projects / deliverable / linux.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
staging: lustre: remove RETURN macro
[deliverable/linux.git] / drivers / staging / lustre / lustre / osc / osc_request.c
1 /*
2 * GPL HEADER START
3 *
4 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
5 *
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.
9 *
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).
15 *
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
19 *
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
22 * have any questions.
23 *
24 * GPL HEADER END
25 */
26 /*
27 * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
28 * Use is subject to license terms.
29 *
30 * Copyright (c) 2011, 2012, Intel Corporation.
31 */
32 /*
33 * This file is part of Lustre, http://www.lustre.org/
34 * Lustre is a trademark of Sun Microsystems, Inc.
35 */
36
37 #define DEBUG_SUBSYSTEM S_OSC
38
39 #include <linux/libcfs/libcfs.h>
40
41
42 #include <lustre_dlm.h>
43 #include <lustre_net.h>
44 #include <lustre/lustre_user.h>
45 #include <obd_cksum.h>
46 #include <obd_ost.h>
47 #include <obd_lov.h>
48
49 #ifdef __CYGWIN__
50 # include <ctype.h>
51 #endif
52
53 #include <lustre_ha.h>
54 #include <lprocfs_status.h>
55 #include <lustre_log.h>
56 #include <lustre_debug.h>
57 #include <lustre_param.h>
58 #include <lustre_fid.h>
59 #include "osc_internal.h"
60 #include "osc_cl_internal.h"
61
62 static void osc_release_ppga(struct brw_page **ppga, obd_count count);
63 static int brw_interpret(const struct lu_env *env,
64 struct ptlrpc_request *req, void *data, int rc);
65 int osc_cleanup(struct obd_device *obd);
66
67 /* Pack OSC object metadata for disk storage (LE byte order). */
68 static int osc_packmd(struct obd_export *exp, struct lov_mds_md **lmmp,
69 struct lov_stripe_md *lsm)
70 {
71 int lmm_size;
72
73 lmm_size = sizeof(**lmmp);
74 if (lmmp == NULL)
75 return lmm_size;
76
77 if (*lmmp != NULL && lsm == NULL) {
78 OBD_FREE(*lmmp, lmm_size);
79 *lmmp = NULL;
80 return 0;
81 } else if (unlikely(lsm != NULL && ostid_id(&lsm->lsm_oi) == 0)) {
82 return -EBADF;
83 }
84
85 if (*lmmp == NULL) {
86 OBD_ALLOC(*lmmp, lmm_size);
87 if (*lmmp == NULL)
88 return -ENOMEM;
89 }
90
91 if (lsm)
92 ostid_cpu_to_le(&lsm->lsm_oi, &(*lmmp)->lmm_oi);
93
94 return lmm_size;
95 }
96
97 /* Unpack OSC object metadata from disk storage (LE byte order). */
98 static int osc_unpackmd(struct obd_export *exp, struct lov_stripe_md **lsmp,
99 struct lov_mds_md *lmm, int lmm_bytes)
100 {
101 int lsm_size;
102 struct obd_import *imp = class_exp2cliimp(exp);
103
104 if (lmm != NULL) {
105 if (lmm_bytes < sizeof(*lmm)) {
106 CERROR("%s: lov_mds_md too small: %d, need %d\n",
107 exp->exp_obd->obd_name, lmm_bytes,
108 (int)sizeof(*lmm));
109 return -EINVAL;
110 }
111 /* XXX LOV_MAGIC etc check? */
112
113 if (unlikely(ostid_id(&lmm->lmm_oi) == 0)) {
114 CERROR("%s: zero lmm_object_id: rc = %d\n",
115 exp->exp_obd->obd_name, -EINVAL);
116 return -EINVAL;
117 }
118 }
119
120 lsm_size = lov_stripe_md_size(1);
121 if (lsmp == NULL)
122 return lsm_size;
123
124 if (*lsmp != NULL && lmm == NULL) {
125 OBD_FREE((*lsmp)->lsm_oinfo[0], sizeof(struct lov_oinfo));
126 OBD_FREE(*lsmp, lsm_size);
127 *lsmp = NULL;
128 return 0;
129 }
130
131 if (*lsmp == NULL) {
132 OBD_ALLOC(*lsmp, lsm_size);
133 if (unlikely(*lsmp == NULL))
134 return -ENOMEM;
135 OBD_ALLOC((*lsmp)->lsm_oinfo[0], sizeof(struct lov_oinfo));
136 if (unlikely((*lsmp)->lsm_oinfo[0] == NULL)) {
137 OBD_FREE(*lsmp, lsm_size);
138 return -ENOMEM;
139 }
140 loi_init((*lsmp)->lsm_oinfo[0]);
141 } else if (unlikely(ostid_id(&(*lsmp)->lsm_oi) == 0)) {
142 return -EBADF;
143 }
144
145 if (lmm != NULL)
146 /* XXX zero *lsmp? */
147 ostid_le_to_cpu(&lmm->lmm_oi, &(*lsmp)->lsm_oi);
148
149 if (imp != NULL &&
150 (imp->imp_connect_data.ocd_connect_flags & OBD_CONNECT_MAXBYTES))
151 (*lsmp)->lsm_maxbytes = imp->imp_connect_data.ocd_maxbytes;
152 else
153 (*lsmp)->lsm_maxbytes = LUSTRE_STRIPE_MAXBYTES;
154
155 return lsm_size;
156 }
157
158 static inline void osc_pack_capa(struct ptlrpc_request *req,
159 struct ost_body *body, void *capa)
160 {
161 struct obd_capa *oc = (struct obd_capa *)capa;
162 struct lustre_capa *c;
163
164 if (!capa)
165 return;
166
167 c = req_capsule_client_get(&req->rq_pill, &RMF_CAPA1);
168 LASSERT(c);
169 capa_cpy(c, oc);
170 body->oa.o_valid |= OBD_MD_FLOSSCAPA;
171 DEBUG_CAPA(D_SEC, c, "pack");
172 }
173
174 static inline void osc_pack_req_body(struct ptlrpc_request *req,
175 struct obd_info *oinfo)
176 {
177 struct ost_body *body;
178
179 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
180 LASSERT(body);
181
182 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
183 oinfo->oi_oa);
184 osc_pack_capa(req, body, oinfo->oi_capa);
185 }
186
187 static inline void osc_set_capa_size(struct ptlrpc_request *req,
188 const struct req_msg_field *field,
189 struct obd_capa *oc)
190 {
191 if (oc == NULL)
192 req_capsule_set_size(&req->rq_pill, field, RCL_CLIENT, 0);
193 else
194 /* it is already calculated as sizeof struct obd_capa */
195 ;
196 }
197
198 static int osc_getattr_interpret(const struct lu_env *env,
199 struct ptlrpc_request *req,
200 struct osc_async_args *aa, int rc)
201 {
202 struct ost_body *body;
203
204 if (rc != 0)
205 GOTO(out, rc);
206
207 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
208 if (body) {
209 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
210 lustre_get_wire_obdo(&req->rq_import->imp_connect_data,
211 aa->aa_oi->oi_oa, &body->oa);
212
213 /* This should really be sent by the OST */
214 aa->aa_oi->oi_oa->o_blksize = DT_MAX_BRW_SIZE;
215 aa->aa_oi->oi_oa->o_valid |= OBD_MD_FLBLKSZ;
216 } else {
217 CDEBUG(D_INFO, "can't unpack ost_body\n");
218 rc = -EPROTO;
219 aa->aa_oi->oi_oa->o_valid = 0;
220 }
221 out:
222 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
223 return rc;
224 }
225
226 static int osc_getattr_async(struct obd_export *exp, struct obd_info *oinfo,
227 struct ptlrpc_request_set *set)
228 {
229 struct ptlrpc_request *req;
230 struct osc_async_args *aa;
231 int rc;
232
233 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
234 if (req == NULL)
235 return -ENOMEM;
236
237 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
238 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
239 if (rc) {
240 ptlrpc_request_free(req);
241 return rc;
242 }
243
244 osc_pack_req_body(req, oinfo);
245
246 ptlrpc_request_set_replen(req);
247 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_getattr_interpret;
248
249 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
250 aa = ptlrpc_req_async_args(req);
251 aa->aa_oi = oinfo;
252
253 ptlrpc_set_add_req(set, req);
254 return 0;
255 }
256
257 static int osc_getattr(const struct lu_env *env, struct obd_export *exp,
258 struct obd_info *oinfo)
259 {
260 struct ptlrpc_request *req;
261 struct ost_body *body;
262 int rc;
263
264 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_GETATTR);
265 if (req == NULL)
266 return -ENOMEM;
267
268 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
269 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GETATTR);
270 if (rc) {
271 ptlrpc_request_free(req);
272 return rc;
273 }
274
275 osc_pack_req_body(req, oinfo);
276
277 ptlrpc_request_set_replen(req);
278
279 rc = ptlrpc_queue_wait(req);
280 if (rc)
281 GOTO(out, rc);
282
283 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
284 if (body == NULL)
285 GOTO(out, rc = -EPROTO);
286
287 CDEBUG(D_INODE, "mode: %o\n", body->oa.o_mode);
288 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oinfo->oi_oa,
289 &body->oa);
290
291 oinfo->oi_oa->o_blksize = cli_brw_size(exp->exp_obd);
292 oinfo->oi_oa->o_valid |= OBD_MD_FLBLKSZ;
293
294 out:
295 ptlrpc_req_finished(req);
296 return rc;
297 }
298
299 static int osc_setattr(const struct lu_env *env, struct obd_export *exp,
300 struct obd_info *oinfo, struct obd_trans_info *oti)
301 {
302 struct ptlrpc_request *req;
303 struct ost_body *body;
304 int rc;
305
306 LASSERT(oinfo->oi_oa->o_valid & OBD_MD_FLGROUP);
307
308 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
309 if (req == NULL)
310 return -ENOMEM;
311
312 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
313 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
314 if (rc) {
315 ptlrpc_request_free(req);
316 return rc;
317 }
318
319 osc_pack_req_body(req, oinfo);
320
321 ptlrpc_request_set_replen(req);
322
323 rc = ptlrpc_queue_wait(req);
324 if (rc)
325 GOTO(out, rc);
326
327 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
328 if (body == NULL)
329 GOTO(out, rc = -EPROTO);
330
331 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oinfo->oi_oa,
332 &body->oa);
333
334 out:
335 ptlrpc_req_finished(req);
336 return rc;
337 }
338
339 static int osc_setattr_interpret(const struct lu_env *env,
340 struct ptlrpc_request *req,
341 struct osc_setattr_args *sa, int rc)
342 {
343 struct ost_body *body;
344
345 if (rc != 0)
346 GOTO(out, rc);
347
348 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
349 if (body == NULL)
350 GOTO(out, rc = -EPROTO);
351
352 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, sa->sa_oa,
353 &body->oa);
354 out:
355 rc = sa->sa_upcall(sa->sa_cookie, rc);
356 return rc;
357 }
358
359 int osc_setattr_async_base(struct obd_export *exp, struct obd_info *oinfo,
360 struct obd_trans_info *oti,
361 obd_enqueue_update_f upcall, void *cookie,
362 struct ptlrpc_request_set *rqset)
363 {
364 struct ptlrpc_request *req;
365 struct osc_setattr_args *sa;
366 int rc;
367
368 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SETATTR);
369 if (req == NULL)
370 return -ENOMEM;
371
372 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
373 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SETATTR);
374 if (rc) {
375 ptlrpc_request_free(req);
376 return rc;
377 }
378
379 if (oti && oinfo->oi_oa->o_valid & OBD_MD_FLCOOKIE)
380 oinfo->oi_oa->o_lcookie = *oti->oti_logcookies;
381
382 osc_pack_req_body(req, oinfo);
383
384 ptlrpc_request_set_replen(req);
385
386 /* do mds to ost setattr asynchronously */
387 if (!rqset) {
388 /* Do not wait for response. */
389 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
390 } else {
391 req->rq_interpret_reply =
392 (ptlrpc_interpterer_t)osc_setattr_interpret;
393
394 CLASSERT (sizeof(*sa) <= sizeof(req->rq_async_args));
395 sa = ptlrpc_req_async_args(req);
396 sa->sa_oa = oinfo->oi_oa;
397 sa->sa_upcall = upcall;
398 sa->sa_cookie = cookie;
399
400 if (rqset == PTLRPCD_SET)
401 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
402 else
403 ptlrpc_set_add_req(rqset, req);
404 }
405
406 return 0;
407 }
408
409 static int osc_setattr_async(struct obd_export *exp, struct obd_info *oinfo,
410 struct obd_trans_info *oti,
411 struct ptlrpc_request_set *rqset)
412 {
413 return osc_setattr_async_base(exp, oinfo, oti,
414 oinfo->oi_cb_up, oinfo, rqset);
415 }
416
417 int osc_real_create(struct obd_export *exp, struct obdo *oa,
418 struct lov_stripe_md **ea, struct obd_trans_info *oti)
419 {
420 struct ptlrpc_request *req;
421 struct ost_body *body;
422 struct lov_stripe_md *lsm;
423 int rc;
424
425 LASSERT(oa);
426 LASSERT(ea);
427
428 lsm = *ea;
429 if (!lsm) {
430 rc = obd_alloc_memmd(exp, &lsm);
431 if (rc < 0)
432 return rc;
433 }
434
435 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_CREATE);
436 if (req == NULL)
437 GOTO(out, rc = -ENOMEM);
438
439 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_CREATE);
440 if (rc) {
441 ptlrpc_request_free(req);
442 GOTO(out, rc);
443 }
444
445 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
446 LASSERT(body);
447
448 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
449
450 ptlrpc_request_set_replen(req);
451
452 if ((oa->o_valid & OBD_MD_FLFLAGS) &&
453 oa->o_flags == OBD_FL_DELORPHAN) {
454 DEBUG_REQ(D_HA, req,
455 "delorphan from OST integration");
456 /* Don't resend the delorphan req */
457 req->rq_no_resend = req->rq_no_delay = 1;
458 }
459
460 rc = ptlrpc_queue_wait(req);
461 if (rc)
462 GOTO(out_req, rc);
463
464 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
465 if (body == NULL)
466 GOTO(out_req, rc = -EPROTO);
467
468 CDEBUG(D_INFO, "oa flags %x\n", oa->o_flags);
469 lustre_get_wire_obdo(&req->rq_import->imp_connect_data, oa, &body->oa);
470
471 oa->o_blksize = cli_brw_size(exp->exp_obd);
472 oa->o_valid |= OBD_MD_FLBLKSZ;
473
474 /* XXX LOV STACKING: the lsm that is passed to us from LOV does not
475 * have valid lsm_oinfo data structs, so don't go touching that.
476 * This needs to be fixed in a big way.
477 */
478 lsm->lsm_oi = oa->o_oi;
479 *ea = lsm;
480
481 if (oti != NULL) {
482 oti->oti_transno = lustre_msg_get_transno(req->rq_repmsg);
483
484 if (oa->o_valid & OBD_MD_FLCOOKIE) {
485 if (!oti->oti_logcookies)
486 oti_alloc_cookies(oti, 1);
487 *oti->oti_logcookies = oa->o_lcookie;
488 }
489 }
490
491 CDEBUG(D_HA, "transno: "LPD64"\n",
492 lustre_msg_get_transno(req->rq_repmsg));
493 out_req:
494 ptlrpc_req_finished(req);
495 out:
496 if (rc && !*ea)
497 obd_free_memmd(exp, &lsm);
498 return rc;
499 }
500
501 int osc_punch_base(struct obd_export *exp, struct obd_info *oinfo,
502 obd_enqueue_update_f upcall, void *cookie,
503 struct ptlrpc_request_set *rqset)
504 {
505 struct ptlrpc_request *req;
506 struct osc_setattr_args *sa;
507 struct ost_body *body;
508 int rc;
509
510 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_PUNCH);
511 if (req == NULL)
512 return -ENOMEM;
513
514 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
515 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_PUNCH);
516 if (rc) {
517 ptlrpc_request_free(req);
518 return rc;
519 }
520 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
521 ptlrpc_at_set_req_timeout(req);
522
523 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
524 LASSERT(body);
525 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
526 oinfo->oi_oa);
527 osc_pack_capa(req, body, oinfo->oi_capa);
528
529 ptlrpc_request_set_replen(req);
530
531 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_setattr_interpret;
532 CLASSERT (sizeof(*sa) <= sizeof(req->rq_async_args));
533 sa = ptlrpc_req_async_args(req);
534 sa->sa_oa = oinfo->oi_oa;
535 sa->sa_upcall = upcall;
536 sa->sa_cookie = cookie;
537 if (rqset == PTLRPCD_SET)
538 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
539 else
540 ptlrpc_set_add_req(rqset, req);
541
542 return 0;
543 }
544
545 static int osc_punch(const struct lu_env *env, struct obd_export *exp,
546 struct obd_info *oinfo, struct obd_trans_info *oti,
547 struct ptlrpc_request_set *rqset)
548 {
549 oinfo->oi_oa->o_size = oinfo->oi_policy.l_extent.start;
550 oinfo->oi_oa->o_blocks = oinfo->oi_policy.l_extent.end;
551 oinfo->oi_oa->o_valid |= OBD_MD_FLSIZE | OBD_MD_FLBLOCKS;
552 return osc_punch_base(exp, oinfo,
553 oinfo->oi_cb_up, oinfo, rqset);
554 }
555
556 static int osc_sync_interpret(const struct lu_env *env,
557 struct ptlrpc_request *req,
558 void *arg, int rc)
559 {
560 struct osc_fsync_args *fa = arg;
561 struct ost_body *body;
562
563 if (rc)
564 GOTO(out, rc);
565
566 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
567 if (body == NULL) {
568 CERROR ("can't unpack ost_body\n");
569 GOTO(out, rc = -EPROTO);
570 }
571
572 *fa->fa_oi->oi_oa = body->oa;
573 out:
574 rc = fa->fa_upcall(fa->fa_cookie, rc);
575 return rc;
576 }
577
578 int osc_sync_base(struct obd_export *exp, struct obd_info *oinfo,
579 obd_enqueue_update_f upcall, void *cookie,
580 struct ptlrpc_request_set *rqset)
581 {
582 struct ptlrpc_request *req;
583 struct ost_body *body;
584 struct osc_fsync_args *fa;
585 int rc;
586
587 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_SYNC);
588 if (req == NULL)
589 return -ENOMEM;
590
591 osc_set_capa_size(req, &RMF_CAPA1, oinfo->oi_capa);
592 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SYNC);
593 if (rc) {
594 ptlrpc_request_free(req);
595 return rc;
596 }
597
598 /* overload the size and blocks fields in the oa with start/end */
599 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
600 LASSERT(body);
601 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa,
602 oinfo->oi_oa);
603 osc_pack_capa(req, body, oinfo->oi_capa);
604
605 ptlrpc_request_set_replen(req);
606 req->rq_interpret_reply = osc_sync_interpret;
607
608 CLASSERT(sizeof(*fa) <= sizeof(req->rq_async_args));
609 fa = ptlrpc_req_async_args(req);
610 fa->fa_oi = oinfo;
611 fa->fa_upcall = upcall;
612 fa->fa_cookie = cookie;
613
614 if (rqset == PTLRPCD_SET)
615 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
616 else
617 ptlrpc_set_add_req(rqset, req);
618
619 return 0;
620 }
621
622 static int osc_sync(const struct lu_env *env, struct obd_export *exp,
623 struct obd_info *oinfo, obd_size start, obd_size end,
624 struct ptlrpc_request_set *set)
625 {
626 if (!oinfo->oi_oa) {
627 CDEBUG(D_INFO, "oa NULL\n");
628 return -EINVAL;
629 }
630
631 oinfo->oi_oa->o_size = start;
632 oinfo->oi_oa->o_blocks = end;
633 oinfo->oi_oa->o_valid |= (OBD_MD_FLSIZE | OBD_MD_FLBLOCKS);
634
635 return osc_sync_base(exp, oinfo, oinfo->oi_cb_up, oinfo, set);
636 }
637
638 /* Find and cancel locally locks matched by @mode in the resource found by
639 * @objid. Found locks are added into @cancel list. Returns the amount of
640 * locks added to @cancels list. */
641 static int osc_resource_get_unused(struct obd_export *exp, struct obdo *oa,
642 struct list_head *cancels,
643 ldlm_mode_t mode, int lock_flags)
644 {
645 struct ldlm_namespace *ns = exp->exp_obd->obd_namespace;
646 struct ldlm_res_id res_id;
647 struct ldlm_resource *res;
648 int count;
649
650 /* Return, i.e. cancel nothing, only if ELC is supported (flag in
651 * export) but disabled through procfs (flag in NS).
652 *
653 * This distinguishes from a case when ELC is not supported originally,
654 * when we still want to cancel locks in advance and just cancel them
655 * locally, without sending any RPC. */
656 if (exp_connect_cancelset(exp) && !ns_connect_cancelset(ns))
657 return 0;
658
659 ostid_build_res_name(&oa->o_oi, &res_id);
660 res = ldlm_resource_get(ns, NULL, &res_id, 0, 0);
661 if (res == NULL)
662 return 0;
663
664 LDLM_RESOURCE_ADDREF(res);
665 count = ldlm_cancel_resource_local(res, cancels, NULL, mode,
666 lock_flags, 0, NULL);
667 LDLM_RESOURCE_DELREF(res);
668 ldlm_resource_putref(res);
669 return count;
670 }
671
672 static int osc_destroy_interpret(const struct lu_env *env,
673 struct ptlrpc_request *req, void *data,
674 int rc)
675 {
676 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
677
678 atomic_dec(&cli->cl_destroy_in_flight);
679 wake_up(&cli->cl_destroy_waitq);
680 return 0;
681 }
682
683 static int osc_can_send_destroy(struct client_obd *cli)
684 {
685 if (atomic_inc_return(&cli->cl_destroy_in_flight) <=
686 cli->cl_max_rpcs_in_flight) {
687 /* The destroy request can be sent */
688 return 1;
689 }
690 if (atomic_dec_return(&cli->cl_destroy_in_flight) <
691 cli->cl_max_rpcs_in_flight) {
692 /*
693 * The counter has been modified between the two atomic
694 * operations.
695 */
696 wake_up(&cli->cl_destroy_waitq);
697 }
698 return 0;
699 }
700
701 int osc_create(const struct lu_env *env, struct obd_export *exp,
702 struct obdo *oa, struct lov_stripe_md **ea,
703 struct obd_trans_info *oti)
704 {
705 int rc = 0;
706
707 LASSERT(oa);
708 LASSERT(ea);
709 LASSERT(oa->o_valid & OBD_MD_FLGROUP);
710
711 if ((oa->o_valid & OBD_MD_FLFLAGS) &&
712 oa->o_flags == OBD_FL_RECREATE_OBJS) {
713 return osc_real_create(exp, oa, ea, oti);
714 }
715
716 if (!fid_seq_is_mdt(ostid_seq(&oa->o_oi)))
717 return osc_real_create(exp, oa, ea, oti);
718
719 /* we should not get here anymore */
720 LBUG();
721
722 return rc;
723 }
724
725 /* Destroy requests can be async always on the client, and we don't even really
726 * care about the return code since the client cannot do anything at all about
727 * a destroy failure.
728 * When the MDS is unlinking a filename, it saves the file objects into a
729 * recovery llog, and these object records are cancelled when the OST reports
730 * they were destroyed and sync'd to disk (i.e. transaction committed).
731 * If the client dies, or the OST is down when the object should be destroyed,
732 * the records are not cancelled, and when the OST reconnects to the MDS next,
733 * it will retrieve the llog unlink logs and then sends the log cancellation
734 * cookies to the MDS after committing destroy transactions. */
735 static int osc_destroy(const struct lu_env *env, struct obd_export *exp,
736 struct obdo *oa, struct lov_stripe_md *ea,
737 struct obd_trans_info *oti, struct obd_export *md_export,
738 void *capa)
739 {
740 struct client_obd *cli = &exp->exp_obd->u.cli;
741 struct ptlrpc_request *req;
742 struct ost_body *body;
743 LIST_HEAD(cancels);
744 int rc, count;
745
746 if (!oa) {
747 CDEBUG(D_INFO, "oa NULL\n");
748 return -EINVAL;
749 }
750
751 count = osc_resource_get_unused(exp, oa, &cancels, LCK_PW,
752 LDLM_FL_DISCARD_DATA);
753
754 req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_OST_DESTROY);
755 if (req == NULL) {
756 ldlm_lock_list_put(&cancels, l_bl_ast, count);
757 return -ENOMEM;
758 }
759
760 osc_set_capa_size(req, &RMF_CAPA1, (struct obd_capa *)capa);
761 rc = ldlm_prep_elc_req(exp, req, LUSTRE_OST_VERSION, OST_DESTROY,
762 0, &cancels, count);
763 if (rc) {
764 ptlrpc_request_free(req);
765 return rc;
766 }
767
768 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
769 ptlrpc_at_set_req_timeout(req);
770
771 if (oti != NULL && oa->o_valid & OBD_MD_FLCOOKIE)
772 oa->o_lcookie = *oti->oti_logcookies;
773 body = req_capsule_client_get(&req->rq_pill, &RMF_OST_BODY);
774 LASSERT(body);
775 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
776
777 osc_pack_capa(req, body, (struct obd_capa *)capa);
778 ptlrpc_request_set_replen(req);
779
780 /* If osc_destory is for destroying the unlink orphan,
781 * sent from MDT to OST, which should not be blocked here,
782 * because the process might be triggered by ptlrpcd, and
783 * it is not good to block ptlrpcd thread (b=16006)*/
784 if (!(oa->o_flags & OBD_FL_DELORPHAN)) {
785 req->rq_interpret_reply = osc_destroy_interpret;
786 if (!osc_can_send_destroy(cli)) {
787 struct l_wait_info lwi = LWI_INTR(LWI_ON_SIGNAL_NOOP,
788 NULL);
789
790 /*
791 * Wait until the number of on-going destroy RPCs drops
792 * under max_rpc_in_flight
793 */
794 l_wait_event_exclusive(cli->cl_destroy_waitq,
795 osc_can_send_destroy(cli), &lwi);
796 }
797 }
798
799 /* Do not wait for response */
800 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
801 return 0;
802 }
803
804 static void osc_announce_cached(struct client_obd *cli, struct obdo *oa,
805 long writing_bytes)
806 {
807 obd_flag bits = OBD_MD_FLBLOCKS|OBD_MD_FLGRANT;
808
809 LASSERT(!(oa->o_valid & bits));
810
811 oa->o_valid |= bits;
812 client_obd_list_lock(&cli->cl_loi_list_lock);
813 oa->o_dirty = cli->cl_dirty;
814 if (unlikely(cli->cl_dirty - cli->cl_dirty_transit >
815 cli->cl_dirty_max)) {
816 CERROR("dirty %lu - %lu > dirty_max %lu\n",
817 cli->cl_dirty, cli->cl_dirty_transit, cli->cl_dirty_max);
818 oa->o_undirty = 0;
819 } else if (unlikely(atomic_read(&obd_dirty_pages) -
820 atomic_read(&obd_dirty_transit_pages) >
821 (long)(obd_max_dirty_pages + 1))) {
822 /* The atomic_read() allowing the atomic_inc() are
823 * not covered by a lock thus they may safely race and trip
824 * this CERROR() unless we add in a small fudge factor (+1). */
825 CERROR("dirty %d - %d > system dirty_max %d\n",
826 atomic_read(&obd_dirty_pages),
827 atomic_read(&obd_dirty_transit_pages),
828 obd_max_dirty_pages);
829 oa->o_undirty = 0;
830 } else if (unlikely(cli->cl_dirty_max - cli->cl_dirty > 0x7fffffff)) {
831 CERROR("dirty %lu - dirty_max %lu too big???\n",
832 cli->cl_dirty, cli->cl_dirty_max);
833 oa->o_undirty = 0;
834 } else {
835 long max_in_flight = (cli->cl_max_pages_per_rpc <<
836 PAGE_CACHE_SHIFT)*
837 (cli->cl_max_rpcs_in_flight + 1);
838 oa->o_undirty = max(cli->cl_dirty_max, max_in_flight);
839 }
840 oa->o_grant = cli->cl_avail_grant + cli->cl_reserved_grant;
841 oa->o_dropped = cli->cl_lost_grant;
842 cli->cl_lost_grant = 0;
843 client_obd_list_unlock(&cli->cl_loi_list_lock);
844 CDEBUG(D_CACHE,"dirty: "LPU64" undirty: %u dropped %u grant: "LPU64"\n",
845 oa->o_dirty, oa->o_undirty, oa->o_dropped, oa->o_grant);
846
847 }
848
849 void osc_update_next_shrink(struct client_obd *cli)
850 {
851 cli->cl_next_shrink_grant =
852 cfs_time_shift(cli->cl_grant_shrink_interval);
853 CDEBUG(D_CACHE, "next time %ld to shrink grant \n",
854 cli->cl_next_shrink_grant);
855 }
856
857 static void __osc_update_grant(struct client_obd *cli, obd_size grant)
858 {
859 client_obd_list_lock(&cli->cl_loi_list_lock);
860 cli->cl_avail_grant += grant;
861 client_obd_list_unlock(&cli->cl_loi_list_lock);
862 }
863
864 static void osc_update_grant(struct client_obd *cli, struct ost_body *body)
865 {
866 if (body->oa.o_valid & OBD_MD_FLGRANT) {
867 CDEBUG(D_CACHE, "got "LPU64" extra grant\n", body->oa.o_grant);
868 __osc_update_grant(cli, body->oa.o_grant);
869 }
870 }
871
872 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
873 obd_count keylen, void *key, obd_count vallen,
874 void *val, struct ptlrpc_request_set *set);
875
876 static int osc_shrink_grant_interpret(const struct lu_env *env,
877 struct ptlrpc_request *req,
878 void *aa, int rc)
879 {
880 struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
881 struct obdo *oa = ((struct osc_grant_args *)aa)->aa_oa;
882 struct ost_body *body;
883
884 if (rc != 0) {
885 __osc_update_grant(cli, oa->o_grant);
886 GOTO(out, rc);
887 }
888
889 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
890 LASSERT(body);
891 osc_update_grant(cli, body);
892 out:
893 OBDO_FREE(oa);
894 return rc;
895 }
896
897 static void osc_shrink_grant_local(struct client_obd *cli, struct obdo *oa)
898 {
899 client_obd_list_lock(&cli->cl_loi_list_lock);
900 oa->o_grant = cli->cl_avail_grant / 4;
901 cli->cl_avail_grant -= oa->o_grant;
902 client_obd_list_unlock(&cli->cl_loi_list_lock);
903 if (!(oa->o_valid & OBD_MD_FLFLAGS)) {
904 oa->o_valid |= OBD_MD_FLFLAGS;
905 oa->o_flags = 0;
906 }
907 oa->o_flags |= OBD_FL_SHRINK_GRANT;
908 osc_update_next_shrink(cli);
909 }
910
911 /* Shrink the current grant, either from some large amount to enough for a
912 * full set of in-flight RPCs, or if we have already shrunk to that limit
913 * then to enough for a single RPC. This avoids keeping more grant than
914 * needed, and avoids shrinking the grant piecemeal. */
915 static int osc_shrink_grant(struct client_obd *cli)
916 {
917 __u64 target_bytes = (cli->cl_max_rpcs_in_flight + 1) *
918 (cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT);
919
920 client_obd_list_lock(&cli->cl_loi_list_lock);
921 if (cli->cl_avail_grant <= target_bytes)
922 target_bytes = cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT;
923 client_obd_list_unlock(&cli->cl_loi_list_lock);
924
925 return osc_shrink_grant_to_target(cli, target_bytes);
926 }
927
928 int osc_shrink_grant_to_target(struct client_obd *cli, __u64 target_bytes)
929 {
930 int rc = 0;
931 struct ost_body *body;
932
933 client_obd_list_lock(&cli->cl_loi_list_lock);
934 /* Don't shrink if we are already above or below the desired limit
935 * We don't want to shrink below a single RPC, as that will negatively
936 * impact block allocation and long-term performance. */
937 if (target_bytes < cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT)
938 target_bytes = cli->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT;
939
940 if (target_bytes >= cli->cl_avail_grant) {
941 client_obd_list_unlock(&cli->cl_loi_list_lock);
942 return 0;
943 }
944 client_obd_list_unlock(&cli->cl_loi_list_lock);
945
946 OBD_ALLOC_PTR(body);
947 if (!body)
948 return -ENOMEM;
949
950 osc_announce_cached(cli, &body->oa, 0);
951
952 client_obd_list_lock(&cli->cl_loi_list_lock);
953 body->oa.o_grant = cli->cl_avail_grant - target_bytes;
954 cli->cl_avail_grant = target_bytes;
955 client_obd_list_unlock(&cli->cl_loi_list_lock);
956 if (!(body->oa.o_valid & OBD_MD_FLFLAGS)) {
957 body->oa.o_valid |= OBD_MD_FLFLAGS;
958 body->oa.o_flags = 0;
959 }
960 body->oa.o_flags |= OBD_FL_SHRINK_GRANT;
961 osc_update_next_shrink(cli);
962
963 rc = osc_set_info_async(NULL, cli->cl_import->imp_obd->obd_self_export,
964 sizeof(KEY_GRANT_SHRINK), KEY_GRANT_SHRINK,
965 sizeof(*body), body, NULL);
966 if (rc != 0)
967 __osc_update_grant(cli, body->oa.o_grant);
968 OBD_FREE_PTR(body);
969 return rc;
970 }
971
972 static int osc_should_shrink_grant(struct client_obd *client)
973 {
974 cfs_time_t time = cfs_time_current();
975 cfs_time_t next_shrink = client->cl_next_shrink_grant;
976
977 if ((client->cl_import->imp_connect_data.ocd_connect_flags &
978 OBD_CONNECT_GRANT_SHRINK) == 0)
979 return 0;
980
981 if (cfs_time_aftereq(time, next_shrink - 5 * CFS_TICK)) {
982 /* Get the current RPC size directly, instead of going via:
983 * cli_brw_size(obd->u.cli.cl_import->imp_obd->obd_self_export)
984 * Keep comment here so that it can be found by searching. */
985 int brw_size = client->cl_max_pages_per_rpc << PAGE_CACHE_SHIFT;
986
987 if (client->cl_import->imp_state == LUSTRE_IMP_FULL &&
988 client->cl_avail_grant > brw_size)
989 return 1;
990 else
991 osc_update_next_shrink(client);
992 }
993 return 0;
994 }
995
996 static int osc_grant_shrink_grant_cb(struct timeout_item *item, void *data)
997 {
998 struct client_obd *client;
999
1000 list_for_each_entry(client, &item->ti_obd_list,
1001 cl_grant_shrink_list) {
1002 if (osc_should_shrink_grant(client))
1003 osc_shrink_grant(client);
1004 }
1005 return 0;
1006 }
1007
1008 static int osc_add_shrink_grant(struct client_obd *client)
1009 {
1010 int rc;
1011
1012 rc = ptlrpc_add_timeout_client(client->cl_grant_shrink_interval,
1013 TIMEOUT_GRANT,
1014 osc_grant_shrink_grant_cb, NULL,
1015 &client->cl_grant_shrink_list);
1016 if (rc) {
1017 CERROR("add grant client %s error %d\n",
1018 client->cl_import->imp_obd->obd_name, rc);
1019 return rc;
1020 }
1021 CDEBUG(D_CACHE, "add grant client %s \n",
1022 client->cl_import->imp_obd->obd_name);
1023 osc_update_next_shrink(client);
1024 return 0;
1025 }
1026
1027 static int osc_del_shrink_grant(struct client_obd *client)
1028 {
1029 return ptlrpc_del_timeout_client(&client->cl_grant_shrink_list,
1030 TIMEOUT_GRANT);
1031 }
1032
1033 static void osc_init_grant(struct client_obd *cli, struct obd_connect_data *ocd)
1034 {
1035 /*
1036 * ocd_grant is the total grant amount we're expect to hold: if we've
1037 * been evicted, it's the new avail_grant amount, cl_dirty will drop
1038 * to 0 as inflight RPCs fail out; otherwise, it's avail_grant + dirty.
1039 *
1040 * race is tolerable here: if we're evicted, but imp_state already
1041 * left EVICTED state, then cl_dirty must be 0 already.
1042 */
1043 client_obd_list_lock(&cli->cl_loi_list_lock);
1044 if (cli->cl_import->imp_state == LUSTRE_IMP_EVICTED)
1045 cli->cl_avail_grant = ocd->ocd_grant;
1046 else
1047 cli->cl_avail_grant = ocd->ocd_grant - cli->cl_dirty;
1048
1049 if (cli->cl_avail_grant < 0) {
1050 CWARN("%s: available grant < 0: avail/ocd/dirty %ld/%u/%ld\n",
1051 cli->cl_import->imp_obd->obd_name, cli->cl_avail_grant,
1052 ocd->ocd_grant, cli->cl_dirty);
1053 /* workaround for servers which do not have the patch from
1054 * LU-2679 */
1055 cli->cl_avail_grant = ocd->ocd_grant;
1056 }
1057
1058 /* determine the appropriate chunk size used by osc_extent. */
1059 cli->cl_chunkbits = max_t(int, PAGE_CACHE_SHIFT, ocd->ocd_blocksize);
1060 client_obd_list_unlock(&cli->cl_loi_list_lock);
1061
1062 CDEBUG(D_CACHE, "%s, setting cl_avail_grant: %ld cl_lost_grant: %ld."
1063 "chunk bits: %d.\n", cli->cl_import->imp_obd->obd_name,
1064 cli->cl_avail_grant, cli->cl_lost_grant, cli->cl_chunkbits);
1065
1066 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT_SHRINK &&
1067 list_empty(&cli->cl_grant_shrink_list))
1068 osc_add_shrink_grant(cli);
1069 }
1070
1071 /* We assume that the reason this OSC got a short read is because it read
1072 * beyond the end of a stripe file; i.e. lustre is reading a sparse file
1073 * via the LOV, and it _knows_ it's reading inside the file, it's just that
1074 * this stripe never got written at or beyond this stripe offset yet. */
1075 static void handle_short_read(int nob_read, obd_count page_count,
1076 struct brw_page **pga)
1077 {
1078 char *ptr;
1079 int i = 0;
1080
1081 /* skip bytes read OK */
1082 while (nob_read > 0) {
1083 LASSERT (page_count > 0);
1084
1085 if (pga[i]->count > nob_read) {
1086 /* EOF inside this page */
1087 ptr = kmap(pga[i]->pg) +
1088 (pga[i]->off & ~CFS_PAGE_MASK);
1089 memset(ptr + nob_read, 0, pga[i]->count - nob_read);
1090 kunmap(pga[i]->pg);
1091 page_count--;
1092 i++;
1093 break;
1094 }
1095
1096 nob_read -= pga[i]->count;
1097 page_count--;
1098 i++;
1099 }
1100
1101 /* zero remaining pages */
1102 while (page_count-- > 0) {
1103 ptr = kmap(pga[i]->pg) + (pga[i]->off & ~CFS_PAGE_MASK);
1104 memset(ptr, 0, pga[i]->count);
1105 kunmap(pga[i]->pg);
1106 i++;
1107 }
1108 }
1109
1110 static int check_write_rcs(struct ptlrpc_request *req,
1111 int requested_nob, int niocount,
1112 obd_count page_count, struct brw_page **pga)
1113 {
1114 int i;
1115 __u32 *remote_rcs;
1116
1117 remote_rcs = req_capsule_server_sized_get(&req->rq_pill, &RMF_RCS,
1118 sizeof(*remote_rcs) *
1119 niocount);
1120 if (remote_rcs == NULL) {
1121 CDEBUG(D_INFO, "Missing/short RC vector on BRW_WRITE reply\n");
1122 return(-EPROTO);
1123 }
1124
1125 /* return error if any niobuf was in error */
1126 for (i = 0; i < niocount; i++) {
1127 if ((int)remote_rcs[i] < 0)
1128 return(remote_rcs[i]);
1129
1130 if (remote_rcs[i] != 0) {
1131 CDEBUG(D_INFO, "rc[%d] invalid (%d) req %p\n",
1132 i, remote_rcs[i], req);
1133 return(-EPROTO);
1134 }
1135 }
1136
1137 if (req->rq_bulk->bd_nob_transferred != requested_nob) {
1138 CERROR("Unexpected # bytes transferred: %d (requested %d)\n",
1139 req->rq_bulk->bd_nob_transferred, requested_nob);
1140 return(-EPROTO);
1141 }
1142
1143 return (0);
1144 }
1145
1146 static inline int can_merge_pages(struct brw_page *p1, struct brw_page *p2)
1147 {
1148 if (p1->flag != p2->flag) {
1149 unsigned mask = ~(OBD_BRW_FROM_GRANT| OBD_BRW_NOCACHE|
1150 OBD_BRW_SYNC|OBD_BRW_ASYNC|OBD_BRW_NOQUOTA);
1151
1152 /* warn if we try to combine flags that we don't know to be
1153 * safe to combine */
1154 if (unlikely((p1->flag & mask) != (p2->flag & mask))) {
1155 CWARN("Saw flags 0x%x and 0x%x in the same brw, please "
1156 "report this at http://bugs.whamcloud.com/\n",
1157 p1->flag, p2->flag);
1158 }
1159 return 0;
1160 }
1161
1162 return (p1->off + p1->count == p2->off);
1163 }
1164
1165 static obd_count osc_checksum_bulk(int nob, obd_count pg_count,
1166 struct brw_page **pga, int opc,
1167 cksum_type_t cksum_type)
1168 {
1169 __u32 cksum;
1170 int i = 0;
1171 struct cfs_crypto_hash_desc *hdesc;
1172 unsigned int bufsize;
1173 int err;
1174 unsigned char cfs_alg = cksum_obd2cfs(cksum_type);
1175
1176 LASSERT(pg_count > 0);
1177
1178 hdesc = cfs_crypto_hash_init(cfs_alg, NULL, 0);
1179 if (IS_ERR(hdesc)) {
1180 CERROR("Unable to initialize checksum hash %s\n",
1181 cfs_crypto_hash_name(cfs_alg));
1182 return PTR_ERR(hdesc);
1183 }
1184
1185 while (nob > 0 && pg_count > 0) {
1186 int count = pga[i]->count > nob ? nob : pga[i]->count;
1187
1188 /* corrupt the data before we compute the checksum, to
1189 * simulate an OST->client data error */
1190 if (i == 0 && opc == OST_READ &&
1191 OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_RECEIVE)) {
1192 unsigned char *ptr = kmap(pga[i]->pg);
1193 int off = pga[i]->off & ~CFS_PAGE_MASK;
1194 memcpy(ptr + off, "bad1", min(4, nob));
1195 kunmap(pga[i]->pg);
1196 }
1197 cfs_crypto_hash_update_page(hdesc, pga[i]->pg,
1198 pga[i]->off & ~CFS_PAGE_MASK,
1199 count);
1200 LL_CDEBUG_PAGE(D_PAGE, pga[i]->pg, "off %d\n",
1201 (int)(pga[i]->off & ~CFS_PAGE_MASK));
1202
1203 nob -= pga[i]->count;
1204 pg_count--;
1205 i++;
1206 }
1207
1208 bufsize = 4;
1209 err = cfs_crypto_hash_final(hdesc, (unsigned char *)&cksum, &bufsize);
1210
1211 if (err)
1212 cfs_crypto_hash_final(hdesc, NULL, NULL);
1213
1214 /* For sending we only compute the wrong checksum instead
1215 * of corrupting the data so it is still correct on a redo */
1216 if (opc == OST_WRITE && OBD_FAIL_CHECK(OBD_FAIL_OSC_CHECKSUM_SEND))
1217 cksum++;
1218
1219 return cksum;
1220 }
1221
1222 static int osc_brw_prep_request(int cmd, struct client_obd *cli,struct obdo *oa,
1223 struct lov_stripe_md *lsm, obd_count page_count,
1224 struct brw_page **pga,
1225 struct ptlrpc_request **reqp,
1226 struct obd_capa *ocapa, int reserve,
1227 int resend)
1228 {
1229 struct ptlrpc_request *req;
1230 struct ptlrpc_bulk_desc *desc;
1231 struct ost_body *body;
1232 struct obd_ioobj *ioobj;
1233 struct niobuf_remote *niobuf;
1234 int niocount, i, requested_nob, opc, rc;
1235 struct osc_brw_async_args *aa;
1236 struct req_capsule *pill;
1237 struct brw_page *pg_prev;
1238
1239 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ))
1240 return -ENOMEM; /* Recoverable */
1241 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_BRW_PREP_REQ2))
1242 return -EINVAL; /* Fatal */
1243
1244 if ((cmd & OBD_BRW_WRITE) != 0) {
1245 opc = OST_WRITE;
1246 req = ptlrpc_request_alloc_pool(cli->cl_import,
1247 cli->cl_import->imp_rq_pool,
1248 &RQF_OST_BRW_WRITE);
1249 } else {
1250 opc = OST_READ;
1251 req = ptlrpc_request_alloc(cli->cl_import, &RQF_OST_BRW_READ);
1252 }
1253 if (req == NULL)
1254 return -ENOMEM;
1255
1256 for (niocount = i = 1; i < page_count; i++) {
1257 if (!can_merge_pages(pga[i - 1], pga[i]))
1258 niocount++;
1259 }
1260
1261 pill = &req->rq_pill;
1262 req_capsule_set_size(pill, &RMF_OBD_IOOBJ, RCL_CLIENT,
1263 sizeof(*ioobj));
1264 req_capsule_set_size(pill, &RMF_NIOBUF_REMOTE, RCL_CLIENT,
1265 niocount * sizeof(*niobuf));
1266 osc_set_capa_size(req, &RMF_CAPA1, ocapa);
1267
1268 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, opc);
1269 if (rc) {
1270 ptlrpc_request_free(req);
1271 return rc;
1272 }
1273 req->rq_request_portal = OST_IO_PORTAL; /* bug 7198 */
1274 ptlrpc_at_set_req_timeout(req);
1275 /* ask ptlrpc not to resend on EINPROGRESS since BRWs have their own
1276 * retry logic */
1277 req->rq_no_retry_einprogress = 1;
1278
1279 desc = ptlrpc_prep_bulk_imp(req, page_count,
1280 cli->cl_import->imp_connect_data.ocd_brw_size >> LNET_MTU_BITS,
1281 opc == OST_WRITE ? BULK_GET_SOURCE : BULK_PUT_SINK,
1282 OST_BULK_PORTAL);
1283
1284 if (desc == NULL)
1285 GOTO(out, rc = -ENOMEM);
1286 /* NB request now owns desc and will free it when it gets freed */
1287
1288 body = req_capsule_client_get(pill, &RMF_OST_BODY);
1289 ioobj = req_capsule_client_get(pill, &RMF_OBD_IOOBJ);
1290 niobuf = req_capsule_client_get(pill, &RMF_NIOBUF_REMOTE);
1291 LASSERT(body != NULL && ioobj != NULL && niobuf != NULL);
1292
1293 lustre_set_wire_obdo(&req->rq_import->imp_connect_data, &body->oa, oa);
1294
1295 obdo_to_ioobj(oa, ioobj);
1296 ioobj->ioo_bufcnt = niocount;
1297 /* The high bits of ioo_max_brw tells server _maximum_ number of bulks
1298 * that might be send for this request. The actual number is decided
1299 * when the RPC is finally sent in ptlrpc_register_bulk(). It sends
1300 * "max - 1" for old client compatibility sending "0", and also so the
1301 * the actual maximum is a power-of-two number, not one less. LU-1431 */
1302 ioobj_max_brw_set(ioobj, desc->bd_md_max_brw);
1303 osc_pack_capa(req, body, ocapa);
1304 LASSERT(page_count > 0);
1305 pg_prev = pga[0];
1306 for (requested_nob = i = 0; i < page_count; i++, niobuf++) {
1307 struct brw_page *pg = pga[i];
1308 int poff = pg->off & ~CFS_PAGE_MASK;
1309
1310 LASSERT(pg->count > 0);
1311 /* make sure there is no gap in the middle of page array */
1312 LASSERTF(page_count == 1 ||
1313 (ergo(i == 0, poff + pg->count == PAGE_CACHE_SIZE) &&
1314 ergo(i > 0 && i < page_count - 1,
1315 poff == 0 && pg->count == PAGE_CACHE_SIZE) &&
1316 ergo(i == page_count - 1, poff == 0)),
1317 "i: %d/%d pg: %p off: "LPU64", count: %u\n",
1318 i, page_count, pg, pg->off, pg->count);
1319 LASSERTF(i == 0 || pg->off > pg_prev->off,
1320 "i %d p_c %u pg %p [pri %lu ind %lu] off "LPU64
1321 " prev_pg %p [pri %lu ind %lu] off "LPU64"\n",
1322 i, page_count,
1323 pg->pg, page_private(pg->pg), pg->pg->index, pg->off,
1324 pg_prev->pg, page_private(pg_prev->pg),
1325 pg_prev->pg->index, pg_prev->off);
1326 LASSERT((pga[0]->flag & OBD_BRW_SRVLOCK) ==
1327 (pg->flag & OBD_BRW_SRVLOCK));
1328
1329 ptlrpc_prep_bulk_page_pin(desc, pg->pg, poff, pg->count);
1330 requested_nob += pg->count;
1331
1332 if (i > 0 && can_merge_pages(pg_prev, pg)) {
1333 niobuf--;
1334 niobuf->len += pg->count;
1335 } else {
1336 niobuf->offset = pg->off;
1337 niobuf->len = pg->count;
1338 niobuf->flags = pg->flag;
1339 }
1340 pg_prev = pg;
1341 }
1342
1343 LASSERTF((void *)(niobuf - niocount) ==
1344 req_capsule_client_get(&req->rq_pill, &RMF_NIOBUF_REMOTE),
1345 "want %p - real %p\n", req_capsule_client_get(&req->rq_pill,
1346 &RMF_NIOBUF_REMOTE), (void *)(niobuf - niocount));
1347
1348 osc_announce_cached(cli, &body->oa, opc == OST_WRITE ? requested_nob:0);
1349 if (resend) {
1350 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1351 body->oa.o_valid |= OBD_MD_FLFLAGS;
1352 body->oa.o_flags = 0;
1353 }
1354 body->oa.o_flags |= OBD_FL_RECOV_RESEND;
1355 }
1356
1357 if (osc_should_shrink_grant(cli))
1358 osc_shrink_grant_local(cli, &body->oa);
1359
1360 /* size[REQ_REC_OFF] still sizeof (*body) */
1361 if (opc == OST_WRITE) {
1362 if (cli->cl_checksum &&
1363 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1364 /* store cl_cksum_type in a local variable since
1365 * it can be changed via lprocfs */
1366 cksum_type_t cksum_type = cli->cl_cksum_type;
1367
1368 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0) {
1369 oa->o_flags &= OBD_FL_LOCAL_MASK;
1370 body->oa.o_flags = 0;
1371 }
1372 body->oa.o_flags |= cksum_type_pack(cksum_type);
1373 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1374 body->oa.o_cksum = osc_checksum_bulk(requested_nob,
1375 page_count, pga,
1376 OST_WRITE,
1377 cksum_type);
1378 CDEBUG(D_PAGE, "checksum at write origin: %x\n",
1379 body->oa.o_cksum);
1380 /* save this in 'oa', too, for later checking */
1381 oa->o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1382 oa->o_flags |= cksum_type_pack(cksum_type);
1383 } else {
1384 /* clear out the checksum flag, in case this is a
1385 * resend but cl_checksum is no longer set. b=11238 */
1386 oa->o_valid &= ~OBD_MD_FLCKSUM;
1387 }
1388 oa->o_cksum = body->oa.o_cksum;
1389 /* 1 RC per niobuf */
1390 req_capsule_set_size(pill, &RMF_RCS, RCL_SERVER,
1391 sizeof(__u32) * niocount);
1392 } else {
1393 if (cli->cl_checksum &&
1394 !sptlrpc_flavor_has_bulk(&req->rq_flvr)) {
1395 if ((body->oa.o_valid & OBD_MD_FLFLAGS) == 0)
1396 body->oa.o_flags = 0;
1397 body->oa.o_flags |= cksum_type_pack(cli->cl_cksum_type);
1398 body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
1399 }
1400 }
1401 ptlrpc_request_set_replen(req);
1402
1403 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
1404 aa = ptlrpc_req_async_args(req);
1405 aa->aa_oa = oa;
1406 aa->aa_requested_nob = requested_nob;
1407 aa->aa_nio_count = niocount;
1408 aa->aa_page_count = page_count;
1409 aa->aa_resends = 0;
1410 aa->aa_ppga = pga;
1411 aa->aa_cli = cli;
1412 INIT_LIST_HEAD(&aa->aa_oaps);
1413 if (ocapa && reserve)
1414 aa->aa_ocapa = capa_get(ocapa);
1415
1416 *reqp = req;
1417 return 0;
1418
1419 out:
1420 ptlrpc_req_finished(req);
1421 return rc;
1422 }
1423
1424 static int check_write_checksum(struct obdo *oa, const lnet_process_id_t *peer,
1425 __u32 client_cksum, __u32 server_cksum, int nob,
1426 obd_count page_count, struct brw_page **pga,
1427 cksum_type_t client_cksum_type)
1428 {
1429 __u32 new_cksum;
1430 char *msg;
1431 cksum_type_t cksum_type;
1432
1433 if (server_cksum == client_cksum) {
1434 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1435 return 0;
1436 }
1437
1438 cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
1439 oa->o_flags : 0);
1440 new_cksum = osc_checksum_bulk(nob, page_count, pga, OST_WRITE,
1441 cksum_type);
1442
1443 if (cksum_type != client_cksum_type)
1444 msg = "the server did not use the checksum type specified in "
1445 "the original request - likely a protocol problem";
1446 else if (new_cksum == server_cksum)
1447 msg = "changed on the client after we checksummed it - "
1448 "likely false positive due to mmap IO (bug 11742)";
1449 else if (new_cksum == client_cksum)
1450 msg = "changed in transit before arrival at OST";
1451 else
1452 msg = "changed in transit AND doesn't match the original - "
1453 "likely false positive due to mmap IO (bug 11742)";
1454
1455 LCONSOLE_ERROR_MSG(0x132, "BAD WRITE CHECKSUM: %s: from %s inode "DFID
1456 " object "DOSTID" extent ["LPU64"-"LPU64"]\n",
1457 msg, libcfs_nid2str(peer->nid),
1458 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
1459 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
1460 oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
1461 POSTID(&oa->o_oi), pga[0]->off,
1462 pga[page_count-1]->off + pga[page_count-1]->count - 1);
1463 CERROR("original client csum %x (type %x), server csum %x (type %x), "
1464 "client csum now %x\n", client_cksum, client_cksum_type,
1465 server_cksum, cksum_type, new_cksum);
1466 return 1;
1467 }
1468
1469 /* Note rc enters this function as number of bytes transferred */
1470 static int osc_brw_fini_request(struct ptlrpc_request *req, int rc)
1471 {
1472 struct osc_brw_async_args *aa = (void *)&req->rq_async_args;
1473 const lnet_process_id_t *peer =
1474 &req->rq_import->imp_connection->c_peer;
1475 struct client_obd *cli = aa->aa_cli;
1476 struct ost_body *body;
1477 __u32 client_cksum = 0;
1478
1479 if (rc < 0 && rc != -EDQUOT) {
1480 DEBUG_REQ(D_INFO, req, "Failed request with rc = %d\n", rc);
1481 return rc;
1482 }
1483
1484 LASSERTF(req->rq_repmsg != NULL, "rc = %d\n", rc);
1485 body = req_capsule_server_get(&req->rq_pill, &RMF_OST_BODY);
1486 if (body == NULL) {
1487 DEBUG_REQ(D_INFO, req, "Can't unpack body\n");
1488 return -EPROTO;
1489 }
1490
1491 /* set/clear over quota flag for a uid/gid */
1492 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE &&
1493 body->oa.o_valid & (OBD_MD_FLUSRQUOTA | OBD_MD_FLGRPQUOTA)) {
1494 unsigned int qid[MAXQUOTAS] = { body->oa.o_uid, body->oa.o_gid };
1495
1496 CDEBUG(D_QUOTA, "setdq for [%u %u] with valid "LPX64", flags %x\n",
1497 body->oa.o_uid, body->oa.o_gid, body->oa.o_valid,
1498 body->oa.o_flags);
1499 osc_quota_setdq(cli, qid, body->oa.o_valid, body->oa.o_flags);
1500 }
1501
1502 osc_update_grant(cli, body);
1503
1504 if (rc < 0)
1505 return rc;
1506
1507 if (aa->aa_oa->o_valid & OBD_MD_FLCKSUM)
1508 client_cksum = aa->aa_oa->o_cksum; /* save for later */
1509
1510 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE) {
1511 if (rc > 0) {
1512 CERROR("Unexpected +ve rc %d\n", rc);
1513 return -EPROTO;
1514 }
1515 LASSERT(req->rq_bulk->bd_nob == aa->aa_requested_nob);
1516
1517 if (sptlrpc_cli_unwrap_bulk_write(req, req->rq_bulk))
1518 return -EAGAIN;
1519
1520 if ((aa->aa_oa->o_valid & OBD_MD_FLCKSUM) && client_cksum &&
1521 check_write_checksum(&body->oa, peer, client_cksum,
1522 body->oa.o_cksum, aa->aa_requested_nob,
1523 aa->aa_page_count, aa->aa_ppga,
1524 cksum_type_unpack(aa->aa_oa->o_flags)))
1525 return -EAGAIN;
1526
1527 rc = check_write_rcs(req, aa->aa_requested_nob,aa->aa_nio_count,
1528 aa->aa_page_count, aa->aa_ppga);
1529 GOTO(out, rc);
1530 }
1531
1532 /* The rest of this function executes only for OST_READs */
1533
1534 /* if unwrap_bulk failed, return -EAGAIN to retry */
1535 rc = sptlrpc_cli_unwrap_bulk_read(req, req->rq_bulk, rc);
1536 if (rc < 0)
1537 GOTO(out, rc = -EAGAIN);
1538
1539 if (rc > aa->aa_requested_nob) {
1540 CERROR("Unexpected rc %d (%d requested)\n", rc,
1541 aa->aa_requested_nob);
1542 return -EPROTO;
1543 }
1544
1545 if (rc != req->rq_bulk->bd_nob_transferred) {
1546 CERROR ("Unexpected rc %d (%d transferred)\n",
1547 rc, req->rq_bulk->bd_nob_transferred);
1548 return (-EPROTO);
1549 }
1550
1551 if (rc < aa->aa_requested_nob)
1552 handle_short_read(rc, aa->aa_page_count, aa->aa_ppga);
1553
1554 if (body->oa.o_valid & OBD_MD_FLCKSUM) {
1555 static int cksum_counter;
1556 __u32 server_cksum = body->oa.o_cksum;
1557 char *via;
1558 char *router;
1559 cksum_type_t cksum_type;
1560
1561 cksum_type = cksum_type_unpack(body->oa.o_valid &OBD_MD_FLFLAGS?
1562 body->oa.o_flags : 0);
1563 client_cksum = osc_checksum_bulk(rc, aa->aa_page_count,
1564 aa->aa_ppga, OST_READ,
1565 cksum_type);
1566
1567 if (peer->nid == req->rq_bulk->bd_sender) {
1568 via = router = "";
1569 } else {
1570 via = " via ";
1571 router = libcfs_nid2str(req->rq_bulk->bd_sender);
1572 }
1573
1574 if (server_cksum == ~0 && rc > 0) {
1575 CERROR("Protocol error: server %s set the 'checksum' "
1576 "bit, but didn't send a checksum. Not fatal, "
1577 "but please notify on http://bugs.whamcloud.com/\n",
1578 libcfs_nid2str(peer->nid));
1579 } else if (server_cksum != client_cksum) {
1580 LCONSOLE_ERROR_MSG(0x133, "%s: BAD READ CHECKSUM: from "
1581 "%s%s%s inode "DFID" object "DOSTID
1582 " extent ["LPU64"-"LPU64"]\n",
1583 req->rq_import->imp_obd->obd_name,
1584 libcfs_nid2str(peer->nid),
1585 via, router,
1586 body->oa.o_valid & OBD_MD_FLFID ?
1587 body->oa.o_parent_seq : (__u64)0,
1588 body->oa.o_valid & OBD_MD_FLFID ?
1589 body->oa.o_parent_oid : 0,
1590 body->oa.o_valid & OBD_MD_FLFID ?
1591 body->oa.o_parent_ver : 0,
1592 POSTID(&body->oa.o_oi),
1593 aa->aa_ppga[0]->off,
1594 aa->aa_ppga[aa->aa_page_count-1]->off +
1595 aa->aa_ppga[aa->aa_page_count-1]->count -
1596 1);
1597 CERROR("client %x, server %x, cksum_type %x\n",
1598 client_cksum, server_cksum, cksum_type);
1599 cksum_counter = 0;
1600 aa->aa_oa->o_cksum = client_cksum;
1601 rc = -EAGAIN;
1602 } else {
1603 cksum_counter++;
1604 CDEBUG(D_PAGE, "checksum %x confirmed\n", client_cksum);
1605 rc = 0;
1606 }
1607 } else if (unlikely(client_cksum)) {
1608 static int cksum_missed;
1609
1610 cksum_missed++;
1611 if ((cksum_missed & (-cksum_missed)) == cksum_missed)
1612 CERROR("Checksum %u requested from %s but not sent\n",
1613 cksum_missed, libcfs_nid2str(peer->nid));
1614 } else {
1615 rc = 0;
1616 }
1617 out:
1618 if (rc >= 0)
1619 lustre_get_wire_obdo(&req->rq_import->imp_connect_data,
1620 aa->aa_oa, &body->oa);
1621
1622 return rc;
1623 }
1624
1625 static int osc_brw_internal(int cmd, struct obd_export *exp, struct obdo *oa,
1626 struct lov_stripe_md *lsm,
1627 obd_count page_count, struct brw_page **pga,
1628 struct obd_capa *ocapa)
1629 {
1630 struct ptlrpc_request *req;
1631 int rc;
1632 wait_queue_head_t waitq;
1633 int generation, resends = 0;
1634 struct l_wait_info lwi;
1635
1636 init_waitqueue_head(&waitq);
1637 generation = exp->exp_obd->u.cli.cl_import->imp_generation;
1638
1639 restart_bulk:
1640 rc = osc_brw_prep_request(cmd, &exp->exp_obd->u.cli, oa, lsm,
1641 page_count, pga, &req, ocapa, 0, resends);
1642 if (rc != 0)
1643 return (rc);
1644
1645 if (resends) {
1646 req->rq_generation_set = 1;
1647 req->rq_import_generation = generation;
1648 req->rq_sent = cfs_time_current_sec() + resends;
1649 }
1650
1651 rc = ptlrpc_queue_wait(req);
1652
1653 if (rc == -ETIMEDOUT && req->rq_resend) {
1654 DEBUG_REQ(D_HA, req, "BULK TIMEOUT");
1655 ptlrpc_req_finished(req);
1656 goto restart_bulk;
1657 }
1658
1659 rc = osc_brw_fini_request(req, rc);
1660
1661 ptlrpc_req_finished(req);
1662 /* When server return -EINPROGRESS, client should always retry
1663 * regardless of the number of times the bulk was resent already.*/
1664 if (osc_recoverable_error(rc)) {
1665 resends++;
1666 if (rc != -EINPROGRESS &&
1667 !client_should_resend(resends, &exp->exp_obd->u.cli)) {
1668 CERROR("%s: too many resend retries for object: "
1669 ""DOSTID", rc = %d.\n", exp->exp_obd->obd_name,
1670 POSTID(&oa->o_oi), rc);
1671 goto out;
1672 }
1673 if (generation !=
1674 exp->exp_obd->u.cli.cl_import->imp_generation) {
1675 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1676 ""DOSTID", rc = %d.\n", exp->exp_obd->obd_name,
1677 POSTID(&oa->o_oi), rc);
1678 goto out;
1679 }
1680
1681 lwi = LWI_TIMEOUT_INTR(cfs_time_seconds(resends), NULL, NULL,
1682 NULL);
1683 l_wait_event(waitq, 0, &lwi);
1684
1685 goto restart_bulk;
1686 }
1687 out:
1688 if (rc == -EAGAIN || rc == -EINPROGRESS)
1689 rc = -EIO;
1690 return rc;
1691 }
1692
1693 static int osc_brw_redo_request(struct ptlrpc_request *request,
1694 struct osc_brw_async_args *aa, int rc)
1695 {
1696 struct ptlrpc_request *new_req;
1697 struct osc_brw_async_args *new_aa;
1698 struct osc_async_page *oap;
1699
1700 DEBUG_REQ(rc == -EINPROGRESS ? D_RPCTRACE : D_ERROR, request,
1701 "redo for recoverable error %d", rc);
1702
1703 rc = osc_brw_prep_request(lustre_msg_get_opc(request->rq_reqmsg) ==
1704 OST_WRITE ? OBD_BRW_WRITE :OBD_BRW_READ,
1705 aa->aa_cli, aa->aa_oa,
1706 NULL /* lsm unused by osc currently */,
1707 aa->aa_page_count, aa->aa_ppga,
1708 &new_req, aa->aa_ocapa, 0, 1);
1709 if (rc)
1710 return rc;
1711
1712 list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
1713 if (oap->oap_request != NULL) {
1714 LASSERTF(request == oap->oap_request,
1715 "request %p != oap_request %p\n",
1716 request, oap->oap_request);
1717 if (oap->oap_interrupted) {
1718 ptlrpc_req_finished(new_req);
1719 return -EINTR;
1720 }
1721 }
1722 }
1723 /* New request takes over pga and oaps from old request.
1724 * Note that copying a list_head doesn't work, need to move it... */
1725 aa->aa_resends++;
1726 new_req->rq_interpret_reply = request->rq_interpret_reply;
1727 new_req->rq_async_args = request->rq_async_args;
1728 /* cap resend delay to the current request timeout, this is similar to
1729 * what ptlrpc does (see after_reply()) */
1730 if (aa->aa_resends > new_req->rq_timeout)
1731 new_req->rq_sent = cfs_time_current_sec() + new_req->rq_timeout;
1732 else
1733 new_req->rq_sent = cfs_time_current_sec() + aa->aa_resends;
1734 new_req->rq_generation_set = 1;
1735 new_req->rq_import_generation = request->rq_import_generation;
1736
1737 new_aa = ptlrpc_req_async_args(new_req);
1738
1739 INIT_LIST_HEAD(&new_aa->aa_oaps);
1740 list_splice_init(&aa->aa_oaps, &new_aa->aa_oaps);
1741 INIT_LIST_HEAD(&new_aa->aa_exts);
1742 list_splice_init(&aa->aa_exts, &new_aa->aa_exts);
1743 new_aa->aa_resends = aa->aa_resends;
1744
1745 list_for_each_entry(oap, &new_aa->aa_oaps, oap_rpc_item) {
1746 if (oap->oap_request) {
1747 ptlrpc_req_finished(oap->oap_request);
1748 oap->oap_request = ptlrpc_request_addref(new_req);
1749 }
1750 }
1751
1752 new_aa->aa_ocapa = aa->aa_ocapa;
1753 aa->aa_ocapa = NULL;
1754
1755 /* XXX: This code will run into problem if we're going to support
1756 * to add a series of BRW RPCs into a self-defined ptlrpc_request_set
1757 * and wait for all of them to be finished. We should inherit request
1758 * set from old request. */
1759 ptlrpcd_add_req(new_req, PDL_POLICY_SAME, -1);
1760
1761 DEBUG_REQ(D_INFO, new_req, "new request");
1762 return 0;
1763 }
1764
1765 /*
1766 * ugh, we want disk allocation on the target to happen in offset order. we'll
1767 * follow sedgewicks advice and stick to the dead simple shellsort -- it'll do
1768 * fine for our small page arrays and doesn't require allocation. its an
1769 * insertion sort that swaps elements that are strides apart, shrinking the
1770 * stride down until its '1' and the array is sorted.
1771 */
1772 static void sort_brw_pages(struct brw_page **array, int num)
1773 {
1774 int stride, i, j;
1775 struct brw_page *tmp;
1776
1777 if (num == 1)
1778 return;
1779 for (stride = 1; stride < num ; stride = (stride * 3) + 1)
1780 ;
1781
1782 do {
1783 stride /= 3;
1784 for (i = stride ; i < num ; i++) {
1785 tmp = array[i];
1786 j = i;
1787 while (j >= stride && array[j - stride]->off > tmp->off) {
1788 array[j] = array[j - stride];
1789 j -= stride;
1790 }
1791 array[j] = tmp;
1792 }
1793 } while (stride > 1);
1794 }
1795
1796 static obd_count max_unfragmented_pages(struct brw_page **pg, obd_count pages)
1797 {
1798 int count = 1;
1799 int offset;
1800 int i = 0;
1801
1802 LASSERT (pages > 0);
1803 offset = pg[i]->off & ~CFS_PAGE_MASK;
1804
1805 for (;;) {
1806 pages--;
1807 if (pages == 0) /* that's all */
1808 return count;
1809
1810 if (offset + pg[i]->count < PAGE_CACHE_SIZE)
1811 return count; /* doesn't end on page boundary */
1812
1813 i++;
1814 offset = pg[i]->off & ~CFS_PAGE_MASK;
1815 if (offset != 0) /* doesn't start on page boundary */
1816 return count;
1817
1818 count++;
1819 }
1820 }
1821
1822 static struct brw_page **osc_build_ppga(struct brw_page *pga, obd_count count)
1823 {
1824 struct brw_page **ppga;
1825 int i;
1826
1827 OBD_ALLOC(ppga, sizeof(*ppga) * count);
1828 if (ppga == NULL)
1829 return NULL;
1830
1831 for (i = 0; i < count; i++)
1832 ppga[i] = pga + i;
1833 return ppga;
1834 }
1835
1836 static void osc_release_ppga(struct brw_page **ppga, obd_count count)
1837 {
1838 LASSERT(ppga != NULL);
1839 OBD_FREE(ppga, sizeof(*ppga) * count);
1840 }
1841
1842 static int osc_brw(int cmd, struct obd_export *exp, struct obd_info *oinfo,
1843 obd_count page_count, struct brw_page *pga,
1844 struct obd_trans_info *oti)
1845 {
1846 struct obdo *saved_oa = NULL;
1847 struct brw_page **ppga, **orig;
1848 struct obd_import *imp = class_exp2cliimp(exp);
1849 struct client_obd *cli;
1850 int rc, page_count_orig;
1851
1852 LASSERT((imp != NULL) && (imp->imp_obd != NULL));
1853 cli = &imp->imp_obd->u.cli;
1854
1855 if (cmd & OBD_BRW_CHECK) {
1856 /* The caller just wants to know if there's a chance that this
1857 * I/O can succeed */
1858
1859 if (imp->imp_invalid)
1860 return -EIO;
1861 return 0;
1862 }
1863
1864 /* test_brw with a failed create can trip this, maybe others. */
1865 LASSERT(cli->cl_max_pages_per_rpc);
1866
1867 rc = 0;
1868
1869 orig = ppga = osc_build_ppga(pga, page_count);
1870 if (ppga == NULL)
1871 return -ENOMEM;
1872 page_count_orig = page_count;
1873
1874 sort_brw_pages(ppga, page_count);
1875 while (page_count) {
1876 obd_count pages_per_brw;
1877
1878 if (page_count > cli->cl_max_pages_per_rpc)
1879 pages_per_brw = cli->cl_max_pages_per_rpc;
1880 else
1881 pages_per_brw = page_count;
1882
1883 pages_per_brw = max_unfragmented_pages(ppga, pages_per_brw);
1884
1885 if (saved_oa != NULL) {
1886 /* restore previously saved oa */
1887 *oinfo->oi_oa = *saved_oa;
1888 } else if (page_count > pages_per_brw) {
1889 /* save a copy of oa (brw will clobber it) */
1890 OBDO_ALLOC(saved_oa);
1891 if (saved_oa == NULL)
1892 GOTO(out, rc = -ENOMEM);
1893 *saved_oa = *oinfo->oi_oa;
1894 }
1895
1896 rc = osc_brw_internal(cmd, exp, oinfo->oi_oa, oinfo->oi_md,
1897 pages_per_brw, ppga, oinfo->oi_capa);
1898
1899 if (rc != 0)
1900 break;
1901
1902 page_count -= pages_per_brw;
1903 ppga += pages_per_brw;
1904 }
1905
1906 out:
1907 osc_release_ppga(orig, page_count_orig);
1908
1909 if (saved_oa != NULL)
1910 OBDO_FREE(saved_oa);
1911
1912 return rc;
1913 }
1914
1915 static int brw_interpret(const struct lu_env *env,
1916 struct ptlrpc_request *req, void *data, int rc)
1917 {
1918 struct osc_brw_async_args *aa = data;
1919 struct osc_extent *ext;
1920 struct osc_extent *tmp;
1921 struct cl_object *obj = NULL;
1922 struct client_obd *cli = aa->aa_cli;
1923
1924 rc = osc_brw_fini_request(req, rc);
1925 CDEBUG(D_INODE, "request %p aa %p rc %d\n", req, aa, rc);
1926 /* When server return -EINPROGRESS, client should always retry
1927 * regardless of the number of times the bulk was resent already. */
1928 if (osc_recoverable_error(rc)) {
1929 if (req->rq_import_generation !=
1930 req->rq_import->imp_generation) {
1931 CDEBUG(D_HA, "%s: resend cross eviction for object: "
1932 ""DOSTID", rc = %d.\n",
1933 req->rq_import->imp_obd->obd_name,
1934 POSTID(&aa->aa_oa->o_oi), rc);
1935 } else if (rc == -EINPROGRESS ||
1936 client_should_resend(aa->aa_resends, aa->aa_cli)) {
1937 rc = osc_brw_redo_request(req, aa, rc);
1938 } else {
1939 CERROR("%s: too many resent retries for object: "
1940 ""LPU64":"LPU64", rc = %d.\n",
1941 req->rq_import->imp_obd->obd_name,
1942 POSTID(&aa->aa_oa->o_oi), rc);
1943 }
1944
1945 if (rc == 0)
1946 return 0;
1947 else if (rc == -EAGAIN || rc == -EINPROGRESS)
1948 rc = -EIO;
1949 }
1950
1951 if (aa->aa_ocapa) {
1952 capa_put(aa->aa_ocapa);
1953 aa->aa_ocapa = NULL;
1954 }
1955
1956 list_for_each_entry_safe(ext, tmp, &aa->aa_exts, oe_link) {
1957 if (obj == NULL && rc == 0) {
1958 obj = osc2cl(ext->oe_obj);
1959 cl_object_get(obj);
1960 }
1961
1962 list_del_init(&ext->oe_link);
1963 osc_extent_finish(env, ext, 1, rc);
1964 }
1965 LASSERT(list_empty(&aa->aa_exts));
1966 LASSERT(list_empty(&aa->aa_oaps));
1967
1968 if (obj != NULL) {
1969 struct obdo *oa = aa->aa_oa;
1970 struct cl_attr *attr = &osc_env_info(env)->oti_attr;
1971 unsigned long valid = 0;
1972
1973 LASSERT(rc == 0);
1974 if (oa->o_valid & OBD_MD_FLBLOCKS) {
1975 attr->cat_blocks = oa->o_blocks;
1976 valid |= CAT_BLOCKS;
1977 }
1978 if (oa->o_valid & OBD_MD_FLMTIME) {
1979 attr->cat_mtime = oa->o_mtime;
1980 valid |= CAT_MTIME;
1981 }
1982 if (oa->o_valid & OBD_MD_FLATIME) {
1983 attr->cat_atime = oa->o_atime;
1984 valid |= CAT_ATIME;
1985 }
1986 if (oa->o_valid & OBD_MD_FLCTIME) {
1987 attr->cat_ctime = oa->o_ctime;
1988 valid |= CAT_CTIME;
1989 }
1990 if (valid != 0) {
1991 cl_object_attr_lock(obj);
1992 cl_object_attr_set(env, obj, attr, valid);
1993 cl_object_attr_unlock(obj);
1994 }
1995 cl_object_put(env, obj);
1996 }
1997 OBDO_FREE(aa->aa_oa);
1998
1999 cl_req_completion(env, aa->aa_clerq, rc < 0 ? rc :
2000 req->rq_bulk->bd_nob_transferred);
2001 osc_release_ppga(aa->aa_ppga, aa->aa_page_count);
2002 ptlrpc_lprocfs_brw(req, req->rq_bulk->bd_nob_transferred);
2003
2004 client_obd_list_lock(&cli->cl_loi_list_lock);
2005 /* We need to decrement before osc_ap_completion->osc_wake_cache_waiters
2006 * is called so we know whether to go to sync BRWs or wait for more
2007 * RPCs to complete */
2008 if (lustre_msg_get_opc(req->rq_reqmsg) == OST_WRITE)
2009 cli->cl_w_in_flight--;
2010 else
2011 cli->cl_r_in_flight--;
2012 osc_wake_cache_waiters(cli);
2013 client_obd_list_unlock(&cli->cl_loi_list_lock);
2014
2015 osc_io_unplug(env, cli, NULL, PDL_POLICY_SAME);
2016 return rc;
2017 }
2018
2019 /**
2020 * Build an RPC by the list of extent @ext_list. The caller must ensure
2021 * that the total pages in this list are NOT over max pages per RPC.
2022 * Extents in the list must be in OES_RPC state.
2023 */
2024 int osc_build_rpc(const struct lu_env *env, struct client_obd *cli,
2025 struct list_head *ext_list, int cmd, pdl_policy_t pol)
2026 {
2027 struct ptlrpc_request *req = NULL;
2028 struct osc_extent *ext;
2029 struct brw_page **pga = NULL;
2030 struct osc_brw_async_args *aa = NULL;
2031 struct obdo *oa = NULL;
2032 struct osc_async_page *oap;
2033 struct osc_async_page *tmp;
2034 struct cl_req *clerq = NULL;
2035 enum cl_req_type crt = (cmd & OBD_BRW_WRITE) ? CRT_WRITE :
2036 CRT_READ;
2037 struct ldlm_lock *lock = NULL;
2038 struct cl_req_attr *crattr = NULL;
2039 obd_off starting_offset = OBD_OBJECT_EOF;
2040 obd_off ending_offset = 0;
2041 int mpflag = 0;
2042 int mem_tight = 0;
2043 int page_count = 0;
2044 int i;
2045 int rc;
2046 LIST_HEAD(rpc_list);
2047
2048 LASSERT(!list_empty(ext_list));
2049
2050 /* add pages into rpc_list to build BRW rpc */
2051 list_for_each_entry(ext, ext_list, oe_link) {
2052 LASSERT(ext->oe_state == OES_RPC);
2053 mem_tight |= ext->oe_memalloc;
2054 list_for_each_entry(oap, &ext->oe_pages, oap_pending_item) {
2055 ++page_count;
2056 list_add_tail(&oap->oap_rpc_item, &rpc_list);
2057 if (starting_offset > oap->oap_obj_off)
2058 starting_offset = oap->oap_obj_off;
2059 else
2060 LASSERT(oap->oap_page_off == 0);
2061 if (ending_offset < oap->oap_obj_off + oap->oap_count)
2062 ending_offset = oap->oap_obj_off +
2063 oap->oap_count;
2064 else
2065 LASSERT(oap->oap_page_off + oap->oap_count ==
2066 PAGE_CACHE_SIZE);
2067 }
2068 }
2069
2070 if (mem_tight)
2071 mpflag = cfs_memory_pressure_get_and_set();
2072
2073 OBD_ALLOC(crattr, sizeof(*crattr));
2074 if (crattr == NULL)
2075 GOTO(out, rc = -ENOMEM);
2076
2077 OBD_ALLOC(pga, sizeof(*pga) * page_count);
2078 if (pga == NULL)
2079 GOTO(out, rc = -ENOMEM);
2080
2081 OBDO_ALLOC(oa);
2082 if (oa == NULL)
2083 GOTO(out, rc = -ENOMEM);
2084
2085 i = 0;
2086 list_for_each_entry(oap, &rpc_list, oap_rpc_item) {
2087 struct cl_page *page = oap2cl_page(oap);
2088 if (clerq == NULL) {
2089 clerq = cl_req_alloc(env, page, crt,
2090 1 /* only 1-object rpcs for now */);
2091 if (IS_ERR(clerq))
2092 GOTO(out, rc = PTR_ERR(clerq));
2093 lock = oap->oap_ldlm_lock;
2094 }
2095 if (mem_tight)
2096 oap->oap_brw_flags |= OBD_BRW_MEMALLOC;
2097 pga[i] = &oap->oap_brw_page;
2098 pga[i]->off = oap->oap_obj_off + oap->oap_page_off;
2099 CDEBUG(0, "put page %p index %lu oap %p flg %x to pga\n",
2100 pga[i]->pg, page_index(oap->oap_page), oap,
2101 pga[i]->flag);
2102 i++;
2103 cl_req_page_add(env, clerq, page);
2104 }
2105
2106 /* always get the data for the obdo for the rpc */
2107 LASSERT(clerq != NULL);
2108 crattr->cra_oa = oa;
2109 cl_req_attr_set(env, clerq, crattr, ~0ULL);
2110 if (lock) {
2111 oa->o_handle = lock->l_remote_handle;
2112 oa->o_valid |= OBD_MD_FLHANDLE;
2113 }
2114
2115 rc = cl_req_prep(env, clerq);
2116 if (rc != 0) {
2117 CERROR("cl_req_prep failed: %d\n", rc);
2118 GOTO(out, rc);
2119 }
2120
2121 sort_brw_pages(pga, page_count);
2122 rc = osc_brw_prep_request(cmd, cli, oa, NULL, page_count,
2123 pga, &req, crattr->cra_capa, 1, 0);
2124 if (rc != 0) {
2125 CERROR("prep_req failed: %d\n", rc);
2126 GOTO(out, rc);
2127 }
2128
2129 req->rq_interpret_reply = brw_interpret;
2130
2131 if (mem_tight != 0)
2132 req->rq_memalloc = 1;
2133
2134 /* Need to update the timestamps after the request is built in case
2135 * we race with setattr (locally or in queue at OST). If OST gets
2136 * later setattr before earlier BRW (as determined by the request xid),
2137 * the OST will not use BRW timestamps. Sadly, there is no obvious
2138 * way to do this in a single call. bug 10150 */
2139 cl_req_attr_set(env, clerq, crattr,
2140 OBD_MD_FLMTIME|OBD_MD_FLCTIME|OBD_MD_FLATIME);
2141
2142 lustre_msg_set_jobid(req->rq_reqmsg, crattr->cra_jobid);
2143
2144 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
2145 aa = ptlrpc_req_async_args(req);
2146 INIT_LIST_HEAD(&aa->aa_oaps);
2147 list_splice_init(&rpc_list, &aa->aa_oaps);
2148 INIT_LIST_HEAD(&aa->aa_exts);
2149 list_splice_init(ext_list, &aa->aa_exts);
2150 aa->aa_clerq = clerq;
2151
2152 /* queued sync pages can be torn down while the pages
2153 * were between the pending list and the rpc */
2154 tmp = NULL;
2155 list_for_each_entry(oap, &aa->aa_oaps, oap_rpc_item) {
2156 /* only one oap gets a request reference */
2157 if (tmp == NULL)
2158 tmp = oap;
2159 if (oap->oap_interrupted && !req->rq_intr) {
2160 CDEBUG(D_INODE, "oap %p in req %p interrupted\n",
2161 oap, req);
2162 ptlrpc_mark_interrupted(req);
2163 }
2164 }
2165 if (tmp != NULL)
2166 tmp->oap_request = ptlrpc_request_addref(req);
2167
2168 client_obd_list_lock(&cli->cl_loi_list_lock);
2169 starting_offset >>= PAGE_CACHE_SHIFT;
2170 if (cmd == OBD_BRW_READ) {
2171 cli->cl_r_in_flight++;
2172 lprocfs_oh_tally_log2(&cli->cl_read_page_hist, page_count);
2173 lprocfs_oh_tally(&cli->cl_read_rpc_hist, cli->cl_r_in_flight);
2174 lprocfs_oh_tally_log2(&cli->cl_read_offset_hist,
2175 starting_offset + 1);
2176 } else {
2177 cli->cl_w_in_flight++;
2178 lprocfs_oh_tally_log2(&cli->cl_write_page_hist, page_count);
2179 lprocfs_oh_tally(&cli->cl_write_rpc_hist, cli->cl_w_in_flight);
2180 lprocfs_oh_tally_log2(&cli->cl_write_offset_hist,
2181 starting_offset + 1);
2182 }
2183 client_obd_list_unlock(&cli->cl_loi_list_lock);
2184
2185 DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %dr/%dw in flight",
2186 page_count, aa, cli->cl_r_in_flight,
2187 cli->cl_w_in_flight);
2188
2189 /* XXX: Maybe the caller can check the RPC bulk descriptor to
2190 * see which CPU/NUMA node the majority of pages were allocated
2191 * on, and try to assign the async RPC to the CPU core
2192 * (PDL_POLICY_PREFERRED) to reduce cross-CPU memory traffic.
2193 *
2194 * But on the other hand, we expect that multiple ptlrpcd
2195 * threads and the initial write sponsor can run in parallel,
2196 * especially when data checksum is enabled, which is CPU-bound
2197 * operation and single ptlrpcd thread cannot process in time.
2198 * So more ptlrpcd threads sharing BRW load
2199 * (with PDL_POLICY_ROUND) seems better.
2200 */
2201 ptlrpcd_add_req(req, pol, -1);
2202 rc = 0;
2203
2204 out:
2205 if (mem_tight != 0)
2206 cfs_memory_pressure_restore(mpflag);
2207
2208 if (crattr != NULL) {
2209 capa_put(crattr->cra_capa);
2210 OBD_FREE(crattr, sizeof(*crattr));
2211 }
2212
2213 if (rc != 0) {
2214 LASSERT(req == NULL);
2215
2216 if (oa)
2217 OBDO_FREE(oa);
2218 if (pga)
2219 OBD_FREE(pga, sizeof(*pga) * page_count);
2220 /* this should happen rarely and is pretty bad, it makes the
2221 * pending list not follow the dirty order */
2222 while (!list_empty(ext_list)) {
2223 ext = list_entry(ext_list->next, struct osc_extent,
2224 oe_link);
2225 list_del_init(&ext->oe_link);
2226 osc_extent_finish(env, ext, 0, rc);
2227 }
2228 if (clerq && !IS_ERR(clerq))
2229 cl_req_completion(env, clerq, rc);
2230 }
2231 return rc;
2232 }
2233
2234 static int osc_set_lock_data_with_check(struct ldlm_lock *lock,
2235 struct ldlm_enqueue_info *einfo)
2236 {
2237 void *data = einfo->ei_cbdata;
2238 int set = 0;
2239
2240 LASSERT(lock != NULL);
2241 LASSERT(lock->l_blocking_ast == einfo->ei_cb_bl);
2242 LASSERT(lock->l_resource->lr_type == einfo->ei_type);
2243 LASSERT(lock->l_completion_ast == einfo->ei_cb_cp);
2244 LASSERT(lock->l_glimpse_ast == einfo->ei_cb_gl);
2245
2246 lock_res_and_lock(lock);
2247 spin_lock(&osc_ast_guard);
2248
2249 if (lock->l_ast_data == NULL)
2250 lock->l_ast_data = data;
2251 if (lock->l_ast_data == data)
2252 set = 1;
2253
2254 spin_unlock(&osc_ast_guard);
2255 unlock_res_and_lock(lock);
2256
2257 return set;
2258 }
2259
2260 static int osc_set_data_with_check(struct lustre_handle *lockh,
2261 struct ldlm_enqueue_info *einfo)
2262 {
2263 struct ldlm_lock *lock = ldlm_handle2lock(lockh);
2264 int set = 0;
2265
2266 if (lock != NULL) {
2267 set = osc_set_lock_data_with_check(lock, einfo);
2268 LDLM_LOCK_PUT(lock);
2269 } else
2270 CERROR("lockh %p, data %p - client evicted?\n",
2271 lockh, einfo->ei_cbdata);
2272 return set;
2273 }
2274
2275 static int osc_change_cbdata(struct obd_export *exp, struct lov_stripe_md *lsm,
2276 ldlm_iterator_t replace, void *data)
2277 {
2278 struct ldlm_res_id res_id;
2279 struct obd_device *obd = class_exp2obd(exp);
2280
2281 ostid_build_res_name(&lsm->lsm_oi, &res_id);
2282 ldlm_resource_iterate(obd->obd_namespace, &res_id, replace, data);
2283 return 0;
2284 }
2285
2286 /* find any ldlm lock of the inode in osc
2287 * return 0 not find
2288 * 1 find one
2289 * < 0 error */
2290 static int osc_find_cbdata(struct obd_export *exp, struct lov_stripe_md *lsm,
2291 ldlm_iterator_t replace, void *data)
2292 {
2293 struct ldlm_res_id res_id;
2294 struct obd_device *obd = class_exp2obd(exp);
2295 int rc = 0;
2296
2297 ostid_build_res_name(&lsm->lsm_oi, &res_id);
2298 rc = ldlm_resource_iterate(obd->obd_namespace, &res_id, replace, data);
2299 if (rc == LDLM_ITER_STOP)
2300 return(1);
2301 if (rc == LDLM_ITER_CONTINUE)
2302 return(0);
2303 return(rc);
2304 }
2305
2306 static int osc_enqueue_fini(struct ptlrpc_request *req, struct ost_lvb *lvb,
2307 obd_enqueue_update_f upcall, void *cookie,
2308 __u64 *flags, int agl, int rc)
2309 {
2310 int intent = *flags & LDLM_FL_HAS_INTENT;
2311
2312 if (intent) {
2313 /* The request was created before ldlm_cli_enqueue call. */
2314 if (rc == ELDLM_LOCK_ABORTED) {
2315 struct ldlm_reply *rep;
2316 rep = req_capsule_server_get(&req->rq_pill,
2317 &RMF_DLM_REP);
2318
2319 LASSERT(rep != NULL);
2320 rep->lock_policy_res1 =
2321 ptlrpc_status_ntoh(rep->lock_policy_res1);
2322 if (rep->lock_policy_res1)
2323 rc = rep->lock_policy_res1;
2324 }
2325 }
2326
2327 if ((intent != 0 && rc == ELDLM_LOCK_ABORTED && agl == 0) ||
2328 (rc == 0)) {
2329 *flags |= LDLM_FL_LVB_READY;
2330 CDEBUG(D_INODE,"got kms "LPU64" blocks "LPU64" mtime "LPU64"\n",
2331 lvb->lvb_size, lvb->lvb_blocks, lvb->lvb_mtime);
2332 }
2333
2334 /* Call the update callback. */
2335 rc = (*upcall)(cookie, rc);
2336 return rc;
2337 }
2338
2339 static int osc_enqueue_interpret(const struct lu_env *env,
2340 struct ptlrpc_request *req,
2341 struct osc_enqueue_args *aa, int rc)
2342 {
2343 struct ldlm_lock *lock;
2344 struct lustre_handle handle;
2345 __u32 mode;
2346 struct ost_lvb *lvb;
2347 __u32 lvb_len;
2348 __u64 *flags = aa->oa_flags;
2349
2350 /* Make a local copy of a lock handle and a mode, because aa->oa_*
2351 * might be freed anytime after lock upcall has been called. */
2352 lustre_handle_copy(&handle, aa->oa_lockh);
2353 mode = aa->oa_ei->ei_mode;
2354
2355 /* ldlm_cli_enqueue is holding a reference on the lock, so it must
2356 * be valid. */
2357 lock = ldlm_handle2lock(&handle);
2358
2359 /* Take an additional reference so that a blocking AST that
2360 * ldlm_cli_enqueue_fini() might post for a failed lock, is guaranteed
2361 * to arrive after an upcall has been executed by
2362 * osc_enqueue_fini(). */
2363 ldlm_lock_addref(&handle, mode);
2364
2365 /* Let CP AST to grant the lock first. */
2366 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_ENQ_RACE, 1);
2367
2368 if (aa->oa_agl && rc == ELDLM_LOCK_ABORTED) {
2369 lvb = NULL;
2370 lvb_len = 0;
2371 } else {
2372 lvb = aa->oa_lvb;
2373 lvb_len = sizeof(*aa->oa_lvb);
2374 }
2375
2376 /* Complete obtaining the lock procedure. */
2377 rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, aa->oa_ei->ei_type, 1,
2378 mode, flags, lvb, lvb_len, &handle, rc);
2379 /* Complete osc stuff. */
2380 rc = osc_enqueue_fini(req, aa->oa_lvb, aa->oa_upcall, aa->oa_cookie,
2381 flags, aa->oa_agl, rc);
2382
2383 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_CP_CANCEL_RACE, 10);
2384
2385 /* Release the lock for async request. */
2386 if (lustre_handle_is_used(&handle) && rc == ELDLM_OK)
2387 /*
2388 * Releases a reference taken by ldlm_cli_enqueue(), if it is
2389 * not already released by
2390 * ldlm_cli_enqueue_fini()->failed_lock_cleanup()
2391 */
2392 ldlm_lock_decref(&handle, mode);
2393
2394 LASSERTF(lock != NULL, "lockh %p, req %p, aa %p - client evicted?\n",
2395 aa->oa_lockh, req, aa);
2396 ldlm_lock_decref(&handle, mode);
2397 LDLM_LOCK_PUT(lock);
2398 return rc;
2399 }
2400
2401 void osc_update_enqueue(struct lustre_handle *lov_lockhp,
2402 struct lov_oinfo *loi, int flags,
2403 struct ost_lvb *lvb, __u32 mode, int rc)
2404 {
2405 struct ldlm_lock *lock = ldlm_handle2lock(lov_lockhp);
2406
2407 if (rc == ELDLM_OK) {
2408 __u64 tmp;
2409
2410 LASSERT(lock != NULL);
2411 loi->loi_lvb = *lvb;
2412 tmp = loi->loi_lvb.lvb_size;
2413 /* Extend KMS up to the end of this lock and no further
2414 * A lock on [x,y] means a KMS of up to y + 1 bytes! */
2415 if (tmp > lock->l_policy_data.l_extent.end)
2416 tmp = lock->l_policy_data.l_extent.end + 1;
2417 if (tmp >= loi->loi_kms) {
2418 LDLM_DEBUG(lock, "lock acquired, setting rss="LPU64
2419 ", kms="LPU64, loi->loi_lvb.lvb_size, tmp);
2420 loi_kms_set(loi, tmp);
2421 } else {
2422 LDLM_DEBUG(lock, "lock acquired, setting rss="
2423 LPU64"; leaving kms="LPU64", end="LPU64,
2424 loi->loi_lvb.lvb_size, loi->loi_kms,
2425 lock->l_policy_data.l_extent.end);
2426 }
2427 ldlm_lock_allow_match(lock);
2428 } else if (rc == ELDLM_LOCK_ABORTED && (flags & LDLM_FL_HAS_INTENT)) {
2429 LASSERT(lock != NULL);
2430 loi->loi_lvb = *lvb;
2431 ldlm_lock_allow_match(lock);
2432 CDEBUG(D_INODE, "glimpsed, setting rss="LPU64"; leaving"
2433 " kms="LPU64"\n", loi->loi_lvb.lvb_size, loi->loi_kms);
2434 rc = ELDLM_OK;
2435 }
2436
2437 if (lock != NULL) {
2438 if (rc != ELDLM_OK)
2439 ldlm_lock_fail_match(lock);
2440
2441 LDLM_LOCK_PUT(lock);
2442 }
2443 }
2444 EXPORT_SYMBOL(osc_update_enqueue);
2445
2446 struct ptlrpc_request_set *PTLRPCD_SET = (void *)1;
2447
2448 /* When enqueuing asynchronously, locks are not ordered, we can obtain a lock
2449 * from the 2nd OSC before a lock from the 1st one. This does not deadlock with
2450 * other synchronous requests, however keeping some locks and trying to obtain
2451 * others may take a considerable amount of time in a case of ost failure; and
2452 * when other sync requests do not get released lock from a client, the client
2453 * is excluded from the cluster -- such scenarious make the life difficult, so
2454 * release locks just after they are obtained. */
2455 int osc_enqueue_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2456 __u64 *flags, ldlm_policy_data_t *policy,
2457 struct ost_lvb *lvb, int kms_valid,
2458 obd_enqueue_update_f upcall, void *cookie,
2459 struct ldlm_enqueue_info *einfo,
2460 struct lustre_handle *lockh,
2461 struct ptlrpc_request_set *rqset, int async, int agl)
2462 {
2463 struct obd_device *obd = exp->exp_obd;
2464 struct ptlrpc_request *req = NULL;
2465 int intent = *flags & LDLM_FL_HAS_INTENT;
2466 int match_lvb = (agl != 0 ? 0 : LDLM_FL_LVB_READY);
2467 ldlm_mode_t mode;
2468 int rc;
2469
2470 /* Filesystem lock extents are extended to page boundaries so that
2471 * dealing with the page cache is a little smoother. */
2472 policy->l_extent.start -= policy->l_extent.start & ~CFS_PAGE_MASK;
2473 policy->l_extent.end |= ~CFS_PAGE_MASK;
2474
2475 /*
2476 * kms is not valid when either object is completely fresh (so that no
2477 * locks are cached), or object was evicted. In the latter case cached
2478 * lock cannot be used, because it would prime inode state with
2479 * potentially stale LVB.
2480 */
2481 if (!kms_valid)
2482 goto no_match;
2483
2484 /* Next, search for already existing extent locks that will cover us */
2485 /* If we're trying to read, we also search for an existing PW lock. The
2486 * VFS and page cache already protect us locally, so lots of readers/
2487 * writers can share a single PW lock.
2488 *
2489 * There are problems with conversion deadlocks, so instead of
2490 * converting a read lock to a write lock, we'll just enqueue a new
2491 * one.
2492 *
2493 * At some point we should cancel the read lock instead of making them
2494 * send us a blocking callback, but there are problems with canceling
2495 * locks out from other users right now, too. */
2496 mode = einfo->ei_mode;
2497 if (einfo->ei_mode == LCK_PR)
2498 mode |= LCK_PW;
2499 mode = ldlm_lock_match(obd->obd_namespace, *flags | match_lvb, res_id,
2500 einfo->ei_type, policy, mode, lockh, 0);
2501 if (mode) {
2502 struct ldlm_lock *matched = ldlm_handle2lock(lockh);
2503
2504 if ((agl != 0) && !(matched->l_flags & LDLM_FL_LVB_READY)) {
2505 /* For AGL, if enqueue RPC is sent but the lock is not
2506 * granted, then skip to process this strpe.
2507 * Return -ECANCELED to tell the caller. */
2508 ldlm_lock_decref(lockh, mode);
2509 LDLM_LOCK_PUT(matched);
2510 return -ECANCELED;
2511 } else if (osc_set_lock_data_with_check(matched, einfo)) {
2512 *flags |= LDLM_FL_LVB_READY;
2513 /* addref the lock only if not async requests and PW
2514 * lock is matched whereas we asked for PR. */
2515 if (!rqset && einfo->ei_mode != mode)
2516 ldlm_lock_addref(lockh, LCK_PR);
2517 if (intent) {
2518 /* I would like to be able to ASSERT here that
2519 * rss <= kms, but I can't, for reasons which
2520 * are explained in lov_enqueue() */
2521 }
2522
2523 /* We already have a lock, and it's referenced.
2524 *
2525 * At this point, the cl_lock::cll_state is CLS_QUEUING,
2526 * AGL upcall may change it to CLS_HELD directly. */
2527 (*upcall)(cookie, ELDLM_OK);
2528
2529 if (einfo->ei_mode != mode)
2530 ldlm_lock_decref(lockh, LCK_PW);
2531 else if (rqset)
2532 /* For async requests, decref the lock. */
2533 ldlm_lock_decref(lockh, einfo->ei_mode);
2534 LDLM_LOCK_PUT(matched);
2535 return ELDLM_OK;
2536 } else {
2537 ldlm_lock_decref(lockh, mode);
2538 LDLM_LOCK_PUT(matched);
2539 }
2540 }
2541
2542 no_match:
2543 if (intent) {
2544 LIST_HEAD(cancels);
2545 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2546 &RQF_LDLM_ENQUEUE_LVB);
2547 if (req == NULL)
2548 return -ENOMEM;
2549
2550 rc = ldlm_prep_enqueue_req(exp, req, &cancels, 0);
2551 if (rc) {
2552 ptlrpc_request_free(req);
2553 return rc;
2554 }
2555
2556 req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER,
2557 sizeof *lvb);
2558 ptlrpc_request_set_replen(req);
2559 }
2560
2561 /* users of osc_enqueue() can pass this flag for ldlm_lock_match() */
2562 *flags &= ~LDLM_FL_BLOCK_GRANTED;
2563
2564 rc = ldlm_cli_enqueue(exp, &req, einfo, res_id, policy, flags, lvb,
2565 sizeof(*lvb), LVB_T_OST, lockh, async);
2566 if (rqset) {
2567 if (!rc) {
2568 struct osc_enqueue_args *aa;
2569 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2570 aa = ptlrpc_req_async_args(req);
2571 aa->oa_ei = einfo;
2572 aa->oa_exp = exp;
2573 aa->oa_flags = flags;
2574 aa->oa_upcall = upcall;
2575 aa->oa_cookie = cookie;
2576 aa->oa_lvb = lvb;
2577 aa->oa_lockh = lockh;
2578 aa->oa_agl = !!agl;
2579
2580 req->rq_interpret_reply =
2581 (ptlrpc_interpterer_t)osc_enqueue_interpret;
2582 if (rqset == PTLRPCD_SET)
2583 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
2584 else
2585 ptlrpc_set_add_req(rqset, req);
2586 } else if (intent) {
2587 ptlrpc_req_finished(req);
2588 }
2589 return rc;
2590 }
2591
2592 rc = osc_enqueue_fini(req, lvb, upcall, cookie, flags, agl, rc);
2593 if (intent)
2594 ptlrpc_req_finished(req);
2595
2596 return rc;
2597 }
2598
2599 static int osc_enqueue(struct obd_export *exp, struct obd_info *oinfo,
2600 struct ldlm_enqueue_info *einfo,
2601 struct ptlrpc_request_set *rqset)
2602 {
2603 struct ldlm_res_id res_id;
2604 int rc;
2605
2606 ostid_build_res_name(&oinfo->oi_md->lsm_oi, &res_id);
2607 rc = osc_enqueue_base(exp, &res_id, &oinfo->oi_flags, &oinfo->oi_policy,
2608 &oinfo->oi_md->lsm_oinfo[0]->loi_lvb,
2609 oinfo->oi_md->lsm_oinfo[0]->loi_kms_valid,
2610 oinfo->oi_cb_up, oinfo, einfo, oinfo->oi_lockh,
2611 rqset, rqset != NULL, 0);
2612 return rc;
2613 }
2614
2615 int osc_match_base(struct obd_export *exp, struct ldlm_res_id *res_id,
2616 __u32 type, ldlm_policy_data_t *policy, __u32 mode,
2617 int *flags, void *data, struct lustre_handle *lockh,
2618 int unref)
2619 {
2620 struct obd_device *obd = exp->exp_obd;
2621 int lflags = *flags;
2622 ldlm_mode_t rc;
2623
2624 if (OBD_FAIL_CHECK(OBD_FAIL_OSC_MATCH))
2625 return -EIO;
2626
2627 /* Filesystem lock extents are extended to page boundaries so that
2628 * dealing with the page cache is a little smoother */
2629 policy->l_extent.start -= policy->l_extent.start & ~CFS_PAGE_MASK;
2630 policy->l_extent.end |= ~CFS_PAGE_MASK;
2631
2632 /* Next, search for already existing extent locks that will cover us */
2633 /* If we're trying to read, we also search for an existing PW lock. The
2634 * VFS and page cache already protect us locally, so lots of readers/
2635 * writers can share a single PW lock. */
2636 rc = mode;
2637 if (mode == LCK_PR)
2638 rc |= LCK_PW;
2639 rc = ldlm_lock_match(obd->obd_namespace, lflags,
2640 res_id, type, policy, rc, lockh, unref);
2641 if (rc) {
2642 if (data != NULL) {
2643 if (!osc_set_data_with_check(lockh, data)) {
2644 if (!(lflags & LDLM_FL_TEST_LOCK))
2645 ldlm_lock_decref(lockh, rc);
2646 return 0;
2647 }
2648 }
2649 if (!(lflags & LDLM_FL_TEST_LOCK) && mode != rc) {
2650 ldlm_lock_addref(lockh, LCK_PR);
2651 ldlm_lock_decref(lockh, LCK_PW);
2652 }
2653 return rc;
2654 }
2655 return rc;
2656 }
2657
2658 int osc_cancel_base(struct lustre_handle *lockh, __u32 mode)
2659 {
2660 if (unlikely(mode == LCK_GROUP))
2661 ldlm_lock_decref_and_cancel(lockh, mode);
2662 else
2663 ldlm_lock_decref(lockh, mode);
2664
2665 return 0;
2666 }
2667
2668 static int osc_cancel(struct obd_export *exp, struct lov_stripe_md *md,
2669 __u32 mode, struct lustre_handle *lockh)
2670 {
2671 return osc_cancel_base(lockh, mode);
2672 }
2673
2674 static int osc_cancel_unused(struct obd_export *exp,
2675 struct lov_stripe_md *lsm,
2676 ldlm_cancel_flags_t flags,
2677 void *opaque)
2678 {
2679 struct obd_device *obd = class_exp2obd(exp);
2680 struct ldlm_res_id res_id, *resp = NULL;
2681
2682 if (lsm != NULL) {
2683 ostid_build_res_name(&lsm->lsm_oi, &res_id);
2684 resp = &res_id;
2685 }
2686
2687 return ldlm_cli_cancel_unused(obd->obd_namespace, resp, flags, opaque);
2688 }
2689
2690 static int osc_statfs_interpret(const struct lu_env *env,
2691 struct ptlrpc_request *req,
2692 struct osc_async_args *aa, int rc)
2693 {
2694 struct obd_statfs *msfs;
2695
2696 if (rc == -EBADR)
2697 /* The request has in fact never been sent
2698 * due to issues at a higher level (LOV).
2699 * Exit immediately since the caller is
2700 * aware of the problem and takes care
2701 * of the clean up */
2702 return rc;
2703
2704 if ((rc == -ENOTCONN || rc == -EAGAIN) &&
2705 (aa->aa_oi->oi_flags & OBD_STATFS_NODELAY))
2706 GOTO(out, rc = 0);
2707
2708 if (rc != 0)
2709 GOTO(out, rc);
2710
2711 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2712 if (msfs == NULL) {
2713 GOTO(out, rc = -EPROTO);
2714 }
2715
2716 *aa->aa_oi->oi_osfs = *msfs;
2717 out:
2718 rc = aa->aa_oi->oi_cb_up(aa->aa_oi, rc);
2719 return rc;
2720 }
2721
2722 static int osc_statfs_async(struct obd_export *exp,
2723 struct obd_info *oinfo, __u64 max_age,
2724 struct ptlrpc_request_set *rqset)
2725 {
2726 struct obd_device *obd = class_exp2obd(exp);
2727 struct ptlrpc_request *req;
2728 struct osc_async_args *aa;
2729 int rc;
2730
2731 /* We could possibly pass max_age in the request (as an absolute
2732 * timestamp or a "seconds.usec ago") so the target can avoid doing
2733 * extra calls into the filesystem if that isn't necessary (e.g.
2734 * during mount that would help a bit). Having relative timestamps
2735 * is not so great if request processing is slow, while absolute
2736 * timestamps are not ideal because they need time synchronization. */
2737 req = ptlrpc_request_alloc(obd->u.cli.cl_import, &RQF_OST_STATFS);
2738 if (req == NULL)
2739 return -ENOMEM;
2740
2741 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2742 if (rc) {
2743 ptlrpc_request_free(req);
2744 return rc;
2745 }
2746 ptlrpc_request_set_replen(req);
2747 req->rq_request_portal = OST_CREATE_PORTAL;
2748 ptlrpc_at_set_req_timeout(req);
2749
2750 if (oinfo->oi_flags & OBD_STATFS_NODELAY) {
2751 /* procfs requests not want stat in wait for avoid deadlock */
2752 req->rq_no_resend = 1;
2753 req->rq_no_delay = 1;
2754 }
2755
2756 req->rq_interpret_reply = (ptlrpc_interpterer_t)osc_statfs_interpret;
2757 CLASSERT (sizeof(*aa) <= sizeof(req->rq_async_args));
2758 aa = ptlrpc_req_async_args(req);
2759 aa->aa_oi = oinfo;
2760
2761 ptlrpc_set_add_req(rqset, req);
2762 return 0;
2763 }
2764
2765 static int osc_statfs(const struct lu_env *env, struct obd_export *exp,
2766 struct obd_statfs *osfs, __u64 max_age, __u32 flags)
2767 {
2768 struct obd_device *obd = class_exp2obd(exp);
2769 struct obd_statfs *msfs;
2770 struct ptlrpc_request *req;
2771 struct obd_import *imp = NULL;
2772 int rc;
2773
2774 /*Since the request might also come from lprocfs, so we need
2775 *sync this with client_disconnect_export Bug15684*/
2776 down_read(&obd->u.cli.cl_sem);
2777 if (obd->u.cli.cl_import)
2778 imp = class_import_get(obd->u.cli.cl_import);
2779 up_read(&obd->u.cli.cl_sem);
2780 if (!imp)
2781 return -ENODEV;
2782
2783 /* We could possibly pass max_age in the request (as an absolute
2784 * timestamp or a "seconds.usec ago") so the target can avoid doing
2785 * extra calls into the filesystem if that isn't necessary (e.g.
2786 * during mount that would help a bit). Having relative timestamps
2787 * is not so great if request processing is slow, while absolute
2788 * timestamps are not ideal because they need time synchronization. */
2789 req = ptlrpc_request_alloc(imp, &RQF_OST_STATFS);
2790
2791 class_import_put(imp);
2792
2793 if (req == NULL)
2794 return -ENOMEM;
2795
2796 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_STATFS);
2797 if (rc) {
2798 ptlrpc_request_free(req);
2799 return rc;
2800 }
2801 ptlrpc_request_set_replen(req);
2802 req->rq_request_portal = OST_CREATE_PORTAL;
2803 ptlrpc_at_set_req_timeout(req);
2804
2805 if (flags & OBD_STATFS_NODELAY) {
2806 /* procfs requests not want stat in wait for avoid deadlock */
2807 req->rq_no_resend = 1;
2808 req->rq_no_delay = 1;
2809 }
2810
2811 rc = ptlrpc_queue_wait(req);
2812 if (rc)
2813 GOTO(out, rc);
2814
2815 msfs = req_capsule_server_get(&req->rq_pill, &RMF_OBD_STATFS);
2816 if (msfs == NULL) {
2817 GOTO(out, rc = -EPROTO);
2818 }
2819
2820 *osfs = *msfs;
2821
2822 out:
2823 ptlrpc_req_finished(req);
2824 return rc;
2825 }
2826
2827 /* Retrieve object striping information.
2828 *
2829 * @lmmu is a pointer to an in-core struct with lmm_ost_count indicating
2830 * the maximum number of OST indices which will fit in the user buffer.
2831 * lmm_magic must be LOV_MAGIC (we only use 1 slot here).
2832 */
2833 static int osc_getstripe(struct lov_stripe_md *lsm, struct lov_user_md *lump)
2834 {
2835 /* we use lov_user_md_v3 because it is larger than lov_user_md_v1 */
2836 struct lov_user_md_v3 lum, *lumk;
2837 struct lov_user_ost_data_v1 *lmm_objects;
2838 int rc = 0, lum_size;
2839
2840 if (!lsm)
2841 return -ENODATA;
2842
2843 /* we only need the header part from user space to get lmm_magic and
2844 * lmm_stripe_count, (the header part is common to v1 and v3) */
2845 lum_size = sizeof(struct lov_user_md_v1);
2846 if (copy_from_user(&lum, lump, lum_size))
2847 return -EFAULT;
2848
2849 if ((lum.lmm_magic != LOV_USER_MAGIC_V1) &&
2850 (lum.lmm_magic != LOV_USER_MAGIC_V3))
2851 return -EINVAL;
2852
2853 /* lov_user_md_vX and lov_mds_md_vX must have the same size */
2854 LASSERT(sizeof(struct lov_user_md_v1) == sizeof(struct lov_mds_md_v1));
2855 LASSERT(sizeof(struct lov_user_md_v3) == sizeof(struct lov_mds_md_v3));
2856 LASSERT(sizeof(lum.lmm_objects[0]) == sizeof(lumk->lmm_objects[0]));
2857
2858 /* we can use lov_mds_md_size() to compute lum_size
2859 * because lov_user_md_vX and lov_mds_md_vX have the same size */
2860 if (lum.lmm_stripe_count > 0) {
2861 lum_size = lov_mds_md_size(lum.lmm_stripe_count, lum.lmm_magic);
2862 OBD_ALLOC(lumk, lum_size);
2863 if (!lumk)
2864 return -ENOMEM;
2865
2866 if (lum.lmm_magic == LOV_USER_MAGIC_V1)
2867 lmm_objects =
2868 &(((struct lov_user_md_v1 *)lumk)->lmm_objects[0]);
2869 else
2870 lmm_objects = &(lumk->lmm_objects[0]);
2871 lmm_objects->l_ost_oi = lsm->lsm_oi;
2872 } else {
2873 lum_size = lov_mds_md_size(0, lum.lmm_magic);
2874 lumk = &lum;
2875 }
2876
2877 lumk->lmm_oi = lsm->lsm_oi;
2878 lumk->lmm_stripe_count = 1;
2879
2880 if (copy_to_user(lump, lumk, lum_size))
2881 rc = -EFAULT;
2882
2883 if (lumk != &lum)
2884 OBD_FREE(lumk, lum_size);
2885
2886 return rc;
2887 }
2888
2889
2890 static int osc_iocontrol(unsigned int cmd, struct obd_export *exp, int len,
2891 void *karg, void *uarg)
2892 {
2893 struct obd_device *obd = exp->exp_obd;
2894 struct obd_ioctl_data *data = karg;
2895 int err = 0;
2896
2897 if (!try_module_get(THIS_MODULE)) {
2898 CERROR("Can't get module. Is it alive?");
2899 return -EINVAL;
2900 }
2901 switch (cmd) {
2902 case OBD_IOC_LOV_GET_CONFIG: {
2903 char *buf;
2904 struct lov_desc *desc;
2905 struct obd_uuid uuid;
2906
2907 buf = NULL;
2908 len = 0;
2909 if (obd_ioctl_getdata(&buf, &len, (void *)uarg))
2910 GOTO(out, err = -EINVAL);
2911
2912 data = (struct obd_ioctl_data *)buf;
2913
2914 if (sizeof(*desc) > data->ioc_inllen1) {
2915 obd_ioctl_freedata(buf, len);
2916 GOTO(out, err = -EINVAL);
2917 }
2918
2919 if (data->ioc_inllen2 < sizeof(uuid)) {
2920 obd_ioctl_freedata(buf, len);
2921 GOTO(out, err = -EINVAL);
2922 }
2923
2924 desc = (struct lov_desc *)data->ioc_inlbuf1;
2925 desc->ld_tgt_count = 1;
2926 desc->ld_active_tgt_count = 1;
2927 desc->ld_default_stripe_count = 1;
2928 desc->ld_default_stripe_size = 0;
2929 desc->ld_default_stripe_offset = 0;
2930 desc->ld_pattern = 0;
2931 memcpy(&desc->ld_uuid, &obd->obd_uuid, sizeof(uuid));
2932
2933 memcpy(data->ioc_inlbuf2, &obd->obd_uuid, sizeof(uuid));
2934
2935 err = copy_to_user((void *)uarg, buf, len);
2936 if (err)
2937 err = -EFAULT;
2938 obd_ioctl_freedata(buf, len);
2939 GOTO(out, err);
2940 }
2941 case LL_IOC_LOV_SETSTRIPE:
2942 err = obd_alloc_memmd(exp, karg);
2943 if (err > 0)
2944 err = 0;
2945 GOTO(out, err);
2946 case LL_IOC_LOV_GETSTRIPE:
2947 err = osc_getstripe(karg, uarg);
2948 GOTO(out, err);
2949 case OBD_IOC_CLIENT_RECOVER:
2950 err = ptlrpc_recover_import(obd->u.cli.cl_import,
2951 data->ioc_inlbuf1, 0);
2952 if (err > 0)
2953 err = 0;
2954 GOTO(out, err);
2955 case IOC_OSC_SET_ACTIVE:
2956 err = ptlrpc_set_import_active(obd->u.cli.cl_import,
2957 data->ioc_offset);
2958 GOTO(out, err);
2959 case OBD_IOC_POLL_QUOTACHECK:
2960 err = osc_quota_poll_check(exp, (struct if_quotacheck *)karg);
2961 GOTO(out, err);
2962 case OBD_IOC_PING_TARGET:
2963 err = ptlrpc_obd_ping(obd);
2964 GOTO(out, err);
2965 default:
2966 CDEBUG(D_INODE, "unrecognised ioctl %#x by %s\n",
2967 cmd, current_comm());
2968 GOTO(out, err = -ENOTTY);
2969 }
2970 out:
2971 module_put(THIS_MODULE);
2972 return err;
2973 }
2974
2975 static int osc_get_info(const struct lu_env *env, struct obd_export *exp,
2976 obd_count keylen, void *key, __u32 *vallen, void *val,
2977 struct lov_stripe_md *lsm)
2978 {
2979 if (!vallen || !val)
2980 return -EFAULT;
2981
2982 if (KEY_IS(KEY_LOCK_TO_STRIPE)) {
2983 __u32 *stripe = val;
2984 *vallen = sizeof(*stripe);
2985 *stripe = 0;
2986 return 0;
2987 } else if (KEY_IS(KEY_LAST_ID)) {
2988 struct ptlrpc_request *req;
2989 obd_id *reply;
2990 char *tmp;
2991 int rc;
2992
2993 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
2994 &RQF_OST_GET_INFO_LAST_ID);
2995 if (req == NULL)
2996 return -ENOMEM;
2997
2998 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
2999 RCL_CLIENT, keylen);
3000 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GET_INFO);
3001 if (rc) {
3002 ptlrpc_request_free(req);
3003 return rc;
3004 }
3005
3006 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
3007 memcpy(tmp, key, keylen);
3008
3009 req->rq_no_delay = req->rq_no_resend = 1;
3010 ptlrpc_request_set_replen(req);
3011 rc = ptlrpc_queue_wait(req);
3012 if (rc)
3013 GOTO(out, rc);
3014
3015 reply = req_capsule_server_get(&req->rq_pill, &RMF_OBD_ID);
3016 if (reply == NULL)
3017 GOTO(out, rc = -EPROTO);
3018
3019 *((obd_id *)val) = *reply;
3020 out:
3021 ptlrpc_req_finished(req);
3022 return rc;
3023 } else if (KEY_IS(KEY_FIEMAP)) {
3024 struct ll_fiemap_info_key *fm_key =
3025 (struct ll_fiemap_info_key *)key;
3026 struct ldlm_res_id res_id;
3027 ldlm_policy_data_t policy;
3028 struct lustre_handle lockh;
3029 ldlm_mode_t mode = 0;
3030 struct ptlrpc_request *req;
3031 struct ll_user_fiemap *reply;
3032 char *tmp;
3033 int rc;
3034
3035 if (!(fm_key->fiemap.fm_flags & FIEMAP_FLAG_SYNC))
3036 goto skip_locking;
3037
3038 policy.l_extent.start = fm_key->fiemap.fm_start &
3039 CFS_PAGE_MASK;
3040
3041 if (OBD_OBJECT_EOF - fm_key->fiemap.fm_length <=
3042 fm_key->fiemap.fm_start + PAGE_CACHE_SIZE - 1)
3043 policy.l_extent.end = OBD_OBJECT_EOF;
3044 else
3045 policy.l_extent.end = (fm_key->fiemap.fm_start +
3046 fm_key->fiemap.fm_length +
3047 PAGE_CACHE_SIZE - 1) & CFS_PAGE_MASK;
3048
3049 ostid_build_res_name(&fm_key->oa.o_oi, &res_id);
3050 mode = ldlm_lock_match(exp->exp_obd->obd_namespace,
3051 LDLM_FL_BLOCK_GRANTED |
3052 LDLM_FL_LVB_READY,
3053 &res_id, LDLM_EXTENT, &policy,
3054 LCK_PR | LCK_PW, &lockh, 0);
3055 if (mode) { /* lock is cached on client */
3056 if (mode != LCK_PR) {
3057 ldlm_lock_addref(&lockh, LCK_PR);
3058 ldlm_lock_decref(&lockh, LCK_PW);
3059 }
3060 } else { /* no cached lock, needs acquire lock on server side */
3061 fm_key->oa.o_valid |= OBD_MD_FLFLAGS;
3062 fm_key->oa.o_flags |= OBD_FL_SRVLOCK;
3063 }
3064
3065 skip_locking:
3066 req = ptlrpc_request_alloc(class_exp2cliimp(exp),
3067 &RQF_OST_GET_INFO_FIEMAP);
3068 if (req == NULL)
3069 GOTO(drop_lock, rc = -ENOMEM);
3070
3071 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_KEY,
3072 RCL_CLIENT, keylen);
3073 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_VAL,
3074 RCL_CLIENT, *vallen);
3075 req_capsule_set_size(&req->rq_pill, &RMF_FIEMAP_VAL,
3076 RCL_SERVER, *vallen);
3077
3078 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_GET_INFO);
3079 if (rc) {
3080 ptlrpc_request_free(req);
3081 GOTO(drop_lock, rc);
3082 }
3083
3084 tmp = req_capsule_client_get(&req->rq_pill, &RMF_FIEMAP_KEY);
3085 memcpy(tmp, key, keylen);
3086 tmp = req_capsule_client_get(&req->rq_pill, &RMF_FIEMAP_VAL);
3087 memcpy(tmp, val, *vallen);
3088
3089 ptlrpc_request_set_replen(req);
3090 rc = ptlrpc_queue_wait(req);
3091 if (rc)
3092 GOTO(fini_req, rc);
3093
3094 reply = req_capsule_server_get(&req->rq_pill, &RMF_FIEMAP_VAL);
3095 if (reply == NULL)
3096 GOTO(fini_req, rc = -EPROTO);
3097
3098 memcpy(val, reply, *vallen);
3099 fini_req:
3100 ptlrpc_req_finished(req);
3101 drop_lock:
3102 if (mode)
3103 ldlm_lock_decref(&lockh, LCK_PR);
3104 return rc;
3105 }
3106
3107 return -EINVAL;
3108 }
3109
3110 static int osc_set_info_async(const struct lu_env *env, struct obd_export *exp,
3111 obd_count keylen, void *key, obd_count vallen,
3112 void *val, struct ptlrpc_request_set *set)
3113 {
3114 struct ptlrpc_request *req;
3115 struct obd_device *obd = exp->exp_obd;
3116 struct obd_import *imp = class_exp2cliimp(exp);
3117 char *tmp;
3118 int rc;
3119
3120 OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_SHUTDOWN, 10);
3121
3122 if (KEY_IS(KEY_CHECKSUM)) {
3123 if (vallen != sizeof(int))
3124 return -EINVAL;
3125 exp->exp_obd->u.cli.cl_checksum = (*(int *)val) ? 1 : 0;
3126 return 0;
3127 }
3128
3129 if (KEY_IS(KEY_SPTLRPC_CONF)) {
3130 sptlrpc_conf_client_adapt(obd);
3131 return 0;
3132 }
3133
3134 if (KEY_IS(KEY_FLUSH_CTX)) {
3135 sptlrpc_import_flush_my_ctx(imp);
3136 return 0;
3137 }
3138
3139 if (KEY_IS(KEY_CACHE_SET)) {
3140 struct client_obd *cli = &obd->u.cli;
3141
3142 LASSERT(cli->cl_cache == NULL); /* only once */
3143 cli->cl_cache = (struct cl_client_cache *)val;
3144 atomic_inc(&cli->cl_cache->ccc_users);
3145 cli->cl_lru_left = &cli->cl_cache->ccc_lru_left;
3146
3147 /* add this osc into entity list */
3148 LASSERT(list_empty(&cli->cl_lru_osc));
3149 spin_lock(&cli->cl_cache->ccc_lru_lock);
3150 list_add(&cli->cl_lru_osc, &cli->cl_cache->ccc_lru);
3151 spin_unlock(&cli->cl_cache->ccc_lru_lock);
3152
3153 return 0;
3154 }
3155
3156 if (KEY_IS(KEY_CACHE_LRU_SHRINK)) {
3157 struct client_obd *cli = &obd->u.cli;
3158 int nr = atomic_read(&cli->cl_lru_in_list) >> 1;
3159 int target = *(int *)val;
3160
3161 nr = osc_lru_shrink(cli, min(nr, target));
3162 *(int *)val -= nr;
3163 return 0;
3164 }
3165
3166 if (!set && !KEY_IS(KEY_GRANT_SHRINK))
3167 return -EINVAL;
3168
3169 /* We pass all other commands directly to OST. Since nobody calls osc
3170 methods directly and everybody is supposed to go through LOV, we
3171 assume lov checked invalid values for us.
3172 The only recognised values so far are evict_by_nid and mds_conn.
3173 Even if something bad goes through, we'd get a -EINVAL from OST
3174 anyway. */
3175
3176 req = ptlrpc_request_alloc(imp, KEY_IS(KEY_GRANT_SHRINK) ?
3177 &RQF_OST_SET_GRANT_INFO :
3178 &RQF_OBD_SET_INFO);
3179 if (req == NULL)
3180 return -ENOMEM;
3181
3182 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_KEY,
3183 RCL_CLIENT, keylen);
3184 if (!KEY_IS(KEY_GRANT_SHRINK))
3185 req_capsule_set_size(&req->rq_pill, &RMF_SETINFO_VAL,
3186 RCL_CLIENT, vallen);
3187 rc = ptlrpc_request_pack(req, LUSTRE_OST_VERSION, OST_SET_INFO);
3188 if (rc) {
3189 ptlrpc_request_free(req);
3190 return rc;
3191 }
3192
3193 tmp = req_capsule_client_get(&req->rq_pill, &RMF_SETINFO_KEY);
3194 memcpy(tmp, key, keylen);
3195 tmp = req_capsule_client_get(&req->rq_pill, KEY_IS(KEY_GRANT_SHRINK) ?
3196 &RMF_OST_BODY :
3197 &RMF_SETINFO_VAL);
3198 memcpy(tmp, val, vallen);
3199
3200 if (KEY_IS(KEY_GRANT_SHRINK)) {
3201 struct osc_grant_args *aa;
3202 struct obdo *oa;
3203
3204 CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
3205 aa = ptlrpc_req_async_args(req);
3206 OBDO_ALLOC(oa);
3207 if (!oa) {
3208 ptlrpc_req_finished(req);
3209 return -ENOMEM;
3210 }
3211 *oa = ((struct ost_body *)val)->oa;
3212 aa->aa_oa = oa;
3213 req->rq_interpret_reply = osc_shrink_grant_interpret;
3214 }
3215
3216 ptlrpc_request_set_replen(req);
3217 if (!KEY_IS(KEY_GRANT_SHRINK)) {
3218 LASSERT(set != NULL);
3219 ptlrpc_set_add_req(set, req);
3220 ptlrpc_check_set(NULL, set);
3221 } else
3222 ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1);
3223
3224 return 0;
3225 }
3226
3227
3228 static int osc_llog_init(struct obd_device *obd, struct obd_llog_group *olg,
3229 struct obd_device *disk_obd, int *index)
3230 {
3231 /* this code is not supposed to be used with LOD/OSP
3232 * to be removed soon */
3233 LBUG();
3234 return 0;
3235 }
3236
3237 static int osc_llog_finish(struct obd_device *obd, int count)
3238 {
3239 struct llog_ctxt *ctxt;
3240
3241 ctxt = llog_get_context(obd, LLOG_MDS_OST_ORIG_CTXT);
3242 if (ctxt) {
3243 llog_cat_close(NULL, ctxt->loc_handle);
3244 llog_cleanup(NULL, ctxt);
3245 }
3246
3247 ctxt = llog_get_context(obd, LLOG_SIZE_REPL_CTXT);
3248 if (ctxt)
3249 llog_cleanup(NULL, ctxt);
3250 return 0;
3251 }
3252
3253 static int osc_reconnect(const struct lu_env *env,
3254 struct obd_export *exp, struct obd_device *obd,
3255 struct obd_uuid *cluuid,
3256 struct obd_connect_data *data,
3257 void *localdata)
3258 {
3259 struct client_obd *cli = &obd->u.cli;
3260
3261 if (data != NULL && (data->ocd_connect_flags & OBD_CONNECT_GRANT)) {
3262 long lost_grant;
3263
3264 client_obd_list_lock(&cli->cl_loi_list_lock);
3265 data->ocd_grant = (cli->cl_avail_grant + cli->cl_dirty) ?:
3266 2 * cli_brw_size(obd);
3267 lost_grant = cli->cl_lost_grant;
3268 cli->cl_lost_grant = 0;
3269 client_obd_list_unlock(&cli->cl_loi_list_lock);
3270
3271 CDEBUG(D_RPCTRACE, "ocd_connect_flags: "LPX64" ocd_version: %d"
3272 " ocd_grant: %d, lost: %ld.\n", data->ocd_connect_flags,
3273 data->ocd_version, data->ocd_grant, lost_grant);
3274 }
3275
3276 return 0;
3277 }
3278
3279 static int osc_disconnect(struct obd_export *exp)
3280 {
3281 struct obd_device *obd = class_exp2obd(exp);
3282 struct llog_ctxt *ctxt;
3283 int rc;
3284
3285 ctxt = llog_get_context(obd, LLOG_SIZE_REPL_CTXT);
3286 if (ctxt) {
3287 if (obd->u.cli.cl_conn_count == 1) {
3288 /* Flush any remaining cancel messages out to the
3289 * target */
3290 llog_sync(ctxt, exp, 0);
3291 }
3292 llog_ctxt_put(ctxt);
3293 } else {
3294 CDEBUG(D_HA, "No LLOG_SIZE_REPL_CTXT found in obd %p\n",
3295 obd);
3296 }
3297
3298 rc = client_disconnect_export(exp);
3299 /**
3300 * Initially we put del_shrink_grant before disconnect_export, but it
3301 * causes the following problem if setup (connect) and cleanup
3302 * (disconnect) are tangled together.
3303 * connect p1 disconnect p2
3304 * ptlrpc_connect_import
3305 * ............... class_manual_cleanup
3306 * osc_disconnect
3307 * del_shrink_grant
3308 * ptlrpc_connect_interrupt
3309 * init_grant_shrink
3310 * add this client to shrink list
3311 * cleanup_osc
3312 * Bang! pinger trigger the shrink.
3313 * So the osc should be disconnected from the shrink list, after we
3314 * are sure the import has been destroyed. BUG18662
3315 */
3316 if (obd->u.cli.cl_import == NULL)
3317 osc_del_shrink_grant(&obd->u.cli);
3318 return rc;
3319 }
3320
3321 static int osc_import_event(struct obd_device *obd,
3322 struct obd_import *imp,
3323 enum obd_import_event event)
3324 {
3325 struct client_obd *cli;
3326 int rc = 0;
3327
3328 LASSERT(imp->imp_obd == obd);
3329
3330 switch (event) {
3331 case IMP_EVENT_DISCON: {
3332 cli = &obd->u.cli;
3333 client_obd_list_lock(&cli->cl_loi_list_lock);
3334 cli->cl_avail_grant = 0;
3335 cli->cl_lost_grant = 0;
3336 client_obd_list_unlock(&cli->cl_loi_list_lock);
3337 break;
3338 }
3339 case IMP_EVENT_INACTIVE: {
3340 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_INACTIVE, NULL);
3341 break;
3342 }
3343 case IMP_EVENT_INVALIDATE: {
3344 struct ldlm_namespace *ns = obd->obd_namespace;
3345 struct lu_env *env;
3346 int refcheck;
3347
3348 env = cl_env_get(&refcheck);
3349 if (!IS_ERR(env)) {
3350 /* Reset grants */
3351 cli = &obd->u.cli;
3352 /* all pages go to failing rpcs due to the invalid
3353 * import */
3354 osc_io_unplug(env, cli, NULL, PDL_POLICY_ROUND);
3355
3356 ldlm_namespace_cleanup(ns, LDLM_FL_LOCAL_ONLY);
3357 cl_env_put(env, &refcheck);
3358 } else
3359 rc = PTR_ERR(env);
3360 break;
3361 }
3362 case IMP_EVENT_ACTIVE: {
3363 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVE, NULL);
3364 break;
3365 }
3366 case IMP_EVENT_OCD: {
3367 struct obd_connect_data *ocd = &imp->imp_connect_data;
3368
3369 if (ocd->ocd_connect_flags & OBD_CONNECT_GRANT)
3370 osc_init_grant(&obd->u.cli, ocd);
3371
3372 /* See bug 7198 */
3373 if (ocd->ocd_connect_flags & OBD_CONNECT_REQPORTAL)
3374 imp->imp_client->cli_request_portal =OST_REQUEST_PORTAL;
3375
3376 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_OCD, NULL);
3377 break;
3378 }
3379 case IMP_EVENT_DEACTIVATE: {
3380 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_DEACTIVATE, NULL);
3381 break;
3382 }
3383 case IMP_EVENT_ACTIVATE: {
3384 rc = obd_notify_observer(obd, obd, OBD_NOTIFY_ACTIVATE, NULL);
3385 break;
3386 }
3387 default:
3388 CERROR("Unknown import event %d\n", event);
3389 LBUG();
3390 }
3391 return rc;
3392 }
3393
3394 /**
3395 * Determine whether the lock can be canceled before replaying the lock
3396 * during recovery, see bug16774 for detailed information.
3397 *
3398 * \retval zero the lock can't be canceled
3399 * \retval other ok to cancel
3400 */
3401 static int osc_cancel_for_recovery(struct ldlm_lock *lock)
3402 {
3403 check_res_locked(lock->l_resource);
3404
3405 /*
3406 * Cancel all unused extent lock in granted mode LCK_PR or LCK_CR.
3407 *
3408 * XXX as a future improvement, we can also cancel unused write lock
3409 * if it doesn't have dirty data and active mmaps.
3410 */
3411 if (lock->l_resource->lr_type == LDLM_EXTENT &&
3412 (lock->l_granted_mode == LCK_PR ||
3413 lock->l_granted_mode == LCK_CR) &&
3414 (osc_dlm_lock_pageref(lock) == 0))
3415 return 1;
3416
3417 return 0;
3418 }
3419
3420 static int brw_queue_work(const struct lu_env *env, void *data)
3421 {
3422 struct client_obd *cli = data;
3423
3424 CDEBUG(D_CACHE, "Run writeback work for client obd %p.\n", cli);
3425
3426 osc_io_unplug(env, cli, NULL, PDL_POLICY_SAME);
3427 return 0;
3428 }
3429
3430 int osc_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
3431 {
3432 struct lprocfs_static_vars lvars = { 0 };
3433 struct client_obd *cli = &obd->u.cli;
3434 void *handler;
3435 int rc;
3436
3437 rc = ptlrpcd_addref();
3438 if (rc)
3439 return rc;
3440
3441 rc = client_obd_setup(obd, lcfg);
3442 if (rc)
3443 GOTO(out_ptlrpcd, rc);
3444
3445 handler = ptlrpcd_alloc_work(cli->cl_import, brw_queue_work, cli);
3446 if (IS_ERR(handler))
3447 GOTO(out_client_setup, rc = PTR_ERR(handler));
3448 cli->cl_writeback_work = handler;
3449
3450 rc = osc_quota_setup(obd);
3451 if (rc)
3452 GOTO(out_ptlrpcd_work, rc);
3453
3454 cli->cl_grant_shrink_interval = GRANT_SHRINK_INTERVAL;
3455 lprocfs_osc_init_vars(&lvars);
3456 if (lprocfs_obd_setup(obd, lvars.obd_vars) == 0) {
3457 lproc_osc_attach_seqstat(obd);
3458 sptlrpc_lprocfs_cliobd_attach(obd);
3459 ptlrpc_lprocfs_register_obd(obd);
3460 }
3461
3462 /* We need to allocate a few requests more, because
3463 * brw_interpret tries to create new requests before freeing
3464 * previous ones, Ideally we want to have 2x max_rpcs_in_flight
3465 * reserved, but I'm afraid that might be too much wasted RAM
3466 * in fact, so 2 is just my guess and still should work. */
3467 cli->cl_import->imp_rq_pool =
3468 ptlrpc_init_rq_pool(cli->cl_max_rpcs_in_flight + 2,
3469 OST_MAXREQSIZE,
3470 ptlrpc_add_rqs_to_pool);
3471
3472 INIT_LIST_HEAD(&cli->cl_grant_shrink_list);
3473 ns_register_cancel(obd->obd_namespace, osc_cancel_for_recovery);
3474 return rc;
3475
3476 out_ptlrpcd_work:
3477 ptlrpcd_destroy_work(handler);
3478 out_client_setup:
3479 client_obd_cleanup(obd);
3480 out_ptlrpcd:
3481 ptlrpcd_decref();
3482 return rc;
3483 }
3484
3485 static int osc_precleanup(struct obd_device *obd, enum obd_cleanup_stage stage)
3486 {
3487 int rc = 0;
3488
3489 switch (stage) {
3490 case OBD_CLEANUP_EARLY: {
3491 struct obd_import *imp;
3492 imp = obd->u.cli.cl_import;
3493 CDEBUG(D_HA, "Deactivating import %s\n", obd->obd_name);
3494 /* ptlrpc_abort_inflight to stop an mds_lov_synchronize */
3495 ptlrpc_deactivate_import(imp);
3496 spin_lock(&imp->imp_lock);
3497 imp->imp_pingable = 0;
3498 spin_unlock(&imp->imp_lock);
3499 break;
3500 }
3501 case OBD_CLEANUP_EXPORTS: {
3502 struct client_obd *cli = &obd->u.cli;
3503 /* LU-464
3504 * for echo client, export may be on zombie list, wait for
3505 * zombie thread to cull it, because cli.cl_import will be
3506 * cleared in client_disconnect_export():
3507 * class_export_destroy() -> obd_cleanup() ->
3508 * echo_device_free() -> echo_client_cleanup() ->
3509 * obd_disconnect() -> osc_disconnect() ->
3510 * client_disconnect_export()
3511 */
3512 obd_zombie_barrier();
3513 if (cli->cl_writeback_work) {
3514 ptlrpcd_destroy_work(cli->cl_writeback_work);
3515 cli->cl_writeback_work = NULL;
3516 }
3517 obd_cleanup_client_import(obd);
3518 ptlrpc_lprocfs_unregister_obd(obd);
3519 lprocfs_obd_cleanup(obd);
3520 rc = obd_llog_finish(obd, 0);
3521 if (rc != 0)
3522 CERROR("failed to cleanup llogging subsystems\n");
3523 break;
3524 }
3525 }
3526 return rc;
3527 }
3528
3529 int osc_cleanup(struct obd_device *obd)
3530 {
3531 struct client_obd *cli = &obd->u.cli;
3532 int rc;
3533
3534 /* lru cleanup */
3535 if (cli->cl_cache != NULL) {
3536 LASSERT(atomic_read(&cli->cl_cache->ccc_users) > 0);
3537 spin_lock(&cli->cl_cache->ccc_lru_lock);
3538 list_del_init(&cli->cl_lru_osc);
3539 spin_unlock(&cli->cl_cache->ccc_lru_lock);
3540 cli->cl_lru_left = NULL;
3541 atomic_dec(&cli->cl_cache->ccc_users);
3542 cli->cl_cache = NULL;
3543 }
3544
3545 /* free memory of osc quota cache */
3546 osc_quota_cleanup(obd);
3547
3548 rc = client_obd_cleanup(obd);
3549
3550 ptlrpcd_decref();
3551 return rc;
3552 }
3553
3554 int osc_process_config_base(struct obd_device *obd, struct lustre_cfg *lcfg)
3555 {
3556 struct lprocfs_static_vars lvars = { 0 };
3557 int rc = 0;
3558
3559 lprocfs_osc_init_vars(&lvars);
3560
3561 switch (lcfg->lcfg_command) {
3562 default:
3563 rc = class_process_proc_param(PARAM_OSC, lvars.obd_vars,
3564 lcfg, obd);
3565 if (rc > 0)
3566 rc = 0;
3567 break;
3568 }
3569
3570 return(rc);
3571 }
3572
3573 static int osc_process_config(struct obd_device *obd, obd_count len, void *buf)
3574 {
3575 return osc_process_config_base(obd, buf);
3576 }
3577
3578 struct obd_ops osc_obd_ops = {
3579 .o_owner = THIS_MODULE,
3580 .o_setup = osc_setup,
3581 .o_precleanup = osc_precleanup,
3582 .o_cleanup = osc_cleanup,
3583 .o_add_conn = client_import_add_conn,
3584 .o_del_conn = client_import_del_conn,
3585 .o_connect = client_connect_import,
3586 .o_reconnect = osc_reconnect,
3587 .o_disconnect = osc_disconnect,
3588 .o_statfs = osc_statfs,
3589 .o_statfs_async = osc_statfs_async,
3590 .o_packmd = osc_packmd,
3591 .o_unpackmd = osc_unpackmd,
3592 .o_create = osc_create,
3593 .o_destroy = osc_destroy,
3594 .o_getattr = osc_getattr,
3595 .o_getattr_async = osc_getattr_async,
3596 .o_setattr = osc_setattr,
3597 .o_setattr_async = osc_setattr_async,
3598 .o_brw = osc_brw,
3599 .o_punch = osc_punch,
3600 .o_sync = osc_sync,
3601 .o_enqueue = osc_enqueue,
3602 .o_change_cbdata = osc_change_cbdata,
3603 .o_find_cbdata = osc_find_cbdata,
3604 .o_cancel = osc_cancel,
3605 .o_cancel_unused = osc_cancel_unused,
3606 .o_iocontrol = osc_iocontrol,
3607 .o_get_info = osc_get_info,
3608 .o_set_info_async = osc_set_info_async,
3609 .o_import_event = osc_import_event,
3610 .o_llog_init = osc_llog_init,
3611 .o_llog_finish = osc_llog_finish,
3612 .o_process_config = osc_process_config,
3613 .o_quotactl = osc_quotactl,
3614 .o_quotacheck = osc_quotacheck,
3615 };
3616
3617 extern struct lu_kmem_descr osc_caches[];
3618 extern spinlock_t osc_ast_guard;
3619 extern struct lock_class_key osc_ast_guard_class;
3620
3621 int __init osc_init(void)
3622 {
3623 struct lprocfs_static_vars lvars = { 0 };
3624 int rc;
3625
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);
3630
3631 rc = lu_kmem_init(osc_caches);
3632 if (rc)
3633 return rc;
3634
3635 lprocfs_osc_init_vars(&lvars);
3636
3637 rc = class_register_type(&osc_obd_ops, NULL, lvars.module_vars,
3638 LUSTRE_OSC_NAME, &osc_device_type);
3639 if (rc) {
3640 lu_kmem_fini(osc_caches);
3641 return rc;
3642 }
3643
3644 spin_lock_init(&osc_ast_guard);
3645 lockdep_set_class(&osc_ast_guard, &osc_ast_guard_class);
3646
3647 return rc;
3648 }
3649
3650 static void /*__exit*/ osc_exit(void)
3651 {
3652 class_unregister_type(LUSTRE_OSC_NAME);
3653 lu_kmem_fini(osc_caches);
3654 }
3655
3656 MODULE_AUTHOR("Sun Microsystems, Inc. <http://www.lustre.org/>");
3657 MODULE_DESCRIPTION("Lustre Object Storage Client (OSC)");
3658 MODULE_LICENSE("GPL");
3659
3660 cfs_module(osc, LUSTRE_VERSION_STRING, osc_init, osc_exit);
Linux kernel - Deliverables
This page took 0.623222 seconds and 5 git commands to generate.