diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2023-08-28 09:00:09 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2023-08-28 09:00:09 -0700 |
commit | 84ab1277ce5a90a8d1f377707d662ac43cc0918a (patch) | |
tree | ac7ca4f1213806f6489bb85f0049090ffd6a2de1 /drivers/hid | |
parent | 2dde18cd1d8fac735875f2e4987f11817cc0bc2c (diff) | |
parent | 22ed7ecdaefe0cac0c6e6295e83048af60435b13 (diff) | |
download | linux-stable-84ab1277ce5a90a8d1f377707d662ac43cc0918a.tar.gz linux-stable-84ab1277ce5a90a8d1f377707d662ac43cc0918a.tar.bz2 linux-stable-84ab1277ce5a90a8d1f377707d662ac43cc0918a.zip |
Merge tag 'v6.6-vfs.fs_context' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull mount API updates from Christian Brauner:
"This introduces FSCONFIG_CMD_CREATE_EXCL which allows userspace to
implement something like
$ mount -t ext4 --exclusive /dev/sda /B
which fails if a superblock for the requested filesystem does already
exist instead of silently reusing an existing superblock.
Without it, in the sequence
$ move-mount -f xfs -o source=/dev/sda4 /A
$ move-mount -f xfs -o noacl,source=/dev/sda4 /B
the initial mounter will create a superblock. The second mounter will
reuse the existing superblock, creating a bind-mount (see [1] for the
source of the move-mount binary).
The problem is that reusing an existing superblock means all mount
options other than read-only and read-write will be silently ignored
even if they are incompatible requests. For example, the second mount
has requested no POSIX ACL support but since the existing superblock
is reused POSIX ACL support will remain enabled.
Such silent superblock reuse can easily become a security issue.
After adding support for FSCONFIG_CMD_CREATE_EXCL to mount(8) in
util-linux this can be fixed:
$ move-mount -f xfs --exclusive -o source=/dev/sda4 /A
$ move-mount -f xfs --exclusive -o noacl,source=/dev/sda4 /B
Device or resource busy | move-mount.c: 300: do_fsconfig: i xfs: reusing existing filesystem not allowed
This requires the new mount api. With the old mount api it would be
necessary to plumb this through every legacy filesystem's
file_system_type->mount() method. If they want this feature they are
most welcome to switch to the new mount api"
Link: https://github.com/brauner/move-mount-beneath [1]
Link: https://lore.kernel.org/linux-block/20230704-fasching-wertarbeit-7c6ffb01c83d@brauner
Link: https://lore.kernel.org/linux-block/20230705-pumpwerk-vielversprechend-a4b1fd947b65@brauner
Link: https://lore.kernel.org/linux-fsdevel/20230725-einnahmen-warnschilder-17779aec0a97@brauner
Link: https://lore.kernel.org/lkml/20230824-anzog-allheilmittel-e8c63e429a79@brauner/
* tag 'v6.6-vfs.fs_context' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
fs: add FSCONFIG_CMD_CREATE_EXCL
fs: add vfs_cmd_reconfigure()
fs: add vfs_cmd_create()
super: remove get_tree_single_reconf()
Diffstat (limited to 'drivers/hid')
0 files changed, 0 insertions, 0 deletions