Subject: Re: [sv-ac] Informal proposal for #15 ref interface scope.
From: Adam Krolnik (krolnik@lsil.com)
Date: Mon Sep 08 2003 - 06:53:54 PDT
HI Surrendra;
Thanks for reading my proposal.
>But I'm not sure about the assert statement. Two problems I see:
>1) Importing is only done for definitions, not executable statements
The need is for the executable statements as well as the assertion of the
property.
>2) If the assert statement is imported, then it will get triggered twice, one for the
>interface, and another for modport.
Good question - I forgot to explain my intent.
My intent is that assertions imported into a modport that becomes instantiated will
operate and execute the action block from within the scope where the modport was
instantiated.
If an assertion/action block is imported in an instantiated modport, it should
be either removed from the interface scope or not actived to run in the
interface scope. I though about saying that regardless of the instantiation of
the modport, the assertion should not be actived in the interface scope. But
that would not allow the assertions to execute if one is using the interface
with tasks as drivers instead of modports.
The intent is that the import of the assert/cover statement moves the activation from
the interface to the child scope, for the purposes of reporting the scope that created
the error (assuming that the user chose the correct to import the assertion.)
Thanks.
Adam Krolnik
Verification Mgr.
LSI Logic Corp.
Plano TX. 75074
Co-Author "Assertion Based Design"
This archive was generated by hypermail 2b28 : Mon Sep 08 2003 - 06:54:45 PDT