News - Project LinBox

Announcing LinBox version 1.1

September 22, 2008

Version 1.1.6 has been released
This release includes:

• improvemnt of the Chinese remaindering efficiency,
• upgrade to fflas-ffpack-1.3.3, including a better modular balanced finite field implementation, better bound computations, and bug fixes
• suppression of the gmp++ wrapper (duplicated from givaro). Givaro is now a required library for linbox to compile
• Several bugs and compilation problems have been fixed for the debianization of the library, and support of solaris and cygwin.

April 3, 2008

Version 1.1.5 has been released
This release mostly provide numerous bug and memory leak fixes, and support a broader range of systems/architectures (tested on Sparc-Solaris, PPC-OS-X, x86_64-OS-X, x86_64/32-Linux)

October 27, 2007

Version 1.1.4 has been released This release includes:

• numerous bug fixes, corrections, memory leaks removals,... mostly due to intense feed-back and patches of M. Abshoff.
• Support for gcc-4.2 new conventions on template default parameters
• Extended functionalities of the linear solver (rank deficient singular matrices)
• Faster FFLAS routines for triangular system solving (trsm)), and new inplace echelon form computations (cf fflas-ffpack-1.3.2)
• and a much more...

January 31, 2007

LinBox 1.1 is our second "stable" release. The Maple interface has undergone a major revision and extension with the purpose of making the linear algebra functions easily invoked. Also the BLAS support has been extended to Fortran BLAS (e.g. Goto BLAS) and C BLAS (e.g. ATLAS).
It has been checked with g++-4, on linux PIV, WinNT cygwin PIV, Mac G5 ... with givaro-3.2.6, gmp-4.2.1, ntl-5.4, GotoBLAS-1.10, ATLAS-3.7.10, Maple 10.

What's new in this release:

• New Maple interface.
• Goto BLAS other Fortran BLAS support.
• Sigma bases algorithm.
• BlackBox Characteristic polynomial.
• Fast polynomial Matrix multiplication.
• Block Hankel and block Toepliz BLackBox support.
• Introspective determinant.
• Sparse rational solver.
• Enhanced ring support (in particular Polynomials).
• Faster FFLAS and FFPACK.
• More integer matrix capabilities, with simple drivers in the solutions directory.
• ...and miscellaneous.