Re: [sv-bc] SV-BC #110 - 2-State Divide by 0 question

From: Brad Pierce <Brad.Pierce@synopsys.com>
Date: Wed Nov 24 2004 - 09:59:39 PST

'Unspecified' as the two-state substitute for 'X' is not
consistent with other similar 2-state scenarios in the LRM.
Why not leave the default 2-state initialization value unspecified?
Why not leave the result of bit'(1'bX) unspecified?
 
>I agree. I'd much rather have this be an "undefined" result
>in the LRM and let a simulator decide what to do.

Some 4-state arithmetic results are left unspecified.
For example --

   "The result of the power operator is unspecified if the
    first operand is zero and the second operand is non-positive,
    or if the first operand is negative and the second operand
    is not an integral value."

I think that's a mistake and that the result should have been
specified as X.

But when the result of a 4-state operation *is* specified, as is the
result of a 4-state divide-by-zero, then the result of the corresponding
2-state operation should be specified, too.

-- Brad
Received on Wed Nov 24 09:58:29 2004

This archive was generated by hypermail 2.1.8 : Wed Nov 24 2004 - 09:58:33 PST