RE: [sv-ac] Effect of assetion control system tasks

From: Korchemny, Dmitry <dmitry.korchemny_at_.....>
Date: Thu Feb 22 2007 - 12:19:21 PST
Hi Dave,

 

I think that $assertoff is enough because there is an ability to specify
exact statements that you want to turn off. At least I cannot see a
scenario when all assertions (in the module) should be enabled, but
assumptions still checked. The coverage collection is a different thing
and controlling it separately might be useful.

 

The users will certainly like if is possible to specify wildcards in the
assertion name.

 

Thanks,

Dmitry

 

________________________________

From: owner-sv-ac@server.eda.org [mailto:owner-sv-ac@server.eda.org] On
Behalf Of Rich, Dave
Sent: Friday, February 16, 2007 9:35 PM
To: Bassam Tabbara; Surrendra Dudani
Cc: sv-ac@server.eda.org
Subject: RE: [sv-ac] Effect of assetion control system tasks

 

Yes, in the VPI section, it's much clearer that they were designed for
concurrent assertions since immediate assertions have no state and do
not have existing attempts.

 

BTW, why is there no $assumeoff or $coveroff?

 

Dave

 

 

________________________________

From: Bassam Tabbara [mailto:Bassam.Tabbara@synopsys.com] 
Sent: Friday, February 16, 2007 10:36 AM
To: Surrendra Dudani; Rich, Dave
Cc: sv-ac@eda.org
Subject: RE: [sv-ac] Effect of assetion control system tasks

 

Indeed. VPI control does same. In the LRM sentence/whole paragraph
"checking"/"executing" is used interchangeably.

 

Thx.

-Bassam.

 

 

________________________________

From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of
Surrendra Dudani
Sent: Friday, February 16, 2007 10:09 AM
To: Rich, Dave
Cc: sv-ac@eda.org
Subject: RE: [sv-ac] Effect of assetion control system tasks

My understanding of $asserton and $assertoff is to turn on/off assertion
evaluations/executions. It is not about reporting.

Surrendra

 

________________________________

From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of
Rich, Dave
Sent: Friday, February 16, 2007 12:47 PM
To: Bassam Tabbara; Singh, Tej; Kulshrestha, Manisha; sv-ac@eda-stds.org
Subject: RE: [sv-ac] Effect of assetion control system tasks

I don't think it's that clear. 

 

It is understandable that "currently executing" assertions is referring
to concurrent assertions since no immediate assertion could be executing
at the same time as the $assertoff. It would have be clearer to use the
language of clause 17 and say that all current attempted evaluations of
assertions are disabled as well as future attempts.

 

To say "stop checking" at best implies that results of the expression is
not checked and therefore no action shall be taken. It says nothing
about whether the expression is or isn't evaluated.

 

In the software world, people like and have been using the immediate
assertions in their code, and you would be doing them a disservice if
side-effects were not executed as a result of $assertoff. The main
purpose of $assertoff is to filter messages and actions from assertions
during a specific unstable time and possibly improving performance of
dynamic simulation by reducing the overhead of numerous assertion
threads. I don't see much of an impact in explicitly requiring that
immediate assertion expression always be evaluated.

 

Dave

 

 

________________________________

From: owner-sv-ac@server.eda.org [mailto:owner-sv-ac@server.eda.org] On
Behalf Of Bassam Tabbara
Sent: Wednesday, February 14, 2007 3:36 PM
To: Singh, Tej; Bassam Tabbara; Kulshrestha, Manisha;
sv-ac@server.eda-stds.org
Subject: RE: [sv-ac] Effect of assetion control system tasks

 

I believe this is clearly stated in $assertoff clause -- stop checking
meaning stop executing which would stop side-effects if any.

 

Thx.

-Bassam.

 

 

________________________________

From: Singh, Tej [mailto:tej_singh@mentor.com] 
Sent: Wednesday, February 14, 2007 3:19 PM
To: Bassam Tabbara; Kulshrestha, Manisha; sv-ac@eda-stds.org
Subject: RE: [sv-ac] Effect of assetion control system tasks

What does it mean to $assertoff an immediate assertion? For e.g

 

assert (bus.randomize() == 1) else $error("randomization failed");

 

Does $assertoff means it should stop randomizing (i.e. stop checking for
expression) or should it stop reporting error?

I think it should stop checking for the expression but then does it
imply to not

assert on an expression that has side effects?

 

Regards

Tej

 

	 

	
________________________________


	From: owner-sv-ac@server.eda.org
[mailto:owner-sv-ac@server.eda.org] On Behalf Of Bassam Tabbara
	Sent: Wednesday, February 14, 2007 12:40 PM
	To: Kulshrestha, Manisha; sv-ac@server.eda-stds.org
	Subject: RE: [sv-ac] Effect of assetion control system tasks

	Yes they should -- same goes for VPI control.

	 

	Thx.

	-Bassam.

	 

	 

	
________________________________


	From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf
Of Kulshrestha, Manisha
	Sent: Wednesday, February 14, 2007 12:34 PM
	To: sv-ac@eda-stds.org
	Subject: [sv-ac] Effect of assetion control system tasks

	Hi,

	 

	Do assertion control system tasks affect immediate assertions ?
It is not very clear from the LRM. The immediate assertion section only
has the following statement: "Note: The assertion control system tasks
are described in 24.9."

	 

	Thanks.

	Manisha

	
	-- 
	This message has been scanned for viruses and 
	dangerous content by MailScanner <http://www.mailscanner.info/>
, and is 
	believed to be clean. 
	-- 
	This message has been scanned for viruses and 
	dangerous content by MailScanner <http://www.mailscanner.info/>
, and is 
	believed to be clean. 


-- 
This message has been scanned for viruses and 
dangerous content by MailScanner <http://www.mailscanner.info/> , and is

believed to be clean. 
-- 
This message has been scanned for viruses and 
dangerous content by MailScanner <http://www.mailscanner.info/> , and is

believed to be clean. 
-- 
This message has been scanned for viruses and 
dangerous content by MailScanner <http://www.mailscanner.info/> , and is

believed to be clean. 


-- 
This message has been scanned for viruses and 
dangerous content by MailScanner <http://www.mailscanner.info/> , and is

believed to be clean. 

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Thu Feb 22 12:20:29 2007

This archive was generated by hypermail 2.1.8 : Thu Feb 22 2007 - 12:20:42 PST