[PATCH -next -v4 00/17] landlock: Add READ_METADATA and WRITE_METADATA access rights
From: Cai Xinchen
Date: Tue Nov 17 2026 - 09:34:04 EST
This series adds two new Landlock filesystem access rights,
LANDLOCK_ACCESS_FS_READ_METADATA and LANDLOCK_ACCESS_FS_WRITE_METADATA,
which control access to file and directory metadata such as inode
attributes (mode, ownership, timestamps), extended attributes, POSIX
ACLs and chattr(1)-style flags. It picks up the work from the
"landlock: add chmod and chown support" series [1] and follows the
coarse-grained grouping discussed in that thread [2]: instead of
separate chmod/chown rights, metadata operations are grouped into one
read and one write right.
Landlock evaluates access rights on a per-path basis, but the
metadata related LSM hooks (inode_setattr, inode_setsecctx, the inode
xattr, POSIX ACL and fileattr hooks) only receive the dentry of the
accessed object. Patches 1-8 therefore first pass struct path
instead of dentry through the metadata-related VFS helpers and LSM
hooks. The hooks also receive a struct file argument, set by the
VFS to the file through which the operation is requested when there
is one (i.e. for the file descriptor-based system calls such as
fchmod(2), fchown(2), fgetxattr(2), utimensat(2) on an empty path,
and for the FS_IOC_* ioctls and file_[gs]etattr(2) with
AT_EMPTY_PATH), and NULL otherwise (i.e. for the path-based system
calls and for kernel-internal accesses). This is a pure refactoring
with no behavior change for the other LSMs, which only have their
hook implementations' signatures updated; the file argument is only
consumed by Landlock (patch 14). The series is split so that every
patch builds on its own:
1: notify_change() and its callers
2: inode_setsecctx hook (must come before 3: the SELinux and Smack
implementations call __vfs_setxattr_locked internally)
3: xattr helpers, which also drops a redundant EVM xattr size sanity
check whose vfs_getxattr() call only has a dentry and therefore
cannot be migrated to the new path-based signature
4: POSIX ACL helpers
5: inode xattr hooks, threaded through the xattr helpers so that
the f[gs]etxattr(2) and io_uring flows pass their file
6: inode POSIX ACL hooks, threaded through do_set_acl()/do_get_acl()
7: inode fileattr hooks (FS_IOC_[GS]ETFLAGS, FS_IOC_FS[SG]ETXATTR,
file_[gs]etattr(2)), which also adds a __vfs_fileattr_get()
counterpart without security checks for the kernel-internal
readers (nfsd), and removes the ovl_security_fileattr() hack by
forwarding the real upper/lower path to the checked
vfs_fileattr_get()/_set() from ecryptfs and overlayfs
8: inode_setattr hook, which now receives the struct file held by
the fd-based callers (fchmod(2), fchown(2), fchmodat2(2) and
fchownat(2) with AT_EMPTY_PATH, utimensat(2) on an empty path);
do_truncate() passes NULL because size changes are governed by
the truncate hooks. The hook is renamed to path_setattr in the
same patch, so that a stale BPF LSM program attaching to
lsm/inode_setattr fails to load instead of reading dentry->d_flags
where it expects iattr->ia_valid (see below)
Three deliberate scoping decisions for this refactor:
- The hooks consistently take struct path rather than struct file as
their primary object. The VFS call sites involved (chmod(2),
chown(2), utimensat(2), xattr(2) and ACL syscalls) operate on
paths, and several of them (lstat(2), lchown(2), llistxattr(2),
...) have no struct file to begin with. The optional struct file
argument carries the open-time semantic described below.
- struct inode_operations->setattr still receives (idmap, dentry, attr).
Only the VFS boundary (notify_change()) and the LSM hook layer see the
path, which keeps the refactor contained to fs/attr.c and the LSM
infrastructure instead of touching every filesystem.
- The path argument of the hooks is only meant to be handed to VFS
helpers taking a struct path (e.g., notify_change(), the xattr
helpers): the VFS makes no guarantees with regard to path->mnt,
whose layout and implementation details are subject to change, and
implementations must not rely on its properties. The idmap, in
contrast, is passed separately so that the VFS and the filesystem
are guaranteed to see the same idmap.
Unlike what was done for LANDLOCK_ACCESS_FS_TRUNCATE, no new file_*
hooks are introduced for the file descriptor flows: the metadata
hooks are already consumed by other LSMs (SELinux, Smack, EVM, IMA,
AppArmor, TOMOYO) on these flows, and routing them to new hooks
would either silently remove the other LSMs' checks, or keep the
path_* hooks as well and check twice with Landlock. Passing the
struct file to the existing path_* hooks instead keeps the pipeline
and the other LSMs' behavior unchanged, while giving Landlock the
information it needs to apply the open-time semantic below.
Since these hooks now operate on paths rather than inodes, patches
8-13 rename them to the path_* naming convention used by the other
hooks that take a struct path (path_chmod, path_chown, path_link,
...). The inode_setattr hook is renamed in patch 8, in the same patch
that changes its ctx layout: BPF LSM program arguments are retyped
from the hook prototype at verification time and are not checked
against a versioned ABI, so leaving the old name in place for a few
patches would let a stale program read dentry->d_flags where it
expects iattr->ia_valid, silently inverting a deny rule. The
inode_getattr hook, which already received a struct path, is renamed
to path_getattr as well; patch 9 also threads the struct file held by
the fd-based vfs_fstat() and vfs_statx_fd() callers through it, while
the many kernel-internal vfs_getattr() callers keep a wrapper which
passes NULL. Apart from patch 8 and the file threading in patch 9,
these renames are purely mechanical and can be reviewed independently.
Patch 14 then adds the access rights and implements their
enforcement, and patches 15-17 add the selftests, the sandboxer
sample update and the documentation.
Patch 14 applies the LANDLOCK_ACCESS_FS_TRUNCATE mechanic to the
new rights, following Mickaël's confirmation [4]: for operations on
already opened files, it is the access rights at the time of open(2)
which decide whether a metadata operation is permitted, not the
rights of the current thread. READ_METADATA and WRITE_METADATA are
looked up at open time, recorded in the opened file, and consumed by
the metadata hooks when they are called with a file. As a
consequence:
- A file descriptor opened before sandboxing keeps allowing metadata
operations whatever the current domain is, which resolves the
"catch-22" raised by Justin [5]: an already opened file descriptor
keeps working, it is not an escape hatch because its rights were
fixed when it was legitimate to open the file.
- Conversely, a restricted file descriptor passed to another
(possibly non-landlocked) process keeps denying metadata
operations: the file descriptor carries the rights of the process
which opened it.
- ioctl(FS_IOC_GETFLAGS) and friends on a file descriptor opened
without READ_METADATA are now denied, which closes the metadata
leak Justin was concerned about.
- O_PATH file descriptors are an exception: they are not opened
through the open permission checks (cf. do_dentry_open()) and
carry no recorded rights; metadata accesses through them are
checked against the current domain, like path-based accesses.
- File descriptors obtained without going through the file_open
hook (e.g. memfd_create(2), pipes, sockets) keep the full access
rights granted at file allocation time, consistent with the
truncate behavior.
- deny_masks_t is widened from u8 to u16 to keep tracking one
nibble per optional access right for the audit logs.
The Landlock ABI version is incremented from 12 to 13; ABI 12 was
taken by the capability and namespace use restrictions that are
already in linux-next.
The series is based on linux-next commit 749a4de37040 ("Add linux-next
specific files for 20261116").
Changes since v3 [6]:
- The empty-path branch of do_fchmodat(), do_fchownat() and
do_utimes_path() used fdget(), which rejects O_PATH file
descriptors. It now uses fd_raw, the same lookup path_init() uses
for an empty path, so AT_EMPTY_PATH metadata operations work again
on O_PATH descriptors; futimens(2) keeps rejecting O_PATH, and a
NULL pathname returns -EFAULT instead of being turned into an empty
path (patch 8).
- Every hook that gained a struct file argument is declared with a
strong file__nullable definition in kernel/bpf/bpf_lsm_proto.c, so
the verifier requires a NULL check before dereferencing it
(patches 5-12).
- The ntfs named stream setattr no longer bypasses the hooks: it goes
through notify_change_target(), which performs the path-based LSM
checks on the caller-supplied path while the attributes land on the
base inode, and the borrowed path is dropped from the caller-owned
struct iattr before notify_change() returns (patch 1).
- ecryptfs and overlayfs run the fileattr hooks on the real
upper/lower path instead of calling unchecked helpers, and the
ksmbd xattr helper passes the file it was given instead of NULL
(patches 5 and 7).
- Since a trusted struct path lets BPF LSM programs reach path->dentry
and call the dentry xattr kfuncs, the sleepable path hooks that
always run under i_rwsem use the _locked kfunc variants, while the
hooks that can run either way (path_link, path_rename, path_rmdir,
path_unlink and path_getattr, which nfsd reaches with the inode
already locked) reject both variants (patch 8).
- fchmod(2) and fchown(2) are not O_PATH-capable metadata interfaces
and are no longer listed as such in the UAPI header and in the
userspace documentation (patches 14 and 17).
- The setattr hook rename is folded into the patch that changes its
ctx layout, so that a stale BPF LSM program cannot be silently
reinterpreted, and that patch now comes right before the getattr
rename (patch 8).
- ksmbd reports the share root for the ".." entry when follow_up()
cannot leave a mount that was detached, instead of using a parent
which that mount does not cover (patch 3).
- The Landlock changelog states that kernel-internal getattr calls
made under the caller's credentials, such as NFS localio's
post-write attribute refresh, are subject to READ_METADATA too, and
that localio discards the resulting error (patch 14).
[1]
https://lore.kernel.org/all/20220827111215.131442-1-xiujianfeng@xxxxxxxxxx/
[2]
https://lore.kernel.org/all/abc960a1-e66e-792e-6869-cfd201c29dbe@xxxxxxxxxxx/
[3]
https://lore.kernel.org/all/20261015125425.3246251-1-caixinchen1@xxxxxxxxxx/
[4] https://lore.kernel.org/all/20260929.Ahgoo4yeetai@xxxxxxxxxxx/
[5] https://lore.kernel.org/all/arvjb0HJYh1KyEBB@suesslenovo/
[6]
https://lore.kernel.org/all/20261027121151.1887650-1-caixinchen1@xxxxxxxxxx/
Cai Xinchen (17):
fs: pass struct path to notify_change()
LSM: pass struct path to the inode_setsecctx hook
fs: pass struct path to xattr helpers
fs: pass struct path to POSIX ACL helpers
LSM: pass struct path and struct file to the inode xattr hooks
LSM: pass struct path and struct file to the inode posix acl hooks
LSM: pass struct path and struct file to the inode fileattr hooks
LSM: rename inode_setattr to path_setattr and pass struct path and
file
LSM: rename inode_getattr to path_getattr and pass struct file
LSM: rename the inode xattr hooks to path_xattr hooks
LSM: rename the inode posix acl hooks to path_*_acl hooks
LSM: rename the inode fileattr hooks to path_file_* hooks
LSM: rename inode_setsecctx to path_setsecctx
landlock: Add READ_METADATA and WRITE_METADATA access rights
selftests/landlock: Add tests for metadata access rights
samples/landlock: Add metadata rights to sandboxer
Documentation: Update landlock doc for metadata rights
Documentation/admin-guide/LSM/landlock.rst | 2 +
Documentation/bpf/fs_kfuncs.rst | 2 +-
Documentation/filesystems/porting.rst | 63 ++
Documentation/trace/events-landlock.rst | 18 +-
Documentation/userspace-api/landlock.rst | 148 ++-
drivers/base/devtmpfs.c | 6 +-
drivers/block/loop.c | 12 +-
drivers/block/zloop.c | 23 +-
fs/attr.c | 119 ++-
fs/bpf_fs_kfuncs.c | 58 +-
fs/btrfs/volumes.c | 2 +-
fs/cachefiles/xattr.c | 45 +-
fs/coredump.c | 2 +-
fs/ecryptfs/inode.c | 48 +-
fs/exfat/file.c | 5 +-
fs/fat/file.c | 5 +-
fs/file_attr.c | 136 ++-
fs/init.c | 6 +-
fs/inode.c | 7 +-
fs/internal.h | 26 +-
fs/namei.c | 5 +-
fs/nfsd/nfs4ctl.h | 5 +-
fs/nfsd/nfs4state.c | 16 +-
fs/nfsd/nfs4xdr.c | 2 +-
fs/nfsd/state.h | 2 +-
fs/nfsd/vfs.c | 87 +-
fs/ntfs/named_stream.c | 14 +-
fs/open.c | 77 +-
fs/overlayfs/copy_up.c | 4 +-
fs/overlayfs/inode.c | 27 +-
fs/overlayfs/overlayfs.h | 44 +-
fs/overlayfs/xattrs.c | 12 +-
fs/posix_acl.c | 61 +-
fs/smb/server/smb2pdu.c | 59 +-
fs/smb/server/smb_common.c | 59 +-
fs/smb/server/smbacl.c | 12 +-
fs/smb/server/vfs.c | 101 +-
fs/smb/server/vfs.h | 19 +-
fs/stat.c | 82 +-
fs/utimes.c | 38 +-
fs/xattr.c | 114 ++-
fs/xfs/xfs_ioctl.c | 2 +-
include/linux/fileattr.h | 10 +-
include/linux/fs.h | 30 +-
include/linux/landlock.h | 4 +-
include/linux/lsm_hook_defs.h | 66 +-
include/linux/posix_acl.h | 25 +-
include/linux/security.h | 131 +--
include/linux/xattr.h | 32 +-
include/uapi/linux/landlock.h | 47 +-
init/initramfs.c | 2 +-
kernel/bpf/bpf_lsm.c | 10 +-
kernel/bpf/bpf_lsm_proto.c | 108 ++
kernel/bpf/verifier.c | 13 +
kernel/trace/bpf_trace.c | 2 +-
samples/landlock/sandboxer.c | 16 +-
security/apparmor/lsm.c | 5 +-
security/commoncap.c | 17 +-
security/integrity/evm/evm_crypto.c | 8 +-
security/integrity/evm/evm_main.c | 77 +-
security/integrity/ima/ima_appraise.c | 41 +-
security/landlock/access.h | 13 +-
security/landlock/fs.c | 235 ++++-
security/landlock/limits.h | 2 +-
security/landlock/syscalls.c | 2 +-
security/security.c | 179 ++--
security/selinux/hooks.c | 105 +-
security/smack/smack_lsm.c | 128 ++-
security/tomoyo/tomoyo.c | 8 +-
.../testing/selftests/bpf/prog_tests/d_path.c | 2 +-
.../selftests/bpf/prog_tests/fs_kfuncs.c | 10 +-
.../testing/selftests/bpf/progs/test_d_path.c | 5 +-
.../bpf/progs/test_d_path_check_rdonly_mem.c | 5 +-
.../bpf/progs/test_d_path_check_types.c | 5 +-
.../selftests/bpf/progs/test_get_xattr.c | 7 +-
.../bpf/progs/test_set_remove_xattr.c | 34 +-
.../bpf/progs/verifier_lsm_init_xattr.c | 8 +-
.../selftests/bpf/progs/verifier_vfs_accept.c | 2 +-
.../selftests/bpf/progs/verifier_vfs_reject.c | 2 +-
tools/testing/selftests/landlock/base_test.c | 2 +-
tools/testing/selftests/landlock/fs_test.c | 932 +++++++++++++++++-
.../selftests/landlock/trace_fs_test.c | 3 +-
82 files changed, 3003 insertions(+), 905 deletions(-)
--
2.18.0.huawei.25