On Wed, Nov 16, 2022 at 9:37 AM Roberto Sassu
[off-list ref] wrote:

On Tue, 2022-11-15 at 21:35 -0500, Paul Moore wrote:

quoted

If you want to somehow instrument the LSM hook definitions (what I
believe to be the motivation behind patch 3/4) to indicate valid
return values for use by the BPF verifier, I think we could entertain
that, or at least discuss it further, but I'm not inclined to support
any runtime overhead at the LSM layer for a specific LSM.

Ok, yes. Patches 1-3 would help to keep in sync the LSM infrastructure
and eBPF, but it is not strictly needed. I could propose an eBPF-only
alternative to declare sets of functions per interval.

More or less, I developed an eBPF-based alternative also for patch 4.
It is just a proof of concept. Will propose it, to validate the idea.

Thanks, I think that might be the best approach.  Also, please
resubmit patches 1/4 and 2/4 with those small changes; those are nice
improvements that just need a couple of small tweaks to be acceptable
:)

-- 
paul-moore.com