RE: [sv-ac] RE: 1800 draft 9 review: Clause 16 - 16.8 incl.

Hi Folks:
 
My comment is only about the last point.  It is illegal to use the
sequence methods with input or inout local variable formal arguments
because these arguments require local variable values to be passed into
the underlying sequence evaluation.  The fact that a local variable
needs to be passed in requires timing alignment between the evaluation
in the instantiating context and the beginning of the evaluation of the
underlying sequence.  However, we do not get such timing alignment when
using the sequence methods.  The text is correct even if the coloring is
not.
 
J.H.

________________________________

From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of
Korchemny, Dmitry
Sent: Wednesday, July 08, 2009 1:50 PM
To: Eduard Cerny
Cc: sv-ac@eda.org
Subject: [sv-ac] RE: 1800 draft 9 review: Clause 16 - 16.8 incl.



Hi Ed,

 

Please, see my comments below.

 

Thanks,

Dmitry

 

From: Eduard Cerny [mailto:Eduard.Cerny@synopsys.com] 
Sent: Wednesday, July 08, 2009 9:37 PM
To: Korchemny, Dmitry
Cc: sv-ac@eda.org; Eduard Cerny
Subject: 1800 draft 9 review: Clause 16 - 16.8 incl.

 

Hello,

 

I did not find any issues in 16 - 16.8, but I have the following
questions / comments.

 

Best regards,

ed

 

------------------

 

Section 16.2: The statement "Immediate assertions are primarily intended
to be used with simulation." does not quite apply to deferred
assertions, they can be used very well in formal.

[Korchemny, Dmitry] But in some cases immediate assertions can also be
used in formal. You can open a Mantis item to fox this in the future.

 

16.5: It states at the beginning:" All variables in a concurrent
assertion use the value sampled in the Preponed region of a time slot
with the exception of local variables, constant casts and automatic
variables in procedural code (see 16.15.6), and free checker variables
(see 17.7.2)."

Similarly below Figure 16-1.

I think that all checker variables are not sampled, not just free
checker variables.

[Korchemny, Dmitry] Since we have only nonblocking assignments in
checkers, the result should be same whether the checker variables are
sampled or not - they are updated in the Re-NBA region. I would even say
that it is safer to treat checker variables that participate in NBA as
sampled in concurrent assertions.

 

16.8.2: Is it necessary to have this restriction, marked in red?

It shall be an error to apply any of the sequence methods triggered (see
16.9.11, 16.14.6) or

matched (see 16.14.5) to an instance of a named sequence with an input
or inout local variable formal

argument. 

[Korchemny, Dmitry] Maybe the intention was to say output variables?

---------------------------------------------------------------------
Intel Israel (74) Limited

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

-- 
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.