Power

Modules

Name
Power Kernels
RMS Kernels

Functions

	Name
void	plp_power_f32(const float restrict pSrc, uint32_t blockSize, float restrict pRes) Glue code for sum of squares of a 32-bit float vector.
void	plp_power_f32_parallel(const float32_t restrict pSrc, uint32_t blockSize, uint32_t nPE, float32_t restrict pRes) Glue code for parallel power of 32-bit floating point vectors.
void	plp_power_i16(const int16_t restrict pSrc, uint32_t blockSize, int32_t restrict pRes) Glue code for sum of squares of a 16-bit integer vector.
void	plp_power_i32(const int32_t restrict pSrc, uint32_t blockSize, int32_t restrict pRes) Glue code for sum of squares of a 32-bit integer vector.
void	plp_power_i8(const int8_t restrict pSrc, uint32_t blockSize, int32_t restrict pRes) Glue code for sum of squares of a 8-bit integer vector.
void	plp_power_q16(const int16_t restrict pSrc, uint32_t blockSize, uint32_t fracBits, int32_t restrict pRes) Glue code for sum of squares of a 8-bit fixed point vector.
void	plp_power_q32(const int32_t restrict pSrc, uint32_t blockSize, uint32_t fracBits, int32_t restrict pRes) Glue code for sum of squares of a 8-bit fixed point vector.
void	plp_power_q32_parallel(const int32_t restrict pSrc, uint32_t blockSize, uint32_t deciPoint, uint32_t nPE, int32_t restrict pRes) Glue code for parallel power of 32-bit fixed-point vectors.
void	plp_power_q8(const int8_t restrict pSrc, uint32_t blockSize, uint32_t fracBits, int32_t restrict pRes) Glue code for sum of squares of a 8-bit fixed point vector.
void	plp_rms_f32(const float restrict pSrc, uint32_t blockSize, float restrict pRes) Glue code for RMS value of a 32-bit float vector.
void	plp_rms_q16(const int16_t restrict pSrc, uint32_t blockSize, uint32_t fracBits, int16_t restrict pRes) Glue code for RMS value of a 8-bit fixed point vector.
void	plp_rms_q32(const int32_t restrict pSrc, uint32_t blockSize, uint32_t fracBits, int32_t restrict pRes) Glue code for RMS value of a 8-bit fixed point vector.
void	plp_rms_q8(const int8_t restrict pSrc, uint32_t blockSize, uint32_t fracBits, int8_t restrict pRes) Glue code for RMS value of a 8-bit fixed point vector.

Detailed Description

Calculates the sum of squares of the input vector. 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.

Calculates the RMS value of the input vector. 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_power_f32

void plp_power_f32(
    const float *__restrict__ pSrc,
    uint32_t blockSize,
    float *__restrict__ pRes
)

Glue code for sum of squares of a 32-bit float vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes sum of squares returned here

Return: none

Glue code for Sum of squares of a 32-bit float vector.

function plp_power_f32_parallel

void plp_power_f32_parallel(
    const float32_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t nPE,
    float32_t *__restrict__ pRes
)

Glue code for parallel power of 32-bit floating point vectors.

Parameters:

pSrc points to the input vector
blockSize number of samples in each vector
fracBits number of fixed point fractional bits
nPE number of parallel processing units
pRes output result returned here

Return: none

function plp_power_i16

void plp_power_i16(
    const int16_t *__restrict__ pSrc,
    uint32_t blockSize,
    int32_t *__restrict__ pRes
)

Glue code for sum of squares of a 16-bit integer vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes sum of squares returned here

Return: none

Glue code for Sum of squares of a 16-bit integer vector.

function plp_power_i32

void plp_power_i32(
    const int32_t *__restrict__ pSrc,
    uint32_t blockSize,
    int32_t *__restrict__ pRes
)

Glue code for sum of squares of a 32-bit integer vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes sum of squares returned here

Return: none

Glue code for Sum of squares of a 32-bit integer vector.

function plp_power_i8

void plp_power_i8(
    const int8_t *__restrict__ pSrc,
    uint32_t blockSize,
    int32_t *__restrict__ pRes
)

Glue code for sum of squares of a 8-bit integer vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes sum of squares returned here

Return: none

Glue code for Sum of squares of a 8-bit integer vector.

function plp_power_q16

void plp_power_q16(
    const int16_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t fracBits,
    int32_t *__restrict__ pRes
)

Glue code for sum of squares of a 8-bit fixed point vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes sum of squares returned here

Return: none

Glue code for Sum of squares of a 16-bit fixed point vector.

function plp_power_q32

void plp_power_q32(
    const int32_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t fracBits,
    int32_t *__restrict__ pRes
)

Glue code for sum of squares of a 8-bit fixed point vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes sum of squares returned here

Return: none

Glue code for Sum of squares of a 32-bit fixed point vector.

function plp_power_q32_parallel

void plp_power_q32_parallel(
    const int32_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t deciPoint,
    uint32_t nPE,
    int32_t *__restrict__ pRes
)

Glue code for parallel power of 32-bit fixed-point vectors.

Parameters:

pSrc points to the input vector
blockSize number of samples in each vector
deciPoint number of fixed point fractional bits
nPE number of parallel processing units
pRes output result returned here

Return: none

Glue code for parallel power of 32-bit fixed point vectors.

function plp_power_q8

void plp_power_q8(
    const int8_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t fracBits,
    int32_t *__restrict__ pRes
)

Glue code for sum of squares of a 8-bit fixed point vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes sum of squares returned here

Return: none

Glue code for Sum of squares of a 8-bit fixed point vector.

function plp_rms_f32

void plp_rms_f32(
    const float *__restrict__ pSrc,
    uint32_t blockSize,
    float *__restrict__ pRes
)

Glue code for RMS value of a 32-bit float vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes RMS value returned here

Return: none

Glue code for Statisical standard deviation of a 32-bit floating point vector.

function plp_rms_q16

void plp_rms_q16(
    const int16_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t fracBits,
    int16_t *__restrict__ pRes
)

Glue code for RMS value of a 8-bit fixed point vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes RMS value returned here

Return: none

Glue code for Statisical standard deviation of a 16-bit fixed point vector.

function plp_rms_q32

void plp_rms_q32(
    const int32_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t fracBits,
    int32_t *__restrict__ pRes
)

Glue code for RMS value of a 8-bit fixed point vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes RMS value returned here

Return: none

Glue code for Statisical standard deviation of a 32-bit fixed point vector.

function plp_rms_q8

void plp_rms_q8(
    const int8_t *__restrict__ pSrc,
    uint32_t blockSize,
    uint32_t fracBits,
    int8_t *__restrict__ pRes
)

Glue code for RMS value of a 8-bit fixed point vector.

Parameters:

pSrc points to the input vector
blockSize number of samples in input vector
pRes RMS value returned here

Return: none

Glue code for Statisical standard deviation of a 8-bit fixed point vector.

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