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 <uapi/linux/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 rc = l_wait_event_exclusive(cli->cl_destroy_waitq,
651 osc_can_send_destroy(cli), &lwi);
653 ptlrpc_req_finished(req);
658 /* Do not wait for response */
659 ptlrpcd_add_req(req);
663 static void osc_announce_cached(struct client_obd *cli, struct obdo *oa,
666 u64 bits = OBD_MD_FLBLOCKS | OBD_MD_FLGRANT;
668 LASSERT(!(oa->o_valid & bits));
671 spin_lock(&cli->cl_loi_list_lock);
672 if (OCD_HAS_FLAG(&cli->cl_import->imp_connect_data, GRANT_PARAM))
673 oa->o_dirty = cli->cl_dirty_grant;
675 oa->o_dirty = cli->cl_dirty_pages << PAGE_SHIFT;
676 if (unlikely(cli->cl_dirty_pages - cli->cl_dirty_transit >
677 cli->cl_dirty_max_pages)) {
678 CERROR("dirty %lu - %lu > dirty_max %lu\n",
679 cli->cl_dirty_pages, cli->cl_dirty_transit,
680 cli->cl_dirty_max_pages);
682 } else if (unlikely(atomic_long_read(&obd_dirty_pages) -
683 atomic_long_read(&obd_dirty_transit_pages) >
684 (long)(obd_max_dirty_pages + 1))) {
685 /* The atomic_read() allowing the atomic_inc() are
686 * not covered by a lock thus they may safely race and trip
687 * this CERROR() unless we add in a small fudge factor (+1). */
688 CERROR("%s: dirty %ld - %ld > system dirty_max %ld\n",
689 cli_name(cli), atomic_long_read(&obd_dirty_pages),
690 atomic_long_read(&obd_dirty_transit_pages),
691 obd_max_dirty_pages);
693 } else if (unlikely(cli->cl_dirty_max_pages - cli->cl_dirty_pages >
695 CERROR("dirty %lu - dirty_max %lu too big???\n",
696 cli->cl_dirty_pages, cli->cl_dirty_max_pages);
699 unsigned long nrpages;
701 nrpages = cli->cl_max_pages_per_rpc;
702 nrpages *= cli->cl_max_rpcs_in_flight + 1;
703 nrpages = max(nrpages, cli->cl_dirty_max_pages);
704 oa->o_undirty = nrpages << PAGE_SHIFT;
705 if (OCD_HAS_FLAG(&cli->cl_import->imp_connect_data,
709 /* take extent tax into account when asking for more
711 nrextents = (nrpages + cli->cl_max_extent_pages - 1) /
712 cli->cl_max_extent_pages;
713 oa->o_undirty += nrextents * cli->cl_grant_extent_tax;
716 oa->o_grant = cli->cl_avail_grant + cli->cl_reserved_grant;
717 oa->o_dropped = cli->cl_lost_grant;
718 cli->cl_lost_grant = 0;
719 spin_unlock(&cli->cl_loi_list_lock);
720 CDEBUG(D_CACHE, "dirty: %llu undirty: %u dropped %u grant: %llu\n",
721 oa->o_dirty, oa->o_undirty, oa->o_dropped, oa->o_grant);
724 void osc_update_next_shrink(struct client_obd *cli)
726 cli->cl_next_shrink_grant =
727 cfs_time_shift(cli->cl_grant_shrink_interval);
728 CDEBUG(D_CACHE, "next time %ld to shrink grant \n",
729 cli->cl_next_shrink_grant);
732 static void __osc_update_grant(struct client_obd *cli, u64 grant)
734 spin_lock(&cli->cl_loi_list_lock);
735 cli->cl_avail_grant += grant;
736 spin_unlock(&cli->cl_loi_list_lock);
739 static void osc_update_grant(struct client_obd *cli, struct ost_body *body)
741 if (body->oa.o_valid & OBD_MD_FLGRANT) {
742 CDEBUG(D_CACHE, "got %llu extra grant\n", body->oa.o_grant);
743 __osc_update_grant(cli, body->oa.o_grant);
747 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
748 u32 keylen, void *key,
749 u32 vallen, void *val,
750 struct ptlrpc_request_set *set);
752 static int osc_shrink_grant_interpret(const struct lu_env *env,
753 struct ptlrpc_request *req,
756 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
757 struct obdo *oa = ((struct osc_grant_args *)aa)->aa_oa;
758 struct ost_body *body;
761 __osc_update_grant(cli, oa->o_grant);
765 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
767 osc_update_grant(cli, body);
773 static void osc_shrink_grant_local(struct client_obd *cli, struct obdo *oa)
775 spin_lock(&cli->cl_loi_list_lock);
776 oa->o_grant = cli->cl_avail_grant / 4;
777 cli->cl_avail_grant -= oa->o_grant;
778 spin_unlock(&cli->cl_loi_list_lock);
779 if (!(oa->o_valid & OBD_MD_FLFLAGS)) {
780 oa->o_valid |= OBD_MD_FLFLAGS;
783 oa->o_flags |= OBD_FL_SHRINK_GRANT;
784 osc_update_next_shrink(cli);
787 /* Shrink the current grant, either from some large amount to enough for a
788 * full set of in-flight RPCs, or if we have already shrunk to that limit
789 * then to enough for a single RPC. This avoids keeping more grant than
790 * needed, and avoids shrinking the grant piecemeal. */
791 static int osc_shrink_grant(struct client_obd *cli)
793 __u64 target_bytes = (cli->cl_max_rpcs_in_flight + 1) *
794 (cli->cl_max_pages_per_rpc << PAGE_SHIFT);
796 spin_lock(&cli->cl_loi_list_lock);
797 if (cli->cl_avail_grant <= target_bytes)
798 target_bytes = cli->cl_max_pages_per_rpc << PAGE_SHIFT;
799 spin_unlock(&cli->cl_loi_list_lock);
801 return osc_shrink_grant_to_target(cli, target_bytes);
804 int osc_shrink_grant_to_target(struct client_obd *cli, __u64 target_bytes)
807 struct ost_body *body;
810 spin_lock(&cli->cl_loi_list_lock);
811 /* Don't shrink if we are already above or below the desired limit
812 * We don't want to shrink below a single RPC, as that will negatively
813 * impact block allocation and long-term performance. */
814 if (target_bytes < cli->cl_max_pages_per_rpc << PAGE_SHIFT)
815 target_bytes = cli->cl_max_pages_per_rpc << PAGE_SHIFT;
817 if (target_bytes >= cli->cl_avail_grant) {
818 spin_unlock(&cli->cl_loi_list_lock);
821 spin_unlock(&cli->cl_loi_list_lock);
827 osc_announce_cached(cli, &body->oa, 0);
829 spin_lock(&cli->cl_loi_list_lock);
830 body->oa.o_grant = cli->cl_avail_grant - target_bytes;
831 cli->cl_avail_grant = target_bytes;
832 spin_unlock(&cli->cl_loi_list_lock);
833 if (!(body->oa.o_valid & OBD_MD_FLFLAGS)) {
834 body->oa.o_valid |= OBD_MD_FLFLAGS;
835 body->oa.o_flags = 0;
837 body->oa.o_flags |= OBD_FL_SHRINK_GRANT;
838 osc_update_next_shrink(cli);
840 rc = osc_set_info_async(NULL, cli->cl_import->imp_obd->obd_self_export,
841 sizeof(KEY_GRANT_SHRINK), KEY_GRANT_SHRINK,
842 sizeof(*body), body, NULL);
844 __osc_update_grant(cli, body->oa.o_grant);
849 static int osc_should_shrink_grant(struct client_obd *client)
851 cfs_time_t time = cfs_time_current();
852 cfs_time_t next_shrink = client->cl_next_shrink_grant;
854 if ((client->cl_import->imp_connect_data.ocd_connect_flags &
855 OBD_CONNECT_GRANT_SHRINK) == 0)
858 if (cfs_time_aftereq(time, next_shrink - 5 * CFS_TICK)) {
859 /* Get the current RPC size directly, instead of going via:
860 * cli_brw_size(obd->u.cli.cl_import->imp_obd->obd_self_export)
861 * Keep comment here so that it can be found by searching. */
862 int brw_size = client->cl_max_pages_per_rpc << PAGE_SHIFT;
864 if (client->cl_import->imp_state == LUSTRE_IMP_FULL &&
865 client->cl_avail_grant > brw_size)
868 osc_update_next_shrink(client);
873 static int osc_grant_shrink_grant_cb(struct timeout_item *item, void *data)
875 struct client_obd *client;
877 list_for_each_entry(client, &item->ti_obd_list, cl_grant_shrink_list) {
878 if (osc_should_shrink_grant(client))
879 osc_shrink_grant(client);
884 static int osc_add_shrink_grant(struct client_obd *client)
888 rc = ptlrpc_add_timeout_client(client->cl_grant_shrink_interval,
890 osc_grant_shrink_grant_cb, NULL,
891 &client->cl_grant_shrink_list);
893 CERROR("add grant client %s error %d\n", cli_name(client), rc);
896 CDEBUG(D_CACHE, "add grant client %s\n", cli_name(client));
897 osc_update_next_shrink(client);
901 static int osc_del_shrink_grant(struct client_obd *client)
903 return ptlrpc_del_timeout_client(&client->cl_grant_shrink_list,
907 static void osc_init_grant(struct client_obd *cli, struct obd_connect_data *ocd)
910 * ocd_grant is the total grant amount we're expect to hold: if we've
911 * been evicted, it's the new avail_grant amount, cl_dirty_pages will
912 * drop to 0 as inflight RPCs fail out; otherwise, it's avail_grant +
915 * race is tolerable here: if we're evicted, but imp_state already
916 * left EVICTED state, then cl_dirty_pages must be 0 already.
918 spin_lock(&cli->cl_loi_list_lock);
919 cli->cl_avail_grant = ocd->ocd_grant;
920 if (cli->cl_import->imp_state != LUSTRE_IMP_EVICTED) {
921 cli->cl_avail_grant -= cli->cl_reserved_grant;
922 if (OCD_HAS_FLAG(ocd, GRANT_PARAM))
923 cli->cl_avail_grant -= cli->cl_dirty_grant;
925 cli->cl_avail_grant -=
926 cli->cl_dirty_pages << PAGE_SHIFT;
929 if (OCD_HAS_FLAG(ocd, GRANT_PARAM)) {
933 /* overhead for each extent insertion */
934 cli->cl_grant_extent_tax = ocd->ocd_grant_tax_kb << 10;
935 /* determine the appropriate chunk size used by osc_extent. */
936 cli->cl_chunkbits = max_t(int, PAGE_SHIFT,
937 ocd->ocd_grant_blkbits);
938 /* max_pages_per_rpc must be chunk aligned */
939 chunk_mask = ~((1 << (cli->cl_chunkbits - PAGE_SHIFT)) - 1);
940 cli->cl_max_pages_per_rpc = (cli->cl_max_pages_per_rpc +
941 ~chunk_mask) & chunk_mask;
942 /* determine maximum extent size, in #pages */
943 size = (u64)ocd->ocd_grant_max_blks << ocd->ocd_grant_blkbits;
944 cli->cl_max_extent_pages = size >> PAGE_SHIFT;
945 if (cli->cl_max_extent_pages == 0)
946 cli->cl_max_extent_pages = 1;
948 cli->cl_grant_extent_tax = 0;
949 cli->cl_chunkbits = PAGE_SHIFT;
950 cli->cl_max_extent_pages = DT_MAX_BRW_PAGES;
952 spin_unlock(&cli->cl_loi_list_lock);
954 CDEBUG(D_CACHE, "%s, setting cl_avail_grant: %ld cl_lost_grant: %ld."
955 "chunk bits: %d cl_max_extent_pages: %d\n",
957 cli->cl_avail_grant, cli->cl_lost_grant, cli->cl_chunkbits,
958 cli->cl_max_extent_pages);
960 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT_SHRINK &&
961 list_empty(&cli->cl_grant_shrink_list))
962 osc_add_shrink_grant(cli);
965 /* We assume that the reason this OSC got a short read is because it read
966 * beyond the end of a stripe file; i.e. lustre is reading a sparse file
967 * via the LOV, and it _knows_ it's reading inside the file, it's just that
968 * this stripe never got written at or beyond this stripe offset yet. */
969 static void handle_short_read(int nob_read, size_t page_count,
970 struct brw_page **pga)
975 /* skip bytes read OK */
976 while (nob_read > 0) {
977 LASSERT (page_count > 0);
979 if (pga[i]->count > nob_read) {
980 /* EOF inside this page */
981 ptr = kmap(pga[i]->pg) +
982 (pga[i]->off & ~PAGE_MASK);
983 memset(ptr + nob_read, 0, pga[i]->count - nob_read);
990 nob_read -= pga[i]->count;
995 /* zero remaining pages */
996 while (page_count-- > 0) {
997 ptr = kmap(pga[i]->pg) + (pga[i]->off & ~PAGE_MASK);
998 memset(ptr, 0, pga[i]->count);
1004 static int check_write_rcs(struct ptlrpc_request *req,
1005 int requested_nob, int niocount,
1006 size_t page_count, struct brw_page **pga)
1011 remote_rcs = req_capsule_server_sized_get(&req->rq_pill, &RMF_RCS,
1012 sizeof(*remote_rcs) *
1014 if (remote_rcs == NULL) {
1015 CDEBUG(D_INFO, "Missing/short RC vector on BRW_WRITE reply\n");
1019 /* return error if any niobuf was in error */
1020 for (i = 0; i < niocount; i++) {
1021 if ((int)remote_rcs[i] < 0)
1022 return(remote_rcs[i]);
1024 if (remote_rcs[i] != 0) {
1025 CDEBUG(D_INFO, "rc[%d] invalid (%d) req %p\n",
1026 i, remote_rcs[i], req);
1031 if (req->rq_bulk->bd_nob_transferred != requested_nob) {
1032 CERROR("Unexpected # bytes transferred: %d (requested %d)\n",
1033 req->rq_bulk->bd_nob_transferred, requested_nob);
1040 static inline int can_merge_pages(struct brw_page *p1, struct brw_page *p2)
1042 if (p1->flag != p2->flag) {
1043 unsigned mask = ~(OBD_BRW_FROM_GRANT | OBD_BRW_NOCACHE |
1044 OBD_BRW_SYNC | OBD_BRW_ASYNC |
1045 OBD_BRW_NOQUOTA | OBD_BRW_SOFT_SYNC);
1047 /* warn if we try to combine flags that we don't know to be
1048 * safe to combine */
1049 if (unlikely((p1->flag & mask) != (p2->flag & mask))) {
1050 CWARN("Saw flags 0x%x and 0x%x in the same brw, please "
1051 "report this at https://jira.hpdd.intel.com/\n",
1052 p1->flag, p2->flag);
1057 return (p1->off + p1->count == p2->off);
1060 static u32 osc_checksum_bulk(int nob, size_t pg_count,
1061 struct brw_page **pga, int opc,
1062 cksum_type_t cksum_type)
1066 struct cfs_crypto_hash_desc *hdesc;
1067 unsigned int bufsize;
1068 unsigned char cfs_alg = cksum_obd2cfs(cksum_type);
1070 LASSERT(pg_count > 0);
1072 hdesc = cfs_crypto_hash_init(cfs_alg, NULL, 0);
1073 if (IS_ERR(hdesc)) {
1074 CERROR("Unable to initialize checksum hash %s\n",
1075 cfs_crypto_hash_name(cfs_alg));
1076 return PTR_ERR(hdesc);
1079 while (nob > 0 && pg_count > 0) {
1080 unsigned int count = pga[i]->count > nob ? nob : pga[i]->count;
1082 /* corrupt the data before we compute the checksum, to
1083 * simulate an OST->client data error */
1084 if (i == 0 && opc == OST_READ &&
1085 OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_RECEIVE)) {
1086 unsigned char *ptr = kmap(pga[i]->pg);
1087 int off = pga[i]->off & ~PAGE_MASK;
1089 memcpy(ptr + off, "bad1", min_t(typeof(nob), 4, nob));
1092 cfs_crypto_hash_update_page(hdesc, pga[i]->pg,
1093 pga[i]->off & ~PAGE_MASK,
1095 LL_CDEBUG_PAGE(D_PAGE, pga[i]->pg, "off %d\n",
1096 (int)(pga[i]->off & ~PAGE_MASK));
1098 nob -= pga[i]->count;
1103 bufsize = sizeof(cksum);
1104 cfs_crypto_hash_final(hdesc, (unsigned char *)&cksum, &bufsize);
1106 /* For sending we only compute the wrong checksum instead
1107 * of corrupting the data so it is still correct on a redo */
1108 if (opc == OST_WRITE && OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_SEND))
1115 osc_brw_prep_request(int cmd, struct client_obd *cli, struct obdo *oa,
1116 u32 page_count, struct brw_page **pga,
1117 struct ptlrpc_request **reqp, int resend)
1119 struct ptlrpc_request *req;
1120 struct ptlrpc_bulk_desc *desc;
1121 struct ost_body *body;
1122 struct obd_ioobj *ioobj;
1123 struct niobuf_remote *niobuf;
1124 int niocount, i, requested_nob, opc, rc;
1125 struct osc_brw_async_args *aa;
1126 struct req_capsule *pill;
1127 struct brw_page *pg_prev;
1130 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ))
1131 RETURN(-ENOMEM); /* Recoverable */
1132 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ2))
1133 RETURN(-EINVAL); /* Fatal */
1135 if ((cmd & OBD_BRW_WRITE) != 0) {
1137 req = ptlrpc_request_alloc_pool(cli->cl_import,
1139 &RQF_OST_BRW_WRITE);
1142 req = ptlrpc_request_alloc(cli->cl_import, &RQF_OST_BRW_READ);
1147 for (niocount = i = 1; i < page_count; i++) {
1148 if (!can_merge_pages(pga[i - 1], pga[i]))
1152 pill = &req->rq_pill;
1153 req_capsule_set_size(pill, &RMF_OBD_IOOBJ, RCL_CLIENT,
1155 req_capsule_set_size(pill, &RMF_NIOBUF_REMOTE, RCL_CLIENT,
1156 niocount * sizeof(*niobuf));
1158 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, opc);
1160 ptlrpc_request_free(req);
1163 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
1164 ptlrpc_at_set_req_timeout(req);
1165 /* ask ptlrpc not to resend on EINPROGRESS since BRWs have their own
1167 req->rq_no_retry_einprogress = 1;
1169 desc = ptlrpc_prep_bulk_imp(req, page_count,
1170 cli->cl_import->imp_connect_data.ocd_brw_size >> LNET_MTU_BITS,
1171 (opc == OST_WRITE ? PTLRPC_BULK_GET_SOURCE :
1172 PTLRPC_BULK_PUT_SINK) |
1173 PTLRPC_BULK_BUF_KIOV,
1175 &ptlrpc_bulk_kiov_pin_ops);
1178 GOTO(out, rc = -ENOMEM);
1179 /* NB request now owns desc and will free it when it gets freed */
1181 body = req_capsule_client_get(pill, &RMF_OST_BODY);
1182 ioobj = req_capsule_client_get(pill, &RMF_OBD_IOOBJ);
1183 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1184 LASSERT(body != NULL && ioobj != NULL && niobuf != NULL);
1186 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
1188 obdo_to_ioobj(oa, ioobj);
1189 ioobj->ioo_bufcnt = niocount;
1190 /* The high bits of ioo_max_brw tells server _maximum_ number of bulks
1191 * that might be send for this request. The actual number is decided
1192 * when the RPC is finally sent in ptlrpc_register_bulk(). It sends
1193 * "max - 1" for old client compatibility sending "0", and also so the
1194 * the actual maximum is a power-of-two number, not one less. LU-1431 */
1195 ioobj_max_brw_set(ioobj, desc->bd_md_max_brw);
1196 LASSERT(page_count > 0);
1198 for (requested_nob = i = 0; i < page_count; i++, niobuf++) {
1199 struct brw_page *pg = pga[i];
1200 int poff = pg->off & ~PAGE_MASK;
1202 LASSERT(pg->count > 0);
1203 /* make sure there is no gap in the middle of page array */
1204 LASSERTF(page_count == 1 ||
1205 (ergo(i == 0, poff + pg->count == PAGE_SIZE) &&
1206 ergo(i > 0 && i < page_count - 1,
1207 poff == 0 && pg->count == PAGE_SIZE) &&
1208 ergo(i == page_count - 1, poff == 0)),
1209 "i: %d/%d pg: %p off: %llu, count: %u\n",
1210 i, page_count, pg, pg->off, pg->count);
1211 LASSERTF(i == 0 || pg->off > pg_prev->off,
1212 "i %d p_c %u pg %p [pri %lu ind %lu] off %llu"
1213 " prev_pg %p [pri %lu ind %lu] off %llu\n",
1215 pg->pg, page_private(pg->pg), pg->pg->index, pg->off,
1216 pg_prev->pg, page_private(pg_prev->pg),
1217 pg_prev->pg->index, pg_prev->off);
1218 LASSERT((pga[0]->flag & OBD_BRW_SRVLOCK) ==
1219 (pg->flag & OBD_BRW_SRVLOCK));
1221 desc->bd_frag_ops->add_kiov_frag(desc, pg->pg, poff, pg->count);
1222 requested_nob += pg->count;
1224 if (i > 0 && can_merge_pages(pg_prev, pg)) {
1226 niobuf->rnb_len += pg->count;
1228 niobuf->rnb_offset = pg->off;
1229 niobuf->rnb_len = pg->count;
1230 niobuf->rnb_flags = pg->flag;
1235 LASSERTF((void *)(niobuf - niocount) ==
1236 req_capsule_client_get(&req->rq_pill, &RMF_NIOBUF_REMOTE),
1237 "want %p - real %p\n", req_capsule_client_get(&req->rq_pill,
1238 &RMF_NIOBUF_REMOTE), (void *)(niobuf - niocount));
1240 osc_announce_cached(cli, &body->oa, opc == OST_WRITE ? requested_nob:0);
1242 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1243 body->oa.o_valid |= OBD_MD_FLFLAGS;
1244 body->oa.o_flags = 0;
1246 body->oa.o_flags |= OBD_FL_RECOV_RESEND;
1249 if (osc_should_shrink_grant(cli))
1250 osc_shrink_grant_local(cli, &body->oa);
1252 /* size[REQ_REC_OFF] still sizeof (*body) */
1253 if (opc == OST_WRITE) {
1254 if (cli->cl_checksum &&
1255 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1256 /* store cl_cksum_type in a local variable since
1257 * it can be changed via lprocfs */
1258 cksum_type_t cksum_type = cli->cl_cksum_type;
1260 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1261 oa->o_flags &= OBD_FL_LOCAL_MASK;
1262 body->oa.o_flags = 0;
1264 body->oa.o_flags |= cksum_type_pack(cksum_type);
1265 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1266 body->oa.o_cksum = osc_checksum_bulk(requested_nob,
1270 CDEBUG(D_PAGE, "checksum at write origin: %x\n",
1272 /* save this in 'oa', too, for later checking */
1273 oa->o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1274 oa->o_flags |= cksum_type_pack(cksum_type);
1276 /* clear out the checksum flag, in case this is a
1277 * resend but cl_checksum is no longer set. b=11238 */
1278 oa->o_valid &= ~OBD_MD_FLCKSUM;
1280 oa->o_cksum = body->oa.o_cksum;
1281 /* 1 RC per niobuf */
1282 req_capsule_set_size(pill, &RMF_RCS, RCL_SERVER,
1283 sizeof(__u32) * niocount);
1285 if (cli->cl_checksum &&
1286 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1287 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0)
1288 body->oa.o_flags = 0;
1289 body->oa.o_flags |= cksum_type_pack(cli->cl_cksum_type);
1290 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1293 ptlrpc_request_set_replen(req);
1295 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1296 aa = ptlrpc_req_async_args(req);
1298 aa->aa_requested_nob = requested_nob;
1299 aa->aa_nio_count = niocount;
1300 aa->aa_page_count = page_count;
1304 INIT_LIST_HEAD(&aa->aa_oaps);
1307 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1308 CDEBUG(D_RPCTRACE, "brw rpc %p - object "DOSTID" offset %lld<>%lld\n",
1309 req, POSTID(&oa->o_oi), niobuf[0].rnb_offset,
1310 niobuf[niocount - 1].rnb_offset + niobuf[niocount - 1].rnb_len);
1314 ptlrpc_req_finished(req);
1319 check_write_checksum(struct obdo *oa, const struct lnet_process_id *peer,
1320 __u32 client_cksum, __u32 server_cksum, int nob,
1321 size_t page_count, struct brw_page **pga,
1322 cksum_type_t client_cksum_type)
1326 cksum_type_t cksum_type;
1328 if (server_cksum == client_cksum) {
1329 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1333 cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
1335 new_cksum = osc_checksum_bulk(nob, page_count, pga, OST_WRITE,
1338 if (cksum_type != client_cksum_type)
1339 msg = "the server did not use the checksum type specified in "
1340 "the original request - likely a protocol problem";
1341 else if (new_cksum == server_cksum)
1342 msg = "changed on the client after we checksummed it - "
1343 "likely false positive due to mmap IO (bug 11742)";
1344 else if (new_cksum == client_cksum)
1345 msg = "changed in transit before arrival at OST";
1347 msg = "changed in transit AND doesn't match the original - "
1348 "likely false positive due to mmap IO (bug 11742)";
1350 LCONSOLE_ERROR_MSG(0x132, "BAD WRITE CHECKSUM: %s: from %s inode "DFID
1351 " object "DOSTID" extent [%llu-%llu]\n",
1352 msg, libcfs_nid2str(peer->nid),
1353 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
1354 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
1355 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
1356 POSTID(&oa->o_oi), pga[0]->off,
1357 pga[page_count-1]->off + pga[page_count-1]->count - 1);
1358 CERROR("original client csum %x (type %x), server csum %x (type %x), "
1359 "client csum now %x\n", client_cksum, client_cksum_type,
1360 server_cksum, cksum_type, new_cksum);
1364 /* Note rc enters this function as number of bytes transferred */
1365 static int osc_brw_fini_request(struct ptlrpc_request *req, int rc)
1367 struct osc_brw_async_args *aa = (void *)&req->rq_async_args;
1368 const struct lnet_process_id *peer =
1369 &req->rq_import->imp_connection->c_peer;
1370 struct client_obd *cli = aa->aa_cli;
1371 struct ost_body *body;
1372 u32 client_cksum = 0;
1375 if (rc < 0 && rc != -EDQUOT) {
1376 DEBUG_REQ(D_INFO, req, "Failed request with rc = %d\n", rc);
1380 LASSERTF(req->rq_repmsg != NULL, "rc = %d\n", rc);
1381 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
1383 DEBUG_REQ(D_INFO, req, "Can't unpack body\n");
1387 /* set/clear over quota flag for a uid/gid/projid */
1388 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE &&
1389 body->oa.o_valid & (OBD_MD_FLALLQUOTA)) {
1390 unsigned qid[LL_MAXQUOTAS] =
1391 { body->oa.o_uid, body->oa.o_gid,
1392 body->oa.o_projid };
1394 CDEBUG(D_QUOTA, "setdq for [%u %u %u] with valid %#llx, flags %x\n",
1395 body->oa.o_uid, body->oa.o_gid, body->oa.o_projid,
1396 body->oa.o_valid, body->oa.o_flags);
1397 osc_quota_setdq(cli, qid, body->oa.o_valid, body->oa.o_flags);
1400 osc_update_grant(cli, body);
1405 if (aa->aa_oa->o_valid & OBD_MD_FLCKSUM)
1406 client_cksum = aa->aa_oa->o_cksum; /* save for later */
1408 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1410 CERROR("Unexpected +ve rc %d\n", rc);
1413 LASSERT(req->rq_bulk->bd_nob == aa->aa_requested_nob);
1415 if (sptlrpc_cli_unwrap_bulk_write(req, req->rq_bulk))
1418 if ((aa->aa_oa->o_valid & OBD_MD_FLCKSUM) && client_cksum &&
1419 check_write_checksum(&body->oa, peer, client_cksum,
1420 body->oa.o_cksum, aa->aa_requested_nob,
1421 aa->aa_page_count, aa->aa_ppga,
1422 cksum_type_unpack(aa->aa_oa->o_flags)))
1425 rc = check_write_rcs(req, aa->aa_requested_nob,aa->aa_nio_count,
1426 aa->aa_page_count, aa->aa_ppga);
1430 /* The rest of this function executes only for OST_READs */
1432 /* if unwrap_bulk failed, return -EAGAIN to retry */
1433 rc = sptlrpc_cli_unwrap_bulk_read(req, req->rq_bulk, rc);
1435 GOTO(out, rc = -EAGAIN);
1437 if (rc > aa->aa_requested_nob) {
1438 CERROR("Unexpected rc %d (%d requested)\n", rc,
1439 aa->aa_requested_nob);
1443 if (rc != req->rq_bulk->bd_nob_transferred) {
1444 CERROR ("Unexpected rc %d (%d transferred)\n",
1445 rc, req->rq_bulk->bd_nob_transferred);
1449 if (rc < aa->aa_requested_nob)
1450 handle_short_read(rc, aa->aa_page_count, aa->aa_ppga);
1452 if (body->oa.o_valid & OBD_MD_FLCKSUM) {
1453 static int cksum_counter;
1454 u32 server_cksum = body->oa.o_cksum;
1457 cksum_type_t cksum_type;
1459 cksum_type = cksum_type_unpack(body->oa.o_valid &OBD_MD_FLFLAGS?
1460 body->oa.o_flags : 0);
1461 client_cksum = osc_checksum_bulk(rc, aa->aa_page_count,
1462 aa->aa_ppga, OST_READ,
1465 if (peer->nid != req->rq_bulk->bd_sender) {
1467 router = libcfs_nid2str(req->rq_bulk->bd_sender);
1470 if (server_cksum != client_cksum) {
1471 LCONSOLE_ERROR_MSG(0x133, "%s: BAD READ CHECKSUM: from "
1472 "%s%s%s inode "DFID" object "DOSTID
1473 " extent [%llu-%llu]\n",
1474 req->rq_import->imp_obd->obd_name,
1475 libcfs_nid2str(peer->nid),
1477 body->oa.o_valid & OBD_MD_FLFID ?
1478 body->oa.o_parent_seq : (__u64)0,
1479 body->oa.o_valid & OBD_MD_FLFID ?
1480 body->oa.o_parent_oid : 0,
1481 body->oa.o_valid & OBD_MD_FLFID ?
1482 body->oa.o_parent_ver : 0,
1483 POSTID(&body->oa.o_oi),
1484 aa->aa_ppga[0]->off,
1485 aa->aa_ppga[aa->aa_page_count-1]->off +
1486 aa->aa_ppga[aa->aa_page_count-1]->count -
1488 CERROR("client %x, server %x, cksum_type %x\n",
1489 client_cksum, server_cksum, cksum_type);
1491 aa->aa_oa->o_cksum = client_cksum;
1495 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1498 } else if (unlikely(client_cksum)) {
1499 static int cksum_missed;
1502 if ((cksum_missed & (-cksum_missed)) == cksum_missed)
1503 CERROR("Checksum %u requested from %s but not sent\n",
1504 cksum_missed, libcfs_nid2str(peer->nid));
1510 lustre_get_wire_obdo(&req->rq_import->imp_connect_data,
1511 aa->aa_oa, &body->oa);
1516 static int osc_brw_redo_request(struct ptlrpc_request *request,
1517 struct osc_brw_async_args *aa, int rc)
1519 struct ptlrpc_request *new_req;
1520 struct osc_brw_async_args *new_aa;
1521 struct osc_async_page *oap;
1524 DEBUG_REQ(rc == -EINPROGRESS ? D_RPCTRACE : D_ERROR, request,
1525 "redo for recoverable error %d", rc);
1527 rc = osc_brw_prep_request(lustre_msg_get_opc(request->rq_reqmsg) ==
1528 OST_WRITE ? OBD_BRW_WRITE : OBD_BRW_READ,
1529 aa->aa_cli, aa->aa_oa, aa->aa_page_count,
1530 aa->aa_ppga, &new_req, 1);
1534 list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
1535 if (oap->oap_request != NULL) {
1536 LASSERTF(request == oap->oap_request,
1537 "request %p != oap_request %p\n",
1538 request, oap->oap_request);
1539 if (oap->oap_interrupted) {
1540 ptlrpc_req_finished(new_req);
1545 /* New request takes over pga and oaps from old request.
1546 * Note that copying a list_head doesn't work, need to move it... */
1548 new_req->rq_interpret_reply = request->rq_interpret_reply;
1549 new_req->rq_async_args = request->rq_async_args;
1550 new_req->rq_commit_cb = request->rq_commit_cb;
1551 /* cap resend delay to the current request timeout, this is similar to
1552 * what ptlrpc does (see after_reply()) */
1553 if (aa->aa_resends > new_req->rq_timeout)
1554 new_req->rq_sent = cfs_time_current_sec() + new_req->rq_timeout;
1556 new_req->rq_sent = cfs_time_current_sec() + aa->aa_resends;
1557 new_req->rq_generation_set = 1;
1558 new_req->rq_import_generation = request->rq_import_generation;
1560 new_aa = ptlrpc_req_async_args(new_req);
1562 INIT_LIST_HEAD(&new_aa->aa_oaps);
1563 list_splice_init(&aa->aa_oaps, &new_aa->aa_oaps);
1564 INIT_LIST_HEAD(&new_aa->aa_exts);
1565 list_splice_init(&aa->aa_exts, &new_aa->aa_exts);
1566 new_aa->aa_resends = aa->aa_resends;
1568 list_for_each_entry(oap, &new_aa->aa_oaps, oap_rpc_item) {
1569 if (oap->oap_request) {
1570 ptlrpc_req_finished(oap->oap_request);
1571 oap->oap_request = ptlrpc_request_addref(new_req);
1575 /* XXX: This code will run into problem if we're going to support
1576 * to add a series of BRW RPCs into a self-defined ptlrpc_request_set
1577 * and wait for all of them to be finished. We should inherit request
1578 * set from old request. */
1579 ptlrpcd_add_req(new_req);
1581 DEBUG_REQ(D_INFO, new_req, "new request");
1586 * ugh, we want disk allocation on the target to happen in offset order. we'll
1587 * follow sedgewicks advice and stick to the dead simple shellsort -- it'll do
1588 * fine for our small page arrays and doesn't require allocation. its an
1589 * insertion sort that swaps elements that are strides apart, shrinking the
1590 * stride down until its '1' and the array is sorted.
1592 static void sort_brw_pages(struct brw_page **array, int num)
1595 struct brw_page *tmp;
1599 for (stride = 1; stride < num ; stride = (stride * 3) + 1)
1604 for (i = stride ; i < num ; i++) {
1607 while (j >= stride && array[j - stride]->off > tmp->off) {
1608 array[j] = array[j - stride];
1613 } while (stride > 1);
1616 static void osc_release_ppga(struct brw_page **ppga, size_t count)
1618 LASSERT(ppga != NULL);
1619 OBD_FREE(ppga, sizeof(*ppga) * count);
1622 static int brw_interpret(const struct lu_env *env,
1623 struct ptlrpc_request *req, void *data, int rc)
1625 struct osc_brw_async_args *aa = data;
1626 struct osc_extent *ext;
1627 struct osc_extent *tmp;
1628 struct client_obd *cli = aa->aa_cli;
1631 rc = osc_brw_fini_request(req, rc);
1632 CDEBUG(D_INODE, "request %p aa %p rc %d\n", req, aa, rc);
1633 /* When server return -EINPROGRESS, client should always retry
1634 * regardless of the number of times the bulk was resent already. */
1635 if (osc_recoverable_error(rc)) {
1636 if (req->rq_import_generation !=
1637 req->rq_import->imp_generation) {
1638 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1639 ""DOSTID", rc = %d.\n",
1640 req->rq_import->imp_obd->obd_name,
1641 POSTID(&aa->aa_oa->o_oi), rc);
1642 } else if (rc == -EINPROGRESS ||
1643 client_should_resend(aa->aa_resends, aa->aa_cli)) {
1644 rc = osc_brw_redo_request(req, aa, rc);
1646 CERROR("%s: too many resent retries for object: "
1647 "%llu:%llu, rc = %d.\n",
1648 req->rq_import->imp_obd->obd_name,
1649 POSTID(&aa->aa_oa->o_oi), rc);
1654 else if (rc == -EAGAIN || rc == -EINPROGRESS)
1659 struct obdo *oa = aa->aa_oa;
1660 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
1661 unsigned long valid = 0;
1662 struct cl_object *obj;
1663 struct osc_async_page *last;
1665 last = brw_page2oap(aa->aa_ppga[aa->aa_page_count - 1]);
1666 obj = osc2cl(last->oap_obj);
1668 cl_object_attr_lock(obj);
1669 if (oa->o_valid & OBD_MD_FLBLOCKS) {
1670 attr->cat_blocks = oa->o_blocks;
1671 valid |= CAT_BLOCKS;
1673 if (oa->o_valid & OBD_MD_FLMTIME) {
1674 attr->cat_mtime = oa->o_mtime;
1677 if (oa->o_valid & OBD_MD_FLATIME) {
1678 attr->cat_atime = oa->o_atime;
1681 if (oa->o_valid & OBD_MD_FLCTIME) {
1682 attr->cat_ctime = oa->o_ctime;
1686 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1687 struct lov_oinfo *loi = cl2osc(obj)->oo_oinfo;
1688 loff_t last_off = last->oap_count + last->oap_obj_off +
1691 /* Change file size if this is an out of quota or
1692 * direct IO write and it extends the file size */
1693 if (loi->loi_lvb.lvb_size < last_off) {
1694 attr->cat_size = last_off;
1697 /* Extend KMS if it's not a lockless write */
1698 if (loi->loi_kms < last_off &&
1699 oap2osc_page(last)->ops_srvlock == 0) {
1700 attr->cat_kms = last_off;
1706 cl_object_attr_update(env, obj, attr, valid);
1707 cl_object_attr_unlock(obj);
1709 OBDO_FREE(aa->aa_oa);
1711 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE && rc == 0)
1712 osc_inc_unstable_pages(req);
1714 list_for_each_entry_safe(ext, tmp, &aa->aa_exts, oe_link) {
1715 list_del_init(&ext->oe_link);
1716 osc_extent_finish(env, ext, 1, rc);
1718 LASSERT(list_empty(&aa->aa_exts));
1719 LASSERT(list_empty(&aa->aa_oaps));
1721 osc_release_ppga(aa->aa_ppga, aa->aa_page_count);
1722 ptlrpc_lprocfs_brw(req, req->rq_bulk->bd_nob_transferred);
1724 spin_lock(&cli->cl_loi_list_lock);
1725 /* We need to decrement before osc_ap_completion->osc_wake_cache_waiters
1726 * is called so we know whether to go to sync BRWs or wait for more
1727 * RPCs to complete */
1728 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE)
1729 cli->cl_w_in_flight--;
1731 cli->cl_r_in_flight--;
1732 osc_wake_cache_waiters(cli);
1733 spin_unlock(&cli->cl_loi_list_lock);
1735 osc_io_unplug(env, cli, NULL);
1739 static void brw_commit(struct ptlrpc_request *req)
1741 /* If osc_inc_unstable_pages (via osc_extent_finish) races with
1742 * this called via the rq_commit_cb, I need to ensure
1743 * osc_dec_unstable_pages is still called. Otherwise unstable
1744 * pages may be leaked. */
1745 spin_lock(&req->rq_lock);
1746 if (likely(req->rq_unstable)) {
1747 req->rq_unstable = 0;
1748 spin_unlock(&req->rq_lock);
1750 osc_dec_unstable_pages(req);
1752 req->rq_committed = 1;
1753 spin_unlock(&req->rq_lock);
1758 * Build an RPC by the list of extent @ext_list. The caller must ensure
1759 * that the total pages in this list are NOT over max pages per RPC.
1760 * Extents in the list must be in OES_RPC state.
1762 int osc_build_rpc(const struct lu_env *env, struct client_obd *cli,
1763 struct list_head *ext_list, int cmd)
1765 struct ptlrpc_request *req = NULL;
1766 struct osc_extent *ext;
1767 struct brw_page **pga = NULL;
1768 struct osc_brw_async_args *aa = NULL;
1769 struct obdo *oa = NULL;
1770 struct osc_async_page *oap;
1771 struct osc_object *obj = NULL;
1772 struct cl_req_attr *crattr = NULL;
1773 loff_t starting_offset = OBD_OBJECT_EOF;
1774 loff_t ending_offset = 0;
1778 bool soft_sync = false;
1779 bool interrupted = false;
1783 struct list_head rpc_list = LIST_HEAD_INIT(rpc_list);
1784 struct ost_body *body;
1786 LASSERT(!list_empty(ext_list));
1788 /* add pages into rpc_list to build BRW rpc */
1789 list_for_each_entry(ext, ext_list, oe_link) {
1790 LASSERT(ext->oe_state == OES_RPC);
1791 mem_tight |= ext->oe_memalloc;
1792 grant += ext->oe_grants;
1793 page_count += ext->oe_nr_pages;
1798 soft_sync = osc_over_unstable_soft_limit(cli);
1800 mpflag = cfs_memory_pressure_get_and_set();
1802 OBD_ALLOC(pga, sizeof(*pga) * page_count);
1804 GOTO(out, rc = -ENOMEM);
1808 GOTO(out, rc = -ENOMEM);
1811 list_for_each_entry(ext, ext_list, oe_link) {
1812 list_for_each_entry(oap, &ext->oe_pages, oap_pending_item) {
1814 oap->oap_brw_flags |= OBD_BRW_MEMALLOC;
1816 oap->oap_brw_flags |= OBD_BRW_SOFT_SYNC;
1817 pga[i] = &oap->oap_brw_page;
1818 pga[i]->off = oap->oap_obj_off + oap->oap_page_off;
1821 list_add_tail(&oap->oap_rpc_item, &rpc_list);
1822 if (starting_offset == OBD_OBJECT_EOF ||
1823 starting_offset > oap->oap_obj_off)
1824 starting_offset = oap->oap_obj_off;
1826 LASSERT(oap->oap_page_off == 0);
1827 if (ending_offset < oap->oap_obj_off + oap->oap_count)
1828 ending_offset = oap->oap_obj_off +
1831 LASSERT(oap->oap_page_off + oap->oap_count ==
1833 if (oap->oap_interrupted)
1838 /* first page in the list */
1839 oap = list_entry(rpc_list.next, typeof(*oap), oap_rpc_item);
1841 crattr = &osc_env_info(env)->oti_req_attr;
1842 memset(crattr, 0, sizeof(*crattr));
1843 crattr->cra_type = (cmd & OBD_BRW_WRITE) ? CRT_WRITE : CRT_READ;
1844 crattr->cra_flags = ~0ULL;
1845 crattr->cra_page = oap2cl_page(oap);
1846 crattr->cra_oa = oa;
1847 cl_req_attr_set(env, osc2cl(obj), crattr);
1849 if (cmd == OBD_BRW_WRITE)
1850 oa->o_grant_used = grant;
1852 sort_brw_pages(pga, page_count);
1853 rc = osc_brw_prep_request(cmd, cli, oa, page_count, pga, &req, 0);
1855 CERROR("prep_req failed: %d\n", rc);
1859 req->rq_commit_cb = brw_commit;
1860 req->rq_interpret_reply = brw_interpret;
1861 req->rq_memalloc = mem_tight != 0;
1862 oap->oap_request = ptlrpc_request_addref(req);
1863 if (interrupted && !req->rq_intr)
1864 ptlrpc_mark_interrupted(req);
1866 /* Need to update the timestamps after the request is built in case
1867 * we race with setattr (locally or in queue at OST). If OST gets
1868 * later setattr before earlier BRW (as determined by the request xid),
1869 * the OST will not use BRW timestamps. Sadly, there is no obvious
1870 * way to do this in a single call. bug 10150 */
1871 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
1872 crattr->cra_oa = &body->oa;
1873 crattr->cra_flags = OBD_MD_FLMTIME|OBD_MD_FLCTIME|OBD_MD_FLATIME;
1874 cl_req_attr_set(env, osc2cl(obj), crattr);
1875 lustre_msg_set_jobid(req->rq_reqmsg, crattr->cra_jobid);
1877 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1878 aa = ptlrpc_req_async_args(req);
1879 INIT_LIST_HEAD(&aa->aa_oaps);
1880 list_splice_init(&rpc_list, &aa->aa_oaps);
1881 INIT_LIST_HEAD(&aa->aa_exts);
1882 list_splice_init(ext_list, &aa->aa_exts);
1884 spin_lock(&cli->cl_loi_list_lock);
1885 starting_offset >>= PAGE_SHIFT;
1886 if (cmd == OBD_BRW_READ) {
1887 cli->cl_r_in_flight++;
1888 lprocfs_oh_tally_log2(&cli->cl_read_page_hist, page_count);
1889 lprocfs_oh_tally(&cli->cl_read_rpc_hist, cli->cl_r_in_flight);
1890 lprocfs_oh_tally_log2(&cli->cl_read_offset_hist,
1891 starting_offset + 1);
1893 cli->cl_w_in_flight++;
1894 lprocfs_oh_tally_log2(&cli->cl_write_page_hist, page_count);
1895 lprocfs_oh_tally(&cli->cl_write_rpc_hist, cli->cl_w_in_flight);
1896 lprocfs_oh_tally_log2(&cli->cl_write_offset_hist,
1897 starting_offset + 1);
1899 spin_unlock(&cli->cl_loi_list_lock);
1901 DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %ur/%uw in flight",
1902 page_count, aa, cli->cl_r_in_flight,
1903 cli->cl_w_in_flight);
1904 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_DELAY_IO, cfs_fail_val);
1906 ptlrpcd_add_req(req);
1912 cfs_memory_pressure_restore(mpflag);
1915 LASSERT(req == NULL);
1920 OBD_FREE(pga, sizeof(*pga) * page_count);
1921 /* this should happen rarely and is pretty bad, it makes the
1922 * pending list not follow the dirty order */
1923 while (!list_empty(ext_list)) {
1924 ext = list_entry(ext_list->next, struct osc_extent,
1926 list_del_init(&ext->oe_link);
1927 osc_extent_finish(env, ext, 0, rc);
1933 static int osc_set_lock_data(struct ldlm_lock *lock, void *data)
1937 LASSERT(lock != NULL);
1939 lock_res_and_lock(lock);
1941 if (lock->l_ast_data == NULL)
1942 lock->l_ast_data = data;
1943 if (lock->l_ast_data == data)
1946 unlock_res_and_lock(lock);
1951 static int osc_enqueue_fini(struct ptlrpc_request *req,
1952 osc_enqueue_upcall_f upcall, void *cookie,
1953 struct lustre_handle *lockh, enum ldlm_mode mode,
1954 __u64 *flags, int agl, int errcode)
1956 bool intent = *flags & LDLM_FL_HAS_INTENT;
1960 /* The request was created before ldlm_cli_enqueue call. */
1961 if (intent && errcode == ELDLM_LOCK_ABORTED) {
1962 struct ldlm_reply *rep;
1964 rep = req_capsule_server_get(&req->rq_pill, &RMF_DLM_REP);
1965 LASSERT(rep != NULL);
1967 rep->lock_policy_res1 =
1968 ptlrpc_status_ntoh(rep->lock_policy_res1);
1969 if (rep->lock_policy_res1)
1970 errcode = rep->lock_policy_res1;
1972 *flags |= LDLM_FL_LVB_READY;
1973 } else if (errcode == ELDLM_OK) {
1974 *flags |= LDLM_FL_LVB_READY;
1977 /* Call the update callback. */
1978 rc = (*upcall)(cookie, lockh, errcode);
1980 /* release the reference taken in ldlm_cli_enqueue() */
1981 if (errcode == ELDLM_LOCK_MATCHED)
1983 if (errcode == ELDLM_OK && lustre_handle_is_used(lockh))
1984 ldlm_lock_decref(lockh, mode);
1989 static int osc_enqueue_interpret(const struct lu_env *env,
1990 struct ptlrpc_request *req,
1991 struct osc_enqueue_args *aa, int rc)
1993 struct ldlm_lock *lock;
1994 struct lustre_handle *lockh = &aa->oa_lockh;
1995 enum ldlm_mode mode = aa->oa_mode;
1996 struct ost_lvb *lvb = aa->oa_lvb;
1997 __u32 lvb_len = sizeof(*lvb);
2002 /* ldlm_cli_enqueue is holding a reference on the lock, so it must
2004 lock = ldlm_handle2lock(lockh);
2005 LASSERTF(lock != NULL,
2006 "lockh %#llx, req %p, aa %p - client evicted?\n",
2007 lockh->cookie, req, aa);
2009 /* Take an additional reference so that a blocking AST that
2010 * ldlm_cli_enqueue_fini() might post for a failed lock, is guaranteed
2011 * to arrive after an upcall has been executed by
2012 * osc_enqueue_fini(). */
2013 ldlm_lock_addref(lockh, mode);
2015 /* Let cl_lock_state_wait fail with -ERESTARTSYS to unuse sublocks. */
2016 OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_ENQUEUE_HANG, 2);
2018 /* Let CP AST to grant the lock first. */
2019 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_ENQ_RACE, 1);
2022 LASSERT(aa->oa_lvb == NULL);
2023 LASSERT(aa->oa_flags == NULL);
2024 aa->oa_flags = &flags;
2027 /* Complete obtaining the lock procedure. */
2028 rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, aa->oa_type, 1,
2029 aa->oa_mode, aa->oa_flags, lvb, lvb_len,
2031 /* Complete osc stuff. */
2032 rc = osc_enqueue_fini(req, aa->oa_upcall, aa->oa_cookie, lockh, mode,
2033 aa->oa_flags, aa->oa_agl, rc);
2035 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_CANCEL_RACE, 10);
2037 ldlm_lock_decref(lockh, mode);
2038 LDLM_LOCK_PUT(lock);
2042 struct ptlrpc_request_set *PTLRPCD_SET = (void *)1;
2044 /* When enqueuing asynchronously, locks are not ordered, we can obtain a lock
2045 * from the 2nd OSC before a lock from the 1st one. This does not deadlock with
2046 * other synchronous requests, however keeping some locks and trying to obtain
2047 * others may take a considerable amount of time in a case of ost failure; and
2048 * when other sync requests do not get released lock from a client, the client
2049 * is evicted from the cluster -- such scenarious make the life difficult, so
2050 * release locks just after they are obtained. */
2051 int osc_enqueue_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2052 __u64 *flags, union ldlm_policy_data *policy,
2053 struct ost_lvb *lvb, int kms_valid,
2054 osc_enqueue_upcall_f upcall, void *cookie,
2055 struct ldlm_enqueue_info *einfo,
2056 struct ptlrpc_request_set *rqset, int async, int agl)
2058 struct obd_device *obd = exp->exp_obd;
2059 struct lustre_handle lockh = { 0 };
2060 struct ptlrpc_request *req = NULL;
2061 int intent = *flags & LDLM_FL_HAS_INTENT;
2062 __u64 match_flags = *flags;
2063 enum ldlm_mode mode;
2067 /* Filesystem lock extents are extended to page boundaries so that
2068 * dealing with the page cache is a little smoother. */
2069 policy->l_extent.start -= policy->l_extent.start & ~PAGE_MASK;
2070 policy->l_extent.end |= ~PAGE_MASK;
2073 * kms is not valid when either object is completely fresh (so that no
2074 * locks are cached), or object was evicted. In the latter case cached
2075 * lock cannot be used, because it would prime inode state with
2076 * potentially stale LVB.
2081 /* Next, search for already existing extent locks that will cover us */
2082 /* If we're trying to read, we also search for an existing PW lock. The
2083 * VFS and page cache already protect us locally, so lots of readers/
2084 * writers can share a single PW lock.
2086 * There are problems with conversion deadlocks, so instead of
2087 * converting a read lock to a write lock, we'll just enqueue a new
2090 * At some point we should cancel the read lock instead of making them
2091 * send us a blocking callback, but there are problems with canceling
2092 * locks out from other users right now, too. */
2093 mode = einfo->ei_mode;
2094 if (einfo->ei_mode == LCK_PR)
2097 match_flags |= LDLM_FL_LVB_READY;
2099 match_flags |= LDLM_FL_BLOCK_GRANTED;
2100 mode = ldlm_lock_match(obd->obd_namespace, match_flags, res_id,
2101 einfo->ei_type, policy, mode, &lockh, 0);
2103 struct ldlm_lock *matched;
2105 if (*flags & LDLM_FL_TEST_LOCK)
2108 matched = ldlm_handle2lock(&lockh);
2110 /* AGL enqueues DLM locks speculatively. Therefore if
2111 * it already exists a DLM lock, it wll just inform the
2112 * caller to cancel the AGL process for this stripe. */
2113 ldlm_lock_decref(&lockh, mode);
2114 LDLM_LOCK_PUT(matched);
2116 } else if (osc_set_lock_data(matched, einfo->ei_cbdata)) {
2117 *flags |= LDLM_FL_LVB_READY;
2119 /* We already have a lock, and it's referenced. */
2120 (*upcall)(cookie, &lockh, ELDLM_LOCK_MATCHED);
2122 ldlm_lock_decref(&lockh, mode);
2123 LDLM_LOCK_PUT(matched);
2126 ldlm_lock_decref(&lockh, mode);
2127 LDLM_LOCK_PUT(matched);
2132 if (*flags & (LDLM_FL_TEST_LOCK | LDLM_FL_MATCH_LOCK))
2136 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2137 &RQF_LDLM_ENQUEUE_LVB);
2141 rc = ldlm_prep_enqueue_req(exp, req, NULL, 0);
2143 ptlrpc_request_free(req);
2147 req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER,
2149 ptlrpc_request_set_replen(req);
2152 /* users of osc_enqueue() can pass this flag for ldlm_lock_match() */
2153 *flags &= ~LDLM_FL_BLOCK_GRANTED;
2155 rc = ldlm_cli_enqueue(exp, &req, einfo, res_id, policy, flags, lvb,
2156 sizeof(*lvb), LVB_T_OST, &lockh, async);
2159 struct osc_enqueue_args *aa;
2160 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2161 aa = ptlrpc_req_async_args(req);
2163 aa->oa_mode = einfo->ei_mode;
2164 aa->oa_type = einfo->ei_type;
2165 lustre_handle_copy(&aa->oa_lockh, &lockh);
2166 aa->oa_upcall = upcall;
2167 aa->oa_cookie = cookie;
2170 aa->oa_flags = flags;
2173 /* AGL is essentially to enqueue an DLM lock
2174 * in advance, so we don't care about the
2175 * result of AGL enqueue. */
2177 aa->oa_flags = NULL;
2180 req->rq_interpret_reply =
2181 (ptlrpc_interpterer_t)osc_enqueue_interpret;
2182 if (rqset == PTLRPCD_SET)
2183 ptlrpcd_add_req(req);
2185 ptlrpc_set_add_req(rqset, req);
2186 } else if (intent) {
2187 ptlrpc_req_finished(req);
2192 rc = osc_enqueue_fini(req, upcall, cookie, &lockh, einfo->ei_mode,
2195 ptlrpc_req_finished(req);
2200 int osc_match_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2201 enum ldlm_type type, union ldlm_policy_data *policy,
2202 enum ldlm_mode mode, __u64 *flags, void *data,
2203 struct lustre_handle *lockh, int unref)
2205 struct obd_device *obd = exp->exp_obd;
2206 __u64 lflags = *flags;
2210 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_MATCH))
2213 /* Filesystem lock extents are extended to page boundaries so that
2214 * dealing with the page cache is a little smoother */
2215 policy->l_extent.start -= policy->l_extent.start & ~PAGE_MASK;
2216 policy->l_extent.end |= ~PAGE_MASK;
2218 /* Next, search for already existing extent locks that will cover us */
2219 /* If we're trying to read, we also search for an existing PW lock. The
2220 * VFS and page cache already protect us locally, so lots of readers/
2221 * writers can share a single PW lock. */
2225 rc = ldlm_lock_match(obd->obd_namespace, lflags,
2226 res_id, type, policy, rc, lockh, unref);
2227 if (rc == 0 || lflags & LDLM_FL_TEST_LOCK)
2231 struct ldlm_lock *lock = ldlm_handle2lock(lockh);
2233 LASSERT(lock != NULL);
2234 if (!osc_set_lock_data(lock, data)) {
2235 ldlm_lock_decref(lockh, rc);
2238 LDLM_LOCK_PUT(lock);
2243 static int osc_statfs_interpret(const struct lu_env *env,
2244 struct ptlrpc_request *req,
2245 struct osc_async_args *aa, int rc)
2247 struct obd_statfs *msfs;
2251 /* The request has in fact never been sent
2252 * due to issues at a higher level (LOV).
2253 * Exit immediately since the caller is
2254 * aware of the problem and takes care
2255 * of the clean up */
2258 if ((rc == -ENOTCONN || rc == -EAGAIN) &&
2259 (aa->aa_oi->oi_flags & OBD_STATFS_NODELAY))
2265 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2267 GOTO(out, rc = -EPROTO);
2270 *aa->aa_oi->oi_osfs = *msfs;
2272 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
2276 static int osc_statfs_async(struct obd_export *exp,
2277 struct obd_info *oinfo, __u64 max_age,
2278 struct ptlrpc_request_set *rqset)
2280 struct obd_device *obd = class_exp2obd(exp);
2281 struct ptlrpc_request *req;
2282 struct osc_async_args *aa;
2286 /* We could possibly pass max_age in the request (as an absolute
2287 * timestamp or a "seconds.usec ago") so the target can avoid doing
2288 * extra calls into the filesystem if that isn't necessary (e.g.
2289 * during mount that would help a bit). Having relative timestamps
2290 * is not so great if request processing is slow, while absolute
2291 * timestamps are not ideal because they need time synchronization. */
2292 req = ptlrpc_request_alloc(obd->u.cli.cl_import, &RQF_OST_STATFS);
2296 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2298 ptlrpc_request_free(req);
2301 ptlrpc_request_set_replen(req);
2302 req->rq_request_portal = OST_CREATE_PORTAL;
2303 ptlrpc_at_set_req_timeout(req);
2305 if (oinfo->oi_flags & OBD_STATFS_NODELAY) {
2306 /* procfs requests not want stat in wait for avoid deadlock */
2307 req->rq_no_resend = 1;
2308 req->rq_no_delay = 1;
2311 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_statfs_interpret;
2312 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2313 aa = ptlrpc_req_async_args(req);
2316 ptlrpc_set_add_req(rqset, req);
2320 static int osc_statfs(const struct lu_env *env, struct obd_export *exp,
2321 struct obd_statfs *osfs, __u64 max_age, __u32 flags)
2323 struct obd_device *obd = class_exp2obd(exp);
2324 struct obd_statfs *msfs;
2325 struct ptlrpc_request *req;
2326 struct obd_import *imp = NULL;
2330 /*Since the request might also come from lprocfs, so we need
2331 *sync this with client_disconnect_export Bug15684*/
2332 down_read(&obd->u.cli.cl_sem);
2333 if (obd->u.cli.cl_import)
2334 imp = class_import_get(obd->u.cli.cl_import);
2335 up_read(&obd->u.cli.cl_sem);
2339 /* We could possibly pass max_age in the request (as an absolute
2340 * timestamp or a "seconds.usec ago") so the target can avoid doing
2341 * extra calls into the filesystem if that isn't necessary (e.g.
2342 * during mount that would help a bit). Having relative timestamps
2343 * is not so great if request processing is slow, while absolute
2344 * timestamps are not ideal because they need time synchronization. */
2345 req = ptlrpc_request_alloc(imp, &RQF_OST_STATFS);
2347 class_import_put(imp);
2352 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2354 ptlrpc_request_free(req);
2357 ptlrpc_request_set_replen(req);
2358 req->rq_request_portal = OST_CREATE_PORTAL;
2359 ptlrpc_at_set_req_timeout(req);
2361 if (flags & OBD_STATFS_NODELAY) {
2362 /* procfs requests not want stat in wait for avoid deadlock */
2363 req->rq_no_resend = 1;
2364 req->rq_no_delay = 1;
2367 rc = ptlrpc_queue_wait(req);
2371 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2373 GOTO(out, rc = -EPROTO);
2380 ptlrpc_req_finished(req);
2384 static int osc_iocontrol(unsigned int cmd, struct obd_export *exp, int len,
2385 void *karg, void __user *uarg)
2387 struct obd_device *obd = exp->exp_obd;
2388 struct obd_ioctl_data *data = karg;
2392 if (!try_module_get(THIS_MODULE)) {
2393 CERROR("%s: cannot get module '%s'\n", obd->obd_name,
2394 module_name(THIS_MODULE));
2398 case OBD_IOC_CLIENT_RECOVER:
2399 err = ptlrpc_recover_import(obd->u.cli.cl_import,
2400 data->ioc_inlbuf1, 0);
2404 case IOC_OSC_SET_ACTIVE:
2405 err = ptlrpc_set_import_active(obd->u.cli.cl_import,
2408 case OBD_IOC_PING_TARGET:
2409 err = ptlrpc_obd_ping(obd);
2412 CDEBUG(D_INODE, "unrecognised ioctl %#x by %s\n",
2413 cmd, current_comm());
2414 GOTO(out, err = -ENOTTY);
2417 module_put(THIS_MODULE);
2421 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
2422 u32 keylen, void *key,
2423 u32 vallen, void *val,
2424 struct ptlrpc_request_set *set)
2426 struct ptlrpc_request *req;
2427 struct obd_device *obd = exp->exp_obd;
2428 struct obd_import *imp = class_exp2cliimp(exp);
2433 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_SHUTDOWN, 10);
2435 if (KEY_IS(KEY_CHECKSUM)) {
2436 if (vallen != sizeof(int))
2438 exp->exp_obd->u.cli.cl_checksum = (*(int *)val) ? 1 : 0;
2442 if (KEY_IS(KEY_SPTLRPC_CONF)) {
2443 sptlrpc_conf_client_adapt(obd);
2447 if (KEY_IS(KEY_FLUSH_CTX)) {
2448 sptlrpc_import_flush_my_ctx(imp);
2452 if (KEY_IS(KEY_CACHE_SET)) {
2453 struct client_obd *cli = &obd->u.cli;
2455 LASSERT(cli->cl_cache == NULL); /* only once */
2456 cli->cl_cache = (struct cl_client_cache *)val;
2457 cl_cache_incref(cli->cl_cache);
2458 cli->cl_lru_left = &cli->cl_cache->ccc_lru_left;
2460 /* add this osc into entity list */
2461 LASSERT(list_empty(&cli->cl_lru_osc));
2462 spin_lock(&cli->cl_cache->ccc_lru_lock);
2463 list_add(&cli->cl_lru_osc, &cli->cl_cache->ccc_lru);
2464 spin_unlock(&cli->cl_cache->ccc_lru_lock);
2469 if (KEY_IS(KEY_CACHE_LRU_SHRINK)) {
2470 struct client_obd *cli = &obd->u.cli;
2471 long nr = atomic_long_read(&cli->cl_lru_in_list) >> 1;
2472 long target = *(long *)val;
2474 nr = osc_lru_shrink(env, cli, min(nr, target), true);
2479 if (!set && !KEY_IS(KEY_GRANT_SHRINK))
2482 /* We pass all other commands directly to OST. Since nobody calls osc
2483 methods directly and everybody is supposed to go through LOV, we
2484 assume lov checked invalid values for us.
2485 The only recognised values so far are evict_by_nid and mds_conn.
2486 Even if something bad goes through, we'd get a -EINVAL from OST
2489 req = ptlrpc_request_alloc(imp, KEY_IS(KEY_GRANT_SHRINK) ?
2490 &RQF_OST_SET_GRANT_INFO :
2495 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
2496 RCL_CLIENT, keylen);
2497 if (!KEY_IS(KEY_GRANT_SHRINK))
2498 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_VAL,
2499 RCL_CLIENT, vallen);
2500 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SET_INFO);
2502 ptlrpc_request_free(req);
2506 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
2507 memcpy(tmp, key, keylen);
2508 tmp = req_capsule_client_get(&req->rq_pill, KEY_IS(KEY_GRANT_SHRINK) ?
2511 memcpy(tmp, val, vallen);
2513 if (KEY_IS(KEY_GRANT_SHRINK)) {
2514 struct osc_grant_args *aa;
2517 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2518 aa = ptlrpc_req_async_args(req);
2521 ptlrpc_req_finished(req);
2524 *oa = ((struct ost_body *)val)->oa;
2526 req->rq_interpret_reply = osc_shrink_grant_interpret;
2529 ptlrpc_request_set_replen(req);
2530 if (!KEY_IS(KEY_GRANT_SHRINK)) {
2531 LASSERT(set != NULL);
2532 ptlrpc_set_add_req(set, req);
2533 ptlrpc_check_set(NULL, set);
2535 ptlrpcd_add_req(req);
2541 static int osc_reconnect(const struct lu_env *env,
2542 struct obd_export *exp, struct obd_device *obd,
2543 struct obd_uuid *cluuid,
2544 struct obd_connect_data *data,
2547 struct client_obd *cli = &obd->u.cli;
2549 if (data != NULL && (data->ocd_connect_flags & OBD_CONNECT_GRANT)) {
2553 spin_lock(&cli->cl_loi_list_lock);
2554 grant = cli->cl_avail_grant + cli->cl_reserved_grant;
2555 if (data->ocd_connect_flags & OBD_CONNECT_GRANT_PARAM)
2556 grant += cli->cl_dirty_grant;
2558 grant += cli->cl_dirty_pages << PAGE_SHIFT;
2559 data->ocd_grant = grant ? : 2 * cli_brw_size(obd);
2560 lost_grant = cli->cl_lost_grant;
2561 cli->cl_lost_grant = 0;
2562 spin_unlock(&cli->cl_loi_list_lock);
2564 CDEBUG(D_RPCTRACE, "ocd_connect_flags: %#llx ocd_version: %d"
2565 " ocd_grant: %d, lost: %ld.\n", data->ocd_connect_flags,
2566 data->ocd_version, data->ocd_grant, lost_grant);
2572 static int osc_disconnect(struct obd_export *exp)
2574 struct obd_device *obd = class_exp2obd(exp);
2577 rc = client_disconnect_export(exp);
2579 * Initially we put del_shrink_grant before disconnect_export, but it
2580 * causes the following problem if setup (connect) and cleanup
2581 * (disconnect) are tangled together.
2582 * connect p1 disconnect p2
2583 * ptlrpc_connect_import
2584 * ............... class_manual_cleanup
2587 * ptlrpc_connect_interrupt
2589 * add this client to shrink list
2591 * Bang! pinger trigger the shrink.
2592 * So the osc should be disconnected from the shrink list, after we
2593 * are sure the import has been destroyed. BUG18662
2595 if (obd->u.cli.cl_import == NULL)
2596 osc_del_shrink_grant(&obd->u.cli);
2600 static int osc_ldlm_resource_invalidate(struct cfs_hash *hs,
2601 struct cfs_hash_bd *bd, struct hlist_node *hnode, void *arg)
2603 struct lu_env *env = arg;
2604 struct ldlm_resource *res = cfs_hash_object(hs, hnode);
2605 struct ldlm_lock *lock;
2606 struct osc_object *osc = NULL;
2610 list_for_each_entry(lock, &res->lr_granted, l_res_link) {
2611 if (lock->l_ast_data != NULL && osc == NULL) {
2612 osc = lock->l_ast_data;
2613 cl_object_get(osc2cl(osc));
2616 /* clear LDLM_FL_CLEANED flag to make sure it will be canceled
2617 * by the 2nd round of ldlm_namespace_clean() call in
2618 * osc_import_event(). */
2619 ldlm_clear_cleaned(lock);
2624 osc_object_invalidate(env, osc);
2625 cl_object_put(env, osc2cl(osc));
2631 static int osc_import_event(struct obd_device *obd,
2632 struct obd_import *imp,
2633 enum obd_import_event event)
2635 struct client_obd *cli;
2639 LASSERT(imp->imp_obd == obd);
2642 case IMP_EVENT_DISCON: {
2644 spin_lock(&cli->cl_loi_list_lock);
2645 cli->cl_avail_grant = 0;
2646 cli->cl_lost_grant = 0;
2647 spin_unlock(&cli->cl_loi_list_lock);
2650 case IMP_EVENT_INACTIVE: {
2651 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_INACTIVE);
2654 case IMP_EVENT_INVALIDATE: {
2655 struct ldlm_namespace *ns = obd->obd_namespace;
2659 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
2661 env = cl_env_get(&refcheck);
2663 osc_io_unplug(env, &obd->u.cli, NULL);
2665 cfs_hash_for_each_nolock(ns->ns_rs_hash,
2666 osc_ldlm_resource_invalidate,
2668 cl_env_put(env, &refcheck);
2670 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
2675 case IMP_EVENT_ACTIVE: {
2676 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVE);
2679 case IMP_EVENT_OCD: {
2680 struct obd_connect_data *ocd = &imp->imp_connect_data;
2682 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT)
2683 osc_init_grant(&obd->u.cli, ocd);
2686 if (ocd->ocd_connect_flags & OBD_CONNECT_REQPORTAL)
2687 imp->imp_client->cli_request_portal =OST_REQUEST_PORTAL;
2689 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_OCD);
2692 case IMP_EVENT_DEACTIVATE: {
2693 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_DEACTIVATE);
2696 case IMP_EVENT_ACTIVATE: {
2697 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVATE);
2701 CERROR("Unknown import event %d\n", event);
2708 * Determine whether the lock can be canceled before replaying the lock
2709 * during recovery, see bug16774 for detailed information.
2711 * \retval zero the lock can't be canceled
2712 * \retval other ok to cancel
2714 static int osc_cancel_weight(struct ldlm_lock *lock)
2717 * Cancel all unused and granted extent lock.
2719 if (lock->l_resource->lr_type == LDLM_EXTENT &&
2720 lock->l_granted_mode == lock->l_req_mode &&
2721 osc_ldlm_weigh_ast(lock) == 0)
2727 static int brw_queue_work(const struct lu_env *env, void *data)
2729 struct client_obd *cli = data;
2731 CDEBUG(D_CACHE, "Run writeback work for client obd %p.\n", cli);
2733 osc_io_unplug(env, cli, NULL);
2737 int osc_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
2739 struct client_obd *cli = &obd->u.cli;
2740 struct obd_type *type;
2748 rc = ptlrpcd_addref();
2752 rc = client_obd_setup(obd, lcfg);
2754 GOTO(out_ptlrpcd, rc);
2756 handler = ptlrpcd_alloc_work(cli->cl_import, brw_queue_work, cli);
2757 if (IS_ERR(handler))
2758 GOTO(out_client_setup, rc = PTR_ERR(handler));
2759 cli->cl_writeback_work = handler;
2761 handler = ptlrpcd_alloc_work(cli->cl_import, lru_queue_work, cli);
2762 if (IS_ERR(handler))
2763 GOTO(out_ptlrpcd_work, rc = PTR_ERR(handler));
2764 cli->cl_lru_work = handler;
2766 rc = osc_quota_setup(obd);
2768 GOTO(out_ptlrpcd_work, rc);
2770 cli->cl_grant_shrink_interval = GRANT_SHRINK_INTERVAL;
2772 #ifdef CONFIG_PROC_FS
2773 obd->obd_vars = lprocfs_osc_obd_vars;
2775 /* If this is true then both client (osc) and server (osp) are on the
2776 * same node. The osp layer if loaded first will register the osc proc
2777 * directory. In that case this obd_device will be attached its proc
2778 * tree to type->typ_procsym instead of obd->obd_type->typ_procroot. */
2779 type = class_search_type(LUSTRE_OSP_NAME);
2780 if (type && type->typ_procsym) {
2781 obd->obd_proc_entry = lprocfs_register(obd->obd_name,
2783 obd->obd_vars, obd);
2784 if (IS_ERR(obd->obd_proc_entry)) {
2785 rc = PTR_ERR(obd->obd_proc_entry);
2786 CERROR("error %d setting up lprocfs for %s\n", rc,
2788 obd->obd_proc_entry = NULL;
2791 rc = lprocfs_obd_setup(obd);
2794 /* If the basic OSC proc tree construction succeeded then
2795 * lets do the rest. */
2797 lproc_osc_attach_seqstat(obd);
2798 sptlrpc_lprocfs_cliobd_attach(obd);
2799 ptlrpc_lprocfs_register_obd(obd);
2803 * We try to control the total number of requests with a upper limit
2804 * osc_reqpool_maxreqcount. There might be some race which will cause
2805 * over-limit allocation, but it is fine.
2807 req_count = atomic_read(&osc_pool_req_count);
2808 if (req_count < osc_reqpool_maxreqcount) {
2809 adding = cli->cl_max_rpcs_in_flight + 2;
2810 if (req_count + adding > osc_reqpool_maxreqcount)
2811 adding = osc_reqpool_maxreqcount - req_count;
2813 added = ptlrpc_add_rqs_to_pool(osc_rq_pool, adding);
2814 atomic_add(added, &osc_pool_req_count);
2817 INIT_LIST_HEAD(&cli->cl_grant_shrink_list);
2818 ns_register_cancel(obd->obd_namespace, osc_cancel_weight);
2820 spin_lock(&osc_shrink_lock);
2821 list_add_tail(&cli->cl_shrink_list, &osc_shrink_list);
2822 spin_unlock(&osc_shrink_lock);
2827 if (cli->cl_writeback_work != NULL) {
2828 ptlrpcd_destroy_work(cli->cl_writeback_work);
2829 cli->cl_writeback_work = NULL;
2831 if (cli->cl_lru_work != NULL) {
2832 ptlrpcd_destroy_work(cli->cl_lru_work);
2833 cli->cl_lru_work = NULL;
2836 client_obd_cleanup(obd);
2842 static int osc_precleanup(struct obd_device *obd)
2844 struct client_obd *cli = &obd->u.cli;
2848 * for echo client, export may be on zombie list, wait for
2849 * zombie thread to cull it, because cli.cl_import will be
2850 * cleared in client_disconnect_export():
2851 * class_export_destroy() -> obd_cleanup() ->
2852 * echo_device_free() -> echo_client_cleanup() ->
2853 * obd_disconnect() -> osc_disconnect() ->
2854 * client_disconnect_export()
2856 obd_zombie_barrier();
2857 if (cli->cl_writeback_work) {
2858 ptlrpcd_destroy_work(cli->cl_writeback_work);
2859 cli->cl_writeback_work = NULL;
2862 if (cli->cl_lru_work) {
2863 ptlrpcd_destroy_work(cli->cl_lru_work);
2864 cli->cl_lru_work = NULL;
2867 obd_cleanup_client_import(obd);
2868 ptlrpc_lprocfs_unregister_obd(obd);
2869 lprocfs_obd_cleanup(obd);
2873 int osc_cleanup(struct obd_device *obd)
2875 struct client_obd *cli = &obd->u.cli;
2880 spin_lock(&osc_shrink_lock);
2881 list_del(&cli->cl_shrink_list);
2882 spin_unlock(&osc_shrink_lock);
2885 if (cli->cl_cache != NULL) {
2886 LASSERT(atomic_read(&cli->cl_cache->ccc_users) > 0);
2887 spin_lock(&cli->cl_cache->ccc_lru_lock);
2888 list_del_init(&cli->cl_lru_osc);
2889 spin_unlock(&cli->cl_cache->ccc_lru_lock);
2890 cli->cl_lru_left = NULL;
2891 cl_cache_decref(cli->cl_cache);
2892 cli->cl_cache = NULL;
2895 /* free memory of osc quota cache */
2896 osc_quota_cleanup(obd);
2898 rc = client_obd_cleanup(obd);
2904 int osc_process_config_base(struct obd_device *obd, struct lustre_cfg *lcfg)
2906 int rc = class_process_proc_param(PARAM_OSC, obd->obd_vars, lcfg, obd);
2907 return rc > 0 ? 0: rc;
2910 static int osc_process_config(struct obd_device *obd, size_t len, void *buf)
2912 return osc_process_config_base(obd, buf);
2915 static struct obd_ops osc_obd_ops = {
2916 .o_owner = THIS_MODULE,
2917 .o_setup = osc_setup,
2918 .o_precleanup = osc_precleanup,
2919 .o_cleanup = osc_cleanup,
2920 .o_add_conn = client_import_add_conn,
2921 .o_del_conn = client_import_del_conn,
2922 .o_connect = client_connect_import,
2923 .o_reconnect = osc_reconnect,
2924 .o_disconnect = osc_disconnect,
2925 .o_statfs = osc_statfs,
2926 .o_statfs_async = osc_statfs_async,
2927 .o_create = osc_create,
2928 .o_destroy = osc_destroy,
2929 .o_getattr = osc_getattr,
2930 .o_setattr = osc_setattr,
2931 .o_iocontrol = osc_iocontrol,
2932 .o_set_info_async = osc_set_info_async,
2933 .o_import_event = osc_import_event,
2934 .o_process_config = osc_process_config,
2935 .o_quotactl = osc_quotactl,
2938 static struct shrinker *osc_cache_shrinker;
2939 struct list_head osc_shrink_list = LIST_HEAD_INIT(osc_shrink_list);
2940 DEFINE_SPINLOCK(osc_shrink_lock);
2942 #ifndef HAVE_SHRINKER_COUNT
2943 static int osc_cache_shrink(SHRINKER_ARGS(sc, nr_to_scan, gfp_mask))
2945 struct shrink_control scv = {
2946 .nr_to_scan = shrink_param(sc, nr_to_scan),
2947 .gfp_mask = shrink_param(sc, gfp_mask)
2949 #if !defined(HAVE_SHRINKER_WANT_SHRINK_PTR) && !defined(HAVE_SHRINK_CONTROL)
2950 struct shrinker *shrinker = NULL;
2953 (void)osc_cache_shrink_scan(shrinker, &scv);
2955 return osc_cache_shrink_count(shrinker, &scv);
2959 static int __init osc_init(void)
2961 bool enable_proc = true;
2962 struct obd_type *type;
2963 unsigned int reqpool_size;
2964 unsigned int reqsize;
2966 DEF_SHRINKER_VAR(osc_shvar, osc_cache_shrink,
2967 osc_cache_shrink_count, osc_cache_shrink_scan);
2970 /* print an address of _any_ initialized kernel symbol from this
2971 * module, to allow debugging with gdb that doesn't support data
2972 * symbols from modules.*/
2973 CDEBUG(D_INFO, "Lustre OSC module (%p).\n", &osc_caches);
2975 rc = lu_kmem_init(osc_caches);
2979 type = class_search_type(LUSTRE_OSP_NAME);
2980 if (type != NULL && type->typ_procsym != NULL)
2981 enable_proc = false;
2983 rc = class_register_type(&osc_obd_ops, NULL, enable_proc, NULL,
2984 LUSTRE_OSC_NAME, &osc_device_type);
2988 osc_cache_shrinker = set_shrinker(DEFAULT_SEEKS, &osc_shvar);
2990 /* This is obviously too much memory, only prevent overflow here */
2991 if (osc_reqpool_mem_max >= 1 << 12 || osc_reqpool_mem_max == 0)
2992 GOTO(out_type, rc = -EINVAL);
2994 reqpool_size = osc_reqpool_mem_max << 20;
2997 while (reqsize < OST_IO_MAXREQSIZE)
2998 reqsize = reqsize << 1;
3001 * We don't enlarge the request count in OSC pool according to
3002 * cl_max_rpcs_in_flight. The allocation from the pool will only be
3003 * tried after normal allocation failed. So a small OSC pool won't
3004 * cause much performance degression in most of cases.
3006 osc_reqpool_maxreqcount = reqpool_size / reqsize;
3008 atomic_set(&osc_pool_req_count, 0);
3009 osc_rq_pool = ptlrpc_init_rq_pool(0, OST_IO_MAXREQSIZE,
3010 ptlrpc_add_rqs_to_pool);
3012 if (osc_rq_pool != NULL)
3016 class_unregister_type(LUSTRE_OSC_NAME);
3018 lu_kmem_fini(osc_caches);
3023 static void __exit osc_exit(void)
3025 remove_shrinker(osc_cache_shrinker);
3026 class_unregister_type(LUSTRE_OSC_NAME);
3027 lu_kmem_fini(osc_caches);
3028 ptlrpc_free_rq_pool(osc_rq_pool);
3031 MODULE_AUTHOR("OpenSFS, Inc. <http://www.lustre.org/>");
3032 MODULE_DESCRIPTION("Lustre Object Storage Client (OSC)");
3033 MODULE_VERSION(LUSTRE_VERSION_STRING);
3034 MODULE_LICENSE("GPL");
3036 module_init(osc_init);
3037 module_exit(osc_exit);