Re: [PATCH v4 4/6] add statmount(2) syscall

[Christian Brauner]

On Fri, Nov 10, 2023 at 12:00:22PM -0500, Paul Moore wrote:
> On Wed, Nov 8, 2023 at 3:10 PM Paul Moore <paul@xxxxxxxxxxxxxx> wrote:
> > On Wed, Nov 8, 2023 at 2:58 AM Christian Brauner <brauner@xxxxxxxxxx> wrote:
> > > > > +static int do_statmount(struct stmt_state *s)
> > > > > +{
> > > > > +   struct statmnt *sm = &s->sm;
> > > > > +   struct mount *m = real_mount(s->mnt);
> > > > > +   size_t copysize = min_t(size_t, s->bufsize, sizeof(*sm));
> > > > > +   int err;
> > > > > +
> > > > > +   err = security_sb_statfs(s->mnt->mnt_root);
> > > > > +   if (err)
> > > > > +           return err;
> > > > > +
> > > > > +   if (!capable(CAP_SYS_ADMIN) &&
> > > > > +       !is_path_reachable(m, m->mnt.mnt_root, &s->root))
> > > > > +           return -EPERM;
> > > >
> > > > In order to be consistent with our typical access control ordering,
> > > > please move the security_sb_statfs() call down to here, after the
> > > > capability checks.
> > >
> > > I've moved the security_sb_statfs() calls accordingly.
> >
> > Okay, good.  Did I miss a comment or a patch where that happened?  I
> > looked over the patchset and comments yesterday and didn't recall
> > seeing anything about shuffling the access control checks.
> 
> Gentle ping on this.  I'm asking because I know there have been issues
> lately with the lists and some mail providers and I want to make sure
> I'm not missing anything, I double checked lore again and didn't see
> anything there either, but I might be missing it.

Sorry, I'm traveling so I just didn't see this. Please see:

https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/commit/?h=vfs.mount&id=dc14fa93943918bee898d75d7ae72fc3623ce9ce
https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/commit/?h=vfs.mount&id=de17643cbf9b0282990bb9cf0e0bf01710c9ec03

I've folded the fixup into these patches. I probably just accidently
dropped the diff from my reply.