Peter Ashenden wrote:
> You can use time'pos(value) to get the number of primary units in a time
> value. The result is of type universal_integer. It doesn't get around the
> problem of exceeding the representable range of universal_integer, but it
> does avoid using "/".
time'pos gets me around the integer divide problem. Thanks.
Jim Lewis wrote:
> David,
> Did you try to do it piece-wise? They are all
> powers of 10, so it should not be anything
> other than tedious.
This gets me past the "0 ns" problem where the resolution passed
is lower than the base resolution value. It makes an ifthenelse
tree, but it works.
Wow, this SIG really works.
By putting these two suggestions together, I was able to get something
that works. Thanks guys. I should have a new version up soon.
BTW, the VHPI suggestions don't compile. I'll talk more about that
when I do the release.
-- David W. Bishop dbishop@vhdl.org All standard disclaimers apply.Received on Tue Sep 28 07:00:42 2004
This archive was generated by hypermail 2.1.8 : Tue Sep 28 2004 - 07:00:46 PDT