From: Andrew T. Lynch (drew@surefirev.com)
Date: Thu Feb 04 1999 - 10:14:34 PST
Thanks for your quick response. A couple of small issues:
Adam Krolnik writes:
> Good afternoon Drew:
>
> The proposal I read on the 1364 reflector has an additional sentence
> fragment on escaped names within a hierarchical reference. Is it not
> in the document you are reading.
Where is the reflector? I have not visited it yet.
> The algorithm stated for resolving hierarchical references is correct.
> To pharaphrase:
>
> Examine the current scope (named block/module) for the reference.
> If not found examine the parent's scope. Recurse until found or
> at root.
What happens if you go up, match, start downward, and then fail to
match. Do you go up again and try over?
module top;
mid1 m1();
mid2 mid1();
bot b1();
endmodule // top
module mid1;
reg x;
bot b1();
endmodule // mid1
<p>module mid2;
reg y;
endmodule
module bot;
initial $display(mid1.y);
endmodule
Using upward reference, mid1.y first matches module mid1, which does
not have a "y". Is this an error, or does it then go up and find
"top.mid1.y"?
> This correctly describes what answer you should get in your example. The
> answer is to consider hierarchical names as first an instance name
> and second a module name. The instance name found is used.
>
> I agree we need to say that module_instance_name can be repeated:
>
> [module_instance_name .]+ item_name.
>
> Maybe it should be called instance_module_name to reflect the precedence?!
>
I read "module_instance_name" as the name of a module instance, as
opposed to "module or instance name" There is also a reference to
"module_identifier" I take to mean "module definition name".
Thanks!
Drew
BTW, how's the new baby?
-- Andrew T. Lynch, Chief Zymurgist, SureFire Verification (408)374-4100 x301
This archive was generated by hypermail 2.1.4
: Mon Jul 08 2002 - 12:53:25 PDT
and
sponsored by Boyd Technology, Inc.