"The systems are highly constrained. and don't have AES in hardware."  You
know this for an absolute fact?  AES has been a standard cell in cell
libraries for decades.  If it is there, then there is a mechanism which is
already evaluated, and fast. [note I'm not talking about AES 256, just
normal AES 128 running in OFB mode.]

[I'm not super thrilled about trading one unevaluated algorithm for another
- SIPHASH]

Deb

On Mon, Sep 22, 2025 at 10:36 AM Alan DeKok <[email protected]> wrote:

> On Sep 22, 2025, at 9:55 AM, Deb Cooley <[email protected]> wrote:
> > I have no earthly idea why one would propose a protocol, TLS, for a
> > function such as this.  I wouldn't imagine that TLS would be fit for
> > purpose in this use case.
>
>   TLS was used just as an example of a "solution" which would be too high
> overhead.
>
> > The question is how to generate bits in a way that is only known to the
> two
> > endpoints, and do that fast enough to provide a reasonably unguessable
> > value to shove into packets.  Classically, a stream cipher (AES in OFB
> mode
> > for example) is much faster than normal block cipher modes (the normal
> AES
> > GCM).  If the CPUs in question have the AES standard cell in hardware, it
> > should be fast and require less implementation to get wrong. CHA-CHA is
> > also a stream cipher, but it is normally implemented in S/W, so I'm not
> > sure how fast it is.
>
>   The systems are highly constrained. and don't have AES in hardware.
> CHA-CHA is a cipher, and isn't useful here.
>
>   We need a way to authenticate the packet contents.  Right now, that's
> done with MD5 as Hash(packet + key).
>
>   Failing that, the idea is to generates a stream of pseudo-random numbers
> based on shared / secret state.  That's the ISAAC method which we propose
> here.
>
>   Paul suggested SipHash.  Perhaps it is fast enough to replace the use of
> MD5 in BFD.   But doing that would involve a new document.
>
> > If there is an instance of using ISAAC in this use case, there will be
> > gigantic warnings about how this is NOT a general purpose option.  I'm
> not
> > a fan, nor is any other Sec AD I've ever met, of picking an algorithm out
> > of the blue and using it without evaluation, no matter how old it is.
>
>   Sure.
>
> > Before I answer the other comments, I have some homework to do (where
> does
> > RFC5880 get/generate/negotiate keys since apparently I read something
> > wrong).
>
>   The keys are not generated.  The keys are really shared secrets entered
> by humans, in an admin interface.  i.e. ASCII text.
>
>   The keys are not negotiated.  There is no way to signal "you sent me key
> X, but I want to use key Y".  BFD just proposes a key ID.  The other end
> either uses the corresponding key, or else it tears the session down.
>
>   While there are provisions for using multiple keys, in practice, there's
> really only one shared secret for any BFD peers which communicate.
>
>   Alan DeKok.
>
>

Reply via email to