RE: [sv-ec] behavior of program with no initial blocks

From: Steven Sharp <sharp_at_.....>
Date: Tue Aug 29 2006 - 13:52:09 PDT
>From: "Arturo Salz" <Arturo.Salz@synopsys.com>

>I believe it should never implicitly exit (as it never started). The
>effect of the implicit exit would only be visible in one corner case:
>when the program with no initial blocks is the only program in the
>system, and, in that case I believe it would be undesirable to end the
>simulation at time 0.

Sounds fine to me.

I do see another potentially visible effect, at least in some future
revision of the language.

If fork..join_none is allowed in functions, then a variable initializer
could spawn subprocesses, even though there are no initial blocks.  If
a $exit is supposed to terminate those subprocesses, then it is visible
that a $exit occurred.  A similar situation could occur for subprocesses
spawned by function calls from continuous assignments, if that were
allowed.  That would be less likely to come up, as spawning subprocesses
from continuous assignments doesn't seem very sensible.

Steven Sharp
sharp@cadence.com
Received on Tue Aug 29 13:52:13 2006

This archive was generated by hypermail 2.1.8 : Tue Aug 29 2006 - 13:52:18 PDT