RE: [sv-ac] reminder to vote on mantis 1550

<forwarding bounced email from Arturo Salz>

-------- Original Message --------

From: "Arturo Salz" <salz@synopsys.com>
Subject: RE: [sv-ac] reminder to vote on mantis 1550
Date: Thu, 25 Jan 2007 16:00:31 -0800
To: <john.havlicek@freescale.com>, <Eduard.Cerny@synopsys.com>
Cc: <sv-ac@server.eda-stds.org>


John,

I agree with your assessment.

On a side note, I personally believe that sampling at time 0 is a risky
business; a good methodology should avoid clock edges at time 0. Time 0
initialization is merely a crude approximation of what happens in an
actual hardware system - where it's impossible to turn on the system and
expect a clock to latch a deterministic value. Even a clock oscillator
has a finite setup time so it's unlikely to start at the moment the
system is turned on.
Attempting to bring simulation determinism to process that isn't (the
turn on process) may obscure actual bugs, which would not serve our
users well.

I believe you are on the right track by sticking to the other parts of
the LRM. I suggest we avoid adding more determinism than that.

	Arturo

-----Original Message-----
From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of John
Havlicek
Sent: Thursday, January 25, 2007 1:14 PM
To: Eduard.Cerny@synopsys.COM
Cc: john.havlicek@freescale.com; Eduard.Cerny@synopsys.COM;
sv-ac@eda-stds.org
Subject: Re: [sv-ac] reminder to vote on mantis 1550

Hi Ed:

This is the text you have now:

   The sampled value in the preponed region at time 0 of an expression
is
   the value assigned in the declaration of the variable if the
   expression is a variable or in the absence of such an assignment it
is
   the default value of the type. For example, if $sampled(y) is called
   at time 0, and y is of type logic, the value returned is X.


I think we need to rely on other parts of the LRM to define the
preponed value of an expression in the time 0 slot.

Note that the expression might combine terms, some of which have
default initial values and others of which have initial values set by
declaration assignments.

I suggest the following instead:

   The sampled value in the preponed region at time 0 of an expression
   is the initial value of the expression as determined by the rules
   of default initial values and declaration assignments (see
   <appropriate references>).  For example, if y is a static variable
   of type logic with no declaration assignment, then a call to
   $sampled(y) at time 0 returns the value X.

J.H.

> X-MimeOLE: Produced By Microsoft Exchange V6.5
> Content-class: urn:content-classes:message
> Date: Thu, 25 Jan 2007 12:37:53 -0800
> X-MS-Has-Attach: yes
> Thread-Topic: [sv-ac] reminder to vote on mantis 1550
> Thread-Index: AcdAs0bTcrq7EThTRdmyhL1bD8l4HQADUOkg
> From: "Eduard Cerny" <Eduard.Cerny@synopsys.com>
> Cc: <sv-ac@eda-stds.org>
> X-OriginalArrivalTime: 25 Jan 2007 20:37:54.0988 (UTC)
FILETIME=[B0F1BEC0:01C740C0]
> 
> This is a multi-part message in MIME format.
> 
> ------_=_NextPart_001_01C740C0.B0CC2EB7
> Content-Type: text/plain;
> 	charset="us-ascii"
> Content-Transfer-Encoding: 7bit
> X-Former-Content-Transfer-Encoding: quoted-printable
> 
>  Hi John,
> 
> before I deposit it on mantis, would the wording in the attached
updated
> proposal be acceptable?
> 
> Thanks,
> ed
> 


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.