Re: [RFC 4/8] lsm: smack: smack callbacks for kdbus security hooks

On Wednesday, July 08, 2015 09:38:25 AM Casey Schaufler wrote:

On 7/8/2015 6:42 AM, Stephen Smalley wrote:

quoted

On 07/08/2015 06:25 AM, Paul Osmialowski wrote:

...

quoted

If Smack only truly needs 3 hooks, then it begs the question of why
there are so many other hooks defined.  Are the other hooks just to
support finer-grained distinctions, or is Smack's coverage incomplete?

I haven't been following kdbus closely for a while, but the original
intent for Smack and kdbus was that it Smack controls would be on the
objects involved, and that to accomplish that only a small number of
hooks would be necessary. After all, Smack uses fewer hooks than SELinux
on other things. I do agree that without a user there is no point in
having hooks. If SELinux requires the other hooks we might want to
hold off on asking for the hooks until the SELinux implementation is
exposed. I also think that AppArmor should be examined as a potential
user of the hooks, just to make sure the hooks aren't excessively
oriented toward subject/object based security modules.

In Paul O.'s defense, we did have some discussion about the reasons for these 
hooks, although that seems like ages ago and I would need to dig through the 
archives (my inbox?) to find the reasoning for each.

However, I don't remember being very comfortable with the hooks back them 
largely due to uncertainty about how we were treating kdbus with respect to 
subjects/objects.  I think it's worth restarting that discussion now before we 
nit pick the patches themselves.

-- 
paul moore
security @ redhat