiters
1459 list as part of process_requestqueue (e.g. a lookup that has an optimized
1460 request reply on the requestqueue) between dlm_recover_waiters_pre() which
1461 set RESEND and dlm_recover_waiters_post() */
1462
1463static int _remove_from_waiters(struct dlm_lkb *lkb, int mstype,
1464 const struct dlm_message *ms)
1465{
1466 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1467 int overlap_done = 0;
1468
1469 if (mstype == DLM_MSG_UNLOCK_REPLY &&
1470 test_and_clear_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags)) {
1471 log_debug(ls, "remwait %x unlock_reply overlap", lkb->lkb_id);
1472 overlap_done = 1;
1473 goto out_del;
1474 }
1475
1476 if (mstype == DLM_MSG_CANCEL_REPLY &&
1477 test_and_clear_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags)) {
1478 log_debug(ls, "remwait %x cancel_reply overlap", lkb->lkb_id);
1479 overlap_done = 1;
1480 goto out_del;
1481 }
1482
1483 /* Cancel state was preemptively cleared by a successful convert,
1484 see next comment, nothing to do. */
1485
1486 if ((mstype == DLM_MSG_CANCEL_REPLY) &&
1487 (lkb->lkb_wait_type != DLM_MSG_CANCEL)) {
1488 log_debug(ls, "remwait %x cancel_reply wait_type %d",
1489 lkb->lkb_id, lkb->lkb_wait_type);
1490 return -1;
1491 }
1492
1493 /* Remove for the convert reply, and premptively remove for the
1494 cancel reply. A convert has been granted while there's still
1495 an outstanding cancel on it (the cancel is moot and the result
1496 in the cancel reply should be 0). We preempt the cancel reply
1497 because the app gets the convert result and then can follow up
1498 with another op, like convert. This subsequent op would see the
1499 lingering state of the cancel and fail with -EBUSY. */
1500
1501 if ((mstype == DLM_MSG_CONVERT_REPLY) &&
1502 (lkb->lkb_wait_type == DLM_MSG_CONVERT) && ms && !ms->m_result &&
1503 test_and_clear_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags)) {
1504 log_debug(ls, "remwait %x convert_reply zap overlap_cancel",
1505 lkb->lkb_id);
1506 lkb->lkb_wait_type = 0;
1507 lkb->lkb_wait_count--;
1508 unhold_lkb(lkb);
1509 goto out_del;
1510 }
1511
1512 /* N.B. type of reply may not always correspond to type of original
1513 msg due to lookup->request optimization, verify others? */
1514
1515 if (lkb->lkb_wait_type) {
1516 lkb->lkb_wait_type = 0;
1517 goto out_del;
1518 }
1519
1520 log_error(ls, "remwait error %x remote %d %x msg %d flags %x no wait",
1521 lkb->lkb_id, ms ? le32_to_cpu(ms->m_header.h_nodeid) : 0,
1522 lkb->lkb_remid, mstype, dlm_iflags_val(lkb));
1523 return -1;
1524
1525 out_del:
1526 /* the force-unlock/cancel has completed and we haven't recvd a reply
1527 to the op that was in progress prior to the unlock/cancel; we
1528 give up on any reply to the earlier op. FIXME: not sure when/how
1529 this would happen */
1530
1531 if (overlap_done && lkb->lkb_wait_type) {
1532 log_error(ls, "remwait error %x reply %d wait_type %d overlap",
1533 lkb->lkb_id, mstype, lkb->lkb_wait_type);
1534 lkb->lkb_wait_count--;
1535 unhold_lkb(lkb);
1536 lkb->lkb_wait_type = 0;
1537 }
1538
1539 DLM_ASSERT(lkb->lkb_wait_count, dlm_print_lkb(lkb););
1540
1541 clear_bit(DLM_IFL_RESEND_BIT, &lkb->lkb_iflags);
1542 lkb->lkb_wait_count--;
1543 if (!lkb->lkb_wait_count)
1544 list_del_init(&lkb->lkb_wait_reply);
1545 unhold_lkb(lkb);
1546 return 0;
1547}
1548
1549static int remove_from_waiters(struct dlm_lkb *lkb, int mstype)
1550{
1551 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1552 int error;
1553
1554 mutex_lock(&ls->ls_waiters_mutex);
1555 error = _remove_from_waiters(lkb, mstype, NULL);
1556 mutex_unlock(&ls->ls_waiters_mutex);
1557 return error;
1558}
1559
1560/* Handles situations where we might be processing a "fake" or "local" reply in
1561 which we can't try to take waiters_mutex again. */
1562
1563static int remove_from_waiters_ms(struct dlm_lkb *lkb,
1564 const struct dlm_message *ms, bool local)
1565{
1566 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
1567 int error;
1568
1569 if (!local)
1570 mutex_lock(&ls->ls_waiters_mutex);
1571 error = _remove_from_waiters(lkb, le32_to_cpu(ms->m_type), ms);
1572 if (!local)
1573 mutex_unlock(&ls->ls_waiters_mutex);
1574 return error;
1575}
1576
1577static void shrink_bucket(struct dlm_ls *ls, int b)
1578{
1579 struct rb_node *n, *next;
1580 struct dlm_rsb *r;
1581 char *name;
1582 int our_nodeid = dlm_our_nodeid();
1583 int remote_count = 0;
1584 int need_shrink = 0;
1585 int i, len, rv;
1586
1587 memset(&ls->ls_remove_lens, 0, sizeof(int) * DLM_REMOVE_NAMES_MAX);
1588
1589 spin_lock(&ls->ls_rsbtbl[b].lock);
1590
1591 if (!test_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags)) {
1592 spin_unlock(&ls->ls_rsbtbl[b].lock);
1593 return;
1594 }
1595
1596 for (n = rb_first(&ls->ls_rsbtbl[b].toss); n; n = next) {
1597 next = rb_next(n);
1598 r = rb_entry(n, struct dlm_rsb, res_hashnode);
1599
1600 /* If we're the directory record for this rsb, and
1601 we're not the master of it, then we need to wait
1602 for the master node to send us a dir remove for
1603 before removing the dir record. */
1604
1605 if (!dlm_no_directory(ls) &&
1606 (r->res_master_nodeid != our_nodeid) &&
1607 (dlm_dir_nodeid(r) == our_nodeid)) {
1608 continue;
1609 }
1610
1611 need_shrink = 1;
1612
1613 if (!time_after_eq(jiffies, r->res_toss_time +
1614 dlm_config.ci_toss_secs * HZ)) {
1615 continue;
1616 }
1617
1618 if (!dlm_no_directory(ls) &&
1619 (r->res_master_nodeid == our_nodeid) &&
1620 (dlm_dir_nodeid(r) != our_nodeid)) {
1621
1622 /* We're the master of this rsb but we're not
1623 the directory record, so we need to tell the
1624 dir node to remove the dir record. */
1625
1626 ls->ls_remove_lens[remote_count] = r->res_length;
1627 memcpy(ls->ls_remove_names[remote_count], r->res_name,
1628 DLM_RESNAME_MAXLEN);
1629 remote_count++;
1630
1631 if (remote_count >= DLM_REMOVE_NAMES_MAX)
1632 break;
1633 continue;
1634 }
1635
1636 if (!kref_put(&r->res_ref, kill_rsb)) {
1637 log_error(ls, "tossed rsb in use %s", r->res_name);
1638 continue;
1639 }
1640
1641 rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss);
1642 dlm_free_rsb(r);
1643 }
1644
1645 if (need_shrink)
1646 set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags);
1647 else
1648 clear_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags);
1649 spin_unlock(&ls->ls_rsbtbl[b].lock);
1650
1651 /*
1652 * While searching for rsb's to free, we found some that require
1653 * remote removal. We leave them in place and find them again here
1654 * so there is a very small gap between removing them from the toss
1655 * list and sending the removal. Keeping this gap small is
1656 * important to keep us (the master node) from being out of sync
1657 * with the remote dir node for very long.
1658 */
1659
1660 for (i = 0; i < remote_count; i++) {
1661 name = ls->ls_remove_names[i];
1662 len = ls->ls_remove_lens[i];
1663
1664 spin_lock(&ls->ls_rsbtbl[b].lock);
1665 rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r);
1666 if (rv) {
1667 spin_unlock(&ls->ls_rsbtbl[b].lock);
1668 log_debug(ls, "remove_name not toss %s", name);
1669 continue;
1670 }
1671
1672 if (r->res_master_nodeid != our_nodeid) {
1673 spin_unlock(&ls->ls_rsbtbl[b].lock);
1674 log_debug(ls, "remove_name master %d dir %d our %d %s",
1675 r->res_master_nodeid, r->res_dir_nodeid,
1676 our_nodeid, name);
1677 continue;
1678 }
1679
1680 if (r->res_dir_nodeid == our_nodeid) {
1681 /* should never happen */
1682 spin_unlock(&ls->ls_rsbtbl[b].lock);
1683 log_error(ls, "remove_name dir %d master %d our %d %s",
1684 r->res_dir_nodeid, r->res_master_nodeid,
1685 our_nodeid, name);
1686 continue;
1687 }
1688
1689 if (!time_after_eq(jiffies, r->res_toss_time +
1690 dlm_config.ci_toss_secs * HZ)) {
1691 spin_unlock(&ls->ls_rsbtbl[b].lock);
1692 log_debug(ls, "remove_name toss_time %lu now %lu %s",
1693 r->res_toss_time, jiffies, name);
1694 continue;
1695 }
1696
1697 if (!kref_put(&r->res_ref, kill_rsb)) {
1698 spin_unlock(&ls->ls_rsbtbl[b].lock);
1699 log_error(ls, "remove_name in use %s", name);
1700 continue;
1701 }
1702
1703 rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss);
1704 send_remove(r);
1705 spin_unlock(&ls->ls_rsbtbl[b].lock);
1706
1707 dlm_free_rsb(r);
1708 }
1709}
1710
1711void dlm_scan_rsbs(struct dlm_ls *ls)
1712{
1713 int i;
1714
1715 for (i = 0; i < ls->ls_rsbtbl_size; i++) {
1716 shrink_bucket(ls, i);
1717 if (dlm_locking_stopped(ls))
1718 break;
1719 cond_resched();
1720 }
1721}
1722
1723/* lkb is master or local copy */
1724
1725static void set_lvb_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1726{
1727 int b, len = r->res_ls->ls_lvblen;
1728
1729 /* b=1 lvb returned to caller
1730 b=0 lvb written to rsb or invalidated
1731 b=-1 do nothing */
1732
1733 b = dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1];
1734
1735 if (b == 1) {
1736 if (!lkb->lkb_lvbptr)
1737 return;
1738
1739 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1740 return;
1741
1742 if (!r->res_lvbptr)
1743 return;
1744
1745 memcpy(lkb->lkb_lvbptr, r->res_lvbptr, len);
1746 lkb->lkb_lvbseq = r->res_lvbseq;
1747
1748 } else if (b == 0) {
1749 if (lkb->lkb_exflags & DLM_LKF_IVVALBLK) {
1750 rsb_set_flag(r, RSB_VALNOTVALID);
1751 return;
1752 }
1753
1754 if (!lkb->lkb_lvbptr)
1755 return;
1756
1757 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1758 return;
1759
1760 if (!r->res_lvbptr)
1761 r->res_lvbptr = dlm_allocate_lvb(r->res_ls);
1762
1763 if (!r->res_lvbptr)
1764 return;
1765
1766 memcpy(r->res_lvbptr, lkb->lkb_lvbptr, len);
1767 r->res_lvbseq++;
1768 lkb->lkb_lvbseq = r->res_lvbseq;
1769 rsb_clear_flag(r, RSB_VALNOTVALID);
1770 }
1771
1772 if (rsb_flag(r, RSB_VALNOTVALID))
1773 set_bit(DLM_SBF_VALNOTVALID_BIT, &lkb->lkb_sbflags);
1774}
1775
1776static void set_lvb_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1777{
1778 if (lkb->lkb_grmode < DLM_LOCK_PW)
1779 return;
1780
1781 if (lkb->lkb_exflags & DLM_LKF_IVVALBLK) {
1782 rsb_set_flag(r, RSB_VALNOTVALID);
1783 return;
1784 }
1785
1786 if (!lkb->lkb_lvbptr)
1787 return;
1788
1789 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1790 return;
1791
1792 if (!r->res_lvbptr)
1793 r->res_lvbptr = dlm_allocate_lvb(r->res_ls);
1794
1795 if (!r->res_lvbptr)
1796 return;
1797
1798 memcpy(r->res_lvbptr, lkb->lkb_lvbptr, r->res_ls->ls_lvblen);
1799 r->res_lvbseq++;
1800 rsb_clear_flag(r, RSB_VALNOTVALID);
1801}
1802
1803/* lkb is process copy (pc) */
1804
1805static void set_lvb_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb,
1806 const struct dlm_message *ms)
1807{
1808 int b;
1809
1810 if (!lkb->lkb_lvbptr)
1811 return;
1812
1813 if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
1814 return;
1815
1816 b = dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1];
1817 if (b == 1) {
1818 int len = receive_extralen(ms);
1819 if (len > r->res_ls->ls_lvblen)
1820 len = r->res_ls->ls_lvblen;
1821 memcpy(lkb->lkb_lvbptr, ms->m_extra, len);
1822 lkb->lkb_lvbseq = le32_to_cpu(ms->m_lvbseq);
1823 }
1824}
1825
1826/* Manipulate lkb's on rsb's convert/granted/waiting queues
1827 remove_lock -- used for unlock, removes lkb from granted
1828 revert_lock -- used for cancel, moves lkb from convert to granted
1829 grant_lock -- used for request and convert, adds lkb to granted or
1830 moves lkb from convert or waiting to granted
1831
1832 Each of these is used for master or local copy lkb's. There is
1833 also a _pc() variation used to make the corresponding change on
1834 a process copy (pc) lkb. */
1835
1836static void _remove_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1837{
1838 del_lkb(r, lkb);
1839 lkb->lkb_grmode = DLM_LOCK_IV;
1840 /* this unhold undoes the original ref from create_lkb()
1841 so this leads to the lkb being freed */
1842 unhold_lkb(lkb);
1843}
1844
1845static void remove_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1846{
1847 set_lvb_unlock(r, lkb);
1848 _remove_lock(r, lkb);
1849}
1850
1851static void remove_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
1852{
1853 _remove_lock(r, lkb);
1854}
1855
1856/* returns: 0 did nothing
1857 1 moved lock to granted
1858 -1 removed lock */
1859
1860static int revert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1861{
1862 int rv = 0;
1863
1864 lkb->lkb_rqmode = DLM_LOCK_IV;
1865
1866 switch (lkb->lkb_status) {
1867 case DLM_LKSTS_GRANTED:
1868 break;
1869 case DLM_LKSTS_CONVERT:
1870 move_lkb(r, lkb, DLM_LKSTS_GRANTED);
1871 rv = 1;
1872 break;
1873 case DLM_LKSTS_WAITING:
1874 del_lkb(r, lkb);
1875 lkb->lkb_grmode = DLM_LOCK_IV;
1876 /* this unhold undoes the original ref from create_lkb()
1877 so this leads to the lkb being freed */
1878 unhold_lkb(lkb);
1879 rv = -1;
1880 break;
1881 default:
1882 log_print("invalid status for revert %d", lkb->lkb_status);
1883 }
1884 return rv;
1885}
1886
1887static int revert_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
1888{
1889 return revert_lock(r, lkb);
1890}
1891
1892static void _grant_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1893{
1894 if (lkb->lkb_grmode != lkb->lkb_rqmode) {
1895 lkb->lkb_grmode = lkb->lkb_rqmode;
1896 if (lkb->lkb_status)
1897 move_lkb(r, lkb, DLM_LKSTS_GRANTED);
1898 else
1899 add_lkb(r, lkb, DLM_LKSTS_GRANTED);
1900 }
1901
1902 lkb->lkb_rqmode = DLM_LOCK_IV;
1903 lkb->lkb_highbast = 0;
1904}
1905
1906static void grant_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1907{
1908 set_lvb_lock(r, lkb);
1909 _grant_lock(r, lkb);
1910}
1911
1912static void grant_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb,
1913 const struct dlm_message *ms)
1914{
1915 set_lvb_lock_pc(r, lkb, ms);
1916 _grant_lock(r, lkb);
1917}
1918
1919/* called by grant_pending_locks() which means an async grant message must
1920 be sent to the requesting node in addition to granting the lock if the
1921 lkb belongs to a remote node. */
1922
1923static void grant_lock_pending(struct dlm_rsb *r, struct dlm_lkb *lkb)
1924{
1925 grant_lock(r, lkb);
1926 if (is_master_copy(lkb))
1927 send_grant(r, lkb);
1928 else
1929 queue_cast(r, lkb, 0);
1930}
1931
1932/* The special CONVDEADLK, ALTPR and ALTCW flags allow the master to
1933 change the granted/requested modes. We're munging things accordingly in
1934 the process copy.
1935 CONVDEADLK: our grmode may have been forced down to NL to resolve a
1936 conversion deadlock
1937 ALTPR/ALTCW: our rqmode may have been changed to PR or CW to become
1938 compatible with other granted locks */
1939
1940static void munge_demoted(struct dlm_lkb *lkb)
1941{
1942 if (lkb->lkb_rqmode == DLM_LOCK_IV || lkb->lkb_grmode == DLM_LOCK_IV) {
1943 log_print("munge_demoted %x invalid modes gr %d rq %d",
1944 lkb->lkb_id, lkb->lkb_grmode, lkb->lkb_rqmode);
1945 return;
1946 }
1947
1948 lkb->lkb_grmode = DLM_LOCK_NL;
1949}
1950
1951static void munge_altmode(struct dlm_lkb *lkb, const struct dlm_message *ms)
1952{
1953 if (ms->m_type != cpu_to_le32(DLM_MSG_REQUEST_REPLY) &&
1954 ms->m_type != cpu_to_le32(DLM_MSG_GRANT)) {
1955 log_print("munge_altmode %x invalid reply type %d",
1956 lkb->lkb_id, le32_to_cpu(ms->m_type));
1957 return;
1958 }
1959
1960 if (lkb->lkb_exflags & DLM_LKF_ALTPR)
1961 lkb->lkb_rqmode = DLM_LOCK_PR;
1962 else if (lkb->lkb_exflags & DLM_LKF_ALTCW)
1963 lkb->lkb_rqmode = DLM_LOCK_CW;
1964 else {
1965 log_print("munge_altmode invalid exflags %x", lkb->lkb_exflags);
1966 dlm_print_lkb(lkb);
1967 }
1968}
1969
1970static inline int first_in_list(struct dlm_lkb *lkb, struct list_head *head)
1971{
1972 struct dlm_lkb *first = list_entry(head->next, struct dlm_lkb,
1973 lkb_statequeue);
1974 if (lkb->lkb_id == first->lkb_id)
1975 return 1;
1976
1977 return 0;
1978}
1979
1980/* Check if the given lkb conflicts with another lkb on the queue. */
1981
1982static int queue_conflict(struct list_head *head, struct dlm_lkb *lkb)
1983{
1984 struct dlm_lkb *this;
1985
1986 list_for_each_entry(this, head, lkb_statequeue) {
1987 if (this == lkb)
1988 continue;
1989 if (!modes_compat(this, lkb))
1990 return 1;
1991 }
1992 return 0;
1993}
1994
1995/*
1996 * "A conversion deadlock arises with a pair of lock requests in the converting
1997 * queue for one resource. The granted mode of each lock blocks the requested
1998 * mode of the other lock."
1999 *
2000 * Part 2: if the granted mode of lkb is preventing an earlier lkb in the
2001 * convert queue from being granted, then deadlk/demote lkb.
2002 *
2003 * Example:
2004 * Granted Queue: empty
2005 * Convert Queue: NL->EX (first lock)
2006 * PR->EX (second lock)
2007 *
2008 * The first lock can't be granted because of the granted mode of the second
2009 * lock and the second lock can't be granted because it's not first in the
2010 * list. We either cancel lkb's conversion (PR->EX) and return EDEADLK, or we
2011 * demote the granted mode of lkb (from PR to NL) if it has the CONVDEADLK
2012 * flag set and return DEMOTED in the lksb flags.
2013 *
2014 * Originally, this function detected conv-deadlk in a more limited scope:
2015 * - if !modes_compat(lkb1, lkb2) && !modes_compat(lkb2, lkb1), or
2016 * - if lkb1 was the first entry in the queue (not just earlier), and was
2017 * blocked by the granted mode of lkb2, and there was nothing on the
2018 * granted queue preventing lkb1 from being granted immediately, i.e.
2019 * lkb2 was the only thing preventing lkb1 from being granted.
2020 *
2021 * That second condition meant we'd only say there was conv-deadlk if
2022 * resolving it (by demotion) would lead to the first lock on the convert
2023 * queue being granted right away. It allowed conversion deadlocks to exist
2024 * between locks on the convert queue while they couldn't be granted anyway.
2025 *
2026 * Now, we detect and take action on conversion deadlocks immediately when
2027 * they're created, even if they may not be immediately consequential. If
2028 * lkb1 exists anywhere in the convert queue and lkb2 comes in with a granted
2029 * mode that would prevent lkb1's conversion from being granted, we do a
2030 * deadlk/demote on lkb2 right away and don't let it onto the convert queue.
2031 * I think this means that the lkb_is_ahead condition below should always
2032 * be zero, i.e. there will never be conv-deadlk between two locks that are
2033 * both already on the convert queue.
2034 */
2035
2036static int conversion_deadlock_detect(struct dlm_rsb *r, struct dlm_lkb *lkb2)
2037{
2038 struct dlm_lkb *lkb1;
2039 int lkb_is_ahead = 0;
2040
2041 list_for_each_entry(lkb1, &r->res_convertqueue, lkb_statequeue) {
2042 if (lkb1 == lkb2) {
2043 lkb_is_ahead = 1;
2044 continue;
2045 }
2046
2047 if (!lkb_is_ahead) {
2048 if (!modes_compat(lkb2, lkb1))
2049 return 1;
2050 } else {
2051 if (!modes_compat(lkb2, lkb1) &&
2052 !modes_compat(lkb1, lkb2))
2053 return 1;
2054 }
2055 }
2056 return 0;
2057}
2058
2059/*
2060 * Return 1 if the lock can be granted, 0 otherwise.
2061 * Also detect and resolve conversion deadlocks.
2062 *
2063 * lkb is the lock to be granted
2064 *
2065 * now is 1 if the function is being called in the context of the
2066 * immediate request, it is 0 if called later, after the lock has been
2067 * queued.
2068 *
2069 * recover is 1 if dlm_recover_grant() is trying to grant conversions
2070 * after recovery.
2071 *
2072 * References are from chapter 6 of "VAXcluster Principles" by Roy Davis
2073 */
2074
2075static int _can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now,
2076 int recover)
2077{
2078 int8_t conv = (lkb->lkb_grmode != DLM_LOCK_IV);
2079
2080 /*
2081 * 6-10: Version 5.4 introduced an option to address the phenomenon of
2082 * a new request for a NL mode lock being blocked.
2083 *
2084 * 6-11: If the optional EXPEDITE flag is used with the new NL mode
2085 * request, then it would be granted. In essence, the use of this flag
2086 * tells the Lock Manager to expedite theis request by not considering
2087 * what may be in the CONVERTING or WAITING queues... As of this
2088 * writing, the EXPEDITE flag can be used only with new requests for NL
2089 * mode locks. This flag is not valid for conversion requests.
2090 *
2091 * A shortcut. Earlier checks return an error if EXPEDITE is used in a
2092 * conversion or used with a non-NL requested mode. We also know an
2093 * EXPEDITE request is always granted immediately, so now must always
2094 * be 1. The full condition to grant an expedite request: (now &&
2095 * !conv && lkb->rqmode == DLM_LOCK_NL && (flags & EXPEDITE)) can
2096 * therefore be shortened to just checking the flag.
2097 */
2098
2099 if (lkb->lkb_exflags & DLM_LKF_EXPEDITE)
2100 return 1;
2101
2102 /*
2103 * A shortcut. Without this, !queue_conflict(grantqueue, lkb) would be
2104 * added to the remaining conditions.
2105 */
2106
2107 if (queue_conflict(&r->res_grantqueue, lkb))
2108 return 0;
2109
2110 /*
2111 * 6-3: By default, a conversion request is immediately granted if the
2112 * requested mode is compatible with the modes of all other granted
2113 * locks
2114 */
2115
2116 if (queue_conflict(&r->res_convertqueue, lkb))
2117 return 0;
2118
2119 /*
2120 * The RECOVER_GRANT flag means dlm_recover_grant() is granting
2121 * locks for a recovered rsb, on which lkb's have been rebuilt.
2122 * The lkb's may have been rebuilt on the queues in a different
2123 * order than they were in on the previous master. So, granting
2124 * queued conversions in order after recovery doesn't make sense
2125 * since the order hasn't been preserved anyway. The new order
2126 * could also have created a new "in place" conversion deadlock.
2127 * (e.g. old, failed master held granted EX, with PR->EX, NL->EX.
2128 * After recovery, there would be no granted locks, and possibly
2129 * NL->EX, PR->EX, an in-place conversion deadlock.) So, after
2130 * recovery, grant conversions without considering order.
2131 */
2132
2133 if (conv && recover)
2134 return 1;
2135
2136 /*
2137 * 6-5: But the default algorithm for deciding whether to grant or
2138 * queue conversion requests does not by itself guarantee that such
2139 * requests are serviced on a "first come first serve" basis. This, in
2140 * turn, can lead to a phenomenon known as "indefinate postponement".
2141 *
2142 * 6-7: This issue is dealt with by using the optional QUECVT flag with
2143 * the system service employed to request a lock conversion. This flag
2144 * forces certain conversion requests to be queued, even if they are
2145 * compatible with the granted modes of other locks on the same
2146 * resource. Thus, the use of this flag results in conversion requests
2147 * being ordered on a "first come first servce" basis.
2148 *
2149 * DCT: This condition is all about new conversions being able to occur
2150 * "in place" while the lock remains on the granted queue (assuming
2151 * nothing else conflicts.) IOW if QUECVT isn't set, a conversion
2152 * doesn't _have_ to go onto the convert queue where it's processed in
2153 * order. The "now" variable is necessary to distinguish converts
2154 * being received and processed for the first time now, because once a
2155 * convert is moved to the conversion queue the condition below applies
2156 * requiring fifo granting.
2157 */
2158
2159 if (now && conv && !(lkb->lkb_exflags & DLM_LKF_QUECVT))
2160 return 1;
2161
2162 /*
2163 * Even if the convert is compat with all granted locks,
2164 * QUECVT forces it behind other locks on the convert queue.
2165 */
2166
2167 if (now && conv && (lkb->lkb_exflags & DLM_LKF_QUECVT)) {
2168 if (list_empty(&r->res_convertqueue))
2169 return 1;
2170 else
2171 return 0;
2172 }
2173
2174 /*
2175 * The NOORDER flag is set to avoid the standard vms rules on grant
2176 * order.
2177 */
2178
2179 if (lkb->lkb_exflags & DLM_LKF_NOORDER)
2180 return 1;
2181
2182 /*
2183 * 6-3: Once in that queue [CONVERTING], a conversion request cannot be
2184 * granted until all other conversion requests ahead of it are granted
2185 * and/or canceled.
2186 */
2187
2188 if (!now && conv && first_in_list(lkb, &r->res_convertqueue))
2189 return 1;
2190
2191 /*
2192 * 6-4: By default, a new request is immediately granted only if all
2193 * three of the following conditions are satisfied when the request is
2194 * issued:
2195 * - The queue of ungranted conversion requests for the resource is
2196 * empty.
2197 * - The queue of ungranted new requests for the resource is empty.
2198 * - The mode of the new request is compatible with the most
2199 * restrictive mode of all granted locks on the resource.
2200 */
2201
2202 if (now && !conv && list_empty(&r->res_convertqueue) &&
2203 list_empty(&r->res_waitqueue))
2204 return 1;
2205
2206 /*
2207 * 6-4: Once a lock request is in the queue of ungranted new requests,
2208 * it cannot be granted until the queue of ungranted conversion
2209 * requests is empty, all ungranted new requests ahead of it are
2210 * granted and/or canceled, and it is compatible with the granted mode
2211 * of the most restrictive lock granted on the resource.
2212 */
2213
2214 if (!now && !conv && list_empty(&r->res_convertqueue) &&
2215 first_in_list(lkb, &r->res_waitqueue))
2216 return 1;
2217
2218 return 0;
2219}
2220
2221static int can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now,
2222 int recover, int *err)
2223{
2224 int rv;
2225 int8_t alt = 0, rqmode = lkb->lkb_rqmode;
2226 int8_t is_convert = (lkb->lkb_grmode != DLM_LOCK_IV);
2227
2228 if (err)
2229 *err = 0;
2230
2231 rv = _can_be_granted(r, lkb, now, recover);
2232 if (rv)
2233 goto out;
2234
2235 /*
2236 * The CONVDEADLK flag is non-standard and tells the dlm to resolve
2237 * conversion deadlocks by demoting grmode to NL, otherwise the dlm
2238 * cancels one of the locks.
2239 */
2240
2241 if (is_convert && can_be_queued(lkb) &&
2242 conversion_deadlock_detect(r, lkb)) {
2243 if (lkb->lkb_exflags & DLM_LKF_CONVDEADLK) {
2244 lkb->lkb_grmode = DLM_LOCK_NL;
2245 set_bit(DLM_SBF_DEMOTED_BIT, &lkb->lkb_sbflags);
2246 } else if (err) {
2247 *err = -EDEADLK;
2248 } else {
2249 log_print("can_be_granted deadlock %x now %d",
2250 lkb->lkb_id, now);
2251 dlm_dump_rsb(r);
2252 }
2253 goto out;
2254 }
2255
2256 /*
2257 * The ALTPR and ALTCW flags are non-standard and tell the dlm to try
2258 * to grant a request in a mode other than the normal rqmode. It's a
2259 * simple way to provide a big optimization to applications that can
2260 * use them.
2261 */
2262
2263 if (rqmode != DLM_LOCK_PR && (lkb->lkb_exflags & DLM_LKF_ALTPR))
2264 alt = DLM_LOCK_PR;
2265 else if (rqmode != DLM_LOCK_CW && (lkb->lkb_exflags & DLM_LKF_ALTCW))
2266 alt = DLM_LOCK_CW;
2267
2268 if (alt) {
2269 lkb->lkb_rqmode = alt;
2270 rv = _can_be_granted(r, lkb, now, 0);
2271 if (rv)
2272 set_bit(DLM_SBF_ALTMODE_BIT, &lkb->lkb_sbflags);
2273 else
2274 lkb->lkb_rqmode = rqmode;
2275 }
2276 out:
2277 return rv;
2278}
2279
2280/* Returns the highest requested mode of all blocked conversions; sets
2281 cw if there's a blocked conversion to DLM_LOCK_CW. */
2282
2283static int grant_pending_convert(struct dlm_rsb *r, int high, int *cw,
2284 unsigned int *count)
2285{
2286 struct dlm_lkb *lkb, *s;
2287 int recover = rsb_flag(r, RSB_RECOVER_GRANT);
2288 int hi, demoted, quit, grant_restart, demote_restart;
2289 int deadlk;
2290
2291 quit = 0;
2292 restart:
2293 grant_restart = 0;
2294 demote_restart = 0;
2295 hi = DLM_LOCK_IV;
2296
2297 list_for_each_entry_safe(lkb, s, &r->res_convertqueue, lkb_statequeue) {
2298 demoted = is_demoted(lkb);
2299 deadlk = 0;
2300
2301 if (can_be_granted(r, lkb, 0, recover, &deadlk)) {
2302 grant_lock_pending(r, lkb);
2303 grant_restart = 1;
2304 if (count)
2305 (*count)++;
2306 continue;
2307 }
2308
2309 if (!demoted && is_demoted(lkb)) {
2310 log_print("WARN: pending demoted %x node %d %s",
2311 lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
2312 demote_restart = 1;
2313 continue;
2314 }
2315
2316 if (deadlk) {
2317 /*
2318 * If DLM_LKB_NODLKWT flag is set and conversion
2319 * deadlock is detected, we request blocking AST and
2320 * down (or cancel) conversion.
2321 */
2322 if (lkb->lkb_exflags & DLM_LKF_NODLCKWT) {
2323 if (lkb->lkb_highbast < lkb->lkb_rqmode) {
2324 queue_bast(r, lkb, lkb->lkb_rqmode);
2325 lkb->lkb_highbast = lkb->lkb_rqmode;
2326 }
2327 } else {
2328 log_print("WARN: pending deadlock %x node %d %s",
2329 lkb->lkb_id, lkb->lkb_nodeid,
2330 r->res_name);
2331 dlm_dump_rsb(r);
2332 }
2333 continue;
2334 }
2335
2336 hi = max_t(int, lkb->lkb_rqmode, hi);
2337
2338 if (cw && lkb->lkb_rqmode == DLM_LOCK_CW)
2339 *cw = 1;
2340 }
2341
2342 if (grant_restart)
2343 goto restart;
2344 if (demote_restart && !quit) {
2345 quit = 1;
2346 goto restart;
2347 }
2348
2349 return max_t(int, high, hi);
2350}
2351
2352static int grant_pending_wait(struct dlm_rsb *r, int high, int *cw,
2353 unsigned int *count)
2354{
2355 struct dlm_lkb *lkb, *s;
2356
2357 list_for_each_entry_safe(lkb, s, &r->res_waitqueue, lkb_statequeue) {
2358 if (can_be_granted(r, lkb, 0, 0, NULL)) {
2359 grant_lock_pending(r, lkb);
2360 if (count)
2361 (*count)++;
2362 } else {
2363 high = max_t(int, lkb->lkb_rqmode, high);
2364 if (lkb->lkb_rqmode == DLM_LOCK_CW)
2365 *cw = 1;
2366 }
2367 }
2368
2369 return high;
2370}
2371
2372/* cw of 1 means there's a lock with a rqmode of DLM_LOCK_CW that's blocked
2373 on either the convert or waiting queue.
2374 high is the largest rqmode of all locks blocked on the convert or
2375 waiting queue. */
2376
2377static int lock_requires_bast(struct dlm_lkb *gr, int high, int cw)
2378{
2379 if (gr->lkb_grmode == DLM_LOCK_PR && cw) {
2380 if (gr->lkb_highbast < DLM_LOCK_EX)
2381 return 1;
2382 return 0;
2383 }
2384
2385 if (gr->lkb_highbast < high &&
2386 !__dlm_compat_matrix[gr->lkb_grmode+1][high+1])
2387 return 1;
2388 return 0;
2389}
2390
2391static void grant_pending_locks(struct dlm_rsb *r, unsigned int *count)
2392{
2393 struct dlm_lkb *lkb, *s;
2394 int high = DLM_LOCK_IV;
2395 int cw = 0;
2396
2397 if (!is_master(r)) {
2398 log_print("grant_pending_locks r nodeid %d", r->res_nodeid);
2399 dlm_dump_rsb(r);
2400 return;
2401 }
2402
2403 high = grant_pending_convert(r, high, &cw, count);
2404 high = grant_pending_wait(r, high, &cw, count);
2405
2406 if (high == DLM_LOCK_IV)
2407 return;
2408
2409 /*
2410 * If there are locks left on the wait/convert queue then send blocking
2411 * ASTs to granted locks based on the largest requested mode (high)
2412 * found above.
2413 */
2414
2415 list_for_each_entry_safe(lkb, s, &r->res_grantqueue, lkb_statequeue) {
2416 if (lkb->lkb_bastfn && lock_requires_bast(lkb, high, cw)) {
2417 if (cw && high == DLM_LOCK_PR &&
2418 lkb->lkb_grmode == DLM_LOCK_PR)
2419 queue_bast(r, lkb, DLM_LOCK_CW);
2420 else
2421 queue_bast(r, lkb, high);
2422 lkb->lkb_highbast = high;
2423 }
2424 }
2425}
2426
2427static int modes_require_bast(struct dlm_lkb *gr, struct dlm_lkb *rq)
2428{
2429 if ((gr->lkb_grmode == DLM_LOCK_PR && rq->lkb_rqmode == DLM_LOCK_CW) ||
2430 (gr->lkb_grmode == DLM_LOCK_CW && rq->lkb_rqmode == DLM_LOCK_PR)) {
2431 if (gr->lkb_highbast < DLM_LOCK_EX)
2432 return 1;
2433 return 0;
2434 }
2435
2436 if (gr->lkb_highbast < rq->lkb_rqmode && !modes_compat(gr, rq))
2437 return 1;
2438 return 0;
2439}
2440
2441static void send_bast_queue(struct dlm_rsb *r, struct list_head *head,
2442 struct dlm_lkb *lkb)
2443{
2444 struct dlm_lkb *gr;
2445
2446 list_for_each_entry(gr, head, lkb_statequeue) {
2447 /* skip self when sending basts to convertqueue */
2448 if (gr == lkb)
2449 continue;
2450 if (gr->lkb_bastfn && modes_require_bast(gr, lkb)) {
2451 queue_bast(r, gr, lkb->lkb_rqmode);
2452 gr->lkb_highbast = lkb->lkb_rqmode;
2453 }
2454 }
2455}
2456
2457static void send_blocking_asts(struct dlm_rsb *r, struct dlm_lkb *lkb)
2458{
2459 send_bast_queue(r, &r->res_grantqueue, lkb);
2460}
2461
2462static void send_blocking_asts_all(struct dlm_rsb *r, struct dlm_lkb *lkb)
2463{
2464 send_bast_queue(r, &r->res_grantqueue, lkb);
2465 send_bast_queue(r, &r->res_convertqueue, lkb);
2466}
2467
2468/* set_master(r, lkb) -- set the master nodeid of a resource
2469
2470 The purpose of this function is to set the nodeid field in the given
2471 lkb using the nodeid field in the given rsb. If the rsb's nodeid is
2472 known, it can just be copied to the lkb and the function will return
2473 0. If the rsb's nodeid is _not_ known, it needs to be looked up
2474 before it can be copied to the lkb.
2475
2476 When the rsb nodeid is being looked up remotely, the initial lkb
2477 causing the lookup is kept on the ls_waiters list waiting for the
2478 lookup reply. Other lkb's waiting for the same rsb lookup are kept
2479 on the rsb's res_lookup list until the master is verified.
2480
2481 Return values:
2482 0: nodeid is set in rsb/lkb and the caller should go ahead and use it
2483 1: the rsb master is not available and the lkb has been placed on
2484 a wait queue
2485*/
2486
2487static int set_master(struct dlm_rsb *r, struct dlm_lkb *lkb)
2488{
2489 int our_nodeid = dlm_our_nodeid();
2490
2491 if (rsb_flag(r, RSB_MASTER_UNCERTAIN)) {
2492 rsb_clear_flag(r, RSB_MASTER_UNCERTAIN);
2493 r->res_first_lkid = lkb->lkb_id;
2494 lkb->lkb_nodeid = r->res_nodeid;
2495 return 0;
2496 }
2497
2498 if (r->res_first_lkid && r->res_first_lkid != lkb->lkb_id) {
2499 list_add_tail(&lkb->lkb_rsb_lookup, &r->res_lookup);
2500 return 1;
2501 }
2502
2503 if (r->res_master_nodeid == our_nodeid) {
2504 lkb->lkb_nodeid = 0;
2505 return 0;
2506 }
2507
2508 if (r->res_master_nodeid) {
2509 lkb->lkb_nodeid = r->res_master_nodeid;
2510 return 0;
2511 }
2512
2513 if (dlm_dir_nodeid(r) == our_nodeid) {
2514 /* This is a somewhat unusual case; find_rsb will usually
2515 have set res_master_nodeid when dir nodeid is local, but
2516 there are cases where we become the dir node after we've
2517 past find_rsb and go through _request_lock again.
2518 confirm_master() or process_lookup_list() needs to be
2519 called after this. */
2520 log_debug(r->res_ls, "set_master %x self master %d dir %d %s",
2521 lkb->lkb_id, r->res_master_nodeid, r->res_dir_nodeid,
2522 r->res_name);
2523 r->res_master_nodeid = our_nodeid;
2524 r->res_nodeid = 0;
2525 lkb->lkb_nodeid = 0;
2526 return 0;
2527 }
2528
2529 r->res_first_lkid = lkb->lkb_id;
2530 send_lookup(r, lkb);
2531 return 1;
2532}
2533
2534static void process_lookup_list(struct dlm_rsb *r)
2535{
2536 struct dlm_lkb *lkb, *safe;
2537
2538 list_for_each_entry_safe(lkb, safe, &r->res_lookup, lkb_rsb_lookup) {
2539 list_del_init(&lkb->lkb_rsb_lookup);
2540 _request_lock(r, lkb);
2541 schedule();
2542 }
2543}
2544
2545/* confirm_master -- confirm (or deny) an rsb's master nodeid */
2546
2547static void confirm_master(struct dlm_rsb *r, int error)
2548{
2549 struct dlm_lkb *lkb;
2550
2551 if (!r->res_first_lkid)
2552 return;
2553
2554 switch (error) {
2555 case 0:
2556 case -EINPROGRESS:
2557 r->res_first_lkid = 0;
2558 process_lookup_list(r);
2559 break;
2560
2561 case -EAGAIN:
2562 case -EBADR:
2563 case -ENOTBLK:
2564 /* the remote request failed and won't be retried (it was
2565 a NOQUEUE, or has been canceled/unlocked); make a waiting
2566 lkb the first_lkid */
2567
2568 r->res_first_lkid = 0;
2569
2570 if (!list_empty(&r->res_lookup)) {
2571 lkb = list_entry(r->res_lookup.next, struct dlm_lkb,
2572 lkb_rsb_lookup);
2573 list_del_init(&lkb->lkb_rsb_lookup);
2574 r->res_first_lkid = lkb->lkb_id;
2575 _request_lock(r, lkb);
2576 }
2577 break;
2578
2579 default:
2580 log_error(r->res_ls, "confirm_master unknown error %d", error);
2581 }
2582}
2583
2584static int set_lock_args(int mode, struct dlm_lksb *lksb, uint32_t flags,
2585 int namelen, void (*ast)(void *astparam),
2586 void *astparam,
2587 void (*bast)(void *astparam, int mode),
2588 struct dlm_args *args)
2589{
2590 int rv = -EINVAL;
2591
2592 /* check for invalid arg usage */
2593
2594 if (mode < 0 || mode > DLM_LOCK_EX)
2595 goto out;
2596
2597 if (!(flags & DLM_LKF_CONVERT) && (namelen > DLM_RESNAME_MAXLEN))
2598 goto out;
2599
2600 if (flags & DLM_LKF_CANCEL)
2601 goto out;
2602
2603 if (flags & DLM_LKF_QUECVT && !(flags & DLM_LKF_CONVERT))
2604 goto out;
2605
2606 if (flags & DLM_LKF_CONVDEADLK && !(flags & DLM_LKF_CONVERT))
2607 goto out;
2608
2609 if (flags & DLM_LKF_CONVDEADLK && flags & DLM_LKF_NOQUEUE)
2610 goto out;
2611
2612 if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_CONVERT)
2613 goto out;
2614
2615 if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_QUECVT)
2616 goto out;
2617
2618 if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_NOQUEUE)
2619 goto out;
2620
2621 if (flags & DLM_LKF_EXPEDITE && mode != DLM_LOCK_NL)
2622 goto out;
2623
2624 if (!ast || !lksb)
2625 goto out;
2626
2627 if (flags & DLM_LKF_VALBLK && !lksb->sb_lvbptr)
2628 goto out;
2629
2630 if (flags & DLM_LKF_CONVERT && !lksb->sb_lkid)
2631 goto out;
2632
2633 /* these args will be copied to the lkb in validate_lock_args,
2634 it cannot be done now because when converting locks, fields in
2635 an active lkb cannot be modified before locking the rsb */
2636
2637 args->flags = flags;
2638 args->astfn = ast;
2639 args->astparam = astparam;
2640 args->bastfn = bast;
2641 args->mode = mode;
2642 args->lksb = lksb;
2643 rv = 0;
2644 out:
2645 return rv;
2646}
2647
2648static int set_unlock_args(uint32_t flags, void *astarg, struct dlm_args *args)
2649{
2650 if (flags & ~(DLM_LKF_CANCEL | DLM_LKF_VALBLK | DLM_LKF_IVVALBLK |
2651 DLM_LKF_FORCEUNLOCK))
2652 return -EINVAL;
2653
2654 if (flags & DLM_LKF_CANCEL && flags & DLM_LKF_FORCEUNLOCK)
2655 return -EINVAL;
2656
2657 args->flags = flags;
2658 args->astparam = astarg;
2659 return 0;
2660}
2661
2662static int validate_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
2663 struct dlm_args *args)
2664{
2665 int rv = -EBUSY;
2666
2667 if (args->flags & DLM_LKF_CONVERT) {
2668 if (lkb->lkb_status != DLM_LKSTS_GRANTED)
2669 goto out;
2670
2671 /* lock not allowed if there's any op in progress */
2672 if (lkb->lkb_wait_type || lkb->lkb_wait_count)
2673 goto out;
2674
2675 if (is_overlap(lkb))
2676 goto out;
2677
2678 rv = -EINVAL;
2679 if (test_bit(DLM_IFL_MSTCPY_BIT, &lkb->lkb_iflags))
2680 goto out;
2681
2682 if (args->flags & DLM_LKF_QUECVT &&
2683 !__quecvt_compat_matrix[lkb->lkb_grmode+1][args->mode+1])
2684 goto out;
2685 }
2686
2687 lkb->lkb_exflags = args->flags;
2688 dlm_set_sbflags_val(lkb, 0);
2689 lkb->lkb_astfn = args->astfn;
2690 lkb->lkb_astparam = args->astparam;
2691 lkb->lkb_bastfn = args->bastfn;
2692 lkb->lkb_rqmode = args->mode;
2693 lkb->lkb_lksb = args->lksb;
2694 lkb->lkb_lvbptr = args->lksb->sb_lvbptr;
2695 lkb->lkb_ownpid = (int) current->pid;
2696 rv = 0;
2697 out:
2698 switch (rv) {
2699 case 0:
2700 break;
2701 case -EINVAL:
2702 /* annoy the user because dlm usage is wrong */
2703 WARN_ON(1);
2704 log_error(ls, "%s %d %x %x %x %d %d %s", __func__,
2705 rv, lkb->lkb_id, dlm_iflags_val(lkb), args->flags,
2706 lkb->lkb_status, lkb->lkb_wait_type,
2707 lkb->lkb_resource->res_name);
2708 break;
2709 default:
2710 log_debug(ls, "%s %d %x %x %x %d %d %s", __func__,
2711 rv, lkb->lkb_id, dlm_iflags_val(lkb), args->flags,
2712 lkb->lkb_status, lkb->lkb_wait_type,
2713 lkb->lkb_resource->res_name);
2714 break;
2715 }
2716
2717 return rv;
2718}
2719
2720/* when dlm_unlock() sees -EBUSY with CANCEL/FORCEUNLOCK it returns 0
2721 for success */
2722
2723/* note: it's valid for lkb_nodeid/res_nodeid to be -1 when we get here
2724 because there may be a lookup in progress and it's valid to do
2725 cancel/unlockf on it */
2726
2727static int validate_unlock_args(struct dlm_lkb *lkb, struct dlm_args *args)
2728{
2729 struct dlm_ls *ls = lkb->lkb_resource->res_ls;
2730 int rv = -EBUSY;
2731
2732 /* normal unlock not allowed if there's any op in progress */
2733 if (!(args->flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)) &&
2734 (lkb->lkb_wait_type || lkb->lkb_wait_count))
2735 goto out;
2736
2737 /* an lkb may be waiting for an rsb lookup to complete where the
2738 lookup was initiated by another lock */
2739
2740 if (!list_empty(&lkb->lkb_rsb_lookup)) {
2741 if (args->flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)) {
2742 log_debug(ls, "unlock on rsb_lookup %x", lkb->lkb_id);
2743 list_del_init(&lkb->lkb_rsb_lookup);
2744 queue_cast(lkb->lkb_resource, lkb,
2745 args->flags & DLM_LKF_CANCEL ?
2746 -DLM_ECANCEL : -DLM_EUNLOCK);
2747 unhold_lkb(lkb); /* undoes create_lkb() */
2748 }
2749 /* caller changes -EBUSY to 0 for CANCEL and FORCEUNLOCK */
2750 goto out;
2751 }
2752
2753 rv = -EINVAL;
2754 if (test_bit(DLM_IFL_MSTCPY_BIT, &lkb->lkb_iflags)) {
2755 log_error(ls, "unlock on MSTCPY %x", lkb->lkb_id);
2756 dlm_print_lkb(lkb);
2757 goto out;
2758 }
2759
2760 /* an lkb may still exist even though the lock is EOL'ed due to a
2761 * cancel, unlock or failed noqueue request; an app can't use these
2762 * locks; return same error as if the lkid had not been found at all
2763 */
2764
2765 if (test_bit(DLM_IFL_ENDOFLIFE_BIT, &lkb->lkb_iflags)) {
2766 log_debug(ls, "unlock on ENDOFLIFE %x", lkb->lkb_id);
2767 rv = -ENOENT;
2768 goto out;
2769 }
2770
2771 /* cancel not allowed with another cancel/unlock in progress */
2772
2773 if (args->flags & DLM_LKF_CANCEL) {
2774 if (lkb->lkb_exflags & DLM_LKF_CANCEL)
2775 goto out;
2776
2777 if (is_overlap(lkb))
2778 goto out;
2779
2780 if (test_bit(DLM_IFL_RESEND_BIT, &lkb->lkb_iflags)) {
2781 set_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags);
2782 rv = -EBUSY;
2783 goto out;
2784 }
2785
2786 /* there's nothing to cancel */
2787 if (lkb->lkb_status == DLM_LKSTS_GRANTED &&
2788 !lkb->lkb_wait_type) {
2789 rv = -EBUSY;
2790 goto out;
2791 }
2792
2793 switch (lkb->lkb_wait_type) {
2794 case DLM_MSG_LOOKUP:
2795 case DLM_MSG_REQUEST:
2796 set_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags);
2797 rv = -EBUSY;
2798 goto out;
2799 case DLM_MSG_UNLOCK:
2800 case DLM_MSG_CANCEL:
2801 goto out;
2802 }
2803 /* add_to_waiters() will set OVERLAP_CANCEL */
2804 goto out_ok;
2805 }
2806
2807 /* do we need to allow a force-unlock if there's a normal unlock
2808 already in progress? in what conditions could the normal unlock
2809 fail such that we'd want to send a force-unlock to be sure? */
2810
2811 if (args->flags & DLM_LKF_FORCEUNLOCK) {
2812 if (lkb->lkb_exflags & DLM_LKF_FORCEUNLOCK)
2813 goto out;
2814
2815 if (is_overlap_unlock(lkb))
2816 goto out;
2817
2818 if (test_bit(DLM_IFL_RESEND_BIT, &lkb->lkb_iflags)) {
2819 set_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags);
2820 rv = -EBUSY;
2821 goto out;
2822 }
2823
2824 switch (lkb->lkb_wait_type) {
2825 case DLM_MSG_LOOKUP:
2826 case DLM_MSG_REQUEST:
2827 set_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags);
2828 rv = -EBUSY;
2829 goto out;
2830 case DLM_MSG_UNLOCK:
2831 goto out;
2832 }
2833 /* add_to_waiters() will set OVERLAP_UNLOCK */
2834 }
2835
2836 out_ok:
2837 /* an overlapping op shouldn't blow away exflags from other op */
2838 lkb->lkb_exflags |= args->flags;
2839 dlm_set_sbflags_val(lkb, 0);
2840 lkb->lkb_astparam = args->astparam;
2841 rv = 0;
2842 out:
2843 switch (rv) {
2844 case 0:
2845 break;
2846 case -EINVAL:
2847 /* annoy the user because dlm usage is wrong */
2848 WARN_ON(1);
2849 log_error(ls, "%s %d %x %x %x %x %d %s", __func__, rv,
2850 lkb->lkb_id, dlm_iflags_val(lkb), lkb->lkb_exflags,
2851 args->flags, lkb->lkb_wait_type,
2852 lkb->lkb_resource->res_name);
2853 break;
2854 default:
2855 log_debug(ls, "%s %d %x %x %x %x %d %s", __func__, rv,
2856 lkb->lkb_id, dlm_iflags_val(lkb), lkb->lkb_exflags,
2857 args->flags, lkb->lkb_wait_type,
2858 lkb->lkb_resource->res_name);
2859 break;
2860 }
2861
2862 return rv;
2863}
2864
2865/*
2866 * Four stage 4 varieties:
2867 * do_request(), do_convert(), do_unlock(), do_cancel()
2868 * These are called on the master node for the given lock and
2869 * from the central locking logic.
2870 */
2871
2872static int do_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
2873{
2874 int error = 0;
2875
2876 if (can_be_granted(r, lkb, 1, 0, NULL)) {
2877 grant_lock(r, lkb);
2878 queue_cast(r, lkb, 0);
2879 goto out;
2880 }
2881
2882 if (can_be_queued(lkb)) {
2883 error = -EINPROGRESS;
2884 add_lkb(r, lkb, DLM_LKSTS_WAITING);
2885 goto out;
2886 }
2887
2888 error = -EAGAIN;
2889 queue_cast(r, lkb, -EAGAIN);
2890 out:
2891 return error;
2892}
2893
2894static void do_request_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
2895 int error)
2896{
2897 switch (error) {
2898 case -EAGAIN:
2899 if (force_blocking_asts(lkb))
2900 send_blocking_asts_all(r, lkb);
2901 break;
2902 case -EINPROGRESS:
2903 send_blocking_asts(r, lkb);
2904 break;
2905 }
2906}
2907
2908static int do_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
2909{
2910 int error = 0;
2911 int deadlk = 0;
2912
2913 /* changing an existing lock may allow others to be granted */
2914
2915 if (can_be_granted(r, lkb, 1, 0, &deadlk)) {
2916 grant_lock(r, lkb);
2917 queue_cast(r, lkb, 0);
2918 goto out;
2919 }
2920
2921 /* can_be_granted() detected that this lock would block in a conversion
2922 deadlock, so we leave it on the granted queue and return EDEADLK in
2923 the ast for the convert. */
2924
2925 if (deadlk && !(lkb->lkb_exflags & DLM_LKF_NODLCKWT)) {
2926 /* it's left on the granted queue */
2927 revert_lock(r, lkb);
2928 queue_cast(r, lkb, -EDEADLK);
2929 error = -EDEADLK;
2930 goto out;
2931 }
2932
2933 /* is_demoted() means the can_be_granted() above set the grmode
2934 to NL, and left us on the granted queue. This auto-demotion
2935 (due to CONVDEADLK) might mean other locks, and/or this lock, are
2936 now grantable. We have to try to grant other converting locks
2937 before we try again to grant this one. */
2938
2939 if (is_demoted(lkb)) {
2940 grant_pending_convert(r, DLM_LOCK_IV, NULL, NULL);
2941 if (_can_be_granted(r, lkb, 1, 0)) {
2942 grant_lock(r, lkb);
2943 queue_cast(r, lkb, 0);
2944 goto out;
2945 }
2946 /* else fall through and move to convert queue */
2947 }
2948
2949 if (can_be_queued(lkb)) {
2950 error = -EINPROGRESS;
2951 del_lkb(r, lkb);
2952 add_lkb(r, lkb, DLM_LKSTS_CONVERT);
2953 goto out;
2954 }
2955
2956 error = -EAGAIN;
2957 queue_cast(r, lkb, -EAGAIN);
2958 out:
2959 return error;
2960}
2961
2962static void do_convert_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
2963 int error)
2964{
2965 switch (error) {
2966 case 0:
2967 grant_pending_locks(r, NULL);
2968 /* grant_pending_locks also sends basts */
2969 break;
2970 case -EAGAIN:
2971 if (force_blocking_asts(lkb))
2972 send_blocking_asts_all(r, lkb);
2973 break;
2974 case -EINPROGRESS:
2975 send_blocking_asts(r, lkb);
2976 break;
2977 }
2978}
2979
2980static int do_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
2981{
2982 remove_lock(r, lkb);
2983 queue_cast(r, lkb, -DLM_EUNLOCK);
2984 return -DLM_EUNLOCK;
2985}
2986
2987static void do_unlock_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
2988 int error)
2989{
2990 grant_pending_locks(r, NULL);
2991}
2992
2993/* returns: 0 did nothing, -DLM_ECANCEL canceled lock */
2994
2995static int do_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb)
2996{
2997 int error;
2998
2999 error = revert_lock(r, lkb);
3000 if (error) {
3001 queue_cast(r, lkb, -DLM_ECANCEL);
3002 return -DLM_ECANCEL;
3003 }
3004 return 0;
3005}
3006
3007static void do_cancel_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
3008 int error)
3009{
3010 if (error)
3011 grant_pending_locks(r, NULL);
3012}
3013
3014/*
3015 * Four stage 3 varieties:
3016 * _request_lock(), _convert_lock(), _unlock_lock(), _cancel_lock()
3017 */
3018
3019/* add a new lkb to a possibly new rsb, called by requesting process */
3020
3021static int _request_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
3022{
3023 int error;
3024
3025 /* set_master: sets lkb nodeid from r */
3026
3027 error = set_master(r, lkb);
3028 if (error < 0)
3029 goto out;
3030 if (error) {
3031 error = 0;
3032 goto out;
3033 }
3034
3035 if (is_remote(r)) {
3036 /* receive_request() calls do_request() on remote node */
3037 error = send_request(r, lkb);
3038 } else {
3039 error = do_request(r, lkb);
3040 /* for remote locks the request_reply is sent
3041 between do_request and do_request_effects */
3042 do_request_effects(r, lkb, error);
3043 }
3044 out:
3045 return error;
3046}
3047
3048/* change some property of an existing lkb, e.g. mode */
3049
3050static int _convert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
3051{
3052 int error;
3053
3054 if (is_remote(r)) {
3055 /* receive_convert() calls do_convert() on remote node */
3056 error = send_convert(r, lkb);
3057 } else {
3058 error = do_convert(r, lkb);
3059 /* for remote locks the convert_reply is sent
3060 between do_convert and do_convert_effects */
3061 do_convert_effects(r, lkb, error);
3062 }
3063
3064 return error;
3065}
3066
3067/* remove an existing lkb from the granted queue */
3068
3069static int _unlock_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
3070{
3071 int error;
3072
3073 if (is_remote(r)) {
3074 /* receive_unlock() calls do_unlock() on remote node */
3075 error = send_unlock(r, lkb);
3076 } else {
3077 error = do_unlock(r, lkb);
3078 /* for remote locks the unlock_reply is sent
3079 between do_unlock and do_unlock_effects */
3080 do_unlock_effects(r, lkb, error);
3081 }
3082
3083 return error;
3084}
3085
3086/* remove an existing lkb from the convert or wait queue */
3087
3088static int _cancel_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
3089{
3090 int error;
3091
3092 if (is_remote(r)) {
3093 /* receive_cancel() calls do_cancel() on remote node */
3094 error = send_cancel(r, lkb);
3095 } else {
3096 error = do_cancel(r, lkb);
3097 /* for remote locks the cancel_reply is sent
3098 between do_cancel and do_cancel_effects */
3099 do_cancel_effects(r, lkb, error);
3100 }
3101
3102 return error;
3103}
3104
3105/*
3106 * Four stage 2 varieties:
3107 * request_lock(), convert_lock(), unlock_lock(), cancel_lock()
3108 */
3109
3110static int request_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
3111 const void *name, int len,
3112 struct dlm_args *args)
3113{
3114 struct dlm_rsb *r;
3115 int error;
3116
3117 error = validate_lock_args(ls, lkb, args);
3118 if (error)
3119 return error;
3120
3121 error = find_rsb(ls, name, len, 0, R_REQUEST, &r);
3122 if (error)
3123 return error;
3124
3125 lock_rsb(r);
3126
3127 attach_lkb(r, lkb);
3128 lkb->lkb_lksb->sb_lkid = lkb->lkb_id;
3129
3130 error = _request_lock(r, lkb);
3131
3132 unlock_rsb(r);
3133 put_rsb(r);
3134 return error;
3135}
3136
3137static int convert_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
3138 struct dlm_args *args)
3139{
3140 struct dlm_rsb *r;
3141 int error;
3142
3143 r = lkb->lkb_resource;
3144
3145 hold_rsb(r);
3146 lock_rsb(r);
3147
3148 error = validate_lock_args(ls, lkb, args);
3149 if (error)
3150 goto out;
3151
3152 error = _convert_lock(r, lkb);
3153 out:
3154 unlock_rsb(r);
3155 put_rsb(r);
3156 return error;
3157}
3158
3159static int unlock_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
3160 struct dlm_args *args)
3161{
3162 struct dlm_rsb *r;
3163 int error;
3164
3165 r = lkb->lkb_resource;
3166
3167 hold_rsb(r);
3168 lock_rsb(r);
3169
3170 error = validate_unlock_args(lkb, args);
3171 if (error)
3172 goto out;
3173
3174 error = _unlock_lock(r, lkb);
3175 out:
3176 unlock_rsb(r);
3177 put_rsb(r);
3178 return error;
3179}
3180
3181static int cancel_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
3182 struct dlm_args *args)
3183{
3184 struct dlm_rsb *r;
3185 int error;
3186
3187 r = lkb->lkb_resource;
3188
3189 hold_rsb(r);
3190 lock_rsb(r);
3191
3192 error = validate_unlock_args(lkb, args);
3193 if (error)
3194 goto out;
3195
3196 error = _cancel_lock(r, lkb);
3197 out:
3198 unlock_rsb(r);
3199 put_rsb(r);
3200 return error;
3201}
3202
3203/*
3204 * Two stage 1 varieties: dlm_lock() and dlm_unlock()
3205 */
3206
3207int dlm_lock(dlm_lockspace_t *lockspace,
3208 int mode,
3209 struct dlm_lksb *lksb,
3210 uint32_t flags,
3211 const void *name,
3212 unsigned int namelen,
3213 uint32_t parent_lkid,
3214 void (*ast) (void *astarg),
3215 void *astarg,
3216 void (*bast) (void *astarg, int mode))
3217{
3218 struct dlm_ls *ls;
3219 struct dlm_lkb *lkb;
3220 struct dlm_args args;
3221 int error, convert = flags & DLM_LKF_CONVERT;
3222
3223 ls = dlm_find_lockspace_local(lockspace);
3224 if (!ls)
3225 return -EINVAL;
3226
3227 dlm_lock_recovery(ls);
3228
3229 if (convert)
3230 error = find_lkb(ls, lksb->sb_lkid, &lkb);
3231 else
3232 error = create_lkb(ls, &lkb);
3233
3234 if (error)
3235 goto out;
3236
3237 trace_dlm_lock_start(ls, lkb, name, namelen, mode, flags);
3238
3239 error = set_lock_args(mode, lksb, flags, namelen, ast, astarg, bast,
3240 &args);
3241 if (error)
3242 goto out_put;
3243
3244 if (convert)
3245 error = convert_lock(ls, lkb, &args);
3246 else
3247 error = request_lock(ls, lkb, name, namelen, &args);
3248
3249 if (error == -EINPROGRESS)
3250 error = 0;
3251 out_put:
3252 trace_dlm_lock_end(ls, lkb, name, namelen, mode, flags, error, true);
3253
3254 if (convert || error)
3255 __put_lkb(ls, lkb);
3256 if (error == -EAGAIN || error == -EDEADLK)
3257 error = 0;
3258 out:
3259 dlm_unlock_recovery(ls);
3260 dlm_put_lockspace(ls);
3261 return error;
3262}
3263
3264int dlm_unlock(dlm_lockspace_t *lockspace,
3265 uint32_t lkid,
3266 uint32_t flags,
3267 struct dlm_lksb *lksb,
3268 void *astarg)
3269{
3270 struct dlm_ls *ls;
3271 struct dlm_lkb *lkb;
3272 struct dlm_args args;
3273 int error;
3274
3275 ls = dlm_find_lockspace_local(lockspace);
3276 if (!ls)
3277 return -EINVAL;
3278
3279 dlm_lock_recovery(ls);
3280
3281 error = find_lkb(ls, lkid, &lkb);
3282 if (error)
3283 goto out;
3284
3285 trace_dlm_unlock_start(ls, lkb, flags);
3286
3287 error = set_unlock_args(flags, astarg, &args);
3288 if (error)
3289 goto out_put;
3290
3291 if (flags & DLM_LKF_CANCEL)
3292 error = cancel_lock(ls, lkb, &args);
3293 else
3294 error = unlock_lock(ls, lkb, &args);
3295
3296 if (error == -DLM_EUNLOCK || error == -DLM_ECANCEL)
3297 error = 0;
3298 if (error == -EBUSY && (flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)))
3299 error = 0;
3300 out_put:
3301 trace_dlm_unlock_end(ls, lkb, flags, error);
3302
3303 dlm_put_lkb(lkb);
3304 out:
3305 dlm_unlock_recovery(ls);
3306 dlm_put_lockspace(ls);
3307 return error;
3308}
3309
3310/*
3311 * send/receive routines for remote operations and replies
3312 *
3313 * send_args
3314 * send_common
3315 * send_request receive_request
3316 * send_convert receive_convert
3317 * send_unlock receive_unlock
3318 * send_cancel receive_cancel
3319 * send_grant receive_grant
3320 * send_bast receive_bast
3321 * send_lookup receive_lookup
3322 * send_remove receive_remove
3323 *
3324 * send_common_reply
3325 * receive_request_reply send_request_reply
3326 * receive_convert_reply send_convert_reply
3327 * receive_unlock_reply send_unlock_reply
3328 * receive_cancel_reply send_cancel_reply
3329 * receive_lookup_reply send_lookup_reply
3330 */
3331
3332static int _create_message(struct dlm_ls *ls, int mb_len,
3333 int to_nodeid, int mstype,
3334 struct dlm_message **ms_ret,
3335 struct dlm_mhandle **mh_ret,
3336 gfp_t allocation)
3337{
3338 struct dlm_message *ms;
3339 struct dlm_mhandle *mh;
3340 char *mb;
3341
3342 /* get_buffer gives us a message handle (mh) that we need to
3343 pass into midcomms_commit and a message buffer (mb) that we
3344 write our data into */
3345
3346 mh = dlm_midcomms_get_mhandle(to_nodeid, mb_len, allocation, &mb);
3347 if (!mh)
3348 return -ENOBUFS;
3349
3350 ms = (struct dlm_message *) mb;
3351
3352 ms->m_header.h_version = cpu_to_le32(DLM_HEADER_MAJOR | DLM_HEADER_MINOR);
3353 ms->m_header.u.h_lockspace = cpu_to_le32(ls->ls_global_id);
3354 ms->m_header.h_nodeid = cpu_to_le32(dlm_our_nodeid());
3355 ms->m_header.h_length = cpu_to_le16(mb_len);
3356 ms->m_header.h_cmd = DLM_MSG;
3357
3358 ms->m_type = cpu_to_le32(mstype);
3359
3360 *mh_ret = mh;
3361 *ms_ret = ms;
3362 return 0;
3363}
3364
3365static int create_message(struct dlm_rsb *r, struct dlm_lkb *lkb,
3366 int to_nodeid, int mstype,
3367 struct dlm_message **ms_ret,
3368 struct dlm_mhandle **mh_ret,
3369 gfp_t allocation)
3370{
3371 int mb_len = sizeof(struct dlm_message);
3372
3373 switch (mstype) {
3374 case DLM_MSG_REQUEST:
3375 case DLM_MSG_LOOKUP:
3376 case DLM_MSG_REMOVE:
3377 mb_len += r->res_length;
3378 break;
3379 case DLM_MSG_CONVERT:
3380 case DLM_MSG_UNLOCK:
3381 case DLM_MSG_REQUEST_REPLY:
3382 case DLM_MSG_CONVERT_REPLY:
3383 case DLM_MSG_GRANT:
3384 if (lkb && lkb->lkb_lvbptr && (lkb->lkb_exflags & DLM_LKF_VALBLK))
3385 mb_len += r->res_ls->ls_lvblen;
3386 break;
3387 }
3388
3389 return _create_message(r->res_ls, mb_len, to_nodeid, mstype,
3390 ms_ret, mh_ret, allocation);
3391}
3392
3393/* further lowcomms enhancements or alternate implementations may make
3394 the return value from this function useful at some point */
3395
3396static int send_message(struct dlm_mhandle *mh, struct dlm_message *ms,
3397 const void *name, int namelen)
3398{
3399 dlm_midcomms_commit_mhandle(mh, name, namelen);
3400 return 0;
3401}
3402
3403static void send_args(struct dlm_rsb *r, struct dlm_lkb *lkb,
3404 struct dlm_message *ms)
3405{
3406 ms->m_nodeid = cpu_to_le32(lkb->lkb_nodeid);
3407 ms->m_pid = cpu_to_le32(lkb->lkb_ownpid);
3408 ms->m_lkid = cpu_to_le32(lkb->lkb_id);
3409 ms->m_remid = cpu_to_le32(lkb->lkb_remid);
3410 ms->m_exflags = cpu_to_le32(lkb->lkb_exflags);
3411 ms->m_sbflags = cpu_to_le32(dlm_sbflags_val(lkb));
3412 ms->m_flags = cpu_to_le32(dlm_dflags_val(lkb));
3413 ms->m_lvbseq = cpu_to_le32(lkb->lkb_lvbseq);
3414 ms->m_status = cpu_to_le32(lkb->lkb_status);
3415 ms->m_grmode = cpu_to_le32(lkb->lkb_grmode);
3416 ms->m_rqmode = cpu_to_le32(lkb->lkb_rqmode);
3417 ms->m_hash = cpu_to_le32(r->res_hash);
3418
3419 /* m_result and m_bastmode are set from function args,
3420 not from lkb fields */
3421
3422 if (lkb->lkb_bastfn)
3423 ms->m_asts |= cpu_to_le32(DLM_CB_BAST);
3424 if (lkb->lkb_astfn)
3425 ms->m_asts |= cpu_to_le32(DLM_CB_CAST);
3426
3427 /* compare with switch in create_message; send_remove() doesn't
3428 use send_args() */
3429
3430 switch (ms->m_type) {
3431 case cpu_to_le32(DLM_MSG_REQUEST):
3432 case cpu_to_le32(DLM_MSG_LOOKUP):
3433 memcpy(ms->m_extra, r->res_name, r->res_length);
3434 break;
3435 case cpu_to_le32(DLM_MSG_CONVERT):
3436 case cpu_to_le32(DLM_MSG_UNLOCK):
3437 case cpu_to_le32(DLM_MSG_REQUEST_REPLY):
3438 case cpu_to_le32(DLM_MSG_CONVERT_REPLY):
3439 case cpu_to_le32(DLM_MSG_GRANT):
3440 if (!lkb->lkb_lvbptr || !(lkb->lkb_exflags & DLM_LKF_VALBLK))
3441 break;
3442 memcpy(ms->m_extra, lkb->lkb_lvbptr, r->res_ls->ls_lvblen);
3443 break;
3444 }
3445}
3446
3447static int send_common(struct dlm_rsb *r, struct dlm_lkb *lkb, int mstype)
3448{
3449 struct dlm_message *ms;
3450 struct dlm_mhandle *mh;
3451 int to_nodeid, error;
3452
3453 to_nodeid = r->res_nodeid;
3454
3455 error = add_to_waiters(lkb, mstype, to_nodeid);
3456 if (error)
3457 return error;
3458
3459 error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh, GFP_NOFS);
3460 if (error)
3461 goto fail;
3462
3463 send_args(r, lkb, ms);
3464
3465 error = send_message(mh, ms, r->res_name, r->res_length);
3466 if (error)
3467 goto fail;
3468 return 0;
3469
3470 fail:
3471 remove_from_waiters(lkb, msg_reply_type(mstype));
3472 return error;
3473}
3474
3475static int send_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
3476{
3477 return send_common(r, lkb, DLM_MSG_REQUEST);
3478}
3479
3480static int send_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
3481{
3482 int error;
3483
3484 error = send_common(r, lkb, DLM_MSG_CONVERT);
3485
3486 /* down conversions go without a reply from the master */
3487 if (!error && down_conversion(lkb)) {
3488 remove_from_waiters(lkb, DLM_MSG_CONVERT_REPLY);
3489 r->res_ls->ls_local_ms.m_type = cpu_to_le32(DLM_MSG_CONVERT_REPLY);
3490 r->res_ls->ls_local_ms.m_result = 0;
3491 __receive_convert_reply(r, lkb, &r->res_ls->ls_local_ms, true);
3492 }
3493
3494 return error;
3495}
3496
3497/* FIXME: if this lkb is the only lock we hold on the rsb, then set
3498 MASTER_UNCERTAIN to force the next request on the rsb to confirm
3499 that the master is still correct. */
3500
3501static int send_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
3502{
3503 return send_common(r, lkb, DLM_MSG_UNLOCK);
3504}
3505
3506static int send_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb)
3507{
3508 return send_common(r, lkb, DLM_MSG_CANCEL);
3509}
3510
3511static int send_grant(struct dlm_rsb *r, struct dlm_lkb *lkb)
3512{
3513 struct dlm_message *ms;
3514 struct dlm_mhandle *mh;
3515 int to_nodeid, error;
3516
3517 to_nodeid = lkb->lkb_nodeid;
3518
3519 error = create_message(r, lkb, to_nodeid, DLM_MSG_GRANT, &ms, &mh,
3520 GFP_NOFS);
3521 if (error)
3522 goto out;
3523
3524 send_args(r, lkb, ms);
3525
3526 ms->m_result = 0;
3527
3528 error = send_message(mh, ms, r->res_name, r->res_length);
3529 out:
3530 return error;
3531}
3532
3533static int send_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int mode)
3534{
3535 struct dlm_message *ms;
3536 struct dlm_mhandle *mh;
3537 int to_nodeid, error;
3538
3539 to_nodeid = lkb->lkb_nodeid;
3540
3541 error = create_message(r, NULL, to_nodeid, DLM_MSG_BAST, &ms, &mh,
3542 GFP_NOFS);
3543 if (error)
3544 goto out;
3545
3546 send_args(r, lkb, ms);
3547
3548 ms->m_bastmode = cpu_to_le32(mode);
3549
3550 error = send_message(mh, ms, r->res_name, r->res_length);
3551 out:
3552 return error;
3553}
3554
3555static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb)
3556{
3557 struct dlm_message *ms;
3558 struct dlm_mhandle *mh;
3559 int to_nodeid, error;
3560
3561 to_nodeid = dlm_dir_nodeid(r);
3562
3563 error = add_to_waiters(lkb, DLM_MSG_LOOKUP, to_nodeid);
3564 if (error)
3565 return error;
3566
3567 error = create_message(r, NULL, to_nodeid, DLM_MSG_LOOKUP, &ms, &mh,
3568 GFP_NOFS);
3569 if (error)
3570 goto fail;
3571
3572 send_args(r, lkb, ms);
3573
3574 error = send_message(mh, ms, r->res_name, r->res_length);
3575 if (error)
3576 goto fail;
3577 return 0;
3578
3579 fail:
3580 remove_from_waiters(lkb, DLM_MSG_LOOKUP_REPLY);
3581 return error;
3582}
3583
3584static int send_remove(struct dlm_rsb *r)
3585{
3586 struct dlm_message *ms;
3587 struct dlm_mhandle *mh;
3588 int to_nodeid, error;
3589
3590 to_nodeid = dlm_dir_nodeid(r);
3591
3592 error = create_message(r, NULL, to_nodeid, DLM_MSG_REMOVE, &ms, &mh,
3593 GFP_ATOMIC);
3594 if (error)
3595 goto out;
3596
3597 memcpy(ms->m_extra, r->res_name, r->res_length);
3598 ms->m_hash = cpu_to_le32(r->res_hash);
3599
3600 error = send_message(mh, ms, r->res_name, r->res_length);
3601 out:
3602 return error;
3603}
3604
3605static int send_common_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
3606 int mstype, int rv)
3607{
3608 struct dlm_message *ms;
3609 struct dlm_mhandle *mh;
3610 int to_nodeid, error;
3611
3612 to_nodeid = lkb->lkb_nodeid;
3613
3614 error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh, GFP_NOFS);
3615 if (error)
3616 goto out;
3617
3618 send_args(r, lkb, ms);
3619
3620 ms->m_result = cpu_to_le32(to_dlm_errno(rv));
3621
3622 error = send_message(mh, ms, r->res_name, r->res_length);
3623 out:
3624 return error;
3625}
3626
3627static int send_request_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3628{
3629 return send_common_reply(r, lkb, DLM_MSG_REQUEST_REPLY, rv);
3630}
3631
3632static int send_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3633{
3634 return send_common_reply(r, lkb, DLM_MSG_CONVERT_REPLY, rv);
3635}
3636
3637static int send_unlock_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3638{
3639 return send_common_reply(r, lkb, DLM_MSG_UNLOCK_REPLY, rv);
3640}
3641
3642static int send_cancel_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
3643{
3644 return send_common_reply(r, lkb, DLM_MSG_CANCEL_REPLY, rv);
3645}
3646
3647static int send_lookup_reply(struct dlm_ls *ls,
3648 const struct dlm_message *ms_in, int ret_nodeid,
3649 int rv)
3650{
3651 struct dlm_rsb *r = &ls->ls_local_rsb;
3652 struct dlm_message *ms;
3653 struct dlm_mhandle *mh;
3654 int error, nodeid = le32_to_cpu(ms_in->m_header.h_nodeid);
3655
3656 error = create_message(r, NULL, nodeid, DLM_MSG_LOOKUP_REPLY, &ms, &mh,
3657 GFP_NOFS);
3658 if (error)
3659 goto out;
3660
3661 ms->m_lkid = ms_in->m_lkid;
3662 ms->m_result = cpu_to_le32(to_dlm_errno(rv));
3663 ms->m_nodeid = cpu_to_le32(ret_nodeid);
3664
3665 error = send_message(mh, ms, ms_in->m_extra, receive_extralen(ms_in));
3666 out:
3667 return error;
3668}
3669
3670/* which args we save from a received message depends heavily on the type
3671 of message, unlike the send side where we can safely send everything about
3672 the lkb for any type of message */
3673
3674static void receive_flags(struct dlm_lkb *lkb, const struct dlm_message *ms)
3675{
3676 lkb->lkb_exflags = le32_to_cpu(ms->m_exflags);
3677 dlm_set_sbflags_val(lkb, le32_to_cpu(ms->m_sbflags));
3678 dlm_set_dflags_val(lkb, le32_to_cpu(ms->m_flags));
3679}
3680
3681static void receive_flags_reply(struct dlm_lkb *lkb,
3682 const struct dlm_message *ms,
3683 bool local)
3684{
3685 if (local)
3686 return;
3687
3688 dlm_set_sbflags_val(lkb, le32_to_cpu(ms->m_sbflags));
3689 dlm_set_dflags_val(lkb, le32_to_cpu(ms->m_flags));
3690}
3691
3692static int receive_extralen(const struct dlm_message *ms)
3693{
3694 return (le16_to_cpu(ms->m_header.h_length) -
3695 sizeof(struct dlm_message));
3696}
3697
3698static int receive_lvb(struct dlm_ls *ls, struct dlm_lkb *lkb,
3699 const struct dlm_message *ms)
3700{
3701 int len;
3702
3703 if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
3704 if (!lkb->lkb_lvbptr)
3705 lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
3706 if (!lkb->lkb_lvbptr)
3707 return -ENOMEM;
3708 len = receive_extralen(ms);
3709 if (len > ls->ls_lvblen)
3710 len = ls->ls_lvblen;
3711 memcpy(lkb->lkb_lvbptr, ms->m_extra, len);
3712 }
3713 return 0;
3714}
3715
3716static void fake_bastfn(void *astparam, int mode)
3717{
3718 log_print("fake_bastfn should not be called");
3719}
3720
3721static void fake_astfn(void *astparam)
3722{
3723 log_print("fake_astfn should not be called");
3724}
3725
3726static int receive_request_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
3727 const struct dlm_message *ms)
3728{
3729 lkb->lkb_nodeid = le32_to_cpu(ms->m_header.h_nodeid);
3730 lkb->lkb_ownpid = le32_to_cpu(ms->m_pid);
3731 lkb->lkb_remid = le32_to_cpu(ms->m_lkid);
3732 lkb->lkb_grmode = DLM_LOCK_IV;
3733 lkb->lkb_rqmode = le32_to_cpu(ms->m_rqmode);
3734
3735 lkb->lkb_bastfn = (ms->m_asts & cpu_to_le32(DLM_CB_BAST)) ? &fake_bastfn : NULL;
3736 lkb->lkb_astfn = (ms->m_asts & cpu_to_le32(DLM_CB_CAST)) ? &fake_astfn : NULL;
3737
3738 if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
3739 /* lkb was just created so there won't be an lvb yet */
3740 lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
3741 if (!lkb->lkb_lvbptr)
3742 return -ENOMEM;
3743 }
3744
3745 return 0;
3746}
3747
3748static int receive_convert_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
3749 const struct dlm_message *ms)
3750{
3751 if (lkb->lkb_status != DLM_LKSTS_GRANTED)
3752 return -EBUSY;
3753
3754 if (receive_lvb(ls, lkb, ms))
3755 return -ENOMEM;
3756
3757 lkb->lkb_rqmode = le32_to_cpu(ms->m_rqmode);
3758 lkb->lkb_lvbseq = le32_to_cpu(ms->m_lvbseq);
3759
3760 return 0;
3761}
3762
3763static int receive_unlock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
3764 const struct dlm_message *ms)
3765{
3766 if (receive_lvb(ls, lkb, ms))
3767 return -ENOMEM;
3768 return 0;
3769}
3770
3771/* We fill in the local-lkb fields with the info that send_xxxx_reply()
3772 uses to send a reply and that the remote end uses to process the reply. */
3773
3774static void setup_local_lkb(struct dlm_ls *ls, const struct dlm_message *ms)
3775{
3776 struct dlm_lkb *lkb = &ls->ls_local_lkb;
3777 lkb->lkb_nodeid = le32_to_cpu(ms->m_header.h_nodeid);
3778 lkb->lkb_remid = le32_to_cpu(ms->m_lkid);
3779}
3780
3781/* This is called after the rsb is locked so that we can safely inspect
3782 fields in the lkb. */
3783
3784static int validate_message(struct dlm_lkb *lkb, const struct dlm_message *ms)
3785{
3786 int from = le32_to_cpu(ms->m_header.h_nodeid);
3787 int error = 0;
3788
3789 /* currently mixing of user/kernel locks are not supported */
3790 if (ms->m_flags & cpu_to_le32(BIT(DLM_DFL_USER_BIT)) &&
3791 !test_bit(DLM_DFL_USER_BIT, &lkb->lkb_dflags)) {
3792 log_error(lkb->lkb_resource->res_ls,
3793 "got user dlm message for a kernel lock");
3794 error = -EINVAL;
3795 goto out;
3796 }
3797
3798 switch (ms->m_type) {
3799 case cpu_to_le32(DLM_MSG_CONVERT):
3800 case cpu_to_le32(DLM_MSG_UNLOCK):
3801 case cpu_to_le32(DLM_MSG_CANCEL):
3802 if (!is_master_copy(lkb) || lkb->lkb_nodeid != from)
3803 error = -EINVAL;
3804 break;
3805
3806 case cpu_to_le32(DLM_MSG_CONVERT_REPLY):
3807 case cpu_to_le32(DLM_MSG_UNLOCK_REPLY):
3808 case cpu_to_le32(DLM_MSG_CANCEL_REPLY):
3809 case cpu_to_le32(DLM_MSG_GRANT):
3810 case cpu_to_le32(DLM_MSG_BAST):
3811 if (!is_process_copy(lkb) || lkb->lkb_nodeid != from)
3812 error = -EINVAL;
3813 break;
3814
3815 case cpu_to_le32(DLM_MSG_REQUEST_REPLY):
3816 if (!is_process_copy(lkb))
3817 error = -EINVAL;
3818 else if (lkb->lkb_nodeid != -1 && lkb->lkb_nodeid != from)
3819 error = -EINVAL;
3820 break;
3821
3822 default:
3823 error = -EINVAL;
3824 }
3825
3826out:
3827 if (error)
3828 log_error(lkb->lkb_resource->res_ls,
3829 "ignore invalid message %d from %d %x %x %x %d",
3830 le32_to_cpu(ms->m_type), from, lkb->lkb_id,
3831 lkb->lkb_remid, dlm_iflags_val(lkb),
3832 lkb->lkb_nodeid);
3833 return error;
3834}
3835
3836static int receive_request(struct dlm_ls *ls, const struct dlm_message *ms)
3837{
3838 struct dlm_lkb *lkb;
3839 struct dlm_rsb *r;
3840 int from_nodeid;
3841 int error, namelen = 0;
3842
3843 from_nodeid = le32_to_cpu(ms->m_header.h_nodeid);
3844
3845 error = create_lkb(ls, &lkb);
3846 if (error)
3847 goto fail;
3848
3849 receive_flags(lkb, ms);
3850 set_bit(DLM_IFL_MSTCPY_BIT, &lkb->lkb_iflags);
3851 error = receive_request_args(ls, lkb, ms);
3852 if (error) {
3853 __put_lkb(ls, lkb);
3854 goto fail;
3855 }
3856
3857 /* The dir node is the authority on whether we are the master
3858 for this rsb or not, so if the master sends us a request, we should
3859 recreate the rsb if we've destroyed it. This race happens when we
3860 send a remove message to the dir node at the same time that the dir
3861 node sends us a request for the rsb. */
3862
3863 namelen = receive_extralen(ms);
3864
3865 error = find_rsb(ls, ms->m_extra, namelen, from_nodeid,
3866 R_RECEIVE_REQUEST, &r);
3867 if (error) {
3868 __put_lkb(ls, lkb);
3869 goto fail;
3870 }
3871
3872 lock_rsb(r);
3873
3874 if (r->res_master_nodeid != dlm_our_nodeid()) {
3875 error = validate_master_nodeid(ls, r, from_nodeid);
3876 if (error) {
3877 unlock_rsb(r);
3878 put_rsb(r);
3879 __put_lkb(ls, lkb);
3880 goto fail;
3881 }
3882 }
3883
3884 attach_lkb(r, lkb);
3885 error = do_request(r, lkb);
3886 send_request_reply(r, lkb, error);
3887 do_request_effects(r, lkb, error);
3888
3889 unlock_rsb(r);
3890 put_rsb(r);
3891
3892 if (error == -EINPROGRESS)
3893 error = 0;
3894 if (error)
3895 dlm_put_lkb(lkb);
3896 return 0;
3897
3898 fail:
3899 /* TODO: instead of returning ENOTBLK, add the lkb to res_lookup
3900 and do this receive_request again from process_lookup_list once
3901 we get the lookup reply. This would avoid a many repeated
3902 ENOTBLK request failures when the lookup reply designating us
3903 as master is delayed. */
3904
3905 if (error != -ENOTBLK) {
3906 log_limit(ls, "receive_request %x from %d %d",
3907 le32_to_cpu(ms->m_lkid), from_nodeid, error);
3908 }
3909
3910 setup_local_lkb(ls, ms);
3911 send_request_reply(&ls->ls_local_rsb, &ls->ls_local_lkb, error);
3912 return error;
3913}
3914
3915static int receive_convert(struct dlm_ls *ls, const struct dlm_message *ms)
3916{
3917 struct dlm_lkb *lkb;
3918 struct dlm_rsb *r;
3919 int error, reply = 1;
3920
3921 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
3922 if (error)
3923 goto fail;
3924
3925 if (lkb->lkb_remid != le32_to_cpu(ms->m_lkid)) {
3926 log_error(ls, "receive_convert %x remid %x recover_seq %llu "
3927 "remote %d %x", lkb->lkb_id, lkb->lkb_remid,
3928 (unsigned long long)lkb->lkb_recover_seq,
3929 le32_to_cpu(ms->m_header.h_nodeid),
3930 le32_to_cpu(ms->m_lkid));
3931 error = -ENOENT;
3932 dlm_put_lkb(lkb);
3933 goto fail;
3934 }
3935
3936 r = lkb->lkb_resource;
3937
3938 hold_rsb(r);
3939 lock_rsb(r);
3940
3941 error = validate_message(lkb, ms);
3942 if (error)
3943 goto out;
3944
3945 receive_flags(lkb, ms);
3946
3947 error = receive_convert_args(ls, lkb, ms);
3948 if (error) {
3949 send_convert_reply(r, lkb, error);
3950 goto out;
3951 }
3952
3953 reply = !down_conversion(lkb);
3954
3955 error = do_convert(r, lkb);
3956 if (reply)
3957 send_convert_reply(r, lkb, error);
3958 do_convert_effects(r, lkb, error);
3959 out:
3960 unlock_rsb(r);
3961 put_rsb(r);
3962 dlm_put_lkb(lkb);
3963 return 0;
3964
3965 fail:
3966 setup_local_lkb(ls, ms);
3967 send_convert_reply(&ls->ls_local_rsb, &ls->ls_local_lkb, error);
3968 return error;
3969}
3970
3971static int receive_unlock(struct dlm_ls *ls, const struct dlm_message *ms)
3972{
3973 struct dlm_lkb *lkb;
3974 struct dlm_rsb *r;
3975 int error;
3976
3977 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
3978 if (error)
3979 goto fail;
3980
3981 if (lkb->lkb_remid != le32_to_cpu(ms->m_lkid)) {
3982 log_error(ls, "receive_unlock %x remid %x remote %d %x",
3983 lkb->lkb_id, lkb->lkb_remid,
3984 le32_to_cpu(ms->m_header.h_nodeid),
3985 le32_to_cpu(ms->m_lkid));
3986 error = -ENOENT;
3987 dlm_put_lkb(lkb);
3988 goto fail;
3989 }
3990
3991 r = lkb->lkb_resource;
3992
3993 hold_rsb(r);
3994 lock_rsb(r);
3995
3996 error = validate_message(lkb, ms);
3997 if (error)
3998 goto out;
3999
4000 receive_flags(lkb, ms);
4001
4002 error = receive_unlock_args(ls, lkb, ms);
4003 if (error) {
4004 send_unlock_reply(r, lkb, error);
4005 goto out;
4006 }
4007
4008 error = do_unlock(r, lkb);
4009 send_unlock_reply(r, lkb, error);
4010 do_unlock_effects(r, lkb, error);
4011 out:
4012 unlock_rsb(r);
4013 put_rsb(r);
4014 dlm_put_lkb(lkb);
4015 return 0;
4016
4017 fail:
4018 setup_local_lkb(ls, ms);
4019 send_unlock_reply(&ls->ls_local_rsb, &ls->ls_local_lkb, error);
4020 return error;
4021}
4022
4023static int receive_cancel(struct dlm_ls *ls, const struct dlm_message *ms)
4024{
4025 struct dlm_lkb *lkb;
4026 struct dlm_rsb *r;
4027 int error;
4028
4029 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
4030 if (error)
4031 goto fail;
4032
4033 receive_flags(lkb, ms);
4034
4035 r = lkb->lkb_resource;
4036
4037 hold_rsb(r);
4038 lock_rsb(r);
4039
4040 error = validate_message(lkb, ms);
4041 if (error)
4042 goto out;
4043
4044 error = do_cancel(r, lkb);
4045 send_cancel_reply(r, lkb, error);
4046 do_cancel_effects(r, lkb, error);
4047 out:
4048 unlock_rsb(r);
4049 put_rsb(r);
4050 dlm_put_lkb(lkb);
4051 return 0;
4052
4053 fail:
4054 setup_local_lkb(ls, ms);
4055 send_cancel_reply(&ls->ls_local_rsb, &ls->ls_local_lkb, error);
4056 return error;
4057}
4058
4059static int receive_grant(struct dlm_ls *ls, const struct dlm_message *ms)
4060{
4061 struct dlm_lkb *lkb;
4062 struct dlm_rsb *r;
4063 int error;
4064
4065 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
4066 if (error)
4067 return error;
4068
4069 r = lkb->lkb_resource;
4070
4071 hold_rsb(r);
4072 lock_rsb(r);
4073
4074 error = validate_message(lkb, ms);
4075 if (error)
4076 goto out;
4077
4078 receive_flags_reply(lkb, ms, false);
4079 if (is_altmode(lkb))
4080 munge_altmode(lkb, ms);
4081 grant_lock_pc(r, lkb, ms);
4082 queue_cast(r, lkb, 0);
4083 out:
4084 unlock_rsb(r);
4085 put_rsb(r);
4086 dlm_put_lkb(lkb);
4087 return 0;
4088}
4089
4090static int receive_bast(struct dlm_ls *ls, const struct dlm_message *ms)
4091{
4092 struct dlm_lkb *lkb;
4093 struct dlm_rsb *r;
4094 int error;
4095
4096 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
4097 if (error)
4098 return error;
4099
4100 r = lkb->lkb_resource;
4101
4102 hold_rsb(r);
4103 lock_rsb(r);
4104
4105 error = validate_message(lkb, ms);
4106 if (error)
4107 goto out;
4108
4109 queue_bast(r, lkb, le32_to_cpu(ms->m_bastmode));
4110 lkb->lkb_highbast = le32_to_cpu(ms->m_bastmode);
4111 out:
4112 unlock_rsb(r);
4113 put_rsb(r);
4114 dlm_put_lkb(lkb);
4115 return 0;
4116}
4117
4118static void receive_lookup(struct dlm_ls *ls, const struct dlm_message *ms)
4119{
4120 int len, error, ret_nodeid, from_nodeid, our_nodeid;
4121
4122 from_nodeid = le32_to_cpu(ms->m_header.h_nodeid);
4123 our_nodeid = dlm_our_nodeid();
4124
4125 len = receive_extralen(ms);
4126
4127 error = dlm_master_lookup(ls, from_nodeid, ms->m_extra, len, 0,
4128 &ret_nodeid, NULL);
4129
4130 /* Optimization: we're master so treat lookup as a request */
4131 if (!error && ret_nodeid == our_nodeid) {
4132 receive_request(ls, ms);
4133 return;
4134 }
4135 send_lookup_reply(ls, ms, ret_nodeid, error);
4136}
4137
4138static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms)
4139{
4140 char name[DLM_RESNAME_MAXLEN+1];
4141 struct dlm_rsb *r;
4142 uint32_t hash, b;
4143 int rv, len, dir_nodeid, from_nodeid;
4144
4145 from_nodeid = le32_to_cpu(ms->m_header.h_nodeid);
4146
4147 len = receive_extralen(ms);
4148
4149 if (len > DLM_RESNAME_MAXLEN) {
4150 log_error(ls, "receive_remove from %d bad len %d",
4151 from_nodeid, len);
4152 return;
4153 }
4154
4155 dir_nodeid = dlm_hash2nodeid(ls, le32_to_cpu(ms->m_hash));
4156 if (dir_nodeid != dlm_our_nodeid()) {
4157 log_error(ls, "receive_remove from %d bad nodeid %d",
4158 from_nodeid, dir_nodeid);
4159 return;
4160 }
4161
4162 /* Look for name on rsbtbl.toss, if it's there, kill it.
4163 If it's on rsbtbl.keep, it's being used, and we should ignore this
4164 message. This is an expected race between the dir node sending a
4165 request to the master node at the same time as the master node sends
4166 a remove to the dir node. The resolution to that race is for the
4167 dir node to ignore the remove message, and the master node to
4168 recreate the master rsb when it gets a request from the dir node for
4169 an rsb it doesn't have. */
4170
4171 memset(name, 0, sizeof(name));
4172 memcpy(name, ms->m_extra, len);
4173
4174 hash = jhash(name, len, 0);
4175 b = hash & (ls->ls_rsbtbl_size - 1);
4176
4177 spin_lock(&ls->ls_rsbtbl[b].lock);
4178
4179 rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r);
4180 if (rv) {
4181 /* verify the rsb is on keep list per comment above */
4182 rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r);
4183 if (rv) {
4184 /* should not happen */
4185 log_error(ls, "receive_remove from %d not found %s",
4186 from_nodeid, name);
4187 spin_unlock(&ls->ls_rsbtbl[b].lock);
4188 return;
4189 }
4190 if (r->res_master_nodeid != from_nodeid) {
4191 /* should not happen */
4192 log_error(ls, "receive_remove keep from %d master %d",
4193 from_nodeid, r->res_master_nodeid);
4194 dlm_print_rsb(r);
4195 spin_unlock(&ls->ls_rsbtbl[b].lock);
4196 return;
4197 }
4198
4199 log_debug(ls, "receive_remove from %d master %d first %x %s",
4200 from_nodeid, r->res_master_nodeid, r->res_first_lkid,
4201 name);
4202 spin_unlock(&ls->ls_rsbtbl[b].lock);
4203 return;
4204 }
4205
4206 if (r->res_master_nodeid != from_nodeid) {
4207 log_error(ls, "receive_remove toss from %d master %d",
4208 from_nodeid, r->res_master_nodeid);
4209 dlm_print_rsb(r);
4210 spin_unlock(&ls->ls_rsbtbl[b].lock);
4211 return;
4212 }
4213
4214 if (kref_put(&r->res_ref, kill_rsb)) {
4215 rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss);
4216 spin_unlock(&ls->ls_rsbtbl[b].lock);
4217 dlm_free_rsb(r);
4218 } else {
4219 log_error(ls, "receive_remove from %d rsb ref error",
4220 from_nodeid);
4221 dlm_print_rsb(r);
4222 spin_unlock(&ls->ls_rsbtbl[b].lock);
4223 }
4224}
4225
4226static void receive_purge(struct dlm_ls *ls, const struct dlm_message *ms)
4227{
4228 do_purge(ls, le32_to_cpu(ms->m_nodeid), le32_to_cpu(ms->m_pid));
4229}
4230
4231static int receive_request_reply(struct dlm_ls *ls,
4232 const struct dlm_message *ms)
4233{
4234 struct dlm_lkb *lkb;
4235 struct dlm_rsb *r;
4236 int error, mstype, result;
4237 int from_nodeid = le32_to_cpu(ms->m_header.h_nodeid);
4238
4239 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
4240 if (error)
4241 return error;
4242
4243 r = lkb->lkb_resource;
4244 hold_rsb(r);
4245 lock_rsb(r);
4246
4247 error = validate_message(lkb, ms);
4248 if (error)
4249 goto out;
4250
4251 mstype = lkb->lkb_wait_type;
4252 error = remove_from_waiters(lkb, DLM_MSG_REQUEST_REPLY);
4253 if (error) {
4254 log_error(ls, "receive_request_reply %x remote %d %x result %d",
4255 lkb->lkb_id, from_nodeid, le32_to_cpu(ms->m_lkid),
4256 from_dlm_errno(le32_to_cpu(ms->m_result)));
4257 dlm_dump_rsb(r);
4258 goto out;
4259 }
4260
4261 /* Optimization: the dir node was also the master, so it took our
4262 lookup as a request and sent request reply instead of lookup reply */
4263 if (mstype == DLM_MSG_LOOKUP) {
4264 r->res_master_nodeid = from_nodeid;
4265 r->res_nodeid = from_nodeid;
4266 lkb->lkb_nodeid = from_nodeid;
4267 }
4268
4269 /* this is the value returned from do_request() on the master */
4270 result = from_dlm_errno(le32_to_cpu(ms->m_result));
4271
4272 switch (result) {
4273 case -EAGAIN:
4274 /* request would block (be queued) on remote master */
4275 queue_cast(r, lkb, -EAGAIN);
4276 confirm_master(r, -EAGAIN);
4277 unhold_lkb(lkb); /* undoes create_lkb() */
4278 break;
4279
4280 case -EINPROGRESS:
4281 case 0:
4282 /* request was queued or granted on remote master */
4283 receive_flags_reply(lkb, ms, false);
4284 lkb->lkb_remid = le32_to_cpu(ms->m_lkid);
4285 if (is_altmode(lkb))
4286 munge_altmode(lkb, ms);
4287 if (result) {
4288 add_lkb(r, lkb, DLM_LKSTS_WAITING);
4289 } else {
4290 grant_lock_pc(r, lkb, ms);
4291 queue_cast(r, lkb, 0);
4292 }
4293 confirm_master(r, result);
4294 break;
4295
4296 case -EBADR:
4297 case -ENOTBLK:
4298 /* find_rsb failed to find rsb or rsb wasn't master */
4299 log_limit(ls, "receive_request_reply %x from %d %d "
4300 "master %d dir %d first %x %s", lkb->lkb_id,
4301 from_nodeid, result, r->res_master_nodeid,
4302 r->res_dir_nodeid, r->res_first_lkid, r->res_name);
4303
4304 if (r->res_dir_nodeid != dlm_our_nodeid() &&
4305 r->res_master_nodeid != dlm_our_nodeid()) {
4306 /* cause _request_lock->set_master->send_lookup */
4307 r->res_master_nodeid = 0;
4308 r->res_nodeid = -1;
4309 lkb->lkb_nodeid = -1;
4310 }
4311
4312 if (is_overlap(lkb)) {
4313 /* we'll ignore error in cancel/unlock reply */
4314 queue_cast_overlap(r, lkb);
4315 confirm_master(r, result);
4316 unhold_lkb(lkb); /* undoes create_lkb() */
4317 } else {
4318 _request_lock(r, lkb);
4319
4320 if (r->res_master_nodeid == dlm_our_nodeid())
4321 confirm_master(r, 0);
4322 }
4323 break;
4324
4325 default:
4326 log_error(ls, "receive_request_reply %x error %d",
4327 lkb->lkb_id, result);
4328 }
4329
4330 if ((result == 0 || result == -EINPROGRESS) &&
4331 test_and_clear_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags)) {
4332 log_debug(ls, "receive_request_reply %x result %d unlock",
4333 lkb->lkb_id, result);
4334 clear_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags);
4335 send_unlock(r, lkb);
4336 } else if ((result == -EINPROGRESS) &&
4337 test_and_clear_bit(DLM_IFL_OVERLAP_CANCEL_BIT,
4338 &lkb->lkb_iflags)) {
4339 log_debug(ls, "receive_request_reply %x cancel", lkb->lkb_id);
4340 clear_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags);
4341 send_cancel(r, lkb);
4342 } else {
4343 clear_bit(DLM_IFL_OVERLAP_CANCEL_BIT, &lkb->lkb_iflags);
4344 clear_bit(DLM_IFL_OVERLAP_UNLOCK_BIT, &lkb->lkb_iflags);
4345 }
4346 out:
4347 unlock_rsb(r);
4348 put_rsb(r);
4349 dlm_put_lkb(lkb);
4350 return 0;
4351}
4352
4353static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
4354 const struct dlm_message *ms, bool local)
4355{
4356 /* this is the value returned from do_convert() on the master */
4357 switch (from_dlm_errno(le32_to_cpu(ms->m_result))) {
4358 case -EAGAIN:
4359 /* convert would block (be queued) on remote master */
4360 queue_cast(r, lkb, -EAGAIN);
4361 break;
4362
4363 case -EDEADLK:
4364 receive_flags_reply(lkb, ms, local);
4365 revert_lock_pc(r, lkb);
4366 queue_cast(r, lkb, -EDEADLK);
4367 break;
4368
4369 case -EINPROGRESS:
4370 /* convert was queued on remote master */
4371 receive_flags_reply(lkb, ms, local);
4372 if (is_demoted(lkb))
4373 munge_demoted(lkb);
4374 del_lkb(r, lkb);
4375 add_lkb(r, lkb, DLM_LKSTS_CONVERT);
4376 break;
4377
4378 case 0:
4379 /* convert was granted on remote master */
4380 receive_flags_reply(lkb, ms, local);
4381 if (is_demoted(lkb))
4382 munge_demoted(lkb);
4383 grant_lock_pc(r, lkb, ms);
4384 queue_cast(r, lkb, 0);
4385 break;
4386
4387 default:
4388 log_error(r->res_ls, "receive_convert_reply %x remote %d %x %d",
4389 lkb->lkb_id, le32_to_cpu(ms->m_header.h_nodeid),
4390 le32_to_cpu(ms->m_lkid),
4391 from_dlm_errno(le32_to_cpu(ms->m_result)));
4392 dlm_print_rsb(r);
4393 dlm_print_lkb(lkb);
4394 }
4395}
4396
4397static void _receive_convert_reply(struct dlm_lkb *lkb,
4398 const struct dlm_message *ms, bool local)
4399{
4400 struct dlm_rsb *r = lkb->lkb_resource;
4401 int error;
4402
4403 hold_rsb(r);
4404 lock_rsb(r);
4405
4406 error = validate_message(lkb, ms);
4407 if (error)
4408 goto out;
4409
4410 /* local reply can happen with waiters_mutex held */
4411 error = remove_from_waiters_ms(lkb, ms, local);
4412 if (error)
4413 goto out;
4414
4415 __receive_convert_reply(r, lkb, ms, local);
4416 out:
4417 unlock_rsb(r);
4418 put_rsb(r);
4419}
4420
4421static int receive_convert_reply(struct dlm_ls *ls,
4422 const struct dlm_message *ms)
4423{
4424 struct dlm_lkb *lkb;
4425 int error;
4426
4427 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
4428 if (error)
4429 return error;
4430
4431 _receive_convert_reply(lkb, ms, false);
4432 dlm_put_lkb(lkb);
4433 return 0;
4434}
4435
4436static void _receive_unlock_reply(struct dlm_lkb *lkb,
4437 const struct dlm_message *ms, bool local)
4438{
4439 struct dlm_rsb *r = lkb->lkb_resource;
4440 int error;
4441
4442 hold_rsb(r);
4443 lock_rsb(r);
4444
4445 error = validate_message(lkb, ms);
4446 if (error)
4447 goto out;
4448
4449 /* local reply can happen with waiters_mutex held */
4450 error = remove_from_waiters_ms(lkb, ms, local);
4451 if (error)
4452 goto out;
4453
4454 /* this is the value returned from do_unlock() on the master */
4455
4456 switch (from_dlm_errno(le32_to_cpu(ms->m_result))) {
4457 case -DLM_EUNLOCK:
4458 receive_flags_reply(lkb, ms, local);
4459 remove_lock_pc(r, lkb);
4460 queue_cast(r, lkb, -DLM_EUNLOCK);
4461 break;
4462 case -ENOENT:
4463 break;
4464 default:
4465 log_error(r->res_ls, "receive_unlock_reply %x error %d",
4466 lkb->lkb_id, from_dlm_errno(le32_to_cpu(ms->m_result)));
4467 }
4468 out:
4469 unlock_rsb(r);
4470 put_rsb(r);
4471}
4472
4473static int receive_unlock_reply(struct dlm_ls *ls,
4474 const struct dlm_message *ms)
4475{
4476 struct dlm_lkb *lkb;
4477 int error;
4478
4479 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
4480 if (error)
4481 return error;
4482
4483 _receive_unlock_reply(lkb, ms, false);
4484 dlm_put_lkb(lkb);
4485 return 0;
4486}
4487
4488static void _receive_cancel_reply(struct dlm_lkb *lkb,
4489 const struct dlm_message *ms, bool local)
4490{
4491 struct dlm_rsb *r = lkb->lkb_resource;
4492 int error;
4493
4494 hold_rsb(r);
4495 lock_rsb(r);
4496
4497 error = validate_message(lkb, ms);
4498 if (error)
4499 goto out;
4500
4501 /* local reply can happen with waiters_mutex held */
4502 error = remove_from_waiters_ms(lkb, ms, local);
4503 if (error)
4504 goto out;
4505
4506 /* this is the value returned from do_cancel() on the master */
4507
4508 switch (from_dlm_errno(le32_to_cpu(ms->m_result))) {
4509 case -DLM_ECANCEL:
4510 receive_flags_reply(lkb, ms, local);
4511 revert_lock_pc(r, lkb);
4512 queue_cast(r, lkb, -DLM_ECANCEL);
4513 break;
4514 case 0:
4515 break;
4516 default:
4517 log_error(r->res_ls, "receive_cancel_reply %x error %d",
4518 lkb->lkb_id,
4519 from_dlm_errno(le32_to_cpu(ms->m_result)));
4520 }
4521 out:
4522 unlock_rsb(r);
4523 put_rsb(r);
4524}
4525
4526static int receive_cancel_reply(struct dlm_ls *ls,
4527 const struct dlm_message *ms)
4528{
4529 struct dlm_lkb *lkb;
4530 int error;
4531
4532 error = find_lkb(ls, le32_to_cpu(ms->m_remid), &lkb);
4533 if (error)
4534 return error;
4535
4536 _receive_cancel_reply(lkb, ms, false);
4537 dlm_put_lkb(lkb);
4538 return 0;
4539}
4540
4541static void receive_lookup_reply(struct dlm_ls *ls,
4542 const struct dlm_message *ms)
4543{
4544 struct dlm_lkb *lkb;
4545 struct dlm_rsb *r;
4546 int error, ret_nodeid;
4547 int do_lookup_list = 0;
4548
4549 error = find_lkb(ls, le32_to_cpu(ms->m_lkid), &lkb);
4550 if (error) {
4551 log_error(ls, "%s no lkid %x", __func__,
4552 le32_to_cpu(ms->m_lkid));
4553 return;
4554 }
4555
4556 /* ms->m_result is the value returned by dlm_master_lookup on dir node
4557 FIXME: will a non-zero error ever be returned? */
4558
4559 r = lkb->lkb_resource;
4560 hold_rsb(r);
4561 lock_rsb(r);
4562
4563 error = remove_from_waiters(lkb, DLM_MSG_LOOKUP_REPLY);
4564 if (error)
4565 goto out;
4566
4567 ret_nodeid = le32_to_cpu(ms->m_nodeid);
4568
4569 /* We sometimes receive a request from the dir node for this
4570 rsb before we've received the dir node's loookup_reply for it.
4571 The request from the dir node implies we're the master, so we set
4572 ourself as master in receive_request_reply, and verify here that
4573 we are indeed the master. */
4574
4575 if (r->res_master_nodeid && (r->res_master_nodeid != ret_nodeid)) {
4576 /* This should never happen */
4577 log_error(ls, "receive_lookup_reply %x from %d ret %d "
4578 "master %d dir %d our %d first %x %s",
4579 lkb->lkb_id, le32_to_cpu(ms->m_header.h_nodeid),
4580 ret_nodeid, r->res_master_nodeid, r->res_dir_nodeid,
4581 dlm_our_nodeid(), r->res_first_lkid, r->res_name);
4582 }
4583
4584 if (ret_nodeid == dlm_our_nodeid()) {
4585 r->res_master_nodeid = ret_nodeid;
4586 r->res_nodeid = 0;
4587 do_lookup_list = 1;
4588 r->res_first_lkid = 0;
4589 } else if (ret_nodeid == -1) {
4590 /* the remote node doesn't believe it's the dir node */
4591 log_error(ls, "receive_lookup_reply %x from %d bad ret_nodeid",
4592 lkb->lkb_id, le32_to_cpu(ms->m_header.h_nodeid));
4593 r->res_master_nodeid = 0;
4594 r->res_nodeid = -1;
4595 lkb->lkb_nodeid = -1;
4596 } else {
4597 /* set_master() will set lkb_nodeid from r */
4598 r->res_master_nodeid = ret_nodeid;
4599 r->res_nodeid = ret_nodeid;
4600 }
4601
4602 if (is_overlap(lkb)) {
4603 log_debug(ls, "receive_lookup_reply %x unlock %x",
4604 lkb->lkb_id, dlm_iflags_val(lkb));
4605 queue_cast_overlap(r, lkb);
4606 unhold_lkb(lkb); /* undoes create_lkb() */
4607 goto out_list;
4608 }
4609
4610 _request_lock(r, lkb);
4611
4612 out_list:
4613 if (do_lookup_list)
4614 process_lookup_list(r);
4615 out:
4616 unlock_rsb(r);
4617 put_rsb(r);
4618 dlm_put_lkb(lkb);
4619}
4620
4621static void _receive_message(struct dlm_ls *ls, const struct dlm_message *ms,
4622 uint32_t saved_seq)
4623{
4624 int error = 0, noent = 0;
4625
4626 if (WARN_ON_ONCE(!dlm_is_member(ls, le32_to_cpu(ms->m_header.h_nodeid)))) {
4627 log_limit(ls, "receive %d from non-member %d %x %x %d",
4628 le32_to_cpu(ms->m_type),
4629 le32_to_cpu(ms->m_header.h_nodeid),
4630 le32_to_cpu(ms->m_lkid), le32_to_cpu(ms->m_remid),
4631 from_dlm_errno(le32_to_cpu(ms->m_result)));
4632 return;
4633 }
4634
4635 switch (ms->m_type) {
4636
4637 /* messages sent to a master node */
4638
4639 case cpu_to_le32(DLM_MSG_REQUEST):
4640 error = receive_request(ls, ms);
4641 break;
4642
4643 case cpu_to_le32(DLM_MSG_CONVERT):
4644 error = receive_convert(ls, ms);
4645 break;
4646
4647 case cpu_to_le32(DLM_MSG_UNLOCK):
4648 error = receive_unlock(ls, ms);
4649 break;
4650
4651 case cpu_to_le32(DLM_MSG_CANCEL):
4652 noent = 1;
4653 error = receive_cancel(ls, ms);
4654 break;
4655
4656 /* messages sent from a master node (replies to above) */
4657
4658 case cpu_to_le32(DLM_MSG_REQUEST_REPLY):
4659 error = receive_request_reply(ls, ms);
4660 break;
4661
4662 case cpu_to_le32(DLM_MSG_CONVERT_REPLY):
4663 error = receive_convert_reply(ls, ms);
4664 break;
4665
4666 case cpu_to_le32(DLM_MSG_UNLOCK_REPLY):
4667 error = receive_unlock_reply(ls, ms);
4668 break;
4669
4670 case cpu_to_le32(DLM_MSG_CANCEL_REPLY):
4671 error = receive_cancel_reply(ls, ms);
4672 break;
4673
4674 /* messages sent from a master node (only two types of async msg) */
4675
4676 case cpu_to_le32(DLM_MSG_GRANT):
4677 noent = 1;
4678 error = receive_grant(ls, ms);
4679 break;
4680
4681 case cpu_to_le32(DLM_MSG_BAST):
4682 noent = 1;
4683 error = receive_bast(ls, ms);
4684 break;
4685
4686 /* messages sent to a dir node */
4687
4688 case cpu_to_le32(DLM_MSG_LOOKUP):
4689 receive_lookup(ls, ms);
4690 break;
4691
4692 case cpu_to_le32(DLM_MSG_REMOVE):
4693 receive_remove(ls, ms);
4694 break;
4695
4696 /* messages sent from a dir node (remove has no reply) */
4697
4698 case cpu_to_le32(DLM_MSG_LOOKUP_REPLY):
4699 receive_lookup_reply(ls, ms);
4700 break;
4701
4702 /* other messages */
4703
4704 case cpu_to_le32(DLM_MSG_PURGE):
4705 receive_purge(ls, ms);
4706 break;
4707
4708 default:
4709 log_error(ls, "unknown message type %d",
4710 le32_to_cpu(ms->m_type));
4711 }
4712
4713 /*
4714 * When checking for ENOENT, we're checking the result of
4715 * find_lkb(m_remid):
4716 *
4717 * The lock id referenced in the message wasn't found. This may
4718 * happen in normal usage for the async messages and cancel, so
4719 * only use log_debug for them.
4720 *
4721 * Some errors are expected and normal.
4722 */
4723
4724 if (error == -ENOENT && noent) {
4725 log_debug(ls, "receive %d no %x remote %d %x saved_seq %u",
4726 le32_to_cpu(ms->m_type), le32_to_cpu(ms->m_remid),
4727 le32_to_cpu(ms->m_header.h_nodeid),
4728 le32_to_cpu(ms->m_lkid), saved_seq);
4729 } else if (error == -ENOENT) {
4730 log_error(ls, "receive %d no %x remote %d %x saved_seq %u",
4731 le32_to_cpu(ms->m_type), le32_to_cpu(ms->m_remid),
4732 le32_to_cpu(ms->m_header.h_nodeid),
4733 le32_to_cpu(ms->m_lkid), saved_seq);
4734
4735 if (ms->m_type == cpu_to_le32(DLM_MSG_CONVERT))
4736 dlm_dump_rsb_hash(ls, le32_to_cpu(ms->m_hash));
4737 }
4738
4739 if (error == -EINVAL) {
4740 log_error(ls, "receive %d inval from %d lkid %x remid %x "
4741 "saved_seq %u",
4742 le32_to_cpu(ms->m_type),
4743 le32_to_cpu(ms->m_header.h_nodeid),
4744 le32_to_cpu(ms->m_lkid), le32_to_cpu(ms->m_remid),
4745 saved_seq);
4746 }
4747}
4748
4749/* If the lockspace is in recovery mode (locking stopped), then normal
4750 messages are saved on the requestqueue for processing after recovery is
4751 done. When not in recovery mode, we wait for dlm_recoverd to drain saved
4752 messages off the requestqueue before we process new ones. This occurs right
4753 after recovery completes when we transition from saving all messages on
4754 requestqueue, to processing all the saved messages, to processing new
4755 messages as they arrive. */
4756
4757static void dlm_receive_message(struct dlm_ls *ls, const struct dlm_message *ms,
4758 int nodeid)
4759{
4760 if (dlm_locking_stopped(ls)) {
4761 /* If we were a member of this lockspace, left, and rejoined,
4762 other nodes may still be sending us messages from the
4763 lockspace generation before we left. */
4764 if (WARN_ON_ONCE(!ls->ls_generation)) {
4765 log_limit(ls, "receive %d from %d ignore old gen",
4766 le32_to_cpu(ms->m_type), nodeid);
4767 return;
4768 }
4769
4770 dlm_add_requestqueue(ls, nodeid, ms);
4771 } else {
4772 dlm_wait_requestqueue(ls);
4773 _receive_message(ls, ms, 0);
4774 }
4775}
4776
4777/* This is called by dlm_recoverd to process messages that were saved on
4778 the requestqueue. */
4779
4780void dlm_receive_message_saved(struct dlm_ls *ls, const struct dlm_message *ms,
4781 uint32_t saved_seq)
4782{
4783 _receive_message(ls, ms, saved_seq);
4784}
4785
4786/* This is called by the midcomms layer when something is received for
4787 the lockspace. It could be either a MSG (normal message sent as part of
4788 standard locking activity) or an RCOM (recovery message sent as part of
4789 lockspace recovery). */
4790
4791void dlm_receive_buffer(const union dlm_packet *p, int nodeid)
4792{
4793 const struct dlm_header *hd = &p->header;
4794 struct dlm_ls *ls;
4795 int type = 0;
4796
4797 switch (hd->h_cmd) {
4798 case DLM_MSG:
4799 type = le32_to_cpu(p->message.m_type);
4800 break;
4801 case DLM_RCOM:
4802 type = le32_to_cpu(p->rcom.rc_type);
4803 break;
4804 default:
4805 log_print("invalid h_cmd %d from %u", hd->h_cmd, nodeid);
4806 return;
4807 }
4808
4809 if (le32_to_cpu(hd->h_nodeid) != nodeid) {
4810 log_print("invalid h_nodeid %d from %d lockspace %x",
4811 le32_to_cpu(hd->h_nodeid), nodeid,
4812 le32_to_cpu(hd->u.h_lockspace));
4813 return;
4814 }
4815
4816 ls = dlm_find_lockspace_global(le32_to_cpu(hd->u.h_lockspace));
4817 if (!ls) {
4818 if (dlm_config.ci_log_debug) {
4819 printk_ratelimited(KERN_DEBUG "dlm: invalid lockspace "
4820 "%u from %d cmd %d type %d\n",
4821 le32_to_cpu(hd->u.h_lockspace), nodeid,
4822 hd->h_cmd, type);
4823 }
4824
4825 if (hd->h_cmd == DLM_RCOM && type == DLM_RCOM_STATUS)
4826 dlm_send_ls_not_ready(nodeid, &p->rcom);
4827 return;
4828 }
4829
4830 /* this rwsem allows dlm_ls_stop() to wait for all dlm_recv threads to
4831 be inactive (in this ls) before transitioning to recovery mode */
4832
4833 down_read(&ls->ls_recv_active);
4834 if (hd->h_cmd == DLM_MSG)
4835 dlm_receive_message(ls, &p->message, nodeid);
4836 else if (hd->h_cmd == DLM_RCOM)
4837 dlm_receive_rcom(ls, &p->rcom, nodeid);
4838 else
4839 log_error(ls, "invalid h_cmd %d from %d lockspace %x",
4840 hd->h_cmd, nodeid, le32_to_cpu(hd->u.h_lockspace));
4841 up_read(&ls->ls_recv_active);
4842
4843 dlm_put_lockspace(ls);
4844}
4845
4846static void recover_convert_waiter(struct dlm_ls *ls, struct dlm_lkb *lkb,
4847 struct dlm_message *ms_local)
4848{
4849 if (middle_conversion(lkb)) {
4850 hold_lkb(lkb);
4851 memset(ms_local, 0, sizeof(struct dlm_message));
4852 ms_local->m_type = cpu_to_le32(DLM_MSG_CONVERT_REPLY);
4853 ms_local->m_result = cpu_to_le32(to_dlm_errno(-EINPROGRESS));
4854 ms_local->m_header.h_nodeid = cpu_to_le32(lkb->lkb_nodeid);
4855 _receive_convert_reply(lkb, ms_local, true);
4856
4857 /* Same special case as in receive_rcom_lock_args() */
4858 lkb->lkb_grmode = DLM_LOCK_IV;
4859 rsb_set_flag(lkb->lkb_resource, RSB_RECOVER_CONVERT);
4860 unhold_lkb(lkb);
4861
4862 } else if (lkb->lkb_rqmode >= lkb->lkb_grmode) {
4863 set_bit(DLM_IFL_RESEND_BIT, &lkb->lkb_iflags);
4864 }
4865
4866 /* lkb->lkb_rqmode < lkb->lkb_grmode shouldn't happen since down
4867 conversions are async; there's no reply from the remote master */
4868}
4869
4870/* A waiting lkb needs recovery if the master node has failed, or
4871 the master node is changing (only when no directory is used) */
4872
4873static int waiter_needs_recovery(struct dlm_ls *ls, struct dlm_lkb *lkb,
4874 int dir_nodeid)
4875{
4876 if (dlm_no_directory(ls))
4877 return 1;
4878
4879 if (dlm_is_removed(ls, lkb->lkb_wait_nodeid))
4880 return 1;
4881
4882 return 0;
4883}
4884
4885/* Recovery for locks that are waiting for replies from nodes that are now
4886 gone. We can just complete unlocks and cancels by faking a reply from the
4887 dead node. Requests and up-conversions we flag to be resent after
4888 recovery. Down-conversions can just be completed with a fake reply like
4889 unlocks. Conversions between PR and CW need special attention. */
4890
4891void dlm_recover_waiters_pre(struct dlm_ls *ls)
4892{
4893 struct dlm_lkb *lkb, *safe;
4894 struct dlm_message *ms_local;
4895 int wait_type, local_unlock_result, local_cancel_result;
4896 int dir_nodeid;
4897
4898 ms_local = kmalloc(sizeof(*ms_local), GFP_KERNEL);
4899 if (!ms_local)
4900 return;
4901
4902 mutex_lock(&ls->ls_waiters_mutex);
4903
4904 list_for_each_entry_safe(lkb, safe, &ls->ls_waiters, lkb_wait_reply) {
4905
4906 dir_nodeid = dlm_dir_nodeid(lkb->lkb_resource);
4907
4908 /* exclude debug messages about unlocks because there can be so
4909 many and they aren't very interesting */
4910
4911 if (lkb->lkb_wait_type != DLM_MSG_UNLOCK) {
4912 log_debug(ls, "waiter %x remote %x msg %d r_nodeid %d "
4913 "lkb_nodeid %d wait_nodeid %d dir_nodeid %d",
4914 lkb->lkb_id,
4915 lkb->lkb_remid,
4916 lkb->lkb_wait_type,
4917 lkb->lkb_resource->res_nodeid,
4918 lkb->lkb_nodeid,
4919 lkb->lkb_wait_nodeid,
4920 dir_nodeid);
4921 }
4922
4923 /* all outstanding lookups, regardless of destination will be
4924 resent after recovery is done */
4925
4926 if (lkb->lkb_wait_type == DLM_MSG_LOOKUP) {
4927 set_bit(DLM_IFL_RESEND_BIT, &lkb->lkb_iflags);
4928 continue;
4929 }
4930
4931 if (!waiter_needs_recovery(ls, lkb, dir_nodeid))
4932 continue;
4933
4934 wait_type = lkb->lkb_wait_type;
4935 local_unlock_result = -DLM_EUNLOCK;
4936 local_cancel_result = -DLM_ECANCEL;
4937
4938 /* Main reply may have been received leaving a zero wait_type,
4939 but a reply for the overlapping op may not have been
4940 received. In that case we need to fake the appropriate
4941 reply for the overlap op. */
4942
4943 if (!wait_type) {
4944 if (is_overlap_cancel(lkb)) {
4945 wait_type = DLM_MSG_CANCEL;
4946 if (lkb->lkb_grmode == DLM_LOCK_IV)
4947 local_cancel_result = 0;
4948 }
4949 if (is_overlap_unlock(lkb)) {
4950 wait_type = DLM_MSG_UNLOCK;
4951 if (lkb->lkb_grmode == DLM_LOCK_IV)
4952 local_unlock_result = -ENOENT;
4953 }
4954
4955 log_debug(ls, "rwpre overlap %x %x %d %d %d",
4956 lkb->lkb_id, dlm_iflags_val(lkb), wait_type,
4957 local_cancel_result, local_unlock_result);
4958 }
4959
4960 switch (wait_type) {
4961
4962 case DLM_MSG_REQUEST:
4963 set_bit(DLM_IFL_RESEND_BIT, &lkb->lkb_iflags);
4964 break;
4965
4966 case DLM_MSG_CONVERT:
4967 recover_convert_waiter(ls, lkb, ms_local);
4968 break;
4969
4970 case DLM_MSG_UNLOCK:
4971 hold_lkb(lkb);
4972 memset(ms_local, 0, sizeof(struct dlm_message));
4973 ms_local->m_type = cpu_to_le32(DLM_MSG_UNLOCK_REPLY);
4974 ms_local->m_result = cpu_to_le32(to_dlm_errno(local_unlock_result));
4975 ms_local->m_header.h_nodeid = cpu_to_le32(lkb->lkb_nodeid);
4976 _receive_unlock_reply(lkb, ms_local, true);
4977 dlm_put_lkb(lkb);
4978 break;
4979
4980 case DLM_MSG_CANCEL:
4981 hold_lkb(lkb);
4982 memset(ms_local, 0, sizeof(struct dlm_message));
4983 ms_local->m_type = cpu_to_le32(DLM_MSG_CANCEL_REPLY);
4984 ms_local->m_result = cpu_to_le32(to_dlm_errno(local_cancel_result));
4985 ms_local->m_header.h_nodeid = cpu_to_le32(lkb->lkb_nodeid);
4986 _receive_cancel_reply(lkb, ms_local, true);
4987 dlm_put_lkb(lkb);
4988 break;
4989
4990 default:
4991 log_error(ls, "invalid lkb wait_type %d %d",
4992 lkb->lkb_wait_type, wait_type);
4993 }
4994 schedule();
4995 }
4996 mutex_unlock(&ls->ls_waiters_mutex);
4997 kfree(ms_local);
4998}
4999
5000static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls)
5001{
5002 struct dlm_lkb *lkb = NULL, *iter;
5003
5004 mutex_lock(&ls->ls_waiters_mutex);
5005 list_for_each_entry(iter, &ls->ls_waiters, lkb_wait_reply) {
5006 if (test_bit(DLM_IFL_RESEND_BIT, &iter->lkb_iflags)) {
5007 hold_lkb(iter);
5008 lkb = iter;
5009 break;
5010 }
5011 }
5012 mutex_unlock(&ls->ls_waiters_mutex);
5013
5014 return lkb;
5015}
5016
5017/*
5018 * Forced state reset for locks that were in the middle of remote operations
5019 * when recovery happened (i.e. lkbs that were on the waiters list, waiting
5020 * for a reply from a remote operation.) The lkbs remaining on the waiters
5021 * list need to be reevaluated; some may need resending to a different node
5022 * than previously, and some may now need local handling rather than remote.
5023 *
5024 * First, the lkb state for the voided remote operation is forcibly reset,
5025 * equivalent to what remove_from_waiters() would normally do:
5026 * . lkb removed from ls_waiters list
5027 * . lkb wait_type cleared
5028 * . lkb waiters_count cleared
5029 * . lkb ref count decremented for each waiters_count (almost always 1,
5030 * but possibly 2 in case of cancel/unlock overlapping, which means
5031 * two remote replies were being expected for the lkb.)
5032 *
5033 * Second, the lkb is reprocessed like an original operation would be,
5034 * by passing it to _request_lock or _convert_lock, which will either
5035 * process the lkb operation locally, or send it to a remote node again
5036 * and put the lkb back onto the waiters list.
5037 *
5038 * When reprocessing the lkb, we may find that it's flagged for an overlapping
5039 * force-unlock or cancel, either from before recovery began, or after recovery
5040 * finished. If this is the case, the unlock/cancel is done directly, and the
5041 * original operation is not initiated again (no _request_lock/_convert_lock.)
5042 */
5043
5044int dlm_recover_waiters_post(struct dlm_ls *ls)
5045{
5046 struct dlm_lkb *lkb;
5047 struct dlm_rsb *r;
5048 int error = 0, mstype, err, oc, ou;
5049
5050 while (1) {
5051 if (dlm_locking_stopped(ls)) {
5052 log_debug(ls, "recover_waiters_post aborted");
5053 error = -EINTR;
5054 break;
5055 }
5056
5057 /*
5058 * Find an lkb from the waiters list that's been affected by
5059 * recovery node changes, and needs to be reprocessed. Does
5060 * hold_lkb(), adding a refcount.
5061 */
5062 lkb = find_resend_waiter(ls);
5063 if (!lkb)
5064 break;
5065
5066 r = lkb->lkb_resource;
5067 hold_rsb(r);
5068 lock_rsb(r);
5069
5070 /*
5071 * If the lkb has been flagged for a force unlock or cancel,
5072 * then the reprocessing below will be replaced by just doing
5073 * the unlock/cancel directly.
5074 */
5075 mstype = lkb->lkb_wait_type;
5076 oc = test_and_clear_bit(DLM_IFL_OVERLAP_CANCEL_BIT,
5077 &lkb->lkb_iflags);
5078 ou = test_and_clear_bit(DLM_IFL_OVERLAP_UNLOCK_BIT,
5079 &lkb->lkb_iflags);
5080 err = 0;
5081
5082 log_debug(ls, "waiter %x remote %x msg %d r_nodeid %d "
5083 "lkb_nodeid %d wait_nodeid %d dir_nodeid %d "
5084 "overlap %d %d", lkb->lkb_id, lkb->lkb_remid, mstype,
5085 r->res_nodeid, lkb->lkb_nodeid, lkb->lkb_wait_nodeid,
5086 dlm_dir_nodeid(r), oc, ou);
5087
5088 /*
5089 * No reply to the pre-recovery operation will now be received,
5090 * so a forced equivalent of remove_from_waiters() is needed to
5091 * reset the waiters state that was in place before recovery.
5092 */
5093
5094 clear_bit(DLM_IFL_RESEND_BIT, &lkb->lkb_iflags);
5095
5096 /* Forcibly clear wait_type */
5097 lkb->lkb_wait_type = 0;
5098
5099 /*
5100 * Forcibly reset wait_count and associated refcount. The
5101 * wait_count will almost always be 1, but in case of an
5102 * overlapping unlock/cancel it could be 2: see where
5103 * add_to_waiters() finds the lkb is already on the waiters
5104 * list and does lkb_wait_count++; hold_lkb().
5105 */
5106 while (lkb->lkb_wait_count) {
5107 lkb->lkb_wait_count--;
5108 unhold_lkb(lkb);
5109 }
5110
5111 /* Forcibly remove from waiters list */
5112 mutex_lock(&ls->ls_waiters_mutex);
5113 list_del_init(&lkb->lkb_wait_reply);
5114 mutex_unlock(&ls->ls_waiters_mutex);
5115
5116 /*
5117 * The lkb is now clear of all prior waiters state and can be
5118 * processed locally, or sent to remote node again, or directly
5119 * cancelled/unlocked.
5120 */
5121
5122 if (oc || ou) {
5123 /* do an unlock or cancel instead of resending */
5124 switch (mstype) {
5125 case DLM_MSG_LOOKUP:
5126 case DLM_MSG_REQUEST:
5127 queue_cast(r, lkb, ou ? -DLM_EUNLOCK :
5128 -DLM_ECANCEL);
5129 unhold_lkb(lkb); /* undoes create_lkb() */
5130 break;
5131 case DLM_MSG_CONVERT:
5132 if (oc) {
5133 queue_cast(r, lkb, -DLM_ECANCEL);
5134 } else {
5135 lkb->lkb_exflags |= DLM_LKF_FORCEUNLOCK;
5136 _unlock_lock(r, lkb);
5137 }
5138 break;
5139 default:
5140 err = 1;
5141 }
5142 } else {
5143 switch (mstype) {
5144 case DLM_MSG_LOOKUP:
5145 case DLM_MSG_REQUEST:
5146 _request_lock(r, lkb);
5147 if (is_master(r))
5148 confirm_master(r, 0);
5149 break;
5150 case DLM_MSG_CONVERT:
5151 _convert_lock(r, lkb);
5152 break;
5153 default:
5154 err = 1;
5155 }
5156 }
5157
5158 if (err) {
5159 log_error(ls, "waiter %x msg %d r_nodeid %d "
5160 "dir_nodeid %d overlap %d %d",
5161 lkb->lkb_id, mstype, r->res_nodeid,
5162 dlm_dir_nodeid(r), oc, ou);
5163 }
5164 unlock_rsb(r);
5165 put_rsb(r);
5166 dlm_put_lkb(lkb);
5167 }
5168
5169 return error;
5170}
5171
5172static void purge_mstcpy_list(struct dlm_ls *ls, struct dlm_rsb *r,
5173 struct list_head *list)
5174{
5175 struct dlm_lkb *lkb, *safe;
5176
5177 list_for_each_entry_safe(lkb, safe, list, lkb_statequeue) {
5178 if (!is_master_copy(lkb))
5179 continue;
5180
5181 /* don't purge lkbs we've added in recover_master_copy for
5182 the current recovery seq */
5183
5184 if (lkb->lkb_recover_seq == ls->ls_recover_seq)
5185 continue;
5186
5187 del_lkb(r, lkb);
5188
5189 /* this put should free the lkb */
5190 if (!dlm_put_lkb(lkb))
5191 log_error(ls, "purged mstcpy lkb not released");
5192 }
5193}
5194
5195void dlm_purge_mstcpy_locks(struct dlm_rsb *r)
5196{
5197 struct dlm_ls *ls = r->res_ls;
5198
5199 purge_mstcpy_list(ls, r, &r->res_grantqueue);
5200 purge_mstcpy_list(ls, r, &r->res_convertqueue);
5201 purge_mstcpy_list(ls, r, &r->res_waitqueue);
5202}
5203
5204static void purge_dead_list(struct dlm_ls *ls, struct dlm_rsb *r,
5205 struct list_head *list,
5206 int nodeid_gone, unsigned int *count)
5207{
5208 struct dlm_lkb *lkb, *safe;
5209
5210 list_for_each_entry_safe(lkb, safe, list, lkb_statequeue) {
5211 if (!is_master_copy(lkb))
5212 continue;
5213
5214 if ((lkb->lkb_nodeid == nodeid_gone) ||
5215 dlm_is_removed(ls, lkb->lkb_nodeid)) {
5216
5217 /* tell recover_lvb to invalidate the lvb
5218 because a node holding EX/PW failed */
5219 if ((lkb->lkb_exflags & DLM_LKF_VALBLK) &&
5220 (lkb->lkb_grmode >= DLM_LOCK_PW)) {
5221 rsb_set_flag(r, RSB_RECOVER_LVB_INVAL);
5222 }
5223
5224 del_lkb(r, lkb);
5225
5226 /* this put should free the lkb */
5227 if (!dlm_put_lkb(lkb))
5228 log_error(ls, "purged dead lkb not released");
5229
5230 rsb_set_flag(r, RSB_RECOVER_GRANT);
5231
5232 (*count)++;
5233 }
5234 }
5235}
5236
5237/* Get rid of locks held by nodes that are gone. */
5238
5239void dlm_recover_purge(struct dlm_ls *ls)
5240{
5241 struct dlm_rsb *r;
5242 struct dlm_member *memb;
5243 int nodes_count = 0;
5244 int nodeid_gone = 0;
5245 unsigned int lkb_count = 0;
5246
5247 /* cache one removed nodeid to optimize the common
5248 case of a single node removed */
5249
5250 list_for_each_entry(memb, &ls->ls_nodes_gone, list) {
5251 nodes_count++;
5252 nodeid_gone = memb->nodeid;
5253 }
5254
5255 if (!nodes_count)
5256 return;
5257
5258 down_write(&ls->ls_root_sem);
5259 list_for_each_entry(r, &ls->ls_root_list, res_root_list) {
5260 hold_rsb(r);
5261 lock_rsb(r);
5262 if (is_master(r)) {
5263 purge_dead_list(ls, r, &r->res_grantqueue,
5264 nodeid_gone, &lkb_count);
5265 purge_dead_list(ls, r, &r->res_convertqueue,
5266 nodeid_gone, &lkb_count);
5267 purge_dead_list(ls, r, &r->res_waitqueue,
5268 nodeid_gone, &lkb_count);
5269 }
5270 unlock_rsb(r);
5271 unhold_rsb(r);
5272 cond_resched();
5273 }
5274 up_write(&ls->ls_root_sem);
5275
5276 if (lkb_count)
5277 log_rinfo(ls, "dlm_recover_purge %u locks for %u nodes",
5278 lkb_count, nodes_count);
5279}
5280
5281static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket)
5282{
5283 struct rb_node *n;
5284 struct dlm_rsb *r;
5285
5286 spin_lock(&ls->ls_rsbtbl[bucket].lock);
5287 for (n = rb_first(&ls->ls_rsbtbl[bucket].keep); n; n = rb_next(n)) {
5288 r = rb_entry(n, struct dlm_rsb, res_hashnode);
5289
5290 if (!rsb_flag(r, RSB_RECOVER_GRANT))
5291 continue;
5292 if (!is_master(r)) {
5293 rsb_clear_flag(r, RSB_RECOVER_GRANT);
5294 continue;
5295 }
5296 hold_rsb(r);
5297 spin_unlock(&ls->ls_rsbtbl[bucket].lock);
5298 return r;
5299 }
5300 spin_unlock(&ls->ls_rsbtbl[bucket].lock);
5301 return NULL;
5302}
5303
5304/*
5305 * Attempt to grant locks on resources that we are the master of.
5306 * Locks may have become grantable during recovery because locks
5307 * from departed nodes have been purged (or not rebuilt), allowing
5308 * previously blocked locks to now be granted. The subset of rsb's
5309 * we are interested in are those with lkb's on either the convert or
5310 * waiting queues.
5311 *
5312 * Simplest would be to go through each master rsb and check for non-empty
5313 * convert or waiting queues, and attempt to grant on those rsbs.
5314 * Checking the queues requires lock_rsb, though, for which we'd need
5315 * to release the rsbtbl lock. This would make iterating through all
5316 * rsb's very inefficient. So, we rely on earlier recovery routines
5317 * to set RECOVER_GRANT on any rsb's that we should attempt to grant
5318 * locks for.
5319 */
5320
5321void dlm_recover_grant(struct dlm_ls *ls)
5322{
5323 struct dlm_rsb *r;
5324 int bucket = 0;
5325 unsigned int count = 0;
5326 unsigned int rsb_count = 0;
5327 unsigned int lkb_count = 0;
5328
5329 while (1) {
5330 r = find_grant_rsb(ls, bucket);
5331 if (!r) {
5332 if (bucket == ls->ls_rsbtbl_size - 1)
5333 break;
5334 bucket++;
5335 continue;
5336 }
5337 rsb_count++;
5338 count = 0;
5339 lock_rsb(r);
5340 /* the RECOVER_GRANT flag is checked in the grant path */
5341 grant_pending_locks(r, &count);
5342 rsb_clear_flag(r, RSB_RECOVER_GRANT);
5343 lkb_count += count;
5344 confirm_master(r, 0);
5345 unlock_rsb(r);
5346 put_rsb(r);
5347 cond_resched();
5348 }
5349
5350 if (lkb_count)
5351 log_rinfo(ls, "dlm_recover_grant %u locks on %u resources",
5352 lkb_count, rsb_count);
5353}
5354
5355static struct dlm_lkb *search_remid_list(struct list_head *head, int nodeid,
5356 uint32_t remid)
5357{
5358 struct dlm_lkb *lkb;
5359
5360 list_for_each_entry(lkb, head, lkb_statequeue) {
5361 if (lkb->lkb_nodeid == nodeid && lkb->lkb_remid == remid)
5362 return lkb;
5363 }
5364 return NULL;
5365}
5366
5367static struct dlm_lkb *search_remid(struct dlm_rsb *r, int nodeid,
5368 uint32_t remid)
5369{
5370 struct dlm_lkb *lkb;
5371
5372 lkb = search_remid_list(&r->res_grantqueue, nodeid, remid);
5373 if (lkb)
5374 return lkb;
5375 lkb = search_remid_list(&r->res_convertqueue, nodeid, remid);
5376 if (lkb)
5377 return lkb;
5378 lkb = search_remid_list(&r->res_waitqueue, nodeid, remid);
5379 if (lkb)
5380 return lkb;
5381 return NULL;
5382}
5383
5384/* needs at least dlm_rcom + rcom_lock */
5385static int receive_rcom_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
5386 struct dlm_rsb *r, const struct dlm_rcom *rc)
5387{
5388 struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
5389
5390 lkb->lkb_nodeid = le32_to_cpu(rc->rc_header.h_nodeid);
5391 lkb->lkb_ownpid = le32_to_cpu(rl->rl_ownpid);
5392 lkb->lkb_remid = le32_to_cpu(rl->rl_lkid);
5393 lkb->lkb_exflags = le32_to_cpu(rl->rl_exflags);
5394 dlm_set_dflags_val(lkb, le32_to_cpu(rl->rl_flags));
5395 set_bit(DLM_IFL_MSTCPY_BIT, &lkb->lkb_iflags);
5396 lkb->lkb_lvbseq = le32_to_cpu(rl->rl_lvbseq);
5397 lkb->lkb_rqmode = rl->rl_rqmode;
5398 lkb->lkb_grmode = rl->rl_grmode;
5399 /* don't set lkb_status because add_lkb wants to itself */
5400
5401 lkb->lkb_bastfn = (rl->rl_asts & DLM_CB_BAST) ? &fake_bastfn : NULL;
5402 lkb->lkb_astfn = (rl->rl_asts & DLM_CB_CAST) ? &fake_astfn : NULL;
5403
5404 if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
5405 int lvblen = le16_to_cpu(rc->rc_header.h_length) -
5406 sizeof(struct dlm_rcom) - sizeof(struct rcom_lock);
5407 if (lvblen > ls->ls_lvblen)
5408 return -EINVAL;
5409 lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
5410 if (!lkb->lkb_lvbptr)
5411 return -ENOMEM;
5412 memcpy(lkb->lkb_lvbptr, rl->rl_lvb, lvblen);
5413 }
5414
5415 /* Conversions between PR and CW (middle modes) need special handling.
5416 The real granted mode of these converting locks cannot be determined
5417 until all locks have been rebuilt on the rsb (recover_conversion) */
5418
5419 if (rl->rl_wait_type == cpu_to_le16(DLM_MSG_CONVERT) &&
5420 middle_conversion(lkb)) {
5421 rl->rl_status = DLM_LKSTS_CONVERT;
5422 lkb->lkb_grmode = DLM_LOCK_IV;
5423 rsb_set_flag(r, RSB_RECOVER_CONVERT);
5424 }
5425
5426 return 0;
5427}
5428
5429/* This lkb may have been recovered in a previous aborted recovery so we need
5430 to check if the rsb already has an lkb with the given remote nodeid/lkid.
5431 If so we just send back a standard reply. If not, we create a new lkb with
5432 the given values and send back our lkid. We send back our lkid by sending
5433 back the rcom_lock struct we got but with the remid field filled in. */
5434
5435/* needs at least dlm_rcom + rcom_lock */
5436int dlm_recover_master_copy(struct dlm_ls *ls, const struct dlm_rcom *rc,
5437 __le32 *rl_remid, __le32 *rl_result)
5438{
5439 struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
5440 struct dlm_rsb *r;
5441 struct dlm_lkb *lkb;
5442 uint32_t remid = 0;
5443 int from_nodeid = le32_to_cpu(rc->rc_header.h_nodeid);
5444 int error;
5445
5446 /* init rl_remid with rcom lock rl_remid */
5447 *rl_remid = rl->rl_remid;
5448
5449 if (rl->rl_parent_lkid) {
5450 error = -EOPNOTSUPP;
5451 goto out;
5452 }
5453
5454 remid = le32_to_cpu(rl->rl_lkid);
5455
5456 /* In general we expect the rsb returned to be R_MASTER, but we don't
5457 have to require it. Recovery of masters on one node can overlap
5458 recovery of locks on another node, so one node can send us MSTCPY
5459 locks before we've made ourselves master of this rsb. We can still
5460 add new MSTCPY locks that we receive here without any harm; when
5461 we make ourselves master, dlm_recover_masters() won't touch the
5462 MSTCPY locks we've received early. */
5463
5464 error = find_rsb(ls, rl->rl_name, le16_to_cpu(rl->rl_namelen),
5465 from_nodeid, R_RECEIVE_RECOVER, &r);
5466 if (error)
5467 goto out;
5468
5469 lock_rsb(r);
5470
5471 if (dlm_no_directory(ls) && (dlm_dir_nodeid(r) != dlm_our_nodeid())) {
5472 log_error(ls, "dlm_recover_master_copy remote %d %x not dir",
5473 from_nodeid, remid);
5474 error = -EBADR;
5475 goto out_unlock;
5476 }
5477
5478 lkb = search_remid(r, from_nodeid, remid);
5479 if (lkb) {
5480 error = -EEXIST;
5481 goto out_remid;
5482 }
5483
5484 error = create_lkb(ls, &lkb);
5485 if (error)
5486 goto out_unlock;
5487
5488 error = receive_rcom_lock_args(ls, lkb, r, rc);
5489 if (error) {
5490 __put_lkb(ls, lkb);
5491 goto out_unlock;
5492 }
5493
5494 attach_lkb(r, lkb);
5495 add_lkb(r, lkb, rl->rl_status);
5496 ls->ls_recover_locks_in++;
5497
5498 if (!list_empty(&r->res_waitqueue) || !list_empty(&r->res_convertqueue))
5499 rsb_set_flag(r, RSB_RECOVER_GRANT);
5500
5501 out_remid:
5502 /* this is the new value returned to the lock holder for
5503 saving in its process-copy lkb */
5504 *rl_remid = cpu_to_le32(lkb->lkb_id);
5505
5506 lkb->lkb_recover_seq = ls->ls_recover_seq;
5507
5508 out_unlock:
5509 unlock_rsb(r);
5510 put_rsb(r);
5511 out:
5512 if (error && error != -EEXIST)
5513 log_rinfo(ls, "dlm_recover_master_copy remote %d %x error %d",
5514 from_nodeid, remid, error);
5515 *rl_result = cpu_to_le32(error);
5516 return error;
5517}
5518
5519/* needs at least dlm_rcom + rcom_lock */
5520int dlm_recover_process_copy(struct dlm_ls *ls, const struct dlm_rcom *rc,
5521 uint64_t seq)
5522{
5523 struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
5524 struct dlm_rsb *r;
5525 struct dlm_lkb *lkb;
5526 uint32_t lkid, remid;
5527 int error, result;
5528
5529 lkid = le32_to_cpu(rl->rl_lkid);
5530 remid = le32_to_cpu(rl->rl_remid);
5531 result = le32_to_cpu(rl->rl_result);
5532
5533 error = find_lkb(ls, lkid, &lkb);
5534 if (error) {
5535 log_error(ls, "dlm_recover_process_copy no %x remote %d %x %d",
5536 lkid, le32_to_cpu(rc->rc_header.h_nodeid), remid,
5537 result);
5538 return error;
5539 }
5540
5541 r = lkb->lkb_resource;
5542 hold_rsb(r);
5543 lock_rsb(r);
5544
5545 if (!is_process_copy(lkb)) {
5546 log_error(ls, "dlm_recover_process_copy bad %x remote %d %x %d",
5547 lkid, le32_to_cpu(rc->rc_header.h_nodeid), remid,
5548 result);
5549 dlm_dump_rsb(r);
5550 unlock_rsb(r);
5551 put_rsb(r);
5552 dlm_put_lkb(lkb);
5553 return -EINVAL;
5554 }
5555
5556 switch (result) {
5557 case -EBADR:
5558 /* There's a chance the new master received our lock before
5559 dlm_recover_master_reply(), this wouldn't happen if we did
5560 a barrier between recover_masters and recover_locks. */
5561
5562 log_debug(ls, "dlm_recover_process_copy %x remote %d %x %d",
5563 lkid, le32_to_cpu(rc->rc_header.h_nodeid), remid,
5564 result);
5565
5566 dlm_send_rcom_lock(r, lkb, seq);
5567 goto out;
5568 case -EEXIST:
5569 case 0:
5570 lkb->lkb_remid = remid;
5571 break;
5572 default:
5573 log_error(ls, "dlm_recover_process_copy %x remote %d %x %d unk",
5574 lkid, le32_to_cpu(rc->rc_header.h_nodeid), remid,
5575 result);
5576 }
5577
5578 /* an ack for dlm_recover_locks() which waits for replies from
5579 all the locks it sends to new masters */
5580 dlm_recovered_lock(r);
5581 out:
5582 unlock_rsb(r);
5583 put_rsb(r);
5584 dlm_put_lkb(lkb);
5585
5586 return 0;
5587}
5588
5589int dlm_user_request(struct dlm_ls *ls, struct dlm_user_args *ua,
5590 int mode, uint32_t flags, void *name, unsigned int namelen)
5591{
5592 struct dlm_lkb *lkb;
5593 struct dlm_args args;
5594 bool do_put = true;
5595 int error;
5596
5597 dlm_lock_recovery(ls);
5598
5599 error = create_lkb(ls, &lkb);
5600 if (error) {
5601 kfree(ua);
5602 goto out;
5603 }
5604
5605 trace_dlm_lock_start(ls, lkb, name, namelen, mode, flags);
5606
5607 if (flags & DLM_LKF_VALBLK) {
5608 ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
5609 if (!ua->lksb.sb_lvbptr) {
5610 kfree(ua);
5611 error = -ENOMEM;
5612 goto out_put;
5613 }
5614 }
5615 error = set_lock_args(mode, &ua->lksb, flags, namelen, fake_astfn, ua,
5616 fake_bastfn, &args);
5617 if (error) {
5618 kfree(ua->lksb.sb_lvbptr);
5619 ua->lksb.sb_lvbptr = NULL;
5620 kfree(ua);
5621 goto out_put;
5622 }
5623
5624 /* After ua is attached to lkb it will be freed by dlm_free_lkb().
5625 When DLM_DFL_USER_BIT is set, the dlm knows that this is a userspace
5626 lock and that lkb_astparam is the dlm_user_args structure. */
5627 set_bit(DLM_DFL_USER_BIT, &lkb->lkb_dflags);
5628 error = request_lock(ls, lkb, name, namelen, &args);
5629
5630 switch (error) {
5631 case 0:
5632 break;
5633 case -EINPROGRESS:
5634 error = 0;
5635 break;
5636 case -EAGAIN:
5637 error = 0;
5638 fallthrough;
5639 default:
5640 goto out_put;
5641 }
5642
5643 /* add this new lkb to the per-process list of locks */
5644 spin_lock(&ua->proc->locks_spin);
5645 hold_lkb(lkb);
5646 list_add_tail(&lkb->lkb_ownqueue, &ua->proc->locks);
5647 spin_unlock(&ua->proc->locks_spin);
5648 do_put = false;
5649 out_put:
5650 trace_dlm_lock_end(ls, lkb, name, namelen, mode, flags, error, false);
5651 if (do_put)
5652 __put_lkb(ls, lkb);
5653 out:
5654 dlm_unlock_recovery(ls);
5655 return error;
5656}
5657
5658int dlm_user_convert(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
5659 int mode, uint32_t flags, uint32_t lkid, char *lvb_in)
5660{
5661 struct dlm_lkb *lkb;
5662 struct dlm_args args;
5663 struct dlm_user_args *ua;
5664 int error;
5665
5666 dlm_lock_recovery(ls);
5667
5668 error = find_lkb(ls, lkid, &lkb);
5669 if (error)
5670 goto out;
5671
5672 trace_dlm_lock_start(ls, lkb, NULL, 0, mode, flags);
5673
5674 /* user can change the params on its lock when it converts it, or
5675 add an lvb that didn't exist before */
5676
5677 ua = lkb->lkb_ua;
5678
5679 if (flags & DLM_LKF_VALBLK && !ua->lksb.sb_lvbptr) {
5680 ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
5681 if (!ua->lksb.sb_lvbptr) {
5682 error = -ENOMEM;
5683 goto out_put;
5684 }
5685 }
5686 if (lvb_in && ua->lksb.sb_lvbptr)
5687 memcpy(ua->lksb.sb_lvbptr, lvb_in, DLM_USER_LVB_LEN);
5688
5689 ua->xid = ua_tmp->xid;
5690 ua->castparam = ua_tmp->castparam;
5691 ua->castaddr = ua_tmp->castaddr;
5692 ua->bastparam = ua_tmp->bastparam;
5693 ua->bastaddr = ua_tmp->bastaddr;
5694 ua->user_lksb = ua_tmp->user_lksb;
5695
5696 error = set_lock_args(mode, &ua->lksb, flags, 0, fake_astfn, ua,
5697 fake_bastfn, &args);
5698 if (error)
5699 goto out_put;
5700
5701 error = convert_lock(ls, lkb, &args);
5702
5703 if (error == -EINPROGRESS || error == -EAGAIN || error == -EDEADLK)
5704 error = 0;
5705 out_put:
5706 trace_dlm_lock_end(ls, lkb, NULL, 0, mode, flags, error, false);
5707 dlm_put_lkb(lkb);
5708 out:
5709 dlm_unlock_recovery(ls);
5710 kfree(ua_tmp);
5711 return error;
5712}
5713
5714/*
5715 * The caller asks for an orphan lock on a given resource with a given mode.
5716 * If a matching lock exists, it's moved to the owner's list of locks and
5717 * the lkid is returned.
5718 */
5719
5720int dlm_user_adopt_orphan(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
5721 int mode, uint32_t flags, void *name, unsigned int namelen,
5722 uint32_t *lkid)
5723{
5724 struct dlm_lkb *lkb = NULL, *iter;
5725 struct dlm_user_args *ua;
5726 int found_other_mode = 0;
5727 int rv = 0;
5728
5729 mutex_lock(&ls->ls_orphans_mutex);
5730 list_for_each_entry(iter, &ls->ls_orphans, lkb_ownqueue) {
5731 if (iter->lkb_resource->res_length != namelen)
5732 continue;
5733 if (memcmp(iter->lkb_resource->res_name, name, namelen))
5734 continue;
5735 if (iter->lkb_grmode != mode) {
5736 found_other_mode = 1;
5737 continue;
5738 }
5739
5740 lkb = iter;
5741 list_del_init(&iter->lkb_ownqueue);
5742 clear_bit(DLM_DFL_ORPHAN_BIT, &iter->lkb_dflags);
5743 *lkid = iter->lkb_id;
5744 break;
5745 }
5746 mutex_unlock(&ls->ls_orphans_mutex);
5747
5748 if (!lkb && found_other_mode) {
5749 rv = -EAGAIN;
5750 goto out;
5751 }
5752
5753 if (!lkb) {
5754 rv = -ENOENT;
5755 goto out;
5756 }
5757
5758 lkb->lkb_exflags = flags;
5759 lkb->lkb_ownpid = (int) current->pid;
5760
5761 ua = lkb->lkb_ua;
5762
5763 ua->proc = ua_tmp->proc;
5764 ua->xid = ua_tmp->xid;
5765 ua->castparam = ua_tmp->castparam;
5766 ua->castaddr = ua_tmp->castaddr;
5767 ua->bastparam = ua_tmp->bastparam;
5768 ua->bastaddr = ua_tmp->bastaddr;
5769 ua->user_lksb = ua_tmp->user_lksb;
5770
5771 /*
5772 * The lkb reference from the ls_orphans list was not
5773 * removed above, and is now considered the reference
5774 * for the proc locks list.
5775 */
5776
5777 spin_lock(&ua->proc->locks_spin);
5778 list_add_tail(&lkb->lkb_ownqueue, &ua->proc->locks);
5779 spin_unlock(&ua->proc->locks_spin);
5780 out:
5781 kfree(ua_tmp);
5782 return rv;
5783}
5784
5785int dlm_user_unlock(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
5786 uint32_t flags, uint32_t lkid, char *lvb_in)
5787{
5788 struct dlm_lkb *lkb;
5789 struct dlm_args args;
5790 struct dlm_user_args *ua;
5791 int error;
5792
5793 dlm_lock_recovery(ls);
5794
5795 error = find_lkb(ls, lkid, &lkb);
5796 if (error)
5797 goto out;
5798
5799 trace_dlm_unlock_start(ls, lkb, flags);
5800
5801 ua = lkb->lkb_ua;
5802
5803 if (lvb_in && ua->lksb.sb_lvbptr)
5804 memcpy(ua->lksb.sb_lvbptr, lvb_in, DLM_USER_LVB_LEN);
5805 if (ua_tmp->castparam)
5806 ua->castparam = ua_tmp->castparam;
5807 ua->user_lksb = ua_tmp->user_lksb;
5808
5809 error = set_unlock_args(flags, ua, &args);
5810 if (error)
5811 goto out_put;
5812
5813 error = unlock_lock(ls, lkb, &args);
5814
5815 if (error == -DLM_EUNLOCK)
5816 error = 0;
5817 /* from validate_unlock_args() */
5818 if (error == -EBUSY && (flags & DLM_LKF_FORCEUNLOCK))
5819 error = 0;
5820 if (error)
5821 goto out_put;
5822
5823 spin_lock(&ua->proc->locks_spin);
5824 /* dlm_user_add_cb() may have already taken lkb off the proc list */
5825 if (!list_empty(&lkb->lkb_ownqueue))
5826 list_move(&lkb->lkb_ownqueue, &ua->proc->unlocking);
5827 spin_unlock(&ua->proc->locks_spin);
5828 out_put:
5829 trace_dlm_unlock_end(ls, lkb, flags, error);
5830 dlm_put_lkb(lkb);
5831 out:
5832 dlm_unlock_recovery(ls);
5833 kfree(ua_tmp);
5834 return error;
5835}
5836
5837int dlm_user_cancel(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
5838 uint32_t flags, uint32_t lkid)
5839{
5840 struct dlm_lkb *lkb;
5841 struct dlm_args args;
5842 struct dlm_user_args *ua;
5843 int error;
5844
5845 dlm_lock_recovery(ls);
5846
5847 error = find_lkb(ls, lkid, &lkb);
5848 if (error)
5849 goto out;
5850
5851 trace_dlm_unlock_start(ls, lkb, flags);
5852
5853 ua = lkb->lkb_ua;
5854 if (ua_tmp->castparam)
5855 ua->castparam = ua_tmp->castparam;
5856 ua->user_lksb = ua_tmp->user_lksb;
5857
5858 error = set_unlock_args(flags, ua, &args);
5859 if (error)
5860 goto out_put;
5861
5862 error = cancel_lock(ls, lkb, &args);
5863
5864 if (error == -DLM_ECANCEL)
5865 error = 0;
5866 /* from validate_unlock_args() */
5867 if (error == -EBUSY)
5868 error = 0;
5869 out_put:
5870 trace_dlm_unlock_end(ls, lkb, flags, error);
5871 dlm_put_lkb(lkb);
5872 out:
5873 dlm_unlock_recovery(ls);
5874 kfree(ua_tmp);
5875 return error;
5876}
5877
5878int dlm_user_deadlock(struct dlm_ls *ls, uint32_t flags, uint32_t lkid)
5879{
5880 struct dlm_lkb *lkb;
5881 struct dlm_args args;
5882 struct dlm_user_args *ua;
5883 struct dlm_rsb *r;
5884 int error;
5885
5886 dlm_lock_recovery(ls);
5887
5888 error = find_lkb(ls, lkid, &lkb);
5889 if (error)
5890 goto out;
5891
5892 trace_dlm_unlock_start(ls, lkb, flags);
5893
5894 ua = lkb->lkb_ua;
5895
5896 error = set_unlock_args(flags, ua, &args);
5897 if (error)
5898 goto out_put;
5899
5900 /* same as cancel_lock(), but set DEADLOCK_CANCEL after lock_rsb */
5901
5902 r = lkb->lkb_resource;
5903 hold_rsb(r);
5904 lock_rsb(r);
5905
5906 error = validate_unlock_args(lkb, &args);
5907 if (error)
5908 goto out_r;
5909 set_bit(DLM_IFL_DEADLOCK_CANCEL_BIT, &lkb->lkb_iflags);
5910
5911 error = _cancel_lock(r, lkb);
5912 out_r:
5913 unlock_rsb(r);
5914 put_rsb(r);
5915
5916 if (error == -DLM_ECANCEL)
5917 error = 0;
5918 /* from validate_unlock_args() */
5919 if (error == -EBUSY)
5920 error = 0;
5921 out_put:
5922 trace_dlm_unlock_end(ls, lkb, flags, error);
5923 dlm_put_lkb(lkb);
5924 out:
5925 dlm_unlock_recovery(ls);
5926 return error;
5927}
5928
5929/* lkb's that are removed from the waiters list by revert are just left on the
5930 orphans list with the granted orphan locks, to be freed by purge */
5931
5932static int orphan_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb)
5933{
5934 struct dlm_args args;
5935 int error;
5936
5937 hold_lkb(lkb); /* reference for the ls_orphans list */
5938 mutex_lock(&ls->ls_orphans_mutex);
5939 list_add_tail(&lkb->lkb_ownqueue, &ls->ls_orphans);
5940 mutex_unlock(&ls->ls_orphans_mutex);
5941
5942 set_unlock_args(0, lkb->lkb_ua, &args);
5943
5944 error = cancel_lock(ls, lkb, &args);
5945 if (error == -DLM_ECANCEL)
5946 error = 0;
5947 return error;
5948}
5949
5950/* The FORCEUNLOCK flag allows the unlock to go ahead even if the lkb isn't
5951 granted. Regardless of what rsb queue the lock is on, it's removed and
5952 freed. The IVVALBLK flag causes the lvb on the resource to be invalidated
5953 if our lock is PW/EX (it's ignored if our granted mode is smaller.) */
5954
5955static int unlock_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb)
5956{
5957 struct dlm_args args;
5958 int error;
5959
5960 set_unlock_args(DLM_LKF_FORCEUNLOCK | DLM_LKF_IVVALBLK,
5961 lkb->lkb_ua, &args);
5962
5963 error = unlock_lock(ls, lkb, &args);
5964 if (error == -DLM_EUNLOCK)
5965 error = 0;
5966 return error;
5967}
5968
5969/* We have to release clear_proc_locks mutex before calling unlock_proc_lock()
5970 (which does lock_rsb) due to deadlock with receiving a message that does
5971 lock_rsb followed by dlm_user_add_cb() */
5972
5973static struct dlm_lkb *del_proc_lock(struct dlm_ls *ls,
5974 struct dlm_user_proc *proc)
5975{
5976 struct dlm_lkb *lkb = NULL;
5977
5978 spin_lock(&ls->ls_clear_proc_locks);
5979 if (list_empty(&proc->locks))
5980 goto out;
5981
5982 lkb = list_entry(proc->locks.next, struct dlm_lkb, lkb_ownqueue);
5983 list_del_init(&lkb->lkb_ownqueue);
5984
5985 if (lkb->lkb_exflags & DLM_LKF_PERSISTENT)
5986 set_bit(DLM_DFL_ORPHAN_BIT, &lkb->lkb_dflags);
5987 else
5988 set_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags);
5989 out:
5990 spin_unlock(&ls->ls_clear_proc_locks);
5991 return lkb;
5992}
5993
5994/* The ls_clear_proc_locks mutex protects against dlm_user_add_cb() which
5995 1) references lkb->ua which we free here and 2) adds lkbs to proc->asts,
5996 which we clear here. */
5997
5998/* proc CLOSING flag is set so no more device_reads should look at proc->asts
5999 list, and no more device_writes should add lkb's to proc->locks list; so we
6000 shouldn't need to take asts_spin or locks_spin here. this assumes that
6001 device reads/writes/closes are serialized -- FIXME: we may need to serialize
6002 them ourself. */
6003
6004void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc)
6005{
6006 struct dlm_lkb *lkb, *safe;
6007
6008 dlm_lock_recovery(ls);
6009
6010 while (1) {
6011 lkb = del_proc_lock(ls, proc);
6012 if (!lkb)
6013 break;
6014 if (lkb->lkb_exflags & DLM_LKF_PERSISTENT)
6015 orphan_proc_lock(ls, lkb);
6016 else
6017 unlock_proc_lock(ls, lkb);
6018
6019 /* this removes the reference for the proc->locks list
6020 added by dlm_user_request, it may result in the lkb
6021 being freed */
6022
6023 dlm_put_lkb(lkb);
6024 }
6025
6026 spin_lock(&ls->ls_clear_proc_locks);
6027
6028 /* in-progress unlocks */
6029 list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) {
6030 list_del_init(&lkb->lkb_ownqueue);
6031 set_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags);
6032 dlm_put_lkb(lkb);
6033 }
6034
6035 list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_cb_list) {
6036 dlm_purge_lkb_callbacks(lkb);
6037 list_del_init(&lkb->lkb_cb_list);
6038 dlm_put_lkb(lkb);
6039 }
6040
6041 spin_unlock(&ls->ls_clear_proc_locks);
6042 dlm_unlock_recovery(ls);
6043}
6044
6045static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc)
6046{
6047 struct dlm_lkb *lkb, *safe;
6048
6049 while (1) {
6050 lkb = NULL;
6051 spin_lock(&proc->locks_spin);
6052 if (!list_empty(&proc->locks)) {
6053 lkb = list_entry(proc->locks.next, struct dlm_lkb,
6054 lkb_ownqueue);
6055 list_del_init(&lkb->lkb_ownqueue);
6056 }
6057 spin_unlock(&proc->locks_spin);
6058
6059 if (!lkb)
6060 break;
6061
6062 set_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags);
6063 unlock_proc_lock(ls, lkb);
6064 dlm_put_lkb(lkb); /* ref from proc->locks list */
6065 }
6066
6067 spin_lock(&proc->locks_spin);
6068 list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) {
6069 list_del_init(&lkb->lkb_ownqueue);
6070 set_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags);
6071 dlm_put_lkb(lkb);
6072 }
6073 spin_unlock(&proc->locks_spin);
6074
6075 spin_lock(&proc->asts_spin);
6076 list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_cb_list) {
6077 dlm_purge_lkb_callbacks(lkb);
6078 list_del_init(&lkb->lkb_cb_list);
6079 dlm_put_lkb(lkb);
6080 }
6081 spin_unlock(&proc->asts_spin);
6082}
6083
6084/* pid of 0 means purge all orphans */
6085
6086static void do_purge(struct dlm_ls *ls, int nodeid, int pid)
6087{
6088 struct dlm_lkb *lkb, *safe;
6089
6090 mutex_lock(&ls->ls_orphans_mutex);
6091 list_for_each_entry_safe(lkb, safe, &ls->ls_orphans, lkb_ownqueue) {
6092 if (pid && lkb->lkb_ownpid != pid)
6093 continue;
6094 unlock_proc_lock(ls, lkb);
6095 list_del_init(&lkb->lkb_ownqueue);
6096 dlm_put_lkb(lkb);
6097 }
6098 mutex_unlock(&ls->ls_orphans_mutex);
6099}
6100
6101static int send_purge(struct dlm_ls *ls, int nodeid, int pid)
6102{
6103 struct dlm_message *ms;
6104 struct dlm_mhandle *mh;
6105 int error;
6106
6107 error = _create_message(ls, sizeof(struct dlm_message), nodeid,
6108 DLM_MSG_PURGE, &ms, &mh, GFP_NOFS);
6109 if (error)
6110 return error;
6111 ms->m_nodeid = cpu_to_le32(nodeid);
6112 ms->m_pid = cpu_to_le32(pid);
6113
6114 return send_message(mh, ms, NULL, 0);
6115}
6116
6117int dlm_user_purge(struct dlm_ls *ls, struct dlm_user_proc *proc,
6118 int nodeid, int pid)
6119{
6120 int error = 0;
6121
6122 if (nodeid && (nodeid != dlm_our_nodeid())) {
6123 error = send_purge(ls, nodeid, pid);
6124 } else {
6125 dlm_lock_recovery(ls);
6126 if (pid == current->pid)
6127 purge_proc_locks(ls, proc);
6128 else
6129 do_purge(ls, nodeid, pid);
6130 dlm_unlock_recovery(ls);
6131 }
6132 return error;
6133}
6134
6135/* debug functionality */
6136int dlm_debug_add_lkb(struct dlm_ls *ls, uint32_t lkb_id, char *name, int len,
6137 int lkb_nodeid, unsigned int lkb_dflags, int lkb_status)
6138{
6139 struct dlm_lksb *lksb;
6140 struct dlm_lkb *lkb;
6141 struct dlm_rsb *r;
6142 int error;
6143
6144 /* we currently can't set a valid user lock */
6145 if (lkb_dflags & BIT(DLM_DFL_USER_BIT))
6146 return -EOPNOTSUPP;
6147
6148 lksb = kzalloc(sizeof(*lksb), GFP_NOFS);
6149 if (!lksb)
6150 return -ENOMEM;
6151
6152 error = _create_lkb(ls, &lkb, lkb_id, lkb_id + 1);
6153 if (error) {
6154 kfree(lksb);
6155 return error;
6156 }
6157
6158 dlm_set_dflags_val(lkb, lkb_dflags);
6159 lkb->lkb_nodeid = lkb_nodeid;
6160 lkb->lkb_lksb = lksb;
6161 /* user specific pointer, just don't have it NULL for kernel locks */
6162 if (~lkb_dflags & BIT(DLM_DFL_USER_BIT))
6163 lkb->lkb_astparam = (void *)0xDEADBEEF;
6164
6165 error = find_rsb(ls, name, len, 0, R_REQUEST, &r);
6166 if (error) {
6167 kfree(lksb);
6168 __put_lkb(ls, lkb);
6169 return error;
6170 }
6171
6172 lock_rsb(r);
6173 attach_lkb(r, lkb);
6174 add_lkb(r, lkb, lkb_status);
6175 unlock_rsb(r);
6176 put_rsb(r);
6177
6178 return 0;
6179}
6180
6181int dlm_debug_add_lkb_to_waiters(struct dlm_ls *ls, uint32_t lkb_id,
6182 int mstype, int to_nodeid)
6183{
6184 struct dlm_lkb *lkb;
6185 int error;
6186
6187 error = find_lkb(ls, lkb_id, &lkb);
6188 if (error)
6189 return error;
6190
6191 error = add_to_waiters(lkb, mstype, to_nodeid);
6192 dlm_put_lkb(lkb);
6193 return error;
6194}
6195