File size: 5,665 Bytes
25ade36 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 | #include "edge-impulse-sdk/dsp/config.hpp"
#if EIDSP_LOAD_CMSIS_DSP_SOURCES
/* ----------------------------------------------------------------------
* Project: CMSIS DSP Library
* Title: arm_mult_q15.c
* Description: Q15 vector multiplication
*
* $Date: 23 April 2021
* $Revision: V1.9.0
*
* Target Processor: Cortex-M and Cortex-A cores
* -------------------------------------------------------------------- */
/*
* Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
*
* Licensed under the Apache License, Version 2.0 (the License); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an AS IS BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include "edge-impulse-sdk/CMSIS/DSP/Include/dsp/basic_math_functions.h"
/**
@ingroup groupMath
*/
/**
@addtogroup BasicMult
@{
*/
/**
@brief Q15 vector multiplication
@param[in] pSrcA points to first input vector
@param[in] pSrcB points to second input vector
@param[out] pDst points to output vector
@param[in] blockSize number of samples in each vector
@return none
@par Scaling and Overflow Behavior
The function uses saturating arithmetic.
Results outside of the allowable Q15 range [0x8000 0x7FFF] are saturated.
*/
#if defined(ARM_MATH_MVEI) && !defined(ARM_MATH_AUTOVECTORIZE)
#include "edge-impulse-sdk/CMSIS/DSP/Include/arm_helium_utils.h"
void arm_mult_q15(
const q15_t * pSrcA,
const q15_t * pSrcB,
q15_t * pDst,
uint32_t blockSize)
{
uint32_t blkCnt; /* loop counters */
q15x8_t vecA, vecB;
/* Compute 8 outputs at a time */
blkCnt = blockSize >> 3;
while (blkCnt > 0U)
{
/*
* C = A * B
* Multiply the inputs and then store the results in the destination buffer.
*/
vecA = vld1q(pSrcA);
vecB = vld1q(pSrcB);
vst1q(pDst, vqdmulhq(vecA, vecB));
/*
* Decrement the blockSize loop counter
*/
blkCnt--;
/*
* advance vector source and destination pointers
*/
pSrcA += 8;
pSrcB += 8;
pDst += 8;
}
/*
* tail
*/
blkCnt = blockSize & 7;
if (blkCnt > 0U)
{
mve_pred16_t p0 = vctp16q(blkCnt);
vecA = vld1q(pSrcA);
vecB = vld1q(pSrcB);
vstrhq_p(pDst, vqdmulhq(vecA, vecB), p0);
}
}
#else
void arm_mult_q15(
const q15_t * pSrcA,
const q15_t * pSrcB,
q15_t * pDst,
uint32_t blockSize)
{
uint32_t blkCnt; /* Loop counter */
#if defined (ARM_MATH_LOOPUNROLL)
#if defined (ARM_MATH_DSP)
q31_t inA1, inA2, inB1, inB2; /* Temporary input variables */
q15_t out1, out2, out3, out4; /* Temporary output variables */
q31_t mul1, mul2, mul3, mul4; /* Temporary variables */
#endif
/* Loop unrolling: Compute 4 outputs at a time */
blkCnt = blockSize >> 2U;
while (blkCnt > 0U)
{
/* C = A * B */
#if defined (ARM_MATH_DSP)
/* read 2 samples at a time from sourceA */
inA1 = read_q15x2_ia (&pSrcA);
/* read 2 samples at a time from sourceB */
inB1 = read_q15x2_ia (&pSrcB);
/* read 2 samples at a time from sourceA */
inA2 = read_q15x2_ia (&pSrcA);
/* read 2 samples at a time from sourceB */
inB2 = read_q15x2_ia (&pSrcB);
/* multiply mul = sourceA * sourceB */
mul1 = (q31_t) ((q15_t) (inA1 >> 16) * (q15_t) (inB1 >> 16));
mul2 = (q31_t) ((q15_t) (inA1 ) * (q15_t) (inB1 ));
mul3 = (q31_t) ((q15_t) (inA2 >> 16) * (q15_t) (inB2 >> 16));
mul4 = (q31_t) ((q15_t) (inA2 ) * (q15_t) (inB2 ));
/* saturate result to 16 bit */
out1 = (q15_t) __SSAT(mul1 >> 15, 16);
out2 = (q15_t) __SSAT(mul2 >> 15, 16);
out3 = (q15_t) __SSAT(mul3 >> 15, 16);
out4 = (q15_t) __SSAT(mul4 >> 15, 16);
/* store result to destination */
#ifndef ARM_MATH_BIG_ENDIAN
write_q15x2_ia (&pDst, __PKHBT(out2, out1, 16));
write_q15x2_ia (&pDst, __PKHBT(out4, out3, 16));
#else
write_q15x2_ia (&pDst, __PKHBT(out1, out2, 16));
write_q15x2_ia (&pDst, __PKHBT(out3, out4, 16));
#endif /* #ifndef ARM_MATH_BIG_ENDIAN */
#else
*pDst++ = (q15_t) __SSAT((((q31_t) (*pSrcA++) * (*pSrcB++)) >> 15), 16);
*pDst++ = (q15_t) __SSAT((((q31_t) (*pSrcA++) * (*pSrcB++)) >> 15), 16);
*pDst++ = (q15_t) __SSAT((((q31_t) (*pSrcA++) * (*pSrcB++)) >> 15), 16);
*pDst++ = (q15_t) __SSAT((((q31_t) (*pSrcA++) * (*pSrcB++)) >> 15), 16);
#endif
/* Decrement loop counter */
blkCnt--;
}
/* Loop unrolling: Compute remaining outputs */
blkCnt = blockSize % 0x4U;
#else
/* Initialize blkCnt with number of samples */
blkCnt = blockSize;
#endif /* #if defined (ARM_MATH_LOOPUNROLL) */
while (blkCnt > 0U)
{
/* C = A * B */
/* Multiply inputs and store result in destination buffer. */
*pDst++ = (q15_t) __SSAT((((q31_t) (*pSrcA++) * (*pSrcB++)) >> 15), 16);
/* Decrement loop counter */
blkCnt--;
}
}
#endif /* defined(ARM_MATH_MVEI) */
/**
@} end of BasicMult group
*/
#endif // EIDSP_LOAD_CMSIS_DSP_SOURCES
|