On Thu, Feb 16, 2012 at 8:20 AM, Phillip Hallam-Baker <[email protected]> wrote:
> It would be really nice if there was some way to audit RNGs algorithmically...

Separate the HW RNG and other entropy gathering parts from the rest of
the RNG (i.e., the entropy pool, the mixer, the extractor), and you
provide a per-device seed for testing.  To test you put the device in
test mode so that the only entropy will be from the test seed, thus
making the RNG completely deterministic.  The production mode can also
use the test (or another per-device) seed and date/time of boot to
initialize the entropy pool, just in case the HW RNG get stuck on all
ones or all zeros (or all nines).  Testing production mode can also be
done by, e.g., statistical analysis of the RNG outputs (and inputs)
under various operating conditions (e.g., different temperatures,
etc...), and by extracting a copy of the entropy pool contents once to
check that the RNG is not deterministic from that point forward
(because new entropy gets mixed in).  But you have to make sure that
any test modes can't be enabled without tampering with the physical
device.  Once the device is sealed you should not be able to test the
RNG in any way other than by statistical analysis of its outputs.

Nico
--
_______________________________________________
therightkey mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/therightkey

Reply via email to