Technical preview: DFP functionality for XL C/C++ Advanced Edition for Linux, V9.0
Decimal floating-point support for IBM XL C/C++ Advanced Edition for Linux, V9.0, allows you to develop programs that can use the arithmetical operation type that is specified at compile or run time, regardless of the hardware platform on which it is running.
You can also start developing and testing programs in anticipation of hardware upgrades.
If you use IBM XL C/C++ Advanced Edition for Linux, V9.0, you can preview IBM extensions that support decimal floating-point operations on either Red Hat Enterprise Linux® AS 5 or SUSE Linux Enterprise Server 10 Service Pack 1. This document tells you how to do it.
Decimal floating-point arithmetic offers greater computational precision in business and financial applications where numeric data computation and I/O is usually performed in decimal form.
IBM provides two types of decimal floating-point support:
- If you are using the compiler on POWER6 hardware, you can enable the compiler to recognize the syntax used by the built-in instructions that support decimal floating-point operations and to use POWER6 DFP hardware instructions for better DFP performance.
- Alternatively, you can use library functions supplied by the Advanced Toolchain Technical Preview to emulate those built-in instructions.