Re: [sv-ec] RE: [sv-bc] tf_port_item footnote 33

From: Steven Sharp <sharp_at_.....>
Date: Tue Jul 24 2007 - 15:02:52 PDT
>From: "Warmke, Doug" <doug_warmke@mentor.com>

>If a prototype contains an anonymous port declaration, then the
>out-of-block
>
>method must contain a named port declaration, and that name shall be the
>
>authoritative name for the port (used in named argument association,
>etc.).

And what do you propose to do with a call that appears before the
named port declaration, and which uses named argument association?

Would this be illegal, in which case this would require additional
text in the LRM to document, and an extra complication in the rules?
Or would all checking of the named arguments have to wait until the
out-of-block method was seen, in which case the benefits of having
the arguments in a prototype are lost?

I don't see that the minimal convenience of anonymous port declarations
warrants this extra complication.  Good programming practice calls for
the names to be provided with the prototype for documentation purposes,
even if the language didn't have named argument association.

I suspect that this anonymous syntax was added with the idea that it
was OK because C does it.  But C doesn't have named argument association.
In general, callers of a Verilog subroutine need the names of the
arguments, so the prototype should provide it.  The anonymous syntax
should be eliminated.

Steven Sharp
sharp@cadence.com


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Tue Jul 24 15:03:26 2007

This archive was generated by hypermail 2.1.8 : Tue Jul 24 2007 - 15:03:36 PDT