RE: enhancement/386: ANALYZED - user defined functions on instantiations

From: Scott Cranston (cranston@cadence.com)
Date: Wed Aug 06 2003 - 07:50:02 PDT

  • Next message: Brad Pierce: "RE: enhancement/386: ANALYZED - user defined functions on instantiations"

    Precedence: bulk

    The following reply was made to PR enhancement/386; it has been noted by GNATS.

    From: "Scott Cranston" <cranston@cadence.com>
    To: <etf-bugs@boyd.com>
    Cc:
    Subject: RE: enhancement/386: ANALYZED - user defined functions on instantiations
    Date: Wed, 6 Aug 2003 10:48:21 -0400

     Only constant *expressions*. Function calls are not assumed to be
     non-constant. I suppose we need a definition of a constant function
     here, and then add this restriction to the original proposal.
     
     
     From the LRM section on defparams:
     
     ------
     The expression on the right-hand side of the defparam assignments shall
     be a constant expression involving
     only numbers and references to parameters.
     ------
     
     There is no similar restriction mentioned in the instantion override
     sections, but one would assume it is implied.
     
     
     +++-----Original Message-----
     +++From: Shalom Bresticker [mailto:Shalom.Bresticker@motorola.com]
     +++Sent: Wednesday, August 06, 2003 10:30 AM
     +++To: etf-bugs@boyd.com
     +++Subject: Re: enhancement/386: ANALYZED - user defined
     +++functions on instantiations
     +++
     +++
     +++Precedence: bulk
     +++
     +++The following reply was made to PR enhancement/386; it has
     +++been noted by GNATS.
     +++
     +++From: Shalom Bresticker <Shalom.Bresticker@motorola.com>
     +++To: etf-bugs@boyd.com
     +++Cc:
     +++Subject: Re: enhancement/386: ANALYZED - user defined
     +++functions on instantiations
     +++Date: Wed, 06 Aug 2003 17:19:44 +0300
     +++
     +++ Don't we already allow use of constant functions in
     +++parameter overrides?
     +++
     +++ Shalom
     +++
     +++
     +++ etf@boyd.com wrote:
     +++
     +++ > Precedence: bulk
     +++ >
     +++ > First, a correction to the proposal. What the AMS folks
     +++are actually calling for here is the ability to call user
     +++defined *HDL* functions in parameter overrides. For example,
     +++assume we have a Verilog function as follows: > > function
     +++real f_mod(a, b) > input a; real a; > input b; real b; > begin
     +++ > f_mod = (a-b*((a+0.5)/b)) ;
     +++ > end
     +++ > endfunction
     +++ >
     +++ > Then what we want to be able to do is use this function
     +++to set the value of a parameter as follows: >
     +++ > parameter real aa= 10.0;
     +++ > parameter real bb=3.0;
     +++ > my_mod #(.r(a*f_mod(aa,bb)+b)) r1 (n1,n2);
     +++ >
     +++ > The need for this arises from the need to specify more
     +++complex parameter expressions in analog netlists as
     +++represented in Verilog-AMS or Verilog. > > It should be
     +++noted that this is very similar to what is currently allowed
     +++in VHDL-AMS in setting values of generics. This may or may
     +++not be a good reason to support it in Verilog as well. > >
     http://boydtechinc.com/cgi-bin/issueproposal.pl?cmd=view&pr=386
      
      --
      Shalom Bresticker
     Shalom.Bresticker@motorola.com
      Design & Reuse Methodology Tel: +972 9
     9522268
      Motorola Semiconductor Israel, Ltd. Fax: +972 9
     9522890
      POB 2208, Herzlia 46120, ISRAEL Cell: +972 50
     441478
      
      
      

    This archive was generated by hypermail 2.1.4 : Wed Aug 06 2003 - 07:50:55 PDT and
    sponsored by Boyd Technology, Inc.