On 2/5/2021 6:17 AM, Serge E. Hallyn wrote:

On Tue, Feb 02, 2021 at 05:27:03PM +0100, Mickaël Salaün wrote:

quoted

From: Casey Schaufler <casey@schaufler-ca.com>

Move management of the superblock->sb_security blob out of the
individual security modules and into the security infrastructure.
Instead of allocating the blobs from within the modules, the modules
tell the infrastructure how much space is required, and the space is
allocated there.

Cc: Kees Cook <redacted>
Cc: John Johansen <john.johansen@canonical.com>
Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Mickaël Salaün <redacted>
Reviewed-by: Stephen Smalley <stephen.smalley.work@gmail.com>

Acked-by: Serge Hallyn <serge@hallyn.com>

I wonder how many out of tree modules this will impact :)

There are several blobs that have already been converted
to infrastructure management. Not a peep from out-of-tree
module developers/maintainers. I can only speculate that
OOT modules are either less common than we may think, using
alternative data management models (as does eBPF) or
sticking with very old kernels. It's also possible that
they're suffering in silence, which would be sad because
every module that's worth having should be in the tree.

Actually
if some new incoming module does an rcu callback to free the
sb_security, then the security_sb_free will need an update, but
that seems unlikely.

We're already doing that for the inode blob, so it's
really just a small matter of cut-n-paste and s/inode/sb/
to make that happen.