1 /** 2 * attrib.c - NTFS attribute operations. Part of the Linux-NTFS project. 3 * 4 * Copyright (c) 2001-2005 Anton Altaparmakov 5 * Copyright (c) 2002 Richard Russon 6 * 7 * This program/include file is free software; you can redistribute it and/or 8 * modify it under the terms of the GNU General Public License as published 9 * by the Free Software Foundation; either version 2 of the License, or 10 * (at your option) any later version. 11 * 12 * This program/include file is distributed in the hope that it will be 13 * useful, but WITHOUT ANY WARRANTY; without even the implied warranty 14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 15 * GNU General Public License for more details. 16 * 17 * You should have received a copy of the GNU General Public License 18 * along with this program (in the main directory of the Linux-NTFS 19 * distribution in the file COPYING); if not, write to the Free Software 20 * Foundation,Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 21 */ 22 23 #include <linux/buffer_head.h> 24 #include <linux/sched.h> 25 #include <linux/swap.h> 26 #include <linux/writeback.h> 27 28 #include "attrib.h" 29 #include "debug.h" 30 #include "layout.h" 31 #include "lcnalloc.h" 32 #include "malloc.h" 33 #include "mft.h" 34 #include "ntfs.h" 35 #include "types.h" 36 37 /** 38 * ntfs_map_runlist_nolock - map (a part of) a runlist of an ntfs inode 39 * @ni: ntfs inode for which to map (part of) a runlist 40 * @vcn: map runlist part containing this vcn 41 * @ctx: active attribute search context if present or NULL if not 42 * 43 * Map the part of a runlist containing the @vcn of the ntfs inode @ni. 44 * 45 * If @ctx is specified, it is an active search context of @ni and its base mft 46 * record. This is needed when ntfs_map_runlist_nolock() encounters unmapped 47 * runlist fragments and allows their mapping. If you do not have the mft 48 * record mapped, you can specify @ctx as NULL and ntfs_map_runlist_nolock() 49 * will perform the necessary mapping and unmapping. 50 * 51 * Note, ntfs_map_runlist_nolock() saves the state of @ctx on entry and 52 * restores it before returning. Thus, @ctx will be left pointing to the same 53 * attribute on return as on entry. However, the actual pointers in @ctx may 54 * point to different memory locations on return, so you must remember to reset 55 * any cached pointers from the @ctx, i.e. after the call to 56 * ntfs_map_runlist_nolock(), you will probably want to do: 57 * m = ctx->mrec; 58 * a = ctx->attr; 59 * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that 60 * you cache ctx->mrec in a variable @m of type MFT_RECORD *. 61 * 62 * Return 0 on success and -errno on error. There is one special error code 63 * which is not an error as such. This is -ENOENT. It means that @vcn is out 64 * of bounds of the runlist. 65 * 66 * Note the runlist can be NULL after this function returns if @vcn is zero and 67 * the attribute has zero allocated size, i.e. there simply is no runlist. 68 * 69 * WARNING: If @ctx is supplied, regardless of whether success or failure is 70 * returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx 71 * is no longer valid, i.e. you need to either call 72 * ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it. 73 * In that case PTR_ERR(@ctx->mrec) will give you the error code for 74 * why the mapping of the old inode failed. 75 * 76 * Locking: - The runlist described by @ni must be locked for writing on entry 77 * and is locked on return. Note the runlist will be modified. 78 * - If @ctx is NULL, the base mft record of @ni must not be mapped on 79 * entry and it will be left unmapped on return. 80 * - If @ctx is not NULL, the base mft record must be mapped on entry 81 * and it will be left mapped on return. 82 */ 83 int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn, ntfs_attr_search_ctx *ctx) 84 { 85 VCN end_vcn; 86 unsigned long flags; 87 ntfs_inode *base_ni; 88 MFT_RECORD *m; 89 ATTR_RECORD *a; 90 runlist_element *rl; 91 struct page *put_this_page = NULL; 92 int err = 0; 93 BOOL ctx_is_temporary, ctx_needs_reset; 94 ntfs_attr_search_ctx old_ctx = { NULL, }; 95 96 ntfs_debug("Mapping runlist part containing vcn 0x%llx.", 97 (unsigned long long)vcn); 98 if (!NInoAttr(ni)) 99 base_ni = ni; 100 else 101 base_ni = ni->ext.base_ntfs_ino; 102 if (!ctx) { 103 ctx_is_temporary = ctx_needs_reset = TRUE; 104 m = map_mft_record(base_ni); 105 if (IS_ERR(m)) 106 return PTR_ERR(m); 107 ctx = ntfs_attr_get_search_ctx(base_ni, m); 108 if (unlikely(!ctx)) { 109 err = -ENOMEM; 110 goto err_out; 111 } 112 } else { 113 VCN allocated_size_vcn; 114 115 BUG_ON(IS_ERR(ctx->mrec)); 116 a = ctx->attr; 117 BUG_ON(!a->non_resident); 118 ctx_is_temporary = FALSE; 119 end_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn); 120 read_lock_irqsave(&ni->size_lock, flags); 121 allocated_size_vcn = ni->allocated_size >> 122 ni->vol->cluster_size_bits; 123 read_unlock_irqrestore(&ni->size_lock, flags); 124 if (!a->data.non_resident.lowest_vcn && end_vcn <= 0) 125 end_vcn = allocated_size_vcn - 1; 126 /* 127 * If we already have the attribute extent containing @vcn in 128 * @ctx, no need to look it up again. We slightly cheat in 129 * that if vcn exceeds the allocated size, we will refuse to 130 * map the runlist below, so there is definitely no need to get 131 * the right attribute extent. 132 */ 133 if (vcn >= allocated_size_vcn || (a->type == ni->type && 134 a->name_length == ni->name_len && 135 !memcmp((u8*)a + le16_to_cpu(a->name_offset), 136 ni->name, ni->name_len) && 137 sle64_to_cpu(a->data.non_resident.lowest_vcn) 138 <= vcn && end_vcn >= vcn)) 139 ctx_needs_reset = FALSE; 140 else { 141 /* Save the old search context. */ 142 old_ctx = *ctx; 143 /* 144 * If the currently mapped (extent) inode is not the 145 * base inode we will unmap it when we reinitialize the 146 * search context which means we need to get a 147 * reference to the page containing the mapped mft 148 * record so we do not accidentally drop changes to the 149 * mft record when it has not been marked dirty yet. 150 */ 151 if (old_ctx.base_ntfs_ino && old_ctx.ntfs_ino != 152 old_ctx.base_ntfs_ino) { 153 put_this_page = old_ctx.ntfs_ino->page; 154 page_cache_get(put_this_page); 155 } 156 /* 157 * Reinitialize the search context so we can lookup the 158 * needed attribute extent. 159 */ 160 ntfs_attr_reinit_search_ctx(ctx); 161 ctx_needs_reset = TRUE; 162 } 163 } 164 if (ctx_needs_reset) { 165 err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, 166 CASE_SENSITIVE, vcn, NULL, 0, ctx); 167 if (unlikely(err)) { 168 if (err == -ENOENT) 169 err = -EIO; 170 goto err_out; 171 } 172 BUG_ON(!ctx->attr->non_resident); 173 } 174 a = ctx->attr; 175 /* 176 * Only decompress the mapping pairs if @vcn is inside it. Otherwise 177 * we get into problems when we try to map an out of bounds vcn because 178 * we then try to map the already mapped runlist fragment and 179 * ntfs_mapping_pairs_decompress() fails. 180 */ 181 end_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn) + 1; 182 if (!a->data.non_resident.lowest_vcn && end_vcn == 1) 183 end_vcn = sle64_to_cpu(a->data.non_resident.allocated_size) >> 184 ni->vol->cluster_size_bits; 185 if (unlikely(vcn >= end_vcn)) { 186 err = -ENOENT; 187 goto err_out; 188 } 189 rl = ntfs_mapping_pairs_decompress(ni->vol, a, ni->runlist.rl); 190 if (IS_ERR(rl)) 191 err = PTR_ERR(rl); 192 else 193 ni->runlist.rl = rl; 194 err_out: 195 if (ctx_is_temporary) { 196 if (likely(ctx)) 197 ntfs_attr_put_search_ctx(ctx); 198 unmap_mft_record(base_ni); 199 } else if (ctx_needs_reset) { 200 /* 201 * If there is no attribute list, restoring the search context 202 * is acomplished simply by copying the saved context back over 203 * the caller supplied context. If there is an attribute list, 204 * things are more complicated as we need to deal with mapping 205 * of mft records and resulting potential changes in pointers. 206 */ 207 if (NInoAttrList(base_ni)) { 208 /* 209 * If the currently mapped (extent) inode is not the 210 * one we had before, we need to unmap it and map the 211 * old one. 212 */ 213 if (ctx->ntfs_ino != old_ctx.ntfs_ino) { 214 /* 215 * If the currently mapped inode is not the 216 * base inode, unmap it. 217 */ 218 if (ctx->base_ntfs_ino && ctx->ntfs_ino != 219 ctx->base_ntfs_ino) { 220 unmap_extent_mft_record(ctx->ntfs_ino); 221 ctx->mrec = ctx->base_mrec; 222 BUG_ON(!ctx->mrec); 223 } 224 /* 225 * If the old mapped inode is not the base 226 * inode, map it. 227 */ 228 if (old_ctx.base_ntfs_ino && 229 old_ctx.ntfs_ino != 230 old_ctx.base_ntfs_ino) { 231 retry_map: 232 ctx->mrec = map_mft_record( 233 old_ctx.ntfs_ino); 234 /* 235 * Something bad has happened. If out 236 * of memory retry till it succeeds. 237 * Any other errors are fatal and we 238 * return the error code in ctx->mrec. 239 * Let the caller deal with it... We 240 * just need to fudge things so the 241 * caller can reinit and/or put the 242 * search context safely. 243 */ 244 if (IS_ERR(ctx->mrec)) { 245 if (PTR_ERR(ctx->mrec) == 246 -ENOMEM) { 247 schedule(); 248 goto retry_map; 249 } else 250 old_ctx.ntfs_ino = 251 old_ctx. 252 base_ntfs_ino; 253 } 254 } 255 } 256 /* Update the changed pointers in the saved context. */ 257 if (ctx->mrec != old_ctx.mrec) { 258 if (!IS_ERR(ctx->mrec)) 259 old_ctx.attr = (ATTR_RECORD*)( 260 (u8*)ctx->mrec + 261 ((u8*)old_ctx.attr - 262 (u8*)old_ctx.mrec)); 263 old_ctx.mrec = ctx->mrec; 264 } 265 } 266 /* Restore the search context to the saved one. */ 267 *ctx = old_ctx; 268 /* 269 * We drop the reference on the page we took earlier. In the 270 * case that IS_ERR(ctx->mrec) is true this means we might lose 271 * some changes to the mft record that had been made between 272 * the last time it was marked dirty/written out and now. This 273 * at this stage is not a problem as the mapping error is fatal 274 * enough that the mft record cannot be written out anyway and 275 * the caller is very likely to shutdown the whole inode 276 * immediately and mark the volume dirty for chkdsk to pick up 277 * the pieces anyway. 278 */ 279 if (put_this_page) 280 page_cache_release(put_this_page); 281 } 282 return err; 283 } 284 285 /** 286 * ntfs_map_runlist - map (a part of) a runlist of an ntfs inode 287 * @ni: ntfs inode for which to map (part of) a runlist 288 * @vcn: map runlist part containing this vcn 289 * 290 * Map the part of a runlist containing the @vcn of the ntfs inode @ni. 291 * 292 * Return 0 on success and -errno on error. There is one special error code 293 * which is not an error as such. This is -ENOENT. It means that @vcn is out 294 * of bounds of the runlist. 295 * 296 * Locking: - The runlist must be unlocked on entry and is unlocked on return. 297 * - This function takes the runlist lock for writing and may modify 298 * the runlist. 299 */ 300 int ntfs_map_runlist(ntfs_inode *ni, VCN vcn) 301 { 302 int err = 0; 303 304 down_write(&ni->runlist.lock); 305 /* Make sure someone else didn't do the work while we were sleeping. */ 306 if (likely(ntfs_rl_vcn_to_lcn(ni->runlist.rl, vcn) <= 307 LCN_RL_NOT_MAPPED)) 308 err = ntfs_map_runlist_nolock(ni, vcn, NULL); 309 up_write(&ni->runlist.lock); 310 return err; 311 } 312 313 /** 314 * ntfs_attr_vcn_to_lcn_nolock - convert a vcn into a lcn given an ntfs inode 315 * @ni: ntfs inode of the attribute whose runlist to search 316 * @vcn: vcn to convert 317 * @write_locked: true if the runlist is locked for writing 318 * 319 * Find the virtual cluster number @vcn in the runlist of the ntfs attribute 320 * described by the ntfs inode @ni and return the corresponding logical cluster 321 * number (lcn). 322 * 323 * If the @vcn is not mapped yet, the attempt is made to map the attribute 324 * extent containing the @vcn and the vcn to lcn conversion is retried. 325 * 326 * If @write_locked is true the caller has locked the runlist for writing and 327 * if false for reading. 328 * 329 * Since lcns must be >= 0, we use negative return codes with special meaning: 330 * 331 * Return code Meaning / Description 332 * ========================================== 333 * LCN_HOLE Hole / not allocated on disk. 334 * LCN_ENOENT There is no such vcn in the runlist, i.e. @vcn is out of bounds. 335 * LCN_ENOMEM Not enough memory to map runlist. 336 * LCN_EIO Critical error (runlist/file is corrupt, i/o error, etc). 337 * 338 * Locking: - The runlist must be locked on entry and is left locked on return. 339 * - If @write_locked is FALSE, i.e. the runlist is locked for reading, 340 * the lock may be dropped inside the function so you cannot rely on 341 * the runlist still being the same when this function returns. 342 */ 343 LCN ntfs_attr_vcn_to_lcn_nolock(ntfs_inode *ni, const VCN vcn, 344 const BOOL write_locked) 345 { 346 LCN lcn; 347 unsigned long flags; 348 BOOL is_retry = FALSE; 349 350 ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, %s_locked.", 351 ni->mft_no, (unsigned long long)vcn, 352 write_locked ? "write" : "read"); 353 BUG_ON(!ni); 354 BUG_ON(!NInoNonResident(ni)); 355 BUG_ON(vcn < 0); 356 if (!ni->runlist.rl) { 357 read_lock_irqsave(&ni->size_lock, flags); 358 if (!ni->allocated_size) { 359 read_unlock_irqrestore(&ni->size_lock, flags); 360 return LCN_ENOENT; 361 } 362 read_unlock_irqrestore(&ni->size_lock, flags); 363 } 364 retry_remap: 365 /* Convert vcn to lcn. If that fails map the runlist and retry once. */ 366 lcn = ntfs_rl_vcn_to_lcn(ni->runlist.rl, vcn); 367 if (likely(lcn >= LCN_HOLE)) { 368 ntfs_debug("Done, lcn 0x%llx.", (long long)lcn); 369 return lcn; 370 } 371 if (lcn != LCN_RL_NOT_MAPPED) { 372 if (lcn != LCN_ENOENT) 373 lcn = LCN_EIO; 374 } else if (!is_retry) { 375 int err; 376 377 if (!write_locked) { 378 up_read(&ni->runlist.lock); 379 down_write(&ni->runlist.lock); 380 if (unlikely(ntfs_rl_vcn_to_lcn(ni->runlist.rl, vcn) != 381 LCN_RL_NOT_MAPPED)) { 382 up_write(&ni->runlist.lock); 383 down_read(&ni->runlist.lock); 384 goto retry_remap; 385 } 386 } 387 err = ntfs_map_runlist_nolock(ni, vcn, NULL); 388 if (!write_locked) { 389 up_write(&ni->runlist.lock); 390 down_read(&ni->runlist.lock); 391 } 392 if (likely(!err)) { 393 is_retry = TRUE; 394 goto retry_remap; 395 } 396 if (err == -ENOENT) 397 lcn = LCN_ENOENT; 398 else if (err == -ENOMEM) 399 lcn = LCN_ENOMEM; 400 else 401 lcn = LCN_EIO; 402 } 403 if (lcn != LCN_ENOENT) 404 ntfs_error(ni->vol->sb, "Failed with error code %lli.", 405 (long long)lcn); 406 return lcn; 407 } 408 409 /** 410 * ntfs_attr_find_vcn_nolock - find a vcn in the runlist of an ntfs inode 411 * @ni: ntfs inode describing the runlist to search 412 * @vcn: vcn to find 413 * @ctx: active attribute search context if present or NULL if not 414 * 415 * Find the virtual cluster number @vcn in the runlist described by the ntfs 416 * inode @ni and return the address of the runlist element containing the @vcn. 417 * 418 * If the @vcn is not mapped yet, the attempt is made to map the attribute 419 * extent containing the @vcn and the vcn to lcn conversion is retried. 420 * 421 * If @ctx is specified, it is an active search context of @ni and its base mft 422 * record. This is needed when ntfs_attr_find_vcn_nolock() encounters unmapped 423 * runlist fragments and allows their mapping. If you do not have the mft 424 * record mapped, you can specify @ctx as NULL and ntfs_attr_find_vcn_nolock() 425 * will perform the necessary mapping and unmapping. 426 * 427 * Note, ntfs_attr_find_vcn_nolock() saves the state of @ctx on entry and 428 * restores it before returning. Thus, @ctx will be left pointing to the same 429 * attribute on return as on entry. However, the actual pointers in @ctx may 430 * point to different memory locations on return, so you must remember to reset 431 * any cached pointers from the @ctx, i.e. after the call to 432 * ntfs_attr_find_vcn_nolock(), you will probably want to do: 433 * m = ctx->mrec; 434 * a = ctx->attr; 435 * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that 436 * you cache ctx->mrec in a variable @m of type MFT_RECORD *. 437 * Note you need to distinguish between the lcn of the returned runlist element 438 * being >= 0 and LCN_HOLE. In the later case you have to return zeroes on 439 * read and allocate clusters on write. 440 * 441 * Return the runlist element containing the @vcn on success and 442 * ERR_PTR(-errno) on error. You need to test the return value with IS_ERR() 443 * to decide if the return is success or failure and PTR_ERR() to get to the 444 * error code if IS_ERR() is true. 445 * 446 * The possible error return codes are: 447 * -ENOENT - No such vcn in the runlist, i.e. @vcn is out of bounds. 448 * -ENOMEM - Not enough memory to map runlist. 449 * -EIO - Critical error (runlist/file is corrupt, i/o error, etc). 450 * 451 * WARNING: If @ctx is supplied, regardless of whether success or failure is 452 * returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx 453 * is no longer valid, i.e. you need to either call 454 * ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it. 455 * In that case PTR_ERR(@ctx->mrec) will give you the error code for 456 * why the mapping of the old inode failed. 457 * 458 * Locking: - The runlist described by @ni must be locked for writing on entry 459 * and is locked on return. Note the runlist may be modified when 460 * needed runlist fragments need to be mapped. 461 * - If @ctx is NULL, the base mft record of @ni must not be mapped on 462 * entry and it will be left unmapped on return. 463 * - If @ctx is not NULL, the base mft record must be mapped on entry 464 * and it will be left mapped on return. 465 */ 466 runlist_element *ntfs_attr_find_vcn_nolock(ntfs_inode *ni, const VCN vcn, 467 ntfs_attr_search_ctx *ctx) 468 { 469 unsigned long flags; 470 runlist_element *rl; 471 int err = 0; 472 BOOL is_retry = FALSE; 473 474 ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, with%s ctx.", 475 ni->mft_no, (unsigned long long)vcn, ctx ? "" : "out"); 476 BUG_ON(!ni); 477 BUG_ON(!NInoNonResident(ni)); 478 BUG_ON(vcn < 0); 479 if (!ni->runlist.rl) { 480 read_lock_irqsave(&ni->size_lock, flags); 481 if (!ni->allocated_size) { 482 read_unlock_irqrestore(&ni->size_lock, flags); 483 return ERR_PTR(-ENOENT); 484 } 485 read_unlock_irqrestore(&ni->size_lock, flags); 486 } 487 retry_remap: 488 rl = ni->runlist.rl; 489 if (likely(rl && vcn >= rl[0].vcn)) { 490 while (likely(rl->length)) { 491 if (unlikely(vcn < rl[1].vcn)) { 492 if (likely(rl->lcn >= LCN_HOLE)) { 493 ntfs_debug("Done."); 494 return rl; 495 } 496 break; 497 } 498 rl++; 499 } 500 if (likely(rl->lcn != LCN_RL_NOT_MAPPED)) { 501 if (likely(rl->lcn == LCN_ENOENT)) 502 err = -ENOENT; 503 else 504 err = -EIO; 505 } 506 } 507 if (!err && !is_retry) { 508 /* 509 * If the search context is invalid we cannot map the unmapped 510 * region. 511 */ 512 if (IS_ERR(ctx->mrec)) 513 err = PTR_ERR(ctx->mrec); 514 else { 515 /* 516 * The @vcn is in an unmapped region, map the runlist 517 * and retry. 518 */ 519 err = ntfs_map_runlist_nolock(ni, vcn, ctx); 520 if (likely(!err)) { 521 is_retry = TRUE; 522 goto retry_remap; 523 } 524 } 525 if (err == -EINVAL) 526 err = -EIO; 527 } else if (!err) 528 err = -EIO; 529 if (err != -ENOENT) 530 ntfs_error(ni->vol->sb, "Failed with error code %i.", err); 531 return ERR_PTR(err); 532 } 533 534 /** 535 * ntfs_attr_find - find (next) attribute in mft record 536 * @type: attribute type to find 537 * @name: attribute name to find (optional, i.e. NULL means don't care) 538 * @name_len: attribute name length (only needed if @name present) 539 * @ic: IGNORE_CASE or CASE_SENSITIVE (ignored if @name not present) 540 * @val: attribute value to find (optional, resident attributes only) 541 * @val_len: attribute value length 542 * @ctx: search context with mft record and attribute to search from 543 * 544 * You should not need to call this function directly. Use ntfs_attr_lookup() 545 * instead. 546 * 547 * ntfs_attr_find() takes a search context @ctx as parameter and searches the 548 * mft record specified by @ctx->mrec, beginning at @ctx->attr, for an 549 * attribute of @type, optionally @name and @val. 550 * 551 * If the attribute is found, ntfs_attr_find() returns 0 and @ctx->attr will 552 * point to the found attribute. 553 * 554 * If the attribute is not found, ntfs_attr_find() returns -ENOENT and 555 * @ctx->attr will point to the attribute before which the attribute being 556 * searched for would need to be inserted if such an action were to be desired. 557 * 558 * On actual error, ntfs_attr_find() returns -EIO. In this case @ctx->attr is 559 * undefined and in particular do not rely on it not changing. 560 * 561 * If @ctx->is_first is TRUE, the search begins with @ctx->attr itself. If it 562 * is FALSE, the search begins after @ctx->attr. 563 * 564 * If @ic is IGNORE_CASE, the @name comparisson is not case sensitive and 565 * @ctx->ntfs_ino must be set to the ntfs inode to which the mft record 566 * @ctx->mrec belongs. This is so we can get at the ntfs volume and hence at 567 * the upcase table. If @ic is CASE_SENSITIVE, the comparison is case 568 * sensitive. When @name is present, @name_len is the @name length in Unicode 569 * characters. 570 * 571 * If @name is not present (NULL), we assume that the unnamed attribute is 572 * being searched for. 573 * 574 * Finally, the resident attribute value @val is looked for, if present. If 575 * @val is not present (NULL), @val_len is ignored. 576 * 577 * ntfs_attr_find() only searches the specified mft record and it ignores the 578 * presence of an attribute list attribute (unless it is the one being searched 579 * for, obviously). If you need to take attribute lists into consideration, 580 * use ntfs_attr_lookup() instead (see below). This also means that you cannot 581 * use ntfs_attr_find() to search for extent records of non-resident 582 * attributes, as extents with lowest_vcn != 0 are usually described by the 583 * attribute list attribute only. - Note that it is possible that the first 584 * extent is only in the attribute list while the last extent is in the base 585 * mft record, so do not rely on being able to find the first extent in the 586 * base mft record. 587 * 588 * Warning: Never use @val when looking for attribute types which can be 589 * non-resident as this most likely will result in a crash! 590 */ 591 static int ntfs_attr_find(const ATTR_TYPE type, const ntfschar *name, 592 const u32 name_len, const IGNORE_CASE_BOOL ic, 593 const u8 *val, const u32 val_len, ntfs_attr_search_ctx *ctx) 594 { 595 ATTR_RECORD *a; 596 ntfs_volume *vol = ctx->ntfs_ino->vol; 597 ntfschar *upcase = vol->upcase; 598 u32 upcase_len = vol->upcase_len; 599 600 /* 601 * Iterate over attributes in mft record starting at @ctx->attr, or the 602 * attribute following that, if @ctx->is_first is TRUE. 603 */ 604 if (ctx->is_first) { 605 a = ctx->attr; 606 ctx->is_first = FALSE; 607 } else 608 a = (ATTR_RECORD*)((u8*)ctx->attr + 609 le32_to_cpu(ctx->attr->length)); 610 for (;; a = (ATTR_RECORD*)((u8*)a + le32_to_cpu(a->length))) { 611 if ((u8*)a < (u8*)ctx->mrec || (u8*)a > (u8*)ctx->mrec + 612 le32_to_cpu(ctx->mrec->bytes_allocated)) 613 break; 614 ctx->attr = a; 615 if (unlikely(le32_to_cpu(a->type) > le32_to_cpu(type) || 616 a->type == AT_END)) 617 return -ENOENT; 618 if (unlikely(!a->length)) 619 break; 620 if (a->type != type) 621 continue; 622 /* 623 * If @name is present, compare the two names. If @name is 624 * missing, assume we want an unnamed attribute. 625 */ 626 if (!name) { 627 /* The search failed if the found attribute is named. */ 628 if (a->name_length) 629 return -ENOENT; 630 } else if (!ntfs_are_names_equal(name, name_len, 631 (ntfschar*)((u8*)a + le16_to_cpu(a->name_offset)), 632 a->name_length, ic, upcase, upcase_len)) { 633 register int rc; 634 635 rc = ntfs_collate_names(name, name_len, 636 (ntfschar*)((u8*)a + 637 le16_to_cpu(a->name_offset)), 638 a->name_length, 1, IGNORE_CASE, 639 upcase, upcase_len); 640 /* 641 * If @name collates before a->name, there is no 642 * matching attribute. 643 */ 644 if (rc == -1) 645 return -ENOENT; 646 /* If the strings are not equal, continue search. */ 647 if (rc) 648 continue; 649 rc = ntfs_collate_names(name, name_len, 650 (ntfschar*)((u8*)a + 651 le16_to_cpu(a->name_offset)), 652 a->name_length, 1, CASE_SENSITIVE, 653 upcase, upcase_len); 654 if (rc == -1) 655 return -ENOENT; 656 if (rc) 657 continue; 658 } 659 /* 660 * The names match or @name not present and attribute is 661 * unnamed. If no @val specified, we have found the attribute 662 * and are done. 663 */ 664 if (!val) 665 return 0; 666 /* @val is present; compare values. */ 667 else { 668 register int rc; 669 670 rc = memcmp(val, (u8*)a + le16_to_cpu( 671 a->data.resident.value_offset), 672 min_t(u32, val_len, le32_to_cpu( 673 a->data.resident.value_length))); 674 /* 675 * If @val collates before the current attribute's 676 * value, there is no matching attribute. 677 */ 678 if (!rc) { 679 register u32 avl; 680 681 avl = le32_to_cpu( 682 a->data.resident.value_length); 683 if (val_len == avl) 684 return 0; 685 if (val_len < avl) 686 return -ENOENT; 687 } else if (rc < 0) 688 return -ENOENT; 689 } 690 } 691 ntfs_error(vol->sb, "Inode is corrupt. Run chkdsk."); 692 NVolSetErrors(vol); 693 return -EIO; 694 } 695 696 /** 697 * load_attribute_list - load an attribute list into memory 698 * @vol: ntfs volume from which to read 699 * @runlist: runlist of the attribute list 700 * @al_start: destination buffer 701 * @size: size of the destination buffer in bytes 702 * @initialized_size: initialized size of the attribute list 703 * 704 * Walk the runlist @runlist and load all clusters from it copying them into 705 * the linear buffer @al. The maximum number of bytes copied to @al is @size 706 * bytes. Note, @size does not need to be a multiple of the cluster size. If 707 * @initialized_size is less than @size, the region in @al between 708 * @initialized_size and @size will be zeroed and not read from disk. 709 * 710 * Return 0 on success or -errno on error. 711 */ 712 int load_attribute_list(ntfs_volume *vol, runlist *runlist, u8 *al_start, 713 const s64 size, const s64 initialized_size) 714 { 715 LCN lcn; 716 u8 *al = al_start; 717 u8 *al_end = al + initialized_size; 718 runlist_element *rl; 719 struct buffer_head *bh; 720 struct super_block *sb; 721 unsigned long block_size; 722 unsigned long block, max_block; 723 int err = 0; 724 unsigned char block_size_bits; 725 726 ntfs_debug("Entering."); 727 if (!vol || !runlist || !al || size <= 0 || initialized_size < 0 || 728 initialized_size > size) 729 return -EINVAL; 730 if (!initialized_size) { 731 memset(al, 0, size); 732 return 0; 733 } 734 sb = vol->sb; 735 block_size = sb->s_blocksize; 736 block_size_bits = sb->s_blocksize_bits; 737 down_read(&runlist->lock); 738 rl = runlist->rl; 739 if (!rl) { 740 ntfs_error(sb, "Cannot read attribute list since runlist is " 741 "missing."); 742 goto err_out; 743 } 744 /* Read all clusters specified by the runlist one run at a time. */ 745 while (rl->length) { 746 lcn = ntfs_rl_vcn_to_lcn(rl, rl->vcn); 747 ntfs_debug("Reading vcn = 0x%llx, lcn = 0x%llx.", 748 (unsigned long long)rl->vcn, 749 (unsigned long long)lcn); 750 /* The attribute list cannot be sparse. */ 751 if (lcn < 0) { 752 ntfs_error(sb, "ntfs_rl_vcn_to_lcn() failed. Cannot " 753 "read attribute list."); 754 goto err_out; 755 } 756 block = lcn << vol->cluster_size_bits >> block_size_bits; 757 /* Read the run from device in chunks of block_size bytes. */ 758 max_block = block + (rl->length << vol->cluster_size_bits >> 759 block_size_bits); 760 ntfs_debug("max_block = 0x%lx.", max_block); 761 do { 762 ntfs_debug("Reading block = 0x%lx.", block); 763 bh = sb_bread(sb, block); 764 if (!bh) { 765 ntfs_error(sb, "sb_bread() failed. Cannot " 766 "read attribute list."); 767 goto err_out; 768 } 769 if (al + block_size >= al_end) 770 goto do_final; 771 memcpy(al, bh->b_data, block_size); 772 brelse(bh); 773 al += block_size; 774 } while (++block < max_block); 775 rl++; 776 } 777 if (initialized_size < size) { 778 initialize: 779 memset(al_start + initialized_size, 0, size - initialized_size); 780 } 781 done: 782 up_read(&runlist->lock); 783 return err; 784 do_final: 785 if (al < al_end) { 786 /* 787 * Partial block. 788 * 789 * Note: The attribute list can be smaller than its allocation 790 * by multiple clusters. This has been encountered by at least 791 * two people running Windows XP, thus we cannot do any 792 * truncation sanity checking here. (AIA) 793 */ 794 memcpy(al, bh->b_data, al_end - al); 795 brelse(bh); 796 if (initialized_size < size) 797 goto initialize; 798 goto done; 799 } 800 brelse(bh); 801 /* Real overflow! */ 802 ntfs_error(sb, "Attribute list buffer overflow. Read attribute list " 803 "is truncated."); 804 err_out: 805 err = -EIO; 806 goto done; 807 } 808 809 /** 810 * ntfs_external_attr_find - find an attribute in the attribute list of an inode 811 * @type: attribute type to find 812 * @name: attribute name to find (optional, i.e. NULL means don't care) 813 * @name_len: attribute name length (only needed if @name present) 814 * @ic: IGNORE_CASE or CASE_SENSITIVE (ignored if @name not present) 815 * @lowest_vcn: lowest vcn to find (optional, non-resident attributes only) 816 * @val: attribute value to find (optional, resident attributes only) 817 * @val_len: attribute value length 818 * @ctx: search context with mft record and attribute to search from 819 * 820 * You should not need to call this function directly. Use ntfs_attr_lookup() 821 * instead. 822 * 823 * Find an attribute by searching the attribute list for the corresponding 824 * attribute list entry. Having found the entry, map the mft record if the 825 * attribute is in a different mft record/inode, ntfs_attr_find() the attribute 826 * in there and return it. 827 * 828 * On first search @ctx->ntfs_ino must be the base mft record and @ctx must 829 * have been obtained from a call to ntfs_attr_get_search_ctx(). On subsequent 830 * calls @ctx->ntfs_ino can be any extent inode, too (@ctx->base_ntfs_ino is 831 * then the base inode). 832 * 833 * After finishing with the attribute/mft record you need to call 834 * ntfs_attr_put_search_ctx() to cleanup the search context (unmapping any 835 * mapped inodes, etc). 836 * 837 * If the attribute is found, ntfs_external_attr_find() returns 0 and 838 * @ctx->attr will point to the found attribute. @ctx->mrec will point to the 839 * mft record in which @ctx->attr is located and @ctx->al_entry will point to 840 * the attribute list entry for the attribute. 841 * 842 * If the attribute is not found, ntfs_external_attr_find() returns -ENOENT and 843 * @ctx->attr will point to the attribute in the base mft record before which 844 * the attribute being searched for would need to be inserted if such an action 845 * were to be desired. @ctx->mrec will point to the mft record in which 846 * @ctx->attr is located and @ctx->al_entry will point to the attribute list 847 * entry of the attribute before which the attribute being searched for would 848 * need to be inserted if such an action were to be desired. 849 * 850 * Thus to insert the not found attribute, one wants to add the attribute to 851 * @ctx->mrec (the base mft record) and if there is not enough space, the 852 * attribute should be placed in a newly allocated extent mft record. The 853 * attribute list entry for the inserted attribute should be inserted in the 854 * attribute list attribute at @ctx->al_entry. 855 * 856 * On actual error, ntfs_external_attr_find() returns -EIO. In this case 857 * @ctx->attr is undefined and in particular do not rely on it not changing. 858 */ 859 static int ntfs_external_attr_find(const ATTR_TYPE type, 860 const ntfschar *name, const u32 name_len, 861 const IGNORE_CASE_BOOL ic, const VCN lowest_vcn, 862 const u8 *val, const u32 val_len, ntfs_attr_search_ctx *ctx) 863 { 864 ntfs_inode *base_ni, *ni; 865 ntfs_volume *vol; 866 ATTR_LIST_ENTRY *al_entry, *next_al_entry; 867 u8 *al_start, *al_end; 868 ATTR_RECORD *a; 869 ntfschar *al_name; 870 u32 al_name_len; 871 int err = 0; 872 static const char *es = " Unmount and run chkdsk."; 873 874 ni = ctx->ntfs_ino; 875 base_ni = ctx->base_ntfs_ino; 876 ntfs_debug("Entering for inode 0x%lx, type 0x%x.", ni->mft_no, type); 877 if (!base_ni) { 878 /* First call happens with the base mft record. */ 879 base_ni = ctx->base_ntfs_ino = ctx->ntfs_ino; 880 ctx->base_mrec = ctx->mrec; 881 } 882 if (ni == base_ni) 883 ctx->base_attr = ctx->attr; 884 if (type == AT_END) 885 goto not_found; 886 vol = base_ni->vol; 887 al_start = base_ni->attr_list; 888 al_end = al_start + base_ni->attr_list_size; 889 if (!ctx->al_entry) 890 ctx->al_entry = (ATTR_LIST_ENTRY*)al_start; 891 /* 892 * Iterate over entries in attribute list starting at @ctx->al_entry, 893 * or the entry following that, if @ctx->is_first is TRUE. 894 */ 895 if (ctx->is_first) { 896 al_entry = ctx->al_entry; 897 ctx->is_first = FALSE; 898 } else 899 al_entry = (ATTR_LIST_ENTRY*)((u8*)ctx->al_entry + 900 le16_to_cpu(ctx->al_entry->length)); 901 for (;; al_entry = next_al_entry) { 902 /* Out of bounds check. */ 903 if ((u8*)al_entry < base_ni->attr_list || 904 (u8*)al_entry > al_end) 905 break; /* Inode is corrupt. */ 906 ctx->al_entry = al_entry; 907 /* Catch the end of the attribute list. */ 908 if ((u8*)al_entry == al_end) 909 goto not_found; 910 if (!al_entry->length) 911 break; 912 if ((u8*)al_entry + 6 > al_end || (u8*)al_entry + 913 le16_to_cpu(al_entry->length) > al_end) 914 break; 915 next_al_entry = (ATTR_LIST_ENTRY*)((u8*)al_entry + 916 le16_to_cpu(al_entry->length)); 917 if (le32_to_cpu(al_entry->type) > le32_to_cpu(type)) 918 goto not_found; 919 if (type != al_entry->type) 920 continue; 921 /* 922 * If @name is present, compare the two names. If @name is 923 * missing, assume we want an unnamed attribute. 924 */ 925 al_name_len = al_entry->name_length; 926 al_name = (ntfschar*)((u8*)al_entry + al_entry->name_offset); 927 if (!name) { 928 if (al_name_len) 929 goto not_found; 930 } else if (!ntfs_are_names_equal(al_name, al_name_len, name, 931 name_len, ic, vol->upcase, vol->upcase_len)) { 932 register int rc; 933 934 rc = ntfs_collate_names(name, name_len, al_name, 935 al_name_len, 1, IGNORE_CASE, 936 vol->upcase, vol->upcase_len); 937 /* 938 * If @name collates before al_name, there is no 939 * matching attribute. 940 */ 941 if (rc == -1) 942 goto not_found; 943 /* If the strings are not equal, continue search. */ 944 if (rc) 945 continue; 946 /* 947 * FIXME: Reverse engineering showed 0, IGNORE_CASE but 948 * that is inconsistent with ntfs_attr_find(). The 949 * subsequent rc checks were also different. Perhaps I 950 * made a mistake in one of the two. Need to recheck 951 * which is correct or at least see what is going on... 952 * (AIA) 953 */ 954 rc = ntfs_collate_names(name, name_len, al_name, 955 al_name_len, 1, CASE_SENSITIVE, 956 vol->upcase, vol->upcase_len); 957 if (rc == -1) 958 goto not_found; 959 if (rc) 960 continue; 961 } 962 /* 963 * The names match or @name not present and attribute is 964 * unnamed. Now check @lowest_vcn. Continue search if the 965 * next attribute list entry still fits @lowest_vcn. Otherwise 966 * we have reached the right one or the search has failed. 967 */ 968 if (lowest_vcn && (u8*)next_al_entry >= al_start && 969 (u8*)next_al_entry + 6 < al_end && 970 (u8*)next_al_entry + le16_to_cpu( 971 next_al_entry->length) <= al_end && 972 sle64_to_cpu(next_al_entry->lowest_vcn) <= 973 lowest_vcn && 974 next_al_entry->type == al_entry->type && 975 next_al_entry->name_length == al_name_len && 976 ntfs_are_names_equal((ntfschar*)((u8*) 977 next_al_entry + 978 next_al_entry->name_offset), 979 next_al_entry->name_length, 980 al_name, al_name_len, CASE_SENSITIVE, 981 vol->upcase, vol->upcase_len)) 982 continue; 983 if (MREF_LE(al_entry->mft_reference) == ni->mft_no) { 984 if (MSEQNO_LE(al_entry->mft_reference) != ni->seq_no) { 985 ntfs_error(vol->sb, "Found stale mft " 986 "reference in attribute list " 987 "of base inode 0x%lx.%s", 988 base_ni->mft_no, es); 989 err = -EIO; 990 break; 991 } 992 } else { /* Mft references do not match. */ 993 /* If there is a mapped record unmap it first. */ 994 if (ni != base_ni) 995 unmap_extent_mft_record(ni); 996 /* Do we want the base record back? */ 997 if (MREF_LE(al_entry->mft_reference) == 998 base_ni->mft_no) { 999 ni = ctx->ntfs_ino = base_ni; 1000 ctx->mrec = ctx->base_mrec; 1001 } else { 1002 /* We want an extent record. */ 1003 ctx->mrec = map_extent_mft_record(base_ni, 1004 le64_to_cpu( 1005 al_entry->mft_reference), &ni); 1006 if (IS_ERR(ctx->mrec)) { 1007 ntfs_error(vol->sb, "Failed to map " 1008 "extent mft record " 1009 "0x%lx of base inode " 1010 "0x%lx.%s", 1011 MREF_LE(al_entry-> 1012 mft_reference), 1013 base_ni->mft_no, es); 1014 err = PTR_ERR(ctx->mrec); 1015 if (err == -ENOENT) 1016 err = -EIO; 1017 /* Cause @ctx to be sanitized below. */ 1018 ni = NULL; 1019 break; 1020 } 1021 ctx->ntfs_ino = ni; 1022 } 1023 ctx->attr = (ATTR_RECORD*)((u8*)ctx->mrec + 1024 le16_to_cpu(ctx->mrec->attrs_offset)); 1025 } 1026 /* 1027 * ctx->vfs_ino, ctx->mrec, and ctx->attr now point to the 1028 * mft record containing the attribute represented by the 1029 * current al_entry. 1030 */ 1031 /* 1032 * We could call into ntfs_attr_find() to find the right 1033 * attribute in this mft record but this would be less 1034 * efficient and not quite accurate as ntfs_attr_find() ignores 1035 * the attribute instance numbers for example which become 1036 * important when one plays with attribute lists. Also, 1037 * because a proper match has been found in the attribute list 1038 * entry above, the comparison can now be optimized. So it is 1039 * worth re-implementing a simplified ntfs_attr_find() here. 1040 */ 1041 a = ctx->attr; 1042 /* 1043 * Use a manual loop so we can still use break and continue 1044 * with the same meanings as above. 1045 */ 1046 do_next_attr_loop: 1047 if ((u8*)a < (u8*)ctx->mrec || (u8*)a > (u8*)ctx->mrec + 1048 le32_to_cpu(ctx->mrec->bytes_allocated)) 1049 break; 1050 if (a->type == AT_END) 1051 continue; 1052 if (!a->length) 1053 break; 1054 if (al_entry->instance != a->instance) 1055 goto do_next_attr; 1056 /* 1057 * If the type and/or the name are mismatched between the 1058 * attribute list entry and the attribute record, there is 1059 * corruption so we break and return error EIO. 1060 */ 1061 if (al_entry->type != a->type) 1062 break; 1063 if (!ntfs_are_names_equal((ntfschar*)((u8*)a + 1064 le16_to_cpu(a->name_offset)), a->name_length, 1065 al_name, al_name_len, CASE_SENSITIVE, 1066 vol->upcase, vol->upcase_len)) 1067 break; 1068 ctx->attr = a; 1069 /* 1070 * If no @val specified or @val specified and it matches, we 1071 * have found it! 1072 */ 1073 if (!val || (!a->non_resident && le32_to_cpu( 1074 a->data.resident.value_length) == val_len && 1075 !memcmp((u8*)a + 1076 le16_to_cpu(a->data.resident.value_offset), 1077 val, val_len))) { 1078 ntfs_debug("Done, found."); 1079 return 0; 1080 } 1081 do_next_attr: 1082 /* Proceed to the next attribute in the current mft record. */ 1083 a = (ATTR_RECORD*)((u8*)a + le32_to_cpu(a->length)); 1084 goto do_next_attr_loop; 1085 } 1086 if (!err) { 1087 ntfs_error(vol->sb, "Base inode 0x%lx contains corrupt " 1088 "attribute list attribute.%s", base_ni->mft_no, 1089 es); 1090 err = -EIO; 1091 } 1092 if (ni != base_ni) { 1093 if (ni) 1094 unmap_extent_mft_record(ni); 1095 ctx->ntfs_ino = base_ni; 1096 ctx->mrec = ctx->base_mrec; 1097 ctx->attr = ctx->base_attr; 1098 } 1099 if (err != -ENOMEM) 1100 NVolSetErrors(vol); 1101 return err; 1102 not_found: 1103 /* 1104 * If we were looking for AT_END, we reset the search context @ctx and 1105 * use ntfs_attr_find() to seek to the end of the base mft record. 1106 */ 1107 if (type == AT_END) { 1108 ntfs_attr_reinit_search_ctx(ctx); 1109 return ntfs_attr_find(AT_END, name, name_len, ic, val, val_len, 1110 ctx); 1111 } 1112 /* 1113 * The attribute was not found. Before we return, we want to ensure 1114 * @ctx->mrec and @ctx->attr indicate the position at which the 1115 * attribute should be inserted in the base mft record. Since we also 1116 * want to preserve @ctx->al_entry we cannot reinitialize the search 1117 * context using ntfs_attr_reinit_search_ctx() as this would set 1118 * @ctx->al_entry to NULL. Thus we do the necessary bits manually (see 1119 * ntfs_attr_init_search_ctx() below). Note, we _only_ preserve 1120 * @ctx->al_entry as the remaining fields (base_*) are identical to 1121 * their non base_ counterparts and we cannot set @ctx->base_attr 1122 * correctly yet as we do not know what @ctx->attr will be set to by 1123 * the call to ntfs_attr_find() below. 1124 */ 1125 if (ni != base_ni) 1126 unmap_extent_mft_record(ni); 1127 ctx->mrec = ctx->base_mrec; 1128 ctx->attr = (ATTR_RECORD*)((u8*)ctx->mrec + 1129 le16_to_cpu(ctx->mrec->attrs_offset)); 1130 ctx->is_first = TRUE; 1131 ctx->ntfs_ino = base_ni; 1132 ctx->base_ntfs_ino = NULL; 1133 ctx->base_mrec = NULL; 1134 ctx->base_attr = NULL; 1135 /* 1136 * In case there are multiple matches in the base mft record, need to 1137 * keep enumerating until we get an attribute not found response (or 1138 * another error), otherwise we would keep returning the same attribute 1139 * over and over again and all programs using us for enumeration would 1140 * lock up in a tight loop. 1141 */ 1142 do { 1143 err = ntfs_attr_find(type, name, name_len, ic, val, val_len, 1144 ctx); 1145 } while (!err); 1146 ntfs_debug("Done, not found."); 1147 return err; 1148 } 1149 1150 /** 1151 * ntfs_attr_lookup - find an attribute in an ntfs inode 1152 * @type: attribute type to find 1153 * @name: attribute name to find (optional, i.e. NULL means don't care) 1154 * @name_len: attribute name length (only needed if @name present) 1155 * @ic: IGNORE_CASE or CASE_SENSITIVE (ignored if @name not present) 1156 * @lowest_vcn: lowest vcn to find (optional, non-resident attributes only) 1157 * @val: attribute value to find (optional, resident attributes only) 1158 * @val_len: attribute value length 1159 * @ctx: search context with mft record and attribute to search from 1160 * 1161 * Find an attribute in an ntfs inode. On first search @ctx->ntfs_ino must 1162 * be the base mft record and @ctx must have been obtained from a call to 1163 * ntfs_attr_get_search_ctx(). 1164 * 1165 * This function transparently handles attribute lists and @ctx is used to 1166 * continue searches where they were left off at. 1167 * 1168 * After finishing with the attribute/mft record you need to call 1169 * ntfs_attr_put_search_ctx() to cleanup the search context (unmapping any 1170 * mapped inodes, etc). 1171 * 1172 * Return 0 if the search was successful and -errno if not. 1173 * 1174 * When 0, @ctx->attr is the found attribute and it is in mft record 1175 * @ctx->mrec. If an attribute list attribute is present, @ctx->al_entry is 1176 * the attribute list entry of the found attribute. 1177 * 1178 * When -ENOENT, @ctx->attr is the attribute which collates just after the 1179 * attribute being searched for, i.e. if one wants to add the attribute to the 1180 * mft record this is the correct place to insert it into. If an attribute 1181 * list attribute is present, @ctx->al_entry is the attribute list entry which 1182 * collates just after the attribute list entry of the attribute being searched 1183 * for, i.e. if one wants to add the attribute to the mft record this is the 1184 * correct place to insert its attribute list entry into. 1185 * 1186 * When -errno != -ENOENT, an error occured during the lookup. @ctx->attr is 1187 * then undefined and in particular you should not rely on it not changing. 1188 */ 1189 int ntfs_attr_lookup(const ATTR_TYPE type, const ntfschar *name, 1190 const u32 name_len, const IGNORE_CASE_BOOL ic, 1191 const VCN lowest_vcn, const u8 *val, const u32 val_len, 1192 ntfs_attr_search_ctx *ctx) 1193 { 1194 ntfs_inode *base_ni; 1195 1196 ntfs_debug("Entering."); 1197 BUG_ON(IS_ERR(ctx->mrec)); 1198 if (ctx->base_ntfs_ino) 1199 base_ni = ctx->base_ntfs_ino; 1200 else 1201 base_ni = ctx->ntfs_ino; 1202 /* Sanity check, just for debugging really. */ 1203 BUG_ON(!base_ni); 1204 if (!NInoAttrList(base_ni) || type == AT_ATTRIBUTE_LIST) 1205 return ntfs_attr_find(type, name, name_len, ic, val, val_len, 1206 ctx); 1207 return ntfs_external_attr_find(type, name, name_len, ic, lowest_vcn, 1208 val, val_len, ctx); 1209 } 1210 1211 /** 1212 * ntfs_attr_init_search_ctx - initialize an attribute search context 1213 * @ctx: attribute search context to initialize 1214 * @ni: ntfs inode with which to initialize the search context 1215 * @mrec: mft record with which to initialize the search context 1216 * 1217 * Initialize the attribute search context @ctx with @ni and @mrec. 1218 */ 1219 static inline void ntfs_attr_init_search_ctx(ntfs_attr_search_ctx *ctx, 1220 ntfs_inode *ni, MFT_RECORD *mrec) 1221 { 1222 *ctx = (ntfs_attr_search_ctx) { 1223 .mrec = mrec, 1224 /* Sanity checks are performed elsewhere. */ 1225 .attr = (ATTR_RECORD*)((u8*)mrec + 1226 le16_to_cpu(mrec->attrs_offset)), 1227 .is_first = TRUE, 1228 .ntfs_ino = ni, 1229 }; 1230 } 1231 1232 /** 1233 * ntfs_attr_reinit_search_ctx - reinitialize an attribute search context 1234 * @ctx: attribute search context to reinitialize 1235 * 1236 * Reinitialize the attribute search context @ctx, unmapping an associated 1237 * extent mft record if present, and initialize the search context again. 1238 * 1239 * This is used when a search for a new attribute is being started to reset 1240 * the search context to the beginning. 1241 */ 1242 void ntfs_attr_reinit_search_ctx(ntfs_attr_search_ctx *ctx) 1243 { 1244 if (likely(!ctx->base_ntfs_ino)) { 1245 /* No attribute list. */ 1246 ctx->is_first = TRUE; 1247 /* Sanity checks are performed elsewhere. */ 1248 ctx->attr = (ATTR_RECORD*)((u8*)ctx->mrec + 1249 le16_to_cpu(ctx->mrec->attrs_offset)); 1250 /* 1251 * This needs resetting due to ntfs_external_attr_find() which 1252 * can leave it set despite having zeroed ctx->base_ntfs_ino. 1253 */ 1254 ctx->al_entry = NULL; 1255 return; 1256 } /* Attribute list. */ 1257 if (ctx->ntfs_ino != ctx->base_ntfs_ino) 1258 unmap_extent_mft_record(ctx->ntfs_ino); 1259 ntfs_attr_init_search_ctx(ctx, ctx->base_ntfs_ino, ctx->base_mrec); 1260 return; 1261 } 1262 1263 /** 1264 * ntfs_attr_get_search_ctx - allocate/initialize a new attribute search context 1265 * @ni: ntfs inode with which to initialize the search context 1266 * @mrec: mft record with which to initialize the search context 1267 * 1268 * Allocate a new attribute search context, initialize it with @ni and @mrec, 1269 * and return it. Return NULL if allocation failed. 1270 */ 1271 ntfs_attr_search_ctx *ntfs_attr_get_search_ctx(ntfs_inode *ni, MFT_RECORD *mrec) 1272 { 1273 ntfs_attr_search_ctx *ctx; 1274 1275 ctx = kmem_cache_alloc(ntfs_attr_ctx_cache, SLAB_NOFS); 1276 if (ctx) 1277 ntfs_attr_init_search_ctx(ctx, ni, mrec); 1278 return ctx; 1279 } 1280 1281 /** 1282 * ntfs_attr_put_search_ctx - release an attribute search context 1283 * @ctx: attribute search context to free 1284 * 1285 * Release the attribute search context @ctx, unmapping an associated extent 1286 * mft record if present. 1287 */ 1288 void ntfs_attr_put_search_ctx(ntfs_attr_search_ctx *ctx) 1289 { 1290 if (ctx->base_ntfs_ino && ctx->ntfs_ino != ctx->base_ntfs_ino) 1291 unmap_extent_mft_record(ctx->ntfs_ino); 1292 kmem_cache_free(ntfs_attr_ctx_cache, ctx); 1293 return; 1294 } 1295 1296 #ifdef NTFS_RW 1297 1298 /** 1299 * ntfs_attr_find_in_attrdef - find an attribute in the $AttrDef system file 1300 * @vol: ntfs volume to which the attribute belongs 1301 * @type: attribute type which to find 1302 * 1303 * Search for the attribute definition record corresponding to the attribute 1304 * @type in the $AttrDef system file. 1305 * 1306 * Return the attribute type definition record if found and NULL if not found. 1307 */ 1308 static ATTR_DEF *ntfs_attr_find_in_attrdef(const ntfs_volume *vol, 1309 const ATTR_TYPE type) 1310 { 1311 ATTR_DEF *ad; 1312 1313 BUG_ON(!vol->attrdef); 1314 BUG_ON(!type); 1315 for (ad = vol->attrdef; (u8*)ad - (u8*)vol->attrdef < 1316 vol->attrdef_size && ad->type; ++ad) { 1317 /* We have not found it yet, carry on searching. */ 1318 if (likely(le32_to_cpu(ad->type) < le32_to_cpu(type))) 1319 continue; 1320 /* We found the attribute; return it. */ 1321 if (likely(ad->type == type)) 1322 return ad; 1323 /* We have gone too far already. No point in continuing. */ 1324 break; 1325 } 1326 /* Attribute not found. */ 1327 ntfs_debug("Attribute type 0x%x not found in $AttrDef.", 1328 le32_to_cpu(type)); 1329 return NULL; 1330 } 1331 1332 /** 1333 * ntfs_attr_size_bounds_check - check a size of an attribute type for validity 1334 * @vol: ntfs volume to which the attribute belongs 1335 * @type: attribute type which to check 1336 * @size: size which to check 1337 * 1338 * Check whether the @size in bytes is valid for an attribute of @type on the 1339 * ntfs volume @vol. This information is obtained from $AttrDef system file. 1340 * 1341 * Return 0 if valid, -ERANGE if not valid, or -ENOENT if the attribute is not 1342 * listed in $AttrDef. 1343 */ 1344 int ntfs_attr_size_bounds_check(const ntfs_volume *vol, const ATTR_TYPE type, 1345 const s64 size) 1346 { 1347 ATTR_DEF *ad; 1348 1349 BUG_ON(size < 0); 1350 /* 1351 * $ATTRIBUTE_LIST has a maximum size of 256kiB, but this is not 1352 * listed in $AttrDef. 1353 */ 1354 if (unlikely(type == AT_ATTRIBUTE_LIST && size > 256 * 1024)) 1355 return -ERANGE; 1356 /* Get the $AttrDef entry for the attribute @type. */ 1357 ad = ntfs_attr_find_in_attrdef(vol, type); 1358 if (unlikely(!ad)) 1359 return -ENOENT; 1360 /* Do the bounds check. */ 1361 if (((sle64_to_cpu(ad->min_size) > 0) && 1362 size < sle64_to_cpu(ad->min_size)) || 1363 ((sle64_to_cpu(ad->max_size) > 0) && size > 1364 sle64_to_cpu(ad->max_size))) 1365 return -ERANGE; 1366 return 0; 1367 } 1368 1369 /** 1370 * ntfs_attr_can_be_non_resident - check if an attribute can be non-resident 1371 * @vol: ntfs volume to which the attribute belongs 1372 * @type: attribute type which to check 1373 * 1374 * Check whether the attribute of @type on the ntfs volume @vol is allowed to 1375 * be non-resident. This information is obtained from $AttrDef system file. 1376 * 1377 * Return 0 if the attribute is allowed to be non-resident, -EPERM if not, and 1378 * -ENOENT if the attribute is not listed in $AttrDef. 1379 */ 1380 int ntfs_attr_can_be_non_resident(const ntfs_volume *vol, const ATTR_TYPE type) 1381 { 1382 ATTR_DEF *ad; 1383 1384 /* Find the attribute definition record in $AttrDef. */ 1385 ad = ntfs_attr_find_in_attrdef(vol, type); 1386 if (unlikely(!ad)) 1387 return -ENOENT; 1388 /* Check the flags and return the result. */ 1389 if (ad->flags & ATTR_DEF_RESIDENT) 1390 return -EPERM; 1391 return 0; 1392 } 1393 1394 /** 1395 * ntfs_attr_can_be_resident - check if an attribute can be resident 1396 * @vol: ntfs volume to which the attribute belongs 1397 * @type: attribute type which to check 1398 * 1399 * Check whether the attribute of @type on the ntfs volume @vol is allowed to 1400 * be resident. This information is derived from our ntfs knowledge and may 1401 * not be completely accurate, especially when user defined attributes are 1402 * present. Basically we allow everything to be resident except for index 1403 * allocation and $EA attributes. 1404 * 1405 * Return 0 if the attribute is allowed to be non-resident and -EPERM if not. 1406 * 1407 * Warning: In the system file $MFT the attribute $Bitmap must be non-resident 1408 * otherwise windows will not boot (blue screen of death)! We cannot 1409 * check for this here as we do not know which inode's $Bitmap is 1410 * being asked about so the caller needs to special case this. 1411 */ 1412 int ntfs_attr_can_be_resident(const ntfs_volume *vol, const ATTR_TYPE type) 1413 { 1414 if (type == AT_INDEX_ALLOCATION) 1415 return -EPERM; 1416 return 0; 1417 } 1418 1419 /** 1420 * ntfs_attr_record_resize - resize an attribute record 1421 * @m: mft record containing attribute record 1422 * @a: attribute record to resize 1423 * @new_size: new size in bytes to which to resize the attribute record @a 1424 * 1425 * Resize the attribute record @a, i.e. the resident part of the attribute, in 1426 * the mft record @m to @new_size bytes. 1427 * 1428 * Return 0 on success and -errno on error. The following error codes are 1429 * defined: 1430 * -ENOSPC - Not enough space in the mft record @m to perform the resize. 1431 * 1432 * Note: On error, no modifications have been performed whatsoever. 1433 * 1434 * Warning: If you make a record smaller without having copied all the data you 1435 * are interested in the data may be overwritten. 1436 */ 1437 int ntfs_attr_record_resize(MFT_RECORD *m, ATTR_RECORD *a, u32 new_size) 1438 { 1439 ntfs_debug("Entering for new_size %u.", new_size); 1440 /* Align to 8 bytes if it is not already done. */ 1441 if (new_size & 7) 1442 new_size = (new_size + 7) & ~7; 1443 /* If the actual attribute length has changed, move things around. */ 1444 if (new_size != le32_to_cpu(a->length)) { 1445 u32 new_muse = le32_to_cpu(m->bytes_in_use) - 1446 le32_to_cpu(a->length) + new_size; 1447 /* Not enough space in this mft record. */ 1448 if (new_muse > le32_to_cpu(m->bytes_allocated)) 1449 return -ENOSPC; 1450 /* Move attributes following @a to their new location. */ 1451 memmove((u8*)a + new_size, (u8*)a + le32_to_cpu(a->length), 1452 le32_to_cpu(m->bytes_in_use) - ((u8*)a - 1453 (u8*)m) - le32_to_cpu(a->length)); 1454 /* Adjust @m to reflect the change in used space. */ 1455 m->bytes_in_use = cpu_to_le32(new_muse); 1456 /* Adjust @a to reflect the new size. */ 1457 if (new_size >= offsetof(ATTR_REC, length) + sizeof(a->length)) 1458 a->length = cpu_to_le32(new_size); 1459 } 1460 return 0; 1461 } 1462 1463 /** 1464 * ntfs_resident_attr_value_resize - resize the value of a resident attribute 1465 * @m: mft record containing attribute record 1466 * @a: attribute record whose value to resize 1467 * @new_size: new size in bytes to which to resize the attribute value of @a 1468 * 1469 * Resize the value of the attribute @a in the mft record @m to @new_size bytes. 1470 * If the value is made bigger, the newly allocated space is cleared. 1471 * 1472 * Return 0 on success and -errno on error. The following error codes are 1473 * defined: 1474 * -ENOSPC - Not enough space in the mft record @m to perform the resize. 1475 * 1476 * Note: On error, no modifications have been performed whatsoever. 1477 * 1478 * Warning: If you make a record smaller without having copied all the data you 1479 * are interested in the data may be overwritten. 1480 */ 1481 int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a, 1482 const u32 new_size) 1483 { 1484 u32 old_size; 1485 1486 /* Resize the resident part of the attribute record. */ 1487 if (ntfs_attr_record_resize(m, a, 1488 le16_to_cpu(a->data.resident.value_offset) + new_size)) 1489 return -ENOSPC; 1490 /* 1491 * The resize succeeded! If we made the attribute value bigger, clear 1492 * the area between the old size and @new_size. 1493 */ 1494 old_size = le32_to_cpu(a->data.resident.value_length); 1495 if (new_size > old_size) 1496 memset((u8*)a + le16_to_cpu(a->data.resident.value_offset) + 1497 old_size, 0, new_size - old_size); 1498 /* Finally update the length of the attribute value. */ 1499 a->data.resident.value_length = cpu_to_le32(new_size); 1500 return 0; 1501 } 1502 1503 /** 1504 * ntfs_attr_make_non_resident - convert a resident to a non-resident attribute 1505 * @ni: ntfs inode describing the attribute to convert 1506 * @data_size: size of the resident data to copy to the non-resident attribute 1507 * 1508 * Convert the resident ntfs attribute described by the ntfs inode @ni to a 1509 * non-resident one. 1510 * 1511 * @data_size must be equal to the attribute value size. This is needed since 1512 * we need to know the size before we can map the mft record and our callers 1513 * always know it. The reason we cannot simply read the size from the vfs 1514 * inode i_size is that this is not necessarily uptodate. This happens when 1515 * ntfs_attr_make_non_resident() is called in the ->truncate call path(s). 1516 * 1517 * Return 0 on success and -errno on error. The following error return codes 1518 * are defined: 1519 * -EPERM - The attribute is not allowed to be non-resident. 1520 * -ENOMEM - Not enough memory. 1521 * -ENOSPC - Not enough disk space. 1522 * -EINVAL - Attribute not defined on the volume. 1523 * -EIO - I/o error or other error. 1524 * Note that -ENOSPC is also returned in the case that there is not enough 1525 * space in the mft record to do the conversion. This can happen when the mft 1526 * record is already very full. The caller is responsible for trying to make 1527 * space in the mft record and trying again. FIXME: Do we need a separate 1528 * error return code for this kind of -ENOSPC or is it always worth trying 1529 * again in case the attribute may then fit in a resident state so no need to 1530 * make it non-resident at all? Ho-hum... (AIA) 1531 * 1532 * NOTE to self: No changes in the attribute list are required to move from 1533 * a resident to a non-resident attribute. 1534 * 1535 * Locking: - The caller must hold i_sem on the inode. 1536 */ 1537 int ntfs_attr_make_non_resident(ntfs_inode *ni, const u32 data_size) 1538 { 1539 s64 new_size; 1540 struct inode *vi = VFS_I(ni); 1541 ntfs_volume *vol = ni->vol; 1542 ntfs_inode *base_ni; 1543 MFT_RECORD *m; 1544 ATTR_RECORD *a; 1545 ntfs_attr_search_ctx *ctx; 1546 struct page *page; 1547 runlist_element *rl; 1548 u8 *kaddr; 1549 unsigned long flags; 1550 int mp_size, mp_ofs, name_ofs, arec_size, err, err2; 1551 u32 attr_size; 1552 u8 old_res_attr_flags; 1553 1554 /* Check that the attribute is allowed to be non-resident. */ 1555 err = ntfs_attr_can_be_non_resident(vol, ni->type); 1556 if (unlikely(err)) { 1557 if (err == -EPERM) 1558 ntfs_debug("Attribute is not allowed to be " 1559 "non-resident."); 1560 else 1561 ntfs_debug("Attribute not defined on the NTFS " 1562 "volume!"); 1563 return err; 1564 } 1565 /* 1566 * FIXME: Compressed and encrypted attributes are not supported when 1567 * writing and we should never have gotten here for them. 1568 */ 1569 BUG_ON(NInoCompressed(ni)); 1570 BUG_ON(NInoEncrypted(ni)); 1571 /* 1572 * The size needs to be aligned to a cluster boundary for allocation 1573 * purposes. 1574 */ 1575 new_size = (data_size + vol->cluster_size - 1) & 1576 ~(vol->cluster_size - 1); 1577 if (new_size > 0) { 1578 /* 1579 * Will need the page later and since the page lock nests 1580 * outside all ntfs locks, we need to get the page now. 1581 */ 1582 page = find_or_create_page(vi->i_mapping, 0, 1583 mapping_gfp_mask(vi->i_mapping)); 1584 if (unlikely(!page)) 1585 return -ENOMEM; 1586 /* Start by allocating clusters to hold the attribute value. */ 1587 rl = ntfs_cluster_alloc(vol, 0, new_size >> 1588 vol->cluster_size_bits, -1, DATA_ZONE, TRUE); 1589 if (IS_ERR(rl)) { 1590 err = PTR_ERR(rl); 1591 ntfs_debug("Failed to allocate cluster%s, error code " 1592 "%i.", (new_size >> 1593 vol->cluster_size_bits) > 1 ? "s" : "", 1594 err); 1595 goto page_err_out; 1596 } 1597 } else { 1598 rl = NULL; 1599 page = NULL; 1600 } 1601 /* Determine the size of the mapping pairs array. */ 1602 mp_size = ntfs_get_size_for_mapping_pairs(vol, rl, 0, -1); 1603 if (unlikely(mp_size < 0)) { 1604 err = mp_size; 1605 ntfs_debug("Failed to get size for mapping pairs array, error " 1606 "code %i.", err); 1607 goto rl_err_out; 1608 } 1609 down_write(&ni->runlist.lock); 1610 if (!NInoAttr(ni)) 1611 base_ni = ni; 1612 else 1613 base_ni = ni->ext.base_ntfs_ino; 1614 m = map_mft_record(base_ni); 1615 if (IS_ERR(m)) { 1616 err = PTR_ERR(m); 1617 m = NULL; 1618 ctx = NULL; 1619 goto err_out; 1620 } 1621 ctx = ntfs_attr_get_search_ctx(base_ni, m); 1622 if (unlikely(!ctx)) { 1623 err = -ENOMEM; 1624 goto err_out; 1625 } 1626 err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, 1627 CASE_SENSITIVE, 0, NULL, 0, ctx); 1628 if (unlikely(err)) { 1629 if (err == -ENOENT) 1630 err = -EIO; 1631 goto err_out; 1632 } 1633 m = ctx->mrec; 1634 a = ctx->attr; 1635 BUG_ON(NInoNonResident(ni)); 1636 BUG_ON(a->non_resident); 1637 /* 1638 * Calculate new offsets for the name and the mapping pairs array. 1639 */ 1640 if (NInoSparse(ni) || NInoCompressed(ni)) 1641 name_ofs = (offsetof(ATTR_REC, 1642 data.non_resident.compressed_size) + 1643 sizeof(a->data.non_resident.compressed_size) + 1644 7) & ~7; 1645 else 1646 name_ofs = (offsetof(ATTR_REC, 1647 data.non_resident.compressed_size) + 7) & ~7; 1648 mp_ofs = (name_ofs + a->name_length * sizeof(ntfschar) + 7) & ~7; 1649 /* 1650 * Determine the size of the resident part of the now non-resident 1651 * attribute record. 1652 */ 1653 arec_size = (mp_ofs + mp_size + 7) & ~7; 1654 /* 1655 * If the page is not uptodate bring it uptodate by copying from the 1656 * attribute value. 1657 */ 1658 attr_size = le32_to_cpu(a->data.resident.value_length); 1659 BUG_ON(attr_size != data_size); 1660 if (page && !PageUptodate(page)) { 1661 kaddr = kmap_atomic(page, KM_USER0); 1662 memcpy(kaddr, (u8*)a + 1663 le16_to_cpu(a->data.resident.value_offset), 1664 attr_size); 1665 memset(kaddr + attr_size, 0, PAGE_CACHE_SIZE - attr_size); 1666 kunmap_atomic(kaddr, KM_USER0); 1667 flush_dcache_page(page); 1668 SetPageUptodate(page); 1669 } 1670 /* Backup the attribute flag. */ 1671 old_res_attr_flags = a->data.resident.flags; 1672 /* Resize the resident part of the attribute record. */ 1673 err = ntfs_attr_record_resize(m, a, arec_size); 1674 if (unlikely(err)) 1675 goto err_out; 1676 /* 1677 * Convert the resident part of the attribute record to describe a 1678 * non-resident attribute. 1679 */ 1680 a->non_resident = 1; 1681 /* Move the attribute name if it exists and update the offset. */ 1682 if (a->name_length) 1683 memmove((u8*)a + name_ofs, (u8*)a + le16_to_cpu(a->name_offset), 1684 a->name_length * sizeof(ntfschar)); 1685 a->name_offset = cpu_to_le16(name_ofs); 1686 /* Setup the fields specific to non-resident attributes. */ 1687 a->data.non_resident.lowest_vcn = 0; 1688 a->data.non_resident.highest_vcn = cpu_to_sle64((new_size - 1) >> 1689 vol->cluster_size_bits); 1690 a->data.non_resident.mapping_pairs_offset = cpu_to_le16(mp_ofs); 1691 memset(&a->data.non_resident.reserved, 0, 1692 sizeof(a->data.non_resident.reserved)); 1693 a->data.non_resident.allocated_size = cpu_to_sle64(new_size); 1694 a->data.non_resident.data_size = 1695 a->data.non_resident.initialized_size = 1696 cpu_to_sle64(attr_size); 1697 if (NInoSparse(ni) || NInoCompressed(ni)) { 1698 a->data.non_resident.compression_unit = 4; 1699 a->data.non_resident.compressed_size = 1700 a->data.non_resident.allocated_size; 1701 } else 1702 a->data.non_resident.compression_unit = 0; 1703 /* Generate the mapping pairs array into the attribute record. */ 1704 err = ntfs_mapping_pairs_build(vol, (u8*)a + mp_ofs, 1705 arec_size - mp_ofs, rl, 0, -1, NULL); 1706 if (unlikely(err)) { 1707 ntfs_debug("Failed to build mapping pairs, error code %i.", 1708 err); 1709 goto undo_err_out; 1710 } 1711 /* Setup the in-memory attribute structure to be non-resident. */ 1712 ni->runlist.rl = rl; 1713 write_lock_irqsave(&ni->size_lock, flags); 1714 ni->allocated_size = new_size; 1715 if (NInoSparse(ni) || NInoCompressed(ni)) { 1716 ni->itype.compressed.size = ni->allocated_size; 1717 ni->itype.compressed.block_size = 1U << 1718 (a->data.non_resident.compression_unit + 1719 vol->cluster_size_bits); 1720 ni->itype.compressed.block_size_bits = 1721 ffs(ni->itype.compressed.block_size) - 1; 1722 ni->itype.compressed.block_clusters = 1U << 1723 a->data.non_resident.compression_unit; 1724 vi->i_blocks = ni->itype.compressed.size >> 9; 1725 } else 1726 vi->i_blocks = ni->allocated_size >> 9; 1727 write_unlock_irqrestore(&ni->size_lock, flags); 1728 /* 1729 * This needs to be last since the address space operations ->readpage 1730 * and ->writepage can run concurrently with us as they are not 1731 * serialized on i_sem. Note, we are not allowed to fail once we flip 1732 * this switch, which is another reason to do this last. 1733 */ 1734 NInoSetNonResident(ni); 1735 /* Mark the mft record dirty, so it gets written back. */ 1736 flush_dcache_mft_record_page(ctx->ntfs_ino); 1737 mark_mft_record_dirty(ctx->ntfs_ino); 1738 ntfs_attr_put_search_ctx(ctx); 1739 unmap_mft_record(base_ni); 1740 up_write(&ni->runlist.lock); 1741 if (page) { 1742 set_page_dirty(page); 1743 unlock_page(page); 1744 mark_page_accessed(page); 1745 page_cache_release(page); 1746 } 1747 ntfs_debug("Done."); 1748 return 0; 1749 undo_err_out: 1750 /* Convert the attribute back into a resident attribute. */ 1751 a->non_resident = 0; 1752 /* Move the attribute name if it exists and update the offset. */ 1753 name_ofs = (offsetof(ATTR_RECORD, data.resident.reserved) + 1754 sizeof(a->data.resident.reserved) + 7) & ~7; 1755 if (a->name_length) 1756 memmove((u8*)a + name_ofs, (u8*)a + le16_to_cpu(a->name_offset), 1757 a->name_length * sizeof(ntfschar)); 1758 mp_ofs = (name_ofs + a->name_length * sizeof(ntfschar) + 7) & ~7; 1759 a->name_offset = cpu_to_le16(name_ofs); 1760 arec_size = (mp_ofs + attr_size + 7) & ~7; 1761 /* Resize the resident part of the attribute record. */ 1762 err2 = ntfs_attr_record_resize(m, a, arec_size); 1763 if (unlikely(err2)) { 1764 /* 1765 * This cannot happen (well if memory corruption is at work it 1766 * could happen in theory), but deal with it as well as we can. 1767 * If the old size is too small, truncate the attribute, 1768 * otherwise simply give it a larger allocated size. 1769 * FIXME: Should check whether chkdsk complains when the 1770 * allocated size is much bigger than the resident value size. 1771 */ 1772 arec_size = le32_to_cpu(a->length); 1773 if ((mp_ofs + attr_size) > arec_size) { 1774 err2 = attr_size; 1775 attr_size = arec_size - mp_ofs; 1776 ntfs_error(vol->sb, "Failed to undo partial resident " 1777 "to non-resident attribute " 1778 "conversion. Truncating inode 0x%lx, " 1779 "attribute type 0x%x from %i bytes to " 1780 "%i bytes to maintain metadata " 1781 "consistency. THIS MEANS YOU ARE " 1782 "LOSING %i BYTES DATA FROM THIS %s.", 1783 vi->i_ino, 1784 (unsigned)le32_to_cpu(ni->type), 1785 err2, attr_size, err2 - attr_size, 1786 ((ni->type == AT_DATA) && 1787 !ni->name_len) ? "FILE": "ATTRIBUTE"); 1788 write_lock_irqsave(&ni->size_lock, flags); 1789 ni->initialized_size = attr_size; 1790 i_size_write(vi, attr_size); 1791 write_unlock_irqrestore(&ni->size_lock, flags); 1792 } 1793 } 1794 /* Setup the fields specific to resident attributes. */ 1795 a->data.resident.value_length = cpu_to_le32(attr_size); 1796 a->data.resident.value_offset = cpu_to_le16(mp_ofs); 1797 a->data.resident.flags = old_res_attr_flags; 1798 memset(&a->data.resident.reserved, 0, 1799 sizeof(a->data.resident.reserved)); 1800 /* Copy the data from the page back to the attribute value. */ 1801 if (page) { 1802 kaddr = kmap_atomic(page, KM_USER0); 1803 memcpy((u8*)a + mp_ofs, kaddr, attr_size); 1804 kunmap_atomic(kaddr, KM_USER0); 1805 } 1806 /* Setup the allocated size in the ntfs inode in case it changed. */ 1807 write_lock_irqsave(&ni->size_lock, flags); 1808 ni->allocated_size = arec_size - mp_ofs; 1809 write_unlock_irqrestore(&ni->size_lock, flags); 1810 /* Mark the mft record dirty, so it gets written back. */ 1811 flush_dcache_mft_record_page(ctx->ntfs_ino); 1812 mark_mft_record_dirty(ctx->ntfs_ino); 1813 err_out: 1814 if (ctx) 1815 ntfs_attr_put_search_ctx(ctx); 1816 if (m) 1817 unmap_mft_record(base_ni); 1818 ni->runlist.rl = NULL; 1819 up_write(&ni->runlist.lock); 1820 rl_err_out: 1821 if (rl) { 1822 if (ntfs_cluster_free_from_rl(vol, rl) < 0) { 1823 ntfs_error(vol->sb, "Failed to release allocated " 1824 "cluster(s) in error code path. Run " 1825 "chkdsk to recover the lost " 1826 "cluster(s)."); 1827 NVolSetErrors(vol); 1828 } 1829 ntfs_free(rl); 1830 page_err_out: 1831 unlock_page(page); 1832 page_cache_release(page); 1833 } 1834 if (err == -EINVAL) 1835 err = -EIO; 1836 return err; 1837 } 1838 1839 /** 1840 * ntfs_attr_extend_allocation - extend the allocated space of an attribute 1841 * @ni: ntfs inode of the attribute whose allocation to extend 1842 * @new_alloc_size: new size in bytes to which to extend the allocation to 1843 * @new_data_size: new size in bytes to which to extend the data to 1844 * @data_start: beginning of region which is required to be non-sparse 1845 * 1846 * Extend the allocated space of an attribute described by the ntfs inode @ni 1847 * to @new_alloc_size bytes. If @data_start is -1, the whole extension may be 1848 * implemented as a hole in the file (as long as both the volume and the ntfs 1849 * inode @ni have sparse support enabled). If @data_start is >= 0, then the 1850 * region between the old allocated size and @data_start - 1 may be made sparse 1851 * but the regions between @data_start and @new_alloc_size must be backed by 1852 * actual clusters. 1853 * 1854 * If @new_data_size is -1, it is ignored. If it is >= 0, then the data size 1855 * of the attribute is extended to @new_data_size. Note that the i_size of the 1856 * vfs inode is not updated. Only the data size in the base attribute record 1857 * is updated. The caller has to update i_size separately if this is required. 1858 * WARNING: It is a BUG() for @new_data_size to be smaller than the old data 1859 * size as well as for @new_data_size to be greater than @new_alloc_size. 1860 * 1861 * For resident attributes this involves resizing the attribute record and if 1862 * necessary moving it and/or other attributes into extent mft records and/or 1863 * converting the attribute to a non-resident attribute which in turn involves 1864 * extending the allocation of a non-resident attribute as described below. 1865 * 1866 * For non-resident attributes this involves allocating clusters in the data 1867 * zone on the volume (except for regions that are being made sparse) and 1868 * extending the run list to describe the allocated clusters as well as 1869 * updating the mapping pairs array of the attribute. This in turn involves 1870 * resizing the attribute record and if necessary moving it and/or other 1871 * attributes into extent mft records and/or splitting the attribute record 1872 * into multiple extent attribute records. 1873 * 1874 * Also, the attribute list attribute is updated if present and in some of the 1875 * above cases (the ones where extent mft records/attributes come into play), 1876 * an attribute list attribute is created if not already present. 1877 * 1878 * Return the new allocated size on success and -errno on error. In the case 1879 * that an error is encountered but a partial extension at least up to 1880 * @data_start (if present) is possible, the allocation is partially extended 1881 * and this is returned. This means the caller must check the returned size to 1882 * determine if the extension was partial. If @data_start is -1 then partial 1883 * allocations are not performed. 1884 * 1885 * WARNING: Do not call ntfs_attr_extend_allocation() for $MFT/$DATA. 1886 * 1887 * Locking: This function takes the runlist lock of @ni for writing as well as 1888 * locking the mft record of the base ntfs inode. These locks are maintained 1889 * throughout execution of the function. These locks are required so that the 1890 * attribute can be resized safely and so that it can for example be converted 1891 * from resident to non-resident safely. 1892 * 1893 * TODO: At present attribute list attribute handling is not implemented. 1894 * 1895 * TODO: At present it is not safe to call this function for anything other 1896 * than the $DATA attribute(s) of an uncompressed and unencrypted file. 1897 */ 1898 s64 ntfs_attr_extend_allocation(ntfs_inode *ni, s64 new_alloc_size, 1899 const s64 new_data_size, const s64 data_start) 1900 { 1901 VCN vcn; 1902 s64 ll, allocated_size, start = data_start; 1903 struct inode *vi = VFS_I(ni); 1904 ntfs_volume *vol = ni->vol; 1905 ntfs_inode *base_ni; 1906 MFT_RECORD *m; 1907 ATTR_RECORD *a; 1908 ntfs_attr_search_ctx *ctx; 1909 runlist_element *rl, *rl2; 1910 unsigned long flags; 1911 int err, mp_size; 1912 u32 attr_len = 0; /* Silence stupid gcc warning. */ 1913 BOOL mp_rebuilt; 1914 1915 #ifdef NTFS_DEBUG 1916 read_lock_irqsave(&ni->size_lock, flags); 1917 allocated_size = ni->allocated_size; 1918 read_unlock_irqrestore(&ni->size_lock, flags); 1919 ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, " 1920 "old_allocated_size 0x%llx, " 1921 "new_allocated_size 0x%llx, new_data_size 0x%llx, " 1922 "data_start 0x%llx.", vi->i_ino, 1923 (unsigned)le32_to_cpu(ni->type), 1924 (unsigned long long)allocated_size, 1925 (unsigned long long)new_alloc_size, 1926 (unsigned long long)new_data_size, 1927 (unsigned long long)start); 1928 #endif 1929 retry_extend: 1930 /* 1931 * For non-resident attributes, @start and @new_size need to be aligned 1932 * to cluster boundaries for allocation purposes. 1933 */ 1934 if (NInoNonResident(ni)) { 1935 if (start > 0) 1936 start &= ~(s64)vol->cluster_size_mask; 1937 new_alloc_size = (new_alloc_size + vol->cluster_size - 1) & 1938 ~(s64)vol->cluster_size_mask; 1939 } 1940 BUG_ON(new_data_size >= 0 && new_data_size > new_alloc_size); 1941 /* Check if new size is allowed in $AttrDef. */ 1942 err = ntfs_attr_size_bounds_check(vol, ni->type, new_alloc_size); 1943 if (unlikely(err)) { 1944 /* Only emit errors when the write will fail completely. */ 1945 read_lock_irqsave(&ni->size_lock, flags); 1946 allocated_size = ni->allocated_size; 1947 read_unlock_irqrestore(&ni->size_lock, flags); 1948 if (start < 0 || start >= allocated_size) { 1949 if (err == -ERANGE) { 1950 ntfs_error(vol->sb, "Cannot extend allocation " 1951 "of inode 0x%lx, attribute " 1952 "type 0x%x, because the new " 1953 "allocation would exceed the " 1954 "maximum allowed size for " 1955 "this attribute type.", 1956 vi->i_ino, (unsigned) 1957 le32_to_cpu(ni->type)); 1958 } else { 1959 ntfs_error(vol->sb, "Cannot extend allocation " 1960 "of inode 0x%lx, attribute " 1961 "type 0x%x, because this " 1962 "attribute type is not " 1963 "defined on the NTFS volume. " 1964 "Possible corruption! You " 1965 "should run chkdsk!", 1966 vi->i_ino, (unsigned) 1967 le32_to_cpu(ni->type)); 1968 } 1969 } 1970 /* Translate error code to be POSIX conformant for write(2). */ 1971 if (err == -ERANGE) 1972 err = -EFBIG; 1973 else 1974 err = -EIO; 1975 return err; 1976 } 1977 if (!NInoAttr(ni)) 1978 base_ni = ni; 1979 else 1980 base_ni = ni->ext.base_ntfs_ino; 1981 /* 1982 * We will be modifying both the runlist (if non-resident) and the mft 1983 * record so lock them both down. 1984 */ 1985 down_write(&ni->runlist.lock); 1986 m = map_mft_record(base_ni); 1987 if (IS_ERR(m)) { 1988 err = PTR_ERR(m); 1989 m = NULL; 1990 ctx = NULL; 1991 goto err_out; 1992 } 1993 ctx = ntfs_attr_get_search_ctx(base_ni, m); 1994 if (unlikely(!ctx)) { 1995 err = -ENOMEM; 1996 goto err_out; 1997 } 1998 read_lock_irqsave(&ni->size_lock, flags); 1999 allocated_size = ni->allocated_size; 2000 read_unlock_irqrestore(&ni->size_lock, flags); 2001 /* 2002 * If non-resident, seek to the last extent. If resident, there is 2003 * only one extent, so seek to that. 2004 */ 2005 vcn = NInoNonResident(ni) ? allocated_size >> vol->cluster_size_bits : 2006 0; 2007 /* 2008 * Abort if someone did the work whilst we waited for the locks. If we 2009 * just converted the attribute from resident to non-resident it is 2010 * likely that exactly this has happened already. We cannot quite 2011 * abort if we need to update the data size. 2012 */ 2013 if (unlikely(new_alloc_size <= allocated_size)) { 2014 ntfs_debug("Allocated size already exceeds requested size."); 2015 new_alloc_size = allocated_size; 2016 if (new_data_size < 0) 2017 goto done; 2018 /* 2019 * We want the first attribute extent so that we can update the 2020 * data size. 2021 */ 2022 vcn = 0; 2023 } 2024 err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, 2025 CASE_SENSITIVE, vcn, NULL, 0, ctx); 2026 if (unlikely(err)) { 2027 if (err == -ENOENT) 2028 err = -EIO; 2029 goto err_out; 2030 } 2031 m = ctx->mrec; 2032 a = ctx->attr; 2033 /* Use goto to reduce indentation. */ 2034 if (a->non_resident) 2035 goto do_non_resident_extend; 2036 BUG_ON(NInoNonResident(ni)); 2037 /* The total length of the attribute value. */ 2038 attr_len = le32_to_cpu(a->data.resident.value_length); 2039 /* 2040 * Extend the attribute record to be able to store the new attribute 2041 * size. ntfs_attr_record_resize() will not do anything if the size is 2042 * not changing. 2043 */ 2044 if (new_alloc_size < vol->mft_record_size && 2045 !ntfs_attr_record_resize(m, a, 2046 le16_to_cpu(a->data.resident.value_offset) + 2047 new_alloc_size)) { 2048 /* The resize succeeded! */ 2049 write_lock_irqsave(&ni->size_lock, flags); 2050 ni->allocated_size = le32_to_cpu(a->length) - 2051 le16_to_cpu(a->data.resident.value_offset); 2052 write_unlock_irqrestore(&ni->size_lock, flags); 2053 if (new_data_size >= 0) { 2054 BUG_ON(new_data_size < attr_len); 2055 a->data.resident.value_length = 2056 cpu_to_le32((u32)new_data_size); 2057 } 2058 goto flush_done; 2059 } 2060 /* 2061 * We have to drop all the locks so we can call 2062 * ntfs_attr_make_non_resident(). This could be optimised by try- 2063 * locking the first page cache page and only if that fails dropping 2064 * the locks, locking the page, and redoing all the locking and 2065 * lookups. While this would be a huge optimisation, it is not worth 2066 * it as this is definitely a slow code path. 2067 */ 2068 ntfs_attr_put_search_ctx(ctx); 2069 unmap_mft_record(base_ni); 2070 up_write(&ni->runlist.lock); 2071 /* 2072 * Not enough space in the mft record, try to make the attribute 2073 * non-resident and if successful restart the extension process. 2074 */ 2075 err = ntfs_attr_make_non_resident(ni, attr_len); 2076 if (likely(!err)) 2077 goto retry_extend; 2078 /* 2079 * Could not make non-resident. If this is due to this not being 2080 * permitted for this attribute type or there not being enough space, 2081 * try to make other attributes non-resident. Otherwise fail. 2082 */ 2083 if (unlikely(err != -EPERM && err != -ENOSPC)) { 2084 /* Only emit errors when the write will fail completely. */ 2085 read_lock_irqsave(&ni->size_lock, flags); 2086 allocated_size = ni->allocated_size; 2087 read_unlock_irqrestore(&ni->size_lock, flags); 2088 if (start < 0 || start >= allocated_size) 2089 ntfs_error(vol->sb, "Cannot extend allocation of " 2090 "inode 0x%lx, attribute type 0x%x, " 2091 "because the conversion from resident " 2092 "to non-resident attribute failed " 2093 "with error code %i.", vi->i_ino, 2094 (unsigned)le32_to_cpu(ni->type), err); 2095 if (err != -ENOMEM) 2096 err = -EIO; 2097 goto conv_err_out; 2098 } 2099 /* TODO: Not implemented from here, abort. */ 2100 read_lock_irqsave(&ni->size_lock, flags); 2101 allocated_size = ni->allocated_size; 2102 read_unlock_irqrestore(&ni->size_lock, flags); 2103 if (start < 0 || start >= allocated_size) { 2104 if (err == -ENOSPC) 2105 ntfs_error(vol->sb, "Not enough space in the mft " 2106 "record/on disk for the non-resident " 2107 "attribute value. This case is not " 2108 "implemented yet."); 2109 else /* if (err == -EPERM) */ 2110 ntfs_error(vol->sb, "This attribute type may not be " 2111 "non-resident. This case is not " 2112 "implemented yet."); 2113 } 2114 err = -EOPNOTSUPP; 2115 goto conv_err_out; 2116 #if 0 2117 // TODO: Attempt to make other attributes non-resident. 2118 if (!err) 2119 goto do_resident_extend; 2120 /* 2121 * Both the attribute list attribute and the standard information 2122 * attribute must remain in the base inode. Thus, if this is one of 2123 * these attributes, we have to try to move other attributes out into 2124 * extent mft records instead. 2125 */ 2126 if (ni->type == AT_ATTRIBUTE_LIST || 2127 ni->type == AT_STANDARD_INFORMATION) { 2128 // TODO: Attempt to move other attributes into extent mft 2129 // records. 2130 err = -EOPNOTSUPP; 2131 if (!err) 2132 goto do_resident_extend; 2133 goto err_out; 2134 } 2135 // TODO: Attempt to move this attribute to an extent mft record, but 2136 // only if it is not already the only attribute in an mft record in 2137 // which case there would be nothing to gain. 2138 err = -EOPNOTSUPP; 2139 if (!err) 2140 goto do_resident_extend; 2141 /* There is nothing we can do to make enough space. )-: */ 2142 goto err_out; 2143 #endif 2144 do_non_resident_extend: 2145 BUG_ON(!NInoNonResident(ni)); 2146 if (new_alloc_size == allocated_size) { 2147 BUG_ON(vcn); 2148 goto alloc_done; 2149 } 2150 /* 2151 * If the data starts after the end of the old allocation, this is a 2152 * $DATA attribute and sparse attributes are enabled on the volume and 2153 * for this inode, then create a sparse region between the old 2154 * allocated size and the start of the data. Otherwise simply proceed 2155 * with filling the whole space between the old allocated size and the 2156 * new allocated size with clusters. 2157 */ 2158 if ((start >= 0 && start <= allocated_size) || ni->type != AT_DATA || 2159 !NVolSparseEnabled(vol) || NInoSparseDisabled(ni)) 2160 goto skip_sparse; 2161 // TODO: This is not implemented yet. We just fill in with real 2162 // clusters for now... 2163 ntfs_debug("Inserting holes is not-implemented yet. Falling back to " 2164 "allocating real clusters instead."); 2165 skip_sparse: 2166 rl = ni->runlist.rl; 2167 if (likely(rl)) { 2168 /* Seek to the end of the runlist. */ 2169 while (rl->length) 2170 rl++; 2171 } 2172 /* If this attribute extent is not mapped, map it now. */ 2173 if (unlikely(!rl || rl->lcn == LCN_RL_NOT_MAPPED || 2174 (rl->lcn == LCN_ENOENT && rl > ni->runlist.rl && 2175 (rl-1)->lcn == LCN_RL_NOT_MAPPED))) { 2176 if (!rl && !allocated_size) 2177 goto first_alloc; 2178 rl = ntfs_mapping_pairs_decompress(vol, a, ni->runlist.rl); 2179 if (IS_ERR(rl)) { 2180 err = PTR_ERR(rl); 2181 if (start < 0 || start >= allocated_size) 2182 ntfs_error(vol->sb, "Cannot extend allocation " 2183 "of inode 0x%lx, attribute " 2184 "type 0x%x, because the " 2185 "mapping of a runlist " 2186 "fragment failed with error " 2187 "code %i.", vi->i_ino, 2188 (unsigned)le32_to_cpu(ni->type), 2189 err); 2190 if (err != -ENOMEM) 2191 err = -EIO; 2192 goto err_out; 2193 } 2194 ni->runlist.rl = rl; 2195 /* Seek to the end of the runlist. */ 2196 while (rl->length) 2197 rl++; 2198 } 2199 /* 2200 * We now know the runlist of the last extent is mapped and @rl is at 2201 * the end of the runlist. We want to begin allocating clusters 2202 * starting at the last allocated cluster to reduce fragmentation. If 2203 * there are no valid LCNs in the attribute we let the cluster 2204 * allocator choose the starting cluster. 2205 */ 2206 /* If the last LCN is a hole or simillar seek back to last real LCN. */ 2207 while (rl->lcn < 0 && rl > ni->runlist.rl) 2208 rl--; 2209 first_alloc: 2210 // FIXME: Need to implement partial allocations so at least part of the 2211 // write can be performed when start >= 0. (Needed for POSIX write(2) 2212 // conformance.) 2213 rl2 = ntfs_cluster_alloc(vol, allocated_size >> vol->cluster_size_bits, 2214 (new_alloc_size - allocated_size) >> 2215 vol->cluster_size_bits, (rl && (rl->lcn >= 0)) ? 2216 rl->lcn + rl->length : -1, DATA_ZONE, TRUE); 2217 if (IS_ERR(rl2)) { 2218 err = PTR_ERR(rl2); 2219 if (start < 0 || start >= allocated_size) 2220 ntfs_error(vol->sb, "Cannot extend allocation of " 2221 "inode 0x%lx, attribute type 0x%x, " 2222 "because the allocation of clusters " 2223 "failed with error code %i.", vi->i_ino, 2224 (unsigned)le32_to_cpu(ni->type), err); 2225 if (err != -ENOMEM && err != -ENOSPC) 2226 err = -EIO; 2227 goto err_out; 2228 } 2229 rl = ntfs_runlists_merge(ni->runlist.rl, rl2); 2230 if (IS_ERR(rl)) { 2231 err = PTR_ERR(rl); 2232 if (start < 0 || start >= allocated_size) 2233 ntfs_error(vol->sb, "Cannot extend allocation of " 2234 "inode 0x%lx, attribute type 0x%x, " 2235 "because the runlist merge failed " 2236 "with error code %i.", vi->i_ino, 2237 (unsigned)le32_to_cpu(ni->type), err); 2238 if (err != -ENOMEM) 2239 err = -EIO; 2240 if (ntfs_cluster_free_from_rl(vol, rl2)) { 2241 ntfs_error(vol->sb, "Failed to release allocated " 2242 "cluster(s) in error code path. Run " 2243 "chkdsk to recover the lost " 2244 "cluster(s)."); 2245 NVolSetErrors(vol); 2246 } 2247 ntfs_free(rl2); 2248 goto err_out; 2249 } 2250 ni->runlist.rl = rl; 2251 ntfs_debug("Allocated 0x%llx clusters.", (long long)(new_alloc_size - 2252 allocated_size) >> vol->cluster_size_bits); 2253 /* Find the runlist element with which the attribute extent starts. */ 2254 ll = sle64_to_cpu(a->data.non_resident.lowest_vcn); 2255 rl2 = ntfs_rl_find_vcn_nolock(rl, ll); 2256 BUG_ON(!rl2); 2257 BUG_ON(!rl2->length); 2258 BUG_ON(rl2->lcn < LCN_HOLE); 2259 mp_rebuilt = FALSE; 2260 /* Get the size for the new mapping pairs array for this extent. */ 2261 mp_size = ntfs_get_size_for_mapping_pairs(vol, rl2, ll, -1); 2262 if (unlikely(mp_size <= 0)) { 2263 err = mp_size; 2264 if (start < 0 || start >= allocated_size) 2265 ntfs_error(vol->sb, "Cannot extend allocation of " 2266 "inode 0x%lx, attribute type 0x%x, " 2267 "because determining the size for the " 2268 "mapping pairs failed with error code " 2269 "%i.", vi->i_ino, 2270 (unsigned)le32_to_cpu(ni->type), err); 2271 err = -EIO; 2272 goto undo_alloc; 2273 } 2274 /* Extend the attribute record to fit the bigger mapping pairs array. */ 2275 attr_len = le32_to_cpu(a->length); 2276 err = ntfs_attr_record_resize(m, a, mp_size + 2277 le16_to_cpu(a->data.non_resident.mapping_pairs_offset)); 2278 if (unlikely(err)) { 2279 BUG_ON(err != -ENOSPC); 2280 // TODO: Deal with this by moving this extent to a new mft 2281 // record or by starting a new extent in a new mft record, 2282 // possibly by extending this extent partially and filling it 2283 // and creating a new extent for the remainder, or by making 2284 // other attributes non-resident and/or by moving other 2285 // attributes out of this mft record. 2286 if (start < 0 || start >= allocated_size) 2287 ntfs_error(vol->sb, "Not enough space in the mft " 2288 "record for the extended attribute " 2289 "record. This case is not " 2290 "implemented yet."); 2291 err = -EOPNOTSUPP; 2292 goto undo_alloc; 2293 } 2294 mp_rebuilt = TRUE; 2295 /* Generate the mapping pairs array directly into the attr record. */ 2296 err = ntfs_mapping_pairs_build(vol, (u8*)a + 2297 le16_to_cpu(a->data.non_resident.mapping_pairs_offset), 2298 mp_size, rl2, ll, -1, NULL); 2299 if (unlikely(err)) { 2300 if (start < 0 || start >= allocated_size) 2301 ntfs_error(vol->sb, "Cannot extend allocation of " 2302 "inode 0x%lx, attribute type 0x%x, " 2303 "because building the mapping pairs " 2304 "failed with error code %i.", vi->i_ino, 2305 (unsigned)le32_to_cpu(ni->type), err); 2306 err = -EIO; 2307 goto undo_alloc; 2308 } 2309 /* Update the highest_vcn. */ 2310 a->data.non_resident.highest_vcn = cpu_to_sle64((new_alloc_size >> 2311 vol->cluster_size_bits) - 1); 2312 /* 2313 * We now have extended the allocated size of the attribute. Reflect 2314 * this in the ntfs_inode structure and the attribute record. 2315 */ 2316 if (a->data.non_resident.lowest_vcn) { 2317 /* 2318 * We are not in the first attribute extent, switch to it, but 2319 * first ensure the changes will make it to disk later. 2320 */ 2321 flush_dcache_mft_record_page(ctx->ntfs_ino); 2322 mark_mft_record_dirty(ctx->ntfs_ino); 2323 ntfs_attr_reinit_search_ctx(ctx); 2324 err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, 2325 CASE_SENSITIVE, 0, NULL, 0, ctx); 2326 if (unlikely(err)) 2327 goto restore_undo_alloc; 2328 /* @m is not used any more so no need to set it. */ 2329 a = ctx->attr; 2330 } 2331 write_lock_irqsave(&ni->size_lock, flags); 2332 ni->allocated_size = new_alloc_size; 2333 a->data.non_resident.allocated_size = cpu_to_sle64(new_alloc_size); 2334 /* 2335 * FIXME: This would fail if @ni is a directory, $MFT, or an index, 2336 * since those can have sparse/compressed set. For example can be 2337 * set compressed even though it is not compressed itself and in that 2338 * case the bit means that files are to be created compressed in the 2339 * directory... At present this is ok as this code is only called for 2340 * regular files, and only for their $DATA attribute(s). 2341 * FIXME: The calculation is wrong if we created a hole above. For now 2342 * it does not matter as we never create holes. 2343 */ 2344 if (NInoSparse(ni) || NInoCompressed(ni)) { 2345 ni->itype.compressed.size += new_alloc_size - allocated_size; 2346 a->data.non_resident.compressed_size = 2347 cpu_to_sle64(ni->itype.compressed.size); 2348 vi->i_blocks = ni->itype.compressed.size >> 9; 2349 } else 2350 vi->i_blocks = new_alloc_size >> 9; 2351 write_unlock_irqrestore(&ni->size_lock, flags); 2352 alloc_done: 2353 if (new_data_size >= 0) { 2354 BUG_ON(new_data_size < 2355 sle64_to_cpu(a->data.non_resident.data_size)); 2356 a->data.non_resident.data_size = cpu_to_sle64(new_data_size); 2357 } 2358 flush_done: 2359 /* Ensure the changes make it to disk. */ 2360 flush_dcache_mft_record_page(ctx->ntfs_ino); 2361 mark_mft_record_dirty(ctx->ntfs_ino); 2362 done: 2363 ntfs_attr_put_search_ctx(ctx); 2364 unmap_mft_record(base_ni); 2365 up_write(&ni->runlist.lock); 2366 ntfs_debug("Done, new_allocated_size 0x%llx.", 2367 (unsigned long long)new_alloc_size); 2368 return new_alloc_size; 2369 restore_undo_alloc: 2370 if (start < 0 || start >= allocated_size) 2371 ntfs_error(vol->sb, "Cannot complete extension of allocation " 2372 "of inode 0x%lx, attribute type 0x%x, because " 2373 "lookup of first attribute extent failed with " 2374 "error code %i.", vi->i_ino, 2375 (unsigned)le32_to_cpu(ni->type), err); 2376 if (err == -ENOENT) 2377 err = -EIO; 2378 ntfs_attr_reinit_search_ctx(ctx); 2379 if (ntfs_attr_lookup(ni->type, ni->name, ni->name_len, CASE_SENSITIVE, 2380 allocated_size >> vol->cluster_size_bits, NULL, 0, 2381 ctx)) { 2382 ntfs_error(vol->sb, "Failed to find last attribute extent of " 2383 "attribute in error code path. Run chkdsk to " 2384 "recover."); 2385 write_lock_irqsave(&ni->size_lock, flags); 2386 ni->allocated_size = new_alloc_size; 2387 /* 2388 * FIXME: This would fail if @ni is a directory... See above. 2389 * FIXME: The calculation is wrong if we created a hole above. 2390 * For now it does not matter as we never create holes. 2391 */ 2392 if (NInoSparse(ni) || NInoCompressed(ni)) { 2393 ni->itype.compressed.size += new_alloc_size - 2394 allocated_size; 2395 vi->i_blocks = ni->itype.compressed.size >> 9; 2396 } else 2397 vi->i_blocks = new_alloc_size >> 9; 2398 write_unlock_irqrestore(&ni->size_lock, flags); 2399 ntfs_attr_put_search_ctx(ctx); 2400 unmap_mft_record(base_ni); 2401 up_write(&ni->runlist.lock); 2402 /* 2403 * The only thing that is now wrong is the allocated size of the 2404 * base attribute extent which chkdsk should be able to fix. 2405 */ 2406 NVolSetErrors(vol); 2407 return err; 2408 } 2409 ctx->attr->data.non_resident.highest_vcn = cpu_to_sle64( 2410 (allocated_size >> vol->cluster_size_bits) - 1); 2411 undo_alloc: 2412 ll = allocated_size >> vol->cluster_size_bits; 2413 if (ntfs_cluster_free(ni, ll, -1, ctx) < 0) { 2414 ntfs_error(vol->sb, "Failed to release allocated cluster(s) " 2415 "in error code path. Run chkdsk to recover " 2416 "the lost cluster(s)."); 2417 NVolSetErrors(vol); 2418 } 2419 m = ctx->mrec; 2420 a = ctx->attr; 2421 /* 2422 * If the runlist truncation fails and/or the search context is no 2423 * longer valid, we cannot resize the attribute record or build the 2424 * mapping pairs array thus we mark the inode bad so that no access to 2425 * the freed clusters can happen. 2426 */ 2427 if (ntfs_rl_truncate_nolock(vol, &ni->runlist, ll) || IS_ERR(m)) { 2428 ntfs_error(vol->sb, "Failed to %s in error code path. Run " 2429 "chkdsk to recover.", IS_ERR(m) ? 2430 "restore attribute search context" : 2431 "truncate attribute runlist"); 2432 make_bad_inode(vi); 2433 make_bad_inode(VFS_I(base_ni)); 2434 NVolSetErrors(vol); 2435 } else if (mp_rebuilt) { 2436 if (ntfs_attr_record_resize(m, a, attr_len)) { 2437 ntfs_error(vol->sb, "Failed to restore attribute " 2438 "record in error code path. Run " 2439 "chkdsk to recover."); 2440 make_bad_inode(vi); 2441 make_bad_inode(VFS_I(base_ni)); 2442 NVolSetErrors(vol); 2443 } else /* if (success) */ { 2444 if (ntfs_mapping_pairs_build(vol, (u8*)a + le16_to_cpu( 2445 a->data.non_resident. 2446 mapping_pairs_offset), attr_len - 2447 le16_to_cpu(a->data.non_resident. 2448 mapping_pairs_offset), rl2, ll, -1, 2449 NULL)) { 2450 ntfs_error(vol->sb, "Failed to restore " 2451 "mapping pairs array in error " 2452 "code path. Run chkdsk to " 2453 "recover."); 2454 make_bad_inode(vi); 2455 make_bad_inode(VFS_I(base_ni)); 2456 NVolSetErrors(vol); 2457 } 2458 flush_dcache_mft_record_page(ctx->ntfs_ino); 2459 mark_mft_record_dirty(ctx->ntfs_ino); 2460 } 2461 } 2462 err_out: 2463 if (ctx) 2464 ntfs_attr_put_search_ctx(ctx); 2465 if (m) 2466 unmap_mft_record(base_ni); 2467 up_write(&ni->runlist.lock); 2468 conv_err_out: 2469 ntfs_debug("Failed. Returning error code %i.", err); 2470 return err; 2471 } 2472 2473 /** 2474 * ntfs_attr_set - fill (a part of) an attribute with a byte 2475 * @ni: ntfs inode describing the attribute to fill 2476 * @ofs: offset inside the attribute at which to start to fill 2477 * @cnt: number of bytes to fill 2478 * @val: the unsigned 8-bit value with which to fill the attribute 2479 * 2480 * Fill @cnt bytes of the attribute described by the ntfs inode @ni starting at 2481 * byte offset @ofs inside the attribute with the constant byte @val. 2482 * 2483 * This function is effectively like memset() applied to an ntfs attribute. 2484 * Note thie function actually only operates on the page cache pages belonging 2485 * to the ntfs attribute and it marks them dirty after doing the memset(). 2486 * Thus it relies on the vm dirty page write code paths to cause the modified 2487 * pages to be written to the mft record/disk. 2488 * 2489 * Return 0 on success and -errno on error. An error code of -ESPIPE means 2490 * that @ofs + @cnt were outside the end of the attribute and no write was 2491 * performed. 2492 */ 2493 int ntfs_attr_set(ntfs_inode *ni, const s64 ofs, const s64 cnt, const u8 val) 2494 { 2495 ntfs_volume *vol = ni->vol; 2496 struct address_space *mapping; 2497 struct page *page; 2498 u8 *kaddr; 2499 pgoff_t idx, end; 2500 unsigned int start_ofs, end_ofs, size; 2501 2502 ntfs_debug("Entering for ofs 0x%llx, cnt 0x%llx, val 0x%hx.", 2503 (long long)ofs, (long long)cnt, val); 2504 BUG_ON(ofs < 0); 2505 BUG_ON(cnt < 0); 2506 if (!cnt) 2507 goto done; 2508 /* 2509 * FIXME: Compressed and encrypted attributes are not supported when 2510 * writing and we should never have gotten here for them. 2511 */ 2512 BUG_ON(NInoCompressed(ni)); 2513 BUG_ON(NInoEncrypted(ni)); 2514 mapping = VFS_I(ni)->i_mapping; 2515 /* Work out the starting index and page offset. */ 2516 idx = ofs >> PAGE_CACHE_SHIFT; 2517 start_ofs = ofs & ~PAGE_CACHE_MASK; 2518 /* Work out the ending index and page offset. */ 2519 end = ofs + cnt; 2520 end_ofs = end & ~PAGE_CACHE_MASK; 2521 /* If the end is outside the inode size return -ESPIPE. */ 2522 if (unlikely(end > i_size_read(VFS_I(ni)))) { 2523 ntfs_error(vol->sb, "Request exceeds end of attribute."); 2524 return -ESPIPE; 2525 } 2526 end >>= PAGE_CACHE_SHIFT; 2527 /* If there is a first partial page, need to do it the slow way. */ 2528 if (start_ofs) { 2529 page = read_cache_page(mapping, idx, 2530 (filler_t*)mapping->a_ops->readpage, NULL); 2531 if (IS_ERR(page)) { 2532 ntfs_error(vol->sb, "Failed to read first partial " 2533 "page (sync error, index 0x%lx).", idx); 2534 return PTR_ERR(page); 2535 } 2536 wait_on_page_locked(page); 2537 if (unlikely(!PageUptodate(page))) { 2538 ntfs_error(vol->sb, "Failed to read first partial page " 2539 "(async error, index 0x%lx).", idx); 2540 page_cache_release(page); 2541 return PTR_ERR(page); 2542 } 2543 /* 2544 * If the last page is the same as the first page, need to 2545 * limit the write to the end offset. 2546 */ 2547 size = PAGE_CACHE_SIZE; 2548 if (idx == end) 2549 size = end_ofs; 2550 kaddr = kmap_atomic(page, KM_USER0); 2551 memset(kaddr + start_ofs, val, size - start_ofs); 2552 flush_dcache_page(page); 2553 kunmap_atomic(kaddr, KM_USER0); 2554 set_page_dirty(page); 2555 page_cache_release(page); 2556 if (idx == end) 2557 goto done; 2558 idx++; 2559 } 2560 /* Do the whole pages the fast way. */ 2561 for (; idx < end; idx++) { 2562 /* Find or create the current page. (The page is locked.) */ 2563 page = grab_cache_page(mapping, idx); 2564 if (unlikely(!page)) { 2565 ntfs_error(vol->sb, "Insufficient memory to grab " 2566 "page (index 0x%lx).", idx); 2567 return -ENOMEM; 2568 } 2569 kaddr = kmap_atomic(page, KM_USER0); 2570 memset(kaddr, val, PAGE_CACHE_SIZE); 2571 flush_dcache_page(page); 2572 kunmap_atomic(kaddr, KM_USER0); 2573 /* 2574 * If the page has buffers, mark them uptodate since buffer 2575 * state and not page state is definitive in 2.6 kernels. 2576 */ 2577 if (page_has_buffers(page)) { 2578 struct buffer_head *bh, *head; 2579 2580 bh = head = page_buffers(page); 2581 do { 2582 set_buffer_uptodate(bh); 2583 } while ((bh = bh->b_this_page) != head); 2584 } 2585 /* Now that buffers are uptodate, set the page uptodate, too. */ 2586 SetPageUptodate(page); 2587 /* 2588 * Set the page and all its buffers dirty and mark the inode 2589 * dirty, too. The VM will write the page later on. 2590 */ 2591 set_page_dirty(page); 2592 /* Finally unlock and release the page. */ 2593 unlock_page(page); 2594 page_cache_release(page); 2595 balance_dirty_pages_ratelimited(mapping); 2596 cond_resched(); 2597 } 2598 /* If there is a last partial page, need to do it the slow way. */ 2599 if (end_ofs) { 2600 page = read_cache_page(mapping, idx, 2601 (filler_t*)mapping->a_ops->readpage, NULL); 2602 if (IS_ERR(page)) { 2603 ntfs_error(vol->sb, "Failed to read last partial page " 2604 "(sync error, index 0x%lx).", idx); 2605 return PTR_ERR(page); 2606 } 2607 wait_on_page_locked(page); 2608 if (unlikely(!PageUptodate(page))) { 2609 ntfs_error(vol->sb, "Failed to read last partial page " 2610 "(async error, index 0x%lx).", idx); 2611 page_cache_release(page); 2612 return PTR_ERR(page); 2613 } 2614 kaddr = kmap_atomic(page, KM_USER0); 2615 memset(kaddr, val, end_ofs); 2616 flush_dcache_page(page); 2617 kunmap_atomic(kaddr, KM_USER0); 2618 set_page_dirty(page); 2619 page_cache_release(page); 2620 } 2621 done: 2622 ntfs_debug("Done."); 2623 return 0; 2624 } 2625 2626 #endif /* NTFS_RW */ 2627