X-Git-Url: https://git.whamcloud.com/?a=blobdiff_plain;f=e2fsck%2Fpass2.c;h=cb1e587566f6cdeda3a5a01e027a7cb497c87a7b;hb=refs%2Ftags%2Fv1.45.6;hp=4774f0653df0aea1c9a710d4de5d777b3fd62299;hpb=82ad476d4709faf9f39a9aa581f9679e8181bbea;p=tools%2Fe2fsprogs.git diff --git a/e2fsck/pass2.c b/e2fsck/pass2.c index 4774f06..cb1e587 100644 --- a/e2fsck/pass2.c +++ b/e2fsck/pass2.c @@ -47,7 +47,7 @@ #include "e2fsck.h" #include "problem.h" -#include "dict.h" +#include "support/dict.h" #ifdef NO_INLINE_FUNCS #define _INLINE_ @@ -61,6 +61,9 @@ * Keeps track of how many times an inode is referenced. */ static void deallocate_inode(e2fsck_t ctx, ext2_ino_t ino, char* block_buf); +static int check_dir_block2(ext2_filsys fs, + struct ext2_db_entry2 *dir_blocks_info, + void *priv_data); static int check_dir_block(ext2_filsys fs, struct ext2_db_entry2 *dir_blocks_info, void *priv_data); @@ -68,8 +71,8 @@ static int allocate_dir_block(e2fsck_t ctx, struct ext2_db_entry2 *dir_blocks_info, char *buf, struct problem_context *pctx); static void clear_htree(e2fsck_t ctx, ext2_ino_t ino); -static int htree_depth(struct dx_dir_info *dx_dir, - struct dx_dirblock_info *dx_db); +static short htree_depth(struct dx_dir_info *dx_dir, + struct dx_dirblock_info *dx_db); static EXT2_QSORT_TYPE special_dir_block_cmp(const void *a, const void *b); struct check_dir_struct { @@ -77,25 +80,64 @@ struct check_dir_struct { struct problem_context pctx; int count, max; e2fsck_t ctx; + unsigned long long list_offset; + unsigned long long ra_entries; + unsigned long long next_ra_off; }; +static void update_parents(struct dx_dir_info *dx_dir, int type) +{ + struct dx_dirblock_info *dx_db, *dx_parent, *dx_previous; + blk_t b; + + for (b = 0, dx_db = dx_dir->dx_block; + b < dx_dir->numblocks; + b++, dx_db++) { + dx_parent = &dx_dir->dx_block[dx_db->parent]; + if (dx_db->type != type) + continue; + + /* + * XXX Make sure dx_parent->min_hash > dx_db->min_hash + */ + if (dx_db->flags & DX_FLAG_FIRST) { + dx_parent->min_hash = dx_db->min_hash; + if (dx_parent->previous) { + dx_previous = + &dx_dir->dx_block[dx_parent->previous]; + dx_previous->node_max_hash = + dx_parent->min_hash; + } + } + /* + * XXX Make sure dx_parent->max_hash < dx_db->max_hash + */ + if (dx_db->flags & DX_FLAG_LAST) { + dx_parent->max_hash = dx_db->max_hash; + } + } +} + void e2fsck_pass2(e2fsck_t ctx) { struct ext2_super_block *sb = ctx->fs->super; struct problem_context pctx; ext2_filsys fs = ctx->fs; - char *buf; + char *buf = NULL; #ifdef RESOURCE_TRACK struct resource_track rtrack; #endif struct check_dir_struct cd; struct dx_dir_info *dx_dir; - struct dx_dirblock_info *dx_db, *dx_parent; - unsigned int save_type; - int b; - int i, depth; + struct dx_dirblock_info *dx_db; + blk_t b; + ext2_ino_t i; + short depth; problem_t code; int bad_dir; + int (*check_dir_func)(ext2_filsys fs, + struct ext2_db_entry2 *dir_blocks_info, + void *priv_data); init_resource_track(&rtrack, ctx->fs->io); clear_problem_context(&cd.pctx); @@ -107,23 +149,13 @@ void e2fsck_pass2(e2fsck_t ctx) if (!(ctx->options & E2F_OPT_PREEN)) fix_problem(ctx, PR_2_PASS_HEADER, &cd.pctx); - e2fsck_setup_tdb_icount(ctx, EXT2_ICOUNT_OPT_INCREMENT, - &ctx->inode_count); - if (ctx->inode_count) - cd.pctx.errcode = 0; - else { - e2fsck_set_bitmap_type(fs, EXT2FS_BMAP64_RBTREE, - "inode_count", &save_type); - cd.pctx.errcode = ext2fs_create_icount2(fs, - EXT2_ICOUNT_OPT_INCREMENT, - 0, ctx->inode_link_info, - &ctx->inode_count); - fs->default_bitmap_type = save_type; - } + cd.pctx.errcode = e2fsck_setup_icount(ctx, "inode_count", + EXT2_ICOUNT_OPT_INCREMENT, + ctx->inode_link_info, &ctx->inode_count); if (cd.pctx.errcode) { fix_problem(ctx, PR_2_ALLOCATE_ICOUNT, &cd.pctx); ctx->flags |= E2F_FLAG_ABORT; - return; + goto cleanup; } buf = (char *) e2fsck_allocate_memory(ctx, 2*fs->blocksize, "directory scan buffer"); @@ -139,34 +171,38 @@ void e2fsck_pass2(e2fsck_t ctx) cd.ctx = ctx; cd.count = 1; cd.max = ext2fs_dblist_count2(fs->dblist); + cd.list_offset = 0; + cd.ra_entries = ctx->readahead_kb * 1024 / ctx->fs->blocksize; + cd.next_ra_off = 0; if (ctx->progress) (void) (ctx->progress)(ctx, 2, 0, cd.max); - if (fs->super->s_feature_compat & EXT2_FEATURE_COMPAT_DIR_INDEX) + if (ext2fs_has_feature_dir_index(fs->super)) ext2fs_dblist_sort2(fs->dblist, special_dir_block_cmp); - cd.pctx.errcode = ext2fs_dblist_iterate2(fs->dblist, check_dir_block, + check_dir_func = cd.ra_entries ? check_dir_block2 : check_dir_block; + cd.pctx.errcode = ext2fs_dblist_iterate2(fs->dblist, check_dir_func, &cd); - if (ctx->flags & E2F_FLAG_SIGNAL_MASK || ctx->flags & E2F_FLAG_RESTART) - return; - if (ctx->flags & E2F_FLAG_RESTART_LATER) { ctx->flags |= E2F_FLAG_RESTART; - return; + ctx->flags &= ~E2F_FLAG_RESTART_LATER; } + if (ctx->flags & E2F_FLAG_RUN_RETURN) + goto cleanup; + if (cd.pctx.errcode) { fix_problem(ctx, PR_2_DBLIST_ITERATE, &cd.pctx); ctx->flags |= E2F_FLAG_ABORT; - return; + goto cleanup; } -#ifdef ENABLE_HTREE for (i=0; (dx_dir = e2fsck_dx_dir_info_iter(ctx, &i)) != 0;) { if (ctx->flags & E2F_FLAG_SIGNAL_MASK) - return; - if (dx_dir->numblocks == 0) + goto cleanup; + if (e2fsck_dir_will_be_rehashed(ctx, dx_dir->ino) || + dx_dir->numblocks == 0) continue; clear_problem_context(&pctx); bad_dir = 0; @@ -180,24 +216,11 @@ void e2fsck_pass2(e2fsck_t ctx) * Find all of the first and last leaf blocks, and * update their parent's min and max hash values */ - for (b=0, dx_db = dx_dir->dx_block; - b < dx_dir->numblocks; - b++, dx_db++) { - if ((dx_db->type != DX_DIRBLOCK_LEAF) || - !(dx_db->flags & (DX_FLAG_FIRST | DX_FLAG_LAST))) - continue; - dx_parent = &dx_dir->dx_block[dx_db->parent]; - /* - * XXX Make sure dx_parent->min_hash > dx_db->min_hash - */ - if (dx_db->flags & DX_FLAG_FIRST) - dx_parent->min_hash = dx_db->min_hash; - /* - * XXX Make sure dx_parent->max_hash < dx_db->max_hash - */ - if (dx_db->flags & DX_FLAG_LAST) - dx_parent->max_hash = dx_db->max_hash; - } + update_parents(dx_dir, DX_DIRBLOCK_LEAF); + + /* for 3 level htree: update 2 level parent's min + * and max hash values */ + update_parents(dx_dir, DX_DIRBLOCK_NODE); for (b=0, dx_db = dx_dir->dx_block; b < dx_dir->numblocks; @@ -250,7 +273,7 @@ void e2fsck_pass2(e2fsck_t ctx) } } e2fsck_free_dx_dir_info(ctx); -#endif + ext2fs_free_mem(&buf); ext2fs_free_dblist(fs->dblist); @@ -262,14 +285,16 @@ void e2fsck_pass2(e2fsck_t ctx) ext2fs_free_inode_bitmap(ctx->inode_reg_map); ctx->inode_reg_map = 0; } + if (ctx->encrypted_dirs) { + ext2fs_u32_list_free(ctx->encrypted_dirs); + ctx->encrypted_dirs = 0; + } clear_problem_context(&pctx); if (ctx->large_files) { - if (!(sb->s_feature_ro_compat & - EXT2_FEATURE_RO_COMPAT_LARGE_FILE) && + if (!ext2fs_has_feature_large_file(sb) && fix_problem(ctx, PR_2_FEATURE_LARGE_FILES, &pctx)) { - sb->s_feature_ro_compat |= - EXT2_FEATURE_RO_COMPAT_LARGE_FILE; + ext2fs_set_feature_large_file(sb); fs->flags &= ~EXT2_FLAG_MASTER_SB_ONLY; ext2fs_mark_super_dirty(fs); } @@ -281,13 +306,15 @@ void e2fsck_pass2(e2fsck_t ctx) } print_resource_track(ctx, _("Pass 2"), &rtrack, fs->io); +cleanup: + ext2fs_free_mem(&buf); } #define MAX_DEPTH 32000 -static int htree_depth(struct dx_dir_info *dx_dir, - struct dx_dirblock_info *dx_db) +static short htree_depth(struct dx_dir_info *dx_dir, + struct dx_dirblock_info *dx_db) { - int depth = 0; + short depth = 0; while (dx_db->type != DX_DIRBLOCK_ROOT && depth < MAX_DEPTH) { dx_db = &dx_dir->dx_block[dx_db->parent]; @@ -309,7 +336,7 @@ static int dict_de_cmp(const void *a, const void *b) if (a_len != b_len) return (a_len - b_len); - return strncmp(de_a->name, de_b->name, a_len); + return memcmp(de_a->name, de_b->name, a_len); } /* @@ -458,7 +485,6 @@ static int check_dotdot(e2fsck_t ctx, */ static int check_name(e2fsck_t ctx, struct ext2_dir_entry *dirent, - ext2_ino_t dir_ino EXT2FS_ATTR((unused)), struct problem_context *pctx) { int i; @@ -466,19 +492,32 @@ static int check_name(e2fsck_t ctx, int ret = 0; for ( i = 0; i < ext2fs_dirent_name_len(dirent); i++) { - if (dirent->name[i] == '/' || dirent->name[i] == '\0') { - if (fixup < 0) { - fixup = fix_problem(ctx, PR_2_BAD_NAME, pctx); - } - if (fixup) { - dirent->name[i] = '.'; - ret = 1; - } - } + if (dirent->name[i] != '/' && dirent->name[i] != '\0') + continue; + if (fixup < 0) + fixup = fix_problem(ctx, PR_2_BAD_NAME, pctx); + if (fixup == 0) + return 0; + dirent->name[i] = '.'; + ret = 1; } return ret; } +static int encrypted_check_name(e2fsck_t ctx, + struct ext2_dir_entry *dirent, + struct problem_context *pctx) +{ + if (ext2fs_dirent_name_len(dirent) < EXT4_CRYPTO_BLOCK_SIZE) { + if (fix_problem(ctx, PR_2_BAD_ENCRYPTED_NAME, pctx)) { + dirent->inode = 0; + return 1; + } + ext2fs_unmark_valid(ctx->fs); + } + return 0; +} + /* * Check the directory filetype (if present) */ @@ -491,8 +530,7 @@ static _INLINE_ int check_filetype(e2fsck_t ctx, int should_be = EXT2_FT_UNKNOWN; struct ext2_inode inode; - if (!(ctx->fs->super->s_feature_incompat & - EXT2_FEATURE_INCOMPAT_FILETYPE)) { + if (!ext2fs_has_feature_filetype(ctx->fs->super)) { if (filetype == 0 || !fix_problem(ctx, PR_2_CLEAR_FILETYPE, pctx)) return 0; @@ -526,15 +564,14 @@ static _INLINE_ int check_filetype(e2fsck_t ctx, return 1; } -#ifdef ENABLE_HTREE static void parse_int_node(ext2_filsys fs, struct ext2_db_entry2 *db, struct check_dir_struct *cd, struct dx_dir_info *dx_dir, char *block_buf, int failed_csum) { - struct ext2_dx_root_info *root; - struct ext2_dx_entry *ent; + struct ext2_dx_root_info *root; + struct ext2_dx_entry *ent; struct ext2_dx_countlimit *limit; struct dx_dirblock_info *dx_db; int i, expect_limit, count; @@ -550,10 +587,10 @@ static void parse_int_node(ext2_filsys fs, #ifdef DX_DEBUG printf("Root node dump:\n"); printf("\t Reserved zero: %u\n", root->reserved_zero); - printf("\t Hash Version: %d\n", root->hash_version); - printf("\t Info length: %d\n", root->info_length); - printf("\t Indirect levels: %d\n", root->indirect_levels); - printf("\t Flags: %d\n", root->unused_flags); + printf("\t Hash Version: %u\n", root->hash_version); + printf("\t Info length: %u\n", root->info_length); + printf("\t Indirect levels: %u\n", root->indirect_levels); + printf("\t Flags: %x\n", root->unused_flags); #endif ent = (struct ext2_dx_entry *) (block_buf + 24 + root->info_length); @@ -583,8 +620,7 @@ static void parse_int_node(ext2_filsys fs, #endif count = ext2fs_le16_to_cpu(limit->count); - if (EXT2_HAS_RO_COMPAT_FEATURE(fs->super, - EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) + if (ext2fs_has_feature_metadata_csum(fs->super)) csum_size = sizeof(struct ext2_dx_tail); expect_limit = (fs->blocksize - (csum_size + ((char *) ent - block_buf))) / @@ -608,9 +644,9 @@ static void parse_int_node(ext2_filsys fs, printf("Entry #%d: Hash 0x%08x, block %u\n", i, hash, ext2fs_le32_to_cpu(ent[i].block)); #endif - blk = ext2fs_le32_to_cpu(ent[i].block) & 0x0ffffff; + blk = ext2fs_le32_to_cpu(ent[i].block) & EXT4_DX_BLOCK_MASK; /* Check to make sure the block is valid */ - if (blk >= (blk_t) dx_dir->numblocks) { + if (blk >= dx_dir->numblocks) { cd->pctx.blk = blk; if (fix_problem(cd->ctx, PR_2_HTREE_BADBLK, &cd->pctx)) @@ -627,6 +663,11 @@ static void parse_int_node(ext2_filsys fs, dx_db->flags |= DX_FLAG_REFERENCED; dx_db->parent = db->blockcnt; } + + dx_db->previous = + i ? (ext2fs_le32_to_cpu(ent[i-1].block) & + EXT4_DX_BLOCK_MASK) : 0; + if (hash < min_hash) min_hash = hash; if (hash > max_hash) @@ -656,7 +697,6 @@ clear_and_exit: dx_dir->numblocks = 0; e2fsck_rehash_dir_later(cd->ctx, cd->pctx.ino); } -#endif /* ENABLE_HTREE */ /* * Given a busted directory, try to salvage it somehow. @@ -671,18 +711,29 @@ static void salvage_directory(ext2_filsys fs, char *cp = (char *) dirent; int left; unsigned int rec_len, prev_rec_len; - unsigned int name_len = ext2fs_dirent_name_len(dirent); + unsigned int name_len; - (void) ext2fs_get_rec_len(fs, dirent, &rec_len); + /* + * If the space left for the entry is too small to be an entry, + * we can't access dirent's fields, so plumb in the values needed + * so that the previous entry absorbs this one. + */ + if (block_len - *offset < EXT2_DIR_ENTRY_HEADER_LEN) { + name_len = 0; + rec_len = block_len - *offset; + } else { + name_len = ext2fs_dirent_name_len(dirent); + (void) ext2fs_get_rec_len(fs, dirent, &rec_len); + } left = block_len - *offset - rec_len; /* * Special case of directory entry of size 8: copy what's left * of the directory block up to cover up the invalid hole. */ - if ((left >= 12) && (rec_len == 8)) { - memmove(cp, cp+8, left); - memset(cp + left, 0, 8); + if ((left >= 12) && (rec_len == EXT2_DIR_ENTRY_HEADER_LEN)) { + memmove(cp, cp+EXT2_DIR_ENTRY_HEADER_LEN, left); + memset(cp + left, 0, EXT2_DIR_ENTRY_HEADER_LEN); return; } /* @@ -691,8 +742,8 @@ static void salvage_directory(ext2_filsys fs, * record length. */ if ((left < 0) && - ((int) rec_len + left > 8) && - ((int) name_len + 8 <= (int) rec_len + left) && + ((int) rec_len + left > EXT2_DIR_ENTRY_HEADER_LEN) && + ((int) name_len + EXT2_DIR_ENTRY_HEADER_LEN <= (int) rec_len + left) && dirent->inode <= fs->super->s_inodes_count && strnlen(dirent->name, name_len) == name_len) { (void) ext2fs_set_rec_len(fs, (int) rec_len + left, dirent); @@ -731,39 +782,25 @@ static void salvage_directory(ext2_filsys fs, } } -static int is_last_entry(ext2_filsys fs, int inline_data_size, - unsigned int offset, int csum_size) -{ - if (inline_data_size) - return (offset < inline_data_size); - else - return (offset < fs->blocksize - csum_size); -} - +#define NEXT_DIRENT(d) ((void *)((char *)(d) + (d)->rec_len)) static errcode_t insert_dirent_tail(ext2_filsys fs, void *dirbuf) { struct ext2_dir_entry *d; void *top; struct ext2_dir_entry_tail *t; - unsigned int rec_len; d = dirbuf; top = EXT2_DIRENT_TAIL(dirbuf, fs->blocksize); - rec_len = d->rec_len; - while (rec_len && !(rec_len & 0x3)) { - d = (struct ext2_dir_entry *)(((char *)d) + rec_len); - if (((void *)d) + d->rec_len >= top) - break; - rec_len = d->rec_len; - } + while (d->rec_len && !(d->rec_len & 0x3) && NEXT_DIRENT(d) <= top) + d = NEXT_DIRENT(d); if (d != top) { - size_t min_size = EXT2_DIR_REC_LEN( + unsigned int min_size = EXT2_DIR_REC_LEN( ext2fs_dirent_name_len(dirbuf)); - if (min_size > d->rec_len - sizeof(struct ext2_dir_entry_tail)) + if (min_size > (char *)top - (char *)d) return EXT2_ET_DIR_NO_SPACE_FOR_CSUM; - d->rec_len -= sizeof(struct ext2_dir_entry_tail); + d->rec_len = (char *)top - (char *)d; } t = (struct ext2_dir_entry_tail *)top; @@ -774,15 +811,102 @@ static errcode_t insert_dirent_tail(ext2_filsys fs, void *dirbuf) return 0; } +#undef NEXT_DIRENT + +static errcode_t fix_inline_dir_size(e2fsck_t ctx, ext2_ino_t ino, + size_t *inline_data_size, + struct problem_context *pctx, + char *buf) +{ + ext2_filsys fs = ctx->fs; + struct ext2_inode inode; + size_t new_size, old_size; + errcode_t retval; + + old_size = *inline_data_size; + /* + * If there's not enough bytes to start the "second" dir block + * (in the EA space) then truncate everything to the first block. + */ + if (old_size > EXT4_MIN_INLINE_DATA_SIZE && + old_size < EXT4_MIN_INLINE_DATA_SIZE + + EXT2_DIR_REC_LEN(1)) { + old_size = EXT4_MIN_INLINE_DATA_SIZE; + new_size = old_size; + } else + /* Increase to the next four-byte boundary for salvaging */ + new_size = old_size + (4 - (old_size & 3)); + memset(buf + old_size, 0, new_size - old_size); + retval = ext2fs_inline_data_set(fs, ino, 0, buf, new_size); + if (retval == EXT2_ET_INLINE_DATA_NO_SPACE) { + /* Or we can't, so truncate. */ + new_size -= 4; + retval = ext2fs_inline_data_set(fs, ino, 0, buf, new_size); + if (retval) { + if (fix_problem(ctx, PR_2_FIX_INLINE_DIR_FAILED, + pctx)) { + new_size = 0; + goto write_inode; + } + goto err; + } + } else if (retval) { + if (fix_problem(ctx, PR_2_FIX_INLINE_DIR_FAILED, + pctx)) { + new_size = 0; + goto write_inode; + } + goto err; + } + +write_inode: + retval = ext2fs_read_inode(fs, ino, &inode); + if (retval) + goto err; + + retval = ext2fs_inode_size_set(fs, &inode, new_size); + if (retval) + goto err; + if (new_size == 0) + inode.i_flags &= ~EXT4_INLINE_DATA_FL; + retval = ext2fs_write_inode(fs, ino, &inode); + if (retval) + goto err; + *inline_data_size = new_size; + +err: + return retval; +} + +static int check_dir_block2(ext2_filsys fs, + struct ext2_db_entry2 *db, + void *priv_data) +{ + int err; + struct check_dir_struct *cd = priv_data; + + if (cd->ra_entries && cd->list_offset >= cd->next_ra_off) { + err = e2fsck_readahead_dblist(fs, + E2FSCK_RA_DBLIST_IGNORE_BLOCKCNT, + fs->dblist, + cd->list_offset + cd->ra_entries / 8, + cd->ra_entries); + if (err) + cd->ra_entries = 0; + cd->next_ra_off = cd->list_offset + (cd->ra_entries * 7 / 8); + } + + err = check_dir_block(fs, db, priv_data); + cd->list_offset++; + return err; +} static int check_dir_block(ext2_filsys fs, struct ext2_db_entry2 *db, void *priv_data) { struct dx_dir_info *dx_dir; -#ifdef ENABLE_HTREE struct dx_dirblock_info *dx_db = 0; -#endif /* ENABLE_HTREE */ struct ext2_dir_entry *dirent, *prev, dot, dotdot; ext2_dirhash_t hash; unsigned int offset = 0; @@ -794,7 +918,7 @@ static int check_dir_block(ext2_filsys fs, ext2_ino_t subdir_parent; __u16 links; struct check_dir_struct *cd; - char *buf; + char *buf, *ibuf; e2fsck_t ctx; problem_t problem; struct ext2_dx_root_info *root; @@ -808,25 +932,27 @@ static int check_dir_block(ext2_filsys fs, int is_leaf = 1; size_t inline_data_size = 0; int filetype = 0; + int encrypted = 0; + size_t max_block_size; + int hash_flags = 0; + static char *eop_read_dirblock = NULL; cd = (struct check_dir_struct *) priv_data; - buf = cd->buf; + ibuf = buf = cd->buf; ctx = cd->ctx; - if (ctx->flags & E2F_FLAG_SIGNAL_MASK || ctx->flags & E2F_FLAG_RESTART) + if (ctx->flags & E2F_FLAG_RUN_RETURN) return DIRENT_ABORT; if (ctx->progress && (ctx->progress)(ctx, 2, cd->count++, cd->max)) return DIRENT_ABORT; - if (EXT2_HAS_RO_COMPAT_FEATURE(fs->super, - EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) { + if (ext2fs_has_feature_metadata_csum(fs->super)) { dx_csum_size = sizeof(struct ext2_dx_tail); de_csum_size = sizeof(struct ext2_dir_entry_tail); } - if (EXT2_HAS_INCOMPAT_FEATURE(fs->super, - EXT2_FEATURE_INCOMPAT_FILETYPE)) + if (ext2fs_has_feature_filetype(fs->super)) filetype = EXT2_FT_DIR << 8; /* @@ -843,8 +969,7 @@ static int check_dir_block(ext2_filsys fs, cd->pctx.dirent = 0; cd->pctx.num = 0; - if (EXT2_HAS_INCOMPAT_FEATURE(fs->super, - EXT4_FEATURE_INCOMPAT_INLINE_DATA)) { + if (ext2fs_has_feature_inline_data(fs->super)) { errcode_t ec; ec = ext2fs_inline_data_size(fs, ino, &inline_data_size); @@ -852,6 +977,15 @@ static int check_dir_block(ext2_filsys fs, return DIRENT_ABORT; } + /* This will allow (at some point in the future) to punch out empty + * directory blocks and reduce the space used by a directory that grows + * very large and then the files are deleted. For now, all that is + * needed is to avoid e2fsck filling in these holes as part of + * feature flag. */ + if (db->blk == 0 && ext2fs_has_feature_largedir(fs->super) && + !ext2fs_has_feature_inline_data(fs->super)) + return 0; + if (db->blk == 0 && !inline_data_size) { if (allocate_dir_block(ctx, db, buf, &cd->pctx)) return 0; @@ -872,12 +1006,51 @@ static int check_dir_block(ext2_filsys fs, db->blockcnt, ino); #endif - ehandler_operation(_("reading directory block")); - if (inline_data_size) + if (!eop_read_dirblock) + eop_read_dirblock = (char *) _("reading directory block"); + ehandler_operation(eop_read_dirblock); + if (inline_data_size) { + memset(buf, 0, fs->blocksize - inline_data_size); cd->pctx.errcode = ext2fs_inline_data_get(fs, ino, 0, buf, 0); - else + if (cd->pctx.errcode) + goto inline_read_fail; +#ifdef WORDS_BIGENDIAN + if (db->blockcnt) + goto skip_first_read_swab; + *((__u32 *)buf) = ext2fs_le32_to_cpu(*((__u32 *)buf)); + cd->pctx.errcode = ext2fs_dirent_swab_in2(fs, + buf + EXT4_INLINE_DATA_DOTDOT_SIZE, + EXT4_MIN_INLINE_DATA_SIZE - EXT4_INLINE_DATA_DOTDOT_SIZE, + 0); + if (cd->pctx.errcode) + goto inline_read_fail; +skip_first_read_swab: + if (inline_data_size <= EXT4_MIN_INLINE_DATA_SIZE || + !db->blockcnt) + goto inline_read_fail; + cd->pctx.errcode = ext2fs_dirent_swab_in2(fs, + buf + EXT4_MIN_INLINE_DATA_SIZE, + inline_data_size - EXT4_MIN_INLINE_DATA_SIZE, + 0); +#endif + } else cd->pctx.errcode = ext2fs_read_dir_block4(fs, block_nr, buf, 0, ino); +inline_read_fail: + pctx.ino = ino; + pctx.num = inline_data_size; + if (((inline_data_size & 3) || + (inline_data_size > EXT4_MIN_INLINE_DATA_SIZE && + inline_data_size < EXT4_MIN_INLINE_DATA_SIZE + + EXT2_DIR_REC_LEN(1))) && + fix_problem(ctx, PR_2_BAD_INLINE_DIR_SIZE, &pctx)) { + errcode_t err = fix_inline_dir_size(ctx, ino, + &inline_data_size, &pctx, + buf); + if (err) + return DIRENT_ABORT; + + } ehandler_operation(0); if (cd->pctx.errcode == EXT2_ET_DIR_CORRUPTED) cd->pctx.errcode = 0; /* We'll handle this ourselves */ @@ -896,10 +1069,10 @@ static int check_dir_block(ext2_filsys fs, memcpy(buf, buf2, fs->blocksize); ext2fs_free_mem(&buf2); } -#ifdef ENABLE_HTREE dx_dir = e2fsck_get_dx_dir_info(ctx, ino); if (dx_dir && dx_dir->numblocks) { if (db->blockcnt >= dx_dir->numblocks) { + pctx.dir = ino; if (fix_problem(ctx, PR_2_UNEXPECTED_HTREE_BLOCK, &pctx)) { clear_htree(ctx, ino); @@ -924,11 +1097,12 @@ static int check_dir_block(ext2_filsys fs, dx_db->flags |= DX_FLAG_FIRST | DX_FLAG_LAST; if ((root->reserved_zero || root->info_length < 8 || - root->indirect_levels > 1) && + root->indirect_levels >= + ext2_dir_htree_level(fs)) && fix_problem(ctx, PR_2_HTREE_BAD_ROOT, &cd->pctx)) { clear_htree(ctx, ino); dx_dir->numblocks = 0; - dx_db = 0; + dx_db = NULL; } dx_dir->hashversion = root->hash_version; if ((dx_dir->hashversion <= EXT2_HASH_TEA) && @@ -940,45 +1114,49 @@ static int check_dir_block(ext2_filsys fs, (ext2fs_dirent_name_len(dirent) == 0) && (ext2fs_le16_to_cpu(limit->limit) == ((fs->blocksize - (8 + dx_csum_size)) / - sizeof(struct ext2_dx_entry)))) + sizeof(struct ext2_dx_entry)))) { dx_db->type = DX_DIRBLOCK_NODE; - is_leaf = 0; + } + is_leaf = dx_db ? (dx_db->type == DX_DIRBLOCK_LEAF) : 0; } out_htree: -#endif /* ENABLE_HTREE */ - - /* Verify checksum. */ - if (is_leaf && de_csum_size && !inline_data_size) { - /* No space for csum? Rebuild dirs in pass 3A. */ - if (!ext2fs_dirent_has_tail(fs, (struct ext2_dir_entry *)buf)) { - de_csum_size = 0; - if (e2fsck_dir_will_be_rehashed(ctx, ino)) - goto skip_checksum; - if (!fix_problem(cd->ctx, PR_2_LEAF_NODE_MISSING_CSUM, - &cd->pctx)) - goto skip_checksum; - e2fsck_rehash_dir_later(ctx, ino); - goto skip_checksum; - } - if (failed_csum) { - char *buf2; - if (!fix_problem(cd->ctx, PR_2_LEAF_NODE_CSUM_INVALID, - &cd->pctx)) - goto skip_checksum; - ext2fs_new_dir_block(fs, - db->blockcnt == 0 ? ino : 0, - EXT2_ROOT_INO, &buf2); - memcpy(buf, buf2, fs->blocksize); - ext2fs_free_mem(&buf2); - dir_modified++; + + /* Leaf node with no space for csum? Rebuild dirs in pass 3A. */ + if (is_leaf && !inline_data_size && failed_csum && + !ext2fs_dirent_has_tail(fs, (struct ext2_dir_entry *)buf)) { + de_csum_size = 0; + if (e2fsck_dir_will_be_rehashed(ctx, ino)) { failed_csum = 0; + goto skip_checksum; } + if (!fix_problem(cd->ctx, PR_2_LEAF_NODE_MISSING_CSUM, + &cd->pctx)) + goto skip_checksum; + e2fsck_rehash_dir_later(ctx, ino); + failed_csum = 0; + goto skip_checksum; } /* htree nodes don't use fake dirents to store checksums */ if (!is_leaf) de_csum_size = 0; skip_checksum: + if (inline_data_size) { + if (db->blockcnt) { + buf += EXT4_MIN_INLINE_DATA_SIZE; + max_block_size = inline_data_size - EXT4_MIN_INLINE_DATA_SIZE; + /* Zero-length second block, just exit */ + if (max_block_size == 0) + return 0; + } else { + max_block_size = EXT4_MIN_INLINE_DATA_SIZE; + } + } else + max_block_size = fs->blocksize - de_csum_size; + + if (ctx->encrypted_dirs) + encrypted = ext2fs_u32_list_test(ctx->encrypted_dirs, ino); + dict_init(&de_dict, DICTCOUNT_T_MAX, dict_de_cmp); prev = 0; do { @@ -989,19 +1167,63 @@ skip_checksum: problem = 0; if (!inline_data_size || dot_state > 1) { dirent = (struct ext2_dir_entry *) (buf + offset); - (void) ext2fs_get_rec_len(fs, dirent, &rec_len); + /* + * If there's not even space for the entry header, + * force salvaging this dir. + */ + if (max_block_size - offset < EXT2_DIR_ENTRY_HEADER_LEN) + rec_len = EXT2_DIR_REC_LEN(1); + else + (void) ext2fs_get_rec_len(fs, dirent, &rec_len); cd->pctx.dirent = dirent; cd->pctx.num = offset; - if (((offset + rec_len) > fs->blocksize) || + if ((offset + rec_len > max_block_size) || (rec_len < 12) || ((rec_len % 4) != 0) || - ((ext2fs_dirent_name_len(dirent) + 8) > rec_len)) { + (((unsigned) ext2fs_dirent_name_len(dirent) + EXT2_DIR_ENTRY_HEADER_LEN) > rec_len)) { if (fix_problem(ctx, PR_2_DIR_CORRUPTED, &cd->pctx)) { +#ifdef WORDS_BIGENDIAN + /* + * On big-endian systems, if the dirent + * swap routine finds a rec_len that it + * doesn't like, it continues + * processing the block as if rec_len + * == EXT2_DIR_ENTRY_HEADER_LEN. This means that the name + * field gets byte swapped, which means + * that salvage will not detect the + * correct name length (unless the name + * has a length that's an exact + * multiple of four bytes), and it'll + * discard the entry (unnecessarily) + * and the rest of the dirent block. + * Therefore, swap the rest of the + * block back to disk order, run + * salvage, and re-swap anything after + * the salvaged dirent. + */ + int need_reswab = 0; + if (rec_len < EXT2_DIR_ENTRY_HEADER_LEN || rec_len % 4) { + need_reswab = 1; + ext2fs_dirent_swab_in2(fs, + ((char *)dirent) + EXT2_DIR_ENTRY_HEADER_LEN, + max_block_size - offset - EXT2_DIR_ENTRY_HEADER_LEN, + 0); + } +#endif salvage_directory(fs, dirent, prev, &offset, - fs->blocksize - - de_csum_size); + max_block_size); +#ifdef WORDS_BIGENDIAN + if (need_reswab) { + (void) ext2fs_get_rec_len(fs, + dirent, &rec_len); + ext2fs_dirent_swab_in2(fs, + ((char *)dirent) + offset + rec_len, + max_block_size - offset - rec_len, + 0); + } +#endif dir_modified++; continue; } else @@ -1197,23 +1419,31 @@ skip_checksum: } } - if (check_name(ctx, dirent, ino, &cd->pctx)) + if (!encrypted && check_name(ctx, dirent, &cd->pctx)) dir_modified++; + if (encrypted && (dot_state) > 1 && + encrypted_check_name(ctx, dirent, &cd->pctx)) { + dir_modified++; + goto next; + } + if (check_filetype(ctx, dirent, ino, &cd->pctx)) dir_modified++; -#ifdef ENABLE_HTREE if (dx_db) { - ext2fs_dirhash(dx_dir->hashversion, dirent->name, - ext2fs_dirent_name_len(dirent), - fs->super->s_hash_seed, &hash, 0); + if (dx_dir->casefolded_hash) + hash_flags = EXT4_CASEFOLD_FL; + + ext2fs_dirhash2(dx_dir->hashversion, dirent->name, + ext2fs_dirent_name_len(dirent), + fs->encoding, hash_flags, + fs->super->s_hash_seed, &hash, 0); if (hash < dx_db->min_hash) dx_db->min_hash = hash; if (hash > dx_db->max_hash) dx_db->max_hash = hash; } -#endif /* * If this is a directory, then mark its parent in its @@ -1270,15 +1500,25 @@ skip_checksum: if (!inline_data_size || dot_state > 1) { offset += rec_len; } else { - if (dot_state == 1) + if (dot_state == 1) { offset = 4; + /* + * If we get here, we're checking an inline + * directory and we've just checked a (fake) + * dotdot entry that we created on the stack. + * Therefore set 'prev' to NULL so that if we + * call salvage_directory on the next entry, + * it won't try to absorb the next entry into + * the on-stack dotdot entry. + */ + prev = NULL; + } } dot_state++; - } while (is_last_entry(fs, inline_data_size, offset, de_csum_size)); + } while (offset < max_block_size); #if 0 printf("\n"); #endif -#ifdef ENABLE_HTREE if (dx_db) { #ifdef DX_DEBUG printf("db_block %d, type %d, min_hash 0x%0x, max_hash 0x%0x\n", @@ -1290,30 +1530,18 @@ skip_checksum: (dx_db->type == DX_DIRBLOCK_NODE)) parse_int_node(fs, db, cd, dx_dir, buf, failed_csum); } -#endif /* ENABLE_HTREE */ - if (inline_data_size) { - if (offset != inline_data_size) { - cd->pctx.num = rec_len + offset - inline_data_size; - if (fix_problem(ctx, PR_2_FINAL_RECLEN, &cd->pctx)) { - dirent->rec_len = cd->pctx.num; - dir_modified++; - } - } - } else { - if (offset != fs->blocksize - de_csum_size) { - cd->pctx.num = rec_len - (fs->blocksize - de_csum_size) + - offset; - if (fix_problem(ctx, PR_2_FINAL_RECLEN, &cd->pctx)) { - dirent->rec_len = cd->pctx.num; - dir_modified++; - } + if (offset != max_block_size) { + cd->pctx.num = rec_len + offset - max_block_size; + if (fix_problem(ctx, PR_2_FINAL_RECLEN, &cd->pctx)) { + dirent->rec_len = cd->pctx.num; + dir_modified++; } } if (dir_modified) { + int flags, will_rehash; /* leaf block with no tail? Rehash dirs later. */ - if (EXT2_HAS_RO_COMPAT_FEATURE(fs->super, - EXT4_FEATURE_RO_COMPAT_METADATA_CSUM) && + if (ext2fs_has_feature_metadata_csum(fs->super) && is_leaf && !inline_data_size && !ext2fs_dirent_has_tail(fs, (struct ext2_dir_entry *)buf)) { @@ -1323,17 +1551,49 @@ skip_checksum: } write_and_fix: - if (e2fsck_dir_will_be_rehashed(ctx, ino)) + will_rehash = e2fsck_dir_will_be_rehashed(ctx, ino); + if (will_rehash) { + flags = ctx->fs->flags; ctx->fs->flags |= EXT2_FLAG_IGNORE_CSUM_ERRORS; + } if (inline_data_size) { + buf = ibuf; +#ifdef WORDS_BIGENDIAN + if (db->blockcnt) + goto skip_first_write_swab; + *((__u32 *)buf) = ext2fs_le32_to_cpu(*((__u32 *)buf)); + cd->pctx.errcode = ext2fs_dirent_swab_out2(fs, + buf + EXT4_INLINE_DATA_DOTDOT_SIZE, + EXT4_MIN_INLINE_DATA_SIZE - + EXT4_INLINE_DATA_DOTDOT_SIZE, + 0); + if (cd->pctx.errcode) + goto skip_second_write_swab; +skip_first_write_swab: + if (inline_data_size <= EXT4_MIN_INLINE_DATA_SIZE || + !db->blockcnt) + goto skip_second_write_swab; + cd->pctx.errcode = ext2fs_dirent_swab_out2(fs, + buf + EXT4_MIN_INLINE_DATA_SIZE, + inline_data_size - + EXT4_MIN_INLINE_DATA_SIZE, + 0); +skip_second_write_swab: + if (cd->pctx.errcode && + !fix_problem(ctx, PR_2_WRITE_DIRBLOCK, &cd->pctx)) + goto abort_free_dict; +#endif cd->pctx.errcode = ext2fs_inline_data_set(fs, ino, 0, buf, inline_data_size); } else cd->pctx.errcode = ext2fs_write_dir_block4(fs, block_nr, buf, 0, ino); - if (e2fsck_dir_will_be_rehashed(ctx, ino)) - ctx->fs->flags &= ~EXT2_FLAG_IGNORE_CSUM_ERRORS; + if (will_rehash) + ctx->fs->flags = (flags & + EXT2_FLAG_IGNORE_CSUM_ERRORS) | + (ctx->fs->flags & + ~EXT2_FLAG_IGNORE_CSUM_ERRORS); if (cd->pctx.errcode) { if (!fix_problem(ctx, PR_2_WRITE_DIRBLOCK, &cd->pctx)) @@ -1360,6 +1620,7 @@ abort_free_dict: struct del_block { e2fsck_t ctx; e2_blkcnt_t num; + blk64_t last_cluster; }; /* @@ -1374,20 +1635,26 @@ static int deallocate_inode_block(ext2_filsys fs, void *priv_data) { struct del_block *p = priv_data; + blk64_t cluster = EXT2FS_B2C(fs, *block_nr); - if (HOLE_BLKADDR(*block_nr)) + if (*block_nr == 0) return 0; + + if (cluster == p->last_cluster) + return 0; + + p->last_cluster = cluster; if ((*block_nr < fs->super->s_first_data_block) || (*block_nr >= ext2fs_blocks_count(fs->super))) return 0; - if ((*block_nr % EXT2FS_CLUSTER_RATIO(fs)) == 0) - ext2fs_block_alloc_stats2(fs, *block_nr, -1); + + ext2fs_block_alloc_stats2(fs, *block_nr, -1); p->num++; return 0; } /* - * This fuction deallocates an inode + * This function deallocates an inode */ static void deallocate_inode(e2fsck_t ctx, ext2_ino_t ino, char* block_buf) { @@ -1408,7 +1675,7 @@ static void deallocate_inode(e2fsck_t ctx, ext2_ino_t ino, char* block_buf) ext2fs_inode_alloc_stats2(fs, ino, -1, LINUX_S_ISDIR(inode.i_mode)); if (ext2fs_file_acl_block(fs, &inode) && - (fs->super->s_feature_compat & EXT2_FEATURE_COMPAT_EXT_ATTR)) { + ext2fs_has_feature_xattr(fs->super)) { pctx.errcode = ext2fs_adjust_ea_refcount3(fs, ext2fs_file_acl_block(fs, &inode), block_buf, -1, &count, ino); @@ -1432,12 +1699,17 @@ static void deallocate_inode(e2fsck_t ctx, ext2_ino_t ino, char* block_buf) if (!ext2fs_inode_has_valid_blocks2(fs, &inode)) goto clear_inode; + /* Inline data inodes don't have blocks to iterate */ + if (inode.i_flags & EXT4_INLINE_DATA_FL) + goto clear_inode; + if (LINUX_S_ISREG(inode.i_mode) && ext2fs_needs_large_file_feature(EXT2_I_SIZE(&inode))) ctx->large_files--; del_block.ctx = ctx; del_block.num = 0; + del_block.last_cluster = 0; pctx.errcode = ext2fs_block_iterate3(fs, ino, 0, block_buf, deallocate_inode_block, &del_block); @@ -1453,7 +1725,7 @@ clear_inode: } /* - * This fuction clears the htree flag on an inode + * This function clears the htree flag on an inode */ static void clear_htree(e2fsck_t ctx, ext2_ino_t ino) { @@ -1486,7 +1758,7 @@ int e2fsck_process_bad_inode(e2fsck_t ctx, ext2_ino_t dir, pctx.inode = &inode; if (ext2fs_file_acl_block(fs, &inode) && - !(fs->super->s_feature_compat & EXT2_FEATURE_COMPAT_EXT_ATTR)) { + !ext2fs_has_feature_xattr(fs->super)) { if (fix_problem(ctx, PR_2_FILE_ACL_ZERO, &pctx)) { ext2fs_file_acl_block_set(fs, &inode, 0); inode_modified++; @@ -1563,8 +1835,7 @@ int e2fsck_process_bad_inode(e2fsck_t ctx, ext2_ino_t dir, } if ((fs->super->s_creator_os == EXT2_OS_LINUX) && - !(fs->super->s_feature_ro_compat & - EXT4_FEATURE_RO_COMPAT_HUGE_FILE) && + !ext2fs_has_feature_huge_file(fs->super) && (inode.osd2.linux2.l_i_blocks_hi != 0)) { pctx.num = inode.osd2.linux2.l_i_blocks_hi; if (fix_problem(ctx, PR_2_BLOCKS_HI_ZERO, &pctx)) { @@ -1573,8 +1844,8 @@ int e2fsck_process_bad_inode(e2fsck_t ctx, ext2_ino_t dir, } } - if (!(fs->super->s_feature_incompat & - EXT4_FEATURE_INCOMPAT_64BIT) && + if ((fs->super->s_creator_os == EXT2_OS_LINUX) && + !ext2fs_has_feature_64bit(fs->super) && inode.osd2.linux2.l_i_file_acl_high != 0) { pctx.num = inode.osd2.linux2.l_i_file_acl_high; if (fix_problem(ctx, PR_2_I_FILE_ACL_HI_ZERO, &pctx)) { @@ -1593,10 +1864,10 @@ int e2fsck_process_bad_inode(e2fsck_t ctx, ext2_ino_t dir, } else not_fixed++; } - if (inode.i_dir_acl && + if (inode.i_size_high && !ext2fs_has_feature_largedir(fs->super) && LINUX_S_ISDIR(inode.i_mode)) { - if (fix_problem(ctx, PR_2_DIR_ACL_ZERO, &pctx)) { - inode.i_dir_acl = 0; + if (fix_problem(ctx, PR_2_DIR_SIZE_HIGH_ZERO, &pctx)) { + inode.i_size_high = 0; inode_modified++; } else not_fixed++; @@ -1609,7 +1880,6 @@ int e2fsck_process_bad_inode(e2fsck_t ctx, ext2_ino_t dir, return 0; } - /* * allocate_dir_block --- this function allocates a new directory * block for a particular inode; this is done if a directory has @@ -1642,7 +1912,8 @@ static int allocate_dir_block(e2fsck_t ctx, pctx->errcode = ext2fs_map_cluster_block(fs, db->ino, &inode, db->blockcnt, &blk); if (pctx->errcode || blk == 0) { - pctx->errcode = ext2fs_new_block2(fs, 0, + blk = ext2fs_find_inode_goal(fs, db->ino, &inode, db->blockcnt); + pctx->errcode = ext2fs_new_block2(fs, blk, ctx->block_found_map, &blk); if (pctx->errcode) { pctx->str = "ext2fs_new_block"; @@ -1681,7 +1952,7 @@ static int allocate_dir_block(e2fsck_t ctx, * Update the inode block count */ ext2fs_iblk_add_blocks(fs, &inode, 1); - if (EXT2_I_SIZE(&inode) < (db->blockcnt+1) * fs->blocksize) { + if (EXT2_I_SIZE(&inode) < ((__u64) db->blockcnt+1) * fs->blocksize) { pctx->errcode = ext2fs_inode_size_set(fs, &inode, (db->blockcnt+1) * fs->blocksize); if (pctx->errcode) {