kernel/linux.git/fs/overlayfs/dir.c, branch v6.12.102

lsm: add backing_file LSM hooks

2026-07-04T11:43:31+00:00

[ Upstream commit 6af36aeb147a06dea47c49859cd6ca5659aeb987 ] Stacked filesystems such as overlayfs do not currently provide the necessary mechanisms for LSMs to properly enforce access controls on the mmap() and mprotect() operations. In order to resolve this gap, a LSM security blob is being added to the backing_file struct and the following new LSM hooks are being created: security_backing_file_alloc() security_backing_file_free() security_mmap_backing_file() The first two hooks are to manage the lifecycle of the LSM security blob in the backing_file struct, while the third provides a new mmap() access control point for the underlying backing file. It is also expected that LSMs will likely want to update their security_file_mprotect() callback to address issues with their mprotect() controls, but that does not require a change to the security_file_mprotect() LSM hook. There are a three other small changes to support these new LSM hooks: * Pass the user file associated with a backing file down to alloc_empty_backing_file() so it can be included in the security_backing_file_alloc() hook. * Add getter and setter functions for the backing_file struct LSM blob as the backing_file struct remains private to fs/file_table.c. * Constify the file struct field in the LSM common_audit_data struct to better support LSMs that need to pass a const file struct pointer into the common LSM audit code. Thanks to Arnd Bergmann for identifying the missing EXPORT_SYMBOL_GPL() and supplying a fixup. Cc: stable@vger.kernel.org Cc: linux-fsdevel@vger.kernel.org Cc: linux-unionfs@vger.kernel.org Cc: linux-erofs@lists.ozlabs.org Reviewed-by: Amir Goldstein Reviewed-by: Serge Hallyn Reviewed-by: Christian Brauner Signed-off-by: Paul Moore [ Mainline declares lsm_backing_file_cache in security/lsm.h. Linux 6.12.y does not have security/lsm_init.c or security/lsm.h; the cache variable is defined locally as static struct kmem_cache *lsm_backing_file_cache in security/security.c. ] Signed-off-by: Cai Xinchen Signed-off-by: Sasha Levin

ovl: fix copy-up in tmpfile

2024-05-28T08:06:55+00:00

Move ovl_copy_up() call outside of ovl_want_write()/ovl_drop_write() region, since copy up may also call ovl_want_write() resulting in recursive locking on sb->s_writers. Reported-and-tested-by: syzbot+85e58cdf5b3136471d4b@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/000000000000f6865106191c3e58@google.com/ Fixes: 9a87907de359 ("ovl: implement tmpfile") Signed-off-by: Miklos Szeredi

ovl: remove upper umask handling from ovl_create_upper()

2024-05-02T18:35:57+00:00

This is already done by vfs_prepare_mode() when creating the upper object by vfs_create(), vfs_mkdir() and vfs_mknod(). No regressions have been observed in xfstests run with posix acls turned off for the upper filesystem. Fixes: 1639a49ccdce ("fs: move S_ISGID stripping into the vfs_*() helpers") Signed-off-by: Miklos Szeredi

ovl: implement tmpfile

2024-05-02T18:35:57+00:00

Combine inode creation with opening a file. There are six separate objects that are being set up: the backing inode, dentry and file, and the overlay inode, dentry and file. Cleanup in case of an error is a bit of a challenge and is difficult to test, so careful review is needed. All tmpfile testcases except generic/509 now run/pass, and no regressions are observed with full xfstests. Signed-off-by: Miklos Szeredi Reviewed-by: Amir Goldstein

rename(): avoid a deadlock in the case of parents having no common ancestor

2023-11-25T07:54:14+00:00

... and fix the directory locking documentation and proof of correctness. Holding ->s_vfs_rename_mutex *almost* prevents ->d_parent changes; the case where we really don't want it is splicing the root of disconnected tree to somewhere. In other words, ->s_vfs_rename_mutex is sufficient to stabilize "X is an ancestor of Y" only if X and Y are already in the same tree. Otherwise it can go from false to true, and one can construct a deadlock on that. Make lock_two_directories() report an error in such case and update the callers of lock_rename()/lock_rename_child() to handle such errors. And yes, such conditions are not impossible to create ;-/ Reviewed-by: Jan Kara Signed-off-by: Al Viro

ovl: Add an alternative type of whiteout

2023-10-30T22:12:59+00:00

An xattr whiteout (called "xwhiteout" in the code) is a reguar file of zero size with the "overlay.whiteout" xattr set. A file like this in a directory with the "overlay.whiteouts" xattrs set will be treated the same way as a regular whiteout. The "overlay.whiteouts" directory xattr is used in order to efficiently handle overlay checks in readdir(), as we only need to checks xattrs in affected directories. The advantage of this kind of whiteout is that they can be escaped using the standard overlay xattr escaping mechanism. So, a file with a "overlay.overlay.whiteout" xattr would be unescaped to "overlay.whiteout", which could then be consumed by another overlayfs as a whiteout. Overlayfs itself doesn't create whiteouts like this, but a userspace mechanism could use this alternative mechanism to convert images that may contain whiteouts to be used with overlayfs. To work as a whiteout for both regular overlayfs mounts as well as userxattr mounts both the "user.overlay.whiteout*" and the "trusted.overlay.whiteout*" xattrs will need to be created. Signed-off-by: Alexander Larsson Reviewed-by: Amir Goldstein Signed-off-by: Amir Goldstein

ovl: reorder ovl_want_write() after ovl_inode_lock()

2023-10-30T22:12:57+00:00

Make the locking order of ovl_inode_lock() strictly between the two vfs stacked layers, i.e.: - ovl vfs locks: sb_writers, inode_lock, ... - ovl_inode_lock - upper vfs locks: sb_writers, inode_lock, ... To that effect, move ovl_want_write() into the helpers ovl_nlink_start() and ovl_copy_up_start which currently take the ovl_inode_lock() after ovl_want_write(). Signed-off-by: Amir Goldstein

ovl: store enum redirect_mode in config instead of a string

2023-06-19T11:02:01+00:00

Do all the logic to set the mode during mount options parsing and do not keep the option string around. Use a constant_table to translate from enum redirect mode to string in preperation for new mount api option parsing. The mount option "off" is translated to either "follow" or "nofollow", depending on the "redirect_always_follow" build/module config, so in effect, there are only three possible redirect modes. This results in a minor change to the string that is displayed in show_options() - when redirect_dir is enabled by default and the user mounts with the option "redirect_dir=off", instead of displaying the mode "redirect_dir=off" in show_options(), the displayed mode will be either "redirect_dir=follow" or "redirect_dir=nofollow", depending on the value of "redirect_always_follow" build/module config. The displayed mode reflects the effective mode, so mounting overlayfs again with the dispalyed redirect_dir option will result with the same effective and displayed mode. Signed-off-by: Amir Goldstein

ovl: negate the ofs->share_whiteout boolean

2023-06-19T11:02:00+00:00

The default common case is that whiteout sharing is enabled. Change to storing the negated no_shared_whiteout state, so we will not need to initialize it. This is the first step towards removing all config and feature initializations out of ovl_fill_super(). Signed-off-by: Amir Goldstein

ovl: move ovl_entry into ovl_inode

2023-06-19T11:01:13+00:00

The lower stacks of all the ovl inode aliases should be identical and there is redundant information in ovl_entry and ovl_inode. Move lowerstack into ovl_inode and keep only the OVL_E_FLAGS per overlay dentry. Following patches will deduplicate redundant ovl_inode fields. Note that for pure upper and negative dentries, OVL_E(dentry) may be NULL now, so it is imporatnt to use the ovl_numlower() accessor. Reviewed-by: Alexander Larsson Signed-off-by: Amir Goldstein Signed-off-by: Miklos Szeredi