Re: [Ksummit-discuss] [TECH TOPIC] Firmware signing

[Andy Lutomirski <luto@amacapital.net>]

On Wed, Aug 12, 2015 at 3:45 PM, David Howells <dhowells@redhat.com> wrote:
> David Woodhouse <dwmw2@infradead.org> wrote:
>
>> No. Just use a *hash* of the acceptable signing cert(s)¹. Note that the
>> SKID is *usually* a hash of the public key, but isn't guaranteed to be
>> so, so using the SKID to specify the acceptable signing cert isn't
>> secure.
>
> True.  That's one of the reasons I don't like SKIDs - the specification is
> very vague and non-enforcing.  We would need a 'standard' for how to hash the
> public key data.  Some types of public key, for example, have more than one
> integer.  I wonder if we could just take the PGP method as the standard -
> though that does require extra elements.

This is a major benefit of modern (or modernish) ECC schemes: the
public key is pretty much unambiguously represented as a byte string.

(IIRC P-256 has some variable length garbage for part of it, but we
could pick any of the common representations and stick with it.  I
think that pretty much everyone agrees on something along the line of
0x04 plus the big-endian representation of the point in some standard
coordinate system, for 65 unambiguous bytes.)

--Andy