CKD_NULL

Hi Bob,
      

      I missed the first part of that discussion.  Was that an question
      that came in through the mailing lists?  Or a question that came
      to you directly through RedHad?
      

    
    It came directly I've been meaning to post the question to the
    mailing list and never got to it.
    

    
      

      I had to go back through some old emails... but there is the
      BSI-TR-0311 standard which does define ECDH where both X and Y
      coordinates are returned from the ECDH primitive, and where it
      does appear to use the X and Y  coordinates as input to the KDF.
      

      

      We (Safenet) were asked about our compliance with TR-0311, and I
      pointed out that our products only return X, but I never got any
      more response from them on this topic.
      

      

      Not sure if this is of any interest to you or not.
      

    
    Hmm it is. I was thinking of which way to go post 3.0 to get what we
    need:
    

    1) a new KDF that returns X and Y.
    

    2) a new mechanism that returns X and Y.
    

    3) full set of mechanisms needed to support McCallum/Relyea.
    

    

    This input leads me to to think 1 and 2 would be viable options.