Re: $root and "top level" instantiations


Subject: Re: $root and "top level" instantiations
From: Gordon Vreugdenhil (gvreugde@synopsys.com)
Date: Thu Aug 15 2002 - 13:05:49 PDT


Peter Flake wrote:
>
> Hi Gordon,
>
> At 03:03 PM 8/12/02 -0700, Gordon Vreugdenhil wrote:
> >I would appreciate some feedback on the intent of $root
> >and "top level" instantiations as described in Sect 12.2
> >of the SystemVerilog 3.0 specification.
> >
> >The relevant text in the LRM is in Section 12.2 (on pg 51):
> >
> >"The order of elaboration shall be: First, look for explicit
> >instantiations in $root. If none, then look for implicit
> >instantiations (i.e. uninstantiated modules)..."
> >
> >... snip ...
> >
> >"A module can be explicitly instantiated in the $root top-level. All
> >uninstantiated modules become implicitly instantiated within the top
> >level, which is compatible with Verilog."
>
> The qualification "If there is no explicit top level instantiation, then
> all uninstantiated modules .... " was removed in draft 5.
>
> >1) If you explicitly instantiate one or more modules in $root, do
> > uninstantiated modules get implicitly instantiated? The two
> > quotes seem to imply contradictory answers.
>
> Not instantiated according to the original spec.

Peter, thanks for the response.

I am interpreting your above comment to mean that since there is
now an apparent conflict in the spec, you are appealing to
an earlier draft to resolve the conflict and that the above
resolution should be re-introduced in some manner into the spec.
Is that a fair interpretation?

> >2) Do implicitly instantiated modules get instantiated in $root or
> > in some "top level" along with $root? Or does this depend on
> > whether $root has any explicit instantiations? What if $root
> > is non-empty (ie. has global data or type declarations) but
> > has no instantiations?
>
> If a module is implicitly instantiated, its hierarchical name given by %m
> should not contain $root for Verilog compatibility.

Given your answer to (1), this makes sense to me.

Thanks again.

Gord.

-- 
----------------------------------------------------------------------
Gord Vreugdenhil                                 gvreugde@synopsys.com
Staff Engineer, VCS (Verification Tech. Group)   (503) 547-6054
Synopsys Inc., Beaverton OR



This archive was generated by hypermail 2b28 : Thu Aug 15 2002 - 13:09:11 PDT