From: Michael McNamara (mac@verisity.com)
Date: Mon May 17 2004 - 07:44:00 PDT
-- On May 17 2004 at 08:40, Vassilios.Gerousis@infineon.com sent a message:
> To: Shalom.Bresticker@freescale.com, dennisb@model.com, etf@boyd.com
> Subject: "RE: enhancement/350: PROPOSAL - Deprecate configs in
> Verilog source"
>
> Dennis email points out certain facts about backward compatibility of
> IEEE standard versions. Does the committee plan to build an IEEE 2005
> version that is not backward compatible with IEEE 2001 version? Dennis
> pointed out that there are existing models and tools that support the
> current Verilog 2001 version. The question that should be answered by
> this committee, do you want to allow non-compliance to existing
> standards (i.e. Verilog 2001)? Shall tool implementers wait for Verilog
> 2010 to ensure that certain functionality in 2005 is not deprecated?
>
> Backward compatibility is an extremely serious matter and should be
> addressed in an "equitable" manner that give confidence to users and
> tool vendors alike. By obsolescing backward compatibility, you are
> rewarding non-early adopters of Virology 2001 rather than awarding early
> adopters.
And yet we are also looking at the fact that "logic" and "bit" and
"next" and hundreds of other words have been approved for use with an
unconstrained scope as identifiers for the last twenty years in
Verilog, and yet we have a proposal to take these hundreds of key
words away from the users; making their existing models and tools which
support them obsolete.
I suggest it might be useful for you to study proposal 287 which looks
to provide a method to reconcile these issues in a beneficial manner
for all parties.
Michael McNamara, Chairman, IEEE 1364 Working Group <mac@verilog.com>
Sr VP Technology, Verisity Design <mac@verisity.com>
W 650-934-6888 F 650-934-6893 M 408-930-6875
This archive was generated by hypermail 2.1.4
: Mon May 17 2004 - 07:13:20 PDT
and
sponsored by Boyd Technology, Inc.