(Fwd) BTF - BE65 - Real conversion in expressions

From: Thomas Fitzpatrick (tfitz@cadence.com)
Date: Tue Apr 21 1998 - 13:02:40 PDT


As promised, here is my analysis of this enhancement:
X-Lines: 77
Content-Type: text/plain; charset="us-ascii"
Content-Length: 2499
X-Status: $$$$
X-UID: 0000000395
Status: RO

I agree with Mac's suggestion for the proposal. I also add the following to the
end of section 9.3.1 on p. 25:

"If either operand to an arithmetic operator is real, the resulting expression
is of type real."

--- Forwarded mail from mac@silicon-sorcery.com

Date: Mon, 27 Oct 1997 13:43:22 -0800 (PST)
From: <mac@silicon-sorcery.com>
To: "Anders Nordstrom" <andersn@nortel.ca>
Cc: btf@boyd.com
Subject: BTF - BE65 - Real conversion in expressions
Reply-To: mac@silicon-sorcery.com

<p>Steve Meyers writes:
>
> Subject: BTF - BE65 - Real conversion in expressions
>
> Behavioral Task Force - Errata Submission
>
> Assigned Errata Number: BE65
> Errata Name (Description): Real conversion in expressions
> Section: 3.9.2 page 25
> Date Submitted: 970923
> Requestor: Steve Meyer
>
> Status: Submitted (priority not yet assigned)
>
> Errors found in the Verilog LRM (IEEE 1364-1995).
>
> Details:
>

> I do not think the LRM's explanation of conversion of reals in
> expressions is right. Section 3.9.2, paragraph 2 explicitly states
> conversion to real takes place on assignment implying that

        I do not agree that there is a implication as you state it.

> non reals in expressions (where operator is defined for real
> operands) are not converted to real but rather a syntax error has
> occurred, but then one of the examples at bottom of section 3.10
> shows a parameter expression with a real and non real constant. I
> assume section 3.9.2 should say that implicit conversion to real
> occurs for mixed type expressions. Or is the text correct and the
> example wrong? Also I think the LRM should document if there is a
> difference between constant expressions and variable expressions
> i.e. if "3.0 + 4" is legal but "r + i" where r is declared real and
> i declared integer is not legal syntax following text in section
> 3.9.2.
> Steve Meyer

My proposal is to change the sentance from

        Implicit conversions shall take place when a net or register is
        assigned to a real.

to:
        Implicit conversions shall take place when an expression is
        assigned to a real.

<p>-mac

<p>---End of forwarded mail from mac@silicon-sorcery.com

-- 
---------------
Tom Fitzpatrick

Senior Technical Marketing Manager Cadence Design Systems, Inc. Cycle Simulation Products 270 Billerica Rd. Logic Design and Verification Business Unit Chelmsford, MA 01824 x6438 (978)446-6438



This archive was generated by hypermail 2.1.4 : Mon Jul 08 2002 - 12:52:48 PDT and
sponsored by Boyd Technology, Inc.