ISSUE 73

Number 73
Category errata
Synopsis 4.1.14 implies replication legal lvalue
State lrmdraft
Class errata-discuss
Arrival-DateJul 16 2002
Originator sharp@cadence.com
Release 2001b: 4.1.14
Environment
Description
Section 4.1.14 describes the behavior of concatenations when
used in expressions (i.e. rvalues). It also describes
replications as a form of concatenation, and the comments
on the examples give concatenations that they are equivalent
to.

Concatenations are also allowed as lvalues, such as on the
left-hand-side of assignments or attached to output ports.
The statement that replication is equivalent to some
concatenation can be interpreted to mean that replication
is also legal as an lvalue. This was never the intent,
and certainly is not true in Verilog-XL.
Fix
In 4.1.14,

CHANGE

"Another form of concatenation is the replication
operation. The first expression shall be a non-zero,
non-X and non-Z constant expression, the second
follows the rules for concatenation."

TO

"An operator that can be applied only to concatenations
is replication, which is expressed by a concatenation
preceded by a positive, non-X and non-Z constant
expression, enclosed together within brace
characters, and which indicates a joining together
of that many copies of the concatenation. Unlike
regular concatenations, expressions containing
replications shall not appear on the left-hand
side of an assignment and shall not be connected
to output or inout ports."

and in the example that follows these sentences,

CHANGE

"// This is equivalent to"

TO

"// This yields the same value as"

and in the final example of 4.1.14,

CHANGE

"nested concatenations"

TO
"a replication nested within a concatenation"

and

CHANGE

"// This is equivalent to"

TO

"// This yields the same value as"

Audit-Trail
From: Shalom Bresticker <Shalom.Bresticker@motorola.com>
To: Brad.Pierce@synopsys.com
Cc: etf-bugs@boyd.com
Subject: Re: errata/73: PROPOSAL - 4.1.14 implies replication legal lvalue
Date: Mon, 22 Sep 2003 12:07:35 +0300

A few quibbles:

Brad.Pierce@synopsys.com wrote:

> In 4.1.4,

4.1.14


> CHANGE
>
> "Another form of concatenation is the replication
> operation. The first expression shall be a non-zero,
> non-X and non-Z constant expression, the second
> follows the rules for concatenation."
>
> TO
>
> "An operator that can be applied only to concatenations
> is replication, which is expressed by a concatenation
> preceded by a nonzero, non-X and non-Z constant

Should be "positive" instead of "nonzero". See #76.

> expression, enclosed together within brace
> characters, and which indicates a joining together
> of that many copies of the concatenation. Unlike
> true concatenations, including nested concatenations,

What is a "true" concatentation? Maybe "regular" concatenations, instead?

"nested concatenations" have not been mentioned yet.


> expressions containing replications can neither
> be assigned to nor connected to output or inout
> ports."

Sounds like it says,
"can neither be assigned to output or inout ports nor .."

How about something like
"shall not appear on the left-hand size of an assignment"?

(Note that the language should be "shall not" instead of "can not".)

The rest looks OK.

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, Brad.Pierce@synopsys.com
Cc:
Subject: Re: errata/73: PROPOSAL - 4.1.14 implies replication legal lvalue
Date: Mon, 6 Oct 2003 21:24:51 -0400 (EDT)

I have now reviewed Brad's proposal, and I think it makes the situation clear.
It does not list all of the possible lvalue situations where a replication
would not be allowed, but a full list would probably cause more confusion
than it prevented. The longer comment text might cause some formatting
problems. We could use something shorter but less clear, like
"// Same value as", or let the editor worry about it.

Steven Sharp
sharp@cadence.com

From: Shalom Bresticker <Shalom.Bresticker@motorola.com>
To: Steven Sharp <sharp@cadence.com>
Cc: etf-bugs@boyd.com
Subject: Re: errata/73: PROPOSAL - 4.1.14 implies replication legal lvalue
Date: Thu, 09 Oct 2003 12:00:00 +0200

Let the soon-to-be-spun-off editor worry about it.

Steven Sharp wrote:

> I have now reviewed Brad's proposal, and I think it makes the situation clear.
> It does not list all of the possible lvalue situations where a replication
> would not be allowed, but a full list would probably cause more confusion
> than it prevented. The longer comment text might cause some formatting
> problems. We could use something shorter but less clear, like
> "// Same value as", or let the editor worry about it.

--
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