4 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
6 * This program is free software; you can redistribute it and/or modify
7 * it under the terms of the GNU General Public License version 2 only,
8 * as published by the Free Software Foundation.
10 * This program is distributed in the hope that it will be useful, but
11 * WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 * General Public License version 2 for more details (a copy is included
14 * in the LICENSE file that accompanied this code).
16 * You should have received a copy of the GNU General Public License
17 * version 2 along with this program; If not, see
18 * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf
20 * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
21 * CA 95054 USA or visit www.sun.com if you need additional information or
27 * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
28 * Use is subject to license terms.
30 * Copyright (c) 2011, 2013, Intel Corporation.
33 * This file is part of Lustre, http://www.lustre.org/
34 * Lustre is a trademark of Sun Microsystems, Inc.
37 #define DEBUG_SUBSYSTEM S_OSC
39 #include <libcfs/libcfs.h>
42 # include <liblustre.h>
45 #include <lustre_dlm.h>
46 #include <lustre_net.h>
47 #include <lustre/lustre_user.h>
48 #include <obd_cksum.h>
54 #include <lustre_ha.h>
55 #include <lprocfs_status.h>
56 #include <lustre_ioctl.h>
57 #include <lustre_log.h>
58 #include <lustre_debug.h>
59 #include <lustre_param.h>
60 #include <lustre_fid.h>
61 #include "osc_internal.h"
62 #include "osc_cl_internal.h"
64 struct osc_brw_async_args {
68 obd_count aa_page_count;
70 struct brw_page **aa_ppga;
71 struct client_obd *aa_cli;
72 struct list_head aa_oaps;
73 struct list_head aa_exts;
74 struct obd_capa *aa_ocapa;
75 struct cl_req *aa_clerq;
78 #define osc_grant_args osc_brw_async_args
80 struct osc_async_args {
81 struct obd_info *aa_oi;
84 struct osc_setattr_args {
86 obd_enqueue_update_f sa_upcall;
90 struct osc_fsync_args {
91 struct obd_info *fa_oi;
92 obd_enqueue_update_f fa_upcall;
96 struct osc_enqueue_args {
97 struct obd_export *oa_exp;
99 obd_enqueue_update_f oa_upcall;
101 struct ost_lvb *oa_lvb;
102 struct lustre_handle *oa_lockh;
103 struct ldlm_enqueue_info *oa_ei;
104 unsigned int oa_agl:1;
107 static void osc_release_ppga(struct brw_page **ppga, obd_count count);
108 static int brw_interpret(const struct lu_env *env, struct ptlrpc_request *req,
111 /* Unpack OSC object metadata from disk storage (LE byte order). */
112 static int osc_unpackmd(struct obd_export *exp, struct lov_stripe_md **lsmp,
113 struct lov_mds_md *lmm, int lmm_bytes)
116 struct obd_import *imp = class_exp2cliimp(exp);
120 if (lmm_bytes < sizeof(*lmm)) {
121 CERROR("%s: lov_mds_md too small: %d, need %d\n",
122 exp->exp_obd->obd_name, lmm_bytes,
126 /* XXX LOV_MAGIC etc check? */
128 if (unlikely(ostid_id(&lmm->lmm_oi) == 0)) {
129 CERROR("%s: zero lmm_object_id: rc = %d\n",
130 exp->exp_obd->obd_name, -EINVAL);
135 lsm_size = lov_stripe_md_size(1);
139 if (*lsmp != NULL && lmm == NULL) {
140 OBD_FREE((*lsmp)->lsm_oinfo[0], sizeof(struct lov_oinfo));
141 OBD_FREE(*lsmp, lsm_size);
147 OBD_ALLOC(*lsmp, lsm_size);
148 if (unlikely(*lsmp == NULL))
150 OBD_ALLOC((*lsmp)->lsm_oinfo[0], sizeof(struct lov_oinfo));
151 if (unlikely((*lsmp)->lsm_oinfo[0] == NULL)) {
152 OBD_FREE(*lsmp, lsm_size);
155 loi_init((*lsmp)->lsm_oinfo[0]);
156 } else if (unlikely(ostid_id(&(*lsmp)->lsm_oi) == 0)) {
161 /* XXX zero *lsmp? */
162 ostid_le_to_cpu(&lmm->lmm_oi, &(*lsmp)->lsm_oi);
165 (imp->imp_connect_data.ocd_connect_flags & OBD_CONNECT_MAXBYTES))
166 (*lsmp)->lsm_maxbytes = imp->imp_connect_data.ocd_maxbytes;
168 (*lsmp)->lsm_maxbytes = LUSTRE_STRIPE_MAXBYTES;
173 static inline void osc_pack_capa(struct ptlrpc_request *req,
174 struct ost_body *body, void *capa)
176 struct obd_capa *oc = (struct obd_capa *)capa;
177 struct lustre_capa *c;
182 c = req_capsule_client_get(&req->rq_pill, &RMF_CAPA1);
185 body->oa.o_valid |= OBD_MD_FLOSSCAPA;
186 DEBUG_CAPA(D_SEC, c, "pack");
189 static inline void osc_pack_req_body(struct ptlrpc_request *req,
190 struct obd_info *oinfo)
192 struct ost_body *body;
194 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
197 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
199 osc_pack_capa(req, body, oinfo->oi_capa);
202 static inline void osc_set_capa_size(struct ptlrpc_request *req,
203 const struct req_msg_field *field,
207 req_capsule_set_size(&req->rq_pill, field, RCL_CLIENT, 0);
209 /* it is already calculated as sizeof struct obd_capa */
213 static int osc_getattr_interpret(const struct lu_env *env,
214 struct ptlrpc_request *req,
215 struct osc_async_args *aa, int rc)
217 struct ost_body *body;
223 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
225 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
226 lustre_get_wire_obdo(&req->rq_import->imp_connect_data,
227 aa->aa_oi->oi_oa, &body->oa);
229 /* This should really be sent by the OST */
230 aa->aa_oi->oi_oa->o_blksize = DT_MAX_BRW_SIZE;
231 aa->aa_oi->oi_oa->o_valid |= OBD_MD_FLBLKSZ;
233 CDEBUG(D_INFO, "can't unpack ost_body\n");
235 aa->aa_oi->oi_oa->o_valid = 0;
238 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
242 static int osc_getattr_async(struct obd_export *exp, struct obd_info *oinfo,
243 struct ptlrpc_request_set *set)
245 struct ptlrpc_request *req;
246 struct osc_async_args *aa;
250 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
254 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
255 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
257 ptlrpc_request_free(req);
261 osc_pack_req_body(req, oinfo);
263 ptlrpc_request_set_replen(req);
264 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_getattr_interpret;
266 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
267 aa = ptlrpc_req_async_args(req);
270 ptlrpc_set_add_req(set, req);
274 static int osc_getattr(const struct lu_env *env, struct obd_export *exp,
275 struct obd_info *oinfo)
277 struct ptlrpc_request *req;
278 struct ost_body *body;
282 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
286 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
287 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
289 ptlrpc_request_free(req);
293 osc_pack_req_body(req, oinfo);
295 ptlrpc_request_set_replen(req);
297 rc = ptlrpc_queue_wait(req);
301 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
303 GOTO(out, rc = -EPROTO);
305 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
306 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oinfo->oi_oa,
309 oinfo->oi_oa->o_blksize = cli_brw_size(exp->exp_obd);
310 oinfo->oi_oa->o_valid |= OBD_MD_FLBLKSZ;
314 ptlrpc_req_finished(req);
318 static int osc_setattr(const struct lu_env *env, struct obd_export *exp,
319 struct obd_info *oinfo, struct obd_trans_info *oti)
321 struct ptlrpc_request *req;
322 struct ost_body *body;
326 LASSERT(oinfo->oi_oa->o_valid & OBD_MD_FLGROUP);
328 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
332 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
333 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
335 ptlrpc_request_free(req);
339 osc_pack_req_body(req, oinfo);
341 ptlrpc_request_set_replen(req);
343 rc = ptlrpc_queue_wait(req);
347 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
349 GOTO(out, rc = -EPROTO);
351 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oinfo->oi_oa,
356 ptlrpc_req_finished(req);
360 static int osc_setattr_interpret(const struct lu_env *env,
361 struct ptlrpc_request *req,
362 struct osc_setattr_args *sa, int rc)
364 struct ost_body *body;
370 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
372 GOTO(out, rc = -EPROTO);
374 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, sa->sa_oa,
377 rc = sa->sa_upcall(sa->sa_cookie, rc);
381 int osc_setattr_async_base(struct obd_export *exp, struct obd_info *oinfo,
382 struct obd_trans_info *oti,
383 obd_enqueue_update_f upcall, void *cookie,
384 struct ptlrpc_request_set *rqset)
386 struct ptlrpc_request *req;
387 struct osc_setattr_args *sa;
391 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
395 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
396 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
398 ptlrpc_request_free(req);
402 if (oti && oinfo->oi_oa->o_valid & OBD_MD_FLCOOKIE)
403 oinfo->oi_oa->o_lcookie = *oti->oti_logcookies;
405 osc_pack_req_body(req, oinfo);
407 ptlrpc_request_set_replen(req);
409 /* do mds to ost setattr asynchronously */
411 /* Do not wait for response. */
412 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
414 req->rq_interpret_reply =
415 (ptlrpc_interpterer_t)osc_setattr_interpret;
417 CLASSERT (sizeof(*sa) <= sizeof(req->rq_async_args));
418 sa = ptlrpc_req_async_args(req);
419 sa->sa_oa = oinfo->oi_oa;
420 sa->sa_upcall = upcall;
421 sa->sa_cookie = cookie;
423 if (rqset == PTLRPCD_SET)
424 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
426 ptlrpc_set_add_req(rqset, req);
432 static int osc_setattr_async(struct obd_export *exp, struct obd_info *oinfo,
433 struct obd_trans_info *oti,
434 struct ptlrpc_request_set *rqset)
436 return osc_setattr_async_base(exp, oinfo, oti,
437 oinfo->oi_cb_up, oinfo, rqset);
440 int osc_real_create(struct obd_export *exp, struct obdo *oa,
441 struct lov_stripe_md **ea, struct obd_trans_info *oti)
443 struct ptlrpc_request *req;
444 struct ost_body *body;
445 struct lov_stripe_md *lsm;
454 rc = obd_alloc_memmd(exp, &lsm);
459 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_CREATE);
461 GOTO(out, rc = -ENOMEM);
463 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_CREATE);
465 ptlrpc_request_free(req);
469 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
472 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
474 ptlrpc_request_set_replen(req);
476 if ((oa->o_valid & OBD_MD_FLFLAGS) &&
477 oa->o_flags == OBD_FL_DELORPHAN) {
479 "delorphan from OST integration");
480 /* Don't resend the delorphan req */
481 req->rq_no_resend = req->rq_no_delay = 1;
484 rc = ptlrpc_queue_wait(req);
488 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
490 GOTO(out_req, rc = -EPROTO);
492 CDEBUG(D_INFO, "oa flags %x\n", oa->o_flags);
493 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
495 oa->o_blksize = cli_brw_size(exp->exp_obd);
496 oa->o_valid |= OBD_MD_FLBLKSZ;
498 /* XXX LOV STACKING: the lsm that is passed to us from LOV does not
499 * have valid lsm_oinfo data structs, so don't go touching that.
500 * This needs to be fixed in a big way.
502 lsm->lsm_oi = oa->o_oi;
506 if (oa->o_valid & OBD_MD_FLCOOKIE) {
507 if (oti->oti_logcookies == NULL)
508 oti->oti_logcookies = &oti->oti_onecookie;
510 *oti->oti_logcookies = oa->o_lcookie;
514 CDEBUG(D_HA, "transno: "LPD64"\n",
515 lustre_msg_get_transno(req->rq_repmsg));
517 ptlrpc_req_finished(req);
520 obd_free_memmd(exp, &lsm);
524 int osc_punch_base(struct obd_export *exp, struct obd_info *oinfo,
525 obd_enqueue_update_f upcall, void *cookie,
526 struct ptlrpc_request_set *rqset)
528 struct ptlrpc_request *req;
529 struct osc_setattr_args *sa;
530 struct ost_body *body;
534 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_PUNCH);
538 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
539 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_PUNCH);
541 ptlrpc_request_free(req);
544 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
545 ptlrpc_at_set_req_timeout(req);
547 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
549 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
551 osc_pack_capa(req, body, oinfo->oi_capa);
553 ptlrpc_request_set_replen(req);
555 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_setattr_interpret;
556 CLASSERT (sizeof(*sa) <= sizeof(req->rq_async_args));
557 sa = ptlrpc_req_async_args(req);
558 sa->sa_oa = oinfo->oi_oa;
559 sa->sa_upcall = upcall;
560 sa->sa_cookie = cookie;
561 if (rqset == PTLRPCD_SET)
562 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
564 ptlrpc_set_add_req(rqset, req);
569 static int osc_sync_interpret(const struct lu_env *env,
570 struct ptlrpc_request *req,
573 struct osc_fsync_args *fa = arg;
574 struct ost_body *body;
580 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
582 CERROR ("can't unpack ost_body\n");
583 GOTO(out, rc = -EPROTO);
586 *fa->fa_oi->oi_oa = body->oa;
588 rc = fa->fa_upcall(fa->fa_cookie, rc);
592 int osc_sync_base(struct obd_export *exp, struct obd_info *oinfo,
593 obd_enqueue_update_f upcall, void *cookie,
594 struct ptlrpc_request_set *rqset)
596 struct ptlrpc_request *req;
597 struct ost_body *body;
598 struct osc_fsync_args *fa;
602 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SYNC);
606 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
607 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SYNC);
609 ptlrpc_request_free(req);
613 /* overload the size and blocks fields in the oa with start/end */
614 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
616 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
618 osc_pack_capa(req, body, oinfo->oi_capa);
620 ptlrpc_request_set_replen(req);
621 req->rq_interpret_reply = osc_sync_interpret;
623 CLASSERT(sizeof(*fa) <= sizeof(req->rq_async_args));
624 fa = ptlrpc_req_async_args(req);
626 fa->fa_upcall = upcall;
627 fa->fa_cookie = cookie;
629 if (rqset == PTLRPCD_SET)
630 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
632 ptlrpc_set_add_req(rqset, req);
637 /* Find and cancel locally locks matched by @mode in the resource found by
638 * @objid. Found locks are added into @cancel list. Returns the amount of
639 * locks added to @cancels list. */
640 static int osc_resource_get_unused(struct obd_export *exp, struct obdo *oa,
641 struct list_head *cancels,
642 ldlm_mode_t mode, __u64 lock_flags)
644 struct ldlm_namespace *ns = exp->exp_obd->obd_namespace;
645 struct ldlm_res_id res_id;
646 struct ldlm_resource *res;
650 /* Return, i.e. cancel nothing, only if ELC is supported (flag in
651 * export) but disabled through procfs (flag in NS).
653 * This distinguishes from a case when ELC is not supported originally,
654 * when we still want to cancel locks in advance and just cancel them
655 * locally, without sending any RPC. */
656 if (exp_connect_cancelset(exp) && !ns_connect_cancelset(ns))
659 ostid_build_res_name(&oa->o_oi, &res_id);
660 res = ldlm_resource_get(ns, NULL, &res_id, 0, 0);
664 LDLM_RESOURCE_ADDREF(res);
665 count = ldlm_cancel_resource_local(res, cancels, NULL, mode,
666 lock_flags, 0, NULL);
667 LDLM_RESOURCE_DELREF(res);
668 ldlm_resource_putref(res);
672 static int osc_destroy_interpret(const struct lu_env *env,
673 struct ptlrpc_request *req, void *data,
676 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
678 atomic_dec(&cli->cl_destroy_in_flight);
679 wake_up(&cli->cl_destroy_waitq);
683 static int osc_can_send_destroy(struct client_obd *cli)
685 if (atomic_inc_return(&cli->cl_destroy_in_flight) <=
686 cli->cl_max_rpcs_in_flight) {
687 /* The destroy request can be sent */
690 if (atomic_dec_return(&cli->cl_destroy_in_flight) <
691 cli->cl_max_rpcs_in_flight) {
693 * The counter has been modified between the two atomic
696 wake_up(&cli->cl_destroy_waitq);
701 int osc_create(const struct lu_env *env, struct obd_export *exp,
702 struct obdo *oa, struct lov_stripe_md **ea,
703 struct obd_trans_info *oti)
710 LASSERT(oa->o_valid & OBD_MD_FLGROUP);
712 if ((oa->o_valid & OBD_MD_FLFLAGS) &&
713 oa->o_flags == OBD_FL_RECREATE_OBJS) {
714 RETURN(osc_real_create(exp, oa, ea, oti));
717 if (!fid_seq_is_mdt(ostid_seq(&oa->o_oi)))
718 RETURN(osc_real_create(exp, oa, ea, oti));
720 /* we should not get here anymore */
726 /* Destroy requests can be async always on the client, and we don't even really
727 * care about the return code since the client cannot do anything at all about
729 * When the MDS is unlinking a filename, it saves the file objects into a
730 * recovery llog, and these object records are cancelled when the OST reports
731 * they were destroyed and sync'd to disk (i.e. transaction committed).
732 * If the client dies, or the OST is down when the object should be destroyed,
733 * the records are not cancelled, and when the OST reconnects to the MDS next,
734 * it will retrieve the llog unlink logs and then sends the log cancellation
735 * cookies to the MDS after committing destroy transactions. */
736 static int osc_destroy(const struct lu_env *env, struct obd_export *exp,
737 struct obdo *oa, struct lov_stripe_md *ea,
738 struct obd_trans_info *oti, struct obd_export *md_export,
741 struct client_obd *cli = &exp->exp_obd->u.cli;
742 struct ptlrpc_request *req;
743 struct ost_body *body;
744 struct list_head cancels = LIST_HEAD_INIT(cancels);
749 CDEBUG(D_INFO, "oa NULL\n");
753 count = osc_resource_get_unused(exp, oa, &cancels, LCK_PW,
754 LDLM_FL_DISCARD_DATA);
756 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_DESTROY);
758 ldlm_lock_list_put(&cancels, l_bl_ast, count);
762 osc_set_capa_size(req, &RMF_CAPA1, (struct obd_capa *)capa);
763 rc = ldlm_prep_elc_req(exp, req, LUSTRE_OST_VERSION, OST_DESTROY,
766 ptlrpc_request_free(req);
770 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
771 ptlrpc_at_set_req_timeout(req);
773 if (oti != NULL && oa->o_valid & OBD_MD_FLCOOKIE)
774 oa->o_lcookie = *oti->oti_logcookies;
775 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
777 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
779 osc_pack_capa(req, body, (struct obd_capa *)capa);
780 ptlrpc_request_set_replen(req);
782 /* If osc_destory is for destroying the unlink orphan,
783 * sent from MDT to OST, which should not be blocked here,
784 * because the process might be triggered by ptlrpcd, and
785 * it is not good to block ptlrpcd thread (b=16006)*/
786 if (!(oa->o_flags & OBD_FL_DELORPHAN)) {
787 req->rq_interpret_reply = osc_destroy_interpret;
788 if (!osc_can_send_destroy(cli)) {
789 struct l_wait_info lwi = LWI_INTR(LWI_ON_SIGNAL_NOOP,
793 * Wait until the number of on-going destroy RPCs drops
794 * under max_rpc_in_flight
796 l_wait_event_exclusive(cli->cl_destroy_waitq,
797 osc_can_send_destroy(cli), &lwi);
801 /* Do not wait for response */
802 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
806 static void osc_announce_cached(struct client_obd *cli, struct obdo *oa,
809 obd_flag bits = OBD_MD_FLBLOCKS|OBD_MD_FLGRANT;
811 LASSERT(!(oa->o_valid & bits));
814 client_obd_list_lock(&cli->cl_loi_list_lock);
815 oa->o_dirty = cli->cl_dirty;
816 if (unlikely(cli->cl_dirty - cli->cl_dirty_transit >
817 cli->cl_dirty_max)) {
818 CERROR("dirty %lu - %lu > dirty_max %lu\n",
819 cli->cl_dirty, cli->cl_dirty_transit, cli->cl_dirty_max);
821 } else if (unlikely(atomic_read(&obd_dirty_pages) -
822 atomic_read(&obd_dirty_transit_pages) >
823 (long)(obd_max_dirty_pages + 1))) {
824 /* The atomic_read() allowing the atomic_inc() are
825 * not covered by a lock thus they may safely race and trip
826 * this CERROR() unless we add in a small fudge factor (+1). */
827 CERROR("%s: dirty %d - %d > system dirty_max %d\n",
828 cli->cl_import->imp_obd->obd_name,
829 atomic_read(&obd_dirty_pages),
830 atomic_read(&obd_dirty_transit_pages),
831 obd_max_dirty_pages);
833 } else if (unlikely(cli->cl_dirty_max - cli->cl_dirty > 0x7fffffff)) {
834 CERROR("dirty %lu - dirty_max %lu too big???\n",
835 cli->cl_dirty, cli->cl_dirty_max);
838 long max_in_flight = (cli->cl_max_pages_per_rpc <<
840 (cli->cl_max_rpcs_in_flight + 1);
841 oa->o_undirty = max(cli->cl_dirty_max, max_in_flight);
843 oa->o_grant = cli->cl_avail_grant + cli->cl_reserved_grant;
844 oa->o_dropped = cli->cl_lost_grant;
845 cli->cl_lost_grant = 0;
846 client_obd_list_unlock(&cli->cl_loi_list_lock);
847 CDEBUG(D_CACHE,"dirty: "LPU64" undirty: %u dropped %u grant: "LPU64"\n",
848 oa->o_dirty, oa->o_undirty, oa->o_dropped, oa->o_grant);
852 void osc_update_next_shrink(struct client_obd *cli)
854 cli->cl_next_shrink_grant =
855 cfs_time_shift(cli->cl_grant_shrink_interval);
856 CDEBUG(D_CACHE, "next time %ld to shrink grant \n",
857 cli->cl_next_shrink_grant);
860 static void __osc_update_grant(struct client_obd *cli, obd_size grant)
862 client_obd_list_lock(&cli->cl_loi_list_lock);
863 cli->cl_avail_grant += grant;
864 client_obd_list_unlock(&cli->cl_loi_list_lock);
867 static void osc_update_grant(struct client_obd *cli, struct ost_body *body)
869 if (body->oa.o_valid & OBD_MD_FLGRANT) {
870 CDEBUG(D_CACHE, "got "LPU64" extra grant\n", body->oa.o_grant);
871 __osc_update_grant(cli, body->oa.o_grant);
875 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
876 obd_count keylen, void *key, obd_count vallen,
877 void *val, struct ptlrpc_request_set *set);
879 static int osc_shrink_grant_interpret(const struct lu_env *env,
880 struct ptlrpc_request *req,
883 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
884 struct obdo *oa = ((struct osc_grant_args *)aa)->aa_oa;
885 struct ost_body *body;
888 __osc_update_grant(cli, oa->o_grant);
892 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
894 osc_update_grant(cli, body);
900 static void osc_shrink_grant_local(struct client_obd *cli, struct obdo *oa)
902 client_obd_list_lock(&cli->cl_loi_list_lock);
903 oa->o_grant = cli->cl_avail_grant / 4;
904 cli->cl_avail_grant -= oa->o_grant;
905 client_obd_list_unlock(&cli->cl_loi_list_lock);
906 if (!(oa->o_valid & OBD_MD_FLFLAGS)) {
907 oa->o_valid |= OBD_MD_FLFLAGS;
910 oa->o_flags |= OBD_FL_SHRINK_GRANT;
911 osc_update_next_shrink(cli);
914 /* Shrink the current grant, either from some large amount to enough for a
915 * full set of in-flight RPCs, or if we have already shrunk to that limit
916 * then to enough for a single RPC. This avoids keeping more grant than
917 * needed, and avoids shrinking the grant piecemeal. */
918 static int osc_shrink_grant(struct client_obd *cli)
920 __u64 target_bytes = (cli->cl_max_rpcs_in_flight + 1) *
921 (cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT);
923 client_obd_list_lock(&cli->cl_loi_list_lock);
924 if (cli->cl_avail_grant <= target_bytes)
925 target_bytes = cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT;
926 client_obd_list_unlock(&cli->cl_loi_list_lock);
928 return osc_shrink_grant_to_target(cli, target_bytes);
931 int osc_shrink_grant_to_target(struct client_obd *cli, __u64 target_bytes)
934 struct ost_body *body;
937 client_obd_list_lock(&cli->cl_loi_list_lock);
938 /* Don't shrink if we are already above or below the desired limit
939 * We don't want to shrink below a single RPC, as that will negatively
940 * impact block allocation and long-term performance. */
941 if (target_bytes < cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT)
942 target_bytes = cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT;
944 if (target_bytes >= cli->cl_avail_grant) {
945 client_obd_list_unlock(&cli->cl_loi_list_lock);
948 client_obd_list_unlock(&cli->cl_loi_list_lock);
954 osc_announce_cached(cli, &body->oa, 0);
956 client_obd_list_lock(&cli->cl_loi_list_lock);
957 body->oa.o_grant = cli->cl_avail_grant - target_bytes;
958 cli->cl_avail_grant = target_bytes;
959 client_obd_list_unlock(&cli->cl_loi_list_lock);
960 if (!(body->oa.o_valid & OBD_MD_FLFLAGS)) {
961 body->oa.o_valid |= OBD_MD_FLFLAGS;
962 body->oa.o_flags = 0;
964 body->oa.o_flags |= OBD_FL_SHRINK_GRANT;
965 osc_update_next_shrink(cli);
967 rc = osc_set_info_async(NULL, cli->cl_import->imp_obd->obd_self_export,
968 sizeof(KEY_GRANT_SHRINK), KEY_GRANT_SHRINK,
969 sizeof(*body), body, NULL);
971 __osc_update_grant(cli, body->oa.o_grant);
976 static int osc_should_shrink_grant(struct client_obd *client)
978 cfs_time_t time = cfs_time_current();
979 cfs_time_t next_shrink = client->cl_next_shrink_grant;
981 if ((client->cl_import->imp_connect_data.ocd_connect_flags &
982 OBD_CONNECT_GRANT_SHRINK) == 0)
985 if (cfs_time_aftereq(time, next_shrink - 5 * CFS_TICK)) {
986 /* Get the current RPC size directly, instead of going via:
987 * cli_brw_size(obd->u.cli.cl_import->imp_obd->obd_self_export)
988 * Keep comment here so that it can be found by searching. */
989 int brw_size = client->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT;
991 if (client->cl_import->imp_state == LUSTRE_IMP_FULL &&
992 client->cl_avail_grant > brw_size)
995 osc_update_next_shrink(client);
1000 static int osc_grant_shrink_grant_cb(struct timeout_item *item, void *data)
1002 struct client_obd *client;
1004 list_for_each_entry(client, &item->ti_obd_list, cl_grant_shrink_list) {
1005 if (osc_should_shrink_grant(client))
1006 osc_shrink_grant(client);
1011 static int osc_add_shrink_grant(struct client_obd *client)
1015 rc = ptlrpc_add_timeout_client(client->cl_grant_shrink_interval,
1017 osc_grant_shrink_grant_cb, NULL,
1018 &client->cl_grant_shrink_list);
1020 CERROR("add grant client %s error %d\n",
1021 client->cl_import->imp_obd->obd_name, rc);
1024 CDEBUG(D_CACHE, "add grant client %s \n",
1025 client->cl_import->imp_obd->obd_name);
1026 osc_update_next_shrink(client);
1030 static int osc_del_shrink_grant(struct client_obd *client)
1032 return ptlrpc_del_timeout_client(&client->cl_grant_shrink_list,
1036 static void osc_init_grant(struct client_obd *cli, struct obd_connect_data *ocd)
1039 * ocd_grant is the total grant amount we're expect to hold: if we've
1040 * been evicted, it's the new avail_grant amount, cl_dirty will drop
1041 * to 0 as inflight RPCs fail out; otherwise, it's avail_grant + dirty.
1043 * race is tolerable here: if we're evicted, but imp_state already
1044 * left EVICTED state, then cl_dirty must be 0 already.
1046 client_obd_list_lock(&cli->cl_loi_list_lock);
1047 if (cli->cl_import->imp_state == LUSTRE_IMP_EVICTED)
1048 cli->cl_avail_grant = ocd->ocd_grant;
1050 cli->cl_avail_grant = ocd->ocd_grant - cli->cl_dirty;
1052 if (cli->cl_avail_grant < 0) {
1053 CWARN("%s: available grant < 0: avail/ocd/dirty %ld/%u/%ld\n",
1054 cli->cl_import->imp_obd->obd_name, cli->cl_avail_grant,
1055 ocd->ocd_grant, cli->cl_dirty);
1056 /* workaround for servers which do not have the patch from
1058 cli->cl_avail_grant = ocd->ocd_grant;
1061 /* determine the appropriate chunk size used by osc_extent. */
1062 cli->cl_chunkbits = max_t(int, PAGE_CACHE_SHIFT, ocd->ocd_blocksize);
1063 client_obd_list_unlock(&cli->cl_loi_list_lock);
1065 CDEBUG(D_CACHE, "%s, setting cl_avail_grant: %ld cl_lost_grant: %ld."
1066 "chunk bits: %d.\n", cli->cl_import->imp_obd->obd_name,
1067 cli->cl_avail_grant, cli->cl_lost_grant, cli->cl_chunkbits);
1069 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT_SHRINK &&
1070 list_empty(&cli->cl_grant_shrink_list))
1071 osc_add_shrink_grant(cli);
1074 /* We assume that the reason this OSC got a short read is because it read
1075 * beyond the end of a stripe file; i.e. lustre is reading a sparse file
1076 * via the LOV, and it _knows_ it's reading inside the file, it's just that
1077 * this stripe never got written at or beyond this stripe offset yet. */
1078 static void handle_short_read(int nob_read, obd_count page_count,
1079 struct brw_page **pga)
1084 /* skip bytes read OK */
1085 while (nob_read > 0) {
1086 LASSERT (page_count > 0);
1088 if (pga[i]->count > nob_read) {
1089 /* EOF inside this page */
1090 ptr = kmap(pga[i]->pg) +
1091 (pga[i]->off & ~CFS_PAGE_MASK);
1092 memset(ptr + nob_read, 0, pga[i]->count - nob_read);
1099 nob_read -= pga[i]->count;
1104 /* zero remaining pages */
1105 while (page_count-- > 0) {
1106 ptr = kmap(pga[i]->pg) + (pga[i]->off & ~CFS_PAGE_MASK);
1107 memset(ptr, 0, pga[i]->count);
1113 static int check_write_rcs(struct ptlrpc_request *req,
1114 int requested_nob, int niocount,
1115 obd_count page_count, struct brw_page **pga)
1120 remote_rcs = req_capsule_server_sized_get(&req->rq_pill, &RMF_RCS,
1121 sizeof(*remote_rcs) *
1123 if (remote_rcs == NULL) {
1124 CDEBUG(D_INFO, "Missing/short RC vector on BRW_WRITE reply\n");
1128 /* return error if any niobuf was in error */
1129 for (i = 0; i < niocount; i++) {
1130 if ((int)remote_rcs[i] < 0)
1131 return(remote_rcs[i]);
1133 if (remote_rcs[i] != 0) {
1134 CDEBUG(D_INFO, "rc[%d] invalid (%d) req %p\n",
1135 i, remote_rcs[i], req);
1140 if (req->rq_bulk->bd_nob_transferred != requested_nob) {
1141 CERROR("Unexpected # bytes transferred: %d (requested %d)\n",
1142 req->rq_bulk->bd_nob_transferred, requested_nob);
1149 static inline int can_merge_pages(struct brw_page *p1, struct brw_page *p2)
1151 if (p1->flag != p2->flag) {
1152 unsigned mask = ~(OBD_BRW_FROM_GRANT | OBD_BRW_NOCACHE |
1153 OBD_BRW_SYNC | OBD_BRW_ASYNC |
1154 OBD_BRW_NOQUOTA | OBD_BRW_SOFT_SYNC);
1156 /* warn if we try to combine flags that we don't know to be
1157 * safe to combine */
1158 if (unlikely((p1->flag & mask) != (p2->flag & mask))) {
1159 CWARN("Saw flags 0x%x and 0x%x in the same brw, please "
1160 "report this at http://bugs.whamcloud.com/\n",
1161 p1->flag, p2->flag);
1166 return (p1->off + p1->count == p2->off);
1169 static obd_count osc_checksum_bulk(int nob, obd_count pg_count,
1170 struct brw_page **pga, int opc,
1171 cksum_type_t cksum_type)
1175 struct cfs_crypto_hash_desc *hdesc;
1176 unsigned int bufsize;
1178 unsigned char cfs_alg = cksum_obd2cfs(cksum_type);
1180 LASSERT(pg_count > 0);
1182 hdesc = cfs_crypto_hash_init(cfs_alg, NULL, 0);
1183 if (IS_ERR(hdesc)) {
1184 CERROR("Unable to initialize checksum hash %s\n",
1185 cfs_crypto_hash_name(cfs_alg));
1186 return PTR_ERR(hdesc);
1189 while (nob > 0 && pg_count > 0) {
1190 int count = pga[i]->count > nob ? nob : pga[i]->count;
1192 /* corrupt the data before we compute the checksum, to
1193 * simulate an OST->client data error */
1194 if (i == 0 && opc == OST_READ &&
1195 OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_RECEIVE)) {
1196 unsigned char *ptr = kmap(pga[i]->pg);
1197 int off = pga[i]->off & ~CFS_PAGE_MASK;
1198 memcpy(ptr + off, "bad1", min(4, nob));
1201 cfs_crypto_hash_update_page(hdesc, pga[i]->pg,
1202 pga[i]->off & ~CFS_PAGE_MASK,
1204 LL_CDEBUG_PAGE(D_PAGE, pga[i]->pg, "off %d\n",
1205 (int)(pga[i]->off & ~CFS_PAGE_MASK));
1207 nob -= pga[i]->count;
1213 err = cfs_crypto_hash_final(hdesc, (unsigned char *)&cksum, &bufsize);
1216 cfs_crypto_hash_final(hdesc, NULL, NULL);
1218 /* For sending we only compute the wrong checksum instead
1219 * of corrupting the data so it is still correct on a redo */
1220 if (opc == OST_WRITE && OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_SEND))
1226 static int osc_brw_prep_request(int cmd, struct client_obd *cli,struct obdo *oa,
1227 struct lov_stripe_md *lsm, obd_count page_count,
1228 struct brw_page **pga,
1229 struct ptlrpc_request **reqp,
1230 struct obd_capa *ocapa, int reserve,
1233 struct ptlrpc_request *req;
1234 struct ptlrpc_bulk_desc *desc;
1235 struct ost_body *body;
1236 struct obd_ioobj *ioobj;
1237 struct niobuf_remote *niobuf;
1238 int niocount, i, requested_nob, opc, rc;
1239 struct osc_brw_async_args *aa;
1240 struct req_capsule *pill;
1241 struct brw_page *pg_prev;
1244 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ))
1245 RETURN(-ENOMEM); /* Recoverable */
1246 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ2))
1247 RETURN(-EINVAL); /* Fatal */
1249 if ((cmd & OBD_BRW_WRITE) != 0) {
1251 req = ptlrpc_request_alloc_pool(cli->cl_import,
1252 cli->cl_import->imp_rq_pool,
1253 &RQF_OST_BRW_WRITE);
1256 req = ptlrpc_request_alloc(cli->cl_import, &RQF_OST_BRW_READ);
1261 for (niocount = i = 1; i < page_count; i++) {
1262 if (!can_merge_pages(pga[i - 1], pga[i]))
1266 pill = &req->rq_pill;
1267 req_capsule_set_size(pill, &RMF_OBD_IOOBJ, RCL_CLIENT,
1269 req_capsule_set_size(pill, &RMF_NIOBUF_REMOTE, RCL_CLIENT,
1270 niocount * sizeof(*niobuf));
1271 osc_set_capa_size(req, &RMF_CAPA1, ocapa);
1273 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, opc);
1275 ptlrpc_request_free(req);
1278 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
1279 ptlrpc_at_set_req_timeout(req);
1280 /* ask ptlrpc not to resend on EINPROGRESS since BRWs have their own
1282 req->rq_no_retry_einprogress = 1;
1284 desc = ptlrpc_prep_bulk_imp(req, page_count,
1285 cli->cl_import->imp_connect_data.ocd_brw_size >> LNET_MTU_BITS,
1286 opc == OST_WRITE ? BULK_GET_SOURCE : BULK_PUT_SINK,
1290 GOTO(out, rc = -ENOMEM);
1291 /* NB request now owns desc and will free it when it gets freed */
1293 body = req_capsule_client_get(pill, &RMF_OST_BODY);
1294 ioobj = req_capsule_client_get(pill, &RMF_OBD_IOOBJ);
1295 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1296 LASSERT(body != NULL && ioobj != NULL && niobuf != NULL);
1298 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
1300 obdo_to_ioobj(oa, ioobj);
1301 ioobj->ioo_bufcnt = niocount;
1302 /* The high bits of ioo_max_brw tells server _maximum_ number of bulks
1303 * that might be send for this request. The actual number is decided
1304 * when the RPC is finally sent in ptlrpc_register_bulk(). It sends
1305 * "max - 1" for old client compatibility sending "0", and also so the
1306 * the actual maximum is a power-of-two number, not one less. LU-1431 */
1307 ioobj_max_brw_set(ioobj, desc->bd_md_max_brw);
1308 osc_pack_capa(req, body, ocapa);
1309 LASSERT(page_count > 0);
1311 for (requested_nob = i = 0; i < page_count; i++, niobuf++) {
1312 struct brw_page *pg = pga[i];
1313 int poff = pg->off & ~CFS_PAGE_MASK;
1315 LASSERT(pg->count > 0);
1316 /* make sure there is no gap in the middle of page array */
1317 LASSERTF(page_count == 1 ||
1318 (ergo(i == 0, poff + pg->count == PAGE_CACHE_SIZE) &&
1319 ergo(i > 0 && i < page_count - 1,
1320 poff == 0 && pg->count == PAGE_CACHE_SIZE) &&
1321 ergo(i == page_count - 1, poff == 0)),
1322 "i: %d/%d pg: %p off: "LPU64", count: %u\n",
1323 i, page_count, pg, pg->off, pg->count);
1325 LASSERTF(i == 0 || pg->off > pg_prev->off,
1326 "i %d p_c %u pg %p [pri %lu ind %lu] off "LPU64
1327 " prev_pg %p [pri %lu ind %lu] off "LPU64"\n",
1329 pg->pg, page_private(pg->pg), pg->pg->index, pg->off,
1330 pg_prev->pg, page_private(pg_prev->pg),
1331 pg_prev->pg->index, pg_prev->off);
1333 LASSERTF(i == 0 || pg->off > pg_prev->off,
1334 "i %d p_c %u\n", i, page_count);
1336 LASSERT((pga[0]->flag & OBD_BRW_SRVLOCK) ==
1337 (pg->flag & OBD_BRW_SRVLOCK));
1339 ptlrpc_prep_bulk_page_pin(desc, pg->pg, poff, pg->count);
1340 requested_nob += pg->count;
1342 if (i > 0 && can_merge_pages(pg_prev, pg)) {
1344 niobuf->len += pg->count;
1346 niobuf->offset = pg->off;
1347 niobuf->len = pg->count;
1348 niobuf->flags = pg->flag;
1353 LASSERTF((void *)(niobuf - niocount) ==
1354 req_capsule_client_get(&req->rq_pill, &RMF_NIOBUF_REMOTE),
1355 "want %p - real %p\n", req_capsule_client_get(&req->rq_pill,
1356 &RMF_NIOBUF_REMOTE), (void *)(niobuf - niocount));
1358 osc_announce_cached(cli, &body->oa, opc == OST_WRITE ? requested_nob:0);
1360 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1361 body->oa.o_valid |= OBD_MD_FLFLAGS;
1362 body->oa.o_flags = 0;
1364 body->oa.o_flags |= OBD_FL_RECOV_RESEND;
1367 if (osc_should_shrink_grant(cli))
1368 osc_shrink_grant_local(cli, &body->oa);
1370 /* size[REQ_REC_OFF] still sizeof (*body) */
1371 if (opc == OST_WRITE) {
1372 if (cli->cl_checksum &&
1373 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1374 /* store cl_cksum_type in a local variable since
1375 * it can be changed via lprocfs */
1376 cksum_type_t cksum_type = cli->cl_cksum_type;
1378 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1379 oa->o_flags &= OBD_FL_LOCAL_MASK;
1380 body->oa.o_flags = 0;
1382 body->oa.o_flags |= cksum_type_pack(cksum_type);
1383 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1384 body->oa.o_cksum = osc_checksum_bulk(requested_nob,
1388 CDEBUG(D_PAGE, "checksum at write origin: %x\n",
1390 /* save this in 'oa', too, for later checking */
1391 oa->o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1392 oa->o_flags |= cksum_type_pack(cksum_type);
1394 /* clear out the checksum flag, in case this is a
1395 * resend but cl_checksum is no longer set. b=11238 */
1396 oa->o_valid &= ~OBD_MD_FLCKSUM;
1398 oa->o_cksum = body->oa.o_cksum;
1399 /* 1 RC per niobuf */
1400 req_capsule_set_size(pill, &RMF_RCS, RCL_SERVER,
1401 sizeof(__u32) * niocount);
1403 if (cli->cl_checksum &&
1404 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1405 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0)
1406 body->oa.o_flags = 0;
1407 body->oa.o_flags |= cksum_type_pack(cli->cl_cksum_type);
1408 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1411 ptlrpc_request_set_replen(req);
1413 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1414 aa = ptlrpc_req_async_args(req);
1416 aa->aa_requested_nob = requested_nob;
1417 aa->aa_nio_count = niocount;
1418 aa->aa_page_count = page_count;
1422 INIT_LIST_HEAD(&aa->aa_oaps);
1423 if (ocapa && reserve)
1424 aa->aa_ocapa = capa_get(ocapa);
1430 ptlrpc_req_finished(req);
1434 static int check_write_checksum(struct obdo *oa, const lnet_process_id_t *peer,
1435 __u32 client_cksum, __u32 server_cksum, int nob,
1436 obd_count page_count, struct brw_page **pga,
1437 cksum_type_t client_cksum_type)
1441 cksum_type_t cksum_type;
1443 if (server_cksum == client_cksum) {
1444 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1448 cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
1450 new_cksum = osc_checksum_bulk(nob, page_count, pga, OST_WRITE,
1453 if (cksum_type != client_cksum_type)
1454 msg = "the server did not use the checksum type specified in "
1455 "the original request - likely a protocol problem";
1456 else if (new_cksum == server_cksum)
1457 msg = "changed on the client after we checksummed it - "
1458 "likely false positive due to mmap IO (bug 11742)";
1459 else if (new_cksum == client_cksum)
1460 msg = "changed in transit before arrival at OST";
1462 msg = "changed in transit AND doesn't match the original - "
1463 "likely false positive due to mmap IO (bug 11742)";
1465 LCONSOLE_ERROR_MSG(0x132, "BAD WRITE CHECKSUM: %s: from %s inode "DFID
1466 " object "DOSTID" extent ["LPU64"-"LPU64"]\n",
1467 msg, libcfs_nid2str(peer->nid),
1468 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
1469 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
1470 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
1471 POSTID(&oa->o_oi), pga[0]->off,
1472 pga[page_count-1]->off + pga[page_count-1]->count - 1);
1473 CERROR("original client csum %x (type %x), server csum %x (type %x), "
1474 "client csum now %x\n", client_cksum, client_cksum_type,
1475 server_cksum, cksum_type, new_cksum);
1479 /* Note rc enters this function as number of bytes transferred */
1480 static int osc_brw_fini_request(struct ptlrpc_request *req, int rc)
1482 struct osc_brw_async_args *aa = (void *)&req->rq_async_args;
1483 const lnet_process_id_t *peer =
1484 &req->rq_import->imp_connection->c_peer;
1485 struct client_obd *cli = aa->aa_cli;
1486 struct ost_body *body;
1487 __u32 client_cksum = 0;
1490 if (rc < 0 && rc != -EDQUOT) {
1491 DEBUG_REQ(D_INFO, req, "Failed request with rc = %d\n", rc);
1495 LASSERTF(req->rq_repmsg != NULL, "rc = %d\n", rc);
1496 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
1498 DEBUG_REQ(D_INFO, req, "Can't unpack body\n");
1502 /* set/clear over quota flag for a uid/gid */
1503 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE &&
1504 body->oa.o_valid & (OBD_MD_FLUSRQUOTA | OBD_MD_FLGRPQUOTA)) {
1505 unsigned int qid[MAXQUOTAS] = { body->oa.o_uid, body->oa.o_gid };
1507 CDEBUG(D_QUOTA, "setdq for [%u %u] with valid "LPX64", flags %x\n",
1508 body->oa.o_uid, body->oa.o_gid, body->oa.o_valid,
1510 osc_quota_setdq(cli, qid, body->oa.o_valid, body->oa.o_flags);
1513 osc_update_grant(cli, body);
1518 if (aa->aa_oa->o_valid & OBD_MD_FLCKSUM)
1519 client_cksum = aa->aa_oa->o_cksum; /* save for later */
1521 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1523 CERROR("Unexpected +ve rc %d\n", rc);
1526 LASSERT(req->rq_bulk->bd_nob == aa->aa_requested_nob);
1528 if (sptlrpc_cli_unwrap_bulk_write(req, req->rq_bulk))
1531 if ((aa->aa_oa->o_valid & OBD_MD_FLCKSUM) && client_cksum &&
1532 check_write_checksum(&body->oa, peer, client_cksum,
1533 body->oa.o_cksum, aa->aa_requested_nob,
1534 aa->aa_page_count, aa->aa_ppga,
1535 cksum_type_unpack(aa->aa_oa->o_flags)))
1538 rc = check_write_rcs(req, aa->aa_requested_nob,aa->aa_nio_count,
1539 aa->aa_page_count, aa->aa_ppga);
1543 /* The rest of this function executes only for OST_READs */
1545 /* if unwrap_bulk failed, return -EAGAIN to retry */
1546 rc = sptlrpc_cli_unwrap_bulk_read(req, req->rq_bulk, rc);
1548 GOTO(out, rc = -EAGAIN);
1550 if (rc > aa->aa_requested_nob) {
1551 CERROR("Unexpected rc %d (%d requested)\n", rc,
1552 aa->aa_requested_nob);
1556 if (rc != req->rq_bulk->bd_nob_transferred) {
1557 CERROR ("Unexpected rc %d (%d transferred)\n",
1558 rc, req->rq_bulk->bd_nob_transferred);
1562 if (rc < aa->aa_requested_nob)
1563 handle_short_read(rc, aa->aa_page_count, aa->aa_ppga);
1565 if (body->oa.o_valid & OBD_MD_FLCKSUM) {
1566 static int cksum_counter;
1567 __u32 server_cksum = body->oa.o_cksum;
1570 cksum_type_t cksum_type;
1572 cksum_type = cksum_type_unpack(body->oa.o_valid &OBD_MD_FLFLAGS?
1573 body->oa.o_flags : 0);
1574 client_cksum = osc_checksum_bulk(rc, aa->aa_page_count,
1575 aa->aa_ppga, OST_READ,
1578 if (peer->nid == req->rq_bulk->bd_sender) {
1582 router = libcfs_nid2str(req->rq_bulk->bd_sender);
1585 if (server_cksum != client_cksum) {
1586 LCONSOLE_ERROR_MSG(0x133, "%s: BAD READ CHECKSUM: from "
1587 "%s%s%s inode "DFID" object "DOSTID
1588 " extent ["LPU64"-"LPU64"]\n",
1589 req->rq_import->imp_obd->obd_name,
1590 libcfs_nid2str(peer->nid),
1592 body->oa.o_valid & OBD_MD_FLFID ?
1593 body->oa.o_parent_seq : (__u64)0,
1594 body->oa.o_valid & OBD_MD_FLFID ?
1595 body->oa.o_parent_oid : 0,
1596 body->oa.o_valid & OBD_MD_FLFID ?
1597 body->oa.o_parent_ver : 0,
1598 POSTID(&body->oa.o_oi),
1599 aa->aa_ppga[0]->off,
1600 aa->aa_ppga[aa->aa_page_count-1]->off +
1601 aa->aa_ppga[aa->aa_page_count-1]->count -
1603 CERROR("client %x, server %x, cksum_type %x\n",
1604 client_cksum, server_cksum, cksum_type);
1606 aa->aa_oa->o_cksum = client_cksum;
1610 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1613 } else if (unlikely(client_cksum)) {
1614 static int cksum_missed;
1617 if ((cksum_missed & (-cksum_missed)) == cksum_missed)
1618 CERROR("Checksum %u requested from %s but not sent\n",
1619 cksum_missed, libcfs_nid2str(peer->nid));
1625 lustre_get_wire_obdo(&req->rq_import->imp_connect_data,
1626 aa->aa_oa, &body->oa);
1631 static int osc_brw_redo_request(struct ptlrpc_request *request,
1632 struct osc_brw_async_args *aa, int rc)
1634 struct ptlrpc_request *new_req;
1635 struct osc_brw_async_args *new_aa;
1636 struct osc_async_page *oap;
1639 DEBUG_REQ(rc == -EINPROGRESS ? D_RPCTRACE : D_ERROR, request,
1640 "redo for recoverable error %d", rc);
1642 rc = osc_brw_prep_request(lustre_msg_get_opc(request->rq_reqmsg) ==
1643 OST_WRITE ? OBD_BRW_WRITE :OBD_BRW_READ,
1644 aa->aa_cli, aa->aa_oa,
1645 NULL /* lsm unused by osc currently */,
1646 aa->aa_page_count, aa->aa_ppga,
1647 &new_req, aa->aa_ocapa, 0, 1);
1651 list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
1652 if (oap->oap_request != NULL) {
1653 LASSERTF(request == oap->oap_request,
1654 "request %p != oap_request %p\n",
1655 request, oap->oap_request);
1656 if (oap->oap_interrupted) {
1657 ptlrpc_req_finished(new_req);
1662 /* New request takes over pga and oaps from old request.
1663 * Note that copying a list_head doesn't work, need to move it... */
1665 new_req->rq_interpret_reply = request->rq_interpret_reply;
1666 new_req->rq_async_args = request->rq_async_args;
1667 new_req->rq_commit_cb = request->rq_commit_cb;
1668 /* cap resend delay to the current request timeout, this is similar to
1669 * what ptlrpc does (see after_reply()) */
1670 if (aa->aa_resends > new_req->rq_timeout)
1671 new_req->rq_sent = cfs_time_current_sec() + new_req->rq_timeout;
1673 new_req->rq_sent = cfs_time_current_sec() + aa->aa_resends;
1674 new_req->rq_generation_set = 1;
1675 new_req->rq_import_generation = request->rq_import_generation;
1677 new_aa = ptlrpc_req_async_args(new_req);
1679 INIT_LIST_HEAD(&new_aa->aa_oaps);
1680 list_splice_init(&aa->aa_oaps, &new_aa->aa_oaps);
1681 INIT_LIST_HEAD(&new_aa->aa_exts);
1682 list_splice_init(&aa->aa_exts, &new_aa->aa_exts);
1683 new_aa->aa_resends = aa->aa_resends;
1685 list_for_each_entry(oap, &new_aa->aa_oaps, oap_rpc_item) {
1686 if (oap->oap_request) {
1687 ptlrpc_req_finished(oap->oap_request);
1688 oap->oap_request = ptlrpc_request_addref(new_req);
1692 new_aa->aa_ocapa = aa->aa_ocapa;
1693 aa->aa_ocapa = NULL;
1695 /* XXX: This code will run into problem if we're going to support
1696 * to add a series of BRW RPCs into a self-defined ptlrpc_request_set
1697 * and wait for all of them to be finished. We should inherit request
1698 * set from old request. */
1699 ptlrpcd_add_req(new_req, PDL_POLICY_SAME, -1);
1701 DEBUG_REQ(D_INFO, new_req, "new request");
1706 * ugh, we want disk allocation on the target to happen in offset order. we'll
1707 * follow sedgewicks advice and stick to the dead simple shellsort -- it'll do
1708 * fine for our small page arrays and doesn't require allocation. its an
1709 * insertion sort that swaps elements that are strides apart, shrinking the
1710 * stride down until its '1' and the array is sorted.
1712 static void sort_brw_pages(struct brw_page **array, int num)
1715 struct brw_page *tmp;
1719 for (stride = 1; stride < num ; stride = (stride * 3) + 1)
1724 for (i = stride ; i < num ; i++) {
1727 while (j >= stride && array[j - stride]->off > tmp->off) {
1728 array[j] = array[j - stride];
1733 } while (stride > 1);
1736 static void osc_release_ppga(struct brw_page **ppga, obd_count count)
1738 LASSERT(ppga != NULL);
1739 OBD_FREE(ppga, sizeof(*ppga) * count);
1742 static int brw_interpret(const struct lu_env *env,
1743 struct ptlrpc_request *req, void *data, int rc)
1745 struct osc_brw_async_args *aa = data;
1746 struct osc_extent *ext;
1747 struct osc_extent *tmp;
1748 struct client_obd *cli = aa->aa_cli;
1751 rc = osc_brw_fini_request(req, rc);
1752 CDEBUG(D_INODE, "request %p aa %p rc %d\n", req, aa, rc);
1753 /* When server return -EINPROGRESS, client should always retry
1754 * regardless of the number of times the bulk was resent already. */
1755 if (osc_recoverable_error(rc)) {
1756 if (req->rq_import_generation !=
1757 req->rq_import->imp_generation) {
1758 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1759 ""DOSTID", rc = %d.\n",
1760 req->rq_import->imp_obd->obd_name,
1761 POSTID(&aa->aa_oa->o_oi), rc);
1762 } else if (rc == -EINPROGRESS ||
1763 client_should_resend(aa->aa_resends, aa->aa_cli)) {
1764 rc = osc_brw_redo_request(req, aa, rc);
1766 CERROR("%s: too many resent retries for object: "
1767 ""LPU64":"LPU64", rc = %d.\n",
1768 req->rq_import->imp_obd->obd_name,
1769 POSTID(&aa->aa_oa->o_oi), rc);
1774 else if (rc == -EAGAIN || rc == -EINPROGRESS)
1779 capa_put(aa->aa_ocapa);
1780 aa->aa_ocapa = NULL;
1784 struct obdo *oa = aa->aa_oa;
1785 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
1786 unsigned long valid = 0;
1787 struct cl_object *obj;
1788 struct osc_async_page *last;
1790 last = brw_page2oap(aa->aa_ppga[aa->aa_page_count - 1]);
1791 obj = osc2cl(last->oap_obj);
1793 cl_object_attr_lock(obj);
1794 if (oa->o_valid & OBD_MD_FLBLOCKS) {
1795 attr->cat_blocks = oa->o_blocks;
1796 valid |= CAT_BLOCKS;
1798 if (oa->o_valid & OBD_MD_FLMTIME) {
1799 attr->cat_mtime = oa->o_mtime;
1802 if (oa->o_valid & OBD_MD_FLATIME) {
1803 attr->cat_atime = oa->o_atime;
1806 if (oa->o_valid & OBD_MD_FLCTIME) {
1807 attr->cat_ctime = oa->o_ctime;
1811 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1812 struct lov_oinfo *loi = cl2osc(obj)->oo_oinfo;
1813 loff_t last_off = last->oap_count + last->oap_obj_off;
1815 /* Change file size if this is an out of quota or
1816 * direct IO write and it extends the file size */
1817 if (loi->loi_lvb.lvb_size < last_off) {
1818 attr->cat_size = last_off;
1821 /* Extend KMS if it's not a lockless write */
1822 if (loi->loi_kms < last_off &&
1823 oap2osc_page(last)->ops_srvlock == 0) {
1824 attr->cat_kms = last_off;
1830 cl_object_attr_set(env, obj, attr, valid);
1831 cl_object_attr_unlock(obj);
1833 OBDO_FREE(aa->aa_oa);
1835 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE && rc == 0)
1836 osc_inc_unstable_pages(req);
1838 list_for_each_entry_safe(ext, tmp, &aa->aa_exts, oe_link) {
1839 list_del_init(&ext->oe_link);
1840 osc_extent_finish(env, ext, 1, rc);
1842 LASSERT(list_empty(&aa->aa_exts));
1843 LASSERT(list_empty(&aa->aa_oaps));
1845 cl_req_completion(env, aa->aa_clerq, rc < 0 ? rc :
1846 req->rq_bulk->bd_nob_transferred);
1847 osc_release_ppga(aa->aa_ppga, aa->aa_page_count);
1848 ptlrpc_lprocfs_brw(req, req->rq_bulk->bd_nob_transferred);
1850 client_obd_list_lock(&cli->cl_loi_list_lock);
1851 /* We need to decrement before osc_ap_completion->osc_wake_cache_waiters
1852 * is called so we know whether to go to sync BRWs or wait for more
1853 * RPCs to complete */
1854 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE)
1855 cli->cl_w_in_flight--;
1857 cli->cl_r_in_flight--;
1858 osc_wake_cache_waiters(cli);
1859 client_obd_list_unlock(&cli->cl_loi_list_lock);
1861 osc_io_unplug(env, cli, NULL, PDL_POLICY_SAME);
1865 static void brw_commit(struct ptlrpc_request *req)
1867 /* If osc_inc_unstable_pages (via osc_extent_finish) races with
1868 * this called via the rq_commit_cb, I need to ensure
1869 * osc_dec_unstable_pages is still called. Otherwise unstable
1870 * pages may be leaked. */
1871 spin_lock(&req->rq_lock);
1872 if (likely(req->rq_unstable)) {
1873 req->rq_unstable = 0;
1874 spin_unlock(&req->rq_lock);
1876 osc_dec_unstable_pages(req);
1878 req->rq_committed = 1;
1879 spin_unlock(&req->rq_lock);
1884 * Build an RPC by the list of extent @ext_list. The caller must ensure
1885 * that the total pages in this list are NOT over max pages per RPC.
1886 * Extents in the list must be in OES_RPC state.
1888 int osc_build_rpc(const struct lu_env *env, struct client_obd *cli,
1889 struct list_head *ext_list, int cmd, pdl_policy_t pol)
1891 struct ptlrpc_request *req = NULL;
1892 struct osc_extent *ext;
1893 struct brw_page **pga = NULL;
1894 struct osc_brw_async_args *aa = NULL;
1895 struct obdo *oa = NULL;
1896 struct osc_async_page *oap;
1897 struct osc_async_page *tmp;
1898 struct cl_req *clerq = NULL;
1899 enum cl_req_type crt = (cmd & OBD_BRW_WRITE) ? CRT_WRITE :
1901 struct cl_req_attr *crattr = NULL;
1902 obd_off starting_offset = OBD_OBJECT_EOF;
1903 obd_off ending_offset = 0;
1907 bool soft_sync = false;
1910 struct list_head rpc_list = LIST_HEAD_INIT(rpc_list);
1913 LASSERT(!list_empty(ext_list));
1915 /* add pages into rpc_list to build BRW rpc */
1916 list_for_each_entry(ext, ext_list, oe_link) {
1917 LASSERT(ext->oe_state == OES_RPC);
1918 mem_tight |= ext->oe_memalloc;
1919 list_for_each_entry(oap, &ext->oe_pages, oap_pending_item) {
1921 list_add_tail(&oap->oap_rpc_item, &rpc_list);
1922 if (starting_offset > oap->oap_obj_off)
1923 starting_offset = oap->oap_obj_off;
1925 LASSERT(oap->oap_page_off == 0);
1926 if (ending_offset < oap->oap_obj_off + oap->oap_count)
1927 ending_offset = oap->oap_obj_off +
1930 LASSERT(oap->oap_page_off + oap->oap_count ==
1935 soft_sync = osc_over_unstable_soft_limit(cli);
1937 mpflag = cfs_memory_pressure_get_and_set();
1939 OBD_ALLOC(crattr, sizeof(*crattr));
1941 GOTO(out, rc = -ENOMEM);
1943 OBD_ALLOC(pga, sizeof(*pga) * page_count);
1945 GOTO(out, rc = -ENOMEM);
1949 GOTO(out, rc = -ENOMEM);
1952 list_for_each_entry(oap, &rpc_list, oap_rpc_item) {
1953 struct cl_page *page = oap2cl_page(oap);
1954 if (clerq == NULL) {
1955 clerq = cl_req_alloc(env, page, crt,
1956 1 /* only 1-object rpcs for now */);
1958 GOTO(out, rc = PTR_ERR(clerq));
1961 oap->oap_brw_flags |= OBD_BRW_MEMALLOC;
1963 oap->oap_brw_flags |= OBD_BRW_SOFT_SYNC;
1964 pga[i] = &oap->oap_brw_page;
1965 pga[i]->off = oap->oap_obj_off + oap->oap_page_off;
1966 CDEBUG(0, "put page %p index %lu oap %p flg %x to pga\n",
1967 pga[i]->pg, page_index(oap->oap_page), oap,
1970 cl_req_page_add(env, clerq, page);
1973 /* always get the data for the obdo for the rpc */
1974 LASSERT(clerq != NULL);
1975 crattr->cra_oa = oa;
1976 cl_req_attr_set(env, clerq, crattr, ~0ULL);
1978 rc = cl_req_prep(env, clerq);
1980 CERROR("cl_req_prep failed: %d\n", rc);
1984 sort_brw_pages(pga, page_count);
1985 rc = osc_brw_prep_request(cmd, cli, oa, NULL, page_count,
1986 pga, &req, crattr->cra_capa, 1, 0);
1988 CERROR("prep_req failed: %d\n", rc);
1992 req->rq_commit_cb = brw_commit;
1993 req->rq_interpret_reply = brw_interpret;
1996 req->rq_memalloc = 1;
1998 /* Need to update the timestamps after the request is built in case
1999 * we race with setattr (locally or in queue at OST). If OST gets
2000 * later setattr before earlier BRW (as determined by the request xid),
2001 * the OST will not use BRW timestamps. Sadly, there is no obvious
2002 * way to do this in a single call. bug 10150 */
2003 cl_req_attr_set(env, clerq, crattr,
2004 OBD_MD_FLMTIME|OBD_MD_FLCTIME|OBD_MD_FLATIME);
2006 lustre_msg_set_jobid(req->rq_reqmsg, crattr->cra_jobid);
2008 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2009 aa = ptlrpc_req_async_args(req);
2010 INIT_LIST_HEAD(&aa->aa_oaps);
2011 list_splice_init(&rpc_list, &aa->aa_oaps);
2012 INIT_LIST_HEAD(&aa->aa_exts);
2013 list_splice_init(ext_list, &aa->aa_exts);
2014 aa->aa_clerq = clerq;
2016 /* queued sync pages can be torn down while the pages
2017 * were between the pending list and the rpc */
2019 list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
2020 /* only one oap gets a request reference */
2023 if (oap->oap_interrupted && !req->rq_intr) {
2024 CDEBUG(D_INODE, "oap %p in req %p interrupted\n",
2026 ptlrpc_mark_interrupted(req);
2030 tmp->oap_request = ptlrpc_request_addref(req);
2032 client_obd_list_lock(&cli->cl_loi_list_lock);
2033 starting_offset >>= PAGE_CACHE_SHIFT;
2034 if (cmd == OBD_BRW_READ) {
2035 cli->cl_r_in_flight++;
2036 lprocfs_oh_tally_log2(&cli->cl_read_page_hist, page_count);
2037 lprocfs_oh_tally(&cli->cl_read_rpc_hist, cli->cl_r_in_flight);
2038 lprocfs_oh_tally_log2(&cli->cl_read_offset_hist,
2039 starting_offset + 1);
2041 cli->cl_w_in_flight++;
2042 lprocfs_oh_tally_log2(&cli->cl_write_page_hist, page_count);
2043 lprocfs_oh_tally(&cli->cl_write_rpc_hist, cli->cl_w_in_flight);
2044 lprocfs_oh_tally_log2(&cli->cl_write_offset_hist,
2045 starting_offset + 1);
2047 client_obd_list_unlock(&cli->cl_loi_list_lock);
2049 DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %dr/%dw in flight",
2050 page_count, aa, cli->cl_r_in_flight,
2051 cli->cl_w_in_flight);
2053 /* XXX: Maybe the caller can check the RPC bulk descriptor to
2054 * see which CPU/NUMA node the majority of pages were allocated
2055 * on, and try to assign the async RPC to the CPU core
2056 * (PDL_POLICY_PREFERRED) to reduce cross-CPU memory traffic.
2058 * But on the other hand, we expect that multiple ptlrpcd
2059 * threads and the initial write sponsor can run in parallel,
2060 * especially when data checksum is enabled, which is CPU-bound
2061 * operation and single ptlrpcd thread cannot process in time.
2062 * So more ptlrpcd threads sharing BRW load
2063 * (with PDL_POLICY_ROUND) seems better.
2065 ptlrpcd_add_req(req, pol, -1);
2071 cfs_memory_pressure_restore(mpflag);
2073 if (crattr != NULL) {
2074 capa_put(crattr->cra_capa);
2075 OBD_FREE(crattr, sizeof(*crattr));
2079 LASSERT(req == NULL);
2084 OBD_FREE(pga, sizeof(*pga) * page_count);
2085 /* this should happen rarely and is pretty bad, it makes the
2086 * pending list not follow the dirty order */
2087 while (!list_empty(ext_list)) {
2088 ext = list_entry(ext_list->next, struct osc_extent,
2090 list_del_init(&ext->oe_link);
2091 osc_extent_finish(env, ext, 0, rc);
2093 if (clerq && !IS_ERR(clerq))
2094 cl_req_completion(env, clerq, rc);
2099 static int osc_set_lock_data_with_check(struct ldlm_lock *lock,
2100 struct ldlm_enqueue_info *einfo)
2102 void *data = einfo->ei_cbdata;
2105 LASSERT(lock != NULL);
2106 LASSERT(lock->l_blocking_ast == einfo->ei_cb_bl);
2107 LASSERT(lock->l_resource->lr_type == einfo->ei_type);
2108 LASSERT(lock->l_completion_ast == einfo->ei_cb_cp);
2109 LASSERT(lock->l_glimpse_ast == einfo->ei_cb_gl);
2111 lock_res_and_lock(lock);
2112 spin_lock(&osc_ast_guard);
2114 if (lock->l_ast_data == NULL)
2115 lock->l_ast_data = data;
2116 if (lock->l_ast_data == data)
2119 spin_unlock(&osc_ast_guard);
2120 unlock_res_and_lock(lock);
2125 static int osc_set_data_with_check(struct lustre_handle *lockh,
2126 struct ldlm_enqueue_info *einfo)
2128 struct ldlm_lock *lock = ldlm_handle2lock(lockh);
2132 set = osc_set_lock_data_with_check(lock, einfo);
2133 LDLM_LOCK_PUT(lock);
2135 CERROR("lockh %p, data %p - client evicted?\n",
2136 lockh, einfo->ei_cbdata);
2140 static int osc_change_cbdata(struct obd_export *exp, struct lov_stripe_md *lsm,
2141 ldlm_iterator_t replace, void *data)
2143 struct ldlm_res_id res_id;
2144 struct obd_device *obd = class_exp2obd(exp);
2146 ostid_build_res_name(&lsm->lsm_oi, &res_id);
2147 ldlm_resource_iterate(obd->obd_namespace, &res_id, replace, data);
2151 /* find any ldlm lock of the inode in osc
2155 static int osc_find_cbdata(struct obd_export *exp, struct lov_stripe_md *lsm,
2156 ldlm_iterator_t replace, void *data)
2158 struct ldlm_res_id res_id;
2159 struct obd_device *obd = class_exp2obd(exp);
2162 ostid_build_res_name(&lsm->lsm_oi, &res_id);
2163 rc = ldlm_resource_iterate(obd->obd_namespace, &res_id, replace, data);
2164 if (rc == LDLM_ITER_STOP)
2166 if (rc == LDLM_ITER_CONTINUE)
2171 static int osc_enqueue_fini(struct ptlrpc_request *req, struct ost_lvb *lvb,
2172 obd_enqueue_update_f upcall, void *cookie,
2173 __u64 *flags, int agl, int rc)
2175 int intent = *flags & LDLM_FL_HAS_INTENT;
2179 /* The request was created before ldlm_cli_enqueue call. */
2180 if (rc == ELDLM_LOCK_ABORTED) {
2181 struct ldlm_reply *rep;
2182 rep = req_capsule_server_get(&req->rq_pill,
2185 LASSERT(rep != NULL);
2186 rep->lock_policy_res1 =
2187 ptlrpc_status_ntoh(rep->lock_policy_res1);
2188 if (rep->lock_policy_res1)
2189 rc = rep->lock_policy_res1;
2193 if ((intent != 0 && rc == ELDLM_LOCK_ABORTED && agl == 0) ||
2195 *flags |= LDLM_FL_LVB_READY;
2196 CDEBUG(D_INODE,"got kms "LPU64" blocks "LPU64" mtime "LPU64"\n",
2197 lvb->lvb_size, lvb->lvb_blocks, lvb->lvb_mtime);
2200 /* Call the update callback. */
2201 rc = (*upcall)(cookie, rc);
2205 static int osc_enqueue_interpret(const struct lu_env *env,
2206 struct ptlrpc_request *req,
2207 struct osc_enqueue_args *aa, int rc)
2209 struct ldlm_lock *lock;
2210 struct lustre_handle handle;
2212 struct ost_lvb *lvb;
2214 __u64 *flags = aa->oa_flags;
2216 /* Make a local copy of a lock handle and a mode, because aa->oa_*
2217 * might be freed anytime after lock upcall has been called. */
2218 lustre_handle_copy(&handle, aa->oa_lockh);
2219 mode = aa->oa_ei->ei_mode;
2221 /* ldlm_cli_enqueue is holding a reference on the lock, so it must
2223 lock = ldlm_handle2lock(&handle);
2225 /* Take an additional reference so that a blocking AST that
2226 * ldlm_cli_enqueue_fini() might post for a failed lock, is guaranteed
2227 * to arrive after an upcall has been executed by
2228 * osc_enqueue_fini(). */
2229 ldlm_lock_addref(&handle, mode);
2231 /* Let cl_lock_state_wait fail with -ERESTARTSYS to unuse sublocks. */
2232 OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_ENQUEUE_HANG, 2);
2234 /* Let CP AST to grant the lock first. */
2235 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_ENQ_RACE, 1);
2237 if (aa->oa_agl && rc == ELDLM_LOCK_ABORTED) {
2242 lvb_len = sizeof(*aa->oa_lvb);
2245 /* Complete obtaining the lock procedure. */
2246 rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, aa->oa_ei->ei_type, 1,
2247 mode, flags, lvb, lvb_len, &handle, rc);
2248 /* Complete osc stuff. */
2249 rc = osc_enqueue_fini(req, aa->oa_lvb, aa->oa_upcall, aa->oa_cookie,
2250 flags, aa->oa_agl, rc);
2252 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_CANCEL_RACE, 10);
2254 /* Release the lock for async request. */
2255 if (lustre_handle_is_used(&handle) && rc == ELDLM_OK)
2257 * Releases a reference taken by ldlm_cli_enqueue(), if it is
2258 * not already released by
2259 * ldlm_cli_enqueue_fini()->failed_lock_cleanup()
2261 ldlm_lock_decref(&handle, mode);
2263 LASSERTF(lock != NULL, "lockh %p, req %p, aa %p - client evicted?\n",
2264 aa->oa_lockh, req, aa);
2265 ldlm_lock_decref(&handle, mode);
2266 LDLM_LOCK_PUT(lock);
2270 struct ptlrpc_request_set *PTLRPCD_SET = (void *)1;
2272 /* When enqueuing asynchronously, locks are not ordered, we can obtain a lock
2273 * from the 2nd OSC before a lock from the 1st one. This does not deadlock with
2274 * other synchronous requests, however keeping some locks and trying to obtain
2275 * others may take a considerable amount of time in a case of ost failure; and
2276 * when other sync requests do not get released lock from a client, the client
2277 * is excluded from the cluster -- such scenarious make the life difficult, so
2278 * release locks just after they are obtained. */
2279 int osc_enqueue_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2280 __u64 *flags, ldlm_policy_data_t *policy,
2281 struct ost_lvb *lvb, int kms_valid,
2282 obd_enqueue_update_f upcall, void *cookie,
2283 struct ldlm_enqueue_info *einfo,
2284 struct lustre_handle *lockh,
2285 struct ptlrpc_request_set *rqset, int async, int agl)
2287 struct obd_device *obd = exp->exp_obd;
2288 struct ptlrpc_request *req = NULL;
2289 int intent = *flags & LDLM_FL_HAS_INTENT;
2290 __u64 match_lvb = (agl != 0 ? 0 : LDLM_FL_LVB_READY);
2295 /* Filesystem lock extents are extended to page boundaries so that
2296 * dealing with the page cache is a little smoother. */
2297 policy->l_extent.start -= policy->l_extent.start & ~CFS_PAGE_MASK;
2298 policy->l_extent.end |= ~CFS_PAGE_MASK;
2301 * kms is not valid when either object is completely fresh (so that no
2302 * locks are cached), or object was evicted. In the latter case cached
2303 * lock cannot be used, because it would prime inode state with
2304 * potentially stale LVB.
2309 /* Next, search for already existing extent locks that will cover us */
2310 /* If we're trying to read, we also search for an existing PW lock. The
2311 * VFS and page cache already protect us locally, so lots of readers/
2312 * writers can share a single PW lock.
2314 * There are problems with conversion deadlocks, so instead of
2315 * converting a read lock to a write lock, we'll just enqueue a new
2318 * At some point we should cancel the read lock instead of making them
2319 * send us a blocking callback, but there are problems with canceling
2320 * locks out from other users right now, too. */
2321 mode = einfo->ei_mode;
2322 if (einfo->ei_mode == LCK_PR)
2324 mode = ldlm_lock_match(obd->obd_namespace, *flags | match_lvb, res_id,
2325 einfo->ei_type, policy, mode, lockh, 0);
2327 struct ldlm_lock *matched = ldlm_handle2lock(lockh);
2329 if ((agl != 0) && !ldlm_is_lvb_ready(matched)) {
2330 /* For AGL, if enqueue RPC is sent but the lock is not
2331 * granted, then skip to process this strpe.
2332 * Return -ECANCELED to tell the caller. */
2333 ldlm_lock_decref(lockh, mode);
2334 LDLM_LOCK_PUT(matched);
2336 } else if (osc_set_lock_data_with_check(matched, einfo)) {
2337 *flags |= LDLM_FL_LVB_READY;
2338 /* addref the lock only if not async requests and PW
2339 * lock is matched whereas we asked for PR. */
2340 if (!rqset && einfo->ei_mode != mode)
2341 ldlm_lock_addref(lockh, LCK_PR);
2343 /* I would like to be able to ASSERT here that
2344 * rss <= kms, but I can't, for reasons which
2345 * are explained in lov_enqueue() */
2348 /* We already have a lock, and it's referenced.
2350 * At this point, the cl_lock::cll_state is CLS_QUEUING,
2351 * AGL upcall may change it to CLS_HELD directly. */
2352 (*upcall)(cookie, ELDLM_OK);
2354 if (einfo->ei_mode != mode)
2355 ldlm_lock_decref(lockh, LCK_PW);
2357 /* For async requests, decref the lock. */
2358 ldlm_lock_decref(lockh, einfo->ei_mode);
2359 LDLM_LOCK_PUT(matched);
2362 ldlm_lock_decref(lockh, mode);
2363 LDLM_LOCK_PUT(matched);
2369 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2370 &RQF_LDLM_ENQUEUE_LVB);
2374 rc = ptlrpc_request_pack(req, LUSTRE_DLM_VERSION, LDLM_ENQUEUE);
2376 ptlrpc_request_free(req);
2380 req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER,
2382 ptlrpc_request_set_replen(req);
2385 /* users of osc_enqueue() can pass this flag for ldlm_lock_match() */
2386 *flags &= ~LDLM_FL_BLOCK_GRANTED;
2388 rc = ldlm_cli_enqueue(exp, &req, einfo, res_id, policy, flags, lvb,
2389 sizeof(*lvb), LVB_T_OST, lockh, async);
2392 struct osc_enqueue_args *aa;
2393 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2394 aa = ptlrpc_req_async_args(req);
2397 aa->oa_flags = flags;
2398 aa->oa_upcall = upcall;
2399 aa->oa_cookie = cookie;
2401 aa->oa_lockh = lockh;
2404 req->rq_interpret_reply =
2405 (ptlrpc_interpterer_t)osc_enqueue_interpret;
2406 if (rqset == PTLRPCD_SET)
2407 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
2409 ptlrpc_set_add_req(rqset, req);
2410 } else if (intent) {
2411 ptlrpc_req_finished(req);
2416 rc = osc_enqueue_fini(req, lvb, upcall, cookie, flags, agl, rc);
2418 ptlrpc_req_finished(req);
2423 int osc_match_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2424 __u32 type, ldlm_policy_data_t *policy, __u32 mode,
2425 __u64 *flags, void *data, struct lustre_handle *lockh,
2428 struct obd_device *obd = exp->exp_obd;
2429 __u64 lflags = *flags;
2433 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_MATCH))
2436 /* Filesystem lock extents are extended to page boundaries so that
2437 * dealing with the page cache is a little smoother */
2438 policy->l_extent.start -= policy->l_extent.start & ~CFS_PAGE_MASK;
2439 policy->l_extent.end |= ~CFS_PAGE_MASK;
2441 /* Next, search for already existing extent locks that will cover us */
2442 /* If we're trying to read, we also search for an existing PW lock. The
2443 * VFS and page cache already protect us locally, so lots of readers/
2444 * writers can share a single PW lock. */
2448 rc = ldlm_lock_match(obd->obd_namespace, lflags,
2449 res_id, type, policy, rc, lockh, unref);
2452 if (!osc_set_data_with_check(lockh, data)) {
2453 if (!(lflags & LDLM_FL_TEST_LOCK))
2454 ldlm_lock_decref(lockh, rc);
2458 if (!(lflags & LDLM_FL_TEST_LOCK) && mode != rc) {
2459 ldlm_lock_addref(lockh, LCK_PR);
2460 ldlm_lock_decref(lockh, LCK_PW);
2467 int osc_cancel_base(struct lustre_handle *lockh, __u32 mode)
2471 if (unlikely(mode == LCK_GROUP))
2472 ldlm_lock_decref_and_cancel(lockh, mode);
2474 ldlm_lock_decref(lockh, mode);
2479 static int osc_statfs_interpret(const struct lu_env *env,
2480 struct ptlrpc_request *req,
2481 struct osc_async_args *aa, int rc)
2483 struct obd_statfs *msfs;
2487 /* The request has in fact never been sent
2488 * due to issues at a higher level (LOV).
2489 * Exit immediately since the caller is
2490 * aware of the problem and takes care
2491 * of the clean up */
2494 if ((rc == -ENOTCONN || rc == -EAGAIN) &&
2495 (aa->aa_oi->oi_flags & OBD_STATFS_NODELAY))
2501 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2503 GOTO(out, rc = -EPROTO);
2506 *aa->aa_oi->oi_osfs = *msfs;
2508 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
2512 static int osc_statfs_async(struct obd_export *exp,
2513 struct obd_info *oinfo, __u64 max_age,
2514 struct ptlrpc_request_set *rqset)
2516 struct obd_device *obd = class_exp2obd(exp);
2517 struct ptlrpc_request *req;
2518 struct osc_async_args *aa;
2522 /* We could possibly pass max_age in the request (as an absolute
2523 * timestamp or a "seconds.usec ago") so the target can avoid doing
2524 * extra calls into the filesystem if that isn't necessary (e.g.
2525 * during mount that would help a bit). Having relative timestamps
2526 * is not so great if request processing is slow, while absolute
2527 * timestamps are not ideal because they need time synchronization. */
2528 req = ptlrpc_request_alloc(obd->u.cli.cl_import, &RQF_OST_STATFS);
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 (oinfo->oi_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 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_statfs_interpret;
2548 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2549 aa = ptlrpc_req_async_args(req);
2552 ptlrpc_set_add_req(rqset, req);
2556 static int osc_statfs(const struct lu_env *env, struct obd_export *exp,
2557 struct obd_statfs *osfs, __u64 max_age, __u32 flags)
2559 struct obd_device *obd = class_exp2obd(exp);
2560 struct obd_statfs *msfs;
2561 struct ptlrpc_request *req;
2562 struct obd_import *imp = NULL;
2566 /*Since the request might also come from lprocfs, so we need
2567 *sync this with client_disconnect_export Bug15684*/
2568 down_read(&obd->u.cli.cl_sem);
2569 if (obd->u.cli.cl_import)
2570 imp = class_import_get(obd->u.cli.cl_import);
2571 up_read(&obd->u.cli.cl_sem);
2575 /* We could possibly pass max_age in the request (as an absolute
2576 * timestamp or a "seconds.usec ago") so the target can avoid doing
2577 * extra calls into the filesystem if that isn't necessary (e.g.
2578 * during mount that would help a bit). Having relative timestamps
2579 * is not so great if request processing is slow, while absolute
2580 * timestamps are not ideal because they need time synchronization. */
2581 req = ptlrpc_request_alloc(imp, &RQF_OST_STATFS);
2583 class_import_put(imp);
2588 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2590 ptlrpc_request_free(req);
2593 ptlrpc_request_set_replen(req);
2594 req->rq_request_portal = OST_CREATE_PORTAL;
2595 ptlrpc_at_set_req_timeout(req);
2597 if (flags & OBD_STATFS_NODELAY) {
2598 /* procfs requests not want stat in wait for avoid deadlock */
2599 req->rq_no_resend = 1;
2600 req->rq_no_delay = 1;
2603 rc = ptlrpc_queue_wait(req);
2607 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2609 GOTO(out, rc = -EPROTO);
2616 ptlrpc_req_finished(req);
2620 static int osc_iocontrol(unsigned int cmd, struct obd_export *exp, int len,
2621 void *karg, void *uarg)
2623 struct obd_device *obd = exp->exp_obd;
2624 struct obd_ioctl_data *data = karg;
2628 if (!try_module_get(THIS_MODULE)) {
2629 CERROR("Can't get module. Is it alive?");
2633 case OBD_IOC_CLIENT_RECOVER:
2634 err = ptlrpc_recover_import(obd->u.cli.cl_import,
2635 data->ioc_inlbuf1, 0);
2639 case IOC_OSC_SET_ACTIVE:
2640 err = ptlrpc_set_import_active(obd->u.cli.cl_import,
2643 case OBD_IOC_POLL_QUOTACHECK:
2644 err = osc_quota_poll_check(exp, (struct if_quotacheck *)karg);
2646 case OBD_IOC_PING_TARGET:
2647 err = ptlrpc_obd_ping(obd);
2650 CDEBUG(D_INODE, "unrecognised ioctl %#x by %s\n",
2651 cmd, current_comm());
2652 GOTO(out, err = -ENOTTY);
2655 module_put(THIS_MODULE);
2659 static int osc_get_info(const struct lu_env *env, struct obd_export *exp,
2660 obd_count keylen, void *key, __u32 *vallen, void *val,
2661 struct lov_stripe_md *lsm)
2664 if (!vallen || !val)
2667 if (KEY_IS(KEY_LOCK_TO_STRIPE)) {
2668 __u32 *stripe = val;
2669 *vallen = sizeof(*stripe);
2672 } else if (KEY_IS(KEY_LAST_ID)) {
2673 struct ptlrpc_request *req;
2678 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2679 &RQF_OST_GET_INFO_LAST_ID);
2683 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
2684 RCL_CLIENT, keylen);
2685 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GET_INFO);
2687 ptlrpc_request_free(req);
2691 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
2692 memcpy(tmp, key, keylen);
2694 req->rq_no_delay = req->rq_no_resend = 1;
2695 ptlrpc_request_set_replen(req);
2696 rc = ptlrpc_queue_wait(req);
2700 reply = req_capsule_server_get(&req->rq_pill, &RMF_OBD_ID);
2702 GOTO(out, rc = -EPROTO);
2704 *((obd_id *)val) = *reply;
2706 ptlrpc_req_finished(req);
2708 } else if (KEY_IS(KEY_FIEMAP)) {
2709 struct ll_fiemap_info_key *fm_key =
2710 (struct ll_fiemap_info_key *)key;
2711 struct ldlm_res_id res_id;
2712 ldlm_policy_data_t policy;
2713 struct lustre_handle lockh;
2714 ldlm_mode_t mode = 0;
2715 struct ptlrpc_request *req;
2716 struct ll_user_fiemap *reply;
2720 if (!(fm_key->fiemap.fm_flags & FIEMAP_FLAG_SYNC))
2723 policy.l_extent.start = fm_key->fiemap.fm_start &
2726 if (OBD_OBJECT_EOF - fm_key->fiemap.fm_length <=
2727 fm_key->fiemap.fm_start + PAGE_CACHE_SIZE - 1)
2728 policy.l_extent.end = OBD_OBJECT_EOF;
2730 policy.l_extent.end = (fm_key->fiemap.fm_start +
2731 fm_key->fiemap.fm_length +
2732 PAGE_CACHE_SIZE - 1) & CFS_PAGE_MASK;
2734 ostid_build_res_name(&fm_key->oa.o_oi, &res_id);
2735 mode = ldlm_lock_match(exp->exp_obd->obd_namespace,
2736 LDLM_FL_BLOCK_GRANTED |
2738 &res_id, LDLM_EXTENT, &policy,
2739 LCK_PR | LCK_PW, &lockh, 0);
2740 if (mode) { /* lock is cached on client */
2741 if (mode != LCK_PR) {
2742 ldlm_lock_addref(&lockh, LCK_PR);
2743 ldlm_lock_decref(&lockh, LCK_PW);
2745 } else { /* no cached lock, needs acquire lock on server side */
2746 fm_key->oa.o_valid |= OBD_MD_FLFLAGS;
2747 fm_key->oa.o_flags |= OBD_FL_SRVLOCK;
2751 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2752 &RQF_OST_GET_INFO_FIEMAP);
2754 GOTO(drop_lock, rc = -ENOMEM);
2756 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_KEY,
2757 RCL_CLIENT, keylen);
2758 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_VAL,
2759 RCL_CLIENT, *vallen);
2760 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_VAL,
2761 RCL_SERVER, *vallen);
2763 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GET_INFO);
2765 ptlrpc_request_free(req);
2766 GOTO(drop_lock, rc);
2769 tmp = req_capsule_client_get(&req->rq_pill, &RMF_FIEMAP_KEY);
2770 memcpy(tmp, key, keylen);
2771 tmp = req_capsule_client_get(&req->rq_pill, &RMF_FIEMAP_VAL);
2772 memcpy(tmp, val, *vallen);
2774 ptlrpc_request_set_replen(req);
2775 rc = ptlrpc_queue_wait(req);
2779 reply = req_capsule_server_get(&req->rq_pill, &RMF_FIEMAP_VAL);
2781 GOTO(fini_req, rc = -EPROTO);
2783 memcpy(val, reply, *vallen);
2785 ptlrpc_req_finished(req);
2788 ldlm_lock_decref(&lockh, LCK_PR);
2795 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
2796 obd_count keylen, void *key, obd_count vallen,
2797 void *val, struct ptlrpc_request_set *set)
2799 struct ptlrpc_request *req;
2800 struct obd_device *obd = exp->exp_obd;
2801 struct obd_import *imp = class_exp2cliimp(exp);
2806 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_SHUTDOWN, 10);
2808 if (KEY_IS(KEY_CHECKSUM)) {
2809 if (vallen != sizeof(int))
2811 exp->exp_obd->u.cli.cl_checksum = (*(int *)val) ? 1 : 0;
2815 if (KEY_IS(KEY_SPTLRPC_CONF)) {
2816 sptlrpc_conf_client_adapt(obd);
2820 if (KEY_IS(KEY_FLUSH_CTX)) {
2821 sptlrpc_import_flush_my_ctx(imp);
2825 if (KEY_IS(KEY_CACHE_SET)) {
2826 struct client_obd *cli = &obd->u.cli;
2828 LASSERT(cli->cl_cache == NULL); /* only once */
2829 cli->cl_cache = (struct cl_client_cache *)val;
2830 atomic_inc(&cli->cl_cache->ccc_users);
2831 cli->cl_lru_left = &cli->cl_cache->ccc_lru_left;
2833 /* add this osc into entity list */
2834 LASSERT(list_empty(&cli->cl_lru_osc));
2835 spin_lock(&cli->cl_cache->ccc_lru_lock);
2836 list_add(&cli->cl_lru_osc, &cli->cl_cache->ccc_lru);
2837 spin_unlock(&cli->cl_cache->ccc_lru_lock);
2842 if (KEY_IS(KEY_CACHE_LRU_SHRINK)) {
2843 struct client_obd *cli = &obd->u.cli;
2844 int nr = atomic_read(&cli->cl_lru_in_list) >> 1;
2845 int target = *(int *)val;
2847 nr = osc_lru_shrink(env, cli, min(nr, target), true);
2852 if (!set && !KEY_IS(KEY_GRANT_SHRINK))
2855 /* We pass all other commands directly to OST. Since nobody calls osc
2856 methods directly and everybody is supposed to go through LOV, we
2857 assume lov checked invalid values for us.
2858 The only recognised values so far are evict_by_nid and mds_conn.
2859 Even if something bad goes through, we'd get a -EINVAL from OST
2862 req = ptlrpc_request_alloc(imp, KEY_IS(KEY_GRANT_SHRINK) ?
2863 &RQF_OST_SET_GRANT_INFO :
2868 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
2869 RCL_CLIENT, keylen);
2870 if (!KEY_IS(KEY_GRANT_SHRINK))
2871 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_VAL,
2872 RCL_CLIENT, vallen);
2873 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SET_INFO);
2875 ptlrpc_request_free(req);
2879 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
2880 memcpy(tmp, key, keylen);
2881 tmp = req_capsule_client_get(&req->rq_pill, KEY_IS(KEY_GRANT_SHRINK) ?
2884 memcpy(tmp, val, vallen);
2886 if (KEY_IS(KEY_GRANT_SHRINK)) {
2887 struct osc_grant_args *aa;
2890 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2891 aa = ptlrpc_req_async_args(req);
2894 ptlrpc_req_finished(req);
2897 *oa = ((struct ost_body *)val)->oa;
2899 req->rq_interpret_reply = osc_shrink_grant_interpret;
2902 ptlrpc_request_set_replen(req);
2903 if (!KEY_IS(KEY_GRANT_SHRINK)) {
2904 LASSERT(set != NULL);
2905 ptlrpc_set_add_req(set, req);
2906 ptlrpc_check_set(NULL, set);
2908 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
2913 static int osc_reconnect(const struct lu_env *env,
2914 struct obd_export *exp, struct obd_device *obd,
2915 struct obd_uuid *cluuid,
2916 struct obd_connect_data *data,
2919 struct client_obd *cli = &obd->u.cli;
2921 if (data != NULL && (data->ocd_connect_flags & OBD_CONNECT_GRANT)) {
2924 client_obd_list_lock(&cli->cl_loi_list_lock);
2925 data->ocd_grant = (cli->cl_avail_grant + cli->cl_dirty) ?:
2926 2 * cli_brw_size(obd);
2927 lost_grant = cli->cl_lost_grant;
2928 cli->cl_lost_grant = 0;
2929 client_obd_list_unlock(&cli->cl_loi_list_lock);
2931 CDEBUG(D_RPCTRACE, "ocd_connect_flags: "LPX64" ocd_version: %d"
2932 " ocd_grant: %d, lost: %ld.\n", data->ocd_connect_flags,
2933 data->ocd_version, data->ocd_grant, lost_grant);
2939 static int osc_disconnect(struct obd_export *exp)
2941 struct obd_device *obd = class_exp2obd(exp);
2942 struct llog_ctxt *ctxt;
2945 ctxt = llog_get_context(obd, LLOG_SIZE_REPL_CTXT);
2947 if (obd->u.cli.cl_conn_count == 1) {
2948 /* Flush any remaining cancel messages out to the
2950 llog_sync(ctxt, exp, 0);
2952 llog_ctxt_put(ctxt);
2954 CDEBUG(D_HA, "No LLOG_SIZE_REPL_CTXT found in obd %p\n",
2958 rc = client_disconnect_export(exp);
2960 * Initially we put del_shrink_grant before disconnect_export, but it
2961 * causes the following problem if setup (connect) and cleanup
2962 * (disconnect) are tangled together.
2963 * connect p1 disconnect p2
2964 * ptlrpc_connect_import
2965 * ............... class_manual_cleanup
2968 * ptlrpc_connect_interrupt
2970 * add this client to shrink list
2972 * Bang! pinger trigger the shrink.
2973 * So the osc should be disconnected from the shrink list, after we
2974 * are sure the import has been destroyed. BUG18662
2976 if (obd->u.cli.cl_import == NULL)
2977 osc_del_shrink_grant(&obd->u.cli);
2981 static int osc_import_event(struct obd_device *obd,
2982 struct obd_import *imp,
2983 enum obd_import_event event)
2985 struct client_obd *cli;
2989 LASSERT(imp->imp_obd == obd);
2992 case IMP_EVENT_DISCON: {
2994 client_obd_list_lock(&cli->cl_loi_list_lock);
2995 cli->cl_avail_grant = 0;
2996 cli->cl_lost_grant = 0;
2997 client_obd_list_unlock(&cli->cl_loi_list_lock);
3000 case IMP_EVENT_INACTIVE: {
3001 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_INACTIVE, NULL);
3004 case IMP_EVENT_INVALIDATE: {
3005 struct ldlm_namespace *ns = obd->obd_namespace;
3009 env = cl_env_get(&refcheck);
3013 /* all pages go to failing rpcs due to the invalid
3015 osc_io_unplug(env, cli, NULL, PDL_POLICY_ROUND);
3017 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
3018 cl_env_put(env, &refcheck);
3023 case IMP_EVENT_ACTIVE: {
3024 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVE, NULL);
3027 case IMP_EVENT_OCD: {
3028 struct obd_connect_data *ocd = &imp->imp_connect_data;
3030 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT)
3031 osc_init_grant(&obd->u.cli, ocd);
3034 if (ocd->ocd_connect_flags & OBD_CONNECT_REQPORTAL)
3035 imp->imp_client->cli_request_portal =OST_REQUEST_PORTAL;
3037 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_OCD, NULL);
3040 case IMP_EVENT_DEACTIVATE: {
3041 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_DEACTIVATE, NULL);
3044 case IMP_EVENT_ACTIVATE: {
3045 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVATE, NULL);
3049 CERROR("Unknown import event %d\n", event);
3056 * Determine whether the lock can be canceled before replaying the lock
3057 * during recovery, see bug16774 for detailed information.
3059 * \retval zero the lock can't be canceled
3060 * \retval other ok to cancel
3062 static int osc_cancel_weight(struct ldlm_lock *lock)
3065 * Cancel all unused and granted extent lock.
3067 if (lock->l_resource->lr_type == LDLM_EXTENT &&
3068 lock->l_granted_mode == lock->l_req_mode &&
3069 osc_ldlm_weigh_ast(lock) == 0)
3075 static int brw_queue_work(const struct lu_env *env, void *data)
3077 struct client_obd *cli = data;
3079 CDEBUG(D_CACHE, "Run writeback work for client obd %p.\n", cli);
3081 osc_io_unplug(env, cli, NULL, PDL_POLICY_SAME);
3085 int osc_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
3087 struct client_obd *cli = &obd->u.cli;
3088 struct obd_type *type;
3093 rc = ptlrpcd_addref();
3097 rc = client_obd_setup(obd, lcfg);
3099 GOTO(out_ptlrpcd, rc);
3101 handler = ptlrpcd_alloc_work(cli->cl_import, brw_queue_work, cli);
3102 if (IS_ERR(handler))
3103 GOTO(out_client_setup, rc = PTR_ERR(handler));
3104 cli->cl_writeback_work = handler;
3106 handler = ptlrpcd_alloc_work(cli->cl_import, lru_queue_work, cli);
3107 if (IS_ERR(handler))
3108 GOTO(out_ptlrpcd_work, rc = PTR_ERR(handler));
3109 cli->cl_lru_work = handler;
3111 rc = osc_quota_setup(obd);
3113 GOTO(out_ptlrpcd_work, rc);
3115 cli->cl_grant_shrink_interval = GRANT_SHRINK_INTERVAL;
3118 obd->obd_vars = lprocfs_osc_obd_vars;
3120 /* If this is true then both client (osc) and server (osp) are on the
3121 * same node. The osp layer if loaded first will register the osc proc
3122 * directory. In that case this obd_device will be attached its proc
3123 * tree to type->typ_procsym instead of obd->obd_type->typ_procroot. */
3124 type = class_search_type(LUSTRE_OSP_NAME);
3125 if (type && type->typ_procsym) {
3126 obd->obd_proc_entry = lprocfs_seq_register(obd->obd_name,
3128 obd->obd_vars, obd);
3129 if (IS_ERR(obd->obd_proc_entry)) {
3130 rc = PTR_ERR(obd->obd_proc_entry);
3131 CERROR("error %d setting up lprocfs for %s\n", rc,
3133 obd->obd_proc_entry = NULL;
3136 rc = lprocfs_seq_obd_setup(obd);
3139 /* If the basic OSC proc tree construction succeeded then
3140 * lets do the rest. */
3142 lproc_osc_attach_seqstat(obd);
3143 sptlrpc_lprocfs_cliobd_attach(obd);
3144 ptlrpc_lprocfs_register_obd(obd);
3147 /* We need to allocate a few requests more, because
3148 * brw_interpret tries to create new requests before freeing
3149 * previous ones, Ideally we want to have 2x max_rpcs_in_flight
3150 * reserved, but I'm afraid that might be too much wasted RAM
3151 * in fact, so 2 is just my guess and still should work. */
3152 cli->cl_import->imp_rq_pool =
3153 ptlrpc_init_rq_pool(cli->cl_max_rpcs_in_flight + 2,
3155 ptlrpc_add_rqs_to_pool);
3157 INIT_LIST_HEAD(&cli->cl_grant_shrink_list);
3158 ns_register_cancel(obd->obd_namespace, osc_cancel_weight);
3162 if (cli->cl_writeback_work != NULL) {
3163 ptlrpcd_destroy_work(cli->cl_writeback_work);
3164 cli->cl_writeback_work = NULL;
3166 if (cli->cl_lru_work != NULL) {
3167 ptlrpcd_destroy_work(cli->cl_lru_work);
3168 cli->cl_lru_work = NULL;
3171 client_obd_cleanup(obd);
3177 static int osc_precleanup(struct obd_device *obd, enum obd_cleanup_stage stage)
3183 case OBD_CLEANUP_EARLY: {
3184 struct obd_import *imp;
3185 imp = obd->u.cli.cl_import;
3186 CDEBUG(D_HA, "Deactivating import %s\n", obd->obd_name);
3187 /* ptlrpc_abort_inflight to stop an mds_lov_synchronize */
3188 ptlrpc_deactivate_import(imp);
3189 spin_lock(&imp->imp_lock);
3190 imp->imp_pingable = 0;
3191 spin_unlock(&imp->imp_lock);
3194 case OBD_CLEANUP_EXPORTS: {
3195 struct client_obd *cli = &obd->u.cli;
3197 * for echo client, export may be on zombie list, wait for
3198 * zombie thread to cull it, because cli.cl_import will be
3199 * cleared in client_disconnect_export():
3200 * class_export_destroy() -> obd_cleanup() ->
3201 * echo_device_free() -> echo_client_cleanup() ->
3202 * obd_disconnect() -> osc_disconnect() ->
3203 * client_disconnect_export()
3205 obd_zombie_barrier();
3206 if (cli->cl_writeback_work) {
3207 ptlrpcd_destroy_work(cli->cl_writeback_work);
3208 cli->cl_writeback_work = NULL;
3210 if (cli->cl_lru_work) {
3211 ptlrpcd_destroy_work(cli->cl_lru_work);
3212 cli->cl_lru_work = NULL;
3214 obd_cleanup_client_import(obd);
3215 ptlrpc_lprocfs_unregister_obd(obd);
3216 lprocfs_obd_cleanup(obd);
3217 rc = obd_llog_finish(obd, 0);
3219 CERROR("failed to cleanup llogging subsystems\n");
3226 int osc_cleanup(struct obd_device *obd)
3228 struct client_obd *cli = &obd->u.cli;
3234 if (cli->cl_cache != NULL) {
3235 LASSERT(atomic_read(&cli->cl_cache->ccc_users) > 0);
3236 spin_lock(&cli->cl_cache->ccc_lru_lock);
3237 list_del_init(&cli->cl_lru_osc);
3238 spin_unlock(&cli->cl_cache->ccc_lru_lock);
3239 cli->cl_lru_left = NULL;
3240 atomic_dec(&cli->cl_cache->ccc_users);
3241 cli->cl_cache = NULL;
3244 /* free memory of osc quota cache */
3245 osc_quota_cleanup(obd);
3247 rc = client_obd_cleanup(obd);
3253 int osc_process_config_base(struct obd_device *obd, struct lustre_cfg *lcfg)
3255 int rc = class_process_proc_seq_param(PARAM_OSC, obd->obd_vars,
3257 return rc > 0 ? 0: rc;
3260 static int osc_process_config(struct obd_device *obd, obd_count len, void *buf)
3262 return osc_process_config_base(obd, buf);
3265 struct obd_ops osc_obd_ops = {
3266 .o_owner = THIS_MODULE,
3267 .o_setup = osc_setup,
3268 .o_precleanup = osc_precleanup,
3269 .o_cleanup = osc_cleanup,
3270 .o_add_conn = client_import_add_conn,
3271 .o_del_conn = client_import_del_conn,
3272 .o_connect = client_connect_import,
3273 .o_reconnect = osc_reconnect,
3274 .o_disconnect = osc_disconnect,
3275 .o_statfs = osc_statfs,
3276 .o_statfs_async = osc_statfs_async,
3277 .o_unpackmd = osc_unpackmd,
3278 .o_create = osc_create,
3279 .o_destroy = osc_destroy,
3280 .o_getattr = osc_getattr,
3281 .o_getattr_async = osc_getattr_async,
3282 .o_setattr = osc_setattr,
3283 .o_setattr_async = osc_setattr_async,
3284 .o_change_cbdata = osc_change_cbdata,
3285 .o_find_cbdata = osc_find_cbdata,
3286 .o_iocontrol = osc_iocontrol,
3287 .o_get_info = osc_get_info,
3288 .o_set_info_async = osc_set_info_async,
3289 .o_import_event = osc_import_event,
3290 .o_process_config = osc_process_config,
3291 .o_quotactl = osc_quotactl,
3292 .o_quotacheck = osc_quotacheck,
3295 extern struct lu_kmem_descr osc_caches[];
3296 extern spinlock_t osc_ast_guard;
3297 extern struct lock_class_key osc_ast_guard_class;
3299 int __init osc_init(void)
3301 bool enable_proc = true;
3302 struct obd_type *type;
3306 /* print an address of _any_ initialized kernel symbol from this
3307 * module, to allow debugging with gdb that doesn't support data
3308 * symbols from modules.*/
3309 CDEBUG(D_INFO, "Lustre OSC module (%p).\n", &osc_caches);
3311 rc = lu_kmem_init(osc_caches);
3315 type = class_search_type(LUSTRE_OSP_NAME);
3316 if (type != NULL && type->typ_procsym != NULL)
3317 enable_proc = false;
3319 rc = class_register_type(&osc_obd_ops, NULL, enable_proc, NULL,
3320 #ifndef HAVE_ONLY_PROCFS_SEQ
3323 LUSTRE_OSC_NAME, &osc_device_type);
3325 lu_kmem_fini(osc_caches);
3329 spin_lock_init(&osc_ast_guard);
3330 lockdep_set_class(&osc_ast_guard, &osc_ast_guard_class);
3336 static void /*__exit*/ osc_exit(void)
3338 class_unregister_type(LUSTRE_OSC_NAME);
3339 lu_kmem_fini(osc_caches);
3342 MODULE_AUTHOR("Sun Microsystems, Inc. <http://www.lustre.org/>");
3343 MODULE_DESCRIPTION("Lustre Object Storage Client (OSC)");
3344 MODULE_LICENSE("GPL");
3346 cfs_module(osc, LUSTRE_VERSION_STRING, osc_init, osc_exit);