This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
Re: 128 bit floating point arithmetic
- From: Eljay Love-Jensen <eljay at adobe dot com>
- To: hladky dot jiri at seznam dot cz, gcc-help at gcc dot gnu dot org
- Date: Mon, 16 Feb 2004 12:55:47 -0600
- Subject: Re: 128 bit floating point arithmetic
Hi Jirka,
A quick Google search indicates that the Pentium 4 does not natively support 128-bit floating point arithmetic. It does support 64-bit precision mantissa, which would (I presume) be utilized in full via long double (80-bit on Intel architecture, 96-bit on Motorolla 680x0 architecture).
Unless I'm mistaken, to use 128-bit floating point, you'll need to write your own "software FPU" or download a floating point bignum to suite your needs.
Chongo's calc software can do that at even higher precision, and more <http://freshmeat.net/projects/calc/>... mind the licensing restrictions.
Performance will be far less than that of the Pentium 4's FPU, however.
HTH,
--Eljay