Go to most recent revision | Details | Last modification | View Log | RSS feed
| Rev | Author | Line No. | Line |
|---|---|---|---|
| 56 | mjames | 1 | /* ---------------------------------------------------------------------- |
| 2 | * Project: CMSIS DSP Library |
||
| 3 | * Title: arm_abs_q15.c |
||
| 4 | * Description: Q15 vector absolute value |
||
| 5 | * |
||
| 6 | * $Date: 27. January 2017 |
||
| 7 | * $Revision: V.1.5.1 |
||
| 8 | * |
||
| 9 | * Target Processor: Cortex-M cores |
||
| 10 | * -------------------------------------------------------------------- */ |
||
| 11 | /* |
||
| 12 | * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved. |
||
| 13 | * |
||
| 14 | * SPDX-License-Identifier: Apache-2.0 |
||
| 15 | * |
||
| 16 | * Licensed under the Apache License, Version 2.0 (the License); you may |
||
| 17 | * not use this file except in compliance with the License. |
||
| 18 | * You may obtain a copy of the License at |
||
| 19 | * |
||
| 20 | * www.apache.org/licenses/LICENSE-2.0 |
||
| 21 | * |
||
| 22 | * Unless required by applicable law or agreed to in writing, software |
||
| 23 | * distributed under the License is distributed on an AS IS BASIS, WITHOUT |
||
| 24 | * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
||
| 25 | * See the License for the specific language governing permissions and |
||
| 26 | * limitations under the License. |
||
| 27 | */ |
||
| 28 | |||
| 29 | #include "arm_math.h" |
||
| 30 | |||
| 31 | /** |
||
| 32 | * @ingroup groupMath |
||
| 33 | */ |
||
| 34 | |||
| 35 | /** |
||
| 36 | * @addtogroup BasicAbs |
||
| 37 | * @{ |
||
| 38 | */ |
||
| 39 | |||
| 40 | /** |
||
| 41 | * @brief Q15 vector absolute value. |
||
| 42 | * @param[in] *pSrc points to the input buffer |
||
| 43 | * @param[out] *pDst points to the output buffer |
||
| 44 | * @param[in] blockSize number of samples in each vector |
||
| 45 | * @return none. |
||
| 46 | * |
||
| 47 | * <b>Scaling and Overflow Behavior:</b> |
||
| 48 | * \par |
||
| 49 | * The function uses saturating arithmetic. |
||
| 50 | * The Q15 value -1 (0x8000) will be saturated to the maximum allowable positive value 0x7FFF. |
||
| 51 | */ |
||
| 52 | |||
| 53 | void arm_abs_q15( |
||
| 54 | q15_t * pSrc, |
||
| 55 | q15_t * pDst, |
||
| 56 | uint32_t blockSize) |
||
| 57 | { |
||
| 58 | uint32_t blkCnt; /* loop counter */ |
||
| 59 | |||
| 60 | #if defined (ARM_MATH_DSP) |
||
| 61 | __SIMD32_TYPE *simd; |
||
| 62 | |||
| 63 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
||
| 64 | |||
| 65 | q15_t in1; /* Input value1 */ |
||
| 66 | q15_t in2; /* Input value2 */ |
||
| 67 | |||
| 68 | |||
| 69 | /*loop Unrolling */ |
||
| 70 | blkCnt = blockSize >> 2U; |
||
| 71 | |||
| 72 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
||
| 73 | ** a second loop below computes the remaining 1 to 3 samples. */ |
||
| 74 | simd = __SIMD32_CONST(pDst); |
||
| 75 | while (blkCnt > 0U) |
||
| 76 | { |
||
| 77 | /* C = |A| */ |
||
| 78 | /* Read two inputs */ |
||
| 79 | in1 = *pSrc++; |
||
| 80 | in2 = *pSrc++; |
||
| 81 | |||
| 82 | |||
| 83 | /* Store the Absolute result in the destination buffer by packing the two values, in a single cycle */ |
||
| 84 | #ifndef ARM_MATH_BIG_ENDIAN |
||
| 85 | *simd++ = |
||
| 86 | __PKHBT(((in1 > 0) ? in1 : (q15_t)__QSUB16(0, in1)), |
||
| 87 | ((in2 > 0) ? in2 : (q15_t)__QSUB16(0, in2)), 16); |
||
| 88 | |||
| 89 | #else |
||
| 90 | |||
| 91 | |||
| 92 | *simd++ = |
||
| 93 | __PKHBT(((in2 > 0) ? in2 : (q15_t)__QSUB16(0, in2)), |
||
| 94 | ((in1 > 0) ? in1 : (q15_t)__QSUB16(0, in1)), 16); |
||
| 95 | |||
| 96 | #endif /* #ifndef ARM_MATH_BIG_ENDIAN */ |
||
| 97 | |||
| 98 | in1 = *pSrc++; |
||
| 99 | in2 = *pSrc++; |
||
| 100 | |||
| 101 | |||
| 102 | #ifndef ARM_MATH_BIG_ENDIAN |
||
| 103 | |||
| 104 | *simd++ = |
||
| 105 | __PKHBT(((in1 > 0) ? in1 : (q15_t)__QSUB16(0, in1)), |
||
| 106 | ((in2 > 0) ? in2 : (q15_t)__QSUB16(0, in2)), 16); |
||
| 107 | |||
| 108 | #else |
||
| 109 | |||
| 110 | |||
| 111 | *simd++ = |
||
| 112 | __PKHBT(((in2 > 0) ? in2 : (q15_t)__QSUB16(0, in2)), |
||
| 113 | ((in1 > 0) ? in1 : (q15_t)__QSUB16(0, in1)), 16); |
||
| 114 | |||
| 115 | #endif /* #ifndef ARM_MATH_BIG_ENDIAN */ |
||
| 116 | |||
| 117 | /* Decrement the loop counter */ |
||
| 118 | blkCnt--; |
||
| 119 | } |
||
| 120 | pDst = (q15_t *)simd; |
||
| 121 | |||
| 122 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
||
| 123 | ** No loop unrolling is used. */ |
||
| 124 | blkCnt = blockSize % 0x4U; |
||
| 125 | |||
| 126 | while (blkCnt > 0U) |
||
| 127 | { |
||
| 128 | /* C = |A| */ |
||
| 129 | /* Read the input */ |
||
| 130 | in1 = *pSrc++; |
||
| 131 | |||
| 132 | /* Calculate absolute value of input and then store the result in the destination buffer. */ |
||
| 133 | *pDst++ = (in1 > 0) ? in1 : (q15_t)__QSUB16(0, in1); |
||
| 134 | |||
| 135 | /* Decrement the loop counter */ |
||
| 136 | blkCnt--; |
||
| 137 | } |
||
| 138 | |||
| 139 | #else |
||
| 140 | |||
| 141 | /* Run the below code for Cortex-M0 */ |
||
| 142 | |||
| 143 | q15_t in; /* Temporary input variable */ |
||
| 144 | |||
| 145 | /* Initialize blkCnt with number of samples */ |
||
| 146 | blkCnt = blockSize; |
||
| 147 | |||
| 148 | while (blkCnt > 0U) |
||
| 149 | { |
||
| 150 | /* C = |A| */ |
||
| 151 | /* Read the input */ |
||
| 152 | in = *pSrc++; |
||
| 153 | |||
| 154 | /* Calculate absolute value of input and then store the result in the destination buffer. */ |
||
| 155 | *pDst++ = (in > 0) ? in : ((in == (q15_t) 0x8000) ? 0x7fff : -in); |
||
| 156 | |||
| 157 | /* Decrement the loop counter */ |
||
| 158 | blkCnt--; |
||
| 159 | } |
||
| 160 | |||
| 161 | #endif /* #if defined (ARM_MATH_DSP) */ |
||
| 162 | |||
| 163 | } |
||
| 164 | |||
| 165 | /** |
||
| 166 | * @} end of BasicAbs group |
||
| 167 | */ |