// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) Qu Wenruo 2017. All rights reserved.
*/
/* * The module is used to catch unexpected/corrupted tree block data. * Such behavior can be caused either by a fuzzed image or bugs. * * The objective is to do leaf/node validation checks when tree block is read * from disk, and check *every* possible member, so other code won't * need to checking them again. * * Due to the potential and unwanted damage, every checker needs to be * carefully reviewed otherwise so it does not prevent mount of valid images.
*/
/* * Error message should follow the following format: * corrupt <type>: <identifier>, <reason>[, <bad_value>] * * @type: leaf or node * @identifier: the necessary info to locate the leaf/node. * It's recommended to decode key.objecitd/offset if it's * meaningful. * @reason: describe the error * @bad_value: optional, it's recommended to output bad value and its * expected value (range). * * Since comma is used to separate the components, only space is allowed * inside each component.
*/
if (btrfs_file_extent_type(leaf, extent) == BTRFS_FILE_EXTENT_INLINE) {
len = btrfs_file_extent_ram_bytes(leaf, extent);
end = ALIGN(key->offset + len, leaf->fs_info->sectorsize);
} else {
len = btrfs_file_extent_num_bytes(leaf, extent);
end = key->offset + len;
} return end;
}
/* * Customized report for dir_item, the only new important information is * key->objectid, which represents inode number
*/
__printf(3, 4)
__cold staticvoid dir_item_err(conststruct extent_buffer *eb, int slot, constchar *fmt, ...)
{ conststruct btrfs_fs_info *fs_info = eb->fs_info; struct btrfs_key key; struct va_format vaf;
va_list args;
/* * This functions checks prev_key->objectid, to ensure current key and prev_key * share the same objectid as inode number. * * This is to detect missing INODE_ITEM in subvolume trees. * * Return true if everything is OK or we don't need to check. * Return false if anything is wrong.
*/ staticbool check_prev_ino(struct extent_buffer *leaf, struct btrfs_key *key, int slot, struct btrfs_key *prev_key)
{ /* No prev key, skip check */ if (slot == 0) returntrue;
/* Only these key->types needs to be checked */
ASSERT(key->type == BTRFS_XATTR_ITEM_KEY ||
key->type == BTRFS_INODE_REF_KEY ||
key->type == BTRFS_INODE_EXTREF_KEY ||
key->type == BTRFS_DIR_INDEX_KEY ||
key->type == BTRFS_DIR_ITEM_KEY ||
key->type == BTRFS_EXTENT_DATA_KEY);
/* * Only subvolume trees along with their reloc trees need this check. * Things like log tree doesn't follow this ino requirement.
*/ if (!btrfs_is_fstree(btrfs_header_owner(leaf))) returntrue;
if (key->objectid == prev_key->objectid) returntrue;
if (unlikely(!IS_ALIGNED(key->offset, sectorsize))) {
file_extent_err(leaf, slot, "unaligned file_offset for file extent, have %llu should be aligned to %u",
key->offset, sectorsize); return -EUCLEAN;
}
/* * Previous key must have the same key->objectid (ino). * It can be XATTR_ITEM, INODE_ITEM or just another EXTENT_DATA. * But if objectids mismatch, it means we have a missing * INODE_ITEM.
*/ if (unlikely(!check_prev_ino(leaf, key, slot, prev_key))) return -EUCLEAN;
fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
/* * Make sure the item contains at least inline header, so the file * extent type is not some garbage.
*/ if (unlikely(item_size < BTRFS_FILE_EXTENT_INLINE_DATA_START)) {
file_extent_err(leaf, slot, "invalid item size, have %u expect [%zu, %u)",
item_size, BTRFS_FILE_EXTENT_INLINE_DATA_START,
SZ_4K); return -EUCLEAN;
} if (unlikely(btrfs_file_extent_type(leaf, fi) >=
BTRFS_NR_FILE_EXTENT_TYPES)) {
file_extent_err(leaf, slot, "invalid type for file extent, have %u expect range [0, %u]",
btrfs_file_extent_type(leaf, fi),
BTRFS_NR_FILE_EXTENT_TYPES - 1); return -EUCLEAN;
}
/* * Support for new compression/encryption must introduce incompat flag, * and must be caught in open_ctree().
*/ if (unlikely(btrfs_file_extent_compression(leaf, fi) >=
BTRFS_NR_COMPRESS_TYPES)) {
file_extent_err(leaf, slot, "invalid compression for file extent, have %u expect range [0, %u]",
btrfs_file_extent_compression(leaf, fi),
BTRFS_NR_COMPRESS_TYPES - 1); return -EUCLEAN;
} if (unlikely(btrfs_file_extent_encryption(leaf, fi))) {
file_extent_err(leaf, slot, "invalid encryption for file extent, have %u expect 0",
btrfs_file_extent_encryption(leaf, fi)); return -EUCLEAN;
} if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) { /* Inline extent must have 0 as key offset */ if (unlikely(key->offset)) {
file_extent_err(leaf, slot, "invalid file_offset for inline file extent, have %llu expect 0",
key->offset); return -EUCLEAN;
}
/* Compressed inline extent has no on-disk size, skip it */ if (btrfs_file_extent_compression(leaf, fi) !=
BTRFS_COMPRESS_NONE) return 0;
/* Uncompressed inline extent size must match item size */ if (unlikely(item_size != BTRFS_FILE_EXTENT_INLINE_DATA_START +
btrfs_file_extent_ram_bytes(leaf, fi))) {
file_extent_err(leaf, slot, "invalid ram_bytes for uncompressed inline extent, have %u expect %llu",
item_size, BTRFS_FILE_EXTENT_INLINE_DATA_START +
btrfs_file_extent_ram_bytes(leaf, fi)); return -EUCLEAN;
} return 0;
}
/* Catch extent end overflow */ if (unlikely(check_add_overflow(btrfs_file_extent_num_bytes(leaf, fi),
key->offset, &extent_end))) {
file_extent_err(leaf, slot, "extent end overflow, have file offset %llu extent num bytes %llu",
key->offset,
btrfs_file_extent_num_bytes(leaf, fi)); return -EUCLEAN;
}
/* * Check that no two consecutive file extent items, in the same leaf, * present ranges that overlap each other.
*/ if (slot > 0 &&
prev_key->objectid == key->objectid &&
prev_key->type == BTRFS_EXTENT_DATA_KEY) { struct btrfs_file_extent_item *prev_fi;
u64 prev_end;
prev_fi = btrfs_item_ptr(leaf, slot - 1, struct btrfs_file_extent_item);
prev_end = file_extent_end(leaf, prev_key, prev_fi); if (unlikely(prev_end > key->offset)) {
file_extent_err(leaf, slot - 1, "file extent end range (%llu) goes beyond start offset (%llu) of the next file extent",
prev_end, key->offset); return -EUCLEAN;
}
}
/* * For non-compressed data extents, ram_bytes should match its * disk_num_bytes. * However we do not really utilize ram_bytes in this case, so this check * is only optional for DEBUG builds for developers to catch the * unexpected behaviors.
*/ if (IS_ENABLED(CONFIG_BTRFS_DEBUG) &&
btrfs_file_extent_compression(leaf, fi) == BTRFS_COMPRESS_NONE &&
btrfs_file_extent_disk_bytenr(leaf, fi)) { if (WARN_ON(btrfs_file_extent_ram_bytes(leaf, fi) !=
btrfs_file_extent_disk_num_bytes(leaf, fi)))
file_extent_err(leaf, slot, "mismatch ram_bytes (%llu) and disk_num_bytes (%llu) for non-compressed extent",
btrfs_file_extent_ram_bytes(leaf, fi),
btrfs_file_extent_disk_num_bytes(leaf, fi));
}
if (unlikely(key->objectid != BTRFS_EXTENT_CSUM_OBJECTID)) {
generic_err(leaf, slot, "invalid key objectid for csum item, have %llu expect %llu",
key->objectid, BTRFS_EXTENT_CSUM_OBJECTID); return -EUCLEAN;
} if (unlikely(!IS_ALIGNED(key->offset, sectorsize))) {
generic_err(leaf, slot, "unaligned key offset for csum item, have %llu should be aligned to %u",
key->offset, sectorsize); return -EUCLEAN;
} if (unlikely(!IS_ALIGNED(btrfs_item_size(leaf, slot), csumsize))) {
generic_err(leaf, slot, "unaligned item size for csum item, have %u should be aligned to %u",
btrfs_item_size(leaf, slot), csumsize); return -EUCLEAN;
} if (slot > 0 && prev_key->type == BTRFS_EXTENT_CSUM_KEY) {
u64 prev_csum_end;
u32 prev_item_size;
prev_item_size = btrfs_item_size(leaf, slot - 1);
prev_csum_end = (prev_item_size / csumsize) * sectorsize;
prev_csum_end += prev_key->offset; if (unlikely(prev_csum_end > key->offset)) {
generic_err(leaf, slot - 1, "csum end range (%llu) goes beyond the start range (%llu) of the next csum item",
prev_csum_end, key->offset); return -EUCLEAN;
}
} return 0;
}
/* Inode item error output has the same format as dir_item_err() */ #define inode_item_err(eb, slot, fmt, ...) \
dir_item_err(eb, slot, fmt, __VA_ARGS__)
/* * Bad rootid for reloc trees. * * Reloc trees are only for subvolume trees, other trees only need * to be COWed to be relocated.
*/ if (unlikely(is_root_item && key->objectid == BTRFS_TREE_RELOC_OBJECTID &&
!btrfs_is_fstree(key->offset))) {
generic_err(leaf, slot, "invalid reloc tree for root %lld, root id is not a subvolume tree",
key->offset); return -EUCLEAN;
}
/* No such tree id */ if (unlikely(key->objectid == 0)) { if (is_root_item)
generic_err(leaf, slot, "invalid root id 0"); else
dir_item_err(leaf, slot, "invalid location key root id 0"); return -EUCLEAN;
}
/* DIR_ITEM/INDEX/INODE_REF is not allowed to point to non-fs trees */ if (unlikely(!btrfs_is_fstree(key->objectid) && !is_root_item)) {
dir_item_err(leaf, slot, "invalid location key objectid, have %llu expect [%llu, %llu]",
key->objectid, BTRFS_FIRST_FREE_OBJECTID,
BTRFS_LAST_FREE_OBJECTID); return -EUCLEAN;
}
/* * ROOT_ITEM with non-zero offset means this is a snapshot, created at * @offset transid. * Furthermore, for location key in DIR_ITEM, its offset is always -1. * * So here we only check offset for reloc tree whose key->offset must * be a valid tree.
*/ if (unlikely(key->objectid == BTRFS_TREE_RELOC_OBJECTID &&
key->offset == 0)) {
generic_err(leaf, slot, "invalid root id 0 for reloc tree"); return -EUCLEAN;
} return 0;
}
if (unlikely(!check_prev_ino(leaf, key, slot, prev_key))) return -EUCLEAN;
di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item); while (cur < item_size) { struct btrfs_key location_key;
u32 name_len;
u32 data_len;
u32 max_name_len;
u32 total_size;
u32 name_hash;
u8 dir_type; int ret;
/* header itself should not cross item boundary */ if (unlikely(cur + sizeof(*di) > item_size)) {
dir_item_err(leaf, slot, "dir item header crosses item boundary, have %zu boundary %u",
cur + sizeof(*di), item_size); return -EUCLEAN;
}
/* Location key check */
btrfs_dir_item_key_to_cpu(leaf, di, &location_key); if (location_key.type == BTRFS_ROOT_ITEM_KEY) {
ret = check_root_key(leaf, &location_key, slot); if (unlikely(ret < 0)) return ret;
} elseif (location_key.type == BTRFS_INODE_ITEM_KEY ||
location_key.type == 0) {
ret = check_inode_key(leaf, &location_key, slot); if (unlikely(ret < 0)) return ret;
} else {
dir_item_err(leaf, slot, "invalid location key type, have %u, expect %u or %u",
location_key.type, BTRFS_ROOT_ITEM_KEY,
BTRFS_INODE_ITEM_KEY); return -EUCLEAN;
}
/* dir type check */
dir_type = btrfs_dir_ftype(leaf, di); if (unlikely(dir_type <= BTRFS_FT_UNKNOWN ||
dir_type >= BTRFS_FT_MAX)) {
dir_item_err(leaf, slot, "invalid dir item type, have %u expect (0, %u)",
dir_type, BTRFS_FT_MAX); return -EUCLEAN;
}
if (unlikely(key->type == BTRFS_XATTR_ITEM_KEY &&
dir_type != BTRFS_FT_XATTR)) {
dir_item_err(leaf, slot, "invalid dir item type for XATTR key, have %u expect %u",
dir_type, BTRFS_FT_XATTR); return -EUCLEAN;
} if (unlikely(dir_type == BTRFS_FT_XATTR &&
key->type != BTRFS_XATTR_ITEM_KEY)) {
dir_item_err(leaf, slot, "xattr dir type found for non-XATTR key"); return -EUCLEAN;
} if (dir_type == BTRFS_FT_XATTR)
max_name_len = XATTR_NAME_MAX; else
max_name_len = BTRFS_NAME_LEN;
/* Name/data length check */
name_len = btrfs_dir_name_len(leaf, di);
data_len = btrfs_dir_data_len(leaf, di); if (unlikely(name_len > max_name_len)) {
dir_item_err(leaf, slot, "dir item name len too long, have %u max %u",
name_len, max_name_len); return -EUCLEAN;
} if (unlikely(name_len + data_len > BTRFS_MAX_XATTR_SIZE(fs_info))) {
dir_item_err(leaf, slot, "dir item name and data len too long, have %u max %u",
name_len + data_len,
BTRFS_MAX_XATTR_SIZE(fs_info)); return -EUCLEAN;
}
if (unlikely(data_len && dir_type != BTRFS_FT_XATTR)) {
dir_item_err(leaf, slot, "dir item with invalid data len, have %u expect 0",
data_len); return -EUCLEAN;
}
total_size = sizeof(*di) + name_len + data_len;
/* header and name/data should not cross item boundary */ if (unlikely(cur + total_size > item_size)) {
dir_item_err(leaf, slot, "dir item data crosses item boundary, have %u boundary %u",
cur + total_size, item_size); return -EUCLEAN;
}
/* * Special check for XATTR/DIR_ITEM, as key->offset is name * hash, should match its name
*/ if (key->type == BTRFS_DIR_ITEM_KEY ||
key->type == BTRFS_XATTR_ITEM_KEY) { char namebuf[MAX(BTRFS_NAME_LEN, XATTR_NAME_MAX)];
read_extent_buffer(leaf, namebuf,
(unsignedlong)(di + 1), name_len);
name_hash = btrfs_name_hash(namebuf, name_len); if (unlikely(key->offset != name_hash)) {
dir_item_err(leaf, slot, "name hash mismatch with key, have 0x%016x expect 0x%016llx",
name_hash, key->offset); return -EUCLEAN;
}
}
cur += total_size;
di = (struct btrfs_dir_item *)((void *)di + total_size);
} return 0;
}
/* * Here we don't really care about alignment since extent allocator can * handle it. We care more about the size.
*/ if (unlikely(key->offset == 0)) {
block_group_err(leaf, slot, "invalid block group size 0"); return -EUCLEAN;
}
if (unlikely(item_size != sizeof(bgi))) {
block_group_err(leaf, slot, "invalid item size, have %u expect %zu",
item_size, sizeof(bgi)); return -EUCLEAN;
}
read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot), sizeof(bgi));
chunk_objectid = btrfs_stack_block_group_chunk_objectid(&bgi); if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) { /* * We don't init the nr_global_roots until we load the global * roots, so this could be 0 at mount time. If it's 0 we'll * just assume we're fine, and later we'll check against our * actual value.
*/ if (unlikely(fs_info->nr_global_roots &&
chunk_objectid >= fs_info->nr_global_roots)) {
block_group_err(leaf, slot, "invalid block group global root id, have %llu, needs to be <= %llu",
chunk_objectid,
fs_info->nr_global_roots); return -EUCLEAN;
}
} elseif (unlikely(chunk_objectid != BTRFS_FIRST_CHUNK_TREE_OBJECTID)) {
block_group_err(leaf, slot, "invalid block group chunk objectid, have %llu expect %llu",
btrfs_stack_block_group_chunk_objectid(&bgi),
BTRFS_FIRST_CHUNK_TREE_OBJECTID); return -EUCLEAN;
}
if (unlikely(btrfs_stack_block_group_used(&bgi) > key->offset)) {
block_group_err(leaf, slot, "invalid block group used, have %llu expect [0, %llu)",
btrfs_stack_block_group_used(&bgi), key->offset); return -EUCLEAN;
}
flags = btrfs_stack_block_group_flags(&bgi); if (unlikely(hweight64(flags & BTRFS_BLOCK_GROUP_PROFILE_MASK) > 1)) {
block_group_err(leaf, slot, "invalid profile flags, have 0x%llx (%lu bits set) expect no more than 1 bit set",
flags & BTRFS_BLOCK_GROUP_PROFILE_MASK,
hweight64(flags & BTRFS_BLOCK_GROUP_PROFILE_MASK)); return -EUCLEAN;
}
type = flags & BTRFS_BLOCK_GROUP_TYPE_MASK; if (unlikely(type != BTRFS_BLOCK_GROUP_DATA &&
type != BTRFS_BLOCK_GROUP_METADATA &&
type != BTRFS_BLOCK_GROUP_SYSTEM &&
type != (BTRFS_BLOCK_GROUP_METADATA |
BTRFS_BLOCK_GROUP_DATA))) {
block_group_err(leaf, slot, "invalid type, have 0x%llx (%lu bits set) expect either 0x%llx, 0x%llx, 0x%llx or 0x%llx",
type, hweight64(type),
BTRFS_BLOCK_GROUP_DATA, BTRFS_BLOCK_GROUP_METADATA,
BTRFS_BLOCK_GROUP_SYSTEM,
BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA); return -EUCLEAN;
} return 0;
}
if (!is_sb) { /* * Get the slot number by iterating through all slots, this * would provide better readability.
*/ for (i = 0; i < btrfs_header_nritems(leaf); i++) { if (btrfs_item_ptr_offset(leaf, i) ==
(unsignedlong)chunk) {
slot = i; break;
}
}
}
va_start(args, fmt);
vaf.fmt = fmt;
vaf.va = &args;
/* * The common chunk check which could also work on super block sys chunk array. * * If @leaf is NULL, then @chunk must be an on-stack chunk item. * (For superblock sys_chunk array, and fs_info->sectorsize is unreliable) * * Return -EUCLEAN if anything is corrupted. * Return 0 if everything is OK.
*/ int btrfs_check_chunk_valid(conststruct btrfs_fs_info *fs_info, conststruct extent_buffer *leaf, conststruct btrfs_chunk *chunk, u64 logical,
u32 sectorsize)
{
u64 length;
u64 chunk_end;
u64 stripe_len;
u16 num_stripes;
u16 sub_stripes;
u64 type;
u64 features;
u32 chunk_sector_size; bool mixed = false; int raid_index; int nparity; int ncopies;
if (unlikely(!num_stripes)) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk num_stripes, have %u", num_stripes); return -EUCLEAN;
} if (unlikely(num_stripes < ncopies)) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk num_stripes < ncopies, have %u < %d",
num_stripes, ncopies); return -EUCLEAN;
} if (unlikely(nparity && num_stripes == nparity)) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk num_stripes == nparity, have %u == %d",
num_stripes, nparity); return -EUCLEAN;
} if (unlikely(!IS_ALIGNED(logical, sectorsize))) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk logical, have %llu should aligned to %u",
logical, sectorsize); return -EUCLEAN;
} if (unlikely(chunk_sector_size != sectorsize)) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk sectorsize, have %u expect %u",
chunk_sector_size, sectorsize); return -EUCLEAN;
} if (unlikely(!length || !IS_ALIGNED(length, sectorsize))) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk length, have %llu", length); return -EUCLEAN;
} if (unlikely(check_add_overflow(logical, length, &chunk_end))) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk logical start and length, have logical start %llu length %llu",
logical, length); return -EUCLEAN;
} if (unlikely(!is_power_of_2(stripe_len) || stripe_len != BTRFS_STRIPE_LEN)) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk stripe length: %llu",
stripe_len); return -EUCLEAN;
} /* * We artificially limit the chunk size, so that the number of stripes * inside a chunk can be fit into a U32. The current limit (256G) is * way too large for real world usage anyway, and it's also much larger * than our existing limit (10G). * * Thus it should be a good way to catch obvious bitflips.
*/ if (unlikely(length >= btrfs_stripe_nr_to_offset(U32_MAX))) {
chunk_err(fs_info, leaf, chunk, logical, "chunk length too large: have %llu limit %llu",
length, btrfs_stripe_nr_to_offset(U32_MAX)); return -EUCLEAN;
} if (unlikely(type & ~(BTRFS_BLOCK_GROUP_TYPE_MASK |
BTRFS_BLOCK_GROUP_PROFILE_MASK))) {
chunk_err(fs_info, leaf, chunk, logical, "unrecognized chunk type: 0x%llx",
~(BTRFS_BLOCK_GROUP_TYPE_MASK |
BTRFS_BLOCK_GROUP_PROFILE_MASK) & type); return -EUCLEAN;
}
if (unlikely(!has_single_bit_set(type & BTRFS_BLOCK_GROUP_PROFILE_MASK) &&
(type & BTRFS_BLOCK_GROUP_PROFILE_MASK) != 0)) {
chunk_err(fs_info, leaf, chunk, logical, "invalid chunk profile flag: 0x%llx, expect 0 or 1 bit set",
type & BTRFS_BLOCK_GROUP_PROFILE_MASK); return -EUCLEAN;
} if (unlikely((type & BTRFS_BLOCK_GROUP_TYPE_MASK) == 0)) {
chunk_err(fs_info, leaf, chunk, logical, "missing chunk type flag, have 0x%llx one bit must be set in 0x%llx",
type, BTRFS_BLOCK_GROUP_TYPE_MASK); return -EUCLEAN;
}
if (unlikely((type & BTRFS_BLOCK_GROUP_SYSTEM) &&
(type & (BTRFS_BLOCK_GROUP_METADATA |
BTRFS_BLOCK_GROUP_DATA)))) {
chunk_err(fs_info, leaf, chunk, logical, "system chunk with data or metadata type: 0x%llx",
type); return -EUCLEAN;
}
features = btrfs_super_incompat_flags(fs_info->super_copy); if (features & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS)
mixed = true;
if (!mixed) { if (unlikely((type & BTRFS_BLOCK_GROUP_METADATA) &&
(type & BTRFS_BLOCK_GROUP_DATA))) {
chunk_err(fs_info, leaf, chunk, logical, "mixed chunk type in non-mixed mode: 0x%llx", type); return -EUCLEAN;
}
}
/* * Enhanced version of chunk item checker. * * The common btrfs_check_chunk_valid() doesn't check item size since it needs * to work on super block sys_chunk_array which doesn't have full item ptr.
*/ staticint check_leaf_chunk_item(struct extent_buffer *leaf, struct btrfs_chunk *chunk, struct btrfs_key *key, int slot)
{ struct btrfs_fs_info *fs_info = leaf->fs_info; int num_stripes;
/* * For device total_bytes, we don't have reliable way to check it, as * it can be 0 for device removal. Device size check can only be done * by dev extents check.
*/ if (unlikely(btrfs_device_bytes_used(leaf, ditem) >
btrfs_device_total_bytes(leaf, ditem))) {
dev_item_err(leaf, slot, "invalid bytes used: have %llu expect [0, %llu]",
btrfs_device_bytes_used(leaf, ditem),
btrfs_device_total_bytes(leaf, ditem)); return -EUCLEAN;
} /* * Remaining members like io_align/type/gen/dev_group aren't really * utilized. Skip them to make later usage of them easier.
*/ return 0;
}
/* Here we use super block generation + 1 to handle log tree */ if (unlikely(btrfs_inode_generation(leaf, iitem) > super_gen + 1)) {
inode_item_err(leaf, slot, "invalid inode generation: has %llu expect (0, %llu]",
btrfs_inode_generation(leaf, iitem),
super_gen + 1); return -EUCLEAN;
} /* Note for ROOT_TREE_DIR_ITEM, mkfs could set its transid 0 */ if (unlikely(btrfs_inode_transid(leaf, iitem) > super_gen + 1)) {
inode_item_err(leaf, slot, "invalid inode transid: has %llu expect [0, %llu]",
btrfs_inode_transid(leaf, iitem), super_gen + 1); return -EUCLEAN;
}
/* * For size and nbytes it's better not to be too strict, as for dir * item its size/nbytes can easily get wrong, but doesn't affect * anything in the fs. So here we skip the check.
*/
mode = btrfs_inode_mode(leaf, iitem); if (unlikely(mode & ~valid_mask)) {
inode_item_err(leaf, slot, "unknown mode bit detected: 0x%x",
mode & ~valid_mask); return -EUCLEAN;
}
/* * S_IFMT is not bit mapped so we can't completely rely on * is_power_of_2/has_single_bit_set, but it can save us from checking * FIFO/CHR/DIR/REG. Only needs to check BLK, LNK and SOCKS
*/ if (!has_single_bit_set(mode & S_IFMT)) { if (unlikely(!S_ISLNK(mode) && !S_ISBLK(mode) && !S_ISSOCK(mode))) {
inode_item_err(leaf, slot, "invalid mode: has 0%o expect valid S_IF* bit(s)",
mode & S_IFMT); return -EUCLEAN;
}
} if (unlikely(S_ISDIR(mode) && btrfs_inode_nlink(leaf, iitem) > 1)) {
inode_item_err(leaf, slot, "invalid nlink: has %u expect no more than 1 for dir",
btrfs_inode_nlink(leaf, iitem)); return -EUCLEAN;
}
btrfs_inode_split_flags(btrfs_inode_flags(leaf, iitem), &flags, &ro_flags); if (unlikely(flags & ~BTRFS_INODE_FLAG_MASK)) {
inode_item_err(leaf, slot, "unknown incompat flags detected: 0x%x", flags); return -EUCLEAN;
} if (unlikely(!sb_rdonly(fs_info->sb) &&
(ro_flags & ~BTRFS_INODE_RO_FLAG_MASK))) {
inode_item_err(leaf, slot, "unknown ro-compat flags detected on writeable mount: 0x%x",
ro_flags); return -EUCLEAN;
} return 0;
}
ret = check_root_key(leaf, key, slot); if (unlikely(ret < 0)) return ret;
if (unlikely(btrfs_item_size(leaf, slot) != sizeof(ri) &&
btrfs_item_size(leaf, slot) !=
btrfs_legacy_root_item_size())) {
generic_err(leaf, slot, "invalid root item size, have %u expect %zu or %u",
btrfs_item_size(leaf, slot), sizeof(ri),
btrfs_legacy_root_item_size()); return -EUCLEAN;
}
/* * For legacy root item, the members starting at generation_v2 will be * all filled with 0. * And since we allow geneartion_v2 as 0, it will still pass the check.
*/
read_extent_buffer(leaf, &ri, btrfs_item_ptr_offset(leaf, slot),
btrfs_item_size(leaf, slot));
staticbool is_valid_dref_root(u64 rootid)
{ /* * The following tree root objectids are allowed to have a data backref: * - subvolume trees * - data reloc tree * - tree root * For v1 space cache
*/ return btrfs_is_fstree(rootid) || rootid == BTRFS_DATA_RELOC_TREE_OBJECTID ||
rootid == BTRFS_ROOT_TREE_OBJECTID;
}
if (unlikely(key->type == BTRFS_METADATA_ITEM_KEY &&
!btrfs_fs_incompat(fs_info, SKINNY_METADATA))) {
generic_err(leaf, slot, "invalid key type, METADATA_ITEM type invalid when SKINNY_METADATA feature disabled"); return -EUCLEAN;
} /* key->objectid is the bytenr for both key types */ if (unlikely(!IS_ALIGNED(key->objectid, fs_info->sectorsize))) {
generic_err(leaf, slot, "invalid key objectid, have %llu expect to be aligned to %u",
key->objectid, fs_info->sectorsize); return -EUCLEAN;
}
/* key->offset is tree level for METADATA_ITEM_KEY */ if (unlikely(key->type == BTRFS_METADATA_ITEM_KEY &&
key->offset >= BTRFS_MAX_LEVEL)) {
extent_err(leaf, slot, "invalid tree level, have %llu expect [0, %u]",
key->offset, BTRFS_MAX_LEVEL - 1); return -EUCLEAN;
}
/* * EXTENT/METADATA_ITEM consists of: * 1) One btrfs_extent_item * Records the total refs, type and generation of the extent. * * 2) One btrfs_tree_block_info (for EXTENT_ITEM and tree backref only) * Records the first key and level of the tree block. * * 2) Zero or more btrfs_extent_inline_ref(s) * Each inline ref has one btrfs_extent_inline_ref shows: * 2.1) The ref type, one of the 4 * TREE_BLOCK_REF Tree block only * SHARED_BLOCK_REF Tree block only * EXTENT_DATA_REF Data only * SHARED_DATA_REF Data only * 2.2) Ref type specific data * Either using btrfs_extent_inline_ref::offset, or specific * data structure. * * All above inline items should follow the order: * * - All btrfs_extent_inline_ref::type should be in an ascending * order * * - Within the same type, the items should follow a descending * order by their sequence number. The sequence number is * determined by: * * btrfs_extent_inline_ref::offset for all types other than * EXTENT_DATA_REF * * hash_extent_data_ref() for EXTENT_DATA_REF
*/ if (unlikely(item_size < sizeof(*ei))) {
extent_err(leaf, slot, "invalid item size, have %u expect [%zu, %u)",
item_size, sizeof(*ei),
BTRFS_LEAF_DATA_SIZE(fs_info)); return -EUCLEAN;
}
end = item_size + btrfs_item_ptr_offset(leaf, slot);
/* Checks against extent_item */
ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
flags = btrfs_extent_flags(leaf, ei);
total_refs = btrfs_extent_refs(leaf, ei);
generation = btrfs_extent_generation(leaf, ei); if (unlikely(generation >
btrfs_super_generation(fs_info->super_copy) + 1)) {
extent_err(leaf, slot, "invalid generation, have %llu expect (0, %llu]",
generation,
btrfs_super_generation(fs_info->super_copy) + 1); return -EUCLEAN;
} if (unlikely(!has_single_bit_set(flags & (BTRFS_EXTENT_FLAG_DATA |
BTRFS_EXTENT_FLAG_TREE_BLOCK)))) {
extent_err(leaf, slot, "invalid extent flag, have 0x%llx expect 1 bit set in 0x%llx",
flags, BTRFS_EXTENT_FLAG_DATA |
BTRFS_EXTENT_FLAG_TREE_BLOCK); return -EUCLEAN;
}
is_tree_block = !!(flags & BTRFS_EXTENT_FLAG_TREE_BLOCK); if (is_tree_block) { if (unlikely(key->type == BTRFS_EXTENT_ITEM_KEY &&
key->offset != fs_info->nodesize)) {
extent_err(leaf, slot, "invalid extent length, have %llu expect %u",
key->offset, fs_info->nodesize); return -EUCLEAN;
}
} else { if (unlikely(key->type != BTRFS_EXTENT_ITEM_KEY)) {
extent_err(leaf, slot, "invalid key type, have %u expect %u for data backref",
key->type, BTRFS_EXTENT_ITEM_KEY); return -EUCLEAN;
} if (unlikely(!IS_ALIGNED(key->offset, fs_info->sectorsize))) {
extent_err(leaf, slot, "invalid extent length, have %llu expect aligned to %u",
key->offset, fs_info->sectorsize); return -EUCLEAN;
} if (unlikely(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)) {
extent_err(leaf, slot, "invalid extent flag, data has full backref set"); return -EUCLEAN;
}
}
ptr = (unsignedlong)(struct btrfs_extent_item *)(ei + 1);
/* Check the special case of btrfs_tree_block_info */ if (is_tree_block && key->type != BTRFS_METADATA_ITEM_KEY) { struct btrfs_tree_block_info *info;
info = (struct btrfs_tree_block_info *)ptr; if (unlikely(btrfs_tree_block_level(leaf, info) >= BTRFS_MAX_LEVEL)) {
extent_err(leaf, slot, "invalid tree block info level, have %u expect [0, %u]",
btrfs_tree_block_level(leaf, info),
BTRFS_MAX_LEVEL - 1); return -EUCLEAN;
}
ptr = (unsignedlong)(struct btrfs_tree_block_info *)(info + 1);
}
/* * NOTE: In theory we should record all found index numbers * to find any duplicated indexes, but that will be too time * consuming for inodes with too many hard links.
*/
ptr += sizeof(*iref) + namelen;
} return 0;
}
/* * Common point to switch the item-specific validation.
*/ staticenum btrfs_tree_block_status check_leaf_item(struct extent_buffer *leaf, struct btrfs_key *key, int slot, struct btrfs_key *prev_key)
{ int ret = 0; struct btrfs_chunk *chunk;
switch (key->type) { case BTRFS_EXTENT_DATA_KEY:
ret = check_extent_data_item(leaf, key, slot, prev_key); break; case BTRFS_EXTENT_CSUM_KEY:
ret = check_csum_item(leaf, key, slot, prev_key); break; case BTRFS_DIR_ITEM_KEY: case BTRFS_DIR_INDEX_KEY: case BTRFS_XATTR_ITEM_KEY:
ret = check_dir_item(leaf, key, prev_key, slot); break; case BTRFS_INODE_REF_KEY:
ret = check_inode_ref(leaf, key, prev_key, slot); break; case BTRFS_INODE_EXTREF_KEY:
ret = check_inode_extref(leaf, key, prev_key, slot); break; case BTRFS_BLOCK_GROUP_ITEM_KEY:
ret = check_block_group_item(leaf, key, slot); break; case BTRFS_CHUNK_ITEM_KEY:
chunk = btrfs_item_ptr(leaf, slot, struct btrfs_chunk);
ret = check_leaf_chunk_item(leaf, chunk, key, slot); break; case BTRFS_DEV_ITEM_KEY:
ret = check_dev_item(leaf, key, slot); break; case BTRFS_DEV_EXTENT_KEY:
ret = check_dev_extent_item(leaf, key, slot, prev_key); break; case BTRFS_INODE_ITEM_KEY:
ret = check_inode_item(leaf, key, slot); break; case BTRFS_ROOT_ITEM_KEY:
ret = check_root_item(leaf, key, slot); break; case BTRFS_EXTENT_ITEM_KEY: case BTRFS_METADATA_ITEM_KEY:
ret = check_extent_item(leaf, key, slot, prev_key); break; case BTRFS_TREE_BLOCK_REF_KEY: case BTRFS_SHARED_DATA_REF_KEY: case BTRFS_SHARED_BLOCK_REF_KEY:
ret = check_simple_keyed_refs(leaf, key, slot); break; case BTRFS_EXTENT_DATA_REF_KEY:
ret = check_extent_data_ref(leaf, key, slot); break; case BTRFS_RAID_STRIPE_KEY:
ret = check_raid_stripe_extent(leaf, key, slot); break;
}
if (unlikely(ret)) return BTRFS_TREE_BLOCK_INVALID_ITEM; return BTRFS_TREE_BLOCK_CLEAN;
}
enum btrfs_tree_block_status __btrfs_check_leaf(struct extent_buffer *leaf)
{ struct btrfs_fs_info *fs_info = leaf->fs_info; /* No valid key type is 0, so all key should be larger than this key */ struct btrfs_key prev_key = {0, 0, 0}; struct btrfs_key key;
u32 nritems = btrfs_header_nritems(leaf); int slot;
if (unlikely(btrfs_header_level(leaf) != 0)) {
generic_err(leaf, 0, "invalid level for leaf, have %d expect 0",
btrfs_header_level(leaf)); return BTRFS_TREE_BLOCK_INVALID_LEVEL;
}
if (unlikely(!btrfs_header_flag(leaf, BTRFS_HEADER_FLAG_WRITTEN))) {
generic_err(leaf, 0, "invalid flag for leaf, WRITTEN not set"); return BTRFS_TREE_BLOCK_WRITTEN_NOT_SET;
}
/* * Extent buffers from a relocation tree have a owner field that * corresponds to the subvolume tree they are based on. So just from an * extent buffer alone we can not find out what is the id of the * corresponding subvolume tree, so we can not figure out if the extent * buffer corresponds to the root of the relocation tree or not. So * skip this check for relocation trees.
*/ if (nritems == 0 && !btrfs_header_flag(leaf, BTRFS_HEADER_FLAG_RELOC)) {
u64 owner = btrfs_header_owner(leaf);
/* These trees must never be empty */ if (unlikely(owner == BTRFS_ROOT_TREE_OBJECTID ||
owner == BTRFS_CHUNK_TREE_OBJECTID ||
owner == BTRFS_DEV_TREE_OBJECTID ||
owner == BTRFS_FS_TREE_OBJECTID ||
owner == BTRFS_DATA_RELOC_TREE_OBJECTID)) {
generic_err(leaf, 0, "invalid root, root %llu must never be empty",
owner); return BTRFS_TREE_BLOCK_INVALID_NRITEMS;
}
/* Unknown tree */ if (unlikely(owner == 0)) {
generic_err(leaf, 0, "invalid owner, root 0 is not defined"); return BTRFS_TREE_BLOCK_INVALID_OWNER;
}
/* EXTENT_TREE_V2 can have empty extent trees. */ if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) return BTRFS_TREE_BLOCK_CLEAN;
if (unlikely(owner == BTRFS_EXTENT_TREE_OBJECTID)) {
generic_err(leaf, 0, "invalid root, root %llu must never be empty",
owner); return BTRFS_TREE_BLOCK_INVALID_NRITEMS;
}
return BTRFS_TREE_BLOCK_CLEAN;
}
if (unlikely(nritems == 0)) return BTRFS_TREE_BLOCK_CLEAN;
/* * Check the following things to make sure this is a good leaf, and * leaf users won't need to bother with similar sanity checks: * * 1) key ordering * 2) item offset and size * No overlap, no hole, all inside the leaf. * 3) item content * If possible, do comprehensive sanity check. * NOTE: All checks must only rely on the item data itself.
*/ for (slot = 0; slot < nritems; slot++) {
u32 item_end_expected;
u64 item_data_end; enum btrfs_tree_block_status ret;
btrfs_item_key_to_cpu(leaf, &key, slot);
/* Make sure the keys are in the right order */ if (unlikely(btrfs_comp_cpu_keys(&prev_key, &key) >= 0)) {
generic_err(leaf, slot, "bad key order, prev (%llu %u %llu) current (%llu %u %llu)",
prev_key.objectid, prev_key.type,
prev_key.offset, key.objectid, key.type,
key.offset); return BTRFS_TREE_BLOCK_BAD_KEY_ORDER;
}
item_data_end = (u64)btrfs_item_offset(leaf, slot) +
btrfs_item_size(leaf, slot); /* * Make sure the offset and ends are right, remember that the * item data starts at the end of the leaf and grows towards the * front.
*/ if (slot == 0)
item_end_expected = BTRFS_LEAF_DATA_SIZE(fs_info); else
item_end_expected = btrfs_item_offset(leaf,
slot - 1); if (unlikely(item_data_end != item_end_expected)) {
generic_err(leaf, slot, "unexpected item end, have %llu expect %u",
item_data_end, item_end_expected); return BTRFS_TREE_BLOCK_INVALID_OFFSETS;
}
/* * Check to make sure that we don't point outside of the leaf, * just in case all the items are consistent to each other, but
--> --------------------
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.
