Segger’s stand-alone Floating-Point Library has now been extended to include an assembly-optimized variant for RISC-V implementations. The library contains a complete set of high-level mathematical functions that have been written in C, and uses advanced algorithms to maximize performance levels.
All of the functionality is fully verified, for both single and double-precision operations. The RISC-V variant, like the existing variant for ARM, is optimized for both high-speed operation and small code size. The balance between size and speed can be configured at library build time.
The SEGGER Floating-Point Library for RISC-V is much smaller than equivalent open-source libraries currently available while achieving up to 100 times the performance on some key operations. This library is also a part of the company’s Runtime Library, which is already included in the widely-used Embedded Studio platform.