4 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
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.
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).
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
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
27 * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
28 * Use is subject to license terms.
30 * Copyright (c) 2010, 2012, Intel Corporation.
33 * This file is part of Lustre, http://www.lustre.org/
34 * Lustre is a trademark of Sun Microsystems, Inc.
36 * lustre/ldlm/ldlm_lockd.c
38 * Author: Peter Braam <braam@clusterfs.com>
39 * Author: Phil Schwan <phil@clusterfs.com>
42 #define DEBUG_SUBSYSTEM S_LDLM
44 #include "../../include/linux/libcfs/libcfs.h"
45 #include "../include/lustre_dlm.h"
46 #include "../include/obd_class.h"
47 #include <linux/list.h>
48 #include "ldlm_internal.h"
50 static int ldlm_num_threads;
51 module_param(ldlm_num_threads, int, 0444);
52 MODULE_PARM_DESC(ldlm_num_threads, "number of DLM service threads to start");
54 static char *ldlm_cpts;
55 module_param(ldlm_cpts, charp, 0444);
56 MODULE_PARM_DESC(ldlm_cpts, "CPU partitions ldlm threads should run on");
58 static struct mutex ldlm_ref_mutex;
59 static int ldlm_refcount;
61 static struct kobject *ldlm_kobj;
62 struct kset *ldlm_ns_kset;
63 static struct kset *ldlm_svc_kset;
65 struct ldlm_cb_async_args {
66 struct ldlm_cb_set_arg *ca_set_arg;
67 struct ldlm_lock *ca_lock;
72 static struct ldlm_state *ldlm_state;
76 #define ELT_TERMINATE 2
82 * blp_prio_list is used for callbacks that should be handled
83 * as a priority. It is used for LDLM_FL_DISCARD_DATA requests.
86 struct list_head blp_prio_list;
89 * blp_list is used for all other callbacks which are likely
90 * to take longer to process.
92 struct list_head blp_list;
94 wait_queue_head_t blp_waitq;
95 struct completion blp_comp;
96 atomic_t blp_num_threads;
97 atomic_t blp_busy_threads;
102 struct ldlm_bl_work_item {
103 struct list_head blwi_entry;
104 struct ldlm_namespace *blwi_ns;
105 struct ldlm_lock_desc blwi_ld;
106 struct ldlm_lock *blwi_lock;
107 struct list_head blwi_head;
109 struct completion blwi_comp;
110 ldlm_cancel_flags_t blwi_flags;
111 int blwi_mem_pressure;
115 * Callback handler for receiving incoming blocking ASTs.
117 * This can only happen on client side.
119 void ldlm_handle_bl_callback(struct ldlm_namespace *ns,
120 struct ldlm_lock_desc *ld, struct ldlm_lock *lock)
124 LDLM_DEBUG(lock, "client blocking AST callback handler");
126 lock_res_and_lock(lock);
127 lock->l_flags |= LDLM_FL_CBPENDING;
129 if (lock->l_flags & LDLM_FL_CANCEL_ON_BLOCK)
130 lock->l_flags |= LDLM_FL_CANCEL;
132 do_ast = !lock->l_readers && !lock->l_writers;
133 unlock_res_and_lock(lock);
137 "Lock %p already unused, calling callback (%p)\n", lock,
138 lock->l_blocking_ast);
139 if (lock->l_blocking_ast != NULL)
140 lock->l_blocking_ast(lock, ld, lock->l_ast_data,
144 "Lock %p is referenced, will be cancelled later\n",
148 LDLM_DEBUG(lock, "client blocking callback handler END");
149 LDLM_LOCK_RELEASE(lock);
153 * Callback handler for receiving incoming completion ASTs.
155 * This only can happen on client side.
157 static void ldlm_handle_cp_callback(struct ptlrpc_request *req,
158 struct ldlm_namespace *ns,
159 struct ldlm_request *dlm_req,
160 struct ldlm_lock *lock)
166 LDLM_DEBUG(lock, "client completion callback handler START");
168 if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_CANCEL_BL_CB_RACE)) {
169 int to = cfs_time_seconds(1);
172 set_current_state(TASK_INTERRUPTIBLE);
173 schedule_timeout(to);
174 if (lock->l_granted_mode == lock->l_req_mode ||
175 lock->l_flags & LDLM_FL_DESTROYED)
180 lvb_len = req_capsule_get_size(&req->rq_pill, &RMF_DLM_LVB, RCL_CLIENT);
182 LDLM_ERROR(lock, "Fail to get lvb_len, rc = %d", lvb_len);
185 } else if (lvb_len > 0) {
186 if (lock->l_lvb_len > 0) {
187 /* for extent lock, lvb contains ost_lvb{}. */
188 LASSERT(lock->l_lvb_data != NULL);
190 if (unlikely(lock->l_lvb_len < lvb_len)) {
191 LDLM_ERROR(lock, "Replied LVB is larger than expectation, expected = %d, replied = %d",
192 lock->l_lvb_len, lvb_len);
196 } else if (ldlm_has_layout(lock)) { /* for layout lock, lvb has
200 lvb_data = kzalloc(lvb_len, GFP_NOFS);
202 LDLM_ERROR(lock, "No memory: %d.\n", lvb_len);
207 lock_res_and_lock(lock);
208 LASSERT(lock->l_lvb_data == NULL);
209 lock->l_lvb_type = LVB_T_LAYOUT;
210 lock->l_lvb_data = lvb_data;
211 lock->l_lvb_len = lvb_len;
212 unlock_res_and_lock(lock);
216 lock_res_and_lock(lock);
217 if ((lock->l_flags & LDLM_FL_DESTROYED) ||
218 lock->l_granted_mode == lock->l_req_mode) {
219 /* bug 11300: the lock has already been granted */
220 unlock_res_and_lock(lock);
221 LDLM_DEBUG(lock, "Double grant race happened");
226 /* If we receive the completion AST before the actual enqueue returned,
227 * then we might need to switch lock modes, resources, or extents. */
228 if (dlm_req->lock_desc.l_granted_mode != lock->l_req_mode) {
229 lock->l_req_mode = dlm_req->lock_desc.l_granted_mode;
230 LDLM_DEBUG(lock, "completion AST, new lock mode");
233 if (lock->l_resource->lr_type != LDLM_PLAIN) {
234 ldlm_convert_policy_to_local(req->rq_export,
235 dlm_req->lock_desc.l_resource.lr_type,
236 &dlm_req->lock_desc.l_policy_data,
237 &lock->l_policy_data);
238 LDLM_DEBUG(lock, "completion AST, new policy data");
241 ldlm_resource_unlink_lock(lock);
242 if (memcmp(&dlm_req->lock_desc.l_resource.lr_name,
243 &lock->l_resource->lr_name,
244 sizeof(lock->l_resource->lr_name)) != 0) {
245 unlock_res_and_lock(lock);
246 rc = ldlm_lock_change_resource(ns, lock,
247 &dlm_req->lock_desc.l_resource.lr_name);
249 LDLM_ERROR(lock, "Failed to allocate resource");
252 LDLM_DEBUG(lock, "completion AST, new resource");
253 CERROR("change resource!\n");
254 lock_res_and_lock(lock);
257 if (dlm_req->lock_flags & LDLM_FL_AST_SENT) {
258 /* BL_AST locks are not needed in LRU.
259 * Let ldlm_cancel_lru() be fast. */
260 ldlm_lock_remove_from_lru(lock);
261 lock->l_flags |= LDLM_FL_CBPENDING | LDLM_FL_BL_AST;
262 LDLM_DEBUG(lock, "completion AST includes blocking AST");
265 if (lock->l_lvb_len > 0) {
266 rc = ldlm_fill_lvb(lock, &req->rq_pill, RCL_CLIENT,
267 lock->l_lvb_data, lvb_len);
269 unlock_res_and_lock(lock);
274 ldlm_grant_lock(lock, &ast_list);
275 unlock_res_and_lock(lock);
277 LDLM_DEBUG(lock, "callback handler finished, about to run_ast_work");
279 /* Let Enqueue to call osc_lock_upcall() and initialize
281 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_ENQ_RACE, 2);
283 ldlm_run_ast_work(ns, &ast_list, LDLM_WORK_CP_AST);
285 LDLM_DEBUG_NOLOCK("client completion callback handler END (lock %p)",
291 lock_res_and_lock(lock);
292 lock->l_flags |= LDLM_FL_FAILED;
293 unlock_res_and_lock(lock);
294 wake_up(&lock->l_waitq);
296 LDLM_LOCK_RELEASE(lock);
300 * Callback handler for receiving incoming glimpse ASTs.
302 * This only can happen on client side. After handling the glimpse AST
303 * we also consider dropping the lock here if it is unused locally for a
306 static void ldlm_handle_gl_callback(struct ptlrpc_request *req,
307 struct ldlm_namespace *ns,
308 struct ldlm_request *dlm_req,
309 struct ldlm_lock *lock)
313 LDLM_DEBUG(lock, "client glimpse AST callback handler");
315 if (lock->l_glimpse_ast != NULL)
316 rc = lock->l_glimpse_ast(lock, req);
318 if (req->rq_repmsg != NULL) {
325 lock_res_and_lock(lock);
326 if (lock->l_granted_mode == LCK_PW &&
327 !lock->l_readers && !lock->l_writers &&
328 cfs_time_after(cfs_time_current(),
329 cfs_time_add(lock->l_last_used,
330 cfs_time_seconds(10)))) {
331 unlock_res_and_lock(lock);
332 if (ldlm_bl_to_thread_lock(ns, NULL, lock))
333 ldlm_handle_bl_callback(ns, NULL, lock);
337 unlock_res_and_lock(lock);
338 LDLM_LOCK_RELEASE(lock);
341 static int ldlm_callback_reply(struct ptlrpc_request *req, int rc)
343 if (req->rq_no_reply)
347 if (!req->rq_packed_final) {
348 rc = lustre_pack_reply(req, 1, NULL, NULL);
352 return ptlrpc_reply(req);
355 static int __ldlm_bl_to_thread(struct ldlm_bl_work_item *blwi,
356 ldlm_cancel_flags_t cancel_flags)
358 struct ldlm_bl_pool *blp = ldlm_state->ldlm_bl_pool;
360 spin_lock(&blp->blp_lock);
361 if (blwi->blwi_lock &&
362 blwi->blwi_lock->l_flags & LDLM_FL_DISCARD_DATA) {
363 /* add LDLM_FL_DISCARD_DATA requests to the priority list */
364 list_add_tail(&blwi->blwi_entry, &blp->blp_prio_list);
366 /* other blocking callbacks are added to the regular list */
367 list_add_tail(&blwi->blwi_entry, &blp->blp_list);
369 spin_unlock(&blp->blp_lock);
371 wake_up(&blp->blp_waitq);
373 /* can not check blwi->blwi_flags as blwi could be already freed in
375 if (!(cancel_flags & LCF_ASYNC))
376 wait_for_completion(&blwi->blwi_comp);
381 static inline void init_blwi(struct ldlm_bl_work_item *blwi,
382 struct ldlm_namespace *ns,
383 struct ldlm_lock_desc *ld,
384 struct list_head *cancels, int count,
385 struct ldlm_lock *lock,
386 ldlm_cancel_flags_t cancel_flags)
388 init_completion(&blwi->blwi_comp);
389 INIT_LIST_HEAD(&blwi->blwi_head);
391 if (memory_pressure_get())
392 blwi->blwi_mem_pressure = 1;
395 blwi->blwi_flags = cancel_flags;
399 list_add(&blwi->blwi_head, cancels);
400 list_del_init(cancels);
401 blwi->blwi_count = count;
403 blwi->blwi_lock = lock;
408 * Queues a list of locks \a cancels containing \a count locks
409 * for later processing by a blocking thread. If \a count is zero,
410 * then the lock referenced as \a lock is queued instead.
412 * The blocking thread would then call ->l_blocking_ast callback in the lock.
413 * If list addition fails an error is returned and caller is supposed to
414 * call ->l_blocking_ast itself.
416 static int ldlm_bl_to_thread(struct ldlm_namespace *ns,
417 struct ldlm_lock_desc *ld,
418 struct ldlm_lock *lock,
419 struct list_head *cancels, int count,
420 ldlm_cancel_flags_t cancel_flags)
422 if (cancels && count == 0)
425 if (cancel_flags & LCF_ASYNC) {
426 struct ldlm_bl_work_item *blwi;
428 blwi = kzalloc(sizeof(*blwi), GFP_NOFS);
431 init_blwi(blwi, ns, ld, cancels, count, lock, cancel_flags);
433 return __ldlm_bl_to_thread(blwi, cancel_flags);
435 /* if it is synchronous call do minimum mem alloc, as it could
436 * be triggered from kernel shrinker
438 struct ldlm_bl_work_item blwi;
440 memset(&blwi, 0, sizeof(blwi));
441 init_blwi(&blwi, ns, ld, cancels, count, lock, cancel_flags);
442 return __ldlm_bl_to_thread(&blwi, cancel_flags);
446 int ldlm_bl_to_thread_lock(struct ldlm_namespace *ns, struct ldlm_lock_desc *ld,
447 struct ldlm_lock *lock)
449 return ldlm_bl_to_thread(ns, ld, lock, NULL, 0, LCF_ASYNC);
452 int ldlm_bl_to_thread_list(struct ldlm_namespace *ns, struct ldlm_lock_desc *ld,
453 struct list_head *cancels, int count,
454 ldlm_cancel_flags_t cancel_flags)
456 return ldlm_bl_to_thread(ns, ld, NULL, cancels, count, cancel_flags);
459 /* Setinfo coming from Server (eg MDT) to Client (eg MDC)! */
460 static int ldlm_handle_setinfo(struct ptlrpc_request *req)
462 struct obd_device *obd = req->rq_export->exp_obd;
468 DEBUG_REQ(D_HSM, req, "%s: handle setinfo\n", obd->obd_name);
470 req_capsule_set(&req->rq_pill, &RQF_OBD_SET_INFO);
472 key = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
474 DEBUG_REQ(D_IOCTL, req, "no set_info key");
477 keylen = req_capsule_get_size(&req->rq_pill, &RMF_SETINFO_KEY,
479 val = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_VAL);
481 DEBUG_REQ(D_IOCTL, req, "no set_info val");
484 vallen = req_capsule_get_size(&req->rq_pill, &RMF_SETINFO_VAL,
487 /* We are responsible for swabbing contents of val */
489 if (KEY_IS(KEY_HSM_COPYTOOL_SEND))
490 /* Pass it on to mdc (the "export" in this case) */
491 rc = obd_set_info_async(req->rq_svc_thread->t_env,
493 sizeof(KEY_HSM_COPYTOOL_SEND),
494 KEY_HSM_COPYTOOL_SEND,
497 DEBUG_REQ(D_WARNING, req, "ignoring unknown key %s", key);
502 static inline void ldlm_callback_errmsg(struct ptlrpc_request *req,
503 const char *msg, int rc,
504 struct lustre_handle *handle)
506 DEBUG_REQ((req->rq_no_reply || rc) ? D_WARNING : D_DLMTRACE, req,
507 "%s: [nid %s] [rc %d] [lock %#llx]",
508 msg, libcfs_id2str(req->rq_peer), rc,
509 handle ? handle->cookie : 0);
510 if (req->rq_no_reply)
511 CWARN("No reply was sent, maybe cause bug 21636.\n");
513 CWARN("Send reply failed, maybe cause bug 21636.\n");
516 static int ldlm_handle_qc_callback(struct ptlrpc_request *req)
518 struct obd_quotactl *oqctl;
519 struct client_obd *cli = &req->rq_export->exp_obd->u.cli;
521 oqctl = req_capsule_client_get(&req->rq_pill, &RMF_OBD_QUOTACTL);
523 CERROR("Can't unpack obd_quotactl\n");
527 oqctl->qc_stat = ptlrpc_status_ntoh(oqctl->qc_stat);
529 cli->cl_qchk_stat = oqctl->qc_stat;
533 /* TODO: handle requests in a similar way as MDT: see mdt_handle_common() */
534 static int ldlm_callback_handler(struct ptlrpc_request *req)
536 struct ldlm_namespace *ns;
537 struct ldlm_request *dlm_req;
538 struct ldlm_lock *lock;
541 /* Requests arrive in sender's byte order. The ptlrpc service
542 * handler has already checked and, if necessary, byte-swapped the
543 * incoming request message body, but I am responsible for the
544 * message buffers. */
546 /* do nothing for sec context finalize */
547 if (lustre_msg_get_opc(req->rq_reqmsg) == SEC_CTX_FINI)
550 req_capsule_init(&req->rq_pill, req, RCL_SERVER);
552 if (req->rq_export == NULL) {
553 rc = ldlm_callback_reply(req, -ENOTCONN);
554 ldlm_callback_errmsg(req, "Operate on unconnected server",
559 LASSERT(req->rq_export != NULL);
560 LASSERT(req->rq_export->exp_obd != NULL);
562 switch (lustre_msg_get_opc(req->rq_reqmsg)) {
563 case LDLM_BL_CALLBACK:
564 if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_BL_CALLBACK_NET))
567 case LDLM_CP_CALLBACK:
568 if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_CP_CALLBACK_NET))
571 case LDLM_GL_CALLBACK:
572 if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_GL_CALLBACK_NET))
576 rc = ldlm_handle_setinfo(req);
577 ldlm_callback_reply(req, rc);
579 case OBD_QC_CALLBACK:
580 req_capsule_set(&req->rq_pill, &RQF_QC_CALLBACK);
581 if (OBD_FAIL_CHECK(OBD_FAIL_OBD_QC_CALLBACK_NET))
583 rc = ldlm_handle_qc_callback(req);
584 ldlm_callback_reply(req, rc);
587 CERROR("unknown opcode %u\n",
588 lustre_msg_get_opc(req->rq_reqmsg));
589 ldlm_callback_reply(req, -EPROTO);
593 ns = req->rq_export->exp_obd->obd_namespace;
596 req_capsule_set(&req->rq_pill, &RQF_LDLM_CALLBACK);
598 dlm_req = req_capsule_client_get(&req->rq_pill, &RMF_DLM_REQ);
599 if (dlm_req == NULL) {
600 rc = ldlm_callback_reply(req, -EPROTO);
601 ldlm_callback_errmsg(req, "Operate without parameter", rc,
606 /* Force a known safe race, send a cancel to the server for a lock
607 * which the server has already started a blocking callback on. */
608 if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_CANCEL_BL_CB_RACE) &&
609 lustre_msg_get_opc(req->rq_reqmsg) == LDLM_BL_CALLBACK) {
610 rc = ldlm_cli_cancel(&dlm_req->lock_handle[0], 0);
612 CERROR("ldlm_cli_cancel: %d\n", rc);
615 lock = ldlm_handle2lock_long(&dlm_req->lock_handle[0], 0);
617 CDEBUG(D_DLMTRACE, "callback on lock %#llx - lock disappeared\n",
618 dlm_req->lock_handle[0].cookie);
619 rc = ldlm_callback_reply(req, -EINVAL);
620 ldlm_callback_errmsg(req, "Operate with invalid parameter", rc,
621 &dlm_req->lock_handle[0]);
625 if ((lock->l_flags & LDLM_FL_FAIL_LOC) &&
626 lustre_msg_get_opc(req->rq_reqmsg) == LDLM_BL_CALLBACK)
627 OBD_RACE(OBD_FAIL_LDLM_CP_BL_RACE);
629 /* Copy hints/flags (e.g. LDLM_FL_DISCARD_DATA) from AST. */
630 lock_res_and_lock(lock);
631 lock->l_flags |= ldlm_flags_from_wire(dlm_req->lock_flags &
633 if (lustre_msg_get_opc(req->rq_reqmsg) == LDLM_BL_CALLBACK) {
634 /* If somebody cancels lock and cache is already dropped,
635 * or lock is failed before cp_ast received on client,
636 * we can tell the server we have no lock. Otherwise, we
637 * should send cancel after dropping the cache. */
638 if (((lock->l_flags & LDLM_FL_CANCELING) &&
639 (lock->l_flags & LDLM_FL_BL_DONE)) ||
640 (lock->l_flags & LDLM_FL_FAILED)) {
641 LDLM_DEBUG(lock, "callback on lock %#llx - lock disappeared\n",
642 dlm_req->lock_handle[0].cookie);
643 unlock_res_and_lock(lock);
644 LDLM_LOCK_RELEASE(lock);
645 rc = ldlm_callback_reply(req, -EINVAL);
646 ldlm_callback_errmsg(req, "Operate on stale lock", rc,
647 &dlm_req->lock_handle[0]);
650 /* BL_AST locks are not needed in LRU.
651 * Let ldlm_cancel_lru() be fast. */
652 ldlm_lock_remove_from_lru(lock);
653 lock->l_flags |= LDLM_FL_BL_AST;
655 unlock_res_and_lock(lock);
657 /* We want the ost thread to get this reply so that it can respond
658 * to ost requests (write cache writeback) that might be triggered
661 * But we'd also like to be able to indicate in the reply that we're
662 * cancelling right now, because it's unused, or have an intent result
663 * in the reply, so we might have to push the responsibility for sending
664 * the reply down into the AST handlers, alas. */
666 switch (lustre_msg_get_opc(req->rq_reqmsg)) {
667 case LDLM_BL_CALLBACK:
668 CDEBUG(D_INODE, "blocking ast\n");
669 req_capsule_extend(&req->rq_pill, &RQF_LDLM_BL_CALLBACK);
670 if (!(lock->l_flags & LDLM_FL_CANCEL_ON_BLOCK)) {
671 rc = ldlm_callback_reply(req, 0);
672 if (req->rq_no_reply || rc)
673 ldlm_callback_errmsg(req, "Normal process", rc,
674 &dlm_req->lock_handle[0]);
676 if (ldlm_bl_to_thread_lock(ns, &dlm_req->lock_desc, lock))
677 ldlm_handle_bl_callback(ns, &dlm_req->lock_desc, lock);
679 case LDLM_CP_CALLBACK:
680 CDEBUG(D_INODE, "completion ast\n");
681 req_capsule_extend(&req->rq_pill, &RQF_LDLM_CP_CALLBACK);
682 ldlm_callback_reply(req, 0);
683 ldlm_handle_cp_callback(req, ns, dlm_req, lock);
685 case LDLM_GL_CALLBACK:
686 CDEBUG(D_INODE, "glimpse ast\n");
687 req_capsule_extend(&req->rq_pill, &RQF_LDLM_GL_CALLBACK);
688 ldlm_handle_gl_callback(req, ns, dlm_req, lock);
691 LBUG(); /* checked above */
697 static struct ldlm_bl_work_item *ldlm_bl_get_work(struct ldlm_bl_pool *blp)
699 struct ldlm_bl_work_item *blwi = NULL;
700 static unsigned int num_bl;
702 spin_lock(&blp->blp_lock);
703 /* process a request from the blp_list at least every blp_num_threads */
704 if (!list_empty(&blp->blp_list) &&
705 (list_empty(&blp->blp_prio_list) || num_bl == 0))
706 blwi = list_entry(blp->blp_list.next,
707 struct ldlm_bl_work_item, blwi_entry);
709 if (!list_empty(&blp->blp_prio_list))
710 blwi = list_entry(blp->blp_prio_list.next,
711 struct ldlm_bl_work_item,
715 if (++num_bl >= atomic_read(&blp->blp_num_threads))
717 list_del(&blwi->blwi_entry);
719 spin_unlock(&blp->blp_lock);
724 /* This only contains temporary data until the thread starts */
725 struct ldlm_bl_thread_data {
726 char bltd_name[CFS_CURPROC_COMM_MAX];
727 struct ldlm_bl_pool *bltd_blp;
728 struct completion bltd_comp;
732 static int ldlm_bl_thread_main(void *arg);
734 static int ldlm_bl_thread_start(struct ldlm_bl_pool *blp)
736 struct ldlm_bl_thread_data bltd = { .bltd_blp = blp };
737 struct task_struct *task;
739 init_completion(&bltd.bltd_comp);
740 bltd.bltd_num = atomic_read(&blp->blp_num_threads);
741 snprintf(bltd.bltd_name, sizeof(bltd.bltd_name),
742 "ldlm_bl_%02d", bltd.bltd_num);
743 task = kthread_run(ldlm_bl_thread_main, &bltd, "%s", bltd.bltd_name);
745 CERROR("cannot start LDLM thread ldlm_bl_%02d: rc %ld\n",
746 atomic_read(&blp->blp_num_threads), PTR_ERR(task));
747 return PTR_ERR(task);
749 wait_for_completion(&bltd.bltd_comp);
755 * Main blocking requests processing thread.
757 * Callers put locks into its queue by calling ldlm_bl_to_thread.
758 * This thread in the end ends up doing actual call to ->l_blocking_ast
761 static int ldlm_bl_thread_main(void *arg)
763 struct ldlm_bl_pool *blp;
766 struct ldlm_bl_thread_data *bltd = arg;
768 blp = bltd->bltd_blp;
770 atomic_inc(&blp->blp_num_threads);
771 atomic_inc(&blp->blp_busy_threads);
773 complete(&bltd->bltd_comp);
774 /* cannot use bltd after this, it is only on caller's stack */
778 struct l_wait_info lwi = { 0 };
779 struct ldlm_bl_work_item *blwi = NULL;
782 blwi = ldlm_bl_get_work(blp);
785 atomic_dec(&blp->blp_busy_threads);
786 l_wait_event_exclusive(blp->blp_waitq,
787 (blwi = ldlm_bl_get_work(blp)) != NULL,
789 busy = atomic_inc_return(&blp->blp_busy_threads);
791 busy = atomic_read(&blp->blp_busy_threads);
794 if (blwi->blwi_ns == NULL)
795 /* added by ldlm_cleanup() */
798 /* Not fatal if racy and have a few too many threads */
799 if (unlikely(busy < blp->blp_max_threads &&
800 busy >= atomic_read(&blp->blp_num_threads) &&
801 !blwi->blwi_mem_pressure))
802 /* discard the return value, we tried */
803 ldlm_bl_thread_start(blp);
805 if (blwi->blwi_mem_pressure)
806 memory_pressure_set();
808 if (blwi->blwi_count) {
810 /* The special case when we cancel locks in LRU
811 * asynchronously, we pass the list of locks here.
812 * Thus locks are marked LDLM_FL_CANCELING, but NOT
813 * canceled locally yet. */
814 count = ldlm_cli_cancel_list_local(&blwi->blwi_head,
817 ldlm_cli_cancel_list(&blwi->blwi_head, count, NULL,
820 ldlm_handle_bl_callback(blwi->blwi_ns, &blwi->blwi_ld,
823 if (blwi->blwi_mem_pressure)
824 memory_pressure_clr();
826 if (blwi->blwi_flags & LCF_ASYNC)
829 complete(&blwi->blwi_comp);
832 atomic_dec(&blp->blp_busy_threads);
833 atomic_dec(&blp->blp_num_threads);
834 complete(&blp->blp_comp);
838 static int ldlm_setup(void);
839 static int ldlm_cleanup(void);
841 int ldlm_get_ref(void)
845 mutex_lock(&ldlm_ref_mutex);
846 if (++ldlm_refcount == 1) {
851 mutex_unlock(&ldlm_ref_mutex);
855 EXPORT_SYMBOL(ldlm_get_ref);
857 void ldlm_put_ref(void)
859 mutex_lock(&ldlm_ref_mutex);
860 if (ldlm_refcount == 1) {
861 int rc = ldlm_cleanup();
864 CERROR("ldlm_cleanup failed: %d\n", rc);
870 mutex_unlock(&ldlm_ref_mutex);
872 EXPORT_SYMBOL(ldlm_put_ref);
874 extern unsigned int ldlm_cancel_unused_locks_before_replay;
876 static ssize_t cancel_unused_locks_before_replay_show(struct kobject *kobj,
877 struct attribute *attr,
880 return sprintf(buf, "%d\n", ldlm_cancel_unused_locks_before_replay);
883 static ssize_t cancel_unused_locks_before_replay_store(struct kobject *kobj,
884 struct attribute *attr,
891 rc = kstrtoul(buffer, 10, &val);
895 ldlm_cancel_unused_locks_before_replay = val;
899 LUSTRE_RW_ATTR(cancel_unused_locks_before_replay);
901 /* These are for root of /sys/fs/lustre/ldlm */
902 static struct attribute *ldlm_attrs[] = {
903 &lustre_attr_cancel_unused_locks_before_replay.attr,
907 static struct attribute_group ldlm_attr_group = {
911 static int ldlm_setup(void)
913 static struct ptlrpc_service_conf conf;
914 struct ldlm_bl_pool *blp = NULL;
918 if (ldlm_state != NULL)
921 ldlm_state = kzalloc(sizeof(*ldlm_state), GFP_NOFS);
925 ldlm_kobj = kobject_create_and_add("ldlm", lustre_kobj);
931 rc = sysfs_create_group(ldlm_kobj, &ldlm_attr_group);
935 ldlm_ns_kset = kset_create_and_add("namespaces", NULL, ldlm_kobj);
941 ldlm_svc_kset = kset_create_and_add("services", NULL, ldlm_kobj);
942 if (!ldlm_svc_kset) {
947 rc = ldlm_debugfs_setup();
951 memset(&conf, 0, sizeof(conf));
952 conf = (typeof(conf)) {
953 .psc_name = "ldlm_cbd",
954 .psc_watchdog_factor = 2,
956 .bc_nbufs = LDLM_CLIENT_NBUFS,
957 .bc_buf_size = LDLM_BUFSIZE,
958 .bc_req_max_size = LDLM_MAXREQSIZE,
959 .bc_rep_max_size = LDLM_MAXREPSIZE,
960 .bc_req_portal = LDLM_CB_REQUEST_PORTAL,
961 .bc_rep_portal = LDLM_CB_REPLY_PORTAL,
964 .tc_thr_name = "ldlm_cb",
965 .tc_thr_factor = LDLM_THR_FACTOR,
966 .tc_nthrs_init = LDLM_NTHRS_INIT,
967 .tc_nthrs_base = LDLM_NTHRS_BASE,
968 .tc_nthrs_max = LDLM_NTHRS_MAX,
969 .tc_nthrs_user = ldlm_num_threads,
970 .tc_cpu_affinity = 1,
971 .tc_ctx_tags = LCT_MD_THREAD | LCT_DT_THREAD,
974 .cc_pattern = ldlm_cpts,
977 .so_req_handler = ldlm_callback_handler,
980 ldlm_state->ldlm_cb_service =
981 ptlrpc_register_service(&conf, ldlm_svc_kset,
982 ldlm_svc_debugfs_dir);
983 if (IS_ERR(ldlm_state->ldlm_cb_service)) {
984 CERROR("failed to start service\n");
985 rc = PTR_ERR(ldlm_state->ldlm_cb_service);
986 ldlm_state->ldlm_cb_service = NULL;
990 blp = kzalloc(sizeof(*blp), GFP_NOFS);
995 ldlm_state->ldlm_bl_pool = blp;
997 spin_lock_init(&blp->blp_lock);
998 INIT_LIST_HEAD(&blp->blp_list);
999 INIT_LIST_HEAD(&blp->blp_prio_list);
1000 init_waitqueue_head(&blp->blp_waitq);
1001 atomic_set(&blp->blp_num_threads, 0);
1002 atomic_set(&blp->blp_busy_threads, 0);
1004 if (ldlm_num_threads == 0) {
1005 blp->blp_min_threads = LDLM_NTHRS_INIT;
1006 blp->blp_max_threads = LDLM_NTHRS_MAX;
1008 blp->blp_min_threads = blp->blp_max_threads =
1009 min_t(int, LDLM_NTHRS_MAX, max_t(int, LDLM_NTHRS_INIT,
1013 for (i = 0; i < blp->blp_min_threads; i++) {
1014 rc = ldlm_bl_thread_start(blp);
1019 rc = ldlm_pools_init();
1021 CERROR("Failed to initialize LDLM pools: %d\n", rc);
1031 static int ldlm_cleanup(void)
1033 if (!list_empty(ldlm_namespace_list(LDLM_NAMESPACE_SERVER)) ||
1034 !list_empty(ldlm_namespace_list(LDLM_NAMESPACE_CLIENT))) {
1035 CERROR("ldlm still has namespaces; clean these up first.\n");
1036 ldlm_dump_all_namespaces(LDLM_NAMESPACE_SERVER, D_DLMTRACE);
1037 ldlm_dump_all_namespaces(LDLM_NAMESPACE_CLIENT, D_DLMTRACE);
1043 if (ldlm_state->ldlm_bl_pool != NULL) {
1044 struct ldlm_bl_pool *blp = ldlm_state->ldlm_bl_pool;
1046 while (atomic_read(&blp->blp_num_threads) > 0) {
1047 struct ldlm_bl_work_item blwi = { .blwi_ns = NULL };
1049 init_completion(&blp->blp_comp);
1051 spin_lock(&blp->blp_lock);
1052 list_add_tail(&blwi.blwi_entry, &blp->blp_list);
1053 wake_up(&blp->blp_waitq);
1054 spin_unlock(&blp->blp_lock);
1056 wait_for_completion(&blp->blp_comp);
1062 if (ldlm_state->ldlm_cb_service != NULL)
1063 ptlrpc_unregister_service(ldlm_state->ldlm_cb_service);
1066 kset_unregister(ldlm_ns_kset);
1068 kset_unregister(ldlm_svc_kset);
1070 kobject_put(ldlm_kobj);
1072 ldlm_debugfs_cleanup();
1082 mutex_init(&ldlm_ref_mutex);
1083 mutex_init(ldlm_namespace_lock(LDLM_NAMESPACE_SERVER));
1084 mutex_init(ldlm_namespace_lock(LDLM_NAMESPACE_CLIENT));
1085 ldlm_resource_slab = kmem_cache_create("ldlm_resources",
1086 sizeof(struct ldlm_resource), 0,
1087 SLAB_HWCACHE_ALIGN, NULL);
1088 if (ldlm_resource_slab == NULL)
1091 ldlm_lock_slab = kmem_cache_create("ldlm_locks",
1092 sizeof(struct ldlm_lock), 0,
1093 SLAB_HWCACHE_ALIGN | SLAB_DESTROY_BY_RCU, NULL);
1094 if (ldlm_lock_slab == NULL) {
1095 kmem_cache_destroy(ldlm_resource_slab);
1099 ldlm_interval_slab = kmem_cache_create("interval_node",
1100 sizeof(struct ldlm_interval),
1101 0, SLAB_HWCACHE_ALIGN, NULL);
1102 if (ldlm_interval_slab == NULL) {
1103 kmem_cache_destroy(ldlm_resource_slab);
1104 kmem_cache_destroy(ldlm_lock_slab);
1107 #if LUSTRE_TRACKS_LOCK_EXP_REFS
1108 class_export_dump_hook = ldlm_dump_export_locks;
1113 void ldlm_exit(void)
1116 CERROR("ldlm_refcount is %d in ldlm_exit!\n", ldlm_refcount);
1117 kmem_cache_destroy(ldlm_resource_slab);
1118 /* ldlm_lock_put() use RCU to call ldlm_lock_free, so need call
1119 * synchronize_rcu() to wait a grace period elapsed, so that
1120 * ldlm_lock_free() get a chance to be called. */
1122 kmem_cache_destroy(ldlm_lock_slab);
1123 kmem_cache_destroy(ldlm_interval_slab);