[quicwg/base-drafts] ClientFinished is HoL blocking, but has no normative text (#2572)

[ianswett <[e-mail]>]

I'm not saying I want the CFIN to be HoL blocking, but previous discussions have established it is.  I don't think the current TLS text is sufficiently clear(and normative) about that.

Section 4.1.1 has:
" Important:  Until the handshake is reported as complete, the
      connection and key exchange are not properly authenticated at the
      server.  Even though 1-RTT keys are available to a server after
      receiving the first handshake messages from a client, the server
      cannot consider the client to be authenticated until it receives
      and validates the client's Finished message.

      The requirement for the server to wait for the client Finished
      message creates a dependency on that message being delivered.  A
      client can avoid the potential for head-of-line blocking that this
      implies by sending a copy of the CRYPTO frame that carries the
      Finished message in multiple packets.  This enables immediate
      server processing for those packets."

https://tools.ietf.org/html/draft-ietf-quic-tls-19#section-4.1.1

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/2572