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.gnu.org/licenses/gpl-2.0.html
23 * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
24 * Use is subject to license terms.
26 * Copyright (c) 2011, 2016, Intel Corporation.
29 * This file is part of Lustre, http://www.lustre.org/
30 * Lustre is a trademark of Sun Microsystems, Inc.
33 #define DEBUG_SUBSYSTEM S_OSC
35 #include <libcfs/libcfs.h>
37 #include <lustre/lustre_user.h>
39 #include <lprocfs_status.h>
40 #include <lustre_debug.h>
41 #include <lustre_dlm.h>
42 #include <lustre_fid.h>
43 #include <lustre_ha.h>
44 #include <lustre_ioctl.h>
45 #include <lustre_net.h>
46 #include <lustre_obdo.h>
47 #include <lustre_param.h>
49 #include <obd_cksum.h>
50 #include <obd_class.h>
52 #include "osc_cl_internal.h"
53 #include "osc_internal.h"
55 atomic_t osc_pool_req_count;
56 unsigned int osc_reqpool_maxreqcount;
57 struct ptlrpc_request_pool *osc_rq_pool;
59 /* max memory used for request pool, unit is MB */
60 static unsigned int osc_reqpool_mem_max = 5;
61 module_param(osc_reqpool_mem_max, uint, 0444);
63 struct osc_brw_async_args {
69 struct brw_page **aa_ppga;
70 struct client_obd *aa_cli;
71 struct list_head aa_oaps;
72 struct list_head aa_exts;
75 #define osc_grant_args osc_brw_async_args
77 struct osc_setattr_args {
79 obd_enqueue_update_f sa_upcall;
83 struct osc_fsync_args {
84 struct osc_object *fa_obj;
86 obd_enqueue_update_f fa_upcall;
90 struct osc_ladvise_args {
92 obd_enqueue_update_f la_upcall;
96 struct osc_enqueue_args {
97 struct obd_export *oa_exp;
98 enum ldlm_type oa_type;
99 enum ldlm_mode oa_mode;
101 osc_enqueue_upcall_f oa_upcall;
103 struct ost_lvb *oa_lvb;
104 struct lustre_handle oa_lockh;
105 unsigned int oa_agl:1;
108 static void osc_release_ppga(struct brw_page **ppga, size_t count);
109 static int brw_interpret(const struct lu_env *env, struct ptlrpc_request *req,
112 void osc_pack_req_body(struct ptlrpc_request *req, struct obdo *oa)
114 struct ost_body *body;
116 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
119 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
122 static int osc_getattr(const struct lu_env *env, struct obd_export *exp,
125 struct ptlrpc_request *req;
126 struct ost_body *body;
130 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
134 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
136 ptlrpc_request_free(req);
140 osc_pack_req_body(req, oa);
142 ptlrpc_request_set_replen(req);
144 rc = ptlrpc_queue_wait(req);
148 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
150 GOTO(out, rc = -EPROTO);
152 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
153 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
155 oa->o_blksize = cli_brw_size(exp->exp_obd);
156 oa->o_valid |= OBD_MD_FLBLKSZ;
160 ptlrpc_req_finished(req);
165 static int osc_setattr(const struct lu_env *env, struct obd_export *exp,
168 struct ptlrpc_request *req;
169 struct ost_body *body;
173 LASSERT(oa->o_valid & OBD_MD_FLGROUP);
175 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
179 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
181 ptlrpc_request_free(req);
185 osc_pack_req_body(req, oa);
187 ptlrpc_request_set_replen(req);
189 rc = ptlrpc_queue_wait(req);
193 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
195 GOTO(out, rc = -EPROTO);
197 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
201 ptlrpc_req_finished(req);
206 static int osc_setattr_interpret(const struct lu_env *env,
207 struct ptlrpc_request *req,
208 struct osc_setattr_args *sa, int rc)
210 struct ost_body *body;
216 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
218 GOTO(out, rc = -EPROTO);
220 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, sa->sa_oa,
223 rc = sa->sa_upcall(sa->sa_cookie, rc);
227 int osc_setattr_async(struct obd_export *exp, struct obdo *oa,
228 obd_enqueue_update_f upcall, void *cookie,
229 struct ptlrpc_request_set *rqset)
231 struct ptlrpc_request *req;
232 struct osc_setattr_args *sa;
237 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
241 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
243 ptlrpc_request_free(req);
247 osc_pack_req_body(req, oa);
249 ptlrpc_request_set_replen(req);
251 /* do mds to ost setattr asynchronously */
253 /* Do not wait for response. */
254 ptlrpcd_add_req(req);
256 req->rq_interpret_reply =
257 (ptlrpc_interpterer_t)osc_setattr_interpret;
259 CLASSERT(sizeof(*sa) <= sizeof(req->rq_async_args));
260 sa = ptlrpc_req_async_args(req);
262 sa->sa_upcall = upcall;
263 sa->sa_cookie = cookie;
265 if (rqset == PTLRPCD_SET)
266 ptlrpcd_add_req(req);
268 ptlrpc_set_add_req(rqset, req);
274 static int osc_ladvise_interpret(const struct lu_env *env,
275 struct ptlrpc_request *req,
278 struct osc_ladvise_args *la = arg;
279 struct ost_body *body;
285 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
287 GOTO(out, rc = -EPROTO);
289 *la->la_oa = body->oa;
291 rc = la->la_upcall(la->la_cookie, rc);
296 * If rqset is NULL, do not wait for response. Upcall and cookie could also
297 * be NULL in this case
299 int osc_ladvise_base(struct obd_export *exp, struct obdo *oa,
300 struct ladvise_hdr *ladvise_hdr,
301 obd_enqueue_update_f upcall, void *cookie,
302 struct ptlrpc_request_set *rqset)
304 struct ptlrpc_request *req;
305 struct ost_body *body;
306 struct osc_ladvise_args *la;
308 struct lu_ladvise *req_ladvise;
309 struct lu_ladvise *ladvise = ladvise_hdr->lah_advise;
310 int num_advise = ladvise_hdr->lah_count;
311 struct ladvise_hdr *req_ladvise_hdr;
314 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_LADVISE);
318 req_capsule_set_size(&req->rq_pill, &RMF_OST_LADVISE, RCL_CLIENT,
319 num_advise * sizeof(*ladvise));
320 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_LADVISE);
322 ptlrpc_request_free(req);
325 req->rq_request_portal = OST_IO_PORTAL;
326 ptlrpc_at_set_req_timeout(req);
328 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
330 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
333 req_ladvise_hdr = req_capsule_client_get(&req->rq_pill,
334 &RMF_OST_LADVISE_HDR);
335 memcpy(req_ladvise_hdr, ladvise_hdr, sizeof(*ladvise_hdr));
337 req_ladvise = req_capsule_client_get(&req->rq_pill, &RMF_OST_LADVISE);
338 memcpy(req_ladvise, ladvise, sizeof(*ladvise) * num_advise);
339 ptlrpc_request_set_replen(req);
342 /* Do not wait for response. */
343 ptlrpcd_add_req(req);
347 req->rq_interpret_reply = osc_ladvise_interpret;
348 CLASSERT(sizeof(*la) <= sizeof(req->rq_async_args));
349 la = ptlrpc_req_async_args(req);
351 la->la_upcall = upcall;
352 la->la_cookie = cookie;
354 if (rqset == PTLRPCD_SET)
355 ptlrpcd_add_req(req);
357 ptlrpc_set_add_req(rqset, req);
362 static int osc_create(const struct lu_env *env, struct obd_export *exp,
365 struct ptlrpc_request *req;
366 struct ost_body *body;
371 LASSERT(oa->o_valid & OBD_MD_FLGROUP);
372 LASSERT(fid_seq_is_echo(ostid_seq(&oa->o_oi)));
374 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_CREATE);
376 GOTO(out, rc = -ENOMEM);
378 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_CREATE);
380 ptlrpc_request_free(req);
384 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
387 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
389 ptlrpc_request_set_replen(req);
391 rc = ptlrpc_queue_wait(req);
395 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
397 GOTO(out_req, rc = -EPROTO);
399 CDEBUG(D_INFO, "oa flags %x\n", oa->o_flags);
400 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
402 oa->o_blksize = cli_brw_size(exp->exp_obd);
403 oa->o_valid |= OBD_MD_FLBLKSZ;
405 CDEBUG(D_HA, "transno: %lld\n",
406 lustre_msg_get_transno(req->rq_repmsg));
408 ptlrpc_req_finished(req);
413 int osc_punch_base(struct obd_export *exp, struct obdo *oa,
414 obd_enqueue_update_f upcall, void *cookie,
415 struct ptlrpc_request_set *rqset)
417 struct ptlrpc_request *req;
418 struct osc_setattr_args *sa;
419 struct ost_body *body;
423 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_PUNCH);
427 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_PUNCH);
429 ptlrpc_request_free(req);
432 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
433 ptlrpc_at_set_req_timeout(req);
435 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
437 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
439 ptlrpc_request_set_replen(req);
441 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_setattr_interpret;
442 CLASSERT(sizeof(*sa) <= sizeof(req->rq_async_args));
443 sa = ptlrpc_req_async_args(req);
445 sa->sa_upcall = upcall;
446 sa->sa_cookie = cookie;
447 if (rqset == PTLRPCD_SET)
448 ptlrpcd_add_req(req);
450 ptlrpc_set_add_req(rqset, req);
455 static int osc_sync_interpret(const struct lu_env *env,
456 struct ptlrpc_request *req,
459 struct osc_fsync_args *fa = arg;
460 struct ost_body *body;
461 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
462 unsigned long valid = 0;
463 struct cl_object *obj;
469 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
471 CERROR("can't unpack ost_body\n");
472 GOTO(out, rc = -EPROTO);
475 *fa->fa_oa = body->oa;
476 obj = osc2cl(fa->fa_obj);
478 /* Update osc object's blocks attribute */
479 cl_object_attr_lock(obj);
480 if (body->oa.o_valid & OBD_MD_FLBLOCKS) {
481 attr->cat_blocks = body->oa.o_blocks;
486 cl_object_attr_update(env, obj, attr, valid);
487 cl_object_attr_unlock(obj);
490 rc = fa->fa_upcall(fa->fa_cookie, rc);
494 int osc_sync_base(struct osc_object *obj, struct obdo *oa,
495 obd_enqueue_update_f upcall, void *cookie,
496 struct ptlrpc_request_set *rqset)
498 struct obd_export *exp = osc_export(obj);
499 struct ptlrpc_request *req;
500 struct ost_body *body;
501 struct osc_fsync_args *fa;
505 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SYNC);
509 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SYNC);
511 ptlrpc_request_free(req);
515 /* overload the size and blocks fields in the oa with start/end */
516 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
518 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
520 ptlrpc_request_set_replen(req);
521 req->rq_interpret_reply = osc_sync_interpret;
523 CLASSERT(sizeof(*fa) <= sizeof(req->rq_async_args));
524 fa = ptlrpc_req_async_args(req);
527 fa->fa_upcall = upcall;
528 fa->fa_cookie = cookie;
530 if (rqset == PTLRPCD_SET)
531 ptlrpcd_add_req(req);
533 ptlrpc_set_add_req(rqset, req);
538 /* Find and cancel locally locks matched by @mode in the resource found by
539 * @objid. Found locks are added into @cancel list. Returns the amount of
540 * locks added to @cancels list. */
541 static int osc_resource_get_unused(struct obd_export *exp, struct obdo *oa,
542 struct list_head *cancels,
543 enum ldlm_mode mode, __u64 lock_flags)
545 struct ldlm_namespace *ns = exp->exp_obd->obd_namespace;
546 struct ldlm_res_id res_id;
547 struct ldlm_resource *res;
551 /* Return, i.e. cancel nothing, only if ELC is supported (flag in
552 * export) but disabled through procfs (flag in NS).
554 * This distinguishes from a case when ELC is not supported originally,
555 * when we still want to cancel locks in advance and just cancel them
556 * locally, without sending any RPC. */
557 if (exp_connect_cancelset(exp) && !ns_connect_cancelset(ns))
560 ostid_build_res_name(&oa->o_oi, &res_id);
561 res = ldlm_resource_get(ns, NULL, &res_id, 0, 0);
565 LDLM_RESOURCE_ADDREF(res);
566 count = ldlm_cancel_resource_local(res, cancels, NULL, mode,
567 lock_flags, 0, NULL);
568 LDLM_RESOURCE_DELREF(res);
569 ldlm_resource_putref(res);
573 static int osc_destroy_interpret(const struct lu_env *env,
574 struct ptlrpc_request *req, void *data,
577 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
579 atomic_dec(&cli->cl_destroy_in_flight);
580 wake_up(&cli->cl_destroy_waitq);
584 static int osc_can_send_destroy(struct client_obd *cli)
586 if (atomic_inc_return(&cli->cl_destroy_in_flight) <=
587 cli->cl_max_rpcs_in_flight) {
588 /* The destroy request can be sent */
591 if (atomic_dec_return(&cli->cl_destroy_in_flight) <
592 cli->cl_max_rpcs_in_flight) {
594 * The counter has been modified between the two atomic
597 wake_up(&cli->cl_destroy_waitq);
602 static int osc_destroy(const struct lu_env *env, struct obd_export *exp,
605 struct client_obd *cli = &exp->exp_obd->u.cli;
606 struct ptlrpc_request *req;
607 struct ost_body *body;
608 struct list_head cancels = LIST_HEAD_INIT(cancels);
613 CDEBUG(D_INFO, "oa NULL\n");
617 count = osc_resource_get_unused(exp, oa, &cancels, LCK_PW,
618 LDLM_FL_DISCARD_DATA);
620 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_DESTROY);
622 ldlm_lock_list_put(&cancels, l_bl_ast, count);
626 rc = ldlm_prep_elc_req(exp, req, LUSTRE_OST_VERSION, OST_DESTROY,
629 ptlrpc_request_free(req);
633 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
634 ptlrpc_at_set_req_timeout(req);
636 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
638 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
640 ptlrpc_request_set_replen(req);
642 req->rq_interpret_reply = osc_destroy_interpret;
643 if (!osc_can_send_destroy(cli)) {
644 struct l_wait_info lwi = LWI_INTR(LWI_ON_SIGNAL_NOOP, NULL);
647 * Wait until the number of on-going destroy RPCs drops
648 * under max_rpc_in_flight
650 l_wait_event_exclusive(cli->cl_destroy_waitq,
651 osc_can_send_destroy(cli), &lwi);
654 /* Do not wait for response */
655 ptlrpcd_add_req(req);
659 static void osc_announce_cached(struct client_obd *cli, struct obdo *oa,
662 u64 bits = OBD_MD_FLBLOCKS | OBD_MD_FLGRANT;
664 LASSERT(!(oa->o_valid & bits));
667 spin_lock(&cli->cl_loi_list_lock);
668 if (OCD_HAS_FLAG(&cli->cl_import->imp_connect_data, GRANT_PARAM))
669 oa->o_dirty = cli->cl_dirty_grant;
671 oa->o_dirty = cli->cl_dirty_pages << PAGE_SHIFT;
672 if (unlikely(cli->cl_dirty_pages - cli->cl_dirty_transit >
673 cli->cl_dirty_max_pages)) {
674 CERROR("dirty %lu - %lu > dirty_max %lu\n",
675 cli->cl_dirty_pages, cli->cl_dirty_transit,
676 cli->cl_dirty_max_pages);
678 } else if (unlikely(atomic_long_read(&obd_dirty_pages) -
679 atomic_long_read(&obd_dirty_transit_pages) >
680 (long)(obd_max_dirty_pages + 1))) {
681 /* The atomic_read() allowing the atomic_inc() are
682 * not covered by a lock thus they may safely race and trip
683 * this CERROR() unless we add in a small fudge factor (+1). */
684 CERROR("%s: dirty %ld - %ld > system dirty_max %ld\n",
685 cli_name(cli), atomic_long_read(&obd_dirty_pages),
686 atomic_long_read(&obd_dirty_transit_pages),
687 obd_max_dirty_pages);
689 } else if (unlikely(cli->cl_dirty_max_pages - cli->cl_dirty_pages >
691 CERROR("dirty %lu - dirty_max %lu too big???\n",
692 cli->cl_dirty_pages, cli->cl_dirty_max_pages);
695 unsigned long nrpages;
697 nrpages = cli->cl_max_pages_per_rpc;
698 nrpages *= cli->cl_max_rpcs_in_flight + 1;
699 nrpages = max(nrpages, cli->cl_dirty_max_pages);
700 oa->o_undirty = nrpages << PAGE_SHIFT;
701 if (OCD_HAS_FLAG(&cli->cl_import->imp_connect_data,
705 /* take extent tax into account when asking for more
707 nrextents = (nrpages + cli->cl_max_extent_pages - 1) /
708 cli->cl_max_extent_pages;
709 oa->o_undirty += nrextents * cli->cl_grant_extent_tax;
712 oa->o_grant = cli->cl_avail_grant + cli->cl_reserved_grant;
713 oa->o_dropped = cli->cl_lost_grant;
714 cli->cl_lost_grant = 0;
715 spin_unlock(&cli->cl_loi_list_lock);
716 CDEBUG(D_CACHE, "dirty: %llu undirty: %u dropped %u grant: %llu\n",
717 oa->o_dirty, oa->o_undirty, oa->o_dropped, oa->o_grant);
720 void osc_update_next_shrink(struct client_obd *cli)
722 cli->cl_next_shrink_grant =
723 cfs_time_shift(cli->cl_grant_shrink_interval);
724 CDEBUG(D_CACHE, "next time %ld to shrink grant \n",
725 cli->cl_next_shrink_grant);
728 static void __osc_update_grant(struct client_obd *cli, u64 grant)
730 spin_lock(&cli->cl_loi_list_lock);
731 cli->cl_avail_grant += grant;
732 spin_unlock(&cli->cl_loi_list_lock);
735 static void osc_update_grant(struct client_obd *cli, struct ost_body *body)
737 if (body->oa.o_valid & OBD_MD_FLGRANT) {
738 CDEBUG(D_CACHE, "got %llu extra grant\n", body->oa.o_grant);
739 __osc_update_grant(cli, body->oa.o_grant);
743 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
744 u32 keylen, void *key,
745 u32 vallen, void *val,
746 struct ptlrpc_request_set *set);
748 static int osc_shrink_grant_interpret(const struct lu_env *env,
749 struct ptlrpc_request *req,
752 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
753 struct obdo *oa = ((struct osc_grant_args *)aa)->aa_oa;
754 struct ost_body *body;
757 __osc_update_grant(cli, oa->o_grant);
761 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
763 osc_update_grant(cli, body);
769 static void osc_shrink_grant_local(struct client_obd *cli, struct obdo *oa)
771 spin_lock(&cli->cl_loi_list_lock);
772 oa->o_grant = cli->cl_avail_grant / 4;
773 cli->cl_avail_grant -= oa->o_grant;
774 spin_unlock(&cli->cl_loi_list_lock);
775 if (!(oa->o_valid & OBD_MD_FLFLAGS)) {
776 oa->o_valid |= OBD_MD_FLFLAGS;
779 oa->o_flags |= OBD_FL_SHRINK_GRANT;
780 osc_update_next_shrink(cli);
783 /* Shrink the current grant, either from some large amount to enough for a
784 * full set of in-flight RPCs, or if we have already shrunk to that limit
785 * then to enough for a single RPC. This avoids keeping more grant than
786 * needed, and avoids shrinking the grant piecemeal. */
787 static int osc_shrink_grant(struct client_obd *cli)
789 __u64 target_bytes = (cli->cl_max_rpcs_in_flight + 1) *
790 (cli->cl_max_pages_per_rpc << PAGE_SHIFT);
792 spin_lock(&cli->cl_loi_list_lock);
793 if (cli->cl_avail_grant <= target_bytes)
794 target_bytes = cli->cl_max_pages_per_rpc << PAGE_SHIFT;
795 spin_unlock(&cli->cl_loi_list_lock);
797 return osc_shrink_grant_to_target(cli, target_bytes);
800 int osc_shrink_grant_to_target(struct client_obd *cli, __u64 target_bytes)
803 struct ost_body *body;
806 spin_lock(&cli->cl_loi_list_lock);
807 /* Don't shrink if we are already above or below the desired limit
808 * We don't want to shrink below a single RPC, as that will negatively
809 * impact block allocation and long-term performance. */
810 if (target_bytes < cli->cl_max_pages_per_rpc << PAGE_SHIFT)
811 target_bytes = cli->cl_max_pages_per_rpc << PAGE_SHIFT;
813 if (target_bytes >= cli->cl_avail_grant) {
814 spin_unlock(&cli->cl_loi_list_lock);
817 spin_unlock(&cli->cl_loi_list_lock);
823 osc_announce_cached(cli, &body->oa, 0);
825 spin_lock(&cli->cl_loi_list_lock);
826 body->oa.o_grant = cli->cl_avail_grant - target_bytes;
827 cli->cl_avail_grant = target_bytes;
828 spin_unlock(&cli->cl_loi_list_lock);
829 if (!(body->oa.o_valid & OBD_MD_FLFLAGS)) {
830 body->oa.o_valid |= OBD_MD_FLFLAGS;
831 body->oa.o_flags = 0;
833 body->oa.o_flags |= OBD_FL_SHRINK_GRANT;
834 osc_update_next_shrink(cli);
836 rc = osc_set_info_async(NULL, cli->cl_import->imp_obd->obd_self_export,
837 sizeof(KEY_GRANT_SHRINK), KEY_GRANT_SHRINK,
838 sizeof(*body), body, NULL);
840 __osc_update_grant(cli, body->oa.o_grant);
845 static int osc_should_shrink_grant(struct client_obd *client)
847 cfs_time_t time = cfs_time_current();
848 cfs_time_t next_shrink = client->cl_next_shrink_grant;
850 if ((client->cl_import->imp_connect_data.ocd_connect_flags &
851 OBD_CONNECT_GRANT_SHRINK) == 0)
854 if (cfs_time_aftereq(time, next_shrink - 5 * CFS_TICK)) {
855 /* Get the current RPC size directly, instead of going via:
856 * cli_brw_size(obd->u.cli.cl_import->imp_obd->obd_self_export)
857 * Keep comment here so that it can be found by searching. */
858 int brw_size = client->cl_max_pages_per_rpc << PAGE_SHIFT;
860 if (client->cl_import->imp_state == LUSTRE_IMP_FULL &&
861 client->cl_avail_grant > brw_size)
864 osc_update_next_shrink(client);
869 static int osc_grant_shrink_grant_cb(struct timeout_item *item, void *data)
871 struct client_obd *client;
873 list_for_each_entry(client, &item->ti_obd_list, cl_grant_shrink_list) {
874 if (osc_should_shrink_grant(client))
875 osc_shrink_grant(client);
880 static int osc_add_shrink_grant(struct client_obd *client)
884 rc = ptlrpc_add_timeout_client(client->cl_grant_shrink_interval,
886 osc_grant_shrink_grant_cb, NULL,
887 &client->cl_grant_shrink_list);
889 CERROR("add grant client %s error %d\n", cli_name(client), rc);
892 CDEBUG(D_CACHE, "add grant client %s\n", cli_name(client));
893 osc_update_next_shrink(client);
897 static int osc_del_shrink_grant(struct client_obd *client)
899 return ptlrpc_del_timeout_client(&client->cl_grant_shrink_list,
903 static void osc_init_grant(struct client_obd *cli, struct obd_connect_data *ocd)
906 * ocd_grant is the total grant amount we're expect to hold: if we've
907 * been evicted, it's the new avail_grant amount, cl_dirty_pages will
908 * drop to 0 as inflight RPCs fail out; otherwise, it's avail_grant +
911 * race is tolerable here: if we're evicted, but imp_state already
912 * left EVICTED state, then cl_dirty_pages must be 0 already.
914 spin_lock(&cli->cl_loi_list_lock);
915 cli->cl_avail_grant = ocd->ocd_grant;
916 if (cli->cl_import->imp_state != LUSTRE_IMP_EVICTED) {
917 cli->cl_avail_grant -= cli->cl_reserved_grant;
918 if (OCD_HAS_FLAG(ocd, GRANT_PARAM))
919 cli->cl_avail_grant -= cli->cl_dirty_grant;
921 cli->cl_avail_grant -=
922 cli->cl_dirty_pages << PAGE_SHIFT;
925 if (OCD_HAS_FLAG(ocd, GRANT_PARAM)) {
929 /* overhead for each extent insertion */
930 cli->cl_grant_extent_tax = ocd->ocd_grant_tax_kb << 10;
931 /* determine the appropriate chunk size used by osc_extent. */
932 cli->cl_chunkbits = max_t(int, PAGE_SHIFT,
933 ocd->ocd_grant_blkbits);
934 /* max_pages_per_rpc must be chunk aligned */
935 chunk_mask = ~((1 << (cli->cl_chunkbits - PAGE_SHIFT)) - 1);
936 cli->cl_max_pages_per_rpc = (cli->cl_max_pages_per_rpc +
937 ~chunk_mask) & chunk_mask;
938 /* determine maximum extent size, in #pages */
939 size = (u64)ocd->ocd_grant_max_blks << ocd->ocd_grant_blkbits;
940 cli->cl_max_extent_pages = size >> PAGE_SHIFT;
941 if (cli->cl_max_extent_pages == 0)
942 cli->cl_max_extent_pages = 1;
944 cli->cl_grant_extent_tax = 0;
945 cli->cl_chunkbits = PAGE_SHIFT;
946 cli->cl_max_extent_pages = DT_MAX_BRW_PAGES;
948 spin_unlock(&cli->cl_loi_list_lock);
950 CDEBUG(D_CACHE, "%s, setting cl_avail_grant: %ld cl_lost_grant: %ld."
951 "chunk bits: %d cl_max_extent_pages: %d\n",
953 cli->cl_avail_grant, cli->cl_lost_grant, cli->cl_chunkbits,
954 cli->cl_max_extent_pages);
956 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT_SHRINK &&
957 list_empty(&cli->cl_grant_shrink_list))
958 osc_add_shrink_grant(cli);
961 /* We assume that the reason this OSC got a short read is because it read
962 * beyond the end of a stripe file; i.e. lustre is reading a sparse file
963 * via the LOV, and it _knows_ it's reading inside the file, it's just that
964 * this stripe never got written at or beyond this stripe offset yet. */
965 static void handle_short_read(int nob_read, size_t page_count,
966 struct brw_page **pga)
971 /* skip bytes read OK */
972 while (nob_read > 0) {
973 LASSERT (page_count > 0);
975 if (pga[i]->count > nob_read) {
976 /* EOF inside this page */
977 ptr = kmap(pga[i]->pg) +
978 (pga[i]->off & ~PAGE_MASK);
979 memset(ptr + nob_read, 0, pga[i]->count - nob_read);
986 nob_read -= pga[i]->count;
991 /* zero remaining pages */
992 while (page_count-- > 0) {
993 ptr = kmap(pga[i]->pg) + (pga[i]->off & ~PAGE_MASK);
994 memset(ptr, 0, pga[i]->count);
1000 static int check_write_rcs(struct ptlrpc_request *req,
1001 int requested_nob, int niocount,
1002 size_t page_count, struct brw_page **pga)
1007 remote_rcs = req_capsule_server_sized_get(&req->rq_pill, &RMF_RCS,
1008 sizeof(*remote_rcs) *
1010 if (remote_rcs == NULL) {
1011 CDEBUG(D_INFO, "Missing/short RC vector on BRW_WRITE reply\n");
1015 /* return error if any niobuf was in error */
1016 for (i = 0; i < niocount; i++) {
1017 if ((int)remote_rcs[i] < 0)
1018 return(remote_rcs[i]);
1020 if (remote_rcs[i] != 0) {
1021 CDEBUG(D_INFO, "rc[%d] invalid (%d) req %p\n",
1022 i, remote_rcs[i], req);
1027 if (req->rq_bulk->bd_nob_transferred != requested_nob) {
1028 CERROR("Unexpected # bytes transferred: %d (requested %d)\n",
1029 req->rq_bulk->bd_nob_transferred, requested_nob);
1036 static inline int can_merge_pages(struct brw_page *p1, struct brw_page *p2)
1038 if (p1->flag != p2->flag) {
1039 unsigned mask = ~(OBD_BRW_FROM_GRANT | OBD_BRW_NOCACHE |
1040 OBD_BRW_SYNC | OBD_BRW_ASYNC |
1041 OBD_BRW_NOQUOTA | OBD_BRW_SOFT_SYNC);
1043 /* warn if we try to combine flags that we don't know to be
1044 * safe to combine */
1045 if (unlikely((p1->flag & mask) != (p2->flag & mask))) {
1046 CWARN("Saw flags 0x%x and 0x%x in the same brw, please "
1047 "report this at https://jira.hpdd.intel.com/\n",
1048 p1->flag, p2->flag);
1053 return (p1->off + p1->count == p2->off);
1056 static u32 osc_checksum_bulk(int nob, size_t pg_count,
1057 struct brw_page **pga, int opc,
1058 cksum_type_t cksum_type)
1062 struct cfs_crypto_hash_desc *hdesc;
1063 unsigned int bufsize;
1064 unsigned char cfs_alg = cksum_obd2cfs(cksum_type);
1066 LASSERT(pg_count > 0);
1068 hdesc = cfs_crypto_hash_init(cfs_alg, NULL, 0);
1069 if (IS_ERR(hdesc)) {
1070 CERROR("Unable to initialize checksum hash %s\n",
1071 cfs_crypto_hash_name(cfs_alg));
1072 return PTR_ERR(hdesc);
1075 while (nob > 0 && pg_count > 0) {
1076 unsigned int count = pga[i]->count > nob ? nob : pga[i]->count;
1078 /* corrupt the data before we compute the checksum, to
1079 * simulate an OST->client data error */
1080 if (i == 0 && opc == OST_READ &&
1081 OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_RECEIVE)) {
1082 unsigned char *ptr = kmap(pga[i]->pg);
1083 int off = pga[i]->off & ~PAGE_MASK;
1085 memcpy(ptr + off, "bad1", min_t(typeof(nob), 4, nob));
1088 cfs_crypto_hash_update_page(hdesc, pga[i]->pg,
1089 pga[i]->off & ~PAGE_MASK,
1091 LL_CDEBUG_PAGE(D_PAGE, pga[i]->pg, "off %d\n",
1092 (int)(pga[i]->off & ~PAGE_MASK));
1094 nob -= pga[i]->count;
1099 bufsize = sizeof(cksum);
1100 cfs_crypto_hash_final(hdesc, (unsigned char *)&cksum, &bufsize);
1102 /* For sending we only compute the wrong checksum instead
1103 * of corrupting the data so it is still correct on a redo */
1104 if (opc == OST_WRITE && OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_SEND))
1111 osc_brw_prep_request(int cmd, struct client_obd *cli, struct obdo *oa,
1112 u32 page_count, struct brw_page **pga,
1113 struct ptlrpc_request **reqp, int resend)
1115 struct ptlrpc_request *req;
1116 struct ptlrpc_bulk_desc *desc;
1117 struct ost_body *body;
1118 struct obd_ioobj *ioobj;
1119 struct niobuf_remote *niobuf;
1120 int niocount, i, requested_nob, opc, rc;
1121 struct osc_brw_async_args *aa;
1122 struct req_capsule *pill;
1123 struct brw_page *pg_prev;
1126 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ))
1127 RETURN(-ENOMEM); /* Recoverable */
1128 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ2))
1129 RETURN(-EINVAL); /* Fatal */
1131 if ((cmd & OBD_BRW_WRITE) != 0) {
1133 req = ptlrpc_request_alloc_pool(cli->cl_import,
1135 &RQF_OST_BRW_WRITE);
1138 req = ptlrpc_request_alloc(cli->cl_import, &RQF_OST_BRW_READ);
1143 for (niocount = i = 1; i < page_count; i++) {
1144 if (!can_merge_pages(pga[i - 1], pga[i]))
1148 pill = &req->rq_pill;
1149 req_capsule_set_size(pill, &RMF_OBD_IOOBJ, RCL_CLIENT,
1151 req_capsule_set_size(pill, &RMF_NIOBUF_REMOTE, RCL_CLIENT,
1152 niocount * sizeof(*niobuf));
1154 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, opc);
1156 ptlrpc_request_free(req);
1159 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
1160 ptlrpc_at_set_req_timeout(req);
1161 /* ask ptlrpc not to resend on EINPROGRESS since BRWs have their own
1163 req->rq_no_retry_einprogress = 1;
1165 desc = ptlrpc_prep_bulk_imp(req, page_count,
1166 cli->cl_import->imp_connect_data.ocd_brw_size >> LNET_MTU_BITS,
1167 (opc == OST_WRITE ? PTLRPC_BULK_GET_SOURCE :
1168 PTLRPC_BULK_PUT_SINK) |
1169 PTLRPC_BULK_BUF_KIOV,
1171 &ptlrpc_bulk_kiov_pin_ops);
1174 GOTO(out, rc = -ENOMEM);
1175 /* NB request now owns desc and will free it when it gets freed */
1177 body = req_capsule_client_get(pill, &RMF_OST_BODY);
1178 ioobj = req_capsule_client_get(pill, &RMF_OBD_IOOBJ);
1179 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1180 LASSERT(body != NULL && ioobj != NULL && niobuf != NULL);
1182 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
1184 obdo_to_ioobj(oa, ioobj);
1185 ioobj->ioo_bufcnt = niocount;
1186 /* The high bits of ioo_max_brw tells server _maximum_ number of bulks
1187 * that might be send for this request. The actual number is decided
1188 * when the RPC is finally sent in ptlrpc_register_bulk(). It sends
1189 * "max - 1" for old client compatibility sending "0", and also so the
1190 * the actual maximum is a power-of-two number, not one less. LU-1431 */
1191 ioobj_max_brw_set(ioobj, desc->bd_md_max_brw);
1192 LASSERT(page_count > 0);
1194 for (requested_nob = i = 0; i < page_count; i++, niobuf++) {
1195 struct brw_page *pg = pga[i];
1196 int poff = pg->off & ~PAGE_MASK;
1198 LASSERT(pg->count > 0);
1199 /* make sure there is no gap in the middle of page array */
1200 LASSERTF(page_count == 1 ||
1201 (ergo(i == 0, poff + pg->count == PAGE_SIZE) &&
1202 ergo(i > 0 && i < page_count - 1,
1203 poff == 0 && pg->count == PAGE_SIZE) &&
1204 ergo(i == page_count - 1, poff == 0)),
1205 "i: %d/%d pg: %p off: %llu, count: %u\n",
1206 i, page_count, pg, pg->off, pg->count);
1207 LASSERTF(i == 0 || pg->off > pg_prev->off,
1208 "i %d p_c %u pg %p [pri %lu ind %lu] off %llu"
1209 " prev_pg %p [pri %lu ind %lu] off %llu\n",
1211 pg->pg, page_private(pg->pg), pg->pg->index, pg->off,
1212 pg_prev->pg, page_private(pg_prev->pg),
1213 pg_prev->pg->index, pg_prev->off);
1214 LASSERT((pga[0]->flag & OBD_BRW_SRVLOCK) ==
1215 (pg->flag & OBD_BRW_SRVLOCK));
1217 desc->bd_frag_ops->add_kiov_frag(desc, pg->pg, poff, pg->count);
1218 requested_nob += pg->count;
1220 if (i > 0 && can_merge_pages(pg_prev, pg)) {
1222 niobuf->rnb_len += pg->count;
1224 niobuf->rnb_offset = pg->off;
1225 niobuf->rnb_len = pg->count;
1226 niobuf->rnb_flags = pg->flag;
1231 LASSERTF((void *)(niobuf - niocount) ==
1232 req_capsule_client_get(&req->rq_pill, &RMF_NIOBUF_REMOTE),
1233 "want %p - real %p\n", req_capsule_client_get(&req->rq_pill,
1234 &RMF_NIOBUF_REMOTE), (void *)(niobuf - niocount));
1236 osc_announce_cached(cli, &body->oa, opc == OST_WRITE ? requested_nob:0);
1238 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1239 body->oa.o_valid |= OBD_MD_FLFLAGS;
1240 body->oa.o_flags = 0;
1242 body->oa.o_flags |= OBD_FL_RECOV_RESEND;
1245 if (osc_should_shrink_grant(cli))
1246 osc_shrink_grant_local(cli, &body->oa);
1248 /* size[REQ_REC_OFF] still sizeof (*body) */
1249 if (opc == OST_WRITE) {
1250 if (cli->cl_checksum &&
1251 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1252 /* store cl_cksum_type in a local variable since
1253 * it can be changed via lprocfs */
1254 cksum_type_t cksum_type = cli->cl_cksum_type;
1256 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1257 oa->o_flags &= OBD_FL_LOCAL_MASK;
1258 body->oa.o_flags = 0;
1260 body->oa.o_flags |= cksum_type_pack(cksum_type);
1261 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1262 body->oa.o_cksum = osc_checksum_bulk(requested_nob,
1266 CDEBUG(D_PAGE, "checksum at write origin: %x\n",
1268 /* save this in 'oa', too, for later checking */
1269 oa->o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1270 oa->o_flags |= cksum_type_pack(cksum_type);
1272 /* clear out the checksum flag, in case this is a
1273 * resend but cl_checksum is no longer set. b=11238 */
1274 oa->o_valid &= ~OBD_MD_FLCKSUM;
1276 oa->o_cksum = body->oa.o_cksum;
1277 /* 1 RC per niobuf */
1278 req_capsule_set_size(pill, &RMF_RCS, RCL_SERVER,
1279 sizeof(__u32) * niocount);
1281 if (cli->cl_checksum &&
1282 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1283 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0)
1284 body->oa.o_flags = 0;
1285 body->oa.o_flags |= cksum_type_pack(cli->cl_cksum_type);
1286 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1289 ptlrpc_request_set_replen(req);
1291 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1292 aa = ptlrpc_req_async_args(req);
1294 aa->aa_requested_nob = requested_nob;
1295 aa->aa_nio_count = niocount;
1296 aa->aa_page_count = page_count;
1300 INIT_LIST_HEAD(&aa->aa_oaps);
1303 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1304 CDEBUG(D_RPCTRACE, "brw rpc %p - object "DOSTID" offset %lld<>%lld\n",
1305 req, POSTID(&oa->o_oi), niobuf[0].rnb_offset,
1306 niobuf[niocount - 1].rnb_offset + niobuf[niocount - 1].rnb_len);
1310 ptlrpc_req_finished(req);
1315 check_write_checksum(struct obdo *oa, const struct lnet_process_id *peer,
1316 __u32 client_cksum, __u32 server_cksum, int nob,
1317 size_t page_count, struct brw_page **pga,
1318 cksum_type_t client_cksum_type)
1322 cksum_type_t cksum_type;
1324 if (server_cksum == client_cksum) {
1325 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1329 cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
1331 new_cksum = osc_checksum_bulk(nob, page_count, pga, OST_WRITE,
1334 if (cksum_type != client_cksum_type)
1335 msg = "the server did not use the checksum type specified in "
1336 "the original request - likely a protocol problem";
1337 else if (new_cksum == server_cksum)
1338 msg = "changed on the client after we checksummed it - "
1339 "likely false positive due to mmap IO (bug 11742)";
1340 else if (new_cksum == client_cksum)
1341 msg = "changed in transit before arrival at OST";
1343 msg = "changed in transit AND doesn't match the original - "
1344 "likely false positive due to mmap IO (bug 11742)";
1346 LCONSOLE_ERROR_MSG(0x132, "BAD WRITE CHECKSUM: %s: from %s inode "DFID
1347 " object "DOSTID" extent [%llu-%llu]\n",
1348 msg, libcfs_nid2str(peer->nid),
1349 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
1350 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
1351 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
1352 POSTID(&oa->o_oi), pga[0]->off,
1353 pga[page_count-1]->off + pga[page_count-1]->count - 1);
1354 CERROR("original client csum %x (type %x), server csum %x (type %x), "
1355 "client csum now %x\n", client_cksum, client_cksum_type,
1356 server_cksum, cksum_type, new_cksum);
1360 /* Note rc enters this function as number of bytes transferred */
1361 static int osc_brw_fini_request(struct ptlrpc_request *req, int rc)
1363 struct osc_brw_async_args *aa = (void *)&req->rq_async_args;
1364 const struct lnet_process_id *peer =
1365 &req->rq_import->imp_connection->c_peer;
1366 struct client_obd *cli = aa->aa_cli;
1367 struct ost_body *body;
1368 u32 client_cksum = 0;
1371 if (rc < 0 && rc != -EDQUOT) {
1372 DEBUG_REQ(D_INFO, req, "Failed request with rc = %d\n", rc);
1376 LASSERTF(req->rq_repmsg != NULL, "rc = %d\n", rc);
1377 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
1379 DEBUG_REQ(D_INFO, req, "Can't unpack body\n");
1383 /* set/clear over quota flag for a uid/gid */
1384 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE &&
1385 body->oa.o_valid & (OBD_MD_FLUSRQUOTA | OBD_MD_FLGRPQUOTA)) {
1386 unsigned int qid[LL_MAXQUOTAS] =
1387 {body->oa.o_uid, body->oa.o_gid};
1389 CDEBUG(D_QUOTA, "setdq for [%u %u] with valid %#llx, flags %x\n",
1390 body->oa.o_uid, body->oa.o_gid, body->oa.o_valid,
1392 osc_quota_setdq(cli, qid, body->oa.o_valid, body->oa.o_flags);
1395 osc_update_grant(cli, body);
1400 if (aa->aa_oa->o_valid & OBD_MD_FLCKSUM)
1401 client_cksum = aa->aa_oa->o_cksum; /* save for later */
1403 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1405 CERROR("Unexpected +ve rc %d\n", rc);
1408 LASSERT(req->rq_bulk->bd_nob == aa->aa_requested_nob);
1410 if (sptlrpc_cli_unwrap_bulk_write(req, req->rq_bulk))
1413 if ((aa->aa_oa->o_valid & OBD_MD_FLCKSUM) && client_cksum &&
1414 check_write_checksum(&body->oa, peer, client_cksum,
1415 body->oa.o_cksum, aa->aa_requested_nob,
1416 aa->aa_page_count, aa->aa_ppga,
1417 cksum_type_unpack(aa->aa_oa->o_flags)))
1420 rc = check_write_rcs(req, aa->aa_requested_nob,aa->aa_nio_count,
1421 aa->aa_page_count, aa->aa_ppga);
1425 /* The rest of this function executes only for OST_READs */
1427 /* if unwrap_bulk failed, return -EAGAIN to retry */
1428 rc = sptlrpc_cli_unwrap_bulk_read(req, req->rq_bulk, rc);
1430 GOTO(out, rc = -EAGAIN);
1432 if (rc > aa->aa_requested_nob) {
1433 CERROR("Unexpected rc %d (%d requested)\n", rc,
1434 aa->aa_requested_nob);
1438 if (rc != req->rq_bulk->bd_nob_transferred) {
1439 CERROR ("Unexpected rc %d (%d transferred)\n",
1440 rc, req->rq_bulk->bd_nob_transferred);
1444 if (rc < aa->aa_requested_nob)
1445 handle_short_read(rc, aa->aa_page_count, aa->aa_ppga);
1447 if (body->oa.o_valid & OBD_MD_FLCKSUM) {
1448 static int cksum_counter;
1449 u32 server_cksum = body->oa.o_cksum;
1452 cksum_type_t cksum_type;
1454 cksum_type = cksum_type_unpack(body->oa.o_valid &OBD_MD_FLFLAGS?
1455 body->oa.o_flags : 0);
1456 client_cksum = osc_checksum_bulk(rc, aa->aa_page_count,
1457 aa->aa_ppga, OST_READ,
1460 if (peer->nid != req->rq_bulk->bd_sender) {
1462 router = libcfs_nid2str(req->rq_bulk->bd_sender);
1465 if (server_cksum != client_cksum) {
1466 LCONSOLE_ERROR_MSG(0x133, "%s: BAD READ CHECKSUM: from "
1467 "%s%s%s inode "DFID" object "DOSTID
1468 " extent [%llu-%llu]\n",
1469 req->rq_import->imp_obd->obd_name,
1470 libcfs_nid2str(peer->nid),
1472 body->oa.o_valid & OBD_MD_FLFID ?
1473 body->oa.o_parent_seq : (__u64)0,
1474 body->oa.o_valid & OBD_MD_FLFID ?
1475 body->oa.o_parent_oid : 0,
1476 body->oa.o_valid & OBD_MD_FLFID ?
1477 body->oa.o_parent_ver : 0,
1478 POSTID(&body->oa.o_oi),
1479 aa->aa_ppga[0]->off,
1480 aa->aa_ppga[aa->aa_page_count-1]->off +
1481 aa->aa_ppga[aa->aa_page_count-1]->count -
1483 CERROR("client %x, server %x, cksum_type %x\n",
1484 client_cksum, server_cksum, cksum_type);
1486 aa->aa_oa->o_cksum = client_cksum;
1490 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1493 } else if (unlikely(client_cksum)) {
1494 static int cksum_missed;
1497 if ((cksum_missed & (-cksum_missed)) == cksum_missed)
1498 CERROR("Checksum %u requested from %s but not sent\n",
1499 cksum_missed, libcfs_nid2str(peer->nid));
1505 lustre_get_wire_obdo(&req->rq_import->imp_connect_data,
1506 aa->aa_oa, &body->oa);
1511 static int osc_brw_redo_request(struct ptlrpc_request *request,
1512 struct osc_brw_async_args *aa, int rc)
1514 struct ptlrpc_request *new_req;
1515 struct osc_brw_async_args *new_aa;
1516 struct osc_async_page *oap;
1519 DEBUG_REQ(rc == -EINPROGRESS ? D_RPCTRACE : D_ERROR, request,
1520 "redo for recoverable error %d", rc);
1522 rc = osc_brw_prep_request(lustre_msg_get_opc(request->rq_reqmsg) ==
1523 OST_WRITE ? OBD_BRW_WRITE : OBD_BRW_READ,
1524 aa->aa_cli, aa->aa_oa, aa->aa_page_count,
1525 aa->aa_ppga, &new_req, 1);
1529 list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
1530 if (oap->oap_request != NULL) {
1531 LASSERTF(request == oap->oap_request,
1532 "request %p != oap_request %p\n",
1533 request, oap->oap_request);
1534 if (oap->oap_interrupted) {
1535 ptlrpc_req_finished(new_req);
1540 /* New request takes over pga and oaps from old request.
1541 * Note that copying a list_head doesn't work, need to move it... */
1543 new_req->rq_interpret_reply = request->rq_interpret_reply;
1544 new_req->rq_async_args = request->rq_async_args;
1545 new_req->rq_commit_cb = request->rq_commit_cb;
1546 /* cap resend delay to the current request timeout, this is similar to
1547 * what ptlrpc does (see after_reply()) */
1548 if (aa->aa_resends > new_req->rq_timeout)
1549 new_req->rq_sent = cfs_time_current_sec() + new_req->rq_timeout;
1551 new_req->rq_sent = cfs_time_current_sec() + aa->aa_resends;
1552 new_req->rq_generation_set = 1;
1553 new_req->rq_import_generation = request->rq_import_generation;
1555 new_aa = ptlrpc_req_async_args(new_req);
1557 INIT_LIST_HEAD(&new_aa->aa_oaps);
1558 list_splice_init(&aa->aa_oaps, &new_aa->aa_oaps);
1559 INIT_LIST_HEAD(&new_aa->aa_exts);
1560 list_splice_init(&aa->aa_exts, &new_aa->aa_exts);
1561 new_aa->aa_resends = aa->aa_resends;
1563 list_for_each_entry(oap, &new_aa->aa_oaps, oap_rpc_item) {
1564 if (oap->oap_request) {
1565 ptlrpc_req_finished(oap->oap_request);
1566 oap->oap_request = ptlrpc_request_addref(new_req);
1570 /* XXX: This code will run into problem if we're going to support
1571 * to add a series of BRW RPCs into a self-defined ptlrpc_request_set
1572 * and wait for all of them to be finished. We should inherit request
1573 * set from old request. */
1574 ptlrpcd_add_req(new_req);
1576 DEBUG_REQ(D_INFO, new_req, "new request");
1581 * ugh, we want disk allocation on the target to happen in offset order. we'll
1582 * follow sedgewicks advice and stick to the dead simple shellsort -- it'll do
1583 * fine for our small page arrays and doesn't require allocation. its an
1584 * insertion sort that swaps elements that are strides apart, shrinking the
1585 * stride down until its '1' and the array is sorted.
1587 static void sort_brw_pages(struct brw_page **array, int num)
1590 struct brw_page *tmp;
1594 for (stride = 1; stride < num ; stride = (stride * 3) + 1)
1599 for (i = stride ; i < num ; i++) {
1602 while (j >= stride && array[j - stride]->off > tmp->off) {
1603 array[j] = array[j - stride];
1608 } while (stride > 1);
1611 static void osc_release_ppga(struct brw_page **ppga, size_t count)
1613 LASSERT(ppga != NULL);
1614 OBD_FREE(ppga, sizeof(*ppga) * count);
1617 static int brw_interpret(const struct lu_env *env,
1618 struct ptlrpc_request *req, void *data, int rc)
1620 struct osc_brw_async_args *aa = data;
1621 struct osc_extent *ext;
1622 struct osc_extent *tmp;
1623 struct client_obd *cli = aa->aa_cli;
1626 rc = osc_brw_fini_request(req, rc);
1627 CDEBUG(D_INODE, "request %p aa %p rc %d\n", req, aa, rc);
1628 /* When server return -EINPROGRESS, client should always retry
1629 * regardless of the number of times the bulk was resent already. */
1630 if (osc_recoverable_error(rc)) {
1631 if (req->rq_import_generation !=
1632 req->rq_import->imp_generation) {
1633 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1634 ""DOSTID", rc = %d.\n",
1635 req->rq_import->imp_obd->obd_name,
1636 POSTID(&aa->aa_oa->o_oi), rc);
1637 } else if (rc == -EINPROGRESS ||
1638 client_should_resend(aa->aa_resends, aa->aa_cli)) {
1639 rc = osc_brw_redo_request(req, aa, rc);
1641 CERROR("%s: too many resent retries for object: "
1642 "%llu:%llu, rc = %d.\n",
1643 req->rq_import->imp_obd->obd_name,
1644 POSTID(&aa->aa_oa->o_oi), rc);
1649 else if (rc == -EAGAIN || rc == -EINPROGRESS)
1654 struct obdo *oa = aa->aa_oa;
1655 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
1656 unsigned long valid = 0;
1657 struct cl_object *obj;
1658 struct osc_async_page *last;
1660 last = brw_page2oap(aa->aa_ppga[aa->aa_page_count - 1]);
1661 obj = osc2cl(last->oap_obj);
1663 cl_object_attr_lock(obj);
1664 if (oa->o_valid & OBD_MD_FLBLOCKS) {
1665 attr->cat_blocks = oa->o_blocks;
1666 valid |= CAT_BLOCKS;
1668 if (oa->o_valid & OBD_MD_FLMTIME) {
1669 attr->cat_mtime = oa->o_mtime;
1672 if (oa->o_valid & OBD_MD_FLATIME) {
1673 attr->cat_atime = oa->o_atime;
1676 if (oa->o_valid & OBD_MD_FLCTIME) {
1677 attr->cat_ctime = oa->o_ctime;
1681 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1682 struct lov_oinfo *loi = cl2osc(obj)->oo_oinfo;
1683 loff_t last_off = last->oap_count + last->oap_obj_off +
1686 /* Change file size if this is an out of quota or
1687 * direct IO write and it extends the file size */
1688 if (loi->loi_lvb.lvb_size < last_off) {
1689 attr->cat_size = last_off;
1692 /* Extend KMS if it's not a lockless write */
1693 if (loi->loi_kms < last_off &&
1694 oap2osc_page(last)->ops_srvlock == 0) {
1695 attr->cat_kms = last_off;
1701 cl_object_attr_update(env, obj, attr, valid);
1702 cl_object_attr_unlock(obj);
1704 OBDO_FREE(aa->aa_oa);
1706 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE && rc == 0)
1707 osc_inc_unstable_pages(req);
1709 list_for_each_entry_safe(ext, tmp, &aa->aa_exts, oe_link) {
1710 list_del_init(&ext->oe_link);
1711 osc_extent_finish(env, ext, 1, rc);
1713 LASSERT(list_empty(&aa->aa_exts));
1714 LASSERT(list_empty(&aa->aa_oaps));
1716 osc_release_ppga(aa->aa_ppga, aa->aa_page_count);
1717 ptlrpc_lprocfs_brw(req, req->rq_bulk->bd_nob_transferred);
1719 spin_lock(&cli->cl_loi_list_lock);
1720 /* We need to decrement before osc_ap_completion->osc_wake_cache_waiters
1721 * is called so we know whether to go to sync BRWs or wait for more
1722 * RPCs to complete */
1723 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE)
1724 cli->cl_w_in_flight--;
1726 cli->cl_r_in_flight--;
1727 osc_wake_cache_waiters(cli);
1728 spin_unlock(&cli->cl_loi_list_lock);
1730 osc_io_unplug(env, cli, NULL);
1734 static void brw_commit(struct ptlrpc_request *req)
1736 /* If osc_inc_unstable_pages (via osc_extent_finish) races with
1737 * this called via the rq_commit_cb, I need to ensure
1738 * osc_dec_unstable_pages is still called. Otherwise unstable
1739 * pages may be leaked. */
1740 spin_lock(&req->rq_lock);
1741 if (likely(req->rq_unstable)) {
1742 req->rq_unstable = 0;
1743 spin_unlock(&req->rq_lock);
1745 osc_dec_unstable_pages(req);
1747 req->rq_committed = 1;
1748 spin_unlock(&req->rq_lock);
1753 * Build an RPC by the list of extent @ext_list. The caller must ensure
1754 * that the total pages in this list are NOT over max pages per RPC.
1755 * Extents in the list must be in OES_RPC state.
1757 int osc_build_rpc(const struct lu_env *env, struct client_obd *cli,
1758 struct list_head *ext_list, int cmd)
1760 struct ptlrpc_request *req = NULL;
1761 struct osc_extent *ext;
1762 struct brw_page **pga = NULL;
1763 struct osc_brw_async_args *aa = NULL;
1764 struct obdo *oa = NULL;
1765 struct osc_async_page *oap;
1766 struct osc_object *obj = NULL;
1767 struct cl_req_attr *crattr = NULL;
1768 loff_t starting_offset = OBD_OBJECT_EOF;
1769 loff_t ending_offset = 0;
1773 bool soft_sync = false;
1774 bool interrupted = false;
1778 struct list_head rpc_list = LIST_HEAD_INIT(rpc_list);
1779 struct ost_body *body;
1781 LASSERT(!list_empty(ext_list));
1783 /* add pages into rpc_list to build BRW rpc */
1784 list_for_each_entry(ext, ext_list, oe_link) {
1785 LASSERT(ext->oe_state == OES_RPC);
1786 mem_tight |= ext->oe_memalloc;
1787 grant += ext->oe_grants;
1788 page_count += ext->oe_nr_pages;
1793 soft_sync = osc_over_unstable_soft_limit(cli);
1795 mpflag = cfs_memory_pressure_get_and_set();
1797 OBD_ALLOC(pga, sizeof(*pga) * page_count);
1799 GOTO(out, rc = -ENOMEM);
1803 GOTO(out, rc = -ENOMEM);
1806 list_for_each_entry(ext, ext_list, oe_link) {
1807 list_for_each_entry(oap, &ext->oe_pages, oap_pending_item) {
1809 oap->oap_brw_flags |= OBD_BRW_MEMALLOC;
1811 oap->oap_brw_flags |= OBD_BRW_SOFT_SYNC;
1812 pga[i] = &oap->oap_brw_page;
1813 pga[i]->off = oap->oap_obj_off + oap->oap_page_off;
1816 list_add_tail(&oap->oap_rpc_item, &rpc_list);
1817 if (starting_offset == OBD_OBJECT_EOF ||
1818 starting_offset > oap->oap_obj_off)
1819 starting_offset = oap->oap_obj_off;
1821 LASSERT(oap->oap_page_off == 0);
1822 if (ending_offset < oap->oap_obj_off + oap->oap_count)
1823 ending_offset = oap->oap_obj_off +
1826 LASSERT(oap->oap_page_off + oap->oap_count ==
1828 if (oap->oap_interrupted)
1833 /* first page in the list */
1834 oap = list_entry(rpc_list.next, typeof(*oap), oap_rpc_item);
1836 crattr = &osc_env_info(env)->oti_req_attr;
1837 memset(crattr, 0, sizeof(*crattr));
1838 crattr->cra_type = (cmd & OBD_BRW_WRITE) ? CRT_WRITE : CRT_READ;
1839 crattr->cra_flags = ~0ULL;
1840 crattr->cra_page = oap2cl_page(oap);
1841 crattr->cra_oa = oa;
1842 cl_req_attr_set(env, osc2cl(obj), crattr);
1844 if (cmd == OBD_BRW_WRITE)
1845 oa->o_grant_used = grant;
1847 sort_brw_pages(pga, page_count);
1848 rc = osc_brw_prep_request(cmd, cli, oa, page_count, pga, &req, 0);
1850 CERROR("prep_req failed: %d\n", rc);
1854 req->rq_commit_cb = brw_commit;
1855 req->rq_interpret_reply = brw_interpret;
1856 req->rq_memalloc = mem_tight != 0;
1857 oap->oap_request = ptlrpc_request_addref(req);
1858 if (interrupted && !req->rq_intr)
1859 ptlrpc_mark_interrupted(req);
1861 /* Need to update the timestamps after the request is built in case
1862 * we race with setattr (locally or in queue at OST). If OST gets
1863 * later setattr before earlier BRW (as determined by the request xid),
1864 * the OST will not use BRW timestamps. Sadly, there is no obvious
1865 * way to do this in a single call. bug 10150 */
1866 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
1867 crattr->cra_oa = &body->oa;
1868 crattr->cra_flags = OBD_MD_FLMTIME|OBD_MD_FLCTIME|OBD_MD_FLATIME;
1869 cl_req_attr_set(env, osc2cl(obj), crattr);
1870 lustre_msg_set_jobid(req->rq_reqmsg, crattr->cra_jobid);
1872 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1873 aa = ptlrpc_req_async_args(req);
1874 INIT_LIST_HEAD(&aa->aa_oaps);
1875 list_splice_init(&rpc_list, &aa->aa_oaps);
1876 INIT_LIST_HEAD(&aa->aa_exts);
1877 list_splice_init(ext_list, &aa->aa_exts);
1879 spin_lock(&cli->cl_loi_list_lock);
1880 starting_offset >>= PAGE_SHIFT;
1881 if (cmd == OBD_BRW_READ) {
1882 cli->cl_r_in_flight++;
1883 lprocfs_oh_tally_log2(&cli->cl_read_page_hist, page_count);
1884 lprocfs_oh_tally(&cli->cl_read_rpc_hist, cli->cl_r_in_flight);
1885 lprocfs_oh_tally_log2(&cli->cl_read_offset_hist,
1886 starting_offset + 1);
1888 cli->cl_w_in_flight++;
1889 lprocfs_oh_tally_log2(&cli->cl_write_page_hist, page_count);
1890 lprocfs_oh_tally(&cli->cl_write_rpc_hist, cli->cl_w_in_flight);
1891 lprocfs_oh_tally_log2(&cli->cl_write_offset_hist,
1892 starting_offset + 1);
1894 spin_unlock(&cli->cl_loi_list_lock);
1896 DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %ur/%uw in flight",
1897 page_count, aa, cli->cl_r_in_flight,
1898 cli->cl_w_in_flight);
1899 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_DELAY_IO, cfs_fail_val);
1901 ptlrpcd_add_req(req);
1907 cfs_memory_pressure_restore(mpflag);
1910 LASSERT(req == NULL);
1915 OBD_FREE(pga, sizeof(*pga) * page_count);
1916 /* this should happen rarely and is pretty bad, it makes the
1917 * pending list not follow the dirty order */
1918 while (!list_empty(ext_list)) {
1919 ext = list_entry(ext_list->next, struct osc_extent,
1921 list_del_init(&ext->oe_link);
1922 osc_extent_finish(env, ext, 0, rc);
1928 static int osc_set_lock_data(struct ldlm_lock *lock, void *data)
1932 LASSERT(lock != NULL);
1934 lock_res_and_lock(lock);
1936 if (lock->l_ast_data == NULL)
1937 lock->l_ast_data = data;
1938 if (lock->l_ast_data == data)
1941 unlock_res_and_lock(lock);
1946 static int osc_enqueue_fini(struct ptlrpc_request *req,
1947 osc_enqueue_upcall_f upcall, void *cookie,
1948 struct lustre_handle *lockh, enum ldlm_mode mode,
1949 __u64 *flags, int agl, int errcode)
1951 bool intent = *flags & LDLM_FL_HAS_INTENT;
1955 /* The request was created before ldlm_cli_enqueue call. */
1956 if (intent && errcode == ELDLM_LOCK_ABORTED) {
1957 struct ldlm_reply *rep;
1959 rep = req_capsule_server_get(&req->rq_pill, &RMF_DLM_REP);
1960 LASSERT(rep != NULL);
1962 rep->lock_policy_res1 =
1963 ptlrpc_status_ntoh(rep->lock_policy_res1);
1964 if (rep->lock_policy_res1)
1965 errcode = rep->lock_policy_res1;
1967 *flags |= LDLM_FL_LVB_READY;
1968 } else if (errcode == ELDLM_OK) {
1969 *flags |= LDLM_FL_LVB_READY;
1972 /* Call the update callback. */
1973 rc = (*upcall)(cookie, lockh, errcode);
1975 /* release the reference taken in ldlm_cli_enqueue() */
1976 if (errcode == ELDLM_LOCK_MATCHED)
1978 if (errcode == ELDLM_OK && lustre_handle_is_used(lockh))
1979 ldlm_lock_decref(lockh, mode);
1984 static int osc_enqueue_interpret(const struct lu_env *env,
1985 struct ptlrpc_request *req,
1986 struct osc_enqueue_args *aa, int rc)
1988 struct ldlm_lock *lock;
1989 struct lustre_handle *lockh = &aa->oa_lockh;
1990 enum ldlm_mode mode = aa->oa_mode;
1991 struct ost_lvb *lvb = aa->oa_lvb;
1992 __u32 lvb_len = sizeof(*lvb);
1997 /* ldlm_cli_enqueue is holding a reference on the lock, so it must
1999 lock = ldlm_handle2lock(lockh);
2000 LASSERTF(lock != NULL,
2001 "lockh %#llx, req %p, aa %p - client evicted?\n",
2002 lockh->cookie, req, aa);
2004 /* Take an additional reference so that a blocking AST that
2005 * ldlm_cli_enqueue_fini() might post for a failed lock, is guaranteed
2006 * to arrive after an upcall has been executed by
2007 * osc_enqueue_fini(). */
2008 ldlm_lock_addref(lockh, mode);
2010 /* Let cl_lock_state_wait fail with -ERESTARTSYS to unuse sublocks. */
2011 OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_ENQUEUE_HANG, 2);
2013 /* Let CP AST to grant the lock first. */
2014 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_ENQ_RACE, 1);
2017 LASSERT(aa->oa_lvb == NULL);
2018 LASSERT(aa->oa_flags == NULL);
2019 aa->oa_flags = &flags;
2022 /* Complete obtaining the lock procedure. */
2023 rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, aa->oa_type, 1,
2024 aa->oa_mode, aa->oa_flags, lvb, lvb_len,
2026 /* Complete osc stuff. */
2027 rc = osc_enqueue_fini(req, aa->oa_upcall, aa->oa_cookie, lockh, mode,
2028 aa->oa_flags, aa->oa_agl, rc);
2030 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_CANCEL_RACE, 10);
2032 ldlm_lock_decref(lockh, mode);
2033 LDLM_LOCK_PUT(lock);
2037 struct ptlrpc_request_set *PTLRPCD_SET = (void *)1;
2039 /* When enqueuing asynchronously, locks are not ordered, we can obtain a lock
2040 * from the 2nd OSC before a lock from the 1st one. This does not deadlock with
2041 * other synchronous requests, however keeping some locks and trying to obtain
2042 * others may take a considerable amount of time in a case of ost failure; and
2043 * when other sync requests do not get released lock from a client, the client
2044 * is evicted from the cluster -- such scenarious make the life difficult, so
2045 * release locks just after they are obtained. */
2046 int osc_enqueue_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2047 __u64 *flags, union ldlm_policy_data *policy,
2048 struct ost_lvb *lvb, int kms_valid,
2049 osc_enqueue_upcall_f upcall, void *cookie,
2050 struct ldlm_enqueue_info *einfo,
2051 struct ptlrpc_request_set *rqset, int async, int agl)
2053 struct obd_device *obd = exp->exp_obd;
2054 struct lustre_handle lockh = { 0 };
2055 struct ptlrpc_request *req = NULL;
2056 int intent = *flags & LDLM_FL_HAS_INTENT;
2057 __u64 match_flags = *flags;
2058 enum ldlm_mode mode;
2062 /* Filesystem lock extents are extended to page boundaries so that
2063 * dealing with the page cache is a little smoother. */
2064 policy->l_extent.start -= policy->l_extent.start & ~PAGE_MASK;
2065 policy->l_extent.end |= ~PAGE_MASK;
2068 * kms is not valid when either object is completely fresh (so that no
2069 * locks are cached), or object was evicted. In the latter case cached
2070 * lock cannot be used, because it would prime inode state with
2071 * potentially stale LVB.
2076 /* Next, search for already existing extent locks that will cover us */
2077 /* If we're trying to read, we also search for an existing PW lock. The
2078 * VFS and page cache already protect us locally, so lots of readers/
2079 * writers can share a single PW lock.
2081 * There are problems with conversion deadlocks, so instead of
2082 * converting a read lock to a write lock, we'll just enqueue a new
2085 * At some point we should cancel the read lock instead of making them
2086 * send us a blocking callback, but there are problems with canceling
2087 * locks out from other users right now, too. */
2088 mode = einfo->ei_mode;
2089 if (einfo->ei_mode == LCK_PR)
2092 match_flags |= LDLM_FL_LVB_READY;
2094 match_flags |= LDLM_FL_BLOCK_GRANTED;
2095 mode = ldlm_lock_match(obd->obd_namespace, match_flags, res_id,
2096 einfo->ei_type, policy, mode, &lockh, 0);
2098 struct ldlm_lock *matched;
2100 if (*flags & LDLM_FL_TEST_LOCK)
2103 matched = ldlm_handle2lock(&lockh);
2105 /* AGL enqueues DLM locks speculatively. Therefore if
2106 * it already exists a DLM lock, it wll just inform the
2107 * caller to cancel the AGL process for this stripe. */
2108 ldlm_lock_decref(&lockh, mode);
2109 LDLM_LOCK_PUT(matched);
2111 } else if (osc_set_lock_data(matched, einfo->ei_cbdata)) {
2112 *flags |= LDLM_FL_LVB_READY;
2114 /* We already have a lock, and it's referenced. */
2115 (*upcall)(cookie, &lockh, ELDLM_LOCK_MATCHED);
2117 ldlm_lock_decref(&lockh, mode);
2118 LDLM_LOCK_PUT(matched);
2121 ldlm_lock_decref(&lockh, mode);
2122 LDLM_LOCK_PUT(matched);
2127 if (*flags & (LDLM_FL_TEST_LOCK | LDLM_FL_MATCH_LOCK))
2131 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2132 &RQF_LDLM_ENQUEUE_LVB);
2136 rc = ldlm_prep_enqueue_req(exp, req, NULL, 0);
2138 ptlrpc_request_free(req);
2142 req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER,
2144 ptlrpc_request_set_replen(req);
2147 /* users of osc_enqueue() can pass this flag for ldlm_lock_match() */
2148 *flags &= ~LDLM_FL_BLOCK_GRANTED;
2150 rc = ldlm_cli_enqueue(exp, &req, einfo, res_id, policy, flags, lvb,
2151 sizeof(*lvb), LVB_T_OST, &lockh, async);
2154 struct osc_enqueue_args *aa;
2155 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2156 aa = ptlrpc_req_async_args(req);
2158 aa->oa_mode = einfo->ei_mode;
2159 aa->oa_type = einfo->ei_type;
2160 lustre_handle_copy(&aa->oa_lockh, &lockh);
2161 aa->oa_upcall = upcall;
2162 aa->oa_cookie = cookie;
2165 aa->oa_flags = flags;
2168 /* AGL is essentially to enqueue an DLM lock
2169 * in advance, so we don't care about the
2170 * result of AGL enqueue. */
2172 aa->oa_flags = NULL;
2175 req->rq_interpret_reply =
2176 (ptlrpc_interpterer_t)osc_enqueue_interpret;
2177 if (rqset == PTLRPCD_SET)
2178 ptlrpcd_add_req(req);
2180 ptlrpc_set_add_req(rqset, req);
2181 } else if (intent) {
2182 ptlrpc_req_finished(req);
2187 rc = osc_enqueue_fini(req, upcall, cookie, &lockh, einfo->ei_mode,
2190 ptlrpc_req_finished(req);
2195 int osc_match_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2196 enum ldlm_type type, union ldlm_policy_data *policy,
2197 enum ldlm_mode mode, __u64 *flags, void *data,
2198 struct lustre_handle *lockh, int unref)
2200 struct obd_device *obd = exp->exp_obd;
2201 __u64 lflags = *flags;
2205 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_MATCH))
2208 /* Filesystem lock extents are extended to page boundaries so that
2209 * dealing with the page cache is a little smoother */
2210 policy->l_extent.start -= policy->l_extent.start & ~PAGE_MASK;
2211 policy->l_extent.end |= ~PAGE_MASK;
2213 /* Next, search for already existing extent locks that will cover us */
2214 /* If we're trying to read, we also search for an existing PW lock. The
2215 * VFS and page cache already protect us locally, so lots of readers/
2216 * writers can share a single PW lock. */
2220 rc = ldlm_lock_match(obd->obd_namespace, lflags,
2221 res_id, type, policy, rc, lockh, unref);
2222 if (rc == 0 || lflags & LDLM_FL_TEST_LOCK)
2226 struct ldlm_lock *lock = ldlm_handle2lock(lockh);
2228 LASSERT(lock != NULL);
2229 if (!osc_set_lock_data(lock, data)) {
2230 ldlm_lock_decref(lockh, rc);
2233 LDLM_LOCK_PUT(lock);
2238 static int osc_statfs_interpret(const struct lu_env *env,
2239 struct ptlrpc_request *req,
2240 struct osc_async_args *aa, int rc)
2242 struct obd_statfs *msfs;
2246 /* The request has in fact never been sent
2247 * due to issues at a higher level (LOV).
2248 * Exit immediately since the caller is
2249 * aware of the problem and takes care
2250 * of the clean up */
2253 if ((rc == -ENOTCONN || rc == -EAGAIN) &&
2254 (aa->aa_oi->oi_flags & OBD_STATFS_NODELAY))
2260 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2262 GOTO(out, rc = -EPROTO);
2265 *aa->aa_oi->oi_osfs = *msfs;
2267 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
2271 static int osc_statfs_async(struct obd_export *exp,
2272 struct obd_info *oinfo, __u64 max_age,
2273 struct ptlrpc_request_set *rqset)
2275 struct obd_device *obd = class_exp2obd(exp);
2276 struct ptlrpc_request *req;
2277 struct osc_async_args *aa;
2281 /* We could possibly pass max_age in the request (as an absolute
2282 * timestamp or a "seconds.usec ago") so the target can avoid doing
2283 * extra calls into the filesystem if that isn't necessary (e.g.
2284 * during mount that would help a bit). Having relative timestamps
2285 * is not so great if request processing is slow, while absolute
2286 * timestamps are not ideal because they need time synchronization. */
2287 req = ptlrpc_request_alloc(obd->u.cli.cl_import, &RQF_OST_STATFS);
2291 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2293 ptlrpc_request_free(req);
2296 ptlrpc_request_set_replen(req);
2297 req->rq_request_portal = OST_CREATE_PORTAL;
2298 ptlrpc_at_set_req_timeout(req);
2300 if (oinfo->oi_flags & OBD_STATFS_NODELAY) {
2301 /* procfs requests not want stat in wait for avoid deadlock */
2302 req->rq_no_resend = 1;
2303 req->rq_no_delay = 1;
2306 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_statfs_interpret;
2307 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2308 aa = ptlrpc_req_async_args(req);
2311 ptlrpc_set_add_req(rqset, req);
2315 static int osc_statfs(const struct lu_env *env, struct obd_export *exp,
2316 struct obd_statfs *osfs, __u64 max_age, __u32 flags)
2318 struct obd_device *obd = class_exp2obd(exp);
2319 struct obd_statfs *msfs;
2320 struct ptlrpc_request *req;
2321 struct obd_import *imp = NULL;
2325 /*Since the request might also come from lprocfs, so we need
2326 *sync this with client_disconnect_export Bug15684*/
2327 down_read(&obd->u.cli.cl_sem);
2328 if (obd->u.cli.cl_import)
2329 imp = class_import_get(obd->u.cli.cl_import);
2330 up_read(&obd->u.cli.cl_sem);
2334 /* We could possibly pass max_age in the request (as an absolute
2335 * timestamp or a "seconds.usec ago") so the target can avoid doing
2336 * extra calls into the filesystem if that isn't necessary (e.g.
2337 * during mount that would help a bit). Having relative timestamps
2338 * is not so great if request processing is slow, while absolute
2339 * timestamps are not ideal because they need time synchronization. */
2340 req = ptlrpc_request_alloc(imp, &RQF_OST_STATFS);
2342 class_import_put(imp);
2347 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2349 ptlrpc_request_free(req);
2352 ptlrpc_request_set_replen(req);
2353 req->rq_request_portal = OST_CREATE_PORTAL;
2354 ptlrpc_at_set_req_timeout(req);
2356 if (flags & OBD_STATFS_NODELAY) {
2357 /* procfs requests not want stat in wait for avoid deadlock */
2358 req->rq_no_resend = 1;
2359 req->rq_no_delay = 1;
2362 rc = ptlrpc_queue_wait(req);
2366 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2368 GOTO(out, rc = -EPROTO);
2375 ptlrpc_req_finished(req);
2379 static int osc_iocontrol(unsigned int cmd, struct obd_export *exp, int len,
2380 void *karg, void __user *uarg)
2382 struct obd_device *obd = exp->exp_obd;
2383 struct obd_ioctl_data *data = karg;
2387 if (!try_module_get(THIS_MODULE)) {
2388 CERROR("%s: cannot get module '%s'\n", obd->obd_name,
2389 module_name(THIS_MODULE));
2393 case OBD_IOC_CLIENT_RECOVER:
2394 err = ptlrpc_recover_import(obd->u.cli.cl_import,
2395 data->ioc_inlbuf1, 0);
2399 case IOC_OSC_SET_ACTIVE:
2400 err = ptlrpc_set_import_active(obd->u.cli.cl_import,
2403 case OBD_IOC_PING_TARGET:
2404 err = ptlrpc_obd_ping(obd);
2407 CDEBUG(D_INODE, "unrecognised ioctl %#x by %s\n",
2408 cmd, current_comm());
2409 GOTO(out, err = -ENOTTY);
2412 module_put(THIS_MODULE);
2416 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
2417 u32 keylen, void *key,
2418 u32 vallen, void *val,
2419 struct ptlrpc_request_set *set)
2421 struct ptlrpc_request *req;
2422 struct obd_device *obd = exp->exp_obd;
2423 struct obd_import *imp = class_exp2cliimp(exp);
2428 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_SHUTDOWN, 10);
2430 if (KEY_IS(KEY_CHECKSUM)) {
2431 if (vallen != sizeof(int))
2433 exp->exp_obd->u.cli.cl_checksum = (*(int *)val) ? 1 : 0;
2437 if (KEY_IS(KEY_SPTLRPC_CONF)) {
2438 sptlrpc_conf_client_adapt(obd);
2442 if (KEY_IS(KEY_FLUSH_CTX)) {
2443 sptlrpc_import_flush_my_ctx(imp);
2447 if (KEY_IS(KEY_CACHE_SET)) {
2448 struct client_obd *cli = &obd->u.cli;
2450 LASSERT(cli->cl_cache == NULL); /* only once */
2451 cli->cl_cache = (struct cl_client_cache *)val;
2452 cl_cache_incref(cli->cl_cache);
2453 cli->cl_lru_left = &cli->cl_cache->ccc_lru_left;
2455 /* add this osc into entity list */
2456 LASSERT(list_empty(&cli->cl_lru_osc));
2457 spin_lock(&cli->cl_cache->ccc_lru_lock);
2458 list_add(&cli->cl_lru_osc, &cli->cl_cache->ccc_lru);
2459 spin_unlock(&cli->cl_cache->ccc_lru_lock);
2464 if (KEY_IS(KEY_CACHE_LRU_SHRINK)) {
2465 struct client_obd *cli = &obd->u.cli;
2466 long nr = atomic_long_read(&cli->cl_lru_in_list) >> 1;
2467 long target = *(long *)val;
2469 nr = osc_lru_shrink(env, cli, min(nr, target), true);
2474 if (!set && !KEY_IS(KEY_GRANT_SHRINK))
2477 /* We pass all other commands directly to OST. Since nobody calls osc
2478 methods directly and everybody is supposed to go through LOV, we
2479 assume lov checked invalid values for us.
2480 The only recognised values so far are evict_by_nid and mds_conn.
2481 Even if something bad goes through, we'd get a -EINVAL from OST
2484 req = ptlrpc_request_alloc(imp, KEY_IS(KEY_GRANT_SHRINK) ?
2485 &RQF_OST_SET_GRANT_INFO :
2490 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
2491 RCL_CLIENT, keylen);
2492 if (!KEY_IS(KEY_GRANT_SHRINK))
2493 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_VAL,
2494 RCL_CLIENT, vallen);
2495 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SET_INFO);
2497 ptlrpc_request_free(req);
2501 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
2502 memcpy(tmp, key, keylen);
2503 tmp = req_capsule_client_get(&req->rq_pill, KEY_IS(KEY_GRANT_SHRINK) ?
2506 memcpy(tmp, val, vallen);
2508 if (KEY_IS(KEY_GRANT_SHRINK)) {
2509 struct osc_grant_args *aa;
2512 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2513 aa = ptlrpc_req_async_args(req);
2516 ptlrpc_req_finished(req);
2519 *oa = ((struct ost_body *)val)->oa;
2521 req->rq_interpret_reply = osc_shrink_grant_interpret;
2524 ptlrpc_request_set_replen(req);
2525 if (!KEY_IS(KEY_GRANT_SHRINK)) {
2526 LASSERT(set != NULL);
2527 ptlrpc_set_add_req(set, req);
2528 ptlrpc_check_set(NULL, set);
2530 ptlrpcd_add_req(req);
2536 static int osc_reconnect(const struct lu_env *env,
2537 struct obd_export *exp, struct obd_device *obd,
2538 struct obd_uuid *cluuid,
2539 struct obd_connect_data *data,
2542 struct client_obd *cli = &obd->u.cli;
2544 if (data != NULL && (data->ocd_connect_flags & OBD_CONNECT_GRANT)) {
2548 spin_lock(&cli->cl_loi_list_lock);
2549 grant = cli->cl_avail_grant + cli->cl_reserved_grant;
2550 if (data->ocd_connect_flags & OBD_CONNECT_GRANT_PARAM)
2551 grant += cli->cl_dirty_grant;
2553 grant += cli->cl_dirty_pages << PAGE_SHIFT;
2554 data->ocd_grant = grant ? : 2 * cli_brw_size(obd);
2555 lost_grant = cli->cl_lost_grant;
2556 cli->cl_lost_grant = 0;
2557 spin_unlock(&cli->cl_loi_list_lock);
2559 CDEBUG(D_RPCTRACE, "ocd_connect_flags: %#llx ocd_version: %d"
2560 " ocd_grant: %d, lost: %ld.\n", data->ocd_connect_flags,
2561 data->ocd_version, data->ocd_grant, lost_grant);
2567 static int osc_disconnect(struct obd_export *exp)
2569 struct obd_device *obd = class_exp2obd(exp);
2572 rc = client_disconnect_export(exp);
2574 * Initially we put del_shrink_grant before disconnect_export, but it
2575 * causes the following problem if setup (connect) and cleanup
2576 * (disconnect) are tangled together.
2577 * connect p1 disconnect p2
2578 * ptlrpc_connect_import
2579 * ............... class_manual_cleanup
2582 * ptlrpc_connect_interrupt
2584 * add this client to shrink list
2586 * Bang! pinger trigger the shrink.
2587 * So the osc should be disconnected from the shrink list, after we
2588 * are sure the import has been destroyed. BUG18662
2590 if (obd->u.cli.cl_import == NULL)
2591 osc_del_shrink_grant(&obd->u.cli);
2595 static int osc_ldlm_resource_invalidate(struct cfs_hash *hs,
2596 struct cfs_hash_bd *bd, struct hlist_node *hnode, void *arg)
2598 struct lu_env *env = arg;
2599 struct ldlm_resource *res = cfs_hash_object(hs, hnode);
2600 struct ldlm_lock *lock;
2601 struct osc_object *osc = NULL;
2605 list_for_each_entry(lock, &res->lr_granted, l_res_link) {
2606 if (lock->l_ast_data != NULL && osc == NULL) {
2607 osc = lock->l_ast_data;
2608 cl_object_get(osc2cl(osc));
2611 /* clear LDLM_FL_CLEANED flag to make sure it will be canceled
2612 * by the 2nd round of ldlm_namespace_clean() call in
2613 * osc_import_event(). */
2614 ldlm_clear_cleaned(lock);
2619 osc_object_invalidate(env, osc);
2620 cl_object_put(env, osc2cl(osc));
2626 static int osc_import_event(struct obd_device *obd,
2627 struct obd_import *imp,
2628 enum obd_import_event event)
2630 struct client_obd *cli;
2634 LASSERT(imp->imp_obd == obd);
2637 case IMP_EVENT_DISCON: {
2639 spin_lock(&cli->cl_loi_list_lock);
2640 cli->cl_avail_grant = 0;
2641 cli->cl_lost_grant = 0;
2642 spin_unlock(&cli->cl_loi_list_lock);
2645 case IMP_EVENT_INACTIVE: {
2646 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_INACTIVE, NULL);
2649 case IMP_EVENT_INVALIDATE: {
2650 struct ldlm_namespace *ns = obd->obd_namespace;
2654 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
2656 env = cl_env_get(&refcheck);
2658 osc_io_unplug(env, &obd->u.cli, NULL);
2660 cfs_hash_for_each_nolock(ns->ns_rs_hash,
2661 osc_ldlm_resource_invalidate,
2663 cl_env_put(env, &refcheck);
2665 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
2670 case IMP_EVENT_ACTIVE: {
2671 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVE, NULL);
2674 case IMP_EVENT_OCD: {
2675 struct obd_connect_data *ocd = &imp->imp_connect_data;
2677 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT)
2678 osc_init_grant(&obd->u.cli, ocd);
2681 if (ocd->ocd_connect_flags & OBD_CONNECT_REQPORTAL)
2682 imp->imp_client->cli_request_portal =OST_REQUEST_PORTAL;
2684 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_OCD, NULL);
2687 case IMP_EVENT_DEACTIVATE: {
2688 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_DEACTIVATE, NULL);
2691 case IMP_EVENT_ACTIVATE: {
2692 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVATE, NULL);
2696 CERROR("Unknown import event %d\n", event);
2703 * Determine whether the lock can be canceled before replaying the lock
2704 * during recovery, see bug16774 for detailed information.
2706 * \retval zero the lock can't be canceled
2707 * \retval other ok to cancel
2709 static int osc_cancel_weight(struct ldlm_lock *lock)
2712 * Cancel all unused and granted extent lock.
2714 if (lock->l_resource->lr_type == LDLM_EXTENT &&
2715 lock->l_granted_mode == lock->l_req_mode &&
2716 osc_ldlm_weigh_ast(lock) == 0)
2722 static int brw_queue_work(const struct lu_env *env, void *data)
2724 struct client_obd *cli = data;
2726 CDEBUG(D_CACHE, "Run writeback work for client obd %p.\n", cli);
2728 osc_io_unplug(env, cli, NULL);
2732 int osc_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
2734 struct client_obd *cli = &obd->u.cli;
2735 struct obd_type *type;
2743 rc = ptlrpcd_addref();
2747 rc = client_obd_setup(obd, lcfg);
2749 GOTO(out_ptlrpcd, rc);
2751 handler = ptlrpcd_alloc_work(cli->cl_import, brw_queue_work, cli);
2752 if (IS_ERR(handler))
2753 GOTO(out_client_setup, rc = PTR_ERR(handler));
2754 cli->cl_writeback_work = handler;
2756 handler = ptlrpcd_alloc_work(cli->cl_import, lru_queue_work, cli);
2757 if (IS_ERR(handler))
2758 GOTO(out_ptlrpcd_work, rc = PTR_ERR(handler));
2759 cli->cl_lru_work = handler;
2761 rc = osc_quota_setup(obd);
2763 GOTO(out_ptlrpcd_work, rc);
2765 cli->cl_grant_shrink_interval = GRANT_SHRINK_INTERVAL;
2767 #ifdef CONFIG_PROC_FS
2768 obd->obd_vars = lprocfs_osc_obd_vars;
2770 /* If this is true then both client (osc) and server (osp) are on the
2771 * same node. The osp layer if loaded first will register the osc proc
2772 * directory. In that case this obd_device will be attached its proc
2773 * tree to type->typ_procsym instead of obd->obd_type->typ_procroot. */
2774 type = class_search_type(LUSTRE_OSP_NAME);
2775 if (type && type->typ_procsym) {
2776 obd->obd_proc_entry = lprocfs_register(obd->obd_name,
2778 obd->obd_vars, obd);
2779 if (IS_ERR(obd->obd_proc_entry)) {
2780 rc = PTR_ERR(obd->obd_proc_entry);
2781 CERROR("error %d setting up lprocfs for %s\n", rc,
2783 obd->obd_proc_entry = NULL;
2786 rc = lprocfs_obd_setup(obd);
2789 /* If the basic OSC proc tree construction succeeded then
2790 * lets do the rest. */
2792 lproc_osc_attach_seqstat(obd);
2793 sptlrpc_lprocfs_cliobd_attach(obd);
2794 ptlrpc_lprocfs_register_obd(obd);
2798 * We try to control the total number of requests with a upper limit
2799 * osc_reqpool_maxreqcount. There might be some race which will cause
2800 * over-limit allocation, but it is fine.
2802 req_count = atomic_read(&osc_pool_req_count);
2803 if (req_count < osc_reqpool_maxreqcount) {
2804 adding = cli->cl_max_rpcs_in_flight + 2;
2805 if (req_count + adding > osc_reqpool_maxreqcount)
2806 adding = osc_reqpool_maxreqcount - req_count;
2808 added = ptlrpc_add_rqs_to_pool(osc_rq_pool, adding);
2809 atomic_add(added, &osc_pool_req_count);
2812 INIT_LIST_HEAD(&cli->cl_grant_shrink_list);
2813 ns_register_cancel(obd->obd_namespace, osc_cancel_weight);
2815 spin_lock(&osc_shrink_lock);
2816 list_add_tail(&cli->cl_shrink_list, &osc_shrink_list);
2817 spin_unlock(&osc_shrink_lock);
2822 if (cli->cl_writeback_work != NULL) {
2823 ptlrpcd_destroy_work(cli->cl_writeback_work);
2824 cli->cl_writeback_work = NULL;
2826 if (cli->cl_lru_work != NULL) {
2827 ptlrpcd_destroy_work(cli->cl_lru_work);
2828 cli->cl_lru_work = NULL;
2831 client_obd_cleanup(obd);
2837 static int osc_precleanup(struct obd_device *obd)
2839 struct client_obd *cli = &obd->u.cli;
2843 * for echo client, export may be on zombie list, wait for
2844 * zombie thread to cull it, because cli.cl_import will be
2845 * cleared in client_disconnect_export():
2846 * class_export_destroy() -> obd_cleanup() ->
2847 * echo_device_free() -> echo_client_cleanup() ->
2848 * obd_disconnect() -> osc_disconnect() ->
2849 * client_disconnect_export()
2851 obd_zombie_barrier();
2852 if (cli->cl_writeback_work) {
2853 ptlrpcd_destroy_work(cli->cl_writeback_work);
2854 cli->cl_writeback_work = NULL;
2857 if (cli->cl_lru_work) {
2858 ptlrpcd_destroy_work(cli->cl_lru_work);
2859 cli->cl_lru_work = NULL;
2862 obd_cleanup_client_import(obd);
2863 ptlrpc_lprocfs_unregister_obd(obd);
2864 lprocfs_obd_cleanup(obd);
2868 int osc_cleanup(struct obd_device *obd)
2870 struct client_obd *cli = &obd->u.cli;
2875 spin_lock(&osc_shrink_lock);
2876 list_del(&cli->cl_shrink_list);
2877 spin_unlock(&osc_shrink_lock);
2880 if (cli->cl_cache != NULL) {
2881 LASSERT(atomic_read(&cli->cl_cache->ccc_users) > 0);
2882 spin_lock(&cli->cl_cache->ccc_lru_lock);
2883 list_del_init(&cli->cl_lru_osc);
2884 spin_unlock(&cli->cl_cache->ccc_lru_lock);
2885 cli->cl_lru_left = NULL;
2886 cl_cache_decref(cli->cl_cache);
2887 cli->cl_cache = NULL;
2890 /* free memory of osc quota cache */
2891 osc_quota_cleanup(obd);
2893 rc = client_obd_cleanup(obd);
2899 int osc_process_config_base(struct obd_device *obd, struct lustre_cfg *lcfg)
2901 int rc = class_process_proc_param(PARAM_OSC, obd->obd_vars, lcfg, obd);
2902 return rc > 0 ? 0: rc;
2905 static int osc_process_config(struct obd_device *obd, size_t len, void *buf)
2907 return osc_process_config_base(obd, buf);
2910 static struct obd_ops osc_obd_ops = {
2911 .o_owner = THIS_MODULE,
2912 .o_setup = osc_setup,
2913 .o_precleanup = osc_precleanup,
2914 .o_cleanup = osc_cleanup,
2915 .o_add_conn = client_import_add_conn,
2916 .o_del_conn = client_import_del_conn,
2917 .o_connect = client_connect_import,
2918 .o_reconnect = osc_reconnect,
2919 .o_disconnect = osc_disconnect,
2920 .o_statfs = osc_statfs,
2921 .o_statfs_async = osc_statfs_async,
2922 .o_create = osc_create,
2923 .o_destroy = osc_destroy,
2924 .o_getattr = osc_getattr,
2925 .o_setattr = osc_setattr,
2926 .o_iocontrol = osc_iocontrol,
2927 .o_set_info_async = osc_set_info_async,
2928 .o_import_event = osc_import_event,
2929 .o_process_config = osc_process_config,
2930 .o_quotactl = osc_quotactl,
2933 static struct shrinker *osc_cache_shrinker;
2934 struct list_head osc_shrink_list = LIST_HEAD_INIT(osc_shrink_list);
2935 DEFINE_SPINLOCK(osc_shrink_lock);
2937 #ifndef HAVE_SHRINKER_COUNT
2938 static int osc_cache_shrink(SHRINKER_ARGS(sc, nr_to_scan, gfp_mask))
2940 struct shrink_control scv = {
2941 .nr_to_scan = shrink_param(sc, nr_to_scan),
2942 .gfp_mask = shrink_param(sc, gfp_mask)
2944 #if !defined(HAVE_SHRINKER_WANT_SHRINK_PTR) && !defined(HAVE_SHRINK_CONTROL)
2945 struct shrinker *shrinker = NULL;
2948 (void)osc_cache_shrink_scan(shrinker, &scv);
2950 return osc_cache_shrink_count(shrinker, &scv);
2954 static int __init osc_init(void)
2956 bool enable_proc = true;
2957 struct obd_type *type;
2958 unsigned int reqpool_size;
2959 unsigned int reqsize;
2961 DEF_SHRINKER_VAR(osc_shvar, osc_cache_shrink,
2962 osc_cache_shrink_count, osc_cache_shrink_scan);
2965 /* print an address of _any_ initialized kernel symbol from this
2966 * module, to allow debugging with gdb that doesn't support data
2967 * symbols from modules.*/
2968 CDEBUG(D_INFO, "Lustre OSC module (%p).\n", &osc_caches);
2970 rc = lu_kmem_init(osc_caches);
2974 type = class_search_type(LUSTRE_OSP_NAME);
2975 if (type != NULL && type->typ_procsym != NULL)
2976 enable_proc = false;
2978 rc = class_register_type(&osc_obd_ops, NULL, enable_proc, NULL,
2979 LUSTRE_OSC_NAME, &osc_device_type);
2983 osc_cache_shrinker = set_shrinker(DEFAULT_SEEKS, &osc_shvar);
2985 /* This is obviously too much memory, only prevent overflow here */
2986 if (osc_reqpool_mem_max >= 1 << 12 || osc_reqpool_mem_max == 0)
2987 GOTO(out_type, rc = -EINVAL);
2989 reqpool_size = osc_reqpool_mem_max << 20;
2992 while (reqsize < OST_IO_MAXREQSIZE)
2993 reqsize = reqsize << 1;
2996 * We don't enlarge the request count in OSC pool according to
2997 * cl_max_rpcs_in_flight. The allocation from the pool will only be
2998 * tried after normal allocation failed. So a small OSC pool won't
2999 * cause much performance degression in most of cases.
3001 osc_reqpool_maxreqcount = reqpool_size / reqsize;
3003 atomic_set(&osc_pool_req_count, 0);
3004 osc_rq_pool = ptlrpc_init_rq_pool(0, OST_IO_MAXREQSIZE,
3005 ptlrpc_add_rqs_to_pool);
3007 if (osc_rq_pool != NULL)
3011 class_unregister_type(LUSTRE_OSC_NAME);
3013 lu_kmem_fini(osc_caches);
3018 static void __exit osc_exit(void)
3020 remove_shrinker(osc_cache_shrinker);
3021 class_unregister_type(LUSTRE_OSC_NAME);
3022 lu_kmem_fini(osc_caches);
3023 ptlrpc_free_rq_pool(osc_rq_pool);
3026 MODULE_AUTHOR("OpenSFS, Inc. <http://www.lustre.org/>");
3027 MODULE_DESCRIPTION("Lustre Object Storage Client (OSC)");
3028 MODULE_VERSION(LUSTRE_VERSION_STRING);
3029 MODULE_LICENSE("GPL");
3031 module_init(osc_init);
3032 module_exit(osc_exit);