lustre/ldlm/ldlm_flock.c

   1 /*
   2  * GPL HEADER START
   3  *
   4  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License version 2 only,
   8  * as published by the Free Software Foundation.
   9  *
  10  * This program is distributed in the hope that it will be useful, but
  11  * WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  * General Public License version 2 for more details (a copy is included
  14  * in the LICENSE file that accompanied this code).
  15  *
  16  * You should have received a copy of the GNU General Public License
  17  * version 2 along with this program; If not, see
  18  * http://www.gnu.org/licenses/gpl-2.0.html
  19  *
  20  * GPL HEADER END
  21  */
  22 /*
  23  * Copyright (c) 2003 Hewlett-Packard Development Company LP.
  24  * Developed under the sponsorship of the US Government under
  25  * Subcontract No. B514193
  26  *
  27  * Copyright (c) 2003, 2010, Oracle and/or its affiliates. All rights reserved.
  28  * Use is subject to license terms.
  29  *
  30  * Copyright (c) 2010, 2014, Intel Corporation.
  31  */
  32 /*
  33  * This file is part of Lustre, http://www.lustre.org/
  34  * Lustre is a trademark of Sun Microsystems, Inc.
  35  */
  36
  37 /**
  38  * This file implements POSIX lock type for Lustre.
  39  * Its policy properties are start and end of extent and PID.
  40  *
  41  * These locks are only done through MDS due to POSIX semantics requiring
  42  * e.g. that locks could be only partially released and as such split into
  43  * two parts, and also that two adjacent locks from the same process may be
  44  * merged into a single wider lock.
  45  *
  46  * Lock modes are mapped like this:
  47  * PR and PW for READ and WRITE locks
  48  * NL to request a releasing of a portion of the lock
  49  *
  50  * These flock locks never timeout.
  51  */
  52
  53 #define DEBUG_SUBSYSTEM S_LDLM
  54
  55 #include <linux/list.h>
  56 #include <lustre_dlm.h>
  57 #include <obd_support.h>
  58 #include <obd_class.h>
  59 #include <lustre_lib.h>
  60
  61 #include "ldlm_internal.h"
  62
  63 int ldlm_flock_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc,
  64                             void *data, int flag);
  65
  66 /**
  67  * list_for_remaining_safe - iterate over the remaining entries in a list
  68  *              and safeguard against removal of a list entry.
  69  * \param pos   the &struct list_head to use as a loop counter. pos MUST
  70  *              have been initialized prior to using it in this macro.
  71  * \param n     another &struct list_head to use as temporary storage
  72  * \param head  the head for your list.
  73  */
  74 #define list_for_remaining_safe(pos, n, head) \
  75         for (n = pos->next; pos != (head); pos = n, n = pos->next)
  76
  77 static inline int
  78 ldlm_same_flock_owner(struct ldlm_lock *lock, struct ldlm_lock *new)
  79 {
  80         return((new->l_policy_data.l_flock.owner ==
  81                 lock->l_policy_data.l_flock.owner) &&
  82                (new->l_export == lock->l_export));
  83 }
  84
  85 static inline int
  86 ldlm_flocks_overlap(struct ldlm_lock *lock, struct ldlm_lock *new)
  87 {
  88         return((new->l_policy_data.l_flock.start <=
  89                 lock->l_policy_data.l_flock.end) &&
  90                (new->l_policy_data.l_flock.end >=
  91                 lock->l_policy_data.l_flock.start));
  92 }
  93
  94 static inline void ldlm_flock_blocking_link(struct ldlm_lock *req,
  95                                             struct ldlm_lock *lock)
  96 {
  97         /* For server only */
  98         if (req->l_export == NULL)
  99                 return;
 100
 101         LASSERT(hlist_unhashed(&req->l_exp_flock_hash));
 102
 103         req->l_policy_data.l_flock.blocking_owner =
 104                 lock->l_policy_data.l_flock.owner;
 105         req->l_policy_data.l_flock.blocking_export =
 106                 lock->l_export;
 107         atomic_set(&req->l_policy_data.l_flock.blocking_refs, 0);
 108
 109         cfs_hash_add(req->l_export->exp_flock_hash,
 110                      &req->l_policy_data.l_flock.owner,
 111                      &req->l_exp_flock_hash);
 112 }
 113
 114 static inline void ldlm_flock_blocking_unlink(struct ldlm_lock *req)
 115 {
 116         /* For server only */
 117         if (req->l_export == NULL)
 118                 return;
 119
 120         check_res_locked(req->l_resource);
 121         if (req->l_export->exp_flock_hash != NULL &&
 122             !hlist_unhashed(&req->l_exp_flock_hash))
 123                 cfs_hash_del(req->l_export->exp_flock_hash,
 124                              &req->l_policy_data.l_flock.owner,
 125                              &req->l_exp_flock_hash);
 126 }
 127
 128 static inline void
 129 ldlm_flock_destroy(struct ldlm_lock *lock, enum ldlm_mode mode, __u64 flags)
 130 {
 131         ENTRY;
 132
 133         LDLM_DEBUG(lock, "ldlm_flock_destroy(mode: %d, flags: %#llx)",
 134                    mode, flags);
 135
 136         /* Safe to not lock here, since it should be empty anyway */
 137         LASSERT(hlist_unhashed(&lock->l_exp_flock_hash));
 138
 139         list_del_init(&lock->l_res_link);
 140         if (flags == LDLM_FL_WAIT_NOREPROC) {
 141                 /* client side - set a flag to prevent sending a CANCEL */
 142                 lock->l_flags |= LDLM_FL_LOCAL_ONLY | LDLM_FL_CBPENDING;
 143
 144                 /* when reaching here, it is under lock_res_and_lock(). Thus,
 145                    need call the nolock version of ldlm_lock_decref_internal*/
 146                 ldlm_lock_decref_internal_nolock(lock, mode);
 147         }
 148
 149         ldlm_lock_destroy_nolock(lock);
 150         EXIT;
 151 }
 152
 153 /**
 154  * POSIX locks deadlock detection code.
 155  *
 156  * Given a new lock \a req and an existing lock \a bl_lock it conflicts
 157  * with, we need to iterate through all blocked POSIX locks for this
 158  * export and see if there is a deadlock condition arising. (i.e. when
 159  * one client holds a lock on something and want a lock on something
 160  * else and at the same time another client has the opposite situation).
 161  */
 162
 163 struct ldlm_flock_lookup_cb_data {
 164         __u64 *bl_owner;
 165         struct ldlm_lock *lock;
 166         struct obd_export *exp;
 167 };
 168
 169 static int ldlm_flock_lookup_cb(struct cfs_hash *hs, struct cfs_hash_bd *bd,
 170                                 struct hlist_node *hnode, void *data)
 171 {
 172         struct ldlm_flock_lookup_cb_data *cb_data = data;
 173         struct obd_export *exp = cfs_hash_object(hs, hnode);
 174         struct ldlm_lock *lock;
 175
 176         lock = cfs_hash_lookup(exp->exp_flock_hash, cb_data->bl_owner);
 177         if (lock == NULL)
 178                 return 0;
 179
 180         /* Stop on first found lock. Same process can't sleep twice */
 181         cb_data->lock = lock;
 182         cb_data->exp = class_export_get(exp);
 183
 184         return 1;
 185 }
 186
 187 static int
 188 ldlm_flock_deadlock(struct ldlm_lock *req, struct ldlm_lock *bl_lock)
 189 {
 190         struct obd_export *req_exp = req->l_export;
 191         struct obd_export *bl_exp = bl_lock->l_export;
 192         __u64 req_owner = req->l_policy_data.l_flock.owner;
 193         __u64 bl_owner = bl_lock->l_policy_data.l_flock.owner;
 194
 195         /* For server only */
 196         if (req_exp == NULL)
 197                 return 0;
 198
 199         class_export_get(bl_exp);
 200         while (1) {
 201                 struct ldlm_flock_lookup_cb_data cb_data = {
 202                                         .bl_owner = &bl_owner,
 203                                         .lock = NULL,
 204                                         .exp = NULL };
 205                 struct obd_export *bl_exp_new;
 206                 struct ldlm_lock *lock = NULL;
 207                 struct ldlm_flock *flock;
 208
 209                 if (bl_exp->exp_flock_hash != NULL) {
 210                         cfs_hash_for_each_key(bl_exp->exp_obd->obd_nid_hash,
 211                                 &bl_exp->exp_connection->c_peer.nid,
 212                                 ldlm_flock_lookup_cb, &cb_data);
 213                         lock = cb_data.lock;
 214                 }
 215                 if (lock == NULL)
 216                         break;
 217
 218                 class_export_put(bl_exp);
 219                 bl_exp = cb_data.exp;
 220
 221                 LASSERT(req != lock);
 222                 flock = &lock->l_policy_data.l_flock;
 223                 LASSERT(flock->owner == bl_owner);
 224                 bl_owner = flock->blocking_owner;
 225                 bl_exp_new = class_export_get(flock->blocking_export);
 226                 class_export_put(bl_exp);
 227
 228                 cfs_hash_put(bl_exp->exp_flock_hash, &lock->l_exp_flock_hash);
 229                 bl_exp = bl_exp_new;
 230
 231                 if (bl_exp->exp_failed)
 232                         break;
 233
 234                 if (bl_owner == req_owner &&
 235                     (bl_exp->exp_connection->c_peer.nid ==
 236                      req_exp->exp_connection->c_peer.nid)) {
 237                         class_export_put(bl_exp);
 238                         return 1;
 239                 }
 240         }
 241         class_export_put(bl_exp);
 242
 243         return 0;
 244 }
 245
 246 static void ldlm_flock_cancel_on_deadlock(struct ldlm_lock *lock,
 247                                           struct list_head *work_list)
 248 {
 249         CDEBUG(D_INFO, "reprocess deadlock req=%p\n", lock);
 250
 251         if ((exp_connect_flags(lock->l_export) &
 252                                 OBD_CONNECT_FLOCK_DEAD) == 0) {
 253                 CERROR("deadlock found, but client doesn't "
 254                                 "support flock canceliation\n");
 255         } else {
 256                 LASSERT(lock->l_completion_ast);
 257                 LASSERT(!ldlm_is_ast_sent(lock));
 258                 lock->l_flags |= LDLM_FL_AST_SENT | LDLM_FL_CANCEL_ON_BLOCK |
 259                         LDLM_FL_FLOCK_DEADLOCK;
 260                 ldlm_flock_blocking_unlink(lock);
 261                 ldlm_resource_unlink_lock(lock);
 262                 ldlm_add_ast_work_item(lock, NULL, work_list);
 263         }
 264 }
 265
 266 /**
 267  * Process a granting attempt for flock lock.
 268  * Must be called under ns lock held.
 269  *
 270  * This function looks for any conflicts for \a lock in the granted or
 271  * waiting queues. The lock is granted if no conflicts are found in
 272  * either queue.
 273  *
 274  * It is also responsible for splitting a lock if a portion of the lock
 275  * is released.
 276  *
 277  * If \a first_enq is 0 (ie, called from ldlm_reprocess_queue):
 278  *   - blocking ASTs have already been sent
 279  *
 280  * If \a first_enq is 1 (ie, called from ldlm_lock_enqueue):
 281  *   - blocking ASTs have not been sent yet, so list of conflicting locks
 282  *     would be collected and ASTs sent.
 283  */
 284 int
 285 ldlm_process_flock_lock(struct ldlm_lock *req, __u64 *flags, int first_enq,
 286                         enum ldlm_error *err, struct list_head *work_list)
 287 {
 288         struct ldlm_resource *res = req->l_resource;
 289         struct ldlm_namespace *ns = ldlm_res_to_ns(res);
 290         struct list_head *tmp;
 291         struct list_head *ownlocks = NULL;
 292         struct ldlm_lock *lock = NULL;
 293         struct ldlm_lock *new = req;
 294         struct ldlm_lock *new2 = NULL;
 295         enum ldlm_mode mode = req->l_req_mode;
 296         int local = ns_is_client(ns);
 297         int added = (mode == LCK_NL);
 298         int overlaps = 0;
 299         int splitted = 0;
 300         const struct ldlm_callback_suite null_cbs = { NULL };
 301         ENTRY;
 302
 303         CDEBUG(D_DLMTRACE, "flags %#llx owner %llu pid %u mode %u start "
 304                "%llu end %llu\n", *flags,
 305                new->l_policy_data.l_flock.owner,
 306                new->l_policy_data.l_flock.pid, mode,
 307                req->l_policy_data.l_flock.start,
 308                req->l_policy_data.l_flock.end);
 309
 310         *err = ELDLM_OK;
 311
 312         if (local) {
 313                 /* No blocking ASTs are sent to the clients for
 314                  * Posix file & record locks */
 315                 req->l_blocking_ast = NULL;
 316         } else {
 317                 /* Called on the server for lock cancels. */
 318                 req->l_blocking_ast = ldlm_flock_blocking_ast;
 319         }
 320
 321 reprocess:
 322         if ((*flags == LDLM_FL_WAIT_NOREPROC) || (mode == LCK_NL)) {
 323                 /* This loop determines where this processes locks start
 324                  * in the resource lr_granted list. */
 325                 list_for_each(tmp, &res->lr_granted) {
 326                         lock = list_entry(tmp, struct ldlm_lock,
 327                                               l_res_link);
 328                         if (ldlm_same_flock_owner(lock, req)) {
 329                                 ownlocks = tmp;
 330                                 break;
 331                         }
 332                 }
 333         } else {
 334                 int reprocess_failed = 0;
 335                 lockmode_verify(mode);
 336
 337                 /* This loop determines if there are existing locks
 338                  * that conflict with the new lock request. */
 339                 list_for_each(tmp, &res->lr_granted) {
 340                         lock = list_entry(tmp, struct ldlm_lock,
 341                                               l_res_link);
 342
 343                         if (ldlm_same_flock_owner(lock, req)) {
 344                                 if (!ownlocks)
 345                                         ownlocks = tmp;
 346                                 continue;
 347                         }
 348
 349                         /* locks are compatible, overlap doesn't matter */
 350                         if (lockmode_compat(lock->l_granted_mode, mode))
 351                                 continue;
 352
 353                         if (!ldlm_flocks_overlap(lock, req))
 354                                 continue;
 355
 356                         if (!first_enq) {
 357                                 reprocess_failed = 1;
 358                                 if (ldlm_flock_deadlock(req, lock)) {
 359                                         ldlm_flock_cancel_on_deadlock(req,
 360                                                         work_list);
 361                                         RETURN(LDLM_ITER_CONTINUE);
 362                                 }
 363                                 continue;
 364                         }
 365
 366                         if (*flags & LDLM_FL_BLOCK_NOWAIT) {
 367                                 ldlm_flock_destroy(req, mode, *flags);
 368                                 *err = -EAGAIN;
 369                                 RETURN(LDLM_ITER_STOP);
 370                         }
 371
 372                         if (*flags & LDLM_FL_TEST_LOCK) {
 373                                 ldlm_flock_destroy(req, mode, *flags);
 374                                 req->l_req_mode = lock->l_granted_mode;
 375                                 req->l_policy_data.l_flock.pid =
 376                                         lock->l_policy_data.l_flock.pid;
 377                                 req->l_policy_data.l_flock.start =
 378                                         lock->l_policy_data.l_flock.start;
 379                                 req->l_policy_data.l_flock.end =
 380                                         lock->l_policy_data.l_flock.end;
 381                                 *flags |= LDLM_FL_LOCK_CHANGED;
 382                                 RETURN(LDLM_ITER_STOP);
 383                         }
 384
 385                         /* add lock to blocking list before deadlock
 386                          * check to prevent race */
 387                         ldlm_flock_blocking_link(req, lock);
 388
 389                         if (ldlm_flock_deadlock(req, lock)) {
 390                                 ldlm_flock_blocking_unlink(req);
 391                                 ldlm_flock_destroy(req, mode, *flags);
 392                                 *err = -EDEADLK;
 393                                 RETURN(LDLM_ITER_STOP);
 394                         }
 395
 396                         ldlm_resource_add_lock(res, &res->lr_waiting, req);
 397                         *flags |= LDLM_FL_BLOCK_GRANTED;
 398                         RETURN(LDLM_ITER_STOP);
 399                 }
 400                 if (reprocess_failed)
 401                         RETURN(LDLM_ITER_CONTINUE);
 402         }
 403
 404         if (*flags & LDLM_FL_TEST_LOCK) {
 405                 ldlm_flock_destroy(req, mode, *flags);
 406                 req->l_req_mode = LCK_NL;
 407                 *flags |= LDLM_FL_LOCK_CHANGED;
 408                 RETURN(LDLM_ITER_STOP);
 409         }
 410
 411         /* In case we had slept on this lock request take it off of the
 412          * deadlock detection hash list. */
 413         ldlm_flock_blocking_unlink(req);
 414
 415         /* Scan the locks owned by this process that overlap this request.
 416          * We may have to merge or split existing locks. */
 417
 418         if (!ownlocks)
 419                 ownlocks = &res->lr_granted;
 420
 421         list_for_remaining_safe(ownlocks, tmp, &res->lr_granted) {
 422                 lock = list_entry(ownlocks, struct ldlm_lock, l_res_link);
 423
 424                 if (!ldlm_same_flock_owner(lock, new))
 425                         break;
 426
 427                 if (lock->l_granted_mode == mode) {
 428                         /* If the modes are the same then we need to process
 429                          * locks that overlap OR adjoin the new lock. The extra
 430                          * logic condition is necessary to deal with arithmetic
 431                          * overflow and underflow. */
 432                         if ((new->l_policy_data.l_flock.start >
 433                              (lock->l_policy_data.l_flock.end + 1))
 434                             && (lock->l_policy_data.l_flock.end !=
 435                                 OBD_OBJECT_EOF))
 436                                 continue;
 437
 438                         if ((new->l_policy_data.l_flock.end <
 439                              (lock->l_policy_data.l_flock.start - 1))
 440                             && (lock->l_policy_data.l_flock.start != 0))
 441                                 break;
 442
 443                         if (new->l_policy_data.l_flock.start <
 444                             lock->l_policy_data.l_flock.start) {
 445                                 lock->l_policy_data.l_flock.start =
 446                                         new->l_policy_data.l_flock.start;
 447                         } else {
 448                                 new->l_policy_data.l_flock.start =
 449                                         lock->l_policy_data.l_flock.start;
 450                         }
 451
 452                         if (new->l_policy_data.l_flock.end >
 453                             lock->l_policy_data.l_flock.end) {
 454                                 lock->l_policy_data.l_flock.end =
 455                                         new->l_policy_data.l_flock.end;
 456                         } else {
 457                                 new->l_policy_data.l_flock.end =
 458                                         lock->l_policy_data.l_flock.end;
 459                         }
 460
 461                         if (added) {
 462                                 ldlm_flock_destroy(lock, mode, *flags);
 463                         } else {
 464                                 new = lock;
 465                                 added = 1;
 466                         }
 467                         continue;
 468                 }
 469
 470                 if (new->l_policy_data.l_flock.start >
 471                     lock->l_policy_data.l_flock.end)
 472                         continue;
 473
 474                 if (new->l_policy_data.l_flock.end <
 475                     lock->l_policy_data.l_flock.start)
 476                         break;
 477
 478                 ++overlaps;
 479
 480                 if (new->l_policy_data.l_flock.start <=
 481                     lock->l_policy_data.l_flock.start) {
 482                         if (new->l_policy_data.l_flock.end <
 483                             lock->l_policy_data.l_flock.end) {
 484                                 lock->l_policy_data.l_flock.start =
 485                                         new->l_policy_data.l_flock.end + 1;
 486                                 break;
 487                         }
 488                         ldlm_flock_destroy(lock, lock->l_req_mode, *flags);
 489                         continue;
 490                 }
 491                 if (new->l_policy_data.l_flock.end >=
 492                     lock->l_policy_data.l_flock.end) {
 493                         lock->l_policy_data.l_flock.end =
 494                                 new->l_policy_data.l_flock.start - 1;
 495                         continue;
 496                 }
 497
 498                 /* split the existing lock into two locks */
 499
 500                 /* if this is an F_UNLCK operation then we could avoid
 501                  * allocating a new lock and use the req lock passed in
 502                  * with the request but this would complicate the reply
 503                  * processing since updates to req get reflected in the
 504                  * reply. The client side replays the lock request so
 505                  * it must see the original lock data in the reply. */
 506
 507                 /* XXX - if ldlm_lock_new() can sleep we should
 508                  * release the lr_lock, allocate the new lock,
 509                  * and restart processing this lock. */
 510                 if (new2 == NULL) {
 511                         unlock_res_and_lock(req);
 512                         new2 = ldlm_lock_create(ns, &res->lr_name, LDLM_FLOCK,
 513                                                 lock->l_granted_mode, &null_cbs,
 514                                                 NULL, 0, LVB_T_NONE);
 515                         lock_res_and_lock(req);
 516                         if (IS_ERR(new2)) {
 517                                 ldlm_flock_destroy(req, lock->l_granted_mode,
 518                                                    *flags);
 519                                 *err = PTR_ERR(new2);
 520                                 RETURN(LDLM_ITER_STOP);
 521                         }
 522                         goto reprocess;
 523                 }
 524
 525                 splitted = 1;
 526
 527                 new2->l_granted_mode = lock->l_granted_mode;
 528                 new2->l_policy_data.l_flock.pid =
 529                         new->l_policy_data.l_flock.pid;
 530                 new2->l_policy_data.l_flock.owner =
 531                         new->l_policy_data.l_flock.owner;
 532                 new2->l_policy_data.l_flock.start =
 533                         lock->l_policy_data.l_flock.start;
 534                 new2->l_policy_data.l_flock.end =
 535                         new->l_policy_data.l_flock.start - 1;
 536                 lock->l_policy_data.l_flock.start =
 537                         new->l_policy_data.l_flock.end + 1;
 538                 new2->l_conn_export = lock->l_conn_export;
 539                 if (lock->l_export != NULL) {
 540                         new2->l_export = class_export_lock_get(lock->l_export, new2);
 541                         if (new2->l_export->exp_lock_hash &&
 542                             hlist_unhashed(&new2->l_exp_hash))
 543                                 cfs_hash_add(new2->l_export->exp_lock_hash,
 544                                              &new2->l_remote_handle,
 545                                              &new2->l_exp_hash);
 546                 }
 547                 if (*flags == LDLM_FL_WAIT_NOREPROC)
 548                         ldlm_lock_addref_internal_nolock(new2,
 549                                                          lock->l_granted_mode);
 550
 551                 /* insert new2 at lock */
 552                 ldlm_resource_add_lock(res, ownlocks, new2);
 553                 LDLM_LOCK_RELEASE(new2);
 554                 break;
 555         }
 556
 557         /* if new2 is created but never used, destroy it*/
 558         if (splitted == 0 && new2 != NULL)
 559                 ldlm_lock_destroy_nolock(new2);
 560
 561         /* At this point we're granting the lock request. */
 562         req->l_granted_mode = req->l_req_mode;
 563
 564         /* Add req to the granted queue before calling ldlm_reprocess_all(). */
 565         if (!added) {
 566                 list_del_init(&req->l_res_link);
 567                 /* insert new lock before ownlocks in list. */
 568                 ldlm_resource_add_lock(res, ownlocks, req);
 569         }
 570
 571         if (*flags != LDLM_FL_WAIT_NOREPROC) {
 572 #ifdef HAVE_SERVER_SUPPORT
 573                 if (first_enq) {
 574                         /* If this is an unlock, reprocess the waitq and
 575                          * send completions ASTs for locks that can now be
 576                          * granted. The only problem with doing this
 577                          * reprocessing here is that the completion ASTs for
 578                          * newly granted locks will be sent before the unlock
 579                          * completion is sent. It shouldn't be an issue. Also
 580                          * note that ldlm_process_flock_lock() will recurse,
 581                          * but only once because first_enq will be false from
 582                          * ldlm_reprocess_queue. */
 583                         if ((mode == LCK_NL) && overlaps) {
 584                                 struct list_head rpc_list;
 585                                 int rc;
 586
 587                                 INIT_LIST_HEAD(&rpc_list);
 588 restart:
 589                                 ldlm_reprocess_queue(res, &res->lr_waiting,
 590                                                      &rpc_list);
 591
 592                                 unlock_res_and_lock(req);
 593                                 rc = ldlm_run_ast_work(ns, &rpc_list,
 594                                                        LDLM_WORK_CP_AST);
 595                                 lock_res_and_lock(req);
 596                                 if (rc == -ERESTART)
 597                                         GOTO(restart, rc);
 598                        }
 599                 } else {
 600                         LASSERT(req->l_completion_ast);
 601                         ldlm_add_ast_work_item(req, NULL, work_list);
 602                 }
 603 #else /* !HAVE_SERVER_SUPPORT */
 604                 /* The only one possible case for client-side calls flock
 605                  * policy function is ldlm_flock_completion_ast inside which
 606                  * carries LDLM_FL_WAIT_NOREPROC flag. */
 607                 CERROR("Illegal parameter for client-side-only module.\n");
 608                 LBUG();
 609 #endif /* HAVE_SERVER_SUPPORT */
 610         }
 611
 612         /* In case we're reprocessing the requested lock we can't destroy
 613          * it until after calling ldlm_add_ast_work_item() above so that laawi()
 614          * can bump the reference count on \a req. Otherwise \a req
 615          * could be freed before the completion AST can be sent.  */
 616         if (added)
 617                 ldlm_flock_destroy(req, mode, *flags);
 618
 619         ldlm_resource_dump(D_INFO, res);
 620         RETURN(LDLM_ITER_CONTINUE);
 621 }
 622
 623 struct ldlm_flock_wait_data {
 624         struct ldlm_lock *fwd_lock;
 625         int               fwd_generation;
 626 };
 627
 628 static void
 629 ldlm_flock_interrupted_wait(void *data)
 630 {
 631         struct ldlm_lock *lock;
 632         ENTRY;
 633
 634         lock = ((struct ldlm_flock_wait_data *)data)->fwd_lock;
 635
 636         /* take lock off the deadlock detection hash list. */
 637         lock_res_and_lock(lock);
 638         ldlm_flock_blocking_unlink(lock);
 639
 640         /* client side - set flag to prevent lock from being put on LRU list */
 641         ldlm_set_cbpending(lock);
 642         unlock_res_and_lock(lock);
 643
 644         EXIT;
 645 }
 646
 647 /**
 648  * Flock completion callback function.
 649  *
 650  * \param lock [in,out]: A lock to be handled
 651  * \param flags    [in]: flags
 652  * \param *data    [in]: ldlm_work_cp_ast_lock() will use ldlm_cb_set_arg
 653  *
 654  * \retval 0    : success
 655  * \retval <0   : failure
 656  */
 657 int
 658 ldlm_flock_completion_ast(struct ldlm_lock *lock, __u64 flags, void *data)
 659 {
 660         struct file_lock *getlk = lock->l_ast_data;
 661         struct obd_device *obd;
 662         struct obd_import *imp = NULL;
 663         struct ldlm_flock_wait_data fwd;
 664         struct l_wait_info lwi;
 665         enum ldlm_error err;
 666         int rc = 0;
 667         ENTRY;
 668
 669         OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_CP_CB_WAIT2, 4);
 670         if (OBD_FAIL_PRECHECK(OBD_FAIL_LDLM_CP_CB_WAIT3)) {
 671                 lock_res_and_lock(lock);
 672                 lock->l_flags |= LDLM_FL_FAIL_LOC;
 673                 unlock_res_and_lock(lock);
 674                 OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_CP_CB_WAIT3, 4);
 675         }
 676         CDEBUG(D_DLMTRACE, "flags: %#llx data: %p getlk: %p\n",
 677                flags, data, getlk);
 678
 679         LASSERT(flags != LDLM_FL_WAIT_NOREPROC);
 680
 681         if (flags & LDLM_FL_FAILED)
 682                 goto granted;
 683
 684         if (!(flags & LDLM_FL_BLOCKED_MASK)) {
 685                 if (NULL == data)
 686                         /* mds granted the lock in the reply */
 687                         goto granted;
 688                 /* CP AST RPC: lock get granted, wake it up */
 689                 wake_up(&lock->l_waitq);
 690                 RETURN(0);
 691         }
 692
 693         LDLM_DEBUG(lock, "client-side enqueue returned a blocked lock, "
 694                    "sleeping");
 695         fwd.fwd_lock = lock;
 696         obd = class_exp2obd(lock->l_conn_export);
 697
 698         /* if this is a local lock, there is no import */
 699         if (NULL != obd)
 700                 imp = obd->u.cli.cl_import;
 701
 702         if (NULL != imp) {
 703                 spin_lock(&imp->imp_lock);
 704                 fwd.fwd_generation = imp->imp_generation;
 705                 spin_unlock(&imp->imp_lock);
 706         }
 707
 708         lwi = LWI_TIMEOUT_INTR(0, NULL, ldlm_flock_interrupted_wait, &fwd);
 709
 710         /* Go to sleep until the lock is granted. */
 711         rc = l_wait_event(lock->l_waitq, is_granted_or_cancelled(lock), &lwi);
 712
 713         if (rc) {
 714                 LDLM_DEBUG(lock, "client-side enqueue waking up: failed (%d)",
 715                            rc);
 716                 RETURN(rc);
 717         }
 718
 719 granted:
 720         OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_CP_CB_WAIT, 10);
 721
 722         if (OBD_FAIL_PRECHECK(OBD_FAIL_LDLM_CP_CB_WAIT4)) {
 723                 lock_res_and_lock(lock);
 724                 /* DEADLOCK is always set with CBPENDING */
 725                 lock->l_flags |= LDLM_FL_FLOCK_DEADLOCK | LDLM_FL_CBPENDING;
 726                 unlock_res_and_lock(lock);
 727                 OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_CP_CB_WAIT4, 4);
 728         }
 729         if (OBD_FAIL_PRECHECK(OBD_FAIL_LDLM_CP_CB_WAIT5)) {
 730                 lock_res_and_lock(lock);
 731                 /* DEADLOCK is always set with CBPENDING */
 732                 lock->l_flags |= LDLM_FL_FAIL_LOC |
 733                                  LDLM_FL_FLOCK_DEADLOCK | LDLM_FL_CBPENDING;
 734                 unlock_res_and_lock(lock);
 735                 OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_CP_CB_WAIT5, 4);
 736         }
 737
 738         lock_res_and_lock(lock);
 739
 740
 741         /* Protect against race where lock could have been just destroyed
 742          * due to overlap in ldlm_process_flock_lock().
 743          */
 744         if (ldlm_is_destroyed(lock)) {
 745                 unlock_res_and_lock(lock);
 746                 LDLM_DEBUG(lock, "client-side enqueue waking up: destroyed");
 747
 748                 /* An error is still to be returned, to propagate it up to
 749                  * ldlm_cli_enqueue_fini() caller. */
 750                 RETURN(-EIO);
 751         }
 752
 753         /* ldlm_lock_enqueue() has already placed lock on the granted list. */
 754         ldlm_resource_unlink_lock(lock);
 755
 756         /* Import invalidation. We need to actually release the lock
 757          * references being held, so that it can go away. No point in
 758          * holding the lock even if app still believes it has it, since
 759          * server already dropped it anyway. Only for granted locks too. */
 760         /* Do the same for DEADLOCK'ed locks. */
 761         if (ldlm_is_failed(lock) || ldlm_is_flock_deadlock(lock)) {
 762                 int mode;
 763
 764                 if (flags & LDLM_FL_TEST_LOCK)
 765                         LASSERT(ldlm_is_test_lock(lock));
 766
 767                 if (ldlm_is_test_lock(lock) || ldlm_is_flock_deadlock(lock))
 768                         mode = flock_type(getlk);
 769                 else
 770                         mode = lock->l_granted_mode;
 771
 772                 if (ldlm_is_flock_deadlock(lock)) {
 773                         LDLM_DEBUG(lock, "client-side enqueue deadlock "
 774                                    "received");
 775                         rc = -EDEADLK;
 776                 }
 777                 ldlm_flock_destroy(lock, mode, LDLM_FL_WAIT_NOREPROC);
 778                 unlock_res_and_lock(lock);
 779
 780                 /* Need to wake up the waiter if we were evicted */
 781                 wake_up(&lock->l_waitq);
 782
 783                 /* An error is still to be returned, to propagate it up to
 784                  * ldlm_cli_enqueue_fini() caller. */
 785                 RETURN(rc ? : -EIO);
 786         }
 787
 788         LDLM_DEBUG(lock, "client-side enqueue granted");
 789
 790         if (flags & LDLM_FL_TEST_LOCK) {
 791                 /* fcntl(F_GETLK) request */
 792                 /* The old mode was saved in getlk->fl_type so that if the mode
 793                  * in the lock changes we can decref the appropriate refcount.*/
 794                 LASSERT(ldlm_is_test_lock(lock));
 795                 ldlm_flock_destroy(lock, flock_type(getlk),
 796                                    LDLM_FL_WAIT_NOREPROC);
 797                 switch (lock->l_granted_mode) {
 798                 case LCK_PR:
 799                         flock_set_type(getlk, F_RDLCK);
 800                         break;
 801                 case LCK_PW:
 802                         flock_set_type(getlk, F_WRLCK);
 803                         break;
 804                 default:
 805                         flock_set_type(getlk, F_UNLCK);
 806                 }
 807                 flock_set_pid(getlk, (pid_t)lock->l_policy_data.l_flock.pid);
 808                 flock_set_start(getlk,
 809                                 (loff_t)lock->l_policy_data.l_flock.start);
 810                 flock_set_end(getlk,
 811                               (loff_t)lock->l_policy_data.l_flock.end);
 812         } else {
 813                 __u64 noreproc = LDLM_FL_WAIT_NOREPROC;
 814
 815                 /* We need to reprocess the lock to do merges or splits
 816                  * with existing locks owned by this process. */
 817                 ldlm_process_flock_lock(lock, &noreproc, 1, &err, NULL);
 818         }
 819         unlock_res_and_lock(lock);
 820         RETURN(rc);
 821 }
 822 EXPORT_SYMBOL(ldlm_flock_completion_ast);
 823
 824 int ldlm_flock_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc,
 825                             void *data, int flag)
 826 {
 827         ENTRY;
 828
 829         LASSERT(lock);
 830         LASSERT(flag == LDLM_CB_CANCELING);
 831
 832         /* take lock off the deadlock detection hash list. */
 833         lock_res_and_lock(lock);
 834         ldlm_flock_blocking_unlink(lock);
 835         unlock_res_and_lock(lock);
 836         RETURN(0);
 837 }
 838
 839 void ldlm_flock_policy_wire_to_local(const union ldlm_wire_policy_data *wpolicy,
 840                                      union ldlm_policy_data *lpolicy)
 841 {
 842         lpolicy->l_flock.start = wpolicy->l_flock.lfw_start;
 843         lpolicy->l_flock.end = wpolicy->l_flock.lfw_end;
 844         lpolicy->l_flock.pid = wpolicy->l_flock.lfw_pid;
 845         lpolicy->l_flock.owner = wpolicy->l_flock.lfw_owner;
 846 }
 847
 848 void ldlm_flock_policy_local_to_wire(const union ldlm_policy_data *lpolicy,
 849                                      union ldlm_wire_policy_data *wpolicy)
 850 {
 851         memset(wpolicy, 0, sizeof(*wpolicy));
 852         wpolicy->l_flock.lfw_start = lpolicy->l_flock.start;
 853         wpolicy->l_flock.lfw_end = lpolicy->l_flock.end;
 854         wpolicy->l_flock.lfw_pid = lpolicy->l_flock.pid;
 855         wpolicy->l_flock.lfw_owner = lpolicy->l_flock.owner;
 856 }
 857
 858 /*
 859  * Export handle<->flock hash operations.
 860  */
 861 static unsigned
 862 ldlm_export_flock_hash(struct cfs_hash *hs, const void *key, unsigned mask)
 863 {
 864         return cfs_hash_u64_hash(*(__u64 *)key, mask);
 865 }
 866
 867 static void *
 868 ldlm_export_flock_key(struct hlist_node *hnode)
 869 {
 870         struct ldlm_lock *lock;
 871
 872         lock = hlist_entry(hnode, struct ldlm_lock, l_exp_flock_hash);
 873         return &lock->l_policy_data.l_flock.owner;
 874 }
 875
 876 static int
 877 ldlm_export_flock_keycmp(const void *key, struct hlist_node *hnode)
 878 {
 879         return !memcmp(ldlm_export_flock_key(hnode), key, sizeof(__u64));
 880 }
 881
 882 static void *
 883 ldlm_export_flock_object(struct hlist_node *hnode)
 884 {
 885         return hlist_entry(hnode, struct ldlm_lock, l_exp_flock_hash);
 886 }
 887
 888 static void
 889 ldlm_export_flock_get(struct cfs_hash *hs, struct hlist_node *hnode)
 890 {
 891         struct ldlm_lock *lock;
 892         struct ldlm_flock *flock;
 893
 894         lock = hlist_entry(hnode, struct ldlm_lock, l_exp_flock_hash);
 895         LDLM_LOCK_GET(lock);
 896
 897         flock = &lock->l_policy_data.l_flock;
 898         LASSERT(flock->blocking_export != NULL);
 899         class_export_get(flock->blocking_export);
 900         atomic_inc(&flock->blocking_refs);
 901 }
 902
 903 static void
 904 ldlm_export_flock_put(struct cfs_hash *hs, struct hlist_node *hnode)
 905 {
 906         struct ldlm_lock *lock;
 907         struct ldlm_flock *flock;
 908
 909         lock = hlist_entry(hnode, struct ldlm_lock, l_exp_flock_hash);
 910
 911         flock = &lock->l_policy_data.l_flock;
 912         LASSERT(flock->blocking_export != NULL);
 913         class_export_put(flock->blocking_export);
 914         if (atomic_dec_and_test(&flock->blocking_refs)) {
 915                 flock->blocking_owner = 0;
 916                 flock->blocking_export = NULL;
 917         }
 918         LDLM_LOCK_RELEASE(lock);
 919 }
 920
 921 static struct cfs_hash_ops ldlm_export_flock_ops = {
 922         .hs_hash        = ldlm_export_flock_hash,
 923         .hs_key         = ldlm_export_flock_key,
 924         .hs_keycmp      = ldlm_export_flock_keycmp,
 925         .hs_object      = ldlm_export_flock_object,
 926         .hs_get         = ldlm_export_flock_get,
 927         .hs_put         = ldlm_export_flock_put,
 928         .hs_put_locked  = ldlm_export_flock_put,
 929 };
 930
 931 int ldlm_init_flock_export(struct obd_export *exp)
 932 {
 933         if( strcmp(exp->exp_obd->obd_type->typ_name, LUSTRE_MDT_NAME) != 0)
 934                 RETURN(0);
 935
 936         exp->exp_flock_hash =
 937                 cfs_hash_create(obd_uuid2str(&exp->exp_client_uuid),
 938                                 HASH_EXP_LOCK_CUR_BITS,
 939                                 HASH_EXP_LOCK_MAX_BITS,
 940                                 HASH_EXP_LOCK_BKT_BITS, 0,
 941                                 CFS_HASH_MIN_THETA, CFS_HASH_MAX_THETA,
 942                                 &ldlm_export_flock_ops,
 943                                 CFS_HASH_DEFAULT | CFS_HASH_NBLK_CHANGE);
 944         if (!exp->exp_flock_hash)
 945                 RETURN(-ENOMEM);
 946
 947         RETURN(0);
 948 }
 949
 950 void ldlm_destroy_flock_export(struct obd_export *exp)
 951 {
 952         ENTRY;
 953         if (exp->exp_flock_hash) {
 954                 cfs_hash_putref(exp->exp_flock_hash);
 955                 exp->exp_flock_hash = NULL;
 956         }
 957         EXIT;
 958 }