Section: Software

FLIP: a floating-point library for integer processors

Participants : Christman Fagbohoun, Claude-Pierre Jeannerod, Jingyan Jourdan Lu, Guillaume Revy.

FLIP is a C library for the efficient software support of binary32 IEEE 754-2008 floating-point arithmetic on processors without floating-point hardware units, such as VLIW or DSP processors for embedded applications. The current target architecture is the VLIW ST200 family from STMicroelectronics (especially the ST231 cores).

FLIP 1.0 has been released in 2009. Most notably, it implements highly-parallel algorithms for correctly-rounded division and square root. Also, more functions have been added to the FLIP project this year, such as square, reciprocal, reciprocal square root, cube root, reciprocal cube root, reciprocal fourth root, fused multiply-add, fused square-add, as well as lower level functions for comparisons, multiplication by 2, min, max, etc. A systematic validation framework has further been set up.

Status: Stable, version 1.0 / Target: VLIW processors (ST200 family from STMicroelectronics) / License: LGPL / OS: Linux / Programming Language: C / URL: http://flip.gforge.inria.fr/