On Oct 11, 2024 Song Liu [off-list ref] wrote:

Currently, fsnotify_open_perm() is called from security_file_open(). This
is not right for CONFIG_SECURITY=n and CONFIG_FSNOTIFY=y case, as
security_file_open() in this combination will be a no-op and not call
fsnotify_open_perm(). Fix this by calling fsnotify_open_perm() directly.

Signed-off-by: Song Liu <song@kernel.org>
---
PS: I didn't included a Fixes tag. This issue was probably introduced 15
years ago in [1]. If we want to back port this to stable, we will need
another version for older kernel because of [2].

[1] c4ec54b40d33 ("fsnotify: new fsnotify hooks and events types for access decisions")
[2] 36e28c42187c ("fsnotify: split fsnotify_perm() into two hooks")
---
 fs/open.c           | 4 ++++
 security/security.c | 9 +--------
 2 files changed, 5 insertions(+), 8 deletions(-)

This looks fine to me, if we can get an ACK from the VFS folks I can
merge this into the lsm/stable-6.12 tree and send it to Linus, or the
VFS folks can do it if they prefer (my ACK is below just in case).

As far as stable prior to v6.8 is concerned, once this hits Linus'
tree you can submit an adjusted backport for the older kernels to the
stable team.

Acked-by: Paul Moore <paul@paul-moore.com>

--
paul-moore.com