[RFC PATCH 3/3] add listmnt(2) syscall

[Paul Moore]

On Mon, Sep 18, 2023 at 12:52 PM Christian Brauner <brauner at kernel.org> wrote:
> On Sun, Sep 17, 2023 at 04:32:04PM +0200, Miklos Szeredi wrote:
> > On Sun, Sep 17, 2023 at 2:54 AM Matthew House <mattlloydhouse at gmail.com> wrote:
> >
> > > > +       list_for_each_entry(r, &m->mnt_mounts, mnt_child) {
> > > > +               if (!capable(CAP_SYS_ADMIN) &&
>
>
> > Good point.  That issue was nagging at the back of my mind.  Having an
> > explicit flag nicely solves the issue.
>
> Ideally we avoid multiple capable(CAP_SYS_ADMIN) calls by only doing it
> once and saving the return value. capable() call's aren't that cheap.

Agreed.  The capability check doesn't do any subject/object
comparisons so calling it for each mount is overkill.  However, I
would think we would want the LSM hook called from inside the loop as
that could involve a subject (@current) and object (individual mount
point) comparison.

> Plus, we should decide whether this should trigger an audit event or
> not: capable(CAP_SYS_ADMIN) triggers an audit event,
> ns_capable_noaudit(&init_user_ns, CAP_SYS_ADMIN) wouldn't.

Why would we not want to audit the capable() call?

-- 
paul-moore.com