Re: [sv-bc] is unbased, unsized literal allowed in concat?

From: Steven Sharp <sharp_at_.....>
Date: Tue Mar 22 2005 - 12:25:25 PST
>The LRM says, 
>"In a self-determined context these literals have a width of 1 bit".
>
>(I think there should be a comma after the word 'context'.)
>
>A concat is a self-determined context.
>
>So this would be like {1'b1, 1'b0}, which would then be extended to 4 bits,
>giving 4'b0010.

Yes, if it is legal, this is what it would mean.  The question is whether
it should be legal.  Note that the fact that these literals would have a
width of 1 bit, which is different from the based unsized literals, means
that the potential for confusion is increased.  I think that is an
argument for making them illegal.  Furthermore, a user might think that
because they have used a '1 at the top of the concatenation, and the
concatenation is eventually extended, this will somehow result in it being
extended with 1s.

Because of the potential for confusion, I think that these unsized literals
should not be allowed in concatenations.  They provide no benefit there over
a normal 1-bit literal anyway.

Steven Sharp
sharp@cadence.com
Received on Tue Mar 22 12:25:33 2005

This archive was generated by hypermail 2.1.8 : Tue Mar 22 2005 - 12:25:44 PST