fs/ubifs/orphan.c

   1 // SPDX-License-Identifier: GPL-2.0-only
   2 /*
   3  * This file is part of UBIFS.
   4  *
   5  * Copyright (C) 2006-2008 Nokia Corporation.
   6  *
   7  * Author: Adrian Hunter
   8  */
   9
  10 #include "ubifs.h"
  11
  12 /*
  13  * An orphan is an inode number whose inode node has been committed to the index
  14  * with a link count of zero. That happens when an open file is deleted
  15  * (unlinked) and then a commit is run. In the normal course of events the inode
  16  * would be deleted when the file is closed. However in the case of an unclean
  17  * unmount, orphans need to be accounted for. After an unclean unmount, the
  18  * orphans' inodes must be deleted which means either scanning the entire index
  19  * looking for them, or keeping a list on flash somewhere. This unit implements
  20  * the latter approach.
  21  *
  22  * The orphan area is a fixed number of LEBs situated between the LPT area and
  23  * the main area. The number of orphan area LEBs is specified when the file
  24  * system is created. The minimum number is 1. The size of the orphan area
  25  * should be so that it can hold the maximum number of orphans that are expected
  26  * to ever exist at one time.
  27  *
  28  * The number of orphans that can fit in a LEB is:
  29  *
  30  *         (c->leb_size - UBIFS_ORPH_NODE_SZ) / sizeof(__le64)
  31  *
  32  * For example: a 15872 byte LEB can fit 1980 orphans so 1 LEB may be enough.
  33  *
  34  * Orphans are accumulated in a rb-tree. When an inode's link count drops to
  35  * zero, the inode number is added to the rb-tree. It is removed from the tree
  36  * when the inode is deleted.  Any new orphans that are in the orphan tree when
  37  * the commit is run, are written to the orphan area in 1 or more orphan nodes.
  38  * If the orphan area is full, it is consolidated to make space.  There is
  39  * always enough space because validation prevents the user from creating more
  40  * than the maximum number of orphans allowed.
  41  */
  42
  43 static int dbg_check_orphans(struct ubifs_info *c);
  44
  45 /**
  46  * ubifs_add_orphan - add an orphan.
  47  * @c: UBIFS file-system description object
  48  * @inum: orphan inode number
  49  *
  50  * Add an orphan. This function is called when an inodes link count drops to
  51  * zero.
  52  */
  53 int ubifs_add_orphan(struct ubifs_info *c, ino_t inum)
  54 {
  55         struct ubifs_orphan *orphan, *o;
  56         struct rb_node **p, *parent = NULL;
  57
  58         orphan = kzalloc(sizeof(struct ubifs_orphan), GFP_NOFS);
  59         if (!orphan)
  60                 return -ENOMEM;
  61         orphan->inum = inum;
  62         orphan->new = 1;
  63
  64         spin_lock(&c->orphan_lock);
  65         if (c->tot_orphans >= c->max_orphans) {
  66                 spin_unlock(&c->orphan_lock);
  67                 kfree(orphan);
  68                 return -ENFILE;
  69         }
  70         p = &c->orph_tree.rb_node;
  71         while (*p) {
  72                 parent = *p;
  73                 o = rb_entry(parent, struct ubifs_orphan, rb);
  74                 if (inum < o->inum)
  75                         p = &(*p)->rb_left;
  76                 else if (inum > o->inum)
  77                         p = &(*p)->rb_right;
  78                 else {
  79                         ubifs_err(c, "ino %lu orphaned twice", (unsigned long)inum);
  80                         spin_unlock(&c->orphan_lock);
  81                         kfree(orphan);
  82                         return -EINVAL;
  83                 }
  84         }
  85         c->tot_orphans += 1;
  86         c->new_orphans += 1;
  87         rb_link_node(&orphan->rb, parent, p);
  88         rb_insert_color(&orphan->rb, &c->orph_tree);
  89         list_add_tail(&orphan->list, &c->orph_list);
  90         list_add_tail(&orphan->new_list, &c->orph_new);
  91
  92         spin_unlock(&c->orphan_lock);
  93         dbg_gen("ino %lu", (unsigned long)inum);
  94         return 0;
  95 }
  96
  97 static struct ubifs_orphan *lookup_orphan(struct ubifs_info *c, ino_t inum)
  98 {
  99         struct ubifs_orphan *o;
 100         struct rb_node *p;
 101
 102         p = c->orph_tree.rb_node;
 103         while (p) {
 104                 o = rb_entry(p, struct ubifs_orphan, rb);
 105                 if (inum < o->inum)
 106                         p = p->rb_left;
 107                 else if (inum > o->inum)
 108                         p = p->rb_right;
 109                 else {
 110                         return o;
 111                 }
 112         }
 113         return NULL;
 114 }
 115
 116 static void __orphan_drop(struct ubifs_info *c, struct ubifs_orphan *o)
 117 {
 118         rb_erase(&o->rb, &c->orph_tree);
 119         list_del(&o->list);
 120         c->tot_orphans -= 1;
 121
 122         if (o->new) {
 123                 list_del(&o->new_list);
 124                 c->new_orphans -= 1;
 125         }
 126
 127         kfree(o);
 128 }
 129
 130 static void orphan_delete(struct ubifs_info *c, struct ubifs_orphan *orph)
 131 {
 132         if (orph->del) {
 133                 dbg_gen("deleted twice ino %lu", (unsigned long)orph->inum);
 134                 return;
 135         }
 136
 137         if (orph->cmt) {
 138                 orph->del = 1;
 139                 rb_erase(&orph->rb, &c->orph_tree);
 140                 orph->dnext = c->orph_dnext;
 141                 c->orph_dnext = orph;
 142                 dbg_gen("delete later ino %lu", (unsigned long)orph->inum);
 143                 return;
 144         }
 145
 146         __orphan_drop(c, orph);
 147 }
 148
 149 /**
 150  * ubifs_delete_orphan - delete an orphan.
 151  * @c: UBIFS file-system description object
 152  * @inum: orphan inode number
 153  *
 154  * Delete an orphan. This function is called when an inode is deleted.
 155  */
 156 void ubifs_delete_orphan(struct ubifs_info *c, ino_t inum)
 157 {
 158         struct ubifs_orphan *orph;
 159
 160         spin_lock(&c->orphan_lock);
 161
 162         orph = lookup_orphan(c, inum);
 163         if (!orph) {
 164                 spin_unlock(&c->orphan_lock);
 165                 ubifs_err(c, "missing orphan ino %lu", (unsigned long)inum);
 166                 dump_stack();
 167
 168                 return;
 169         }
 170
 171         orphan_delete(c, orph);
 172
 173         spin_unlock(&c->orphan_lock);
 174 }
 175
 176 /**
 177  * ubifs_orphan_start_commit - start commit of orphans.
 178  * @c: UBIFS file-system description object
 179  *
 180  * Start commit of orphans.
 181  */
 182 int ubifs_orphan_start_commit(struct ubifs_info *c)
 183 {
 184         struct ubifs_orphan *orphan, **last;
 185
 186         spin_lock(&c->orphan_lock);
 187         last = &c->orph_cnext;
 188         list_for_each_entry(orphan, &c->orph_new, new_list) {
 189                 ubifs_assert(c, orphan->new);
 190                 ubifs_assert(c, !orphan->cmt);
 191                 orphan->new = 0;
 192                 orphan->cmt = 1;
 193                 *last = orphan;
 194                 last = &orphan->cnext;
 195         }
 196         *last = NULL;
 197         c->cmt_orphans = c->new_orphans;
 198         c->new_orphans = 0;
 199         dbg_cmt("%d orphans to commit", c->cmt_orphans);
 200         INIT_LIST_HEAD(&c->orph_new);
 201         if (c->tot_orphans == 0)
 202                 c->no_orphs = 1;
 203         else
 204                 c->no_orphs = 0;
 205         spin_unlock(&c->orphan_lock);
 206         return 0;
 207 }
 208
 209 /**
 210  * avail_orphs - calculate available space.
 211  * @c: UBIFS file-system description object
 212  *
 213  * This function returns the number of orphans that can be written in the
 214  * available space.
 215  */
 216 static int avail_orphs(struct ubifs_info *c)
 217 {
 218         int avail_lebs, avail, gap;
 219
 220         avail_lebs = c->orph_lebs - (c->ohead_lnum - c->orph_first) - 1;
 221         avail = avail_lebs *
 222                ((c->leb_size - UBIFS_ORPH_NODE_SZ) / sizeof(__le64));
 223         gap = c->leb_size - c->ohead_offs;
 224         if (gap >= UBIFS_ORPH_NODE_SZ + sizeof(__le64))
 225                 avail += (gap - UBIFS_ORPH_NODE_SZ) / sizeof(__le64);
 226         return avail;
 227 }
 228
 229 /**
 230  * tot_avail_orphs - calculate total space.
 231  * @c: UBIFS file-system description object
 232  *
 233  * This function returns the number of orphans that can be written in half
 234  * the total space. That leaves half the space for adding new orphans.
 235  */
 236 static int tot_avail_orphs(struct ubifs_info *c)
 237 {
 238         int avail_lebs, avail;
 239
 240         avail_lebs = c->orph_lebs;
 241         avail = avail_lebs *
 242                ((c->leb_size - UBIFS_ORPH_NODE_SZ) / sizeof(__le64));
 243         return avail / 2;
 244 }
 245
 246 /**
 247  * do_write_orph_node - write a node to the orphan head.
 248  * @c: UBIFS file-system description object
 249  * @len: length of node
 250  * @atomic: write atomically
 251  *
 252  * This function writes a node to the orphan head from the orphan buffer. If
 253  * %atomic is not zero, then the write is done atomically. On success, %0 is
 254  * returned, otherwise a negative error code is returned.
 255  */
 256 static int do_write_orph_node(struct ubifs_info *c, int len, int atomic)
 257 {
 258         int err = 0;
 259
 260         if (atomic) {
 261                 ubifs_assert(c, c->ohead_offs == 0);
 262                 ubifs_prepare_node(c, c->orph_buf, len, 1);
 263                 len = ALIGN(len, c->min_io_size);
 264                 err = ubifs_leb_change(c, c->ohead_lnum, c->orph_buf, len);
 265         } else {
 266                 if (c->ohead_offs == 0) {
 267                         /* Ensure LEB has been unmapped */
 268                         err = ubifs_leb_unmap(c, c->ohead_lnum);
 269                         if (err)
 270                                 return err;
 271                 }
 272                 err = ubifs_write_node(c, c->orph_buf, len, c->ohead_lnum,
 273                                        c->ohead_offs);
 274         }
 275         return err;
 276 }
 277
 278 /**
 279  * write_orph_node - write an orphan node.
 280  * @c: UBIFS file-system description object
 281  * @atomic: write atomically
 282  *
 283  * This function builds an orphan node from the cnext list and writes it to the
 284  * orphan head. On success, %0 is returned, otherwise a negative error code
 285  * is returned.
 286  */
 287 static int write_orph_node(struct ubifs_info *c, int atomic)
 288 {
 289         struct ubifs_orphan *orphan, *cnext;
 290         struct ubifs_orph_node *orph;
 291         int gap, err, len, cnt, i;
 292
 293         ubifs_assert(c, c->cmt_orphans > 0);
 294         gap = c->leb_size - c->ohead_offs;
 295         if (gap < UBIFS_ORPH_NODE_SZ + sizeof(__le64)) {
 296                 c->ohead_lnum += 1;
 297                 c->ohead_offs = 0;
 298                 gap = c->leb_size;
 299                 if (c->ohead_lnum > c->orph_last) {
 300                         /*
 301                          * We limit the number of orphans so that this should
 302                          * never happen.
 303                          */
 304                         ubifs_err(c, "out of space in orphan area");
 305                         return -EINVAL;
 306                 }
 307         }
 308         cnt = (gap - UBIFS_ORPH_NODE_SZ) / sizeof(__le64);
 309         if (cnt > c->cmt_orphans)
 310                 cnt = c->cmt_orphans;
 311         len = UBIFS_ORPH_NODE_SZ + cnt * sizeof(__le64);
 312         ubifs_assert(c, c->orph_buf);
 313         orph = c->orph_buf;
 314         orph->ch.node_type = UBIFS_ORPH_NODE;
 315         spin_lock(&c->orphan_lock);
 316         cnext = c->orph_cnext;
 317         for (i = 0; i < cnt; i++) {
 318                 orphan = cnext;
 319                 ubifs_assert(c, orphan->cmt);
 320                 orph->inos[i] = cpu_to_le64(orphan->inum);
 321                 orphan->cmt = 0;
 322                 cnext = orphan->cnext;
 323                 orphan->cnext = NULL;
 324         }
 325         c->orph_cnext = cnext;
 326         c->cmt_orphans -= cnt;
 327         spin_unlock(&c->orphan_lock);
 328         if (c->cmt_orphans)
 329                 orph->cmt_no = cpu_to_le64(c->cmt_no);
 330         else
 331                 /* Mark the last node of the commit */
 332                 orph->cmt_no = cpu_to_le64((c->cmt_no) | (1ULL << 63));
 333         ubifs_assert(c, c->ohead_offs + len <= c->leb_size);
 334         ubifs_assert(c, c->ohead_lnum >= c->orph_first);
 335         ubifs_assert(c, c->ohead_lnum <= c->orph_last);
 336         err = do_write_orph_node(c, len, atomic);
 337         c->ohead_offs += ALIGN(len, c->min_io_size);
 338         c->ohead_offs = ALIGN(c->ohead_offs, 8);
 339         return err;
 340 }
 341
 342 /**
 343  * write_orph_nodes - write orphan nodes until there are no more to commit.
 344  * @c: UBIFS file-system description object
 345  * @atomic: write atomically
 346  *
 347  * This function writes orphan nodes for all the orphans to commit. On success,
 348  * %0 is returned, otherwise a negative error code is returned.
 349  */
 350 static int write_orph_nodes(struct ubifs_info *c, int atomic)
 351 {
 352         int err;
 353
 354         while (c->cmt_orphans > 0) {
 355                 err = write_orph_node(c, atomic);
 356                 if (err)
 357                         return err;
 358         }
 359         if (atomic) {
 360                 int lnum;
 361
 362                 /* Unmap any unused LEBs after consolidation */
 363                 for (lnum = c->ohead_lnum + 1; lnum <= c->orph_last; lnum++) {
 364                         err = ubifs_leb_unmap(c, lnum);
 365                         if (err)
 366                                 return err;
 367                 }
 368         }
 369         return 0;
 370 }
 371
 372 /**
 373  * consolidate - consolidate the orphan area.
 374  * @c: UBIFS file-system description object
 375  *
 376  * This function enables consolidation by putting all the orphans into the list
 377  * to commit. The list is in the order that the orphans were added, and the
 378  * LEBs are written atomically in order, so at no time can orphans be lost by
 379  * an unclean unmount.
 380  *
 381  * This function returns %0 on success and a negative error code on failure.
 382  */
 383 static int consolidate(struct ubifs_info *c)
 384 {
 385         int tot_avail = tot_avail_orphs(c), err = 0;
 386
 387         spin_lock(&c->orphan_lock);
 388         dbg_cmt("there is space for %d orphans and there are %d",
 389                 tot_avail, c->tot_orphans);
 390         if (c->tot_orphans - c->new_orphans <= tot_avail) {
 391                 struct ubifs_orphan *orphan, **last;
 392                 int cnt = 0;
 393
 394                 /* Change the cnext list to include all non-new orphans */
 395                 last = &c->orph_cnext;
 396                 list_for_each_entry(orphan, &c->orph_list, list) {
 397                         if (orphan->new)
 398                                 continue;
 399                         orphan->cmt = 1;
 400                         *last = orphan;
 401                         last = &orphan->cnext;
 402                         cnt += 1;
 403                 }
 404                 *last = NULL;
 405                 ubifs_assert(c, cnt == c->tot_orphans - c->new_orphans);
 406                 c->cmt_orphans = cnt;
 407                 c->ohead_lnum = c->orph_first;
 408                 c->ohead_offs = 0;
 409         } else {
 410                 /*
 411                  * We limit the number of orphans so that this should
 412                  * never happen.
 413                  */
 414                 ubifs_err(c, "out of space in orphan area");
 415                 err = -EINVAL;
 416         }
 417         spin_unlock(&c->orphan_lock);
 418         return err;
 419 }
 420
 421 /**
 422  * commit_orphans - commit orphans.
 423  * @c: UBIFS file-system description object
 424  *
 425  * This function commits orphans to flash. On success, %0 is returned,
 426  * otherwise a negative error code is returned.
 427  */
 428 static int commit_orphans(struct ubifs_info *c)
 429 {
 430         int avail, atomic = 0, err;
 431
 432         ubifs_assert(c, c->cmt_orphans > 0);
 433         avail = avail_orphs(c);
 434         if (avail < c->cmt_orphans) {
 435                 /* Not enough space to write new orphans, so consolidate */
 436                 err = consolidate(c);
 437                 if (err)
 438                         return err;
 439                 atomic = 1;
 440         }
 441         err = write_orph_nodes(c, atomic);
 442         return err;
 443 }
 444
 445 /**
 446  * erase_deleted - erase the orphans marked for deletion.
 447  * @c: UBIFS file-system description object
 448  *
 449  * During commit, the orphans being committed cannot be deleted, so they are
 450  * marked for deletion and deleted by this function. Also, the recovery
 451  * adds killed orphans to the deletion list, and therefore they are deleted
 452  * here too.
 453  */
 454 static void erase_deleted(struct ubifs_info *c)
 455 {
 456         struct ubifs_orphan *orphan, *dnext;
 457
 458         spin_lock(&c->orphan_lock);
 459         dnext = c->orph_dnext;
 460         while (dnext) {
 461                 orphan = dnext;
 462                 dnext = orphan->dnext;
 463                 ubifs_assert(c, !orphan->new);
 464                 ubifs_assert(c, orphan->del);
 465                 list_del(&orphan->list);
 466                 c->tot_orphans -= 1;
 467                 dbg_gen("deleting orphan ino %lu", (unsigned long)orphan->inum);
 468                 kfree(orphan);
 469         }
 470         c->orph_dnext = NULL;
 471         spin_unlock(&c->orphan_lock);
 472 }
 473
 474 /**
 475  * ubifs_orphan_end_commit - end commit of orphans.
 476  * @c: UBIFS file-system description object
 477  *
 478  * End commit of orphans.
 479  */
 480 int ubifs_orphan_end_commit(struct ubifs_info *c)
 481 {
 482         int err;
 483
 484         if (c->cmt_orphans != 0) {
 485                 err = commit_orphans(c);
 486                 if (err)
 487                         return err;
 488         }
 489         erase_deleted(c);
 490         err = dbg_check_orphans(c);
 491         return err;
 492 }
 493
 494 /**
 495  * ubifs_clear_orphans - erase all LEBs used for orphans.
 496  * @c: UBIFS file-system description object
 497  *
 498  * If recovery is not required, then the orphans from the previous session
 499  * are not needed. This function locates the LEBs used to record
 500  * orphans, and un-maps them.
 501  */
 502 int ubifs_clear_orphans(struct ubifs_info *c)
 503 {
 504         int lnum, err;
 505
 506         for (lnum = c->orph_first; lnum <= c->orph_last; lnum++) {
 507                 err = ubifs_leb_unmap(c, lnum);
 508                 if (err)
 509                         return err;
 510         }
 511         c->ohead_lnum = c->orph_first;
 512         c->ohead_offs = 0;
 513         return 0;
 514 }
 515
 516 /**
 517  * do_kill_orphans - remove orphan inodes from the index.
 518  * @c: UBIFS file-system description object
 519  * @sleb: scanned LEB
 520  * @last_cmt_no: cmt_no of last orphan node read is passed and returned here
 521  * @outofdate: whether the LEB is out of date is returned here
 522  * @last_flagged: whether the end orphan node is encountered
 523  *
 524  * This function is a helper to the 'kill_orphans()' function. It goes through
 525  * every orphan node in a LEB and for every inode number recorded, removes
 526  * all keys for that inode from the TNC.
 527  */
 528 static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 529                            unsigned long long *last_cmt_no, int *outofdate,
 530                            int *last_flagged)
 531 {
 532         struct ubifs_scan_node *snod;
 533         struct ubifs_orph_node *orph;
 534         struct ubifs_ino_node *ino = NULL;
 535         unsigned long long cmt_no;
 536         ino_t inum;
 537         int i, n, err, first = 1;
 538
 539         ino = kmalloc(UBIFS_MAX_INO_NODE_SZ, GFP_NOFS);
 540         if (!ino)
 541                 return -ENOMEM;
 542
 543         list_for_each_entry(snod, &sleb->nodes, list) {
 544                 if (snod->type != UBIFS_ORPH_NODE) {
 545                         ubifs_err(c, "invalid node type %d in orphan area at %d:%d",
 546                                   snod->type, sleb->lnum, snod->offs);
 547                         ubifs_dump_node(c, snod->node,
 548                                         c->leb_size - snod->offs);
 549                         err = -EINVAL;
 550                         goto out_free;
 551                 }
 552
 553                 orph = snod->node;
 554
 555                 /* Check commit number */
 556                 cmt_no = le64_to_cpu(orph->cmt_no) & LLONG_MAX;
 557                 /*
 558                  * The commit number on the master node may be less, because
 559                  * of a failed commit. If there are several failed commits in a
 560                  * row, the commit number written on orphan nodes will continue
 561                  * to increase (because the commit number is adjusted here) even
 562                  * though the commit number on the master node stays the same
 563                  * because the master node has not been re-written.
 564                  */
 565                 if (cmt_no > c->cmt_no)
 566                         c->cmt_no = cmt_no;
 567                 if (cmt_no < *last_cmt_no && *last_flagged) {
 568                         /*
 569                          * The last orphan node had a higher commit number and
 570                          * was flagged as the last written for that commit
 571                          * number. That makes this orphan node, out of date.
 572                          */
 573                         if (!first) {
 574                                 ubifs_err(c, "out of order commit number %llu in orphan node at %d:%d",
 575                                           cmt_no, sleb->lnum, snod->offs);
 576                                 ubifs_dump_node(c, snod->node,
 577                                                 c->leb_size - snod->offs);
 578                                 err = -EINVAL;
 579                                 goto out_free;
 580                         }
 581                         dbg_rcvry("out of date LEB %d", sleb->lnum);
 582                         *outofdate = 1;
 583                         err = 0;
 584                         goto out_free;
 585                 }
 586
 587                 if (first)
 588                         first = 0;
 589
 590                 n = (le32_to_cpu(orph->ch.len) - UBIFS_ORPH_NODE_SZ) >> 3;
 591                 for (i = 0; i < n; i++) {
 592                         union ubifs_key key;
 593
 594                         inum = le64_to_cpu(orph->inos[i]);
 595
 596                         ino_key_init(c, &key, inum);
 597                         err = ubifs_tnc_lookup(c, &key, ino);
 598                         if (err && err != -ENOENT)
 599                                 goto out_free;
 600
 601                         /*
 602                          * Check whether an inode can really get deleted.
 603                          * linkat() with O_TMPFILE allows rebirth of an inode.
 604                          */
 605                         if (err == 0 && ino->nlink == 0) {
 606                                 dbg_rcvry("deleting orphaned inode %lu",
 607                                           (unsigned long)inum);
 608
 609                                 err = ubifs_tnc_remove_ino(c, inum);
 610                                 if (err)
 611                                         goto out_ro;
 612                         }
 613                 }
 614
 615                 *last_cmt_no = cmt_no;
 616                 if (le64_to_cpu(orph->cmt_no) & (1ULL << 63)) {
 617                         dbg_rcvry("last orph node for commit %llu at %d:%d",
 618                                   cmt_no, sleb->lnum, snod->offs);
 619                         *last_flagged = 1;
 620                 } else
 621                         *last_flagged = 0;
 622         }
 623
 624         err = 0;
 625 out_free:
 626         kfree(ino);
 627         return err;
 628
 629 out_ro:
 630         ubifs_ro_mode(c, err);
 631         kfree(ino);
 632         return err;
 633 }
 634
 635 /**
 636  * kill_orphans - remove all orphan inodes from the index.
 637  * @c: UBIFS file-system description object
 638  *
 639  * If recovery is required, then orphan inodes recorded during the previous
 640  * session (which ended with an unclean unmount) must be deleted from the index.
 641  * This is done by updating the TNC, but since the index is not updated until
 642  * the next commit, the LEBs where the orphan information is recorded are not
 643  * erased until the next commit.
 644  */
 645 static int kill_orphans(struct ubifs_info *c)
 646 {
 647         unsigned long long last_cmt_no = 0;
 648         int lnum, err = 0, outofdate = 0, last_flagged = 0;
 649
 650         c->ohead_lnum = c->orph_first;
 651         c->ohead_offs = 0;
 652         /* Check no-orphans flag and skip this if no orphans */
 653         if (c->no_orphs) {
 654                 dbg_rcvry("no orphans");
 655                 return 0;
 656         }
 657         /*
 658          * Orph nodes always start at c->orph_first and are written to each
 659          * successive LEB in turn. Generally unused LEBs will have been unmapped
 660          * but may contain out of date orphan nodes if the unmap didn't go
 661          * through. In addition, the last orphan node written for each commit is
 662          * marked (top bit of orph->cmt_no is set to 1). It is possible that
 663          * there are orphan nodes from the next commit (i.e. the commit did not
 664          * complete successfully). In that case, no orphans will have been lost
 665          * due to the way that orphans are written, and any orphans added will
 666          * be valid orphans anyway and so can be deleted.
 667          */
 668         for (lnum = c->orph_first; lnum <= c->orph_last; lnum++) {
 669                 struct ubifs_scan_leb *sleb;
 670
 671                 dbg_rcvry("LEB %d", lnum);
 672                 sleb = ubifs_scan(c, lnum, 0, c->sbuf, 1);
 673                 if (IS_ERR(sleb)) {
 674                         if (PTR_ERR(sleb) == -EUCLEAN)
 675                                 sleb = ubifs_recover_leb(c, lnum, 0,
 676                                                          c->sbuf, -1);
 677                         if (IS_ERR(sleb)) {
 678                                 err = PTR_ERR(sleb);
 679                                 break;
 680                         }
 681                 }
 682                 err = do_kill_orphans(c, sleb, &last_cmt_no, &outofdate,
 683                                       &last_flagged);
 684                 if (err || outofdate) {
 685                         ubifs_scan_destroy(sleb);
 686                         break;
 687                 }
 688                 if (sleb->endpt) {
 689                         c->ohead_lnum = lnum;
 690                         c->ohead_offs = sleb->endpt;
 691                 }
 692                 ubifs_scan_destroy(sleb);
 693         }
 694         return err;
 695 }
 696
 697 /**
 698  * ubifs_mount_orphans - delete orphan inodes and erase LEBs that recorded them.
 699  * @c: UBIFS file-system description object
 700  * @unclean: indicates recovery from unclean unmount
 701  * @read_only: indicates read only mount
 702  *
 703  * This function is called when mounting to erase orphans from the previous
 704  * session. If UBIFS was not unmounted cleanly, then the inodes recorded as
 705  * orphans are deleted.
 706  */
 707 int ubifs_mount_orphans(struct ubifs_info *c, int unclean, int read_only)
 708 {
 709         int err = 0;
 710
 711         c->max_orphans = tot_avail_orphs(c);
 712
 713         if (!read_only) {
 714                 c->orph_buf = vmalloc(c->leb_size);
 715                 if (!c->orph_buf)
 716                         return -ENOMEM;
 717         }
 718
 719         if (unclean)
 720                 err = kill_orphans(c);
 721         else if (!read_only)
 722                 err = ubifs_clear_orphans(c);
 723
 724         return err;
 725 }
 726
 727 /*
 728  * Everything below is related to debugging.
 729  */
 730
 731 struct check_orphan {
 732         struct rb_node rb;
 733         ino_t inum;
 734 };
 735
 736 struct check_info {
 737         unsigned long last_ino;
 738         unsigned long tot_inos;
 739         unsigned long missing;
 740         unsigned long long leaf_cnt;
 741         struct ubifs_ino_node *node;
 742         struct rb_root root;
 743 };
 744
 745 static bool dbg_find_orphan(struct ubifs_info *c, ino_t inum)
 746 {
 747         bool found = false;
 748
 749         spin_lock(&c->orphan_lock);
 750         found = !!lookup_orphan(c, inum);
 751         spin_unlock(&c->orphan_lock);
 752
 753         return found;
 754 }
 755
 756 static int dbg_ins_check_orphan(struct rb_root *root, ino_t inum)
 757 {
 758         struct check_orphan *orphan, *o;
 759         struct rb_node **p, *parent = NULL;
 760
 761         orphan = kzalloc(sizeof(struct check_orphan), GFP_NOFS);
 762         if (!orphan)
 763                 return -ENOMEM;
 764         orphan->inum = inum;
 765
 766         p = &root->rb_node;
 767         while (*p) {
 768                 parent = *p;
 769                 o = rb_entry(parent, struct check_orphan, rb);
 770                 if (inum < o->inum)
 771                         p = &(*p)->rb_left;
 772                 else if (inum > o->inum)
 773                         p = &(*p)->rb_right;
 774                 else {
 775                         kfree(orphan);
 776                         return 0;
 777                 }
 778         }
 779         rb_link_node(&orphan->rb, parent, p);
 780         rb_insert_color(&orphan->rb, root);
 781         return 0;
 782 }
 783
 784 static int dbg_find_check_orphan(struct rb_root *root, ino_t inum)
 785 {
 786         struct check_orphan *o;
 787         struct rb_node *p;
 788
 789         p = root->rb_node;
 790         while (p) {
 791                 o = rb_entry(p, struct check_orphan, rb);
 792                 if (inum < o->inum)
 793                         p = p->rb_left;
 794                 else if (inum > o->inum)
 795                         p = p->rb_right;
 796                 else
 797                         return 1;
 798         }
 799         return 0;
 800 }
 801
 802 static void dbg_free_check_tree(struct rb_root *root)
 803 {
 804         struct check_orphan *o, *n;
 805
 806         rbtree_postorder_for_each_entry_safe(o, n, root, rb)
 807                 kfree(o);
 808 }
 809
 810 static int dbg_orphan_check(struct ubifs_info *c, struct ubifs_zbranch *zbr,
 811                             void *priv)
 812 {
 813         struct check_info *ci = priv;
 814         ino_t inum;
 815         int err;
 816
 817         inum = key_inum(c, &zbr->key);
 818         if (inum != ci->last_ino) {
 819                 /*
 820                  * Lowest node type is the inode node or xattr entry(when
 821                  * selinux/encryption is enabled), so it comes first
 822                  */
 823                 if (key_type(c, &zbr->key) != UBIFS_INO_KEY &&
 824                     key_type(c, &zbr->key) != UBIFS_XENT_KEY)
 825                         ubifs_err(c, "found orphan node ino %lu, type %d",
 826                                   (unsigned long)inum, key_type(c, &zbr->key));
 827                 ci->last_ino = inum;
 828                 ci->tot_inos += 1;
 829                 err = ubifs_tnc_read_node(c, zbr, ci->node);
 830                 if (err) {
 831                         ubifs_err(c, "node read failed, error %d", err);
 832                         return err;
 833                 }
 834                 if (ci->node->nlink == 0)
 835                         /* Must be recorded as an orphan */
 836                         if (!dbg_find_check_orphan(&ci->root, inum) &&
 837                             !dbg_find_orphan(c, inum)) {
 838                                 ubifs_err(c, "missing orphan, ino %lu",
 839                                           (unsigned long)inum);
 840                                 ci->missing += 1;
 841                         }
 842         }
 843         ci->leaf_cnt += 1;
 844         return 0;
 845 }
 846
 847 static int dbg_read_orphans(struct check_info *ci, struct ubifs_scan_leb *sleb)
 848 {
 849         struct ubifs_scan_node *snod;
 850         struct ubifs_orph_node *orph;
 851         ino_t inum;
 852         int i, n, err;
 853
 854         list_for_each_entry(snod, &sleb->nodes, list) {
 855                 cond_resched();
 856                 if (snod->type != UBIFS_ORPH_NODE)
 857                         continue;
 858                 orph = snod->node;
 859                 n = (le32_to_cpu(orph->ch.len) - UBIFS_ORPH_NODE_SZ) >> 3;
 860                 for (i = 0; i < n; i++) {
 861                         inum = le64_to_cpu(orph->inos[i]);
 862                         err = dbg_ins_check_orphan(&ci->root, inum);
 863                         if (err)
 864                                 return err;
 865                 }
 866         }
 867         return 0;
 868 }
 869
 870 static int dbg_scan_orphans(struct ubifs_info *c, struct check_info *ci)
 871 {
 872         int lnum, err = 0;
 873         void *buf;
 874
 875         /* Check no-orphans flag and skip this if no orphans */
 876         if (c->no_orphs)
 877                 return 0;
 878
 879         buf = __vmalloc(c->leb_size, GFP_NOFS);
 880         if (!buf) {
 881                 ubifs_err(c, "cannot allocate memory to check orphans");
 882                 return 0;
 883         }
 884
 885         for (lnum = c->orph_first; lnum <= c->orph_last; lnum++) {
 886                 struct ubifs_scan_leb *sleb;
 887
 888                 sleb = ubifs_scan(c, lnum, 0, buf, 0);
 889                 if (IS_ERR(sleb)) {
 890                         err = PTR_ERR(sleb);
 891                         break;
 892                 }
 893
 894                 err = dbg_read_orphans(ci, sleb);
 895                 ubifs_scan_destroy(sleb);
 896                 if (err)
 897                         break;
 898         }
 899
 900         vfree(buf);
 901         return err;
 902 }
 903
 904 static int dbg_check_orphans(struct ubifs_info *c)
 905 {
 906         struct check_info ci;
 907         int err;
 908
 909         if (!dbg_is_chk_orph(c))
 910                 return 0;
 911
 912         ci.last_ino = 0;
 913         ci.tot_inos = 0;
 914         ci.missing  = 0;
 915         ci.leaf_cnt = 0;
 916         ci.root = RB_ROOT;
 917         ci.node = kmalloc(UBIFS_MAX_INO_NODE_SZ, GFP_NOFS);
 918         if (!ci.node) {
 919                 ubifs_err(c, "out of memory");
 920                 return -ENOMEM;
 921         }
 922
 923         err = dbg_scan_orphans(c, &ci);
 924         if (err)
 925                 goto out;
 926
 927         err = dbg_walk_index(c, &dbg_orphan_check, NULL, &ci);
 928         if (err) {
 929                 ubifs_err(c, "cannot scan TNC, error %d", err);
 930                 goto out;
 931         }
 932
 933         if (ci.missing) {
 934                 ubifs_err(c, "%lu missing orphan(s)", ci.missing);
 935                 err = -EINVAL;
 936                 goto out;
 937         }
 938
 939         dbg_cmt("last inode number is %lu", ci.last_ino);
 940         dbg_cmt("total number of inodes is %lu", ci.tot_inos);
 941         dbg_cmt("total number of leaf nodes is %llu", ci.leaf_cnt);
 942
 943 out:
 944         dbg_free_check_tree(&ci.root);
 945         kfree(ci.node);
 946         return err;
 947 }