Re: [sv-ac] Re: [sv-bc] Re: if-else

Hi Ed:

'b0 is another possibility.  I am trying to understand the
technical distinction, if any, from the different choices.

Is it possible for "expression" to have no bits at all in

   if (expression != 'b0)

If so, what does the inequality comparison mean?

J.H.

> X-MimeOLE: Produced By Microsoft Exchange V6.5
> Content-class: urn:content-classes:message
> Date: Thu, 1 Nov 2007 09:18:01 -0700
> Thread-Topic: [sv-ac] Re: [sv-bc] Re: if-else
> thread-index: Acgcoi150OVUrIY6TjSJCBnw3Fw2gAAAJKmg
> From: "Eduard Cerny" <Eduard.Cerny@synopsys.com>
> Cc: <sv-ac@eda.org>, <sv-bc@eda.org>
> X-OriginalArrivalTime: 01 Nov 2007 16:18:02.0861 (UTC) FILETIME=[C6F9F9D0:01C81CA2]
> 
> I thought we would use 'b0?
> ed=20
> 
> > -----Original Message-----
> > From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On=20
> > Behalf Of John Havlicek
> > Sent: Thursday, November 01, 2007 12:13 PM
> > To: sharp@cadence.com
> > Cc: sharp@cadence.com; john.havlicek@freescale.com;=20
> > sv-ac@eda.org; sv-bc@eda.org
> > Subject: [sv-ac] Re: [sv-bc] Re: if-else
> >=20
> > Hi Steven:
> >=20
> > Thanks for your clarification.
> >=20
> > The only other issues that are nagging me involve stuff that
> > I guess is near the boundary of the realm of what is useful. =20
> > For example:
> >=20
> > - In "expresson !=3D 0", the constant 0 is signed 32-bit.  Are there
> >   any cases in which we should use 1'b0 instead of 0?
> >=20
> > - Are there any cases where "expression" has no bit at all?  If so,
> >   is sign-extension defined?  I am worried about void members of=20
> >   tagged unions and similar things.  Perhaps a tagged union with=20
> >   only one member and such that the one member is void type.
> >=20
> >=20
> > J.H.
> >=20
> > > Date: Thu, 1 Nov 2007 10:51:06 -0400 (EDT)
> > > From: Steven Sharp <sharp@cadence.com>
> > > Reply-To: Steven Sharp <sharp@cadence.com>
> > > Cc: sv-ac@eda.org, sv-bc@eda.org
> > > Content-MD5: ogJLgUzqnuP7SOfkNcmc3A=3D=3D
> > > X-Received: By mx-sanjose.cadence.com as lA1EpoHs026736 at=20
> > Thu Nov  1 07:51:50 2007
> > > X-OriginalArrivalTime: 01 Nov 2007 14:55:02.0620 (UTC)=20
> > FILETIME=3D[2E8581C0:01C81C97]
> > >=20
> > > John,
> > >=20
> > > As you have noted,
> > >=20
> > > 	if (expression)
> > > =09
> > > is equivalent to
> > >=20
> > > 	if (expression !=3D 0)
> > >=20
> > > The problem is then that the results of the equality=20
> > operator are not
> > > clearly specified when unknowns are involved.  The correct answer is
> > > that (value1 !=3D value2) is equivalent to (|(value1 ^=20
> > value2)).  So for
> > > (expression !=3D 0), that is equivalent to (|expression), as you =
> have
> > > suggested.
> > >=20
> > > This definition makes sense.  If any of the bits are 1,=20
> > then the value
> > > is definitely not equal to zero, no matter what any unknown=20
> > bits are.
> > > If all of the bits are 0, then the value is definitely=20
> > equal to zero.
> > > If the bits are a mixture of 0 and X/Z, then it is unknown=20
> > whether the
> > > value is equal to zero.  It depends on the value of the=20
> > unknown bits.
> > >=20
> > > After that, there is the issue of how if-else deals with an unknown
> > > condition.  It rather arbitrarily treats an unknown=20
> > condition as false.
> > > Jonathan's suggestion of a cast to "bit" has this same effect.  This
> > > treatment can cause simulation to produce optimistic=20
> > results, which is
> > > undesirable, but that is how it works.
> > > =09
> > > Steven Sharp
> > > sharp@cadence.com
> > >=20
> >=20
> > --=20
> > This message has been scanned for viruses and
> > dangerous content by MailScanner, and is
> > believed to be clean.
> >=20
> >=20

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