We would like to announce the availability of the "Intel IEEE 754r Decimal Floating-Point BID Library" (beta version), at http://www3.intel.com/cd/software/products/asmo-na/eng/219861.htm under 'Decimal Floating-Point', packaged as a tar file which includes a README file besides the source code. The library was tested on several Intel Architecture platforms, with two operating systems (Linux and Microsoft Windows).
The library, written in C, implements the functions defined for decimal floating-point arithmetic operations in 'DRAFT Standard for Floating-Point Arithmetic P754', which is a revision of the IEEE Standard 754-1985 for Binary Floating-Point Arithmetic. It supports primarily the operations on decimal floating-point values encoded using the binary encoding method from the standard draft, but the decimal encoding method is supported as well through conversion functions: next to the link for the library there is another link that provides access to conversion routines between decimal floating-point values encoded using the binary and the decimal encoding methods. This set of conversion routines was contributed to GCC 4.3 by Intel Corporation in January 2007. The source code provided at this web site can be used in any way that does not conflict with the license notice found at the top of each file. If you decide to use the library and have any feedback, please use the feedback link from the web page. Please note however that the source code posted on this web site has a BSD license associated to it (the exact terms are specified in the license notice). We would like to contribute the library to GCC as well, and that version of the code would be released under a GPL license + FSF exception. Thank you, Marius Cornea Intel Corporation
