Re: [Dcrup] I-D draft-ietf-dcrup-dkim-crypto-06

Jeremy,

I believe the exact thing you describe (whether to pass original headers to
the library to hash and sign) was discussed in many back-and-forths. That
is the obvious alternative to re-hash, so the entire discussion was about
that, vs. re-hash.

denis

On Wed, Nov 22, 2017 at 3:25 AM, Jeremy Harris <jgh@wizmail.org> wrote:

> On 22/11/17 04:35, John Levine wrote:
> > In article <00d888c1-64c9-9f26-0426-fbbb17fc5bdc@wizmail.org> you write:
> >> On 21/11/17 22:03, John Levine wrote:
> >>> Keeping in mind that you're responding to a message I wrote two months
> ago ...
> >>>
> >>> In article <383fef94-84c4-9b54-5566-a6fa1279aa38@wizmail.org> you
> write:
> >>>> Wouldn't it be more aesthetically pleasing to decouple, for dkim
> >>>> a=ed25519-sha256 signing, the hash used for the body (sha256 as
> >>>> specified by the 'a' tag) from the hash used for signing headers
> >>>> (sha512, I think, but whatever the libraries have tied to
> >>>> Ed25519 signing)?
> >>>
> >>> No.
> >>
> >> I think you need to further support your position.
> >
> > You might want to review some of the 100 messages posted to this list
> > since the one you were responding to.
>
> > PS: look for the ones about many libraries don't plan to provide the
> > pure version of ed25519, and how we agreed to deal with that.
>
> I already did.  All assumed immediately the re-hash, once it was
> accepted that we had no chance of the libraries providing a pure
> signing facility.  I didn't see one that suggested that the headers
> be passed in original unhashed form to the library hash-and-sign.
>
> Is your position "it's too late, we already decided"?  You
> final sentence above implies that, but perhaps I misread you.
> --
> Cheers,
>   Jeremy
>
> _______________________________________________
> Dcrup mailing list
> Dcrup@ietf.org
> https://www.ietf.org/mailman/listinfo/dcrup
>