Subject: Re: fullcase/parallel case (fwd)
From: Joseph P. Wetstein (jpw@cbis.ece.drexel.edu)
Date: Wed Jul 11 2001 - 10:34:36 PDT
I have been playing with forms that look like this:
(* rtl_synthesis, full_case [=1] [; <other non synthesis attributes>])
in other words, allowing some attributes to be associated with other
atributes, and some to not be associated. Is there ever conceivably a need
for this? If you have one attribute related to synth and another
specifically not - it creates potentially a heirarchy of attributes.
Comments?
> Joe:
>
> One of the action items from the last meeting for you to decide is the
> format of the attributes.
>
> Currently, they are of the form:
> (* full_case *)
> (* parallel_case *)
> (* full_case, parallel_case *)
>
> Cliff suggested something like:
> (* rtl_synthesis, full_case *)
> (* rtl_synthesis, parallel_case *)
> (* rtl_synthesis, full_case, parallel_case *)
>
> What do you think?
>
> My thinking is that if we want to include the "rtl_synthesis"
> keyword, then we are better off with the following syntax:
>
> (* rtl_synthesis_full_case *)
> (* rtl_synthesis_parallel_case *)
> (* rtl_synthesis_full_case, rtl_synthesis_parallel_case *)
>
> - bhasker
>
> > From: "Joseph P. Wetstein" <jpw@cbis.ece.drexel.edu>
> > Subject: fullcase/parallel case (fwd)
> > To: jbhasker
> > Date: Mon, 9 Jul 2001 13:30:00 -0400 (EDT)
> > CC: joseph wetstein <jpw@cbis.ece.drexel.edu>
> > MIME-Version: 1.0
> > Content-Transfer-Encoding: 7bit
> > X-Received: By mailgate.Cadence.COM as KAA00943 at Mon Jul 9 10:30:15 2001
> >
> > hi -
> >
> > For some reason, my email to the group bounced... perhaps you can help...
> >
> > Joe
> >
> > ----- Forwarded message from Joseph P. Wetstein -----
> >
> > >From jpw Mon Jul 9 13:28:44 2001
> > From: "Joseph P. Wetstein" <jpw>
> > Message-Id: <200107091728.f69HSfn15471@coe.drexel.edu>
> > Subject: fullcase/parallel case
> > To: vlog-synth@eda.org
> > Date: Mon, 9 Jul 2001 13:28:41 -0400 (EDT)
> > CC: joseph wetstein <jpw@cbis.ece.drexel.edu>
> > Reply-To: j.wetstein@ieee.org
> > Organization: self-affiliated
> > X-Mailer: ELM [version 2.4ME+ PL60 (25)]
> >
> > All,
> >
> > There seem to be some big issues, as I'm sure most of you know by now,
> > with the differences in simulation and synthesis when using full case and
> > parallel case attributes in case-statement design.
> >
> > While these attributes should be synthesizable, they are very dangerous.
> > The irony is that, especially in the 'case' of the parallel case (sorry
> > for the pun), it is most dangerous when it actually works correctly!
> >
> > I will go into this in more depth if it is required by the group, but it
> > seems to me that the attributes should be allowed and should be
> > synthesizable, but there should be error checking performed by the tools
> > to determine if the statements are more deterimental than not.
> > Additionally a WARNING should be issued whenever they are present. At
> > least, when a default statement is used in conjunction with a full-case,
> > and the latter is then ignored, it seems that a warning should be
> > generated.
> >
> > Do these provide some reasonable points for discussion?
> >
> > -Joe
> >
> >
> >
> > --
> > Joseph P. Wetstein, P.E.
> > j.wetstein@ieee.org
> > (707) 202-0600 fax
> > PP/ASEL & KA3VJY [Tech+]
> >
> > ----- End of forwarded message from Joseph P. Wetstein -----
> >
> > --
> > Joseph P. Wetstein, P.E.
> > j.wetstein@ieee.org
> > (707) 202-0600 fax
> > PP/ASEL & KA3VJY [Tech+]
> >
>
> J. Bhasker
> Cadence Design Systems
> 7535 Windsor Drive, Suite A200, Allentown, PA 18195
> 610.398.6312, 610.530.7985 (fax), jbhasker@cadence.com
>
-- Joseph P. Wetstein, P.E. j.wetstein@ieee.org (707) 202-0600 fax PP/ASEL & KA3VJY [Tech+]
This archive was generated by hypermail 2b28 : Wed Jul 11 2001 - 10:51:11 PDT