Re: [TLS] I-D Action: draft-ietf-tls-curve25519-00.txt

[Ilari Liusvaara <[e-mail]>]

On Mon, Jun 15, 2015 at 11:40:11PM +0200, Simon Josefsson wrote:
> Martin Thomson <[e-mail]> writes:
> 
> > On 15 June 2015 at 14:12, Simon Josefsson <[e-mail]> wrote:
> >
> > You don't have to add a 4 at the front, or mess around with forms.
> > Just define (by reference to draft-irtf-cfrg-curves, preferably) the
> > format.  We have plenty of contextual information available in TLS, we
> > just need the bytes.
> 
> I agree!  Committed as follows:
> 
> https://gitlab.com/jas/ietf-tls-newcurves/commit/e7b449ee5af70c2a0851f617ce225f93cc86d993#diff-1

+1.

But seemingly examples of encoding got stripped in process. Could be
a good idea to take the DH test vectors from cfrg-curves and also
show the resulting premaster secret (note: The hexadecimal versions
are already little-endian, as can be seen by computing mod 16 on
the decimal integer versions).

Also, could be a good idea to add Curve448 while at it too (Note: The
DH final result in testcase is wrong, the correct one seems to be:

07fff4181ac6cc95ec1c16a94a0f74d12da232ce40a77552281d282b
b60c0b56fd2464c335543936521c24403085d59a449a5037514a879d

As per https://www.ietf.org/mail-archive/web/cfrg/current/msg06770.html )


-Ilari