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, 2017, 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 <lprocfs_status.h>
36 #include <lustre_debug.h>
37 #include <lustre_dlm.h>
38 #include <lustre_fid.h>
39 #include <lustre_ha.h>
40 #include <uapi/linux/lustre/lustre_ioctl.h>
41 #include <lustre_net.h>
42 #include <lustre_obdo.h>
43 #include <uapi/linux/lustre/lustre_param.h>
45 #include <obd_cksum.h>
46 #include <obd_class.h>
47 #include <lustre_osc.h>
49 #include "osc_internal.h"
51 atomic_t osc_pool_req_count;
52 unsigned int osc_reqpool_maxreqcount;
53 struct ptlrpc_request_pool *osc_rq_pool;
55 /* max memory used for request pool, unit is MB */
56 static unsigned int osc_reqpool_mem_max = 5;
57 module_param(osc_reqpool_mem_max, uint, 0444);
59 #define osc_grant_args osc_brw_async_args
61 struct osc_setattr_args {
63 obd_enqueue_update_f sa_upcall;
67 struct osc_fsync_args {
68 struct osc_object *fa_obj;
70 obd_enqueue_update_f fa_upcall;
74 struct osc_ladvise_args {
76 obd_enqueue_update_f la_upcall;
80 static void osc_release_ppga(struct brw_page **ppga, size_t count);
81 static int brw_interpret(const struct lu_env *env, struct ptlrpc_request *req,
84 void osc_pack_req_body(struct ptlrpc_request *req, struct obdo *oa)
86 struct ost_body *body;
88 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
91 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
94 static int osc_getattr(const struct lu_env *env, struct obd_export *exp,
97 struct ptlrpc_request *req;
98 struct ost_body *body;
102 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
106 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
108 ptlrpc_request_free(req);
112 osc_pack_req_body(req, oa);
114 ptlrpc_request_set_replen(req);
116 rc = ptlrpc_queue_wait(req);
120 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
122 GOTO(out, rc = -EPROTO);
124 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
125 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
127 oa->o_blksize = cli_brw_size(exp->exp_obd);
128 oa->o_valid |= OBD_MD_FLBLKSZ;
132 ptlrpc_req_finished(req);
137 static int osc_setattr(const struct lu_env *env, struct obd_export *exp,
140 struct ptlrpc_request *req;
141 struct ost_body *body;
145 LASSERT(oa->o_valid & OBD_MD_FLGROUP);
147 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
151 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
153 ptlrpc_request_free(req);
157 osc_pack_req_body(req, oa);
159 ptlrpc_request_set_replen(req);
161 rc = ptlrpc_queue_wait(req);
165 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
167 GOTO(out, rc = -EPROTO);
169 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
173 ptlrpc_req_finished(req);
178 static int osc_setattr_interpret(const struct lu_env *env,
179 struct ptlrpc_request *req,
180 struct osc_setattr_args *sa, int rc)
182 struct ost_body *body;
188 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
190 GOTO(out, rc = -EPROTO);
192 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, sa->sa_oa,
195 rc = sa->sa_upcall(sa->sa_cookie, rc);
199 int osc_setattr_async(struct obd_export *exp, struct obdo *oa,
200 obd_enqueue_update_f upcall, void *cookie,
201 struct ptlrpc_request_set *rqset)
203 struct ptlrpc_request *req;
204 struct osc_setattr_args *sa;
209 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
213 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
215 ptlrpc_request_free(req);
219 osc_pack_req_body(req, oa);
221 ptlrpc_request_set_replen(req);
223 /* do mds to ost setattr asynchronously */
225 /* Do not wait for response. */
226 ptlrpcd_add_req(req);
228 req->rq_interpret_reply =
229 (ptlrpc_interpterer_t)osc_setattr_interpret;
231 CLASSERT(sizeof(*sa) <= sizeof(req->rq_async_args));
232 sa = ptlrpc_req_async_args(req);
234 sa->sa_upcall = upcall;
235 sa->sa_cookie = cookie;
237 if (rqset == PTLRPCD_SET)
238 ptlrpcd_add_req(req);
240 ptlrpc_set_add_req(rqset, req);
246 static int osc_ladvise_interpret(const struct lu_env *env,
247 struct ptlrpc_request *req,
250 struct osc_ladvise_args *la = arg;
251 struct ost_body *body;
257 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
259 GOTO(out, rc = -EPROTO);
261 *la->la_oa = body->oa;
263 rc = la->la_upcall(la->la_cookie, rc);
268 * If rqset is NULL, do not wait for response. Upcall and cookie could also
269 * be NULL in this case
271 int osc_ladvise_base(struct obd_export *exp, struct obdo *oa,
272 struct ladvise_hdr *ladvise_hdr,
273 obd_enqueue_update_f upcall, void *cookie,
274 struct ptlrpc_request_set *rqset)
276 struct ptlrpc_request *req;
277 struct ost_body *body;
278 struct osc_ladvise_args *la;
280 struct lu_ladvise *req_ladvise;
281 struct lu_ladvise *ladvise = ladvise_hdr->lah_advise;
282 int num_advise = ladvise_hdr->lah_count;
283 struct ladvise_hdr *req_ladvise_hdr;
286 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_LADVISE);
290 req_capsule_set_size(&req->rq_pill, &RMF_OST_LADVISE, RCL_CLIENT,
291 num_advise * sizeof(*ladvise));
292 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_LADVISE);
294 ptlrpc_request_free(req);
297 req->rq_request_portal = OST_IO_PORTAL;
298 ptlrpc_at_set_req_timeout(req);
300 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
302 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
305 req_ladvise_hdr = req_capsule_client_get(&req->rq_pill,
306 &RMF_OST_LADVISE_HDR);
307 memcpy(req_ladvise_hdr, ladvise_hdr, sizeof(*ladvise_hdr));
309 req_ladvise = req_capsule_client_get(&req->rq_pill, &RMF_OST_LADVISE);
310 memcpy(req_ladvise, ladvise, sizeof(*ladvise) * num_advise);
311 ptlrpc_request_set_replen(req);
314 /* Do not wait for response. */
315 ptlrpcd_add_req(req);
319 req->rq_interpret_reply = osc_ladvise_interpret;
320 CLASSERT(sizeof(*la) <= sizeof(req->rq_async_args));
321 la = ptlrpc_req_async_args(req);
323 la->la_upcall = upcall;
324 la->la_cookie = cookie;
326 if (rqset == PTLRPCD_SET)
327 ptlrpcd_add_req(req);
329 ptlrpc_set_add_req(rqset, req);
334 static int osc_create(const struct lu_env *env, struct obd_export *exp,
337 struct ptlrpc_request *req;
338 struct ost_body *body;
343 LASSERT(oa->o_valid & OBD_MD_FLGROUP);
344 LASSERT(fid_seq_is_echo(ostid_seq(&oa->o_oi)));
346 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_CREATE);
348 GOTO(out, rc = -ENOMEM);
350 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_CREATE);
352 ptlrpc_request_free(req);
356 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
359 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
361 ptlrpc_request_set_replen(req);
363 rc = ptlrpc_queue_wait(req);
367 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
369 GOTO(out_req, rc = -EPROTO);
371 CDEBUG(D_INFO, "oa flags %x\n", oa->o_flags);
372 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
374 oa->o_blksize = cli_brw_size(exp->exp_obd);
375 oa->o_valid |= OBD_MD_FLBLKSZ;
377 CDEBUG(D_HA, "transno: %lld\n",
378 lustre_msg_get_transno(req->rq_repmsg));
380 ptlrpc_req_finished(req);
385 int osc_punch_send(struct obd_export *exp, struct obdo *oa,
386 obd_enqueue_update_f upcall, void *cookie)
388 struct ptlrpc_request *req;
389 struct osc_setattr_args *sa;
390 struct obd_import *imp = class_exp2cliimp(exp);
391 struct ost_body *body;
396 req = ptlrpc_request_alloc(imp, &RQF_OST_PUNCH);
400 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_PUNCH);
402 ptlrpc_request_free(req);
406 osc_set_io_portal(req);
408 ptlrpc_at_set_req_timeout(req);
410 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
412 lustre_set_wire_obdo(&imp->imp_connect_data, &body->oa, oa);
414 ptlrpc_request_set_replen(req);
416 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_setattr_interpret;
417 CLASSERT(sizeof(*sa) <= sizeof(req->rq_async_args));
418 sa = ptlrpc_req_async_args(req);
420 sa->sa_upcall = upcall;
421 sa->sa_cookie = cookie;
423 ptlrpcd_add_req(req);
427 EXPORT_SYMBOL(osc_punch_send);
429 static int osc_sync_interpret(const struct lu_env *env,
430 struct ptlrpc_request *req,
433 struct osc_fsync_args *fa = arg;
434 struct ost_body *body;
435 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
436 unsigned long valid = 0;
437 struct cl_object *obj;
443 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
445 CERROR("can't unpack ost_body\n");
446 GOTO(out, rc = -EPROTO);
449 *fa->fa_oa = body->oa;
450 obj = osc2cl(fa->fa_obj);
452 /* Update osc object's blocks attribute */
453 cl_object_attr_lock(obj);
454 if (body->oa.o_valid & OBD_MD_FLBLOCKS) {
455 attr->cat_blocks = body->oa.o_blocks;
460 cl_object_attr_update(env, obj, attr, valid);
461 cl_object_attr_unlock(obj);
464 rc = fa->fa_upcall(fa->fa_cookie, rc);
468 int osc_sync_base(struct osc_object *obj, struct obdo *oa,
469 obd_enqueue_update_f upcall, void *cookie,
470 struct ptlrpc_request_set *rqset)
472 struct obd_export *exp = osc_export(obj);
473 struct ptlrpc_request *req;
474 struct ost_body *body;
475 struct osc_fsync_args *fa;
479 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SYNC);
483 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SYNC);
485 ptlrpc_request_free(req);
489 /* overload the size and blocks fields in the oa with start/end */
490 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
492 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
494 ptlrpc_request_set_replen(req);
495 req->rq_interpret_reply = osc_sync_interpret;
497 CLASSERT(sizeof(*fa) <= sizeof(req->rq_async_args));
498 fa = ptlrpc_req_async_args(req);
501 fa->fa_upcall = upcall;
502 fa->fa_cookie = cookie;
504 if (rqset == PTLRPCD_SET)
505 ptlrpcd_add_req(req);
507 ptlrpc_set_add_req(rqset, req);
512 /* Find and cancel locally locks matched by @mode in the resource found by
513 * @objid. Found locks are added into @cancel list. Returns the amount of
514 * locks added to @cancels list. */
515 static int osc_resource_get_unused(struct obd_export *exp, struct obdo *oa,
516 struct list_head *cancels,
517 enum ldlm_mode mode, __u64 lock_flags)
519 struct ldlm_namespace *ns = exp->exp_obd->obd_namespace;
520 struct ldlm_res_id res_id;
521 struct ldlm_resource *res;
525 /* Return, i.e. cancel nothing, only if ELC is supported (flag in
526 * export) but disabled through procfs (flag in NS).
528 * This distinguishes from a case when ELC is not supported originally,
529 * when we still want to cancel locks in advance and just cancel them
530 * locally, without sending any RPC. */
531 if (exp_connect_cancelset(exp) && !ns_connect_cancelset(ns))
534 ostid_build_res_name(&oa->o_oi, &res_id);
535 res = ldlm_resource_get(ns, NULL, &res_id, 0, 0);
539 LDLM_RESOURCE_ADDREF(res);
540 count = ldlm_cancel_resource_local(res, cancels, NULL, mode,
541 lock_flags, 0, NULL);
542 LDLM_RESOURCE_DELREF(res);
543 ldlm_resource_putref(res);
547 static int osc_destroy_interpret(const struct lu_env *env,
548 struct ptlrpc_request *req, void *data,
551 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
553 atomic_dec(&cli->cl_destroy_in_flight);
554 wake_up(&cli->cl_destroy_waitq);
558 static int osc_can_send_destroy(struct client_obd *cli)
560 if (atomic_inc_return(&cli->cl_destroy_in_flight) <=
561 cli->cl_max_rpcs_in_flight) {
562 /* The destroy request can be sent */
565 if (atomic_dec_return(&cli->cl_destroy_in_flight) <
566 cli->cl_max_rpcs_in_flight) {
568 * The counter has been modified between the two atomic
571 wake_up(&cli->cl_destroy_waitq);
576 static int osc_destroy(const struct lu_env *env, struct obd_export *exp,
579 struct client_obd *cli = &exp->exp_obd->u.cli;
580 struct ptlrpc_request *req;
581 struct ost_body *body;
582 struct list_head cancels = LIST_HEAD_INIT(cancels);
587 CDEBUG(D_INFO, "oa NULL\n");
591 count = osc_resource_get_unused(exp, oa, &cancels, LCK_PW,
592 LDLM_FL_DISCARD_DATA);
594 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_DESTROY);
596 ldlm_lock_list_put(&cancels, l_bl_ast, count);
600 rc = ldlm_prep_elc_req(exp, req, LUSTRE_OST_VERSION, OST_DESTROY,
603 ptlrpc_request_free(req);
607 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
608 ptlrpc_at_set_req_timeout(req);
610 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
612 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
614 ptlrpc_request_set_replen(req);
616 req->rq_interpret_reply = osc_destroy_interpret;
617 if (!osc_can_send_destroy(cli)) {
618 struct l_wait_info lwi = LWI_INTR(LWI_ON_SIGNAL_NOOP, NULL);
621 * Wait until the number of on-going destroy RPCs drops
622 * under max_rpc_in_flight
624 rc = l_wait_event_exclusive(cli->cl_destroy_waitq,
625 osc_can_send_destroy(cli), &lwi);
627 ptlrpc_req_finished(req);
632 /* Do not wait for response */
633 ptlrpcd_add_req(req);
637 static void osc_announce_cached(struct client_obd *cli, struct obdo *oa,
640 u64 bits = OBD_MD_FLBLOCKS | OBD_MD_FLGRANT;
642 LASSERT(!(oa->o_valid & bits));
645 spin_lock(&cli->cl_loi_list_lock);
646 if (OCD_HAS_FLAG(&cli->cl_import->imp_connect_data, GRANT_PARAM))
647 oa->o_dirty = cli->cl_dirty_grant;
649 oa->o_dirty = cli->cl_dirty_pages << PAGE_SHIFT;
650 if (unlikely(cli->cl_dirty_pages - cli->cl_dirty_transit >
651 cli->cl_dirty_max_pages)) {
652 CERROR("dirty %lu - %lu > dirty_max %lu\n",
653 cli->cl_dirty_pages, cli->cl_dirty_transit,
654 cli->cl_dirty_max_pages);
656 } else if (unlikely(atomic_long_read(&obd_dirty_pages) -
657 atomic_long_read(&obd_dirty_transit_pages) >
658 (long)(obd_max_dirty_pages + 1))) {
659 /* The atomic_read() allowing the atomic_inc() are
660 * not covered by a lock thus they may safely race and trip
661 * this CERROR() unless we add in a small fudge factor (+1). */
662 CERROR("%s: dirty %ld - %ld > system dirty_max %ld\n",
663 cli_name(cli), atomic_long_read(&obd_dirty_pages),
664 atomic_long_read(&obd_dirty_transit_pages),
665 obd_max_dirty_pages);
667 } else if (unlikely(cli->cl_dirty_max_pages - cli->cl_dirty_pages >
669 CERROR("dirty %lu - dirty_max %lu too big???\n",
670 cli->cl_dirty_pages, cli->cl_dirty_max_pages);
673 unsigned long nrpages;
675 nrpages = cli->cl_max_pages_per_rpc;
676 nrpages *= cli->cl_max_rpcs_in_flight + 1;
677 nrpages = max(nrpages, cli->cl_dirty_max_pages);
678 oa->o_undirty = nrpages << PAGE_SHIFT;
679 if (OCD_HAS_FLAG(&cli->cl_import->imp_connect_data,
683 /* take extent tax into account when asking for more
685 nrextents = (nrpages + cli->cl_max_extent_pages - 1) /
686 cli->cl_max_extent_pages;
687 oa->o_undirty += nrextents * cli->cl_grant_extent_tax;
690 oa->o_grant = cli->cl_avail_grant + cli->cl_reserved_grant;
691 oa->o_dropped = cli->cl_lost_grant;
692 cli->cl_lost_grant = 0;
693 spin_unlock(&cli->cl_loi_list_lock);
694 CDEBUG(D_CACHE, "dirty: %llu undirty: %u dropped %u grant: %llu\n",
695 oa->o_dirty, oa->o_undirty, oa->o_dropped, oa->o_grant);
698 void osc_update_next_shrink(struct client_obd *cli)
700 cli->cl_next_shrink_grant = ktime_get_seconds() +
701 cli->cl_grant_shrink_interval;
703 CDEBUG(D_CACHE, "next time %lld to shrink grant\n",
704 cli->cl_next_shrink_grant);
707 static void __osc_update_grant(struct client_obd *cli, u64 grant)
709 spin_lock(&cli->cl_loi_list_lock);
710 cli->cl_avail_grant += grant;
711 spin_unlock(&cli->cl_loi_list_lock);
714 static void osc_update_grant(struct client_obd *cli, struct ost_body *body)
716 if (body->oa.o_valid & OBD_MD_FLGRANT) {
717 CDEBUG(D_CACHE, "got %llu extra grant\n", body->oa.o_grant);
718 __osc_update_grant(cli, body->oa.o_grant);
722 static int osc_shrink_grant_interpret(const struct lu_env *env,
723 struct ptlrpc_request *req,
726 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
727 struct obdo *oa = ((struct osc_grant_args *)aa)->aa_oa;
728 struct ost_body *body;
731 __osc_update_grant(cli, oa->o_grant);
735 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
737 osc_update_grant(cli, body);
743 static void osc_shrink_grant_local(struct client_obd *cli, struct obdo *oa)
745 spin_lock(&cli->cl_loi_list_lock);
746 oa->o_grant = cli->cl_avail_grant / 4;
747 cli->cl_avail_grant -= oa->o_grant;
748 spin_unlock(&cli->cl_loi_list_lock);
749 if (!(oa->o_valid & OBD_MD_FLFLAGS)) {
750 oa->o_valid |= OBD_MD_FLFLAGS;
753 oa->o_flags |= OBD_FL_SHRINK_GRANT;
754 osc_update_next_shrink(cli);
757 /* Shrink the current grant, either from some large amount to enough for a
758 * full set of in-flight RPCs, or if we have already shrunk to that limit
759 * then to enough for a single RPC. This avoids keeping more grant than
760 * needed, and avoids shrinking the grant piecemeal. */
761 static int osc_shrink_grant(struct client_obd *cli)
763 __u64 target_bytes = (cli->cl_max_rpcs_in_flight + 1) *
764 (cli->cl_max_pages_per_rpc << PAGE_SHIFT);
766 spin_lock(&cli->cl_loi_list_lock);
767 if (cli->cl_avail_grant <= target_bytes)
768 target_bytes = cli->cl_max_pages_per_rpc << PAGE_SHIFT;
769 spin_unlock(&cli->cl_loi_list_lock);
771 return osc_shrink_grant_to_target(cli, target_bytes);
774 int osc_shrink_grant_to_target(struct client_obd *cli, __u64 target_bytes)
777 struct ost_body *body;
780 spin_lock(&cli->cl_loi_list_lock);
781 /* Don't shrink if we are already above or below the desired limit
782 * We don't want to shrink below a single RPC, as that will negatively
783 * impact block allocation and long-term performance. */
784 if (target_bytes < cli->cl_max_pages_per_rpc << PAGE_SHIFT)
785 target_bytes = cli->cl_max_pages_per_rpc << PAGE_SHIFT;
787 if (target_bytes >= cli->cl_avail_grant) {
788 spin_unlock(&cli->cl_loi_list_lock);
791 spin_unlock(&cli->cl_loi_list_lock);
797 osc_announce_cached(cli, &body->oa, 0);
799 spin_lock(&cli->cl_loi_list_lock);
800 body->oa.o_grant = cli->cl_avail_grant - target_bytes;
801 cli->cl_avail_grant = target_bytes;
802 spin_unlock(&cli->cl_loi_list_lock);
803 if (!(body->oa.o_valid & OBD_MD_FLFLAGS)) {
804 body->oa.o_valid |= OBD_MD_FLFLAGS;
805 body->oa.o_flags = 0;
807 body->oa.o_flags |= OBD_FL_SHRINK_GRANT;
808 osc_update_next_shrink(cli);
810 rc = osc_set_info_async(NULL, cli->cl_import->imp_obd->obd_self_export,
811 sizeof(KEY_GRANT_SHRINK), KEY_GRANT_SHRINK,
812 sizeof(*body), body, NULL);
814 __osc_update_grant(cli, body->oa.o_grant);
819 static int osc_should_shrink_grant(struct client_obd *client)
821 time64_t next_shrink = client->cl_next_shrink_grant;
823 if ((client->cl_import->imp_connect_data.ocd_connect_flags &
824 OBD_CONNECT_GRANT_SHRINK) == 0)
827 if (ktime_get_seconds() >= next_shrink - 5) {
828 /* Get the current RPC size directly, instead of going via:
829 * cli_brw_size(obd->u.cli.cl_import->imp_obd->obd_self_export)
830 * Keep comment here so that it can be found by searching. */
831 int brw_size = client->cl_max_pages_per_rpc << PAGE_SHIFT;
833 if (client->cl_import->imp_state == LUSTRE_IMP_FULL &&
834 client->cl_avail_grant > brw_size)
837 osc_update_next_shrink(client);
842 static int osc_grant_shrink_grant_cb(struct timeout_item *item, void *data)
844 struct client_obd *client;
846 list_for_each_entry(client, &item->ti_obd_list, cl_grant_shrink_list) {
847 if (osc_should_shrink_grant(client))
848 osc_shrink_grant(client);
853 static int osc_add_shrink_grant(struct client_obd *client)
857 rc = ptlrpc_add_timeout_client(client->cl_grant_shrink_interval,
859 osc_grant_shrink_grant_cb, NULL,
860 &client->cl_grant_shrink_list);
862 CERROR("add grant client %s error %d\n", cli_name(client), rc);
865 CDEBUG(D_CACHE, "add grant client %s\n", cli_name(client));
866 osc_update_next_shrink(client);
870 static int osc_del_shrink_grant(struct client_obd *client)
872 return ptlrpc_del_timeout_client(&client->cl_grant_shrink_list,
876 void osc_init_grant(struct client_obd *cli, struct obd_connect_data *ocd)
879 * ocd_grant is the total grant amount we're expect to hold: if we've
880 * been evicted, it's the new avail_grant amount, cl_dirty_pages will
881 * drop to 0 as inflight RPCs fail out; otherwise, it's avail_grant +
884 * race is tolerable here: if we're evicted, but imp_state already
885 * left EVICTED state, then cl_dirty_pages must be 0 already.
887 spin_lock(&cli->cl_loi_list_lock);
888 cli->cl_avail_grant = ocd->ocd_grant;
889 if (cli->cl_import->imp_state != LUSTRE_IMP_EVICTED) {
890 cli->cl_avail_grant -= cli->cl_reserved_grant;
891 if (OCD_HAS_FLAG(ocd, GRANT_PARAM))
892 cli->cl_avail_grant -= cli->cl_dirty_grant;
894 cli->cl_avail_grant -=
895 cli->cl_dirty_pages << PAGE_SHIFT;
898 if (OCD_HAS_FLAG(ocd, GRANT_PARAM)) {
902 /* overhead for each extent insertion */
903 cli->cl_grant_extent_tax = ocd->ocd_grant_tax_kb << 10;
904 /* determine the appropriate chunk size used by osc_extent. */
905 cli->cl_chunkbits = max_t(int, PAGE_SHIFT,
906 ocd->ocd_grant_blkbits);
907 /* max_pages_per_rpc must be chunk aligned */
908 chunk_mask = ~((1 << (cli->cl_chunkbits - PAGE_SHIFT)) - 1);
909 cli->cl_max_pages_per_rpc = (cli->cl_max_pages_per_rpc +
910 ~chunk_mask) & chunk_mask;
911 /* determine maximum extent size, in #pages */
912 size = (u64)ocd->ocd_grant_max_blks << ocd->ocd_grant_blkbits;
913 cli->cl_max_extent_pages = size >> PAGE_SHIFT;
914 if (cli->cl_max_extent_pages == 0)
915 cli->cl_max_extent_pages = 1;
917 cli->cl_grant_extent_tax = 0;
918 cli->cl_chunkbits = PAGE_SHIFT;
919 cli->cl_max_extent_pages = DT_MAX_BRW_PAGES;
921 spin_unlock(&cli->cl_loi_list_lock);
923 CDEBUG(D_CACHE, "%s, setting cl_avail_grant: %ld cl_lost_grant: %ld."
924 "chunk bits: %d cl_max_extent_pages: %d\n",
926 cli->cl_avail_grant, cli->cl_lost_grant, cli->cl_chunkbits,
927 cli->cl_max_extent_pages);
929 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT_SHRINK &&
930 list_empty(&cli->cl_grant_shrink_list))
931 osc_add_shrink_grant(cli);
933 EXPORT_SYMBOL(osc_init_grant);
935 /* We assume that the reason this OSC got a short read is because it read
936 * beyond the end of a stripe file; i.e. lustre is reading a sparse file
937 * via the LOV, and it _knows_ it's reading inside the file, it's just that
938 * this stripe never got written at or beyond this stripe offset yet. */
939 static void handle_short_read(int nob_read, size_t page_count,
940 struct brw_page **pga)
945 /* skip bytes read OK */
946 while (nob_read > 0) {
947 LASSERT (page_count > 0);
949 if (pga[i]->count > nob_read) {
950 /* EOF inside this page */
951 ptr = kmap(pga[i]->pg) +
952 (pga[i]->off & ~PAGE_MASK);
953 memset(ptr + nob_read, 0, pga[i]->count - nob_read);
960 nob_read -= pga[i]->count;
965 /* zero remaining pages */
966 while (page_count-- > 0) {
967 ptr = kmap(pga[i]->pg) + (pga[i]->off & ~PAGE_MASK);
968 memset(ptr, 0, pga[i]->count);
974 static int check_write_rcs(struct ptlrpc_request *req,
975 int requested_nob, int niocount,
976 size_t page_count, struct brw_page **pga)
981 remote_rcs = req_capsule_server_sized_get(&req->rq_pill, &RMF_RCS,
982 sizeof(*remote_rcs) *
984 if (remote_rcs == NULL) {
985 CDEBUG(D_INFO, "Missing/short RC vector on BRW_WRITE reply\n");
989 /* return error if any niobuf was in error */
990 for (i = 0; i < niocount; i++) {
991 if ((int)remote_rcs[i] < 0)
992 return(remote_rcs[i]);
994 if (remote_rcs[i] != 0) {
995 CDEBUG(D_INFO, "rc[%d] invalid (%d) req %p\n",
996 i, remote_rcs[i], req);
1000 if (req->rq_bulk != NULL &&
1001 req->rq_bulk->bd_nob_transferred != requested_nob) {
1002 CERROR("Unexpected # bytes transferred: %d (requested %d)\n",
1003 req->rq_bulk->bd_nob_transferred, requested_nob);
1010 static inline int can_merge_pages(struct brw_page *p1, struct brw_page *p2)
1012 if (p1->flag != p2->flag) {
1013 unsigned mask = ~(OBD_BRW_FROM_GRANT | OBD_BRW_NOCACHE |
1014 OBD_BRW_SYNC | OBD_BRW_ASYNC |
1015 OBD_BRW_NOQUOTA | OBD_BRW_SOFT_SYNC);
1017 /* warn if we try to combine flags that we don't know to be
1018 * safe to combine */
1019 if (unlikely((p1->flag & mask) != (p2->flag & mask))) {
1020 CWARN("Saw flags 0x%x and 0x%x in the same brw, please "
1021 "report this at https://jira.hpdd.intel.com/\n",
1022 p1->flag, p2->flag);
1027 return (p1->off + p1->count == p2->off);
1030 static u32 osc_checksum_bulk(int nob, size_t pg_count,
1031 struct brw_page **pga, int opc,
1032 enum cksum_types cksum_type)
1036 struct cfs_crypto_hash_desc *hdesc;
1037 unsigned int bufsize;
1038 unsigned char cfs_alg = cksum_obd2cfs(cksum_type);
1040 LASSERT(pg_count > 0);
1042 hdesc = cfs_crypto_hash_init(cfs_alg, NULL, 0);
1043 if (IS_ERR(hdesc)) {
1044 CERROR("Unable to initialize checksum hash %s\n",
1045 cfs_crypto_hash_name(cfs_alg));
1046 return PTR_ERR(hdesc);
1049 while (nob > 0 && pg_count > 0) {
1050 unsigned int count = pga[i]->count > nob ? nob : pga[i]->count;
1052 /* corrupt the data before we compute the checksum, to
1053 * simulate an OST->client data error */
1054 if (i == 0 && opc == OST_READ &&
1055 OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_RECEIVE)) {
1056 unsigned char *ptr = kmap(pga[i]->pg);
1057 int off = pga[i]->off & ~PAGE_MASK;
1059 memcpy(ptr + off, "bad1", min_t(typeof(nob), 4, nob));
1062 cfs_crypto_hash_update_page(hdesc, pga[i]->pg,
1063 pga[i]->off & ~PAGE_MASK,
1065 LL_CDEBUG_PAGE(D_PAGE, pga[i]->pg, "off %d\n",
1066 (int)(pga[i]->off & ~PAGE_MASK));
1068 nob -= pga[i]->count;
1073 bufsize = sizeof(cksum);
1074 cfs_crypto_hash_final(hdesc, (unsigned char *)&cksum, &bufsize);
1076 /* For sending we only compute the wrong checksum instead
1077 * of corrupting the data so it is still correct on a redo */
1078 if (opc == OST_WRITE && OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_SEND))
1085 osc_brw_prep_request(int cmd, struct client_obd *cli, struct obdo *oa,
1086 u32 page_count, struct brw_page **pga,
1087 struct ptlrpc_request **reqp, int resend)
1089 struct ptlrpc_request *req;
1090 struct ptlrpc_bulk_desc *desc;
1091 struct ost_body *body;
1092 struct obd_ioobj *ioobj;
1093 struct niobuf_remote *niobuf;
1094 int niocount, i, requested_nob, opc, rc, short_io_size = 0;
1095 struct osc_brw_async_args *aa;
1096 struct req_capsule *pill;
1097 struct brw_page *pg_prev;
1101 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ))
1102 RETURN(-ENOMEM); /* Recoverable */
1103 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ2))
1104 RETURN(-EINVAL); /* Fatal */
1106 if ((cmd & OBD_BRW_WRITE) != 0) {
1108 req = ptlrpc_request_alloc_pool(cli->cl_import,
1110 &RQF_OST_BRW_WRITE);
1113 req = ptlrpc_request_alloc(cli->cl_import, &RQF_OST_BRW_READ);
1118 for (niocount = i = 1; i < page_count; i++) {
1119 if (!can_merge_pages(pga[i - 1], pga[i]))
1123 pill = &req->rq_pill;
1124 req_capsule_set_size(pill, &RMF_OBD_IOOBJ, RCL_CLIENT,
1126 req_capsule_set_size(pill, &RMF_NIOBUF_REMOTE, RCL_CLIENT,
1127 niocount * sizeof(*niobuf));
1129 for (i = 0; i < page_count; i++)
1130 short_io_size += pga[i]->count;
1132 /* Check if we can do a short io. */
1133 if (!(short_io_size <= cli->cl_short_io_bytes && niocount == 1 &&
1134 imp_connect_shortio(cli->cl_import)))
1137 req_capsule_set_size(pill, &RMF_SHORT_IO, RCL_CLIENT,
1138 opc == OST_READ ? 0 : short_io_size);
1139 if (opc == OST_READ)
1140 req_capsule_set_size(pill, &RMF_SHORT_IO, RCL_SERVER,
1143 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, opc);
1145 ptlrpc_request_free(req);
1148 osc_set_io_portal(req);
1150 ptlrpc_at_set_req_timeout(req);
1151 /* ask ptlrpc not to resend on EINPROGRESS since BRWs have their own
1153 req->rq_no_retry_einprogress = 1;
1155 if (short_io_size != 0) {
1157 short_io_buf = NULL;
1161 desc = ptlrpc_prep_bulk_imp(req, page_count,
1162 cli->cl_import->imp_connect_data.ocd_brw_size >> LNET_MTU_BITS,
1163 (opc == OST_WRITE ? PTLRPC_BULK_GET_SOURCE :
1164 PTLRPC_BULK_PUT_SINK) |
1165 PTLRPC_BULK_BUF_KIOV,
1167 &ptlrpc_bulk_kiov_pin_ops);
1170 GOTO(out, rc = -ENOMEM);
1171 /* NB request now owns desc and will free it when it gets freed */
1173 body = req_capsule_client_get(pill, &RMF_OST_BODY);
1174 ioobj = req_capsule_client_get(pill, &RMF_OBD_IOOBJ);
1175 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1176 LASSERT(body != NULL && ioobj != NULL && niobuf != NULL);
1178 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
1180 obdo_to_ioobj(oa, ioobj);
1181 ioobj->ioo_bufcnt = niocount;
1182 /* The high bits of ioo_max_brw tells server _maximum_ number of bulks
1183 * that might be send for this request. The actual number is decided
1184 * when the RPC is finally sent in ptlrpc_register_bulk(). It sends
1185 * "max - 1" for old client compatibility sending "0", and also so the
1186 * the actual maximum is a power-of-two number, not one less. LU-1431 */
1188 ioobj_max_brw_set(ioobj, desc->bd_md_max_brw);
1190 ioobj_max_brw_set(ioobj, 0);
1192 if (short_io_size != 0) {
1193 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1194 body->oa.o_valid |= OBD_MD_FLFLAGS;
1195 body->oa.o_flags = 0;
1197 body->oa.o_flags |= OBD_FL_SHORT_IO;
1198 CDEBUG(D_CACHE, "Using short io for data transfer, size = %d\n",
1200 if (opc == OST_WRITE) {
1201 short_io_buf = req_capsule_client_get(pill,
1203 LASSERT(short_io_buf != NULL);
1207 LASSERT(page_count > 0);
1209 for (requested_nob = i = 0; i < page_count; i++, niobuf++) {
1210 struct brw_page *pg = pga[i];
1211 int poff = pg->off & ~PAGE_MASK;
1213 LASSERT(pg->count > 0);
1214 /* make sure there is no gap in the middle of page array */
1215 LASSERTF(page_count == 1 ||
1216 (ergo(i == 0, poff + pg->count == PAGE_SIZE) &&
1217 ergo(i > 0 && i < page_count - 1,
1218 poff == 0 && pg->count == PAGE_SIZE) &&
1219 ergo(i == page_count - 1, poff == 0)),
1220 "i: %d/%d pg: %p off: %llu, count: %u\n",
1221 i, page_count, pg, pg->off, pg->count);
1222 LASSERTF(i == 0 || pg->off > pg_prev->off,
1223 "i %d p_c %u pg %p [pri %lu ind %lu] off %llu"
1224 " prev_pg %p [pri %lu ind %lu] off %llu\n",
1226 pg->pg, page_private(pg->pg), pg->pg->index, pg->off,
1227 pg_prev->pg, page_private(pg_prev->pg),
1228 pg_prev->pg->index, pg_prev->off);
1229 LASSERT((pga[0]->flag & OBD_BRW_SRVLOCK) ==
1230 (pg->flag & OBD_BRW_SRVLOCK));
1231 if (short_io_size != 0 && opc == OST_WRITE) {
1232 unsigned char *ptr = ll_kmap_atomic(pg->pg, KM_USER0);
1234 LASSERT(short_io_size >= requested_nob + pg->count);
1235 memcpy(short_io_buf + requested_nob,
1238 ll_kunmap_atomic(ptr, KM_USER0);
1239 } else if (short_io_size == 0) {
1240 desc->bd_frag_ops->add_kiov_frag(desc, pg->pg, poff,
1243 requested_nob += pg->count;
1245 if (i > 0 && can_merge_pages(pg_prev, pg)) {
1247 niobuf->rnb_len += pg->count;
1249 niobuf->rnb_offset = pg->off;
1250 niobuf->rnb_len = pg->count;
1251 niobuf->rnb_flags = pg->flag;
1256 LASSERTF((void *)(niobuf - niocount) ==
1257 req_capsule_client_get(&req->rq_pill, &RMF_NIOBUF_REMOTE),
1258 "want %p - real %p\n", req_capsule_client_get(&req->rq_pill,
1259 &RMF_NIOBUF_REMOTE), (void *)(niobuf - niocount));
1261 osc_announce_cached(cli, &body->oa, opc == OST_WRITE ? requested_nob:0);
1263 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1264 body->oa.o_valid |= OBD_MD_FLFLAGS;
1265 body->oa.o_flags = 0;
1267 body->oa.o_flags |= OBD_FL_RECOV_RESEND;
1270 if (osc_should_shrink_grant(cli))
1271 osc_shrink_grant_local(cli, &body->oa);
1273 /* size[REQ_REC_OFF] still sizeof (*body) */
1274 if (opc == OST_WRITE) {
1275 if (cli->cl_checksum &&
1276 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1277 /* store cl_cksum_type in a local variable since
1278 * it can be changed via lprocfs */
1279 enum cksum_types cksum_type = cli->cl_cksum_type;
1281 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0)
1282 body->oa.o_flags = 0;
1284 body->oa.o_flags |= cksum_type_pack(cksum_type);
1285 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1286 body->oa.o_cksum = osc_checksum_bulk(requested_nob,
1290 CDEBUG(D_PAGE, "checksum at write origin: %x\n",
1292 /* save this in 'oa', too, for later checking */
1293 oa->o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1294 oa->o_flags |= cksum_type_pack(cksum_type);
1296 /* clear out the checksum flag, in case this is a
1297 * resend but cl_checksum is no longer set. b=11238 */
1298 oa->o_valid &= ~OBD_MD_FLCKSUM;
1300 oa->o_cksum = body->oa.o_cksum;
1301 /* 1 RC per niobuf */
1302 req_capsule_set_size(pill, &RMF_RCS, RCL_SERVER,
1303 sizeof(__u32) * niocount);
1305 if (cli->cl_checksum &&
1306 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1307 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0)
1308 body->oa.o_flags = 0;
1309 body->oa.o_flags |= cksum_type_pack(cli->cl_cksum_type);
1310 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1313 /* Client cksum has been already copied to wire obdo in previous
1314 * lustre_set_wire_obdo(), and in the case a bulk-read is being
1315 * resent due to cksum error, this will allow Server to
1316 * check+dump pages on its side */
1318 ptlrpc_request_set_replen(req);
1320 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1321 aa = ptlrpc_req_async_args(req);
1323 aa->aa_requested_nob = requested_nob;
1324 aa->aa_nio_count = niocount;
1325 aa->aa_page_count = page_count;
1329 INIT_LIST_HEAD(&aa->aa_oaps);
1332 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1333 CDEBUG(D_RPCTRACE, "brw rpc %p - object "DOSTID" offset %lld<>%lld\n",
1334 req, POSTID(&oa->o_oi), niobuf[0].rnb_offset,
1335 niobuf[niocount - 1].rnb_offset + niobuf[niocount - 1].rnb_len);
1339 ptlrpc_req_finished(req);
1343 char dbgcksum_file_name[PATH_MAX];
1345 static void dump_all_bulk_pages(struct obdo *oa, __u32 page_count,
1346 struct brw_page **pga, __u32 server_cksum,
1355 /* will only keep dump of pages on first error for the same range in
1356 * file/fid, not during the resends/retries. */
1357 snprintf(dbgcksum_file_name, sizeof(dbgcksum_file_name),
1358 "%s-checksum_dump-osc-"DFID":[%llu-%llu]-%x-%x",
1359 (strncmp(libcfs_debug_file_path_arr, "NONE", 4) != 0 ?
1360 libcfs_debug_file_path_arr :
1361 LIBCFS_DEBUG_FILE_PATH_DEFAULT),
1362 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : 0ULL,
1363 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
1364 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
1366 pga[page_count-1]->off + pga[page_count-1]->count - 1,
1367 client_cksum, server_cksum);
1368 filp = filp_open(dbgcksum_file_name,
1369 O_CREAT | O_EXCL | O_WRONLY | O_LARGEFILE, 0600);
1373 CDEBUG(D_INFO, "%s: can't open to dump pages with "
1374 "checksum error: rc = %d\n", dbgcksum_file_name,
1377 CERROR("%s: can't open to dump pages with checksum "
1378 "error: rc = %d\n", dbgcksum_file_name, rc);
1384 for (i = 0; i < page_count; i++) {
1385 len = pga[i]->count;
1386 buf = kmap(pga[i]->pg);
1388 rc = vfs_write(filp, (__force const char __user *)buf,
1391 CERROR("%s: wanted to write %u but got %d "
1392 "error\n", dbgcksum_file_name, len, rc);
1397 CDEBUG(D_INFO, "%s: wrote %d bytes\n",
1398 dbgcksum_file_name, rc);
1404 rc = ll_vfs_fsync_range(filp, 0, LLONG_MAX, 1);
1406 CERROR("%s: sync returns %d\n", dbgcksum_file_name, rc);
1407 filp_close(filp, NULL);
1412 check_write_checksum(struct obdo *oa, const struct lnet_process_id *peer,
1413 __u32 client_cksum, __u32 server_cksum,
1414 struct osc_brw_async_args *aa)
1418 enum cksum_types cksum_type;
1420 if (server_cksum == client_cksum) {
1421 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1425 if (aa->aa_cli->cl_checksum_dump)
1426 dump_all_bulk_pages(oa, aa->aa_page_count, aa->aa_ppga,
1427 server_cksum, client_cksum);
1429 cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
1431 new_cksum = osc_checksum_bulk(aa->aa_requested_nob, aa->aa_page_count,
1432 aa->aa_ppga, OST_WRITE, cksum_type);
1434 if (cksum_type != cksum_type_unpack(aa->aa_oa->o_flags))
1435 msg = "the server did not use the checksum type specified in "
1436 "the original request - likely a protocol problem";
1437 else if (new_cksum == server_cksum)
1438 msg = "changed on the client after we checksummed it - "
1439 "likely false positive due to mmap IO (bug 11742)";
1440 else if (new_cksum == client_cksum)
1441 msg = "changed in transit before arrival at OST";
1443 msg = "changed in transit AND doesn't match the original - "
1444 "likely false positive due to mmap IO (bug 11742)";
1446 LCONSOLE_ERROR_MSG(0x132, "%s: BAD WRITE CHECKSUM: %s: from %s inode "
1447 DFID " object "DOSTID" extent [%llu-%llu], original "
1448 "client csum %x (type %x), server csum %x (type %x),"
1449 " client csum now %x\n",
1450 aa->aa_cli->cl_import->imp_obd->obd_name,
1451 msg, libcfs_nid2str(peer->nid),
1452 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
1453 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
1454 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
1455 POSTID(&oa->o_oi), aa->aa_ppga[0]->off,
1456 aa->aa_ppga[aa->aa_page_count - 1]->off +
1457 aa->aa_ppga[aa->aa_page_count-1]->count - 1,
1458 client_cksum, cksum_type_unpack(aa->aa_oa->o_flags),
1459 server_cksum, cksum_type, new_cksum);
1463 /* Note rc enters this function as number of bytes transferred */
1464 static int osc_brw_fini_request(struct ptlrpc_request *req, int rc)
1466 struct osc_brw_async_args *aa = (void *)&req->rq_async_args;
1467 const struct lnet_process_id *peer =
1468 &req->rq_import->imp_connection->c_peer;
1469 struct client_obd *cli = aa->aa_cli;
1470 struct ost_body *body;
1471 u32 client_cksum = 0;
1474 if (rc < 0 && rc != -EDQUOT) {
1475 DEBUG_REQ(D_INFO, req, "Failed request with rc = %d\n", rc);
1479 LASSERTF(req->rq_repmsg != NULL, "rc = %d\n", rc);
1480 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
1482 DEBUG_REQ(D_INFO, req, "Can't unpack body\n");
1486 /* set/clear over quota flag for a uid/gid/projid */
1487 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE &&
1488 body->oa.o_valid & (OBD_MD_FLALLQUOTA)) {
1489 unsigned qid[LL_MAXQUOTAS] = {
1490 body->oa.o_uid, body->oa.o_gid,
1491 body->oa.o_projid };
1492 CDEBUG(D_QUOTA, "setdq for [%u %u %u] with valid %#llx, flags %x\n",
1493 body->oa.o_uid, body->oa.o_gid, body->oa.o_projid,
1494 body->oa.o_valid, body->oa.o_flags);
1495 osc_quota_setdq(cli, qid, body->oa.o_valid,
1499 osc_update_grant(cli, body);
1504 if (aa->aa_oa->o_valid & OBD_MD_FLCKSUM)
1505 client_cksum = aa->aa_oa->o_cksum; /* save for later */
1507 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1509 CERROR("Unexpected +ve rc %d\n", rc);
1513 if (req->rq_bulk != NULL &&
1514 sptlrpc_cli_unwrap_bulk_write(req, req->rq_bulk))
1517 if ((aa->aa_oa->o_valid & OBD_MD_FLCKSUM) && client_cksum &&
1518 check_write_checksum(&body->oa, peer, client_cksum,
1519 body->oa.o_cksum, aa))
1522 rc = check_write_rcs(req, aa->aa_requested_nob,aa->aa_nio_count,
1523 aa->aa_page_count, aa->aa_ppga);
1527 /* The rest of this function executes only for OST_READs */
1529 if (req->rq_bulk == NULL) {
1530 rc = req_capsule_get_size(&req->rq_pill, &RMF_SHORT_IO,
1532 LASSERT(rc == req->rq_status);
1534 /* if unwrap_bulk failed, return -EAGAIN to retry */
1535 rc = sptlrpc_cli_unwrap_bulk_read(req, req->rq_bulk, rc);
1538 GOTO(out, rc = -EAGAIN);
1540 if (rc > aa->aa_requested_nob) {
1541 CERROR("Unexpected rc %d (%d requested)\n", rc,
1542 aa->aa_requested_nob);
1546 if (req->rq_bulk != NULL && rc != req->rq_bulk->bd_nob_transferred) {
1547 CERROR ("Unexpected rc %d (%d transferred)\n",
1548 rc, req->rq_bulk->bd_nob_transferred);
1552 if (req->rq_bulk == NULL) {
1554 int nob, pg_count, i = 0;
1557 CDEBUG(D_CACHE, "Using short io read, size %d\n", rc);
1558 pg_count = aa->aa_page_count;
1559 buf = req_capsule_server_sized_get(&req->rq_pill, &RMF_SHORT_IO,
1562 while (nob > 0 && pg_count > 0) {
1564 int count = aa->aa_ppga[i]->count > nob ?
1565 nob : aa->aa_ppga[i]->count;
1567 CDEBUG(D_CACHE, "page %p count %d\n",
1568 aa->aa_ppga[i]->pg, count);
1569 ptr = ll_kmap_atomic(aa->aa_ppga[i]->pg, KM_USER0);
1570 memcpy(ptr + (aa->aa_ppga[i]->off & ~PAGE_MASK), buf,
1572 ll_kunmap_atomic((void *) ptr, KM_USER0);
1581 if (rc < aa->aa_requested_nob)
1582 handle_short_read(rc, aa->aa_page_count, aa->aa_ppga);
1584 if (body->oa.o_valid & OBD_MD_FLCKSUM) {
1585 static int cksum_counter;
1586 u32 server_cksum = body->oa.o_cksum;
1589 enum cksum_types cksum_type;
1591 cksum_type = cksum_type_unpack(body->oa.o_valid &OBD_MD_FLFLAGS?
1592 body->oa.o_flags : 0);
1593 client_cksum = osc_checksum_bulk(rc, aa->aa_page_count,
1594 aa->aa_ppga, OST_READ,
1597 if (req->rq_bulk != NULL &&
1598 peer->nid != req->rq_bulk->bd_sender) {
1600 router = libcfs_nid2str(req->rq_bulk->bd_sender);
1603 if (server_cksum != client_cksum) {
1604 struct ost_body *clbody;
1605 u32 page_count = aa->aa_page_count;
1607 clbody = req_capsule_client_get(&req->rq_pill,
1609 if (cli->cl_checksum_dump)
1610 dump_all_bulk_pages(&clbody->oa, page_count,
1611 aa->aa_ppga, server_cksum,
1614 LCONSOLE_ERROR_MSG(0x133, "%s: BAD READ CHECKSUM: from "
1615 "%s%s%s inode "DFID" object "DOSTID
1616 " extent [%llu-%llu], client %x, "
1617 "server %x, cksum_type %x\n",
1618 req->rq_import->imp_obd->obd_name,
1619 libcfs_nid2str(peer->nid),
1621 clbody->oa.o_valid & OBD_MD_FLFID ?
1622 clbody->oa.o_parent_seq : 0ULL,
1623 clbody->oa.o_valid & OBD_MD_FLFID ?
1624 clbody->oa.o_parent_oid : 0,
1625 clbody->oa.o_valid & OBD_MD_FLFID ?
1626 clbody->oa.o_parent_ver : 0,
1627 POSTID(&body->oa.o_oi),
1628 aa->aa_ppga[0]->off,
1629 aa->aa_ppga[page_count-1]->off +
1630 aa->aa_ppga[page_count-1]->count - 1,
1631 client_cksum, server_cksum,
1634 aa->aa_oa->o_cksum = client_cksum;
1638 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1641 } else if (unlikely(client_cksum)) {
1642 static int cksum_missed;
1645 if ((cksum_missed & (-cksum_missed)) == cksum_missed)
1646 CERROR("Checksum %u requested from %s but not sent\n",
1647 cksum_missed, libcfs_nid2str(peer->nid));
1653 lustre_get_wire_obdo(&req->rq_import->imp_connect_data,
1654 aa->aa_oa, &body->oa);
1659 static int osc_brw_redo_request(struct ptlrpc_request *request,
1660 struct osc_brw_async_args *aa, int rc)
1662 struct ptlrpc_request *new_req;
1663 struct osc_brw_async_args *new_aa;
1664 struct osc_async_page *oap;
1667 DEBUG_REQ(rc == -EINPROGRESS ? D_RPCTRACE : D_ERROR, request,
1668 "redo for recoverable error %d", rc);
1670 rc = osc_brw_prep_request(lustre_msg_get_opc(request->rq_reqmsg) ==
1671 OST_WRITE ? OBD_BRW_WRITE : OBD_BRW_READ,
1672 aa->aa_cli, aa->aa_oa, aa->aa_page_count,
1673 aa->aa_ppga, &new_req, 1);
1677 list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
1678 if (oap->oap_request != NULL) {
1679 LASSERTF(request == oap->oap_request,
1680 "request %p != oap_request %p\n",
1681 request, oap->oap_request);
1682 if (oap->oap_interrupted) {
1683 ptlrpc_req_finished(new_req);
1688 /* New request takes over pga and oaps from old request.
1689 * Note that copying a list_head doesn't work, need to move it... */
1691 new_req->rq_interpret_reply = request->rq_interpret_reply;
1692 new_req->rq_async_args = request->rq_async_args;
1693 new_req->rq_commit_cb = request->rq_commit_cb;
1694 /* cap resend delay to the current request timeout, this is similar to
1695 * what ptlrpc does (see after_reply()) */
1696 if (aa->aa_resends > new_req->rq_timeout)
1697 new_req->rq_sent = ktime_get_real_seconds() + new_req->rq_timeout;
1699 new_req->rq_sent = ktime_get_real_seconds() + aa->aa_resends;
1700 new_req->rq_generation_set = 1;
1701 new_req->rq_import_generation = request->rq_import_generation;
1703 new_aa = ptlrpc_req_async_args(new_req);
1705 INIT_LIST_HEAD(&new_aa->aa_oaps);
1706 list_splice_init(&aa->aa_oaps, &new_aa->aa_oaps);
1707 INIT_LIST_HEAD(&new_aa->aa_exts);
1708 list_splice_init(&aa->aa_exts, &new_aa->aa_exts);
1709 new_aa->aa_resends = aa->aa_resends;
1711 list_for_each_entry(oap, &new_aa->aa_oaps, oap_rpc_item) {
1712 if (oap->oap_request) {
1713 ptlrpc_req_finished(oap->oap_request);
1714 oap->oap_request = ptlrpc_request_addref(new_req);
1718 /* XXX: This code will run into problem if we're going to support
1719 * to add a series of BRW RPCs into a self-defined ptlrpc_request_set
1720 * and wait for all of them to be finished. We should inherit request
1721 * set from old request. */
1722 ptlrpcd_add_req(new_req);
1724 DEBUG_REQ(D_INFO, new_req, "new request");
1729 * ugh, we want disk allocation on the target to happen in offset order. we'll
1730 * follow sedgewicks advice and stick to the dead simple shellsort -- it'll do
1731 * fine for our small page arrays and doesn't require allocation. its an
1732 * insertion sort that swaps elements that are strides apart, shrinking the
1733 * stride down until its '1' and the array is sorted.
1735 static void sort_brw_pages(struct brw_page **array, int num)
1738 struct brw_page *tmp;
1742 for (stride = 1; stride < num ; stride = (stride * 3) + 1)
1747 for (i = stride ; i < num ; i++) {
1750 while (j >= stride && array[j - stride]->off > tmp->off) {
1751 array[j] = array[j - stride];
1756 } while (stride > 1);
1759 static void osc_release_ppga(struct brw_page **ppga, size_t count)
1761 LASSERT(ppga != NULL);
1762 OBD_FREE(ppga, sizeof(*ppga) * count);
1765 static int brw_interpret(const struct lu_env *env,
1766 struct ptlrpc_request *req, void *data, int rc)
1768 struct osc_brw_async_args *aa = data;
1769 struct osc_extent *ext;
1770 struct osc_extent *tmp;
1771 struct client_obd *cli = aa->aa_cli;
1772 unsigned long transferred = 0;
1775 rc = osc_brw_fini_request(req, rc);
1776 CDEBUG(D_INODE, "request %p aa %p rc %d\n", req, aa, rc);
1777 /* When server return -EINPROGRESS, client should always retry
1778 * regardless of the number of times the bulk was resent already. */
1779 if (osc_recoverable_error(rc) && !req->rq_no_delay) {
1780 if (req->rq_import_generation !=
1781 req->rq_import->imp_generation) {
1782 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1783 ""DOSTID", rc = %d.\n",
1784 req->rq_import->imp_obd->obd_name,
1785 POSTID(&aa->aa_oa->o_oi), rc);
1786 } else if (rc == -EINPROGRESS ||
1787 client_should_resend(aa->aa_resends, aa->aa_cli)) {
1788 rc = osc_brw_redo_request(req, aa, rc);
1790 CERROR("%s: too many resent retries for object: "
1791 "%llu:%llu, rc = %d.\n",
1792 req->rq_import->imp_obd->obd_name,
1793 POSTID(&aa->aa_oa->o_oi), rc);
1798 else if (rc == -EAGAIN || rc == -EINPROGRESS)
1803 struct obdo *oa = aa->aa_oa;
1804 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
1805 unsigned long valid = 0;
1806 struct cl_object *obj;
1807 struct osc_async_page *last;
1809 last = brw_page2oap(aa->aa_ppga[aa->aa_page_count - 1]);
1810 obj = osc2cl(last->oap_obj);
1812 cl_object_attr_lock(obj);
1813 if (oa->o_valid & OBD_MD_FLBLOCKS) {
1814 attr->cat_blocks = oa->o_blocks;
1815 valid |= CAT_BLOCKS;
1817 if (oa->o_valid & OBD_MD_FLMTIME) {
1818 attr->cat_mtime = oa->o_mtime;
1821 if (oa->o_valid & OBD_MD_FLATIME) {
1822 attr->cat_atime = oa->o_atime;
1825 if (oa->o_valid & OBD_MD_FLCTIME) {
1826 attr->cat_ctime = oa->o_ctime;
1830 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1831 struct lov_oinfo *loi = cl2osc(obj)->oo_oinfo;
1832 loff_t last_off = last->oap_count + last->oap_obj_off +
1835 /* Change file size if this is an out of quota or
1836 * direct IO write and it extends the file size */
1837 if (loi->loi_lvb.lvb_size < last_off) {
1838 attr->cat_size = last_off;
1841 /* Extend KMS if it's not a lockless write */
1842 if (loi->loi_kms < last_off &&
1843 oap2osc_page(last)->ops_srvlock == 0) {
1844 attr->cat_kms = last_off;
1850 cl_object_attr_update(env, obj, attr, valid);
1851 cl_object_attr_unlock(obj);
1853 OBDO_FREE(aa->aa_oa);
1855 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE && rc == 0)
1856 osc_inc_unstable_pages(req);
1858 list_for_each_entry_safe(ext, tmp, &aa->aa_exts, oe_link) {
1859 list_del_init(&ext->oe_link);
1860 osc_extent_finish(env, ext, 1,
1861 rc && req->rq_no_delay ? -EWOULDBLOCK : rc);
1863 LASSERT(list_empty(&aa->aa_exts));
1864 LASSERT(list_empty(&aa->aa_oaps));
1866 transferred = (req->rq_bulk == NULL ? /* short io */
1867 aa->aa_requested_nob :
1868 req->rq_bulk->bd_nob_transferred);
1870 osc_release_ppga(aa->aa_ppga, aa->aa_page_count);
1871 ptlrpc_lprocfs_brw(req, transferred);
1873 spin_lock(&cli->cl_loi_list_lock);
1874 /* We need to decrement before osc_ap_completion->osc_wake_cache_waiters
1875 * is called so we know whether to go to sync BRWs or wait for more
1876 * RPCs to complete */
1877 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE)
1878 cli->cl_w_in_flight--;
1880 cli->cl_r_in_flight--;
1881 osc_wake_cache_waiters(cli);
1882 spin_unlock(&cli->cl_loi_list_lock);
1884 osc_io_unplug(env, cli, NULL);
1888 static void brw_commit(struct ptlrpc_request *req)
1890 /* If osc_inc_unstable_pages (via osc_extent_finish) races with
1891 * this called via the rq_commit_cb, I need to ensure
1892 * osc_dec_unstable_pages is still called. Otherwise unstable
1893 * pages may be leaked. */
1894 spin_lock(&req->rq_lock);
1895 if (likely(req->rq_unstable)) {
1896 req->rq_unstable = 0;
1897 spin_unlock(&req->rq_lock);
1899 osc_dec_unstable_pages(req);
1901 req->rq_committed = 1;
1902 spin_unlock(&req->rq_lock);
1907 * Build an RPC by the list of extent @ext_list. The caller must ensure
1908 * that the total pages in this list are NOT over max pages per RPC.
1909 * Extents in the list must be in OES_RPC state.
1911 int osc_build_rpc(const struct lu_env *env, struct client_obd *cli,
1912 struct list_head *ext_list, int cmd)
1914 struct ptlrpc_request *req = NULL;
1915 struct osc_extent *ext;
1916 struct brw_page **pga = NULL;
1917 struct osc_brw_async_args *aa = NULL;
1918 struct obdo *oa = NULL;
1919 struct osc_async_page *oap;
1920 struct osc_object *obj = NULL;
1921 struct cl_req_attr *crattr = NULL;
1922 loff_t starting_offset = OBD_OBJECT_EOF;
1923 loff_t ending_offset = 0;
1927 bool soft_sync = false;
1928 bool interrupted = false;
1929 bool ndelay = false;
1933 __u32 layout_version = 0;
1934 struct list_head rpc_list = LIST_HEAD_INIT(rpc_list);
1935 struct ost_body *body;
1937 LASSERT(!list_empty(ext_list));
1939 /* add pages into rpc_list to build BRW rpc */
1940 list_for_each_entry(ext, ext_list, oe_link) {
1941 LASSERT(ext->oe_state == OES_RPC);
1942 mem_tight |= ext->oe_memalloc;
1943 grant += ext->oe_grants;
1944 page_count += ext->oe_nr_pages;
1945 layout_version = MAX(layout_version, ext->oe_layout_version);
1950 soft_sync = osc_over_unstable_soft_limit(cli);
1952 mpflag = cfs_memory_pressure_get_and_set();
1954 OBD_ALLOC(pga, sizeof(*pga) * page_count);
1956 GOTO(out, rc = -ENOMEM);
1960 GOTO(out, rc = -ENOMEM);
1963 list_for_each_entry(ext, ext_list, oe_link) {
1964 list_for_each_entry(oap, &ext->oe_pages, oap_pending_item) {
1966 oap->oap_brw_flags |= OBD_BRW_MEMALLOC;
1968 oap->oap_brw_flags |= OBD_BRW_SOFT_SYNC;
1969 pga[i] = &oap->oap_brw_page;
1970 pga[i]->off = oap->oap_obj_off + oap->oap_page_off;
1973 list_add_tail(&oap->oap_rpc_item, &rpc_list);
1974 if (starting_offset == OBD_OBJECT_EOF ||
1975 starting_offset > oap->oap_obj_off)
1976 starting_offset = oap->oap_obj_off;
1978 LASSERT(oap->oap_page_off == 0);
1979 if (ending_offset < oap->oap_obj_off + oap->oap_count)
1980 ending_offset = oap->oap_obj_off +
1983 LASSERT(oap->oap_page_off + oap->oap_count ==
1985 if (oap->oap_interrupted)
1992 /* first page in the list */
1993 oap = list_entry(rpc_list.next, typeof(*oap), oap_rpc_item);
1995 crattr = &osc_env_info(env)->oti_req_attr;
1996 memset(crattr, 0, sizeof(*crattr));
1997 crattr->cra_type = (cmd & OBD_BRW_WRITE) ? CRT_WRITE : CRT_READ;
1998 crattr->cra_flags = ~0ULL;
1999 crattr->cra_page = oap2cl_page(oap);
2000 crattr->cra_oa = oa;
2001 cl_req_attr_set(env, osc2cl(obj), crattr);
2003 if (cmd == OBD_BRW_WRITE) {
2004 oa->o_grant_used = grant;
2005 if (layout_version > 0) {
2006 CDEBUG(D_LAYOUT, DFID": write with layout version %u\n",
2007 PFID(&oa->o_oi.oi_fid), layout_version);
2009 oa->o_layout_version = layout_version;
2010 oa->o_valid |= OBD_MD_LAYOUT_VERSION;
2014 sort_brw_pages(pga, page_count);
2015 rc = osc_brw_prep_request(cmd, cli, oa, page_count, pga, &req, 0);
2017 CERROR("prep_req failed: %d\n", rc);
2021 req->rq_commit_cb = brw_commit;
2022 req->rq_interpret_reply = brw_interpret;
2023 req->rq_memalloc = mem_tight != 0;
2024 oap->oap_request = ptlrpc_request_addref(req);
2025 if (interrupted && !req->rq_intr)
2026 ptlrpc_mark_interrupted(req);
2028 req->rq_no_resend = req->rq_no_delay = 1;
2029 /* probably set a shorter timeout value.
2030 * to handle ETIMEDOUT in brw_interpret() correctly. */
2031 /* lustre_msg_set_timeout(req, req->rq_timeout / 2); */
2034 /* Need to update the timestamps after the request is built in case
2035 * we race with setattr (locally or in queue at OST). If OST gets
2036 * later setattr before earlier BRW (as determined by the request xid),
2037 * the OST will not use BRW timestamps. Sadly, there is no obvious
2038 * way to do this in a single call. bug 10150 */
2039 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
2040 crattr->cra_oa = &body->oa;
2041 crattr->cra_flags = OBD_MD_FLMTIME|OBD_MD_FLCTIME|OBD_MD_FLATIME;
2042 cl_req_attr_set(env, osc2cl(obj), crattr);
2043 lustre_msg_set_jobid(req->rq_reqmsg, crattr->cra_jobid);
2045 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2046 aa = ptlrpc_req_async_args(req);
2047 INIT_LIST_HEAD(&aa->aa_oaps);
2048 list_splice_init(&rpc_list, &aa->aa_oaps);
2049 INIT_LIST_HEAD(&aa->aa_exts);
2050 list_splice_init(ext_list, &aa->aa_exts);
2052 spin_lock(&cli->cl_loi_list_lock);
2053 starting_offset >>= PAGE_SHIFT;
2054 if (cmd == OBD_BRW_READ) {
2055 cli->cl_r_in_flight++;
2056 lprocfs_oh_tally_log2(&cli->cl_read_page_hist, page_count);
2057 lprocfs_oh_tally(&cli->cl_read_rpc_hist, cli->cl_r_in_flight);
2058 lprocfs_oh_tally_log2(&cli->cl_read_offset_hist,
2059 starting_offset + 1);
2061 cli->cl_w_in_flight++;
2062 lprocfs_oh_tally_log2(&cli->cl_write_page_hist, page_count);
2063 lprocfs_oh_tally(&cli->cl_write_rpc_hist, cli->cl_w_in_flight);
2064 lprocfs_oh_tally_log2(&cli->cl_write_offset_hist,
2065 starting_offset + 1);
2067 spin_unlock(&cli->cl_loi_list_lock);
2069 DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %ur/%uw in flight",
2070 page_count, aa, cli->cl_r_in_flight,
2071 cli->cl_w_in_flight);
2072 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_DELAY_IO, cfs_fail_val);
2074 ptlrpcd_add_req(req);
2080 cfs_memory_pressure_restore(mpflag);
2083 LASSERT(req == NULL);
2088 OBD_FREE(pga, sizeof(*pga) * page_count);
2089 /* this should happen rarely and is pretty bad, it makes the
2090 * pending list not follow the dirty order */
2091 while (!list_empty(ext_list)) {
2092 ext = list_entry(ext_list->next, struct osc_extent,
2094 list_del_init(&ext->oe_link);
2095 osc_extent_finish(env, ext, 0, rc);
2101 static int osc_set_lock_data(struct ldlm_lock *lock, void *data)
2105 LASSERT(lock != NULL);
2107 lock_res_and_lock(lock);
2109 if (lock->l_ast_data == NULL)
2110 lock->l_ast_data = data;
2111 if (lock->l_ast_data == data)
2114 unlock_res_and_lock(lock);
2119 int osc_enqueue_fini(struct ptlrpc_request *req, osc_enqueue_upcall_f upcall,
2120 void *cookie, struct lustre_handle *lockh,
2121 enum ldlm_mode mode, __u64 *flags, bool speculative,
2124 bool intent = *flags & LDLM_FL_HAS_INTENT;
2128 /* The request was created before ldlm_cli_enqueue call. */
2129 if (intent && errcode == ELDLM_LOCK_ABORTED) {
2130 struct ldlm_reply *rep;
2132 rep = req_capsule_server_get(&req->rq_pill, &RMF_DLM_REP);
2133 LASSERT(rep != NULL);
2135 rep->lock_policy_res1 =
2136 ptlrpc_status_ntoh(rep->lock_policy_res1);
2137 if (rep->lock_policy_res1)
2138 errcode = rep->lock_policy_res1;
2140 *flags |= LDLM_FL_LVB_READY;
2141 } else if (errcode == ELDLM_OK) {
2142 *flags |= LDLM_FL_LVB_READY;
2145 /* Call the update callback. */
2146 rc = (*upcall)(cookie, lockh, errcode);
2148 /* release the reference taken in ldlm_cli_enqueue() */
2149 if (errcode == ELDLM_LOCK_MATCHED)
2151 if (errcode == ELDLM_OK && lustre_handle_is_used(lockh))
2152 ldlm_lock_decref(lockh, mode);
2157 int osc_enqueue_interpret(const struct lu_env *env, struct ptlrpc_request *req,
2158 struct osc_enqueue_args *aa, int rc)
2160 struct ldlm_lock *lock;
2161 struct lustre_handle *lockh = &aa->oa_lockh;
2162 enum ldlm_mode mode = aa->oa_mode;
2163 struct ost_lvb *lvb = aa->oa_lvb;
2164 __u32 lvb_len = sizeof(*lvb);
2169 /* ldlm_cli_enqueue is holding a reference on the lock, so it must
2171 lock = ldlm_handle2lock(lockh);
2172 LASSERTF(lock != NULL,
2173 "lockh %#llx, req %p, aa %p - client evicted?\n",
2174 lockh->cookie, req, aa);
2176 /* Take an additional reference so that a blocking AST that
2177 * ldlm_cli_enqueue_fini() might post for a failed lock, is guaranteed
2178 * to arrive after an upcall has been executed by
2179 * osc_enqueue_fini(). */
2180 ldlm_lock_addref(lockh, mode);
2182 /* Let cl_lock_state_wait fail with -ERESTARTSYS to unuse sublocks. */
2183 OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_ENQUEUE_HANG, 2);
2185 /* Let CP AST to grant the lock first. */
2186 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_ENQ_RACE, 1);
2188 if (aa->oa_speculative) {
2189 LASSERT(aa->oa_lvb == NULL);
2190 LASSERT(aa->oa_flags == NULL);
2191 aa->oa_flags = &flags;
2194 /* Complete obtaining the lock procedure. */
2195 rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, aa->oa_type, 1,
2196 aa->oa_mode, aa->oa_flags, lvb, lvb_len,
2198 /* Complete osc stuff. */
2199 rc = osc_enqueue_fini(req, aa->oa_upcall, aa->oa_cookie, lockh, mode,
2200 aa->oa_flags, aa->oa_speculative, rc);
2202 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_CANCEL_RACE, 10);
2204 ldlm_lock_decref(lockh, mode);
2205 LDLM_LOCK_PUT(lock);
2209 struct ptlrpc_request_set *PTLRPCD_SET = (void *)1;
2211 /* When enqueuing asynchronously, locks are not ordered, we can obtain a lock
2212 * from the 2nd OSC before a lock from the 1st one. This does not deadlock with
2213 * other synchronous requests, however keeping some locks and trying to obtain
2214 * others may take a considerable amount of time in a case of ost failure; and
2215 * when other sync requests do not get released lock from a client, the client
2216 * is evicted from the cluster -- such scenarious make the life difficult, so
2217 * release locks just after they are obtained. */
2218 int osc_enqueue_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2219 __u64 *flags, union ldlm_policy_data *policy,
2220 struct ost_lvb *lvb, int kms_valid,
2221 osc_enqueue_upcall_f upcall, void *cookie,
2222 struct ldlm_enqueue_info *einfo,
2223 struct ptlrpc_request_set *rqset, int async,
2226 struct obd_device *obd = exp->exp_obd;
2227 struct lustre_handle lockh = { 0 };
2228 struct ptlrpc_request *req = NULL;
2229 int intent = *flags & LDLM_FL_HAS_INTENT;
2230 __u64 match_flags = *flags;
2231 enum ldlm_mode mode;
2235 /* Filesystem lock extents are extended to page boundaries so that
2236 * dealing with the page cache is a little smoother. */
2237 policy->l_extent.start -= policy->l_extent.start & ~PAGE_MASK;
2238 policy->l_extent.end |= ~PAGE_MASK;
2241 * kms is not valid when either object is completely fresh (so that no
2242 * locks are cached), or object was evicted. In the latter case cached
2243 * lock cannot be used, because it would prime inode state with
2244 * potentially stale LVB.
2249 /* Next, search for already existing extent locks that will cover us */
2250 /* If we're trying to read, we also search for an existing PW lock. The
2251 * VFS and page cache already protect us locally, so lots of readers/
2252 * writers can share a single PW lock.
2254 * There are problems with conversion deadlocks, so instead of
2255 * converting a read lock to a write lock, we'll just enqueue a new
2258 * At some point we should cancel the read lock instead of making them
2259 * send us a blocking callback, but there are problems with canceling
2260 * locks out from other users right now, too. */
2261 mode = einfo->ei_mode;
2262 if (einfo->ei_mode == LCK_PR)
2264 /* Normal lock requests must wait for the LVB to be ready before
2265 * matching a lock; speculative lock requests do not need to,
2266 * because they will not actually use the lock. */
2268 match_flags |= LDLM_FL_LVB_READY;
2270 match_flags |= LDLM_FL_BLOCK_GRANTED;
2271 mode = ldlm_lock_match(obd->obd_namespace, match_flags, res_id,
2272 einfo->ei_type, policy, mode, &lockh, 0);
2274 struct ldlm_lock *matched;
2276 if (*flags & LDLM_FL_TEST_LOCK)
2279 matched = ldlm_handle2lock(&lockh);
2281 /* This DLM lock request is speculative, and does not
2282 * have an associated IO request. Therefore if there
2283 * is already a DLM lock, it wll just inform the
2284 * caller to cancel the request for this stripe.*/
2285 lock_res_and_lock(matched);
2286 if (ldlm_extent_equal(&policy->l_extent,
2287 &matched->l_policy_data.l_extent))
2291 unlock_res_and_lock(matched);
2293 ldlm_lock_decref(&lockh, mode);
2294 LDLM_LOCK_PUT(matched);
2296 } else if (osc_set_lock_data(matched, einfo->ei_cbdata)) {
2297 *flags |= LDLM_FL_LVB_READY;
2299 /* We already have a lock, and it's referenced. */
2300 (*upcall)(cookie, &lockh, ELDLM_LOCK_MATCHED);
2302 ldlm_lock_decref(&lockh, mode);
2303 LDLM_LOCK_PUT(matched);
2306 ldlm_lock_decref(&lockh, mode);
2307 LDLM_LOCK_PUT(matched);
2312 if (*flags & (LDLM_FL_TEST_LOCK | LDLM_FL_MATCH_LOCK))
2316 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2317 &RQF_LDLM_ENQUEUE_LVB);
2321 rc = ldlm_prep_enqueue_req(exp, req, NULL, 0);
2323 ptlrpc_request_free(req);
2327 req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER,
2329 ptlrpc_request_set_replen(req);
2332 /* users of osc_enqueue() can pass this flag for ldlm_lock_match() */
2333 *flags &= ~LDLM_FL_BLOCK_GRANTED;
2335 rc = ldlm_cli_enqueue(exp, &req, einfo, res_id, policy, flags, lvb,
2336 sizeof(*lvb), LVB_T_OST, &lockh, async);
2339 struct osc_enqueue_args *aa;
2340 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2341 aa = ptlrpc_req_async_args(req);
2343 aa->oa_mode = einfo->ei_mode;
2344 aa->oa_type = einfo->ei_type;
2345 lustre_handle_copy(&aa->oa_lockh, &lockh);
2346 aa->oa_upcall = upcall;
2347 aa->oa_cookie = cookie;
2348 aa->oa_speculative = speculative;
2350 aa->oa_flags = flags;
2353 /* speculative locks are essentially to enqueue
2354 * a DLM lock in advance, so we don't care
2355 * about the result of the enqueue. */
2357 aa->oa_flags = NULL;
2360 req->rq_interpret_reply =
2361 (ptlrpc_interpterer_t)osc_enqueue_interpret;
2362 if (rqset == PTLRPCD_SET)
2363 ptlrpcd_add_req(req);
2365 ptlrpc_set_add_req(rqset, req);
2366 } else if (intent) {
2367 ptlrpc_req_finished(req);
2372 rc = osc_enqueue_fini(req, upcall, cookie, &lockh, einfo->ei_mode,
2373 flags, speculative, rc);
2375 ptlrpc_req_finished(req);
2380 int osc_match_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2381 enum ldlm_type type, union ldlm_policy_data *policy,
2382 enum ldlm_mode mode, __u64 *flags, void *data,
2383 struct lustre_handle *lockh, int unref)
2385 struct obd_device *obd = exp->exp_obd;
2386 __u64 lflags = *flags;
2390 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_MATCH))
2393 /* Filesystem lock extents are extended to page boundaries so that
2394 * dealing with the page cache is a little smoother */
2395 policy->l_extent.start -= policy->l_extent.start & ~PAGE_MASK;
2396 policy->l_extent.end |= ~PAGE_MASK;
2398 /* Next, search for already existing extent locks that will cover us */
2399 /* If we're trying to read, we also search for an existing PW lock. The
2400 * VFS and page cache already protect us locally, so lots of readers/
2401 * writers can share a single PW lock. */
2405 rc = ldlm_lock_match(obd->obd_namespace, lflags,
2406 res_id, type, policy, rc, lockh, unref);
2407 if (rc == 0 || lflags & LDLM_FL_TEST_LOCK)
2411 struct ldlm_lock *lock = ldlm_handle2lock(lockh);
2413 LASSERT(lock != NULL);
2414 if (!osc_set_lock_data(lock, data)) {
2415 ldlm_lock_decref(lockh, rc);
2418 LDLM_LOCK_PUT(lock);
2423 static int osc_statfs_interpret(const struct lu_env *env,
2424 struct ptlrpc_request *req,
2425 struct osc_async_args *aa, int rc)
2427 struct obd_statfs *msfs;
2431 /* The request has in fact never been sent
2432 * due to issues at a higher level (LOV).
2433 * Exit immediately since the caller is
2434 * aware of the problem and takes care
2435 * of the clean up */
2438 if ((rc == -ENOTCONN || rc == -EAGAIN) &&
2439 (aa->aa_oi->oi_flags & OBD_STATFS_NODELAY))
2445 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2447 GOTO(out, rc = -EPROTO);
2450 *aa->aa_oi->oi_osfs = *msfs;
2452 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
2456 static int osc_statfs_async(struct obd_export *exp,
2457 struct obd_info *oinfo, __u64 max_age,
2458 struct ptlrpc_request_set *rqset)
2460 struct obd_device *obd = class_exp2obd(exp);
2461 struct ptlrpc_request *req;
2462 struct osc_async_args *aa;
2466 /* We could possibly pass max_age in the request (as an absolute
2467 * timestamp or a "seconds.usec ago") so the target can avoid doing
2468 * extra calls into the filesystem if that isn't necessary (e.g.
2469 * during mount that would help a bit). Having relative timestamps
2470 * is not so great if request processing is slow, while absolute
2471 * timestamps are not ideal because they need time synchronization. */
2472 req = ptlrpc_request_alloc(obd->u.cli.cl_import, &RQF_OST_STATFS);
2476 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2478 ptlrpc_request_free(req);
2481 ptlrpc_request_set_replen(req);
2482 req->rq_request_portal = OST_CREATE_PORTAL;
2483 ptlrpc_at_set_req_timeout(req);
2485 if (oinfo->oi_flags & OBD_STATFS_NODELAY) {
2486 /* procfs requests not want stat in wait for avoid deadlock */
2487 req->rq_no_resend = 1;
2488 req->rq_no_delay = 1;
2491 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_statfs_interpret;
2492 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2493 aa = ptlrpc_req_async_args(req);
2496 ptlrpc_set_add_req(rqset, req);
2500 static int osc_statfs(const struct lu_env *env, struct obd_export *exp,
2501 struct obd_statfs *osfs, __u64 max_age, __u32 flags)
2503 struct obd_device *obd = class_exp2obd(exp);
2504 struct obd_statfs *msfs;
2505 struct ptlrpc_request *req;
2506 struct obd_import *imp = NULL;
2510 /*Since the request might also come from lprocfs, so we need
2511 *sync this with client_disconnect_export Bug15684*/
2512 down_read(&obd->u.cli.cl_sem);
2513 if (obd->u.cli.cl_import)
2514 imp = class_import_get(obd->u.cli.cl_import);
2515 up_read(&obd->u.cli.cl_sem);
2519 /* We could possibly pass max_age in the request (as an absolute
2520 * timestamp or a "seconds.usec ago") so the target can avoid doing
2521 * extra calls into the filesystem if that isn't necessary (e.g.
2522 * during mount that would help a bit). Having relative timestamps
2523 * is not so great if request processing is slow, while absolute
2524 * timestamps are not ideal because they need time synchronization. */
2525 req = ptlrpc_request_alloc(imp, &RQF_OST_STATFS);
2527 class_import_put(imp);
2532 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2534 ptlrpc_request_free(req);
2537 ptlrpc_request_set_replen(req);
2538 req->rq_request_portal = OST_CREATE_PORTAL;
2539 ptlrpc_at_set_req_timeout(req);
2541 if (flags & OBD_STATFS_NODELAY) {
2542 /* procfs requests not want stat in wait for avoid deadlock */
2543 req->rq_no_resend = 1;
2544 req->rq_no_delay = 1;
2547 rc = ptlrpc_queue_wait(req);
2551 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2553 GOTO(out, rc = -EPROTO);
2560 ptlrpc_req_finished(req);
2564 static int osc_iocontrol(unsigned int cmd, struct obd_export *exp, int len,
2565 void *karg, void __user *uarg)
2567 struct obd_device *obd = exp->exp_obd;
2568 struct obd_ioctl_data *data = karg;
2572 if (!try_module_get(THIS_MODULE)) {
2573 CERROR("%s: cannot get module '%s'\n", obd->obd_name,
2574 module_name(THIS_MODULE));
2578 case OBD_IOC_CLIENT_RECOVER:
2579 err = ptlrpc_recover_import(obd->u.cli.cl_import,
2580 data->ioc_inlbuf1, 0);
2584 case IOC_OSC_SET_ACTIVE:
2585 err = ptlrpc_set_import_active(obd->u.cli.cl_import,
2588 case OBD_IOC_PING_TARGET:
2589 err = ptlrpc_obd_ping(obd);
2592 CDEBUG(D_INODE, "unrecognised ioctl %#x by %s\n",
2593 cmd, current_comm());
2594 GOTO(out, err = -ENOTTY);
2597 module_put(THIS_MODULE);
2601 int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
2602 u32 keylen, void *key, u32 vallen, void *val,
2603 struct ptlrpc_request_set *set)
2605 struct ptlrpc_request *req;
2606 struct obd_device *obd = exp->exp_obd;
2607 struct obd_import *imp = class_exp2cliimp(exp);
2612 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_SHUTDOWN, 10);
2614 if (KEY_IS(KEY_CHECKSUM)) {
2615 if (vallen != sizeof(int))
2617 exp->exp_obd->u.cli.cl_checksum = (*(int *)val) ? 1 : 0;
2621 if (KEY_IS(KEY_SPTLRPC_CONF)) {
2622 sptlrpc_conf_client_adapt(obd);
2626 if (KEY_IS(KEY_FLUSH_CTX)) {
2627 sptlrpc_import_flush_my_ctx(imp);
2631 if (KEY_IS(KEY_CACHE_SET)) {
2632 struct client_obd *cli = &obd->u.cli;
2634 LASSERT(cli->cl_cache == NULL); /* only once */
2635 cli->cl_cache = (struct cl_client_cache *)val;
2636 cl_cache_incref(cli->cl_cache);
2637 cli->cl_lru_left = &cli->cl_cache->ccc_lru_left;
2639 /* add this osc into entity list */
2640 LASSERT(list_empty(&cli->cl_lru_osc));
2641 spin_lock(&cli->cl_cache->ccc_lru_lock);
2642 list_add(&cli->cl_lru_osc, &cli->cl_cache->ccc_lru);
2643 spin_unlock(&cli->cl_cache->ccc_lru_lock);
2648 if (KEY_IS(KEY_CACHE_LRU_SHRINK)) {
2649 struct client_obd *cli = &obd->u.cli;
2650 long nr = atomic_long_read(&cli->cl_lru_in_list) >> 1;
2651 long target = *(long *)val;
2653 nr = osc_lru_shrink(env, cli, min(nr, target), true);
2658 if (!set && !KEY_IS(KEY_GRANT_SHRINK))
2661 /* We pass all other commands directly to OST. Since nobody calls osc
2662 methods directly and everybody is supposed to go through LOV, we
2663 assume lov checked invalid values for us.
2664 The only recognised values so far are evict_by_nid and mds_conn.
2665 Even if something bad goes through, we'd get a -EINVAL from OST
2668 req = ptlrpc_request_alloc(imp, KEY_IS(KEY_GRANT_SHRINK) ?
2669 &RQF_OST_SET_GRANT_INFO :
2674 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
2675 RCL_CLIENT, keylen);
2676 if (!KEY_IS(KEY_GRANT_SHRINK))
2677 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_VAL,
2678 RCL_CLIENT, vallen);
2679 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SET_INFO);
2681 ptlrpc_request_free(req);
2685 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
2686 memcpy(tmp, key, keylen);
2687 tmp = req_capsule_client_get(&req->rq_pill, KEY_IS(KEY_GRANT_SHRINK) ?
2690 memcpy(tmp, val, vallen);
2692 if (KEY_IS(KEY_GRANT_SHRINK)) {
2693 struct osc_grant_args *aa;
2696 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2697 aa = ptlrpc_req_async_args(req);
2700 ptlrpc_req_finished(req);
2703 *oa = ((struct ost_body *)val)->oa;
2705 req->rq_interpret_reply = osc_shrink_grant_interpret;
2708 ptlrpc_request_set_replen(req);
2709 if (!KEY_IS(KEY_GRANT_SHRINK)) {
2710 LASSERT(set != NULL);
2711 ptlrpc_set_add_req(set, req);
2712 ptlrpc_check_set(NULL, set);
2714 ptlrpcd_add_req(req);
2719 EXPORT_SYMBOL(osc_set_info_async);
2721 int osc_reconnect(const struct lu_env *env, struct obd_export *exp,
2722 struct obd_device *obd, struct obd_uuid *cluuid,
2723 struct obd_connect_data *data, void *localdata)
2725 struct client_obd *cli = &obd->u.cli;
2727 if (data != NULL && (data->ocd_connect_flags & OBD_CONNECT_GRANT)) {
2731 spin_lock(&cli->cl_loi_list_lock);
2732 grant = cli->cl_avail_grant + cli->cl_reserved_grant;
2733 if (data->ocd_connect_flags & OBD_CONNECT_GRANT_PARAM)
2734 grant += cli->cl_dirty_grant;
2736 grant += cli->cl_dirty_pages << PAGE_SHIFT;
2737 data->ocd_grant = grant ? : 2 * cli_brw_size(obd);
2738 lost_grant = cli->cl_lost_grant;
2739 cli->cl_lost_grant = 0;
2740 spin_unlock(&cli->cl_loi_list_lock);
2742 CDEBUG(D_RPCTRACE, "ocd_connect_flags: %#llx ocd_version: %d"
2743 " ocd_grant: %d, lost: %ld.\n", data->ocd_connect_flags,
2744 data->ocd_version, data->ocd_grant, lost_grant);
2749 EXPORT_SYMBOL(osc_reconnect);
2751 int osc_disconnect(struct obd_export *exp)
2753 struct obd_device *obd = class_exp2obd(exp);
2756 rc = client_disconnect_export(exp);
2758 * Initially we put del_shrink_grant before disconnect_export, but it
2759 * causes the following problem if setup (connect) and cleanup
2760 * (disconnect) are tangled together.
2761 * connect p1 disconnect p2
2762 * ptlrpc_connect_import
2763 * ............... class_manual_cleanup
2766 * ptlrpc_connect_interrupt
2768 * add this client to shrink list
2770 * Bang! pinger trigger the shrink.
2771 * So the osc should be disconnected from the shrink list, after we
2772 * are sure the import has been destroyed. BUG18662
2774 if (obd->u.cli.cl_import == NULL)
2775 osc_del_shrink_grant(&obd->u.cli);
2778 EXPORT_SYMBOL(osc_disconnect);
2780 int osc_ldlm_resource_invalidate(struct cfs_hash *hs, struct cfs_hash_bd *bd,
2781 struct hlist_node *hnode, void *arg)
2783 struct lu_env *env = arg;
2784 struct ldlm_resource *res = cfs_hash_object(hs, hnode);
2785 struct ldlm_lock *lock;
2786 struct osc_object *osc = NULL;
2790 list_for_each_entry(lock, &res->lr_granted, l_res_link) {
2791 if (lock->l_ast_data != NULL && osc == NULL) {
2792 osc = lock->l_ast_data;
2793 cl_object_get(osc2cl(osc));
2796 /* clear LDLM_FL_CLEANED flag to make sure it will be canceled
2797 * by the 2nd round of ldlm_namespace_clean() call in
2798 * osc_import_event(). */
2799 ldlm_clear_cleaned(lock);
2804 osc_object_invalidate(env, osc);
2805 cl_object_put(env, osc2cl(osc));
2810 EXPORT_SYMBOL(osc_ldlm_resource_invalidate);
2812 static int osc_import_event(struct obd_device *obd,
2813 struct obd_import *imp,
2814 enum obd_import_event event)
2816 struct client_obd *cli;
2820 LASSERT(imp->imp_obd == obd);
2823 case IMP_EVENT_DISCON: {
2825 spin_lock(&cli->cl_loi_list_lock);
2826 cli->cl_avail_grant = 0;
2827 cli->cl_lost_grant = 0;
2828 spin_unlock(&cli->cl_loi_list_lock);
2831 case IMP_EVENT_INACTIVE: {
2832 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_INACTIVE);
2835 case IMP_EVENT_INVALIDATE: {
2836 struct ldlm_namespace *ns = obd->obd_namespace;
2840 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
2842 env = cl_env_get(&refcheck);
2844 osc_io_unplug(env, &obd->u.cli, NULL);
2846 cfs_hash_for_each_nolock(ns->ns_rs_hash,
2847 osc_ldlm_resource_invalidate,
2849 cl_env_put(env, &refcheck);
2851 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
2856 case IMP_EVENT_ACTIVE: {
2857 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVE);
2860 case IMP_EVENT_OCD: {
2861 struct obd_connect_data *ocd = &imp->imp_connect_data;
2863 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT)
2864 osc_init_grant(&obd->u.cli, ocd);
2867 if (ocd->ocd_connect_flags & OBD_CONNECT_REQPORTAL)
2868 imp->imp_client->cli_request_portal =OST_REQUEST_PORTAL;
2870 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_OCD);
2873 case IMP_EVENT_DEACTIVATE: {
2874 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_DEACTIVATE);
2877 case IMP_EVENT_ACTIVATE: {
2878 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVATE);
2882 CERROR("Unknown import event %d\n", event);
2889 * Determine whether the lock can be canceled before replaying the lock
2890 * during recovery, see bug16774 for detailed information.
2892 * \retval zero the lock can't be canceled
2893 * \retval other ok to cancel
2895 static int osc_cancel_weight(struct ldlm_lock *lock)
2898 * Cancel all unused and granted extent lock.
2900 if (lock->l_resource->lr_type == LDLM_EXTENT &&
2901 lock->l_granted_mode == lock->l_req_mode &&
2902 osc_ldlm_weigh_ast(lock) == 0)
2908 static int brw_queue_work(const struct lu_env *env, void *data)
2910 struct client_obd *cli = data;
2912 CDEBUG(D_CACHE, "Run writeback work for client obd %p.\n", cli);
2914 osc_io_unplug(env, cli, NULL);
2918 int osc_setup_common(struct obd_device *obd, struct lustre_cfg *lcfg)
2920 struct client_obd *cli = &obd->u.cli;
2926 rc = ptlrpcd_addref();
2930 rc = client_obd_setup(obd, lcfg);
2932 GOTO(out_ptlrpcd, rc);
2935 handler = ptlrpcd_alloc_work(cli->cl_import, brw_queue_work, cli);
2936 if (IS_ERR(handler))
2937 GOTO(out_ptlrpcd_work, rc = PTR_ERR(handler));
2938 cli->cl_writeback_work = handler;
2940 handler = ptlrpcd_alloc_work(cli->cl_import, lru_queue_work, cli);
2941 if (IS_ERR(handler))
2942 GOTO(out_ptlrpcd_work, rc = PTR_ERR(handler));
2943 cli->cl_lru_work = handler;
2945 rc = osc_quota_setup(obd);
2947 GOTO(out_ptlrpcd_work, rc);
2949 cli->cl_grant_shrink_interval = GRANT_SHRINK_INTERVAL;
2951 INIT_LIST_HEAD(&cli->cl_grant_shrink_list);
2955 if (cli->cl_writeback_work != NULL) {
2956 ptlrpcd_destroy_work(cli->cl_writeback_work);
2957 cli->cl_writeback_work = NULL;
2959 if (cli->cl_lru_work != NULL) {
2960 ptlrpcd_destroy_work(cli->cl_lru_work);
2961 cli->cl_lru_work = NULL;
2963 client_obd_cleanup(obd);
2968 EXPORT_SYMBOL(osc_setup_common);
2970 int osc_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
2972 struct client_obd *cli = &obd->u.cli;
2973 struct obd_type *type;
2981 rc = osc_setup_common(obd, lcfg);
2985 #ifdef CONFIG_PROC_FS
2986 obd->obd_vars = lprocfs_osc_obd_vars;
2988 /* If this is true then both client (osc) and server (osp) are on the
2989 * same node. The osp layer if loaded first will register the osc proc
2990 * directory. In that case this obd_device will be attached its proc
2991 * tree to type->typ_procsym instead of obd->obd_type->typ_procroot.
2993 type = class_search_type(LUSTRE_OSP_NAME);
2994 if (type && type->typ_procsym) {
2995 obd->obd_proc_entry = lprocfs_register(obd->obd_name,
2997 obd->obd_vars, obd);
2998 if (IS_ERR(obd->obd_proc_entry)) {
2999 rc = PTR_ERR(obd->obd_proc_entry);
3000 CERROR("error %d setting up lprocfs for %s\n", rc,
3002 obd->obd_proc_entry = NULL;
3006 rc = lprocfs_obd_setup(obd, false);
3008 /* If the basic OSC proc tree construction succeeded then
3011 lproc_osc_attach_seqstat(obd);
3012 sptlrpc_lprocfs_cliobd_attach(obd);
3013 ptlrpc_lprocfs_register_obd(obd);
3017 * We try to control the total number of requests with a upper limit
3018 * osc_reqpool_maxreqcount. There might be some race which will cause
3019 * over-limit allocation, but it is fine.
3021 req_count = atomic_read(&osc_pool_req_count);
3022 if (req_count < osc_reqpool_maxreqcount) {
3023 adding = cli->cl_max_rpcs_in_flight + 2;
3024 if (req_count + adding > osc_reqpool_maxreqcount)
3025 adding = osc_reqpool_maxreqcount - req_count;
3027 added = ptlrpc_add_rqs_to_pool(osc_rq_pool, adding);
3028 atomic_add(added, &osc_pool_req_count);
3031 INIT_LIST_HEAD(&cli->cl_grant_shrink_list);
3032 ns_register_cancel(obd->obd_namespace, osc_cancel_weight);
3034 spin_lock(&osc_shrink_lock);
3035 list_add_tail(&cli->cl_shrink_list, &osc_shrink_list);
3036 spin_unlock(&osc_shrink_lock);
3041 int osc_precleanup_common(struct obd_device *obd)
3043 struct client_obd *cli = &obd->u.cli;
3047 * for echo client, export may be on zombie list, wait for
3048 * zombie thread to cull it, because cli.cl_import will be
3049 * cleared in client_disconnect_export():
3050 * class_export_destroy() -> obd_cleanup() ->
3051 * echo_device_free() -> echo_client_cleanup() ->
3052 * obd_disconnect() -> osc_disconnect() ->
3053 * client_disconnect_export()
3055 obd_zombie_barrier();
3056 if (cli->cl_writeback_work) {
3057 ptlrpcd_destroy_work(cli->cl_writeback_work);
3058 cli->cl_writeback_work = NULL;
3061 if (cli->cl_lru_work) {
3062 ptlrpcd_destroy_work(cli->cl_lru_work);
3063 cli->cl_lru_work = NULL;
3066 obd_cleanup_client_import(obd);
3069 EXPORT_SYMBOL(osc_precleanup_common);
3071 static int osc_precleanup(struct obd_device *obd)
3075 osc_precleanup_common(obd);
3077 ptlrpc_lprocfs_unregister_obd(obd);
3081 int osc_cleanup_common(struct obd_device *obd)
3083 struct client_obd *cli = &obd->u.cli;
3088 spin_lock(&osc_shrink_lock);
3089 list_del(&cli->cl_shrink_list);
3090 spin_unlock(&osc_shrink_lock);
3093 if (cli->cl_cache != NULL) {
3094 LASSERT(atomic_read(&cli->cl_cache->ccc_users) > 0);
3095 spin_lock(&cli->cl_cache->ccc_lru_lock);
3096 list_del_init(&cli->cl_lru_osc);
3097 spin_unlock(&cli->cl_cache->ccc_lru_lock);
3098 cli->cl_lru_left = NULL;
3099 cl_cache_decref(cli->cl_cache);
3100 cli->cl_cache = NULL;
3103 /* free memory of osc quota cache */
3104 osc_quota_cleanup(obd);
3106 rc = client_obd_cleanup(obd);
3111 EXPORT_SYMBOL(osc_cleanup_common);
3113 int osc_process_config_base(struct obd_device *obd, struct lustre_cfg *lcfg)
3115 int rc = class_process_proc_param(PARAM_OSC, obd->obd_vars, lcfg, obd);
3116 return rc > 0 ? 0: rc;
3119 static int osc_process_config(struct obd_device *obd, size_t len, void *buf)
3121 return osc_process_config_base(obd, buf);
3124 static struct obd_ops osc_obd_ops = {
3125 .o_owner = THIS_MODULE,
3126 .o_setup = osc_setup,
3127 .o_precleanup = osc_precleanup,
3128 .o_cleanup = osc_cleanup_common,
3129 .o_add_conn = client_import_add_conn,
3130 .o_del_conn = client_import_del_conn,
3131 .o_connect = client_connect_import,
3132 .o_reconnect = osc_reconnect,
3133 .o_disconnect = osc_disconnect,
3134 .o_statfs = osc_statfs,
3135 .o_statfs_async = osc_statfs_async,
3136 .o_create = osc_create,
3137 .o_destroy = osc_destroy,
3138 .o_getattr = osc_getattr,
3139 .o_setattr = osc_setattr,
3140 .o_iocontrol = osc_iocontrol,
3141 .o_set_info_async = osc_set_info_async,
3142 .o_import_event = osc_import_event,
3143 .o_process_config = osc_process_config,
3144 .o_quotactl = osc_quotactl,
3147 static struct shrinker *osc_cache_shrinker;
3148 struct list_head osc_shrink_list = LIST_HEAD_INIT(osc_shrink_list);
3149 DEFINE_SPINLOCK(osc_shrink_lock);
3151 #ifndef HAVE_SHRINKER_COUNT
3152 static int osc_cache_shrink(SHRINKER_ARGS(sc, nr_to_scan, gfp_mask))
3154 struct shrink_control scv = {
3155 .nr_to_scan = shrink_param(sc, nr_to_scan),
3156 .gfp_mask = shrink_param(sc, gfp_mask)
3158 #if !defined(HAVE_SHRINKER_WANT_SHRINK_PTR) && !defined(HAVE_SHRINK_CONTROL)
3159 struct shrinker *shrinker = NULL;
3162 (void)osc_cache_shrink_scan(shrinker, &scv);
3164 return osc_cache_shrink_count(shrinker, &scv);
3168 static int __init osc_init(void)
3170 bool enable_proc = true;
3171 struct obd_type *type;
3172 unsigned int reqpool_size;
3173 unsigned int reqsize;
3175 DEF_SHRINKER_VAR(osc_shvar, osc_cache_shrink,
3176 osc_cache_shrink_count, osc_cache_shrink_scan);
3179 /* print an address of _any_ initialized kernel symbol from this
3180 * module, to allow debugging with gdb that doesn't support data
3181 * symbols from modules.*/
3182 CDEBUG(D_INFO, "Lustre OSC module (%p).\n", &osc_caches);
3184 rc = lu_kmem_init(osc_caches);
3188 type = class_search_type(LUSTRE_OSP_NAME);
3189 if (type != NULL && type->typ_procsym != NULL)
3190 enable_proc = false;
3192 rc = class_register_type(&osc_obd_ops, NULL, enable_proc, NULL,
3193 LUSTRE_OSC_NAME, &osc_device_type);
3197 osc_cache_shrinker = set_shrinker(DEFAULT_SEEKS, &osc_shvar);
3199 /* This is obviously too much memory, only prevent overflow here */
3200 if (osc_reqpool_mem_max >= 1 << 12 || osc_reqpool_mem_max == 0)
3201 GOTO(out_type, rc = -EINVAL);
3203 reqpool_size = osc_reqpool_mem_max << 20;
3206 while (reqsize < OST_IO_MAXREQSIZE)
3207 reqsize = reqsize << 1;
3210 * We don't enlarge the request count in OSC pool according to
3211 * cl_max_rpcs_in_flight. The allocation from the pool will only be
3212 * tried after normal allocation failed. So a small OSC pool won't
3213 * cause much performance degression in most of cases.
3215 osc_reqpool_maxreqcount = reqpool_size / reqsize;
3217 atomic_set(&osc_pool_req_count, 0);
3218 osc_rq_pool = ptlrpc_init_rq_pool(0, OST_IO_MAXREQSIZE,
3219 ptlrpc_add_rqs_to_pool);
3221 if (osc_rq_pool != NULL)
3225 class_unregister_type(LUSTRE_OSC_NAME);
3227 lu_kmem_fini(osc_caches);
3232 static void __exit osc_exit(void)
3234 remove_shrinker(osc_cache_shrinker);
3235 class_unregister_type(LUSTRE_OSC_NAME);
3236 lu_kmem_fini(osc_caches);
3237 ptlrpc_free_rq_pool(osc_rq_pool);
3240 MODULE_AUTHOR("OpenSFS, Inc. <http://www.lustre.org/>");
3241 MODULE_DESCRIPTION("Lustre Object Storage Client (OSC)");
3242 MODULE_VERSION(LUSTRE_VERSION_STRING);
3243 MODULE_LICENSE("GPL");
3245 module_init(osc_init);
3246 module_exit(osc_exit);