README.md

mlr

mlr - single-header-only C++26 linear algebra math library

About

mlr implements aligned array arr : std::array<T,N> based vector vec::type<T,N,A> type for N-dimensional linear algebra math using OpenGL/KHR providing a base compatibility type interface for Quake C++ vec_t based types.

Dependencies

• C++26
• OpenGL/KHR

Building

cmake . --install-prefix=/usr
make install

Optimization

Change CMakeLists.txt compile flags to fit your needs:

add_compile_options(-march=native -mfpmath=[your SIMD instruction set] -O3)

Usage

#include <mlr/vector.hpp>

using namespace math;

int main(int argc, char** argv)
{
    vec::f32<3> A = { vec::f32<3>::identity(0),
                      vec::f32<3>::identity(1),
                      vec::f32<3>::identity(2) };

    A[2] = vec::f32<3>::cross3(A[0],A[1]);

    A[2].print_header(3, "cross3");
    A[2].print(3);

	exit(EXIT_SUCCESS);
}

|vector                                 | typ|alg|vec|alg|mode           |cnt
|+1.00e+00 +0.00e+00 +0.00e+00 +0.00e+00|   4|  4| 16| 16|adaptive/vector|4
|+0.00e+00 +1.00e+00 +0.00e+00 +0.00e+00|   4|  4| 16| 16|adaptive/vector|4
|cross3                                 | typ|alg|vec|alg|mode           |cnt
|+0.00e+00 +0.00e+00 +1.00e+00 +0.00e+00|   4|  4| 16| 16|adaptive/vector|4
|cross4                                 | typ|alg|vec|alg|mode           |cnt
|-0.00e+00 +0.00e+00 -0.00e+00 +1.00e+00|   4|  4| 16| 16|adaptive/vector|4
|cross2 GL_CCW/GL_CW                    | typ|alg|vec|alg|mode           |cnt
|-0.00e+00 +1.00e+00 +0.00e+00 -1.00e+00|   4|  4| 16| 16|adaptive/vector|4
|dot4                                   | typ|alg|vec|alg|mode           |cnt
|+0.00e+00 +0.00e+00 +0.00e+00 +0.00e+00|   4|  4| 16| 16|adaptive/vector|4
|sum                                    | typ|alg|vec|alg|mode           |cnt
|+1.00e+00 +1.00e+00 +1.00e+00 +1.00e+00|   4|  4| 16| 16|adaptive/vector|4

|vector                       | typ|alg|vec|alg|mode           |cnt
|+1.00e+00 +0.00e+00 +0.00e+00|   4|  4| 12|  4|adaptive/scalar|3
|+0.00e+00 +1.00e+00 +0.00e+00|   4|  4| 12|  4|         scalar|3
|+0.00e+00 +0.00e+00 +1.00e+00|   4|  4| 16| 16|         vector|3
|un/aligned add               | typ|alg|vec|alg|mode           |cnt
|+0.00e+00 +1.00e+00 +1.00e+00|   4|  4| 12|  4|adaptive/scalar|3
|+1.00e+00 +0.00e+00 +1.00e+00|   4|  4| 12|  4|         scalar|3
|+1.00e+00 +1.00e+00 +0.00e+00|   4|  4| 16| 16|         vector|3

Links

Other C++ Math Libraries

• ITK
• vxl/vnl
• Terathon-Math-Library
• mr-math
• versor

Other OpenCL/SYCL/CUDA linear algebra vector type implementations

• SYCL Overview
• OpenCL
• CUDA C++ Programming Guide

Quake C++ Math Libraries

• TrenchBroom/vm
• ericwa/ericw-tools
• a1batross/fakk2-sdk
• Prey2006/neo/idlib/math

War for the Overworld C++ Math Library

• CML

Math code/tutorials

• Permutation
• Parity Theorem for Permutations
• Levi-Civita symbol
• Hodge star operator
• How to Generate a Cube's vertices and tris
• par
• RGB565 to RGB888