1 /* 2 * Copyright 2000 by Hans Reiser, licensing governed by reiserfs/README 3 */ 4 /* Reiserfs block (de)allocator, bitmap-based. */ 5 6 #include <linux/time.h> 7 #include <linux/reiserfs_fs.h> 8 #include <linux/errno.h> 9 #include <linux/buffer_head.h> 10 #include <linux/kernel.h> 11 #include <linux/pagemap.h> 12 #include <linux/vmalloc.h> 13 #include <linux/reiserfs_fs_sb.h> 14 #include <linux/reiserfs_fs_i.h> 15 #include <linux/quotaops.h> 16 17 #define PREALLOCATION_SIZE 9 18 19 /* different reiserfs block allocator options */ 20 21 #define SB_ALLOC_OPTS(s) (REISERFS_SB(s)->s_alloc_options.bits) 22 23 #define _ALLOC_concentrating_formatted_nodes 0 24 #define _ALLOC_displacing_large_files 1 25 #define _ALLOC_displacing_new_packing_localities 2 26 #define _ALLOC_old_hashed_relocation 3 27 #define _ALLOC_new_hashed_relocation 4 28 #define _ALLOC_skip_busy 5 29 #define _ALLOC_displace_based_on_dirid 6 30 #define _ALLOC_hashed_formatted_nodes 7 31 #define _ALLOC_old_way 8 32 #define _ALLOC_hundredth_slices 9 33 #define _ALLOC_dirid_groups 10 34 #define _ALLOC_oid_groups 11 35 #define _ALLOC_packing_groups 12 36 37 #define concentrating_formatted_nodes(s) test_bit(_ALLOC_concentrating_formatted_nodes, &SB_ALLOC_OPTS(s)) 38 #define displacing_large_files(s) test_bit(_ALLOC_displacing_large_files, &SB_ALLOC_OPTS(s)) 39 #define displacing_new_packing_localities(s) test_bit(_ALLOC_displacing_new_packing_localities, &SB_ALLOC_OPTS(s)) 40 41 #define SET_OPTION(optname) \ 42 do { \ 43 reiserfs_info(s, "block allocator option \"%s\" is set", #optname); \ 44 set_bit(_ALLOC_ ## optname , &SB_ALLOC_OPTS(s)); \ 45 } while(0) 46 #define TEST_OPTION(optname, s) \ 47 test_bit(_ALLOC_ ## optname , &SB_ALLOC_OPTS(s)) 48 49 static inline void get_bit_address(struct super_block *s, 50 b_blocknr_t block, 51 unsigned int *bmap_nr, 52 unsigned int *offset) 53 { 54 /* It is in the bitmap block number equal to the block 55 * number divided by the number of bits in a block. */ 56 *bmap_nr = block >> (s->s_blocksize_bits + 3); 57 /* Within that bitmap block it is located at bit offset *offset. */ 58 *offset = block & ((s->s_blocksize << 3) - 1); 59 } 60 61 int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value) 62 { 63 unsigned int bmap, offset; 64 unsigned int bmap_count = reiserfs_bmap_count(s); 65 66 if (block == 0 || block >= SB_BLOCK_COUNT(s)) { 67 reiserfs_error(s, "vs-4010", 68 "block number is out of range %lu (%u)", 69 block, SB_BLOCK_COUNT(s)); 70 return 0; 71 } 72 73 get_bit_address(s, block, &bmap, &offset); 74 75 /* Old format filesystem? Unlikely, but the bitmaps are all up front so 76 * we need to account for it. */ 77 if (unlikely(test_bit(REISERFS_OLD_FORMAT, 78 &(REISERFS_SB(s)->s_properties)))) { 79 b_blocknr_t bmap1 = REISERFS_SB(s)->s_sbh->b_blocknr + 1; 80 if (block >= bmap1 && 81 block <= bmap1 + bmap_count) { 82 reiserfs_error(s, "vs-4019", "bitmap block %lu(%u) " 83 "can't be freed or reused", 84 block, bmap_count); 85 return 0; 86 } 87 } else { 88 if (offset == 0) { 89 reiserfs_error(s, "vs-4020", "bitmap block %lu(%u) " 90 "can't be freed or reused", 91 block, bmap_count); 92 return 0; 93 } 94 } 95 96 if (bmap >= bmap_count) { 97 reiserfs_error(s, "vs-4030", "bitmap for requested block " 98 "is out of range: block=%lu, bitmap_nr=%u", 99 block, bmap); 100 return 0; 101 } 102 103 if (bit_value == 0 && block == SB_ROOT_BLOCK(s)) { 104 reiserfs_error(s, "vs-4050", "this is root block (%u), " 105 "it must be busy", SB_ROOT_BLOCK(s)); 106 return 0; 107 } 108 109 return 1; 110 } 111 112 /* searches in journal structures for a given block number (bmap, off). If block 113 is found in reiserfs journal it suggests next free block candidate to test. */ 114 static inline int is_block_in_journal(struct super_block *s, unsigned int bmap, 115 int off, int *next) 116 { 117 b_blocknr_t tmp; 118 119 if (reiserfs_in_journal(s, bmap, off, 1, &tmp)) { 120 if (tmp) { /* hint supplied */ 121 *next = tmp; 122 PROC_INFO_INC(s, scan_bitmap.in_journal_hint); 123 } else { 124 (*next) = off + 1; /* inc offset to avoid looping. */ 125 PROC_INFO_INC(s, scan_bitmap.in_journal_nohint); 126 } 127 PROC_INFO_INC(s, scan_bitmap.retry); 128 return 1; 129 } 130 return 0; 131 } 132 133 /* it searches for a window of zero bits with given minimum and maximum lengths in one bitmap 134 * block; */ 135 static int scan_bitmap_block(struct reiserfs_transaction_handle *th, 136 unsigned int bmap_n, int *beg, int boundary, 137 int min, int max, int unfm) 138 { 139 struct super_block *s = th->t_super; 140 struct reiserfs_bitmap_info *bi = &SB_AP_BITMAP(s)[bmap_n]; 141 struct buffer_head *bh; 142 int end, next; 143 int org = *beg; 144 145 BUG_ON(!th->t_trans_id); 146 147 RFALSE(bmap_n >= reiserfs_bmap_count(s), "Bitmap %u is out of " 148 "range (0..%u)", bmap_n, reiserfs_bmap_count(s) - 1); 149 PROC_INFO_INC(s, scan_bitmap.bmap); 150 /* this is unclear and lacks comments, explain how journal bitmaps 151 work here for the reader. Convey a sense of the design here. What 152 is a window? */ 153 /* - I mean `a window of zero bits' as in description of this function - Zam. */ 154 155 if (!bi) { 156 reiserfs_error(s, "jdm-4055", "NULL bitmap info pointer " 157 "for bitmap %d", bmap_n); 158 return 0; 159 } 160 161 bh = reiserfs_read_bitmap_block(s, bmap_n); 162 if (bh == NULL) 163 return 0; 164 165 while (1) { 166 cont: 167 if (bi->free_count < min) { 168 brelse(bh); 169 return 0; // No free blocks in this bitmap 170 } 171 172 /* search for a first zero bit -- beginning of a window */ 173 *beg = reiserfs_find_next_zero_le_bit 174 ((unsigned long *)(bh->b_data), boundary, *beg); 175 176 if (*beg + min > boundary) { /* search for a zero bit fails or the rest of bitmap block 177 * cannot contain a zero window of minimum size */ 178 brelse(bh); 179 return 0; 180 } 181 182 if (unfm && is_block_in_journal(s, bmap_n, *beg, beg)) 183 continue; 184 /* first zero bit found; we check next bits */ 185 for (end = *beg + 1;; end++) { 186 if (end >= *beg + max || end >= boundary 187 || reiserfs_test_le_bit(end, bh->b_data)) { 188 next = end; 189 break; 190 } 191 /* finding the other end of zero bit window requires looking into journal structures (in 192 * case of searching for free blocks for unformatted nodes) */ 193 if (unfm && is_block_in_journal(s, bmap_n, end, &next)) 194 break; 195 } 196 197 /* now (*beg) points to beginning of zero bits window, 198 * (end) points to one bit after the window end */ 199 if (end - *beg >= min) { /* it seems we have found window of proper size */ 200 int i; 201 reiserfs_prepare_for_journal(s, bh, 1); 202 /* try to set all blocks used checking are they still free */ 203 for (i = *beg; i < end; i++) { 204 /* It seems that we should not check in journal again. */ 205 if (reiserfs_test_and_set_le_bit 206 (i, bh->b_data)) { 207 /* bit was set by another process 208 * while we slept in prepare_for_journal() */ 209 PROC_INFO_INC(s, scan_bitmap.stolen); 210 if (i >= *beg + min) { /* we can continue with smaller set of allocated blocks, 211 * if length of this set is more or equal to `min' */ 212 end = i; 213 break; 214 } 215 /* otherwise we clear all bit were set ... */ 216 while (--i >= *beg) 217 reiserfs_test_and_clear_le_bit 218 (i, bh->b_data); 219 reiserfs_restore_prepared_buffer(s, bh); 220 *beg = org; 221 /* ... and search again in current block from beginning */ 222 goto cont; 223 } 224 } 225 bi->free_count -= (end - *beg); 226 journal_mark_dirty(th, s, bh); 227 brelse(bh); 228 229 /* free block count calculation */ 230 reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 231 1); 232 PUT_SB_FREE_BLOCKS(s, SB_FREE_BLOCKS(s) - (end - *beg)); 233 journal_mark_dirty(th, s, SB_BUFFER_WITH_SB(s)); 234 235 return end - (*beg); 236 } else { 237 *beg = next; 238 } 239 } 240 } 241 242 static int bmap_hash_id(struct super_block *s, u32 id) 243 { 244 char *hash_in = NULL; 245 unsigned long hash; 246 unsigned bm; 247 248 if (id <= 2) { 249 bm = 1; 250 } else { 251 hash_in = (char *)(&id); 252 hash = keyed_hash(hash_in, 4); 253 bm = hash % reiserfs_bmap_count(s); 254 if (!bm) 255 bm = 1; 256 } 257 /* this can only be true when SB_BMAP_NR = 1 */ 258 if (bm >= reiserfs_bmap_count(s)) 259 bm = 0; 260 return bm; 261 } 262 263 /* 264 * hashes the id and then returns > 0 if the block group for the 265 * corresponding hash is full 266 */ 267 static inline int block_group_used(struct super_block *s, u32 id) 268 { 269 int bm = bmap_hash_id(s, id); 270 struct reiserfs_bitmap_info *info = &SB_AP_BITMAP(s)[bm]; 271 272 /* If we don't have cached information on this bitmap block, we're 273 * going to have to load it later anyway. Loading it here allows us 274 * to make a better decision. This favors long-term performance gain 275 * with a better on-disk layout vs. a short term gain of skipping the 276 * read and potentially having a bad placement. */ 277 if (info->free_count == UINT_MAX) { 278 struct buffer_head *bh = reiserfs_read_bitmap_block(s, bm); 279 brelse(bh); 280 } 281 282 if (info->free_count > ((s->s_blocksize << 3) * 60 / 100)) { 283 return 0; 284 } 285 return 1; 286 } 287 288 /* 289 * the packing is returned in disk byte order 290 */ 291 __le32 reiserfs_choose_packing(struct inode * dir) 292 { 293 __le32 packing; 294 if (TEST_OPTION(packing_groups, dir->i_sb)) { 295 u32 parent_dir = le32_to_cpu(INODE_PKEY(dir)->k_dir_id); 296 /* 297 * some versions of reiserfsck expect packing locality 1 to be 298 * special 299 */ 300 if (parent_dir == 1 || block_group_used(dir->i_sb, parent_dir)) 301 packing = INODE_PKEY(dir)->k_objectid; 302 else 303 packing = INODE_PKEY(dir)->k_dir_id; 304 } else 305 packing = INODE_PKEY(dir)->k_objectid; 306 return packing; 307 } 308 309 /* Tries to find contiguous zero bit window (given size) in given region of 310 * bitmap and place new blocks there. Returns number of allocated blocks. */ 311 static int scan_bitmap(struct reiserfs_transaction_handle *th, 312 b_blocknr_t * start, b_blocknr_t finish, 313 int min, int max, int unfm, sector_t file_block) 314 { 315 int nr_allocated = 0; 316 struct super_block *s = th->t_super; 317 /* find every bm and bmap and bmap_nr in this file, and change them all to bitmap_blocknr 318 * - Hans, it is not a block number - Zam. */ 319 320 unsigned int bm, off; 321 unsigned int end_bm, end_off; 322 unsigned int off_max = s->s_blocksize << 3; 323 324 BUG_ON(!th->t_trans_id); 325 326 PROC_INFO_INC(s, scan_bitmap.call); 327 if (SB_FREE_BLOCKS(s) <= 0) 328 return 0; // No point in looking for more free blocks 329 330 get_bit_address(s, *start, &bm, &off); 331 get_bit_address(s, finish, &end_bm, &end_off); 332 if (bm > reiserfs_bmap_count(s)) 333 return 0; 334 if (end_bm > reiserfs_bmap_count(s)) 335 end_bm = reiserfs_bmap_count(s); 336 337 /* When the bitmap is more than 10% free, anyone can allocate. 338 * When it's less than 10% free, only files that already use the 339 * bitmap are allowed. Once we pass 80% full, this restriction 340 * is lifted. 341 * 342 * We do this so that files that grow later still have space close to 343 * their original allocation. This improves locality, and presumably 344 * performance as a result. 345 * 346 * This is only an allocation policy and does not make up for getting a 347 * bad hint. Decent hinting must be implemented for this to work well. 348 */ 349 if (TEST_OPTION(skip_busy, s) 350 && SB_FREE_BLOCKS(s) > SB_BLOCK_COUNT(s) / 20) { 351 for (; bm < end_bm; bm++, off = 0) { 352 if ((off && (!unfm || (file_block != 0))) 353 || SB_AP_BITMAP(s)[bm].free_count > 354 (s->s_blocksize << 3) / 10) 355 nr_allocated = 356 scan_bitmap_block(th, bm, &off, off_max, 357 min, max, unfm); 358 if (nr_allocated) 359 goto ret; 360 } 361 /* we know from above that start is a reasonable number */ 362 get_bit_address(s, *start, &bm, &off); 363 } 364 365 for (; bm < end_bm; bm++, off = 0) { 366 nr_allocated = 367 scan_bitmap_block(th, bm, &off, off_max, min, max, unfm); 368 if (nr_allocated) 369 goto ret; 370 } 371 372 nr_allocated = 373 scan_bitmap_block(th, bm, &off, end_off + 1, min, max, unfm); 374 375 ret: 376 *start = bm * off_max + off; 377 return nr_allocated; 378 379 } 380 381 static void _reiserfs_free_block(struct reiserfs_transaction_handle *th, 382 struct inode *inode, b_blocknr_t block, 383 int for_unformatted) 384 { 385 struct super_block *s = th->t_super; 386 struct reiserfs_super_block *rs; 387 struct buffer_head *sbh, *bmbh; 388 struct reiserfs_bitmap_info *apbi; 389 unsigned int nr, offset; 390 391 BUG_ON(!th->t_trans_id); 392 393 PROC_INFO_INC(s, free_block); 394 395 rs = SB_DISK_SUPER_BLOCK(s); 396 sbh = SB_BUFFER_WITH_SB(s); 397 apbi = SB_AP_BITMAP(s); 398 399 get_bit_address(s, block, &nr, &offset); 400 401 if (nr >= reiserfs_bmap_count(s)) { 402 reiserfs_error(s, "vs-4075", "block %lu is out of range", 403 block); 404 return; 405 } 406 407 bmbh = reiserfs_read_bitmap_block(s, nr); 408 if (!bmbh) 409 return; 410 411 reiserfs_prepare_for_journal(s, bmbh, 1); 412 413 /* clear bit for the given block in bit map */ 414 if (!reiserfs_test_and_clear_le_bit(offset, bmbh->b_data)) { 415 reiserfs_error(s, "vs-4080", 416 "block %lu: bit already cleared", block); 417 } 418 apbi[nr].free_count++; 419 journal_mark_dirty(th, s, bmbh); 420 brelse(bmbh); 421 422 reiserfs_prepare_for_journal(s, sbh, 1); 423 /* update super block */ 424 set_sb_free_blocks(rs, sb_free_blocks(rs) + 1); 425 426 journal_mark_dirty(th, s, sbh); 427 if (for_unformatted) 428 dquot_free_block_nodirty(inode, 1); 429 } 430 431 void reiserfs_free_block(struct reiserfs_transaction_handle *th, 432 struct inode *inode, b_blocknr_t block, 433 int for_unformatted) 434 { 435 struct super_block *s = th->t_super; 436 BUG_ON(!th->t_trans_id); 437 438 RFALSE(!s, "vs-4061: trying to free block on nonexistent device"); 439 if (!is_reusable(s, block, 1)) 440 return; 441 442 if (block > sb_block_count(REISERFS_SB(s)->s_rs)) { 443 reiserfs_error(th->t_super, "bitmap-4072", 444 "Trying to free block outside file system " 445 "boundaries (%lu > %lu)", 446 block, sb_block_count(REISERFS_SB(s)->s_rs)); 447 return; 448 } 449 /* mark it before we clear it, just in case */ 450 journal_mark_freed(th, s, block); 451 _reiserfs_free_block(th, inode, block, for_unformatted); 452 } 453 454 /* preallocated blocks don't need to be run through journal_mark_freed */ 455 static void reiserfs_free_prealloc_block(struct reiserfs_transaction_handle *th, 456 struct inode *inode, b_blocknr_t block) 457 { 458 BUG_ON(!th->t_trans_id); 459 RFALSE(!th->t_super, 460 "vs-4060: trying to free block on nonexistent device"); 461 if (!is_reusable(th->t_super, block, 1)) 462 return; 463 _reiserfs_free_block(th, inode, block, 1); 464 } 465 466 static void __discard_prealloc(struct reiserfs_transaction_handle *th, 467 struct reiserfs_inode_info *ei) 468 { 469 unsigned long save = ei->i_prealloc_block; 470 int dirty = 0; 471 struct inode *inode = &ei->vfs_inode; 472 BUG_ON(!th->t_trans_id); 473 #ifdef CONFIG_REISERFS_CHECK 474 if (ei->i_prealloc_count < 0) 475 reiserfs_error(th->t_super, "zam-4001", 476 "inode has negative prealloc blocks count."); 477 #endif 478 while (ei->i_prealloc_count > 0) { 479 reiserfs_free_prealloc_block(th, inode, ei->i_prealloc_block); 480 ei->i_prealloc_block++; 481 ei->i_prealloc_count--; 482 dirty = 1; 483 } 484 if (dirty) 485 reiserfs_update_sd(th, inode); 486 ei->i_prealloc_block = save; 487 list_del_init(&(ei->i_prealloc_list)); 488 } 489 490 /* FIXME: It should be inline function */ 491 void reiserfs_discard_prealloc(struct reiserfs_transaction_handle *th, 492 struct inode *inode) 493 { 494 struct reiserfs_inode_info *ei = REISERFS_I(inode); 495 BUG_ON(!th->t_trans_id); 496 if (ei->i_prealloc_count) 497 __discard_prealloc(th, ei); 498 } 499 500 void reiserfs_discard_all_prealloc(struct reiserfs_transaction_handle *th) 501 { 502 struct list_head *plist = &SB_JOURNAL(th->t_super)->j_prealloc_list; 503 504 BUG_ON(!th->t_trans_id); 505 506 while (!list_empty(plist)) { 507 struct reiserfs_inode_info *ei; 508 ei = list_entry(plist->next, struct reiserfs_inode_info, 509 i_prealloc_list); 510 #ifdef CONFIG_REISERFS_CHECK 511 if (!ei->i_prealloc_count) { 512 reiserfs_error(th->t_super, "zam-4001", 513 "inode is in prealloc list but has " 514 "no preallocated blocks."); 515 } 516 #endif 517 __discard_prealloc(th, ei); 518 } 519 } 520 521 void reiserfs_init_alloc_options(struct super_block *s) 522 { 523 set_bit(_ALLOC_skip_busy, &SB_ALLOC_OPTS(s)); 524 set_bit(_ALLOC_dirid_groups, &SB_ALLOC_OPTS(s)); 525 set_bit(_ALLOC_packing_groups, &SB_ALLOC_OPTS(s)); 526 } 527 528 /* block allocator related options are parsed here */ 529 int reiserfs_parse_alloc_options(struct super_block *s, char *options) 530 { 531 char *this_char, *value; 532 533 REISERFS_SB(s)->s_alloc_options.bits = 0; /* clear default settings */ 534 535 while ((this_char = strsep(&options, ":")) != NULL) { 536 if ((value = strchr(this_char, '=')) != NULL) 537 *value++ = 0; 538 539 if (!strcmp(this_char, "concentrating_formatted_nodes")) { 540 int temp; 541 SET_OPTION(concentrating_formatted_nodes); 542 temp = (value 543 && *value) ? simple_strtoul(value, &value, 544 0) : 10; 545 if (temp <= 0 || temp > 100) { 546 REISERFS_SB(s)->s_alloc_options.border = 10; 547 } else { 548 REISERFS_SB(s)->s_alloc_options.border = 549 100 / temp; 550 } 551 continue; 552 } 553 if (!strcmp(this_char, "displacing_large_files")) { 554 SET_OPTION(displacing_large_files); 555 REISERFS_SB(s)->s_alloc_options.large_file_size = 556 (value 557 && *value) ? simple_strtoul(value, &value, 0) : 16; 558 continue; 559 } 560 if (!strcmp(this_char, "displacing_new_packing_localities")) { 561 SET_OPTION(displacing_new_packing_localities); 562 continue; 563 }; 564 565 if (!strcmp(this_char, "old_hashed_relocation")) { 566 SET_OPTION(old_hashed_relocation); 567 continue; 568 } 569 570 if (!strcmp(this_char, "new_hashed_relocation")) { 571 SET_OPTION(new_hashed_relocation); 572 continue; 573 } 574 575 if (!strcmp(this_char, "dirid_groups")) { 576 SET_OPTION(dirid_groups); 577 continue; 578 } 579 if (!strcmp(this_char, "oid_groups")) { 580 SET_OPTION(oid_groups); 581 continue; 582 } 583 if (!strcmp(this_char, "packing_groups")) { 584 SET_OPTION(packing_groups); 585 continue; 586 } 587 if (!strcmp(this_char, "hashed_formatted_nodes")) { 588 SET_OPTION(hashed_formatted_nodes); 589 continue; 590 } 591 592 if (!strcmp(this_char, "skip_busy")) { 593 SET_OPTION(skip_busy); 594 continue; 595 } 596 597 if (!strcmp(this_char, "hundredth_slices")) { 598 SET_OPTION(hundredth_slices); 599 continue; 600 } 601 602 if (!strcmp(this_char, "old_way")) { 603 SET_OPTION(old_way); 604 continue; 605 } 606 607 if (!strcmp(this_char, "displace_based_on_dirid")) { 608 SET_OPTION(displace_based_on_dirid); 609 continue; 610 } 611 612 if (!strcmp(this_char, "preallocmin")) { 613 REISERFS_SB(s)->s_alloc_options.preallocmin = 614 (value 615 && *value) ? simple_strtoul(value, &value, 0) : 4; 616 continue; 617 } 618 619 if (!strcmp(this_char, "preallocsize")) { 620 REISERFS_SB(s)->s_alloc_options.preallocsize = 621 (value 622 && *value) ? simple_strtoul(value, &value, 623 0) : 624 PREALLOCATION_SIZE; 625 continue; 626 } 627 628 reiserfs_warning(s, "zam-4001", "unknown option - %s", 629 this_char); 630 return 1; 631 } 632 633 reiserfs_info(s, "allocator options = [%08x]\n", SB_ALLOC_OPTS(s)); 634 return 0; 635 } 636 637 static inline void new_hashed_relocation(reiserfs_blocknr_hint_t * hint) 638 { 639 char *hash_in; 640 if (hint->formatted_node) { 641 hash_in = (char *)&hint->key.k_dir_id; 642 } else { 643 if (!hint->inode) { 644 //hint->search_start = hint->beg; 645 hash_in = (char *)&hint->key.k_dir_id; 646 } else 647 if (TEST_OPTION(displace_based_on_dirid, hint->th->t_super)) 648 hash_in = (char *)(&INODE_PKEY(hint->inode)->k_dir_id); 649 else 650 hash_in = 651 (char *)(&INODE_PKEY(hint->inode)->k_objectid); 652 } 653 654 hint->search_start = 655 hint->beg + keyed_hash(hash_in, 4) % (hint->end - hint->beg); 656 } 657 658 /* 659 * Relocation based on dirid, hashing them into a given bitmap block 660 * files. Formatted nodes are unaffected, a separate policy covers them 661 */ 662 static void dirid_groups(reiserfs_blocknr_hint_t * hint) 663 { 664 unsigned long hash; 665 __u32 dirid = 0; 666 int bm = 0; 667 struct super_block *sb = hint->th->t_super; 668 if (hint->inode) 669 dirid = le32_to_cpu(INODE_PKEY(hint->inode)->k_dir_id); 670 else if (hint->formatted_node) 671 dirid = hint->key.k_dir_id; 672 673 if (dirid) { 674 bm = bmap_hash_id(sb, dirid); 675 hash = bm * (sb->s_blocksize << 3); 676 /* give a portion of the block group to metadata */ 677 if (hint->inode) 678 hash += sb->s_blocksize / 2; 679 hint->search_start = hash; 680 } 681 } 682 683 /* 684 * Relocation based on oid, hashing them into a given bitmap block 685 * files. Formatted nodes are unaffected, a separate policy covers them 686 */ 687 static void oid_groups(reiserfs_blocknr_hint_t * hint) 688 { 689 if (hint->inode) { 690 unsigned long hash; 691 __u32 oid; 692 __u32 dirid; 693 int bm; 694 695 dirid = le32_to_cpu(INODE_PKEY(hint->inode)->k_dir_id); 696 697 /* keep the root dir and it's first set of subdirs close to 698 * the start of the disk 699 */ 700 if (dirid <= 2) 701 hash = (hint->inode->i_sb->s_blocksize << 3); 702 else { 703 oid = le32_to_cpu(INODE_PKEY(hint->inode)->k_objectid); 704 bm = bmap_hash_id(hint->inode->i_sb, oid); 705 hash = bm * (hint->inode->i_sb->s_blocksize << 3); 706 } 707 hint->search_start = hash; 708 } 709 } 710 711 /* returns 1 if it finds an indirect item and gets valid hint info 712 * from it, otherwise 0 713 */ 714 static int get_left_neighbor(reiserfs_blocknr_hint_t * hint) 715 { 716 struct treepath *path; 717 struct buffer_head *bh; 718 struct item_head *ih; 719 int pos_in_item; 720 __le32 *item; 721 int ret = 0; 722 723 if (!hint->path) /* reiserfs code can call this function w/o pointer to path 724 * structure supplied; then we rely on supplied search_start */ 725 return 0; 726 727 path = hint->path; 728 bh = get_last_bh(path); 729 RFALSE(!bh, "green-4002: Illegal path specified to get_left_neighbor"); 730 ih = get_ih(path); 731 pos_in_item = path->pos_in_item; 732 item = get_item(path); 733 734 hint->search_start = bh->b_blocknr; 735 736 if (!hint->formatted_node && is_indirect_le_ih(ih)) { 737 /* for indirect item: go to left and look for the first non-hole entry 738 in the indirect item */ 739 if (pos_in_item == I_UNFM_NUM(ih)) 740 pos_in_item--; 741 // pos_in_item = I_UNFM_NUM (ih) - 1; 742 while (pos_in_item >= 0) { 743 int t = get_block_num(item, pos_in_item); 744 if (t) { 745 hint->search_start = t; 746 ret = 1; 747 break; 748 } 749 pos_in_item--; 750 } 751 } 752 753 /* does result value fit into specified region? */ 754 return ret; 755 } 756 757 /* should be, if formatted node, then try to put on first part of the device 758 specified as number of percent with mount option device, else try to put 759 on last of device. This is not to say it is good code to do so, 760 but the effect should be measured. */ 761 static inline void set_border_in_hint(struct super_block *s, 762 reiserfs_blocknr_hint_t * hint) 763 { 764 b_blocknr_t border = 765 SB_BLOCK_COUNT(s) / REISERFS_SB(s)->s_alloc_options.border; 766 767 if (hint->formatted_node) 768 hint->end = border - 1; 769 else 770 hint->beg = border; 771 } 772 773 static inline void displace_large_file(reiserfs_blocknr_hint_t * hint) 774 { 775 if (TEST_OPTION(displace_based_on_dirid, hint->th->t_super)) 776 hint->search_start = 777 hint->beg + 778 keyed_hash((char *)(&INODE_PKEY(hint->inode)->k_dir_id), 779 4) % (hint->end - hint->beg); 780 else 781 hint->search_start = 782 hint->beg + 783 keyed_hash((char *)(&INODE_PKEY(hint->inode)->k_objectid), 784 4) % (hint->end - hint->beg); 785 } 786 787 static inline void hash_formatted_node(reiserfs_blocknr_hint_t * hint) 788 { 789 char *hash_in; 790 791 if (!hint->inode) 792 hash_in = (char *)&hint->key.k_dir_id; 793 else if (TEST_OPTION(displace_based_on_dirid, hint->th->t_super)) 794 hash_in = (char *)(&INODE_PKEY(hint->inode)->k_dir_id); 795 else 796 hash_in = (char *)(&INODE_PKEY(hint->inode)->k_objectid); 797 798 hint->search_start = 799 hint->beg + keyed_hash(hash_in, 4) % (hint->end - hint->beg); 800 } 801 802 static inline int 803 this_blocknr_allocation_would_make_it_a_large_file(reiserfs_blocknr_hint_t * 804 hint) 805 { 806 return hint->block == 807 REISERFS_SB(hint->th->t_super)->s_alloc_options.large_file_size; 808 } 809 810 #ifdef DISPLACE_NEW_PACKING_LOCALITIES 811 static inline void displace_new_packing_locality(reiserfs_blocknr_hint_t * hint) 812 { 813 struct in_core_key *key = &hint->key; 814 815 hint->th->displace_new_blocks = 0; 816 hint->search_start = 817 hint->beg + keyed_hash((char *)(&key->k_objectid), 818 4) % (hint->end - hint->beg); 819 } 820 #endif 821 822 static inline int old_hashed_relocation(reiserfs_blocknr_hint_t * hint) 823 { 824 b_blocknr_t border; 825 u32 hash_in; 826 827 if (hint->formatted_node || hint->inode == NULL) { 828 return 0; 829 } 830 831 hash_in = le32_to_cpu((INODE_PKEY(hint->inode))->k_dir_id); 832 border = 833 hint->beg + (u32) keyed_hash(((char *)(&hash_in)), 834 4) % (hint->end - hint->beg - 1); 835 if (border > hint->search_start) 836 hint->search_start = border; 837 838 return 1; 839 } 840 841 static inline int old_way(reiserfs_blocknr_hint_t * hint) 842 { 843 b_blocknr_t border; 844 845 if (hint->formatted_node || hint->inode == NULL) { 846 return 0; 847 } 848 849 border = 850 hint->beg + 851 le32_to_cpu(INODE_PKEY(hint->inode)->k_dir_id) % (hint->end - 852 hint->beg); 853 if (border > hint->search_start) 854 hint->search_start = border; 855 856 return 1; 857 } 858 859 static inline void hundredth_slices(reiserfs_blocknr_hint_t * hint) 860 { 861 struct in_core_key *key = &hint->key; 862 b_blocknr_t slice_start; 863 864 slice_start = 865 (keyed_hash((char *)(&key->k_dir_id), 4) % 100) * (hint->end / 100); 866 if (slice_start > hint->search_start 867 || slice_start + (hint->end / 100) <= hint->search_start) { 868 hint->search_start = slice_start; 869 } 870 } 871 872 static void determine_search_start(reiserfs_blocknr_hint_t * hint, 873 int amount_needed) 874 { 875 struct super_block *s = hint->th->t_super; 876 int unfm_hint; 877 878 hint->beg = 0; 879 hint->end = SB_BLOCK_COUNT(s) - 1; 880 881 /* This is former border algorithm. Now with tunable border offset */ 882 if (concentrating_formatted_nodes(s)) 883 set_border_in_hint(s, hint); 884 885 #ifdef DISPLACE_NEW_PACKING_LOCALITIES 886 /* whenever we create a new directory, we displace it. At first we will 887 hash for location, later we might look for a moderately empty place for 888 it */ 889 if (displacing_new_packing_localities(s) 890 && hint->th->displace_new_blocks) { 891 displace_new_packing_locality(hint); 892 893 /* we do not continue determine_search_start, 894 * if new packing locality is being displaced */ 895 return; 896 } 897 #endif 898 899 /* all persons should feel encouraged to add more special cases here and 900 * test them */ 901 902 if (displacing_large_files(s) && !hint->formatted_node 903 && this_blocknr_allocation_would_make_it_a_large_file(hint)) { 904 displace_large_file(hint); 905 return; 906 } 907 908 /* if none of our special cases is relevant, use the left neighbor in the 909 tree order of the new node we are allocating for */ 910 if (hint->formatted_node && TEST_OPTION(hashed_formatted_nodes, s)) { 911 hash_formatted_node(hint); 912 return; 913 } 914 915 unfm_hint = get_left_neighbor(hint); 916 917 /* Mimic old block allocator behaviour, that is if VFS allowed for preallocation, 918 new blocks are displaced based on directory ID. Also, if suggested search_start 919 is less than last preallocated block, we start searching from it, assuming that 920 HDD dataflow is faster in forward direction */ 921 if (TEST_OPTION(old_way, s)) { 922 if (!hint->formatted_node) { 923 if (!reiserfs_hashed_relocation(s)) 924 old_way(hint); 925 else if (!reiserfs_no_unhashed_relocation(s)) 926 old_hashed_relocation(hint); 927 928 if (hint->inode 929 && hint->search_start < 930 REISERFS_I(hint->inode)->i_prealloc_block) 931 hint->search_start = 932 REISERFS_I(hint->inode)->i_prealloc_block; 933 } 934 return; 935 } 936 937 /* This is an approach proposed by Hans */ 938 if (TEST_OPTION(hundredth_slices, s) 939 && !(displacing_large_files(s) && !hint->formatted_node)) { 940 hundredth_slices(hint); 941 return; 942 } 943 944 /* old_hashed_relocation only works on unformatted */ 945 if (!unfm_hint && !hint->formatted_node && 946 TEST_OPTION(old_hashed_relocation, s)) { 947 old_hashed_relocation(hint); 948 } 949 /* new_hashed_relocation works with both formatted/unformatted nodes */ 950 if ((!unfm_hint || hint->formatted_node) && 951 TEST_OPTION(new_hashed_relocation, s)) { 952 new_hashed_relocation(hint); 953 } 954 /* dirid grouping works only on unformatted nodes */ 955 if (!unfm_hint && !hint->formatted_node && TEST_OPTION(dirid_groups, s)) { 956 dirid_groups(hint); 957 } 958 #ifdef DISPLACE_NEW_PACKING_LOCALITIES 959 if (hint->formatted_node && TEST_OPTION(dirid_groups, s)) { 960 dirid_groups(hint); 961 } 962 #endif 963 964 /* oid grouping works only on unformatted nodes */ 965 if (!unfm_hint && !hint->formatted_node && TEST_OPTION(oid_groups, s)) { 966 oid_groups(hint); 967 } 968 return; 969 } 970 971 static int determine_prealloc_size(reiserfs_blocknr_hint_t * hint) 972 { 973 /* make minimum size a mount option and benchmark both ways */ 974 /* we preallocate blocks only for regular files, specific size */ 975 /* benchmark preallocating always and see what happens */ 976 977 hint->prealloc_size = 0; 978 979 if (!hint->formatted_node && hint->preallocate) { 980 if (S_ISREG(hint->inode->i_mode) 981 && hint->inode->i_size >= 982 REISERFS_SB(hint->th->t_super)->s_alloc_options. 983 preallocmin * hint->inode->i_sb->s_blocksize) 984 hint->prealloc_size = 985 REISERFS_SB(hint->th->t_super)->s_alloc_options. 986 preallocsize - 1; 987 } 988 return CARRY_ON; 989 } 990 991 /* XXX I know it could be merged with upper-level function; 992 but may be result function would be too complex. */ 993 static inline int allocate_without_wrapping_disk(reiserfs_blocknr_hint_t * hint, 994 b_blocknr_t * new_blocknrs, 995 b_blocknr_t start, 996 b_blocknr_t finish, int min, 997 int amount_needed, 998 int prealloc_size) 999 { 1000 int rest = amount_needed; 1001 int nr_allocated; 1002 1003 while (rest > 0 && start <= finish) { 1004 nr_allocated = scan_bitmap(hint->th, &start, finish, min, 1005 rest + prealloc_size, 1006 !hint->formatted_node, hint->block); 1007 1008 if (nr_allocated == 0) /* no new blocks allocated, return */ 1009 break; 1010 1011 /* fill free_blocknrs array first */ 1012 while (rest > 0 && nr_allocated > 0) { 1013 *new_blocknrs++ = start++; 1014 rest--; 1015 nr_allocated--; 1016 } 1017 1018 /* do we have something to fill prealloc. array also ? */ 1019 if (nr_allocated > 0) { 1020 /* it means prealloc_size was greater that 0 and we do preallocation */ 1021 list_add(&REISERFS_I(hint->inode)->i_prealloc_list, 1022 &SB_JOURNAL(hint->th->t_super)-> 1023 j_prealloc_list); 1024 REISERFS_I(hint->inode)->i_prealloc_block = start; 1025 REISERFS_I(hint->inode)->i_prealloc_count = 1026 nr_allocated; 1027 break; 1028 } 1029 } 1030 1031 return (amount_needed - rest); 1032 } 1033 1034 static inline int blocknrs_and_prealloc_arrays_from_search_start 1035 (reiserfs_blocknr_hint_t * hint, b_blocknr_t * new_blocknrs, 1036 int amount_needed) { 1037 struct super_block *s = hint->th->t_super; 1038 b_blocknr_t start = hint->search_start; 1039 b_blocknr_t finish = SB_BLOCK_COUNT(s) - 1; 1040 int passno = 0; 1041 int nr_allocated = 0; 1042 1043 determine_prealloc_size(hint); 1044 if (!hint->formatted_node) { 1045 int quota_ret; 1046 #ifdef REISERQUOTA_DEBUG 1047 reiserfs_debug(s, REISERFS_DEBUG_CODE, 1048 "reiserquota: allocating %d blocks id=%u", 1049 amount_needed, hint->inode->i_uid); 1050 #endif 1051 quota_ret = 1052 dquot_alloc_block_nodirty(hint->inode, amount_needed); 1053 if (quota_ret) /* Quota exceeded? */ 1054 return QUOTA_EXCEEDED; 1055 if (hint->preallocate && hint->prealloc_size) { 1056 #ifdef REISERQUOTA_DEBUG 1057 reiserfs_debug(s, REISERFS_DEBUG_CODE, 1058 "reiserquota: allocating (prealloc) %d blocks id=%u", 1059 hint->prealloc_size, hint->inode->i_uid); 1060 #endif 1061 quota_ret = dquot_prealloc_block_nodirty(hint->inode, 1062 hint->prealloc_size); 1063 if (quota_ret) 1064 hint->preallocate = hint->prealloc_size = 0; 1065 } 1066 /* for unformatted nodes, force large allocations */ 1067 } 1068 1069 do { 1070 switch (passno++) { 1071 case 0: /* Search from hint->search_start to end of disk */ 1072 start = hint->search_start; 1073 finish = SB_BLOCK_COUNT(s) - 1; 1074 break; 1075 case 1: /* Search from hint->beg to hint->search_start */ 1076 start = hint->beg; 1077 finish = hint->search_start; 1078 break; 1079 case 2: /* Last chance: Search from 0 to hint->beg */ 1080 start = 0; 1081 finish = hint->beg; 1082 break; 1083 default: /* We've tried searching everywhere, not enough space */ 1084 /* Free the blocks */ 1085 if (!hint->formatted_node) { 1086 #ifdef REISERQUOTA_DEBUG 1087 reiserfs_debug(s, REISERFS_DEBUG_CODE, 1088 "reiserquota: freeing (nospace) %d blocks id=%u", 1089 amount_needed + 1090 hint->prealloc_size - 1091 nr_allocated, 1092 hint->inode->i_uid); 1093 #endif 1094 /* Free not allocated blocks */ 1095 dquot_free_block_nodirty(hint->inode, 1096 amount_needed + hint->prealloc_size - 1097 nr_allocated); 1098 } 1099 while (nr_allocated--) 1100 reiserfs_free_block(hint->th, hint->inode, 1101 new_blocknrs[nr_allocated], 1102 !hint->formatted_node); 1103 1104 return NO_DISK_SPACE; 1105 } 1106 } while ((nr_allocated += allocate_without_wrapping_disk(hint, 1107 new_blocknrs + 1108 nr_allocated, 1109 start, finish, 1110 1, 1111 amount_needed - 1112 nr_allocated, 1113 hint-> 1114 prealloc_size)) 1115 < amount_needed); 1116 if (!hint->formatted_node && 1117 amount_needed + hint->prealloc_size > 1118 nr_allocated + REISERFS_I(hint->inode)->i_prealloc_count) { 1119 /* Some of preallocation blocks were not allocated */ 1120 #ifdef REISERQUOTA_DEBUG 1121 reiserfs_debug(s, REISERFS_DEBUG_CODE, 1122 "reiserquota: freeing (failed prealloc) %d blocks id=%u", 1123 amount_needed + hint->prealloc_size - 1124 nr_allocated - 1125 REISERFS_I(hint->inode)->i_prealloc_count, 1126 hint->inode->i_uid); 1127 #endif 1128 dquot_free_block_nodirty(hint->inode, amount_needed + 1129 hint->prealloc_size - nr_allocated - 1130 REISERFS_I(hint->inode)-> 1131 i_prealloc_count); 1132 } 1133 1134 return CARRY_ON; 1135 } 1136 1137 /* grab new blocknrs from preallocated list */ 1138 /* return amount still needed after using them */ 1139 static int use_preallocated_list_if_available(reiserfs_blocknr_hint_t * hint, 1140 b_blocknr_t * new_blocknrs, 1141 int amount_needed) 1142 { 1143 struct inode *inode = hint->inode; 1144 1145 if (REISERFS_I(inode)->i_prealloc_count > 0) { 1146 while (amount_needed) { 1147 1148 *new_blocknrs++ = REISERFS_I(inode)->i_prealloc_block++; 1149 REISERFS_I(inode)->i_prealloc_count--; 1150 1151 amount_needed--; 1152 1153 if (REISERFS_I(inode)->i_prealloc_count <= 0) { 1154 list_del(&REISERFS_I(inode)->i_prealloc_list); 1155 break; 1156 } 1157 } 1158 } 1159 /* return amount still needed after using preallocated blocks */ 1160 return amount_needed; 1161 } 1162 1163 int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t * hint, b_blocknr_t * new_blocknrs, int amount_needed, int reserved_by_us /* Amount of blocks we have 1164 already reserved */ ) 1165 { 1166 int initial_amount_needed = amount_needed; 1167 int ret; 1168 struct super_block *s = hint->th->t_super; 1169 1170 /* Check if there is enough space, taking into account reserved space */ 1171 if (SB_FREE_BLOCKS(s) - REISERFS_SB(s)->reserved_blocks < 1172 amount_needed - reserved_by_us) 1173 return NO_DISK_SPACE; 1174 /* should this be if !hint->inode && hint->preallocate? */ 1175 /* do you mean hint->formatted_node can be removed ? - Zam */ 1176 /* hint->formatted_node cannot be removed because we try to access 1177 inode information here, and there is often no inode assotiated with 1178 metadata allocations - green */ 1179 1180 if (!hint->formatted_node && hint->preallocate) { 1181 amount_needed = use_preallocated_list_if_available 1182 (hint, new_blocknrs, amount_needed); 1183 if (amount_needed == 0) /* all blocknrs we need we got from 1184 prealloc. list */ 1185 return CARRY_ON; 1186 new_blocknrs += (initial_amount_needed - amount_needed); 1187 } 1188 1189 /* find search start and save it in hint structure */ 1190 determine_search_start(hint, amount_needed); 1191 if (hint->search_start >= SB_BLOCK_COUNT(s)) 1192 hint->search_start = SB_BLOCK_COUNT(s) - 1; 1193 1194 /* allocation itself; fill new_blocknrs and preallocation arrays */ 1195 ret = blocknrs_and_prealloc_arrays_from_search_start 1196 (hint, new_blocknrs, amount_needed); 1197 1198 /* we used prealloc. list to fill (partially) new_blocknrs array. If final allocation fails we 1199 * need to return blocks back to prealloc. list or just free them. -- Zam (I chose second 1200 * variant) */ 1201 1202 if (ret != CARRY_ON) { 1203 while (amount_needed++ < initial_amount_needed) { 1204 reiserfs_free_block(hint->th, hint->inode, 1205 *(--new_blocknrs), 1); 1206 } 1207 } 1208 return ret; 1209 } 1210 1211 void reiserfs_cache_bitmap_metadata(struct super_block *sb, 1212 struct buffer_head *bh, 1213 struct reiserfs_bitmap_info *info) 1214 { 1215 unsigned long *cur = (unsigned long *)(bh->b_data + bh->b_size); 1216 1217 /* The first bit must ALWAYS be 1 */ 1218 if (!reiserfs_test_le_bit(0, (unsigned long *)bh->b_data)) 1219 reiserfs_error(sb, "reiserfs-2025", "bitmap block %lu is " 1220 "corrupted: first bit must be 1", bh->b_blocknr); 1221 1222 info->free_count = 0; 1223 1224 while (--cur >= (unsigned long *)bh->b_data) { 1225 int i; 1226 1227 /* 0 and ~0 are special, we can optimize for them */ 1228 if (*cur == 0) 1229 info->free_count += BITS_PER_LONG; 1230 else if (*cur != ~0L) /* A mix, investigate */ 1231 for (i = BITS_PER_LONG - 1; i >= 0; i--) 1232 if (!reiserfs_test_le_bit(i, cur)) 1233 info->free_count++; 1234 } 1235 } 1236 1237 struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb, 1238 unsigned int bitmap) 1239 { 1240 b_blocknr_t block = (sb->s_blocksize << 3) * bitmap; 1241 struct reiserfs_bitmap_info *info = SB_AP_BITMAP(sb) + bitmap; 1242 struct buffer_head *bh; 1243 1244 /* Way old format filesystems had the bitmaps packed up front. 1245 * I doubt there are any of these left, but just in case... */ 1246 if (unlikely(test_bit(REISERFS_OLD_FORMAT, 1247 &(REISERFS_SB(sb)->s_properties)))) 1248 block = REISERFS_SB(sb)->s_sbh->b_blocknr + 1 + bitmap; 1249 else if (bitmap == 0) 1250 block = (REISERFS_DISK_OFFSET_IN_BYTES >> sb->s_blocksize_bits) + 1; 1251 1252 reiserfs_write_unlock(sb); 1253 bh = sb_bread(sb, block); 1254 reiserfs_write_lock(sb); 1255 if (bh == NULL) 1256 reiserfs_warning(sb, "sh-2029: %s: bitmap block (#%u) " 1257 "reading failed", __func__, block); 1258 else { 1259 if (buffer_locked(bh)) { 1260 PROC_INFO_INC(sb, scan_bitmap.wait); 1261 reiserfs_write_unlock(sb); 1262 __wait_on_buffer(bh); 1263 reiserfs_write_lock(sb); 1264 } 1265 BUG_ON(!buffer_uptodate(bh)); 1266 BUG_ON(atomic_read(&bh->b_count) == 0); 1267 1268 if (info->free_count == UINT_MAX) 1269 reiserfs_cache_bitmap_metadata(sb, bh, info); 1270 } 1271 1272 return bh; 1273 } 1274 1275 int reiserfs_init_bitmap_cache(struct super_block *sb) 1276 { 1277 struct reiserfs_bitmap_info *bitmap; 1278 unsigned int bmap_nr = reiserfs_bmap_count(sb); 1279 1280 /* Avoid lock recursion in fault case */ 1281 reiserfs_write_unlock(sb); 1282 bitmap = vmalloc(sizeof(*bitmap) * bmap_nr); 1283 reiserfs_write_lock(sb); 1284 if (bitmap == NULL) 1285 return -ENOMEM; 1286 1287 memset(bitmap, 0xff, sizeof(*bitmap) * bmap_nr); 1288 1289 SB_AP_BITMAP(sb) = bitmap; 1290 1291 return 0; 1292 } 1293 1294 void reiserfs_free_bitmap_cache(struct super_block *sb) 1295 { 1296 if (SB_AP_BITMAP(sb)) { 1297 vfree(SB_AP_BITMAP(sb)); 1298 SB_AP_BITMAP(sb) = NULL; 1299 } 1300 } 1301