diff options
Diffstat (limited to 'Drivers/CMSIS/DSP/Source/BasicMathFunctions/arm_dot_prod_q7.c')
| -rw-r--r-- | Drivers/CMSIS/DSP/Source/BasicMathFunctions/arm_dot_prod_q7.c | 118 |
1 files changed, 55 insertions, 63 deletions
diff --git a/Drivers/CMSIS/DSP/Source/BasicMathFunctions/arm_dot_prod_q7.c b/Drivers/CMSIS/DSP/Source/BasicMathFunctions/arm_dot_prod_q7.c index 487efe397..8e18a7397 100644 --- a/Drivers/CMSIS/DSP/Source/BasicMathFunctions/arm_dot_prod_q7.c +++ b/Drivers/CMSIS/DSP/Source/BasicMathFunctions/arm_dot_prod_q7.c @@ -3,13 +3,13 @@ * Title: arm_dot_prod_q7.c * Description: Q7 dot product * - * $Date: 27. January 2017 - * $Revision: V.1.5.1 + * $Date: 18. March 2019 + * $Revision: V1.6.0 * * Target Processor: Cortex-M cores * -------------------------------------------------------------------- */ /* - * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved. + * Copyright (C) 2010-2019 ARM Limited or its affiliates. All rights reserved. * * SPDX-License-Identifier: Apache-2.0 * @@ -29,61 +29,58 @@ #include "arm_math.h" /** - * @ingroup groupMath + @ingroup groupMath */ /** - * @addtogroup dot_prod - * @{ + @addtogroup BasicDotProd + @{ */ /** - * @brief Dot product of Q7 vectors. - * @param[in] *pSrcA points to the first input vector - * @param[in] *pSrcB points to the second input vector - * @param[in] blockSize number of samples in each vector - * @param[out] *result output result returned here - * @return none. - * - * <b>Scaling and Overflow Behavior:</b> - * \par - * 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. + @brief Dot product of Q7 vectors. + @param[in] pSrcA points to the first input vector + @param[in] pSrcB points to the second input vector + @param[in] blockSize number of samples in each vector + @param[out] result output result returned here + @return none + + @par 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. */ void arm_dot_prod_q7( - q7_t * pSrcA, - q7_t * pSrcB, - uint32_t blockSize, - q31_t * result) + const q7_t * pSrcA, + const q7_t * pSrcB, + uint32_t blockSize, + q31_t * result) { - uint32_t blkCnt; /* loop counter */ + uint32_t blkCnt; /* Loop counter */ + q31_t sum = 0; /* Temporary return variable */ - q31_t sum = 0; /* Temporary variables to store output */ +#if defined (ARM_MATH_LOOPUNROLL) #if defined (ARM_MATH_DSP) + q31_t input1, input2; /* Temporary variables */ + q31_t inA1, inA2, inB1, inB2; /* Temporary variables */ +#endif -/* Run the below code for Cortex-M4 and Cortex-M3 */ - - q31_t input1, input2; /* Temporary variables to store input */ - q31_t inA1, inA2, inB1, inB2; /* Temporary variables to store input */ - - - - /*loop Unrolling */ + /* Loop unrolling: Compute 4 outputs at a time */ blkCnt = blockSize >> 2U; - /* First part of the processing with loop unrolling. Compute 4 outputs at a time. - ** a second loop below computes the remaining 1 to 3 samples. */ while (blkCnt > 0U) { + /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */ + +#if defined (ARM_MATH_DSP) /* read 4 samples at a time from sourceA */ - input1 = *__SIMD32(pSrcA)++; + input1 = read_q7x4_ia ((q7_t **) &pSrcA); /* read 4 samples at a time from sourceB */ - input2 = *__SIMD32(pSrcB)++; + input2 = read_q7x4_ia ((q7_t **) &pSrcB); /* extract two q7_t samples to q15_t samples */ inA1 = __SXTB16(__ROR(input1, 8)); @@ -97,51 +94,46 @@ void arm_dot_prod_q7( /* multiply and accumulate two samples at a time */ sum = __SMLAD(inA1, inB1, sum); sum = __SMLAD(inA2, inB2, sum); +#else + sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); + sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); + sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); + sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); +#endif - /* Decrement the loop counter */ + /* Decrement loop counter */ blkCnt--; } - /* If the blockSize is not a multiple of 4, compute any remaining output samples here. - ** No loop unrolling is used. */ + /* Loop unrolling: Compute remaining outputs */ blkCnt = blockSize % 0x4U; - while (blkCnt > 0U) - { - /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */ - /* Dot product and then store the results in a temporary buffer. */ - sum = __SMLAD(*pSrcA++, *pSrcB++, sum); - - /* Decrement the loop counter */ - blkCnt--; - } - #else - /* Run the below code for Cortex-M0 */ - - - /* Initialize blkCnt with number of samples */ blkCnt = blockSize; +#endif /* #if defined (ARM_MATH_LOOPUNROLL) */ + while (blkCnt > 0U) { /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */ - /* Dot product and then store the results in a temporary buffer. */ - sum += (q31_t) ((q15_t) * pSrcA++ * *pSrcB++); - /* Decrement the loop counter */ + /* Calculate dot product and store result in a temporary buffer. */ +//#if defined (ARM_MATH_DSP) +// sum = __SMLAD(*pSrcA++, *pSrcB++, sum); +//#else + sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); +//#endif + + /* Decrement loop counter */ blkCnt--; } -#endif /* #if defined (ARM_MATH_DSP) */ - - - /* Store the result in the destination buffer in 18.14 format */ + /* Store result in destination buffer in 18.14 format */ *result = sum; } /** - * @} end of dot_prod group + @} end of BasicDotProd group */ |