Re: [pkcs11] New Mechanisms subgroup Part 3 or 4, new AES

[Michael StJohns <msj@nthpermutation.com>]

On 5/1/2013 11:01 AM, mark powers wrote:
> > I think that adding a mode simply because you don't want to move bits 
> > around is probably a bad idea.
>
> I believe there is value in being able to connect mechanisms mentioned 
> in a spec to mechanisms supported
> by PKCS11. Not everyone is going to dig deeper to figure out that 
> mechanism X is the same as Y when
> certain parameters are set. I'm obviously wearing "marketing" hat today.

The issue here is there are lots of mechanisms that use truncated hashes 
or MACs.  Rather than put in possibly 100 new mechanisms, I'd much 
rather generalize the mechanisms so that we can subsume all of those.

So instead of CKM_AES_XCBC_MAC_96 let's do CKM_AES_XCBC_MAC_GENERAL as 
that seems to be the common approach.

In the same way, I'm thinking that it might be useful to define a 
generic signature mechanism in which you specify separately the summary 
(hash mechanism with padding) and the actual public key operation as 
part of the mechanism info.  That would give a path for new hash/PK 
combos without waiting for PKCS11 to act.

Mike