Template Project

Overview

This project is the proverbial Hello World for the Authors algorithm development activities and is used as a starting point for a platform independent code development.

Credits:

Creadits for the build enviorement goto the authors of the lwext4 project.

Matlab Toolbox

Matlab is the workhorse for algorithm development, and much of the Authors algorithms begins as a Matlab implementation which subsequently is migrated to C, often targeting embedded application. As part of this design flow, a mex wrapper is used to develop a Matlab executable version of the algorithm. This approach accelerates the Matlab runtime and provides a test-bench to the same C code that is eventually cross-compiled for the target platform. These C reference becomes the 'Gold-Standard' for the algorithms and will form the based of subsequent implementation.

Toolchains and Target Architectures:

The project and associated libraries compiled for many targets. Here are an examples for 8/16/32/64 bit architectures.

• generic for x86 or amd64
• arm-none-eabi-gcc for ARM cortex-m0/m3/m4 microcontrollers
• avr-gcc for AVR xmega microcontrollers

Cross compile for embedded architectures

Build cortex-m0 library:

---

make cortex-m0
cd build_cortex-m0
make lib

Build cortex-m3 library:

---

make cortex-m3
cd build_cortex-m3
make lib

Build cortex-m4 library:

---

make cortex-m4
cd build_cortex-m4
make lib

Build an xcode project:

One of the nice feature of CMake is the ability to generate project files for favorite IDEs.

make xcode
cd xcode
open mytoolbox.xcodeproj/