math_utilities.h File Reference

Contains various functions for common mathematical operations. More...

#include <cmath>
#include <tuple>
#include "../global/global.h"
#include "../global/global_cuda.h"
#include "../utils/basic_structs.h"
#include "../utils/gpu.hpp"

Include dependency graph for math_utilities.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Functions
template<typename T >
std::tuple< T, T, T >	math_utils::rotateCoords (Real const &x_1, Real const &x_2, Real const &x_3, Real const &pitch, Real const &yaw)
	Rotate cartesian coordinates. All arguments are cast to double then rotated. If the type is 'int' then the value is rounded to the nearest int.
__device__ __host__ Real	math_utils::dotProduct (Real const &a1, Real const &a2, Real const &a3, Real const &b1, Real const &b2, Real const &b3)
	Compute the dot product of a and b.
__device__ __host__ Real	math_utils::SquareMagnitude (Real const &v1, Real const &v2, Real const &v3)
	Compute the magnitude of a vector.
__device__ __host__ void	math_utils::Cyclic_Permute_Once (hydro_utilities::VectorXYZ< Real > &vec)
	Cyclically permute a Vector once. i.e. (x,y,z) becomes (y,z,x)
__device__ __host__ void	math_utils::Cyclic_Permute_Twice (hydro_utilities::VectorXYZ< Real > &vec)
	Cyclically permute a Vector twice. i.e. (x,y,z) becomes (z,x,y)
template<typename T >
__device__ __host__ T	math_utils::clamp (T val, T lo, T hi)
	When val lies within the inclusive range [lo, hi], returns val. Otherwise, return the closest value in the range.

Detailed Description

Contains various functions for common mathematical operations.

Author

Robert 'Bob' Caddy (rvc@p.nosp@m.itt..nosp@m.edu)

Function Documentation

clamp()

template<typename T >

__device__ __host__ T math_utils::clamp	(	T	val,
		T	lo,
		T	hi
	)

When val lies within the inclusive range [lo, hi], returns val. Otherwise, return the closest value in the range.

Warning

The behavior is undefined when lo>hi or any of the values are std::nan.

Note

Instead of using this implementation, we may be able to use std::clamp (unclear if supported by CUDA/Rocm). We could also consider using CUDA/Rocm intrinsics to speed this up

Cyclic_Permute_Once()

__device__ __host__ void math_utils::Cyclic_Permute_Once ( hydro_utilities::VectorXYZ< Real > &  vec )

inline

Cyclically permute a Vector once. i.e. (x,y,z) becomes (y,z,x)

Parameters

[in,out]

vec

The vector to permute

Cyclic_Permute_Twice()

__device__ __host__ void math_utils::Cyclic_Permute_Twice ( hydro_utilities::VectorXYZ< Real > &  vec )

inline

Cyclically permute a Vector twice. i.e. (x,y,z) becomes (z,x,y)

Parameters

[in,out]

vec

The vector to permute

dotProduct()

__device__ __host__ Real math_utils::dotProduct ( Real const &  a1, Real const &  a2, Real const &  a3, Real const &  b1, Real const &  b2, Real const &  b3  )

inline

Compute the dot product of a and b.

Parameters

[in]	a1	The first element of a
[in]	a2	The second element of a
[in]	a3	The third element of a
[in]	b1	The first element of b
[in]	b2	The second element of b
[in]	b3	The third element of b

Returns

Real The dot product of a and b

rotateCoords()

template<typename T >

std::tuple< T, T, T > math_utils::rotateCoords ( Real const &  x_1, Real const &  x_2, Real const &  x_3, Real const &  pitch, Real const &  yaw  )

inline

Rotate cartesian coordinates. All arguments are cast to double then rotated. If the type is 'int' then the value is rounded to the nearest int.

Rotation such that when pitch=90 and yaw=0 x1_rot = -x3 and when pitch=0 and yaw=90 x1_rot = -x2

Template Parameters

T	The return type

Parameters

[in]	x_1	x1 coordinate
[in]	x_2	x2 coordinate
[in]	x_3	x3 coordinate
[in]	pitch	Pitch angle in radians
[in]	yaw	Yaw angle in radians

Returns

std::tuple<T, T, T> The new, rotated, coordinates in the order <x1, x2, x2>. Intended to be captured with structured binding

SquareMagnitude()

__device__ __host__ Real math_utils::SquareMagnitude ( Real const &  v1, Real const &  v2, Real const &  v3  )

inline

Compute the magnitude of a vector.

Parameters

[in]	v1	The first element of the vector
[in]	v2	The second element of the vector
[in]	v3	The third element of the vector

Returns

Real The dot product of a and b