Details | Last modification | View Log | RSS feed
| Rev | Author | Line No. | Line |
|---|---|---|---|
| 2 | mjames | 1 | /* ---------------------------------------------------------------------- |
| 2 | * Project: CMSIS DSP Library |
||
| 3 | * Title: arm_abs_f32.c |
||
| 4 | * Description: Floating-point 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 | #include <math.h> |
||
| 31 | |||
| 32 | /** |
||
| 33 | * @ingroup groupMath |
||
| 34 | */ |
||
| 35 | |||
| 36 | /** |
||
| 37 | * @defgroup BasicAbs Vector Absolute Value |
||
| 38 | * |
||
| 39 | * Computes the absolute value of a vector on an element-by-element basis. |
||
| 40 | * |
||
| 41 | * <pre> |
||
| 42 | * pDst[n] = abs(pSrc[n]), 0 <= n < blockSize. |
||
| 43 | * </pre> |
||
| 44 | * |
||
| 45 | * The functions support in-place computation allowing the source and |
||
| 46 | * destination pointers to reference the same memory buffer. |
||
| 47 | * There are separate functions for floating-point, Q7, Q15, and Q31 data types. |
||
| 48 | */ |
||
| 49 | |||
| 50 | /** |
||
| 51 | * @addtogroup BasicAbs |
||
| 52 | * @{ |
||
| 53 | */ |
||
| 54 | |||
| 55 | /** |
||
| 56 | * @brief Floating-point vector absolute value. |
||
| 57 | * @param[in] *pSrc points to the input buffer |
||
| 58 | * @param[out] *pDst points to the output buffer |
||
| 59 | * @param[in] blockSize number of samples in each vector |
||
| 60 | * @return none. |
||
| 61 | */ |
||
| 62 | |||
| 63 | void arm_abs_f32( |
||
| 64 | float32_t * pSrc, |
||
| 65 | float32_t * pDst, |
||
| 66 | uint32_t blockSize) |
||
| 67 | { |
||
| 68 | uint32_t blkCnt; /* loop counter */ |
||
| 69 | |||
| 70 | #if defined (ARM_MATH_DSP) |
||
| 71 | |||
| 72 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
||
| 73 | float32_t in1, in2, in3, in4; /* temporary variables */ |
||
| 74 | |||
| 75 | /*loop Unrolling */ |
||
| 76 | blkCnt = blockSize >> 2U; |
||
| 77 | |||
| 78 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
||
| 79 | ** a second loop below computes the remaining 1 to 3 samples. */ |
||
| 80 | while (blkCnt > 0U) |
||
| 81 | { |
||
| 82 | /* C = |A| */ |
||
| 83 | /* Calculate absolute and then store the results in the destination buffer. */ |
||
| 84 | /* read sample from source */ |
||
| 85 | in1 = *pSrc; |
||
| 86 | in2 = *(pSrc + 1); |
||
| 87 | in3 = *(pSrc + 2); |
||
| 88 | |||
| 89 | /* find absolute value */ |
||
| 90 | in1 = fabsf(in1); |
||
| 91 | |||
| 92 | /* read sample from source */ |
||
| 93 | in4 = *(pSrc + 3); |
||
| 94 | |||
| 95 | /* find absolute value */ |
||
| 96 | in2 = fabsf(in2); |
||
| 97 | |||
| 98 | /* read sample from source */ |
||
| 99 | *pDst = in1; |
||
| 100 | |||
| 101 | /* find absolute value */ |
||
| 102 | in3 = fabsf(in3); |
||
| 103 | |||
| 104 | /* find absolute value */ |
||
| 105 | in4 = fabsf(in4); |
||
| 106 | |||
| 107 | /* store result to destination */ |
||
| 108 | *(pDst + 1) = in2; |
||
| 109 | |||
| 110 | /* store result to destination */ |
||
| 111 | *(pDst + 2) = in3; |
||
| 112 | |||
| 113 | /* store result to destination */ |
||
| 114 | *(pDst + 3) = in4; |
||
| 115 | |||
| 116 | |||
| 117 | /* Update source pointer to process next sampels */ |
||
| 118 | pSrc += 4U; |
||
| 119 | |||
| 120 | /* Update destination pointer to process next sampels */ |
||
| 121 | pDst += 4U; |
||
| 122 | |||
| 123 | /* Decrement the loop counter */ |
||
| 124 | blkCnt--; |
||
| 125 | } |
||
| 126 | |||
| 127 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
||
| 128 | ** No loop unrolling is used. */ |
||
| 129 | blkCnt = blockSize % 0x4U; |
||
| 130 | |||
| 131 | #else |
||
| 132 | |||
| 133 | /* Run the below code for Cortex-M0 */ |
||
| 134 | |||
| 135 | /* Initialize blkCnt with number of samples */ |
||
| 136 | blkCnt = blockSize; |
||
| 137 | |||
| 138 | #endif /* #if defined (ARM_MATH_DSP) */ |
||
| 139 | |||
| 140 | while (blkCnt > 0U) |
||
| 141 | { |
||
| 142 | /* C = |A| */ |
||
| 143 | /* Calculate absolute and then store the results in the destination buffer. */ |
||
| 144 | *pDst++ = fabsf(*pSrc++); |
||
| 145 | |||
| 146 | /* Decrement the loop counter */ |
||
| 147 | blkCnt--; |
||
| 148 | } |
||
| 149 | } |
||
| 150 | |||
| 151 | /** |
||
| 152 | * @} end of BasicAbs group |
||
| 153 | */ |