Re: Division by zero

[Mathieu Langlois <realsoftlists at gmail dot com>]

Check this out on page 388.  It's the IBM PPC 32bit documentation
describing the divw instruction:
http://www-3.ibm.com/chips/techlib/techlib.nsf/techdocs/852569B20050FF778525699600719DF2/$file/6xx_pem.pdf

Quote: "If an attempt is made to perform either of the
divisions—0x8000_0000 / -1 or
<anything> / 0, then the contents of rD are undefined, as are the
contents of the LT, GT,
and EQ bits of the CR0 field (if Rc = 1). In this case, if OE = 1 then
OV is set."

As you can see, the result from the instruction on a division by 0 is
undefined, and there are no error flags whatsoever.  If someone wants
to research the Intel documentation, I suspect the behavior is
similar.  RB is simply passing the instruction to the processor and
returning the result.

Should RS implement internal checkings?  Maybe.  At the cost of
performance?  Not if it can't be disabled.

Math

On Wed, 30 Mar 2005 20:22:03 -0600, Jonathan Johnson
<jonj at realsoftware dot com> wrote:
> REALbasic simply compiles in a divw instruction on PPC for integer
> division, and the equivalent on x86.
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>