lib_xcore_math: VPU-optimized math for the xcore architecture

FAVOURITE VERSION RELEASED COMMENTS DOCUMENTATION SOFTWARE
3.0.0 2026-05-27 v3 adds preliminary support for XCORE architecture VX4B product and removes obsolete build system support VIEW HTML VIEW PDF ZIP
View older versions
2.4.1 2026-02-16 2.4.1 brings minor bug fixes (XMATH_BFP_DEBUG_CHECK_LENGTHS now enabled at build, FFT Lookup script path in CMakeLists.txt), and change linking libm & enabling position-independent code for native x86 builds ZIP ZIP
2.4.0 2024-11-06 ZIP ZIP
v2.2.0 2024-06-04 ZIP ZIP
INFO

Summary

lib_xcore_math is an arithmetic and DSP library which leverages the Vector Processing Unit (VPU) of XS3/xcore.ai based devices in order to accelerate costly operations on vectors of 16- or 32-bit data. Included are functions for block floating-point arithmetic, fast Fourier transforms, discrete cosine transforms, linear filtering and more.

Features

  • Block Floating-Point Vector API

  • Vector/Array API

  • Scalar API

  • Linear Filtering API

  • Fast Fourier Transform API

  • Discrete Cosine Transform API

Known issues

  • Native build only supports xs3a VPU emulation

Development repo

Required tools

  • XMOS XTC Tools: 15.3.1

Required libraries (dependencies)

  • None

Support

This package is supported by XMOS Ltd. Issues can be raised against the software at www.xmos.com/support or using GitHub issues.

Scroll to Top
Cookie Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.

Strictly Necessary Cookies

Strictly Necessary Cookie should be enabled at all times so that we can save your preferences for cookie settings.

Analytics Cookies

This website uses Google Analytics to collect anonymous information such as the number of visitors to the site, and the most popular pages.

Keeping this cookie enabled helps us to improve our website.