Re: [TLS] Server Signature Algorithms

[Wan-Teh Chang <wtc at google.com>]

On Sun, Nov 1, 2009 at 1:34 PM, Michael D'Errico <mike-list at pobox.com> wrote:
> So instead of changing the existing signature_algorithms extension,
> nothing would break by creating a new server_signature_algorithms
> extension.  Client support would be advertised by sending an empty
> extension to the server.  The server would show support by replying
> with its list.

Michael: I agree this would be the safest solution which
eliminates the concern of breaking some TLS 1.2 implementation,
at the cost of adding a new TLS extension.

This would be an example of a "server-oriented" extension
described in Sec. 2.3, page 7 of RFC 4366.

Nikos: GnuTLS is one of the first TLS 1.2 implementations.
How does GnuTLS deal with this issue now?  Does it compute
hashes of all the hash algorithms it supports on the handshake
messages, or does it buffer the handshake messages until
the hash algorithm has been selected?

Wan-Teh