On 08/21/10 12:00 PM, Jeroen Demeyer wrote:
On 2010-08-21 07:55, Carl Witty wrote:
On Fri, Aug 20, 2010 at 9:26 PM, Dr. David Kirkby
<david.kir...@onetel.net> wrote:
Unless OS X rounds by default to 64-bits, I can't understand how this would
have ever worked. Why was it not necessary to change the rounding behavior
of an Intel based OS X system?
Modern x86 family chips actually have two totally separate
floating-point units (at least logically, I don't know if they share
hardware). The older uses the "x87" instructions, dating back to the
8087; this rounds to 80 bits by default. SYMPOW's fpu.c changes the
x87 instructions to round to 64 bits. The newer is the SSE/SSE2
floating-point unit (SSE only had single-precision floating point,
SSE2 extended this to double precision); this rounds to 64 bits (or 32
bits for single-precision instructions).
gcc uses SSE by default on 64-bit Intel systems and FPU by default on
32-bit Intel systems. However, this can be changed with gcc's -mfpmath
option: you can choose -mfpmath=387 or -mfpmath=sse
Thank you Carl and Jeroen. Adding the compiler option -mfpmath=sse would
probably allow SYMPOW to build on Cygwin, but of course that would only work for
CPUs which support SSE instructions.
That said, I just tried testing a 64-bit version of sympow I'd made before on
OpenSolaris on an Intel Xeon. In theory that should have worked, but it dumped
core. But nothing surprises me about SYMPOW. Any code that starts:
#! /bin/sh
then has code to test if the 'sh' shell exists or not, leaves a lot to be
desired.
I dug out an old copy of the Intel 387 programmers reference manual I have, and
looked at that last night so I understand what's going on more. I'll have a
patch soon, which will hopefully be portable across compilers and operating
systems.
Dave
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
