RE: [sv-bc] Local parameters in parameter-port-list (Mantis 1134)

Jonathan,

I am OK with your proposed changes, and the "side-effect" you describe.  The
side effect is the same as with module ports and task/function formal
arguments, where the direction is sticky.  Once declared, it affects all
subsequent ports or arguments until a new direction is defined.  It is also
similar to tasks and functions which have a default direction of "input",
until a direction is declared.  The parameter port list has a default of
"parameter" until a different constant type is declared.

While this sticky nature of declarations can be a gotcha to users, it is
consistent with the nature of Verilog.  IMHO, being inconsistent (e.g.
requiring a specific declaration order) would be a greater gotcha.

Stu
~~~~~~~~~~~~~~~~~~~~~~~~~
Stuart Sutherland
Sutherland HDL, Inc.
stuart@sutherland-hdl.com
503-692-0898
 

> -----Original Message-----
> From: owner-sv-bc@server.eda.org 
> [mailto:owner-sv-bc@server.eda.org] On Behalf Of Jonathan Bromley
> Sent: Tuesday, August 21, 2007 8:22 AM
> To: sv-bc@server.eda-stds.org
> Subject: RE: [sv-bc] Local parameters in parameter-port-list 
> (Mantis 1134)
> 
> hi BC,
> 
> The minutes of your meeting of 23 July say the following about 1134:
> 
>    Consensus that localparams be skipped when dealing with
>    override by position.  This is the same as non-ANSI-style
>    parameter usage currently enabled and relaxes requirement
>    in the current proposal that they appear at the end of the list.
>    Update wording and BNF accordingly.
> 
> This can be achieved rather easily by deleting all my proposed
> BNF changes and instead making this change:
> 
> parameter_port_declaration ::=
>     parameter_declaration
>   | local_parameter_declaration           // added branch
>   | data_type list_of_param_assignments
>   | *type* list_of_type_assignments
> 
> provided the changes in Mantis 979 are also implemented.
> 
> However, I think this has an unfortunate side-effect.
> Consider
> 
>   module foo
>    #(
>       parameter P = 3,
>       localparam L = P+1,
>       type T = logic [P:0],
>       T PT = 5
>     ) ...
> 
> Would you say that T is a localparam or a parameter?
> What about PT?  And what would users think?
> (For what it's worth, I think that both T and PT 
> are localparams, but I think users might be very 
> surprised to find they can't override T.)
> 
> This is precisely the confusion I was trying to avoid by 
> requiring all localparams to be at the end of the list.
> 
> If you're OK with it, then I'll go ahead and update the
> proposal; the new proposal will state as a prerequisite
> that the changes of Mantis 979 be implemented.
> -- 
> Jonathan Bromley, Consultant
> 
> DOULOS - Developing Design Know-how
> VHDL * Verilog * SystemC * e * Perl * Tcl/Tk * Project Services
> 
> Doulos Ltd. Church Hatch, 22 Market Place, Ringwood, 
> Hampshire, BH24 1AW, UK
> Tel: +44 (0)1425 471223                   Email: 
> jonathan.bromley@doulos.com
> Fax: +44 (0)1425 471573                           Web: 
> http://www.doulos.com
> 
> The contents of this message may contain personal views which 
> are not the views of Doulos Ltd., unless specifically stated.
> 
> -- 
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
> 
> 
> 
> 



-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.