[MLton-user] minor 32bit vs. 64bit differences in floating-point calculations with large numbers

Wesley W. Terpstra wesley at terpstra.ca
Sat Oct 16 06:28:06 PDT 2010


On Sat, Oct 16, 2010 at 5:09 AM, Henry Cejtin <henry.cejtin at sbcglobal.net>wrote:

> It is that it is more correct because when you spill from  a
> register to memory and then load it back into a register, it is the same
> as it was before the spill.
>

Yes, it lets you perform all operations with exactly 32/64 bits of
precision.

Thus the SSE2 instructions correctly do
> what IEEE mandates while the x86 mode does not.  (At least  that  is  my
> understanding.)
>

The x87 conforms to the IEEE specification for extended precision. As for
performing intermediate operations at a higher precision, depending on who
you talk to [1], this is a good thing or a bad thing. Both ways are
certainly allowed.

[1] How Java’s floating-point hurts everyone everywhere. Retrieved September
5, 2003 from http://www.cs.berkeley.edu/~wkahan/JAVAhurt.pdf<http://www.cs.berkeley.edu/%7Ewkahan/JAVAhurt.pdf>
.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mlton.org/pipermail/mlton-user/attachments/20101016/80b2d1d9/attachment.htm


More information about the MLton-user mailing list