From 70eff1bca49b0e1f5a37b8b89416d0c062ed0a91 Mon Sep 17 00:00:00 2001 From: Torbjorn Granlund Date: Mon, 13 Apr 2009 18:59:40 +0200 Subject: *** empty log message *** --- NEWS | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) diff --git a/NEWS b/NEWS index a35ec0e57..e1771e541 100644 --- a/NEWS +++ b/NEWS @@ -26,25 +26,26 @@ Changes between GMP version 4.2.X and 4.3.0 new code in this release. Speedups: - * GCD and Extended GCD now use the best known sub-quadratic algorithms. - * Division is now O(M(n)), losing a log(n) term compared to GMP 4.2. + * Vastly improved assembly code for x86-64 processors from AMD and Intel. + * Major improvements also for many other processor families, such as + alpha, powerpc, and itanium. + * New sub-quadratic mpn_gcd and mpn_gcdext, as well as improved basecase + gcd code. + * The multiply FFT code has been slightly improved. + * Balanced multiplication now uses 4-way Toom in addition to schoolbook, + Karatsuba, 3-way Toom, and FFT. + * Unbalanced multiplication has been vastly improved. * Improved schoolbook division by means of faster quotient approximation. - * Multiplication of operands of different sizes now runs much faster. - * About 5% to 10% faster FFT multiply code. - * 4-way toom multiply code added between 3-way toom and FFT. - * Well-optimised assembly code for the x86_64 architecture for K8-K10, - Pentium 4 and "Core 2", resulting in around 75% better speed. - * Much improved, but still far from optimal assembly code for Itanium. - * Well-optimised assembly for POWER4, POWER5, and PPC970. - * Many assembly optimizations for other chips. * Several new algoritms for division and mod by single limbs, giving many-fold speedups. - * The mpz_nextprime function uses sieving. - * Vastly improved nth root code, slightly improved square root code. + * Improved nth root computations. + * The mpz_nextprime function uses sieving and is much faster. * Countless minor tweaks. Features: - * New mpn functions: XXX + * Updated support for fat binaries for x86_32 include current processors + * Lots of new mpn internal interfaces. Some of them will become public + in a future GMP release. * Support for the 32-bit ABI under x86-apple-darwin. * x86 CPU recognition code should now default better for future processors. -- cgit v1.2.1