1 // SPDX-License-Identifier: GPL-2.0 2 /* 3 * fs/f2fs/dir.c 4 * 5 * Copyright (c) 2012 Samsung Electronics Co., Ltd. 6 * http://www.samsung.com/ 7 */ 8 #include <asm/unaligned.h> 9 #include <linux/fs.h> 10 #include <linux/f2fs_fs.h> 11 #include <linux/sched/signal.h> 12 #include <linux/unicode.h> 13 #include "f2fs.h" 14 #include "node.h" 15 #include "acl.h" 16 #include "xattr.h" 17 #include <trace/events/f2fs.h> 18 19 #if IS_ENABLED(CONFIG_UNICODE) 20 extern struct kmem_cache *f2fs_cf_name_slab; 21 #endif 22 23 static unsigned long dir_blocks(struct inode *inode) 24 { 25 return ((unsigned long long) (i_size_read(inode) + PAGE_SIZE - 1)) 26 >> PAGE_SHIFT; 27 } 28 29 static unsigned int dir_buckets(unsigned int level, int dir_level) 30 { 31 if (level + dir_level < MAX_DIR_HASH_DEPTH / 2) 32 return 1 << (level + dir_level); 33 else 34 return MAX_DIR_BUCKETS; 35 } 36 37 static unsigned int bucket_blocks(unsigned int level) 38 { 39 if (level < MAX_DIR_HASH_DEPTH / 2) 40 return 2; 41 else 42 return 4; 43 } 44 45 static unsigned char f2fs_filetype_table[F2FS_FT_MAX] = { 46 [F2FS_FT_UNKNOWN] = DT_UNKNOWN, 47 [F2FS_FT_REG_FILE] = DT_REG, 48 [F2FS_FT_DIR] = DT_DIR, 49 [F2FS_FT_CHRDEV] = DT_CHR, 50 [F2FS_FT_BLKDEV] = DT_BLK, 51 [F2FS_FT_FIFO] = DT_FIFO, 52 [F2FS_FT_SOCK] = DT_SOCK, 53 [F2FS_FT_SYMLINK] = DT_LNK, 54 }; 55 56 static unsigned char f2fs_type_by_mode[S_IFMT >> S_SHIFT] = { 57 [S_IFREG >> S_SHIFT] = F2FS_FT_REG_FILE, 58 [S_IFDIR >> S_SHIFT] = F2FS_FT_DIR, 59 [S_IFCHR >> S_SHIFT] = F2FS_FT_CHRDEV, 60 [S_IFBLK >> S_SHIFT] = F2FS_FT_BLKDEV, 61 [S_IFIFO >> S_SHIFT] = F2FS_FT_FIFO, 62 [S_IFSOCK >> S_SHIFT] = F2FS_FT_SOCK, 63 [S_IFLNK >> S_SHIFT] = F2FS_FT_SYMLINK, 64 }; 65 66 static void set_de_type(struct f2fs_dir_entry *de, umode_t mode) 67 { 68 de->file_type = f2fs_type_by_mode[(mode & S_IFMT) >> S_SHIFT]; 69 } 70 71 unsigned char f2fs_get_de_type(struct f2fs_dir_entry *de) 72 { 73 if (de->file_type < F2FS_FT_MAX) 74 return f2fs_filetype_table[de->file_type]; 75 return DT_UNKNOWN; 76 } 77 78 /* If @dir is casefolded, initialize @fname->cf_name from @fname->usr_fname. */ 79 int f2fs_init_casefolded_name(const struct inode *dir, 80 struct f2fs_filename *fname) 81 { 82 #if IS_ENABLED(CONFIG_UNICODE) 83 struct super_block *sb = dir->i_sb; 84 85 if (IS_CASEFOLDED(dir) && 86 !is_dot_dotdot(fname->usr_fname->name, fname->usr_fname->len)) { 87 fname->cf_name.name = f2fs_kmem_cache_alloc(f2fs_cf_name_slab, 88 GFP_NOFS, false, F2FS_SB(sb)); 89 if (!fname->cf_name.name) 90 return -ENOMEM; 91 fname->cf_name.len = utf8_casefold(sb->s_encoding, 92 fname->usr_fname, 93 fname->cf_name.name, 94 F2FS_NAME_LEN); 95 if ((int)fname->cf_name.len <= 0) { 96 kmem_cache_free(f2fs_cf_name_slab, fname->cf_name.name); 97 fname->cf_name.name = NULL; 98 if (sb_has_strict_encoding(sb)) 99 return -EINVAL; 100 /* fall back to treating name as opaque byte sequence */ 101 } 102 } 103 #endif 104 return 0; 105 } 106 107 static int __f2fs_setup_filename(const struct inode *dir, 108 const struct fscrypt_name *crypt_name, 109 struct f2fs_filename *fname) 110 { 111 int err; 112 113 memset(fname, 0, sizeof(*fname)); 114 115 fname->usr_fname = crypt_name->usr_fname; 116 fname->disk_name = crypt_name->disk_name; 117 #ifdef CONFIG_FS_ENCRYPTION 118 fname->crypto_buf = crypt_name->crypto_buf; 119 #endif 120 if (crypt_name->is_nokey_name) { 121 /* hash was decoded from the no-key name */ 122 fname->hash = cpu_to_le32(crypt_name->hash); 123 } else { 124 err = f2fs_init_casefolded_name(dir, fname); 125 if (err) { 126 f2fs_free_filename(fname); 127 return err; 128 } 129 f2fs_hash_filename(dir, fname); 130 } 131 return 0; 132 } 133 134 /* 135 * Prepare to search for @iname in @dir. This is similar to 136 * fscrypt_setup_filename(), but this also handles computing the casefolded name 137 * and the f2fs dirhash if needed, then packing all the information about this 138 * filename up into a 'struct f2fs_filename'. 139 */ 140 int f2fs_setup_filename(struct inode *dir, const struct qstr *iname, 141 int lookup, struct f2fs_filename *fname) 142 { 143 struct fscrypt_name crypt_name; 144 int err; 145 146 err = fscrypt_setup_filename(dir, iname, lookup, &crypt_name); 147 if (err) 148 return err; 149 150 return __f2fs_setup_filename(dir, &crypt_name, fname); 151 } 152 153 /* 154 * Prepare to look up @dentry in @dir. This is similar to 155 * fscrypt_prepare_lookup(), but this also handles computing the casefolded name 156 * and the f2fs dirhash if needed, then packing all the information about this 157 * filename up into a 'struct f2fs_filename'. 158 */ 159 int f2fs_prepare_lookup(struct inode *dir, struct dentry *dentry, 160 struct f2fs_filename *fname) 161 { 162 struct fscrypt_name crypt_name; 163 int err; 164 165 err = fscrypt_prepare_lookup(dir, dentry, &crypt_name); 166 if (err) 167 return err; 168 169 return __f2fs_setup_filename(dir, &crypt_name, fname); 170 } 171 172 void f2fs_free_filename(struct f2fs_filename *fname) 173 { 174 #ifdef CONFIG_FS_ENCRYPTION 175 kfree(fname->crypto_buf.name); 176 fname->crypto_buf.name = NULL; 177 #endif 178 #if IS_ENABLED(CONFIG_UNICODE) 179 if (fname->cf_name.name) { 180 kmem_cache_free(f2fs_cf_name_slab, fname->cf_name.name); 181 fname->cf_name.name = NULL; 182 } 183 #endif 184 } 185 186 static unsigned long dir_block_index(unsigned int level, 187 int dir_level, unsigned int idx) 188 { 189 unsigned long i; 190 unsigned long bidx = 0; 191 192 for (i = 0; i < level; i++) 193 bidx += dir_buckets(i, dir_level) * bucket_blocks(i); 194 bidx += idx * bucket_blocks(level); 195 return bidx; 196 } 197 198 static struct f2fs_dir_entry *find_in_block(struct inode *dir, 199 struct page *dentry_page, 200 const struct f2fs_filename *fname, 201 int *max_slots) 202 { 203 struct f2fs_dentry_block *dentry_blk; 204 struct f2fs_dentry_ptr d; 205 206 dentry_blk = (struct f2fs_dentry_block *)page_address(dentry_page); 207 208 make_dentry_ptr_block(dir, &d, dentry_blk); 209 return f2fs_find_target_dentry(&d, fname, max_slots); 210 } 211 212 #if IS_ENABLED(CONFIG_UNICODE) 213 /* 214 * Test whether a case-insensitive directory entry matches the filename 215 * being searched for. 216 * 217 * Returns 1 for a match, 0 for no match, and -errno on an error. 218 */ 219 static int f2fs_match_ci_name(const struct inode *dir, const struct qstr *name, 220 const u8 *de_name, u32 de_name_len) 221 { 222 const struct super_block *sb = dir->i_sb; 223 const struct unicode_map *um = sb->s_encoding; 224 struct fscrypt_str decrypted_name = FSTR_INIT(NULL, de_name_len); 225 struct qstr entry = QSTR_INIT(de_name, de_name_len); 226 int res; 227 228 if (IS_ENCRYPTED(dir)) { 229 const struct fscrypt_str encrypted_name = 230 FSTR_INIT((u8 *)de_name, de_name_len); 231 232 if (WARN_ON_ONCE(!fscrypt_has_encryption_key(dir))) 233 return -EINVAL; 234 235 decrypted_name.name = kmalloc(de_name_len, GFP_KERNEL); 236 if (!decrypted_name.name) 237 return -ENOMEM; 238 res = fscrypt_fname_disk_to_usr(dir, 0, 0, &encrypted_name, 239 &decrypted_name); 240 if (res < 0) 241 goto out; 242 entry.name = decrypted_name.name; 243 entry.len = decrypted_name.len; 244 } 245 246 res = utf8_strncasecmp_folded(um, name, &entry); 247 /* 248 * In strict mode, ignore invalid names. In non-strict mode, 249 * fall back to treating them as opaque byte sequences. 250 */ 251 if (res < 0 && !sb_has_strict_encoding(sb)) { 252 res = name->len == entry.len && 253 memcmp(name->name, entry.name, name->len) == 0; 254 } else { 255 /* utf8_strncasecmp_folded returns 0 on match */ 256 res = (res == 0); 257 } 258 out: 259 kfree(decrypted_name.name); 260 return res; 261 } 262 #endif /* CONFIG_UNICODE */ 263 264 static inline int f2fs_match_name(const struct inode *dir, 265 const struct f2fs_filename *fname, 266 const u8 *de_name, u32 de_name_len) 267 { 268 struct fscrypt_name f; 269 270 #if IS_ENABLED(CONFIG_UNICODE) 271 if (fname->cf_name.name) { 272 struct qstr cf = FSTR_TO_QSTR(&fname->cf_name); 273 274 return f2fs_match_ci_name(dir, &cf, de_name, de_name_len); 275 } 276 #endif 277 f.usr_fname = fname->usr_fname; 278 f.disk_name = fname->disk_name; 279 #ifdef CONFIG_FS_ENCRYPTION 280 f.crypto_buf = fname->crypto_buf; 281 #endif 282 return fscrypt_match_name(&f, de_name, de_name_len); 283 } 284 285 struct f2fs_dir_entry *f2fs_find_target_dentry(const struct f2fs_dentry_ptr *d, 286 const struct f2fs_filename *fname, int *max_slots) 287 { 288 struct f2fs_dir_entry *de; 289 unsigned long bit_pos = 0; 290 int max_len = 0; 291 int res = 0; 292 293 if (max_slots) 294 *max_slots = 0; 295 while (bit_pos < d->max) { 296 if (!test_bit_le(bit_pos, d->bitmap)) { 297 bit_pos++; 298 max_len++; 299 continue; 300 } 301 302 de = &d->dentry[bit_pos]; 303 304 if (unlikely(!de->name_len)) { 305 bit_pos++; 306 continue; 307 } 308 309 if (de->hash_code == fname->hash) { 310 res = f2fs_match_name(d->inode, fname, 311 d->filename[bit_pos], 312 le16_to_cpu(de->name_len)); 313 if (res < 0) 314 return ERR_PTR(res); 315 if (res) 316 goto found; 317 } 318 319 if (max_slots && max_len > *max_slots) 320 *max_slots = max_len; 321 max_len = 0; 322 323 bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len)); 324 } 325 326 de = NULL; 327 found: 328 if (max_slots && max_len > *max_slots) 329 *max_slots = max_len; 330 return de; 331 } 332 333 static struct f2fs_dir_entry *find_in_level(struct inode *dir, 334 unsigned int level, 335 const struct f2fs_filename *fname, 336 struct page **res_page) 337 { 338 int s = GET_DENTRY_SLOTS(fname->disk_name.len); 339 unsigned int nbucket, nblock; 340 unsigned int bidx, end_block; 341 struct page *dentry_page; 342 struct f2fs_dir_entry *de = NULL; 343 pgoff_t next_pgofs; 344 bool room = false; 345 int max_slots; 346 347 nbucket = dir_buckets(level, F2FS_I(dir)->i_dir_level); 348 nblock = bucket_blocks(level); 349 350 bidx = dir_block_index(level, F2FS_I(dir)->i_dir_level, 351 le32_to_cpu(fname->hash) % nbucket); 352 end_block = bidx + nblock; 353 354 while (bidx < end_block) { 355 /* no need to allocate new dentry pages to all the indices */ 356 dentry_page = f2fs_find_data_page(dir, bidx, &next_pgofs); 357 if (IS_ERR(dentry_page)) { 358 if (PTR_ERR(dentry_page) == -ENOENT) { 359 room = true; 360 bidx = next_pgofs; 361 continue; 362 } else { 363 *res_page = dentry_page; 364 break; 365 } 366 } 367 368 de = find_in_block(dir, dentry_page, fname, &max_slots); 369 if (IS_ERR(de)) { 370 *res_page = ERR_CAST(de); 371 de = NULL; 372 break; 373 } else if (de) { 374 *res_page = dentry_page; 375 break; 376 } 377 378 if (max_slots >= s) 379 room = true; 380 f2fs_put_page(dentry_page, 0); 381 382 bidx++; 383 } 384 385 if (!de && room && F2FS_I(dir)->chash != fname->hash) { 386 F2FS_I(dir)->chash = fname->hash; 387 F2FS_I(dir)->clevel = level; 388 } 389 390 return de; 391 } 392 393 struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir, 394 const struct f2fs_filename *fname, 395 struct page **res_page) 396 { 397 unsigned long npages = dir_blocks(dir); 398 struct f2fs_dir_entry *de = NULL; 399 unsigned int max_depth; 400 unsigned int level; 401 402 *res_page = NULL; 403 404 if (f2fs_has_inline_dentry(dir)) { 405 de = f2fs_find_in_inline_dir(dir, fname, res_page); 406 goto out; 407 } 408 409 if (npages == 0) 410 goto out; 411 412 max_depth = F2FS_I(dir)->i_current_depth; 413 if (unlikely(max_depth > MAX_DIR_HASH_DEPTH)) { 414 f2fs_warn(F2FS_I_SB(dir), "Corrupted max_depth of %lu: %u", 415 dir->i_ino, max_depth); 416 max_depth = MAX_DIR_HASH_DEPTH; 417 f2fs_i_depth_write(dir, max_depth); 418 } 419 420 for (level = 0; level < max_depth; level++) { 421 de = find_in_level(dir, level, fname, res_page); 422 if (de || IS_ERR(*res_page)) 423 break; 424 } 425 out: 426 /* This is to increase the speed of f2fs_create */ 427 if (!de) 428 F2FS_I(dir)->task = current; 429 return de; 430 } 431 432 /* 433 * Find an entry in the specified directory with the wanted name. 434 * It returns the page where the entry was found (as a parameter - res_page), 435 * and the entry itself. Page is returned mapped and unlocked. 436 * Entry is guaranteed to be valid. 437 */ 438 struct f2fs_dir_entry *f2fs_find_entry(struct inode *dir, 439 const struct qstr *child, struct page **res_page) 440 { 441 struct f2fs_dir_entry *de = NULL; 442 struct f2fs_filename fname; 443 int err; 444 445 err = f2fs_setup_filename(dir, child, 1, &fname); 446 if (err) { 447 if (err == -ENOENT) 448 *res_page = NULL; 449 else 450 *res_page = ERR_PTR(err); 451 return NULL; 452 } 453 454 de = __f2fs_find_entry(dir, &fname, res_page); 455 456 f2fs_free_filename(&fname); 457 return de; 458 } 459 460 struct f2fs_dir_entry *f2fs_parent_dir(struct inode *dir, struct page **p) 461 { 462 return f2fs_find_entry(dir, &dotdot_name, p); 463 } 464 465 ino_t f2fs_inode_by_name(struct inode *dir, const struct qstr *qstr, 466 struct page **page) 467 { 468 ino_t res = 0; 469 struct f2fs_dir_entry *de; 470 471 de = f2fs_find_entry(dir, qstr, page); 472 if (de) { 473 res = le32_to_cpu(de->ino); 474 f2fs_put_page(*page, 0); 475 } 476 477 return res; 478 } 479 480 void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de, 481 struct page *page, struct inode *inode) 482 { 483 enum page_type type = f2fs_has_inline_dentry(dir) ? NODE : DATA; 484 485 lock_page(page); 486 f2fs_wait_on_page_writeback(page, type, true, true); 487 de->ino = cpu_to_le32(inode->i_ino); 488 set_de_type(de, inode->i_mode); 489 set_page_dirty(page); 490 491 dir->i_mtime = dir->i_ctime = current_time(dir); 492 f2fs_mark_inode_dirty_sync(dir, false); 493 f2fs_put_page(page, 1); 494 } 495 496 static void init_dent_inode(struct inode *dir, struct inode *inode, 497 const struct f2fs_filename *fname, 498 struct page *ipage) 499 { 500 struct f2fs_inode *ri; 501 502 if (!fname) /* tmpfile case? */ 503 return; 504 505 f2fs_wait_on_page_writeback(ipage, NODE, true, true); 506 507 /* copy name info. to this inode page */ 508 ri = F2FS_INODE(ipage); 509 ri->i_namelen = cpu_to_le32(fname->disk_name.len); 510 memcpy(ri->i_name, fname->disk_name.name, fname->disk_name.len); 511 if (IS_ENCRYPTED(dir)) { 512 file_set_enc_name(inode); 513 /* 514 * Roll-forward recovery doesn't have encryption keys available, 515 * so it can't compute the dirhash for encrypted+casefolded 516 * filenames. Append it to i_name if possible. Else, disable 517 * roll-forward recovery of the dentry (i.e., make fsync'ing the 518 * file force a checkpoint) by setting LOST_PINO. 519 */ 520 if (IS_CASEFOLDED(dir)) { 521 if (fname->disk_name.len + sizeof(f2fs_hash_t) <= 522 F2FS_NAME_LEN) 523 put_unaligned(fname->hash, (f2fs_hash_t *) 524 &ri->i_name[fname->disk_name.len]); 525 else 526 file_lost_pino(inode); 527 } 528 } 529 set_page_dirty(ipage); 530 } 531 532 void f2fs_do_make_empty_dir(struct inode *inode, struct inode *parent, 533 struct f2fs_dentry_ptr *d) 534 { 535 struct fscrypt_str dot = FSTR_INIT(".", 1); 536 struct fscrypt_str dotdot = FSTR_INIT("..", 2); 537 538 /* update dirent of "." */ 539 f2fs_update_dentry(inode->i_ino, inode->i_mode, d, &dot, 0, 0); 540 541 /* update dirent of ".." */ 542 f2fs_update_dentry(parent->i_ino, parent->i_mode, d, &dotdot, 0, 1); 543 } 544 545 static int make_empty_dir(struct inode *inode, 546 struct inode *parent, struct page *page) 547 { 548 struct page *dentry_page; 549 struct f2fs_dentry_block *dentry_blk; 550 struct f2fs_dentry_ptr d; 551 552 if (f2fs_has_inline_dentry(inode)) 553 return f2fs_make_empty_inline_dir(inode, parent, page); 554 555 dentry_page = f2fs_get_new_data_page(inode, page, 0, true); 556 if (IS_ERR(dentry_page)) 557 return PTR_ERR(dentry_page); 558 559 dentry_blk = page_address(dentry_page); 560 561 make_dentry_ptr_block(NULL, &d, dentry_blk); 562 f2fs_do_make_empty_dir(inode, parent, &d); 563 564 set_page_dirty(dentry_page); 565 f2fs_put_page(dentry_page, 1); 566 return 0; 567 } 568 569 struct page *f2fs_init_inode_metadata(struct inode *inode, struct inode *dir, 570 const struct f2fs_filename *fname, struct page *dpage) 571 { 572 struct page *page; 573 int err; 574 575 if (is_inode_flag_set(inode, FI_NEW_INODE)) { 576 page = f2fs_new_inode_page(inode); 577 if (IS_ERR(page)) 578 return page; 579 580 if (S_ISDIR(inode->i_mode)) { 581 /* in order to handle error case */ 582 get_page(page); 583 err = make_empty_dir(inode, dir, page); 584 if (err) { 585 lock_page(page); 586 goto put_error; 587 } 588 put_page(page); 589 } 590 591 err = f2fs_init_acl(inode, dir, page, dpage); 592 if (err) 593 goto put_error; 594 595 err = f2fs_init_security(inode, dir, 596 fname ? fname->usr_fname : NULL, page); 597 if (err) 598 goto put_error; 599 600 if (IS_ENCRYPTED(inode)) { 601 err = fscrypt_set_context(inode, page); 602 if (err) 603 goto put_error; 604 } 605 } else { 606 page = f2fs_get_node_page(F2FS_I_SB(dir), inode->i_ino); 607 if (IS_ERR(page)) 608 return page; 609 } 610 611 init_dent_inode(dir, inode, fname, page); 612 613 /* 614 * This file should be checkpointed during fsync. 615 * We lost i_pino from now on. 616 */ 617 if (is_inode_flag_set(inode, FI_INC_LINK)) { 618 if (!S_ISDIR(inode->i_mode)) 619 file_lost_pino(inode); 620 /* 621 * If link the tmpfile to alias through linkat path, 622 * we should remove this inode from orphan list. 623 */ 624 if (inode->i_nlink == 0) 625 f2fs_remove_orphan_inode(F2FS_I_SB(dir), inode->i_ino); 626 f2fs_i_links_write(inode, true); 627 } 628 return page; 629 630 put_error: 631 clear_nlink(inode); 632 f2fs_update_inode(inode, page); 633 f2fs_put_page(page, 1); 634 return ERR_PTR(err); 635 } 636 637 void f2fs_update_parent_metadata(struct inode *dir, struct inode *inode, 638 unsigned int current_depth) 639 { 640 if (inode && is_inode_flag_set(inode, FI_NEW_INODE)) { 641 if (S_ISDIR(inode->i_mode)) 642 f2fs_i_links_write(dir, true); 643 clear_inode_flag(inode, FI_NEW_INODE); 644 } 645 dir->i_mtime = dir->i_ctime = current_time(dir); 646 f2fs_mark_inode_dirty_sync(dir, false); 647 648 if (F2FS_I(dir)->i_current_depth != current_depth) 649 f2fs_i_depth_write(dir, current_depth); 650 651 if (inode && is_inode_flag_set(inode, FI_INC_LINK)) 652 clear_inode_flag(inode, FI_INC_LINK); 653 } 654 655 int f2fs_room_for_filename(const void *bitmap, int slots, int max_slots) 656 { 657 int bit_start = 0; 658 int zero_start, zero_end; 659 next: 660 zero_start = find_next_zero_bit_le(bitmap, max_slots, bit_start); 661 if (zero_start >= max_slots) 662 return max_slots; 663 664 zero_end = find_next_bit_le(bitmap, max_slots, zero_start); 665 if (zero_end - zero_start >= slots) 666 return zero_start; 667 668 bit_start = zero_end + 1; 669 670 if (zero_end + 1 >= max_slots) 671 return max_slots; 672 goto next; 673 } 674 675 bool f2fs_has_enough_room(struct inode *dir, struct page *ipage, 676 const struct f2fs_filename *fname) 677 { 678 struct f2fs_dentry_ptr d; 679 unsigned int bit_pos; 680 int slots = GET_DENTRY_SLOTS(fname->disk_name.len); 681 682 make_dentry_ptr_inline(dir, &d, inline_data_addr(dir, ipage)); 683 684 bit_pos = f2fs_room_for_filename(d.bitmap, slots, d.max); 685 686 return bit_pos < d.max; 687 } 688 689 void f2fs_update_dentry(nid_t ino, umode_t mode, struct f2fs_dentry_ptr *d, 690 const struct fscrypt_str *name, f2fs_hash_t name_hash, 691 unsigned int bit_pos) 692 { 693 struct f2fs_dir_entry *de; 694 int slots = GET_DENTRY_SLOTS(name->len); 695 int i; 696 697 de = &d->dentry[bit_pos]; 698 de->hash_code = name_hash; 699 de->name_len = cpu_to_le16(name->len); 700 memcpy(d->filename[bit_pos], name->name, name->len); 701 de->ino = cpu_to_le32(ino); 702 set_de_type(de, mode); 703 for (i = 0; i < slots; i++) { 704 __set_bit_le(bit_pos + i, (void *)d->bitmap); 705 /* avoid wrong garbage data for readdir */ 706 if (i) 707 (de + i)->name_len = 0; 708 } 709 } 710 711 int f2fs_add_regular_entry(struct inode *dir, const struct f2fs_filename *fname, 712 struct inode *inode, nid_t ino, umode_t mode) 713 { 714 unsigned int bit_pos; 715 unsigned int level; 716 unsigned int current_depth; 717 unsigned long bidx, block; 718 unsigned int nbucket, nblock; 719 struct page *dentry_page = NULL; 720 struct f2fs_dentry_block *dentry_blk = NULL; 721 struct f2fs_dentry_ptr d; 722 struct page *page = NULL; 723 int slots, err = 0; 724 725 level = 0; 726 slots = GET_DENTRY_SLOTS(fname->disk_name.len); 727 728 current_depth = F2FS_I(dir)->i_current_depth; 729 if (F2FS_I(dir)->chash == fname->hash) { 730 level = F2FS_I(dir)->clevel; 731 F2FS_I(dir)->chash = 0; 732 } 733 734 start: 735 if (time_to_inject(F2FS_I_SB(dir), FAULT_DIR_DEPTH)) { 736 f2fs_show_injection_info(F2FS_I_SB(dir), FAULT_DIR_DEPTH); 737 return -ENOSPC; 738 } 739 740 if (unlikely(current_depth == MAX_DIR_HASH_DEPTH)) 741 return -ENOSPC; 742 743 /* Increase the depth, if required */ 744 if (level == current_depth) 745 ++current_depth; 746 747 nbucket = dir_buckets(level, F2FS_I(dir)->i_dir_level); 748 nblock = bucket_blocks(level); 749 750 bidx = dir_block_index(level, F2FS_I(dir)->i_dir_level, 751 (le32_to_cpu(fname->hash) % nbucket)); 752 753 for (block = bidx; block <= (bidx + nblock - 1); block++) { 754 dentry_page = f2fs_get_new_data_page(dir, NULL, block, true); 755 if (IS_ERR(dentry_page)) 756 return PTR_ERR(dentry_page); 757 758 dentry_blk = page_address(dentry_page); 759 bit_pos = f2fs_room_for_filename(&dentry_blk->dentry_bitmap, 760 slots, NR_DENTRY_IN_BLOCK); 761 if (bit_pos < NR_DENTRY_IN_BLOCK) 762 goto add_dentry; 763 764 f2fs_put_page(dentry_page, 1); 765 } 766 767 /* Move to next level to find the empty slot for new dentry */ 768 ++level; 769 goto start; 770 add_dentry: 771 f2fs_wait_on_page_writeback(dentry_page, DATA, true, true); 772 773 if (inode) { 774 f2fs_down_write(&F2FS_I(inode)->i_sem); 775 page = f2fs_init_inode_metadata(inode, dir, fname, NULL); 776 if (IS_ERR(page)) { 777 err = PTR_ERR(page); 778 goto fail; 779 } 780 } 781 782 make_dentry_ptr_block(NULL, &d, dentry_blk); 783 f2fs_update_dentry(ino, mode, &d, &fname->disk_name, fname->hash, 784 bit_pos); 785 786 set_page_dirty(dentry_page); 787 788 if (inode) { 789 f2fs_i_pino_write(inode, dir->i_ino); 790 791 /* synchronize inode page's data from inode cache */ 792 if (is_inode_flag_set(inode, FI_NEW_INODE)) 793 f2fs_update_inode(inode, page); 794 795 f2fs_put_page(page, 1); 796 } 797 798 f2fs_update_parent_metadata(dir, inode, current_depth); 799 fail: 800 if (inode) 801 f2fs_up_write(&F2FS_I(inode)->i_sem); 802 803 f2fs_put_page(dentry_page, 1); 804 805 return err; 806 } 807 808 int f2fs_add_dentry(struct inode *dir, const struct f2fs_filename *fname, 809 struct inode *inode, nid_t ino, umode_t mode) 810 { 811 int err = -EAGAIN; 812 813 if (f2fs_has_inline_dentry(dir)) 814 err = f2fs_add_inline_entry(dir, fname, inode, ino, mode); 815 if (err == -EAGAIN) 816 err = f2fs_add_regular_entry(dir, fname, inode, ino, mode); 817 818 f2fs_update_time(F2FS_I_SB(dir), REQ_TIME); 819 return err; 820 } 821 822 /* 823 * Caller should grab and release a rwsem by calling f2fs_lock_op() and 824 * f2fs_unlock_op(). 825 */ 826 int f2fs_do_add_link(struct inode *dir, const struct qstr *name, 827 struct inode *inode, nid_t ino, umode_t mode) 828 { 829 struct f2fs_filename fname; 830 struct page *page = NULL; 831 struct f2fs_dir_entry *de = NULL; 832 int err; 833 834 err = f2fs_setup_filename(dir, name, 0, &fname); 835 if (err) 836 return err; 837 838 /* 839 * An immature stackable filesystem shows a race condition between lookup 840 * and create. If we have same task when doing lookup and create, it's 841 * definitely fine as expected by VFS normally. Otherwise, let's just 842 * verify on-disk dentry one more time, which guarantees filesystem 843 * consistency more. 844 */ 845 if (current != F2FS_I(dir)->task) { 846 de = __f2fs_find_entry(dir, &fname, &page); 847 F2FS_I(dir)->task = NULL; 848 } 849 if (de) { 850 f2fs_put_page(page, 0); 851 err = -EEXIST; 852 } else if (IS_ERR(page)) { 853 err = PTR_ERR(page); 854 } else { 855 err = f2fs_add_dentry(dir, &fname, inode, ino, mode); 856 } 857 f2fs_free_filename(&fname); 858 return err; 859 } 860 861 int f2fs_do_tmpfile(struct inode *inode, struct inode *dir) 862 { 863 struct page *page; 864 int err = 0; 865 866 f2fs_down_write(&F2FS_I(inode)->i_sem); 867 page = f2fs_init_inode_metadata(inode, dir, NULL, NULL); 868 if (IS_ERR(page)) { 869 err = PTR_ERR(page); 870 goto fail; 871 } 872 f2fs_put_page(page, 1); 873 874 clear_inode_flag(inode, FI_NEW_INODE); 875 f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); 876 fail: 877 f2fs_up_write(&F2FS_I(inode)->i_sem); 878 return err; 879 } 880 881 void f2fs_drop_nlink(struct inode *dir, struct inode *inode) 882 { 883 struct f2fs_sb_info *sbi = F2FS_I_SB(dir); 884 885 f2fs_down_write(&F2FS_I(inode)->i_sem); 886 887 if (S_ISDIR(inode->i_mode)) 888 f2fs_i_links_write(dir, false); 889 inode->i_ctime = current_time(inode); 890 891 f2fs_i_links_write(inode, false); 892 if (S_ISDIR(inode->i_mode)) { 893 f2fs_i_links_write(inode, false); 894 f2fs_i_size_write(inode, 0); 895 } 896 f2fs_up_write(&F2FS_I(inode)->i_sem); 897 898 if (inode->i_nlink == 0) 899 f2fs_add_orphan_inode(inode); 900 else 901 f2fs_release_orphan_inode(sbi); 902 } 903 904 /* 905 * It only removes the dentry from the dentry page, corresponding name 906 * entry in name page does not need to be touched during deletion. 907 */ 908 void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page, 909 struct inode *dir, struct inode *inode) 910 { 911 struct f2fs_dentry_block *dentry_blk; 912 unsigned int bit_pos; 913 int slots = GET_DENTRY_SLOTS(le16_to_cpu(dentry->name_len)); 914 int i; 915 916 f2fs_update_time(F2FS_I_SB(dir), REQ_TIME); 917 918 if (F2FS_OPTION(F2FS_I_SB(dir)).fsync_mode == FSYNC_MODE_STRICT) 919 f2fs_add_ino_entry(F2FS_I_SB(dir), dir->i_ino, TRANS_DIR_INO); 920 921 if (f2fs_has_inline_dentry(dir)) 922 return f2fs_delete_inline_entry(dentry, page, dir, inode); 923 924 lock_page(page); 925 f2fs_wait_on_page_writeback(page, DATA, true, true); 926 927 dentry_blk = page_address(page); 928 bit_pos = dentry - dentry_blk->dentry; 929 for (i = 0; i < slots; i++) 930 __clear_bit_le(bit_pos + i, &dentry_blk->dentry_bitmap); 931 932 /* Let's check and deallocate this dentry page */ 933 bit_pos = find_next_bit_le(&dentry_blk->dentry_bitmap, 934 NR_DENTRY_IN_BLOCK, 935 0); 936 set_page_dirty(page); 937 938 if (bit_pos == NR_DENTRY_IN_BLOCK && 939 !f2fs_truncate_hole(dir, page->index, page->index + 1)) { 940 f2fs_clear_page_cache_dirty_tag(page); 941 clear_page_dirty_for_io(page); 942 ClearPageUptodate(page); 943 944 clear_page_private_gcing(page); 945 946 inode_dec_dirty_pages(dir); 947 f2fs_remove_dirty_inode(dir); 948 949 detach_page_private(page); 950 set_page_private(page, 0); 951 } 952 f2fs_put_page(page, 1); 953 954 dir->i_ctime = dir->i_mtime = current_time(dir); 955 f2fs_mark_inode_dirty_sync(dir, false); 956 957 if (inode) 958 f2fs_drop_nlink(dir, inode); 959 } 960 961 bool f2fs_empty_dir(struct inode *dir) 962 { 963 unsigned long bidx = 0; 964 struct page *dentry_page; 965 unsigned int bit_pos; 966 struct f2fs_dentry_block *dentry_blk; 967 unsigned long nblock = dir_blocks(dir); 968 969 if (f2fs_has_inline_dentry(dir)) 970 return f2fs_empty_inline_dir(dir); 971 972 while (bidx < nblock) { 973 pgoff_t next_pgofs; 974 975 dentry_page = f2fs_find_data_page(dir, bidx, &next_pgofs); 976 if (IS_ERR(dentry_page)) { 977 if (PTR_ERR(dentry_page) == -ENOENT) { 978 bidx = next_pgofs; 979 continue; 980 } else { 981 return false; 982 } 983 } 984 985 dentry_blk = page_address(dentry_page); 986 if (bidx == 0) 987 bit_pos = 2; 988 else 989 bit_pos = 0; 990 bit_pos = find_next_bit_le(&dentry_blk->dentry_bitmap, 991 NR_DENTRY_IN_BLOCK, 992 bit_pos); 993 994 f2fs_put_page(dentry_page, 0); 995 996 if (bit_pos < NR_DENTRY_IN_BLOCK) 997 return false; 998 999 bidx++; 1000 } 1001 return true; 1002 } 1003 1004 int f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d, 1005 unsigned int start_pos, struct fscrypt_str *fstr) 1006 { 1007 unsigned char d_type = DT_UNKNOWN; 1008 unsigned int bit_pos; 1009 struct f2fs_dir_entry *de = NULL; 1010 struct fscrypt_str de_name = FSTR_INIT(NULL, 0); 1011 struct f2fs_sb_info *sbi = F2FS_I_SB(d->inode); 1012 struct blk_plug plug; 1013 bool readdir_ra = sbi->readdir_ra; 1014 bool found_valid_dirent = false; 1015 int err = 0; 1016 1017 bit_pos = ((unsigned long)ctx->pos % d->max); 1018 1019 if (readdir_ra) 1020 blk_start_plug(&plug); 1021 1022 while (bit_pos < d->max) { 1023 bit_pos = find_next_bit_le(d->bitmap, d->max, bit_pos); 1024 if (bit_pos >= d->max) 1025 break; 1026 1027 de = &d->dentry[bit_pos]; 1028 if (de->name_len == 0) { 1029 if (found_valid_dirent || !bit_pos) { 1030 printk_ratelimited( 1031 "%sF2FS-fs (%s): invalid namelen(0), ino:%u, run fsck to fix.", 1032 KERN_WARNING, sbi->sb->s_id, 1033 le32_to_cpu(de->ino)); 1034 set_sbi_flag(sbi, SBI_NEED_FSCK); 1035 } 1036 bit_pos++; 1037 ctx->pos = start_pos + bit_pos; 1038 continue; 1039 } 1040 1041 d_type = f2fs_get_de_type(de); 1042 1043 de_name.name = d->filename[bit_pos]; 1044 de_name.len = le16_to_cpu(de->name_len); 1045 1046 /* check memory boundary before moving forward */ 1047 bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len)); 1048 if (unlikely(bit_pos > d->max || 1049 le16_to_cpu(de->name_len) > F2FS_NAME_LEN)) { 1050 f2fs_warn(sbi, "%s: corrupted namelen=%d, run fsck to fix.", 1051 __func__, le16_to_cpu(de->name_len)); 1052 set_sbi_flag(sbi, SBI_NEED_FSCK); 1053 err = -EFSCORRUPTED; 1054 f2fs_handle_error(sbi, ERROR_CORRUPTED_DIRENT); 1055 goto out; 1056 } 1057 1058 if (IS_ENCRYPTED(d->inode)) { 1059 int save_len = fstr->len; 1060 1061 err = fscrypt_fname_disk_to_usr(d->inode, 1062 (u32)le32_to_cpu(de->hash_code), 1063 0, &de_name, fstr); 1064 if (err) 1065 goto out; 1066 1067 de_name = *fstr; 1068 fstr->len = save_len; 1069 } 1070 1071 if (!dir_emit(ctx, de_name.name, de_name.len, 1072 le32_to_cpu(de->ino), d_type)) { 1073 err = 1; 1074 goto out; 1075 } 1076 1077 if (readdir_ra) 1078 f2fs_ra_node_page(sbi, le32_to_cpu(de->ino)); 1079 1080 ctx->pos = start_pos + bit_pos; 1081 found_valid_dirent = true; 1082 } 1083 out: 1084 if (readdir_ra) 1085 blk_finish_plug(&plug); 1086 return err; 1087 } 1088 1089 static int f2fs_readdir(struct file *file, struct dir_context *ctx) 1090 { 1091 struct inode *inode = file_inode(file); 1092 unsigned long npages = dir_blocks(inode); 1093 struct f2fs_dentry_block *dentry_blk = NULL; 1094 struct page *dentry_page = NULL; 1095 struct file_ra_state *ra = &file->f_ra; 1096 loff_t start_pos = ctx->pos; 1097 unsigned int n = ((unsigned long)ctx->pos / NR_DENTRY_IN_BLOCK); 1098 struct f2fs_dentry_ptr d; 1099 struct fscrypt_str fstr = FSTR_INIT(NULL, 0); 1100 int err = 0; 1101 1102 if (IS_ENCRYPTED(inode)) { 1103 err = fscrypt_prepare_readdir(inode); 1104 if (err) 1105 goto out; 1106 1107 err = fscrypt_fname_alloc_buffer(F2FS_NAME_LEN, &fstr); 1108 if (err < 0) 1109 goto out; 1110 } 1111 1112 if (f2fs_has_inline_dentry(inode)) { 1113 err = f2fs_read_inline_dir(file, ctx, &fstr); 1114 goto out_free; 1115 } 1116 1117 for (; n < npages; ctx->pos = n * NR_DENTRY_IN_BLOCK) { 1118 pgoff_t next_pgofs; 1119 1120 /* allow readdir() to be interrupted */ 1121 if (fatal_signal_pending(current)) { 1122 err = -ERESTARTSYS; 1123 goto out_free; 1124 } 1125 cond_resched(); 1126 1127 /* readahead for multi pages of dir */ 1128 if (npages - n > 1 && !ra_has_index(ra, n)) 1129 page_cache_sync_readahead(inode->i_mapping, ra, file, n, 1130 min(npages - n, (pgoff_t)MAX_DIR_RA_PAGES)); 1131 1132 dentry_page = f2fs_find_data_page(inode, n, &next_pgofs); 1133 if (IS_ERR(dentry_page)) { 1134 err = PTR_ERR(dentry_page); 1135 if (err == -ENOENT) { 1136 err = 0; 1137 n = next_pgofs; 1138 continue; 1139 } else { 1140 goto out_free; 1141 } 1142 } 1143 1144 dentry_blk = page_address(dentry_page); 1145 1146 make_dentry_ptr_block(inode, &d, dentry_blk); 1147 1148 err = f2fs_fill_dentries(ctx, &d, 1149 n * NR_DENTRY_IN_BLOCK, &fstr); 1150 if (err) { 1151 f2fs_put_page(dentry_page, 0); 1152 break; 1153 } 1154 1155 f2fs_put_page(dentry_page, 0); 1156 1157 n++; 1158 } 1159 out_free: 1160 fscrypt_fname_free_buffer(&fstr); 1161 out: 1162 trace_f2fs_readdir(inode, start_pos, ctx->pos, err); 1163 return err < 0 ? err : 0; 1164 } 1165 1166 const struct file_operations f2fs_dir_operations = { 1167 .llseek = generic_file_llseek, 1168 .read = generic_read_dir, 1169 .iterate_shared = f2fs_readdir, 1170 .fsync = f2fs_sync_file, 1171 .unlocked_ioctl = f2fs_ioctl, 1172 #ifdef CONFIG_COMPAT 1173 .compat_ioctl = f2fs_compat_ioctl, 1174 #endif 1175 }; 1176