RE: [sv-ac] SVA: What is the defintion of "identical clock"?

So what does "syntactically identical" mean? To me, from a user's
perspective, that is just as ambiguous as the original wording. I still
have no idea what is required when assertions refer to "identical clocks".
The LRM needs to define this term with total clarity for both implementers
and users.

Stu
~~~~~~~~~~~~~~~~~~~~~~~~~
Stuart Sutherland
stuart@sutherland-hdl.com
+1-503-692-0898
  

> -----Original Message-----
> From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On
> Behalf Of Eduard Cerny
> Sent: Wednesday, October 27, 2004 8:12 AM
> To: VhdlCohen@aol.com; sv-ac@eda.org
> Cc: stuart@sutherland-hdl.com; cliffc@sunburst-design.com
> Subject: RE: [sv-ac] SVA: What is the defintion of "identical clock"?
>
> Defining semantic identity could be quite tricky. Perhaps we
> could just state that the clocks should be syntactically
> identical. The tools can do better if so desired.
>
> ed
>
>
> > -----Original Message-----
> > From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org]On Behalf Of
> > VhdlCohen@aol.com
> > Sent: Tuesday, October 26, 2004 8:39 PM
> > To: sv-ac@eda.org
> > Cc: stuart@sutherland-hdl.com; cliffc@sunburst-design.com
> > Subject: [sv-ac] SVA: What is the defintion of "identical clock"?
> >
> >
> > The LRM discusses at great length the issue of "identical" or
> > "non-identical" clocks, but the LRM fails anywhere to
> define what is
> > an "identical clock". LRM Std P1364 also fails to define identical.
> > Specifically, Identical in what way? The exact same signal?
> The same
> > period? Identical in some other way? Is clk1 aliased to clk2
> > identical? Are signals connected via a module port so that they are
> > the same signal? Does "assign clk1=clk2;" create identical clocks?
> >
> > Below is a typical reference in the LRM to identical clocks.
> > LRM 17.12.1 Multiply-clocked sequences states the following:
> > "
> > @(posedge clk0) sig0 ##1 @(posedge clk1) sig1 ....
> > If clk0 and clk1 are identical, then the clocking event does not
> > change after ##1 and the above sequence is equivalent to the
> > singly-clocked sequence @(posedge clk0) sig0 ##1 sig1
> >
> > Thanks,
> > Ben Cohen
> > --
> >
>
>
>
Received on Wed Oct 27 10:36:17 2004