Vector Multiplication

Modules

Name
Vector Multiplication Kernels

Functions

	Name
void	plp_mult_f32(const float32_t * pSrcA, const float32_t * pSrcB, float32_t * pDst, uint32_t blockSize) Glue code for element-by-element multiplication of 32-bit float vectors.
void	plp_mult_f32_parallel(const float32_t restrict pSrcA, const float32_t restrict pSrcB, uint32_t blockSize, uint32_t nPE, float32_t *restrict pDst) Glue code for parallel dot product of 32-bit float vectors.
void	plp_mult_i16(const int16_t * pSrcA, const int16_t * pSrcB, int32_t * pDst, uint32_t blockSize) Glue code for element-by-element multiplication of 16-bit integer vectors.
void	plp_mult_i32(const int32_t * pSrcA, const int32_t * pSrcB, int32_t * pDst, uint32_t blockSize) Glue code for element-by-element multiplication of 32-bit integer vectors.
void	plp_mult_i8(const int8_t * pSrcA, const int8_t * pSrcB, int32_t * pDst, uint32_t blockSize) Glue code for element-by-element multiplication of 8-bit integer vectors.

Detailed Description

This module contains the glue code for Vector Multiplication. The kernel codes (kernels) are in the Moducle Vector Multiplication Kernels.

The Vector Multiplication computes element-by-element multiplication of two vectors.

pDst[n] = pSrcA[n] * pSrcB[n], 0 <= n < blockSize.

There are separate functions for floating point, integer, and fixed point 32- 16- 8-bit data types. For lower precision integers (16- and 8-bit), functions exploiting SIMD instructions are provided.

The naming scheme of the functions follows the following pattern (for example plp_dot_prod_i32s): _ _ _ , with

data type = {f, i, q} respectively for floats, integers, fixed points

precision = {32, 16, 8} bits

method = {s, p} respectively meaning single core or parallel multicore implementation.

isa extension = rv32im, xpulpv2, etc. of which rv32im is the most general one.

Functions Documentation

function plp_mult_f32

void plp_mult_f32(
    const float32_t * pSrcA,
    const float32_t * pSrcB,
    float32_t * pDst,
    uint32_t blockSize
)

Glue code for element-by-element multiplication of 32-bit float vectors.

Parameters:

pSrcA points to first input vector
pSrcB points to second input vector
pDst points to output vector
blockSize number of samples in each vector

Return: none

function plp_mult_f32_parallel

void plp_mult_f32_parallel(
    const float32_t *__restrict__ pSrcA,
    const float32_t *__restrict__ pSrcB,
    uint32_t blockSize,
    uint32_t nPE,
    float32_t *__restrict__ pDst
)

Glue code for parallel dot product of 32-bit float vectors.

Parameters:

pSrcA points to the first input vector
pSrcB points to the second input vector
blockSize number of samples in each vector
nPE number of parallel processing units
pDst points to output vector

Return: none

function plp_mult_i16

void plp_mult_i16(
    const int16_t * pSrcA,
    const int16_t * pSrcB,
    int32_t * pDst,
    uint32_t blockSize
)

Glue code for element-by-element multiplication of 16-bit integer vectors.

Parameters:

pSrcA points to first input vector
pSrcB points to second input vector
pDst points to output vector
blockSize number of samples in each vector

Return: none

function plp_mult_i32

void plp_mult_i32(
    const int32_t * pSrcA,
    const int32_t * pSrcB,
    int32_t * pDst,
    uint32_t blockSize
)

Glue code for element-by-element multiplication of 32-bit integer vectors.

Parameters:

pSrcA points to first input vector
pSrcB points to second input vector
pDst points to output vector
blockSize number of samples in each vector

Return: none

function plp_mult_i8

void plp_mult_i8(
    const int8_t * pSrcA,
    const int8_t * pSrcB,
    int32_t * pDst,
    uint32_t blockSize
)

Glue code for element-by-element multiplication of 8-bit integer vectors.

Parameters:

pSrcA points to first input vector
pSrcB points to second input vector
pDst points to output vector
blockSize number of samples in each vector

Return: none

Updated on 2023-03-01 at 16:16:32 +0000