lustre/ldlm/ldlm_lock.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.sun.com/software/products/lustre/docs/GPLv2.pdf
  19  *
  20  * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
  21  * CA 95054 USA or visit www.sun.com if you need additional information or
  22  * have any questions.
  23  *
  24  * GPL HEADER END
  25  */
  26 /*
  27  * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
  28  * Use is subject to license terms.
  29  *
  30  * Copyright (c) 2011, 2012, Whamcloud, Inc.
  31  */
  32 /*
  33  * This file is part of Lustre, http://www.lustre.org/
  34  * Lustre is a trademark of Sun Microsystems, Inc.
  35  *
  36  * lustre/ldlm/ldlm_lock.c
  37  *
  38  * Author: Peter Braam <braam@clusterfs.com>
  39  * Author: Phil Schwan <phil@clusterfs.com>
  40  */
  41
  42 #define DEBUG_SUBSYSTEM S_LDLM
  43
  44 #ifdef __KERNEL__
  45 # include <libcfs/libcfs.h>
  46 # include <linux/lustre_intent.h>
  47 #else
  48 # include <liblustre.h>
  49 #endif
  50
  51 #include <obd_class.h>
  52 #include "ldlm_internal.h"
  53
  54 /* lock types */
  55 char *ldlm_lockname[] = {
  56         [0] "--",
  57         [LCK_EX] "EX",
  58         [LCK_PW] "PW",
  59         [LCK_PR] "PR",
  60         [LCK_CW] "CW",
  61         [LCK_CR] "CR",
  62         [LCK_NL] "NL",
  63         [LCK_GROUP] "GROUP",
  64         [LCK_COS] "COS"
  65 };
  66 EXPORT_SYMBOL(ldlm_lockname);
  67
  68 char *ldlm_typename[] = {
  69         [LDLM_PLAIN] "PLN",
  70         [LDLM_EXTENT] "EXT",
  71         [LDLM_FLOCK] "FLK",
  72         [LDLM_IBITS] "IBT",
  73 };
  74 EXPORT_SYMBOL(ldlm_typename);
  75
  76 static ldlm_policy_wire_to_local_t ldlm_policy_wire18_to_local[] = {
  77         [LDLM_PLAIN - LDLM_MIN_TYPE] ldlm_plain_policy_wire_to_local,
  78         [LDLM_EXTENT - LDLM_MIN_TYPE] ldlm_extent_policy_wire_to_local,
  79         [LDLM_FLOCK - LDLM_MIN_TYPE] ldlm_flock_policy_wire18_to_local,
  80         [LDLM_IBITS - LDLM_MIN_TYPE] ldlm_ibits_policy_wire_to_local,
  81 };
  82
  83 static ldlm_policy_wire_to_local_t ldlm_policy_wire21_to_local[] = {
  84         [LDLM_PLAIN - LDLM_MIN_TYPE] ldlm_plain_policy_wire_to_local,
  85         [LDLM_EXTENT - LDLM_MIN_TYPE] ldlm_extent_policy_wire_to_local,
  86         [LDLM_FLOCK - LDLM_MIN_TYPE] ldlm_flock_policy_wire21_to_local,
  87         [LDLM_IBITS - LDLM_MIN_TYPE] ldlm_ibits_policy_wire_to_local,
  88 };
  89
  90 static ldlm_policy_local_to_wire_t ldlm_policy_local_to_wire[] = {
  91         [LDLM_PLAIN - LDLM_MIN_TYPE] ldlm_plain_policy_local_to_wire,
  92         [LDLM_EXTENT - LDLM_MIN_TYPE] ldlm_extent_policy_local_to_wire,
  93         [LDLM_FLOCK - LDLM_MIN_TYPE] ldlm_flock_policy_local_to_wire,
  94         [LDLM_IBITS - LDLM_MIN_TYPE] ldlm_ibits_policy_local_to_wire,
  95 };
  96
  97 /**
  98  * Converts lock policy from local format to on the wire lock_desc format
  99  */
 100 void ldlm_convert_policy_to_wire(ldlm_type_t type,
 101                                  const ldlm_policy_data_t *lpolicy,
 102                                  ldlm_wire_policy_data_t *wpolicy)
 103 {
 104         ldlm_policy_local_to_wire_t convert;
 105
 106         convert = ldlm_policy_local_to_wire[type - LDLM_MIN_TYPE];
 107
 108         convert(lpolicy, wpolicy);
 109 }
 110
 111 /**
 112  * Converts lock policy from on the wire lock_desc format to local format
 113  */
 114 void ldlm_convert_policy_to_local(struct obd_export *exp, ldlm_type_t type,
 115                                   const ldlm_wire_policy_data_t *wpolicy,
 116                                   ldlm_policy_data_t *lpolicy)
 117 {
 118         ldlm_policy_wire_to_local_t convert;
 119         int new_client;
 120
 121         /** some badnes for 2.0.0 clients, but 2.0.0 isn't supported */
 122         new_client = (exp->exp_connect_flags & OBD_CONNECT_FULL20) != 0;
 123         if (new_client)
 124                convert = ldlm_policy_wire21_to_local[type - LDLM_MIN_TYPE];
 125         else
 126                convert = ldlm_policy_wire18_to_local[type - LDLM_MIN_TYPE];
 127
 128         convert(wpolicy, lpolicy);
 129 }
 130
 131 char *ldlm_it2str(int it)
 132 {
 133         switch (it) {
 134         case IT_OPEN:
 135                 return "open";
 136         case IT_CREAT:
 137                 return "creat";
 138         case (IT_OPEN | IT_CREAT):
 139                 return "open|creat";
 140         case IT_READDIR:
 141                 return "readdir";
 142         case IT_GETATTR:
 143                 return "getattr";
 144         case IT_LOOKUP:
 145                 return "lookup";
 146         case IT_UNLINK:
 147                 return "unlink";
 148         case IT_GETXATTR:
 149                 return "getxattr";
 150         case IT_LAYOUT:
 151                 return "layout";
 152         default:
 153                 CERROR("Unknown intent %d\n", it);
 154                 return "UNKNOWN";
 155         }
 156 }
 157 EXPORT_SYMBOL(ldlm_it2str);
 158
 159 extern cfs_mem_cache_t *ldlm_lock_slab;
 160
 161 #ifdef HAVE_SERVER_SUPPORT
 162 static ldlm_processing_policy ldlm_processing_policy_table[] = {
 163         [LDLM_PLAIN] ldlm_process_plain_lock,
 164         [LDLM_EXTENT] ldlm_process_extent_lock,
 165 # ifdef __KERNEL__
 166         [LDLM_FLOCK] ldlm_process_flock_lock,
 167 # endif
 168         [LDLM_IBITS] ldlm_process_inodebits_lock,
 169 };
 170
 171 ldlm_processing_policy ldlm_get_processing_policy(struct ldlm_resource *res)
 172 {
 173         return ldlm_processing_policy_table[res->lr_type];
 174 }
 175 EXPORT_SYMBOL(ldlm_get_processing_policy);
 176 #endif /* HAVE_SERVER_SUPPORT */
 177
 178 void ldlm_register_intent(struct ldlm_namespace *ns, ldlm_res_policy arg)
 179 {
 180         ns->ns_policy = arg;
 181 }
 182 EXPORT_SYMBOL(ldlm_register_intent);
 183
 184 /*
 185  * REFCOUNTED LOCK OBJECTS
 186  */
 187
 188
 189 /*
 190  * Lock refcounts, during creation:
 191  *   - one special one for allocation, dec'd only once in destroy
 192  *   - one for being a lock that's in-use
 193  *   - one for the addref associated with a new lock
 194  */
 195 struct ldlm_lock *ldlm_lock_get(struct ldlm_lock *lock)
 196 {
 197         cfs_atomic_inc(&lock->l_refc);
 198         return lock;
 199 }
 200 EXPORT_SYMBOL(ldlm_lock_get);
 201
 202 void ldlm_lock_put(struct ldlm_lock *lock)
 203 {
 204         ENTRY;
 205
 206         LASSERT(lock->l_resource != LP_POISON);
 207         LASSERT(cfs_atomic_read(&lock->l_refc) > 0);
 208         if (cfs_atomic_dec_and_test(&lock->l_refc)) {
 209                 struct ldlm_resource *res;
 210
 211                 LDLM_DEBUG(lock,
 212                            "final lock_put on destroyed lock, freeing it.");
 213
 214                 res = lock->l_resource;
 215                 LASSERT(lock->l_destroyed);
 216                 LASSERT(cfs_list_empty(&lock->l_res_link));
 217                 LASSERT(cfs_list_empty(&lock->l_pending_chain));
 218
 219                 lprocfs_counter_decr(ldlm_res_to_ns(res)->ns_stats,
 220                                      LDLM_NSS_LOCKS);
 221                 lu_ref_del(&res->lr_reference, "lock", lock);
 222                 ldlm_resource_putref(res);
 223                 lock->l_resource = NULL;
 224                 if (lock->l_export) {
 225                         class_export_lock_put(lock->l_export, lock);
 226                         lock->l_export = NULL;
 227                 }
 228
 229                 if (lock->l_lvb_data != NULL)
 230                         OBD_FREE(lock->l_lvb_data, lock->l_lvb_len);
 231
 232                 ldlm_interval_free(ldlm_interval_detach(lock));
 233                 lu_ref_fini(&lock->l_reference);
 234                 OBD_FREE_RCU(lock, sizeof(*lock), &lock->l_handle);
 235         }
 236
 237         EXIT;
 238 }
 239 EXPORT_SYMBOL(ldlm_lock_put);
 240
 241 int ldlm_lock_remove_from_lru_nolock(struct ldlm_lock *lock)
 242 {
 243         int rc = 0;
 244         if (!cfs_list_empty(&lock->l_lru)) {
 245                 struct ldlm_namespace *ns = ldlm_lock_to_ns(lock);
 246
 247                 LASSERT(lock->l_resource->lr_type != LDLM_FLOCK);
 248                 cfs_list_del_init(&lock->l_lru);
 249                 if (lock->l_flags & LDLM_FL_SKIPPED)
 250                         lock->l_flags &= ~LDLM_FL_SKIPPED;
 251                 LASSERT(ns->ns_nr_unused > 0);
 252                 ns->ns_nr_unused--;
 253                 rc = 1;
 254         }
 255         return rc;
 256 }
 257
 258 int ldlm_lock_remove_from_lru(struct ldlm_lock *lock)
 259 {
 260         struct ldlm_namespace *ns = ldlm_lock_to_ns(lock);
 261         int rc;
 262
 263         ENTRY;
 264         if (lock->l_ns_srv) {
 265                 LASSERT(cfs_list_empty(&lock->l_lru));
 266                 RETURN(0);
 267         }
 268
 269         cfs_spin_lock(&ns->ns_lock);
 270         rc = ldlm_lock_remove_from_lru_nolock(lock);
 271         cfs_spin_unlock(&ns->ns_lock);
 272         EXIT;
 273         return rc;
 274 }
 275
 276 void ldlm_lock_add_to_lru_nolock(struct ldlm_lock *lock)
 277 {
 278         struct ldlm_namespace *ns = ldlm_lock_to_ns(lock);
 279
 280         lock->l_last_used = cfs_time_current();
 281         LASSERT(cfs_list_empty(&lock->l_lru));
 282         LASSERT(lock->l_resource->lr_type != LDLM_FLOCK);
 283         cfs_list_add_tail(&lock->l_lru, &ns->ns_unused_list);
 284         LASSERT(ns->ns_nr_unused >= 0);
 285         ns->ns_nr_unused++;
 286 }
 287
 288 void ldlm_lock_add_to_lru(struct ldlm_lock *lock)
 289 {
 290         struct ldlm_namespace *ns = ldlm_lock_to_ns(lock);
 291
 292         ENTRY;
 293         cfs_spin_lock(&ns->ns_lock);
 294         ldlm_lock_add_to_lru_nolock(lock);
 295         cfs_spin_unlock(&ns->ns_lock);
 296         EXIT;
 297 }
 298
 299 void ldlm_lock_touch_in_lru(struct ldlm_lock *lock)
 300 {
 301         struct ldlm_namespace *ns = ldlm_lock_to_ns(lock);
 302
 303         ENTRY;
 304         if (lock->l_ns_srv) {
 305                 LASSERT(cfs_list_empty(&lock->l_lru));
 306                 EXIT;
 307                 return;
 308         }
 309
 310         cfs_spin_lock(&ns->ns_lock);
 311         if (!cfs_list_empty(&lock->l_lru)) {
 312                 ldlm_lock_remove_from_lru_nolock(lock);
 313                 ldlm_lock_add_to_lru_nolock(lock);
 314         }
 315         cfs_spin_unlock(&ns->ns_lock);
 316         EXIT;
 317 }
 318
 319 /* This used to have a 'strict' flag, which recovery would use to mark an
 320  * in-use lock as needing-to-die.  Lest I am ever tempted to put it back, I
 321  * shall explain why it's gone: with the new hash table scheme, once you call
 322  * ldlm_lock_destroy, you can never drop your final references on this lock.
 323  * Because it's not in the hash table anymore.  -phil */
 324 int ldlm_lock_destroy_internal(struct ldlm_lock *lock)
 325 {
 326         ENTRY;
 327
 328         if (lock->l_readers || lock->l_writers) {
 329                 LDLM_ERROR(lock, "lock still has references");
 330                 LBUG();
 331         }
 332
 333         if (!cfs_list_empty(&lock->l_res_link)) {
 334                 LDLM_ERROR(lock, "lock still on resource");
 335                 LBUG();
 336         }
 337
 338         if (lock->l_destroyed) {
 339                 LASSERT(cfs_list_empty(&lock->l_lru));
 340                 EXIT;
 341                 return 0;
 342         }
 343         lock->l_destroyed = 1;
 344
 345         if (lock->l_export && lock->l_export->exp_lock_hash) {
 346                 /* NB: it's safe to call cfs_hash_del() even lock isn't
 347                  * in exp_lock_hash. */
 348                 /* In the function below, .hs_keycmp resolves to
 349                  * ldlm_export_lock_keycmp() */
 350                 /* coverity[overrun-buffer-val] */
 351                 cfs_hash_del(lock->l_export->exp_lock_hash,
 352                              &lock->l_remote_handle, &lock->l_exp_hash);
 353         }
 354
 355         ldlm_lock_remove_from_lru(lock);
 356         class_handle_unhash(&lock->l_handle);
 357
 358 #if 0
 359         /* Wake anyone waiting for this lock */
 360         /* FIXME: I should probably add yet another flag, instead of using
 361          * l_export to only call this on clients */
 362         if (lock->l_export)
 363                 class_export_put(lock->l_export);
 364         lock->l_export = NULL;
 365         if (lock->l_export && lock->l_completion_ast)
 366                 lock->l_completion_ast(lock, 0);
 367 #endif
 368         EXIT;
 369         return 1;
 370 }
 371
 372 void ldlm_lock_destroy(struct ldlm_lock *lock)
 373 {
 374         int first;
 375         ENTRY;
 376         lock_res_and_lock(lock);
 377         first = ldlm_lock_destroy_internal(lock);
 378         unlock_res_and_lock(lock);
 379
 380         /* drop reference from hashtable only for first destroy */
 381         if (first) {
 382                 lu_ref_del(&lock->l_reference, "hash", lock);
 383                 LDLM_LOCK_RELEASE(lock);
 384         }
 385         EXIT;
 386 }
 387
 388 void ldlm_lock_destroy_nolock(struct ldlm_lock *lock)
 389 {
 390         int first;
 391         ENTRY;
 392         first = ldlm_lock_destroy_internal(lock);
 393         /* drop reference from hashtable only for first destroy */
 394         if (first) {
 395                 lu_ref_del(&lock->l_reference, "hash", lock);
 396                 LDLM_LOCK_RELEASE(lock);
 397         }
 398         EXIT;
 399 }
 400
 401 /* this is called by portals_handle2object with the handle lock taken */
 402 static void lock_handle_addref(void *lock)
 403 {
 404         LDLM_LOCK_GET((struct ldlm_lock *)lock);
 405 }
 406
 407 static void lock_handle_free(void *lock, int size)
 408 {
 409         LASSERT(size == sizeof(struct ldlm_lock));
 410         OBD_SLAB_FREE(lock, ldlm_lock_slab, size);
 411 }
 412
 413 struct portals_handle_ops lock_handle_ops = {
 414         .hop_addref = lock_handle_addref,
 415         .hop_free   = lock_handle_free,
 416 };
 417
 418 /*
 419  * usage: pass in a resource on which you have done ldlm_resource_get
 420  *        new lock will take over the refcount.
 421  * returns: lock with refcount 2 - one for current caller and one for remote
 422  */
 423 static struct ldlm_lock *ldlm_lock_new(struct ldlm_resource *resource)
 424 {
 425         struct ldlm_lock *lock;
 426         ENTRY;
 427
 428         if (resource == NULL)
 429                 LBUG();
 430
 431         OBD_SLAB_ALLOC_PTR_GFP(lock, ldlm_lock_slab, CFS_ALLOC_IO);
 432         if (lock == NULL)
 433                 RETURN(NULL);
 434
 435         cfs_spin_lock_init(&lock->l_lock);
 436         lock->l_resource = resource;
 437         lu_ref_add(&resource->lr_reference, "lock", lock);
 438
 439         cfs_atomic_set(&lock->l_refc, 2);
 440         CFS_INIT_LIST_HEAD(&lock->l_res_link);
 441         CFS_INIT_LIST_HEAD(&lock->l_lru);
 442         CFS_INIT_LIST_HEAD(&lock->l_pending_chain);
 443         CFS_INIT_LIST_HEAD(&lock->l_bl_ast);
 444         CFS_INIT_LIST_HEAD(&lock->l_cp_ast);
 445         CFS_INIT_LIST_HEAD(&lock->l_rk_ast);
 446         cfs_waitq_init(&lock->l_waitq);
 447         lock->l_blocking_lock = NULL;
 448         CFS_INIT_LIST_HEAD(&lock->l_sl_mode);
 449         CFS_INIT_LIST_HEAD(&lock->l_sl_policy);
 450         CFS_INIT_HLIST_NODE(&lock->l_exp_hash);
 451         CFS_INIT_HLIST_NODE(&lock->l_exp_flock_hash);
 452
 453         lprocfs_counter_incr(ldlm_res_to_ns(resource)->ns_stats,
 454                              LDLM_NSS_LOCKS);
 455         CFS_INIT_LIST_HEAD(&lock->l_handle.h_link);
 456         class_handle_hash(&lock->l_handle, &lock_handle_ops);
 457
 458         lu_ref_init(&lock->l_reference);
 459         lu_ref_add(&lock->l_reference, "hash", lock);
 460         lock->l_callback_timeout = 0;
 461
 462 #if LUSTRE_TRACKS_LOCK_EXP_REFS
 463         CFS_INIT_LIST_HEAD(&lock->l_exp_refs_link);
 464         lock->l_exp_refs_nr = 0;
 465         lock->l_exp_refs_target = NULL;
 466 #endif
 467         CFS_INIT_LIST_HEAD(&lock->l_exp_list);
 468
 469         RETURN(lock);
 470 }
 471
 472 int ldlm_lock_change_resource(struct ldlm_namespace *ns, struct ldlm_lock *lock,
 473                               const struct ldlm_res_id *new_resid)
 474 {
 475         struct ldlm_resource *oldres = lock->l_resource;
 476         struct ldlm_resource *newres;
 477         int type;
 478         ENTRY;
 479
 480         LASSERT(ns_is_client(ns));
 481
 482         lock_res_and_lock(lock);
 483         if (memcmp(new_resid, &lock->l_resource->lr_name,
 484                    sizeof(lock->l_resource->lr_name)) == 0) {
 485                 /* Nothing to do */
 486                 unlock_res_and_lock(lock);
 487                 RETURN(0);
 488         }
 489
 490         LASSERT(new_resid->name[0] != 0);
 491
 492         /* This function assumes that the lock isn't on any lists */
 493         LASSERT(cfs_list_empty(&lock->l_res_link));
 494
 495         type = oldres->lr_type;
 496         unlock_res_and_lock(lock);
 497
 498         newres = ldlm_resource_get(ns, NULL, new_resid, type, 1);
 499         if (newres == NULL)
 500                 RETURN(-ENOMEM);
 501
 502         lu_ref_add(&newres->lr_reference, "lock", lock);
 503         /*
 504          * To flip the lock from the old to the new resource, lock, oldres and
 505          * newres have to be locked. Resource spin-locks are nested within
 506          * lock->l_lock, and are taken in the memory address order to avoid
 507          * dead-locks.
 508          */
 509         cfs_spin_lock(&lock->l_lock);
 510         oldres = lock->l_resource;
 511         if (oldres < newres) {
 512                 lock_res(oldres);
 513                 lock_res_nested(newres, LRT_NEW);
 514         } else {
 515                 lock_res(newres);
 516                 lock_res_nested(oldres, LRT_NEW);
 517         }
 518         LASSERT(memcmp(new_resid, &oldres->lr_name,
 519                        sizeof oldres->lr_name) != 0);
 520         lock->l_resource = newres;
 521         unlock_res(oldres);
 522         unlock_res_and_lock(lock);
 523
 524         /* ...and the flowers are still standing! */
 525         lu_ref_del(&oldres->lr_reference, "lock", lock);
 526         ldlm_resource_putref(oldres);
 527
 528         RETURN(0);
 529 }
 530 EXPORT_SYMBOL(ldlm_lock_change_resource);
 531
 532 /*
 533  *  HANDLES
 534  */
 535
 536 void ldlm_lock2handle(const struct ldlm_lock *lock, struct lustre_handle *lockh)
 537 {
 538         lockh->cookie = lock->l_handle.h_cookie;
 539 }
 540 EXPORT_SYMBOL(ldlm_lock2handle);
 541
 542 /* if flags: atomically get the lock and set the flags.
 543  *           Return NULL if flag already set
 544  */
 545
 546 struct ldlm_lock *__ldlm_handle2lock(const struct lustre_handle *handle,
 547                                      int flags)
 548 {
 549         struct ldlm_lock *lock;
 550         ENTRY;
 551
 552         LASSERT(handle);
 553
 554         lock = class_handle2object(handle->cookie);
 555         if (lock == NULL)
 556                 RETURN(NULL);
 557
 558         /* It's unlikely but possible that someone marked the lock as
 559          * destroyed after we did handle2object on it */
 560         if (flags == 0 && !lock->l_destroyed) {
 561                 lu_ref_add(&lock->l_reference, "handle", cfs_current());
 562                 RETURN(lock);
 563         }
 564
 565         lock_res_and_lock(lock);
 566
 567         LASSERT(lock->l_resource != NULL);
 568
 569         lu_ref_add_atomic(&lock->l_reference, "handle", cfs_current());
 570         if (unlikely(lock->l_destroyed)) {
 571                 unlock_res_and_lock(lock);
 572                 CDEBUG(D_INFO, "lock already destroyed: lock %p\n", lock);
 573                 LDLM_LOCK_PUT(lock);
 574                 RETURN(NULL);
 575         }
 576
 577         if (flags && (lock->l_flags & flags)) {
 578                 unlock_res_and_lock(lock);
 579                 LDLM_LOCK_PUT(lock);
 580                 RETURN(NULL);
 581         }
 582
 583         if (flags)
 584                 lock->l_flags |= flags;
 585
 586         unlock_res_and_lock(lock);
 587         RETURN(lock);
 588 }
 589 EXPORT_SYMBOL(__ldlm_handle2lock);
 590
 591 void ldlm_lock2desc(struct ldlm_lock *lock, struct ldlm_lock_desc *desc)
 592 {
 593         struct obd_export *exp = lock->l_export?:lock->l_conn_export;
 594         /* INODEBITS_INTEROP: If the other side does not support
 595          * inodebits, reply with a plain lock descriptor.
 596          */
 597         if ((lock->l_resource->lr_type == LDLM_IBITS) &&
 598             (exp && !(exp->exp_connect_flags & OBD_CONNECT_IBITS))) {
 599                 /* Make sure all the right bits are set in this lock we
 600                    are going to pass to client */
 601                 LASSERTF(lock->l_policy_data.l_inodebits.bits ==
 602                          (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_UPDATE |
 603                           MDS_INODELOCK_LAYOUT),
 604                          "Inappropriate inode lock bits during "
 605                          "conversion " LPU64 "\n",
 606                          lock->l_policy_data.l_inodebits.bits);
 607
 608                 ldlm_res2desc(lock->l_resource, &desc->l_resource);
 609                 desc->l_resource.lr_type = LDLM_PLAIN;
 610
 611                 /* Convert "new" lock mode to something old client can
 612                    understand */
 613                 if ((lock->l_req_mode == LCK_CR) ||
 614                     (lock->l_req_mode == LCK_CW))
 615                         desc->l_req_mode = LCK_PR;
 616                 else
 617                         desc->l_req_mode = lock->l_req_mode;
 618                 if ((lock->l_granted_mode == LCK_CR) ||
 619                     (lock->l_granted_mode == LCK_CW)) {
 620                         desc->l_granted_mode = LCK_PR;
 621                 } else {
 622                         /* We never grant PW/EX locks to clients */
 623                         LASSERT((lock->l_granted_mode != LCK_PW) &&
 624                                 (lock->l_granted_mode != LCK_EX));
 625                         desc->l_granted_mode = lock->l_granted_mode;
 626                 }
 627
 628                 /* We do not copy policy here, because there is no
 629                    policy for plain locks */
 630         } else {
 631                 ldlm_res2desc(lock->l_resource, &desc->l_resource);
 632                 desc->l_req_mode = lock->l_req_mode;
 633                 desc->l_granted_mode = lock->l_granted_mode;
 634                 ldlm_convert_policy_to_wire(lock->l_resource->lr_type,
 635                                             &lock->l_policy_data,
 636                                             &desc->l_policy_data);
 637         }
 638 }
 639 EXPORT_SYMBOL(ldlm_lock2desc);
 640
 641 void ldlm_add_bl_work_item(struct ldlm_lock *lock, struct ldlm_lock *new,
 642                            cfs_list_t *work_list)
 643 {
 644         if ((lock->l_flags & LDLM_FL_AST_SENT) == 0) {
 645                 LDLM_DEBUG(lock, "lock incompatible; sending blocking AST.");
 646                 lock->l_flags |= LDLM_FL_AST_SENT;
 647                 /* If the enqueuing client said so, tell the AST recipient to
 648                  * discard dirty data, rather than writing back. */
 649                 if (new->l_flags & LDLM_AST_DISCARD_DATA)
 650                         lock->l_flags |= LDLM_FL_DISCARD_DATA;
 651                 LASSERT(cfs_list_empty(&lock->l_bl_ast));
 652                 cfs_list_add(&lock->l_bl_ast, work_list);
 653                 LDLM_LOCK_GET(lock);
 654                 LASSERT(lock->l_blocking_lock == NULL);
 655                 lock->l_blocking_lock = LDLM_LOCK_GET(new);
 656         }
 657 }
 658
 659 void ldlm_add_cp_work_item(struct ldlm_lock *lock, cfs_list_t *work_list)
 660 {
 661         if ((lock->l_flags & LDLM_FL_CP_REQD) == 0) {
 662                 lock->l_flags |= LDLM_FL_CP_REQD;
 663                 LDLM_DEBUG(lock, "lock granted; sending completion AST.");
 664                 LASSERT(cfs_list_empty(&lock->l_cp_ast));
 665                 cfs_list_add(&lock->l_cp_ast, work_list);
 666                 LDLM_LOCK_GET(lock);
 667         }
 668 }
 669
 670 /* must be called with lr_lock held */
 671 void ldlm_add_ast_work_item(struct ldlm_lock *lock, struct ldlm_lock *new,
 672                             cfs_list_t *work_list)
 673 {
 674         ENTRY;
 675         check_res_locked(lock->l_resource);
 676         if (new)
 677                 ldlm_add_bl_work_item(lock, new, work_list);
 678         else
 679                 ldlm_add_cp_work_item(lock, work_list);
 680         EXIT;
 681 }
 682
 683 void ldlm_lock_addref(struct lustre_handle *lockh, __u32 mode)
 684 {
 685         struct ldlm_lock *lock;
 686
 687         lock = ldlm_handle2lock(lockh);
 688         LASSERT(lock != NULL);
 689         ldlm_lock_addref_internal(lock, mode);
 690         LDLM_LOCK_PUT(lock);
 691 }
 692 EXPORT_SYMBOL(ldlm_lock_addref);
 693
 694 void ldlm_lock_addref_internal_nolock(struct ldlm_lock *lock, __u32 mode)
 695 {
 696         ldlm_lock_remove_from_lru(lock);
 697         if (mode & (LCK_NL | LCK_CR | LCK_PR)) {
 698                 lock->l_readers++;
 699                 lu_ref_add_atomic(&lock->l_reference, "reader", lock);
 700         }
 701         if (mode & (LCK_EX | LCK_CW | LCK_PW | LCK_GROUP | LCK_COS)) {
 702                 lock->l_writers++;
 703                 lu_ref_add_atomic(&lock->l_reference, "writer", lock);
 704         }
 705         LDLM_LOCK_GET(lock);
 706         lu_ref_add_atomic(&lock->l_reference, "user", lock);
 707         LDLM_DEBUG(lock, "ldlm_lock_addref(%s)", ldlm_lockname[mode]);
 708 }
 709
 710 /**
 711  * Attempts to addref a lock, and fails if lock is already LDLM_FL_CBPENDING
 712  * or destroyed.
 713  *
 714  * \retval 0 success, lock was addref-ed
 715  *
 716  * \retval -EAGAIN lock is being canceled.
 717  */
 718 int ldlm_lock_addref_try(struct lustre_handle *lockh, __u32 mode)
 719 {
 720         struct ldlm_lock *lock;
 721         int               result;
 722
 723         result = -EAGAIN;
 724         lock = ldlm_handle2lock(lockh);
 725         if (lock != NULL) {
 726                 lock_res_and_lock(lock);
 727                 if (lock->l_readers != 0 || lock->l_writers != 0 ||
 728                     !(lock->l_flags & LDLM_FL_CBPENDING)) {
 729                         ldlm_lock_addref_internal_nolock(lock, mode);
 730                         result = 0;
 731                 }
 732                 unlock_res_and_lock(lock);
 733                 LDLM_LOCK_PUT(lock);
 734         }
 735         return result;
 736 }
 737 EXPORT_SYMBOL(ldlm_lock_addref_try);
 738
 739 /* only called for local locks */
 740 void ldlm_lock_addref_internal(struct ldlm_lock *lock, __u32 mode)
 741 {
 742         lock_res_and_lock(lock);
 743         ldlm_lock_addref_internal_nolock(lock, mode);
 744         unlock_res_and_lock(lock);
 745 }
 746
 747 /* only called in ldlm_flock_destroy and for local locks.
 748  *  * for LDLM_FLOCK type locks, l_blocking_ast is null, and
 749  *   * ldlm_lock_remove_from_lru() does nothing, it is safe
 750  *    * for ldlm_flock_destroy usage by dropping some code */
 751 void ldlm_lock_decref_internal_nolock(struct ldlm_lock *lock, __u32 mode)
 752 {
 753         LDLM_DEBUG(lock, "ldlm_lock_decref(%s)", ldlm_lockname[mode]);
 754         if (mode & (LCK_NL | LCK_CR | LCK_PR)) {
 755                 LASSERT(lock->l_readers > 0);
 756                 lu_ref_del(&lock->l_reference, "reader", lock);
 757                 lock->l_readers--;
 758         }
 759         if (mode & (LCK_EX | LCK_CW | LCK_PW | LCK_GROUP | LCK_COS)) {
 760                 LASSERT(lock->l_writers > 0);
 761                 lu_ref_del(&lock->l_reference, "writer", lock);
 762                 lock->l_writers--;
 763         }
 764
 765         lu_ref_del(&lock->l_reference, "user", lock);
 766         LDLM_LOCK_RELEASE(lock);    /* matches the LDLM_LOCK_GET() in addref */
 767 }
 768
 769 void ldlm_lock_decref_internal(struct ldlm_lock *lock, __u32 mode)
 770 {
 771         struct ldlm_namespace *ns;
 772         ENTRY;
 773
 774         lock_res_and_lock(lock);
 775
 776         ns = ldlm_lock_to_ns(lock);
 777
 778         ldlm_lock_decref_internal_nolock(lock, mode);
 779
 780         if (lock->l_flags & LDLM_FL_LOCAL &&
 781             !lock->l_readers && !lock->l_writers) {
 782                 /* If this is a local lock on a server namespace and this was
 783                  * the last reference, cancel the lock. */
 784                 CDEBUG(D_INFO, "forcing cancel of local lock\n");
 785                 lock->l_flags |= LDLM_FL_CBPENDING;
 786         }
 787
 788         if (!lock->l_readers && !lock->l_writers &&
 789             (lock->l_flags & LDLM_FL_CBPENDING)) {
 790                 /* If we received a blocked AST and this was the last reference,
 791                  * run the callback. */
 792                 if (lock->l_ns_srv && lock->l_export)
 793                         CERROR("FL_CBPENDING set on non-local lock--just a "
 794                                "warning\n");
 795
 796                 LDLM_DEBUG(lock, "final decref done on cbpending lock");
 797
 798                 LDLM_LOCK_GET(lock); /* dropped by bl thread */
 799                 ldlm_lock_remove_from_lru(lock);
 800                 unlock_res_and_lock(lock);
 801
 802                 if (lock->l_flags & LDLM_FL_FAIL_LOC)
 803                         OBD_RACE(OBD_FAIL_LDLM_CP_BL_RACE);
 804
 805                 if ((lock->l_flags & LDLM_FL_ATOMIC_CB) ||
 806                     ldlm_bl_to_thread_lock(ns, NULL, lock) != 0)
 807                         ldlm_handle_bl_callback(ns, NULL, lock);
 808         } else if (ns_is_client(ns) &&
 809                    !lock->l_readers && !lock->l_writers &&
 810                    !(lock->l_flags & LDLM_FL_NO_LRU) &&
 811                    !(lock->l_flags & LDLM_FL_BL_AST)) {
 812
 813                 LDLM_DEBUG(lock, "add lock into lru list");
 814
 815                 /* If this is a client-side namespace and this was the last
 816                  * reference, put it on the LRU. */
 817                 ldlm_lock_add_to_lru(lock);
 818                 unlock_res_and_lock(lock);
 819
 820                 if (lock->l_flags & LDLM_FL_FAIL_LOC)
 821                         OBD_RACE(OBD_FAIL_LDLM_CP_BL_RACE);
 822
 823                 /* Call ldlm_cancel_lru() only if EARLY_CANCEL and LRU RESIZE
 824                  * are not supported by the server, otherwise, it is done on
 825                  * enqueue. */
 826                 if (!exp_connect_cancelset(lock->l_conn_export) &&
 827                     !ns_connect_lru_resize(ns))
 828                         ldlm_cancel_lru(ns, 0, LDLM_ASYNC, 0);
 829         } else {
 830                 LDLM_DEBUG(lock, "do not add lock into lru list");
 831                 unlock_res_and_lock(lock);
 832         }
 833
 834         EXIT;
 835 }
 836
 837 void ldlm_lock_decref(struct lustre_handle *lockh, __u32 mode)
 838 {
 839         struct ldlm_lock *lock = __ldlm_handle2lock(lockh, 0);
 840         LASSERTF(lock != NULL, "Non-existing lock: "LPX64"\n", lockh->cookie);
 841         ldlm_lock_decref_internal(lock, mode);
 842         LDLM_LOCK_PUT(lock);
 843 }
 844 EXPORT_SYMBOL(ldlm_lock_decref);
 845
 846 /* This will drop a lock reference and mark it for destruction, but will not
 847  * necessarily cancel the lock before returning. */
 848 void ldlm_lock_decref_and_cancel(struct lustre_handle *lockh, __u32 mode)
 849 {
 850         struct ldlm_lock *lock = __ldlm_handle2lock(lockh, 0);
 851         ENTRY;
 852
 853         LASSERT(lock != NULL);
 854
 855         LDLM_DEBUG(lock, "ldlm_lock_decref(%s)", ldlm_lockname[mode]);
 856         lock_res_and_lock(lock);
 857         lock->l_flags |= LDLM_FL_CBPENDING;
 858         unlock_res_and_lock(lock);
 859         ldlm_lock_decref_internal(lock, mode);
 860         LDLM_LOCK_PUT(lock);
 861 }
 862 EXPORT_SYMBOL(ldlm_lock_decref_and_cancel);
 863
 864 struct sl_insert_point {
 865         cfs_list_t *res_link;
 866         cfs_list_t *mode_link;
 867         cfs_list_t *policy_link;
 868 };
 869
 870 /*
 871  * search_granted_lock
 872  *
 873  * Description:
 874  *      Finds a position to insert the new lock.
 875  * Parameters:
 876  *      queue [input]:  the granted list where search acts on;
 877  *      req [input]:    the lock whose position to be located;
 878  *      prev [output]:  positions within 3 lists to insert @req to
 879  * Return Value:
 880  *      filled @prev
 881  * NOTE: called by
 882  *  - ldlm_grant_lock_with_skiplist
 883  */
 884 static void search_granted_lock(cfs_list_t *queue,
 885                                 struct ldlm_lock *req,
 886                                 struct sl_insert_point *prev)
 887 {
 888         cfs_list_t *tmp;
 889         struct ldlm_lock *lock, *mode_end, *policy_end;
 890         ENTRY;
 891
 892         cfs_list_for_each(tmp, queue) {
 893                 lock = cfs_list_entry(tmp, struct ldlm_lock, l_res_link);
 894
 895                 mode_end = cfs_list_entry(lock->l_sl_mode.prev,
 896                                           struct ldlm_lock, l_sl_mode);
 897
 898                 if (lock->l_req_mode != req->l_req_mode) {
 899                         /* jump to last lock of mode group */
 900                         tmp = &mode_end->l_res_link;
 901                         continue;
 902                 }
 903
 904                 /* suitable mode group is found */
 905                 if (lock->l_resource->lr_type == LDLM_PLAIN) {
 906                         /* insert point is last lock of the mode group */
 907                         prev->res_link = &mode_end->l_res_link;
 908                         prev->mode_link = &mode_end->l_sl_mode;
 909                         prev->policy_link = &req->l_sl_policy;
 910                         EXIT;
 911                         return;
 912                 } else if (lock->l_resource->lr_type == LDLM_IBITS) {
 913                         for (;;) {
 914                                 policy_end =
 915                                         cfs_list_entry(lock->l_sl_policy.prev,
 916                                                        struct ldlm_lock,
 917                                                        l_sl_policy);
 918
 919                                 if (lock->l_policy_data.l_inodebits.bits ==
 920                                     req->l_policy_data.l_inodebits.bits) {
 921                                         /* insert point is last lock of
 922                                          * the policy group */
 923                                         prev->res_link =
 924                                                 &policy_end->l_res_link;
 925                                         prev->mode_link =
 926                                                 &policy_end->l_sl_mode;
 927                                         prev->policy_link =
 928                                                 &policy_end->l_sl_policy;
 929                                         EXIT;
 930                                         return;
 931                                 }
 932
 933                                 if (policy_end == mode_end)
 934                                         /* done with mode group */
 935                                         break;
 936
 937                                 /* go to next policy group within mode group */
 938                                 tmp = policy_end->l_res_link.next;
 939                                 lock = cfs_list_entry(tmp, struct ldlm_lock,
 940                                                       l_res_link);
 941                         }  /* loop over policy groups within the mode group */
 942
 943                         /* insert point is last lock of the mode group,
 944                          * new policy group is started */
 945                         prev->res_link = &mode_end->l_res_link;
 946                         prev->mode_link = &mode_end->l_sl_mode;
 947                         prev->policy_link = &req->l_sl_policy;
 948                         EXIT;
 949                         return;
 950                 } else {
 951                         LDLM_ERROR(lock,"is not LDLM_PLAIN or LDLM_IBITS lock");
 952                         LBUG();
 953                 }
 954         }
 955
 956         /* insert point is last lock on the queue,
 957          * new mode group and new policy group are started */
 958         prev->res_link = queue->prev;
 959         prev->mode_link = &req->l_sl_mode;
 960         prev->policy_link = &req->l_sl_policy;
 961         EXIT;
 962         return;
 963 }
 964
 965 static void ldlm_granted_list_add_lock(struct ldlm_lock *lock,
 966                                        struct sl_insert_point *prev)
 967 {
 968         struct ldlm_resource *res = lock->l_resource;
 969         ENTRY;
 970
 971         check_res_locked(res);
 972
 973         ldlm_resource_dump(D_INFO, res);
 974         LDLM_DEBUG(lock, "About to add lock:");
 975
 976         if (lock->l_destroyed) {
 977                 CDEBUG(D_OTHER, "Lock destroyed, not adding to resource\n");
 978                 return;
 979         }
 980
 981         LASSERT(cfs_list_empty(&lock->l_res_link));
 982         LASSERT(cfs_list_empty(&lock->l_sl_mode));
 983         LASSERT(cfs_list_empty(&lock->l_sl_policy));
 984
 985         cfs_list_add(&lock->l_res_link, prev->res_link);
 986         cfs_list_add(&lock->l_sl_mode, prev->mode_link);
 987         cfs_list_add(&lock->l_sl_policy, prev->policy_link);
 988
 989         EXIT;
 990 }
 991
 992 static void ldlm_grant_lock_with_skiplist(struct ldlm_lock *lock)
 993 {
 994         struct sl_insert_point prev;
 995         ENTRY;
 996
 997         LASSERT(lock->l_req_mode == lock->l_granted_mode);
 998
 999         search_granted_lock(&lock->l_resource->lr_granted, lock, &prev);
1000         ldlm_granted_list_add_lock(lock, &prev);
1001         EXIT;
1002 }
1003
1004 /* NOTE: called by
1005  *  - ldlm_lock_enqueue
1006  *  - ldlm_reprocess_queue
1007  *  - ldlm_lock_convert
1008  *
1009  * must be called with lr_lock held
1010  */
1011 void ldlm_grant_lock(struct ldlm_lock *lock, cfs_list_t *work_list)
1012 {
1013         struct ldlm_resource *res = lock->l_resource;
1014         ENTRY;
1015
1016         check_res_locked(res);
1017
1018         lock->l_granted_mode = lock->l_req_mode;
1019         if (res->lr_type == LDLM_PLAIN || res->lr_type == LDLM_IBITS)
1020                 ldlm_grant_lock_with_skiplist(lock);
1021         else if (res->lr_type == LDLM_EXTENT)
1022                 ldlm_extent_add_lock(res, lock);
1023         else
1024                 ldlm_resource_add_lock(res, &res->lr_granted, lock);
1025
1026         if (lock->l_granted_mode < res->lr_most_restr)
1027                 res->lr_most_restr = lock->l_granted_mode;
1028
1029         if (work_list && lock->l_completion_ast != NULL)
1030                 ldlm_add_ast_work_item(lock, NULL, work_list);
1031
1032         ldlm_pool_add(&ldlm_res_to_ns(res)->ns_pool, lock);
1033         EXIT;
1034 }
1035
1036 /* returns a referenced lock or NULL.  See the flag descriptions below, in the
1037  * comment above ldlm_lock_match */
1038 static struct ldlm_lock *search_queue(cfs_list_t *queue,
1039                                       ldlm_mode_t *mode,
1040                                       ldlm_policy_data_t *policy,
1041                                       struct ldlm_lock *old_lock,
1042                                       int flags, int unref)
1043 {
1044         struct ldlm_lock *lock;
1045         cfs_list_t       *tmp;
1046
1047         cfs_list_for_each(tmp, queue) {
1048                 ldlm_mode_t match;
1049
1050                 lock = cfs_list_entry(tmp, struct ldlm_lock, l_res_link);
1051
1052                 if (lock == old_lock)
1053                         break;
1054
1055                 /* llite sometimes wants to match locks that will be
1056                  * canceled when their users drop, but we allow it to match
1057                  * if it passes in CBPENDING and the lock still has users.
1058                  * this is generally only going to be used by children
1059                  * whose parents already hold a lock so forward progress
1060                  * can still happen. */
1061                 if (lock->l_flags & LDLM_FL_CBPENDING &&
1062                     !(flags & LDLM_FL_CBPENDING))
1063                         continue;
1064                 if (!unref && lock->l_flags & LDLM_FL_CBPENDING &&
1065                     lock->l_readers == 0 && lock->l_writers == 0)
1066                         continue;
1067
1068                 if (!(lock->l_req_mode & *mode))
1069                         continue;
1070                 match = lock->l_req_mode;
1071
1072                 if (lock->l_resource->lr_type == LDLM_EXTENT &&
1073                     (lock->l_policy_data.l_extent.start >
1074                      policy->l_extent.start ||
1075                      lock->l_policy_data.l_extent.end < policy->l_extent.end))
1076                         continue;
1077
1078                 if (unlikely(match == LCK_GROUP) &&
1079                     lock->l_resource->lr_type == LDLM_EXTENT &&
1080                     lock->l_policy_data.l_extent.gid != policy->l_extent.gid)
1081                         continue;
1082
1083                 /* We match if we have existing lock with same or wider set
1084                    of bits. */
1085                 if (lock->l_resource->lr_type == LDLM_IBITS &&
1086                      ((lock->l_policy_data.l_inodebits.bits &
1087                       policy->l_inodebits.bits) !=
1088                       policy->l_inodebits.bits))
1089                         continue;
1090
1091                 if (!unref &&
1092                     (lock->l_destroyed || lock->l_flags & LDLM_FL_FAILED ||
1093                      lock->l_failed))
1094                         continue;
1095
1096                 if ((flags & LDLM_FL_LOCAL_ONLY) &&
1097                     !(lock->l_flags & LDLM_FL_LOCAL))
1098                         continue;
1099
1100                 if (flags & LDLM_FL_TEST_LOCK) {
1101                         LDLM_LOCK_GET(lock);
1102                         ldlm_lock_touch_in_lru(lock);
1103                 } else {
1104                         ldlm_lock_addref_internal_nolock(lock, match);
1105                 }
1106                 *mode = match;
1107                 return lock;
1108         }
1109
1110         return NULL;
1111 }
1112
1113 void ldlm_lock_fail_match_locked(struct ldlm_lock *lock)
1114 {
1115         if (!lock->l_failed) {
1116                 lock->l_failed = 1;
1117                 cfs_waitq_broadcast(&lock->l_waitq);
1118         }
1119 }
1120 EXPORT_SYMBOL(ldlm_lock_fail_match_locked);
1121
1122 void ldlm_lock_fail_match(struct ldlm_lock *lock)
1123 {
1124         lock_res_and_lock(lock);
1125         ldlm_lock_fail_match_locked(lock);
1126         unlock_res_and_lock(lock);
1127 }
1128 EXPORT_SYMBOL(ldlm_lock_fail_match);
1129
1130 void ldlm_lock_allow_match_locked(struct ldlm_lock *lock)
1131 {
1132         lock->l_flags |= LDLM_FL_LVB_READY;
1133         cfs_waitq_broadcast(&lock->l_waitq);
1134 }
1135 EXPORT_SYMBOL(ldlm_lock_allow_match_locked);
1136
1137 void ldlm_lock_allow_match(struct ldlm_lock *lock)
1138 {
1139         lock_res_and_lock(lock);
1140         ldlm_lock_allow_match_locked(lock);
1141         unlock_res_and_lock(lock);
1142 }
1143 EXPORT_SYMBOL(ldlm_lock_allow_match);
1144
1145 /* Can be called in two ways:
1146  *
1147  * If 'ns' is NULL, then lockh describes an existing lock that we want to look
1148  * for a duplicate of.
1149  *
1150  * Otherwise, all of the fields must be filled in, to match against.
1151  *
1152  * If 'flags' contains LDLM_FL_LOCAL_ONLY, then only match local locks on the
1153  *     server (ie, connh is NULL)
1154  * If 'flags' contains LDLM_FL_BLOCK_GRANTED, then only locks on the granted
1155  *     list will be considered
1156  * If 'flags' contains LDLM_FL_CBPENDING, then locks that have been marked
1157  *     to be canceled can still be matched as long as they still have reader
1158  *     or writer refernces
1159  * If 'flags' contains LDLM_FL_TEST_LOCK, then don't actually reference a lock,
1160  *     just tell us if we would have matched.
1161  *
1162  * Returns 1 if it finds an already-existing lock that is compatible; in this
1163  * case, lockh is filled in with a addref()ed lock
1164  *
1165  * we also check security context, if that failed we simply return 0 (to keep
1166  * caller code unchanged), the context failure will be discovered by caller
1167  * sometime later.
1168  */
1169 ldlm_mode_t ldlm_lock_match(struct ldlm_namespace *ns, int flags,
1170                             const struct ldlm_res_id *res_id, ldlm_type_t type,
1171                             ldlm_policy_data_t *policy, ldlm_mode_t mode,
1172                             struct lustre_handle *lockh, int unref)
1173 {
1174         struct ldlm_resource *res;
1175         struct ldlm_lock *lock, *old_lock = NULL;
1176         int rc = 0;
1177         ENTRY;
1178
1179         if (ns == NULL) {
1180                 old_lock = ldlm_handle2lock(lockh);
1181                 LASSERT(old_lock);
1182
1183                 ns = ldlm_lock_to_ns(old_lock);
1184                 res_id = &old_lock->l_resource->lr_name;
1185                 type = old_lock->l_resource->lr_type;
1186                 mode = old_lock->l_req_mode;
1187         }
1188
1189         res = ldlm_resource_get(ns, NULL, res_id, type, 0);
1190         if (res == NULL) {
1191                 LASSERT(old_lock == NULL);
1192                 RETURN(0);
1193         }
1194
1195         LDLM_RESOURCE_ADDREF(res);
1196         lock_res(res);
1197
1198         lock = search_queue(&res->lr_granted, &mode, policy, old_lock,
1199                             flags, unref);
1200         if (lock != NULL)
1201                 GOTO(out, rc = 1);
1202         if (flags & LDLM_FL_BLOCK_GRANTED)
1203                 GOTO(out, rc = 0);
1204         lock = search_queue(&res->lr_converting, &mode, policy, old_lock,
1205                             flags, unref);
1206         if (lock != NULL)
1207                 GOTO(out, rc = 1);
1208         lock = search_queue(&res->lr_waiting, &mode, policy, old_lock,
1209                             flags, unref);
1210         if (lock != NULL)
1211                 GOTO(out, rc = 1);
1212
1213         EXIT;
1214  out:
1215         unlock_res(res);
1216         LDLM_RESOURCE_DELREF(res);
1217         ldlm_resource_putref(res);
1218
1219         if (lock) {
1220                 ldlm_lock2handle(lock, lockh);
1221                 if ((flags & LDLM_FL_LVB_READY) &&
1222                     (!(lock->l_flags & LDLM_FL_LVB_READY))) {
1223                         struct l_wait_info lwi;
1224                         if (lock->l_completion_ast) {
1225                                 int err = lock->l_completion_ast(lock,
1226                                                           LDLM_FL_WAIT_NOREPROC,
1227                                                                  NULL);
1228                                 if (err) {
1229                                         if (flags & LDLM_FL_TEST_LOCK)
1230                                                 LDLM_LOCK_RELEASE(lock);
1231                                         else
1232                                                 ldlm_lock_decref_internal(lock,
1233                                                                           mode);
1234                                         rc = 0;
1235                                         goto out2;
1236                                 }
1237                         }
1238
1239                         lwi = LWI_TIMEOUT_INTR(cfs_time_seconds(obd_timeout),
1240                                                NULL, LWI_ON_SIGNAL_NOOP, NULL);
1241
1242                         /* XXX FIXME see comment on CAN_MATCH in lustre_dlm.h */
1243                         l_wait_event(lock->l_waitq,
1244                                      lock->l_flags & LDLM_FL_LVB_READY ||
1245                                      lock->l_failed,
1246                                      &lwi);
1247                         if (!(lock->l_flags & LDLM_FL_LVB_READY)) {
1248                                 if (flags & LDLM_FL_TEST_LOCK)
1249                                         LDLM_LOCK_RELEASE(lock);
1250                                 else
1251                                         ldlm_lock_decref_internal(lock, mode);
1252                                 rc = 0;
1253                         }
1254                 }
1255         }
1256  out2:
1257         if (rc) {
1258                 LDLM_DEBUG(lock, "matched ("LPU64" "LPU64")",
1259                            (type == LDLM_PLAIN || type == LDLM_IBITS) ?
1260                                 res_id->name[2] : policy->l_extent.start,
1261                            (type == LDLM_PLAIN || type == LDLM_IBITS) ?
1262                                 res_id->name[3] : policy->l_extent.end);
1263
1264                 /* check user's security context */
1265                 if (lock->l_conn_export &&
1266                     sptlrpc_import_check_ctx(
1267                                 class_exp2cliimp(lock->l_conn_export))) {
1268                         if (!(flags & LDLM_FL_TEST_LOCK))
1269                                 ldlm_lock_decref_internal(lock, mode);
1270                         rc = 0;
1271                 }
1272
1273                 if (flags & LDLM_FL_TEST_LOCK)
1274                         LDLM_LOCK_RELEASE(lock);
1275
1276         } else if (!(flags & LDLM_FL_TEST_LOCK)) {/*less verbose for test-only*/
1277                 LDLM_DEBUG_NOLOCK("not matched ns %p type %u mode %u res "
1278                                   LPU64"/"LPU64" ("LPU64" "LPU64")", ns,
1279                                   type, mode, res_id->name[0], res_id->name[1],
1280                                   (type == LDLM_PLAIN || type == LDLM_IBITS) ?
1281                                         res_id->name[2] :policy->l_extent.start,
1282                                   (type == LDLM_PLAIN || type == LDLM_IBITS) ?
1283                                         res_id->name[3] : policy->l_extent.end);
1284         }
1285         if (old_lock)
1286                 LDLM_LOCK_PUT(old_lock);
1287
1288         return rc ? mode : 0;
1289 }
1290 EXPORT_SYMBOL(ldlm_lock_match);
1291
1292 ldlm_mode_t ldlm_revalidate_lock_handle(struct lustre_handle *lockh,
1293                                         __u64 *bits)
1294 {
1295         struct ldlm_lock *lock;
1296         ldlm_mode_t mode = 0;
1297         ENTRY;
1298
1299         lock = ldlm_handle2lock(lockh);
1300         if (lock != NULL) {
1301                 lock_res_and_lock(lock);
1302                 if (lock->l_destroyed || lock->l_flags & LDLM_FL_FAILED ||
1303                     lock->l_failed)
1304                         GOTO(out, mode);
1305
1306                 if (lock->l_flags & LDLM_FL_CBPENDING &&
1307                     lock->l_readers == 0 && lock->l_writers == 0)
1308                         GOTO(out, mode);
1309
1310                 if (bits)
1311                         *bits = lock->l_policy_data.l_inodebits.bits;
1312                 mode = lock->l_granted_mode;
1313                 ldlm_lock_addref_internal_nolock(lock, mode);
1314         }
1315
1316         EXIT;
1317
1318 out:
1319         if (lock != NULL) {
1320                 unlock_res_and_lock(lock);
1321                 LDLM_LOCK_PUT(lock);
1322         }
1323         return mode;
1324 }
1325 EXPORT_SYMBOL(ldlm_revalidate_lock_handle);
1326
1327 /* Returns a referenced lock */
1328 struct ldlm_lock *ldlm_lock_create(struct ldlm_namespace *ns,
1329                                    const struct ldlm_res_id *res_id,
1330                                    ldlm_type_t type,
1331                                    ldlm_mode_t mode,
1332                                    const struct ldlm_callback_suite *cbs,
1333                                    void *data, __u32 lvb_len)
1334 {
1335         struct ldlm_lock *lock;
1336         struct ldlm_resource *res;
1337         ENTRY;
1338
1339         res = ldlm_resource_get(ns, NULL, res_id, type, 1);
1340         if (res == NULL)
1341                 RETURN(NULL);
1342
1343         lock = ldlm_lock_new(res);
1344
1345         if (lock == NULL)
1346                 RETURN(NULL);
1347
1348         lock->l_req_mode = mode;
1349         lock->l_ast_data = data;
1350         lock->l_pid = cfs_curproc_pid();
1351         lock->l_ns_srv = !!ns_is_server(ns);
1352         if (cbs) {
1353                 lock->l_blocking_ast = cbs->lcs_blocking;
1354                 lock->l_completion_ast = cbs->lcs_completion;
1355                 lock->l_glimpse_ast = cbs->lcs_glimpse;
1356                 lock->l_weigh_ast = cbs->lcs_weigh;
1357         }
1358
1359         lock->l_tree_node = NULL;
1360         /* if this is the extent lock, allocate the interval tree node */
1361         if (type == LDLM_EXTENT) {
1362                 if (ldlm_interval_alloc(lock) == NULL)
1363                         GOTO(out, 0);
1364         }
1365
1366         if (lvb_len) {
1367                 lock->l_lvb_len = lvb_len;
1368                 OBD_ALLOC(lock->l_lvb_data, lvb_len);
1369                 if (lock->l_lvb_data == NULL)
1370                         GOTO(out, 0);
1371         }
1372
1373         if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_NEW_LOCK))
1374                 GOTO(out, 0);
1375
1376         RETURN(lock);
1377
1378 out:
1379         ldlm_lock_destroy(lock);
1380         LDLM_LOCK_RELEASE(lock);
1381         return NULL;
1382 }
1383
1384 ldlm_error_t ldlm_lock_enqueue(struct ldlm_namespace *ns,
1385                                struct ldlm_lock **lockp,
1386                                void *cookie, int *flags)
1387 {
1388         struct ldlm_lock *lock = *lockp;
1389         struct ldlm_resource *res = lock->l_resource;
1390         int local = ns_is_client(ldlm_res_to_ns(res));
1391 #ifdef HAVE_SERVER_SUPPORT
1392         ldlm_processing_policy policy;
1393 #endif
1394         ldlm_error_t rc = ELDLM_OK;
1395         struct ldlm_interval *node = NULL;
1396         ENTRY;
1397
1398         lock->l_last_activity = cfs_time_current_sec();
1399         /* policies are not executed on the client or during replay */
1400         if ((*flags & (LDLM_FL_HAS_INTENT|LDLM_FL_REPLAY)) == LDLM_FL_HAS_INTENT
1401             && !local && ns->ns_policy) {
1402                 rc = ns->ns_policy(ns, lockp, cookie, lock->l_req_mode, *flags,
1403                                    NULL);
1404                 if (rc == ELDLM_LOCK_REPLACED) {
1405                         /* The lock that was returned has already been granted,
1406                          * and placed into lockp.  If it's not the same as the
1407                          * one we passed in, then destroy the old one and our
1408                          * work here is done. */
1409                         if (lock != *lockp) {
1410                                 ldlm_lock_destroy(lock);
1411                                 LDLM_LOCK_RELEASE(lock);
1412                         }
1413                         *flags |= LDLM_FL_LOCK_CHANGED;
1414                         RETURN(0);
1415                 } else if (rc != ELDLM_OK ||
1416                            (rc == ELDLM_OK && (*flags & LDLM_FL_INTENT_ONLY))) {
1417                         ldlm_lock_destroy(lock);
1418                         RETURN(rc);
1419                 }
1420         }
1421
1422         /* For a replaying lock, it might be already in granted list. So
1423          * unlinking the lock will cause the interval node to be freed, we
1424          * have to allocate the interval node early otherwise we can't regrant
1425          * this lock in the future. - jay */
1426         if (!local && (*flags & LDLM_FL_REPLAY) && res->lr_type == LDLM_EXTENT)
1427                 OBD_SLAB_ALLOC_PTR_GFP(node, ldlm_interval_slab, CFS_ALLOC_IO);
1428
1429         lock_res_and_lock(lock);
1430         if (local && lock->l_req_mode == lock->l_granted_mode) {
1431                 /* The server returned a blocked lock, but it was granted
1432                  * before we got a chance to actually enqueue it.  We don't
1433                  * need to do anything else. */
1434                 *flags &= ~(LDLM_FL_BLOCK_GRANTED |
1435                             LDLM_FL_BLOCK_CONV | LDLM_FL_BLOCK_WAIT);
1436                 GOTO(out, ELDLM_OK);
1437         }
1438
1439         ldlm_resource_unlink_lock(lock);
1440         if (res->lr_type == LDLM_EXTENT && lock->l_tree_node == NULL) {
1441                 if (node == NULL) {
1442                         ldlm_lock_destroy_nolock(lock);
1443                         GOTO(out, rc = -ENOMEM);
1444                 }
1445
1446                 CFS_INIT_LIST_HEAD(&node->li_group);
1447                 ldlm_interval_attach(node, lock);
1448                 node = NULL;
1449         }
1450
1451         /* Some flags from the enqueue want to make it into the AST, via the
1452          * lock's l_flags. */
1453         lock->l_flags |= *flags & LDLM_AST_DISCARD_DATA;
1454
1455         /* This distinction between local lock trees is very important; a client
1456          * namespace only has information about locks taken by that client, and
1457          * thus doesn't have enough information to decide for itself if it can
1458          * be granted (below).  In this case, we do exactly what the server
1459          * tells us to do, as dictated by the 'flags'.
1460          *
1461          * We do exactly the same thing during recovery, when the server is
1462          * more or less trusting the clients not to lie.
1463          *
1464          * FIXME (bug 268): Detect obvious lies by checking compatibility in
1465          * granted/converting queues. */
1466         if (local) {
1467                 if (*flags & LDLM_FL_BLOCK_CONV)
1468                         ldlm_resource_add_lock(res, &res->lr_converting, lock);
1469                 else if (*flags & (LDLM_FL_BLOCK_WAIT | LDLM_FL_BLOCK_GRANTED))
1470                         ldlm_resource_add_lock(res, &res->lr_waiting, lock);
1471                 else
1472                         ldlm_grant_lock(lock, NULL);
1473                 GOTO(out, ELDLM_OK);
1474 #ifdef HAVE_SERVER_SUPPORT
1475         } else if (*flags & LDLM_FL_REPLAY) {
1476                 if (*flags & LDLM_FL_BLOCK_CONV) {
1477                         ldlm_resource_add_lock(res, &res->lr_converting, lock);
1478                         GOTO(out, ELDLM_OK);
1479                 } else if (*flags & LDLM_FL_BLOCK_WAIT) {
1480                         ldlm_resource_add_lock(res, &res->lr_waiting, lock);
1481                         GOTO(out, ELDLM_OK);
1482                 } else if (*flags & LDLM_FL_BLOCK_GRANTED) {
1483                         ldlm_grant_lock(lock, NULL);
1484                         GOTO(out, ELDLM_OK);
1485                 }
1486                 /* If no flags, fall through to normal enqueue path. */
1487         }
1488
1489         policy = ldlm_processing_policy_table[res->lr_type];
1490         policy(lock, flags, 1, &rc, NULL);
1491         GOTO(out, rc);
1492 #else
1493         } else {
1494                 CERROR("This is client-side-only module, cannot handle "
1495                        "LDLM_NAMESPACE_SERVER resource type lock.\n");
1496                 LBUG();
1497         }
1498 #endif
1499
1500 out:
1501         unlock_res_and_lock(lock);
1502         if (node)
1503                 OBD_SLAB_FREE(node, ldlm_interval_slab, sizeof(*node));
1504         return rc;
1505 }
1506
1507 #ifdef HAVE_SERVER_SUPPORT
1508 /* Must be called with namespace taken: queue is waiting or converting. */
1509 int ldlm_reprocess_queue(struct ldlm_resource *res, cfs_list_t *queue,
1510                          cfs_list_t *work_list)
1511 {
1512         cfs_list_t *tmp, *pos;
1513         ldlm_processing_policy policy;
1514         int flags;
1515         int rc = LDLM_ITER_CONTINUE;
1516         ldlm_error_t err;
1517         ENTRY;
1518
1519         check_res_locked(res);
1520
1521         policy = ldlm_processing_policy_table[res->lr_type];
1522         LASSERT(policy);
1523
1524         cfs_list_for_each_safe(tmp, pos, queue) {
1525                 struct ldlm_lock *pending;
1526                 pending = cfs_list_entry(tmp, struct ldlm_lock, l_res_link);
1527
1528                 CDEBUG(D_INFO, "Reprocessing lock %p\n", pending);
1529
1530                 flags = 0;
1531                 rc = policy(pending, &flags, 0, &err, work_list);
1532                 if (rc != LDLM_ITER_CONTINUE)
1533                         break;
1534         }
1535
1536         RETURN(rc);
1537 }
1538 #endif
1539
1540 static int
1541 ldlm_work_bl_ast_lock(struct ptlrpc_request_set *rqset, void *opaq)
1542 {
1543         struct ldlm_cb_set_arg *arg = opaq;
1544         struct ldlm_lock_desc   d;
1545         int                     rc;
1546         struct ldlm_lock       *lock;
1547         ENTRY;
1548
1549         if (cfs_list_empty(arg->list))
1550                 RETURN(-ENOENT);
1551
1552         lock = cfs_list_entry(arg->list->next, struct ldlm_lock, l_bl_ast);
1553
1554         /* nobody should touch l_bl_ast */
1555         lock_res_and_lock(lock);
1556         cfs_list_del_init(&lock->l_bl_ast);
1557
1558         LASSERT(lock->l_flags & LDLM_FL_AST_SENT);
1559         LASSERT(lock->l_bl_ast_run == 0);
1560         LASSERT(lock->l_blocking_lock);
1561         lock->l_bl_ast_run++;
1562         unlock_res_and_lock(lock);
1563
1564         ldlm_lock2desc(lock->l_blocking_lock, &d);
1565
1566         rc = lock->l_blocking_ast(lock, &d, (void *)arg, LDLM_CB_BLOCKING);
1567         LDLM_LOCK_RELEASE(lock->l_blocking_lock);
1568         lock->l_blocking_lock = NULL;
1569         LDLM_LOCK_RELEASE(lock);
1570
1571         RETURN(rc);
1572 }
1573
1574 static int
1575 ldlm_work_cp_ast_lock(struct ptlrpc_request_set *rqset, void *opaq)
1576 {
1577         struct ldlm_cb_set_arg  *arg = opaq;
1578         int                      rc = 0;
1579         struct ldlm_lock        *lock;
1580         ldlm_completion_callback completion_callback;
1581         ENTRY;
1582
1583         if (cfs_list_empty(arg->list))
1584                 RETURN(-ENOENT);
1585
1586         lock = cfs_list_entry(arg->list->next, struct ldlm_lock, l_cp_ast);
1587
1588         /* It's possible to receive a completion AST before we've set
1589          * the l_completion_ast pointer: either because the AST arrived
1590          * before the reply, or simply because there's a small race
1591          * window between receiving the reply and finishing the local
1592          * enqueue. (bug 842)
1593          *
1594          * This can't happen with the blocking_ast, however, because we
1595          * will never call the local blocking_ast until we drop our
1596          * reader/writer reference, which we won't do until we get the
1597          * reply and finish enqueueing. */
1598
1599         /* nobody should touch l_cp_ast */
1600         lock_res_and_lock(lock);
1601         cfs_list_del_init(&lock->l_cp_ast);
1602         LASSERT(lock->l_flags & LDLM_FL_CP_REQD);
1603         /* save l_completion_ast since it can be changed by
1604          * mds_intent_policy(), see bug 14225 */
1605         completion_callback = lock->l_completion_ast;
1606         lock->l_flags &= ~LDLM_FL_CP_REQD;
1607         unlock_res_and_lock(lock);
1608
1609         if (completion_callback != NULL)
1610                 rc = completion_callback(lock, 0, (void *)arg);
1611         LDLM_LOCK_RELEASE(lock);
1612
1613         RETURN(rc);
1614 }
1615
1616 static int
1617 ldlm_work_revoke_ast_lock(struct ptlrpc_request_set *rqset, void *opaq)
1618 {
1619         struct ldlm_cb_set_arg *arg = opaq;
1620         struct ldlm_lock_desc   desc;
1621         int                     rc;
1622         struct ldlm_lock       *lock;
1623         ENTRY;
1624
1625         if (cfs_list_empty(arg->list))
1626                 RETURN(-ENOENT);
1627
1628         lock = cfs_list_entry(arg->list->next, struct ldlm_lock, l_rk_ast);
1629         cfs_list_del_init(&lock->l_rk_ast);
1630
1631         /* the desc just pretend to exclusive */
1632         ldlm_lock2desc(lock, &desc);
1633         desc.l_req_mode = LCK_EX;
1634         desc.l_granted_mode = 0;
1635
1636         rc = lock->l_blocking_ast(lock, &desc, (void*)arg, LDLM_CB_BLOCKING);
1637         LDLM_LOCK_RELEASE(lock);
1638
1639         RETURN(rc);
1640 }
1641
1642 int ldlm_work_gl_ast_lock(struct ptlrpc_request_set *rqset, void *opaq)
1643 {
1644         struct ldlm_cb_set_arg          *arg = opaq;
1645         struct ldlm_glimpse_work        *gl_work;
1646         struct ldlm_lock                *lock;
1647         int                              rc = 0;
1648         ENTRY;
1649
1650         if (cfs_list_empty(arg->list))
1651                 RETURN(-ENOENT);
1652
1653         gl_work = cfs_list_entry(arg->list->next, struct ldlm_glimpse_work,
1654                                  gl_list);
1655         cfs_list_del_init(&gl_work->gl_list);
1656
1657         lock = gl_work->gl_lock;
1658
1659         /* transfer the glimpse descriptor to ldlm_cb_set_arg */
1660         arg->gl_desc = gl_work->gl_desc;
1661
1662         /* invoke the actual glimpse callback */
1663         if (lock->l_glimpse_ast(lock, (void*)arg) == 0)
1664                 rc = 1;
1665
1666         LDLM_LOCK_RELEASE(lock);
1667
1668         if ((gl_work->gl_flags & LDLM_GL_WORK_NOFREE) == 0)
1669                 OBD_FREE_PTR(gl_work);
1670
1671         RETURN(rc);
1672 }
1673
1674 int ldlm_run_ast_work(struct ldlm_namespace *ns, cfs_list_t *rpc_list,
1675                       ldlm_desc_ast_t ast_type)
1676 {
1677         struct ldlm_cb_set_arg *arg;
1678         set_producer_func       work_ast_lock;
1679         int                     rc;
1680
1681         if (cfs_list_empty(rpc_list))
1682                 RETURN(0);
1683
1684         OBD_ALLOC_PTR(arg);
1685         if (arg == NULL)
1686                 RETURN(-ENOMEM);
1687
1688         cfs_atomic_set(&arg->restart, 0);
1689         arg->list = rpc_list;
1690
1691         switch (ast_type) {
1692                 case LDLM_WORK_BL_AST:
1693                         arg->type = LDLM_BL_CALLBACK;
1694                         work_ast_lock = ldlm_work_bl_ast_lock;
1695                         break;
1696                 case LDLM_WORK_CP_AST:
1697                         arg->type = LDLM_CP_CALLBACK;
1698                         work_ast_lock = ldlm_work_cp_ast_lock;
1699                         break;
1700                 case LDLM_WORK_REVOKE_AST:
1701                         arg->type = LDLM_BL_CALLBACK;
1702                         work_ast_lock = ldlm_work_revoke_ast_lock;
1703                         break;
1704                 case LDLM_WORK_GL_AST:
1705                         arg->type = LDLM_GL_CALLBACK;
1706                         work_ast_lock = ldlm_work_gl_ast_lock;
1707                         break;
1708                 default:
1709                         LBUG();
1710         }
1711
1712         /* We create a ptlrpc request set with flow control extension.
1713          * This request set will use the work_ast_lock function to produce new
1714          * requests and will send a new request each time one completes in order
1715          * to keep the number of requests in flight to ns_max_parallel_ast */
1716         arg->set = ptlrpc_prep_fcset(ns->ns_max_parallel_ast ? : UINT_MAX,
1717                                      work_ast_lock, arg);
1718         if (arg->set == NULL)
1719                 GOTO(out, rc = -ENOMEM);
1720
1721         ptlrpc_set_wait(arg->set);
1722         ptlrpc_set_destroy(arg->set);
1723
1724         rc = cfs_atomic_read(&arg->restart) ? -ERESTART : 0;
1725         GOTO(out, rc);
1726 out:
1727         OBD_FREE_PTR(arg);
1728         return rc;
1729 }
1730
1731 static int reprocess_one_queue(struct ldlm_resource *res, void *closure)
1732 {
1733         ldlm_reprocess_all(res);
1734         return LDLM_ITER_CONTINUE;
1735 }
1736
1737 static int ldlm_reprocess_res(cfs_hash_t *hs, cfs_hash_bd_t *bd,
1738                               cfs_hlist_node_t *hnode, void *arg)
1739 {
1740         struct ldlm_resource *res = cfs_hash_object(hs, hnode);
1741         int    rc;
1742
1743         rc = reprocess_one_queue(res, arg);
1744
1745         return rc == LDLM_ITER_STOP;
1746 }
1747
1748 void ldlm_reprocess_all_ns(struct ldlm_namespace *ns)
1749 {
1750         ENTRY;
1751
1752         if (ns != NULL) {
1753                 cfs_hash_for_each_nolock(ns->ns_rs_hash,
1754                                          ldlm_reprocess_res, NULL);
1755         }
1756         EXIT;
1757 }
1758 EXPORT_SYMBOL(ldlm_reprocess_all_ns);
1759
1760 void ldlm_reprocess_all(struct ldlm_resource *res)
1761 {
1762         CFS_LIST_HEAD(rpc_list);
1763
1764 #ifdef HAVE_SERVER_SUPPORT
1765         int rc;
1766         ENTRY;
1767         /* Local lock trees don't get reprocessed. */
1768         if (ns_is_client(ldlm_res_to_ns(res))) {
1769                 EXIT;
1770                 return;
1771         }
1772
1773 restart:
1774         lock_res(res);
1775         rc = ldlm_reprocess_queue(res, &res->lr_converting, &rpc_list);
1776         if (rc == LDLM_ITER_CONTINUE)
1777                 ldlm_reprocess_queue(res, &res->lr_waiting, &rpc_list);
1778         unlock_res(res);
1779
1780         rc = ldlm_run_ast_work(ldlm_res_to_ns(res), &rpc_list,
1781                                LDLM_WORK_CP_AST);
1782         if (rc == -ERESTART) {
1783                 LASSERT(cfs_list_empty(&rpc_list));
1784                 goto restart;
1785         }
1786 #else
1787         ENTRY;
1788         if (!ns_is_client(ldlm_res_to_ns(res))) {
1789                 CERROR("This is client-side-only module, cannot handle "
1790                        "LDLM_NAMESPACE_SERVER resource type lock.\n");
1791                 LBUG();
1792         }
1793 #endif
1794         EXIT;
1795 }
1796
1797 void ldlm_cancel_callback(struct ldlm_lock *lock)
1798 {
1799         check_res_locked(lock->l_resource);
1800         if (!(lock->l_flags & LDLM_FL_CANCEL)) {
1801                 lock->l_flags |= LDLM_FL_CANCEL;
1802                 if (lock->l_blocking_ast) {
1803                         // l_check_no_ns_lock(ns);
1804                         unlock_res_and_lock(lock);
1805                         lock->l_blocking_ast(lock, NULL, lock->l_ast_data,
1806                                              LDLM_CB_CANCELING);
1807                         lock_res_and_lock(lock);
1808                 } else {
1809                         LDLM_DEBUG(lock, "no blocking ast");
1810                 }
1811         }
1812         lock->l_flags |= LDLM_FL_BL_DONE;
1813 }
1814
1815 void ldlm_unlink_lock_skiplist(struct ldlm_lock *req)
1816 {
1817         if (req->l_resource->lr_type != LDLM_PLAIN &&
1818             req->l_resource->lr_type != LDLM_IBITS)
1819                 return;
1820
1821         cfs_list_del_init(&req->l_sl_policy);
1822         cfs_list_del_init(&req->l_sl_mode);
1823 }
1824
1825 void ldlm_lock_cancel(struct ldlm_lock *lock)
1826 {
1827         struct ldlm_resource *res;
1828         struct ldlm_namespace *ns;
1829         ENTRY;
1830
1831         lock_res_and_lock(lock);
1832
1833         res = lock->l_resource;
1834         ns  = ldlm_res_to_ns(res);
1835
1836         /* Please do not, no matter how tempting, remove this LBUG without
1837          * talking to me first. -phik */
1838         if (lock->l_readers || lock->l_writers) {
1839                 LDLM_ERROR(lock, "lock still has references");
1840                 LBUG();
1841         }
1842
1843         if (lock->l_waited)
1844                 ldlm_del_waiting_lock(lock);
1845
1846         /* Releases cancel callback. */
1847         ldlm_cancel_callback(lock);
1848
1849         /* Yes, second time, just in case it was added again while we were
1850            running with no res lock in ldlm_cancel_callback */
1851         if (lock->l_waited)
1852                 ldlm_del_waiting_lock(lock);
1853
1854         ldlm_resource_unlink_lock(lock);
1855         ldlm_lock_destroy_nolock(lock);
1856
1857         if (lock->l_granted_mode == lock->l_req_mode)
1858                 ldlm_pool_del(&ns->ns_pool, lock);
1859
1860         /* Make sure we will not be called again for same lock what is possible
1861          * if not to zero out lock->l_granted_mode */
1862         lock->l_granted_mode = LCK_MINMODE;
1863         unlock_res_and_lock(lock);
1864
1865         EXIT;
1866 }
1867 EXPORT_SYMBOL(ldlm_lock_cancel);
1868
1869 int ldlm_lock_set_data(struct lustre_handle *lockh, void *data)
1870 {
1871         struct ldlm_lock *lock = ldlm_handle2lock(lockh);
1872         int rc = -EINVAL;
1873         ENTRY;
1874
1875         if (lock) {
1876                 if (lock->l_ast_data == NULL)
1877                         lock->l_ast_data = data;
1878                 if (lock->l_ast_data == data)
1879                         rc = 0;
1880                 LDLM_LOCK_PUT(lock);
1881         }
1882         RETURN(rc);
1883 }
1884 EXPORT_SYMBOL(ldlm_lock_set_data);
1885
1886 struct export_cl_data {
1887         struct obd_export       *ecl_exp;
1888         int                     ecl_loop;
1889 };
1890
1891 int ldlm_cancel_locks_for_export_cb(cfs_hash_t *hs, cfs_hash_bd_t *bd,
1892                                     cfs_hlist_node_t *hnode, void *data)
1893
1894 {
1895         struct export_cl_data   *ecl = (struct export_cl_data *)data;
1896         struct obd_export       *exp  = ecl->ecl_exp;
1897         struct ldlm_lock     *lock = cfs_hash_object(hs, hnode);
1898         struct ldlm_resource *res;
1899
1900         res = ldlm_resource_getref(lock->l_resource);
1901         LDLM_LOCK_GET(lock);
1902
1903         LDLM_DEBUG(lock, "export %p", exp);
1904         ldlm_res_lvbo_update(res, NULL, 1);
1905         ldlm_lock_cancel(lock);
1906         ldlm_reprocess_all(res);
1907         ldlm_resource_putref(res);
1908         LDLM_LOCK_RELEASE(lock);
1909
1910         ecl->ecl_loop++;
1911         if ((ecl->ecl_loop & -ecl->ecl_loop) == ecl->ecl_loop) {
1912                 CDEBUG(D_INFO,
1913                        "Cancel lock %p for export %p (loop %d), still have "
1914                        "%d locks left on hash table.\n",
1915                        lock, exp, ecl->ecl_loop,
1916                        cfs_atomic_read(&hs->hs_count));
1917         }
1918
1919         return 0;
1920 }
1921
1922 void ldlm_cancel_locks_for_export(struct obd_export *exp)
1923 {
1924         struct export_cl_data   ecl = {
1925                 .ecl_exp        = exp,
1926                 .ecl_loop       = 0,
1927         };
1928
1929         cfs_hash_for_each_empty(exp->exp_lock_hash,
1930                                 ldlm_cancel_locks_for_export_cb, &ecl);
1931 }
1932
1933 /**
1934  * Downgrade an exclusive lock.
1935  *
1936  * A fast variant of ldlm_lock_convert for convertion of exclusive
1937  * locks. The convertion is always successful.
1938  *
1939  * \param lock A lock to convert
1940  * \param new_mode new lock mode
1941  */
1942 void ldlm_lock_downgrade(struct ldlm_lock *lock, int new_mode)
1943 {
1944         ENTRY;
1945
1946         LASSERT(lock->l_granted_mode & (LCK_PW | LCK_EX));
1947         LASSERT(new_mode == LCK_COS);
1948
1949         lock_res_and_lock(lock);
1950         ldlm_resource_unlink_lock(lock);
1951         /*
1952          * Remove the lock from pool as it will be added again in
1953          * ldlm_grant_lock() called below.
1954          */
1955         ldlm_pool_del(&ldlm_lock_to_ns(lock)->ns_pool, lock);
1956
1957         lock->l_req_mode = new_mode;
1958         ldlm_grant_lock(lock, NULL);
1959         unlock_res_and_lock(lock);
1960         ldlm_reprocess_all(lock->l_resource);
1961
1962         EXIT;
1963 }
1964 EXPORT_SYMBOL(ldlm_lock_downgrade);
1965
1966 struct ldlm_resource *ldlm_lock_convert(struct ldlm_lock *lock, int new_mode,
1967                                         __u32 *flags)
1968 {
1969         CFS_LIST_HEAD(rpc_list);
1970         struct ldlm_resource *res;
1971         struct ldlm_namespace *ns;
1972         int granted = 0;
1973 #ifdef HAVE_SERVER_SUPPORT
1974         int old_mode;
1975         struct sl_insert_point prev;
1976 #endif
1977         struct ldlm_interval *node;
1978         ENTRY;
1979
1980         if (new_mode == lock->l_granted_mode) { // No changes? Just return.
1981                 *flags |= LDLM_FL_BLOCK_GRANTED;
1982                 RETURN(lock->l_resource);
1983         }
1984
1985         /* I can't check the type of lock here because the bitlock of lock
1986          * is not held here, so do the allocation blindly. -jay */
1987         OBD_SLAB_ALLOC_PTR_GFP(node, ldlm_interval_slab, CFS_ALLOC_IO);
1988         if (node == NULL)  /* Actually, this causes EDEADLOCK to be returned */
1989                 RETURN(NULL);
1990
1991         LASSERTF((new_mode == LCK_PW && lock->l_granted_mode == LCK_PR),
1992                  "new_mode %u, granted %u\n", new_mode, lock->l_granted_mode);
1993
1994         lock_res_and_lock(lock);
1995
1996         res = lock->l_resource;
1997         ns  = ldlm_res_to_ns(res);
1998
1999 #ifdef HAVE_SERVER_SUPPORT
2000         old_mode = lock->l_req_mode;
2001 #endif
2002         lock->l_req_mode = new_mode;
2003         if (res->lr_type == LDLM_PLAIN || res->lr_type == LDLM_IBITS) {
2004 #ifdef HAVE_SERVER_SUPPORT
2005                 /* remember the lock position where the lock might be
2006                  * added back to the granted list later and also
2007                  * remember the join mode for skiplist fixing. */
2008                 prev.res_link = lock->l_res_link.prev;
2009                 prev.mode_link = lock->l_sl_mode.prev;
2010                 prev.policy_link = lock->l_sl_policy.prev;
2011 #endif
2012                 ldlm_resource_unlink_lock(lock);
2013         } else {
2014                 ldlm_resource_unlink_lock(lock);
2015                 if (res->lr_type == LDLM_EXTENT) {
2016                         /* FIXME: ugly code, I have to attach the lock to a
2017                          * interval node again since perhaps it will be granted
2018                          * soon */
2019                         CFS_INIT_LIST_HEAD(&node->li_group);
2020                         ldlm_interval_attach(node, lock);
2021                         node = NULL;
2022                 }
2023         }
2024
2025         /*
2026          * Remove old lock from the pool before adding the lock with new
2027          * mode below in ->policy()
2028          */
2029         ldlm_pool_del(&ns->ns_pool, lock);
2030
2031         /* If this is a local resource, put it on the appropriate list. */
2032         if (ns_is_client(ldlm_res_to_ns(res))) {
2033                 if (*flags & (LDLM_FL_BLOCK_CONV | LDLM_FL_BLOCK_GRANTED)) {
2034                         ldlm_resource_add_lock(res, &res->lr_converting, lock);
2035                 } else {
2036                         /* This should never happen, because of the way the
2037                          * server handles conversions. */
2038                         LDLM_ERROR(lock, "Erroneous flags %d on local lock\n",
2039                                    *flags);
2040                         LBUG();
2041
2042                         ldlm_grant_lock(lock, &rpc_list);
2043                         granted = 1;
2044                         /* FIXME: completion handling not with lr_lock held ! */
2045                         if (lock->l_completion_ast)
2046                                 lock->l_completion_ast(lock, 0, NULL);
2047                 }
2048 #ifdef HAVE_SERVER_SUPPORT
2049         } else {
2050                 int rc;
2051                 ldlm_error_t err;
2052                 int pflags = 0;
2053                 ldlm_processing_policy policy;
2054                 policy = ldlm_processing_policy_table[res->lr_type];
2055                 rc = policy(lock, &pflags, 0, &err, &rpc_list);
2056                 if (rc == LDLM_ITER_STOP) {
2057                         lock->l_req_mode = old_mode;
2058                         if (res->lr_type == LDLM_EXTENT)
2059                                 ldlm_extent_add_lock(res, lock);
2060                         else
2061                                 ldlm_granted_list_add_lock(lock, &prev);
2062
2063                         res = NULL;
2064                 } else {
2065                         *flags |= LDLM_FL_BLOCK_GRANTED;
2066                         granted = 1;
2067                 }
2068         }
2069 #else
2070         } else {
2071                 CERROR("This is client-side-only module, cannot handle "
2072                        "LDLM_NAMESPACE_SERVER resource type lock.\n");
2073                 LBUG();
2074         }
2075 #endif
2076         unlock_res_and_lock(lock);
2077
2078         if (granted)
2079                 ldlm_run_ast_work(ns, &rpc_list, LDLM_WORK_CP_AST);
2080         if (node)
2081                 OBD_SLAB_FREE(node, ldlm_interval_slab, sizeof(*node));
2082         RETURN(res);
2083 }
2084 EXPORT_SYMBOL(ldlm_lock_convert);
2085
2086 void ldlm_lock_dump_handle(int level, struct lustre_handle *lockh)
2087 {
2088         struct ldlm_lock *lock;
2089
2090         if (!((libcfs_debug | D_ERROR) & level))
2091                 return;
2092
2093         lock = ldlm_handle2lock(lockh);
2094         if (lock == NULL)
2095                 return;
2096
2097         LDLM_DEBUG_LIMIT(level, lock, "###");
2098
2099         LDLM_LOCK_PUT(lock);
2100 }
2101 EXPORT_SYMBOL(ldlm_lock_dump_handle);
2102
2103 void _ldlm_lock_debug(struct ldlm_lock *lock,
2104                       struct libcfs_debug_msg_data *msgdata,
2105                       const char *fmt, ...)
2106 {
2107         va_list args;
2108         struct obd_export *exp = lock->l_export;
2109         struct ldlm_resource *resource = lock->l_resource;
2110         char *nid = "local";
2111
2112         va_start(args, fmt);
2113
2114         if (exp && exp->exp_connection) {
2115                 nid = libcfs_nid2str(exp->exp_connection->c_peer.nid);
2116         } else if (exp && exp->exp_obd != NULL) {
2117                 struct obd_import *imp = exp->exp_obd->u.cli.cl_import;
2118                 nid = libcfs_nid2str(imp->imp_connection->c_peer.nid);
2119         }
2120
2121         if (resource == NULL) {
2122                 libcfs_debug_vmsg2(msgdata, fmt, args,
2123                        " ns: \?\? lock: %p/"LPX64" lrc: %d/%d,%d mode: %s/%s "
2124                        "res: \?\? rrc=\?\? type: \?\?\? flags: "LPX64" nid: %s "
2125                        "remote: "LPX64" expref: %d pid: %u timeout: %lu\n",
2126                        lock,
2127                        lock->l_handle.h_cookie, cfs_atomic_read(&lock->l_refc),
2128                        lock->l_readers, lock->l_writers,
2129                        ldlm_lockname[lock->l_granted_mode],
2130                        ldlm_lockname[lock->l_req_mode],
2131                        lock->l_flags, nid, lock->l_remote_handle.cookie,
2132                        exp ? cfs_atomic_read(&exp->exp_refcount) : -99,
2133                        lock->l_pid, lock->l_callback_timeout);
2134                 va_end(args);
2135                 return;
2136         }
2137
2138         switch (resource->lr_type) {
2139         case LDLM_EXTENT:
2140                 libcfs_debug_vmsg2(msgdata, fmt, args,
2141                        " ns: %s lock: %p/"LPX64" lrc: %d/%d,%d mode: %s/%s "
2142                        "res: "LPU64"/"LPU64" rrc: %d type: %s ["LPU64"->"LPU64
2143                        "] (req "LPU64"->"LPU64") flags: "LPX64" nid: %s remote:"
2144                        " "LPX64" expref: %d pid: %u timeout %lu\n",
2145                        ldlm_lock_to_ns_name(lock), lock,
2146                        lock->l_handle.h_cookie, cfs_atomic_read(&lock->l_refc),
2147                        lock->l_readers, lock->l_writers,
2148                        ldlm_lockname[lock->l_granted_mode],
2149                        ldlm_lockname[lock->l_req_mode],
2150                        resource->lr_name.name[0],
2151                        resource->lr_name.name[1],
2152                        cfs_atomic_read(&resource->lr_refcount),
2153                        ldlm_typename[resource->lr_type],
2154                        lock->l_policy_data.l_extent.start,
2155                        lock->l_policy_data.l_extent.end,
2156                        lock->l_req_extent.start, lock->l_req_extent.end,
2157                        lock->l_flags, nid, lock->l_remote_handle.cookie,
2158                        exp ? cfs_atomic_read(&exp->exp_refcount) : -99,
2159                        lock->l_pid, lock->l_callback_timeout);
2160                 break;
2161
2162         case LDLM_FLOCK:
2163                 libcfs_debug_vmsg2(msgdata, fmt, args,
2164                        " ns: %s lock: %p/"LPX64" lrc: %d/%d,%d mode: %s/%s "
2165                        "res: "LPU64"/"LPU64" rrc: %d type: %s pid: %d "
2166                        "["LPU64"->"LPU64"] flags: "LPX64" nid: %s remote: "LPX64
2167                        " expref: %d pid: %u timeout: %lu\n",
2168                        ldlm_lock_to_ns_name(lock), lock,
2169                        lock->l_handle.h_cookie, cfs_atomic_read(&lock->l_refc),
2170                        lock->l_readers, lock->l_writers,
2171                        ldlm_lockname[lock->l_granted_mode],
2172                        ldlm_lockname[lock->l_req_mode],
2173                        resource->lr_name.name[0],
2174                        resource->lr_name.name[1],
2175                        cfs_atomic_read(&resource->lr_refcount),
2176                        ldlm_typename[resource->lr_type],
2177                        lock->l_policy_data.l_flock.pid,
2178                        lock->l_policy_data.l_flock.start,
2179                        lock->l_policy_data.l_flock.end,
2180                        lock->l_flags, nid, lock->l_remote_handle.cookie,
2181                        exp ? cfs_atomic_read(&exp->exp_refcount) : -99,
2182                        lock->l_pid, lock->l_callback_timeout);
2183                 break;
2184
2185         case LDLM_IBITS:
2186                 libcfs_debug_vmsg2(msgdata, fmt, args,
2187                        " ns: %s lock: %p/"LPX64" lrc: %d/%d,%d mode: %s/%s "
2188                        "res: "LPU64"/"LPU64" bits "LPX64" rrc: %d type: %s "
2189                        "flags: "LPX64" nid: %s remote: "LPX64" expref: %d "
2190                        "pid: %u timeout: %lu\n",
2191                        ldlm_lock_to_ns_name(lock),
2192                        lock, lock->l_handle.h_cookie,
2193                        cfs_atomic_read (&lock->l_refc),
2194                        lock->l_readers, lock->l_writers,
2195                        ldlm_lockname[lock->l_granted_mode],
2196                        ldlm_lockname[lock->l_req_mode],
2197                        resource->lr_name.name[0],
2198                        resource->lr_name.name[1],
2199                        lock->l_policy_data.l_inodebits.bits,
2200                        cfs_atomic_read(&resource->lr_refcount),
2201                        ldlm_typename[resource->lr_type],
2202                        lock->l_flags, nid, lock->l_remote_handle.cookie,
2203                        exp ? cfs_atomic_read(&exp->exp_refcount) : -99,
2204                        lock->l_pid, lock->l_callback_timeout);
2205                 break;
2206
2207         default:
2208                 libcfs_debug_vmsg2(msgdata, fmt, args,
2209                        " ns: %s lock: %p/"LPX64" lrc: %d/%d,%d mode: %s/%s "
2210                        "res: "LPU64"/"LPU64" rrc: %d type: %s flags: "LPX64" "
2211                        "nid: %s remote: "LPX64" expref: %d pid: %u timeout %lu"
2212                        "\n",
2213                        ldlm_lock_to_ns_name(lock),
2214                        lock, lock->l_handle.h_cookie,
2215                        cfs_atomic_read (&lock->l_refc),
2216                        lock->l_readers, lock->l_writers,
2217                        ldlm_lockname[lock->l_granted_mode],
2218                        ldlm_lockname[lock->l_req_mode],
2219                        resource->lr_name.name[0],
2220                        resource->lr_name.name[1],
2221                        cfs_atomic_read(&resource->lr_refcount),
2222                        ldlm_typename[resource->lr_type],
2223                        lock->l_flags, nid, lock->l_remote_handle.cookie,
2224                        exp ? cfs_atomic_read(&exp->exp_refcount) : -99,
2225                        lock->l_pid, lock->l_callback_timeout);
2226                 break;
2227         }
2228         va_end(args);
2229 }
2230 EXPORT_SYMBOL(_ldlm_lock_debug);