Subject: Re: [sv-ac] R29a/b Optional or mandatory name for assertions/properties/assumptions.
From: Simon Davidmann (simond@co-design.com)
Date: Tue Sep 17 2002 - 23:32:49 PDT
I think that they should be optional, and that tool can a) choose to auto
generate one if it is not named and a name is given, or b) tool can decide
to ignore un-named assertions.
ie the language is flexible but the tools enforce a methodology.
Simon
At 01:24 PM 9/17/2002, Prakash Narain wrote:
>I completely support Adam's position. Automatically generated
>names are not a convenient way to refer to assertions in tool
>control commands. Also, it is more intuitive to understand
>reports on assertions when they refer to a user given name and
>not an automatic name.
>
>Best Regards,
>
>Prakash
>
>Adam Krolnik wrote:
>
>>
>>Good afternoon;
>>
>>I would like to understand how you intend to use names placed on
>>properties/assertions/assumptions in tools.
>>
>>I would also like to understand what you would do in a tool
>>with an assert/etc without a name. How would you refer to this
>>element?
>>
>>
>>My position is that a mandatory name is best for:
>> Ease of reference to the element by external means (tools, etc.)
>> referral in conversations about failures, etc.
>> referral in documents for coverage, testplans, etc.
>> Ease of location of the assertion (within editor, etc.)
>>
>>
>>
>> Thanks.
>>
>> Adam Krolnik
>> Verification Mgr.
>> LSI Logic Corp.
>> Plano TX. 75074
>>
>>.
>
This archive was generated by hypermail 2b28 : Tue Sep 17 2002 - 23:37:20 PDT