[noise] Resumption PSKs

[Trevor Perrin]

On Wed, Jun 20, 2018 at 3:25 PM, str4d <str4d at i2pmail.org> wrote:
>
> Sounds good to me; let's go with HKDF everywhere. That means
> create_chains() is now:
>
> create_chains(labels):
>     if not (enable_ask and ask_master):
>         return Err()
>     ask_chains = {}
>     for label in labels:
>         ask_chains[label] = HKDF(ask_master, h || label)
>     delete ask_master
>
> I'll update my branch to implement this. I will also go ahead with using
> the above key chain derivation in NTCP2 (which is being finalized this
> week).

Cool, I started a separate thread just to summarize this and maybe
draw comments from people who haven't followed all these twists:

https://moderncrypto.org/mail-archive/noise/2018/001713.html

Is there anything else you'd like to stabilize in Noise soon?  For
example, the "Ephemeral key obfuscation" discussion didn't really
reach a conclusion?

Trevor