math library for ARM?

Kai Ruottu
Fri Dec 17 06:24:00 GMT 1999

Bill Su wrote:
> Hi,
>   I have used arm-elf from the development version of binutils.
> It can be read if the output is stripped.  Thanks for help. Now
> I have a problem on floating library.  I use newlib from Cygnus.
> However, it's very slow on whetstone and my project is timing
> critical on floating point performance.  Is there any better
> floating library?  I saw the libm in NetBSD/OpenBSD comes from
> SunPro.  Is it better?  Where can I get the archive?  Thanks.

 For the basic soft-float routines, instead of the 'fp-bit/dp-bit' in
the GCC sources:

------------------ clip -------------------------
New IEEE P854 emulation library

   * To:
   * Subject: New IEEE P854 emulation library
   * From: Torbjorn Granlund <>
   * Date: 15 Jul 1999 20:05:58 +0200

We wrote a new soft floating-point emulation library a while ago, since the
old one (fp-bit.c) is big, buggy, and slow.  This new library is 3 to 15
times faster, depending on the operands, host, and operation.  It is also
smaller, and we believe it is very accurate.

We (Swox AB) hereby donate this to the GNU project.

We have run hundreds of billions of tests for this, comparing the results to
the arithmetic of a Solaris machine.  Testing vectors were pseudorandom,
with a pattern that more likely triggers corner cases.  (I'll give more
information on testing on demand.)
------------------ clip -------------------------

 Please see the egcs-maillist archives for the actual code...

 For the Hitachi H/300 the SWOX-routines are about two times faster than the

 Please see also ' http://HTTP.CS.Berkeley.EDU/~jhauser/ '

Cheers, Kai

Want more information?  See the CrossGCC FAQ,
Want to unsubscribe? Send a note to

More information about the crossgcc mailing list