ISSUE 281

Number 281
Category errata
Synopsis fork..join in automatic tasks/functions
State closed
Class mistaken
Arrival-DateJan 31 2003
Originator Karen Pieper <Karen.Pieper@synopsys.com>
Release 2001b
Environment
Description

>
>Message-ID: <D58D987E20BAD2118A420090273BF41F05871DAD@out.model.com>
>From: Jamie LaFlamme <jamiel@model.com>
>To: "'btf@boyd.com'" <btf@boyd.com>
>Subject: fork..join in automatic tasks/functions
>Date: Fri, 31 Jan 2003 12:00:15 -0800
>
>
>
>I haven't found any information in the 1364-2001 LRM that describes the
>expected behavior of fork..join statements within an automatic task or
>function. Has there been any discussion about whether each statement within
>the fork-join block should have its own copy of the automatic variables or
>whether the statments should behave as if all automatic variables are
>shared? Maybe the behavior intended to be implementation-defined?
>
>Thanks,
>-Jamie LaFlamme


Fix

Close. No change needed.

Audit-Trail

From: Shalom Bresticker <Shalom.Bresticker@motorola.com>
To: jamiel@model.com
Cc: etf-bugs@boyd.com
Subject: Re: errata/281: fork..join in automatic tasks/functions
Date: Sun, 02 Feb 2003 15:15:03 +0200

I don't see here any room for question.

fork...join is no different than any other statement in tasks/functions.
The variables declared with an automatic t/fn are allocated upon entrance to the
t/f,
and then used within the body of the t/f. Once they are allocated, that's it.

Entrance into a fork...join block does not change the variables in any way.

> >I haven't found any information in the 1364-2001 LRM that describes the
> >expected behavior of fork..join statements within an automatic task or
> >function. Has there been any discussion about whether each statement within
> >the fork-join block should have its own copy of the automatic variables or
> >whether the statments should behave as if all automatic variables are
> >shared? Maybe the behavior intended to be implementation-defined?

Shalom

--
Shalom Bresticker Shalom.Bresticker@motorola.com
Design & Reuse Methodology Tel: +972 9 9522268
Motorola Semiconductor Israel, Ltd. Fax: +972 9 9522890
POB 2208, Herzlia 46120, ISRAEL Cell: +972 50 441478




From: Steven Sharp <sharp@cadence.com>
To: etf-bugs@boyd.com, jamiel@model.com
Cc:
Subject: Re: errata/281: fork..join in automatic tasks/functions
Date: Mon, 3 Feb 2003 18:30:07 -0500 (EST)

I agree with Shalom that this one is clear.

All of the statements are in the same scope and will share the same
variables. Automatic variables are only allocated by invoking/calling
the automatic task or function, as stated in the LRM.

Nor would it be desirable for it to work differently.

Steven Sharp
sharp@cadence.com


From: Shalom.Bresticker@motorola.com
To: etf-bugs@boyd.com
Cc:
Subject: Re: errata/281: PROPOSAL - fork..join in automatic tasks/functions
Date: Wed, 27 Aug 2003 11:03:02 +0300 (IDT)

I would like to reopen this issue.
Although the conclusion is the same, the problem is that the issue keeps
coming up, indicating that the text is not clear enough.
So I think we nevertheless should add some text to clarify the behavior.

Shalom


> Date: Mon, 10 Mar 2003 22:11:40 -0800
> Subject: errata/281: PROPOSAL - fork..join in automatic tasks/functions
>
> Close. No change needed.
>
> http://boydtechinc.com/cgi-bin/issueproposal.pl?cmd=view&pr=281

--
Shalom Bresticker Shalom.Bresticker@motorola.com
Design & Reuse Methodology Tel: +972 9 9522268
Motorola Semiconductor Israel, Ltd. Fax: +972 9 9522890
POB 2208, Herzlia 46120, ISRAEL Cell: +972 50 441478

Unformatted


Hosted by Boyd Technology