Re: [sv-bc] Is '1 > 1 ?

>In regard to
>
>> ’0, ’1, ’X, ’x, ’Z, ’z
>
>P1800 says:
>
>> In a self-determined context these literals have a width of 1 bit,
>> and the value is treated as unsigned.
>
>
>This leaves me curious about how they behave in context-determined
>settings.  In particular, as an operand of ">", or anywhere that
>the signedness of the operands informs the context, do these literals
>"poison" the signed waters by being an unsigned term in the expression?

I believe that they are unsigned.  This sentence was added when I asked
what their width was in a self-determined context, and whether they were
signed or unsigned.  It may be that the only reason the part about them
being unsigned appears together with the part about the self-determined
width is that I asked the two questions together.  I don't recall.

Or it may have been an attempt to say that they are unsigned without the
confusion between the signedness that they contribute to the expression
and the signedness that they end up with from context.  If you specify
their signedness in a self-determined context, then you have ruled out
anything inherited from context.

At any rate, without any special rules in the LRM, they must follow the
existing rules.  If they are unsigned in a self-determined context, then
they are unsigned.

Steven Sharp
sharp@cadence.com