linux/fs/f2fs/namei.c
Linus Torvalds 81d8e5e213 f2fs-for-6.15-rc1
In this round, there are three major updates: 1) folio conversion, 2) refactor
 for mount API conversion, 3) some performance improvement such as direct IO,
 checkpoint speed, and IO priority hints. For stability, there are patches
 which add more sanity checks and fixes some major issues like i_size in
 atomic write operations and write pointer recovery in zoned devices.
 
 Enhancement:
  - huge folio converion work by Matthew Wilcox
  - clean up for mount API conversion by Eric Sandeen
  - improve direct IO speed in the overwrite case
  - add some sanity check on node consistency
  - set highest IO priority for checkpoint thread
  - keep POSIX_FADV_NOREUSE ranges and add sysfs entry to reclaim pages
  - add ioctl to get IO priority hint
  - add carve_out sysfs node for fsstat
 
 Bug fix:
  - disable nat_bits during umount to avoid potential nat entry corruption
  - fix missing i_size update on atomic writes
  - fix missing discard for active segments
  - fix running out of free segments
  - fix out-of-bounds access in f2fs_truncate_inode_blocks()
  - call f2fs_recover_quota_end() correctly
  - fix potential deadloop in prepare_compress_overwrite()
  - fix the missing write pointer correction for zoned device
  - fix to avoid panic once fallocation fails for pinfile
  - don't retry IO for corrupted data scenario
 
 There are many other clean up patches and minor bug fixes as usual.
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCgAdFiEE00UqedjCtOrGVvQiQBSofoJIUNIFAmfhjuYACgkQQBSofoJI
 UNJF+hAAip0Sf6bXwt43KR3lmbXg/YBTtPDACOs375Pn8pfRVsAPIAf5e1AnlBET
 rA0KpqUrEZHXenQFF9n4LIoHYqfuUw3EMempuvp4qgQcx15Kaajw+EGWVLYstNy2
 9dELc9DA55f/i1uvHezGfQFy6hMfXasf+tkaYk0z0ZYDStYboLgkVAY+F869q1Dl
 D16Y7Lna12h4eCxDdssIPDsLjFH/2LDn7SmhXsnpZxwK0Zx8JSo83WxXHnzXHxz4
 vUkukInKgqcBDvf6ufW/YF3/tqSs20XEXNK3cI1vyHx1dwij6Us+G0n0WJHL30QE
 zsliecR0X28JP1rJ3ldCjr+4Kxm6/u/Uwpinm2Fm1jL67UY4TZcaARBE8k20I6ND
 j/L1+sXrIdZ1aILM9/bwCjgXiVFdZbvlfGfpTj1duAkQgRd+/s9cjPlo5j5HTad5
 XJmzJz6YOaUNarhP/E31Z9SV9M2kEcmoDOTxKBg6ZcMWXZ27B6Z0ag92prd0GWk6
 rWDuVj0eP/LDY1QedbHRPbg1D84jVgcnldfPaf9ptln993skJGdgS0dkegTqMR0L
 H8RgpWOzWZI53gnQdePdej8diHmD8uRTrLf/oABm//GzTHC5BdwVpArOl1DCuLFF
 YkMtVEicgnWRmL3PgODAdTJXaDi3uGvT116i+lfMlUn9p+t93r0=
 =E+TE
 -----END PGP SIGNATURE-----

Merge tag 'f2fs-for-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "In this round, there are three major updates: (1) folio conversion,
  (2) refactoring for mount API conversion, (3) some performance
  improvement such as direct IO, checkpoint speed, and IO priority
  hints.

  For stability, there are patches which add more sanity checks and
  fixes some major issues like i_size in atomic write operations and
  write pointer recovery in zoned devices.

  Enhancements:
   - huge folio converion work by Matthew Wilcox
   - clean up for mount API conversion by Eric Sandeen
   - improve direct IO speed in the overwrite case
   - add some sanity check on node consistency
   - set highest IO priority for checkpoint thread
   - keep POSIX_FADV_NOREUSE ranges and add sysfs entry to reclaim pages
   - add ioctl to get IO priority hint
   - add carve_out sysfs node for fsstat

  Bug fixes:
   - disable nat_bits during umount to avoid potential nat entry corruption
   - fix missing i_size update on atomic writes
   - fix missing discard for active segments
   - fix running out of free segments
   - fix out-of-bounds access in f2fs_truncate_inode_blocks()
   - call f2fs_recover_quota_end() correctly
   - fix potential deadloop in prepare_compress_overwrite()
   - fix the missing write pointer correction for zoned device
   - fix to avoid panic once fallocation fails for pinfile
   - don't retry IO for corrupted data scenario

  There are many other clean up patches and minor bug fixes as usual"

* tag 'f2fs-for-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (68 commits)
  f2fs: fix missing discard for active segments
  f2fs: optimize f2fs DIO overwrites
  f2fs: fix to avoid atomicity corruption of atomic file
  f2fs: pass sbi rather than sb to parse_options()
  f2fs: pass sbi rather than sb to quota qf_name helpers
  f2fs: defer readonly check vs norecovery
  f2fs: Pass sbi rather than sb to f2fs_set_test_dummy_encryption
  f2fs: make LAZYTIME a mount option flag
  f2fs: make INLINECRYPT a mount option flag
  f2fs: factor out an f2fs_default_check function
  f2fs: consolidate unsupported option handling errors
  f2fs: use f2fs_sb_has_device_alias during option parsing
  f2fs: add carve_out sysfs node
  f2fs: fix to avoid running out of free segments
  f2fs: Remove f2fs_write_node_page()
  f2fs: Remove f2fs_write_meta_page()
  f2fs: Remove f2fs_write_data_page()
  f2fs: Remove check for ->writepage
  Revert "f2fs: rebuild nat_bits during umount"
  f2fs: fix to avoid accessing uninitialized curseg
  ...
2025-03-27 12:55:54 -07:00

1360 lines
32 KiB
C

// SPDX-License-Identifier: GPL-2.0
/*
* fs/f2fs/namei.c
*
* Copyright (c) 2012 Samsung Electronics Co., Ltd.
* http://www.samsung.com/
*/
#include <linux/fs.h>
#include <linux/f2fs_fs.h>
#include <linux/pagemap.h>
#include <linux/sched.h>
#include <linux/ctype.h>
#include <linux/random.h>
#include <linux/dcache.h>
#include <linux/namei.h>
#include <linux/quotaops.h>
#include "f2fs.h"
#include "node.h"
#include "segment.h"
#include "xattr.h"
#include "acl.h"
#include <trace/events/f2fs.h>
static inline bool is_extension_exist(const unsigned char *s, const char *sub,
bool tmp_ext, bool tmp_dot)
{
size_t slen = strlen(s);
size_t sublen = strlen(sub);
int i;
if (sublen == 1 && *sub == '*')
return true;
/*
* filename format of multimedia file should be defined as:
* "filename + '.' + extension + (optional: '.' + temp extension)".
*/
if (slen < sublen + 2)
return false;
if (!tmp_ext) {
/* file has no temp extension */
if (s[slen - sublen - 1] != '.')
return false;
return !strncasecmp(s + slen - sublen, sub, sublen);
}
for (i = 1; i < slen - sublen; i++) {
if (s[i] != '.')
continue;
if (!strncasecmp(s + i + 1, sub, sublen)) {
if (!tmp_dot)
return true;
if (i == slen - sublen - 1 || s[i + 1 + sublen] == '.')
return true;
}
}
return false;
}
static inline bool is_temperature_extension(const unsigned char *s, const char *sub)
{
return is_extension_exist(s, sub, true, false);
}
static inline bool is_compress_extension(const unsigned char *s, const char *sub)
{
return is_extension_exist(s, sub, true, true);
}
int f2fs_update_extension_list(struct f2fs_sb_info *sbi, const char *name,
bool hot, bool set)
{
__u8 (*extlist)[F2FS_EXTENSION_LEN] = sbi->raw_super->extension_list;
int cold_count = le32_to_cpu(sbi->raw_super->extension_count);
int hot_count = sbi->raw_super->hot_ext_count;
int total_count = cold_count + hot_count;
int start, count;
int i;
if (set) {
if (total_count == F2FS_MAX_EXTENSION)
return -EINVAL;
} else {
if (!hot && !cold_count)
return -EINVAL;
if (hot && !hot_count)
return -EINVAL;
}
if (hot) {
start = cold_count;
count = total_count;
} else {
start = 0;
count = cold_count;
}
for (i = start; i < count; i++) {
if (strcmp(name, extlist[i]))
continue;
if (set)
return -EINVAL;
memcpy(extlist[i], extlist[i + 1],
F2FS_EXTENSION_LEN * (total_count - i - 1));
memset(extlist[total_count - 1], 0, F2FS_EXTENSION_LEN);
if (hot)
sbi->raw_super->hot_ext_count = hot_count - 1;
else
sbi->raw_super->extension_count =
cpu_to_le32(cold_count - 1);
return 0;
}
if (!set)
return -EINVAL;
if (hot) {
memcpy(extlist[count], name, strlen(name));
sbi->raw_super->hot_ext_count = hot_count + 1;
} else {
char buf[F2FS_MAX_EXTENSION][F2FS_EXTENSION_LEN];
memcpy(buf, &extlist[cold_count],
F2FS_EXTENSION_LEN * hot_count);
memset(extlist[cold_count], 0, F2FS_EXTENSION_LEN);
memcpy(extlist[cold_count], name, strlen(name));
memcpy(&extlist[cold_count + 1], buf,
F2FS_EXTENSION_LEN * hot_count);
sbi->raw_super->extension_count = cpu_to_le32(cold_count + 1);
}
return 0;
}
static void set_compress_new_inode(struct f2fs_sb_info *sbi, struct inode *dir,
struct inode *inode, const unsigned char *name)
{
__u8 (*extlist)[F2FS_EXTENSION_LEN] = sbi->raw_super->extension_list;
unsigned char (*noext)[F2FS_EXTENSION_LEN] =
F2FS_OPTION(sbi).noextensions;
unsigned char (*ext)[F2FS_EXTENSION_LEN] = F2FS_OPTION(sbi).extensions;
unsigned char ext_cnt = F2FS_OPTION(sbi).compress_ext_cnt;
unsigned char noext_cnt = F2FS_OPTION(sbi).nocompress_ext_cnt;
int i, cold_count, hot_count;
if (!f2fs_sb_has_compression(sbi))
return;
if (S_ISDIR(inode->i_mode))
goto inherit_comp;
/* This name comes only from normal files. */
if (!name)
return;
/* Don't compress hot files. */
f2fs_down_read(&sbi->sb_lock);
cold_count = le32_to_cpu(sbi->raw_super->extension_count);
hot_count = sbi->raw_super->hot_ext_count;
for (i = cold_count; i < cold_count + hot_count; i++)
if (is_temperature_extension(name, extlist[i]))
break;
f2fs_up_read(&sbi->sb_lock);
if (i < (cold_count + hot_count))
return;
/* Don't compress unallowed extension. */
for (i = 0; i < noext_cnt; i++)
if (is_compress_extension(name, noext[i]))
return;
/* Compress wanting extension. */
for (i = 0; i < ext_cnt; i++) {
if (is_compress_extension(name, ext[i])) {
set_compress_context(inode);
return;
}
}
inherit_comp:
/* Inherit the {no-}compression flag in directory */
if (F2FS_I(dir)->i_flags & F2FS_NOCOMP_FL) {
F2FS_I(inode)->i_flags |= F2FS_NOCOMP_FL;
f2fs_mark_inode_dirty_sync(inode, true);
} else if (F2FS_I(dir)->i_flags & F2FS_COMPR_FL) {
set_compress_context(inode);
}
}
/*
* Set file's temperature for hot/cold data separation
*/
static void set_file_temperature(struct f2fs_sb_info *sbi, struct inode *inode,
const unsigned char *name)
{
__u8 (*extlist)[F2FS_EXTENSION_LEN] = sbi->raw_super->extension_list;
int i, cold_count, hot_count;
f2fs_down_read(&sbi->sb_lock);
cold_count = le32_to_cpu(sbi->raw_super->extension_count);
hot_count = sbi->raw_super->hot_ext_count;
for (i = 0; i < cold_count + hot_count; i++)
if (is_temperature_extension(name, extlist[i]))
break;
f2fs_up_read(&sbi->sb_lock);
if (i == cold_count + hot_count)
return;
if (i < cold_count)
file_set_cold(inode);
else
file_set_hot(inode);
}
static struct inode *f2fs_new_inode(struct mnt_idmap *idmap,
struct inode *dir, umode_t mode,
const char *name)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct f2fs_inode_info *fi;
nid_t ino;
struct inode *inode;
bool nid_free = false;
bool encrypt = false;
int xattr_size = 0;
int err;
inode = new_inode(dir->i_sb);
if (!inode)
return ERR_PTR(-ENOMEM);
if (!f2fs_alloc_nid(sbi, &ino)) {
err = -ENOSPC;
goto fail;
}
nid_free = true;
inode_init_owner(idmap, inode, dir, mode);
fi = F2FS_I(inode);
inode->i_ino = ino;
inode->i_blocks = 0;
simple_inode_init_ts(inode);
fi->i_crtime = inode_get_mtime(inode);
inode->i_generation = get_random_u32();
if (S_ISDIR(inode->i_mode))
fi->i_current_depth = 1;
err = insert_inode_locked(inode);
if (err) {
err = -EINVAL;
goto fail;
}
if (f2fs_sb_has_project_quota(sbi) &&
(F2FS_I(dir)->i_flags & F2FS_PROJINHERIT_FL))
fi->i_projid = F2FS_I(dir)->i_projid;
else
fi->i_projid = make_kprojid(&init_user_ns,
F2FS_DEF_PROJID);
err = fscrypt_prepare_new_inode(dir, inode, &encrypt);
if (err)
goto fail_drop;
err = f2fs_dquot_initialize(inode);
if (err)
goto fail_drop;
set_inode_flag(inode, FI_NEW_INODE);
if (encrypt)
f2fs_set_encrypted_inode(inode);
if (f2fs_sb_has_extra_attr(sbi)) {
set_inode_flag(inode, FI_EXTRA_ATTR);
fi->i_extra_isize = F2FS_TOTAL_EXTRA_ATTR_SIZE;
}
if (test_opt(sbi, INLINE_XATTR))
set_inode_flag(inode, FI_INLINE_XATTR);
if (f2fs_may_inline_dentry(inode))
set_inode_flag(inode, FI_INLINE_DENTRY);
if (f2fs_sb_has_flexible_inline_xattr(sbi)) {
f2fs_bug_on(sbi, !f2fs_has_extra_attr(inode));
if (f2fs_has_inline_xattr(inode))
xattr_size = F2FS_OPTION(sbi).inline_xattr_size;
/* Otherwise, will be 0 */
} else if (f2fs_has_inline_xattr(inode) ||
f2fs_has_inline_dentry(inode)) {
xattr_size = DEFAULT_INLINE_XATTR_ADDRS;
}
fi->i_inline_xattr_size = xattr_size;
fi->i_flags =
f2fs_mask_flags(mode, F2FS_I(dir)->i_flags & F2FS_FL_INHERITED);
if (S_ISDIR(inode->i_mode))
fi->i_flags |= F2FS_INDEX_FL;
if (fi->i_flags & F2FS_PROJINHERIT_FL)
set_inode_flag(inode, FI_PROJ_INHERIT);
/* Check compression first. */
set_compress_new_inode(sbi, dir, inode, name);
/* Should enable inline_data after compression set */
if (test_opt(sbi, INLINE_DATA) && f2fs_may_inline_data(inode))
set_inode_flag(inode, FI_INLINE_DATA);
if (name && !test_opt(sbi, DISABLE_EXT_IDENTIFY))
set_file_temperature(sbi, inode, name);
stat_inc_inline_xattr(inode);
stat_inc_inline_inode(inode);
stat_inc_inline_dir(inode);
f2fs_set_inode_flags(inode);
f2fs_init_extent_tree(inode);
trace_f2fs_new_inode(inode, 0);
return inode;
fail:
trace_f2fs_new_inode(inode, err);
make_bad_inode(inode);
if (nid_free)
set_inode_flag(inode, FI_FREE_NID);
iput(inode);
return ERR_PTR(err);
fail_drop:
trace_f2fs_new_inode(inode, err);
dquot_drop(inode);
inode->i_flags |= S_NOQUOTA;
make_bad_inode(inode);
if (nid_free)
set_inode_flag(inode, FI_FREE_NID);
clear_nlink(inode);
unlock_new_inode(inode);
iput(inode);
return ERR_PTR(err);
}
static int f2fs_create(struct mnt_idmap *idmap, struct inode *dir,
struct dentry *dentry, umode_t mode, bool excl)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
nid_t ino = 0;
int err;
if (unlikely(f2fs_cp_error(sbi)))
return -EIO;
if (!f2fs_is_checkpoint_ready(sbi))
return -ENOSPC;
err = f2fs_dquot_initialize(dir);
if (err)
return err;
inode = f2fs_new_inode(idmap, dir, mode, dentry->d_name.name);
if (IS_ERR(inode))
return PTR_ERR(inode);
inode->i_op = &f2fs_file_inode_operations;
inode->i_fop = &f2fs_file_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
ino = inode->i_ino;
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out;
f2fs_unlock_op(sbi);
f2fs_alloc_nid_done(sbi, ino);
d_instantiate_new(dentry, inode);
if (IS_DIRSYNC(dir))
f2fs_sync_fs(sbi->sb, 1);
f2fs_balance_fs(sbi, true);
return 0;
out:
f2fs_handle_failed_inode(inode);
return err;
}
static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
struct dentry *dentry)
{
struct inode *inode = d_inode(old_dentry);
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
int err;
if (unlikely(f2fs_cp_error(sbi)))
return -EIO;
if (!f2fs_is_checkpoint_ready(sbi))
return -ENOSPC;
err = fscrypt_prepare_link(old_dentry, dir, dentry);
if (err)
return err;
if (is_inode_flag_set(dir, FI_PROJ_INHERIT) &&
(!projid_eq(F2FS_I(dir)->i_projid,
F2FS_I(old_dentry->d_inode)->i_projid)))
return -EXDEV;
err = f2fs_dquot_initialize(dir);
if (err)
return err;
f2fs_balance_fs(sbi, true);
inode_set_ctime_current(inode);
ihold(inode);
set_inode_flag(inode, FI_INC_LINK);
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out;
f2fs_unlock_op(sbi);
d_instantiate(dentry, inode);
if (IS_DIRSYNC(dir))
f2fs_sync_fs(sbi->sb, 1);
return 0;
out:
clear_inode_flag(inode, FI_INC_LINK);
iput(inode);
f2fs_unlock_op(sbi);
return err;
}
struct dentry *f2fs_get_parent(struct dentry *child)
{
struct page *page;
unsigned long ino = f2fs_inode_by_name(d_inode(child), &dotdot_name, &page);
if (!ino) {
if (IS_ERR(page))
return ERR_CAST(page);
return ERR_PTR(-ENOENT);
}
return d_obtain_alias(f2fs_iget(child->d_sb, ino));
}
static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry,
unsigned int flags)
{
struct inode *inode = NULL;
struct f2fs_dir_entry *de;
struct page *page;
struct dentry *new;
nid_t ino = -1;
int err = 0;
struct f2fs_filename fname;
trace_f2fs_lookup_start(dir, dentry, flags);
if (dentry->d_name.len > F2FS_NAME_LEN) {
err = -ENAMETOOLONG;
goto out;
}
err = f2fs_prepare_lookup(dir, dentry, &fname);
if (err == -ENOENT)
goto out_splice;
if (err)
goto out;
de = __f2fs_find_entry(dir, &fname, &page);
f2fs_free_filename(&fname);
if (!de) {
if (IS_ERR(page)) {
err = PTR_ERR(page);
goto out;
}
err = -ENOENT;
goto out_splice;
}
ino = le32_to_cpu(de->ino);
f2fs_put_page(page, 0);
inode = f2fs_iget(dir->i_sb, ino);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto out;
}
if (inode->i_nlink == 0) {
f2fs_warn(F2FS_I_SB(inode), "%s: inode (ino=%lx) has zero i_nlink",
__func__, inode->i_ino);
err = -EFSCORRUPTED;
set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK);
goto out_iput;
}
if (IS_ENCRYPTED(dir) &&
(S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) &&
!fscrypt_has_permitted_context(dir, inode)) {
f2fs_warn(F2FS_I_SB(inode), "Inconsistent encryption contexts: %lu/%lu",
dir->i_ino, inode->i_ino);
err = -EPERM;
goto out_iput;
}
out_splice:
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.
*/
trace_f2fs_lookup_end(dir, dentry, ino, err);
return NULL;
}
new = d_splice_alias(inode, dentry);
trace_f2fs_lookup_end(dir, !IS_ERR_OR_NULL(new) ? new : dentry,
ino, IS_ERR(new) ? PTR_ERR(new) : err);
return new;
out_iput:
iput(inode);
out:
trace_f2fs_lookup_end(dir, dentry, ino, err);
return ERR_PTR(err);
}
static int f2fs_unlink(struct inode *dir, struct dentry *dentry)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode = d_inode(dentry);
struct f2fs_dir_entry *de;
struct page *page;
int err;
trace_f2fs_unlink_enter(dir, dentry);
if (unlikely(f2fs_cp_error(sbi))) {
err = -EIO;
goto fail;
}
err = f2fs_dquot_initialize(dir);
if (err)
goto fail;
err = f2fs_dquot_initialize(inode);
if (err)
goto fail;
de = f2fs_find_entry(dir, &dentry->d_name, &page);
if (!de) {
if (IS_ERR(page))
err = PTR_ERR(page);
goto fail;
}
f2fs_balance_fs(sbi, true);
f2fs_lock_op(sbi);
err = f2fs_acquire_orphan_inode(sbi);
if (err) {
f2fs_unlock_op(sbi);
f2fs_put_page(page, 0);
goto fail;
}
f2fs_delete_entry(de, page, dir, inode);
f2fs_unlock_op(sbi);
/* VFS negative dentries are incompatible with Encoding and
* Case-insensitiveness. Eventually we'll want avoid
* invalidating the dentries here, alongside with returning the
* negative dentries at f2fs_lookup(), when it is better
* supported by the VFS for the CI case.
*/
if (IS_ENABLED(CONFIG_UNICODE) && IS_CASEFOLDED(dir))
d_invalidate(dentry);
if (IS_DIRSYNC(dir))
f2fs_sync_fs(sbi->sb, 1);
fail:
trace_f2fs_unlink_exit(inode, err);
return err;
}
static const char *f2fs_get_link(struct dentry *dentry,
struct inode *inode,
struct delayed_call *done)
{
const char *link = page_get_link(dentry, inode, done);
if (!IS_ERR(link) && !*link) {
/* this is broken symlink case */
do_delayed_call(done);
clear_delayed_call(done);
link = ERR_PTR(-ENOENT);
}
return link;
}
static int f2fs_symlink(struct mnt_idmap *idmap, struct inode *dir,
struct dentry *dentry, const char *symname)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
size_t len = strlen(symname);
struct fscrypt_str disk_link;
int err;
if (unlikely(f2fs_cp_error(sbi)))
return -EIO;
if (!f2fs_is_checkpoint_ready(sbi))
return -ENOSPC;
err = fscrypt_prepare_symlink(dir, symname, len, dir->i_sb->s_blocksize,
&disk_link);
if (err)
return err;
err = f2fs_dquot_initialize(dir);
if (err)
return err;
inode = f2fs_new_inode(idmap, dir, S_IFLNK | S_IRWXUGO, NULL);
if (IS_ERR(inode))
return PTR_ERR(inode);
if (IS_ENCRYPTED(inode))
inode->i_op = &f2fs_encrypted_symlink_inode_operations;
else
inode->i_op = &f2fs_symlink_inode_operations;
inode_nohighmem(inode);
inode->i_mapping->a_ops = &f2fs_dblock_aops;
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out_f2fs_handle_failed_inode;
f2fs_unlock_op(sbi);
f2fs_alloc_nid_done(sbi, inode->i_ino);
err = fscrypt_encrypt_symlink(inode, symname, len, &disk_link);
if (err)
goto err_out;
err = page_symlink(inode, disk_link.name, disk_link.len);
err_out:
d_instantiate_new(dentry, inode);
/*
* Let's flush symlink data in order to avoid broken symlink as much as
* possible. Nevertheless, fsyncing is the best way, but there is no
* way to get a file descriptor in order to flush that.
*
* Note that, it needs to do dir->fsync to make this recoverable.
* If the symlink path is stored into inline_data, there is no
* performance regression.
*/
if (!err) {
filemap_write_and_wait_range(inode->i_mapping, 0,
disk_link.len - 1);
if (IS_DIRSYNC(dir))
f2fs_sync_fs(sbi->sb, 1);
} else {
f2fs_unlink(dir, dentry);
}
f2fs_balance_fs(sbi, true);
goto out_free_encrypted_link;
out_f2fs_handle_failed_inode:
f2fs_handle_failed_inode(inode);
out_free_encrypted_link:
if (disk_link.name != (unsigned char *)symname)
kfree(disk_link.name);
return err;
}
static struct dentry *f2fs_mkdir(struct mnt_idmap *idmap, struct inode *dir,
struct dentry *dentry, umode_t mode)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
int err;
if (unlikely(f2fs_cp_error(sbi)))
return ERR_PTR(-EIO);
err = f2fs_dquot_initialize(dir);
if (err)
return ERR_PTR(err);
inode = f2fs_new_inode(idmap, dir, S_IFDIR | mode, NULL);
if (IS_ERR(inode))
return ERR_CAST(inode);
inode->i_op = &f2fs_dir_inode_operations;
inode->i_fop = &f2fs_dir_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);
set_inode_flag(inode, FI_INC_LINK);
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out_fail;
f2fs_unlock_op(sbi);
f2fs_alloc_nid_done(sbi, inode->i_ino);
d_instantiate_new(dentry, inode);
if (IS_DIRSYNC(dir))
f2fs_sync_fs(sbi->sb, 1);
f2fs_balance_fs(sbi, true);
return NULL;
out_fail:
clear_inode_flag(inode, FI_INC_LINK);
f2fs_handle_failed_inode(inode);
return ERR_PTR(err);
}
static int f2fs_rmdir(struct inode *dir, struct dentry *dentry)
{
struct inode *inode = d_inode(dentry);
if (f2fs_empty_dir(inode))
return f2fs_unlink(dir, dentry);
return -ENOTEMPTY;
}
static int f2fs_mknod(struct mnt_idmap *idmap, struct inode *dir,
struct dentry *dentry, umode_t mode, dev_t rdev)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
int err = 0;
if (unlikely(f2fs_cp_error(sbi)))
return -EIO;
if (!f2fs_is_checkpoint_ready(sbi))
return -ENOSPC;
err = f2fs_dquot_initialize(dir);
if (err)
return err;
inode = f2fs_new_inode(idmap, dir, mode, NULL);
if (IS_ERR(inode))
return PTR_ERR(inode);
init_special_inode(inode, inode->i_mode, rdev);
inode->i_op = &f2fs_special_inode_operations;
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out;
f2fs_unlock_op(sbi);
f2fs_alloc_nid_done(sbi, inode->i_ino);
d_instantiate_new(dentry, inode);
if (IS_DIRSYNC(dir))
f2fs_sync_fs(sbi->sb, 1);
f2fs_balance_fs(sbi, true);
return 0;
out:
f2fs_handle_failed_inode(inode);
return err;
}
static int __f2fs_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
struct file *file, umode_t mode, bool is_whiteout,
struct inode **new_inode, struct f2fs_filename *fname)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
int err;
err = f2fs_dquot_initialize(dir);
if (err)
return err;
inode = f2fs_new_inode(idmap, dir, mode, NULL);
if (IS_ERR(inode))
return PTR_ERR(inode);
if (is_whiteout) {
init_special_inode(inode, inode->i_mode, WHITEOUT_DEV);
inode->i_op = &f2fs_special_inode_operations;
} else {
inode->i_op = &f2fs_file_inode_operations;
inode->i_fop = &f2fs_file_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
}
f2fs_lock_op(sbi);
err = f2fs_acquire_orphan_inode(sbi);
if (err)
goto out;
err = f2fs_do_tmpfile(inode, dir, fname);
if (err)
goto release_out;
/*
* add this non-linked tmpfile to orphan list, in this way we could
* remove all unused data of tmpfile after abnormal power-off.
*/
f2fs_add_orphan_inode(inode);
f2fs_alloc_nid_done(sbi, inode->i_ino);
if (is_whiteout) {
f2fs_i_links_write(inode, false);
spin_lock(&inode->i_lock);
inode->i_state |= I_LINKABLE;
spin_unlock(&inode->i_lock);
} else {
if (file)
d_tmpfile(file, inode);
else
f2fs_i_links_write(inode, false);
}
/* link_count was changed by d_tmpfile as well. */
f2fs_unlock_op(sbi);
unlock_new_inode(inode);
if (new_inode)
*new_inode = inode;
f2fs_balance_fs(sbi, true);
return 0;
release_out:
f2fs_release_orphan_inode(sbi);
out:
f2fs_handle_failed_inode(inode);
return err;
}
static int f2fs_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
struct file *file, umode_t mode)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
int err;
if (unlikely(f2fs_cp_error(sbi)))
return -EIO;
if (!f2fs_is_checkpoint_ready(sbi))
return -ENOSPC;
err = __f2fs_tmpfile(idmap, dir, file, mode, false, NULL, NULL);
return finish_open_simple(file, err);
}
static int f2fs_create_whiteout(struct mnt_idmap *idmap,
struct inode *dir, struct inode **whiteout,
struct f2fs_filename *fname)
{
return __f2fs_tmpfile(idmap, dir, NULL, S_IFCHR | WHITEOUT_MODE,
true, whiteout, fname);
}
int f2fs_get_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
struct inode **new_inode)
{
return __f2fs_tmpfile(idmap, dir, NULL, S_IFREG,
false, new_inode, NULL);
}
static int f2fs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
struct dentry *old_dentry, struct inode *new_dir,
struct dentry *new_dentry, unsigned int flags)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(old_dir);
struct inode *old_inode = d_inode(old_dentry);
struct inode *new_inode = d_inode(new_dentry);
struct inode *whiteout = NULL;
struct page *old_dir_page = NULL;
struct page *old_page, *new_page = NULL;
struct f2fs_dir_entry *old_dir_entry = NULL;
struct f2fs_dir_entry *old_entry;
struct f2fs_dir_entry *new_entry;
bool old_is_dir = S_ISDIR(old_inode->i_mode);
int err;
if (unlikely(f2fs_cp_error(sbi)))
return -EIO;
if (!f2fs_is_checkpoint_ready(sbi))
return -ENOSPC;
if (is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
(!projid_eq(F2FS_I(new_dir)->i_projid,
F2FS_I(old_dentry->d_inode)->i_projid)))
return -EXDEV;
/*
* If new_inode is null, the below renaming flow will
* add a link in old_dir which can convert inline_dir.
* After then, if we failed to get the entry due to other
* reasons like ENOMEM, we had to remove the new entry.
* Instead of adding such the error handling routine, let's
* simply convert first here.
*/
if (old_dir == new_dir && !new_inode) {
err = f2fs_try_convert_inline_dir(old_dir, new_dentry);
if (err)
return err;
}
if (flags & RENAME_WHITEOUT) {
struct f2fs_filename fname;
err = f2fs_setup_filename(old_dir, &old_dentry->d_name,
0, &fname);
if (err)
return err;
err = f2fs_create_whiteout(idmap, old_dir, &whiteout, &fname);
if (err)
return err;
}
err = f2fs_dquot_initialize(old_dir);
if (err)
goto out;
err = f2fs_dquot_initialize(new_dir);
if (err)
goto out;
if (new_inode) {
err = f2fs_dquot_initialize(new_inode);
if (err)
goto out;
}
err = -ENOENT;
old_entry = f2fs_find_entry(old_dir, &old_dentry->d_name, &old_page);
if (!old_entry) {
if (IS_ERR(old_page))
err = PTR_ERR(old_page);
goto out;
}
if (old_is_dir && old_dir != new_dir) {
old_dir_entry = f2fs_parent_dir(old_inode, &old_dir_page);
if (!old_dir_entry) {
if (IS_ERR(old_dir_page))
err = PTR_ERR(old_dir_page);
goto out_old;
}
}
if (new_inode) {
err = -ENOTEMPTY;
if (old_is_dir && !f2fs_empty_dir(new_inode))
goto out_dir;
err = -ENOENT;
new_entry = f2fs_find_entry(new_dir, &new_dentry->d_name,
&new_page);
if (!new_entry) {
if (IS_ERR(new_page))
err = PTR_ERR(new_page);
goto out_dir;
}
f2fs_balance_fs(sbi, true);
f2fs_lock_op(sbi);
err = f2fs_acquire_orphan_inode(sbi);
if (err)
goto put_out_dir;
f2fs_set_link(new_dir, new_entry, new_page, old_inode);
new_page = NULL;
inode_set_ctime_current(new_inode);
f2fs_down_write(&F2FS_I(new_inode)->i_sem);
if (old_is_dir)
f2fs_i_links_write(new_inode, false);
f2fs_i_links_write(new_inode, false);
f2fs_up_write(&F2FS_I(new_inode)->i_sem);
if (!new_inode->i_nlink)
f2fs_add_orphan_inode(new_inode);
else
f2fs_release_orphan_inode(sbi);
} else {
f2fs_balance_fs(sbi, true);
f2fs_lock_op(sbi);
err = f2fs_add_link(new_dentry, old_inode);
if (err) {
f2fs_unlock_op(sbi);
goto out_dir;
}
if (old_is_dir)
f2fs_i_links_write(new_dir, true);
}
f2fs_down_write(&F2FS_I(old_inode)->i_sem);
if (!old_is_dir || whiteout)
file_lost_pino(old_inode);
else
/* adjust dir's i_pino to pass fsck check */
f2fs_i_pino_write(old_inode, new_dir->i_ino);
f2fs_up_write(&F2FS_I(old_inode)->i_sem);
inode_set_ctime_current(old_inode);
f2fs_mark_inode_dirty_sync(old_inode, false);
f2fs_delete_entry(old_entry, old_page, old_dir, NULL);
old_page = NULL;
if (whiteout) {
set_inode_flag(whiteout, FI_INC_LINK);
err = f2fs_add_link(old_dentry, whiteout);
if (err)
goto put_out_dir;
spin_lock(&whiteout->i_lock);
whiteout->i_state &= ~I_LINKABLE;
spin_unlock(&whiteout->i_lock);
iput(whiteout);
}
if (old_dir_entry)
f2fs_set_link(old_inode, old_dir_entry, old_dir_page, new_dir);
if (old_is_dir)
f2fs_i_links_write(old_dir, false);
if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_STRICT) {
f2fs_add_ino_entry(sbi, new_dir->i_ino, TRANS_DIR_INO);
if (S_ISDIR(old_inode->i_mode))
f2fs_add_ino_entry(sbi, old_inode->i_ino,
TRANS_DIR_INO);
}
f2fs_unlock_op(sbi);
if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir))
f2fs_sync_fs(sbi->sb, 1);
f2fs_update_time(sbi, REQ_TIME);
return 0;
put_out_dir:
f2fs_unlock_op(sbi);
f2fs_put_page(new_page, 0);
out_dir:
if (old_dir_entry)
f2fs_put_page(old_dir_page, 0);
out_old:
f2fs_put_page(old_page, 0);
out:
iput(whiteout);
return err;
}
static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
struct inode *new_dir, struct dentry *new_dentry)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(old_dir);
struct inode *old_inode = d_inode(old_dentry);
struct inode *new_inode = d_inode(new_dentry);
struct page *old_dir_page, *new_dir_page;
struct page *old_page, *new_page;
struct f2fs_dir_entry *old_dir_entry = NULL, *new_dir_entry = NULL;
struct f2fs_dir_entry *old_entry, *new_entry;
int old_nlink = 0, new_nlink = 0;
int err;
if (unlikely(f2fs_cp_error(sbi)))
return -EIO;
if (!f2fs_is_checkpoint_ready(sbi))
return -ENOSPC;
if ((is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
!projid_eq(F2FS_I(new_dir)->i_projid,
F2FS_I(old_dentry->d_inode)->i_projid)) ||
(is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
!projid_eq(F2FS_I(old_dir)->i_projid,
F2FS_I(new_dentry->d_inode)->i_projid)))
return -EXDEV;
err = f2fs_dquot_initialize(old_dir);
if (err)
goto out;
err = f2fs_dquot_initialize(new_dir);
if (err)
goto out;
err = -ENOENT;
old_entry = f2fs_find_entry(old_dir, &old_dentry->d_name, &old_page);
if (!old_entry) {
if (IS_ERR(old_page))
err = PTR_ERR(old_page);
goto out;
}
new_entry = f2fs_find_entry(new_dir, &new_dentry->d_name, &new_page);
if (!new_entry) {
if (IS_ERR(new_page))
err = PTR_ERR(new_page);
goto out_old;
}
/* prepare for updating ".." directory entry info later */
if (old_dir != new_dir) {
if (S_ISDIR(old_inode->i_mode)) {
old_dir_entry = f2fs_parent_dir(old_inode,
&old_dir_page);
if (!old_dir_entry) {
if (IS_ERR(old_dir_page))
err = PTR_ERR(old_dir_page);
goto out_new;
}
}
if (S_ISDIR(new_inode->i_mode)) {
new_dir_entry = f2fs_parent_dir(new_inode,
&new_dir_page);
if (!new_dir_entry) {
if (IS_ERR(new_dir_page))
err = PTR_ERR(new_dir_page);
goto out_old_dir;
}
}
}
/*
* If cross rename between file and directory those are not
* in the same directory, we will inc nlink of file's parent
* later, so we should check upper boundary of its nlink.
*/
if ((!old_dir_entry || !new_dir_entry) &&
old_dir_entry != new_dir_entry) {
old_nlink = old_dir_entry ? -1 : 1;
new_nlink = -old_nlink;
err = -EMLINK;
if ((old_nlink > 0 && old_dir->i_nlink >= F2FS_LINK_MAX) ||
(new_nlink > 0 && new_dir->i_nlink >= F2FS_LINK_MAX))
goto out_new_dir;
}
f2fs_balance_fs(sbi, true);
f2fs_lock_op(sbi);
/* update ".." directory entry info of old dentry */
if (old_dir_entry)
f2fs_set_link(old_inode, old_dir_entry, old_dir_page, new_dir);
/* update ".." directory entry info of new dentry */
if (new_dir_entry)
f2fs_set_link(new_inode, new_dir_entry, new_dir_page, old_dir);
/* update directory entry info of old dir inode */
f2fs_set_link(old_dir, old_entry, old_page, new_inode);
f2fs_down_write(&F2FS_I(old_inode)->i_sem);
if (!old_dir_entry)
file_lost_pino(old_inode);
else
/* adjust dir's i_pino to pass fsck check */
f2fs_i_pino_write(old_inode, new_dir->i_ino);
f2fs_up_write(&F2FS_I(old_inode)->i_sem);
inode_set_ctime_current(old_dir);
if (old_nlink) {
f2fs_down_write(&F2FS_I(old_dir)->i_sem);
f2fs_i_links_write(old_dir, old_nlink > 0);
f2fs_up_write(&F2FS_I(old_dir)->i_sem);
}
f2fs_mark_inode_dirty_sync(old_dir, false);
/* update directory entry info of new dir inode */
f2fs_set_link(new_dir, new_entry, new_page, old_inode);
f2fs_down_write(&F2FS_I(new_inode)->i_sem);
if (!new_dir_entry)
file_lost_pino(new_inode);
else
/* adjust dir's i_pino to pass fsck check */
f2fs_i_pino_write(new_inode, old_dir->i_ino);
f2fs_up_write(&F2FS_I(new_inode)->i_sem);
inode_set_ctime_current(new_dir);
if (new_nlink) {
f2fs_down_write(&F2FS_I(new_dir)->i_sem);
f2fs_i_links_write(new_dir, new_nlink > 0);
f2fs_up_write(&F2FS_I(new_dir)->i_sem);
}
f2fs_mark_inode_dirty_sync(new_dir, false);
if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_STRICT) {
f2fs_add_ino_entry(sbi, old_dir->i_ino, TRANS_DIR_INO);
f2fs_add_ino_entry(sbi, new_dir->i_ino, TRANS_DIR_INO);
}
f2fs_unlock_op(sbi);
if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir))
f2fs_sync_fs(sbi->sb, 1);
f2fs_update_time(sbi, REQ_TIME);
return 0;
out_new_dir:
if (new_dir_entry) {
f2fs_put_page(new_dir_page, 0);
}
out_old_dir:
if (old_dir_entry) {
f2fs_put_page(old_dir_page, 0);
}
out_new:
f2fs_put_page(new_page, 0);
out_old:
f2fs_put_page(old_page, 0);
out:
return err;
}
static int f2fs_rename2(struct mnt_idmap *idmap,
struct inode *old_dir, struct dentry *old_dentry,
struct inode *new_dir, struct dentry *new_dentry,
unsigned int flags)
{
int err;
if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT))
return -EINVAL;
trace_f2fs_rename_start(old_dir, old_dentry, new_dir, new_dentry,
flags);
err = fscrypt_prepare_rename(old_dir, old_dentry, new_dir, new_dentry,
flags);
if (err)
return err;
if (flags & RENAME_EXCHANGE)
err = f2fs_cross_rename(old_dir, old_dentry,
new_dir, new_dentry);
else
/*
* VFS has already handled the new dentry existence case,
* here, we just deal with "RENAME_NOREPLACE" as regular rename.
*/
err = f2fs_rename(idmap, old_dir, old_dentry,
new_dir, new_dentry, flags);
trace_f2fs_rename_end(old_dentry, new_dentry, flags, err);
return err;
}
static const char *f2fs_encrypted_get_link(struct dentry *dentry,
struct inode *inode,
struct delayed_call *done)
{
struct page *page;
const char *target;
if (!dentry)
return ERR_PTR(-ECHILD);
page = read_mapping_page(inode->i_mapping, 0, NULL);
if (IS_ERR(page))
return ERR_CAST(page);
target = fscrypt_get_symlink(inode, page_address(page),
inode->i_sb->s_blocksize, done);
put_page(page);
return target;
}
static int f2fs_encrypted_symlink_getattr(struct mnt_idmap *idmap,
const struct path *path,
struct kstat *stat, u32 request_mask,
unsigned int query_flags)
{
f2fs_getattr(idmap, path, stat, request_mask, query_flags);
return fscrypt_symlink_getattr(path, stat);
}
const struct inode_operations f2fs_encrypted_symlink_inode_operations = {
.get_link = f2fs_encrypted_get_link,
.getattr = f2fs_encrypted_symlink_getattr,
.setattr = f2fs_setattr,
.listxattr = f2fs_listxattr,
};
const struct inode_operations f2fs_dir_inode_operations = {
.create = f2fs_create,
.lookup = f2fs_lookup,
.link = f2fs_link,
.unlink = f2fs_unlink,
.symlink = f2fs_symlink,
.mkdir = f2fs_mkdir,
.rmdir = f2fs_rmdir,
.mknod = f2fs_mknod,
.rename = f2fs_rename2,
.tmpfile = f2fs_tmpfile,
.getattr = f2fs_getattr,
.setattr = f2fs_setattr,
.get_inode_acl = f2fs_get_acl,
.set_acl = f2fs_set_acl,
.listxattr = f2fs_listxattr,
.fiemap = f2fs_fiemap,
.fileattr_get = f2fs_fileattr_get,
.fileattr_set = f2fs_fileattr_set,
};
const struct inode_operations f2fs_symlink_inode_operations = {
.get_link = f2fs_get_link,
.getattr = f2fs_getattr,
.setattr = f2fs_setattr,
.listxattr = f2fs_listxattr,
};
const struct inode_operations f2fs_special_inode_operations = {
.getattr = f2fs_getattr,
.setattr = f2fs_setattr,
.get_inode_acl = f2fs_get_acl,
.set_acl = f2fs_set_acl,
.listxattr = f2fs_listxattr,
};