RE: [sv-ac] update on 805

 

 

________________________________

From: Kulshrestha, Manisha [mailto:Manisha_Kulshrestha@mentor.com] 
Sent: Friday, September 01, 2006 5:21 PM
To: Lisa Piper; sv-ac@eda-stds.org
Subject: RE: [sv-ac] update on 805

 

My comments are included:

 

________________________________

From: Lisa Piper [mailto:piper@cadence.com] 
Sent: Friday, September 01, 2006 2:13 PM
To: Kulshrestha, Manisha; sv-ac@eda-stds.org
Subject: RE: [sv-ac] update on 805

Hi Minisha,

 

Thanks for doing this - it is needed.  My comments below:

 

lisa

 

________________________________

From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of
Kulshrestha, Manisha
Sent: Friday, September 01, 2006 4:46 PM
To: sv-ac@eda-stds.org
Subject: [sv-ac] update on 805

 

Hi All,

 

Ed and Lisa have raised some questions in the past about the coverage
information from cover directives and corresponding vpi routines.
Currently those two do not match. After discussions with Bassam, we
propose the following changes in the vpi routines section:

 

1. In the statement in the LRM "For assertion handle, the coverable
entities are assertions.", we'll add something like: "The assertions
include assert and  cover directives (cover on properties and
sequences). Some of the vpi properties are not valid for certain cover
directives. The limitations are listed in description for each vpi
property.".

 

2. The description for "vpi_get(vpiAssertVacuousSuccessCovered,
assertion_handle)" will mention that this vpi property can only be used
for assert directives and cover directives on properties. 

 

3. The description for "vpi_get(vpiAssertFailureCovered,
assertion_handle)" will mention that this vpi property can only be used
for assert directives and cover directives on properties.

 

4. We'll state that the formula "in progress = attempts - (successes +
vacuous success + failures)" is not valid for cover directives on
sequences.

[Lisa Piper >>>] where does a "disabled or assert-killed" assertion fit
in?  805 says it is disabled, not vacuous if I remember correctly.   

 

Manisha: In 805, the formula is "in progress = attempts - (successes +
vacuous success + disabled + failures)". This new formula will not be
valid for covers on sequences.

[Lisa Piper >>>] that answers the question of disabled but not
assertkill (an attempt is started but not completed - is that vacuous?)


 

 

5. In the description of the new vpi property
"vpi_get(vpiAssertDisableCovered, assertion_handle)", will mention that
this vpi property can only be used for assert directives.

[Lisa Piper >>>] and that disabled refers to the "disable iff"
construct, not assertoff or assertkill 

 

Manisha: That is correct. The description makes that clear in the
proposal. 

 

Similarly, we'll update the vpi callback part to match with this.

 

Please send your feedback. We can discuss it further in the coming
meeting.

 

Thanks.

Manisha