1 // SPDX-License-Identifier: GPL-2.0 2 /* 3 * fs/ext4/extents_status.c 4 * 5 * Written by Yongqiang Yang <xiaoqiangnk@gmail.com> 6 * Modified by 7 * Allison Henderson <achender@linux.vnet.ibm.com> 8 * Hugh Dickins <hughd@google.com> 9 * Zheng Liu <wenqing.lz@taobao.com> 10 * 11 * Ext4 extents status tree core functions. 12 */ 13 #include <linux/list_sort.h> 14 #include <linux/proc_fs.h> 15 #include <linux/seq_file.h> 16 #include "ext4.h" 17 18 #include <trace/events/ext4.h> 19 20 /* 21 * According to previous discussion in Ext4 Developer Workshop, we 22 * will introduce a new structure called io tree to track all extent 23 * status in order to solve some problems that we have met 24 * (e.g. Reservation space warning), and provide extent-level locking. 25 * Delay extent tree is the first step to achieve this goal. It is 26 * original built by Yongqiang Yang. At that time it is called delay 27 * extent tree, whose goal is only track delayed extents in memory to 28 * simplify the implementation of fiemap and bigalloc, and introduce 29 * lseek SEEK_DATA/SEEK_HOLE support. That is why it is still called 30 * delay extent tree at the first commit. But for better understand 31 * what it does, it has been rename to extent status tree. 32 * 33 * Step1: 34 * Currently the first step has been done. All delayed extents are 35 * tracked in the tree. It maintains the delayed extent when a delayed 36 * allocation is issued, and the delayed extent is written out or 37 * invalidated. Therefore the implementation of fiemap and bigalloc 38 * are simplified, and SEEK_DATA/SEEK_HOLE are introduced. 39 * 40 * The following comment describes the implemenmtation of extent 41 * status tree and future works. 42 * 43 * Step2: 44 * In this step all extent status are tracked by extent status tree. 45 * Thus, we can first try to lookup a block mapping in this tree before 46 * finding it in extent tree. Hence, single extent cache can be removed 47 * because extent status tree can do a better job. Extents in status 48 * tree are loaded on-demand. Therefore, the extent status tree may not 49 * contain all of the extents in a file. Meanwhile we define a shrinker 50 * to reclaim memory from extent status tree because fragmented extent 51 * tree will make status tree cost too much memory. written/unwritten/- 52 * hole extents in the tree will be reclaimed by this shrinker when we 53 * are under high memory pressure. Delayed extents will not be 54 * reclimed because fiemap, bigalloc, and seek_data/hole need it. 55 */ 56 57 /* 58 * Extent status tree implementation for ext4. 59 * 60 * 61 * ========================================================================== 62 * Extent status tree tracks all extent status. 63 * 64 * 1. Why we need to implement extent status tree? 65 * 66 * Without extent status tree, ext4 identifies a delayed extent by looking 67 * up page cache, this has several deficiencies - complicated, buggy, 68 * and inefficient code. 69 * 70 * FIEMAP, SEEK_HOLE/DATA, bigalloc, and writeout all need to know if a 71 * block or a range of blocks are belonged to a delayed extent. 72 * 73 * Let us have a look at how they do without extent status tree. 74 * -- FIEMAP 75 * FIEMAP looks up page cache to identify delayed allocations from holes. 76 * 77 * -- SEEK_HOLE/DATA 78 * SEEK_HOLE/DATA has the same problem as FIEMAP. 79 * 80 * -- bigalloc 81 * bigalloc looks up page cache to figure out if a block is 82 * already under delayed allocation or not to determine whether 83 * quota reserving is needed for the cluster. 84 * 85 * -- writeout 86 * Writeout looks up whole page cache to see if a buffer is 87 * mapped, If there are not very many delayed buffers, then it is 88 * time consuming. 89 * 90 * With extent status tree implementation, FIEMAP, SEEK_HOLE/DATA, 91 * bigalloc and writeout can figure out if a block or a range of 92 * blocks is under delayed allocation(belonged to a delayed extent) or 93 * not by searching the extent tree. 94 * 95 * 96 * ========================================================================== 97 * 2. Ext4 extent status tree impelmentation 98 * 99 * -- extent 100 * A extent is a range of blocks which are contiguous logically and 101 * physically. Unlike extent in extent tree, this extent in ext4 is 102 * a in-memory struct, there is no corresponding on-disk data. There 103 * is no limit on length of extent, so an extent can contain as many 104 * blocks as they are contiguous logically and physically. 105 * 106 * -- extent status tree 107 * Every inode has an extent status tree and all allocation blocks 108 * are added to the tree with different status. The extent in the 109 * tree are ordered by logical block no. 110 * 111 * -- operations on a extent status tree 112 * There are three important operations on a delayed extent tree: find 113 * next extent, adding a extent(a range of blocks) and removing a extent. 114 * 115 * -- race on a extent status tree 116 * Extent status tree is protected by inode->i_es_lock. 117 * 118 * -- memory consumption 119 * Fragmented extent tree will make extent status tree cost too much 120 * memory. Hence, we will reclaim written/unwritten/hole extents from 121 * the tree under a heavy memory pressure. 122 * 123 * 124 * ========================================================================== 125 * 3. Performance analysis 126 * 127 * -- overhead 128 * 1. There is a cache extent for write access, so if writes are 129 * not very random, adding space operaions are in O(1) time. 130 * 131 * -- gain 132 * 2. Code is much simpler, more readable, more maintainable and 133 * more efficient. 134 * 135 * 136 * ========================================================================== 137 * 4. TODO list 138 * 139 * -- Refactor delayed space reservation 140 * 141 * -- Extent-level locking 142 */ 143 144 static struct kmem_cache *ext4_es_cachep; 145 static struct kmem_cache *ext4_pending_cachep; 146 147 static int __es_insert_extent(struct inode *inode, struct extent_status *newes); 148 static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk, 149 ext4_lblk_t end, int *reserved); 150 static int es_reclaim_extents(struct ext4_inode_info *ei, int *nr_to_scan); 151 static int __es_shrink(struct ext4_sb_info *sbi, int nr_to_scan, 152 struct ext4_inode_info *locked_ei); 153 static void __revise_pending(struct inode *inode, ext4_lblk_t lblk, 154 ext4_lblk_t len); 155 156 int __init ext4_init_es(void) 157 { 158 ext4_es_cachep = kmem_cache_create("ext4_extent_status", 159 sizeof(struct extent_status), 160 0, (SLAB_RECLAIM_ACCOUNT), NULL); 161 if (ext4_es_cachep == NULL) 162 return -ENOMEM; 163 return 0; 164 } 165 166 void ext4_exit_es(void) 167 { 168 kmem_cache_destroy(ext4_es_cachep); 169 } 170 171 void ext4_es_init_tree(struct ext4_es_tree *tree) 172 { 173 tree->root = RB_ROOT; 174 tree->cache_es = NULL; 175 } 176 177 #ifdef ES_DEBUG__ 178 static void ext4_es_print_tree(struct inode *inode) 179 { 180 struct ext4_es_tree *tree; 181 struct rb_node *node; 182 183 printk(KERN_DEBUG "status extents for inode %lu:", inode->i_ino); 184 tree = &EXT4_I(inode)->i_es_tree; 185 node = rb_first(&tree->root); 186 while (node) { 187 struct extent_status *es; 188 es = rb_entry(node, struct extent_status, rb_node); 189 printk(KERN_DEBUG " [%u/%u) %llu %x", 190 es->es_lblk, es->es_len, 191 ext4_es_pblock(es), ext4_es_status(es)); 192 node = rb_next(node); 193 } 194 printk(KERN_DEBUG "\n"); 195 } 196 #else 197 #define ext4_es_print_tree(inode) 198 #endif 199 200 static inline ext4_lblk_t ext4_es_end(struct extent_status *es) 201 { 202 BUG_ON(es->es_lblk + es->es_len < es->es_lblk); 203 return es->es_lblk + es->es_len - 1; 204 } 205 206 /* 207 * search through the tree for an delayed extent with a given offset. If 208 * it can't be found, try to find next extent. 209 */ 210 static struct extent_status *__es_tree_search(struct rb_root *root, 211 ext4_lblk_t lblk) 212 { 213 struct rb_node *node = root->rb_node; 214 struct extent_status *es = NULL; 215 216 while (node) { 217 es = rb_entry(node, struct extent_status, rb_node); 218 if (lblk < es->es_lblk) 219 node = node->rb_left; 220 else if (lblk > ext4_es_end(es)) 221 node = node->rb_right; 222 else 223 return es; 224 } 225 226 if (es && lblk < es->es_lblk) 227 return es; 228 229 if (es && lblk > ext4_es_end(es)) { 230 node = rb_next(&es->rb_node); 231 return node ? rb_entry(node, struct extent_status, rb_node) : 232 NULL; 233 } 234 235 return NULL; 236 } 237 238 /* 239 * ext4_es_find_extent_range - find extent with specified status within block 240 * range or next extent following block range in 241 * extents status tree 242 * 243 * @inode - file containing the range 244 * @matching_fn - pointer to function that matches extents with desired status 245 * @lblk - logical block defining start of range 246 * @end - logical block defining end of range 247 * @es - extent found, if any 248 * 249 * Find the first extent within the block range specified by @lblk and @end 250 * in the extents status tree that satisfies @matching_fn. If a match 251 * is found, it's returned in @es. If not, and a matching extent is found 252 * beyond the block range, it's returned in @es. If no match is found, an 253 * extent is returned in @es whose es_lblk, es_len, and es_pblk components 254 * are 0. 255 */ 256 static void __es_find_extent_range(struct inode *inode, 257 int (*matching_fn)(struct extent_status *es), 258 ext4_lblk_t lblk, ext4_lblk_t end, 259 struct extent_status *es) 260 { 261 struct ext4_es_tree *tree = NULL; 262 struct extent_status *es1 = NULL; 263 struct rb_node *node; 264 265 WARN_ON(es == NULL); 266 WARN_ON(end < lblk); 267 268 tree = &EXT4_I(inode)->i_es_tree; 269 270 /* see if the extent has been cached */ 271 es->es_lblk = es->es_len = es->es_pblk = 0; 272 if (tree->cache_es) { 273 es1 = tree->cache_es; 274 if (in_range(lblk, es1->es_lblk, es1->es_len)) { 275 es_debug("%u cached by [%u/%u) %llu %x\n", 276 lblk, es1->es_lblk, es1->es_len, 277 ext4_es_pblock(es1), ext4_es_status(es1)); 278 goto out; 279 } 280 } 281 282 es1 = __es_tree_search(&tree->root, lblk); 283 284 out: 285 if (es1 && !matching_fn(es1)) { 286 while ((node = rb_next(&es1->rb_node)) != NULL) { 287 es1 = rb_entry(node, struct extent_status, rb_node); 288 if (es1->es_lblk > end) { 289 es1 = NULL; 290 break; 291 } 292 if (matching_fn(es1)) 293 break; 294 } 295 } 296 297 if (es1 && matching_fn(es1)) { 298 tree->cache_es = es1; 299 es->es_lblk = es1->es_lblk; 300 es->es_len = es1->es_len; 301 es->es_pblk = es1->es_pblk; 302 } 303 304 } 305 306 /* 307 * Locking for __es_find_extent_range() for external use 308 */ 309 void ext4_es_find_extent_range(struct inode *inode, 310 int (*matching_fn)(struct extent_status *es), 311 ext4_lblk_t lblk, ext4_lblk_t end, 312 struct extent_status *es) 313 { 314 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 315 return; 316 317 trace_ext4_es_find_extent_range_enter(inode, lblk); 318 319 read_lock(&EXT4_I(inode)->i_es_lock); 320 __es_find_extent_range(inode, matching_fn, lblk, end, es); 321 read_unlock(&EXT4_I(inode)->i_es_lock); 322 323 trace_ext4_es_find_extent_range_exit(inode, es); 324 } 325 326 /* 327 * __es_scan_range - search block range for block with specified status 328 * in extents status tree 329 * 330 * @inode - file containing the range 331 * @matching_fn - pointer to function that matches extents with desired status 332 * @lblk - logical block defining start of range 333 * @end - logical block defining end of range 334 * 335 * Returns true if at least one block in the specified block range satisfies 336 * the criterion specified by @matching_fn, and false if not. If at least 337 * one extent has the specified status, then there is at least one block 338 * in the cluster with that status. Should only be called by code that has 339 * taken i_es_lock. 340 */ 341 static bool __es_scan_range(struct inode *inode, 342 int (*matching_fn)(struct extent_status *es), 343 ext4_lblk_t start, ext4_lblk_t end) 344 { 345 struct extent_status es; 346 347 __es_find_extent_range(inode, matching_fn, start, end, &es); 348 if (es.es_len == 0) 349 return false; /* no matching extent in the tree */ 350 else if (es.es_lblk <= start && 351 start < es.es_lblk + es.es_len) 352 return true; 353 else if (start <= es.es_lblk && es.es_lblk <= end) 354 return true; 355 else 356 return false; 357 } 358 /* 359 * Locking for __es_scan_range() for external use 360 */ 361 bool ext4_es_scan_range(struct inode *inode, 362 int (*matching_fn)(struct extent_status *es), 363 ext4_lblk_t lblk, ext4_lblk_t end) 364 { 365 bool ret; 366 367 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 368 return false; 369 370 read_lock(&EXT4_I(inode)->i_es_lock); 371 ret = __es_scan_range(inode, matching_fn, lblk, end); 372 read_unlock(&EXT4_I(inode)->i_es_lock); 373 374 return ret; 375 } 376 377 /* 378 * __es_scan_clu - search cluster for block with specified status in 379 * extents status tree 380 * 381 * @inode - file containing the cluster 382 * @matching_fn - pointer to function that matches extents with desired status 383 * @lblk - logical block in cluster to be searched 384 * 385 * Returns true if at least one extent in the cluster containing @lblk 386 * satisfies the criterion specified by @matching_fn, and false if not. If at 387 * least one extent has the specified status, then there is at least one block 388 * in the cluster with that status. Should only be called by code that has 389 * taken i_es_lock. 390 */ 391 static bool __es_scan_clu(struct inode *inode, 392 int (*matching_fn)(struct extent_status *es), 393 ext4_lblk_t lblk) 394 { 395 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 396 ext4_lblk_t lblk_start, lblk_end; 397 398 lblk_start = EXT4_LBLK_CMASK(sbi, lblk); 399 lblk_end = lblk_start + sbi->s_cluster_ratio - 1; 400 401 return __es_scan_range(inode, matching_fn, lblk_start, lblk_end); 402 } 403 404 /* 405 * Locking for __es_scan_clu() for external use 406 */ 407 bool ext4_es_scan_clu(struct inode *inode, 408 int (*matching_fn)(struct extent_status *es), 409 ext4_lblk_t lblk) 410 { 411 bool ret; 412 413 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 414 return false; 415 416 read_lock(&EXT4_I(inode)->i_es_lock); 417 ret = __es_scan_clu(inode, matching_fn, lblk); 418 read_unlock(&EXT4_I(inode)->i_es_lock); 419 420 return ret; 421 } 422 423 static void ext4_es_list_add(struct inode *inode) 424 { 425 struct ext4_inode_info *ei = EXT4_I(inode); 426 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 427 428 if (!list_empty(&ei->i_es_list)) 429 return; 430 431 spin_lock(&sbi->s_es_lock); 432 if (list_empty(&ei->i_es_list)) { 433 list_add_tail(&ei->i_es_list, &sbi->s_es_list); 434 sbi->s_es_nr_inode++; 435 } 436 spin_unlock(&sbi->s_es_lock); 437 } 438 439 static void ext4_es_list_del(struct inode *inode) 440 { 441 struct ext4_inode_info *ei = EXT4_I(inode); 442 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 443 444 spin_lock(&sbi->s_es_lock); 445 if (!list_empty(&ei->i_es_list)) { 446 list_del_init(&ei->i_es_list); 447 sbi->s_es_nr_inode--; 448 WARN_ON_ONCE(sbi->s_es_nr_inode < 0); 449 } 450 spin_unlock(&sbi->s_es_lock); 451 } 452 453 static struct extent_status * 454 ext4_es_alloc_extent(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t len, 455 ext4_fsblk_t pblk) 456 { 457 struct extent_status *es; 458 es = kmem_cache_alloc(ext4_es_cachep, GFP_ATOMIC); 459 if (es == NULL) 460 return NULL; 461 es->es_lblk = lblk; 462 es->es_len = len; 463 es->es_pblk = pblk; 464 465 /* 466 * We don't count delayed extent because we never try to reclaim them 467 */ 468 if (!ext4_es_is_delayed(es)) { 469 if (!EXT4_I(inode)->i_es_shk_nr++) 470 ext4_es_list_add(inode); 471 percpu_counter_inc(&EXT4_SB(inode->i_sb)-> 472 s_es_stats.es_stats_shk_cnt); 473 } 474 475 EXT4_I(inode)->i_es_all_nr++; 476 percpu_counter_inc(&EXT4_SB(inode->i_sb)->s_es_stats.es_stats_all_cnt); 477 478 return es; 479 } 480 481 static void ext4_es_free_extent(struct inode *inode, struct extent_status *es) 482 { 483 EXT4_I(inode)->i_es_all_nr--; 484 percpu_counter_dec(&EXT4_SB(inode->i_sb)->s_es_stats.es_stats_all_cnt); 485 486 /* Decrease the shrink counter when this es is not delayed */ 487 if (!ext4_es_is_delayed(es)) { 488 BUG_ON(EXT4_I(inode)->i_es_shk_nr == 0); 489 if (!--EXT4_I(inode)->i_es_shk_nr) 490 ext4_es_list_del(inode); 491 percpu_counter_dec(&EXT4_SB(inode->i_sb)-> 492 s_es_stats.es_stats_shk_cnt); 493 } 494 495 kmem_cache_free(ext4_es_cachep, es); 496 } 497 498 /* 499 * Check whether or not two extents can be merged 500 * Condition: 501 * - logical block number is contiguous 502 * - physical block number is contiguous 503 * - status is equal 504 */ 505 static int ext4_es_can_be_merged(struct extent_status *es1, 506 struct extent_status *es2) 507 { 508 if (ext4_es_type(es1) != ext4_es_type(es2)) 509 return 0; 510 511 if (((__u64) es1->es_len) + es2->es_len > EXT_MAX_BLOCKS) { 512 pr_warn("ES assertion failed when merging extents. " 513 "The sum of lengths of es1 (%d) and es2 (%d) " 514 "is bigger than allowed file size (%d)\n", 515 es1->es_len, es2->es_len, EXT_MAX_BLOCKS); 516 WARN_ON(1); 517 return 0; 518 } 519 520 if (((__u64) es1->es_lblk) + es1->es_len != es2->es_lblk) 521 return 0; 522 523 if ((ext4_es_is_written(es1) || ext4_es_is_unwritten(es1)) && 524 (ext4_es_pblock(es1) + es1->es_len == ext4_es_pblock(es2))) 525 return 1; 526 527 if (ext4_es_is_hole(es1)) 528 return 1; 529 530 /* we need to check delayed extent is without unwritten status */ 531 if (ext4_es_is_delayed(es1) && !ext4_es_is_unwritten(es1)) 532 return 1; 533 534 return 0; 535 } 536 537 static struct extent_status * 538 ext4_es_try_to_merge_left(struct inode *inode, struct extent_status *es) 539 { 540 struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree; 541 struct extent_status *es1; 542 struct rb_node *node; 543 544 node = rb_prev(&es->rb_node); 545 if (!node) 546 return es; 547 548 es1 = rb_entry(node, struct extent_status, rb_node); 549 if (ext4_es_can_be_merged(es1, es)) { 550 es1->es_len += es->es_len; 551 if (ext4_es_is_referenced(es)) 552 ext4_es_set_referenced(es1); 553 rb_erase(&es->rb_node, &tree->root); 554 ext4_es_free_extent(inode, es); 555 es = es1; 556 } 557 558 return es; 559 } 560 561 static struct extent_status * 562 ext4_es_try_to_merge_right(struct inode *inode, struct extent_status *es) 563 { 564 struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree; 565 struct extent_status *es1; 566 struct rb_node *node; 567 568 node = rb_next(&es->rb_node); 569 if (!node) 570 return es; 571 572 es1 = rb_entry(node, struct extent_status, rb_node); 573 if (ext4_es_can_be_merged(es, es1)) { 574 es->es_len += es1->es_len; 575 if (ext4_es_is_referenced(es1)) 576 ext4_es_set_referenced(es); 577 rb_erase(node, &tree->root); 578 ext4_es_free_extent(inode, es1); 579 } 580 581 return es; 582 } 583 584 #ifdef ES_AGGRESSIVE_TEST 585 #include "ext4_extents.h" /* Needed when ES_AGGRESSIVE_TEST is defined */ 586 587 static void ext4_es_insert_extent_ext_check(struct inode *inode, 588 struct extent_status *es) 589 { 590 struct ext4_ext_path *path = NULL; 591 struct ext4_extent *ex; 592 ext4_lblk_t ee_block; 593 ext4_fsblk_t ee_start; 594 unsigned short ee_len; 595 int depth, ee_status, es_status; 596 597 path = ext4_find_extent(inode, es->es_lblk, NULL, EXT4_EX_NOCACHE); 598 if (IS_ERR(path)) 599 return; 600 601 depth = ext_depth(inode); 602 ex = path[depth].p_ext; 603 604 if (ex) { 605 606 ee_block = le32_to_cpu(ex->ee_block); 607 ee_start = ext4_ext_pblock(ex); 608 ee_len = ext4_ext_get_actual_len(ex); 609 610 ee_status = ext4_ext_is_unwritten(ex) ? 1 : 0; 611 es_status = ext4_es_is_unwritten(es) ? 1 : 0; 612 613 /* 614 * Make sure ex and es are not overlap when we try to insert 615 * a delayed/hole extent. 616 */ 617 if (!ext4_es_is_written(es) && !ext4_es_is_unwritten(es)) { 618 if (in_range(es->es_lblk, ee_block, ee_len)) { 619 pr_warn("ES insert assertion failed for " 620 "inode: %lu we can find an extent " 621 "at block [%d/%d/%llu/%c], but we " 622 "want to add a delayed/hole extent " 623 "[%d/%d/%llu/%x]\n", 624 inode->i_ino, ee_block, ee_len, 625 ee_start, ee_status ? 'u' : 'w', 626 es->es_lblk, es->es_len, 627 ext4_es_pblock(es), ext4_es_status(es)); 628 } 629 goto out; 630 } 631 632 /* 633 * We don't check ee_block == es->es_lblk, etc. because es 634 * might be a part of whole extent, vice versa. 635 */ 636 if (es->es_lblk < ee_block || 637 ext4_es_pblock(es) != ee_start + es->es_lblk - ee_block) { 638 pr_warn("ES insert assertion failed for inode: %lu " 639 "ex_status [%d/%d/%llu/%c] != " 640 "es_status [%d/%d/%llu/%c]\n", inode->i_ino, 641 ee_block, ee_len, ee_start, 642 ee_status ? 'u' : 'w', es->es_lblk, es->es_len, 643 ext4_es_pblock(es), es_status ? 'u' : 'w'); 644 goto out; 645 } 646 647 if (ee_status ^ es_status) { 648 pr_warn("ES insert assertion failed for inode: %lu " 649 "ex_status [%d/%d/%llu/%c] != " 650 "es_status [%d/%d/%llu/%c]\n", inode->i_ino, 651 ee_block, ee_len, ee_start, 652 ee_status ? 'u' : 'w', es->es_lblk, es->es_len, 653 ext4_es_pblock(es), es_status ? 'u' : 'w'); 654 } 655 } else { 656 /* 657 * We can't find an extent on disk. So we need to make sure 658 * that we don't want to add an written/unwritten extent. 659 */ 660 if (!ext4_es_is_delayed(es) && !ext4_es_is_hole(es)) { 661 pr_warn("ES insert assertion failed for inode: %lu " 662 "can't find an extent at block %d but we want " 663 "to add a written/unwritten extent " 664 "[%d/%d/%llu/%x]\n", inode->i_ino, 665 es->es_lblk, es->es_lblk, es->es_len, 666 ext4_es_pblock(es), ext4_es_status(es)); 667 } 668 } 669 out: 670 ext4_ext_drop_refs(path); 671 kfree(path); 672 } 673 674 static void ext4_es_insert_extent_ind_check(struct inode *inode, 675 struct extent_status *es) 676 { 677 struct ext4_map_blocks map; 678 int retval; 679 680 /* 681 * Here we call ext4_ind_map_blocks to lookup a block mapping because 682 * 'Indirect' structure is defined in indirect.c. So we couldn't 683 * access direct/indirect tree from outside. It is too dirty to define 684 * this function in indirect.c file. 685 */ 686 687 map.m_lblk = es->es_lblk; 688 map.m_len = es->es_len; 689 690 retval = ext4_ind_map_blocks(NULL, inode, &map, 0); 691 if (retval > 0) { 692 if (ext4_es_is_delayed(es) || ext4_es_is_hole(es)) { 693 /* 694 * We want to add a delayed/hole extent but this 695 * block has been allocated. 696 */ 697 pr_warn("ES insert assertion failed for inode: %lu " 698 "We can find blocks but we want to add a " 699 "delayed/hole extent [%d/%d/%llu/%x]\n", 700 inode->i_ino, es->es_lblk, es->es_len, 701 ext4_es_pblock(es), ext4_es_status(es)); 702 return; 703 } else if (ext4_es_is_written(es)) { 704 if (retval != es->es_len) { 705 pr_warn("ES insert assertion failed for " 706 "inode: %lu retval %d != es_len %d\n", 707 inode->i_ino, retval, es->es_len); 708 return; 709 } 710 if (map.m_pblk != ext4_es_pblock(es)) { 711 pr_warn("ES insert assertion failed for " 712 "inode: %lu m_pblk %llu != " 713 "es_pblk %llu\n", 714 inode->i_ino, map.m_pblk, 715 ext4_es_pblock(es)); 716 return; 717 } 718 } else { 719 /* 720 * We don't need to check unwritten extent because 721 * indirect-based file doesn't have it. 722 */ 723 BUG(); 724 } 725 } else if (retval == 0) { 726 if (ext4_es_is_written(es)) { 727 pr_warn("ES insert assertion failed for inode: %lu " 728 "We can't find the block but we want to add " 729 "a written extent [%d/%d/%llu/%x]\n", 730 inode->i_ino, es->es_lblk, es->es_len, 731 ext4_es_pblock(es), ext4_es_status(es)); 732 return; 733 } 734 } 735 } 736 737 static inline void ext4_es_insert_extent_check(struct inode *inode, 738 struct extent_status *es) 739 { 740 /* 741 * We don't need to worry about the race condition because 742 * caller takes i_data_sem locking. 743 */ 744 BUG_ON(!rwsem_is_locked(&EXT4_I(inode)->i_data_sem)); 745 if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) 746 ext4_es_insert_extent_ext_check(inode, es); 747 else 748 ext4_es_insert_extent_ind_check(inode, es); 749 } 750 #else 751 static inline void ext4_es_insert_extent_check(struct inode *inode, 752 struct extent_status *es) 753 { 754 } 755 #endif 756 757 static int __es_insert_extent(struct inode *inode, struct extent_status *newes) 758 { 759 struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree; 760 struct rb_node **p = &tree->root.rb_node; 761 struct rb_node *parent = NULL; 762 struct extent_status *es; 763 764 while (*p) { 765 parent = *p; 766 es = rb_entry(parent, struct extent_status, rb_node); 767 768 if (newes->es_lblk < es->es_lblk) { 769 if (ext4_es_can_be_merged(newes, es)) { 770 /* 771 * Here we can modify es_lblk directly 772 * because it isn't overlapped. 773 */ 774 es->es_lblk = newes->es_lblk; 775 es->es_len += newes->es_len; 776 if (ext4_es_is_written(es) || 777 ext4_es_is_unwritten(es)) 778 ext4_es_store_pblock(es, 779 newes->es_pblk); 780 es = ext4_es_try_to_merge_left(inode, es); 781 goto out; 782 } 783 p = &(*p)->rb_left; 784 } else if (newes->es_lblk > ext4_es_end(es)) { 785 if (ext4_es_can_be_merged(es, newes)) { 786 es->es_len += newes->es_len; 787 es = ext4_es_try_to_merge_right(inode, es); 788 goto out; 789 } 790 p = &(*p)->rb_right; 791 } else { 792 BUG(); 793 return -EINVAL; 794 } 795 } 796 797 es = ext4_es_alloc_extent(inode, newes->es_lblk, newes->es_len, 798 newes->es_pblk); 799 if (!es) 800 return -ENOMEM; 801 rb_link_node(&es->rb_node, parent, p); 802 rb_insert_color(&es->rb_node, &tree->root); 803 804 out: 805 tree->cache_es = es; 806 return 0; 807 } 808 809 /* 810 * ext4_es_insert_extent() adds information to an inode's extent 811 * status tree. 812 * 813 * Return 0 on success, error code on failure. 814 */ 815 int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk, 816 ext4_lblk_t len, ext4_fsblk_t pblk, 817 unsigned int status) 818 { 819 struct extent_status newes; 820 ext4_lblk_t end = lblk + len - 1; 821 int err = 0; 822 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 823 824 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 825 return 0; 826 827 es_debug("add [%u/%u) %llu %x to extent status tree of inode %lu\n", 828 lblk, len, pblk, status, inode->i_ino); 829 830 if (!len) 831 return 0; 832 833 BUG_ON(end < lblk); 834 835 if ((status & EXTENT_STATUS_DELAYED) && 836 (status & EXTENT_STATUS_WRITTEN)) { 837 ext4_warning(inode->i_sb, "Inserting extent [%u/%u] as " 838 " delayed and written which can potentially " 839 " cause data loss.", lblk, len); 840 WARN_ON(1); 841 } 842 843 newes.es_lblk = lblk; 844 newes.es_len = len; 845 ext4_es_store_pblock_status(&newes, pblk, status); 846 trace_ext4_es_insert_extent(inode, &newes); 847 848 ext4_es_insert_extent_check(inode, &newes); 849 850 write_lock(&EXT4_I(inode)->i_es_lock); 851 err = __es_remove_extent(inode, lblk, end, NULL); 852 if (err != 0) 853 goto error; 854 retry: 855 err = __es_insert_extent(inode, &newes); 856 if (err == -ENOMEM && __es_shrink(EXT4_SB(inode->i_sb), 857 128, EXT4_I(inode))) 858 goto retry; 859 if (err == -ENOMEM && !ext4_es_is_delayed(&newes)) 860 err = 0; 861 862 if (sbi->s_cluster_ratio > 1 && test_opt(inode->i_sb, DELALLOC) && 863 (status & EXTENT_STATUS_WRITTEN || 864 status & EXTENT_STATUS_UNWRITTEN)) 865 __revise_pending(inode, lblk, len); 866 867 error: 868 write_unlock(&EXT4_I(inode)->i_es_lock); 869 870 ext4_es_print_tree(inode); 871 872 return err; 873 } 874 875 /* 876 * ext4_es_cache_extent() inserts information into the extent status 877 * tree if and only if there isn't information about the range in 878 * question already. 879 */ 880 void ext4_es_cache_extent(struct inode *inode, ext4_lblk_t lblk, 881 ext4_lblk_t len, ext4_fsblk_t pblk, 882 unsigned int status) 883 { 884 struct extent_status *es; 885 struct extent_status newes; 886 ext4_lblk_t end = lblk + len - 1; 887 888 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 889 return; 890 891 newes.es_lblk = lblk; 892 newes.es_len = len; 893 ext4_es_store_pblock_status(&newes, pblk, status); 894 trace_ext4_es_cache_extent(inode, &newes); 895 896 if (!len) 897 return; 898 899 BUG_ON(end < lblk); 900 901 write_lock(&EXT4_I(inode)->i_es_lock); 902 903 es = __es_tree_search(&EXT4_I(inode)->i_es_tree.root, lblk); 904 if (!es || es->es_lblk > end) 905 __es_insert_extent(inode, &newes); 906 write_unlock(&EXT4_I(inode)->i_es_lock); 907 } 908 909 /* 910 * ext4_es_lookup_extent() looks up an extent in extent status tree. 911 * 912 * ext4_es_lookup_extent is called by ext4_map_blocks/ext4_da_map_blocks. 913 * 914 * Return: 1 on found, 0 on not 915 */ 916 int ext4_es_lookup_extent(struct inode *inode, ext4_lblk_t lblk, 917 ext4_lblk_t *next_lblk, 918 struct extent_status *es) 919 { 920 struct ext4_es_tree *tree; 921 struct ext4_es_stats *stats; 922 struct extent_status *es1 = NULL; 923 struct rb_node *node; 924 int found = 0; 925 926 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 927 return 0; 928 929 trace_ext4_es_lookup_extent_enter(inode, lblk); 930 es_debug("lookup extent in block %u\n", lblk); 931 932 tree = &EXT4_I(inode)->i_es_tree; 933 read_lock(&EXT4_I(inode)->i_es_lock); 934 935 /* find extent in cache firstly */ 936 es->es_lblk = es->es_len = es->es_pblk = 0; 937 if (tree->cache_es) { 938 es1 = tree->cache_es; 939 if (in_range(lblk, es1->es_lblk, es1->es_len)) { 940 es_debug("%u cached by [%u/%u)\n", 941 lblk, es1->es_lblk, es1->es_len); 942 found = 1; 943 goto out; 944 } 945 } 946 947 node = tree->root.rb_node; 948 while (node) { 949 es1 = rb_entry(node, struct extent_status, rb_node); 950 if (lblk < es1->es_lblk) 951 node = node->rb_left; 952 else if (lblk > ext4_es_end(es1)) 953 node = node->rb_right; 954 else { 955 found = 1; 956 break; 957 } 958 } 959 960 out: 961 stats = &EXT4_SB(inode->i_sb)->s_es_stats; 962 if (found) { 963 BUG_ON(!es1); 964 es->es_lblk = es1->es_lblk; 965 es->es_len = es1->es_len; 966 es->es_pblk = es1->es_pblk; 967 if (!ext4_es_is_referenced(es1)) 968 ext4_es_set_referenced(es1); 969 percpu_counter_inc(&stats->es_stats_cache_hits); 970 if (next_lblk) { 971 node = rb_next(&es1->rb_node); 972 if (node) { 973 es1 = rb_entry(node, struct extent_status, 974 rb_node); 975 *next_lblk = es1->es_lblk; 976 } else 977 *next_lblk = 0; 978 } 979 } else { 980 percpu_counter_inc(&stats->es_stats_cache_misses); 981 } 982 983 read_unlock(&EXT4_I(inode)->i_es_lock); 984 985 trace_ext4_es_lookup_extent_exit(inode, es, found); 986 return found; 987 } 988 989 struct rsvd_count { 990 int ndelonly; 991 bool first_do_lblk_found; 992 ext4_lblk_t first_do_lblk; 993 ext4_lblk_t last_do_lblk; 994 struct extent_status *left_es; 995 bool partial; 996 ext4_lblk_t lclu; 997 }; 998 999 /* 1000 * init_rsvd - initialize reserved count data before removing block range 1001 * in file from extent status tree 1002 * 1003 * @inode - file containing range 1004 * @lblk - first block in range 1005 * @es - pointer to first extent in range 1006 * @rc - pointer to reserved count data 1007 * 1008 * Assumes es is not NULL 1009 */ 1010 static void init_rsvd(struct inode *inode, ext4_lblk_t lblk, 1011 struct extent_status *es, struct rsvd_count *rc) 1012 { 1013 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 1014 struct rb_node *node; 1015 1016 rc->ndelonly = 0; 1017 1018 /* 1019 * for bigalloc, note the first delonly block in the range has not 1020 * been found, record the extent containing the block to the left of 1021 * the region to be removed, if any, and note that there's no partial 1022 * cluster to track 1023 */ 1024 if (sbi->s_cluster_ratio > 1) { 1025 rc->first_do_lblk_found = false; 1026 if (lblk > es->es_lblk) { 1027 rc->left_es = es; 1028 } else { 1029 node = rb_prev(&es->rb_node); 1030 rc->left_es = node ? rb_entry(node, 1031 struct extent_status, 1032 rb_node) : NULL; 1033 } 1034 rc->partial = false; 1035 } 1036 } 1037 1038 /* 1039 * count_rsvd - count the clusters containing delayed and not unwritten 1040 * (delonly) blocks in a range within an extent and add to 1041 * the running tally in rsvd_count 1042 * 1043 * @inode - file containing extent 1044 * @lblk - first block in range 1045 * @len - length of range in blocks 1046 * @es - pointer to extent containing clusters to be counted 1047 * @rc - pointer to reserved count data 1048 * 1049 * Tracks partial clusters found at the beginning and end of extents so 1050 * they aren't overcounted when they span adjacent extents 1051 */ 1052 static void count_rsvd(struct inode *inode, ext4_lblk_t lblk, long len, 1053 struct extent_status *es, struct rsvd_count *rc) 1054 { 1055 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 1056 ext4_lblk_t i, end, nclu; 1057 1058 if (!ext4_es_is_delonly(es)) 1059 return; 1060 1061 WARN_ON(len <= 0); 1062 1063 if (sbi->s_cluster_ratio == 1) { 1064 rc->ndelonly += (int) len; 1065 return; 1066 } 1067 1068 /* bigalloc */ 1069 1070 i = (lblk < es->es_lblk) ? es->es_lblk : lblk; 1071 end = lblk + (ext4_lblk_t) len - 1; 1072 end = (end > ext4_es_end(es)) ? ext4_es_end(es) : end; 1073 1074 /* record the first block of the first delonly extent seen */ 1075 if (!rc->first_do_lblk_found) { 1076 rc->first_do_lblk = i; 1077 rc->first_do_lblk_found = true; 1078 } 1079 1080 /* update the last lblk in the region seen so far */ 1081 rc->last_do_lblk = end; 1082 1083 /* 1084 * if we're tracking a partial cluster and the current extent 1085 * doesn't start with it, count it and stop tracking 1086 */ 1087 if (rc->partial && (rc->lclu != EXT4_B2C(sbi, i))) { 1088 rc->ndelonly++; 1089 rc->partial = false; 1090 } 1091 1092 /* 1093 * if the first cluster doesn't start on a cluster boundary but 1094 * ends on one, count it 1095 */ 1096 if (EXT4_LBLK_COFF(sbi, i) != 0) { 1097 if (end >= EXT4_LBLK_CFILL(sbi, i)) { 1098 rc->ndelonly++; 1099 rc->partial = false; 1100 i = EXT4_LBLK_CFILL(sbi, i) + 1; 1101 } 1102 } 1103 1104 /* 1105 * if the current cluster starts on a cluster boundary, count the 1106 * number of whole delonly clusters in the extent 1107 */ 1108 if ((i + sbi->s_cluster_ratio - 1) <= end) { 1109 nclu = (end - i + 1) >> sbi->s_cluster_bits; 1110 rc->ndelonly += nclu; 1111 i += nclu << sbi->s_cluster_bits; 1112 } 1113 1114 /* 1115 * start tracking a partial cluster if there's a partial at the end 1116 * of the current extent and we're not already tracking one 1117 */ 1118 if (!rc->partial && i <= end) { 1119 rc->partial = true; 1120 rc->lclu = EXT4_B2C(sbi, i); 1121 } 1122 } 1123 1124 /* 1125 * __pr_tree_search - search for a pending cluster reservation 1126 * 1127 * @root - root of pending reservation tree 1128 * @lclu - logical cluster to search for 1129 * 1130 * Returns the pending reservation for the cluster identified by @lclu 1131 * if found. If not, returns a reservation for the next cluster if any, 1132 * and if not, returns NULL. 1133 */ 1134 static struct pending_reservation *__pr_tree_search(struct rb_root *root, 1135 ext4_lblk_t lclu) 1136 { 1137 struct rb_node *node = root->rb_node; 1138 struct pending_reservation *pr = NULL; 1139 1140 while (node) { 1141 pr = rb_entry(node, struct pending_reservation, rb_node); 1142 if (lclu < pr->lclu) 1143 node = node->rb_left; 1144 else if (lclu > pr->lclu) 1145 node = node->rb_right; 1146 else 1147 return pr; 1148 } 1149 if (pr && lclu < pr->lclu) 1150 return pr; 1151 if (pr && lclu > pr->lclu) { 1152 node = rb_next(&pr->rb_node); 1153 return node ? rb_entry(node, struct pending_reservation, 1154 rb_node) : NULL; 1155 } 1156 return NULL; 1157 } 1158 1159 /* 1160 * get_rsvd - calculates and returns the number of cluster reservations to be 1161 * released when removing a block range from the extent status tree 1162 * and releases any pending reservations within the range 1163 * 1164 * @inode - file containing block range 1165 * @end - last block in range 1166 * @right_es - pointer to extent containing next block beyond end or NULL 1167 * @rc - pointer to reserved count data 1168 * 1169 * The number of reservations to be released is equal to the number of 1170 * clusters containing delayed and not unwritten (delonly) blocks within 1171 * the range, minus the number of clusters still containing delonly blocks 1172 * at the ends of the range, and minus the number of pending reservations 1173 * within the range. 1174 */ 1175 static unsigned int get_rsvd(struct inode *inode, ext4_lblk_t end, 1176 struct extent_status *right_es, 1177 struct rsvd_count *rc) 1178 { 1179 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 1180 struct pending_reservation *pr; 1181 struct ext4_pending_tree *tree = &EXT4_I(inode)->i_pending_tree; 1182 struct rb_node *node; 1183 ext4_lblk_t first_lclu, last_lclu; 1184 bool left_delonly, right_delonly, count_pending; 1185 struct extent_status *es; 1186 1187 if (sbi->s_cluster_ratio > 1) { 1188 /* count any remaining partial cluster */ 1189 if (rc->partial) 1190 rc->ndelonly++; 1191 1192 if (rc->ndelonly == 0) 1193 return 0; 1194 1195 first_lclu = EXT4_B2C(sbi, rc->first_do_lblk); 1196 last_lclu = EXT4_B2C(sbi, rc->last_do_lblk); 1197 1198 /* 1199 * decrease the delonly count by the number of clusters at the 1200 * ends of the range that still contain delonly blocks - 1201 * these clusters still need to be reserved 1202 */ 1203 left_delonly = right_delonly = false; 1204 1205 es = rc->left_es; 1206 while (es && ext4_es_end(es) >= 1207 EXT4_LBLK_CMASK(sbi, rc->first_do_lblk)) { 1208 if (ext4_es_is_delonly(es)) { 1209 rc->ndelonly--; 1210 left_delonly = true; 1211 break; 1212 } 1213 node = rb_prev(&es->rb_node); 1214 if (!node) 1215 break; 1216 es = rb_entry(node, struct extent_status, rb_node); 1217 } 1218 if (right_es && (!left_delonly || first_lclu != last_lclu)) { 1219 if (end < ext4_es_end(right_es)) { 1220 es = right_es; 1221 } else { 1222 node = rb_next(&right_es->rb_node); 1223 es = node ? rb_entry(node, struct extent_status, 1224 rb_node) : NULL; 1225 } 1226 while (es && es->es_lblk <= 1227 EXT4_LBLK_CFILL(sbi, rc->last_do_lblk)) { 1228 if (ext4_es_is_delonly(es)) { 1229 rc->ndelonly--; 1230 right_delonly = true; 1231 break; 1232 } 1233 node = rb_next(&es->rb_node); 1234 if (!node) 1235 break; 1236 es = rb_entry(node, struct extent_status, 1237 rb_node); 1238 } 1239 } 1240 1241 /* 1242 * Determine the block range that should be searched for 1243 * pending reservations, if any. Clusters on the ends of the 1244 * original removed range containing delonly blocks are 1245 * excluded. They've already been accounted for and it's not 1246 * possible to determine if an associated pending reservation 1247 * should be released with the information available in the 1248 * extents status tree. 1249 */ 1250 if (first_lclu == last_lclu) { 1251 if (left_delonly | right_delonly) 1252 count_pending = false; 1253 else 1254 count_pending = true; 1255 } else { 1256 if (left_delonly) 1257 first_lclu++; 1258 if (right_delonly) 1259 last_lclu--; 1260 if (first_lclu <= last_lclu) 1261 count_pending = true; 1262 else 1263 count_pending = false; 1264 } 1265 1266 /* 1267 * a pending reservation found between first_lclu and last_lclu 1268 * represents an allocated cluster that contained at least one 1269 * delonly block, so the delonly total must be reduced by one 1270 * for each pending reservation found and released 1271 */ 1272 if (count_pending) { 1273 pr = __pr_tree_search(&tree->root, first_lclu); 1274 while (pr && pr->lclu <= last_lclu) { 1275 rc->ndelonly--; 1276 node = rb_next(&pr->rb_node); 1277 rb_erase(&pr->rb_node, &tree->root); 1278 kmem_cache_free(ext4_pending_cachep, pr); 1279 if (!node) 1280 break; 1281 pr = rb_entry(node, struct pending_reservation, 1282 rb_node); 1283 } 1284 } 1285 } 1286 return rc->ndelonly; 1287 } 1288 1289 1290 /* 1291 * __es_remove_extent - removes block range from extent status tree 1292 * 1293 * @inode - file containing range 1294 * @lblk - first block in range 1295 * @end - last block in range 1296 * @reserved - number of cluster reservations released 1297 * 1298 * If @reserved is not NULL and delayed allocation is enabled, counts 1299 * block/cluster reservations freed by removing range and if bigalloc 1300 * enabled cancels pending reservations as needed. Returns 0 on success, 1301 * error code on failure. 1302 */ 1303 static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk, 1304 ext4_lblk_t end, int *reserved) 1305 { 1306 struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree; 1307 struct rb_node *node; 1308 struct extent_status *es; 1309 struct extent_status orig_es; 1310 ext4_lblk_t len1, len2; 1311 ext4_fsblk_t block; 1312 int err; 1313 bool count_reserved = true; 1314 struct rsvd_count rc; 1315 1316 if (reserved == NULL || !test_opt(inode->i_sb, DELALLOC)) 1317 count_reserved = false; 1318 retry: 1319 err = 0; 1320 1321 es = __es_tree_search(&tree->root, lblk); 1322 if (!es) 1323 goto out; 1324 if (es->es_lblk > end) 1325 goto out; 1326 1327 /* Simply invalidate cache_es. */ 1328 tree->cache_es = NULL; 1329 if (count_reserved) 1330 init_rsvd(inode, lblk, es, &rc); 1331 1332 orig_es.es_lblk = es->es_lblk; 1333 orig_es.es_len = es->es_len; 1334 orig_es.es_pblk = es->es_pblk; 1335 1336 len1 = lblk > es->es_lblk ? lblk - es->es_lblk : 0; 1337 len2 = ext4_es_end(es) > end ? ext4_es_end(es) - end : 0; 1338 if (len1 > 0) 1339 es->es_len = len1; 1340 if (len2 > 0) { 1341 if (len1 > 0) { 1342 struct extent_status newes; 1343 1344 newes.es_lblk = end + 1; 1345 newes.es_len = len2; 1346 block = 0x7FDEADBEEFULL; 1347 if (ext4_es_is_written(&orig_es) || 1348 ext4_es_is_unwritten(&orig_es)) 1349 block = ext4_es_pblock(&orig_es) + 1350 orig_es.es_len - len2; 1351 ext4_es_store_pblock_status(&newes, block, 1352 ext4_es_status(&orig_es)); 1353 err = __es_insert_extent(inode, &newes); 1354 if (err) { 1355 es->es_lblk = orig_es.es_lblk; 1356 es->es_len = orig_es.es_len; 1357 if ((err == -ENOMEM) && 1358 __es_shrink(EXT4_SB(inode->i_sb), 1359 128, EXT4_I(inode))) 1360 goto retry; 1361 goto out; 1362 } 1363 } else { 1364 es->es_lblk = end + 1; 1365 es->es_len = len2; 1366 if (ext4_es_is_written(es) || 1367 ext4_es_is_unwritten(es)) { 1368 block = orig_es.es_pblk + orig_es.es_len - len2; 1369 ext4_es_store_pblock(es, block); 1370 } 1371 } 1372 if (count_reserved) 1373 count_rsvd(inode, lblk, orig_es.es_len - len1 - len2, 1374 &orig_es, &rc); 1375 goto out; 1376 } 1377 1378 if (len1 > 0) { 1379 if (count_reserved) 1380 count_rsvd(inode, lblk, orig_es.es_len - len1, 1381 &orig_es, &rc); 1382 node = rb_next(&es->rb_node); 1383 if (node) 1384 es = rb_entry(node, struct extent_status, rb_node); 1385 else 1386 es = NULL; 1387 } 1388 1389 while (es && ext4_es_end(es) <= end) { 1390 if (count_reserved) 1391 count_rsvd(inode, es->es_lblk, es->es_len, es, &rc); 1392 node = rb_next(&es->rb_node); 1393 rb_erase(&es->rb_node, &tree->root); 1394 ext4_es_free_extent(inode, es); 1395 if (!node) { 1396 es = NULL; 1397 break; 1398 } 1399 es = rb_entry(node, struct extent_status, rb_node); 1400 } 1401 1402 if (es && es->es_lblk < end + 1) { 1403 ext4_lblk_t orig_len = es->es_len; 1404 1405 len1 = ext4_es_end(es) - end; 1406 if (count_reserved) 1407 count_rsvd(inode, es->es_lblk, orig_len - len1, 1408 es, &rc); 1409 es->es_lblk = end + 1; 1410 es->es_len = len1; 1411 if (ext4_es_is_written(es) || ext4_es_is_unwritten(es)) { 1412 block = es->es_pblk + orig_len - len1; 1413 ext4_es_store_pblock(es, block); 1414 } 1415 } 1416 1417 if (count_reserved) 1418 *reserved = get_rsvd(inode, end, es, &rc); 1419 out: 1420 return err; 1421 } 1422 1423 /* 1424 * ext4_es_remove_extent - removes block range from extent status tree 1425 * 1426 * @inode - file containing range 1427 * @lblk - first block in range 1428 * @len - number of blocks to remove 1429 * 1430 * Reduces block/cluster reservation count and for bigalloc cancels pending 1431 * reservations as needed. Returns 0 on success, error code on failure. 1432 */ 1433 int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk, 1434 ext4_lblk_t len) 1435 { 1436 ext4_lblk_t end; 1437 int err = 0; 1438 int reserved = 0; 1439 1440 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 1441 return 0; 1442 1443 trace_ext4_es_remove_extent(inode, lblk, len); 1444 es_debug("remove [%u/%u) from extent status tree of inode %lu\n", 1445 lblk, len, inode->i_ino); 1446 1447 if (!len) 1448 return err; 1449 1450 end = lblk + len - 1; 1451 BUG_ON(end < lblk); 1452 1453 /* 1454 * ext4_clear_inode() depends on us taking i_es_lock unconditionally 1455 * so that we are sure __es_shrink() is done with the inode before it 1456 * is reclaimed. 1457 */ 1458 write_lock(&EXT4_I(inode)->i_es_lock); 1459 err = __es_remove_extent(inode, lblk, end, &reserved); 1460 write_unlock(&EXT4_I(inode)->i_es_lock); 1461 ext4_es_print_tree(inode); 1462 ext4_da_release_space(inode, reserved); 1463 return err; 1464 } 1465 1466 static int __es_shrink(struct ext4_sb_info *sbi, int nr_to_scan, 1467 struct ext4_inode_info *locked_ei) 1468 { 1469 struct ext4_inode_info *ei; 1470 struct ext4_es_stats *es_stats; 1471 ktime_t start_time; 1472 u64 scan_time; 1473 int nr_to_walk; 1474 int nr_shrunk = 0; 1475 int retried = 0, nr_skipped = 0; 1476 1477 es_stats = &sbi->s_es_stats; 1478 start_time = ktime_get(); 1479 1480 retry: 1481 spin_lock(&sbi->s_es_lock); 1482 nr_to_walk = sbi->s_es_nr_inode; 1483 while (nr_to_walk-- > 0) { 1484 if (list_empty(&sbi->s_es_list)) { 1485 spin_unlock(&sbi->s_es_lock); 1486 goto out; 1487 } 1488 ei = list_first_entry(&sbi->s_es_list, struct ext4_inode_info, 1489 i_es_list); 1490 /* Move the inode to the tail */ 1491 list_move_tail(&ei->i_es_list, &sbi->s_es_list); 1492 1493 /* 1494 * Normally we try hard to avoid shrinking precached inodes, 1495 * but we will as a last resort. 1496 */ 1497 if (!retried && ext4_test_inode_state(&ei->vfs_inode, 1498 EXT4_STATE_EXT_PRECACHED)) { 1499 nr_skipped++; 1500 continue; 1501 } 1502 1503 if (ei == locked_ei || !write_trylock(&ei->i_es_lock)) { 1504 nr_skipped++; 1505 continue; 1506 } 1507 /* 1508 * Now we hold i_es_lock which protects us from inode reclaim 1509 * freeing inode under us 1510 */ 1511 spin_unlock(&sbi->s_es_lock); 1512 1513 nr_shrunk += es_reclaim_extents(ei, &nr_to_scan); 1514 write_unlock(&ei->i_es_lock); 1515 1516 if (nr_to_scan <= 0) 1517 goto out; 1518 spin_lock(&sbi->s_es_lock); 1519 } 1520 spin_unlock(&sbi->s_es_lock); 1521 1522 /* 1523 * If we skipped any inodes, and we weren't able to make any 1524 * forward progress, try again to scan precached inodes. 1525 */ 1526 if ((nr_shrunk == 0) && nr_skipped && !retried) { 1527 retried++; 1528 goto retry; 1529 } 1530 1531 if (locked_ei && nr_shrunk == 0) 1532 nr_shrunk = es_reclaim_extents(locked_ei, &nr_to_scan); 1533 1534 out: 1535 scan_time = ktime_to_ns(ktime_sub(ktime_get(), start_time)); 1536 if (likely(es_stats->es_stats_scan_time)) 1537 es_stats->es_stats_scan_time = (scan_time + 1538 es_stats->es_stats_scan_time*3) / 4; 1539 else 1540 es_stats->es_stats_scan_time = scan_time; 1541 if (scan_time > es_stats->es_stats_max_scan_time) 1542 es_stats->es_stats_max_scan_time = scan_time; 1543 if (likely(es_stats->es_stats_shrunk)) 1544 es_stats->es_stats_shrunk = (nr_shrunk + 1545 es_stats->es_stats_shrunk*3) / 4; 1546 else 1547 es_stats->es_stats_shrunk = nr_shrunk; 1548 1549 trace_ext4_es_shrink(sbi->s_sb, nr_shrunk, scan_time, 1550 nr_skipped, retried); 1551 return nr_shrunk; 1552 } 1553 1554 static unsigned long ext4_es_count(struct shrinker *shrink, 1555 struct shrink_control *sc) 1556 { 1557 unsigned long nr; 1558 struct ext4_sb_info *sbi; 1559 1560 sbi = container_of(shrink, struct ext4_sb_info, s_es_shrinker); 1561 nr = percpu_counter_read_positive(&sbi->s_es_stats.es_stats_shk_cnt); 1562 trace_ext4_es_shrink_count(sbi->s_sb, sc->nr_to_scan, nr); 1563 return nr; 1564 } 1565 1566 static unsigned long ext4_es_scan(struct shrinker *shrink, 1567 struct shrink_control *sc) 1568 { 1569 struct ext4_sb_info *sbi = container_of(shrink, 1570 struct ext4_sb_info, s_es_shrinker); 1571 int nr_to_scan = sc->nr_to_scan; 1572 int ret, nr_shrunk; 1573 1574 ret = percpu_counter_read_positive(&sbi->s_es_stats.es_stats_shk_cnt); 1575 trace_ext4_es_shrink_scan_enter(sbi->s_sb, nr_to_scan, ret); 1576 1577 nr_shrunk = __es_shrink(sbi, nr_to_scan, NULL); 1578 1579 ret = percpu_counter_read_positive(&sbi->s_es_stats.es_stats_shk_cnt); 1580 trace_ext4_es_shrink_scan_exit(sbi->s_sb, nr_shrunk, ret); 1581 return nr_shrunk; 1582 } 1583 1584 int ext4_seq_es_shrinker_info_show(struct seq_file *seq, void *v) 1585 { 1586 struct ext4_sb_info *sbi = EXT4_SB((struct super_block *) seq->private); 1587 struct ext4_es_stats *es_stats = &sbi->s_es_stats; 1588 struct ext4_inode_info *ei, *max = NULL; 1589 unsigned int inode_cnt = 0; 1590 1591 if (v != SEQ_START_TOKEN) 1592 return 0; 1593 1594 /* here we just find an inode that has the max nr. of objects */ 1595 spin_lock(&sbi->s_es_lock); 1596 list_for_each_entry(ei, &sbi->s_es_list, i_es_list) { 1597 inode_cnt++; 1598 if (max && max->i_es_all_nr < ei->i_es_all_nr) 1599 max = ei; 1600 else if (!max) 1601 max = ei; 1602 } 1603 spin_unlock(&sbi->s_es_lock); 1604 1605 seq_printf(seq, "stats:\n %lld objects\n %lld reclaimable objects\n", 1606 percpu_counter_sum_positive(&es_stats->es_stats_all_cnt), 1607 percpu_counter_sum_positive(&es_stats->es_stats_shk_cnt)); 1608 seq_printf(seq, " %lld/%lld cache hits/misses\n", 1609 percpu_counter_sum_positive(&es_stats->es_stats_cache_hits), 1610 percpu_counter_sum_positive(&es_stats->es_stats_cache_misses)); 1611 if (inode_cnt) 1612 seq_printf(seq, " %d inodes on list\n", inode_cnt); 1613 1614 seq_printf(seq, "average:\n %llu us scan time\n", 1615 div_u64(es_stats->es_stats_scan_time, 1000)); 1616 seq_printf(seq, " %lu shrunk objects\n", es_stats->es_stats_shrunk); 1617 if (inode_cnt) 1618 seq_printf(seq, 1619 "maximum:\n %lu inode (%u objects, %u reclaimable)\n" 1620 " %llu us max scan time\n", 1621 max->vfs_inode.i_ino, max->i_es_all_nr, max->i_es_shk_nr, 1622 div_u64(es_stats->es_stats_max_scan_time, 1000)); 1623 1624 return 0; 1625 } 1626 1627 int ext4_es_register_shrinker(struct ext4_sb_info *sbi) 1628 { 1629 int err; 1630 1631 /* Make sure we have enough bits for physical block number */ 1632 BUILD_BUG_ON(ES_SHIFT < 48); 1633 INIT_LIST_HEAD(&sbi->s_es_list); 1634 sbi->s_es_nr_inode = 0; 1635 spin_lock_init(&sbi->s_es_lock); 1636 sbi->s_es_stats.es_stats_shrunk = 0; 1637 err = percpu_counter_init(&sbi->s_es_stats.es_stats_cache_hits, 0, 1638 GFP_KERNEL); 1639 if (err) 1640 return err; 1641 err = percpu_counter_init(&sbi->s_es_stats.es_stats_cache_misses, 0, 1642 GFP_KERNEL); 1643 if (err) 1644 goto err1; 1645 sbi->s_es_stats.es_stats_scan_time = 0; 1646 sbi->s_es_stats.es_stats_max_scan_time = 0; 1647 err = percpu_counter_init(&sbi->s_es_stats.es_stats_all_cnt, 0, GFP_KERNEL); 1648 if (err) 1649 goto err2; 1650 err = percpu_counter_init(&sbi->s_es_stats.es_stats_shk_cnt, 0, GFP_KERNEL); 1651 if (err) 1652 goto err3; 1653 1654 sbi->s_es_shrinker.scan_objects = ext4_es_scan; 1655 sbi->s_es_shrinker.count_objects = ext4_es_count; 1656 sbi->s_es_shrinker.seeks = DEFAULT_SEEKS; 1657 err = register_shrinker(&sbi->s_es_shrinker); 1658 if (err) 1659 goto err4; 1660 1661 return 0; 1662 err4: 1663 percpu_counter_destroy(&sbi->s_es_stats.es_stats_shk_cnt); 1664 err3: 1665 percpu_counter_destroy(&sbi->s_es_stats.es_stats_all_cnt); 1666 err2: 1667 percpu_counter_destroy(&sbi->s_es_stats.es_stats_cache_misses); 1668 err1: 1669 percpu_counter_destroy(&sbi->s_es_stats.es_stats_cache_hits); 1670 return err; 1671 } 1672 1673 void ext4_es_unregister_shrinker(struct ext4_sb_info *sbi) 1674 { 1675 percpu_counter_destroy(&sbi->s_es_stats.es_stats_cache_hits); 1676 percpu_counter_destroy(&sbi->s_es_stats.es_stats_cache_misses); 1677 percpu_counter_destroy(&sbi->s_es_stats.es_stats_all_cnt); 1678 percpu_counter_destroy(&sbi->s_es_stats.es_stats_shk_cnt); 1679 unregister_shrinker(&sbi->s_es_shrinker); 1680 } 1681 1682 /* 1683 * Shrink extents in given inode from ei->i_es_shrink_lblk till end. Scan at 1684 * most *nr_to_scan extents, update *nr_to_scan accordingly. 1685 * 1686 * Return 0 if we hit end of tree / interval, 1 if we exhausted nr_to_scan. 1687 * Increment *nr_shrunk by the number of reclaimed extents. Also update 1688 * ei->i_es_shrink_lblk to where we should continue scanning. 1689 */ 1690 static int es_do_reclaim_extents(struct ext4_inode_info *ei, ext4_lblk_t end, 1691 int *nr_to_scan, int *nr_shrunk) 1692 { 1693 struct inode *inode = &ei->vfs_inode; 1694 struct ext4_es_tree *tree = &ei->i_es_tree; 1695 struct extent_status *es; 1696 struct rb_node *node; 1697 1698 es = __es_tree_search(&tree->root, ei->i_es_shrink_lblk); 1699 if (!es) 1700 goto out_wrap; 1701 1702 while (*nr_to_scan > 0) { 1703 if (es->es_lblk > end) { 1704 ei->i_es_shrink_lblk = end + 1; 1705 return 0; 1706 } 1707 1708 (*nr_to_scan)--; 1709 node = rb_next(&es->rb_node); 1710 /* 1711 * We can't reclaim delayed extent from status tree because 1712 * fiemap, bigallic, and seek_data/hole need to use it. 1713 */ 1714 if (ext4_es_is_delayed(es)) 1715 goto next; 1716 if (ext4_es_is_referenced(es)) { 1717 ext4_es_clear_referenced(es); 1718 goto next; 1719 } 1720 1721 rb_erase(&es->rb_node, &tree->root); 1722 ext4_es_free_extent(inode, es); 1723 (*nr_shrunk)++; 1724 next: 1725 if (!node) 1726 goto out_wrap; 1727 es = rb_entry(node, struct extent_status, rb_node); 1728 } 1729 ei->i_es_shrink_lblk = es->es_lblk; 1730 return 1; 1731 out_wrap: 1732 ei->i_es_shrink_lblk = 0; 1733 return 0; 1734 } 1735 1736 static int es_reclaim_extents(struct ext4_inode_info *ei, int *nr_to_scan) 1737 { 1738 struct inode *inode = &ei->vfs_inode; 1739 int nr_shrunk = 0; 1740 ext4_lblk_t start = ei->i_es_shrink_lblk; 1741 static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL, 1742 DEFAULT_RATELIMIT_BURST); 1743 1744 if (ei->i_es_shk_nr == 0) 1745 return 0; 1746 1747 if (ext4_test_inode_state(inode, EXT4_STATE_EXT_PRECACHED) && 1748 __ratelimit(&_rs)) 1749 ext4_warning(inode->i_sb, "forced shrink of precached extents"); 1750 1751 if (!es_do_reclaim_extents(ei, EXT_MAX_BLOCKS, nr_to_scan, &nr_shrunk) && 1752 start != 0) 1753 es_do_reclaim_extents(ei, start - 1, nr_to_scan, &nr_shrunk); 1754 1755 ei->i_es_tree.cache_es = NULL; 1756 return nr_shrunk; 1757 } 1758 1759 /* 1760 * Called to support EXT4_IOC_CLEAR_ES_CACHE. We can only remove 1761 * discretionary entries from the extent status cache. (Some entries 1762 * must be present for proper operations.) 1763 */ 1764 void ext4_clear_inode_es(struct inode *inode) 1765 { 1766 struct ext4_inode_info *ei = EXT4_I(inode); 1767 struct extent_status *es; 1768 struct ext4_es_tree *tree; 1769 struct rb_node *node; 1770 1771 write_lock(&ei->i_es_lock); 1772 tree = &EXT4_I(inode)->i_es_tree; 1773 tree->cache_es = NULL; 1774 node = rb_first(&tree->root); 1775 while (node) { 1776 es = rb_entry(node, struct extent_status, rb_node); 1777 node = rb_next(node); 1778 if (!ext4_es_is_delayed(es)) { 1779 rb_erase(&es->rb_node, &tree->root); 1780 ext4_es_free_extent(inode, es); 1781 } 1782 } 1783 ext4_clear_inode_state(inode, EXT4_STATE_EXT_PRECACHED); 1784 write_unlock(&ei->i_es_lock); 1785 } 1786 1787 #ifdef ES_DEBUG__ 1788 static void ext4_print_pending_tree(struct inode *inode) 1789 { 1790 struct ext4_pending_tree *tree; 1791 struct rb_node *node; 1792 struct pending_reservation *pr; 1793 1794 printk(KERN_DEBUG "pending reservations for inode %lu:", inode->i_ino); 1795 tree = &EXT4_I(inode)->i_pending_tree; 1796 node = rb_first(&tree->root); 1797 while (node) { 1798 pr = rb_entry(node, struct pending_reservation, rb_node); 1799 printk(KERN_DEBUG " %u", pr->lclu); 1800 node = rb_next(node); 1801 } 1802 printk(KERN_DEBUG "\n"); 1803 } 1804 #else 1805 #define ext4_print_pending_tree(inode) 1806 #endif 1807 1808 int __init ext4_init_pending(void) 1809 { 1810 ext4_pending_cachep = kmem_cache_create("ext4_pending_reservation", 1811 sizeof(struct pending_reservation), 1812 0, (SLAB_RECLAIM_ACCOUNT), NULL); 1813 if (ext4_pending_cachep == NULL) 1814 return -ENOMEM; 1815 return 0; 1816 } 1817 1818 void ext4_exit_pending(void) 1819 { 1820 kmem_cache_destroy(ext4_pending_cachep); 1821 } 1822 1823 void ext4_init_pending_tree(struct ext4_pending_tree *tree) 1824 { 1825 tree->root = RB_ROOT; 1826 } 1827 1828 /* 1829 * __get_pending - retrieve a pointer to a pending reservation 1830 * 1831 * @inode - file containing the pending cluster reservation 1832 * @lclu - logical cluster of interest 1833 * 1834 * Returns a pointer to a pending reservation if it's a member of 1835 * the set, and NULL if not. Must be called holding i_es_lock. 1836 */ 1837 static struct pending_reservation *__get_pending(struct inode *inode, 1838 ext4_lblk_t lclu) 1839 { 1840 struct ext4_pending_tree *tree; 1841 struct rb_node *node; 1842 struct pending_reservation *pr = NULL; 1843 1844 tree = &EXT4_I(inode)->i_pending_tree; 1845 node = (&tree->root)->rb_node; 1846 1847 while (node) { 1848 pr = rb_entry(node, struct pending_reservation, rb_node); 1849 if (lclu < pr->lclu) 1850 node = node->rb_left; 1851 else if (lclu > pr->lclu) 1852 node = node->rb_right; 1853 else if (lclu == pr->lclu) 1854 return pr; 1855 } 1856 return NULL; 1857 } 1858 1859 /* 1860 * __insert_pending - adds a pending cluster reservation to the set of 1861 * pending reservations 1862 * 1863 * @inode - file containing the cluster 1864 * @lblk - logical block in the cluster to be added 1865 * 1866 * Returns 0 on successful insertion and -ENOMEM on failure. If the 1867 * pending reservation is already in the set, returns successfully. 1868 */ 1869 static int __insert_pending(struct inode *inode, ext4_lblk_t lblk) 1870 { 1871 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 1872 struct ext4_pending_tree *tree = &EXT4_I(inode)->i_pending_tree; 1873 struct rb_node **p = &tree->root.rb_node; 1874 struct rb_node *parent = NULL; 1875 struct pending_reservation *pr; 1876 ext4_lblk_t lclu; 1877 int ret = 0; 1878 1879 lclu = EXT4_B2C(sbi, lblk); 1880 /* search to find parent for insertion */ 1881 while (*p) { 1882 parent = *p; 1883 pr = rb_entry(parent, struct pending_reservation, rb_node); 1884 1885 if (lclu < pr->lclu) { 1886 p = &(*p)->rb_left; 1887 } else if (lclu > pr->lclu) { 1888 p = &(*p)->rb_right; 1889 } else { 1890 /* pending reservation already inserted */ 1891 goto out; 1892 } 1893 } 1894 1895 pr = kmem_cache_alloc(ext4_pending_cachep, GFP_ATOMIC); 1896 if (pr == NULL) { 1897 ret = -ENOMEM; 1898 goto out; 1899 } 1900 pr->lclu = lclu; 1901 1902 rb_link_node(&pr->rb_node, parent, p); 1903 rb_insert_color(&pr->rb_node, &tree->root); 1904 1905 out: 1906 return ret; 1907 } 1908 1909 /* 1910 * __remove_pending - removes a pending cluster reservation from the set 1911 * of pending reservations 1912 * 1913 * @inode - file containing the cluster 1914 * @lblk - logical block in the pending cluster reservation to be removed 1915 * 1916 * Returns successfully if pending reservation is not a member of the set. 1917 */ 1918 static void __remove_pending(struct inode *inode, ext4_lblk_t lblk) 1919 { 1920 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 1921 struct pending_reservation *pr; 1922 struct ext4_pending_tree *tree; 1923 1924 pr = __get_pending(inode, EXT4_B2C(sbi, lblk)); 1925 if (pr != NULL) { 1926 tree = &EXT4_I(inode)->i_pending_tree; 1927 rb_erase(&pr->rb_node, &tree->root); 1928 kmem_cache_free(ext4_pending_cachep, pr); 1929 } 1930 } 1931 1932 /* 1933 * ext4_remove_pending - removes a pending cluster reservation from the set 1934 * of pending reservations 1935 * 1936 * @inode - file containing the cluster 1937 * @lblk - logical block in the pending cluster reservation to be removed 1938 * 1939 * Locking for external use of __remove_pending. 1940 */ 1941 void ext4_remove_pending(struct inode *inode, ext4_lblk_t lblk) 1942 { 1943 struct ext4_inode_info *ei = EXT4_I(inode); 1944 1945 write_lock(&ei->i_es_lock); 1946 __remove_pending(inode, lblk); 1947 write_unlock(&ei->i_es_lock); 1948 } 1949 1950 /* 1951 * ext4_is_pending - determine whether a cluster has a pending reservation 1952 * on it 1953 * 1954 * @inode - file containing the cluster 1955 * @lblk - logical block in the cluster 1956 * 1957 * Returns true if there's a pending reservation for the cluster in the 1958 * set of pending reservations, and false if not. 1959 */ 1960 bool ext4_is_pending(struct inode *inode, ext4_lblk_t lblk) 1961 { 1962 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 1963 struct ext4_inode_info *ei = EXT4_I(inode); 1964 bool ret; 1965 1966 read_lock(&ei->i_es_lock); 1967 ret = (bool)(__get_pending(inode, EXT4_B2C(sbi, lblk)) != NULL); 1968 read_unlock(&ei->i_es_lock); 1969 1970 return ret; 1971 } 1972 1973 /* 1974 * ext4_es_insert_delayed_block - adds a delayed block to the extents status 1975 * tree, adding a pending reservation where 1976 * needed 1977 * 1978 * @inode - file containing the newly added block 1979 * @lblk - logical block to be added 1980 * @allocated - indicates whether a physical cluster has been allocated for 1981 * the logical cluster that contains the block 1982 * 1983 * Returns 0 on success, negative error code on failure. 1984 */ 1985 int ext4_es_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk, 1986 bool allocated) 1987 { 1988 struct extent_status newes; 1989 int err = 0; 1990 1991 if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) 1992 return 0; 1993 1994 es_debug("add [%u/1) delayed to extent status tree of inode %lu\n", 1995 lblk, inode->i_ino); 1996 1997 newes.es_lblk = lblk; 1998 newes.es_len = 1; 1999 ext4_es_store_pblock_status(&newes, ~0, EXTENT_STATUS_DELAYED); 2000 trace_ext4_es_insert_delayed_block(inode, &newes, allocated); 2001 2002 ext4_es_insert_extent_check(inode, &newes); 2003 2004 write_lock(&EXT4_I(inode)->i_es_lock); 2005 2006 err = __es_remove_extent(inode, lblk, lblk, NULL); 2007 if (err != 0) 2008 goto error; 2009 retry: 2010 err = __es_insert_extent(inode, &newes); 2011 if (err == -ENOMEM && __es_shrink(EXT4_SB(inode->i_sb), 2012 128, EXT4_I(inode))) 2013 goto retry; 2014 if (err != 0) 2015 goto error; 2016 2017 if (allocated) 2018 __insert_pending(inode, lblk); 2019 2020 error: 2021 write_unlock(&EXT4_I(inode)->i_es_lock); 2022 2023 ext4_es_print_tree(inode); 2024 ext4_print_pending_tree(inode); 2025 2026 return err; 2027 } 2028 2029 /* 2030 * __es_delayed_clu - count number of clusters containing blocks that 2031 * are delayed only 2032 * 2033 * @inode - file containing block range 2034 * @start - logical block defining start of range 2035 * @end - logical block defining end of range 2036 * 2037 * Returns the number of clusters containing only delayed (not delayed 2038 * and unwritten) blocks in the range specified by @start and @end. Any 2039 * cluster or part of a cluster within the range and containing a delayed 2040 * and not unwritten block within the range is counted as a whole cluster. 2041 */ 2042 static unsigned int __es_delayed_clu(struct inode *inode, ext4_lblk_t start, 2043 ext4_lblk_t end) 2044 { 2045 struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree; 2046 struct extent_status *es; 2047 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 2048 struct rb_node *node; 2049 ext4_lblk_t first_lclu, last_lclu; 2050 unsigned long long last_counted_lclu; 2051 unsigned int n = 0; 2052 2053 /* guaranteed to be unequal to any ext4_lblk_t value */ 2054 last_counted_lclu = ~0ULL; 2055 2056 es = __es_tree_search(&tree->root, start); 2057 2058 while (es && (es->es_lblk <= end)) { 2059 if (ext4_es_is_delonly(es)) { 2060 if (es->es_lblk <= start) 2061 first_lclu = EXT4_B2C(sbi, start); 2062 else 2063 first_lclu = EXT4_B2C(sbi, es->es_lblk); 2064 2065 if (ext4_es_end(es) >= end) 2066 last_lclu = EXT4_B2C(sbi, end); 2067 else 2068 last_lclu = EXT4_B2C(sbi, ext4_es_end(es)); 2069 2070 if (first_lclu == last_counted_lclu) 2071 n += last_lclu - first_lclu; 2072 else 2073 n += last_lclu - first_lclu + 1; 2074 last_counted_lclu = last_lclu; 2075 } 2076 node = rb_next(&es->rb_node); 2077 if (!node) 2078 break; 2079 es = rb_entry(node, struct extent_status, rb_node); 2080 } 2081 2082 return n; 2083 } 2084 2085 /* 2086 * ext4_es_delayed_clu - count number of clusters containing blocks that 2087 * are both delayed and unwritten 2088 * 2089 * @inode - file containing block range 2090 * @lblk - logical block defining start of range 2091 * @len - number of blocks in range 2092 * 2093 * Locking for external use of __es_delayed_clu(). 2094 */ 2095 unsigned int ext4_es_delayed_clu(struct inode *inode, ext4_lblk_t lblk, 2096 ext4_lblk_t len) 2097 { 2098 struct ext4_inode_info *ei = EXT4_I(inode); 2099 ext4_lblk_t end; 2100 unsigned int n; 2101 2102 if (len == 0) 2103 return 0; 2104 2105 end = lblk + len - 1; 2106 WARN_ON(end < lblk); 2107 2108 read_lock(&ei->i_es_lock); 2109 2110 n = __es_delayed_clu(inode, lblk, end); 2111 2112 read_unlock(&ei->i_es_lock); 2113 2114 return n; 2115 } 2116 2117 /* 2118 * __revise_pending - makes, cancels, or leaves unchanged pending cluster 2119 * reservations for a specified block range depending 2120 * upon the presence or absence of delayed blocks 2121 * outside the range within clusters at the ends of the 2122 * range 2123 * 2124 * @inode - file containing the range 2125 * @lblk - logical block defining the start of range 2126 * @len - length of range in blocks 2127 * 2128 * Used after a newly allocated extent is added to the extents status tree. 2129 * Requires that the extents in the range have either written or unwritten 2130 * status. Must be called while holding i_es_lock. 2131 */ 2132 static void __revise_pending(struct inode *inode, ext4_lblk_t lblk, 2133 ext4_lblk_t len) 2134 { 2135 struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); 2136 ext4_lblk_t end = lblk + len - 1; 2137 ext4_lblk_t first, last; 2138 bool f_del = false, l_del = false; 2139 2140 if (len == 0) 2141 return; 2142 2143 /* 2144 * Two cases - block range within single cluster and block range 2145 * spanning two or more clusters. Note that a cluster belonging 2146 * to a range starting and/or ending on a cluster boundary is treated 2147 * as if it does not contain a delayed extent. The new range may 2148 * have allocated space for previously delayed blocks out to the 2149 * cluster boundary, requiring that any pre-existing pending 2150 * reservation be canceled. Because this code only looks at blocks 2151 * outside the range, it should revise pending reservations 2152 * correctly even if the extent represented by the range can't be 2153 * inserted in the extents status tree due to ENOSPC. 2154 */ 2155 2156 if (EXT4_B2C(sbi, lblk) == EXT4_B2C(sbi, end)) { 2157 first = EXT4_LBLK_CMASK(sbi, lblk); 2158 if (first != lblk) 2159 f_del = __es_scan_range(inode, &ext4_es_is_delonly, 2160 first, lblk - 1); 2161 if (f_del) { 2162 __insert_pending(inode, first); 2163 } else { 2164 last = EXT4_LBLK_CMASK(sbi, end) + 2165 sbi->s_cluster_ratio - 1; 2166 if (last != end) 2167 l_del = __es_scan_range(inode, 2168 &ext4_es_is_delonly, 2169 end + 1, last); 2170 if (l_del) 2171 __insert_pending(inode, last); 2172 else 2173 __remove_pending(inode, last); 2174 } 2175 } else { 2176 first = EXT4_LBLK_CMASK(sbi, lblk); 2177 if (first != lblk) 2178 f_del = __es_scan_range(inode, &ext4_es_is_delonly, 2179 first, lblk - 1); 2180 if (f_del) 2181 __insert_pending(inode, first); 2182 else 2183 __remove_pending(inode, first); 2184 2185 last = EXT4_LBLK_CMASK(sbi, end) + sbi->s_cluster_ratio - 1; 2186 if (last != end) 2187 l_del = __es_scan_range(inode, &ext4_es_is_delonly, 2188 end + 1, last); 2189 if (l_del) 2190 __insert_pending(inode, last); 2191 else 2192 __remove_pending(inode, last); 2193 } 2194 } 2195