RE: [sv-bc] E-mail Ballot due 8am PDT, Monday, June 11, 2007

Actually, I agree with Steven.

In the other places where this type of description appears, at least in
the 'Operator descriptions' sub-clause, the operators are those where
the context of the operands is self-determined, e.g., equality and
relational operators.

This place is different.

The [section_number] required in Draft 3 is probably 11.7.2. Note that
11.6, while describing size extension, does not discuss the type of
extension, i.e., zero-extension vs. sign-extension. That is introduced
in 11.7.

Shalom
 

> >SVDB 1004 ___Yes   _X_No
> >http://www.eda.org/svdb/bug_view_page.php?bug_id=1004
> 
> The proposed description is incorrect.  It is only valid in the
> special
> case where the conditional operator appears in a self-determined
> context.
> This was always a flaw of most of these separate descriptions under
> the
> individual operators, as opposed to the full description in the width
> rules.
> 
> The shorter operand is not lengthened to match the longer.  Both
> operands
> are lengthened to match the length of the expression, which may be
> longer
> than either of them.
> 
> And it is not necessarily true that the extension is sign-extension if
> both operands are signed.  It is sign-extension if the expression is
> signed.  There could be another unsigned operand elsewhere in the
> context
> of this expression, which would cause the expression to be unsigned.
> In
> that case they would be zero-extended.
> 
> I would suggest that the replacement text be something like
> 
> "The first and second expressions are extended to the same width, as
> described in [section_number]."

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