Collaboration diagram for [apu.h] APU Vector Operations:

Functions
void	APUVectorDot (uint16_t N, void pInputA, void pInputB, void *pResult)
	APU accelerator for vector dot product c = A dot B. More...

void	APUVectorDotConj (uint16_t N, void pInputA, void pInputB, void *pResult)
	APU accelerator for vector dot product c = A dot conj(B) More...

void	APUVectorMult (uint16_t N, void pInputA, void pInputB, void *pResult)
	APU accelerator for element-wise product of two vectors. More...

void	APUVectorScalarMult (uint16_t N, void pInputA, void pInputB, void *pResult)
	APU accelerator for product of a vector and a scalar. More...

void	APUVectorMultConj (uint16_t N, void pInputA, void pInputB, void *pResult)
	APU accelerator for element-wise product of vector A with the conjugate of vector B. More...

void	APUVectorSum (uint16_t N, void pInputA, void pInputB, uint16_t op, void *pResult)
	APU accelerator for addition/subtraction of two vectors. More...

void	APUVectorScalarSum (uint16_t N, void pInputA, void pInputB, uint16_t op, void *pResult)
	APU accelerator for addition/subtraction of a vector and a scalar. More...

void	APUVectorCart2Pol (uint16_t N, void pInput, void pResult)
	APU accelerator for element-wise Cartesian-to-Polar transformation of a vector. More...

void	APUVectorPol2Cart (uint16_t N, void pInput, void pResult, void *pTemp)
	APU accelerator for element-wise Polar-to-Cartesian transformation of a vector. More...

void	APUVectorSort (uint16_t N, void *pInput)
	APU accelerator for in-place sorting of a vector. More...

Detailed Description

Function Documentation

§ APUVectorDot()

void APUVectorDot	(	uint16_t	N,
		void *	pInputA,
		void *	pInputB,
		void *	pResult
	)

APU accelerator for vector dot product c = A dot B.

Calculate the scalar product (dot product/inner product) of two vectors.

Defined as:

c = A dot B = sum(A[i] * B[i]), i = 0 to N-1

in which, A, B are complex vectors, c is a complex scalar.

Parameters

N	is the size of the input vectors. Both vectors must be the same size
pInputA	a pointer to the base of the first input vector, in APU memory
pInputB	a pointer to the base of the second input vector, in APU memory
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_DOTPROD, and APU_GET_DATA_MEM_OFFSET.

§ APUVectorDotConj()

void APUVectorDotConj	(	uint16_t	N,
		void *	pInputA,
		void *	pInputB,
		void *	pResult
	)

APU accelerator for vector dot product c = A dot conj(B)

Calculate the scalar product (dot product/inner product) of vector A and conjugate of vector B.

Defined as:

c = A dot conj(B) = sum(A[i] * conj(B[i])), i = 0 to N-1

in which, A, B are complex vectors, c is a complex scalar.

Parameters

N	is the size of the input vectors. Both vectors must be the same size
pInputA	a pointer to the base of the first input vector, in APU memory
pInputB	a pointer to the base of the second input vector, in APU memory
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_DOTPROD, and APU_GET_DATA_MEM_OFFSET.

§ APUVectorMult()

void APUVectorMult	(	uint16_t	N,
		void *	pInputA,
		void *	pInputB,
		void *	pResult
	)

APU accelerator for element-wise product of two vectors.

Calculate the element-wise product (also called Hadamard product) of two vectors.

Defined as:

C = A .* B = [A[0] * B[0], A[1] * B[1], .., A[N-1] * B[N-1]]

in which, A, B and C are complex vectors.

Parameters

N	is the size of the input vectors. Both vectors must be the same size
pInputA	a pointer to the base of the first input vector, in APU memory
pInputB	a pointer to the base of the second input vector, in APU memory
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_VECTMULT, and APU_GET_DATA_MEM_OFFSET.

§ APUVectorScalarMult()

void APUVectorScalarMult	(	uint16_t	N,
		void *	pInputA,
		void *	pInputB,
		void *	pResult
	)

APU accelerator for product of a vector and a scalar.

Calculate the product of a vector and a scalar.

Defined as:

C = A * b = [A[0] * b, A[1] * b, .., A[N-1] * b]

in which, A, C are complex vectors and b is a complex scalar.

Parameters

N	is the size of the input and output vectors
pInputA	a pointer to the base of the first input vector, in APU memory
pInputB	a pointer to the base of the scalar, in APU memory
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_VECTMULT, and APU_GET_DATA_MEM_OFFSET.

Referenced by APUMatrixScalarMult().

§ APUVectorMultConj()

void APUVectorMultConj	(	uint16_t	N,
		void *	pInputA,
		void *	pInputB,
		void *	pResult
	)

APU accelerator for element-wise product of vector A with the conjugate of vector B.

Defined as:

C = A .* conj(B) = [A[0] * conj(B[0]), A[1] * conj(B[1]), .., A[N-1] *

conj(B[N-1])]

in which, A, B and C are complex vectors.

Parameters

N	is the size of the input vectors. Both vectors must be the same size
pInputA	a pointer to the base of the first input vector, in APU memory
pInputB	a pointer to the base of the second input vector, in APU memory
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_VECTMULT, and APU_GET_DATA_MEM_OFFSET.

§ APUVectorSum()

void APUVectorSum	(	uint16_t	N,
		void *	pInputA,
		void *	pInputB,
		uint16_t	op,
		void *	pResult
	)

APU accelerator for addition/subtraction of two vectors.

Calculate vector addition of two vectors.

Defined as:

C = A + B = [A[0] + B[0], A[1] + B[1], .., A[N-1] + B[N-1]]

in which, A, B and C are complex vectors.

Parameters

N	is the size of the input vectors. Both vectors must be the same size
pInputA	a pointer to the base of the first input vector, in APU memory
pInputB	a pointer to the base of the second input vector, in APU memory
op	select the operator, addition or substraction APU_OP_ADD APU_OP_SUB
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_VECTSUM, and APU_GET_DATA_MEM_OFFSET.

§ APUVectorScalarSum()

void APUVectorScalarSum	(	uint16_t	N,
		void *	pInputA,
		void *	pInputB,
		uint16_t	op,
		void *	pResult
	)

APU accelerator for addition/subtraction of a vector and a scalar.

Calculate vector addition/subtraction of a vector and a scalar.

Defined as:

C = A + b = [A[0] + b, A[1] + b, .., A[N-1] + b]

in which, A, and C are complex vectors, and b is a scalar.

Parameters

N	is the size of the input vectors. Both vectors must be the same size
pInputA	a pointer to the base of the first input vector, in APU memory
pInputB	a pointer to the base of the second input vector, in APU memory
op	select the operator, addition or substraction APU_OP_ADD APU_OP_SUB
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_VECTSUM, and APU_GET_DATA_MEM_OFFSET.

Referenced by APUMatrixScalarSum().

§ APUVectorCart2Pol()

void APUVectorCart2Pol	(	uint16_t	N,
		void *	pInput,
		void *	pResult
	)

APU accelerator for element-wise Cartesian-to-Polar transformation of a vector.

Calculate element-wise Cartesian-to-Polar transformation.

Defined as:

C[i] = cartesian_to_polar(A[i]), i = 0 to N-1

in which A, C are two complex vectors.

Note: The elements of input vector A are assumed in Cartesian representation. The elements of output vector C are in Polar representation.

Parameters

N	length of the vector
pInput	a pointer to the base of the first input vector A, in APU memory
pResult	a pointer to where the result will be placed, in APU memory

Returns: None

References APU_API_POLAR, and APU_GET_DATA_MEM_OFFSET.

§ APUVectorPol2Cart()

void APUVectorPol2Cart	(	uint16_t	N,
		void *	pInput,
		void *	pResult,
		void *	pTemp
	)

APU accelerator for element-wise Polar-to-Cartesian transformation of a vector.

Calculate element-wise Polar-to-Cartesian transformation.

Defined as:

C[i] = polar_to_cartesian(A[i]), i = 0 to N-1

in which A, C are two complex vectors.

Note: The elements of input vector A are assumed in Polar representation. The elements of output vector C are in Cartesian representation. This API requires a temporary vector pTemp of length N to store temporary results. Instead of using fixed and pre-defined allocation, this API allows users flexibly select the location of this temporary vector for a better utilization of available memory.

Parameters

N	length of the vector
pInput	a pointer to the base of the first input vector A, in APU memory
pResult	a pointer to where the result will be placed, in APU memory
pTemp	a pointer to a temporary vector of length N in APU memory

Returns: None

References APU_API_CARTESIAN, and APU_GET_DATA_MEM_OFFSET.

§ APUVectorSort()

void APUVectorSort	(	uint16_t	N,
		void *	pInput
	)

APU accelerator for in-place sorting of a vector.

Sort (in-place) an input vector in descending order based on the real component of each element.

Defined as:

C = sort(A, "descend", "ComparisonMethod", "real")

in which A, C are two complex vectors.

Note: This is an in-place sorting, which means the input vector will be overwritten by its sorted version.

Parameters

N	length of the vector A/C
pInput	a pointer to the base of the input vector A, in APU memory

Returns: None

References APU_API_SORT, and APU_GET_DATA_MEM_OFFSET.

Functions

Detailed Description

Function Documentation

§ APUVectorDot()

§ APUVectorDotConj()

§ APUVectorMult()

§ APUVectorScalarMult()

§ APUVectorMultConj()

§ APUVectorSum()

§ APUVectorScalarSum()

§ APUVectorCart2Pol()

§ APUVectorPol2Cart()

§ APUVectorSort()