Message-Id: <200206112142.g5BLgHS06172@delorie.com>
Sender: rich AT phekda DOT freeserve DOT co DOT uk
Date: Tue, 11 Jun 2002 22:12:27 +0100
From: Richard Dawe <rich AT phekda DOT freeserve DOT co DOT uk>
X-Accept-Language: de,fr
To: DJGPP announcements <djgpp-announce AT delorie DOT com>
CC: GMP mailing list <gmp AT swox DOT com>
Subject: ANNOUNCE: DJGPP packages of GMP 4.1
Content-Type: text/plain; charset=us-ascii
Reply-To: djgpp AT delorie DOT com

Hello.

I have made DJGPP packages of GMP 4.1. But what is GMP? Here is a
description from its README:

"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."

Binary (gmp41b.zip), documentation (gmp41d.zip) and source (gmp41s.zip)
packages are available from Simtel.NET and its mirrors:

http://www.simtel.net/gnudlpage.php?product=/gnu/djgpp/v2gnu/gmp41b.zip&name=gmp41b.zip
http://www.simtel.net/gnudlpage.php?product=/gnu/djgpp/v2gnu/gmp41d.zip&name=gmp41d.zip
http://www.simtel.net/gnudlpage.php?product=/gnu/djgpp/v2gnu/gmp41s.zip&name=gmp41s.zip

If you just wish to program with GMP, please download the binary package.
The binary package was built for i386 and above processors.

If you wish to build GMP from scratch, please download the source package.
No source changes were needed to build with DJGPP. You may wish to build
from scratch, to get more performance from GMP, because the library can
use optimisations for your particular processor, e.g. if you have an AMD
Athlon processor.

If you wish to view the documentation in other formats (DVI, PostScript,
HTML, text), please download the binary package.

Here is a list of changes since GMP 4.0.1 from GMP 4.1's NEWS file:

"* Bug fixes.
* Speed improvements.
* Upwardly binary compatible with 4.0, 4.0.1, and 3.x versions.
* Asymptotically fast conversion to/from strings (mpz, mpq, mpn levels), but
  also major speed improvements for tiny operands.
* mpn_get_str parameter restrictions relaxed.
* Major speed improvments for HPPA 2.0 systems.
* Major speed improvments for UltraSPARC systems.
* Major speed improvments for IA-64 systems (but still sub-optimal code).
* Extended test suite.
* mpfr is back, with many bug fixes and portability improvements.
* New function: mpz_ui_sub.
* New functions: mpz_export, mpz_import.
* Optimization for nth root functions (mpz_root, mpz_perfect_power_p).
* Optimization for extended gcd (mpz_gcdext, mpz_invert, mpn_gcdext).
* Generalized low-level number format, reserving a `nails' part of each
  limb.  (Please note that this is really experimental; some functions
  are likely to compute garbage when nails are enabled.)
* Nails-enabled Alpha 21264 assembly code, allowing up to 75% better
  performance.  (Use --enable-nails=4 to enable it.)"

Finally if you have any problems, comments or suggestions for the DJGPP
port, please send me mail. Please note that I do not use GMP. The packages
are provided for your convenience. But I will do my best to help.

Thanks, bye,

-- 
Richard Dawe [ http://www.phekda.freeserve.co.uk/richdawe/ ]