On Wed, Jun 18, 2008 at 6:59 PM, Gabriel Sechan <[EMAIL PROTECTED]> wrote:
>
> Does anyone have or know where to find the algorithm for floor(double)? I
> have to implement this on a platform without a working implementation and I'd
> rather not have to come up with it myself- working with mantissas and
> exponents hurts my brain.
You don't mention the software and hardware environment you are working in.
I break open my copy of _The Standard C Library_, P.J.Plauger,
Prentice Hall 1992.
Turning to page 141 I see one-line definitions of double (ceil)
(double x) and double (floor) (double x).
They both depend on an internal function _Dint which "tests and clears
all fraction bits less than a threshold value". Note "The code for
clearing fraction bits is a bit tricky."
Turning over to page 142, I see the C code for function _Dint. It
takes up the whole page. About 50 lines. Possibly this is on line
somewhere, but I found it easier to open the book.
carl
--
carl lowenstein marine physical lab u.c. san diego
[EMAIL PROTECTED]
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
