Re: continued 1364 BTF work

From: Brad Pierce (Brad.Pierce@synopsys.com)
Date: Thu Oct 14 2004 - 10:31:00 PDT

  • Next message: Steven Sharp: "Re: continued 1364 BTF work"

    Steven,

    Jonathan Bradford's proposal, at the end of

                  http://www.eda.org/sv-bc/hm/1910.html

    seems like one reasonable approach to consider. The idea is to
    allow a packed type identifier, enclosed in angle braces < >,
    to be used in net declarations where now only signing and packed
    dimensions are allowed.

       wire_type ::=
            [ signing ] { packed_dimension }
       | '<' type_identifier '>'

       net declaration ::=
          net_type_or_trireg
             [ drive_strength | charge_strength ]
             [ vectored | scalared ]
             wire_type
             [ delay3 ]
             list_of_net_decl_assignments ';'
     
    -- Brad

    -----Original Message-----
    From: owner-btf@boyd.com [mailto:owner-btf@boyd.com]On Behalf Of Steven
    Sharp
    Sent: Wednesday, October 13, 2004 5:31 PM
    To: sharp@cadence.com; krishna@synplicity.com
    Cc: btf@boyd.com
    Subject: Re: continued 1364 BTF work

    >I was attending the datatypes subgroup sometime back. Can you be a
    >little more
    >specific what you mean by "to get datatypes on nets specified". Does
    >this include
    >the datatypes already defined under the P1800 spec ??

    Yes. It would not involve adding any new datatypes, but rather allowing
    nets of the datatypes already defined under P1800 (and probably only the
    most useful and sensible subset of those, given schedule constraints).
    The main difference between P1800 and the datatype proposals being
    considered by the datatype subgroup, was this useful ability to put
    datatypes on nets.

    Steven Sharp
    sharp@cadence.com

    This archive was generated by hypermail 2.1.4 : Thu Oct 14 2004 - 10:21:29 PDT and
    sponsored by Boyd Technology, Inc.