Vector Dot Product

void riscv_dot_prod_f16(const float16_t *pSrcA, const float16_t *pSrcB, uint32_t blockSize, float16_t *result)

void riscv_dot_prod_f32(const float32_t *pSrcA, const float32_t *pSrcB, uint32_t blockSize, float32_t *result)

void riscv_dot_prod_f64(const float64_t *pSrcA, const float64_t *pSrcB, uint32_t blockSize, float64_t *result)

void riscv_dot_prod_q15(const q15_t *pSrcA, const q15_t *pSrcB, uint32_t blockSize, q63_t *result)

void riscv_dot_prod_q31(const q31_t *pSrcA, const q31_t *pSrcB, uint32_t blockSize, q63_t *result)

void riscv_dot_prod_q7(const q7_t *pSrcA, const q7_t *pSrcB, uint32_t blockSize, q31_t *result)

group BasicDotProd

Computes the dot product of two vectors. The vectors are multiplied element-by-element and then summed.

There are separate functions for floating-point, Q7, Q15, and Q31 data types.

Functions

void riscv_dot_prod_f16(const float16_t *pSrcA, const float16_t *pSrcB, uint32_t blockSize, float16_t *result)

Dot product of floating-point vectors.

Parameters

pSrcA – [in] points to the first input vector.
pSrcB – [in] points to the second input vector.
blockSize – [in] number of samples in each vector.
result – [out] output result returned here.

Returns

none

void riscv_dot_prod_f32(const float32_t *pSrcA, const float32_t *pSrcB, uint32_t blockSize, float32_t *result)

Dot product of floating-point vectors.

Parameters

pSrcA – [in] points to the first input vector.
pSrcB – [in] points to the second input vector.
blockSize – [in] number of samples in each vector.
result – [out] output result returned here.

Returns

none

void riscv_dot_prod_f64(const float64_t *pSrcA, const float64_t *pSrcB, uint32_t blockSize, float64_t *result)

Dot product of floating-point vectors.

Parameters

pSrcA – [in] points to the first input vector.
pSrcB – [in] points to the second input vector.
blockSize – [in] number of samples in each vector.
result – [out] output result returned here.

Returns

none

void riscv_dot_prod_q15(const q15_t *pSrcA, const q15_t *pSrcB, uint32_t blockSize, q63_t *result)

Dot product of Q15 vectors.

Scaling and Overflow Behavior: The intermediate multiplications are in 1.15 x 1.15 = 2.30 format and these results are added to a 64-bit accumulator in 34.30 format. Nonsaturating additions are used and given that there are 33 guard bits in the accumulator there is no risk of overflow. The return result is in 34.30 format.

Parameters

pSrcA – [in] points to the first input vector
pSrcB – [in] points to the second input vector
blockSize – [in] number of samples in each vector
result – [out] output result returned here

Returns

none

void riscv_dot_prod_q31(const q31_t *pSrcA, const q31_t *pSrcB, uint32_t blockSize, q63_t *result)

Dot product of Q31 vectors.

Scaling and Overflow Behavior: The intermediate multiplications are in 1.31 x 1.31 = 2.62 format and these are truncated to 2.48 format by discarding the lower 14 bits. The 2.48 result is then added without saturation to a 64-bit accumulator in 16.48 format. There are 15 guard bits in the accumulator and there is no risk of overflow as long as the length of the vectors is less than 2^16 elements. The return result is in 16.48 format.

Parameters

pSrcA – [in] points to the first input vector.
pSrcB – [in] points to the second input vector.
blockSize – [in] number of samples in each vector.
result – [out] output result returned here.

Returns

none

void riscv_dot_prod_q7(const q7_t *pSrcA, const q7_t *pSrcB, uint32_t blockSize, q31_t *result)

Dot product of Q7 vectors.

Scaling and Overflow Behavior: The intermediate multiplications are in 1.7 x 1.7 = 2.14 format and these results are added to an accumulator in 18.14 format. Nonsaturating additions are used and there is no danger of wrap around as long as the vectors are less than 2^18 elements long. The return result is in 18.14 format.

Parameters

pSrcA – [in] points to the first input vector
pSrcB – [in] points to the second input vector
blockSize – [in] number of samples in each vector
result – [out] output result returned here

Returns

none