/* * We're appending new directory block. Make sure the block is not * allocated yet, otherwise we will end up corrupting the * directory.
*/
err = ext4_map_blocks(NULL, inode, &map, 0); if (err < 0) return ERR_PTR(err); if (err) {
EXT4_ERROR_INODE(inode, "Logical block already allocated"); return ERR_PTR(-EFSCORRUPTED);
}
/* * Hints to ext4_read_dirblock regarding whether we expect a directory * block being read to be an index block, or a block containing * directory entries (and if the latter, whether it was found via a * logical block in an htree index block). This is used to control * what sort of sanity checkinig ext4_read_dirblock() will do on the * directory block read from the storage device. EITHER will means * the caller doesn't know what kind of directory block will be read, * so no specific verification will be done.
*/ typedefenum {
EITHER, INDEX, DIRENT, DIRENT_HTREE
} dirblock_type_t;
return bh;
} /* The first directory block must not be a hole. */ if (!bh && (type == INDEX || type == DIRENT_HTREE || block == 0)) {
ext4_error_inode(inode, func, line, block, "Directory hole found for htree %s block %u",
(type == INDEX) ? "index" : "leaf", block); return ERR_PTR(-EFSCORRUPTED);
} if (!bh) return NULL;
dirent = (struct ext4_dir_entry *) bh->b_data; /* Determine whether or not we have an index block */ if (is_dx(inode)) { if (block == 0)
is_dx_block = 1; elseif (ext4_rec_len_from_disk(dirent->rec_len,
inode->i_sb->s_blocksize) ==
inode->i_sb->s_blocksize)
is_dx_block = 1;
} if (!is_dx_block && type == INDEX) {
ext4_error_inode(inode, func, line, block, "directory leaf block found instead of index block");
brelse(bh); return ERR_PTR(-EFSCORRUPTED);
} if (!ext4_has_feature_metadata_csum(inode->i_sb) ||
buffer_verified(bh)) return bh;
/* * An empty leaf block can get mistaken for a index block; for * this reason, we can only check the index checksum when the * caller is sure it should be an index block.
*/ if (is_dx_block && type == INDEX) { if (ext4_dx_csum_verify(inode, dirent) &&
!ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_CRC))
set_buffer_verified(bh); else {
ext4_error_inode_err(inode, func, line, block,
EFSBADCRC, "Directory index failed checksum");
brelse(bh); return ERR_PTR(-EFSBADCRC);
}
} if (!is_dx_block) { if (ext4_dirblock_csum_verify(inode, bh) &&
!ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_CRC))
set_buffer_verified(bh); else {
ext4_error_inode_err(inode, func, line, block,
EFSBADCRC, "Directory block failed checksum");
brelse(bh); return ERR_PTR(-EFSBADCRC);
}
} return bh;
}
/* * dx_root_info is laid out so that if it should somehow get overlaid by a * dirent the two low bits of the hash version will be zero. Therefore, the * hash version mod 4 should never be 0. Sincerely, the paranoia department.
*/
/* Walk through a dirent block to find a checksum "dirent" at the tail */ staticstruct ext4_dir_entry_tail *get_dirent_tail(struct inode *inode, struct buffer_head *bh)
{ struct ext4_dir_entry_tail *t; int blocksize = EXT4_BLOCK_SIZE(inode->i_sb);
#ifdef PARANOID struct ext4_dir_entry *d, *top;
d = (struct ext4_dir_entry *)bh->b_data;
top = (struct ext4_dir_entry *)(bh->b_data +
(blocksize - sizeof(struct ext4_dir_entry_tail))); while (d < top && ext4_rec_len_from_disk(d->rec_len, blocksize))
d = (struct ext4_dir_entry *)(((void *)d) +
ext4_rec_len_from_disk(d->rec_len, blocksize));
if (d != top) return NULL;
t = (struct ext4_dir_entry_tail *)d; #else
t = EXT4_DIRENT_TAIL(bh->b_data, EXT4_BLOCK_SIZE(inode->i_sb)); #endif
/* * p is at least 6 bytes before the end of page
*/ staticinlinestruct ext4_dir_entry_2 *
ext4_next_entry(struct ext4_dir_entry_2 *p, unsignedlong blocksize)
{ return (struct ext4_dir_entry_2 *)((char *)p +
ext4_rec_len_from_disk(p->rec_len, blocksize));
}
/* * Future: use high four bits of block for coalesce-on-delete flags * Mask them off for now.
*/
staticstruct stats dx_show_leaf(struct inode *dir, struct dx_hash_info *hinfo, struct ext4_dir_entry_2 *de, int size, int show_names)
{ unsigned names = 0, space = 0; char *base = (char *) de; struct dx_hash_info h = *hinfo;
printk("names: "); while ((char *) de < base + size)
{ if (de->inode)
{ if (show_names)
{ #ifdef CONFIG_FS_ENCRYPTION int len; char *name; struct fscrypt_str fname_crypto_str =
FSTR_INIT(NULL, 0); int res = 0;
name = de->name;
len = de->name_len; if (!IS_ENCRYPTED(dir)) { /* Directory is not encrypted */
(void) ext4fs_dirhash(dir, de->name,
de->name_len, &h);
printk("%*.s:(U)%x.%u ", len,
name, h.hash,
(unsigned) ((char *) de
- base));
} else { struct fscrypt_str de_name =
FSTR_INIT(name, len);
/* Directory is encrypted */
res = fscrypt_fname_alloc_buffer(
len, &fname_crypto_str); if (res)
printk(KERN_WARNING "Error " "allocating crypto " "buffer--skipping " "crypto\n");
res = fscrypt_fname_disk_to_usr(dir,
0, 0, &de_name,
&fname_crypto_str); if (res) {
printk(KERN_WARNING "Error " "converting filename " "from disk to usr" "\n");
name = "??";
len = 2;
} else {
name = fname_crypto_str.name;
len = fname_crypto_str.len;
} if (IS_CASEFOLDED(dir))
h.hash = EXT4_DIRENT_HASH(de); else
(void) ext4fs_dirhash(dir,
de->name,
de->name_len, &h);
printk("%*.s:(E)%x.%u ", len, name,
h.hash, (unsigned) ((char *) de
- base));
fscrypt_fname_free_buffer(
&fname_crypto_str);
} #else int len = de->name_len; char *name = de->name;
(void) ext4fs_dirhash(dir, de->name,
de->name_len, &h);
printk("%*.s:%x.%u ", len, name, h.hash,
(unsigned) ((char *) de - base)); #endif
}
space += ext4_dir_rec_len(de->name_len, dir);
names++;
}
de = ext4_next_entry(de, size);
}
printk(KERN_CONT "(%i)\n", names); return (struct stats) { names, space, 1 };
}
/* * Probe for a directory leaf block to search. * * dx_probe can return ERR_BAD_DX_DIR, which means there was a format * error in the directory index, and the caller should fall back to * searching the directory normally. The callers of dx_probe **MUST** * check for this error code, and make sure it never gets reflected * back to userspace.
*/ staticstruct dx_frame *
dx_probe(struct ext4_filename *fname, struct inode *dir, struct dx_hash_info *hinfo, struct dx_frame *frame_in)
{ unsigned count, indirect, level, i; struct dx_entry *at, *entries, *p, *q, *m; struct dx_root *root; struct dx_frame *frame = frame_in; struct dx_frame *ret_err = ERR_PTR(ERR_BAD_DX_DIR);
u32 hash;
ext4_lblk_t block;
ext4_lblk_t blocks[EXT4_HTREE_LEVEL];
info = &((struct dx_root *)frames[0].bh->b_data)->info; /* save local copy, "info" may be freed after brelse() */
indirect_levels = info->indirect_levels; for (i = 0; i <= indirect_levels; i++) { if (frames[i].bh == NULL) break;
brelse(frames[i].bh);
frames[i].bh = NULL;
}
}
/* * This function increments the frame pointer to search the next leaf * block, and reads in the necessary intervening nodes if the search * should be necessary. Whether or not the search is necessary is * controlled by the hash parameter. If the hash value is even, then * the search is only continued if the next block starts with that * hash value. This is used if we are searching for a specific file. * * If the hash value is HASH_NB_ALWAYS, then always go to the next block. * * This function returns 1 if the caller should continue to search, * or 0 if it should not. If there is an error reading one of the * index blocks, it will a negative error code. * * If start_hash is non-null, it will be filled in with the starting * hash of the next page.
*/ staticint ext4_htree_next_block(struct inode *dir, __u32 hash, struct dx_frame *frame, struct dx_frame *frames,
__u32 *start_hash)
{ struct dx_frame *p; struct buffer_head *bh; int num_frames = 0;
__u32 bhash;
p = frame; /* * Find the next leaf page by incrementing the frame pointer. * If we run out of entries in the interior node, loop around and * increment pointer in the parent node. When we break out of * this loop, num_frames indicates the number of interior * nodes need to be read.
*/ while (1) { if (++(p->at) < p->entries + dx_get_count(p->entries)) break; if (p == frames) return 0;
num_frames++;
p--;
}
/* * If the hash is 1, then continue only if the next page has a * continuation hash of any value. This is used for readdir * handling. Otherwise, check to see if the hash matches the * desired continuation hash. If it doesn't, return since * there's no point to read in the successive index pages.
*/
bhash = dx_get_hash(p->at); if (start_hash)
*start_hash = bhash; if ((hash & 1) == 0) { if ((bhash & ~1) != hash) return 0;
} /* * If the hash is HASH_NB_ALWAYS, we always go to the next * block so no check is necessary
*/ while (num_frames--) {
bh = ext4_read_dirblock(dir, dx_get_block(p->at), INDEX); if (IS_ERR(bh)) return PTR_ERR(bh);
p++;
brelse(p->bh);
p->bh = bh;
p->at = p->entries = ((struct dx_node *) bh->b_data)->entries;
} return 1;
}
/* * This function fills a red-black tree with information from a * directory block. It returns the number directory entries loaded * into the tree. If there is an error it is returned in err.
*/ staticint htree_dirblock_to_tree(struct file *dir_file, struct inode *dir, ext4_lblk_t block, struct dx_hash_info *hinfo,
__u32 start_hash, __u32 start_minor_hash)
{ struct buffer_head *bh; struct ext4_dir_entry_2 *de, *top; int err = 0, count = 0; struct fscrypt_str fname_crypto_str = FSTR_INIT(NULL, 0), tmp_str; int csum = ext4_has_feature_metadata_csum(dir->i_sb);
dxtrace(printk(KERN_INFO "In htree dirblock_to_tree: block %lu\n",
(unsignedlong)block));
bh = ext4_read_dirblock(dir, block, DIRENT_HTREE); if (IS_ERR(bh)) return PTR_ERR(bh);
de = (struct ext4_dir_entry_2 *) bh->b_data; /* csum entries are not larger in the casefolded encrypted case */
top = (struct ext4_dir_entry_2 *) ((char *) de +
dir->i_sb->s_blocksize -
ext4_dir_rec_len(0,
csum ? NULL : dir)); /* Check if the directory is encrypted */ if (IS_ENCRYPTED(dir)) {
err = fscrypt_prepare_readdir(dir); if (err < 0) {
brelse(bh); return err;
}
err = fscrypt_fname_alloc_buffer(EXT4_NAME_LEN,
&fname_crypto_str); if (err < 0) {
brelse(bh); return err;
}
}
for (; de < top; de = ext4_next_entry(de, dir->i_sb->s_blocksize)) { if (ext4_check_dir_entry(dir, NULL, de, bh,
bh->b_data, bh->b_size,
(block<<EXT4_BLOCK_SIZE_BITS(dir->i_sb))
+ ((char *)de - bh->b_data))) { /* silently ignore the rest of the block */ break;
} if (ext4_hash_in_dirent(dir)) { if (de->name_len && de->inode) {
hinfo->hash = EXT4_DIRENT_HASH(de);
hinfo->minor_hash = EXT4_DIRENT_MINOR_HASH(de);
} else {
hinfo->hash = 0;
hinfo->minor_hash = 0;
}
} else {
err = ext4fs_dirhash(dir, de->name,
de->name_len, hinfo); if (err < 0) {
count = err; goto errout;
}
} if ((hinfo->hash < start_hash) ||
((hinfo->hash == start_hash) &&
(hinfo->minor_hash < start_minor_hash))) continue; if (de->inode == 0) continue; if (!IS_ENCRYPTED(dir)) {
tmp_str.name = de->name;
tmp_str.len = de->name_len;
err = ext4_htree_store_dirent(dir_file,
hinfo->hash, hinfo->minor_hash, de,
&tmp_str);
} else { int save_len = fname_crypto_str.len; struct fscrypt_str de_name = FSTR_INIT(de->name,
de->name_len);
/* * This function fills a red-black tree with information from a * directory. We start scanning the directory in hash order, starting * at start_hash and start_minor_hash. * * This function returns the number of entries inserted into the tree, * or a negative error code.
*/ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash,
__u32 start_minor_hash, __u32 *next_hash)
{ struct dx_hash_info hinfo; struct ext4_dir_entry_2 *de; struct dx_frame frames[EXT4_HTREE_LEVEL], *frame; struct inode *dir;
ext4_lblk_t block; int count = 0; int ret, err;
__u32 hashval; struct fscrypt_str tmp_str;
dxtrace(printk(KERN_DEBUG "In htree_fill_tree, start hash: %x:%x\n",
start_hash, start_minor_hash));
dir = file_inode(dir_file); if (!(ext4_test_inode_flag(dir, EXT4_INODE_INDEX))) { if (ext4_hash_in_dirent(dir))
hinfo.hash_version = DX_HASH_SIPHASH; else
hinfo.hash_version =
EXT4_SB(dir->i_sb)->s_def_hash_version; if (hinfo.hash_version <= DX_HASH_TEA)
hinfo.hash_version +=
EXT4_SB(dir->i_sb)->s_hash_unsigned;
hinfo.seed = EXT4_SB(dir->i_sb)->s_hash_seed; if (ext4_has_inline_data(dir)) { int has_inline_data = 1;
count = ext4_inlinedir_to_tree(dir_file, dir, 0,
&hinfo, start_hash,
start_minor_hash,
&has_inline_data); if (has_inline_data) {
*next_hash = ~0; return count;
}
}
count = htree_dirblock_to_tree(dir_file, dir, 0, &hinfo,
start_hash, start_minor_hash);
*next_hash = ~0; return count;
}
hinfo.hash = start_hash;
hinfo.minor_hash = 0;
frame = dx_probe(NULL, dir, &hinfo, frames); if (IS_ERR(frame)) return PTR_ERR(frame);
/* Add '.' and '..' from the htree header */ if (!start_hash && !start_minor_hash) {
de = (struct ext4_dir_entry_2 *) frames[0].bh->b_data;
tmp_str.name = de->name;
tmp_str.len = de->name_len;
err = ext4_htree_store_dirent(dir_file, 0, 0,
de, &tmp_str); if (err != 0) goto errout;
count++;
} if (start_hash < 2 || (start_hash ==2 && start_minor_hash==0)) {
de = (struct ext4_dir_entry_2 *) frames[0].bh->b_data;
de = ext4_next_entry(de, dir->i_sb->s_blocksize);
tmp_str.name = de->name;
tmp_str.len = de->name_len;
err = ext4_htree_store_dirent(dir_file, 2, 0,
de, &tmp_str); if (err != 0) goto errout;
count++;
}
while (1) { if (fatal_signal_pending(current)) {
err = -ERESTARTSYS; goto errout;
}
cond_resched();
block = dx_get_block(frame->at);
ret = htree_dirblock_to_tree(dir_file, dir, block, &hinfo,
start_hash, start_minor_hash); if (ret < 0) {
err = ret; goto errout;
}
count += ret;
hashval = ~0;
ret = ext4_htree_next_block(dir, HASH_NB_ALWAYS,
frame, frames, &hashval);
*next_hash = hashval; if (ret < 0) {
err = ret; goto errout;
} /* * Stop if: (a) there are no more entries, or * (b) we have inserted at least one entry and the * next hash value is not a continuation
*/ if ((ret == 0) ||
(count && ((hashval & 1) == 0))) break;
}
dx_release(frames);
dxtrace(printk(KERN_DEBUG "Fill tree: returned %d entries, " "next hash: %x\n", count, *next_hash)); return count;
errout:
dx_release(frames); return (err);
}
#if IS_ENABLED(CONFIG_UNICODE) if (IS_CASEFOLDED(parent) &&
(!IS_ENCRYPTED(parent) || fscrypt_has_encryption_key(parent))) { /* * Just checking IS_ENCRYPTED(parent) below is not * sufficient to decide whether one can use the hash for * skipping the string comparison, because the key might * have been added right after * ext4_fname_setup_ci_filename(). In this case, a hash * mismatch will be a false negative. Therefore, make * sure cf_name was properly initialized before * considering the calculated hash.
*/ if (sb_no_casefold_compat_fallback(parent->i_sb) &&
IS_ENCRYPTED(parent) && fname->cf_name.name &&
(fname->hinfo.hash != EXT4_DIRENT_HASH(de) ||
fname->hinfo.minor_hash != EXT4_DIRENT_MINOR_HASH(de))) returnfalse; /* * Treat comparison errors as not a match. The * only case where it happens is on a disk * corruption or ENOMEM.
*/
/* * Returns 0 if not found, -EFSCORRUPTED on failure, and 1 on success
*/ int ext4_search_dir(struct buffer_head *bh, char *search_buf, int buf_size, struct inode *dir, struct ext4_filename *fname, unsignedint offset, struct ext4_dir_entry_2 **res_dir)
{ struct ext4_dir_entry_2 * de; char * dlimit; int de_len;
de = (struct ext4_dir_entry_2 *)search_buf;
dlimit = search_buf + buf_size; while ((char *) de < dlimit - EXT4_BASE_DIR_LEN) { /* this code is executed quadratically often */ /* do minimal checking `by hand' */ if (de->name + de->name_len <= dlimit &&
ext4_match(dir, fname, de)) { /* found a match - just to be sure, do
* a full check */ if (ext4_check_dir_entry(dir, NULL, de, bh, search_buf,
buf_size, offset)) return -EFSCORRUPTED;
*res_dir = de; return 1;
} /* prevent looping on a bad block */
de_len = ext4_rec_len_from_disk(de->rec_len,
dir->i_sb->s_blocksize); if (de_len <= 0) return -EFSCORRUPTED;
offset += de_len;
de = (struct ext4_dir_entry_2 *) ((char *) de + de_len);
} return 0;
}
if (!is_dx(dir)) return 0; if (block == 0) return 1; if (de->inode == 0 &&
ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize) ==
sb->s_blocksize) return 1; return 0;
}
/* * __ext4_find_entry() * * finds an entry in the specified directory with the wanted name. It * returns the cache buffer in which the entry was found, and the entry * itself (as a parameter - res_dir). It does NOT read the inode of the * entry - you'll have to do that yourself if you want to. * * The returned buffer_head has ->b_count elevated. The caller is expected * to brelse() it when appropriate.
*/ staticstruct buffer_head *__ext4_find_entry(struct inode *dir, struct ext4_filename *fname, struct ext4_dir_entry_2 **res_dir, int *inlined)
{ struct super_block *sb; struct buffer_head *bh_use[NAMEI_RA_SIZE]; struct buffer_head *bh, *ret = NULL;
ext4_lblk_t start, block; const u8 *name = fname->usr_fname->name;
size_t ra_max = 0; /* Number of bh's in the readahead
buffer, bh_use[] */
size_t ra_ptr = 0; /* Current index into readahead
buffer */
ext4_lblk_t nblocks; int i, namelen, retval;
if (ext4_has_inline_data(dir)) { int has_inline_data = 1;
ret = ext4_find_inline_entry(dir, fname, res_dir,
&has_inline_data); if (inlined)
*inlined = has_inline_data; if (has_inline_data || IS_ERR(ret)) goto cleanup_and_exit;
}
if ((namelen <= 2) && (name[0] == '.') &&
(name[1] == '.' || name[1] == '\0')) { /* * "." or ".." will only be in the first block * NFS may look up ".."; "." should be handled by the VFS
*/
block = start = 0;
nblocks = 1; goto restart;
} if (is_dx(dir)) {
ret = ext4_dx_find_entry(dir, fname, res_dir); /* * On success, or if the error was file not found, * return. Otherwise, fall back to doing a search the * old fashioned way.
*/ if (IS_ERR(ret) && PTR_ERR(ret) == ERR_BAD_DX_DIR)
dxtrace(printk(KERN_DEBUG "ext4_find_entry: dx failed, " "falling back\n")); elseif (!sb_no_casefold_compat_fallback(dir->i_sb) &&
*res_dir == NULL && IS_CASEFOLDED(dir))
dxtrace(printk(KERN_DEBUG "ext4_find_entry: casefold " "failed, falling back\n")); else goto cleanup_and_exit;
ret = NULL;
}
nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); if (!nblocks) {
ret = NULL; goto cleanup_and_exit;
}
start = EXT4_I(dir)->i_dir_start_lookup; if (start >= nblocks)
start = 0;
block = start;
restart: do { /* * We deal with the read-ahead logic here.
*/
cond_resched(); if (ra_ptr >= ra_max) { /* Refill the readahead buffer */
ra_ptr = 0; if (block < start)
ra_max = start - block; else
ra_max = nblocks - block;
ra_max = min(ra_max, ARRAY_SIZE(bh_use));
retval = ext4_bread_batch(dir, block, ra_max, false/* wait */, bh_use); if (retval) {
ret = ERR_PTR(retval);
ra_max = 0; goto cleanup_and_exit;
}
} if ((bh = bh_use[ra_ptr++]) == NULL) goto next;
wait_on_buffer(bh); if (!buffer_uptodate(bh)) {
EXT4_ERROR_INODE_ERR(dir, EIO, "reading directory lblock %lu",
(unsignedlong) block);
brelse(bh);
ret = ERR_PTR(-EIO); goto cleanup_and_exit;
} if (!buffer_verified(bh) &&
!is_dx_internal_node(dir, block,
(struct ext4_dir_entry *)bh->b_data) &&
!ext4_dirblock_csum_verify(dir, bh)) {
EXT4_ERROR_INODE_ERR(dir, EFSBADCRC, "checksumming directory " "block %lu", (unsignedlong)block);
brelse(bh);
ret = ERR_PTR(-EFSBADCRC); goto cleanup_and_exit;
}
set_buffer_verified(bh);
i = search_dirblock(bh, dir, fname,
block << EXT4_BLOCK_SIZE_BITS(sb), res_dir); if (i == 1) {
EXT4_I(dir)->i_dir_start_lookup = block;
ret = bh; goto cleanup_and_exit;
} else {
brelse(bh); if (i < 0) {
ret = ERR_PTR(i); goto cleanup_and_exit;
}
}
next: if (++block >= nblocks)
block = 0;
} while (block != start);
/* * If the directory has grown while we were searching, then * search the last part of the directory before giving up.
*/
block = nblocks;
nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); if (block < nblocks) {
start = 0; goto restart;
}
cleanup_and_exit: /* Clean up the read-ahead blocks */ for (; ra_ptr < ra_max; ra_ptr++)
brelse(bh_use[ra_ptr]); return ret;
}
/* Check to see if we should continue to search */
retval = ext4_htree_next_block(dir, fname->hinfo.hash, frame,
frames, NULL); if (retval < 0) {
ext4_warning_inode(dir, "error %d reading directory index block",
retval);
bh = ERR_PTR(retval); goto errout;
}
} while (retval == 1);
if (IS_ENABLED(CONFIG_UNICODE) && !inode && IS_CASEFOLDED(dir)) { /* Eventually we want to call d_add_ci(dentry, NULL) * for negative dentries in the encoding case as * well. For now, prevent the negative dentry * from being cached.
*/ return NULL;
}
/* * Compact each dir entry in the range to the minimal rec_len. * Returns pointer to last entry in range.
*/ staticstruct ext4_dir_entry_2 *dx_pack_dirents(struct inode *dir, char *base, unsignedint blocksize)
{ struct ext4_dir_entry_2 *next, *to, *prev, *de = (struct ext4_dir_entry_2 *) base; unsigned rec_len = 0;
prev = to = de; while ((char*)de < base + blocksize) {
next = ext4_next_entry(de, blocksize); if (de->inode && de->name_len) {
rec_len = ext4_dir_rec_len(de->name_len, dir); if (de > to)
memmove(to, de, rec_len);
to->rec_len = ext4_rec_len_to_disk(rec_len, blocksize);
prev = to;
to = (struct ext4_dir_entry_2 *) (((char *) to) + rec_len);
}
de = next;
} return prev;
}
/* * Split a full leaf block to make room for a new dir entry. * Allocate a new block, and move entries so that they are approx. equally full. * Returns pointer to de in block into which the new entry will be inserted.
*/ staticstruct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, struct buffer_head **bh,struct dx_frame *frame, struct dx_hash_info *hinfo)
{ unsigned blocksize = dir->i_sb->s_blocksize; unsigned continued; int count; struct buffer_head *bh2;
ext4_lblk_t newblock;
u32 hash2; struct dx_map_entry *map; char *data1 = (*bh)->b_data, *data2; unsigned split, move, size; struct ext4_dir_entry_2 *de = NULL, *de2; int csum_size = 0; int err = 0, i;
if (ext4_has_feature_metadata_csum(dir->i_sb))
csum_size = sizeof(struct ext4_dir_entry_tail);
/* create map in the end of data2 block */
map = (struct dx_map_entry *) (data2 + blocksize);
count = dx_make_map(dir, *bh, hinfo, map); if (count < 0) {
err = count; goto journal_error;
}
map -= count;
dx_sort_map(map, count); /* Ensure that neither split block is over half full */
size = 0;
move = 0; for (i = count-1; i >= 0; i--) { /* is more than half of this entry in 2nd half of the block? */ if (size + map[i].size/2 > blocksize/2) break;
size += map[i].size;
move++;
} /* * map index at which we will split * * If the sum of active entries didn't exceed half the block size, just * split it in half by count; each resulting block will have at least * half the space free.
*/ if (i >= 0)
split = count - move; else
split = count/2;
if (WARN_ON_ONCE(split == 0)) { /* Should never happen, but avoid out-of-bounds access below */
ext4_error_inode_block(dir, (*bh)->b_blocknr, 0, "bad indexed directory? hash=%08x:%08x count=%d move=%u",
hinfo->hash, hinfo->minor_hash, count, move);
err = -EFSCORRUPTED; goto out;
}
/* * Add a new entry into a directory (leaf) block. If de is non-NULL, * it points to a directory entry which is guaranteed to be large * enough for new directory entry. If de is NULL, then * add_dirent_to_buf will attempt search the directory block for * space. It will return -ENOSPC if no space is available, and -EIO * and -EEXIST if directory entry already exists.
*/ staticint add_dirent_to_buf(handle_t *handle, struct ext4_filename *fname, struct inode *dir, struct inode *inode, struct ext4_dir_entry_2 *de, struct buffer_head *bh)
{ unsignedint blocksize = dir->i_sb->s_blocksize; int csum_size = 0; int err, err2;
if (ext4_has_feature_metadata_csum(inode->i_sb))
csum_size = sizeof(struct ext4_dir_entry_tail);
/* By now the buffer is marked for journaling */
ext4_insert_dentry(dir, inode, de, blocksize, fname);
/* * XXX shouldn't update any times until successful * completion of syscall, but too many callers depend * on this. * * XXX similarly, too many callers depend on * ext4_new_inode() setting the times, but error * recovery deletes the inode, so the worst that can * happen is that the times are slightly out of date * and/or different from the directory change time.
*/
inode_set_mtime_to_ts(dir, inode_set_ctime_current(dir));
ext4_update_dx_flag(dir);
inode_inc_iversion(dir);
err2 = ext4_mark_inode_dirty(handle, dir);
BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_dirblock(handle, dir, bh); if (err)
ext4_std_error(dir->i_sb, err); return err ? err : err2;
}
/* Initialize as for dx_probe */
fname->hinfo.hash_version = root->info.hash_version; if (fname->hinfo.hash_version <= DX_HASH_TEA)
fname->hinfo.hash_version += EXT4_SB(dir->i_sb)->s_hash_unsigned;
fname->hinfo.seed = EXT4_SB(dir->i_sb)->s_hash_seed;
/* casefolded encrypted hashes are computed on fname setup */ if (!ext4_hash_in_dirent(dir)) { int err = ext4fs_dirhash(dir, fname_name(fname),
fname_len(fname), &fname->hinfo); if (err < 0) {
brelse(bh2);
brelse(bh); return err;
}
}
memset(frames, 0, sizeof(frames));
frame = frames;
frame->entries = entries;
frame->at = entries;
frame->bh = bh;
retval = ext4_handle_dirty_dx_node(handle, dir, frame->bh); if (retval) goto out_frames;
retval = ext4_handle_dirty_dirblock(handle, dir, bh2); if (retval) goto out_frames;
de = do_split(handle,dir, &bh2, frame, &fname->hinfo); if (IS_ERR(de)) {
retval = PTR_ERR(de); goto out_frames;
}
retval = add_dirent_to_buf(handle, fname, dir, inode, de, bh2);
out_frames: /* * Even if the block split failed, we have to properly write * out all the changes we did so far. Otherwise we can end up * with corrupted filesystem.
*/ if (retval)
ext4_mark_inode_dirty(handle, dir);
dx_release(frames);
brelse(bh2); return retval;
}
/* * ext4_add_entry() * * adds a file entry to the specified directory, using the same * semantics as ext4_find_entry(). It returns NULL if it failed. * * NOTE!! The inode part of 'de' is left at 0 - which means you * may not sleep between calling this and putting something into * the entry, as someone else might have used it while you slept.
*/ staticint ext4_add_entry(handle_t *handle, struct dentry *dentry, struct inode *inode)
{ struct inode *dir = d_inode(dentry->d_parent); struct buffer_head *bh = NULL; struct ext4_dir_entry_2 *de; struct super_block *sb; struct ext4_filename fname; int retval; int dx_fallback=0; unsigned blocksize;
ext4_lblk_t block, blocks; int csum_size = 0;
if (ext4_has_feature_metadata_csum(inode->i_sb))
csum_size = sizeof(struct ext4_dir_entry_tail);
sb = dir->i_sb;
blocksize = sb->s_blocksize;
if (fscrypt_is_nokey_name(dentry)) return -ENOKEY;
if (!generic_ci_validate_strict_name(dir, &dentry->d_name)) return -EINVAL;
retval = ext4_fname_setup_filename(dir, &dentry->d_name, 0, &fname); if (retval) return retval;
if (ext4_has_inline_data(dir)) {
retval = ext4_try_add_inline_entry(handle, &fname, dir, inode); if (retval < 0) goto out; if (retval == 1) {
retval = 0; goto out;
}
}
if (is_dx(dir)) {
retval = ext4_dx_add_entry(handle, &fname, dir, inode); if (!retval || (retval != ERR_BAD_DX_DIR)) goto out; /* Can we just ignore htree data? */ if (ext4_has_feature_metadata_csum(sb)) {
EXT4_ERROR_INODE(dir, "Directory has corrupted htree index.");
retval = -EFSCORRUPTED; goto out;
}
ext4_clear_inode_flag(dir, EXT4_INODE_INDEX);
dx_fallback++;
retval = ext4_mark_inode_dirty(handle, dir); if (unlikely(retval)) goto out;
}
blocks = dir->i_size >> sb->s_blocksize_bits; for (block = 0; block < blocks; block++) {
bh = ext4_read_dirblock(dir, block, DIRENT); if (bh == NULL) {
bh = ext4_bread(handle, dir, block,
EXT4_GET_BLOCKS_CREATE); goto add_to_new_block;
} if (IS_ERR(bh)) {
retval = PTR_ERR(bh);
bh = NULL; goto out;
}
retval = add_dirent_to_buf(handle, &fname, dir, inode,
NULL, bh); if (retval != -ENOSPC) goto out;
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.