projects / deliverable / linux.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
writeback: Fix performance regression in wb_over_bg_thresh()
[deliverable/linux.git] / drivers / staging / lustre / lustre / lmv / lmv_obd.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) 2004, 2010, Oracle and/or its affiliates. All rights reserved.
28 * Use is subject to license terms.
29 *
30 * Copyright (c) 2011, 2015, 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_LMV
38 #include <linux/slab.h>
39 #include <linux/module.h>
40 #include <linux/init.h>
41 #include <linux/pagemap.h>
42 #include <linux/mm.h>
43 #include <asm/div64.h>
44 #include <linux/seq_file.h>
45 #include <linux/namei.h>
46 #include <linux/uaccess.h>
47
48 #include "../include/lustre/lustre_idl.h"
49 #include "../include/obd_support.h"
50 #include "../include/lustre_lib.h"
51 #include "../include/lustre_net.h"
52 #include "../include/obd_class.h"
53 #include "../include/lprocfs_status.h"
54 #include "../include/lustre_lite.h"
55 #include "../include/lustre_fid.h"
56 #include "../include/lustre_kernelcomm.h"
57 #include "lmv_internal.h"
58
59 static void lmv_activate_target(struct lmv_obd *lmv,
60 struct lmv_tgt_desc *tgt,
61 int activate)
62 {
63 if (tgt->ltd_active == activate)
64 return;
65
66 tgt->ltd_active = activate;
67 lmv->desc.ld_active_tgt_count += (activate ? 1 : -1);
68 }
69
70 /**
71 * Error codes:
72 *
73 * -EINVAL : UUID can't be found in the LMV's target list
74 * -ENOTCONN: The UUID is found, but the target connection is bad (!)
75 * -EBADF : The UUID is found, but the OBD of the wrong type (!)
76 */
77 static int lmv_set_mdc_active(struct lmv_obd *lmv, struct obd_uuid *uuid,
78 int activate)
79 {
80 struct lmv_tgt_desc *uninitialized_var(tgt);
81 struct obd_device *obd;
82 int i;
83 int rc = 0;
84
85 CDEBUG(D_INFO, "Searching in lmv %p for uuid %s (activate=%d)\n",
86 lmv, uuid->uuid, activate);
87
88 spin_lock(&lmv->lmv_lock);
89 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
90 tgt = lmv->tgts[i];
91 if (!tgt || !tgt->ltd_exp)
92 continue;
93
94 CDEBUG(D_INFO, "Target idx %d is %s conn %#llx\n", i,
95 tgt->ltd_uuid.uuid, tgt->ltd_exp->exp_handle.h_cookie);
96
97 if (obd_uuid_equals(uuid, &tgt->ltd_uuid))
98 break;
99 }
100
101 if (i == lmv->desc.ld_tgt_count) {
102 rc = -EINVAL;
103 goto out_lmv_lock;
104 }
105
106 obd = class_exp2obd(tgt->ltd_exp);
107 if (!obd) {
108 rc = -ENOTCONN;
109 goto out_lmv_lock;
110 }
111
112 CDEBUG(D_INFO, "Found OBD %s=%s device %d (%p) type %s at LMV idx %d\n",
113 obd->obd_name, obd->obd_uuid.uuid, obd->obd_minor, obd,
114 obd->obd_type->typ_name, i);
115 LASSERT(strcmp(obd->obd_type->typ_name, LUSTRE_MDC_NAME) == 0);
116
117 if (tgt->ltd_active == activate) {
118 CDEBUG(D_INFO, "OBD %p already %sactive!\n", obd,
119 activate ? "" : "in");
120 goto out_lmv_lock;
121 }
122
123 CDEBUG(D_INFO, "Marking OBD %p %sactive\n", obd,
124 activate ? "" : "in");
125 lmv_activate_target(lmv, tgt, activate);
126
127 out_lmv_lock:
128 spin_unlock(&lmv->lmv_lock);
129 return rc;
130 }
131
132 static struct obd_uuid *lmv_get_uuid(struct obd_export *exp)
133 {
134 struct lmv_obd *lmv = &exp->exp_obd->u.lmv;
135
136 return obd_get_uuid(lmv->tgts[0]->ltd_exp);
137 }
138
139 static int lmv_notify(struct obd_device *obd, struct obd_device *watched,
140 enum obd_notify_event ev, void *data)
141 {
142 struct obd_connect_data *conn_data;
143 struct lmv_obd *lmv = &obd->u.lmv;
144 struct obd_uuid *uuid;
145 int rc = 0;
146
147 if (strcmp(watched->obd_type->typ_name, LUSTRE_MDC_NAME)) {
148 CERROR("unexpected notification of %s %s!\n",
149 watched->obd_type->typ_name,
150 watched->obd_name);
151 return -EINVAL;
152 }
153
154 uuid = &watched->u.cli.cl_target_uuid;
155 if (ev == OBD_NOTIFY_ACTIVE || ev == OBD_NOTIFY_INACTIVE) {
156 /*
157 * Set MDC as active before notifying the observer, so the
158 * observer can use the MDC normally.
159 */
160 rc = lmv_set_mdc_active(lmv, uuid,
161 ev == OBD_NOTIFY_ACTIVE);
162 if (rc) {
163 CERROR("%sactivation of %s failed: %d\n",
164 ev == OBD_NOTIFY_ACTIVE ? "" : "de",
165 uuid->uuid, rc);
166 return rc;
167 }
168 } else if (ev == OBD_NOTIFY_OCD) {
169 conn_data = &watched->u.cli.cl_import->imp_connect_data;
170 /*
171 * XXX: Make sure that ocd_connect_flags from all targets are
172 * the same. Otherwise one of MDTs runs wrong version or
173 * something like this. --umka
174 */
175 obd->obd_self_export->exp_connect_data = *conn_data;
176 }
177 #if 0
178 else if (ev == OBD_NOTIFY_DISCON) {
179 /*
180 * For disconnect event, flush fld cache for failout MDS case.
181 */
182 fld_client_flush(&lmv->lmv_fld);
183 }
184 #endif
185 /*
186 * Pass the notification up the chain.
187 */
188 if (obd->obd_observer)
189 rc = obd_notify(obd->obd_observer, watched, ev, data);
190
191 return rc;
192 }
193
194 /**
195 * This is fake connect function. Its purpose is to initialize lmv and say
196 * caller that everything is okay. Real connection will be performed later.
197 */
198 static int lmv_connect(const struct lu_env *env,
199 struct obd_export **exp, struct obd_device *obd,
200 struct obd_uuid *cluuid, struct obd_connect_data *data,
201 void *localdata)
202 {
203 struct lmv_obd *lmv = &obd->u.lmv;
204 struct lustre_handle conn = { 0 };
205 int rc = 0;
206
207 /*
208 * We don't want to actually do the underlying connections more than
209 * once, so keep track.
210 */
211 lmv->refcount++;
212 if (lmv->refcount > 1) {
213 *exp = NULL;
214 return 0;
215 }
216
217 rc = class_connect(&conn, obd, cluuid);
218 if (rc) {
219 CERROR("class_connection() returned %d\n", rc);
220 return rc;
221 }
222
223 *exp = class_conn2export(&conn);
224 class_export_get(*exp);
225
226 lmv->exp = *exp;
227 lmv->connected = 0;
228 lmv->cluuid = *cluuid;
229
230 if (data)
231 lmv->conn_data = *data;
232
233 lmv->lmv_tgts_kobj = kobject_create_and_add("target_obds",
234 &obd->obd_kobj);
235 /*
236 * All real clients should perform actual connection right away, because
237 * it is possible, that LMV will not have opportunity to connect targets
238 * and MDC stuff will be called directly, for instance while reading
239 * ../mdc/../kbytesfree procfs file, etc.
240 */
241 if (data && data->ocd_connect_flags & OBD_CONNECT_REAL)
242 rc = lmv_check_connect(obd);
243
244 if (rc && lmv->lmv_tgts_kobj)
245 kobject_put(lmv->lmv_tgts_kobj);
246
247 return rc;
248 }
249
250 static void lmv_set_timeouts(struct obd_device *obd)
251 {
252 struct lmv_tgt_desc *tgt;
253 struct lmv_obd *lmv;
254 int i;
255
256 lmv = &obd->u.lmv;
257 if (lmv->server_timeout == 0)
258 return;
259
260 if (lmv->connected == 0)
261 return;
262
263 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
264 tgt = lmv->tgts[i];
265 if (!tgt || !tgt->ltd_exp || tgt->ltd_active == 0)
266 continue;
267
268 obd_set_info_async(NULL, tgt->ltd_exp, sizeof(KEY_INTERMDS),
269 KEY_INTERMDS, 0, NULL, NULL);
270 }
271 }
272
273 static int lmv_init_ea_size(struct obd_export *exp, int easize,
274 int def_easize, int cookiesize, int def_cookiesize)
275 {
276 struct obd_device *obd = exp->exp_obd;
277 struct lmv_obd *lmv = &obd->u.lmv;
278 int i;
279 int rc = 0;
280 int change = 0;
281
282 if (lmv->max_easize < easize) {
283 lmv->max_easize = easize;
284 change = 1;
285 }
286 if (lmv->max_def_easize < def_easize) {
287 lmv->max_def_easize = def_easize;
288 change = 1;
289 }
290 if (lmv->max_cookiesize < cookiesize) {
291 lmv->max_cookiesize = cookiesize;
292 change = 1;
293 }
294 if (lmv->max_def_cookiesize < def_cookiesize) {
295 lmv->max_def_cookiesize = def_cookiesize;
296 change = 1;
297 }
298 if (change == 0)
299 return 0;
300
301 if (lmv->connected == 0)
302 return 0;
303
304 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
305 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp ||
306 lmv->tgts[i]->ltd_active == 0) {
307 CWARN("%s: NULL export for %d\n", obd->obd_name, i);
308 continue;
309 }
310
311 rc = md_init_ea_size(lmv->tgts[i]->ltd_exp, easize, def_easize,
312 cookiesize, def_cookiesize);
313 if (rc) {
314 CERROR("%s: obd_init_ea_size() failed on MDT target %d: rc = %d\n",
315 obd->obd_name, i, rc);
316 break;
317 }
318 }
319 return rc;
320 }
321
322 #define MAX_STRING_SIZE 128
323
324 static int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt)
325 {
326 struct lmv_obd *lmv = &obd->u.lmv;
327 struct obd_uuid *cluuid = &lmv->cluuid;
328 struct obd_uuid lmv_mdc_uuid = { "LMV_MDC_UUID" };
329 struct obd_device *mdc_obd;
330 struct obd_export *mdc_exp;
331 struct lu_fld_target target;
332 int rc;
333
334 mdc_obd = class_find_client_obd(&tgt->ltd_uuid, LUSTRE_MDC_NAME,
335 &obd->obd_uuid);
336 if (!mdc_obd) {
337 CERROR("target %s not attached\n", tgt->ltd_uuid.uuid);
338 return -EINVAL;
339 }
340
341 CDEBUG(D_CONFIG, "connect to %s(%s) - %s, %s FOR %s\n",
342 mdc_obd->obd_name, mdc_obd->obd_uuid.uuid,
343 tgt->ltd_uuid.uuid, obd->obd_uuid.uuid, cluuid->uuid);
344
345 if (!mdc_obd->obd_set_up) {
346 CERROR("target %s is not set up\n", tgt->ltd_uuid.uuid);
347 return -EINVAL;
348 }
349
350 rc = obd_connect(NULL, &mdc_exp, mdc_obd, &lmv_mdc_uuid,
351 &lmv->conn_data, NULL);
352 if (rc) {
353 CERROR("target %s connect error %d\n", tgt->ltd_uuid.uuid, rc);
354 return rc;
355 }
356
357 /*
358 * Init fid sequence client for this mdc and add new fld target.
359 */
360 rc = obd_fid_init(mdc_obd, mdc_exp, LUSTRE_SEQ_METADATA);
361 if (rc)
362 return rc;
363
364 target.ft_srv = NULL;
365 target.ft_exp = mdc_exp;
366 target.ft_idx = tgt->ltd_idx;
367
368 fld_client_add_target(&lmv->lmv_fld, &target);
369
370 rc = obd_register_observer(mdc_obd, obd);
371 if (rc) {
372 obd_disconnect(mdc_exp);
373 CERROR("target %s register_observer error %d\n",
374 tgt->ltd_uuid.uuid, rc);
375 return rc;
376 }
377
378 if (obd->obd_observer) {
379 /*
380 * Tell the observer about the new target.
381 */
382 rc = obd_notify(obd->obd_observer, mdc_exp->exp_obd,
383 OBD_NOTIFY_ACTIVE,
384 (void *)(tgt - lmv->tgts[0]));
385 if (rc) {
386 obd_disconnect(mdc_exp);
387 return rc;
388 }
389 }
390
391 tgt->ltd_active = 1;
392 tgt->ltd_exp = mdc_exp;
393 lmv->desc.ld_active_tgt_count++;
394
395 md_init_ea_size(tgt->ltd_exp, lmv->max_easize, lmv->max_def_easize,
396 lmv->max_cookiesize, lmv->max_def_cookiesize);
397
398 CDEBUG(D_CONFIG, "Connected to %s(%s) successfully (%d)\n",
399 mdc_obd->obd_name, mdc_obd->obd_uuid.uuid,
400 atomic_read(&obd->obd_refcount));
401
402 if (lmv->lmv_tgts_kobj)
403 /* Even if we failed to create the link, that's fine */
404 rc = sysfs_create_link(lmv->lmv_tgts_kobj, &mdc_obd->obd_kobj,
405 mdc_obd->obd_name);
406 return 0;
407 }
408
409 static void lmv_del_target(struct lmv_obd *lmv, int index)
410 {
411 if (!lmv->tgts[index])
412 return;
413
414 kfree(lmv->tgts[index]);
415 lmv->tgts[index] = NULL;
416 return;
417 }
418
419 static int lmv_add_target(struct obd_device *obd, struct obd_uuid *uuidp,
420 __u32 index, int gen)
421 {
422 struct lmv_obd *lmv = &obd->u.lmv;
423 struct lmv_tgt_desc *tgt;
424 int rc = 0;
425
426 CDEBUG(D_CONFIG, "Target uuid: %s. index %d\n", uuidp->uuid, index);
427
428 lmv_init_lock(lmv);
429
430 if (lmv->desc.ld_tgt_count == 0) {
431 struct obd_device *mdc_obd;
432
433 mdc_obd = class_find_client_obd(uuidp, LUSTRE_MDC_NAME,
434 &obd->obd_uuid);
435 if (!mdc_obd) {
436 lmv_init_unlock(lmv);
437 CERROR("%s: Target %s not attached: rc = %d\n",
438 obd->obd_name, uuidp->uuid, -EINVAL);
439 return -EINVAL;
440 }
441 }
442
443 if ((index < lmv->tgts_size) && lmv->tgts[index]) {
444 tgt = lmv->tgts[index];
445 CERROR("%s: UUID %s already assigned at LOV target index %d: rc = %d\n",
446 obd->obd_name,
447 obd_uuid2str(&tgt->ltd_uuid), index, -EEXIST);
448 lmv_init_unlock(lmv);
449 return -EEXIST;
450 }
451
452 if (index >= lmv->tgts_size) {
453 /* We need to reallocate the lmv target array. */
454 struct lmv_tgt_desc **newtgts, **old = NULL;
455 __u32 newsize = 1;
456 __u32 oldsize = 0;
457
458 while (newsize < index + 1)
459 newsize <<= 1;
460 newtgts = kcalloc(newsize, sizeof(*newtgts), GFP_NOFS);
461 if (!newtgts) {
462 lmv_init_unlock(lmv);
463 return -ENOMEM;
464 }
465
466 if (lmv->tgts_size) {
467 memcpy(newtgts, lmv->tgts,
468 sizeof(*newtgts) * lmv->tgts_size);
469 old = lmv->tgts;
470 oldsize = lmv->tgts_size;
471 }
472
473 lmv->tgts = newtgts;
474 lmv->tgts_size = newsize;
475 smp_rmb();
476 kfree(old);
477
478 CDEBUG(D_CONFIG, "tgts: %p size: %d\n", lmv->tgts,
479 lmv->tgts_size);
480 }
481
482 tgt = kzalloc(sizeof(*tgt), GFP_NOFS);
483 if (!tgt) {
484 lmv_init_unlock(lmv);
485 return -ENOMEM;
486 }
487
488 mutex_init(&tgt->ltd_fid_mutex);
489 tgt->ltd_idx = index;
490 tgt->ltd_uuid = *uuidp;
491 tgt->ltd_active = 0;
492 lmv->tgts[index] = tgt;
493 if (index >= lmv->desc.ld_tgt_count)
494 lmv->desc.ld_tgt_count = index + 1;
495
496 if (lmv->connected) {
497 rc = lmv_connect_mdc(obd, tgt);
498 if (rc) {
499 spin_lock(&lmv->lmv_lock);
500 lmv->desc.ld_tgt_count--;
501 memset(tgt, 0, sizeof(*tgt));
502 spin_unlock(&lmv->lmv_lock);
503 } else {
504 int easize = sizeof(struct lmv_stripe_md) +
505 lmv->desc.ld_tgt_count * sizeof(struct lu_fid);
506 lmv_init_ea_size(obd->obd_self_export, easize, 0, 0, 0);
507 }
508 }
509
510 lmv_init_unlock(lmv);
511 return rc;
512 }
513
514 int lmv_check_connect(struct obd_device *obd)
515 {
516 struct lmv_obd *lmv = &obd->u.lmv;
517 struct lmv_tgt_desc *tgt;
518 int i;
519 int rc;
520 int easize;
521
522 if (lmv->connected)
523 return 0;
524
525 lmv_init_lock(lmv);
526 if (lmv->connected) {
527 lmv_init_unlock(lmv);
528 return 0;
529 }
530
531 if (lmv->desc.ld_tgt_count == 0) {
532 lmv_init_unlock(lmv);
533 CERROR("%s: no targets configured.\n", obd->obd_name);
534 return -EINVAL;
535 }
536
537 CDEBUG(D_CONFIG, "Time to connect %s to %s\n",
538 lmv->cluuid.uuid, obd->obd_name);
539
540 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
541 tgt = lmv->tgts[i];
542 if (!tgt)
543 continue;
544 rc = lmv_connect_mdc(obd, tgt);
545 if (rc)
546 goto out_disc;
547 }
548
549 lmv_set_timeouts(obd);
550 class_export_put(lmv->exp);
551 lmv->connected = 1;
552 easize = lmv_get_easize(lmv);
553 lmv_init_ea_size(obd->obd_self_export, easize, 0, 0, 0);
554 lmv_init_unlock(lmv);
555 return 0;
556
557 out_disc:
558 while (i-- > 0) {
559 int rc2;
560
561 tgt = lmv->tgts[i];
562 if (!tgt)
563 continue;
564 tgt->ltd_active = 0;
565 if (tgt->ltd_exp) {
566 --lmv->desc.ld_active_tgt_count;
567 rc2 = obd_disconnect(tgt->ltd_exp);
568 if (rc2) {
569 CERROR("LMV target %s disconnect on MDC idx %d: error %d\n",
570 tgt->ltd_uuid.uuid, i, rc2);
571 }
572 }
573 }
574 class_disconnect(lmv->exp);
575 lmv_init_unlock(lmv);
576 return rc;
577 }
578
579 static int lmv_disconnect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt)
580 {
581 struct lmv_obd *lmv = &obd->u.lmv;
582 struct obd_device *mdc_obd;
583 int rc;
584
585 mdc_obd = class_exp2obd(tgt->ltd_exp);
586
587 if (mdc_obd) {
588 mdc_obd->obd_force = obd->obd_force;
589 mdc_obd->obd_fail = obd->obd_fail;
590 mdc_obd->obd_no_recov = obd->obd_no_recov;
591
592 if (lmv->lmv_tgts_kobj)
593 sysfs_remove_link(lmv->lmv_tgts_kobj,
594 mdc_obd->obd_name);
595 }
596
597 rc = obd_fid_fini(tgt->ltd_exp->exp_obd);
598 if (rc)
599 CERROR("Can't finalize fids factory\n");
600
601 CDEBUG(D_INFO, "Disconnected from %s(%s) successfully\n",
602 tgt->ltd_exp->exp_obd->obd_name,
603 tgt->ltd_exp->exp_obd->obd_uuid.uuid);
604
605 obd_register_observer(tgt->ltd_exp->exp_obd, NULL);
606 rc = obd_disconnect(tgt->ltd_exp);
607 if (rc) {
608 if (tgt->ltd_active) {
609 CERROR("Target %s disconnect error %d\n",
610 tgt->ltd_uuid.uuid, rc);
611 }
612 }
613
614 lmv_activate_target(lmv, tgt, 0);
615 tgt->ltd_exp = NULL;
616 return 0;
617 }
618
619 static int lmv_disconnect(struct obd_export *exp)
620 {
621 struct obd_device *obd = class_exp2obd(exp);
622 struct lmv_obd *lmv = &obd->u.lmv;
623 int rc;
624 int i;
625
626 if (!lmv->tgts)
627 goto out_local;
628
629 /*
630 * Only disconnect the underlying layers on the final disconnect.
631 */
632 lmv->refcount--;
633 if (lmv->refcount != 0)
634 goto out_local;
635
636 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
637 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp)
638 continue;
639
640 lmv_disconnect_mdc(obd, lmv->tgts[i]);
641 }
642
643 if (lmv->lmv_tgts_kobj)
644 kobject_put(lmv->lmv_tgts_kobj);
645
646 out_local:
647 /*
648 * This is the case when no real connection is established by
649 * lmv_check_connect().
650 */
651 if (!lmv->connected)
652 class_export_put(exp);
653 rc = class_disconnect(exp);
654 if (lmv->refcount == 0)
655 lmv->connected = 0;
656 return rc;
657 }
658
659 static int lmv_fid2path(struct obd_export *exp, int len, void *karg,
660 void __user *uarg)
661 {
662 struct obd_device *obddev = class_exp2obd(exp);
663 struct lmv_obd *lmv = &obddev->u.lmv;
664 struct getinfo_fid2path *gf;
665 struct lmv_tgt_desc *tgt;
666 struct getinfo_fid2path *remote_gf = NULL;
667 int remote_gf_size = 0;
668 int rc;
669
670 gf = (struct getinfo_fid2path *)karg;
671 tgt = lmv_find_target(lmv, &gf->gf_fid);
672 if (IS_ERR(tgt))
673 return PTR_ERR(tgt);
674
675 repeat_fid2path:
676 rc = obd_iocontrol(OBD_IOC_FID2PATH, tgt->ltd_exp, len, gf, uarg);
677 if (rc != 0 && rc != -EREMOTE)
678 goto out_fid2path;
679
680 /* If remote_gf != NULL, it means just building the
681 * path on the remote MDT, copy this path segment to gf
682 */
683 if (remote_gf) {
684 struct getinfo_fid2path *ori_gf;
685 char *ptr;
686
687 ori_gf = (struct getinfo_fid2path *)karg;
688 if (strlen(ori_gf->gf_path) +
689 strlen(gf->gf_path) > ori_gf->gf_pathlen) {
690 rc = -EOVERFLOW;
691 goto out_fid2path;
692 }
693
694 ptr = ori_gf->gf_path;
695
696 memmove(ptr + strlen(gf->gf_path) + 1, ptr,
697 strlen(ori_gf->gf_path));
698
699 strncpy(ptr, gf->gf_path, strlen(gf->gf_path));
700 ptr += strlen(gf->gf_path);
701 *ptr = '/';
702 }
703
704 CDEBUG(D_INFO, "%s: get path %s "DFID" rec: %llu ln: %u\n",
705 tgt->ltd_exp->exp_obd->obd_name,
706 gf->gf_path, PFID(&gf->gf_fid), gf->gf_recno,
707 gf->gf_linkno);
708
709 if (rc == 0)
710 goto out_fid2path;
711
712 /* sigh, has to go to another MDT to do path building further */
713 if (!remote_gf) {
714 remote_gf_size = sizeof(*remote_gf) + PATH_MAX;
715 remote_gf = kzalloc(remote_gf_size, GFP_NOFS);
716 if (!remote_gf) {
717 rc = -ENOMEM;
718 goto out_fid2path;
719 }
720 remote_gf->gf_pathlen = PATH_MAX;
721 }
722
723 if (!fid_is_sane(&gf->gf_fid)) {
724 CERROR("%s: invalid FID "DFID": rc = %d\n",
725 tgt->ltd_exp->exp_obd->obd_name,
726 PFID(&gf->gf_fid), -EINVAL);
727 rc = -EINVAL;
728 goto out_fid2path;
729 }
730
731 tgt = lmv_find_target(lmv, &gf->gf_fid);
732 if (IS_ERR(tgt)) {
733 rc = -EINVAL;
734 goto out_fid2path;
735 }
736
737 remote_gf->gf_fid = gf->gf_fid;
738 remote_gf->gf_recno = -1;
739 remote_gf->gf_linkno = -1;
740 memset(remote_gf->gf_path, 0, remote_gf->gf_pathlen);
741 gf = remote_gf;
742 goto repeat_fid2path;
743
744 out_fid2path:
745 kfree(remote_gf);
746 return rc;
747 }
748
749 static int lmv_hsm_req_count(struct lmv_obd *lmv,
750 const struct hsm_user_request *hur,
751 const struct lmv_tgt_desc *tgt_mds)
752 {
753 int i, nr = 0;
754 struct lmv_tgt_desc *curr_tgt;
755
756 /* count how many requests must be sent to the given target */
757 for (i = 0; i < hur->hur_request.hr_itemcount; i++) {
758 curr_tgt = lmv_find_target(lmv, &hur->hur_user_item[i].hui_fid);
759 if (obd_uuid_equals(&curr_tgt->ltd_uuid, &tgt_mds->ltd_uuid))
760 nr++;
761 }
762 return nr;
763 }
764
765 static void lmv_hsm_req_build(struct lmv_obd *lmv,
766 struct hsm_user_request *hur_in,
767 const struct lmv_tgt_desc *tgt_mds,
768 struct hsm_user_request *hur_out)
769 {
770 int i, nr_out;
771 struct lmv_tgt_desc *curr_tgt;
772
773 /* build the hsm_user_request for the given target */
774 hur_out->hur_request = hur_in->hur_request;
775 nr_out = 0;
776 for (i = 0; i < hur_in->hur_request.hr_itemcount; i++) {
777 curr_tgt = lmv_find_target(lmv,
778 &hur_in->hur_user_item[i].hui_fid);
779 if (obd_uuid_equals(&curr_tgt->ltd_uuid, &tgt_mds->ltd_uuid)) {
780 hur_out->hur_user_item[nr_out] =
781 hur_in->hur_user_item[i];
782 nr_out++;
783 }
784 }
785 hur_out->hur_request.hr_itemcount = nr_out;
786 memcpy(hur_data(hur_out), hur_data(hur_in),
787 hur_in->hur_request.hr_data_len);
788 }
789
790 static int lmv_hsm_ct_unregister(struct lmv_obd *lmv, unsigned int cmd, int len,
791 struct lustre_kernelcomm *lk,
792 void __user *uarg)
793 {
794 int rc = 0;
795 __u32 i;
796
797 /* unregister request (call from llapi_hsm_copytool_fini) */
798 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
799 /* best effort: try to clean as much as possible
800 * (continue on error)
801 */
802 obd_iocontrol(cmd, lmv->tgts[i]->ltd_exp, len, lk, uarg);
803 }
804
805 /* Whatever the result, remove copytool from kuc groups.
806 * Unreached coordinators will get EPIPE on next requests
807 * and will unregister automatically.
808 */
809 rc = libcfs_kkuc_group_rem(lk->lk_uid, lk->lk_group);
810
811 return rc;
812 }
813
814 static int lmv_hsm_ct_register(struct lmv_obd *lmv, unsigned int cmd, int len,
815 struct lustre_kernelcomm *lk, void __user *uarg)
816 {
817 struct file *filp;
818 __u32 i, j;
819 int err, rc = 0;
820 bool any_set = false;
821 struct kkuc_ct_data kcd = { 0 };
822
823 /* All or nothing: try to register to all MDS.
824 * In case of failure, unregister from previous MDS,
825 * except if it because of inactive target.
826 */
827 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
828 err = obd_iocontrol(cmd, lmv->tgts[i]->ltd_exp, len, lk, uarg);
829 if (err) {
830 if (lmv->tgts[i]->ltd_active) {
831 /* permanent error */
832 CERROR("error: iocontrol MDC %s on MDTidx %d cmd %x: err = %d\n",
833 lmv->tgts[i]->ltd_uuid.uuid,
834 i, cmd, err);
835 rc = err;
836 lk->lk_flags |= LK_FLG_STOP;
837 /* unregister from previous MDS */
838 for (j = 0; j < i; j++)
839 obd_iocontrol(cmd,
840 lmv->tgts[j]->ltd_exp,
841 len, lk, uarg);
842 return rc;
843 }
844 /* else: transient error.
845 * kuc will register to the missing MDT when it is back
846 */
847 } else {
848 any_set = true;
849 }
850 }
851
852 if (!any_set)
853 /* no registration done: return error */
854 return -ENOTCONN;
855
856 /* at least one registration done, with no failure */
857 filp = fget(lk->lk_wfd);
858 if (!filp)
859 return -EBADF;
860
861 kcd.kcd_magic = KKUC_CT_DATA_MAGIC;
862 kcd.kcd_uuid = lmv->cluuid;
863 kcd.kcd_archive = lk->lk_data;
864
865 rc = libcfs_kkuc_group_add(filp, lk->lk_uid, lk->lk_group,
866 &kcd, sizeof(kcd));
867 if (rc) {
868 if (filp)
869 fput(filp);
870 }
871
872 return rc;
873 }
874
875 static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp,
876 int len, void *karg, void __user *uarg)
877 {
878 struct obd_device *obddev = class_exp2obd(exp);
879 struct lmv_obd *lmv = &obddev->u.lmv;
880 int i = 0;
881 int rc = 0;
882 int set = 0;
883 int count = lmv->desc.ld_tgt_count;
884
885 if (count == 0)
886 return -ENOTTY;
887
888 switch (cmd) {
889 case IOC_OBD_STATFS: {
890 struct obd_ioctl_data *data = karg;
891 struct obd_device *mdc_obd;
892 struct obd_statfs stat_buf = {0};
893 __u32 index;
894
895 memcpy(&index, data->ioc_inlbuf2, sizeof(__u32));
896 if (index >= count)
897 return -ENODEV;
898
899 if (!lmv->tgts[index] || lmv->tgts[index]->ltd_active == 0)
900 return -ENODATA;
901
902 mdc_obd = class_exp2obd(lmv->tgts[index]->ltd_exp);
903 if (!mdc_obd)
904 return -EINVAL;
905
906 /* copy UUID */
907 if (copy_to_user(data->ioc_pbuf2, obd2cli_tgt(mdc_obd),
908 min((int)data->ioc_plen2,
909 (int)sizeof(struct obd_uuid))))
910 return -EFAULT;
911
912 rc = obd_statfs(NULL, lmv->tgts[index]->ltd_exp, &stat_buf,
913 cfs_time_shift_64(-OBD_STATFS_CACHE_SECONDS),
914 0);
915 if (rc)
916 return rc;
917 if (copy_to_user(data->ioc_pbuf1, &stat_buf,
918 min((int)data->ioc_plen1,
919 (int)sizeof(stat_buf))))
920 return -EFAULT;
921 break;
922 }
923 case OBD_IOC_QUOTACTL: {
924 struct if_quotactl *qctl = karg;
925 struct lmv_tgt_desc *tgt = NULL;
926 struct obd_quotactl *oqctl;
927
928 if (qctl->qc_valid == QC_MDTIDX) {
929 if (qctl->qc_idx < 0 || count <= qctl->qc_idx)
930 return -EINVAL;
931
932 tgt = lmv->tgts[qctl->qc_idx];
933 if (!tgt || !tgt->ltd_exp)
934 return -EINVAL;
935 } else if (qctl->qc_valid == QC_UUID) {
936 for (i = 0; i < count; i++) {
937 tgt = lmv->tgts[i];
938 if (!tgt)
939 continue;
940 if (!obd_uuid_equals(&tgt->ltd_uuid,
941 &qctl->obd_uuid))
942 continue;
943
944 if (!tgt->ltd_exp)
945 return -EINVAL;
946
947 break;
948 }
949 } else {
950 return -EINVAL;
951 }
952
953 if (i >= count)
954 return -EAGAIN;
955
956 LASSERT(tgt && tgt->ltd_exp);
957 oqctl = kzalloc(sizeof(*oqctl), GFP_NOFS);
958 if (!oqctl)
959 return -ENOMEM;
960
961 QCTL_COPY(oqctl, qctl);
962 rc = obd_quotactl(tgt->ltd_exp, oqctl);
963 if (rc == 0) {
964 QCTL_COPY(qctl, oqctl);
965 qctl->qc_valid = QC_MDTIDX;
966 qctl->obd_uuid = tgt->ltd_uuid;
967 }
968 kfree(oqctl);
969 break;
970 }
971 case OBD_IOC_CHANGELOG_SEND:
972 case OBD_IOC_CHANGELOG_CLEAR: {
973 struct ioc_changelog *icc = karg;
974
975 if (icc->icc_mdtindex >= count)
976 return -ENODEV;
977
978 if (!lmv->tgts[icc->icc_mdtindex] ||
979 !lmv->tgts[icc->icc_mdtindex]->ltd_exp ||
980 lmv->tgts[icc->icc_mdtindex]->ltd_active == 0)
981 return -ENODEV;
982 rc = obd_iocontrol(cmd, lmv->tgts[icc->icc_mdtindex]->ltd_exp,
983 sizeof(*icc), icc, NULL);
984 break;
985 }
986 case LL_IOC_GET_CONNECT_FLAGS: {
987 if (!lmv->tgts[0])
988 return -ENODATA;
989 rc = obd_iocontrol(cmd, lmv->tgts[0]->ltd_exp, len, karg, uarg);
990 break;
991 }
992 case OBD_IOC_FID2PATH: {
993 rc = lmv_fid2path(exp, len, karg, uarg);
994 break;
995 }
996 case LL_IOC_HSM_STATE_GET:
997 case LL_IOC_HSM_STATE_SET:
998 case LL_IOC_HSM_ACTION: {
999 struct md_op_data *op_data = karg;
1000 struct lmv_tgt_desc *tgt;
1001
1002 tgt = lmv_find_target(lmv, &op_data->op_fid1);
1003 if (IS_ERR(tgt))
1004 return PTR_ERR(tgt);
1005
1006 if (!tgt->ltd_exp)
1007 return -EINVAL;
1008
1009 rc = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg);
1010 break;
1011 }
1012 case LL_IOC_HSM_PROGRESS: {
1013 const struct hsm_progress_kernel *hpk = karg;
1014 struct lmv_tgt_desc *tgt;
1015
1016 tgt = lmv_find_target(lmv, &hpk->hpk_fid);
1017 if (IS_ERR(tgt))
1018 return PTR_ERR(tgt);
1019 rc = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg);
1020 break;
1021 }
1022 case LL_IOC_HSM_REQUEST: {
1023 struct hsm_user_request *hur = karg;
1024 struct lmv_tgt_desc *tgt;
1025 unsigned int reqcount = hur->hur_request.hr_itemcount;
1026
1027 if (reqcount == 0)
1028 return 0;
1029
1030 /* if the request is about a single fid
1031 * or if there is a single MDS, no need to split
1032 * the request.
1033 */
1034 if (reqcount == 1 || count == 1) {
1035 tgt = lmv_find_target(lmv,
1036 &hur->hur_user_item[0].hui_fid);
1037 if (IS_ERR(tgt))
1038 return PTR_ERR(tgt);
1039 rc = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg);
1040 } else {
1041 /* split fid list to their respective MDS */
1042 for (i = 0; i < count; i++) {
1043 unsigned int nr, reqlen;
1044 int rc1;
1045 struct hsm_user_request *req;
1046
1047 nr = lmv_hsm_req_count(lmv, hur, lmv->tgts[i]);
1048 if (nr == 0) /* nothing for this MDS */
1049 continue;
1050
1051 /* build a request with fids for this MDS */
1052 reqlen = offsetof(typeof(*hur),
1053 hur_user_item[nr])
1054 + hur->hur_request.hr_data_len;
1055 req = libcfs_kvzalloc(reqlen, GFP_NOFS);
1056 if (!req)
1057 return -ENOMEM;
1058
1059 lmv_hsm_req_build(lmv, hur, lmv->tgts[i], req);
1060
1061 rc1 = obd_iocontrol(cmd, lmv->tgts[i]->ltd_exp,
1062 reqlen, req, uarg);
1063 if (rc1 != 0 && rc == 0)
1064 rc = rc1;
1065 kvfree(req);
1066 }
1067 }
1068 break;
1069 }
1070 case LL_IOC_LOV_SWAP_LAYOUTS: {
1071 struct md_op_data *op_data = karg;
1072 struct lmv_tgt_desc *tgt1, *tgt2;
1073
1074 tgt1 = lmv_find_target(lmv, &op_data->op_fid1);
1075 if (IS_ERR(tgt1))
1076 return PTR_ERR(tgt1);
1077
1078 tgt2 = lmv_find_target(lmv, &op_data->op_fid2);
1079 if (IS_ERR(tgt2))
1080 return PTR_ERR(tgt2);
1081
1082 if (!tgt1->ltd_exp || !tgt2->ltd_exp)
1083 return -EINVAL;
1084
1085 /* only files on same MDT can have their layouts swapped */
1086 if (tgt1->ltd_idx != tgt2->ltd_idx)
1087 return -EPERM;
1088
1089 rc = obd_iocontrol(cmd, tgt1->ltd_exp, len, karg, uarg);
1090 break;
1091 }
1092 case LL_IOC_HSM_CT_START: {
1093 struct lustre_kernelcomm *lk = karg;
1094
1095 if (lk->lk_flags & LK_FLG_STOP)
1096 rc = lmv_hsm_ct_unregister(lmv, cmd, len, lk, uarg);
1097 else
1098 rc = lmv_hsm_ct_register(lmv, cmd, len, lk, uarg);
1099 break;
1100 }
1101 default:
1102 for (i = 0; i < count; i++) {
1103 struct obd_device *mdc_obd;
1104 int err;
1105
1106 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp)
1107 continue;
1108 /* ll_umount_begin() sets force flag but for lmv, not
1109 * mdc. Let's pass it through
1110 */
1111 mdc_obd = class_exp2obd(lmv->tgts[i]->ltd_exp);
1112 mdc_obd->obd_force = obddev->obd_force;
1113 err = obd_iocontrol(cmd, lmv->tgts[i]->ltd_exp, len,
1114 karg, uarg);
1115 if (err == -ENODATA && cmd == OBD_IOC_POLL_QUOTACHECK) {
1116 return err;
1117 } else if (err) {
1118 if (lmv->tgts[i]->ltd_active) {
1119 CERROR("error: iocontrol MDC %s on MDTidx %d cmd %x: err = %d\n",
1120 lmv->tgts[i]->ltd_uuid.uuid,
1121 i, cmd, err);
1122 if (!rc)
1123 rc = err;
1124 }
1125 } else
1126 set = 1;
1127 }
1128 if (!set && !rc)
1129 rc = -EIO;
1130 }
1131 return rc;
1132 }
1133
1134 /**
1135 * This is _inode_ placement policy function (not name).
1136 */
1137 static int lmv_placement_policy(struct obd_device *obd,
1138 struct md_op_data *op_data, u32 *mds)
1139 {
1140 struct lmv_obd *lmv = &obd->u.lmv;
1141
1142 LASSERT(mds);
1143
1144 if (lmv->desc.ld_tgt_count == 1) {
1145 *mds = 0;
1146 return 0;
1147 }
1148
1149 /**
1150 * If stripe_offset is provided during setdirstripe
1151 * (setdirstripe -i xx), xx MDS will be chosen.
1152 */
1153 if (op_data->op_cli_flags & CLI_SET_MEA) {
1154 struct lmv_user_md *lum;
1155
1156 lum = (struct lmv_user_md *)op_data->op_data;
1157 if (lum->lum_type == LMV_STRIPE_TYPE &&
1158 lum->lum_stripe_offset != -1) {
1159 if (lum->lum_stripe_offset >= lmv->desc.ld_tgt_count) {
1160 CERROR("%s: Stripe_offset %d > MDT count %d: rc = %d\n",
1161 obd->obd_name,
1162 lum->lum_stripe_offset,
1163 lmv->desc.ld_tgt_count, -ERANGE);
1164 return -ERANGE;
1165 }
1166 *mds = lum->lum_stripe_offset;
1167 return 0;
1168 }
1169 }
1170
1171 /* Allocate new fid on target according to operation type and parent
1172 * home mds.
1173 */
1174 *mds = op_data->op_mds;
1175 return 0;
1176 }
1177
1178 int __lmv_fid_alloc(struct lmv_obd *lmv, struct lu_fid *fid, u32 mds)
1179 {
1180 struct lmv_tgt_desc *tgt;
1181 int rc;
1182
1183 tgt = lmv_get_target(lmv, mds);
1184 if (IS_ERR(tgt))
1185 return PTR_ERR(tgt);
1186
1187 /*
1188 * New seq alloc and FLD setup should be atomic. Otherwise we may find
1189 * on server that seq in new allocated fid is not yet known.
1190 */
1191 mutex_lock(&tgt->ltd_fid_mutex);
1192
1193 if (tgt->ltd_active == 0 || !tgt->ltd_exp) {
1194 rc = -ENODEV;
1195 goto out;
1196 }
1197
1198 /*
1199 * Asking underlaying tgt layer to allocate new fid.
1200 */
1201 rc = obd_fid_alloc(tgt->ltd_exp, fid, NULL);
1202 if (rc > 0) {
1203 LASSERT(fid_is_sane(fid));
1204 rc = 0;
1205 }
1206
1207 out:
1208 mutex_unlock(&tgt->ltd_fid_mutex);
1209 return rc;
1210 }
1211
1212 int lmv_fid_alloc(struct obd_export *exp, struct lu_fid *fid,
1213 struct md_op_data *op_data)
1214 {
1215 struct obd_device *obd = class_exp2obd(exp);
1216 struct lmv_obd *lmv = &obd->u.lmv;
1217 u32 mds = 0;
1218 int rc;
1219
1220 LASSERT(op_data);
1221 LASSERT(fid);
1222
1223 rc = lmv_placement_policy(obd, op_data, &mds);
1224 if (rc) {
1225 CERROR("Can't get target for allocating fid, rc %d\n",
1226 rc);
1227 return rc;
1228 }
1229
1230 rc = __lmv_fid_alloc(lmv, fid, mds);
1231 if (rc) {
1232 CERROR("Can't alloc new fid, rc %d\n", rc);
1233 return rc;
1234 }
1235
1236 return rc;
1237 }
1238
1239 static int lmv_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
1240 {
1241 struct lmv_obd *lmv = &obd->u.lmv;
1242 struct lprocfs_static_vars lvars = { NULL };
1243 struct lmv_desc *desc;
1244 int rc;
1245
1246 if (LUSTRE_CFG_BUFLEN(lcfg, 1) < 1) {
1247 CERROR("LMV setup requires a descriptor\n");
1248 return -EINVAL;
1249 }
1250
1251 desc = (struct lmv_desc *)lustre_cfg_buf(lcfg, 1);
1252 if (sizeof(*desc) > LUSTRE_CFG_BUFLEN(lcfg, 1)) {
1253 CERROR("Lmv descriptor size wrong: %d > %d\n",
1254 (int)sizeof(*desc), LUSTRE_CFG_BUFLEN(lcfg, 1));
1255 return -EINVAL;
1256 }
1257
1258 lmv->tgts = kcalloc(32, sizeof(*lmv->tgts), GFP_NOFS);
1259 if (!lmv->tgts)
1260 return -ENOMEM;
1261 lmv->tgts_size = 32;
1262
1263 obd_str2uuid(&lmv->desc.ld_uuid, desc->ld_uuid.uuid);
1264 lmv->desc.ld_tgt_count = 0;
1265 lmv->desc.ld_active_tgt_count = 0;
1266 lmv->max_cookiesize = 0;
1267 lmv->max_def_easize = 0;
1268 lmv->max_easize = 0;
1269 lmv->lmv_placement = PLACEMENT_CHAR_POLICY;
1270
1271 spin_lock_init(&lmv->lmv_lock);
1272 mutex_init(&lmv->init_mutex);
1273
1274 lprocfs_lmv_init_vars(&lvars);
1275
1276 lprocfs_obd_setup(obd, lvars.obd_vars, lvars.sysfs_vars);
1277 rc = ldebugfs_seq_create(obd->obd_debugfs_entry, "target_obd",
1278 0444, &lmv_proc_target_fops, obd);
1279 if (rc)
1280 CWARN("%s: error adding LMV target_obd file: rc = %d\n",
1281 obd->obd_name, rc);
1282 rc = fld_client_init(&lmv->lmv_fld, obd->obd_name,
1283 LUSTRE_CLI_FLD_HASH_DHT);
1284 if (rc) {
1285 CERROR("Can't init FLD, err %d\n", rc);
1286 goto out;
1287 }
1288
1289 return 0;
1290
1291 out:
1292 return rc;
1293 }
1294
1295 static int lmv_cleanup(struct obd_device *obd)
1296 {
1297 struct lmv_obd *lmv = &obd->u.lmv;
1298
1299 fld_client_fini(&lmv->lmv_fld);
1300 if (lmv->tgts) {
1301 int i;
1302
1303 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
1304 if (!lmv->tgts[i])
1305 continue;
1306 lmv_del_target(lmv, i);
1307 }
1308 kfree(lmv->tgts);
1309 lmv->tgts_size = 0;
1310 }
1311 return 0;
1312 }
1313
1314 static int lmv_process_config(struct obd_device *obd, u32 len, void *buf)
1315 {
1316 struct lustre_cfg *lcfg = buf;
1317 struct obd_uuid obd_uuid;
1318 int gen;
1319 __u32 index;
1320 int rc;
1321
1322 switch (lcfg->lcfg_command) {
1323 case LCFG_ADD_MDC:
1324 /* modify_mdc_tgts add 0:lustre-clilmv 1:lustre-MDT0000_UUID
1325 * 2:0 3:1 4:lustre-MDT0000-mdc_UUID
1326 */
1327 if (LUSTRE_CFG_BUFLEN(lcfg, 1) > sizeof(obd_uuid.uuid)) {
1328 rc = -EINVAL;
1329 goto out;
1330 }
1331
1332 obd_str2uuid(&obd_uuid, lustre_cfg_buf(lcfg, 1));
1333
1334 if (sscanf(lustre_cfg_buf(lcfg, 2), "%d", &index) != 1) {
1335 rc = -EINVAL;
1336 goto out;
1337 }
1338 if (sscanf(lustre_cfg_buf(lcfg, 3), "%d", &gen) != 1) {
1339 rc = -EINVAL;
1340 goto out;
1341 }
1342 rc = lmv_add_target(obd, &obd_uuid, index, gen);
1343 goto out;
1344 default:
1345 CERROR("Unknown command: %d\n", lcfg->lcfg_command);
1346 rc = -EINVAL;
1347 goto out;
1348 }
1349 out:
1350 return rc;
1351 }
1352
1353 static int lmv_statfs(const struct lu_env *env, struct obd_export *exp,
1354 struct obd_statfs *osfs, __u64 max_age, __u32 flags)
1355 {
1356 struct obd_device *obd = class_exp2obd(exp);
1357 struct lmv_obd *lmv = &obd->u.lmv;
1358 struct obd_statfs *temp;
1359 int rc = 0;
1360 int i;
1361
1362 rc = lmv_check_connect(obd);
1363 if (rc)
1364 return rc;
1365
1366 temp = kzalloc(sizeof(*temp), GFP_NOFS);
1367 if (!temp)
1368 return -ENOMEM;
1369
1370 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
1371 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp)
1372 continue;
1373
1374 rc = obd_statfs(env, lmv->tgts[i]->ltd_exp, temp,
1375 max_age, flags);
1376 if (rc) {
1377 CERROR("can't stat MDS #%d (%s), error %d\n", i,
1378 lmv->tgts[i]->ltd_exp->exp_obd->obd_name,
1379 rc);
1380 goto out_free_temp;
1381 }
1382
1383 if (i == 0) {
1384 *osfs = *temp;
1385 /* If the statfs is from mount, it will needs
1386 * retrieve necessary information from MDT0.
1387 * i.e. mount does not need the merged osfs
1388 * from all of MDT.
1389 * And also clients can be mounted as long as
1390 * MDT0 is in service
1391 */
1392 if (flags & OBD_STATFS_FOR_MDT0)
1393 goto out_free_temp;
1394 } else {
1395 osfs->os_bavail += temp->os_bavail;
1396 osfs->os_blocks += temp->os_blocks;
1397 osfs->os_ffree += temp->os_ffree;
1398 osfs->os_files += temp->os_files;
1399 }
1400 }
1401
1402 out_free_temp:
1403 kfree(temp);
1404 return rc;
1405 }
1406
1407 static int lmv_getstatus(struct obd_export *exp,
1408 struct lu_fid *fid)
1409 {
1410 struct obd_device *obd = exp->exp_obd;
1411 struct lmv_obd *lmv = &obd->u.lmv;
1412 int rc;
1413
1414 rc = lmv_check_connect(obd);
1415 if (rc)
1416 return rc;
1417
1418 rc = md_getstatus(lmv->tgts[0]->ltd_exp, fid);
1419 return rc;
1420 }
1421
1422 static int lmv_getxattr(struct obd_export *exp, const struct lu_fid *fid,
1423 u64 valid, const char *name,
1424 const char *input, int input_size, int output_size,
1425 int flags, struct ptlrpc_request **request)
1426 {
1427 struct obd_device *obd = exp->exp_obd;
1428 struct lmv_obd *lmv = &obd->u.lmv;
1429 struct lmv_tgt_desc *tgt;
1430 int rc;
1431
1432 rc = lmv_check_connect(obd);
1433 if (rc)
1434 return rc;
1435
1436 tgt = lmv_find_target(lmv, fid);
1437 if (IS_ERR(tgt))
1438 return PTR_ERR(tgt);
1439
1440 rc = md_getxattr(tgt->ltd_exp, fid, valid, name, input,
1441 input_size, output_size, flags, request);
1442
1443 return rc;
1444 }
1445
1446 static int lmv_setxattr(struct obd_export *exp, const struct lu_fid *fid,
1447 u64 valid, const char *name,
1448 const char *input, int input_size, int output_size,
1449 int flags, __u32 suppgid,
1450 struct ptlrpc_request **request)
1451 {
1452 struct obd_device *obd = exp->exp_obd;
1453 struct lmv_obd *lmv = &obd->u.lmv;
1454 struct lmv_tgt_desc *tgt;
1455 int rc;
1456
1457 rc = lmv_check_connect(obd);
1458 if (rc)
1459 return rc;
1460
1461 tgt = lmv_find_target(lmv, fid);
1462 if (IS_ERR(tgt))
1463 return PTR_ERR(tgt);
1464
1465 rc = md_setxattr(tgt->ltd_exp, fid, valid, name, input,
1466 input_size, output_size, flags, suppgid,
1467 request);
1468
1469 return rc;
1470 }
1471
1472 static int lmv_getattr(struct obd_export *exp, struct md_op_data *op_data,
1473 struct ptlrpc_request **request)
1474 {
1475 struct obd_device *obd = exp->exp_obd;
1476 struct lmv_obd *lmv = &obd->u.lmv;
1477 struct lmv_tgt_desc *tgt;
1478 int rc;
1479
1480 rc = lmv_check_connect(obd);
1481 if (rc)
1482 return rc;
1483
1484 tgt = lmv_find_target(lmv, &op_data->op_fid1);
1485 if (IS_ERR(tgt))
1486 return PTR_ERR(tgt);
1487
1488 if (op_data->op_flags & MF_GET_MDT_IDX) {
1489 op_data->op_mds = tgt->ltd_idx;
1490 return 0;
1491 }
1492
1493 rc = md_getattr(tgt->ltd_exp, op_data, request);
1494
1495 return rc;
1496 }
1497
1498 static int lmv_null_inode(struct obd_export *exp, const struct lu_fid *fid)
1499 {
1500 struct obd_device *obd = exp->exp_obd;
1501 struct lmv_obd *lmv = &obd->u.lmv;
1502 int i;
1503 int rc;
1504
1505 rc = lmv_check_connect(obd);
1506 if (rc)
1507 return rc;
1508
1509 CDEBUG(D_INODE, "CBDATA for "DFID"\n", PFID(fid));
1510
1511 /*
1512 * With DNE every object can have two locks in different namespaces:
1513 * lookup lock in space of MDT storing direntry and update/open lock in
1514 * space of MDT storing inode.
1515 */
1516 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
1517 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp)
1518 continue;
1519 md_null_inode(lmv->tgts[i]->ltd_exp, fid);
1520 }
1521
1522 return 0;
1523 }
1524
1525 static int lmv_find_cbdata(struct obd_export *exp, const struct lu_fid *fid,
1526 ldlm_iterator_t it, void *data)
1527 {
1528 struct obd_device *obd = exp->exp_obd;
1529 struct lmv_obd *lmv = &obd->u.lmv;
1530 int i;
1531 int rc;
1532
1533 rc = lmv_check_connect(obd);
1534 if (rc)
1535 return rc;
1536
1537 CDEBUG(D_INODE, "CBDATA for "DFID"\n", PFID(fid));
1538
1539 /*
1540 * With DNE every object can have two locks in different namespaces:
1541 * lookup lock in space of MDT storing direntry and update/open lock in
1542 * space of MDT storing inode.
1543 */
1544 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
1545 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp)
1546 continue;
1547 rc = md_find_cbdata(lmv->tgts[i]->ltd_exp, fid, it, data);
1548 if (rc)
1549 return rc;
1550 }
1551
1552 return rc;
1553 }
1554
1555 static int lmv_close(struct obd_export *exp, struct md_op_data *op_data,
1556 struct md_open_data *mod, struct ptlrpc_request **request)
1557 {
1558 struct obd_device *obd = exp->exp_obd;
1559 struct lmv_obd *lmv = &obd->u.lmv;
1560 struct lmv_tgt_desc *tgt;
1561 int rc;
1562
1563 rc = lmv_check_connect(obd);
1564 if (rc)
1565 return rc;
1566
1567 tgt = lmv_find_target(lmv, &op_data->op_fid1);
1568 if (IS_ERR(tgt))
1569 return PTR_ERR(tgt);
1570
1571 CDEBUG(D_INODE, "CLOSE "DFID"\n", PFID(&op_data->op_fid1));
1572 rc = md_close(tgt->ltd_exp, op_data, mod, request);
1573 return rc;
1574 }
1575
1576 struct lmv_tgt_desc
1577 *lmv_locate_mds(struct lmv_obd *lmv, struct md_op_data *op_data,
1578 struct lu_fid *fid)
1579 {
1580 struct lmv_tgt_desc *tgt;
1581
1582 tgt = lmv_find_target(lmv, fid);
1583 if (IS_ERR(tgt))
1584 return tgt;
1585
1586 op_data->op_mds = tgt->ltd_idx;
1587
1588 return tgt;
1589 }
1590
1591 static int lmv_create(struct obd_export *exp, struct md_op_data *op_data,
1592 const void *data, int datalen, int mode, __u32 uid,
1593 __u32 gid, cfs_cap_t cap_effective, __u64 rdev,
1594 struct ptlrpc_request **request)
1595 {
1596 struct obd_device *obd = exp->exp_obd;
1597 struct lmv_obd *lmv = &obd->u.lmv;
1598 struct lmv_tgt_desc *tgt;
1599 int rc;
1600
1601 rc = lmv_check_connect(obd);
1602 if (rc)
1603 return rc;
1604
1605 if (!lmv->desc.ld_active_tgt_count)
1606 return -EIO;
1607
1608 tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1);
1609 if (IS_ERR(tgt))
1610 return PTR_ERR(tgt);
1611
1612 rc = lmv_fid_alloc(exp, &op_data->op_fid2, op_data);
1613 if (rc)
1614 return rc;
1615
1616 CDEBUG(D_INODE, "CREATE '%*s' on "DFID" -> mds #%x\n",
1617 op_data->op_namelen, op_data->op_name, PFID(&op_data->op_fid1),
1618 op_data->op_mds);
1619
1620 op_data->op_flags |= MF_MDC_CANCEL_FID1;
1621 rc = md_create(tgt->ltd_exp, op_data, data, datalen, mode, uid, gid,
1622 cap_effective, rdev, request);
1623
1624 if (rc == 0) {
1625 if (!*request)
1626 return rc;
1627 CDEBUG(D_INODE, "Created - "DFID"\n", PFID(&op_data->op_fid2));
1628 }
1629 return rc;
1630 }
1631
1632 static int lmv_done_writing(struct obd_export *exp,
1633 struct md_op_data *op_data,
1634 struct md_open_data *mod)
1635 {
1636 struct obd_device *obd = exp->exp_obd;
1637 struct lmv_obd *lmv = &obd->u.lmv;
1638 struct lmv_tgt_desc *tgt;
1639 int rc;
1640
1641 rc = lmv_check_connect(obd);
1642 if (rc)
1643 return rc;
1644
1645 tgt = lmv_find_target(lmv, &op_data->op_fid1);
1646 if (IS_ERR(tgt))
1647 return PTR_ERR(tgt);
1648
1649 rc = md_done_writing(tgt->ltd_exp, op_data, mod);
1650 return rc;
1651 }
1652
1653 static int
1654 lmv_enqueue_remote(struct obd_export *exp, struct ldlm_enqueue_info *einfo,
1655 struct lookup_intent *it, struct md_op_data *op_data,
1656 struct lustre_handle *lockh, void *lmm, int lmmsize,
1657 __u64 extra_lock_flags)
1658 {
1659 struct ptlrpc_request *req = it->d.lustre.it_data;
1660 struct obd_device *obd = exp->exp_obd;
1661 struct lmv_obd *lmv = &obd->u.lmv;
1662 struct lustre_handle plock;
1663 struct lmv_tgt_desc *tgt;
1664 struct md_op_data *rdata;
1665 struct lu_fid fid1;
1666 struct mdt_body *body;
1667 int rc = 0;
1668 int pmode;
1669
1670 body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY);
1671
1672 if (!(body->valid & OBD_MD_MDS))
1673 return 0;
1674
1675 CDEBUG(D_INODE, "REMOTE_ENQUEUE '%s' on "DFID" -> "DFID"\n",
1676 LL_IT2STR(it), PFID(&op_data->op_fid1), PFID(&body->fid1));
1677
1678 /*
1679 * We got LOOKUP lock, but we really need attrs.
1680 */
1681 pmode = it->d.lustre.it_lock_mode;
1682 LASSERT(pmode != 0);
1683 memcpy(&plock, lockh, sizeof(plock));
1684 it->d.lustre.it_lock_mode = 0;
1685 it->d.lustre.it_data = NULL;
1686 fid1 = body->fid1;
1687
1688 ptlrpc_req_finished(req);
1689
1690 tgt = lmv_find_target(lmv, &fid1);
1691 if (IS_ERR(tgt)) {
1692 rc = PTR_ERR(tgt);
1693 goto out;
1694 }
1695
1696 rdata = kzalloc(sizeof(*rdata), GFP_NOFS);
1697 if (!rdata) {
1698 rc = -ENOMEM;
1699 goto out;
1700 }
1701
1702 rdata->op_fid1 = fid1;
1703 rdata->op_bias = MDS_CROSS_REF;
1704
1705 rc = md_enqueue(tgt->ltd_exp, einfo, it, rdata, lockh,
1706 lmm, lmmsize, NULL, extra_lock_flags);
1707 kfree(rdata);
1708 out:
1709 ldlm_lock_decref(&plock, pmode);
1710 return rc;
1711 }
1712
1713 static int
1714 lmv_enqueue(struct obd_export *exp, struct ldlm_enqueue_info *einfo,
1715 struct lookup_intent *it, struct md_op_data *op_data,
1716 struct lustre_handle *lockh, void *lmm, int lmmsize,
1717 struct ptlrpc_request **req, __u64 extra_lock_flags)
1718 {
1719 struct obd_device *obd = exp->exp_obd;
1720 struct lmv_obd *lmv = &obd->u.lmv;
1721 struct lmv_tgt_desc *tgt;
1722 int rc;
1723
1724 rc = lmv_check_connect(obd);
1725 if (rc)
1726 return rc;
1727
1728 CDEBUG(D_INODE, "ENQUEUE '%s' on "DFID"\n",
1729 LL_IT2STR(it), PFID(&op_data->op_fid1));
1730
1731 tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1);
1732 if (IS_ERR(tgt))
1733 return PTR_ERR(tgt);
1734
1735 CDEBUG(D_INODE, "ENQUEUE '%s' on "DFID" -> mds #%d\n",
1736 LL_IT2STR(it), PFID(&op_data->op_fid1), tgt->ltd_idx);
1737
1738 rc = md_enqueue(tgt->ltd_exp, einfo, it, op_data, lockh,
1739 lmm, lmmsize, req, extra_lock_flags);
1740
1741 if (rc == 0 && it && it->it_op == IT_OPEN) {
1742 rc = lmv_enqueue_remote(exp, einfo, it, op_data, lockh,
1743 lmm, lmmsize, extra_lock_flags);
1744 }
1745 return rc;
1746 }
1747
1748 static int
1749 lmv_getattr_name(struct obd_export *exp, struct md_op_data *op_data,
1750 struct ptlrpc_request **request)
1751 {
1752 struct ptlrpc_request *req = NULL;
1753 struct obd_device *obd = exp->exp_obd;
1754 struct lmv_obd *lmv = &obd->u.lmv;
1755 struct lmv_tgt_desc *tgt;
1756 struct mdt_body *body;
1757 int rc;
1758
1759 rc = lmv_check_connect(obd);
1760 if (rc)
1761 return rc;
1762
1763 tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1);
1764 if (IS_ERR(tgt))
1765 return PTR_ERR(tgt);
1766
1767 CDEBUG(D_INODE, "GETATTR_NAME for %*s on "DFID" -> mds #%d\n",
1768 op_data->op_namelen, op_data->op_name, PFID(&op_data->op_fid1),
1769 tgt->ltd_idx);
1770
1771 rc = md_getattr_name(tgt->ltd_exp, op_data, request);
1772 if (rc != 0)
1773 return rc;
1774
1775 body = req_capsule_server_get(&(*request)->rq_pill,
1776 &RMF_MDT_BODY);
1777
1778 if (body->valid & OBD_MD_MDS) {
1779 struct lu_fid rid = body->fid1;
1780
1781 CDEBUG(D_INODE, "Request attrs for "DFID"\n",
1782 PFID(&rid));
1783
1784 tgt = lmv_find_target(lmv, &rid);
1785 if (IS_ERR(tgt)) {
1786 ptlrpc_req_finished(*request);
1787 return PTR_ERR(tgt);
1788 }
1789
1790 op_data->op_fid1 = rid;
1791 op_data->op_valid |= OBD_MD_FLCROSSREF;
1792 op_data->op_namelen = 0;
1793 op_data->op_name = NULL;
1794 rc = md_getattr_name(tgt->ltd_exp, op_data, &req);
1795 ptlrpc_req_finished(*request);
1796 *request = req;
1797 }
1798
1799 return rc;
1800 }
1801
1802 #define md_op_data_fid(op_data, fl) \
1803 (fl == MF_MDC_CANCEL_FID1 ? &op_data->op_fid1 : \
1804 fl == MF_MDC_CANCEL_FID2 ? &op_data->op_fid2 : \
1805 fl == MF_MDC_CANCEL_FID3 ? &op_data->op_fid3 : \
1806 fl == MF_MDC_CANCEL_FID4 ? &op_data->op_fid4 : \
1807 NULL)
1808
1809 static int lmv_early_cancel(struct obd_export *exp, struct md_op_data *op_data,
1810 int op_tgt, enum ldlm_mode mode, int bits,
1811 int flag)
1812 {
1813 struct lu_fid *fid = md_op_data_fid(op_data, flag);
1814 struct obd_device *obd = exp->exp_obd;
1815 struct lmv_obd *lmv = &obd->u.lmv;
1816 struct lmv_tgt_desc *tgt;
1817 ldlm_policy_data_t policy = { {0} };
1818 int rc = 0;
1819
1820 if (!fid_is_sane(fid))
1821 return 0;
1822
1823 tgt = lmv_find_target(lmv, fid);
1824 if (IS_ERR(tgt))
1825 return PTR_ERR(tgt);
1826
1827 if (tgt->ltd_idx != op_tgt) {
1828 CDEBUG(D_INODE, "EARLY_CANCEL on "DFID"\n", PFID(fid));
1829 policy.l_inodebits.bits = bits;
1830 rc = md_cancel_unused(tgt->ltd_exp, fid, &policy,
1831 mode, LCF_ASYNC, NULL);
1832 } else {
1833 CDEBUG(D_INODE,
1834 "EARLY_CANCEL skip operation target %d on "DFID"\n",
1835 op_tgt, PFID(fid));
1836 op_data->op_flags |= flag;
1837 rc = 0;
1838 }
1839
1840 return rc;
1841 }
1842
1843 /*
1844 * llite passes fid of an target inode in op_data->op_fid1 and id of directory in
1845 * op_data->op_fid2
1846 */
1847 static int lmv_link(struct obd_export *exp, struct md_op_data *op_data,
1848 struct ptlrpc_request **request)
1849 {
1850 struct obd_device *obd = exp->exp_obd;
1851 struct lmv_obd *lmv = &obd->u.lmv;
1852 struct lmv_tgt_desc *tgt;
1853 int rc;
1854
1855 rc = lmv_check_connect(obd);
1856 if (rc)
1857 return rc;
1858
1859 LASSERT(op_data->op_namelen != 0);
1860
1861 CDEBUG(D_INODE, "LINK "DFID":%*s to "DFID"\n",
1862 PFID(&op_data->op_fid2), op_data->op_namelen,
1863 op_data->op_name, PFID(&op_data->op_fid1));
1864
1865 op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid());
1866 op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid());
1867 op_data->op_cap = cfs_curproc_cap_pack();
1868 tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid2);
1869 if (IS_ERR(tgt))
1870 return PTR_ERR(tgt);
1871
1872 /*
1873 * Cancel UPDATE lock on child (fid1).
1874 */
1875 op_data->op_flags |= MF_MDC_CANCEL_FID2;
1876 rc = lmv_early_cancel(exp, op_data, tgt->ltd_idx, LCK_EX,
1877 MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID1);
1878 if (rc != 0)
1879 return rc;
1880
1881 rc = md_link(tgt->ltd_exp, op_data, request);
1882
1883 return rc;
1884 }
1885
1886 static int lmv_rename(struct obd_export *exp, struct md_op_data *op_data,
1887 const char *old, int oldlen, const char *new, int newlen,
1888 struct ptlrpc_request **request)
1889 {
1890 struct obd_device *obd = exp->exp_obd;
1891 struct lmv_obd *lmv = &obd->u.lmv;
1892 struct lmv_tgt_desc *src_tgt;
1893 struct lmv_tgt_desc *tgt_tgt;
1894 int rc;
1895
1896 LASSERT(oldlen != 0);
1897
1898 CDEBUG(D_INODE, "RENAME %*s in "DFID" to %*s in "DFID"\n",
1899 oldlen, old, PFID(&op_data->op_fid1),
1900 newlen, new, PFID(&op_data->op_fid2));
1901
1902 rc = lmv_check_connect(obd);
1903 if (rc)
1904 return rc;
1905
1906 op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid());
1907 op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid());
1908 op_data->op_cap = cfs_curproc_cap_pack();
1909 src_tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1);
1910 if (IS_ERR(src_tgt))
1911 return PTR_ERR(src_tgt);
1912
1913 tgt_tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid2);
1914 if (IS_ERR(tgt_tgt))
1915 return PTR_ERR(tgt_tgt);
1916 /*
1917 * LOOKUP lock on src child (fid3) should also be cancelled for
1918 * src_tgt in mdc_rename.
1919 */
1920 op_data->op_flags |= MF_MDC_CANCEL_FID1 | MF_MDC_CANCEL_FID3;
1921
1922 /*
1923 * Cancel UPDATE locks on tgt parent (fid2), tgt_tgt is its
1924 * own target.
1925 */
1926 rc = lmv_early_cancel(exp, op_data, src_tgt->ltd_idx,
1927 LCK_EX, MDS_INODELOCK_UPDATE,
1928 MF_MDC_CANCEL_FID2);
1929
1930 /*
1931 * Cancel LOOKUP locks on tgt child (fid4) for parent tgt_tgt.
1932 */
1933 if (rc == 0) {
1934 rc = lmv_early_cancel(exp, op_data, src_tgt->ltd_idx,
1935 LCK_EX, MDS_INODELOCK_LOOKUP,
1936 MF_MDC_CANCEL_FID4);
1937 }
1938
1939 /*
1940 * Cancel all the locks on tgt child (fid4).
1941 */
1942 if (rc == 0)
1943 rc = lmv_early_cancel(exp, op_data, src_tgt->ltd_idx,
1944 LCK_EX, MDS_INODELOCK_FULL,
1945 MF_MDC_CANCEL_FID4);
1946
1947 if (rc == 0)
1948 rc = md_rename(src_tgt->ltd_exp, op_data, old, oldlen,
1949 new, newlen, request);
1950 return rc;
1951 }
1952
1953 static int lmv_setattr(struct obd_export *exp, struct md_op_data *op_data,
1954 void *ea, int ealen, void *ea2, int ea2len,
1955 struct ptlrpc_request **request,
1956 struct md_open_data **mod)
1957 {
1958 struct obd_device *obd = exp->exp_obd;
1959 struct lmv_obd *lmv = &obd->u.lmv;
1960 struct lmv_tgt_desc *tgt;
1961 int rc;
1962
1963 rc = lmv_check_connect(obd);
1964 if (rc)
1965 return rc;
1966
1967 CDEBUG(D_INODE, "SETATTR for "DFID", valid 0x%x\n",
1968 PFID(&op_data->op_fid1), op_data->op_attr.ia_valid);
1969
1970 op_data->op_flags |= MF_MDC_CANCEL_FID1;
1971 tgt = lmv_find_target(lmv, &op_data->op_fid1);
1972 if (IS_ERR(tgt))
1973 return PTR_ERR(tgt);
1974
1975 rc = md_setattr(tgt->ltd_exp, op_data, ea, ealen, ea2,
1976 ea2len, request, mod);
1977
1978 return rc;
1979 }
1980
1981 static int lmv_sync(struct obd_export *exp, const struct lu_fid *fid,
1982 struct ptlrpc_request **request)
1983 {
1984 struct obd_device *obd = exp->exp_obd;
1985 struct lmv_obd *lmv = &obd->u.lmv;
1986 struct lmv_tgt_desc *tgt;
1987 int rc;
1988
1989 rc = lmv_check_connect(obd);
1990 if (rc)
1991 return rc;
1992
1993 tgt = lmv_find_target(lmv, fid);
1994 if (IS_ERR(tgt))
1995 return PTR_ERR(tgt);
1996
1997 rc = md_sync(tgt->ltd_exp, fid, request);
1998 return rc;
1999 }
2000
2001 /*
2002 * Adjust a set of pages, each page containing an array of lu_dirpages,
2003 * so that each page can be used as a single logical lu_dirpage.
2004 *
2005 * A lu_dirpage is laid out as follows, where s = ldp_hash_start,
2006 * e = ldp_hash_end, f = ldp_flags, p = padding, and each "ent" is a
2007 * struct lu_dirent. It has size up to LU_PAGE_SIZE. The ldp_hash_end
2008 * value is used as a cookie to request the next lu_dirpage in a
2009 * directory listing that spans multiple pages (two in this example):
2010 * ________
2011 * | |
2012 * .|--------v------- -----.
2013 * |s|e|f|p|ent|ent| ... |ent|
2014 * '--|-------------- -----' Each CFS_PAGE contains a single
2015 * '------. lu_dirpage.
2016 * .---------v------- -----.
2017 * |s|e|f|p|ent| 0 | ... | 0 |
2018 * '----------------- -----'
2019 *
2020 * However, on hosts where the native VM page size (PAGE_CACHE_SIZE) is
2021 * larger than LU_PAGE_SIZE, a single host page may contain multiple
2022 * lu_dirpages. After reading the lu_dirpages from the MDS, the
2023 * ldp_hash_end of the first lu_dirpage refers to the one immediately
2024 * after it in the same CFS_PAGE (arrows simplified for brevity, but
2025 * in general e0==s1, e1==s2, etc.):
2026 *
2027 * .-------------------- -----.
2028 * |s0|e0|f0|p|ent|ent| ... |ent|
2029 * |---v---------------- -----|
2030 * |s1|e1|f1|p|ent|ent| ... |ent|
2031 * |---v---------------- -----| Here, each CFS_PAGE contains
2032 * ... multiple lu_dirpages.
2033 * |---v---------------- -----|
2034 * |s'|e'|f'|p|ent|ent| ... |ent|
2035 * '---|---------------- -----'
2036 * v
2037 * .----------------------------.
2038 * | next CFS_PAGE |
2039 *
2040 * This structure is transformed into a single logical lu_dirpage as follows:
2041 *
2042 * - Replace e0 with e' so the request for the next lu_dirpage gets the page
2043 * labeled 'next CFS_PAGE'.
2044 *
2045 * - Copy the LDF_COLLIDE flag from f' to f0 to correctly reflect whether
2046 * a hash collision with the next page exists.
2047 *
2048 * - Adjust the lde_reclen of the ending entry of each lu_dirpage to span
2049 * to the first entry of the next lu_dirpage.
2050 */
2051 #if PAGE_CACHE_SIZE > LU_PAGE_SIZE
2052 static void lmv_adjust_dirpages(struct page **pages, int ncfspgs, int nlupgs)
2053 {
2054 int i;
2055
2056 for (i = 0; i < ncfspgs; i++) {
2057 struct lu_dirpage *dp = kmap(pages[i]);
2058 struct lu_dirpage *first = dp;
2059 struct lu_dirent *end_dirent = NULL;
2060 struct lu_dirent *ent;
2061 __u64 hash_end = dp->ldp_hash_end;
2062 __u32 flags = dp->ldp_flags;
2063
2064 while (--nlupgs > 0) {
2065 ent = lu_dirent_start(dp);
2066 for (end_dirent = ent; ent;
2067 end_dirent = ent, ent = lu_dirent_next(ent))
2068 ;
2069
2070 /* Advance dp to next lu_dirpage. */
2071 dp = (struct lu_dirpage *)((char *)dp + LU_PAGE_SIZE);
2072
2073 /* Check if we've reached the end of the CFS_PAGE. */
2074 if (!((unsigned long)dp & ~CFS_PAGE_MASK))
2075 break;
2076
2077 /* Save the hash and flags of this lu_dirpage. */
2078 hash_end = dp->ldp_hash_end;
2079 flags = dp->ldp_flags;
2080
2081 /* Check if lu_dirpage contains no entries. */
2082 if (!end_dirent)
2083 break;
2084
2085 /* Enlarge the end entry lde_reclen from 0 to
2086 * first entry of next lu_dirpage.
2087 */
2088 LASSERT(le16_to_cpu(end_dirent->lde_reclen) == 0);
2089 end_dirent->lde_reclen =
2090 cpu_to_le16((char *)(dp->ldp_entries) -
2091 (char *)end_dirent);
2092 }
2093
2094 first->ldp_hash_end = hash_end;
2095 first->ldp_flags &= ~cpu_to_le32(LDF_COLLIDE);
2096 first->ldp_flags |= flags & cpu_to_le32(LDF_COLLIDE);
2097
2098 kunmap(pages[i]);
2099 }
2100 LASSERTF(nlupgs == 0, "left = %d", nlupgs);
2101 }
2102 #else
2103 #define lmv_adjust_dirpages(pages, ncfspgs, nlupgs) do {} while (0)
2104 #endif /* PAGE_CACHE_SIZE > LU_PAGE_SIZE */
2105
2106 static int lmv_readpage(struct obd_export *exp, struct md_op_data *op_data,
2107 struct page **pages, struct ptlrpc_request **request)
2108 {
2109 struct obd_device *obd = exp->exp_obd;
2110 struct lmv_obd *lmv = &obd->u.lmv;
2111 __u64 offset = op_data->op_offset;
2112 int rc;
2113 int ncfspgs; /* pages read in PAGE_CACHE_SIZE */
2114 int nlupgs; /* pages read in LU_PAGE_SIZE */
2115 struct lmv_tgt_desc *tgt;
2116
2117 rc = lmv_check_connect(obd);
2118 if (rc)
2119 return rc;
2120
2121 CDEBUG(D_INODE, "READPAGE at %#llx from "DFID"\n",
2122 offset, PFID(&op_data->op_fid1));
2123
2124 tgt = lmv_find_target(lmv, &op_data->op_fid1);
2125 if (IS_ERR(tgt))
2126 return PTR_ERR(tgt);
2127
2128 rc = md_readpage(tgt->ltd_exp, op_data, pages, request);
2129 if (rc != 0)
2130 return rc;
2131
2132 ncfspgs = ((*request)->rq_bulk->bd_nob_transferred + PAGE_CACHE_SIZE - 1)
2133 >> PAGE_CACHE_SHIFT;
2134 nlupgs = (*request)->rq_bulk->bd_nob_transferred >> LU_PAGE_SHIFT;
2135 LASSERT(!((*request)->rq_bulk->bd_nob_transferred & ~LU_PAGE_MASK));
2136 LASSERT(ncfspgs > 0 && ncfspgs <= op_data->op_npages);
2137
2138 CDEBUG(D_INODE, "read %d(%d)/%d pages\n", ncfspgs, nlupgs,
2139 op_data->op_npages);
2140
2141 lmv_adjust_dirpages(pages, ncfspgs, nlupgs);
2142
2143 return rc;
2144 }
2145
2146 static int lmv_unlink(struct obd_export *exp, struct md_op_data *op_data,
2147 struct ptlrpc_request **request)
2148 {
2149 struct obd_device *obd = exp->exp_obd;
2150 struct lmv_obd *lmv = &obd->u.lmv;
2151 struct lmv_tgt_desc *tgt = NULL;
2152 struct mdt_body *body;
2153 int rc;
2154
2155 rc = lmv_check_connect(obd);
2156 if (rc)
2157 return rc;
2158 retry:
2159 /* Send unlink requests to the MDT where the child is located */
2160 if (likely(!fid_is_zero(&op_data->op_fid2)))
2161 tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid2);
2162 else
2163 tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1);
2164 if (IS_ERR(tgt))
2165 return PTR_ERR(tgt);
2166
2167 op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid());
2168 op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid());
2169 op_data->op_cap = cfs_curproc_cap_pack();
2170
2171 /*
2172 * If child's fid is given, cancel unused locks for it if it is from
2173 * another export than parent.
2174 *
2175 * LOOKUP lock for child (fid3) should also be cancelled on parent
2176 * tgt_tgt in mdc_unlink().
2177 */
2178 op_data->op_flags |= MF_MDC_CANCEL_FID1 | MF_MDC_CANCEL_FID3;
2179
2180 /*
2181 * Cancel FULL locks on child (fid3).
2182 */
2183 rc = lmv_early_cancel(exp, op_data, tgt->ltd_idx, LCK_EX,
2184 MDS_INODELOCK_FULL, MF_MDC_CANCEL_FID3);
2185
2186 if (rc != 0)
2187 return rc;
2188
2189 CDEBUG(D_INODE, "unlink with fid="DFID"/"DFID" -> mds #%d\n",
2190 PFID(&op_data->op_fid1), PFID(&op_data->op_fid2), tgt->ltd_idx);
2191
2192 rc = md_unlink(tgt->ltd_exp, op_data, request);
2193 if (rc != 0 && rc != -EREMOTE)
2194 return rc;
2195
2196 body = req_capsule_server_get(&(*request)->rq_pill, &RMF_MDT_BODY);
2197 if (!body)
2198 return -EPROTO;
2199
2200 /* Not cross-ref case, just get out of here. */
2201 if (likely(!(body->valid & OBD_MD_MDS)))
2202 return 0;
2203
2204 CDEBUG(D_INODE, "%s: try unlink to another MDT for "DFID"\n",
2205 exp->exp_obd->obd_name, PFID(&body->fid1));
2206
2207 /* This is a remote object, try remote MDT, Note: it may
2208 * try more than 1 time here, Considering following case
2209 * /mnt/lustre is root on MDT0, remote1 is on MDT1
2210 * 1. Initially A does not know where remote1 is, it send
2211 * unlink RPC to MDT0, MDT0 return -EREMOTE, it will
2212 * resend unlink RPC to MDT1 (retry 1st time).
2213 *
2214 * 2. During the unlink RPC in flight,
2215 * client B mv /mnt/lustre/remote1 /mnt/lustre/remote2
2216 * and create new remote1, but on MDT0
2217 *
2218 * 3. MDT1 get unlink RPC(from A), then do remote lock on
2219 * /mnt/lustre, then lookup get fid of remote1, and find
2220 * it is remote dir again, and replay -EREMOTE again.
2221 *
2222 * 4. Then A will resend unlink RPC to MDT0. (retry 2nd times).
2223 *
2224 * In theory, it might try unlimited time here, but it should
2225 * be very rare case.
2226 */
2227 op_data->op_fid2 = body->fid1;
2228 ptlrpc_req_finished(*request);
2229 *request = NULL;
2230
2231 goto retry;
2232 }
2233
2234 static int lmv_precleanup(struct obd_device *obd, enum obd_cleanup_stage stage)
2235 {
2236 struct lmv_obd *lmv = &obd->u.lmv;
2237
2238 switch (stage) {
2239 case OBD_CLEANUP_EARLY:
2240 /* XXX: here should be calling obd_precleanup() down to
2241 * stack.
2242 */
2243 break;
2244 case OBD_CLEANUP_EXPORTS:
2245 fld_client_debugfs_fini(&lmv->lmv_fld);
2246 lprocfs_obd_cleanup(obd);
2247 break;
2248 default:
2249 break;
2250 }
2251 return 0;
2252 }
2253
2254 static int lmv_get_info(const struct lu_env *env, struct obd_export *exp,
2255 __u32 keylen, void *key, __u32 *vallen, void *val,
2256 struct lov_stripe_md *lsm)
2257 {
2258 struct obd_device *obd;
2259 struct lmv_obd *lmv;
2260 int rc = 0;
2261
2262 obd = class_exp2obd(exp);
2263 if (!obd) {
2264 CDEBUG(D_IOCTL, "Invalid client cookie %#llx\n",
2265 exp->exp_handle.h_cookie);
2266 return -EINVAL;
2267 }
2268
2269 lmv = &obd->u.lmv;
2270 if (keylen >= strlen("remote_flag") && !strcmp(key, "remote_flag")) {
2271 struct lmv_tgt_desc *tgt;
2272 int i;
2273
2274 rc = lmv_check_connect(obd);
2275 if (rc)
2276 return rc;
2277
2278 LASSERT(*vallen == sizeof(__u32));
2279 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
2280 tgt = lmv->tgts[i];
2281 /*
2282 * All tgts should be connected when this gets called.
2283 */
2284 if (!tgt || !tgt->ltd_exp)
2285 continue;
2286
2287 if (!obd_get_info(env, tgt->ltd_exp, keylen, key,
2288 vallen, val, NULL))
2289 return 0;
2290 }
2291 return -EINVAL;
2292 } else if (KEY_IS(KEY_MAX_EASIZE) ||
2293 KEY_IS(KEY_DEFAULT_EASIZE) ||
2294 KEY_IS(KEY_CONN_DATA)) {
2295 rc = lmv_check_connect(obd);
2296 if (rc)
2297 return rc;
2298
2299 /*
2300 * Forwarding this request to first MDS, it should know LOV
2301 * desc.
2302 */
2303 rc = obd_get_info(env, lmv->tgts[0]->ltd_exp, keylen, key,
2304 vallen, val, NULL);
2305 if (!rc && KEY_IS(KEY_CONN_DATA))
2306 exp->exp_connect_data = *(struct obd_connect_data *)val;
2307 return rc;
2308 } else if (KEY_IS(KEY_TGT_COUNT)) {
2309 *((int *)val) = lmv->desc.ld_tgt_count;
2310 return 0;
2311 }
2312
2313 CDEBUG(D_IOCTL, "Invalid key\n");
2314 return -EINVAL;
2315 }
2316
2317 static int lmv_set_info_async(const struct lu_env *env, struct obd_export *exp,
2318 u32 keylen, void *key, u32 vallen,
2319 void *val, struct ptlrpc_request_set *set)
2320 {
2321 struct lmv_tgt_desc *tgt;
2322 struct obd_device *obd;
2323 struct lmv_obd *lmv;
2324 int rc = 0;
2325
2326 obd = class_exp2obd(exp);
2327 if (!obd) {
2328 CDEBUG(D_IOCTL, "Invalid client cookie %#llx\n",
2329 exp->exp_handle.h_cookie);
2330 return -EINVAL;
2331 }
2332 lmv = &obd->u.lmv;
2333
2334 if (KEY_IS(KEY_READ_ONLY) || KEY_IS(KEY_FLUSH_CTX)) {
2335 int i, err = 0;
2336
2337 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
2338 tgt = lmv->tgts[i];
2339
2340 if (!tgt || !tgt->ltd_exp)
2341 continue;
2342
2343 err = obd_set_info_async(env, tgt->ltd_exp,
2344 keylen, key, vallen, val, set);
2345 if (err && rc == 0)
2346 rc = err;
2347 }
2348
2349 return rc;
2350 }
2351
2352 return -EINVAL;
2353 }
2354
2355 static int lmv_packmd(struct obd_export *exp, struct lov_mds_md **lmmp,
2356 struct lov_stripe_md *lsm)
2357 {
2358 struct obd_device *obd = class_exp2obd(exp);
2359 struct lmv_obd *lmv = &obd->u.lmv;
2360 struct lmv_stripe_md *meap;
2361 struct lmv_stripe_md *lsmp;
2362 int mea_size;
2363 int i;
2364
2365 mea_size = lmv_get_easize(lmv);
2366 if (!lmmp)
2367 return mea_size;
2368
2369 if (*lmmp && !lsm) {
2370 kvfree(*lmmp);
2371 *lmmp = NULL;
2372 return 0;
2373 }
2374
2375 if (!*lmmp) {
2376 *lmmp = libcfs_kvzalloc(mea_size, GFP_NOFS);
2377 if (!*lmmp)
2378 return -ENOMEM;
2379 }
2380
2381 if (!lsm)
2382 return mea_size;
2383
2384 lsmp = (struct lmv_stripe_md *)lsm;
2385 meap = (struct lmv_stripe_md *)*lmmp;
2386
2387 if (lsmp->mea_magic != MEA_MAGIC_LAST_CHAR &&
2388 lsmp->mea_magic != MEA_MAGIC_ALL_CHARS)
2389 return -EINVAL;
2390
2391 meap->mea_magic = cpu_to_le32(lsmp->mea_magic);
2392 meap->mea_count = cpu_to_le32(lsmp->mea_count);
2393 meap->mea_master = cpu_to_le32(lsmp->mea_master);
2394
2395 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
2396 meap->mea_ids[i] = lsmp->mea_ids[i];
2397 fid_cpu_to_le(&meap->mea_ids[i], &lsmp->mea_ids[i]);
2398 }
2399
2400 return mea_size;
2401 }
2402
2403 static int lmv_unpackmd(struct obd_export *exp, struct lov_stripe_md **lsmp,
2404 struct lov_mds_md *lmm, int lmm_size)
2405 {
2406 struct obd_device *obd = class_exp2obd(exp);
2407 struct lmv_stripe_md **tmea = (struct lmv_stripe_md **)lsmp;
2408 struct lmv_stripe_md *mea = (struct lmv_stripe_md *)lmm;
2409 struct lmv_obd *lmv = &obd->u.lmv;
2410 int mea_size;
2411 int i;
2412 __u32 magic;
2413
2414 mea_size = lmv_get_easize(lmv);
2415 if (!lsmp)
2416 return mea_size;
2417
2418 if (*lsmp && !lmm) {
2419 kvfree(*tmea);
2420 *lsmp = NULL;
2421 return 0;
2422 }
2423
2424 LASSERT(mea_size == lmm_size);
2425
2426 *tmea = libcfs_kvzalloc(mea_size, GFP_NOFS);
2427 if (!*tmea)
2428 return -ENOMEM;
2429
2430 if (!lmm)
2431 return mea_size;
2432
2433 if (mea->mea_magic == MEA_MAGIC_LAST_CHAR ||
2434 mea->mea_magic == MEA_MAGIC_ALL_CHARS ||
2435 mea->mea_magic == MEA_MAGIC_HASH_SEGMENT) {
2436 magic = le32_to_cpu(mea->mea_magic);
2437 } else {
2438 /*
2439 * Old mea is not handled here.
2440 */
2441 CERROR("Old not supportable EA is found\n");
2442 LBUG();
2443 }
2444
2445 (*tmea)->mea_magic = magic;
2446 (*tmea)->mea_count = le32_to_cpu(mea->mea_count);
2447 (*tmea)->mea_master = le32_to_cpu(mea->mea_master);
2448
2449 for (i = 0; i < (*tmea)->mea_count; i++) {
2450 (*tmea)->mea_ids[i] = mea->mea_ids[i];
2451 fid_le_to_cpu(&(*tmea)->mea_ids[i], &(*tmea)->mea_ids[i]);
2452 }
2453 return mea_size;
2454 }
2455
2456 static int lmv_cancel_unused(struct obd_export *exp, const struct lu_fid *fid,
2457 ldlm_policy_data_t *policy, enum ldlm_mode mode,
2458 enum ldlm_cancel_flags flags, void *opaque)
2459 {
2460 struct obd_device *obd = exp->exp_obd;
2461 struct lmv_obd *lmv = &obd->u.lmv;
2462 int rc = 0;
2463 int err;
2464 int i;
2465
2466 LASSERT(fid);
2467
2468 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
2469 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp ||
2470 lmv->tgts[i]->ltd_active == 0)
2471 continue;
2472
2473 err = md_cancel_unused(lmv->tgts[i]->ltd_exp, fid,
2474 policy, mode, flags, opaque);
2475 if (!rc)
2476 rc = err;
2477 }
2478 return rc;
2479 }
2480
2481 static int lmv_set_lock_data(struct obd_export *exp, __u64 *lockh, void *data,
2482 __u64 *bits)
2483 {
2484 struct lmv_obd *lmv = &exp->exp_obd->u.lmv;
2485 int rc;
2486
2487 rc = md_set_lock_data(lmv->tgts[0]->ltd_exp, lockh, data, bits);
2488 return rc;
2489 }
2490
2491 static enum ldlm_mode lmv_lock_match(struct obd_export *exp, __u64 flags,
2492 const struct lu_fid *fid,
2493 enum ldlm_type type,
2494 ldlm_policy_data_t *policy,
2495 enum ldlm_mode mode,
2496 struct lustre_handle *lockh)
2497 {
2498 struct obd_device *obd = exp->exp_obd;
2499 struct lmv_obd *lmv = &obd->u.lmv;
2500 enum ldlm_mode rc;
2501 int i;
2502
2503 CDEBUG(D_INODE, "Lock match for "DFID"\n", PFID(fid));
2504
2505 /*
2506 * With CMD every object can have two locks in different namespaces:
2507 * lookup lock in space of mds storing direntry and update/open lock in
2508 * space of mds storing inode. Thus we check all targets, not only that
2509 * one fid was created in.
2510 */
2511 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
2512 if (!lmv->tgts[i] || !lmv->tgts[i]->ltd_exp ||
2513 lmv->tgts[i]->ltd_active == 0)
2514 continue;
2515
2516 rc = md_lock_match(lmv->tgts[i]->ltd_exp, flags, fid,
2517 type, policy, mode, lockh);
2518 if (rc)
2519 return rc;
2520 }
2521
2522 return 0;
2523 }
2524
2525 static int lmv_get_lustre_md(struct obd_export *exp,
2526 struct ptlrpc_request *req,
2527 struct obd_export *dt_exp,
2528 struct obd_export *md_exp,
2529 struct lustre_md *md)
2530 {
2531 struct lmv_obd *lmv = &exp->exp_obd->u.lmv;
2532
2533 return md_get_lustre_md(lmv->tgts[0]->ltd_exp, req, dt_exp, md_exp, md);
2534 }
2535
2536 static int lmv_free_lustre_md(struct obd_export *exp, struct lustre_md *md)
2537 {
2538 struct obd_device *obd = exp->exp_obd;
2539 struct lmv_obd *lmv = &obd->u.lmv;
2540
2541 if (md->mea)
2542 obd_free_memmd(exp, (void *)&md->mea);
2543 return md_free_lustre_md(lmv->tgts[0]->ltd_exp, md);
2544 }
2545
2546 static int lmv_set_open_replay_data(struct obd_export *exp,
2547 struct obd_client_handle *och,
2548 struct lookup_intent *it)
2549 {
2550 struct obd_device *obd = exp->exp_obd;
2551 struct lmv_obd *lmv = &obd->u.lmv;
2552 struct lmv_tgt_desc *tgt;
2553
2554 tgt = lmv_find_target(lmv, &och->och_fid);
2555 if (IS_ERR(tgt))
2556 return PTR_ERR(tgt);
2557
2558 return md_set_open_replay_data(tgt->ltd_exp, och, it);
2559 }
2560
2561 static int lmv_clear_open_replay_data(struct obd_export *exp,
2562 struct obd_client_handle *och)
2563 {
2564 struct obd_device *obd = exp->exp_obd;
2565 struct lmv_obd *lmv = &obd->u.lmv;
2566 struct lmv_tgt_desc *tgt;
2567
2568 tgt = lmv_find_target(lmv, &och->och_fid);
2569 if (IS_ERR(tgt))
2570 return PTR_ERR(tgt);
2571
2572 return md_clear_open_replay_data(tgt->ltd_exp, och);
2573 }
2574
2575 static int lmv_get_remote_perm(struct obd_export *exp,
2576 const struct lu_fid *fid,
2577 __u32 suppgid, struct ptlrpc_request **request)
2578 {
2579 struct obd_device *obd = exp->exp_obd;
2580 struct lmv_obd *lmv = &obd->u.lmv;
2581 struct lmv_tgt_desc *tgt;
2582 int rc;
2583
2584 rc = lmv_check_connect(obd);
2585 if (rc)
2586 return rc;
2587
2588 tgt = lmv_find_target(lmv, fid);
2589 if (IS_ERR(tgt))
2590 return PTR_ERR(tgt);
2591
2592 rc = md_get_remote_perm(tgt->ltd_exp, fid, suppgid, request);
2593 return rc;
2594 }
2595
2596 static int lmv_intent_getattr_async(struct obd_export *exp,
2597 struct md_enqueue_info *minfo,
2598 struct ldlm_enqueue_info *einfo)
2599 {
2600 struct md_op_data *op_data = &minfo->mi_data;
2601 struct obd_device *obd = exp->exp_obd;
2602 struct lmv_obd *lmv = &obd->u.lmv;
2603 struct lmv_tgt_desc *tgt = NULL;
2604 int rc;
2605
2606 rc = lmv_check_connect(obd);
2607 if (rc)
2608 return rc;
2609
2610 tgt = lmv_find_target(lmv, &op_data->op_fid1);
2611 if (IS_ERR(tgt))
2612 return PTR_ERR(tgt);
2613
2614 rc = md_intent_getattr_async(tgt->ltd_exp, minfo, einfo);
2615 return rc;
2616 }
2617
2618 static int lmv_revalidate_lock(struct obd_export *exp, struct lookup_intent *it,
2619 struct lu_fid *fid, __u64 *bits)
2620 {
2621 struct obd_device *obd = exp->exp_obd;
2622 struct lmv_obd *lmv = &obd->u.lmv;
2623 struct lmv_tgt_desc *tgt;
2624 int rc;
2625
2626 rc = lmv_check_connect(obd);
2627 if (rc)
2628 return rc;
2629
2630 tgt = lmv_find_target(lmv, fid);
2631 if (IS_ERR(tgt))
2632 return PTR_ERR(tgt);
2633
2634 rc = md_revalidate_lock(tgt->ltd_exp, it, fid, bits);
2635 return rc;
2636 }
2637
2638 /**
2639 * For lmv, only need to send request to master MDT, and the master MDT will
2640 * process with other slave MDTs. The only exception is Q_GETOQUOTA for which
2641 * we directly fetch data from the slave MDTs.
2642 */
2643 static int lmv_quotactl(struct obd_device *unused, struct obd_export *exp,
2644 struct obd_quotactl *oqctl)
2645 {
2646 struct obd_device *obd = class_exp2obd(exp);
2647 struct lmv_obd *lmv = &obd->u.lmv;
2648 struct lmv_tgt_desc *tgt = lmv->tgts[0];
2649 int rc = 0, i;
2650 __u64 curspace, curinodes;
2651
2652 if (!lmv->desc.ld_tgt_count || !tgt->ltd_active) {
2653 CERROR("master lmv inactive\n");
2654 return -EIO;
2655 }
2656
2657 if (oqctl->qc_cmd != Q_GETOQUOTA) {
2658 rc = obd_quotactl(tgt->ltd_exp, oqctl);
2659 return rc;
2660 }
2661
2662 curspace = curinodes = 0;
2663 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
2664 int err;
2665
2666 tgt = lmv->tgts[i];
2667
2668 if (!tgt || !tgt->ltd_exp || tgt->ltd_active == 0)
2669 continue;
2670 if (!tgt->ltd_active) {
2671 CDEBUG(D_HA, "mdt %d is inactive.\n", i);
2672 continue;
2673 }
2674
2675 err = obd_quotactl(tgt->ltd_exp, oqctl);
2676 if (err) {
2677 CERROR("getquota on mdt %d failed. %d\n", i, err);
2678 if (!rc)
2679 rc = err;
2680 } else {
2681 curspace += oqctl->qc_dqblk.dqb_curspace;
2682 curinodes += oqctl->qc_dqblk.dqb_curinodes;
2683 }
2684 }
2685 oqctl->qc_dqblk.dqb_curspace = curspace;
2686 oqctl->qc_dqblk.dqb_curinodes = curinodes;
2687
2688 return rc;
2689 }
2690
2691 static int lmv_quotacheck(struct obd_device *unused, struct obd_export *exp,
2692 struct obd_quotactl *oqctl)
2693 {
2694 struct obd_device *obd = class_exp2obd(exp);
2695 struct lmv_obd *lmv = &obd->u.lmv;
2696 struct lmv_tgt_desc *tgt;
2697 int i, rc = 0;
2698
2699 for (i = 0; i < lmv->desc.ld_tgt_count; i++) {
2700 int err;
2701
2702 tgt = lmv->tgts[i];
2703 if (!tgt || !tgt->ltd_exp || !tgt->ltd_active) {
2704 CERROR("lmv idx %d inactive\n", i);
2705 return -EIO;
2706 }
2707
2708 err = obd_quotacheck(tgt->ltd_exp, oqctl);
2709 if (err && !rc)
2710 rc = err;
2711 }
2712
2713 return rc;
2714 }
2715
2716 static struct obd_ops lmv_obd_ops = {
2717 .owner = THIS_MODULE,
2718 .setup = lmv_setup,
2719 .cleanup = lmv_cleanup,
2720 .precleanup = lmv_precleanup,
2721 .process_config = lmv_process_config,
2722 .connect = lmv_connect,
2723 .disconnect = lmv_disconnect,
2724 .statfs = lmv_statfs,
2725 .get_info = lmv_get_info,
2726 .set_info_async = lmv_set_info_async,
2727 .packmd = lmv_packmd,
2728 .unpackmd = lmv_unpackmd,
2729 .notify = lmv_notify,
2730 .get_uuid = lmv_get_uuid,
2731 .iocontrol = lmv_iocontrol,
2732 .quotacheck = lmv_quotacheck,
2733 .quotactl = lmv_quotactl
2734 };
2735
2736 static struct md_ops lmv_md_ops = {
2737 .getstatus = lmv_getstatus,
2738 .null_inode = lmv_null_inode,
2739 .find_cbdata = lmv_find_cbdata,
2740 .close = lmv_close,
2741 .create = lmv_create,
2742 .done_writing = lmv_done_writing,
2743 .enqueue = lmv_enqueue,
2744 .getattr = lmv_getattr,
2745 .getxattr = lmv_getxattr,
2746 .getattr_name = lmv_getattr_name,
2747 .intent_lock = lmv_intent_lock,
2748 .link = lmv_link,
2749 .rename = lmv_rename,
2750 .setattr = lmv_setattr,
2751 .setxattr = lmv_setxattr,
2752 .sync = lmv_sync,
2753 .readpage = lmv_readpage,
2754 .unlink = lmv_unlink,
2755 .init_ea_size = lmv_init_ea_size,
2756 .cancel_unused = lmv_cancel_unused,
2757 .set_lock_data = lmv_set_lock_data,
2758 .lock_match = lmv_lock_match,
2759 .get_lustre_md = lmv_get_lustre_md,
2760 .free_lustre_md = lmv_free_lustre_md,
2761 .set_open_replay_data = lmv_set_open_replay_data,
2762 .clear_open_replay_data = lmv_clear_open_replay_data,
2763 .get_remote_perm = lmv_get_remote_perm,
2764 .intent_getattr_async = lmv_intent_getattr_async,
2765 .revalidate_lock = lmv_revalidate_lock
2766 };
2767
2768 static int __init lmv_init(void)
2769 {
2770 struct lprocfs_static_vars lvars;
2771 int rc;
2772
2773 lprocfs_lmv_init_vars(&lvars);
2774
2775 rc = class_register_type(&lmv_obd_ops, &lmv_md_ops,
2776 LUSTRE_LMV_NAME, NULL);
2777 return rc;
2778 }
2779
2780 static void lmv_exit(void)
2781 {
2782 class_unregister_type(LUSTRE_LMV_NAME);
2783 }
2784
2785 MODULE_AUTHOR("OpenSFS, Inc. <http://www.lustre.org/>");
2786 MODULE_DESCRIPTION("Lustre Logical Metadata Volume");
2787 MODULE_VERSION(LUSTRE_VERSION_STRING);
2788 MODULE_LICENSE("GPL");
2789
2790 module_init(lmv_init);
2791 module_exit(lmv_exit);
Linux kernel - Deliverables
This page took 0.130536 seconds and 5 git commands to generate.