Re: [TLS] DSA should die

[Dave Garrett <[e-mail]>]

On Wednesday, April 01, 2015 09:15:08 pm Dave Garrett wrote:
> There's plenty of space in the current registry if you want to start
> over. Just say 0xD000 and up is for TLS2 suites, and all below are TLS1.
> (at this point it's definitely time for 2.0) Define TLS2 codepoints for
> valid combinations for TLS2 and the TLS1 would be invalid to negotiate
> for TLS2 but in there for TLS1. This would cost a few more bytes to list
> AES-GCM in both TLS1 & TLS2 format in the same list, but resorting to a
> whole new system would add more bytes anyway, along with another registry
> to keep track of.
[...]
> I'm definitely in favor of reducing the combinatorial explosion in suites
> and moving hash, KEA, & etc. into a separate negotiation. Instead of an
> extension, you could just do the same thing again, really. Reserve a
> sizable block of the 16-bit suite space for hashes and another for KEA
> and list them all in the same cipher_suites array. TLS1 would ignore
> them, and TLS2 would just see points between [a,b] are hash, [b+1,c] are
> KEA, etc. So, new registries in the existing space, just dumped all into
> the same pile.

On Wednesday, April 01, 2015 10:17:23 pm Martin Rex wrote:
> But how about using the Cipher Suites registry in a more creative
> fashion.
> 
> For TLSv1.3, we could do the negotiation through he cipher suites list
> more like this:
> 
>    0x10,0xXX    specifies a key exchange algorithm (XX) 256 codepoints
>    0x12,0xYY    specifies an authentication algorithm (YY) 256 codepoints
>    0x14,0xZZ    specifies a symmetric encryption scheme (ZZ) 256 codepoints
>    0x16,0xQQ    specifies a mac algorithm (QQ) 256 codepoints
>    0x18,0xPP    specifies a PRF algorithm (PP) 256 codepoints
> 
> (with a little room in between if we ever exceed the 256 codepoints)

Yes, that's basically what I'm proposing. Myself, Nico, and you are all just describing the same sort of scheme differently. Yours is the clearest thus far, as it covers everything we might want here.

The only difference I suggest is picking points all above a threshold that separates new from old so filtering out the old system trivial. That looks to be around 0xD000, or 0xC100 if you want to get closer to the current highest assigned value.

https://www.iana.org/assignments/tls-parameters/tls-parameters.xhtml#tls-parameters-4


Dave