[sv-ec] RE: negative delays

Sorry to beat this beyond death.

1364-2005 9.7.1 says,
"If the delay expression evaluates to a negative value, it shall be
interpreted as a twos-complement unsigned integer of the same size as a
time variable."

I'm not sure how to interpret that.

I don't know how "two's complement" and "unsigned" go together.

It's not clear from the wording that it is first sign-extended to 64
bits and then treated as unsigned. I'm not arguing with the intent. I
just think the wording is foggy.

> > My understanding has always been that for # unit delays, a
> > negative delay value is treated as an unsigned value.
> > I also thought that was in the 1364 LRM, somewhere.
> 
> It is; a colleague has pointed out to me that 1364-2005
> clause 9.7.1 clearly states that the value is sign-extended
> to the width of "time" and then treated as unsigned.  So
> the simulator that treated the negative value as being zero
> was definitely violating 1364-2005.

Thanks,
Shalom

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