[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

OS X Real Exponentiation

From: Joe Koski
Subject: OS X Real Exponentiation
Date: Thu, 07 Sep 2006 10:49:51 -0600
User-agent: Microsoft-Entourage/

The following interesting (to OS X users, at least) statement appeared on
Andy Vaught's g95 blog at http://g95.sourceforge.net/

September 6
Bill McKie pointed out that x86/OSX wasn't trapping overflows during real
exponentiation. It turns out that OSX uses the MMX unit to do exponentiation
instead of the x87. So I added code to set the bits for MMX traps. Not only
does the MMX exponentiation use about six times the code that the x87 uses,
but it turns out that OSX doesn't handle traps from the MMX unit correctly
either, reporting only a generic floating point exception. To add insult to
injury, it takes about 3/4 of second (!) to process a floating point
exception on OSX, be it x86 or powerpc.

Over the last couple of years I've been coming around to a new philosophy
regarding OSX-- OSX sucks. Hard.

I don't know how this affects octave, but as a long ago 8086/8087 PC user,
it's a big surprise to me.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]