RE: migration: 1364-2001 vs SV

Hello Sri,

Sri Chandra wrote:
> > Geoffrey Coram Wrote:
> > How much work is there in a 2001 migration? My sense is that most of the changes in the syntax Graham
> > has worked on were changes from syntactic to semantic restrictions: ie, everything that worked before
> > will still work, and everything that was disallowed before will be disallowed (though for a different
> > reason).
>
> The work that's remaining with regards to 2001 migration is the following:
> * reviewing changes done by Graham to integrate the compact modelling BNF into the updated syntax. I
> don't think this is a big change as we retain most of the semantic restrictions from the original
> proposal, but just making it consistent and inline with the digital syntax.
> * Migrating the changes into the individual chapters. I think this would be a bigger change in the
> documentation to update all the syntax boxes and references to reflect the new BNF. Also part of the
> change is to document the semantic restrictions on the individual items in a better fashion. I think
> this will be a bigger change.

Migrating the merged AMS/2001 syntax into the body of the LRM is not simple as it initially appears. It is not as straight forward as updating the syntax blocks and documenting a few new semantic checks. A number of the sections will require further changes based on a more detailed syntax (e.g. net declarations).

> > Is the 1364 committee amenable to accepting a donation?
>
> I am not sure. I need to get in touch with the IEEE 1364 committee chair and see whether this is
> workable. Also, as part of updating the BNF, there were some updates to the digital syntax of
> IEEE2001, to be able to merge the two languages from a syntax point of view, and this needs to be
> ratified by the 1364 committee. I think the changes and the number of them were very minor if I
> remember correctly, but Graham can add to this and correct me if I have mistated it.
> We should also look into what parts of AMS could be integrated piece by piece, such as the trig
> functions, without disrupting Verilog.

There should be no changes to the IEEE2001 BNF, only extensions to the existing digital syntax items to include analog syntax.

If the AMS LRM is donated to the 1364 committee, will they merge the IEEE2001 and AMS documents into one as some point in the future? If so, will they accept the document as it is or will they want it to be more closely aligned with the IEEE2001 document format and structure?

Regards
Graham
Received on Fri Jan 21 03:02:23 2005