fs/ceph/mds_client.c

   1 // SPDX-License-Identifier: GPL-2.0
   2 #include <linux/ceph/ceph_debug.h>
   3
   4 #include <linux/fs.h>
   5 #include <linux/wait.h>
   6 #include <linux/slab.h>
   7 #include <linux/gfp.h>
   8 #include <linux/sched.h>
   9 #include <linux/debugfs.h>
  10 #include <linux/seq_file.h>
  11 #include <linux/ratelimit.h>
  12 #include <linux/bits.h>
  13 #include <linux/ktime.h>
  14
  15 #include "super.h"
  16 #include "mds_client.h"
  17
  18 #include <linux/ceph/ceph_features.h>
  19 #include <linux/ceph/messenger.h>
  20 #include <linux/ceph/decode.h>
  21 #include <linux/ceph/pagelist.h>
  22 #include <linux/ceph/auth.h>
  23 #include <linux/ceph/debugfs.h>
  24
  25 #define RECONNECT_MAX_SIZE (INT_MAX - PAGE_SIZE)
  26
  27 /*
  28  * A cluster of MDS (metadata server) daemons is responsible for
  29  * managing the file system namespace (the directory hierarchy and
  30  * inodes) and for coordinating shared access to storage.  Metadata is
  31  * partitioning hierarchically across a number of servers, and that
  32  * partition varies over time as the cluster adjusts the distribution
  33  * in order to balance load.
  34  *
  35  * The MDS client is primarily responsible to managing synchronous
  36  * metadata requests for operations like open, unlink, and so forth.
  37  * If there is a MDS failure, we find out about it when we (possibly
  38  * request and) receive a new MDS map, and can resubmit affected
  39  * requests.
  40  *
  41  * For the most part, though, we take advantage of a lossless
  42  * communications channel to the MDS, and do not need to worry about
  43  * timing out or resubmitting requests.
  44  *
  45  * We maintain a stateful "session" with each MDS we interact with.
  46  * Within each session, we sent periodic heartbeat messages to ensure
  47  * any capabilities or leases we have been issues remain valid.  If
  48  * the session times out and goes stale, our leases and capabilities
  49  * are no longer valid.
  50  */
  51
  52 struct ceph_reconnect_state {
  53         struct ceph_mds_session *session;
  54         int nr_caps, nr_realms;
  55         struct ceph_pagelist *pagelist;
  56         unsigned msg_version;
  57         bool allow_multi;
  58 };
  59
  60 static void __wake_requests(struct ceph_mds_client *mdsc,
  61                             struct list_head *head);
  62 static void ceph_cap_release_work(struct work_struct *work);
  63 static void ceph_cap_reclaim_work(struct work_struct *work);
  64
  65 static const struct ceph_connection_operations mds_con_ops;
  66
  67
  68 /*
  69  * mds reply parsing
  70  */
  71
  72 static int parse_reply_info_quota(void **p, void *end,
  73                                   struct ceph_mds_reply_info_in *info)
  74 {
  75         u8 struct_v, struct_compat;
  76         u32 struct_len;
  77
  78         ceph_decode_8_safe(p, end, struct_v, bad);
  79         ceph_decode_8_safe(p, end, struct_compat, bad);
  80         /* struct_v is expected to be >= 1. we only
  81          * understand encoding with struct_compat == 1. */
  82         if (!struct_v || struct_compat != 1)
  83                 goto bad;
  84         ceph_decode_32_safe(p, end, struct_len, bad);
  85         ceph_decode_need(p, end, struct_len, bad);
  86         end = *p + struct_len;
  87         ceph_decode_64_safe(p, end, info->max_bytes, bad);
  88         ceph_decode_64_safe(p, end, info->max_files, bad);
  89         *p = end;
  90         return 0;
  91 bad:
  92         return -EIO;
  93 }
  94
  95 /*
  96  * parse individual inode info
  97  */
  98 static int parse_reply_info_in(void **p, void *end,
  99                                struct ceph_mds_reply_info_in *info,
 100                                u64 features)
 101 {
 102         int err = 0;
 103         u8 struct_v = 0;
 104
 105         if (features == (u64)-1) {
 106                 u32 struct_len;
 107                 u8 struct_compat;
 108                 ceph_decode_8_safe(p, end, struct_v, bad);
 109                 ceph_decode_8_safe(p, end, struct_compat, bad);
 110                 /* struct_v is expected to be >= 1. we only understand
 111                  * encoding with struct_compat == 1. */
 112                 if (!struct_v || struct_compat != 1)
 113                         goto bad;
 114                 ceph_decode_32_safe(p, end, struct_len, bad);
 115                 ceph_decode_need(p, end, struct_len, bad);
 116                 end = *p + struct_len;
 117         }
 118
 119         ceph_decode_need(p, end, sizeof(struct ceph_mds_reply_inode), bad);
 120         info->in = *p;
 121         *p += sizeof(struct ceph_mds_reply_inode) +
 122                 sizeof(*info->in->fragtree.splits) *
 123                 le32_to_cpu(info->in->fragtree.nsplits);
 124
 125         ceph_decode_32_safe(p, end, info->symlink_len, bad);
 126         ceph_decode_need(p, end, info->symlink_len, bad);
 127         info->symlink = *p;
 128         *p += info->symlink_len;
 129
 130         ceph_decode_copy_safe(p, end, &info->dir_layout,
 131                               sizeof(info->dir_layout), bad);
 132         ceph_decode_32_safe(p, end, info->xattr_len, bad);
 133         ceph_decode_need(p, end, info->xattr_len, bad);
 134         info->xattr_data = *p;
 135         *p += info->xattr_len;
 136
 137         if (features == (u64)-1) {
 138                 /* inline data */
 139                 ceph_decode_64_safe(p, end, info->inline_version, bad);
 140                 ceph_decode_32_safe(p, end, info->inline_len, bad);
 141                 ceph_decode_need(p, end, info->inline_len, bad);
 142                 info->inline_data = *p;
 143                 *p += info->inline_len;
 144                 /* quota */
 145                 err = parse_reply_info_quota(p, end, info);
 146                 if (err < 0)
 147                         goto out_bad;
 148                 /* pool namespace */
 149                 ceph_decode_32_safe(p, end, info->pool_ns_len, bad);
 150                 if (info->pool_ns_len > 0) {
 151                         ceph_decode_need(p, end, info->pool_ns_len, bad);
 152                         info->pool_ns_data = *p;
 153                         *p += info->pool_ns_len;
 154                 }
 155
 156                 /* btime */
 157                 ceph_decode_need(p, end, sizeof(info->btime), bad);
 158                 ceph_decode_copy(p, &info->btime, sizeof(info->btime));
 159
 160                 /* change attribute */
 161                 ceph_decode_64_safe(p, end, info->change_attr, bad);
 162
 163                 /* dir pin */
 164                 if (struct_v >= 2) {
 165                         ceph_decode_32_safe(p, end, info->dir_pin, bad);
 166                 } else {
 167                         info->dir_pin = -ENODATA;
 168                 }
 169
 170                 /* snapshot birth time, remains zero for v<=2 */
 171                 if (struct_v >= 3) {
 172                         ceph_decode_need(p, end, sizeof(info->snap_btime), bad);
 173                         ceph_decode_copy(p, &info->snap_btime,
 174                                          sizeof(info->snap_btime));
 175                 } else {
 176                         memset(&info->snap_btime, 0, sizeof(info->snap_btime));
 177                 }
 178
 179                 *p = end;
 180         } else {
 181                 if (features & CEPH_FEATURE_MDS_INLINE_DATA) {
 182                         ceph_decode_64_safe(p, end, info->inline_version, bad);
 183                         ceph_decode_32_safe(p, end, info->inline_len, bad);
 184                         ceph_decode_need(p, end, info->inline_len, bad);
 185                         info->inline_data = *p;
 186                         *p += info->inline_len;
 187                 } else
 188                         info->inline_version = CEPH_INLINE_NONE;
 189
 190                 if (features & CEPH_FEATURE_MDS_QUOTA) {
 191                         err = parse_reply_info_quota(p, end, info);
 192                         if (err < 0)
 193                                 goto out_bad;
 194                 } else {
 195                         info->max_bytes = 0;
 196                         info->max_files = 0;
 197                 }
 198
 199                 info->pool_ns_len = 0;
 200                 info->pool_ns_data = NULL;
 201                 if (features & CEPH_FEATURE_FS_FILE_LAYOUT_V2) {
 202                         ceph_decode_32_safe(p, end, info->pool_ns_len, bad);
 203                         if (info->pool_ns_len > 0) {
 204                                 ceph_decode_need(p, end, info->pool_ns_len, bad);
 205                                 info->pool_ns_data = *p;
 206                                 *p += info->pool_ns_len;
 207                         }
 208                 }
 209
 210                 if (features & CEPH_FEATURE_FS_BTIME) {
 211                         ceph_decode_need(p, end, sizeof(info->btime), bad);
 212                         ceph_decode_copy(p, &info->btime, sizeof(info->btime));
 213                         ceph_decode_64_safe(p, end, info->change_attr, bad);
 214                 }
 215
 216                 info->dir_pin = -ENODATA;
 217                 /* info->snap_btime remains zero */
 218         }
 219         return 0;
 220 bad:
 221         err = -EIO;
 222 out_bad:
 223         return err;
 224 }
 225
 226 static int parse_reply_info_dir(void **p, void *end,
 227                                 struct ceph_mds_reply_dirfrag **dirfrag,
 228                                 u64 features)
 229 {
 230         if (features == (u64)-1) {
 231                 u8 struct_v, struct_compat;
 232                 u32 struct_len;
 233                 ceph_decode_8_safe(p, end, struct_v, bad);
 234                 ceph_decode_8_safe(p, end, struct_compat, bad);
 235                 /* struct_v is expected to be >= 1. we only understand
 236                  * encoding whose struct_compat == 1. */
 237                 if (!struct_v || struct_compat != 1)
 238                         goto bad;
 239                 ceph_decode_32_safe(p, end, struct_len, bad);
 240                 ceph_decode_need(p, end, struct_len, bad);
 241                 end = *p + struct_len;
 242         }
 243
 244         ceph_decode_need(p, end, sizeof(**dirfrag), bad);
 245         *dirfrag = *p;
 246         *p += sizeof(**dirfrag) + sizeof(u32) * le32_to_cpu((*dirfrag)->ndist);
 247         if (unlikely(*p > end))
 248                 goto bad;
 249         if (features == (u64)-1)
 250                 *p = end;
 251         return 0;
 252 bad:
 253         return -EIO;
 254 }
 255
 256 static int parse_reply_info_lease(void **p, void *end,
 257                                   struct ceph_mds_reply_lease **lease,
 258                                   u64 features)
 259 {
 260         if (features == (u64)-1) {
 261                 u8 struct_v, struct_compat;
 262                 u32 struct_len;
 263                 ceph_decode_8_safe(p, end, struct_v, bad);
 264                 ceph_decode_8_safe(p, end, struct_compat, bad);
 265                 /* struct_v is expected to be >= 1. we only understand
 266                  * encoding whose struct_compat == 1. */
 267                 if (!struct_v || struct_compat != 1)
 268                         goto bad;
 269                 ceph_decode_32_safe(p, end, struct_len, bad);
 270                 ceph_decode_need(p, end, struct_len, bad);
 271                 end = *p + struct_len;
 272         }
 273
 274         ceph_decode_need(p, end, sizeof(**lease), bad);
 275         *lease = *p;
 276         *p += sizeof(**lease);
 277         if (features == (u64)-1)
 278                 *p = end;
 279         return 0;
 280 bad:
 281         return -EIO;
 282 }
 283
 284 /*
 285  * parse a normal reply, which may contain a (dir+)dentry and/or a
 286  * target inode.
 287  */
 288 static int parse_reply_info_trace(void **p, void *end,
 289                                   struct ceph_mds_reply_info_parsed *info,
 290                                   u64 features)
 291 {
 292         int err;
 293
 294         if (info->head->is_dentry) {
 295                 err = parse_reply_info_in(p, end, &info->diri, features);
 296                 if (err < 0)
 297                         goto out_bad;
 298
 299                 err = parse_reply_info_dir(p, end, &info->dirfrag, features);
 300                 if (err < 0)
 301                         goto out_bad;
 302
 303                 ceph_decode_32_safe(p, end, info->dname_len, bad);
 304                 ceph_decode_need(p, end, info->dname_len, bad);
 305                 info->dname = *p;
 306                 *p += info->dname_len;
 307
 308                 err = parse_reply_info_lease(p, end, &info->dlease, features);
 309                 if (err < 0)
 310                         goto out_bad;
 311         }
 312
 313         if (info->head->is_target) {
 314                 err = parse_reply_info_in(p, end, &info->targeti, features);
 315                 if (err < 0)
 316                         goto out_bad;
 317         }
 318
 319         if (unlikely(*p != end))
 320                 goto bad;
 321         return 0;
 322
 323 bad:
 324         err = -EIO;
 325 out_bad:
 326         pr_err("problem parsing mds trace %d\n", err);
 327         return err;
 328 }
 329
 330 /*
 331  * parse readdir results
 332  */
 333 static int parse_reply_info_readdir(void **p, void *end,
 334                                 struct ceph_mds_reply_info_parsed *info,
 335                                 u64 features)
 336 {
 337         u32 num, i = 0;
 338         int err;
 339
 340         err = parse_reply_info_dir(p, end, &info->dir_dir, features);
 341         if (err < 0)
 342                 goto out_bad;
 343
 344         ceph_decode_need(p, end, sizeof(num) + 2, bad);
 345         num = ceph_decode_32(p);
 346         {
 347                 u16 flags = ceph_decode_16(p);
 348                 info->dir_end = !!(flags & CEPH_READDIR_FRAG_END);
 349                 info->dir_complete = !!(flags & CEPH_READDIR_FRAG_COMPLETE);
 350                 info->hash_order = !!(flags & CEPH_READDIR_HASH_ORDER);
 351                 info->offset_hash = !!(flags & CEPH_READDIR_OFFSET_HASH);
 352         }
 353         if (num == 0)
 354                 goto done;
 355
 356         BUG_ON(!info->dir_entries);
 357         if ((unsigned long)(info->dir_entries + num) >
 358             (unsigned long)info->dir_entries + info->dir_buf_size) {
 359                 pr_err("dir contents are larger than expected\n");
 360                 WARN_ON(1);
 361                 goto bad;
 362         }
 363
 364         info->dir_nr = num;
 365         while (num) {
 366                 struct ceph_mds_reply_dir_entry *rde = info->dir_entries + i;
 367                 /* dentry */
 368                 ceph_decode_32_safe(p, end, rde->name_len, bad);
 369                 ceph_decode_need(p, end, rde->name_len, bad);
 370                 rde->name = *p;
 371                 *p += rde->name_len;
 372                 dout("parsed dir dname '%.*s'\n", rde->name_len, rde->name);
 373
 374                 /* dentry lease */
 375                 err = parse_reply_info_lease(p, end, &rde->lease, features);
 376                 if (err)
 377                         goto out_bad;
 378                 /* inode */
 379                 err = parse_reply_info_in(p, end, &rde->inode, features);
 380                 if (err < 0)
 381                         goto out_bad;
 382                 /* ceph_readdir_prepopulate() will update it */
 383                 rde->offset = 0;
 384                 i++;
 385                 num--;
 386         }
 387
 388 done:
 389         /* Skip over any unrecognized fields */
 390         *p = end;
 391         return 0;
 392
 393 bad:
 394         err = -EIO;
 395 out_bad:
 396         pr_err("problem parsing dir contents %d\n", err);
 397         return err;
 398 }
 399
 400 /*
 401  * parse fcntl F_GETLK results
 402  */
 403 static int parse_reply_info_filelock(void **p, void *end,
 404                                      struct ceph_mds_reply_info_parsed *info,
 405                                      u64 features)
 406 {
 407         if (*p + sizeof(*info->filelock_reply) > end)
 408                 goto bad;
 409
 410         info->filelock_reply = *p;
 411
 412         /* Skip over any unrecognized fields */
 413         *p = end;
 414         return 0;
 415 bad:
 416         return -EIO;
 417 }
 418
 419
 420 #if BITS_PER_LONG == 64
 421
 422 #define DELEGATED_INO_AVAILABLE         xa_mk_value(1)
 423
 424 static int ceph_parse_deleg_inos(void **p, void *end,
 425                                  struct ceph_mds_session *s)
 426 {
 427         u32 sets;
 428
 429         ceph_decode_32_safe(p, end, sets, bad);
 430         dout("got %u sets of delegated inodes\n", sets);
 431         while (sets--) {
 432                 u64 start, len, ino;
 433
 434                 ceph_decode_64_safe(p, end, start, bad);
 435                 ceph_decode_64_safe(p, end, len, bad);
 436                 while (len--) {
 437                         int err = xa_insert(&s->s_delegated_inos, ino = start++,
 438                                             DELEGATED_INO_AVAILABLE,
 439                                             GFP_KERNEL);
 440                         if (!err) {
 441                                 dout("added delegated inode 0x%llx\n",
 442                                      start - 1);
 443                         } else if (err == -EBUSY) {
 444                                 pr_warn("ceph: MDS delegated inode 0x%llx more than once.\n",
 445                                         start - 1);
 446                         } else {
 447                                 return err;
 448                         }
 449                 }
 450         }
 451         return 0;
 452 bad:
 453         return -EIO;
 454 }
 455
 456 u64 ceph_get_deleg_ino(struct ceph_mds_session *s)
 457 {
 458         unsigned long ino;
 459         void *val;
 460
 461         xa_for_each(&s->s_delegated_inos, ino, val) {
 462                 val = xa_erase(&s->s_delegated_inos, ino);
 463                 if (val == DELEGATED_INO_AVAILABLE)
 464                         return ino;
 465         }
 466         return 0;
 467 }
 468
 469 int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino)
 470 {
 471         return xa_insert(&s->s_delegated_inos, ino, DELEGATED_INO_AVAILABLE,
 472                          GFP_KERNEL);
 473 }
 474 #else /* BITS_PER_LONG == 64 */
 475 /*
 476  * FIXME: xarrays can't handle 64-bit indexes on a 32-bit arch. For now, just
 477  * ignore delegated_inos on 32 bit arch. Maybe eventually add xarrays for top
 478  * and bottom words?
 479  */
 480 static int ceph_parse_deleg_inos(void **p, void *end,
 481                                  struct ceph_mds_session *s)
 482 {
 483         u32 sets;
 484
 485         ceph_decode_32_safe(p, end, sets, bad);
 486         if (sets)
 487                 ceph_decode_skip_n(p, end, sets * 2 * sizeof(__le64), bad);
 488         return 0;
 489 bad:
 490         return -EIO;
 491 }
 492
 493 u64 ceph_get_deleg_ino(struct ceph_mds_session *s)
 494 {
 495         return 0;
 496 }
 497
 498 int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino)
 499 {
 500         return 0;
 501 }
 502 #endif /* BITS_PER_LONG == 64 */
 503
 504 /*
 505  * parse create results
 506  */
 507 static int parse_reply_info_create(void **p, void *end,
 508                                   struct ceph_mds_reply_info_parsed *info,
 509                                   u64 features, struct ceph_mds_session *s)
 510 {
 511         int ret;
 512
 513         if (features == (u64)-1 ||
 514             (features & CEPH_FEATURE_REPLY_CREATE_INODE)) {
 515                 if (*p == end) {
 516                         /* Malformed reply? */
 517                         info->has_create_ino = false;
 518                 } else if (test_bit(CEPHFS_FEATURE_DELEG_INO, &s->s_features)) {
 519                         info->has_create_ino = true;
 520                         /* struct_v, struct_compat, and len */
 521                         ceph_decode_skip_n(p, end, 2 + sizeof(u32), bad);
 522                         ceph_decode_64_safe(p, end, info->ino, bad);
 523                         ret = ceph_parse_deleg_inos(p, end, s);
 524                         if (ret)
 525                                 return ret;
 526                 } else {
 527                         /* legacy */
 528                         ceph_decode_64_safe(p, end, info->ino, bad);
 529                         info->has_create_ino = true;
 530                 }
 531         } else {
 532                 if (*p != end)
 533                         goto bad;
 534         }
 535
 536         /* Skip over any unrecognized fields */
 537         *p = end;
 538         return 0;
 539 bad:
 540         return -EIO;
 541 }
 542
 543 /*
 544  * parse extra results
 545  */
 546 static int parse_reply_info_extra(void **p, void *end,
 547                                   struct ceph_mds_reply_info_parsed *info,
 548                                   u64 features, struct ceph_mds_session *s)
 549 {
 550         u32 op = le32_to_cpu(info->head->op);
 551
 552         if (op == CEPH_MDS_OP_GETFILELOCK)
 553                 return parse_reply_info_filelock(p, end, info, features);
 554         else if (op == CEPH_MDS_OP_READDIR || op == CEPH_MDS_OP_LSSNAP)
 555                 return parse_reply_info_readdir(p, end, info, features);
 556         else if (op == CEPH_MDS_OP_CREATE)
 557                 return parse_reply_info_create(p, end, info, features, s);
 558         else
 559                 return -EIO;
 560 }
 561
 562 /*
 563  * parse entire mds reply
 564  */
 565 static int parse_reply_info(struct ceph_mds_session *s, struct ceph_msg *msg,
 566                             struct ceph_mds_reply_info_parsed *info,
 567                             u64 features)
 568 {
 569         void *p, *end;
 570         u32 len;
 571         int err;
 572
 573         info->head = msg->front.iov_base;
 574         p = msg->front.iov_base + sizeof(struct ceph_mds_reply_head);
 575         end = p + msg->front.iov_len - sizeof(struct ceph_mds_reply_head);
 576
 577         /* trace */
 578         ceph_decode_32_safe(&p, end, len, bad);
 579         if (len > 0) {
 580                 ceph_decode_need(&p, end, len, bad);
 581                 err = parse_reply_info_trace(&p, p+len, info, features);
 582                 if (err < 0)
 583                         goto out_bad;
 584         }
 585
 586         /* extra */
 587         ceph_decode_32_safe(&p, end, len, bad);
 588         if (len > 0) {
 589                 ceph_decode_need(&p, end, len, bad);
 590                 err = parse_reply_info_extra(&p, p+len, info, features, s);
 591                 if (err < 0)
 592                         goto out_bad;
 593         }
 594
 595         /* snap blob */
 596         ceph_decode_32_safe(&p, end, len, bad);
 597         info->snapblob_len = len;
 598         info->snapblob = p;
 599         p += len;
 600
 601         if (p != end)
 602                 goto bad;
 603         return 0;
 604
 605 bad:
 606         err = -EIO;
 607 out_bad:
 608         pr_err("mds parse_reply err %d\n", err);
 609         return err;
 610 }
 611
 612 static void destroy_reply_info(struct ceph_mds_reply_info_parsed *info)
 613 {
 614         if (!info->dir_entries)
 615                 return;
 616         free_pages((unsigned long)info->dir_entries, get_order(info->dir_buf_size));
 617 }
 618
 619
 620 /*
 621  * sessions
 622  */
 623 const char *ceph_session_state_name(int s)
 624 {
 625         switch (s) {
 626         case CEPH_MDS_SESSION_NEW: return "new";
 627         case CEPH_MDS_SESSION_OPENING: return "opening";
 628         case CEPH_MDS_SESSION_OPEN: return "open";
 629         case CEPH_MDS_SESSION_HUNG: return "hung";
 630         case CEPH_MDS_SESSION_CLOSING: return "closing";
 631         case CEPH_MDS_SESSION_CLOSED: return "closed";
 632         case CEPH_MDS_SESSION_RESTARTING: return "restarting";
 633         case CEPH_MDS_SESSION_RECONNECTING: return "reconnecting";
 634         case CEPH_MDS_SESSION_REJECTED: return "rejected";
 635         default: return "???";
 636         }
 637 }
 638
 639 struct ceph_mds_session *ceph_get_mds_session(struct ceph_mds_session *s)
 640 {
 641         if (refcount_inc_not_zero(&s->s_ref)) {
 642                 dout("mdsc get_session %p %d -> %d\n", s,
 643                      refcount_read(&s->s_ref)-1, refcount_read(&s->s_ref));
 644                 return s;
 645         } else {
 646                 dout("mdsc get_session %p 0 -- FAIL\n", s);
 647                 return NULL;
 648         }
 649 }
 650
 651 void ceph_put_mds_session(struct ceph_mds_session *s)
 652 {
 653         dout("mdsc put_session %p %d -> %d\n", s,
 654              refcount_read(&s->s_ref), refcount_read(&s->s_ref)-1);
 655         if (refcount_dec_and_test(&s->s_ref)) {
 656                 if (s->s_auth.authorizer)
 657                         ceph_auth_destroy_authorizer(s->s_auth.authorizer);
 658                 WARN_ON(mutex_is_locked(&s->s_mutex));
 659                 xa_destroy(&s->s_delegated_inos);
 660                 kfree(s);
 661         }
 662 }
 663
 664 /*
 665  * called under mdsc->mutex
 666  */
 667 struct ceph_mds_session *__ceph_lookup_mds_session(struct ceph_mds_client *mdsc,
 668                                                    int mds)
 669 {
 670         if (mds >= mdsc->max_sessions || !mdsc->sessions[mds])
 671                 return NULL;
 672         return ceph_get_mds_session(mdsc->sessions[mds]);
 673 }
 674
 675 static bool __have_session(struct ceph_mds_client *mdsc, int mds)
 676 {
 677         if (mds >= mdsc->max_sessions || !mdsc->sessions[mds])
 678                 return false;
 679         else
 680                 return true;
 681 }
 682
 683 static int __verify_registered_session(struct ceph_mds_client *mdsc,
 684                                        struct ceph_mds_session *s)
 685 {
 686         if (s->s_mds >= mdsc->max_sessions ||
 687             mdsc->sessions[s->s_mds] != s)
 688                 return -ENOENT;
 689         return 0;
 690 }
 691
 692 /*
 693  * create+register a new session for given mds.
 694  * called under mdsc->mutex.
 695  */
 696 static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
 697                                                  int mds)
 698 {
 699         struct ceph_mds_session *s;
 700
 701         if (mds >= mdsc->mdsmap->possible_max_rank)
 702                 return ERR_PTR(-EINVAL);
 703
 704         s = kzalloc(sizeof(*s), GFP_NOFS);
 705         if (!s)
 706                 return ERR_PTR(-ENOMEM);
 707
 708         if (mds >= mdsc->max_sessions) {
 709                 int newmax = 1 << get_count_order(mds + 1);
 710                 struct ceph_mds_session **sa;
 711
 712                 dout("%s: realloc to %d\n", __func__, newmax);
 713                 sa = kcalloc(newmax, sizeof(void *), GFP_NOFS);
 714                 if (!sa)
 715                         goto fail_realloc;
 716                 if (mdsc->sessions) {
 717                         memcpy(sa, mdsc->sessions,
 718                                mdsc->max_sessions * sizeof(void *));
 719                         kfree(mdsc->sessions);
 720                 }
 721                 mdsc->sessions = sa;
 722                 mdsc->max_sessions = newmax;
 723         }
 724
 725         dout("%s: mds%d\n", __func__, mds);
 726         s->s_mdsc = mdsc;
 727         s->s_mds = mds;
 728         s->s_state = CEPH_MDS_SESSION_NEW;
 729         s->s_ttl = 0;
 730         s->s_seq = 0;
 731         mutex_init(&s->s_mutex);
 732
 733         ceph_con_init(&s->s_con, s, &mds_con_ops, &mdsc->fsc->client->msgr);
 734
 735         spin_lock_init(&s->s_gen_ttl_lock);
 736         s->s_cap_gen = 1;
 737         s->s_cap_ttl = jiffies - 1;
 738
 739         spin_lock_init(&s->s_cap_lock);
 740         s->s_renew_requested = 0;
 741         s->s_renew_seq = 0;
 742         INIT_LIST_HEAD(&s->s_caps);
 743         s->s_nr_caps = 0;
 744         refcount_set(&s->s_ref, 1);
 745         INIT_LIST_HEAD(&s->s_waiting);
 746         INIT_LIST_HEAD(&s->s_unsafe);
 747         xa_init(&s->s_delegated_inos);
 748         s->s_num_cap_releases = 0;
 749         s->s_cap_reconnect = 0;
 750         s->s_cap_iterator = NULL;
 751         INIT_LIST_HEAD(&s->s_cap_releases);
 752         INIT_WORK(&s->s_cap_release_work, ceph_cap_release_work);
 753
 754         INIT_LIST_HEAD(&s->s_cap_dirty);
 755         INIT_LIST_HEAD(&s->s_cap_flushing);
 756
 757         mdsc->sessions[mds] = s;
 758         atomic_inc(&mdsc->num_sessions);
 759         refcount_inc(&s->s_ref);  /* one ref to sessions[], one to caller */
 760
 761         ceph_con_open(&s->s_con, CEPH_ENTITY_TYPE_MDS, mds,
 762                       ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
 763
 764         return s;
 765
 766 fail_realloc:
 767         kfree(s);
 768         return ERR_PTR(-ENOMEM);
 769 }
 770
 771 /*
 772  * called under mdsc->mutex
 773  */
 774 static void __unregister_session(struct ceph_mds_client *mdsc,
 775                                struct ceph_mds_session *s)
 776 {
 777         dout("__unregister_session mds%d %p\n", s->s_mds, s);
 778         BUG_ON(mdsc->sessions[s->s_mds] != s);
 779         mdsc->sessions[s->s_mds] = NULL;
 780         ceph_con_close(&s->s_con);
 781         ceph_put_mds_session(s);
 782         atomic_dec(&mdsc->num_sessions);
 783 }
 784
 785 /*
 786  * drop session refs in request.
 787  *
 788  * should be last request ref, or hold mdsc->mutex
 789  */
 790 static void put_request_session(struct ceph_mds_request *req)
 791 {
 792         if (req->r_session) {
 793                 ceph_put_mds_session(req->r_session);
 794                 req->r_session = NULL;
 795         }
 796 }
 797
 798 void ceph_mdsc_release_request(struct kref *kref)
 799 {
 800         struct ceph_mds_request *req = container_of(kref,
 801                                                     struct ceph_mds_request,
 802                                                     r_kref);
 803         ceph_mdsc_release_dir_caps_no_check(req);
 804         destroy_reply_info(&req->r_reply_info);
 805         if (req->r_request)
 806                 ceph_msg_put(req->r_request);
 807         if (req->r_reply)
 808                 ceph_msg_put(req->r_reply);
 809         if (req->r_inode) {
 810                 ceph_put_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
 811                 /* avoid calling iput_final() in mds dispatch threads */
 812                 ceph_async_iput(req->r_inode);
 813         }
 814         if (req->r_parent) {
 815                 ceph_put_cap_refs(ceph_inode(req->r_parent), CEPH_CAP_PIN);
 816                 ceph_async_iput(req->r_parent);
 817         }
 818         ceph_async_iput(req->r_target_inode);
 819         if (req->r_dentry)
 820                 dput(req->r_dentry);
 821         if (req->r_old_dentry)
 822                 dput(req->r_old_dentry);
 823         if (req->r_old_dentry_dir) {
 824                 /*
 825                  * track (and drop pins for) r_old_dentry_dir
 826                  * separately, since r_old_dentry's d_parent may have
 827                  * changed between the dir mutex being dropped and
 828                  * this request being freed.
 829                  */
 830                 ceph_put_cap_refs(ceph_inode(req->r_old_dentry_dir),
 831                                   CEPH_CAP_PIN);
 832                 ceph_async_iput(req->r_old_dentry_dir);
 833         }
 834         kfree(req->r_path1);
 835         kfree(req->r_path2);
 836         put_cred(req->r_cred);
 837         if (req->r_pagelist)
 838                 ceph_pagelist_release(req->r_pagelist);
 839         put_request_session(req);
 840         ceph_unreserve_caps(req->r_mdsc, &req->r_caps_reservation);
 841         WARN_ON_ONCE(!list_empty(&req->r_wait));
 842         kmem_cache_free(ceph_mds_request_cachep, req);
 843 }
 844
 845 DEFINE_RB_FUNCS(request, struct ceph_mds_request, r_tid, r_node)
 846
 847 /*
 848  * lookup session, bump ref if found.
 849  *
 850  * called under mdsc->mutex.
 851  */
 852 static struct ceph_mds_request *
 853 lookup_get_request(struct ceph_mds_client *mdsc, u64 tid)
 854 {
 855         struct ceph_mds_request *req;
 856
 857         req = lookup_request(&mdsc->request_tree, tid);
 858         if (req)
 859                 ceph_mdsc_get_request(req);
 860
 861         return req;
 862 }
 863
 864 /*
 865  * Register an in-flight request, and assign a tid.  Link to directory
 866  * are modifying (if any).
 867  *
 868  * Called under mdsc->mutex.
 869  */
 870 static void __register_request(struct ceph_mds_client *mdsc,
 871                                struct ceph_mds_request *req,
 872                                struct inode *dir)
 873 {
 874         int ret = 0;
 875
 876         req->r_tid = ++mdsc->last_tid;
 877         if (req->r_num_caps) {
 878                 ret = ceph_reserve_caps(mdsc, &req->r_caps_reservation,
 879                                         req->r_num_caps);
 880                 if (ret < 0) {
 881                         pr_err("__register_request %p "
 882                                "failed to reserve caps: %d\n", req, ret);
 883                         /* set req->r_err to fail early from __do_request */
 884                         req->r_err = ret;
 885                         return;
 886                 }
 887         }
 888         dout("__register_request %p tid %lld\n", req, req->r_tid);
 889         ceph_mdsc_get_request(req);
 890         insert_request(&mdsc->request_tree, req);
 891
 892         req->r_cred = get_current_cred();
 893
 894         if (mdsc->oldest_tid == 0 && req->r_op != CEPH_MDS_OP_SETFILELOCK)
 895                 mdsc->oldest_tid = req->r_tid;
 896
 897         if (dir) {
 898                 struct ceph_inode_info *ci = ceph_inode(dir);
 899
 900                 ihold(dir);
 901                 req->r_unsafe_dir = dir;
 902                 spin_lock(&ci->i_unsafe_lock);
 903                 list_add_tail(&req->r_unsafe_dir_item, &ci->i_unsafe_dirops);
 904                 spin_unlock(&ci->i_unsafe_lock);
 905         }
 906 }
 907
 908 static void __unregister_request(struct ceph_mds_client *mdsc,
 909                                  struct ceph_mds_request *req)
 910 {
 911         dout("__unregister_request %p tid %lld\n", req, req->r_tid);
 912
 913         /* Never leave an unregistered request on an unsafe list! */
 914         list_del_init(&req->r_unsafe_item);
 915
 916         if (req->r_tid == mdsc->oldest_tid) {
 917                 struct rb_node *p = rb_next(&req->r_node);
 918                 mdsc->oldest_tid = 0;
 919                 while (p) {
 920                         struct ceph_mds_request *next_req =
 921                                 rb_entry(p, struct ceph_mds_request, r_node);
 922                         if (next_req->r_op != CEPH_MDS_OP_SETFILELOCK) {
 923                                 mdsc->oldest_tid = next_req->r_tid;
 924                                 break;
 925                         }
 926                         p = rb_next(p);
 927                 }
 928         }
 929
 930         erase_request(&mdsc->request_tree, req);
 931
 932         if (req->r_unsafe_dir) {
 933                 struct ceph_inode_info *ci = ceph_inode(req->r_unsafe_dir);
 934                 spin_lock(&ci->i_unsafe_lock);
 935                 list_del_init(&req->r_unsafe_dir_item);
 936                 spin_unlock(&ci->i_unsafe_lock);
 937         }
 938         if (req->r_target_inode &&
 939             test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
 940                 struct ceph_inode_info *ci = ceph_inode(req->r_target_inode);
 941                 spin_lock(&ci->i_unsafe_lock);
 942                 list_del_init(&req->r_unsafe_target_item);
 943                 spin_unlock(&ci->i_unsafe_lock);
 944         }
 945
 946         if (req->r_unsafe_dir) {
 947                 /* avoid calling iput_final() in mds dispatch threads */
 948                 ceph_async_iput(req->r_unsafe_dir);
 949                 req->r_unsafe_dir = NULL;
 950         }
 951
 952         complete_all(&req->r_safe_completion);
 953
 954         ceph_mdsc_put_request(req);
 955 }
 956
 957 /*
 958  * Walk back up the dentry tree until we hit a dentry representing a
 959  * non-snapshot inode. We do this using the rcu_read_lock (which must be held
 960  * when calling this) to ensure that the objects won't disappear while we're
 961  * working with them. Once we hit a candidate dentry, we attempt to take a
 962  * reference to it, and return that as the result.
 963  */
 964 static struct inode *get_nonsnap_parent(struct dentry *dentry)
 965 {
 966         struct inode *inode = NULL;
 967
 968         while (dentry && !IS_ROOT(dentry)) {
 969                 inode = d_inode_rcu(dentry);
 970                 if (!inode || ceph_snap(inode) == CEPH_NOSNAP)
 971                         break;
 972                 dentry = dentry->d_parent;
 973         }
 974         if (inode)
 975                 inode = igrab(inode);
 976         return inode;
 977 }
 978
 979 /*
 980  * Choose mds to send request to next.  If there is a hint set in the
 981  * request (e.g., due to a prior forward hint from the mds), use that.
 982  * Otherwise, consult frag tree and/or caps to identify the
 983  * appropriate mds.  If all else fails, choose randomly.
 984  *
 985  * Called under mdsc->mutex.
 986  */
 987 static int __choose_mds(struct ceph_mds_client *mdsc,
 988                         struct ceph_mds_request *req,
 989                         bool *random)
 990 {
 991         struct inode *inode;
 992         struct ceph_inode_info *ci;
 993         struct ceph_cap *cap;
 994         int mode = req->r_direct_mode;
 995         int mds = -1;
 996         u32 hash = req->r_direct_hash;
 997         bool is_hash = test_bit(CEPH_MDS_R_DIRECT_IS_HASH, &req->r_req_flags);
 998
 999         if (random)
1000                 *random = false;
1001
1002         /*
1003          * is there a specific mds we should try?  ignore hint if we have
1004          * no session and the mds is not up (active or recovering).
1005          */
1006         if (req->r_resend_mds >= 0 &&
1007             (__have_session(mdsc, req->r_resend_mds) ||
1008              ceph_mdsmap_get_state(mdsc->mdsmap, req->r_resend_mds) > 0)) {
1009                 dout("%s using resend_mds mds%d\n", __func__,
1010                      req->r_resend_mds);
1011                 return req->r_resend_mds;
1012         }
1013
1014         if (mode == USE_RANDOM_MDS)
1015                 goto random;
1016
1017         inode = NULL;
1018         if (req->r_inode) {
1019                 if (ceph_snap(req->r_inode) != CEPH_SNAPDIR) {
1020                         inode = req->r_inode;
1021                         ihold(inode);
1022                 } else {
1023                         /* req->r_dentry is non-null for LSSNAP request */
1024                         rcu_read_lock();
1025                         inode = get_nonsnap_parent(req->r_dentry);
1026                         rcu_read_unlock();
1027                         dout("%s using snapdir's parent %p\n", __func__, inode);
1028                 }
1029         } else if (req->r_dentry) {
1030                 /* ignore race with rename; old or new d_parent is okay */
1031                 struct dentry *parent;
1032                 struct inode *dir;
1033
1034                 rcu_read_lock();
1035                 parent = READ_ONCE(req->r_dentry->d_parent);
1036                 dir = req->r_parent ? : d_inode_rcu(parent);
1037
1038                 if (!dir || dir->i_sb != mdsc->fsc->sb) {
1039                         /*  not this fs or parent went negative */
1040                         inode = d_inode(req->r_dentry);
1041                         if (inode)
1042                                 ihold(inode);
1043                 } else if (ceph_snap(dir) != CEPH_NOSNAP) {
1044                         /* direct snapped/virtual snapdir requests
1045                          * based on parent dir inode */
1046                         inode = get_nonsnap_parent(parent);
1047                         dout("%s using nonsnap parent %p\n", __func__, inode);
1048                 } else {
1049                         /* dentry target */
1050                         inode = d_inode(req->r_dentry);
1051                         if (!inode || mode == USE_AUTH_MDS) {
1052                                 /* dir + name */
1053                                 inode = igrab(dir);
1054                                 hash = ceph_dentry_hash(dir, req->r_dentry);
1055                                 is_hash = true;
1056                         } else {
1057                                 ihold(inode);
1058                         }
1059                 }
1060                 rcu_read_unlock();
1061         }
1062
1063         dout("%s %p is_hash=%d (0x%x) mode %d\n", __func__, inode, (int)is_hash,
1064              hash, mode);
1065         if (!inode)
1066                 goto random;
1067         ci = ceph_inode(inode);
1068
1069         if (is_hash && S_ISDIR(inode->i_mode)) {
1070                 struct ceph_inode_frag frag;
1071                 int found;
1072
1073                 ceph_choose_frag(ci, hash, &frag, &found);
1074                 if (found) {
1075                         if (mode == USE_ANY_MDS && frag.ndist > 0) {
1076                                 u8 r;
1077
1078                                 /* choose a random replica */
1079                                 get_random_bytes(&r, 1);
1080                                 r %= frag.ndist;
1081                                 mds = frag.dist[r];
1082                                 dout("%s %p %llx.%llx frag %u mds%d (%d/%d)\n",
1083                                      __func__, inode, ceph_vinop(inode),
1084                                      frag.frag, mds, (int)r, frag.ndist);
1085                                 if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
1086                                     CEPH_MDS_STATE_ACTIVE &&
1087                                     !ceph_mdsmap_is_laggy(mdsc->mdsmap, mds))
1088                                         goto out;
1089                         }
1090
1091                         /* since this file/dir wasn't known to be
1092                          * replicated, then we want to look for the
1093                          * authoritative mds. */
1094                         if (frag.mds >= 0) {
1095                                 /* choose auth mds */
1096                                 mds = frag.mds;
1097                                 dout("%s %p %llx.%llx frag %u mds%d (auth)\n",
1098                                      __func__, inode, ceph_vinop(inode),
1099                                      frag.frag, mds);
1100                                 if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
1101                                     CEPH_MDS_STATE_ACTIVE) {
1102                                         if (!ceph_mdsmap_is_laggy(mdsc->mdsmap,
1103                                                                   mds))
1104                                                 goto out;
1105                                 }
1106                         }
1107                         mode = USE_AUTH_MDS;
1108                 }
1109         }
1110
1111         spin_lock(&ci->i_ceph_lock);
1112         cap = NULL;
1113         if (mode == USE_AUTH_MDS)
1114                 cap = ci->i_auth_cap;
1115         if (!cap && !RB_EMPTY_ROOT(&ci->i_caps))
1116                 cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node);
1117         if (!cap) {
1118                 spin_unlock(&ci->i_ceph_lock);
1119                 ceph_async_iput(inode);
1120                 goto random;
1121         }
1122         mds = cap->session->s_mds;
1123         dout("%s %p %llx.%llx mds%d (%scap %p)\n", __func__,
1124              inode, ceph_vinop(inode), mds,
1125              cap == ci->i_auth_cap ? "auth " : "", cap);
1126         spin_unlock(&ci->i_ceph_lock);
1127 out:
1128         /* avoid calling iput_final() while holding mdsc->mutex or
1129          * in mds dispatch threads */
1130         ceph_async_iput(inode);
1131         return mds;
1132
1133 random:
1134         if (random)
1135                 *random = true;
1136
1137         mds = ceph_mdsmap_get_random_mds(mdsc->mdsmap);
1138         dout("%s chose random mds%d\n", __func__, mds);
1139         return mds;
1140 }
1141
1142
1143 /*
1144  * session messages
1145  */
1146 static struct ceph_msg *create_session_msg(u32 op, u64 seq)
1147 {
1148         struct ceph_msg *msg;
1149         struct ceph_mds_session_head *h;
1150
1151         msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h), GFP_NOFS,
1152                            false);
1153         if (!msg) {
1154                 pr_err("create_session_msg ENOMEM creating msg\n");
1155                 return NULL;
1156         }
1157         h = msg->front.iov_base;
1158         h->op = cpu_to_le32(op);
1159         h->seq = cpu_to_le64(seq);
1160
1161         return msg;
1162 }
1163
1164 static const unsigned char feature_bits[] = CEPHFS_FEATURES_CLIENT_SUPPORTED;
1165 #define FEATURE_BYTES(c) (DIV_ROUND_UP((size_t)feature_bits[c - 1] + 1, 64) * 8)
1166 static int encode_supported_features(void **p, void *end)
1167 {
1168         static const size_t count = ARRAY_SIZE(feature_bits);
1169
1170         if (count > 0) {
1171                 size_t i;
1172                 size_t size = FEATURE_BYTES(count);
1173
1174                 if (WARN_ON_ONCE(*p + 4 + size > end))
1175                         return -ERANGE;
1176
1177                 ceph_encode_32(p, size);
1178                 memset(*p, 0, size);
1179                 for (i = 0; i < count; i++)
1180                         ((unsigned char*)(*p))[i / 8] |= BIT(feature_bits[i] % 8);
1181                 *p += size;
1182         } else {
1183                 if (WARN_ON_ONCE(*p + 4 > end))
1184                         return -ERANGE;
1185
1186                 ceph_encode_32(p, 0);
1187         }
1188
1189         return 0;
1190 }
1191
1192 static const unsigned char metric_bits[] = CEPHFS_METRIC_SPEC_CLIENT_SUPPORTED;
1193 #define METRIC_BYTES(cnt) (DIV_ROUND_UP((size_t)metric_bits[cnt - 1] + 1, 64) * 8)
1194 static int encode_metric_spec(void **p, void *end)
1195 {
1196         static const size_t count = ARRAY_SIZE(metric_bits);
1197
1198         /* header */
1199         if (WARN_ON_ONCE(*p + 2 > end))
1200                 return -ERANGE;
1201
1202         ceph_encode_8(p, 1); /* version */
1203         ceph_encode_8(p, 1); /* compat */
1204
1205         if (count > 0) {
1206                 size_t i;
1207                 size_t size = METRIC_BYTES(count);
1208
1209                 if (WARN_ON_ONCE(*p + 4 + 4 + size > end))
1210                         return -ERANGE;
1211
1212                 /* metric spec info length */
1213                 ceph_encode_32(p, 4 + size);
1214
1215                 /* metric spec */
1216                 ceph_encode_32(p, size);
1217                 memset(*p, 0, size);
1218                 for (i = 0; i < count; i++)
1219                         ((unsigned char *)(*p))[i / 8] |= BIT(metric_bits[i] % 8);
1220                 *p += size;
1221         } else {
1222                 if (WARN_ON_ONCE(*p + 4 + 4 > end))
1223                         return -ERANGE;
1224
1225                 /* metric spec info length */
1226                 ceph_encode_32(p, 4);
1227                 /* metric spec */
1228                 ceph_encode_32(p, 0);
1229         }
1230
1231         return 0;
1232 }
1233
1234 /*
1235  * session message, specialization for CEPH_SESSION_REQUEST_OPEN
1236  * to include additional client metadata fields.
1237  */
1238 static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u64 seq)
1239 {
1240         struct ceph_msg *msg;
1241         struct ceph_mds_session_head *h;
1242         int i;
1243         int extra_bytes = 0;
1244         int metadata_key_count = 0;
1245         struct ceph_options *opt = mdsc->fsc->client->options;
1246         struct ceph_mount_options *fsopt = mdsc->fsc->mount_options;
1247         size_t size, count;
1248         void *p, *end;
1249         int ret;
1250
1251         const char* metadata[][2] = {
1252                 {"hostname", mdsc->nodename},
1253                 {"kernel_version", init_utsname()->release},
1254                 {"entity_id", opt->name ? : ""},
1255                 {"root", fsopt->server_path ? : "/"},
1256                 {NULL, NULL}
1257         };
1258
1259         /* Calculate serialized length of metadata */
1260         extra_bytes = 4;  /* map length */
1261         for (i = 0; metadata[i][0]; ++i) {
1262                 extra_bytes += 8 + strlen(metadata[i][0]) +
1263                         strlen(metadata[i][1]);
1264                 metadata_key_count++;
1265         }
1266
1267         /* supported feature */
1268         size = 0;
1269         count = ARRAY_SIZE(feature_bits);
1270         if (count > 0)
1271                 size = FEATURE_BYTES(count);
1272         extra_bytes += 4 + size;
1273
1274         /* metric spec */
1275         size = 0;
1276         count = ARRAY_SIZE(metric_bits);
1277         if (count > 0)
1278                 size = METRIC_BYTES(count);
1279         extra_bytes += 2 + 4 + 4 + size;
1280
1281         /* Allocate the message */
1282         msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h) + extra_bytes,
1283                            GFP_NOFS, false);
1284         if (!msg) {
1285                 pr_err("create_session_msg ENOMEM creating msg\n");
1286                 return ERR_PTR(-ENOMEM);
1287         }
1288         p = msg->front.iov_base;
1289         end = p + msg->front.iov_len;
1290
1291         h = p;
1292         h->op = cpu_to_le32(CEPH_SESSION_REQUEST_OPEN);
1293         h->seq = cpu_to_le64(seq);
1294
1295         /*
1296          * Serialize client metadata into waiting buffer space, using
1297          * the format that userspace expects for map<string, string>
1298          *
1299          * ClientSession messages with metadata are v4
1300          */
1301         msg->hdr.version = cpu_to_le16(4);
1302         msg->hdr.compat_version = cpu_to_le16(1);
1303
1304         /* The write pointer, following the session_head structure */
1305         p += sizeof(*h);
1306
1307         /* Number of entries in the map */
1308         ceph_encode_32(&p, metadata_key_count);
1309
1310         /* Two length-prefixed strings for each entry in the map */
1311         for (i = 0; metadata[i][0]; ++i) {
1312                 size_t const key_len = strlen(metadata[i][0]);
1313                 size_t const val_len = strlen(metadata[i][1]);
1314
1315                 ceph_encode_32(&p, key_len);
1316                 memcpy(p, metadata[i][0], key_len);
1317                 p += key_len;
1318                 ceph_encode_32(&p, val_len);
1319                 memcpy(p, metadata[i][1], val_len);
1320                 p += val_len;
1321         }
1322
1323         ret = encode_supported_features(&p, end);
1324         if (ret) {
1325                 pr_err("encode_supported_features failed!\n");
1326                 ceph_msg_put(msg);
1327                 return ERR_PTR(ret);
1328         }
1329
1330         ret = encode_metric_spec(&p, end);
1331         if (ret) {
1332                 pr_err("encode_metric_spec failed!\n");
1333                 ceph_msg_put(msg);
1334                 return ERR_PTR(ret);
1335         }
1336
1337         msg->front.iov_len = p - msg->front.iov_base;
1338         msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
1339
1340         return msg;
1341 }
1342
1343 /*
1344  * send session open request.
1345  *
1346  * called under mdsc->mutex
1347  */
1348 static int __open_session(struct ceph_mds_client *mdsc,
1349                           struct ceph_mds_session *session)
1350 {
1351         struct ceph_msg *msg;
1352         int mstate;
1353         int mds = session->s_mds;
1354
1355         /* wait for mds to go active? */
1356         mstate = ceph_mdsmap_get_state(mdsc->mdsmap, mds);
1357         dout("open_session to mds%d (%s)\n", mds,
1358              ceph_mds_state_name(mstate));
1359         session->s_state = CEPH_MDS_SESSION_OPENING;
1360         session->s_renew_requested = jiffies;
1361
1362         /* send connect message */
1363         msg = create_session_open_msg(mdsc, session->s_seq);
1364         if (IS_ERR(msg))
1365                 return PTR_ERR(msg);
1366         ceph_con_send(&session->s_con, msg);
1367         return 0;
1368 }
1369
1370 /*
1371  * open sessions for any export targets for the given mds
1372  *
1373  * called under mdsc->mutex
1374  */
1375 static struct ceph_mds_session *
1376 __open_export_target_session(struct ceph_mds_client *mdsc, int target)
1377 {
1378         struct ceph_mds_session *session;
1379         int ret;
1380
1381         session = __ceph_lookup_mds_session(mdsc, target);
1382         if (!session) {
1383                 session = register_session(mdsc, target);
1384                 if (IS_ERR(session))
1385                         return session;
1386         }
1387         if (session->s_state == CEPH_MDS_SESSION_NEW ||
1388             session->s_state == CEPH_MDS_SESSION_CLOSING) {
1389                 ret = __open_session(mdsc, session);
1390                 if (ret)
1391                         return ERR_PTR(ret);
1392         }
1393
1394         return session;
1395 }
1396
1397 struct ceph_mds_session *
1398 ceph_mdsc_open_export_target_session(struct ceph_mds_client *mdsc, int target)
1399 {
1400         struct ceph_mds_session *session;
1401
1402         dout("open_export_target_session to mds%d\n", target);
1403
1404         mutex_lock(&mdsc->mutex);
1405         session = __open_export_target_session(mdsc, target);
1406         mutex_unlock(&mdsc->mutex);
1407
1408         return session;
1409 }
1410
1411 static void __open_export_target_sessions(struct ceph_mds_client *mdsc,
1412                                           struct ceph_mds_session *session)
1413 {
1414         struct ceph_mds_info *mi;
1415         struct ceph_mds_session *ts;
1416         int i, mds = session->s_mds;
1417
1418         if (mds >= mdsc->mdsmap->possible_max_rank)
1419                 return;
1420
1421         mi = &mdsc->mdsmap->m_info[mds];
1422         dout("open_export_target_sessions for mds%d (%d targets)\n",
1423              session->s_mds, mi->num_export_targets);
1424
1425         for (i = 0; i < mi->num_export_targets; i++) {
1426                 ts = __open_export_target_session(mdsc, mi->export_targets[i]);
1427                 if (!IS_ERR(ts))
1428                         ceph_put_mds_session(ts);
1429         }
1430 }
1431
1432 void ceph_mdsc_open_export_target_sessions(struct ceph_mds_client *mdsc,
1433                                            struct ceph_mds_session *session)
1434 {
1435         mutex_lock(&mdsc->mutex);
1436         __open_export_target_sessions(mdsc, session);
1437         mutex_unlock(&mdsc->mutex);
1438 }
1439
1440 /*
1441  * session caps
1442  */
1443
1444 static void detach_cap_releases(struct ceph_mds_session *session,
1445                                 struct list_head *target)
1446 {
1447         lockdep_assert_held(&session->s_cap_lock);
1448
1449         list_splice_init(&session->s_cap_releases, target);
1450         session->s_num_cap_releases = 0;
1451         dout("dispose_cap_releases mds%d\n", session->s_mds);
1452 }
1453
1454 static void dispose_cap_releases(struct ceph_mds_client *mdsc,
1455                                  struct list_head *dispose)
1456 {
1457         while (!list_empty(dispose)) {
1458                 struct ceph_cap *cap;
1459                 /* zero out the in-progress message */
1460                 cap = list_first_entry(dispose, struct ceph_cap, session_caps);
1461                 list_del(&cap->session_caps);
1462                 ceph_put_cap(mdsc, cap);
1463         }
1464 }
1465
1466 static void cleanup_session_requests(struct ceph_mds_client *mdsc,
1467                                      struct ceph_mds_session *session)
1468 {
1469         struct ceph_mds_request *req;
1470         struct rb_node *p;
1471         struct ceph_inode_info *ci;
1472
1473         dout("cleanup_session_requests mds%d\n", session->s_mds);
1474         mutex_lock(&mdsc->mutex);
1475         while (!list_empty(&session->s_unsafe)) {
1476                 req = list_first_entry(&session->s_unsafe,
1477                                        struct ceph_mds_request, r_unsafe_item);
1478                 pr_warn_ratelimited(" dropping unsafe request %llu\n",
1479                                     req->r_tid);
1480                 if (req->r_target_inode) {
1481                         /* dropping unsafe change of inode's attributes */
1482                         ci = ceph_inode(req->r_target_inode);
1483                         errseq_set(&ci->i_meta_err, -EIO);
1484                 }
1485                 if (req->r_unsafe_dir) {
1486                         /* dropping unsafe directory operation */
1487                         ci = ceph_inode(req->r_unsafe_dir);
1488                         errseq_set(&ci->i_meta_err, -EIO);
1489                 }
1490                 __unregister_request(mdsc, req);
1491         }
1492         /* zero r_attempts, so kick_requests() will re-send requests */
1493         p = rb_first(&mdsc->request_tree);
1494         while (p) {
1495                 req = rb_entry(p, struct ceph_mds_request, r_node);
1496                 p = rb_next(p);
1497                 if (req->r_session &&
1498                     req->r_session->s_mds == session->s_mds)
1499                         req->r_attempts = 0;
1500         }
1501         mutex_unlock(&mdsc->mutex);
1502 }
1503
1504 /*
1505  * Helper to safely iterate over all caps associated with a session, with
1506  * special care taken to handle a racing __ceph_remove_cap().
1507  *
1508  * Caller must hold session s_mutex.
1509  */
1510 int ceph_iterate_session_caps(struct ceph_mds_session *session,
1511                               int (*cb)(struct inode *, struct ceph_cap *,
1512                                         void *), void *arg)
1513 {
1514         struct list_head *p;
1515         struct ceph_cap *cap;
1516         struct inode *inode, *last_inode = NULL;
1517         struct ceph_cap *old_cap = NULL;
1518         int ret;
1519
1520         dout("iterate_session_caps %p mds%d\n", session, session->s_mds);
1521         spin_lock(&session->s_cap_lock);
1522         p = session->s_caps.next;
1523         while (p != &session->s_caps) {
1524                 cap = list_entry(p, struct ceph_cap, session_caps);
1525                 inode = igrab(&cap->ci->vfs_inode);
1526                 if (!inode) {
1527                         p = p->next;
1528                         continue;
1529                 }
1530                 session->s_cap_iterator = cap;
1531                 spin_unlock(&session->s_cap_lock);
1532
1533                 if (last_inode) {
1534                         /* avoid calling iput_final() while holding
1535                          * s_mutex or in mds dispatch threads */
1536                         ceph_async_iput(last_inode);
1537                         last_inode = NULL;
1538                 }
1539                 if (old_cap) {
1540                         ceph_put_cap(session->s_mdsc, old_cap);
1541                         old_cap = NULL;
1542                 }
1543
1544                 ret = cb(inode, cap, arg);
1545                 last_inode = inode;
1546
1547                 spin_lock(&session->s_cap_lock);
1548                 p = p->next;
1549                 if (!cap->ci) {
1550                         dout("iterate_session_caps  finishing cap %p removal\n",
1551                              cap);
1552                         BUG_ON(cap->session != session);
1553                         cap->session = NULL;
1554                         list_del_init(&cap->session_caps);
1555                         session->s_nr_caps--;
1556                         atomic64_dec(&session->s_mdsc->metric.total_caps);
1557                         if (cap->queue_release)
1558                                 __ceph_queue_cap_release(session, cap);
1559                         else
1560                                 old_cap = cap;  /* put_cap it w/o locks held */
1561                 }
1562                 if (ret < 0)
1563                         goto out;
1564         }
1565         ret = 0;
1566 out:
1567         session->s_cap_iterator = NULL;
1568         spin_unlock(&session->s_cap_lock);
1569
1570         ceph_async_iput(last_inode);
1571         if (old_cap)
1572                 ceph_put_cap(session->s_mdsc, old_cap);
1573
1574         return ret;
1575 }
1576
1577 static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
1578                                   void *arg)
1579 {
1580         struct ceph_fs_client *fsc = (struct ceph_fs_client *)arg;
1581         struct ceph_inode_info *ci = ceph_inode(inode);
1582         LIST_HEAD(to_remove);
1583         bool dirty_dropped = false;
1584         bool invalidate = false;
1585
1586         dout("removing cap %p, ci is %p, inode is %p\n",
1587              cap, ci, &ci->vfs_inode);
1588         spin_lock(&ci->i_ceph_lock);
1589         __ceph_remove_cap(cap, false);
1590         if (!ci->i_auth_cap) {
1591                 struct ceph_cap_flush *cf;
1592                 struct ceph_mds_client *mdsc = fsc->mdsc;
1593
1594                 if (READ_ONCE(fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN) {
1595                         if (inode->i_data.nrpages > 0)
1596                                 invalidate = true;
1597                         if (ci->i_wrbuffer_ref > 0)
1598                                 mapping_set_error(&inode->i_data, -EIO);
1599                 }
1600
1601                 while (!list_empty(&ci->i_cap_flush_list)) {
1602                         cf = list_first_entry(&ci->i_cap_flush_list,
1603                                               struct ceph_cap_flush, i_list);
1604                         list_move(&cf->i_list, &to_remove);
1605                 }
1606
1607                 spin_lock(&mdsc->cap_dirty_lock);
1608
1609                 list_for_each_entry(cf, &to_remove, i_list)
1610                         list_del(&cf->g_list);
1611
1612                 if (!list_empty(&ci->i_dirty_item)) {
1613                         pr_warn_ratelimited(
1614                                 " dropping dirty %s state for %p %lld\n",
1615                                 ceph_cap_string(ci->i_dirty_caps),
1616                                 inode, ceph_ino(inode));
1617                         ci->i_dirty_caps = 0;
1618                         list_del_init(&ci->i_dirty_item);
1619                         dirty_dropped = true;
1620                 }
1621                 if (!list_empty(&ci->i_flushing_item)) {
1622                         pr_warn_ratelimited(
1623                                 " dropping dirty+flushing %s state for %p %lld\n",
1624                                 ceph_cap_string(ci->i_flushing_caps),
1625                                 inode, ceph_ino(inode));
1626                         ci->i_flushing_caps = 0;
1627                         list_del_init(&ci->i_flushing_item);
1628                         mdsc->num_cap_flushing--;
1629                         dirty_dropped = true;
1630                 }
1631                 spin_unlock(&mdsc->cap_dirty_lock);
1632
1633                 if (dirty_dropped) {
1634                         errseq_set(&ci->i_meta_err, -EIO);
1635
1636                         if (ci->i_wrbuffer_ref_head == 0 &&
1637                             ci->i_wr_ref == 0 &&
1638                             ci->i_dirty_caps == 0 &&
1639                             ci->i_flushing_caps == 0) {
1640                                 ceph_put_snap_context(ci->i_head_snapc);
1641                                 ci->i_head_snapc = NULL;
1642                         }
1643                 }
1644
1645                 if (atomic_read(&ci->i_filelock_ref) > 0) {
1646                         /* make further file lock syscall return -EIO */
1647                         ci->i_ceph_flags |= CEPH_I_ERROR_FILELOCK;
1648                         pr_warn_ratelimited(" dropping file locks for %p %lld\n",
1649                                             inode, ceph_ino(inode));
1650                 }
1651
1652                 if (!ci->i_dirty_caps && ci->i_prealloc_cap_flush) {
1653                         list_add(&ci->i_prealloc_cap_flush->i_list, &to_remove);
1654                         ci->i_prealloc_cap_flush = NULL;
1655                 }
1656         }
1657         spin_unlock(&ci->i_ceph_lock);
1658         while (!list_empty(&to_remove)) {
1659                 struct ceph_cap_flush *cf;
1660                 cf = list_first_entry(&to_remove,
1661                                       struct ceph_cap_flush, i_list);
1662                 list_del(&cf->i_list);
1663                 ceph_free_cap_flush(cf);
1664         }
1665
1666         wake_up_all(&ci->i_cap_wq);
1667         if (invalidate)
1668                 ceph_queue_invalidate(inode);
1669         if (dirty_dropped)
1670                 iput(inode);
1671         return 0;
1672 }
1673
1674 /*
1675  * caller must hold session s_mutex
1676  */
1677 static void remove_session_caps(struct ceph_mds_session *session)
1678 {
1679         struct ceph_fs_client *fsc = session->s_mdsc->fsc;
1680         struct super_block *sb = fsc->sb;
1681         LIST_HEAD(dispose);
1682
1683         dout("remove_session_caps on %p\n", session);
1684         ceph_iterate_session_caps(session, remove_session_caps_cb, fsc);
1685
1686         wake_up_all(&fsc->mdsc->cap_flushing_wq);
1687
1688         spin_lock(&session->s_cap_lock);
1689         if (session->s_nr_caps > 0) {
1690                 struct inode *inode;
1691                 struct ceph_cap *cap, *prev = NULL;
1692                 struct ceph_vino vino;
1693                 /*
1694                  * iterate_session_caps() skips inodes that are being
1695                  * deleted, we need to wait until deletions are complete.
1696                  * __wait_on_freeing_inode() is designed for the job,
1697                  * but it is not exported, so use lookup inode function
1698                  * to access it.
1699                  */
1700                 while (!list_empty(&session->s_caps)) {
1701                         cap = list_entry(session->s_caps.next,
1702                                          struct ceph_cap, session_caps);
1703                         if (cap == prev)
1704                                 break;
1705                         prev = cap;
1706                         vino = cap->ci->i_vino;
1707                         spin_unlock(&session->s_cap_lock);
1708
1709                         inode = ceph_find_inode(sb, vino);
1710                          /* avoid calling iput_final() while holding s_mutex */
1711                         ceph_async_iput(inode);
1712
1713                         spin_lock(&session->s_cap_lock);
1714                 }
1715         }
1716
1717         // drop cap expires and unlock s_cap_lock
1718         detach_cap_releases(session, &dispose);
1719
1720         BUG_ON(session->s_nr_caps > 0);
1721         BUG_ON(!list_empty(&session->s_cap_flushing));
1722         spin_unlock(&session->s_cap_lock);
1723         dispose_cap_releases(session->s_mdsc, &dispose);
1724 }
1725
1726 enum {
1727         RECONNECT,
1728         RENEWCAPS,
1729         FORCE_RO,
1730 };
1731
1732 /*
1733  * wake up any threads waiting on this session's caps.  if the cap is
1734  * old (didn't get renewed on the client reconnect), remove it now.
1735  *
1736  * caller must hold s_mutex.
1737  */
1738 static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap,
1739                               void *arg)
1740 {
1741         struct ceph_inode_info *ci = ceph_inode(inode);
1742         unsigned long ev = (unsigned long)arg;
1743
1744         if (ev == RECONNECT) {
1745                 spin_lock(&ci->i_ceph_lock);
1746                 ci->i_wanted_max_size = 0;
1747                 ci->i_requested_max_size = 0;
1748                 spin_unlock(&ci->i_ceph_lock);
1749         } else if (ev == RENEWCAPS) {
1750                 if (cap->cap_gen < cap->session->s_cap_gen) {
1751                         /* mds did not re-issue stale cap */
1752                         spin_lock(&ci->i_ceph_lock);
1753                         cap->issued = cap->implemented = CEPH_CAP_PIN;
1754                         spin_unlock(&ci->i_ceph_lock);
1755                 }
1756         } else if (ev == FORCE_RO) {
1757         }
1758         wake_up_all(&ci->i_cap_wq);
1759         return 0;
1760 }
1761
1762 static void wake_up_session_caps(struct ceph_mds_session *session, int ev)
1763 {
1764         dout("wake_up_session_caps %p mds%d\n", session, session->s_mds);
1765         ceph_iterate_session_caps(session, wake_up_session_cb,
1766                                   (void *)(unsigned long)ev);
1767 }
1768
1769 /*
1770  * Send periodic message to MDS renewing all currently held caps.  The
1771  * ack will reset the expiration for all caps from this session.
1772  *
1773  * caller holds s_mutex
1774  */
1775 static int send_renew_caps(struct ceph_mds_client *mdsc,
1776                            struct ceph_mds_session *session)
1777 {
1778         struct ceph_msg *msg;
1779         int state;
1780
1781         if (time_after_eq(jiffies, session->s_cap_ttl) &&
1782             time_after_eq(session->s_cap_ttl, session->s_renew_requested))
1783                 pr_info("mds%d caps stale\n", session->s_mds);
1784         session->s_renew_requested = jiffies;
1785
1786         /* do not try to renew caps until a recovering mds has reconnected
1787          * with its clients. */
1788         state = ceph_mdsmap_get_state(mdsc->mdsmap, session->s_mds);
1789         if (state < CEPH_MDS_STATE_RECONNECT) {
1790                 dout("send_renew_caps ignoring mds%d (%s)\n",
1791                      session->s_mds, ceph_mds_state_name(state));
1792                 return 0;
1793         }
1794
1795         dout("send_renew_caps to mds%d (%s)\n", session->s_mds,
1796                 ceph_mds_state_name(state));
1797         msg = create_session_msg(CEPH_SESSION_REQUEST_RENEWCAPS,
1798                                  ++session->s_renew_seq);
1799         if (!msg)
1800                 return -ENOMEM;
1801         ceph_con_send(&session->s_con, msg);
1802         return 0;
1803 }
1804
1805 static int send_flushmsg_ack(struct ceph_mds_client *mdsc,
1806                              struct ceph_mds_session *session, u64 seq)
1807 {
1808         struct ceph_msg *msg;
1809
1810         dout("send_flushmsg_ack to mds%d (%s)s seq %lld\n",
1811              session->s_mds, ceph_session_state_name(session->s_state), seq);
1812         msg = create_session_msg(CEPH_SESSION_FLUSHMSG_ACK, seq);
1813         if (!msg)
1814                 return -ENOMEM;
1815         ceph_con_send(&session->s_con, msg);
1816         return 0;
1817 }
1818
1819
1820 /*
1821  * Note new cap ttl, and any transition from stale -> not stale (fresh?).
1822  *
1823  * Called under session->s_mutex
1824  */
1825 static void renewed_caps(struct ceph_mds_client *mdsc,
1826                          struct ceph_mds_session *session, int is_renew)
1827 {
1828         int was_stale;
1829         int wake = 0;
1830
1831         spin_lock(&session->s_cap_lock);
1832         was_stale = is_renew && time_after_eq(jiffies, session->s_cap_ttl);
1833
1834         session->s_cap_ttl = session->s_renew_requested +
1835                 mdsc->mdsmap->m_session_timeout*HZ;
1836
1837         if (was_stale) {
1838                 if (time_before(jiffies, session->s_cap_ttl)) {
1839                         pr_info("mds%d caps renewed\n", session->s_mds);
1840                         wake = 1;
1841                 } else {
1842                         pr_info("mds%d caps still stale\n", session->s_mds);
1843                 }
1844         }
1845         dout("renewed_caps mds%d ttl now %lu, was %s, now %s\n",
1846              session->s_mds, session->s_cap_ttl, was_stale ? "stale" : "fresh",
1847              time_before(jiffies, session->s_cap_ttl) ? "stale" : "fresh");
1848         spin_unlock(&session->s_cap_lock);
1849
1850         if (wake)
1851                 wake_up_session_caps(session, RENEWCAPS);
1852 }
1853
1854 /*
1855  * send a session close request
1856  */
1857 static int request_close_session(struct ceph_mds_session *session)
1858 {
1859         struct ceph_msg *msg;
1860
1861         dout("request_close_session mds%d state %s seq %lld\n",
1862              session->s_mds, ceph_session_state_name(session->s_state),
1863              session->s_seq);
1864         msg = create_session_msg(CEPH_SESSION_REQUEST_CLOSE, session->s_seq);
1865         if (!msg)
1866                 return -ENOMEM;
1867         ceph_con_send(&session->s_con, msg);
1868         return 1;
1869 }
1870
1871 /*
1872  * Called with s_mutex held.
1873  */
1874 static int __close_session(struct ceph_mds_client *mdsc,
1875                          struct ceph_mds_session *session)
1876 {
1877         if (session->s_state >= CEPH_MDS_SESSION_CLOSING)
1878                 return 0;
1879         session->s_state = CEPH_MDS_SESSION_CLOSING;
1880         return request_close_session(session);
1881 }
1882
1883 static bool drop_negative_children(struct dentry *dentry)
1884 {
1885         struct dentry *child;
1886         bool all_negative = true;
1887
1888         if (!d_is_dir(dentry))
1889                 goto out;
1890
1891         spin_lock(&dentry->d_lock);
1892         list_for_each_entry(child, &dentry->d_subdirs, d_child) {
1893                 if (d_really_is_positive(child)) {
1894                         all_negative = false;
1895                         break;
1896                 }
1897         }
1898         spin_unlock(&dentry->d_lock);
1899
1900         if (all_negative)
1901                 shrink_dcache_parent(dentry);
1902 out:
1903         return all_negative;
1904 }
1905
1906 /*
1907  * Trim old(er) caps.
1908  *
1909  * Because we can't cache an inode without one or more caps, we do
1910  * this indirectly: if a cap is unused, we prune its aliases, at which
1911  * point the inode will hopefully get dropped to.
1912  *
1913  * Yes, this is a bit sloppy.  Our only real goal here is to respond to
1914  * memory pressure from the MDS, though, so it needn't be perfect.
1915  */
1916 static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg)
1917 {
1918         int *remaining = arg;
1919         struct ceph_inode_info *ci = ceph_inode(inode);
1920         int used, wanted, oissued, mine;
1921
1922         if (*remaining <= 0)
1923                 return -1;
1924
1925         spin_lock(&ci->i_ceph_lock);
1926         mine = cap->issued | cap->implemented;
1927         used = __ceph_caps_used(ci);
1928         wanted = __ceph_caps_file_wanted(ci);
1929         oissued = __ceph_caps_issued_other(ci, cap);
1930
1931         dout("trim_caps_cb %p cap %p mine %s oissued %s used %s wanted %s\n",
1932              inode, cap, ceph_cap_string(mine), ceph_cap_string(oissued),
1933              ceph_cap_string(used), ceph_cap_string(wanted));
1934         if (cap == ci->i_auth_cap) {
1935                 if (ci->i_dirty_caps || ci->i_flushing_caps ||
1936                     !list_empty(&ci->i_cap_snaps))
1937                         goto out;
1938                 if ((used | wanted) & CEPH_CAP_ANY_WR)
1939                         goto out;
1940                 /* Note: it's possible that i_filelock_ref becomes non-zero
1941                  * after dropping auth caps. It doesn't hurt because reply
1942                  * of lock mds request will re-add auth caps. */
1943                 if (atomic_read(&ci->i_filelock_ref) > 0)
1944                         goto out;
1945         }
1946         /* The inode has cached pages, but it's no longer used.
1947          * we can safely drop it */
1948         if (S_ISREG(inode->i_mode) &&
1949             wanted == 0 && used == CEPH_CAP_FILE_CACHE &&
1950             !(oissued & CEPH_CAP_FILE_CACHE)) {
1951           used = 0;
1952           oissued = 0;
1953         }
1954         if ((used | wanted) & ~oissued & mine)
1955                 goto out;   /* we need these caps */
1956
1957         if (oissued) {
1958                 /* we aren't the only cap.. just remove us */
1959                 __ceph_remove_cap(cap, true);
1960                 (*remaining)--;
1961         } else {
1962                 struct dentry *dentry;
1963                 /* try dropping referring dentries */
1964                 spin_unlock(&ci->i_ceph_lock);
1965                 dentry = d_find_any_alias(inode);
1966                 if (dentry && drop_negative_children(dentry)) {
1967                         int count;
1968                         dput(dentry);
1969                         d_prune_aliases(inode);
1970                         count = atomic_read(&inode->i_count);
1971                         if (count == 1)
1972                                 (*remaining)--;
1973                         dout("trim_caps_cb %p cap %p pruned, count now %d\n",
1974                              inode, cap, count);
1975                 } else {
1976                         dput(dentry);
1977                 }
1978                 return 0;
1979         }
1980
1981 out:
1982         spin_unlock(&ci->i_ceph_lock);
1983         return 0;
1984 }
1985
1986 /*
1987  * Trim session cap count down to some max number.
1988  */
1989 int ceph_trim_caps(struct ceph_mds_client *mdsc,
1990                    struct ceph_mds_session *session,
1991                    int max_caps)
1992 {
1993         int trim_caps = session->s_nr_caps - max_caps;
1994
1995         dout("trim_caps mds%d start: %d / %d, trim %d\n",
1996              session->s_mds, session->s_nr_caps, max_caps, trim_caps);
1997         if (trim_caps > 0) {
1998                 int remaining = trim_caps;
1999
2000                 ceph_iterate_session_caps(session, trim_caps_cb, &remaining);
2001                 dout("trim_caps mds%d done: %d / %d, trimmed %d\n",
2002                      session->s_mds, session->s_nr_caps, max_caps,
2003                         trim_caps - remaining);
2004         }
2005
2006         ceph_flush_cap_releases(mdsc, session);
2007         return 0;
2008 }
2009
2010 static int check_caps_flush(struct ceph_mds_client *mdsc,
2011                             u64 want_flush_tid)
2012 {
2013         int ret = 1;
2014
2015         spin_lock(&mdsc->cap_dirty_lock);
2016         if (!list_empty(&mdsc->cap_flush_list)) {
2017                 struct ceph_cap_flush *cf =
2018                         list_first_entry(&mdsc->cap_flush_list,
2019                                          struct ceph_cap_flush, g_list);
2020                 if (cf->tid <= want_flush_tid) {
2021                         dout("check_caps_flush still flushing tid "
2022                              "%llu <= %llu\n", cf->tid, want_flush_tid);
2023                         ret = 0;
2024                 }
2025         }
2026         spin_unlock(&mdsc->cap_dirty_lock);
2027         return ret;
2028 }
2029
2030 /*
2031  * flush all dirty inode data to disk.
2032  *
2033  * returns true if we've flushed through want_flush_tid
2034  */
2035 static void wait_caps_flush(struct ceph_mds_client *mdsc,
2036                             u64 want_flush_tid)
2037 {
2038         dout("check_caps_flush want %llu\n", want_flush_tid);
2039
2040         wait_event(mdsc->cap_flushing_wq,
2041                    check_caps_flush(mdsc, want_flush_tid));
2042
2043         dout("check_caps_flush ok, flushed thru %llu\n", want_flush_tid);
2044 }
2045
2046 /*
2047  * called under s_mutex
2048  */
2049 static void ceph_send_cap_releases(struct ceph_mds_client *mdsc,
2050                                    struct ceph_mds_session *session)
2051 {
2052         struct ceph_msg *msg = NULL;
2053         struct ceph_mds_cap_release *head;
2054         struct ceph_mds_cap_item *item;
2055         struct ceph_osd_client *osdc = &mdsc->fsc->client->osdc;
2056         struct ceph_cap *cap;
2057         LIST_HEAD(tmp_list);
2058         int num_cap_releases;
2059         __le32  barrier, *cap_barrier;
2060
2061         down_read(&osdc->lock);
2062         barrier = cpu_to_le32(osdc->epoch_barrier);
2063         up_read(&osdc->lock);
2064
2065         spin_lock(&session->s_cap_lock);
2066 again:
2067         list_splice_init(&session->s_cap_releases, &tmp_list);
2068         num_cap_releases = session->s_num_cap_releases;
2069         session->s_num_cap_releases = 0;
2070         spin_unlock(&session->s_cap_lock);
2071
2072         while (!list_empty(&tmp_list)) {
2073                 if (!msg) {
2074                         msg = ceph_msg_new(CEPH_MSG_CLIENT_CAPRELEASE,
2075                                         PAGE_SIZE, GFP_NOFS, false);
2076                         if (!msg)
2077                                 goto out_err;
2078                         head = msg->front.iov_base;
2079                         head->num = cpu_to_le32(0);
2080                         msg->front.iov_len = sizeof(*head);
2081
2082                         msg->hdr.version = cpu_to_le16(2);
2083                         msg->hdr.compat_version = cpu_to_le16(1);
2084                 }
2085
2086                 cap = list_first_entry(&tmp_list, struct ceph_cap,
2087                                         session_caps);
2088                 list_del(&cap->session_caps);
2089                 num_cap_releases--;
2090
2091                 head = msg->front.iov_base;
2092                 put_unaligned_le32(get_unaligned_le32(&head->num) + 1,
2093                                    &head->num);
2094                 item = msg->front.iov_base + msg->front.iov_len;
2095                 item->ino = cpu_to_le64(cap->cap_ino);
2096                 item->cap_id = cpu_to_le64(cap->cap_id);
2097                 item->migrate_seq = cpu_to_le32(cap->mseq);
2098                 item->seq = cpu_to_le32(cap->issue_seq);
2099                 msg->front.iov_len += sizeof(*item);
2100
2101                 ceph_put_cap(mdsc, cap);
2102
2103                 if (le32_to_cpu(head->num) == CEPH_CAPS_PER_RELEASE) {
2104                         // Append cap_barrier field
2105                         cap_barrier = msg->front.iov_base + msg->front.iov_len;
2106                         *cap_barrier = barrier;
2107                         msg->front.iov_len += sizeof(*cap_barrier);
2108
2109                         msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
2110                         dout("send_cap_releases mds%d %p\n", session->s_mds, msg);
2111                         ceph_con_send(&session->s_con, msg);
2112                         msg = NULL;
2113                 }
2114         }
2115
2116         BUG_ON(num_cap_releases != 0);
2117
2118         spin_lock(&session->s_cap_lock);
2119         if (!list_empty(&session->s_cap_releases))
2120                 goto again;
2121         spin_unlock(&session->s_cap_lock);
2122
2123         if (msg) {
2124                 // Append cap_barrier field
2125                 cap_barrier = msg->front.iov_base + msg->front.iov_len;
2126                 *cap_barrier = barrier;
2127                 msg->front.iov_len += sizeof(*cap_barrier);
2128
2129                 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
2130                 dout("send_cap_releases mds%d %p\n", session->s_mds, msg);
2131                 ceph_con_send(&session->s_con, msg);
2132         }
2133         return;
2134 out_err:
2135         pr_err("send_cap_releases mds%d, failed to allocate message\n",
2136                 session->s_mds);
2137         spin_lock(&session->s_cap_lock);
2138         list_splice(&tmp_list, &session->s_cap_releases);
2139         session->s_num_cap_releases += num_cap_releases;
2140         spin_unlock(&session->s_cap_lock);
2141 }
2142
2143 static void ceph_cap_release_work(struct work_struct *work)
2144 {
2145         struct ceph_mds_session *session =
2146                 container_of(work, struct ceph_mds_session, s_cap_release_work);
2147
2148         mutex_lock(&session->s_mutex);
2149         if (session->s_state == CEPH_MDS_SESSION_OPEN ||
2150             session->s_state == CEPH_MDS_SESSION_HUNG)
2151                 ceph_send_cap_releases(session->s_mdsc, session);
2152         mutex_unlock(&session->s_mutex);
2153         ceph_put_mds_session(session);
2154 }
2155
2156 void ceph_flush_cap_releases(struct ceph_mds_client *mdsc,
2157                              struct ceph_mds_session *session)
2158 {
2159         if (mdsc->stopping)
2160                 return;
2161
2162         ceph_get_mds_session(session);
2163         if (queue_work(mdsc->fsc->cap_wq,
2164                        &session->s_cap_release_work)) {
2165                 dout("cap release work queued\n");
2166         } else {
2167                 ceph_put_mds_session(session);
2168                 dout("failed to queue cap release work\n");
2169         }
2170 }
2171
2172 /*
2173  * caller holds session->s_cap_lock
2174  */
2175 void __ceph_queue_cap_release(struct ceph_mds_session *session,
2176                               struct ceph_cap *cap)
2177 {
2178         list_add_tail(&cap->session_caps, &session->s_cap_releases);
2179         session->s_num_cap_releases++;
2180
2181         if (!(session->s_num_cap_releases % CEPH_CAPS_PER_RELEASE))
2182                 ceph_flush_cap_releases(session->s_mdsc, session);
2183 }
2184
2185 static void ceph_cap_reclaim_work(struct work_struct *work)
2186 {
2187         struct ceph_mds_client *mdsc =
2188                 container_of(work, struct ceph_mds_client, cap_reclaim_work);
2189         int ret = ceph_trim_dentries(mdsc);
2190         if (ret == -EAGAIN)
2191                 ceph_queue_cap_reclaim_work(mdsc);
2192 }
2193
2194 void ceph_queue_cap_reclaim_work(struct ceph_mds_client *mdsc)
2195 {
2196         if (mdsc->stopping)
2197                 return;
2198
2199         if (queue_work(mdsc->fsc->cap_wq, &mdsc->cap_reclaim_work)) {
2200                 dout("caps reclaim work queued\n");
2201         } else {
2202                 dout("failed to queue caps release work\n");
2203         }
2204 }
2205
2206 void ceph_reclaim_caps_nr(struct ceph_mds_client *mdsc, int nr)
2207 {
2208         int val;
2209         if (!nr)
2210                 return;
2211         val = atomic_add_return(nr, &mdsc->cap_reclaim_pending);
2212         if ((val % CEPH_CAPS_PER_RELEASE) < nr) {
2213                 atomic_set(&mdsc->cap_reclaim_pending, 0);
2214                 ceph_queue_cap_reclaim_work(mdsc);
2215         }
2216 }
2217
2218 /*
2219  * requests
2220  */
2221
2222 int ceph_alloc_readdir_reply_buffer(struct ceph_mds_request *req,
2223                                     struct inode *dir)
2224 {
2225         struct ceph_inode_info *ci = ceph_inode(dir);
2226         struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
2227         struct ceph_mount_options *opt = req->r_mdsc->fsc->mount_options;
2228         size_t size = sizeof(struct ceph_mds_reply_dir_entry);
2229         unsigned int num_entries;
2230         int order;
2231
2232         spin_lock(&ci->i_ceph_lock);
2233         num_entries = ci->i_files + ci->i_subdirs;
2234         spin_unlock(&ci->i_ceph_lock);
2235         num_entries = max(num_entries, 1U);
2236         num_entries = min(num_entries, opt->max_readdir);
2237
2238         order = get_order(size * num_entries);
2239         while (order >= 0) {
2240                 rinfo->dir_entries = (void*)__get_free_pages(GFP_KERNEL |
2241                                                              __GFP_NOWARN,
2242                                                              order);
2243                 if (rinfo->dir_entries)
2244                         break;
2245                 order--;
2246         }
2247         if (!rinfo->dir_entries)
2248                 return -ENOMEM;
2249
2250         num_entries = (PAGE_SIZE << order) / size;
2251         num_entries = min(num_entries, opt->max_readdir);
2252
2253         rinfo->dir_buf_size = PAGE_SIZE << order;
2254         req->r_num_caps = num_entries + 1;
2255         req->r_args.readdir.max_entries = cpu_to_le32(num_entries);
2256         req->r_args.readdir.max_bytes = cpu_to_le32(opt->max_readdir_bytes);
2257         return 0;
2258 }
2259
2260 /*
2261  * Create an mds request.
2262  */
2263 struct ceph_mds_request *
2264 ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
2265 {
2266         struct ceph_mds_request *req;
2267
2268         req = kmem_cache_zalloc(ceph_mds_request_cachep, GFP_NOFS);
2269         if (!req)
2270                 return ERR_PTR(-ENOMEM);
2271
2272         mutex_init(&req->r_fill_mutex);
2273         req->r_mdsc = mdsc;
2274         req->r_started = jiffies;
2275         req->r_start_latency = ktime_get();
2276         req->r_resend_mds = -1;
2277         INIT_LIST_HEAD(&req->r_unsafe_dir_item);
2278         INIT_LIST_HEAD(&req->r_unsafe_target_item);
2279         req->r_fmode = -1;
2280         kref_init(&req->r_kref);
2281         RB_CLEAR_NODE(&req->r_node);
2282         INIT_LIST_HEAD(&req->r_wait);
2283         init_completion(&req->r_completion);
2284         init_completion(&req->r_safe_completion);
2285         INIT_LIST_HEAD(&req->r_unsafe_item);
2286
2287         ktime_get_coarse_real_ts64(&req->r_stamp);
2288
2289         req->r_op = op;
2290         req->r_direct_mode = mode;
2291         return req;
2292 }
2293
2294 /*
2295  * return oldest (lowest) request, tid in request tree, 0 if none.
2296  *
2297  * called under mdsc->mutex.
2298  */
2299 static struct ceph_mds_request *__get_oldest_req(struct ceph_mds_client *mdsc)
2300 {
2301         if (RB_EMPTY_ROOT(&mdsc->request_tree))
2302                 return NULL;
2303         return rb_entry(rb_first(&mdsc->request_tree),
2304                         struct ceph_mds_request, r_node);
2305 }
2306
2307 static inline  u64 __get_oldest_tid(struct ceph_mds_client *mdsc)
2308 {
2309         return mdsc->oldest_tid;
2310 }
2311
2312 /*
2313  * Build a dentry's path.  Allocate on heap; caller must kfree.  Based
2314  * on build_path_from_dentry in fs/cifs/dir.c.
2315  *
2316  * If @stop_on_nosnap, generate path relative to the first non-snapped
2317  * inode.
2318  *
2319  * Encode hidden .snap dirs as a double /, i.e.
2320  *   foo/.snap/bar -> foo//bar
2321  */
2322 char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *pbase,
2323                            int stop_on_nosnap)
2324 {
2325         struct dentry *temp;
2326         char *path;
2327         int pos;
2328         unsigned seq;
2329         u64 base;
2330
2331         if (!dentry)
2332                 return ERR_PTR(-EINVAL);
2333
2334         path = __getname();
2335         if (!path)
2336                 return ERR_PTR(-ENOMEM);
2337 retry:
2338         pos = PATH_MAX - 1;
2339         path[pos] = '\0';
2340
2341         seq = read_seqbegin(&rename_lock);
2342         rcu_read_lock();
2343         temp = dentry;
2344         for (;;) {
2345                 struct inode *inode;
2346
2347                 spin_lock(&temp->d_lock);
2348                 inode = d_inode(temp);
2349                 if (inode && ceph_snap(inode) == CEPH_SNAPDIR) {
2350                         dout("build_path path+%d: %p SNAPDIR\n",
2351                              pos, temp);
2352                 } else if (stop_on_nosnap && inode && dentry != temp &&
2353                            ceph_snap(inode) == CEPH_NOSNAP) {
2354                         spin_unlock(&temp->d_lock);
2355                         pos++; /* get rid of any prepended '/' */
2356                         break;
2357                 } else {
2358                         pos -= temp->d_name.len;
2359                         if (pos < 0) {
2360                                 spin_unlock(&temp->d_lock);
2361                                 break;
2362                         }
2363                         memcpy(path + pos, temp->d_name.name, temp->d_name.len);
2364                 }
2365                 spin_unlock(&temp->d_lock);
2366                 temp = READ_ONCE(temp->d_parent);
2367
2368                 /* Are we at the root? */
2369                 if (IS_ROOT(temp))
2370                         break;
2371
2372                 /* Are we out of buffer? */
2373                 if (--pos < 0)
2374                         break;
2375
2376                 path[pos] = '/';
2377         }
2378         base = ceph_ino(d_inode(temp));
2379         rcu_read_unlock();
2380
2381         if (read_seqretry(&rename_lock, seq))
2382                 goto retry;
2383
2384         if (pos < 0) {
2385                 /*
2386                  * A rename didn't occur, but somehow we didn't end up where
2387                  * we thought we would. Throw a warning and try again.
2388                  */
2389                 pr_warn("build_path did not end path lookup where "
2390                         "expected, pos is %d\n", pos);
2391                 goto retry;
2392         }
2393
2394         *pbase = base;
2395         *plen = PATH_MAX - 1 - pos;
2396         dout("build_path on %p %d built %llx '%.*s'\n",
2397              dentry, d_count(dentry), base, *plen, path + pos);
2398         return path + pos;
2399 }
2400
2401 static int build_dentry_path(struct dentry *dentry, struct inode *dir,
2402                              const char **ppath, int *ppathlen, u64 *pino,
2403                              bool *pfreepath, bool parent_locked)
2404 {
2405         char *path;
2406
2407         rcu_read_lock();
2408         if (!dir)
2409                 dir = d_inode_rcu(dentry->d_parent);
2410         if (dir && parent_locked && ceph_snap(dir) == CEPH_NOSNAP) {
2411                 *pino = ceph_ino(dir);
2412                 rcu_read_unlock();
2413                 *ppath = dentry->d_name.name;
2414                 *ppathlen = dentry->d_name.len;
2415                 return 0;
2416         }
2417         rcu_read_unlock();
2418         path = ceph_mdsc_build_path(dentry, ppathlen, pino, 1);
2419         if (IS_ERR(path))
2420                 return PTR_ERR(path);
2421         *ppath = path;
2422         *pfreepath = true;
2423         return 0;
2424 }
2425
2426 static int build_inode_path(struct inode *inode,
2427                             const char **ppath, int *ppathlen, u64 *pino,
2428                             bool *pfreepath)
2429 {
2430         struct dentry *dentry;
2431         char *path;
2432
2433         if (ceph_snap(inode) == CEPH_NOSNAP) {
2434                 *pino = ceph_ino(inode);
2435                 *ppathlen = 0;
2436                 return 0;
2437         }
2438         dentry = d_find_alias(inode);
2439         path = ceph_mdsc_build_path(dentry, ppathlen, pino, 1);
2440         dput(dentry);
2441         if (IS_ERR(path))
2442                 return PTR_ERR(path);
2443         *ppath = path;
2444         *pfreepath = true;
2445         return 0;
2446 }
2447
2448 /*
2449  * request arguments may be specified via an inode *, a dentry *, or
2450  * an explicit ino+path.
2451  */
2452 static int set_request_path_attr(struct inode *rinode, struct dentry *rdentry,
2453                                   struct inode *rdiri, const char *rpath,
2454                                   u64 rino, const char **ppath, int *pathlen,
2455                                   u64 *ino, bool *freepath, bool parent_locked)
2456 {
2457         int r = 0;
2458
2459         if (rinode) {
2460                 r = build_inode_path(rinode, ppath, pathlen, ino, freepath);
2461                 dout(" inode %p %llx.%llx\n", rinode, ceph_ino(rinode),
2462                      ceph_snap(rinode));
2463         } else if (rdentry) {
2464                 r = build_dentry_path(rdentry, rdiri, ppath, pathlen, ino,
2465                                         freepath, parent_locked);
2466                 dout(" dentry %p %llx/%.*s\n", rdentry, *ino, *pathlen,
2467                      *ppath);
2468         } else if (rpath || rino) {
2469                 *ino = rino;
2470                 *ppath = rpath;
2471                 *pathlen = rpath ? strlen(rpath) : 0;
2472                 dout(" path %.*s\n", *pathlen, rpath);
2473         }
2474
2475         return r;
2476 }
2477
2478 static void encode_timestamp_and_gids(void **p,
2479                                       const struct ceph_mds_request *req)
2480 {
2481         struct ceph_timespec ts;
2482         int i;
2483
2484         ceph_encode_timespec64(&ts, &req->r_stamp);
2485         ceph_encode_copy(p, &ts, sizeof(ts));
2486
2487         /* gid_list */
2488         ceph_encode_32(p, req->r_cred->group_info->ngroups);
2489         for (i = 0; i < req->r_cred->group_info->ngroups; i++)
2490                 ceph_encode_64(p, from_kgid(&init_user_ns,
2491                                             req->r_cred->group_info->gid[i]));
2492 }
2493
2494 /*
2495  * called under mdsc->mutex
2496  */
2497 static struct ceph_msg *create_request_message(struct ceph_mds_session *session,
2498                                                struct ceph_mds_request *req,
2499                                                bool drop_cap_releases)
2500 {
2501         int mds = session->s_mds;
2502         struct ceph_mds_client *mdsc = session->s_mdsc;
2503         struct ceph_msg *msg;
2504         struct ceph_mds_request_head_old *head;
2505         const char *path1 = NULL;
2506         const char *path2 = NULL;
2507         u64 ino1 = 0, ino2 = 0;
2508         int pathlen1 = 0, pathlen2 = 0;
2509         bool freepath1 = false, freepath2 = false;
2510         int len;
2511         u16 releases;
2512         void *p, *end;
2513         int ret;
2514         bool legacy = !(session->s_con.peer_features & CEPH_FEATURE_FS_BTIME);
2515
2516         ret = set_request_path_attr(req->r_inode, req->r_dentry,
2517                               req->r_parent, req->r_path1, req->r_ino1.ino,
2518                               &path1, &pathlen1, &ino1, &freepath1,
2519                               test_bit(CEPH_MDS_R_PARENT_LOCKED,
2520                                         &req->r_req_flags));
2521         if (ret < 0) {
2522                 msg = ERR_PTR(ret);
2523                 goto out;
2524         }
2525
2526         /* If r_old_dentry is set, then assume that its parent is locked */
2527         ret = set_request_path_attr(NULL, req->r_old_dentry,
2528                               req->r_old_dentry_dir,
2529                               req->r_path2, req->r_ino2.ino,
2530                               &path2, &pathlen2, &ino2, &freepath2, true);
2531         if (ret < 0) {
2532                 msg = ERR_PTR(ret);
2533                 goto out_free1;
2534         }
2535
2536         len = legacy ? sizeof(*head) : sizeof(struct ceph_mds_request_head);
2537         len += pathlen1 + pathlen2 + 2*(1 + sizeof(u32) + sizeof(u64)) +
2538                 sizeof(struct ceph_timespec);
2539         len += sizeof(u32) + (sizeof(u64) * req->r_cred->group_info->ngroups);
2540
2541         /* calculate (max) length for cap releases */
2542         len += sizeof(struct ceph_mds_request_release) *
2543                 (!!req->r_inode_drop + !!req->r_dentry_drop +
2544                  !!req->r_old_inode_drop + !!req->r_old_dentry_drop);
2545
2546         if (req->r_dentry_drop)
2547                 len += pathlen1;
2548         if (req->r_old_dentry_drop)
2549                 len += pathlen2;
2550
2551         msg = ceph_msg_new2(CEPH_MSG_CLIENT_REQUEST, len, 1, GFP_NOFS, false);
2552         if (!msg) {
2553                 msg = ERR_PTR(-ENOMEM);
2554                 goto out_free2;
2555         }
2556
2557         msg->hdr.tid = cpu_to_le64(req->r_tid);
2558
2559         /*
2560          * The old ceph_mds_request_head didn't contain a version field, and
2561          * one was added when we moved the message version from 3->4.
2562          */
2563         if (legacy) {
2564                 msg->hdr.version = cpu_to_le16(3);
2565                 head = msg->front.iov_base;
2566                 p = msg->front.iov_base + sizeof(*head);
2567         } else {
2568                 struct ceph_mds_request_head *new_head = msg->front.iov_base;
2569
2570                 msg->hdr.version = cpu_to_le16(4);
2571                 new_head->version = cpu_to_le16(CEPH_MDS_REQUEST_HEAD_VERSION);
2572                 head = (struct ceph_mds_request_head_old *)&new_head->oldest_client_tid;
2573                 p = msg->front.iov_base + sizeof(*new_head);
2574         }
2575
2576         end = msg->front.iov_base + msg->front.iov_len;
2577
2578         head->mdsmap_epoch = cpu_to_le32(mdsc->mdsmap->m_epoch);
2579         head->op = cpu_to_le32(req->r_op);
2580         head->caller_uid = cpu_to_le32(from_kuid(&init_user_ns,
2581                                                  req->r_cred->fsuid));
2582         head->caller_gid = cpu_to_le32(from_kgid(&init_user_ns,
2583                                                  req->r_cred->fsgid));
2584         head->ino = cpu_to_le64(req->r_deleg_ino);
2585         head->args = req->r_args;
2586
2587         ceph_encode_filepath(&p, end, ino1, path1);
2588         ceph_encode_filepath(&p, end, ino2, path2);
2589
2590         /* make note of release offset, in case we need to replay */
2591         req->r_request_release_offset = p - msg->front.iov_base;
2592
2593         /* cap releases */
2594         releases = 0;
2595         if (req->r_inode_drop)
2596                 releases += ceph_encode_inode_release(&p,
2597                       req->r_inode ? req->r_inode : d_inode(req->r_dentry),
2598                       mds, req->r_inode_drop, req->r_inode_unless,
2599                       req->r_op == CEPH_MDS_OP_READDIR);
2600         if (req->r_dentry_drop)
2601                 releases += ceph_encode_dentry_release(&p, req->r_dentry,
2602                                 req->r_parent, mds, req->r_dentry_drop,
2603                                 req->r_dentry_unless);
2604         if (req->r_old_dentry_drop)
2605                 releases += ceph_encode_dentry_release(&p, req->r_old_dentry,
2606                                 req->r_old_dentry_dir, mds,
2607                                 req->r_old_dentry_drop,
2608                                 req->r_old_dentry_unless);
2609         if (req->r_old_inode_drop)
2610                 releases += ceph_encode_inode_release(&p,
2611                       d_inode(req->r_old_dentry),
2612                       mds, req->r_old_inode_drop, req->r_old_inode_unless, 0);
2613
2614         if (drop_cap_releases) {
2615                 releases = 0;
2616                 p = msg->front.iov_base + req->r_request_release_offset;
2617         }
2618
2619         head->num_releases = cpu_to_le16(releases);
2620
2621         encode_timestamp_and_gids(&p, req);
2622
2623         if (WARN_ON_ONCE(p > end)) {
2624                 ceph_msg_put(msg);
2625                 msg = ERR_PTR(-ERANGE);
2626                 goto out_free2;
2627         }
2628
2629         msg->front.iov_len = p - msg->front.iov_base;
2630         msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
2631
2632         if (req->r_pagelist) {
2633                 struct ceph_pagelist *pagelist = req->r_pagelist;
2634                 ceph_msg_data_add_pagelist(msg, pagelist);
2635                 msg->hdr.data_len = cpu_to_le32(pagelist->length);
2636         } else {
2637                 msg->hdr.data_len = 0;
2638         }
2639
2640         msg->hdr.data_off = cpu_to_le16(0);
2641
2642 out_free2:
2643         if (freepath2)
2644                 ceph_mdsc_free_path((char *)path2, pathlen2);
2645 out_free1:
2646         if (freepath1)
2647                 ceph_mdsc_free_path((char *)path1, pathlen1);
2648 out:
2649         return msg;
2650 }
2651
2652 /*
2653  * called under mdsc->mutex if error, under no mutex if
2654  * success.
2655  */
2656 static void complete_request(struct ceph_mds_client *mdsc,
2657                              struct ceph_mds_request *req)
2658 {
2659         req->r_end_latency = ktime_get();
2660
2661         if (req->r_callback)
2662                 req->r_callback(mdsc, req);
2663         complete_all(&req->r_completion);
2664 }
2665
2666 static struct ceph_mds_request_head_old *
2667 find_old_request_head(void *p, u64 features)
2668 {
2669         bool legacy = !(features & CEPH_FEATURE_FS_BTIME);
2670         struct ceph_mds_request_head *new_head;
2671
2672         if (legacy)
2673                 return (struct ceph_mds_request_head_old *)p;
2674         new_head = (struct ceph_mds_request_head *)p;
2675         return (struct ceph_mds_request_head_old *)&new_head->oldest_client_tid;
2676 }
2677
2678 /*
2679  * called under mdsc->mutex
2680  */
2681 static int __prepare_send_request(struct ceph_mds_session *session,
2682                                   struct ceph_mds_request *req,
2683                                   bool drop_cap_releases)
2684 {
2685         int mds = session->s_mds;
2686         struct ceph_mds_client *mdsc = session->s_mdsc;
2687         struct ceph_mds_request_head_old *rhead;
2688         struct ceph_msg *msg;
2689         int flags = 0;
2690
2691         req->r_attempts++;
2692         if (req->r_inode) {
2693                 struct ceph_cap *cap =
2694                         ceph_get_cap_for_mds(ceph_inode(req->r_inode), mds);
2695
2696                 if (cap)
2697                         req->r_sent_on_mseq = cap->mseq;
2698                 else
2699                         req->r_sent_on_mseq = -1;
2700         }
2701         dout("prepare_send_request %p tid %lld %s (attempt %d)\n", req,
2702              req->r_tid, ceph_mds_op_name(req->r_op), req->r_attempts);
2703
2704         if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
2705                 void *p;
2706
2707                 /*
2708                  * Replay.  Do not regenerate message (and rebuild
2709                  * paths, etc.); just use the original message.
2710                  * Rebuilding paths will break for renames because
2711                  * d_move mangles the src name.
2712                  */
2713                 msg = req->r_request;
2714                 rhead = find_old_request_head(msg->front.iov_base,
2715                                               session->s_con.peer_features);
2716
2717                 flags = le32_to_cpu(rhead->flags);
2718                 flags |= CEPH_MDS_FLAG_REPLAY;
2719                 rhead->flags = cpu_to_le32(flags);
2720
2721                 if (req->r_target_inode)
2722                         rhead->ino = cpu_to_le64(ceph_ino(req->r_target_inode));
2723
2724                 rhead->num_retry = req->r_attempts - 1;
2725
2726                 /* remove cap/dentry releases from message */
2727                 rhead->num_releases = 0;
2728
2729                 p = msg->front.iov_base + req->r_request_release_offset;
2730                 encode_timestamp_and_gids(&p, req);
2731
2732                 msg->front.iov_len = p - msg->front.iov_base;
2733                 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
2734                 return 0;
2735         }
2736
2737         if (req->r_request) {
2738                 ceph_msg_put(req->r_request);
2739                 req->r_request = NULL;
2740         }
2741         msg = create_request_message(session, req, drop_cap_releases);
2742         if (IS_ERR(msg)) {
2743                 req->r_err = PTR_ERR(msg);
2744                 return PTR_ERR(msg);
2745         }
2746         req->r_request = msg;
2747
2748         rhead = find_old_request_head(msg->front.iov_base,
2749                                       session->s_con.peer_features);
2750         rhead->oldest_client_tid = cpu_to_le64(__get_oldest_tid(mdsc));
2751         if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags))
2752                 flags |= CEPH_MDS_FLAG_REPLAY;
2753         if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags))
2754                 flags |= CEPH_MDS_FLAG_ASYNC;
2755         if (req->r_parent)
2756                 flags |= CEPH_MDS_FLAG_WANT_DENTRY;
2757         rhead->flags = cpu_to_le32(flags);
2758         rhead->num_fwd = req->r_num_fwd;
2759         rhead->num_retry = req->r_attempts - 1;
2760
2761         dout(" r_parent = %p\n", req->r_parent);
2762         return 0;
2763 }
2764
2765 /*
2766  * called under mdsc->mutex
2767  */
2768 static int __send_request(struct ceph_mds_session *session,
2769                           struct ceph_mds_request *req,
2770                           bool drop_cap_releases)
2771 {
2772         int err;
2773
2774         err = __prepare_send_request(session, req, drop_cap_releases);
2775         if (!err) {
2776                 ceph_msg_get(req->r_request);
2777                 ceph_con_send(&session->s_con, req->r_request);
2778         }
2779
2780         return err;
2781 }
2782
2783 /*
2784  * send request, or put it on the appropriate wait list.
2785  */
2786 static void __do_request(struct ceph_mds_client *mdsc,
2787                         struct ceph_mds_request *req)
2788 {
2789         struct ceph_mds_session *session = NULL;
2790         int mds = -1;
2791         int err = 0;
2792         bool random;
2793
2794         if (req->r_err || test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags)) {
2795                 if (test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags))
2796                         __unregister_request(mdsc, req);
2797                 return;
2798         }
2799
2800         if (req->r_timeout &&
2801             time_after_eq(jiffies, req->r_started + req->r_timeout)) {
2802                 dout("do_request timed out\n");
2803                 err = -ETIMEDOUT;
2804                 goto finish;
2805         }
2806         if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_SHUTDOWN) {
2807                 dout("do_request forced umount\n");
2808                 err = -EIO;
2809                 goto finish;
2810         }
2811         if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_MOUNTING) {
2812                 if (mdsc->mdsmap_err) {
2813                         err = mdsc->mdsmap_err;
2814                         dout("do_request mdsmap err %d\n", err);
2815                         goto finish;
2816                 }
2817                 if (mdsc->mdsmap->m_epoch == 0) {
2818                         dout("do_request no mdsmap, waiting for map\n");
2819                         list_add(&req->r_wait, &mdsc->waiting_for_map);
2820                         return;
2821                 }
2822                 if (!(mdsc->fsc->mount_options->flags &
2823                       CEPH_MOUNT_OPT_MOUNTWAIT) &&
2824                     !ceph_mdsmap_is_cluster_available(mdsc->mdsmap)) {
2825                         err = -EHOSTUNREACH;
2826                         goto finish;
2827                 }
2828         }
2829
2830         put_request_session(req);
2831
2832         mds = __choose_mds(mdsc, req, &random);
2833         if (mds < 0 ||
2834             ceph_mdsmap_get_state(mdsc->mdsmap, mds) < CEPH_MDS_STATE_ACTIVE) {
2835                 if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags)) {
2836                         err = -EJUKEBOX;
2837                         goto finish;
2838                 }
2839                 dout("do_request no mds or not active, waiting for map\n");
2840                 list_add(&req->r_wait, &mdsc->waiting_for_map);
2841                 return;
2842         }
2843
2844         /* get, open session */
2845         session = __ceph_lookup_mds_session(mdsc, mds);
2846         if (!session) {
2847                 session = register_session(mdsc, mds);
2848                 if (IS_ERR(session)) {
2849                         err = PTR_ERR(session);
2850                         goto finish;
2851                 }
2852         }
2853         req->r_session = ceph_get_mds_session(session);
2854
2855         dout("do_request mds%d session %p state %s\n", mds, session,
2856              ceph_session_state_name(session->s_state));
2857         if (session->s_state != CEPH_MDS_SESSION_OPEN &&
2858             session->s_state != CEPH_MDS_SESSION_HUNG) {
2859                 /*
2860                  * We cannot queue async requests since the caps and delegated
2861                  * inodes are bound to the session. Just return -EJUKEBOX and
2862                  * let the caller retry a sync request in that case.
2863                  */
2864                 if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags)) {
2865                         err = -EJUKEBOX;
2866                         goto out_session;
2867                 }
2868
2869                 /*
2870                  * If the session has been REJECTED, then return a hard error,
2871                  * unless it's a CLEANRECOVER mount, in which case we'll queue
2872                  * it to the mdsc queue.
2873                  */
2874                 if (session->s_state == CEPH_MDS_SESSION_REJECTED) {
2875                         if (ceph_test_mount_opt(mdsc->fsc, CLEANRECOVER))
2876                                 list_add(&req->r_wait, &mdsc->waiting_for_map);
2877                         else
2878                                 err = -EACCES;
2879                         goto out_session;
2880                 }
2881
2882                 if (session->s_state == CEPH_MDS_SESSION_NEW ||
2883                     session->s_state == CEPH_MDS_SESSION_CLOSING) {
2884                         err = __open_session(mdsc, session);
2885                         if (err)
2886                                 goto out_session;
2887                         /* retry the same mds later */
2888                         if (random)
2889                                 req->r_resend_mds = mds;
2890                 }
2891                 list_add(&req->r_wait, &session->s_waiting);
2892                 goto out_session;
2893         }
2894
2895         /* send request */
2896         req->r_resend_mds = -1;   /* forget any previous mds hint */
2897
2898         if (req->r_request_started == 0)   /* note request start time */
2899                 req->r_request_started = jiffies;
2900
2901         err = __send_request(session, req, false);
2902
2903 out_session:
2904         ceph_put_mds_session(session);
2905 finish:
2906         if (err) {
2907                 dout("__do_request early error %d\n", err);
2908                 req->r_err = err;
2909                 complete_request(mdsc, req);
2910                 __unregister_request(mdsc, req);
2911         }
2912         return;
2913 }
2914
2915 /*
2916  * called under mdsc->mutex
2917  */
2918 static void __wake_requests(struct ceph_mds_client *mdsc,
2919                             struct list_head *head)
2920 {
2921         struct ceph_mds_request *req;
2922         LIST_HEAD(tmp_list);
2923
2924         list_splice_init(head, &tmp_list);
2925
2926         while (!list_empty(&tmp_list)) {
2927                 req = list_entry(tmp_list.next,
2928                                  struct ceph_mds_request, r_wait);
2929                 list_del_init(&req->r_wait);
2930                 dout(" wake request %p tid %llu\n", req, req->r_tid);
2931                 __do_request(mdsc, req);
2932         }
2933 }
2934
2935 /*
2936  * Wake up threads with requests pending for @mds, so that they can
2937  * resubmit their requests to a possibly different mds.
2938  */
2939 static void kick_requests(struct ceph_mds_client *mdsc, int mds)
2940 {
2941         struct ceph_mds_request *req;
2942         struct rb_node *p = rb_first(&mdsc->request_tree);
2943
2944         dout("kick_requests mds%d\n", mds);
2945         while (p) {
2946                 req = rb_entry(p, struct ceph_mds_request, r_node);
2947                 p = rb_next(p);
2948                 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags))
2949                         continue;
2950                 if (req->r_attempts > 0)
2951                         continue; /* only new requests */
2952                 if (req->r_session &&
2953                     req->r_session->s_mds == mds) {
2954                         dout(" kicking tid %llu\n", req->r_tid);
2955                         list_del_init(&req->r_wait);
2956                         __do_request(mdsc, req);
2957                 }
2958         }
2959 }
2960
2961 int ceph_mdsc_submit_request(struct ceph_mds_client *mdsc, struct inode *dir,
2962                               struct ceph_mds_request *req)
2963 {
2964         int err = 0;
2965
2966         /* take CAP_PIN refs for r_inode, r_parent, r_old_dentry */
2967         if (req->r_inode)
2968                 ceph_get_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
2969         if (req->r_parent) {
2970                 struct ceph_inode_info *ci = ceph_inode(req->r_parent);
2971                 int fmode = (req->r_op & CEPH_MDS_OP_WRITE) ?
2972                             CEPH_FILE_MODE_WR : CEPH_FILE_MODE_RD;
2973                 spin_lock(&ci->i_ceph_lock);
2974                 ceph_take_cap_refs(ci, CEPH_CAP_PIN, false);
2975                 __ceph_touch_fmode(ci, mdsc, fmode);
2976                 spin_unlock(&ci->i_ceph_lock);
2977                 ihold(req->r_parent);
2978         }
2979         if (req->r_old_dentry_dir)
2980                 ceph_get_cap_refs(ceph_inode(req->r_old_dentry_dir),
2981                                   CEPH_CAP_PIN);
2982
2983         if (req->r_inode) {
2984                 err = ceph_wait_on_async_create(req->r_inode);
2985                 if (err) {
2986                         dout("%s: wait for async create returned: %d\n",
2987                              __func__, err);
2988                         return err;
2989                 }
2990         }
2991
2992         if (!err && req->r_old_inode) {
2993                 err = ceph_wait_on_async_create(req->r_old_inode);
2994                 if (err) {
2995                         dout("%s: wait for async create returned: %d\n",
2996                              __func__, err);
2997                         return err;
2998                 }
2999         }
3000
3001         dout("submit_request on %p for inode %p\n", req, dir);
3002         mutex_lock(&mdsc->mutex);
3003         __register_request(mdsc, req, dir);
3004         __do_request(mdsc, req);
3005         err = req->r_err;
3006         mutex_unlock(&mdsc->mutex);
3007         return err;
3008 }
3009
3010 static int ceph_mdsc_wait_request(struct ceph_mds_client *mdsc,
3011                                   struct ceph_mds_request *req)
3012 {
3013         int err;
3014
3015         /* wait */
3016         dout("do_request waiting\n");
3017         if (!req->r_timeout && req->r_wait_for_completion) {
3018                 err = req->r_wait_for_completion(mdsc, req);
3019         } else {
3020                 long timeleft = wait_for_completion_killable_timeout(
3021                                         &req->r_completion,
3022                                         ceph_timeout_jiffies(req->r_timeout));
3023                 if (timeleft > 0)
3024                         err = 0;
3025                 else if (!timeleft)
3026                         err = -ETIMEDOUT;  /* timed out */
3027                 else
3028                         err = timeleft;  /* killed */
3029         }
3030         dout("do_request waited, got %d\n", err);
3031         mutex_lock(&mdsc->mutex);
3032
3033         /* only abort if we didn't race with a real reply */
3034         if (test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags)) {
3035                 err = le32_to_cpu(req->r_reply_info.head->result);
3036         } else if (err < 0) {
3037                 dout("aborted request %lld with %d\n", req->r_tid, err);
3038
3039                 /*
3040                  * ensure we aren't running concurrently with
3041                  * ceph_fill_trace or ceph_readdir_prepopulate, which
3042                  * rely on locks (dir mutex) held by our caller.
3043                  */
3044                 mutex_lock(&req->r_fill_mutex);
3045                 req->r_err = err;
3046                 set_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags);
3047                 mutex_unlock(&req->r_fill_mutex);
3048
3049                 if (req->r_parent &&
3050                     (req->r_op & CEPH_MDS_OP_WRITE))
3051                         ceph_invalidate_dir_request(req);
3052         } else {
3053                 err = req->r_err;
3054         }
3055
3056         mutex_unlock(&mdsc->mutex);
3057         return err;
3058 }
3059
3060 /*
3061  * Synchrously perform an mds request.  Take care of all of the
3062  * session setup, forwarding, retry details.
3063  */
3064 int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
3065                          struct inode *dir,
3066                          struct ceph_mds_request *req)
3067 {
3068         int err;
3069
3070         dout("do_request on %p\n", req);
3071
3072         /* issue */
3073         err = ceph_mdsc_submit_request(mdsc, dir, req);
3074         if (!err)
3075                 err = ceph_mdsc_wait_request(mdsc, req);
3076         dout("do_request %p done, result %d\n", req, err);
3077         return err;
3078 }
3079
3080 /*
3081  * Invalidate dir's completeness, dentry lease state on an aborted MDS
3082  * namespace request.
3083  */
3084 void ceph_invalidate_dir_request(struct ceph_mds_request *req)
3085 {
3086         struct inode *dir = req->r_parent;
3087         struct inode *old_dir = req->r_old_dentry_dir;
3088
3089         dout("invalidate_dir_request %p %p (complete, lease(s))\n", dir, old_dir);
3090
3091         ceph_dir_clear_complete(dir);
3092         if (old_dir)
3093                 ceph_dir_clear_complete(old_dir);
3094         if (req->r_dentry)
3095                 ceph_invalidate_dentry_lease(req->r_dentry);
3096         if (req->r_old_dentry)
3097                 ceph_invalidate_dentry_lease(req->r_old_dentry);
3098 }
3099
3100 /*
3101  * Handle mds reply.
3102  *
3103  * We take the session mutex and parse and process the reply immediately.
3104  * This preserves the logical ordering of replies, capabilities, etc., sent
3105  * by the MDS as they are applied to our local cache.
3106  */
3107 static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
3108 {
3109         struct ceph_mds_client *mdsc = session->s_mdsc;
3110         struct ceph_mds_request *req;
3111         struct ceph_mds_reply_head *head = msg->front.iov_base;
3112         struct ceph_mds_reply_info_parsed *rinfo;  /* parsed reply info */
3113         struct ceph_snap_realm *realm;
3114         u64 tid;
3115         int err, result;
3116         int mds = session->s_mds;
3117
3118         if (msg->front.iov_len < sizeof(*head)) {
3119                 pr_err("mdsc_handle_reply got corrupt (short) reply\n");
3120                 ceph_msg_dump(msg);
3121                 return;
3122         }
3123
3124         /* get request, session */
3125         tid = le64_to_cpu(msg->hdr.tid);
3126         mutex_lock(&mdsc->mutex);
3127         req = lookup_get_request(mdsc, tid);
3128         if (!req) {
3129                 dout("handle_reply on unknown tid %llu\n", tid);
3130                 mutex_unlock(&mdsc->mutex);
3131                 return;
3132         }
3133         dout("handle_reply %p\n", req);
3134
3135         /* correct session? */
3136         if (req->r_session != session) {
3137                 pr_err("mdsc_handle_reply got %llu on session mds%d"
3138                        " not mds%d\n", tid, session->s_mds,
3139                        req->r_session ? req->r_session->s_mds : -1);
3140                 mutex_unlock(&mdsc->mutex);
3141                 goto out;
3142         }
3143
3144         /* dup? */
3145         if ((test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags) && !head->safe) ||
3146             (test_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags) && head->safe)) {
3147                 pr_warn("got a dup %s reply on %llu from mds%d\n",
3148                            head->safe ? "safe" : "unsafe", tid, mds);
3149                 mutex_unlock(&mdsc->mutex);
3150                 goto out;
3151         }
3152         if (test_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags)) {
3153                 pr_warn("got unsafe after safe on %llu from mds%d\n",
3154                            tid, mds);
3155                 mutex_unlock(&mdsc->mutex);
3156                 goto out;
3157         }
3158
3159         result = le32_to_cpu(head->result);
3160
3161         /*
3162          * Handle an ESTALE
3163          * if we're not talking to the authority, send to them
3164          * if the authority has changed while we weren't looking,
3165          * send to new authority
3166          * Otherwise we just have to return an ESTALE
3167          */
3168         if (result == -ESTALE) {
3169                 dout("got ESTALE on request %llu\n", req->r_tid);
3170                 req->r_resend_mds = -1;
3171                 if (req->r_direct_mode != USE_AUTH_MDS) {
3172                         dout("not using auth, setting for that now\n");
3173                         req->r_direct_mode = USE_AUTH_MDS;
3174                         __do_request(mdsc, req);
3175                         mutex_unlock(&mdsc->mutex);
3176                         goto out;
3177                 } else  {
3178                         int mds = __choose_mds(mdsc, req, NULL);
3179                         if (mds >= 0 && mds != req->r_session->s_mds) {
3180                                 dout("but auth changed, so resending\n");
3181                                 __do_request(mdsc, req);
3182                                 mutex_unlock(&mdsc->mutex);
3183                                 goto out;
3184                         }
3185                 }
3186                 dout("have to return ESTALE on request %llu\n", req->r_tid);
3187         }
3188
3189
3190         if (head->safe) {
3191                 set_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags);
3192                 __unregister_request(mdsc, req);
3193
3194                 /* last request during umount? */
3195                 if (mdsc->stopping && !__get_oldest_req(mdsc))
3196                         complete_all(&mdsc->safe_umount_waiters);
3197
3198                 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
3199                         /*
3200                          * We already handled the unsafe response, now do the
3201                          * cleanup.  No need to examine the response; the MDS
3202                          * doesn't include any result info in the safe
3203                          * response.  And even if it did, there is nothing
3204                          * useful we could do with a revised return value.
3205                          */
3206                         dout("got safe reply %llu, mds%d\n", tid, mds);
3207
3208                         mutex_unlock(&mdsc->mutex);
3209                         goto out;
3210                 }
3211         } else {
3212                 set_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags);
3213                 list_add_tail(&req->r_unsafe_item, &req->r_session->s_unsafe);
3214         }
3215
3216         dout("handle_reply tid %lld result %d\n", tid, result);
3217         rinfo = &req->r_reply_info;
3218         if (test_bit(CEPHFS_FEATURE_REPLY_ENCODING, &session->s_features))
3219                 err = parse_reply_info(session, msg, rinfo, (u64)-1);
3220         else
3221                 err = parse_reply_info(session, msg, rinfo, session->s_con.peer_features);
3222         mutex_unlock(&mdsc->mutex);
3223
3224         /* Must find target inode outside of mutexes to avoid deadlocks */
3225         if ((err >= 0) && rinfo->head->is_target) {
3226                 struct inode *in;
3227                 struct ceph_vino tvino = {
3228                         .ino  = le64_to_cpu(rinfo->targeti.in->ino),
3229                         .snap = le64_to_cpu(rinfo->targeti.in->snapid)
3230                 };
3231
3232                 in = ceph_get_inode(mdsc->fsc->sb, tvino);
3233                 if (IS_ERR(in)) {
3234                         err = PTR_ERR(in);
3235                         mutex_lock(&session->s_mutex);
3236                         goto out_err;
3237                 }
3238                 req->r_target_inode = in;
3239         }
3240
3241         mutex_lock(&session->s_mutex);
3242         if (err < 0) {
3243                 pr_err("mdsc_handle_reply got corrupt reply mds%d(tid:%lld)\n", mds, tid);
3244                 ceph_msg_dump(msg);
3245                 goto out_err;
3246         }
3247
3248         /* snap trace */
3249         realm = NULL;
3250         if (rinfo->snapblob_len) {
3251                 down_write(&mdsc->snap_rwsem);
3252                 ceph_update_snap_trace(mdsc, rinfo->snapblob,
3253                                 rinfo->snapblob + rinfo->snapblob_len,
3254                                 le32_to_cpu(head->op) == CEPH_MDS_OP_RMSNAP,
3255                                 &realm);
3256                 downgrade_write(&mdsc->snap_rwsem);
3257         } else {
3258                 down_read(&mdsc->snap_rwsem);
3259         }
3260
3261         /* insert trace into our cache */
3262         mutex_lock(&req->r_fill_mutex);
3263         current->journal_info = req;
3264         err = ceph_fill_trace(mdsc->fsc->sb, req);
3265         if (err == 0) {
3266                 if (result == 0 && (req->r_op == CEPH_MDS_OP_READDIR ||
3267                                     req->r_op == CEPH_MDS_OP_LSSNAP))
3268                         ceph_readdir_prepopulate(req, req->r_session);
3269         }
3270         current->journal_info = NULL;
3271         mutex_unlock(&req->r_fill_mutex);
3272
3273         up_read(&mdsc->snap_rwsem);
3274         if (realm)
3275                 ceph_put_snap_realm(mdsc, realm);
3276
3277         if (err == 0) {
3278                 if (req->r_target_inode &&
3279                     test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
3280                         struct ceph_inode_info *ci =
3281                                 ceph_inode(req->r_target_inode);
3282                         spin_lock(&ci->i_unsafe_lock);
3283                         list_add_tail(&req->r_unsafe_target_item,
3284                                       &ci->i_unsafe_iops);
3285                         spin_unlock(&ci->i_unsafe_lock);
3286                 }
3287
3288                 ceph_unreserve_caps(mdsc, &req->r_caps_reservation);
3289         }
3290 out_err:
3291         mutex_lock(&mdsc->mutex);
3292         if (!test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags)) {
3293                 if (err) {
3294                         req->r_err = err;
3295                 } else {
3296                         req->r_reply =  ceph_msg_get(msg);
3297                         set_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags);
3298                 }
3299         } else {
3300                 dout("reply arrived after request %lld was aborted\n", tid);
3301         }
3302         mutex_unlock(&mdsc->mutex);
3303
3304         mutex_unlock(&session->s_mutex);
3305
3306         /* kick calling process */
3307         complete_request(mdsc, req);
3308
3309         ceph_update_metadata_latency(&mdsc->metric, req->r_start_latency,
3310                                      req->r_end_latency, err);
3311 out:
3312         ceph_mdsc_put_request(req);
3313         return;
3314 }
3315
3316
3317
3318 /*
3319  * handle mds notification that our request has been forwarded.
3320  */
3321 static void handle_forward(struct ceph_mds_client *mdsc,
3322                            struct ceph_mds_session *session,
3323                            struct ceph_msg *msg)
3324 {
3325         struct ceph_mds_request *req;
3326         u64 tid = le64_to_cpu(msg->hdr.tid);
3327         u32 next_mds;
3328         u32 fwd_seq;
3329         int err = -EINVAL;
3330         void *p = msg->front.iov_base;
3331         void *end = p + msg->front.iov_len;
3332
3333         ceph_decode_need(&p, end, 2*sizeof(u32), bad);
3334         next_mds = ceph_decode_32(&p);
3335         fwd_seq = ceph_decode_32(&p);
3336
3337         mutex_lock(&mdsc->mutex);
3338         req = lookup_get_request(mdsc, tid);
3339         if (!req) {
3340                 dout("forward tid %llu to mds%d - req dne\n", tid, next_mds);
3341                 goto out;  /* dup reply? */
3342         }
3343
3344         if (test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags)) {
3345                 dout("forward tid %llu aborted, unregistering\n", tid);
3346                 __unregister_request(mdsc, req);
3347         } else if (fwd_seq <= req->r_num_fwd) {
3348                 dout("forward tid %llu to mds%d - old seq %d <= %d\n",
3349                      tid, next_mds, req->r_num_fwd, fwd_seq);
3350         } else {
3351                 /* resend. forward race not possible; mds would drop */
3352                 dout("forward tid %llu to mds%d (we resend)\n", tid, next_mds);
3353                 BUG_ON(req->r_err);
3354                 BUG_ON(test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags));
3355                 req->r_attempts = 0;
3356                 req->r_num_fwd = fwd_seq;
3357                 req->r_resend_mds = next_mds;
3358                 put_request_session(req);
3359                 __do_request(mdsc, req);
3360         }
3361         ceph_mdsc_put_request(req);
3362 out:
3363         mutex_unlock(&mdsc->mutex);
3364         return;
3365
3366 bad:
3367         pr_err("mdsc_handle_forward decode error err=%d\n", err);
3368 }
3369
3370 static int __decode_session_metadata(void **p, void *end,
3371                                      bool *blocklisted)
3372 {
3373         /* map<string,string> */
3374         u32 n;
3375         bool err_str;
3376         ceph_decode_32_safe(p, end, n, bad);
3377         while (n-- > 0) {
3378                 u32 len;
3379                 ceph_decode_32_safe(p, end, len, bad);
3380                 ceph_decode_need(p, end, len, bad);
3381                 err_str = !strncmp(*p, "error_string", len);
3382                 *p += len;
3383                 ceph_decode_32_safe(p, end, len, bad);
3384                 ceph_decode_need(p, end, len, bad);
3385                 /*
3386                  * Match "blocklisted (blacklisted)" from newer MDSes,
3387                  * or "blacklisted" from older MDSes.
3388                  */
3389                 if (err_str && strnstr(*p, "blacklisted", len))
3390                         *blocklisted = true;
3391                 *p += len;
3392         }
3393         return 0;
3394 bad:
3395         return -1;
3396 }
3397
3398 /*
3399  * handle a mds session control message
3400  */
3401 static void handle_session(struct ceph_mds_session *session,
3402                            struct ceph_msg *msg)
3403 {
3404         struct ceph_mds_client *mdsc = session->s_mdsc;
3405         int mds = session->s_mds;
3406         int msg_version = le16_to_cpu(msg->hdr.version);
3407         void *p = msg->front.iov_base;
3408         void *end = p + msg->front.iov_len;
3409         struct ceph_mds_session_head *h;
3410         u32 op;
3411         u64 seq, features = 0;
3412         int wake = 0;
3413         bool blocklisted = false;
3414
3415         /* decode */
3416         ceph_decode_need(&p, end, sizeof(*h), bad);
3417         h = p;
3418         p += sizeof(*h);
3419
3420         op = le32_to_cpu(h->op);
3421         seq = le64_to_cpu(h->seq);
3422
3423         if (msg_version >= 3) {
3424                 u32 len;
3425                 /* version >= 2, metadata */
3426                 if (__decode_session_metadata(&p, end, &blocklisted) < 0)
3427                         goto bad;
3428                 /* version >= 3, feature bits */
3429                 ceph_decode_32_safe(&p, end, len, bad);
3430                 if (len) {
3431                         ceph_decode_64_safe(&p, end, features, bad);
3432                         p += len - sizeof(features);
3433                 }
3434         }
3435
3436         mutex_lock(&mdsc->mutex);
3437         if (op == CEPH_SESSION_CLOSE) {
3438                 ceph_get_mds_session(session);
3439                 __unregister_session(mdsc, session);
3440         }
3441         /* FIXME: this ttl calculation is generous */
3442         session->s_ttl = jiffies + HZ*mdsc->mdsmap->m_session_autoclose;
3443         mutex_unlock(&mdsc->mutex);
3444
3445         mutex_lock(&session->s_mutex);
3446
3447         dout("handle_session mds%d %s %p state %s seq %llu\n",
3448              mds, ceph_session_op_name(op), session,
3449              ceph_session_state_name(session->s_state), seq);
3450
3451         if (session->s_state == CEPH_MDS_SESSION_HUNG) {
3452                 session->s_state = CEPH_MDS_SESSION_OPEN;
3453                 pr_info("mds%d came back\n", session->s_mds);
3454         }
3455
3456         switch (op) {
3457         case CEPH_SESSION_OPEN:
3458                 if (session->s_state == CEPH_MDS_SESSION_RECONNECTING)
3459                         pr_info("mds%d reconnect success\n", session->s_mds);
3460                 session->s_state = CEPH_MDS_SESSION_OPEN;
3461                 session->s_features = features;
3462                 renewed_caps(mdsc, session, 0);
3463                 if (test_bit(CEPHFS_FEATURE_METRIC_COLLECT, &session->s_features))
3464                         metric_schedule_delayed(&mdsc->metric);
3465                 wake = 1;
3466                 if (mdsc->stopping)
3467                         __close_session(mdsc, session);
3468                 break;
3469
3470         case CEPH_SESSION_RENEWCAPS:
3471                 if (session->s_renew_seq == seq)
3472                         renewed_caps(mdsc, session, 1);
3473                 break;
3474
3475         case CEPH_SESSION_CLOSE:
3476                 if (session->s_state == CEPH_MDS_SESSION_RECONNECTING)
3477                         pr_info("mds%d reconnect denied\n", session->s_mds);
3478                 session->s_state = CEPH_MDS_SESSION_CLOSED;
3479                 cleanup_session_requests(mdsc, session);
3480                 remove_session_caps(session);
3481                 wake = 2; /* for good measure */
3482                 wake_up_all(&mdsc->session_close_wq);
3483                 break;
3484
3485         case CEPH_SESSION_STALE:
3486                 pr_info("mds%d caps went stale, renewing\n",
3487                         session->s_mds);
3488                 spin_lock(&session->s_gen_ttl_lock);
3489                 session->s_cap_gen++;
3490                 session->s_cap_ttl = jiffies - 1;
3491                 spin_unlock(&session->s_gen_ttl_lock);
3492                 send_renew_caps(mdsc, session);
3493                 break;
3494
3495         case CEPH_SESSION_RECALL_STATE:
3496                 ceph_trim_caps(mdsc, session, le32_to_cpu(h->max_caps));
3497                 break;
3498
3499         case CEPH_SESSION_FLUSHMSG:
3500                 send_flushmsg_ack(mdsc, session, seq);
3501                 break;
3502
3503         case CEPH_SESSION_FORCE_RO:
3504                 dout("force_session_readonly %p\n", session);
3505                 spin_lock(&session->s_cap_lock);
3506                 session->s_readonly = true;
3507                 spin_unlock(&session->s_cap_lock);
3508                 wake_up_session_caps(session, FORCE_RO);
3509                 break;
3510
3511         case CEPH_SESSION_REJECT:
3512                 WARN_ON(session->s_state != CEPH_MDS_SESSION_OPENING);
3513                 pr_info("mds%d rejected session\n", session->s_mds);
3514                 session->s_state = CEPH_MDS_SESSION_REJECTED;
3515                 cleanup_session_requests(mdsc, session);
3516                 remove_session_caps(session);
3517                 if (blocklisted)
3518                         mdsc->fsc->blocklisted = true;
3519                 wake = 2; /* for good measure */
3520                 break;
3521
3522         default:
3523                 pr_err("mdsc_handle_session bad op %d mds%d\n", op, mds);
3524                 WARN_ON(1);
3525         }
3526
3527         mutex_unlock(&session->s_mutex);
3528         if (wake) {
3529                 mutex_lock(&mdsc->mutex);
3530                 __wake_requests(mdsc, &session->s_waiting);
3531                 if (wake == 2)
3532                         kick_requests(mdsc, mds);
3533                 mutex_unlock(&mdsc->mutex);
3534         }
3535         if (op == CEPH_SESSION_CLOSE)
3536                 ceph_put_mds_session(session);
3537         return;
3538
3539 bad:
3540         pr_err("mdsc_handle_session corrupt message mds%d len %d\n", mds,
3541                (int)msg->front.iov_len);
3542         ceph_msg_dump(msg);
3543         return;
3544 }
3545
3546 void ceph_mdsc_release_dir_caps(struct ceph_mds_request *req)
3547 {
3548         int dcaps;
3549
3550         dcaps = xchg(&req->r_dir_caps, 0);
3551         if (dcaps) {
3552                 dout("releasing r_dir_caps=%s\n", ceph_cap_string(dcaps));
3553                 ceph_put_cap_refs(ceph_inode(req->r_parent), dcaps);
3554         }
3555 }
3556
3557 void ceph_mdsc_release_dir_caps_no_check(struct ceph_mds_request *req)
3558 {
3559         int dcaps;
3560
3561         dcaps = xchg(&req->r_dir_caps, 0);
3562         if (dcaps) {
3563                 dout("releasing r_dir_caps=%s\n", ceph_cap_string(dcaps));
3564                 ceph_put_cap_refs_no_check_caps(ceph_inode(req->r_parent),
3565                                                 dcaps);
3566         }
3567 }
3568
3569 /*
3570  * called under session->mutex.
3571  */
3572 static void replay_unsafe_requests(struct ceph_mds_client *mdsc,
3573                                    struct ceph_mds_session *session)
3574 {
3575         struct ceph_mds_request *req, *nreq;
3576         struct rb_node *p;
3577
3578         dout("replay_unsafe_requests mds%d\n", session->s_mds);
3579
3580         mutex_lock(&mdsc->mutex);
3581         list_for_each_entry_safe(req, nreq, &session->s_unsafe, r_unsafe_item)
3582                 __send_request(session, req, true);
3583
3584         /*
3585          * also re-send old requests when MDS enters reconnect stage. So that MDS
3586          * can process completed request in clientreplay stage.
3587          */
3588         p = rb_first(&mdsc->request_tree);
3589         while (p) {
3590                 req = rb_entry(p, struct ceph_mds_request, r_node);
3591                 p = rb_next(p);
3592                 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags))
3593                         continue;
3594                 if (req->r_attempts == 0)
3595                         continue; /* only old requests */
3596                 if (!req->r_session)
3597                         continue;
3598                 if (req->r_session->s_mds != session->s_mds)
3599                         continue;
3600
3601                 ceph_mdsc_release_dir_caps_no_check(req);
3602
3603                 __send_request(session, req, true);
3604         }
3605         mutex_unlock(&mdsc->mutex);
3606 }
3607
3608 static int send_reconnect_partial(struct ceph_reconnect_state *recon_state)
3609 {
3610         struct ceph_msg *reply;
3611         struct ceph_pagelist *_pagelist;
3612         struct page *page;
3613         __le32 *addr;
3614         int err = -ENOMEM;
3615
3616         if (!recon_state->allow_multi)
3617                 return -ENOSPC;
3618
3619         /* can't handle message that contains both caps and realm */
3620         BUG_ON(!recon_state->nr_caps == !recon_state->nr_realms);
3621
3622         /* pre-allocate new pagelist */
3623         _pagelist = ceph_pagelist_alloc(GFP_NOFS);
3624         if (!_pagelist)
3625                 return -ENOMEM;
3626
3627         reply = ceph_msg_new2(CEPH_MSG_CLIENT_RECONNECT, 0, 1, GFP_NOFS, false);
3628         if (!reply)
3629                 goto fail_msg;
3630
3631         /* placeholder for nr_caps */
3632         err = ceph_pagelist_encode_32(_pagelist, 0);
3633         if (err < 0)
3634                 goto fail;
3635
3636         if (recon_state->nr_caps) {
3637                 /* currently encoding caps */
3638                 err = ceph_pagelist_encode_32(recon_state->pagelist, 0);
3639                 if (err)
3640                         goto fail;
3641         } else {
3642                 /* placeholder for nr_realms (currently encoding relams) */
3643                 err = ceph_pagelist_encode_32(_pagelist, 0);
3644                 if (err < 0)
3645                         goto fail;
3646         }
3647
3648         err = ceph_pagelist_encode_8(recon_state->pagelist, 1);
3649         if (err)
3650                 goto fail;
3651
3652         page = list_first_entry(&recon_state->pagelist->head, struct page, lru);
3653         addr = kmap_atomic(page);
3654         if (recon_state->nr_caps) {
3655                 /* currently encoding caps */
3656                 *addr = cpu_to_le32(recon_state->nr_caps);
3657         } else {
3658                 /* currently encoding relams */
3659                 *(addr + 1) = cpu_to_le32(recon_state->nr_realms);
3660         }
3661         kunmap_atomic(addr);
3662
3663         reply->hdr.version = cpu_to_le16(5);
3664         reply->hdr.compat_version = cpu_to_le16(4);
3665
3666         reply->hdr.data_len = cpu_to_le32(recon_state->pagelist->length);
3667         ceph_msg_data_add_pagelist(reply, recon_state->pagelist);
3668
3669         ceph_con_send(&recon_state->session->s_con, reply);
3670         ceph_pagelist_release(recon_state->pagelist);
3671
3672         recon_state->pagelist = _pagelist;
3673         recon_state->nr_caps = 0;
3674         recon_state->nr_realms = 0;
3675         recon_state->msg_version = 5;
3676         return 0;
3677 fail:
3678         ceph_msg_put(reply);
3679 fail_msg:
3680         ceph_pagelist_release(_pagelist);
3681         return err;
3682 }
3683
3684 static struct dentry* d_find_primary(struct inode *inode)
3685 {
3686         struct dentry *alias, *dn = NULL;
3687
3688         if (hlist_empty(&inode->i_dentry))
3689                 return NULL;
3690
3691         spin_lock(&inode->i_lock);
3692         if (hlist_empty(&inode->i_dentry))
3693                 goto out_unlock;
3694
3695         if (S_ISDIR(inode->i_mode)) {
3696                 alias = hlist_entry(inode->i_dentry.first, struct dentry, d_u.d_alias);
3697                 if (!IS_ROOT(alias))
3698                         dn = dget(alias);
3699                 goto out_unlock;
3700         }
3701
3702         hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) {
3703                 spin_lock(&alias->d_lock);
3704                 if (!d_unhashed(alias) &&
3705                     (ceph_dentry(alias)->flags & CEPH_DENTRY_PRIMARY_LINK)) {
3706                         dn = dget_dlock(alias);
3707                 }
3708                 spin_unlock(&alias->d_lock);
3709                 if (dn)
3710                         break;
3711         }
3712 out_unlock:
3713         spin_unlock(&inode->i_lock);
3714         return dn;
3715 }
3716
3717 /*
3718  * Encode information about a cap for a reconnect with the MDS.
3719  */
3720 static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap,
3721                           void *arg)
3722 {
3723         union {
3724                 struct ceph_mds_cap_reconnect v2;
3725                 struct ceph_mds_cap_reconnect_v1 v1;
3726         } rec;
3727         struct ceph_inode_info *ci = cap->ci;
3728         struct ceph_reconnect_state *recon_state = arg;
3729         struct ceph_pagelist *pagelist = recon_state->pagelist;
3730         struct dentry *dentry;
3731         char *path;
3732         int pathlen, err;
3733         u64 pathbase;
3734         u64 snap_follows;
3735
3736         dout(" adding %p ino %llx.%llx cap %p %lld %s\n",
3737              inode, ceph_vinop(inode), cap, cap->cap_id,
3738              ceph_cap_string(cap->issued));
3739
3740         dentry = d_find_primary(inode);
3741         if (dentry) {
3742                 /* set pathbase to parent dir when msg_version >= 2 */
3743                 path = ceph_mdsc_build_path(dentry, &pathlen, &pathbase,
3744                                             recon_state->msg_version >= 2);
3745                 dput(dentry);
3746                 if (IS_ERR(path)) {
3747                         err = PTR_ERR(path);
3748                         goto out_err;
3749                 }
3750         } else {
3751                 path = NULL;
3752                 pathlen = 0;
3753                 pathbase = 0;
3754         }
3755
3756         spin_lock(&ci->i_ceph_lock);
3757         cap->seq = 0;        /* reset cap seq */
3758         cap->issue_seq = 0;  /* and issue_seq */
3759         cap->mseq = 0;       /* and migrate_seq */
3760         cap->cap_gen = cap->session->s_cap_gen;
3761
3762         /* These are lost when the session goes away */
3763         if (S_ISDIR(inode->i_mode)) {
3764                 if (cap->issued & CEPH_CAP_DIR_CREATE) {
3765                         ceph_put_string(rcu_dereference_raw(ci->i_cached_layout.pool_ns));
3766                         memset(&ci->i_cached_layout, 0, sizeof(ci->i_cached_layout));
3767                 }
3768                 cap->issued &= ~CEPH_CAP_ANY_DIR_OPS;
3769         }
3770
3771         if (recon_state->msg_version >= 2) {
3772                 rec.v2.cap_id = cpu_to_le64(cap->cap_id);
3773                 rec.v2.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
3774                 rec.v2.issued = cpu_to_le32(cap->issued);
3775                 rec.v2.snaprealm = cpu_to_le64(ci->i_snap_realm->ino);
3776                 rec.v2.pathbase = cpu_to_le64(pathbase);
3777                 rec.v2.flock_len = (__force __le32)
3778                         ((ci->i_ceph_flags & CEPH_I_ERROR_FILELOCK) ? 0 : 1);
3779         } else {
3780                 rec.v1.cap_id = cpu_to_le64(cap->cap_id);
3781                 rec.v1.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
3782                 rec.v1.issued = cpu_to_le32(cap->issued);
3783                 rec.v1.size = cpu_to_le64(inode->i_size);
3784                 ceph_encode_timespec64(&rec.v1.mtime, &inode->i_mtime);
3785                 ceph_encode_timespec64(&rec.v1.atime, &inode->i_atime);
3786                 rec.v1.snaprealm = cpu_to_le64(ci->i_snap_realm->ino);
3787                 rec.v1.pathbase = cpu_to_le64(pathbase);
3788         }
3789
3790         if (list_empty(&ci->i_cap_snaps)) {
3791                 snap_follows = ci->i_head_snapc ? ci->i_head_snapc->seq : 0;
3792         } else {
3793                 struct ceph_cap_snap *capsnap =
3794                         list_first_entry(&ci->i_cap_snaps,
3795                                          struct ceph_cap_snap, ci_item);
3796                 snap_follows = capsnap->follows;
3797         }
3798         spin_unlock(&ci->i_ceph_lock);
3799
3800         if (recon_state->msg_version >= 2) {
3801                 int num_fcntl_locks, num_flock_locks;
3802                 struct ceph_filelock *flocks = NULL;
3803                 size_t struct_len, total_len = sizeof(u64);
3804                 u8 struct_v = 0;
3805
3806 encode_again:
3807                 if (rec.v2.flock_len) {
3808                         ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks);
3809                 } else {
3810                         num_fcntl_locks = 0;
3811                         num_flock_locks = 0;
3812                 }
3813                 if (num_fcntl_locks + num_flock_locks > 0) {
3814                         flocks = kmalloc_array(num_fcntl_locks + num_flock_locks,
3815                                                sizeof(struct ceph_filelock),
3816                                                GFP_NOFS);
3817                         if (!flocks) {
3818                                 err = -ENOMEM;
3819                                 goto out_err;
3820                         }
3821                         err = ceph_encode_locks_to_buffer(inode, flocks,
3822                                                           num_fcntl_locks,
3823                                                           num_flock_locks);
3824                         if (err) {
3825                                 kfree(flocks);
3826                                 flocks = NULL;
3827                                 if (err == -ENOSPC)
3828                                         goto encode_again;
3829                                 goto out_err;
3830                         }
3831                 } else {
3832                         kfree(flocks);
3833                         flocks = NULL;
3834                 }
3835
3836                 if (recon_state->msg_version >= 3) {
3837                         /* version, compat_version and struct_len */
3838                         total_len += 2 * sizeof(u8) + sizeof(u32);
3839                         struct_v = 2;
3840                 }
3841                 /*
3842                  * number of encoded locks is stable, so copy to pagelist
3843                  */
3844                 struct_len = 2 * sizeof(u32) +
3845                             (num_fcntl_locks + num_flock_locks) *
3846                             sizeof(struct ceph_filelock);
3847                 rec.v2.flock_len = cpu_to_le32(struct_len);
3848
3849                 struct_len += sizeof(u32) + pathlen + sizeof(rec.v2);
3850
3851                 if (struct_v >= 2)
3852                         struct_len += sizeof(u64); /* snap_follows */
3853
3854                 total_len += struct_len;
3855
3856                 if (pagelist->length + total_len > RECONNECT_MAX_SIZE) {
3857                         err = send_reconnect_partial(recon_state);
3858                         if (err)
3859                                 goto out_freeflocks;
3860                         pagelist = recon_state->pagelist;
3861                 }
3862
3863                 err = ceph_pagelist_reserve(pagelist, total_len);
3864                 if (err)
3865                         goto out_freeflocks;
3866
3867                 ceph_pagelist_encode_64(pagelist, ceph_ino(inode));
3868                 if (recon_state->msg_version >= 3) {
3869                         ceph_pagelist_encode_8(pagelist, struct_v);
3870                         ceph_pagelist_encode_8(pagelist, 1);
3871                         ceph_pagelist_encode_32(pagelist, struct_len);
3872                 }
3873                 ceph_pagelist_encode_string(pagelist, path, pathlen);
3874                 ceph_pagelist_append(pagelist, &rec, sizeof(rec.v2));
3875                 ceph_locks_to_pagelist(flocks, pagelist,
3876                                        num_fcntl_locks, num_flock_locks);
3877                 if (struct_v >= 2)
3878                         ceph_pagelist_encode_64(pagelist, snap_follows);
3879 out_freeflocks:
3880                 kfree(flocks);
3881         } else {
3882                 err = ceph_pagelist_reserve(pagelist,
3883                                             sizeof(u64) + sizeof(u32) +
3884                                             pathlen + sizeof(rec.v1));
3885                 if (err)
3886                         goto out_err;
3887
3888                 ceph_pagelist_encode_64(pagelist, ceph_ino(inode));
3889                 ceph_pagelist_encode_string(pagelist, path, pathlen);
3890                 ceph_pagelist_append(pagelist, &rec, sizeof(rec.v1));
3891         }
3892
3893 out_err:
3894         ceph_mdsc_free_path(path, pathlen);
3895         if (!err)
3896                 recon_state->nr_caps++;
3897         return err;
3898 }
3899
3900 static int encode_snap_realms(struct ceph_mds_client *mdsc,
3901                               struct ceph_reconnect_state *recon_state)
3902 {
3903         struct rb_node *p;
3904         struct ceph_pagelist *pagelist = recon_state->pagelist;
3905         int err = 0;
3906
3907         if (recon_state->msg_version >= 4) {
3908                 err = ceph_pagelist_encode_32(pagelist, mdsc->num_snap_realms);
3909                 if (err < 0)
3910                         goto fail;
3911         }
3912
3913         /*
3914          * snaprealms.  we provide mds with the ino, seq (version), and
3915          * parent for all of our realms.  If the mds has any newer info,
3916          * it will tell us.
3917          */
3918         for (p = rb_first(&mdsc->snap_realms); p; p = rb_next(p)) {
3919                 struct ceph_snap_realm *realm =
3920                        rb_entry(p, struct ceph_snap_realm, node);
3921                 struct ceph_mds_snaprealm_reconnect sr_rec;
3922
3923                 if (recon_state->msg_version >= 4) {
3924                         size_t need = sizeof(u8) * 2 + sizeof(u32) +
3925                                       sizeof(sr_rec);
3926
3927                         if (pagelist->length + need > RECONNECT_MAX_SIZE) {
3928                                 err = send_reconnect_partial(recon_state);
3929                                 if (err)
3930                                         goto fail;
3931                                 pagelist = recon_state->pagelist;
3932                         }
3933
3934                         err = ceph_pagelist_reserve(pagelist, need);
3935                         if (err)
3936                                 goto fail;
3937
3938                         ceph_pagelist_encode_8(pagelist, 1);
3939                         ceph_pagelist_encode_8(pagelist, 1);
3940                         ceph_pagelist_encode_32(pagelist, sizeof(sr_rec));
3941                 }
3942
3943                 dout(" adding snap realm %llx seq %lld parent %llx\n",
3944                      realm->ino, realm->seq, realm->parent_ino);
3945                 sr_rec.ino = cpu_to_le64(realm->ino);
3946                 sr_rec.seq = cpu_to_le64(realm->seq);
3947                 sr_rec.parent = cpu_to_le64(realm->parent_ino);
3948
3949                 err = ceph_pagelist_append(pagelist, &sr_rec, sizeof(sr_rec));
3950                 if (err)
3951                         goto fail;
3952
3953                 recon_state->nr_realms++;
3954         }
3955 fail:
3956         return err;
3957 }
3958
3959
3960 /*
3961  * If an MDS fails and recovers, clients need to reconnect in order to
3962  * reestablish shared state.  This includes all caps issued through
3963  * this session _and_ the snap_realm hierarchy.  Because it's not
3964  * clear which snap realms the mds cares about, we send everything we
3965  * know about.. that ensures we'll then get any new info the
3966  * recovering MDS might have.
3967  *
3968  * This is a relatively heavyweight operation, but it's rare.
3969  */
3970 static void send_mds_reconnect(struct ceph_mds_client *mdsc,
3971                                struct ceph_mds_session *session)
3972 {
3973         struct ceph_msg *reply;
3974         int mds = session->s_mds;
3975         int err = -ENOMEM;
3976         struct ceph_reconnect_state recon_state = {
3977                 .session = session,
3978         };
3979         LIST_HEAD(dispose);
3980
3981         pr_info("mds%d reconnect start\n", mds);
3982
3983         recon_state.pagelist = ceph_pagelist_alloc(GFP_NOFS);
3984         if (!recon_state.pagelist)
3985                 goto fail_nopagelist;
3986
3987         reply = ceph_msg_new2(CEPH_MSG_CLIENT_RECONNECT, 0, 1, GFP_NOFS, false);
3988         if (!reply)
3989                 goto fail_nomsg;
3990
3991         xa_destroy(&session->s_delegated_inos);
3992
3993         mutex_lock(&session->s_mutex);
3994         session->s_state = CEPH_MDS_SESSION_RECONNECTING;
3995         session->s_seq = 0;
3996
3997         dout("session %p state %s\n", session,
3998              ceph_session_state_name(session->s_state));
3999
4000         spin_lock(&session->s_gen_ttl_lock);
4001         session->s_cap_gen++;
4002         spin_unlock(&session->s_gen_ttl_lock);
4003
4004         spin_lock(&session->s_cap_lock);
4005         /* don't know if session is readonly */
4006         session->s_readonly = 0;
4007         /*
4008          * notify __ceph_remove_cap() that we are composing cap reconnect.
4009          * If a cap get released before being added to the cap reconnect,
4010          * __ceph_remove_cap() should skip queuing cap release.
4011          */
4012         session->s_cap_reconnect = 1;
4013         /* drop old cap expires; we're about to reestablish that state */
4014         detach_cap_releases(session, &dispose);
4015         spin_unlock(&session->s_cap_lock);
4016         dispose_cap_releases(mdsc, &dispose);
4017
4018         /* trim unused caps to reduce MDS's cache rejoin time */
4019         if (mdsc->fsc->sb->s_root)
4020                 shrink_dcache_parent(mdsc->fsc->sb->s_root);
4021
4022         ceph_con_close(&session->s_con);
4023         ceph_con_open(&session->s_con,
4024                       CEPH_ENTITY_TYPE_MDS, mds,
4025                       ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
4026
4027         /* replay unsafe requests */
4028         replay_unsafe_requests(mdsc, session);
4029
4030         ceph_early_kick_flushing_caps(mdsc, session);
4031
4032         down_read(&mdsc->snap_rwsem);
4033
4034         /* placeholder for nr_caps */
4035         err = ceph_pagelist_encode_32(recon_state.pagelist, 0);
4036         if (err)
4037                 goto fail;
4038
4039         if (test_bit(CEPHFS_FEATURE_MULTI_RECONNECT, &session->s_features)) {
4040                 recon_state.msg_version = 3;
4041                 recon_state.allow_multi = true;
4042         } else if (session->s_con.peer_features & CEPH_FEATURE_MDSENC) {
4043                 recon_state.msg_version = 3;
4044         } else {
4045                 recon_state.msg_version = 2;
4046         }
4047         /* trsaverse this session's caps */
4048         err = ceph_iterate_session_caps(session, reconnect_caps_cb, &recon_state);
4049
4050         spin_lock(&session->s_cap_lock);
4051         session->s_cap_reconnect = 0;
4052         spin_unlock(&session->s_cap_lock);
4053
4054         if (err < 0)
4055                 goto fail;
4056
4057         /* check if all realms can be encoded into current message */
4058         if (mdsc->num_snap_realms) {
4059                 size_t total_len =
4060                         recon_state.pagelist->length +
4061                         mdsc->num_snap_realms *
4062                         sizeof(struct ceph_mds_snaprealm_reconnect);
4063                 if (recon_state.msg_version >= 4) {
4064                         /* number of realms */
4065                         total_len += sizeof(u32);
4066                         /* version, compat_version and struct_len */
4067                         total_len += mdsc->num_snap_realms *
4068                                      (2 * sizeof(u8) + sizeof(u32));
4069                 }
4070                 if (total_len > RECONNECT_MAX_SIZE) {
4071                         if (!recon_state.allow_multi) {
4072                                 err = -ENOSPC;
4073                                 goto fail;
4074                         }
4075                         if (recon_state.nr_caps) {
4076                                 err = send_reconnect_partial(&recon_state);
4077                                 if (err)
4078                                         goto fail;
4079                         }
4080                         recon_state.msg_version = 5;
4081                 }
4082         }
4083
4084         err = encode_snap_realms(mdsc, &recon_state);
4085         if (err < 0)
4086                 goto fail;
4087
4088         if (recon_state.msg_version >= 5) {
4089                 err = ceph_pagelist_encode_8(recon_state.pagelist, 0);
4090                 if (err < 0)
4091                         goto fail;
4092         }
4093
4094         if (recon_state.nr_caps || recon_state.nr_realms) {
4095                 struct page *page =
4096                         list_first_entry(&recon_state.pagelist->head,
4097                                         struct page, lru);
4098                 __le32 *addr = kmap_atomic(page);
4099                 if (recon_state.nr_caps) {
4100                         WARN_ON(recon_state.nr_realms != mdsc->num_snap_realms);
4101                         *addr = cpu_to_le32(recon_state.nr_caps);
4102                 } else if (recon_state.msg_version >= 4) {
4103                         *(addr + 1) = cpu_to_le32(recon_state.nr_realms);
4104                 }
4105                 kunmap_atomic(addr);
4106         }
4107
4108         reply->hdr.version = cpu_to_le16(recon_state.msg_version);
4109         if (recon_state.msg_version >= 4)
4110                 reply->hdr.compat_version = cpu_to_le16(4);
4111
4112         reply->hdr.data_len = cpu_to_le32(recon_state.pagelist->length);
4113         ceph_msg_data_add_pagelist(reply, recon_state.pagelist);
4114
4115         ceph_con_send(&session->s_con, reply);
4116
4117         mutex_unlock(&session->s_mutex);
4118
4119         mutex_lock(&mdsc->mutex);
4120         __wake_requests(mdsc, &session->s_waiting);
4121         mutex_unlock(&mdsc->mutex);
4122
4123         up_read(&mdsc->snap_rwsem);
4124         ceph_pagelist_release(recon_state.pagelist);
4125         return;
4126
4127 fail:
4128         ceph_msg_put(reply);
4129         up_read(&mdsc->snap_rwsem);
4130         mutex_unlock(&session->s_mutex);
4131 fail_nomsg:
4132         ceph_pagelist_release(recon_state.pagelist);
4133 fail_nopagelist:
4134         pr_err("error %d preparing reconnect for mds%d\n", err, mds);
4135         return;
4136 }
4137
4138
4139 /*
4140  * compare old and new mdsmaps, kicking requests
4141  * and closing out old connections as necessary
4142  *
4143  * called under mdsc->mutex.
4144  */
4145 static void check_new_map(struct ceph_mds_client *mdsc,
4146                           struct ceph_mdsmap *newmap,
4147                           struct ceph_mdsmap *oldmap)
4148 {
4149         int i;
4150         int oldstate, newstate;
4151         struct ceph_mds_session *s;
4152
4153         dout("check_new_map new %u old %u\n",
4154              newmap->m_epoch, oldmap->m_epoch);
4155
4156         for (i = 0; i < oldmap->possible_max_rank && i < mdsc->max_sessions; i++) {
4157                 if (!mdsc->sessions[i])
4158                         continue;
4159                 s = mdsc->sessions[i];
4160                 oldstate = ceph_mdsmap_get_state(oldmap, i);
4161                 newstate = ceph_mdsmap_get_state(newmap, i);
4162
4163                 dout("check_new_map mds%d state %s%s -> %s%s (session %s)\n",
4164                      i, ceph_mds_state_name(oldstate),
4165                      ceph_mdsmap_is_laggy(oldmap, i) ? " (laggy)" : "",
4166                      ceph_mds_state_name(newstate),
4167                      ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "",
4168                      ceph_session_state_name(s->s_state));
4169
4170                 if (i >= newmap->possible_max_rank) {
4171                         /* force close session for stopped mds */
4172                         ceph_get_mds_session(s);
4173                         __unregister_session(mdsc, s);
4174                         __wake_requests(mdsc, &s->s_waiting);
4175                         mutex_unlock(&mdsc->mutex);
4176
4177                         mutex_lock(&s->s_mutex);
4178                         cleanup_session_requests(mdsc, s);
4179                         remove_session_caps(s);
4180                         mutex_unlock(&s->s_mutex);
4181
4182                         ceph_put_mds_session(s);
4183
4184                         mutex_lock(&mdsc->mutex);
4185                         kick_requests(mdsc, i);
4186                         continue;
4187                 }
4188
4189                 if (memcmp(ceph_mdsmap_get_addr(oldmap, i),
4190                            ceph_mdsmap_get_addr(newmap, i),
4191                            sizeof(struct ceph_entity_addr))) {
4192                         /* just close it */
4193                         mutex_unlock(&mdsc->mutex);
4194                         mutex_lock(&s->s_mutex);
4195                         mutex_lock(&mdsc->mutex);
4196                         ceph_con_close(&s->s_con);
4197                         mutex_unlock(&s->s_mutex);
4198                         s->s_state = CEPH_MDS_SESSION_RESTARTING;
4199                 } else if (oldstate == newstate) {
4200                         continue;  /* nothing new with this mds */
4201                 }
4202
4203                 /*
4204                  * send reconnect?
4205                  */
4206                 if (s->s_state == CEPH_MDS_SESSION_RESTARTING &&
4207                     newstate >= CEPH_MDS_STATE_RECONNECT) {
4208                         mutex_unlock(&mdsc->mutex);
4209                         send_mds_reconnect(mdsc, s);
4210                         mutex_lock(&mdsc->mutex);
4211                 }
4212
4213                 /*
4214                  * kick request on any mds that has gone active.
4215                  */
4216                 if (oldstate < CEPH_MDS_STATE_ACTIVE &&
4217                     newstate >= CEPH_MDS_STATE_ACTIVE) {
4218                         if (oldstate != CEPH_MDS_STATE_CREATING &&
4219                             oldstate != CEPH_MDS_STATE_STARTING)
4220                                 pr_info("mds%d recovery completed\n", s->s_mds);
4221                         kick_requests(mdsc, i);
4222                         mutex_unlock(&mdsc->mutex);
4223                         mutex_lock(&s->s_mutex);
4224                         mutex_lock(&mdsc->mutex);
4225                         ceph_kick_flushing_caps(mdsc, s);
4226                         mutex_unlock(&s->s_mutex);
4227                         wake_up_session_caps(s, RECONNECT);
4228                 }
4229         }
4230
4231         for (i = 0; i < newmap->possible_max_rank && i < mdsc->max_sessions; i++) {
4232                 s = mdsc->sessions[i];
4233                 if (!s)
4234                         continue;
4235                 if (!ceph_mdsmap_is_laggy(newmap, i))
4236                         continue;
4237                 if (s->s_state == CEPH_MDS_SESSION_OPEN ||
4238                     s->s_state == CEPH_MDS_SESSION_HUNG ||
4239                     s->s_state == CEPH_MDS_SESSION_CLOSING) {
4240                         dout(" connecting to export targets of laggy mds%d\n",
4241                              i);
4242                         __open_export_target_sessions(mdsc, s);
4243                 }
4244         }
4245 }
4246
4247
4248
4249 /*
4250  * leases
4251  */
4252
4253 /*
4254  * caller must hold session s_mutex, dentry->d_lock
4255  */
4256 void __ceph_mdsc_drop_dentry_lease(struct dentry *dentry)
4257 {
4258         struct ceph_dentry_info *di = ceph_dentry(dentry);
4259
4260         ceph_put_mds_session(di->lease_session);
4261         di->lease_session = NULL;
4262 }
4263
4264 static void handle_lease(struct ceph_mds_client *mdsc,
4265                          struct ceph_mds_session *session,
4266                          struct ceph_msg *msg)
4267 {
4268         struct super_block *sb = mdsc->fsc->sb;
4269         struct inode *inode;
4270         struct dentry *parent, *dentry;
4271         struct ceph_dentry_info *di;
4272         int mds = session->s_mds;
4273         struct ceph_mds_lease *h = msg->front.iov_base;
4274         u32 seq;
4275         struct ceph_vino vino;
4276         struct qstr dname;
4277         int release = 0;
4278
4279         dout("handle_lease from mds%d\n", mds);
4280
4281         /* decode */
4282         if (msg->front.iov_len < sizeof(*h) + sizeof(u32))
4283                 goto bad;
4284         vino.ino = le64_to_cpu(h->ino);
4285         vino.snap = CEPH_NOSNAP;
4286         seq = le32_to_cpu(h->seq);
4287         dname.len = get_unaligned_le32(h + 1);
4288         if (msg->front.iov_len < sizeof(*h) + sizeof(u32) + dname.len)
4289                 goto bad;
4290         dname.name = (void *)(h + 1) + sizeof(u32);
4291
4292         /* lookup inode */
4293         inode = ceph_find_inode(sb, vino);
4294         dout("handle_lease %s, ino %llx %p %.*s\n",
4295              ceph_lease_op_name(h->action), vino.ino, inode,
4296              dname.len, dname.name);
4297
4298         mutex_lock(&session->s_mutex);
4299         inc_session_sequence(session);
4300
4301         if (!inode) {
4302                 dout("handle_lease no inode %llx\n", vino.ino);
4303                 goto release;
4304         }
4305
4306         /* dentry */
4307         parent = d_find_alias(inode);
4308         if (!parent) {
4309                 dout("no parent dentry on inode %p\n", inode);
4310                 WARN_ON(1);
4311                 goto release;  /* hrm... */
4312         }
4313         dname.hash = full_name_hash(parent, dname.name, dname.len);
4314         dentry = d_lookup(parent, &dname);
4315         dput(parent);
4316         if (!dentry)
4317                 goto release;
4318
4319         spin_lock(&dentry->d_lock);
4320         di = ceph_dentry(dentry);
4321         switch (h->action) {
4322         case CEPH_MDS_LEASE_REVOKE:
4323                 if (di->lease_session == session) {
4324                         if (ceph_seq_cmp(di->lease_seq, seq) > 0)
4325                                 h->seq = cpu_to_le32(di->lease_seq);
4326                         __ceph_mdsc_drop_dentry_lease(dentry);
4327                 }
4328                 release = 1;
4329                 break;
4330
4331         case CEPH_MDS_LEASE_RENEW:
4332                 if (di->lease_session == session &&
4333                     di->lease_gen == session->s_cap_gen &&
4334                     di->lease_renew_from &&
4335                     di->lease_renew_after == 0) {
4336                         unsigned long duration =
4337                                 msecs_to_jiffies(le32_to_cpu(h->duration_ms));
4338
4339                         di->lease_seq = seq;
4340                         di->time = di->lease_renew_from + duration;
4341                         di->lease_renew_after = di->lease_renew_from +
4342                                 (duration >> 1);
4343                         di->lease_renew_from = 0;
4344                 }
4345                 break;
4346         }
4347         spin_unlock(&dentry->d_lock);
4348         dput(dentry);
4349
4350         if (!release)
4351                 goto out;
4352
4353 release:
4354         /* let's just reuse the same message */
4355         h->action = CEPH_MDS_LEASE_REVOKE_ACK;
4356         ceph_msg_get(msg);
4357         ceph_con_send(&session->s_con, msg);
4358
4359 out:
4360         mutex_unlock(&session->s_mutex);
4361         /* avoid calling iput_final() in mds dispatch threads */
4362         ceph_async_iput(inode);
4363         return;
4364
4365 bad:
4366         pr_err("corrupt lease message\n");
4367         ceph_msg_dump(msg);
4368 }
4369
4370 void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
4371                               struct dentry *dentry, char action,
4372                               u32 seq)
4373 {
4374         struct ceph_msg *msg;
4375         struct ceph_mds_lease *lease;
4376         struct inode *dir;
4377         int len = sizeof(*lease) + sizeof(u32) + NAME_MAX;
4378
4379         dout("lease_send_msg identry %p %s to mds%d\n",
4380              dentry, ceph_lease_op_name(action), session->s_mds);
4381
4382         msg = ceph_msg_new(CEPH_MSG_CLIENT_LEASE, len, GFP_NOFS, false);
4383         if (!msg)
4384                 return;
4385         lease = msg->front.iov_base;
4386         lease->action = action;
4387         lease->seq = cpu_to_le32(seq);
4388
4389         spin_lock(&dentry->d_lock);
4390         dir = d_inode(dentry->d_parent);
4391         lease->ino = cpu_to_le64(ceph_ino(dir));
4392         lease->first = lease->last = cpu_to_le64(ceph_snap(dir));
4393
4394         put_unaligned_le32(dentry->d_name.len, lease + 1);
4395         memcpy((void *)(lease + 1) + 4,
4396                dentry->d_name.name, dentry->d_name.len);
4397         spin_unlock(&dentry->d_lock);
4398         /*
4399          * if this is a preemptive lease RELEASE, no need to
4400          * flush request stream, since the actual request will
4401          * soon follow.
4402          */
4403         msg->more_to_follow = (action == CEPH_MDS_LEASE_RELEASE);
4404
4405         ceph_con_send(&session->s_con, msg);
4406 }
4407
4408 /*
4409  * lock unlock sessions, to wait ongoing session activities
4410  */
4411 static void lock_unlock_sessions(struct ceph_mds_client *mdsc)
4412 {
4413         int i;
4414
4415         mutex_lock(&mdsc->mutex);
4416         for (i = 0; i < mdsc->max_sessions; i++) {
4417                 struct ceph_mds_session *s = __ceph_lookup_mds_session(mdsc, i);
4418                 if (!s)
4419                         continue;
4420                 mutex_unlock(&mdsc->mutex);
4421                 mutex_lock(&s->s_mutex);
4422                 mutex_unlock(&s->s_mutex);
4423                 ceph_put_mds_session(s);
4424                 mutex_lock(&mdsc->mutex);
4425         }
4426         mutex_unlock(&mdsc->mutex);
4427 }
4428
4429 static void maybe_recover_session(struct ceph_mds_client *mdsc)
4430 {
4431         struct ceph_fs_client *fsc = mdsc->fsc;
4432
4433         if (!ceph_test_mount_opt(fsc, CLEANRECOVER))
4434                 return;
4435
4436         if (READ_ONCE(fsc->mount_state) != CEPH_MOUNT_MOUNTED)
4437                 return;
4438
4439         if (!READ_ONCE(fsc->blocklisted))
4440                 return;
4441
4442         pr_info("auto reconnect after blocklisted\n");
4443         ceph_force_reconnect(fsc->sb);
4444 }
4445
4446 bool check_session_state(struct ceph_mds_session *s)
4447 {
4448         switch (s->s_state) {
4449         case CEPH_MDS_SESSION_OPEN:
4450                 if (s->s_ttl && time_after(jiffies, s->s_ttl)) {
4451                         s->s_state = CEPH_MDS_SESSION_HUNG;
4452                         pr_info("mds%d hung\n", s->s_mds);
4453                 }
4454                 break;
4455         case CEPH_MDS_SESSION_CLOSING:
4456                 /* Should never reach this when we're unmounting */
4457                 WARN_ON_ONCE(true);
4458                 fallthrough;
4459         case CEPH_MDS_SESSION_NEW:
4460         case CEPH_MDS_SESSION_RESTARTING:
4461         case CEPH_MDS_SESSION_CLOSED:
4462         case CEPH_MDS_SESSION_REJECTED:
4463                 return false;
4464         }
4465
4466         return true;
4467 }
4468
4469 /*
4470  * If the sequence is incremented while we're waiting on a REQUEST_CLOSE reply,
4471  * then we need to retransmit that request.
4472  */
4473 void inc_session_sequence(struct ceph_mds_session *s)
4474 {
4475         lockdep_assert_held(&s->s_mutex);
4476
4477         s->s_seq++;
4478
4479         if (s->s_state == CEPH_MDS_SESSION_CLOSING) {
4480                 int ret;
4481
4482                 dout("resending session close request for mds%d\n", s->s_mds);
4483                 ret = request_close_session(s);
4484                 if (ret < 0)
4485                         pr_err("unable to close session to mds%d: %d\n",
4486                                s->s_mds, ret);
4487         }
4488 }
4489
4490 /*
4491  * delayed work -- periodically trim expired leases, renew caps with mds
4492  */
4493 static void schedule_delayed(struct ceph_mds_client *mdsc)
4494 {
4495         int delay = 5;
4496         unsigned hz = round_jiffies_relative(HZ * delay);
4497         schedule_delayed_work(&mdsc->delayed_work, hz);
4498 }
4499
4500 static void delayed_work(struct work_struct *work)
4501 {
4502         int i;
4503         struct ceph_mds_client *mdsc =
4504                 container_of(work, struct ceph_mds_client, delayed_work.work);
4505         int renew_interval;
4506         int renew_caps;
4507
4508         dout("mdsc delayed_work\n");
4509
4510         if (mdsc->stopping)
4511                 return;
4512
4513         mutex_lock(&mdsc->mutex);
4514         renew_interval = mdsc->mdsmap->m_session_timeout >> 2;
4515         renew_caps = time_after_eq(jiffies, HZ*renew_interval +
4516                                    mdsc->last_renew_caps);
4517         if (renew_caps)
4518                 mdsc->last_renew_caps = jiffies;
4519
4520         for (i = 0; i < mdsc->max_sessions; i++) {
4521                 struct ceph_mds_session *s = __ceph_lookup_mds_session(mdsc, i);
4522                 if (!s)
4523                         continue;
4524
4525                 if (!check_session_state(s)) {
4526                         ceph_put_mds_session(s);
4527                         continue;
4528                 }
4529                 mutex_unlock(&mdsc->mutex);
4530
4531                 mutex_lock(&s->s_mutex);
4532                 if (renew_caps)
4533                         send_renew_caps(mdsc, s);
4534                 else
4535                         ceph_con_keepalive(&s->s_con);
4536                 if (s->s_state == CEPH_MDS_SESSION_OPEN ||
4537                     s->s_state == CEPH_MDS_SESSION_HUNG)
4538                         ceph_send_cap_releases(mdsc, s);
4539                 mutex_unlock(&s->s_mutex);
4540                 ceph_put_mds_session(s);
4541
4542                 mutex_lock(&mdsc->mutex);
4543         }
4544         mutex_unlock(&mdsc->mutex);
4545
4546         ceph_check_delayed_caps(mdsc);
4547
4548         ceph_queue_cap_reclaim_work(mdsc);
4549
4550         ceph_trim_snapid_map(mdsc);
4551
4552         maybe_recover_session(mdsc);
4553
4554         schedule_delayed(mdsc);
4555 }
4556
4557 int ceph_mdsc_init(struct ceph_fs_client *fsc)
4558
4559 {
4560         struct ceph_mds_client *mdsc;
4561         int err;
4562
4563         mdsc = kzalloc(sizeof(struct ceph_mds_client), GFP_NOFS);
4564         if (!mdsc)
4565                 return -ENOMEM;
4566         mdsc->fsc = fsc;
4567         mutex_init(&mdsc->mutex);
4568         mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
4569         if (!mdsc->mdsmap) {
4570                 err = -ENOMEM;
4571                 goto err_mdsc;
4572         }
4573
4574         init_completion(&mdsc->safe_umount_waiters);
4575         init_waitqueue_head(&mdsc->session_close_wq);
4576         INIT_LIST_HEAD(&mdsc->waiting_for_map);
4577         mdsc->sessions = NULL;
4578         atomic_set(&mdsc->num_sessions, 0);
4579         mdsc->max_sessions = 0;
4580         mdsc->stopping = 0;
4581         atomic64_set(&mdsc->quotarealms_count, 0);
4582         mdsc->quotarealms_inodes = RB_ROOT;
4583         mutex_init(&mdsc->quotarealms_inodes_mutex);
4584         mdsc->last_snap_seq = 0;
4585         init_rwsem(&mdsc->snap_rwsem);
4586         mdsc->snap_realms = RB_ROOT;
4587         INIT_LIST_HEAD(&mdsc->snap_empty);
4588         mdsc->num_snap_realms = 0;
4589         spin_lock_init(&mdsc->snap_empty_lock);
4590         mdsc->last_tid = 0;
4591         mdsc->oldest_tid = 0;
4592         mdsc->request_tree = RB_ROOT;
4593         INIT_DELAYED_WORK(&mdsc->delayed_work, delayed_work);
4594         mdsc->last_renew_caps = jiffies;
4595         INIT_LIST_HEAD(&mdsc->cap_delay_list);
4596         INIT_LIST_HEAD(&mdsc->cap_wait_list);
4597         spin_lock_init(&mdsc->cap_delay_lock);
4598         INIT_LIST_HEAD(&mdsc->snap_flush_list);
4599         spin_lock_init(&mdsc->snap_flush_lock);
4600         mdsc->last_cap_flush_tid = 1;
4601         INIT_LIST_HEAD(&mdsc->cap_flush_list);
4602         INIT_LIST_HEAD(&mdsc->cap_dirty_migrating);
4603         mdsc->num_cap_flushing = 0;
4604         spin_lock_init(&mdsc->cap_dirty_lock);
4605         init_waitqueue_head(&mdsc->cap_flushing_wq);
4606         INIT_WORK(&mdsc->cap_reclaim_work, ceph_cap_reclaim_work);
4607         atomic_set(&mdsc->cap_reclaim_pending, 0);
4608         err = ceph_metric_init(&mdsc->metric);
4609         if (err)
4610                 goto err_mdsmap;
4611
4612         spin_lock_init(&mdsc->dentry_list_lock);
4613         INIT_LIST_HEAD(&mdsc->dentry_leases);
4614         INIT_LIST_HEAD(&mdsc->dentry_dir_leases);
4615
4616         ceph_caps_init(mdsc);
4617         ceph_adjust_caps_max_min(mdsc, fsc->mount_options);
4618
4619         spin_lock_init(&mdsc->snapid_map_lock);
4620         mdsc->snapid_map_tree = RB_ROOT;
4621         INIT_LIST_HEAD(&mdsc->snapid_map_lru);
4622
4623         init_rwsem(&mdsc->pool_perm_rwsem);
4624         mdsc->pool_perm_tree = RB_ROOT;
4625
4626         strscpy(mdsc->nodename, utsname()->nodename,
4627                 sizeof(mdsc->nodename));
4628
4629         fsc->mdsc = mdsc;
4630         return 0;
4631
4632 err_mdsmap:
4633         kfree(mdsc->mdsmap);
4634 err_mdsc:
4635         kfree(mdsc);
4636         return err;
4637 }
4638
4639 /*
4640  * Wait for safe replies on open mds requests.  If we time out, drop
4641  * all requests from the tree to avoid dangling dentry refs.
4642  */
4643 static void wait_requests(struct ceph_mds_client *mdsc)
4644 {
4645         struct ceph_options *opts = mdsc->fsc->client->options;
4646         struct ceph_mds_request *req;
4647
4648         mutex_lock(&mdsc->mutex);
4649         if (__get_oldest_req(mdsc)) {
4650                 mutex_unlock(&mdsc->mutex);
4651
4652                 dout("wait_requests waiting for requests\n");
4653                 wait_for_completion_timeout(&mdsc->safe_umount_waiters,
4654                                     ceph_timeout_jiffies(opts->mount_timeout));
4655
4656                 /* tear down remaining requests */
4657                 mutex_lock(&mdsc->mutex);
4658                 while ((req = __get_oldest_req(mdsc))) {
4659                         dout("wait_requests timed out on tid %llu\n",
4660                              req->r_tid);
4661                         list_del_init(&req->r_wait);
4662                         __unregister_request(mdsc, req);
4663                 }
4664         }
4665         mutex_unlock(&mdsc->mutex);
4666         dout("wait_requests done\n");
4667 }
4668
4669 /*
4670  * called before mount is ro, and before dentries are torn down.
4671  * (hmm, does this still race with new lookups?)
4672  */
4673 void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
4674 {
4675         dout("pre_umount\n");
4676         mdsc->stopping = 1;
4677
4678         lock_unlock_sessions(mdsc);
4679         ceph_flush_dirty_caps(mdsc);
4680         wait_requests(mdsc);
4681
4682         /*
4683          * wait for reply handlers to drop their request refs and
4684          * their inode/dcache refs
4685          */
4686         ceph_msgr_flush();
4687
4688         ceph_cleanup_quotarealms_inodes(mdsc);
4689 }
4690
4691 /*
4692  * wait for all write mds requests to flush.
4693  */
4694 static void wait_unsafe_requests(struct ceph_mds_client *mdsc, u64 want_tid)
4695 {
4696         struct ceph_mds_request *req = NULL, *nextreq;
4697         struct rb_node *n;
4698
4699         mutex_lock(&mdsc->mutex);
4700         dout("wait_unsafe_requests want %lld\n", want_tid);
4701 restart:
4702         req = __get_oldest_req(mdsc);
4703         while (req && req->r_tid <= want_tid) {
4704                 /* find next request */
4705                 n = rb_next(&req->r_node);
4706                 if (n)
4707                         nextreq = rb_entry(n, struct ceph_mds_request, r_node);
4708                 else
4709                         nextreq = NULL;
4710                 if (req->r_op != CEPH_MDS_OP_SETFILELOCK &&
4711                     (req->r_op & CEPH_MDS_OP_WRITE)) {
4712                         /* write op */
4713                         ceph_mdsc_get_request(req);
4714                         if (nextreq)
4715                                 ceph_mdsc_get_request(nextreq);
4716                         mutex_unlock(&mdsc->mutex);
4717                         dout("wait_unsafe_requests  wait on %llu (want %llu)\n",
4718                              req->r_tid, want_tid);
4719                         wait_for_completion(&req->r_safe_completion);
4720                         mutex_lock(&mdsc->mutex);
4721                         ceph_mdsc_put_request(req);
4722                         if (!nextreq)
4723                                 break;  /* next dne before, so we're done! */
4724                         if (RB_EMPTY_NODE(&nextreq->r_node)) {
4725                                 /* next request was removed from tree */
4726                                 ceph_mdsc_put_request(nextreq);
4727                                 goto restart;
4728                         }
4729                         ceph_mdsc_put_request(nextreq);  /* won't go away */
4730                 }
4731                 req = nextreq;
4732         }
4733         mutex_unlock(&mdsc->mutex);
4734         dout("wait_unsafe_requests done\n");
4735 }
4736
4737 void ceph_mdsc_sync(struct ceph_mds_client *mdsc)
4738 {
4739         u64 want_tid, want_flush;
4740
4741         if (READ_ONCE(mdsc->fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN)
4742                 return;
4743
4744         dout("sync\n");
4745         mutex_lock(&mdsc->mutex);
4746         want_tid = mdsc->last_tid;
4747         mutex_unlock(&mdsc->mutex);
4748
4749         ceph_flush_dirty_caps(mdsc);
4750         spin_lock(&mdsc->cap_dirty_lock);
4751         want_flush = mdsc->last_cap_flush_tid;
4752         if (!list_empty(&mdsc->cap_flush_list)) {
4753                 struct ceph_cap_flush *cf =
4754                         list_last_entry(&mdsc->cap_flush_list,
4755                                         struct ceph_cap_flush, g_list);
4756                 cf->wake = true;
4757         }
4758         spin_unlock(&mdsc->cap_dirty_lock);
4759
4760         dout("sync want tid %lld flush_seq %lld\n",
4761              want_tid, want_flush);
4762
4763         wait_unsafe_requests(mdsc, want_tid);
4764         wait_caps_flush(mdsc, want_flush);
4765 }
4766
4767 /*
4768  * true if all sessions are closed, or we force unmount
4769  */
4770 static bool done_closing_sessions(struct ceph_mds_client *mdsc, int skipped)
4771 {
4772         if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_SHUTDOWN)
4773                 return true;
4774         return atomic_read(&mdsc->num_sessions) <= skipped;
4775 }
4776
4777 /*
4778  * called after sb is ro.
4779  */
4780 void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
4781 {
4782         struct ceph_options *opts = mdsc->fsc->client->options;
4783         struct ceph_mds_session *session;
4784         int i;
4785         int skipped = 0;
4786
4787         dout("close_sessions\n");
4788
4789         /* close sessions */
4790         mutex_lock(&mdsc->mutex);
4791         for (i = 0; i < mdsc->max_sessions; i++) {
4792                 session = __ceph_lookup_mds_session(mdsc, i);
4793                 if (!session)
4794                         continue;
4795                 mutex_unlock(&mdsc->mutex);
4796                 mutex_lock(&session->s_mutex);
4797                 if (__close_session(mdsc, session) <= 0)
4798                         skipped++;
4799                 mutex_unlock(&session->s_mutex);
4800                 ceph_put_mds_session(session);
4801                 mutex_lock(&mdsc->mutex);
4802         }
4803         mutex_unlock(&mdsc->mutex);
4804
4805         dout("waiting for sessions to close\n");
4806         wait_event_timeout(mdsc->session_close_wq,
4807                            done_closing_sessions(mdsc, skipped),
4808                            ceph_timeout_jiffies(opts->mount_timeout));
4809
4810         /* tear down remaining sessions */
4811         mutex_lock(&mdsc->mutex);
4812         for (i = 0; i < mdsc->max_sessions; i++) {
4813                 if (mdsc->sessions[i]) {
4814                         session = ceph_get_mds_session(mdsc->sessions[i]);
4815                         __unregister_session(mdsc, session);
4816                         mutex_unlock(&mdsc->mutex);
4817                         mutex_lock(&session->s_mutex);
4818                         remove_session_caps(session);
4819                         mutex_unlock(&session->s_mutex);
4820                         ceph_put_mds_session(session);
4821                         mutex_lock(&mdsc->mutex);
4822                 }
4823         }
4824         WARN_ON(!list_empty(&mdsc->cap_delay_list));
4825         mutex_unlock(&mdsc->mutex);
4826
4827         ceph_cleanup_snapid_map(mdsc);
4828         ceph_cleanup_empty_realms(mdsc);
4829
4830         cancel_work_sync(&mdsc->cap_reclaim_work);
4831         cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */
4832
4833         dout("stopped\n");
4834 }
4835
4836 void ceph_mdsc_force_umount(struct ceph_mds_client *mdsc)
4837 {
4838         struct ceph_mds_session *session;
4839         int mds;
4840
4841         dout("force umount\n");
4842
4843         mutex_lock(&mdsc->mutex);
4844         for (mds = 0; mds < mdsc->max_sessions; mds++) {
4845                 session = __ceph_lookup_mds_session(mdsc, mds);
4846                 if (!session)
4847                         continue;
4848
4849                 if (session->s_state == CEPH_MDS_SESSION_REJECTED)
4850                         __unregister_session(mdsc, session);
4851                 __wake_requests(mdsc, &session->s_waiting);
4852                 mutex_unlock(&mdsc->mutex);
4853
4854                 mutex_lock(&session->s_mutex);
4855                 __close_session(mdsc, session);
4856                 if (session->s_state == CEPH_MDS_SESSION_CLOSING) {
4857                         cleanup_session_requests(mdsc, session);
4858                         remove_session_caps(session);
4859                 }
4860                 mutex_unlock(&session->s_mutex);
4861                 ceph_put_mds_session(session);
4862
4863                 mutex_lock(&mdsc->mutex);
4864                 kick_requests(mdsc, mds);
4865         }
4866         __wake_requests(mdsc, &mdsc->waiting_for_map);
4867         mutex_unlock(&mdsc->mutex);
4868 }
4869
4870 static void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
4871 {
4872         dout("stop\n");
4873         /*
4874          * Make sure the delayed work stopped before releasing
4875          * the resources.
4876          *
4877          * Because the cancel_delayed_work_sync() will only
4878          * guarantee that the work finishes executing. But the
4879          * delayed work will re-arm itself again after that.
4880          */
4881         flush_delayed_work(&mdsc->delayed_work);
4882
4883         if (mdsc->mdsmap)
4884                 ceph_mdsmap_destroy(mdsc->mdsmap);
4885         kfree(mdsc->sessions);
4886         ceph_caps_finalize(mdsc);
4887         ceph_pool_perm_destroy(mdsc);
4888 }
4889
4890 void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
4891 {
4892         struct ceph_mds_client *mdsc = fsc->mdsc;
4893         dout("mdsc_destroy %p\n", mdsc);
4894
4895         if (!mdsc)
4896                 return;
4897
4898         /* flush out any connection work with references to us */
4899         ceph_msgr_flush();
4900
4901         ceph_mdsc_stop(mdsc);
4902
4903         ceph_metric_destroy(&mdsc->metric);
4904
4905         flush_delayed_work(&mdsc->metric.delayed_work);
4906         fsc->mdsc = NULL;
4907         kfree(mdsc);
4908         dout("mdsc_destroy %p done\n", mdsc);
4909 }
4910
4911 void ceph_mdsc_handle_fsmap(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
4912 {
4913         struct ceph_fs_client *fsc = mdsc->fsc;
4914         const char *mds_namespace = fsc->mount_options->mds_namespace;
4915         void *p = msg->front.iov_base;
4916         void *end = p + msg->front.iov_len;
4917         u32 epoch;
4918         u32 num_fs;
4919         u32 mount_fscid = (u32)-1;
4920         int err = -EINVAL;
4921
4922         ceph_decode_need(&p, end, sizeof(u32), bad);
4923         epoch = ceph_decode_32(&p);
4924
4925         dout("handle_fsmap epoch %u\n", epoch);
4926
4927         /* struct_v, struct_cv, map_len, epoch, legacy_client_fscid */
4928         ceph_decode_skip_n(&p, end, 2 + sizeof(u32) * 3, bad);
4929
4930         ceph_decode_32_safe(&p, end, num_fs, bad);
4931         while (num_fs-- > 0) {
4932                 void *info_p, *info_end;
4933                 u32 info_len;
4934                 u32 fscid, namelen;
4935
4936                 ceph_decode_need(&p, end, 2 + sizeof(u32), bad);
4937                 p += 2;         // info_v, info_cv
4938                 info_len = ceph_decode_32(&p);
4939                 ceph_decode_need(&p, end, info_len, bad);
4940                 info_p = p;
4941                 info_end = p + info_len;
4942                 p = info_end;
4943
4944                 ceph_decode_need(&info_p, info_end, sizeof(u32) * 2, bad);
4945                 fscid = ceph_decode_32(&info_p);
4946                 namelen = ceph_decode_32(&info_p);
4947                 ceph_decode_need(&info_p, info_end, namelen, bad);
4948
4949                 if (mds_namespace &&
4950                     strlen(mds_namespace) == namelen &&
4951                     !strncmp(mds_namespace, (char *)info_p, namelen)) {
4952                         mount_fscid = fscid;
4953                         break;
4954                 }
4955         }
4956
4957         ceph_monc_got_map(&fsc->client->monc, CEPH_SUB_FSMAP, epoch);
4958         if (mount_fscid != (u32)-1) {
4959                 fsc->client->monc.fs_cluster_id = mount_fscid;
4960                 ceph_monc_want_map(&fsc->client->monc, CEPH_SUB_MDSMAP,
4961                                    0, true);
4962                 ceph_monc_renew_subs(&fsc->client->monc);
4963         } else {
4964                 err = -ENOENT;
4965                 goto err_out;
4966         }
4967         return;
4968
4969 bad:
4970         pr_err("error decoding fsmap\n");
4971 err_out:
4972         mutex_lock(&mdsc->mutex);
4973         mdsc->mdsmap_err = err;
4974         __wake_requests(mdsc, &mdsc->waiting_for_map);
4975         mutex_unlock(&mdsc->mutex);
4976 }
4977
4978 /*
4979  * handle mds map update.
4980  */
4981 void ceph_mdsc_handle_mdsmap(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
4982 {
4983         u32 epoch;
4984         u32 maplen;
4985         void *p = msg->front.iov_base;
4986         void *end = p + msg->front.iov_len;
4987         struct ceph_mdsmap *newmap, *oldmap;
4988         struct ceph_fsid fsid;
4989         int err = -EINVAL;
4990
4991         ceph_decode_need(&p, end, sizeof(fsid)+2*sizeof(u32), bad);
4992         ceph_decode_copy(&p, &fsid, sizeof(fsid));
4993         if (ceph_check_fsid(mdsc->fsc->client, &fsid) < 0)
4994                 return;
4995         epoch = ceph_decode_32(&p);
4996         maplen = ceph_decode_32(&p);
4997         dout("handle_map epoch %u len %d\n", epoch, (int)maplen);
4998
4999         /* do we need it? */
5000         mutex_lock(&mdsc->mutex);
5001         if (mdsc->mdsmap && epoch <= mdsc->mdsmap->m_epoch) {
5002                 dout("handle_map epoch %u <= our %u\n",
5003                      epoch, mdsc->mdsmap->m_epoch);
5004                 mutex_unlock(&mdsc->mutex);
5005                 return;
5006         }
5007
5008         newmap = ceph_mdsmap_decode(&p, end, ceph_msgr2(mdsc->fsc->client));
5009         if (IS_ERR(newmap)) {
5010                 err = PTR_ERR(newmap);
5011                 goto bad_unlock;
5012         }
5013
5014         /* swap into place */
5015         if (mdsc->mdsmap) {
5016                 oldmap = mdsc->mdsmap;
5017                 mdsc->mdsmap = newmap;
5018                 check_new_map(mdsc, newmap, oldmap);
5019                 ceph_mdsmap_destroy(oldmap);
5020         } else {
5021                 mdsc->mdsmap = newmap;  /* first mds map */
5022         }
5023         mdsc->fsc->max_file_size = min((loff_t)mdsc->mdsmap->m_max_file_size,
5024                                         MAX_LFS_FILESIZE);
5025
5026         __wake_requests(mdsc, &mdsc->waiting_for_map);
5027         ceph_monc_got_map(&mdsc->fsc->client->monc, CEPH_SUB_MDSMAP,
5028                           mdsc->mdsmap->m_epoch);
5029
5030         mutex_unlock(&mdsc->mutex);
5031         schedule_delayed(mdsc);
5032         return;
5033
5034 bad_unlock:
5035         mutex_unlock(&mdsc->mutex);
5036 bad:
5037         pr_err("error decoding mdsmap %d\n", err);
5038         return;
5039 }
5040
5041 static struct ceph_connection *mds_get_con(struct ceph_connection *con)
5042 {
5043         struct ceph_mds_session *s = con->private;
5044
5045         if (ceph_get_mds_session(s))
5046                 return con;
5047         return NULL;
5048 }
5049
5050 static void mds_put_con(struct ceph_connection *con)
5051 {
5052         struct ceph_mds_session *s = con->private;
5053
5054         ceph_put_mds_session(s);
5055 }
5056
5057 /*
5058  * if the client is unresponsive for long enough, the mds will kill
5059  * the session entirely.
5060  */
5061 static void mds_peer_reset(struct ceph_connection *con)
5062 {
5063         struct ceph_mds_session *s = con->private;
5064         struct ceph_mds_client *mdsc = s->s_mdsc;
5065
5066         pr_warn("mds%d closed our session\n", s->s_mds);
5067         send_mds_reconnect(mdsc, s);
5068 }
5069
5070 static void mds_dispatch(struct ceph_connection *con, struct ceph_msg *msg)
5071 {
5072         struct ceph_mds_session *s = con->private;
5073         struct ceph_mds_client *mdsc = s->s_mdsc;
5074         int type = le16_to_cpu(msg->hdr.type);
5075
5076         mutex_lock(&mdsc->mutex);
5077         if (__verify_registered_session(mdsc, s) < 0) {
5078                 mutex_unlock(&mdsc->mutex);
5079                 goto out;
5080         }
5081         mutex_unlock(&mdsc->mutex);
5082
5083         switch (type) {
5084         case CEPH_MSG_MDS_MAP:
5085                 ceph_mdsc_handle_mdsmap(mdsc, msg);
5086                 break;
5087         case CEPH_MSG_FS_MAP_USER:
5088                 ceph_mdsc_handle_fsmap(mdsc, msg);
5089                 break;
5090         case CEPH_MSG_CLIENT_SESSION:
5091                 handle_session(s, msg);
5092                 break;
5093         case CEPH_MSG_CLIENT_REPLY:
5094                 handle_reply(s, msg);
5095                 break;
5096         case CEPH_MSG_CLIENT_REQUEST_FORWARD:
5097                 handle_forward(mdsc, s, msg);
5098                 break;
5099         case CEPH_MSG_CLIENT_CAPS:
5100                 ceph_handle_caps(s, msg);
5101                 break;
5102         case CEPH_MSG_CLIENT_SNAP:
5103                 ceph_handle_snap(mdsc, s, msg);
5104                 break;
5105         case CEPH_MSG_CLIENT_LEASE:
5106                 handle_lease(mdsc, s, msg);
5107                 break;
5108         case CEPH_MSG_CLIENT_QUOTA:
5109                 ceph_handle_quota(mdsc, s, msg);
5110                 break;
5111
5112         default:
5113                 pr_err("received unknown message type %d %s\n", type,
5114                        ceph_msg_type_name(type));
5115         }
5116 out:
5117         ceph_msg_put(msg);
5118 }
5119
5120 /*
5121  * authentication
5122  */
5123
5124 /*
5125  * Note: returned pointer is the address of a structure that's
5126  * managed separately.  Caller must *not* attempt to free it.
5127  */
5128 static struct ceph_auth_handshake *
5129 mds_get_authorizer(struct ceph_connection *con, int *proto, int force_new)
5130 {
5131         struct ceph_mds_session *s = con->private;
5132         struct ceph_mds_client *mdsc = s->s_mdsc;
5133         struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
5134         struct ceph_auth_handshake *auth = &s->s_auth;
5135         int ret;
5136
5137         ret = __ceph_auth_get_authorizer(ac, auth, CEPH_ENTITY_TYPE_MDS,
5138                                          force_new, proto, NULL, NULL);
5139         if (ret)
5140                 return ERR_PTR(ret);
5141
5142         return auth;
5143 }
5144
5145 static int mds_add_authorizer_challenge(struct ceph_connection *con,
5146                                     void *challenge_buf, int challenge_buf_len)
5147 {
5148         struct ceph_mds_session *s = con->private;
5149         struct ceph_mds_client *mdsc = s->s_mdsc;
5150         struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
5151
5152         return ceph_auth_add_authorizer_challenge(ac, s->s_auth.authorizer,
5153                                             challenge_buf, challenge_buf_len);
5154 }
5155
5156 static int mds_verify_authorizer_reply(struct ceph_connection *con)
5157 {
5158         struct ceph_mds_session *s = con->private;
5159         struct ceph_mds_client *mdsc = s->s_mdsc;
5160         struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
5161         struct ceph_auth_handshake *auth = &s->s_auth;
5162
5163         return ceph_auth_verify_authorizer_reply(ac, auth->authorizer,
5164                 auth->authorizer_reply_buf, auth->authorizer_reply_buf_len,
5165                 NULL, NULL, NULL, NULL);
5166 }
5167
5168 static int mds_invalidate_authorizer(struct ceph_connection *con)
5169 {
5170         struct ceph_mds_session *s = con->private;
5171         struct ceph_mds_client *mdsc = s->s_mdsc;
5172         struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
5173
5174         ceph_auth_invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);
5175
5176         return ceph_monc_validate_auth(&mdsc->fsc->client->monc);
5177 }
5178
5179 static int mds_get_auth_request(struct ceph_connection *con,
5180                                 void *buf, int *buf_len,
5181                                 void **authorizer, int *authorizer_len)
5182 {
5183         struct ceph_mds_session *s = con->private;
5184         struct ceph_auth_client *ac = s->s_mdsc->fsc->client->monc.auth;
5185         struct ceph_auth_handshake *auth = &s->s_auth;
5186         int ret;
5187
5188         ret = ceph_auth_get_authorizer(ac, auth, CEPH_ENTITY_TYPE_MDS,
5189                                        buf, buf_len);
5190         if (ret)
5191                 return ret;
5192
5193         *authorizer = auth->authorizer_buf;
5194         *authorizer_len = auth->authorizer_buf_len;
5195         return 0;
5196 }
5197
5198 static int mds_handle_auth_reply_more(struct ceph_connection *con,
5199                                       void *reply, int reply_len,
5200                                       void *buf, int *buf_len,
5201                                       void **authorizer, int *authorizer_len)
5202 {
5203         struct ceph_mds_session *s = con->private;
5204         struct ceph_auth_client *ac = s->s_mdsc->fsc->client->monc.auth;
5205         struct ceph_auth_handshake *auth = &s->s_auth;
5206         int ret;
5207
5208         ret = ceph_auth_handle_svc_reply_more(ac, auth, reply, reply_len,
5209                                               buf, buf_len);
5210         if (ret)
5211                 return ret;
5212
5213         *authorizer = auth->authorizer_buf;
5214         *authorizer_len = auth->authorizer_buf_len;
5215         return 0;
5216 }
5217
5218 static int mds_handle_auth_done(struct ceph_connection *con,
5219                                 u64 global_id, void *reply, int reply_len,
5220                                 u8 *session_key, int *session_key_len,
5221                                 u8 *con_secret, int *con_secret_len)
5222 {
5223         struct ceph_mds_session *s = con->private;
5224         struct ceph_auth_client *ac = s->s_mdsc->fsc->client->monc.auth;
5225         struct ceph_auth_handshake *auth = &s->s_auth;
5226
5227         return ceph_auth_handle_svc_reply_done(ac, auth, reply, reply_len,
5228                                                session_key, session_key_len,
5229                                                con_secret, con_secret_len);
5230 }
5231
5232 static int mds_handle_auth_bad_method(struct ceph_connection *con,
5233                                       int used_proto, int result,
5234                                       const int *allowed_protos, int proto_cnt,
5235                                       const int *allowed_modes, int mode_cnt)
5236 {
5237         struct ceph_mds_session *s = con->private;
5238         struct ceph_mon_client *monc = &s->s_mdsc->fsc->client->monc;
5239         int ret;
5240
5241         if (ceph_auth_handle_bad_authorizer(monc->auth, CEPH_ENTITY_TYPE_MDS,
5242                                             used_proto, result,
5243                                             allowed_protos, proto_cnt,
5244                                             allowed_modes, mode_cnt)) {
5245                 ret = ceph_monc_validate_auth(monc);
5246                 if (ret)
5247                         return ret;
5248         }
5249
5250         return -EACCES;
5251 }
5252
5253 static struct ceph_msg *mds_alloc_msg(struct ceph_connection *con,
5254                                 struct ceph_msg_header *hdr, int *skip)
5255 {
5256         struct ceph_msg *msg;
5257         int type = (int) le16_to_cpu(hdr->type);
5258         int front_len = (int) le32_to_cpu(hdr->front_len);
5259
5260         if (con->in_msg)
5261                 return con->in_msg;
5262
5263         *skip = 0;
5264         msg = ceph_msg_new(type, front_len, GFP_NOFS, false);
5265         if (!msg) {
5266                 pr_err("unable to allocate msg type %d len %d\n",
5267                        type, front_len);
5268                 return NULL;
5269         }
5270
5271         return msg;
5272 }
5273
5274 static int mds_sign_message(struct ceph_msg *msg)
5275 {
5276        struct ceph_mds_session *s = msg->con->private;
5277        struct ceph_auth_handshake *auth = &s->s_auth;
5278
5279        return ceph_auth_sign_message(auth, msg);
5280 }
5281
5282 static int mds_check_message_signature(struct ceph_msg *msg)
5283 {
5284        struct ceph_mds_session *s = msg->con->private;
5285        struct ceph_auth_handshake *auth = &s->s_auth;
5286
5287        return ceph_auth_check_message_signature(auth, msg);
5288 }
5289
5290 static const struct ceph_connection_operations mds_con_ops = {
5291         .get = mds_get_con,
5292         .put = mds_put_con,
5293         .alloc_msg = mds_alloc_msg,
5294         .dispatch = mds_dispatch,
5295         .peer_reset = mds_peer_reset,
5296         .get_authorizer = mds_get_authorizer,
5297         .add_authorizer_challenge = mds_add_authorizer_challenge,
5298         .verify_authorizer_reply = mds_verify_authorizer_reply,
5299         .invalidate_authorizer = mds_invalidate_authorizer,
5300         .sign_message = mds_sign_message,
5301         .check_message_signature = mds_check_message_signature,
5302         .get_auth_request = mds_get_auth_request,
5303         .handle_auth_reply_more = mds_handle_auth_reply_more,
5304         .handle_auth_done = mds_handle_auth_done,
5305         .handle_auth_bad_method = mds_handle_auth_bad_method,
5306 };
5307
5308 /* eof */