Subject: Re: Proposed PAR for Verification Language Study Group
From: Paul J. Menchini (mench@mench.com)
Date: Tue Feb 04 2003 - 06:31:28 PST
Don,
> Even after all of the previous round of discussion on the topic and
> with the PAR itself, I am left wondering:
Thanks for your questions. I'll provide my personal perspective below.
> 1. How do we avoid a collision between this PAR and the work already
> underway at Accelera?
I'm not sure what you mean by collision, but in general, I do not
believe that the DASC can force our control over work done in similar
scopes by non-IEEE-sanctioned groups (nor are we required to). I would
hope that, through common membership and interests, the groups will work
synergistically or become part of our efforts. (Historically, this has
happened in the majority of cases.) But, I do not see a lack of
cooperation (should that occur) as a problem that we can or must
address.
> 2. How does e and how should any functional verification language or
> formal property language or system verification language or ...,
> relate to Verilog and VHDL?
This topic was discussed extensively by the SG when formulating their
draft PAR. While I do not believe that any definitive conclusion (as to
approach) was reached (but several were proposed and discussed), the
group does believe that the topic is important and does intend to
address the topic.
> 3. Is there a clear and complete statement of the problem that this
> PAR is to address that separates this work from other efforts?
I believe that the scope statement of the draft PAR expresses best the
definition of the work to be done. Whether this statement is clear and
complete is subjective and, indeed, is one of the reasons that this
draft (as is true with all drafts) is being presented to the SC for
approval.
As to separation, again I believe that your question presupposes that
the DASC either can or must somehow "carve out" areas of work that are
not potentially addressed by others or can somehow prevent non-IEEE
groups from working in areas within DASC's scope. I do not believe that
DASC can do either.
What DASC does, in distinct contrast to other some standardization
groups, is to provide standardized technology. We do not dictate or
prescribe the use of standardized technology. Thus, an IEEE standard
does not preclude the creation and use of other (IEEE or non-IEEE)
standards with similar scopes but with distinct characteristics. (For
example, 1076 and 1364.) So, there's nothing to prevent other
verification projects from coming forward to the DASC and gaining
approval.
Note that, the DASC itself provides separation by context. Within the
IEEE, we have exclusive domain over EDA standards (through the PAR
approval process). Thus, when potentially overlapping standardization
efforts are proposed, we have control over them and can now require
cooperation or alignment.
But, we lack the ability to do so for non-IEEE groups. If we decide
that there can be no potential overlap between DASC efforts and the
efforts of others, the only thing that I believe that we can do is to
not work in those areas or abandon our efforts when "competition"
develops. We cannot force them to join us or abandon their efforts. I
do not think that this is how the DASC should operate.
> In net, is there more that the DASC can and should do in attempt to
> get what appear (to me) to be competing "standards" efforts on a
> common course?
Again, I do not believe that DASC can control what non-IEEE groups do in
DASC's area of interest. In the past, we've relied on common membership
to work with such outside groups and, indeed, have enjoyed good
cooperation and support from such groups. Only the future will tell
whether such cooperation will occur in this case.
Nevertheless, as you apparently feel that DASC should be working to
align its efforts with other standardization efforts with similar
scopes, I'd be willing to continue this dicussion to develop a strategy
for creating such alignment.
Paul
This archive was generated by hypermail 2b28 : Tue Feb 04 2003 - 06:32:10 PST