Rekey regression test: How is GCM magical?

[Markus Friedl]

Am 16.11.2013 um 17:43 schrieb Christian Weisgerber <naddy at mips.inka.de>:

> Revision 1.10 of regress/usr.bin/ssh/rekey.sh has added this:
> 
> # GCM is magical so test with all KexAlgorithms
> [...]
> 
> How is GCM magical?

just because there is no MAC. so it makes sense to trigger more than
the default KEX.

> Is chacha20-poly1305 equally magical and needs testing with all
> KexAlgorithms as well?

in this sense, yes.

> 
> (chacha20-poly1305 doesn't work with diffie-hellman-group1-sha1
> because the latter can't provide the required 512 bits of key
> material.)

why should it not work? in this case the key gets expanded,
even if the security margin of the kex is smaller.

-m