Faster floating-point square root for integer processors

This paper presents some work in progress on fast and accurate floating-point arithmetic software for ST200-based embedded systems. We show how to use some key architectural features to design codes that achieve correct rounding-to-nearest without sacrificing for efficiency. This is illustrated with the square root function, whose implementation given here is faster by over 35% than the previously best one for such systems. I. INTRODUCTION The STMicroelectronics ST200 is an embedded media pro- cessor derived from the Lx technology platform (7), designed to implement advanced audio and video codecs in consumer devices such as set-top boxes for HD-IPTV (High Definition Internet Protocol Television), cell phones, wireless terminals, and PDAs. The integration of several instances of the ST200 in large Systems On Chip (SOC) provides an unprecedented level of media processing ability, enabling for instance single chip H.264 solutions.