projects / deliverable / linux.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
ocfs2: Add ->set_lvb callback in dlmglue
[deliverable/linux.git] / fs / ocfs2 / dlmglue.c
CommitLineData
ccd979bd
MF		 1/* -*- mode: c; c-basic-offset: 8; -*-
2 * vim: noexpandtab sw=8 ts=8 sts=0:
3 *
4 * dlmglue.c
5 *
6 * Code which implements an OCFS2 specific interface to our DLM.
7 *
8 * Copyright (C) 2003, 2004 Oracle. All rights reserved.
9 *
10 * This program is free software; you can redistribute it and/or
11 * modify it under the terms of the GNU General Public
12 * License as published by the Free Software Foundation; either
13 * version 2 of the License, or (at your option) any later version.
14 *
15 * This program is distributed in the hope that it will be useful,
16 * but WITHOUT ANY WARRANTY; without even the implied warranty of
17 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18 * General Public License for more details.
19 *
20 * You should have received a copy of the GNU General Public
21 * License along with this program; if not, write to the
22 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 * Boston, MA 021110-1307, USA.
24 */
25
26#include <linux/types.h>
27#include <linux/slab.h>
28#include <linux/highmem.h>
29#include <linux/mm.h>
30#include <linux/smp_lock.h>
31#include <linux/crc32.h>
32#include <linux/kthread.h>
33#include <linux/pagemap.h>
34#include <linux/debugfs.h>
35#include <linux/seq_file.h>
36
37#include <cluster/heartbeat.h>
38#include <cluster/nodemanager.h>
39#include <cluster/tcp.h>
40
41#include <dlm/dlmapi.h>
42
43#define MLOG_MASK_PREFIX ML_DLM_GLUE
44#include <cluster/masklog.h>
45
46#include "ocfs2.h"
47
48#include "alloc.h"
d680efe9 49#include "dcache.h"
ccd979bd
MF		 50#include "dlmglue.h"
51#include "extent_map.h"
52#include "heartbeat.h"
53#include "inode.h"
54#include "journal.h"
55#include "slot_map.h"
56#include "super.h"
57#include "uptodate.h"
58#include "vote.h"
59
60#include "buffer_head_io.h"
61
62struct ocfs2_mask_waiter {
63 struct list_head mw_item;
64 int mw_status;
65 struct completion mw_complete;
66 unsigned long mw_mask;
67 unsigned long mw_goal;
68};
69
54a7e755
MF		 70static struct ocfs2_super *ocfs2_get_dentry_osb(struct ocfs2_lock_res *lockres);
71static struct ocfs2_super *ocfs2_get_inode_osb(struct ocfs2_lock_res *lockres);
ccd979bd 72
d680efe9
MF		 73/*
74 * Return value from ocfs2_convert_worker_t functions.
75 *
76 * These control the precise actions of ocfs2_generic_unblock_lock()
77 * and ocfs2_process_blocked_lock()
78 *
79 */
80enum ocfs2_unblock_action {
81 UNBLOCK_CONTINUE = 0, /* Continue downconvert */
82 UNBLOCK_CONTINUE_POST = 1, /* Continue downconvert, fire
83 * ->post_unlock callback */
84 UNBLOCK_STOP_POST = 2, /* Do not downconvert, fire
85 * ->post_unlock() callback. */
86};
87
88struct ocfs2_unblock_ctl {
89 int requeue;
90 enum ocfs2_unblock_action unblock_action;
91};
92
ccd979bd 93static int ocfs2_unblock_meta(struct ocfs2_lock_res *lockres,
d680efe9 94 struct ocfs2_unblock_ctl *ctl);
ccd979bd 95static int ocfs2_unblock_data(struct ocfs2_lock_res *lockres,
d680efe9 96 struct ocfs2_unblock_ctl *ctl);
ccd979bd 97static int ocfs2_unblock_inode_lock(struct ocfs2_lock_res *lockres,
d680efe9
MF		 98 struct ocfs2_unblock_ctl *ctl);
99static int ocfs2_unblock_dentry_lock(struct ocfs2_lock_res *lockres,
100 struct ocfs2_unblock_ctl *ctl);
ccd979bd 101static int ocfs2_unblock_osb_lock(struct ocfs2_lock_res *lockres,
d680efe9
MF		 102 struct ocfs2_unblock_ctl *ctl);
103
104static void ocfs2_dentry_post_unlock(struct ocfs2_super *osb,
105 struct ocfs2_lock_res *lockres);
ccd979bd 106
f625c979
MF		 107/*
108 * OCFS2 Lock Resource Operations
109 *
110 * These fine tune the behavior of the generic dlmglue locking infrastructure.
111 */
ccd979bd 112struct ocfs2_lock_res_ops {
54a7e755
MF		 113 /*
114 * Translate an ocfs2_lock_res * into an ocfs2_super *. Define
115 * this callback if ->l_priv is not an ocfs2_super pointer
116 */
117 struct ocfs2_super * (*get_osb)(struct ocfs2_lock_res *);
d680efe9
MF		 118 int (*unblock)(struct ocfs2_lock_res *, struct ocfs2_unblock_ctl *);
119 void (*post_unlock)(struct ocfs2_super *, struct ocfs2_lock_res *);
f625c979 120
16d5b956
MF		 121 /*
122 * Allow a lock type to add checks to determine whether it is
123 * safe to downconvert a lock. Return 0 to re-queue the
124 * downconvert at a later time, nonzero to continue.
125 *
126 * For most locks, the default checks that there are no
127 * incompatible holders are sufficient.
128 *
129 * Called with the lockres spinlock held.
130 */
131 int (*check_downconvert)(struct ocfs2_lock_res *, int);
132
5ef0d4ea
MF		 133 /*
134 * Allows a lock type to populate the lock value block. This
135 * is called on downconvert, and when we drop a lock.
136 *
137 * Locks that want to use this should set LOCK_TYPE_USES_LVB
138 * in the flags field.
139 *
140 * Called with the lockres spinlock held.
141 */
142 void (*set_lvb)(struct ocfs2_lock_res *);
143
f625c979
MF		 144 /*
145 * LOCK_TYPE_* flags which describe the specific requirements
146 * of a lock type. Descriptions of each individual flag follow.
147 */
148 int flags;
ccd979bd
MF		 149};
150
f625c979
MF		 151/*
152 * Some locks want to "refresh" potentially stale data when a
153 * meaningful (PRMODE or EXMODE) lock level is first obtained. If this
154 * flag is set, the OCFS2_LOCK_NEEDS_REFRESH flag will be set on the
155 * individual lockres l_flags member from the ast function. It is
156 * expected that the locking wrapper will clear the
157 * OCFS2_LOCK_NEEDS_REFRESH flag when done.
158 */
159#define LOCK_TYPE_REQUIRES_REFRESH 0x1
160
b80fc012 161/*
5ef0d4ea
MF		 162 * Indicate that a lock type makes use of the lock value block. The
163 * ->set_lvb lock type callback must be defined.
b80fc012
MF		 164 */
165#define LOCK_TYPE_USES_LVB 0x2
166
d680efe9
MF		 167typedef int (ocfs2_convert_worker_t)(struct ocfs2_lock_res *, int);
168static int ocfs2_generic_unblock_lock(struct ocfs2_super *osb,
169 struct ocfs2_lock_res *lockres,
170 struct ocfs2_unblock_ctl *ctl,
171 ocfs2_convert_worker_t *worker);
172
ccd979bd 173static struct ocfs2_lock_res_ops ocfs2_inode_rw_lops = {
54a7e755 174 .get_osb = ocfs2_get_inode_osb,
ccd979bd 175 .unblock = ocfs2_unblock_inode_lock,
f625c979 176 .flags = 0,
ccd979bd
MF		 177};
178
179static struct ocfs2_lock_res_ops ocfs2_inode_meta_lops = {
54a7e755 180 .get_osb = ocfs2_get_inode_osb,
ccd979bd 181 .unblock = ocfs2_unblock_meta,
b80fc012 182 .flags = LOCK_TYPE_REQUIRES_REFRESH|LOCK_TYPE_USES_LVB,
ccd979bd
MF		 183};
184
ccd979bd 185static struct ocfs2_lock_res_ops ocfs2_inode_data_lops = {
54a7e755 186 .get_osb = ocfs2_get_inode_osb,
ccd979bd 187 .unblock = ocfs2_unblock_data,
f625c979 188 .flags = 0,
ccd979bd
MF		 189};
190
191static struct ocfs2_lock_res_ops ocfs2_super_lops = {
ccd979bd 192 .unblock = ocfs2_unblock_osb_lock,
f625c979 193 .flags = LOCK_TYPE_REQUIRES_REFRESH,
ccd979bd
MF		 194};
195
196static struct ocfs2_lock_res_ops ocfs2_rename_lops = {
ccd979bd 197 .unblock = ocfs2_unblock_osb_lock,
f625c979 198 .flags = 0,
ccd979bd
MF		 199};
200
d680efe9 201static struct ocfs2_lock_res_ops ocfs2_dentry_lops = {
54a7e755 202 .get_osb = ocfs2_get_dentry_osb,
d680efe9
MF		 203 .unblock = ocfs2_unblock_dentry_lock,
204 .post_unlock = ocfs2_dentry_post_unlock,
f625c979 205 .flags = 0,
d680efe9
MF		 206};
207
ccd979bd
MF		 208static inline int ocfs2_is_inode_lock(struct ocfs2_lock_res *lockres)
209{
210 return lockres->l_type == OCFS2_LOCK_TYPE_META ||
211 lockres->l_type == OCFS2_LOCK_TYPE_DATA ||
212 lockres->l_type == OCFS2_LOCK_TYPE_RW;
213}
214
ccd979bd
MF		 215static inline struct inode *ocfs2_lock_res_inode(struct ocfs2_lock_res *lockres)
216{
217 BUG_ON(!ocfs2_is_inode_lock(lockres));
218
219 return (struct inode *) lockres->l_priv;
220}
221
d680efe9
MF		 222static inline struct ocfs2_dentry_lock *ocfs2_lock_res_dl(struct ocfs2_lock_res *lockres)
223{
224 BUG_ON(lockres->l_type != OCFS2_LOCK_TYPE_DENTRY);
225
226 return (struct ocfs2_dentry_lock *)lockres->l_priv;
227}
228
54a7e755
MF		 229static inline struct ocfs2_super *ocfs2_get_lockres_osb(struct ocfs2_lock_res *lockres)
230{
231 if (lockres->l_ops->get_osb)
232 return lockres->l_ops->get_osb(lockres);
233
234 return (struct ocfs2_super *)lockres->l_priv;
235}
236
ccd979bd
MF		 237static int ocfs2_lock_create(struct ocfs2_super *osb,
238 struct ocfs2_lock_res *lockres,
239 int level,
240 int dlm_flags);
241static inline int ocfs2_may_continue_on_blocked_lock(struct ocfs2_lock_res *lockres,
242 int wanted);
243static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
244 struct ocfs2_lock_res *lockres,
245 int level);
246static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res *lockres);
247static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lockres);
248static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *lockres);
249static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres, int level);
250static void ocfs2_schedule_blocked_lock(struct ocfs2_super *osb,
251 struct ocfs2_lock_res *lockres);
252static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres,
253 int convert);
254#define ocfs2_log_dlm_error(_func, _stat, _lockres) do { \
255 mlog(ML_ERROR, "Dlm error \"%s\" while calling %s on " \
256 "resource %s: %s\n", dlm_errname(_stat), _func, \
257 _lockres->l_name, dlm_errmsg(_stat)); \
258} while (0)
259static void ocfs2_vote_on_unlock(struct ocfs2_super *osb,
260 struct ocfs2_lock_res *lockres);
261static int ocfs2_meta_lock_update(struct inode *inode,
262 struct buffer_head **bh);
263static void ocfs2_drop_osb_locks(struct ocfs2_super *osb);
264static inline int ocfs2_highest_compat_lock_level(int level);
265static inline int ocfs2_can_downconvert_meta_lock(struct inode *inode,
266 struct ocfs2_lock_res *lockres,
267 int new_level);
268
ccd979bd
MF		 269static void ocfs2_build_lock_name(enum ocfs2_lock_type type,
270 u64 blkno,
271 u32 generation,
272 char *name)
273{
274 int len;
275
276 mlog_entry_void();
277
278 BUG_ON(type >= OCFS2_NUM_LOCK_TYPES);
279
b0697053
MF		 280 len = snprintf(name, OCFS2_LOCK_ID_MAX_LEN, "%c%s%016llx%08x",
281 ocfs2_lock_type_char(type), OCFS2_LOCK_ID_PAD,
282 (long long)blkno, generation);
ccd979bd
MF		 283
284 BUG_ON(len != (OCFS2_LOCK_ID_MAX_LEN - 1));
285
286 mlog(0, "built lock resource with name: %s\n", name);
287
288 mlog_exit_void();
289}
290
34af946a 291static DEFINE_SPINLOCK(ocfs2_dlm_tracking_lock);
ccd979bd
MF		 292
293static void ocfs2_add_lockres_tracking(struct ocfs2_lock_res *res,
294 struct ocfs2_dlm_debug *dlm_debug)
295{
296 mlog(0, "Add tracking for lockres %s\n", res->l_name);
297
298 spin_lock(&ocfs2_dlm_tracking_lock);
299 list_add(&res->l_debug_list, &dlm_debug->d_lockres_tracking);
300 spin_unlock(&ocfs2_dlm_tracking_lock);
301}
302
303static void ocfs2_remove_lockres_tracking(struct ocfs2_lock_res *res)
304{
305 spin_lock(&ocfs2_dlm_tracking_lock);
306 if (!list_empty(&res->l_debug_list))
307 list_del_init(&res->l_debug_list);
308 spin_unlock(&ocfs2_dlm_tracking_lock);
309}
310
311static void ocfs2_lock_res_init_common(struct ocfs2_super *osb,
312 struct ocfs2_lock_res *res,
313 enum ocfs2_lock_type type,
ccd979bd
MF		 314 struct ocfs2_lock_res_ops *ops,
315 void *priv)
316{
ccd979bd
MF		 317 res->l_type = type;
318 res->l_ops = ops;
319 res->l_priv = priv;
320
321 res->l_level = LKM_IVMODE;
322 res->l_requested = LKM_IVMODE;
323 res->l_blocking = LKM_IVMODE;
324 res->l_action = OCFS2_AST_INVALID;
325 res->l_unlock_action = OCFS2_UNLOCK_INVALID;
326
327 res->l_flags = OCFS2_LOCK_INITIALIZED;
328
329 ocfs2_add_lockres_tracking(res, osb->osb_dlm_debug);
330}
331
332void ocfs2_lock_res_init_once(struct ocfs2_lock_res *res)
333{
334 /* This also clears out the lock status block */
335 memset(res, 0, sizeof(struct ocfs2_lock_res));
336 spin_lock_init(&res->l_lock);
337 init_waitqueue_head(&res->l_event);
338 INIT_LIST_HEAD(&res->l_blocked_list);
339 INIT_LIST_HEAD(&res->l_mask_waiters);
340}
341
342void ocfs2_inode_lock_res_init(struct ocfs2_lock_res *res,
343 enum ocfs2_lock_type type,
24c19ef4 344 unsigned int generation,
ccd979bd
MF		 345 struct inode *inode)
346{
347 struct ocfs2_lock_res_ops *ops;
348
349 switch(type) {
350 case OCFS2_LOCK_TYPE_RW:
351 ops = &ocfs2_inode_rw_lops;
352 break;
353 case OCFS2_LOCK_TYPE_META:
354 ops = &ocfs2_inode_meta_lops;
355 break;
356 case OCFS2_LOCK_TYPE_DATA:
357 ops = &ocfs2_inode_data_lops;
358 break;
359 default:
360 mlog_bug_on_msg(1, "type: %d\n", type);
361 ops = NULL; /* thanks, gcc */
362 break;
363 };
364
d680efe9 365 ocfs2_build_lock_name(type, OCFS2_I(inode)->ip_blkno,
24c19ef4 366 generation, res->l_name);
d680efe9
MF		 367 ocfs2_lock_res_init_common(OCFS2_SB(inode->i_sb), res, type, ops, inode);
368}
369
54a7e755
MF		 370static struct ocfs2_super *ocfs2_get_inode_osb(struct ocfs2_lock_res *lockres)
371{
372 struct inode *inode = ocfs2_lock_res_inode(lockres);
373
374 return OCFS2_SB(inode->i_sb);
375}
376
d680efe9
MF		 377static __u64 ocfs2_get_dentry_lock_ino(struct ocfs2_lock_res *lockres)
378{
379 __be64 inode_blkno_be;
380
381 memcpy(&inode_blkno_be, &lockres->l_name[OCFS2_DENTRY_LOCK_INO_START],
382 sizeof(__be64));
383
384 return be64_to_cpu(inode_blkno_be);
385}
386
54a7e755
MF		 387static struct ocfs2_super *ocfs2_get_dentry_osb(struct ocfs2_lock_res *lockres)
388{
389 struct ocfs2_dentry_lock *dl = lockres->l_priv;
390
391 return OCFS2_SB(dl->dl_inode->i_sb);
392}
393
d680efe9
MF		 394void ocfs2_dentry_lock_res_init(struct ocfs2_dentry_lock *dl,
395 u64 parent, struct inode *inode)
396{
397 int len;
398 u64 inode_blkno = OCFS2_I(inode)->ip_blkno;
399 __be64 inode_blkno_be = cpu_to_be64(inode_blkno);
400 struct ocfs2_lock_res *lockres = &dl->dl_lockres;
401
402 ocfs2_lock_res_init_once(lockres);
403
404 /*
405 * Unfortunately, the standard lock naming scheme won't work
406 * here because we have two 16 byte values to use. Instead,
407 * we'll stuff the inode number as a binary value. We still
408 * want error prints to show something without garbling the
409 * display, so drop a null byte in there before the inode
410 * number. A future version of OCFS2 will likely use all
411 * binary lock names. The stringified names have been a
412 * tremendous aid in debugging, but now that the debugfs
413 * interface exists, we can mangle things there if need be.
414 *
415 * NOTE: We also drop the standard "pad" value (the total lock
416 * name size stays the same though - the last part is all
417 * zeros due to the memset in ocfs2_lock_res_init_once()
418 */
419 len = snprintf(lockres->l_name, OCFS2_DENTRY_LOCK_INO_START,
420 "%c%016llx",
421 ocfs2_lock_type_char(OCFS2_LOCK_TYPE_DENTRY),
422 (long long)parent);
423
424 BUG_ON(len != (OCFS2_DENTRY_LOCK_INO_START - 1));
425
426 memcpy(&lockres->l_name[OCFS2_DENTRY_LOCK_INO_START], &inode_blkno_be,
427 sizeof(__be64));
428
429 ocfs2_lock_res_init_common(OCFS2_SB(inode->i_sb), lockres,
430 OCFS2_LOCK_TYPE_DENTRY, &ocfs2_dentry_lops,
431 dl);
ccd979bd
MF		 432}
433
434static void ocfs2_super_lock_res_init(struct ocfs2_lock_res *res,
435 struct ocfs2_super *osb)
436{
437 /* Superblock lockres doesn't come from a slab so we call init
438 * once on it manually. */
439 ocfs2_lock_res_init_once(res);
d680efe9
MF		 440 ocfs2_build_lock_name(OCFS2_LOCK_TYPE_SUPER, OCFS2_SUPER_BLOCK_BLKNO,
441 0, res->l_name);
ccd979bd 442 ocfs2_lock_res_init_common(osb, res, OCFS2_LOCK_TYPE_SUPER,
ccd979bd
MF		 443 &ocfs2_super_lops, osb);
444}
445
446static void ocfs2_rename_lock_res_init(struct ocfs2_lock_res *res,
447 struct ocfs2_super *osb)
448{
449 /* Rename lockres doesn't come from a slab so we call init
450 * once on it manually. */
451 ocfs2_lock_res_init_once(res);
d680efe9
MF		 452 ocfs2_build_lock_name(OCFS2_LOCK_TYPE_RENAME, 0, 0, res->l_name);
453 ocfs2_lock_res_init_common(osb, res, OCFS2_LOCK_TYPE_RENAME,
ccd979bd
MF		 454 &ocfs2_rename_lops, osb);
455}
456
457void ocfs2_lock_res_free(struct ocfs2_lock_res *res)
458{
459 mlog_entry_void();
460
461 if (!(res->l_flags & OCFS2_LOCK_INITIALIZED))
462 return;
463
464 ocfs2_remove_lockres_tracking(res);
465
466 mlog_bug_on_msg(!list_empty(&res->l_blocked_list),
467 "Lockres %s is on the blocked list\n",
468 res->l_name);
469 mlog_bug_on_msg(!list_empty(&res->l_mask_waiters),
470 "Lockres %s has mask waiters pending\n",
471 res->l_name);
472 mlog_bug_on_msg(spin_is_locked(&res->l_lock),
473 "Lockres %s is locked\n",
474 res->l_name);
475 mlog_bug_on_msg(res->l_ro_holders,
476 "Lockres %s has %u ro holders\n",
477 res->l_name, res->l_ro_holders);
478 mlog_bug_on_msg(res->l_ex_holders,
479 "Lockres %s has %u ex holders\n",
480 res->l_name, res->l_ex_holders);
481
482 /* Need to clear out the lock status block for the dlm */
483 memset(&res->l_lksb, 0, sizeof(res->l_lksb));
484
485 res->l_flags = 0UL;
486 mlog_exit_void();
487}
488
489static inline void ocfs2_inc_holders(struct ocfs2_lock_res *lockres,
490 int level)
491{
492 mlog_entry_void();
493
494 BUG_ON(!lockres);
495
496 switch(level) {
497 case LKM_EXMODE:
498 lockres->l_ex_holders++;
499 break;
500 case LKM_PRMODE:
501 lockres->l_ro_holders++;
502 break;
503 default:
504 BUG();
505 }
506
507 mlog_exit_void();
508}
509
510static inline void ocfs2_dec_holders(struct ocfs2_lock_res *lockres,
511 int level)
512{
513 mlog_entry_void();
514
515 BUG_ON(!lockres);
516
517 switch(level) {
518 case LKM_EXMODE:
519 BUG_ON(!lockres->l_ex_holders);
520 lockres->l_ex_holders--;
521 break;
522 case LKM_PRMODE:
523 BUG_ON(!lockres->l_ro_holders);
524 lockres->l_ro_holders--;
525 break;
526 default:
527 BUG();
528 }
529 mlog_exit_void();
530}
531
532/* WARNING: This function lives in a world where the only three lock
533 * levels are EX, PR, and NL. It *will* have to be adjusted when more
534 * lock types are added. */
535static inline int ocfs2_highest_compat_lock_level(int level)
536{
537 int new_level = LKM_EXMODE;
538
539 if (level == LKM_EXMODE)
540 new_level = LKM_NLMODE;
541 else if (level == LKM_PRMODE)
542 new_level = LKM_PRMODE;
543 return new_level;
544}
545
546static void lockres_set_flags(struct ocfs2_lock_res *lockres,
547 unsigned long newflags)
548{
549 struct list_head *pos, *tmp;
550 struct ocfs2_mask_waiter *mw;
551
552 assert_spin_locked(&lockres->l_lock);
553
554 lockres->l_flags = newflags;
555
556 list_for_each_safe(pos, tmp, &lockres->l_mask_waiters) {
557 mw = list_entry(pos, struct ocfs2_mask_waiter, mw_item);
558 if ((lockres->l_flags & mw->mw_mask) != mw->mw_goal)
559 continue;
560
561 list_del_init(&mw->mw_item);
562 mw->mw_status = 0;
563 complete(&mw->mw_complete);
564 }
565}
566static void lockres_or_flags(struct ocfs2_lock_res *lockres, unsigned long or)
567{
568 lockres_set_flags(lockres, lockres->l_flags | or);
569}
570static void lockres_clear_flags(struct ocfs2_lock_res *lockres,
571 unsigned long clear)
572{
573 lockres_set_flags(lockres, lockres->l_flags & ~clear);
574}
575
576static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res *lockres)
577{
578 mlog_entry_void();
579
580 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BUSY));
581 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_ATTACHED));
582 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));
583 BUG_ON(lockres->l_blocking <= LKM_NLMODE);
584
585 lockres->l_level = lockres->l_requested;
586 if (lockres->l_level <=
587 ocfs2_highest_compat_lock_level(lockres->l_blocking)) {
588 lockres->l_blocking = LKM_NLMODE;
589 lockres_clear_flags(lockres, OCFS2_LOCK_BLOCKED);
590 }
591 lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
592
593 mlog_exit_void();
594}
595
596static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lockres)
597{
598 mlog_entry_void();
599
600 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BUSY));
601 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_ATTACHED));
602
603 /* Convert from RO to EX doesn't really need anything as our
604 * information is already up to data. Convert from NL to
605 * *anything* however should mark ourselves as needing an
606 * update */
f625c979
MF		 607 if (lockres->l_level == LKM_NLMODE &&
608 lockres->l_ops->flags & LOCK_TYPE_REQUIRES_REFRESH)
ccd979bd
MF		 609 lockres_or_flags(lockres, OCFS2_LOCK_NEEDS_REFRESH);
610
611 lockres->l_level = lockres->l_requested;
612 lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
613
614 mlog_exit_void();
615}
616
617static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *lockres)
618{
619 mlog_entry_void();
620
621 BUG_ON((!lockres->l_flags & OCFS2_LOCK_BUSY));
622 BUG_ON(lockres->l_flags & OCFS2_LOCK_ATTACHED);
623
624 if (lockres->l_requested > LKM_NLMODE &&
f625c979
MF		 625 !(lockres->l_flags & OCFS2_LOCK_LOCAL) &&
626 lockres->l_ops->flags & LOCK_TYPE_REQUIRES_REFRESH)
ccd979bd
MF		 627 lockres_or_flags(lockres, OCFS2_LOCK_NEEDS_REFRESH);
628
629 lockres->l_level = lockres->l_requested;
630 lockres_or_flags(lockres, OCFS2_LOCK_ATTACHED);
631 lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
632
633 mlog_exit_void();
634}
635
ccd979bd
MF		 636static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres,
637 int level)
638{
639 int needs_downconvert = 0;
640 mlog_entry_void();
641
642 assert_spin_locked(&lockres->l_lock);
643
644 lockres_or_flags(lockres, OCFS2_LOCK_BLOCKED);
645
646 if (level > lockres->l_blocking) {
647 /* only schedule a downconvert if we haven't already scheduled
648 * one that goes low enough to satisfy the level we're
649 * blocking. this also catches the case where we get
650 * duplicate BASTs */
651 if (ocfs2_highest_compat_lock_level(level) <
652 ocfs2_highest_compat_lock_level(lockres->l_blocking))
653 needs_downconvert = 1;
654
655 lockres->l_blocking = level;
656 }
657
658 mlog_exit(needs_downconvert);
659 return needs_downconvert;
660}
661
aa2623ad 662static void ocfs2_blocking_ast(void *opaque, int level)
ccd979bd 663{
aa2623ad
MF		 664 struct ocfs2_lock_res *lockres = opaque;
665 struct ocfs2_super *osb = ocfs2_get_lockres_osb(lockres);
ccd979bd
MF		 666 int needs_downconvert;
667 unsigned long flags;
668
ccd979bd
MF		 669 BUG_ON(level <= LKM_NLMODE);
670
aa2623ad
MF		 671 mlog(0, "BAST fired for lockres %s, blocking %d, level %d type %s\n",
672 lockres->l_name, level, lockres->l_level,
673 ocfs2_lock_type_string(lockres->l_type));
674
ccd979bd
MF		 675 spin_lock_irqsave(&lockres->l_lock, flags);
676 needs_downconvert = ocfs2_generic_handle_bast(lockres, level);
677 if (needs_downconvert)
678 ocfs2_schedule_blocked_lock(osb, lockres);
679 spin_unlock_irqrestore(&lockres->l_lock, flags);
680
d680efe9
MF		 681 wake_up(&lockres->l_event);
682
ccd979bd 683 ocfs2_kick_vote_thread(osb);
ccd979bd
MF		 684}
685
e92d57df 686static void ocfs2_locking_ast(void *opaque)
ccd979bd 687{
e92d57df 688 struct ocfs2_lock_res *lockres = opaque;
ccd979bd
MF		 689 struct dlm_lockstatus *lksb = &lockres->l_lksb;
690 unsigned long flags;
691
692 spin_lock_irqsave(&lockres->l_lock, flags);
693
694 if (lksb->status != DLM_NORMAL) {
695 mlog(ML_ERROR, "lockres %s: lksb status value of %u!\n",
696 lockres->l_name, lksb->status);
697 spin_unlock_irqrestore(&lockres->l_lock, flags);
698 return;
699 }
700
701 switch(lockres->l_action) {
702 case OCFS2_AST_ATTACH:
703 ocfs2_generic_handle_attach_action(lockres);
e92d57df 704 lockres_clear_flags(lockres, OCFS2_LOCK_LOCAL);
ccd979bd
MF		 705 break;
706 case OCFS2_AST_CONVERT:
707 ocfs2_generic_handle_convert_action(lockres);
708 break;
709 case OCFS2_AST_DOWNCONVERT:
710 ocfs2_generic_handle_downconvert_action(lockres);
711 break;
712 default:
e92d57df
MF		 713 mlog(ML_ERROR, "lockres %s: ast fired with invalid action: %u "
714 "lockres flags = 0x%lx, unlock action: %u\n",
715 lockres->l_name, lockres->l_action, lockres->l_flags,
716 lockres->l_unlock_action);
ccd979bd
MF		 717 BUG();
718 }
719
ccd979bd
MF		 720 /* set it to something invalid so if we get called again we
721 * can catch it. */
722 lockres->l_action = OCFS2_AST_INVALID;
ccd979bd
MF		 723
724 wake_up(&lockres->l_event);
d680efe9 725 spin_unlock_irqrestore(&lockres->l_lock, flags);
ccd979bd
MF		 726}
727
ccd979bd
MF		 728static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres,
729 int convert)
730{
731 unsigned long flags;
732
733 mlog_entry_void();
734 spin_lock_irqsave(&lockres->l_lock, flags);
735 lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
736 if (convert)
737 lockres->l_action = OCFS2_AST_INVALID;
738 else
739 lockres->l_unlock_action = OCFS2_UNLOCK_INVALID;
740 spin_unlock_irqrestore(&lockres->l_lock, flags);
741
742 wake_up(&lockres->l_event);
743 mlog_exit_void();
744}
745
746/* Note: If we detect another process working on the lock (i.e.,
747 * OCFS2_LOCK_BUSY), we'll bail out returning 0. It's up to the caller
748 * to do the right thing in that case.
749 */
750static int ocfs2_lock_create(struct ocfs2_super *osb,
751 struct ocfs2_lock_res *lockres,
752 int level,
753 int dlm_flags)
754{
755 int ret = 0;
756 enum dlm_status status;
757 unsigned long flags;
758
759 mlog_entry_void();
760
761 mlog(0, "lock %s, level = %d, flags = %d\n", lockres->l_name, level,
762 dlm_flags);
763
764 spin_lock_irqsave(&lockres->l_lock, flags);
765 if ((lockres->l_flags & OCFS2_LOCK_ATTACHED) ||
766 (lockres->l_flags & OCFS2_LOCK_BUSY)) {
767 spin_unlock_irqrestore(&lockres->l_lock, flags);
768 goto bail;
769 }
770
771 lockres->l_action = OCFS2_AST_ATTACH;
772 lockres->l_requested = level;
773 lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
774 spin_unlock_irqrestore(&lockres->l_lock, flags);
775
776 status = dlmlock(osb->dlm,
777 level,
778 &lockres->l_lksb,
779 dlm_flags,
780 lockres->l_name,
f0681062 781 OCFS2_LOCK_ID_MAX_LEN - 1,
e92d57df 782 ocfs2_locking_ast,
ccd979bd 783 lockres,
aa2623ad 784 ocfs2_blocking_ast);
ccd979bd
MF		 785 if (status != DLM_NORMAL) {
786 ocfs2_log_dlm_error("dlmlock", status, lockres);
787 ret = -EINVAL;
788 ocfs2_recover_from_dlm_error(lockres, 1);
789 }
790
791 mlog(0, "lock %s, successfull return from dlmlock\n", lockres->l_name);
792
793bail:
794 mlog_exit(ret);
795 return ret;
796}
797
798static inline int ocfs2_check_wait_flag(struct ocfs2_lock_res *lockres,
799 int flag)
800{
801 unsigned long flags;
802 int ret;
803
804 spin_lock_irqsave(&lockres->l_lock, flags);
805 ret = lockres->l_flags & flag;
806 spin_unlock_irqrestore(&lockres->l_lock, flags);
807
808 return ret;
809}
810
811static inline void ocfs2_wait_on_busy_lock(struct ocfs2_lock_res *lockres)
812
813{
814 wait_event(lockres->l_event,
815 !ocfs2_check_wait_flag(lockres, OCFS2_LOCK_BUSY));
816}
817
818static inline void ocfs2_wait_on_refreshing_lock(struct ocfs2_lock_res *lockres)
819
820{
821 wait_event(lockres->l_event,
822 !ocfs2_check_wait_flag(lockres, OCFS2_LOCK_REFRESHING));
823}
824
825/* predict what lock level we'll be dropping down to on behalf
826 * of another node, and return true if the currently wanted
827 * level will be compatible with it. */
828static inline int ocfs2_may_continue_on_blocked_lock(struct ocfs2_lock_res *lockres,
829 int wanted)
830{
831 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));
832
833 return wanted <= ocfs2_highest_compat_lock_level(lockres->l_blocking);
834}
835
836static void ocfs2_init_mask_waiter(struct ocfs2_mask_waiter *mw)
837{
838 INIT_LIST_HEAD(&mw->mw_item);
839 init_completion(&mw->mw_complete);
840}
841
842static int ocfs2_wait_for_mask(struct ocfs2_mask_waiter *mw)
843{
844 wait_for_completion(&mw->mw_complete);
845 /* Re-arm the completion in case we want to wait on it again */
846 INIT_COMPLETION(mw->mw_complete);
847 return mw->mw_status;
848}
849
850static void lockres_add_mask_waiter(struct ocfs2_lock_res *lockres,
851 struct ocfs2_mask_waiter *mw,
852 unsigned long mask,
853 unsigned long goal)
854{
855 BUG_ON(!list_empty(&mw->mw_item));
856
857 assert_spin_locked(&lockres->l_lock);
858
859 list_add_tail(&mw->mw_item, &lockres->l_mask_waiters);
860 mw->mw_mask = mask;
861 mw->mw_goal = goal;
862}
863
864/* returns 0 if the mw that was removed was already satisfied, -EBUSY
865 * if the mask still hadn't reached its goal */
866static int lockres_remove_mask_waiter(struct ocfs2_lock_res *lockres,
867 struct ocfs2_mask_waiter *mw)
868{
869 unsigned long flags;
870 int ret = 0;
871
872 spin_lock_irqsave(&lockres->l_lock, flags);
873 if (!list_empty(&mw->mw_item)) {
874 if ((lockres->l_flags & mw->mw_mask) != mw->mw_goal)
875 ret = -EBUSY;
876
877 list_del_init(&mw->mw_item);
878 init_completion(&mw->mw_complete);
879 }
880 spin_unlock_irqrestore(&lockres->l_lock, flags);
881
882 return ret;
883
884}
885
886static int ocfs2_cluster_lock(struct ocfs2_super *osb,
887 struct ocfs2_lock_res *lockres,
888 int level,
889 int lkm_flags,
890 int arg_flags)
891{
892 struct ocfs2_mask_waiter mw;
893 enum dlm_status status;
894 int wait, catch_signals = !(osb->s_mount_opt & OCFS2_MOUNT_NOINTR);
895 int ret = 0; /* gcc doesn't realize wait = 1 guarantees ret is set */
896 unsigned long flags;
897
898 mlog_entry_void();
899
900 ocfs2_init_mask_waiter(&mw);
901
b80fc012
MF		 902 if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB)
903 lkm_flags |= LKM_VALBLK;
904
ccd979bd
MF		 905again:
906 wait = 0;
907
908 if (catch_signals && signal_pending(current)) {
909 ret = -ERESTARTSYS;
910 goto out;
911 }
912
913 spin_lock_irqsave(&lockres->l_lock, flags);
914
915 mlog_bug_on_msg(lockres->l_flags & OCFS2_LOCK_FREEING,
916 "Cluster lock called on freeing lockres %s! flags "
917 "0x%lx\n", lockres->l_name, lockres->l_flags);
918
919 /* We only compare against the currently granted level
920 * here. If the lock is blocked waiting on a downconvert,
921 * we'll get caught below. */
922 if (lockres->l_flags & OCFS2_LOCK_BUSY &&
923 level > lockres->l_level) {
924 /* is someone sitting in dlm_lock? If so, wait on
925 * them. */
926 lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BUSY, 0);
927 wait = 1;
928 goto unlock;
929 }
930
931 if (!(lockres->l_flags & OCFS2_LOCK_ATTACHED)) {
932 /* lock has not been created yet. */
933 spin_unlock_irqrestore(&lockres->l_lock, flags);
934
935 ret = ocfs2_lock_create(osb, lockres, LKM_NLMODE, 0);
936 if (ret < 0) {
937 mlog_errno(ret);
938 goto out;
939 }
940 goto again;
941 }
942
943 if (lockres->l_flags & OCFS2_LOCK_BLOCKED &&
944 !ocfs2_may_continue_on_blocked_lock(lockres, level)) {
945 /* is the lock is currently blocked on behalf of
946 * another node */
947 lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BLOCKED, 0);
948 wait = 1;
949 goto unlock;
950 }
951
952 if (level > lockres->l_level) {
953 if (lockres->l_action != OCFS2_AST_INVALID)
954 mlog(ML_ERROR, "lockres %s has action %u pending\n",
955 lockres->l_name, lockres->l_action);
956
957 lockres->l_action = OCFS2_AST_CONVERT;
958 lockres->l_requested = level;
959 lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
960 spin_unlock_irqrestore(&lockres->l_lock, flags);
961
962 BUG_ON(level == LKM_IVMODE);
963 BUG_ON(level == LKM_NLMODE);
964
965 mlog(0, "lock %s, convert from %d to level = %d\n",
966 lockres->l_name, lockres->l_level, level);
967
968 /* call dlm_lock to upgrade lock now */
969 status = dlmlock(osb->dlm,
970 level,
971 &lockres->l_lksb,
b80fc012 972 lkm_flags|LKM_CONVERT,
ccd979bd 973 lockres->l_name,
f0681062 974 OCFS2_LOCK_ID_MAX_LEN - 1,
e92d57df 975 ocfs2_locking_ast,
ccd979bd 976 lockres,
aa2623ad 977 ocfs2_blocking_ast);
ccd979bd
MF		 978 if (status != DLM_NORMAL) {
979 if ((lkm_flags & LKM_NOQUEUE) &&
980 (status == DLM_NOTQUEUED))
981 ret = -EAGAIN;
982 else {
983 ocfs2_log_dlm_error("dlmlock", status,
984 lockres);
985 ret = -EINVAL;
986 }
987 ocfs2_recover_from_dlm_error(lockres, 1);
988 goto out;
989 }
990
991 mlog(0, "lock %s, successfull return from dlmlock\n",
992 lockres->l_name);
993
994 /* At this point we've gone inside the dlm and need to
995 * complete our work regardless. */
996 catch_signals = 0;
997
998 /* wait for busy to clear and carry on */
999 goto again;
1000 }
1001
1002 /* Ok, if we get here then we're good to go. */
1003 ocfs2_inc_holders(lockres, level);
1004
1005 ret = 0;
1006unlock:
1007 spin_unlock_irqrestore(&lockres->l_lock, flags);
1008out:
1009 /*
1010 * This is helping work around a lock inversion between the page lock
1011 * and dlm locks. One path holds the page lock while calling aops
1012 * which block acquiring dlm locks. The voting thread holds dlm
1013 * locks while acquiring page locks while down converting data locks.
1014 * This block is helping an aop path notice the inversion and back
1015 * off to unlock its page lock before trying the dlm lock again.
1016 */
1017 if (wait && arg_flags & OCFS2_LOCK_NONBLOCK &&
1018 mw.mw_mask & (OCFS2_LOCK_BUSY|OCFS2_LOCK_BLOCKED)) {
1019 wait = 0;
1020 if (lockres_remove_mask_waiter(lockres, &mw))
1021 ret = -EAGAIN;
1022 else
1023 goto again;
1024 }
1025 if (wait) {
1026 ret = ocfs2_wait_for_mask(&mw);
1027 if (ret == 0)
1028 goto again;
1029 mlog_errno(ret);
1030 }
1031
1032 mlog_exit(ret);
1033 return ret;
1034}
1035
1036static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
1037 struct ocfs2_lock_res *lockres,
1038 int level)
1039{
1040 unsigned long flags;
1041
1042 mlog_entry_void();
1043 spin_lock_irqsave(&lockres->l_lock, flags);
1044 ocfs2_dec_holders(lockres, level);
1045 ocfs2_vote_on_unlock(osb, lockres);
1046 spin_unlock_irqrestore(&lockres->l_lock, flags);
1047 mlog_exit_void();
1048}
1049
d680efe9
MF		 1050int ocfs2_create_new_lock(struct ocfs2_super *osb,
1051 struct ocfs2_lock_res *lockres,
24c19ef4
MF		 1052 int ex,
1053 int local)
ccd979bd 1054{
d680efe9 1055 int level = ex ? LKM_EXMODE : LKM_PRMODE;
ccd979bd 1056 unsigned long flags;
24c19ef4 1057 int lkm_flags = local ? LKM_LOCAL : 0;
ccd979bd
MF		 1058
1059 spin_lock_irqsave(&lockres->l_lock, flags);
1060 BUG_ON(lockres->l_flags & OCFS2_LOCK_ATTACHED);
1061 lockres_or_flags(lockres, OCFS2_LOCK_LOCAL);
1062 spin_unlock_irqrestore(&lockres->l_lock, flags);
1063
24c19ef4 1064 return ocfs2_lock_create(osb, lockres, level, lkm_flags);
ccd979bd
MF		 1065}
1066
1067/* Grants us an EX lock on the data and metadata resources, skipping
1068 * the normal cluster directory lookup. Use this ONLY on newly created
1069 * inodes which other nodes can't possibly see, and which haven't been
1070 * hashed in the inode hash yet. This can give us a good performance
1071 * increase as it'll skip the network broadcast normally associated
1072 * with creating a new lock resource. */
1073int ocfs2_create_new_inode_locks(struct inode *inode)
1074{
1075 int ret;
d680efe9 1076 struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
ccd979bd
MF		 1077
1078 BUG_ON(!inode);
1079 BUG_ON(!ocfs2_inode_is_new(inode));
1080
1081 mlog_entry_void();
1082
b0697053 1083 mlog(0, "Inode %llu\n", (unsigned long long)OCFS2_I(inode)->ip_blkno);
ccd979bd
MF		 1084
1085 /* NOTE: That we don't increment any of the holder counts, nor
1086 * do we add anything to a journal handle. Since this is
1087 * supposed to be a new inode which the cluster doesn't know
1088 * about yet, there is no need to. As far as the LVB handling
1089 * is concerned, this is basically like acquiring an EX lock
1090 * on a resource which has an invalid one -- we'll set it
1091 * valid when we release the EX. */
1092
24c19ef4 1093 ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_rw_lockres, 1, 1);
ccd979bd
MF		 1094 if (ret) {
1095 mlog_errno(ret);
1096 goto bail;
1097 }
1098
24c19ef4
MF		 1099 /*
1100 * We don't want to use LKM_LOCAL on a meta data lock as they
1101 * don't use a generation in their lock names.
1102 */
1103 ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_meta_lockres, 1, 0);
ccd979bd
MF		 1104 if (ret) {
1105 mlog_errno(ret);
1106 goto bail;
1107 }
1108
24c19ef4 1109 ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_data_lockres, 1, 1);
ccd979bd
MF		 1110 if (ret) {
1111 mlog_errno(ret);
1112 goto bail;
1113 }
1114
1115bail:
1116 mlog_exit(ret);
1117 return ret;
1118}
1119
1120int ocfs2_rw_lock(struct inode *inode, int write)
1121{
1122 int status, level;
1123 struct ocfs2_lock_res *lockres;
1124
1125 BUG_ON(!inode);
1126
1127 mlog_entry_void();
1128
b0697053
MF		 1129 mlog(0, "inode %llu take %s RW lock\n",
1130 (unsigned long long)OCFS2_I(inode)->ip_blkno,
ccd979bd
MF		 1131 write ? "EXMODE" : "PRMODE");
1132
1133 lockres = &OCFS2_I(inode)->ip_rw_lockres;
1134
1135 level = write ? LKM_EXMODE : LKM_PRMODE;
1136
1137 status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, level, 0,
1138 0);
1139 if (status < 0)
1140 mlog_errno(status);
1141
1142 mlog_exit(status);
1143 return status;
1144}
1145
1146void ocfs2_rw_unlock(struct inode *inode, int write)
1147{
1148 int level = write ? LKM_EXMODE : LKM_PRMODE;
1149 struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_rw_lockres;
1150
1151 mlog_entry_void();
1152
b0697053
MF		 1153 mlog(0, "inode %llu drop %s RW lock\n",
1154 (unsigned long long)OCFS2_I(inode)->ip_blkno,
ccd979bd
MF		 1155 write ? "EXMODE" : "PRMODE");
1156
1157 ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);
1158
1159 mlog_exit_void();
1160}
1161
1162int ocfs2_data_lock_full(struct inode *inode,
1163 int write,
1164 int arg_flags)
1165{
1166 int status = 0, level;
1167 struct ocfs2_lock_res *lockres;
1168
1169 BUG_ON(!inode);
1170
1171 mlog_entry_void();
1172
b0697053
MF		 1173 mlog(0, "inode %llu take %s DATA lock\n",
1174 (unsigned long long)OCFS2_I(inode)->ip_blkno,
ccd979bd
MF		 1175 write ? "EXMODE" : "PRMODE");
1176
1177 /* We'll allow faking a readonly data lock for
1178 * rodevices. */
1179 if (ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb))) {
1180 if (write) {
1181 status = -EROFS;
1182 mlog_errno(status);
1183 }
1184 goto out;
1185 }
1186
1187 lockres = &OCFS2_I(inode)->ip_data_lockres;
1188
1189 level = write ? LKM_EXMODE : LKM_PRMODE;
1190
1191 status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, level,
1192 0, arg_flags);
1193 if (status < 0 && status != -EAGAIN)
1194 mlog_errno(status);
1195
1196out:
1197 mlog_exit(status);
1198 return status;
1199}
1200
1201/* see ocfs2_meta_lock_with_page() */
1202int ocfs2_data_lock_with_page(struct inode *inode,
1203 int write,
1204 struct page *page)
1205{
1206 int ret;
1207
1208 ret = ocfs2_data_lock_full(inode, write, OCFS2_LOCK_NONBLOCK);
1209 if (ret == -EAGAIN) {
1210 unlock_page(page);
1211 if (ocfs2_data_lock(inode, write) == 0)
1212 ocfs2_data_unlock(inode, write);
1213 ret = AOP_TRUNCATED_PAGE;
1214 }
1215
1216 return ret;
1217}
1218
1219static void ocfs2_vote_on_unlock(struct ocfs2_super *osb,
1220 struct ocfs2_lock_res *lockres)
1221{
1222 int kick = 0;
1223
1224 mlog_entry_void();
1225
1226 /* If we know that another node is waiting on our lock, kick
1227 * the vote thread * pre-emptively when we reach a release
1228 * condition. */
1229 if (lockres->l_flags & OCFS2_LOCK_BLOCKED) {
1230 switch(lockres->l_blocking) {
1231 case LKM_EXMODE:
1232 if (!lockres->l_ex_holders && !lockres->l_ro_holders)
1233 kick = 1;
1234 break;
1235 case LKM_PRMODE:
1236 if (!lockres->l_ex_holders)
1237 kick = 1;
1238 break;
1239 default:
1240 BUG();
1241 }
1242 }
1243
1244 if (kick)
1245 ocfs2_kick_vote_thread(osb);
1246
1247 mlog_exit_void();
1248}
1249
1250void ocfs2_data_unlock(struct inode *inode,
1251 int write)
1252{
1253 int level = write ? LKM_EXMODE : LKM_PRMODE;
1254 struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_data_lockres;
1255
1256 mlog_entry_void();
1257
b0697053
MF		 1258 mlog(0, "inode %llu drop %s DATA lock\n",
1259 (unsigned long long)OCFS2_I(inode)->ip_blkno,
ccd979bd
MF		 1260 write ? "EXMODE" : "PRMODE");
1261
1262 if (!ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb)))
1263 ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);
1264
1265 mlog_exit_void();
1266}
1267
1268#define OCFS2_SEC_BITS 34
1269#define OCFS2_SEC_SHIFT (64 - 34)
1270#define OCFS2_NSEC_MASK ((1ULL << OCFS2_SEC_SHIFT) - 1)
1271
1272/* LVB only has room for 64 bits of time here so we pack it for
1273 * now. */
1274static u64 ocfs2_pack_timespec(struct timespec *spec)
1275{
1276 u64 res;
1277 u64 sec = spec->tv_sec;
1278 u32 nsec = spec->tv_nsec;
1279
1280 res = (sec << OCFS2_SEC_SHIFT) | (nsec & OCFS2_NSEC_MASK);
1281
1282 return res;
1283}
1284
1285/* Call this with the lockres locked. I am reasonably sure we don't
1286 * need ip_lock in this function as anyone who would be changing those
1287 * values is supposed to be blocked in ocfs2_meta_lock right now. */
1288static void __ocfs2_stuff_meta_lvb(struct inode *inode)
1289{
1290 struct ocfs2_inode_info *oi = OCFS2_I(inode);
1291 struct ocfs2_lock_res *lockres = &oi->ip_meta_lockres;
1292 struct ocfs2_meta_lvb *lvb;
1293
1294 mlog_entry_void();
1295
1296 lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;
1297
24c19ef4
MF		 1298 /*
1299 * Invalidate the LVB of a deleted inode - this way other
1300 * nodes are forced to go to disk and discover the new inode
1301 * status.
1302 */
1303 if (oi->ip_flags & OCFS2_INODE_DELETED) {
1304 lvb->lvb_version = 0;
1305 goto out;
1306 }
1307
4d3b83f7 1308 lvb->lvb_version = OCFS2_LVB_VERSION;
ccd979bd
MF		 1309 lvb->lvb_isize = cpu_to_be64(i_size_read(inode));
1310 lvb->lvb_iclusters = cpu_to_be32(oi->ip_clusters);
1311 lvb->lvb_iuid = cpu_to_be32(inode->i_uid);
1312 lvb->lvb_igid = cpu_to_be32(inode->i_gid);
1313 lvb->lvb_imode = cpu_to_be16(inode->i_mode);
1314 lvb->lvb_inlink = cpu_to_be16(inode->i_nlink);
1315 lvb->lvb_iatime_packed =
1316 cpu_to_be64(ocfs2_pack_timespec(&inode->i_atime));
1317 lvb->lvb_ictime_packed =
1318 cpu_to_be64(ocfs2_pack_timespec(&inode->i_ctime));
1319 lvb->lvb_imtime_packed =
1320 cpu_to_be64(ocfs2_pack_timespec(&inode->i_mtime));
ca4d147e 1321 lvb->lvb_iattr = cpu_to_be32(oi->ip_attr);
f9e2d82e 1322 lvb->lvb_igeneration = cpu_to_be32(inode->i_generation);
ccd979bd 1323
24c19ef4 1324out:
ccd979bd
MF		 1325 mlog_meta_lvb(0, lockres);
1326
1327 mlog_exit_void();
1328}
1329
1330static void ocfs2_unpack_timespec(struct timespec *spec,
1331 u64 packed_time)
1332{
1333 spec->tv_sec = packed_time >> OCFS2_SEC_SHIFT;
1334 spec->tv_nsec = packed_time & OCFS2_NSEC_MASK;
1335}
1336
1337static void ocfs2_refresh_inode_from_lvb(struct inode *inode)
1338{
1339 struct ocfs2_inode_info *oi = OCFS2_I(inode);
1340 struct ocfs2_lock_res *lockres = &oi->ip_meta_lockres;
1341 struct ocfs2_meta_lvb *lvb;
1342
1343 mlog_entry_void();
1344
1345 mlog_meta_lvb(0, lockres);
1346
1347 lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;
1348
1349 /* We're safe here without the lockres lock... */
1350 spin_lock(&oi->ip_lock);
1351 oi->ip_clusters = be32_to_cpu(lvb->lvb_iclusters);
1352 i_size_write(inode, be64_to_cpu(lvb->lvb_isize));
1353
ca4d147e
HP		 1354 oi->ip_attr = be32_to_cpu(lvb->lvb_iattr);
1355 ocfs2_set_inode_flags(inode);
1356
ccd979bd
MF		 1357 /* fast-symlinks are a special case */
1358 if (S_ISLNK(inode->i_mode) && !oi->ip_clusters)
1359 inode->i_blocks = 0;
1360 else
1361 inode->i_blocks =
1362 ocfs2_align_bytes_to_sectors(i_size_read(inode));
1363
1364 inode->i_uid = be32_to_cpu(lvb->lvb_iuid);
1365 inode->i_gid = be32_to_cpu(lvb->lvb_igid);
1366 inode->i_mode = be16_to_cpu(lvb->lvb_imode);
1367 inode->i_nlink = be16_to_cpu(lvb->lvb_inlink);
1368 ocfs2_unpack_timespec(&inode->i_atime,
1369 be64_to_cpu(lvb->lvb_iatime_packed));
1370 ocfs2_unpack_timespec(&inode->i_mtime,
1371 be64_to_cpu(lvb->lvb_imtime_packed));
1372 ocfs2_unpack_timespec(&inode->i_ctime,
1373 be64_to_cpu(lvb->lvb_ictime_packed));
1374 spin_unlock(&oi->ip_lock);
1375
1376 mlog_exit_void();
1377}
1378
f9e2d82e
MF		 1379static inline int ocfs2_meta_lvb_is_trustable(struct inode *inode,
1380 struct ocfs2_lock_res *lockres)
ccd979bd
MF		 1381{
1382 struct ocfs2_meta_lvb *lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;
1383
f9e2d82e
MF		 1384 if (lvb->lvb_version == OCFS2_LVB_VERSION
1385 && be32_to_cpu(lvb->lvb_igeneration) == inode->i_generation)
ccd979bd
MF		 1386 return 1;
1387 return 0;
1388}
1389
1390/* Determine whether a lock resource needs to be refreshed, and
1391 * arbitrate who gets to refresh it.
1392 *
1393 * 0 means no refresh needed.
1394 *
1395 * > 0 means you need to refresh this and you MUST call
1396 * ocfs2_complete_lock_res_refresh afterwards. */
1397static int ocfs2_should_refresh_lock_res(struct ocfs2_lock_res *lockres)
1398{
1399 unsigned long flags;
1400 int status = 0;
1401
1402 mlog_entry_void();
1403
1404refresh_check:
1405 spin_lock_irqsave(&lockres->l_lock, flags);
1406 if (!(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH)) {
1407 spin_unlock_irqrestore(&lockres->l_lock, flags);
1408 goto bail;
1409 }
1410
1411 if (lockres->l_flags & OCFS2_LOCK_REFRESHING) {
1412 spin_unlock_irqrestore(&lockres->l_lock, flags);
1413
1414 ocfs2_wait_on_refreshing_lock(lockres);
1415 goto refresh_check;
1416 }
1417
1418 /* Ok, I'll be the one to refresh this lock. */
1419 lockres_or_flags(lockres, OCFS2_LOCK_REFRESHING);
1420 spin_unlock_irqrestore(&lockres->l_lock, flags);
1421
1422 status = 1;
1423bail:
1424 mlog_exit(status);
1425 return status;
1426}
1427
1428/* If status is non zero, I'll mark it as not being in refresh
1429 * anymroe, but i won't clear the needs refresh flag. */
1430static inline void ocfs2_complete_lock_res_refresh(struct ocfs2_lock_res *lockres,
1431 int status)
1432{
1433 unsigned long flags;
1434 mlog_entry_void();
1435
1436 spin_lock_irqsave(&lockres->l_lock, flags);
1437 lockres_clear_flags(lockres, OCFS2_LOCK_REFRESHING);
1438 if (!status)
1439 lockres_clear_flags(lockres, OCFS2_LOCK_NEEDS_REFRESH);
1440 spin_unlock_irqrestore(&lockres->l_lock, flags);
1441
1442 wake_up(&lockres->l_event);
1443
1444 mlog_exit_void();
1445}
1446
1447/* may or may not return a bh if it went to disk. */
1448static int ocfs2_meta_lock_update(struct inode *inode,
1449 struct buffer_head **bh)
1450{
1451 int status = 0;
1452 struct ocfs2_inode_info *oi = OCFS2_I(inode);
1453 struct ocfs2_lock_res *lockres;
1454 struct ocfs2_dinode *fe;
1455
1456 mlog_entry_void();
1457
1458 spin_lock(&oi->ip_lock);
1459 if (oi->ip_flags & OCFS2_INODE_DELETED) {
b0697053 1460 mlog(0, "Orphaned inode %llu was deleted while we "
ccd979bd 1461 "were waiting on a lock. ip_flags = 0x%x\n",
b0697053 1462 (unsigned long long)oi->ip_blkno, oi->ip_flags);
ccd979bd
MF		 1463 spin_unlock(&oi->ip_lock);
1464 status = -ENOENT;
1465 goto bail;
1466 }
1467 spin_unlock(&oi->ip_lock);
1468
1469 lockres = &oi->ip_meta_lockres;
1470
1471 if (!ocfs2_should_refresh_lock_res(lockres))
1472 goto bail;
1473
1474 /* This will discard any caching information we might have had
1475 * for the inode metadata. */
1476 ocfs2_metadata_cache_purge(inode);
1477
1478 /* will do nothing for inode types that don't use the extent
1479 * map (directories, bitmap files, etc) */
1480 ocfs2_extent_map_trunc(inode, 0);
1481
f9e2d82e 1482 if (ocfs2_meta_lvb_is_trustable(inode, lockres)) {
b0697053
MF		 1483 mlog(0, "Trusting LVB on inode %llu\n",
1484 (unsigned long long)oi->ip_blkno);
ccd979bd
MF		 1485 ocfs2_refresh_inode_from_lvb(inode);
1486 } else {
1487 /* Boo, we have to go to disk. */
1488 /* read bh, cast, ocfs2_refresh_inode */
1489 status = ocfs2_read_block(OCFS2_SB(inode->i_sb), oi->ip_blkno,
1490 bh, OCFS2_BH_CACHED, inode);
1491 if (status < 0) {
1492 mlog_errno(status);
1493 goto bail_refresh;
1494 }
1495 fe = (struct ocfs2_dinode *) (*bh)->b_data;
1496
1497 /* This is a good chance to make sure we're not
1498 * locking an invalid object.
1499 *
1500 * We bug on a stale inode here because we checked
1501 * above whether it was wiped from disk. The wiping
1502 * node provides a guarantee that we receive that
1503 * message and can mark the inode before dropping any
1504 * locks associated with it. */
1505 if (!OCFS2_IS_VALID_DINODE(fe)) {
1506 OCFS2_RO_ON_INVALID_DINODE(inode->i_sb, fe);
1507 status = -EIO;
1508 goto bail_refresh;
1509 }
1510 mlog_bug_on_msg(inode->i_generation !=
1511 le32_to_cpu(fe->i_generation),
b0697053 1512 "Invalid dinode %llu disk generation: %u "
ccd979bd 1513 "inode->i_generation: %u\n",
b0697053
MF		 1514 (unsigned long long)oi->ip_blkno,
1515 le32_to_cpu(fe->i_generation),
ccd979bd
MF		 1516 inode->i_generation);
1517 mlog_bug_on_msg(le64_to_cpu(fe->i_dtime) ||
1518 !(fe->i_flags & cpu_to_le32(OCFS2_VALID_FL)),
b0697053
MF		 1519 "Stale dinode %llu dtime: %llu flags: 0x%x\n",
1520 (unsigned long long)oi->ip_blkno,
1521 (unsigned long long)le64_to_cpu(fe->i_dtime),
ccd979bd
MF		 1522 le32_to_cpu(fe->i_flags));
1523
1524 ocfs2_refresh_inode(inode, fe);
1525 }
1526
1527 status = 0;
1528bail_refresh:
1529 ocfs2_complete_lock_res_refresh(lockres, status);
1530bail:
1531 mlog_exit(status);
1532 return status;
1533}
1534
1535static int ocfs2_assign_bh(struct inode *inode,
1536 struct buffer_head **ret_bh,
1537 struct buffer_head *passed_bh)
1538{
1539 int status;
1540
1541 if (passed_bh) {
1542 /* Ok, the update went to disk for us, use the
1543 * returned bh. */
1544 *ret_bh = passed_bh;
1545 get_bh(*ret_bh);
1546
1547 return 0;
1548 }
1549
1550 status = ocfs2_read_block(OCFS2_SB(inode->i_sb),
1551 OCFS2_I(inode)->ip_blkno,
1552 ret_bh,
1553 OCFS2_BH_CACHED,
1554 inode);
1555 if (status < 0)
1556 mlog_errno(status);
1557
1558 return status;
1559}
1560
1561/*
1562 * returns < 0 error if the callback will never be called, otherwise
1563 * the result of the lock will be communicated via the callback.
1564 */
1565int ocfs2_meta_lock_full(struct inode *inode,
1566 struct ocfs2_journal_handle *handle,
1567 struct buffer_head **ret_bh,
1568 int ex,
1569 int arg_flags)
1570{
1571 int status, level, dlm_flags, acquired;
1572 struct ocfs2_lock_res *lockres;
1573 struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
1574 struct buffer_head *local_bh = NULL;
1575
1576 BUG_ON(!inode);
1577
1578 mlog_entry_void();
1579
b0697053
MF		 1580 mlog(0, "inode %llu, take %s META lock\n",
1581 (unsigned long long)OCFS2_I(inode)->ip_blkno,
ccd979bd
MF		 1582 ex ? "EXMODE" : "PRMODE");
1583
1584 status = 0;
1585 acquired = 0;
1586 /* We'll allow faking a readonly metadata lock for
1587 * rodevices. */
1588 if (ocfs2_is_hard_readonly(osb)) {
1589 if (ex)
1590 status = -EROFS;
1591 goto bail;
1592 }
1593
1594 if (!(arg_flags & OCFS2_META_LOCK_RECOVERY))
1595 wait_event(osb->recovery_event,
1596 ocfs2_node_map_is_empty(osb, &osb->recovery_map));
1597
1598 acquired = 0;
1599 lockres = &OCFS2_I(inode)->ip_meta_lockres;
1600 level = ex ? LKM_EXMODE : LKM_PRMODE;
1601 dlm_flags = 0;
1602 if (arg_flags & OCFS2_META_LOCK_NOQUEUE)
1603 dlm_flags |= LKM_NOQUEUE;
1604
1605 status = ocfs2_cluster_lock(osb, lockres, level, dlm_flags, arg_flags);
1606 if (status < 0) {
1607 if (status != -EAGAIN && status != -EIOCBRETRY)
1608 mlog_errno(status);
1609 goto bail;
1610 }
1611
1612 /* Notify the error cleanup path to drop the cluster lock. */
1613 acquired = 1;
1614
1615 /* We wait twice because a node may have died while we were in
1616 * the lower dlm layers. The second time though, we've
1617 * committed to owning this lock so we don't allow signals to
1618 * abort the operation. */
1619 if (!(arg_flags & OCFS2_META_LOCK_RECOVERY))
1620 wait_event(osb->recovery_event,
1621 ocfs2_node_map_is_empty(osb, &osb->recovery_map));
1622
24c19ef4
MF		 1623 /*
1624 * We only see this flag if we're being called from
1625 * ocfs2_read_locked_inode(). It means we're locking an inode
1626 * which hasn't been populated yet, so clear the refresh flag
1627 * and let the caller handle it.
1628 */
1629 if (inode->i_state & I_NEW) {
1630 status = 0;
1631 ocfs2_complete_lock_res_refresh(lockres, 0);
1632 goto bail;
1633 }
1634
ccd979bd
MF		 1635 /* This is fun. The caller may want a bh back, or it may
1636 * not. ocfs2_meta_lock_update definitely wants one in, but
1637 * may or may not read one, depending on what's in the
1638 * LVB. The result of all of this is that we've *only* gone to
1639 * disk if we have to, so the complexity is worthwhile. */
1640 status = ocfs2_meta_lock_update(inode, &local_bh);
1641 if (status < 0) {
1642 if (status != -ENOENT)
1643 mlog_errno(status);
1644 goto bail;
1645 }
1646
1647 if (ret_bh) {
1648 status = ocfs2_assign_bh(inode, ret_bh, local_bh);
1649 if (status < 0) {
1650 mlog_errno(status);
1651 goto bail;
1652 }
1653 }
1654
1655 if (handle) {
1656 status = ocfs2_handle_add_lock(handle, inode);
1657 if (status < 0)
1658 mlog_errno(status);
1659 }
1660
1661bail:
1662 if (status < 0) {
1663 if (ret_bh && (*ret_bh)) {
1664 brelse(*ret_bh);
1665 *ret_bh = NULL;
1666 }
1667 if (acquired)
1668 ocfs2_meta_unlock(inode, ex);
1669 }
1670
1671 if (local_bh)
1672 brelse(local_bh);
1673
1674 mlog_exit(status);
1675 return status;
1676}
1677
1678/*
1679 * This is working around a lock inversion between tasks acquiring DLM locks
1680 * while holding a page lock and the vote thread which blocks dlm lock acquiry
1681 * while acquiring page locks.
1682 *
1683 * ** These _with_page variantes are only intended to be called from aop
1684 * methods that hold page locks and return a very specific *positive* error
1685 * code that aop methods pass up to the VFS -- test for errors with != 0. **
1686 *
1687 * The DLM is called such that it returns -EAGAIN if it would have blocked
1688 * waiting for the vote thread. In that case we unlock our page so the vote
1689 * thread can make progress. Once we've done this we have to return
1690 * AOP_TRUNCATED_PAGE so the aop method that called us can bubble that back up
1691 * into the VFS who will then immediately retry the aop call.
1692 *
1693 * We do a blocking lock and immediate unlock before returning, though, so that
1694 * the lock has a great chance of being cached on this node by the time the VFS
1695 * calls back to retry the aop. This has a potential to livelock as nodes
1696 * ping locks back and forth, but that's a risk we're willing to take to avoid
1697 * the lock inversion simply.
1698 */
1699int ocfs2_meta_lock_with_page(struct inode *inode,
1700 struct ocfs2_journal_handle *handle,
1701 struct buffer_head **ret_bh,
1702 int ex,
1703 struct page *page)
1704{
1705 int ret;
1706
1707 ret = ocfs2_meta_lock_full(inode, handle, ret_bh, ex,
1708 OCFS2_LOCK_NONBLOCK);
1709 if (ret == -EAGAIN) {
1710 unlock_page(page);
1711 if (ocfs2_meta_lock(inode, handle, ret_bh, ex) == 0)
1712 ocfs2_meta_unlock(inode, ex);
1713 ret = AOP_TRUNCATED_PAGE;
1714 }
1715
1716 return ret;
1717}
1718
1719void ocfs2_meta_unlock(struct inode *inode,
1720 int ex)
1721{
1722 int level = ex ? LKM_EXMODE : LKM_PRMODE;
1723 struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_meta_lockres;
1724
1725 mlog_entry_void();
1726
b0697053
MF		 1727 mlog(0, "inode %llu drop %s META lock\n",
1728 (unsigned long long)OCFS2_I(inode)->ip_blkno,
ccd979bd
MF		 1729 ex ? "EXMODE" : "PRMODE");
1730
1731 if (!ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb)))
1732 ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);
1733
1734 mlog_exit_void();
1735}
1736
1737int ocfs2_super_lock(struct ocfs2_super *osb,
1738 int ex)
1739{
1740 int status;
1741 int level = ex ? LKM_EXMODE : LKM_PRMODE;
1742 struct ocfs2_lock_res *lockres = &osb->osb_super_lockres;
1743 struct buffer_head *bh;
1744 struct ocfs2_slot_info *si = osb->slot_info;
1745
1746 mlog_entry_void();
1747
1748 if (ocfs2_is_hard_readonly(osb))
1749 return -EROFS;
1750
1751 status = ocfs2_cluster_lock(osb, lockres, level, 0, 0);
1752 if (status < 0) {
1753 mlog_errno(status);
1754 goto bail;
1755 }
1756
1757 /* The super block lock path is really in the best position to
1758 * know when resources covered by the lock need to be
1759 * refreshed, so we do it here. Of course, making sense of
1760 * everything is up to the caller :) */
1761 status = ocfs2_should_refresh_lock_res(lockres);
1762 if (status < 0) {
1763 mlog_errno(status);
1764 goto bail;
1765 }
1766 if (status) {
1767 bh = si->si_bh;
1768 status = ocfs2_read_block(osb, bh->b_blocknr, &bh, 0,
1769 si->si_inode);
1770 if (status == 0)
1771 ocfs2_update_slot_info(si);
1772
1773 ocfs2_complete_lock_res_refresh(lockres, status);
1774
1775 if (status < 0)
1776 mlog_errno(status);
1777 }
1778bail:
1779 mlog_exit(status);
1780 return status;
1781}
1782
1783void ocfs2_super_unlock(struct ocfs2_super *osb,
1784 int ex)
1785{
1786 int level = ex ? LKM_EXMODE : LKM_PRMODE;
1787 struct ocfs2_lock_res *lockres = &osb->osb_super_lockres;
1788
1789 ocfs2_cluster_unlock(osb, lockres, level);
1790}
1791
1792int ocfs2_rename_lock(struct ocfs2_super *osb)
1793{
1794 int status;
1795 struct ocfs2_lock_res *lockres = &osb->osb_rename_lockres;
1796
1797 if (ocfs2_is_hard_readonly(osb))
1798 return -EROFS;
1799
1800 status = ocfs2_cluster_lock(osb, lockres, LKM_EXMODE, 0, 0);
1801 if (status < 0)
1802 mlog_errno(status);
1803
1804 return status;
1805}
1806
1807void ocfs2_rename_unlock(struct ocfs2_super *osb)
1808{
1809 struct ocfs2_lock_res *lockres = &osb->osb_rename_lockres;
1810
1811 ocfs2_cluster_unlock(osb, lockres, LKM_EXMODE);
1812}
1813
d680efe9
MF		 1814int ocfs2_dentry_lock(struct dentry *dentry, int ex)
1815{
1816 int ret;
1817 int level = ex ? LKM_EXMODE : LKM_PRMODE;
1818 struct ocfs2_dentry_lock *dl = dentry->d_fsdata;
1819 struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
1820
1821 BUG_ON(!dl);
1822
1823 if (ocfs2_is_hard_readonly(osb))
1824 return -EROFS;
1825
1826 ret = ocfs2_cluster_lock(osb, &dl->dl_lockres, level, 0, 0);
1827 if (ret < 0)
1828 mlog_errno(ret);
1829
1830 return ret;
1831}
1832
1833void ocfs2_dentry_unlock(struct dentry *dentry, int ex)
1834{
1835 int level = ex ? LKM_EXMODE : LKM_PRMODE;
1836 struct ocfs2_dentry_lock *dl = dentry->d_fsdata;
1837 struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
1838
1839 ocfs2_cluster_unlock(osb, &dl->dl_lockres, level);
1840}
1841
ccd979bd
MF		 1842/* Reference counting of the dlm debug structure. We want this because
1843 * open references on the debug inodes can live on after a mount, so
1844 * we can't rely on the ocfs2_super to always exist. */
1845static void ocfs2_dlm_debug_free(struct kref *kref)
1846{
1847 struct ocfs2_dlm_debug *dlm_debug;
1848
1849 dlm_debug = container_of(kref, struct ocfs2_dlm_debug, d_refcnt);
1850
1851 kfree(dlm_debug);
1852}
1853
1854void ocfs2_put_dlm_debug(struct ocfs2_dlm_debug *dlm_debug)
1855{
1856 if (dlm_debug)
1857 kref_put(&dlm_debug->d_refcnt, ocfs2_dlm_debug_free);
1858}
1859
1860static void ocfs2_get_dlm_debug(struct ocfs2_dlm_debug *debug)
1861{
1862 kref_get(&debug->d_refcnt);
1863}
1864
1865struct ocfs2_dlm_debug *ocfs2_new_dlm_debug(void)
1866{
1867 struct ocfs2_dlm_debug *dlm_debug;
1868
1869 dlm_debug = kmalloc(sizeof(struct ocfs2_dlm_debug), GFP_KERNEL);
1870 if (!dlm_debug) {
1871 mlog_errno(-ENOMEM);
1872 goto out;
1873 }
1874
1875 kref_init(&dlm_debug->d_refcnt);
1876 INIT_LIST_HEAD(&dlm_debug->d_lockres_tracking);
1877 dlm_debug->d_locking_state = NULL;
1878out:
1879 return dlm_debug;
1880}
1881
1882/* Access to this is arbitrated for us via seq_file->sem. */
1883struct ocfs2_dlm_seq_priv {
1884 struct ocfs2_dlm_debug *p_dlm_debug;
1885 struct ocfs2_lock_res p_iter_res;
1886 struct ocfs2_lock_res p_tmp_res;
1887};
1888
1889static struct ocfs2_lock_res *ocfs2_dlm_next_res(struct ocfs2_lock_res *start,
1890 struct ocfs2_dlm_seq_priv *priv)
1891{
1892 struct ocfs2_lock_res *iter, *ret = NULL;
1893 struct ocfs2_dlm_debug *dlm_debug = priv->p_dlm_debug;
1894
1895 assert_spin_locked(&ocfs2_dlm_tracking_lock);
1896
1897 list_for_each_entry(iter, &start->l_debug_list, l_debug_list) {
1898 /* discover the head of the list */
1899 if (&iter->l_debug_list == &dlm_debug->d_lockres_tracking) {
1900 mlog(0, "End of list found, %p\n", ret);
1901 break;
1902 }
1903
1904 /* We track our "dummy" iteration lockres' by a NULL
1905 * l_ops field. */
1906 if (iter->l_ops != NULL) {
1907 ret = iter;
1908 break;
1909 }
1910 }
1911
1912 return ret;
1913}
1914
1915static void *ocfs2_dlm_seq_start(struct seq_file *m, loff_t *pos)
1916{
1917 struct ocfs2_dlm_seq_priv *priv = m->private;
1918 struct ocfs2_lock_res *iter;
1919
1920 spin_lock(&ocfs2_dlm_tracking_lock);
1921 iter = ocfs2_dlm_next_res(&priv->p_iter_res, priv);
1922 if (iter) {
1923 /* Since lockres' have the lifetime of their container
1924 * (which can be inodes, ocfs2_supers, etc) we want to
1925 * copy this out to a temporary lockres while still
1926 * under the spinlock. Obviously after this we can't
1927 * trust any pointers on the copy returned, but that's
1928 * ok as the information we want isn't typically held
1929 * in them. */
1930 priv->p_tmp_res = *iter;
1931 iter = &priv->p_tmp_res;
1932 }
1933 spin_unlock(&ocfs2_dlm_tracking_lock);
1934
1935 return iter;
1936}
1937
1938static void ocfs2_dlm_seq_stop(struct seq_file *m, void *v)
1939{
1940}
1941
1942static void *ocfs2_dlm_seq_next(struct seq_file *m, void *v, loff_t *pos)
1943{
1944 struct ocfs2_dlm_seq_priv *priv = m->private;
1945 struct ocfs2_lock_res *iter = v;
1946 struct ocfs2_lock_res *dummy = &priv->p_iter_res;
1947
1948 spin_lock(&ocfs2_dlm_tracking_lock);
1949 iter = ocfs2_dlm_next_res(iter, priv);
1950 list_del_init(&dummy->l_debug_list);
1951 if (iter) {
1952 list_add(&dummy->l_debug_list, &iter->l_debug_list);
1953 priv->p_tmp_res = *iter;
1954 iter = &priv->p_tmp_res;
1955 }
1956 spin_unlock(&ocfs2_dlm_tracking_lock);
1957
1958 return iter;
1959}
1960
1961/* So that debugfs.ocfs2 can determine which format is being used */
1962#define OCFS2_DLM_DEBUG_STR_VERSION 1
1963static int ocfs2_dlm_seq_show(struct seq_file *m, void *v)
1964{
1965 int i;
1966 char *lvb;
1967 struct ocfs2_lock_res *lockres = v;
1968
1969 if (!lockres)
1970 return -EINVAL;
1971
d680efe9
MF		 1972 seq_printf(m, "0x%x\t", OCFS2_DLM_DEBUG_STR_VERSION);
1973
1974 if (lockres->l_type == OCFS2_LOCK_TYPE_DENTRY)
1975 seq_printf(m, "%.*s%08x\t", OCFS2_DENTRY_LOCK_INO_START - 1,
1976 lockres->l_name,
1977 (unsigned int)ocfs2_get_dentry_lock_ino(lockres));
1978 else
1979 seq_printf(m, "%.*s\t", OCFS2_LOCK_ID_MAX_LEN, lockres->l_name);
1980
1981 seq_printf(m, "%d\t"
ccd979bd
MF		 1982 "0x%lx\t"
1983 "0x%x\t"
1984 "0x%x\t"
1985 "%u\t"
1986 "%u\t"
1987 "%d\t"
1988 "%d\t",
ccd979bd
MF		 1989 lockres->l_level,
1990 lockres->l_flags,
1991 lockres->l_action,
1992 lockres->l_unlock_action,
1993 lockres->l_ro_holders,
1994 lockres->l_ex_holders,
1995 lockres->l_requested,
1996 lockres->l_blocking);
1997
1998 /* Dump the raw LVB */
1999 lvb = lockres->l_lksb.lvb;
2000 for(i = 0; i < DLM_LVB_LEN; i++)
2001 seq_printf(m, "0x%x\t", lvb[i]);
2002
2003 /* End the line */
2004 seq_printf(m, "\n");
2005 return 0;
2006}
2007
2008static struct seq_operations ocfs2_dlm_seq_ops = {
2009 .start = ocfs2_dlm_seq_start,
2010 .stop = ocfs2_dlm_seq_stop,
2011 .next = ocfs2_dlm_seq_next,
2012 .show = ocfs2_dlm_seq_show,
2013};
2014
2015static int ocfs2_dlm_debug_release(struct inode *inode, struct file *file)
2016{
2017 struct seq_file *seq = (struct seq_file *) file->private_data;
2018 struct ocfs2_dlm_seq_priv *priv = seq->private;
2019 struct ocfs2_lock_res *res = &priv->p_iter_res;
2020
2021 ocfs2_remove_lockres_tracking(res);
2022 ocfs2_put_dlm_debug(priv->p_dlm_debug);
2023 return seq_release_private(inode, file);
2024}
2025
2026static int ocfs2_dlm_debug_open(struct inode *inode, struct file *file)
2027{
2028 int ret;
2029 struct ocfs2_dlm_seq_priv *priv;
2030 struct seq_file *seq;
2031 struct ocfs2_super *osb;
2032
2033 priv = kzalloc(sizeof(struct ocfs2_dlm_seq_priv), GFP_KERNEL);
2034 if (!priv) {
2035 ret = -ENOMEM;
2036 mlog_errno(ret);
2037 goto out;
2038 }
2039 osb = (struct ocfs2_super *) inode->u.generic_ip;
2040 ocfs2_get_dlm_debug(osb->osb_dlm_debug);
2041 priv->p_dlm_debug = osb->osb_dlm_debug;
2042 INIT_LIST_HEAD(&priv->p_iter_res.l_debug_list);
2043
2044 ret = seq_open(file, &ocfs2_dlm_seq_ops);
2045 if (ret) {
2046 kfree(priv);
2047 mlog_errno(ret);
2048 goto out;
2049 }
2050
2051 seq = (struct seq_file *) file->private_data;
2052 seq->private = priv;
2053
2054 ocfs2_add_lockres_tracking(&priv->p_iter_res,
2055 priv->p_dlm_debug);
2056
2057out:
2058 return ret;
2059}
2060
4b6f5d20 2061static const struct file_operations ocfs2_dlm_debug_fops = {
ccd979bd
MF		 2062 .open = ocfs2_dlm_debug_open,
2063 .release = ocfs2_dlm_debug_release,
2064 .read = seq_read,
2065 .llseek = seq_lseek,
2066};
2067
2068static int ocfs2_dlm_init_debug(struct ocfs2_super *osb)
2069{
2070 int ret = 0;
2071 struct ocfs2_dlm_debug *dlm_debug = osb->osb_dlm_debug;
2072
2073 dlm_debug->d_locking_state = debugfs_create_file("locking_state",
2074 S_IFREG|S_IRUSR,
2075 osb->osb_debug_root,
2076 osb,
2077 &ocfs2_dlm_debug_fops);
2078 if (!dlm_debug->d_locking_state) {
2079 ret = -EINVAL;
2080 mlog(ML_ERROR,
2081 "Unable to create locking state debugfs file.\n");
2082 goto out;
2083 }
2084
2085 ocfs2_get_dlm_debug(dlm_debug);
2086out:
2087 return ret;
2088}
2089
2090static void ocfs2_dlm_shutdown_debug(struct ocfs2_super *osb)
2091{
2092 struct ocfs2_dlm_debug *dlm_debug = osb->osb_dlm_debug;
2093
2094 if (dlm_debug) {
2095 debugfs_remove(dlm_debug->d_locking_state);
2096 ocfs2_put_dlm_debug(dlm_debug);
2097 }
2098}
2099
2100int ocfs2_dlm_init(struct ocfs2_super *osb)
2101{
2102 int status;
2103 u32 dlm_key;
2104 struct dlm_ctxt *dlm;
2105
2106 mlog_entry_void();
2107
2108 status = ocfs2_dlm_init_debug(osb);
2109 if (status < 0) {
2110 mlog_errno(status);
2111 goto bail;
2112 }
2113
2114 /* launch vote thread */
78427043 2115 osb->vote_task = kthread_run(ocfs2_vote_thread, osb, "ocfs2vote");
ccd979bd
MF		 2116 if (IS_ERR(osb->vote_task)) {
2117 status = PTR_ERR(osb->vote_task);
2118 osb->vote_task = NULL;
2119 mlog_errno(status);
2120 goto bail;
2121 }
2122
2123 /* used by the dlm code to make message headers unique, each
2124 * node in this domain must agree on this. */
2125 dlm_key = crc32_le(0, osb->uuid_str, strlen(osb->uuid_str));
2126
2127 /* for now, uuid == domain */
2128 dlm = dlm_register_domain(osb->uuid_str, dlm_key);
2129 if (IS_ERR(dlm)) {
2130 status = PTR_ERR(dlm);
2131 mlog_errno(status);
2132 goto bail;
2133 }
2134
2135 ocfs2_super_lock_res_init(&osb->osb_super_lockres, osb);
2136 ocfs2_rename_lock_res_init(&osb->osb_rename_lockres, osb);
2137
2138 dlm_register_eviction_cb(dlm, &osb->osb_eviction_cb);
2139
2140 osb->dlm = dlm;
2141
2142 status = 0;
2143bail:
2144 if (status < 0) {
2145 ocfs2_dlm_shutdown_debug(osb);
2146 if (osb->vote_task)
2147 kthread_stop(osb->vote_task);
2148 }
2149
2150 mlog_exit(status);
2151 return status;
2152}
2153
2154void ocfs2_dlm_shutdown(struct ocfs2_super *osb)
2155{
2156 mlog_entry_void();
2157
2158 dlm_unregister_eviction_cb(&osb->osb_eviction_cb);
2159
2160 ocfs2_drop_osb_locks(osb);
2161
2162 if (osb->vote_task) {
2163 kthread_stop(osb->vote_task);
2164 osb->vote_task = NULL;
2165 }
2166
2167 ocfs2_lock_res_free(&osb->osb_super_lockres);
2168 ocfs2_lock_res_free(&osb->osb_rename_lockres);
2169
2170 dlm_unregister_domain(osb->dlm);
2171 osb->dlm = NULL;
2172
2173 ocfs2_dlm_shutdown_debug(osb);
2174
2175 mlog_exit_void();
2176}
2177
2a45f2d1 2178static void ocfs2_unlock_ast(void *opaque, enum dlm_status status)
ccd979bd
MF		 2179{
2180 struct ocfs2_lock_res *lockres = opaque;
2181 unsigned long flags;
2182
2183 mlog_entry_void();
2184
2185 mlog(0, "UNLOCK AST called on lock %s, action = %d\n", lockres->l_name,
2186 lockres->l_unlock_action);
2187
2188 spin_lock_irqsave(&lockres->l_lock, flags);
2189 /* We tried to cancel a convert request, but it was already
2190 * granted. All we want to do here is clear our unlock
2191 * state. The wake_up call done at the bottom is redundant
2192 * (ocfs2_prepare_cancel_convert doesn't sleep on this) but doesn't
2193 * hurt anything anyway */
2194 if (status == DLM_CANCELGRANT &&
2195 lockres->l_unlock_action == OCFS2_UNLOCK_CANCEL_CONVERT) {
2196 mlog(0, "Got cancelgrant for %s\n", lockres->l_name);
2197
2198 /* We don't clear the busy flag in this case as it
2199 * should have been cleared by the ast which the dlm
2200 * has called. */
2201 goto complete_unlock;
2202 }
2203
2204 if (status != DLM_NORMAL) {
2205 mlog(ML_ERROR, "Dlm passes status %d for lock %s, "
2206 "unlock_action %d\n", status, lockres->l_name,
2207 lockres->l_unlock_action);
2208 spin_unlock_irqrestore(&lockres->l_lock, flags);
2209 return;
2210 }
2211
2212 switch(lockres->l_unlock_action) {
2213 case OCFS2_UNLOCK_CANCEL_CONVERT:
2214 mlog(0, "Cancel convert success for %s\n", lockres->l_name);
2215 lockres->l_action = OCFS2_AST_INVALID;
2216 break;
2217 case OCFS2_UNLOCK_DROP_LOCK:
2218 lockres->l_level = LKM_IVMODE;
2219 break;
2220 default:
2221 BUG();
2222 }
2223
2224 lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
2225complete_unlock:
2226 lockres->l_unlock_action = OCFS2_UNLOCK_INVALID;
2227 spin_unlock_irqrestore(&lockres->l_lock, flags);
2228
2229 wake_up(&lockres->l_event);
2230
2231 mlog_exit_void();
2232}
2233
2234typedef void (ocfs2_pre_drop_cb_t)(struct ocfs2_lock_res *, void *);
2235
2236struct drop_lock_cb {
2237 ocfs2_pre_drop_cb_t *drop_func;
2238 void *drop_data;
2239};
2240
2241static int ocfs2_drop_lock(struct ocfs2_super *osb,
2242 struct ocfs2_lock_res *lockres,
2243 struct drop_lock_cb *dcb)
2244{
2245 enum dlm_status status;
2246 unsigned long flags;
b80fc012 2247 int lkm_flags = 0;
ccd979bd
MF		 2248
2249 /* We didn't get anywhere near actually using this lockres. */
2250 if (!(lockres->l_flags & OCFS2_LOCK_INITIALIZED))
2251 goto out;
2252
b80fc012
MF		 2253 if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB)
2254 lkm_flags |= LKM_VALBLK;
2255
ccd979bd
MF		 2256 spin_lock_irqsave(&lockres->l_lock, flags);
2257
2258 mlog_bug_on_msg(!(lockres->l_flags & OCFS2_LOCK_FREEING),
2259 "lockres %s, flags 0x%lx\n",
2260 lockres->l_name, lockres->l_flags);
2261
2262 while (lockres->l_flags & OCFS2_LOCK_BUSY) {
2263 mlog(0, "waiting on busy lock \"%s\": flags = %lx, action = "
2264 "%u, unlock_action = %u\n",
2265 lockres->l_name, lockres->l_flags, lockres->l_action,
2266 lockres->l_unlock_action);
2267
2268 spin_unlock_irqrestore(&lockres->l_lock, flags);
2269
2270 /* XXX: Today we just wait on any busy
2271 * locks... Perhaps we need to cancel converts in the
2272 * future? */
2273 ocfs2_wait_on_busy_lock(lockres);
2274
2275 spin_lock_irqsave(&lockres->l_lock, flags);
2276 }
2277
2278 if (dcb)
2279 dcb->drop_func(lockres, dcb->drop_data);
2280
2281 if (lockres->l_flags & OCFS2_LOCK_BUSY)
2282 mlog(ML_ERROR, "destroying busy lock: \"%s\"\n",
2283 lockres->l_name);
2284 if (lockres->l_flags & OCFS2_LOCK_BLOCKED)
2285 mlog(0, "destroying blocked lock: \"%s\"\n", lockres->l_name);
2286
2287 if (!(lockres->l_flags & OCFS2_LOCK_ATTACHED)) {
2288 spin_unlock_irqrestore(&lockres->l_lock, flags);
2289 goto out;
2290 }
2291
2292 lockres_clear_flags(lockres, OCFS2_LOCK_ATTACHED);
2293
2294 /* make sure we never get here while waiting for an ast to
2295 * fire. */
2296 BUG_ON(lockres->l_action != OCFS2_AST_INVALID);
2297
2298 /* is this necessary? */
2299 lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
2300 lockres->l_unlock_action = OCFS2_UNLOCK_DROP_LOCK;
2301 spin_unlock_irqrestore(&lockres->l_lock, flags);
2302
2303 mlog(0, "lock %s\n", lockres->l_name);
2304
b80fc012 2305 status = dlmunlock(osb->dlm, &lockres->l_lksb, lkm_flags,
2a45f2d1 2306 ocfs2_unlock_ast, lockres);
ccd979bd
MF		 2307 if (status != DLM_NORMAL) {
2308 ocfs2_log_dlm_error("dlmunlock", status, lockres);
2309 mlog(ML_ERROR, "lockres flags: %lu\n", lockres->l_flags);
2310 dlm_print_one_lock(lockres->l_lksb.lockid);
2311 BUG();
2312 }
2313 mlog(0, "lock %s, successfull return from dlmunlock\n",
2314 lockres->l_name);
2315
2316 ocfs2_wait_on_busy_lock(lockres);
2317out:
2318 mlog_exit(0);
2319 return 0;
2320}
2321
2322/* Mark the lockres as being dropped. It will no longer be
2323 * queued if blocking, but we still may have to wait on it
2324 * being dequeued from the vote thread before we can consider
2325 * it safe to drop.
2326 *
2327 * You can *not* attempt to call cluster_lock on this lockres anymore. */
2328void ocfs2_mark_lockres_freeing(struct ocfs2_lock_res *lockres)
2329{
2330 int status;
2331 struct ocfs2_mask_waiter mw;
2332 unsigned long flags;
2333
2334 ocfs2_init_mask_waiter(&mw);
2335
2336 spin_lock_irqsave(&lockres->l_lock, flags);
2337 lockres->l_flags |= OCFS2_LOCK_FREEING;
2338 while (lockres->l_flags & OCFS2_LOCK_QUEUED) {
2339 lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_QUEUED, 0);
2340 spin_unlock_irqrestore(&lockres->l_lock, flags);
2341
2342 mlog(0, "Waiting on lockres %s\n", lockres->l_name);
2343
2344 status = ocfs2_wait_for_mask(&mw);
2345 if (status)
2346 mlog_errno(status);
2347
2348 spin_lock_irqsave(&lockres->l_lock, flags);
2349 }
2350 spin_unlock_irqrestore(&lockres->l_lock, flags);
2351}
2352
d680efe9
MF		 2353void ocfs2_simple_drop_lockres(struct ocfs2_super *osb,
2354 struct ocfs2_lock_res *lockres)
ccd979bd 2355{
d680efe9 2356 int ret;
ccd979bd 2357
d680efe9
MF		 2358 ocfs2_mark_lockres_freeing(lockres);
2359 ret = ocfs2_drop_lock(osb, lockres, NULL);
2360 if (ret)
2361 mlog_errno(ret);
2362}
ccd979bd 2363
d680efe9
MF		 2364static void ocfs2_drop_osb_locks(struct ocfs2_super *osb)
2365{
2366 ocfs2_simple_drop_lockres(osb, &osb->osb_super_lockres);
2367 ocfs2_simple_drop_lockres(osb, &osb->osb_rename_lockres);
ccd979bd
MF		 2368}
2369
2370static void ocfs2_meta_pre_drop(struct ocfs2_lock_res *lockres, void *data)
2371{
2372 struct inode *inode = data;
2373
2374 /* the metadata lock requires a bit more work as we have an
2375 * LVB to worry about. */
2376 if (lockres->l_flags & OCFS2_LOCK_ATTACHED &&
2377 lockres->l_level == LKM_EXMODE &&
2378 !(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH))
2379 __ocfs2_stuff_meta_lvb(inode);
2380}
2381
2382int ocfs2_drop_inode_locks(struct inode *inode)
2383{
2384 int status, err;
2385 struct drop_lock_cb meta_dcb = { ocfs2_meta_pre_drop, inode, };
2386
2387 mlog_entry_void();
2388
2389 /* No need to call ocfs2_mark_lockres_freeing here -
2390 * ocfs2_clear_inode has done it for us. */
2391
2392 err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb),
2393 &OCFS2_I(inode)->ip_data_lockres,
2394 NULL);
2395 if (err < 0)
2396 mlog_errno(err);
2397
2398 status = err;
2399
2400 err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb),
2401 &OCFS2_I(inode)->ip_meta_lockres,
2402 &meta_dcb);
2403 if (err < 0)
2404 mlog_errno(err);
2405 if (err < 0 && !status)
2406 status = err;
2407
2408 err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb),
2409 &OCFS2_I(inode)->ip_rw_lockres,
2410 NULL);
2411 if (err < 0)
2412 mlog_errno(err);
2413 if (err < 0 && !status)
2414 status = err;
2415
2416 mlog_exit(status);
2417 return status;
2418}
2419
2420static void ocfs2_prepare_downconvert(struct ocfs2_lock_res *lockres,
2421 int new_level)
2422{
2423 assert_spin_locked(&lockres->l_lock);
2424
2425 BUG_ON(lockres->l_blocking <= LKM_NLMODE);
2426
2427 if (lockres->l_level <= new_level) {
2428 mlog(ML_ERROR, "lockres->l_level (%u) <= new_level (%u)\n",
2429 lockres->l_level, new_level);
2430 BUG();
2431 }
2432
2433 mlog(0, "lock %s, new_level = %d, l_blocking = %d\n",
2434 lockres->l_name, new_level, lockres->l_blocking);
2435
2436 lockres->l_action = OCFS2_AST_DOWNCONVERT;
2437 lockres->l_requested = new_level;
2438 lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
2439}
2440
2441static int ocfs2_downconvert_lock(struct ocfs2_super *osb,
2442 struct ocfs2_lock_res *lockres,
2443 int new_level,
2444 int lvb)
2445{
2446 int ret, dlm_flags = LKM_CONVERT;
2447 enum dlm_status status;
2448
2449 mlog_entry_void();
2450
2451 if (lvb)
2452 dlm_flags |= LKM_VALBLK;
2453
2454 status = dlmlock(osb->dlm,
2455 new_level,
2456 &lockres->l_lksb,
2457 dlm_flags,
2458 lockres->l_name,
f0681062 2459 OCFS2_LOCK_ID_MAX_LEN - 1,
e92d57df 2460 ocfs2_locking_ast,
ccd979bd 2461 lockres,
aa2623ad 2462 ocfs2_blocking_ast);
ccd979bd
MF		 2463 if (status != DLM_NORMAL) {
2464 ocfs2_log_dlm_error("dlmlock", status, lockres);
2465 ret = -EINVAL;
2466 ocfs2_recover_from_dlm_error(lockres, 1);
2467 goto bail;
2468 }
2469
2470 ret = 0;
2471bail:
2472 mlog_exit(ret);
2473 return ret;
2474}
2475
2476/* returns 1 when the caller should unlock and call dlmunlock */
2477static int ocfs2_prepare_cancel_convert(struct ocfs2_super *osb,
2478 struct ocfs2_lock_res *lockres)
2479{
2480 assert_spin_locked(&lockres->l_lock);
2481
2482 mlog_entry_void();
2483 mlog(0, "lock %s\n", lockres->l_name);
2484
2485 if (lockres->l_unlock_action == OCFS2_UNLOCK_CANCEL_CONVERT) {
2486 /* If we're already trying to cancel a lock conversion
2487 * then just drop the spinlock and allow the caller to
2488 * requeue this lock. */
2489
2490 mlog(0, "Lockres %s, skip convert\n", lockres->l_name);
2491 return 0;
2492 }
2493
2494 /* were we in a convert when we got the bast fire? */
2495 BUG_ON(lockres->l_action != OCFS2_AST_CONVERT &&
2496 lockres->l_action != OCFS2_AST_DOWNCONVERT);
2497 /* set things up for the unlockast to know to just
2498 * clear out the ast_action and unset busy, etc. */
2499 lockres->l_unlock_action = OCFS2_UNLOCK_CANCEL_CONVERT;
2500
2501 mlog_bug_on_msg(!(lockres->l_flags & OCFS2_LOCK_BUSY),
2502 "lock %s, invalid flags: 0x%lx\n",
2503 lockres->l_name, lockres->l_flags);
2504
2505 return 1;
2506}
2507
2508static int ocfs2_cancel_convert(struct ocfs2_super *osb,
2509 struct ocfs2_lock_res *lockres)
2510{
2511 int ret;
2512 enum dlm_status status;
2513
2514 mlog_entry_void();
2515 mlog(0, "lock %s\n", lockres->l_name);
2516
2517 ret = 0;
2518 status = dlmunlock(osb->dlm,
2519 &lockres->l_lksb,
2520 LKM_CANCEL,
2a45f2d1 2521 ocfs2_unlock_ast,
ccd979bd
MF		 2522 lockres);
2523 if (status != DLM_NORMAL) {
2524 ocfs2_log_dlm_error("dlmunlock", status, lockres);
2525 ret = -EINVAL;
2526 ocfs2_recover_from_dlm_error(lockres, 0);
2527 }
2528
2529 mlog(0, "lock %s return from dlmunlock\n", lockres->l_name);
2530
2531 mlog_exit(ret);
2532 return ret;
2533}
2534
2535static inline int ocfs2_can_downconvert_meta_lock(struct inode *inode,
2536 struct ocfs2_lock_res *lockres,
2537 int new_level)
2538{
2539 int ret;
2540
2541 mlog_entry_void();
2542
2543 BUG_ON(new_level != LKM_NLMODE && new_level != LKM_PRMODE);
2544
2545 if (lockres->l_flags & OCFS2_LOCK_REFRESHING) {
2546 ret = 0;
2547 mlog(0, "lockres %s currently being refreshed -- backing "
2548 "off!\n", lockres->l_name);
2549 } else if (new_level == LKM_PRMODE)
2550 ret = !lockres->l_ex_holders &&
2551 ocfs2_inode_fully_checkpointed(inode);
2552 else /* Must be NLMODE we're converting to. */
2553 ret = !lockres->l_ro_holders && !lockres->l_ex_holders &&
2554 ocfs2_inode_fully_checkpointed(inode);
2555
2556 mlog_exit(ret);
2557 return ret;
2558}
2559
2560static int ocfs2_do_unblock_meta(struct inode *inode,
2561 int *requeue)
2562{
2563 int new_level;
2564 int set_lvb = 0;
2565 int ret = 0;
2566 struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_meta_lockres;
2567 unsigned long flags;
2568
2569 struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
2570
2571 mlog_entry_void();
2572
2573 spin_lock_irqsave(&lockres->l_lock, flags);
2574
2575 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));
2576
2577 mlog(0, "l_level=%d, l_blocking=%d\n", lockres->l_level,
2578 lockres->l_blocking);
2579
2580 BUG_ON(lockres->l_level != LKM_EXMODE &&
2581 lockres->l_level != LKM_PRMODE);
2582
2583 if (lockres->l_flags & OCFS2_LOCK_BUSY) {
2584 *requeue = 1;
2585 ret = ocfs2_prepare_cancel_convert(osb, lockres);
2586 spin_unlock_irqrestore(&lockres->l_lock, flags);
2587 if (ret) {
2588 ret = ocfs2_cancel_convert(osb, lockres);
2589 if (ret < 0)
2590 mlog_errno(ret);
2591 }
2592 goto leave;
2593 }
2594
2595 new_level = ocfs2_highest_compat_lock_level(lockres->l_blocking);
2596
2597 mlog(0, "l_level=%d, l_blocking=%d, new_level=%d\n",
2598 lockres->l_level, lockres->l_blocking, new_level);
2599
2600 if (ocfs2_can_downconvert_meta_lock(inode, lockres, new_level)) {
2601 if (lockres->l_level == LKM_EXMODE)
2602 set_lvb = 1;
2603
2604 /* If the lock hasn't been refreshed yet (rare), then
2605 * our memory inode values are old and we skip
2606 * stuffing the lvb. There's no need to actually clear
2607 * out the lvb here as it's value is still valid. */
2608 if (!(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH)) {
2609 if (set_lvb)
2610 __ocfs2_stuff_meta_lvb(inode);
2611 } else
2612 mlog(0, "lockres %s: downconverting stale lock!\n",
2613 lockres->l_name);
2614
2615 mlog(0, "calling ocfs2_downconvert_lock with l_level=%d, "
2616 "l_blocking=%d, new_level=%d\n",
2617 lockres->l_level, lockres->l_blocking, new_level);
2618
2619 ocfs2_prepare_downconvert(lockres, new_level);
2620 spin_unlock_irqrestore(&lockres->l_lock, flags);
2621 ret = ocfs2_downconvert_lock(osb, lockres, new_level, set_lvb);
2622 goto leave;
2623 }
2624 if (!ocfs2_inode_fully_checkpointed(inode))
2625 ocfs2_start_checkpoint(osb);
2626
2627 *requeue = 1;
2628 spin_unlock_irqrestore(&lockres->l_lock, flags);
2629 ret = 0;
2630leave:
2631 mlog_exit(ret);
2632 return ret;
2633}
2634
2635static int ocfs2_generic_unblock_lock(struct ocfs2_super *osb,
2636 struct ocfs2_lock_res *lockres,
d680efe9 2637 struct ocfs2_unblock_ctl *ctl,
ccd979bd
MF		 2638 ocfs2_convert_worker_t *worker)
2639{
2640 unsigned long flags;
2641 int blocking;
2642 int new_level;
2643 int ret = 0;
5ef0d4ea 2644 int set_lvb = 0;
ccd979bd
MF		 2645
2646 mlog_entry_void();
2647
2648 spin_lock_irqsave(&lockres->l_lock, flags);
2649
2650 BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));
2651
2652recheck:
2653 if (lockres->l_flags & OCFS2_LOCK_BUSY) {
d680efe9 2654 ctl->requeue = 1;
ccd979bd
MF		 2655 ret = ocfs2_prepare_cancel_convert(osb, lockres);
2656 spin_unlock_irqrestore(&lockres->l_lock, flags);
2657 if (ret) {
2658 ret = ocfs2_cancel_convert(osb, lockres);
2659 if (ret < 0)
2660 mlog_errno(ret);
2661 }
2662 goto leave;
2663 }
2664
2665 /* if we're blocking an exclusive and we have *any* holders,
2666 * then requeue. */
2667 if ((lockres->l_blocking == LKM_EXMODE)
f7fbfdd1
MF		 2668 && (lockres->l_ex_holders || lockres->l_ro_holders))
2669 goto leave_requeue;
ccd979bd
MF		 2670
2671 /* If it's a PR we're blocking, then only
2672 * requeue if we've got any EX holders */
2673 if (lockres->l_blocking == LKM_PRMODE &&
f7fbfdd1
MF		 2674 lockres->l_ex_holders)
2675 goto leave_requeue;
2676
2677 /*
2678 * Can we get a lock in this state if the holder counts are
2679 * zero? The meta data unblock code used to check this.
2680 */
2681 if ((lockres->l_ops->flags & LOCK_TYPE_REQUIRES_REFRESH)
2682 && (lockres->l_flags & OCFS2_LOCK_REFRESHING))
2683 goto leave_requeue;
ccd979bd 2684
16d5b956
MF		 2685 new_level = ocfs2_highest_compat_lock_level(lockres->l_blocking);
2686
2687 if (lockres->l_ops->check_downconvert
2688 && !lockres->l_ops->check_downconvert(lockres, new_level))
2689 goto leave_requeue;
2690
ccd979bd
MF		 2691 /* If we get here, then we know that there are no more
2692 * incompatible holders (and anyone asking for an incompatible
2693 * lock is blocked). We can now downconvert the lock */
2694 if (!worker)
2695 goto downconvert;
2696
2697 /* Some lockres types want to do a bit of work before
2698 * downconverting a lock. Allow that here. The worker function
2699 * may sleep, so we save off a copy of what we're blocking as
2700 * it may change while we're not holding the spin lock. */
2701 blocking = lockres->l_blocking;
2702 spin_unlock_irqrestore(&lockres->l_lock, flags);
2703
d680efe9
MF		 2704 ctl->unblock_action = worker(lockres, blocking);
2705
2706 if (ctl->unblock_action == UNBLOCK_STOP_POST)
2707 goto leave;
ccd979bd
MF		 2708
2709 spin_lock_irqsave(&lockres->l_lock, flags);
2710 if (blocking != lockres->l_blocking) {
2711 /* If this changed underneath us, then we can't drop
2712 * it just yet. */
2713 goto recheck;
2714 }
2715
2716downconvert:
d680efe9 2717 ctl->requeue = 0;
ccd979bd 2718
5ef0d4ea
MF		 2719 if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB) {
2720 if (lockres->l_level == LKM_EXMODE)
2721 set_lvb = 1;
2722
2723 /*
2724 * We only set the lvb if the lock has been fully
2725 * refreshed - otherwise we risk setting stale
2726 * data. Otherwise, there's no need to actually clear
2727 * out the lvb here as it's value is still valid.
2728 */
2729 if (set_lvb && !(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH))
2730 lockres->l_ops->set_lvb(lockres);
2731 }
2732
ccd979bd
MF		 2733 ocfs2_prepare_downconvert(lockres, new_level);
2734 spin_unlock_irqrestore(&lockres->l_lock, flags);
5ef0d4ea 2735 ret = ocfs2_downconvert_lock(osb, lockres, new_level, set_lvb);
ccd979bd
MF		 2736leave:
2737 mlog_exit(ret);
2738 return ret;
f7fbfdd1
MF		 2739
2740leave_requeue:
2741 spin_unlock_irqrestore(&lockres->l_lock, flags);
2742 ctl->requeue = 1;
2743
2744 mlog_exit(0);
2745 return 0;
ccd979bd
MF		 2746}
2747
d680efe9
MF		 2748static int ocfs2_data_convert_worker(struct ocfs2_lock_res *lockres,
2749 int blocking)
ccd979bd
MF		 2750{
2751 struct inode *inode;
2752 struct address_space *mapping;
2753
ccd979bd
MF		 2754 inode = ocfs2_lock_res_inode(lockres);
2755 mapping = inode->i_mapping;
2756
2757 if (filemap_fdatawrite(mapping)) {
b0697053
MF		 2758 mlog(ML_ERROR, "Could not sync inode %llu for downconvert!",
2759 (unsigned long long)OCFS2_I(inode)->ip_blkno);
ccd979bd
MF		 2760 }
2761 sync_mapping_buffers(mapping);
2762 if (blocking == LKM_EXMODE) {
2763 truncate_inode_pages(mapping, 0);
2764 unmap_mapping_range(mapping, 0, 0, 0);
2765 } else {
2766 /* We only need to wait on the I/O if we're not also
2767 * truncating pages because truncate_inode_pages waits
2768 * for us above. We don't truncate pages if we're
2769 * blocking anything < EXMODE because we want to keep
2770 * them around in that case. */
2771 filemap_fdatawait(mapping);
2772 }
2773
d680efe9 2774 return UNBLOCK_CONTINUE;
ccd979bd
MF		 2775}
2776
2777int ocfs2_unblock_data(struct ocfs2_lock_res *lockres,
d680efe9 2778 struct ocfs2_unblock_ctl *ctl)
ccd979bd
MF		 2779{
2780 int status;
2781 struct inode *inode;
2782 struct ocfs2_super *osb;
2783
2784 mlog_entry_void();
2785
2786 inode = ocfs2_lock_res_inode(lockres);
2787 osb = OCFS2_SB(inode->i_sb);
2788
b0697053
MF		 2789 mlog(0, "unblock inode %llu\n",
2790 (unsigned long long)OCFS2_I(inode)->ip_blkno);
ccd979bd 2791
d680efe9 2792 status = ocfs2_generic_unblock_lock(osb, lockres, ctl,
ccd979bd
MF		 2793 ocfs2_data_convert_worker);
2794 if (status < 0)
2795 mlog_errno(status);
2796
b0697053 2797 mlog(0, "inode %llu, requeue = %d\n",
d680efe9 2798 (unsigned long long)OCFS2_I(inode)->ip_blkno, ctl->requeue);
ccd979bd
MF		 2799
2800 mlog_exit(status);
2801 return status;
2802}
2803
2804static int ocfs2_unblock_inode_lock(struct ocfs2_lock_res *lockres,
d680efe9 2805 struct ocfs2_unblock_ctl *ctl)
ccd979bd
MF		 2806{
2807 int status;
2808 struct inode *inode;
2809
2810 mlog_entry_void();
2811
2812 mlog(0, "Unblock lockres %s\n", lockres->l_name);
2813
2814 inode = ocfs2_lock_res_inode(lockres);
2815
2816 status = ocfs2_generic_unblock_lock(OCFS2_SB(inode->i_sb),
d680efe9 2817 lockres, ctl, NULL);
ccd979bd
MF		 2818 if (status < 0)
2819 mlog_errno(status);
2820
2821 mlog_exit(status);
2822 return status;
2823}
2824
d680efe9
MF		 2825static int ocfs2_unblock_meta(struct ocfs2_lock_res *lockres,
2826 struct ocfs2_unblock_ctl *ctl)
ccd979bd
MF		 2827{
2828 int status;
2829 struct inode *inode;
2830
2831 mlog_entry_void();
2832
2833 inode = ocfs2_lock_res_inode(lockres);
2834
b0697053
MF		 2835 mlog(0, "unblock inode %llu\n",
2836 (unsigned long long)OCFS2_I(inode)->ip_blkno);
ccd979bd 2837
d680efe9 2838 status = ocfs2_do_unblock_meta(inode, &ctl->requeue);
ccd979bd
MF		 2839 if (status < 0)
2840 mlog_errno(status);
2841
b0697053 2842 mlog(0, "inode %llu, requeue = %d\n",
d680efe9 2843 (unsigned long long)OCFS2_I(inode)->ip_blkno, ctl->requeue);
ccd979bd
MF		 2844
2845 mlog_exit(status);
2846 return status;
2847}
2848
d680efe9
MF		 2849/*
2850 * Does the final reference drop on our dentry lock. Right now this
2851 * happens in the vote thread, but we could choose to simplify the
2852 * dlmglue API and push these off to the ocfs2_wq in the future.
2853 */
2854static void ocfs2_dentry_post_unlock(struct ocfs2_super *osb,
2855 struct ocfs2_lock_res *lockres)
2856{
2857 struct ocfs2_dentry_lock *dl = ocfs2_lock_res_dl(lockres);
2858 ocfs2_dentry_lock_put(osb, dl);
2859}
2860
2861/*
2862 * d_delete() matching dentries before the lock downconvert.
2863 *
2864 * At this point, any process waiting to destroy the
2865 * dentry_lock due to last ref count is stopped by the
2866 * OCFS2_LOCK_QUEUED flag.
2867 *
2868 * We have two potential problems
2869 *
2870 * 1) If we do the last reference drop on our dentry_lock (via dput)
2871 * we'll wind up in ocfs2_release_dentry_lock(), waiting on
2872 * the downconvert to finish. Instead we take an elevated
2873 * reference and push the drop until after we've completed our
2874 * unblock processing.
2875 *
2876 * 2) There might be another process with a final reference,
2877 * waiting on us to finish processing. If this is the case, we
2878 * detect it and exit out - there's no more dentries anyway.
2879 */
2880static int ocfs2_dentry_convert_worker(struct ocfs2_lock_res *lockres,
2881 int blocking)
2882{
2883 struct ocfs2_dentry_lock *dl = ocfs2_lock_res_dl(lockres);
2884 struct ocfs2_inode_info *oi = OCFS2_I(dl->dl_inode);
2885 struct dentry *dentry;
2886 unsigned long flags;
2887 int extra_ref = 0;
2888
2889 /*
2890 * This node is blocking another node from getting a read
2891 * lock. This happens when we've renamed within a
2892 * directory. We've forced the other nodes to d_delete(), but
2893 * we never actually dropped our lock because it's still
2894 * valid. The downconvert code will retain a PR for this node,
2895 * so there's no further work to do.
2896 */
2897 if (blocking == LKM_PRMODE)
2898 return UNBLOCK_CONTINUE;
2899
2900 /*
2901 * Mark this inode as potentially orphaned. The code in
2902 * ocfs2_delete_inode() will figure out whether it actually
2903 * needs to be freed or not.
2904 */
2905 spin_lock(&oi->ip_lock);
2906 oi->ip_flags |= OCFS2_INODE_MAYBE_ORPHANED;
2907 spin_unlock(&oi->ip_lock);
2908
2909 /*
2910 * Yuck. We need to make sure however that the check of
2911 * OCFS2_LOCK_FREEING and the extra reference are atomic with
2912 * respect to a reference decrement or the setting of that
2913 * flag.
2914 */
2915 spin_lock_irqsave(&lockres->l_lock, flags);
2916 spin_lock(&dentry_attach_lock);
2917 if (!(lockres->l_flags & OCFS2_LOCK_FREEING)
2918 && dl->dl_count) {
2919 dl->dl_count++;
2920 extra_ref = 1;
2921 }
2922 spin_unlock(&dentry_attach_lock);
2923 spin_unlock_irqrestore(&lockres->l_lock, flags);
2924
2925 mlog(0, "extra_ref = %d\n", extra_ref);
2926
2927 /*
2928 * We have a process waiting on us in ocfs2_dentry_iput(),
2929 * which means we can't have any more outstanding
2930 * aliases. There's no need to do any more work.
2931 */
2932 if (!extra_ref)
2933 return UNBLOCK_CONTINUE;
2934
2935 spin_lock(&dentry_attach_lock);
2936 while (1) {
2937 dentry = ocfs2_find_local_alias(dl->dl_inode,
2938 dl->dl_parent_blkno, 1);
2939 if (!dentry)
2940 break;
2941 spin_unlock(&dentry_attach_lock);
2942
2943 mlog(0, "d_delete(%.*s);\n", dentry->d_name.len,
2944 dentry->d_name.name);
2945
2946 /*
2947 * The following dcache calls may do an
2948 * iput(). Normally we don't want that from the
2949 * downconverting thread, but in this case it's ok
2950 * because the requesting node already has an
2951 * exclusive lock on the inode, so it can't be queued
2952 * for a downconvert.
2953 */
2954 d_delete(dentry);
2955 dput(dentry);
2956
2957 spin_lock(&dentry_attach_lock);
2958 }
2959 spin_unlock(&dentry_attach_lock);
2960
2961 /*
2962 * If we are the last holder of this dentry lock, there is no
2963 * reason to downconvert so skip straight to the unlock.
2964 */
2965 if (dl->dl_count == 1)
2966 return UNBLOCK_STOP_POST;
2967
2968 return UNBLOCK_CONTINUE_POST;
2969}
2970
2971static int ocfs2_unblock_dentry_lock(struct ocfs2_lock_res *lockres,
2972 struct ocfs2_unblock_ctl *ctl)
2973{
2974 int ret;
2975 struct ocfs2_dentry_lock *dl = ocfs2_lock_res_dl(lockres);
2976 struct ocfs2_super *osb = OCFS2_SB(dl->dl_inode->i_sb);
2977
2978 mlog(0, "unblock dentry lock: %llu\n",
2979 (unsigned long long)OCFS2_I(dl->dl_inode)->ip_blkno);
2980
2981 ret = ocfs2_generic_unblock_lock(osb,
2982 lockres,
2983 ctl,
2984 ocfs2_dentry_convert_worker);
2985 if (ret < 0)
2986 mlog_errno(ret);
2987
2988 mlog(0, "requeue = %d, post = %d\n", ctl->requeue, ctl->unblock_action);
2989
2990 return ret;
2991}
2992
ccd979bd
MF		 2993/* Generic unblock function for any lockres whose private data is an
2994 * ocfs2_super pointer. */
2995static int ocfs2_unblock_osb_lock(struct ocfs2_lock_res *lockres,
d680efe9 2996 struct ocfs2_unblock_ctl *ctl)
ccd979bd
MF		 2997{
2998 int status;
2999 struct ocfs2_super *osb;
3000
3001 mlog_entry_void();
3002
3003 mlog(0, "Unblock lockres %s\n", lockres->l_name);
3004
aa2623ad 3005 osb = ocfs2_get_lockres_osb(lockres);
ccd979bd
MF		 3006
3007 status = ocfs2_generic_unblock_lock(osb,
3008 lockres,
d680efe9 3009 ctl,
ccd979bd
MF		 3010 NULL);
3011 if (status < 0)
3012 mlog_errno(status);
3013
3014 mlog_exit(status);
3015 return status;
3016}
3017
3018void ocfs2_process_blocked_lock(struct ocfs2_super *osb,
3019 struct ocfs2_lock_res *lockres)
3020{
3021 int status;
d680efe9 3022 struct ocfs2_unblock_ctl ctl = {0, 0,};
ccd979bd
MF		 3023 unsigned long flags;
3024
3025 /* Our reference to the lockres in this function can be
3026 * considered valid until we remove the OCFS2_LOCK_QUEUED
3027 * flag. */
3028
3029 mlog_entry_void();
3030
3031 BUG_ON(!lockres);
3032 BUG_ON(!lockres->l_ops);
3033 BUG_ON(!lockres->l_ops->unblock);
3034
3035 mlog(0, "lockres %s blocked.\n", lockres->l_name);
3036
3037 /* Detect whether a lock has been marked as going away while
3038 * the vote thread was processing other things. A lock can
3039 * still be marked with OCFS2_LOCK_FREEING after this check,
3040 * but short circuiting here will still save us some
3041 * performance. */
3042 spin_lock_irqsave(&lockres->l_lock, flags);
3043 if (lockres->l_flags & OCFS2_LOCK_FREEING)
3044 goto unqueue;
3045 spin_unlock_irqrestore(&lockres->l_lock, flags);
3046
d680efe9 3047 status = lockres->l_ops->unblock(lockres, &ctl);
ccd979bd
MF		 3048 if (status < 0)
3049 mlog_errno(status);
3050
3051 spin_lock_irqsave(&lockres->l_lock, flags);
3052unqueue:
d680efe9 3053 if (lockres->l_flags & OCFS2_LOCK_FREEING || !ctl.requeue) {
ccd979bd
MF		 3054 lockres_clear_flags(lockres, OCFS2_LOCK_QUEUED);
3055 } else
3056 ocfs2_schedule_blocked_lock(osb, lockres);
3057
3058 mlog(0, "lockres %s, requeue = %s.\n", lockres->l_name,
d680efe9 3059 ctl.requeue ? "yes" : "no");
ccd979bd
MF		 3060 spin_unlock_irqrestore(&lockres->l_lock, flags);
3061
d680efe9
MF		 3062 if (ctl.unblock_action != UNBLOCK_CONTINUE
3063 && lockres->l_ops->post_unlock)
3064 lockres->l_ops->post_unlock(osb, lockres);
3065
ccd979bd
MF		 3066 mlog_exit_void();
3067}
3068
3069static void ocfs2_schedule_blocked_lock(struct ocfs2_super *osb,
3070 struct ocfs2_lock_res *lockres)
3071{
3072 mlog_entry_void();
3073
3074 assert_spin_locked(&lockres->l_lock);
3075
3076 if (lockres->l_flags & OCFS2_LOCK_FREEING) {
3077 /* Do not schedule a lock for downconvert when it's on
3078 * the way to destruction - any nodes wanting access
3079 * to the resource will get it soon. */
3080 mlog(0, "Lockres %s won't be scheduled: flags 0x%lx\n",
3081 lockres->l_name, lockres->l_flags);
3082 return;
3083 }
3084
3085 lockres_or_flags(lockres, OCFS2_LOCK_QUEUED);
3086
3087 spin_lock(&osb->vote_task_lock);
3088 if (list_empty(&lockres->l_blocked_list)) {
3089 list_add_tail(&lockres->l_blocked_list,
3090 &osb->blocked_lock_list);
3091 osb->blocked_lock_count++;
3092 }
3093 spin_unlock(&osb->vote_task_lock);
3094
3095 mlog_exit_void();
3096}
3097
3098/* This aids in debugging situations where a bad LVB might be involved. */
3099void ocfs2_dump_meta_lvb_info(u64 level,
3100 const char *function,
3101 unsigned int line,
3102 struct ocfs2_lock_res *lockres)
3103{
3104 struct ocfs2_meta_lvb *lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;
3105
3106 mlog(level, "LVB information for %s (called from %s:%u):\n",
3107 lockres->l_name, function, line);
f9e2d82e
MF		 3108 mlog(level, "version: %u, clusters: %u, generation: 0x%x\n",
3109 lvb->lvb_version, be32_to_cpu(lvb->lvb_iclusters),
3110 be32_to_cpu(lvb->lvb_igeneration));
b0697053
MF		 3111 mlog(level, "size: %llu, uid %u, gid %u, mode 0x%x\n",
3112 (unsigned long long)be64_to_cpu(lvb->lvb_isize),
3113 be32_to_cpu(lvb->lvb_iuid), be32_to_cpu(lvb->lvb_igid),
3114 be16_to_cpu(lvb->lvb_imode));
3115 mlog(level, "nlink %u, atime_packed 0x%llx, ctime_packed 0x%llx, "
ca4d147e 3116 "mtime_packed 0x%llx iattr 0x%x\n", be16_to_cpu(lvb->lvb_inlink),
b0697053
MF		 3117 (long long)be64_to_cpu(lvb->lvb_iatime_packed),
3118 (long long)be64_to_cpu(lvb->lvb_ictime_packed),
ca4d147e
HP		 3119 (long long)be64_to_cpu(lvb->lvb_imtime_packed),
3120 be32_to_cpu(lvb->lvb_iattr));
ccd979bd 3121}
Linux kernel - Deliverables
This page took 0.249182 seconds and 5 git commands to generate.