Message-Id:	<200011072347.SAA24725@delorie.com>
Sender:	richdawe AT bigfoot DOT com
Date:	Tue, 07 Nov 2000 21:58:39 +0000
From:	Richard Dawe <richdawe AT bigfoot DOT com>
X-Accept-Language:	de,fr
To:	DJGPP announcements <djgpp-announce AT delorie DOT com>
Subject:	ANNOUNCE: DJGPP packages of GMP 3.1.1 library
Reply-To:	djgpp AT delorie DOT com

Hello.

I have packaged the GNU MP Library version 3.1.1 for use with DJGPP - no
changes were necessary to get it to build with DJGPP. It is available from
the DJGPP archive on Simtelnet:

ftp://ftp.simtel.net/pub/simtelnet/gnu/djgpp/v2gnu/gmp311b.zip
ftp://ftp.simtel.net/pub/simtelnet/gnu/djgpp/v2gnu/gmp311d.zip
ftp://ftp.simtel.net/pub/simtelnet/gnu/djgpp/v2gnu/gmp311s.zip
ftp://ftp.simtel.net/pub/simtelnet/gnu/djgpp/v2gnu/gmp311xb.zip

These are packages of the binaries, documentation (*), sources and
experimental binaries respectively.

(*) texinfo, DVI, PostScript, HTML, plain text - info docs are included
with the binaries.

If you just wish to use GMP in your programs, please download the binary
distribution. You may wish to download the documentation distribution too,
if you want the other documentation formats. Otherwise, download the
source distribution. The binaries were built for i386 processors - you may
want to rebuild from sources to get a speed improvement for your
particular processor.

Here is a description of GNU MP, taken from the Readme file:

"GNU MP is a library for arbitrary precision arithmetic, operating on
signed integers, rational numbers, and floating point numbers.  It has a
rich set of functions, and the functions have a regular interface.

GNU MP is designed to be as fast as possible, both for small operands and
huge operands.  The speed is achieved by using fullwords as the basic
arithmetic type, by using fast algorithms, with carefully optimized
assembly code for the most common inner loops for lots of CPUs, and by a
general emphasis on speed (instead of simplicity or elegance).

GNU MP is believed to be faster than any other similar library.  Its
advantage increases with operand sizes for certain operations, since GNU
MP in many cases has asymptotically faster algorithms."

The experimental distribution has experimental code enabled:

- FFT-based multiplication;
- accurate floating-point arithmetic with rounding control.

This code is still in development and should be considered unstable.

Both sets of binaries were built to use malloc()-allocated memory (rather
than stack-allocated memory used in previous packages). It should be less
prone to overflowing the stack than the GMP 3.0, 3.1 packages.

I have not tested the package extensively, but it passed all its
self-tests. Currently GMP can only be built with Long File Names. If you
have any problems with the package, please mail me.

Thanks to those people who have given me feedback on the previous
packages. Please continue giving me feedback.

Thanks, bye,

-- 
Richard Dawe
[ mailto:richdawe AT bigfoot DOT com | http://www.bigfoot.com/~richdawe/ ]

- Raw text -