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, 2012, 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>
56 #include <lustre_ha.h>
57 #include <lprocfs_status.h>
58 #include <lustre_log.h>
59 #include <lustre_debug.h>
60 #include <lustre_param.h>
61 #include <lustre_fid.h>
62 #include "osc_internal.h"
63 #include "osc_cl_internal.h"
65 static void osc_release_ppga(struct brw_page **ppga, obd_count count);
66 static int brw_interpret(const struct lu_env *env,
67 struct ptlrpc_request *req, void *data, int rc);
68 int osc_cleanup(struct obd_device *obd);
70 /* Pack OSC object metadata for disk storage (LE byte order). */
71 static int osc_packmd(struct obd_export *exp, struct lov_mds_md **lmmp,
72 struct lov_stripe_md *lsm)
77 lmm_size = sizeof(**lmmp);
82 OBD_FREE(*lmmp, lmm_size);
88 OBD_ALLOC(*lmmp, lmm_size);
94 LASSERT(lsm->lsm_object_id);
95 LASSERT_SEQ_IS_MDT(lsm->lsm_object_seq);
96 (*lmmp)->lmm_object_id = cpu_to_le64(lsm->lsm_object_id);
97 (*lmmp)->lmm_object_seq = cpu_to_le64(lsm->lsm_object_seq);
103 /* Unpack OSC object metadata from disk storage (LE byte order). */
104 static int osc_unpackmd(struct obd_export *exp, struct lov_stripe_md **lsmp,
105 struct lov_mds_md *lmm, int lmm_bytes)
108 struct obd_import *imp = class_exp2cliimp(exp);
112 if (lmm_bytes < sizeof (*lmm)) {
113 CERROR("lov_mds_md too small: %d, need %d\n",
114 lmm_bytes, (int)sizeof(*lmm));
117 /* XXX LOV_MAGIC etc check? */
119 if (lmm->lmm_object_id == 0) {
120 CERROR("lov_mds_md: zero lmm_object_id\n");
125 lsm_size = lov_stripe_md_size(1);
129 if (*lsmp != NULL && lmm == NULL) {
130 OBD_FREE((*lsmp)->lsm_oinfo[0], sizeof(struct lov_oinfo));
131 OBD_FREE(*lsmp, lsm_size);
137 OBD_ALLOC(*lsmp, lsm_size);
140 OBD_ALLOC((*lsmp)->lsm_oinfo[0], sizeof(struct lov_oinfo));
141 if ((*lsmp)->lsm_oinfo[0] == NULL) {
142 OBD_FREE(*lsmp, lsm_size);
145 loi_init((*lsmp)->lsm_oinfo[0]);
149 /* XXX zero *lsmp? */
150 (*lsmp)->lsm_object_id = le64_to_cpu (lmm->lmm_object_id);
151 (*lsmp)->lsm_object_seq = le64_to_cpu (lmm->lmm_object_seq);
152 LASSERT((*lsmp)->lsm_object_id);
153 LASSERT_SEQ_IS_MDT((*lsmp)->lsm_object_seq);
157 (imp->imp_connect_data.ocd_connect_flags & OBD_CONNECT_MAXBYTES))
158 (*lsmp)->lsm_maxbytes = imp->imp_connect_data.ocd_maxbytes;
160 (*lsmp)->lsm_maxbytes = LUSTRE_STRIPE_MAXBYTES;
165 static inline void osc_pack_capa(struct ptlrpc_request *req,
166 struct ost_body *body, void *capa)
168 struct obd_capa *oc = (struct obd_capa *)capa;
169 struct lustre_capa *c;
174 c = req_capsule_client_get(&req->rq_pill, &RMF_CAPA1);
177 body->oa.o_valid |= OBD_MD_FLOSSCAPA;
178 DEBUG_CAPA(D_SEC, c, "pack");
181 static inline void osc_pack_req_body(struct ptlrpc_request *req,
182 struct obd_info *oinfo)
184 struct ost_body *body;
186 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
189 lustre_set_wire_obdo(&body->oa, oinfo->oi_oa);
190 osc_pack_capa(req, body, oinfo->oi_capa);
193 static inline void osc_set_capa_size(struct ptlrpc_request *req,
194 const struct req_msg_field *field,
198 req_capsule_set_size(&req->rq_pill, field, RCL_CLIENT, 0);
200 /* it is already calculated as sizeof struct obd_capa */
204 static int osc_getattr_interpret(const struct lu_env *env,
205 struct ptlrpc_request *req,
206 struct osc_async_args *aa, int rc)
208 struct ost_body *body;
214 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
216 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
217 lustre_get_wire_obdo(aa->aa_oi->oi_oa, &body->oa);
219 /* This should really be sent by the OST */
220 aa->aa_oi->oi_oa->o_blksize = PTLRPC_MAX_BRW_SIZE;
221 aa->aa_oi->oi_oa->o_valid |= OBD_MD_FLBLKSZ;
223 CDEBUG(D_INFO, "can't unpack ost_body\n");
225 aa->aa_oi->oi_oa->o_valid = 0;
228 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
232 static int osc_getattr_async(struct obd_export *exp, struct obd_info *oinfo,
233 struct ptlrpc_request_set *set)
235 struct ptlrpc_request *req;
236 struct osc_async_args *aa;
240 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
244 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
245 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
247 ptlrpc_request_free(req);
251 osc_pack_req_body(req, oinfo);
253 ptlrpc_request_set_replen(req);
254 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_getattr_interpret;
256 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
257 aa = ptlrpc_req_async_args(req);
260 ptlrpc_set_add_req(set, req);
264 static int osc_getattr(const struct lu_env *env, struct obd_export *exp,
265 struct obd_info *oinfo)
267 struct ptlrpc_request *req;
268 struct ost_body *body;
272 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
276 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
277 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
279 ptlrpc_request_free(req);
283 osc_pack_req_body(req, oinfo);
285 ptlrpc_request_set_replen(req);
287 rc = ptlrpc_queue_wait(req);
291 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
293 GOTO(out, rc = -EPROTO);
295 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
296 lustre_get_wire_obdo(oinfo->oi_oa, &body->oa);
298 /* This should really be sent by the OST */
299 oinfo->oi_oa->o_blksize = PTLRPC_MAX_BRW_SIZE;
300 oinfo->oi_oa->o_valid |= OBD_MD_FLBLKSZ;
304 ptlrpc_req_finished(req);
308 static int osc_setattr(const struct lu_env *env, struct obd_export *exp,
309 struct obd_info *oinfo, struct obd_trans_info *oti)
311 struct ptlrpc_request *req;
312 struct ost_body *body;
316 LASSERT(oinfo->oi_oa->o_valid & OBD_MD_FLGROUP);
318 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
322 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
323 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
325 ptlrpc_request_free(req);
329 osc_pack_req_body(req, oinfo);
331 ptlrpc_request_set_replen(req);
333 rc = ptlrpc_queue_wait(req);
337 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
339 GOTO(out, rc = -EPROTO);
341 lustre_get_wire_obdo(oinfo->oi_oa, &body->oa);
345 ptlrpc_req_finished(req);
349 static int osc_setattr_interpret(const struct lu_env *env,
350 struct ptlrpc_request *req,
351 struct osc_setattr_args *sa, int rc)
353 struct ost_body *body;
359 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
361 GOTO(out, rc = -EPROTO);
363 lustre_get_wire_obdo(sa->sa_oa, &body->oa);
365 rc = sa->sa_upcall(sa->sa_cookie, rc);
369 int osc_setattr_async_base(struct obd_export *exp, struct obd_info *oinfo,
370 struct obd_trans_info *oti,
371 obd_enqueue_update_f upcall, void *cookie,
372 struct ptlrpc_request_set *rqset)
374 struct ptlrpc_request *req;
375 struct osc_setattr_args *sa;
379 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
383 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
384 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
386 ptlrpc_request_free(req);
390 if (oti && oinfo->oi_oa->o_valid & OBD_MD_FLCOOKIE)
391 oinfo->oi_oa->o_lcookie = *oti->oti_logcookies;
393 osc_pack_req_body(req, oinfo);
395 ptlrpc_request_set_replen(req);
397 /* do mds to ost setattr asynchronously */
399 /* Do not wait for response. */
400 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
402 req->rq_interpret_reply =
403 (ptlrpc_interpterer_t)osc_setattr_interpret;
405 CLASSERT (sizeof(*sa) <= sizeof(req->rq_async_args));
406 sa = ptlrpc_req_async_args(req);
407 sa->sa_oa = oinfo->oi_oa;
408 sa->sa_upcall = upcall;
409 sa->sa_cookie = cookie;
411 if (rqset == PTLRPCD_SET)
412 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
414 ptlrpc_set_add_req(rqset, req);
420 static int osc_setattr_async(struct obd_export *exp, struct obd_info *oinfo,
421 struct obd_trans_info *oti,
422 struct ptlrpc_request_set *rqset)
424 return osc_setattr_async_base(exp, oinfo, oti,
425 oinfo->oi_cb_up, oinfo, rqset);
428 int osc_real_create(struct obd_export *exp, struct obdo *oa,
429 struct lov_stripe_md **ea, struct obd_trans_info *oti)
431 struct ptlrpc_request *req;
432 struct ost_body *body;
433 struct lov_stripe_md *lsm;
442 rc = obd_alloc_memmd(exp, &lsm);
447 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_CREATE);
449 GOTO(out, rc = -ENOMEM);
451 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_CREATE);
453 ptlrpc_request_free(req);
457 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
459 lustre_set_wire_obdo(&body->oa, oa);
461 ptlrpc_request_set_replen(req);
463 if ((oa->o_valid & OBD_MD_FLFLAGS) &&
464 oa->o_flags == OBD_FL_DELORPHAN) {
466 "delorphan from OST integration");
467 /* Don't resend the delorphan req */
468 req->rq_no_resend = req->rq_no_delay = 1;
471 rc = ptlrpc_queue_wait(req);
475 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
477 GOTO(out_req, rc = -EPROTO);
479 lustre_get_wire_obdo(oa, &body->oa);
481 /* This should really be sent by the OST */
482 oa->o_blksize = PTLRPC_MAX_BRW_SIZE;
483 oa->o_valid |= OBD_MD_FLBLKSZ;
485 /* XXX LOV STACKING: the lsm that is passed to us from LOV does not
486 * have valid lsm_oinfo data structs, so don't go touching that.
487 * This needs to be fixed in a big way.
489 lsm->lsm_object_id = oa->o_id;
490 lsm->lsm_object_seq = oa->o_seq;
494 oti->oti_transno = lustre_msg_get_transno(req->rq_repmsg);
496 if (oa->o_valid & OBD_MD_FLCOOKIE) {
497 if (!oti->oti_logcookies)
498 oti_alloc_cookies(oti, 1);
499 *oti->oti_logcookies = oa->o_lcookie;
503 CDEBUG(D_HA, "transno: "LPD64"\n",
504 lustre_msg_get_transno(req->rq_repmsg));
506 ptlrpc_req_finished(req);
509 obd_free_memmd(exp, &lsm);
513 int osc_punch_base(struct obd_export *exp, struct obd_info *oinfo,
514 obd_enqueue_update_f upcall, void *cookie,
515 struct ptlrpc_request_set *rqset)
517 struct ptlrpc_request *req;
518 struct osc_setattr_args *sa;
519 struct ost_body *body;
523 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_PUNCH);
527 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
528 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_PUNCH);
530 ptlrpc_request_free(req);
533 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
534 ptlrpc_at_set_req_timeout(req);
536 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
538 lustre_set_wire_obdo(&body->oa, oinfo->oi_oa);
539 osc_pack_capa(req, body, oinfo->oi_capa);
541 ptlrpc_request_set_replen(req);
543 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_setattr_interpret;
544 CLASSERT (sizeof(*sa) <= sizeof(req->rq_async_args));
545 sa = ptlrpc_req_async_args(req);
546 sa->sa_oa = oinfo->oi_oa;
547 sa->sa_upcall = upcall;
548 sa->sa_cookie = cookie;
549 if (rqset == PTLRPCD_SET)
550 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
552 ptlrpc_set_add_req(rqset, req);
557 static int osc_punch(const struct lu_env *env, struct obd_export *exp,
558 struct obd_info *oinfo, struct obd_trans_info *oti,
559 struct ptlrpc_request_set *rqset)
561 oinfo->oi_oa->o_size = oinfo->oi_policy.l_extent.start;
562 oinfo->oi_oa->o_blocks = oinfo->oi_policy.l_extent.end;
563 oinfo->oi_oa->o_valid |= OBD_MD_FLSIZE | OBD_MD_FLBLOCKS;
564 return osc_punch_base(exp, oinfo,
565 oinfo->oi_cb_up, oinfo, rqset);
568 static int osc_sync_interpret(const struct lu_env *env,
569 struct ptlrpc_request *req,
572 struct osc_fsync_args *fa = arg;
573 struct ost_body *body;
579 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
581 CERROR ("can't unpack ost_body\n");
582 GOTO(out, rc = -EPROTO);
585 *fa->fa_oi->oi_oa = body->oa;
587 rc = fa->fa_upcall(fa->fa_cookie, rc);
591 int osc_sync_base(struct obd_export *exp, struct obd_info *oinfo,
592 obd_enqueue_update_f upcall, void *cookie,
593 struct ptlrpc_request_set *rqset)
595 struct ptlrpc_request *req;
596 struct ost_body *body;
597 struct osc_fsync_args *fa;
601 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SYNC);
605 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
606 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SYNC);
608 ptlrpc_request_free(req);
612 /* overload the size and blocks fields in the oa with start/end */
613 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
615 lustre_set_wire_obdo(&body->oa, oinfo->oi_oa);
616 osc_pack_capa(req, body, oinfo->oi_capa);
618 ptlrpc_request_set_replen(req);
619 req->rq_interpret_reply = osc_sync_interpret;
621 CLASSERT(sizeof(*fa) <= sizeof(req->rq_async_args));
622 fa = ptlrpc_req_async_args(req);
624 fa->fa_upcall = upcall;
625 fa->fa_cookie = cookie;
627 if (rqset == PTLRPCD_SET)
628 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
630 ptlrpc_set_add_req(rqset, req);
635 static int osc_sync(const struct lu_env *env, struct obd_export *exp,
636 struct obd_info *oinfo, obd_size start, obd_size end,
637 struct ptlrpc_request_set *set)
642 CDEBUG(D_INFO, "oa NULL\n");
646 oinfo->oi_oa->o_size = start;
647 oinfo->oi_oa->o_blocks = end;
648 oinfo->oi_oa->o_valid |= (OBD_MD_FLSIZE | OBD_MD_FLBLOCKS);
650 RETURN(osc_sync_base(exp, oinfo, oinfo->oi_cb_up, oinfo, set));
653 /* Find and cancel locally locks matched by @mode in the resource found by
654 * @objid. Found locks are added into @cancel list. Returns the amount of
655 * locks added to @cancels list. */
656 static int osc_resource_get_unused(struct obd_export *exp, struct obdo *oa,
658 ldlm_mode_t mode, int lock_flags)
660 struct ldlm_namespace *ns = exp->exp_obd->obd_namespace;
661 struct ldlm_res_id res_id;
662 struct ldlm_resource *res;
666 /* Return, i.e. cancel nothing, only if ELC is supported (flag in
667 * export) but disabled through procfs (flag in NS).
669 * This distinguishes from a case when ELC is not supported originally,
670 * when we still want to cancel locks in advance and just cancel them
671 * locally, without sending any RPC. */
672 if (exp_connect_cancelset(exp) && !ns_connect_cancelset(ns))
675 ostid_build_res_name(&oa->o_oi, &res_id);
676 res = ldlm_resource_get(ns, NULL, &res_id, 0, 0);
680 LDLM_RESOURCE_ADDREF(res);
681 count = ldlm_cancel_resource_local(res, cancels, NULL, mode,
682 lock_flags, 0, NULL);
683 LDLM_RESOURCE_DELREF(res);
684 ldlm_resource_putref(res);
688 static int osc_destroy_interpret(const struct lu_env *env,
689 struct ptlrpc_request *req, void *data,
692 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
694 cfs_atomic_dec(&cli->cl_destroy_in_flight);
695 cfs_waitq_signal(&cli->cl_destroy_waitq);
699 static int osc_can_send_destroy(struct client_obd *cli)
701 if (cfs_atomic_inc_return(&cli->cl_destroy_in_flight) <=
702 cli->cl_max_rpcs_in_flight) {
703 /* The destroy request can be sent */
706 if (cfs_atomic_dec_return(&cli->cl_destroy_in_flight) <
707 cli->cl_max_rpcs_in_flight) {
709 * The counter has been modified between the two atomic
712 cfs_waitq_signal(&cli->cl_destroy_waitq);
717 int osc_create(const struct lu_env *env, struct obd_export *exp,
718 struct obdo *oa, struct lov_stripe_md **ea,
719 struct obd_trans_info *oti)
726 LASSERT(oa->o_valid & OBD_MD_FLGROUP);
728 if ((oa->o_valid & OBD_MD_FLFLAGS) &&
729 oa->o_flags == OBD_FL_RECREATE_OBJS) {
730 RETURN(osc_real_create(exp, oa, ea, oti));
733 if (!fid_seq_is_mdt(oa->o_seq))
734 RETURN(osc_real_create(exp, oa, ea, oti));
736 /* we should not get here anymore */
742 /* Destroy requests can be async always on the client, and we don't even really
743 * care about the return code since the client cannot do anything at all about
745 * When the MDS is unlinking a filename, it saves the file objects into a
746 * recovery llog, and these object records are cancelled when the OST reports
747 * they were destroyed and sync'd to disk (i.e. transaction committed).
748 * If the client dies, or the OST is down when the object should be destroyed,
749 * the records are not cancelled, and when the OST reconnects to the MDS next,
750 * it will retrieve the llog unlink logs and then sends the log cancellation
751 * cookies to the MDS after committing destroy transactions. */
752 static int osc_destroy(const struct lu_env *env, struct obd_export *exp,
753 struct obdo *oa, struct lov_stripe_md *ea,
754 struct obd_trans_info *oti, struct obd_export *md_export,
757 struct client_obd *cli = &exp->exp_obd->u.cli;
758 struct ptlrpc_request *req;
759 struct ost_body *body;
760 CFS_LIST_HEAD(cancels);
765 CDEBUG(D_INFO, "oa NULL\n");
769 count = osc_resource_get_unused(exp, oa, &cancels, LCK_PW,
770 LDLM_FL_DISCARD_DATA);
772 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_DESTROY);
774 ldlm_lock_list_put(&cancels, l_bl_ast, count);
778 osc_set_capa_size(req, &RMF_CAPA1, (struct obd_capa *)capa);
779 rc = ldlm_prep_elc_req(exp, req, LUSTRE_OST_VERSION, OST_DESTROY,
782 ptlrpc_request_free(req);
786 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
787 ptlrpc_at_set_req_timeout(req);
789 if (oti != NULL && oa->o_valid & OBD_MD_FLCOOKIE)
790 oa->o_lcookie = *oti->oti_logcookies;
791 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
793 lustre_set_wire_obdo(&body->oa, oa);
795 osc_pack_capa(req, body, (struct obd_capa *)capa);
796 ptlrpc_request_set_replen(req);
798 /* If osc_destory is for destroying the unlink orphan,
799 * sent from MDT to OST, which should not be blocked here,
800 * because the process might be triggered by ptlrpcd, and
801 * it is not good to block ptlrpcd thread (b=16006)*/
802 if (!(oa->o_flags & OBD_FL_DELORPHAN)) {
803 req->rq_interpret_reply = osc_destroy_interpret;
804 if (!osc_can_send_destroy(cli)) {
805 struct l_wait_info lwi = LWI_INTR(LWI_ON_SIGNAL_NOOP,
809 * Wait until the number of on-going destroy RPCs drops
810 * under max_rpc_in_flight
812 l_wait_event_exclusive(cli->cl_destroy_waitq,
813 osc_can_send_destroy(cli), &lwi);
817 /* Do not wait for response */
818 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
822 static void osc_announce_cached(struct client_obd *cli, struct obdo *oa,
825 obd_flag bits = OBD_MD_FLBLOCKS|OBD_MD_FLGRANT;
827 LASSERT(!(oa->o_valid & bits));
830 client_obd_list_lock(&cli->cl_loi_list_lock);
831 oa->o_dirty = cli->cl_dirty;
832 if (unlikely(cli->cl_dirty - cli->cl_dirty_transit >
833 cli->cl_dirty_max)) {
834 CERROR("dirty %lu - %lu > dirty_max %lu\n",
835 cli->cl_dirty, cli->cl_dirty_transit, cli->cl_dirty_max);
837 } else if (unlikely(cfs_atomic_read(&obd_dirty_pages) -
838 cfs_atomic_read(&obd_dirty_transit_pages) >
839 (long)(obd_max_dirty_pages + 1))) {
840 /* The cfs_atomic_read() allowing the cfs_atomic_inc() are
841 * not covered by a lock thus they may safely race and trip
842 * this CERROR() unless we add in a small fudge factor (+1). */
843 CERROR("dirty %d - %d > system dirty_max %d\n",
844 cfs_atomic_read(&obd_dirty_pages),
845 cfs_atomic_read(&obd_dirty_transit_pages),
846 obd_max_dirty_pages);
848 } else if (unlikely(cli->cl_dirty_max - cli->cl_dirty > 0x7fffffff)) {
849 CERROR("dirty %lu - dirty_max %lu too big???\n",
850 cli->cl_dirty, cli->cl_dirty_max);
853 long max_in_flight = (cli->cl_max_pages_per_rpc <<
855 (cli->cl_max_rpcs_in_flight + 1);
856 oa->o_undirty = max(cli->cl_dirty_max, max_in_flight);
858 oa->o_grant = cli->cl_avail_grant + cli->cl_reserved_grant;
859 oa->o_dropped = cli->cl_lost_grant;
860 cli->cl_lost_grant = 0;
861 client_obd_list_unlock(&cli->cl_loi_list_lock);
862 CDEBUG(D_CACHE,"dirty: "LPU64" undirty: %u dropped %u grant: "LPU64"\n",
863 oa->o_dirty, oa->o_undirty, oa->o_dropped, oa->o_grant);
867 void osc_update_next_shrink(struct client_obd *cli)
869 cli->cl_next_shrink_grant =
870 cfs_time_shift(cli->cl_grant_shrink_interval);
871 CDEBUG(D_CACHE, "next time %ld to shrink grant \n",
872 cli->cl_next_shrink_grant);
875 static void __osc_update_grant(struct client_obd *cli, obd_size grant)
877 client_obd_list_lock(&cli->cl_loi_list_lock);
878 cli->cl_avail_grant += grant;
879 client_obd_list_unlock(&cli->cl_loi_list_lock);
882 static void osc_update_grant(struct client_obd *cli, struct ost_body *body)
884 if (body->oa.o_valid & OBD_MD_FLGRANT) {
885 CDEBUG(D_CACHE, "got "LPU64" extra grant\n", body->oa.o_grant);
886 __osc_update_grant(cli, body->oa.o_grant);
890 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
891 obd_count keylen, void *key, obd_count vallen,
892 void *val, struct ptlrpc_request_set *set);
894 static int osc_shrink_grant_interpret(const struct lu_env *env,
895 struct ptlrpc_request *req,
898 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
899 struct obdo *oa = ((struct osc_grant_args *)aa)->aa_oa;
900 struct ost_body *body;
903 __osc_update_grant(cli, oa->o_grant);
907 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
909 osc_update_grant(cli, body);
915 static void osc_shrink_grant_local(struct client_obd *cli, struct obdo *oa)
917 client_obd_list_lock(&cli->cl_loi_list_lock);
918 oa->o_grant = cli->cl_avail_grant / 4;
919 cli->cl_avail_grant -= oa->o_grant;
920 client_obd_list_unlock(&cli->cl_loi_list_lock);
921 if (!(oa->o_valid & OBD_MD_FLFLAGS)) {
922 oa->o_valid |= OBD_MD_FLFLAGS;
925 oa->o_flags |= OBD_FL_SHRINK_GRANT;
926 osc_update_next_shrink(cli);
929 /* Shrink the current grant, either from some large amount to enough for a
930 * full set of in-flight RPCs, or if we have already shrunk to that limit
931 * then to enough for a single RPC. This avoids keeping more grant than
932 * needed, and avoids shrinking the grant piecemeal. */
933 static int osc_shrink_grant(struct client_obd *cli)
935 long target = (cli->cl_max_rpcs_in_flight + 1) *
936 cli->cl_max_pages_per_rpc;
938 client_obd_list_lock(&cli->cl_loi_list_lock);
939 if (cli->cl_avail_grant <= target)
940 target = cli->cl_max_pages_per_rpc;
941 client_obd_list_unlock(&cli->cl_loi_list_lock);
943 return osc_shrink_grant_to_target(cli, target);
946 int osc_shrink_grant_to_target(struct client_obd *cli, long target)
949 struct ost_body *body;
952 client_obd_list_lock(&cli->cl_loi_list_lock);
953 /* Don't shrink if we are already above or below the desired limit
954 * We don't want to shrink below a single RPC, as that will negatively
955 * impact block allocation and long-term performance. */
956 if (target < cli->cl_max_pages_per_rpc)
957 target = cli->cl_max_pages_per_rpc;
959 if (target >= cli->cl_avail_grant) {
960 client_obd_list_unlock(&cli->cl_loi_list_lock);
963 client_obd_list_unlock(&cli->cl_loi_list_lock);
969 osc_announce_cached(cli, &body->oa, 0);
971 client_obd_list_lock(&cli->cl_loi_list_lock);
972 body->oa.o_grant = cli->cl_avail_grant - target;
973 cli->cl_avail_grant = target;
974 client_obd_list_unlock(&cli->cl_loi_list_lock);
975 if (!(body->oa.o_valid & OBD_MD_FLFLAGS)) {
976 body->oa.o_valid |= OBD_MD_FLFLAGS;
977 body->oa.o_flags = 0;
979 body->oa.o_flags |= OBD_FL_SHRINK_GRANT;
980 osc_update_next_shrink(cli);
982 rc = osc_set_info_async(NULL, cli->cl_import->imp_obd->obd_self_export,
983 sizeof(KEY_GRANT_SHRINK), KEY_GRANT_SHRINK,
984 sizeof(*body), body, NULL);
986 __osc_update_grant(cli, body->oa.o_grant);
991 #define GRANT_SHRINK_LIMIT PTLRPC_MAX_BRW_SIZE
992 static int osc_should_shrink_grant(struct client_obd *client)
994 cfs_time_t time = cfs_time_current();
995 cfs_time_t next_shrink = client->cl_next_shrink_grant;
997 if ((client->cl_import->imp_connect_data.ocd_connect_flags &
998 OBD_CONNECT_GRANT_SHRINK) == 0)
1001 if (cfs_time_aftereq(time, next_shrink - 5 * CFS_TICK)) {
1002 if (client->cl_import->imp_state == LUSTRE_IMP_FULL &&
1003 client->cl_avail_grant > GRANT_SHRINK_LIMIT)
1006 osc_update_next_shrink(client);
1011 static int osc_grant_shrink_grant_cb(struct timeout_item *item, void *data)
1013 struct client_obd *client;
1015 cfs_list_for_each_entry(client, &item->ti_obd_list,
1016 cl_grant_shrink_list) {
1017 if (osc_should_shrink_grant(client))
1018 osc_shrink_grant(client);
1023 static int osc_add_shrink_grant(struct client_obd *client)
1027 rc = ptlrpc_add_timeout_client(client->cl_grant_shrink_interval,
1029 osc_grant_shrink_grant_cb, NULL,
1030 &client->cl_grant_shrink_list);
1032 CERROR("add grant client %s error %d\n",
1033 client->cl_import->imp_obd->obd_name, rc);
1036 CDEBUG(D_CACHE, "add grant client %s \n",
1037 client->cl_import->imp_obd->obd_name);
1038 osc_update_next_shrink(client);
1042 static int osc_del_shrink_grant(struct client_obd *client)
1044 return ptlrpc_del_timeout_client(&client->cl_grant_shrink_list,
1048 static void osc_init_grant(struct client_obd *cli, struct obd_connect_data *ocd)
1051 * ocd_grant is the total grant amount we're expect to hold: if we've
1052 * been evicted, it's the new avail_grant amount, cl_dirty will drop
1053 * to 0 as inflight RPCs fail out; otherwise, it's avail_grant + dirty.
1055 * race is tolerable here: if we're evicted, but imp_state already
1056 * left EVICTED state, then cl_dirty must be 0 already.
1058 client_obd_list_lock(&cli->cl_loi_list_lock);
1059 if (cli->cl_import->imp_state == LUSTRE_IMP_EVICTED)
1060 cli->cl_avail_grant = ocd->ocd_grant;
1062 cli->cl_avail_grant = ocd->ocd_grant - cli->cl_dirty;
1064 if (cli->cl_avail_grant < 0) {
1065 CWARN("%s: available grant < 0, the OSS is probably not running"
1066 " with patch from bug20278 (%ld) \n",
1067 cli->cl_import->imp_obd->obd_name, cli->cl_avail_grant);
1068 /* workaround for 1.6 servers which do not have
1069 * the patch from bug20278 */
1070 cli->cl_avail_grant = ocd->ocd_grant;
1073 /* determine the appropriate chunk size used by osc_extent. */
1074 cli->cl_chunkbits = max_t(int, CFS_PAGE_SHIFT, ocd->ocd_blocksize);
1075 client_obd_list_unlock(&cli->cl_loi_list_lock);
1077 CDEBUG(D_CACHE, "%s, setting cl_avail_grant: %ld cl_lost_grant: %ld."
1078 "chunk bits: %d.\n", cli->cl_import->imp_obd->obd_name,
1079 cli->cl_avail_grant, cli->cl_lost_grant, cli->cl_chunkbits);
1081 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT_SHRINK &&
1082 cfs_list_empty(&cli->cl_grant_shrink_list))
1083 osc_add_shrink_grant(cli);
1086 /* We assume that the reason this OSC got a short read is because it read
1087 * beyond the end of a stripe file; i.e. lustre is reading a sparse file
1088 * via the LOV, and it _knows_ it's reading inside the file, it's just that
1089 * this stripe never got written at or beyond this stripe offset yet. */
1090 static void handle_short_read(int nob_read, obd_count page_count,
1091 struct brw_page **pga)
1096 /* skip bytes read OK */
1097 while (nob_read > 0) {
1098 LASSERT (page_count > 0);
1100 if (pga[i]->count > nob_read) {
1101 /* EOF inside this page */
1102 ptr = cfs_kmap(pga[i]->pg) +
1103 (pga[i]->off & ~CFS_PAGE_MASK);
1104 memset(ptr + nob_read, 0, pga[i]->count - nob_read);
1105 cfs_kunmap(pga[i]->pg);
1111 nob_read -= pga[i]->count;
1116 /* zero remaining pages */
1117 while (page_count-- > 0) {
1118 ptr = cfs_kmap(pga[i]->pg) + (pga[i]->off & ~CFS_PAGE_MASK);
1119 memset(ptr, 0, pga[i]->count);
1120 cfs_kunmap(pga[i]->pg);
1125 static int check_write_rcs(struct ptlrpc_request *req,
1126 int requested_nob, int niocount,
1127 obd_count page_count, struct brw_page **pga)
1132 remote_rcs = req_capsule_server_sized_get(&req->rq_pill, &RMF_RCS,
1133 sizeof(*remote_rcs) *
1135 if (remote_rcs == NULL) {
1136 CDEBUG(D_INFO, "Missing/short RC vector on BRW_WRITE reply\n");
1140 /* return error if any niobuf was in error */
1141 for (i = 0; i < niocount; i++) {
1142 if ((int)remote_rcs[i] < 0)
1143 return(remote_rcs[i]);
1145 if (remote_rcs[i] != 0) {
1146 CDEBUG(D_INFO, "rc[%d] invalid (%d) req %p\n",
1147 i, remote_rcs[i], req);
1152 if (req->rq_bulk->bd_nob_transferred != requested_nob) {
1153 CERROR("Unexpected # bytes transferred: %d (requested %d)\n",
1154 req->rq_bulk->bd_nob_transferred, requested_nob);
1161 static inline int can_merge_pages(struct brw_page *p1, struct brw_page *p2)
1163 if (p1->flag != p2->flag) {
1164 unsigned mask = ~(OBD_BRW_FROM_GRANT| OBD_BRW_NOCACHE|
1165 OBD_BRW_SYNC|OBD_BRW_ASYNC|OBD_BRW_NOQUOTA);
1167 /* warn if we try to combine flags that we don't know to be
1168 * safe to combine */
1169 if (unlikely((p1->flag & mask) != (p2->flag & mask))) {
1170 CWARN("Saw flags 0x%x and 0x%x in the same brw, please "
1171 "report this at http://bugs.whamcloud.com/\n",
1172 p1->flag, p2->flag);
1177 return (p1->off + p1->count == p2->off);
1180 static obd_count osc_checksum_bulk(int nob, obd_count pg_count,
1181 struct brw_page **pga, int opc,
1182 cksum_type_t cksum_type)
1186 struct cfs_crypto_hash_desc *hdesc;
1187 unsigned int bufsize;
1189 unsigned char cfs_alg = cksum_obd2cfs(cksum_type);
1191 LASSERT(pg_count > 0);
1193 hdesc = cfs_crypto_hash_init(cfs_alg, NULL, 0);
1194 if (IS_ERR(hdesc)) {
1195 CERROR("Unable to initialize checksum hash %s\n",
1196 cfs_crypto_hash_name(cfs_alg));
1197 return PTR_ERR(hdesc);
1200 while (nob > 0 && pg_count > 0) {
1201 int count = pga[i]->count > nob ? nob : pga[i]->count;
1203 /* corrupt the data before we compute the checksum, to
1204 * simulate an OST->client data error */
1205 if (i == 0 && opc == OST_READ &&
1206 OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_RECEIVE)) {
1207 unsigned char *ptr = cfs_kmap(pga[i]->pg);
1208 int off = pga[i]->off & ~CFS_PAGE_MASK;
1209 memcpy(ptr + off, "bad1", min(4, nob));
1210 cfs_kunmap(pga[i]->pg);
1212 cfs_crypto_hash_update_page(hdesc, pga[i]->pg,
1213 pga[i]->off & ~CFS_PAGE_MASK,
1215 LL_CDEBUG_PAGE(D_PAGE, pga[i]->pg, "off %d checksum %x\n",
1216 (int)(pga[i]->off & ~CFS_PAGE_MASK), cksum);
1218 nob -= pga[i]->count;
1224 err = cfs_crypto_hash_final(hdesc, (unsigned char *)&cksum, &bufsize);
1227 cfs_crypto_hash_final(hdesc, NULL, NULL);
1229 /* For sending we only compute the wrong checksum instead
1230 * of corrupting the data so it is still correct on a redo */
1231 if (opc == OST_WRITE && OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_SEND))
1237 static int osc_brw_prep_request(int cmd, struct client_obd *cli,struct obdo *oa,
1238 struct lov_stripe_md *lsm, obd_count page_count,
1239 struct brw_page **pga,
1240 struct ptlrpc_request **reqp,
1241 struct obd_capa *ocapa, int reserve,
1244 struct ptlrpc_request *req;
1245 struct ptlrpc_bulk_desc *desc;
1246 struct ost_body *body;
1247 struct obd_ioobj *ioobj;
1248 struct niobuf_remote *niobuf;
1249 int niocount, i, requested_nob, opc, rc;
1250 struct osc_brw_async_args *aa;
1251 struct req_capsule *pill;
1252 struct brw_page *pg_prev;
1255 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ))
1256 RETURN(-ENOMEM); /* Recoverable */
1257 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ2))
1258 RETURN(-EINVAL); /* Fatal */
1260 if ((cmd & OBD_BRW_WRITE) != 0) {
1262 req = ptlrpc_request_alloc_pool(cli->cl_import,
1263 cli->cl_import->imp_rq_pool,
1264 &RQF_OST_BRW_WRITE);
1267 req = ptlrpc_request_alloc(cli->cl_import, &RQF_OST_BRW_READ);
1272 for (niocount = i = 1; i < page_count; i++) {
1273 if (!can_merge_pages(pga[i - 1], pga[i]))
1277 pill = &req->rq_pill;
1278 req_capsule_set_size(pill, &RMF_OBD_IOOBJ, RCL_CLIENT,
1280 req_capsule_set_size(pill, &RMF_NIOBUF_REMOTE, RCL_CLIENT,
1281 niocount * sizeof(*niobuf));
1282 osc_set_capa_size(req, &RMF_CAPA1, ocapa);
1284 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, opc);
1286 ptlrpc_request_free(req);
1289 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
1290 ptlrpc_at_set_req_timeout(req);
1291 /* ask ptlrpc not to resend on EINPROGRESS since BRWs have their own
1293 req->rq_no_retry_einprogress = 1;
1295 if (opc == OST_WRITE)
1296 desc = ptlrpc_prep_bulk_imp(req, page_count,
1297 BULK_GET_SOURCE, OST_BULK_PORTAL);
1299 desc = ptlrpc_prep_bulk_imp(req, page_count,
1300 BULK_PUT_SINK, OST_BULK_PORTAL);
1303 GOTO(out, rc = -ENOMEM);
1304 /* NB request now owns desc and will free it when it gets freed */
1306 body = req_capsule_client_get(pill, &RMF_OST_BODY);
1307 ioobj = req_capsule_client_get(pill, &RMF_OBD_IOOBJ);
1308 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1309 LASSERT(body != NULL && ioobj != NULL && niobuf != NULL);
1311 lustre_set_wire_obdo(&body->oa, oa);
1313 obdo_to_ioobj(oa, ioobj);
1314 ioobj->ioo_bufcnt = niocount;
1315 osc_pack_capa(req, body, ocapa);
1316 LASSERT (page_count > 0);
1318 for (requested_nob = i = 0; i < page_count; i++, niobuf++) {
1319 struct brw_page *pg = pga[i];
1320 int poff = pg->off & ~CFS_PAGE_MASK;
1322 LASSERT(pg->count > 0);
1323 /* make sure there is no gap in the middle of page array */
1324 LASSERTF(page_count == 1 ||
1325 (ergo(i == 0, poff + pg->count == CFS_PAGE_SIZE) &&
1326 ergo(i > 0 && i < page_count - 1,
1327 poff == 0 && pg->count == CFS_PAGE_SIZE) &&
1328 ergo(i == page_count - 1, poff == 0)),
1329 "i: %d/%d pg: %p off: "LPU64", count: %u\n",
1330 i, page_count, pg, pg->off, pg->count);
1332 LASSERTF(i == 0 || pg->off > pg_prev->off,
1333 "i %d p_c %u pg %p [pri %lu ind %lu] off "LPU64
1334 " prev_pg %p [pri %lu ind %lu] off "LPU64"\n",
1336 pg->pg, page_private(pg->pg), pg->pg->index, pg->off,
1337 pg_prev->pg, page_private(pg_prev->pg),
1338 pg_prev->pg->index, pg_prev->off);
1340 LASSERTF(i == 0 || pg->off > pg_prev->off,
1341 "i %d p_c %u\n", i, page_count);
1343 LASSERT((pga[0]->flag & OBD_BRW_SRVLOCK) ==
1344 (pg->flag & OBD_BRW_SRVLOCK));
1346 ptlrpc_prep_bulk_page_pin(desc, pg->pg, poff, pg->count);
1347 requested_nob += pg->count;
1349 if (i > 0 && can_merge_pages(pg_prev, pg)) {
1351 niobuf->len += pg->count;
1353 niobuf->offset = pg->off;
1354 niobuf->len = pg->count;
1355 niobuf->flags = pg->flag;
1360 LASSERTF((void *)(niobuf - niocount) ==
1361 req_capsule_client_get(&req->rq_pill, &RMF_NIOBUF_REMOTE),
1362 "want %p - real %p\n", req_capsule_client_get(&req->rq_pill,
1363 &RMF_NIOBUF_REMOTE), (void *)(niobuf - niocount));
1365 osc_announce_cached(cli, &body->oa, opc == OST_WRITE ? requested_nob:0);
1367 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1368 body->oa.o_valid |= OBD_MD_FLFLAGS;
1369 body->oa.o_flags = 0;
1371 body->oa.o_flags |= OBD_FL_RECOV_RESEND;
1374 if (osc_should_shrink_grant(cli))
1375 osc_shrink_grant_local(cli, &body->oa);
1377 /* size[REQ_REC_OFF] still sizeof (*body) */
1378 if (opc == OST_WRITE) {
1379 if (cli->cl_checksum &&
1380 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1381 /* store cl_cksum_type in a local variable since
1382 * it can be changed via lprocfs */
1383 cksum_type_t cksum_type = cli->cl_cksum_type;
1385 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1386 oa->o_flags &= OBD_FL_LOCAL_MASK;
1387 body->oa.o_flags = 0;
1389 body->oa.o_flags |= cksum_type_pack(cksum_type);
1390 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1391 body->oa.o_cksum = osc_checksum_bulk(requested_nob,
1395 CDEBUG(D_PAGE, "checksum at write origin: %x\n",
1397 /* save this in 'oa', too, for later checking */
1398 oa->o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1399 oa->o_flags |= cksum_type_pack(cksum_type);
1401 /* clear out the checksum flag, in case this is a
1402 * resend but cl_checksum is no longer set. b=11238 */
1403 oa->o_valid &= ~OBD_MD_FLCKSUM;
1405 oa->o_cksum = body->oa.o_cksum;
1406 /* 1 RC per niobuf */
1407 req_capsule_set_size(pill, &RMF_RCS, RCL_SERVER,
1408 sizeof(__u32) * niocount);
1410 if (cli->cl_checksum &&
1411 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1412 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0)
1413 body->oa.o_flags = 0;
1414 body->oa.o_flags |= cksum_type_pack(cli->cl_cksum_type);
1415 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1418 ptlrpc_request_set_replen(req);
1420 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1421 aa = ptlrpc_req_async_args(req);
1423 aa->aa_requested_nob = requested_nob;
1424 aa->aa_nio_count = niocount;
1425 aa->aa_page_count = page_count;
1429 CFS_INIT_LIST_HEAD(&aa->aa_oaps);
1430 if (ocapa && reserve)
1431 aa->aa_ocapa = capa_get(ocapa);
1437 ptlrpc_req_finished(req);
1441 static int check_write_checksum(struct obdo *oa, const lnet_process_id_t *peer,
1442 __u32 client_cksum, __u32 server_cksum, int nob,
1443 obd_count page_count, struct brw_page **pga,
1444 cksum_type_t client_cksum_type)
1448 cksum_type_t cksum_type;
1450 if (server_cksum == client_cksum) {
1451 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1455 cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
1457 new_cksum = osc_checksum_bulk(nob, page_count, pga, OST_WRITE,
1460 if (cksum_type != client_cksum_type)
1461 msg = "the server did not use the checksum type specified in "
1462 "the original request - likely a protocol problem";
1463 else if (new_cksum == server_cksum)
1464 msg = "changed on the client after we checksummed it - "
1465 "likely false positive due to mmap IO (bug 11742)";
1466 else if (new_cksum == client_cksum)
1467 msg = "changed in transit before arrival at OST";
1469 msg = "changed in transit AND doesn't match the original - "
1470 "likely false positive due to mmap IO (bug 11742)";
1472 LCONSOLE_ERROR_MSG(0x132, "BAD WRITE CHECKSUM: %s: from %s inode "DFID
1473 " object "LPU64"/"LPU64" extent ["LPU64"-"LPU64"]\n",
1474 msg, libcfs_nid2str(peer->nid),
1475 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
1476 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
1477 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
1479 oa->o_valid & OBD_MD_FLGROUP ? oa->o_seq : (__u64)0,
1481 pga[page_count-1]->off + pga[page_count-1]->count - 1);
1482 CERROR("original client csum %x (type %x), server csum %x (type %x), "
1483 "client csum now %x\n", client_cksum, client_cksum_type,
1484 server_cksum, cksum_type, new_cksum);
1488 /* Note rc enters this function as number of bytes transferred */
1489 static int osc_brw_fini_request(struct ptlrpc_request *req, int rc)
1491 struct osc_brw_async_args *aa = (void *)&req->rq_async_args;
1492 const lnet_process_id_t *peer =
1493 &req->rq_import->imp_connection->c_peer;
1494 struct client_obd *cli = aa->aa_cli;
1495 struct ost_body *body;
1496 __u32 client_cksum = 0;
1499 if (rc < 0 && rc != -EDQUOT) {
1500 DEBUG_REQ(D_INFO, req, "Failed request with rc = %d\n", rc);
1504 LASSERTF(req->rq_repmsg != NULL, "rc = %d\n", rc);
1505 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
1507 DEBUG_REQ(D_INFO, req, "Can't unpack body\n");
1511 /* set/clear over quota flag for a uid/gid */
1512 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE &&
1513 body->oa.o_valid & (OBD_MD_FLUSRQUOTA | OBD_MD_FLGRPQUOTA)) {
1514 unsigned int qid[MAXQUOTAS] = { body->oa.o_uid, body->oa.o_gid };
1516 CDEBUG(D_QUOTA, "setdq for [%u %u] with valid "LPX64", flags %x\n",
1517 body->oa.o_uid, body->oa.o_gid, body->oa.o_valid,
1519 osc_quota_setdq(cli, qid, body->oa.o_valid, body->oa.o_flags);
1522 osc_update_grant(cli, body);
1527 if (aa->aa_oa->o_valid & OBD_MD_FLCKSUM)
1528 client_cksum = aa->aa_oa->o_cksum; /* save for later */
1530 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1532 CERROR("Unexpected +ve rc %d\n", rc);
1535 LASSERT(req->rq_bulk->bd_nob == aa->aa_requested_nob);
1537 if (sptlrpc_cli_unwrap_bulk_write(req, req->rq_bulk))
1540 if ((aa->aa_oa->o_valid & OBD_MD_FLCKSUM) && client_cksum &&
1541 check_write_checksum(&body->oa, peer, client_cksum,
1542 body->oa.o_cksum, aa->aa_requested_nob,
1543 aa->aa_page_count, aa->aa_ppga,
1544 cksum_type_unpack(aa->aa_oa->o_flags)))
1547 rc = check_write_rcs(req, aa->aa_requested_nob,aa->aa_nio_count,
1548 aa->aa_page_count, aa->aa_ppga);
1552 /* The rest of this function executes only for OST_READs */
1554 /* if unwrap_bulk failed, return -EAGAIN to retry */
1555 rc = sptlrpc_cli_unwrap_bulk_read(req, req->rq_bulk, rc);
1557 GOTO(out, rc = -EAGAIN);
1559 if (rc > aa->aa_requested_nob) {
1560 CERROR("Unexpected rc %d (%d requested)\n", rc,
1561 aa->aa_requested_nob);
1565 if (rc != req->rq_bulk->bd_nob_transferred) {
1566 CERROR ("Unexpected rc %d (%d transferred)\n",
1567 rc, req->rq_bulk->bd_nob_transferred);
1571 if (rc < aa->aa_requested_nob)
1572 handle_short_read(rc, aa->aa_page_count, aa->aa_ppga);
1574 if (body->oa.o_valid & OBD_MD_FLCKSUM) {
1575 static int cksum_counter;
1576 __u32 server_cksum = body->oa.o_cksum;
1579 cksum_type_t cksum_type;
1581 cksum_type = cksum_type_unpack(body->oa.o_valid &OBD_MD_FLFLAGS?
1582 body->oa.o_flags : 0);
1583 client_cksum = osc_checksum_bulk(rc, aa->aa_page_count,
1584 aa->aa_ppga, OST_READ,
1587 if (peer->nid == req->rq_bulk->bd_sender) {
1591 router = libcfs_nid2str(req->rq_bulk->bd_sender);
1594 if (server_cksum == ~0 && rc > 0) {
1595 CERROR("Protocol error: server %s set the 'checksum' "
1596 "bit, but didn't send a checksum. Not fatal, "
1597 "but please notify on http://bugs.whamcloud.com/\n",
1598 libcfs_nid2str(peer->nid));
1599 } else if (server_cksum != client_cksum) {
1600 LCONSOLE_ERROR_MSG(0x133, "%s: BAD READ CHECKSUM: from "
1601 "%s%s%s inode "DFID" object "
1602 LPU64"/"LPU64" extent "
1603 "["LPU64"-"LPU64"]\n",
1604 req->rq_import->imp_obd->obd_name,
1605 libcfs_nid2str(peer->nid),
1607 body->oa.o_valid & OBD_MD_FLFID ?
1608 body->oa.o_parent_seq : (__u64)0,
1609 body->oa.o_valid & OBD_MD_FLFID ?
1610 body->oa.o_parent_oid : 0,
1611 body->oa.o_valid & OBD_MD_FLFID ?
1612 body->oa.o_parent_ver : 0,
1614 body->oa.o_valid & OBD_MD_FLGROUP ?
1615 body->oa.o_seq : (__u64)0,
1616 aa->aa_ppga[0]->off,
1617 aa->aa_ppga[aa->aa_page_count-1]->off +
1618 aa->aa_ppga[aa->aa_page_count-1]->count -
1620 CERROR("client %x, server %x, cksum_type %x\n",
1621 client_cksum, server_cksum, cksum_type);
1623 aa->aa_oa->o_cksum = client_cksum;
1627 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1630 } else if (unlikely(client_cksum)) {
1631 static int cksum_missed;
1634 if ((cksum_missed & (-cksum_missed)) == cksum_missed)
1635 CERROR("Checksum %u requested from %s but not sent\n",
1636 cksum_missed, libcfs_nid2str(peer->nid));
1642 lustre_get_wire_obdo(aa->aa_oa, &body->oa);
1647 static int osc_brw_internal(int cmd, struct obd_export *exp, struct obdo *oa,
1648 struct lov_stripe_md *lsm,
1649 obd_count page_count, struct brw_page **pga,
1650 struct obd_capa *ocapa)
1652 struct ptlrpc_request *req;
1655 int generation, resends = 0;
1656 struct l_wait_info lwi;
1660 cfs_waitq_init(&waitq);
1661 generation = exp->exp_obd->u.cli.cl_import->imp_generation;
1664 rc = osc_brw_prep_request(cmd, &exp->exp_obd->u.cli, oa, lsm,
1665 page_count, pga, &req, ocapa, 0, resends);
1670 req->rq_generation_set = 1;
1671 req->rq_import_generation = generation;
1672 req->rq_sent = cfs_time_current_sec() + resends;
1675 rc = ptlrpc_queue_wait(req);
1677 if (rc == -ETIMEDOUT && req->rq_resend) {
1678 DEBUG_REQ(D_HA, req, "BULK TIMEOUT");
1679 ptlrpc_req_finished(req);
1683 rc = osc_brw_fini_request(req, rc);
1685 ptlrpc_req_finished(req);
1686 /* When server return -EINPROGRESS, client should always retry
1687 * regardless of the number of times the bulk was resent already.*/
1688 if (osc_recoverable_error(rc)) {
1690 if (rc != -EINPROGRESS &&
1691 !client_should_resend(resends, &exp->exp_obd->u.cli)) {
1692 CERROR("%s: too many resend retries for object: "
1693 ""LPU64":"LPU64", rc = %d.\n",
1694 exp->exp_obd->obd_name, oa->o_id, oa->o_seq, rc);
1698 exp->exp_obd->u.cli.cl_import->imp_generation) {
1699 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1700 ""LPU64":"LPU64", rc = %d.\n",
1701 exp->exp_obd->obd_name, oa->o_id, oa->o_seq, rc);
1705 lwi = LWI_TIMEOUT_INTR(cfs_time_seconds(resends), NULL, NULL,
1707 l_wait_event(waitq, 0, &lwi);
1712 if (rc == -EAGAIN || rc == -EINPROGRESS)
1717 static int osc_brw_redo_request(struct ptlrpc_request *request,
1718 struct osc_brw_async_args *aa, int rc)
1720 struct ptlrpc_request *new_req;
1721 struct osc_brw_async_args *new_aa;
1722 struct osc_async_page *oap;
1725 DEBUG_REQ(rc == -EINPROGRESS ? D_RPCTRACE : D_ERROR, request,
1726 "redo for recoverable error %d", rc);
1728 rc = osc_brw_prep_request(lustre_msg_get_opc(request->rq_reqmsg) ==
1729 OST_WRITE ? OBD_BRW_WRITE :OBD_BRW_READ,
1730 aa->aa_cli, aa->aa_oa,
1731 NULL /* lsm unused by osc currently */,
1732 aa->aa_page_count, aa->aa_ppga,
1733 &new_req, aa->aa_ocapa, 0, 1);
1737 cfs_list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
1738 if (oap->oap_request != NULL) {
1739 LASSERTF(request == oap->oap_request,
1740 "request %p != oap_request %p\n",
1741 request, oap->oap_request);
1742 if (oap->oap_interrupted) {
1743 ptlrpc_req_finished(new_req);
1748 /* New request takes over pga and oaps from old request.
1749 * Note that copying a list_head doesn't work, need to move it... */
1751 new_req->rq_interpret_reply = request->rq_interpret_reply;
1752 new_req->rq_async_args = request->rq_async_args;
1753 /* cap resend delay to the current request timeout, this is similar to
1754 * what ptlrpc does (see after_reply()) */
1755 if (aa->aa_resends > new_req->rq_timeout)
1756 new_req->rq_sent = cfs_time_current_sec() + new_req->rq_timeout;
1758 new_req->rq_sent = cfs_time_current_sec() + aa->aa_resends;
1759 new_req->rq_generation_set = 1;
1760 new_req->rq_import_generation = request->rq_import_generation;
1762 new_aa = ptlrpc_req_async_args(new_req);
1764 CFS_INIT_LIST_HEAD(&new_aa->aa_oaps);
1765 cfs_list_splice_init(&aa->aa_oaps, &new_aa->aa_oaps);
1766 CFS_INIT_LIST_HEAD(&new_aa->aa_exts);
1767 cfs_list_splice_init(&aa->aa_exts, &new_aa->aa_exts);
1768 new_aa->aa_resends = aa->aa_resends;
1770 cfs_list_for_each_entry(oap, &new_aa->aa_oaps, oap_rpc_item) {
1771 if (oap->oap_request) {
1772 ptlrpc_req_finished(oap->oap_request);
1773 oap->oap_request = ptlrpc_request_addref(new_req);
1777 new_aa->aa_ocapa = aa->aa_ocapa;
1778 aa->aa_ocapa = NULL;
1780 /* XXX: This code will run into problem if we're going to support
1781 * to add a series of BRW RPCs into a self-defined ptlrpc_request_set
1782 * and wait for all of them to be finished. We should inherit request
1783 * set from old request. */
1784 ptlrpcd_add_req(new_req, PDL_POLICY_SAME, -1);
1786 DEBUG_REQ(D_INFO, new_req, "new request");
1791 * ugh, we want disk allocation on the target to happen in offset order. we'll
1792 * follow sedgewicks advice and stick to the dead simple shellsort -- it'll do
1793 * fine for our small page arrays and doesn't require allocation. its an
1794 * insertion sort that swaps elements that are strides apart, shrinking the
1795 * stride down until its '1' and the array is sorted.
1797 static void sort_brw_pages(struct brw_page **array, int num)
1800 struct brw_page *tmp;
1804 for (stride = 1; stride < num ; stride = (stride * 3) + 1)
1809 for (i = stride ; i < num ; i++) {
1812 while (j >= stride && array[j - stride]->off > tmp->off) {
1813 array[j] = array[j - stride];
1818 } while (stride > 1);
1821 static obd_count max_unfragmented_pages(struct brw_page **pg, obd_count pages)
1827 LASSERT (pages > 0);
1828 offset = pg[i]->off & ~CFS_PAGE_MASK;
1832 if (pages == 0) /* that's all */
1835 if (offset + pg[i]->count < CFS_PAGE_SIZE)
1836 return count; /* doesn't end on page boundary */
1839 offset = pg[i]->off & ~CFS_PAGE_MASK;
1840 if (offset != 0) /* doesn't start on page boundary */
1847 static struct brw_page **osc_build_ppga(struct brw_page *pga, obd_count count)
1849 struct brw_page **ppga;
1852 OBD_ALLOC(ppga, sizeof(*ppga) * count);
1856 for (i = 0; i < count; i++)
1861 static void osc_release_ppga(struct brw_page **ppga, obd_count count)
1863 LASSERT(ppga != NULL);
1864 OBD_FREE(ppga, sizeof(*ppga) * count);
1867 static int osc_brw(int cmd, struct obd_export *exp, struct obd_info *oinfo,
1868 obd_count page_count, struct brw_page *pga,
1869 struct obd_trans_info *oti)
1871 struct obdo *saved_oa = NULL;
1872 struct brw_page **ppga, **orig;
1873 struct obd_import *imp = class_exp2cliimp(exp);
1874 struct client_obd *cli;
1875 int rc, page_count_orig;
1878 LASSERT((imp != NULL) && (imp->imp_obd != NULL));
1879 cli = &imp->imp_obd->u.cli;
1881 if (cmd & OBD_BRW_CHECK) {
1882 /* The caller just wants to know if there's a chance that this
1883 * I/O can succeed */
1885 if (imp->imp_invalid)
1890 /* test_brw with a failed create can trip this, maybe others. */
1891 LASSERT(cli->cl_max_pages_per_rpc);
1895 orig = ppga = osc_build_ppga(pga, page_count);
1898 page_count_orig = page_count;
1900 sort_brw_pages(ppga, page_count);
1901 while (page_count) {
1902 obd_count pages_per_brw;
1904 if (page_count > cli->cl_max_pages_per_rpc)
1905 pages_per_brw = cli->cl_max_pages_per_rpc;
1907 pages_per_brw = page_count;
1909 pages_per_brw = max_unfragmented_pages(ppga, pages_per_brw);
1911 if (saved_oa != NULL) {
1912 /* restore previously saved oa */
1913 *oinfo->oi_oa = *saved_oa;
1914 } else if (page_count > pages_per_brw) {
1915 /* save a copy of oa (brw will clobber it) */
1916 OBDO_ALLOC(saved_oa);
1917 if (saved_oa == NULL)
1918 GOTO(out, rc = -ENOMEM);
1919 *saved_oa = *oinfo->oi_oa;
1922 rc = osc_brw_internal(cmd, exp, oinfo->oi_oa, oinfo->oi_md,
1923 pages_per_brw, ppga, oinfo->oi_capa);
1928 page_count -= pages_per_brw;
1929 ppga += pages_per_brw;
1933 osc_release_ppga(orig, page_count_orig);
1935 if (saved_oa != NULL)
1936 OBDO_FREE(saved_oa);
1941 static int brw_interpret(const struct lu_env *env,
1942 struct ptlrpc_request *req, void *data, int rc)
1944 struct osc_brw_async_args *aa = data;
1945 struct osc_extent *ext;
1946 struct osc_extent *tmp;
1947 struct cl_object *obj = NULL;
1948 struct client_obd *cli = aa->aa_cli;
1951 rc = osc_brw_fini_request(req, rc);
1952 CDEBUG(D_INODE, "request %p aa %p rc %d\n", req, aa, rc);
1953 /* When server return -EINPROGRESS, client should always retry
1954 * regardless of the number of times the bulk was resent already. */
1955 if (osc_recoverable_error(rc)) {
1956 if (req->rq_import_generation !=
1957 req->rq_import->imp_generation) {
1958 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1959 ""LPU64":"LPU64", rc = %d.\n",
1960 req->rq_import->imp_obd->obd_name,
1961 aa->aa_oa->o_id, aa->aa_oa->o_seq, rc);
1962 } else if (rc == -EINPROGRESS ||
1963 client_should_resend(aa->aa_resends, aa->aa_cli)) {
1964 rc = osc_brw_redo_request(req, aa, rc);
1966 CERROR("%s: too many resent retries for object: "
1967 ""LPU64":"LPU64", rc = %d.\n",
1968 req->rq_import->imp_obd->obd_name,
1969 aa->aa_oa->o_id, aa->aa_oa->o_seq, rc);
1974 else if (rc == -EAGAIN || rc == -EINPROGRESS)
1979 capa_put(aa->aa_ocapa);
1980 aa->aa_ocapa = NULL;
1983 cfs_list_for_each_entry_safe(ext, tmp, &aa->aa_exts, oe_link) {
1984 if (obj == NULL && rc == 0) {
1985 obj = osc2cl(ext->oe_obj);
1989 cfs_list_del_init(&ext->oe_link);
1990 osc_extent_finish(env, ext, 1, rc);
1992 LASSERT(cfs_list_empty(&aa->aa_exts));
1993 LASSERT(cfs_list_empty(&aa->aa_oaps));
1996 struct obdo *oa = aa->aa_oa;
1997 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
1998 unsigned long valid = 0;
2001 if (oa->o_valid & OBD_MD_FLBLOCKS) {
2002 attr->cat_blocks = oa->o_blocks;
2003 valid |= CAT_BLOCKS;
2005 if (oa->o_valid & OBD_MD_FLMTIME) {
2006 attr->cat_mtime = oa->o_mtime;
2009 if (oa->o_valid & OBD_MD_FLATIME) {
2010 attr->cat_atime = oa->o_atime;
2013 if (oa->o_valid & OBD_MD_FLCTIME) {
2014 attr->cat_ctime = oa->o_ctime;
2018 cl_object_attr_lock(obj);
2019 cl_object_attr_set(env, obj, attr, valid);
2020 cl_object_attr_unlock(obj);
2022 cl_object_put(env, obj);
2024 OBDO_FREE(aa->aa_oa);
2026 cl_req_completion(env, aa->aa_clerq, rc < 0 ? rc :
2027 req->rq_bulk->bd_nob_transferred);
2028 osc_release_ppga(aa->aa_ppga, aa->aa_page_count);
2029 ptlrpc_lprocfs_brw(req, req->rq_bulk->bd_nob_transferred);
2031 client_obd_list_lock(&cli->cl_loi_list_lock);
2032 /* We need to decrement before osc_ap_completion->osc_wake_cache_waiters
2033 * is called so we know whether to go to sync BRWs or wait for more
2034 * RPCs to complete */
2035 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE)
2036 cli->cl_w_in_flight--;
2038 cli->cl_r_in_flight--;
2039 osc_wake_cache_waiters(cli);
2040 client_obd_list_unlock(&cli->cl_loi_list_lock);
2042 osc_io_unplug(env, cli, NULL, PDL_POLICY_SAME);
2047 * Build an RPC by the list of extent @ext_list. The caller must ensure
2048 * that the total pages in this list are NOT over max pages per RPC.
2049 * Extents in the list must be in OES_RPC state.
2051 int osc_build_rpc(const struct lu_env *env, struct client_obd *cli,
2052 cfs_list_t *ext_list, int cmd, pdl_policy_t pol)
2054 struct ptlrpc_request *req = NULL;
2055 struct osc_extent *ext;
2056 CFS_LIST_HEAD(rpc_list);
2057 struct brw_page **pga = NULL;
2058 struct osc_brw_async_args *aa = NULL;
2059 struct obdo *oa = NULL;
2060 struct osc_async_page *oap;
2061 struct osc_async_page *tmp;
2062 struct cl_req *clerq = NULL;
2063 enum cl_req_type crt = (cmd & OBD_BRW_WRITE) ? CRT_WRITE : CRT_READ;
2064 struct ldlm_lock *lock = NULL;
2065 struct cl_req_attr crattr;
2066 obd_off starting_offset = OBD_OBJECT_EOF;
2067 obd_off ending_offset = 0;
2068 int i, rc, mpflag = 0, mem_tight = 0, page_count = 0;
2071 LASSERT(!cfs_list_empty(ext_list));
2073 /* add pages into rpc_list to build BRW rpc */
2074 cfs_list_for_each_entry(ext, ext_list, oe_link) {
2075 LASSERT(ext->oe_state == OES_RPC);
2076 mem_tight |= ext->oe_memalloc;
2077 cfs_list_for_each_entry(oap, &ext->oe_pages, oap_pending_item) {
2079 cfs_list_add_tail(&oap->oap_rpc_item, &rpc_list);
2080 if (starting_offset > oap->oap_obj_off)
2081 starting_offset = oap->oap_obj_off;
2083 LASSERT(oap->oap_page_off == 0);
2084 if (ending_offset < oap->oap_obj_off + oap->oap_count)
2085 ending_offset = oap->oap_obj_off +
2088 LASSERT(oap->oap_page_off + oap->oap_count ==
2094 mpflag = cfs_memory_pressure_get_and_set();
2096 memset(&crattr, 0, sizeof crattr);
2097 OBD_ALLOC(pga, sizeof(*pga) * page_count);
2099 GOTO(out, rc = -ENOMEM);
2103 GOTO(out, rc = -ENOMEM);
2106 cfs_list_for_each_entry(oap, &rpc_list, oap_rpc_item) {
2107 struct cl_page *page = oap2cl_page(oap);
2108 if (clerq == NULL) {
2109 clerq = cl_req_alloc(env, page, crt,
2110 1 /* only 1-object rpcs for
2113 GOTO(out, rc = PTR_ERR(clerq));
2114 lock = oap->oap_ldlm_lock;
2117 oap->oap_brw_flags |= OBD_BRW_MEMALLOC;
2118 pga[i] = &oap->oap_brw_page;
2119 pga[i]->off = oap->oap_obj_off + oap->oap_page_off;
2120 CDEBUG(0, "put page %p index %lu oap %p flg %x to pga\n",
2121 pga[i]->pg, cfs_page_index(oap->oap_page), oap, pga[i]->flag);
2123 cl_req_page_add(env, clerq, page);
2126 /* always get the data for the obdo for the rpc */
2127 LASSERT(clerq != NULL);
2129 crattr.cra_capa = NULL;
2130 memset(crattr.cra_jobid, 0, JOBSTATS_JOBID_SIZE);
2131 cl_req_attr_set(env, clerq, &crattr, ~0ULL);
2133 oa->o_handle = lock->l_remote_handle;
2134 oa->o_valid |= OBD_MD_FLHANDLE;
2137 rc = cl_req_prep(env, clerq);
2139 CERROR("cl_req_prep failed: %d\n", rc);
2143 sort_brw_pages(pga, page_count);
2144 rc = osc_brw_prep_request(cmd, cli, oa, NULL, page_count,
2145 pga, &req, crattr.cra_capa, 1, 0);
2147 CERROR("prep_req failed: %d\n", rc);
2151 req->rq_interpret_reply = brw_interpret;
2153 req->rq_memalloc = 1;
2155 /* Need to update the timestamps after the request is built in case
2156 * we race with setattr (locally or in queue at OST). If OST gets
2157 * later setattr before earlier BRW (as determined by the request xid),
2158 * the OST will not use BRW timestamps. Sadly, there is no obvious
2159 * way to do this in a single call. bug 10150 */
2160 cl_req_attr_set(env, clerq, &crattr,
2161 OBD_MD_FLMTIME|OBD_MD_FLCTIME|OBD_MD_FLATIME);
2163 lustre_msg_set_jobid(req->rq_reqmsg, crattr.cra_jobid);
2165 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2166 aa = ptlrpc_req_async_args(req);
2167 CFS_INIT_LIST_HEAD(&aa->aa_oaps);
2168 cfs_list_splice_init(&rpc_list, &aa->aa_oaps);
2169 CFS_INIT_LIST_HEAD(&aa->aa_exts);
2170 cfs_list_splice_init(ext_list, &aa->aa_exts);
2171 aa->aa_clerq = clerq;
2173 /* queued sync pages can be torn down while the pages
2174 * were between the pending list and the rpc */
2176 cfs_list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
2177 /* only one oap gets a request reference */
2180 if (oap->oap_interrupted && !req->rq_intr) {
2181 CDEBUG(D_INODE, "oap %p in req %p interrupted\n",
2183 ptlrpc_mark_interrupted(req);
2187 tmp->oap_request = ptlrpc_request_addref(req);
2189 client_obd_list_lock(&cli->cl_loi_list_lock);
2190 starting_offset >>= CFS_PAGE_SHIFT;
2191 if (cmd == OBD_BRW_READ) {
2192 cli->cl_r_in_flight++;
2193 lprocfs_oh_tally_log2(&cli->cl_read_page_hist, page_count);
2194 lprocfs_oh_tally(&cli->cl_read_rpc_hist, cli->cl_r_in_flight);
2195 lprocfs_oh_tally_log2(&cli->cl_read_offset_hist,
2196 starting_offset + 1);
2198 cli->cl_w_in_flight++;
2199 lprocfs_oh_tally_log2(&cli->cl_write_page_hist, page_count);
2200 lprocfs_oh_tally(&cli->cl_write_rpc_hist, cli->cl_w_in_flight);
2201 lprocfs_oh_tally_log2(&cli->cl_write_offset_hist,
2202 starting_offset + 1);
2204 client_obd_list_unlock(&cli->cl_loi_list_lock);
2206 DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %dr/%dw in flight",
2207 page_count, aa, cli->cl_r_in_flight,
2208 cli->cl_w_in_flight);
2210 /* XXX: Maybe the caller can check the RPC bulk descriptor to
2211 * see which CPU/NUMA node the majority of pages were allocated
2212 * on, and try to assign the async RPC to the CPU core
2213 * (PDL_POLICY_PREFERRED) to reduce cross-CPU memory traffic.
2215 * But on the other hand, we expect that multiple ptlrpcd
2216 * threads and the initial write sponsor can run in parallel,
2217 * especially when data checksum is enabled, which is CPU-bound
2218 * operation and single ptlrpcd thread cannot process in time.
2219 * So more ptlrpcd threads sharing BRW load
2220 * (with PDL_POLICY_ROUND) seems better.
2222 ptlrpcd_add_req(req, pol, -1);
2228 cfs_memory_pressure_restore(mpflag);
2230 capa_put(crattr.cra_capa);
2232 LASSERT(req == NULL);
2237 OBD_FREE(pga, sizeof(*pga) * page_count);
2238 /* this should happen rarely and is pretty bad, it makes the
2239 * pending list not follow the dirty order */
2240 while (!cfs_list_empty(ext_list)) {
2241 ext = cfs_list_entry(ext_list->next, struct osc_extent,
2243 cfs_list_del_init(&ext->oe_link);
2244 osc_extent_finish(env, ext, 0, rc);
2246 if (clerq && !IS_ERR(clerq))
2247 cl_req_completion(env, clerq, rc);
2252 static int osc_set_lock_data_with_check(struct ldlm_lock *lock,
2253 struct ldlm_enqueue_info *einfo)
2255 void *data = einfo->ei_cbdata;
2258 LASSERT(lock != NULL);
2259 LASSERT(lock->l_blocking_ast == einfo->ei_cb_bl);
2260 LASSERT(lock->l_resource->lr_type == einfo->ei_type);
2261 LASSERT(lock->l_completion_ast == einfo->ei_cb_cp);
2262 LASSERT(lock->l_glimpse_ast == einfo->ei_cb_gl);
2264 lock_res_and_lock(lock);
2265 spin_lock(&osc_ast_guard);
2267 if (lock->l_ast_data == NULL)
2268 lock->l_ast_data = data;
2269 if (lock->l_ast_data == data)
2272 spin_unlock(&osc_ast_guard);
2273 unlock_res_and_lock(lock);
2278 static int osc_set_data_with_check(struct lustre_handle *lockh,
2279 struct ldlm_enqueue_info *einfo)
2281 struct ldlm_lock *lock = ldlm_handle2lock(lockh);
2285 set = osc_set_lock_data_with_check(lock, einfo);
2286 LDLM_LOCK_PUT(lock);
2288 CERROR("lockh %p, data %p - client evicted?\n",
2289 lockh, einfo->ei_cbdata);
2293 static int osc_change_cbdata(struct obd_export *exp, struct lov_stripe_md *lsm,
2294 ldlm_iterator_t replace, void *data)
2296 struct ldlm_res_id res_id;
2297 struct obd_device *obd = class_exp2obd(exp);
2299 ostid_build_res_name(&lsm->lsm_object_oid, &res_id);
2300 ldlm_resource_iterate(obd->obd_namespace, &res_id, replace, data);
2304 /* find any ldlm lock of the inode in osc
2308 static int osc_find_cbdata(struct obd_export *exp, struct lov_stripe_md *lsm,
2309 ldlm_iterator_t replace, void *data)
2311 struct ldlm_res_id res_id;
2312 struct obd_device *obd = class_exp2obd(exp);
2315 ostid_build_res_name(&lsm->lsm_object_oid, &res_id);
2316 rc = ldlm_resource_iterate(obd->obd_namespace, &res_id, replace, data);
2317 if (rc == LDLM_ITER_STOP)
2319 if (rc == LDLM_ITER_CONTINUE)
2324 static int osc_enqueue_fini(struct ptlrpc_request *req, struct ost_lvb *lvb,
2325 obd_enqueue_update_f upcall, void *cookie,
2326 __u64 *flags, int agl, int rc)
2328 int intent = *flags & LDLM_FL_HAS_INTENT;
2332 /* The request was created before ldlm_cli_enqueue call. */
2333 if (rc == ELDLM_LOCK_ABORTED) {
2334 struct ldlm_reply *rep;
2335 rep = req_capsule_server_get(&req->rq_pill,
2338 LASSERT(rep != NULL);
2339 if (rep->lock_policy_res1)
2340 rc = rep->lock_policy_res1;
2344 if ((intent != 0 && rc == ELDLM_LOCK_ABORTED && agl == 0) ||
2346 *flags |= LDLM_FL_LVB_READY;
2347 CDEBUG(D_INODE,"got kms "LPU64" blocks "LPU64" mtime "LPU64"\n",
2348 lvb->lvb_size, lvb->lvb_blocks, lvb->lvb_mtime);
2351 /* Call the update callback. */
2352 rc = (*upcall)(cookie, rc);
2356 static int osc_enqueue_interpret(const struct lu_env *env,
2357 struct ptlrpc_request *req,
2358 struct osc_enqueue_args *aa, int rc)
2360 struct ldlm_lock *lock;
2361 struct lustre_handle handle;
2363 struct ost_lvb *lvb;
2365 __u64 *flags = aa->oa_flags;
2367 /* Make a local copy of a lock handle and a mode, because aa->oa_*
2368 * might be freed anytime after lock upcall has been called. */
2369 lustre_handle_copy(&handle, aa->oa_lockh);
2370 mode = aa->oa_ei->ei_mode;
2372 /* ldlm_cli_enqueue is holding a reference on the lock, so it must
2374 lock = ldlm_handle2lock(&handle);
2376 /* Take an additional reference so that a blocking AST that
2377 * ldlm_cli_enqueue_fini() might post for a failed lock, is guaranteed
2378 * to arrive after an upcall has been executed by
2379 * osc_enqueue_fini(). */
2380 ldlm_lock_addref(&handle, mode);
2382 /* Let CP AST to grant the lock first. */
2383 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_ENQ_RACE, 1);
2385 if (aa->oa_agl && rc == ELDLM_LOCK_ABORTED) {
2390 lvb_len = sizeof(*aa->oa_lvb);
2393 /* Complete obtaining the lock procedure. */
2394 rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, aa->oa_ei->ei_type, 1,
2395 mode, flags, lvb, lvb_len, &handle, rc);
2396 /* Complete osc stuff. */
2397 rc = osc_enqueue_fini(req, aa->oa_lvb, aa->oa_upcall, aa->oa_cookie,
2398 flags, aa->oa_agl, rc);
2400 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_CANCEL_RACE, 10);
2402 /* Release the lock for async request. */
2403 if (lustre_handle_is_used(&handle) && rc == ELDLM_OK)
2405 * Releases a reference taken by ldlm_cli_enqueue(), if it is
2406 * not already released by
2407 * ldlm_cli_enqueue_fini()->failed_lock_cleanup()
2409 ldlm_lock_decref(&handle, mode);
2411 LASSERTF(lock != NULL, "lockh %p, req %p, aa %p - client evicted?\n",
2412 aa->oa_lockh, req, aa);
2413 ldlm_lock_decref(&handle, mode);
2414 LDLM_LOCK_PUT(lock);
2418 void osc_update_enqueue(struct lustre_handle *lov_lockhp,
2419 struct lov_oinfo *loi, int flags,
2420 struct ost_lvb *lvb, __u32 mode, int rc)
2422 struct ldlm_lock *lock = ldlm_handle2lock(lov_lockhp);
2424 if (rc == ELDLM_OK) {
2427 LASSERT(lock != NULL);
2428 loi->loi_lvb = *lvb;
2429 tmp = loi->loi_lvb.lvb_size;
2430 /* Extend KMS up to the end of this lock and no further
2431 * A lock on [x,y] means a KMS of up to y + 1 bytes! */
2432 if (tmp > lock->l_policy_data.l_extent.end)
2433 tmp = lock->l_policy_data.l_extent.end + 1;
2434 if (tmp >= loi->loi_kms) {
2435 LDLM_DEBUG(lock, "lock acquired, setting rss="LPU64
2436 ", kms="LPU64, loi->loi_lvb.lvb_size, tmp);
2437 loi_kms_set(loi, tmp);
2439 LDLM_DEBUG(lock, "lock acquired, setting rss="
2440 LPU64"; leaving kms="LPU64", end="LPU64,
2441 loi->loi_lvb.lvb_size, loi->loi_kms,
2442 lock->l_policy_data.l_extent.end);
2444 ldlm_lock_allow_match(lock);
2445 } else if (rc == ELDLM_LOCK_ABORTED && (flags & LDLM_FL_HAS_INTENT)) {
2446 LASSERT(lock != NULL);
2447 loi->loi_lvb = *lvb;
2448 ldlm_lock_allow_match(lock);
2449 CDEBUG(D_INODE, "glimpsed, setting rss="LPU64"; leaving"
2450 " kms="LPU64"\n", loi->loi_lvb.lvb_size, loi->loi_kms);
2456 ldlm_lock_fail_match(lock);
2458 LDLM_LOCK_PUT(lock);
2461 EXPORT_SYMBOL(osc_update_enqueue);
2463 struct ptlrpc_request_set *PTLRPCD_SET = (void *)1;
2465 /* When enqueuing asynchronously, locks are not ordered, we can obtain a lock
2466 * from the 2nd OSC before a lock from the 1st one. This does not deadlock with
2467 * other synchronous requests, however keeping some locks and trying to obtain
2468 * others may take a considerable amount of time in a case of ost failure; and
2469 * when other sync requests do not get released lock from a client, the client
2470 * is excluded from the cluster -- such scenarious make the life difficult, so
2471 * release locks just after they are obtained. */
2472 int osc_enqueue_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2473 __u64 *flags, ldlm_policy_data_t *policy,
2474 struct ost_lvb *lvb, int kms_valid,
2475 obd_enqueue_update_f upcall, void *cookie,
2476 struct ldlm_enqueue_info *einfo,
2477 struct lustre_handle *lockh,
2478 struct ptlrpc_request_set *rqset, int async, int agl)
2480 struct obd_device *obd = exp->exp_obd;
2481 struct ptlrpc_request *req = NULL;
2482 int intent = *flags & LDLM_FL_HAS_INTENT;
2483 int match_lvb = (agl != 0 ? 0 : LDLM_FL_LVB_READY);
2488 /* Filesystem lock extents are extended to page boundaries so that
2489 * dealing with the page cache is a little smoother. */
2490 policy->l_extent.start -= policy->l_extent.start & ~CFS_PAGE_MASK;
2491 policy->l_extent.end |= ~CFS_PAGE_MASK;
2494 * kms is not valid when either object is completely fresh (so that no
2495 * locks are cached), or object was evicted. In the latter case cached
2496 * lock cannot be used, because it would prime inode state with
2497 * potentially stale LVB.
2502 /* Next, search for already existing extent locks that will cover us */
2503 /* If we're trying to read, we also search for an existing PW lock. The
2504 * VFS and page cache already protect us locally, so lots of readers/
2505 * writers can share a single PW lock.
2507 * There are problems with conversion deadlocks, so instead of
2508 * converting a read lock to a write lock, we'll just enqueue a new
2511 * At some point we should cancel the read lock instead of making them
2512 * send us a blocking callback, but there are problems with canceling
2513 * locks out from other users right now, too. */
2514 mode = einfo->ei_mode;
2515 if (einfo->ei_mode == LCK_PR)
2517 mode = ldlm_lock_match(obd->obd_namespace, *flags | match_lvb, res_id,
2518 einfo->ei_type, policy, mode, lockh, 0);
2520 struct ldlm_lock *matched = ldlm_handle2lock(lockh);
2522 if ((agl != 0) && !(matched->l_flags & LDLM_FL_LVB_READY)) {
2523 /* For AGL, if enqueue RPC is sent but the lock is not
2524 * granted, then skip to process this strpe.
2525 * Return -ECANCELED to tell the caller. */
2526 ldlm_lock_decref(lockh, mode);
2527 LDLM_LOCK_PUT(matched);
2529 } else if (osc_set_lock_data_with_check(matched, einfo)) {
2530 *flags |= LDLM_FL_LVB_READY;
2531 /* addref the lock only if not async requests and PW
2532 * lock is matched whereas we asked for PR. */
2533 if (!rqset && einfo->ei_mode != mode)
2534 ldlm_lock_addref(lockh, LCK_PR);
2536 /* I would like to be able to ASSERT here that
2537 * rss <= kms, but I can't, for reasons which
2538 * are explained in lov_enqueue() */
2541 /* We already have a lock, and it's referenced.
2543 * At this point, the cl_lock::cll_state is CLS_QUEUING,
2544 * AGL upcall may change it to CLS_HELD directly. */
2545 (*upcall)(cookie, ELDLM_OK);
2547 if (einfo->ei_mode != mode)
2548 ldlm_lock_decref(lockh, LCK_PW);
2550 /* For async requests, decref the lock. */
2551 ldlm_lock_decref(lockh, einfo->ei_mode);
2552 LDLM_LOCK_PUT(matched);
2555 ldlm_lock_decref(lockh, mode);
2556 LDLM_LOCK_PUT(matched);
2562 CFS_LIST_HEAD(cancels);
2563 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2564 &RQF_LDLM_ENQUEUE_LVB);
2568 rc = ldlm_prep_enqueue_req(exp, req, &cancels, 0);
2570 ptlrpc_request_free(req);
2574 req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER,
2576 ptlrpc_request_set_replen(req);
2579 /* users of osc_enqueue() can pass this flag for ldlm_lock_match() */
2580 *flags &= ~LDLM_FL_BLOCK_GRANTED;
2582 rc = ldlm_cli_enqueue(exp, &req, einfo, res_id, policy, flags, lvb,
2583 sizeof(*lvb), LVB_T_OST, lockh, async);
2586 struct osc_enqueue_args *aa;
2587 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2588 aa = ptlrpc_req_async_args(req);
2591 aa->oa_flags = flags;
2592 aa->oa_upcall = upcall;
2593 aa->oa_cookie = cookie;
2595 aa->oa_lockh = lockh;
2598 req->rq_interpret_reply =
2599 (ptlrpc_interpterer_t)osc_enqueue_interpret;
2600 if (rqset == PTLRPCD_SET)
2601 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
2603 ptlrpc_set_add_req(rqset, req);
2604 } else if (intent) {
2605 ptlrpc_req_finished(req);
2610 rc = osc_enqueue_fini(req, lvb, upcall, cookie, flags, agl, rc);
2612 ptlrpc_req_finished(req);
2617 static int osc_enqueue(struct obd_export *exp, struct obd_info *oinfo,
2618 struct ldlm_enqueue_info *einfo,
2619 struct ptlrpc_request_set *rqset)
2621 struct ldlm_res_id res_id;
2625 ostid_build_res_name(&oinfo->oi_md->lsm_object_oid, &res_id);
2626 rc = osc_enqueue_base(exp, &res_id, &oinfo->oi_flags, &oinfo->oi_policy,
2627 &oinfo->oi_md->lsm_oinfo[0]->loi_lvb,
2628 oinfo->oi_md->lsm_oinfo[0]->loi_kms_valid,
2629 oinfo->oi_cb_up, oinfo, einfo, oinfo->oi_lockh,
2630 rqset, rqset != NULL, 0);
2634 int osc_match_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2635 __u32 type, ldlm_policy_data_t *policy, __u32 mode,
2636 int *flags, void *data, struct lustre_handle *lockh,
2639 struct obd_device *obd = exp->exp_obd;
2640 int lflags = *flags;
2644 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_MATCH))
2647 /* Filesystem lock extents are extended to page boundaries so that
2648 * dealing with the page cache is a little smoother */
2649 policy->l_extent.start -= policy->l_extent.start & ~CFS_PAGE_MASK;
2650 policy->l_extent.end |= ~CFS_PAGE_MASK;
2652 /* Next, search for already existing extent locks that will cover us */
2653 /* If we're trying to read, we also search for an existing PW lock. The
2654 * VFS and page cache already protect us locally, so lots of readers/
2655 * writers can share a single PW lock. */
2659 rc = ldlm_lock_match(obd->obd_namespace, lflags,
2660 res_id, type, policy, rc, lockh, unref);
2663 if (!osc_set_data_with_check(lockh, data)) {
2664 if (!(lflags & LDLM_FL_TEST_LOCK))
2665 ldlm_lock_decref(lockh, rc);
2669 if (!(lflags & LDLM_FL_TEST_LOCK) && mode != rc) {
2670 ldlm_lock_addref(lockh, LCK_PR);
2671 ldlm_lock_decref(lockh, LCK_PW);
2678 int osc_cancel_base(struct lustre_handle *lockh, __u32 mode)
2682 if (unlikely(mode == LCK_GROUP))
2683 ldlm_lock_decref_and_cancel(lockh, mode);
2685 ldlm_lock_decref(lockh, mode);
2690 static int osc_cancel(struct obd_export *exp, struct lov_stripe_md *md,
2691 __u32 mode, struct lustre_handle *lockh)
2694 RETURN(osc_cancel_base(lockh, mode));
2697 static int osc_cancel_unused(struct obd_export *exp,
2698 struct lov_stripe_md *lsm,
2699 ldlm_cancel_flags_t flags,
2702 struct obd_device *obd = class_exp2obd(exp);
2703 struct ldlm_res_id res_id, *resp = NULL;
2706 ostid_build_res_name(&lsm->lsm_object_oid, &res_id);
2710 return ldlm_cli_cancel_unused(obd->obd_namespace, resp, flags, opaque);
2713 static int osc_statfs_interpret(const struct lu_env *env,
2714 struct ptlrpc_request *req,
2715 struct osc_async_args *aa, int rc)
2717 struct obd_statfs *msfs;
2721 /* The request has in fact never been sent
2722 * due to issues at a higher level (LOV).
2723 * Exit immediately since the caller is
2724 * aware of the problem and takes care
2725 * of the clean up */
2728 if ((rc == -ENOTCONN || rc == -EAGAIN) &&
2729 (aa->aa_oi->oi_flags & OBD_STATFS_NODELAY))
2735 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2737 GOTO(out, rc = -EPROTO);
2740 *aa->aa_oi->oi_osfs = *msfs;
2742 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
2746 static int osc_statfs_async(struct obd_export *exp,
2747 struct obd_info *oinfo, __u64 max_age,
2748 struct ptlrpc_request_set *rqset)
2750 struct obd_device *obd = class_exp2obd(exp);
2751 struct ptlrpc_request *req;
2752 struct osc_async_args *aa;
2756 /* We could possibly pass max_age in the request (as an absolute
2757 * timestamp or a "seconds.usec ago") so the target can avoid doing
2758 * extra calls into the filesystem if that isn't necessary (e.g.
2759 * during mount that would help a bit). Having relative timestamps
2760 * is not so great if request processing is slow, while absolute
2761 * timestamps are not ideal because they need time synchronization. */
2762 req = ptlrpc_request_alloc(obd->u.cli.cl_import, &RQF_OST_STATFS);
2766 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2768 ptlrpc_request_free(req);
2771 ptlrpc_request_set_replen(req);
2772 req->rq_request_portal = OST_CREATE_PORTAL;
2773 ptlrpc_at_set_req_timeout(req);
2775 if (oinfo->oi_flags & OBD_STATFS_NODELAY) {
2776 /* procfs requests not want stat in wait for avoid deadlock */
2777 req->rq_no_resend = 1;
2778 req->rq_no_delay = 1;
2781 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_statfs_interpret;
2782 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2783 aa = ptlrpc_req_async_args(req);
2786 ptlrpc_set_add_req(rqset, req);
2790 static int osc_statfs(const struct lu_env *env, struct obd_export *exp,
2791 struct obd_statfs *osfs, __u64 max_age, __u32 flags)
2793 struct obd_device *obd = class_exp2obd(exp);
2794 struct obd_statfs *msfs;
2795 struct ptlrpc_request *req;
2796 struct obd_import *imp = NULL;
2800 /*Since the request might also come from lprocfs, so we need
2801 *sync this with client_disconnect_export Bug15684*/
2802 down_read(&obd->u.cli.cl_sem);
2803 if (obd->u.cli.cl_import)
2804 imp = class_import_get(obd->u.cli.cl_import);
2805 up_read(&obd->u.cli.cl_sem);
2809 /* We could possibly pass max_age in the request (as an absolute
2810 * timestamp or a "seconds.usec ago") so the target can avoid doing
2811 * extra calls into the filesystem if that isn't necessary (e.g.
2812 * during mount that would help a bit). Having relative timestamps
2813 * is not so great if request processing is slow, while absolute
2814 * timestamps are not ideal because they need time synchronization. */
2815 req = ptlrpc_request_alloc(imp, &RQF_OST_STATFS);
2817 class_import_put(imp);
2822 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2824 ptlrpc_request_free(req);
2827 ptlrpc_request_set_replen(req);
2828 req->rq_request_portal = OST_CREATE_PORTAL;
2829 ptlrpc_at_set_req_timeout(req);
2831 if (flags & OBD_STATFS_NODELAY) {
2832 /* procfs requests not want stat in wait for avoid deadlock */
2833 req->rq_no_resend = 1;
2834 req->rq_no_delay = 1;
2837 rc = ptlrpc_queue_wait(req);
2841 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2843 GOTO(out, rc = -EPROTO);
2850 ptlrpc_req_finished(req);
2854 /* Retrieve object striping information.
2856 * @lmmu is a pointer to an in-core struct with lmm_ost_count indicating
2857 * the maximum number of OST indices which will fit in the user buffer.
2858 * lmm_magic must be LOV_MAGIC (we only use 1 slot here).
2860 static int osc_getstripe(struct lov_stripe_md *lsm, struct lov_user_md *lump)
2862 /* we use lov_user_md_v3 because it is larger than lov_user_md_v1 */
2863 struct lov_user_md_v3 lum, *lumk;
2864 struct lov_user_ost_data_v1 *lmm_objects;
2865 int rc = 0, lum_size;
2871 /* we only need the header part from user space to get lmm_magic and
2872 * lmm_stripe_count, (the header part is common to v1 and v3) */
2873 lum_size = sizeof(struct lov_user_md_v1);
2874 if (cfs_copy_from_user(&lum, lump, lum_size))
2877 if ((lum.lmm_magic != LOV_USER_MAGIC_V1) &&
2878 (lum.lmm_magic != LOV_USER_MAGIC_V3))
2881 /* lov_user_md_vX and lov_mds_md_vX must have the same size */
2882 LASSERT(sizeof(struct lov_user_md_v1) == sizeof(struct lov_mds_md_v1));
2883 LASSERT(sizeof(struct lov_user_md_v3) == sizeof(struct lov_mds_md_v3));
2884 LASSERT(sizeof(lum.lmm_objects[0]) == sizeof(lumk->lmm_objects[0]));
2886 /* we can use lov_mds_md_size() to compute lum_size
2887 * because lov_user_md_vX and lov_mds_md_vX have the same size */
2888 if (lum.lmm_stripe_count > 0) {
2889 lum_size = lov_mds_md_size(lum.lmm_stripe_count, lum.lmm_magic);
2890 OBD_ALLOC(lumk, lum_size);
2894 if (lum.lmm_magic == LOV_USER_MAGIC_V1)
2895 lmm_objects = &(((struct lov_user_md_v1 *)lumk)->lmm_objects[0]);
2897 lmm_objects = &(lumk->lmm_objects[0]);
2898 lmm_objects->l_object_id = lsm->lsm_object_id;
2900 lum_size = lov_mds_md_size(0, lum.lmm_magic);
2904 lumk->lmm_object_id = lsm->lsm_object_id;
2905 lumk->lmm_object_seq = lsm->lsm_object_seq;
2906 lumk->lmm_stripe_count = 1;
2908 if (cfs_copy_to_user(lump, lumk, lum_size))
2912 OBD_FREE(lumk, lum_size);
2918 static int osc_iocontrol(unsigned int cmd, struct obd_export *exp, int len,
2919 void *karg, void *uarg)
2921 struct obd_device *obd = exp->exp_obd;
2922 struct obd_ioctl_data *data = karg;
2926 if (!cfs_try_module_get(THIS_MODULE)) {
2927 CERROR("Can't get module. Is it alive?");
2931 case OBD_IOC_LOV_GET_CONFIG: {
2933 struct lov_desc *desc;
2934 struct obd_uuid uuid;
2938 if (obd_ioctl_getdata(&buf, &len, (void *)uarg))
2939 GOTO(out, err = -EINVAL);
2941 data = (struct obd_ioctl_data *)buf;
2943 if (sizeof(*desc) > data->ioc_inllen1) {
2944 obd_ioctl_freedata(buf, len);
2945 GOTO(out, err = -EINVAL);
2948 if (data->ioc_inllen2 < sizeof(uuid)) {
2949 obd_ioctl_freedata(buf, len);
2950 GOTO(out, err = -EINVAL);
2953 desc = (struct lov_desc *)data->ioc_inlbuf1;
2954 desc->ld_tgt_count = 1;
2955 desc->ld_active_tgt_count = 1;
2956 desc->ld_default_stripe_count = 1;
2957 desc->ld_default_stripe_size = 0;
2958 desc->ld_default_stripe_offset = 0;
2959 desc->ld_pattern = 0;
2960 memcpy(&desc->ld_uuid, &obd->obd_uuid, sizeof(uuid));
2962 memcpy(data->ioc_inlbuf2, &obd->obd_uuid, sizeof(uuid));
2964 err = cfs_copy_to_user((void *)uarg, buf, len);
2967 obd_ioctl_freedata(buf, len);
2970 case LL_IOC_LOV_SETSTRIPE:
2971 err = obd_alloc_memmd(exp, karg);
2975 case LL_IOC_LOV_GETSTRIPE:
2976 err = osc_getstripe(karg, uarg);
2978 case OBD_IOC_CLIENT_RECOVER:
2979 err = ptlrpc_recover_import(obd->u.cli.cl_import,
2980 data->ioc_inlbuf1, 0);
2984 case IOC_OSC_SET_ACTIVE:
2985 err = ptlrpc_set_import_active(obd->u.cli.cl_import,
2988 case OBD_IOC_POLL_QUOTACHECK:
2989 err = osc_quota_poll_check(exp, (struct if_quotacheck *)karg);
2991 case OBD_IOC_PING_TARGET:
2992 err = ptlrpc_obd_ping(obd);
2995 CDEBUG(D_INODE, "unrecognised ioctl %#x by %s\n",
2996 cmd, cfs_curproc_comm());
2997 GOTO(out, err = -ENOTTY);
3000 cfs_module_put(THIS_MODULE);
3004 static int osc_get_info(const struct lu_env *env, struct obd_export *exp,
3005 obd_count keylen, void *key, __u32 *vallen, void *val,
3006 struct lov_stripe_md *lsm)
3009 if (!vallen || !val)
3012 if (KEY_IS(KEY_LOCK_TO_STRIPE)) {
3013 __u32 *stripe = val;
3014 *vallen = sizeof(*stripe);
3017 } else if (KEY_IS(KEY_LAST_ID)) {
3018 struct ptlrpc_request *req;
3023 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
3024 &RQF_OST_GET_INFO_LAST_ID);
3028 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
3029 RCL_CLIENT, keylen);
3030 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GET_INFO);
3032 ptlrpc_request_free(req);
3036 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
3037 memcpy(tmp, key, keylen);
3039 req->rq_no_delay = req->rq_no_resend = 1;
3040 ptlrpc_request_set_replen(req);
3041 rc = ptlrpc_queue_wait(req);
3045 reply = req_capsule_server_get(&req->rq_pill, &RMF_OBD_ID);
3047 GOTO(out, rc = -EPROTO);
3049 *((obd_id *)val) = *reply;
3051 ptlrpc_req_finished(req);
3053 } else if (KEY_IS(KEY_FIEMAP)) {
3054 struct ptlrpc_request *req;
3055 struct ll_user_fiemap *reply;
3059 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
3060 &RQF_OST_GET_INFO_FIEMAP);
3064 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_KEY,
3065 RCL_CLIENT, keylen);
3066 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_VAL,
3067 RCL_CLIENT, *vallen);
3068 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_VAL,
3069 RCL_SERVER, *vallen);
3071 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GET_INFO);
3073 ptlrpc_request_free(req);
3077 tmp = req_capsule_client_get(&req->rq_pill, &RMF_FIEMAP_KEY);
3078 memcpy(tmp, key, keylen);
3079 tmp = req_capsule_client_get(&req->rq_pill, &RMF_FIEMAP_VAL);
3080 memcpy(tmp, val, *vallen);
3082 ptlrpc_request_set_replen(req);
3083 rc = ptlrpc_queue_wait(req);
3087 reply = req_capsule_server_get(&req->rq_pill, &RMF_FIEMAP_VAL);
3089 GOTO(out1, rc = -EPROTO);
3091 memcpy(val, reply, *vallen);
3093 ptlrpc_req_finished(req);
3101 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
3102 obd_count keylen, void *key, obd_count vallen,
3103 void *val, struct ptlrpc_request_set *set)
3105 struct ptlrpc_request *req;
3106 struct obd_device *obd = exp->exp_obd;
3107 struct obd_import *imp = class_exp2cliimp(exp);
3112 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_SHUTDOWN, 10);
3114 if (KEY_IS(KEY_CHECKSUM)) {
3115 if (vallen != sizeof(int))
3117 exp->exp_obd->u.cli.cl_checksum = (*(int *)val) ? 1 : 0;
3121 if (KEY_IS(KEY_SPTLRPC_CONF)) {
3122 sptlrpc_conf_client_adapt(obd);
3126 if (KEY_IS(KEY_FLUSH_CTX)) {
3127 sptlrpc_import_flush_my_ctx(imp);
3131 if (KEY_IS(KEY_CACHE_SET)) {
3132 struct client_obd *cli = &obd->u.cli;
3134 LASSERT(cli->cl_cache == NULL); /* only once */
3135 cli->cl_cache = (struct cl_client_cache *)val;
3136 cfs_atomic_inc(&cli->cl_cache->ccc_users);
3137 cli->cl_lru_left = &cli->cl_cache->ccc_lru_left;
3139 /* add this osc into entity list */
3140 LASSERT(cfs_list_empty(&cli->cl_lru_osc));
3141 spin_lock(&cli->cl_cache->ccc_lru_lock);
3142 cfs_list_add(&cli->cl_lru_osc, &cli->cl_cache->ccc_lru);
3143 spin_unlock(&cli->cl_cache->ccc_lru_lock);
3148 if (KEY_IS(KEY_CACHE_LRU_SHRINK)) {
3149 struct client_obd *cli = &obd->u.cli;
3150 int nr = cfs_atomic_read(&cli->cl_lru_in_list) >> 1;
3151 int target = *(int *)val;
3153 nr = osc_lru_shrink(cli, min(nr, target));
3158 if (!set && !KEY_IS(KEY_GRANT_SHRINK))
3161 /* We pass all other commands directly to OST. Since nobody calls osc
3162 methods directly and everybody is supposed to go through LOV, we
3163 assume lov checked invalid values for us.
3164 The only recognised values so far are evict_by_nid and mds_conn.
3165 Even if something bad goes through, we'd get a -EINVAL from OST
3168 req = ptlrpc_request_alloc(imp, KEY_IS(KEY_GRANT_SHRINK) ?
3169 &RQF_OST_SET_GRANT_INFO :
3174 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
3175 RCL_CLIENT, keylen);
3176 if (!KEY_IS(KEY_GRANT_SHRINK))
3177 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_VAL,
3178 RCL_CLIENT, vallen);
3179 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SET_INFO);
3181 ptlrpc_request_free(req);
3185 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
3186 memcpy(tmp, key, keylen);
3187 tmp = req_capsule_client_get(&req->rq_pill, KEY_IS(KEY_GRANT_SHRINK) ?
3190 memcpy(tmp, val, vallen);
3192 if (KEY_IS(KEY_GRANT_SHRINK)) {
3193 struct osc_grant_args *aa;
3196 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
3197 aa = ptlrpc_req_async_args(req);
3200 ptlrpc_req_finished(req);
3203 *oa = ((struct ost_body *)val)->oa;
3205 req->rq_interpret_reply = osc_shrink_grant_interpret;
3208 ptlrpc_request_set_replen(req);
3209 if (!KEY_IS(KEY_GRANT_SHRINK)) {
3210 LASSERT(set != NULL);
3211 ptlrpc_set_add_req(set, req);
3212 ptlrpc_check_set(NULL, set);
3214 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
3220 static int osc_llog_init(struct obd_device *obd, struct obd_llog_group *olg,
3221 struct obd_device *disk_obd, int *index)
3223 /* this code is not supposed to be used with LOD/OSP
3224 * to be removed soon */
3229 static int osc_llog_finish(struct obd_device *obd, int count)
3231 struct llog_ctxt *ctxt;
3235 ctxt = llog_get_context(obd, LLOG_MDS_OST_ORIG_CTXT);
3237 llog_cat_close(NULL, ctxt->loc_handle);
3238 llog_cleanup(NULL, ctxt);
3241 ctxt = llog_get_context(obd, LLOG_SIZE_REPL_CTXT);
3243 llog_cleanup(NULL, ctxt);
3247 static int osc_reconnect(const struct lu_env *env,
3248 struct obd_export *exp, struct obd_device *obd,
3249 struct obd_uuid *cluuid,
3250 struct obd_connect_data *data,
3253 struct client_obd *cli = &obd->u.cli;
3255 if (data != NULL && (data->ocd_connect_flags & OBD_CONNECT_GRANT)) {
3258 client_obd_list_lock(&cli->cl_loi_list_lock);
3259 data->ocd_grant = (cli->cl_avail_grant + cli->cl_dirty) ?:
3260 2 * cli->cl_max_pages_per_rpc << CFS_PAGE_SHIFT;
3261 lost_grant = cli->cl_lost_grant;
3262 cli->cl_lost_grant = 0;
3263 client_obd_list_unlock(&cli->cl_loi_list_lock);
3265 CDEBUG(D_RPCTRACE, "ocd_connect_flags: "LPX64" ocd_version: %d"
3266 " ocd_grant: %d, lost: %ld.\n", data->ocd_connect_flags,
3267 data->ocd_version, data->ocd_grant, lost_grant);
3273 static int osc_disconnect(struct obd_export *exp)
3275 struct obd_device *obd = class_exp2obd(exp);
3276 struct llog_ctxt *ctxt;
3279 ctxt = llog_get_context(obd, LLOG_SIZE_REPL_CTXT);
3281 if (obd->u.cli.cl_conn_count == 1) {
3282 /* Flush any remaining cancel messages out to the
3284 llog_sync(ctxt, exp, 0);
3286 llog_ctxt_put(ctxt);
3288 CDEBUG(D_HA, "No LLOG_SIZE_REPL_CTXT found in obd %p\n",
3292 rc = client_disconnect_export(exp);
3294 * Initially we put del_shrink_grant before disconnect_export, but it
3295 * causes the following problem if setup (connect) and cleanup
3296 * (disconnect) are tangled together.
3297 * connect p1 disconnect p2
3298 * ptlrpc_connect_import
3299 * ............... class_manual_cleanup
3302 * ptlrpc_connect_interrupt
3304 * add this client to shrink list
3306 * Bang! pinger trigger the shrink.
3307 * So the osc should be disconnected from the shrink list, after we
3308 * are sure the import has been destroyed. BUG18662
3310 if (obd->u.cli.cl_import == NULL)
3311 osc_del_shrink_grant(&obd->u.cli);
3315 static int osc_import_event(struct obd_device *obd,
3316 struct obd_import *imp,
3317 enum obd_import_event event)
3319 struct client_obd *cli;
3323 LASSERT(imp->imp_obd == obd);
3326 case IMP_EVENT_DISCON: {
3328 client_obd_list_lock(&cli->cl_loi_list_lock);
3329 cli->cl_avail_grant = 0;
3330 cli->cl_lost_grant = 0;
3331 client_obd_list_unlock(&cli->cl_loi_list_lock);
3334 case IMP_EVENT_INACTIVE: {
3335 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_INACTIVE, NULL);
3338 case IMP_EVENT_INVALIDATE: {
3339 struct ldlm_namespace *ns = obd->obd_namespace;
3343 env = cl_env_get(&refcheck);
3347 /* all pages go to failing rpcs due to the invalid
3349 osc_io_unplug(env, cli, NULL, PDL_POLICY_ROUND);
3351 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
3352 cl_env_put(env, &refcheck);
3357 case IMP_EVENT_ACTIVE: {
3358 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVE, NULL);
3361 case IMP_EVENT_OCD: {
3362 struct obd_connect_data *ocd = &imp->imp_connect_data;
3364 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT)
3365 osc_init_grant(&obd->u.cli, ocd);
3368 if (ocd->ocd_connect_flags & OBD_CONNECT_REQPORTAL)
3369 imp->imp_client->cli_request_portal =OST_REQUEST_PORTAL;
3371 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_OCD, NULL);
3374 case IMP_EVENT_DEACTIVATE: {
3375 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_DEACTIVATE, NULL);
3378 case IMP_EVENT_ACTIVATE: {
3379 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVATE, NULL);
3383 CERROR("Unknown import event %d\n", event);
3390 * Determine whether the lock can be canceled before replaying the lock
3391 * during recovery, see bug16774 for detailed information.
3393 * \retval zero the lock can't be canceled
3394 * \retval other ok to cancel
3396 static int osc_cancel_for_recovery(struct ldlm_lock *lock)
3398 check_res_locked(lock->l_resource);
3401 * Cancel all unused extent lock in granted mode LCK_PR or LCK_CR.
3403 * XXX as a future improvement, we can also cancel unused write lock
3404 * if it doesn't have dirty data and active mmaps.
3406 if (lock->l_resource->lr_type == LDLM_EXTENT &&
3407 (lock->l_granted_mode == LCK_PR ||
3408 lock->l_granted_mode == LCK_CR) &&
3409 (osc_dlm_lock_pageref(lock) == 0))
3415 static int brw_queue_work(const struct lu_env *env, void *data)
3417 struct client_obd *cli = data;
3419 CDEBUG(D_CACHE, "Run writeback work for client obd %p.\n", cli);
3421 osc_io_unplug(env, cli, NULL, PDL_POLICY_SAME);
3425 int osc_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
3427 struct lprocfs_static_vars lvars = { 0 };
3428 struct client_obd *cli = &obd->u.cli;
3433 rc = ptlrpcd_addref();
3437 rc = client_obd_setup(obd, lcfg);
3439 GOTO(out_ptlrpcd, rc);
3441 handler = ptlrpcd_alloc_work(cli->cl_import, brw_queue_work, cli);
3442 if (IS_ERR(handler))
3443 GOTO(out_client_setup, rc = PTR_ERR(handler));
3444 cli->cl_writeback_work = handler;
3446 rc = osc_quota_setup(obd);
3448 GOTO(out_ptlrpcd_work, rc);
3450 cli->cl_grant_shrink_interval = GRANT_SHRINK_INTERVAL;
3451 lprocfs_osc_init_vars(&lvars);
3452 if (lprocfs_obd_setup(obd, lvars.obd_vars) == 0) {
3453 lproc_osc_attach_seqstat(obd);
3454 sptlrpc_lprocfs_cliobd_attach(obd);
3455 ptlrpc_lprocfs_register_obd(obd);
3458 /* We need to allocate a few requests more, because
3459 * brw_interpret tries to create new requests before freeing
3460 * previous ones, Ideally we want to have 2x max_rpcs_in_flight
3461 * reserved, but I'm afraid that might be too much wasted RAM
3462 * in fact, so 2 is just my guess and still should work. */
3463 cli->cl_import->imp_rq_pool =
3464 ptlrpc_init_rq_pool(cli->cl_max_rpcs_in_flight + 2,
3466 ptlrpc_add_rqs_to_pool);
3468 CFS_INIT_LIST_HEAD(&cli->cl_grant_shrink_list);
3469 ns_register_cancel(obd->obd_namespace, osc_cancel_for_recovery);
3473 ptlrpcd_destroy_work(handler);
3475 client_obd_cleanup(obd);
3481 static int osc_precleanup(struct obd_device *obd, enum obd_cleanup_stage stage)
3487 case OBD_CLEANUP_EARLY: {
3488 struct obd_import *imp;
3489 imp = obd->u.cli.cl_import;
3490 CDEBUG(D_HA, "Deactivating import %s\n", obd->obd_name);
3491 /* ptlrpc_abort_inflight to stop an mds_lov_synchronize */
3492 ptlrpc_deactivate_import(imp);
3493 spin_lock(&imp->imp_lock);
3494 imp->imp_pingable = 0;
3495 spin_unlock(&imp->imp_lock);
3498 case OBD_CLEANUP_EXPORTS: {
3499 struct client_obd *cli = &obd->u.cli;
3501 * for echo client, export may be on zombie list, wait for
3502 * zombie thread to cull it, because cli.cl_import will be
3503 * cleared in client_disconnect_export():
3504 * class_export_destroy() -> obd_cleanup() ->
3505 * echo_device_free() -> echo_client_cleanup() ->
3506 * obd_disconnect() -> osc_disconnect() ->
3507 * client_disconnect_export()
3509 obd_zombie_barrier();
3510 if (cli->cl_writeback_work) {
3511 ptlrpcd_destroy_work(cli->cl_writeback_work);
3512 cli->cl_writeback_work = NULL;
3514 obd_cleanup_client_import(obd);
3515 ptlrpc_lprocfs_unregister_obd(obd);
3516 lprocfs_obd_cleanup(obd);
3517 rc = obd_llog_finish(obd, 0);
3519 CERROR("failed to cleanup llogging subsystems\n");
3526 int osc_cleanup(struct obd_device *obd)
3528 struct client_obd *cli = &obd->u.cli;
3534 if (cli->cl_cache != NULL) {
3535 LASSERT(cfs_atomic_read(&cli->cl_cache->ccc_users) > 0);
3536 spin_lock(&cli->cl_cache->ccc_lru_lock);
3537 cfs_list_del_init(&cli->cl_lru_osc);
3538 spin_unlock(&cli->cl_cache->ccc_lru_lock);
3539 cli->cl_lru_left = NULL;
3540 cfs_atomic_dec(&cli->cl_cache->ccc_users);
3541 cli->cl_cache = NULL;
3544 /* free memory of osc quota cache */
3545 osc_quota_cleanup(obd);
3547 rc = client_obd_cleanup(obd);
3553 int osc_process_config_base(struct obd_device *obd, struct lustre_cfg *lcfg)
3555 struct lprocfs_static_vars lvars = { 0 };
3558 lprocfs_osc_init_vars(&lvars);
3560 switch (lcfg->lcfg_command) {
3562 rc = class_process_proc_param(PARAM_OSC, lvars.obd_vars,
3572 static int osc_process_config(struct obd_device *obd, obd_count len, void *buf)
3574 return osc_process_config_base(obd, buf);
3577 struct obd_ops osc_obd_ops = {
3578 .o_owner = THIS_MODULE,
3579 .o_setup = osc_setup,
3580 .o_precleanup = osc_precleanup,
3581 .o_cleanup = osc_cleanup,
3582 .o_add_conn = client_import_add_conn,
3583 .o_del_conn = client_import_del_conn,
3584 .o_connect = client_connect_import,
3585 .o_reconnect = osc_reconnect,
3586 .o_disconnect = osc_disconnect,
3587 .o_statfs = osc_statfs,
3588 .o_statfs_async = osc_statfs_async,
3589 .o_packmd = osc_packmd,
3590 .o_unpackmd = osc_unpackmd,
3591 .o_create = osc_create,
3592 .o_destroy = osc_destroy,
3593 .o_getattr = osc_getattr,
3594 .o_getattr_async = osc_getattr_async,
3595 .o_setattr = osc_setattr,
3596 .o_setattr_async = osc_setattr_async,
3598 .o_punch = osc_punch,
3600 .o_enqueue = osc_enqueue,
3601 .o_change_cbdata = osc_change_cbdata,
3602 .o_find_cbdata = osc_find_cbdata,
3603 .o_cancel = osc_cancel,
3604 .o_cancel_unused = osc_cancel_unused,
3605 .o_iocontrol = osc_iocontrol,
3606 .o_get_info = osc_get_info,
3607 .o_set_info_async = osc_set_info_async,
3608 .o_import_event = osc_import_event,
3609 .o_llog_init = osc_llog_init,
3610 .o_llog_finish = osc_llog_finish,
3611 .o_process_config = osc_process_config,
3612 .o_quotactl = osc_quotactl,
3613 .o_quotacheck = osc_quotacheck,
3616 extern struct lu_kmem_descr osc_caches[];
3617 extern spinlock_t osc_ast_guard;
3618 extern struct lock_class_key osc_ast_guard_class;
3620 int __init osc_init(void)
3622 struct lprocfs_static_vars lvars = { 0 };
3626 /* print an address of _any_ initialized kernel symbol from this
3627 * module, to allow debugging with gdb that doesn't support data
3628 * symbols from modules.*/
3629 CDEBUG(D_INFO, "Lustre OSC module (%p).\n", &osc_caches);
3631 rc = lu_kmem_init(osc_caches);
3633 lprocfs_osc_init_vars(&lvars);
3635 rc = class_register_type(&osc_obd_ops, NULL, lvars.module_vars,
3636 LUSTRE_OSC_NAME, &osc_device_type);
3638 lu_kmem_fini(osc_caches);
3642 spin_lock_init(&osc_ast_guard);
3643 lockdep_set_class(&osc_ast_guard, &osc_ast_guard_class);
3649 static void /*__exit*/ osc_exit(void)
3651 class_unregister_type(LUSTRE_OSC_NAME);
3652 lu_kmem_fini(osc_caches);
3655 MODULE_AUTHOR("Sun Microsystems, Inc. <http://www.lustre.org/>");
3656 MODULE_DESCRIPTION("Lustre Object Storage Client (OSC)");
3657 MODULE_LICENSE("GPL");
3659 cfs_module(osc, LUSTRE_VERSION_STRING, osc_init, osc_exit);
git://git.whamcloud.com / fs / lustre-release.git / blob