Loading...
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 | /*
* GPL HEADER START
*
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 only,
* as published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License version 2 for more details (a copy is included
* in the LICENSE file that accompanied this code).
*
* You should have received a copy of the GNU General Public License
* version 2 along with this program; If not, see
* http://www.gnu.org/licenses/gpl-2.0.html
*
* GPL HEADER END
*/
/*
* Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
* Copyright (c) 2011, 2015, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
* Lustre is a trademark of Sun Microsystems, Inc.
*/
#define DEBUG_SUBSYSTEM S_MDC
# include <linux/module.h>
# include <linux/kernel.h>
#include "../include/obd_class.h"
#include "mdc_internal.h"
#include "../include/lustre_fid.h"
/* mdc_setattr does its own semaphore handling */
static int mdc_reint(struct ptlrpc_request *request,
struct mdc_rpc_lock *rpc_lock,
int level)
{
int rc;
request->rq_send_state = level;
mdc_get_rpc_lock(rpc_lock, NULL);
rc = ptlrpc_queue_wait(request);
mdc_put_rpc_lock(rpc_lock, NULL);
if (rc)
CDEBUG(D_INFO, "error in handling %d\n", rc);
else if (!req_capsule_server_get(&request->rq_pill, &RMF_MDT_BODY))
rc = -EPROTO;
return rc;
}
/* Find and cancel locally locks matched by inode @bits & @mode in the resource
* found by @fid. Found locks are added into @cancel list. Returns the amount of
* locks added to @cancels list.
*/
int mdc_resource_get_unused(struct obd_export *exp, const struct lu_fid *fid,
struct list_head *cancels, enum ldlm_mode mode,
__u64 bits)
{
struct ldlm_namespace *ns = exp->exp_obd->obd_namespace;
ldlm_policy_data_t policy = {};
struct ldlm_res_id res_id;
struct ldlm_resource *res;
int count;
/* Return, i.e. cancel nothing, only if ELC is supported (flag in
* export) but disabled through procfs (flag in NS).
*
* This distinguishes from a case when ELC is not supported originally,
* when we still want to cancel locks in advance and just cancel them
* locally, without sending any RPC.
*/
if (exp_connect_cancelset(exp) && !ns_connect_cancelset(ns))
return 0;
fid_build_reg_res_name(fid, &res_id);
res = ldlm_resource_get(exp->exp_obd->obd_namespace,
NULL, &res_id, 0, 0);
if (IS_ERR(res))
return 0;
LDLM_RESOURCE_ADDREF(res);
/* Initialize ibits lock policy. */
policy.l_inodebits.bits = bits;
count = ldlm_cancel_resource_local(res, cancels, &policy,
mode, 0, 0, NULL);
LDLM_RESOURCE_DELREF(res);
ldlm_resource_putref(res);
return count;
}
int mdc_setattr(struct obd_export *exp, struct md_op_data *op_data,
void *ea, size_t ealen, void *ea2, size_t ea2len,
struct ptlrpc_request **request, struct md_open_data **mod)
{
LIST_HEAD(cancels);
struct ptlrpc_request *req;
struct mdc_rpc_lock *rpc_lock;
struct obd_device *obd = exp->exp_obd;
int count = 0, rc;
__u64 bits;
bits = MDS_INODELOCK_UPDATE;
if (op_data->op_attr.ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID))
bits |= MDS_INODELOCK_LOOKUP;
if ((op_data->op_flags & MF_MDC_CANCEL_FID1) &&
(fid_is_sane(&op_data->op_fid1)))
count = mdc_resource_get_unused(exp, &op_data->op_fid1,
&cancels, LCK_EX, bits);
req = ptlrpc_request_alloc(class_exp2cliimp(exp),
&RQF_MDS_REINT_SETATTR);
if (!req) {
ldlm_lock_list_put(&cancels, l_bl_ast, count);
return -ENOMEM;
}
if ((op_data->op_flags & (MF_SOM_CHANGE | MF_EPOCH_OPEN)) == 0)
req_capsule_set_size(&req->rq_pill, &RMF_MDT_EPOCH, RCL_CLIENT,
0);
req_capsule_set_size(&req->rq_pill, &RMF_EADATA, RCL_CLIENT, ealen);
req_capsule_set_size(&req->rq_pill, &RMF_LOGCOOKIES, RCL_CLIENT,
ea2len);
rc = mdc_prep_elc_req(exp, req, MDS_REINT, &cancels, count);
if (rc) {
ptlrpc_request_free(req);
return rc;
}
rpc_lock = obd->u.cli.cl_rpc_lock;
if (op_data->op_attr.ia_valid & (ATTR_MTIME | ATTR_CTIME))
CDEBUG(D_INODE, "setting mtime %ld, ctime %ld\n",
LTIME_S(op_data->op_attr.ia_mtime),
LTIME_S(op_data->op_attr.ia_ctime));
mdc_setattr_pack(req, op_data, ea, ealen, ea2, ea2len);
ptlrpc_request_set_replen(req);
if (mod && (op_data->op_flags & MF_EPOCH_OPEN) &&
req->rq_import->imp_replayable) {
LASSERT(!*mod);
*mod = obd_mod_alloc();
if (!*mod) {
DEBUG_REQ(D_ERROR, req, "Can't allocate md_open_data");
} else {
req->rq_replay = 1;
req->rq_cb_data = *mod;
(*mod)->mod_open_req = req;
req->rq_commit_cb = mdc_commit_open;
(*mod)->mod_is_create = true;
/**
* Take an extra reference on \var mod, it protects \var
* mod from being freed on eviction (commit callback is
* called despite rq_replay flag).
* Will be put on mdc_done_writing().
*/
obd_mod_get(*mod);
}
}
rc = mdc_reint(req, rpc_lock, LUSTRE_IMP_FULL);
/* Save the obtained info in the original RPC for the replay case. */
if (rc == 0 && (op_data->op_flags & MF_EPOCH_OPEN)) {
struct mdt_ioepoch *epoch;
struct mdt_body *body;
epoch = req_capsule_client_get(&req->rq_pill, &RMF_MDT_EPOCH);
body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY);
epoch->handle = body->mbo_handle;
epoch->ioepoch = body->mbo_ioepoch;
req->rq_replay_cb = mdc_replay_open;
/** bug 3633, open may be committed and estale answer is not error */
} else if (rc == -ESTALE && (op_data->op_flags & MF_SOM_CHANGE)) {
rc = 0;
} else if (rc == -ERESTARTSYS) {
rc = 0;
}
*request = req;
if (rc && req->rq_commit_cb) {
/* Put an extra reference on \var mod on error case. */
if (mod && *mod)
obd_mod_put(*mod);
req->rq_commit_cb(req);
}
return rc;
}
int mdc_create(struct obd_export *exp, struct md_op_data *op_data,
const void *data, size_t datalen, umode_t mode,
uid_t uid, gid_t gid, cfs_cap_t cap_effective,
__u64 rdev, struct ptlrpc_request **request)
{
struct ptlrpc_request *req;
int level, rc;
int count, resends = 0;
struct obd_import *import = exp->exp_obd->u.cli.cl_import;
int generation = import->imp_generation;
LIST_HEAD(cancels);
/* For case if upper layer did not alloc fid, do it now. */
if (!fid_is_sane(&op_data->op_fid2)) {
/*
* mdc_fid_alloc() may return errno 1 in case of switch to new
* sequence, handle this.
*/
rc = mdc_fid_alloc(NULL, exp, &op_data->op_fid2, op_data);
if (rc < 0)
return rc;
}
rebuild:
count = 0;
if ((op_data->op_flags & MF_MDC_CANCEL_FID1) &&
(fid_is_sane(&op_data->op_fid1)))
count = mdc_resource_get_unused(exp, &op_data->op_fid1,
&cancels, LCK_EX,
MDS_INODELOCK_UPDATE);
req = ptlrpc_request_alloc(class_exp2cliimp(exp),
&RQF_MDS_REINT_CREATE_ACL);
if (!req) {
ldlm_lock_list_put(&cancels, l_bl_ast, count);
return -ENOMEM;
}
req_capsule_set_size(&req->rq_pill, &RMF_NAME, RCL_CLIENT,
op_data->op_namelen + 1);
req_capsule_set_size(&req->rq_pill, &RMF_EADATA, RCL_CLIENT,
data && datalen ? datalen : 0);
rc = mdc_prep_elc_req(exp, req, MDS_REINT, &cancels, count);
if (rc) {
ptlrpc_request_free(req);
return rc;
}
/*
* mdc_create_pack() fills msg->bufs[1] with name and msg->bufs[2] with
* tgt, for symlinks or lov MD data.
*/
mdc_create_pack(req, op_data, data, datalen, mode, uid,
gid, cap_effective, rdev);
ptlrpc_request_set_replen(req);
/* ask ptlrpc not to resend on EINPROGRESS since we have our own retry
* logic here
*/
req->rq_no_retry_einprogress = 1;
if (resends) {
req->rq_generation_set = 1;
req->rq_import_generation = generation;
req->rq_sent = ktime_get_real_seconds() + resends;
}
level = LUSTRE_IMP_FULL;
resend:
rc = mdc_reint(req, exp->exp_obd->u.cli.cl_rpc_lock, level);
/* Resend if we were told to. */
if (rc == -ERESTARTSYS) {
level = LUSTRE_IMP_RECOVER;
goto resend;
} else if (rc == -EINPROGRESS) {
/* Retry create infinitely until succeed or get other
* error code.
*/
ptlrpc_req_finished(req);
resends++;
CDEBUG(D_HA, "%s: resend:%d create on "DFID"/"DFID"\n",
exp->exp_obd->obd_name, resends,
PFID(&op_data->op_fid1), PFID(&op_data->op_fid2));
if (generation == import->imp_generation) {
goto rebuild;
} else {
CDEBUG(D_HA, "resend cross eviction\n");
return -EIO;
}
}
*request = req;
return rc;
}
int mdc_unlink(struct obd_export *exp, struct md_op_data *op_data,
struct ptlrpc_request **request)
{
LIST_HEAD(cancels);
struct obd_device *obd = class_exp2obd(exp);
struct ptlrpc_request *req = *request;
int count = 0, rc;
LASSERT(!req);
if ((op_data->op_flags & MF_MDC_CANCEL_FID1) &&
(fid_is_sane(&op_data->op_fid1)))
count = mdc_resource_get_unused(exp, &op_data->op_fid1,
&cancels, LCK_EX,
MDS_INODELOCK_UPDATE);
if ((op_data->op_flags & MF_MDC_CANCEL_FID3) &&
(fid_is_sane(&op_data->op_fid3)))
count += mdc_resource_get_unused(exp, &op_data->op_fid3,
&cancels, LCK_EX,
MDS_INODELOCK_FULL);
req = ptlrpc_request_alloc(class_exp2cliimp(exp),
&RQF_MDS_REINT_UNLINK);
if (!req) {
ldlm_lock_list_put(&cancels, l_bl_ast, count);
return -ENOMEM;
}
req_capsule_set_size(&req->rq_pill, &RMF_NAME, RCL_CLIENT,
op_data->op_namelen + 1);
rc = mdc_prep_elc_req(exp, req, MDS_REINT, &cancels, count);
if (rc) {
ptlrpc_request_free(req);
return rc;
}
mdc_unlink_pack(req, op_data);
req_capsule_set_size(&req->rq_pill, &RMF_MDT_MD, RCL_SERVER,
obd->u.cli.cl_default_mds_easize);
req_capsule_set_size(&req->rq_pill, &RMF_LOGCOOKIES, RCL_SERVER,
obd->u.cli.cl_default_mds_cookiesize);
ptlrpc_request_set_replen(req);
*request = req;
rc = mdc_reint(req, obd->u.cli.cl_rpc_lock, LUSTRE_IMP_FULL);
if (rc == -ERESTARTSYS)
rc = 0;
return rc;
}
int mdc_link(struct obd_export *exp, struct md_op_data *op_data,
struct ptlrpc_request **request)
{
LIST_HEAD(cancels);
struct obd_device *obd = exp->exp_obd;
struct ptlrpc_request *req;
int count = 0, rc;
if ((op_data->op_flags & MF_MDC_CANCEL_FID2) &&
(fid_is_sane(&op_data->op_fid2)))
count = mdc_resource_get_unused(exp, &op_data->op_fid2,
&cancels, LCK_EX,
MDS_INODELOCK_UPDATE);
if ((op_data->op_flags & MF_MDC_CANCEL_FID1) &&
(fid_is_sane(&op_data->op_fid1)))
count += mdc_resource_get_unused(exp, &op_data->op_fid1,
&cancels, LCK_EX,
MDS_INODELOCK_UPDATE);
req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_MDS_REINT_LINK);
if (!req) {
ldlm_lock_list_put(&cancels, l_bl_ast, count);
return -ENOMEM;
}
req_capsule_set_size(&req->rq_pill, &RMF_NAME, RCL_CLIENT,
op_data->op_namelen + 1);
rc = mdc_prep_elc_req(exp, req, MDS_REINT, &cancels, count);
if (rc) {
ptlrpc_request_free(req);
return rc;
}
mdc_link_pack(req, op_data);
ptlrpc_request_set_replen(req);
rc = mdc_reint(req, obd->u.cli.cl_rpc_lock, LUSTRE_IMP_FULL);
*request = req;
if (rc == -ERESTARTSYS)
rc = 0;
return rc;
}
int mdc_rename(struct obd_export *exp, struct md_op_data *op_data,
const char *old, size_t oldlen, const char *new, size_t newlen,
struct ptlrpc_request **request)
{
LIST_HEAD(cancels);
struct obd_device *obd = exp->exp_obd;
struct ptlrpc_request *req;
int count = 0, rc;
if ((op_data->op_flags & MF_MDC_CANCEL_FID1) &&
(fid_is_sane(&op_data->op_fid1)))
count = mdc_resource_get_unused(exp, &op_data->op_fid1,
&cancels, LCK_EX,
MDS_INODELOCK_UPDATE);
if ((op_data->op_flags & MF_MDC_CANCEL_FID2) &&
(fid_is_sane(&op_data->op_fid2)))
count += mdc_resource_get_unused(exp, &op_data->op_fid2,
&cancels, LCK_EX,
MDS_INODELOCK_UPDATE);
if ((op_data->op_flags & MF_MDC_CANCEL_FID3) &&
(fid_is_sane(&op_data->op_fid3)))
count += mdc_resource_get_unused(exp, &op_data->op_fid3,
&cancels, LCK_EX,
MDS_INODELOCK_LOOKUP);
if ((op_data->op_flags & MF_MDC_CANCEL_FID4) &&
(fid_is_sane(&op_data->op_fid4)))
count += mdc_resource_get_unused(exp, &op_data->op_fid4,
&cancels, LCK_EX,
MDS_INODELOCK_FULL);
req = ptlrpc_request_alloc(class_exp2cliimp(exp),
&RQF_MDS_REINT_RENAME);
if (!req) {
ldlm_lock_list_put(&cancels, l_bl_ast, count);
return -ENOMEM;
}
req_capsule_set_size(&req->rq_pill, &RMF_NAME, RCL_CLIENT, oldlen + 1);
req_capsule_set_size(&req->rq_pill, &RMF_SYMTGT, RCL_CLIENT,
newlen + 1);
rc = mdc_prep_elc_req(exp, req, MDS_REINT, &cancels, count);
if (rc) {
ptlrpc_request_free(req);
return rc;
}
if (exp_connect_cancelset(exp) && req)
ldlm_cli_cancel_list(&cancels, count, req, 0);
mdc_rename_pack(req, op_data, old, oldlen, new, newlen);
req_capsule_set_size(&req->rq_pill, &RMF_MDT_MD, RCL_SERVER,
obd->u.cli.cl_default_mds_easize);
req_capsule_set_size(&req->rq_pill, &RMF_LOGCOOKIES, RCL_SERVER,
obd->u.cli.cl_default_mds_cookiesize);
ptlrpc_request_set_replen(req);
rc = mdc_reint(req, obd->u.cli.cl_rpc_lock, LUSTRE_IMP_FULL);
*request = req;
if (rc == -ERESTARTSYS)
rc = 0;
return rc;
}
|