fs/reiserfs/stree.c

   1 /*
   2  *  Copyright 2000 by Hans Reiser, licensing governed by reiserfs/README
   3  */
   4
   5 /*
   6  *  Written by Anatoly P. Pinchuk pap@namesys.botik.ru
   7  *  Programm System Institute
   8  *  Pereslavl-Zalessky Russia
   9  */
  10
  11 /*
  12  *  This file contains functions dealing with S+tree
  13  *
  14  * B_IS_IN_TREE
  15  * copy_item_head
  16  * comp_short_keys
  17  * comp_keys
  18  * comp_short_le_keys
  19  * le_key2cpu_key
  20  * comp_le_keys
  21  * bin_search
  22  * get_lkey
  23  * get_rkey
  24  * key_in_buffer
  25  * decrement_bcount
  26  * reiserfs_check_path
  27  * pathrelse_and_restore
  28  * pathrelse
  29  * search_by_key_reada
  30  * search_by_key
  31  * search_for_position_by_key
  32  * comp_items
  33  * prepare_for_direct_item
  34  * prepare_for_direntry_item
  35  * prepare_for_delete_or_cut
  36  * calc_deleted_bytes_number
  37  * init_tb_struct
  38  * padd_item
  39  * reiserfs_delete_item
  40  * reiserfs_delete_solid_item
  41  * reiserfs_delete_object
  42  * maybe_indirect_to_direct
  43  * indirect_to_direct_roll_back
  44  * reiserfs_cut_from_item
  45  * truncate_directory
  46  * reiserfs_do_truncate
  47  * reiserfs_paste_into_item
  48  * reiserfs_insert_item
  49  */
  50
  51 #include <linux/time.h>
  52 #include <linux/string.h>
  53 #include <linux/pagemap.h>
  54 #include <linux/reiserfs_fs.h>
  55 #include <linux/buffer_head.h>
  56 #include <linux/quotaops.h>
  57
  58 /* Does the buffer contain a disk block which is in the tree. */
  59 inline int B_IS_IN_TREE(const struct buffer_head *p_s_bh)
  60 {
  61
  62         RFALSE(B_LEVEL(p_s_bh) > MAX_HEIGHT,
  63                "PAP-1010: block (%b) has too big level (%z)", p_s_bh, p_s_bh);
  64
  65         return (B_LEVEL(p_s_bh) != FREE_LEVEL);
  66 }
  67
  68 //
  69 // to gets item head in le form
  70 //
  71 inline void copy_item_head(struct item_head *p_v_to,
  72                            const struct item_head *p_v_from)
  73 {
  74         memcpy(p_v_to, p_v_from, IH_SIZE);
  75 }
  76
  77 /* k1 is pointer to on-disk structure which is stored in little-endian
  78    form. k2 is pointer to cpu variable. For key of items of the same
  79    object this returns 0.
  80    Returns: -1 if key1 < key2
  81    0 if key1 == key2
  82    1 if key1 > key2 */
  83 inline int comp_short_keys(const struct reiserfs_key *le_key,
  84                            const struct cpu_key *cpu_key)
  85 {
  86         __u32 n;
  87         n = le32_to_cpu(le_key->k_dir_id);
  88         if (n < cpu_key->on_disk_key.k_dir_id)
  89                 return -1;
  90         if (n > cpu_key->on_disk_key.k_dir_id)
  91                 return 1;
  92         n = le32_to_cpu(le_key->k_objectid);
  93         if (n < cpu_key->on_disk_key.k_objectid)
  94                 return -1;
  95         if (n > cpu_key->on_disk_key.k_objectid)
  96                 return 1;
  97         return 0;
  98 }
  99
 100 /* k1 is pointer to on-disk structure which is stored in little-endian
 101    form. k2 is pointer to cpu variable.
 102    Compare keys using all 4 key fields.
 103    Returns: -1 if key1 < key2 0
 104    if key1 = key2 1 if key1 > key2 */
 105 static inline int comp_keys(const struct reiserfs_key *le_key,
 106                             const struct cpu_key *cpu_key)
 107 {
 108         int retval;
 109
 110         retval = comp_short_keys(le_key, cpu_key);
 111         if (retval)
 112                 return retval;
 113         if (le_key_k_offset(le_key_version(le_key), le_key) <
 114             cpu_key_k_offset(cpu_key))
 115                 return -1;
 116         if (le_key_k_offset(le_key_version(le_key), le_key) >
 117             cpu_key_k_offset(cpu_key))
 118                 return 1;
 119
 120         if (cpu_key->key_length == 3)
 121                 return 0;
 122
 123         /* this part is needed only when tail conversion is in progress */
 124         if (le_key_k_type(le_key_version(le_key), le_key) <
 125             cpu_key_k_type(cpu_key))
 126                 return -1;
 127
 128         if (le_key_k_type(le_key_version(le_key), le_key) >
 129             cpu_key_k_type(cpu_key))
 130                 return 1;
 131
 132         return 0;
 133 }
 134
 135 inline int comp_short_le_keys(const struct reiserfs_key *key1,
 136                               const struct reiserfs_key *key2)
 137 {
 138         __u32 *p_s_1_u32, *p_s_2_u32;
 139         int n_key_length = REISERFS_SHORT_KEY_LEN;
 140
 141         p_s_1_u32 = (__u32 *) key1;
 142         p_s_2_u32 = (__u32 *) key2;
 143         for (; n_key_length--; ++p_s_1_u32, ++p_s_2_u32) {
 144                 if (le32_to_cpu(*p_s_1_u32) < le32_to_cpu(*p_s_2_u32))
 145                         return -1;
 146                 if (le32_to_cpu(*p_s_1_u32) > le32_to_cpu(*p_s_2_u32))
 147                         return 1;
 148         }
 149         return 0;
 150 }
 151
 152 inline void le_key2cpu_key(struct cpu_key *to, const struct reiserfs_key *from)
 153 {
 154         int version;
 155         to->on_disk_key.k_dir_id = le32_to_cpu(from->k_dir_id);
 156         to->on_disk_key.k_objectid = le32_to_cpu(from->k_objectid);
 157
 158         // find out version of the key
 159         version = le_key_version(from);
 160         to->version = version;
 161         to->on_disk_key.k_offset = le_key_k_offset(version, from);
 162         to->on_disk_key.k_type = le_key_k_type(version, from);
 163 }
 164
 165 // this does not say which one is bigger, it only returns 1 if keys
 166 // are not equal, 0 otherwise
 167 inline int comp_le_keys(const struct reiserfs_key *k1,
 168                         const struct reiserfs_key *k2)
 169 {
 170         return memcmp(k1, k2, sizeof(struct reiserfs_key));
 171 }
 172
 173 /**************************************************************************
 174  *  Binary search toolkit function                                        *
 175  *  Search for an item in the array by the item key                       *
 176  *  Returns:    1 if found,  0 if not found;                              *
 177  *        *p_n_pos = number of the searched element if found, else the    *
 178  *        number of the first element that is larger than p_v_key.        *
 179  **************************************************************************/
 180 /* For those not familiar with binary search: n_lbound is the leftmost item that it
 181  could be, n_rbound the rightmost item that it could be.  We examine the item
 182  halfway between n_lbound and n_rbound, and that tells us either that we can increase
 183  n_lbound, or decrease n_rbound, or that we have found it, or if n_lbound <= n_rbound that
 184  there are no possible items, and we have not found it. With each examination we
 185  cut the number of possible items it could be by one more than half rounded down,
 186  or we find it. */
 187 static inline int bin_search(const void *p_v_key,       /* Key to search for.                   */
 188                              const void *p_v_base,      /* First item in the array.             */
 189                              int p_n_num,       /* Number of items in the array.        */
 190                              int p_n_width,     /* Item size in the array.
 191                                                    searched. Lest the reader be
 192                                                    confused, note that this is crafted
 193                                                    as a general function, and when it
 194                                                    is applied specifically to the array
 195                                                    of item headers in a node, p_n_width
 196                                                    is actually the item header size not
 197                                                    the item size.                      */
 198                              int *p_n_pos       /* Number of the searched for element. */
 199     )
 200 {
 201         int n_rbound, n_lbound, n_j;
 202
 203         for (n_j = ((n_rbound = p_n_num - 1) + (n_lbound = 0)) / 2;
 204              n_lbound <= n_rbound; n_j = (n_rbound + n_lbound) / 2)
 205                 switch (comp_keys
 206                         ((struct reiserfs_key *)((char *)p_v_base +
 207                                                  n_j * p_n_width),
 208                          (struct cpu_key *)p_v_key)) {
 209                 case -1:
 210                         n_lbound = n_j + 1;
 211                         continue;
 212                 case 1:
 213                         n_rbound = n_j - 1;
 214                         continue;
 215                 case 0:
 216                         *p_n_pos = n_j;
 217                         return ITEM_FOUND;      /* Key found in the array.  */
 218                 }
 219
 220         /* bin_search did not find given key, it returns position of key,
 221            that is minimal and greater than the given one. */
 222         *p_n_pos = n_lbound;
 223         return ITEM_NOT_FOUND;
 224 }
 225
 226 #ifdef CONFIG_REISERFS_CHECK
 227 extern struct tree_balance *cur_tb;
 228 #endif
 229
 230 /* Minimal possible key. It is never in the tree. */
 231 const struct reiserfs_key MIN_KEY = { 0, 0, {{0, 0},} };
 232
 233 /* Maximal possible key. It is never in the tree. */
 234 static const struct reiserfs_key MAX_KEY = {
 235         __constant_cpu_to_le32(0xffffffff),
 236         __constant_cpu_to_le32(0xffffffff),
 237         {{__constant_cpu_to_le32(0xffffffff),
 238           __constant_cpu_to_le32(0xffffffff)},}
 239 };
 240
 241 /* Get delimiting key of the buffer by looking for it in the buffers in the path, starting from the bottom
 242    of the path, and going upwards.  We must check the path's validity at each step.  If the key is not in
 243    the path, there is no delimiting key in the tree (buffer is first or last buffer in tree), and in this
 244    case we return a special key, either MIN_KEY or MAX_KEY. */
 245 static inline const struct reiserfs_key *get_lkey(const struct treepath
 246                                                   *p_s_chk_path,
 247                                                   const struct super_block
 248                                                   *sb)
 249 {
 250         int n_position, n_path_offset = p_s_chk_path->path_length;
 251         struct buffer_head *p_s_parent;
 252
 253         RFALSE(n_path_offset < FIRST_PATH_ELEMENT_OFFSET,
 254                "PAP-5010: invalid offset in the path");
 255
 256         /* While not higher in path than first element. */
 257         while (n_path_offset-- > FIRST_PATH_ELEMENT_OFFSET) {
 258
 259                 RFALSE(!buffer_uptodate
 260                        (PATH_OFFSET_PBUFFER(p_s_chk_path, n_path_offset)),
 261                        "PAP-5020: parent is not uptodate");
 262
 263                 /* Parent at the path is not in the tree now. */
 264                 if (!B_IS_IN_TREE
 265                     (p_s_parent =
 266                      PATH_OFFSET_PBUFFER(p_s_chk_path, n_path_offset)))
 267                         return &MAX_KEY;
 268                 /* Check whether position in the parent is correct. */
 269                 if ((n_position =
 270                      PATH_OFFSET_POSITION(p_s_chk_path,
 271                                           n_path_offset)) >
 272                     B_NR_ITEMS(p_s_parent))
 273                         return &MAX_KEY;
 274                 /* Check whether parent at the path really points to the child. */
 275                 if (B_N_CHILD_NUM(p_s_parent, n_position) !=
 276                     PATH_OFFSET_PBUFFER(p_s_chk_path,
 277                                         n_path_offset + 1)->b_blocknr)
 278                         return &MAX_KEY;
 279                 /* Return delimiting key if position in the parent is not equal to zero. */
 280                 if (n_position)
 281                         return B_N_PDELIM_KEY(p_s_parent, n_position - 1);
 282         }
 283         /* Return MIN_KEY if we are in the root of the buffer tree. */
 284         if (PATH_OFFSET_PBUFFER(p_s_chk_path, FIRST_PATH_ELEMENT_OFFSET)->
 285             b_blocknr == SB_ROOT_BLOCK(sb))
 286                 return &MIN_KEY;
 287         return &MAX_KEY;
 288 }
 289
 290 /* Get delimiting key of the buffer at the path and its right neighbor. */
 291 inline const struct reiserfs_key *get_rkey(const struct treepath *p_s_chk_path,
 292                                            const struct super_block *sb)
 293 {
 294         int n_position, n_path_offset = p_s_chk_path->path_length;
 295         struct buffer_head *p_s_parent;
 296
 297         RFALSE(n_path_offset < FIRST_PATH_ELEMENT_OFFSET,
 298                "PAP-5030: invalid offset in the path");
 299
 300         while (n_path_offset-- > FIRST_PATH_ELEMENT_OFFSET) {
 301
 302                 RFALSE(!buffer_uptodate
 303                        (PATH_OFFSET_PBUFFER(p_s_chk_path, n_path_offset)),
 304                        "PAP-5040: parent is not uptodate");
 305
 306                 /* Parent at the path is not in the tree now. */
 307                 if (!B_IS_IN_TREE
 308                     (p_s_parent =
 309                      PATH_OFFSET_PBUFFER(p_s_chk_path, n_path_offset)))
 310                         return &MIN_KEY;
 311                 /* Check whether position in the parent is correct. */
 312                 if ((n_position =
 313                      PATH_OFFSET_POSITION(p_s_chk_path,
 314                                           n_path_offset)) >
 315                     B_NR_ITEMS(p_s_parent))
 316                         return &MIN_KEY;
 317                 /* Check whether parent at the path really points to the child. */
 318                 if (B_N_CHILD_NUM(p_s_parent, n_position) !=
 319                     PATH_OFFSET_PBUFFER(p_s_chk_path,
 320                                         n_path_offset + 1)->b_blocknr)
 321                         return &MIN_KEY;
 322                 /* Return delimiting key if position in the parent is not the last one. */
 323                 if (n_position != B_NR_ITEMS(p_s_parent))
 324                         return B_N_PDELIM_KEY(p_s_parent, n_position);
 325         }
 326         /* Return MAX_KEY if we are in the root of the buffer tree. */
 327         if (PATH_OFFSET_PBUFFER(p_s_chk_path, FIRST_PATH_ELEMENT_OFFSET)->
 328             b_blocknr == SB_ROOT_BLOCK(sb))
 329                 return &MAX_KEY;
 330         return &MIN_KEY;
 331 }
 332
 333 /* Check whether a key is contained in the tree rooted from a buffer at a path. */
 334 /* This works by looking at the left and right delimiting keys for the buffer in the last path_element in
 335    the path.  These delimiting keys are stored at least one level above that buffer in the tree. If the
 336    buffer is the first or last node in the tree order then one of the delimiting keys may be absent, and in
 337    this case get_lkey and get_rkey return a special key which is MIN_KEY or MAX_KEY. */
 338 static inline int key_in_buffer(struct treepath *p_s_chk_path,  /* Path which should be checked.  */
 339                                 const struct cpu_key *p_s_key,  /* Key which should be checked.   */
 340                                 struct super_block *sb  /* Super block pointer.    */
 341     )
 342 {
 343
 344         RFALSE(!p_s_key || p_s_chk_path->path_length < FIRST_PATH_ELEMENT_OFFSET
 345                || p_s_chk_path->path_length > MAX_HEIGHT,
 346                "PAP-5050: pointer to the key(%p) is NULL or invalid path length(%d)",
 347                p_s_key, p_s_chk_path->path_length);
 348         RFALSE(!PATH_PLAST_BUFFER(p_s_chk_path)->b_bdev,
 349                "PAP-5060: device must not be NODEV");
 350
 351         if (comp_keys(get_lkey(p_s_chk_path, sb), p_s_key) == 1)
 352                 /* left delimiting key is bigger, that the key we look for */
 353                 return 0;
 354         //  if ( comp_keys(p_s_key, get_rkey(p_s_chk_path, sb)) != -1 )
 355         if (comp_keys(get_rkey(p_s_chk_path, sb), p_s_key) != 1)
 356                 /* p_s_key must be less than right delimitiing key */
 357                 return 0;
 358         return 1;
 359 }
 360
 361 int reiserfs_check_path(struct treepath *p)
 362 {
 363         RFALSE(p->path_length != ILLEGAL_PATH_ELEMENT_OFFSET,
 364                "path not properly relsed");
 365         return 0;
 366 }
 367
 368 /* Drop the reference to each buffer in a path and restore
 369  * dirty bits clean when preparing the buffer for the log.
 370  * This version should only be called from fix_nodes() */
 371 void pathrelse_and_restore(struct super_block *sb,
 372                            struct treepath *p_s_search_path)
 373 {
 374         int n_path_offset = p_s_search_path->path_length;
 375
 376         RFALSE(n_path_offset < ILLEGAL_PATH_ELEMENT_OFFSET,
 377                "clm-4000: invalid path offset");
 378
 379         while (n_path_offset > ILLEGAL_PATH_ELEMENT_OFFSET) {
 380                 struct buffer_head *bh;
 381                 bh = PATH_OFFSET_PBUFFER(p_s_search_path, n_path_offset--);
 382                 reiserfs_restore_prepared_buffer(sb, bh);
 383                 brelse(bh);
 384         }
 385         p_s_search_path->path_length = ILLEGAL_PATH_ELEMENT_OFFSET;
 386 }
 387
 388 /* Drop the reference to each buffer in a path */
 389 void pathrelse(struct treepath *p_s_search_path)
 390 {
 391         int n_path_offset = p_s_search_path->path_length;
 392
 393         RFALSE(n_path_offset < ILLEGAL_PATH_ELEMENT_OFFSET,
 394                "PAP-5090: invalid path offset");
 395
 396         while (n_path_offset > ILLEGAL_PATH_ELEMENT_OFFSET)
 397                 brelse(PATH_OFFSET_PBUFFER(p_s_search_path, n_path_offset--));
 398
 399         p_s_search_path->path_length = ILLEGAL_PATH_ELEMENT_OFFSET;
 400 }
 401
 402 static int is_leaf(char *buf, int blocksize, struct buffer_head *bh)
 403 {
 404         struct block_head *blkh;
 405         struct item_head *ih;
 406         int used_space;
 407         int prev_location;
 408         int i;
 409         int nr;
 410
 411         blkh = (struct block_head *)buf;
 412         if (blkh_level(blkh) != DISK_LEAF_NODE_LEVEL) {
 413                 reiserfs_warning(NULL, "reiserfs-5080",
 414                                  "this should be caught earlier");
 415                 return 0;
 416         }
 417
 418         nr = blkh_nr_item(blkh);
 419         if (nr < 1 || nr > ((blocksize - BLKH_SIZE) / (IH_SIZE + MIN_ITEM_LEN))) {
 420                 /* item number is too big or too small */
 421                 reiserfs_warning(NULL, "reiserfs-5081",
 422                                  "nr_item seems wrong: %z", bh);
 423                 return 0;
 424         }
 425         ih = (struct item_head *)(buf + BLKH_SIZE) + nr - 1;
 426         used_space = BLKH_SIZE + IH_SIZE * nr + (blocksize - ih_location(ih));
 427         if (used_space != blocksize - blkh_free_space(blkh)) {
 428                 /* free space does not match to calculated amount of use space */
 429                 reiserfs_warning(NULL, "reiserfs-5082",
 430                                  "free space seems wrong: %z", bh);
 431                 return 0;
 432         }
 433         // FIXME: it is_leaf will hit performance too much - we may have
 434         // return 1 here
 435
 436         /* check tables of item heads */
 437         ih = (struct item_head *)(buf + BLKH_SIZE);
 438         prev_location = blocksize;
 439         for (i = 0; i < nr; i++, ih++) {
 440                 if (le_ih_k_type(ih) == TYPE_ANY) {
 441                         reiserfs_warning(NULL, "reiserfs-5083",
 442                                          "wrong item type for item %h",
 443                                          ih);
 444                         return 0;
 445                 }
 446                 if (ih_location(ih) >= blocksize
 447                     || ih_location(ih) < IH_SIZE * nr) {
 448                         reiserfs_warning(NULL, "reiserfs-5084",
 449                                          "item location seems wrong: %h",
 450                                          ih);
 451                         return 0;
 452                 }
 453                 if (ih_item_len(ih) < 1
 454                     || ih_item_len(ih) > MAX_ITEM_LEN(blocksize)) {
 455                         reiserfs_warning(NULL, "reiserfs-5085",
 456                                          "item length seems wrong: %h",
 457                                          ih);
 458                         return 0;
 459                 }
 460                 if (prev_location - ih_location(ih) != ih_item_len(ih)) {
 461                         reiserfs_warning(NULL, "reiserfs-5086",
 462                                          "item location seems wrong "
 463                                          "(second one): %h", ih);
 464                         return 0;
 465                 }
 466                 prev_location = ih_location(ih);
 467         }
 468
 469         // one may imagine much more checks
 470         return 1;
 471 }
 472
 473 /* returns 1 if buf looks like an internal node, 0 otherwise */
 474 static int is_internal(char *buf, int blocksize, struct buffer_head *bh)
 475 {
 476         struct block_head *blkh;
 477         int nr;
 478         int used_space;
 479
 480         blkh = (struct block_head *)buf;
 481         nr = blkh_level(blkh);
 482         if (nr <= DISK_LEAF_NODE_LEVEL || nr > MAX_HEIGHT) {
 483                 /* this level is not possible for internal nodes */
 484                 reiserfs_warning(NULL, "reiserfs-5087",
 485                                  "this should be caught earlier");
 486                 return 0;
 487         }
 488
 489         nr = blkh_nr_item(blkh);
 490         if (nr > (blocksize - BLKH_SIZE - DC_SIZE) / (KEY_SIZE + DC_SIZE)) {
 491                 /* for internal which is not root we might check min number of keys */
 492                 reiserfs_warning(NULL, "reiserfs-5088",
 493                                  "number of key seems wrong: %z", bh);
 494                 return 0;
 495         }
 496
 497         used_space = BLKH_SIZE + KEY_SIZE * nr + DC_SIZE * (nr + 1);
 498         if (used_space != blocksize - blkh_free_space(blkh)) {
 499                 reiserfs_warning(NULL, "reiserfs-5089",
 500                                  "free space seems wrong: %z", bh);
 501                 return 0;
 502         }
 503         // one may imagine much more checks
 504         return 1;
 505 }
 506
 507 // make sure that bh contains formatted node of reiserfs tree of
 508 // 'level'-th level
 509 static int is_tree_node(struct buffer_head *bh, int level)
 510 {
 511         if (B_LEVEL(bh) != level) {
 512                 reiserfs_warning(NULL, "reiserfs-5090", "node level %d does "
 513                                  "not match to the expected one %d",
 514                                  B_LEVEL(bh), level);
 515                 return 0;
 516         }
 517         if (level == DISK_LEAF_NODE_LEVEL)
 518                 return is_leaf(bh->b_data, bh->b_size, bh);
 519
 520         return is_internal(bh->b_data, bh->b_size, bh);
 521 }
 522
 523 #define SEARCH_BY_KEY_READA 16
 524
 525 /* The function is NOT SCHEDULE-SAFE! */
 526 static void search_by_key_reada(struct super_block *s,
 527                                 struct buffer_head **bh,
 528                                 b_blocknr_t *b, int num)
 529 {
 530         int i, j;
 531
 532         for (i = 0; i < num; i++) {
 533                 bh[i] = sb_getblk(s, b[i]);
 534         }
 535         for (j = 0; j < i; j++) {
 536                 /*
 537                  * note, this needs attention if we are getting rid of the BKL
 538                  * you have to make sure the prepared bit isn't set on this buffer
 539                  */
 540                 if (!buffer_uptodate(bh[j]))
 541                         ll_rw_block(READA, 1, bh + j);
 542                 brelse(bh[j]);
 543         }
 544 }
 545
 546 /**************************************************************************
 547  * Algorithm   SearchByKey                                                *
 548  *             look for item in the Disk S+Tree by its key                *
 549  * Input:  sb   -  super block                                            *
 550  *         p_s_key  - pointer to the key to search                        *
 551  * Output: ITEM_FOUND, ITEM_NOT_FOUND or IO_ERROR                         *
 552  *         p_s_search_path - path from the root to the needed leaf        *
 553  **************************************************************************/
 554
 555 /* This function fills up the path from the root to the leaf as it
 556    descends the tree looking for the key.  It uses reiserfs_bread to
 557    try to find buffers in the cache given their block number.  If it
 558    does not find them in the cache it reads them from disk.  For each
 559    node search_by_key finds using reiserfs_bread it then uses
 560    bin_search to look through that node.  bin_search will find the
 561    position of the block_number of the next node if it is looking
 562    through an internal node.  If it is looking through a leaf node
 563    bin_search will find the position of the item which has key either
 564    equal to given key, or which is the maximal key less than the given
 565    key.  search_by_key returns a path that must be checked for the
 566    correctness of the top of the path but need not be checked for the
 567    correctness of the bottom of the path */
 568 /* The function is NOT SCHEDULE-SAFE! */
 569 int search_by_key(struct super_block *sb, const struct cpu_key *p_s_key,        /* Key to search. */
 570                   struct treepath *p_s_search_path,/* This structure was
 571                                                    allocated and initialized
 572                                                    by the calling
 573                                                    function. It is filled up
 574                                                    by this function.  */
 575                   int n_stop_level      /* How far down the tree to search. To
 576                                            stop at leaf level - set to
 577                                            DISK_LEAF_NODE_LEVEL */
 578     )
 579 {
 580         b_blocknr_t n_block_number;
 581         int expected_level;
 582         struct buffer_head *p_s_bh;
 583         struct path_element *p_s_last_element;
 584         int n_node_level, n_retval;
 585         int right_neighbor_of_leaf_node;
 586         int fs_gen;
 587         struct buffer_head *reada_bh[SEARCH_BY_KEY_READA];
 588         b_blocknr_t reada_blocks[SEARCH_BY_KEY_READA];
 589         int reada_count = 0;
 590
 591 #ifdef CONFIG_REISERFS_CHECK
 592         int n_repeat_counter = 0;
 593 #endif
 594
 595         PROC_INFO_INC(sb, search_by_key);
 596
 597         /* As we add each node to a path we increase its count.  This means that
 598            we must be careful to release all nodes in a path before we either
 599            discard the path struct or re-use the path struct, as we do here. */
 600
 601         pathrelse(p_s_search_path);
 602
 603         right_neighbor_of_leaf_node = 0;
 604
 605         /* With each iteration of this loop we search through the items in the
 606            current node, and calculate the next current node(next path element)
 607            for the next iteration of this loop.. */
 608         n_block_number = SB_ROOT_BLOCK(sb);
 609         expected_level = -1;
 610         while (1) {
 611
 612 #ifdef CONFIG_REISERFS_CHECK
 613                 if (!(++n_repeat_counter % 50000))
 614                         reiserfs_warning(sb, "PAP-5100",
 615                                          "%s: there were %d iterations of "
 616                                          "while loop looking for key %K",
 617                                          current->comm, n_repeat_counter,
 618                                          p_s_key);
 619 #endif
 620
 621                 /* prep path to have another element added to it. */
 622                 p_s_last_element =
 623                     PATH_OFFSET_PELEMENT(p_s_search_path,
 624                                          ++p_s_search_path->path_length);
 625                 fs_gen = get_generation(sb);
 626
 627                 /* Read the next tree node, and set the last element in the path to
 628                    have a pointer to it. */
 629                 if ((p_s_bh = p_s_last_element->pe_buffer =
 630                      sb_getblk(sb, n_block_number))) {
 631                         if (!buffer_uptodate(p_s_bh) && reada_count > 1) {
 632                                 search_by_key_reada(sb, reada_bh,
 633                                                     reada_blocks, reada_count);
 634                         }
 635                         ll_rw_block(READ, 1, &p_s_bh);
 636                         wait_on_buffer(p_s_bh);
 637                         if (!buffer_uptodate(p_s_bh))
 638                                 goto io_error;
 639                 } else {
 640                       io_error:
 641                         p_s_search_path->path_length--;
 642                         pathrelse(p_s_search_path);
 643                         return IO_ERROR;
 644                 }
 645                 reada_count = 0;
 646                 if (expected_level == -1)
 647                         expected_level = SB_TREE_HEIGHT(sb);
 648                 expected_level--;
 649
 650                 /* It is possible that schedule occurred. We must check whether the key
 651                    to search is still in the tree rooted from the current buffer. If
 652                    not then repeat search from the root. */
 653                 if (fs_changed(fs_gen, sb) &&
 654                     (!B_IS_IN_TREE(p_s_bh) ||
 655                      B_LEVEL(p_s_bh) != expected_level ||
 656                      !key_in_buffer(p_s_search_path, p_s_key, sb))) {
 657                         PROC_INFO_INC(sb, search_by_key_fs_changed);
 658                         PROC_INFO_INC(sb, search_by_key_restarted);
 659                         PROC_INFO_INC(sb,
 660                                       sbk_restarted[expected_level - 1]);
 661                         pathrelse(p_s_search_path);
 662
 663                         /* Get the root block number so that we can repeat the search
 664                            starting from the root. */
 665                         n_block_number = SB_ROOT_BLOCK(sb);
 666                         expected_level = -1;
 667                         right_neighbor_of_leaf_node = 0;
 668
 669                         /* repeat search from the root */
 670                         continue;
 671                 }
 672
 673                 /* only check that the key is in the buffer if p_s_key is not
 674                    equal to the MAX_KEY. Latter case is only possible in
 675                    "finish_unfinished()" processing during mount. */
 676                 RFALSE(comp_keys(&MAX_KEY, p_s_key) &&
 677                        !key_in_buffer(p_s_search_path, p_s_key, sb),
 678                        "PAP-5130: key is not in the buffer");
 679 #ifdef CONFIG_REISERFS_CHECK
 680                 if (cur_tb) {
 681                         print_cur_tb("5140");
 682                         reiserfs_panic(sb, "PAP-5140",
 683                                        "schedule occurred in do_balance!");
 684                 }
 685 #endif
 686
 687                 // make sure, that the node contents look like a node of
 688                 // certain level
 689                 if (!is_tree_node(p_s_bh, expected_level)) {
 690                         reiserfs_error(sb, "vs-5150",
 691                                        "invalid format found in block %ld. "
 692                                        "Fsck?", p_s_bh->b_blocknr);
 693                         pathrelse(p_s_search_path);
 694                         return IO_ERROR;
 695                 }
 696
 697                 /* ok, we have acquired next formatted node in the tree */
 698                 n_node_level = B_LEVEL(p_s_bh);
 699
 700                 PROC_INFO_BH_STAT(sb, p_s_bh, n_node_level - 1);
 701
 702                 RFALSE(n_node_level < n_stop_level,
 703                        "vs-5152: tree level (%d) is less than stop level (%d)",
 704                        n_node_level, n_stop_level);
 705
 706                 n_retval = bin_search(p_s_key, B_N_PITEM_HEAD(p_s_bh, 0),
 707                                       B_NR_ITEMS(p_s_bh),
 708                                       (n_node_level ==
 709                                        DISK_LEAF_NODE_LEVEL) ? IH_SIZE :
 710                                       KEY_SIZE,
 711                                       &(p_s_last_element->pe_position));
 712                 if (n_node_level == n_stop_level) {
 713                         return n_retval;
 714                 }
 715
 716                 /* we are not in the stop level */
 717                 if (n_retval == ITEM_FOUND)
 718                         /* item has been found, so we choose the pointer which is to the right of the found one */
 719                         p_s_last_element->pe_position++;
 720
 721                 /* if item was not found we choose the position which is to
 722                    the left of the found item. This requires no code,
 723                    bin_search did it already. */
 724
 725                 /* So we have chosen a position in the current node which is
 726                    an internal node.  Now we calculate child block number by
 727                    position in the node. */
 728                 n_block_number =
 729                     B_N_CHILD_NUM(p_s_bh, p_s_last_element->pe_position);
 730
 731                 /* if we are going to read leaf nodes, try for read ahead as well */
 732                 if ((p_s_search_path->reada & PATH_READA) &&
 733                     n_node_level == DISK_LEAF_NODE_LEVEL + 1) {
 734                         int pos = p_s_last_element->pe_position;
 735                         int limit = B_NR_ITEMS(p_s_bh);
 736                         struct reiserfs_key *le_key;
 737
 738                         if (p_s_search_path->reada & PATH_READA_BACK)
 739                                 limit = 0;
 740                         while (reada_count < SEARCH_BY_KEY_READA) {
 741                                 if (pos == limit)
 742                                         break;
 743                                 reada_blocks[reada_count++] =
 744                                     B_N_CHILD_NUM(p_s_bh, pos);
 745                                 if (p_s_search_path->reada & PATH_READA_BACK)
 746                                         pos--;
 747                                 else
 748                                         pos++;
 749
 750                                 /*
 751                                  * check to make sure we're in the same object
 752                                  */
 753                                 le_key = B_N_PDELIM_KEY(p_s_bh, pos);
 754                                 if (le32_to_cpu(le_key->k_objectid) !=
 755                                     p_s_key->on_disk_key.k_objectid) {
 756                                         break;
 757                                 }
 758                         }
 759                 }
 760         }
 761 }
 762
 763 /* Form the path to an item and position in this item which contains
 764    file byte defined by p_s_key. If there is no such item
 765    corresponding to the key, we point the path to the item with
 766    maximal key less than p_s_key, and *p_n_pos_in_item is set to one
 767    past the last entry/byte in the item.  If searching for entry in a
 768    directory item, and it is not found, *p_n_pos_in_item is set to one
 769    entry more than the entry with maximal key which is less than the
 770    sought key.
 771
 772    Note that if there is no entry in this same node which is one more,
 773    then we point to an imaginary entry.  for direct items, the
 774    position is in units of bytes, for indirect items the position is
 775    in units of blocknr entries, for directory items the position is in
 776    units of directory entries.  */
 777
 778 /* The function is NOT SCHEDULE-SAFE! */
 779 int search_for_position_by_key(struct super_block *sb,  /* Pointer to the super block.          */
 780                                const struct cpu_key *p_cpu_key, /* Key to search (cpu variable)         */
 781                                struct treepath *p_s_search_path /* Filled up by this function.          */
 782     )
 783 {
 784         struct item_head *p_le_ih;      /* pointer to on-disk structure */
 785         int n_blk_size;
 786         loff_t item_offset, offset;
 787         struct reiserfs_dir_entry de;
 788         int retval;
 789
 790         /* If searching for directory entry. */
 791         if (is_direntry_cpu_key(p_cpu_key))
 792                 return search_by_entry_key(sb, p_cpu_key, p_s_search_path,
 793                                            &de);
 794
 795         /* If not searching for directory entry. */
 796
 797         /* If item is found. */
 798         retval = search_item(sb, p_cpu_key, p_s_search_path);
 799         if (retval == IO_ERROR)
 800                 return retval;
 801         if (retval == ITEM_FOUND) {
 802
 803                 RFALSE(!ih_item_len
 804                        (B_N_PITEM_HEAD
 805                         (PATH_PLAST_BUFFER(p_s_search_path),
 806                          PATH_LAST_POSITION(p_s_search_path))),
 807                        "PAP-5165: item length equals zero");
 808
 809                 pos_in_item(p_s_search_path) = 0;
 810                 return POSITION_FOUND;
 811         }
 812
 813         RFALSE(!PATH_LAST_POSITION(p_s_search_path),
 814                "PAP-5170: position equals zero");
 815
 816         /* Item is not found. Set path to the previous item. */
 817         p_le_ih =
 818             B_N_PITEM_HEAD(PATH_PLAST_BUFFER(p_s_search_path),
 819                            --PATH_LAST_POSITION(p_s_search_path));
 820         n_blk_size = sb->s_blocksize;
 821
 822         if (comp_short_keys(&(p_le_ih->ih_key), p_cpu_key)) {
 823                 return FILE_NOT_FOUND;
 824         }
 825         // FIXME: quite ugly this far
 826
 827         item_offset = le_ih_k_offset(p_le_ih);
 828         offset = cpu_key_k_offset(p_cpu_key);
 829
 830         /* Needed byte is contained in the item pointed to by the path. */
 831         if (item_offset <= offset &&
 832             item_offset + op_bytes_number(p_le_ih, n_blk_size) > offset) {
 833                 pos_in_item(p_s_search_path) = offset - item_offset;
 834                 if (is_indirect_le_ih(p_le_ih)) {
 835                         pos_in_item(p_s_search_path) /= n_blk_size;
 836                 }
 837                 return POSITION_FOUND;
 838         }
 839
 840         /* Needed byte is not contained in the item pointed to by the
 841            path. Set pos_in_item out of the item. */
 842         if (is_indirect_le_ih(p_le_ih))
 843                 pos_in_item(p_s_search_path) =
 844                     ih_item_len(p_le_ih) / UNFM_P_SIZE;
 845         else
 846                 pos_in_item(p_s_search_path) = ih_item_len(p_le_ih);
 847
 848         return POSITION_NOT_FOUND;
 849 }
 850
 851 /* Compare given item and item pointed to by the path. */
 852 int comp_items(const struct item_head *stored_ih, const struct treepath *p_s_path)
 853 {
 854         struct buffer_head *p_s_bh;
 855         struct item_head *ih;
 856
 857         /* Last buffer at the path is not in the tree. */
 858         if (!B_IS_IN_TREE(p_s_bh = PATH_PLAST_BUFFER(p_s_path)))
 859                 return 1;
 860
 861         /* Last path position is invalid. */
 862         if (PATH_LAST_POSITION(p_s_path) >= B_NR_ITEMS(p_s_bh))
 863                 return 1;
 864
 865         /* we need only to know, whether it is the same item */
 866         ih = get_ih(p_s_path);
 867         return memcmp(stored_ih, ih, IH_SIZE);
 868 }
 869
 870 /* unformatted nodes are not logged anymore, ever.  This is safe
 871 ** now
 872 */
 873 #define held_by_others(bh) (atomic_read(&(bh)->b_count) > 1)
 874
 875 // block can not be forgotten as it is in I/O or held by someone
 876 #define block_in_use(bh) (buffer_locked(bh) || (held_by_others(bh)))
 877
 878 // prepare for delete or cut of direct item
 879 static inline int prepare_for_direct_item(struct treepath *path,
 880                                           struct item_head *le_ih,
 881                                           struct inode *inode,
 882                                           loff_t new_file_length, int *cut_size)
 883 {
 884         loff_t round_len;
 885
 886         if (new_file_length == max_reiserfs_offset(inode)) {
 887                 /* item has to be deleted */
 888                 *cut_size = -(IH_SIZE + ih_item_len(le_ih));
 889                 return M_DELETE;
 890         }
 891         // new file gets truncated
 892         if (get_inode_item_key_version(inode) == KEY_FORMAT_3_6) {
 893                 //
 894                 round_len = ROUND_UP(new_file_length);
 895                 /* this was n_new_file_length < le_ih ... */
 896                 if (round_len < le_ih_k_offset(le_ih)) {
 897                         *cut_size = -(IH_SIZE + ih_item_len(le_ih));
 898                         return M_DELETE;        /* Delete this item. */
 899                 }
 900                 /* Calculate first position and size for cutting from item. */
 901                 pos_in_item(path) = round_len - (le_ih_k_offset(le_ih) - 1);
 902                 *cut_size = -(ih_item_len(le_ih) - pos_in_item(path));
 903
 904                 return M_CUT;   /* Cut from this item. */
 905         }
 906
 907         // old file: items may have any length
 908
 909         if (new_file_length < le_ih_k_offset(le_ih)) {
 910                 *cut_size = -(IH_SIZE + ih_item_len(le_ih));
 911                 return M_DELETE;        /* Delete this item. */
 912         }
 913         /* Calculate first position and size for cutting from item. */
 914         *cut_size = -(ih_item_len(le_ih) -
 915                       (pos_in_item(path) =
 916                        new_file_length + 1 - le_ih_k_offset(le_ih)));
 917         return M_CUT;           /* Cut from this item. */
 918 }
 919
 920 static inline int prepare_for_direntry_item(struct treepath *path,
 921                                             struct item_head *le_ih,
 922                                             struct inode *inode,
 923                                             loff_t new_file_length,
 924                                             int *cut_size)
 925 {
 926         if (le_ih_k_offset(le_ih) == DOT_OFFSET &&
 927             new_file_length == max_reiserfs_offset(inode)) {
 928                 RFALSE(ih_entry_count(le_ih) != 2,
 929                        "PAP-5220: incorrect empty directory item (%h)", le_ih);
 930                 *cut_size = -(IH_SIZE + ih_item_len(le_ih));
 931                 return M_DELETE;        /* Delete the directory item containing "." and ".." entry. */
 932         }
 933
 934         if (ih_entry_count(le_ih) == 1) {
 935                 /* Delete the directory item such as there is one record only
 936                    in this item */
 937                 *cut_size = -(IH_SIZE + ih_item_len(le_ih));
 938                 return M_DELETE;
 939         }
 940
 941         /* Cut one record from the directory item. */
 942         *cut_size =
 943             -(DEH_SIZE +
 944               entry_length(get_last_bh(path), le_ih, pos_in_item(path)));
 945         return M_CUT;
 946 }
 947
 948 #define JOURNAL_FOR_FREE_BLOCK_AND_UPDATE_SD (2 * JOURNAL_PER_BALANCE_CNT + 1)
 949
 950 /*  If the path points to a directory or direct item, calculate mode and the size cut, for balance.
 951     If the path points to an indirect item, remove some number of its unformatted nodes.
 952     In case of file truncate calculate whether this item must be deleted/truncated or last
 953     unformatted node of this item will be converted to a direct item.
 954     This function returns a determination of what balance mode the calling function should employ. */
 955 static char prepare_for_delete_or_cut(struct reiserfs_transaction_handle *th, struct inode *inode, struct treepath *p_s_path, const struct cpu_key *p_s_item_key, int *p_n_removed,     /* Number of unformatted nodes which were removed
 956                                                                                                                                                                                    from end of the file. */
 957                                       int *p_n_cut_size, unsigned long long n_new_file_length   /* MAX_KEY_OFFSET in case of delete. */
 958     )
 959 {
 960         struct super_block *sb = inode->i_sb;
 961         struct item_head *p_le_ih = PATH_PITEM_HEAD(p_s_path);
 962         struct buffer_head *p_s_bh = PATH_PLAST_BUFFER(p_s_path);
 963
 964         BUG_ON(!th->t_trans_id);
 965
 966         /* Stat_data item. */
 967         if (is_statdata_le_ih(p_le_ih)) {
 968
 969                 RFALSE(n_new_file_length != max_reiserfs_offset(inode),
 970                        "PAP-5210: mode must be M_DELETE");
 971
 972                 *p_n_cut_size = -(IH_SIZE + ih_item_len(p_le_ih));
 973                 return M_DELETE;
 974         }
 975
 976         /* Directory item. */
 977         if (is_direntry_le_ih(p_le_ih))
 978                 return prepare_for_direntry_item(p_s_path, p_le_ih, inode,
 979                                                  n_new_file_length,
 980                                                  p_n_cut_size);
 981
 982         /* Direct item. */
 983         if (is_direct_le_ih(p_le_ih))
 984                 return prepare_for_direct_item(p_s_path, p_le_ih, inode,
 985                                                n_new_file_length, p_n_cut_size);
 986
 987         /* Case of an indirect item. */
 988         {
 989             int blk_size = sb->s_blocksize;
 990             struct item_head s_ih;
 991             int need_re_search;
 992             int delete = 0;
 993             int result = M_CUT;
 994             int pos = 0;
 995
 996             if ( n_new_file_length == max_reiserfs_offset (inode) ) {
 997                 /* prepare_for_delete_or_cut() is called by
 998                  * reiserfs_delete_item() */
 999                 n_new_file_length = 0;
1000                 delete = 1;
1001             }
1002
1003             do {
1004                 need_re_search = 0;
1005                 *p_n_cut_size = 0;
1006                 p_s_bh = PATH_PLAST_BUFFER(p_s_path);
1007                 copy_item_head(&s_ih, PATH_PITEM_HEAD(p_s_path));
1008                 pos = I_UNFM_NUM(&s_ih);
1009
1010                 while (le_ih_k_offset (&s_ih) + (pos - 1) * blk_size > n_new_file_length) {
1011                     __le32 *unfm;
1012                     __u32 block;
1013
1014                     /* Each unformatted block deletion may involve one additional
1015                      * bitmap block into the transaction, thereby the initial
1016                      * journal space reservation might not be enough. */
1017                     if (!delete && (*p_n_cut_size) != 0 &&
1018                         reiserfs_transaction_free_space(th) < JOURNAL_FOR_FREE_BLOCK_AND_UPDATE_SD) {
1019                         break;
1020                     }
1021
1022                     unfm = (__le32 *)B_I_PITEM(p_s_bh, &s_ih) + pos - 1;
1023                     block = get_block_num(unfm, 0);
1024
1025                     if (block != 0) {
1026                         reiserfs_prepare_for_journal(sb, p_s_bh, 1);
1027                         put_block_num(unfm, 0, 0);
1028                         journal_mark_dirty (th, sb, p_s_bh);
1029                         reiserfs_free_block(th, inode, block, 1);
1030                     }
1031
1032                     cond_resched();
1033
1034                     if (item_moved (&s_ih, p_s_path))  {
1035                         need_re_search = 1;
1036                         break;
1037                     }
1038
1039                     pos --;
1040                     (*p_n_removed) ++;
1041                     (*p_n_cut_size) -= UNFM_P_SIZE;
1042
1043                     if (pos == 0) {
1044                         (*p_n_cut_size) -= IH_SIZE;
1045                         result = M_DELETE;
1046                         break;
1047                     }
1048                 }
1049                 /* a trick.  If the buffer has been logged, this will do nothing.  If
1050                 ** we've broken the loop without logging it, it will restore the
1051                 ** buffer */
1052                 reiserfs_restore_prepared_buffer(sb, p_s_bh);
1053             } while (need_re_search &&
1054                      search_for_position_by_key(sb, p_s_item_key, p_s_path) == POSITION_FOUND);
1055             pos_in_item(p_s_path) = pos * UNFM_P_SIZE;
1056
1057             if (*p_n_cut_size == 0) {
1058                 /* Nothing were cut. maybe convert last unformatted node to the
1059                  * direct item? */
1060                 result = M_CONVERT;
1061             }
1062             return result;
1063         }
1064 }
1065
1066 /* Calculate number of bytes which will be deleted or cut during balance */
1067 static int calc_deleted_bytes_number(struct tree_balance *p_s_tb, char c_mode)
1068 {
1069         int n_del_size;
1070         struct item_head *p_le_ih = PATH_PITEM_HEAD(p_s_tb->tb_path);
1071
1072         if (is_statdata_le_ih(p_le_ih))
1073                 return 0;
1074
1075         n_del_size =
1076             (c_mode ==
1077              M_DELETE) ? ih_item_len(p_le_ih) : -p_s_tb->insert_size[0];
1078         if (is_direntry_le_ih(p_le_ih)) {
1079                 // return EMPTY_DIR_SIZE; /* We delete emty directoris only. */
1080                 // we can't use EMPTY_DIR_SIZE, as old format dirs have a different
1081                 // empty size.  ick. FIXME, is this right?
1082                 //
1083                 return n_del_size;
1084         }
1085
1086         if (is_indirect_le_ih(p_le_ih))
1087                 n_del_size = (n_del_size / UNFM_P_SIZE) * (PATH_PLAST_BUFFER(p_s_tb->tb_path)->b_size); // - get_ih_free_space (p_le_ih);
1088         return n_del_size;
1089 }
1090
1091 static void init_tb_struct(struct reiserfs_transaction_handle *th,
1092                            struct tree_balance *p_s_tb,
1093                            struct super_block *sb,
1094                            struct treepath *p_s_path, int n_size)
1095 {
1096
1097         BUG_ON(!th->t_trans_id);
1098
1099         memset(p_s_tb, '\0', sizeof(struct tree_balance));
1100         p_s_tb->transaction_handle = th;
1101         p_s_tb->tb_sb = sb;
1102         p_s_tb->tb_path = p_s_path;
1103         PATH_OFFSET_PBUFFER(p_s_path, ILLEGAL_PATH_ELEMENT_OFFSET) = NULL;
1104         PATH_OFFSET_POSITION(p_s_path, ILLEGAL_PATH_ELEMENT_OFFSET) = 0;
1105         p_s_tb->insert_size[0] = n_size;
1106 }
1107
1108 void padd_item(char *item, int total_length, int length)
1109 {
1110         int i;
1111
1112         for (i = total_length; i > length;)
1113                 item[--i] = 0;
1114 }
1115
1116 #ifdef REISERQUOTA_DEBUG
1117 char key2type(struct reiserfs_key *ih)
1118 {
1119         if (is_direntry_le_key(2, ih))
1120                 return 'd';
1121         if (is_direct_le_key(2, ih))
1122                 return 'D';
1123         if (is_indirect_le_key(2, ih))
1124                 return 'i';
1125         if (is_statdata_le_key(2, ih))
1126                 return 's';
1127         return 'u';
1128 }
1129
1130 char head2type(struct item_head *ih)
1131 {
1132         if (is_direntry_le_ih(ih))
1133                 return 'd';
1134         if (is_direct_le_ih(ih))
1135                 return 'D';
1136         if (is_indirect_le_ih(ih))
1137                 return 'i';
1138         if (is_statdata_le_ih(ih))
1139                 return 's';
1140         return 'u';
1141 }
1142 #endif
1143
1144 /* Delete object item. */
1145 int reiserfs_delete_item(struct reiserfs_transaction_handle *th, struct treepath *p_s_path,     /* Path to the deleted item. */
1146                          const struct cpu_key *p_s_item_key,    /* Key to search for the deleted item.  */
1147                          struct inode *p_s_inode,       /* inode is here just to update i_blocks and quotas */
1148                          struct buffer_head *p_s_un_bh)
1149 {                               /* NULL or unformatted node pointer.    */
1150         struct super_block *sb = p_s_inode->i_sb;
1151         struct tree_balance s_del_balance;
1152         struct item_head s_ih;
1153         struct item_head *q_ih;
1154         int quota_cut_bytes;
1155         int n_ret_value, n_del_size, n_removed;
1156
1157 #ifdef CONFIG_REISERFS_CHECK
1158         char c_mode;
1159         int n_iter = 0;
1160 #endif
1161
1162         BUG_ON(!th->t_trans_id);
1163
1164         init_tb_struct(th, &s_del_balance, sb, p_s_path,
1165                        0 /*size is unknown */ );
1166
1167         while (1) {
1168                 n_removed = 0;
1169
1170 #ifdef CONFIG_REISERFS_CHECK
1171                 n_iter++;
1172                 c_mode =
1173 #endif
1174                     prepare_for_delete_or_cut(th, p_s_inode, p_s_path,
1175                                               p_s_item_key, &n_removed,
1176                                               &n_del_size,
1177                                               max_reiserfs_offset(p_s_inode));
1178
1179                 RFALSE(c_mode != M_DELETE, "PAP-5320: mode must be M_DELETE");
1180
1181                 copy_item_head(&s_ih, PATH_PITEM_HEAD(p_s_path));
1182                 s_del_balance.insert_size[0] = n_del_size;
1183
1184                 n_ret_value = fix_nodes(M_DELETE, &s_del_balance, NULL, NULL);
1185                 if (n_ret_value != REPEAT_SEARCH)
1186                         break;
1187
1188                 PROC_INFO_INC(sb, delete_item_restarted);
1189
1190                 // file system changed, repeat search
1191                 n_ret_value =
1192                     search_for_position_by_key(sb, p_s_item_key, p_s_path);
1193                 if (n_ret_value == IO_ERROR)
1194                         break;
1195                 if (n_ret_value == FILE_NOT_FOUND) {
1196                         reiserfs_warning(sb, "vs-5340",
1197                                          "no items of the file %K found",
1198                                          p_s_item_key);
1199                         break;
1200                 }
1201         }                       /* while (1) */
1202
1203         if (n_ret_value != CARRY_ON) {
1204                 unfix_nodes(&s_del_balance);
1205                 return 0;
1206         }
1207         // reiserfs_delete_item returns item length when success
1208         n_ret_value = calc_deleted_bytes_number(&s_del_balance, M_DELETE);
1209         q_ih = get_ih(p_s_path);
1210         quota_cut_bytes = ih_item_len(q_ih);
1211
1212         /* hack so the quota code doesn't have to guess if the file
1213          ** has a tail.  On tail insert, we allocate quota for 1 unformatted node.
1214          ** We test the offset because the tail might have been
1215          ** split into multiple items, and we only want to decrement for
1216          ** the unfm node once
1217          */
1218         if (!S_ISLNK(p_s_inode->i_mode) && is_direct_le_ih(q_ih)) {
1219                 if ((le_ih_k_offset(q_ih) & (sb->s_blocksize - 1)) == 1) {
1220                         quota_cut_bytes = sb->s_blocksize + UNFM_P_SIZE;
1221                 } else {
1222                         quota_cut_bytes = 0;
1223                 }
1224         }
1225
1226         if (p_s_un_bh) {
1227                 int off;
1228                 char *data;
1229
1230                 /* We are in direct2indirect conversion, so move tail contents
1231                    to the unformatted node */
1232                 /* note, we do the copy before preparing the buffer because we
1233                  ** don't care about the contents of the unformatted node yet.
1234                  ** the only thing we really care about is the direct item's data
1235                  ** is in the unformatted node.
1236                  **
1237                  ** Otherwise, we would have to call reiserfs_prepare_for_journal on
1238                  ** the unformatted node, which might schedule, meaning we'd have to
1239                  ** loop all the way back up to the start of the while loop.
1240                  **
1241                  ** The unformatted node must be dirtied later on.  We can't be
1242                  ** sure here if the entire tail has been deleted yet.
1243                  **
1244                  ** p_s_un_bh is from the page cache (all unformatted nodes are
1245                  ** from the page cache) and might be a highmem page.  So, we
1246                  ** can't use p_s_un_bh->b_data.
1247                  ** -clm
1248                  */
1249
1250                 data = kmap_atomic(p_s_un_bh->b_page, KM_USER0);
1251                 off = ((le_ih_k_offset(&s_ih) - 1) & (PAGE_CACHE_SIZE - 1));
1252                 memcpy(data + off,
1253                        B_I_PITEM(PATH_PLAST_BUFFER(p_s_path), &s_ih),
1254                        n_ret_value);
1255                 kunmap_atomic(data, KM_USER0);
1256         }
1257         /* Perform balancing after all resources have been collected at once. */
1258         do_balance(&s_del_balance, NULL, NULL, M_DELETE);
1259
1260 #ifdef REISERQUOTA_DEBUG
1261         reiserfs_debug(sb, REISERFS_DEBUG_CODE,
1262                        "reiserquota delete_item(): freeing %u, id=%u type=%c",
1263                        quota_cut_bytes, p_s_inode->i_uid, head2type(&s_ih));
1264 #endif
1265         DQUOT_FREE_SPACE_NODIRTY(p_s_inode, quota_cut_bytes);
1266
1267         /* Return deleted body length */
1268         return n_ret_value;
1269 }
1270
1271 /* Summary Of Mechanisms For Handling Collisions Between Processes:
1272
1273  deletion of the body of the object is performed by iput(), with the
1274  result that if multiple processes are operating on a file, the
1275  deletion of the body of the file is deferred until the last process
1276  that has an open inode performs its iput().
1277
1278  writes and truncates are protected from collisions by use of
1279  semaphores.
1280
1281  creates, linking, and mknod are protected from collisions with other
1282  processes by making the reiserfs_add_entry() the last step in the
1283  creation, and then rolling back all changes if there was a collision.
1284  - Hans
1285 */
1286
1287 /* this deletes item which never gets split */
1288 void reiserfs_delete_solid_item(struct reiserfs_transaction_handle *th,
1289                                 struct inode *inode, struct reiserfs_key *key)
1290 {
1291         struct tree_balance tb;
1292         INITIALIZE_PATH(path);
1293         int item_len = 0;
1294         int tb_init = 0;
1295         struct cpu_key cpu_key;
1296         int retval;
1297         int quota_cut_bytes = 0;
1298
1299         BUG_ON(!th->t_trans_id);
1300
1301         le_key2cpu_key(&cpu_key, key);
1302
1303         while (1) {
1304                 retval = search_item(th->t_super, &cpu_key, &path);
1305                 if (retval == IO_ERROR) {
1306                         reiserfs_error(th->t_super, "vs-5350",
1307                                        "i/o failure occurred trying "
1308                                        "to delete %K", &cpu_key);
1309                         break;
1310                 }
1311                 if (retval != ITEM_FOUND) {
1312                         pathrelse(&path);
1313                         // No need for a warning, if there is just no free space to insert '..' item into the newly-created subdir
1314                         if (!
1315                             ((unsigned long long)
1316                              GET_HASH_VALUE(le_key_k_offset
1317                                             (le_key_version(key), key)) == 0
1318                              && (unsigned long long)
1319                              GET_GENERATION_NUMBER(le_key_k_offset
1320                                                    (le_key_version(key),
1321                                                     key)) == 1))
1322                                 reiserfs_warning(th->t_super, "vs-5355",
1323                                                  "%k not found", key);
1324                         break;
1325                 }
1326                 if (!tb_init) {
1327                         tb_init = 1;
1328                         item_len = ih_item_len(PATH_PITEM_HEAD(&path));
1329                         init_tb_struct(th, &tb, th->t_super, &path,
1330                                        -(IH_SIZE + item_len));
1331                 }
1332                 quota_cut_bytes = ih_item_len(PATH_PITEM_HEAD(&path));
1333
1334                 retval = fix_nodes(M_DELETE, &tb, NULL, NULL);
1335                 if (retval == REPEAT_SEARCH) {
1336                         PROC_INFO_INC(th->t_super, delete_solid_item_restarted);
1337                         continue;
1338                 }
1339
1340                 if (retval == CARRY_ON) {
1341                         do_balance(&tb, NULL, NULL, M_DELETE);
1342                         if (inode) {    /* Should we count quota for item? (we don't count quotas for save-links) */
1343 #ifdef REISERQUOTA_DEBUG
1344                                 reiserfs_debug(th->t_super, REISERFS_DEBUG_CODE,
1345                                                "reiserquota delete_solid_item(): freeing %u id=%u type=%c",
1346                                                quota_cut_bytes, inode->i_uid,
1347                                                key2type(key));
1348 #endif
1349                                 DQUOT_FREE_SPACE_NODIRTY(inode,
1350                                                          quota_cut_bytes);
1351                         }
1352                         break;
1353                 }
1354                 // IO_ERROR, NO_DISK_SPACE, etc
1355                 reiserfs_warning(th->t_super, "vs-5360",
1356                                  "could not delete %K due to fix_nodes failure",
1357                                  &cpu_key);
1358                 unfix_nodes(&tb);
1359                 break;
1360         }
1361
1362         reiserfs_check_path(&path);
1363 }
1364
1365 int reiserfs_delete_object(struct reiserfs_transaction_handle *th,
1366                            struct inode *inode)
1367 {
1368         int err;
1369         inode->i_size = 0;
1370         BUG_ON(!th->t_trans_id);
1371
1372         /* for directory this deletes item containing "." and ".." */
1373         err =
1374             reiserfs_do_truncate(th, inode, NULL, 0 /*no timestamp updates */ );
1375         if (err)
1376                 return err;
1377
1378 #if defined( USE_INODE_GENERATION_COUNTER )
1379         if (!old_format_only(th->t_super)) {
1380                 __le32 *inode_generation;
1381
1382                 inode_generation =
1383                     &REISERFS_SB(th->t_super)->s_rs->s_inode_generation;
1384                 le32_add_cpu(inode_generation, 1);
1385         }
1386 /* USE_INODE_GENERATION_COUNTER */
1387 #endif
1388         reiserfs_delete_solid_item(th, inode, INODE_PKEY(inode));
1389
1390         return err;
1391 }
1392
1393 static void unmap_buffers(struct page *page, loff_t pos)
1394 {
1395         struct buffer_head *bh;
1396         struct buffer_head *head;
1397         struct buffer_head *next;
1398         unsigned long tail_index;
1399         unsigned long cur_index;
1400
1401         if (page) {
1402                 if (page_has_buffers(page)) {
1403                         tail_index = pos & (PAGE_CACHE_SIZE - 1);
1404                         cur_index = 0;
1405                         head = page_buffers(page);
1406                         bh = head;
1407                         do {
1408                                 next = bh->b_this_page;
1409
1410                                 /* we want to unmap the buffers that contain the tail, and
1411                                  ** all the buffers after it (since the tail must be at the
1412                                  ** end of the file).  We don't want to unmap file data
1413                                  ** before the tail, since it might be dirty and waiting to
1414                                  ** reach disk
1415                                  */
1416                                 cur_index += bh->b_size;
1417                                 if (cur_index > tail_index) {
1418                                         reiserfs_unmap_buffer(bh);
1419                                 }
1420                                 bh = next;
1421                         } while (bh != head);
1422                 }
1423         }
1424 }
1425
1426 static int maybe_indirect_to_direct(struct reiserfs_transaction_handle *th,
1427                                     struct inode *p_s_inode,
1428                                     struct page *page,
1429                                     struct treepath *p_s_path,
1430                                     const struct cpu_key *p_s_item_key,
1431                                     loff_t n_new_file_size, char *p_c_mode)
1432 {
1433         struct super_block *sb = p_s_inode->i_sb;
1434         int n_block_size = sb->s_blocksize;
1435         int cut_bytes;
1436         BUG_ON(!th->t_trans_id);
1437         BUG_ON(n_new_file_size != p_s_inode->i_size);
1438
1439         /* the page being sent in could be NULL if there was an i/o error
1440          ** reading in the last block.  The user will hit problems trying to
1441          ** read the file, but for now we just skip the indirect2direct
1442          */
1443         if (atomic_read(&p_s_inode->i_count) > 1 ||
1444             !tail_has_to_be_packed(p_s_inode) ||
1445             !page || (REISERFS_I(p_s_inode)->i_flags & i_nopack_mask)) {
1446                 /* leave tail in an unformatted node */
1447                 *p_c_mode = M_SKIP_BALANCING;
1448                 cut_bytes =
1449                     n_block_size - (n_new_file_size & (n_block_size - 1));
1450                 pathrelse(p_s_path);
1451                 return cut_bytes;
1452         }
1453         /* Permorm the conversion to a direct_item. */
1454         /*return indirect_to_direct (p_s_inode, p_s_path, p_s_item_key, n_new_file_size, p_c_mode); */
1455         return indirect2direct(th, p_s_inode, page, p_s_path, p_s_item_key,
1456                                n_new_file_size, p_c_mode);
1457 }
1458
1459 /* we did indirect_to_direct conversion. And we have inserted direct
1460    item successesfully, but there were no disk space to cut unfm
1461    pointer being converted. Therefore we have to delete inserted
1462    direct item(s) */
1463 static void indirect_to_direct_roll_back(struct reiserfs_transaction_handle *th,
1464                                          struct inode *inode, struct treepath *path)
1465 {
1466         struct cpu_key tail_key;
1467         int tail_len;
1468         int removed;
1469         BUG_ON(!th->t_trans_id);
1470
1471         make_cpu_key(&tail_key, inode, inode->i_size + 1, TYPE_DIRECT, 4);      // !!!!
1472         tail_key.key_length = 4;
1473
1474         tail_len =
1475             (cpu_key_k_offset(&tail_key) & (inode->i_sb->s_blocksize - 1)) - 1;
1476         while (tail_len) {
1477                 /* look for the last byte of the tail */
1478                 if (search_for_position_by_key(inode->i_sb, &tail_key, path) ==
1479                     POSITION_NOT_FOUND)
1480                         reiserfs_panic(inode->i_sb, "vs-5615",
1481                                        "found invalid item");
1482                 RFALSE(path->pos_in_item !=
1483                        ih_item_len(PATH_PITEM_HEAD(path)) - 1,
1484                        "vs-5616: appended bytes found");
1485                 PATH_LAST_POSITION(path)--;
1486
1487                 removed =
1488                     reiserfs_delete_item(th, path, &tail_key, inode,
1489                                          NULL /*unbh not needed */ );
1490                 RFALSE(removed <= 0
1491                        || removed > tail_len,
1492                        "vs-5617: there was tail %d bytes, removed item length %d bytes",
1493                        tail_len, removed);
1494                 tail_len -= removed;
1495                 set_cpu_key_k_offset(&tail_key,
1496                                      cpu_key_k_offset(&tail_key) - removed);
1497         }
1498         reiserfs_warning(inode->i_sb, "reiserfs-5091", "indirect_to_direct "
1499                          "conversion has been rolled back due to "
1500                          "lack of disk space");
1501         //mark_file_without_tail (inode);
1502         mark_inode_dirty(inode);
1503 }
1504
1505 /* (Truncate or cut entry) or delete object item. Returns < 0 on failure */
1506 int reiserfs_cut_from_item(struct reiserfs_transaction_handle *th,
1507                            struct treepath *p_s_path,
1508                            struct cpu_key *p_s_item_key,
1509                            struct inode *p_s_inode,
1510                            struct page *page, loff_t n_new_file_size)
1511 {
1512         struct super_block *sb = p_s_inode->i_sb;
1513         /* Every function which is going to call do_balance must first
1514            create a tree_balance structure.  Then it must fill up this
1515            structure by using the init_tb_struct and fix_nodes functions.
1516            After that we can make tree balancing. */
1517         struct tree_balance s_cut_balance;
1518         struct item_head *p_le_ih;
1519         int n_cut_size = 0,     /* Amount to be cut. */
1520             n_ret_value = CARRY_ON, n_removed = 0,      /* Number of the removed unformatted nodes. */
1521             n_is_inode_locked = 0;
1522         char c_mode;            /* Mode of the balance. */
1523         int retval2 = -1;
1524         int quota_cut_bytes;
1525         loff_t tail_pos = 0;
1526
1527         BUG_ON(!th->t_trans_id);
1528
1529         init_tb_struct(th, &s_cut_balance, p_s_inode->i_sb, p_s_path,
1530                        n_cut_size);
1531
1532         /* Repeat this loop until we either cut the item without needing
1533            to balance, or we fix_nodes without schedule occurring */
1534         while (1) {
1535                 /* Determine the balance mode, position of the first byte to
1536                    be cut, and size to be cut.  In case of the indirect item
1537                    free unformatted nodes which are pointed to by the cut
1538                    pointers. */
1539
1540                 c_mode =
1541                     prepare_for_delete_or_cut(th, p_s_inode, p_s_path,
1542                                               p_s_item_key, &n_removed,
1543                                               &n_cut_size, n_new_file_size);
1544                 if (c_mode == M_CONVERT) {
1545                         /* convert last unformatted node to direct item or leave
1546                            tail in the unformatted node */
1547                         RFALSE(n_ret_value != CARRY_ON,
1548                                "PAP-5570: can not convert twice");
1549
1550                         n_ret_value =
1551                             maybe_indirect_to_direct(th, p_s_inode, page,
1552                                                      p_s_path, p_s_item_key,
1553                                                      n_new_file_size, &c_mode);
1554                         if (c_mode == M_SKIP_BALANCING)
1555                                 /* tail has been left in the unformatted node */
1556                                 return n_ret_value;
1557
1558                         n_is_inode_locked = 1;
1559
1560                         /* removing of last unformatted node will change value we
1561                            have to return to truncate. Save it */
1562                         retval2 = n_ret_value;
1563                         /*retval2 = sb->s_blocksize - (n_new_file_size & (sb->s_blocksize - 1)); */
1564
1565                         /* So, we have performed the first part of the conversion:
1566                            inserting the new direct item.  Now we are removing the
1567                            last unformatted node pointer. Set key to search for
1568                            it. */
1569                         set_cpu_key_k_type(p_s_item_key, TYPE_INDIRECT);
1570                         p_s_item_key->key_length = 4;
1571                         n_new_file_size -=
1572                             (n_new_file_size & (sb->s_blocksize - 1));
1573                         tail_pos = n_new_file_size;
1574                         set_cpu_key_k_offset(p_s_item_key, n_new_file_size + 1);
1575                         if (search_for_position_by_key
1576                             (sb, p_s_item_key,
1577                              p_s_path) == POSITION_NOT_FOUND) {
1578                                 print_block(PATH_PLAST_BUFFER(p_s_path), 3,
1579                                             PATH_LAST_POSITION(p_s_path) - 1,
1580                                             PATH_LAST_POSITION(p_s_path) + 1);
1581                                 reiserfs_panic(sb, "PAP-5580", "item to "
1582                                                "convert does not exist (%K)",
1583                                                p_s_item_key);
1584                         }
1585                         continue;
1586                 }
1587                 if (n_cut_size == 0) {
1588                         pathrelse(p_s_path);
1589                         return 0;
1590                 }
1591
1592                 s_cut_balance.insert_size[0] = n_cut_size;
1593
1594                 n_ret_value = fix_nodes(c_mode, &s_cut_balance, NULL, NULL);
1595                 if (n_ret_value != REPEAT_SEARCH)
1596                         break;
1597
1598                 PROC_INFO_INC(sb, cut_from_item_restarted);
1599
1600                 n_ret_value =
1601                     search_for_position_by_key(sb, p_s_item_key, p_s_path);
1602                 if (n_ret_value == POSITION_FOUND)
1603                         continue;
1604
1605                 reiserfs_warning(sb, "PAP-5610", "item %K not found",
1606                                  p_s_item_key);
1607                 unfix_nodes(&s_cut_balance);
1608                 return (n_ret_value == IO_ERROR) ? -EIO : -ENOENT;
1609         }                       /* while */
1610
1611         // check fix_nodes results (IO_ERROR or NO_DISK_SPACE)
1612         if (n_ret_value != CARRY_ON) {
1613                 if (n_is_inode_locked) {
1614                         // FIXME: this seems to be not needed: we are always able
1615                         // to cut item
1616                         indirect_to_direct_roll_back(th, p_s_inode, p_s_path);
1617                 }
1618                 if (n_ret_value == NO_DISK_SPACE)
1619                         reiserfs_warning(sb, "reiserfs-5092",
1620                                          "NO_DISK_SPACE");
1621                 unfix_nodes(&s_cut_balance);
1622                 return -EIO;
1623         }
1624
1625         /* go ahead and perform balancing */
1626
1627         RFALSE(c_mode == M_PASTE || c_mode == M_INSERT, "invalid mode");
1628
1629         /* Calculate number of bytes that need to be cut from the item. */
1630         quota_cut_bytes =
1631             (c_mode ==
1632              M_DELETE) ? ih_item_len(get_ih(p_s_path)) : -s_cut_balance.
1633             insert_size[0];
1634         if (retval2 == -1)
1635                 n_ret_value = calc_deleted_bytes_number(&s_cut_balance, c_mode);
1636         else
1637                 n_ret_value = retval2;
1638
1639         /* For direct items, we only change the quota when deleting the last
1640          ** item.
1641          */
1642         p_le_ih = PATH_PITEM_HEAD(s_cut_balance.tb_path);
1643         if (!S_ISLNK(p_s_inode->i_mode) && is_direct_le_ih(p_le_ih)) {
1644                 if (c_mode == M_DELETE &&
1645                     (le_ih_k_offset(p_le_ih) & (sb->s_blocksize - 1)) ==
1646                     1) {
1647                         // FIXME: this is to keep 3.5 happy
1648                         REISERFS_I(p_s_inode)->i_first_direct_byte = U32_MAX;
1649                         quota_cut_bytes = sb->s_blocksize + UNFM_P_SIZE;
1650                 } else {
1651                         quota_cut_bytes = 0;
1652                 }
1653         }
1654 #ifdef CONFIG_REISERFS_CHECK
1655         if (n_is_inode_locked) {
1656                 struct item_head *le_ih =
1657                     PATH_PITEM_HEAD(s_cut_balance.tb_path);
1658                 /* we are going to complete indirect2direct conversion. Make
1659                    sure, that we exactly remove last unformatted node pointer
1660                    of the item */
1661                 if (!is_indirect_le_ih(le_ih))
1662                         reiserfs_panic(sb, "vs-5652",
1663                                        "item must be indirect %h", le_ih);
1664
1665                 if (c_mode == M_DELETE && ih_item_len(le_ih) != UNFM_P_SIZE)
1666                         reiserfs_panic(sb, "vs-5653", "completing "
1667                                        "indirect2direct conversion indirect "
1668                                        "item %h being deleted must be of "
1669                                        "4 byte long", le_ih);
1670
1671                 if (c_mode == M_CUT
1672                     && s_cut_balance.insert_size[0] != -UNFM_P_SIZE) {
1673                         reiserfs_panic(sb, "vs-5654", "can not complete "
1674                                        "indirect2direct conversion of %h "
1675                                        "(CUT, insert_size==%d)",
1676                                        le_ih, s_cut_balance.insert_size[0]);
1677                 }
1678                 /* it would be useful to make sure, that right neighboring
1679                    item is direct item of this file */
1680         }
1681 #endif
1682
1683         do_balance(&s_cut_balance, NULL, NULL, c_mode);
1684         if (n_is_inode_locked) {
1685                 /* we've done an indirect->direct conversion.  when the data block
1686                  ** was freed, it was removed from the list of blocks that must
1687                  ** be flushed before the transaction commits, make sure to
1688                  ** unmap and invalidate it
1689                  */
1690                 unmap_buffers(page, tail_pos);
1691                 REISERFS_I(p_s_inode)->i_flags &= ~i_pack_on_close_mask;
1692         }
1693 #ifdef REISERQUOTA_DEBUG
1694         reiserfs_debug(p_s_inode->i_sb, REISERFS_DEBUG_CODE,
1695                        "reiserquota cut_from_item(): freeing %u id=%u type=%c",
1696                        quota_cut_bytes, p_s_inode->i_uid, '?');
1697 #endif
1698         DQUOT_FREE_SPACE_NODIRTY(p_s_inode, quota_cut_bytes);
1699         return n_ret_value;
1700 }
1701
1702 static void truncate_directory(struct reiserfs_transaction_handle *th,
1703                                struct inode *inode)
1704 {
1705         BUG_ON(!th->t_trans_id);
1706         if (inode->i_nlink)
1707                 reiserfs_error(inode->i_sb, "vs-5655", "link count != 0");
1708
1709         set_le_key_k_offset(KEY_FORMAT_3_5, INODE_PKEY(inode), DOT_OFFSET);
1710         set_le_key_k_type(KEY_FORMAT_3_5, INODE_PKEY(inode), TYPE_DIRENTRY);
1711         reiserfs_delete_solid_item(th, inode, INODE_PKEY(inode));
1712         reiserfs_update_sd(th, inode);
1713         set_le_key_k_offset(KEY_FORMAT_3_5, INODE_PKEY(inode), SD_OFFSET);
1714         set_le_key_k_type(KEY_FORMAT_3_5, INODE_PKEY(inode), TYPE_STAT_DATA);
1715 }
1716
1717 /* Truncate file to the new size. Note, this must be called with a transaction
1718    already started */
1719 int reiserfs_do_truncate(struct reiserfs_transaction_handle *th, struct inode *p_s_inode,       /* ->i_size contains new
1720                                                                                                    size */
1721                          struct page *page,     /* up to date for last block */
1722                          int update_timestamps  /* when it is called by
1723                                                    file_release to convert
1724                                                    the tail - no timestamps
1725                                                    should be updated */
1726     )
1727 {
1728         INITIALIZE_PATH(s_search_path); /* Path to the current object item. */
1729         struct item_head *p_le_ih;      /* Pointer to an item header. */
1730         struct cpu_key s_item_key;      /* Key to search for a previous file item. */
1731         loff_t n_file_size,     /* Old file size. */
1732          n_new_file_size;       /* New file size. */
1733         int n_deleted;          /* Number of deleted or truncated bytes. */
1734         int retval;
1735         int err = 0;
1736
1737         BUG_ON(!th->t_trans_id);
1738         if (!
1739             (S_ISREG(p_s_inode->i_mode) || S_ISDIR(p_s_inode->i_mode)
1740              || S_ISLNK(p_s_inode->i_mode)))
1741                 return 0;
1742
1743         if (S_ISDIR(p_s_inode->i_mode)) {
1744                 // deletion of directory - no need to update timestamps
1745                 truncate_directory(th, p_s_inode);
1746                 return 0;
1747         }
1748
1749         /* Get new file size. */
1750         n_new_file_size = p_s_inode->i_size;
1751
1752         // FIXME: note, that key type is unimportant here
1753         make_cpu_key(&s_item_key, p_s_inode, max_reiserfs_offset(p_s_inode),
1754                      TYPE_DIRECT, 3);
1755
1756         retval =
1757             search_for_position_by_key(p_s_inode->i_sb, &s_item_key,
1758                                        &s_search_path);
1759         if (retval == IO_ERROR) {
1760                 reiserfs_error(p_s_inode->i_sb, "vs-5657",
1761                                "i/o failure occurred trying to truncate %K",
1762                                &s_item_key);
1763                 err = -EIO;
1764                 goto out;
1765         }
1766         if (retval == POSITION_FOUND || retval == FILE_NOT_FOUND) {
1767                 reiserfs_error(p_s_inode->i_sb, "PAP-5660",
1768                                "wrong result %d of search for %K", retval,
1769                                &s_item_key);
1770
1771                 err = -EIO;
1772                 goto out;
1773         }
1774
1775         s_search_path.pos_in_item--;
1776
1777         /* Get real file size (total length of all file items) */
1778         p_le_ih = PATH_PITEM_HEAD(&s_search_path);
1779         if (is_statdata_le_ih(p_le_ih))
1780                 n_file_size = 0;
1781         else {
1782                 loff_t offset = le_ih_k_offset(p_le_ih);
1783                 int bytes =
1784                     op_bytes_number(p_le_ih, p_s_inode->i_sb->s_blocksize);
1785
1786                 /* this may mismatch with real file size: if last direct item
1787                    had no padding zeros and last unformatted node had no free
1788                    space, this file would have this file size */
1789                 n_file_size = offset + bytes - 1;
1790         }
1791         /*
1792          * are we doing a full truncate or delete, if so
1793          * kick in the reada code
1794          */
1795         if (n_new_file_size == 0)
1796                 s_search_path.reada = PATH_READA | PATH_READA_BACK;
1797
1798         if (n_file_size == 0 || n_file_size < n_new_file_size) {
1799                 goto update_and_out;
1800         }
1801
1802         /* Update key to search for the last file item. */
1803         set_cpu_key_k_offset(&s_item_key, n_file_size);
1804
1805         do {
1806                 /* Cut or delete file item. */
1807                 n_deleted =
1808                     reiserfs_cut_from_item(th, &s_search_path, &s_item_key,
1809                                            p_s_inode, page, n_new_file_size);
1810                 if (n_deleted < 0) {
1811                         reiserfs_warning(p_s_inode->i_sb, "vs-5665",
1812                                          "reiserfs_cut_from_item failed");
1813                         reiserfs_check_path(&s_search_path);
1814                         return 0;
1815                 }
1816
1817                 RFALSE(n_deleted > n_file_size,
1818                        "PAP-5670: reiserfs_cut_from_item: too many bytes deleted: deleted %d, file_size %lu, item_key %K",
1819                        n_deleted, n_file_size, &s_item_key);
1820
1821                 /* Change key to search the last file item. */
1822                 n_file_size -= n_deleted;
1823
1824                 set_cpu_key_k_offset(&s_item_key, n_file_size);
1825
1826                 /* While there are bytes to truncate and previous file item is presented in the tree. */
1827
1828                 /*
1829                  ** This loop could take a really long time, and could log
1830                  ** many more blocks than a transaction can hold.  So, we do a polite
1831                  ** journal end here, and if the transaction needs ending, we make
1832                  ** sure the file is consistent before ending the current trans
1833                  ** and starting a new one
1834                  */
1835                 if (journal_transaction_should_end(th, 0) ||
1836                     reiserfs_transaction_free_space(th) <= JOURNAL_FOR_FREE_BLOCK_AND_UPDATE_SD) {
1837                         int orig_len_alloc = th->t_blocks_allocated;
1838                         pathrelse(&s_search_path);
1839
1840                         if (update_timestamps) {
1841                                 p_s_inode->i_mtime = p_s_inode->i_ctime =
1842                                     CURRENT_TIME_SEC;
1843                         }
1844                         reiserfs_update_sd(th, p_s_inode);
1845
1846                         err = journal_end(th, p_s_inode->i_sb, orig_len_alloc);
1847                         if (err)
1848                                 goto out;
1849                         err = journal_begin(th, p_s_inode->i_sb,
1850                                             JOURNAL_FOR_FREE_BLOCK_AND_UPDATE_SD + JOURNAL_PER_BALANCE_CNT * 4) ;
1851                         if (err)
1852                                 goto out;
1853                         reiserfs_update_inode_transaction(p_s_inode);
1854                 }
1855         } while (n_file_size > ROUND_UP(n_new_file_size) &&
1856                  search_for_position_by_key(p_s_inode->i_sb, &s_item_key,
1857                                             &s_search_path) == POSITION_FOUND);
1858
1859         RFALSE(n_file_size > ROUND_UP(n_new_file_size),
1860                "PAP-5680: truncate did not finish: new_file_size %Ld, current %Ld, oid %d",
1861                n_new_file_size, n_file_size, s_item_key.on_disk_key.k_objectid);
1862
1863       update_and_out:
1864         if (update_timestamps) {
1865                 // this is truncate, not file closing
1866                 p_s_inode->i_mtime = p_s_inode->i_ctime = CURRENT_TIME_SEC;
1867         }
1868         reiserfs_update_sd(th, p_s_inode);
1869
1870       out:
1871         pathrelse(&s_search_path);
1872         return err;
1873 }
1874
1875 #ifdef CONFIG_REISERFS_CHECK
1876 // this makes sure, that we __append__, not overwrite or add holes
1877 static void check_research_for_paste(struct treepath *path,
1878                                      const struct cpu_key *p_s_key)
1879 {
1880         struct item_head *found_ih = get_ih(path);
1881
1882         if (is_direct_le_ih(found_ih)) {
1883                 if (le_ih_k_offset(found_ih) +
1884                     op_bytes_number(found_ih,
1885                                     get_last_bh(path)->b_size) !=
1886                     cpu_key_k_offset(p_s_key)
1887                     || op_bytes_number(found_ih,
1888                                        get_last_bh(path)->b_size) !=
1889                     pos_in_item(path))
1890                         reiserfs_panic(NULL, "PAP-5720", "found direct item "
1891                                        "%h or position (%d) does not match "
1892                                        "to key %K", found_ih,
1893                                        pos_in_item(path), p_s_key);
1894         }
1895         if (is_indirect_le_ih(found_ih)) {
1896                 if (le_ih_k_offset(found_ih) +
1897                     op_bytes_number(found_ih,
1898                                     get_last_bh(path)->b_size) !=
1899                     cpu_key_k_offset(p_s_key)
1900                     || I_UNFM_NUM(found_ih) != pos_in_item(path)
1901                     || get_ih_free_space(found_ih) != 0)
1902                         reiserfs_panic(NULL, "PAP-5730", "found indirect "
1903                                        "item (%h) or position (%d) does not "
1904                                        "match to key (%K)",
1905                                        found_ih, pos_in_item(path), p_s_key);
1906         }
1907 }
1908 #endif                          /* config reiserfs check */
1909
1910 /* Paste bytes to the existing item. Returns bytes number pasted into the item. */
1911 int reiserfs_paste_into_item(struct reiserfs_transaction_handle *th, struct treepath *p_s_search_path,  /* Path to the pasted item.          */
1912                              const struct cpu_key *p_s_key,     /* Key to search for the needed item. */
1913                              struct inode *inode,       /* Inode item belongs to */
1914                              const char *p_c_body,      /* Pointer to the bytes to paste.    */
1915                              int n_pasted_size)
1916 {                               /* Size of pasted bytes.             */
1917         struct tree_balance s_paste_balance;
1918         int retval;
1919         int fs_gen;
1920
1921         BUG_ON(!th->t_trans_id);
1922
1923         fs_gen = get_generation(inode->i_sb);
1924
1925 #ifdef REISERQUOTA_DEBUG
1926         reiserfs_debug(inode->i_sb, REISERFS_DEBUG_CODE,
1927                        "reiserquota paste_into_item(): allocating %u id=%u type=%c",
1928                        n_pasted_size, inode->i_uid,
1929                        key2type(&(p_s_key->on_disk_key)));
1930 #endif
1931
1932         if (DQUOT_ALLOC_SPACE_NODIRTY(inode, n_pasted_size)) {
1933                 pathrelse(p_s_search_path);
1934                 return -EDQUOT;
1935         }
1936         init_tb_struct(th, &s_paste_balance, th->t_super, p_s_search_path,
1937                        n_pasted_size);
1938 #ifdef DISPLACE_NEW_PACKING_LOCALITIES
1939         s_paste_balance.key = p_s_key->on_disk_key;
1940 #endif
1941
1942         /* DQUOT_* can schedule, must check before the fix_nodes */
1943         if (fs_changed(fs_gen, inode->i_sb)) {
1944                 goto search_again;
1945         }
1946
1947         while ((retval =
1948                 fix_nodes(M_PASTE, &s_paste_balance, NULL,
1949                           p_c_body)) == REPEAT_SEARCH) {
1950               search_again:
1951                 /* file system changed while we were in the fix_nodes */
1952                 PROC_INFO_INC(th->t_super, paste_into_item_restarted);
1953                 retval =
1954                     search_for_position_by_key(th->t_super, p_s_key,
1955                                                p_s_search_path);
1956                 if (retval == IO_ERROR) {
1957                         retval = -EIO;
1958                         goto error_out;
1959                 }
1960                 if (retval == POSITION_FOUND) {
1961                         reiserfs_warning(inode->i_sb, "PAP-5710",
1962                                          "entry or pasted byte (%K) exists",
1963                                          p_s_key);
1964                         retval = -EEXIST;
1965                         goto error_out;
1966                 }
1967 #ifdef CONFIG_REISERFS_CHECK
1968                 check_research_for_paste(p_s_search_path, p_s_key);
1969 #endif
1970         }
1971
1972         /* Perform balancing after all resources are collected by fix_nodes, and
1973            accessing them will not risk triggering schedule. */
1974         if (retval == CARRY_ON) {
1975                 do_balance(&s_paste_balance, NULL /*ih */ , p_c_body, M_PASTE);
1976                 return 0;
1977         }
1978         retval = (retval == NO_DISK_SPACE) ? -ENOSPC : -EIO;
1979       error_out:
1980         /* this also releases the path */
1981         unfix_nodes(&s_paste_balance);
1982 #ifdef REISERQUOTA_DEBUG
1983         reiserfs_debug(inode->i_sb, REISERFS_DEBUG_CODE,
1984                        "reiserquota paste_into_item(): freeing %u id=%u type=%c",
1985                        n_pasted_size, inode->i_uid,
1986                        key2type(&(p_s_key->on_disk_key)));
1987 #endif
1988         DQUOT_FREE_SPACE_NODIRTY(inode, n_pasted_size);
1989         return retval;
1990 }
1991
1992 /* Insert new item into the buffer at the path. */
1993 int reiserfs_insert_item(struct reiserfs_transaction_handle *th, struct treepath *p_s_path,     /* Path to the inserteded item.         */
1994                          const struct cpu_key *key, struct item_head *p_s_ih,   /* Pointer to the item header to insert. */
1995                          struct inode *inode, const char *p_c_body)
1996 {                               /* Pointer to the bytes to insert.      */
1997         struct tree_balance s_ins_balance;
1998         int retval;
1999         int fs_gen = 0;
2000         int quota_bytes = 0;
2001
2002         BUG_ON(!th->t_trans_id);
2003
2004         if (inode) {            /* Do we count quotas for item? */
2005                 fs_gen = get_generation(inode->i_sb);
2006                 quota_bytes = ih_item_len(p_s_ih);
2007
2008                 /* hack so the quota code doesn't have to guess if the file has
2009                  ** a tail, links are always tails, so there's no guessing needed
2010                  */
2011                 if (!S_ISLNK(inode->i_mode) && is_direct_le_ih(p_s_ih)) {
2012                         quota_bytes = inode->i_sb->s_blocksize + UNFM_P_SIZE;
2013                 }
2014 #ifdef REISERQUOTA_DEBUG
2015                 reiserfs_debug(inode->i_sb, REISERFS_DEBUG_CODE,
2016                                "reiserquota insert_item(): allocating %u id=%u type=%c",
2017                                quota_bytes, inode->i_uid, head2type(p_s_ih));
2018 #endif
2019                 /* We can't dirty inode here. It would be immediately written but
2020                  * appropriate stat item isn't inserted yet... */
2021                 if (DQUOT_ALLOC_SPACE_NODIRTY(inode, quota_bytes)) {
2022                         pathrelse(p_s_path);
2023                         return -EDQUOT;
2024                 }
2025         }
2026         init_tb_struct(th, &s_ins_balance, th->t_super, p_s_path,
2027                        IH_SIZE + ih_item_len(p_s_ih));
2028 #ifdef DISPLACE_NEW_PACKING_LOCALITIES
2029         s_ins_balance.key = key->on_disk_key;
2030 #endif
2031         /* DQUOT_* can schedule, must check to be sure calling fix_nodes is safe */
2032         if (inode && fs_changed(fs_gen, inode->i_sb)) {
2033                 goto search_again;
2034         }
2035
2036         while ((retval =
2037                 fix_nodes(M_INSERT, &s_ins_balance, p_s_ih,
2038                           p_c_body)) == REPEAT_SEARCH) {
2039               search_again:
2040                 /* file system changed while we were in the fix_nodes */
2041                 PROC_INFO_INC(th->t_super, insert_item_restarted);
2042                 retval = search_item(th->t_super, key, p_s_path);
2043                 if (retval == IO_ERROR) {
2044                         retval = -EIO;
2045                         goto error_out;
2046                 }
2047                 if (retval == ITEM_FOUND) {
2048                         reiserfs_warning(th->t_super, "PAP-5760",
2049                                          "key %K already exists in the tree",
2050                                          key);
2051                         retval = -EEXIST;
2052                         goto error_out;
2053                 }
2054         }
2055
2056         /* make balancing after all resources will be collected at a time */
2057         if (retval == CARRY_ON) {
2058                 do_balance(&s_ins_balance, p_s_ih, p_c_body, M_INSERT);
2059                 return 0;
2060         }
2061
2062         retval = (retval == NO_DISK_SPACE) ? -ENOSPC : -EIO;
2063       error_out:
2064         /* also releases the path */
2065         unfix_nodes(&s_ins_balance);
2066 #ifdef REISERQUOTA_DEBUG
2067         reiserfs_debug(th->t_super, REISERFS_DEBUG_CODE,
2068                        "reiserquota insert_item(): freeing %u id=%u type=%c",
2069                        quota_bytes, inode->i_uid, head2type(p_s_ih));
2070 #endif
2071         if (inode)
2072                 DQUOT_FREE_SPACE_NODIRTY(inode, quota_bytes);
2073         return retval;
2074 }