Re: [tcpinc] new drafts of TCP-ENO and tcpcrypt

[iang <[e-mail]>]

On 08/10/2017 00:33, Gregorio Guidi wrote:

> ...
> Having ECDHE-Curve25519 and ECDHE-Curve448 as MTI was suggested, but 
> the lack of implementations for Curve448 was mentioned. Unfortunately 
> this is still an issue: there are implementations available but no 
> stable and well-proved implementation of Curve448 is there yet, as 
> shown here:
> https://en.wikipedia.org/wiki/Comparison_of_TLS_implementations#Supported_elliptic_curves 
>
>
> Nonetheless, in the time passed since that exchange, the adoption of 
> Curve25519 has consolidated further, so the option to have 
> ECDHE-Curve25519 as the only MTI would not look so bad in my view.

+1.  The notion that we need two MTIs with one longer than 256 bits in 
an embedded optimistic half-measure needs some serious justification.  
And the one that fits best in kernel, isolated from external libraries, 
is likely best for attack surface.  I'd put my money on Curve25519 every 
day of the month for that.

iang